diff -urNp linux-2.6.32.48/arch/alpha/include/asm/elf.h linux-2.6.32.48-openvz/arch/alpha/include/asm/elf.h --- linux-2.6.32.48/arch/alpha/include/asm/elf.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/alpha/include/asm/elf.h 2011-11-17 20:08:53.000000000 -0500 @@ -91,6 +91,13 @@ typedef elf_fpreg_t elf_fpregset_t[ELF_N #define ELF_ET_DYN_BASE (TASK_UNMAPPED_BASE + 0x1000000) +#ifdef CONFIG_PAX_ASLR +#define PAX_ELF_ET_DYN_BASE (current->personality & ADDR_LIMIT_32BIT ? 0x10000 : 0x120000000UL) + +#define PAX_DELTA_MMAP_LEN (current->personality & ADDR_LIMIT_32BIT ? 14 : 28) +#define PAX_DELTA_STACK_LEN (current->personality & ADDR_LIMIT_32BIT ? 14 : 19) +#endif + /* $0 is set by ld.so to a pointer to a function which might be registered using atexit. This provides a mean for the dynamic linker to call DT_FINI functions for shared libraries that have diff -urNp linux-2.6.32.48/arch/alpha/include/asm/pgtable.h linux-2.6.32.48-openvz/arch/alpha/include/asm/pgtable.h --- linux-2.6.32.48/arch/alpha/include/asm/pgtable.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/alpha/include/asm/pgtable.h 2011-11-17 20:08:53.000000000 -0500 @@ -101,6 +101,17 @@ struct vm_area_struct; #define PAGE_SHARED __pgprot(_PAGE_VALID | __ACCESS_BITS) #define PAGE_COPY __pgprot(_PAGE_VALID | __ACCESS_BITS | _PAGE_FOW) #define PAGE_READONLY __pgprot(_PAGE_VALID | __ACCESS_BITS | _PAGE_FOW) + +#ifdef CONFIG_PAX_PAGEEXEC +# define PAGE_SHARED_NOEXEC __pgprot(_PAGE_VALID | __ACCESS_BITS | _PAGE_FOE) +# define PAGE_COPY_NOEXEC __pgprot(_PAGE_VALID | __ACCESS_BITS | _PAGE_FOW | _PAGE_FOE) +# define PAGE_READONLY_NOEXEC __pgprot(_PAGE_VALID | __ACCESS_BITS | _PAGE_FOW | _PAGE_FOE) +#else +# define PAGE_SHARED_NOEXEC PAGE_SHARED +# define PAGE_COPY_NOEXEC PAGE_COPY +# define PAGE_READONLY_NOEXEC PAGE_READONLY +#endif + #define PAGE_KERNEL __pgprot(_PAGE_VALID | _PAGE_ASM | _PAGE_KRE | _PAGE_KWE) #define _PAGE_NORMAL(x) __pgprot(_PAGE_VALID | __ACCESS_BITS | (x)) diff -urNp linux-2.6.32.48/arch/alpha/kernel/module.c linux-2.6.32.48-openvz/arch/alpha/kernel/module.c --- linux-2.6.32.48/arch/alpha/kernel/module.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/alpha/kernel/module.c 2011-11-17 20:08:53.000000000 -0500 @@ -182,7 +182,7 @@ apply_relocate_add(Elf64_Shdr *sechdrs, /* The small sections were sorted to the end of the segment. The following should definitely cover them. */ - gp = (u64)me->module_core + me->core_size - 0x8000; + gp = (u64)me->module_core_rw + me->core_size_rw - 0x8000; got = sechdrs[me->arch.gotsecindex].sh_addr; for (i = 0; i < n; i++) { diff -urNp linux-2.6.32.48/arch/alpha/kernel/osf_sys.c linux-2.6.32.48-openvz/arch/alpha/kernel/osf_sys.c --- linux-2.6.32.48/arch/alpha/kernel/osf_sys.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/alpha/kernel/osf_sys.c 2011-11-17 20:09:04.000000000 -0500 @@ -1172,7 +1172,7 @@ arch_get_unmapped_area_1(unsigned long a /* At this point: (!vma || addr < vma->vm_end). */ if (limit - len < addr) return -ENOMEM; - if (!vma || addr + len <= vma->vm_start) + if (check_heap_stack_gap(vma, addr, len)) return addr; addr = vma->vm_end; vma = vma->vm_next; @@ -1208,6 +1208,10 @@ arch_get_unmapped_area(struct file *filp merely specific addresses, but regions of memory -- perhaps this feature should be incorporated into all ports? */ +#ifdef CONFIG_PAX_RANDMMAP + if (!(current->mm->pax_flags & MF_PAX_RANDMMAP)) +#endif + if (addr) { addr = arch_get_unmapped_area_1 (PAGE_ALIGN(addr), len, limit); if (addr != (unsigned long) -ENOMEM) @@ -1215,8 +1219,8 @@ arch_get_unmapped_area(struct file *filp } /* Next, try allocating at TASK_UNMAPPED_BASE. */ - addr = arch_get_unmapped_area_1 (PAGE_ALIGN(TASK_UNMAPPED_BASE), - len, limit); + addr = arch_get_unmapped_area_1 (PAGE_ALIGN(current->mm->mmap_base), len, limit); + if (addr != (unsigned long) -ENOMEM) return addr; diff -urNp linux-2.6.32.48/arch/alpha/mm/fault.c linux-2.6.32.48-openvz/arch/alpha/mm/fault.c --- linux-2.6.32.48/arch/alpha/mm/fault.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/alpha/mm/fault.c 2011-11-17 20:08:53.000000000 -0500 @@ -54,6 +54,124 @@ __load_new_mm_context(struct mm_struct * __reload_thread(pcb); } +#ifdef CONFIG_PAX_PAGEEXEC +/* + * PaX: decide what to do with offenders (regs->pc = fault address) + * + * returns 1 when task should be killed + * 2 when patched PLT trampoline was detected + * 3 when unpatched PLT trampoline was detected + */ +static int pax_handle_fetch_fault(struct pt_regs *regs) +{ + +#ifdef CONFIG_PAX_EMUPLT + int err; + + do { /* PaX: patched PLT emulation #1 */ + unsigned int ldah, ldq, jmp; + + err = get_user(ldah, (unsigned int *)regs->pc); + err |= get_user(ldq, (unsigned int *)(regs->pc+4)); + err |= get_user(jmp, (unsigned int *)(regs->pc+8)); + + if (err) + break; + + if ((ldah & 0xFFFF0000U) == 0x277B0000U && + (ldq & 0xFFFF0000U) == 0xA77B0000U && + jmp == 0x6BFB0000U) + { + unsigned long r27, addr; + unsigned long addrh = (ldah | 0xFFFFFFFFFFFF0000UL) << 16; + unsigned long addrl = ldq | 0xFFFFFFFFFFFF0000UL; + + addr = regs->r27 + ((addrh ^ 0x80000000UL) + 0x80000000UL) + ((addrl ^ 0x8000UL) + 0x8000UL); + err = get_user(r27, (unsigned long *)addr); + if (err) + break; + + regs->r27 = r27; + regs->pc = r27; + return 2; + } + } while (0); + + do { /* PaX: patched PLT emulation #2 */ + unsigned int ldah, lda, br; + + err = get_user(ldah, (unsigned int *)regs->pc); + err |= get_user(lda, (unsigned int *)(regs->pc+4)); + err |= get_user(br, (unsigned int *)(regs->pc+8)); + + if (err) + break; + + if ((ldah & 0xFFFF0000U) == 0x277B0000U && + (lda & 0xFFFF0000U) == 0xA77B0000U && + (br & 0xFFE00000U) == 0xC3E00000U) + { + unsigned long addr = br | 0xFFFFFFFFFFE00000UL; + unsigned long addrh = (ldah | 0xFFFFFFFFFFFF0000UL) << 16; + unsigned long addrl = lda | 0xFFFFFFFFFFFF0000UL; + + regs->r27 += ((addrh ^ 0x80000000UL) + 0x80000000UL) + ((addrl ^ 0x8000UL) + 0x8000UL); + regs->pc += 12 + (((addr ^ 0x00100000UL) + 0x00100000UL) << 2); + return 2; + } + } while (0); + + do { /* PaX: unpatched PLT emulation */ + unsigned int br; + + err = get_user(br, (unsigned int *)regs->pc); + + if (!err && (br & 0xFFE00000U) == 0xC3800000U) { + unsigned int br2, ldq, nop, jmp; + unsigned long addr = br | 0xFFFFFFFFFFE00000UL, resolver; + + addr = regs->pc + 4 + (((addr ^ 0x00100000UL) + 0x00100000UL) << 2); + err = get_user(br2, (unsigned int *)addr); + err |= get_user(ldq, (unsigned int *)(addr+4)); + err |= get_user(nop, (unsigned int *)(addr+8)); + err |= get_user(jmp, (unsigned int *)(addr+12)); + err |= get_user(resolver, (unsigned long *)(addr+16)); + + if (err) + break; + + if (br2 == 0xC3600000U && + ldq == 0xA77B000CU && + nop == 0x47FF041FU && + jmp == 0x6B7B0000U) + { + regs->r28 = regs->pc+4; + regs->r27 = addr+16; + regs->pc = resolver; + return 3; + } + } + } while (0); +#endif + + return 1; +} + +void pax_report_insns(void *pc, void *sp) +{ + unsigned long i; + + printk(KERN_ERR "PAX: bytes at PC: "); + for (i = 0; i < 5; i++) { + unsigned int c; + if (get_user(c, (unsigned int *)pc+i)) + printk(KERN_CONT "???????? "); + else + printk(KERN_CONT "%08x ", c); + } + printk("\n"); +} +#endif /* * This routine handles page faults. It determines the address, @@ -131,8 +249,29 @@ do_page_fault(unsigned long address, uns good_area: si_code = SEGV_ACCERR; if (cause < 0) { - if (!(vma->vm_flags & VM_EXEC)) + if (!(vma->vm_flags & VM_EXEC)) { + +#ifdef CONFIG_PAX_PAGEEXEC + if (!(mm->pax_flags & MF_PAX_PAGEEXEC) || address != regs->pc) + goto bad_area; + + up_read(&mm->mmap_sem); + switch (pax_handle_fetch_fault(regs)) { + +#ifdef CONFIG_PAX_EMUPLT + case 2: + case 3: + return; +#endif + + } + pax_report_fault(regs, (void *)regs->pc, (void *)rdusp()); + do_group_exit(SIGKILL); +#else goto bad_area; +#endif + + } } else if (!cause) { /* Allow reads even for write-only mappings */ if (!(vma->vm_flags & (VM_READ | VM_WRITE))) diff -urNp linux-2.6.32.48/arch/arm/include/asm/elf.h linux-2.6.32.48-openvz/arch/arm/include/asm/elf.h --- linux-2.6.32.48/arch/arm/include/asm/elf.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/arm/include/asm/elf.h 2011-11-17 20:08:53.000000000 -0500 @@ -109,7 +109,14 @@ int dump_task_regs(struct task_struct *t the loader. We need to make sure that it is out of the way of the program that it will "exec", and that there is sufficient room for the brk. */ -#define ELF_ET_DYN_BASE (2 * TASK_SIZE / 3) +#define ELF_ET_DYN_BASE (TASK_SIZE / 3 * 2) + +#ifdef CONFIG_PAX_ASLR +#define PAX_ELF_ET_DYN_BASE 0x00008000UL + +#define PAX_DELTA_MMAP_LEN ((current->personality == PER_LINUX_32BIT) ? 16 : 10) +#define PAX_DELTA_STACK_LEN ((current->personality == PER_LINUX_32BIT) ? 16 : 10) +#endif /* When the program starts, a1 contains a pointer to a function to be registered with atexit, as per the SVR4 ABI. A value of 0 means we diff -urNp linux-2.6.32.48/arch/arm/include/asm/kmap_types.h linux-2.6.32.48-openvz/arch/arm/include/asm/kmap_types.h --- linux-2.6.32.48/arch/arm/include/asm/kmap_types.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/arm/include/asm/kmap_types.h 2011-11-17 20:08:53.000000000 -0500 @@ -19,6 +19,7 @@ enum km_type { KM_SOFTIRQ0, KM_SOFTIRQ1, KM_L2_CACHE, + KM_CLEARPAGE, KM_TYPE_NR }; diff -urNp linux-2.6.32.48/arch/arm/include/asm/uaccess.h linux-2.6.32.48-openvz/arch/arm/include/asm/uaccess.h --- linux-2.6.32.48/arch/arm/include/asm/uaccess.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/arm/include/asm/uaccess.h 2011-11-17 20:09:04.000000000 -0500 @@ -22,6 +22,8 @@ #define VERIFY_READ 0 #define VERIFY_WRITE 1 +extern void check_object_size(const void *ptr, unsigned long n, bool to); + /* * The exception table consists of pairs of addresses: the first is the * address of an instruction that is allowed to fault, and the second is @@ -387,8 +389,23 @@ do { \ #ifdef CONFIG_MMU -extern unsigned long __must_check __copy_from_user(void *to, const void __user *from, unsigned long n); -extern unsigned long __must_check __copy_to_user(void __user *to, const void *from, unsigned long n); +extern unsigned long __must_check ___copy_from_user(void *to, const void __user *from, unsigned long n); +extern unsigned long __must_check ___copy_to_user(void __user *to, const void *from, unsigned long n); + +static inline unsigned long __must_check __copy_from_user(void *to, const void __user *from, unsigned long n) +{ + if (!__builtin_constant_p(n)) + check_object_size(to, n, false); + return ___copy_from_user(to, from, n); +} + +static inline unsigned long __must_check __copy_to_user(void __user *to, const void *from, unsigned long n) +{ + if (!__builtin_constant_p(n)) + check_object_size(from, n, true); + return ___copy_to_user(to, from, n); +} + extern unsigned long __must_check __copy_to_user_std(void __user *to, const void *from, unsigned long n); extern unsigned long __must_check __clear_user(void __user *addr, unsigned long n); extern unsigned long __must_check __clear_user_std(void __user *addr, unsigned long n); @@ -403,6 +420,9 @@ extern unsigned long __must_check __strn static inline unsigned long __must_check copy_from_user(void *to, const void __user *from, unsigned long n) { + if ((long)n < 0) + return n; + if (access_ok(VERIFY_READ, from, n)) n = __copy_from_user(to, from, n); else /* security hole - plug it */ @@ -412,6 +432,9 @@ static inline unsigned long __must_check static inline unsigned long __must_check copy_to_user(void __user *to, const void *from, unsigned long n) { + if ((long)n < 0) + return n; + if (access_ok(VERIFY_WRITE, to, n)) n = __copy_to_user(to, from, n); return n; diff -urNp linux-2.6.32.48/arch/arm/kernel/armksyms.c linux-2.6.32.48-openvz/arch/arm/kernel/armksyms.c --- linux-2.6.32.48/arch/arm/kernel/armksyms.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/arm/kernel/armksyms.c 2011-11-17 20:09:04.000000000 -0500 @@ -118,8 +118,8 @@ EXPORT_SYMBOL(__strncpy_from_user); #ifdef CONFIG_MMU EXPORT_SYMBOL(copy_page); -EXPORT_SYMBOL(__copy_from_user); -EXPORT_SYMBOL(__copy_to_user); +EXPORT_SYMBOL(___copy_from_user); +EXPORT_SYMBOL(___copy_to_user); EXPORT_SYMBOL(__clear_user); EXPORT_SYMBOL(__get_user_1); diff -urNp linux-2.6.32.48/arch/arm/kernel/kgdb.c linux-2.6.32.48-openvz/arch/arm/kernel/kgdb.c --- linux-2.6.32.48/arch/arm/kernel/kgdb.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/arm/kernel/kgdb.c 2011-11-17 20:08:53.000000000 -0500 @@ -190,7 +190,7 @@ void kgdb_arch_exit(void) * and we handle the normal undef case within the do_undefinstr * handler. */ -struct kgdb_arch arch_kgdb_ops = { +const struct kgdb_arch arch_kgdb_ops = { #ifndef __ARMEB__ .gdb_bpt_instr = {0xfe, 0xde, 0xff, 0xe7} #else /* ! __ARMEB__ */ diff -urNp linux-2.6.32.48/arch/arm/kernel/traps.c linux-2.6.32.48-openvz/arch/arm/kernel/traps.c --- linux-2.6.32.48/arch/arm/kernel/traps.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/arm/kernel/traps.c 2011-11-17 20:09:04.000000000 -0500 @@ -247,6 +247,8 @@ static void __die(const char *str, int e DEFINE_SPINLOCK(die_lock); +extern void gr_handle_kernel_exploit(void); + /* * This function is protected against re-entrancy. */ @@ -271,6 +273,8 @@ NORET_TYPE void die(const char *str, str if (panic_on_oops) panic("Fatal exception"); + gr_handle_kernel_exploit(); + do_exit(SIGSEGV); } diff -urNp linux-2.6.32.48/arch/arm/lib/copy_from_user.S linux-2.6.32.48-openvz/arch/arm/lib/copy_from_user.S --- linux-2.6.32.48/arch/arm/lib/copy_from_user.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/arm/lib/copy_from_user.S 2011-11-17 20:09:04.000000000 -0500 @@ -16,7 +16,7 @@ /* * Prototype: * - * size_t __copy_from_user(void *to, const void *from, size_t n) + * size_t ___copy_from_user(void *to, const void *from, size_t n) * * Purpose: * @@ -84,11 +84,11 @@ .text -ENTRY(__copy_from_user) +ENTRY(___copy_from_user) #include "copy_template.S" -ENDPROC(__copy_from_user) +ENDPROC(___copy_from_user) .section .fixup,"ax" .align 0 diff -urNp linux-2.6.32.48/arch/arm/lib/copy_to_user.S linux-2.6.32.48-openvz/arch/arm/lib/copy_to_user.S --- linux-2.6.32.48/arch/arm/lib/copy_to_user.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/arm/lib/copy_to_user.S 2011-11-17 20:09:04.000000000 -0500 @@ -16,7 +16,7 @@ /* * Prototype: * - * size_t __copy_to_user(void *to, const void *from, size_t n) + * size_t ___copy_to_user(void *to, const void *from, size_t n) * * Purpose: * @@ -88,11 +88,11 @@ .text ENTRY(__copy_to_user_std) -WEAK(__copy_to_user) +WEAK(___copy_to_user) #include "copy_template.S" -ENDPROC(__copy_to_user) +ENDPROC(___copy_to_user) .section .fixup,"ax" .align 0 diff -urNp linux-2.6.32.48/arch/arm/lib/uaccess.S linux-2.6.32.48-openvz/arch/arm/lib/uaccess.S --- linux-2.6.32.48/arch/arm/lib/uaccess.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/arm/lib/uaccess.S 2011-11-17 20:09:04.000000000 -0500 @@ -19,7 +19,7 @@ #define PAGE_SHIFT 12 -/* Prototype: int __copy_to_user(void *to, const char *from, size_t n) +/* Prototype: int ___copy_to_user(void *to, const char *from, size_t n) * Purpose : copy a block to user memory from kernel memory * Params : to - user memory * : from - kernel memory @@ -39,7 +39,7 @@ USER( strgtbt r3, [r0], #1) @ May fau sub r2, r2, ip b .Lc2u_dest_aligned -ENTRY(__copy_to_user) +ENTRY(___copy_to_user) stmfd sp!, {r2, r4 - r7, lr} cmp r2, #4 blt .Lc2u_not_enough @@ -277,14 +277,14 @@ USER( strgebt r3, [r0], #1) @ May fau ldrgtb r3, [r1], #0 USER( strgtbt r3, [r0], #1) @ May fault b .Lc2u_finished -ENDPROC(__copy_to_user) +ENDPROC(___copy_to_user) .section .fixup,"ax" .align 0 9001: ldmfd sp!, {r0, r4 - r7, pc} .previous -/* Prototype: unsigned long __copy_from_user(void *to,const void *from,unsigned long n); +/* Prototype: unsigned long ___copy_from_user(void *to,const void *from,unsigned long n); * Purpose : copy a block from user memory to kernel memory * Params : to - kernel memory * : from - user memory @@ -303,7 +303,7 @@ USER( ldrgtbt r3, [r1], #1) @ May fau sub r2, r2, ip b .Lcfu_dest_aligned -ENTRY(__copy_from_user) +ENTRY(___copy_from_user) stmfd sp!, {r0, r2, r4 - r7, lr} cmp r2, #4 blt .Lcfu_not_enough @@ -543,7 +543,7 @@ USER( ldrgebt r3, [r1], #1) @ May fau USER( ldrgtbt r3, [r1], #1) @ May fault strgtb r3, [r0], #1 b .Lcfu_finished -ENDPROC(__copy_from_user) +ENDPROC(___copy_from_user) .section .fixup,"ax" .align 0 diff -urNp linux-2.6.32.48/arch/arm/lib/uaccess_with_memcpy.c linux-2.6.32.48-openvz/arch/arm/lib/uaccess_with_memcpy.c --- linux-2.6.32.48/arch/arm/lib/uaccess_with_memcpy.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/arm/lib/uaccess_with_memcpy.c 2011-11-17 20:09:04.000000000 -0500 @@ -97,7 +97,7 @@ out: } unsigned long -__copy_to_user(void __user *to, const void *from, unsigned long n) +___copy_to_user(void __user *to, const void *from, unsigned long n) { /* * This test is stubbed out of the main function above to keep diff -urNp linux-2.6.32.48/arch/arm/mach-at91/pm.c linux-2.6.32.48-openvz/arch/arm/mach-at91/pm.c --- linux-2.6.32.48/arch/arm/mach-at91/pm.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/arm/mach-at91/pm.c 2011-11-17 20:08:53.000000000 -0500 @@ -348,7 +348,7 @@ static void at91_pm_end(void) } -static struct platform_suspend_ops at91_pm_ops ={ +static const struct platform_suspend_ops at91_pm_ops ={ .valid = at91_pm_valid_state, .begin = at91_pm_begin, .enter = at91_pm_enter, diff -urNp linux-2.6.32.48/arch/arm/mach-omap1/pm.c linux-2.6.32.48-openvz/arch/arm/mach-omap1/pm.c --- linux-2.6.32.48/arch/arm/mach-omap1/pm.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/arm/mach-omap1/pm.c 2011-11-17 20:08:53.000000000 -0500 @@ -647,7 +647,7 @@ static struct irqaction omap_wakeup_irq -static struct platform_suspend_ops omap_pm_ops ={ +static const struct platform_suspend_ops omap_pm_ops ={ .prepare = omap_pm_prepare, .enter = omap_pm_enter, .finish = omap_pm_finish, diff -urNp linux-2.6.32.48/arch/arm/mach-omap2/pm24xx.c linux-2.6.32.48-openvz/arch/arm/mach-omap2/pm24xx.c --- linux-2.6.32.48/arch/arm/mach-omap2/pm24xx.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/arm/mach-omap2/pm24xx.c 2011-11-17 20:08:53.000000000 -0500 @@ -326,7 +326,7 @@ static void omap2_pm_finish(void) enable_hlt(); } -static struct platform_suspend_ops omap_pm_ops = { +static const struct platform_suspend_ops omap_pm_ops = { .prepare = omap2_pm_prepare, .enter = omap2_pm_enter, .finish = omap2_pm_finish, diff -urNp linux-2.6.32.48/arch/arm/mach-omap2/pm34xx.c linux-2.6.32.48-openvz/arch/arm/mach-omap2/pm34xx.c --- linux-2.6.32.48/arch/arm/mach-omap2/pm34xx.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/arm/mach-omap2/pm34xx.c 2011-11-17 20:08:53.000000000 -0500 @@ -401,7 +401,7 @@ static void omap3_pm_end(void) return; } -static struct platform_suspend_ops omap_pm_ops = { +static const struct platform_suspend_ops omap_pm_ops = { .begin = omap3_pm_begin, .end = omap3_pm_end, .prepare = omap3_pm_prepare, diff -urNp linux-2.6.32.48/arch/arm/mach-pnx4008/pm.c linux-2.6.32.48-openvz/arch/arm/mach-pnx4008/pm.c --- linux-2.6.32.48/arch/arm/mach-pnx4008/pm.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/arm/mach-pnx4008/pm.c 2011-11-17 20:08:53.000000000 -0500 @@ -116,7 +116,7 @@ static int pnx4008_pm_valid(suspend_stat (state == PM_SUSPEND_MEM); } -static struct platform_suspend_ops pnx4008_pm_ops = { +static const struct platform_suspend_ops pnx4008_pm_ops = { .enter = pnx4008_pm_enter, .valid = pnx4008_pm_valid, }; diff -urNp linux-2.6.32.48/arch/arm/mach-pxa/pm.c linux-2.6.32.48-openvz/arch/arm/mach-pxa/pm.c --- linux-2.6.32.48/arch/arm/mach-pxa/pm.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/arm/mach-pxa/pm.c 2011-11-17 20:08:53.000000000 -0500 @@ -95,7 +95,7 @@ void pxa_pm_finish(void) pxa_cpu_pm_fns->finish(); } -static struct platform_suspend_ops pxa_pm_ops = { +static const struct platform_suspend_ops pxa_pm_ops = { .valid = pxa_pm_valid, .enter = pxa_pm_enter, .prepare = pxa_pm_prepare, diff -urNp linux-2.6.32.48/arch/arm/mach-pxa/sharpsl_pm.c linux-2.6.32.48-openvz/arch/arm/mach-pxa/sharpsl_pm.c --- linux-2.6.32.48/arch/arm/mach-pxa/sharpsl_pm.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/arm/mach-pxa/sharpsl_pm.c 2011-11-17 20:08:53.000000000 -0500 @@ -891,7 +891,7 @@ static void sharpsl_apm_get_power_status } #ifdef CONFIG_PM -static struct platform_suspend_ops sharpsl_pm_ops = { +static const struct platform_suspend_ops sharpsl_pm_ops = { .prepare = pxa_pm_prepare, .finish = pxa_pm_finish, .enter = corgi_pxa_pm_enter, diff -urNp linux-2.6.32.48/arch/arm/mach-sa1100/pm.c linux-2.6.32.48-openvz/arch/arm/mach-sa1100/pm.c --- linux-2.6.32.48/arch/arm/mach-sa1100/pm.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/arm/mach-sa1100/pm.c 2011-11-17 20:08:53.000000000 -0500 @@ -120,7 +120,7 @@ unsigned long sleep_phys_sp(void *sp) return virt_to_phys(sp); } -static struct platform_suspend_ops sa11x0_pm_ops = { +static const struct platform_suspend_ops sa11x0_pm_ops = { .enter = sa11x0_pm_enter, .valid = suspend_valid_only_mem, }; diff -urNp linux-2.6.32.48/arch/arm/mm/fault.c linux-2.6.32.48-openvz/arch/arm/mm/fault.c --- linux-2.6.32.48/arch/arm/mm/fault.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/arm/mm/fault.c 2011-11-17 20:08:53.000000000 -0500 @@ -166,6 +166,13 @@ __do_user_fault(struct task_struct *tsk, } #endif +#ifdef CONFIG_PAX_PAGEEXEC + if (fsr & FSR_LNX_PF) { + pax_report_fault(regs, (void *)regs->ARM_pc, (void *)regs->ARM_sp); + do_group_exit(SIGKILL); + } +#endif + tsk->thread.address = addr; tsk->thread.error_code = fsr; tsk->thread.trap_no = 14; @@ -357,6 +364,33 @@ do_page_fault(unsigned long addr, unsign } #endif /* CONFIG_MMU */ +#ifdef CONFIG_PAX_PAGEEXEC +void pax_report_insns(void *pc, void *sp) +{ + long i; + + printk(KERN_ERR "PAX: bytes at PC: "); + for (i = 0; i < 20; i++) { + unsigned char c; + if (get_user(c, (__force unsigned char __user *)pc+i)) + printk(KERN_CONT "?? "); + else + printk(KERN_CONT "%02x ", c); + } + printk("\n"); + + printk(KERN_ERR "PAX: bytes at SP-4: "); + for (i = -1; i < 20; i++) { + unsigned long c; + if (get_user(c, (__force unsigned long __user *)sp+i)) + printk(KERN_CONT "???????? "); + else + printk(KERN_CONT "%08lx ", c); + } + printk("\n"); +} +#endif + /* * First Level Translation Fault Handler * diff -urNp linux-2.6.32.48/arch/arm/mm/mmap.c linux-2.6.32.48-openvz/arch/arm/mm/mmap.c --- linux-2.6.32.48/arch/arm/mm/mmap.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/arm/mm/mmap.c 2011-11-17 20:08:53.000000000 -0500 @@ -63,6 +63,10 @@ arch_get_unmapped_area(struct file *filp if (len > TASK_SIZE) return -ENOMEM; +#ifdef CONFIG_PAX_RANDMMAP + if (!(mm->pax_flags & MF_PAX_RANDMMAP)) +#endif + if (addr) { if (do_align) addr = COLOUR_ALIGN(addr, pgoff); @@ -70,15 +74,14 @@ arch_get_unmapped_area(struct file *filp addr = PAGE_ALIGN(addr); vma = find_vma(mm, addr); - if (TASK_SIZE - len >= addr && - (!vma || addr + len <= vma->vm_start)) + if (TASK_SIZE - len >= addr && check_heap_stack_gap(vma, addr, len)) return addr; } if (len > mm->cached_hole_size) { - start_addr = addr = mm->free_area_cache; + start_addr = addr = mm->free_area_cache; } else { - start_addr = addr = TASK_UNMAPPED_BASE; - mm->cached_hole_size = 0; + start_addr = addr = mm->mmap_base; + mm->cached_hole_size = 0; } full_search: @@ -94,14 +97,14 @@ full_search: * Start a new search - just in case we missed * some holes. */ - if (start_addr != TASK_UNMAPPED_BASE) { - start_addr = addr = TASK_UNMAPPED_BASE; + if (start_addr != mm->mmap_base) { + start_addr = addr = mm->mmap_base; mm->cached_hole_size = 0; goto full_search; } return -ENOMEM; } - if (!vma || addr + len <= vma->vm_start) { + if (check_heap_stack_gap(vma, addr, len)) { /* * Remember the place where we stopped the search: */ diff -urNp linux-2.6.32.48/arch/arm/plat-s3c/pm.c linux-2.6.32.48-openvz/arch/arm/plat-s3c/pm.c --- linux-2.6.32.48/arch/arm/plat-s3c/pm.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/arm/plat-s3c/pm.c 2011-11-17 20:08:53.000000000 -0500 @@ -355,7 +355,7 @@ static void s3c_pm_finish(void) s3c_pm_check_cleanup(); } -static struct platform_suspend_ops s3c_pm_ops = { +static const struct platform_suspend_ops s3c_pm_ops = { .enter = s3c_pm_enter, .prepare = s3c_pm_prepare, .finish = s3c_pm_finish, diff -urNp linux-2.6.32.48/arch/avr32/include/asm/elf.h linux-2.6.32.48-openvz/arch/avr32/include/asm/elf.h --- linux-2.6.32.48/arch/avr32/include/asm/elf.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/avr32/include/asm/elf.h 2011-11-17 20:08:53.000000000 -0500 @@ -85,8 +85,14 @@ typedef struct user_fpu_struct elf_fpreg the loader. We need to make sure that it is out of the way of the program that it will "exec", and that there is sufficient room for the brk. */ -#define ELF_ET_DYN_BASE (2 * TASK_SIZE / 3) +#define ELF_ET_DYN_BASE (TASK_SIZE / 3 * 2) +#ifdef CONFIG_PAX_ASLR +#define PAX_ELF_ET_DYN_BASE 0x00001000UL + +#define PAX_DELTA_MMAP_LEN 15 +#define PAX_DELTA_STACK_LEN 15 +#endif /* This yields a mask that user programs can use to figure out what instruction set this CPU supports. This could be done in user space, diff -urNp linux-2.6.32.48/arch/avr32/include/asm/kmap_types.h linux-2.6.32.48-openvz/arch/avr32/include/asm/kmap_types.h --- linux-2.6.32.48/arch/avr32/include/asm/kmap_types.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/avr32/include/asm/kmap_types.h 2011-11-17 20:08:53.000000000 -0500 @@ -22,7 +22,8 @@ D(10) KM_IRQ0, D(11) KM_IRQ1, D(12) KM_SOFTIRQ0, D(13) KM_SOFTIRQ1, -D(14) KM_TYPE_NR +D(14) KM_CLEARPAGE, +D(15) KM_TYPE_NR }; #undef D diff -urNp linux-2.6.32.48/arch/avr32/mach-at32ap/pm.c linux-2.6.32.48-openvz/arch/avr32/mach-at32ap/pm.c --- linux-2.6.32.48/arch/avr32/mach-at32ap/pm.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/avr32/mach-at32ap/pm.c 2011-11-17 20:08:53.000000000 -0500 @@ -176,7 +176,7 @@ out: return 0; } -static struct platform_suspend_ops avr32_pm_ops = { +static const struct platform_suspend_ops avr32_pm_ops = { .valid = avr32_pm_valid_state, .enter = avr32_pm_enter, }; diff -urNp linux-2.6.32.48/arch/avr32/mm/fault.c linux-2.6.32.48-openvz/arch/avr32/mm/fault.c --- linux-2.6.32.48/arch/avr32/mm/fault.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/avr32/mm/fault.c 2011-11-17 20:08:53.000000000 -0500 @@ -41,6 +41,23 @@ static inline int notify_page_fault(stru int exception_trace = 1; +#ifdef CONFIG_PAX_PAGEEXEC +void pax_report_insns(void *pc, void *sp) +{ + unsigned long i; + + printk(KERN_ERR "PAX: bytes at PC: "); + for (i = 0; i < 20; i++) { + unsigned char c; + if (get_user(c, (unsigned char *)pc+i)) + printk(KERN_CONT "???????? "); + else + printk(KERN_CONT "%02x ", c); + } + printk("\n"); +} +#endif + /* * This routine handles page faults. It determines the address and the * problem, and then passes it off to one of the appropriate routines. @@ -157,6 +174,16 @@ bad_area: up_read(&mm->mmap_sem); if (user_mode(regs)) { + +#ifdef CONFIG_PAX_PAGEEXEC + if (mm->pax_flags & MF_PAX_PAGEEXEC) { + if (ecr == ECR_PROTECTION_X || ecr == ECR_TLB_MISS_X) { + pax_report_fault(regs, (void *)regs->pc, (void *)regs->sp); + do_group_exit(SIGKILL); + } + } +#endif + if (exception_trace && printk_ratelimit()) printk("%s%s[%d]: segfault at %08lx pc %08lx " "sp %08lx ecr %lu\n", diff -urNp linux-2.6.32.48/arch/blackfin/kernel/kgdb.c linux-2.6.32.48-openvz/arch/blackfin/kernel/kgdb.c --- linux-2.6.32.48/arch/blackfin/kernel/kgdb.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/blackfin/kernel/kgdb.c 2011-11-17 20:08:53.000000000 -0500 @@ -428,7 +428,7 @@ int kgdb_arch_handle_exception(int vecto return -1; /* this means that we do not want to exit from the handler */ } -struct kgdb_arch arch_kgdb_ops = { +const struct kgdb_arch arch_kgdb_ops = { .gdb_bpt_instr = {0xa1}, #ifdef CONFIG_SMP .flags = KGDB_HW_BREAKPOINT|KGDB_THR_PROC_SWAP, diff -urNp linux-2.6.32.48/arch/blackfin/mach-common/pm.c linux-2.6.32.48-openvz/arch/blackfin/mach-common/pm.c --- linux-2.6.32.48/arch/blackfin/mach-common/pm.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/blackfin/mach-common/pm.c 2011-11-17 20:08:53.000000000 -0500 @@ -255,7 +255,7 @@ static int bfin_pm_enter(suspend_state_t return 0; } -struct platform_suspend_ops bfin_pm_ops = { +const struct platform_suspend_ops bfin_pm_ops = { .enter = bfin_pm_enter, .valid = bfin_pm_valid, }; diff -urNp linux-2.6.32.48/arch/frv/include/asm/kmap_types.h linux-2.6.32.48-openvz/arch/frv/include/asm/kmap_types.h --- linux-2.6.32.48/arch/frv/include/asm/kmap_types.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/frv/include/asm/kmap_types.h 2011-11-17 20:08:53.000000000 -0500 @@ -23,6 +23,7 @@ enum km_type { KM_IRQ1, KM_SOFTIRQ0, KM_SOFTIRQ1, + KM_CLEARPAGE, KM_TYPE_NR }; diff -urNp linux-2.6.32.48/arch/frv/mm/elf-fdpic.c linux-2.6.32.48-openvz/arch/frv/mm/elf-fdpic.c --- linux-2.6.32.48/arch/frv/mm/elf-fdpic.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/frv/mm/elf-fdpic.c 2011-11-17 20:08:53.000000000 -0500 @@ -73,8 +73,7 @@ unsigned long arch_get_unmapped_area(str if (addr) { addr = PAGE_ALIGN(addr); vma = find_vma(current->mm, addr); - if (TASK_SIZE - len >= addr && - (!vma || addr + len <= vma->vm_start)) + if (TASK_SIZE - len >= addr && check_heap_stack_gap(vma, addr, len)) goto success; } @@ -89,7 +88,7 @@ unsigned long arch_get_unmapped_area(str for (; vma; vma = vma->vm_next) { if (addr > limit) break; - if (addr + len <= vma->vm_start) + if (check_heap_stack_gap(vma, addr, len)) goto success; addr = vma->vm_end; } @@ -104,7 +103,7 @@ unsigned long arch_get_unmapped_area(str for (; vma; vma = vma->vm_next) { if (addr > limit) break; - if (addr + len <= vma->vm_start) + if (check_heap_stack_gap(vma, addr, len)) goto success; addr = vma->vm_end; } diff -urNp linux-2.6.32.48/arch/ia64/hp/common/hwsw_iommu.c linux-2.6.32.48-openvz/arch/ia64/hp/common/hwsw_iommu.c --- linux-2.6.32.48/arch/ia64/hp/common/hwsw_iommu.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/ia64/hp/common/hwsw_iommu.c 2011-11-17 20:08:53.000000000 -0500 @@ -17,7 +17,7 @@ #include #include -extern struct dma_map_ops sba_dma_ops, swiotlb_dma_ops; +extern const struct dma_map_ops sba_dma_ops, swiotlb_dma_ops; /* swiotlb declarations & definitions: */ extern int swiotlb_late_init_with_default_size (size_t size); @@ -33,7 +33,7 @@ static inline int use_swiotlb(struct dev !sba_dma_ops.dma_supported(dev, *dev->dma_mask); } -struct dma_map_ops *hwsw_dma_get_ops(struct device *dev) +const struct dma_map_ops *hwsw_dma_get_ops(struct device *dev) { if (use_swiotlb(dev)) return &swiotlb_dma_ops; diff -urNp linux-2.6.32.48/arch/ia64/hp/common/sba_iommu.c linux-2.6.32.48-openvz/arch/ia64/hp/common/sba_iommu.c --- linux-2.6.32.48/arch/ia64/hp/common/sba_iommu.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/ia64/hp/common/sba_iommu.c 2011-11-17 20:08:53.000000000 -0500 @@ -2097,7 +2097,7 @@ static struct acpi_driver acpi_sba_ioc_d }, }; -extern struct dma_map_ops swiotlb_dma_ops; +extern const struct dma_map_ops swiotlb_dma_ops; static int __init sba_init(void) @@ -2211,7 +2211,7 @@ sba_page_override(char *str) __setup("sbapagesize=",sba_page_override); -struct dma_map_ops sba_dma_ops = { +const struct dma_map_ops sba_dma_ops = { .alloc_coherent = sba_alloc_coherent, .free_coherent = sba_free_coherent, .map_page = sba_map_page, diff -urNp linux-2.6.32.48/arch/ia64/ia32/binfmt_elf32.c linux-2.6.32.48-openvz/arch/ia64/ia32/binfmt_elf32.c --- linux-2.6.32.48/arch/ia64/ia32/binfmt_elf32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/ia64/ia32/binfmt_elf32.c 2011-11-17 20:08:53.000000000 -0500 @@ -45,6 +45,13 @@ randomize_stack_top(unsigned long stack_ #define elf_read_implies_exec(ex, have_pt_gnu_stack) (!(have_pt_gnu_stack)) +#ifdef CONFIG_PAX_ASLR +#define PAX_ELF_ET_DYN_BASE (current->personality == PER_LINUX32 ? 0x08048000UL : 0x4000000000000000UL) + +#define PAX_DELTA_MMAP_LEN (current->personality == PER_LINUX32 ? 16 : 3*PAGE_SHIFT - 13) +#define PAX_DELTA_STACK_LEN (current->personality == PER_LINUX32 ? 16 : 3*PAGE_SHIFT - 13) +#endif + /* Ugly but avoids duplication */ #include "../../../fs/binfmt_elf.c" diff -urNp linux-2.6.32.48/arch/ia64/ia32/ia32priv.h linux-2.6.32.48-openvz/arch/ia64/ia32/ia32priv.h --- linux-2.6.32.48/arch/ia64/ia32/ia32priv.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/ia64/ia32/ia32priv.h 2011-11-17 20:08:53.000000000 -0500 @@ -296,7 +296,14 @@ typedef struct compat_siginfo { #define ELF_DATA ELFDATA2LSB #define ELF_ARCH EM_386 -#define IA32_STACK_TOP IA32_PAGE_OFFSET +#ifdef CONFIG_PAX_RANDUSTACK +#define __IA32_DELTA_STACK (current->mm->delta_stack) +#else +#define __IA32_DELTA_STACK 0UL +#endif + +#define IA32_STACK_TOP (IA32_PAGE_OFFSET - __IA32_DELTA_STACK) + #define IA32_GATE_OFFSET IA32_PAGE_OFFSET #define IA32_GATE_END IA32_PAGE_OFFSET + PAGE_SIZE diff -urNp linux-2.6.32.48/arch/ia64/include/asm/dma-mapping.h linux-2.6.32.48-openvz/arch/ia64/include/asm/dma-mapping.h --- linux-2.6.32.48/arch/ia64/include/asm/dma-mapping.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/ia64/include/asm/dma-mapping.h 2011-11-17 20:08:53.000000000 -0500 @@ -12,7 +12,7 @@ #define ARCH_HAS_DMA_GET_REQUIRED_MASK -extern struct dma_map_ops *dma_ops; +extern const struct dma_map_ops *dma_ops; extern struct ia64_machine_vector ia64_mv; extern void set_iommu_machvec(void); @@ -24,7 +24,7 @@ extern void machvec_dma_sync_sg(struct d static inline void *dma_alloc_coherent(struct device *dev, size_t size, dma_addr_t *daddr, gfp_t gfp) { - struct dma_map_ops *ops = platform_dma_get_ops(dev); + const struct dma_map_ops *ops = platform_dma_get_ops(dev); void *caddr; caddr = ops->alloc_coherent(dev, size, daddr, gfp); @@ -35,7 +35,7 @@ static inline void *dma_alloc_coherent(s static inline void dma_free_coherent(struct device *dev, size_t size, void *caddr, dma_addr_t daddr) { - struct dma_map_ops *ops = platform_dma_get_ops(dev); + const struct dma_map_ops *ops = platform_dma_get_ops(dev); debug_dma_free_coherent(dev, size, caddr, daddr); ops->free_coherent(dev, size, caddr, daddr); } @@ -49,13 +49,13 @@ static inline void dma_free_coherent(str static inline int dma_mapping_error(struct device *dev, dma_addr_t daddr) { - struct dma_map_ops *ops = platform_dma_get_ops(dev); + const struct dma_map_ops *ops = platform_dma_get_ops(dev); return ops->mapping_error(dev, daddr); } static inline int dma_supported(struct device *dev, u64 mask) { - struct dma_map_ops *ops = platform_dma_get_ops(dev); + const struct dma_map_ops *ops = platform_dma_get_ops(dev); return ops->dma_supported(dev, mask); } diff -urNp linux-2.6.32.48/arch/ia64/include/asm/elf.h linux-2.6.32.48-openvz/arch/ia64/include/asm/elf.h --- linux-2.6.32.48/arch/ia64/include/asm/elf.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/ia64/include/asm/elf.h 2011-11-17 20:08:53.000000000 -0500 @@ -43,6 +43,13 @@ */ #define ELF_ET_DYN_BASE (TASK_UNMAPPED_BASE + 0x800000000UL) +#ifdef CONFIG_PAX_ASLR +#define PAX_ELF_ET_DYN_BASE (current->personality == PER_LINUX32 ? 0x08048000UL : 0x4000000000000000UL) + +#define PAX_DELTA_MMAP_LEN (current->personality == PER_LINUX32 ? 16 : 3*PAGE_SHIFT - 13) +#define PAX_DELTA_STACK_LEN (current->personality == PER_LINUX32 ? 16 : 3*PAGE_SHIFT - 13) +#endif + #define PT_IA_64_UNWIND 0x70000001 /* IA-64 relocations: */ diff -urNp linux-2.6.32.48/arch/ia64/include/asm/machvec.h linux-2.6.32.48-openvz/arch/ia64/include/asm/machvec.h --- linux-2.6.32.48/arch/ia64/include/asm/machvec.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/ia64/include/asm/machvec.h 2011-11-17 20:08:53.000000000 -0500 @@ -45,7 +45,7 @@ typedef void ia64_mv_kernel_launch_event /* DMA-mapping interface: */ typedef void ia64_mv_dma_init (void); typedef u64 ia64_mv_dma_get_required_mask (struct device *); -typedef struct dma_map_ops *ia64_mv_dma_get_ops(struct device *); +typedef const struct dma_map_ops *ia64_mv_dma_get_ops(struct device *); /* * WARNING: The legacy I/O space is _architected_. Platforms are @@ -251,7 +251,7 @@ extern void machvec_init_from_cmdline(co # endif /* CONFIG_IA64_GENERIC */ extern void swiotlb_dma_init(void); -extern struct dma_map_ops *dma_get_ops(struct device *); +extern const struct dma_map_ops *dma_get_ops(struct device *); /* * Define default versions so we can extend machvec for new platforms without having diff -urNp linux-2.6.32.48/arch/ia64/include/asm/pgtable.h linux-2.6.32.48-openvz/arch/ia64/include/asm/pgtable.h --- linux-2.6.32.48/arch/ia64/include/asm/pgtable.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/ia64/include/asm/pgtable.h 2011-11-17 20:08:53.000000000 -0500 @@ -12,7 +12,7 @@ * David Mosberger-Tang */ - +#include #include #include #include @@ -143,6 +143,17 @@ #define PAGE_READONLY __pgprot(__ACCESS_BITS | _PAGE_PL_3 | _PAGE_AR_R) #define PAGE_COPY __pgprot(__ACCESS_BITS | _PAGE_PL_3 | _PAGE_AR_R) #define PAGE_COPY_EXEC __pgprot(__ACCESS_BITS | _PAGE_PL_3 | _PAGE_AR_RX) + +#ifdef CONFIG_PAX_PAGEEXEC +# define PAGE_SHARED_NOEXEC __pgprot(__ACCESS_BITS | _PAGE_PL_3 | _PAGE_AR_RW) +# define PAGE_READONLY_NOEXEC __pgprot(__ACCESS_BITS | _PAGE_PL_3 | _PAGE_AR_R) +# define PAGE_COPY_NOEXEC __pgprot(__ACCESS_BITS | _PAGE_PL_3 | _PAGE_AR_R) +#else +# define PAGE_SHARED_NOEXEC PAGE_SHARED +# define PAGE_READONLY_NOEXEC PAGE_READONLY +# define PAGE_COPY_NOEXEC PAGE_COPY +#endif + #define PAGE_GATE __pgprot(__ACCESS_BITS | _PAGE_PL_0 | _PAGE_AR_X_RX) #define PAGE_KERNEL __pgprot(__DIRTY_BITS | _PAGE_PL_0 | _PAGE_AR_RWX) #define PAGE_KERNELRX __pgprot(__ACCESS_BITS | _PAGE_PL_0 | _PAGE_AR_RX) diff -urNp linux-2.6.32.48/arch/ia64/include/asm/spinlock.h linux-2.6.32.48-openvz/arch/ia64/include/asm/spinlock.h --- linux-2.6.32.48/arch/ia64/include/asm/spinlock.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/ia64/include/asm/spinlock.h 2011-11-17 20:08:53.000000000 -0500 @@ -72,7 +72,7 @@ static __always_inline void __ticket_spi unsigned short *p = (unsigned short *)&lock->lock + 1, tmp; asm volatile ("ld2.bias %0=[%1]" : "=r"(tmp) : "r"(p)); - ACCESS_ONCE(*p) = (tmp + 2) & ~1; + ACCESS_ONCE_RW(*p) = (tmp + 2) & ~1; } static __always_inline void __ticket_spin_unlock_wait(raw_spinlock_t *lock) diff -urNp linux-2.6.32.48/arch/ia64/include/asm/uaccess.h linux-2.6.32.48-openvz/arch/ia64/include/asm/uaccess.h --- linux-2.6.32.48/arch/ia64/include/asm/uaccess.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/ia64/include/asm/uaccess.h 2011-11-17 20:08:53.000000000 -0500 @@ -257,7 +257,7 @@ __copy_from_user (void *to, const void _ const void *__cu_from = (from); \ long __cu_len = (n); \ \ - if (__access_ok(__cu_to, __cu_len, get_fs())) \ + if (__cu_len > 0 && __cu_len <= INT_MAX && __access_ok(__cu_to, __cu_len, get_fs())) \ __cu_len = __copy_user(__cu_to, (__force void __user *) __cu_from, __cu_len); \ __cu_len; \ }) @@ -269,7 +269,7 @@ __copy_from_user (void *to, const void _ long __cu_len = (n); \ \ __chk_user_ptr(__cu_from); \ - if (__access_ok(__cu_from, __cu_len, get_fs())) \ + if (__cu_len > 0 && __cu_len <= INT_MAX && __access_ok(__cu_from, __cu_len, get_fs())) \ __cu_len = __copy_user((__force void __user *) __cu_to, __cu_from, __cu_len); \ __cu_len; \ }) diff -urNp linux-2.6.32.48/arch/ia64/kernel/dma-mapping.c linux-2.6.32.48-openvz/arch/ia64/kernel/dma-mapping.c --- linux-2.6.32.48/arch/ia64/kernel/dma-mapping.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/ia64/kernel/dma-mapping.c 2011-11-17 20:08:53.000000000 -0500 @@ -3,7 +3,7 @@ /* Set this to 1 if there is a HW IOMMU in the system */ int iommu_detected __read_mostly; -struct dma_map_ops *dma_ops; +const struct dma_map_ops *dma_ops; EXPORT_SYMBOL(dma_ops); #define PREALLOC_DMA_DEBUG_ENTRIES (1 << 16) @@ -16,7 +16,7 @@ static int __init dma_init(void) } fs_initcall(dma_init); -struct dma_map_ops *dma_get_ops(struct device *dev) +const struct dma_map_ops *dma_get_ops(struct device *dev) { return dma_ops; } diff -urNp linux-2.6.32.48/arch/ia64/kernel/module.c linux-2.6.32.48-openvz/arch/ia64/kernel/module.c --- linux-2.6.32.48/arch/ia64/kernel/module.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/ia64/kernel/module.c 2011-11-17 20:08:53.000000000 -0500 @@ -315,8 +315,7 @@ module_alloc (unsigned long size) void module_free (struct module *mod, void *module_region) { - if (mod && mod->arch.init_unw_table && - module_region == mod->module_init) { + if (mod && mod->arch.init_unw_table && module_region == mod->module_init_rx) { unw_remove_unwind_table(mod->arch.init_unw_table); mod->arch.init_unw_table = NULL; } @@ -502,15 +501,39 @@ module_frob_arch_sections (Elf_Ehdr *ehd } static inline int +in_init_rx (const struct module *mod, uint64_t addr) +{ + return addr - (uint64_t) mod->module_init_rx < mod->init_size_rx; +} + +static inline int +in_init_rw (const struct module *mod, uint64_t addr) +{ + return addr - (uint64_t) mod->module_init_rw < mod->init_size_rw; +} + +static inline int in_init (const struct module *mod, uint64_t addr) { - return addr - (uint64_t) mod->module_init < mod->init_size; + return in_init_rx(mod, addr) || in_init_rw(mod, addr); +} + +static inline int +in_core_rx (const struct module *mod, uint64_t addr) +{ + return addr - (uint64_t) mod->module_core_rx < mod->core_size_rx; +} + +static inline int +in_core_rw (const struct module *mod, uint64_t addr) +{ + return addr - (uint64_t) mod->module_core_rw < mod->core_size_rw; } static inline int in_core (const struct module *mod, uint64_t addr) { - return addr - (uint64_t) mod->module_core < mod->core_size; + return in_core_rx(mod, addr) || in_core_rw(mod, addr); } static inline int @@ -693,7 +716,14 @@ do_reloc (struct module *mod, uint8_t r_ break; case RV_BDREL: - val -= (uint64_t) (in_init(mod, val) ? mod->module_init : mod->module_core); + if (in_init_rx(mod, val)) + val -= (uint64_t) mod->module_init_rx; + else if (in_init_rw(mod, val)) + val -= (uint64_t) mod->module_init_rw; + else if (in_core_rx(mod, val)) + val -= (uint64_t) mod->module_core_rx; + else if (in_core_rw(mod, val)) + val -= (uint64_t) mod->module_core_rw; break; case RV_LTV: @@ -828,15 +858,15 @@ apply_relocate_add (Elf64_Shdr *sechdrs, * addresses have been selected... */ uint64_t gp; - if (mod->core_size > MAX_LTOFF) + if (mod->core_size_rx + mod->core_size_rw > MAX_LTOFF) /* * This takes advantage of fact that SHF_ARCH_SMALL gets allocated * at the end of the module. */ - gp = mod->core_size - MAX_LTOFF / 2; + gp = mod->core_size_rx + mod->core_size_rw - MAX_LTOFF / 2; else - gp = mod->core_size / 2; - gp = (uint64_t) mod->module_core + ((gp + 7) & -8); + gp = (mod->core_size_rx + mod->core_size_rw) / 2; + gp = (uint64_t) mod->module_core_rx + ((gp + 7) & -8); mod->arch.gp = gp; DEBUGP("%s: placing gp at 0x%lx\n", __func__, gp); } diff -urNp linux-2.6.32.48/arch/ia64/kernel/pci-dma.c linux-2.6.32.48-openvz/arch/ia64/kernel/pci-dma.c --- linux-2.6.32.48/arch/ia64/kernel/pci-dma.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/ia64/kernel/pci-dma.c 2011-11-17 20:08:53.000000000 -0500 @@ -43,7 +43,7 @@ struct device fallback_dev = { .dma_mask = &fallback_dev.coherent_dma_mask, }; -extern struct dma_map_ops intel_dma_ops; +extern const struct dma_map_ops intel_dma_ops; static int __init pci_iommu_init(void) { @@ -96,15 +96,34 @@ int iommu_dma_supported(struct device *d } EXPORT_SYMBOL(iommu_dma_supported); +extern void *intel_alloc_coherent(struct device *hwdev, size_t size, dma_addr_t *dma_handle, gfp_t flags); +extern void intel_free_coherent(struct device *hwdev, size_t size, void *vaddr, dma_addr_t dma_handle); +extern int intel_map_sg(struct device *hwdev, struct scatterlist *sglist, int nelems, enum dma_data_direction dir, struct dma_attrs *attrs); +extern void intel_unmap_sg(struct device *hwdev, struct scatterlist *sglist, int nelems, enum dma_data_direction dir, struct dma_attrs *attrs); +extern dma_addr_t intel_map_page(struct device *dev, struct page *page, unsigned long offset, size_t size, enum dma_data_direction dir, struct dma_attrs *attrs); +extern void intel_unmap_page(struct device *dev, dma_addr_t dev_addr, size_t size, enum dma_data_direction dir, struct dma_attrs *attrs); +extern int intel_mapping_error(struct device *dev, dma_addr_t dma_addr); + +static const struct dma_map_ops intel_iommu_dma_ops = { + /* from drivers/pci/intel-iommu.c:intel_dma_ops */ + .alloc_coherent = intel_alloc_coherent, + .free_coherent = intel_free_coherent, + .map_sg = intel_map_sg, + .unmap_sg = intel_unmap_sg, + .map_page = intel_map_page, + .unmap_page = intel_unmap_page, + .mapping_error = intel_mapping_error, + + .sync_single_for_cpu = machvec_dma_sync_single, + .sync_sg_for_cpu = machvec_dma_sync_sg, + .sync_single_for_device = machvec_dma_sync_single, + .sync_sg_for_device = machvec_dma_sync_sg, + .dma_supported = iommu_dma_supported, +}; + void __init pci_iommu_alloc(void) { - dma_ops = &intel_dma_ops; - - dma_ops->sync_single_for_cpu = machvec_dma_sync_single; - dma_ops->sync_sg_for_cpu = machvec_dma_sync_sg; - dma_ops->sync_single_for_device = machvec_dma_sync_single; - dma_ops->sync_sg_for_device = machvec_dma_sync_sg; - dma_ops->dma_supported = iommu_dma_supported; + dma_ops = &intel_iommu_dma_ops; /* * The order of these functions is important for diff -urNp linux-2.6.32.48/arch/ia64/kernel/pci-swiotlb.c linux-2.6.32.48-openvz/arch/ia64/kernel/pci-swiotlb.c --- linux-2.6.32.48/arch/ia64/kernel/pci-swiotlb.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/ia64/kernel/pci-swiotlb.c 2011-11-17 20:08:53.000000000 -0500 @@ -21,7 +21,7 @@ static void *ia64_swiotlb_alloc_coherent return swiotlb_alloc_coherent(dev, size, dma_handle, gfp); } -struct dma_map_ops swiotlb_dma_ops = { +const struct dma_map_ops swiotlb_dma_ops = { .alloc_coherent = ia64_swiotlb_alloc_coherent, .free_coherent = swiotlb_free_coherent, .map_page = swiotlb_map_page, diff -urNp linux-2.6.32.48/arch/ia64/kernel/sys_ia64.c linux-2.6.32.48-openvz/arch/ia64/kernel/sys_ia64.c --- linux-2.6.32.48/arch/ia64/kernel/sys_ia64.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/ia64/kernel/sys_ia64.c 2011-11-17 20:08:53.000000000 -0500 @@ -43,6 +43,13 @@ arch_get_unmapped_area (struct file *fil if (REGION_NUMBER(addr) == RGN_HPAGE) addr = 0; #endif + +#ifdef CONFIG_PAX_RANDMMAP + if (mm->pax_flags & MF_PAX_RANDMMAP) + addr = mm->free_area_cache; + else +#endif + if (!addr) addr = mm->free_area_cache; @@ -61,14 +68,14 @@ arch_get_unmapped_area (struct file *fil for (vma = find_vma(mm, addr); ; vma = vma->vm_next) { /* At this point: (!vma || addr < vma->vm_end). */ if (TASK_SIZE - len < addr || RGN_MAP_LIMIT - len < REGION_OFFSET(addr)) { - if (start_addr != TASK_UNMAPPED_BASE) { + if (start_addr != mm->mmap_base) { /* Start a new search --- just in case we missed some holes. */ - addr = TASK_UNMAPPED_BASE; + addr = mm->mmap_base; goto full_search; } return -ENOMEM; } - if (!vma || addr + len <= vma->vm_start) { + if (check_heap_stack_gap(vma, addr, len)) { /* Remember the address where we stopped this search: */ mm->free_area_cache = addr + len; return addr; diff -urNp linux-2.6.32.48/arch/ia64/kernel/topology.c linux-2.6.32.48-openvz/arch/ia64/kernel/topology.c --- linux-2.6.32.48/arch/ia64/kernel/topology.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/ia64/kernel/topology.c 2011-11-17 20:08:53.000000000 -0500 @@ -282,7 +282,7 @@ static ssize_t cache_show(struct kobject return ret; } -static struct sysfs_ops cache_sysfs_ops = { +static const struct sysfs_ops cache_sysfs_ops = { .show = cache_show }; diff -urNp linux-2.6.32.48/arch/ia64/kernel/vmlinux.lds.S linux-2.6.32.48-openvz/arch/ia64/kernel/vmlinux.lds.S --- linux-2.6.32.48/arch/ia64/kernel/vmlinux.lds.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/ia64/kernel/vmlinux.lds.S 2011-11-17 20:08:53.000000000 -0500 @@ -190,7 +190,7 @@ SECTIONS /* Per-cpu data: */ . = ALIGN(PERCPU_PAGE_SIZE); PERCPU_VADDR(PERCPU_ADDR, :percpu) - __phys_per_cpu_start = __per_cpu_load; + __phys_per_cpu_start = per_cpu_load; . = __phys_per_cpu_start + PERCPU_PAGE_SIZE; /* ensure percpu data fits * into percpu page size */ diff -urNp linux-2.6.32.48/arch/ia64/mm/fault.c linux-2.6.32.48-openvz/arch/ia64/mm/fault.c --- linux-2.6.32.48/arch/ia64/mm/fault.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/ia64/mm/fault.c 2011-11-17 20:08:53.000000000 -0500 @@ -72,6 +72,23 @@ mapped_kernel_page_is_present (unsigned return pte_present(pte); } +#ifdef CONFIG_PAX_PAGEEXEC +void pax_report_insns(void *pc, void *sp) +{ + unsigned long i; + + printk(KERN_ERR "PAX: bytes at PC: "); + for (i = 0; i < 8; i++) { + unsigned int c; + if (get_user(c, (unsigned int *)pc+i)) + printk(KERN_CONT "???????? "); + else + printk(KERN_CONT "%08x ", c); + } + printk("\n"); +} +#endif + void __kprobes ia64_do_page_fault (unsigned long address, unsigned long isr, struct pt_regs *regs) { @@ -145,9 +162,23 @@ ia64_do_page_fault (unsigned long addres mask = ( (((isr >> IA64_ISR_X_BIT) & 1UL) << VM_EXEC_BIT) | (((isr >> IA64_ISR_W_BIT) & 1UL) << VM_WRITE_BIT)); - if ((vma->vm_flags & mask) != mask) + if ((vma->vm_flags & mask) != mask) { + +#ifdef CONFIG_PAX_PAGEEXEC + if (!(vma->vm_flags & VM_EXEC) && (mask & VM_EXEC)) { + if (!(mm->pax_flags & MF_PAX_PAGEEXEC) || address != regs->cr_iip) + goto bad_area; + + up_read(&mm->mmap_sem); + pax_report_fault(regs, (void *)regs->cr_iip, (void *)regs->r12); + do_group_exit(SIGKILL); + } +#endif + goto bad_area; + } + survive: /* * If for any reason at all we couldn't handle the fault, make diff -urNp linux-2.6.32.48/arch/ia64/mm/hugetlbpage.c linux-2.6.32.48-openvz/arch/ia64/mm/hugetlbpage.c --- linux-2.6.32.48/arch/ia64/mm/hugetlbpage.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/ia64/mm/hugetlbpage.c 2011-11-17 20:08:53.000000000 -0500 @@ -172,7 +172,7 @@ unsigned long hugetlb_get_unmapped_area( /* At this point: (!vmm || addr < vmm->vm_end). */ if (REGION_OFFSET(addr) + len > RGN_MAP_LIMIT) return -ENOMEM; - if (!vmm || (addr + len) <= vmm->vm_start) + if (check_heap_stack_gap(vmm, addr, len)) return addr; addr = ALIGN(vmm->vm_end, HPAGE_SIZE); } diff -urNp linux-2.6.32.48/arch/ia64/mm/init.c linux-2.6.32.48-openvz/arch/ia64/mm/init.c --- linux-2.6.32.48/arch/ia64/mm/init.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/ia64/mm/init.c 2011-11-17 20:08:53.000000000 -0500 @@ -122,6 +122,19 @@ ia64_init_addr_space (void) vma->vm_start = current->thread.rbs_bot & PAGE_MASK; vma->vm_end = vma->vm_start + PAGE_SIZE; vma->vm_flags = VM_DATA_DEFAULT_FLAGS|VM_GROWSUP|VM_ACCOUNT; + +#ifdef CONFIG_PAX_PAGEEXEC + if (current->mm->pax_flags & MF_PAX_PAGEEXEC) { + vma->vm_flags &= ~VM_EXEC; + +#ifdef CONFIG_PAX_MPROTECT + if (current->mm->pax_flags & MF_PAX_MPROTECT) + vma->vm_flags &= ~VM_MAYEXEC; +#endif + + } +#endif + vma->vm_page_prot = vm_get_page_prot(vma->vm_flags); down_write(¤t->mm->mmap_sem); if (insert_vm_struct(current->mm, vma)) { diff -urNp linux-2.6.32.48/arch/ia64/sn/pci/pci_dma.c linux-2.6.32.48-openvz/arch/ia64/sn/pci/pci_dma.c --- linux-2.6.32.48/arch/ia64/sn/pci/pci_dma.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/ia64/sn/pci/pci_dma.c 2011-11-17 20:08:53.000000000 -0500 @@ -464,7 +464,7 @@ int sn_pci_legacy_write(struct pci_bus * return ret; } -static struct dma_map_ops sn_dma_ops = { +static const struct dma_map_ops sn_dma_ops = { .alloc_coherent = sn_dma_alloc_coherent, .free_coherent = sn_dma_free_coherent, .map_page = sn_dma_map_page, diff -urNp linux-2.6.32.48/arch/m32r/lib/usercopy.c linux-2.6.32.48-openvz/arch/m32r/lib/usercopy.c --- linux-2.6.32.48/arch/m32r/lib/usercopy.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/m32r/lib/usercopy.c 2011-11-17 20:08:53.000000000 -0500 @@ -14,6 +14,9 @@ unsigned long __generic_copy_to_user(void __user *to, const void *from, unsigned long n) { + if ((long)n < 0) + return n; + prefetch(from); if (access_ok(VERIFY_WRITE, to, n)) __copy_user(to,from,n); @@ -23,6 +26,9 @@ __generic_copy_to_user(void __user *to, unsigned long __generic_copy_from_user(void *to, const void __user *from, unsigned long n) { + if ((long)n < 0) + return n; + prefetchw(to); if (access_ok(VERIFY_READ, from, n)) __copy_user_zeroing(to,from,n); diff -urNp linux-2.6.32.48/arch/mips/alchemy/devboards/pm.c linux-2.6.32.48-openvz/arch/mips/alchemy/devboards/pm.c --- linux-2.6.32.48/arch/mips/alchemy/devboards/pm.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/mips/alchemy/devboards/pm.c 2011-11-17 20:08:53.000000000 -0500 @@ -78,7 +78,7 @@ static void db1x_pm_end(void) } -static struct platform_suspend_ops db1x_pm_ops = { +static const struct platform_suspend_ops db1x_pm_ops = { .valid = suspend_valid_only_mem, .begin = db1x_pm_begin, .enter = db1x_pm_enter, diff -urNp linux-2.6.32.48/arch/mips/include/asm/elf.h linux-2.6.32.48-openvz/arch/mips/include/asm/elf.h --- linux-2.6.32.48/arch/mips/include/asm/elf.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/mips/include/asm/elf.h 2011-11-17 20:08:53.000000000 -0500 @@ -368,4 +368,11 @@ extern int dump_task_fpu(struct task_str #define ELF_ET_DYN_BASE (TASK_SIZE / 3 * 2) #endif +#ifdef CONFIG_PAX_ASLR +#define PAX_ELF_ET_DYN_BASE (test_thread_flag(TIF_32BIT_ADDR) ? 0x00400000UL : 0x00400000UL) + +#define PAX_DELTA_MMAP_LEN (test_thread_flag(TIF_32BIT_ADDR) ? 27-PAGE_SHIFT : 36-PAGE_SHIFT) +#define PAX_DELTA_STACK_LEN (test_thread_flag(TIF_32BIT_ADDR) ? 27-PAGE_SHIFT : 36-PAGE_SHIFT) +#endif + #endif /* _ASM_ELF_H */ diff -urNp linux-2.6.32.48/arch/mips/include/asm/page.h linux-2.6.32.48-openvz/arch/mips/include/asm/page.h --- linux-2.6.32.48/arch/mips/include/asm/page.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/mips/include/asm/page.h 2011-11-17 20:08:53.000000000 -0500 @@ -93,7 +93,7 @@ extern void copy_user_highpage(struct pa #ifdef CONFIG_CPU_MIPS32 typedef struct { unsigned long pte_low, pte_high; } pte_t; #define pte_val(x) ((x).pte_low | ((unsigned long long)(x).pte_high << 32)) - #define __pte(x) ({ pte_t __pte = {(x), ((unsigned long long)(x)) >> 32}; __pte; }) + #define __pte(x) ({ pte_t __pte = {(x), (x) >> 32}; __pte; }) #else typedef struct { unsigned long long pte; } pte_t; #define pte_val(x) ((x).pte) diff -urNp linux-2.6.32.48/arch/mips/include/asm/reboot.h linux-2.6.32.48-openvz/arch/mips/include/asm/reboot.h --- linux-2.6.32.48/arch/mips/include/asm/reboot.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/mips/include/asm/reboot.h 2011-11-17 20:09:04.000000000 -0500 @@ -9,7 +9,7 @@ #ifndef _ASM_REBOOT_H #define _ASM_REBOOT_H -extern void (*_machine_restart)(char *command); -extern void (*_machine_halt)(void); +extern void (*__noreturn _machine_restart)(char *command); +extern void (*__noreturn _machine_halt)(void); #endif /* _ASM_REBOOT_H */ diff -urNp linux-2.6.32.48/arch/mips/include/asm/system.h linux-2.6.32.48-openvz/arch/mips/include/asm/system.h --- linux-2.6.32.48/arch/mips/include/asm/system.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/mips/include/asm/system.h 2011-11-17 20:08:53.000000000 -0500 @@ -230,6 +230,6 @@ extern void per_cpu_trap_init(void); */ #define __ARCH_WANT_UNLOCKED_CTXSW -extern unsigned long arch_align_stack(unsigned long sp); +#define arch_align_stack(x) ((x) & ~0xfUL) #endif /* _ASM_SYSTEM_H */ diff -urNp linux-2.6.32.48/arch/mips/kernel/binfmt_elfn32.c linux-2.6.32.48-openvz/arch/mips/kernel/binfmt_elfn32.c --- linux-2.6.32.48/arch/mips/kernel/binfmt_elfn32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/mips/kernel/binfmt_elfn32.c 2011-11-17 20:08:53.000000000 -0500 @@ -50,6 +50,13 @@ typedef elf_fpreg_t elf_fpregset_t[ELF_N #undef ELF_ET_DYN_BASE #define ELF_ET_DYN_BASE (TASK32_SIZE / 3 * 2) +#ifdef CONFIG_PAX_ASLR +#define PAX_ELF_ET_DYN_BASE (test_thread_flag(TIF_32BIT_ADDR) ? 0x00400000UL : 0x00400000UL) + +#define PAX_DELTA_MMAP_LEN (test_thread_flag(TIF_32BIT_ADDR) ? 27-PAGE_SHIFT : 36-PAGE_SHIFT) +#define PAX_DELTA_STACK_LEN (test_thread_flag(TIF_32BIT_ADDR) ? 27-PAGE_SHIFT : 36-PAGE_SHIFT) +#endif + #include #include #include diff -urNp linux-2.6.32.48/arch/mips/kernel/binfmt_elfo32.c linux-2.6.32.48-openvz/arch/mips/kernel/binfmt_elfo32.c --- linux-2.6.32.48/arch/mips/kernel/binfmt_elfo32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/mips/kernel/binfmt_elfo32.c 2011-11-17 20:08:53.000000000 -0500 @@ -52,6 +52,13 @@ typedef elf_fpreg_t elf_fpregset_t[ELF_N #undef ELF_ET_DYN_BASE #define ELF_ET_DYN_BASE (TASK32_SIZE / 3 * 2) +#ifdef CONFIG_PAX_ASLR +#define PAX_ELF_ET_DYN_BASE (test_thread_flag(TIF_32BIT_ADDR) ? 0x00400000UL : 0x00400000UL) + +#define PAX_DELTA_MMAP_LEN (test_thread_flag(TIF_32BIT_ADDR) ? 27-PAGE_SHIFT : 36-PAGE_SHIFT) +#define PAX_DELTA_STACK_LEN (test_thread_flag(TIF_32BIT_ADDR) ? 27-PAGE_SHIFT : 36-PAGE_SHIFT) +#endif + #include /* diff -urNp linux-2.6.32.48/arch/mips/kernel/kgdb.c linux-2.6.32.48-openvz/arch/mips/kernel/kgdb.c --- linux-2.6.32.48/arch/mips/kernel/kgdb.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/mips/kernel/kgdb.c 2011-11-17 20:08:53.000000000 -0500 @@ -245,6 +245,7 @@ int kgdb_arch_handle_exception(int vecto return -1; } +/* cannot be const */ struct kgdb_arch arch_kgdb_ops; /* diff -urNp linux-2.6.32.48/arch/mips/kernel/process.c linux-2.6.32.48-openvz/arch/mips/kernel/process.c --- linux-2.6.32.48/arch/mips/kernel/process.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/mips/kernel/process.c 2011-11-17 20:08:53.000000000 -0500 @@ -470,15 +470,3 @@ unsigned long get_wchan(struct task_stru out: return pc; } - -/* - * Don't forget that the stack pointer must be aligned on a 8 bytes - * boundary for 32-bits ABI and 16 bytes for 64-bits ABI. - */ -unsigned long arch_align_stack(unsigned long sp) -{ - if (!(current->personality & ADDR_NO_RANDOMIZE) && randomize_va_space) - sp -= get_random_int() & ~PAGE_MASK; - - return sp & ALMASK; -} diff -urNp linux-2.6.32.48/arch/mips/kernel/reset.c linux-2.6.32.48-openvz/arch/mips/kernel/reset.c --- linux-2.6.32.48/arch/mips/kernel/reset.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/mips/kernel/reset.c 2011-11-17 20:09:04.000000000 -0500 @@ -19,8 +19,8 @@ * So handle all using function pointers to machine specific * functions. */ -void (*_machine_restart)(char *command); -void (*_machine_halt)(void); +void (*__noreturn _machine_restart)(char *command); +void (*__noreturn _machine_halt)(void); void (*pm_power_off)(void); EXPORT_SYMBOL(pm_power_off); @@ -29,16 +29,19 @@ void machine_restart(char *command) { if (_machine_restart) _machine_restart(command); + BUG(); } void machine_halt(void) { if (_machine_halt) _machine_halt(); + BUG(); } void machine_power_off(void) { if (pm_power_off) pm_power_off(); + BUG(); } diff -urNp linux-2.6.32.48/arch/mips/kernel/syscall.c linux-2.6.32.48-openvz/arch/mips/kernel/syscall.c --- linux-2.6.32.48/arch/mips/kernel/syscall.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/mips/kernel/syscall.c 2011-11-17 20:08:53.000000000 -0500 @@ -102,17 +102,21 @@ unsigned long arch_get_unmapped_area(str do_color_align = 0; if (filp || (flags & MAP_SHARED)) do_color_align = 1; + +#ifdef CONFIG_PAX_RANDMMAP + if (!(current->mm->pax_flags & MF_PAX_RANDMMAP)) +#endif + if (addr) { if (do_color_align) addr = COLOUR_ALIGN(addr, pgoff); else addr = PAGE_ALIGN(addr); vmm = find_vma(current->mm, addr); - if (task_size - len >= addr && - (!vmm || addr + len <= vmm->vm_start)) + if (task_size - len >= addr && check_heap_stack_gap(vmm, addr, len)) return addr; } - addr = TASK_UNMAPPED_BASE; + addr = current->mm->mmap_base; if (do_color_align) addr = COLOUR_ALIGN(addr, pgoff); else @@ -122,7 +126,7 @@ unsigned long arch_get_unmapped_area(str /* At this point: (!vmm || addr < vmm->vm_end). */ if (task_size - len < addr) return -ENOMEM; - if (!vmm || addr + len <= vmm->vm_start) + if (check_heap_stack_gap(vmm, addr, len)) return addr; addr = vmm->vm_end; if (do_color_align) diff -urNp linux-2.6.32.48/arch/mips/Makefile linux-2.6.32.48-openvz/arch/mips/Makefile --- linux-2.6.32.48/arch/mips/Makefile 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/mips/Makefile 2011-11-17 20:09:04.000000000 -0500 @@ -51,6 +51,8 @@ endif cflags-y := -ffunction-sections cflags-y += $(call cc-option, -mno-check-zero-division) +cflags-y += -Wno-sign-compare -Wno-extra + ifdef CONFIG_32BIT ld-emul = $(32bit-emul) vmlinux-32 = vmlinux diff -urNp linux-2.6.32.48/arch/mips/mm/fault.c linux-2.6.32.48-openvz/arch/mips/mm/fault.c --- linux-2.6.32.48/arch/mips/mm/fault.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/mips/mm/fault.c 2011-11-17 20:08:53.000000000 -0500 @@ -26,6 +26,23 @@ #include #include /* For VMALLOC_END */ +#ifdef CONFIG_PAX_PAGEEXEC +void pax_report_insns(void *pc, void *sp) +{ + unsigned long i; + + printk(KERN_ERR "PAX: bytes at PC: "); + for (i = 0; i < 5; i++) { + unsigned int c; + if (get_user(c, (unsigned int *)pc+i)) + printk(KERN_CONT "???????? "); + else + printk(KERN_CONT "%08x ", c); + } + printk("\n"); +} +#endif + /* * This routine handles page faults. It determines the address, * and the problem, and then passes it off to one of the appropriate diff -urNp linux-2.6.32.48/arch/parisc/include/asm/elf.h linux-2.6.32.48-openvz/arch/parisc/include/asm/elf.h --- linux-2.6.32.48/arch/parisc/include/asm/elf.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/parisc/include/asm/elf.h 2011-11-17 20:08:53.000000000 -0500 @@ -343,6 +343,13 @@ struct pt_regs; /* forward declaration.. #define ELF_ET_DYN_BASE (TASK_UNMAPPED_BASE + 0x01000000) +#ifdef CONFIG_PAX_ASLR +#define PAX_ELF_ET_DYN_BASE 0x10000UL + +#define PAX_DELTA_MMAP_LEN 16 +#define PAX_DELTA_STACK_LEN 16 +#endif + /* This yields a mask that user programs can use to figure out what instruction set this CPU supports. This could be done in user space, but it's not easy, and we've already done it here. */ diff -urNp linux-2.6.32.48/arch/parisc/include/asm/pgtable.h linux-2.6.32.48-openvz/arch/parisc/include/asm/pgtable.h --- linux-2.6.32.48/arch/parisc/include/asm/pgtable.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/parisc/include/asm/pgtable.h 2011-11-17 20:08:53.000000000 -0500 @@ -207,6 +207,17 @@ #define PAGE_EXECREAD __pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_READ | _PAGE_EXEC |_PAGE_ACCESSED) #define PAGE_COPY PAGE_EXECREAD #define PAGE_RWX __pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_READ | _PAGE_WRITE | _PAGE_EXEC |_PAGE_ACCESSED) + +#ifdef CONFIG_PAX_PAGEEXEC +# define PAGE_SHARED_NOEXEC __pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_READ | _PAGE_WRITE | _PAGE_ACCESSED) +# define PAGE_COPY_NOEXEC __pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_READ | _PAGE_ACCESSED) +# define PAGE_READONLY_NOEXEC __pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_READ | _PAGE_ACCESSED) +#else +# define PAGE_SHARED_NOEXEC PAGE_SHARED +# define PAGE_COPY_NOEXEC PAGE_COPY +# define PAGE_READONLY_NOEXEC PAGE_READONLY +#endif + #define PAGE_KERNEL __pgprot(_PAGE_KERNEL) #define PAGE_KERNEL_RO __pgprot(_PAGE_KERNEL & ~_PAGE_WRITE) #define PAGE_KERNEL_UNC __pgprot(_PAGE_KERNEL | _PAGE_NO_CACHE) diff -urNp linux-2.6.32.48/arch/parisc/kernel/module.c linux-2.6.32.48-openvz/arch/parisc/kernel/module.c --- linux-2.6.32.48/arch/parisc/kernel/module.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/parisc/kernel/module.c 2011-11-17 20:08:53.000000000 -0500 @@ -95,16 +95,38 @@ /* three functions to determine where in the module core * or init pieces the location is */ +static inline int in_init_rx(struct module *me, void *loc) +{ + return (loc >= me->module_init_rx && + loc < (me->module_init_rx + me->init_size_rx)); +} + +static inline int in_init_rw(struct module *me, void *loc) +{ + return (loc >= me->module_init_rw && + loc < (me->module_init_rw + me->init_size_rw)); +} + static inline int in_init(struct module *me, void *loc) { - return (loc >= me->module_init && - loc <= (me->module_init + me->init_size)); + return in_init_rx(me, loc) || in_init_rw(me, loc); +} + +static inline int in_core_rx(struct module *me, void *loc) +{ + return (loc >= me->module_core_rx && + loc < (me->module_core_rx + me->core_size_rx)); +} + +static inline int in_core_rw(struct module *me, void *loc) +{ + return (loc >= me->module_core_rw && + loc < (me->module_core_rw + me->core_size_rw)); } static inline int in_core(struct module *me, void *loc) { - return (loc >= me->module_core && - loc <= (me->module_core + me->core_size)); + return in_core_rx(me, loc) || in_core_rw(me, loc); } static inline int in_local(struct module *me, void *loc) @@ -364,13 +386,13 @@ int module_frob_arch_sections(CONST Elf_ } /* align things a bit */ - me->core_size = ALIGN(me->core_size, 16); - me->arch.got_offset = me->core_size; - me->core_size += gots * sizeof(struct got_entry); - - me->core_size = ALIGN(me->core_size, 16); - me->arch.fdesc_offset = me->core_size; - me->core_size += fdescs * sizeof(Elf_Fdesc); + me->core_size_rw = ALIGN(me->core_size_rw, 16); + me->arch.got_offset = me->core_size_rw; + me->core_size_rw += gots * sizeof(struct got_entry); + + me->core_size_rw = ALIGN(me->core_size_rw, 16); + me->arch.fdesc_offset = me->core_size_rw; + me->core_size_rw += fdescs * sizeof(Elf_Fdesc); me->arch.got_max = gots; me->arch.fdesc_max = fdescs; @@ -388,7 +410,7 @@ static Elf64_Word get_got(struct module BUG_ON(value == 0); - got = me->module_core + me->arch.got_offset; + got = me->module_core_rw + me->arch.got_offset; for (i = 0; got[i].addr; i++) if (got[i].addr == value) goto out; @@ -406,7 +428,7 @@ static Elf64_Word get_got(struct module #ifdef CONFIG_64BIT static Elf_Addr get_fdesc(struct module *me, unsigned long value) { - Elf_Fdesc *fdesc = me->module_core + me->arch.fdesc_offset; + Elf_Fdesc *fdesc = me->module_core_rw + me->arch.fdesc_offset; if (!value) { printk(KERN_ERR "%s: zero OPD requested!\n", me->name); @@ -424,7 +446,7 @@ static Elf_Addr get_fdesc(struct module /* Create new one */ fdesc->addr = value; - fdesc->gp = (Elf_Addr)me->module_core + me->arch.got_offset; + fdesc->gp = (Elf_Addr)me->module_core_rw + me->arch.got_offset; return (Elf_Addr)fdesc; } #endif /* CONFIG_64BIT */ @@ -848,7 +870,7 @@ register_unwind_table(struct module *me, table = (unsigned char *)sechdrs[me->arch.unwind_section].sh_addr; end = table + sechdrs[me->arch.unwind_section].sh_size; - gp = (Elf_Addr)me->module_core + me->arch.got_offset; + gp = (Elf_Addr)me->module_core_rw + me->arch.got_offset; DEBUGP("register_unwind_table(), sect = %d at 0x%p - 0x%p (gp=0x%lx)\n", me->arch.unwind_section, table, end, gp); diff -urNp linux-2.6.32.48/arch/parisc/kernel/sys_parisc.c linux-2.6.32.48-openvz/arch/parisc/kernel/sys_parisc.c --- linux-2.6.32.48/arch/parisc/kernel/sys_parisc.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/parisc/kernel/sys_parisc.c 2011-11-17 20:08:53.000000000 -0500 @@ -43,7 +43,7 @@ static unsigned long get_unshared_area(u /* At this point: (!vma || addr < vma->vm_end). */ if (TASK_SIZE - len < addr) return -ENOMEM; - if (!vma || addr + len <= vma->vm_start) + if (check_heap_stack_gap(vma, addr, len)) return addr; addr = vma->vm_end; } @@ -79,7 +79,7 @@ static unsigned long get_shared_area(str /* At this point: (!vma || addr < vma->vm_end). */ if (TASK_SIZE - len < addr) return -ENOMEM; - if (!vma || addr + len <= vma->vm_start) + if (check_heap_stack_gap(vma, addr, len)) return addr; addr = DCACHE_ALIGN(vma->vm_end - offset) + offset; if (addr < vma->vm_end) /* handle wraparound */ @@ -98,7 +98,7 @@ unsigned long arch_get_unmapped_area(str if (flags & MAP_FIXED) return addr; if (!addr) - addr = TASK_UNMAPPED_BASE; + addr = current->mm->mmap_base; if (filp) { addr = get_shared_area(filp->f_mapping, addr, len, pgoff); diff -urNp linux-2.6.32.48/arch/parisc/kernel/traps.c linux-2.6.32.48-openvz/arch/parisc/kernel/traps.c --- linux-2.6.32.48/arch/parisc/kernel/traps.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/parisc/kernel/traps.c 2011-11-17 20:08:53.000000000 -0500 @@ -733,9 +733,7 @@ void notrace handle_interruption(int cod down_read(¤t->mm->mmap_sem); vma = find_vma(current->mm,regs->iaoq[0]); - if (vma && (regs->iaoq[0] >= vma->vm_start) - && (vma->vm_flags & VM_EXEC)) { - + if (vma && (regs->iaoq[0] >= vma->vm_start)) { fault_address = regs->iaoq[0]; fault_space = regs->iasq[0]; diff -urNp linux-2.6.32.48/arch/parisc/mm/fault.c linux-2.6.32.48-openvz/arch/parisc/mm/fault.c --- linux-2.6.32.48/arch/parisc/mm/fault.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/parisc/mm/fault.c 2011-11-17 20:08:53.000000000 -0500 @@ -15,6 +15,7 @@ #include #include #include +#include #include #include @@ -52,7 +53,7 @@ DEFINE_PER_CPU(struct exception_data, ex static unsigned long parisc_acctyp(unsigned long code, unsigned int inst) { - if (code == 6 || code == 16) + if (code == 6 || code == 7 || code == 16) return VM_EXEC; switch (inst & 0xf0000000) { @@ -138,6 +139,116 @@ parisc_acctyp(unsigned long code, unsign } #endif +#ifdef CONFIG_PAX_PAGEEXEC +/* + * PaX: decide what to do with offenders (instruction_pointer(regs) = fault address) + * + * returns 1 when task should be killed + * 2 when rt_sigreturn trampoline was detected + * 3 when unpatched PLT trampoline was detected + */ +static int pax_handle_fetch_fault(struct pt_regs *regs) +{ + +#ifdef CONFIG_PAX_EMUPLT + int err; + + do { /* PaX: unpatched PLT emulation */ + unsigned int bl, depwi; + + err = get_user(bl, (unsigned int *)instruction_pointer(regs)); + err |= get_user(depwi, (unsigned int *)(instruction_pointer(regs)+4)); + + if (err) + break; + + if (bl == 0xEA9F1FDDU && depwi == 0xD6801C1EU) { + unsigned int ldw, bv, ldw2, addr = instruction_pointer(regs)-12; + + err = get_user(ldw, (unsigned int *)addr); + err |= get_user(bv, (unsigned int *)(addr+4)); + err |= get_user(ldw2, (unsigned int *)(addr+8)); + + if (err) + break; + + if (ldw == 0x0E801096U && + bv == 0xEAC0C000U && + ldw2 == 0x0E881095U) + { + unsigned int resolver, map; + + err = get_user(resolver, (unsigned int *)(instruction_pointer(regs)+8)); + err |= get_user(map, (unsigned int *)(instruction_pointer(regs)+12)); + if (err) + break; + + regs->gr[20] = instruction_pointer(regs)+8; + regs->gr[21] = map; + regs->gr[22] = resolver; + regs->iaoq[0] = resolver | 3UL; + regs->iaoq[1] = regs->iaoq[0] + 4; + return 3; + } + } + } while (0); +#endif + +#ifdef CONFIG_PAX_EMUTRAMP + +#ifndef CONFIG_PAX_EMUSIGRT + if (!(current->mm->pax_flags & MF_PAX_EMUTRAMP)) + return 1; +#endif + + do { /* PaX: rt_sigreturn emulation */ + unsigned int ldi1, ldi2, bel, nop; + + err = get_user(ldi1, (unsigned int *)instruction_pointer(regs)); + err |= get_user(ldi2, (unsigned int *)(instruction_pointer(regs)+4)); + err |= get_user(bel, (unsigned int *)(instruction_pointer(regs)+8)); + err |= get_user(nop, (unsigned int *)(instruction_pointer(regs)+12)); + + if (err) + break; + + if ((ldi1 == 0x34190000U || ldi1 == 0x34190002U) && + ldi2 == 0x3414015AU && + bel == 0xE4008200U && + nop == 0x08000240U) + { + regs->gr[25] = (ldi1 & 2) >> 1; + regs->gr[20] = __NR_rt_sigreturn; + regs->gr[31] = regs->iaoq[1] + 16; + regs->sr[0] = regs->iasq[1]; + regs->iaoq[0] = 0x100UL; + regs->iaoq[1] = regs->iaoq[0] + 4; + regs->iasq[0] = regs->sr[2]; + regs->iasq[1] = regs->sr[2]; + return 2; + } + } while (0); +#endif + + return 1; +} + +void pax_report_insns(void *pc, void *sp) +{ + unsigned long i; + + printk(KERN_ERR "PAX: bytes at PC: "); + for (i = 0; i < 5; i++) { + unsigned int c; + if (get_user(c, (unsigned int *)pc+i)) + printk(KERN_CONT "???????? "); + else + printk(KERN_CONT "%08x ", c); + } + printk("\n"); +} +#endif + int fixup_exception(struct pt_regs *regs) { const struct exception_table_entry *fix; @@ -192,8 +303,33 @@ good_area: acc_type = parisc_acctyp(code,regs->iir); - if ((vma->vm_flags & acc_type) != acc_type) + if ((vma->vm_flags & acc_type) != acc_type) { + +#ifdef CONFIG_PAX_PAGEEXEC + if ((mm->pax_flags & MF_PAX_PAGEEXEC) && (acc_type & VM_EXEC) && + (address & ~3UL) == instruction_pointer(regs)) + { + up_read(&mm->mmap_sem); + switch (pax_handle_fetch_fault(regs)) { + +#ifdef CONFIG_PAX_EMUPLT + case 3: + return; +#endif + +#ifdef CONFIG_PAX_EMUTRAMP + case 2: + return; +#endif + + } + pax_report_fault(regs, (void *)instruction_pointer(regs), (void *)regs->gr[30]); + do_group_exit(SIGKILL); + } +#endif + goto bad_area; + } /* * If for any reason at all we couldn't handle the fault, make diff -urNp linux-2.6.32.48/arch/powerpc/include/asm/device.h linux-2.6.32.48-openvz/arch/powerpc/include/asm/device.h --- linux-2.6.32.48/arch/powerpc/include/asm/device.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/include/asm/device.h 2011-11-17 20:08:53.000000000 -0500 @@ -14,7 +14,7 @@ struct dev_archdata { struct device_node *of_node; /* DMA operations on that device */ - struct dma_map_ops *dma_ops; + const struct dma_map_ops *dma_ops; /* * When an iommu is in use, dma_data is used as a ptr to the base of the diff -urNp linux-2.6.32.48/arch/powerpc/include/asm/dma-mapping.h linux-2.6.32.48-openvz/arch/powerpc/include/asm/dma-mapping.h --- linux-2.6.32.48/arch/powerpc/include/asm/dma-mapping.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/include/asm/dma-mapping.h 2011-11-17 20:08:53.000000000 -0500 @@ -69,9 +69,9 @@ static inline unsigned long device_to_ma #ifdef CONFIG_PPC64 extern struct dma_map_ops dma_iommu_ops; #endif -extern struct dma_map_ops dma_direct_ops; +extern const struct dma_map_ops dma_direct_ops; -static inline struct dma_map_ops *get_dma_ops(struct device *dev) +static inline const struct dma_map_ops *get_dma_ops(struct device *dev) { /* We don't handle the NULL dev case for ISA for now. We could * do it via an out of line call but it is not needed for now. The @@ -84,7 +84,7 @@ static inline struct dma_map_ops *get_dm return dev->archdata.dma_ops; } -static inline void set_dma_ops(struct device *dev, struct dma_map_ops *ops) +static inline void set_dma_ops(struct device *dev, const struct dma_map_ops *ops) { dev->archdata.dma_ops = ops; } @@ -118,7 +118,7 @@ static inline void set_dma_offset(struct static inline int dma_supported(struct device *dev, u64 mask) { - struct dma_map_ops *dma_ops = get_dma_ops(dev); + const struct dma_map_ops *dma_ops = get_dma_ops(dev); if (unlikely(dma_ops == NULL)) return 0; @@ -132,7 +132,7 @@ static inline int dma_supported(struct d static inline int dma_set_mask(struct device *dev, u64 dma_mask) { - struct dma_map_ops *dma_ops = get_dma_ops(dev); + const struct dma_map_ops *dma_ops = get_dma_ops(dev); if (unlikely(dma_ops == NULL)) return -EIO; @@ -147,7 +147,7 @@ static inline int dma_set_mask(struct de static inline void *dma_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_handle, gfp_t flag) { - struct dma_map_ops *dma_ops = get_dma_ops(dev); + const struct dma_map_ops *dma_ops = get_dma_ops(dev); void *cpu_addr; BUG_ON(!dma_ops); @@ -162,7 +162,7 @@ static inline void *dma_alloc_coherent(s static inline void dma_free_coherent(struct device *dev, size_t size, void *cpu_addr, dma_addr_t dma_handle) { - struct dma_map_ops *dma_ops = get_dma_ops(dev); + const struct dma_map_ops *dma_ops = get_dma_ops(dev); BUG_ON(!dma_ops); @@ -173,7 +173,7 @@ static inline void dma_free_coherent(str static inline int dma_mapping_error(struct device *dev, dma_addr_t dma_addr) { - struct dma_map_ops *dma_ops = get_dma_ops(dev); + const struct dma_map_ops *dma_ops = get_dma_ops(dev); if (dma_ops->mapping_error) return dma_ops->mapping_error(dev, dma_addr); diff -urNp linux-2.6.32.48/arch/powerpc/include/asm/elf.h linux-2.6.32.48-openvz/arch/powerpc/include/asm/elf.h --- linux-2.6.32.48/arch/powerpc/include/asm/elf.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/include/asm/elf.h 2011-11-17 20:08:53.000000000 -0500 @@ -179,8 +179,19 @@ typedef elf_fpreg_t elf_vsrreghalf_t32[E the loader. We need to make sure that it is out of the way of the program that it will "exec", and that there is sufficient room for the brk. */ -extern unsigned long randomize_et_dyn(unsigned long base); -#define ELF_ET_DYN_BASE (randomize_et_dyn(0x20000000)) +#define ELF_ET_DYN_BASE (0x20000000) + +#ifdef CONFIG_PAX_ASLR +#define PAX_ELF_ET_DYN_BASE (0x10000000UL) + +#ifdef __powerpc64__ +#define PAX_DELTA_MMAP_LEN (test_thread_flag(TIF_32BIT) ? 16 : 28) +#define PAX_DELTA_STACK_LEN (test_thread_flag(TIF_32BIT) ? 16 : 28) +#else +#define PAX_DELTA_MMAP_LEN 15 +#define PAX_DELTA_STACK_LEN 15 +#endif +#endif /* * Our registers are always unsigned longs, whether we're a 32 bit @@ -275,9 +286,6 @@ extern int arch_setup_additional_pages(s (0x7ff >> (PAGE_SHIFT - 12)) : \ (0x3ffff >> (PAGE_SHIFT - 12))) -extern unsigned long arch_randomize_brk(struct mm_struct *mm); -#define arch_randomize_brk arch_randomize_brk - #endif /* __KERNEL__ */ /* diff -urNp linux-2.6.32.48/arch/powerpc/include/asm/iommu.h linux-2.6.32.48-openvz/arch/powerpc/include/asm/iommu.h --- linux-2.6.32.48/arch/powerpc/include/asm/iommu.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/include/asm/iommu.h 2011-11-17 20:08:53.000000000 -0500 @@ -116,6 +116,9 @@ extern void iommu_init_early_iSeries(voi extern void iommu_init_early_dart(void); extern void iommu_init_early_pasemi(void); +/* dma-iommu.c */ +extern int dma_iommu_dma_supported(struct device *dev, u64 mask); + #ifdef CONFIG_PCI extern void pci_iommu_init(void); extern void pci_direct_iommu_init(void); diff -urNp linux-2.6.32.48/arch/powerpc/include/asm/kmap_types.h linux-2.6.32.48-openvz/arch/powerpc/include/asm/kmap_types.h --- linux-2.6.32.48/arch/powerpc/include/asm/kmap_types.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/include/asm/kmap_types.h 2011-11-17 20:08:53.000000000 -0500 @@ -26,6 +26,7 @@ enum km_type { KM_SOFTIRQ1, KM_PPC_SYNC_PAGE, KM_PPC_SYNC_ICACHE, + KM_CLEARPAGE, KM_TYPE_NR }; diff -urNp linux-2.6.32.48/arch/powerpc/include/asm/page_64.h linux-2.6.32.48-openvz/arch/powerpc/include/asm/page_64.h --- linux-2.6.32.48/arch/powerpc/include/asm/page_64.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/include/asm/page_64.h 2011-11-17 20:08:53.000000000 -0500 @@ -180,15 +180,18 @@ do { \ * stack by default, so in the absense of a PT_GNU_STACK program header * we turn execute permission off. */ -#define VM_STACK_DEFAULT_FLAGS32 (VM_READ | VM_WRITE | VM_EXEC | \ - VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC) +#define VM_STACK_DEFAULT_FLAGS32 \ + (((current->personality & READ_IMPLIES_EXEC) ? VM_EXEC : 0) | \ + VM_READ | VM_WRITE | VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC) #define VM_STACK_DEFAULT_FLAGS64 (VM_READ | VM_WRITE | \ VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC) +#ifndef CONFIG_PAX_PAGEEXEC #define VM_STACK_DEFAULT_FLAGS \ (test_thread_flag(TIF_32BIT) ? \ VM_STACK_DEFAULT_FLAGS32 : VM_STACK_DEFAULT_FLAGS64) +#endif #include diff -urNp linux-2.6.32.48/arch/powerpc/include/asm/page.h linux-2.6.32.48-openvz/arch/powerpc/include/asm/page.h --- linux-2.6.32.48/arch/powerpc/include/asm/page.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/include/asm/page.h 2011-11-17 20:08:53.000000000 -0500 @@ -116,8 +116,9 @@ extern phys_addr_t kernstart_addr; * and needs to be executable. This means the whole heap ends * up being executable. */ -#define VM_DATA_DEFAULT_FLAGS32 (VM_READ | VM_WRITE | VM_EXEC | \ - VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC) +#define VM_DATA_DEFAULT_FLAGS32 \ + (((current->personality & READ_IMPLIES_EXEC) ? VM_EXEC : 0) | \ + VM_READ | VM_WRITE | VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC) #define VM_DATA_DEFAULT_FLAGS64 (VM_READ | VM_WRITE | \ VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC) @@ -145,6 +146,9 @@ extern phys_addr_t kernstart_addr; #define is_kernel_addr(x) ((x) >= PAGE_OFFSET) #endif +#define ktla_ktva(addr) (addr) +#define ktva_ktla(addr) (addr) + #ifndef __ASSEMBLY__ #undef STRICT_MM_TYPECHECKS diff -urNp linux-2.6.32.48/arch/powerpc/include/asm/pci.h linux-2.6.32.48-openvz/arch/powerpc/include/asm/pci.h --- linux-2.6.32.48/arch/powerpc/include/asm/pci.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/include/asm/pci.h 2011-11-17 20:08:53.000000000 -0500 @@ -65,8 +65,8 @@ static inline int pci_get_legacy_ide_irq } #ifdef CONFIG_PCI -extern void set_pci_dma_ops(struct dma_map_ops *dma_ops); -extern struct dma_map_ops *get_pci_dma_ops(void); +extern void set_pci_dma_ops(const struct dma_map_ops *dma_ops); +extern const struct dma_map_ops *get_pci_dma_ops(void); #else /* CONFIG_PCI */ #define set_pci_dma_ops(d) #define get_pci_dma_ops() NULL diff -urNp linux-2.6.32.48/arch/powerpc/include/asm/pgtable.h linux-2.6.32.48-openvz/arch/powerpc/include/asm/pgtable.h --- linux-2.6.32.48/arch/powerpc/include/asm/pgtable.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/include/asm/pgtable.h 2011-11-17 20:08:53.000000000 -0500 @@ -2,6 +2,7 @@ #define _ASM_POWERPC_PGTABLE_H #ifdef __KERNEL__ +#include #ifndef __ASSEMBLY__ #include /* For TASK_SIZE */ #include diff -urNp linux-2.6.32.48/arch/powerpc/include/asm/pte-hash32.h linux-2.6.32.48-openvz/arch/powerpc/include/asm/pte-hash32.h --- linux-2.6.32.48/arch/powerpc/include/asm/pte-hash32.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/include/asm/pte-hash32.h 2011-11-17 20:08:53.000000000 -0500 @@ -21,6 +21,7 @@ #define _PAGE_FILE 0x004 /* when !present: nonlinear file mapping */ #define _PAGE_USER 0x004 /* usermode access allowed */ #define _PAGE_GUARDED 0x008 /* G: prohibit speculative access */ +#define _PAGE_EXEC _PAGE_GUARDED #define _PAGE_COHERENT 0x010 /* M: enforce memory coherence (SMP systems) */ #define _PAGE_NO_CACHE 0x020 /* I: cache inhibit */ #define _PAGE_WRITETHRU 0x040 /* W: cache write-through */ diff -urNp linux-2.6.32.48/arch/powerpc/include/asm/ptrace.h linux-2.6.32.48-openvz/arch/powerpc/include/asm/ptrace.h --- linux-2.6.32.48/arch/powerpc/include/asm/ptrace.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/include/asm/ptrace.h 2011-11-17 20:09:04.000000000 -0500 @@ -103,7 +103,7 @@ extern unsigned long profile_pc(struct p } while(0) struct task_struct; -extern unsigned long ptrace_get_reg(struct task_struct *task, int regno); +extern unsigned long ptrace_get_reg(struct task_struct *task, unsigned int regno); extern int ptrace_put_reg(struct task_struct *task, int regno, unsigned long data); diff -urNp linux-2.6.32.48/arch/powerpc/include/asm/reg.h linux-2.6.32.48-openvz/arch/powerpc/include/asm/reg.h --- linux-2.6.32.48/arch/powerpc/include/asm/reg.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/include/asm/reg.h 2011-11-17 20:08:53.000000000 -0500 @@ -191,6 +191,7 @@ #define SPRN_DBCR 0x136 /* e300 Data Breakpoint Control Reg */ #define SPRN_DSISR 0x012 /* Data Storage Interrupt Status Register */ #define DSISR_NOHPTE 0x40000000 /* no translation found */ +#define DSISR_GUARDED 0x10000000 /* fetch from guarded storage */ #define DSISR_PROTFAULT 0x08000000 /* protection fault */ #define DSISR_ISSTORE 0x02000000 /* access was a store */ #define DSISR_DABRMATCH 0x00400000 /* hit data breakpoint */ diff -urNp linux-2.6.32.48/arch/powerpc/include/asm/swiotlb.h linux-2.6.32.48-openvz/arch/powerpc/include/asm/swiotlb.h --- linux-2.6.32.48/arch/powerpc/include/asm/swiotlb.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/include/asm/swiotlb.h 2011-11-17 20:08:53.000000000 -0500 @@ -13,7 +13,7 @@ #include -extern struct dma_map_ops swiotlb_dma_ops; +extern const struct dma_map_ops swiotlb_dma_ops; static inline void dma_mark_clean(void *addr, size_t size) {} diff -urNp linux-2.6.32.48/arch/powerpc/include/asm/system.h linux-2.6.32.48-openvz/arch/powerpc/include/asm/system.h --- linux-2.6.32.48/arch/powerpc/include/asm/system.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/include/asm/system.h 2011-11-17 20:08:53.000000000 -0500 @@ -531,7 +531,7 @@ __cmpxchg_local(volatile void *ptr, unsi #define cmpxchg64_local(ptr, o, n) __cmpxchg64_local_generic((ptr), (o), (n)) #endif -extern unsigned long arch_align_stack(unsigned long sp); +#define arch_align_stack(x) ((x) & ~0xfUL) /* Used in very early kernel initialization. */ extern unsigned long reloc_offset(void); diff -urNp linux-2.6.32.48/arch/powerpc/include/asm/uaccess.h linux-2.6.32.48-openvz/arch/powerpc/include/asm/uaccess.h --- linux-2.6.32.48/arch/powerpc/include/asm/uaccess.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/include/asm/uaccess.h 2011-11-17 20:08:53.000000000 -0500 @@ -13,6 +13,8 @@ #define VERIFY_READ 0 #define VERIFY_WRITE 1 +extern void check_object_size(const void *ptr, unsigned long n, bool to); + /* * The fs value determines whether argument validity checking should be * performed or not. If get_fs() == USER_DS, checking is performed, with @@ -327,52 +329,6 @@ do { \ extern unsigned long __copy_tofrom_user(void __user *to, const void __user *from, unsigned long size); -#ifndef __powerpc64__ - -static inline unsigned long copy_from_user(void *to, - const void __user *from, unsigned long n) -{ - unsigned long over; - - if (access_ok(VERIFY_READ, from, n)) - return __copy_tofrom_user((__force void __user *)to, from, n); - if ((unsigned long)from < TASK_SIZE) { - over = (unsigned long)from + n - TASK_SIZE; - return __copy_tofrom_user((__force void __user *)to, from, - n - over) + over; - } - return n; -} - -static inline unsigned long copy_to_user(void __user *to, - const void *from, unsigned long n) -{ - unsigned long over; - - if (access_ok(VERIFY_WRITE, to, n)) - return __copy_tofrom_user(to, (__force void __user *)from, n); - if ((unsigned long)to < TASK_SIZE) { - over = (unsigned long)to + n - TASK_SIZE; - return __copy_tofrom_user(to, (__force void __user *)from, - n - over) + over; - } - return n; -} - -#else /* __powerpc64__ */ - -#define __copy_in_user(to, from, size) \ - __copy_tofrom_user((to), (from), (size)) - -extern unsigned long copy_from_user(void *to, const void __user *from, - unsigned long n); -extern unsigned long copy_to_user(void __user *to, const void *from, - unsigned long n); -extern unsigned long copy_in_user(void __user *to, const void __user *from, - unsigned long n); - -#endif /* __powerpc64__ */ - static inline unsigned long __copy_from_user_inatomic(void *to, const void __user *from, unsigned long n) { @@ -396,6 +352,10 @@ static inline unsigned long __copy_from_ if (ret == 0) return 0; } + + if (!__builtin_constant_p(n)) + check_object_size(to, n, false); + return __copy_tofrom_user((__force void __user *)to, from, n); } @@ -422,6 +382,10 @@ static inline unsigned long __copy_to_us if (ret == 0) return 0; } + + if (!__builtin_constant_p(n)) + check_object_size(from, n, true); + return __copy_tofrom_user(to, (__force const void __user *)from, n); } @@ -439,6 +403,92 @@ static inline unsigned long __copy_to_us return __copy_to_user_inatomic(to, from, size); } +#ifndef __powerpc64__ + +static inline unsigned long __must_check copy_from_user(void *to, + const void __user *from, unsigned long n) +{ + unsigned long over; + + if ((long)n < 0) + return n; + + if (access_ok(VERIFY_READ, from, n)) { + if (!__builtin_constant_p(n)) + check_object_size(to, n, false); + return __copy_tofrom_user((__force void __user *)to, from, n); + } + if ((unsigned long)from < TASK_SIZE) { + over = (unsigned long)from + n - TASK_SIZE; + if (!__builtin_constant_p(n - over)) + check_object_size(to, n - over, false); + return __copy_tofrom_user((__force void __user *)to, from, + n - over) + over; + } + return n; +} + +static inline unsigned long __must_check copy_to_user(void __user *to, + const void *from, unsigned long n) +{ + unsigned long over; + + if ((long)n < 0) + return n; + + if (access_ok(VERIFY_WRITE, to, n)) { + if (!__builtin_constant_p(n)) + check_object_size(from, n, true); + return __copy_tofrom_user(to, (__force void __user *)from, n); + } + if ((unsigned long)to < TASK_SIZE) { + over = (unsigned long)to + n - TASK_SIZE; + if (!__builtin_constant_p(n)) + check_object_size(from, n - over, true); + return __copy_tofrom_user(to, (__force void __user *)from, + n - over) + over; + } + return n; +} + +#else /* __powerpc64__ */ + +#define __copy_in_user(to, from, size) \ + __copy_tofrom_user((to), (from), (size)) + +static inline unsigned long __must_check copy_from_user(void *to, const void __user *from, unsigned long n) +{ + if ((long)n < 0 || n > INT_MAX) + return n; + + if (!__builtin_constant_p(n)) + check_object_size(to, n, false); + + if (likely(access_ok(VERIFY_READ, from, n))) + n = __copy_from_user(to, from, n); + else + memset(to, 0, n); + return n; +} + +static inline unsigned long __must_check copy_to_user(void __user *to, const void *from, unsigned long n) +{ + if ((long)n < 0 || n > INT_MAX) + return n; + + if (likely(access_ok(VERIFY_WRITE, to, n))) { + if (!__builtin_constant_p(n)) + check_object_size(from, n, true); + n = __copy_to_user(to, from, n); + } + return n; +} + +extern unsigned long copy_in_user(void __user *to, const void __user *from, + unsigned long n); + +#endif /* __powerpc64__ */ + extern unsigned long __clear_user(void __user *addr, unsigned long size); static inline unsigned long clear_user(void __user *addr, unsigned long size) diff -urNp linux-2.6.32.48/arch/powerpc/kernel/cacheinfo.c linux-2.6.32.48-openvz/arch/powerpc/kernel/cacheinfo.c --- linux-2.6.32.48/arch/powerpc/kernel/cacheinfo.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/kernel/cacheinfo.c 2011-11-17 20:08:53.000000000 -0500 @@ -642,7 +642,7 @@ static struct kobj_attribute *cache_inde &cache_assoc_attr, }; -static struct sysfs_ops cache_index_ops = { +static const struct sysfs_ops cache_index_ops = { .show = cache_index_show, }; diff -urNp linux-2.6.32.48/arch/powerpc/kernel/dma.c linux-2.6.32.48-openvz/arch/powerpc/kernel/dma.c --- linux-2.6.32.48/arch/powerpc/kernel/dma.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/kernel/dma.c 2011-11-17 20:08:53.000000000 -0500 @@ -134,7 +134,7 @@ static inline void dma_direct_sync_singl } #endif -struct dma_map_ops dma_direct_ops = { +const struct dma_map_ops dma_direct_ops = { .alloc_coherent = dma_direct_alloc_coherent, .free_coherent = dma_direct_free_coherent, .map_sg = dma_direct_map_sg, diff -urNp linux-2.6.32.48/arch/powerpc/kernel/dma-iommu.c linux-2.6.32.48-openvz/arch/powerpc/kernel/dma-iommu.c --- linux-2.6.32.48/arch/powerpc/kernel/dma-iommu.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/kernel/dma-iommu.c 2011-11-17 20:08:53.000000000 -0500 @@ -70,7 +70,7 @@ static void dma_iommu_unmap_sg(struct de } /* We support DMA to/from any memory page via the iommu */ -static int dma_iommu_dma_supported(struct device *dev, u64 mask) +int dma_iommu_dma_supported(struct device *dev, u64 mask) { struct iommu_table *tbl = get_iommu_table_base(dev); diff -urNp linux-2.6.32.48/arch/powerpc/kernel/dma-swiotlb.c linux-2.6.32.48-openvz/arch/powerpc/kernel/dma-swiotlb.c --- linux-2.6.32.48/arch/powerpc/kernel/dma-swiotlb.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/kernel/dma-swiotlb.c 2011-11-17 20:08:53.000000000 -0500 @@ -31,7 +31,7 @@ unsigned int ppc_swiotlb_enable; * map_page, and unmap_page on highmem, use normal dma_ops * for everything else. */ -struct dma_map_ops swiotlb_dma_ops = { +const struct dma_map_ops swiotlb_dma_ops = { .alloc_coherent = dma_direct_alloc_coherent, .free_coherent = dma_direct_free_coherent, .map_sg = swiotlb_map_sg_attrs, diff -urNp linux-2.6.32.48/arch/powerpc/kernel/exceptions-64e.S linux-2.6.32.48-openvz/arch/powerpc/kernel/exceptions-64e.S --- linux-2.6.32.48/arch/powerpc/kernel/exceptions-64e.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/kernel/exceptions-64e.S 2011-11-17 20:08:53.000000000 -0500 @@ -455,6 +455,7 @@ storage_fault_common: std r14,_DAR(r1) std r15,_DSISR(r1) addi r3,r1,STACK_FRAME_OVERHEAD + bl .save_nvgprs mr r4,r14 mr r5,r15 ld r14,PACA_EXGEN+EX_R14(r13) @@ -464,8 +465,7 @@ storage_fault_common: cmpdi r3,0 bne- 1f b .ret_from_except_lite -1: bl .save_nvgprs - mr r5,r3 +1: mr r5,r3 addi r3,r1,STACK_FRAME_OVERHEAD ld r4,_DAR(r1) bl .bad_page_fault diff -urNp linux-2.6.32.48/arch/powerpc/kernel/exceptions-64s.S linux-2.6.32.48-openvz/arch/powerpc/kernel/exceptions-64s.S --- linux-2.6.32.48/arch/powerpc/kernel/exceptions-64s.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/kernel/exceptions-64s.S 2011-11-17 20:08:53.000000000 -0500 @@ -818,10 +818,10 @@ handle_page_fault: 11: ld r4,_DAR(r1) ld r5,_DSISR(r1) addi r3,r1,STACK_FRAME_OVERHEAD + bl .save_nvgprs bl .do_page_fault cmpdi r3,0 beq+ 13f - bl .save_nvgprs mr r5,r3 addi r3,r1,STACK_FRAME_OVERHEAD lwz r4,_DAR(r1) diff -urNp linux-2.6.32.48/arch/powerpc/kernel/ibmebus.c linux-2.6.32.48-openvz/arch/powerpc/kernel/ibmebus.c --- linux-2.6.32.48/arch/powerpc/kernel/ibmebus.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/kernel/ibmebus.c 2011-11-17 20:08:53.000000000 -0500 @@ -127,7 +127,7 @@ static int ibmebus_dma_supported(struct return 1; } -static struct dma_map_ops ibmebus_dma_ops = { +static const struct dma_map_ops ibmebus_dma_ops = { .alloc_coherent = ibmebus_alloc_coherent, .free_coherent = ibmebus_free_coherent, .map_sg = ibmebus_map_sg, diff -urNp linux-2.6.32.48/arch/powerpc/kernel/kgdb.c linux-2.6.32.48-openvz/arch/powerpc/kernel/kgdb.c --- linux-2.6.32.48/arch/powerpc/kernel/kgdb.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/kernel/kgdb.c 2011-11-17 20:08:53.000000000 -0500 @@ -126,7 +126,7 @@ static int kgdb_handle_breakpoint(struct if (kgdb_handle_exception(0, SIGTRAP, 0, regs) != 0) return 0; - if (*(u32 *) (regs->nip) == *(u32 *) (&arch_kgdb_ops.gdb_bpt_instr)) + if (*(u32 *) (regs->nip) == *(const u32 *) (&arch_kgdb_ops.gdb_bpt_instr)) regs->nip += 4; return 1; @@ -353,7 +353,7 @@ int kgdb_arch_handle_exception(int vecto /* * Global data */ -struct kgdb_arch arch_kgdb_ops = { +const struct kgdb_arch arch_kgdb_ops = { .gdb_bpt_instr = {0x7d, 0x82, 0x10, 0x08}, }; diff -urNp linux-2.6.32.48/arch/powerpc/kernel/module_32.c linux-2.6.32.48-openvz/arch/powerpc/kernel/module_32.c --- linux-2.6.32.48/arch/powerpc/kernel/module_32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/kernel/module_32.c 2011-11-17 20:08:53.000000000 -0500 @@ -162,7 +162,7 @@ int module_frob_arch_sections(Elf32_Ehdr me->arch.core_plt_section = i; } if (!me->arch.core_plt_section || !me->arch.init_plt_section) { - printk("Module doesn't contain .plt or .init.plt sections.\n"); + printk("Module %s doesn't contain .plt or .init.plt sections.\n", me->name); return -ENOEXEC; } @@ -203,11 +203,16 @@ static uint32_t do_plt_call(void *locati DEBUGP("Doing plt for call to 0x%x at 0x%x\n", val, (unsigned int)location); /* Init, or core PLT? */ - if (location >= mod->module_core - && location < mod->module_core + mod->core_size) + if ((location >= mod->module_core_rx && location < mod->module_core_rx + mod->core_size_rx) || + (location >= mod->module_core_rw && location < mod->module_core_rw + mod->core_size_rw)) entry = (void *)sechdrs[mod->arch.core_plt_section].sh_addr; - else + else if ((location >= mod->module_init_rx && location < mod->module_init_rx + mod->init_size_rx) || + (location >= mod->module_init_rw && location < mod->module_init_rw + mod->init_size_rw)) entry = (void *)sechdrs[mod->arch.init_plt_section].sh_addr; + else { + printk(KERN_ERR "%s: invalid R_PPC_REL24 entry found\n", mod->name); + return ~0UL; + } /* Find this entry, or if that fails, the next avail. entry */ while (entry->jump[0]) { diff -urNp linux-2.6.32.48/arch/powerpc/kernel/module.c linux-2.6.32.48-openvz/arch/powerpc/kernel/module.c --- linux-2.6.32.48/arch/powerpc/kernel/module.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/kernel/module.c 2011-11-17 20:08:53.000000000 -0500 @@ -31,11 +31,24 @@ LIST_HEAD(module_bug_list); +#ifdef CONFIG_PAX_KERNEXEC void *module_alloc(unsigned long size) { if (size == 0) return NULL; + return vmalloc(size); +} + +void *module_alloc_exec(unsigned long size) +#else +void *module_alloc(unsigned long size) +#endif + +{ + if (size == 0) + return NULL; + return vmalloc_exec(size); } @@ -45,6 +58,13 @@ void module_free(struct module *mod, voi vfree(module_region); } +#ifdef CONFIG_PAX_KERNEXEC +void module_free_exec(struct module *mod, void *module_region) +{ + module_free(mod, module_region); +} +#endif + static const Elf_Shdr *find_section(const Elf_Ehdr *hdr, const Elf_Shdr *sechdrs, const char *name) diff -urNp linux-2.6.32.48/arch/powerpc/kernel/pci-common.c linux-2.6.32.48-openvz/arch/powerpc/kernel/pci-common.c --- linux-2.6.32.48/arch/powerpc/kernel/pci-common.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/kernel/pci-common.c 2011-11-17 20:08:53.000000000 -0500 @@ -50,14 +50,14 @@ resource_size_t isa_mem_base; unsigned int ppc_pci_flags = 0; -static struct dma_map_ops *pci_dma_ops = &dma_direct_ops; +static const struct dma_map_ops *pci_dma_ops = &dma_direct_ops; -void set_pci_dma_ops(struct dma_map_ops *dma_ops) +void set_pci_dma_ops(const struct dma_map_ops *dma_ops) { pci_dma_ops = dma_ops; } -struct dma_map_ops *get_pci_dma_ops(void) +const struct dma_map_ops *get_pci_dma_ops(void) { return pci_dma_ops; } diff -urNp linux-2.6.32.48/arch/powerpc/kernel/process.c linux-2.6.32.48-openvz/arch/powerpc/kernel/process.c --- linux-2.6.32.48/arch/powerpc/kernel/process.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/kernel/process.c 2011-11-17 20:08:53.000000000 -0500 @@ -539,8 +539,8 @@ void show_regs(struct pt_regs * regs) * Lookup NIP late so we have the best change of getting the * above info out without failing */ - printk("NIP ["REG"] %pS\n", regs->nip, (void *)regs->nip); - printk("LR ["REG"] %pS\n", regs->link, (void *)regs->link); + printk("NIP ["REG"] %pA\n", regs->nip, (void *)regs->nip); + printk("LR ["REG"] %pA\n", regs->link, (void *)regs->link); #endif show_stack(current, (unsigned long *) regs->gpr[1]); if (!user_mode(regs)) @@ -1034,10 +1034,10 @@ void show_stack(struct task_struct *tsk, newsp = stack[0]; ip = stack[STACK_FRAME_LR_SAVE]; if (!firstframe || ip != lr) { - printk("["REG"] ["REG"] %pS", sp, ip, (void *)ip); + printk("["REG"] ["REG"] %pA", sp, ip, (void *)ip); #ifdef CONFIG_FUNCTION_GRAPH_TRACER if ((ip == rth || ip == mrth) && curr_frame >= 0) { - printk(" (%pS)", + printk(" (%pA)", (void *)current->ret_stack[curr_frame].ret); curr_frame--; } @@ -1057,7 +1057,7 @@ void show_stack(struct task_struct *tsk, struct pt_regs *regs = (struct pt_regs *) (sp + STACK_FRAME_OVERHEAD); lr = regs->link; - printk("--- Exception: %lx at %pS\n LR = %pS\n", + printk("--- Exception: %lx at %pA\n LR = %pA\n", regs->trap, (void *)regs->nip, (void *)lr); firstframe = 1; } @@ -1134,58 +1134,3 @@ void thread_info_cache_init(void) } #endif /* THREAD_SHIFT < PAGE_SHIFT */ - -unsigned long arch_align_stack(unsigned long sp) -{ - if (!(current->personality & ADDR_NO_RANDOMIZE) && randomize_va_space) - sp -= get_random_int() & ~PAGE_MASK; - return sp & ~0xf; -} - -static inline unsigned long brk_rnd(void) -{ - unsigned long rnd = 0; - - /* 8MB for 32bit, 1GB for 64bit */ - if (is_32bit_task()) - rnd = (long)(get_random_int() % (1<<(23-PAGE_SHIFT))); - else - rnd = (long)(get_random_int() % (1<<(30-PAGE_SHIFT))); - - return rnd << PAGE_SHIFT; -} - -unsigned long arch_randomize_brk(struct mm_struct *mm) -{ - unsigned long base = mm->brk; - unsigned long ret; - -#ifdef CONFIG_PPC_STD_MMU_64 - /* - * If we are using 1TB segments and we are allowed to randomise - * the heap, we can put it above 1TB so it is backed by a 1TB - * segment. Otherwise the heap will be in the bottom 1TB - * which always uses 256MB segments and this may result in a - * performance penalty. - */ - if (!is_32bit_task() && (mmu_highuser_ssize == MMU_SEGSIZE_1T)) - base = max_t(unsigned long, mm->brk, 1UL << SID_SHIFT_1T); -#endif - - ret = PAGE_ALIGN(base + brk_rnd()); - - if (ret < mm->brk) - return mm->brk; - - return ret; -} - -unsigned long randomize_et_dyn(unsigned long base) -{ - unsigned long ret = PAGE_ALIGN(base + brk_rnd()); - - if (ret < base) - return base; - - return ret; -} diff -urNp linux-2.6.32.48/arch/powerpc/kernel/ptrace.c linux-2.6.32.48-openvz/arch/powerpc/kernel/ptrace.c --- linux-2.6.32.48/arch/powerpc/kernel/ptrace.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/kernel/ptrace.c 2011-11-17 20:09:04.000000000 -0500 @@ -86,7 +86,7 @@ static int set_user_trap(struct task_str /* * Get contents of register REGNO in task TASK. */ -unsigned long ptrace_get_reg(struct task_struct *task, int regno) +unsigned long ptrace_get_reg(struct task_struct *task, unsigned int regno) { if (task->thread.regs == NULL) return -EIO; @@ -894,7 +894,7 @@ long arch_ptrace(struct task_struct *chi CHECK_FULL_REGS(child->thread.regs); if (index < PT_FPR0) { - tmp = ptrace_get_reg(child, (int) index); + tmp = ptrace_get_reg(child, index); } else { flush_fp_to_thread(child); tmp = ((unsigned long *)child->thread.fpr) diff -urNp linux-2.6.32.48/arch/powerpc/kernel/signal_32.c linux-2.6.32.48-openvz/arch/powerpc/kernel/signal_32.c --- linux-2.6.32.48/arch/powerpc/kernel/signal_32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/kernel/signal_32.c 2011-11-17 20:08:53.000000000 -0500 @@ -857,7 +857,7 @@ int handle_rt_signal32(unsigned long sig /* Save user registers on the stack */ frame = &rt_sf->uc.uc_mcontext; addr = frame; - if (vdso32_rt_sigtramp && current->mm->context.vdso_base) { + if (vdso32_rt_sigtramp && current->mm->context.vdso_base != ~0UL) { if (save_user_regs(regs, frame, 0, 1)) goto badframe; regs->link = current->mm->context.vdso_base + vdso32_rt_sigtramp; diff -urNp linux-2.6.32.48/arch/powerpc/kernel/signal_64.c linux-2.6.32.48-openvz/arch/powerpc/kernel/signal_64.c --- linux-2.6.32.48/arch/powerpc/kernel/signal_64.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/kernel/signal_64.c 2011-11-17 20:08:53.000000000 -0500 @@ -429,7 +429,7 @@ int handle_rt_signal64(int signr, struct current->thread.fpscr.val = 0; /* Set up to return from userspace. */ - if (vdso64_rt_sigtramp && current->mm->context.vdso_base) { + if (vdso64_rt_sigtramp && current->mm->context.vdso_base != ~0UL) { regs->link = current->mm->context.vdso_base + vdso64_rt_sigtramp; } else { err |= setup_trampoline(__NR_rt_sigreturn, &frame->tramp[0]); diff -urNp linux-2.6.32.48/arch/powerpc/kernel/sys_ppc32.c linux-2.6.32.48-openvz/arch/powerpc/kernel/sys_ppc32.c --- linux-2.6.32.48/arch/powerpc/kernel/sys_ppc32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/kernel/sys_ppc32.c 2011-11-17 20:08:53.000000000 -0500 @@ -563,10 +563,10 @@ asmlinkage long compat_sys_sysctl(struct if (oldlenp) { if (!error) { if (get_user(oldlen, oldlenp) || - put_user(oldlen, (compat_size_t __user *)compat_ptr(tmp.oldlenp))) + put_user(oldlen, (compat_size_t __user *)compat_ptr(tmp.oldlenp)) || + copy_to_user(args->__unused, tmp.__unused, sizeof(tmp.__unused))) error = -EFAULT; } - copy_to_user(args->__unused, tmp.__unused, sizeof(tmp.__unused)); } return error; } diff -urNp linux-2.6.32.48/arch/powerpc/kernel/traps.c linux-2.6.32.48-openvz/arch/powerpc/kernel/traps.c --- linux-2.6.32.48/arch/powerpc/kernel/traps.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/kernel/traps.c 2011-11-17 20:09:04.000000000 -0500 @@ -99,6 +99,8 @@ static void pmac_backlight_unblank(void) static inline void pmac_backlight_unblank(void) { } #endif +extern void gr_handle_kernel_exploit(void); + int die(const char *str, struct pt_regs *regs, long err) { static struct { @@ -168,6 +170,8 @@ int die(const char *str, struct pt_regs if (panic_on_oops) panic("Fatal exception"); + gr_handle_kernel_exploit(); + oops_exit(); do_exit(err); diff -urNp linux-2.6.32.48/arch/powerpc/kernel/vdso.c linux-2.6.32.48-openvz/arch/powerpc/kernel/vdso.c --- linux-2.6.32.48/arch/powerpc/kernel/vdso.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/kernel/vdso.c 2011-11-17 20:08:53.000000000 -0500 @@ -36,6 +36,7 @@ #include #include #include +#include #include "setup.h" @@ -220,7 +221,7 @@ int arch_setup_additional_pages(struct l vdso_base = VDSO32_MBASE; #endif - current->mm->context.vdso_base = 0; + current->mm->context.vdso_base = ~0UL; /* vDSO has a problem and was disabled, just don't "enable" it for the * process @@ -240,7 +241,7 @@ int arch_setup_additional_pages(struct l vdso_base = get_unmapped_area(NULL, vdso_base, (vdso_pages << PAGE_SHIFT) + ((VDSO_ALIGNMENT - 1) & PAGE_MASK), - 0, 0); + 0, MAP_PRIVATE | MAP_EXECUTABLE); if (IS_ERR_VALUE(vdso_base)) { rc = vdso_base; goto fail_mmapsem; diff -urNp linux-2.6.32.48/arch/powerpc/kernel/vio.c linux-2.6.32.48-openvz/arch/powerpc/kernel/vio.c --- linux-2.6.32.48/arch/powerpc/kernel/vio.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/kernel/vio.c 2011-11-17 20:08:53.000000000 -0500 @@ -601,11 +601,12 @@ static void vio_dma_iommu_unmap_sg(struc vio_cmo_dealloc(viodev, alloc_size); } -struct dma_map_ops vio_dma_mapping_ops = { +static const struct dma_map_ops vio_dma_mapping_ops = { .alloc_coherent = vio_dma_iommu_alloc_coherent, .free_coherent = vio_dma_iommu_free_coherent, .map_sg = vio_dma_iommu_map_sg, .unmap_sg = vio_dma_iommu_unmap_sg, + .dma_supported = dma_iommu_dma_supported, .map_page = vio_dma_iommu_map_page, .unmap_page = vio_dma_iommu_unmap_page, @@ -857,7 +858,6 @@ static void vio_cmo_bus_remove(struct vi static void vio_cmo_set_dma_ops(struct vio_dev *viodev) { - vio_dma_mapping_ops.dma_supported = dma_iommu_ops.dma_supported; viodev->dev.archdata.dma_ops = &vio_dma_mapping_ops; } diff -urNp linux-2.6.32.48/arch/powerpc/lib/usercopy_64.c linux-2.6.32.48-openvz/arch/powerpc/lib/usercopy_64.c --- linux-2.6.32.48/arch/powerpc/lib/usercopy_64.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/lib/usercopy_64.c 2011-11-17 20:08:53.000000000 -0500 @@ -9,22 +9,6 @@ #include #include -unsigned long copy_from_user(void *to, const void __user *from, unsigned long n) -{ - if (likely(access_ok(VERIFY_READ, from, n))) - n = __copy_from_user(to, from, n); - else - memset(to, 0, n); - return n; -} - -unsigned long copy_to_user(void __user *to, const void *from, unsigned long n) -{ - if (likely(access_ok(VERIFY_WRITE, to, n))) - n = __copy_to_user(to, from, n); - return n; -} - unsigned long copy_in_user(void __user *to, const void __user *from, unsigned long n) { @@ -35,7 +19,5 @@ unsigned long copy_in_user(void __user * return n; } -EXPORT_SYMBOL(copy_from_user); -EXPORT_SYMBOL(copy_to_user); EXPORT_SYMBOL(copy_in_user); diff -urNp linux-2.6.32.48/arch/powerpc/Makefile linux-2.6.32.48-openvz/arch/powerpc/Makefile --- linux-2.6.32.48/arch/powerpc/Makefile 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/Makefile 2011-11-17 20:09:04.000000000 -0500 @@ -74,6 +74,8 @@ KBUILD_AFLAGS += -Iarch/$(ARCH) KBUILD_CFLAGS += -msoft-float -pipe -Iarch/$(ARCH) $(CFLAGS-y) CPP = $(CC) -E $(KBUILD_CFLAGS) +cflags-y += -Wno-sign-compare -Wno-extra + CHECKFLAGS += -m$(CONFIG_WORD_SIZE) -D__powerpc__ -D__powerpc$(CONFIG_WORD_SIZE)__ ifeq ($(CONFIG_PPC64),y) diff -urNp linux-2.6.32.48/arch/powerpc/mm/fault.c linux-2.6.32.48-openvz/arch/powerpc/mm/fault.c --- linux-2.6.32.48/arch/powerpc/mm/fault.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/mm/fault.c 2011-11-17 20:08:53.000000000 -0500 @@ -30,6 +30,10 @@ #include #include #include +#include +#include +#include +#include #include #include @@ -40,6 +44,7 @@ #include #include #include +#include #ifdef CONFIG_KPROBES @@ -64,6 +69,33 @@ static inline int notify_page_fault(stru } #endif +#ifdef CONFIG_PAX_PAGEEXEC +/* + * PaX: decide what to do with offenders (regs->nip = fault address) + * + * returns 1 when task should be killed + */ +static int pax_handle_fetch_fault(struct pt_regs *regs) +{ + return 1; +} + +void pax_report_insns(void *pc, void *sp) +{ + unsigned long i; + + printk(KERN_ERR "PAX: bytes at PC: "); + for (i = 0; i < 5; i++) { + unsigned int c; + if (get_user(c, (unsigned int __user *)pc+i)) + printk(KERN_CONT "???????? "); + else + printk(KERN_CONT "%08x ", c); + } + printk("\n"); +} +#endif + /* * Check whether the instruction at regs->nip is a store using * an update addressing form which will update r1. @@ -134,7 +166,7 @@ int __kprobes do_page_fault(struct pt_re * indicate errors in DSISR but can validly be set in SRR1. */ if (trap == 0x400) - error_code &= 0x48200000; + error_code &= 0x58200000; else is_write = error_code & DSISR_ISSTORE; #else @@ -250,7 +282,7 @@ good_area: * "undefined". Of those that can be set, this is the only * one which seems bad. */ - if (error_code & 0x10000000) + if (error_code & DSISR_GUARDED) /* Guarded storage error. */ goto bad_area; #endif /* CONFIG_8xx */ @@ -265,7 +297,7 @@ good_area: * processors use the same I/D cache coherency mechanism * as embedded. */ - if (error_code & DSISR_PROTFAULT) + if (error_code & (DSISR_PROTFAULT | DSISR_GUARDED)) goto bad_area; #endif /* CONFIG_PPC_STD_MMU */ @@ -335,6 +367,23 @@ bad_area: bad_area_nosemaphore: /* User mode accesses cause a SIGSEGV */ if (user_mode(regs)) { + +#ifdef CONFIG_PAX_PAGEEXEC + if (mm->pax_flags & MF_PAX_PAGEEXEC) { +#ifdef CONFIG_PPC_STD_MMU + if (is_exec && (error_code & (DSISR_PROTFAULT | DSISR_GUARDED))) { +#else + if (is_exec && regs->nip == address) { +#endif + switch (pax_handle_fetch_fault(regs)) { + } + + pax_report_fault(regs, (void *)regs->nip, (void *)regs->gpr[PT_R1]); + do_group_exit(SIGKILL); + } + } +#endif + _exception(SIGSEGV, regs, code, address); return 0; } diff -urNp linux-2.6.32.48/arch/powerpc/mm/mem.c linux-2.6.32.48-openvz/arch/powerpc/mm/mem.c --- linux-2.6.32.48/arch/powerpc/mm/mem.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/mm/mem.c 2011-11-17 20:09:04.000000000 -0500 @@ -250,7 +250,7 @@ static int __init mark_nonram_nosave(voi { unsigned long lmb_next_region_start_pfn, lmb_region_max_pfn; - int i; + unsigned int i; for (i = 0; i < lmb.memory.cnt - 1; i++) { lmb_region_max_pfn = diff -urNp linux-2.6.32.48/arch/powerpc/mm/mmap_64.c linux-2.6.32.48-openvz/arch/powerpc/mm/mmap_64.c --- linux-2.6.32.48/arch/powerpc/mm/mmap_64.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/mm/mmap_64.c 2011-11-17 20:08:53.000000000 -0500 @@ -99,10 +99,22 @@ void arch_pick_mmap_layout(struct mm_str */ if (mmap_is_legacy()) { mm->mmap_base = TASK_UNMAPPED_BASE; + +#ifdef CONFIG_PAX_RANDMMAP + if (mm->pax_flags & MF_PAX_RANDMMAP) + mm->mmap_base += mm->delta_mmap; +#endif + mm->get_unmapped_area = arch_get_unmapped_area; mm->unmap_area = arch_unmap_area; } else { mm->mmap_base = mmap_base(); + +#ifdef CONFIG_PAX_RANDMMAP + if (mm->pax_flags & MF_PAX_RANDMMAP) + mm->mmap_base -= mm->delta_mmap + mm->delta_stack; +#endif + mm->get_unmapped_area = arch_get_unmapped_area_topdown; mm->unmap_area = arch_unmap_area_topdown; } diff -urNp linux-2.6.32.48/arch/powerpc/mm/slice.c linux-2.6.32.48-openvz/arch/powerpc/mm/slice.c --- linux-2.6.32.48/arch/powerpc/mm/slice.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/mm/slice.c 2011-11-17 20:08:53.000000000 -0500 @@ -98,7 +98,7 @@ static int slice_area_is_free(struct mm_ if ((mm->task_size - len) < addr) return 0; vma = find_vma(mm, addr); - return (!vma || (addr + len) <= vma->vm_start); + return check_heap_stack_gap(vma, addr, len); } static int slice_low_has_vma(struct mm_struct *mm, unsigned long slice) @@ -256,7 +256,7 @@ full_search: addr = _ALIGN_UP(addr + 1, 1ul << SLICE_HIGH_SHIFT); continue; } - if (!vma || addr + len <= vma->vm_start) { + if (check_heap_stack_gap(vma, addr, len)) { /* * Remember the place where we stopped the search: */ @@ -313,10 +313,14 @@ static unsigned long slice_find_area_top } } - addr = mm->mmap_base; - while (addr > len) { + if (mm->mmap_base < len) + addr = -ENOMEM; + else + addr = mm->mmap_base - len; + + while (!IS_ERR_VALUE(addr)) { /* Go down by chunk size */ - addr = _ALIGN_DOWN(addr - len, 1ul << pshift); + addr = _ALIGN_DOWN(addr, 1ul << pshift); /* Check for hit with different page size */ mask = slice_range_to_mask(addr, len); @@ -336,7 +340,7 @@ static unsigned long slice_find_area_top * return with success: */ vma = find_vma(mm, addr); - if (!vma || (addr + len) <= vma->vm_start) { + if (check_heap_stack_gap(vma, addr, len)) { /* remember the address as a hint for next time */ if (use_cache) mm->free_area_cache = addr; @@ -348,7 +352,7 @@ static unsigned long slice_find_area_top mm->cached_hole_size = vma->vm_start - addr; /* try just below the current vma->vm_start */ - addr = vma->vm_start; + addr = skip_heap_stack_gap(vma, len); } /* @@ -426,6 +430,11 @@ unsigned long slice_get_unmapped_area(un if (fixed && addr > (mm->task_size - len)) return -EINVAL; +#ifdef CONFIG_PAX_RANDMMAP + if (!fixed && (mm->pax_flags & MF_PAX_RANDMMAP)) + addr = 0; +#endif + /* If hint, make sure it matches our alignment restrictions */ if (!fixed && addr) { addr = _ALIGN_UP(addr, 1ul << pshift); diff -urNp linux-2.6.32.48/arch/powerpc/platforms/52xx/lite5200_pm.c linux-2.6.32.48-openvz/arch/powerpc/platforms/52xx/lite5200_pm.c --- linux-2.6.32.48/arch/powerpc/platforms/52xx/lite5200_pm.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/platforms/52xx/lite5200_pm.c 2011-11-17 20:08:53.000000000 -0500 @@ -235,7 +235,7 @@ static void lite5200_pm_end(void) lite5200_pm_target_state = PM_SUSPEND_ON; } -static struct platform_suspend_ops lite5200_pm_ops = { +static const struct platform_suspend_ops lite5200_pm_ops = { .valid = lite5200_pm_valid, .begin = lite5200_pm_begin, .prepare = lite5200_pm_prepare, diff -urNp linux-2.6.32.48/arch/powerpc/platforms/52xx/mpc52xx_pm.c linux-2.6.32.48-openvz/arch/powerpc/platforms/52xx/mpc52xx_pm.c --- linux-2.6.32.48/arch/powerpc/platforms/52xx/mpc52xx_pm.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/platforms/52xx/mpc52xx_pm.c 2011-11-17 20:08:53.000000000 -0500 @@ -180,7 +180,7 @@ void mpc52xx_pm_finish(void) iounmap(mbar); } -static struct platform_suspend_ops mpc52xx_pm_ops = { +static const struct platform_suspend_ops mpc52xx_pm_ops = { .valid = mpc52xx_pm_valid, .prepare = mpc52xx_pm_prepare, .enter = mpc52xx_pm_enter, diff -urNp linux-2.6.32.48/arch/powerpc/platforms/83xx/suspend.c linux-2.6.32.48-openvz/arch/powerpc/platforms/83xx/suspend.c --- linux-2.6.32.48/arch/powerpc/platforms/83xx/suspend.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/platforms/83xx/suspend.c 2011-11-17 20:08:53.000000000 -0500 @@ -273,7 +273,7 @@ static int mpc83xx_is_pci_agent(void) return ret; } -static struct platform_suspend_ops mpc83xx_suspend_ops = { +static const struct platform_suspend_ops mpc83xx_suspend_ops = { .valid = mpc83xx_suspend_valid, .begin = mpc83xx_suspend_begin, .enter = mpc83xx_suspend_enter, diff -urNp linux-2.6.32.48/arch/powerpc/platforms/cell/iommu.c linux-2.6.32.48-openvz/arch/powerpc/platforms/cell/iommu.c --- linux-2.6.32.48/arch/powerpc/platforms/cell/iommu.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/platforms/cell/iommu.c 2011-11-17 20:08:53.000000000 -0500 @@ -642,7 +642,7 @@ static int dma_fixed_dma_supported(struc static int dma_set_mask_and_switch(struct device *dev, u64 dma_mask); -struct dma_map_ops dma_iommu_fixed_ops = { +const struct dma_map_ops dma_iommu_fixed_ops = { .alloc_coherent = dma_fixed_alloc_coherent, .free_coherent = dma_fixed_free_coherent, .map_sg = dma_fixed_map_sg, diff -urNp linux-2.6.32.48/arch/powerpc/platforms/ps3/system-bus.c linux-2.6.32.48-openvz/arch/powerpc/platforms/ps3/system-bus.c --- linux-2.6.32.48/arch/powerpc/platforms/ps3/system-bus.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/platforms/ps3/system-bus.c 2011-11-17 20:08:53.000000000 -0500 @@ -694,7 +694,7 @@ static int ps3_dma_supported(struct devi return mask >= DMA_BIT_MASK(32); } -static struct dma_map_ops ps3_sb_dma_ops = { +static const struct dma_map_ops ps3_sb_dma_ops = { .alloc_coherent = ps3_alloc_coherent, .free_coherent = ps3_free_coherent, .map_sg = ps3_sb_map_sg, @@ -704,7 +704,7 @@ static struct dma_map_ops ps3_sb_dma_ops .unmap_page = ps3_unmap_page, }; -static struct dma_map_ops ps3_ioc0_dma_ops = { +static const struct dma_map_ops ps3_ioc0_dma_ops = { .alloc_coherent = ps3_alloc_coherent, .free_coherent = ps3_free_coherent, .map_sg = ps3_ioc0_map_sg, diff -urNp linux-2.6.32.48/arch/powerpc/platforms/pseries/Kconfig linux-2.6.32.48-openvz/arch/powerpc/platforms/pseries/Kconfig --- linux-2.6.32.48/arch/powerpc/platforms/pseries/Kconfig 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/powerpc/platforms/pseries/Kconfig 2011-11-17 20:08:53.000000000 -0500 @@ -2,6 +2,8 @@ config PPC_PSERIES depends on PPC64 && PPC_BOOK3S bool "IBM pSeries & new (POWER5-based) iSeries" select MPIC + select PCI_MSI + select XICS select PPC_I8259 select PPC_RTAS select RTAS_ERROR_LOGGING diff -urNp linux-2.6.32.48/arch/s390/include/asm/elf.h linux-2.6.32.48-openvz/arch/s390/include/asm/elf.h --- linux-2.6.32.48/arch/s390/include/asm/elf.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/s390/include/asm/elf.h 2011-11-17 20:08:53.000000000 -0500 @@ -164,6 +164,13 @@ extern unsigned int vdso_enabled; that it will "exec", and that there is sufficient room for the brk. */ #define ELF_ET_DYN_BASE (STACK_TOP / 3 * 2) +#ifdef CONFIG_PAX_ASLR +#define PAX_ELF_ET_DYN_BASE (test_thread_flag(TIF_31BIT) ? 0x10000UL : 0x80000000UL) + +#define PAX_DELTA_MMAP_LEN (test_thread_flag(TIF_31BIT) ? 15 : 26 ) +#define PAX_DELTA_STACK_LEN (test_thread_flag(TIF_31BIT) ? 15 : 26 ) +#endif + /* This yields a mask that user programs can use to figure out what instruction set this CPU supports. */ diff -urNp linux-2.6.32.48/arch/s390/include/asm/setup.h linux-2.6.32.48-openvz/arch/s390/include/asm/setup.h --- linux-2.6.32.48/arch/s390/include/asm/setup.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/s390/include/asm/setup.h 2011-11-17 20:08:53.000000000 -0500 @@ -50,13 +50,13 @@ extern unsigned long memory_end; void detect_memory_layout(struct mem_chunk chunk[]); #ifdef CONFIG_S390_SWITCH_AMODE -extern unsigned int switch_amode; +#define switch_amode (1) #else #define switch_amode (0) #endif #ifdef CONFIG_S390_EXEC_PROTECT -extern unsigned int s390_noexec; +#define s390_noexec (1) #else #define s390_noexec (0) #endif diff -urNp linux-2.6.32.48/arch/s390/include/asm/uaccess.h linux-2.6.32.48-openvz/arch/s390/include/asm/uaccess.h --- linux-2.6.32.48/arch/s390/include/asm/uaccess.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/s390/include/asm/uaccess.h 2011-11-17 20:08:53.000000000 -0500 @@ -232,6 +232,10 @@ static inline unsigned long __must_check copy_to_user(void __user *to, const void *from, unsigned long n) { might_fault(); + + if ((long)n < 0) + return n; + if (access_ok(VERIFY_WRITE, to, n)) n = __copy_to_user(to, from, n); return n; @@ -257,6 +261,9 @@ copy_to_user(void __user *to, const void static inline unsigned long __must_check __copy_from_user(void *to, const void __user *from, unsigned long n) { + if ((long)n < 0) + return n; + if (__builtin_constant_p(n) && (n <= 256)) return uaccess.copy_from_user_small(n, from, to); else @@ -283,6 +290,10 @@ static inline unsigned long __must_check copy_from_user(void *to, const void __user *from, unsigned long n) { might_fault(); + + if ((long)n < 0) + return n; + if (access_ok(VERIFY_READ, from, n)) n = __copy_from_user(to, from, n); else diff -urNp linux-2.6.32.48/arch/s390/Kconfig linux-2.6.32.48-openvz/arch/s390/Kconfig --- linux-2.6.32.48/arch/s390/Kconfig 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/s390/Kconfig 2011-11-17 20:08:53.000000000 -0500 @@ -194,28 +194,26 @@ config AUDIT_ARCH config S390_SWITCH_AMODE bool "Switch kernel/user addressing modes" + default y help This option allows to switch the addressing modes of kernel and user - space. The kernel parameter switch_amode=on will enable this feature, - default is disabled. Enabling this (via kernel parameter) on machines - earlier than IBM System z9-109 EC/BC will reduce system performance. + space. Enabling this on machines earlier than IBM System z9-109 EC/BC + will reduce system performance. Note that this option will also be selected by selecting the execute - protection option below. Enabling the execute protection via the - noexec kernel parameter will also switch the addressing modes, - independent of the switch_amode kernel parameter. + protection option below. Enabling the execute protection will also + switch the addressing modes, independent of this option. config S390_EXEC_PROTECT bool "Data execute protection" + default y select S390_SWITCH_AMODE help This option allows to enable a buffer overflow protection for user space programs and it also selects the addressing mode option above. - The kernel parameter noexec=on will enable this feature and also - switch the addressing modes, default is disabled. Enabling this (via - kernel parameter) on machines earlier than IBM System z9-109 EC/BC - will reduce system performance. + Enabling this on machines earlier than IBM System z9-109 EC/BC will + reduce system performance. comment "Code generation options" diff -urNp linux-2.6.32.48/arch/s390/kernel/module.c linux-2.6.32.48-openvz/arch/s390/kernel/module.c --- linux-2.6.32.48/arch/s390/kernel/module.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/s390/kernel/module.c 2011-11-17 20:08:53.000000000 -0500 @@ -166,11 +166,11 @@ module_frob_arch_sections(Elf_Ehdr *hdr, /* Increase core size by size of got & plt and set start offsets for got and plt. */ - me->core_size = ALIGN(me->core_size, 4); - me->arch.got_offset = me->core_size; - me->core_size += me->arch.got_size; - me->arch.plt_offset = me->core_size; - me->core_size += me->arch.plt_size; + me->core_size_rw = ALIGN(me->core_size_rw, 4); + me->arch.got_offset = me->core_size_rw; + me->core_size_rw += me->arch.got_size; + me->arch.plt_offset = me->core_size_rx; + me->core_size_rx += me->arch.plt_size; return 0; } @@ -256,7 +256,7 @@ apply_rela(Elf_Rela *rela, Elf_Addr base if (info->got_initialized == 0) { Elf_Addr *gotent; - gotent = me->module_core + me->arch.got_offset + + gotent = me->module_core_rw + me->arch.got_offset + info->got_offset; *gotent = val; info->got_initialized = 1; @@ -280,7 +280,7 @@ apply_rela(Elf_Rela *rela, Elf_Addr base else if (r_type == R_390_GOTENT || r_type == R_390_GOTPLTENT) *(unsigned int *) loc = - (val + (Elf_Addr) me->module_core - loc) >> 1; + (val + (Elf_Addr) me->module_core_rw - loc) >> 1; else if (r_type == R_390_GOT64 || r_type == R_390_GOTPLT64) *(unsigned long *) loc = val; @@ -294,7 +294,7 @@ apply_rela(Elf_Rela *rela, Elf_Addr base case R_390_PLTOFF64: /* 16 bit offset from GOT to PLT. */ if (info->plt_initialized == 0) { unsigned int *ip; - ip = me->module_core + me->arch.plt_offset + + ip = me->module_core_rx + me->arch.plt_offset + info->plt_offset; #ifndef CONFIG_64BIT ip[0] = 0x0d105810; /* basr 1,0; l 1,6(1); br 1 */ @@ -319,7 +319,7 @@ apply_rela(Elf_Rela *rela, Elf_Addr base val - loc + 0xffffUL < 0x1ffffeUL) || (r_type == R_390_PLT32DBL && val - loc + 0xffffffffULL < 0x1fffffffeULL))) - val = (Elf_Addr) me->module_core + + val = (Elf_Addr) me->module_core_rx + me->arch.plt_offset + info->plt_offset; val += rela->r_addend - loc; @@ -341,7 +341,7 @@ apply_rela(Elf_Rela *rela, Elf_Addr base case R_390_GOTOFF32: /* 32 bit offset to GOT. */ case R_390_GOTOFF64: /* 64 bit offset to GOT. */ val = val + rela->r_addend - - ((Elf_Addr) me->module_core + me->arch.got_offset); + ((Elf_Addr) me->module_core_rw + me->arch.got_offset); if (r_type == R_390_GOTOFF16) *(unsigned short *) loc = val; else if (r_type == R_390_GOTOFF32) @@ -351,7 +351,7 @@ apply_rela(Elf_Rela *rela, Elf_Addr base break; case R_390_GOTPC: /* 32 bit PC relative offset to GOT. */ case R_390_GOTPCDBL: /* 32 bit PC rel. off. to GOT shifted by 1. */ - val = (Elf_Addr) me->module_core + me->arch.got_offset + + val = (Elf_Addr) me->module_core_rw + me->arch.got_offset + rela->r_addend - loc; if (r_type == R_390_GOTPC) *(unsigned int *) loc = val; diff -urNp linux-2.6.32.48/arch/s390/kernel/setup.c linux-2.6.32.48-openvz/arch/s390/kernel/setup.c --- linux-2.6.32.48/arch/s390/kernel/setup.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/s390/kernel/setup.c 2011-11-17 20:08:53.000000000 -0500 @@ -306,9 +306,6 @@ static int __init early_parse_mem(char * early_param("mem", early_parse_mem); #ifdef CONFIG_S390_SWITCH_AMODE -unsigned int switch_amode = 0; -EXPORT_SYMBOL_GPL(switch_amode); - static int set_amode_and_uaccess(unsigned long user_amode, unsigned long user32_amode) { @@ -334,17 +331,6 @@ static int set_amode_and_uaccess(unsigne return 0; } } - -/* - * Switch kernel/user addressing modes? - */ -static int __init early_parse_switch_amode(char *p) -{ - switch_amode = 1; - return 0; -} -early_param("switch_amode", early_parse_switch_amode); - #else /* CONFIG_S390_SWITCH_AMODE */ static inline int set_amode_and_uaccess(unsigned long user_amode, unsigned long user32_amode) @@ -353,24 +339,6 @@ static inline int set_amode_and_uaccess( } #endif /* CONFIG_S390_SWITCH_AMODE */ -#ifdef CONFIG_S390_EXEC_PROTECT -unsigned int s390_noexec = 0; -EXPORT_SYMBOL_GPL(s390_noexec); - -/* - * Enable execute protection? - */ -static int __init early_parse_noexec(char *p) -{ - if (!strncmp(p, "off", 3)) - return 0; - switch_amode = 1; - s390_noexec = 1; - return 0; -} -early_param("noexec", early_parse_noexec); -#endif /* CONFIG_S390_EXEC_PROTECT */ - static void setup_addressing_mode(void) { if (s390_noexec) { diff -urNp linux-2.6.32.48/arch/s390/mm/mmap.c linux-2.6.32.48-openvz/arch/s390/mm/mmap.c --- linux-2.6.32.48/arch/s390/mm/mmap.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/s390/mm/mmap.c 2011-11-17 20:08:53.000000000 -0500 @@ -78,10 +78,22 @@ void arch_pick_mmap_layout(struct mm_str */ if (mmap_is_legacy()) { mm->mmap_base = TASK_UNMAPPED_BASE; + +#ifdef CONFIG_PAX_RANDMMAP + if (mm->pax_flags & MF_PAX_RANDMMAP) + mm->mmap_base += mm->delta_mmap; +#endif + mm->get_unmapped_area = arch_get_unmapped_area; mm->unmap_area = arch_unmap_area; } else { mm->mmap_base = mmap_base(); + +#ifdef CONFIG_PAX_RANDMMAP + if (mm->pax_flags & MF_PAX_RANDMMAP) + mm->mmap_base -= mm->delta_mmap + mm->delta_stack; +#endif + mm->get_unmapped_area = arch_get_unmapped_area_topdown; mm->unmap_area = arch_unmap_area_topdown; } @@ -153,10 +165,22 @@ void arch_pick_mmap_layout(struct mm_str */ if (mmap_is_legacy()) { mm->mmap_base = TASK_UNMAPPED_BASE; + +#ifdef CONFIG_PAX_RANDMMAP + if (mm->pax_flags & MF_PAX_RANDMMAP) + mm->mmap_base += mm->delta_mmap; +#endif + mm->get_unmapped_area = s390_get_unmapped_area; mm->unmap_area = arch_unmap_area; } else { mm->mmap_base = mmap_base(); + +#ifdef CONFIG_PAX_RANDMMAP + if (mm->pax_flags & MF_PAX_RANDMMAP) + mm->mmap_base -= mm->delta_mmap + mm->delta_stack; +#endif + mm->get_unmapped_area = s390_get_unmapped_area_topdown; mm->unmap_area = arch_unmap_area_topdown; } diff -urNp linux-2.6.32.48/arch/score/include/asm/system.h linux-2.6.32.48-openvz/arch/score/include/asm/system.h --- linux-2.6.32.48/arch/score/include/asm/system.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/score/include/asm/system.h 2011-11-17 20:08:53.000000000 -0500 @@ -17,7 +17,7 @@ do { \ #define finish_arch_switch(prev) do {} while (0) typedef void (*vi_handler_t)(void); -extern unsigned long arch_align_stack(unsigned long sp); +#define arch_align_stack(x) (x) #define mb() barrier() #define rmb() barrier() diff -urNp linux-2.6.32.48/arch/score/kernel/process.c linux-2.6.32.48-openvz/arch/score/kernel/process.c --- linux-2.6.32.48/arch/score/kernel/process.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/score/kernel/process.c 2011-11-17 20:08:53.000000000 -0500 @@ -161,8 +161,3 @@ unsigned long get_wchan(struct task_stru return task_pt_regs(task)->cp0_epc; } - -unsigned long arch_align_stack(unsigned long sp) -{ - return sp; -} diff -urNp linux-2.6.32.48/arch/sh/boards/mach-hp6xx/pm.c linux-2.6.32.48-openvz/arch/sh/boards/mach-hp6xx/pm.c --- linux-2.6.32.48/arch/sh/boards/mach-hp6xx/pm.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sh/boards/mach-hp6xx/pm.c 2011-11-17 20:08:53.000000000 -0500 @@ -143,7 +143,7 @@ static int hp6x0_pm_enter(suspend_state_ return 0; } -static struct platform_suspend_ops hp6x0_pm_ops = { +static const struct platform_suspend_ops hp6x0_pm_ops = { .enter = hp6x0_pm_enter, .valid = suspend_valid_only_mem, }; diff -urNp linux-2.6.32.48/arch/sh/kernel/cpu/sh4/sq.c linux-2.6.32.48-openvz/arch/sh/kernel/cpu/sh4/sq.c --- linux-2.6.32.48/arch/sh/kernel/cpu/sh4/sq.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sh/kernel/cpu/sh4/sq.c 2011-11-17 20:08:53.000000000 -0500 @@ -327,7 +327,7 @@ static struct attribute *sq_sysfs_attrs[ NULL, }; -static struct sysfs_ops sq_sysfs_ops = { +static const struct sysfs_ops sq_sysfs_ops = { .show = sq_sysfs_show, .store = sq_sysfs_store, }; diff -urNp linux-2.6.32.48/arch/sh/kernel/cpu/shmobile/pm.c linux-2.6.32.48-openvz/arch/sh/kernel/cpu/shmobile/pm.c --- linux-2.6.32.48/arch/sh/kernel/cpu/shmobile/pm.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sh/kernel/cpu/shmobile/pm.c 2011-11-17 20:08:53.000000000 -0500 @@ -58,7 +58,7 @@ static int sh_pm_enter(suspend_state_t s return 0; } -static struct platform_suspend_ops sh_pm_ops = { +static const struct platform_suspend_ops sh_pm_ops = { .enter = sh_pm_enter, .valid = suspend_valid_only_mem, }; diff -urNp linux-2.6.32.48/arch/sh/kernel/kgdb.c linux-2.6.32.48-openvz/arch/sh/kernel/kgdb.c --- linux-2.6.32.48/arch/sh/kernel/kgdb.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sh/kernel/kgdb.c 2011-11-17 20:08:53.000000000 -0500 @@ -271,7 +271,7 @@ void kgdb_arch_exit(void) { } -struct kgdb_arch arch_kgdb_ops = { +const struct kgdb_arch arch_kgdb_ops = { /* Breakpoint instruction: trapa #0x3c */ #ifdef CONFIG_CPU_LITTLE_ENDIAN .gdb_bpt_instr = { 0x3c, 0xc3 }, diff -urNp linux-2.6.32.48/arch/sh/mm/mmap.c linux-2.6.32.48-openvz/arch/sh/mm/mmap.c --- linux-2.6.32.48/arch/sh/mm/mmap.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sh/mm/mmap.c 2011-11-17 20:08:53.000000000 -0500 @@ -74,8 +74,7 @@ unsigned long arch_get_unmapped_area(str addr = PAGE_ALIGN(addr); vma = find_vma(mm, addr); - if (TASK_SIZE - len >= addr && - (!vma || addr + len <= vma->vm_start)) + if (TASK_SIZE - len >= addr && check_heap_stack_gap(vma, addr, len)) return addr; } @@ -106,7 +105,7 @@ full_search: } return -ENOMEM; } - if (likely(!vma || addr + len <= vma->vm_start)) { + if (likely(check_heap_stack_gap(vma, addr, len))) { /* * Remember the place where we stopped the search: */ @@ -157,8 +156,7 @@ arch_get_unmapped_area_topdown(struct fi addr = PAGE_ALIGN(addr); vma = find_vma(mm, addr); - if (TASK_SIZE - len >= addr && - (!vma || addr + len <= vma->vm_start)) + if (TASK_SIZE - len >= addr && check_heap_stack_gap(vma, addr, len)) return addr; } @@ -179,7 +177,7 @@ arch_get_unmapped_area_topdown(struct fi /* make sure it can fit in the remaining address space */ if (likely(addr > len)) { vma = find_vma(mm, addr-len); - if (!vma || addr <= vma->vm_start) { + if (check_heap_stack_gap(vma, addr - len, len)) { /* remember the address as a hint for next time */ return (mm->free_area_cache = addr-len); } @@ -188,18 +186,18 @@ arch_get_unmapped_area_topdown(struct fi if (unlikely(mm->mmap_base < len)) goto bottomup; - addr = mm->mmap_base-len; - if (do_colour_align) - addr = COLOUR_ALIGN_DOWN(addr, pgoff); + addr = mm->mmap_base - len; do { + if (do_colour_align) + addr = COLOUR_ALIGN_DOWN(addr, pgoff); /* * Lookup failure means no vma is above this address, * else if new region fits below vma->vm_start, * return with success: */ vma = find_vma(mm, addr); - if (likely(!vma || addr+len <= vma->vm_start)) { + if (likely(check_heap_stack_gap(vma, addr, len))) { /* remember the address as a hint for next time */ return (mm->free_area_cache = addr); } @@ -209,10 +207,8 @@ arch_get_unmapped_area_topdown(struct fi mm->cached_hole_size = vma->vm_start - addr; /* try just below the current vma->vm_start */ - addr = vma->vm_start-len; - if (do_colour_align) - addr = COLOUR_ALIGN_DOWN(addr, pgoff); - } while (likely(len < vma->vm_start)); + addr = skip_heap_stack_gap(vma, len); + } while (!IS_ERR_VALUE(addr)); bottomup: /* diff -urNp linux-2.6.32.48/arch/sparc/include/asm/atomic_64.h linux-2.6.32.48-openvz/arch/sparc/include/asm/atomic_64.h --- linux-2.6.32.48/arch/sparc/include/asm/atomic_64.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/include/asm/atomic_64.h 2011-11-17 20:09:04.000000000 -0500 @@ -14,18 +14,40 @@ #define ATOMIC64_INIT(i) { (i) } #define atomic_read(v) ((v)->counter) +static inline int atomic_read_unchecked(const atomic_unchecked_t *v) +{ + return v->counter; +} #define atomic64_read(v) ((v)->counter) +static inline long atomic64_read_unchecked(const atomic64_unchecked_t *v) +{ + return v->counter; +} #define atomic_set(v, i) (((v)->counter) = i) +static inline void atomic_set_unchecked(atomic_unchecked_t *v, int i) +{ + v->counter = i; +} #define atomic64_set(v, i) (((v)->counter) = i) +static inline void atomic64_set_unchecked(atomic64_unchecked_t *v, long i) +{ + v->counter = i; +} extern void atomic_add(int, atomic_t *); +extern void atomic_add_unchecked(int, atomic_unchecked_t *); extern void atomic64_add(long, atomic64_t *); +extern void atomic64_add_unchecked(long, atomic64_unchecked_t *); extern void atomic_sub(int, atomic_t *); +extern void atomic_sub_unchecked(int, atomic_unchecked_t *); extern void atomic64_sub(long, atomic64_t *); +extern void atomic64_sub_unchecked(long, atomic64_unchecked_t *); extern int atomic_add_ret(int, atomic_t *); +extern int atomic_add_ret_unchecked(int, atomic_unchecked_t *); extern long atomic64_add_ret(long, atomic64_t *); +extern long atomic64_add_ret_unchecked(long, atomic64_unchecked_t *); extern int atomic_sub_ret(int, atomic_t *); extern long atomic64_sub_ret(long, atomic64_t *); @@ -33,13 +55,29 @@ extern long atomic64_sub_ret(long, atomi #define atomic64_dec_return(v) atomic64_sub_ret(1, v) #define atomic_inc_return(v) atomic_add_ret(1, v) +static inline int atomic_inc_return_unchecked(atomic_unchecked_t *v) +{ + return atomic_add_ret_unchecked(1, v); +} #define atomic64_inc_return(v) atomic64_add_ret(1, v) +static inline long atomic64_inc_return_unchecked(atomic64_unchecked_t *v) +{ + return atomic64_add_ret_unchecked(1, v); +} #define atomic_sub_return(i, v) atomic_sub_ret(i, v) #define atomic64_sub_return(i, v) atomic64_sub_ret(i, v) #define atomic_add_return(i, v) atomic_add_ret(i, v) +static inline int atomic_add_return_unchecked(int i, atomic_unchecked_t *v) +{ + return atomic_add_ret_unchecked(i, v); +} #define atomic64_add_return(i, v) atomic64_add_ret(i, v) +static inline long atomic64_add_return_unchecked(long i, atomic64_unchecked_t *v) +{ + return atomic64_add_ret_unchecked(i, v); +} /* * atomic_inc_and_test - increment and test @@ -50,6 +88,10 @@ extern long atomic64_sub_ret(long, atomi * other cases. */ #define atomic_inc_and_test(v) (atomic_inc_return(v) == 0) +static inline int atomic_inc_and_test_unchecked(atomic_unchecked_t *v) +{ + return atomic_inc_return_unchecked(v) == 0; +} #define atomic64_inc_and_test(v) (atomic64_inc_return(v) == 0) #define atomic_sub_and_test(i, v) (atomic_sub_ret(i, v) == 0) @@ -59,30 +101,65 @@ extern long atomic64_sub_ret(long, atomi #define atomic64_dec_and_test(v) (atomic64_sub_ret(1, v) == 0) #define atomic_inc(v) atomic_add(1, v) +static inline void atomic_inc_unchecked(atomic_unchecked_t *v) +{ + atomic_add_unchecked(1, v); +} #define atomic64_inc(v) atomic64_add(1, v) +static inline void atomic64_inc_unchecked(atomic64_unchecked_t *v) +{ + atomic64_add_unchecked(1, v); +} #define atomic_dec(v) atomic_sub(1, v) +static inline void atomic_dec_unchecked(atomic_unchecked_t *v) +{ + atomic_sub_unchecked(1, v); +} #define atomic64_dec(v) atomic64_sub(1, v) +static inline void atomic64_dec_unchecked(atomic64_unchecked_t *v) +{ + atomic64_sub_unchecked(1, v); +} #define atomic_add_negative(i, v) (atomic_add_ret(i, v) < 0) #define atomic64_add_negative(i, v) (atomic64_add_ret(i, v) < 0) #define atomic_cmpxchg(v, o, n) (cmpxchg(&((v)->counter), (o), (n))) +static inline int atomic_cmpxchg_unchecked(atomic_unchecked_t *v, int old, int new) +{ + return cmpxchg(&v->counter, old, new); +} #define atomic_xchg(v, new) (xchg(&((v)->counter), new)) +static inline int atomic_xchg_unchecked(atomic_unchecked_t *v, int new) +{ + return xchg(&v->counter, new); +} static inline int atomic_add_unless(atomic_t *v, int a, int u) { - int c, old; + int c, old, new; c = atomic_read(v); for (;;) { - if (unlikely(c == (u))) + if (unlikely(c == u)) break; - old = atomic_cmpxchg((v), c, c + (a)); + + asm volatile("addcc %2, %0, %0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "tvs %%icc, 6\n" +#endif + + : "=r" (new) + : "0" (c), "ir" (a) + : "cc"); + + old = atomic_cmpxchg(v, c, new); if (likely(old == c)) break; c = old; } - return c != (u); + return c != u; } #define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0) @@ -90,20 +167,35 @@ static inline int atomic_add_unless(atom #define atomic64_cmpxchg(v, o, n) \ ((__typeof__((v)->counter))cmpxchg(&((v)->counter), (o), (n))) #define atomic64_xchg(v, new) (xchg(&((v)->counter), new)) +static inline long atomic64_xchg_unchecked(atomic64_unchecked_t *v, long new) +{ + return xchg(&v->counter, new); +} static inline long atomic64_add_unless(atomic64_t *v, long a, long u) { - long c, old; + long c, old, new; c = atomic64_read(v); for (;;) { - if (unlikely(c == (u))) + if (unlikely(c == u)) break; - old = atomic64_cmpxchg((v), c, c + (a)); + + asm volatile("addcc %2, %0, %0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "tvs %%xcc, 6\n" +#endif + + : "=r" (new) + : "0" (c), "ir" (a) + : "cc"); + + old = atomic64_cmpxchg(v, c, new); if (likely(old == c)) break; c = old; } - return c != (u); + return c != u; } #define atomic64_inc_not_zero(v) atomic64_add_unless((v), 1, 0) diff -urNp linux-2.6.32.48/arch/sparc/include/asm/cache.h linux-2.6.32.48-openvz/arch/sparc/include/asm/cache.h --- linux-2.6.32.48/arch/sparc/include/asm/cache.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/include/asm/cache.h 2011-11-17 20:09:04.000000000 -0500 @@ -8,7 +8,7 @@ #define _SPARC_CACHE_H #define L1_CACHE_SHIFT 5 -#define L1_CACHE_BYTES 32 +#define L1_CACHE_BYTES 32UL #define L1_CACHE_ALIGN(x) ((((x)+(L1_CACHE_BYTES-1))&~(L1_CACHE_BYTES-1))) #ifdef CONFIG_SPARC32 diff -urNp linux-2.6.32.48/arch/sparc/include/asm/dma-mapping.h linux-2.6.32.48-openvz/arch/sparc/include/asm/dma-mapping.h --- linux-2.6.32.48/arch/sparc/include/asm/dma-mapping.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/include/asm/dma-mapping.h 2011-11-17 20:08:53.000000000 -0500 @@ -14,10 +14,10 @@ extern int dma_set_mask(struct device *d #define dma_free_noncoherent(d, s, v, h) dma_free_coherent(d, s, v, h) #define dma_is_consistent(d, h) (1) -extern struct dma_map_ops *dma_ops, pci32_dma_ops; +extern const struct dma_map_ops *dma_ops, pci32_dma_ops; extern struct bus_type pci_bus_type; -static inline struct dma_map_ops *get_dma_ops(struct device *dev) +static inline const struct dma_map_ops *get_dma_ops(struct device *dev) { #if defined(CONFIG_SPARC32) && defined(CONFIG_PCI) if (dev->bus == &pci_bus_type) @@ -31,7 +31,7 @@ static inline struct dma_map_ops *get_dm static inline void *dma_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_handle, gfp_t flag) { - struct dma_map_ops *ops = get_dma_ops(dev); + const struct dma_map_ops *ops = get_dma_ops(dev); void *cpu_addr; cpu_addr = ops->alloc_coherent(dev, size, dma_handle, flag); @@ -42,7 +42,7 @@ static inline void *dma_alloc_coherent(s static inline void dma_free_coherent(struct device *dev, size_t size, void *cpu_addr, dma_addr_t dma_handle) { - struct dma_map_ops *ops = get_dma_ops(dev); + const struct dma_map_ops *ops = get_dma_ops(dev); debug_dma_free_coherent(dev, size, cpu_addr, dma_handle); ops->free_coherent(dev, size, cpu_addr, dma_handle); diff -urNp linux-2.6.32.48/arch/sparc/include/asm/elf_32.h linux-2.6.32.48-openvz/arch/sparc/include/asm/elf_32.h --- linux-2.6.32.48/arch/sparc/include/asm/elf_32.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/include/asm/elf_32.h 2011-11-17 20:08:53.000000000 -0500 @@ -116,6 +116,13 @@ typedef struct { #define ELF_ET_DYN_BASE (TASK_UNMAPPED_BASE) +#ifdef CONFIG_PAX_ASLR +#define PAX_ELF_ET_DYN_BASE 0x10000UL + +#define PAX_DELTA_MMAP_LEN 16 +#define PAX_DELTA_STACK_LEN 16 +#endif + /* This yields a mask that user programs can use to figure out what instruction set this cpu supports. This can NOT be done in userspace on Sparc. */ diff -urNp linux-2.6.32.48/arch/sparc/include/asm/elf_64.h linux-2.6.32.48-openvz/arch/sparc/include/asm/elf_64.h --- linux-2.6.32.48/arch/sparc/include/asm/elf_64.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/include/asm/elf_64.h 2011-11-17 20:08:53.000000000 -0500 @@ -163,6 +163,12 @@ typedef struct { #define ELF_ET_DYN_BASE 0x0000010000000000UL #define COMPAT_ELF_ET_DYN_BASE 0x0000000070000000UL +#ifdef CONFIG_PAX_ASLR +#define PAX_ELF_ET_DYN_BASE (test_thread_flag(TIF_32BIT) ? 0x10000UL : 0x100000UL) + +#define PAX_DELTA_MMAP_LEN (test_thread_flag(TIF_32BIT) ? 14 : 28) +#define PAX_DELTA_STACK_LEN (test_thread_flag(TIF_32BIT) ? 15 : 29) +#endif /* This yields a mask that user programs can use to figure out what instruction set this cpu supports. */ diff -urNp linux-2.6.32.48/arch/sparc/include/asm/pgtable_32.h linux-2.6.32.48-openvz/arch/sparc/include/asm/pgtable_32.h --- linux-2.6.32.48/arch/sparc/include/asm/pgtable_32.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/include/asm/pgtable_32.h 2011-11-17 20:08:53.000000000 -0500 @@ -43,6 +43,13 @@ BTFIXUPDEF_SIMM13(user_ptrs_per_pgd) BTFIXUPDEF_INT(page_none) BTFIXUPDEF_INT(page_copy) BTFIXUPDEF_INT(page_readonly) + +#ifdef CONFIG_PAX_PAGEEXEC +BTFIXUPDEF_INT(page_shared_noexec) +BTFIXUPDEF_INT(page_copy_noexec) +BTFIXUPDEF_INT(page_readonly_noexec) +#endif + BTFIXUPDEF_INT(page_kernel) #define PMD_SHIFT SUN4C_PMD_SHIFT @@ -64,6 +71,16 @@ extern pgprot_t PAGE_SHARED; #define PAGE_COPY __pgprot(BTFIXUP_INT(page_copy)) #define PAGE_READONLY __pgprot(BTFIXUP_INT(page_readonly)) +#ifdef CONFIG_PAX_PAGEEXEC +extern pgprot_t PAGE_SHARED_NOEXEC; +# define PAGE_COPY_NOEXEC __pgprot(BTFIXUP_INT(page_copy_noexec)) +# define PAGE_READONLY_NOEXEC __pgprot(BTFIXUP_INT(page_readonly_noexec)) +#else +# define PAGE_SHARED_NOEXEC PAGE_SHARED +# define PAGE_COPY_NOEXEC PAGE_COPY +# define PAGE_READONLY_NOEXEC PAGE_READONLY +#endif + extern unsigned long page_kernel; #ifdef MODULE diff -urNp linux-2.6.32.48/arch/sparc/include/asm/pgtsrmmu.h linux-2.6.32.48-openvz/arch/sparc/include/asm/pgtsrmmu.h --- linux-2.6.32.48/arch/sparc/include/asm/pgtsrmmu.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/include/asm/pgtsrmmu.h 2011-11-17 20:08:53.000000000 -0500 @@ -115,6 +115,13 @@ SRMMU_EXEC | SRMMU_REF) #define SRMMU_PAGE_RDONLY __pgprot(SRMMU_VALID | SRMMU_CACHE | \ SRMMU_EXEC | SRMMU_REF) + +#ifdef CONFIG_PAX_PAGEEXEC +#define SRMMU_PAGE_SHARED_NOEXEC __pgprot(SRMMU_VALID | SRMMU_CACHE | SRMMU_WRITE | SRMMU_REF) +#define SRMMU_PAGE_COPY_NOEXEC __pgprot(SRMMU_VALID | SRMMU_CACHE | SRMMU_REF) +#define SRMMU_PAGE_RDONLY_NOEXEC __pgprot(SRMMU_VALID | SRMMU_CACHE | SRMMU_REF) +#endif + #define SRMMU_PAGE_KERNEL __pgprot(SRMMU_VALID | SRMMU_CACHE | SRMMU_PRIV | \ SRMMU_DIRTY | SRMMU_REF) diff -urNp linux-2.6.32.48/arch/sparc/include/asm/spinlock_64.h linux-2.6.32.48-openvz/arch/sparc/include/asm/spinlock_64.h --- linux-2.6.32.48/arch/sparc/include/asm/spinlock_64.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/include/asm/spinlock_64.h 2011-11-17 20:09:04.000000000 -0500 @@ -92,14 +92,19 @@ static inline void __raw_spin_lock_flags /* Multi-reader locks, these are much saner than the 32-bit Sparc ones... */ -static void inline arch_read_lock(raw_rwlock_t *lock) +static inline void arch_read_lock(raw_rwlock_t *lock) { unsigned long tmp1, tmp2; __asm__ __volatile__ ( "1: ldsw [%2], %0\n" " brlz,pn %0, 2f\n" -"4: add %0, 1, %1\n" +"4: addcc %0, 1, %1\n" + +#ifdef CONFIG_PAX_REFCOUNT +" tvs %%icc, 6\n" +#endif + " cas [%2], %0, %1\n" " cmp %0, %1\n" " bne,pn %%icc, 1b\n" @@ -112,10 +117,10 @@ static void inline arch_read_lock(raw_rw " .previous" : "=&r" (tmp1), "=&r" (tmp2) : "r" (lock) - : "memory"); + : "memory", "cc"); } -static int inline arch_read_trylock(raw_rwlock_t *lock) +static inline int arch_read_trylock(raw_rwlock_t *lock) { int tmp1, tmp2; @@ -123,7 +128,12 @@ static int inline arch_read_trylock(raw_ "1: ldsw [%2], %0\n" " brlz,a,pn %0, 2f\n" " mov 0, %0\n" -" add %0, 1, %1\n" +" addcc %0, 1, %1\n" + +#ifdef CONFIG_PAX_REFCOUNT +" tvs %%icc, 6\n" +#endif + " cas [%2], %0, %1\n" " cmp %0, %1\n" " bne,pn %%icc, 1b\n" @@ -136,13 +146,18 @@ static int inline arch_read_trylock(raw_ return tmp1; } -static void inline arch_read_unlock(raw_rwlock_t *lock) +static inline void arch_read_unlock(raw_rwlock_t *lock) { unsigned long tmp1, tmp2; __asm__ __volatile__( "1: lduw [%2], %0\n" -" sub %0, 1, %1\n" +" subcc %0, 1, %1\n" + +#ifdef CONFIG_PAX_REFCOUNT +" tvs %%icc, 6\n" +#endif + " cas [%2], %0, %1\n" " cmp %0, %1\n" " bne,pn %%xcc, 1b\n" @@ -152,7 +167,7 @@ static void inline arch_read_unlock(raw_ : "memory"); } -static void inline arch_write_lock(raw_rwlock_t *lock) +static inline void arch_write_lock(raw_rwlock_t *lock) { unsigned long mask, tmp1, tmp2; @@ -177,7 +192,7 @@ static void inline arch_write_lock(raw_r : "memory"); } -static void inline arch_write_unlock(raw_rwlock_t *lock) +static inline void arch_write_unlock(raw_rwlock_t *lock) { __asm__ __volatile__( " stw %%g0, [%0]" @@ -186,7 +201,7 @@ static void inline arch_write_unlock(raw : "memory"); } -static int inline arch_write_trylock(raw_rwlock_t *lock) +static inline int arch_write_trylock(raw_rwlock_t *lock) { unsigned long mask, tmp1, tmp2, result; diff -urNp linux-2.6.32.48/arch/sparc/include/asm/thread_info_32.h linux-2.6.32.48-openvz/arch/sparc/include/asm/thread_info_32.h --- linux-2.6.32.48/arch/sparc/include/asm/thread_info_32.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/include/asm/thread_info_32.h 2011-11-17 20:09:04.000000000 -0500 @@ -50,6 +50,8 @@ struct thread_info { unsigned long w_saved; struct restart_block restart_block; + + unsigned long lowest_stack; }; /* diff -urNp linux-2.6.32.48/arch/sparc/include/asm/thread_info_64.h linux-2.6.32.48-openvz/arch/sparc/include/asm/thread_info_64.h --- linux-2.6.32.48/arch/sparc/include/asm/thread_info_64.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/include/asm/thread_info_64.h 2011-11-17 20:09:04.000000000 -0500 @@ -68,6 +68,8 @@ struct thread_info { struct pt_regs *kern_una_regs; unsigned int kern_una_insn; + unsigned long lowest_stack; + unsigned long fpregs[0] __attribute__ ((aligned(64))); }; diff -urNp linux-2.6.32.48/arch/sparc/include/asm/uaccess_32.h linux-2.6.32.48-openvz/arch/sparc/include/asm/uaccess_32.h --- linux-2.6.32.48/arch/sparc/include/asm/uaccess_32.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/include/asm/uaccess_32.h 2011-11-17 20:08:53.000000000 -0500 @@ -249,27 +249,46 @@ extern unsigned long __copy_user(void __ static inline unsigned long copy_to_user(void __user *to, const void *from, unsigned long n) { - if (n && __access_ok((unsigned long) to, n)) + if ((long)n < 0) + return n; + + if (n && __access_ok((unsigned long) to, n)) { + if (!__builtin_constant_p(n)) + check_object_size(from, n, true); return __copy_user(to, (__force void __user *) from, n); - else + } else return n; } static inline unsigned long __copy_to_user(void __user *to, const void *from, unsigned long n) { + if ((long)n < 0) + return n; + + if (!__builtin_constant_p(n)) + check_object_size(from, n, true); + return __copy_user(to, (__force void __user *) from, n); } static inline unsigned long copy_from_user(void *to, const void __user *from, unsigned long n) { - if (n && __access_ok((unsigned long) from, n)) + if ((long)n < 0) + return n; + + if (n && __access_ok((unsigned long) from, n)) { + if (!__builtin_constant_p(n)) + check_object_size(to, n, false); return __copy_user((__force void __user *) to, from, n); - else + } else return n; } static inline unsigned long __copy_from_user(void *to, const void __user *from, unsigned long n) { + if ((long)n < 0) + return n; + return __copy_user((__force void __user *) to, from, n); } diff -urNp linux-2.6.32.48/arch/sparc/include/asm/uaccess_64.h linux-2.6.32.48-openvz/arch/sparc/include/asm/uaccess_64.h --- linux-2.6.32.48/arch/sparc/include/asm/uaccess_64.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/include/asm/uaccess_64.h 2011-11-17 20:08:53.000000000 -0500 @@ -9,6 +9,7 @@ #include #include #include +#include #include #include #include @@ -212,8 +213,15 @@ extern unsigned long copy_from_user_fixu static inline unsigned long __must_check copy_from_user(void *to, const void __user *from, unsigned long size) { - unsigned long ret = ___copy_from_user(to, from, size); + unsigned long ret; + if ((long)size < 0 || size > INT_MAX) + return size; + + if (!__builtin_constant_p(size)) + check_object_size(to, size, false); + + ret = ___copy_from_user(to, from, size); if (unlikely(ret)) ret = copy_from_user_fixup(to, from, size); return ret; @@ -228,8 +236,15 @@ extern unsigned long copy_to_user_fixup( static inline unsigned long __must_check copy_to_user(void __user *to, const void *from, unsigned long size) { - unsigned long ret = ___copy_to_user(to, from, size); + unsigned long ret; + + if ((long)size < 0 || size > INT_MAX) + return size; + + if (!__builtin_constant_p(size)) + check_object_size(from, size, true); + ret = ___copy_to_user(to, from, size); if (unlikely(ret)) ret = copy_to_user_fixup(to, from, size); return ret; diff -urNp linux-2.6.32.48/arch/sparc/include/asm/uaccess.h linux-2.6.32.48-openvz/arch/sparc/include/asm/uaccess.h --- linux-2.6.32.48/arch/sparc/include/asm/uaccess.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/include/asm/uaccess.h 2011-11-17 20:08:53.000000000 -0500 @@ -1,5 +1,13 @@ #ifndef ___ASM_SPARC_UACCESS_H #define ___ASM_SPARC_UACCESS_H + +#ifdef __KERNEL__ +#ifndef __ASSEMBLY__ +#include +extern void check_object_size(const void *ptr, unsigned long n, bool to); +#endif +#endif + #if defined(__sparc__) && defined(__arch64__) #include #else diff -urNp linux-2.6.32.48/arch/sparc/kernel/iommu.c linux-2.6.32.48-openvz/arch/sparc/kernel/iommu.c --- linux-2.6.32.48/arch/sparc/kernel/iommu.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/kernel/iommu.c 2011-11-17 20:08:53.000000000 -0500 @@ -826,7 +826,7 @@ static void dma_4u_sync_sg_for_cpu(struc spin_unlock_irqrestore(&iommu->lock, flags); } -static struct dma_map_ops sun4u_dma_ops = { +static const struct dma_map_ops sun4u_dma_ops = { .alloc_coherent = dma_4u_alloc_coherent, .free_coherent = dma_4u_free_coherent, .map_page = dma_4u_map_page, @@ -837,7 +837,7 @@ static struct dma_map_ops sun4u_dma_ops .sync_sg_for_cpu = dma_4u_sync_sg_for_cpu, }; -struct dma_map_ops *dma_ops = &sun4u_dma_ops; +const struct dma_map_ops *dma_ops = &sun4u_dma_ops; EXPORT_SYMBOL(dma_ops); extern int pci64_dma_supported(struct pci_dev *pdev, u64 device_mask); diff -urNp linux-2.6.32.48/arch/sparc/kernel/ioport.c linux-2.6.32.48-openvz/arch/sparc/kernel/ioport.c --- linux-2.6.32.48/arch/sparc/kernel/ioport.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/kernel/ioport.c 2011-11-17 20:08:53.000000000 -0500 @@ -392,7 +392,7 @@ static void sbus_sync_sg_for_device(stru BUG(); } -struct dma_map_ops sbus_dma_ops = { +const struct dma_map_ops sbus_dma_ops = { .alloc_coherent = sbus_alloc_coherent, .free_coherent = sbus_free_coherent, .map_page = sbus_map_page, @@ -403,7 +403,7 @@ struct dma_map_ops sbus_dma_ops = { .sync_sg_for_device = sbus_sync_sg_for_device, }; -struct dma_map_ops *dma_ops = &sbus_dma_ops; +const struct dma_map_ops *dma_ops = &sbus_dma_ops; EXPORT_SYMBOL(dma_ops); static int __init sparc_register_ioport(void) @@ -640,7 +640,7 @@ static void pci32_sync_sg_for_device(str } } -struct dma_map_ops pci32_dma_ops = { +const struct dma_map_ops pci32_dma_ops = { .alloc_coherent = pci32_alloc_coherent, .free_coherent = pci32_free_coherent, .map_page = pci32_map_page, diff -urNp linux-2.6.32.48/arch/sparc/kernel/kgdb_32.c linux-2.6.32.48-openvz/arch/sparc/kernel/kgdb_32.c --- linux-2.6.32.48/arch/sparc/kernel/kgdb_32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/kernel/kgdb_32.c 2011-11-17 20:08:53.000000000 -0500 @@ -158,7 +158,7 @@ void kgdb_arch_exit(void) { } -struct kgdb_arch arch_kgdb_ops = { +const struct kgdb_arch arch_kgdb_ops = { /* Breakpoint instruction: ta 0x7d */ .gdb_bpt_instr = { 0x91, 0xd0, 0x20, 0x7d }, }; diff -urNp linux-2.6.32.48/arch/sparc/kernel/kgdb_64.c linux-2.6.32.48-openvz/arch/sparc/kernel/kgdb_64.c --- linux-2.6.32.48/arch/sparc/kernel/kgdb_64.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/kernel/kgdb_64.c 2011-11-17 20:08:53.000000000 -0500 @@ -180,7 +180,7 @@ void kgdb_arch_exit(void) { } -struct kgdb_arch arch_kgdb_ops = { +const struct kgdb_arch arch_kgdb_ops = { /* Breakpoint instruction: ta 0x72 */ .gdb_bpt_instr = { 0x91, 0xd0, 0x20, 0x72 }, }; diff -urNp linux-2.6.32.48/arch/sparc/kernel/Makefile linux-2.6.32.48-openvz/arch/sparc/kernel/Makefile --- linux-2.6.32.48/arch/sparc/kernel/Makefile 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/kernel/Makefile 2011-11-17 20:09:04.000000000 -0500 @@ -3,7 +3,7 @@ # asflags-y := -ansi -ccflags-y := -Werror +#ccflags-y := -Werror extra-y := head_$(BITS).o extra-y += init_task.o diff -urNp linux-2.6.32.48/arch/sparc/kernel/pci_sun4v.c linux-2.6.32.48-openvz/arch/sparc/kernel/pci_sun4v.c --- linux-2.6.32.48/arch/sparc/kernel/pci_sun4v.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/kernel/pci_sun4v.c 2011-11-17 20:08:53.000000000 -0500 @@ -525,7 +525,7 @@ static void dma_4v_unmap_sg(struct devic spin_unlock_irqrestore(&iommu->lock, flags); } -static struct dma_map_ops sun4v_dma_ops = { +static const struct dma_map_ops sun4v_dma_ops = { .alloc_coherent = dma_4v_alloc_coherent, .free_coherent = dma_4v_free_coherent, .map_page = dma_4v_map_page, diff -urNp linux-2.6.32.48/arch/sparc/kernel/process_32.c linux-2.6.32.48-openvz/arch/sparc/kernel/process_32.c --- linux-2.6.32.48/arch/sparc/kernel/process_32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/kernel/process_32.c 2011-11-17 20:08:53.000000000 -0500 @@ -196,7 +196,7 @@ void __show_backtrace(unsigned long fp) rw->ins[4], rw->ins[5], rw->ins[6], rw->ins[7]); - printk("%pS\n", (void *) rw->ins[7]); + printk("%pA\n", (void *) rw->ins[7]); rw = (struct reg_window32 *) rw->ins[6]; } spin_unlock_irqrestore(&sparc_backtrace_lock, flags); @@ -263,14 +263,14 @@ void show_regs(struct pt_regs *r) printk("PSR: %08lx PC: %08lx NPC: %08lx Y: %08lx %s\n", r->psr, r->pc, r->npc, r->y, print_tainted()); - printk("PC: <%pS>\n", (void *) r->pc); + printk("PC: <%pA>\n", (void *) r->pc); printk("%%G: %08lx %08lx %08lx %08lx %08lx %08lx %08lx %08lx\n", r->u_regs[0], r->u_regs[1], r->u_regs[2], r->u_regs[3], r->u_regs[4], r->u_regs[5], r->u_regs[6], r->u_regs[7]); printk("%%O: %08lx %08lx %08lx %08lx %08lx %08lx %08lx %08lx\n", r->u_regs[8], r->u_regs[9], r->u_regs[10], r->u_regs[11], r->u_regs[12], r->u_regs[13], r->u_regs[14], r->u_regs[15]); - printk("RPC: <%pS>\n", (void *) r->u_regs[15]); + printk("RPC: <%pA>\n", (void *) r->u_regs[15]); printk("%%L: %08lx %08lx %08lx %08lx %08lx %08lx %08lx %08lx\n", rw->locals[0], rw->locals[1], rw->locals[2], rw->locals[3], @@ -305,7 +305,7 @@ void show_stack(struct task_struct *tsk, rw = (struct reg_window32 *) fp; pc = rw->ins[7]; printk("[%08lx : ", pc); - printk("%pS ] ", (void *) pc); + printk("%pA ] ", (void *) pc); fp = rw->ins[6]; } while (++count < 16); printk("\n"); diff -urNp linux-2.6.32.48/arch/sparc/kernel/process_64.c linux-2.6.32.48-openvz/arch/sparc/kernel/process_64.c --- linux-2.6.32.48/arch/sparc/kernel/process_64.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/kernel/process_64.c 2011-11-17 20:08:53.000000000 -0500 @@ -180,14 +180,14 @@ static void show_regwindow(struct pt_reg printk("i4: %016lx i5: %016lx i6: %016lx i7: %016lx\n", rwk->ins[4], rwk->ins[5], rwk->ins[6], rwk->ins[7]); if (regs->tstate & TSTATE_PRIV) - printk("I7: <%pS>\n", (void *) rwk->ins[7]); + printk("I7: <%pA>\n", (void *) rwk->ins[7]); } void show_regs(struct pt_regs *regs) { printk("TSTATE: %016lx TPC: %016lx TNPC: %016lx Y: %08x %s\n", regs->tstate, regs->tpc, regs->tnpc, regs->y, print_tainted()); - printk("TPC: <%pS>\n", (void *) regs->tpc); + printk("TPC: <%pA>\n", (void *) regs->tpc); printk("g0: %016lx g1: %016lx g2: %016lx g3: %016lx\n", regs->u_regs[0], regs->u_regs[1], regs->u_regs[2], regs->u_regs[3]); @@ -200,7 +200,7 @@ void show_regs(struct pt_regs *regs) printk("o4: %016lx o5: %016lx sp: %016lx ret_pc: %016lx\n", regs->u_regs[12], regs->u_regs[13], regs->u_regs[14], regs->u_regs[15]); - printk("RPC: <%pS>\n", (void *) regs->u_regs[15]); + printk("RPC: <%pA>\n", (void *) regs->u_regs[15]); show_regwindow(regs); } @@ -284,7 +284,7 @@ void arch_trigger_all_cpu_backtrace(void ((tp && tp->task) ? tp->task->pid : -1)); if (gp->tstate & TSTATE_PRIV) { - printk(" TPC[%pS] O7[%pS] I7[%pS] RPC[%pS]\n", + printk(" TPC[%pA] O7[%pA] I7[%pA] RPC[%pA]\n", (void *) gp->tpc, (void *) gp->o7, (void *) gp->i7, diff -urNp linux-2.6.32.48/arch/sparc/kernel/sys_sparc_32.c linux-2.6.32.48-openvz/arch/sparc/kernel/sys_sparc_32.c --- linux-2.6.32.48/arch/sparc/kernel/sys_sparc_32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/kernel/sys_sparc_32.c 2011-11-17 20:08:53.000000000 -0500 @@ -57,7 +57,7 @@ unsigned long arch_get_unmapped_area(str if (ARCH_SUN4C && len > 0x20000000) return -ENOMEM; if (!addr) - addr = TASK_UNMAPPED_BASE; + addr = current->mm->mmap_base; if (flags & MAP_SHARED) addr = COLOUR_ALIGN(addr); @@ -72,7 +72,7 @@ unsigned long arch_get_unmapped_area(str } if (TASK_SIZE - PAGE_SIZE - len < addr) return -ENOMEM; - if (!vmm || addr + len <= vmm->vm_start) + if (check_heap_stack_gap(vmm, addr, len)) return addr; addr = vmm->vm_end; if (flags & MAP_SHARED) diff -urNp linux-2.6.32.48/arch/sparc/kernel/sys_sparc_64.c linux-2.6.32.48-openvz/arch/sparc/kernel/sys_sparc_64.c --- linux-2.6.32.48/arch/sparc/kernel/sys_sparc_64.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/kernel/sys_sparc_64.c 2011-11-17 20:08:53.000000000 -0500 @@ -125,7 +125,7 @@ unsigned long arch_get_unmapped_area(str /* We do not accept a shared mapping if it would violate * cache aliasing constraints. */ - if ((flags & MAP_SHARED) && + if ((filp || (flags & MAP_SHARED)) && ((addr - (pgoff << PAGE_SHIFT)) & (SHMLBA - 1))) return -EINVAL; return addr; @@ -140,6 +140,10 @@ unsigned long arch_get_unmapped_area(str if (filp || (flags & MAP_SHARED)) do_color_align = 1; +#ifdef CONFIG_PAX_RANDMMAP + if (!(mm->pax_flags & MF_PAX_RANDMMAP)) +#endif + if (addr) { if (do_color_align) addr = COLOUR_ALIGN(addr, pgoff); @@ -147,15 +151,14 @@ unsigned long arch_get_unmapped_area(str addr = PAGE_ALIGN(addr); vma = find_vma(mm, addr); - if (task_size - len >= addr && - (!vma || addr + len <= vma->vm_start)) + if (task_size - len >= addr && check_heap_stack_gap(vma, addr, len)) return addr; } if (len > mm->cached_hole_size) { - start_addr = addr = mm->free_area_cache; + start_addr = addr = mm->free_area_cache; } else { - start_addr = addr = TASK_UNMAPPED_BASE; + start_addr = addr = mm->mmap_base; mm->cached_hole_size = 0; } @@ -175,14 +178,14 @@ full_search: vma = find_vma(mm, VA_EXCLUDE_END); } if (unlikely(task_size < addr)) { - if (start_addr != TASK_UNMAPPED_BASE) { - start_addr = addr = TASK_UNMAPPED_BASE; + if (start_addr != mm->mmap_base) { + start_addr = addr = mm->mmap_base; mm->cached_hole_size = 0; goto full_search; } return -ENOMEM; } - if (likely(!vma || addr + len <= vma->vm_start)) { + if (likely(check_heap_stack_gap(vma, addr, len))) { /* * Remember the place where we stopped the search: */ @@ -216,7 +219,7 @@ arch_get_unmapped_area_topdown(struct fi /* We do not accept a shared mapping if it would violate * cache aliasing constraints. */ - if ((flags & MAP_SHARED) && + if ((filp || (flags & MAP_SHARED)) && ((addr - (pgoff << PAGE_SHIFT)) & (SHMLBA - 1))) return -EINVAL; return addr; @@ -237,8 +240,7 @@ arch_get_unmapped_area_topdown(struct fi addr = PAGE_ALIGN(addr); vma = find_vma(mm, addr); - if (task_size - len >= addr && - (!vma || addr + len <= vma->vm_start)) + if (task_size - len >= addr && check_heap_stack_gap(vma, addr, len)) return addr; } @@ -259,7 +261,7 @@ arch_get_unmapped_area_topdown(struct fi /* make sure it can fit in the remaining address space */ if (likely(addr > len)) { vma = find_vma(mm, addr-len); - if (!vma || addr <= vma->vm_start) { + if (check_heap_stack_gap(vma, addr - len, len)) { /* remember the address as a hint for next time */ return (mm->free_area_cache = addr-len); } @@ -268,18 +270,18 @@ arch_get_unmapped_area_topdown(struct fi if (unlikely(mm->mmap_base < len)) goto bottomup; - addr = mm->mmap_base-len; - if (do_color_align) - addr = COLOUR_ALIGN_DOWN(addr, pgoff); + addr = mm->mmap_base - len; do { + if (do_color_align) + addr = COLOUR_ALIGN_DOWN(addr, pgoff); /* * Lookup failure means no vma is above this address, * else if new region fits below vma->vm_start, * return with success: */ vma = find_vma(mm, addr); - if (likely(!vma || addr+len <= vma->vm_start)) { + if (likely(check_heap_stack_gap(vma, addr, len))) { /* remember the address as a hint for next time */ return (mm->free_area_cache = addr); } @@ -289,10 +291,8 @@ arch_get_unmapped_area_topdown(struct fi mm->cached_hole_size = vma->vm_start - addr; /* try just below the current vma->vm_start */ - addr = vma->vm_start-len; - if (do_color_align) - addr = COLOUR_ALIGN_DOWN(addr, pgoff); - } while (likely(len < vma->vm_start)); + addr = skip_heap_stack_gap(vma, len); + } while (!IS_ERR_VALUE(addr)); bottomup: /* @@ -384,6 +384,12 @@ void arch_pick_mmap_layout(struct mm_str current->signal->rlim[RLIMIT_STACK].rlim_cur == RLIM_INFINITY || sysctl_legacy_va_layout) { mm->mmap_base = TASK_UNMAPPED_BASE + random_factor; + +#ifdef CONFIG_PAX_RANDMMAP + if (mm->pax_flags & MF_PAX_RANDMMAP) + mm->mmap_base += mm->delta_mmap; +#endif + mm->get_unmapped_area = arch_get_unmapped_area; mm->unmap_area = arch_unmap_area; } else { @@ -398,6 +404,12 @@ void arch_pick_mmap_layout(struct mm_str gap = (task_size / 6 * 5); mm->mmap_base = PAGE_ALIGN(task_size - gap - random_factor); + +#ifdef CONFIG_PAX_RANDMMAP + if (mm->pax_flags & MF_PAX_RANDMMAP) + mm->mmap_base -= mm->delta_mmap + mm->delta_stack; +#endif + mm->get_unmapped_area = arch_get_unmapped_area_topdown; mm->unmap_area = arch_unmap_area_topdown; } diff -urNp linux-2.6.32.48/arch/sparc/kernel/traps_32.c linux-2.6.32.48-openvz/arch/sparc/kernel/traps_32.c --- linux-2.6.32.48/arch/sparc/kernel/traps_32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/kernel/traps_32.c 2011-11-17 20:09:04.000000000 -0500 @@ -44,6 +44,8 @@ static void instruction_dump(unsigned lo #define __SAVE __asm__ __volatile__("save %sp, -0x40, %sp\n\t") #define __RESTORE __asm__ __volatile__("restore %g0, %g0, %g0\n\t") +extern void gr_handle_kernel_exploit(void); + void die_if_kernel(char *str, struct pt_regs *regs) { static int die_counter; @@ -76,15 +78,17 @@ void die_if_kernel(char *str, struct pt_ count++ < 30 && (((unsigned long) rw) >= PAGE_OFFSET) && !(((unsigned long) rw) & 0x7)) { - printk("Caller[%08lx]: %pS\n", rw->ins[7], + printk("Caller[%08lx]: %pA\n", rw->ins[7], (void *) rw->ins[7]); rw = (struct reg_window32 *)rw->ins[6]; } } printk("Instruction DUMP:"); instruction_dump ((unsigned long *) regs->pc); - if(regs->psr & PSR_PS) + if(regs->psr & PSR_PS) { + gr_handle_kernel_exploit(); do_exit(SIGKILL); + } do_exit(SIGSEGV); } diff -urNp linux-2.6.32.48/arch/sparc/kernel/traps_64.c linux-2.6.32.48-openvz/arch/sparc/kernel/traps_64.c --- linux-2.6.32.48/arch/sparc/kernel/traps_64.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/kernel/traps_64.c 2011-11-17 20:09:04.000000000 -0500 @@ -73,7 +73,7 @@ static void dump_tl1_traplog(struct tl1_ i + 1, p->trapstack[i].tstate, p->trapstack[i].tpc, p->trapstack[i].tnpc, p->trapstack[i].tt); - printk("TRAPLOG: TPC<%pS>\n", (void *) p->trapstack[i].tpc); + printk("TRAPLOG: TPC<%pA>\n", (void *) p->trapstack[i].tpc); } } @@ -93,6 +93,12 @@ void bad_trap(struct pt_regs *regs, long lvl -= 0x100; if (regs->tstate & TSTATE_PRIV) { + +#ifdef CONFIG_PAX_REFCOUNT + if (lvl == 6) + pax_report_refcount_overflow(regs); +#endif + sprintf(buffer, "Kernel bad sw trap %lx", lvl); die_if_kernel(buffer, regs); } @@ -111,11 +117,16 @@ void bad_trap(struct pt_regs *regs, long void bad_trap_tl1(struct pt_regs *regs, long lvl) { char buffer[32]; - + if (notify_die(DIE_TRAP_TL1, "bad trap tl1", regs, 0, lvl, SIGTRAP) == NOTIFY_STOP) return; +#ifdef CONFIG_PAX_REFCOUNT + if (lvl == 6) + pax_report_refcount_overflow(regs); +#endif + dump_tl1_traplog((struct tl1_traplog *)(regs + 1)); sprintf (buffer, "Bad trap %lx at tl>0", lvl); @@ -1139,7 +1150,7 @@ static void cheetah_log_errors(struct pt regs->tpc, regs->tnpc, regs->u_regs[UREG_I7], regs->tstate); printk("%s" "ERROR(%d): ", (recoverable ? KERN_WARNING : KERN_CRIT), smp_processor_id()); - printk("TPC<%pS>\n", (void *) regs->tpc); + printk("TPC<%pA>\n", (void *) regs->tpc); printk("%s" "ERROR(%d): M_SYND(%lx), E_SYND(%lx)%s%s\n", (recoverable ? KERN_WARNING : KERN_CRIT), smp_processor_id(), (afsr & CHAFSR_M_SYNDROME) >> CHAFSR_M_SYNDROME_SHIFT, @@ -1746,7 +1757,7 @@ void cheetah_plus_parity_error(int type, smp_processor_id(), (type & 0x1) ? 'I' : 'D', regs->tpc); - printk(KERN_EMERG "TPC<%pS>\n", (void *) regs->tpc); + printk(KERN_EMERG "TPC<%pA>\n", (void *) regs->tpc); panic("Irrecoverable Cheetah+ parity error."); } @@ -1754,7 +1765,7 @@ void cheetah_plus_parity_error(int type, smp_processor_id(), (type & 0x1) ? 'I' : 'D', regs->tpc); - printk(KERN_WARNING "TPC<%pS>\n", (void *) regs->tpc); + printk(KERN_WARNING "TPC<%pA>\n", (void *) regs->tpc); } struct sun4v_error_entry { @@ -1961,9 +1972,9 @@ void sun4v_itlb_error_report(struct pt_r printk(KERN_EMERG "SUN4V-ITLB: Error at TPC[%lx], tl %d\n", regs->tpc, tl); - printk(KERN_EMERG "SUN4V-ITLB: TPC<%pS>\n", (void *) regs->tpc); + printk(KERN_EMERG "SUN4V-ITLB: TPC<%pA>\n", (void *) regs->tpc); printk(KERN_EMERG "SUN4V-ITLB: O7[%lx]\n", regs->u_regs[UREG_I7]); - printk(KERN_EMERG "SUN4V-ITLB: O7<%pS>\n", + printk(KERN_EMERG "SUN4V-ITLB: O7<%pA>\n", (void *) regs->u_regs[UREG_I7]); printk(KERN_EMERG "SUN4V-ITLB: vaddr[%lx] ctx[%lx] " "pte[%lx] error[%lx]\n", @@ -1985,9 +1996,9 @@ void sun4v_dtlb_error_report(struct pt_r printk(KERN_EMERG "SUN4V-DTLB: Error at TPC[%lx], tl %d\n", regs->tpc, tl); - printk(KERN_EMERG "SUN4V-DTLB: TPC<%pS>\n", (void *) regs->tpc); + printk(KERN_EMERG "SUN4V-DTLB: TPC<%pA>\n", (void *) regs->tpc); printk(KERN_EMERG "SUN4V-DTLB: O7[%lx]\n", regs->u_regs[UREG_I7]); - printk(KERN_EMERG "SUN4V-DTLB: O7<%pS>\n", + printk(KERN_EMERG "SUN4V-DTLB: O7<%pA>\n", (void *) regs->u_regs[UREG_I7]); printk(KERN_EMERG "SUN4V-DTLB: vaddr[%lx] ctx[%lx] " "pte[%lx] error[%lx]\n", @@ -2191,7 +2202,7 @@ void show_stack(struct task_struct *tsk, fp = (unsigned long)sf->fp + STACK_BIAS; } - printk(" [%016lx] %pS\n", pc, (void *) pc); + printk(" [%016lx] %pA\n", pc, (void *) pc); } while (++count < 16); } @@ -2233,6 +2244,8 @@ static inline struct reg_window *kernel_ return (struct reg_window *) (fp + STACK_BIAS); } +extern void gr_handle_kernel_exploit(void); + void die_if_kernel(char *str, struct pt_regs *regs) { static int die_counter; @@ -2260,7 +2273,7 @@ void die_if_kernel(char *str, struct pt_ while (rw && count++ < 30&& is_kernel_stack(current, rw)) { - printk("Caller[%016lx]: %pS\n", rw->ins[7], + printk("Caller[%016lx]: %pA\n", rw->ins[7], (void *) rw->ins[7]); rw = kernel_stack_up(rw); @@ -2273,8 +2286,11 @@ void die_if_kernel(char *str, struct pt_ } user_instruction_dump ((unsigned int __user *) regs->tpc); } - if (regs->tstate & TSTATE_PRIV) + if (regs->tstate & TSTATE_PRIV) { + gr_handle_kernel_exploit(); do_exit(SIGKILL); + } + do_exit(SIGSEGV); } EXPORT_SYMBOL(die_if_kernel); diff -urNp linux-2.6.32.48/arch/sparc/kernel/una_asm_64.S linux-2.6.32.48-openvz/arch/sparc/kernel/una_asm_64.S --- linux-2.6.32.48/arch/sparc/kernel/una_asm_64.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/kernel/una_asm_64.S 2011-11-17 20:09:04.000000000 -0500 @@ -127,7 +127,7 @@ do_int_load: wr %o5, 0x0, %asi retl mov 0, %o0 - .size __do_int_load, .-__do_int_load + .size do_int_load, .-do_int_load .section __ex_table,"a" .word 4b, __retl_efault diff -urNp linux-2.6.32.48/arch/sparc/kernel/unaligned_64.c linux-2.6.32.48-openvz/arch/sparc/kernel/unaligned_64.c --- linux-2.6.32.48/arch/sparc/kernel/unaligned_64.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/kernel/unaligned_64.c 2011-11-17 20:08:53.000000000 -0500 @@ -288,7 +288,7 @@ static void log_unaligned(struct pt_regs if (count < 5) { last_time = jiffies; count++; - printk("Kernel unaligned access at TPC[%lx] %pS\n", + printk("Kernel unaligned access at TPC[%lx] %pA\n", regs->tpc, (void *) regs->tpc); } } diff -urNp linux-2.6.32.48/arch/sparc/lib/atomic_64.S linux-2.6.32.48-openvz/arch/sparc/lib/atomic_64.S --- linux-2.6.32.48/arch/sparc/lib/atomic_64.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/lib/atomic_64.S 2011-11-17 20:08:53.000000000 -0500 @@ -18,7 +18,12 @@ atomic_add: /* %o0 = increment, %o1 = atomic_ptr */ BACKOFF_SETUP(%o2) 1: lduw [%o1], %g1 - add %g1, %o0, %g7 + addcc %g1, %o0, %g7 + +#ifdef CONFIG_PAX_REFCOUNT + tvs %icc, 6 +#endif + cas [%o1], %g1, %g7 cmp %g1, %g7 bne,pn %icc, 2f @@ -28,12 +33,32 @@ atomic_add: /* %o0 = increment, %o1 = at 2: BACKOFF_SPIN(%o2, %o3, 1b) .size atomic_add, .-atomic_add + .globl atomic_add_unchecked + .type atomic_add_unchecked,#function +atomic_add_unchecked: /* %o0 = increment, %o1 = atomic_ptr */ + BACKOFF_SETUP(%o2) +1: lduw [%o1], %g1 + add %g1, %o0, %g7 + cas [%o1], %g1, %g7 + cmp %g1, %g7 + bne,pn %icc, 2f + nop + retl + nop +2: BACKOFF_SPIN(%o2, %o3, 1b) + .size atomic_add_unchecked, .-atomic_add_unchecked + .globl atomic_sub .type atomic_sub,#function atomic_sub: /* %o0 = decrement, %o1 = atomic_ptr */ BACKOFF_SETUP(%o2) 1: lduw [%o1], %g1 - sub %g1, %o0, %g7 + subcc %g1, %o0, %g7 + +#ifdef CONFIG_PAX_REFCOUNT + tvs %icc, 6 +#endif + cas [%o1], %g1, %g7 cmp %g1, %g7 bne,pn %icc, 2f @@ -43,12 +68,32 @@ atomic_sub: /* %o0 = decrement, %o1 = at 2: BACKOFF_SPIN(%o2, %o3, 1b) .size atomic_sub, .-atomic_sub + .globl atomic_sub_unchecked + .type atomic_sub_unchecked,#function +atomic_sub_unchecked: /* %o0 = decrement, %o1 = atomic_ptr */ + BACKOFF_SETUP(%o2) +1: lduw [%o1], %g1 + sub %g1, %o0, %g7 + cas [%o1], %g1, %g7 + cmp %g1, %g7 + bne,pn %icc, 2f + nop + retl + nop +2: BACKOFF_SPIN(%o2, %o3, 1b) + .size atomic_sub_unchecked, .-atomic_sub_unchecked + .globl atomic_add_ret .type atomic_add_ret,#function atomic_add_ret: /* %o0 = increment, %o1 = atomic_ptr */ BACKOFF_SETUP(%o2) 1: lduw [%o1], %g1 - add %g1, %o0, %g7 + addcc %g1, %o0, %g7 + +#ifdef CONFIG_PAX_REFCOUNT + tvs %icc, 6 +#endif + cas [%o1], %g1, %g7 cmp %g1, %g7 bne,pn %icc, 2f @@ -59,12 +104,33 @@ atomic_add_ret: /* %o0 = increment, %o1 2: BACKOFF_SPIN(%o2, %o3, 1b) .size atomic_add_ret, .-atomic_add_ret + .globl atomic_add_ret_unchecked + .type atomic_add_ret_unchecked,#function +atomic_add_ret_unchecked: /* %o0 = increment, %o1 = atomic_ptr */ + BACKOFF_SETUP(%o2) +1: lduw [%o1], %g1 + addcc %g1, %o0, %g7 + cas [%o1], %g1, %g7 + cmp %g1, %g7 + bne,pn %icc, 2f + add %g7, %o0, %g7 + sra %g7, 0, %o0 + retl + nop +2: BACKOFF_SPIN(%o2, %o3, 1b) + .size atomic_add_ret_unchecked, .-atomic_add_ret_unchecked + .globl atomic_sub_ret .type atomic_sub_ret,#function atomic_sub_ret: /* %o0 = decrement, %o1 = atomic_ptr */ BACKOFF_SETUP(%o2) 1: lduw [%o1], %g1 - sub %g1, %o0, %g7 + subcc %g1, %o0, %g7 + +#ifdef CONFIG_PAX_REFCOUNT + tvs %icc, 6 +#endif + cas [%o1], %g1, %g7 cmp %g1, %g7 bne,pn %icc, 2f @@ -80,7 +146,12 @@ atomic_sub_ret: /* %o0 = decrement, %o1 atomic64_add: /* %o0 = increment, %o1 = atomic_ptr */ BACKOFF_SETUP(%o2) 1: ldx [%o1], %g1 - add %g1, %o0, %g7 + addcc %g1, %o0, %g7 + +#ifdef CONFIG_PAX_REFCOUNT + tvs %xcc, 6 +#endif + casx [%o1], %g1, %g7 cmp %g1, %g7 bne,pn %xcc, 2f @@ -90,12 +161,32 @@ atomic64_add: /* %o0 = increment, %o1 = 2: BACKOFF_SPIN(%o2, %o3, 1b) .size atomic64_add, .-atomic64_add + .globl atomic64_add_unchecked + .type atomic64_add_unchecked,#function +atomic64_add_unchecked: /* %o0 = increment, %o1 = atomic_ptr */ + BACKOFF_SETUP(%o2) +1: ldx [%o1], %g1 + addcc %g1, %o0, %g7 + casx [%o1], %g1, %g7 + cmp %g1, %g7 + bne,pn %xcc, 2f + nop + retl + nop +2: BACKOFF_SPIN(%o2, %o3, 1b) + .size atomic64_add_unchecked, .-atomic64_add_unchecked + .globl atomic64_sub .type atomic64_sub,#function atomic64_sub: /* %o0 = decrement, %o1 = atomic_ptr */ BACKOFF_SETUP(%o2) 1: ldx [%o1], %g1 - sub %g1, %o0, %g7 + subcc %g1, %o0, %g7 + +#ifdef CONFIG_PAX_REFCOUNT + tvs %xcc, 6 +#endif + casx [%o1], %g1, %g7 cmp %g1, %g7 bne,pn %xcc, 2f @@ -105,12 +196,32 @@ atomic64_sub: /* %o0 = decrement, %o1 = 2: BACKOFF_SPIN(%o2, %o3, 1b) .size atomic64_sub, .-atomic64_sub + .globl atomic64_sub_unchecked + .type atomic64_sub_unchecked,#function +atomic64_sub_unchecked: /* %o0 = decrement, %o1 = atomic_ptr */ + BACKOFF_SETUP(%o2) +1: ldx [%o1], %g1 + subcc %g1, %o0, %g7 + casx [%o1], %g1, %g7 + cmp %g1, %g7 + bne,pn %xcc, 2f + nop + retl + nop +2: BACKOFF_SPIN(%o2, %o3, 1b) + .size atomic64_sub_unchecked, .-atomic64_sub_unchecked + .globl atomic64_add_ret .type atomic64_add_ret,#function atomic64_add_ret: /* %o0 = increment, %o1 = atomic_ptr */ BACKOFF_SETUP(%o2) 1: ldx [%o1], %g1 - add %g1, %o0, %g7 + addcc %g1, %o0, %g7 + +#ifdef CONFIG_PAX_REFCOUNT + tvs %xcc, 6 +#endif + casx [%o1], %g1, %g7 cmp %g1, %g7 bne,pn %xcc, 2f @@ -121,12 +232,33 @@ atomic64_add_ret: /* %o0 = increment, %o 2: BACKOFF_SPIN(%o2, %o3, 1b) .size atomic64_add_ret, .-atomic64_add_ret + .globl atomic64_add_ret_unchecked + .type atomic64_add_ret_unchecked,#function +atomic64_add_ret_unchecked: /* %o0 = increment, %o1 = atomic_ptr */ + BACKOFF_SETUP(%o2) +1: ldx [%o1], %g1 + addcc %g1, %o0, %g7 + casx [%o1], %g1, %g7 + cmp %g1, %g7 + bne,pn %xcc, 2f + add %g7, %o0, %g7 + mov %g7, %o0 + retl + nop +2: BACKOFF_SPIN(%o2, %o3, 1b) + .size atomic64_add_ret_unchecked, .-atomic64_add_ret_unchecked + .globl atomic64_sub_ret .type atomic64_sub_ret,#function atomic64_sub_ret: /* %o0 = decrement, %o1 = atomic_ptr */ BACKOFF_SETUP(%o2) 1: ldx [%o1], %g1 - sub %g1, %o0, %g7 + subcc %g1, %o0, %g7 + +#ifdef CONFIG_PAX_REFCOUNT + tvs %xcc, 6 +#endif + casx [%o1], %g1, %g7 cmp %g1, %g7 bne,pn %xcc, 2f diff -urNp linux-2.6.32.48/arch/sparc/lib/ksyms.c linux-2.6.32.48-openvz/arch/sparc/lib/ksyms.c --- linux-2.6.32.48/arch/sparc/lib/ksyms.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/lib/ksyms.c 2011-11-17 20:09:04.000000000 -0500 @@ -144,12 +144,18 @@ EXPORT_SYMBOL(__downgrade_write); /* Atomic counter implementation. */ EXPORT_SYMBOL(atomic_add); +EXPORT_SYMBOL(atomic_add_unchecked); EXPORT_SYMBOL(atomic_add_ret); +EXPORT_SYMBOL(atomic_add_ret_unchecked); EXPORT_SYMBOL(atomic_sub); +EXPORT_SYMBOL(atomic_sub_unchecked); EXPORT_SYMBOL(atomic_sub_ret); EXPORT_SYMBOL(atomic64_add); +EXPORT_SYMBOL(atomic64_add_unchecked); EXPORT_SYMBOL(atomic64_add_ret); +EXPORT_SYMBOL(atomic64_add_ret_unchecked); EXPORT_SYMBOL(atomic64_sub); +EXPORT_SYMBOL(atomic64_sub_unchecked); EXPORT_SYMBOL(atomic64_sub_ret); /* Atomic bit operations. */ diff -urNp linux-2.6.32.48/arch/sparc/lib/Makefile linux-2.6.32.48-openvz/arch/sparc/lib/Makefile --- linux-2.6.32.48/arch/sparc/lib/Makefile 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/lib/Makefile 2011-11-17 20:08:53.000000000 -0500 @@ -2,7 +2,7 @@ # asflags-y := -ansi -DST_DIV0=0x02 -ccflags-y := -Werror +#ccflags-y := -Werror lib-$(CONFIG_SPARC32) += mul.o rem.o sdiv.o udiv.o umul.o urem.o ashrdi3.o lib-$(CONFIG_SPARC32) += memcpy.o memset.o diff -urNp linux-2.6.32.48/arch/sparc/lib/rwsem_64.S linux-2.6.32.48-openvz/arch/sparc/lib/rwsem_64.S --- linux-2.6.32.48/arch/sparc/lib/rwsem_64.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/lib/rwsem_64.S 2011-11-17 20:08:53.000000000 -0500 @@ -11,7 +11,12 @@ .globl __down_read __down_read: 1: lduw [%o0], %g1 - add %g1, 1, %g7 + addcc %g1, 1, %g7 + +#ifdef CONFIG_PAX_REFCOUNT + tvs %icc, 6 +#endif + cas [%o0], %g1, %g7 cmp %g1, %g7 bne,pn %icc, 1b @@ -33,7 +38,12 @@ __down_read: .globl __down_read_trylock __down_read_trylock: 1: lduw [%o0], %g1 - add %g1, 1, %g7 + addcc %g1, 1, %g7 + +#ifdef CONFIG_PAX_REFCOUNT + tvs %icc, 6 +#endif + cmp %g7, 0 bl,pn %icc, 2f mov 0, %o1 @@ -51,7 +61,12 @@ __down_write: or %g1, %lo(RWSEM_ACTIVE_WRITE_BIAS), %g1 1: lduw [%o0], %g3 - add %g3, %g1, %g7 + addcc %g3, %g1, %g7 + +#ifdef CONFIG_PAX_REFCOUNT + tvs %icc, 6 +#endif + cas [%o0], %g3, %g7 cmp %g3, %g7 bne,pn %icc, 1b @@ -77,7 +92,12 @@ __down_write_trylock: cmp %g3, 0 bne,pn %icc, 2f mov 0, %o1 - add %g3, %g1, %g7 + addcc %g3, %g1, %g7 + +#ifdef CONFIG_PAX_REFCOUNT + tvs %icc, 6 +#endif + cas [%o0], %g3, %g7 cmp %g3, %g7 bne,pn %icc, 1b @@ -90,7 +110,12 @@ __down_write_trylock: __up_read: 1: lduw [%o0], %g1 - sub %g1, 1, %g7 + subcc %g1, 1, %g7 + +#ifdef CONFIG_PAX_REFCOUNT + tvs %icc, 6 +#endif + cas [%o0], %g1, %g7 cmp %g1, %g7 bne,pn %icc, 1b @@ -118,7 +143,12 @@ __up_write: or %g1, %lo(RWSEM_ACTIVE_WRITE_BIAS), %g1 1: lduw [%o0], %g3 - sub %g3, %g1, %g7 + subcc %g3, %g1, %g7 + +#ifdef CONFIG_PAX_REFCOUNT + tvs %icc, 6 +#endif + cas [%o0], %g3, %g7 cmp %g3, %g7 bne,pn %icc, 1b @@ -143,7 +173,12 @@ __downgrade_write: or %g1, %lo(RWSEM_WAITING_BIAS), %g1 1: lduw [%o0], %g3 - sub %g3, %g1, %g7 + subcc %g3, %g1, %g7 + +#ifdef CONFIG_PAX_REFCOUNT + tvs %icc, 6 +#endif + cas [%o0], %g3, %g7 cmp %g3, %g7 bne,pn %icc, 1b diff -urNp linux-2.6.32.48/arch/sparc/Makefile linux-2.6.32.48-openvz/arch/sparc/Makefile --- linux-2.6.32.48/arch/sparc/Makefile 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/Makefile 2011-11-17 20:08:53.000000000 -0500 @@ -75,7 +75,7 @@ drivers-$(CONFIG_OPROFILE) += arch/sparc # Export what is needed by arch/sparc/boot/Makefile export VMLINUX_INIT VMLINUX_MAIN VMLINUX_INIT := $(head-y) $(init-y) -VMLINUX_MAIN := $(core-y) kernel/ mm/ fs/ ipc/ security/ crypto/ block/ +VMLINUX_MAIN := $(core-y) kernel/ mm/ fs/ ipc/ security/ crypto/ block/ grsecurity/ VMLINUX_MAIN += $(patsubst %/, %/lib.a, $(libs-y)) $(libs-y) VMLINUX_MAIN += $(drivers-y) $(net-y) diff -urNp linux-2.6.32.48/arch/sparc/mm/fault_32.c linux-2.6.32.48-openvz/arch/sparc/mm/fault_32.c --- linux-2.6.32.48/arch/sparc/mm/fault_32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/mm/fault_32.c 2011-11-17 20:08:53.000000000 -0500 @@ -21,6 +21,9 @@ #include #include #include +#include +#include +#include #include #include @@ -167,6 +170,267 @@ static unsigned long compute_si_addr(str return safe_compute_effective_address(regs, insn); } +#ifdef CONFIG_PAX_PAGEEXEC +#ifdef CONFIG_PAX_DLRESOLVE +static void pax_emuplt_close(struct vm_area_struct *vma) +{ + vma->vm_mm->call_dl_resolve = 0UL; +} + +static int pax_emuplt_fault(struct vm_area_struct *vma, struct vm_fault *vmf) +{ + unsigned int *kaddr; + + vmf->page = alloc_page(GFP_HIGHUSER); + if (!vmf->page) + return VM_FAULT_OOM; + + kaddr = kmap(vmf->page); + memset(kaddr, 0, PAGE_SIZE); + kaddr[0] = 0x9DE3BFA8U; /* save */ + flush_dcache_page(vmf->page); + kunmap(vmf->page); + return VM_FAULT_MAJOR; +} + +static const struct vm_operations_struct pax_vm_ops = { + .close = pax_emuplt_close, + .fault = pax_emuplt_fault +}; + +static int pax_insert_vma(struct vm_area_struct *vma, unsigned long addr) +{ + int ret; + + vma->vm_mm = current->mm; + vma->vm_start = addr; + vma->vm_end = addr + PAGE_SIZE; + vma->vm_flags = VM_READ | VM_EXEC | VM_MAYREAD | VM_MAYEXEC; + vma->vm_page_prot = vm_get_page_prot(vma->vm_flags); + vma->vm_ops = &pax_vm_ops; + + ret = insert_vm_struct(current->mm, vma); + if (ret) + return ret; + + ++current->mm->total_vm; + return 0; +} +#endif + +/* + * PaX: decide what to do with offenders (regs->pc = fault address) + * + * returns 1 when task should be killed + * 2 when patched PLT trampoline was detected + * 3 when unpatched PLT trampoline was detected + */ +static int pax_handle_fetch_fault(struct pt_regs *regs) +{ + +#ifdef CONFIG_PAX_EMUPLT + int err; + + do { /* PaX: patched PLT emulation #1 */ + unsigned int sethi1, sethi2, jmpl; + + err = get_user(sethi1, (unsigned int *)regs->pc); + err |= get_user(sethi2, (unsigned int *)(regs->pc+4)); + err |= get_user(jmpl, (unsigned int *)(regs->pc+8)); + + if (err) + break; + + if ((sethi1 & 0xFFC00000U) == 0x03000000U && + (sethi2 & 0xFFC00000U) == 0x03000000U && + (jmpl & 0xFFFFE000U) == 0x81C06000U) + { + unsigned int addr; + + regs->u_regs[UREG_G1] = (sethi2 & 0x003FFFFFU) << 10; + addr = regs->u_regs[UREG_G1]; + addr += (((jmpl | 0xFFFFE000U) ^ 0x00001000U) + 0x00001000U); + regs->pc = addr; + regs->npc = addr+4; + return 2; + } + } while (0); + + { /* PaX: patched PLT emulation #2 */ + unsigned int ba; + + err = get_user(ba, (unsigned int *)regs->pc); + + if (!err && (ba & 0xFFC00000U) == 0x30800000U) { + unsigned int addr; + + addr = regs->pc + ((((ba | 0xFFC00000U) ^ 0x00200000U) + 0x00200000U) << 2); + regs->pc = addr; + regs->npc = addr+4; + return 2; + } + } + + do { /* PaX: patched PLT emulation #3 */ + unsigned int sethi, jmpl, nop; + + err = get_user(sethi, (unsigned int *)regs->pc); + err |= get_user(jmpl, (unsigned int *)(regs->pc+4)); + err |= get_user(nop, (unsigned int *)(regs->pc+8)); + + if (err) + break; + + if ((sethi & 0xFFC00000U) == 0x03000000U && + (jmpl & 0xFFFFE000U) == 0x81C06000U && + nop == 0x01000000U) + { + unsigned int addr; + + addr = (sethi & 0x003FFFFFU) << 10; + regs->u_regs[UREG_G1] = addr; + addr += (((jmpl | 0xFFFFE000U) ^ 0x00001000U) + 0x00001000U); + regs->pc = addr; + regs->npc = addr+4; + return 2; + } + } while (0); + + do { /* PaX: unpatched PLT emulation step 1 */ + unsigned int sethi, ba, nop; + + err = get_user(sethi, (unsigned int *)regs->pc); + err |= get_user(ba, (unsigned int *)(regs->pc+4)); + err |= get_user(nop, (unsigned int *)(regs->pc+8)); + + if (err) + break; + + if ((sethi & 0xFFC00000U) == 0x03000000U && + ((ba & 0xFFC00000U) == 0x30800000U || (ba & 0xFFF80000U) == 0x30680000U) && + nop == 0x01000000U) + { + unsigned int addr, save, call; + + if ((ba & 0xFFC00000U) == 0x30800000U) + addr = regs->pc + 4 + ((((ba | 0xFFC00000U) ^ 0x00200000U) + 0x00200000U) << 2); + else + addr = regs->pc + 4 + ((((ba | 0xFFF80000U) ^ 0x00040000U) + 0x00040000U) << 2); + + err = get_user(save, (unsigned int *)addr); + err |= get_user(call, (unsigned int *)(addr+4)); + err |= get_user(nop, (unsigned int *)(addr+8)); + if (err) + break; + +#ifdef CONFIG_PAX_DLRESOLVE + if (save == 0x9DE3BFA8U && + (call & 0xC0000000U) == 0x40000000U && + nop == 0x01000000U) + { + struct vm_area_struct *vma; + unsigned long call_dl_resolve; + + down_read(¤t->mm->mmap_sem); + call_dl_resolve = current->mm->call_dl_resolve; + up_read(¤t->mm->mmap_sem); + if (likely(call_dl_resolve)) + goto emulate; + + vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); + + down_write(¤t->mm->mmap_sem); + if (current->mm->call_dl_resolve) { + call_dl_resolve = current->mm->call_dl_resolve; + up_write(¤t->mm->mmap_sem); + if (vma) + kmem_cache_free(vm_area_cachep, vma); + goto emulate; + } + + call_dl_resolve = get_unmapped_area(NULL, 0UL, PAGE_SIZE, 0UL, MAP_PRIVATE); + if (!vma || (call_dl_resolve & ~PAGE_MASK)) { + up_write(¤t->mm->mmap_sem); + if (vma) + kmem_cache_free(vm_area_cachep, vma); + return 1; + } + + if (pax_insert_vma(vma, call_dl_resolve)) { + up_write(¤t->mm->mmap_sem); + kmem_cache_free(vm_area_cachep, vma); + return 1; + } + + current->mm->call_dl_resolve = call_dl_resolve; + up_write(¤t->mm->mmap_sem); + +emulate: + regs->u_regs[UREG_G1] = (sethi & 0x003FFFFFU) << 10; + regs->pc = call_dl_resolve; + regs->npc = addr+4; + return 3; + } +#endif + + /* PaX: glibc 2.4+ generates sethi/jmpl instead of save/call */ + if ((save & 0xFFC00000U) == 0x05000000U && + (call & 0xFFFFE000U) == 0x85C0A000U && + nop == 0x01000000U) + { + regs->u_regs[UREG_G1] = (sethi & 0x003FFFFFU) << 10; + regs->u_regs[UREG_G2] = addr + 4; + addr = (save & 0x003FFFFFU) << 10; + addr += (((call | 0xFFFFE000U) ^ 0x00001000U) + 0x00001000U); + regs->pc = addr; + regs->npc = addr+4; + return 3; + } + } + } while (0); + + do { /* PaX: unpatched PLT emulation step 2 */ + unsigned int save, call, nop; + + err = get_user(save, (unsigned int *)(regs->pc-4)); + err |= get_user(call, (unsigned int *)regs->pc); + err |= get_user(nop, (unsigned int *)(regs->pc+4)); + if (err) + break; + + if (save == 0x9DE3BFA8U && + (call & 0xC0000000U) == 0x40000000U && + nop == 0x01000000U) + { + unsigned int dl_resolve = regs->pc + ((((call | 0xC0000000U) ^ 0x20000000U) + 0x20000000U) << 2); + + regs->u_regs[UREG_RETPC] = regs->pc; + regs->pc = dl_resolve; + regs->npc = dl_resolve+4; + return 3; + } + } while (0); +#endif + + return 1; +} + +void pax_report_insns(void *pc, void *sp) +{ + unsigned long i; + + printk(KERN_ERR "PAX: bytes at PC: "); + for (i = 0; i < 8; i++) { + unsigned int c; + if (get_user(c, (unsigned int *)pc+i)) + printk(KERN_CONT "???????? "); + else + printk(KERN_CONT "%08x ", c); + } + printk("\n"); +} +#endif + asmlinkage void do_sparc_fault(struct pt_regs *regs, int text_fault, int write, unsigned long address) { @@ -231,6 +495,24 @@ good_area: if(!(vma->vm_flags & VM_WRITE)) goto bad_area; } else { + +#ifdef CONFIG_PAX_PAGEEXEC + if ((mm->pax_flags & MF_PAX_PAGEEXEC) && text_fault && !(vma->vm_flags & VM_EXEC)) { + up_read(&mm->mmap_sem); + switch (pax_handle_fetch_fault(regs)) { + +#ifdef CONFIG_PAX_EMUPLT + case 2: + case 3: + return; +#endif + + } + pax_report_fault(regs, (void *)regs->pc, (void *)regs->u_regs[UREG_FP]); + do_group_exit(SIGKILL); + } +#endif + /* Allow reads even for write-only mappings */ if(!(vma->vm_flags & (VM_READ | VM_EXEC))) goto bad_area; diff -urNp linux-2.6.32.48/arch/sparc/mm/fault_64.c linux-2.6.32.48-openvz/arch/sparc/mm/fault_64.c --- linux-2.6.32.48/arch/sparc/mm/fault_64.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/mm/fault_64.c 2011-11-17 20:08:53.000000000 -0500 @@ -20,6 +20,9 @@ #include #include #include +#include +#include +#include #include #include @@ -78,7 +81,7 @@ static void bad_kernel_pc(struct pt_regs printk(KERN_CRIT "OOPS: Bogus kernel PC [%016lx] in fault handler\n", regs->tpc); printk(KERN_CRIT "OOPS: RPC [%016lx]\n", regs->u_regs[15]); - printk("OOPS: RPC <%pS>\n", (void *) regs->u_regs[15]); + printk("OOPS: RPC <%pA>\n", (void *) regs->u_regs[15]); printk(KERN_CRIT "OOPS: Fault was to vaddr[%lx]\n", vaddr); dump_stack(); unhandled_fault(regs->tpc, current, regs); @@ -249,6 +252,456 @@ static void noinline bogus_32bit_fault_a show_regs(regs); } +#ifdef CONFIG_PAX_PAGEEXEC +#ifdef CONFIG_PAX_DLRESOLVE +static void pax_emuplt_close(struct vm_area_struct *vma) +{ + vma->vm_mm->call_dl_resolve = 0UL; +} + +static int pax_emuplt_fault(struct vm_area_struct *vma, struct vm_fault *vmf) +{ + unsigned int *kaddr; + + vmf->page = alloc_page(GFP_HIGHUSER); + if (!vmf->page) + return VM_FAULT_OOM; + + kaddr = kmap(vmf->page); + memset(kaddr, 0, PAGE_SIZE); + kaddr[0] = 0x9DE3BFA8U; /* save */ + flush_dcache_page(vmf->page); + kunmap(vmf->page); + return VM_FAULT_MAJOR; +} + +static const struct vm_operations_struct pax_vm_ops = { + .close = pax_emuplt_close, + .fault = pax_emuplt_fault +}; + +static int pax_insert_vma(struct vm_area_struct *vma, unsigned long addr) +{ + int ret; + + vma->vm_mm = current->mm; + vma->vm_start = addr; + vma->vm_end = addr + PAGE_SIZE; + vma->vm_flags = VM_READ | VM_EXEC | VM_MAYREAD | VM_MAYEXEC; + vma->vm_page_prot = vm_get_page_prot(vma->vm_flags); + vma->vm_ops = &pax_vm_ops; + + ret = insert_vm_struct(current->mm, vma); + if (ret) + return ret; + + ++current->mm->total_vm; + return 0; +} +#endif + +/* + * PaX: decide what to do with offenders (regs->tpc = fault address) + * + * returns 1 when task should be killed + * 2 when patched PLT trampoline was detected + * 3 when unpatched PLT trampoline was detected + */ +static int pax_handle_fetch_fault(struct pt_regs *regs) +{ + +#ifdef CONFIG_PAX_EMUPLT + int err; + + do { /* PaX: patched PLT emulation #1 */ + unsigned int sethi1, sethi2, jmpl; + + err = get_user(sethi1, (unsigned int *)regs->tpc); + err |= get_user(sethi2, (unsigned int *)(regs->tpc+4)); + err |= get_user(jmpl, (unsigned int *)(regs->tpc+8)); + + if (err) + break; + + if ((sethi1 & 0xFFC00000U) == 0x03000000U && + (sethi2 & 0xFFC00000U) == 0x03000000U && + (jmpl & 0xFFFFE000U) == 0x81C06000U) + { + unsigned long addr; + + regs->u_regs[UREG_G1] = (sethi2 & 0x003FFFFFU) << 10; + addr = regs->u_regs[UREG_G1]; + addr += (((jmpl | 0xFFFFFFFFFFFFE000UL) ^ 0x00001000UL) + 0x00001000UL); + + if (test_thread_flag(TIF_32BIT)) + addr &= 0xFFFFFFFFUL; + + regs->tpc = addr; + regs->tnpc = addr+4; + return 2; + } + } while (0); + + { /* PaX: patched PLT emulation #2 */ + unsigned int ba; + + err = get_user(ba, (unsigned int *)regs->tpc); + + if (!err && (ba & 0xFFC00000U) == 0x30800000U) { + unsigned long addr; + + addr = regs->tpc + ((((ba | 0xFFFFFFFFFFC00000UL) ^ 0x00200000UL) + 0x00200000UL) << 2); + + if (test_thread_flag(TIF_32BIT)) + addr &= 0xFFFFFFFFUL; + + regs->tpc = addr; + regs->tnpc = addr+4; + return 2; + } + } + + do { /* PaX: patched PLT emulation #3 */ + unsigned int sethi, jmpl, nop; + + err = get_user(sethi, (unsigned int *)regs->tpc); + err |= get_user(jmpl, (unsigned int *)(regs->tpc+4)); + err |= get_user(nop, (unsigned int *)(regs->tpc+8)); + + if (err) + break; + + if ((sethi & 0xFFC00000U) == 0x03000000U && + (jmpl & 0xFFFFE000U) == 0x81C06000U && + nop == 0x01000000U) + { + unsigned long addr; + + addr = (sethi & 0x003FFFFFU) << 10; + regs->u_regs[UREG_G1] = addr; + addr += (((jmpl | 0xFFFFFFFFFFFFE000UL) ^ 0x00001000UL) + 0x00001000UL); + + if (test_thread_flag(TIF_32BIT)) + addr &= 0xFFFFFFFFUL; + + regs->tpc = addr; + regs->tnpc = addr+4; + return 2; + } + } while (0); + + do { /* PaX: patched PLT emulation #4 */ + unsigned int sethi, mov1, call, mov2; + + err = get_user(sethi, (unsigned int *)regs->tpc); + err |= get_user(mov1, (unsigned int *)(regs->tpc+4)); + err |= get_user(call, (unsigned int *)(regs->tpc+8)); + err |= get_user(mov2, (unsigned int *)(regs->tpc+12)); + + if (err) + break; + + if ((sethi & 0xFFC00000U) == 0x03000000U && + mov1 == 0x8210000FU && + (call & 0xC0000000U) == 0x40000000U && + mov2 == 0x9E100001U) + { + unsigned long addr; + + regs->u_regs[UREG_G1] = regs->u_regs[UREG_RETPC]; + addr = regs->tpc + 4 + ((((call | 0xFFFFFFFFC0000000UL) ^ 0x20000000UL) + 0x20000000UL) << 2); + + if (test_thread_flag(TIF_32BIT)) + addr &= 0xFFFFFFFFUL; + + regs->tpc = addr; + regs->tnpc = addr+4; + return 2; + } + } while (0); + + do { /* PaX: patched PLT emulation #5 */ + unsigned int sethi, sethi1, sethi2, or1, or2, sllx, jmpl, nop; + + err = get_user(sethi, (unsigned int *)regs->tpc); + err |= get_user(sethi1, (unsigned int *)(regs->tpc+4)); + err |= get_user(sethi2, (unsigned int *)(regs->tpc+8)); + err |= get_user(or1, (unsigned int *)(regs->tpc+12)); + err |= get_user(or2, (unsigned int *)(regs->tpc+16)); + err |= get_user(sllx, (unsigned int *)(regs->tpc+20)); + err |= get_user(jmpl, (unsigned int *)(regs->tpc+24)); + err |= get_user(nop, (unsigned int *)(regs->tpc+28)); + + if (err) + break; + + if ((sethi & 0xFFC00000U) == 0x03000000U && + (sethi1 & 0xFFC00000U) == 0x03000000U && + (sethi2 & 0xFFC00000U) == 0x0B000000U && + (or1 & 0xFFFFE000U) == 0x82106000U && + (or2 & 0xFFFFE000U) == 0x8A116000U && + sllx == 0x83287020U && + jmpl == 0x81C04005U && + nop == 0x01000000U) + { + unsigned long addr; + + regs->u_regs[UREG_G1] = ((sethi1 & 0x003FFFFFU) << 10) | (or1 & 0x000003FFU); + regs->u_regs[UREG_G1] <<= 32; + regs->u_regs[UREG_G5] = ((sethi2 & 0x003FFFFFU) << 10) | (or2 & 0x000003FFU); + addr = regs->u_regs[UREG_G1] + regs->u_regs[UREG_G5]; + regs->tpc = addr; + regs->tnpc = addr+4; + return 2; + } + } while (0); + + do { /* PaX: patched PLT emulation #6 */ + unsigned int sethi, sethi1, sethi2, sllx, or, jmpl, nop; + + err = get_user(sethi, (unsigned int *)regs->tpc); + err |= get_user(sethi1, (unsigned int *)(regs->tpc+4)); + err |= get_user(sethi2, (unsigned int *)(regs->tpc+8)); + err |= get_user(sllx, (unsigned int *)(regs->tpc+12)); + err |= get_user(or, (unsigned int *)(regs->tpc+16)); + err |= get_user(jmpl, (unsigned int *)(regs->tpc+20)); + err |= get_user(nop, (unsigned int *)(regs->tpc+24)); + + if (err) + break; + + if ((sethi & 0xFFC00000U) == 0x03000000U && + (sethi1 & 0xFFC00000U) == 0x03000000U && + (sethi2 & 0xFFC00000U) == 0x0B000000U && + sllx == 0x83287020U && + (or & 0xFFFFE000U) == 0x8A116000U && + jmpl == 0x81C04005U && + nop == 0x01000000U) + { + unsigned long addr; + + regs->u_regs[UREG_G1] = (sethi1 & 0x003FFFFFU) << 10; + regs->u_regs[UREG_G1] <<= 32; + regs->u_regs[UREG_G5] = ((sethi2 & 0x003FFFFFU) << 10) | (or & 0x3FFU); + addr = regs->u_regs[UREG_G1] + regs->u_regs[UREG_G5]; + regs->tpc = addr; + regs->tnpc = addr+4; + return 2; + } + } while (0); + + do { /* PaX: unpatched PLT emulation step 1 */ + unsigned int sethi, ba, nop; + + err = get_user(sethi, (unsigned int *)regs->tpc); + err |= get_user(ba, (unsigned int *)(regs->tpc+4)); + err |= get_user(nop, (unsigned int *)(regs->tpc+8)); + + if (err) + break; + + if ((sethi & 0xFFC00000U) == 0x03000000U && + ((ba & 0xFFC00000U) == 0x30800000U || (ba & 0xFFF80000U) == 0x30680000U) && + nop == 0x01000000U) + { + unsigned long addr; + unsigned int save, call; + unsigned int sethi1, sethi2, or1, or2, sllx, add, jmpl; + + if ((ba & 0xFFC00000U) == 0x30800000U) + addr = regs->tpc + 4 + ((((ba | 0xFFFFFFFFFFC00000UL) ^ 0x00200000UL) + 0x00200000UL) << 2); + else + addr = regs->tpc + 4 + ((((ba | 0xFFFFFFFFFFF80000UL) ^ 0x00040000UL) + 0x00040000UL) << 2); + + if (test_thread_flag(TIF_32BIT)) + addr &= 0xFFFFFFFFUL; + + err = get_user(save, (unsigned int *)addr); + err |= get_user(call, (unsigned int *)(addr+4)); + err |= get_user(nop, (unsigned int *)(addr+8)); + if (err) + break; + +#ifdef CONFIG_PAX_DLRESOLVE + if (save == 0x9DE3BFA8U && + (call & 0xC0000000U) == 0x40000000U && + nop == 0x01000000U) + { + struct vm_area_struct *vma; + unsigned long call_dl_resolve; + + down_read(¤t->mm->mmap_sem); + call_dl_resolve = current->mm->call_dl_resolve; + up_read(¤t->mm->mmap_sem); + if (likely(call_dl_resolve)) + goto emulate; + + vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); + + down_write(¤t->mm->mmap_sem); + if (current->mm->call_dl_resolve) { + call_dl_resolve = current->mm->call_dl_resolve; + up_write(¤t->mm->mmap_sem); + if (vma) + kmem_cache_free(vm_area_cachep, vma); + goto emulate; + } + + call_dl_resolve = get_unmapped_area(NULL, 0UL, PAGE_SIZE, 0UL, MAP_PRIVATE); + if (!vma || (call_dl_resolve & ~PAGE_MASK)) { + up_write(¤t->mm->mmap_sem); + if (vma) + kmem_cache_free(vm_area_cachep, vma); + return 1; + } + + if (pax_insert_vma(vma, call_dl_resolve)) { + up_write(¤t->mm->mmap_sem); + kmem_cache_free(vm_area_cachep, vma); + return 1; + } + + current->mm->call_dl_resolve = call_dl_resolve; + up_write(¤t->mm->mmap_sem); + +emulate: + regs->u_regs[UREG_G1] = (sethi & 0x003FFFFFU) << 10; + regs->tpc = call_dl_resolve; + regs->tnpc = addr+4; + return 3; + } +#endif + + /* PaX: glibc 2.4+ generates sethi/jmpl instead of save/call */ + if ((save & 0xFFC00000U) == 0x05000000U && + (call & 0xFFFFE000U) == 0x85C0A000U && + nop == 0x01000000U) + { + regs->u_regs[UREG_G1] = (sethi & 0x003FFFFFU) << 10; + regs->u_regs[UREG_G2] = addr + 4; + addr = (save & 0x003FFFFFU) << 10; + addr += (((call | 0xFFFFFFFFFFFFE000UL) ^ 0x00001000UL) + 0x00001000UL); + + if (test_thread_flag(TIF_32BIT)) + addr &= 0xFFFFFFFFUL; + + regs->tpc = addr; + regs->tnpc = addr+4; + return 3; + } + + /* PaX: 64-bit PLT stub */ + err = get_user(sethi1, (unsigned int *)addr); + err |= get_user(sethi2, (unsigned int *)(addr+4)); + err |= get_user(or1, (unsigned int *)(addr+8)); + err |= get_user(or2, (unsigned int *)(addr+12)); + err |= get_user(sllx, (unsigned int *)(addr+16)); + err |= get_user(add, (unsigned int *)(addr+20)); + err |= get_user(jmpl, (unsigned int *)(addr+24)); + err |= get_user(nop, (unsigned int *)(addr+28)); + if (err) + break; + + if ((sethi1 & 0xFFC00000U) == 0x09000000U && + (sethi2 & 0xFFC00000U) == 0x0B000000U && + (or1 & 0xFFFFE000U) == 0x88112000U && + (or2 & 0xFFFFE000U) == 0x8A116000U && + sllx == 0x89293020U && + add == 0x8A010005U && + jmpl == 0x89C14000U && + nop == 0x01000000U) + { + regs->u_regs[UREG_G1] = (sethi & 0x003FFFFFU) << 10; + regs->u_regs[UREG_G4] = ((sethi1 & 0x003FFFFFU) << 10) | (or1 & 0x000003FFU); + regs->u_regs[UREG_G4] <<= 32; + regs->u_regs[UREG_G5] = ((sethi2 & 0x003FFFFFU) << 10) | (or2 & 0x000003FFU); + regs->u_regs[UREG_G5] += regs->u_regs[UREG_G4]; + regs->u_regs[UREG_G4] = addr + 24; + addr = regs->u_regs[UREG_G5]; + regs->tpc = addr; + regs->tnpc = addr+4; + return 3; + } + } + } while (0); + +#ifdef CONFIG_PAX_DLRESOLVE + do { /* PaX: unpatched PLT emulation step 2 */ + unsigned int save, call, nop; + + err = get_user(save, (unsigned int *)(regs->tpc-4)); + err |= get_user(call, (unsigned int *)regs->tpc); + err |= get_user(nop, (unsigned int *)(regs->tpc+4)); + if (err) + break; + + if (save == 0x9DE3BFA8U && + (call & 0xC0000000U) == 0x40000000U && + nop == 0x01000000U) + { + unsigned long dl_resolve = regs->tpc + ((((call | 0xFFFFFFFFC0000000UL) ^ 0x20000000UL) + 0x20000000UL) << 2); + + if (test_thread_flag(TIF_32BIT)) + dl_resolve &= 0xFFFFFFFFUL; + + regs->u_regs[UREG_RETPC] = regs->tpc; + regs->tpc = dl_resolve; + regs->tnpc = dl_resolve+4; + return 3; + } + } while (0); +#endif + + do { /* PaX: patched PLT emulation #7, must be AFTER the unpatched PLT emulation */ + unsigned int sethi, ba, nop; + + err = get_user(sethi, (unsigned int *)regs->tpc); + err |= get_user(ba, (unsigned int *)(regs->tpc+4)); + err |= get_user(nop, (unsigned int *)(regs->tpc+8)); + + if (err) + break; + + if ((sethi & 0xFFC00000U) == 0x03000000U && + (ba & 0xFFF00000U) == 0x30600000U && + nop == 0x01000000U) + { + unsigned long addr; + + addr = (sethi & 0x003FFFFFU) << 10; + regs->u_regs[UREG_G1] = addr; + addr = regs->tpc + ((((ba | 0xFFFFFFFFFFF80000UL) ^ 0x00040000UL) + 0x00040000UL) << 2); + + if (test_thread_flag(TIF_32BIT)) + addr &= 0xFFFFFFFFUL; + + regs->tpc = addr; + regs->tnpc = addr+4; + return 2; + } + } while (0); + +#endif + + return 1; +} + +void pax_report_insns(void *pc, void *sp) +{ + unsigned long i; + + printk(KERN_ERR "PAX: bytes at PC: "); + for (i = 0; i < 8; i++) { + unsigned int c; + if (get_user(c, (unsigned int *)pc+i)) + printk(KERN_CONT "???????? "); + else + printk(KERN_CONT "%08x ", c); + } + printk("\n"); +} +#endif + asmlinkage void __kprobes do_sparc64_fault(struct pt_regs *regs) { struct mm_struct *mm = current->mm; @@ -315,6 +768,29 @@ asmlinkage void __kprobes do_sparc64_fau if (!vma) goto bad_area; +#ifdef CONFIG_PAX_PAGEEXEC + /* PaX: detect ITLB misses on non-exec pages */ + if ((mm->pax_flags & MF_PAX_PAGEEXEC) && vma->vm_start <= address && + !(vma->vm_flags & VM_EXEC) && (fault_code & FAULT_CODE_ITLB)) + { + if (address != regs->tpc) + goto good_area; + + up_read(&mm->mmap_sem); + switch (pax_handle_fetch_fault(regs)) { + +#ifdef CONFIG_PAX_EMUPLT + case 2: + case 3: + return; +#endif + + } + pax_report_fault(regs, (void *)regs->tpc, (void *)(regs->u_regs[UREG_FP] + STACK_BIAS)); + do_group_exit(SIGKILL); + } +#endif + /* Pure DTLB misses do not tell us whether the fault causing * load/store/atomic was a write or not, it only says that there * was no match. So in such a case we (carefully) read the diff -urNp linux-2.6.32.48/arch/sparc/mm/hugetlbpage.c linux-2.6.32.48-openvz/arch/sparc/mm/hugetlbpage.c --- linux-2.6.32.48/arch/sparc/mm/hugetlbpage.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/mm/hugetlbpage.c 2011-11-17 20:08:53.000000000 -0500 @@ -69,7 +69,7 @@ full_search: } return -ENOMEM; } - if (likely(!vma || addr + len <= vma->vm_start)) { + if (likely(check_heap_stack_gap(vma, addr, len))) { /* * Remember the place where we stopped the search: */ @@ -108,7 +108,7 @@ hugetlb_get_unmapped_area_topdown(struct /* make sure it can fit in the remaining address space */ if (likely(addr > len)) { vma = find_vma(mm, addr-len); - if (!vma || addr <= vma->vm_start) { + if (check_heap_stack_gap(vma, addr - len, len)) { /* remember the address as a hint for next time */ return (mm->free_area_cache = addr-len); } @@ -117,16 +117,17 @@ hugetlb_get_unmapped_area_topdown(struct if (unlikely(mm->mmap_base < len)) goto bottomup; - addr = (mm->mmap_base-len) & HPAGE_MASK; + addr = mm->mmap_base - len; do { + addr &= HPAGE_MASK; /* * Lookup failure means no vma is above this address, * else if new region fits below vma->vm_start, * return with success: */ vma = find_vma(mm, addr); - if (likely(!vma || addr+len <= vma->vm_start)) { + if (likely(check_heap_stack_gap(vma, addr, len))) { /* remember the address as a hint for next time */ return (mm->free_area_cache = addr); } @@ -136,8 +137,8 @@ hugetlb_get_unmapped_area_topdown(struct mm->cached_hole_size = vma->vm_start - addr; /* try just below the current vma->vm_start */ - addr = (vma->vm_start-len) & HPAGE_MASK; - } while (likely(len < vma->vm_start)); + addr = skip_heap_stack_gap(vma, len); + } while (!IS_ERR_VALUE(addr)); bottomup: /* @@ -183,8 +184,7 @@ hugetlb_get_unmapped_area(struct file *f if (addr) { addr = ALIGN(addr, HPAGE_SIZE); vma = find_vma(mm, addr); - if (task_size - len >= addr && - (!vma || addr + len <= vma->vm_start)) + if (task_size - len >= addr && check_heap_stack_gap(vma, addr, len)) return addr; } if (mm->get_unmapped_area == arch_get_unmapped_area) diff -urNp linux-2.6.32.48/arch/sparc/mm/init_32.c linux-2.6.32.48-openvz/arch/sparc/mm/init_32.c --- linux-2.6.32.48/arch/sparc/mm/init_32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/mm/init_32.c 2011-11-17 20:08:53.000000000 -0500 @@ -317,6 +317,9 @@ extern void device_scan(void); pgprot_t PAGE_SHARED __read_mostly; EXPORT_SYMBOL(PAGE_SHARED); +pgprot_t PAGE_SHARED_NOEXEC __read_mostly; +EXPORT_SYMBOL(PAGE_SHARED_NOEXEC); + void __init paging_init(void) { switch(sparc_cpu_model) { @@ -345,17 +348,17 @@ void __init paging_init(void) /* Initialize the protection map with non-constant, MMU dependent values. */ protection_map[0] = PAGE_NONE; - protection_map[1] = PAGE_READONLY; - protection_map[2] = PAGE_COPY; - protection_map[3] = PAGE_COPY; + protection_map[1] = PAGE_READONLY_NOEXEC; + protection_map[2] = PAGE_COPY_NOEXEC; + protection_map[3] = PAGE_COPY_NOEXEC; protection_map[4] = PAGE_READONLY; protection_map[5] = PAGE_READONLY; protection_map[6] = PAGE_COPY; protection_map[7] = PAGE_COPY; protection_map[8] = PAGE_NONE; - protection_map[9] = PAGE_READONLY; - protection_map[10] = PAGE_SHARED; - protection_map[11] = PAGE_SHARED; + protection_map[9] = PAGE_READONLY_NOEXEC; + protection_map[10] = PAGE_SHARED_NOEXEC; + protection_map[11] = PAGE_SHARED_NOEXEC; protection_map[12] = PAGE_READONLY; protection_map[13] = PAGE_READONLY; protection_map[14] = PAGE_SHARED; diff -urNp linux-2.6.32.48/arch/sparc/mm/Makefile linux-2.6.32.48-openvz/arch/sparc/mm/Makefile --- linux-2.6.32.48/arch/sparc/mm/Makefile 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/mm/Makefile 2011-11-17 20:08:53.000000000 -0500 @@ -2,7 +2,7 @@ # asflags-y := -ansi -ccflags-y := -Werror +#ccflags-y := -Werror obj-$(CONFIG_SPARC64) += ultra.o tlb.o tsb.o obj-y += fault_$(BITS).o diff -urNp linux-2.6.32.48/arch/sparc/mm/srmmu.c linux-2.6.32.48-openvz/arch/sparc/mm/srmmu.c --- linux-2.6.32.48/arch/sparc/mm/srmmu.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/sparc/mm/srmmu.c 2011-11-17 20:08:53.000000000 -0500 @@ -2200,6 +2200,13 @@ void __init ld_mmu_srmmu(void) PAGE_SHARED = pgprot_val(SRMMU_PAGE_SHARED); BTFIXUPSET_INT(page_copy, pgprot_val(SRMMU_PAGE_COPY)); BTFIXUPSET_INT(page_readonly, pgprot_val(SRMMU_PAGE_RDONLY)); + +#ifdef CONFIG_PAX_PAGEEXEC + PAGE_SHARED_NOEXEC = pgprot_val(SRMMU_PAGE_SHARED_NOEXEC); + BTFIXUPSET_INT(page_copy_noexec, pgprot_val(SRMMU_PAGE_COPY_NOEXEC)); + BTFIXUPSET_INT(page_readonly_noexec, pgprot_val(SRMMU_PAGE_RDONLY_NOEXEC)); +#endif + BTFIXUPSET_INT(page_kernel, pgprot_val(SRMMU_PAGE_KERNEL)); page_kernel = pgprot_val(SRMMU_PAGE_KERNEL); diff -urNp linux-2.6.32.48/arch/um/include/asm/kmap_types.h linux-2.6.32.48-openvz/arch/um/include/asm/kmap_types.h --- linux-2.6.32.48/arch/um/include/asm/kmap_types.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/um/include/asm/kmap_types.h 2011-11-17 20:08:53.000000000 -0500 @@ -23,6 +23,7 @@ enum km_type { KM_IRQ1, KM_SOFTIRQ0, KM_SOFTIRQ1, + KM_CLEARPAGE, KM_TYPE_NR }; diff -urNp linux-2.6.32.48/arch/um/include/asm/page.h linux-2.6.32.48-openvz/arch/um/include/asm/page.h --- linux-2.6.32.48/arch/um/include/asm/page.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/um/include/asm/page.h 2011-11-17 20:08:53.000000000 -0500 @@ -14,6 +14,9 @@ #define PAGE_SIZE (_AC(1, UL) << PAGE_SHIFT) #define PAGE_MASK (~(PAGE_SIZE-1)) +#define ktla_ktva(addr) (addr) +#define ktva_ktla(addr) (addr) + #ifndef __ASSEMBLY__ struct page; diff -urNp linux-2.6.32.48/arch/um/kernel/process.c linux-2.6.32.48-openvz/arch/um/kernel/process.c --- linux-2.6.32.48/arch/um/kernel/process.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/um/kernel/process.c 2011-11-17 20:08:53.000000000 -0500 @@ -393,22 +393,6 @@ int singlestepping(void * t) return 2; } -/* - * Only x86 and x86_64 have an arch_align_stack(). - * All other arches have "#define arch_align_stack(x) (x)" - * in their asm/system.h - * As this is included in UML from asm-um/system-generic.h, - * we can use it to behave as the subarch does. - */ -#ifndef arch_align_stack -unsigned long arch_align_stack(unsigned long sp) -{ - if (!(current->personality & ADDR_NO_RANDOMIZE) && randomize_va_space) - sp -= get_random_int() % 8192; - return sp & ~0xf; -} -#endif - unsigned long get_wchan(struct task_struct *p) { unsigned long stack_page, sp, ip; diff -urNp linux-2.6.32.48/arch/um/sys-i386/syscalls.c linux-2.6.32.48-openvz/arch/um/sys-i386/syscalls.c --- linux-2.6.32.48/arch/um/sys-i386/syscalls.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/um/sys-i386/syscalls.c 2011-11-17 20:08:53.000000000 -0500 @@ -11,6 +11,21 @@ #include "asm/uaccess.h" #include "asm/unistd.h" +int i386_mmap_check(unsigned long addr, unsigned long len, unsigned long flags) +{ + unsigned long pax_task_size = TASK_SIZE; + +#ifdef CONFIG_PAX_SEGMEXEC + if (current->mm->pax_flags & MF_PAX_SEGMEXEC) + pax_task_size = SEGMEXEC_TASK_SIZE; +#endif + + if (len > pax_task_size || addr > pax_task_size - len) + return -EINVAL; + + return 0; +} + /* * Perform the select(nd, in, out, ex, tv) and mmap() system * calls. Linux/i386 didn't use to be able to handle more than diff -urNp linux-2.6.32.48/arch/x86/boot/bitops.h linux-2.6.32.48-openvz/arch/x86/boot/bitops.h --- linux-2.6.32.48/arch/x86/boot/bitops.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/boot/bitops.h 2011-11-17 20:08:53.000000000 -0500 @@ -26,7 +26,7 @@ static inline int variable_test_bit(int u8 v; const u32 *p = (const u32 *)addr; - asm("btl %2,%1; setc %0" : "=qm" (v) : "m" (*p), "Ir" (nr)); + asm volatile("btl %2,%1; setc %0" : "=qm" (v) : "m" (*p), "Ir" (nr)); return v; } @@ -37,7 +37,7 @@ static inline int variable_test_bit(int static inline void set_bit(int nr, void *addr) { - asm("btsl %1,%0" : "+m" (*(u32 *)addr) : "Ir" (nr)); + asm volatile("btsl %1,%0" : "+m" (*(u32 *)addr) : "Ir" (nr)); } #endif /* BOOT_BITOPS_H */ diff -urNp linux-2.6.32.48/arch/x86/boot/boot.h linux-2.6.32.48-openvz/arch/x86/boot/boot.h --- linux-2.6.32.48/arch/x86/boot/boot.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/boot/boot.h 2011-11-17 20:08:53.000000000 -0500 @@ -82,7 +82,7 @@ static inline void io_delay(void) static inline u16 ds(void) { u16 seg; - asm("movw %%ds,%0" : "=rm" (seg)); + asm volatile("movw %%ds,%0" : "=rm" (seg)); return seg; } @@ -178,7 +178,7 @@ static inline void wrgs32(u32 v, addr_t static inline int memcmp(const void *s1, const void *s2, size_t len) { u8 diff; - asm("repe; cmpsb; setnz %0" + asm volatile("repe; cmpsb; setnz %0" : "=qm" (diff), "+D" (s1), "+S" (s2), "+c" (len)); return diff; } diff -urNp linux-2.6.32.48/arch/x86/boot/compressed/head_32.S linux-2.6.32.48-openvz/arch/x86/boot/compressed/head_32.S --- linux-2.6.32.48/arch/x86/boot/compressed/head_32.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/boot/compressed/head_32.S 2011-11-17 20:08:53.000000000 -0500 @@ -76,7 +76,7 @@ ENTRY(startup_32) notl %eax andl %eax, %ebx #else - movl $LOAD_PHYSICAL_ADDR, %ebx + movl $____LOAD_PHYSICAL_ADDR, %ebx #endif /* Target address to relocate to for decompression */ @@ -149,7 +149,7 @@ relocated: * and where it was actually loaded. */ movl %ebp, %ebx - subl $LOAD_PHYSICAL_ADDR, %ebx + subl $____LOAD_PHYSICAL_ADDR, %ebx jz 2f /* Nothing to be done if loaded at compiled addr. */ /* * Process relocations. @@ -157,8 +157,7 @@ relocated: 1: subl $4, %edi movl (%edi), %ecx - testl %ecx, %ecx - jz 2f + jecxz 2f addl %ebx, -__PAGE_OFFSET(%ebx, %ecx) jmp 1b 2: diff -urNp linux-2.6.32.48/arch/x86/boot/compressed/head_64.S linux-2.6.32.48-openvz/arch/x86/boot/compressed/head_64.S --- linux-2.6.32.48/arch/x86/boot/compressed/head_64.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/boot/compressed/head_64.S 2011-11-17 20:09:04.000000000 -0500 @@ -91,7 +91,7 @@ ENTRY(startup_32) notl %eax andl %eax, %ebx #else - movl $LOAD_PHYSICAL_ADDR, %ebx + movl $____LOAD_PHYSICAL_ADDR, %ebx #endif /* Target address to relocate to for decompression */ @@ -183,7 +183,7 @@ no_longmode: hlt jmp 1b -#include "../../kernel/verify_cpu_64.S" +#include "../../kernel/verify_cpu.S" /* * Be careful here startup_64 needs to be at a predictable @@ -234,7 +234,7 @@ ENTRY(startup_64) notq %rax andq %rax, %rbp #else - movq $LOAD_PHYSICAL_ADDR, %rbp + movq $____LOAD_PHYSICAL_ADDR, %rbp #endif /* Target address to relocate to for decompression */ diff -urNp linux-2.6.32.48/arch/x86/boot/compressed/Makefile linux-2.6.32.48-openvz/arch/x86/boot/compressed/Makefile --- linux-2.6.32.48/arch/x86/boot/compressed/Makefile 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/boot/compressed/Makefile 2011-11-17 20:09:04.000000000 -0500 @@ -13,6 +13,9 @@ cflags-$(CONFIG_X86_64) := -mcmodel=smal KBUILD_CFLAGS += $(cflags-y) KBUILD_CFLAGS += $(call cc-option,-ffreestanding) KBUILD_CFLAGS += $(call cc-option,-fno-stack-protector) +ifdef CONSTIFY_PLUGIN +KBUILD_CFLAGS += $(CONSTIFY_PLUGIN) -fplugin-arg-constify_plugin-no-constify +endif KBUILD_AFLAGS := $(KBUILD_CFLAGS) -D__ASSEMBLY__ GCOV_PROFILE := n diff -urNp linux-2.6.32.48/arch/x86/boot/compressed/misc.c linux-2.6.32.48-openvz/arch/x86/boot/compressed/misc.c --- linux-2.6.32.48/arch/x86/boot/compressed/misc.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/boot/compressed/misc.c 2011-11-17 20:08:53.000000000 -0500 @@ -288,7 +288,7 @@ static void parse_elf(void *output) case PT_LOAD: #ifdef CONFIG_RELOCATABLE dest = output; - dest += (phdr->p_paddr - LOAD_PHYSICAL_ADDR); + dest += (phdr->p_paddr - ____LOAD_PHYSICAL_ADDR); #else dest = (void *)(phdr->p_paddr); #endif @@ -335,7 +335,7 @@ asmlinkage void decompress_kernel(void * error("Destination address too large"); #endif #ifndef CONFIG_RELOCATABLE - if ((unsigned long)output != LOAD_PHYSICAL_ADDR) + if ((unsigned long)output != ____LOAD_PHYSICAL_ADDR) error("Wrong destination address"); #endif diff -urNp linux-2.6.32.48/arch/x86/boot/compressed/mkpiggy.c linux-2.6.32.48-openvz/arch/x86/boot/compressed/mkpiggy.c --- linux-2.6.32.48/arch/x86/boot/compressed/mkpiggy.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/boot/compressed/mkpiggy.c 2011-11-17 20:08:53.000000000 -0500 @@ -74,7 +74,7 @@ int main(int argc, char *argv[]) offs = (olen > ilen) ? olen - ilen : 0; offs += olen >> 12; /* Add 8 bytes for each 32K block */ - offs += 32*1024 + 18; /* Add 32K + 18 bytes slack */ + offs += 64*1024; /* Add 64K bytes slack */ offs = (offs+4095) & ~4095; /* Round to a 4K boundary */ printf(".section \".rodata.compressed\",\"a\",@progbits\n"); diff -urNp linux-2.6.32.48/arch/x86/boot/compressed/relocs.c linux-2.6.32.48-openvz/arch/x86/boot/compressed/relocs.c --- linux-2.6.32.48/arch/x86/boot/compressed/relocs.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/boot/compressed/relocs.c 2011-11-17 20:08:53.000000000 -0500 @@ -10,8 +10,11 @@ #define USE_BSD #include +#include "../../../../include/linux/autoconf.h" + #define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0])) static Elf32_Ehdr ehdr; +static Elf32_Phdr *phdr; static unsigned long reloc_count, reloc_idx; static unsigned long *relocs; @@ -37,7 +40,7 @@ static const char* safe_abs_relocs[] = { static int is_safe_abs_reloc(const char* sym_name) { - int i; + unsigned int i; for (i = 0; i < ARRAY_SIZE(safe_abs_relocs); i++) { if (!strcmp(sym_name, safe_abs_relocs[i])) @@ -245,9 +248,39 @@ static void read_ehdr(FILE *fp) } } +static void read_phdrs(FILE *fp) +{ + unsigned int i; + + phdr = calloc(ehdr.e_phnum, sizeof(Elf32_Phdr)); + if (!phdr) { + die("Unable to allocate %d program headers\n", + ehdr.e_phnum); + } + if (fseek(fp, ehdr.e_phoff, SEEK_SET) < 0) { + die("Seek to %d failed: %s\n", + ehdr.e_phoff, strerror(errno)); + } + if (fread(phdr, sizeof(*phdr), ehdr.e_phnum, fp) != ehdr.e_phnum) { + die("Cannot read ELF program headers: %s\n", + strerror(errno)); + } + for(i = 0; i < ehdr.e_phnum; i++) { + phdr[i].p_type = elf32_to_cpu(phdr[i].p_type); + phdr[i].p_offset = elf32_to_cpu(phdr[i].p_offset); + phdr[i].p_vaddr = elf32_to_cpu(phdr[i].p_vaddr); + phdr[i].p_paddr = elf32_to_cpu(phdr[i].p_paddr); + phdr[i].p_filesz = elf32_to_cpu(phdr[i].p_filesz); + phdr[i].p_memsz = elf32_to_cpu(phdr[i].p_memsz); + phdr[i].p_flags = elf32_to_cpu(phdr[i].p_flags); + phdr[i].p_align = elf32_to_cpu(phdr[i].p_align); + } + +} + static void read_shdrs(FILE *fp) { - int i; + unsigned int i; Elf32_Shdr shdr; secs = calloc(ehdr.e_shnum, sizeof(struct section)); @@ -282,7 +315,7 @@ static void read_shdrs(FILE *fp) static void read_strtabs(FILE *fp) { - int i; + unsigned int i; for (i = 0; i < ehdr.e_shnum; i++) { struct section *sec = &secs[i]; if (sec->shdr.sh_type != SHT_STRTAB) { @@ -307,7 +340,7 @@ static void read_strtabs(FILE *fp) static void read_symtabs(FILE *fp) { - int i,j; + unsigned int i,j; for (i = 0; i < ehdr.e_shnum; i++) { struct section *sec = &secs[i]; if (sec->shdr.sh_type != SHT_SYMTAB) { @@ -340,7 +373,9 @@ static void read_symtabs(FILE *fp) static void read_relocs(FILE *fp) { - int i,j; + unsigned int i,j; + uint32_t base; + for (i = 0; i < ehdr.e_shnum; i++) { struct section *sec = &secs[i]; if (sec->shdr.sh_type != SHT_REL) { @@ -360,9 +395,18 @@ static void read_relocs(FILE *fp) die("Cannot read symbol table: %s\n", strerror(errno)); } + base = 0; + for (j = 0; j < ehdr.e_phnum; j++) { + if (phdr[j].p_type != PT_LOAD ) + continue; + if (secs[sec->shdr.sh_info].shdr.sh_offset < phdr[j].p_offset || secs[sec->shdr.sh_info].shdr.sh_offset >= phdr[j].p_offset + phdr[j].p_filesz) + continue; + base = CONFIG_PAGE_OFFSET + phdr[j].p_paddr - phdr[j].p_vaddr; + break; + } for (j = 0; j < sec->shdr.sh_size/sizeof(Elf32_Rel); j++) { Elf32_Rel *rel = &sec->reltab[j]; - rel->r_offset = elf32_to_cpu(rel->r_offset); + rel->r_offset = elf32_to_cpu(rel->r_offset) + base; rel->r_info = elf32_to_cpu(rel->r_info); } } @@ -371,14 +415,14 @@ static void read_relocs(FILE *fp) static void print_absolute_symbols(void) { - int i; + unsigned int i; printf("Absolute symbols\n"); printf(" Num: Value Size Type Bind Visibility Name\n"); for (i = 0; i < ehdr.e_shnum; i++) { struct section *sec = &secs[i]; char *sym_strtab; Elf32_Sym *sh_symtab; - int j; + unsigned int j; if (sec->shdr.sh_type != SHT_SYMTAB) { continue; @@ -406,14 +450,14 @@ static void print_absolute_symbols(void) static void print_absolute_relocs(void) { - int i, printed = 0; + unsigned int i, printed = 0; for (i = 0; i < ehdr.e_shnum; i++) { struct section *sec = &secs[i]; struct section *sec_applies, *sec_symtab; char *sym_strtab; Elf32_Sym *sh_symtab; - int j; + unsigned int j; if (sec->shdr.sh_type != SHT_REL) { continue; } @@ -474,13 +518,13 @@ static void print_absolute_relocs(void) static void walk_relocs(void (*visit)(Elf32_Rel *rel, Elf32_Sym *sym)) { - int i; + unsigned int i; /* Walk through the relocations */ for (i = 0; i < ehdr.e_shnum; i++) { char *sym_strtab; Elf32_Sym *sh_symtab; struct section *sec_applies, *sec_symtab; - int j; + unsigned int j; struct section *sec = &secs[i]; if (sec->shdr.sh_type != SHT_REL) { @@ -504,6 +548,21 @@ static void walk_relocs(void (*visit)(El if (sym->st_shndx == SHN_ABS) { continue; } + /* Don't relocate actual per-cpu variables, they are absolute indices, not addresses */ + if (!strcmp(sec_name(sym->st_shndx), ".data.percpu") && strcmp(sym_name(sym_strtab, sym), "__per_cpu_load")) + continue; + +#if defined(CONFIG_PAX_KERNEXEC) && defined(CONFIG_X86_32) + /* Don't relocate actual code, they are relocated implicitly by the base address of KERNEL_CS */ + if (!strcmp(sec_name(sym->st_shndx), ".module.text") && !strcmp(sym_name(sym_strtab, sym), "_etext")) + continue; + if (!strcmp(sec_name(sym->st_shndx), ".init.text")) + continue; + if (!strcmp(sec_name(sym->st_shndx), ".exit.text")) + continue; + if (!strcmp(sec_name(sym->st_shndx), ".text") && strcmp(sym_name(sym_strtab, sym), "__LOAD_PHYSICAL_ADDR")) + continue; +#endif if (r_type == R_386_NONE || r_type == R_386_PC32) { /* * NONE can be ignored and and PC relative @@ -541,7 +600,7 @@ static int cmp_relocs(const void *va, co static void emit_relocs(int as_text) { - int i; + unsigned int i; /* Count how many relocations I have and allocate space for them. */ reloc_count = 0; walk_relocs(count_reloc); @@ -634,6 +693,7 @@ int main(int argc, char **argv) fname, strerror(errno)); } read_ehdr(fp); + read_phdrs(fp); read_shdrs(fp); read_strtabs(fp); read_symtabs(fp); diff -urNp linux-2.6.32.48/arch/x86/boot/cpucheck.c linux-2.6.32.48-openvz/arch/x86/boot/cpucheck.c --- linux-2.6.32.48/arch/x86/boot/cpucheck.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/boot/cpucheck.c 2011-11-17 20:08:53.000000000 -0500 @@ -74,7 +74,7 @@ static int has_fpu(void) u16 fcw = -1, fsw = -1; u32 cr0; - asm("movl %%cr0,%0" : "=r" (cr0)); + asm volatile("movl %%cr0,%0" : "=r" (cr0)); if (cr0 & (X86_CR0_EM|X86_CR0_TS)) { cr0 &= ~(X86_CR0_EM|X86_CR0_TS); asm volatile("movl %0,%%cr0" : : "r" (cr0)); @@ -90,7 +90,7 @@ static int has_eflag(u32 mask) { u32 f0, f1; - asm("pushfl ; " + asm volatile("pushfl ; " "pushfl ; " "popl %0 ; " "movl %0,%1 ; " @@ -115,7 +115,7 @@ static void get_flags(void) set_bit(X86_FEATURE_FPU, cpu.flags); if (has_eflag(X86_EFLAGS_ID)) { - asm("cpuid" + asm volatile("cpuid" : "=a" (max_intel_level), "=b" (cpu_vendor[0]), "=d" (cpu_vendor[1]), @@ -124,7 +124,7 @@ static void get_flags(void) if (max_intel_level >= 0x00000001 && max_intel_level <= 0x0000ffff) { - asm("cpuid" + asm volatile("cpuid" : "=a" (tfms), "=c" (cpu.flags[4]), "=d" (cpu.flags[0]) @@ -136,7 +136,7 @@ static void get_flags(void) cpu.model += ((tfms >> 16) & 0xf) << 4; } - asm("cpuid" + asm volatile("cpuid" : "=a" (max_amd_level) : "a" (0x80000000) : "ebx", "ecx", "edx"); @@ -144,7 +144,7 @@ static void get_flags(void) if (max_amd_level >= 0x80000001 && max_amd_level <= 0x8000ffff) { u32 eax = 0x80000001; - asm("cpuid" + asm volatile("cpuid" : "+a" (eax), "=c" (cpu.flags[6]), "=d" (cpu.flags[1]) @@ -203,9 +203,9 @@ int check_cpu(int *cpu_level_ptr, int *r u32 ecx = MSR_K7_HWCR; u32 eax, edx; - asm("rdmsr" : "=a" (eax), "=d" (edx) : "c" (ecx)); + asm volatile("rdmsr" : "=a" (eax), "=d" (edx) : "c" (ecx)); eax &= ~(1 << 15); - asm("wrmsr" : : "a" (eax), "d" (edx), "c" (ecx)); + asm volatile("wrmsr" : : "a" (eax), "d" (edx), "c" (ecx)); get_flags(); /* Make sure it really did something */ err = check_flags(); @@ -218,9 +218,9 @@ int check_cpu(int *cpu_level_ptr, int *r u32 ecx = MSR_VIA_FCR; u32 eax, edx; - asm("rdmsr" : "=a" (eax), "=d" (edx) : "c" (ecx)); + asm volatile("rdmsr" : "=a" (eax), "=d" (edx) : "c" (ecx)); eax |= (1<<1)|(1<<7); - asm("wrmsr" : : "a" (eax), "d" (edx), "c" (ecx)); + asm volatile("wrmsr" : : "a" (eax), "d" (edx), "c" (ecx)); set_bit(X86_FEATURE_CX8, cpu.flags); err = check_flags(); @@ -231,12 +231,12 @@ int check_cpu(int *cpu_level_ptr, int *r u32 eax, edx; u32 level = 1; - asm("rdmsr" : "=a" (eax), "=d" (edx) : "c" (ecx)); - asm("wrmsr" : : "a" (~0), "d" (edx), "c" (ecx)); - asm("cpuid" + asm volatile("rdmsr" : "=a" (eax), "=d" (edx) : "c" (ecx)); + asm volatile("wrmsr" : : "a" (~0), "d" (edx), "c" (ecx)); + asm volatile("cpuid" : "+a" (level), "=d" (cpu.flags[0]) : : "ecx", "ebx"); - asm("wrmsr" : : "a" (eax), "d" (edx), "c" (ecx)); + asm volatile("wrmsr" : : "a" (eax), "d" (edx), "c" (ecx)); err = check_flags(); } diff -urNp linux-2.6.32.48/arch/x86/boot/header.S linux-2.6.32.48-openvz/arch/x86/boot/header.S --- linux-2.6.32.48/arch/x86/boot/header.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/boot/header.S 2011-11-17 20:08:53.000000000 -0500 @@ -224,7 +224,7 @@ setup_data: .quad 0 # 64-bit physical # single linked list of # struct setup_data -pref_address: .quad LOAD_PHYSICAL_ADDR # preferred load addr +pref_address: .quad ____LOAD_PHYSICAL_ADDR # preferred load addr #define ZO_INIT_SIZE (ZO__end - ZO_startup_32 + ZO_z_extract_offset) #define VO_INIT_SIZE (VO__end - VO__text) diff -urNp linux-2.6.32.48/arch/x86/boot/Makefile linux-2.6.32.48-openvz/arch/x86/boot/Makefile --- linux-2.6.32.48/arch/x86/boot/Makefile 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/boot/Makefile 2011-11-17 20:09:04.000000000 -0500 @@ -69,6 +69,9 @@ KBUILD_CFLAGS := $(LINUXINCLUDE) -g -Os $(call cc-option, -fno-stack-protector) \ $(call cc-option, -mpreferred-stack-boundary=2) KBUILD_CFLAGS += $(call cc-option, -m32) +ifdef CONSTIFY_PLUGIN +KBUILD_CFLAGS += $(CONSTIFY_PLUGIN) -fplugin-arg-constify_plugin-no-constify +endif KBUILD_AFLAGS := $(KBUILD_CFLAGS) -D__ASSEMBLY__ GCOV_PROFILE := n diff -urNp linux-2.6.32.48/arch/x86/boot/memory.c linux-2.6.32.48-openvz/arch/x86/boot/memory.c --- linux-2.6.32.48/arch/x86/boot/memory.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/boot/memory.c 2011-11-17 20:08:53.000000000 -0500 @@ -19,7 +19,7 @@ static int detect_memory_e820(void) { - int count = 0; + unsigned int count = 0; struct biosregs ireg, oreg; struct e820entry *desc = boot_params.e820_map; static struct e820entry buf; /* static so it is zeroed */ diff -urNp linux-2.6.32.48/arch/x86/boot/video.c linux-2.6.32.48-openvz/arch/x86/boot/video.c --- linux-2.6.32.48/arch/x86/boot/video.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/boot/video.c 2011-11-17 20:08:53.000000000 -0500 @@ -90,7 +90,7 @@ static void store_mode_params(void) static unsigned int get_entry(void) { char entry_buf[4]; - int i, len = 0; + unsigned int i, len = 0; int key; unsigned int v; diff -urNp linux-2.6.32.48/arch/x86/boot/video-vesa.c linux-2.6.32.48-openvz/arch/x86/boot/video-vesa.c --- linux-2.6.32.48/arch/x86/boot/video-vesa.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/boot/video-vesa.c 2011-11-17 20:08:53.000000000 -0500 @@ -200,6 +200,7 @@ static void vesa_store_pm_info(void) boot_params.screen_info.vesapm_seg = oreg.es; boot_params.screen_info.vesapm_off = oreg.di; + boot_params.screen_info.vesapm_size = oreg.cx; } /* diff -urNp linux-2.6.32.48/arch/x86/crypto/aes-x86_64-asm_64.S linux-2.6.32.48-openvz/arch/x86/crypto/aes-x86_64-asm_64.S --- linux-2.6.32.48/arch/x86/crypto/aes-x86_64-asm_64.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/crypto/aes-x86_64-asm_64.S 2011-11-17 20:09:04.000000000 -0500 @@ -8,6 +8,8 @@ * including this sentence is retained in full. */ +#include + .extern crypto_ft_tab .extern crypto_it_tab .extern crypto_fl_tab @@ -71,6 +73,8 @@ FUNC: movq r1,r2; \ je B192; \ leaq 32(r9),r9; +#define ret pax_force_retaddr; ret + #define epilogue(r1,r2,r3,r4,r5,r6,r7,r8,r9) \ movq r1,r2; \ movq r3,r4; \ diff -urNp linux-2.6.32.48/arch/x86/crypto/salsa20-x86_64-asm_64.S linux-2.6.32.48-openvz/arch/x86/crypto/salsa20-x86_64-asm_64.S --- linux-2.6.32.48/arch/x86/crypto/salsa20-x86_64-asm_64.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/crypto/salsa20-x86_64-asm_64.S 2011-11-17 20:09:04.000000000 -0500 @@ -1,3 +1,5 @@ +#include + # enter ECRYPT_encrypt_bytes .text .p2align 5 @@ -790,6 +792,7 @@ ECRYPT_encrypt_bytes: add %r11,%rsp mov %rdi,%rax mov %rsi,%rdx + pax_force_retaddr ret # bytesatleast65: ._bytesatleast65: @@ -891,6 +894,7 @@ ECRYPT_keysetup: add %r11,%rsp mov %rdi,%rax mov %rsi,%rdx + pax_force_retaddr ret # enter ECRYPT_ivsetup .text @@ -917,4 +921,5 @@ ECRYPT_ivsetup: add %r11,%rsp mov %rdi,%rax mov %rsi,%rdx + pax_force_retaddr ret diff -urNp linux-2.6.32.48/arch/x86/crypto/twofish-x86_64-asm_64.S linux-2.6.32.48-openvz/arch/x86/crypto/twofish-x86_64-asm_64.S --- linux-2.6.32.48/arch/x86/crypto/twofish-x86_64-asm_64.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/crypto/twofish-x86_64-asm_64.S 2011-11-17 20:09:04.000000000 -0500 @@ -21,6 +21,7 @@ .text #include +#include #define a_offset 0 #define b_offset 4 @@ -269,6 +270,7 @@ twofish_enc_blk: popq R1 movq $1,%rax + pax_force_retaddr ret twofish_dec_blk: @@ -321,4 +323,5 @@ twofish_dec_blk: popq R1 movq $1,%rax + pax_force_retaddr ret diff -urNp linux-2.6.32.48/arch/x86/ia32/ia32_aout.c linux-2.6.32.48-openvz/arch/x86/ia32/ia32_aout.c --- linux-2.6.32.48/arch/x86/ia32/ia32_aout.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/ia32/ia32_aout.c 2011-11-17 20:08:53.000000000 -0500 @@ -169,6 +169,8 @@ static int aout_core_dump(long signr, st unsigned long dump_start, dump_size; struct user32 dump; + memset(&dump, 0, sizeof(dump)); + fs = get_fs(); set_fs(KERNEL_DS); has_dumped = 1; @@ -218,12 +220,6 @@ static int aout_core_dump(long signr, st dump_size = dump.u_ssize << PAGE_SHIFT; DUMP_WRITE(dump_start, dump_size); } - /* - * Finally dump the task struct. Not be used by gdb, but - * could be useful - */ - set_fs(KERNEL_DS); - DUMP_WRITE(current, sizeof(*current)); end_coredump: set_fs(fs); return has_dumped; diff -urNp linux-2.6.32.48/arch/x86/ia32/ia32entry.S linux-2.6.32.48-openvz/arch/x86/ia32/ia32entry.S --- linux-2.6.32.48/arch/x86/ia32/ia32entry.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/ia32/ia32entry.S 2011-11-17 20:12:39.000000000 -0500 @@ -13,6 +13,7 @@ #include #include #include +#include #include /* Avoid __ASSEMBLER__'ifying just for this. */ @@ -93,6 +94,29 @@ ENTRY(native_irq_enable_sysexit) ENDPROC(native_irq_enable_sysexit) #endif + .macro pax_enter_kernel_user +#ifdef CONFIG_PAX_MEMORY_UDEREF + call pax_enter_kernel_user +#endif + .endm + + .macro pax_exit_kernel_user +#ifdef CONFIG_PAX_MEMORY_UDEREF + call pax_exit_kernel_user +#endif +#ifdef CONFIG_PAX_RANDKSTACK + pushq %rax + call pax_randomize_kstack + popq %rax +#endif + .endm + +.macro pax_erase_kstack +#ifdef CONFIG_PAX_MEMORY_STACKLEAK + call pax_erase_kstack +#endif +.endm + /* * 32bit SYSENTER instruction entry. * @@ -119,7 +143,7 @@ ENTRY(ia32_sysenter_target) CFI_REGISTER rsp,rbp SWAPGS_UNSAFE_STACK movq PER_CPU_VAR(kernel_stack), %rsp - addq $(KERNEL_STACK_OFFSET),%rsp + pax_enter_kernel_user /* * No need to follow this irqs on/off section: the syscall * disabled irqs, here we enable it straight after entry: @@ -135,7 +159,8 @@ ENTRY(ia32_sysenter_target) pushfq CFI_ADJUST_CFA_OFFSET 8 /*CFI_REL_OFFSET rflags,0*/ - movl 8*3-THREAD_SIZE+TI_sysenter_return(%rsp), %r10d + GET_THREAD_INFO(%r10) + movl TI_sysenter_return(%r10), %r10d CFI_REGISTER rip,r10 pushq $__USER32_CS CFI_ADJUST_CFA_OFFSET 8 @@ -150,6 +175,12 @@ ENTRY(ia32_sysenter_target) SAVE_ARGS 0,0,1 /* no need to do an access_ok check here because rbp has been 32bit zero extended */ + +#ifdef CONFIG_PAX_MEMORY_UDEREF + mov $PAX_USER_SHADOW_BASE,%r10 + add %r10,%rbp +#endif + 1: movl (%rbp),%ebp .section __ex_table,"a" .quad 1b,ia32_badarg @@ -172,6 +203,8 @@ sysenter_dispatch: testl $_TIF_ALLWORK_MASK,TI_flags(%r10) jnz sysexit_audit sysexit_from_sys_call: + pax_exit_kernel_user + pax_erase_kstack andl $~TS_COMPAT,TI_status(%r10) /* clear IF, that popfq doesn't enable interrupts early */ andl $~0x200,EFLAGS-R11(%rsp) @@ -283,19 +316,24 @@ ENDPROC(ia32_sysenter_target) ENTRY(ia32_cstar_target) CFI_STARTPROC32 simple CFI_SIGNAL_FRAME - CFI_DEF_CFA rsp,KERNEL_STACK_OFFSET + CFI_DEF_CFA rsp,0 CFI_REGISTER rip,rcx /*CFI_REGISTER rflags,r11*/ SWAPGS_UNSAFE_STACK movl %esp,%r8d CFI_REGISTER rsp,r8 movq PER_CPU_VAR(kernel_stack),%rsp + +#ifdef CONFIG_PAX_MEMORY_UDEREF + pax_enter_kernel_user +#endif + /* * No need to follow this irqs on/off section: the syscall * disabled irqs and here we enable it straight after entry: */ ENABLE_INTERRUPTS(CLBR_NONE) - SAVE_ARGS 8,1,1 + SAVE_ARGS 8*6,1,1 movl %eax,%eax /* zero extension */ movq %rax,ORIG_RAX-ARGOFFSET(%rsp) movq %rcx,RIP-ARGOFFSET(%rsp) @@ -311,6 +349,12 @@ ENTRY(ia32_cstar_target) /* no need to do an access_ok check here because r8 has been 32bit zero extended */ /* hardware stack frame is complete now */ + +#ifdef CONFIG_PAX_MEMORY_UDEREF + mov $PAX_USER_SHADOW_BASE,%r10 + add %r10,%r8 +#endif + 1: movl (%r8),%r9d .section __ex_table,"a" .quad 1b,ia32_badarg @@ -333,6 +377,8 @@ cstar_dispatch: testl $_TIF_ALLWORK_MASK,TI_flags(%r10) jnz sysretl_audit sysretl_from_sys_call: + pax_exit_kernel_user + pax_erase_kstack andl $~TS_COMPAT,TI_status(%r10) RESTORE_ARGS 1,-ARG_SKIP,1,1,1 movl RIP-ARGOFFSET(%rsp),%ecx @@ -415,6 +461,7 @@ ENTRY(ia32_syscall) CFI_REL_OFFSET rip,RIP-RIP PARAVIRT_ADJUST_EXCEPTION_FRAME SWAPGS + pax_enter_kernel_user /* * No need to follow this irqs on/off section: the syscall * disabled irqs and here we enable it straight after entry: @@ -623,7 +670,7 @@ ia32_sys_call_table: .quad stub32_iopl /* 110 */ .quad sys_vhangup .quad quiet_ni_syscall /* old "idle" system call */ - .quad sys32_vm86_warning /* vm86old */ + .quad quiet_ni_syscall /* vm86old */ .quad compat_sys_wait4 .quad sys_swapoff /* 115 */ .quad compat_sys_sysinfo @@ -676,7 +723,7 @@ ia32_sys_call_table: .quad sys_mremap .quad sys_setresuid16 .quad sys_getresuid16 /* 165 */ - .quad sys32_vm86_warning /* vm86 */ + .quad quiet_ni_syscall /* vm86 */ .quad quiet_ni_syscall /* query_module */ .quad sys_poll .quad compat_sys_nfsservctl @@ -847,4 +894,25 @@ ia32_sys_call_table: .quad compat_sys_pwritev .quad compat_sys_rt_tgsigqueueinfo /* 335 */ .quad sys_perf_event_open + .rept 500-(.-ia32_sys_call_table)/8 + .quad sys_ni_syscall + .endr + .quad sys_fairsched_mknod /* 500 */ + .quad sys_fairsched_rmnod + .quad sys_fairsched_chwt + .quad sys_fairsched_mvpr + .quad sys_fairsched_rate + .quad sys_fairsched_vcpus /* 505 */ + .quad sys_ni_syscall + .quad sys_ni_syscall + .quad sys_ni_syscall + .quad sys_ni_syscall + .quad sys_getluid /* 510 */ + .quad sys_setluid + .quad compat_sys_setublimit + .quad compat_sys_ubstat + .quad sys_ni_syscall + .quad sys_ni_syscall /* 515 */ + .quad sys_lchmod + .quad compat_sys_lutime ia32_syscall_end: diff -urNp linux-2.6.32.48/arch/x86/ia32/ia32_signal.c linux-2.6.32.48-openvz/arch/x86/ia32/ia32_signal.c --- linux-2.6.32.48/arch/x86/ia32/ia32_signal.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/ia32/ia32_signal.c 2011-11-17 20:09:04.000000000 -0500 @@ -167,7 +167,7 @@ asmlinkage long sys32_sigaltstack(const } seg = get_fs(); set_fs(KERNEL_DS); - ret = do_sigaltstack(uss_ptr ? &uss : NULL, &uoss, regs->sp); + ret = do_sigaltstack(uss_ptr ? (const stack_t __force_user *)&uss : NULL, (stack_t __force_user *)&uoss, regs->sp); set_fs(seg); if (ret >= 0 && uoss_ptr) { if (!access_ok(VERIFY_WRITE, uoss_ptr, sizeof(stack_ia32_t))) @@ -374,7 +374,7 @@ static int ia32_setup_sigcontext(struct */ static void __user *get_sigframe(struct k_sigaction *ka, struct pt_regs *regs, size_t frame_size, - void **fpstate) + void __user **fpstate) { unsigned long sp; @@ -395,7 +395,7 @@ static void __user *get_sigframe(struct if (used_math()) { sp = sp - sig_xstate_ia32_size; - *fpstate = (struct _fpstate_ia32 *) sp; + *fpstate = (struct _fpstate_ia32 __user *) sp; if (save_i387_xstate_ia32(*fpstate) < 0) return (void __user *) -1L; } @@ -403,7 +403,7 @@ static void __user *get_sigframe(struct sp -= frame_size; /* Align the stack pointer according to the i386 ABI, * i.e. so that on function entry ((sp + 4) & 15) == 0. */ - sp = ((sp + 4) & -16ul) - 4; + sp = ((sp - 12) & -16ul) - 4; return (void __user *) sp; } @@ -461,7 +461,7 @@ int ia32_setup_frame(int sig, struct k_s * These are actually not used anymore, but left because some * gdb versions depend on them as a marker. */ - put_user_ex(*((u64 *)&code), (u64 *)frame->retcode); + put_user_ex(*((const u64 *)&code), (u64 __user *)frame->retcode); } put_user_catch(err); if (err) @@ -503,7 +503,7 @@ int ia32_setup_rt_frame(int sig, struct 0xb8, __NR_ia32_rt_sigreturn, 0x80cd, - 0, + 0 }; frame = get_sigframe(ka, regs, sizeof(*frame), &fpstate); @@ -533,16 +533,18 @@ int ia32_setup_rt_frame(int sig, struct if (ka->sa.sa_flags & SA_RESTORER) restorer = ka->sa.sa_restorer; + else if (current->mm->context.vdso) + /* Return stub is in 32bit vsyscall page */ + restorer = VDSO32_SYMBOL(current->mm->context.vdso, rt_sigreturn); else - restorer = VDSO32_SYMBOL(current->mm->context.vdso, - rt_sigreturn); + restorer = &frame->retcode; put_user_ex(ptr_to_compat(restorer), &frame->pretcode); /* * Not actually used anymore, but left because some gdb * versions need it. */ - put_user_ex(*((u64 *)&code), (u64 *)frame->retcode); + put_user_ex(*((const u64 *)&code), (u64 __user *)frame->retcode); } put_user_catch(err); if (err) diff -urNp linux-2.6.32.48/arch/x86/ia32/sys_ia32.c linux-2.6.32.48-openvz/arch/x86/ia32/sys_ia32.c --- linux-2.6.32.48/arch/x86/ia32/sys_ia32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/ia32/sys_ia32.c 2011-11-17 20:09:04.000000000 -0500 @@ -69,8 +69,8 @@ asmlinkage long sys32_ftruncate64(unsign */ static int cp_stat64(struct stat64 __user *ubuf, struct kstat *stat) { - typeof(ubuf->st_uid) uid = 0; - typeof(ubuf->st_gid) gid = 0; + typeof(((struct stat64 *)0)->st_uid) uid = 0; + typeof(((struct stat64 *)0)->st_gid) gid = 0; SET_UID(uid, stat->uid); SET_GID(gid, stat->gid); if (!access_ok(VERIFY_WRITE, ubuf, sizeof(struct stat64)) || @@ -308,8 +308,8 @@ asmlinkage long sys32_rt_sigprocmask(int } set_fs(KERNEL_DS); ret = sys_rt_sigprocmask(how, - set ? (sigset_t __user *)&s : NULL, - oset ? (sigset_t __user *)&s : NULL, + set ? (sigset_t __force_user *)&s : NULL, + oset ? (sigset_t __force_user *)&s : NULL, sigsetsize); set_fs(old_fs); if (ret) @@ -371,7 +371,7 @@ asmlinkage long sys32_sched_rr_get_inter mm_segment_t old_fs = get_fs(); set_fs(KERNEL_DS); - ret = sys_sched_rr_get_interval(pid, (struct timespec __user *)&t); + ret = sys_sched_rr_get_interval(pid, (struct timespec __force_user *)&t); set_fs(old_fs); if (put_compat_timespec(&t, interval)) return -EFAULT; @@ -387,7 +387,7 @@ asmlinkage long sys32_rt_sigpending(comp mm_segment_t old_fs = get_fs(); set_fs(KERNEL_DS); - ret = sys_rt_sigpending((sigset_t __user *)&s, sigsetsize); + ret = sys_rt_sigpending((sigset_t __force_user *)&s, sigsetsize); set_fs(old_fs); if (!ret) { switch (_NSIG_WORDS) { @@ -412,7 +412,7 @@ asmlinkage long sys32_rt_sigqueueinfo(in if (copy_siginfo_from_user32(&info, uinfo)) return -EFAULT; set_fs(KERNEL_DS); - ret = sys_rt_sigqueueinfo(pid, sig, (siginfo_t __user *)&info); + ret = sys_rt_sigqueueinfo(pid, sig, (siginfo_t __force_user *)&info); set_fs(old_fs); return ret; } @@ -513,7 +513,7 @@ asmlinkage long sys32_sendfile(int out_f return -EFAULT; set_fs(KERNEL_DS); - ret = sys_sendfile(out_fd, in_fd, offset ? (off_t __user *)&of : NULL, + ret = sys_sendfile(out_fd, in_fd, offset ? (off_t __force_user *)&of : NULL, count); set_fs(old_fs); @@ -623,20 +623,6 @@ long sys32_fadvise64_64(int fd, __u32 of advice); } -long sys32_vm86_warning(void) -{ - struct task_struct *me = current; - static char lastcomm[sizeof(me->comm)]; - - if (strncmp(lastcomm, me->comm, sizeof(lastcomm))) { - compat_printk(KERN_INFO - "%s: vm86 mode not supported on 64 bit kernel\n", - me->comm); - strncpy(lastcomm, me->comm, sizeof(lastcomm)); - } - return -ENOSYS; -} - long sys32_lookup_dcookie(u32 addr_low, u32 addr_high, char __user *buf, size_t len) { diff -urNp linux-2.6.32.48/arch/x86/include/asm/alternative-asm.h linux-2.6.32.48-openvz/arch/x86/include/asm/alternative-asm.h --- linux-2.6.32.48/arch/x86/include/asm/alternative-asm.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/alternative-asm.h 2011-11-17 20:09:04.000000000 -0500 @@ -19,4 +19,18 @@ .endm #endif +#ifdef CONFIG_PAX_KERNEXEC_PLUGIN + .macro pax_force_retaddr rip=0 + btsq $63,\rip(%rsp) + .endm + .macro pax_force_fptr ptr + btsq $63,\ptr + .endm +#else + .macro pax_force_retaddr rip=0 + .endm + .macro pax_force_fptr ptr + .endm +#endif + #endif /* __ASSEMBLY__ */ diff -urNp linux-2.6.32.48/arch/x86/include/asm/alternative.h linux-2.6.32.48-openvz/arch/x86/include/asm/alternative.h --- linux-2.6.32.48/arch/x86/include/asm/alternative.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/alternative.h 2011-11-17 20:08:53.000000000 -0500 @@ -85,7 +85,7 @@ static inline void alternatives_smp_swit " .byte 662b-661b\n" /* sourcelen */ \ " .byte 664f-663f\n" /* replacementlen */ \ ".previous\n" \ - ".section .altinstr_replacement, \"ax\"\n" \ + ".section .altinstr_replacement, \"a\"\n" \ "663:\n\t" newinstr "\n664:\n" /* replacement */ \ ".previous" diff -urNp linux-2.6.32.48/arch/x86/include/asm/apic.h linux-2.6.32.48-openvz/arch/x86/include/asm/apic.h --- linux-2.6.32.48/arch/x86/include/asm/apic.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/apic.h 2011-11-17 20:09:04.000000000 -0500 @@ -46,7 +46,7 @@ static inline void generic_apic_probe(vo #ifdef CONFIG_X86_LOCAL_APIC -extern unsigned int apic_verbosity; +extern int apic_verbosity; extern int local_apic_timer_c2_ok; extern int disable_apic; diff -urNp linux-2.6.32.48/arch/x86/include/asm/apm.h linux-2.6.32.48-openvz/arch/x86/include/asm/apm.h --- linux-2.6.32.48/arch/x86/include/asm/apm.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/apm.h 2011-11-17 20:08:53.000000000 -0500 @@ -34,7 +34,7 @@ static inline void apm_bios_call_asm(u32 __asm__ __volatile__(APM_DO_ZERO_SEGS "pushl %%edi\n\t" "pushl %%ebp\n\t" - "lcall *%%cs:apm_bios_entry\n\t" + "lcall *%%ss:apm_bios_entry\n\t" "setc %%al\n\t" "popl %%ebp\n\t" "popl %%edi\n\t" @@ -58,7 +58,7 @@ static inline u8 apm_bios_call_simple_as __asm__ __volatile__(APM_DO_ZERO_SEGS "pushl %%edi\n\t" "pushl %%ebp\n\t" - "lcall *%%cs:apm_bios_entry\n\t" + "lcall *%%ss:apm_bios_entry\n\t" "setc %%bl\n\t" "popl %%ebp\n\t" "popl %%edi\n\t" diff -urNp linux-2.6.32.48/arch/x86/include/asm/atomic_32.h linux-2.6.32.48-openvz/arch/x86/include/asm/atomic_32.h --- linux-2.6.32.48/arch/x86/include/asm/atomic_32.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/atomic_32.h 2011-11-17 20:08:53.000000000 -0500 @@ -25,6 +25,17 @@ static inline int atomic_read(const atom } /** + * atomic_read_unchecked - read atomic variable + * @v: pointer of type atomic_unchecked_t + * + * Atomically reads the value of @v. + */ +static inline int atomic_read_unchecked(const atomic_unchecked_t *v) +{ + return v->counter; +} + +/** * atomic_set - set atomic variable * @v: pointer of type atomic_t * @i: required value @@ -37,6 +48,18 @@ static inline void atomic_set(atomic_t * } /** + * atomic_set_unchecked - set atomic variable + * @v: pointer of type atomic_unchecked_t + * @i: required value + * + * Atomically sets the value of @v to @i. + */ +static inline void atomic_set_unchecked(atomic_unchecked_t *v, int i) +{ + v->counter = i; +} + +/** * atomic_add - add integer to atomic variable * @i: integer value to add * @v: pointer of type atomic_t @@ -45,7 +68,29 @@ static inline void atomic_set(atomic_t * */ static inline void atomic_add(int i, atomic_t *v) { - asm volatile(LOCK_PREFIX "addl %1,%0" + asm volatile(LOCK_PREFIX "addl %1,%0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + LOCK_PREFIX "subl %1,%0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + + : "+m" (v->counter) + : "ir" (i)); +} + +/** + * atomic_add_unchecked - add integer to atomic variable + * @i: integer value to add + * @v: pointer of type atomic_unchecked_t + * + * Atomically adds @i to @v. + */ +static inline void atomic_add_unchecked(int i, atomic_unchecked_t *v) +{ + asm volatile(LOCK_PREFIX "addl %1,%0\n" : "+m" (v->counter) : "ir" (i)); } @@ -59,7 +104,29 @@ static inline void atomic_add(int i, ato */ static inline void atomic_sub(int i, atomic_t *v) { - asm volatile(LOCK_PREFIX "subl %1,%0" + asm volatile(LOCK_PREFIX "subl %1,%0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + LOCK_PREFIX "addl %1,%0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + + : "+m" (v->counter) + : "ir" (i)); +} + +/** + * atomic_sub_unchecked - subtract integer from atomic variable + * @i: integer value to subtract + * @v: pointer of type atomic_unchecked_t + * + * Atomically subtracts @i from @v. + */ +static inline void atomic_sub_unchecked(int i, atomic_unchecked_t *v) +{ + asm volatile(LOCK_PREFIX "subl %1,%0\n" : "+m" (v->counter) : "ir" (i)); } @@ -77,7 +144,16 @@ static inline int atomic_sub_and_test(in { unsigned char c; - asm volatile(LOCK_PREFIX "subl %2,%0; sete %1" + asm volatile(LOCK_PREFIX "subl %2,%0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + LOCK_PREFIX "addl %2,%0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + + "sete %1\n" : "+m" (v->counter), "=qm" (c) : "ir" (i) : "memory"); return c; @@ -91,7 +167,27 @@ static inline int atomic_sub_and_test(in */ static inline void atomic_inc(atomic_t *v) { - asm volatile(LOCK_PREFIX "incl %0" + asm volatile(LOCK_PREFIX "incl %0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + LOCK_PREFIX "decl %0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + + : "+m" (v->counter)); +} + +/** + * atomic_inc_unchecked - increment atomic variable + * @v: pointer of type atomic_unchecked_t + * + * Atomically increments @v by 1. + */ +static inline void atomic_inc_unchecked(atomic_unchecked_t *v) +{ + asm volatile(LOCK_PREFIX "incl %0\n" : "+m" (v->counter)); } @@ -103,7 +199,27 @@ static inline void atomic_inc(atomic_t * */ static inline void atomic_dec(atomic_t *v) { - asm volatile(LOCK_PREFIX "decl %0" + asm volatile(LOCK_PREFIX "decl %0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + LOCK_PREFIX "incl %0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + + : "+m" (v->counter)); +} + +/** + * atomic_dec_unchecked - decrement atomic variable + * @v: pointer of type atomic_unchecked_t + * + * Atomically decrements @v by 1. + */ +static inline void atomic_dec_unchecked(atomic_unchecked_t *v) +{ + asm volatile(LOCK_PREFIX "decl %0\n" : "+m" (v->counter)); } @@ -119,7 +235,16 @@ static inline int atomic_dec_and_test(at { unsigned char c; - asm volatile(LOCK_PREFIX "decl %0; sete %1" + asm volatile(LOCK_PREFIX "decl %0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + LOCK_PREFIX "incl %0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + + "sete %1\n" : "+m" (v->counter), "=qm" (c) : : "memory"); return c != 0; @@ -137,7 +262,35 @@ static inline int atomic_inc_and_test(at { unsigned char c; - asm volatile(LOCK_PREFIX "incl %0; sete %1" + asm volatile(LOCK_PREFIX "incl %0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + LOCK_PREFIX "decl %0\n" + "into\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + + "sete %1\n" + : "+m" (v->counter), "=qm" (c) + : : "memory"); + return c != 0; +} + +/** + * atomic_inc_and_test_unchecked - increment and test + * @v: pointer of type atomic_unchecked_t + * + * Atomically increments @v by 1 + * and returns true if the result is zero, or false for all + * other cases. + */ +static inline int atomic_inc_and_test_unchecked(atomic_unchecked_t *v) +{ + unsigned char c; + + asm volatile(LOCK_PREFIX "incl %0\n" + "sete %1\n" : "+m" (v->counter), "=qm" (c) : : "memory"); return c != 0; @@ -156,7 +309,16 @@ static inline int atomic_add_negative(in { unsigned char c; - asm volatile(LOCK_PREFIX "addl %2,%0; sets %1" + asm volatile(LOCK_PREFIX "addl %2,%0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + LOCK_PREFIX "subl %2,%0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + + "sets %1\n" : "+m" (v->counter), "=qm" (c) : "ir" (i) : "memory"); return c; @@ -179,6 +341,46 @@ static inline int atomic_add_return(int #endif /* Modern 486+ processor */ __i = i; + asm volatile(LOCK_PREFIX "xaddl %0, %1\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + "movl %0, %1\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + + : "+r" (i), "+m" (v->counter) + : : "memory"); + return i + __i; + +#ifdef CONFIG_M386 +no_xadd: /* Legacy 386 processor */ + local_irq_save(flags); + __i = atomic_read(v); + atomic_set(v, i + __i); + local_irq_restore(flags); + return i + __i; +#endif +} + +/** + * atomic_add_return_unchecked - add integer and return + * @v: pointer of type atomic_unchecked_t + * @i: integer value to add + * + * Atomically adds @i to @v and returns @i + @v + */ +static inline int atomic_add_return_unchecked(int i, atomic_unchecked_t *v) +{ + int __i; +#ifdef CONFIG_M386 + unsigned long flags; + if (unlikely(boot_cpu_data.x86 <= 3)) + goto no_xadd; +#endif + /* Modern 486+ processor */ + __i = i; asm volatile(LOCK_PREFIX "xaddl %0, %1" : "+r" (i), "+m" (v->counter) : : "memory"); @@ -211,11 +413,21 @@ static inline int atomic_cmpxchg(atomic_ return cmpxchg(&v->counter, old, new); } +static inline int atomic_cmpxchg_unchecked(atomic_unchecked_t *v, int old, int new) +{ + return cmpxchg(&v->counter, old, new); +} + static inline int atomic_xchg(atomic_t *v, int new) { return xchg(&v->counter, new); } +static inline int atomic_xchg_unchecked(atomic_unchecked_t *v, int new) +{ + return xchg(&v->counter, new); +} + /** * atomic_add_unless - add unless the number is already a given value * @v: pointer of type atomic_t @@ -227,22 +439,39 @@ static inline int atomic_xchg(atomic_t * */ static inline int atomic_add_unless(atomic_t *v, int a, int u) { - int c, old; + int c, old, new; c = atomic_read(v); for (;;) { - if (unlikely(c == (u))) + if (unlikely(c == u)) break; - old = atomic_cmpxchg((v), c, c + (a)); + + asm volatile("addl %2,%0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + "subl %2,%0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + + : "=r" (new) + : "0" (c), "ir" (a)); + + old = atomic_cmpxchg(v, c, new); if (likely(old == c)) break; c = old; } - return c != (u); + return c != u; } #define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0) #define atomic_inc_return(v) (atomic_add_return(1, v)) +static inline int atomic_inc_return_unchecked(atomic_unchecked_t *v) +{ + return atomic_add_return_unchecked(1, v); +} #define atomic_dec_return(v) (atomic_sub_return(1, v)) /* These are x86-specific, used by some header files */ @@ -266,9 +495,18 @@ typedef struct { u64 __aligned(8) counter; } atomic64_t; +#ifdef CONFIG_PAX_REFCOUNT +typedef struct { + u64 __aligned(8) counter; +} atomic64_unchecked_t; +#else +typedef atomic64_t atomic64_unchecked_t; +#endif + #define ATOMIC64_INIT(val) { (val) } extern u64 atomic64_cmpxchg(atomic64_t *ptr, u64 old_val, u64 new_val); +extern u64 atomic64_cmpxchg_unchecked(atomic64_unchecked_t *ptr, u64 old_val, u64 new_val); /** * atomic64_xchg - xchg atomic64 variable @@ -279,6 +517,7 @@ extern u64 atomic64_cmpxchg(atomic64_t * * the old value. */ extern u64 atomic64_xchg(atomic64_t *ptr, u64 new_val); +extern u64 atomic64_xchg_unchecked(atomic64_unchecked_t *ptr, u64 new_val); /** * atomic64_set - set atomic64 variable @@ -290,6 +529,15 @@ extern u64 atomic64_xchg(atomic64_t *ptr extern void atomic64_set(atomic64_t *ptr, u64 new_val); /** + * atomic64_unchecked_set - set atomic64 variable + * @ptr: pointer to type atomic64_unchecked_t + * @new_val: value to assign + * + * Atomically sets the value of @ptr to @new_val. + */ +extern void atomic64_set_unchecked(atomic64_unchecked_t *ptr, u64 new_val); + +/** * atomic64_read - read atomic64 variable * @ptr: pointer to type atomic64_t * @@ -317,7 +565,33 @@ static inline u64 atomic64_read(atomic64 return res; } -extern u64 atomic64_read(atomic64_t *ptr); +/** + * atomic64_read_unchecked - read atomic64 variable + * @ptr: pointer to type atomic64_unchecked_t + * + * Atomically reads the value of @ptr and returns it. + */ +static inline u64 atomic64_read_unchecked(atomic64_unchecked_t *ptr) +{ + u64 res; + + /* + * Note, we inline this atomic64_unchecked_t primitive because + * it only clobbers EAX/EDX and leaves the others + * untouched. We also (somewhat subtly) rely on the + * fact that cmpxchg8b returns the current 64-bit value + * of the memory location we are touching: + */ + asm volatile( + "mov %%ebx, %%eax\n\t" + "mov %%ecx, %%edx\n\t" + LOCK_PREFIX "cmpxchg8b %1\n" + : "=&A" (res) + : "m" (*ptr) + ); + + return res; +} /** * atomic64_add_return - add and return @@ -332,8 +606,11 @@ extern u64 atomic64_add_return(u64 delta * Other variants with different arithmetic operators: */ extern u64 atomic64_sub_return(u64 delta, atomic64_t *ptr); +extern u64 atomic64_sub_return_unchecked(u64 delta, atomic64_unchecked_t *ptr); extern u64 atomic64_inc_return(atomic64_t *ptr); +extern u64 atomic64_inc_return_unchecked(atomic64_unchecked_t *ptr); extern u64 atomic64_dec_return(atomic64_t *ptr); +extern u64 atomic64_dec_return_unchecked(atomic64_unchecked_t *ptr); /** * atomic64_add - add integer to atomic64 variable @@ -345,6 +622,15 @@ extern u64 atomic64_dec_return(atomic64_ extern void atomic64_add(u64 delta, atomic64_t *ptr); /** + * atomic64_add_unchecked - add integer to atomic64 variable + * @delta: integer value to add + * @ptr: pointer to type atomic64_unchecked_t + * + * Atomically adds @delta to @ptr. + */ +extern void atomic64_add_unchecked(u64 delta, atomic64_unchecked_t *ptr); + +/** * atomic64_sub - subtract the atomic64 variable * @delta: integer value to subtract * @ptr: pointer to type atomic64_t @@ -354,6 +640,15 @@ extern void atomic64_add(u64 delta, atom extern void atomic64_sub(u64 delta, atomic64_t *ptr); /** + * atomic64_sub_unchecked - subtract the atomic64 variable + * @delta: integer value to subtract + * @ptr: pointer to type atomic64_unchecked_t + * + * Atomically subtracts @delta from @ptr. + */ +extern void atomic64_sub_unchecked(u64 delta, atomic64_unchecked_t *ptr); + +/** * atomic64_sub_and_test - subtract value from variable and test result * @delta: integer value to subtract * @ptr: pointer to type atomic64_t @@ -373,6 +668,14 @@ extern int atomic64_sub_and_test(u64 del extern void atomic64_inc(atomic64_t *ptr); /** + * atomic64_inc_unchecked - increment atomic64 variable + * @ptr: pointer to type atomic64_unchecked_t + * + * Atomically increments @ptr by 1. + */ +extern void atomic64_inc_unchecked(atomic64_unchecked_t *ptr); + +/** * atomic64_dec - decrement atomic64 variable * @ptr: pointer to type atomic64_t * @@ -381,6 +684,14 @@ extern void atomic64_inc(atomic64_t *ptr extern void atomic64_dec(atomic64_t *ptr); /** + * atomic64_dec_unchecked - decrement atomic64 variable + * @ptr: pointer to type atomic64_unchecked_t + * + * Atomically decrements @ptr by 1. + */ +extern void atomic64_dec_unchecked(atomic64_unchecked_t *ptr); + +/** * atomic64_dec_and_test - decrement and test * @ptr: pointer to type atomic64_t * diff -urNp linux-2.6.32.48/arch/x86/include/asm/atomic_64.h linux-2.6.32.48-openvz/arch/x86/include/asm/atomic_64.h --- linux-2.6.32.48/arch/x86/include/asm/atomic_64.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/atomic_64.h 2011-11-17 20:08:53.000000000 -0500 @@ -24,6 +24,17 @@ static inline int atomic_read(const atom } /** + * atomic_read_unchecked - read atomic variable + * @v: pointer of type atomic_unchecked_t + * + * Atomically reads the value of @v. + */ +static inline int atomic_read_unchecked(const atomic_unchecked_t *v) +{ + return v->counter; +} + +/** * atomic_set - set atomic variable * @v: pointer of type atomic_t * @i: required value @@ -36,6 +47,18 @@ static inline void atomic_set(atomic_t * } /** + * atomic_set_unchecked - set atomic variable + * @v: pointer of type atomic_unchecked_t + * @i: required value + * + * Atomically sets the value of @v to @i. + */ +static inline void atomic_set_unchecked(atomic_unchecked_t *v, int i) +{ + v->counter = i; +} + +/** * atomic_add - add integer to atomic variable * @i: integer value to add * @v: pointer of type atomic_t @@ -44,7 +67,29 @@ static inline void atomic_set(atomic_t * */ static inline void atomic_add(int i, atomic_t *v) { - asm volatile(LOCK_PREFIX "addl %1,%0" + asm volatile(LOCK_PREFIX "addl %1,%0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + LOCK_PREFIX "subl %1,%0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + + : "=m" (v->counter) + : "ir" (i), "m" (v->counter)); +} + +/** + * atomic_add_unchecked - add integer to atomic variable + * @i: integer value to add + * @v: pointer of type atomic_unchecked_t + * + * Atomically adds @i to @v. + */ +static inline void atomic_add_unchecked(int i, atomic_unchecked_t *v) +{ + asm volatile(LOCK_PREFIX "addl %1,%0\n" : "=m" (v->counter) : "ir" (i), "m" (v->counter)); } @@ -58,7 +103,29 @@ static inline void atomic_add(int i, ato */ static inline void atomic_sub(int i, atomic_t *v) { - asm volatile(LOCK_PREFIX "subl %1,%0" + asm volatile(LOCK_PREFIX "subl %1,%0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + LOCK_PREFIX "addl %1,%0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + + : "=m" (v->counter) + : "ir" (i), "m" (v->counter)); +} + +/** + * atomic_sub_unchecked - subtract the atomic variable + * @i: integer value to subtract + * @v: pointer of type atomic_unchecked_t + * + * Atomically subtracts @i from @v. + */ +static inline void atomic_sub_unchecked(int i, atomic_unchecked_t *v) +{ + asm volatile(LOCK_PREFIX "subl %1,%0\n" : "=m" (v->counter) : "ir" (i), "m" (v->counter)); } @@ -76,7 +143,16 @@ static inline int atomic_sub_and_test(in { unsigned char c; - asm volatile(LOCK_PREFIX "subl %2,%0; sete %1" + asm volatile(LOCK_PREFIX "subl %2,%0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + LOCK_PREFIX "addl %2,%0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + + "sete %1\n" : "=m" (v->counter), "=qm" (c) : "ir" (i), "m" (v->counter) : "memory"); return c; @@ -90,7 +166,28 @@ static inline int atomic_sub_and_test(in */ static inline void atomic_inc(atomic_t *v) { - asm volatile(LOCK_PREFIX "incl %0" + asm volatile(LOCK_PREFIX "incl %0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + LOCK_PREFIX "decl %0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + + : "=m" (v->counter) + : "m" (v->counter)); +} + +/** + * atomic_inc_unchecked - increment atomic variable + * @v: pointer of type atomic_unchecked_t + * + * Atomically increments @v by 1. + */ +static inline void atomic_inc_unchecked(atomic_unchecked_t *v) +{ + asm volatile(LOCK_PREFIX "incl %0\n" : "=m" (v->counter) : "m" (v->counter)); } @@ -103,7 +200,28 @@ static inline void atomic_inc(atomic_t * */ static inline void atomic_dec(atomic_t *v) { - asm volatile(LOCK_PREFIX "decl %0" + asm volatile(LOCK_PREFIX "decl %0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + LOCK_PREFIX "incl %0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + + : "=m" (v->counter) + : "m" (v->counter)); +} + +/** + * atomic_dec_unchecked - decrement atomic variable + * @v: pointer of type atomic_unchecked_t + * + * Atomically decrements @v by 1. + */ +static inline void atomic_dec_unchecked(atomic_unchecked_t *v) +{ + asm volatile(LOCK_PREFIX "decl %0\n" : "=m" (v->counter) : "m" (v->counter)); } @@ -120,7 +238,16 @@ static inline int atomic_dec_and_test(at { unsigned char c; - asm volatile(LOCK_PREFIX "decl %0; sete %1" + asm volatile(LOCK_PREFIX "decl %0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + LOCK_PREFIX "incl %0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + + "sete %1\n" : "=m" (v->counter), "=qm" (c) : "m" (v->counter) : "memory"); return c != 0; @@ -138,7 +265,35 @@ static inline int atomic_inc_and_test(at { unsigned char c; - asm volatile(LOCK_PREFIX "incl %0; sete %1" + asm volatile(LOCK_PREFIX "incl %0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + LOCK_PREFIX "decl %0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + + "sete %1\n" + : "=m" (v->counter), "=qm" (c) + : "m" (v->counter) : "memory"); + return c != 0; +} + +/** + * atomic_inc_and_test_unchecked - increment and test + * @v: pointer of type atomic_unchecked_t + * + * Atomically increments @v by 1 + * and returns true if the result is zero, or false for all + * other cases. + */ +static inline int atomic_inc_and_test_unchecked(atomic_unchecked_t *v) +{ + unsigned char c; + + asm volatile(LOCK_PREFIX "incl %0\n" + "sete %1\n" : "=m" (v->counter), "=qm" (c) : "m" (v->counter) : "memory"); return c != 0; @@ -157,7 +312,16 @@ static inline int atomic_add_negative(in { unsigned char c; - asm volatile(LOCK_PREFIX "addl %2,%0; sets %1" + asm volatile(LOCK_PREFIX "addl %2,%0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + LOCK_PREFIX "subl %2,%0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + + "sets %1\n" : "=m" (v->counter), "=qm" (c) : "ir" (i), "m" (v->counter) : "memory"); return c; @@ -173,7 +337,31 @@ static inline int atomic_add_negative(in static inline int atomic_add_return(int i, atomic_t *v) { int __i = i; - asm volatile(LOCK_PREFIX "xaddl %0, %1" + asm volatile(LOCK_PREFIX "xaddl %0, %1\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + "movl %0, %1\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + + : "+r" (i), "+m" (v->counter) + : : "memory"); + return i + __i; +} + +/** + * atomic_add_return_unchecked - add and return + * @i: integer value to add + * @v: pointer of type atomic_unchecked_t + * + * Atomically adds @i to @v and returns @i + @v + */ +static inline int atomic_add_return_unchecked(int i, atomic_unchecked_t *v) +{ + int __i = i; + asm volatile(LOCK_PREFIX "xaddl %0, %1\n" : "+r" (i), "+m" (v->counter) : : "memory"); return i + __i; @@ -185,6 +373,10 @@ static inline int atomic_sub_return(int } #define atomic_inc_return(v) (atomic_add_return(1, v)) +static inline int atomic_inc_return_unchecked(atomic_unchecked_t *v) +{ + return atomic_add_return_unchecked(1, v); +} #define atomic_dec_return(v) (atomic_sub_return(1, v)) /* The 64-bit atomic type */ @@ -204,6 +396,18 @@ static inline long atomic64_read(const a } /** + * atomic64_read_unchecked - read atomic64 variable + * @v: pointer of type atomic64_unchecked_t + * + * Atomically reads the value of @v. + * Doesn't imply a read memory barrier. + */ +static inline long atomic64_read_unchecked(const atomic64_unchecked_t *v) +{ + return v->counter; +} + +/** * atomic64_set - set atomic64 variable * @v: pointer to type atomic64_t * @i: required value @@ -216,6 +420,18 @@ static inline void atomic64_set(atomic64 } /** + * atomic64_set_unchecked - set atomic64 variable + * @v: pointer to type atomic64_unchecked_t + * @i: required value + * + * Atomically sets the value of @v to @i. + */ +static inline void atomic64_set_unchecked(atomic64_unchecked_t *v, long i) +{ + v->counter = i; +} + +/** * atomic64_add - add integer to atomic64 variable * @i: integer value to add * @v: pointer to type atomic64_t @@ -224,6 +440,28 @@ static inline void atomic64_set(atomic64 */ static inline void atomic64_add(long i, atomic64_t *v) { + asm volatile(LOCK_PREFIX "addq %1,%0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + LOCK_PREFIX "subq %1,%0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + + : "=m" (v->counter) + : "er" (i), "m" (v->counter)); +} + +/** + * atomic64_add_unchecked - add integer to atomic64 variable + * @i: integer value to add + * @v: pointer to type atomic64_unchecked_t + * + * Atomically adds @i to @v. + */ +static inline void atomic64_add_unchecked(long i, atomic64_unchecked_t *v) +{ asm volatile(LOCK_PREFIX "addq %1,%0" : "=m" (v->counter) : "er" (i), "m" (v->counter)); @@ -238,7 +476,15 @@ static inline void atomic64_add(long i, */ static inline void atomic64_sub(long i, atomic64_t *v) { - asm volatile(LOCK_PREFIX "subq %1,%0" + asm volatile(LOCK_PREFIX "subq %1,%0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + LOCK_PREFIX "addq %1,%0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + : "=m" (v->counter) : "er" (i), "m" (v->counter)); } @@ -256,7 +502,16 @@ static inline int atomic64_sub_and_test( { unsigned char c; - asm volatile(LOCK_PREFIX "subq %2,%0; sete %1" + asm volatile(LOCK_PREFIX "subq %2,%0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + LOCK_PREFIX "addq %2,%0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + + "sete %1\n" : "=m" (v->counter), "=qm" (c) : "er" (i), "m" (v->counter) : "memory"); return c; @@ -270,6 +525,27 @@ static inline int atomic64_sub_and_test( */ static inline void atomic64_inc(atomic64_t *v) { + asm volatile(LOCK_PREFIX "incq %0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + LOCK_PREFIX "decq %0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + + : "=m" (v->counter) + : "m" (v->counter)); +} + +/** + * atomic64_inc_unchecked - increment atomic64 variable + * @v: pointer to type atomic64_unchecked_t + * + * Atomically increments @v by 1. + */ +static inline void atomic64_inc_unchecked(atomic64_unchecked_t *v) +{ asm volatile(LOCK_PREFIX "incq %0" : "=m" (v->counter) : "m" (v->counter)); @@ -283,7 +559,28 @@ static inline void atomic64_inc(atomic64 */ static inline void atomic64_dec(atomic64_t *v) { - asm volatile(LOCK_PREFIX "decq %0" + asm volatile(LOCK_PREFIX "decq %0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + LOCK_PREFIX "incq %0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + + : "=m" (v->counter) + : "m" (v->counter)); +} + +/** + * atomic64_dec_unchecked - decrement atomic64 variable + * @v: pointer to type atomic64_t + * + * Atomically decrements @v by 1. + */ +static inline void atomic64_dec_unchecked(atomic64_unchecked_t *v) +{ + asm volatile(LOCK_PREFIX "decq %0\n" : "=m" (v->counter) : "m" (v->counter)); } @@ -300,7 +597,16 @@ static inline int atomic64_dec_and_test( { unsigned char c; - asm volatile(LOCK_PREFIX "decq %0; sete %1" + asm volatile(LOCK_PREFIX "decq %0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + LOCK_PREFIX "incq %0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + + "sete %1\n" : "=m" (v->counter), "=qm" (c) : "m" (v->counter) : "memory"); return c != 0; @@ -318,7 +624,16 @@ static inline int atomic64_inc_and_test( { unsigned char c; - asm volatile(LOCK_PREFIX "incq %0; sete %1" + asm volatile(LOCK_PREFIX "incq %0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + LOCK_PREFIX "decq %0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + + "sete %1\n" : "=m" (v->counter), "=qm" (c) : "m" (v->counter) : "memory"); return c != 0; @@ -337,7 +652,16 @@ static inline int atomic64_add_negative( { unsigned char c; - asm volatile(LOCK_PREFIX "addq %2,%0; sets %1" + asm volatile(LOCK_PREFIX "addq %2,%0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + LOCK_PREFIX "subq %2,%0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + + "sets %1\n" : "=m" (v->counter), "=qm" (c) : "er" (i), "m" (v->counter) : "memory"); return c; @@ -353,7 +677,31 @@ static inline int atomic64_add_negative( static inline long atomic64_add_return(long i, atomic64_t *v) { long __i = i; - asm volatile(LOCK_PREFIX "xaddq %0, %1;" + asm volatile(LOCK_PREFIX "xaddq %0, %1\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + "movq %0, %1\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + + : "+r" (i), "+m" (v->counter) + : : "memory"); + return i + __i; +} + +/** + * atomic64_add_return_unchecked - add and return + * @i: integer value to add + * @v: pointer to type atomic64_unchecked_t + * + * Atomically adds @i to @v and returns @i + @v + */ +static inline long atomic64_add_return_unchecked(long i, atomic64_unchecked_t *v) +{ + long __i = i; + asm volatile(LOCK_PREFIX "xaddq %0, %1" : "+r" (i), "+m" (v->counter) : : "memory"); return i + __i; @@ -365,6 +713,10 @@ static inline long atomic64_sub_return(l } #define atomic64_inc_return(v) (atomic64_add_return(1, (v))) +static inline long atomic64_inc_return_unchecked(atomic64_unchecked_t *v) +{ + return atomic64_add_return_unchecked(1, v); +} #define atomic64_dec_return(v) (atomic64_sub_return(1, (v))) static inline long atomic64_cmpxchg(atomic64_t *v, long old, long new) @@ -372,21 +724,41 @@ static inline long atomic64_cmpxchg(atom return cmpxchg(&v->counter, old, new); } +static inline long atomic64_cmpxchg_unchecked(atomic64_unchecked_t *v, long old, long new) +{ + return cmpxchg(&v->counter, old, new); +} + static inline long atomic64_xchg(atomic64_t *v, long new) { return xchg(&v->counter, new); } +static inline long atomic64_xchg_unchecked(atomic64_unchecked_t *v, long new) +{ + return xchg(&v->counter, new); +} + static inline long atomic_cmpxchg(atomic_t *v, int old, int new) { return cmpxchg(&v->counter, old, new); } +static inline long atomic_cmpxchg_unchecked(atomic_unchecked_t *v, int old, int new) +{ + return cmpxchg(&v->counter, old, new); +} + static inline long atomic_xchg(atomic_t *v, int new) { return xchg(&v->counter, new); } +static inline long atomic_xchg_unchecked(atomic_unchecked_t *v, int new) +{ + return xchg(&v->counter, new); +} + /** * atomic_add_unless - add unless the number is a given value * @v: pointer of type atomic_t @@ -398,17 +770,30 @@ static inline long atomic_xchg(atomic_t */ static inline int atomic_add_unless(atomic_t *v, int a, int u) { - int c, old; + int c, old, new; c = atomic_read(v); for (;;) { - if (unlikely(c == (u))) + if (unlikely(c == u)) break; - old = atomic_cmpxchg((v), c, c + (a)); + + asm volatile("addl %2,%0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + "subl %2,%0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + + : "=r" (new) + : "0" (c), "ir" (a)); + + old = atomic_cmpxchg(v, c, new); if (likely(old == c)) break; c = old; } - return c != (u); + return c != u; } #define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0) @@ -424,17 +809,30 @@ static inline int atomic_add_unless(atom */ static inline int atomic64_add_unless(atomic64_t *v, long a, long u) { - long c, old; + long c, old, new; c = atomic64_read(v); for (;;) { - if (unlikely(c == (u))) + if (unlikely(c == u)) break; - old = atomic64_cmpxchg((v), c, c + (a)); + + asm volatile("addq %2,%0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + "subq %2,%0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + + : "=r" (new) + : "0" (c), "er" (a)); + + old = atomic64_cmpxchg(v, c, new); if (likely(old == c)) break; c = old; } - return c != (u); + return c != u; } /** diff -urNp linux-2.6.32.48/arch/x86/include/asm/bitops.h linux-2.6.32.48-openvz/arch/x86/include/asm/bitops.h --- linux-2.6.32.48/arch/x86/include/asm/bitops.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/bitops.h 2011-11-17 20:08:53.000000000 -0500 @@ -38,7 +38,7 @@ * a mask operation on a byte. */ #define IS_IMMEDIATE(nr) (__builtin_constant_p(nr)) -#define CONST_MASK_ADDR(nr, addr) BITOP_ADDR((void *)(addr) + ((nr)>>3)) +#define CONST_MASK_ADDR(nr, addr) BITOP_ADDR((volatile void *)(addr) + ((nr)>>3)) #define CONST_MASK(nr) (1 << ((nr) & 7)) /** diff -urNp linux-2.6.32.48/arch/x86/include/asm/boot.h linux-2.6.32.48-openvz/arch/x86/include/asm/boot.h --- linux-2.6.32.48/arch/x86/include/asm/boot.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/boot.h 2011-11-17 20:08:53.000000000 -0500 @@ -11,10 +11,15 @@ #include /* Physical address where kernel should be loaded. */ -#define LOAD_PHYSICAL_ADDR ((CONFIG_PHYSICAL_START \ +#define ____LOAD_PHYSICAL_ADDR ((CONFIG_PHYSICAL_START \ + (CONFIG_PHYSICAL_ALIGN - 1)) \ & ~(CONFIG_PHYSICAL_ALIGN - 1)) +#ifndef __ASSEMBLY__ +extern unsigned char __LOAD_PHYSICAL_ADDR[]; +#define LOAD_PHYSICAL_ADDR ((unsigned long)__LOAD_PHYSICAL_ADDR) +#endif + /* Minimum kernel alignment, as a power of two */ #ifdef CONFIG_X86_64 #define MIN_KERNEL_ALIGN_LG2 PMD_SHIFT diff -urNp linux-2.6.32.48/arch/x86/include/asm/cacheflush.h linux-2.6.32.48-openvz/arch/x86/include/asm/cacheflush.h --- linux-2.6.32.48/arch/x86/include/asm/cacheflush.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/cacheflush.h 2011-11-17 20:08:53.000000000 -0500 @@ -60,7 +60,7 @@ PAGEFLAG(WC, WC) static inline unsigned long get_page_memtype(struct page *pg) { if (!PageUncached(pg) && !PageWC(pg)) - return -1; + return ~0UL; else if (!PageUncached(pg) && PageWC(pg)) return _PAGE_CACHE_WC; else if (PageUncached(pg) && !PageWC(pg)) @@ -85,7 +85,7 @@ static inline void set_page_memtype(stru SetPageWC(pg); break; default: - case -1: + case ~0UL: ClearPageUncached(pg); ClearPageWC(pg); break; diff -urNp linux-2.6.32.48/arch/x86/include/asm/cache.h linux-2.6.32.48-openvz/arch/x86/include/asm/cache.h --- linux-2.6.32.48/arch/x86/include/asm/cache.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/cache.h 2011-11-17 20:09:04.000000000 -0500 @@ -5,9 +5,10 @@ /* L1 cache line size */ #define L1_CACHE_SHIFT (CONFIG_X86_L1_CACHE_SHIFT) -#define L1_CACHE_BYTES (1 << L1_CACHE_SHIFT) +#define L1_CACHE_BYTES (_AC(1,UL) << L1_CACHE_SHIFT) #define __read_mostly __attribute__((__section__(".data.read_mostly"))) +#define __read_only __attribute__((__section__(".data.read_only"))) #ifdef CONFIG_X86_VSMP /* vSMP Internode cacheline shift */ diff -urNp linux-2.6.32.48/arch/x86/include/asm/calling.h linux-2.6.32.48-openvz/arch/x86/include/asm/calling.h --- linux-2.6.32.48/arch/x86/include/asm/calling.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/calling.h 2011-11-17 20:09:04.000000000 -0500 @@ -52,32 +52,32 @@ For 32-bit we have the following convent * for assembly code: */ -#define R15 0 -#define R14 8 -#define R13 16 -#define R12 24 -#define RBP 32 -#define RBX 40 +#define R15 (0) +#define R14 (8) +#define R13 (16) +#define R12 (24) +#define RBP (32) +#define RBX (40) /* arguments: interrupts/non tracing syscalls only save up to here: */ -#define R11 48 -#define R10 56 -#define R9 64 -#define R8 72 -#define RAX 80 -#define RCX 88 -#define RDX 96 -#define RSI 104 -#define RDI 112 -#define ORIG_RAX 120 /* + error_code */ +#define R11 (48) +#define R10 (56) +#define R9 (64) +#define R8 (72) +#define RAX (80) +#define RCX (88) +#define RDX (96) +#define RSI (104) +#define RDI (112) +#define ORIG_RAX (120) /* + error_code */ /* end of arguments */ /* cpu exception frame or undefined in case of fast syscall: */ -#define RIP 128 -#define CS 136 -#define EFLAGS 144 -#define RSP 152 -#define SS 160 +#define RIP (128) +#define CS (136) +#define EFLAGS (144) +#define RSP (152) +#define SS (160) #define ARGOFFSET R11 #define SWFRAME ORIG_RAX diff -urNp linux-2.6.32.48/arch/x86/include/asm/checksum_32.h linux-2.6.32.48-openvz/arch/x86/include/asm/checksum_32.h --- linux-2.6.32.48/arch/x86/include/asm/checksum_32.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/checksum_32.h 2011-11-17 20:08:53.000000000 -0500 @@ -31,6 +31,14 @@ asmlinkage __wsum csum_partial_copy_gene int len, __wsum sum, int *src_err_ptr, int *dst_err_ptr); +asmlinkage __wsum csum_partial_copy_generic_to_user(const void *src, void *dst, + int len, __wsum sum, + int *src_err_ptr, int *dst_err_ptr); + +asmlinkage __wsum csum_partial_copy_generic_from_user(const void *src, void *dst, + int len, __wsum sum, + int *src_err_ptr, int *dst_err_ptr); + /* * Note: when you get a NULL pointer exception here this means someone * passed in an incorrect kernel address to one of these functions. @@ -50,7 +58,7 @@ static inline __wsum csum_partial_copy_f int *err_ptr) { might_sleep(); - return csum_partial_copy_generic((__force void *)src, dst, + return csum_partial_copy_generic_from_user((__force void *)src, dst, len, sum, err_ptr, NULL); } @@ -178,7 +186,7 @@ static inline __wsum csum_and_copy_to_us { might_sleep(); if (access_ok(VERIFY_WRITE, dst, len)) - return csum_partial_copy_generic(src, (__force void *)dst, + return csum_partial_copy_generic_to_user(src, (__force void *)dst, len, sum, NULL, err_ptr); if (len) diff -urNp linux-2.6.32.48/arch/x86/include/asm/desc_defs.h linux-2.6.32.48-openvz/arch/x86/include/asm/desc_defs.h --- linux-2.6.32.48/arch/x86/include/asm/desc_defs.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/desc_defs.h 2011-11-17 20:08:53.000000000 -0500 @@ -31,6 +31,12 @@ struct desc_struct { unsigned base1: 8, type: 4, s: 1, dpl: 2, p: 1; unsigned limit: 4, avl: 1, l: 1, d: 1, g: 1, base2: 8; }; + struct { + u16 offset_low; + u16 seg; + unsigned reserved: 8, type: 4, s: 1, dpl: 2, p: 1; + unsigned offset_high: 16; + } gate; }; } __attribute__((packed)); diff -urNp linux-2.6.32.48/arch/x86/include/asm/desc.h linux-2.6.32.48-openvz/arch/x86/include/asm/desc.h --- linux-2.6.32.48/arch/x86/include/asm/desc.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/desc.h 2011-11-17 20:08:53.000000000 -0500 @@ -4,6 +4,7 @@ #include #include #include +#include #include static inline void fill_ldt(struct desc_struct *desc, @@ -15,6 +16,7 @@ static inline void fill_ldt(struct desc_ desc->base1 = (info->base_addr & 0x00ff0000) >> 16; desc->type = (info->read_exec_only ^ 1) << 1; desc->type |= info->contents << 2; + desc->type |= info->seg_not_present ^ 1; desc->s = 1; desc->dpl = 0x3; desc->p = info->seg_not_present ^ 1; @@ -31,16 +33,12 @@ static inline void fill_ldt(struct desc_ } extern struct desc_ptr idt_descr; -extern gate_desc idt_table[]; - -struct gdt_page { - struct desc_struct gdt[GDT_ENTRIES]; -} __attribute__((aligned(PAGE_SIZE))); -DECLARE_PER_CPU_PAGE_ALIGNED(struct gdt_page, gdt_page); +extern gate_desc idt_table[256]; +extern struct desc_struct cpu_gdt_table[NR_CPUS][PAGE_SIZE / sizeof(struct desc_struct)]; static inline struct desc_struct *get_cpu_gdt_table(unsigned int cpu) { - return per_cpu(gdt_page, cpu).gdt; + return cpu_gdt_table[cpu]; } #ifdef CONFIG_X86_64 @@ -65,9 +63,14 @@ static inline void pack_gate(gate_desc * unsigned long base, unsigned dpl, unsigned flags, unsigned short seg) { - gate->a = (seg << 16) | (base & 0xffff); - gate->b = (base & 0xffff0000) | - (((0x80 | type | (dpl << 5)) & 0xff) << 8); + gate->gate.offset_low = base; + gate->gate.seg = seg; + gate->gate.reserved = 0; + gate->gate.type = type; + gate->gate.s = 0; + gate->gate.dpl = dpl; + gate->gate.p = 1; + gate->gate.offset_high = base >> 16; } #endif @@ -115,13 +118,17 @@ static inline void paravirt_free_ldt(str static inline void native_write_idt_entry(gate_desc *idt, int entry, const gate_desc *gate) { + pax_open_kernel(); memcpy(&idt[entry], gate, sizeof(*gate)); + pax_close_kernel(); } static inline void native_write_ldt_entry(struct desc_struct *ldt, int entry, const void *desc) { + pax_open_kernel(); memcpy(&ldt[entry], desc, 8); + pax_close_kernel(); } static inline void native_write_gdt_entry(struct desc_struct *gdt, int entry, @@ -139,7 +146,10 @@ static inline void native_write_gdt_entr size = sizeof(struct desc_struct); break; } + + pax_open_kernel(); memcpy(&gdt[entry], desc, size); + pax_close_kernel(); } static inline void pack_descriptor(struct desc_struct *desc, unsigned long base, @@ -211,7 +221,9 @@ static inline void native_set_ldt(const static inline void native_load_tr_desc(void) { + pax_open_kernel(); asm volatile("ltr %w0"::"q" (GDT_ENTRY_TSS*8)); + pax_close_kernel(); } static inline void native_load_gdt(const struct desc_ptr *dtr) @@ -246,8 +258,10 @@ static inline void native_load_tls(struc unsigned int i; struct desc_struct *gdt = get_cpu_gdt_table(cpu); + pax_open_kernel(); for (i = 0; i < GDT_ENTRY_TLS_ENTRIES; i++) gdt[GDT_ENTRY_TLS_MIN + i] = t->tls_array[i]; + pax_close_kernel(); } #define _LDT_empty(info) \ @@ -309,7 +323,7 @@ static inline void set_desc_limit(struct desc->limit = (limit >> 16) & 0xf; } -static inline void _set_gate(int gate, unsigned type, void *addr, +static inline void _set_gate(int gate, unsigned type, const void *addr, unsigned dpl, unsigned ist, unsigned seg) { gate_desc s; @@ -327,7 +341,7 @@ static inline void _set_gate(int gate, u * Pentium F0 0F bugfix can have resulted in the mapped * IDT being write-protected. */ -static inline void set_intr_gate(unsigned int n, void *addr) +static inline void set_intr_gate(unsigned int n, const void *addr) { BUG_ON((unsigned)n > 0xFF); _set_gate(n, GATE_INTERRUPT, addr, 0, 0, __KERNEL_CS); @@ -356,19 +370,19 @@ static inline void alloc_intr_gate(unsig /* * This routine sets up an interrupt gate at directory privilege level 3. */ -static inline void set_system_intr_gate(unsigned int n, void *addr) +static inline void set_system_intr_gate(unsigned int n, const void *addr) { BUG_ON((unsigned)n > 0xFF); _set_gate(n, GATE_INTERRUPT, addr, 0x3, 0, __KERNEL_CS); } -static inline void set_system_trap_gate(unsigned int n, void *addr) +static inline void set_system_trap_gate(unsigned int n, const void *addr) { BUG_ON((unsigned)n > 0xFF); _set_gate(n, GATE_TRAP, addr, 0x3, 0, __KERNEL_CS); } -static inline void set_trap_gate(unsigned int n, void *addr) +static inline void set_trap_gate(unsigned int n, const void *addr) { BUG_ON((unsigned)n > 0xFF); _set_gate(n, GATE_TRAP, addr, 0, 0, __KERNEL_CS); @@ -377,19 +391,31 @@ static inline void set_trap_gate(unsigne static inline void set_task_gate(unsigned int n, unsigned int gdt_entry) { BUG_ON((unsigned)n > 0xFF); - _set_gate(n, GATE_TASK, (void *)0, 0, 0, (gdt_entry<<3)); + _set_gate(n, GATE_TASK, (const void *)0, 0, 0, (gdt_entry<<3)); } -static inline void set_intr_gate_ist(int n, void *addr, unsigned ist) +static inline void set_intr_gate_ist(int n, const void *addr, unsigned ist) { BUG_ON((unsigned)n > 0xFF); _set_gate(n, GATE_INTERRUPT, addr, 0, ist, __KERNEL_CS); } -static inline void set_system_intr_gate_ist(int n, void *addr, unsigned ist) +static inline void set_system_intr_gate_ist(int n, const void *addr, unsigned ist) { BUG_ON((unsigned)n > 0xFF); _set_gate(n, GATE_INTERRUPT, addr, 0x3, ist, __KERNEL_CS); } +#ifdef CONFIG_X86_32 +static inline void set_user_cs(unsigned long base, unsigned long limit, int cpu) +{ + struct desc_struct d; + + if (likely(limit)) + limit = (limit - 1UL) >> PAGE_SHIFT; + pack_descriptor(&d, base, limit, 0xFB, 0xC); + write_gdt_entry(get_cpu_gdt_table(cpu), GDT_ENTRY_DEFAULT_USER_CS, &d, DESCTYPE_S); +} +#endif + #endif /* _ASM_X86_DESC_H */ diff -urNp linux-2.6.32.48/arch/x86/include/asm/device.h linux-2.6.32.48-openvz/arch/x86/include/asm/device.h --- linux-2.6.32.48/arch/x86/include/asm/device.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/device.h 2011-11-17 20:08:53.000000000 -0500 @@ -6,7 +6,7 @@ struct dev_archdata { void *acpi_handle; #endif #ifdef CONFIG_X86_64 -struct dma_map_ops *dma_ops; + const struct dma_map_ops *dma_ops; #endif #ifdef CONFIG_DMAR void *iommu; /* hook for IOMMU specific extension */ diff -urNp linux-2.6.32.48/arch/x86/include/asm/dma-mapping.h linux-2.6.32.48-openvz/arch/x86/include/asm/dma-mapping.h --- linux-2.6.32.48/arch/x86/include/asm/dma-mapping.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/dma-mapping.h 2011-11-17 20:08:53.000000000 -0500 @@ -25,9 +25,9 @@ extern int iommu_merge; extern struct device x86_dma_fallback_dev; extern int panic_on_overflow; -extern struct dma_map_ops *dma_ops; +extern const struct dma_map_ops *dma_ops; -static inline struct dma_map_ops *get_dma_ops(struct device *dev) +static inline const struct dma_map_ops *get_dma_ops(struct device *dev) { #ifdef CONFIG_X86_32 return dma_ops; @@ -44,7 +44,7 @@ static inline struct dma_map_ops *get_dm /* Make sure we keep the same behaviour */ static inline int dma_mapping_error(struct device *dev, dma_addr_t dma_addr) { - struct dma_map_ops *ops = get_dma_ops(dev); + const struct dma_map_ops *ops = get_dma_ops(dev); if (ops->mapping_error) return ops->mapping_error(dev, dma_addr); @@ -122,7 +122,7 @@ static inline void * dma_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_handle, gfp_t gfp) { - struct dma_map_ops *ops = get_dma_ops(dev); + const struct dma_map_ops *ops = get_dma_ops(dev); void *memory; gfp &= ~(__GFP_DMA | __GFP_HIGHMEM | __GFP_DMA32); @@ -149,7 +149,7 @@ dma_alloc_coherent(struct device *dev, s static inline void dma_free_coherent(struct device *dev, size_t size, void *vaddr, dma_addr_t bus) { - struct dma_map_ops *ops = get_dma_ops(dev); + const struct dma_map_ops *ops = get_dma_ops(dev); WARN_ON(irqs_disabled()); /* for portability */ diff -urNp linux-2.6.32.48/arch/x86/include/asm/e820.h linux-2.6.32.48-openvz/arch/x86/include/asm/e820.h --- linux-2.6.32.48/arch/x86/include/asm/e820.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/e820.h 2011-11-17 20:08:53.000000000 -0500 @@ -133,7 +133,7 @@ extern char *default_machine_specific_me #define ISA_END_ADDRESS 0x100000 #define is_ISA_range(s, e) ((s) >= ISA_START_ADDRESS && (e) < ISA_END_ADDRESS) -#define BIOS_BEGIN 0x000a0000 +#define BIOS_BEGIN 0x000c0000 #define BIOS_END 0x00100000 #ifdef __KERNEL__ diff -urNp linux-2.6.32.48/arch/x86/include/asm/elf.h linux-2.6.32.48-openvz/arch/x86/include/asm/elf.h --- linux-2.6.32.48/arch/x86/include/asm/elf.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/elf.h 2011-11-17 20:09:04.000000000 -0500 @@ -257,7 +257,25 @@ extern int force_personality32; the loader. We need to make sure that it is out of the way of the program that it will "exec", and that there is sufficient room for the brk. */ +#ifdef CONFIG_PAX_SEGMEXEC +#define ELF_ET_DYN_BASE ((current->mm->pax_flags & MF_PAX_SEGMEXEC) ? SEGMEXEC_TASK_SIZE/3*2 : TASK_SIZE/3*2) +#else #define ELF_ET_DYN_BASE (TASK_SIZE / 3 * 2) +#endif + +#ifdef CONFIG_PAX_ASLR +#ifdef CONFIG_X86_32 +#define PAX_ELF_ET_DYN_BASE 0x10000000UL + +#define PAX_DELTA_MMAP_LEN (current->mm->pax_flags & MF_PAX_SEGMEXEC ? 15 : 16) +#define PAX_DELTA_STACK_LEN (current->mm->pax_flags & MF_PAX_SEGMEXEC ? 15 : 16) +#else +#define PAX_ELF_ET_DYN_BASE 0x400000UL + +#define PAX_DELTA_MMAP_LEN ((test_thread_flag(TIF_IA32)) ? 16 : TASK_SIZE_MAX_SHIFT - PAGE_SHIFT - 3) +#define PAX_DELTA_STACK_LEN ((test_thread_flag(TIF_IA32)) ? 16 : TASK_SIZE_MAX_SHIFT - PAGE_SHIFT - 3) +#endif +#endif /* This yields a mask that user programs can use to figure out what instruction set this CPU supports. This could be done in user space, @@ -285,7 +303,7 @@ struct task_struct; #define ARCH_DLINFO_IA32(vdso_enabled) \ do { \ - if (vdso_enabled) { \ + if (vdso_enabled && sysctl_at_vsyscall) { \ NEW_AUX_ENT(AT_SYSINFO, VDSO_ENTRY); \ NEW_AUX_ENT(AT_SYSINFO_EHDR, VDSO_CURRENT_BASE); \ } \ @@ -310,9 +328,7 @@ do { \ #define ARCH_DLINFO \ do { \ - if (vdso_enabled) \ - NEW_AUX_ENT(AT_SYSINFO_EHDR, \ - (unsigned long)current->mm->context.vdso); \ + NEW_AUX_ENT(AT_SYSINFO_EHDR, current->mm->context.vdso); \ } while (0) #define AT_SYSINFO 32 @@ -323,7 +339,7 @@ do { \ #endif /* !CONFIG_X86_32 */ -#define VDSO_CURRENT_BASE ((unsigned long)current->mm->context.vdso) +#define VDSO_CURRENT_BASE (current->mm->context.vdso) #define VDSO_ENTRY \ ((unsigned long)VDSO32_SYMBOL(VDSO_CURRENT_BASE, vsyscall)) @@ -332,12 +348,11 @@ struct linux_binprm; #define ARCH_HAS_SETUP_ADDITIONAL_PAGES 1 extern int arch_setup_additional_pages(struct linux_binprm *bprm, - int uses_interp); + int uses_interp, + unsigned long map_address); -extern int syscall32_setup_pages(struct linux_binprm *, int exstack); +extern int syscall32_setup_pages(struct linux_binprm *, int exstack, + unsigned long map_address); #define compat_arch_setup_additional_pages syscall32_setup_pages -extern unsigned long arch_randomize_brk(struct mm_struct *mm); -#define arch_randomize_brk arch_randomize_brk - #endif /* _ASM_X86_ELF_H */ diff -urNp linux-2.6.32.48/arch/x86/include/asm/futex.h linux-2.6.32.48-openvz/arch/x86/include/asm/futex.h --- linux-2.6.32.48/arch/x86/include/asm/futex.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/futex.h 2011-11-17 20:09:04.000000000 -0500 @@ -12,16 +12,18 @@ #include #define __futex_atomic_op1(insn, ret, oldval, uaddr, oparg) \ + typecheck(u32 __user *, uaddr); \ asm volatile("1:\t" insn "\n" \ "2:\t.section .fixup,\"ax\"\n" \ "3:\tmov\t%3, %1\n" \ "\tjmp\t2b\n" \ "\t.previous\n" \ _ASM_EXTABLE(1b, 3b) \ - : "=r" (oldval), "=r" (ret), "+m" (*uaddr) \ + : "=r" (oldval), "=r" (ret), "+m" (*(u32 __user *)____m(uaddr))\ : "i" (-EFAULT), "0" (oparg), "1" (0)) #define __futex_atomic_op2(insn, ret, oldval, uaddr, oparg) \ + typecheck(u32 __user *, uaddr); \ asm volatile("1:\tmovl %2, %0\n" \ "\tmovl\t%0, %3\n" \ "\t" insn "\n" \ @@ -34,10 +36,10 @@ _ASM_EXTABLE(1b, 4b) \ _ASM_EXTABLE(2b, 4b) \ : "=&a" (oldval), "=&r" (ret), \ - "+m" (*uaddr), "=&r" (tem) \ + "+m" (*(u32 __user *)____m(uaddr)), "=&r" (tem) \ : "r" (oparg), "i" (-EFAULT), "1" (0)) -static inline int futex_atomic_op_inuser(int encoded_op, int __user *uaddr) +static inline int futex_atomic_op_inuser(int encoded_op, u32 __user *uaddr) { int op = (encoded_op >> 28) & 7; int cmp = (encoded_op >> 24) & 15; @@ -61,10 +63,10 @@ static inline int futex_atomic_op_inuser switch (op) { case FUTEX_OP_SET: - __futex_atomic_op1("xchgl %0, %2", ret, oldval, uaddr, oparg); + __futex_atomic_op1(__copyuser_seg"xchgl %0, %2", ret, oldval, uaddr, oparg); break; case FUTEX_OP_ADD: - __futex_atomic_op1(LOCK_PREFIX "xaddl %0, %2", ret, oldval, + __futex_atomic_op1(LOCK_PREFIX __copyuser_seg"xaddl %0, %2", ret, oldval, uaddr, oparg); break; case FUTEX_OP_OR: @@ -109,7 +111,7 @@ static inline int futex_atomic_op_inuser return ret; } -static inline int futex_atomic_cmpxchg_inatomic(int __user *uaddr, int oldval, +static inline int futex_atomic_cmpxchg_inatomic(u32 __user *uaddr, int oldval, int newval) { @@ -119,16 +121,16 @@ static inline int futex_atomic_cmpxchg_i return -ENOSYS; #endif - if (!access_ok(VERIFY_WRITE, uaddr, sizeof(int))) + if (!access_ok(VERIFY_WRITE, uaddr, sizeof(u32))) return -EFAULT; - asm volatile("1:\t" LOCK_PREFIX "cmpxchgl %3, %1\n" + asm volatile("1:\t" LOCK_PREFIX __copyuser_seg"cmpxchgl %3, %1\n" "2:\t.section .fixup, \"ax\"\n" "3:\tmov %2, %0\n" "\tjmp 2b\n" "\t.previous\n" _ASM_EXTABLE(1b, 3b) - : "=a" (oldval), "+m" (*uaddr) + : "=a" (oldval), "+m" (*(u32 *)____m(uaddr)) : "i" (-EFAULT), "r" (newval), "0" (oldval) : "memory" ); diff -urNp linux-2.6.32.48/arch/x86/include/asm/hw_irq.h linux-2.6.32.48-openvz/arch/x86/include/asm/hw_irq.h --- linux-2.6.32.48/arch/x86/include/asm/hw_irq.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/hw_irq.h 2011-11-17 20:08:53.000000000 -0500 @@ -92,8 +92,8 @@ extern void setup_ioapic_dest(void); extern void enable_IO_APIC(void); /* Statistics */ -extern atomic_t irq_err_count; -extern atomic_t irq_mis_count; +extern atomic_unchecked_t irq_err_count; +extern atomic_unchecked_t irq_mis_count; /* EISA */ extern void eisa_set_level_irq(unsigned int irq); diff -urNp linux-2.6.32.48/arch/x86/include/asm/i387.h linux-2.6.32.48-openvz/arch/x86/include/asm/i387.h --- linux-2.6.32.48/arch/x86/include/asm/i387.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/i387.h 2011-11-17 20:08:53.000000000 -0500 @@ -60,6 +60,11 @@ static inline int fxrstor_checking(struc { int err; +#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) + if ((unsigned long)fx < PAX_USER_SHADOW_BASE) + fx = (struct i387_fxsave_struct *)((void *)fx + PAX_USER_SHADOW_BASE); +#endif + asm volatile("1: rex64/fxrstor (%[fx])\n\t" "2:\n" ".section .fixup,\"ax\"\n" @@ -105,6 +110,11 @@ static inline int fxsave_user(struct i38 { int err; +#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) + if ((unsigned long)fx < PAX_USER_SHADOW_BASE) + fx = (struct i387_fxsave_struct __user *)((void __user *)fx + PAX_USER_SHADOW_BASE); +#endif + asm volatile("1: rex64/fxsave (%[fx])\n\t" "2:\n" ".section .fixup,\"ax\"\n" @@ -195,13 +205,8 @@ static inline int fxrstor_checking(struc } /* We need a safe address that is cheap to find and that is already - in L1 during context switch. The best choices are unfortunately - different for UP and SMP */ -#ifdef CONFIG_SMP -#define safe_address (__per_cpu_offset[0]) -#else -#define safe_address (kstat_cpu(0).cpustat.user) -#endif + in L1 during context switch. */ +#define safe_address (init_tss[smp_processor_id()].x86_tss.sp0) /* * These must be called with preempt disabled @@ -291,7 +296,7 @@ static inline void kernel_fpu_begin(void struct thread_info *me = current_thread_info(); preempt_disable(); if (me->status & TS_USEDFPU) - __save_init_fpu(me->task); + __save_init_fpu(current); else clts(); } diff -urNp linux-2.6.32.48/arch/x86/include/asm/io_32.h linux-2.6.32.48-openvz/arch/x86/include/asm/io_32.h --- linux-2.6.32.48/arch/x86/include/asm/io_32.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/io_32.h 2011-11-17 20:08:53.000000000 -0500 @@ -3,6 +3,7 @@ #include #include +#include /* * This file contains the definitions for the x86 IO instructions @@ -42,6 +43,17 @@ #ifdef __KERNEL__ +#define ARCH_HAS_VALID_PHYS_ADDR_RANGE +static inline int valid_phys_addr_range(unsigned long addr, size_t count) +{ + return ((addr + count + PAGE_SIZE - 1) >> PAGE_SHIFT) < (1ULL << (boot_cpu_data.x86_phys_bits - PAGE_SHIFT)) ? 1 : 0; +} + +static inline int valid_mmap_phys_addr_range(unsigned long pfn, size_t count) +{ + return (pfn + (count >> PAGE_SHIFT)) < (1ULL << (boot_cpu_data.x86_phys_bits - PAGE_SHIFT)) ? 1 : 0; +} + #include #include diff -urNp linux-2.6.32.48/arch/x86/include/asm/io_64.h linux-2.6.32.48-openvz/arch/x86/include/asm/io_64.h --- linux-2.6.32.48/arch/x86/include/asm/io_64.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/io_64.h 2011-11-17 20:08:53.000000000 -0500 @@ -140,6 +140,17 @@ __OUTS(l) #include +#define ARCH_HAS_VALID_PHYS_ADDR_RANGE +static inline int valid_phys_addr_range(unsigned long addr, size_t count) +{ + return ((addr + count + PAGE_SIZE - 1) >> PAGE_SHIFT) < (1ULL << (boot_cpu_data.x86_phys_bits - PAGE_SHIFT)) ? 1 : 0; +} + +static inline int valid_mmap_phys_addr_range(unsigned long pfn, size_t count) +{ + return (pfn + (count >> PAGE_SHIFT)) < (1ULL << (boot_cpu_data.x86_phys_bits - PAGE_SHIFT)) ? 1 : 0; +} + #include void __memcpy_fromio(void *, unsigned long, unsigned); diff -urNp linux-2.6.32.48/arch/x86/include/asm/iommu.h linux-2.6.32.48-openvz/arch/x86/include/asm/iommu.h --- linux-2.6.32.48/arch/x86/include/asm/iommu.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/iommu.h 2011-11-17 20:08:53.000000000 -0500 @@ -3,7 +3,7 @@ extern void pci_iommu_shutdown(void); extern void no_iommu_init(void); -extern struct dma_map_ops nommu_dma_ops; +extern const struct dma_map_ops nommu_dma_ops; extern int force_iommu, no_iommu; extern int iommu_detected; extern int iommu_pass_through; diff -urNp linux-2.6.32.48/arch/x86/include/asm/irqflags.h linux-2.6.32.48-openvz/arch/x86/include/asm/irqflags.h --- linux-2.6.32.48/arch/x86/include/asm/irqflags.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/irqflags.h 2011-11-17 20:08:53.000000000 -0500 @@ -142,6 +142,11 @@ static inline unsigned long __raw_local_ sti; \ sysexit +#define GET_CR0_INTO_RDI mov %cr0, %rdi +#define SET_RDI_INTO_CR0 mov %rdi, %cr0 +#define GET_CR3_INTO_RDI mov %cr3, %rdi +#define SET_RDI_INTO_CR3 mov %rdi, %cr3 + #else #define INTERRUPT_RETURN iret #define ENABLE_INTERRUPTS_SYSEXIT sti; sysexit diff -urNp linux-2.6.32.48/arch/x86/include/asm/kprobes.h linux-2.6.32.48-openvz/arch/x86/include/asm/kprobes.h --- linux-2.6.32.48/arch/x86/include/asm/kprobes.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/kprobes.h 2011-11-17 20:08:53.000000000 -0500 @@ -34,13 +34,8 @@ typedef u8 kprobe_opcode_t; #define BREAKPOINT_INSTRUCTION 0xcc #define RELATIVEJUMP_INSTRUCTION 0xe9 #define MAX_INSN_SIZE 16 -#define MAX_STACK_SIZE 64 -#define MIN_STACK_SIZE(ADDR) \ - (((MAX_STACK_SIZE) < (((unsigned long)current_thread_info()) + \ - THREAD_SIZE - (unsigned long)(ADDR))) \ - ? (MAX_STACK_SIZE) \ - : (((unsigned long)current_thread_info()) + \ - THREAD_SIZE - (unsigned long)(ADDR))) +#define MAX_STACK_SIZE 64UL +#define MIN_STACK_SIZE(ADDR) min(MAX_STACK_SIZE, current->thread.sp0 - (unsigned long)(ADDR)) #define flush_insn_slot(p) do { } while (0) diff -urNp linux-2.6.32.48/arch/x86/include/asm/kvm_host.h linux-2.6.32.48-openvz/arch/x86/include/asm/kvm_host.h --- linux-2.6.32.48/arch/x86/include/asm/kvm_host.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/kvm_host.h 2011-11-17 20:09:04.000000000 -0500 @@ -534,9 +534,9 @@ struct kvm_x86_ops { bool (*gb_page_enable)(void); const struct trace_print_flags *exit_reasons_str; -}; +} __do_const; -extern struct kvm_x86_ops *kvm_x86_ops; +extern const struct kvm_x86_ops *kvm_x86_ops; int kvm_mmu_module_init(void); void kvm_mmu_module_exit(void); diff -urNp linux-2.6.32.48/arch/x86/include/asm/local.h linux-2.6.32.48-openvz/arch/x86/include/asm/local.h --- linux-2.6.32.48/arch/x86/include/asm/local.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/local.h 2011-11-17 20:08:53.000000000 -0500 @@ -18,26 +18,58 @@ typedef struct { static inline void local_inc(local_t *l) { - asm volatile(_ASM_INC "%0" + asm volatile(_ASM_INC "%0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + _ASM_DEC "%0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + : "+m" (l->a.counter)); } static inline void local_dec(local_t *l) { - asm volatile(_ASM_DEC "%0" + asm volatile(_ASM_DEC "%0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + _ASM_INC "%0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + : "+m" (l->a.counter)); } static inline void local_add(long i, local_t *l) { - asm volatile(_ASM_ADD "%1,%0" + asm volatile(_ASM_ADD "%1,%0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + _ASM_SUB "%1,%0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + : "+m" (l->a.counter) : "ir" (i)); } static inline void local_sub(long i, local_t *l) { - asm volatile(_ASM_SUB "%1,%0" + asm volatile(_ASM_SUB "%1,%0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + _ASM_ADD "%1,%0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + : "+m" (l->a.counter) : "ir" (i)); } @@ -55,7 +87,16 @@ static inline int local_sub_and_test(lon { unsigned char c; - asm volatile(_ASM_SUB "%2,%0; sete %1" + asm volatile(_ASM_SUB "%2,%0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + _ASM_ADD "%2,%0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + + "sete %1\n" : "+m" (l->a.counter), "=qm" (c) : "ir" (i) : "memory"); return c; @@ -73,7 +114,16 @@ static inline int local_dec_and_test(loc { unsigned char c; - asm volatile(_ASM_DEC "%0; sete %1" + asm volatile(_ASM_DEC "%0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + _ASM_INC "%0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + + "sete %1\n" : "+m" (l->a.counter), "=qm" (c) : : "memory"); return c != 0; @@ -91,7 +141,16 @@ static inline int local_inc_and_test(loc { unsigned char c; - asm volatile(_ASM_INC "%0; sete %1" + asm volatile(_ASM_INC "%0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + _ASM_DEC "%0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + + "sete %1\n" : "+m" (l->a.counter), "=qm" (c) : : "memory"); return c != 0; @@ -110,7 +169,16 @@ static inline int local_add_negative(lon { unsigned char c; - asm volatile(_ASM_ADD "%2,%0; sets %1" + asm volatile(_ASM_ADD "%2,%0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + _ASM_SUB "%2,%0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + + "sets %1\n" : "+m" (l->a.counter), "=qm" (c) : "ir" (i) : "memory"); return c; @@ -133,7 +201,15 @@ static inline long local_add_return(long #endif /* Modern 486+ processor */ __i = i; - asm volatile(_ASM_XADD "%0, %1;" + asm volatile(_ASM_XADD "%0, %1\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + _ASM_MOV "%0,%1\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + : "+r" (i), "+m" (l->a.counter) : : "memory"); return i + __i; diff -urNp linux-2.6.32.48/arch/x86/include/asm/microcode.h linux-2.6.32.48-openvz/arch/x86/include/asm/microcode.h --- linux-2.6.32.48/arch/x86/include/asm/microcode.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/microcode.h 2011-11-17 20:08:53.000000000 -0500 @@ -12,13 +12,13 @@ struct device; enum ucode_state { UCODE_ERROR, UCODE_OK, UCODE_NFOUND }; struct microcode_ops { - enum ucode_state (*request_microcode_user) (int cpu, + enum ucode_state (* const request_microcode_user) (int cpu, const void __user *buf, size_t size); - enum ucode_state (*request_microcode_fw) (int cpu, + enum ucode_state (* const request_microcode_fw) (int cpu, struct device *device); - void (*microcode_fini_cpu) (int cpu); + void (* const microcode_fini_cpu) (int cpu); /* * The generic 'microcode_core' part guarantees that @@ -38,18 +38,18 @@ struct ucode_cpu_info { extern struct ucode_cpu_info ucode_cpu_info[]; #ifdef CONFIG_MICROCODE_INTEL -extern struct microcode_ops * __init init_intel_microcode(void); +extern const struct microcode_ops * __init init_intel_microcode(void); #else -static inline struct microcode_ops * __init init_intel_microcode(void) +static inline const struct microcode_ops * __init init_intel_microcode(void) { return NULL; } #endif /* CONFIG_MICROCODE_INTEL */ #ifdef CONFIG_MICROCODE_AMD -extern struct microcode_ops * __init init_amd_microcode(void); +extern const struct microcode_ops * __init init_amd_microcode(void); #else -static inline struct microcode_ops * __init init_amd_microcode(void) +static inline const struct microcode_ops * __init init_amd_microcode(void) { return NULL; } diff -urNp linux-2.6.32.48/arch/x86/include/asm/mman.h linux-2.6.32.48-openvz/arch/x86/include/asm/mman.h --- linux-2.6.32.48/arch/x86/include/asm/mman.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/mman.h 2011-11-17 20:08:53.000000000 -0500 @@ -5,4 +5,14 @@ #include +#ifdef __KERNEL__ +#ifndef __ASSEMBLY__ +#ifdef CONFIG_X86_32 +#define arch_mmap_check i386_mmap_check +int i386_mmap_check(unsigned long addr, unsigned long len, + unsigned long flags); +#endif +#endif +#endif + #endif /* _ASM_X86_MMAN_H */ diff -urNp linux-2.6.32.48/arch/x86/include/asm/mmu_context.h linux-2.6.32.48-openvz/arch/x86/include/asm/mmu_context.h --- linux-2.6.32.48/arch/x86/include/asm/mmu_context.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/mmu_context.h 2011-11-17 20:09:04.000000000 -0500 @@ -24,6 +24,18 @@ void destroy_context(struct mm_struct *m static inline void enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk) { + +#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) + unsigned int i; + pgd_t *pgd; + + pax_open_kernel(); + pgd = get_cpu_pgd(smp_processor_id()); + for (i = USER_PGD_PTRS; i < 2 * USER_PGD_PTRS; ++i) + set_pgd_batched(pgd+i, native_make_pgd(0)); + pax_close_kernel(); +#endif + #ifdef CONFIG_SMP if (percpu_read(cpu_tlbstate.state) == TLBSTATE_OK) percpu_write(cpu_tlbstate.state, TLBSTATE_LAZY); @@ -34,16 +46,30 @@ static inline void switch_mm(struct mm_s struct task_struct *tsk) { unsigned cpu = smp_processor_id(); +#if defined(CONFIG_X86_32) && (defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC)) && defined(CONFIG_SMP) + int tlbstate = TLBSTATE_OK; +#endif if (likely(prev != next)) { #ifdef CONFIG_SMP +#if defined(CONFIG_X86_32) && (defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC)) + tlbstate = percpu_read(cpu_tlbstate.state); +#endif percpu_write(cpu_tlbstate.state, TLBSTATE_OK); percpu_write(cpu_tlbstate.active_mm, next); #endif cpumask_set_cpu(cpu, mm_cpumask(next)); /* Re-load page tables */ +#ifdef CONFIG_PAX_PER_CPU_PGD + pax_open_kernel(); + __clone_user_pgds(get_cpu_pgd(cpu), next->pgd, USER_PGD_PTRS); + __shadow_user_pgds(get_cpu_pgd(cpu) + USER_PGD_PTRS, next->pgd, USER_PGD_PTRS); + pax_close_kernel(); + load_cr3(get_cpu_pgd(cpu)); +#else load_cr3(next->pgd); +#endif /* stop flush ipis for the previous mm */ cpumask_clear_cpu(cpu, mm_cpumask(prev)); @@ -53,9 +79,38 @@ static inline void switch_mm(struct mm_s */ if (unlikely(prev->context.ldt != next->context.ldt)) load_LDT_nolock(&next->context); - } + +#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_PAGEEXEC) && defined(CONFIG_SMP) + if (!nx_enabled) { + smp_mb__before_clear_bit(); + cpu_clear(cpu, prev->context.cpu_user_cs_mask); + smp_mb__after_clear_bit(); + cpu_set(cpu, next->context.cpu_user_cs_mask); + } +#endif + +#if defined(CONFIG_X86_32) && (defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC)) + if (unlikely(prev->context.user_cs_base != next->context.user_cs_base || + prev->context.user_cs_limit != next->context.user_cs_limit)) + set_user_cs(next->context.user_cs_base, next->context.user_cs_limit, cpu); #ifdef CONFIG_SMP + else if (unlikely(tlbstate != TLBSTATE_OK)) + set_user_cs(next->context.user_cs_base, next->context.user_cs_limit, cpu); +#endif +#endif + + } else { + +#ifdef CONFIG_PAX_PER_CPU_PGD + pax_open_kernel(); + __clone_user_pgds(get_cpu_pgd(cpu), next->pgd, USER_PGD_PTRS); + __shadow_user_pgds(get_cpu_pgd(cpu) + USER_PGD_PTRS, next->pgd, USER_PGD_PTRS); + pax_close_kernel(); + load_cr3(get_cpu_pgd(cpu)); +#endif + +#ifdef CONFIG_SMP percpu_write(cpu_tlbstate.state, TLBSTATE_OK); BUG_ON(percpu_read(cpu_tlbstate.active_mm) != next); @@ -64,11 +119,28 @@ static inline void switch_mm(struct mm_s * tlb flush IPI delivery. We must reload CR3 * to make sure to use no freed page tables. */ + +#ifndef CONFIG_PAX_PER_CPU_PGD load_cr3(next->pgd); +#endif + load_LDT_nolock(&next->context); + +#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_PAGEEXEC) + if (!nx_enabled) + cpu_set(cpu, next->context.cpu_user_cs_mask); +#endif + +#if defined(CONFIG_X86_32) && (defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC)) +#ifdef CONFIG_PAX_PAGEEXEC + if (!((next->pax_flags & MF_PAX_PAGEEXEC) && nx_enabled)) +#endif + set_user_cs(next->context.user_cs_base, next->context.user_cs_limit, cpu); +#endif + } - } #endif + } } #define activate_mm(prev, next) \ diff -urNp linux-2.6.32.48/arch/x86/include/asm/mmu.h linux-2.6.32.48-openvz/arch/x86/include/asm/mmu.h --- linux-2.6.32.48/arch/x86/include/asm/mmu.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/mmu.h 2011-11-17 20:08:53.000000000 -0500 @@ -9,10 +9,23 @@ * we put the segment information here. */ typedef struct { - void *ldt; + struct desc_struct *ldt; int size; struct mutex lock; - void *vdso; + unsigned long vdso; + +#ifdef CONFIG_X86_32 +#if defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC) + unsigned long user_cs_base; + unsigned long user_cs_limit; + +#if defined(CONFIG_PAX_PAGEEXEC) && defined(CONFIG_SMP) + cpumask_t cpu_user_cs_mask; +#endif + +#endif +#endif + } mm_context_t; #ifdef CONFIG_SMP diff -urNp linux-2.6.32.48/arch/x86/include/asm/module.h linux-2.6.32.48-openvz/arch/x86/include/asm/module.h --- linux-2.6.32.48/arch/x86/include/asm/module.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/module.h 2011-11-17 20:09:04.000000000 -0500 @@ -5,6 +5,7 @@ #ifdef CONFIG_X86_64 /* X86_64 does not define MODULE_PROC_FAMILY */ +#define MODULE_PROC_FAMILY "" #elif defined CONFIG_M386 #define MODULE_PROC_FAMILY "386 " #elif defined CONFIG_M486 @@ -59,13 +60,24 @@ #error unknown processor family #endif -#ifdef CONFIG_X86_32 -# ifdef CONFIG_4KSTACKS -# define MODULE_STACKSIZE "4KSTACKS " -# else -# define MODULE_STACKSIZE "" -# endif -# define MODULE_ARCH_VERMAGIC MODULE_PROC_FAMILY MODULE_STACKSIZE +#if defined(CONFIG_X86_32) && defined(CONFIG_4KSTACKS) +#define MODULE_STACKSIZE "4KSTACKS " +#else +#define MODULE_STACKSIZE "" +#endif + +#ifdef CONFIG_PAX_KERNEXEC +#define MODULE_PAX_KERNEXEC "KERNEXEC " +#else +#define MODULE_PAX_KERNEXEC "" #endif +#ifdef CONFIG_PAX_MEMORY_UDEREF +#define MODULE_PAX_UDEREF "UDEREF " +#else +#define MODULE_PAX_UDEREF "" +#endif + +#define MODULE_ARCH_VERMAGIC MODULE_PROC_FAMILY MODULE_STACKSIZE MODULE_PAX_KERNEXEC MODULE_PAX_UDEREF + #endif /* _ASM_X86_MODULE_H */ diff -urNp linux-2.6.32.48/arch/x86/include/asm/page_64_types.h linux-2.6.32.48-openvz/arch/x86/include/asm/page_64_types.h --- linux-2.6.32.48/arch/x86/include/asm/page_64_types.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/page_64_types.h 2011-11-17 20:08:53.000000000 -0500 @@ -56,7 +56,7 @@ void copy_page(void *to, void *from); /* duplicated to the one in bootmem.h */ extern unsigned long max_pfn; -extern unsigned long phys_base; +extern const unsigned long phys_base; extern unsigned long __phys_addr(unsigned long); #define __phys_reloc_hide(x) (x) diff -urNp linux-2.6.32.48/arch/x86/include/asm/paravirt.h linux-2.6.32.48-openvz/arch/x86/include/asm/paravirt.h --- linux-2.6.32.48/arch/x86/include/asm/paravirt.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/paravirt.h 2011-11-17 20:09:04.000000000 -0500 @@ -648,6 +648,18 @@ static inline void set_pgd(pgd_t *pgdp, val); } +static inline void set_pgd_batched(pgd_t *pgdp, pgd_t pgd) +{ + pgdval_t val = native_pgd_val(pgd); + + if (sizeof(pgdval_t) > sizeof(long)) + PVOP_VCALL3(pv_mmu_ops.set_pgd_batched, pgdp, + val, (u64)val >> 32); + else + PVOP_VCALL2(pv_mmu_ops.set_pgd_batched, pgdp, + val); +} + static inline void pgd_clear(pgd_t *pgdp) { set_pgd(pgdp, __pgd(0)); @@ -729,6 +741,21 @@ static inline void __set_fixmap(unsigned pv_mmu_ops.set_fixmap(idx, phys, flags); } +#ifdef CONFIG_PAX_KERNEXEC +static inline unsigned long pax_open_kernel(void) +{ + return PVOP_CALL0(unsigned long, pv_mmu_ops.pax_open_kernel); +} + +static inline unsigned long pax_close_kernel(void) +{ + return PVOP_CALL0(unsigned long, pv_mmu_ops.pax_close_kernel); +} +#else +static inline unsigned long pax_open_kernel(void) { return 0; } +static inline unsigned long pax_close_kernel(void) { return 0; } +#endif + #if defined(CONFIG_SMP) && defined(CONFIG_PARAVIRT_SPINLOCKS) static inline int __raw_spin_is_locked(struct raw_spinlock *lock) @@ -945,7 +972,7 @@ extern void default_banner(void); #define PARA_PATCH(struct, off) ((PARAVIRT_PATCH_##struct + (off)) / 4) #define PARA_SITE(ptype, clobbers, ops) _PVSITE(ptype, clobbers, ops, .long, 4) -#define PARA_INDIRECT(addr) *%cs:addr +#define PARA_INDIRECT(addr) *%ss:addr #endif #define INTERRUPT_RETURN \ @@ -1022,6 +1049,21 @@ extern void default_banner(void); PARA_SITE(PARA_PATCH(pv_cpu_ops, PV_CPU_irq_enable_sysexit), \ CLBR_NONE, \ jmp PARA_INDIRECT(pv_cpu_ops+PV_CPU_irq_enable_sysexit)) + +#define GET_CR0_INTO_RDI \ + call PARA_INDIRECT(pv_cpu_ops+PV_CPU_read_cr0); \ + mov %rax,%rdi + +#define SET_RDI_INTO_CR0 \ + call PARA_INDIRECT(pv_cpu_ops+PV_CPU_write_cr0) + +#define GET_CR3_INTO_RDI \ + call PARA_INDIRECT(pv_mmu_ops+PV_MMU_read_cr3); \ + mov %rax,%rdi + +#define SET_RDI_INTO_CR3 \ + call PARA_INDIRECT(pv_mmu_ops+PV_MMU_write_cr3) + #endif /* CONFIG_X86_32 */ #endif /* __ASSEMBLY__ */ diff -urNp linux-2.6.32.48/arch/x86/include/asm/paravirt_types.h linux-2.6.32.48-openvz/arch/x86/include/asm/paravirt_types.h --- linux-2.6.32.48/arch/x86/include/asm/paravirt_types.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/paravirt_types.h 2011-11-17 20:09:04.000000000 -0500 @@ -78,19 +78,19 @@ struct pv_init_ops { */ unsigned (*patch)(u8 type, u16 clobber, void *insnbuf, unsigned long addr, unsigned len); -}; +} __no_const; struct pv_lazy_ops { /* Set deferred update mode, used for batching operations. */ void (*enter)(void); void (*leave)(void); -}; +} __no_const; struct pv_time_ops { unsigned long long (*sched_clock)(void); unsigned long (*get_tsc_khz)(void); -}; +} __no_const; struct pv_cpu_ops { /* hooks for various privileged instructions */ @@ -186,7 +186,7 @@ struct pv_cpu_ops { void (*start_context_switch)(struct task_struct *prev); void (*end_context_switch)(struct task_struct *next); -}; +} __no_const; struct pv_irq_ops { /* @@ -217,7 +217,7 @@ struct pv_apic_ops { unsigned long start_eip, unsigned long start_esp); #endif -}; +} __no_const; struct pv_mmu_ops { unsigned long (*read_cr2)(void); @@ -301,6 +301,7 @@ struct pv_mmu_ops { struct paravirt_callee_save make_pud; void (*set_pgd)(pgd_t *pudp, pgd_t pgdval); + void (*set_pgd_batched)(pgd_t *pudp, pgd_t pgdval); #endif /* PAGETABLE_LEVELS == 4 */ #endif /* PAGETABLE_LEVELS >= 3 */ @@ -316,6 +317,12 @@ struct pv_mmu_ops { an mfn. We can tell which is which from the index. */ void (*set_fixmap)(unsigned /* enum fixed_addresses */ idx, phys_addr_t phys, pgprot_t flags); + +#ifdef CONFIG_PAX_KERNEXEC + unsigned long (*pax_open_kernel)(void); + unsigned long (*pax_close_kernel)(void); +#endif + }; struct raw_spinlock; @@ -326,7 +333,7 @@ struct pv_lock_ops { void (*spin_lock_flags)(struct raw_spinlock *lock, unsigned long flags); int (*spin_trylock)(struct raw_spinlock *lock); void (*spin_unlock)(struct raw_spinlock *lock); -}; +} __no_const; /* This contains all the paravirt structures: we get a convenient * number for each function using the offset which we use to indicate diff -urNp linux-2.6.32.48/arch/x86/include/asm/pci_x86.h linux-2.6.32.48-openvz/arch/x86/include/asm/pci_x86.h --- linux-2.6.32.48/arch/x86/include/asm/pci_x86.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/pci_x86.h 2011-11-17 20:08:53.000000000 -0500 @@ -89,16 +89,16 @@ extern int (*pcibios_enable_irq)(struct extern void (*pcibios_disable_irq)(struct pci_dev *dev); struct pci_raw_ops { - int (*read)(unsigned int domain, unsigned int bus, unsigned int devfn, + int (* const read)(unsigned int domain, unsigned int bus, unsigned int devfn, int reg, int len, u32 *val); - int (*write)(unsigned int domain, unsigned int bus, unsigned int devfn, + int (* const write)(unsigned int domain, unsigned int bus, unsigned int devfn, int reg, int len, u32 val); }; -extern struct pci_raw_ops *raw_pci_ops; -extern struct pci_raw_ops *raw_pci_ext_ops; +extern const struct pci_raw_ops *raw_pci_ops; +extern const struct pci_raw_ops *raw_pci_ext_ops; -extern struct pci_raw_ops pci_direct_conf1; +extern const struct pci_raw_ops pci_direct_conf1; extern bool port_cf9_safe; /* arch_initcall level */ diff -urNp linux-2.6.32.48/arch/x86/include/asm/percpu.h linux-2.6.32.48-openvz/arch/x86/include/asm/percpu.h --- linux-2.6.32.48/arch/x86/include/asm/percpu.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/percpu.h 2011-11-17 20:09:04.000000000 -0500 @@ -78,6 +78,7 @@ do { \ if (0) { \ T__ tmp__; \ tmp__ = (val); \ + (void)tmp__; \ } \ switch (sizeof(var)) { \ case 1: \ diff -urNp linux-2.6.32.48/arch/x86/include/asm/pgalloc.h linux-2.6.32.48-openvz/arch/x86/include/asm/pgalloc.h --- linux-2.6.32.48/arch/x86/include/asm/pgalloc.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/pgalloc.h 2011-11-17 20:08:53.000000000 -0500 @@ -63,6 +63,13 @@ static inline void pmd_populate_kernel(s pmd_t *pmd, pte_t *pte) { paravirt_alloc_pte(mm, __pa(pte) >> PAGE_SHIFT); + set_pmd(pmd, __pmd(__pa(pte) | _KERNPG_TABLE)); +} + +static inline void pmd_populate_user(struct mm_struct *mm, + pmd_t *pmd, pte_t *pte) +{ + paravirt_alloc_pte(mm, __pa(pte) >> PAGE_SHIFT); set_pmd(pmd, __pmd(__pa(pte) | _PAGE_TABLE)); } @@ -80,7 +87,7 @@ static inline void pmd_populate(struct m #if PAGETABLE_LEVELS > 2 static inline pmd_t *pmd_alloc_one(struct mm_struct *mm, unsigned long addr) { - return (pmd_t *)get_zeroed_page(GFP_KERNEL|__GFP_REPEAT); + return (pmd_t *)get_zeroed_page(GFP_KERNEL_UBC|__GFP_REPEAT); } static inline void pmd_free(struct mm_struct *mm, pmd_t *pmd) @@ -116,7 +123,7 @@ static inline void pgd_populate(struct m static inline pud_t *pud_alloc_one(struct mm_struct *mm, unsigned long addr) { - return (pud_t *)get_zeroed_page(GFP_KERNEL|__GFP_REPEAT); + return (pud_t *)get_zeroed_page(GFP_KERNEL_UBC|__GFP_REPEAT); } static inline void pud_free(struct mm_struct *mm, pud_t *pud) diff -urNp linux-2.6.32.48/arch/x86/include/asm/pgtable-2level.h linux-2.6.32.48-openvz/arch/x86/include/asm/pgtable-2level.h --- linux-2.6.32.48/arch/x86/include/asm/pgtable-2level.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/pgtable-2level.h 2011-11-17 20:08:53.000000000 -0500 @@ -18,7 +18,9 @@ static inline void native_set_pte(pte_t static inline void native_set_pmd(pmd_t *pmdp, pmd_t pmd) { + pax_open_kernel(); *pmdp = pmd; + pax_close_kernel(); } static inline void native_set_pte_atomic(pte_t *ptep, pte_t pte) diff -urNp linux-2.6.32.48/arch/x86/include/asm/pgtable_32.h linux-2.6.32.48-openvz/arch/x86/include/asm/pgtable_32.h --- linux-2.6.32.48/arch/x86/include/asm/pgtable_32.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/pgtable_32.h 2011-11-17 20:08:53.000000000 -0500 @@ -26,9 +26,6 @@ struct mm_struct; struct vm_area_struct; -extern pgd_t swapper_pg_dir[1024]; -extern pgd_t trampoline_pg_dir[1024]; - static inline void pgtable_cache_init(void) { } static inline void check_pgt_cache(void) { } void paging_init(void); @@ -49,6 +46,12 @@ extern void set_pmd_pfn(unsigned long, u # include #endif +extern pgd_t swapper_pg_dir[PTRS_PER_PGD]; +extern pgd_t trampoline_pg_dir[PTRS_PER_PGD]; +#ifdef CONFIG_X86_PAE +extern pmd_t swapper_pm_dir[PTRS_PER_PGD][PTRS_PER_PMD]; +#endif + #if defined(CONFIG_HIGHPTE) #define __KM_PTE \ (in_nmi() ? KM_NMI_PTE : \ @@ -73,7 +76,9 @@ extern void set_pmd_pfn(unsigned long, u /* Clear a kernel PTE and flush it from the TLB */ #define kpte_clear_flush(ptep, vaddr) \ do { \ + pax_open_kernel(); \ pte_clear(&init_mm, (vaddr), (ptep)); \ + pax_close_kernel(); \ __flush_tlb_one((vaddr)); \ } while (0) @@ -85,6 +90,9 @@ do { \ #endif /* !__ASSEMBLY__ */ +#define HAVE_ARCH_UNMAPPED_AREA +#define HAVE_ARCH_UNMAPPED_AREA_TOPDOWN + /* * kern_addr_valid() is (1) for FLATMEM and (0) for * SPARSEMEM and DISCONTIGMEM diff -urNp linux-2.6.32.48/arch/x86/include/asm/pgtable_32_types.h linux-2.6.32.48-openvz/arch/x86/include/asm/pgtable_32_types.h --- linux-2.6.32.48/arch/x86/include/asm/pgtable_32_types.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/pgtable_32_types.h 2011-11-17 20:08:53.000000000 -0500 @@ -8,7 +8,7 @@ */ #ifdef CONFIG_X86_PAE # include -# define PMD_SIZE (1UL << PMD_SHIFT) +# define PMD_SIZE (_AC(1, UL) << PMD_SHIFT) # define PMD_MASK (~(PMD_SIZE - 1)) #else # include @@ -46,6 +46,19 @@ extern bool __vmalloc_start_set; /* set # define VMALLOC_END (FIXADDR_START - 2 * PAGE_SIZE) #endif +#ifdef CONFIG_PAX_KERNEXEC +#ifndef __ASSEMBLY__ +extern unsigned char MODULES_EXEC_VADDR[]; +extern unsigned char MODULES_EXEC_END[]; +#endif +#include +#define ktla_ktva(addr) (addr + LOAD_PHYSICAL_ADDR + PAGE_OFFSET) +#define ktva_ktla(addr) (addr - LOAD_PHYSICAL_ADDR - PAGE_OFFSET) +#else +#define ktla_ktva(addr) (addr) +#define ktva_ktla(addr) (addr) +#endif + #define MODULES_VADDR VMALLOC_START #define MODULES_END VMALLOC_END #define MODULES_LEN (MODULES_VADDR - MODULES_END) diff -urNp linux-2.6.32.48/arch/x86/include/asm/pgtable-3level.h linux-2.6.32.48-openvz/arch/x86/include/asm/pgtable-3level.h --- linux-2.6.32.48/arch/x86/include/asm/pgtable-3level.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/pgtable-3level.h 2011-11-17 20:08:53.000000000 -0500 @@ -38,12 +38,16 @@ static inline void native_set_pte_atomic static inline void native_set_pmd(pmd_t *pmdp, pmd_t pmd) { + pax_open_kernel(); set_64bit((unsigned long long *)(pmdp), native_pmd_val(pmd)); + pax_close_kernel(); } static inline void native_set_pud(pud_t *pudp, pud_t pud) { + pax_open_kernel(); set_64bit((unsigned long long *)(pudp), native_pud_val(pud)); + pax_close_kernel(); } /* diff -urNp linux-2.6.32.48/arch/x86/include/asm/pgtable_64.h linux-2.6.32.48-openvz/arch/x86/include/asm/pgtable_64.h --- linux-2.6.32.48/arch/x86/include/asm/pgtable_64.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/pgtable_64.h 2011-11-17 20:09:04.000000000 -0500 @@ -16,10 +16,13 @@ extern pud_t level3_kernel_pgt[512]; extern pud_t level3_ident_pgt[512]; +extern pud_t level3_vmalloc_pgt[512]; +extern pud_t level3_vmemmap_pgt[512]; +extern pud_t level2_vmemmap_pgt[512]; extern pmd_t level2_kernel_pgt[512]; extern pmd_t level2_fixmap_pgt[512]; -extern pmd_t level2_ident_pgt[512]; -extern pgd_t init_level4_pgt[]; +extern pmd_t level2_ident_pgt[512*2]; +extern pgd_t init_level4_pgt[512]; #define swapper_pg_dir init_level4_pgt @@ -74,7 +77,9 @@ static inline pte_t native_ptep_get_and_ static inline void native_set_pmd(pmd_t *pmdp, pmd_t pmd) { + pax_open_kernel(); *pmdp = pmd; + pax_close_kernel(); } static inline void native_pmd_clear(pmd_t *pmd) @@ -94,6 +99,13 @@ static inline void native_pud_clear(pud_ static inline void native_set_pgd(pgd_t *pgdp, pgd_t pgd) { + pax_open_kernel(); + *pgdp = pgd; + pax_close_kernel(); +} + +static inline void native_set_pgd_batched(pgd_t *pgdp, pgd_t pgd) +{ *pgdp = pgd; } diff -urNp linux-2.6.32.48/arch/x86/include/asm/pgtable_64_types.h linux-2.6.32.48-openvz/arch/x86/include/asm/pgtable_64_types.h --- linux-2.6.32.48/arch/x86/include/asm/pgtable_64_types.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/pgtable_64_types.h 2011-11-17 20:08:53.000000000 -0500 @@ -59,5 +59,10 @@ typedef struct { pteval_t pte; } pte_t; #define MODULES_VADDR _AC(0xffffffffa0000000, UL) #define MODULES_END _AC(0xffffffffff000000, UL) #define MODULES_LEN (MODULES_END - MODULES_VADDR) +#define MODULES_EXEC_VADDR MODULES_VADDR +#define MODULES_EXEC_END MODULES_END + +#define ktla_ktva(addr) (addr) +#define ktva_ktla(addr) (addr) #endif /* _ASM_X86_PGTABLE_64_DEFS_H */ diff -urNp linux-2.6.32.48/arch/x86/include/asm/pgtable.h linux-2.6.32.48-openvz/arch/x86/include/asm/pgtable.h --- linux-2.6.32.48/arch/x86/include/asm/pgtable.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/pgtable.h 2011-11-17 20:09:04.000000000 -0500 @@ -39,6 +39,7 @@ extern struct list_head pgd_list; #ifndef __PAGETABLE_PUD_FOLDED #define set_pgd(pgdp, pgd) native_set_pgd(pgdp, pgd) +#define set_pgd_batched(pgdp, pgd) native_set_pgd_batched(pgdp, pgd) #define pgd_clear(pgd) native_pgd_clear(pgd) #endif @@ -74,12 +75,51 @@ extern struct list_head pgd_list; #define arch_end_context_switch(prev) do {} while(0) +#define pax_open_kernel() native_pax_open_kernel() +#define pax_close_kernel() native_pax_close_kernel() #endif /* CONFIG_PARAVIRT */ +#define __HAVE_ARCH_PAX_OPEN_KERNEL +#define __HAVE_ARCH_PAX_CLOSE_KERNEL + +#ifdef CONFIG_PAX_KERNEXEC +static inline unsigned long native_pax_open_kernel(void) +{ + unsigned long cr0; + + preempt_disable(); + barrier(); + cr0 = read_cr0() ^ X86_CR0_WP; + BUG_ON(unlikely(cr0 & X86_CR0_WP)); + write_cr0(cr0); + return cr0 ^ X86_CR0_WP; +} + +static inline unsigned long native_pax_close_kernel(void) +{ + unsigned long cr0; + + cr0 = read_cr0() ^ X86_CR0_WP; + BUG_ON(unlikely(!(cr0 & X86_CR0_WP))); + write_cr0(cr0); + barrier(); + preempt_enable_no_resched(); + return cr0 ^ X86_CR0_WP; +} +#else +static inline unsigned long native_pax_open_kernel(void) { return 0; } +static inline unsigned long native_pax_close_kernel(void) { return 0; } +#endif + /* * The following only work if pte_present() is true. * Undefined behaviour if not.. */ +static inline int pte_user(pte_t pte) +{ + return pte_val(pte) & _PAGE_USER; +} + static inline int pte_dirty(pte_t pte) { return pte_flags(pte) & _PAGE_DIRTY; @@ -167,9 +207,29 @@ static inline pte_t pte_wrprotect(pte_t return pte_clear_flags(pte, _PAGE_RW); } +static inline pte_t pte_mkread(pte_t pte) +{ + return __pte(pte_val(pte) | _PAGE_USER); +} + static inline pte_t pte_mkexec(pte_t pte) { - return pte_clear_flags(pte, _PAGE_NX); +#ifdef CONFIG_X86_PAE + if (__supported_pte_mask & _PAGE_NX) + return pte_clear_flags(pte, _PAGE_NX); + else +#endif + return pte_set_flags(pte, _PAGE_USER); +} + +static inline pte_t pte_exprotect(pte_t pte) +{ +#ifdef CONFIG_X86_PAE + if (__supported_pte_mask & _PAGE_NX) + return pte_set_flags(pte, _PAGE_NX); + else +#endif + return pte_clear_flags(pte, _PAGE_USER); } static inline pte_t pte_mkdirty(pte_t pte) @@ -302,6 +362,15 @@ pte_t *populate_extra_pte(unsigned long #endif #ifndef __ASSEMBLY__ + +#ifdef CONFIG_PAX_PER_CPU_PGD +extern pgd_t cpu_pgd[NR_CPUS][PTRS_PER_PGD]; +static inline pgd_t *get_cpu_pgd(unsigned int cpu) +{ + return cpu_pgd[cpu]; +} +#endif + #include static inline int pte_none(pte_t pte) @@ -472,7 +541,7 @@ static inline pud_t *pud_offset(pgd_t *p static inline int pgd_bad(pgd_t pgd) { - return (pgd_flags(pgd) & ~_PAGE_USER) != _KERNPG_TABLE; + return (pgd_flags(pgd) & ~(_PAGE_USER | _PAGE_NX)) != _KERNPG_TABLE; } static inline int pgd_none(pgd_t pgd) @@ -495,7 +564,12 @@ static inline int pgd_none(pgd_t pgd) * pgd_offset() returns a (pgd_t *) * pgd_index() is used get the offset into the pgd page's array of pgd_t's; */ -#define pgd_offset(mm, address) ((mm)->pgd + pgd_index((address))) +#define pgd_offset(mm, address) ((mm)->pgd + pgd_index(address)) + +#ifdef CONFIG_PAX_PER_CPU_PGD +#define pgd_offset_cpu(cpu, address) (get_cpu_pgd(cpu) + pgd_index(address)) +#endif + /* * a shortcut which implies the use of the kernel's pgd, instead * of a process's @@ -506,6 +580,20 @@ static inline int pgd_none(pgd_t pgd) #define KERNEL_PGD_BOUNDARY pgd_index(PAGE_OFFSET) #define KERNEL_PGD_PTRS (PTRS_PER_PGD - KERNEL_PGD_BOUNDARY) +#ifdef CONFIG_X86_32 +#define USER_PGD_PTRS KERNEL_PGD_BOUNDARY +#else +#define TASK_SIZE_MAX_SHIFT CONFIG_TASK_SIZE_MAX_SHIFT +#define USER_PGD_PTRS (_AC(1,UL) << (TASK_SIZE_MAX_SHIFT - PGDIR_SHIFT)) + +#ifdef CONFIG_PAX_MEMORY_UDEREF +#define PAX_USER_SHADOW_BASE (_AC(1,UL) << TASK_SIZE_MAX_SHIFT) +#else +#define PAX_USER_SHADOW_BASE (_AC(0,UL)) +#endif + +#endif + #ifndef __ASSEMBLY__ extern int direct_gbpages; @@ -611,11 +699,23 @@ static inline void ptep_set_wrprotect(st * dst and src can be on the same page, but the range must not overlap, * and must not cross a page boundary. */ -static inline void clone_pgd_range(pgd_t *dst, pgd_t *src, int count) +static inline void clone_pgd_range(pgd_t *dst, const pgd_t *src, int count) { - memcpy(dst, src, count * sizeof(pgd_t)); + pax_open_kernel(); + while (count--) + *dst++ = *src++; + pax_close_kernel(); } +#ifdef CONFIG_PAX_PER_CPU_PGD +extern void __clone_user_pgds(pgd_t *dst, const pgd_t *src, int count); +#endif + +#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) +extern void __shadow_user_pgds(pgd_t *dst, const pgd_t *src, int count); +#else +static inline void __shadow_user_pgds(pgd_t *dst, const pgd_t *src, int count) {} +#endif #include #endif /* __ASSEMBLY__ */ diff -urNp linux-2.6.32.48/arch/x86/include/asm/pgtable_types.h linux-2.6.32.48-openvz/arch/x86/include/asm/pgtable_types.h --- linux-2.6.32.48/arch/x86/include/asm/pgtable_types.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/pgtable_types.h 2011-11-17 20:08:53.000000000 -0500 @@ -16,12 +16,11 @@ #define _PAGE_BIT_PSE 7 /* 4 MB (or 2MB) page */ #define _PAGE_BIT_PAT 7 /* on 4KB pages */ #define _PAGE_BIT_GLOBAL 8 /* Global TLB entry PPro+ */ -#define _PAGE_BIT_UNUSED1 9 /* available for programmer */ +#define _PAGE_BIT_SPECIAL 9 /* special mappings, no associated struct page */ #define _PAGE_BIT_IOMAP 10 /* flag used to indicate IO mapping */ #define _PAGE_BIT_HIDDEN 11 /* hidden by kmemcheck */ #define _PAGE_BIT_PAT_LARGE 12 /* On 2MB or 1GB pages */ -#define _PAGE_BIT_SPECIAL _PAGE_BIT_UNUSED1 -#define _PAGE_BIT_CPA_TEST _PAGE_BIT_UNUSED1 +#define _PAGE_BIT_CPA_TEST _PAGE_BIT_SPECIAL #define _PAGE_BIT_NX 63 /* No execute: only valid after cpuid check */ /* If _PAGE_BIT_PRESENT is clear, we use these: */ @@ -39,7 +38,6 @@ #define _PAGE_DIRTY (_AT(pteval_t, 1) << _PAGE_BIT_DIRTY) #define _PAGE_PSE (_AT(pteval_t, 1) << _PAGE_BIT_PSE) #define _PAGE_GLOBAL (_AT(pteval_t, 1) << _PAGE_BIT_GLOBAL) -#define _PAGE_UNUSED1 (_AT(pteval_t, 1) << _PAGE_BIT_UNUSED1) #define _PAGE_IOMAP (_AT(pteval_t, 1) << _PAGE_BIT_IOMAP) #define _PAGE_PAT (_AT(pteval_t, 1) << _PAGE_BIT_PAT) #define _PAGE_PAT_LARGE (_AT(pteval_t, 1) << _PAGE_BIT_PAT_LARGE) @@ -55,8 +53,10 @@ #if defined(CONFIG_X86_64) || defined(CONFIG_X86_PAE) #define _PAGE_NX (_AT(pteval_t, 1) << _PAGE_BIT_NX) -#else +#elif defined(CONFIG_KMEMCHECK) #define _PAGE_NX (_AT(pteval_t, 0)) +#else +#define _PAGE_NX (_AT(pteval_t, 1) << _PAGE_BIT_HIDDEN) #endif #define _PAGE_FILE (_AT(pteval_t, 1) << _PAGE_BIT_FILE) @@ -93,6 +93,9 @@ #define PAGE_READONLY_EXEC __pgprot(_PAGE_PRESENT | _PAGE_USER | \ _PAGE_ACCESSED) +#define PAGE_READONLY_NOEXEC PAGE_READONLY +#define PAGE_SHARED_NOEXEC PAGE_SHARED + #define __PAGE_KERNEL_EXEC \ (_PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY | _PAGE_ACCESSED | _PAGE_GLOBAL) #define __PAGE_KERNEL (__PAGE_KERNEL_EXEC | _PAGE_NX) @@ -103,8 +106,8 @@ #define __PAGE_KERNEL_WC (__PAGE_KERNEL | _PAGE_CACHE_WC) #define __PAGE_KERNEL_NOCACHE (__PAGE_KERNEL | _PAGE_PCD | _PAGE_PWT) #define __PAGE_KERNEL_UC_MINUS (__PAGE_KERNEL | _PAGE_PCD) -#define __PAGE_KERNEL_VSYSCALL (__PAGE_KERNEL_RX | _PAGE_USER) -#define __PAGE_KERNEL_VSYSCALL_NOCACHE (__PAGE_KERNEL_VSYSCALL | _PAGE_PCD | _PAGE_PWT) +#define __PAGE_KERNEL_VSYSCALL (__PAGE_KERNEL_RO | _PAGE_USER) +#define __PAGE_KERNEL_VSYSCALL_NOCACHE (__PAGE_KERNEL_RO | _PAGE_PCD | _PAGE_PWT | _PAGE_USER) #define __PAGE_KERNEL_LARGE (__PAGE_KERNEL | _PAGE_PSE) #define __PAGE_KERNEL_LARGE_NOCACHE (__PAGE_KERNEL | _PAGE_CACHE_UC | _PAGE_PSE) #define __PAGE_KERNEL_LARGE_EXEC (__PAGE_KERNEL_EXEC | _PAGE_PSE) @@ -163,8 +166,8 @@ * bits are combined, this will alow user to access the high address mapped * VDSO in the presence of CONFIG_COMPAT_VDSO */ -#define PTE_IDENT_ATTR 0x003 /* PRESENT+RW */ -#define PDE_IDENT_ATTR 0x067 /* PRESENT+RW+USER+DIRTY+ACCESSED */ +#define PTE_IDENT_ATTR 0x063 /* PRESENT+RW+DIRTY+ACCESSED */ +#define PDE_IDENT_ATTR 0x063 /* PRESENT+RW+DIRTY+ACCESSED */ #define PGD_IDENT_ATTR 0x001 /* PRESENT (no other attributes) */ #endif @@ -202,7 +205,17 @@ static inline pgdval_t pgd_flags(pgd_t p { return native_pgd_val(pgd) & PTE_FLAGS_MASK; } +#endif +#if PAGETABLE_LEVELS == 3 +#include +#endif + +#if PAGETABLE_LEVELS == 2 +#include +#endif + +#ifndef __ASSEMBLY__ #if PAGETABLE_LEVELS > 3 typedef struct { pudval_t pud; } pud_t; @@ -216,8 +229,6 @@ static inline pudval_t native_pud_val(pu return pud.pud; } #else -#include - static inline pudval_t native_pud_val(pud_t pud) { return native_pgd_val(pud.pgd); @@ -237,8 +248,6 @@ static inline pmdval_t native_pmd_val(pm return pmd.pmd; } #else -#include - static inline pmdval_t native_pmd_val(pmd_t pmd) { return native_pgd_val(pmd.pud.pgd); @@ -278,7 +287,16 @@ typedef struct page *pgtable_t; extern pteval_t __supported_pte_mask; extern void set_nx(void); + +#ifdef CONFIG_X86_32 +#ifdef CONFIG_X86_PAE extern int nx_enabled; +#else +#define nx_enabled (0) +#endif +#else +#define nx_enabled (1) +#endif #define pgprot_writecombine pgprot_writecombine extern pgprot_t pgprot_writecombine(pgprot_t prot); diff -urNp linux-2.6.32.48/arch/x86/include/asm/processor.h linux-2.6.32.48-openvz/arch/x86/include/asm/processor.h --- linux-2.6.32.48/arch/x86/include/asm/processor.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/processor.h 2011-11-17 20:09:04.000000000 -0500 @@ -272,7 +272,7 @@ struct tss_struct { } ____cacheline_aligned; -DECLARE_PER_CPU_SHARED_ALIGNED(struct tss_struct, init_tss); +extern struct tss_struct init_tss[NR_CPUS]; /* * Save the original ist values for checking stack pointers during debugging @@ -911,11 +911,18 @@ static inline void spin_lock_prefetch(co */ #define TASK_SIZE PAGE_OFFSET #define TASK_SIZE_MAX TASK_SIZE + +#ifdef CONFIG_PAX_SEGMEXEC +#define SEGMEXEC_TASK_SIZE (TASK_SIZE / 2) +#define STACK_TOP ((current->mm->pax_flags & MF_PAX_SEGMEXEC)?SEGMEXEC_TASK_SIZE:TASK_SIZE) +#else #define STACK_TOP TASK_SIZE -#define STACK_TOP_MAX STACK_TOP +#endif + +#define STACK_TOP_MAX TASK_SIZE #define INIT_THREAD { \ - .sp0 = sizeof(init_stack) + (long)&init_stack, \ + .sp0 = sizeof(init_stack) + (long)&init_stack - 8, \ .vm86_info = NULL, \ .sysenter_cs = __KERNEL_CS, \ .io_bitmap_ptr = NULL, \ @@ -929,7 +936,7 @@ static inline void spin_lock_prefetch(co */ #define INIT_TSS { \ .x86_tss = { \ - .sp0 = sizeof(init_stack) + (long)&init_stack, \ + .sp0 = sizeof(init_stack) + (long)&init_stack - 8, \ .ss0 = __KERNEL_DS, \ .ss1 = __KERNEL_CS, \ .io_bitmap_base = INVALID_IO_BITMAP_OFFSET, \ @@ -940,11 +947,7 @@ static inline void spin_lock_prefetch(co extern unsigned long thread_saved_pc(struct task_struct *tsk); #define THREAD_SIZE_LONGS (THREAD_SIZE/sizeof(unsigned long)) -#define KSTK_TOP(info) \ -({ \ - unsigned long *__ptr = (unsigned long *)(info); \ - (unsigned long)(&__ptr[THREAD_SIZE_LONGS]); \ -}) +#define KSTK_TOP(info) ((container_of(info, struct task_struct, tinfo))->thread.sp0) /* * The below -8 is to reserve 8 bytes on top of the ring0 stack. @@ -959,7 +962,7 @@ extern unsigned long thread_saved_pc(str #define task_pt_regs(task) \ ({ \ struct pt_regs *__regs__; \ - __regs__ = (struct pt_regs *)(KSTK_TOP(task_stack_page(task))-8); \ + __regs__ = (struct pt_regs *)((task)->thread.sp0); \ __regs__ - 1; \ }) @@ -969,13 +972,12 @@ extern unsigned long thread_saved_pc(str /* * User space process size. 47bits minus one guard page. */ -#define TASK_SIZE_MAX ((1UL << 47) - PAGE_SIZE) +#define TASK_SIZE_MAX ((1UL << TASK_SIZE_MAX_SHIFT) - PAGE_SIZE) /* This decides where the kernel will search for a free chunk of vm * space during mmap's. */ -#define IA32_PAGE_OFFSET ((current->personality & ADDR_LIMIT_3GB) ? \ - 0xc0000000 : 0xFFFFe000) +#define IA32_PAGE_OFFSET 0xc0000000 #define TASK_SIZE (test_thread_flag(TIF_IA32) ? \ IA32_PAGE_OFFSET : TASK_SIZE_MAX) @@ -986,11 +988,11 @@ extern unsigned long thread_saved_pc(str #define STACK_TOP_MAX TASK_SIZE_MAX #define INIT_THREAD { \ - .sp0 = (unsigned long)&init_stack + sizeof(init_stack) \ + .sp0 = (unsigned long)&init_stack + sizeof(init_stack) - 16 \ } #define INIT_TSS { \ - .x86_tss.sp0 = (unsigned long)&init_stack + sizeof(init_stack) \ + .x86_tss.sp0 = (unsigned long)&init_stack + sizeof(init_stack) - 16 \ } /* @@ -1012,6 +1014,10 @@ extern void start_thread(struct pt_regs */ #define TASK_UNMAPPED_BASE (PAGE_ALIGN(TASK_SIZE / 3)) +#ifdef CONFIG_PAX_SEGMEXEC +#define SEGMEXEC_TASK_UNMAPPED_BASE (PAGE_ALIGN(SEGMEXEC_TASK_SIZE / 3)) +#endif + #define KSTK_EIP(task) (task_pt_regs(task)->ip) /* Get/set a process' ability to use the timestamp counter instruction */ diff -urNp linux-2.6.32.48/arch/x86/include/asm/ptrace.h linux-2.6.32.48-openvz/arch/x86/include/asm/ptrace.h --- linux-2.6.32.48/arch/x86/include/asm/ptrace.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/ptrace.h 2011-11-17 20:08:53.000000000 -0500 @@ -151,28 +151,29 @@ static inline unsigned long regs_return_ } /* - * user_mode_vm(regs) determines whether a register set came from user mode. + * user_mode(regs) determines whether a register set came from user mode. * This is true if V8086 mode was enabled OR if the register set was from * protected mode with RPL-3 CS value. This tricky test checks that with * one comparison. Many places in the kernel can bypass this full check - * if they have already ruled out V8086 mode, so user_mode(regs) can be used. + * if they have already ruled out V8086 mode, so user_mode_novm(regs) can + * be used. */ -static inline int user_mode(struct pt_regs *regs) +static inline int user_mode_novm(struct pt_regs *regs) { #ifdef CONFIG_X86_32 return (regs->cs & SEGMENT_RPL_MASK) == USER_RPL; #else - return !!(regs->cs & 3); + return !!(regs->cs & SEGMENT_RPL_MASK); #endif } -static inline int user_mode_vm(struct pt_regs *regs) +static inline int user_mode(struct pt_regs *regs) { #ifdef CONFIG_X86_32 return ((regs->cs & SEGMENT_RPL_MASK) | (regs->flags & X86_VM_MASK)) >= USER_RPL; #else - return user_mode(regs); + return user_mode_novm(regs); #endif } diff -urNp linux-2.6.32.48/arch/x86/include/asm/reboot.h linux-2.6.32.48-openvz/arch/x86/include/asm/reboot.h --- linux-2.6.32.48/arch/x86/include/asm/reboot.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/reboot.h 2011-11-17 20:16:38.000000000 -0500 @@ -11,14 +11,14 @@ struct machine_ops { void (*power_off)(void); void (*shutdown)(void); void (*crash_shutdown)(struct pt_regs *); - void (*emergency_restart)(void); -}; + void (* __noreturn emergency_restart)(void); +} __no_const; extern struct machine_ops machine_ops; void native_machine_crash_shutdown(struct pt_regs *regs); void native_machine_shutdown(void); -void machine_real_restart(const unsigned char *code, int length); +void machine_real_restart(const unsigned char *code, unsigned int length); typedef void (*nmi_shootdown_cb)(int, struct die_args*); void nmi_shootdown_cpus(nmi_shootdown_cb callback); diff -urNp linux-2.6.32.48/arch/x86/include/asm/rwsem.h linux-2.6.32.48-openvz/arch/x86/include/asm/rwsem.h --- linux-2.6.32.48/arch/x86/include/asm/rwsem.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/rwsem.h 2011-11-17 20:08:53.000000000 -0500 @@ -118,6 +118,14 @@ static inline void __down_read(struct rw { asm volatile("# beginning down_read\n\t" LOCK_PREFIX _ASM_INC "(%1)\n\t" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + LOCK_PREFIX _ASM_DEC "(%1)\n\t" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + /* adds 0x00000001, returns the old value */ " jns 1f\n" " call call_rwsem_down_read_failed\n" @@ -139,6 +147,14 @@ static inline int __down_read_trylock(st "1:\n\t" " mov %1,%2\n\t" " add %3,%2\n\t" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + "sub %3,%2\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + " jle 2f\n\t" LOCK_PREFIX " cmpxchg %2,%0\n\t" " jnz 1b\n\t" @@ -160,6 +176,14 @@ static inline void __down_write_nested(s tmp = RWSEM_ACTIVE_WRITE_BIAS; asm volatile("# beginning down_write\n\t" LOCK_PREFIX " xadd %1,(%2)\n\t" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + "mov %1,(%2)\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + /* subtract 0x0000ffff, returns the old value */ " test %1,%1\n\t" /* was the count 0 before? */ @@ -198,6 +222,14 @@ static inline void __up_read(struct rw_s rwsem_count_t tmp = -RWSEM_ACTIVE_READ_BIAS; asm volatile("# beginning __up_read\n\t" LOCK_PREFIX " xadd %1,(%2)\n\t" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + "mov %1,(%2)\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + /* subtracts 1, returns the old value */ " jns 1f\n\t" " call call_rwsem_wake\n" @@ -216,6 +248,14 @@ static inline void __up_write(struct rw_ rwsem_count_t tmp; asm volatile("# beginning __up_write\n\t" LOCK_PREFIX " xadd %1,(%2)\n\t" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + "mov %1,(%2)\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + /* tries to transition 0xffff0001 -> 0x00000000 */ " jz 1f\n" @@ -234,6 +274,14 @@ static inline void __downgrade_write(str { asm volatile("# beginning __downgrade_write\n\t" LOCK_PREFIX _ASM_ADD "%2,(%1)\n\t" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + LOCK_PREFIX _ASM_SUB "%2,(%1)\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + /* * transitions 0xZZZZ0001 -> 0xYYYY0001 (i386) * 0xZZZZZZZZ00000001 -> 0xYYYYYYYY00000001 (x86_64) @@ -253,7 +301,15 @@ static inline void __downgrade_write(str static inline void rwsem_atomic_add(rwsem_count_t delta, struct rw_semaphore *sem) { - asm volatile(LOCK_PREFIX _ASM_ADD "%1,%0" + asm volatile(LOCK_PREFIX _ASM_ADD "%1,%0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + LOCK_PREFIX _ASM_SUB "%1,%0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + : "+m" (sem->count) : "er" (delta)); } @@ -266,7 +322,15 @@ static inline rwsem_count_t rwsem_atomic { rwsem_count_t tmp = delta; - asm volatile(LOCK_PREFIX "xadd %0,%1" + asm volatile(LOCK_PREFIX "xadd %0,%1\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + "mov %0,%1\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + : "+r" (tmp), "+m" (sem->count) : : "memory"); diff -urNp linux-2.6.32.48/arch/x86/include/asm/segment.h linux-2.6.32.48-openvz/arch/x86/include/asm/segment.h --- linux-2.6.32.48/arch/x86/include/asm/segment.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/segment.h 2011-11-17 20:09:04.000000000 -0500 @@ -62,10 +62,15 @@ * 26 - ESPFIX small SS * 27 - per-cpu [ offset to per-cpu data area ] * 28 - stack_canary-20 [ for stack protector ] - * 29 - unused - * 30 - unused + * 29 - PCI BIOS CS + * 30 - PCI BIOS DS * 31 - TSS for double fault handler */ +#define GDT_ENTRY_KERNEXEC_EFI_CS (1) +#define GDT_ENTRY_KERNEXEC_EFI_DS (2) +#define __KERNEXEC_EFI_CS (GDT_ENTRY_KERNEXEC_EFI_CS*8) +#define __KERNEXEC_EFI_DS (GDT_ENTRY_KERNEXEC_EFI_DS*8) + #define GDT_ENTRY_TLS_MIN 6 #define GDT_ENTRY_TLS_MAX (GDT_ENTRY_TLS_MIN + GDT_ENTRY_TLS_ENTRIES - 1) @@ -77,6 +82,8 @@ #define GDT_ENTRY_KERNEL_CS (GDT_ENTRY_KERNEL_BASE + 0) +#define GDT_ENTRY_KERNEXEC_KERNEL_CS (4) + #define GDT_ENTRY_KERNEL_DS (GDT_ENTRY_KERNEL_BASE + 1) #define GDT_ENTRY_TSS (GDT_ENTRY_KERNEL_BASE + 4) @@ -88,7 +95,7 @@ #define GDT_ENTRY_ESPFIX_SS (GDT_ENTRY_KERNEL_BASE + 14) #define __ESPFIX_SS (GDT_ENTRY_ESPFIX_SS * 8) -#define GDT_ENTRY_PERCPU (GDT_ENTRY_KERNEL_BASE + 15) +#define GDT_ENTRY_PERCPU (GDT_ENTRY_KERNEL_BASE + 15) #ifdef CONFIG_SMP #define __KERNEL_PERCPU (GDT_ENTRY_PERCPU * 8) #else @@ -102,6 +109,12 @@ #define __KERNEL_STACK_CANARY 0 #endif +#define GDT_ENTRY_PCIBIOS_CS (GDT_ENTRY_KERNEL_BASE + 17) +#define __PCIBIOS_CS (GDT_ENTRY_PCIBIOS_CS * 8) + +#define GDT_ENTRY_PCIBIOS_DS (GDT_ENTRY_KERNEL_BASE + 18) +#define __PCIBIOS_DS (GDT_ENTRY_PCIBIOS_DS * 8) + #define GDT_ENTRY_DOUBLEFAULT_TSS 31 /* @@ -139,7 +152,7 @@ */ /* Matches PNP_CS32 and PNP_CS16 (they must be consecutive) */ -#define SEGMENT_IS_PNP_CODE(x) (((x) & 0xf4) == GDT_ENTRY_PNPBIOS_BASE * 8) +#define SEGMENT_IS_PNP_CODE(x) (((x) & 0xFFFCU) == PNP_CS32 || ((x) & 0xFFFCU) == PNP_CS16) #else @@ -163,6 +176,8 @@ #define __USER32_CS (GDT_ENTRY_DEFAULT_USER32_CS * 8 + 3) #define __USER32_DS __USER_DS +#define GDT_ENTRY_KERNEXEC_KERNEL_CS 7 + #define GDT_ENTRY_TSS 8 /* needs two entries */ #define GDT_ENTRY_LDT 10 /* needs two entries */ #define GDT_ENTRY_TLS_MIN 12 @@ -183,6 +198,7 @@ #endif #define __KERNEL_CS (GDT_ENTRY_KERNEL_CS * 8) +#define __KERNEXEC_KERNEL_CS (GDT_ENTRY_KERNEXEC_KERNEL_CS * 8) #define __KERNEL_DS (GDT_ENTRY_KERNEL_DS * 8) #define __USER_DS (GDT_ENTRY_DEFAULT_USER_DS* 8 + 3) #define __USER_CS (GDT_ENTRY_DEFAULT_USER_CS* 8 + 3) diff -urNp linux-2.6.32.48/arch/x86/include/asm/smp.h linux-2.6.32.48-openvz/arch/x86/include/asm/smp.h --- linux-2.6.32.48/arch/x86/include/asm/smp.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/smp.h 2011-11-17 20:09:04.000000000 -0500 @@ -24,7 +24,7 @@ extern unsigned int num_processors; DECLARE_PER_CPU(cpumask_var_t, cpu_sibling_map); DECLARE_PER_CPU(cpumask_var_t, cpu_core_map); DECLARE_PER_CPU(u16, cpu_llc_id); -DECLARE_PER_CPU(int, cpu_number); +DECLARE_PER_CPU(unsigned int, cpu_number); static inline struct cpumask *cpu_sibling_mask(int cpu) { @@ -40,10 +40,7 @@ DECLARE_EARLY_PER_CPU(u16, x86_cpu_to_ap DECLARE_EARLY_PER_CPU(u16, x86_bios_cpu_apicid); /* Static state in head.S used to set up a CPU */ -extern struct { - void *sp; - unsigned short ss; -} stack_start; +extern unsigned long stack_start; /* Initial stack pointer address */ struct smp_ops { void (*smp_prepare_boot_cpu)(void); @@ -60,7 +57,7 @@ struct smp_ops { void (*send_call_func_ipi)(const struct cpumask *mask); void (*send_call_func_single_ipi)(int cpu); -}; +} __no_const; /* Globals due to paravirt */ extern void set_cpu_sibling_map(int cpu); @@ -175,14 +172,8 @@ extern unsigned disabled_cpus __cpuinitd extern int safe_smp_processor_id(void); #elif defined(CONFIG_X86_64_SMP) -#define raw_smp_processor_id() (percpu_read(cpu_number)) - -#define stack_smp_processor_id() \ -({ \ - struct thread_info *ti; \ - __asm__("andq %%rsp,%0; ":"=r" (ti) : "0" (CURRENT_MASK)); \ - ti->cpu; \ -}) +#define raw_smp_processor_id() (percpu_read(cpu_number)) +#define stack_smp_processor_id() raw_smp_processor_id() #define safe_smp_processor_id() smp_processor_id() #endif diff -urNp linux-2.6.32.48/arch/x86/include/asm/spinlock.h linux-2.6.32.48-openvz/arch/x86/include/asm/spinlock.h --- linux-2.6.32.48/arch/x86/include/asm/spinlock.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/spinlock.h 2011-11-17 20:08:53.000000000 -0500 @@ -249,6 +249,14 @@ static inline int __raw_write_can_lock(r static inline void __raw_read_lock(raw_rwlock_t *rw) { asm volatile(LOCK_PREFIX " subl $1,(%0)\n\t" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + LOCK_PREFIX " addl $1,(%0)\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + "jns 1f\n" "call __read_lock_failed\n\t" "1:\n" @@ -258,6 +266,14 @@ static inline void __raw_read_lock(raw_r static inline void __raw_write_lock(raw_rwlock_t *rw) { asm volatile(LOCK_PREFIX " subl %1,(%0)\n\t" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + LOCK_PREFIX " addl %1,(%0)\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + "jz 1f\n" "call __write_lock_failed\n\t" "1:\n" @@ -286,12 +302,29 @@ static inline int __raw_write_trylock(ra static inline void __raw_read_unlock(raw_rwlock_t *rw) { - asm volatile(LOCK_PREFIX "incl %0" :"+m" (rw->lock) : : "memory"); + asm volatile(LOCK_PREFIX "incl %0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + LOCK_PREFIX "decl %0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + + :"+m" (rw->lock) : : "memory"); } static inline void __raw_write_unlock(raw_rwlock_t *rw) { - asm volatile(LOCK_PREFIX "addl %1, %0" + asm volatile(LOCK_PREFIX "addl %1, %0\n" + +#ifdef CONFIG_PAX_REFCOUNT + "jno 0f\n" + LOCK_PREFIX "subl %1, %0\n" + "int $4\n0:\n" + _ASM_EXTABLE(0b, 0b) +#endif + : "+m" (rw->lock) : "i" (RW_LOCK_BIAS) : "memory"); } diff -urNp linux-2.6.32.48/arch/x86/include/asm/stackprotector.h linux-2.6.32.48-openvz/arch/x86/include/asm/stackprotector.h --- linux-2.6.32.48/arch/x86/include/asm/stackprotector.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/stackprotector.h 2011-11-17 20:09:04.000000000 -0500 @@ -48,7 +48,7 @@ * head_32 for boot CPU and setup_per_cpu_areas() for others. */ #define GDT_STACK_CANARY_INIT \ - [GDT_ENTRY_STACK_CANARY] = GDT_ENTRY_INIT(0x4090, 0, 0x18), + [GDT_ENTRY_STACK_CANARY] = GDT_ENTRY_INIT(0x4090, 0, 0x17), /* * Initialize the stackprotector canary value. @@ -113,7 +113,7 @@ static inline void setup_stack_canary_se static inline void load_stack_canary_segment(void) { -#ifdef CONFIG_X86_32 +#if defined(CONFIG_X86_32) && !defined(CONFIG_PAX_MEMORY_UDEREF) asm volatile ("mov %0, %%gs" : : "r" (0)); #endif } diff -urNp linux-2.6.32.48/arch/x86/include/asm/system.h linux-2.6.32.48-openvz/arch/x86/include/asm/system.h --- linux-2.6.32.48/arch/x86/include/asm/system.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/system.h 2011-11-17 20:08:53.000000000 -0500 @@ -132,7 +132,7 @@ do { \ "thread_return:\n\t" \ "movq "__percpu_arg([current_task])",%%rsi\n\t" \ __switch_canary \ - "movq %P[thread_info](%%rsi),%%r8\n\t" \ + "movq "__percpu_arg([thread_info])",%%r8\n\t" \ "movq %%rax,%%rdi\n\t" \ "testl %[_tif_fork],%P[ti_flags](%%r8)\n\t" \ "jnz ret_from_fork\n\t" \ @@ -143,7 +143,7 @@ do { \ [threadrsp] "i" (offsetof(struct task_struct, thread.sp)), \ [ti_flags] "i" (offsetof(struct thread_info, flags)), \ [_tif_fork] "i" (_TIF_FORK), \ - [thread_info] "i" (offsetof(struct task_struct, stack)), \ + [thread_info] "m" (per_cpu_var(current_tinfo)), \ [current_task] "m" (per_cpu_var(current_task)) \ __switch_canary_iparam \ : "memory", "cc" __EXTRA_CLOBBER) @@ -200,7 +200,7 @@ static inline unsigned long get_limit(un { unsigned long __limit; asm("lsll %1,%0" : "=r" (__limit) : "r" (segment)); - return __limit + 1; + return __limit; } static inline void native_clts(void) @@ -340,7 +340,7 @@ void enable_hlt(void); void cpu_idle_wait(void); -extern unsigned long arch_align_stack(unsigned long sp); +#define arch_align_stack(x) ((x) & ~0xfUL) extern void free_init_pages(char *what, unsigned long begin, unsigned long end); void default_idle(void); diff -urNp linux-2.6.32.48/arch/x86/include/asm/thread_info.h linux-2.6.32.48-openvz/arch/x86/include/asm/thread_info.h --- linux-2.6.32.48/arch/x86/include/asm/thread_info.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/thread_info.h 2011-11-17 20:08:53.000000000 -0500 @@ -10,6 +10,7 @@ #include #include #include +#include /* * low level task data that entry.S needs immediate access to @@ -24,7 +25,6 @@ struct exec_domain; #include struct thread_info { - struct task_struct *task; /* main task structure */ struct exec_domain *exec_domain; /* execution domain */ __u32 flags; /* low level flags */ __u32 status; /* thread synchronous flags */ @@ -34,18 +34,12 @@ struct thread_info { mm_segment_t addr_limit; struct restart_block restart_block; void __user *sysenter_return; -#ifdef CONFIG_X86_32 - unsigned long previous_esp; /* ESP of the previous stack in - case of nested (IRQ) stacks - */ - __u8 supervisor_stack[0]; -#endif + unsigned long lowest_stack; int uaccess_err; }; -#define INIT_THREAD_INFO(tsk) \ +#define INIT_THREAD_INFO \ { \ - .task = &tsk, \ .exec_domain = &default_exec_domain, \ .flags = 0, \ .cpu = 0, \ @@ -56,7 +50,7 @@ struct thread_info { }, \ } -#define init_thread_info (init_thread_union.thread_info) +#define init_thread_info (init_thread_union.stack) #define init_stack (init_thread_union.stack) #else /* !__ASSEMBLY__ */ @@ -95,6 +89,7 @@ struct thread_info { #define TIF_DS_AREA_MSR 26 /* uses thread_struct.ds_area_msr */ #define TIF_LAZY_MMU_UPDATES 27 /* task is updating the mmu lazily */ #define TIF_SYSCALL_TRACEPOINT 28 /* syscall tracepoint instrumentation */ +#define TIF_RESUME 29 #define _TIF_SYSCALL_TRACE (1 << TIF_SYSCALL_TRACE) #define _TIF_NOTIFY_RESUME (1 << TIF_NOTIFY_RESUME) @@ -117,6 +112,7 @@ struct thread_info { #define _TIF_DS_AREA_MSR (1 << TIF_DS_AREA_MSR) #define _TIF_LAZY_MMU_UPDATES (1 << TIF_LAZY_MMU_UPDATES) #define _TIF_SYSCALL_TRACEPOINT (1 << TIF_SYSCALL_TRACEPOINT) +#define _TIF_RESUME (1< -#define KERNEL_STACK_OFFSET (5*8) - /* * macros/functions for gaining access to the thread information structure * preempt_count needs to be 1 initially, until the scheduler is functional. @@ -209,21 +201,8 @@ static inline struct thread_info *curren #ifndef __ASSEMBLY__ DECLARE_PER_CPU(unsigned long, kernel_stack); -static inline struct thread_info *current_thread_info(void) -{ - struct thread_info *ti; - ti = (void *)(percpu_read_stable(kernel_stack) + - KERNEL_STACK_OFFSET - THREAD_SIZE); - return ti; -} - -#else /* !__ASSEMBLY__ */ - -/* how to get the thread information struct from ASM */ -#define GET_THREAD_INFO(reg) \ - movq PER_CPU_VAR(kernel_stack),reg ; \ - subq $(THREAD_SIZE-KERNEL_STACK_OFFSET),reg - +/* how to get the current stack pointer from C */ +register unsigned long current_stack_pointer asm("rsp") __used; #endif #endif /* !X86_32 */ @@ -260,5 +239,16 @@ extern void arch_task_cache_init(void); extern void free_thread_info(struct thread_info *ti); extern int arch_dup_task_struct(struct task_struct *dst, struct task_struct *src); #define arch_task_cache_init arch_task_cache_init + +#define __HAVE_THREAD_FUNCTIONS +#define task_thread_info(task) (&(task)->tinfo) +#define task_stack_page(task) ((task)->stack) +#define setup_thread_stack(p, org) do {} while (0) +#define end_of_stack(p) ((unsigned long *)task_stack_page(p) + 1) + +#define __HAVE_ARCH_TASK_STRUCT_ALLOCATOR +extern struct task_struct *alloc_task_struct(void); +extern void free_task_struct(struct task_struct *); + #endif #endif /* _ASM_X86_THREAD_INFO_H */ diff -urNp linux-2.6.32.48/arch/x86/include/asm/tsc.h linux-2.6.32.48-openvz/arch/x86/include/asm/tsc.h --- linux-2.6.32.48/arch/x86/include/asm/tsc.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/tsc.h 2011-11-17 20:08:53.000000000 -0500 @@ -24,7 +24,7 @@ static inline cycles_t get_cycles(void) unsigned long long ret = 0; #ifndef CONFIG_X86_TSC - if (!cpu_has_tsc) + if (WARN_ON_ONCE(!cpu_has_tsc)) return 0; #endif rdtscll(ret); diff -urNp linux-2.6.32.48/arch/x86/include/asm/uaccess_32.h linux-2.6.32.48-openvz/arch/x86/include/asm/uaccess_32.h --- linux-2.6.32.48/arch/x86/include/asm/uaccess_32.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/uaccess_32.h 2011-11-17 20:08:53.000000000 -0500 @@ -44,6 +44,11 @@ unsigned long __must_check __copy_from_u static __always_inline unsigned long __must_check __copy_to_user_inatomic(void __user *to, const void *from, unsigned long n) { + pax_track_stack(); + + if ((long)n < 0) + return n; + if (__builtin_constant_p(n)) { unsigned long ret; @@ -62,6 +67,8 @@ __copy_to_user_inatomic(void __user *to, return ret; } } + if (!__builtin_constant_p(n)) + check_object_size(from, n, true); return __copy_to_user_ll(to, from, n); } @@ -83,12 +90,16 @@ static __always_inline unsigned long __m __copy_to_user(void __user *to, const void *from, unsigned long n) { might_fault(); + return __copy_to_user_inatomic(to, from, n); } static __always_inline unsigned long __copy_from_user_inatomic(void *to, const void __user *from, unsigned long n) { + if ((long)n < 0) + return n; + /* Avoid zeroing the tail if the copy fails.. * If 'n' is constant and 1, 2, or 4, we do still zero on a failure, * but as the zeroing behaviour is only significant when n is not @@ -138,6 +149,12 @@ static __always_inline unsigned long __copy_from_user(void *to, const void __user *from, unsigned long n) { might_fault(); + + pax_track_stack(); + + if ((long)n < 0) + return n; + if (__builtin_constant_p(n)) { unsigned long ret; @@ -153,6 +170,8 @@ __copy_from_user(void *to, const void __ return ret; } } + if (!__builtin_constant_p(n)) + check_object_size(to, n, false); return __copy_from_user_ll(to, from, n); } @@ -160,6 +179,10 @@ static __always_inline unsigned long __c const void __user *from, unsigned long n) { might_fault(); + + if ((long)n < 0) + return n; + if (__builtin_constant_p(n)) { unsigned long ret; @@ -182,14 +205,62 @@ static __always_inline unsigned long __copy_from_user_inatomic_nocache(void *to, const void __user *from, unsigned long n) { - return __copy_from_user_ll_nocache_nozero(to, from, n); + if ((long)n < 0) + return n; + + return __copy_from_user_ll_nocache_nozero(to, from, n); +} + +/** + * copy_to_user: - Copy a block of data into user space. + * @to: Destination address, in user space. + * @from: Source address, in kernel space. + * @n: Number of bytes to copy. + * + * Context: User context only. This function may sleep. + * + * Copy data from kernel space to user space. + * + * Returns number of bytes that could not be copied. + * On success, this will be zero. + */ +static __always_inline unsigned long __must_check +copy_to_user(void __user *to, const void *from, unsigned long n) +{ + if (access_ok(VERIFY_WRITE, to, n)) + n = __copy_to_user(to, from, n); + return n; +} + +/** + * copy_from_user: - Copy a block of data from user space. + * @to: Destination address, in kernel space. + * @from: Source address, in user space. + * @n: Number of bytes to copy. + * + * Context: User context only. This function may sleep. + * + * Copy data from user space to kernel space. + * + * Returns number of bytes that could not be copied. + * On success, this will be zero. + * + * If some data could not be copied, this function will pad the copied + * data to the requested size using zero bytes. + */ +static __always_inline unsigned long __must_check +copy_from_user(void *to, const void __user *from, unsigned long n) +{ + if (access_ok(VERIFY_READ, from, n)) + n = __copy_from_user(to, from, n); + else if ((long)n > 0) { + if (!__builtin_constant_p(n)) + check_object_size(to, n, false); + memset(to, 0, n); + } + return n; } -unsigned long __must_check copy_to_user(void __user *to, - const void *from, unsigned long n); -unsigned long __must_check copy_from_user(void *to, - const void __user *from, - unsigned long n); long __must_check strncpy_from_user(char *dst, const char __user *src, long count); long __must_check __strncpy_from_user(char *dst, diff -urNp linux-2.6.32.48/arch/x86/include/asm/uaccess_64.h linux-2.6.32.48-openvz/arch/x86/include/asm/uaccess_64.h --- linux-2.6.32.48/arch/x86/include/asm/uaccess_64.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/uaccess_64.h 2011-11-17 20:09:04.000000000 -0500 @@ -9,6 +9,9 @@ #include #include #include +#include + +#define set_fs(x) (current_thread_info()->addr_limit = (x)) /* * Copy To/From Userspace @@ -19,113 +22,203 @@ __must_check unsigned long copy_user_generic(void *to, const void *from, unsigned len); __must_check unsigned long -copy_to_user(void __user *to, const void *from, unsigned len); -__must_check unsigned long -copy_from_user(void *to, const void __user *from, unsigned len); -__must_check unsigned long copy_in_user(void __user *to, const void __user *from, unsigned len); static __always_inline __must_check -int __copy_from_user(void *dst, const void __user *src, unsigned size) +unsigned long __copy_from_user(void *dst, const void __user *src, unsigned size) { - int ret = 0; + unsigned ret = 0; might_fault(); - if (!__builtin_constant_p(size)) - return copy_user_generic(dst, (__force void *)src, size); + + if ((int)size < 0) + return size; + +#ifdef CONFIG_PAX_MEMORY_UDEREF + if (!__access_ok(VERIFY_READ, src, size)) + return size; +#endif + + if (!__builtin_constant_p(size)) { + check_object_size(dst, size, false); + +#ifdef CONFIG_PAX_MEMORY_UDEREF + if ((unsigned long)src < PAX_USER_SHADOW_BASE) + src += PAX_USER_SHADOW_BASE; +#endif + + return copy_user_generic(dst, (__force_kernel const void *)src, size); + } switch (size) { - case 1:__get_user_asm(*(u8 *)dst, (u8 __user *)src, + case 1:__get_user_asm(*(u8 *)dst, (const u8 __user *)src, ret, "b", "b", "=q", 1); return ret; - case 2:__get_user_asm(*(u16 *)dst, (u16 __user *)src, + case 2:__get_user_asm(*(u16 *)dst, (const u16 __user *)src, ret, "w", "w", "=r", 2); return ret; - case 4:__get_user_asm(*(u32 *)dst, (u32 __user *)src, + case 4:__get_user_asm(*(u32 *)dst, (const u32 __user *)src, ret, "l", "k", "=r", 4); return ret; - case 8:__get_user_asm(*(u64 *)dst, (u64 __user *)src, + case 8:__get_user_asm(*(u64 *)dst, (const u64 __user *)src, ret, "q", "", "=r", 8); return ret; case 10: - __get_user_asm(*(u64 *)dst, (u64 __user *)src, + __get_user_asm(*(u64 *)dst, (const u64 __user *)src, ret, "q", "", "=r", 10); if (unlikely(ret)) return ret; __get_user_asm(*(u16 *)(8 + (char *)dst), - (u16 __user *)(8 + (char __user *)src), + (const u16 __user *)(8 + (const char __user *)src), ret, "w", "w", "=r", 2); return ret; case 16: - __get_user_asm(*(u64 *)dst, (u64 __user *)src, + __get_user_asm(*(u64 *)dst, (const u64 __user *)src, ret, "q", "", "=r", 16); if (unlikely(ret)) return ret; __get_user_asm(*(u64 *)(8 + (char *)dst), - (u64 __user *)(8 + (char __user *)src), + (const u64 __user *)(8 + (const char __user *)src), ret, "q", "", "=r", 8); return ret; default: - return copy_user_generic(dst, (__force void *)src, size); + +#ifdef CONFIG_PAX_MEMORY_UDEREF + if ((unsigned long)src < PAX_USER_SHADOW_BASE) + src += PAX_USER_SHADOW_BASE; +#endif + + return copy_user_generic(dst, (__force_kernel const void *)src, size); } } static __always_inline __must_check -int __copy_to_user(void __user *dst, const void *src, unsigned size) +unsigned long __copy_to_user(void __user *dst, const void *src, unsigned size) { - int ret = 0; + unsigned ret = 0; might_fault(); - if (!__builtin_constant_p(size)) - return copy_user_generic((__force void *)dst, src, size); + + pax_track_stack(); + + if ((int)size < 0) + return size; + +#ifdef CONFIG_PAX_MEMORY_UDEREF + if (!__access_ok(VERIFY_WRITE, dst, size)) + return size; +#endif + + if (!__builtin_constant_p(size)) { + check_object_size(src, size, true); + +#ifdef CONFIG_PAX_MEMORY_UDEREF + if ((unsigned long)dst < PAX_USER_SHADOW_BASE) + dst += PAX_USER_SHADOW_BASE; +#endif + + return copy_user_generic((__force_kernel void *)dst, src, size); + } switch (size) { - case 1:__put_user_asm(*(u8 *)src, (u8 __user *)dst, + case 1:__put_user_asm(*(const u8 *)src, (u8 __user *)dst, ret, "b", "b", "iq", 1); return ret; - case 2:__put_user_asm(*(u16 *)src, (u16 __user *)dst, + case 2:__put_user_asm(*(const u16 *)src, (u16 __user *)dst, ret, "w", "w", "ir", 2); return ret; - case 4:__put_user_asm(*(u32 *)src, (u32 __user *)dst, + case 4:__put_user_asm(*(const u32 *)src, (u32 __user *)dst, ret, "l", "k", "ir", 4); return ret; - case 8:__put_user_asm(*(u64 *)src, (u64 __user *)dst, + case 8:__put_user_asm(*(const u64 *)src, (u64 __user *)dst, ret, "q", "", "er", 8); return ret; case 10: - __put_user_asm(*(u64 *)src, (u64 __user *)dst, + __put_user_asm(*(const u64 *)src, (u64 __user *)dst, ret, "q", "", "er", 10); if (unlikely(ret)) return ret; asm("":::"memory"); - __put_user_asm(4[(u16 *)src], 4 + (u16 __user *)dst, + __put_user_asm(4[(const u16 *)src], 4 + (u16 __user *)dst, ret, "w", "w", "ir", 2); return ret; case 16: - __put_user_asm(*(u64 *)src, (u64 __user *)dst, + __put_user_asm(*(const u64 *)src, (u64 __user *)dst, ret, "q", "", "er", 16); if (unlikely(ret)) return ret; asm("":::"memory"); - __put_user_asm(1[(u64 *)src], 1 + (u64 __user *)dst, + __put_user_asm(1[(const u64 *)src], 1 + (u64 __user *)dst, ret, "q", "", "er", 8); return ret; default: - return copy_user_generic((__force void *)dst, src, size); + +#ifdef CONFIG_PAX_MEMORY_UDEREF + if ((unsigned long)dst < PAX_USER_SHADOW_BASE) + dst += PAX_USER_SHADOW_BASE; +#endif + + return copy_user_generic((__force_kernel void *)dst, src, size); + } +} + +static __always_inline __must_check +unsigned long copy_to_user(void __user *to, const void *from, unsigned len) +{ + if (access_ok(VERIFY_WRITE, to, len)) + len = __copy_to_user(to, from, len); + return len; +} + +static __always_inline __must_check +unsigned long copy_from_user(void *to, const void __user *from, unsigned len) +{ + if ((int)len < 0) + return len; + + if (access_ok(VERIFY_READ, from, len)) + len = __copy_from_user(to, from, len); + else if ((int)len > 0) { + if (!__builtin_constant_p(len)) + check_object_size(to, len, false); + memset(to, 0, len); } + return len; } static __always_inline __must_check -int __copy_in_user(void __user *dst, const void __user *src, unsigned size) +unsigned long __copy_in_user(void __user *dst, const void __user *src, unsigned size) { - int ret = 0; + unsigned ret = 0; might_fault(); - if (!__builtin_constant_p(size)) - return copy_user_generic((__force void *)dst, - (__force void *)src, size); + + pax_track_stack(); + + if ((int)size < 0) + return size; + +#ifdef CONFIG_PAX_MEMORY_UDEREF + if (!__access_ok(VERIFY_READ, src, size)) + return size; + if (!__access_ok(VERIFY_WRITE, dst, size)) + return size; +#endif + + if (!__builtin_constant_p(size)) { + +#ifdef CONFIG_PAX_MEMORY_UDEREF + if ((unsigned long)src < PAX_USER_SHADOW_BASE) + src += PAX_USER_SHADOW_BASE; + if ((unsigned long)dst < PAX_USER_SHADOW_BASE) + dst += PAX_USER_SHADOW_BASE; +#endif + + return copy_user_generic((__force_kernel void *)dst, + (__force_kernel const void *)src, size); + } switch (size) { case 1: { u8 tmp; - __get_user_asm(tmp, (u8 __user *)src, + __get_user_asm(tmp, (const u8 __user *)src, ret, "b", "b", "=q", 1); if (likely(!ret)) __put_user_asm(tmp, (u8 __user *)dst, @@ -134,7 +227,7 @@ int __copy_in_user(void __user *dst, con } case 2: { u16 tmp; - __get_user_asm(tmp, (u16 __user *)src, + __get_user_asm(tmp, (const u16 __user *)src, ret, "w", "w", "=r", 2); if (likely(!ret)) __put_user_asm(tmp, (u16 __user *)dst, @@ -144,7 +237,7 @@ int __copy_in_user(void __user *dst, con case 4: { u32 tmp; - __get_user_asm(tmp, (u32 __user *)src, + __get_user_asm(tmp, (const u32 __user *)src, ret, "l", "k", "=r", 4); if (likely(!ret)) __put_user_asm(tmp, (u32 __user *)dst, @@ -153,7 +246,7 @@ int __copy_in_user(void __user *dst, con } case 8: { u64 tmp; - __get_user_asm(tmp, (u64 __user *)src, + __get_user_asm(tmp, (const u64 __user *)src, ret, "q", "", "=r", 8); if (likely(!ret)) __put_user_asm(tmp, (u64 __user *)dst, @@ -161,8 +254,16 @@ int __copy_in_user(void __user *dst, con return ret; } default: - return copy_user_generic((__force void *)dst, - (__force void *)src, size); + +#ifdef CONFIG_PAX_MEMORY_UDEREF + if ((unsigned long)src < PAX_USER_SHADOW_BASE) + src += PAX_USER_SHADOW_BASE; + if ((unsigned long)dst < PAX_USER_SHADOW_BASE) + dst += PAX_USER_SHADOW_BASE; +#endif + + return copy_user_generic((__force_kernel void *)dst, + (__force_kernel const void *)src, size); } } @@ -176,33 +277,75 @@ __must_check long strlen_user(const char __must_check unsigned long clear_user(void __user *mem, unsigned long len); __must_check unsigned long __clear_user(void __user *mem, unsigned long len); -__must_check long __copy_from_user_inatomic(void *dst, const void __user *src, - unsigned size); +static __must_check __always_inline unsigned long +__copy_from_user_inatomic(void *dst, const void __user *src, unsigned size) +{ + pax_track_stack(); + + if ((int)size < 0) + return size; + +#ifdef CONFIG_PAX_MEMORY_UDEREF + if (!__access_ok(VERIFY_READ, src, size)) + return size; + + if ((unsigned long)src < PAX_USER_SHADOW_BASE) + src += PAX_USER_SHADOW_BASE; +#endif -static __must_check __always_inline int + return copy_user_generic(dst, (__force_kernel const void *)src, size); +} + +static __must_check __always_inline unsigned long __copy_to_user_inatomic(void __user *dst, const void *src, unsigned size) { - return copy_user_generic((__force void *)dst, src, size); + if ((int)size < 0) + return size; + +#ifdef CONFIG_PAX_MEMORY_UDEREF + if (!__access_ok(VERIFY_WRITE, dst, size)) + return size; + + if ((unsigned long)dst < PAX_USER_SHADOW_BASE) + dst += PAX_USER_SHADOW_BASE; +#endif + + return copy_user_generic((__force_kernel void *)dst, src, size); } -extern long __copy_user_nocache(void *dst, const void __user *src, +extern unsigned long __copy_user_nocache(void *dst, const void __user *src, unsigned size, int zerorest); -static inline int -__copy_from_user_nocache(void *dst, const void __user *src, unsigned size) +static inline unsigned long __copy_from_user_nocache(void *dst, const void __user *src, unsigned size) { might_sleep(); + + if ((int)size < 0) + return size; + +#ifdef CONFIG_PAX_MEMORY_UDEREF + if (!__access_ok(VERIFY_READ, src, size)) + return size; +#endif + return __copy_user_nocache(dst, src, size, 1); } -static inline int -__copy_from_user_inatomic_nocache(void *dst, const void __user *src, +static inline unsigned long __copy_from_user_inatomic_nocache(void *dst, const void __user *src, unsigned size) { + if ((int)size < 0) + return size; + +#ifdef CONFIG_PAX_MEMORY_UDEREF + if (!__access_ok(VERIFY_READ, src, size)) + return size; +#endif + return __copy_user_nocache(dst, src, size, 0); } -unsigned long -copy_user_handle_tail(char *to, char *from, unsigned len, unsigned zerorest); +extern unsigned long +copy_user_handle_tail(char __user *to, char __user *from, unsigned len, unsigned zerorest); #endif /* _ASM_X86_UACCESS_64_H */ diff -urNp linux-2.6.32.48/arch/x86/include/asm/uaccess.h linux-2.6.32.48-openvz/arch/x86/include/asm/uaccess.h --- linux-2.6.32.48/arch/x86/include/asm/uaccess.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/uaccess.h 2011-11-17 20:09:04.000000000 -0500 @@ -8,12 +8,15 @@ #include #include #include +#include #include #include #define VERIFY_READ 0 #define VERIFY_WRITE 1 +extern void check_object_size(const void *ptr, unsigned long n, bool to); + /* * The fs value determines whether argument validity checking should be * performed or not. If get_fs() == USER_DS, checking is performed, with @@ -29,7 +32,12 @@ #define get_ds() (KERNEL_DS) #define get_fs() (current_thread_info()->addr_limit) +#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_MEMORY_UDEREF) +void __set_fs(mm_segment_t x); +void set_fs(mm_segment_t x); +#else #define set_fs(x) (current_thread_info()->addr_limit = (x)) +#endif #define segment_eq(a, b) ((a).seg == (b).seg) @@ -77,7 +85,33 @@ * checks that the pointer is in the user space range - after calling * this function, memory access functions may still return -EFAULT. */ -#define access_ok(type, addr, size) (likely(__range_not_ok(addr, size) == 0)) +#define __access_ok(type, addr, size) (likely(__range_not_ok(addr, size) == 0)) +#define access_ok(type, addr, size) \ +({ \ + long __size = size; \ + unsigned long __addr = (unsigned long)addr; \ + unsigned long __addr_ao = __addr & PAGE_MASK; \ + unsigned long __end_ao = __addr + __size - 1; \ + bool __ret_ao = __range_not_ok(__addr, __size) == 0; \ + if (__ret_ao && unlikely((__end_ao ^ __addr_ao) & PAGE_MASK)) { \ + while(__addr_ao <= __end_ao) { \ + char __c_ao; \ + __addr_ao += PAGE_SIZE; \ + if (__size > PAGE_SIZE) \ + cond_resched(); \ + if (__get_user(__c_ao, (char __user *)__addr)) \ + break; \ + if (type != VERIFY_WRITE) { \ + __addr = __addr_ao; \ + continue; \ + } \ + if (__put_user(__c_ao, (char __user *)__addr)) \ + break; \ + __addr = __addr_ao; \ + } \ + } \ + __ret_ao; \ +}) /* * The exception table consists of pairs of addresses: the first is the @@ -183,12 +217,20 @@ extern int __get_user_bad(void); asm volatile("call __put_user_" #size : "=a" (__ret_pu) \ : "0" ((typeof(*(ptr)))(x)), "c" (ptr) : "ebx") - +#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_MEMORY_UDEREF) +#define __copyuser_seg "gs;" +#define __COPYUSER_SET_ES "pushl %%gs; popl %%es\n" +#define __COPYUSER_RESTORE_ES "pushl %%ss; popl %%es\n" +#else +#define __copyuser_seg +#define __COPYUSER_SET_ES +#define __COPYUSER_RESTORE_ES +#endif #ifdef CONFIG_X86_32 #define __put_user_asm_u64(x, addr, err, errret) \ - asm volatile("1: movl %%eax,0(%2)\n" \ - "2: movl %%edx,4(%2)\n" \ + asm volatile("1: "__copyuser_seg"movl %%eax,0(%2)\n" \ + "2: "__copyuser_seg"movl %%edx,4(%2)\n" \ "3:\n" \ ".section .fixup,\"ax\"\n" \ "4: movl %3,%0\n" \ @@ -200,8 +242,8 @@ extern int __get_user_bad(void); : "A" (x), "r" (addr), "i" (errret), "0" (err)) #define __put_user_asm_ex_u64(x, addr) \ - asm volatile("1: movl %%eax,0(%1)\n" \ - "2: movl %%edx,4(%1)\n" \ + asm volatile("1: "__copyuser_seg"movl %%eax,0(%1)\n" \ + "2: "__copyuser_seg"movl %%edx,4(%1)\n" \ "3:\n" \ _ASM_EXTABLE(1b, 2b - 1b) \ _ASM_EXTABLE(2b, 3b - 2b) \ @@ -253,7 +295,7 @@ extern void __put_user_8(void); __typeof__(*(ptr)) __pu_val; \ __chk_user_ptr(ptr); \ might_fault(); \ - __pu_val = x; \ + __pu_val = (x); \ switch (sizeof(*(ptr))) { \ case 1: \ __put_user_x(1, __pu_val, ptr, __ret_pu); \ @@ -374,7 +416,7 @@ do { \ } while (0) #define __get_user_asm(x, addr, err, itype, rtype, ltype, errret) \ - asm volatile("1: mov"itype" %2,%"rtype"1\n" \ + asm volatile("1: "__copyuser_seg"mov"itype" %2,%"rtype"1\n"\ "2:\n" \ ".section .fixup,\"ax\"\n" \ "3: mov %3,%0\n" \ @@ -382,7 +424,7 @@ do { \ " jmp 2b\n" \ ".previous\n" \ _ASM_EXTABLE(1b, 3b) \ - : "=r" (err), ltype(x) \ + : "=r" (err), ltype (x) \ : "m" (__m(addr)), "i" (errret), "0" (err)) #define __get_user_size_ex(x, ptr, size) \ @@ -407,7 +449,7 @@ do { \ } while (0) #define __get_user_asm_ex(x, addr, itype, rtype, ltype) \ - asm volatile("1: mov"itype" %1,%"rtype"0\n" \ + asm volatile("1: "__copyuser_seg"mov"itype" %1,%"rtype"0\n"\ "2:\n" \ _ASM_EXTABLE(1b, 2b - 1b) \ : ltype(x) : "m" (__m(addr))) @@ -424,13 +466,24 @@ do { \ int __gu_err; \ unsigned long __gu_val; \ __get_user_size(__gu_val, (ptr), (size), __gu_err, -EFAULT); \ - (x) = (__force __typeof__(*(ptr)))__gu_val; \ + (x) = (__typeof__(*(ptr)))__gu_val; \ __gu_err; \ }) /* FIXME: this hack is definitely wrong -AK */ struct __large_struct { unsigned long buf[100]; }; -#define __m(x) (*(struct __large_struct __user *)(x)) +#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) +#define ____m(x) \ +({ \ + unsigned long ____x = (unsigned long)(x); \ + if (____x < PAX_USER_SHADOW_BASE) \ + ____x += PAX_USER_SHADOW_BASE; \ + (void __user *)____x; \ +}) +#else +#define ____m(x) (x) +#endif +#define __m(x) (*(struct __large_struct __user *)____m(x)) /* * Tell gcc we read from memory instead of writing: this is because @@ -438,7 +491,7 @@ struct __large_struct { unsigned long bu * aliasing issues. */ #define __put_user_asm(x, addr, err, itype, rtype, ltype, errret) \ - asm volatile("1: mov"itype" %"rtype"1,%2\n" \ + asm volatile("1: "__copyuser_seg"mov"itype" %"rtype"1,%2\n"\ "2:\n" \ ".section .fixup,\"ax\"\n" \ "3: mov %3,%0\n" \ @@ -446,10 +499,10 @@ struct __large_struct { unsigned long bu ".previous\n" \ _ASM_EXTABLE(1b, 3b) \ : "=r"(err) \ - : ltype(x), "m" (__m(addr)), "i" (errret), "0" (err)) + : ltype (x), "m" (__m(addr)), "i" (errret), "0" (err)) #define __put_user_asm_ex(x, addr, itype, rtype, ltype) \ - asm volatile("1: mov"itype" %"rtype"0,%1\n" \ + asm volatile("1: "__copyuser_seg"mov"itype" %"rtype"0,%1\n"\ "2:\n" \ _ASM_EXTABLE(1b, 2b - 1b) \ : : ltype(x), "m" (__m(addr))) @@ -488,8 +541,12 @@ struct __large_struct { unsigned long bu * On error, the variable @x is set to zero. */ +#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) +#define __get_user(x, ptr) get_user((x), (ptr)) +#else #define __get_user(x, ptr) \ __get_user_nocheck((x), (ptr), sizeof(*(ptr))) +#endif /** * __put_user: - Write a simple value into user space, with less checking. @@ -511,8 +568,12 @@ struct __large_struct { unsigned long bu * Returns zero on success, or -EFAULT on error. */ +#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) +#define __put_user(x, ptr) put_user((x), (ptr)) +#else #define __put_user(x, ptr) \ __put_user_nocheck((__typeof__(*(ptr)))(x), (ptr), sizeof(*(ptr))) +#endif #define __get_user_unaligned __get_user #define __put_user_unaligned __put_user @@ -530,7 +591,7 @@ struct __large_struct { unsigned long bu #define get_user_ex(x, ptr) do { \ unsigned long __gue_val; \ __get_user_size_ex((__gue_val), (ptr), (sizeof(*(ptr)))); \ - (x) = (__force __typeof__(*(ptr)))__gue_val; \ + (x) = (__typeof__(*(ptr)))__gue_val; \ } while (0) #ifdef CONFIG_X86_WP_WORKS_OK @@ -567,6 +628,7 @@ extern struct movsl_mask { #define ARCH_HAS_NOCACHE_UACCESS 1 +#define ARCH_HAS_SORT_EXTABLE #ifdef CONFIG_X86_32 # include "uaccess_32.h" #else diff -urNp linux-2.6.32.48/arch/x86/include/asm/unistd_32.h linux-2.6.32.48-openvz/arch/x86/include/asm/unistd_32.h --- linux-2.6.32.48/arch/x86/include/asm/unistd_32.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/unistd_32.h 2011-11-17 20:08:53.000000000 -0500 @@ -342,10 +342,22 @@ #define __NR_pwritev 334 #define __NR_rt_tgsigqueueinfo 335 #define __NR_perf_event_open 336 +#define __NR_fairsched_mknod 500 /* FairScheduler syscalls */ +#define __NR_fairsched_rmnod 501 +#define __NR_fairsched_chwt 502 +#define __NR_fairsched_mvpr 503 +#define __NR_fairsched_rate 504 +#define __NR_fairsched_vcpus 505 +#define __NR_getluid 510 +#define __NR_setluid 511 +#define __NR_setublimit 512 +#define __NR_ubstat 513 +#define __NR_lchmod 516 +#define __NR_lutime 517 #ifdef __KERNEL__ -#define NR_syscalls 337 +#define NR_syscalls 514 #define __ARCH_WANT_IPC_PARSE_VERSION #define __ARCH_WANT_OLD_READDIR diff -urNp linux-2.6.32.48/arch/x86/include/asm/unistd_64.h linux-2.6.32.48-openvz/arch/x86/include/asm/unistd_64.h --- linux-2.6.32.48/arch/x86/include/asm/unistd_64.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/unistd_64.h 2011-11-17 20:08:53.000000000 -0500 @@ -661,6 +661,30 @@ __SYSCALL(__NR_pwritev, sys_pwritev) __SYSCALL(__NR_rt_tgsigqueueinfo, sys_rt_tgsigqueueinfo) #define __NR_perf_event_open 298 __SYSCALL(__NR_perf_event_open, sys_perf_event_open) +#define __NR_fairsched_vcpus 499 +__SYSCALL(__NR_fairsched_vcpus, sys_fairsched_vcpus) +#define __NR_getluid 500 +__SYSCALL(__NR_getluid, sys_getluid) +#define __NR_setluid 501 +__SYSCALL(__NR_setluid, sys_setluid) +#define __NR_setublimit 502 +__SYSCALL(__NR_setublimit, sys_setublimit) +#define __NR_ubstat 503 +__SYSCALL(__NR_ubstat, sys_ubstat) +#define __NR_fairsched_mknod 504 /* FairScheduler syscalls */ +__SYSCALL(__NR_fairsched_mknod, sys_fairsched_mknod) +#define __NR_fairsched_rmnod 505 +__SYSCALL(__NR_fairsched_rmnod, sys_fairsched_rmnod) +#define __NR_fairsched_chwt 506 +__SYSCALL(__NR_fairsched_chwt, sys_fairsched_chwt) +#define __NR_fairsched_mvpr 507 +__SYSCALL(__NR_fairsched_mvpr, sys_fairsched_mvpr) +#define __NR_fairsched_rate 508 +__SYSCALL(__NR_fairsched_rate, sys_fairsched_rate) +#define __NR_lchmod 509 +__SYSCALL(__NR_lchmod, sys_lchmod) +#define __NR_lutime 510 +__SYSCALL(__NR_lutime, sys_lutime) #ifndef __NO_STUBS #define __ARCH_WANT_OLD_READDIR @@ -685,6 +709,7 @@ __SYSCALL(__NR_perf_event_open, sys_perf #define __ARCH_WANT_SYS_RT_SIGSUSPEND #define __ARCH_WANT_SYS_TIME #define __ARCH_WANT_COMPAT_SYS_TIME +#define __ARCH_WANT_SYS_RT_SIGSUSPEND #endif /* __NO_STUBS */ #ifdef __KERNEL__ diff -urNp linux-2.6.32.48/arch/x86/include/asm/vdso.h linux-2.6.32.48-openvz/arch/x86/include/asm/vdso.h --- linux-2.6.32.48/arch/x86/include/asm/vdso.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/vdso.h 2011-11-17 20:09:04.000000000 -0500 @@ -18,6 +18,7 @@ extern const char VDSO64_PRELINK[]; #if defined CONFIG_X86_32 || defined CONFIG_COMPAT extern const char VDSO32_PRELINK[]; +extern const char VDSO32_SYSENTER_RETURN[]; /* * Given a pointer to the vDSO image, find the pointer to VDSO32_name * as that symbol is defined in the vDSO sources or linker script. @@ -25,7 +26,7 @@ extern const char VDSO32_PRELINK[]; #define VDSO32_SYMBOL(base, name) \ ({ \ extern const char VDSO32_##name[]; \ - (void *)(VDSO32_##name - VDSO32_PRELINK + (unsigned long)(base)); \ + (void __user *)(VDSO32_##name - VDSO32_PRELINK + (unsigned long)(base)); \ }) #endif diff -urNp linux-2.6.32.48/arch/x86/include/asm/vgtod.h linux-2.6.32.48-openvz/arch/x86/include/asm/vgtod.h --- linux-2.6.32.48/arch/x86/include/asm/vgtod.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/vgtod.h 2011-11-17 20:08:53.000000000 -0500 @@ -14,6 +14,7 @@ struct vsyscall_gtod_data { int sysctl_enabled; struct timezone sys_tz; struct { /* extract of a clocksource struct */ + char name[8]; cycle_t (*vread)(void); cycle_t cycle_last; cycle_t mask; diff -urNp linux-2.6.32.48/arch/x86/include/asm/vmi.h linux-2.6.32.48-openvz/arch/x86/include/asm/vmi.h --- linux-2.6.32.48/arch/x86/include/asm/vmi.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/vmi.h 2011-11-17 20:08:53.000000000 -0500 @@ -191,6 +191,7 @@ struct vrom_header { u8 reserved[96]; /* Reserved for headers */ char vmi_init[8]; /* VMI_Init jump point */ char get_reloc[8]; /* VMI_GetRelocationInfo jump point */ + char rom_data[8048]; /* rest of the option ROM */ } __attribute__((packed)); struct pnp_header { diff -urNp linux-2.6.32.48/arch/x86/include/asm/vmi_time.h linux-2.6.32.48-openvz/arch/x86/include/asm/vmi_time.h --- linux-2.6.32.48/arch/x86/include/asm/vmi_time.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/vmi_time.h 2011-11-17 20:09:04.000000000 -0500 @@ -43,7 +43,7 @@ extern struct vmi_timer_ops { int (*wallclock_updated)(void); void (*set_alarm)(u32 flags, u64 expiry, u64 period); void (*cancel_alarm)(u32 flags); -} vmi_timer_ops; +} __no_const vmi_timer_ops; /* Prototypes */ extern void __init vmi_time_init(void); diff -urNp linux-2.6.32.48/arch/x86/include/asm/vsyscall.h linux-2.6.32.48-openvz/arch/x86/include/asm/vsyscall.h --- linux-2.6.32.48/arch/x86/include/asm/vsyscall.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/vsyscall.h 2011-11-17 20:08:53.000000000 -0500 @@ -15,9 +15,10 @@ enum vsyscall_num { #ifdef __KERNEL__ #include +#include +#include #define __section_vgetcpu_mode __attribute__ ((unused, __section__ (".vgetcpu_mode"), aligned(16))) -#define __section_jiffies __attribute__ ((unused, __section__ (".jiffies"), aligned(16))) /* Definitions for CONFIG_GENERIC_TIME definitions */ #define __section_vsyscall_gtod_data __attribute__ \ @@ -31,7 +32,6 @@ enum vsyscall_num { #define VGETCPU_LSL 2 extern int __vgetcpu_mode; -extern volatile unsigned long __jiffies; /* kernel space (writeable) */ extern int vgetcpu_mode; @@ -39,6 +39,9 @@ extern struct timezone sys_tz; extern void map_vsyscall(void); +extern int vgettimeofday(struct timeval * tv, struct timezone * tz); +extern time_t vtime(time_t *t); +extern long vgetcpu(unsigned *cpu, unsigned *node, struct getcpu_cache *tcache); #endif /* __KERNEL__ */ #endif /* _ASM_X86_VSYSCALL_H */ diff -urNp linux-2.6.32.48/arch/x86/include/asm/x86_init.h linux-2.6.32.48-openvz/arch/x86/include/asm/x86_init.h --- linux-2.6.32.48/arch/x86/include/asm/x86_init.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/x86_init.h 2011-11-17 20:09:04.000000000 -0500 @@ -28,7 +28,7 @@ struct x86_init_mpparse { void (*mpc_oem_bus_info)(struct mpc_bus *m, char *name); void (*find_smp_config)(unsigned int reserve); void (*get_smp_config)(unsigned int early); -}; +} __no_const; /** * struct x86_init_resources - platform specific resource related ops @@ -42,7 +42,7 @@ struct x86_init_resources { void (*probe_roms)(void); void (*reserve_resources)(void); char *(*memory_setup)(void); -}; +} __no_const; /** * struct x86_init_irqs - platform specific interrupt setup @@ -55,7 +55,7 @@ struct x86_init_irqs { void (*pre_vector_init)(void); void (*intr_init)(void); void (*trap_init)(void); -}; +} __no_const; /** * struct x86_init_oem - oem platform specific customizing functions @@ -65,7 +65,7 @@ struct x86_init_irqs { struct x86_init_oem { void (*arch_setup)(void); void (*banner)(void); -}; +} __no_const; /** * struct x86_init_paging - platform specific paging functions @@ -75,7 +75,7 @@ struct x86_init_oem { struct x86_init_paging { void (*pagetable_setup_start)(pgd_t *base); void (*pagetable_setup_done)(pgd_t *base); -}; +} __no_const; /** * struct x86_init_timers - platform specific timer setup @@ -88,7 +88,7 @@ struct x86_init_timers { void (*setup_percpu_clockev)(void); void (*tsc_pre_init)(void); void (*timer_init)(void); -}; +} __no_const; /** * struct x86_init_ops - functions for platform specific setup @@ -101,7 +101,7 @@ struct x86_init_ops { struct x86_init_oem oem; struct x86_init_paging paging; struct x86_init_timers timers; -}; +} __no_const; /** * struct x86_cpuinit_ops - platform specific cpu hotplug setups @@ -109,7 +109,7 @@ struct x86_init_ops { */ struct x86_cpuinit_ops { void (*setup_percpu_clockev)(void); -}; +} __no_const; /** * struct x86_platform_ops - platform specific runtime functions @@ -121,7 +121,7 @@ struct x86_platform_ops { unsigned long (*calibrate_tsc)(void); unsigned long (*get_wallclock)(void); int (*set_wallclock)(unsigned long nowtime); -}; +} __no_const; extern struct x86_init_ops x86_init; extern struct x86_cpuinit_ops x86_cpuinit; diff -urNp linux-2.6.32.48/arch/x86/include/asm/xsave.h linux-2.6.32.48-openvz/arch/x86/include/asm/xsave.h --- linux-2.6.32.48/arch/x86/include/asm/xsave.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/include/asm/xsave.h 2011-11-17 20:09:04.000000000 -0500 @@ -56,6 +56,12 @@ static inline int xrstor_checking(struct static inline int xsave_user(struct xsave_struct __user *buf) { int err; + +#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) + if ((unsigned long)buf < PAX_USER_SHADOW_BASE) + buf = (struct xsave_struct __user *)((void __user*)buf + PAX_USER_SHADOW_BASE); +#endif + __asm__ __volatile__("1: .byte " REX_PREFIX "0x0f,0xae,0x27\n" "2:\n" ".section .fixup,\"ax\"\n" @@ -78,10 +84,15 @@ static inline int xsave_user(struct xsav static inline int xrestore_user(struct xsave_struct __user *buf, u64 mask) { int err; - struct xsave_struct *xstate = ((__force struct xsave_struct *)buf); + struct xsave_struct *xstate = ((__force_kernel struct xsave_struct *)buf); u32 lmask = mask; u32 hmask = mask >> 32; +#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) + if ((unsigned long)xstate < PAX_USER_SHADOW_BASE) + xstate = (struct xsave_struct *)((void *)xstate + PAX_USER_SHADOW_BASE); +#endif + __asm__ __volatile__("1: .byte " REX_PREFIX "0x0f,0xae,0x2f\n" "2:\n" ".section .fixup,\"ax\"\n" diff -urNp linux-2.6.32.48/arch/x86/Kconfig linux-2.6.32.48-openvz/arch/x86/Kconfig --- linux-2.6.32.48/arch/x86/Kconfig 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/Kconfig 2011-11-17 20:09:04.000000000 -0500 @@ -223,7 +223,7 @@ config X86_TRAMPOLINE config X86_32_LAZY_GS def_bool y - depends on X86_32 && !CC_STACKPROTECTOR + depends on X86_32 && !CC_STACKPROTECTOR && !PAX_MEMORY_UDEREF config KTIME_SCALAR def_bool X86_32 @@ -1008,7 +1008,7 @@ choice config NOHIGHMEM bool "off" - depends on !X86_NUMAQ + depends on !X86_NUMAQ && !(PAX_PAGEEXEC && PAX_ENABLE_PAE) ---help--- Linux can use up to 64 Gigabytes of physical memory on x86 systems. However, the address space of 32-bit x86 processors is only 4 @@ -1045,7 +1045,7 @@ config NOHIGHMEM config HIGHMEM4G bool "4GB" - depends on !X86_NUMAQ + depends on !X86_NUMAQ && !(PAX_PAGEEXEC && PAX_ENABLE_PAE) ---help--- Select this if you have a 32-bit processor and between 1 and 4 gigabytes of physical RAM. @@ -1099,7 +1099,7 @@ config PAGE_OFFSET hex default 0xB0000000 if VMSPLIT_3G_OPT default 0x80000000 if VMSPLIT_2G - default 0x78000000 if VMSPLIT_2G_OPT + default 0x70000000 if VMSPLIT_2G_OPT default 0x40000000 if VMSPLIT_1G default 0xC0000000 depends on X86_32 @@ -1460,6 +1460,7 @@ config SECCOMP config CC_STACKPROTECTOR bool "Enable -fstack-protector buffer overflow detection (EXPERIMENTAL)" + depends on X86_64 || !PAX_MEMORY_UDEREF ---help--- This option turns on the -fstack-protector GCC feature. This feature puts, at the beginning of functions, a canary value on @@ -1517,6 +1518,7 @@ config KEXEC_JUMP config PHYSICAL_START hex "Physical address where the kernel is loaded" if (EMBEDDED || CRASH_DUMP) default "0x1000000" + range 0x400000 0x40000000 ---help--- This gives the physical address where the kernel is loaded. @@ -1581,6 +1583,7 @@ config PHYSICAL_ALIGN hex prompt "Alignment value to which kernel should be aligned" if X86_32 default "0x1000000" + range 0x400000 0x1000000 if PAX_KERNEXEC range 0x2000 0x1000000 ---help--- This value puts the alignment restrictions on physical address @@ -1612,9 +1615,10 @@ config HOTPLUG_CPU Say N if you want to disable CPU hotplug. config COMPAT_VDSO - def_bool y + def_bool n prompt "Compat VDSO support" depends on X86_32 || IA32_EMULATION + depends on !PAX_NOEXEC && !PAX_MEMORY_UDEREF ---help--- Map the 32-bit VDSO to the predictable old-style address too. ---help--- @@ -2090,6 +2094,8 @@ config HAVE_ATOMIC_IOMAP def_bool y depends on X86_32 +source "kernel/Kconfig.openvz" + source "net/Kconfig" source "drivers/Kconfig" @@ -2107,3 +2113,5 @@ source "crypto/Kconfig" source "arch/x86/kvm/Kconfig" source "lib/Kconfig" + +source "kernel/bc/Kconfig" diff -urNp linux-2.6.32.48/arch/x86/Kconfig.cpu linux-2.6.32.48-openvz/arch/x86/Kconfig.cpu --- linux-2.6.32.48/arch/x86/Kconfig.cpu 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/Kconfig.cpu 2011-11-17 20:08:53.000000000 -0500 @@ -340,7 +340,7 @@ config X86_PPRO_FENCE config X86_F00F_BUG def_bool y - depends on M586MMX || M586TSC || M586 || M486 || M386 + depends on (M586MMX || M586TSC || M586 || M486 || M386) && !PAX_KERNEXEC config X86_WP_WORKS_OK def_bool y @@ -360,7 +360,7 @@ config X86_POPAD_OK config X86_ALIGNMENT_16 def_bool y - depends on MWINCHIP3D || MWINCHIPC6 || MCYRIXIII || X86_ELAN || MK6 || M586MMX || M586TSC || M586 || M486 || MVIAC3_2 || MGEODEGX1 + depends on MWINCHIP3D || MWINCHIPC6 || MCYRIXIII || X86_ELAN || MK8 || MK7 || MK6 || MCORE2 || MPENTIUM4 || MPENTIUMIII || MPENTIUMII || M686 || M586MMX || M586TSC || M586 || M486 || MVIAC3_2 || MGEODEGX1 config X86_INTEL_USERCOPY def_bool y @@ -406,7 +406,7 @@ config X86_CMPXCHG64 # generates cmov. config X86_CMOV def_bool y - depends on (MK8 || MK7 || MCORE2 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || MVIAC3_2 || MVIAC7 || MCRUSOE || MEFFICEON || X86_64 || MATOM) + depends on (MK8 || MK7 || MCORE2 || MPSC || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || MVIAC3_2 || MVIAC7 || MCRUSOE || MEFFICEON || X86_64 || MATOM) config X86_MINIMUM_CPU_FAMILY int diff -urNp linux-2.6.32.48/arch/x86/Kconfig.debug linux-2.6.32.48-openvz/arch/x86/Kconfig.debug --- linux-2.6.32.48/arch/x86/Kconfig.debug 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/Kconfig.debug 2011-11-17 20:08:53.000000000 -0500 @@ -99,7 +99,7 @@ config X86_PTDUMP config DEBUG_RODATA bool "Write protect kernel read-only data structures" default y - depends on DEBUG_KERNEL + depends on DEBUG_KERNEL && BROKEN ---help--- Mark the kernel read-only data as write-protected in the pagetables, in order to catch accidental (and incorrect) writes to such const diff -urNp linux-2.6.32.48/arch/x86/kernel/acpi/realmode/Makefile linux-2.6.32.48-openvz/arch/x86/kernel/acpi/realmode/Makefile --- linux-2.6.32.48/arch/x86/kernel/acpi/realmode/Makefile 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/acpi/realmode/Makefile 2011-11-17 20:09:04.000000000 -0500 @@ -41,6 +41,9 @@ KBUILD_CFLAGS := $(LINUXINCLUDE) -g -Os $(call cc-option, -fno-stack-protector) \ $(call cc-option, -mpreferred-stack-boundary=2) KBUILD_CFLAGS += $(call cc-option, -m32) +ifdef CONSTIFY_PLUGIN +KBUILD_CFLAGS += $(CONSTIFY_PLUGIN) -fplugin-arg-constify_plugin-no-constify +endif KBUILD_AFLAGS := $(KBUILD_CFLAGS) -D__ASSEMBLY__ GCOV_PROFILE := n diff -urNp linux-2.6.32.48/arch/x86/kernel/acpi/realmode/wakeup.S linux-2.6.32.48-openvz/arch/x86/kernel/acpi/realmode/wakeup.S --- linux-2.6.32.48/arch/x86/kernel/acpi/realmode/wakeup.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/acpi/realmode/wakeup.S 2011-11-17 20:09:04.000000000 -0500 @@ -91,6 +91,9 @@ _start: /* Do any other stuff... */ #ifndef CONFIG_64BIT + /* Recheck NX bit overrides (64bit path does this in trampoline) */ + call verify_cpu + /* This could also be done in C code... */ movl pmode_cr3, %eax movl %eax, %cr3 @@ -104,7 +107,7 @@ _start: movl %eax, %ecx orl %edx, %ecx jz 1f - movl $0xc0000080, %ecx + mov $MSR_EFER, %ecx wrmsr 1: @@ -114,6 +117,7 @@ _start: movl pmode_cr0, %eax movl %eax, %cr0 jmp pmode_return +# include "../../verify_cpu.S" #else pushw $0 pushw trampoline_segment diff -urNp linux-2.6.32.48/arch/x86/kernel/acpi/sleep.c linux-2.6.32.48-openvz/arch/x86/kernel/acpi/sleep.c --- linux-2.6.32.48/arch/x86/kernel/acpi/sleep.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/acpi/sleep.c 2011-11-17 20:09:04.000000000 -0500 @@ -11,11 +11,12 @@ #include #include #include +#include #include "realmode/wakeup.h" #include "sleep.h" -unsigned long acpi_wakeup_address; +unsigned long acpi_wakeup_address = 0x2000; unsigned long acpi_realmode_flags; /* address in low memory of the wakeup routine. */ @@ -98,9 +99,13 @@ int acpi_save_state_mem(void) #else /* CONFIG_64BIT */ header->trampoline_segment = setup_trampoline() >> 4; #ifdef CONFIG_SMP - stack_start.sp = temp_stack + sizeof(temp_stack); + stack_start = (unsigned long)temp_stack + sizeof(temp_stack); + + pax_open_kernel(); early_gdt_descr.address = (unsigned long)get_cpu_gdt_table(smp_processor_id()); + pax_close_kernel(); + initial_gs = per_cpu_offset(smp_processor_id()); #endif initial_code = (unsigned long)wakeup_long64; @@ -134,14 +139,8 @@ void __init acpi_reserve_bootmem(void) return; } - acpi_realmode = (unsigned long)alloc_bootmem_low(WAKEUP_SIZE); - - if (!acpi_realmode) { - printk(KERN_ERR "ACPI: Cannot allocate lowmem, S3 disabled.\n"); - return; - } - - acpi_wakeup_address = virt_to_phys((void *)acpi_realmode); + reserve_early(acpi_wakeup_address, acpi_wakeup_address + WAKEUP_SIZE, "ACPI Wakeup Code"); + acpi_realmode = (unsigned long)__va(acpi_wakeup_address);; } diff -urNp linux-2.6.32.48/arch/x86/kernel/acpi/wakeup_32.S linux-2.6.32.48-openvz/arch/x86/kernel/acpi/wakeup_32.S --- linux-2.6.32.48/arch/x86/kernel/acpi/wakeup_32.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/acpi/wakeup_32.S 2011-11-17 20:08:53.000000000 -0500 @@ -30,13 +30,11 @@ wakeup_pmode_return: # and restore the stack ... but you need gdt for this to work movl saved_context_esp, %esp - movl %cs:saved_magic, %eax - cmpl $0x12345678, %eax + cmpl $0x12345678, saved_magic jne bogus_magic # jump to place where we left off - movl saved_eip, %eax - jmp *%eax + jmp *(saved_eip) bogus_magic: jmp bogus_magic diff -urNp linux-2.6.32.48/arch/x86/kernel/alternative.c linux-2.6.32.48-openvz/arch/x86/kernel/alternative.c --- linux-2.6.32.48/arch/x86/kernel/alternative.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/alternative.c 2011-11-17 20:08:53.000000000 -0500 @@ -407,7 +407,7 @@ void __init_or_module apply_paravirt(str BUG_ON(p->len > MAX_PATCH_LEN); /* prep the buffer with the original instructions */ - memcpy(insnbuf, p->instr, p->len); + memcpy(insnbuf, ktla_ktva(p->instr), p->len); used = pv_init_ops.patch(p->instrtype, p->clobbers, insnbuf, (unsigned long)p->instr, p->len); @@ -475,7 +475,7 @@ void __init alternative_instructions(voi if (smp_alt_once) free_init_pages("SMP alternatives", (unsigned long)__smp_locks, - (unsigned long)__smp_locks_end); + PAGE_ALIGN((unsigned long)__smp_locks_end)); restart_nmi(); } @@ -492,13 +492,17 @@ void __init alternative_instructions(voi * instructions. And on the local CPU you need to be protected again NMI or MCE * handlers seeing an inconsistent instruction while you patch. */ -static void *__init_or_module text_poke_early(void *addr, const void *opcode, +static void *__kprobes text_poke_early(void *addr, const void *opcode, size_t len) { unsigned long flags; local_irq_save(flags); - memcpy(addr, opcode, len); + + pax_open_kernel(); + memcpy(ktla_ktva(addr), opcode, len); sync_core(); + pax_close_kernel(); + local_irq_restore(flags); /* Could also do a CLFLUSH here to speed up CPU recovery; but that causes hangs on some VIA CPUs. */ @@ -520,35 +524,21 @@ static void *__init_or_module text_poke_ */ void *__kprobes text_poke(void *addr, const void *opcode, size_t len) { - unsigned long flags; - char *vaddr; + unsigned char *vaddr = ktla_ktva(addr); struct page *pages[2]; - int i; + size_t i; if (!core_kernel_text((unsigned long)addr)) { - pages[0] = vmalloc_to_page(addr); - pages[1] = vmalloc_to_page(addr + PAGE_SIZE); + pages[0] = vmalloc_to_page(vaddr); + pages[1] = vmalloc_to_page(vaddr + PAGE_SIZE); } else { - pages[0] = virt_to_page(addr); + pages[0] = virt_to_page(vaddr); WARN_ON(!PageReserved(pages[0])); - pages[1] = virt_to_page(addr + PAGE_SIZE); + pages[1] = virt_to_page(vaddr + PAGE_SIZE); } BUG_ON(!pages[0]); - local_irq_save(flags); - set_fixmap(FIX_TEXT_POKE0, page_to_phys(pages[0])); - if (pages[1]) - set_fixmap(FIX_TEXT_POKE1, page_to_phys(pages[1])); - vaddr = (char *)fix_to_virt(FIX_TEXT_POKE0); - memcpy(&vaddr[(unsigned long)addr & ~PAGE_MASK], opcode, len); - clear_fixmap(FIX_TEXT_POKE0); - if (pages[1]) - clear_fixmap(FIX_TEXT_POKE1); - local_flush_tlb(); - sync_core(); - /* Could also do a CLFLUSH here to speed up CPU recovery; but - that causes hangs on some VIA CPUs. */ + text_poke_early(addr, opcode, len); for (i = 0; i < len; i++) - BUG_ON(((char *)addr)[i] != ((char *)opcode)[i]); - local_irq_restore(flags); + BUG_ON((vaddr)[i] != ((const unsigned char *)opcode)[i]); return addr; } diff -urNp linux-2.6.32.48/arch/x86/kernel/amd_iommu.c linux-2.6.32.48-openvz/arch/x86/kernel/amd_iommu.c --- linux-2.6.32.48/arch/x86/kernel/amd_iommu.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/amd_iommu.c 2011-11-17 20:09:04.000000000 -0500 @@ -2076,7 +2076,7 @@ static void prealloc_protection_domains( } } -static struct dma_map_ops amd_iommu_dma_ops = { +static const struct dma_map_ops amd_iommu_dma_ops = { .alloc_coherent = alloc_coherent, .free_coherent = free_coherent, .map_page = map_page, diff -urNp linux-2.6.32.48/arch/x86/kernel/apic/apic.c linux-2.6.32.48-openvz/arch/x86/kernel/apic/apic.c --- linux-2.6.32.48/arch/x86/kernel/apic/apic.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/apic/apic.c 2011-11-17 20:09:04.000000000 -0500 @@ -170,7 +170,7 @@ int first_system_vector = 0xfe; /* * Debug level, exported for io_apic.c */ -unsigned int apic_verbosity; +int apic_verbosity; int pic_mode; @@ -1794,7 +1794,7 @@ void smp_error_interrupt(struct pt_regs apic_write(APIC_ESR, 0); v1 = apic_read(APIC_ESR); ack_APIC_irq(); - atomic_inc(&irq_err_count); + atomic_inc_unchecked(&irq_err_count); /* * Here is what the APIC error bits mean: @@ -2184,6 +2184,8 @@ static int __cpuinit apic_cluster_num(vo u16 *bios_cpu_apicid; DECLARE_BITMAP(clustermap, NUM_APIC_CLUSTERS); + pax_track_stack(); + bios_cpu_apicid = early_per_cpu_ptr(x86_bios_cpu_apicid); bitmap_zero(clustermap, NUM_APIC_CLUSTERS); diff -urNp linux-2.6.32.48/arch/x86/kernel/apic/io_apic.c linux-2.6.32.48-openvz/arch/x86/kernel/apic/io_apic.c --- linux-2.6.32.48/arch/x86/kernel/apic/io_apic.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/apic/io_apic.c 2011-11-17 20:08:53.000000000 -0500 @@ -716,7 +716,7 @@ struct IO_APIC_route_entry **alloc_ioapi ioapic_entries = kzalloc(sizeof(*ioapic_entries) * nr_ioapics, GFP_ATOMIC); if (!ioapic_entries) - return 0; + return NULL; for (apic = 0; apic < nr_ioapics; apic++) { ioapic_entries[apic] = @@ -733,7 +733,7 @@ nomem: kfree(ioapic_entries[apic]); kfree(ioapic_entries); - return 0; + return NULL; } /* @@ -1150,7 +1150,7 @@ int IO_APIC_get_PCI_irq_vector(int bus, } EXPORT_SYMBOL(IO_APIC_get_PCI_irq_vector); -void lock_vector_lock(void) +void lock_vector_lock(void) __acquires(vector_lock) { /* Used to the online set of cpus does not change * during assign_irq_vector. @@ -1158,7 +1158,7 @@ void lock_vector_lock(void) spin_lock(&vector_lock); } -void unlock_vector_lock(void) +void unlock_vector_lock(void) __releases(vector_lock) { spin_unlock(&vector_lock); } @@ -2542,7 +2542,7 @@ static void ack_apic_edge(unsigned int i ack_APIC_irq(); } -atomic_t irq_mis_count; +atomic_unchecked_t irq_mis_count; static void ack_apic_level(unsigned int irq) { @@ -2626,7 +2626,7 @@ static void ack_apic_level(unsigned int /* Tail end of version 0x11 I/O APIC bug workaround */ if (!(v & (1 << (i & 0x1f)))) { - atomic_inc(&irq_mis_count); + atomic_inc_unchecked(&irq_mis_count); spin_lock(&ioapic_lock); __mask_and_edge_IO_APIC_irq(cfg); __unmask_and_level_IO_APIC_irq(cfg); diff -urNp linux-2.6.32.48/arch/x86/kernel/apic/nmi.c linux-2.6.32.48-openvz/arch/x86/kernel/apic/nmi.c --- linux-2.6.32.48/arch/x86/kernel/apic/nmi.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/apic/nmi.c 2011-11-17 20:08:53.000000000 -0500 @@ -435,10 +435,10 @@ nmi_watchdog_tick(struct pt_regs *regs, if (!touched && __get_cpu_var(last_irq_sum) == sum) { /* * Ayiee, looks like this CPU is stuck ... - * wait a few IRQs (5 seconds) before doing the oops ... + * wait a few IRQs (30 seconds) before doing the oops ... */ local_inc(&__get_cpu_var(alert_counter)); - if (local_read(&__get_cpu_var(alert_counter)) == 5 * nmi_hz) + if (local_read(&__get_cpu_var(alert_counter)) == 30 * nmi_hz) /* * die_nmi will return ONLY if NOTIFY_STOP happens.. */ diff -urNp linux-2.6.32.48/arch/x86/kernel/apm_32.c linux-2.6.32.48-openvz/arch/x86/kernel/apm_32.c --- linux-2.6.32.48/arch/x86/kernel/apm_32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/apm_32.c 2011-11-17 20:08:53.000000000 -0500 @@ -410,7 +410,7 @@ static DEFINE_SPINLOCK(user_list_lock); * This is for buggy BIOS's that refer to (real mode) segment 0x40 * even though they are called in protected mode. */ -static struct desc_struct bad_bios_desc = GDT_ENTRY_INIT(0x4092, +static const struct desc_struct bad_bios_desc = GDT_ENTRY_INIT(0x4093, (unsigned long)__va(0x400UL), PAGE_SIZE - 0x400 - 1); static const char driver_version[] = "1.16ac"; /* no spaces */ @@ -588,7 +588,10 @@ static long __apm_bios_call(void *_call) BUG_ON(cpu != 0); gdt = get_cpu_gdt_table(cpu); save_desc_40 = gdt[0x40 / 8]; + + pax_open_kernel(); gdt[0x40 / 8] = bad_bios_desc; + pax_close_kernel(); apm_irq_save(flags); APM_DO_SAVE_SEGS; @@ -597,7 +600,11 @@ static long __apm_bios_call(void *_call) &call->esi); APM_DO_RESTORE_SEGS; apm_irq_restore(flags); + + pax_open_kernel(); gdt[0x40 / 8] = save_desc_40; + pax_close_kernel(); + put_cpu(); return call->eax & 0xff; @@ -664,7 +671,10 @@ static long __apm_bios_call_simple(void BUG_ON(cpu != 0); gdt = get_cpu_gdt_table(cpu); save_desc_40 = gdt[0x40 / 8]; + + pax_open_kernel(); gdt[0x40 / 8] = bad_bios_desc; + pax_close_kernel(); apm_irq_save(flags); APM_DO_SAVE_SEGS; @@ -672,7 +682,11 @@ static long __apm_bios_call_simple(void &call->eax); APM_DO_RESTORE_SEGS; apm_irq_restore(flags); + + pax_open_kernel(); gdt[0x40 / 8] = save_desc_40; + pax_close_kernel(); + put_cpu(); return error; } @@ -975,7 +989,7 @@ recalc: static void apm_power_off(void) { - unsigned char po_bios_call[] = { + const unsigned char po_bios_call[] = { 0xb8, 0x00, 0x10, /* movw $0x1000,ax */ 0x8e, 0xd0, /* movw ax,ss */ 0xbc, 0x00, 0xf0, /* movw $0xf000,sp */ @@ -2357,12 +2371,15 @@ static int __init apm_init(void) * code to that CPU. */ gdt = get_cpu_gdt_table(0); + + pax_open_kernel(); set_desc_base(&gdt[APM_CS >> 3], (unsigned long)__va((unsigned long)apm_info.bios.cseg << 4)); set_desc_base(&gdt[APM_CS_16 >> 3], (unsigned long)__va((unsigned long)apm_info.bios.cseg_16 << 4)); set_desc_base(&gdt[APM_DS >> 3], (unsigned long)__va((unsigned long)apm_info.bios.dseg << 4)); + pax_close_kernel(); proc_create("apm", 0, NULL, &apm_file_ops); diff -urNp linux-2.6.32.48/arch/x86/kernel/asm-offsets_32.c linux-2.6.32.48-openvz/arch/x86/kernel/asm-offsets_32.c --- linux-2.6.32.48/arch/x86/kernel/asm-offsets_32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/asm-offsets_32.c 2011-11-17 20:08:53.000000000 -0500 @@ -51,7 +51,6 @@ void foo(void) OFFSET(CPUINFO_x86_vendor_id, cpuinfo_x86, x86_vendor_id); BLANK(); - OFFSET(TI_task, thread_info, task); OFFSET(TI_exec_domain, thread_info, exec_domain); OFFSET(TI_flags, thread_info, flags); OFFSET(TI_status, thread_info, status); @@ -60,6 +59,8 @@ void foo(void) OFFSET(TI_restart_block, thread_info, restart_block); OFFSET(TI_sysenter_return, thread_info, sysenter_return); OFFSET(TI_cpu, thread_info, cpu); + OFFSET(TI_lowest_stack, thread_info, lowest_stack); + DEFINE(TI_task_thread_sp0, offsetof(struct task_struct, thread.sp0) - offsetof(struct task_struct, tinfo)); BLANK(); OFFSET(GDS_size, desc_ptr, size); @@ -99,6 +100,7 @@ void foo(void) DEFINE(PAGE_SIZE_asm, PAGE_SIZE); DEFINE(PAGE_SHIFT_asm, PAGE_SHIFT); + DEFINE(THREAD_SIZE_asm, THREAD_SIZE); DEFINE(PTRS_PER_PTE, PTRS_PER_PTE); DEFINE(PTRS_PER_PMD, PTRS_PER_PMD); DEFINE(PTRS_PER_PGD, PTRS_PER_PGD); @@ -115,6 +117,11 @@ void foo(void) OFFSET(PV_CPU_iret, pv_cpu_ops, iret); OFFSET(PV_CPU_irq_enable_sysexit, pv_cpu_ops, irq_enable_sysexit); OFFSET(PV_CPU_read_cr0, pv_cpu_ops, read_cr0); + +#ifdef CONFIG_PAX_KERNEXEC + OFFSET(PV_CPU_write_cr0, pv_cpu_ops, write_cr0); +#endif + #endif #ifdef CONFIG_XEN diff -urNp linux-2.6.32.48/arch/x86/kernel/asm-offsets_64.c linux-2.6.32.48-openvz/arch/x86/kernel/asm-offsets_64.c --- linux-2.6.32.48/arch/x86/kernel/asm-offsets_64.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/asm-offsets_64.c 2011-11-17 20:09:04.000000000 -0500 @@ -44,6 +44,8 @@ int main(void) ENTRY(addr_limit); ENTRY(preempt_count); ENTRY(status); + ENTRY(lowest_stack); + DEFINE(TI_task_thread_sp0, offsetof(struct task_struct, thread.sp0) - offsetof(struct task_struct, tinfo)); #ifdef CONFIG_IA32_EMULATION ENTRY(sysenter_return); #endif @@ -63,6 +65,18 @@ int main(void) OFFSET(PV_CPU_irq_enable_sysexit, pv_cpu_ops, irq_enable_sysexit); OFFSET(PV_CPU_swapgs, pv_cpu_ops, swapgs); OFFSET(PV_MMU_read_cr2, pv_mmu_ops, read_cr2); + +#ifdef CONFIG_PAX_KERNEXEC + OFFSET(PV_CPU_read_cr0, pv_cpu_ops, read_cr0); + OFFSET(PV_CPU_write_cr0, pv_cpu_ops, write_cr0); +#endif + +#ifdef CONFIG_PAX_MEMORY_UDEREF + OFFSET(PV_MMU_read_cr3, pv_mmu_ops, read_cr3); + OFFSET(PV_MMU_write_cr3, pv_mmu_ops, write_cr3); + OFFSET(PV_MMU_set_pgd_batched, pv_mmu_ops, set_pgd_batched); +#endif + #endif @@ -115,6 +129,7 @@ int main(void) ENTRY(cr8); BLANK(); #undef ENTRY + DEFINE(TSS_size, sizeof(struct tss_struct)); DEFINE(TSS_ist, offsetof(struct tss_struct, x86_tss.ist)); BLANK(); DEFINE(crypto_tfm_ctx_offset, offsetof(struct crypto_tfm, __crt_ctx)); @@ -130,6 +145,7 @@ int main(void) BLANK(); DEFINE(PAGE_SIZE_asm, PAGE_SIZE); + DEFINE(THREAD_SIZE_asm, THREAD_SIZE); #ifdef CONFIG_XEN BLANK(); OFFSET(XEN_vcpu_info_mask, vcpu_info, evtchn_upcall_mask); diff -urNp linux-2.6.32.48/arch/x86/kernel/cpu/amd.c linux-2.6.32.48-openvz/arch/x86/kernel/cpu/amd.c --- linux-2.6.32.48/arch/x86/kernel/cpu/amd.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/cpu/amd.c 2011-11-17 20:09:04.000000000 -0500 @@ -602,7 +602,7 @@ static unsigned int __cpuinit amd_size_c unsigned int size) { /* AMD errata T13 (order #21922) */ - if ((c->x86 == 6)) { + if (c->x86 == 6) { /* Duron Rev A0 */ if (c->x86_model == 3 && c->x86_mask == 0) size = 64; diff -urNp linux-2.6.32.48/arch/x86/kernel/cpu/common.c linux-2.6.32.48-openvz/arch/x86/kernel/cpu/common.c --- linux-2.6.32.48/arch/x86/kernel/cpu/common.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/cpu/common.c 2011-11-17 20:08:53.000000000 -0500 @@ -83,60 +83,6 @@ static const struct cpu_dev __cpuinitcon static const struct cpu_dev *this_cpu __cpuinitdata = &default_cpu; -DEFINE_PER_CPU_PAGE_ALIGNED(struct gdt_page, gdt_page) = { .gdt = { -#ifdef CONFIG_X86_64 - /* - * We need valid kernel segments for data and code in long mode too - * IRET will check the segment types kkeil 2000/10/28 - * Also sysret mandates a special GDT layout - * - * TLS descriptors are currently at a different place compared to i386. - * Hopefully nobody expects them at a fixed place (Wine?) - */ - [GDT_ENTRY_KERNEL32_CS] = GDT_ENTRY_INIT(0xc09b, 0, 0xfffff), - [GDT_ENTRY_KERNEL_CS] = GDT_ENTRY_INIT(0xa09b, 0, 0xfffff), - [GDT_ENTRY_KERNEL_DS] = GDT_ENTRY_INIT(0xc093, 0, 0xfffff), - [GDT_ENTRY_DEFAULT_USER32_CS] = GDT_ENTRY_INIT(0xc0fb, 0, 0xfffff), - [GDT_ENTRY_DEFAULT_USER_DS] = GDT_ENTRY_INIT(0xc0f3, 0, 0xfffff), - [GDT_ENTRY_DEFAULT_USER_CS] = GDT_ENTRY_INIT(0xa0fb, 0, 0xfffff), -#else - [GDT_ENTRY_KERNEL_CS] = GDT_ENTRY_INIT(0xc09a, 0, 0xfffff), - [GDT_ENTRY_KERNEL_DS] = GDT_ENTRY_INIT(0xc092, 0, 0xfffff), - [GDT_ENTRY_DEFAULT_USER_CS] = GDT_ENTRY_INIT(0xc0fa, 0, 0xfffff), - [GDT_ENTRY_DEFAULT_USER_DS] = GDT_ENTRY_INIT(0xc0f2, 0, 0xfffff), - /* - * Segments used for calling PnP BIOS have byte granularity. - * They code segments and data segments have fixed 64k limits, - * the transfer segment sizes are set at run time. - */ - /* 32-bit code */ - [GDT_ENTRY_PNPBIOS_CS32] = GDT_ENTRY_INIT(0x409a, 0, 0xffff), - /* 16-bit code */ - [GDT_ENTRY_PNPBIOS_CS16] = GDT_ENTRY_INIT(0x009a, 0, 0xffff), - /* 16-bit data */ - [GDT_ENTRY_PNPBIOS_DS] = GDT_ENTRY_INIT(0x0092, 0, 0xffff), - /* 16-bit data */ - [GDT_ENTRY_PNPBIOS_TS1] = GDT_ENTRY_INIT(0x0092, 0, 0), - /* 16-bit data */ - [GDT_ENTRY_PNPBIOS_TS2] = GDT_ENTRY_INIT(0x0092, 0, 0), - /* - * The APM segments have byte granularity and their bases - * are set at run time. All have 64k limits. - */ - /* 32-bit code */ - [GDT_ENTRY_APMBIOS_BASE] = GDT_ENTRY_INIT(0x409a, 0, 0xffff), - /* 16-bit code */ - [GDT_ENTRY_APMBIOS_BASE+1] = GDT_ENTRY_INIT(0x009a, 0, 0xffff), - /* data */ - [GDT_ENTRY_APMBIOS_BASE+2] = GDT_ENTRY_INIT(0x4092, 0, 0xffff), - - [GDT_ENTRY_ESPFIX_SS] = GDT_ENTRY_INIT(0xc092, 0, 0xfffff), - [GDT_ENTRY_PERCPU] = GDT_ENTRY_INIT(0xc092, 0, 0xfffff), - GDT_STACK_CANARY_INIT -#endif -} }; -EXPORT_PER_CPU_SYMBOL_GPL(gdt_page); - static int __init x86_xsave_setup(char *s) { setup_clear_cpu_cap(X86_FEATURE_XSAVE); @@ -344,7 +290,7 @@ void switch_to_new_gdt(int cpu) { struct desc_ptr gdt_descr; - gdt_descr.address = (long)get_cpu_gdt_table(cpu); + gdt_descr.address = (unsigned long)get_cpu_gdt_table(cpu); gdt_descr.size = GDT_SIZE - 1; load_gdt(&gdt_descr); /* Reload the per-cpu base */ @@ -798,6 +744,10 @@ static void __cpuinit identify_cpu(struc /* Filter out anything that depends on CPUID levels we don't have */ filter_cpuid_features(c, true); +#if defined(CONFIG_PAX_SEGMEXEC) || defined(CONFIG_PAX_KERNEXEC) || (defined(CONFIG_PAX_MEMORY_UDEREF) && defined(CONFIG_X86_32)) + setup_clear_cpu_cap(X86_FEATURE_SEP); +#endif + /* If the model name is still unset, do table lookup. */ if (!c->x86_model_id[0]) { const char *p; @@ -980,6 +930,9 @@ static __init int setup_disablecpuid(cha } __setup("clearcpuid=", setup_disablecpuid); +DEFINE_PER_CPU(struct thread_info *, current_tinfo) = &init_task.tinfo; +EXPORT_PER_CPU_SYMBOL(current_tinfo); + #ifdef CONFIG_X86_64 struct desc_ptr idt_descr = { NR_VECTORS * 16 - 1, (unsigned long) idt_table }; @@ -995,7 +948,7 @@ DEFINE_PER_CPU(struct task_struct *, cur EXPORT_PER_CPU_SYMBOL(current_task); DEFINE_PER_CPU(unsigned long, kernel_stack) = - (unsigned long)&init_thread_union - KERNEL_STACK_OFFSET + THREAD_SIZE; + (unsigned long)&init_thread_union - 16 + THREAD_SIZE; EXPORT_PER_CPU_SYMBOL(kernel_stack); DEFINE_PER_CPU(char *, irq_stack_ptr) = @@ -1060,7 +1013,7 @@ struct pt_regs * __cpuinit idle_regs(str { memset(regs, 0, sizeof(struct pt_regs)); regs->fs = __KERNEL_PERCPU; - regs->gs = __KERNEL_STACK_CANARY; + savesegment(gs, regs->gs); return regs; } @@ -1101,7 +1054,7 @@ void __cpuinit cpu_init(void) int i; cpu = stack_smp_processor_id(); - t = &per_cpu(init_tss, cpu); + t = init_tss + cpu; orig_ist = &per_cpu(orig_ist, cpu); #ifdef CONFIG_NUMA @@ -1127,7 +1080,7 @@ void __cpuinit cpu_init(void) switch_to_new_gdt(cpu); loadsegment(fs, 0); - load_idt((const struct desc_ptr *)&idt_descr); + load_idt(&idt_descr); memset(me->thread.tls_array, 0, GDT_ENTRY_TLS_ENTRIES * 8); syscall_init(); @@ -1136,7 +1089,6 @@ void __cpuinit cpu_init(void) wrmsrl(MSR_KERNEL_GS_BASE, 0); barrier(); - check_efer(); if (cpu != 0) enable_x2apic(); @@ -1199,7 +1151,7 @@ void __cpuinit cpu_init(void) { int cpu = smp_processor_id(); struct task_struct *curr = current; - struct tss_struct *t = &per_cpu(init_tss, cpu); + struct tss_struct *t = init_tss + cpu; struct thread_struct *thread = &curr->thread; if (cpumask_test_and_set_cpu(cpu, cpu_initialized_mask)) { diff -urNp linux-2.6.32.48/arch/x86/kernel/cpu/intel.c linux-2.6.32.48-openvz/arch/x86/kernel/cpu/intel.c --- linux-2.6.32.48/arch/x86/kernel/cpu/intel.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/cpu/intel.c 2011-11-17 20:08:53.000000000 -0500 @@ -162,7 +162,7 @@ static void __cpuinit trap_init_f00f_bug * Update the IDT descriptor and reload the IDT so that * it uses the read-only mapped virtual address. */ - idt_descr.address = fix_to_virt(FIX_F00F_IDT); + idt_descr.address = (struct desc_struct *)fix_to_virt(FIX_F00F_IDT); load_idt(&idt_descr); } #endif diff -urNp linux-2.6.32.48/arch/x86/kernel/cpu/intel_cacheinfo.c linux-2.6.32.48-openvz/arch/x86/kernel/cpu/intel_cacheinfo.c --- linux-2.6.32.48/arch/x86/kernel/cpu/intel_cacheinfo.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/cpu/intel_cacheinfo.c 2011-11-17 20:08:53.000000000 -0500 @@ -921,7 +921,7 @@ static ssize_t store(struct kobject *kob return ret; } -static struct sysfs_ops sysfs_ops = { +static const struct sysfs_ops sysfs_ops = { .show = show, .store = store, }; diff -urNp linux-2.6.32.48/arch/x86/kernel/cpu/Makefile linux-2.6.32.48-openvz/arch/x86/kernel/cpu/Makefile --- linux-2.6.32.48/arch/x86/kernel/cpu/Makefile 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/cpu/Makefile 2011-11-17 20:08:53.000000000 -0500 @@ -7,10 +7,6 @@ ifdef CONFIG_FUNCTION_TRACER CFLAGS_REMOVE_common.o = -pg endif -# Make sure load_percpu_segment has no stackprotector -nostackp := $(call cc-option, -fno-stack-protector) -CFLAGS_common.o := $(nostackp) - obj-y := intel_cacheinfo.o addon_cpuid_features.o obj-y += proc.o capflags.o powerflags.o common.o obj-y += vmware.o hypervisor.o sched.o diff -urNp linux-2.6.32.48/arch/x86/kernel/cpu/mcheck/mce_amd.c linux-2.6.32.48-openvz/arch/x86/kernel/cpu/mcheck/mce_amd.c --- linux-2.6.32.48/arch/x86/kernel/cpu/mcheck/mce_amd.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/cpu/mcheck/mce_amd.c 2011-11-17 20:09:04.000000000 -0500 @@ -385,7 +385,7 @@ static ssize_t store(struct kobject *kob return ret; } -static struct sysfs_ops threshold_ops = { +static const struct sysfs_ops threshold_ops = { .show = show, .store = store, }; diff -urNp linux-2.6.32.48/arch/x86/kernel/cpu/mcheck/mce.c linux-2.6.32.48-openvz/arch/x86/kernel/cpu/mcheck/mce.c --- linux-2.6.32.48/arch/x86/kernel/cpu/mcheck/mce.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/cpu/mcheck/mce.c 2011-11-17 20:08:53.000000000 -0500 @@ -43,6 +43,7 @@ #include #include #include +#include #include "mce-internal.h" @@ -187,7 +188,7 @@ static void print_mce(struct mce *m) !(m->mcgstatus & MCG_STATUS_EIPV) ? " !INEXACT!" : "", m->cs, m->ip); - if (m->cs == __KERNEL_CS) + if (m->cs == __KERNEL_CS || m->cs == __KERNEXEC_KERNEL_CS) print_symbol("{%s}", m->ip); pr_cont("\n"); } @@ -221,10 +222,10 @@ static void print_mce_tail(void) #define PANIC_TIMEOUT 5 /* 5 seconds */ -static atomic_t mce_paniced; +static atomic_unchecked_t mce_paniced; static int fake_panic; -static atomic_t mce_fake_paniced; +static atomic_unchecked_t mce_fake_paniced; /* Panic in progress. Enable interrupts and wait for final IPI */ static void wait_for_panic(void) @@ -248,7 +249,7 @@ static void mce_panic(char *msg, struct /* * Make sure only one CPU runs in machine check panic */ - if (atomic_inc_return(&mce_paniced) > 1) + if (atomic_inc_return_unchecked(&mce_paniced) > 1) wait_for_panic(); barrier(); @@ -256,7 +257,7 @@ static void mce_panic(char *msg, struct console_verbose(); } else { /* Don't log too much for fake panic */ - if (atomic_inc_return(&mce_fake_paniced) > 1) + if (atomic_inc_return_unchecked(&mce_fake_paniced) > 1) return; } print_mce_head(); @@ -616,7 +617,7 @@ static int mce_timed_out(u64 *t) * might have been modified by someone else. */ rmb(); - if (atomic_read(&mce_paniced)) + if (atomic_read_unchecked(&mce_paniced)) wait_for_panic(); if (!monarch_timeout) goto out; @@ -1429,14 +1430,14 @@ void __cpuinit mcheck_init(struct cpuinf */ static DEFINE_SPINLOCK(mce_state_lock); -static int open_count; /* #times opened */ +static local_t open_count; /* #times opened */ static int open_exclu; /* already open exclusive? */ static int mce_open(struct inode *inode, struct file *file) { spin_lock(&mce_state_lock); - if (open_exclu || (open_count && (file->f_flags & O_EXCL))) { + if (open_exclu || (local_read(&open_count) && (file->f_flags & O_EXCL))) { spin_unlock(&mce_state_lock); return -EBUSY; @@ -1444,7 +1445,7 @@ static int mce_open(struct inode *inode, if (file->f_flags & O_EXCL) open_exclu = 1; - open_count++; + local_inc(&open_count); spin_unlock(&mce_state_lock); @@ -1455,7 +1456,7 @@ static int mce_release(struct inode *ino { spin_lock(&mce_state_lock); - open_count--; + local_dec(&open_count); open_exclu = 0; spin_unlock(&mce_state_lock); @@ -2082,7 +2083,7 @@ struct dentry *mce_get_debugfs_dir(void) static void mce_reset(void) { cpu_missing = 0; - atomic_set(&mce_fake_paniced, 0); + atomic_set_unchecked(&mce_fake_paniced, 0); atomic_set(&mce_executing, 0); atomic_set(&mce_callin, 0); atomic_set(&global_nwo, 0); diff -urNp linux-2.6.32.48/arch/x86/kernel/cpu/mcheck/mce-inject.c linux-2.6.32.48-openvz/arch/x86/kernel/cpu/mcheck/mce-inject.c --- linux-2.6.32.48/arch/x86/kernel/cpu/mcheck/mce-inject.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/cpu/mcheck/mce-inject.c 2011-11-17 20:09:04.000000000 -0500 @@ -211,7 +211,9 @@ static ssize_t mce_write(struct file *fi static int inject_init(void) { printk(KERN_INFO "Machine check injector initialized\n"); - mce_chrdev_ops.write = mce_write; + pax_open_kernel(); + *(void **)&mce_chrdev_ops.write = mce_write; + pax_close_kernel(); register_die_notifier(&mce_raise_nb); return 0; } diff -urNp linux-2.6.32.48/arch/x86/kernel/cpu/mtrr/amd.c linux-2.6.32.48-openvz/arch/x86/kernel/cpu/mtrr/amd.c --- linux-2.6.32.48/arch/x86/kernel/cpu/mtrr/amd.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/cpu/mtrr/amd.c 2011-11-17 20:08:53.000000000 -0500 @@ -108,7 +108,7 @@ amd_validate_add_page(unsigned long base return 0; } -static struct mtrr_ops amd_mtrr_ops = { +static const struct mtrr_ops amd_mtrr_ops = { .vendor = X86_VENDOR_AMD, .set = amd_set_mtrr, .get = amd_get_mtrr, diff -urNp linux-2.6.32.48/arch/x86/kernel/cpu/mtrr/centaur.c linux-2.6.32.48-openvz/arch/x86/kernel/cpu/mtrr/centaur.c --- linux-2.6.32.48/arch/x86/kernel/cpu/mtrr/centaur.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/cpu/mtrr/centaur.c 2011-11-17 20:08:53.000000000 -0500 @@ -110,7 +110,7 @@ centaur_validate_add_page(unsigned long return 0; } -static struct mtrr_ops centaur_mtrr_ops = { +static const struct mtrr_ops centaur_mtrr_ops = { .vendor = X86_VENDOR_CENTAUR, .set = centaur_set_mcr, .get = centaur_get_mcr, diff -urNp linux-2.6.32.48/arch/x86/kernel/cpu/mtrr/cyrix.c linux-2.6.32.48-openvz/arch/x86/kernel/cpu/mtrr/cyrix.c --- linux-2.6.32.48/arch/x86/kernel/cpu/mtrr/cyrix.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/cpu/mtrr/cyrix.c 2011-11-17 20:08:53.000000000 -0500 @@ -265,7 +265,7 @@ static void cyrix_set_all(void) post_set(); } -static struct mtrr_ops cyrix_mtrr_ops = { +static const struct mtrr_ops cyrix_mtrr_ops = { .vendor = X86_VENDOR_CYRIX, .set_all = cyrix_set_all, .set = cyrix_set_arr, diff -urNp linux-2.6.32.48/arch/x86/kernel/cpu/mtrr/generic.c linux-2.6.32.48-openvz/arch/x86/kernel/cpu/mtrr/generic.c --- linux-2.6.32.48/arch/x86/kernel/cpu/mtrr/generic.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/cpu/mtrr/generic.c 2011-11-17 20:08:53.000000000 -0500 @@ -752,7 +752,7 @@ int positive_have_wrcomb(void) /* * Generic structure... */ -struct mtrr_ops generic_mtrr_ops = { +const struct mtrr_ops generic_mtrr_ops = { .use_intel_if = 1, .set_all = generic_set_all, .get = generic_get_mtrr, diff -urNp linux-2.6.32.48/arch/x86/kernel/cpu/mtrr/main.c linux-2.6.32.48-openvz/arch/x86/kernel/cpu/mtrr/main.c --- linux-2.6.32.48/arch/x86/kernel/cpu/mtrr/main.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/cpu/mtrr/main.c 2011-11-17 20:08:53.000000000 -0500 @@ -60,14 +60,14 @@ static DEFINE_MUTEX(mtrr_mutex); u64 size_or_mask, size_and_mask; static bool mtrr_aps_delayed_init; -static struct mtrr_ops *mtrr_ops[X86_VENDOR_NUM]; +static const struct mtrr_ops *mtrr_ops[X86_VENDOR_NUM] __read_only; -struct mtrr_ops *mtrr_if; +const struct mtrr_ops *mtrr_if; static void set_mtrr(unsigned int reg, unsigned long base, unsigned long size, mtrr_type type); -void set_mtrr_ops(struct mtrr_ops *ops) +void set_mtrr_ops(const struct mtrr_ops *ops) { if (ops->vendor && ops->vendor < X86_VENDOR_NUM) mtrr_ops[ops->vendor] = ops; diff -urNp linux-2.6.32.48/arch/x86/kernel/cpu/mtrr/mtrr.h linux-2.6.32.48-openvz/arch/x86/kernel/cpu/mtrr/mtrr.h --- linux-2.6.32.48/arch/x86/kernel/cpu/mtrr/mtrr.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/cpu/mtrr/mtrr.h 2011-11-17 20:09:04.000000000 -0500 @@ -25,14 +25,14 @@ struct mtrr_ops { int (*validate_add_page)(unsigned long base, unsigned long size, unsigned int type); int (*have_wrcomb)(void); -}; +} __do_const; extern int generic_get_free_region(unsigned long base, unsigned long size, int replace_reg); extern int generic_validate_add_page(unsigned long base, unsigned long size, unsigned int type); -extern struct mtrr_ops generic_mtrr_ops; +extern const struct mtrr_ops generic_mtrr_ops; extern int positive_have_wrcomb(void); @@ -53,10 +53,10 @@ void fill_mtrr_var_range(unsigned int in u32 base_lo, u32 base_hi, u32 mask_lo, u32 mask_hi); void get_mtrr_state(void); -extern void set_mtrr_ops(struct mtrr_ops *ops); +extern void set_mtrr_ops(const struct mtrr_ops *ops); extern u64 size_or_mask, size_and_mask; -extern struct mtrr_ops *mtrr_if; +extern const struct mtrr_ops *mtrr_if; #define is_cpu(vnd) (mtrr_if && mtrr_if->vendor == X86_VENDOR_##vnd) #define use_intel() (mtrr_if && mtrr_if->use_intel_if == 1) diff -urNp linux-2.6.32.48/arch/x86/kernel/cpu/perfctr-watchdog.c linux-2.6.32.48-openvz/arch/x86/kernel/cpu/perfctr-watchdog.c --- linux-2.6.32.48/arch/x86/kernel/cpu/perfctr-watchdog.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/cpu/perfctr-watchdog.c 2011-11-17 20:08:53.000000000 -0500 @@ -30,11 +30,11 @@ struct nmi_watchdog_ctlblk { /* Interface defining a CPU specific perfctr watchdog */ struct wd_ops { - int (*reserve)(void); - void (*unreserve)(void); - int (*setup)(unsigned nmi_hz); - void (*rearm)(struct nmi_watchdog_ctlblk *wd, unsigned nmi_hz); - void (*stop)(void); + int (* const reserve)(void); + void (* const unreserve)(void); + int (* const setup)(unsigned nmi_hz); + void (* const rearm)(struct nmi_watchdog_ctlblk *wd, unsigned nmi_hz); + void (* const stop)(void); unsigned perfctr; unsigned evntsel; u64 checkbit; @@ -645,6 +645,7 @@ static const struct wd_ops p4_wd_ops = { #define ARCH_PERFMON_NMI_EVENT_SEL ARCH_PERFMON_UNHALTED_CORE_CYCLES_SEL #define ARCH_PERFMON_NMI_EVENT_UMASK ARCH_PERFMON_UNHALTED_CORE_CYCLES_UMASK +/* cannot be const */ static struct wd_ops intel_arch_wd_ops; static int setup_intel_arch_watchdog(unsigned nmi_hz) @@ -697,6 +698,7 @@ static int setup_intel_arch_watchdog(uns return 1; } +/* cannot be const */ static struct wd_ops intel_arch_wd_ops __read_mostly = { .reserve = single_msr_reserve, .unreserve = single_msr_unreserve, diff -urNp linux-2.6.32.48/arch/x86/kernel/cpu/perf_event.c linux-2.6.32.48-openvz/arch/x86/kernel/cpu/perf_event.c --- linux-2.6.32.48/arch/x86/kernel/cpu/perf_event.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/cpu/perf_event.c 2011-11-17 20:08:53.000000000 -0500 @@ -723,10 +723,10 @@ x86_perf_event_update(struct perf_event * count to the generic event atomically: */ again: - prev_raw_count = atomic64_read(&hwc->prev_count); + prev_raw_count = atomic64_read_unchecked(&hwc->prev_count); rdmsrl(hwc->event_base + idx, new_raw_count); - if (atomic64_cmpxchg(&hwc->prev_count, prev_raw_count, + if (atomic64_cmpxchg_unchecked(&hwc->prev_count, prev_raw_count, new_raw_count) != prev_raw_count) goto again; @@ -741,7 +741,7 @@ again: delta = (new_raw_count << shift) - (prev_raw_count << shift); delta >>= shift; - atomic64_add(delta, &event->count); + atomic64_add_unchecked(delta, &event->count); atomic64_sub(delta, &hwc->period_left); return new_raw_count; @@ -1353,7 +1353,7 @@ x86_perf_event_set_period(struct perf_ev * The hw event starts counting from this event offset, * mark it to be able to extra future deltas: */ - atomic64_set(&hwc->prev_count, (u64)-left); + atomic64_set_unchecked(&hwc->prev_count, (u64)-left); err = checking_wrmsrl(hwc->event_base + idx, (u64)(-left) & x86_pmu.event_mask); @@ -2357,7 +2357,7 @@ perf_callchain_user(struct pt_regs *regs break; callchain_store(entry, frame.return_address); - fp = frame.next_frame; + fp = (__force const void __user *)frame.next_frame; } } diff -urNp linux-2.6.32.48/arch/x86/kernel/cpu/transmeta.c linux-2.6.32.48-openvz/arch/x86/kernel/cpu/transmeta.c --- linux-2.6.32.48/arch/x86/kernel/cpu/transmeta.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/cpu/transmeta.c 2011-11-17 20:08:53.000000000 -0500 @@ -1,6 +1,7 @@ #include #include #include +#include #include #include #include "cpu.h" diff -urNp linux-2.6.32.48/arch/x86/kernel/crash.c linux-2.6.32.48-openvz/arch/x86/kernel/crash.c --- linux-2.6.32.48/arch/x86/kernel/crash.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/crash.c 2011-11-17 20:08:53.000000000 -0500 @@ -41,7 +41,7 @@ static void kdump_nmi_callback(int cpu, regs = args->regs; #ifdef CONFIG_X86_32 - if (!user_mode_vm(regs)) { + if (!user_mode(regs)) { crash_fixup_ss_esp(&fixed_regs, regs); regs = &fixed_regs; } diff -urNp linux-2.6.32.48/arch/x86/kernel/doublefault_32.c linux-2.6.32.48-openvz/arch/x86/kernel/doublefault_32.c --- linux-2.6.32.48/arch/x86/kernel/doublefault_32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/doublefault_32.c 2011-11-17 20:08:53.000000000 -0500 @@ -11,7 +11,7 @@ #define DOUBLEFAULT_STACKSIZE (1024) static unsigned long doublefault_stack[DOUBLEFAULT_STACKSIZE]; -#define STACK_START (unsigned long)(doublefault_stack+DOUBLEFAULT_STACKSIZE) +#define STACK_START (unsigned long)(doublefault_stack+DOUBLEFAULT_STACKSIZE-2) #define ptr_ok(x) ((x) > PAGE_OFFSET && (x) < PAGE_OFFSET + MAXMEM) @@ -21,7 +21,7 @@ static void doublefault_fn(void) unsigned long gdt, tss; store_gdt(&gdt_desc); - gdt = gdt_desc.address; + gdt = (unsigned long)gdt_desc.address; printk(KERN_EMERG "PANIC: double fault, gdt at %08lx [%d bytes]\n", gdt, gdt_desc.size); @@ -58,10 +58,10 @@ struct tss_struct doublefault_tss __cach /* 0x2 bit is always set */ .flags = X86_EFLAGS_SF | 0x2, .sp = STACK_START, - .es = __USER_DS, + .es = __KERNEL_DS, .cs = __KERNEL_CS, .ss = __KERNEL_DS, - .ds = __USER_DS, + .ds = __KERNEL_DS, .fs = __KERNEL_PERCPU, .__cr3 = __pa_nodebug(swapper_pg_dir), diff -urNp linux-2.6.32.48/arch/x86/kernel/dumpstack_32.c linux-2.6.32.48-openvz/arch/x86/kernel/dumpstack_32.c --- linux-2.6.32.48/arch/x86/kernel/dumpstack_32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/dumpstack_32.c 2011-11-17 20:08:53.000000000 -0500 @@ -53,16 +53,12 @@ void dump_trace(struct task_struct *task #endif for (;;) { - struct thread_info *context; + void *stack_start = (void *)((unsigned long)stack & ~(THREAD_SIZE-1)); + bp = print_context_stack(task, stack_start, stack, bp, ops, data, NULL, &graph); - context = (struct thread_info *) - ((unsigned long)stack & (~(THREAD_SIZE - 1))); - bp = print_context_stack(context, stack, bp, ops, - data, NULL, &graph); - - stack = (unsigned long *)context->previous_esp; - if (!stack) + if (stack_start == task_stack_page(task)) break; + stack = *(unsigned long **)stack_start; if (ops->stack(data, "IRQ") < 0) break; touch_nmi_watchdog(); @@ -105,18 +101,20 @@ void show_registers(struct pt_regs *regs print_modules(); __show_regs(regs, 0); - printk(KERN_EMERG "Process %.*s (pid: %d, ti=%p task=%p task.ti=%p)\n", + printk(KERN_EMERG "Process %.*s (pid: %d, veid: %d, ti=%p task=%p task.ti=%p)\n", TASK_COMM_LEN, current->comm, task_pid_nr(current), + VEID(current->ve_task_info.owner_env), current_thread_info(), current, task_thread_info(current)); /* * When in-kernel, we also print out the stack and code at the * time of the fault.. */ - if (!user_mode_vm(regs)) { + if (!user_mode(regs)) { unsigned int code_prologue = code_bytes * 43 / 64; unsigned int code_len = code_bytes; unsigned char c; u8 *ip; + unsigned long cs_base = get_desc_base(&get_cpu_gdt_table(smp_processor_id())[(0xffff & regs->cs) >> 3]); printk(KERN_EMERG "Stack:\n"); show_stack_log_lvl(NULL, regs, ®s->sp, @@ -124,10 +122,10 @@ void show_registers(struct pt_regs *regs printk(KERN_EMERG "Code: "); - ip = (u8 *)regs->ip - code_prologue; + ip = (u8 *)regs->ip - code_prologue + cs_base; if (ip < (u8 *)PAGE_OFFSET || probe_kernel_address(ip, c)) { /* try starting at IP */ - ip = (u8 *)regs->ip; + ip = (u8 *)regs->ip + cs_base; code_len = code_len - code_prologue + 1; } for (i = 0; i < code_len; i++, ip++) { @@ -136,7 +134,7 @@ void show_registers(struct pt_regs *regs printk(" Bad EIP value."); break; } - if (ip == (u8 *)regs->ip) + if (ip == (u8 *)regs->ip + cs_base) printk("<%02x> ", c); else printk("%02x ", c); @@ -149,6 +147,7 @@ int is_valid_bugaddr(unsigned long ip) { unsigned short ud2; + ip = ktla_ktva(ip); if (ip < PAGE_OFFSET) return 0; if (probe_kernel_address((unsigned short *)ip, ud2)) diff -urNp linux-2.6.32.48/arch/x86/kernel/dumpstack_64.c linux-2.6.32.48-openvz/arch/x86/kernel/dumpstack_64.c --- linux-2.6.32.48/arch/x86/kernel/dumpstack_64.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/dumpstack_64.c 2011-11-17 20:08:53.000000000 -0500 @@ -116,8 +116,8 @@ void dump_trace(struct task_struct *task unsigned long *irq_stack_end = (unsigned long *)per_cpu(irq_stack_ptr, cpu); unsigned used = 0; - struct thread_info *tinfo; int graph = 0; + void *stack_start; if (!task) task = current; @@ -146,10 +146,10 @@ void dump_trace(struct task_struct *task * current stack address. If the stacks consist of nested * exceptions */ - tinfo = task_thread_info(task); for (;;) { char *id; unsigned long *estack_end; + estack_end = in_exception_stack(cpu, (unsigned long)stack, &used, &id); @@ -157,7 +157,7 @@ void dump_trace(struct task_struct *task if (ops->stack(data, id) < 0) break; - bp = print_context_stack(tinfo, stack, bp, ops, + bp = print_context_stack(task, estack_end - EXCEPTION_STKSZ, stack, bp, ops, data, estack_end, &graph); ops->stack(data, ""); /* @@ -176,7 +176,7 @@ void dump_trace(struct task_struct *task if (stack >= irq_stack && stack < irq_stack_end) { if (ops->stack(data, "IRQ") < 0) break; - bp = print_context_stack(tinfo, stack, bp, + bp = print_context_stack(task, irq_stack, stack, bp, ops, data, irq_stack_end, &graph); /* * We link to the next stack (which would be @@ -195,7 +195,8 @@ void dump_trace(struct task_struct *task /* * This handles the process stack: */ - bp = print_context_stack(tinfo, stack, bp, ops, data, NULL, &graph); + stack_start = (void *)((unsigned long)stack & ~(THREAD_SIZE-1)); + bp = print_context_stack(task, stack_start, stack, bp, ops, data, NULL, &graph); put_cpu(); } EXPORT_SYMBOL(dump_trace); @@ -254,8 +255,10 @@ void show_registers(struct pt_regs *regs sp = regs->sp; printk("CPU %d ", cpu); __show_regs(regs, 1); - printk("Process %s (pid: %d, threadinfo %p, task %p)\n", - cur->comm, cur->pid, task_thread_info(cur), cur); + printk("Process %s (pid: %d, veid=%d, threadinfo %p, task %p)\n", + cur->comm, cur->pid, + VEID(VE_TASK_INFO(current)->owner_env), + task_thread_info(cur), cur); /* * When in-kernel, we also print out the stack and code at the diff -urNp linux-2.6.32.48/arch/x86/kernel/dumpstack.c linux-2.6.32.48-openvz/arch/x86/kernel/dumpstack.c --- linux-2.6.32.48/arch/x86/kernel/dumpstack.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/dumpstack.c 2011-11-17 20:08:53.000000000 -0500 @@ -2,6 +2,9 @@ * Copyright (C) 1991, 1992 Linus Torvalds * Copyright (C) 2000, 2001, 2002 Andi Kleen, SuSE Labs */ +#ifdef CONFIG_GRKERNSEC_HIDESYM +#define __INCLUDED_BY_HIDESYM 1 +#endif #include #include #include @@ -28,7 +31,7 @@ static int die_counter; void printk_address(unsigned long address, int reliable) { - printk(" [<%p>] %s%pS\n", (void *) address, + printk(" [<%p>] %s%pA\n", (void *) address, reliable ? "" : "? ", (void *) address); } @@ -36,9 +39,8 @@ void printk_address(unsigned long addres static void print_ftrace_graph_addr(unsigned long addr, void *data, const struct stacktrace_ops *ops, - struct thread_info *tinfo, int *graph) + struct task_struct *task, int *graph) { - struct task_struct *task = tinfo->task; unsigned long ret_addr; int index = task->curr_ret_stack; @@ -59,7 +61,7 @@ print_ftrace_graph_addr(unsigned long ad static inline void print_ftrace_graph_addr(unsigned long addr, void *data, const struct stacktrace_ops *ops, - struct thread_info *tinfo, int *graph) + struct task_struct *task, int *graph) { } #endif @@ -70,10 +72,8 @@ print_ftrace_graph_addr(unsigned long ad * severe exception (double fault, nmi, stack fault, debug, mce) hardware stack */ -static inline int valid_stack_ptr(struct thread_info *tinfo, - void *p, unsigned int size, void *end) +static inline int valid_stack_ptr(void *t, void *p, unsigned int size, void *end) { - void *t = tinfo; if (end) { if (p < end && p >= (end-THREAD_SIZE)) return 1; @@ -84,14 +84,14 @@ static inline int valid_stack_ptr(struct } unsigned long -print_context_stack(struct thread_info *tinfo, +print_context_stack(struct task_struct *task, void *stack_start, unsigned long *stack, unsigned long bp, const struct stacktrace_ops *ops, void *data, unsigned long *end, int *graph) { struct stack_frame *frame = (struct stack_frame *)bp; - while (valid_stack_ptr(tinfo, stack, sizeof(*stack), end)) { + while (valid_stack_ptr(stack_start, stack, sizeof(*stack), end)) { unsigned long addr; addr = *stack; @@ -103,7 +103,7 @@ print_context_stack(struct thread_info * } else { ops->address(data, addr, 0); } - print_ftrace_graph_addr(addr, data, ops, tinfo, graph); + print_ftrace_graph_addr(addr, data, ops, task, graph); } stack++; } @@ -180,7 +180,7 @@ void dump_stack(void) #endif printk("Pid: %d, comm: %.20s %s %s %.*s\n", - current->pid, current->comm, print_tainted(), + task_pid_nr(current), current->comm, print_tainted(), init_utsname()->release, (int)strcspn(init_utsname()->version, " "), init_utsname()->version); @@ -220,6 +220,8 @@ unsigned __kprobes long oops_begin(void) return flags; } +extern void gr_handle_kernel_exploit(void); + void __kprobes oops_end(unsigned long flags, struct pt_regs *regs, int signr) { if (regs && kexec_should_crash(current)) @@ -241,7 +243,10 @@ void __kprobes oops_end(unsigned long fl panic("Fatal exception in interrupt"); if (panic_on_oops) panic("Fatal exception"); - do_exit(signr); + + gr_handle_kernel_exploit(); + + do_group_exit(signr); } int __kprobes __die(const char *str, struct pt_regs *regs, long err) @@ -295,7 +300,7 @@ void die(const char *str, struct pt_regs unsigned long flags = oops_begin(); int sig = SIGSEGV; - if (!user_mode_vm(regs)) + if (!user_mode(regs)) report_bug(regs->ip, regs); if (__die(str, regs, err)) @@ -320,6 +325,7 @@ die_nmi(char *str, struct pt_regs *regs, printk(" on CPU%d, ip %08lx, registers:\n", smp_processor_id(), regs->ip); show_registers(regs); + nmi_show_regs(regs, 1); oops_end(flags, regs, 0); if (do_panic || panic_on_oops) panic("Non maskable interrupt"); diff -urNp linux-2.6.32.48/arch/x86/kernel/dumpstack.h linux-2.6.32.48-openvz/arch/x86/kernel/dumpstack.h --- linux-2.6.32.48/arch/x86/kernel/dumpstack.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/dumpstack.h 2011-11-17 20:08:53.000000000 -0500 @@ -15,7 +15,7 @@ #endif extern unsigned long -print_context_stack(struct thread_info *tinfo, +print_context_stack(struct task_struct *task, void *stack_start, unsigned long *stack, unsigned long bp, const struct stacktrace_ops *ops, void *data, unsigned long *end, int *graph); diff -urNp linux-2.6.32.48/arch/x86/kernel/e820.c linux-2.6.32.48-openvz/arch/x86/kernel/e820.c --- linux-2.6.32.48/arch/x86/kernel/e820.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/e820.c 2011-11-17 20:08:53.000000000 -0500 @@ -733,7 +733,7 @@ struct early_res { }; static struct early_res early_res[MAX_EARLY_RES] __initdata = { { 0, PAGE_SIZE, "BIOS data page" }, /* BIOS data page */ - {} + { 0, 0, {0}, 0 } }; static int __init find_overlapped_early(u64 start, u64 end) diff -urNp linux-2.6.32.48/arch/x86/kernel/early_printk.c linux-2.6.32.48-openvz/arch/x86/kernel/early_printk.c --- linux-2.6.32.48/arch/x86/kernel/early_printk.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/early_printk.c 2011-11-17 20:08:53.000000000 -0500 @@ -7,6 +7,7 @@ #include #include #include +#include #include #include #include @@ -170,6 +171,8 @@ asmlinkage void early_printk(const char int n; va_list ap; + pax_track_stack(); + va_start(ap, fmt); n = vscnprintf(buf, sizeof(buf), fmt, ap); early_console->write(early_console, buf, n); diff -urNp linux-2.6.32.48/arch/x86/kernel/efi_32.c linux-2.6.32.48-openvz/arch/x86/kernel/efi_32.c --- linux-2.6.32.48/arch/x86/kernel/efi_32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/efi_32.c 2011-11-17 20:09:04.000000000 -0500 @@ -38,70 +38,56 @@ */ static unsigned long efi_rt_eflags; -static pgd_t efi_bak_pg_dir_pointer[2]; +static pgd_t __initdata efi_bak_pg_dir_pointer[KERNEL_PGD_PTRS]; -void efi_call_phys_prelog(void) +void __init efi_call_phys_prelog(void) { - unsigned long cr4; - unsigned long temp; struct desc_ptr gdt_descr; - local_irq_save(efi_rt_eflags); +#ifdef CONFIG_PAX_KERNEXEC + struct desc_struct d; +#endif - /* - * If I don't have PAE, I should just duplicate two entries in page - * directory. If I have PAE, I just need to duplicate one entry in - * page directory. - */ - cr4 = read_cr4_safe(); + local_irq_save(efi_rt_eflags); - if (cr4 & X86_CR4_PAE) { - efi_bak_pg_dir_pointer[0].pgd = - swapper_pg_dir[pgd_index(0)].pgd; - swapper_pg_dir[0].pgd = - swapper_pg_dir[pgd_index(PAGE_OFFSET)].pgd; - } else { - efi_bak_pg_dir_pointer[0].pgd = - swapper_pg_dir[pgd_index(0)].pgd; - efi_bak_pg_dir_pointer[1].pgd = - swapper_pg_dir[pgd_index(0x400000)].pgd; - swapper_pg_dir[pgd_index(0)].pgd = - swapper_pg_dir[pgd_index(PAGE_OFFSET)].pgd; - temp = PAGE_OFFSET + 0x400000; - swapper_pg_dir[pgd_index(0x400000)].pgd = - swapper_pg_dir[pgd_index(temp)].pgd; - } + clone_pgd_range(efi_bak_pg_dir_pointer, swapper_pg_dir, KERNEL_PGD_PTRS); + clone_pgd_range(swapper_pg_dir, swapper_pg_dir + KERNEL_PGD_BOUNDARY, + min_t(unsigned long, KERNEL_PGD_PTRS, KERNEL_PGD_BOUNDARY)); /* * After the lock is released, the original page table is restored. */ __flush_tlb_all(); +#ifdef CONFIG_PAX_KERNEXEC + pack_descriptor(&d, 0, 0xFFFFF, 0x9B, 0xC); + write_gdt_entry(get_cpu_gdt_table(0), GDT_ENTRY_KERNEXEC_EFI_CS, &d, DESCTYPE_S); + pack_descriptor(&d, 0, 0xFFFFF, 0x93, 0xC); + write_gdt_entry(get_cpu_gdt_table(0), GDT_ENTRY_KERNEXEC_EFI_DS, &d, DESCTYPE_S); +#endif + gdt_descr.address = __pa(get_cpu_gdt_table(0)); gdt_descr.size = GDT_SIZE - 1; load_gdt(&gdt_descr); } -void efi_call_phys_epilog(void) +void __init efi_call_phys_epilog(void) { - unsigned long cr4; struct desc_ptr gdt_descr; +#ifdef CONFIG_PAX_KERNEXEC + struct desc_struct d; + + memset(&d, 0, sizeof d); + write_gdt_entry(get_cpu_gdt_table(0), GDT_ENTRY_KERNEXEC_EFI_CS, &d, DESCTYPE_S); + write_gdt_entry(get_cpu_gdt_table(0), GDT_ENTRY_KERNEXEC_EFI_DS, &d, DESCTYPE_S); +#endif + gdt_descr.address = (unsigned long)get_cpu_gdt_table(0); gdt_descr.size = GDT_SIZE - 1; load_gdt(&gdt_descr); - cr4 = read_cr4_safe(); - - if (cr4 & X86_CR4_PAE) { - swapper_pg_dir[pgd_index(0)].pgd = - efi_bak_pg_dir_pointer[0].pgd; - } else { - swapper_pg_dir[pgd_index(0)].pgd = - efi_bak_pg_dir_pointer[0].pgd; - swapper_pg_dir[pgd_index(0x400000)].pgd = - efi_bak_pg_dir_pointer[1].pgd; - } + clone_pgd_range(swapper_pg_dir, efi_bak_pg_dir_pointer, KERNEL_PGD_PTRS); /* * After the lock is released, the original page table is restored. diff -urNp linux-2.6.32.48/arch/x86/kernel/efi_stub_32.S linux-2.6.32.48-openvz/arch/x86/kernel/efi_stub_32.S --- linux-2.6.32.48/arch/x86/kernel/efi_stub_32.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/efi_stub_32.S 2011-11-17 20:09:04.000000000 -0500 @@ -6,7 +6,9 @@ */ #include +#include #include +#include /* * efi_call_phys(void *, ...) is a function with variable parameters. @@ -20,7 +22,7 @@ * service functions will comply with gcc calling convention, too. */ -.text +__INIT ENTRY(efi_call_phys) /* * 0. The function can only be called in Linux kernel. So CS has been @@ -36,9 +38,11 @@ ENTRY(efi_call_phys) * The mapping of lower virtual memory has been created in prelog and * epilog. */ - movl $1f, %edx - subl $__PAGE_OFFSET, %edx - jmp *%edx + movl $(__KERNEXEC_EFI_DS), %edx + mov %edx, %ds + mov %edx, %es + mov %edx, %ss + ljmp $(__KERNEXEC_EFI_CS),$1f-__PAGE_OFFSET 1: /* @@ -47,14 +51,8 @@ ENTRY(efi_call_phys) * parameter 2, ..., param n. To make things easy, we save the return * address of efi_call_phys in a global variable. */ - popl %edx - movl %edx, saved_return_addr - /* get the function pointer into ECX*/ - popl %ecx - movl %ecx, efi_rt_function_ptr - movl $2f, %edx - subl $__PAGE_OFFSET, %edx - pushl %edx + popl (saved_return_addr) + popl (efi_rt_function_ptr) /* * 3. Clear PG bit in %CR0. @@ -73,9 +71,8 @@ ENTRY(efi_call_phys) /* * 5. Call the physical function. */ - jmp *%ecx + call *(efi_rt_function_ptr-__PAGE_OFFSET) -2: /* * 6. After EFI runtime service returns, control will return to * following instruction. We'd better readjust stack pointer first. @@ -88,35 +85,32 @@ ENTRY(efi_call_phys) movl %cr0, %edx orl $0x80000000, %edx movl %edx, %cr0 - jmp 1f -1: + /* * 8. Now restore the virtual mode from flat mode by * adding EIP with PAGE_OFFSET. */ - movl $1f, %edx - jmp *%edx + ljmp $(__KERNEL_CS),$1f+__PAGE_OFFSET 1: + movl $(__KERNEL_DS), %edx + mov %edx, %ds + mov %edx, %es + mov %edx, %ss /* * 9. Balance the stack. And because EAX contain the return value, * we'd better not clobber it. */ - leal efi_rt_function_ptr, %edx - movl (%edx), %ecx - pushl %ecx + pushl (efi_rt_function_ptr) /* - * 10. Push the saved return address onto the stack and return. + * 10. Return to the saved return address. */ - leal saved_return_addr, %edx - movl (%edx), %ecx - pushl %ecx - ret + jmpl *(saved_return_addr) ENDPROC(efi_call_phys) .previous -.data +__INITDATA saved_return_addr: .long 0 efi_rt_function_ptr: diff -urNp linux-2.6.32.48/arch/x86/kernel/efi_stub_64.S linux-2.6.32.48-openvz/arch/x86/kernel/efi_stub_64.S --- linux-2.6.32.48/arch/x86/kernel/efi_stub_64.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/efi_stub_64.S 2011-11-17 20:09:04.000000000 -0500 @@ -7,6 +7,7 @@ */ #include +#include #define SAVE_XMM \ mov %rsp, %rax; \ @@ -40,6 +41,7 @@ ENTRY(efi_call0) call *%rdi addq $32, %rsp RESTORE_XMM + pax_force_retaddr ret ENDPROC(efi_call0) @@ -50,6 +52,7 @@ ENTRY(efi_call1) call *%rdi addq $32, %rsp RESTORE_XMM + pax_force_retaddr ret ENDPROC(efi_call1) @@ -60,6 +63,7 @@ ENTRY(efi_call2) call *%rdi addq $32, %rsp RESTORE_XMM + pax_force_retaddr ret ENDPROC(efi_call2) @@ -71,6 +75,7 @@ ENTRY(efi_call3) call *%rdi addq $32, %rsp RESTORE_XMM + pax_force_retaddr ret ENDPROC(efi_call3) @@ -83,6 +88,7 @@ ENTRY(efi_call4) call *%rdi addq $32, %rsp RESTORE_XMM + pax_force_retaddr ret ENDPROC(efi_call4) @@ -96,6 +102,7 @@ ENTRY(efi_call5) call *%rdi addq $48, %rsp RESTORE_XMM + pax_force_retaddr ret ENDPROC(efi_call5) @@ -112,5 +119,6 @@ ENTRY(efi_call6) call *%rdi addq $48, %rsp RESTORE_XMM + pax_force_retaddr ret ENDPROC(efi_call6) diff -urNp linux-2.6.32.48/arch/x86/kernel/entry_32.S linux-2.6.32.48-openvz/arch/x86/kernel/entry_32.S --- linux-2.6.32.48/arch/x86/kernel/entry_32.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/entry_32.S 2011-11-17 20:09:04.000000000 -0500 @@ -185,13 +185,146 @@ /*CFI_REL_OFFSET gs, PT_GS*/ .endm .macro SET_KERNEL_GS reg + +#ifdef CONFIG_CC_STACKPROTECTOR movl $(__KERNEL_STACK_CANARY), \reg +#elif defined(CONFIG_PAX_MEMORY_UDEREF) + movl $(__USER_DS), \reg +#else + xorl \reg, \reg +#endif + movl \reg, %gs .endm #endif /* CONFIG_X86_32_LAZY_GS */ -.macro SAVE_ALL +.macro pax_enter_kernel +#ifdef CONFIG_PAX_KERNEXEC + call pax_enter_kernel +#endif +.endm + +.macro pax_exit_kernel +#ifdef CONFIG_PAX_KERNEXEC + call pax_exit_kernel +#endif +.endm + +#ifdef CONFIG_PAX_KERNEXEC +ENTRY(pax_enter_kernel) +#ifdef CONFIG_PARAVIRT + pushl %eax + pushl %ecx + call PARA_INDIRECT(pv_cpu_ops+PV_CPU_read_cr0) + mov %eax, %esi +#else + mov %cr0, %esi +#endif + bts $16, %esi + jnc 1f + mov %cs, %esi + cmp $__KERNEL_CS, %esi + jz 3f + ljmp $__KERNEL_CS, $3f +1: ljmp $__KERNEXEC_KERNEL_CS, $2f +2: +#ifdef CONFIG_PARAVIRT + mov %esi, %eax + call PARA_INDIRECT(pv_cpu_ops+PV_CPU_write_cr0) +#else + mov %esi, %cr0 +#endif +3: +#ifdef CONFIG_PARAVIRT + popl %ecx + popl %eax +#endif + ret +ENDPROC(pax_enter_kernel) + +ENTRY(pax_exit_kernel) +#ifdef CONFIG_PARAVIRT + pushl %eax + pushl %ecx +#endif + mov %cs, %esi + cmp $__KERNEXEC_KERNEL_CS, %esi + jnz 2f +#ifdef CONFIG_PARAVIRT + call PARA_INDIRECT(pv_cpu_ops+PV_CPU_read_cr0); + mov %eax, %esi +#else + mov %cr0, %esi +#endif + btr $16, %esi + ljmp $__KERNEL_CS, $1f +1: +#ifdef CONFIG_PARAVIRT + mov %esi, %eax + call PARA_INDIRECT(pv_cpu_ops+PV_CPU_write_cr0); +#else + mov %esi, %cr0 +#endif +2: +#ifdef CONFIG_PARAVIRT + popl %ecx + popl %eax +#endif + ret +ENDPROC(pax_exit_kernel) +#endif + +.macro pax_erase_kstack +#ifdef CONFIG_PAX_MEMORY_STACKLEAK + call pax_erase_kstack +#endif +.endm + +#ifdef CONFIG_PAX_MEMORY_STACKLEAK +/* + * ebp: thread_info + * ecx, edx: can be clobbered + */ +ENTRY(pax_erase_kstack) + pushl %edi + pushl %eax + + mov TI_lowest_stack(%ebp), %edi + mov $-0xBEEF, %eax + std + +1: mov %edi, %ecx + and $THREAD_SIZE_asm - 1, %ecx + shr $2, %ecx + repne scasl + jecxz 2f + + cmp $2*16, %ecx + jc 2f + + mov $2*16, %ecx + repe scasl + jecxz 2f + jne 1b + +2: cld + mov %esp, %ecx + sub %edi, %ecx + shr $2, %ecx + rep stosl + + mov TI_task_thread_sp0(%ebp), %edi + sub $128, %edi + mov %edi, TI_lowest_stack(%ebp) + + popl %eax + popl %edi + ret +ENDPROC(pax_erase_kstack) +#endif + +.macro __SAVE_ALL _DS cld PUSH_GS pushl %fs @@ -224,7 +357,7 @@ pushl %ebx CFI_ADJUST_CFA_OFFSET 4 CFI_REL_OFFSET ebx, 0 - movl $(__USER_DS), %edx + movl $\_DS, %edx movl %edx, %ds movl %edx, %es movl $(__KERNEL_PERCPU), %edx @@ -232,6 +365,15 @@ SET_KERNEL_GS %edx .endm +.macro SAVE_ALL +#if defined(CONFIG_PAX_KERNEXEC) || defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC) || defined(CONFIG_PAX_MEMORY_UDEREF) + __SAVE_ALL __KERNEL_DS + pax_enter_kernel +#else + __SAVE_ALL __USER_DS +#endif +.endm + .macro RESTORE_INT_REGS popl %ebx CFI_ADJUST_CFA_OFFSET -4 @@ -325,6 +467,7 @@ ENTRY(ret_from_fork) GET_THREAD_INFO(%ebp) popl %eax CFI_ADJUST_CFA_OFFSET -4 +ret_from_fork_tail: pushl $0x0202 # Reset kernel eflags CFI_ADJUST_CFA_OFFSET 4 popfl @@ -333,6 +476,25 @@ ENTRY(ret_from_fork) CFI_ENDPROC END(ret_from_fork) +ENTRY(i386_ret_from_resume) + CFI_STARTPROC + pushl %eax + CFI_ADJUST_CFA_OFFSET 4 + call schedule_tail + GET_THREAD_INFO(%ebp) + popl %eax + CFI_ADJUST_CFA_OFFSET -4 + movl (%esp),%eax + testl %eax,%eax + jz 1f + pushl %esp + call *%eax + addl $4,%esp +1: + addl $256,%esp + jmp ret_from_fork_tail + CFI_ENDPROC + /* * Return to user mode is not as complex as all this looks, * but we want the default path for a system call return to @@ -352,7 +514,15 @@ check_userspace: movb PT_CS(%esp), %al andl $(X86_EFLAGS_VM | SEGMENT_RPL_MASK), %eax cmpl $USER_RPL, %eax + +#ifdef CONFIG_PAX_KERNEXEC + jae resume_userspace + + PAX_EXIT_KERNEL + jmp resume_kernel +#else jb resume_kernel # not returning to v8086 or userspace +#endif ENTRY(resume_userspace) LOCKDEP_SYS_EXIT @@ -414,25 +584,36 @@ sysenter_past_esp: /*CFI_REL_OFFSET cs, 0*/ /* * Push current_thread_info()->sysenter_return to the stack. - * A tiny bit of offset fixup is necessary - 4*4 means the 4 words - * pushed above; +8 corresponds to copy_thread's esp0 setting. */ - pushl (TI_sysenter_return-THREAD_SIZE+8+4*4)(%esp) + pushl $0 CFI_ADJUST_CFA_OFFSET 4 CFI_REL_OFFSET eip, 0 pushl %eax CFI_ADJUST_CFA_OFFSET 4 SAVE_ALL + GET_THREAD_INFO(%ebp) + movl TI_sysenter_return(%ebp),%ebp + movl %ebp,PT_EIP(%esp) ENABLE_INTERRUPTS(CLBR_NONE) /* * Load the potential sixth argument from user stack. * Careful about security. */ + movl PT_OLDESP(%esp),%ebp + +#ifdef CONFIG_PAX_MEMORY_UDEREF + mov PT_OLDSS(%esp),%ds +1: movl %ds:(%ebp),%ebp + push %ss + pop %ds +#else cmpl $__PAGE_OFFSET-3,%ebp jae syscall_fault 1: movl (%ebp),%ebp +#endif + movl %ebp,PT_EBP(%esp) .section __ex_table,"a" .align 4 @@ -455,12 +636,24 @@ sysenter_do_call: testl $_TIF_ALLWORK_MASK, %ecx jne sysexit_audit sysenter_exit: + +#ifdef CONFIG_PAX_RANDKSTACK + pushl_cfi %eax + movl %esp, %eax + call pax_randomize_kstack + popl_cfi %eax +#endif + + pax_erase_kstack + /* if something modifies registers it must also disable sysexit */ movl PT_EIP(%esp), %edx movl PT_OLDESP(%esp), %ecx xorl %ebp,%ebp TRACE_IRQS_ON 1: mov PT_FS(%esp), %fs +2: mov PT_DS(%esp), %ds +3: mov PT_ES(%esp), %es PTGS_TO_GS ENABLE_INTERRUPTS_SYSEXIT @@ -504,11 +697,17 @@ sysexit_audit: CFI_ENDPROC .pushsection .fixup,"ax" -2: movl $0,PT_FS(%esp) +4: movl $0,PT_FS(%esp) + jmp 1b +5: movl $0,PT_DS(%esp) + jmp 1b +6: movl $0,PT_ES(%esp) jmp 1b .section __ex_table,"a" .align 4 - .long 1b,2b + .long 1b,4b + .long 2b,5b + .long 3b,6b .popsection PTGS_TO_GS_EX ENDPROC(ia32_sysenter_target) @@ -538,6 +737,13 @@ syscall_exit: testl $_TIF_ALLWORK_MASK, %ecx # current->work jne syscall_exit_work +#ifdef CONFIG_PAX_RANDKSTACK + movl %esp, %eax + call pax_randomize_kstack +#endif + + pax_erase_kstack + restore_all: TRACE_IRQS_IRET restore_all_notrace: @@ -602,10 +808,29 @@ ldt_ss: mov PT_OLDESP(%esp), %eax /* load userspace esp */ mov %dx, %ax /* eax: new kernel esp */ sub %eax, %edx /* offset (low word is 0) */ - PER_CPU(gdt_page, %ebx) +#ifdef CONFIG_SMP + movl PER_CPU_VAR(cpu_number), %ebx + shll $PAGE_SHIFT_asm, %ebx + addl $cpu_gdt_table, %ebx +#else + movl $cpu_gdt_table, %ebx +#endif shr $16, %edx + +#ifdef CONFIG_PAX_KERNEXEC + mov %cr0, %esi + btr $16, %esi + mov %esi, %cr0 +#endif + mov %dl, GDT_ENTRY_ESPFIX_SS * 8 + 4(%ebx) /* bits 16..23 */ mov %dh, GDT_ENTRY_ESPFIX_SS * 8 + 7(%ebx) /* bits 24..31 */ + +#ifdef CONFIG_PAX_KERNEXEC + bts $16, %esi + mov %esi, %cr0 +#endif + pushl $__ESPFIX_SS CFI_ADJUST_CFA_OFFSET 4 push %eax /* new kernel esp */ @@ -642,25 +867,19 @@ work_resched: work_notifysig: # deal with pending signals and # notify-resume requests + movl %esp, %eax #ifdef CONFIG_VM86 testl $X86_EFLAGS_VM, PT_EFLAGS(%esp) - movl %esp, %eax - jne work_notifysig_v86 # returning to kernel-space or + jz 1f # returning to kernel-space or # vm86-space - xorl %edx, %edx - call do_notify_resume - jmp resume_userspace_sig - ALIGN -work_notifysig_v86: pushl %ecx # save ti_flags for do_notify_resume CFI_ADJUST_CFA_OFFSET 4 call save_v86_state # %eax contains pt_regs pointer popl %ecx CFI_ADJUST_CFA_OFFSET -4 movl %eax, %esp -#else - movl %esp, %eax +1: #endif xorl %edx, %edx call do_notify_resume @@ -695,6 +914,10 @@ END(syscall_exit_work) RING0_INT_FRAME # can't unwind into user space anyway syscall_fault: +#ifdef CONFIG_PAX_MEMORY_UDEREF + push %ss + pop %ds +#endif GET_THREAD_INFO(%ebp) movl $-EFAULT,PT_EAX(%esp) jmp resume_userspace @@ -726,6 +949,33 @@ PTREGSCALL(rt_sigreturn) PTREGSCALL(vm86) PTREGSCALL(vm86old) + ALIGN; +ENTRY(kernel_execve) + push %ebp + sub $PT_OLDSS+4,%esp + push %edi + push %ecx + push %eax + lea 3*4(%esp),%edi + mov $PT_OLDSS/4+1,%ecx + xorl %eax,%eax + rep stosl + pop %eax + pop %ecx + pop %edi + movl $X86_EFLAGS_IF,PT_EFLAGS(%esp) + mov %eax,PT_EBX(%esp) + mov %edx,PT_ECX(%esp) + mov %ecx,PT_EDX(%esp) + mov %esp,%eax + call sys_execve + GET_THREAD_INFO(%ebp) + test %eax,%eax + jz syscall_exit + add $PT_OLDSS+4,%esp + pop %ebp + ret + .macro FIXUP_ESPFIX_STACK /* * Switch back for ESPFIX stack to the normal zerobased stack @@ -735,7 +985,13 @@ PTREGSCALL(vm86old) * normal stack and adjusts ESP with the matching offset. */ /* fixup the stack */ - PER_CPU(gdt_page, %ebx) +#ifdef CONFIG_SMP + movl PER_CPU_VAR(cpu_number), %ebx + shll $PAGE_SHIFT_asm, %ebx + addl $cpu_gdt_table, %ebx +#else + movl $cpu_gdt_table, %ebx +#endif mov GDT_ENTRY_ESPFIX_SS * 8 + 4(%ebx), %al /* bits 16..23 */ mov GDT_ENTRY_ESPFIX_SS * 8 + 7(%ebx), %ah /* bits 24..31 */ shl $16, %eax @@ -1198,7 +1454,6 @@ return_to_handler: ret #endif -.section .rodata,"a" #include "syscall_table_32.S" syscall_table_size=(.-sys_call_table) @@ -1255,9 +1510,12 @@ error_code: movl $-1, PT_ORIG_EAX(%esp) # no syscall to restart REG_TO_PTGS %ecx SET_KERNEL_GS %ecx - movl $(__USER_DS), %ecx + movl $(__KERNEL_DS), %ecx movl %ecx, %ds movl %ecx, %es + + pax_enter_kernel + TRACE_IRQS_OFF movl %esp,%eax # pt_regs pointer call *%edi @@ -1351,6 +1609,9 @@ nmi_stack_correct: xorl %edx,%edx # zero error code movl %esp,%eax # pt_regs pointer call do_nmi + + pax_exit_kernel + jmp restore_all_notrace CFI_ENDPROC @@ -1391,6 +1652,9 @@ nmi_espfix_stack: FIXUP_ESPFIX_STACK # %eax == %esp xorl %edx,%edx # zero error code call do_nmi + + pax_exit_kernel + RESTORE_REGS lss 12+4(%esp), %esp # back to espfix stack CFI_ADJUST_CFA_OFFSET -24 diff -urNp linux-2.6.32.48/arch/x86/kernel/entry_64.S linux-2.6.32.48-openvz/arch/x86/kernel/entry_64.S --- linux-2.6.32.48/arch/x86/kernel/entry_64.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/entry_64.S 2011-11-17 20:15:35.000000000 -0500 @@ -53,6 +53,8 @@ #include #include #include +#include +#include /* Avoid __ASSEMBLER__'ifying just for this. */ #include @@ -64,6 +66,7 @@ #ifdef CONFIG_FUNCTION_TRACER #ifdef CONFIG_DYNAMIC_FTRACE ENTRY(mcount) + pax_force_retaddr retq END(mcount) @@ -88,6 +91,7 @@ GLOBAL(ftrace_graph_call) #endif GLOBAL(ftrace_stub) + pax_force_retaddr retq END(ftrace_caller) @@ -108,6 +112,7 @@ ENTRY(mcount) #endif GLOBAL(ftrace_stub) + pax_force_retaddr retq trace: @@ -117,6 +122,7 @@ trace: movq 8(%rbp), %rsi subq $MCOUNT_INSN_SIZE, %rdi + pax_force_fptr ftrace_trace_function call *ftrace_trace_function MCOUNT_RESTORE_FRAME @@ -142,6 +148,7 @@ ENTRY(ftrace_graph_caller) MCOUNT_RESTORE_FRAME + pax_force_retaddr retq END(ftrace_graph_caller) @@ -159,6 +166,7 @@ GLOBAL(return_to_handler) movq 8(%rsp), %rdx movq (%rsp), %rax addq $16, %rsp + pax_force_retaddr retq #endif @@ -174,6 +182,264 @@ ENTRY(native_usergs_sysret64) ENDPROC(native_usergs_sysret64) #endif /* CONFIG_PARAVIRT */ + .macro ljmpq sel, off +#if defined(CONFIG_MPSC) || defined(CONFIG_MCORE2) || defined (CONFIG_MATOM) + .byte 0x48; ljmp *1234f(%rip) + .pushsection .rodata + .align 16 + 1234: .quad \off; .word \sel + .popsection +#else + pushq $\sel + pushq $\off + lretq +#endif + .endm + + .macro pax_enter_kernel +#ifdef CONFIG_PAX_KERNEXEC + call pax_enter_kernel +#endif + .endm + + .macro pax_exit_kernel +#ifdef CONFIG_PAX_KERNEXEC + call pax_exit_kernel +#endif + .endm + +#ifdef CONFIG_PAX_KERNEXEC +ENTRY(pax_enter_kernel) + pushq %rdi + +#ifdef CONFIG_PARAVIRT + PV_SAVE_REGS(CLBR_RDI) +#endif + + GET_CR0_INTO_RDI + bts $16,%rdi + jnc 1f + mov %cs,%edi + cmp $__KERNEL_CS,%edi + jz 3f + ljmpq __KERNEL_CS,3f +1: ljmpq __KERNEXEC_KERNEL_CS,2f +2: SET_RDI_INTO_CR0 +3: + +#ifdef CONFIG_PARAVIRT + PV_RESTORE_REGS(CLBR_RDI) +#endif + + popq %rdi + pax_force_retaddr + retq +ENDPROC(pax_enter_kernel) + +ENTRY(pax_exit_kernel) + pushq %rdi + +#ifdef CONFIG_PARAVIRT + PV_SAVE_REGS(CLBR_RDI) +#endif + + mov %cs,%rdi + cmp $__KERNEXEC_KERNEL_CS,%edi + jnz 2f + GET_CR0_INTO_RDI + btr $16,%rdi + ljmpq __KERNEL_CS,1f +1: SET_RDI_INTO_CR0 +2: + +#ifdef CONFIG_PARAVIRT + PV_RESTORE_REGS(CLBR_RDI); +#endif + + popq %rdi + pax_force_retaddr + retq +ENDPROC(pax_exit_kernel) +#endif + + .macro pax_enter_kernel_user +#ifdef CONFIG_PAX_MEMORY_UDEREF + call pax_enter_kernel_user +#endif + .endm + + .macro pax_exit_kernel_user +#ifdef CONFIG_PAX_MEMORY_UDEREF + call pax_exit_kernel_user +#endif + .endm + +#ifdef CONFIG_PAX_MEMORY_UDEREF +ENTRY(pax_enter_kernel_user) + pushq %rdi + pushq %rbx + +#ifdef CONFIG_PARAVIRT + PV_SAVE_REGS(CLBR_RDI) +#endif + + GET_CR3_INTO_RDI + mov %rdi,%rbx + add $__START_KERNEL_map,%rbx + sub phys_base(%rip),%rbx + +#ifdef CONFIG_PARAVIRT + pushq %rdi + cmpl $0, pv_info+PARAVIRT_enabled + jz 1f + i = 0 + .rept USER_PGD_PTRS + mov i*8(%rbx),%rsi + mov $0,%sil + lea i*8(%rbx),%rdi + call PARA_INDIRECT(pv_mmu_ops+PV_MMU_set_pgd_batched) + i = i + 1 + .endr + jmp 2f +1: +#endif + + i = 0 + .rept USER_PGD_PTRS + movb $0,i*8(%rbx) + i = i + 1 + .endr + +#ifdef CONFIG_PARAVIRT +2: popq %rdi +#endif + SET_RDI_INTO_CR3 + +#ifdef CONFIG_PAX_KERNEXEC + GET_CR0_INTO_RDI + bts $16,%rdi + SET_RDI_INTO_CR0 +#endif + +#ifdef CONFIG_PARAVIRT + PV_RESTORE_REGS(CLBR_RDI) +#endif + + popq %rbx + popq %rdi + pax_force_retaddr + retq +ENDPROC(pax_enter_kernel_user) + +ENTRY(pax_exit_kernel_user) + push %rdi + +#ifdef CONFIG_PARAVIRT + pushq %rbx + PV_SAVE_REGS(CLBR_RDI) +#endif + +#ifdef CONFIG_PAX_KERNEXEC + GET_CR0_INTO_RDI + btr $16,%rdi + SET_RDI_INTO_CR0 +#endif + + GET_CR3_INTO_RDI + add $__START_KERNEL_map,%rdi + sub phys_base(%rip),%rdi + +#ifdef CONFIG_PARAVIRT + cmpl $0, pv_info+PARAVIRT_enabled + jz 1f + mov %rdi,%rbx + i = 0 + .rept USER_PGD_PTRS + mov i*8(%rbx),%rsi + mov $0x67,%sil + lea i*8(%rbx),%rdi + call PARA_INDIRECT(pv_mmu_ops+PV_MMU_set_pgd_batched) + i = i + 1 + .endr + jmp 2f +1: +#endif + + i = 0 + .rept USER_PGD_PTRS + movb $0x67,i*8(%rdi) + i = i + 1 + .endr + +#ifdef CONFIG_PARAVIRT +2: PV_RESTORE_REGS(CLBR_RDI) + popq %rbx +#endif + + popq %rdi + pax_force_retaddr + retq +ENDPROC(pax_exit_kernel_user) +#endif + +.macro pax_erase_kstack +#ifdef CONFIG_PAX_RANDKSTACK + call pax_randomize_kstack +#endif +.endm + +#ifdef CONFIG_PAX_MEMORY_STACKLEAK +/* + * r10: thread_info + * rcx, rdx: can be clobbered + */ +ENTRY(pax_erase_kstack) + pushq %rdi + pushq %rax + pushq %r10 + + GET_THREAD_INFO(%r10) + mov TI_lowest_stack(%r10), %rdi + mov $-0xBEEF, %rax + std + +1: mov %edi, %ecx + and $THREAD_SIZE_asm - 1, %ecx + shr $3, %ecx + repne scasq + jecxz 2f + + cmp $2*8, %ecx + jc 2f + + mov $2*8, %ecx + repe scasq + jecxz 2f + jne 1b + +2: cld + mov %esp, %ecx + sub %edi, %ecx + + cmp $THREAD_SIZE_asm, %rcx + jb 3f + ud2 +3: + + shr $3, %ecx + rep stosq + + mov TI_task_thread_sp0(%r10), %rdi + sub $256, %rdi + mov %rdi, TI_lowest_stack(%r10) + + popq %r10 + popq %rax + popq %rdi + pax_force_retaddr + ret +ENDPROC(pax_erase_kstack) +#endif .macro TRACE_IRQS_IRETQ offset=ARGOFFSET #ifdef CONFIG_TRACE_IRQFLAGS @@ -317,7 +583,7 @@ ENTRY(save_args) leaq -ARGOFFSET+16(%rsp),%rdi /* arg1 for handler */ movq_cfi rbp, 8 /* push %rbp */ leaq 8(%rsp), %rbp /* mov %rsp, %ebp */ - testl $3, CS(%rdi) + testb $3, CS(%rdi) je 1f SWAPGS /* @@ -337,6 +603,7 @@ ENTRY(save_args) * We entered an interrupt context - irqs are off: */ 2: TRACE_IRQS_OFF + pax_force_retaddr ret CFI_ENDPROC END(save_args) @@ -352,6 +619,7 @@ ENTRY(save_rest) movq_cfi r15, R15+16 movq %r11, 8(%rsp) /* return address */ FIXUP_TOP_OF_STACK %r11, 16 + pax_force_retaddr ret CFI_ENDPROC END(save_rest) @@ -383,7 +651,8 @@ ENTRY(save_paranoid) js 1f /* negative -> in kernel */ SWAPGS xorl %ebx,%ebx -1: ret +1: pax_force_retaddr + ret CFI_ENDPROC END(save_paranoid) .popsection @@ -405,11 +674,15 @@ ENTRY(ret_from_fork) call schedule_tail # rdi: 'prev' task parameter +ret_from_fork_tail: GET_THREAD_INFO(%rcx) + btr $TIF_RESUME,TI_flags(%rcx) + jc x86_64_ret_from_resume +ret_from_fork_check: RESTORE_REST - testl $3, CS-ARGOFFSET(%rsp) # from kernel_thread? + testb $3, CS-ARGOFFSET(%rsp) # from kernel_thread? je int_ret_from_sys_call testl $_TIF_IA32, TI_flags(%rcx) # 32-bit compat task needs IRET @@ -418,6 +691,18 @@ ENTRY(ret_from_fork) RESTORE_TOP_OF_STACK %rdi, -ARGOFFSET jmp ret_from_sys_call # go to the SYSRET fastpath +x86_64_ret_from_resume: + movq (%rsp),%rax + testq %rax,%rax + jz 1f + movq %rsp,%rdi + call *%rax +1: + addq $256,%rsp + cmpq $0,ORIG_RAX(%rsp) + jge ret_from_fork_tail + RESTORE_REST + jmp int_ret_from_sys_call CFI_ENDPROC END(ret_from_fork) @@ -455,7 +740,7 @@ END(ret_from_fork) ENTRY(system_call) CFI_STARTPROC simple CFI_SIGNAL_FRAME - CFI_DEF_CFA rsp,KERNEL_STACK_OFFSET + CFI_DEF_CFA rsp,0 CFI_REGISTER rip,rcx /*CFI_REGISTER rflags,r11*/ SWAPGS_UNSAFE_STACK @@ -468,12 +753,13 @@ ENTRY(system_call_after_swapgs) movq %rsp,PER_CPU_VAR(old_rsp) movq PER_CPU_VAR(kernel_stack),%rsp + pax_enter_kernel_user /* * No need to follow this irqs off/on section - it's straight * and short: */ ENABLE_INTERRUPTS(CLBR_NONE) - SAVE_ARGS 8,1 + SAVE_ARGS 8*6,1 movq %rax,ORIG_RAX-ARGOFFSET(%rsp) movq %rcx,RIP-ARGOFFSET(%rsp) CFI_REL_OFFSET rip,RIP-ARGOFFSET @@ -502,6 +788,8 @@ sysret_check: andl %edi,%edx jnz sysret_careful CFI_REMEMBER_STATE + pax_exit_kernel_user + pax_erase_kstack /* * sysretq will re-enable interrupts: */ @@ -613,7 +901,7 @@ tracesys: GLOBAL(int_ret_from_sys_call) DISABLE_INTERRUPTS(CLBR_NONE) TRACE_IRQS_OFF - testl $3,CS-ARGOFFSET(%rsp) + testb $3,CS-ARGOFFSET(%rsp) je retint_restore_args movl $_TIF_ALLWORK_MASK,%edi /* edi: mask to check */ @@ -708,6 +996,7 @@ ENTRY(ptregscall_common) movq_cfi_restore R12+8, r12 movq_cfi_restore RBP+8, rbp movq_cfi_restore RBX+8, rbx + pax_force_retaddr ret $REST_SKIP /* pop extended registers */ CFI_ENDPROC END(ptregscall_common) @@ -800,6 +1089,16 @@ END(interrupt) CFI_ADJUST_CFA_OFFSET 10*8 call save_args PARTIAL_FRAME 0 +#ifdef CONFIG_PAX_MEMORY_UDEREF + testb $3, CS(%rdi) + jnz 1f + pax_enter_kernel + jmp 2f +1: pax_enter_kernel_user +2: +#else + pax_enter_kernel +#endif call \func .endm @@ -822,7 +1121,7 @@ ret_from_intr: CFI_ADJUST_CFA_OFFSET -8 exit_intr: GET_THREAD_INFO(%rcx) - testl $3,CS-ARGOFFSET(%rsp) + testb $3,CS-ARGOFFSET(%rsp) je retint_kernel /* Interrupt came from user space */ @@ -844,12 +1143,16 @@ retint_swapgs: /* return to user-space * The iretq could re-enable interrupts: */ DISABLE_INTERRUPTS(CLBR_ANY) + pax_exit_kernel_user + pax_erase_kstack TRACE_IRQS_IRETQ SWAPGS jmp restore_args retint_restore_args: /* return to kernel space */ DISABLE_INTERRUPTS(CLBR_ANY) + pax_exit_kernel + pax_force_retaddr RIP-ARGOFFSET /* * The iretq could re-enable interrupts: */ @@ -1032,6 +1335,16 @@ ENTRY(\sym) CFI_ADJUST_CFA_OFFSET 15*8 call error_entry DEFAULT_FRAME 0 +#ifdef CONFIG_PAX_MEMORY_UDEREF + testb $3, CS(%rsp) + jnz 1f + pax_enter_kernel + jmp 2f +1: pax_enter_kernel_user +2: +#else + pax_enter_kernel +#endif movq %rsp,%rdi /* pt_regs pointer */ xorl %esi,%esi /* no error code */ call \do_sym @@ -1049,6 +1362,16 @@ ENTRY(\sym) subq $15*8, %rsp call save_paranoid TRACE_IRQS_OFF +#ifdef CONFIG_PAX_MEMORY_UDEREF + testb $3, CS(%rsp) + jnz 1f + pax_enter_kernel + jmp 2f +1: pax_enter_kernel_user +2: +#else + pax_enter_kernel +#endif movq %rsp,%rdi /* pt_regs pointer */ xorl %esi,%esi /* no error code */ call \do_sym @@ -1066,9 +1389,24 @@ ENTRY(\sym) subq $15*8, %rsp call save_paranoid TRACE_IRQS_OFF +#ifdef CONFIG_PAX_MEMORY_UDEREF + testb $3, CS(%rsp) + jnz 1f + pax_enter_kernel + jmp 2f +1: pax_enter_kernel_user +2: +#else + pax_enter_kernel +#endif movq %rsp,%rdi /* pt_regs pointer */ xorl %esi,%esi /* no error code */ - PER_CPU(init_tss, %rbp) +#ifdef CONFIG_SMP + imul $TSS_size, PER_CPU_VAR(cpu_number), %ebp + lea init_tss(%rbp), %rbp +#else + lea init_tss(%rip), %rbp +#endif subq $EXCEPTION_STKSZ, TSS_ist + (\ist - 1) * 8(%rbp) call \do_sym addq $EXCEPTION_STKSZ, TSS_ist + (\ist - 1) * 8(%rbp) @@ -1085,6 +1423,16 @@ ENTRY(\sym) CFI_ADJUST_CFA_OFFSET 15*8 call error_entry DEFAULT_FRAME 0 +#ifdef CONFIG_PAX_MEMORY_UDEREF + testb $3, CS(%rsp) + jnz 1f + pax_enter_kernel + jmp 2f +1: pax_enter_kernel_user +2: +#else + pax_enter_kernel +#endif movq %rsp,%rdi /* pt_regs pointer */ movq ORIG_RAX(%rsp),%rsi /* get error code */ movq $-1,ORIG_RAX(%rsp) /* no syscall to restart */ @@ -1104,6 +1452,16 @@ ENTRY(\sym) call save_paranoid DEFAULT_FRAME 0 TRACE_IRQS_OFF +#ifdef CONFIG_PAX_MEMORY_UDEREF + testb $3, CS(%rsp) + jnz 1f + pax_enter_kernel + jmp 2f +1: pax_enter_kernel_user +2: +#else + pax_enter_kernel +#endif movq %rsp,%rdi /* pt_regs pointer */ movq ORIG_RAX(%rsp),%rsi /* get error code */ movq $-1,ORIG_RAX(%rsp) /* no syscall to restart */ @@ -1141,6 +1499,7 @@ gs_change: SWAPGS popf CFI_ADJUST_CFA_OFFSET -8 + pax_force_retaddr ret CFI_ENDPROC END(native_load_gs_index) @@ -1182,7 +1541,7 @@ ENTRY(kernel_thread) xorl %r9d,%r9d # clone now - call do_fork + call do_fork_kthread movq %rax,RAX(%rsp) xorl %edi,%edi @@ -1195,6 +1554,7 @@ ENTRY(kernel_thread) */ RESTORE_ALL UNFAKE_STACK_FRAME + pax_force_retaddr ret CFI_ENDPROC END(kernel_thread) @@ -1208,6 +1568,7 @@ ENTRY(child_rip) */ movq %rdi, %rax movq %rsi, %rdi + pax_force_fptr %rax call *%rax # exit mov %eax, %edi @@ -1243,6 +1604,7 @@ ENTRY(kernel_execve) je int_ret_from_sys_call RESTORE_ARGS UNFAKE_STACK_FRAME + pax_force_retaddr ret CFI_ENDPROC END(kernel_execve) @@ -1263,6 +1625,7 @@ ENTRY(call_softirq) CFI_DEF_CFA_REGISTER rsp CFI_ADJUST_CFA_OFFSET -8 decl PER_CPU_VAR(irq_count) + pax_force_retaddr ret CFI_ENDPROC END(call_softirq) @@ -1405,16 +1768,31 @@ ENTRY(paranoid_exit) TRACE_IRQS_OFF testl %ebx,%ebx /* swapgs needed? */ jnz paranoid_restore - testl $3,CS(%rsp) + testb $3,CS(%rsp) jnz paranoid_userspace +#ifdef CONFIG_PAX_MEMORY_UDEREF + pax_exit_kernel + TRACE_IRQS_IRETQ 0 + SWAPGS_UNSAFE_STACK + RESTORE_ALL 8 + pax_force_retaddr + jmp irq_return +#endif paranoid_swapgs: +#ifdef CONFIG_PAX_MEMORY_UDEREF + pax_exit_kernel_user +#else + pax_exit_kernel +#endif TRACE_IRQS_IRETQ 0 SWAPGS_UNSAFE_STACK RESTORE_ALL 8 jmp irq_return paranoid_restore: + pax_exit_kernel TRACE_IRQS_IRETQ 0 RESTORE_ALL 8 + pax_force_retaddr jmp irq_return paranoid_userspace: GET_THREAD_INFO(%rcx) @@ -1470,12 +1848,13 @@ ENTRY(error_entry) movq_cfi r14, R14+8 movq_cfi r15, R15+8 xorl %ebx,%ebx - testl $3,CS+8(%rsp) + testb $3,CS+8(%rsp) je error_kernelspace error_swapgs: SWAPGS error_sti: TRACE_IRQS_OFF + pax_force_retaddr ret CFI_ENDPROC @@ -1529,6 +1908,16 @@ ENTRY(nmi) CFI_ADJUST_CFA_OFFSET 15*8 call save_paranoid DEFAULT_FRAME 0 +#ifdef CONFIG_PAX_MEMORY_UDEREF + testb $3, CS(%rsp) + jnz 1f + pax_enter_kernel + jmp 2f +1: pax_enter_kernel_user +2: +#else + pax_enter_kernel +#endif /* paranoidentry do_nmi, 0; without TRACE_IRQS_OFF */ movq %rsp,%rdi movq $-1,%rsi @@ -1539,12 +1928,28 @@ ENTRY(nmi) DISABLE_INTERRUPTS(CLBR_NONE) testl %ebx,%ebx /* swapgs needed? */ jnz nmi_restore - testl $3,CS(%rsp) + testb $3,CS(%rsp) jnz nmi_userspace +#ifdef CONFIG_PAX_MEMORY_UDEREF + pax_exit_kernel + SWAPGS_UNSAFE_STACK + RESTORE_ALL 8 + pax_force_retaddr + jmp irq_return +#endif nmi_swapgs: +#ifdef CONFIG_PAX_MEMORY_UDEREF + pax_exit_kernel_user +#else + pax_exit_kernel +#endif SWAPGS_UNSAFE_STACK + RESTORE_ALL 8 + jmp irq_return nmi_restore: + pax_exit_kernel RESTORE_ALL 8 + pax_force_retaddr jmp irq_return nmi_userspace: GET_THREAD_INFO(%rcx) diff -urNp linux-2.6.32.48/arch/x86/kernel/ftrace.c linux-2.6.32.48-openvz/arch/x86/kernel/ftrace.c --- linux-2.6.32.48/arch/x86/kernel/ftrace.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/ftrace.c 2011-11-17 20:08:53.000000000 -0500 @@ -103,7 +103,7 @@ static void *mod_code_ip; /* holds the static void *mod_code_newcode; /* holds the text to write to the IP */ static unsigned nmi_wait_count; -static atomic_t nmi_update_count = ATOMIC_INIT(0); +static atomic_unchecked_t nmi_update_count = ATOMIC_INIT(0); int ftrace_arch_read_dyn_info(char *buf, int size) { @@ -111,7 +111,7 @@ int ftrace_arch_read_dyn_info(char *buf, r = snprintf(buf, size, "%u %u", nmi_wait_count, - atomic_read(&nmi_update_count)); + atomic_read_unchecked(&nmi_update_count)); return r; } @@ -149,8 +149,10 @@ void ftrace_nmi_enter(void) { if (atomic_inc_return(&nmi_running) & MOD_CODE_WRITE_FLAG) { smp_rmb(); + pax_open_kernel(); ftrace_mod_code(); - atomic_inc(&nmi_update_count); + pax_close_kernel(); + atomic_inc_unchecked(&nmi_update_count); } /* Must have previous changes seen before executions */ smp_mb(); @@ -215,7 +217,7 @@ do_ftrace_mod_code(unsigned long ip, voi -static unsigned char ftrace_nop[MCOUNT_INSN_SIZE]; +static unsigned char ftrace_nop[MCOUNT_INSN_SIZE] __read_only; static unsigned char *ftrace_nop_replace(void) { @@ -228,6 +230,8 @@ ftrace_modify_code(unsigned long ip, uns { unsigned char replaced[MCOUNT_INSN_SIZE]; + ip = ktla_ktva(ip); + /* * Note: Due to modules and __init, code can * disappear and change, we need to protect against faulting @@ -284,7 +288,7 @@ int ftrace_update_ftrace_func(ftrace_fun unsigned char old[MCOUNT_INSN_SIZE], *new; int ret; - memcpy(old, &ftrace_call, MCOUNT_INSN_SIZE); + memcpy(old, (void *)ktla_ktva((unsigned long)ftrace_call), MCOUNT_INSN_SIZE); new = ftrace_call_replace(ip, (unsigned long)func); ret = ftrace_modify_code(ip, old, new); @@ -337,15 +341,15 @@ int __init ftrace_dyn_arch_init(void *da switch (faulted) { case 0: pr_info("ftrace: converting mcount calls to 0f 1f 44 00 00\n"); - memcpy(ftrace_nop, ftrace_test_p6nop, MCOUNT_INSN_SIZE); + memcpy(ftrace_nop, ktla_ktva(ftrace_test_p6nop), MCOUNT_INSN_SIZE); break; case 1: pr_info("ftrace: converting mcount calls to 66 66 66 66 90\n"); - memcpy(ftrace_nop, ftrace_test_nop5, MCOUNT_INSN_SIZE); + memcpy(ftrace_nop, ktla_ktva(ftrace_test_nop5), MCOUNT_INSN_SIZE); break; case 2: pr_info("ftrace: converting mcount calls to jmp . + 5\n"); - memcpy(ftrace_nop, ftrace_test_jmp, MCOUNT_INSN_SIZE); + memcpy(ftrace_nop, ktla_ktva(ftrace_test_jmp), MCOUNT_INSN_SIZE); break; } @@ -366,6 +370,8 @@ static int ftrace_mod_jmp(unsigned long { unsigned char code[MCOUNT_INSN_SIZE]; + ip = ktla_ktva(ip); + if (probe_kernel_read(code, (void *)ip, MCOUNT_INSN_SIZE)) return -EFAULT; diff -urNp linux-2.6.32.48/arch/x86/kernel/head32.c linux-2.6.32.48-openvz/arch/x86/kernel/head32.c --- linux-2.6.32.48/arch/x86/kernel/head32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/head32.c 2011-11-17 20:08:53.000000000 -0500 @@ -16,6 +16,7 @@ #include #include #include +#include static void __init i386_default_early_setup(void) { @@ -31,7 +32,7 @@ void __init i386_start_kernel(void) { reserve_trampoline_memory(); - reserve_early(__pa_symbol(&_text), __pa_symbol(&__bss_stop), "TEXT DATA BSS"); + reserve_early(LOAD_PHYSICAL_ADDR, __pa_symbol(&__bss_stop), "TEXT DATA BSS"); #ifdef CONFIG_BLK_DEV_INITRD /* Reserve INITRD */ diff -urNp linux-2.6.32.48/arch/x86/kernel/head_32.S linux-2.6.32.48-openvz/arch/x86/kernel/head_32.S --- linux-2.6.32.48/arch/x86/kernel/head_32.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/head_32.S 2011-11-17 20:09:04.000000000 -0500 @@ -19,10 +19,17 @@ #include #include #include +#include /* Physical address */ #define pa(X) ((X) - __PAGE_OFFSET) +#ifdef CONFIG_PAX_KERNEXEC +#define ta(X) (X) +#else +#define ta(X) ((X) - __PAGE_OFFSET) +#endif + /* * References to members of the new_cpu_data structure. */ @@ -52,11 +59,7 @@ * and small than max_low_pfn, otherwise will waste some page table entries */ -#if PTRS_PER_PMD > 1 -#define PAGE_TABLE_SIZE(pages) (((pages) / PTRS_PER_PMD) + PTRS_PER_PGD) -#else -#define PAGE_TABLE_SIZE(pages) ((pages) / PTRS_PER_PGD) -#endif +#define PAGE_TABLE_SIZE(pages) ((pages) / PTRS_PER_PTE) /* Enough space to fit pagetables for the low memory linear map */ MAPPING_BEYOND_END = \ @@ -73,6 +76,12 @@ INIT_MAP_SIZE = PAGE_TABLE_SIZE(KERNEL_P RESERVE_BRK(pagetables, INIT_MAP_SIZE) /* + * Real beginning of normal "text" segment + */ +ENTRY(stext) +ENTRY(_stext) + +/* * 32-bit kernel entrypoint; only used by the boot CPU. On entry, * %esi points to the real-mode code as a 32-bit pointer. * CS and DS must be 4 GB flat segments, but we don't depend on @@ -80,7 +89,16 @@ RESERVE_BRK(pagetables, INIT_MAP_SIZE) * can. */ __HEAD + +#ifdef CONFIG_PAX_KERNEXEC + jmp startup_32 +/* PaX: fill first page in .text with int3 to catch NULL derefs in kernel mode */ +.fill PAGE_SIZE-5,1,0xcc +#endif + ENTRY(startup_32) + movl pa(stack_start),%ecx + /* test KEEP_SEGMENTS flag to see if the bootloader is asking us to not reload segments */ testb $(1<<6), BP_loadflags(%esi) @@ -95,7 +113,60 @@ ENTRY(startup_32) movl %eax,%es movl %eax,%fs movl %eax,%gs + movl %eax,%ss 2: + leal -__PAGE_OFFSET(%ecx),%esp + +#ifdef CONFIG_SMP + movl $pa(cpu_gdt_table),%edi + movl $__per_cpu_load,%eax + movw %ax,__KERNEL_PERCPU + 2(%edi) + rorl $16,%eax + movb %al,__KERNEL_PERCPU + 4(%edi) + movb %ah,__KERNEL_PERCPU + 7(%edi) + movl $__per_cpu_end - 1,%eax + subl $__per_cpu_start,%eax + movw %ax,__KERNEL_PERCPU + 0(%edi) +#endif + +#ifdef CONFIG_PAX_MEMORY_UDEREF + movl $NR_CPUS,%ecx + movl $pa(cpu_gdt_table),%edi +1: + movl $((((__PAGE_OFFSET-1) & 0xf0000000) >> 12) | 0x00c09700),GDT_ENTRY_KERNEL_DS * 8 + 4(%edi) + movl $((((__PAGE_OFFSET-1) & 0xf0000000) >> 12) | 0x00c0fb00),GDT_ENTRY_DEFAULT_USER_CS * 8 + 4(%edi) + movl $((((__PAGE_OFFSET-1) & 0xf0000000) >> 12) | 0x00c0f300),GDT_ENTRY_DEFAULT_USER_DS * 8 + 4(%edi) + addl $PAGE_SIZE_asm,%edi + loop 1b +#endif + +#ifdef CONFIG_PAX_KERNEXEC + movl $pa(boot_gdt),%edi + movl $__LOAD_PHYSICAL_ADDR,%eax + movw %ax,__BOOT_CS + 2(%edi) + rorl $16,%eax + movb %al,__BOOT_CS + 4(%edi) + movb %ah,__BOOT_CS + 7(%edi) + rorl $16,%eax + + ljmp $(__BOOT_CS),$1f +1: + + movl $NR_CPUS,%ecx + movl $pa(cpu_gdt_table),%edi + addl $__PAGE_OFFSET,%eax +1: + movw %ax,__KERNEL_CS + 2(%edi) + movw %ax,__KERNEXEC_KERNEL_CS + 2(%edi) + rorl $16,%eax + movb %al,__KERNEL_CS + 4(%edi) + movb %al,__KERNEXEC_KERNEL_CS + 4(%edi) + movb %ah,__KERNEL_CS + 7(%edi) + movb %ah,__KERNEXEC_KERNEL_CS + 7(%edi) + rorl $16,%eax + addl $PAGE_SIZE_asm,%edi + loop 1b +#endif /* * Clear BSS first so that there are no surprises... @@ -140,9 +211,7 @@ ENTRY(startup_32) cmpl $num_subarch_entries, %eax jae bad_subarch - movl pa(subarch_entries)(,%eax,4), %eax - subl $__PAGE_OFFSET, %eax - jmp *%eax + jmp *pa(subarch_entries)(,%eax,4) bad_subarch: WEAK(lguest_entry) @@ -154,10 +223,10 @@ WEAK(xen_entry) __INITDATA subarch_entries: - .long default_entry /* normal x86/PC */ - .long lguest_entry /* lguest hypervisor */ - .long xen_entry /* Xen hypervisor */ - .long default_entry /* Moorestown MID */ + .long ta(default_entry) /* normal x86/PC */ + .long ta(lguest_entry) /* lguest hypervisor */ + .long ta(xen_entry) /* Xen hypervisor */ + .long ta(default_entry) /* Moorestown MID */ num_subarch_entries = (. - subarch_entries) / 4 .previous #endif /* CONFIG_PARAVIRT */ @@ -218,8 +287,11 @@ default_entry: movl %eax, pa(max_pfn_mapped) /* Do early initialization of the fixmap area */ - movl $pa(swapper_pg_fixmap)+PDE_IDENT_ATTR,%eax - movl %eax,pa(swapper_pg_pmd+0x1000*KPMDS-8) +#ifdef CONFIG_COMPAT_VDSO + movl $pa(swapper_pg_fixmap)+PDE_IDENT_ATTR+_PAGE_USER,pa(swapper_pg_pmd+0x1000*KPMDS-8) +#else + movl $pa(swapper_pg_fixmap)+PDE_IDENT_ATTR,pa(swapper_pg_pmd+0x1000*KPMDS-8) +#endif #else /* Not PAE */ page_pde_offset = (__PAGE_OFFSET >> 20); @@ -249,8 +321,11 @@ page_pde_offset = (__PAGE_OFFSET >> 20); movl %eax, pa(max_pfn_mapped) /* Do early initialization of the fixmap area */ - movl $pa(swapper_pg_fixmap)+PDE_IDENT_ATTR,%eax - movl %eax,pa(swapper_pg_dir+0xffc) +#ifdef CONFIG_COMPAT_VDSO + movl $pa(swapper_pg_fixmap)+PDE_IDENT_ATTR+_PAGE_USER,pa(swapper_pg_dir+0xffc) +#else + movl $pa(swapper_pg_fixmap)+PDE_IDENT_ATTR,pa(swapper_pg_dir+0xffc) +#endif #endif jmp 3f /* @@ -272,6 +347,9 @@ ENTRY(startup_32_smp) movl %eax,%es movl %eax,%fs movl %eax,%gs + movl pa(stack_start),%ecx + movl %eax,%ss + leal -__PAGE_OFFSET(%ecx),%esp #endif /* CONFIG_SMP */ 3: @@ -297,6 +375,7 @@ ENTRY(startup_32_smp) orl %edx,%eax movl %eax,%cr4 +#ifdef CONFIG_X86_PAE btl $5, %eax # check if PAE is enabled jnc 6f @@ -305,6 +384,10 @@ ENTRY(startup_32_smp) cpuid cmpl $0x80000000, %eax jbe 6f + + /* Clear bogus XD_DISABLE bits */ + call verify_cpu + mov $0x80000001, %eax cpuid /* Execute Disable bit supported? */ @@ -312,13 +395,17 @@ ENTRY(startup_32_smp) jnc 6f /* Setup EFER (Extended Feature Enable Register) */ - movl $0xc0000080, %ecx + movl $MSR_EFER, %ecx rdmsr btsl $11, %eax /* Make changes effective */ wrmsr + btsl $_PAGE_BIT_NX-32,pa(__supported_pte_mask+4) + movl $1,pa(nx_enabled) +#endif + 6: /* @@ -331,8 +418,8 @@ ENTRY(startup_32_smp) movl %eax,%cr0 /* ..and set paging (PG) bit */ ljmp $__BOOT_CS,$1f /* Clear prefetch and normalize %eip */ 1: - /* Set up the stack pointer */ - lss stack_start,%esp + /* Shift the stack pointer to a virtual address */ + addl $__PAGE_OFFSET, %esp /* * Initialize eflags. Some BIOS's leave bits like NT set. This would @@ -344,9 +431,7 @@ ENTRY(startup_32_smp) #ifdef CONFIG_SMP cmpb $0, ready - jz 1f /* Initial CPU cleans BSS */ - jmp checkCPUtype -1: + jnz checkCPUtype #endif /* CONFIG_SMP */ /* @@ -424,7 +509,7 @@ is386: movl $2,%ecx # set MP 1: movl $(__KERNEL_DS),%eax # reload all the segment registers movl %eax,%ss # after changing gdt. - movl $(__USER_DS),%eax # DS/ES contains default USER segment +# movl $(__KERNEL_DS),%eax # DS/ES contains default KERNEL segment movl %eax,%ds movl %eax,%es @@ -438,15 +523,22 @@ is386: movl $2,%ecx # set MP */ cmpb $0,ready jne 1f - movl $per_cpu__gdt_page,%eax + movl $cpu_gdt_table,%eax movl $per_cpu__stack_canary,%ecx +#ifdef CONFIG_SMP + addl $__per_cpu_load,%ecx +#endif movw %cx, 8 * GDT_ENTRY_STACK_CANARY + 2(%eax) shrl $16, %ecx movb %cl, 8 * GDT_ENTRY_STACK_CANARY + 4(%eax) movb %ch, 8 * GDT_ENTRY_STACK_CANARY + 7(%eax) 1: -#endif movl $(__KERNEL_STACK_CANARY),%eax +#elif defined(CONFIG_PAX_MEMORY_UDEREF) + movl $(__USER_DS),%eax +#else + xorl %eax,%eax +#endif movl %eax,%gs xorl %eax,%eax # Clear LDT @@ -454,14 +546,7 @@ is386: movl $2,%ecx # set MP cld # gcc2 wants the direction flag cleared at all times pushl $0 # fake return address for unwinder -#ifdef CONFIG_SMP - movb ready, %cl movb $1, ready - cmpb $0,%cl # the first CPU calls start_kernel - je 1f - movl (stack_start), %esp -1: -#endif /* CONFIG_SMP */ jmp *(initial_code) /* @@ -546,22 +631,22 @@ early_page_fault: jmp early_fault early_fault: - cld #ifdef CONFIG_PRINTK + cmpl $1,%ss:early_recursion_flag + je hlt_loop + incl %ss:early_recursion_flag + cld pusha movl $(__KERNEL_DS),%eax movl %eax,%ds movl %eax,%es - cmpl $2,early_recursion_flag - je hlt_loop - incl early_recursion_flag movl %cr2,%eax pushl %eax pushl %edx /* trapno */ pushl $fault_msg call printk +; call dump_stack #endif - call dump_stack hlt_loop: hlt jmp hlt_loop @@ -569,8 +654,11 @@ hlt_loop: /* This is the default interrupt "handler" :-) */ ALIGN ignore_int: - cld #ifdef CONFIG_PRINTK + cmpl $2,%ss:early_recursion_flag + je hlt_loop + incl %ss:early_recursion_flag + cld pushl %eax pushl %ecx pushl %edx @@ -579,9 +667,6 @@ ignore_int: movl $(__KERNEL_DS),%eax movl %eax,%ds movl %eax,%es - cmpl $2,early_recursion_flag - je hlt_loop - incl early_recursion_flag pushl 16(%esp) pushl 24(%esp) pushl 32(%esp) @@ -600,6 +685,8 @@ ignore_int: #endif iret +#include "verify_cpu.S" + __REFDATA .align 4 ENTRY(initial_code) @@ -610,31 +697,47 @@ ENTRY(initial_page_table) /* * BSS section */ -__PAGE_ALIGNED_BSS - .align PAGE_SIZE_asm #ifdef CONFIG_X86_PAE +.section .swapper_pg_pmd,"a",@progbits swapper_pg_pmd: .fill 1024*KPMDS,4,0 #else +.section .swapper_pg_dir,"a",@progbits ENTRY(swapper_pg_dir) .fill 1024,4,0 #endif +.section .swapper_pg_fixmap,"a",@progbits swapper_pg_fixmap: .fill 1024,4,0 #ifdef CONFIG_X86_TRAMPOLINE +.section .trampoline_pg_dir,"a",@progbits ENTRY(trampoline_pg_dir) +#ifdef CONFIG_X86_PAE + .fill 4,8,0 +#else .fill 1024,4,0 #endif +#endif + +.section .empty_zero_page,"a",@progbits ENTRY(empty_zero_page) .fill 4096,1,0 /* + * The IDT has to be page-aligned to simplify the Pentium + * F0 0F bug workaround.. We have a special link segment + * for this. + */ +.section .idt,"a",@progbits +ENTRY(idt_table) + .fill 256,8,0 + +/* * This starts the data section. */ #ifdef CONFIG_X86_PAE -__PAGE_ALIGNED_DATA - /* Page-aligned for the benefit of paravirt? */ - .align PAGE_SIZE_asm +.section .swapper_pg_dir,"a",@progbits + ENTRY(swapper_pg_dir) .long pa(swapper_pg_pmd+PGD_IDENT_ATTR),0 /* low identity map */ # if KPMDS == 3 @@ -653,15 +756,24 @@ ENTRY(swapper_pg_dir) # error "Kernel PMDs should be 1, 2 or 3" # endif .align PAGE_SIZE_asm /* needs to be page-sized too */ + +#ifdef CONFIG_PAX_PER_CPU_PGD +ENTRY(cpu_pgd) + .rept NR_CPUS + .fill 4,8,0 + .endr +#endif + #endif .data +.balign 4 ENTRY(stack_start) - .long init_thread_union+THREAD_SIZE - .long __BOOT_DS + .long init_thread_union+THREAD_SIZE-8 ready: .byte 0 +.section .rodata,"a",@progbits early_recursion_flag: .long 0 @@ -697,7 +809,7 @@ fault_msg: .word 0 # 32 bit align gdt_desc.address boot_gdt_descr: .word __BOOT_DS+7 - .long boot_gdt - __PAGE_OFFSET + .long pa(boot_gdt) .word 0 # 32-bit align idt_desc.address idt_descr: @@ -708,7 +820,7 @@ idt_descr: .word 0 # 32 bit align gdt_desc.address ENTRY(early_gdt_descr) .word GDT_ENTRIES*8-1 - .long per_cpu__gdt_page /* Overwritten for secondary CPUs */ + .long cpu_gdt_table /* Overwritten for secondary CPUs */ /* * The boot_gdt must mirror the equivalent in setup.S and is @@ -717,5 +829,65 @@ ENTRY(early_gdt_descr) .align L1_CACHE_BYTES ENTRY(boot_gdt) .fill GDT_ENTRY_BOOT_CS,8,0 - .quad 0x00cf9a000000ffff /* kernel 4GB code at 0x00000000 */ - .quad 0x00cf92000000ffff /* kernel 4GB data at 0x00000000 */ + .quad 0x00cf9b000000ffff /* kernel 4GB code at 0x00000000 */ + .quad 0x00cf93000000ffff /* kernel 4GB data at 0x00000000 */ + + .align PAGE_SIZE_asm +ENTRY(cpu_gdt_table) + .rept NR_CPUS + .quad 0x0000000000000000 /* NULL descriptor */ + .quad 0x0000000000000000 /* 0x0b reserved */ + .quad 0x0000000000000000 /* 0x13 reserved */ + .quad 0x0000000000000000 /* 0x1b reserved */ + +#ifdef CONFIG_PAX_KERNEXEC + .quad 0x00cf9b000000ffff /* 0x20 alternate kernel 4GB code at 0x00000000 */ +#else + .quad 0x0000000000000000 /* 0x20 unused */ +#endif + + .quad 0x0000000000000000 /* 0x28 unused */ + .quad 0x0000000000000000 /* 0x33 TLS entry 1 */ + .quad 0x0000000000000000 /* 0x3b TLS entry 2 */ + .quad 0x0000000000000000 /* 0x43 TLS entry 3 */ + .quad 0x0000000000000000 /* 0x4b reserved */ + .quad 0x0000000000000000 /* 0x53 reserved */ + .quad 0x0000000000000000 /* 0x5b reserved */ + + .quad 0x00cf9b000000ffff /* 0x60 kernel 4GB code at 0x00000000 */ + .quad 0x00cf93000000ffff /* 0x68 kernel 4GB data at 0x00000000 */ + .quad 0x00cffb000000ffff /* 0x73 user 4GB code at 0x00000000 */ + .quad 0x00cff3000000ffff /* 0x7b user 4GB data at 0x00000000 */ + + .quad 0x0000000000000000 /* 0x80 TSS descriptor */ + .quad 0x0000000000000000 /* 0x88 LDT descriptor */ + + /* + * Segments used for calling PnP BIOS have byte granularity. + * The code segments and data segments have fixed 64k limits, + * the transfer segment sizes are set at run time. + */ + .quad 0x00409b000000ffff /* 0x90 32-bit code */ + .quad 0x00009b000000ffff /* 0x98 16-bit code */ + .quad 0x000093000000ffff /* 0xa0 16-bit data */ + .quad 0x0000930000000000 /* 0xa8 16-bit data */ + .quad 0x0000930000000000 /* 0xb0 16-bit data */ + + /* + * The APM segments have byte granularity and their bases + * are set at run time. All have 64k limits. + */ + .quad 0x00409b000000ffff /* 0xb8 APM CS code */ + .quad 0x00009b000000ffff /* 0xc0 APM CS 16 code (16 bit) */ + .quad 0x004093000000ffff /* 0xc8 APM DS data */ + + .quad 0x00c0930000000000 /* 0xd0 - ESPFIX SS */ + .quad 0x0040930000000000 /* 0xd8 - PERCPU */ + .quad 0x0040910000000017 /* 0xe0 - STACK_CANARY */ + .quad 0x0000000000000000 /* 0xe8 - PCIBIOS_CS */ + .quad 0x0000000000000000 /* 0xf0 - PCIBIOS_DS */ + .quad 0x0000000000000000 /* 0xf8 - GDT entry 31: double-fault TSS */ + + /* Be sure this is zeroed to avoid false validations in Xen */ + .fill PAGE_SIZE_asm - GDT_SIZE,1,0 + .endr diff -urNp linux-2.6.32.48/arch/x86/kernel/head_64.S linux-2.6.32.48-openvz/arch/x86/kernel/head_64.S --- linux-2.6.32.48/arch/x86/kernel/head_64.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/head_64.S 2011-11-17 20:08:53.000000000 -0500 @@ -19,6 +19,7 @@ #include #include #include +#include #ifdef CONFIG_PARAVIRT #include @@ -38,6 +39,10 @@ L4_PAGE_OFFSET = pgd_index(__PAGE_OFFSET L3_PAGE_OFFSET = pud_index(__PAGE_OFFSET) L4_START_KERNEL = pgd_index(__START_KERNEL_map) L3_START_KERNEL = pud_index(__START_KERNEL_map) +L4_VMALLOC_START = pgd_index(VMALLOC_START) +L3_VMALLOC_START = pud_index(VMALLOC_START) +L4_VMEMMAP_START = pgd_index(VMEMMAP_START) +L3_VMEMMAP_START = pud_index(VMEMMAP_START) .text __HEAD @@ -85,35 +90,22 @@ startup_64: */ addq %rbp, init_level4_pgt + 0(%rip) addq %rbp, init_level4_pgt + (L4_PAGE_OFFSET*8)(%rip) + addq %rbp, init_level4_pgt + (L4_VMALLOC_START*8)(%rip) + addq %rbp, init_level4_pgt + (L4_VMEMMAP_START*8)(%rip) addq %rbp, init_level4_pgt + (L4_START_KERNEL*8)(%rip) addq %rbp, level3_ident_pgt + 0(%rip) +#ifndef CONFIG_XEN + addq %rbp, level3_ident_pgt + 8(%rip) +#endif - addq %rbp, level3_kernel_pgt + (510*8)(%rip) - addq %rbp, level3_kernel_pgt + (511*8)(%rip) + addq %rbp, level3_vmemmap_pgt + (L3_VMEMMAP_START*8)(%rip) - addq %rbp, level2_fixmap_pgt + (506*8)(%rip) + addq %rbp, level3_kernel_pgt + (L3_START_KERNEL*8)(%rip) + addq %rbp, level3_kernel_pgt + (L3_START_KERNEL*8+8)(%rip) - /* Add an Identity mapping if I am above 1G */ - leaq _text(%rip), %rdi - andq $PMD_PAGE_MASK, %rdi - - movq %rdi, %rax - shrq $PUD_SHIFT, %rax - andq $(PTRS_PER_PUD - 1), %rax - jz ident_complete - - leaq (level2_spare_pgt - __START_KERNEL_map + _KERNPG_TABLE)(%rbp), %rdx - leaq level3_ident_pgt(%rip), %rbx - movq %rdx, 0(%rbx, %rax, 8) - - movq %rdi, %rax - shrq $PMD_SHIFT, %rax - andq $(PTRS_PER_PMD - 1), %rax - leaq __PAGE_KERNEL_IDENT_LARGE_EXEC(%rdi), %rdx - leaq level2_spare_pgt(%rip), %rbx - movq %rdx, 0(%rbx, %rax, 8) -ident_complete: + addq %rbp, level2_fixmap_pgt + (506*8)(%rip) + addq %rbp, level2_fixmap_pgt + (507*8)(%rip) /* * Fixup the kernel text+data virtual addresses. Note that @@ -161,8 +153,8 @@ ENTRY(secondary_startup_64) * after the boot processor executes this code. */ - /* Enable PAE mode and PGE */ - movl $(X86_CR4_PAE | X86_CR4_PGE), %eax + /* Enable PAE mode and PSE/PGE */ + movl $(X86_CR4_PSE | X86_CR4_PAE | X86_CR4_PGE), %eax movq %rax, %cr4 /* Setup early boot stage 4 level pagetables. */ @@ -184,9 +176,13 @@ ENTRY(secondary_startup_64) movl $MSR_EFER, %ecx rdmsr btsl $_EFER_SCE, %eax /* Enable System Call */ - btl $20,%edi /* No Execute supported? */ + btl $(X86_FEATURE_NX & 31),%edi /* No Execute supported? */ jnc 1f btsl $_EFER_NX, %eax + leaq init_level4_pgt(%rip), %rdi + btsq $_PAGE_BIT_NX, 8*L4_PAGE_OFFSET(%rdi) + btsq $_PAGE_BIT_NX, 8*L4_VMALLOC_START(%rdi) + btsq $_PAGE_BIT_NX, 8*L4_VMEMMAP_START(%rdi) 1: wrmsr /* Make changes effective */ /* Setup cr0 */ @@ -262,16 +258,16 @@ ENTRY(secondary_startup_64) .quad x86_64_start_kernel ENTRY(initial_gs) .quad INIT_PER_CPU_VAR(irq_stack_union) - __FINITDATA ENTRY(stack_start) .quad init_thread_union+THREAD_SIZE-8 .word 0 + __FINITDATA bad_address: jmp bad_address - .section ".init.text","ax" + __INIT #ifdef CONFIG_EARLY_PRINTK .globl early_idt_handlers early_idt_handlers: @@ -316,18 +312,23 @@ ENTRY(early_idt_handler) #endif /* EARLY_PRINTK */ 1: hlt jmp 1b + .previous #ifdef CONFIG_EARLY_PRINTK + __INITDATA early_recursion_flag: .long 0 + .previous + .section .rodata,"a",@progbits early_idt_msg: .asciz "PANIC: early exception %02lx rip %lx:%lx error %lx cr2 %lx\n" early_idt_ripmsg: .asciz "RIP %s\n" -#endif /* CONFIG_EARLY_PRINTK */ .previous +#endif /* CONFIG_EARLY_PRINTK */ + .section .rodata,"a",@progbits #define NEXT_PAGE(name) \ .balign PAGE_SIZE; \ ENTRY(name) @@ -350,13 +351,36 @@ NEXT_PAGE(init_level4_pgt) .quad level3_ident_pgt - __START_KERNEL_map + _KERNPG_TABLE .org init_level4_pgt + L4_PAGE_OFFSET*8, 0 .quad level3_ident_pgt - __START_KERNEL_map + _KERNPG_TABLE + .org init_level4_pgt + L4_VMALLOC_START*8, 0 + .quad level3_vmalloc_pgt - __START_KERNEL_map + _KERNPG_TABLE + .org init_level4_pgt + L4_VMEMMAP_START*8, 0 + .quad level3_vmemmap_pgt - __START_KERNEL_map + _KERNPG_TABLE .org init_level4_pgt + L4_START_KERNEL*8, 0 /* (2^48-(2*1024*1024*1024))/(2^39) = 511 */ .quad level3_kernel_pgt - __START_KERNEL_map + _PAGE_TABLE +#ifdef CONFIG_PAX_PER_CPU_PGD +NEXT_PAGE(cpu_pgd) + .rept NR_CPUS + .fill 512,8,0 + .endr +#endif + NEXT_PAGE(level3_ident_pgt) .quad level2_ident_pgt - __START_KERNEL_map + _KERNPG_TABLE +#ifdef CONFIG_XEN .fill 511,8,0 +#else + .quad level2_ident_pgt + PAGE_SIZE - __START_KERNEL_map + _KERNPG_TABLE + .fill 510,8,0 +#endif + +NEXT_PAGE(level3_vmalloc_pgt) + .fill 512,8,0 + +NEXT_PAGE(level3_vmemmap_pgt) + .fill L3_VMEMMAP_START,8,0 + .quad level2_vmemmap_pgt - __START_KERNEL_map + _KERNPG_TABLE NEXT_PAGE(level3_kernel_pgt) .fill L3_START_KERNEL,8,0 @@ -364,20 +388,23 @@ NEXT_PAGE(level3_kernel_pgt) .quad level2_kernel_pgt - __START_KERNEL_map + _KERNPG_TABLE .quad level2_fixmap_pgt - __START_KERNEL_map + _PAGE_TABLE +NEXT_PAGE(level2_vmemmap_pgt) + .fill 512,8,0 + NEXT_PAGE(level2_fixmap_pgt) - .fill 506,8,0 - .quad level1_fixmap_pgt - __START_KERNEL_map + _PAGE_TABLE - /* 8MB reserved for vsyscalls + a 2MB hole = 4 + 1 entries */ - .fill 5,8,0 + .fill 507,8,0 + .quad level1_vsyscall_pgt - __START_KERNEL_map + _PAGE_TABLE + /* 6MB reserved for vsyscalls + a 2MB hole = 3 + 1 entries */ + .fill 4,8,0 -NEXT_PAGE(level1_fixmap_pgt) +NEXT_PAGE(level1_vsyscall_pgt) .fill 512,8,0 -NEXT_PAGE(level2_ident_pgt) - /* Since I easily can, map the first 1G. + /* Since I easily can, map the first 2G. * Don't set NX because code runs from these pages. */ - PMDS(0, __PAGE_KERNEL_IDENT_LARGE_EXEC, PTRS_PER_PMD) +NEXT_PAGE(level2_ident_pgt) + PMDS(0, __PAGE_KERNEL_IDENT_LARGE_EXEC, 2*PTRS_PER_PMD) NEXT_PAGE(level2_kernel_pgt) /* @@ -390,33 +417,55 @@ NEXT_PAGE(level2_kernel_pgt) * If you want to increase this then increase MODULES_VADDR * too.) */ - PMDS(0, __PAGE_KERNEL_LARGE_EXEC, - KERNEL_IMAGE_SIZE/PMD_SIZE) - -NEXT_PAGE(level2_spare_pgt) - .fill 512, 8, 0 + PMDS(0, __PAGE_KERNEL_LARGE_EXEC, KERNEL_IMAGE_SIZE/PMD_SIZE) #undef PMDS #undef NEXT_PAGE - .data + .align PAGE_SIZE +ENTRY(cpu_gdt_table) + .rept NR_CPUS + .quad 0x0000000000000000 /* NULL descriptor */ + .quad 0x00cf9b000000ffff /* __KERNEL32_CS */ + .quad 0x00af9b000000ffff /* __KERNEL_CS */ + .quad 0x00cf93000000ffff /* __KERNEL_DS */ + .quad 0x00cffb000000ffff /* __USER32_CS */ + .quad 0x00cff3000000ffff /* __USER_DS, __USER32_DS */ + .quad 0x00affb000000ffff /* __USER_CS */ + +#ifdef CONFIG_PAX_KERNEXEC + .quad 0x00af9b000000ffff /* __KERNEXEC_KERNEL_CS */ +#else + .quad 0x0 /* unused */ +#endif + + .quad 0,0 /* TSS */ + .quad 0,0 /* LDT */ + .quad 0,0,0 /* three TLS descriptors */ + .quad 0x0000f40000000000 /* node/CPU stored in limit */ + /* asm/segment.h:GDT_ENTRIES must match this */ + + /* zero the remaining page */ + .fill PAGE_SIZE / 8 - GDT_ENTRIES,8,0 + .endr + .align 16 .globl early_gdt_descr early_gdt_descr: .word GDT_ENTRIES*8-1 early_gdt_descr_base: - .quad INIT_PER_CPU_VAR(gdt_page) + .quad cpu_gdt_table ENTRY(phys_base) /* This must match the first entry in level2_kernel_pgt */ .quad 0x0000000000000000 #include "../../x86/xen/xen-head.S" - - .section .bss, "aw", @nobits + + .section .rodata,"a",@progbits .align L1_CACHE_BYTES ENTRY(idt_table) - .skip IDT_ENTRIES * 16 + .fill 512,8,0 __PAGE_ALIGNED_BSS .align PAGE_SIZE diff -urNp linux-2.6.32.48/arch/x86/kernel/i386_ksyms_32.c linux-2.6.32.48-openvz/arch/x86/kernel/i386_ksyms_32.c --- linux-2.6.32.48/arch/x86/kernel/i386_ksyms_32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/i386_ksyms_32.c 2011-11-17 20:08:53.000000000 -0500 @@ -20,8 +20,12 @@ extern void cmpxchg8b_emu(void); EXPORT_SYMBOL(cmpxchg8b_emu); #endif +EXPORT_SYMBOL_GPL(cpu_gdt_table); + /* Networking helper routines. */ EXPORT_SYMBOL(csum_partial_copy_generic); +EXPORT_SYMBOL(csum_partial_copy_generic_to_user); +EXPORT_SYMBOL(csum_partial_copy_generic_from_user); EXPORT_SYMBOL(__get_user_1); EXPORT_SYMBOL(__get_user_2); @@ -36,3 +40,7 @@ EXPORT_SYMBOL(strstr); EXPORT_SYMBOL(csum_partial); EXPORT_SYMBOL(empty_zero_page); + +#ifdef CONFIG_PAX_KERNEXEC +EXPORT_SYMBOL(__LOAD_PHYSICAL_ADDR); +#endif diff -urNp linux-2.6.32.48/arch/x86/kernel/i387.c linux-2.6.32.48-openvz/arch/x86/kernel/i387.c --- linux-2.6.32.48/arch/x86/kernel/i387.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/i387.c 2011-11-17 20:08:53.000000000 -0500 @@ -163,6 +163,7 @@ int init_fpu(struct task_struct *tsk) set_stopped_child_used_math(tsk); return 0; } +EXPORT_SYMBOL(init_fpu); int fpregs_active(struct task_struct *target, const struct user_regset *regset) { diff -urNp linux-2.6.32.48/arch/x86/kernel/i8259.c linux-2.6.32.48-openvz/arch/x86/kernel/i8259.c --- linux-2.6.32.48/arch/x86/kernel/i8259.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/i8259.c 2011-11-17 20:08:53.000000000 -0500 @@ -208,7 +208,7 @@ spurious_8259A_irq: "spurious 8259A interrupt: IRQ%d.\n", irq); spurious_irq_mask |= irqmask; } - atomic_inc(&irq_err_count); + atomic_inc_unchecked(&irq_err_count); /* * Theoretically we do not have to handle this IRQ, * but in Linux this does not cause problems and is diff -urNp linux-2.6.32.48/arch/x86/kernel/init_task.c linux-2.6.32.48-openvz/arch/x86/kernel/init_task.c --- linux-2.6.32.48/arch/x86/kernel/init_task.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/init_task.c 2011-11-17 20:08:53.000000000 -0500 @@ -20,8 +20,7 @@ static struct sighand_struct init_sighan * way process stacks are handled. This is done by having a special * "init_task" linker map entry.. */ -union thread_union init_thread_union __init_task_data = - { INIT_THREAD_INFO(init_task) }; +union thread_union init_thread_union __init_task_data; /* * Initial task structure. @@ -38,5 +37,5 @@ EXPORT_SYMBOL(init_task); * section. Since TSS's are completely CPU-local, we want them * on exact cacheline boundaries, to eliminate cacheline ping-pong. */ -DEFINE_PER_CPU_SHARED_ALIGNED(struct tss_struct, init_tss) = INIT_TSS; - +struct tss_struct init_tss[NR_CPUS] ____cacheline_internodealigned_in_smp = { [0 ... NR_CPUS-1] = INIT_TSS }; +EXPORT_SYMBOL(init_tss); diff -urNp linux-2.6.32.48/arch/x86/kernel/ioport.c linux-2.6.32.48-openvz/arch/x86/kernel/ioport.c --- linux-2.6.32.48/arch/x86/kernel/ioport.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/ioport.c 2011-11-17 20:08:53.000000000 -0500 @@ -6,6 +6,7 @@ #include #include #include +#include #include #include #include @@ -41,6 +42,12 @@ asmlinkage long sys_ioperm(unsigned long if ((from + num <= from) || (from + num > IO_BITMAP_BITS)) return -EINVAL; +#ifdef CONFIG_GRKERNSEC_IO + if (turn_on && grsec_disable_privio) { + gr_handle_ioperm(); + return -EPERM; + } +#endif if (turn_on && !capable(CAP_SYS_RAWIO)) return -EPERM; @@ -67,7 +74,7 @@ asmlinkage long sys_ioperm(unsigned long * because the ->io_bitmap_max value must match the bitmap * contents: */ - tss = &per_cpu(init_tss, get_cpu()); + tss = init_tss + get_cpu(); set_bitmap(t->io_bitmap_ptr, from, num, !turn_on); @@ -111,6 +118,12 @@ static int do_iopl(unsigned int level, s return -EINVAL; /* Trying to gain more privileges? */ if (level > old) { +#ifdef CONFIG_GRKERNSEC_IO + if (grsec_disable_privio) { + gr_handle_iopl(); + return -EPERM; + } +#endif if (!capable(CAP_SYS_RAWIO)) return -EPERM; } diff -urNp linux-2.6.32.48/arch/x86/kernel/irq_32.c linux-2.6.32.48-openvz/arch/x86/kernel/irq_32.c --- linux-2.6.32.48/arch/x86/kernel/irq_32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/irq_32.c 2011-11-17 20:09:04.000000000 -0500 @@ -35,7 +35,7 @@ static int check_stack_overflow(void) __asm__ __volatile__("andl %%esp,%0" : "=r" (sp) : "0" (THREAD_SIZE - 1)); - return sp < (sizeof(struct thread_info) + STACK_WARN); + return sp < STACK_WARN; } static void print_stack_overflow(void) @@ -54,9 +54,9 @@ static inline void print_stack_overflow( * per-CPU IRQ handling contexts (thread information and stack) */ union irq_ctx { - struct thread_info tinfo; - u32 stack[THREAD_SIZE/sizeof(u32)]; -} __attribute__((aligned(PAGE_SIZE))); + unsigned long previous_esp; + u32 stack[THREAD_SIZE/sizeof(u32)]; +} __attribute__((aligned(THREAD_SIZE))); static DEFINE_PER_CPU(union irq_ctx *, hardirq_ctx); static DEFINE_PER_CPU(union irq_ctx *, softirq_ctx); @@ -78,10 +78,9 @@ static void call_on_stack(void *func, vo static inline int execute_on_irq_stack(int overflow, struct irq_desc *desc, int irq) { - union irq_ctx *curctx, *irqctx; + union irq_ctx *irqctx; u32 *isp, arg1, arg2; - curctx = (union irq_ctx *) current_thread_info(); irqctx = __get_cpu_var(hardirq_ctx); /* @@ -90,21 +89,16 @@ execute_on_irq_stack(int overflow, struc * handler) we can't do that and just have to keep using the * current stack (which is the irq stack already after all) */ - if (unlikely(curctx == irqctx)) + if (unlikely((void *)current_stack_pointer - (void *)irqctx < THREAD_SIZE)) return 0; /* build the stack frame on the IRQ stack */ - isp = (u32 *) ((char *)irqctx + sizeof(*irqctx)); - irqctx->tinfo.task = curctx->tinfo.task; - irqctx->tinfo.previous_esp = current_stack_pointer; + isp = (u32 *) ((char *)irqctx + sizeof(*irqctx) - 8); + irqctx->previous_esp = current_stack_pointer; - /* - * Copy the softirq bits in preempt_count so that the - * softirq checks work in the hardirq context. - */ - irqctx->tinfo.preempt_count = - (irqctx->tinfo.preempt_count & ~SOFTIRQ_MASK) | - (curctx->tinfo.preempt_count & SOFTIRQ_MASK); +#ifdef CONFIG_PAX_MEMORY_UDEREF + __set_fs(MAKE_MM_SEG(0)); +#endif if (unlikely(overflow)) call_on_stack(print_stack_overflow, isp); @@ -116,6 +110,11 @@ execute_on_irq_stack(int overflow, struc : "0" (irq), "1" (desc), "2" (isp), "D" (desc->handle_irq) : "memory", "cc", "ecx"); + +#ifdef CONFIG_PAX_MEMORY_UDEREF + __set_fs(current_thread_info()->addr_limit); +#endif + return 1; } @@ -124,28 +123,11 @@ execute_on_irq_stack(int overflow, struc */ void __cpuinit irq_ctx_init(int cpu) { - union irq_ctx *irqctx; - if (per_cpu(hardirq_ctx, cpu)) return; - irqctx = &per_cpu(hardirq_stack, cpu); - irqctx->tinfo.task = NULL; - irqctx->tinfo.exec_domain = NULL; - irqctx->tinfo.cpu = cpu; - irqctx->tinfo.preempt_count = HARDIRQ_OFFSET; - irqctx->tinfo.addr_limit = MAKE_MM_SEG(0); - - per_cpu(hardirq_ctx, cpu) = irqctx; - - irqctx = &per_cpu(softirq_stack, cpu); - irqctx->tinfo.task = NULL; - irqctx->tinfo.exec_domain = NULL; - irqctx->tinfo.cpu = cpu; - irqctx->tinfo.preempt_count = 0; - irqctx->tinfo.addr_limit = MAKE_MM_SEG(0); - - per_cpu(softirq_ctx, cpu) = irqctx; + per_cpu(hardirq_ctx, cpu) = &per_cpu(hardirq_stack, cpu); + per_cpu(softirq_ctx, cpu) = &per_cpu(softirq_stack, cpu); printk(KERN_DEBUG "CPU %u irqstacks, hard=%p soft=%p\n", cpu, per_cpu(hardirq_ctx, cpu), per_cpu(softirq_ctx, cpu)); @@ -159,7 +141,6 @@ void irq_ctx_exit(int cpu) asmlinkage void do_softirq(void) { unsigned long flags; - struct thread_info *curctx; union irq_ctx *irqctx; u32 *isp; @@ -169,15 +150,22 @@ asmlinkage void do_softirq(void) local_irq_save(flags); if (local_softirq_pending()) { - curctx = current_thread_info(); irqctx = __get_cpu_var(softirq_ctx); - irqctx->tinfo.task = curctx->task; - irqctx->tinfo.previous_esp = current_stack_pointer; + irqctx->previous_esp = current_stack_pointer; /* build the stack frame on the softirq stack */ - isp = (u32 *) ((char *)irqctx + sizeof(*irqctx)); + isp = (u32 *) ((char *)irqctx + sizeof(*irqctx) - 8); + +#ifdef CONFIG_PAX_MEMORY_UDEREF + __set_fs(MAKE_MM_SEG(0)); +#endif call_on_stack(__do_softirq, isp); + +#ifdef CONFIG_PAX_MEMORY_UDEREF + __set_fs(current_thread_info()->addr_limit); +#endif + /* * Shouldnt happen, we returned above if in_interrupt(): */ diff -urNp linux-2.6.32.48/arch/x86/kernel/irq.c linux-2.6.32.48-openvz/arch/x86/kernel/irq.c --- linux-2.6.32.48/arch/x86/kernel/irq.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/irq.c 2011-11-17 20:08:53.000000000 -0500 @@ -15,7 +15,7 @@ #include #include -atomic_t irq_err_count; +atomic_unchecked_t irq_err_count; /* Function pointer for generic interrupt vector handling */ void (*generic_interrupt_extension)(void) = NULL; @@ -114,9 +114,9 @@ static int show_other_interrupts(struct seq_printf(p, "%10u ", per_cpu(mce_poll_count, j)); seq_printf(p, " Machine check polls\n"); #endif - seq_printf(p, "%*s: %10u\n", prec, "ERR", atomic_read(&irq_err_count)); + seq_printf(p, "%*s: %10u\n", prec, "ERR", atomic_read_unchecked(&irq_err_count)); #if defined(CONFIG_X86_IO_APIC) - seq_printf(p, "%*s: %10u\n", prec, "MIS", atomic_read(&irq_mis_count)); + seq_printf(p, "%*s: %10u\n", prec, "MIS", atomic_read_unchecked(&irq_mis_count)); #endif return 0; } @@ -209,10 +209,10 @@ u64 arch_irq_stat_cpu(unsigned int cpu) u64 arch_irq_stat(void) { - u64 sum = atomic_read(&irq_err_count); + u64 sum = atomic_read_unchecked(&irq_err_count); #ifdef CONFIG_X86_IO_APIC - sum += atomic_read(&irq_mis_count); + sum += atomic_read_unchecked(&irq_mis_count); #endif return sum; } diff -urNp linux-2.6.32.48/arch/x86/kernel/kgdb.c linux-2.6.32.48-openvz/arch/x86/kernel/kgdb.c --- linux-2.6.32.48/arch/x86/kernel/kgdb.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/kgdb.c 2011-11-17 20:08:53.000000000 -0500 @@ -390,13 +390,13 @@ int kgdb_arch_handle_exception(int e_vec /* clear the trace bit */ linux_regs->flags &= ~X86_EFLAGS_TF; - atomic_set(&kgdb_cpu_doing_single_step, -1); + atomic_set_unchecked(&kgdb_cpu_doing_single_step, -1); /* set the trace bit if we're stepping */ if (remcomInBuffer[0] == 's') { linux_regs->flags |= X86_EFLAGS_TF; kgdb_single_step = 1; - atomic_set(&kgdb_cpu_doing_single_step, + atomic_set_unchecked(&kgdb_cpu_doing_single_step, raw_smp_processor_id()); } @@ -476,7 +476,7 @@ static int __kgdb_notify(struct die_args break; case DIE_DEBUG: - if (atomic_read(&kgdb_cpu_doing_single_step) == + if (atomic_read_unchecked(&kgdb_cpu_doing_single_step) == raw_smp_processor_id()) { if (user_mode(regs)) return single_step_cont(regs, args); @@ -573,7 +573,7 @@ unsigned long kgdb_arch_pc(int exception return instruction_pointer(regs); } -struct kgdb_arch arch_kgdb_ops = { +const struct kgdb_arch arch_kgdb_ops = { /* Breakpoint instruction: */ .gdb_bpt_instr = { 0xcc }, .flags = KGDB_HW_BREAKPOINT, diff -urNp linux-2.6.32.48/arch/x86/kernel/kprobes.c linux-2.6.32.48-openvz/arch/x86/kernel/kprobes.c --- linux-2.6.32.48/arch/x86/kernel/kprobes.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/kprobes.c 2011-11-17 20:09:04.000000000 -0500 @@ -168,9 +168,13 @@ static void __kprobes set_jmp_op(void *f char op; s32 raddr; } __attribute__((packed)) * jop; - jop = (struct __arch_jmp_op *)from; + + jop = (struct __arch_jmp_op *)(ktla_ktva(from)); + + pax_open_kernel(); jop->raddr = (s32)((long)(to) - ((long)(from) + 5)); jop->op = RELATIVEJUMP_INSTRUCTION; + pax_close_kernel(); } /* @@ -195,7 +199,7 @@ static int __kprobes can_boost(kprobe_op kprobe_opcode_t opcode; kprobe_opcode_t *orig_opcodes = opcodes; - if (search_exception_tables((unsigned long)opcodes)) + if (search_exception_tables(ktva_ktla((unsigned long)opcodes))) return 0; /* Page fault may occur on this address. */ retry: @@ -339,7 +343,9 @@ static void __kprobes fix_riprel(struct disp = (u8 *) p->addr + *((s32 *) insn) - (u8 *) p->ainsn.insn; BUG_ON((s64) (s32) disp != disp); /* Sanity check. */ + pax_open_kernel(); *(s32 *)insn = (s32) disp; + pax_close_kernel(); } } #endif @@ -347,16 +353,18 @@ static void __kprobes fix_riprel(struct static void __kprobes arch_copy_kprobe(struct kprobe *p) { - memcpy(p->ainsn.insn, p->addr, MAX_INSN_SIZE * sizeof(kprobe_opcode_t)); + pax_open_kernel(); + memcpy(p->ainsn.insn, ktla_ktva(p->addr), MAX_INSN_SIZE * sizeof(kprobe_opcode_t)); + pax_close_kernel(); fix_riprel(p); - if (can_boost(p->addr)) + if (can_boost(ktla_ktva(p->addr))) p->ainsn.boostable = 0; else p->ainsn.boostable = -1; - p->opcode = *p->addr; + p->opcode = *(ktla_ktva(p->addr)); } int __kprobes arch_prepare_kprobe(struct kprobe *p) @@ -434,7 +442,7 @@ static void __kprobes prepare_singlestep if (p->opcode == BREAKPOINT_INSTRUCTION) regs->ip = (unsigned long)p->addr; else - regs->ip = (unsigned long)p->ainsn.insn; + regs->ip = ktva_ktla((unsigned long)p->ainsn.insn); } void __kprobes arch_prepare_kretprobe(struct kretprobe_instance *ri, @@ -455,7 +463,7 @@ static void __kprobes setup_singlestep(s if (p->ainsn.boostable == 1 && !p->post_handler) { /* Boost up -- we can execute copied instructions directly */ reset_current_kprobe(); - regs->ip = (unsigned long)p->ainsn.insn; + regs->ip = ktva_ktla((unsigned long)p->ainsn.insn); preempt_enable_no_resched(); return; } @@ -525,7 +533,7 @@ static int __kprobes kprobe_handler(stru struct kprobe_ctlblk *kcb; addr = (kprobe_opcode_t *)(regs->ip - sizeof(kprobe_opcode_t)); - if (*addr != BREAKPOINT_INSTRUCTION) { + if (*(kprobe_opcode_t *)ktla_ktva((unsigned long)addr) != BREAKPOINT_INSTRUCTION) { /* * The breakpoint instruction was removed right * after we hit it. Another cpu has removed @@ -777,7 +785,7 @@ static void __kprobes resume_execution(s struct pt_regs *regs, struct kprobe_ctlblk *kcb) { unsigned long *tos = stack_addr(regs); - unsigned long copy_ip = (unsigned long)p->ainsn.insn; + unsigned long copy_ip = ktva_ktla((unsigned long)p->ainsn.insn); unsigned long orig_ip = (unsigned long)p->addr; kprobe_opcode_t *insn = p->ainsn.insn; @@ -960,7 +968,7 @@ int __kprobes kprobe_exceptions_notify(s struct die_args *args = data; int ret = NOTIFY_DONE; - if (args->regs && user_mode_vm(args->regs)) + if (args->regs && user_mode(args->regs)) return ret; switch (val) { diff -urNp linux-2.6.32.48/arch/x86/kernel/kvm.c linux-2.6.32.48-openvz/arch/x86/kernel/kvm.c --- linux-2.6.32.48/arch/x86/kernel/kvm.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/kvm.c 2011-11-17 20:09:04.000000000 -0500 @@ -216,6 +216,7 @@ static void __init paravirt_ops_setup(vo pv_mmu_ops.set_pud = kvm_set_pud; #if PAGETABLE_LEVELS == 4 pv_mmu_ops.set_pgd = kvm_set_pgd; + pv_mmu_ops.set_pgd_batched = kvm_set_pgd; #endif #endif pv_mmu_ops.flush_tlb_user = kvm_flush_tlb; diff -urNp linux-2.6.32.48/arch/x86/kernel/ldt.c linux-2.6.32.48-openvz/arch/x86/kernel/ldt.c --- linux-2.6.32.48/arch/x86/kernel/ldt.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/ldt.c 2011-11-17 20:08:53.000000000 -0500 @@ -13,6 +13,8 @@ #include #include #include +#include +#include #include #include @@ -39,9 +41,9 @@ static int alloc_ldt(mm_context_t *pc, i mincount = (mincount + (PAGE_SIZE / LDT_ENTRY_SIZE - 1)) & (~(PAGE_SIZE / LDT_ENTRY_SIZE - 1)); if (mincount * LDT_ENTRY_SIZE > PAGE_SIZE) - newldt = vmalloc(mincount * LDT_ENTRY_SIZE); + newldt = ub_vmalloc(mincount * LDT_ENTRY_SIZE); else - newldt = (void *)__get_free_page(GFP_KERNEL); + newldt = (void *)__get_free_page(GFP_KERNEL_UBC); if (!newldt) return -ENOMEM; @@ -66,13 +68,13 @@ static int alloc_ldt(mm_context_t *pc, i if (reload) { #ifdef CONFIG_SMP preempt_disable(); - load_LDT(pc); + load_LDT_nolock(pc); if (!cpumask_equal(mm_cpumask(current->mm), cpumask_of(smp_processor_id()))) smp_call_function(flush_ldt, current->mm, 1); preempt_enable(); #else - load_LDT(pc); + load_LDT_nolock(pc); #endif } if (oldsize) { @@ -94,7 +96,7 @@ static inline int copy_ldt(mm_context_t return err; for (i = 0; i < old->size; i++) - write_ldt_entry(new->ldt, i, old->ldt + i * LDT_ENTRY_SIZE); + write_ldt_entry(new->ldt, i, old->ldt + i); return 0; } @@ -115,8 +117,27 @@ int init_new_context(struct task_struct retval = copy_ldt(&mm->context, &old_mm->context); mutex_unlock(&old_mm->context.lock); } + + if (tsk == current) { + mm->context.vdso = 0; + +#ifdef CONFIG_X86_32 +#if defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC) + mm->context.user_cs_base = 0UL; + mm->context.user_cs_limit = ~0UL; + +#if defined(CONFIG_PAX_PAGEEXEC) && defined(CONFIG_SMP) + cpus_clear(mm->context.cpu_user_cs_mask); +#endif + +#endif +#endif + + } + return retval; } +EXPORT_SYMBOL_GPL(init_new_context); /* * No need to lock the MM as we are the last user @@ -229,6 +250,13 @@ static int write_ldt(void __user *ptr, u } } +#ifdef CONFIG_PAX_SEGMEXEC + if ((mm->pax_flags & MF_PAX_SEGMEXEC) && (ldt_info.contents & MODIFY_LDT_CONTENTS_CODE)) { + error = -EINVAL; + goto out_unlock; + } +#endif + fill_ldt(&ldt, &ldt_info); if (oldmode) ldt.avl = 0; diff -urNp linux-2.6.32.48/arch/x86/kernel/machine_kexec_32.c linux-2.6.32.48-openvz/arch/x86/kernel/machine_kexec_32.c --- linux-2.6.32.48/arch/x86/kernel/machine_kexec_32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/machine_kexec_32.c 2011-11-17 20:08:53.000000000 -0500 @@ -26,7 +26,7 @@ #include #include -static void set_idt(void *newidt, __u16 limit) +static void set_idt(struct desc_struct *newidt, __u16 limit) { struct desc_ptr curidt; @@ -38,7 +38,7 @@ static void set_idt(void *newidt, __u16 } -static void set_gdt(void *newgdt, __u16 limit) +static void set_gdt(struct desc_struct *newgdt, __u16 limit) { struct desc_ptr curgdt; @@ -217,7 +217,7 @@ void machine_kexec(struct kimage *image) } control_page = page_address(image->control_code_page); - memcpy(control_page, relocate_kernel, KEXEC_CONTROL_CODE_MAX_SIZE); + memcpy(control_page, (void *)ktla_ktva((unsigned long)relocate_kernel), KEXEC_CONTROL_CODE_MAX_SIZE); relocate_kernel_ptr = control_page; page_list[PA_CONTROL_PAGE] = __pa(control_page); diff -urNp linux-2.6.32.48/arch/x86/kernel/microcode_amd.c linux-2.6.32.48-openvz/arch/x86/kernel/microcode_amd.c --- linux-2.6.32.48/arch/x86/kernel/microcode_amd.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/microcode_amd.c 2011-11-17 20:08:53.000000000 -0500 @@ -364,7 +364,7 @@ static void microcode_fini_cpu_amd(int c uci->mc = NULL; } -static struct microcode_ops microcode_amd_ops = { +static const struct microcode_ops microcode_amd_ops = { .request_microcode_user = request_microcode_user, .request_microcode_fw = request_microcode_fw, .collect_cpu_info = collect_cpu_info_amd, @@ -372,7 +372,7 @@ static struct microcode_ops microcode_am .microcode_fini_cpu = microcode_fini_cpu_amd, }; -struct microcode_ops * __init init_amd_microcode(void) +const struct microcode_ops * __init init_amd_microcode(void) { return µcode_amd_ops; } diff -urNp linux-2.6.32.48/arch/x86/kernel/microcode_core.c linux-2.6.32.48-openvz/arch/x86/kernel/microcode_core.c --- linux-2.6.32.48/arch/x86/kernel/microcode_core.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/microcode_core.c 2011-11-17 20:08:53.000000000 -0500 @@ -90,7 +90,7 @@ MODULE_LICENSE("GPL"); #define MICROCODE_VERSION "2.00" -static struct microcode_ops *microcode_ops; +static const struct microcode_ops *microcode_ops; /* * Synchronization. diff -urNp linux-2.6.32.48/arch/x86/kernel/microcode_intel.c linux-2.6.32.48-openvz/arch/x86/kernel/microcode_intel.c --- linux-2.6.32.48/arch/x86/kernel/microcode_intel.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/microcode_intel.c 2011-11-17 20:09:04.000000000 -0500 @@ -443,13 +443,13 @@ static enum ucode_state request_microcod static int get_ucode_user(void *to, const void *from, size_t n) { - return copy_from_user(to, from, n); + return copy_from_user(to, (const void __force_user *)from, n); } static enum ucode_state request_microcode_user(int cpu, const void __user *buf, size_t size) { - return generic_load_microcode(cpu, (void *)buf, size, &get_ucode_user); + return generic_load_microcode(cpu, (__force_kernel void *)buf, size, &get_ucode_user); } static void microcode_fini_cpu(int cpu) @@ -460,7 +460,7 @@ static void microcode_fini_cpu(int cpu) uci->mc = NULL; } -static struct microcode_ops microcode_intel_ops = { +static const struct microcode_ops microcode_intel_ops = { .request_microcode_user = request_microcode_user, .request_microcode_fw = request_microcode_fw, .collect_cpu_info = collect_cpu_info, @@ -468,7 +468,7 @@ static struct microcode_ops microcode_in .microcode_fini_cpu = microcode_fini_cpu, }; -struct microcode_ops * __init init_intel_microcode(void) +const struct microcode_ops * __init init_intel_microcode(void) { return µcode_intel_ops; } diff -urNp linux-2.6.32.48/arch/x86/kernel/module.c linux-2.6.32.48-openvz/arch/x86/kernel/module.c --- linux-2.6.32.48/arch/x86/kernel/module.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/module.c 2011-11-17 20:08:53.000000000 -0500 @@ -34,7 +34,7 @@ #define DEBUGP(fmt...) #endif -void *module_alloc(unsigned long size) +static void *__module_alloc(unsigned long size, pgprot_t prot) { struct vm_struct *area; @@ -48,8 +48,18 @@ void *module_alloc(unsigned long size) if (!area) return NULL; - return __vmalloc_area(area, GFP_KERNEL | __GFP_HIGHMEM, - PAGE_KERNEL_EXEC); + return __vmalloc_area(area, GFP_KERNEL | __GFP_HIGHMEM | __GFP_ZERO, prot); +} + +void *module_alloc(unsigned long size) +{ + +#ifdef CONFIG_PAX_KERNEXEC + return __module_alloc(size, PAGE_KERNEL); +#else + return __module_alloc(size, PAGE_KERNEL_EXEC); +#endif + } /* Free memory returned from module_alloc */ @@ -58,6 +68,40 @@ void module_free(struct module *mod, voi vfree(module_region); } +#ifdef CONFIG_PAX_KERNEXEC +#ifdef CONFIG_X86_32 +void *module_alloc_exec(unsigned long size) +{ + struct vm_struct *area; + + if (size == 0) + return NULL; + + area = __get_vm_area(size, VM_ALLOC, (unsigned long)&MODULES_EXEC_VADDR, (unsigned long)&MODULES_EXEC_END); + return area ? area->addr : NULL; +} +EXPORT_SYMBOL(module_alloc_exec); + +void module_free_exec(struct module *mod, void *module_region) +{ + vunmap(module_region); +} +EXPORT_SYMBOL(module_free_exec); +#else +void module_free_exec(struct module *mod, void *module_region) +{ + module_free(mod, module_region); +} +EXPORT_SYMBOL(module_free_exec); + +void *module_alloc_exec(unsigned long size) +{ + return __module_alloc(size, PAGE_KERNEL_RX); +} +EXPORT_SYMBOL(module_alloc_exec); +#endif +#endif + /* We don't need anything special. */ int module_frob_arch_sections(Elf_Ehdr *hdr, Elf_Shdr *sechdrs, @@ -77,14 +121,16 @@ int apply_relocate(Elf32_Shdr *sechdrs, unsigned int i; Elf32_Rel *rel = (void *)sechdrs[relsec].sh_addr; Elf32_Sym *sym; - uint32_t *location; + uint32_t *plocation, location; DEBUGP("Applying relocate section %u to %u\n", relsec, sechdrs[relsec].sh_info); for (i = 0; i < sechdrs[relsec].sh_size / sizeof(*rel); i++) { /* This is where to make the change */ - location = (void *)sechdrs[sechdrs[relsec].sh_info].sh_addr - + rel[i].r_offset; + plocation = (void *)sechdrs[sechdrs[relsec].sh_info].sh_addr + rel[i].r_offset; + location = (uint32_t)plocation; + if (sechdrs[sechdrs[relsec].sh_info].sh_flags & SHF_EXECINSTR) + plocation = ktla_ktva((void *)plocation); /* This is the symbol it is referring to. Note that all undefined symbols have been resolved. */ sym = (Elf32_Sym *)sechdrs[symindex].sh_addr @@ -93,11 +139,15 @@ int apply_relocate(Elf32_Shdr *sechdrs, switch (ELF32_R_TYPE(rel[i].r_info)) { case R_386_32: /* We add the value into the location given */ - *location += sym->st_value; + pax_open_kernel(); + *plocation += sym->st_value; + pax_close_kernel(); break; case R_386_PC32: /* Add the value, subtract its postition */ - *location += sym->st_value - (uint32_t)location; + pax_open_kernel(); + *plocation += sym->st_value - location; + pax_close_kernel(); break; default: printk(KERN_ERR "module %s: Unknown relocation: %u\n", @@ -153,21 +203,30 @@ int apply_relocate_add(Elf64_Shdr *sechd case R_X86_64_NONE: break; case R_X86_64_64: + pax_open_kernel(); *(u64 *)loc = val; + pax_close_kernel(); break; case R_X86_64_32: + pax_open_kernel(); *(u32 *)loc = val; + pax_close_kernel(); if (val != *(u32 *)loc) goto overflow; break; case R_X86_64_32S: + pax_open_kernel(); *(s32 *)loc = val; + pax_close_kernel(); if ((s64)val != *(s32 *)loc) goto overflow; break; case R_X86_64_PC32: val -= (u64)loc; + pax_open_kernel(); *(u32 *)loc = val; + pax_close_kernel(); + #if 0 if ((s64)val != *(s32 *)loc) goto overflow; diff -urNp linux-2.6.32.48/arch/x86/kernel/paravirt.c linux-2.6.32.48-openvz/arch/x86/kernel/paravirt.c --- linux-2.6.32.48/arch/x86/kernel/paravirt.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/paravirt.c 2011-11-17 20:09:04.000000000 -0500 @@ -53,6 +53,9 @@ u64 _paravirt_ident_64(u64 x) { return x; } +#if defined(CONFIG_X86_32) && defined(CONFIG_X86_PAE) +PV_CALLEE_SAVE_REGS_THUNK(_paravirt_ident_64); +#endif void __init default_banner(void) { @@ -122,7 +125,7 @@ unsigned paravirt_patch_jmp(void *insnbu * corresponding structure. */ static void *get_call_destination(u8 type) { - struct paravirt_patch_template tmpl = { + const struct paravirt_patch_template tmpl = { .pv_init_ops = pv_init_ops, .pv_time_ops = pv_time_ops, .pv_cpu_ops = pv_cpu_ops, @@ -133,6 +136,8 @@ static void *get_call_destination(u8 typ .pv_lock_ops = pv_lock_ops, #endif }; + + pax_track_stack(); return *((void **)&tmpl + type); } @@ -145,15 +150,19 @@ unsigned paravirt_patch_default(u8 type, if (opfunc == NULL) /* If there's no function, patch it with a ud2a (BUG) */ ret = paravirt_patch_insns(insnbuf, len, ud2a, ud2a+sizeof(ud2a)); - else if (opfunc == _paravirt_nop) + else if (opfunc == (void *)_paravirt_nop) /* If the operation is a nop, then nop the callsite */ ret = paravirt_patch_nop(); /* identity functions just return their single argument */ - else if (opfunc == _paravirt_ident_32) + else if (opfunc == (void *)_paravirt_ident_32) ret = paravirt_patch_ident_32(insnbuf, len); - else if (opfunc == _paravirt_ident_64) + else if (opfunc == (void *)_paravirt_ident_64) + ret = paravirt_patch_ident_64(insnbuf, len); +#if defined(CONFIG_X86_32) && defined(CONFIG_X86_PAE) + else if (opfunc == (void *)__raw_callee_save__paravirt_ident_64) ret = paravirt_patch_ident_64(insnbuf, len); +#endif else if (type == PARAVIRT_PATCH(pv_cpu_ops.iret) || type == PARAVIRT_PATCH(pv_cpu_ops.irq_enable_sysexit) || @@ -178,7 +187,7 @@ unsigned paravirt_patch_insns(void *insn if (insn_len > len || start == NULL) insn_len = len; else - memcpy(insnbuf, start, insn_len); + memcpy(insnbuf, ktla_ktva(start), insn_len); return insn_len; } @@ -294,22 +303,22 @@ void arch_flush_lazy_mmu_mode(void) preempt_enable(); } -struct pv_info pv_info = { +struct pv_info pv_info __read_only = { .name = "bare hardware", .paravirt_enabled = 0, .kernel_rpl = 0, .shared_kernel_pmd = 1, /* Only used when CONFIG_X86_PAE is set */ }; -struct pv_init_ops pv_init_ops = { +struct pv_init_ops pv_init_ops __read_only = { .patch = native_patch, }; -struct pv_time_ops pv_time_ops = { +struct pv_time_ops pv_time_ops __read_only = { .sched_clock = native_sched_clock, }; -struct pv_irq_ops pv_irq_ops = { +struct pv_irq_ops pv_irq_ops __read_only = { .save_fl = __PV_IS_CALLEE_SAVE(native_save_fl), .restore_fl = __PV_IS_CALLEE_SAVE(native_restore_fl), .irq_disable = __PV_IS_CALLEE_SAVE(native_irq_disable), @@ -321,7 +330,7 @@ struct pv_irq_ops pv_irq_ops = { #endif }; -struct pv_cpu_ops pv_cpu_ops = { +struct pv_cpu_ops pv_cpu_ops __read_only = { .cpuid = native_cpuid, .get_debugreg = native_get_debugreg, .set_debugreg = native_set_debugreg, @@ -382,21 +391,26 @@ struct pv_cpu_ops pv_cpu_ops = { .end_context_switch = paravirt_nop, }; -struct pv_apic_ops pv_apic_ops = { +struct pv_apic_ops pv_apic_ops __read_only = { #ifdef CONFIG_X86_LOCAL_APIC .startup_ipi_hook = paravirt_nop, #endif }; -#if defined(CONFIG_X86_32) && !defined(CONFIG_X86_PAE) +#ifdef CONFIG_X86_32 +#ifdef CONFIG_X86_PAE +/* 64-bit pagetable entries */ +#define PTE_IDENT PV_CALLEE_SAVE(_paravirt_ident_64) +#else /* 32-bit pagetable entries */ #define PTE_IDENT __PV_IS_CALLEE_SAVE(_paravirt_ident_32) +#endif #else /* 64-bit pagetable entries */ #define PTE_IDENT __PV_IS_CALLEE_SAVE(_paravirt_ident_64) #endif -struct pv_mmu_ops pv_mmu_ops = { +struct pv_mmu_ops pv_mmu_ops __read_only = { .read_cr2 = native_read_cr2, .write_cr2 = native_write_cr2, @@ -448,6 +462,7 @@ struct pv_mmu_ops pv_mmu_ops = { .make_pud = PTE_IDENT, .set_pgd = native_set_pgd, + .set_pgd_batched = native_set_pgd_batched, #endif #endif /* PAGETABLE_LEVELS >= 3 */ @@ -467,6 +482,12 @@ struct pv_mmu_ops pv_mmu_ops = { }, .set_fixmap = native_set_fixmap, + +#ifdef CONFIG_PAX_KERNEXEC + .pax_open_kernel = native_pax_open_kernel, + .pax_close_kernel = native_pax_close_kernel, +#endif + }; EXPORT_SYMBOL_GPL(pv_time_ops); diff -urNp linux-2.6.32.48/arch/x86/kernel/paravirt-spinlocks.c linux-2.6.32.48-openvz/arch/x86/kernel/paravirt-spinlocks.c --- linux-2.6.32.48/arch/x86/kernel/paravirt-spinlocks.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/paravirt-spinlocks.c 2011-11-17 20:08:53.000000000 -0500 @@ -13,7 +13,7 @@ default_spin_lock_flags(raw_spinlock_t * __raw_spin_lock(lock); } -struct pv_lock_ops pv_lock_ops = { +struct pv_lock_ops pv_lock_ops __read_only = { #ifdef CONFIG_SMP .spin_is_locked = __ticket_spin_is_locked, .spin_is_contended = __ticket_spin_is_contended, diff -urNp linux-2.6.32.48/arch/x86/kernel/pci-calgary_64.c linux-2.6.32.48-openvz/arch/x86/kernel/pci-calgary_64.c --- linux-2.6.32.48/arch/x86/kernel/pci-calgary_64.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/pci-calgary_64.c 2011-11-17 20:08:53.000000000 -0500 @@ -477,7 +477,7 @@ static void calgary_free_coherent(struct free_pages((unsigned long)vaddr, get_order(size)); } -static struct dma_map_ops calgary_dma_ops = { +static const struct dma_map_ops calgary_dma_ops = { .alloc_coherent = calgary_alloc_coherent, .free_coherent = calgary_free_coherent, .map_sg = calgary_map_sg, diff -urNp linux-2.6.32.48/arch/x86/kernel/pci-dma.c linux-2.6.32.48-openvz/arch/x86/kernel/pci-dma.c --- linux-2.6.32.48/arch/x86/kernel/pci-dma.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/pci-dma.c 2011-11-17 20:08:53.000000000 -0500 @@ -14,7 +14,7 @@ static int forbid_dac __read_mostly; -struct dma_map_ops *dma_ops; +const struct dma_map_ops *dma_ops; EXPORT_SYMBOL(dma_ops); static int iommu_sac_force __read_mostly; @@ -243,7 +243,7 @@ early_param("iommu", iommu_setup); int dma_supported(struct device *dev, u64 mask) { - struct dma_map_ops *ops = get_dma_ops(dev); + const struct dma_map_ops *ops = get_dma_ops(dev); #ifdef CONFIG_PCI if (mask > 0xffffffff && forbid_dac > 0) { diff -urNp linux-2.6.32.48/arch/x86/kernel/pci-gart_64.c linux-2.6.32.48-openvz/arch/x86/kernel/pci-gart_64.c --- linux-2.6.32.48/arch/x86/kernel/pci-gart_64.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/pci-gart_64.c 2011-11-17 20:08:53.000000000 -0500 @@ -682,7 +682,7 @@ static __init int init_k8_gatt(struct ag return -1; } -static struct dma_map_ops gart_dma_ops = { +static const struct dma_map_ops gart_dma_ops = { .map_sg = gart_map_sg, .unmap_sg = gart_unmap_sg, .map_page = gart_map_page, diff -urNp linux-2.6.32.48/arch/x86/kernel/pci-nommu.c linux-2.6.32.48-openvz/arch/x86/kernel/pci-nommu.c --- linux-2.6.32.48/arch/x86/kernel/pci-nommu.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/pci-nommu.c 2011-11-17 20:08:53.000000000 -0500 @@ -94,7 +94,7 @@ static void nommu_sync_sg_for_device(str flush_write_buffers(); } -struct dma_map_ops nommu_dma_ops = { +const struct dma_map_ops nommu_dma_ops = { .alloc_coherent = dma_generic_alloc_coherent, .free_coherent = nommu_free_coherent, .map_sg = nommu_map_sg, diff -urNp linux-2.6.32.48/arch/x86/kernel/pci-swiotlb.c linux-2.6.32.48-openvz/arch/x86/kernel/pci-swiotlb.c --- linux-2.6.32.48/arch/x86/kernel/pci-swiotlb.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/pci-swiotlb.c 2011-11-17 20:08:53.000000000 -0500 @@ -25,7 +25,7 @@ static void *x86_swiotlb_alloc_coherent( return swiotlb_alloc_coherent(hwdev, size, dma_handle, flags); } -static struct dma_map_ops swiotlb_dma_ops = { +static const struct dma_map_ops swiotlb_dma_ops = { .mapping_error = swiotlb_dma_mapping_error, .alloc_coherent = x86_swiotlb_alloc_coherent, .free_coherent = swiotlb_free_coherent, diff -urNp linux-2.6.32.48/arch/x86/kernel/process_32.c linux-2.6.32.48-openvz/arch/x86/kernel/process_32.c --- linux-2.6.32.48/arch/x86/kernel/process_32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/process_32.c 2011-11-17 20:09:04.000000000 -0500 @@ -40,6 +40,8 @@ #include #include #include +#include +#include #include #include @@ -60,6 +62,9 @@ #include asmlinkage void ret_from_fork(void) __asm__("ret_from_fork"); +EXPORT_SYMBOL(ret_from_fork); +asmlinkage void i386_ret_from_resume(void) __asm__("i386_ret_from_resume"); +EXPORT_SYMBOL_GPL(i386_ret_from_resume); /* * Return saved PC of a blocked thread. @@ -67,6 +72,7 @@ asmlinkage void ret_from_fork(void) __as unsigned long thread_saved_pc(struct task_struct *tsk) { return ((unsigned long *)tsk->thread.sp)[3]; +//XXX return tsk->thread.eip; } #ifndef CONFIG_SMP @@ -129,31 +135,31 @@ void __show_regs(struct pt_regs *regs, i unsigned short ss, gs; const char *board; - if (user_mode_vm(regs)) { + if (user_mode(regs)) { sp = regs->sp; ss = regs->ss & 0xffff; - gs = get_user_gs(regs); } else { sp = (unsigned long) (®s->sp); savesegment(ss, ss); - savesegment(gs, gs); } + gs = get_user_gs(regs); printk("\n"); board = dmi_get_system_info(DMI_PRODUCT_NAME); if (!board) board = ""; - printk("Pid: %d, comm: %s %s (%s %.*s) %s\n", + printk("Pid: %d, comm: %s %s (%s %.*s) %s %s)\n", task_pid_nr(current), current->comm, print_tainted(), init_utsname()->release, (int)strcspn(init_utsname()->version, " "), - init_utsname()->version, board); + init_utsname()->version, VZVERSION, board); printk("EIP: %04x:[<%08lx>] EFLAGS: %08lx CPU: %d\n", (u16)regs->cs, regs->ip, regs->flags, smp_processor_id()); - print_symbol("EIP is at %s\n", regs->ip); + if (decode_call_traces) + print_symbol("EIP is at %s\n", regs->ip); printk("EAX: %08lx EBX: %08lx ECX: %08lx EDX: %08lx\n", regs->ax, regs->bx, regs->cx, regs->dx); @@ -189,6 +195,8 @@ void show_regs(struct pt_regs *regs) { __show_regs(regs, 1); show_trace(NULL, regs, ®s->sp, regs->bp); + if (!decode_call_traces) + printk(" EIP: [<%08lx>]\n", regs->ip); } /* @@ -197,6 +205,7 @@ void show_regs(struct pt_regs *regs) * the "args". */ extern void kernel_thread_helper(void); +EXPORT_SYMBOL(kernel_thread_helper); /* * Create a kernel thread @@ -205,15 +214,22 @@ int kernel_thread(int (*fn)(void *), voi { struct pt_regs regs; + /* Don't allow kernel_thread() inside VE */ + if (!ve_allow_kthreads && !ve_is_super(get_exec_env())) { + printk("kernel_thread call inside container\n"); + dump_stack(); + return -EPERM; + } + memset(®s, 0, sizeof(regs)); regs.bx = (unsigned long) fn; regs.dx = (unsigned long) arg; - regs.ds = __USER_DS; - regs.es = __USER_DS; + regs.ds = __KERNEL_DS; + regs.es = __KERNEL_DS; regs.fs = __KERNEL_PERCPU; - regs.gs = __KERNEL_STACK_CANARY; + savesegment(gs, regs.gs); regs.orig_ax = -1; regs.ip = (unsigned long) kernel_thread_helper; regs.cs = __KERNEL_CS | get_kernel_rpl(); @@ -247,13 +263,14 @@ int copy_thread(unsigned long clone_flag struct task_struct *tsk; int err; - childregs = task_pt_regs(p); + childregs = task_stack_page(p) + THREAD_SIZE - sizeof(struct pt_regs) - 8; *childregs = *regs; childregs->ax = 0; childregs->sp = sp; p->thread.sp = (unsigned long) childregs; p->thread.sp0 = (unsigned long) (childregs+1); + p->tinfo.lowest_stack = (unsigned long)task_stack_page(p); p->thread.ip = (unsigned long) ret_from_fork; @@ -345,7 +362,7 @@ __switch_to(struct task_struct *prev_p, struct thread_struct *prev = &prev_p->thread, *next = &next_p->thread; int cpu = smp_processor_id(); - struct tss_struct *tss = &per_cpu(init_tss, cpu); + struct tss_struct *tss = init_tss + cpu; bool preload_fpu; /* never put a printk in __switch_to... printk() calls wake_up*() indirectly */ @@ -380,6 +397,10 @@ __switch_to(struct task_struct *prev_p, */ lazy_save_gs(prev->gs); +#ifdef CONFIG_PAX_MEMORY_UDEREF + __set_fs(task_thread_info(next_p)->addr_limit); +#endif + /* * Load the per-thread Thread-Local Storage descriptor. */ @@ -415,6 +436,9 @@ __switch_to(struct task_struct *prev_p, */ arch_end_context_switch(next_p); + percpu_write(current_task, next_p); + percpu_write(current_tinfo, &next_p->tinfo); + if (preload_fpu) __math_state_restore(); @@ -424,8 +448,6 @@ __switch_to(struct task_struct *prev_p, if (prev->gs | next->gs) lazy_load_gs(next->gs); - percpu_write(current_task, next_p); - return prev_p; } @@ -495,4 +517,3 @@ unsigned long get_wchan(struct task_stru } while (count++ < 16); return 0; } - diff -urNp linux-2.6.32.48/arch/x86/kernel/process_64.c linux-2.6.32.48-openvz/arch/x86/kernel/process_64.c --- linux-2.6.32.48/arch/x86/kernel/process_64.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/process_64.c 2011-11-17 20:09:04.000000000 -0500 @@ -25,8 +25,10 @@ #include #include #include +#include #include #include +#include #include #include #include @@ -53,8 +55,6 @@ #include #include -asmlinkage extern void ret_from_fork(void); - DEFINE_PER_CPU(unsigned long, old_rsp); static DEFINE_PER_CPU(unsigned char, is_idle); @@ -91,7 +91,7 @@ static void __exit_idle(void) void exit_idle(void) { /* idle loop has pid 0 */ - if (current->pid) + if (task_pid_nr(current)) return; __exit_idle(); } @@ -169,13 +169,14 @@ void __show_regs(struct pt_regs *regs, i board = dmi_get_system_info(DMI_PRODUCT_NAME); if (!board) board = ""; - printk(KERN_INFO "Pid: %d, comm: %.20s %s %s %.*s %s\n", - current->pid, current->comm, print_tainted(), + printk(KERN_INFO "Pid: %d, comm: %.20s %s %s %.*s %s %s\n", + task_pid_nr(current), current->comm, print_tainted(), init_utsname()->release, (int)strcspn(init_utsname()->version, " "), - init_utsname()->version, board); + init_utsname()->version, VZVERSION, board); printk(KERN_INFO "RIP: %04lx:[<%016lx>] ", regs->cs & 0xffff, regs->ip); - printk_address(regs->ip, 1); + if (decode_call_traces) + printk_address(regs->ip, 1); printk(KERN_INFO "RSP: %04lx:%016lx EFLAGS: %08lx\n", regs->ss, regs->sp, regs->flags); printk(KERN_INFO "RAX: %016lx RBX: %016lx RCX: %016lx\n", @@ -228,7 +229,9 @@ void show_regs(struct pt_regs *regs) { printk(KERN_INFO "CPU %d:", smp_processor_id()); __show_regs(regs, 1); - show_trace(NULL, regs, (void *)(regs + 1), regs->bp); + show_trace(NULL, regs, ®s->sp, regs->bp); + if (!decode_call_traces) + printk(" EIP: [<%08lx>]\n", regs->ip); } void release_thread(struct task_struct *dead_task) @@ -280,8 +283,7 @@ int copy_thread(unsigned long clone_flag struct pt_regs *childregs; struct task_struct *me = current; - childregs = ((struct pt_regs *) - (THREAD_SIZE + task_stack_page(p))) - 1; + childregs = task_stack_page(p) + THREAD_SIZE - sizeof(struct pt_regs) - 16; *childregs = *regs; childregs->ax = 0; @@ -292,6 +294,7 @@ int copy_thread(unsigned long clone_flag p->thread.sp = (unsigned long) childregs; p->thread.sp0 = (unsigned long) (childregs+1); p->thread.usersp = me->thread.usersp; + p->tinfo.lowest_stack = (unsigned long)task_stack_page(p); set_tsk_thread_flag(p, TIF_FORK); @@ -379,7 +382,7 @@ __switch_to(struct task_struct *prev_p, struct thread_struct *prev = &prev_p->thread; struct thread_struct *next = &next_p->thread; int cpu = smp_processor_id(); - struct tss_struct *tss = &per_cpu(init_tss, cpu); + struct tss_struct *tss = init_tss + cpu; unsigned fsindex, gsindex; bool preload_fpu; @@ -475,10 +478,9 @@ __switch_to(struct task_struct *prev_p, prev->usersp = percpu_read(old_rsp); percpu_write(old_rsp, next->usersp); percpu_write(current_task, next_p); + percpu_write(current_tinfo, &next_p->tinfo); - percpu_write(kernel_stack, - (unsigned long)task_stack_page(next_p) + - THREAD_SIZE - KERNEL_STACK_OFFSET); + percpu_write(kernel_stack, next->sp0); /* * Now maybe reload the debug registers and handle I/O bitmaps @@ -559,12 +561,11 @@ unsigned long get_wchan(struct task_stru if (!p || p == current || p->state == TASK_RUNNING) return 0; stack = (unsigned long)task_stack_page(p); - if (p->thread.sp < stack || p->thread.sp >= stack+THREAD_SIZE) + if (p->thread.sp < stack || p->thread.sp > stack+THREAD_SIZE-16-sizeof(u64)) return 0; fp = *(u64 *)(p->thread.sp); do { - if (fp < (unsigned long)stack || - fp >= (unsigned long)stack+THREAD_SIZE) + if (fp < stack || fp > stack+THREAD_SIZE-16-sizeof(u64)) return 0; ip = *(u64 *)(fp+8); if (!in_sched_functions(ip)) @@ -679,3 +680,20 @@ unsigned long KSTK_ESP(struct task_struc return (test_tsk_thread_flag(task, TIF_IA32)) ? (task_pt_regs(task)->sp) : ((task)->thread.usersp); } + +long do_fork_kthread(unsigned long clone_flags, + unsigned long stack_start, + struct pt_regs *regs, + unsigned long stack_size, + int __user *parent_tidptr, + int __user *child_tidptr) +{ + if (ve_allow_kthreads || ve_is_super(get_exec_env())) + return do_fork(clone_flags, stack_start, regs, stack_size, + parent_tidptr, child_tidptr); + + /* Don't allow kernel_thread() inside VE */ + printk("kernel_thread call inside container\n"); + dump_stack(); + return -EPERM; +} diff -urNp linux-2.6.32.48/arch/x86/kernel/process.c linux-2.6.32.48-openvz/arch/x86/kernel/process.c --- linux-2.6.32.48/arch/x86/kernel/process.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/process.c 2011-11-17 20:09:04.000000000 -0500 @@ -51,16 +51,33 @@ void free_thread_xstate(struct task_stru void free_thread_info(struct thread_info *ti) { - free_thread_xstate(ti->task); free_pages((unsigned long)ti, get_order(THREAD_SIZE)); } +static struct kmem_cache *task_struct_cachep; + void arch_task_cache_init(void) { - task_xstate_cachep = - kmem_cache_create("task_xstate", xstate_size, + /* create a slab on which task_structs can be allocated */ + task_struct_cachep = + kmem_cache_create("task_struct", sizeof(struct task_struct), + ARCH_MIN_TASKALIGN, SLAB_PANIC | SLAB_NOTRACK, NULL); + + task_xstate_cachep = + kmem_cache_create("task_xstate", xstate_size, __alignof__(union thread_xstate), - SLAB_PANIC | SLAB_NOTRACK, NULL); + SLAB_PANIC | SLAB_NOTRACK | SLAB_USERCOPY, NULL); +} + +struct task_struct *alloc_task_struct(void) +{ + return kmem_cache_alloc(task_struct_cachep, GFP_KERNEL); +} + +void free_task_struct(struct task_struct *task) +{ + free_thread_xstate(task); + kmem_cache_free(task_struct_cachep, task); } /* @@ -73,7 +90,7 @@ void exit_thread(void) unsigned long *bp = t->io_bitmap_ptr; if (bp) { - struct tss_struct *tss = &per_cpu(init_tss, get_cpu()); + struct tss_struct *tss = init_tss + get_cpu(); t->io_bitmap_ptr = NULL; clear_thread_flag(TIF_IO_BITMAP); @@ -93,6 +110,9 @@ void flush_thread(void) clear_tsk_thread_flag(tsk, TIF_DEBUG); +#if defined(CONFIG_X86_32) && !defined(CONFIG_CC_STACKPROTECTOR) && !defined(CONFIG_PAX_MEMORY_UDEREF) + loadsegment(gs, 0); +#endif tsk->thread.debugreg0 = 0; tsk->thread.debugreg1 = 0; tsk->thread.debugreg2 = 0; @@ -568,16 +588,38 @@ static int __init idle_setup(char *str) } early_param("idle", idle_setup); -unsigned long arch_align_stack(unsigned long sp) +#ifdef CONFIG_PAX_RANDKSTACK +void pax_randomize_kstack(struct pt_regs *regs) { - if (!(current->personality & ADDR_NO_RANDOMIZE) && randomize_va_space) - sp -= get_random_int() % 8192; - return sp & ~0xf; -} + struct thread_struct *thread = ¤t->thread; + unsigned long time; -unsigned long arch_randomize_brk(struct mm_struct *mm) -{ - unsigned long range_end = mm->brk + 0x02000000; - return randomize_range(mm->brk, range_end, 0) ? : mm->brk; + if (!randomize_va_space) + return; + + if (v8086_mode(regs)) + return; + + rdtscl(time); + + /* P4 seems to return a 0 LSB, ignore it */ +#ifdef CONFIG_MPENTIUM4 + time &= 0x3EUL; + time <<= 2; +#elif defined(CONFIG_X86_64) + time &= 0xFUL; + time <<= 4; +#else + time &= 0x1FUL; + time <<= 3; +#endif + + thread->sp0 ^= time; + load_sp0(init_tss + smp_processor_id(), thread); + +#ifdef CONFIG_X86_64 + percpu_write(kernel_stack, thread->sp0); +#endif } +#endif diff -urNp linux-2.6.32.48/arch/x86/kernel/ptrace.c linux-2.6.32.48-openvz/arch/x86/kernel/ptrace.c --- linux-2.6.32.48/arch/x86/kernel/ptrace.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/ptrace.c 2011-11-17 20:08:53.000000000 -0500 @@ -925,7 +925,7 @@ static const struct user_regset_view use long arch_ptrace(struct task_struct *child, long request, long addr, long data) { int ret; - unsigned long __user *datap = (unsigned long __user *)data; + unsigned long __user *datap = (__force unsigned long __user *)data; switch (request) { /* read the word at location addr in the USER area. */ @@ -1012,14 +1012,14 @@ long arch_ptrace(struct task_struct *chi if (addr < 0) return -EIO; ret = do_get_thread_area(child, addr, - (struct user_desc __user *) data); + (__force struct user_desc __user *) data); break; case PTRACE_SET_THREAD_AREA: if (addr < 0) return -EIO; ret = do_set_thread_area(child, addr, - (struct user_desc __user *) data, 0); + (__force struct user_desc __user *) data, 0); break; #endif @@ -1038,12 +1038,12 @@ long arch_ptrace(struct task_struct *chi #ifdef CONFIG_X86_PTRACE_BTS case PTRACE_BTS_CONFIG: ret = ptrace_bts_config - (child, data, (struct ptrace_bts_config __user *)addr); + (child, data, (__force struct ptrace_bts_config __user *)addr); break; case PTRACE_BTS_STATUS: ret = ptrace_bts_status - (child, data, (struct ptrace_bts_config __user *)addr); + (child, data, (__force struct ptrace_bts_config __user *)addr); break; case PTRACE_BTS_SIZE: @@ -1052,7 +1052,7 @@ long arch_ptrace(struct task_struct *chi case PTRACE_BTS_GET: ret = ptrace_bts_read_record - (child, data, (struct bts_struct __user *) addr); + (child, data, (__force struct bts_struct __user *) addr); break; case PTRACE_BTS_CLEAR: @@ -1061,7 +1061,7 @@ long arch_ptrace(struct task_struct *chi case PTRACE_BTS_DRAIN: ret = ptrace_bts_drain - (child, data, (struct bts_struct __user *) addr); + (child, data, (__force struct bts_struct __user *) addr); break; #endif /* CONFIG_X86_PTRACE_BTS */ @@ -1450,7 +1450,7 @@ void send_sigtrap(struct task_struct *ts info.si_code = si_code; /* User-mode ip? */ - info.si_addr = user_mode_vm(regs) ? (void __user *) regs->ip : NULL; + info.si_addr = user_mode(regs) ? (__force void __user *) regs->ip : NULL; /* Send us the fake SIGTRAP */ force_sig_info(SIGTRAP, &info, tsk); @@ -1469,7 +1469,7 @@ void send_sigtrap(struct task_struct *ts * We must return the syscall number to actually look up in the table. * This can be -1L to skip running any syscall at all. */ -asmregparm long syscall_trace_enter(struct pt_regs *regs) +long syscall_trace_enter(struct pt_regs *regs) { long ret = 0; @@ -1514,7 +1514,7 @@ asmregparm long syscall_trace_enter(stru return ret ?: regs->orig_ax; } -asmregparm void syscall_trace_leave(struct pt_regs *regs) +void syscall_trace_leave(struct pt_regs *regs) { if (unlikely(current->audit_context)) audit_syscall_exit(AUDITSC_RESULT(regs->ax), regs->ax); diff -urNp linux-2.6.32.48/arch/x86/kernel/reboot.c linux-2.6.32.48-openvz/arch/x86/kernel/reboot.c --- linux-2.6.32.48/arch/x86/kernel/reboot.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/reboot.c 2011-11-17 20:09:04.000000000 -0500 @@ -33,7 +33,7 @@ void (*pm_power_off)(void); EXPORT_SYMBOL(pm_power_off); static const struct desc_ptr no_idt = {}; -static int reboot_mode; +static unsigned short reboot_mode; enum reboot_type reboot_type = BOOT_KBD; int reboot_force; @@ -292,12 +292,12 @@ core_initcall(reboot_init); controller to pulse the CPU reset line, which is more thorough, but doesn't work with at least one type of 486 motherboard. It is easy to stop this code working; hence the copious comments. */ -static const unsigned long long -real_mode_gdt_entries [3] = +static struct desc_struct +real_mode_gdt_entries [3] __read_only = { - 0x0000000000000000ULL, /* Null descriptor */ - 0x00009b000000ffffULL, /* 16-bit real-mode 64k code at 0x00000000 */ - 0x000093000100ffffULL /* 16-bit real-mode 64k data at 0x00000100 */ + GDT_ENTRY_INIT(0, 0, 0), /* Null descriptor */ + GDT_ENTRY_INIT(0x9b, 0, 0xffff), /* 16-bit real-mode 64k code at 0x00000000 */ + GDT_ENTRY_INIT(0x93, 0x100, 0xffff) /* 16-bit real-mode 64k data at 0x00000100 */ }; static const struct desc_ptr @@ -346,7 +346,7 @@ static const unsigned char jump_to_bios * specified by the code and length parameters. * We assume that length will aways be less that 100! */ -void machine_real_restart(const unsigned char *code, int length) +void machine_real_restart(const unsigned char *code, unsigned int length) { local_irq_disable(); @@ -366,8 +366,8 @@ void machine_real_restart(const unsigned /* Remap the kernel at virtual address zero, as well as offset zero from the kernel segment. This assumes the kernel segment starts at virtual address PAGE_OFFSET. */ - memcpy(swapper_pg_dir, swapper_pg_dir + KERNEL_PGD_BOUNDARY, - sizeof(swapper_pg_dir [0]) * KERNEL_PGD_PTRS); + clone_pgd_range(swapper_pg_dir, swapper_pg_dir + KERNEL_PGD_BOUNDARY, + min_t(unsigned long, KERNEL_PGD_PTRS, KERNEL_PGD_BOUNDARY)); /* * Use `swapper_pg_dir' as our page directory. @@ -379,16 +379,15 @@ void machine_real_restart(const unsigned boot)". This seems like a fairly standard thing that gets set by REBOOT.COM programs, and the previous reset routine did this too. */ - *((unsigned short *)0x472) = reboot_mode; + *(unsigned short *)(__va(0x472)) = reboot_mode; /* For the switch to real mode, copy some code to low memory. It has to be in the first 64k because it is running in 16-bit mode, and it has to have the same physical and virtual address, because it turns off paging. Copy it near the end of the first page, out of the way of BIOS variables. */ - memcpy((void *)(0x1000 - sizeof(real_mode_switch) - 100), - real_mode_switch, sizeof (real_mode_switch)); - memcpy((void *)(0x1000 - 100), code, length); + memcpy(__va(0x1000 - sizeof (real_mode_switch) - 100), real_mode_switch, sizeof (real_mode_switch)); + memcpy(__va(0x1000 - 100), code, length); /* Set up the IDT for real mode. */ load_idt(&real_mode_idt); diff -urNp linux-2.6.32.48/arch/x86/kernel/setup.c linux-2.6.32.48-openvz/arch/x86/kernel/setup.c --- linux-2.6.32.48/arch/x86/kernel/setup.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/setup.c 2011-11-17 20:08:53.000000000 -0500 @@ -783,14 +783,14 @@ void __init setup_arch(char **cmdline_p) if (!boot_params.hdr.root_flags) root_mountflags &= ~MS_RDONLY; - init_mm.start_code = (unsigned long) _text; - init_mm.end_code = (unsigned long) _etext; + init_mm.start_code = ktla_ktva((unsigned long) _text); + init_mm.end_code = ktla_ktva((unsigned long) _etext); init_mm.end_data = (unsigned long) _edata; init_mm.brk = _brk_end; - code_resource.start = virt_to_phys(_text); - code_resource.end = virt_to_phys(_etext)-1; - data_resource.start = virt_to_phys(_etext); + code_resource.start = virt_to_phys(ktla_ktva(_text)); + code_resource.end = virt_to_phys(ktla_ktva(_etext))-1; + data_resource.start = virt_to_phys(_sdata); data_resource.end = virt_to_phys(_edata)-1; bss_resource.start = virt_to_phys(&__bss_start); bss_resource.end = virt_to_phys(&__bss_stop)-1; diff -urNp linux-2.6.32.48/arch/x86/kernel/setup_percpu.c linux-2.6.32.48-openvz/arch/x86/kernel/setup_percpu.c --- linux-2.6.32.48/arch/x86/kernel/setup_percpu.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/setup_percpu.c 2011-11-17 20:09:04.000000000 -0500 @@ -25,19 +25,17 @@ # define DBG(x...) #endif -DEFINE_PER_CPU(int, cpu_number); +#ifdef CONFIG_SMP +DEFINE_PER_CPU(unsigned int, cpu_number); EXPORT_PER_CPU_SYMBOL(cpu_number); +#endif -#ifdef CONFIG_X86_64 #define BOOT_PERCPU_OFFSET ((unsigned long)__per_cpu_load) -#else -#define BOOT_PERCPU_OFFSET 0 -#endif DEFINE_PER_CPU(unsigned long, this_cpu_off) = BOOT_PERCPU_OFFSET; EXPORT_PER_CPU_SYMBOL(this_cpu_off); -unsigned long __per_cpu_offset[NR_CPUS] __read_mostly = { +unsigned long __per_cpu_offset[NR_CPUS] __read_only = { [0 ... NR_CPUS-1] = BOOT_PERCPU_OFFSET, }; EXPORT_SYMBOL(__per_cpu_offset); @@ -159,10 +157,10 @@ static inline void setup_percpu_segment( { #ifdef CONFIG_X86_32 struct desc_struct gdt; + unsigned long base = per_cpu_offset(cpu); - pack_descriptor(&gdt, per_cpu_offset(cpu), 0xFFFFF, - 0x2 | DESCTYPE_S, 0x8); - gdt.s = 1; + pack_descriptor(&gdt, base, (VMALLOC_END - base - 1) >> PAGE_SHIFT, + 0x83 | DESCTYPE_S, 0xC); write_gdt_entry(get_cpu_gdt_table(cpu), GDT_ENTRY_PERCPU, &gdt, DESCTYPE_S); #endif @@ -212,6 +210,11 @@ void __init setup_per_cpu_areas(void) /* alrighty, percpu areas up and running */ delta = (unsigned long)pcpu_base_addr - (unsigned long)__per_cpu_start; for_each_possible_cpu(cpu) { +#ifdef CONFIG_CC_STACKPROTECTOR +#ifdef CONFIG_X86_32 + unsigned long canary = per_cpu(stack_canary.canary, cpu); +#endif +#endif per_cpu_offset(cpu) = delta + pcpu_unit_offsets[cpu]; per_cpu(this_cpu_off, cpu) = per_cpu_offset(cpu); per_cpu(cpu_number, cpu) = cpu; @@ -239,6 +242,12 @@ void __init setup_per_cpu_areas(void) early_per_cpu_map(x86_cpu_to_node_map, cpu); #endif #endif +#ifdef CONFIG_CC_STACKPROTECTOR +#ifdef CONFIG_X86_32 + if (!cpu) + per_cpu(stack_canary.canary, cpu) = canary; +#endif +#endif /* * Up to this point, the boot CPU has been using .data.init * area. Reload any changed state for the boot CPU. diff -urNp linux-2.6.32.48/arch/x86/kernel/signal.c linux-2.6.32.48-openvz/arch/x86/kernel/signal.c --- linux-2.6.32.48/arch/x86/kernel/signal.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/signal.c 2011-11-17 20:08:53.000000000 -0500 @@ -19,6 +19,7 @@ #include #include #include +#include #include #include @@ -197,7 +198,7 @@ static unsigned long align_sigframe(unsi * Align the stack pointer according to the i386 ABI, * i.e. so that on function entry ((sp + 4) & 15) == 0. */ - sp = ((sp + 4) & -16ul) - 4; + sp = ((sp - 12) & -16ul) - 4; #else /* !CONFIG_X86_32 */ sp = round_down(sp, 16) - 8; #endif @@ -248,11 +249,11 @@ get_sigframe(struct k_sigaction *ka, str * Return an always-bogus address instead so we will die with SIGSEGV. */ if (onsigstack && !likely(on_sig_stack(sp))) - return (void __user *)-1L; + return (__force void __user *)-1L; /* save i387 state */ if (used_math() && save_i387_xstate(*fpstate) < 0) - return (void __user *)-1L; + return (__force void __user *)-1L; return (void __user *)sp; } @@ -307,9 +308,9 @@ __setup_frame(int sig, struct k_sigactio } if (current->mm->context.vdso) - restorer = VDSO32_SYMBOL(current->mm->context.vdso, sigreturn); + restorer = (__force void __user *)VDSO32_SYMBOL(current->mm->context.vdso, sigreturn); else - restorer = &frame->retcode; + restorer = (void __user *)&frame->retcode; if (ka->sa.sa_flags & SA_RESTORER) restorer = ka->sa.sa_restorer; @@ -323,7 +324,7 @@ __setup_frame(int sig, struct k_sigactio * reasons and because gdb uses it as a signature to notice * signal handler stack frames. */ - err |= __put_user(*((u64 *)&retcode), (u64 *)frame->retcode); + err |= __put_user(*((u64 *)&retcode), (u64 __user *)frame->retcode); if (err) return -EFAULT; @@ -377,7 +378,10 @@ static int __setup_rt_frame(int sig, str err |= __copy_to_user(&frame->uc.uc_sigmask, set, sizeof(*set)); /* Set up to return from userspace. */ - restorer = VDSO32_SYMBOL(current->mm->context.vdso, rt_sigreturn); + if (current->mm->context.vdso) + restorer = (__force void __user *)VDSO32_SYMBOL(current->mm->context.vdso, rt_sigreturn); + else + restorer = (void __user *)&frame->retcode; if (ka->sa.sa_flags & SA_RESTORER) restorer = ka->sa.sa_restorer; put_user_ex(restorer, &frame->pretcode); @@ -389,7 +393,7 @@ static int __setup_rt_frame(int sig, str * reasons and because gdb uses it as a signature to notice * signal handler stack frames. */ - put_user_ex(*((u64 *)&rt_retcode), (u64 *)frame->retcode); + put_user_ex(*((u64 *)&rt_retcode), (u64 __user *)frame->retcode); } put_user_catch(err); if (err) @@ -789,9 +793,12 @@ static void do_signal(struct pt_regs *re * X86_32: vm86 regs switched out by assembly code before reaching * here, so testing against kernel CS suffices. */ - if (!user_mode(regs)) + if (!user_mode_novm(regs)) return; + if (try_to_freeze() && !signal_pending(current)) + goto no_signal; + if (current_thread_info()->status & TS_RESTORE_SIGMASK) oldset = ¤t->saved_sigmask; else @@ -821,6 +828,7 @@ static void do_signal(struct pt_regs *re return; } +no_signal: /* Did we come from a system call? */ if (syscall_get_nr(current, regs) >= 0) { /* Restart the system call - no handlers present */ diff -urNp linux-2.6.32.48/arch/x86/kernel/smpboot.c linux-2.6.32.48-openvz/arch/x86/kernel/smpboot.c --- linux-2.6.32.48/arch/x86/kernel/smpboot.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/smpboot.c 2011-11-17 20:09:04.000000000 -0500 @@ -94,14 +94,14 @@ static DEFINE_PER_CPU(struct task_struct */ static DEFINE_MUTEX(x86_cpu_hotplug_driver_mutex); -void cpu_hotplug_driver_lock() +void cpu_hotplug_driver_lock(void) { - mutex_lock(&x86_cpu_hotplug_driver_mutex); + mutex_lock(&x86_cpu_hotplug_driver_mutex); } -void cpu_hotplug_driver_unlock() +void cpu_hotplug_driver_unlock(void) { - mutex_unlock(&x86_cpu_hotplug_driver_mutex); + mutex_unlock(&x86_cpu_hotplug_driver_mutex); } ssize_t arch_cpu_probe(const char *buf, size_t count) { return -1; } @@ -625,7 +625,7 @@ wakeup_secondary_cpu_via_init(int phys_a * target processor state. */ startup_ipi_hook(phys_apicid, (unsigned long) start_secondary, - (unsigned long)stack_start.sp); + stack_start); /* * Run STARTUP IPI loop. @@ -743,6 +743,7 @@ static int __cpuinit do_boot_cpu(int api set_idle_for_cpu(cpu, c_idle.idle); do_rest: per_cpu(current_task, cpu) = c_idle.idle; + per_cpu(current_tinfo, cpu) = &c_idle.idle->tinfo; #ifdef CONFIG_X86_32 /* Stack for startup_32 can be just as for start_secondary onwards */ irq_ctx_init(cpu); @@ -750,13 +751,21 @@ do_rest: #else clear_tsk_thread_flag(c_idle.idle, TIF_FORK); initial_gs = per_cpu_offset(cpu); - per_cpu(kernel_stack, cpu) = - (unsigned long)task_stack_page(c_idle.idle) - - KERNEL_STACK_OFFSET + THREAD_SIZE; + per_cpu(kernel_stack, cpu) = (unsigned long)task_stack_page(c_idle.idle) - 16 + THREAD_SIZE; #endif + + pax_open_kernel(); early_gdt_descr.address = (unsigned long)get_cpu_gdt_table(cpu); + pax_close_kernel(); + initial_code = (unsigned long)start_secondary; - stack_start.sp = (void *) c_idle.idle->thread.sp; + stack_start = c_idle.idle->thread.sp; + +#ifdef CONFIG_VE + /* Cosmetic: sleep_time won't be changed afterwards for the idle + * thread; keep it 0 rather than -cycles. */ + VE_TASK_INFO(c_idle.idle)->sleep_time = 0; +#endif /* start_ip had better be page-aligned! */ start_ip = setup_trampoline(); @@ -891,6 +900,12 @@ int __cpuinit native_cpu_up(unsigned int per_cpu(cpu_state, cpu) = CPU_UP_PREPARE; +#ifdef CONFIG_PAX_PER_CPU_PGD + clone_pgd_range(get_cpu_pgd(cpu) + KERNEL_PGD_BOUNDARY, + swapper_pg_dir + KERNEL_PGD_BOUNDARY, + KERNEL_PGD_PTRS); +#endif + err = do_boot_cpu(apicid, cpu); if (err) { diff -urNp linux-2.6.32.48/arch/x86/kernel/smp.c linux-2.6.32.48-openvz/arch/x86/kernel/smp.c --- linux-2.6.32.48/arch/x86/kernel/smp.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/smp.c 2011-11-17 20:08:53.000000000 -0500 @@ -224,6 +224,11 @@ void smp_call_function_single_interrupt( irq_exit(); } +void send_nmi_ipi_allbutself(void) +{ + apic->send_IPI_allbutself(NMI_VECTOR); +} + struct smp_ops smp_ops = { .smp_prepare_boot_cpu = native_smp_prepare_boot_cpu, .smp_prepare_cpus = native_smp_prepare_cpus, diff -urNp linux-2.6.32.48/arch/x86/kernel/step.c linux-2.6.32.48-openvz/arch/x86/kernel/step.c --- linux-2.6.32.48/arch/x86/kernel/step.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/step.c 2011-11-17 20:08:53.000000000 -0500 @@ -27,10 +27,10 @@ unsigned long convert_ip_to_linear(struc struct desc_struct *desc; unsigned long base; - seg &= ~7UL; + seg >>= 3; mutex_lock(&child->mm->context.lock); - if (unlikely((seg >> 3) >= child->mm->context.size)) + if (unlikely(seg >= child->mm->context.size)) addr = -1L; /* bogus selector, access would fault */ else { desc = child->mm->context.ldt + seg; @@ -42,7 +42,8 @@ unsigned long convert_ip_to_linear(struc addr += base; } mutex_unlock(&child->mm->context.lock); - } + } else if (seg == __KERNEL_CS || seg == __KERNEXEC_KERNEL_CS) + addr = ktla_ktva(addr); return addr; } @@ -53,6 +54,9 @@ static int is_setting_trap_flag(struct t unsigned char opcode[15]; unsigned long addr = convert_ip_to_linear(child, regs); + if (addr == -EINVAL) + return 0; + copied = access_process_vm(child, addr, opcode, sizeof(opcode), 0); for (i = 0; i < copied; i++) { switch (opcode[i]) { @@ -74,7 +78,7 @@ static int is_setting_trap_flag(struct t #ifdef CONFIG_X86_64 case 0x40 ... 0x4f: - if (regs->cs != __USER_CS) + if ((regs->cs & 0xffff) != __USER_CS) /* 32-bit mode: register increment */ return 0; /* 64-bit mode: REX prefix */ diff -urNp linux-2.6.32.48/arch/x86/kernel/syscall_table_32.S linux-2.6.32.48-openvz/arch/x86/kernel/syscall_table_32.S --- linux-2.6.32.48/arch/x86/kernel/syscall_table_32.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/syscall_table_32.S 2011-11-17 20:08:53.000000000 -0500 @@ -1,3 +1,4 @@ +.section .rodata,"a",@progbits ENTRY(sys_call_table) .long sys_restart_syscall /* 0 - old "setup()" system call, used for restarting */ .long sys_exit @@ -336,3 +337,24 @@ ENTRY(sys_call_table) .long sys_pwritev .long sys_rt_tgsigqueueinfo /* 335 */ .long sys_perf_event_open + .rept 500-(.-sys_call_table)/4 + .long sys_ni_syscall + .endr + .long sys_fairsched_mknod /* 500 */ + .long sys_fairsched_rmnod + .long sys_fairsched_chwt + .long sys_fairsched_mvpr + .long sys_fairsched_rate + .long sys_fairsched_vcpus /* 505 */ + .long sys_ni_syscall + .long sys_ni_syscall + .long sys_ni_syscall + .long sys_ni_syscall + .long sys_getluid /* 510 */ + .long sys_setluid + .long sys_setublimit + .long sys_ubstat + .long sys_ni_syscall + .long sys_ni_syscall /* 515 */ + .long sys_lchmod + .long sys_lutime diff -urNp linux-2.6.32.48/arch/x86/kernel/sys_i386_32.c linux-2.6.32.48-openvz/arch/x86/kernel/sys_i386_32.c --- linux-2.6.32.48/arch/x86/kernel/sys_i386_32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/sys_i386_32.c 2011-11-17 20:08:53.000000000 -0500 @@ -24,6 +24,21 @@ #include +int i386_mmap_check(unsigned long addr, unsigned long len, unsigned long flags) +{ + unsigned long pax_task_size = TASK_SIZE; + +#ifdef CONFIG_PAX_SEGMEXEC + if (current->mm->pax_flags & MF_PAX_SEGMEXEC) + pax_task_size = SEGMEXEC_TASK_SIZE; +#endif + + if (len > pax_task_size || addr > pax_task_size - len) + return -EINVAL; + + return 0; +} + /* * Perform the select(nd, in, out, ex, tv) and mmap() system * calls. Linux/i386 didn't use to be able to handle more than @@ -58,6 +73,212 @@ out: return err; } +unsigned long +arch_get_unmapped_area(struct file *filp, unsigned long addr, + unsigned long len, unsigned long pgoff, unsigned long flags) +{ + struct mm_struct *mm = current->mm; + struct vm_area_struct *vma; + unsigned long start_addr, pax_task_size = TASK_SIZE; + +#ifdef CONFIG_PAX_SEGMEXEC + if (mm->pax_flags & MF_PAX_SEGMEXEC) + pax_task_size = SEGMEXEC_TASK_SIZE; +#endif + + pax_task_size -= PAGE_SIZE; + + if (len > pax_task_size) + return -ENOMEM; + + if (flags & MAP_FIXED) + return addr; + +#ifdef CONFIG_PAX_RANDMMAP + if (!(mm->pax_flags & MF_PAX_RANDMMAP)) +#endif + + if (addr) { + addr = PAGE_ALIGN(addr); + if (pax_task_size - len >= addr) { + vma = find_vma(mm, addr); + if (check_heap_stack_gap(vma, addr, len)) + return addr; + } + } + if (len > mm->cached_hole_size) { + start_addr = addr = mm->free_area_cache; + } else { + start_addr = addr = mm->mmap_base; + mm->cached_hole_size = 0; + } + +#ifdef CONFIG_PAX_PAGEEXEC + if (!nx_enabled && (mm->pax_flags & MF_PAX_PAGEEXEC) && (flags & MAP_EXECUTABLE) && start_addr >= mm->mmap_base) { + start_addr = 0x00110000UL; + +#ifdef CONFIG_PAX_RANDMMAP + if (mm->pax_flags & MF_PAX_RANDMMAP) + start_addr += mm->delta_mmap & 0x03FFF000UL; +#endif + + if (mm->start_brk <= start_addr && start_addr < mm->mmap_base) + start_addr = addr = mm->mmap_base; + else + addr = start_addr; + } +#endif + +full_search: + for (vma = find_vma(mm, addr); ; vma = vma->vm_next) { + /* At this point: (!vma || addr < vma->vm_end). */ + if (pax_task_size - len < addr) { + /* + * Start a new search - just in case we missed + * some holes. + */ + if (start_addr != mm->mmap_base) { + start_addr = addr = mm->mmap_base; + mm->cached_hole_size = 0; + goto full_search; + } + return -ENOMEM; + } + if (check_heap_stack_gap(vma, addr, len)) + break; + if (addr + mm->cached_hole_size < vma->vm_start) + mm->cached_hole_size = vma->vm_start - addr; + addr = vma->vm_end; + if (mm->start_brk <= addr && addr < mm->mmap_base) { + start_addr = addr = mm->mmap_base; + mm->cached_hole_size = 0; + goto full_search; + } + } + + /* + * Remember the place where we stopped the search: + */ + mm->free_area_cache = addr + len; + return addr; +} + +unsigned long +arch_get_unmapped_area_topdown(struct file *filp, const unsigned long addr0, + const unsigned long len, const unsigned long pgoff, + const unsigned long flags) +{ + struct vm_area_struct *vma; + struct mm_struct *mm = current->mm; + unsigned long base = mm->mmap_base, addr = addr0, pax_task_size = TASK_SIZE; + +#ifdef CONFIG_PAX_SEGMEXEC + if (mm->pax_flags & MF_PAX_SEGMEXEC) + pax_task_size = SEGMEXEC_TASK_SIZE; +#endif + + pax_task_size -= PAGE_SIZE; + + /* requested length too big for entire address space */ + if (len > pax_task_size) + return -ENOMEM; + + if (flags & MAP_FIXED) + return addr; + +#ifdef CONFIG_PAX_PAGEEXEC + if (!nx_enabled && (mm->pax_flags & MF_PAX_PAGEEXEC) && (flags & MAP_EXECUTABLE)) + goto bottomup; +#endif + +#ifdef CONFIG_PAX_RANDMMAP + if (!(mm->pax_flags & MF_PAX_RANDMMAP)) +#endif + + /* requesting a specific address */ + if (addr) { + addr = PAGE_ALIGN(addr); + if (pax_task_size - len >= addr) { + vma = find_vma(mm, addr); + if (check_heap_stack_gap(vma, addr, len)) + return addr; + } + } + + /* check if free_area_cache is useful for us */ + if (len <= mm->cached_hole_size) { + mm->cached_hole_size = 0; + mm->free_area_cache = mm->mmap_base; + } + + /* either no address requested or can't fit in requested address hole */ + addr = mm->free_area_cache; + + /* make sure it can fit in the remaining address space */ + if (addr > len) { + vma = find_vma(mm, addr-len); + if (check_heap_stack_gap(vma, addr - len, len)) + /* remember the address as a hint for next time */ + return (mm->free_area_cache = addr-len); + } + + if (mm->mmap_base < len) + goto bottomup; + + addr = mm->mmap_base-len; + + do { + /* + * Lookup failure means no vma is above this address, + * else if new region fits below vma->vm_start, + * return with success: + */ + vma = find_vma(mm, addr); + if (check_heap_stack_gap(vma, addr, len)) + /* remember the address as a hint for next time */ + return (mm->free_area_cache = addr); + + /* remember the largest hole we saw so far */ + if (addr + mm->cached_hole_size < vma->vm_start) + mm->cached_hole_size = vma->vm_start - addr; + + /* try just below the current vma->vm_start */ + addr = skip_heap_stack_gap(vma, len); + } while (!IS_ERR_VALUE(addr)); + +bottomup: + /* + * A failed mmap() very likely causes application failure, + * so fall back to the bottom-up function here. This scenario + * can happen with large stack limits and large mmap() + * allocations. + */ + +#ifdef CONFIG_PAX_SEGMEXEC + if (mm->pax_flags & MF_PAX_SEGMEXEC) + mm->mmap_base = SEGMEXEC_TASK_UNMAPPED_BASE; + else +#endif + + mm->mmap_base = TASK_UNMAPPED_BASE; + +#ifdef CONFIG_PAX_RANDMMAP + if (mm->pax_flags & MF_PAX_RANDMMAP) + mm->mmap_base += mm->delta_mmap; +#endif + + mm->free_area_cache = mm->mmap_base; + mm->cached_hole_size = ~0UL; + addr = arch_get_unmapped_area(filp, addr0, len, pgoff, flags); + /* + * Restore the topdown base: + */ + mm->mmap_base = base; + mm->free_area_cache = base; + mm->cached_hole_size = ~0UL; + + return addr; +} struct sel_arg_struct { unsigned long n; @@ -93,7 +314,7 @@ asmlinkage int sys_ipc(uint call, int fi return sys_semtimedop(first, (struct sembuf __user *)ptr, second, NULL); case SEMTIMEDOP: return sys_semtimedop(first, (struct sembuf __user *)ptr, second, - (const struct timespec __user *)fifth); + (__force const struct timespec __user *)fifth); case SEMGET: return sys_semget(first, second, third); @@ -140,7 +361,7 @@ asmlinkage int sys_ipc(uint call, int fi ret = do_shmat(first, (char __user *) ptr, second, &raddr); if (ret) return ret; - return put_user(raddr, (ulong __user *) third); + return put_user(raddr, (__force ulong __user *) third); } case 1: /* iBCS2 emulator entry point */ if (!segment_eq(get_fs(), get_ds())) @@ -207,17 +428,3 @@ asmlinkage int sys_olduname(struct oldol return error; } - - -/* - * Do a system call from kernel instead of calling sys_execve so we - * end up with proper pt_regs. - */ -int kernel_execve(const char *filename, char *const argv[], char *const envp[]) -{ - long __res; - asm volatile ("push %%ebx ; movl %2,%%ebx ; int $0x80 ; pop %%ebx" - : "=a" (__res) - : "0" (__NR_execve), "ri" (filename), "c" (argv), "d" (envp) : "memory"); - return __res; -} diff -urNp linux-2.6.32.48/arch/x86/kernel/sys_x86_64.c linux-2.6.32.48-openvz/arch/x86/kernel/sys_x86_64.c --- linux-2.6.32.48/arch/x86/kernel/sys_x86_64.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/sys_x86_64.c 2011-11-17 20:08:53.000000000 -0500 @@ -32,8 +32,8 @@ out: return error; } -static void find_start_end(unsigned long flags, unsigned long *begin, - unsigned long *end) +static void find_start_end(struct mm_struct *mm, unsigned long flags, + unsigned long *begin, unsigned long *end) { if (!test_thread_flag(TIF_IA32) && (flags & MAP_32BIT)) { unsigned long new_begin; @@ -52,7 +52,7 @@ static void find_start_end(unsigned long *begin = new_begin; } } else { - *begin = TASK_UNMAPPED_BASE; + *begin = mm->mmap_base; *end = TASK_SIZE; } } @@ -69,16 +69,19 @@ arch_get_unmapped_area(struct file *filp if (flags & MAP_FIXED) return addr; - find_start_end(flags, &begin, &end); + find_start_end(mm, flags, &begin, &end); if (len > end) return -ENOMEM; +#ifdef CONFIG_PAX_RANDMMAP + if (!(mm->pax_flags & MF_PAX_RANDMMAP)) +#endif + if (addr) { addr = PAGE_ALIGN(addr); vma = find_vma(mm, addr); - if (end - len >= addr && - (!vma || addr + len <= vma->vm_start)) + if (end - len >= addr && check_heap_stack_gap(vma, addr, len)) return addr; } if (((flags & MAP_32BIT) || test_thread_flag(TIF_IA32)) @@ -106,7 +109,7 @@ full_search: } return -ENOMEM; } - if (!vma || addr + len <= vma->vm_start) { + if (check_heap_stack_gap(vma, addr, len)) { /* * Remember the place where we stopped the search: */ @@ -128,7 +131,7 @@ arch_get_unmapped_area_topdown(struct fi { struct vm_area_struct *vma; struct mm_struct *mm = current->mm; - unsigned long addr = addr0; + unsigned long base = mm->mmap_base, addr = addr0; /* requested length too big for entire address space */ if (len > TASK_SIZE) @@ -141,13 +144,18 @@ arch_get_unmapped_area_topdown(struct fi if (!test_thread_flag(TIF_IA32) && (flags & MAP_32BIT)) goto bottomup; +#ifdef CONFIG_PAX_RANDMMAP + if (!(mm->pax_flags & MF_PAX_RANDMMAP)) +#endif + /* requesting a specific address */ if (addr) { addr = PAGE_ALIGN(addr); - vma = find_vma(mm, addr); - if (TASK_SIZE - len >= addr && - (!vma || addr + len <= vma->vm_start)) - return addr; + if (TASK_SIZE - len >= addr) { + vma = find_vma(mm, addr); + if (check_heap_stack_gap(vma, addr, len)) + return addr; + } } /* check if free_area_cache is useful for us */ @@ -162,7 +170,7 @@ arch_get_unmapped_area_topdown(struct fi /* make sure it can fit in the remaining address space */ if (addr > len) { vma = find_vma(mm, addr-len); - if (!vma || addr <= vma->vm_start) + if (check_heap_stack_gap(vma, addr - len, len)) /* remember the address as a hint for next time */ return mm->free_area_cache = addr-len; } @@ -179,7 +187,7 @@ arch_get_unmapped_area_topdown(struct fi * return with success: */ vma = find_vma(mm, addr); - if (!vma || addr+len <= vma->vm_start) + if (check_heap_stack_gap(vma, addr, len)) /* remember the address as a hint for next time */ return mm->free_area_cache = addr; @@ -188,8 +196,8 @@ arch_get_unmapped_area_topdown(struct fi mm->cached_hole_size = vma->vm_start - addr; /* try just below the current vma->vm_start */ - addr = vma->vm_start-len; - } while (len < vma->vm_start); + addr = skip_heap_stack_gap(vma, len); + } while (!IS_ERR_VALUE(addr)); bottomup: /* @@ -198,13 +206,21 @@ bottomup: * can happen with large stack limits and large mmap() * allocations. */ + mm->mmap_base = TASK_UNMAPPED_BASE; + +#ifdef CONFIG_PAX_RANDMMAP + if (mm->pax_flags & MF_PAX_RANDMMAP) + mm->mmap_base += mm->delta_mmap; +#endif + + mm->free_area_cache = mm->mmap_base; mm->cached_hole_size = ~0UL; - mm->free_area_cache = TASK_UNMAPPED_BASE; addr = arch_get_unmapped_area(filp, addr0, len, pgoff, flags); /* * Restore the topdown base: */ - mm->free_area_cache = mm->mmap_base; + mm->mmap_base = base; + mm->free_area_cache = base; mm->cached_hole_size = ~0UL; return addr; diff -urNp linux-2.6.32.48/arch/x86/kernel/tboot.c linux-2.6.32.48-openvz/arch/x86/kernel/tboot.c --- linux-2.6.32.48/arch/x86/kernel/tboot.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/tboot.c 2011-11-17 20:08:53.000000000 -0500 @@ -295,7 +295,7 @@ void tboot_sleep(u8 sleep_state, u32 pm1 tboot_shutdown(acpi_shutdown_map[sleep_state]); } -static atomic_t ap_wfs_count; +static atomic_unchecked_t ap_wfs_count; static int tboot_wait_for_aps(int num_aps) { @@ -319,9 +319,9 @@ static int __cpuinit tboot_cpu_callback( { switch (action) { case CPU_DYING: - atomic_inc(&ap_wfs_count); + atomic_inc_unchecked(&ap_wfs_count); if (num_online_cpus() == 1) - if (tboot_wait_for_aps(atomic_read(&ap_wfs_count))) + if (tboot_wait_for_aps(atomic_read_unchecked(&ap_wfs_count))) return NOTIFY_BAD; break; } @@ -340,7 +340,7 @@ static __init int tboot_late_init(void) tboot_create_trampoline(); - atomic_set(&ap_wfs_count, 0); + atomic_set_unchecked(&ap_wfs_count, 0); register_hotcpu_notifier(&tboot_cpu_notifier); return 0; } diff -urNp linux-2.6.32.48/arch/x86/kernel/time.c linux-2.6.32.48-openvz/arch/x86/kernel/time.c --- linux-2.6.32.48/arch/x86/kernel/time.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/time.c 2011-11-17 20:08:53.000000000 -0500 @@ -26,17 +26,13 @@ int timer_ack; #endif -#ifdef CONFIG_X86_64 -volatile unsigned long __jiffies __section_jiffies = INITIAL_JIFFIES; -#endif - unsigned long profile_pc(struct pt_regs *regs) { unsigned long pc = instruction_pointer(regs); - if (!user_mode_vm(regs) && in_lock_functions(pc)) { + if (!user_mode(regs) && in_lock_functions(pc)) { #ifdef CONFIG_FRAME_POINTER - return *(unsigned long *)(regs->bp + sizeof(long)); + return ktla_ktva(*(unsigned long *)(regs->bp + sizeof(long))); #else unsigned long *sp = (unsigned long *)kernel_stack_pointer(regs); @@ -45,11 +41,17 @@ unsigned long profile_pc(struct pt_regs * or above a saved flags. Eflags has bits 22-31 zero, * kernel addresses don't. */ + +#ifdef CONFIG_PAX_KERNEXEC + return ktla_ktva(sp[0]); +#else if (sp[0] >> 22) return sp[0]; if (sp[1] >> 22) return sp[1]; #endif + +#endif } return pc; } diff -urNp linux-2.6.32.48/arch/x86/kernel/tls.c linux-2.6.32.48-openvz/arch/x86/kernel/tls.c --- linux-2.6.32.48/arch/x86/kernel/tls.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/tls.c 2011-11-17 20:08:53.000000000 -0500 @@ -85,6 +85,11 @@ int do_set_thread_area(struct task_struc if (idx < GDT_ENTRY_TLS_MIN || idx > GDT_ENTRY_TLS_MAX) return -EINVAL; +#ifdef CONFIG_PAX_SEGMEXEC + if ((p->mm->pax_flags & MF_PAX_SEGMEXEC) && (info.contents & MODIFY_LDT_CONTENTS_CODE)) + return -EINVAL; +#endif + set_tls_desc(p, idx, &info, 1); return 0; diff -urNp linux-2.6.32.48/arch/x86/kernel/trampoline_32.S linux-2.6.32.48-openvz/arch/x86/kernel/trampoline_32.S --- linux-2.6.32.48/arch/x86/kernel/trampoline_32.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/trampoline_32.S 2011-11-17 20:08:53.000000000 -0500 @@ -32,6 +32,12 @@ #include #include +#ifdef CONFIG_PAX_KERNEXEC +#define ta(X) (X) +#else +#define ta(X) ((X) - __PAGE_OFFSET) +#endif + /* We can free up trampoline after bootup if cpu hotplug is not supported. */ __CPUINITRODATA .code16 @@ -60,7 +66,7 @@ r_base = . inc %ax # protected mode (PE) bit lmsw %ax # into protected mode # flush prefetch and jump to startup_32_smp in arch/i386/kernel/head.S - ljmpl $__BOOT_CS, $(startup_32_smp-__PAGE_OFFSET) + ljmpl $__BOOT_CS, $ta(startup_32_smp) # These need to be in the same 64K segment as the above; # hence we don't use the boot_gdt_descr defined in head.S diff -urNp linux-2.6.32.48/arch/x86/kernel/trampoline_64.S linux-2.6.32.48-openvz/arch/x86/kernel/trampoline_64.S --- linux-2.6.32.48/arch/x86/kernel/trampoline_64.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/trampoline_64.S 2011-11-17 20:09:04.000000000 -0500 @@ -91,7 +91,7 @@ startup_32: movl $__KERNEL_DS, %eax # Initialize the %ds segment register movl %eax, %ds - movl $X86_CR4_PAE, %eax + movl $(X86_CR4_PSE | X86_CR4_PAE | X86_CR4_PGE), %eax movl %eax, %cr4 # Enable PAE mode # Setup trampoline 4 level pagetables @@ -127,7 +127,7 @@ startup_64: no_longmode: hlt jmp no_longmode -#include "verify_cpu_64.S" +#include "verify_cpu.S" # Careful these need to be in the same 64K segment as the above; tidt: @@ -138,7 +138,7 @@ tidt: # so the kernel can live anywhere .balign 4 tgdt: - .short tgdt_end - tgdt # gdt limit + .short tgdt_end - tgdt - 1 # gdt limit .long tgdt - r_base .short 0 .quad 0x00cf9b000000ffff # __KERNEL32_CS diff -urNp linux-2.6.32.48/arch/x86/kernel/traps.c linux-2.6.32.48-openvz/arch/x86/kernel/traps.c --- linux-2.6.32.48/arch/x86/kernel/traps.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/traps.c 2011-11-17 20:09:04.000000000 -0500 @@ -69,12 +69,6 @@ asmlinkage int system_call(void); /* Do we ignore FPU interrupts ? */ char ignore_fpu_irq; - -/* - * The IDT has to be page-aligned to simplify the Pentium - * F0 0F bug workaround. - */ -gate_desc idt_table[NR_VECTORS] __page_aligned_data = { { { { 0, 0 } } }, }; #endif DECLARE_BITMAP(used_vectors, NR_VECTORS); @@ -112,19 +106,19 @@ static inline void preempt_conditional_c static inline void die_if_kernel(const char *str, struct pt_regs *regs, long err) { - if (!user_mode_vm(regs)) + if (!user_mode(regs)) die(str, regs, err); } #endif static void __kprobes -do_trap(int trapnr, int signr, char *str, struct pt_regs *regs, +do_trap(int trapnr, int signr, const char *str, struct pt_regs *regs, long error_code, siginfo_t *info) { struct task_struct *tsk = current; #ifdef CONFIG_X86_32 - if (regs->flags & X86_VM_MASK) { + if (v8086_mode(regs)) { /* * traps 0, 1, 3, 4, and 5 should be forwarded to vm86. * On nmi (interrupt 2), do_trap should not be called. @@ -135,7 +129,7 @@ do_trap(int trapnr, int signr, char *str } #endif - if (!user_mode(regs)) + if (!user_mode_novm(regs)) goto kernel_trap; #ifdef CONFIG_X86_32 @@ -158,7 +152,7 @@ trap_signal: printk_ratelimit()) { printk(KERN_INFO "%s[%d] trap %s ip:%lx sp:%lx error:%lx", - tsk->comm, tsk->pid, str, + tsk->comm, task_pid_nr(tsk), str, regs->ip, regs->sp, error_code); print_vma_addr(" in ", regs->ip); printk("\n"); @@ -175,8 +169,20 @@ kernel_trap: if (!fixup_exception(regs)) { tsk->thread.error_code = error_code; tsk->thread.trap_no = trapnr; + +#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_KERNEXEC) + if (trapnr == 12 && ((regs->cs & 0xFFFF) == __KERNEL_CS || (regs->cs & 0xFFFF) == __KERNEXEC_KERNEL_CS)) + str = "PAX: suspicious stack segment fault"; +#endif + die(str, regs, error_code); } + +#ifdef CONFIG_PAX_REFCOUNT + if (trapnr == 4) + pax_report_refcount_overflow(regs); +#endif + return; #ifdef CONFIG_X86_32 @@ -265,14 +271,30 @@ do_general_protection(struct pt_regs *re conditional_sti(regs); #ifdef CONFIG_X86_32 - if (regs->flags & X86_VM_MASK) + if (v8086_mode(regs)) goto gp_in_vm86; #endif tsk = current; - if (!user_mode(regs)) + if (!user_mode_novm(regs)) goto gp_in_kernel; +#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_PAGEEXEC) + if (!nx_enabled && tsk->mm && (tsk->mm->pax_flags & MF_PAX_PAGEEXEC)) { + struct mm_struct *mm = tsk->mm; + unsigned long limit; + + down_write(&mm->mmap_sem); + limit = mm->context.user_cs_limit; + if (limit < TASK_SIZE) { + track_exec_limit(mm, limit, TASK_SIZE, VM_EXEC); + up_write(&mm->mmap_sem); + return; + } + up_write(&mm->mmap_sem); + } +#endif + tsk->thread.error_code = error_code; tsk->thread.trap_no = 13; @@ -305,6 +327,13 @@ gp_in_kernel: if (notify_die(DIE_GPF, "general protection fault", regs, error_code, 13, SIGSEGV) == NOTIFY_STOP) return; + +#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_KERNEXEC) + if ((regs->cs & 0xFFFF) == __KERNEL_CS || (regs->cs & 0xFFFF) == __KERNEXEC_KERNEL_CS) + die("PAX: suspicious general protection fault", regs, error_code); + else +#endif + die("general protection fault", regs, error_code); } @@ -405,7 +434,8 @@ static notrace __kprobes void default_do * Ok, so this is none of the documented NMI sources, * so it must be the NMI watchdog. */ - if (nmi_watchdog_tick(regs, reason)) + if (nmi_watchdog_tick(regs, reason) + + do_nmi_show_regs(regs, cpu)) return; if (!do_nmi_callback(regs, cpu)) unknown_nmi_error(reason, regs); @@ -435,6 +465,17 @@ static notrace __kprobes void default_do dotraplinkage notrace __kprobes void do_nmi(struct pt_regs *regs, long error_code) { + +#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_KERNEXEC) + if (!user_mode(regs)) { + unsigned long cs = regs->cs & 0xFFFF; + unsigned long ip = ktva_ktla(regs->ip); + + if ((cs == __KERNEL_CS || cs == __KERNEXEC_KERNEL_CS) && ip <= (unsigned long)_etext) + regs->ip = ip; + } +#endif + nmi_enter(); inc_irq_stat(__nmi_count); @@ -558,7 +599,7 @@ dotraplinkage void __kprobes do_debug(st } #ifdef CONFIG_X86_32 - if (regs->flags & X86_VM_MASK) + if (v8086_mode(regs)) goto debug_vm86; #endif @@ -570,7 +611,7 @@ dotraplinkage void __kprobes do_debug(st * kernel space (but re-enable TF when returning to user mode). */ if (condition & DR_STEP) { - if (!user_mode(regs)) + if (!user_mode_novm(regs)) goto clear_TF_reenable; } @@ -757,7 +798,7 @@ do_simd_coprocessor_error(struct pt_regs * Handle strange cache flush from user space exception * in all other cases. This is undocumented behaviour. */ - if (regs->flags & X86_VM_MASK) { + if (v8086_mode(regs)) { handle_vm86_fault((struct kernel_vm86_regs *)regs, error_code); return; } @@ -798,7 +839,7 @@ asmlinkage void __attribute__((weak)) sm void __math_state_restore(void) { struct thread_info *thread = current_thread_info(); - struct task_struct *tsk = thread->task; + struct task_struct *tsk = current; /* * Paranoid restore. send a SIGSEGV if we fail to restore the state. @@ -825,8 +866,7 @@ void __math_state_restore(void) */ asmlinkage void math_state_restore(void) { - struct thread_info *thread = current_thread_info(); - struct task_struct *tsk = thread->task; + struct task_struct *tsk = current; if (!tsk_used_math(tsk)) { local_irq_enable(); diff -urNp linux-2.6.32.48/arch/x86/kernel/tsc_sync.c linux-2.6.32.48-openvz/arch/x86/kernel/tsc_sync.c --- linux-2.6.32.48/arch/x86/kernel/tsc_sync.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/tsc_sync.c 2011-11-17 20:08:53.000000000 -0500 @@ -150,6 +150,10 @@ void __cpuinit check_tsc_sync_source(int printk(" passed.\n"); } +#ifdef CONFIG_VE + /* TSC reset. kill whatever might rely on old values */ + VE_TASK_INFO(current)->wakeup_stamp = 0; +#endif /* * Reset it - just in case we boot another CPU later: */ diff -urNp linux-2.6.32.48/arch/x86/kernel/verify_cpu_64.S linux-2.6.32.48-openvz/arch/x86/kernel/verify_cpu_64.S --- linux-2.6.32.48/arch/x86/kernel/verify_cpu_64.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/verify_cpu_64.S 1969-12-31 19:00:00.000000000 -0500 @@ -1,105 +0,0 @@ -/* - * - * verify_cpu.S - Code for cpu long mode and SSE verification. This - * code has been borrowed from boot/setup.S and was introduced by - * Andi Kleen. - * - * Copyright (c) 2007 Andi Kleen (ak@suse.de) - * Copyright (c) 2007 Eric Biederman (ebiederm@xmission.com) - * Copyright (c) 2007 Vivek Goyal (vgoyal@in.ibm.com) - * - * This source code is licensed under the GNU General Public License, - * Version 2. See the file COPYING for more details. - * - * This is a common code for verification whether CPU supports - * long mode and SSE or not. It is not called directly instead this - * file is included at various places and compiled in that context. - * Following are the current usage. - * - * This file is included by both 16bit and 32bit code. - * - * arch/x86_64/boot/setup.S : Boot cpu verification (16bit) - * arch/x86_64/boot/compressed/head.S: Boot cpu verification (32bit) - * arch/x86_64/kernel/trampoline.S: secondary processor verfication (16bit) - * arch/x86_64/kernel/acpi/wakeup.S:Verfication at resume (16bit) - * - * verify_cpu, returns the status of cpu check in register %eax. - * 0: Success 1: Failure - * - * The caller needs to check for the error code and take the action - * appropriately. Either display a message or halt. - */ - -#include - -verify_cpu: - pushfl # Save caller passed flags - pushl $0 # Kill any dangerous flags - popfl - - pushfl # standard way to check for cpuid - popl %eax - movl %eax,%ebx - xorl $0x200000,%eax - pushl %eax - popfl - pushfl - popl %eax - cmpl %eax,%ebx - jz verify_cpu_no_longmode # cpu has no cpuid - - movl $0x0,%eax # See if cpuid 1 is implemented - cpuid - cmpl $0x1,%eax - jb verify_cpu_no_longmode # no cpuid 1 - - xor %di,%di - cmpl $0x68747541,%ebx # AuthenticAMD - jnz verify_cpu_noamd - cmpl $0x69746e65,%edx - jnz verify_cpu_noamd - cmpl $0x444d4163,%ecx - jnz verify_cpu_noamd - mov $1,%di # cpu is from AMD - -verify_cpu_noamd: - movl $0x1,%eax # Does the cpu have what it takes - cpuid - andl $REQUIRED_MASK0,%edx - xorl $REQUIRED_MASK0,%edx - jnz verify_cpu_no_longmode - - movl $0x80000000,%eax # See if extended cpuid is implemented - cpuid - cmpl $0x80000001,%eax - jb verify_cpu_no_longmode # no extended cpuid - - movl $0x80000001,%eax # Does the cpu have what it takes - cpuid - andl $REQUIRED_MASK1,%edx - xorl $REQUIRED_MASK1,%edx - jnz verify_cpu_no_longmode - -verify_cpu_sse_test: - movl $1,%eax - cpuid - andl $SSE_MASK,%edx - cmpl $SSE_MASK,%edx - je verify_cpu_sse_ok - test %di,%di - jz verify_cpu_no_longmode # only try to force SSE on AMD - movl $0xc0010015,%ecx # HWCR - rdmsr - btr $15,%eax # enable SSE - wrmsr - xor %di,%di # don't loop - jmp verify_cpu_sse_test # try again - -verify_cpu_no_longmode: - popfl # Restore caller passed flags - movl $1,%eax - ret -verify_cpu_sse_ok: - popfl # Restore caller passed flags - xorl %eax, %eax - ret diff -urNp linux-2.6.32.48/arch/x86/kernel/verify_cpu.S linux-2.6.32.48-openvz/arch/x86/kernel/verify_cpu.S --- linux-2.6.32.48/arch/x86/kernel/verify_cpu.S 1969-12-31 19:00:00.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/verify_cpu.S 2011-11-17 20:09:04.000000000 -0500 @@ -0,0 +1,140 @@ +/* + * + * verify_cpu.S - Code for cpu long mode and SSE verification. This + * code has been borrowed from boot/setup.S and was introduced by + * Andi Kleen. + * + * Copyright (c) 2007 Andi Kleen (ak@suse.de) + * Copyright (c) 2007 Eric Biederman (ebiederm@xmission.com) + * Copyright (c) 2007 Vivek Goyal (vgoyal@in.ibm.com) + * Copyright (c) 2010 Kees Cook (kees.cook@canonical.com) + * + * This source code is licensed under the GNU General Public License, + * Version 2. See the file COPYING for more details. + * + * This is a common code for verification whether CPU supports + * long mode and SSE or not. It is not called directly instead this + * file is included at various places and compiled in that context. + * This file is expected to run in 32bit code. Currently: + * + * arch/x86/boot/compressed/head_64.S: Boot cpu verification + * arch/x86/kernel/trampoline_64.S: secondary processor verification + * arch/x86/kernel/head_32.S: processor startup + * arch/x86/kernel/acpi/realmode/wakeup.S: 32bit processor resume + * + * verify_cpu, returns the status of longmode and SSE in register %eax. + * 0: Success 1: Failure + * + * On Intel, the XD_DISABLE flag will be cleared as a side-effect. + * + * The caller needs to check for the error code and take the action + * appropriately. Either display a message or halt. + */ + +#include +#include + +verify_cpu: + pushfl # Save caller passed flags + pushl $0 # Kill any dangerous flags + popfl + + pushfl # standard way to check for cpuid + popl %eax + movl %eax,%ebx + xorl $0x200000,%eax + pushl %eax + popfl + pushfl + popl %eax + cmpl %eax,%ebx + jz verify_cpu_no_longmode # cpu has no cpuid + + movl $0x0,%eax # See if cpuid 1 is implemented + cpuid + cmpl $0x1,%eax + jb verify_cpu_no_longmode # no cpuid 1 + + xor %di,%di + cmpl $0x68747541,%ebx # AuthenticAMD + jnz verify_cpu_noamd + cmpl $0x69746e65,%edx + jnz verify_cpu_noamd + cmpl $0x444d4163,%ecx + jnz verify_cpu_noamd + mov $1,%di # cpu is from AMD + jmp verify_cpu_check + +verify_cpu_noamd: + cmpl $0x756e6547,%ebx # GenuineIntel? + jnz verify_cpu_check + cmpl $0x49656e69,%edx + jnz verify_cpu_check + cmpl $0x6c65746e,%ecx + jnz verify_cpu_check + + # only call IA32_MISC_ENABLE when: + # family > 6 || (family == 6 && model >= 0xd) + movl $0x1, %eax # check CPU family and model + cpuid + movl %eax, %ecx + + andl $0x0ff00f00, %eax # mask family and extended family + shrl $8, %eax + cmpl $6, %eax + ja verify_cpu_clear_xd # family > 6, ok + jb verify_cpu_check # family < 6, skip + + andl $0x000f00f0, %ecx # mask model and extended model + shrl $4, %ecx + cmpl $0xd, %ecx + jb verify_cpu_check # family == 6, model < 0xd, skip + +verify_cpu_clear_xd: + movl $MSR_IA32_MISC_ENABLE, %ecx + rdmsr + btrl $2, %edx # clear MSR_IA32_MISC_ENABLE_XD_DISABLE + jnc verify_cpu_check # only write MSR if bit was changed + wrmsr + +verify_cpu_check: + movl $0x1,%eax # Does the cpu have what it takes + cpuid + andl $REQUIRED_MASK0,%edx + xorl $REQUIRED_MASK0,%edx + jnz verify_cpu_no_longmode + + movl $0x80000000,%eax # See if extended cpuid is implemented + cpuid + cmpl $0x80000001,%eax + jb verify_cpu_no_longmode # no extended cpuid + + movl $0x80000001,%eax # Does the cpu have what it takes + cpuid + andl $REQUIRED_MASK1,%edx + xorl $REQUIRED_MASK1,%edx + jnz verify_cpu_no_longmode + +verify_cpu_sse_test: + movl $1,%eax + cpuid + andl $SSE_MASK,%edx + cmpl $SSE_MASK,%edx + je verify_cpu_sse_ok + test %di,%di + jz verify_cpu_no_longmode # only try to force SSE on AMD + movl $MSR_K7_HWCR,%ecx + rdmsr + btr $15,%eax # enable SSE + wrmsr + xor %di,%di # don't loop + jmp verify_cpu_sse_test # try again + +verify_cpu_no_longmode: + popfl # Restore caller passed flags + movl $1,%eax + ret +verify_cpu_sse_ok: + popfl # Restore caller passed flags + xorl %eax, %eax + ret diff -urNp linux-2.6.32.48/arch/x86/kernel/vm86_32.c linux-2.6.32.48-openvz/arch/x86/kernel/vm86_32.c --- linux-2.6.32.48/arch/x86/kernel/vm86_32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/vm86_32.c 2011-11-17 20:08:53.000000000 -0500 @@ -41,6 +41,7 @@ #include #include #include +#include #include #include @@ -148,7 +149,7 @@ struct pt_regs *save_v86_state(struct ke do_exit(SIGSEGV); } - tss = &per_cpu(init_tss, get_cpu()); + tss = init_tss + get_cpu(); current->thread.sp0 = current->thread.saved_sp0; current->thread.sysenter_cs = __KERNEL_CS; load_sp0(tss, ¤t->thread); @@ -208,6 +209,13 @@ int sys_vm86old(struct pt_regs *regs) struct task_struct *tsk; int tmp, ret = -EPERM; +#ifdef CONFIG_GRKERNSEC_VM86 + if (!capable(CAP_SYS_RAWIO)) { + gr_handle_vm86(); + goto out; + } +#endif + tsk = current; if (tsk->thread.saved_sp0) goto out; @@ -238,6 +246,14 @@ int sys_vm86(struct pt_regs *regs) int tmp, ret; struct vm86plus_struct __user *v86; +#ifdef CONFIG_GRKERNSEC_VM86 + if (!capable(CAP_SYS_RAWIO)) { + gr_handle_vm86(); + ret = -EPERM; + goto out; + } +#endif + tsk = current; switch (regs->bx) { case VM86_REQUEST_IRQ: @@ -324,7 +340,7 @@ static void do_sys_vm86(struct kernel_vm tsk->thread.saved_fs = info->regs32->fs; tsk->thread.saved_gs = get_user_gs(info->regs32); - tss = &per_cpu(init_tss, get_cpu()); + tss = init_tss + get_cpu(); tsk->thread.sp0 = (unsigned long) &info->VM86_TSS_ESP0; if (cpu_has_sep) tsk->thread.sysenter_cs = 0; @@ -529,7 +545,7 @@ static void do_int(struct kernel_vm86_re goto cannot_handle; if (i == 0x21 && is_revectored(AH(regs), &KVM86->int21_revectored)) goto cannot_handle; - intr_ptr = (unsigned long __user *) (i << 2); + intr_ptr = (__force unsigned long __user *) (i << 2); if (get_user(segoffs, intr_ptr)) goto cannot_handle; if ((segoffs >> 16) == BIOSSEG) diff -urNp linux-2.6.32.48/arch/x86/kernel/vmi_32.c linux-2.6.32.48-openvz/arch/x86/kernel/vmi_32.c --- linux-2.6.32.48/arch/x86/kernel/vmi_32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/vmi_32.c 2011-11-17 20:09:04.000000000 -0500 @@ -44,12 +44,17 @@ typedef u32 __attribute__((regparm(1))) typedef u64 __attribute__((regparm(2))) (VROMLONGFUNC)(int); #define call_vrom_func(rom,func) \ - (((VROMFUNC *)(rom->func))()) + (((VROMFUNC *)(ktva_ktla(rom.func)))()) #define call_vrom_long_func(rom,func,arg) \ - (((VROMLONGFUNC *)(rom->func)) (arg)) +({\ + u64 __reloc = ((VROMLONGFUNC *)(ktva_ktla(rom.func))) (arg);\ + struct vmi_relocation_info *const __rel = (struct vmi_relocation_info *)&__reloc;\ + __rel->eip = (unsigned char *)ktva_ktla((unsigned long)__rel->eip);\ + __reloc;\ +}) -static struct vrom_header *vmi_rom; +static struct vrom_header vmi_rom __attribute((__section__(".vmi.rom"), __aligned__(PAGE_SIZE))); static int disable_pge; static int disable_pse; static int disable_sep; @@ -76,10 +81,10 @@ static struct { void (*set_initial_ap_state)(int, int); void (*halt)(void); void (*set_lazy_mode)(int mode); -} vmi_ops; +} __no_const vmi_ops __read_only; /* Cached VMI operations */ -struct vmi_timer_ops vmi_timer_ops; +struct vmi_timer_ops vmi_timer_ops __read_only; /* * VMI patching routines. @@ -94,7 +99,7 @@ struct vmi_timer_ops vmi_timer_ops; static inline void patch_offset(void *insnbuf, unsigned long ip, unsigned long dest) { - *(unsigned long *)(insnbuf+1) = dest-ip-5; + *(unsigned long *)(insnbuf+1) = dest-ip-5; } static unsigned patch_internal(int call, unsigned len, void *insnbuf, @@ -102,6 +107,7 @@ static unsigned patch_internal(int call, { u64 reloc; struct vmi_relocation_info *const rel = (struct vmi_relocation_info *)&reloc; + reloc = call_vrom_long_func(vmi_rom, get_reloc, call); switch(rel->type) { case VMI_RELOCATION_CALL_REL: @@ -404,13 +410,13 @@ static void vmi_set_pud(pud_t *pudp, pud static void vmi_pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep) { - const pte_t pte = { .pte = 0 }; + const pte_t pte = __pte(0ULL); vmi_ops.set_pte(pte, ptep, vmi_flags_addr(mm, addr, VMI_PAGE_PT, 0)); } static void vmi_pmd_clear(pmd_t *pmd) { - const pte_t pte = { .pte = 0 }; + const pte_t pte = __pte(0ULL); vmi_ops.set_pte(pte, (pte_t *)pmd, VMI_PAGE_PD); } #endif @@ -438,10 +444,10 @@ vmi_startup_ipi_hook(int phys_apicid, un ap.ss = __KERNEL_DS; ap.esp = (unsigned long) start_esp; - ap.ds = __USER_DS; - ap.es = __USER_DS; + ap.ds = __KERNEL_DS; + ap.es = __KERNEL_DS; ap.fs = __KERNEL_PERCPU; - ap.gs = __KERNEL_STACK_CANARY; + savesegment(gs, ap.gs); ap.eflags = 0; @@ -486,6 +492,18 @@ static void vmi_leave_lazy_mmu(void) paravirt_leave_lazy_mmu(); } +#ifdef CONFIG_PAX_KERNEXEC +static unsigned long vmi_pax_open_kernel(void) +{ + return 0; +} + +static unsigned long vmi_pax_close_kernel(void) +{ + return 0; +} +#endif + static inline int __init check_vmi_rom(struct vrom_header *rom) { struct pci_header *pci; @@ -498,6 +516,10 @@ static inline int __init check_vmi_rom(s return 0; if (rom->vrom_signature != VMI_SIGNATURE) return 0; + if (rom->rom_length * 512 > sizeof(*rom)) { + printk(KERN_WARNING "PAX: VMI: ROM size too big: %x\n", rom->rom_length * 512); + return 0; + } if (rom->api_version_maj != VMI_API_REV_MAJOR || rom->api_version_min+1 < VMI_API_REV_MINOR+1) { printk(KERN_WARNING "VMI: Found mismatched rom version %d.%d\n", @@ -562,7 +584,7 @@ static inline int __init probe_vmi_rom(v struct vrom_header *romstart; romstart = (struct vrom_header *)isa_bus_to_virt(base); if (check_vmi_rom(romstart)) { - vmi_rom = romstart; + vmi_rom = *romstart; return 1; } } @@ -836,6 +858,11 @@ static inline int __init activate_vmi(vo para_fill(pv_irq_ops.safe_halt, Halt); +#ifdef CONFIG_PAX_KERNEXEC + pv_mmu_ops.pax_open_kernel = vmi_pax_open_kernel; + pv_mmu_ops.pax_close_kernel = vmi_pax_close_kernel; +#endif + /* * Alternative instruction rewriting doesn't happen soon enough * to convert VMI_IRET to a call instead of a jump; so we have @@ -853,16 +880,16 @@ static inline int __init activate_vmi(vo void __init vmi_init(void) { - if (!vmi_rom) + if (!vmi_rom.rom_signature) probe_vmi_rom(); else - check_vmi_rom(vmi_rom); + check_vmi_rom(&vmi_rom); /* In case probing for or validating the ROM failed, basil */ - if (!vmi_rom) + if (!vmi_rom.rom_signature) return; - reserve_top_address(-vmi_rom->virtual_top); + reserve_top_address(-vmi_rom.virtual_top); #ifdef CONFIG_X86_IO_APIC /* This is virtual hardware; timer routing is wired correctly */ @@ -874,7 +901,7 @@ void __init vmi_activate(void) { unsigned long flags; - if (!vmi_rom) + if (!vmi_rom.rom_signature) return; local_irq_save(flags); diff -urNp linux-2.6.32.48/arch/x86/kernel/vmlinux.lds.S linux-2.6.32.48-openvz/arch/x86/kernel/vmlinux.lds.S --- linux-2.6.32.48/arch/x86/kernel/vmlinux.lds.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/vmlinux.lds.S 2011-11-17 20:08:53.000000000 -0500 @@ -26,6 +26,13 @@ #include #include #include +#include + +#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_KERNEXEC) +#define __KERNEL_TEXT_OFFSET (LOAD_OFFSET + ____LOAD_PHYSICAL_ADDR) +#else +#define __KERNEL_TEXT_OFFSET 0 +#endif #undef i386 /* in case the preprocessor is a 32bit one */ @@ -34,40 +41,53 @@ OUTPUT_FORMAT(CONFIG_OUTPUT_FORMAT, CONF #ifdef CONFIG_X86_32 OUTPUT_ARCH(i386) ENTRY(phys_startup_32) -jiffies = jiffies_64; #else OUTPUT_ARCH(i386:x86-64) ENTRY(phys_startup_64) -jiffies_64 = jiffies; #endif PHDRS { text PT_LOAD FLAGS(5); /* R_E */ - data PT_LOAD FLAGS(7); /* RWE */ +#ifdef CONFIG_X86_32 + module PT_LOAD FLAGS(5); /* R_E */ +#endif +#ifdef CONFIG_XEN + rodata PT_LOAD FLAGS(5); /* R_E */ +#else + rodata PT_LOAD FLAGS(4); /* R__ */ +#endif + data PT_LOAD FLAGS(6); /* RW_ */ #ifdef CONFIG_X86_64 user PT_LOAD FLAGS(5); /* R_E */ +#endif + init.begin PT_LOAD FLAGS(6); /* RW_ */ #ifdef CONFIG_SMP percpu PT_LOAD FLAGS(6); /* RW_ */ #endif + text.init PT_LOAD FLAGS(5); /* R_E */ + text.exit PT_LOAD FLAGS(5); /* R_E */ init PT_LOAD FLAGS(7); /* RWE */ -#endif note PT_NOTE FLAGS(0); /* ___ */ } SECTIONS { #ifdef CONFIG_X86_32 - . = LOAD_OFFSET + LOAD_PHYSICAL_ADDR; - phys_startup_32 = startup_32 - LOAD_OFFSET; + . = LOAD_OFFSET + ____LOAD_PHYSICAL_ADDR; #else - . = __START_KERNEL; - phys_startup_64 = startup_64 - LOAD_OFFSET; + . = __START_KERNEL; #endif /* Text and read-only data */ - .text : AT(ADDR(.text) - LOAD_OFFSET) { - _text = .; + .text (. - __KERNEL_TEXT_OFFSET): AT(ADDR(.text) - LOAD_OFFSET + __KERNEL_TEXT_OFFSET) { /* bootstrapping code */ +#ifdef CONFIG_X86_32 + phys_startup_32 = startup_32 - LOAD_OFFSET + __KERNEL_TEXT_OFFSET; +#else + phys_startup_64 = startup_64 - LOAD_OFFSET + __KERNEL_TEXT_OFFSET; +#endif + __LOAD_PHYSICAL_ADDR = . - LOAD_OFFSET + __KERNEL_TEXT_OFFSET; + _text = .; HEAD_TEXT #ifdef CONFIG_X86_32 . = ALIGN(PAGE_SIZE); @@ -82,28 +102,71 @@ SECTIONS IRQENTRY_TEXT *(.fixup) *(.gnu.warning) - /* End of text section */ - _etext = .; } :text = 0x9090 - NOTES :text :note + . += __KERNEL_TEXT_OFFSET; + +#ifdef CONFIG_X86_32 + . = ALIGN(PAGE_SIZE); + .vmi.rom : AT(ADDR(.vmi.rom) - LOAD_OFFSET) { + *(.vmi.rom) + } :module + + . = ALIGN(PAGE_SIZE); + .module.text : AT(ADDR(.module.text) - LOAD_OFFSET) { + +#if defined(CONFIG_PAX_KERNEXEC) && defined(CONFIG_MODULES) + MODULES_EXEC_VADDR = .; + BYTE(0) + . += (CONFIG_PAX_KERNEXEC_MODULE_TEXT * 1024 * 1024); + . = ALIGN(HPAGE_SIZE); + MODULES_EXEC_END = . - 1; +#endif + + } :module +#endif - EXCEPTION_TABLE(16) :text = 0x9090 + .text.end : AT(ADDR(.text.end) - LOAD_OFFSET) { + /* End of text section */ + _etext = . - __KERNEL_TEXT_OFFSET; + } + +#ifdef CONFIG_X86_32 + . = ALIGN(PAGE_SIZE); + .rodata.page_aligned : AT(ADDR(.rodata.page_aligned) - LOAD_OFFSET) { + *(.idt) + . = ALIGN(PAGE_SIZE); + *(.empty_zero_page) + *(.swapper_pg_fixmap) + *(.swapper_pg_pmd) + *(.swapper_pg_dir) + *(.trampoline_pg_dir) + } :rodata +#endif + + . = ALIGN(PAGE_SIZE); + NOTES :rodata :note + + EXCEPTION_TABLE(16) :rodata RO_DATA(PAGE_SIZE) /* Data */ .data : AT(ADDR(.data) - LOAD_OFFSET) { + +#ifdef CONFIG_PAX_KERNEXEC + . = ALIGN(HPAGE_SIZE); +#else + . = ALIGN(PAGE_SIZE); +#endif + /* Start of data section */ _sdata = .; /* init_task */ INIT_TASK_DATA(THREAD_SIZE) -#ifdef CONFIG_X86_32 - /* 32 bit has nosave before _edata */ NOSAVE_DATA -#endif PAGE_ALIGNED_DATA(PAGE_SIZE) @@ -112,6 +175,8 @@ SECTIONS DATA_DATA CONSTRUCTORS + jiffies = jiffies_64; + /* rarely changed data like cpu maps */ READ_MOSTLY_DATA(CONFIG_X86_INTERNODE_CACHE_BYTES) @@ -166,12 +231,6 @@ SECTIONS } vgetcpu_mode = VVIRT(.vgetcpu_mode); - . = ALIGN(CONFIG_X86_L1_CACHE_BYTES); - .jiffies : AT(VLOAD(.jiffies)) { - *(.jiffies) - } - jiffies = VVIRT(.jiffies); - .vsyscall_3 ADDR(.vsyscall_0) + 3072: AT(VLOAD(.vsyscall_3)) { *(.vsyscall_3) } @@ -187,12 +246,19 @@ SECTIONS #endif /* CONFIG_X86_64 */ /* Init code and data - will be freed after init */ - . = ALIGN(PAGE_SIZE); .init.begin : AT(ADDR(.init.begin) - LOAD_OFFSET) { + BYTE(0) + +#ifdef CONFIG_PAX_KERNEXEC + . = ALIGN(HPAGE_SIZE); +#else + . = ALIGN(PAGE_SIZE); +#endif + __init_begin = .; /* paired with __init_end */ - } + } :init.begin -#if defined(CONFIG_X86_64) && defined(CONFIG_SMP) +#ifdef CONFIG_SMP /* * percpu offsets are zero-based on SMP. PERCPU_VADDR() changes the * output PHDR, so the next output section - .init.text - should @@ -201,12 +267,27 @@ SECTIONS PERCPU_VADDR(0, :percpu) #endif - INIT_TEXT_SECTION(PAGE_SIZE) -#ifdef CONFIG_X86_64 - :init -#endif + . = ALIGN(PAGE_SIZE); + init_begin = .; + .init.text (. - __KERNEL_TEXT_OFFSET): AT(init_begin - LOAD_OFFSET) { + VMLINUX_SYMBOL(_sinittext) = .; + INIT_TEXT + VMLINUX_SYMBOL(_einittext) = .; + . = ALIGN(PAGE_SIZE); + } :text.init - INIT_DATA_SECTION(16) + /* + * .exit.text is discard at runtime, not link time, to deal with + * references from .altinstructions and .eh_frame + */ + .exit.text : AT(ADDR(.exit.text) - LOAD_OFFSET + __KERNEL_TEXT_OFFSET) { + EXIT_TEXT + . = ALIGN(16); + } :text.exit + . = init_begin + SIZEOF(.init.text) + SIZEOF(.exit.text); + + . = ALIGN(PAGE_SIZE); + INIT_DATA_SECTION(16) :init .x86_cpu_dev.init : AT(ADDR(.x86_cpu_dev.init) - LOAD_OFFSET) { __x86_cpu_dev_start = .; @@ -232,19 +313,11 @@ SECTIONS *(.altinstr_replacement) } - /* - * .exit.text is discard at runtime, not link time, to deal with - * references from .altinstructions and .eh_frame - */ - .exit.text : AT(ADDR(.exit.text) - LOAD_OFFSET) { - EXIT_TEXT - } - .exit.data : AT(ADDR(.exit.data) - LOAD_OFFSET) { EXIT_DATA } -#if !defined(CONFIG_X86_64) || !defined(CONFIG_SMP) +#ifndef CONFIG_SMP PERCPU(PAGE_SIZE) #endif @@ -267,12 +340,6 @@ SECTIONS . = ALIGN(PAGE_SIZE); } -#ifdef CONFIG_X86_64 - .data_nosave : AT(ADDR(.data_nosave) - LOAD_OFFSET) { - NOSAVE_DATA - } -#endif - /* BSS */ . = ALIGN(PAGE_SIZE); .bss : AT(ADDR(.bss) - LOAD_OFFSET) { @@ -288,6 +355,7 @@ SECTIONS __brk_base = .; . += 64 * 1024; /* 64k alignment slop space */ *(.brk_reservation) /* areas brk users have reserved */ + . = ALIGN(HPAGE_SIZE); __brk_limit = .; } @@ -316,13 +384,12 @@ SECTIONS * for the boot processor. */ #define INIT_PER_CPU(x) init_per_cpu__##x = per_cpu__##x + __per_cpu_load -INIT_PER_CPU(gdt_page); INIT_PER_CPU(irq_stack_union); /* * Build-time check on the image size: */ -. = ASSERT((_end - _text <= KERNEL_IMAGE_SIZE), +. = ASSERT((_end - _text - __KERNEL_TEXT_OFFSET <= KERNEL_IMAGE_SIZE), "kernel image bigger than KERNEL_IMAGE_SIZE"); #ifdef CONFIG_SMP diff -urNp linux-2.6.32.48/arch/x86/kernel/vsyscall_64.c linux-2.6.32.48-openvz/arch/x86/kernel/vsyscall_64.c --- linux-2.6.32.48/arch/x86/kernel/vsyscall_64.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/vsyscall_64.c 2011-11-17 20:08:53.000000000 -0500 @@ -80,6 +80,7 @@ void update_vsyscall(struct timespec *wa write_seqlock_irqsave(&vsyscall_gtod_data.lock, flags); /* copy vsyscall data */ + strlcpy(vsyscall_gtod_data.clock.name, clock->name, sizeof vsyscall_gtod_data.clock.name); vsyscall_gtod_data.clock.vread = clock->vread; vsyscall_gtod_data.clock.cycle_last = clock->cycle_last; vsyscall_gtod_data.clock.mask = clock->mask; @@ -203,7 +204,7 @@ vgetcpu(unsigned *cpu, unsigned *node, s We do this here because otherwise user space would do it on its own in a likely inferior way (no access to jiffies). If you don't like it pass NULL. */ - if (tcache && tcache->blob[0] == (j = __jiffies)) { + if (tcache && tcache->blob[0] == (j = jiffies)) { p = tcache->blob[1]; } else if (__vgetcpu_mode == VGETCPU_RDTSCP) { /* Load per CPU data from RDTSCP */ diff -urNp linux-2.6.32.48/arch/x86/kernel/x8664_ksyms_64.c linux-2.6.32.48-openvz/arch/x86/kernel/x8664_ksyms_64.c --- linux-2.6.32.48/arch/x86/kernel/x8664_ksyms_64.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/x8664_ksyms_64.c 2011-11-17 20:08:53.000000000 -0500 @@ -3,6 +3,7 @@ #include #include +#include #include @@ -17,8 +18,11 @@ EXPORT_SYMBOL(mcount); #endif +EXPORT_SYMBOL(kernel_execve); EXPORT_SYMBOL(kernel_thread); +EXPORT_SYMBOL_GPL(cpu_gdt_table); + EXPORT_SYMBOL(__get_user_1); EXPORT_SYMBOL(__get_user_2); EXPORT_SYMBOL(__get_user_4); @@ -30,8 +34,6 @@ EXPORT_SYMBOL(__put_user_8); EXPORT_SYMBOL(copy_user_generic); EXPORT_SYMBOL(__copy_user_nocache); -EXPORT_SYMBOL(copy_from_user); -EXPORT_SYMBOL(copy_to_user); EXPORT_SYMBOL(__copy_from_user_inatomic); EXPORT_SYMBOL(copy_page); diff -urNp linux-2.6.32.48/arch/x86/kernel/xsave.c linux-2.6.32.48-openvz/arch/x86/kernel/xsave.c --- linux-2.6.32.48/arch/x86/kernel/xsave.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kernel/xsave.c 2011-11-17 20:09:04.000000000 -0500 @@ -54,7 +54,7 @@ int check_for_xstate(struct i387_fxsave_ fx_sw_user->xstate_size > fx_sw_user->extended_size) return -1; - err = __get_user(magic2, (__u32 *) (((void *)fpstate) + + err = __get_user(magic2, (__u32 __user *) (((void __user *)fpstate) + fx_sw_user->extended_size - FP_XSTATE_MAGIC2_SIZE)); /* @@ -196,7 +196,7 @@ fx_only: * the other extended state. */ xrstor_state(init_xstate_buf, pcntxt_mask & ~XSTATE_FPSSE); - return fxrstor_checking((__force struct i387_fxsave_struct *)buf); + return fxrstor_checking((struct i387_fxsave_struct __force_kernel *)buf); } /* @@ -228,7 +228,7 @@ int restore_i387_xstate(void __user *buf if (task_thread_info(tsk)->status & TS_XSAVE) err = restore_user_xstate(buf); else - err = fxrstor_checking((__force struct i387_fxsave_struct *) + err = fxrstor_checking((struct i387_fxsave_struct __user *) buf); if (unlikely(err)) { /* diff -urNp linux-2.6.32.48/arch/x86/kvm/emulate.c linux-2.6.32.48-openvz/arch/x86/kvm/emulate.c --- linux-2.6.32.48/arch/x86/kvm/emulate.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kvm/emulate.c 2011-11-17 20:08:53.000000000 -0500 @@ -81,8 +81,8 @@ #define Src2CL (1<<29) #define Src2ImmByte (2<<29) #define Src2One (3<<29) -#define Src2Imm16 (4<<29) -#define Src2Mask (7<<29) +#define Src2Imm16 (4U<<29) +#define Src2Mask (7U<<29) enum { Group1_80, Group1_81, Group1_82, Group1_83, @@ -411,6 +411,7 @@ static u32 group2_table[] = { #define ____emulate_2op(_op, _src, _dst, _eflags, _x, _y, _suffix) \ do { \ + unsigned long _tmp; \ __asm__ __volatile__ ( \ _PRE_EFLAGS("0", "4", "2") \ _op _suffix " %"_x"3,%1; " \ @@ -424,8 +425,6 @@ static u32 group2_table[] = { /* Raw emulation: instruction has two explicit operands. */ #define __emulate_2op_nobyte(_op,_src,_dst,_eflags,_wx,_wy,_lx,_ly,_qx,_qy) \ do { \ - unsigned long _tmp; \ - \ switch ((_dst).bytes) { \ case 2: \ ____emulate_2op(_op,_src,_dst,_eflags,_wx,_wy,"w"); \ @@ -441,7 +440,6 @@ static u32 group2_table[] = { #define __emulate_2op(_op,_src,_dst,_eflags,_bx,_by,_wx,_wy,_lx,_ly,_qx,_qy) \ do { \ - unsigned long _tmp; \ switch ((_dst).bytes) { \ case 1: \ ____emulate_2op(_op,_src,_dst,_eflags,_bx,_by,"b"); \ diff -urNp linux-2.6.32.48/arch/x86/kvm/lapic.c linux-2.6.32.48-openvz/arch/x86/kvm/lapic.c --- linux-2.6.32.48/arch/x86/kvm/lapic.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kvm/lapic.c 2011-11-17 20:08:53.000000000 -0500 @@ -52,7 +52,7 @@ #define APIC_BUS_CYCLE_NS 1 /* #define apic_debug(fmt,arg...) printk(KERN_WARNING fmt,##arg) */ -#define apic_debug(fmt, arg...) +#define apic_debug(fmt, arg...) do {} while (0) #define APIC_LVT_NUM 6 /* 14 is the version for Xeon and Pentium 8.4.8*/ diff -urNp linux-2.6.32.48/arch/x86/kvm/paging_tmpl.h linux-2.6.32.48-openvz/arch/x86/kvm/paging_tmpl.h --- linux-2.6.32.48/arch/x86/kvm/paging_tmpl.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kvm/paging_tmpl.h 2011-11-17 20:08:53.000000000 -0500 @@ -416,6 +416,8 @@ static int FNAME(page_fault)(struct kvm_ int level = PT_PAGE_TABLE_LEVEL; unsigned long mmu_seq; + pax_track_stack(); + pgprintk("%s: addr %lx err %x\n", __func__, addr, error_code); kvm_mmu_audit(vcpu, "pre page fault"); diff -urNp linux-2.6.32.48/arch/x86/kvm/svm.c linux-2.6.32.48-openvz/arch/x86/kvm/svm.c --- linux-2.6.32.48/arch/x86/kvm/svm.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kvm/svm.c 2011-11-17 20:09:04.000000000 -0500 @@ -2486,7 +2486,11 @@ static void reload_tss(struct kvm_vcpu * int cpu = raw_smp_processor_id(); struct svm_cpu_data *svm_data = per_cpu(svm_data, cpu); + + pax_open_kernel(); svm_data->tss_desc->type = 9; /* available 32/64-bit TSS */ + pax_close_kernel(); + load_TR_desc(); } @@ -2947,7 +2951,7 @@ static bool svm_gb_page_enable(void) return true; } -static struct kvm_x86_ops svm_x86_ops = { +static const struct kvm_x86_ops svm_x86_ops = { .cpu_has_kvm_support = has_svm, .disabled_by_bios = is_disabled, .hardware_setup = svm_hardware_setup, diff -urNp linux-2.6.32.48/arch/x86/kvm/vmx.c linux-2.6.32.48-openvz/arch/x86/kvm/vmx.c --- linux-2.6.32.48/arch/x86/kvm/vmx.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kvm/vmx.c 2011-11-17 20:09:04.000000000 -0500 @@ -570,7 +570,11 @@ static void reload_tss(void) kvm_get_gdt(&gdt); descs = (void *)gdt.base; + + pax_open_kernel(); descs[GDT_ENTRY_TSS].type = 9; /* available TSS */ + pax_close_kernel(); + load_TR_desc(); } @@ -1410,8 +1414,11 @@ static __init int hardware_setup(void) if (!cpu_has_vmx_flexpriority()) flexpriority_enabled = 0; - if (!cpu_has_vmx_tpr_shadow()) - kvm_x86_ops->update_cr8_intercept = NULL; + if (!cpu_has_vmx_tpr_shadow()) { + pax_open_kernel(); + *(void **)&kvm_x86_ops->update_cr8_intercept = NULL; + pax_close_kernel(); + } if (enable_ept && !cpu_has_vmx_ept_2m_page()) kvm_disable_largepages(); @@ -2362,7 +2369,7 @@ static int vmx_vcpu_setup(struct vcpu_vm vmcs_writel(HOST_IDTR_BASE, dt.base); /* 22.2.4 */ asm("mov $.Lkvm_vmx_return, %0" : "=r"(kvm_vmx_return)); - vmcs_writel(HOST_RIP, kvm_vmx_return); /* 22.2.5 */ + vmcs_writel(HOST_RIP, ktla_ktva(kvm_vmx_return)); /* 22.2.5 */ vmcs_write32(VM_EXIT_MSR_STORE_COUNT, 0); vmcs_write32(VM_EXIT_MSR_LOAD_COUNT, 0); vmcs_write32(VM_ENTRY_MSR_LOAD_COUNT, 0); @@ -3718,6 +3725,12 @@ static void vmx_vcpu_run(struct kvm_vcpu "jmp .Lkvm_vmx_return \n\t" ".Llaunched: " __ex(ASM_VMX_VMRESUME) "\n\t" ".Lkvm_vmx_return: " + +#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_KERNEXEC) + "ljmp %[cs],$.Lkvm_vmx_return2\n\t" + ".Lkvm_vmx_return2: " +#endif + /* Save guest registers, load host registers, keep flags */ "xchg %0, (%%"R"sp) \n\t" "mov %%"R"ax, %c[rax](%0) \n\t" @@ -3764,8 +3777,13 @@ static void vmx_vcpu_run(struct kvm_vcpu [r15]"i"(offsetof(struct vcpu_vmx, vcpu.arch.regs[VCPU_REGS_R15])), #endif [cr2]"i"(offsetof(struct vcpu_vmx, vcpu.arch.cr2)) + +#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_KERNEXEC) + ,[cs]"i"(__KERNEL_CS) +#endif + : "cc", "memory" - , R"bx", R"di", R"si" + , R"ax", R"bx", R"di", R"si" #ifdef CONFIG_X86_64 , "r8", "r9", "r10", "r11", "r12", "r13", "r14", "r15" #endif @@ -3782,7 +3800,16 @@ static void vmx_vcpu_run(struct kvm_vcpu if (vmx->rmode.irq.pending) fixup_rmode_irq(vmx); - asm("mov %0, %%ds; mov %0, %%es" : : "r"(__USER_DS)); + asm("mov %0, %%ds; mov %0, %%es; mov %0, %%ss" : : "r"(__KERNEL_DS)); + +#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_KERNEXEC) + loadsegment(fs, __KERNEL_PERCPU); +#endif + +#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_MEMORY_UDEREF) + __set_fs(current_thread_info()->addr_limit); +#endif + vmx->launched = 1; vmx_complete_interrupts(vmx); @@ -3957,7 +3984,7 @@ static bool vmx_gb_page_enable(void) return false; } -static struct kvm_x86_ops vmx_x86_ops = { +static const struct kvm_x86_ops vmx_x86_ops = { .cpu_has_kvm_support = cpu_has_kvm_support, .disabled_by_bios = vmx_disabled_by_bios, .hardware_setup = hardware_setup, diff -urNp linux-2.6.32.48/arch/x86/kvm/x86.c linux-2.6.32.48-openvz/arch/x86/kvm/x86.c --- linux-2.6.32.48/arch/x86/kvm/x86.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/kvm/x86.c 2011-11-17 20:08:53.000000000 -0500 @@ -82,7 +82,7 @@ static void update_cr8_intercept(struct static int kvm_dev_ioctl_get_supported_cpuid(struct kvm_cpuid2 *cpuid, struct kvm_cpuid_entry2 __user *entries); -struct kvm_x86_ops *kvm_x86_ops; +const struct kvm_x86_ops *kvm_x86_ops; EXPORT_SYMBOL_GPL(kvm_x86_ops); int ignore_msrs = 0; @@ -1430,15 +1430,20 @@ static int kvm_vcpu_ioctl_set_cpuid2(str struct kvm_cpuid2 *cpuid, struct kvm_cpuid_entry2 __user *entries) { - int r; + int r, i; r = -E2BIG; if (cpuid->nent > KVM_MAX_CPUID_ENTRIES) goto out; r = -EFAULT; - if (copy_from_user(&vcpu->arch.cpuid_entries, entries, - cpuid->nent * sizeof(struct kvm_cpuid_entry2))) + if (!access_ok(VERIFY_READ, entries, cpuid->nent * sizeof(struct kvm_cpuid_entry2))) goto out; + for (i = 0; i < cpuid->nent; ++i) { + struct kvm_cpuid_entry2 cpuid_entry; + if (__copy_from_user(&cpuid_entry, entries + i, sizeof(cpuid_entry))) + goto out; + vcpu->arch.cpuid_entries[i] = cpuid_entry; + } vcpu->arch.cpuid_nent = cpuid->nent; kvm_apic_set_version(vcpu); return 0; @@ -1451,16 +1456,20 @@ static int kvm_vcpu_ioctl_get_cpuid2(str struct kvm_cpuid2 *cpuid, struct kvm_cpuid_entry2 __user *entries) { - int r; + int r, i; vcpu_load(vcpu); r = -E2BIG; if (cpuid->nent < vcpu->arch.cpuid_nent) goto out; r = -EFAULT; - if (copy_to_user(entries, &vcpu->arch.cpuid_entries, - vcpu->arch.cpuid_nent * sizeof(struct kvm_cpuid_entry2))) + if (!access_ok(VERIFY_WRITE, entries, vcpu->arch.cpuid_nent * sizeof(struct kvm_cpuid_entry2))) goto out; + for (i = 0; i < vcpu->arch.cpuid_nent; ++i) { + struct kvm_cpuid_entry2 cpuid_entry = vcpu->arch.cpuid_entries[i]; + if (__copy_to_user(entries + i, &cpuid_entry, sizeof(cpuid_entry))) + goto out; + } return 0; out: @@ -1678,7 +1687,7 @@ static int kvm_vcpu_ioctl_set_lapic(stru static int kvm_vcpu_ioctl_interrupt(struct kvm_vcpu *vcpu, struct kvm_interrupt *irq) { - if (irq->irq < 0 || irq->irq >= 256) + if (irq->irq >= 256) return -EINVAL; if (irqchip_in_kernel(vcpu->kvm)) return -ENXIO; @@ -3260,10 +3269,10 @@ static struct notifier_block kvmclock_cp .notifier_call = kvmclock_cpufreq_notifier }; -int kvm_arch_init(void *opaque) +int kvm_arch_init(const void *opaque) { int r, cpu; - struct kvm_x86_ops *ops = (struct kvm_x86_ops *)opaque; + const struct kvm_x86_ops *ops = (const struct kvm_x86_ops *)opaque; if (kvm_x86_ops) { printk(KERN_ERR "kvm: already loaded the other module\n"); diff -urNp linux-2.6.32.48/arch/x86/lguest/boot.c linux-2.6.32.48-openvz/arch/x86/lguest/boot.c --- linux-2.6.32.48/arch/x86/lguest/boot.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/lguest/boot.c 2011-11-17 20:09:04.000000000 -0500 @@ -1172,9 +1172,10 @@ static __init int early_put_chars(u32 vt * Rebooting also tells the Host we're finished, but the RESTART flag tells the * Launcher to reboot us. */ -static void lguest_restart(char *reason) +static __noreturn void lguest_restart(char *reason) { kvm_hypercall2(LHCALL_SHUTDOWN, __pa(reason), LGUEST_SHUTDOWN_RESTART); + BUG(); } /*G:050 diff -urNp linux-2.6.32.48/arch/x86/lib/atomic64_32.c linux-2.6.32.48-openvz/arch/x86/lib/atomic64_32.c --- linux-2.6.32.48/arch/x86/lib/atomic64_32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/lib/atomic64_32.c 2011-11-17 20:08:53.000000000 -0500 @@ -25,6 +25,12 @@ u64 atomic64_cmpxchg(atomic64_t *ptr, u6 } EXPORT_SYMBOL(atomic64_cmpxchg); +u64 atomic64_cmpxchg_unchecked(atomic64_unchecked_t *ptr, u64 old_val, u64 new_val) +{ + return cmpxchg8b(&ptr->counter, old_val, new_val); +} +EXPORT_SYMBOL(atomic64_cmpxchg_unchecked); + /** * atomic64_xchg - xchg atomic64 variable * @ptr: pointer to type atomic64_t @@ -56,6 +62,36 @@ u64 atomic64_xchg(atomic64_t *ptr, u64 n EXPORT_SYMBOL(atomic64_xchg); /** + * atomic64_xchg_unchecked - xchg atomic64 variable + * @ptr: pointer to type atomic64_unchecked_t + * @new_val: value to assign + * + * Atomically xchgs the value of @ptr to @new_val and returns + * the old value. + */ +u64 atomic64_xchg_unchecked(atomic64_unchecked_t *ptr, u64 new_val) +{ + /* + * Try first with a (possibly incorrect) assumption about + * what we have there. We'll do two loops most likely, + * but we'll get an ownership MESI transaction straight away + * instead of a read transaction followed by a + * flush-for-ownership transaction: + */ + u64 old_val, real_val = 0; + + do { + old_val = real_val; + + real_val = atomic64_cmpxchg_unchecked(ptr, old_val, new_val); + + } while (real_val != old_val); + + return old_val; +} +EXPORT_SYMBOL(atomic64_xchg_unchecked); + +/** * atomic64_set - set atomic64 variable * @ptr: pointer to type atomic64_t * @new_val: value to assign @@ -69,7 +105,19 @@ void atomic64_set(atomic64_t *ptr, u64 n EXPORT_SYMBOL(atomic64_set); /** -EXPORT_SYMBOL(atomic64_read); + * atomic64_unchecked_set - set atomic64 variable + * @ptr: pointer to type atomic64_unchecked_t + * @new_val: value to assign + * + * Atomically sets the value of @ptr to @new_val. + */ +void atomic64_set_unchecked(atomic64_unchecked_t *ptr, u64 new_val) +{ + atomic64_xchg_unchecked(ptr, new_val); +} +EXPORT_SYMBOL(atomic64_set_unchecked); + +/** * atomic64_add_return - add and return * @delta: integer value to add * @ptr: pointer to type atomic64_t @@ -99,24 +147,72 @@ noinline u64 atomic64_add_return(u64 del } EXPORT_SYMBOL(atomic64_add_return); +/** + * atomic64_add_return_unchecked - add and return + * @delta: integer value to add + * @ptr: pointer to type atomic64_unchecked_t + * + * Atomically adds @delta to @ptr and returns @delta + *@ptr + */ +noinline u64 atomic64_add_return_unchecked(u64 delta, atomic64_unchecked_t *ptr) +{ + /* + * Try first with a (possibly incorrect) assumption about + * what we have there. We'll do two loops most likely, + * but we'll get an ownership MESI transaction straight away + * instead of a read transaction followed by a + * flush-for-ownership transaction: + */ + u64 old_val, new_val, real_val = 0; + + do { + old_val = real_val; + new_val = old_val + delta; + + real_val = atomic64_cmpxchg_unchecked(ptr, old_val, new_val); + + } while (real_val != old_val); + + return new_val; +} +EXPORT_SYMBOL(atomic64_add_return_unchecked); + u64 atomic64_sub_return(u64 delta, atomic64_t *ptr) { return atomic64_add_return(-delta, ptr); } EXPORT_SYMBOL(atomic64_sub_return); +u64 atomic64_sub_return_unchecked(u64 delta, atomic64_unchecked_t *ptr) +{ + return atomic64_add_return_unchecked(-delta, ptr); +} +EXPORT_SYMBOL(atomic64_sub_return_unchecked); + u64 atomic64_inc_return(atomic64_t *ptr) { return atomic64_add_return(1, ptr); } EXPORT_SYMBOL(atomic64_inc_return); +u64 atomic64_inc_return_unchecked(atomic64_unchecked_t *ptr) +{ + return atomic64_add_return_unchecked(1, ptr); +} +EXPORT_SYMBOL(atomic64_inc_return_unchecked); + u64 atomic64_dec_return(atomic64_t *ptr) { return atomic64_sub_return(1, ptr); } EXPORT_SYMBOL(atomic64_dec_return); +u64 atomic64_dec_return_unchecked(atomic64_unchecked_t *ptr) +{ + return atomic64_sub_return_unchecked(1, ptr); +} +EXPORT_SYMBOL(atomic64_dec_return_unchecked); + /** * atomic64_add - add integer to atomic64 variable * @delta: integer value to add @@ -131,6 +227,19 @@ void atomic64_add(u64 delta, atomic64_t EXPORT_SYMBOL(atomic64_add); /** + * atomic64_add_unchecked - add integer to atomic64 variable + * @delta: integer value to add + * @ptr: pointer to type atomic64_unchecked_t + * + * Atomically adds @delta to @ptr. + */ +void atomic64_add_unchecked(u64 delta, atomic64_unchecked_t *ptr) +{ + atomic64_add_return_unchecked(delta, ptr); +} +EXPORT_SYMBOL(atomic64_add_unchecked); + +/** * atomic64_sub - subtract the atomic64 variable * @delta: integer value to subtract * @ptr: pointer to type atomic64_t @@ -144,6 +253,19 @@ void atomic64_sub(u64 delta, atomic64_t EXPORT_SYMBOL(atomic64_sub); /** + * atomic64_sub_unchecked - subtract the atomic64 variable + * @delta: integer value to subtract + * @ptr: pointer to type atomic64_unchecked_t + * + * Atomically subtracts @delta from @ptr. + */ +void atomic64_sub_unchecked(u64 delta, atomic64_unchecked_t *ptr) +{ + atomic64_add_unchecked(-delta, ptr); +} +EXPORT_SYMBOL(atomic64_sub_unchecked); + +/** * atomic64_sub_and_test - subtract value from variable and test result * @delta: integer value to subtract * @ptr: pointer to type atomic64_t @@ -173,6 +295,18 @@ void atomic64_inc(atomic64_t *ptr) EXPORT_SYMBOL(atomic64_inc); /** + * atomic64_inc_unchecked - increment atomic64 variable + * @ptr: pointer to type atomic64_unchecked_t + * + * Atomically increments @ptr by 1. + */ +void atomic64_inc_unchecked(atomic64_unchecked_t *ptr) +{ + atomic64_add_unchecked(1, ptr); +} +EXPORT_SYMBOL(atomic64_inc_unchecked); + +/** * atomic64_dec - decrement atomic64 variable * @ptr: pointer to type atomic64_t * @@ -185,6 +319,18 @@ void atomic64_dec(atomic64_t *ptr) EXPORT_SYMBOL(atomic64_dec); /** + * atomic64_dec_unchecked - decrement atomic64 variable + * @ptr: pointer to type atomic64_unchecked_t + * + * Atomically decrements @ptr by 1. + */ +void atomic64_dec_unchecked(atomic64_unchecked_t *ptr) +{ + atomic64_sub_unchecked(1, ptr); +} +EXPORT_SYMBOL(atomic64_dec_unchecked); + +/** * atomic64_dec_and_test - decrement and test * @ptr: pointer to type atomic64_t * diff -urNp linux-2.6.32.48/arch/x86/lib/checksum_32.S linux-2.6.32.48-openvz/arch/x86/lib/checksum_32.S --- linux-2.6.32.48/arch/x86/lib/checksum_32.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/lib/checksum_32.S 2011-11-17 20:08:53.000000000 -0500 @@ -28,7 +28,8 @@ #include #include #include - +#include + /* * computes a partial checksum, e.g. for TCP/UDP fragments */ @@ -304,9 +305,28 @@ unsigned int csum_partial_copy_generic ( #define ARGBASE 16 #define FP 12 - -ENTRY(csum_partial_copy_generic) + +ENTRY(csum_partial_copy_generic_to_user) CFI_STARTPROC + +#ifdef CONFIG_PAX_MEMORY_UDEREF + pushl %gs + CFI_ADJUST_CFA_OFFSET 4 + popl %es + CFI_ADJUST_CFA_OFFSET -4 + jmp csum_partial_copy_generic +#endif + +ENTRY(csum_partial_copy_generic_from_user) + +#ifdef CONFIG_PAX_MEMORY_UDEREF + pushl %gs + CFI_ADJUST_CFA_OFFSET 4 + popl %ds + CFI_ADJUST_CFA_OFFSET -4 +#endif + +ENTRY(csum_partial_copy_generic) subl $4,%esp CFI_ADJUST_CFA_OFFSET 4 pushl %edi @@ -331,7 +351,7 @@ ENTRY(csum_partial_copy_generic) jmp 4f SRC(1: movw (%esi), %bx ) addl $2, %esi -DST( movw %bx, (%edi) ) +DST( movw %bx, %es:(%edi) ) addl $2, %edi addw %bx, %ax adcl $0, %eax @@ -343,30 +363,30 @@ DST( movw %bx, (%edi) ) SRC(1: movl (%esi), %ebx ) SRC( movl 4(%esi), %edx ) adcl %ebx, %eax -DST( movl %ebx, (%edi) ) +DST( movl %ebx, %es:(%edi) ) adcl %edx, %eax -DST( movl %edx, 4(%edi) ) +DST( movl %edx, %es:4(%edi) ) SRC( movl 8(%esi), %ebx ) SRC( movl 12(%esi), %edx ) adcl %ebx, %eax -DST( movl %ebx, 8(%edi) ) +DST( movl %ebx, %es:8(%edi) ) adcl %edx, %eax -DST( movl %edx, 12(%edi) ) +DST( movl %edx, %es:12(%edi) ) SRC( movl 16(%esi), %ebx ) SRC( movl 20(%esi), %edx ) adcl %ebx, %eax -DST( movl %ebx, 16(%edi) ) +DST( movl %ebx, %es:16(%edi) ) adcl %edx, %eax -DST( movl %edx, 20(%edi) ) +DST( movl %edx, %es:20(%edi) ) SRC( movl 24(%esi), %ebx ) SRC( movl 28(%esi), %edx ) adcl %ebx, %eax -DST( movl %ebx, 24(%edi) ) +DST( movl %ebx, %es:24(%edi) ) adcl %edx, %eax -DST( movl %edx, 28(%edi) ) +DST( movl %edx, %es:28(%edi) ) lea 32(%esi), %esi lea 32(%edi), %edi @@ -380,7 +400,7 @@ DST( movl %edx, 28(%edi) ) shrl $2, %edx # This clears CF SRC(3: movl (%esi), %ebx ) adcl %ebx, %eax -DST( movl %ebx, (%edi) ) +DST( movl %ebx, %es:(%edi) ) lea 4(%esi), %esi lea 4(%edi), %edi dec %edx @@ -392,12 +412,12 @@ DST( movl %ebx, (%edi) ) jb 5f SRC( movw (%esi), %cx ) leal 2(%esi), %esi -DST( movw %cx, (%edi) ) +DST( movw %cx, %es:(%edi) ) leal 2(%edi), %edi je 6f shll $16,%ecx SRC(5: movb (%esi), %cl ) -DST( movb %cl, (%edi) ) +DST( movb %cl, %es:(%edi) ) 6: addl %ecx, %eax adcl $0, %eax 7: @@ -408,7 +428,7 @@ DST( movb %cl, (%edi) ) 6001: movl ARGBASE+20(%esp), %ebx # src_err_ptr - movl $-EFAULT, (%ebx) + movl $-EFAULT, %ss:(%ebx) # zero the complete destination - computing the rest # is too much work @@ -421,11 +441,19 @@ DST( movb %cl, (%edi) ) 6002: movl ARGBASE+24(%esp), %ebx # dst_err_ptr - movl $-EFAULT,(%ebx) + movl $-EFAULT,%ss:(%ebx) jmp 5000b .previous + pushl %ss + CFI_ADJUST_CFA_OFFSET 4 + popl %ds + CFI_ADJUST_CFA_OFFSET -4 + pushl %ss + CFI_ADJUST_CFA_OFFSET 4 + popl %es + CFI_ADJUST_CFA_OFFSET -4 popl %ebx CFI_ADJUST_CFA_OFFSET -4 CFI_RESTORE ebx @@ -439,26 +467,47 @@ DST( movb %cl, (%edi) ) CFI_ADJUST_CFA_OFFSET -4 ret CFI_ENDPROC -ENDPROC(csum_partial_copy_generic) +ENDPROC(csum_partial_copy_generic_to_user) #else /* Version for PentiumII/PPro */ #define ROUND1(x) \ + nop; nop; nop; \ SRC(movl x(%esi), %ebx ) ; \ addl %ebx, %eax ; \ - DST(movl %ebx, x(%edi) ) ; + DST(movl %ebx, %es:x(%edi)) ; #define ROUND(x) \ + nop; nop; nop; \ SRC(movl x(%esi), %ebx ) ; \ adcl %ebx, %eax ; \ - DST(movl %ebx, x(%edi) ) ; + DST(movl %ebx, %es:x(%edi)) ; #define ARGBASE 12 - -ENTRY(csum_partial_copy_generic) + +ENTRY(csum_partial_copy_generic_to_user) CFI_STARTPROC + +#ifdef CONFIG_PAX_MEMORY_UDEREF + pushl %gs + CFI_ADJUST_CFA_OFFSET 4 + popl %es + CFI_ADJUST_CFA_OFFSET -4 + jmp csum_partial_copy_generic +#endif + +ENTRY(csum_partial_copy_generic_from_user) + +#ifdef CONFIG_PAX_MEMORY_UDEREF + pushl %gs + CFI_ADJUST_CFA_OFFSET 4 + popl %ds + CFI_ADJUST_CFA_OFFSET -4 +#endif + +ENTRY(csum_partial_copy_generic) pushl %ebx CFI_ADJUST_CFA_OFFSET 4 CFI_REL_OFFSET ebx, 0 @@ -482,7 +531,7 @@ ENTRY(csum_partial_copy_generic) subl %ebx, %edi lea -1(%esi),%edx andl $-32,%edx - lea 3f(%ebx,%ebx), %ebx + lea 3f(%ebx,%ebx,2), %ebx testl %esi, %esi jmp *%ebx 1: addl $64,%esi @@ -503,19 +552,19 @@ ENTRY(csum_partial_copy_generic) jb 5f SRC( movw (%esi), %dx ) leal 2(%esi), %esi -DST( movw %dx, (%edi) ) +DST( movw %dx, %es:(%edi) ) leal 2(%edi), %edi je 6f shll $16,%edx 5: SRC( movb (%esi), %dl ) -DST( movb %dl, (%edi) ) +DST( movb %dl, %es:(%edi) ) 6: addl %edx, %eax adcl $0, %eax 7: .section .fixup, "ax" 6001: movl ARGBASE+20(%esp), %ebx # src_err_ptr - movl $-EFAULT, (%ebx) + movl $-EFAULT, %ss:(%ebx) # zero the complete destination (computing the rest is too much work) movl ARGBASE+8(%esp),%edi # dst movl ARGBASE+12(%esp),%ecx # len @@ -523,10 +572,21 @@ DST( movb %dl, (%edi) ) rep; stosb jmp 7b 6002: movl ARGBASE+24(%esp), %ebx # dst_err_ptr - movl $-EFAULT, (%ebx) + movl $-EFAULT, %ss:(%ebx) jmp 7b .previous +#ifdef CONFIG_PAX_MEMORY_UDEREF + pushl %ss + CFI_ADJUST_CFA_OFFSET 4 + popl %ds + CFI_ADJUST_CFA_OFFSET -4 + pushl %ss + CFI_ADJUST_CFA_OFFSET 4 + popl %es + CFI_ADJUST_CFA_OFFSET -4 +#endif + popl %esi CFI_ADJUST_CFA_OFFSET -4 CFI_RESTORE esi @@ -538,7 +598,7 @@ DST( movb %dl, (%edi) ) CFI_RESTORE ebx ret CFI_ENDPROC -ENDPROC(csum_partial_copy_generic) +ENDPROC(csum_partial_copy_generic_to_user) #undef ROUND #undef ROUND1 diff -urNp linux-2.6.32.48/arch/x86/lib/clear_page_64.S linux-2.6.32.48-openvz/arch/x86/lib/clear_page_64.S --- linux-2.6.32.48/arch/x86/lib/clear_page_64.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/lib/clear_page_64.S 2011-11-17 20:09:04.000000000 -0500 @@ -1,5 +1,6 @@ #include #include +#include /* * Zero a page. @@ -10,6 +11,7 @@ ENTRY(clear_page_c) movl $4096/8,%ecx xorl %eax,%eax rep stosq + pax_force_retaddr ret CFI_ENDPROC ENDPROC(clear_page_c) @@ -33,6 +35,7 @@ ENTRY(clear_page) leaq 64(%rdi),%rdi jnz .Lloop nop + pax_force_retaddr ret CFI_ENDPROC .Lclear_page_end: @@ -43,7 +46,7 @@ ENDPROC(clear_page) #include - .section .altinstr_replacement,"ax" + .section .altinstr_replacement,"a" 1: .byte 0xeb /* jmp */ .byte (clear_page_c - clear_page) - (2f - 1b) /* offset */ 2: diff -urNp linux-2.6.32.48/arch/x86/lib/copy_page_64.S linux-2.6.32.48-openvz/arch/x86/lib/copy_page_64.S --- linux-2.6.32.48/arch/x86/lib/copy_page_64.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/lib/copy_page_64.S 2011-11-17 20:09:04.000000000 -0500 @@ -2,12 +2,14 @@ #include #include +#include ALIGN copy_page_c: CFI_STARTPROC movl $4096/8,%ecx rep movsq + pax_force_retaddr ret CFI_ENDPROC ENDPROC(copy_page_c) @@ -94,6 +96,7 @@ ENTRY(copy_page) CFI_RESTORE r13 addq $3*8,%rsp CFI_ADJUST_CFA_OFFSET -3*8 + pax_force_retaddr ret .Lcopy_page_end: CFI_ENDPROC @@ -104,7 +107,7 @@ ENDPROC(copy_page) #include - .section .altinstr_replacement,"ax" + .section .altinstr_replacement,"a" 1: .byte 0xeb /* jmp */ .byte (copy_page_c - copy_page) - (2f - 1b) /* offset */ 2: diff -urNp linux-2.6.32.48/arch/x86/lib/copy_user_64.S linux-2.6.32.48-openvz/arch/x86/lib/copy_user_64.S --- linux-2.6.32.48/arch/x86/lib/copy_user_64.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/lib/copy_user_64.S 2011-11-17 20:09:04.000000000 -0500 @@ -15,13 +15,15 @@ #include #include #include +#include +#include .macro ALTERNATIVE_JUMP feature,orig,alt 0: .byte 0xe9 /* 32bit jump */ .long \orig-1f /* by default jump to orig */ 1: - .section .altinstr_replacement,"ax" + .section .altinstr_replacement,"a" 2: .byte 0xe9 /* near jump with 32bit immediate */ .long \alt-1b /* offset */ /* or alternatively to alt */ .previous @@ -64,55 +66,26 @@ #endif .endm -/* Standard copy_to_user with segment limit checking */ -ENTRY(copy_to_user) - CFI_STARTPROC - GET_THREAD_INFO(%rax) - movq %rdi,%rcx - addq %rdx,%rcx - jc bad_to_user - cmpq TI_addr_limit(%rax),%rcx - ja bad_to_user - ALTERNATIVE_JUMP X86_FEATURE_REP_GOOD,copy_user_generic_unrolled,copy_user_generic_string - CFI_ENDPROC -ENDPROC(copy_to_user) - -/* Standard copy_from_user with segment limit checking */ -ENTRY(copy_from_user) - CFI_STARTPROC - GET_THREAD_INFO(%rax) - movq %rsi,%rcx - addq %rdx,%rcx - jc bad_from_user - cmpq TI_addr_limit(%rax),%rcx - ja bad_from_user - ALTERNATIVE_JUMP X86_FEATURE_REP_GOOD,copy_user_generic_unrolled,copy_user_generic_string - CFI_ENDPROC -ENDPROC(copy_from_user) - ENTRY(copy_user_generic) CFI_STARTPROC ALTERNATIVE_JUMP X86_FEATURE_REP_GOOD,copy_user_generic_unrolled,copy_user_generic_string CFI_ENDPROC ENDPROC(copy_user_generic) -ENTRY(__copy_from_user_inatomic) - CFI_STARTPROC - ALTERNATIVE_JUMP X86_FEATURE_REP_GOOD,copy_user_generic_unrolled,copy_user_generic_string - CFI_ENDPROC -ENDPROC(__copy_from_user_inatomic) - .section .fixup,"ax" /* must zero dest */ ENTRY(bad_from_user) bad_from_user: CFI_STARTPROC + testl %edx,%edx + js bad_to_user movl %edx,%ecx xorl %eax,%eax rep stosb bad_to_user: movl %edx,%eax + pax_force_retaddr ret CFI_ENDPROC ENDPROC(bad_from_user) @@ -180,6 +153,7 @@ ENTRY(copy_user_generic_unrolled) decl %ecx jnz 21b 23: xor %eax,%eax + pax_force_retaddr ret .section .fixup,"ax" @@ -252,6 +226,7 @@ ENTRY(copy_user_generic_string) 3: rep movsb 4: xorl %eax,%eax + pax_force_retaddr ret .section .fixup,"ax" diff -urNp linux-2.6.32.48/arch/x86/lib/copy_user_nocache_64.S linux-2.6.32.48-openvz/arch/x86/lib/copy_user_nocache_64.S --- linux-2.6.32.48/arch/x86/lib/copy_user_nocache_64.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/lib/copy_user_nocache_64.S 2011-11-17 20:09:04.000000000 -0500 @@ -8,12 +8,14 @@ #include #include +#include #define FIX_ALIGNMENT 1 #include #include #include +#include .macro ALIGN_DESTINATION #ifdef FIX_ALIGNMENT @@ -50,6 +52,15 @@ */ ENTRY(__copy_user_nocache) CFI_STARTPROC + +#ifdef CONFIG_PAX_MEMORY_UDEREF + mov $PAX_USER_SHADOW_BASE,%rcx + cmp %rcx,%rsi + jae 1f + add %rcx,%rsi +1: +#endif + cmpl $8,%edx jb 20f /* less then 8 bytes, go to byte copy loop */ ALIGN_DESTINATION @@ -98,6 +109,7 @@ ENTRY(__copy_user_nocache) jnz 21b 23: xorl %eax,%eax sfence + pax_force_retaddr ret .section .fixup,"ax" diff -urNp linux-2.6.32.48/arch/x86/lib/csum-copy_64.S linux-2.6.32.48-openvz/arch/x86/lib/csum-copy_64.S --- linux-2.6.32.48/arch/x86/lib/csum-copy_64.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/lib/csum-copy_64.S 2011-11-17 20:09:04.000000000 -0500 @@ -8,6 +8,7 @@ #include #include #include +#include /* * Checksum copy with exception handling. @@ -228,6 +229,7 @@ ENTRY(csum_partial_copy_generic) CFI_RESTORE rbp addq $7*8,%rsp CFI_ADJUST_CFA_OFFSET -7*8 + pax_force_retaddr ret CFI_RESTORE_STATE diff -urNp linux-2.6.32.48/arch/x86/lib/csum-wrappers_64.c linux-2.6.32.48-openvz/arch/x86/lib/csum-wrappers_64.c --- linux-2.6.32.48/arch/x86/lib/csum-wrappers_64.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/lib/csum-wrappers_64.c 2011-11-17 20:09:04.000000000 -0500 @@ -52,7 +52,13 @@ csum_partial_copy_from_user(const void _ len -= 2; } } - isum = csum_partial_copy_generic((__force const void *)src, + +#ifdef CONFIG_PAX_MEMORY_UDEREF + if ((unsigned long)src < PAX_USER_SHADOW_BASE) + src += PAX_USER_SHADOW_BASE; +#endif + + isum = csum_partial_copy_generic((const void __force_kernel *)src, dst, len, isum, errp, NULL); if (unlikely(*errp)) goto out_err; @@ -105,7 +111,13 @@ csum_partial_copy_to_user(const void *sr } *errp = 0; - return csum_partial_copy_generic(src, (void __force *)dst, + +#ifdef CONFIG_PAX_MEMORY_UDEREF + if ((unsigned long)dst < PAX_USER_SHADOW_BASE) + dst += PAX_USER_SHADOW_BASE; +#endif + + return csum_partial_copy_generic(src, (void __force_kernel *)dst, len, isum, NULL, errp); } EXPORT_SYMBOL(csum_partial_copy_to_user); diff -urNp linux-2.6.32.48/arch/x86/lib/getuser.S linux-2.6.32.48-openvz/arch/x86/lib/getuser.S --- linux-2.6.32.48/arch/x86/lib/getuser.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/lib/getuser.S 2011-11-17 20:09:04.000000000 -0500 @@ -33,15 +33,38 @@ #include #include #include +#include +#include +#include + +#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_MEMORY_UDEREF) +#define __copyuser_seg gs; +#else +#define __copyuser_seg +#endif .text ENTRY(__get_user_1) CFI_STARTPROC + +#if !defined(CONFIG_X86_32) || !defined(CONFIG_PAX_MEMORY_UDEREF) GET_THREAD_INFO(%_ASM_DX) cmp TI_addr_limit(%_ASM_DX),%_ASM_AX jae bad_get_user -1: movzb (%_ASM_AX),%edx + +#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) + mov $PAX_USER_SHADOW_BASE,%_ASM_DX + cmp %_ASM_DX,%_ASM_AX + jae 1234f + add %_ASM_DX,%_ASM_AX +1234: +#endif + +#endif + +1: __copyuser_seg movzb (%_ASM_AX),%edx xor %eax,%eax + pax_force_retaddr ret CFI_ENDPROC ENDPROC(__get_user_1) @@ -49,12 +72,26 @@ ENDPROC(__get_user_1) ENTRY(__get_user_2) CFI_STARTPROC add $1,%_ASM_AX + +#if !defined(CONFIG_X86_32) || !defined(CONFIG_PAX_MEMORY_UDEREF) jc bad_get_user GET_THREAD_INFO(%_ASM_DX) cmp TI_addr_limit(%_ASM_DX),%_ASM_AX jae bad_get_user -2: movzwl -1(%_ASM_AX),%edx + +#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) + mov $PAX_USER_SHADOW_BASE,%_ASM_DX + cmp %_ASM_DX,%_ASM_AX + jae 1234f + add %_ASM_DX,%_ASM_AX +1234: +#endif + +#endif + +2: __copyuser_seg movzwl -1(%_ASM_AX),%edx xor %eax,%eax + pax_force_retaddr ret CFI_ENDPROC ENDPROC(__get_user_2) @@ -62,12 +99,26 @@ ENDPROC(__get_user_2) ENTRY(__get_user_4) CFI_STARTPROC add $3,%_ASM_AX + +#if !defined(CONFIG_X86_32) || !defined(CONFIG_PAX_MEMORY_UDEREF) jc bad_get_user GET_THREAD_INFO(%_ASM_DX) cmp TI_addr_limit(%_ASM_DX),%_ASM_AX jae bad_get_user -3: mov -3(%_ASM_AX),%edx + +#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) + mov $PAX_USER_SHADOW_BASE,%_ASM_DX + cmp %_ASM_DX,%_ASM_AX + jae 1234f + add %_ASM_DX,%_ASM_AX +1234: +#endif + +#endif + +3: __copyuser_seg mov -3(%_ASM_AX),%edx xor %eax,%eax + pax_force_retaddr ret CFI_ENDPROC ENDPROC(__get_user_4) @@ -80,8 +131,18 @@ ENTRY(__get_user_8) GET_THREAD_INFO(%_ASM_DX) cmp TI_addr_limit(%_ASM_DX),%_ASM_AX jae bad_get_user + +#ifdef CONFIG_PAX_MEMORY_UDEREF + mov $PAX_USER_SHADOW_BASE,%_ASM_DX + cmp %_ASM_DX,%_ASM_AX + jae 1234f + add %_ASM_DX,%_ASM_AX +1234: +#endif + 4: movq -7(%_ASM_AX),%_ASM_DX xor %eax,%eax + pax_force_retaddr ret CFI_ENDPROC ENDPROC(__get_user_8) @@ -91,6 +152,7 @@ bad_get_user: CFI_STARTPROC xor %edx,%edx mov $(-EFAULT),%_ASM_AX + pax_force_retaddr ret CFI_ENDPROC END(bad_get_user) diff -urNp linux-2.6.32.48/arch/x86/lib/iomap_copy_64.S linux-2.6.32.48-openvz/arch/x86/lib/iomap_copy_64.S --- linux-2.6.32.48/arch/x86/lib/iomap_copy_64.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/lib/iomap_copy_64.S 2011-11-17 20:09:04.000000000 -0500 @@ -17,6 +17,7 @@ #include #include +#include /* * override generic version in lib/iomap_copy.c @@ -25,6 +26,7 @@ ENTRY(__iowrite32_copy) CFI_STARTPROC movl %edx,%ecx rep movsd + pax_force_retaddr ret CFI_ENDPROC ENDPROC(__iowrite32_copy) diff -urNp linux-2.6.32.48/arch/x86/lib/memcpy_64.S linux-2.6.32.48-openvz/arch/x86/lib/memcpy_64.S --- linux-2.6.32.48/arch/x86/lib/memcpy_64.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/lib/memcpy_64.S 2011-11-17 20:09:04.000000000 -0500 @@ -4,6 +4,7 @@ #include #include +#include /* * memcpy - Copy a memory block. @@ -34,6 +35,7 @@ memcpy_c: rep movsq movl %edx, %ecx rep movsb + pax_force_retaddr ret CFI_ENDPROC ENDPROC(memcpy_c) @@ -118,6 +120,7 @@ ENTRY(memcpy) jnz .Lloop_1 .Lend: + pax_force_retaddr ret CFI_ENDPROC ENDPROC(memcpy) @@ -128,7 +131,7 @@ ENDPROC(__memcpy) * It is also a lot simpler. Use this when possible: */ - .section .altinstr_replacement, "ax" + .section .altinstr_replacement, "a" 1: .byte 0xeb /* jmp */ .byte (memcpy_c - memcpy) - (2f - 1b) /* offset */ 2: diff -urNp linux-2.6.32.48/arch/x86/lib/memset_64.S linux-2.6.32.48-openvz/arch/x86/lib/memset_64.S --- linux-2.6.32.48/arch/x86/lib/memset_64.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/lib/memset_64.S 2011-11-17 20:09:04.000000000 -0500 @@ -2,6 +2,7 @@ #include #include +#include /* * ISO C memset - set a memory block to a byte value. @@ -28,6 +29,7 @@ memset_c: movl %r8d,%ecx rep stosb movq %r9,%rax + pax_force_retaddr ret CFI_ENDPROC ENDPROC(memset_c) @@ -96,6 +98,7 @@ ENTRY(__memset) .Lende: movq %r10,%rax + pax_force_retaddr ret CFI_RESTORE_STATE @@ -118,7 +121,7 @@ ENDPROC(__memset) #include - .section .altinstr_replacement,"ax" + .section .altinstr_replacement,"a" 1: .byte 0xeb /* jmp */ .byte (memset_c - memset) - (2f - 1b) /* offset */ 2: diff -urNp linux-2.6.32.48/arch/x86/lib/mmx_32.c linux-2.6.32.48-openvz/arch/x86/lib/mmx_32.c --- linux-2.6.32.48/arch/x86/lib/mmx_32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/lib/mmx_32.c 2011-11-17 20:08:53.000000000 -0500 @@ -29,6 +29,7 @@ void *_mmx_memcpy(void *to, const void * { void *p; int i; + unsigned long cr0; if (unlikely(in_interrupt())) return __memcpy(to, from, len); @@ -39,44 +40,72 @@ void *_mmx_memcpy(void *to, const void * kernel_fpu_begin(); __asm__ __volatile__ ( - "1: prefetch (%0)\n" /* This set is 28 bytes */ - " prefetch 64(%0)\n" - " prefetch 128(%0)\n" - " prefetch 192(%0)\n" - " prefetch 256(%0)\n" + "1: prefetch (%1)\n" /* This set is 28 bytes */ + " prefetch 64(%1)\n" + " prefetch 128(%1)\n" + " prefetch 192(%1)\n" + " prefetch 256(%1)\n" "2: \n" ".section .fixup, \"ax\"\n" - "3: movw $0x1AEB, 1b\n" /* jmp on 26 bytes */ + "3: \n" + +#ifdef CONFIG_PAX_KERNEXEC + " movl %%cr0, %0\n" + " movl %0, %%eax\n" + " andl $0xFFFEFFFF, %%eax\n" + " movl %%eax, %%cr0\n" +#endif + + " movw $0x1AEB, 1b\n" /* jmp on 26 bytes */ + +#ifdef CONFIG_PAX_KERNEXEC + " movl %0, %%cr0\n" +#endif + " jmp 2b\n" ".previous\n" _ASM_EXTABLE(1b, 3b) - : : "r" (from)); + : "=&r" (cr0) : "r" (from) : "ax"); for ( ; i > 5; i--) { __asm__ __volatile__ ( - "1: prefetch 320(%0)\n" - "2: movq (%0), %%mm0\n" - " movq 8(%0), %%mm1\n" - " movq 16(%0), %%mm2\n" - " movq 24(%0), %%mm3\n" - " movq %%mm0, (%1)\n" - " movq %%mm1, 8(%1)\n" - " movq %%mm2, 16(%1)\n" - " movq %%mm3, 24(%1)\n" - " movq 32(%0), %%mm0\n" - " movq 40(%0), %%mm1\n" - " movq 48(%0), %%mm2\n" - " movq 56(%0), %%mm3\n" - " movq %%mm0, 32(%1)\n" - " movq %%mm1, 40(%1)\n" - " movq %%mm2, 48(%1)\n" - " movq %%mm3, 56(%1)\n" + "1: prefetch 320(%1)\n" + "2: movq (%1), %%mm0\n" + " movq 8(%1), %%mm1\n" + " movq 16(%1), %%mm2\n" + " movq 24(%1), %%mm3\n" + " movq %%mm0, (%2)\n" + " movq %%mm1, 8(%2)\n" + " movq %%mm2, 16(%2)\n" + " movq %%mm3, 24(%2)\n" + " movq 32(%1), %%mm0\n" + " movq 40(%1), %%mm1\n" + " movq 48(%1), %%mm2\n" + " movq 56(%1), %%mm3\n" + " movq %%mm0, 32(%2)\n" + " movq %%mm1, 40(%2)\n" + " movq %%mm2, 48(%2)\n" + " movq %%mm3, 56(%2)\n" ".section .fixup, \"ax\"\n" - "3: movw $0x05EB, 1b\n" /* jmp on 5 bytes */ + "3:\n" + +#ifdef CONFIG_PAX_KERNEXEC + " movl %%cr0, %0\n" + " movl %0, %%eax\n" + " andl $0xFFFEFFFF, %%eax\n" + " movl %%eax, %%cr0\n" +#endif + + " movw $0x05EB, 1b\n" /* jmp on 5 bytes */ + +#ifdef CONFIG_PAX_KERNEXEC + " movl %0, %%cr0\n" +#endif + " jmp 2b\n" ".previous\n" _ASM_EXTABLE(1b, 3b) - : : "r" (from), "r" (to) : "memory"); + : "=&r" (cr0) : "r" (from), "r" (to) : "memory", "ax"); from += 64; to += 64; @@ -158,6 +187,7 @@ static void fast_clear_page(void *page) static void fast_copy_page(void *to, void *from) { int i; + unsigned long cr0; kernel_fpu_begin(); @@ -166,42 +196,70 @@ static void fast_copy_page(void *to, voi * but that is for later. -AV */ __asm__ __volatile__( - "1: prefetch (%0)\n" - " prefetch 64(%0)\n" - " prefetch 128(%0)\n" - " prefetch 192(%0)\n" - " prefetch 256(%0)\n" + "1: prefetch (%1)\n" + " prefetch 64(%1)\n" + " prefetch 128(%1)\n" + " prefetch 192(%1)\n" + " prefetch 256(%1)\n" "2: \n" ".section .fixup, \"ax\"\n" - "3: movw $0x1AEB, 1b\n" /* jmp on 26 bytes */ + "3: \n" + +#ifdef CONFIG_PAX_KERNEXEC + " movl %%cr0, %0\n" + " movl %0, %%eax\n" + " andl $0xFFFEFFFF, %%eax\n" + " movl %%eax, %%cr0\n" +#endif + + " movw $0x1AEB, 1b\n" /* jmp on 26 bytes */ + +#ifdef CONFIG_PAX_KERNEXEC + " movl %0, %%cr0\n" +#endif + " jmp 2b\n" ".previous\n" - _ASM_EXTABLE(1b, 3b) : : "r" (from)); + _ASM_EXTABLE(1b, 3b) : "=&r" (cr0) : "r" (from) : "ax"); for (i = 0; i < (4096-320)/64; i++) { __asm__ __volatile__ ( - "1: prefetch 320(%0)\n" - "2: movq (%0), %%mm0\n" - " movntq %%mm0, (%1)\n" - " movq 8(%0), %%mm1\n" - " movntq %%mm1, 8(%1)\n" - " movq 16(%0), %%mm2\n" - " movntq %%mm2, 16(%1)\n" - " movq 24(%0), %%mm3\n" - " movntq %%mm3, 24(%1)\n" - " movq 32(%0), %%mm4\n" - " movntq %%mm4, 32(%1)\n" - " movq 40(%0), %%mm5\n" - " movntq %%mm5, 40(%1)\n" - " movq 48(%0), %%mm6\n" - " movntq %%mm6, 48(%1)\n" - " movq 56(%0), %%mm7\n" - " movntq %%mm7, 56(%1)\n" + "1: prefetch 320(%1)\n" + "2: movq (%1), %%mm0\n" + " movntq %%mm0, (%2)\n" + " movq 8(%1), %%mm1\n" + " movntq %%mm1, 8(%2)\n" + " movq 16(%1), %%mm2\n" + " movntq %%mm2, 16(%2)\n" + " movq 24(%1), %%mm3\n" + " movntq %%mm3, 24(%2)\n" + " movq 32(%1), %%mm4\n" + " movntq %%mm4, 32(%2)\n" + " movq 40(%1), %%mm5\n" + " movntq %%mm5, 40(%2)\n" + " movq 48(%1), %%mm6\n" + " movntq %%mm6, 48(%2)\n" + " movq 56(%1), %%mm7\n" + " movntq %%mm7, 56(%2)\n" ".section .fixup, \"ax\"\n" - "3: movw $0x05EB, 1b\n" /* jmp on 5 bytes */ + "3:\n" + +#ifdef CONFIG_PAX_KERNEXEC + " movl %%cr0, %0\n" + " movl %0, %%eax\n" + " andl $0xFFFEFFFF, %%eax\n" + " movl %%eax, %%cr0\n" +#endif + + " movw $0x05EB, 1b\n" /* jmp on 5 bytes */ + +#ifdef CONFIG_PAX_KERNEXEC + " movl %0, %%cr0\n" +#endif + " jmp 2b\n" ".previous\n" - _ASM_EXTABLE(1b, 3b) : : "r" (from), "r" (to) : "memory"); + _ASM_EXTABLE(1b, 3b) : "=&r" (cr0) : "r" (from), "r" (to) : "memory", "ax"); from += 64; to += 64; @@ -280,47 +338,76 @@ static void fast_clear_page(void *page) static void fast_copy_page(void *to, void *from) { int i; + unsigned long cr0; kernel_fpu_begin(); __asm__ __volatile__ ( - "1: prefetch (%0)\n" - " prefetch 64(%0)\n" - " prefetch 128(%0)\n" - " prefetch 192(%0)\n" - " prefetch 256(%0)\n" + "1: prefetch (%1)\n" + " prefetch 64(%1)\n" + " prefetch 128(%1)\n" + " prefetch 192(%1)\n" + " prefetch 256(%1)\n" "2: \n" ".section .fixup, \"ax\"\n" - "3: movw $0x1AEB, 1b\n" /* jmp on 26 bytes */ + "3: \n" + +#ifdef CONFIG_PAX_KERNEXEC + " movl %%cr0, %0\n" + " movl %0, %%eax\n" + " andl $0xFFFEFFFF, %%eax\n" + " movl %%eax, %%cr0\n" +#endif + + " movw $0x1AEB, 1b\n" /* jmp on 26 bytes */ + +#ifdef CONFIG_PAX_KERNEXEC + " movl %0, %%cr0\n" +#endif + " jmp 2b\n" ".previous\n" - _ASM_EXTABLE(1b, 3b) : : "r" (from)); + _ASM_EXTABLE(1b, 3b) : "=&r" (cr0) : "r" (from) : "ax"); for (i = 0; i < 4096/64; i++) { __asm__ __volatile__ ( - "1: prefetch 320(%0)\n" - "2: movq (%0), %%mm0\n" - " movq 8(%0), %%mm1\n" - " movq 16(%0), %%mm2\n" - " movq 24(%0), %%mm3\n" - " movq %%mm0, (%1)\n" - " movq %%mm1, 8(%1)\n" - " movq %%mm2, 16(%1)\n" - " movq %%mm3, 24(%1)\n" - " movq 32(%0), %%mm0\n" - " movq 40(%0), %%mm1\n" - " movq 48(%0), %%mm2\n" - " movq 56(%0), %%mm3\n" - " movq %%mm0, 32(%1)\n" - " movq %%mm1, 40(%1)\n" - " movq %%mm2, 48(%1)\n" - " movq %%mm3, 56(%1)\n" + "1: prefetch 320(%1)\n" + "2: movq (%1), %%mm0\n" + " movq 8(%1), %%mm1\n" + " movq 16(%1), %%mm2\n" + " movq 24(%1), %%mm3\n" + " movq %%mm0, (%2)\n" + " movq %%mm1, 8(%2)\n" + " movq %%mm2, 16(%2)\n" + " movq %%mm3, 24(%2)\n" + " movq 32(%1), %%mm0\n" + " movq 40(%1), %%mm1\n" + " movq 48(%1), %%mm2\n" + " movq 56(%1), %%mm3\n" + " movq %%mm0, 32(%2)\n" + " movq %%mm1, 40(%2)\n" + " movq %%mm2, 48(%2)\n" + " movq %%mm3, 56(%2)\n" ".section .fixup, \"ax\"\n" - "3: movw $0x05EB, 1b\n" /* jmp on 5 bytes */ + "3:\n" + +#ifdef CONFIG_PAX_KERNEXEC + " movl %%cr0, %0\n" + " movl %0, %%eax\n" + " andl $0xFFFEFFFF, %%eax\n" + " movl %%eax, %%cr0\n" +#endif + + " movw $0x05EB, 1b\n" /* jmp on 5 bytes */ + +#ifdef CONFIG_PAX_KERNEXEC + " movl %0, %%cr0\n" +#endif + " jmp 2b\n" ".previous\n" _ASM_EXTABLE(1b, 3b) - : : "r" (from), "r" (to) : "memory"); + : "=&r" (cr0) : "r" (from), "r" (to) : "memory", "ax"); from += 64; to += 64; diff -urNp linux-2.6.32.48/arch/x86/lib/msr-reg.S linux-2.6.32.48-openvz/arch/x86/lib/msr-reg.S --- linux-2.6.32.48/arch/x86/lib/msr-reg.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/lib/msr-reg.S 2011-11-17 20:09:04.000000000 -0500 @@ -3,6 +3,7 @@ #include #include #include +#include #ifdef CONFIG_X86_64 /* @@ -37,6 +38,7 @@ ENTRY(native_\op\()_safe_regs) movl %edi, 28(%r10) popq_cfi %rbp popq_cfi %rbx + pax_force_retaddr ret 3: CFI_RESTORE_STATE diff -urNp linux-2.6.32.48/arch/x86/lib/putuser.S linux-2.6.32.48-openvz/arch/x86/lib/putuser.S --- linux-2.6.32.48/arch/x86/lib/putuser.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/lib/putuser.S 2011-11-17 20:09:04.000000000 -0500 @@ -15,7 +15,9 @@ #include #include #include - +#include +#include +#include /* * __put_user_X @@ -29,52 +31,119 @@ * as they get called from within inline assembly. */ -#define ENTER CFI_STARTPROC ; \ - GET_THREAD_INFO(%_ASM_BX) -#define EXIT ret ; \ +#define ENTER CFI_STARTPROC +#define EXIT pax_force_retaddr; ret ; \ CFI_ENDPROC +#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) +#define _DEST %_ASM_CX,%_ASM_BX +#else +#define _DEST %_ASM_CX +#endif + +#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_MEMORY_UDEREF) +#define __copyuser_seg gs; +#else +#define __copyuser_seg +#endif + .text ENTRY(__put_user_1) ENTER + +#if !defined(CONFIG_X86_32) || !defined(CONFIG_PAX_MEMORY_UDEREF) + GET_THREAD_INFO(%_ASM_BX) cmp TI_addr_limit(%_ASM_BX),%_ASM_CX jae bad_put_user -1: movb %al,(%_ASM_CX) + +#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) + mov $PAX_USER_SHADOW_BASE,%_ASM_BX + cmp %_ASM_BX,%_ASM_CX + jb 1234f + xor %ebx,%ebx +1234: +#endif + +#endif + +1: __copyuser_seg movb %al,(_DEST) xor %eax,%eax EXIT ENDPROC(__put_user_1) ENTRY(__put_user_2) ENTER + +#if !defined(CONFIG_X86_32) || !defined(CONFIG_PAX_MEMORY_UDEREF) + GET_THREAD_INFO(%_ASM_BX) mov TI_addr_limit(%_ASM_BX),%_ASM_BX sub $1,%_ASM_BX cmp %_ASM_BX,%_ASM_CX jae bad_put_user -2: movw %ax,(%_ASM_CX) + +#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) + mov $PAX_USER_SHADOW_BASE,%_ASM_BX + cmp %_ASM_BX,%_ASM_CX + jb 1234f + xor %ebx,%ebx +1234: +#endif + +#endif + +2: __copyuser_seg movw %ax,(_DEST) xor %eax,%eax EXIT ENDPROC(__put_user_2) ENTRY(__put_user_4) ENTER + +#if !defined(CONFIG_X86_32) || !defined(CONFIG_PAX_MEMORY_UDEREF) + GET_THREAD_INFO(%_ASM_BX) mov TI_addr_limit(%_ASM_BX),%_ASM_BX sub $3,%_ASM_BX cmp %_ASM_BX,%_ASM_CX jae bad_put_user -3: movl %eax,(%_ASM_CX) + +#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) + mov $PAX_USER_SHADOW_BASE,%_ASM_BX + cmp %_ASM_BX,%_ASM_CX + jb 1234f + xor %ebx,%ebx +1234: +#endif + +#endif + +3: __copyuser_seg movl %eax,(_DEST) xor %eax,%eax EXIT ENDPROC(__put_user_4) ENTRY(__put_user_8) ENTER + +#if !defined(CONFIG_X86_32) || !defined(CONFIG_PAX_MEMORY_UDEREF) + GET_THREAD_INFO(%_ASM_BX) mov TI_addr_limit(%_ASM_BX),%_ASM_BX sub $7,%_ASM_BX cmp %_ASM_BX,%_ASM_CX jae bad_put_user -4: mov %_ASM_AX,(%_ASM_CX) + +#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) + mov $PAX_USER_SHADOW_BASE,%_ASM_BX + cmp %_ASM_BX,%_ASM_CX + jb 1234f + xor %ebx,%ebx +1234: +#endif + +#endif + +4: __copyuser_seg mov %_ASM_AX,(_DEST) #ifdef CONFIG_X86_32 -5: movl %edx,4(%_ASM_CX) +5: __copyuser_seg movl %edx,4(_DEST) #endif xor %eax,%eax EXIT diff -urNp linux-2.6.32.48/arch/x86/lib/rwlock_64.S linux-2.6.32.48-openvz/arch/x86/lib/rwlock_64.S --- linux-2.6.32.48/arch/x86/lib/rwlock_64.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/lib/rwlock_64.S 2011-11-17 20:09:04.000000000 -0500 @@ -17,6 +17,7 @@ ENTRY(__write_lock_failed) LOCK_PREFIX subl $RW_LOCK_BIAS,(%rdi) jnz __write_lock_failed + pax_force_retaddr ret CFI_ENDPROC END(__write_lock_failed) @@ -33,6 +34,7 @@ ENTRY(__read_lock_failed) LOCK_PREFIX decl (%rdi) js __read_lock_failed + pax_force_retaddr ret CFI_ENDPROC END(__read_lock_failed) diff -urNp linux-2.6.32.48/arch/x86/lib/rwsem_64.S linux-2.6.32.48-openvz/arch/x86/lib/rwsem_64.S --- linux-2.6.32.48/arch/x86/lib/rwsem_64.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/lib/rwsem_64.S 2011-11-17 20:09:04.000000000 -0500 @@ -48,6 +48,7 @@ ENTRY(call_rwsem_down_read_failed) call rwsem_down_read_failed popq %rdx restore_common_regs + pax_force_retaddr ret ENDPROC(call_rwsem_down_read_failed) @@ -56,6 +57,7 @@ ENTRY(call_rwsem_down_write_failed) movq %rax,%rdi call rwsem_down_write_failed restore_common_regs + pax_force_retaddr ret ENDPROC(call_rwsem_down_write_failed) @@ -66,7 +68,8 @@ ENTRY(call_rwsem_wake) movq %rax,%rdi call rwsem_wake restore_common_regs -1: ret +1: pax_force_retaddr + ret ENDPROC(call_rwsem_wake) /* Fix up special calling conventions */ @@ -77,5 +80,6 @@ ENTRY(call_rwsem_downgrade_wake) call rwsem_downgrade_wake popq %rdx restore_common_regs + pax_force_retaddr ret ENDPROC(call_rwsem_downgrade_wake) diff -urNp linux-2.6.32.48/arch/x86/lib/thunk_64.S linux-2.6.32.48-openvz/arch/x86/lib/thunk_64.S --- linux-2.6.32.48/arch/x86/lib/thunk_64.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/lib/thunk_64.S 2011-11-17 20:09:04.000000000 -0500 @@ -10,7 +10,8 @@ #include #include #include - + #include + /* rdi: arg1 ... normal C conventions. rax is saved/restored. */ .macro thunk name,func .globl \name @@ -70,6 +71,7 @@ SAVE_ARGS restore: RESTORE_ARGS + pax_force_retaddr ret CFI_ENDPROC @@ -77,5 +79,6 @@ restore: SAVE_ARGS restore_norax: RESTORE_ARGS 1 + pax_force_retaddr ret CFI_ENDPROC diff -urNp linux-2.6.32.48/arch/x86/lib/usercopy_32.c linux-2.6.32.48-openvz/arch/x86/lib/usercopy_32.c --- linux-2.6.32.48/arch/x86/lib/usercopy_32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/lib/usercopy_32.c 2011-11-17 20:08:53.000000000 -0500 @@ -43,7 +43,7 @@ do { \ __asm__ __volatile__( \ " testl %1,%1\n" \ " jz 2f\n" \ - "0: lodsb\n" \ + "0: "__copyuser_seg"lodsb\n" \ " stosb\n" \ " testb %%al,%%al\n" \ " jz 1f\n" \ @@ -128,10 +128,12 @@ do { \ int __d0; \ might_fault(); \ __asm__ __volatile__( \ + __COPYUSER_SET_ES \ "0: rep; stosl\n" \ " movl %2,%0\n" \ "1: rep; stosb\n" \ "2:\n" \ + __COPYUSER_RESTORE_ES \ ".section .fixup,\"ax\"\n" \ "3: lea 0(%2,%0,4),%0\n" \ " jmp 2b\n" \ @@ -200,6 +202,7 @@ long strnlen_user(const char __user *s, might_fault(); __asm__ __volatile__( + __COPYUSER_SET_ES " testl %0, %0\n" " jz 3f\n" " andl %0,%%ecx\n" @@ -208,6 +211,7 @@ long strnlen_user(const char __user *s, " subl %%ecx,%0\n" " addl %0,%%eax\n" "1:\n" + __COPYUSER_RESTORE_ES ".section .fixup,\"ax\"\n" "2: xorl %%eax,%%eax\n" " jmp 1b\n" @@ -227,7 +231,7 @@ EXPORT_SYMBOL(strnlen_user); #ifdef CONFIG_X86_INTEL_USERCOPY static unsigned long -__copy_user_intel(void __user *to, const void *from, unsigned long size) +__generic_copy_to_user_intel(void __user *to, const void *from, unsigned long size) { int d0, d1; __asm__ __volatile__( @@ -239,36 +243,36 @@ __copy_user_intel(void __user *to, const " .align 2,0x90\n" "3: movl 0(%4), %%eax\n" "4: movl 4(%4), %%edx\n" - "5: movl %%eax, 0(%3)\n" - "6: movl %%edx, 4(%3)\n" + "5: "__copyuser_seg" movl %%eax, 0(%3)\n" + "6: "__copyuser_seg" movl %%edx, 4(%3)\n" "7: movl 8(%4), %%eax\n" "8: movl 12(%4),%%edx\n" - "9: movl %%eax, 8(%3)\n" - "10: movl %%edx, 12(%3)\n" + "9: "__copyuser_seg" movl %%eax, 8(%3)\n" + "10: "__copyuser_seg" movl %%edx, 12(%3)\n" "11: movl 16(%4), %%eax\n" "12: movl 20(%4), %%edx\n" - "13: movl %%eax, 16(%3)\n" - "14: movl %%edx, 20(%3)\n" + "13: "__copyuser_seg" movl %%eax, 16(%3)\n" + "14: "__copyuser_seg" movl %%edx, 20(%3)\n" "15: movl 24(%4), %%eax\n" "16: movl 28(%4), %%edx\n" - "17: movl %%eax, 24(%3)\n" - "18: movl %%edx, 28(%3)\n" + "17: "__copyuser_seg" movl %%eax, 24(%3)\n" + "18: "__copyuser_seg" movl %%edx, 28(%3)\n" "19: movl 32(%4), %%eax\n" "20: movl 36(%4), %%edx\n" - "21: movl %%eax, 32(%3)\n" - "22: movl %%edx, 36(%3)\n" + "21: "__copyuser_seg" movl %%eax, 32(%3)\n" + "22: "__copyuser_seg" movl %%edx, 36(%3)\n" "23: movl 40(%4), %%eax\n" "24: movl 44(%4), %%edx\n" - "25: movl %%eax, 40(%3)\n" - "26: movl %%edx, 44(%3)\n" + "25: "__copyuser_seg" movl %%eax, 40(%3)\n" + "26: "__copyuser_seg" movl %%edx, 44(%3)\n" "27: movl 48(%4), %%eax\n" "28: movl 52(%4), %%edx\n" - "29: movl %%eax, 48(%3)\n" - "30: movl %%edx, 52(%3)\n" + "29: "__copyuser_seg" movl %%eax, 48(%3)\n" + "30: "__copyuser_seg" movl %%edx, 52(%3)\n" "31: movl 56(%4), %%eax\n" "32: movl 60(%4), %%edx\n" - "33: movl %%eax, 56(%3)\n" - "34: movl %%edx, 60(%3)\n" + "33: "__copyuser_seg" movl %%eax, 56(%3)\n" + "34: "__copyuser_seg" movl %%edx, 60(%3)\n" " addl $-64, %0\n" " addl $64, %4\n" " addl $64, %3\n" @@ -278,10 +282,119 @@ __copy_user_intel(void __user *to, const " shrl $2, %0\n" " andl $3, %%eax\n" " cld\n" + __COPYUSER_SET_ES "99: rep; movsl\n" "36: movl %%eax, %0\n" "37: rep; movsb\n" "100:\n" + __COPYUSER_RESTORE_ES + ".section .fixup,\"ax\"\n" + "101: lea 0(%%eax,%0,4),%0\n" + " jmp 100b\n" + ".previous\n" + ".section __ex_table,\"a\"\n" + " .align 4\n" + " .long 1b,100b\n" + " .long 2b,100b\n" + " .long 3b,100b\n" + " .long 4b,100b\n" + " .long 5b,100b\n" + " .long 6b,100b\n" + " .long 7b,100b\n" + " .long 8b,100b\n" + " .long 9b,100b\n" + " .long 10b,100b\n" + " .long 11b,100b\n" + " .long 12b,100b\n" + " .long 13b,100b\n" + " .long 14b,100b\n" + " .long 15b,100b\n" + " .long 16b,100b\n" + " .long 17b,100b\n" + " .long 18b,100b\n" + " .long 19b,100b\n" + " .long 20b,100b\n" + " .long 21b,100b\n" + " .long 22b,100b\n" + " .long 23b,100b\n" + " .long 24b,100b\n" + " .long 25b,100b\n" + " .long 26b,100b\n" + " .long 27b,100b\n" + " .long 28b,100b\n" + " .long 29b,100b\n" + " .long 30b,100b\n" + " .long 31b,100b\n" + " .long 32b,100b\n" + " .long 33b,100b\n" + " .long 34b,100b\n" + " .long 35b,100b\n" + " .long 36b,100b\n" + " .long 37b,100b\n" + " .long 99b,101b\n" + ".previous" + : "=&c"(size), "=&D" (d0), "=&S" (d1) + : "1"(to), "2"(from), "0"(size) + : "eax", "edx", "memory"); + return size; +} + +static unsigned long +__generic_copy_from_user_intel(void *to, const void __user *from, unsigned long size) +{ + int d0, d1; + __asm__ __volatile__( + " .align 2,0x90\n" + "1: "__copyuser_seg" movl 32(%4), %%eax\n" + " cmpl $67, %0\n" + " jbe 3f\n" + "2: "__copyuser_seg" movl 64(%4), %%eax\n" + " .align 2,0x90\n" + "3: "__copyuser_seg" movl 0(%4), %%eax\n" + "4: "__copyuser_seg" movl 4(%4), %%edx\n" + "5: movl %%eax, 0(%3)\n" + "6: movl %%edx, 4(%3)\n" + "7: "__copyuser_seg" movl 8(%4), %%eax\n" + "8: "__copyuser_seg" movl 12(%4),%%edx\n" + "9: movl %%eax, 8(%3)\n" + "10: movl %%edx, 12(%3)\n" + "11: "__copyuser_seg" movl 16(%4), %%eax\n" + "12: "__copyuser_seg" movl 20(%4), %%edx\n" + "13: movl %%eax, 16(%3)\n" + "14: movl %%edx, 20(%3)\n" + "15: "__copyuser_seg" movl 24(%4), %%eax\n" + "16: "__copyuser_seg" movl 28(%4), %%edx\n" + "17: movl %%eax, 24(%3)\n" + "18: movl %%edx, 28(%3)\n" + "19: "__copyuser_seg" movl 32(%4), %%eax\n" + "20: "__copyuser_seg" movl 36(%4), %%edx\n" + "21: movl %%eax, 32(%3)\n" + "22: movl %%edx, 36(%3)\n" + "23: "__copyuser_seg" movl 40(%4), %%eax\n" + "24: "__copyuser_seg" movl 44(%4), %%edx\n" + "25: movl %%eax, 40(%3)\n" + "26: movl %%edx, 44(%3)\n" + "27: "__copyuser_seg" movl 48(%4), %%eax\n" + "28: "__copyuser_seg" movl 52(%4), %%edx\n" + "29: movl %%eax, 48(%3)\n" + "30: movl %%edx, 52(%3)\n" + "31: "__copyuser_seg" movl 56(%4), %%eax\n" + "32: "__copyuser_seg" movl 60(%4), %%edx\n" + "33: movl %%eax, 56(%3)\n" + "34: movl %%edx, 60(%3)\n" + " addl $-64, %0\n" + " addl $64, %4\n" + " addl $64, %3\n" + " cmpl $63, %0\n" + " ja 1b\n" + "35: movl %0, %%eax\n" + " shrl $2, %0\n" + " andl $3, %%eax\n" + " cld\n" + "99: rep; "__copyuser_seg" movsl\n" + "36: movl %%eax, %0\n" + "37: rep; "__copyuser_seg" movsb\n" + "100:\n" ".section .fixup,\"ax\"\n" "101: lea 0(%%eax,%0,4),%0\n" " jmp 100b\n" @@ -339,41 +452,41 @@ __copy_user_zeroing_intel(void *to, cons int d0, d1; __asm__ __volatile__( " .align 2,0x90\n" - "0: movl 32(%4), %%eax\n" + "0: "__copyuser_seg" movl 32(%4), %%eax\n" " cmpl $67, %0\n" " jbe 2f\n" - "1: movl 64(%4), %%eax\n" + "1: "__copyuser_seg" movl 64(%4), %%eax\n" " .align 2,0x90\n" - "2: movl 0(%4), %%eax\n" - "21: movl 4(%4), %%edx\n" + "2: "__copyuser_seg" movl 0(%4), %%eax\n" + "21: "__copyuser_seg" movl 4(%4), %%edx\n" " movl %%eax, 0(%3)\n" " movl %%edx, 4(%3)\n" - "3: movl 8(%4), %%eax\n" - "31: movl 12(%4),%%edx\n" + "3: "__copyuser_seg" movl 8(%4), %%eax\n" + "31: "__copyuser_seg" movl 12(%4),%%edx\n" " movl %%eax, 8(%3)\n" " movl %%edx, 12(%3)\n" - "4: movl 16(%4), %%eax\n" - "41: movl 20(%4), %%edx\n" + "4: "__copyuser_seg" movl 16(%4), %%eax\n" + "41: "__copyuser_seg" movl 20(%4), %%edx\n" " movl %%eax, 16(%3)\n" " movl %%edx, 20(%3)\n" - "10: movl 24(%4), %%eax\n" - "51: movl 28(%4), %%edx\n" + "10: "__copyuser_seg" movl 24(%4), %%eax\n" + "51: "__copyuser_seg" movl 28(%4), %%edx\n" " movl %%eax, 24(%3)\n" " movl %%edx, 28(%3)\n" - "11: movl 32(%4), %%eax\n" - "61: movl 36(%4), %%edx\n" + "11: "__copyuser_seg" movl 32(%4), %%eax\n" + "61: "__copyuser_seg" movl 36(%4), %%edx\n" " movl %%eax, 32(%3)\n" " movl %%edx, 36(%3)\n" - "12: movl 40(%4), %%eax\n" - "71: movl 44(%4), %%edx\n" + "12: "__copyuser_seg" movl 40(%4), %%eax\n" + "71: "__copyuser_seg" movl 44(%4), %%edx\n" " movl %%eax, 40(%3)\n" " movl %%edx, 44(%3)\n" - "13: movl 48(%4), %%eax\n" - "81: movl 52(%4), %%edx\n" + "13: "__copyuser_seg" movl 48(%4), %%eax\n" + "81: "__copyuser_seg" movl 52(%4), %%edx\n" " movl %%eax, 48(%3)\n" " movl %%edx, 52(%3)\n" - "14: movl 56(%4), %%eax\n" - "91: movl 60(%4), %%edx\n" + "14: "__copyuser_seg" movl 56(%4), %%eax\n" + "91: "__copyuser_seg" movl 60(%4), %%edx\n" " movl %%eax, 56(%3)\n" " movl %%edx, 60(%3)\n" " addl $-64, %0\n" @@ -385,9 +498,9 @@ __copy_user_zeroing_intel(void *to, cons " shrl $2, %0\n" " andl $3, %%eax\n" " cld\n" - "6: rep; movsl\n" + "6: rep; "__copyuser_seg" movsl\n" " movl %%eax,%0\n" - "7: rep; movsb\n" + "7: rep; "__copyuser_seg" movsb\n" "8:\n" ".section .fixup,\"ax\"\n" "9: lea 0(%%eax,%0,4),%0\n" @@ -440,41 +553,41 @@ static unsigned long __copy_user_zeroing __asm__ __volatile__( " .align 2,0x90\n" - "0: movl 32(%4), %%eax\n" + "0: "__copyuser_seg" movl 32(%4), %%eax\n" " cmpl $67, %0\n" " jbe 2f\n" - "1: movl 64(%4), %%eax\n" + "1: "__copyuser_seg" movl 64(%4), %%eax\n" " .align 2,0x90\n" - "2: movl 0(%4), %%eax\n" - "21: movl 4(%4), %%edx\n" + "2: "__copyuser_seg" movl 0(%4), %%eax\n" + "21: "__copyuser_seg" movl 4(%4), %%edx\n" " movnti %%eax, 0(%3)\n" " movnti %%edx, 4(%3)\n" - "3: movl 8(%4), %%eax\n" - "31: movl 12(%4),%%edx\n" + "3: "__copyuser_seg" movl 8(%4), %%eax\n" + "31: "__copyuser_seg" movl 12(%4),%%edx\n" " movnti %%eax, 8(%3)\n" " movnti %%edx, 12(%3)\n" - "4: movl 16(%4), %%eax\n" - "41: movl 20(%4), %%edx\n" + "4: "__copyuser_seg" movl 16(%4), %%eax\n" + "41: "__copyuser_seg" movl 20(%4), %%edx\n" " movnti %%eax, 16(%3)\n" " movnti %%edx, 20(%3)\n" - "10: movl 24(%4), %%eax\n" - "51: movl 28(%4), %%edx\n" + "10: "__copyuser_seg" movl 24(%4), %%eax\n" + "51: "__copyuser_seg" movl 28(%4), %%edx\n" " movnti %%eax, 24(%3)\n" " movnti %%edx, 28(%3)\n" - "11: movl 32(%4), %%eax\n" - "61: movl 36(%4), %%edx\n" + "11: "__copyuser_seg" movl 32(%4), %%eax\n" + "61: "__copyuser_seg" movl 36(%4), %%edx\n" " movnti %%eax, 32(%3)\n" " movnti %%edx, 36(%3)\n" - "12: movl 40(%4), %%eax\n" - "71: movl 44(%4), %%edx\n" + "12: "__copyuser_seg" movl 40(%4), %%eax\n" + "71: "__copyuser_seg" movl 44(%4), %%edx\n" " movnti %%eax, 40(%3)\n" " movnti %%edx, 44(%3)\n" - "13: movl 48(%4), %%eax\n" - "81: movl 52(%4), %%edx\n" + "13: "__copyuser_seg" movl 48(%4), %%eax\n" + "81: "__copyuser_seg" movl 52(%4), %%edx\n" " movnti %%eax, 48(%3)\n" " movnti %%edx, 52(%3)\n" - "14: movl 56(%4), %%eax\n" - "91: movl 60(%4), %%edx\n" + "14: "__copyuser_seg" movl 56(%4), %%eax\n" + "91: "__copyuser_seg" movl 60(%4), %%edx\n" " movnti %%eax, 56(%3)\n" " movnti %%edx, 60(%3)\n" " addl $-64, %0\n" @@ -487,9 +600,9 @@ static unsigned long __copy_user_zeroing " shrl $2, %0\n" " andl $3, %%eax\n" " cld\n" - "6: rep; movsl\n" + "6: rep; "__copyuser_seg" movsl\n" " movl %%eax,%0\n" - "7: rep; movsb\n" + "7: rep; "__copyuser_seg" movsb\n" "8:\n" ".section .fixup,\"ax\"\n" "9: lea 0(%%eax,%0,4),%0\n" @@ -537,41 +650,41 @@ static unsigned long __copy_user_intel_n __asm__ __volatile__( " .align 2,0x90\n" - "0: movl 32(%4), %%eax\n" + "0: "__copyuser_seg" movl 32(%4), %%eax\n" " cmpl $67, %0\n" " jbe 2f\n" - "1: movl 64(%4), %%eax\n" + "1: "__copyuser_seg" movl 64(%4), %%eax\n" " .align 2,0x90\n" - "2: movl 0(%4), %%eax\n" - "21: movl 4(%4), %%edx\n" + "2: "__copyuser_seg" movl 0(%4), %%eax\n" + "21: "__copyuser_seg" movl 4(%4), %%edx\n" " movnti %%eax, 0(%3)\n" " movnti %%edx, 4(%3)\n" - "3: movl 8(%4), %%eax\n" - "31: movl 12(%4),%%edx\n" + "3: "__copyuser_seg" movl 8(%4), %%eax\n" + "31: "__copyuser_seg" movl 12(%4),%%edx\n" " movnti %%eax, 8(%3)\n" " movnti %%edx, 12(%3)\n" - "4: movl 16(%4), %%eax\n" - "41: movl 20(%4), %%edx\n" + "4: "__copyuser_seg" movl 16(%4), %%eax\n" + "41: "__copyuser_seg" movl 20(%4), %%edx\n" " movnti %%eax, 16(%3)\n" " movnti %%edx, 20(%3)\n" - "10: movl 24(%4), %%eax\n" - "51: movl 28(%4), %%edx\n" + "10: "__copyuser_seg" movl 24(%4), %%eax\n" + "51: "__copyuser_seg" movl 28(%4), %%edx\n" " movnti %%eax, 24(%3)\n" " movnti %%edx, 28(%3)\n" - "11: movl 32(%4), %%eax\n" - "61: movl 36(%4), %%edx\n" + "11: "__copyuser_seg" movl 32(%4), %%eax\n" + "61: "__copyuser_seg" movl 36(%4), %%edx\n" " movnti %%eax, 32(%3)\n" " movnti %%edx, 36(%3)\n" - "12: movl 40(%4), %%eax\n" - "71: movl 44(%4), %%edx\n" + "12: "__copyuser_seg" movl 40(%4), %%eax\n" + "71: "__copyuser_seg" movl 44(%4), %%edx\n" " movnti %%eax, 40(%3)\n" " movnti %%edx, 44(%3)\n" - "13: movl 48(%4), %%eax\n" - "81: movl 52(%4), %%edx\n" + "13: "__copyuser_seg" movl 48(%4), %%eax\n" + "81: "__copyuser_seg" movl 52(%4), %%edx\n" " movnti %%eax, 48(%3)\n" " movnti %%edx, 52(%3)\n" - "14: movl 56(%4), %%eax\n" - "91: movl 60(%4), %%edx\n" + "14: "__copyuser_seg" movl 56(%4), %%eax\n" + "91: "__copyuser_seg" movl 60(%4), %%edx\n" " movnti %%eax, 56(%3)\n" " movnti %%edx, 60(%3)\n" " addl $-64, %0\n" @@ -584,9 +697,9 @@ static unsigned long __copy_user_intel_n " shrl $2, %0\n" " andl $3, %%eax\n" " cld\n" - "6: rep; movsl\n" + "6: rep; "__copyuser_seg" movsl\n" " movl %%eax,%0\n" - "7: rep; movsb\n" + "7: rep; "__copyuser_seg" movsb\n" "8:\n" ".section .fixup,\"ax\"\n" "9: lea 0(%%eax,%0,4),%0\n" @@ -629,32 +742,36 @@ static unsigned long __copy_user_intel_n */ unsigned long __copy_user_zeroing_intel(void *to, const void __user *from, unsigned long size); -unsigned long __copy_user_intel(void __user *to, const void *from, +unsigned long __generic_copy_to_user_intel(void __user *to, const void *from, + unsigned long size); +unsigned long __generic_copy_from_user_intel(void *to, const void __user *from, unsigned long size); unsigned long __copy_user_zeroing_intel_nocache(void *to, const void __user *from, unsigned long size); #endif /* CONFIG_X86_INTEL_USERCOPY */ /* Generic arbitrary sized copy. */ -#define __copy_user(to, from, size) \ +#define __copy_user(to, from, size, prefix, set, restore) \ do { \ int __d0, __d1, __d2; \ __asm__ __volatile__( \ + set \ " cmp $7,%0\n" \ " jbe 1f\n" \ " movl %1,%0\n" \ " negl %0\n" \ " andl $7,%0\n" \ " subl %0,%3\n" \ - "4: rep; movsb\n" \ + "4: rep; "prefix"movsb\n" \ " movl %3,%0\n" \ " shrl $2,%0\n" \ " andl $3,%3\n" \ " .align 2,0x90\n" \ - "0: rep; movsl\n" \ + "0: rep; "prefix"movsl\n" \ " movl %3,%0\n" \ - "1: rep; movsb\n" \ + "1: rep; "prefix"movsb\n" \ "2:\n" \ + restore \ ".section .fixup,\"ax\"\n" \ "5: addl %3,%0\n" \ " jmp 2b\n" \ @@ -682,14 +799,14 @@ do { \ " negl %0\n" \ " andl $7,%0\n" \ " subl %0,%3\n" \ - "4: rep; movsb\n" \ + "4: rep; "__copyuser_seg"movsb\n" \ " movl %3,%0\n" \ " shrl $2,%0\n" \ " andl $3,%3\n" \ " .align 2,0x90\n" \ - "0: rep; movsl\n" \ + "0: rep; "__copyuser_seg"movsl\n" \ " movl %3,%0\n" \ - "1: rep; movsb\n" \ + "1: rep; "__copyuser_seg"movsb\n" \ "2:\n" \ ".section .fixup,\"ax\"\n" \ "5: addl %3,%0\n" \ @@ -775,9 +892,9 @@ survive: } #endif if (movsl_is_ok(to, from, n)) - __copy_user(to, from, n); + __copy_user(to, from, n, "", __COPYUSER_SET_ES, __COPYUSER_RESTORE_ES); else - n = __copy_user_intel(to, from, n); + n = __generic_copy_to_user_intel(to, from, n); return n; } EXPORT_SYMBOL(__copy_to_user_ll); @@ -797,10 +914,9 @@ unsigned long __copy_from_user_ll_nozero unsigned long n) { if (movsl_is_ok(to, from, n)) - __copy_user(to, from, n); + __copy_user(to, from, n, __copyuser_seg, "", ""); else - n = __copy_user_intel((void __user *)to, - (const void *)from, n); + n = __generic_copy_from_user_intel(to, from, n); return n; } EXPORT_SYMBOL(__copy_from_user_ll_nozero); @@ -827,59 +943,38 @@ unsigned long __copy_from_user_ll_nocach if (n > 64 && cpu_has_xmm2) n = __copy_user_intel_nocache(to, from, n); else - __copy_user(to, from, n); + __copy_user(to, from, n, __copyuser_seg, "", ""); #else - __copy_user(to, from, n); + __copy_user(to, from, n, __copyuser_seg, "", ""); #endif return n; } EXPORT_SYMBOL(__copy_from_user_ll_nocache_nozero); -/** - * copy_to_user: - Copy a block of data into user space. - * @to: Destination address, in user space. - * @from: Source address, in kernel space. - * @n: Number of bytes to copy. - * - * Context: User context only. This function may sleep. - * - * Copy data from kernel space to user space. - * - * Returns number of bytes that could not be copied. - * On success, this will be zero. - */ -unsigned long -copy_to_user(void __user *to, const void *from, unsigned long n) +#ifdef CONFIG_PAX_MEMORY_UDEREF +void __set_fs(mm_segment_t x) { - if (access_ok(VERIFY_WRITE, to, n)) - n = __copy_to_user(to, from, n); - return n; + switch (x.seg) { + case 0: + loadsegment(gs, 0); + break; + case TASK_SIZE_MAX: + loadsegment(gs, __USER_DS); + break; + case -1UL: + loadsegment(gs, __KERNEL_DS); + break; + default: + BUG(); + } + return; } -EXPORT_SYMBOL(copy_to_user); +EXPORT_SYMBOL(__set_fs); -/** - * copy_from_user: - Copy a block of data from user space. - * @to: Destination address, in kernel space. - * @from: Source address, in user space. - * @n: Number of bytes to copy. - * - * Context: User context only. This function may sleep. - * - * Copy data from user space to kernel space. - * - * Returns number of bytes that could not be copied. - * On success, this will be zero. - * - * If some data could not be copied, this function will pad the copied - * data to the requested size using zero bytes. - */ -unsigned long -copy_from_user(void *to, const void __user *from, unsigned long n) +void set_fs(mm_segment_t x) { - if (access_ok(VERIFY_READ, from, n)) - n = __copy_from_user(to, from, n); - else - memset(to, 0, n); - return n; + current_thread_info()->addr_limit = x; + __set_fs(x); } -EXPORT_SYMBOL(copy_from_user); +EXPORT_SYMBOL(set_fs); +#endif diff -urNp linux-2.6.32.48/arch/x86/lib/usercopy_64.c linux-2.6.32.48-openvz/arch/x86/lib/usercopy_64.c --- linux-2.6.32.48/arch/x86/lib/usercopy_64.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/lib/usercopy_64.c 2011-11-17 20:09:04.000000000 -0500 @@ -42,6 +42,12 @@ long __strncpy_from_user(char *dst, const char __user *src, long count) { long res; + +#ifdef CONFIG_PAX_MEMORY_UDEREF + if ((unsigned long)src < PAX_USER_SHADOW_BASE) + src += PAX_USER_SHADOW_BASE; +#endif + __do_strncpy_from_user(dst, src, count, res); return res; } @@ -65,6 +71,12 @@ unsigned long __clear_user(void __user * { long __d0; might_fault(); + +#ifdef CONFIG_PAX_MEMORY_UDEREF + if ((unsigned long)addr < PAX_USER_SHADOW_BASE) + addr += PAX_USER_SHADOW_BASE; +#endif + /* no memory constraint because it doesn't change any memory gcc knows about */ asm volatile( @@ -151,10 +163,18 @@ EXPORT_SYMBOL(strlen_user); unsigned long copy_in_user(void __user *to, const void __user *from, unsigned len) { - if (access_ok(VERIFY_WRITE, to, len) && access_ok(VERIFY_READ, from, len)) { - return copy_user_generic((__force void *)to, (__force void *)from, len); - } - return len; + if (access_ok(VERIFY_WRITE, to, len) && access_ok(VERIFY_READ, from, len)) { + +#ifdef CONFIG_PAX_MEMORY_UDEREF + if ((unsigned long)to < PAX_USER_SHADOW_BASE) + to += PAX_USER_SHADOW_BASE; + if ((unsigned long)from < PAX_USER_SHADOW_BASE) + from += PAX_USER_SHADOW_BASE; +#endif + + return copy_user_generic((void __force_kernel *)to, (void __force_kernel *)from, len); + } + return len; } EXPORT_SYMBOL(copy_in_user); @@ -164,7 +184,7 @@ EXPORT_SYMBOL(copy_in_user); * it is not necessary to optimize tail handling. */ unsigned long -copy_user_handle_tail(char *to, char *from, unsigned len, unsigned zerorest) +copy_user_handle_tail(char __user *to, char __user *from, unsigned len, unsigned zerorest) { char c; unsigned zero_len; diff -urNp linux-2.6.32.48/arch/x86/Makefile linux-2.6.32.48-openvz/arch/x86/Makefile --- linux-2.6.32.48/arch/x86/Makefile 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/Makefile 2011-11-17 20:09:04.000000000 -0500 @@ -44,6 +44,7 @@ ifeq ($(CONFIG_X86_32),y) else BITS := 64 UTS_MACHINE := x86_64 + biarch := $(call cc-option,-m64) CHECKFLAGS += -D__x86_64__ -m64 KBUILD_AFLAGS += -m64 @@ -189,3 +190,12 @@ define archhelp echo ' FDARGS="..." arguments for the booted kernel' echo ' FDINITRD=file initrd for the booted kernel' endef + +define OLD_LD + +*** ${VERSION}.${PATCHLEVEL} PaX kernels no longer build correctly with old versions of binutils. +*** Please upgrade your binutils to 2.18 or newer +endef + +archprepare: + $(if $(LDFLAGS_BUILD_ID),,$(error $(OLD_LD))) diff -urNp linux-2.6.32.48/arch/x86/mm/extable.c linux-2.6.32.48-openvz/arch/x86/mm/extable.c --- linux-2.6.32.48/arch/x86/mm/extable.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/mm/extable.c 2011-11-17 20:08:53.000000000 -0500 @@ -1,14 +1,71 @@ #include #include +#include #include +#include +/* + * The exception table needs to be sorted so that the binary + * search that we use to find entries in it works properly. + * This is used both for the kernel exception table and for + * the exception tables of modules that get loaded. + */ +static int cmp_ex(const void *a, const void *b) +{ + const struct exception_table_entry *x = a, *y = b; + + /* avoid overflow */ + if (x->insn > y->insn) + return 1; + if (x->insn < y->insn) + return -1; + return 0; +} + +static void swap_ex(void *a, void *b, int size) +{ + struct exception_table_entry t, *x = a, *y = b; + + t = *x; + + pax_open_kernel(); + *x = *y; + *y = t; + pax_close_kernel(); +} + +void sort_extable(struct exception_table_entry *start, + struct exception_table_entry *finish) +{ + sort(start, finish - start, sizeof(struct exception_table_entry), + cmp_ex, swap_ex); +} + +#ifdef CONFIG_MODULES +/* + * If the exception table is sorted, any referring to the module init + * will be at the beginning or the end. + */ +void trim_init_extable(struct module *m) +{ + /*trim the beginning*/ + while (m->num_exentries && within_module_init(m->extable[0].insn, m)) { + m->extable++; + m->num_exentries--; + } + /*trim the end*/ + while (m->num_exentries && + within_module_init(m->extable[m->num_exentries-1].insn, m)) + m->num_exentries--; +} +#endif /* CONFIG_MODULES */ int fixup_exception(struct pt_regs *regs) { const struct exception_table_entry *fixup; #ifdef CONFIG_PNPBIOS - if (unlikely(SEGMENT_IS_PNP_CODE(regs->cs))) { + if (unlikely(!v8086_mode(regs) && SEGMENT_IS_PNP_CODE(regs->cs))) { extern u32 pnp_bios_fault_eip, pnp_bios_fault_esp; extern u32 pnp_bios_is_utter_crap; pnp_bios_is_utter_crap = 1; diff -urNp linux-2.6.32.48/arch/x86/mm/fault.c linux-2.6.32.48-openvz/arch/x86/mm/fault.c --- linux-2.6.32.48/arch/x86/mm/fault.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/mm/fault.c 2011-11-17 20:09:04.000000000 -0500 @@ -11,10 +11,19 @@ #include /* __kprobes, ... */ #include /* kmmio_handler, ... */ #include /* perf_sw_event */ +#include +#include #include /* dotraplinkage, ... */ #include /* pgd_*(), ... */ #include /* kmemcheck_*(), ... */ +#include +#include + +#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) +#include +#include "../kernel/dumpstack.h" +#endif /* * Page fault error code bits: @@ -51,7 +60,7 @@ static inline int notify_page_fault(stru int ret = 0; /* kprobe_running() needs smp_processor_id() */ - if (kprobes_built_in() && !user_mode_vm(regs)) { + if (kprobes_built_in() && !user_mode(regs)) { preempt_disable(); if (kprobe_running() && kprobe_fault_handler(regs, 14)) ret = 1; @@ -112,7 +121,10 @@ check_prefetch_opcode(struct pt_regs *re return !instr_lo || (instr_lo>>1) == 1; case 0x00: /* Prefetch instruction is 0x0F0D or 0x0F18 */ - if (probe_kernel_address(instr, opcode)) + if (user_mode(regs)) { + if (__copy_from_user_inatomic(&opcode, (unsigned char __force_user *)(instr), 1)) + return 0; + } else if (probe_kernel_address(instr, opcode)) return 0; *prefetch = (instr_lo == 0xF) && @@ -146,7 +158,10 @@ is_prefetch(struct pt_regs *regs, unsign while (instr < max_instr) { unsigned char opcode; - if (probe_kernel_address(instr, opcode)) + if (user_mode(regs)) { + if (__copy_from_user_inatomic(&opcode, (unsigned char __force_user *)(instr), 1)) + break; + } else if (probe_kernel_address(instr, opcode)) break; instr++; @@ -172,6 +187,30 @@ force_sig_info_fault(int si_signo, int s force_sig_info(si_signo, &info, tsk); } +#ifdef CONFIG_PAX_EMUTRAMP +static int pax_handle_fetch_fault(struct pt_regs *regs); +#endif + +#ifdef CONFIG_PAX_PAGEEXEC +static inline pmd_t * pax_get_pmd(struct mm_struct *mm, unsigned long address) +{ + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd; + + pgd = pgd_offset(mm, address); + if (!pgd_present(*pgd)) + return NULL; + pud = pud_offset(pgd, address); + if (!pud_present(*pud)) + return NULL; + pmd = pmd_offset(pud, address); + if (!pmd_present(*pmd)) + return NULL; + return pmd; +} +#endif + DEFINE_SPINLOCK(pgd_lock); LIST_HEAD(pgd_list); @@ -224,11 +263,24 @@ void vmalloc_sync_all(void) address += PMD_SIZE) { unsigned long flags; + +#ifdef CONFIG_PAX_PER_CPU_PGD + unsigned long cpu; +#else struct page *page; +#endif spin_lock_irqsave(&pgd_lock, flags); + +#ifdef CONFIG_PAX_PER_CPU_PGD + for (cpu = 0; cpu < NR_CPUS; ++cpu) { + pgd_t *pgd = get_cpu_pgd(cpu); +#else list_for_each_entry(page, &pgd_list, lru) { - if (!vmalloc_sync_one(page_address(page), address)) + pgd_t *pgd = page_address(page); +#endif + + if (!vmalloc_sync_one(pgd, address)) break; } spin_unlock_irqrestore(&pgd_lock, flags); @@ -258,6 +310,11 @@ static noinline int vmalloc_fault(unsign * an interrupt in the middle of a task switch.. */ pgd_paddr = read_cr3(); + +#ifdef CONFIG_PAX_PER_CPU_PGD + BUG_ON(__pa(get_cpu_pgd(smp_processor_id())) != (pgd_paddr & PHYSICAL_PAGE_MASK)); +#endif + pmd_k = vmalloc_sync_one(__va(pgd_paddr), address); if (!pmd_k) return -1; @@ -332,15 +389,27 @@ void vmalloc_sync_all(void) const pgd_t *pgd_ref = pgd_offset_k(address); unsigned long flags; + +#ifdef CONFIG_PAX_PER_CPU_PGD + unsigned long cpu; +#else struct page *page; +#endif if (pgd_none(*pgd_ref)) continue; spin_lock_irqsave(&pgd_lock, flags); + +#ifdef CONFIG_PAX_PER_CPU_PGD + for (cpu = 0; cpu < NR_CPUS; ++cpu) { + pgd_t *pgd = pgd_offset_cpu(cpu, address); +#else list_for_each_entry(page, &pgd_list, lru) { pgd_t *pgd; pgd = (pgd_t *)page_address(page) + pgd_index(address); +#endif + if (pgd_none(*pgd)) set_pgd(pgd, *pgd_ref); else @@ -373,7 +442,14 @@ static noinline int vmalloc_fault(unsign * happen within a race in page table update. In the later * case just flush: */ + +#ifdef CONFIG_PAX_PER_CPU_PGD + BUG_ON(__pa(get_cpu_pgd(smp_processor_id())) != (read_cr3() & PHYSICAL_PAGE_MASK)); + pgd = pgd_offset_cpu(smp_processor_id(), address); +#else pgd = pgd_offset(current->active_mm, address); +#endif + pgd_ref = pgd_offset_k(address); if (pgd_none(*pgd_ref)) return -1; @@ -535,7 +611,7 @@ static int is_errata93(struct pt_regs *r static int is_errata100(struct pt_regs *regs, unsigned long address) { #ifdef CONFIG_X86_64 - if ((regs->cs == __USER32_CS || (regs->cs & (1<<2))) && (address >> 32)) + if ((regs->cs == __USER32_CS || (regs->cs & SEGMENT_LDT)) && (address >> 32)) return 1; #endif return 0; @@ -562,7 +638,7 @@ static int is_f00f_bug(struct pt_regs *r } static const char nx_warning[] = KERN_CRIT -"kernel tried to execute NX-protected page - exploit attempt? (uid: %d)\n"; +"kernel tried to execute NX-protected page - exploit attempt? (uid: %d, task: %s, pid: %d)\n"; static void show_fault_oops(struct pt_regs *regs, unsigned long error_code, @@ -571,15 +647,26 @@ show_fault_oops(struct pt_regs *regs, un if (!oops_may_print()) return; - if (error_code & PF_INSTR) { + if (nx_enabled && (error_code & PF_INSTR)) { unsigned int level; pte_t *pte = lookup_address(address, &level); if (pte && pte_present(*pte) && !pte_exec(*pte)) - printk(nx_warning, current_uid()); + printk(nx_warning, current_uid(), current->comm, task_pid_nr(current)); } +#ifdef CONFIG_PAX_KERNEXEC + if (init_mm.start_code <= address && address < init_mm.end_code) { + if (current->signal->curr_ip) + printk(KERN_ERR "PAX: From %pI4: %s:%d, uid/euid: %u/%u, attempted to modify kernel code\n", + ¤t->signal->curr_ip, current->comm, task_pid_nr(current), current_uid(), current_euid()); + else + printk(KERN_ERR "PAX: %s:%d, uid/euid: %u/%u, attempted to modify kernel code\n", + current->comm, task_pid_nr(current), current_uid(), current_euid()); + } +#endif + printk(KERN_ALERT "BUG: unable to handle kernel "); if (address < PAGE_SIZE) printk(KERN_CONT "NULL pointer dereference"); @@ -689,7 +776,7 @@ show_signal_msg(struct pt_regs *regs, un if (!printk_ratelimit()) return; - printk("%s%s[%d]: segfault at %lx ip %p sp %p error %lx", + ve_printk(VE_LOG, "%s%s[%d]: segfault at %lx ip %p sp %p error %lx", task_pid_nr(tsk) > 1 ? KERN_INFO : KERN_EMERG, tsk->comm, task_pid_nr(tsk), address, (void *)regs->ip, (void *)regs->sp, error_code); @@ -704,6 +791,70 @@ __bad_area_nosemaphore(struct pt_regs *r unsigned long address, int si_code) { struct task_struct *tsk = current; +#if defined(CONFIG_X86_64) || defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC) + struct mm_struct *mm = tsk->mm; +#endif + +#ifdef CONFIG_X86_64 + if (mm && (error_code & PF_INSTR) && mm->context.vdso) { + if (regs->ip == (unsigned long)vgettimeofday) { + regs->ip = (unsigned long)VDSO64_SYMBOL(mm->context.vdso, fallback_gettimeofday); + return; + } else if (regs->ip == (unsigned long)vtime) { + regs->ip = (unsigned long)VDSO64_SYMBOL(mm->context.vdso, fallback_time); + return; + } else if (regs->ip == (unsigned long)vgetcpu) { + regs->ip = (unsigned long)VDSO64_SYMBOL(mm->context.vdso, getcpu); + return; + } + } +#endif + +#if defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC) + if (mm && (error_code & PF_USER)) { + unsigned long ip = regs->ip; + + if (v8086_mode(regs)) + ip = ((regs->cs & 0xffff) << 4) + (ip & 0xffff); + + /* + * It's possible to have interrupts off here: + */ + local_irq_enable(); + +#ifdef CONFIG_PAX_PAGEEXEC + if ((mm->pax_flags & MF_PAX_PAGEEXEC) && + ((nx_enabled && (error_code & PF_INSTR)) || (!(error_code & (PF_PROT | PF_WRITE)) && ip == address))) { + +#ifdef CONFIG_PAX_EMUTRAMP + switch (pax_handle_fetch_fault(regs)) { + case 2: + return; + } +#endif + + pax_report_fault(regs, (void *)ip, (void *)regs->sp); + do_group_exit(SIGKILL); + } +#endif + +#ifdef CONFIG_PAX_SEGMEXEC + if ((mm->pax_flags & MF_PAX_SEGMEXEC) && !(error_code & (PF_PROT | PF_WRITE)) && (ip + SEGMEXEC_TASK_SIZE == address)) { + +#ifdef CONFIG_PAX_EMUTRAMP + switch (pax_handle_fetch_fault(regs)) { + case 2: + return; + } +#endif + + pax_report_fault(regs, (void *)ip, (void *)regs->sp); + do_group_exit(SIGKILL); + } +#endif + + } +#endif /* User mode accesses just cause a SIGSEGV */ if (error_code & PF_USER) { @@ -857,6 +1008,99 @@ static int spurious_fault_check(unsigned return 1; } +#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_PAGEEXEC) +static int pax_handle_pageexec_fault(struct pt_regs *regs, struct mm_struct *mm, unsigned long address, unsigned long error_code) +{ + pte_t *pte; + pmd_t *pmd; + spinlock_t *ptl; + unsigned char pte_mask; + + if (nx_enabled || (error_code & (PF_PROT|PF_USER)) != (PF_PROT|PF_USER) || v8086_mode(regs) || + !(mm->pax_flags & MF_PAX_PAGEEXEC)) + return 0; + + /* PaX: it's our fault, let's handle it if we can */ + + /* PaX: take a look at read faults before acquiring any locks */ + if (unlikely(!(error_code & PF_WRITE) && (regs->ip == address))) { + /* instruction fetch attempt from a protected page in user mode */ + up_read(&mm->mmap_sem); + +#ifdef CONFIG_PAX_EMUTRAMP + switch (pax_handle_fetch_fault(regs)) { + case 2: + return 1; + } +#endif + + pax_report_fault(regs, (void *)regs->ip, (void *)regs->sp); + do_group_exit(SIGKILL); + } + + pmd = pax_get_pmd(mm, address); + if (unlikely(!pmd)) + return 0; + + pte = pte_offset_map_lock(mm, pmd, address, &ptl); + if (unlikely(!(pte_val(*pte) & _PAGE_PRESENT) || pte_user(*pte))) { + pte_unmap_unlock(pte, ptl); + return 0; + } + + if (unlikely((error_code & PF_WRITE) && !pte_write(*pte))) { + /* write attempt to a protected page in user mode */ + pte_unmap_unlock(pte, ptl); + return 0; + } + +#ifdef CONFIG_SMP + if (likely(address > get_limit(regs->cs) && cpu_isset(smp_processor_id(), mm->context.cpu_user_cs_mask))) +#else + if (likely(address > get_limit(regs->cs))) +#endif + { + set_pte(pte, pte_mkread(*pte)); + __flush_tlb_one(address); + pte_unmap_unlock(pte, ptl); + up_read(&mm->mmap_sem); + return 1; + } + + pte_mask = _PAGE_ACCESSED | _PAGE_USER | ((error_code & PF_WRITE) << (_PAGE_BIT_DIRTY-1)); + + /* + * PaX: fill DTLB with user rights and retry + */ + __asm__ __volatile__ ( + "orb %2,(%1)\n" +#if defined(CONFIG_M586) || defined(CONFIG_M586TSC) +/* + * PaX: let this uncommented 'invlpg' remind us on the behaviour of Intel's + * (and AMD's) TLBs. namely, they do not cache PTEs that would raise *any* + * page fault when examined during a TLB load attempt. this is true not only + * for PTEs holding a non-present entry but also present entries that will + * raise a page fault (such as those set up by PaX, or the copy-on-write + * mechanism). in effect it means that we do *not* need to flush the TLBs + * for our target pages since their PTEs are simply not in the TLBs at all. + + * the best thing in omitting it is that we gain around 15-20% speed in the + * fast path of the page fault handler and can get rid of tracing since we + * can no longer flush unintended entries. + */ + "invlpg (%0)\n" +#endif + __copyuser_seg"testb $0,(%0)\n" + "xorb %3,(%1)\n" + : + : "r" (address), "r" (pte), "q" (pte_mask), "i" (_PAGE_USER) + : "memory", "cc"); + pte_unmap_unlock(pte, ptl); + up_read(&mm->mmap_sem); + return 1; +} +#endif + /* * Handle a spurious fault caused by a stale TLB entry. * @@ -918,11 +1162,14 @@ spurious_fault(unsigned long error_code, return ret; } -int show_unhandled_signals = 1; +int show_unhandled_signals = 0; static inline int access_error(unsigned long error_code, int write, struct vm_area_struct *vma) { + if (nx_enabled && (error_code & PF_INSTR) && !(vma->vm_flags & VM_EXEC)) + return 1; + if (write) { /* write, present and write, not present: */ if (unlikely(!(vma->vm_flags & VM_WRITE))) @@ -956,17 +1203,31 @@ do_page_fault(struct pt_regs *regs, unsi { struct vm_area_struct *vma; struct task_struct *tsk; - unsigned long address; struct mm_struct *mm; int write; int fault; + /* Get the faulting address: */ + unsigned long address = read_cr2(); + +#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) + if (!user_mode(regs) && address < 2 * PAX_USER_SHADOW_BASE) { + if (!search_exception_tables(regs->ip)) { + bad_area_nosemaphore(regs, error_code, address); + return; + } + if (address < PAX_USER_SHADOW_BASE) { + printk(KERN_ERR "PAX: please report this to pageexec@freemail.hu\n"); + printk(KERN_ERR "PAX: faulting IP: %pA\n", (void *)regs->ip); + show_trace_log_lvl(NULL, NULL, (void *)regs->sp, regs->bp, KERN_ERR); + } else + address -= PAX_USER_SHADOW_BASE; + } +#endif + tsk = current; mm = tsk->mm; - /* Get the faulting address: */ - address = read_cr2(); - /* * Detect and handle instructions that would cause a page fault for * both a tracked kernel page and a userspace page. @@ -1026,7 +1287,7 @@ do_page_fault(struct pt_regs *regs, unsi * User-mode registers count as a user access even for any * potential system fault or CPU buglet: */ - if (user_mode_vm(regs)) { + if (user_mode(regs)) { local_irq_enable(); error_code |= PF_USER; } else { @@ -1080,6 +1341,11 @@ do_page_fault(struct pt_regs *regs, unsi might_sleep(); } +#if defined(CONFIG_X86_32) && defined(CONFIG_PAX_PAGEEXEC) + if (pax_handle_pageexec_fault(regs, mm, address, error_code)) + return; +#endif + vma = find_vma(mm, address); if (unlikely(!vma)) { bad_area(regs, error_code, address); @@ -1091,18 +1357,24 @@ do_page_fault(struct pt_regs *regs, unsi bad_area(regs, error_code, address); return; } - if (error_code & PF_USER) { - /* - * Accessing the stack below %sp is always a bug. - * The large cushion allows instructions like enter - * and pusha to work. ("enter $65535, $31" pushes - * 32 pointers and then decrements %sp by 65535.) - */ - if (unlikely(address + 65536 + 32 * sizeof(unsigned long) < regs->sp)) { - bad_area(regs, error_code, address); - return; - } + /* + * Accessing the stack below %sp is always a bug. + * The large cushion allows instructions like enter + * and pusha to work. ("enter $65535, $31" pushes + * 32 pointers and then decrements %sp by 65535.) + */ + if (unlikely(address + 65536 + 32 * sizeof(unsigned long) < task_pt_regs(tsk)->sp)) { + bad_area(regs, error_code, address); + return; } + +#ifdef CONFIG_PAX_SEGMEXEC + if (unlikely((mm->pax_flags & MF_PAX_SEGMEXEC) && vma->vm_end - SEGMEXEC_TASK_SIZE - 1 < address - SEGMEXEC_TASK_SIZE - 1)) { + bad_area(regs, error_code, address); + return; + } +#endif + if (unlikely(expand_stack(vma, address))) { bad_area(regs, error_code, address); return; @@ -1146,3 +1418,199 @@ good_area: up_read(&mm->mmap_sem); } + +#ifdef CONFIG_PAX_EMUTRAMP +static int pax_handle_fetch_fault_32(struct pt_regs *regs) +{ + int err; + + do { /* PaX: gcc trampoline emulation #1 */ + unsigned char mov1, mov2; + unsigned short jmp; + unsigned int addr1, addr2; + +#ifdef CONFIG_X86_64 + if ((regs->ip + 11) >> 32) + break; +#endif + + err = get_user(mov1, (unsigned char __user *)regs->ip); + err |= get_user(addr1, (unsigned int __user *)(regs->ip + 1)); + err |= get_user(mov2, (unsigned char __user *)(regs->ip + 5)); + err |= get_user(addr2, (unsigned int __user *)(regs->ip + 6)); + err |= get_user(jmp, (unsigned short __user *)(regs->ip + 10)); + + if (err) + break; + + if (mov1 == 0xB9 && mov2 == 0xB8 && jmp == 0xE0FF) { + regs->cx = addr1; + regs->ax = addr2; + regs->ip = addr2; + return 2; + } + } while (0); + + do { /* PaX: gcc trampoline emulation #2 */ + unsigned char mov, jmp; + unsigned int addr1, addr2; + +#ifdef CONFIG_X86_64 + if ((regs->ip + 9) >> 32) + break; +#endif + + err = get_user(mov, (unsigned char __user *)regs->ip); + err |= get_user(addr1, (unsigned int __user *)(regs->ip + 1)); + err |= get_user(jmp, (unsigned char __user *)(regs->ip + 5)); + err |= get_user(addr2, (unsigned int __user *)(regs->ip + 6)); + + if (err) + break; + + if (mov == 0xB9 && jmp == 0xE9) { + regs->cx = addr1; + regs->ip = (unsigned int)(regs->ip + addr2 + 10); + return 2; + } + } while (0); + + return 1; /* PaX in action */ +} + +#ifdef CONFIG_X86_64 +static int pax_handle_fetch_fault_64(struct pt_regs *regs) +{ + int err; + + do { /* PaX: gcc trampoline emulation #1 */ + unsigned short mov1, mov2, jmp1; + unsigned char jmp2; + unsigned int addr1; + unsigned long addr2; + + err = get_user(mov1, (unsigned short __user *)regs->ip); + err |= get_user(addr1, (unsigned int __user *)(regs->ip + 2)); + err |= get_user(mov2, (unsigned short __user *)(regs->ip + 6)); + err |= get_user(addr2, (unsigned long __user *)(regs->ip + 8)); + err |= get_user(jmp1, (unsigned short __user *)(regs->ip + 16)); + err |= get_user(jmp2, (unsigned char __user *)(regs->ip + 18)); + + if (err) + break; + + if (mov1 == 0xBB41 && mov2 == 0xBA49 && jmp1 == 0xFF49 && jmp2 == 0xE3) { + regs->r11 = addr1; + regs->r10 = addr2; + regs->ip = addr1; + return 2; + } + } while (0); + + do { /* PaX: gcc trampoline emulation #2 */ + unsigned short mov1, mov2, jmp1; + unsigned char jmp2; + unsigned long addr1, addr2; + + err = get_user(mov1, (unsigned short __user *)regs->ip); + err |= get_user(addr1, (unsigned long __user *)(regs->ip + 2)); + err |= get_user(mov2, (unsigned short __user *)(regs->ip + 10)); + err |= get_user(addr2, (unsigned long __user *)(regs->ip + 12)); + err |= get_user(jmp1, (unsigned short __user *)(regs->ip + 20)); + err |= get_user(jmp2, (unsigned char __user *)(regs->ip + 22)); + + if (err) + break; + + if (mov1 == 0xBB49 && mov2 == 0xBA49 && jmp1 == 0xFF49 && jmp2 == 0xE3) { + regs->r11 = addr1; + regs->r10 = addr2; + regs->ip = addr1; + return 2; + } + } while (0); + + return 1; /* PaX in action */ +} +#endif + +/* + * PaX: decide what to do with offenders (regs->ip = fault address) + * + * returns 1 when task should be killed + * 2 when gcc trampoline was detected + */ +static int pax_handle_fetch_fault(struct pt_regs *regs) +{ + if (v8086_mode(regs)) + return 1; + + if (!(current->mm->pax_flags & MF_PAX_EMUTRAMP)) + return 1; + +#ifdef CONFIG_X86_32 + return pax_handle_fetch_fault_32(regs); +#else + if (regs->cs == __USER32_CS || (regs->cs & SEGMENT_LDT)) + return pax_handle_fetch_fault_32(regs); + else + return pax_handle_fetch_fault_64(regs); +#endif +} +#endif + +#if defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC) +void pax_report_insns(void *pc, void *sp) +{ + long i; + + printk(KERN_ERR "PAX: bytes at PC: "); + for (i = 0; i < 20; i++) { + unsigned char c; + if (get_user(c, (unsigned char __force_user *)pc+i)) + printk(KERN_CONT "?? "); + else + printk(KERN_CONT "%02x ", c); + } + printk("\n"); + + printk(KERN_ERR "PAX: bytes at SP-%lu: ", (unsigned long)sizeof(long)); + for (i = -1; i < 80 / (long)sizeof(long); i++) { + unsigned long c; + if (get_user(c, (unsigned long __force_user *)sp+i)) +#ifdef CONFIG_X86_32 + printk(KERN_CONT "???????? "); +#else + printk(KERN_CONT "???????????????? "); +#endif + else + printk(KERN_CONT "%0*lx ", 2 * (int)sizeof(long), c); + } + printk("\n"); +} +#endif + +/** + * probe_kernel_write(): safely attempt to write to a location + * @dst: address to write to + * @src: pointer to the data that shall be written + * @size: size of the data chunk + * + * Safely write to address @dst from the buffer at @src. If a kernel fault + * happens, handle that and return -EFAULT. + */ +long notrace probe_kernel_write(void *dst, const void *src, size_t size) +{ + long ret; + mm_segment_t old_fs = get_fs(); + + set_fs(KERNEL_DS); + pagefault_disable(); + pax_open_kernel(); + ret = __copy_to_user_inatomic((void __force_user *)dst, src, size); + pax_close_kernel(); + pagefault_enable(); + set_fs(old_fs); + + return ret ? -EFAULT : 0; +} diff -urNp linux-2.6.32.48/arch/x86/mm/gup.c linux-2.6.32.48-openvz/arch/x86/mm/gup.c --- linux-2.6.32.48/arch/x86/mm/gup.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/mm/gup.c 2011-11-17 20:08:53.000000000 -0500 @@ -237,7 +237,7 @@ int __get_user_pages_fast(unsigned long addr = start; len = (unsigned long) nr_pages << PAGE_SHIFT; end = start + len; - if (unlikely(!access_ok(write ? VERIFY_WRITE : VERIFY_READ, + if (unlikely(!__access_ok(write ? VERIFY_WRITE : VERIFY_READ, (void __user *)start, len))) return 0; diff -urNp linux-2.6.32.48/arch/x86/mm/highmem_32.c linux-2.6.32.48-openvz/arch/x86/mm/highmem_32.c --- linux-2.6.32.48/arch/x86/mm/highmem_32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/mm/highmem_32.c 2011-11-17 20:08:53.000000000 -0500 @@ -43,7 +43,10 @@ void *kmap_atomic_prot(struct page *page idx = type + KM_TYPE_NR*smp_processor_id(); vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx); BUG_ON(!pte_none(*(kmap_pte-idx))); + + pax_open_kernel(); set_pte(kmap_pte-idx, mk_pte(page, prot)); + pax_close_kernel(); return (void *)vaddr; } diff -urNp linux-2.6.32.48/arch/x86/mm/hugetlbpage.c linux-2.6.32.48-openvz/arch/x86/mm/hugetlbpage.c --- linux-2.6.32.48/arch/x86/mm/hugetlbpage.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/mm/hugetlbpage.c 2011-11-17 20:08:53.000000000 -0500 @@ -12,6 +12,7 @@ #include #include #include +#include #include #include #include @@ -230,6 +231,7 @@ int pud_huge(pud_t pud) { return !!(pud_val(pud) & _PAGE_PSE); } +EXPORT_SYMBOL(pmd_huge); struct page * follow_huge_pmd(struct mm_struct *mm, unsigned long address, @@ -267,13 +269,20 @@ static unsigned long hugetlb_get_unmappe struct hstate *h = hstate_file(file); struct mm_struct *mm = current->mm; struct vm_area_struct *vma; - unsigned long start_addr; + unsigned long start_addr, pax_task_size = TASK_SIZE; + +#ifdef CONFIG_PAX_SEGMEXEC + if (mm->pax_flags & MF_PAX_SEGMEXEC) + pax_task_size = SEGMEXEC_TASK_SIZE; +#endif + + pax_task_size -= PAGE_SIZE; if (len > mm->cached_hole_size) { - start_addr = mm->free_area_cache; + start_addr = mm->free_area_cache; } else { - start_addr = TASK_UNMAPPED_BASE; - mm->cached_hole_size = 0; + start_addr = mm->mmap_base; + mm->cached_hole_size = 0; } full_search: @@ -281,26 +290,27 @@ full_search: for (vma = find_vma(mm, addr); ; vma = vma->vm_next) { /* At this point: (!vma || addr < vma->vm_end). */ - if (TASK_SIZE - len < addr) { + if (pax_task_size - len < addr) { /* * Start a new search - just in case we missed * some holes. */ - if (start_addr != TASK_UNMAPPED_BASE) { - start_addr = TASK_UNMAPPED_BASE; + if (start_addr != mm->mmap_base) { + start_addr = mm->mmap_base; mm->cached_hole_size = 0; goto full_search; } return -ENOMEM; } - if (!vma || addr + len <= vma->vm_start) { - mm->free_area_cache = addr + len; - return addr; - } + if (check_heap_stack_gap(vma, addr, len)) + break; if (addr + mm->cached_hole_size < vma->vm_start) mm->cached_hole_size = vma->vm_start - addr; addr = ALIGN(vma->vm_end, huge_page_size(h)); } + + mm->free_area_cache = addr + len; + return addr; } static unsigned long hugetlb_get_unmapped_area_topdown(struct file *file, @@ -309,10 +319,9 @@ static unsigned long hugetlb_get_unmappe { struct hstate *h = hstate_file(file); struct mm_struct *mm = current->mm; - struct vm_area_struct *vma, *prev_vma; - unsigned long base = mm->mmap_base, addr = addr0; + struct vm_area_struct *vma; + unsigned long base = mm->mmap_base, addr; unsigned long largest_hole = mm->cached_hole_size; - int first_time = 1; /* don't allow allocations above current base */ if (mm->free_area_cache > base) @@ -322,64 +331,63 @@ static unsigned long hugetlb_get_unmappe largest_hole = 0; mm->free_area_cache = base; } -try_again: + /* make sure it can fit in the remaining address space */ if (mm->free_area_cache < len) goto fail; /* either no address requested or cant fit in requested address hole */ - addr = (mm->free_area_cache - len) & huge_page_mask(h); + addr = (mm->free_area_cache - len); do { + addr &= huge_page_mask(h); + vma = find_vma(mm, addr); /* * Lookup failure means no vma is above this address, * i.e. return with success: - */ - if (!(vma = find_vma_prev(mm, addr, &prev_vma))) - return addr; - - /* * new region fits between prev_vma->vm_end and * vma->vm_start, use it: */ - if (addr + len <= vma->vm_start && - (!prev_vma || (addr >= prev_vma->vm_end))) { + if (check_heap_stack_gap(vma, addr, len)) { /* remember the address as a hint for next time */ - mm->cached_hole_size = largest_hole; - return (mm->free_area_cache = addr); - } else { - /* pull free_area_cache down to the first hole */ - if (mm->free_area_cache == vma->vm_end) { - mm->free_area_cache = vma->vm_start; - mm->cached_hole_size = largest_hole; - } + mm->cached_hole_size = largest_hole; + return (mm->free_area_cache = addr); + } + /* pull free_area_cache down to the first hole */ + if (mm->free_area_cache == vma->vm_end) { + mm->free_area_cache = vma->vm_start; + mm->cached_hole_size = largest_hole; } /* remember the largest hole we saw so far */ if (addr + largest_hole < vma->vm_start) - largest_hole = vma->vm_start - addr; + largest_hole = vma->vm_start - addr; /* try just below the current vma->vm_start */ - addr = (vma->vm_start - len) & huge_page_mask(h); - } while (len <= vma->vm_start); + addr = skip_heap_stack_gap(vma, len); + } while (!IS_ERR_VALUE(addr)); fail: /* - * if hint left us with no space for the requested - * mapping then try again: - */ - if (first_time) { - mm->free_area_cache = base; - largest_hole = 0; - first_time = 0; - goto try_again; - } - /* * A failed mmap() very likely causes application failure, * so fall back to the bottom-up function here. This scenario * can happen with large stack limits and large mmap() * allocations. */ - mm->free_area_cache = TASK_UNMAPPED_BASE; + +#ifdef CONFIG_PAX_SEGMEXEC + if (mm->pax_flags & MF_PAX_SEGMEXEC) + mm->mmap_base = SEGMEXEC_TASK_UNMAPPED_BASE; + else +#endif + + mm->mmap_base = TASK_UNMAPPED_BASE; + +#ifdef CONFIG_PAX_RANDMMAP + if (mm->pax_flags & MF_PAX_RANDMMAP) + mm->mmap_base += mm->delta_mmap; +#endif + + mm->free_area_cache = mm->mmap_base; mm->cached_hole_size = ~0UL; addr = hugetlb_get_unmapped_area_bottomup(file, addr0, len, pgoff, flags); @@ -387,6 +395,7 @@ fail: /* * Restore the topdown base: */ + mm->mmap_base = base; mm->free_area_cache = base; mm->cached_hole_size = ~0UL; @@ -400,10 +409,19 @@ hugetlb_get_unmapped_area(struct file *f struct hstate *h = hstate_file(file); struct mm_struct *mm = current->mm; struct vm_area_struct *vma; + unsigned long pax_task_size = TASK_SIZE; if (len & ~huge_page_mask(h)) return -EINVAL; - if (len > TASK_SIZE) + +#ifdef CONFIG_PAX_SEGMEXEC + if (mm->pax_flags & MF_PAX_SEGMEXEC) + pax_task_size = SEGMEXEC_TASK_SIZE; +#endif + + pax_task_size -= PAGE_SIZE; + + if (len > pax_task_size) return -ENOMEM; if (flags & MAP_FIXED) { @@ -415,8 +433,7 @@ hugetlb_get_unmapped_area(struct file *f if (addr) { addr = ALIGN(addr, huge_page_size(h)); vma = find_vma(mm, addr); - if (TASK_SIZE - len >= addr && - (!vma || addr + len <= vma->vm_start)) + if (pax_task_size - len >= addr && check_heap_stack_gap(vma, addr, len)) return addr; } if (mm->get_unmapped_area == arch_get_unmapped_area) diff -urNp linux-2.6.32.48/arch/x86/mm/init_32.c linux-2.6.32.48-openvz/arch/x86/mm/init_32.c --- linux-2.6.32.48/arch/x86/mm/init_32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/mm/init_32.c 2011-11-17 20:08:53.000000000 -0500 @@ -72,36 +72,6 @@ static __init void *alloc_low_page(void) } /* - * Creates a middle page table and puts a pointer to it in the - * given global directory entry. This only returns the gd entry - * in non-PAE compilation mode, since the middle layer is folded. - */ -static pmd_t * __init one_md_table_init(pgd_t *pgd) -{ - pud_t *pud; - pmd_t *pmd_table; - -#ifdef CONFIG_X86_PAE - if (!(pgd_val(*pgd) & _PAGE_PRESENT)) { - if (after_bootmem) - pmd_table = (pmd_t *)alloc_bootmem_pages(PAGE_SIZE); - else - pmd_table = (pmd_t *)alloc_low_page(); - paravirt_alloc_pmd(&init_mm, __pa(pmd_table) >> PAGE_SHIFT); - set_pgd(pgd, __pgd(__pa(pmd_table) | _PAGE_PRESENT)); - pud = pud_offset(pgd, 0); - BUG_ON(pmd_table != pmd_offset(pud, 0)); - - return pmd_table; - } -#endif - pud = pud_offset(pgd, 0); - pmd_table = pmd_offset(pud, 0); - - return pmd_table; -} - -/* * Create a page table and place a pointer to it in a middle page * directory entry: */ @@ -121,13 +91,28 @@ static pte_t * __init one_page_table_ini page_table = (pte_t *)alloc_low_page(); paravirt_alloc_pte(&init_mm, __pa(page_table) >> PAGE_SHIFT); +#if defined(CONFIG_PAX_PAGEEXEC) || defined(CONFIG_PAX_SEGMEXEC) + set_pmd(pmd, __pmd(__pa(page_table) | _KERNPG_TABLE)); +#else set_pmd(pmd, __pmd(__pa(page_table) | _PAGE_TABLE)); +#endif BUG_ON(page_table != pte_offset_kernel(pmd, 0)); } return pte_offset_kernel(pmd, 0); } +static pmd_t * __init one_md_table_init(pgd_t *pgd) +{ + pud_t *pud; + pmd_t *pmd_table; + + pud = pud_offset(pgd, 0); + pmd_table = pmd_offset(pud, 0); + + return pmd_table; +} + pmd_t * __init populate_extra_pmd(unsigned long vaddr) { int pgd_idx = pgd_index(vaddr); @@ -201,6 +186,7 @@ page_table_range_init(unsigned long star int pgd_idx, pmd_idx; unsigned long vaddr; pgd_t *pgd; + pud_t *pud; pmd_t *pmd; pte_t *pte = NULL; @@ -210,8 +196,13 @@ page_table_range_init(unsigned long star pgd = pgd_base + pgd_idx; for ( ; (pgd_idx < PTRS_PER_PGD) && (vaddr != end); pgd++, pgd_idx++) { - pmd = one_md_table_init(pgd); - pmd = pmd + pmd_index(vaddr); + pud = pud_offset(pgd, vaddr); + pmd = pmd_offset(pud, vaddr); + +#ifdef CONFIG_X86_PAE + paravirt_alloc_pmd(&init_mm, __pa(pmd) >> PAGE_SHIFT); +#endif + for (; (pmd_idx < PTRS_PER_PMD) && (vaddr != end); pmd++, pmd_idx++) { pte = page_table_kmap_check(one_page_table_init(pmd), @@ -223,11 +214,20 @@ page_table_range_init(unsigned long star } } -static inline int is_kernel_text(unsigned long addr) +static inline int is_kernel_text(unsigned long start, unsigned long end) { - if (addr >= PAGE_OFFSET && addr <= (unsigned long)__init_end) - return 1; - return 0; + if ((start > ktla_ktva((unsigned long)_etext) || + end <= ktla_ktva((unsigned long)_stext)) && + (start > ktla_ktva((unsigned long)_einittext) || + end <= ktla_ktva((unsigned long)_sinittext)) && + +#ifdef CONFIG_ACPI_SLEEP + (start > (unsigned long)__va(acpi_wakeup_address) + 0x4000 || end <= (unsigned long)__va(acpi_wakeup_address)) && +#endif + + (start > (unsigned long)__va(0xfffff) || end <= (unsigned long)__va(0xc0000))) + return 0; + return 1; } /* @@ -243,9 +243,10 @@ kernel_physical_mapping_init(unsigned lo int use_pse = page_size_mask == (1<> PAGE_SHIFT); +#endif if (pfn >= end_pfn) continue; @@ -291,14 +297,13 @@ repeat: #endif for (; pmd_idx < PTRS_PER_PMD && pfn < end_pfn; pmd++, pmd_idx++) { - unsigned int addr = pfn * PAGE_SIZE + PAGE_OFFSET; + unsigned long address = pfn * PAGE_SIZE + PAGE_OFFSET; /* * Map with big pages if possible, otherwise * create normal page tables: */ if (use_pse) { - unsigned int addr2; pgprot_t prot = PAGE_KERNEL_LARGE; /* * first pass will use the same initial @@ -308,11 +313,7 @@ repeat: __pgprot(PTE_IDENT_ATTR | _PAGE_PSE); - addr2 = (pfn + PTRS_PER_PTE-1) * PAGE_SIZE + - PAGE_OFFSET + PAGE_SIZE-1; - - if (is_kernel_text(addr) || - is_kernel_text(addr2)) + if (is_kernel_text(address, address + PMD_SIZE)) prot = PAGE_KERNEL_LARGE_EXEC; pages_2m++; @@ -329,7 +330,7 @@ repeat: pte_ofs = pte_index((pfn<> 10, - (unsigned long)&_etext, (unsigned long)&_edata, - ((unsigned long)&_edata - (unsigned long)&_etext) >> 10, + (unsigned long)&_sdata, (unsigned long)&_edata, + ((unsigned long)&_edata - (unsigned long)&_sdata) >> 10, - (unsigned long)&_text, (unsigned long)&_etext, + ktla_ktva((unsigned long)&_text), ktla_ktva((unsigned long)&_etext), ((unsigned long)&_etext - (unsigned long)&_text) >> 10); /* @@ -1007,6 +1012,7 @@ void set_kernel_text_rw(void) if (!kernel_set_to_readonly) return; + start = ktla_ktva(start); pr_debug("Set kernel text: %lx - %lx for read write\n", start, start+size); @@ -1021,6 +1027,7 @@ void set_kernel_text_ro(void) if (!kernel_set_to_readonly) return; + start = ktla_ktva(start); pr_debug("Set kernel text: %lx - %lx for read only\n", start, start+size); @@ -1032,6 +1039,7 @@ void mark_rodata_ro(void) unsigned long start = PFN_ALIGN(_text); unsigned long size = PFN_ALIGN(_etext) - start; + start = ktla_ktva(start); set_pages_ro(virt_to_page(start), size >> PAGE_SHIFT); printk(KERN_INFO "Write protecting the kernel text: %luk\n", size >> 10); diff -urNp linux-2.6.32.48/arch/x86/mm/init_64.c linux-2.6.32.48-openvz/arch/x86/mm/init_64.c --- linux-2.6.32.48/arch/x86/mm/init_64.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/mm/init_64.c 2011-11-17 20:08:53.000000000 -0500 @@ -164,7 +164,9 @@ void set_pte_vaddr_pud(pud_t *pud_page, pmd = fill_pmd(pud, vaddr); pte = fill_pte(pmd, vaddr); + pax_open_kernel(); set_pte(pte, new_pte); + pax_close_kernel(); /* * It's enough to flush this one mapping. @@ -223,14 +225,12 @@ static void __init __init_extra_mapping( pgd = pgd_offset_k((unsigned long)__va(phys)); if (pgd_none(*pgd)) { pud = (pud_t *) spp_getpage(); - set_pgd(pgd, __pgd(__pa(pud) | _KERNPG_TABLE | - _PAGE_USER)); + set_pgd(pgd, __pgd(__pa(pud) | _PAGE_TABLE)); } pud = pud_offset(pgd, (unsigned long)__va(phys)); if (pud_none(*pud)) { pmd = (pmd_t *) spp_getpage(); - set_pud(pud, __pud(__pa(pmd) | _KERNPG_TABLE | - _PAGE_USER)); + set_pud(pud, __pud(__pa(pmd) | _PAGE_TABLE)); } pmd = pmd_offset(pud, phys); BUG_ON(!pmd_none(*pmd)); @@ -675,6 +675,12 @@ void __init mem_init(void) pci_iommu_alloc(); +#ifdef CONFIG_PAX_PER_CPU_PGD + clone_pgd_range(get_cpu_pgd(0) + KERNEL_PGD_BOUNDARY, + swapper_pg_dir + KERNEL_PGD_BOUNDARY, + KERNEL_PGD_PTRS); +#endif + /* clear_bss() already clear the empty_zero_page */ reservedpages = 0; @@ -861,8 +867,8 @@ int kern_addr_valid(unsigned long addr) static struct vm_area_struct gate_vma = { .vm_start = VSYSCALL_START, .vm_end = VSYSCALL_START + (VSYSCALL_MAPPED_PAGES * PAGE_SIZE), - .vm_page_prot = PAGE_READONLY_EXEC, - .vm_flags = VM_READ | VM_EXEC + .vm_page_prot = PAGE_READONLY, + .vm_flags = VM_READ }; struct vm_area_struct *get_gate_vma(struct task_struct *tsk) @@ -896,7 +902,7 @@ int in_gate_area_no_task(unsigned long a const char *arch_vma_name(struct vm_area_struct *vma) { - if (vma->vm_mm && vma->vm_start == (long)vma->vm_mm->context.vdso) + if (vma->vm_mm && vma->vm_start == vma->vm_mm->context.vdso) return "[vdso]"; if (vma == &gate_vma) return "[vsyscall]"; diff -urNp linux-2.6.32.48/arch/x86/mm/init.c linux-2.6.32.48-openvz/arch/x86/mm/init.c --- linux-2.6.32.48/arch/x86/mm/init.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/mm/init.c 2011-11-17 20:09:04.000000000 -0500 @@ -69,11 +69,7 @@ static void __init find_early_table_spac * cause a hotspot and fill up ZONE_DMA. The page tables * need roughly 0.5KB per GB. */ -#ifdef CONFIG_X86_32 - start = 0x7000; -#else - start = 0x8000; -#endif + start = 0x100000; e820_table_start = find_e820_area(start, max_pfn_mapped<> PAGE_SHIFT) == pagenr) + return 1; + /* allow ISA/video mem */ + if ((ISA_START_ADDRESS >> PAGE_SHIFT) <= pagenr && pagenr < (ISA_END_ADDRESS >> PAGE_SHIFT)) + return 1; + /* throw out everything else below 1MB */ + if (pagenr <= 256) + return 0; +#else if (pagenr <= 256) return 1; +#endif + if (iomem_is_exclusive(pagenr << PAGE_SHIFT)) return 0; if (!page_is_ram(pagenr)) @@ -379,6 +392,86 @@ void free_init_pages(char *what, unsigne void free_initmem(void) { + +#ifdef CONFIG_PAX_KERNEXEC +#ifdef CONFIG_X86_32 + /* PaX: limit KERNEL_CS to actual size */ + unsigned long addr, limit; + struct desc_struct d; + int cpu; + + limit = paravirt_enabled() ? ktva_ktla(0xffffffff) : (unsigned long)&_etext; + limit = (limit - 1UL) >> PAGE_SHIFT; + + memset(__LOAD_PHYSICAL_ADDR + PAGE_OFFSET, POISON_FREE_INITMEM, PAGE_SIZE); + for (cpu = 0; cpu < NR_CPUS; cpu++) { + pack_descriptor(&d, get_desc_base(&get_cpu_gdt_table(cpu)[GDT_ENTRY_KERNEL_CS]), limit, 0x9B, 0xC); + write_gdt_entry(get_cpu_gdt_table(cpu), GDT_ENTRY_KERNEL_CS, &d, DESCTYPE_S); + } + + /* PaX: make KERNEL_CS read-only */ + addr = PFN_ALIGN(ktla_ktva((unsigned long)&_text)); + if (!paravirt_enabled()) + set_memory_ro(addr, (PFN_ALIGN(_sdata) - addr) >> PAGE_SHIFT); +/* + for (addr = ktla_ktva((unsigned long)&_text); addr < (unsigned long)&_sdata; addr += PMD_SIZE) { + pgd = pgd_offset_k(addr); + pud = pud_offset(pgd, addr); + pmd = pmd_offset(pud, addr); + set_pmd(pmd, __pmd(pmd_val(*pmd) & ~_PAGE_RW)); + } +*/ +#ifdef CONFIG_X86_PAE + set_memory_nx(PFN_ALIGN(__init_begin), (PFN_ALIGN(__init_end) - PFN_ALIGN(__init_begin)) >> PAGE_SHIFT); +/* + for (addr = (unsigned long)&__init_begin; addr < (unsigned long)&__init_end; addr += PMD_SIZE) { + pgd = pgd_offset_k(addr); + pud = pud_offset(pgd, addr); + pmd = pmd_offset(pud, addr); + set_pmd(pmd, __pmd(pmd_val(*pmd) | (_PAGE_NX & __supported_pte_mask))); + } +*/ +#endif + +#ifdef CONFIG_MODULES + set_memory_4k((unsigned long)MODULES_EXEC_VADDR, (MODULES_EXEC_END - MODULES_EXEC_VADDR) >> PAGE_SHIFT); +#endif + +#else + pgd_t *pgd; + pud_t *pud; + pmd_t *pmd; + unsigned long addr, end; + + /* PaX: make kernel code/rodata read-only, rest non-executable */ + for (addr = __START_KERNEL_map; addr < __START_KERNEL_map + KERNEL_IMAGE_SIZE; addr += PMD_SIZE) { + pgd = pgd_offset_k(addr); + pud = pud_offset(pgd, addr); + pmd = pmd_offset(pud, addr); + if (!pmd_present(*pmd)) + continue; + if ((unsigned long)_text <= addr && addr < (unsigned long)_sdata) + set_pmd(pmd, __pmd(pmd_val(*pmd) & ~_PAGE_RW)); + else + set_pmd(pmd, __pmd(pmd_val(*pmd) | (_PAGE_NX & __supported_pte_mask))); + } + + addr = (unsigned long)__va(__pa(__START_KERNEL_map)); + end = addr + KERNEL_IMAGE_SIZE; + for (; addr < end; addr += PMD_SIZE) { + pgd = pgd_offset_k(addr); + pud = pud_offset(pgd, addr); + pmd = pmd_offset(pud, addr); + if (!pmd_present(*pmd)) + continue; + if ((unsigned long)__va(__pa(_text)) <= addr && addr < (unsigned long)__va(__pa(_sdata))) + set_pmd(pmd, __pmd(pmd_val(*pmd) & ~_PAGE_RW)); + } +#endif + + flush_tlb_all(); +#endif + free_init_pages("unused kernel memory", (unsigned long)(&__init_begin), (unsigned long)(&__init_end)); diff -urNp linux-2.6.32.48/arch/x86/mm/iomap_32.c linux-2.6.32.48-openvz/arch/x86/mm/iomap_32.c --- linux-2.6.32.48/arch/x86/mm/iomap_32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/mm/iomap_32.c 2011-11-17 20:08:53.000000000 -0500 @@ -65,7 +65,11 @@ void *kmap_atomic_prot_pfn(unsigned long debug_kmap_atomic(type); idx = type + KM_TYPE_NR * smp_processor_id(); vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx); + + pax_open_kernel(); set_pte(kmap_pte - idx, pfn_pte(pfn, prot)); + pax_close_kernel(); + arch_flush_lazy_mmu_mode(); return (void *)vaddr; diff -urNp linux-2.6.32.48/arch/x86/mm/ioremap.c linux-2.6.32.48-openvz/arch/x86/mm/ioremap.c --- linux-2.6.32.48/arch/x86/mm/ioremap.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/mm/ioremap.c 2011-11-17 20:08:53.000000000 -0500 @@ -41,8 +41,8 @@ int page_is_ram(unsigned long pagenr) * Second special case: Some BIOSen report the PC BIOS * area (640->1Mb) as ram even though it is not. */ - if (pagenr >= (BIOS_BEGIN >> PAGE_SHIFT) && - pagenr < (BIOS_END >> PAGE_SHIFT)) + if (pagenr >= (ISA_START_ADDRESS >> PAGE_SHIFT) && + pagenr < (ISA_END_ADDRESS >> PAGE_SHIFT)) return 0; for (i = 0; i < e820.nr_map; i++) { @@ -137,13 +137,10 @@ static void __iomem *__ioremap_caller(re /* * Don't allow anybody to remap normal RAM that we're using.. */ - for (pfn = phys_addr >> PAGE_SHIFT; - (pfn << PAGE_SHIFT) < (last_addr & PAGE_MASK); - pfn++) { - + for (pfn = phys_addr >> PAGE_SHIFT; ((resource_size_t)pfn << PAGE_SHIFT) < (last_addr & PAGE_MASK); pfn++) { int is_ram = page_is_ram(pfn); - if (is_ram && pfn_valid(pfn) && !PageReserved(pfn_to_page(pfn))) + if (is_ram && pfn_valid(pfn) && (pfn >= 0x100 || !PageReserved(pfn_to_page(pfn)))) return NULL; WARN_ON_ONCE(is_ram); } @@ -407,7 +404,7 @@ static int __init early_ioremap_debug_se early_param("early_ioremap_debug", early_ioremap_debug_setup); static __initdata int after_paging_init; -static pte_t bm_pte[PAGE_SIZE/sizeof(pte_t)] __page_aligned_bss; +static pte_t bm_pte[PAGE_SIZE/sizeof(pte_t)] __read_only __aligned(PAGE_SIZE); static inline pmd_t * __init early_ioremap_pmd(unsigned long addr) { @@ -439,8 +436,7 @@ void __init early_ioremap_init(void) slot_virt[i] = __fix_to_virt(FIX_BTMAP_BEGIN - NR_FIX_BTMAPS*i); pmd = early_ioremap_pmd(fix_to_virt(FIX_BTMAP_BEGIN)); - memset(bm_pte, 0, sizeof(bm_pte)); - pmd_populate_kernel(&init_mm, pmd, bm_pte); + pmd_populate_user(&init_mm, pmd, bm_pte); /* * The boot-ioremap range spans multiple pmds, for which diff -urNp linux-2.6.32.48/arch/x86/mm/kmemcheck/kmemcheck.c linux-2.6.32.48-openvz/arch/x86/mm/kmemcheck/kmemcheck.c --- linux-2.6.32.48/arch/x86/mm/kmemcheck/kmemcheck.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/mm/kmemcheck/kmemcheck.c 2011-11-17 20:08:53.000000000 -0500 @@ -622,9 +622,9 @@ bool kmemcheck_fault(struct pt_regs *reg * memory (e.g. tracked pages)? For now, we need this to avoid * invoking kmemcheck for PnP BIOS calls. */ - if (regs->flags & X86_VM_MASK) + if (v8086_mode(regs)) return false; - if (regs->cs != __KERNEL_CS) + if (regs->cs != __KERNEL_CS && regs->cs != __KERNEXEC_KERNEL_CS) return false; pte = kmemcheck_pte_lookup(address); diff -urNp linux-2.6.32.48/arch/x86/mm/mmap.c linux-2.6.32.48-openvz/arch/x86/mm/mmap.c --- linux-2.6.32.48/arch/x86/mm/mmap.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/mm/mmap.c 2011-11-17 20:08:53.000000000 -0500 @@ -49,7 +49,7 @@ static unsigned int stack_maxrandom_size * Leave an at least ~128 MB hole with possible stack randomization. */ #define MIN_GAP (128*1024*1024UL + stack_maxrandom_size()) -#define MAX_GAP (TASK_SIZE/6*5) +#define MAX_GAP (pax_task_size/6*5) /* * True on X86_32 or when emulating IA32 on X86_64 @@ -94,27 +94,40 @@ static unsigned long mmap_rnd(void) return rnd << PAGE_SHIFT; } -static unsigned long mmap_base(void) +static unsigned long mmap_base(struct mm_struct *mm) { unsigned long gap = current->signal->rlim[RLIMIT_STACK].rlim_cur; + unsigned long pax_task_size = TASK_SIZE; + +#ifdef CONFIG_PAX_SEGMEXEC + if (mm->pax_flags & MF_PAX_SEGMEXEC) + pax_task_size = SEGMEXEC_TASK_SIZE; +#endif if (gap < MIN_GAP) gap = MIN_GAP; else if (gap > MAX_GAP) gap = MAX_GAP; - return PAGE_ALIGN(TASK_SIZE - gap - mmap_rnd()); + return PAGE_ALIGN(pax_task_size - gap - mmap_rnd()); } /* * Bottom-up (legacy) layout on X86_32 did not support randomization, X86_64 * does, but not when emulating X86_32 */ -static unsigned long mmap_legacy_base(void) +static unsigned long mmap_legacy_base(struct mm_struct *mm) { - if (mmap_is_ia32()) + if (mmap_is_ia32()) { + +#ifdef CONFIG_PAX_SEGMEXEC + if (mm->pax_flags & MF_PAX_SEGMEXEC) + return SEGMEXEC_TASK_UNMAPPED_BASE; + else +#endif + return TASK_UNMAPPED_BASE; - else + } else return TASK_UNMAPPED_BASE + mmap_rnd(); } @@ -125,11 +138,23 @@ static unsigned long mmap_legacy_base(vo void arch_pick_mmap_layout(struct mm_struct *mm) { if (mmap_is_legacy()) { - mm->mmap_base = mmap_legacy_base(); + mm->mmap_base = mmap_legacy_base(mm); + +#ifdef CONFIG_PAX_RANDMMAP + if (mm->pax_flags & MF_PAX_RANDMMAP) + mm->mmap_base += mm->delta_mmap; +#endif + mm->get_unmapped_area = arch_get_unmapped_area; mm->unmap_area = arch_unmap_area; } else { - mm->mmap_base = mmap_base(); + mm->mmap_base = mmap_base(mm); + +#ifdef CONFIG_PAX_RANDMMAP + if (mm->pax_flags & MF_PAX_RANDMMAP) + mm->mmap_base -= mm->delta_mmap + mm->delta_stack; +#endif + mm->get_unmapped_area = arch_get_unmapped_area_topdown; mm->unmap_area = arch_unmap_area_topdown; } diff -urNp linux-2.6.32.48/arch/x86/mm/mmio-mod.c linux-2.6.32.48-openvz/arch/x86/mm/mmio-mod.c --- linux-2.6.32.48/arch/x86/mm/mmio-mod.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/mm/mmio-mod.c 2011-11-17 20:09:04.000000000 -0500 @@ -193,7 +193,7 @@ static void pre(struct kmmio_probe *p, s break; default: { - unsigned char *ip = (unsigned char *)instptr; + unsigned char *ip = (unsigned char *)ktla_ktva(instptr); my_trace->opcode = MMIO_UNKNOWN_OP; my_trace->width = 0; my_trace->value = (*ip) << 16 | *(ip + 1) << 8 | @@ -233,7 +233,7 @@ static void post(struct kmmio_probe *p, static void ioremap_trace_core(resource_size_t offset, unsigned long size, void __iomem *addr) { - static atomic_t next_id; + static atomic_unchecked_t next_id; struct remap_trace *trace = kmalloc(sizeof(*trace), GFP_KERNEL); /* These are page-unaligned. */ struct mmiotrace_map map = { @@ -257,7 +257,7 @@ static void ioremap_trace_core(resource_ .private = trace }, .phys = offset, - .id = atomic_inc_return(&next_id) + .id = atomic_inc_return_unchecked(&next_id) }; map.map_id = trace->id; diff -urNp linux-2.6.32.48/arch/x86/mm/numa_32.c linux-2.6.32.48-openvz/arch/x86/mm/numa_32.c --- linux-2.6.32.48/arch/x86/mm/numa_32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/mm/numa_32.c 2011-11-17 20:08:53.000000000 -0500 @@ -98,7 +98,6 @@ unsigned long node_memmap_size_bytes(int } #endif -extern unsigned long find_max_low_pfn(void); extern unsigned long highend_pfn, highstart_pfn; #define LARGE_PAGE_BYTES (PTRS_PER_PTE * PAGE_SIZE) diff -urNp linux-2.6.32.48/arch/x86/mm/pageattr.c linux-2.6.32.48-openvz/arch/x86/mm/pageattr.c --- linux-2.6.32.48/arch/x86/mm/pageattr.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/mm/pageattr.c 2011-11-17 20:08:53.000000000 -0500 @@ -261,16 +261,17 @@ static inline pgprot_t static_protection * PCI BIOS based config access (CONFIG_PCI_GOBIOS) support. */ if (within(pfn, BIOS_BEGIN >> PAGE_SHIFT, BIOS_END >> PAGE_SHIFT)) - pgprot_val(forbidden) |= _PAGE_NX; + pgprot_val(forbidden) |= _PAGE_NX & __supported_pte_mask; /* * The kernel text needs to be executable for obvious reasons * Does not cover __inittext since that is gone later on. On * 64bit we do not enforce !NX on the low mapping */ - if (within(address, (unsigned long)_text, (unsigned long)_etext)) - pgprot_val(forbidden) |= _PAGE_NX; + if (within(address, ktla_ktva((unsigned long)_text), ktla_ktva((unsigned long)_etext))) + pgprot_val(forbidden) |= _PAGE_NX & __supported_pte_mask; +#ifdef CONFIG_DEBUG_RODATA /* * The .rodata section needs to be read-only. Using the pfn * catches all aliases. @@ -278,6 +279,14 @@ static inline pgprot_t static_protection if (within(pfn, __pa((unsigned long)__start_rodata) >> PAGE_SHIFT, __pa((unsigned long)__end_rodata) >> PAGE_SHIFT)) pgprot_val(forbidden) |= _PAGE_RW; +#endif + +#ifdef CONFIG_PAX_KERNEXEC + if (within(pfn, __pa((unsigned long)&_text), __pa((unsigned long)&_sdata))) { + pgprot_val(forbidden) |= _PAGE_RW; + pgprot_val(forbidden) |= _PAGE_NX & __supported_pte_mask; + } +#endif prot = __pgprot(pgprot_val(prot) & ~pgprot_val(forbidden)); @@ -331,23 +340,37 @@ EXPORT_SYMBOL_GPL(lookup_address); static void __set_pmd_pte(pte_t *kpte, unsigned long address, pte_t pte) { /* change init_mm */ + pax_open_kernel(); set_pte_atomic(kpte, pte); + #ifdef CONFIG_X86_32 if (!SHARED_KERNEL_PMD) { + +#ifdef CONFIG_PAX_PER_CPU_PGD + unsigned long cpu; +#else struct page *page; +#endif +#ifdef CONFIG_PAX_PER_CPU_PGD + for (cpu = 0; cpu < NR_CPUS; ++cpu) { + pgd_t *pgd = get_cpu_pgd(cpu); +#else list_for_each_entry(page, &pgd_list, lru) { - pgd_t *pgd; + pgd_t *pgd = (pgd_t *)page_address(page); +#endif + pud_t *pud; pmd_t *pmd; - pgd = (pgd_t *)page_address(page) + pgd_index(address); + pgd += pgd_index(address); pud = pud_offset(pgd, address); pmd = pmd_offset(pud, address); set_pte_atomic((pte_t *)pmd, pte); } } #endif + pax_close_kernel(); } static int diff -urNp linux-2.6.32.48/arch/x86/mm/pageattr-test.c linux-2.6.32.48-openvz/arch/x86/mm/pageattr-test.c --- linux-2.6.32.48/arch/x86/mm/pageattr-test.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/mm/pageattr-test.c 2011-11-17 20:08:53.000000000 -0500 @@ -36,7 +36,7 @@ enum { static int pte_testbit(pte_t pte) { - return pte_flags(pte) & _PAGE_UNUSED1; + return pte_flags(pte) & _PAGE_CPA_TEST; } struct split_state { diff -urNp linux-2.6.32.48/arch/x86/mm/pat.c linux-2.6.32.48-openvz/arch/x86/mm/pat.c --- linux-2.6.32.48/arch/x86/mm/pat.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/mm/pat.c 2011-11-17 20:08:53.000000000 -0500 @@ -258,7 +258,7 @@ chk_conflict(struct memtype *new, struct conflict: printk(KERN_INFO "%s:%d conflicting memory types " - "%Lx-%Lx %s<->%s\n", current->comm, current->pid, new->start, + "%Lx-%Lx %s<->%s\n", current->comm, task_pid_nr(current), new->start, new->end, cattr_name(new->type), cattr_name(entry->type)); return -EBUSY; } @@ -559,7 +559,7 @@ unlock_ret: if (err) { printk(KERN_INFO "%s:%d freeing invalid memtype %Lx-%Lx\n", - current->comm, current->pid, start, end); + current->comm, task_pid_nr(current), start, end); } dprintk("free_memtype request 0x%Lx-0x%Lx\n", start, end); @@ -689,8 +689,8 @@ static inline int range_is_allowed(unsig while (cursor < to) { if (!devmem_is_allowed(pfn)) { printk(KERN_INFO - "Program %s tried to access /dev/mem between %Lx->%Lx.\n", - current->comm, from, to); + "Program %s tried to access /dev/mem between %Lx->%Lx (%Lx).\n", + current->comm, from, to, cursor); return 0; } cursor += PAGE_SIZE; @@ -755,7 +755,7 @@ int kernel_map_sync_memtype(u64 base, un printk(KERN_INFO "%s:%d ioremap_change_attr failed %s " "for %Lx-%Lx\n", - current->comm, current->pid, + current->comm, task_pid_nr(current), cattr_name(flags), base, (unsigned long long)(base + size)); return -EINVAL; @@ -813,7 +813,7 @@ static int reserve_pfn_range(u64 paddr, free_memtype(paddr, paddr + size); printk(KERN_ERR "%s:%d map pfn expected mapping type %s" " for %Lx-%Lx, got %s\n", - current->comm, current->pid, + current->comm, task_pid_nr(current), cattr_name(want_flags), (unsigned long long)paddr, (unsigned long long)(paddr + size), diff -urNp linux-2.6.32.48/arch/x86/mm/pf_in.c linux-2.6.32.48-openvz/arch/x86/mm/pf_in.c --- linux-2.6.32.48/arch/x86/mm/pf_in.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/mm/pf_in.c 2011-11-17 20:09:04.000000000 -0500 @@ -148,7 +148,7 @@ enum reason_type get_ins_type(unsigned l int i; enum reason_type rv = OTHERS; - p = (unsigned char *)ins_addr; + p = (unsigned char *)ktla_ktva(ins_addr); p += skip_prefix(p, &prf); p += get_opcode(p, &opcode); @@ -168,7 +168,7 @@ static unsigned int get_ins_reg_width(un struct prefix_bits prf; int i; - p = (unsigned char *)ins_addr; + p = (unsigned char *)ktla_ktva(ins_addr); p += skip_prefix(p, &prf); p += get_opcode(p, &opcode); @@ -191,7 +191,7 @@ unsigned int get_ins_mem_width(unsigned struct prefix_bits prf; int i; - p = (unsigned char *)ins_addr; + p = (unsigned char *)ktla_ktva(ins_addr); p += skip_prefix(p, &prf); p += get_opcode(p, &opcode); @@ -417,7 +417,7 @@ unsigned long get_ins_reg_val(unsigned l int i; unsigned long rv; - p = (unsigned char *)ins_addr; + p = (unsigned char *)ktla_ktva(ins_addr); p += skip_prefix(p, &prf); p += get_opcode(p, &opcode); for (i = 0; i < ARRAY_SIZE(reg_rop); i++) @@ -472,7 +472,7 @@ unsigned long get_ins_imm_val(unsigned l int i; unsigned long rv; - p = (unsigned char *)ins_addr; + p = (unsigned char *)ktla_ktva(ins_addr); p += skip_prefix(p, &prf); p += get_opcode(p, &opcode); for (i = 0; i < ARRAY_SIZE(imm_wop); i++) diff -urNp linux-2.6.32.48/arch/x86/mm/pgtable_32.c linux-2.6.32.48-openvz/arch/x86/mm/pgtable_32.c --- linux-2.6.32.48/arch/x86/mm/pgtable_32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/mm/pgtable_32.c 2011-11-17 20:08:53.000000000 -0500 @@ -49,10 +49,13 @@ void set_pte_vaddr(unsigned long vaddr, return; } pte = pte_offset_kernel(pmd, vaddr); + + pax_open_kernel(); if (pte_val(pteval)) set_pte_at(&init_mm, vaddr, pte, pteval); else pte_clear(&init_mm, vaddr, pte); + pax_close_kernel(); /* * It's enough to flush this one mapping. diff -urNp linux-2.6.32.48/arch/x86/mm/pgtable.c linux-2.6.32.48-openvz/arch/x86/mm/pgtable.c --- linux-2.6.32.48/arch/x86/mm/pgtable.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/mm/pgtable.c 2011-11-17 20:08:53.000000000 -0500 @@ -4,7 +4,8 @@ #include #include -#define PGALLOC_GFP GFP_KERNEL | __GFP_NOTRACK | __GFP_REPEAT | __GFP_ZERO +#define PGALLOC_GFP GFP_KERNEL | __GFP_NOTRACK | __GFP_REPEAT | __GFP_ZERO | __GFP_UBC +#define PGALLOC_KERN_GFP GFP_KERNEL | __GFP_NOTRACK | __GFP_REPEAT | __GFP_ZERO #ifdef CONFIG_HIGHPTE #define PGALLOC_USER_GFP __GFP_HIGHMEM @@ -16,7 +17,7 @@ gfp_t __userpte_alloc_gfp = PGALLOC_GFP pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address) { - return (pte_t *)__get_free_page(PGALLOC_GFP); + return (pte_t *)__get_free_page(PGALLOC_KERN_GFP); } pgtable_t pte_alloc_one(struct mm_struct *mm, unsigned long address) @@ -83,9 +84,52 @@ static inline void pgd_list_del(pgd_t *p list_del(&page->lru); } -#define UNSHARED_PTRS_PER_PGD \ - (SHARED_KERNEL_PMD ? KERNEL_PGD_BOUNDARY : PTRS_PER_PGD) +#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) +pgdval_t clone_pgd_mask __read_only = ~_PAGE_PRESENT; +void __shadow_user_pgds(pgd_t *dst, const pgd_t *src, int count) +{ + while (count--) + *dst++ = __pgd((pgd_val(*src++) | (_PAGE_NX & __supported_pte_mask)) & ~_PAGE_USER); +} +#endif + +#ifdef CONFIG_PAX_PER_CPU_PGD +void __clone_user_pgds(pgd_t *dst, const pgd_t *src, int count) +{ + while (count--) + +#if defined(CONFIG_X86_64) && defined(CONFIG_PAX_MEMORY_UDEREF) + *dst++ = __pgd(pgd_val(*src++) & clone_pgd_mask); +#else + *dst++ = *src++; +#endif + +} +#endif + +#ifdef CONFIG_X86_64 +#define pxd_t pud_t +#define pyd_t pgd_t +#define paravirt_release_pxd(pfn) paravirt_release_pud(pfn) +#define pxd_free(mm, pud) pud_free((mm), (pud)) +#define pyd_populate(mm, pgd, pud) pgd_populate((mm), (pgd), (pud)) +#define pyd_offset(mm ,address) pgd_offset((mm), (address)) +#define PYD_SIZE PGDIR_SIZE +#else +#define pxd_t pmd_t +#define pyd_t pud_t +#define paravirt_release_pxd(pfn) paravirt_release_pmd(pfn) +#define pxd_free(mm, pud) pmd_free((mm), (pud)) +#define pyd_populate(mm, pgd, pud) pud_populate((mm), (pgd), (pud)) +#define pyd_offset(mm ,address) pud_offset((mm), (address)) +#define PYD_SIZE PUD_SIZE +#endif + +#ifdef CONFIG_PAX_PER_CPU_PGD +static inline void pgd_ctor(pgd_t *pgd) {} +static inline void pgd_dtor(pgd_t *pgd) {} +#else static void pgd_ctor(pgd_t *pgd) { /* If the pgd points to a shared pagetable level (either the @@ -119,6 +163,7 @@ static void pgd_dtor(pgd_t *pgd) pgd_list_del(pgd); spin_unlock_irqrestore(&pgd_lock, flags); } +#endif /* * List of all pgd's needed for non-PAE so it can invalidate entries @@ -131,7 +176,7 @@ static void pgd_dtor(pgd_t *pgd) * -- wli */ -#ifdef CONFIG_X86_PAE +#if defined(CONFIG_X86_32) && defined(CONFIG_X86_PAE) /* * In PAE mode, we need to do a cr3 reload (=tlb flush) when * updating the top-level pagetable entries to guarantee the @@ -143,7 +188,7 @@ static void pgd_dtor(pgd_t *pgd) * not shared between pagetables (!SHARED_KERNEL_PMDS), we allocate * and initialize the kernel pmds here. */ -#define PREALLOCATED_PMDS UNSHARED_PTRS_PER_PGD +#define PREALLOCATED_PXDS (SHARED_KERNEL_PMD ? KERNEL_PGD_BOUNDARY : PTRS_PER_PGD) void pud_populate(struct mm_struct *mm, pud_t *pudp, pmd_t *pmd) { @@ -161,36 +206,38 @@ void pud_populate(struct mm_struct *mm, */ flush_tlb_mm(mm); } +#elif defined(CONFIG_X86_64) && defined(CONFIG_PAX_PER_CPU_PGD) +#define PREALLOCATED_PXDS USER_PGD_PTRS #else /* !CONFIG_X86_PAE */ /* No need to prepopulate any pagetable entries in non-PAE modes. */ -#define PREALLOCATED_PMDS 0 +#define PREALLOCATED_PXDS 0 #endif /* CONFIG_X86_PAE */ -static void free_pmds(pmd_t *pmds[]) +static void free_pxds(pxd_t *pxds[]) { int i; - for(i = 0; i < PREALLOCATED_PMDS; i++) - if (pmds[i]) - free_page((unsigned long)pmds[i]); + for(i = 0; i < PREALLOCATED_PXDS; i++) + if (pxds[i]) + free_page((unsigned long)pxds[i]); } -static int preallocate_pmds(pmd_t *pmds[]) +static int preallocate_pxds(pxd_t *pxds[]) { int i; bool failed = false; - for(i = 0; i < PREALLOCATED_PMDS; i++) { - pmd_t *pmd = (pmd_t *)__get_free_page(PGALLOC_GFP); - if (pmd == NULL) + for(i = 0; i < PREALLOCATED_PXDS; i++) { + pxd_t *pxd = (pxd_t *)__get_free_page(PGALLOC_GFP); + if (pxd == NULL) failed = true; - pmds[i] = pmd; + pxds[i] = pxd; } if (failed) { - free_pmds(pmds); + free_pxds(pxds); return -ENOMEM; } @@ -203,51 +250,56 @@ static int preallocate_pmds(pmd_t *pmds[ * preallocate which never got a corresponding vma will need to be * freed manually. */ -static void pgd_mop_up_pmds(struct mm_struct *mm, pgd_t *pgdp) +static void pgd_mop_up_pxds(struct mm_struct *mm, pgd_t *pgdp) { int i; - for(i = 0; i < PREALLOCATED_PMDS; i++) { + for(i = 0; i < PREALLOCATED_PXDS; i++) { pgd_t pgd = pgdp[i]; if (pgd_val(pgd) != 0) { - pmd_t *pmd = (pmd_t *)pgd_page_vaddr(pgd); + pxd_t *pxd = (pxd_t *)pgd_page_vaddr(pgd); - pgdp[i] = native_make_pgd(0); + set_pgd(pgdp + i, native_make_pgd(0)); - paravirt_release_pmd(pgd_val(pgd) >> PAGE_SHIFT); - pmd_free(mm, pmd); + paravirt_release_pxd(pgd_val(pgd) >> PAGE_SHIFT); + pxd_free(mm, pxd); } } } -static void pgd_prepopulate_pmd(struct mm_struct *mm, pgd_t *pgd, pmd_t *pmds[]) +static void pgd_prepopulate_pxd(struct mm_struct *mm, pgd_t *pgd, pxd_t *pxds[]) { - pud_t *pud; + pyd_t *pyd; unsigned long addr; int i; - if (PREALLOCATED_PMDS == 0) /* Work around gcc-3.4.x bug */ + if (PREALLOCATED_PXDS == 0) /* Work around gcc-3.4.x bug */ return; - pud = pud_offset(pgd, 0); +#ifdef CONFIG_X86_64 + pyd = pyd_offset(mm, 0L); +#else + pyd = pyd_offset(pgd, 0L); +#endif - for (addr = i = 0; i < PREALLOCATED_PMDS; - i++, pud++, addr += PUD_SIZE) { - pmd_t *pmd = pmds[i]; + for (addr = i = 0; i < PREALLOCATED_PXDS; + i++, pyd++, addr += PYD_SIZE) { + pxd_t *pxd = pxds[i]; if (i >= KERNEL_PGD_BOUNDARY) - memcpy(pmd, (pmd_t *)pgd_page_vaddr(swapper_pg_dir[i]), - sizeof(pmd_t) * PTRS_PER_PMD); + memcpy(pxd, (pxd_t *)pgd_page_vaddr(swapper_pg_dir[i]), + sizeof(pxd_t) * PTRS_PER_PMD); - pud_populate(mm, pud, pmd); + pyd_populate(mm, pyd, pxd); } } pgd_t *pgd_alloc(struct mm_struct *mm) { pgd_t *pgd; - pmd_t *pmds[PREALLOCATED_PMDS]; + pxd_t *pxds[PREALLOCATED_PXDS]; + unsigned long flags; pgd = (pgd_t *)__get_free_page(PGALLOC_GFP); @@ -257,11 +309,11 @@ pgd_t *pgd_alloc(struct mm_struct *mm) mm->pgd = pgd; - if (preallocate_pmds(pmds) != 0) + if (preallocate_pxds(pxds) != 0) goto out_free_pgd; if (paravirt_pgd_alloc(mm) != 0) - goto out_free_pmds; + goto out_free_pxds; /* * Make sure that pre-populating the pmds is atomic with @@ -271,14 +323,14 @@ pgd_t *pgd_alloc(struct mm_struct *mm) spin_lock_irqsave(&pgd_lock, flags); pgd_ctor(pgd); - pgd_prepopulate_pmd(mm, pgd, pmds); + pgd_prepopulate_pxd(mm, pgd, pxds); spin_unlock_irqrestore(&pgd_lock, flags); return pgd; -out_free_pmds: - free_pmds(pmds); +out_free_pxds: + free_pxds(pxds); out_free_pgd: free_page((unsigned long)pgd); out: @@ -287,7 +339,7 @@ out: void pgd_free(struct mm_struct *mm, pgd_t *pgd) { - pgd_mop_up_pmds(mm, pgd); + pgd_mop_up_pxds(mm, pgd); pgd_dtor(pgd); paravirt_pgd_free(mm, pgd); free_page((unsigned long)pgd); diff -urNp linux-2.6.32.48/arch/x86/mm/setup_nx.c linux-2.6.32.48-openvz/arch/x86/mm/setup_nx.c --- linux-2.6.32.48/arch/x86/mm/setup_nx.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/mm/setup_nx.c 2011-11-17 20:08:53.000000000 -0500 @@ -4,11 +4,10 @@ #include +#if defined(CONFIG_X86_32) && defined(CONFIG_X86_PAE) int nx_enabled; -#if defined(CONFIG_X86_64) || defined(CONFIG_X86_PAE) -static int disable_nx __cpuinitdata; - +#ifndef CONFIG_PAX_PAGEEXEC /* * noexec = on|off * @@ -22,32 +21,26 @@ static int __init noexec_setup(char *str if (!str) return -EINVAL; if (!strncmp(str, "on", 2)) { - __supported_pte_mask |= _PAGE_NX; - disable_nx = 0; + nx_enabled = 1; } else if (!strncmp(str, "off", 3)) { - disable_nx = 1; - __supported_pte_mask &= ~_PAGE_NX; + nx_enabled = 0; } return 0; } early_param("noexec", noexec_setup); #endif +#endif #ifdef CONFIG_X86_PAE void __init set_nx(void) { - unsigned int v[4], l, h; + if (!nx_enabled && cpu_has_nx) { + unsigned l, h; - if (cpu_has_pae && (cpuid_eax(0x80000000) > 0x80000001)) { - cpuid(0x80000001, &v[0], &v[1], &v[2], &v[3]); - - if ((v[3] & (1 << 20)) && !disable_nx) { - rdmsr(MSR_EFER, l, h); - l |= EFER_NX; - wrmsr(MSR_EFER, l, h); - nx_enabled = 1; - __supported_pte_mask |= _PAGE_NX; - } + __supported_pte_mask &= ~_PAGE_NX; + rdmsr(MSR_EFER, l, h); + l &= ~EFER_NX; + wrmsr(MSR_EFER, l, h); } } #else @@ -62,7 +55,7 @@ void __cpuinit check_efer(void) unsigned long efer; rdmsrl(MSR_EFER, efer); - if (!(efer & EFER_NX) || disable_nx) + if (!(efer & EFER_NX) || !nx_enabled) __supported_pte_mask &= ~_PAGE_NX; } #endif diff -urNp linux-2.6.32.48/arch/x86/mm/tlb.c linux-2.6.32.48-openvz/arch/x86/mm/tlb.c --- linux-2.6.32.48/arch/x86/mm/tlb.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/mm/tlb.c 2011-11-17 20:08:53.000000000 -0500 @@ -61,7 +61,11 @@ void leave_mm(int cpu) BUG(); cpumask_clear_cpu(cpu, mm_cpumask(percpu_read(cpu_tlbstate.active_mm))); + +#ifndef CONFIG_PAX_PER_CPU_PGD load_cr3(swapper_pg_dir); +#endif + } EXPORT_SYMBOL_GPL(leave_mm); @@ -256,6 +260,8 @@ void flush_tlb_mm(struct mm_struct *mm) preempt_enable(); } +EXPORT_SYMBOL(flush_tlb_mm); + void flush_tlb_page(struct vm_area_struct *vma, unsigned long va) { struct mm_struct *mm = vma->vm_mm; diff -urNp linux-2.6.32.48/arch/x86/oprofile/backtrace.c linux-2.6.32.48-openvz/arch/x86/oprofile/backtrace.c --- linux-2.6.32.48/arch/x86/oprofile/backtrace.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/oprofile/backtrace.c 2011-11-17 20:08:53.000000000 -0500 @@ -57,7 +57,7 @@ static struct frame_head *dump_user_back struct frame_head bufhead[2]; /* Also check accessibility of one struct frame_head beyond */ - if (!access_ok(VERIFY_READ, head, sizeof(bufhead))) + if (!__access_ok(VERIFY_READ, head, sizeof(bufhead))) return NULL; if (__copy_from_user_inatomic(bufhead, head, sizeof(bufhead))) return NULL; @@ -77,7 +77,7 @@ x86_backtrace(struct pt_regs * const reg { struct frame_head *head = (struct frame_head *)frame_pointer(regs); - if (!user_mode_vm(regs)) { + if (!user_mode(regs)) { unsigned long stack = kernel_stack_pointer(regs); if (depth) dump_trace(NULL, regs, (unsigned long *)stack, 0, diff -urNp linux-2.6.32.48/arch/x86/oprofile/op_model_p4.c linux-2.6.32.48-openvz/arch/x86/oprofile/op_model_p4.c --- linux-2.6.32.48/arch/x86/oprofile/op_model_p4.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/oprofile/op_model_p4.c 2011-11-17 20:08:53.000000000 -0500 @@ -50,7 +50,7 @@ static inline void setup_num_counters(vo #endif } -static int inline addr_increment(void) +static inline int addr_increment(void) { #ifdef CONFIG_SMP return smp_num_siblings == 2 ? 2 : 1; diff -urNp linux-2.6.32.48/arch/x86/pci/common.c linux-2.6.32.48-openvz/arch/x86/pci/common.c --- linux-2.6.32.48/arch/x86/pci/common.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/pci/common.c 2011-11-17 20:08:53.000000000 -0500 @@ -31,8 +31,8 @@ int noioapicreroute = 1; int pcibios_last_bus = -1; unsigned long pirq_table_addr; struct pci_bus *pci_root_bus; -struct pci_raw_ops *raw_pci_ops; -struct pci_raw_ops *raw_pci_ext_ops; +const struct pci_raw_ops *raw_pci_ops; +const struct pci_raw_ops *raw_pci_ext_ops; int raw_pci_read(unsigned int domain, unsigned int bus, unsigned int devfn, int reg, int len, u32 *val) diff -urNp linux-2.6.32.48/arch/x86/pci/direct.c linux-2.6.32.48-openvz/arch/x86/pci/direct.c --- linux-2.6.32.48/arch/x86/pci/direct.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/pci/direct.c 2011-11-17 20:08:53.000000000 -0500 @@ -79,7 +79,7 @@ static int pci_conf1_write(unsigned int #undef PCI_CONF1_ADDRESS -struct pci_raw_ops pci_direct_conf1 = { +const struct pci_raw_ops pci_direct_conf1 = { .read = pci_conf1_read, .write = pci_conf1_write, }; @@ -173,7 +173,7 @@ static int pci_conf2_write(unsigned int #undef PCI_CONF2_ADDRESS -struct pci_raw_ops pci_direct_conf2 = { +const struct pci_raw_ops pci_direct_conf2 = { .read = pci_conf2_read, .write = pci_conf2_write, }; @@ -189,7 +189,7 @@ struct pci_raw_ops pci_direct_conf2 = { * This should be close to trivial, but it isn't, because there are buggy * chipsets (yes, you guessed it, by Intel and Compaq) that have no class ID. */ -static int __init pci_sanity_check(struct pci_raw_ops *o) +static int __init pci_sanity_check(const struct pci_raw_ops *o) { u32 x = 0; int year, devfn; diff -urNp linux-2.6.32.48/arch/x86/pci/mmconfig_32.c linux-2.6.32.48-openvz/arch/x86/pci/mmconfig_32.c --- linux-2.6.32.48/arch/x86/pci/mmconfig_32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/pci/mmconfig_32.c 2011-11-17 20:08:53.000000000 -0500 @@ -125,7 +125,7 @@ static int pci_mmcfg_write(unsigned int return 0; } -static struct pci_raw_ops pci_mmcfg = { +static const struct pci_raw_ops pci_mmcfg = { .read = pci_mmcfg_read, .write = pci_mmcfg_write, }; diff -urNp linux-2.6.32.48/arch/x86/pci/mmconfig_64.c linux-2.6.32.48-openvz/arch/x86/pci/mmconfig_64.c --- linux-2.6.32.48/arch/x86/pci/mmconfig_64.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/pci/mmconfig_64.c 2011-11-17 20:08:53.000000000 -0500 @@ -104,7 +104,7 @@ static int pci_mmcfg_write(unsigned int return 0; } -static struct pci_raw_ops pci_mmcfg = { +static const struct pci_raw_ops pci_mmcfg = { .read = pci_mmcfg_read, .write = pci_mmcfg_write, }; diff -urNp linux-2.6.32.48/arch/x86/pci/numaq_32.c linux-2.6.32.48-openvz/arch/x86/pci/numaq_32.c --- linux-2.6.32.48/arch/x86/pci/numaq_32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/pci/numaq_32.c 2011-11-17 20:08:53.000000000 -0500 @@ -112,7 +112,7 @@ static int pci_conf1_mq_write(unsigned i #undef PCI_CONF1_MQ_ADDRESS -static struct pci_raw_ops pci_direct_conf1_mq = { +static const struct pci_raw_ops pci_direct_conf1_mq = { .read = pci_conf1_mq_read, .write = pci_conf1_mq_write }; diff -urNp linux-2.6.32.48/arch/x86/pci/olpc.c linux-2.6.32.48-openvz/arch/x86/pci/olpc.c --- linux-2.6.32.48/arch/x86/pci/olpc.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/pci/olpc.c 2011-11-17 20:08:53.000000000 -0500 @@ -297,7 +297,7 @@ static int pci_olpc_write(unsigned int s return 0; } -static struct pci_raw_ops pci_olpc_conf = { +static const struct pci_raw_ops pci_olpc_conf = { .read = pci_olpc_read, .write = pci_olpc_write, }; diff -urNp linux-2.6.32.48/arch/x86/pci/pcbios.c linux-2.6.32.48-openvz/arch/x86/pci/pcbios.c --- linux-2.6.32.48/arch/x86/pci/pcbios.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/pci/pcbios.c 2011-11-17 20:08:53.000000000 -0500 @@ -56,50 +56,93 @@ union bios32 { static struct { unsigned long address; unsigned short segment; -} bios32_indirect = { 0, __KERNEL_CS }; +} bios32_indirect __read_only = { 0, __PCIBIOS_CS }; /* * Returns the entry point for the given service, NULL on error */ -static unsigned long bios32_service(unsigned long service) +static unsigned long __devinit bios32_service(unsigned long service) { unsigned char return_code; /* %al */ unsigned long address; /* %ebx */ unsigned long length; /* %ecx */ unsigned long entry; /* %edx */ unsigned long flags; + struct desc_struct d, *gdt; local_irq_save(flags); - __asm__("lcall *(%%edi); cld" + + gdt = get_cpu_gdt_table(smp_processor_id()); + + pack_descriptor(&d, 0UL, 0xFFFFFUL, 0x9B, 0xC); + write_gdt_entry(gdt, GDT_ENTRY_PCIBIOS_CS, &d, DESCTYPE_S); + pack_descriptor(&d, 0UL, 0xFFFFFUL, 0x93, 0xC); + write_gdt_entry(gdt, GDT_ENTRY_PCIBIOS_DS, &d, DESCTYPE_S); + + __asm__("movw %w7, %%ds; lcall *(%%edi); push %%ss; pop %%ds; cld" : "=a" (return_code), "=b" (address), "=c" (length), "=d" (entry) : "0" (service), "1" (0), - "D" (&bios32_indirect)); + "D" (&bios32_indirect), + "r"(__PCIBIOS_DS) + : "memory"); + + pax_open_kernel(); + gdt[GDT_ENTRY_PCIBIOS_CS].a = 0; + gdt[GDT_ENTRY_PCIBIOS_CS].b = 0; + gdt[GDT_ENTRY_PCIBIOS_DS].a = 0; + gdt[GDT_ENTRY_PCIBIOS_DS].b = 0; + pax_close_kernel(); + local_irq_restore(flags); switch (return_code) { - case 0: - return address + entry; - case 0x80: /* Not present */ - printk(KERN_WARNING "bios32_service(0x%lx): not present\n", service); - return 0; - default: /* Shouldn't happen */ - printk(KERN_WARNING "bios32_service(0x%lx): returned 0x%x -- BIOS bug!\n", - service, return_code); + case 0: { + int cpu; + unsigned char flags; + + printk(KERN_INFO "bios32_service: base:%08lx length:%08lx entry:%08lx\n", address, length, entry); + if (address >= 0xFFFF0 || length > 0x100000 - address || length <= entry) { + printk(KERN_WARNING "bios32_service: not valid\n"); return 0; + } + address = address + PAGE_OFFSET; + length += 16UL; /* some BIOSs underreport this... */ + flags = 4; + if (length >= 64*1024*1024) { + length >>= PAGE_SHIFT; + flags |= 8; + } + + for (cpu = 0; cpu < NR_CPUS; cpu++) { + gdt = get_cpu_gdt_table(cpu); + pack_descriptor(&d, address, length, 0x9b, flags); + write_gdt_entry(gdt, GDT_ENTRY_PCIBIOS_CS, &d, DESCTYPE_S); + pack_descriptor(&d, address, length, 0x93, flags); + write_gdt_entry(gdt, GDT_ENTRY_PCIBIOS_DS, &d, DESCTYPE_S); + } + return entry; + } + case 0x80: /* Not present */ + printk(KERN_WARNING "bios32_service(0x%lx): not present\n", service); + return 0; + default: /* Shouldn't happen */ + printk(KERN_WARNING "bios32_service(0x%lx): returned 0x%x -- BIOS bug!\n", + service, return_code); + return 0; } } static struct { unsigned long address; unsigned short segment; -} pci_indirect = { 0, __KERNEL_CS }; +} pci_indirect __read_only = { 0, __PCIBIOS_CS }; -static int pci_bios_present; +static int pci_bios_present __read_only; static int __devinit check_pcibios(void) { @@ -108,11 +151,13 @@ static int __devinit check_pcibios(void) unsigned long flags, pcibios_entry; if ((pcibios_entry = bios32_service(PCI_SERVICE))) { - pci_indirect.address = pcibios_entry + PAGE_OFFSET; + pci_indirect.address = pcibios_entry; local_irq_save(flags); - __asm__( - "lcall *(%%edi); cld\n\t" + __asm__("movw %w6, %%ds\n\t" + "lcall *%%ss:(%%edi); cld\n\t" + "push %%ss\n\t" + "pop %%ds\n\t" "jc 1f\n\t" "xor %%ah, %%ah\n" "1:" @@ -121,7 +166,8 @@ static int __devinit check_pcibios(void) "=b" (ebx), "=c" (ecx) : "1" (PCIBIOS_PCI_BIOS_PRESENT), - "D" (&pci_indirect) + "D" (&pci_indirect), + "r" (__PCIBIOS_DS) : "memory"); local_irq_restore(flags); @@ -165,7 +211,10 @@ static int pci_bios_read(unsigned int se switch (len) { case 1: - __asm__("lcall *(%%esi); cld\n\t" + __asm__("movw %w6, %%ds\n\t" + "lcall *%%ss:(%%esi); cld\n\t" + "push %%ss\n\t" + "pop %%ds\n\t" "jc 1f\n\t" "xor %%ah, %%ah\n" "1:" @@ -174,7 +223,8 @@ static int pci_bios_read(unsigned int se : "1" (PCIBIOS_READ_CONFIG_BYTE), "b" (bx), "D" ((long)reg), - "S" (&pci_indirect)); + "S" (&pci_indirect), + "r" (__PCIBIOS_DS)); /* * Zero-extend the result beyond 8 bits, do not trust the * BIOS having done it: @@ -182,7 +232,10 @@ static int pci_bios_read(unsigned int se *value &= 0xff; break; case 2: - __asm__("lcall *(%%esi); cld\n\t" + __asm__("movw %w6, %%ds\n\t" + "lcall *%%ss:(%%esi); cld\n\t" + "push %%ss\n\t" + "pop %%ds\n\t" "jc 1f\n\t" "xor %%ah, %%ah\n" "1:" @@ -191,7 +244,8 @@ static int pci_bios_read(unsigned int se : "1" (PCIBIOS_READ_CONFIG_WORD), "b" (bx), "D" ((long)reg), - "S" (&pci_indirect)); + "S" (&pci_indirect), + "r" (__PCIBIOS_DS)); /* * Zero-extend the result beyond 16 bits, do not trust the * BIOS having done it: @@ -199,7 +253,10 @@ static int pci_bios_read(unsigned int se *value &= 0xffff; break; case 4: - __asm__("lcall *(%%esi); cld\n\t" + __asm__("movw %w6, %%ds\n\t" + "lcall *%%ss:(%%esi); cld\n\t" + "push %%ss\n\t" + "pop %%ds\n\t" "jc 1f\n\t" "xor %%ah, %%ah\n" "1:" @@ -208,7 +265,8 @@ static int pci_bios_read(unsigned int se : "1" (PCIBIOS_READ_CONFIG_DWORD), "b" (bx), "D" ((long)reg), - "S" (&pci_indirect)); + "S" (&pci_indirect), + "r" (__PCIBIOS_DS)); break; } @@ -231,7 +289,10 @@ static int pci_bios_write(unsigned int s switch (len) { case 1: - __asm__("lcall *(%%esi); cld\n\t" + __asm__("movw %w6, %%ds\n\t" + "lcall *%%ss:(%%esi); cld\n\t" + "push %%ss\n\t" + "pop %%ds\n\t" "jc 1f\n\t" "xor %%ah, %%ah\n" "1:" @@ -240,10 +301,14 @@ static int pci_bios_write(unsigned int s "c" (value), "b" (bx), "D" ((long)reg), - "S" (&pci_indirect)); + "S" (&pci_indirect), + "r" (__PCIBIOS_DS)); break; case 2: - __asm__("lcall *(%%esi); cld\n\t" + __asm__("movw %w6, %%ds\n\t" + "lcall *%%ss:(%%esi); cld\n\t" + "push %%ss\n\t" + "pop %%ds\n\t" "jc 1f\n\t" "xor %%ah, %%ah\n" "1:" @@ -252,10 +317,14 @@ static int pci_bios_write(unsigned int s "c" (value), "b" (bx), "D" ((long)reg), - "S" (&pci_indirect)); + "S" (&pci_indirect), + "r" (__PCIBIOS_DS)); break; case 4: - __asm__("lcall *(%%esi); cld\n\t" + __asm__("movw %w6, %%ds\n\t" + "lcall *%%ss:(%%esi); cld\n\t" + "push %%ss\n\t" + "pop %%ds\n\t" "jc 1f\n\t" "xor %%ah, %%ah\n" "1:" @@ -264,7 +333,8 @@ static int pci_bios_write(unsigned int s "c" (value), "b" (bx), "D" ((long)reg), - "S" (&pci_indirect)); + "S" (&pci_indirect), + "r" (__PCIBIOS_DS)); break; } @@ -278,7 +348,7 @@ static int pci_bios_write(unsigned int s * Function table for BIOS32 access */ -static struct pci_raw_ops pci_bios_access = { +static const struct pci_raw_ops pci_bios_access = { .read = pci_bios_read, .write = pci_bios_write }; @@ -287,7 +357,7 @@ static struct pci_raw_ops pci_bios_acces * Try to find PCI BIOS. */ -static struct pci_raw_ops * __devinit pci_find_bios(void) +static const struct pci_raw_ops * __devinit pci_find_bios(void) { union bios32 *check; unsigned char sum; @@ -368,10 +438,13 @@ struct irq_routing_table * pcibios_get_i DBG("PCI: Fetching IRQ routing table... "); __asm__("push %%es\n\t" + "movw %w8, %%ds\n\t" "push %%ds\n\t" "pop %%es\n\t" - "lcall *(%%esi); cld\n\t" + "lcall *%%ss:(%%esi); cld\n\t" "pop %%es\n\t" + "push %%ss\n\t" + "pop %%ds\n" "jc 1f\n\t" "xor %%ah, %%ah\n" "1:" @@ -382,7 +455,8 @@ struct irq_routing_table * pcibios_get_i "1" (0), "D" ((long) &opt), "S" (&pci_indirect), - "m" (opt) + "m" (opt), + "r" (__PCIBIOS_DS) : "memory"); DBG("OK ret=%d, size=%d, map=%x\n", ret, opt.size, map); if (ret & 0xff00) @@ -406,7 +480,10 @@ int pcibios_set_irq_routing(struct pci_d { int ret; - __asm__("lcall *(%%esi); cld\n\t" + __asm__("movw %w5, %%ds\n\t" + "lcall *%%ss:(%%esi); cld\n\t" + "push %%ss\n\t" + "pop %%ds\n" "jc 1f\n\t" "xor %%ah, %%ah\n" "1:" @@ -414,7 +491,8 @@ int pcibios_set_irq_routing(struct pci_d : "0" (PCIBIOS_SET_PCI_HW_INT), "b" ((dev->bus->number << 8) | dev->devfn), "c" ((irq << 8) | (pin + 10)), - "S" (&pci_indirect)); + "S" (&pci_indirect), + "r" (__PCIBIOS_DS)); return !(ret & 0xff00); } EXPORT_SYMBOL(pcibios_set_irq_routing); diff -urNp linux-2.6.32.48/arch/x86/power/cpu.c linux-2.6.32.48-openvz/arch/x86/power/cpu.c --- linux-2.6.32.48/arch/x86/power/cpu.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/power/cpu.c 2011-11-17 20:08:53.000000000 -0500 @@ -129,7 +129,7 @@ static void do_fpu_end(void) static void fix_processor_context(void) { int cpu = smp_processor_id(); - struct tss_struct *t = &per_cpu(init_tss, cpu); + struct tss_struct *t = init_tss + cpu; set_tss_desc(cpu, t); /* * This just modifies memory; should not be @@ -139,7 +139,9 @@ static void fix_processor_context(void) */ #ifdef CONFIG_X86_64 + pax_open_kernel(); get_cpu_gdt_table(cpu)[GDT_ENTRY_TSS].type = 9; + pax_close_kernel(); syscall_init(); /* This sets MSR_*STAR and related */ #endif diff -urNp linux-2.6.32.48/arch/x86/vdso/Makefile linux-2.6.32.48-openvz/arch/x86/vdso/Makefile --- linux-2.6.32.48/arch/x86/vdso/Makefile 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/vdso/Makefile 2011-11-17 20:08:53.000000000 -0500 @@ -122,7 +122,7 @@ quiet_cmd_vdso = VDSO $@ $(VDSO_LDFLAGS) $(VDSO_LDFLAGS_$(filter %.lds,$(^F))) \ -Wl,-T,$(filter %.lds,$^) $(filter %.o,$^) -VDSO_LDFLAGS = -fPIC -shared $(call cc-ldoption, -Wl$(comma)--hash-style=sysv) +VDSO_LDFLAGS = -fPIC -shared -Wl,--no-undefined $(call cc-ldoption, -Wl$(comma)--hash-style=sysv) GCOV_PROFILE := n # diff -urNp linux-2.6.32.48/arch/x86/vdso/vclock_gettime.c linux-2.6.32.48-openvz/arch/x86/vdso/vclock_gettime.c --- linux-2.6.32.48/arch/x86/vdso/vclock_gettime.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/vdso/vclock_gettime.c 2011-11-17 20:08:53.000000000 -0500 @@ -22,24 +22,48 @@ #include #include #include +#include #include "vextern.h" #define gtod vdso_vsyscall_gtod_data +notrace noinline long __vdso_fallback_time(long *t) +{ + long secs; + asm volatile("syscall" + : "=a" (secs) + : "0" (__NR_time),"D" (t) : "r11", "cx", "memory"); + return secs; +} + notrace static long vdso_fallback_gettime(long clock, struct timespec *ts) { long ret; asm("syscall" : "=a" (ret) : - "0" (__NR_clock_gettime),"D" (clock), "S" (ts) : "memory"); + "0" (__NR_clock_gettime),"D" (clock), "S" (ts) : "r11", "cx", "memory"); return ret; } +notrace static inline cycle_t __vdso_vread_hpet(void) +{ + return readl((const void __iomem *)fix_to_virt(VSYSCALL_HPET) + 0xf0); +} + +notrace static inline cycle_t __vdso_vread_tsc(void) +{ + cycle_t ret = (cycle_t)vget_cycles(); + + return ret >= gtod->clock.cycle_last ? ret : gtod->clock.cycle_last; +} + notrace static inline long vgetns(void) { long v; - cycles_t (*vread)(void); - vread = gtod->clock.vread; - v = (vread() - gtod->clock.cycle_last) & gtod->clock.mask; + if (gtod->clock.name[0] == 't' && gtod->clock.name[1] == 's' && gtod->clock.name[2] == 'c' && !gtod->clock.name[3]) + v = __vdso_vread_tsc(); + else + v = __vdso_vread_hpet(); + v = (v - gtod->clock.cycle_last) & gtod->clock.mask; return (v * gtod->clock.mult) >> gtod->clock.shift; } @@ -113,7 +137,9 @@ notrace static noinline int do_monotonic notrace int __vdso_clock_gettime(clockid_t clock, struct timespec *ts) { - if (likely(gtod->sysctl_enabled)) + if (likely(gtod->sysctl_enabled && + ((gtod->clock.name[0] == 'h' && gtod->clock.name[1] == 'p' && gtod->clock.name[2] == 'e' && gtod->clock.name[3] == 't' && !gtod->clock.name[4]) || + (gtod->clock.name[0] == 't' && gtod->clock.name[1] == 's' && gtod->clock.name[2] == 'c' && !gtod->clock.name[3])))) switch (clock) { case CLOCK_REALTIME: if (likely(gtod->clock.vread)) @@ -133,10 +159,20 @@ notrace int __vdso_clock_gettime(clockid int clock_gettime(clockid_t, struct timespec *) __attribute__((weak, alias("__vdso_clock_gettime"))); -notrace int __vdso_gettimeofday(struct timeval *tv, struct timezone *tz) +notrace noinline int __vdso_fallback_gettimeofday(struct timeval *tv, struct timezone *tz) { long ret; - if (likely(gtod->sysctl_enabled && gtod->clock.vread)) { + asm("syscall" : "=a" (ret) : + "0" (__NR_gettimeofday), "D" (tv), "S" (tz) : "r11", "cx", "memory"); + return ret; +} + +notrace int __vdso_gettimeofday(struct timeval *tv, struct timezone *tz) +{ + if (likely(gtod->sysctl_enabled && + ((gtod->clock.name[0] == 'h' && gtod->clock.name[1] == 'p' && gtod->clock.name[2] == 'e' && gtod->clock.name[3] == 't' && !gtod->clock.name[4]) || + (gtod->clock.name[0] == 't' && gtod->clock.name[1] == 's' && gtod->clock.name[2] == 'c' && !gtod->clock.name[3])))) + { if (likely(tv != NULL)) { BUILD_BUG_ON(offsetof(struct timeval, tv_usec) != offsetof(struct timespec, tv_nsec) || @@ -151,9 +187,7 @@ notrace int __vdso_gettimeofday(struct t } return 0; } - asm("syscall" : "=a" (ret) : - "0" (__NR_gettimeofday), "D" (tv), "S" (tz) : "memory"); - return ret; + return __vdso_fallback_gettimeofday(tv, tz); } int gettimeofday(struct timeval *, struct timezone *) __attribute__((weak, alias("__vdso_gettimeofday"))); diff -urNp linux-2.6.32.48/arch/x86/vdso/vdso32-setup.c linux-2.6.32.48-openvz/arch/x86/vdso/vdso32-setup.c --- linux-2.6.32.48/arch/x86/vdso/vdso32-setup.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/vdso/vdso32-setup.c 2011-11-17 20:08:53.000000000 -0500 @@ -17,6 +17,8 @@ #include #include +#include + #include #include #include @@ -25,6 +27,7 @@ #include #include #include +#include enum { VDSO_DISABLED = 0, @@ -37,6 +40,8 @@ enum { #else #define VDSO_DEFAULT VDSO_ENABLED #endif +#undef VDSO_DEFAULT +#define VDSO_DEFAULT VDSO_DISABLED #ifdef CONFIG_X86_64 #define vdso_enabled sysctl_vsyscall32 @@ -193,7 +198,8 @@ static __init void relocate_vdso(Elf32_E } } -static struct page *vdso32_pages[1]; +struct page *vdso32_pages[1]; +EXPORT_SYMBOL_GPL(vdso32_pages); #ifdef CONFIG_X86_64 @@ -226,7 +232,7 @@ static inline void map_compat_vdso(int m void enable_sep_cpu(void) { int cpu = get_cpu(); - struct tss_struct *tss = &per_cpu(init_tss, cpu); + struct tss_struct *tss = init_tss + cpu; if (!boot_cpu_has(X86_FEATURE_SEP)) { put_cpu(); @@ -249,7 +255,7 @@ static int __init gate_vma_init(void) gate_vma.vm_start = FIXADDR_USER_START; gate_vma.vm_end = FIXADDR_USER_END; gate_vma.vm_flags = VM_READ | VM_MAYREAD | VM_EXEC | VM_MAYEXEC; - gate_vma.vm_page_prot = __P101; + gate_vma.vm_page_prot = vm_get_page_prot(gate_vma.vm_flags); /* * Make sure the vDSO gets into every core dump. * Dumping its contents makes post-mortem fully interpretable later @@ -309,16 +315,30 @@ int __init sysenter_setup(void) return 0; } +EXPORT_SYMBOL_GPL(VDSO32_SYSENTER_RETURN); +EXPORT_SYMBOL_GPL(VDSO32_PRELINK); + /* Setup a VMA at program startup for the vsyscall page */ -int arch_setup_additional_pages(struct linux_binprm *bprm, int uses_interp) +int arch_setup_additional_pages(struct linux_binprm *bprm, int uses_interp, + unsigned long map_address) { struct mm_struct *mm = current->mm; - unsigned long addr; + unsigned long addr = map_address; int ret = 0; bool compat; + unsigned long flags; - if (vdso_enabled == VDSO_DISABLED) + if (vdso_enabled == VDSO_DISABLED && map_address == 0) { + current->mm->context.vdso = NULL; return 0; + } + + flags = VM_READ | VM_EXEC | VM_MAYREAD | VM_MAYEXEC | VM_MAYWRITE | + mm->def_flags; + + ret = -ENOMEM; + if (ub_memory_charge(mm, PAGE_SIZE, flags, NULL, UB_SOFT)) + goto err_charge; down_write(&mm->mmap_sem); @@ -328,19 +348,18 @@ int arch_setup_additional_pages(struct l map_compat_vdso(compat); - if (compat) - addr = VDSO_HIGH_BASE; - else { - addr = get_unmapped_area(NULL, 0, PAGE_SIZE, 0, 0); + if (!compat || map_address) { + addr = get_unmapped_area(NULL, addr, PAGE_SIZE, 0, MAP_EXECUTABLE); if (IS_ERR_VALUE(addr)) { ret = addr; goto up_fail; } - } + } else + addr = VDSO_HIGH_BASE; - current->mm->context.vdso = (void *)addr; + current->mm->context.vdso = addr; - if (compat_uses_vma || !compat) { + if (compat_uses_vma || !compat || map_address) { /* * MAYWRITE to allow gdb to COW and set breakpoints * @@ -361,16 +380,20 @@ int arch_setup_additional_pages(struct l } current_thread_info()->sysenter_return = - VDSO32_SYMBOL(addr, SYSENTER_RETURN); + (__force void __user *)VDSO32_SYMBOL(addr, SYSENTER_RETURN); up_fail: if (ret) - current->mm->context.vdso = NULL; + current->mm->context.vdso = 0; up_write(&mm->mmap_sem); + if (ret < 0) + ub_memory_uncharge(mm, PAGE_SIZE, flags, NULL); +err_charge: return ret; } +EXPORT_SYMBOL_GPL(arch_setup_additional_pages); #ifdef CONFIG_X86_64 @@ -413,8 +436,14 @@ __initcall(ia32_binfmt_init); const char *arch_vma_name(struct vm_area_struct *vma) { - if (vma->vm_mm && vma->vm_start == (long)vma->vm_mm->context.vdso) + if (vma->vm_mm && vma->vm_start == vma->vm_mm->context.vdso) return "[vdso]"; + +#ifdef CONFIG_PAX_SEGMEXEC + if (vma->vm_mm && vma->vm_mirror && vma->vm_mirror->vm_start == vma->vm_mm->context.vdso) + return "[vdso]"; +#endif + return NULL; } @@ -423,7 +452,7 @@ struct vm_area_struct *get_gate_vma(stru struct mm_struct *mm = tsk->mm; /* Check to see if this task was created in compat vdso mode */ - if (mm && mm->context.vdso == (void *)VDSO_HIGH_BASE) + if (mm && mm->context.vdso == VDSO_HIGH_BASE) return &gate_vma; return NULL; } diff -urNp linux-2.6.32.48/arch/x86/vdso/vdso.lds.S linux-2.6.32.48-openvz/arch/x86/vdso/vdso.lds.S --- linux-2.6.32.48/arch/x86/vdso/vdso.lds.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/vdso/vdso.lds.S 2011-11-17 20:08:53.000000000 -0500 @@ -35,3 +35,9 @@ VDSO64_PRELINK = VDSO_PRELINK; #define VEXTERN(x) VDSO64_ ## x = vdso_ ## x; #include "vextern.h" #undef VEXTERN + +#define VEXTERN(x) VDSO64_ ## x = __vdso_ ## x; +VEXTERN(fallback_gettimeofday) +VEXTERN(fallback_time) +VEXTERN(getcpu) +#undef VEXTERN diff -urNp linux-2.6.32.48/arch/x86/vdso/vextern.h linux-2.6.32.48-openvz/arch/x86/vdso/vextern.h --- linux-2.6.32.48/arch/x86/vdso/vextern.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/vdso/vextern.h 2011-11-17 20:08:53.000000000 -0500 @@ -11,6 +11,5 @@ put into vextern.h and be referenced as a pointer with vdso prefix. The main kernel later fills in the values. */ -VEXTERN(jiffies) VEXTERN(vgetcpu_mode) VEXTERN(vsyscall_gtod_data) diff -urNp linux-2.6.32.48/arch/x86/vdso/vma.c linux-2.6.32.48-openvz/arch/x86/vdso/vma.c --- linux-2.6.32.48/arch/x86/vdso/vma.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/vdso/vma.c 2011-11-17 21:09:26.000000000 -0500 @@ -4,6 +4,7 @@ * Subject to the GPL, v.2 */ #include +#include #include #include #include @@ -27,10 +28,8 @@ static unsigned vdso_size; static inline void *var_ref(void *p, char *name) { - if (*(void **)p != (void *)VMAGIC) { - printk("VDSO: variable %s broken\n", name); - vdso_enabled = 0; - } + if (*(void **)p != (void *)VMAGIC) + panic("VDSO: variable %s broken\n", name); return p; } @@ -57,21 +56,18 @@ static int __init init_vdso_vars(void) if (!vbase) goto oom; - if (memcmp(vbase, "\177ELF", 4)) { - printk("VDSO: I'm broken; not ELF\n"); - vdso_enabled = 0; - } + if (memcmp(vbase, ELFMAG, SELFMAG)) + panic("VDSO: I'm broken; not ELF\n"); #define VEXTERN(x) \ *(typeof(__ ## x) **) var_ref(VDSO64_SYMBOL(vbase, x), #x) = &__ ## x; #include "vextern.h" #undef VEXTERN + vunmap(vbase); return 0; oom: - printk("Cannot allocate vdso\n"); - vdso_enabled = 0; - return -ENOMEM; + panic("Cannot allocate vdso\n"); } __initcall(init_vdso_vars); @@ -99,24 +95,30 @@ static unsigned long vdso_addr(unsigned /* Setup a VMA at program startup for the vsyscall page. Not called for compat tasks */ -int arch_setup_additional_pages(struct linux_binprm *bprm, int uses_interp) +int arch_setup_additional_pages(struct linux_binprm *bprm, int uses_interp, + unsigned long map_address) { struct mm_struct *mm = current->mm; unsigned long addr; int ret; - if (!vdso_enabled) + if (!vdso_enabled && map_address == 0) { + current->mm->context.vdso = 0; return 0; + } down_write(&mm->mmap_sem); - addr = vdso_addr(mm->start_stack, vdso_size); + if (map_address) + addr = map_address; + else + addr = vdso_addr(mm->start_stack, vdso_size); addr = get_unmapped_area(NULL, addr, vdso_size, 0, 0); if (IS_ERR_VALUE(addr)) { ret = addr; goto up_fail; } - current->mm->context.vdso = (void *)addr; + current->mm->context.vdso = addr; ret = install_special_mapping(mm, addr, vdso_size, VM_READ|VM_EXEC| @@ -124,7 +126,7 @@ int arch_setup_additional_pages(struct l VM_ALWAYSDUMP, vdso_pages); if (ret) { - current->mm->context.vdso = NULL; + current->mm->context.vdso = 0; goto up_fail; } @@ -132,10 +134,4 @@ up_fail: up_write(&mm->mmap_sem); return ret; } - -static __init int vdso_setup(char *s) -{ - vdso_enabled = simple_strtoul(s, NULL, 0); - return 0; -} -__setup("vdso=", vdso_setup); +EXPORT_SYMBOL_GPL(arch_setup_additional_pages); diff -urNp linux-2.6.32.48/arch/x86/xen/enlighten.c linux-2.6.32.48-openvz/arch/x86/xen/enlighten.c --- linux-2.6.32.48/arch/x86/xen/enlighten.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/xen/enlighten.c 2011-11-17 20:08:53.000000000 -0500 @@ -71,8 +71,6 @@ EXPORT_SYMBOL_GPL(xen_start_info); struct shared_info xen_dummy_shared_info; -void *xen_initial_gdt; - /* * Point at some empty memory to start with. We map the real shared_info * page as soon as fixmap is up and running. @@ -548,7 +546,7 @@ static void xen_write_idt_entry(gate_des preempt_disable(); - start = __get_cpu_var(idt_desc).address; + start = (unsigned long)__get_cpu_var(idt_desc).address; end = start + __get_cpu_var(idt_desc).size + 1; xen_mc_flush(); @@ -1095,9 +1093,20 @@ asmlinkage void __init xen_start_kernel( */ __userpte_alloc_gfp &= ~__GFP_HIGHMEM; -#ifdef CONFIG_X86_64 /* Work out if we support NX */ - check_efer(); +#if defined(CONFIG_X86_64) || defined(CONFIG_X86_PAE) + if ((cpuid_eax(0x80000000) & 0xffff0000) == 0x80000000 && + (cpuid_edx(0x80000001) & (1U << (X86_FEATURE_NX & 31)))) { + unsigned l, h; + +#ifdef CONFIG_X86_PAE + nx_enabled = 1; +#endif + __supported_pte_mask |= _PAGE_NX; + rdmsr(MSR_EFER, l, h); + l |= EFER_NX; + wrmsr(MSR_EFER, l, h); + } #endif xen_setup_features(); @@ -1129,13 +1138,6 @@ asmlinkage void __init xen_start_kernel( machine_ops = xen_machine_ops; - /* - * The only reliable way to retain the initial address of the - * percpu gdt_page is to remember it here, so we can go and - * mark it RW later, when the initial percpu area is freed. - */ - xen_initial_gdt = &per_cpu(gdt_page, 0); - xen_smp_init(); pgd = (pgd_t *)xen_start_info->pt_base; diff -urNp linux-2.6.32.48/arch/x86/xen/mmu.c linux-2.6.32.48-openvz/arch/x86/xen/mmu.c --- linux-2.6.32.48/arch/x86/xen/mmu.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/xen/mmu.c 2011-11-17 20:09:04.000000000 -0500 @@ -1719,6 +1719,8 @@ __init pgd_t *xen_setup_kernel_pagetable convert_pfn_mfn(init_level4_pgt); convert_pfn_mfn(level3_ident_pgt); convert_pfn_mfn(level3_kernel_pgt); + convert_pfn_mfn(level3_vmalloc_pgt); + convert_pfn_mfn(level3_vmemmap_pgt); l3 = m2v(pgd[pgd_index(__START_KERNEL_map)].pgd); l2 = m2v(l3[pud_index(__START_KERNEL_map)].pud); @@ -1737,7 +1739,10 @@ __init pgd_t *xen_setup_kernel_pagetable set_page_prot(init_level4_pgt, PAGE_KERNEL_RO); set_page_prot(level3_ident_pgt, PAGE_KERNEL_RO); set_page_prot(level3_kernel_pgt, PAGE_KERNEL_RO); + set_page_prot(level3_vmalloc_pgt, PAGE_KERNEL_RO); + set_page_prot(level3_vmemmap_pgt, PAGE_KERNEL_RO); set_page_prot(level3_user_vsyscall, PAGE_KERNEL_RO); + set_page_prot(level2_vmemmap_pgt, PAGE_KERNEL_RO); set_page_prot(level2_kernel_pgt, PAGE_KERNEL_RO); set_page_prot(level2_fixmap_pgt, PAGE_KERNEL_RO); @@ -1860,6 +1865,7 @@ static __init void xen_post_allocator_in pv_mmu_ops.set_pud = xen_set_pud; #if PAGETABLE_LEVELS == 4 pv_mmu_ops.set_pgd = xen_set_pgd; + pv_mmu_ops.set_pgd_batched = xen_set_pgd; #endif /* This will work as long as patching hasn't happened yet @@ -1946,6 +1952,7 @@ static const struct pv_mmu_ops xen_mmu_o .pud_val = PV_CALLEE_SAVE(xen_pud_val), .make_pud = PV_CALLEE_SAVE(xen_make_pud), .set_pgd = xen_set_pgd_hyper, + .set_pgd_batched = xen_set_pgd_hyper, .alloc_pud = xen_alloc_pmd_init, .release_pud = xen_release_pmd_init, diff -urNp linux-2.6.32.48/arch/x86/xen/smp.c linux-2.6.32.48-openvz/arch/x86/xen/smp.c --- linux-2.6.32.48/arch/x86/xen/smp.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/xen/smp.c 2011-11-17 20:09:04.000000000 -0500 @@ -168,11 +168,6 @@ static void __init xen_smp_prepare_boot_ { BUG_ON(smp_processor_id() != 0); native_smp_prepare_boot_cpu(); - - /* We've switched to the "real" per-cpu gdt, so make sure the - old memory can be recycled */ - make_lowmem_page_readwrite(xen_initial_gdt); - xen_setup_vcpu_info_placement(); } @@ -241,12 +236,12 @@ cpu_initialize_context(unsigned int cpu, gdt = get_cpu_gdt_table(cpu); ctxt->flags = VGCF_IN_KERNEL; - ctxt->user_regs.ds = __USER_DS; - ctxt->user_regs.es = __USER_DS; + ctxt->user_regs.ds = __KERNEL_DS; + ctxt->user_regs.es = __KERNEL_DS; ctxt->user_regs.ss = __KERNEL_DS; #ifdef CONFIG_X86_32 ctxt->user_regs.fs = __KERNEL_PERCPU; - ctxt->user_regs.gs = __KERNEL_STACK_CANARY; + savesegment(gs, ctxt->user_regs.gs); #else ctxt->gs_base_kernel = per_cpu_offset(cpu); #endif @@ -297,13 +292,12 @@ static int __cpuinit xen_cpu_up(unsigned int rc; per_cpu(current_task, cpu) = idle; + per_cpu(current_tinfo, cpu) = &idle->tinfo; #ifdef CONFIG_X86_32 irq_ctx_init(cpu); #else clear_tsk_thread_flag(idle, TIF_FORK); - per_cpu(kernel_stack, cpu) = - (unsigned long)task_stack_page(idle) - - KERNEL_STACK_OFFSET + THREAD_SIZE; + per_cpu(kernel_stack, cpu) = (unsigned long)task_stack_page(idle) - 16 + THREAD_SIZE; #endif xen_setup_runstate_info(cpu); xen_setup_timer(cpu); diff -urNp linux-2.6.32.48/arch/x86/xen/xen-asm_32.S linux-2.6.32.48-openvz/arch/x86/xen/xen-asm_32.S --- linux-2.6.32.48/arch/x86/xen/xen-asm_32.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/xen/xen-asm_32.S 2011-11-17 20:09:04.000000000 -0500 @@ -83,14 +83,14 @@ ENTRY(xen_iret) ESP_OFFSET=4 # bytes pushed onto stack /* - * Store vcpu_info pointer for easy access. Do it this way to - * avoid having to reload %fs + * Store vcpu_info pointer for easy access. */ #ifdef CONFIG_SMP - GET_THREAD_INFO(%eax) - movl TI_cpu(%eax), %eax - movl __per_cpu_offset(,%eax,4), %eax - mov per_cpu__xen_vcpu(%eax), %eax + push %fs + mov $(__KERNEL_PERCPU), %eax + mov %eax, %fs + mov PER_CPU_VAR(xen_vcpu), %eax + pop %fs #else movl per_cpu__xen_vcpu, %eax #endif diff -urNp linux-2.6.32.48/arch/x86/xen/xen-head.S linux-2.6.32.48-openvz/arch/x86/xen/xen-head.S --- linux-2.6.32.48/arch/x86/xen/xen-head.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/xen/xen-head.S 2011-11-17 20:08:53.000000000 -0500 @@ -19,6 +19,17 @@ ENTRY(startup_xen) #ifdef CONFIG_X86_32 mov %esi,xen_start_info mov $init_thread_union+THREAD_SIZE,%esp +#ifdef CONFIG_SMP + movl $cpu_gdt_table,%edi + movl $__per_cpu_load,%eax + movw %ax,__KERNEL_PERCPU + 2(%edi) + rorl $16,%eax + movb %al,__KERNEL_PERCPU + 4(%edi) + movb %ah,__KERNEL_PERCPU + 7(%edi) + movl $__per_cpu_end - 1,%eax + subl $__per_cpu_start,%eax + movw %ax,__KERNEL_PERCPU + 0(%edi) +#endif #else mov %rsi,xen_start_info mov $init_thread_union+THREAD_SIZE,%rsp diff -urNp linux-2.6.32.48/arch/x86/xen/xen-ops.h linux-2.6.32.48-openvz/arch/x86/xen/xen-ops.h --- linux-2.6.32.48/arch/x86/xen/xen-ops.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/arch/x86/xen/xen-ops.h 2011-11-17 20:08:53.000000000 -0500 @@ -10,8 +10,6 @@ extern const char xen_hypervisor_callback[]; extern const char xen_failsafe_callback[]; -extern void *xen_initial_gdt; - struct trap_info; void xen_copy_trap_info(struct trap_info *traps); diff -urNp linux-2.6.32.48/block/blk-cgroup.c linux-2.6.32.48-openvz/block/blk-cgroup.c --- linux-2.6.32.48/block/blk-cgroup.c 1969-12-31 19:00:00.000000000 -0500 +++ linux-2.6.32.48-openvz/block/blk-cgroup.c 2011-11-17 20:08:53.000000000 -0500 @@ -0,0 +1,366 @@ +/* + * Common Block IO controller cgroup interface + * + * Based on ideas and code from CFQ, CFS and BFQ: + * Copyright (C) 2003 Jens Axboe + * + * Copyright (C) 2008 Fabio Checconi + * Paolo Valente + * + * Copyright (C) 2009 Vivek Goyal + * Nauman Rafique + */ +#include +#include +#include +#include +#include +#include "blk-cgroup.h" + +static DEFINE_SPINLOCK(blkio_list_lock); +static LIST_HEAD(blkio_list); + +struct blkio_cgroup blkio_root_cgroup = { .weight = 2*BLKIO_WEIGHT_DEFAULT }; +EXPORT_SYMBOL_GPL(blkio_root_cgroup); + +bool blkiocg_css_tryget(struct blkio_cgroup *blkcg) +{ + if (!css_tryget(&blkcg->css)) + return false; + return true; +} +EXPORT_SYMBOL_GPL(blkiocg_css_tryget); + +void blkiocg_css_put(struct blkio_cgroup *blkcg) +{ + css_put(&blkcg->css); +} +EXPORT_SYMBOL_GPL(blkiocg_css_put); + +struct blkio_cgroup *cgroup_to_blkio_cgroup(struct cgroup *cgroup) +{ + return container_of(cgroup_subsys_state(cgroup, blkio_subsys_id), + struct blkio_cgroup, css); +} +EXPORT_SYMBOL_GPL(cgroup_to_blkio_cgroup); + +void blkiocg_update_blkio_group_stats(struct blkio_group *blkg, + unsigned long time, unsigned long sectors) +{ + blkg->time += time; + blkg->sectors += sectors; +} +EXPORT_SYMBOL_GPL(blkiocg_update_blkio_group_stats); + +void blkiocg_add_blkio_group(struct blkio_cgroup *blkcg, + struct blkio_group *blkg, void *key, dev_t dev) +{ + unsigned long flags; + + spin_lock_irqsave(&blkcg->lock, flags); + rcu_assign_pointer(blkg->key, key); + blkg->blkcg_id = css_id(&blkcg->css); + hlist_add_head_rcu(&blkg->blkcg_node, &blkcg->blkg_list); + spin_unlock_irqrestore(&blkcg->lock, flags); +#ifdef CONFIG_DEBUG_BLK_CGROUP + /* Need to take css reference ? */ + cgroup_path(blkcg->css.cgroup, blkg->path, sizeof(blkg->path)); +#endif + blkg->dev = dev; +} +EXPORT_SYMBOL_GPL(blkiocg_add_blkio_group); + +static void __blkiocg_del_blkio_group(struct blkio_group *blkg) +{ + hlist_del_init_rcu(&blkg->blkcg_node); + blkg->blkcg_id = 0; +} + +/* + * returns 0 if blkio_group was still on cgroup list. Otherwise returns 1 + * indicating that blk_group was unhashed by the time we got to it. + */ +int blkiocg_del_blkio_group(struct blkio_group *blkg) +{ + struct blkio_cgroup *blkcg; + unsigned long flags; + struct cgroup_subsys_state *css; + int ret = 1; + + rcu_read_lock(); + css = css_lookup(&blkio_subsys, blkg->blkcg_id); + if (!css) + goto out; + + blkcg = container_of(css, struct blkio_cgroup, css); + spin_lock_irqsave(&blkcg->lock, flags); + if (!hlist_unhashed(&blkg->blkcg_node)) { + __blkiocg_del_blkio_group(blkg); + ret = 0; + } + spin_unlock_irqrestore(&blkcg->lock, flags); +out: + rcu_read_unlock(); + return ret; +} +EXPORT_SYMBOL_GPL(blkiocg_del_blkio_group); + +/* called under rcu_read_lock(). */ +struct blkio_group *blkiocg_lookup_group(struct blkio_cgroup *blkcg, void *key) +{ + struct blkio_group *blkg; + struct hlist_node *n; + void *__key; + + hlist_for_each_entry_rcu(blkg, n, &blkcg->blkg_list, blkcg_node) { + __key = blkg->key; + if (__key == key) + return blkg; + } + + return NULL; +} +EXPORT_SYMBOL_GPL(blkiocg_lookup_group); + +#define SHOW_FUNCTION(__VAR) \ +static u64 blkiocg_##__VAR##_read(struct cgroup *cgroup, \ + struct cftype *cftype) \ +{ \ + struct blkio_cgroup *blkcg; \ + \ + blkcg = cgroup_to_blkio_cgroup(cgroup); \ + return (u64)blkcg->__VAR; \ +} + +SHOW_FUNCTION(weight); +#undef SHOW_FUNCTION + +static int +blkiocg_weight_write(struct cgroup *cgroup, struct cftype *cftype, u64 val) +{ + struct blkio_cgroup *blkcg; + struct blkio_group *blkg; + struct hlist_node *n; + struct blkio_policy_type *blkiop; + + if (val < BLKIO_WEIGHT_MIN || val > BLKIO_WEIGHT_MAX) + return -EINVAL; + + blkcg = cgroup_to_blkio_cgroup(cgroup); + spin_lock(&blkio_list_lock); + spin_lock_irq(&blkcg->lock); + blkcg->weight = (unsigned int)val; + hlist_for_each_entry(blkg, n, &blkcg->blkg_list, blkcg_node) { + list_for_each_entry(blkiop, &blkio_list, list) + blkiop->ops.blkio_update_group_weight_fn(blkg, + blkcg->weight); + } + spin_unlock_irq(&blkcg->lock); + spin_unlock(&blkio_list_lock); + return 0; +} + +int blkiocg_set_weight(struct cgroup *cgroup, u64 val) +{ + return blkiocg_weight_write(cgroup, NULL, val); +} + +#define SHOW_FUNCTION_PER_GROUP(__VAR) \ +static int blkiocg_##__VAR##_read(struct cgroup *cgroup, \ + struct cftype *cftype, struct seq_file *m) \ +{ \ + struct blkio_cgroup *blkcg; \ + struct blkio_group *blkg; \ + struct hlist_node *n; \ + \ + if (!cgroup_lock_live_group(cgroup)) \ + return -ENODEV; \ + \ + blkcg = cgroup_to_blkio_cgroup(cgroup); \ + rcu_read_lock(); \ + hlist_for_each_entry_rcu(blkg, n, &blkcg->blkg_list, blkcg_node) {\ + if (blkg->dev) \ + seq_printf(m, "%u:%u %lu\n", MAJOR(blkg->dev), \ + MINOR(blkg->dev), blkg->__VAR); \ + } \ + rcu_read_unlock(); \ + cgroup_unlock(); \ + return 0; \ +} + +SHOW_FUNCTION_PER_GROUP(time); +SHOW_FUNCTION_PER_GROUP(sectors); +#ifdef CONFIG_DEBUG_BLK_CGROUP +SHOW_FUNCTION_PER_GROUP(dequeue); +#endif +#undef SHOW_FUNCTION_PER_GROUP + +#ifdef CONFIG_DEBUG_BLK_CGROUP +void blkiocg_update_blkio_group_dequeue_stats(struct blkio_group *blkg, + unsigned long dequeue) +{ + blkg->dequeue += dequeue; +} +EXPORT_SYMBOL_GPL(blkiocg_update_blkio_group_dequeue_stats); +#endif + +struct cftype blkio_files[] = { + { + .name = "weight", + .read_u64 = blkiocg_weight_read, + .write_u64 = blkiocg_weight_write, + }, + { + .name = "time", + .read_seq_string = blkiocg_time_read, + }, + { + .name = "sectors", + .read_seq_string = blkiocg_sectors_read, + }, +#ifdef CONFIG_DEBUG_BLK_CGROUP + { + .name = "dequeue", + .read_seq_string = blkiocg_dequeue_read, + }, +#endif +}; + +static int blkiocg_populate(struct cgroup_subsys *subsys, struct cgroup *cgroup) +{ + return cgroup_add_files(cgroup, subsys, blkio_files, + ARRAY_SIZE(blkio_files)); +} + +static void blkiocg_destroy(struct cgroup_subsys *subsys, struct cgroup *cgroup) +{ + struct blkio_cgroup *blkcg = cgroup_to_blkio_cgroup(cgroup); + unsigned long flags; + struct blkio_group *blkg; + void *key; + struct blkio_policy_type *blkiop; + + rcu_read_lock(); +remove_entry: + spin_lock_irqsave(&blkcg->lock, flags); + + if (hlist_empty(&blkcg->blkg_list)) { + spin_unlock_irqrestore(&blkcg->lock, flags); + goto done; + } + + blkg = hlist_entry(blkcg->blkg_list.first, struct blkio_group, + blkcg_node); + key = rcu_dereference(blkg->key); + __blkiocg_del_blkio_group(blkg); + + spin_unlock_irqrestore(&blkcg->lock, flags); + + /* + * This blkio_group is being unlinked as associated cgroup is going + * away. Let all the IO controlling policies know about this event. + * + * Currently this is static call to one io controlling policy. Once + * we have more policies in place, we need some dynamic registration + * of callback function. + */ + spin_lock(&blkio_list_lock); + list_for_each_entry(blkiop, &blkio_list, list) + blkiop->ops.blkio_unlink_group_fn(key, blkg); + spin_unlock(&blkio_list_lock); + goto remove_entry; +done: + free_css_id(&blkio_subsys, &blkcg->css); + rcu_read_unlock(); + kfree(blkcg); +} + +static struct cgroup_subsys_state * +blkiocg_create(struct cgroup_subsys *subsys, struct cgroup *cgroup) +{ + struct blkio_cgroup *blkcg, *parent_blkcg; + + if (!cgroup->parent) { + blkcg = &blkio_root_cgroup; + goto done; + } + + /* Currently we do not support hierarchy deeper than two level (0,1) */ + parent_blkcg = cgroup_to_blkio_cgroup(cgroup->parent); + if (css_depth(&parent_blkcg->css) > 0) + return ERR_PTR(-EINVAL); + + blkcg = kzalloc(sizeof(*blkcg), GFP_KERNEL); + if (!blkcg) + return ERR_PTR(-ENOMEM); + + blkcg->weight = BLKIO_WEIGHT_DEFAULT; +done: + spin_lock_init(&blkcg->lock); + INIT_HLIST_HEAD(&blkcg->blkg_list); + + return &blkcg->css; +} + +/* + * We cannot support shared io contexts, as we have no mean to support + * two tasks with the same ioc in two different groups without major rework + * of the main cic data structures. For now we allow a task to change + * its cgroup only if it's the only owner of its ioc. + */ +static int blkiocg_can_attach(struct cgroup_subsys *subsys, + struct cgroup *cgroup, struct task_struct *tsk, + bool threadgroup) +{ + struct io_context *ioc; + int ret = 0; + + /* task_lock() is needed to avoid races with exit_io_context() */ + task_lock(tsk); + ioc = tsk->io_context; + if (ioc && atomic_read(&ioc->nr_tasks) > 1) + ret = -EINVAL; + task_unlock(tsk); + + return ret; +} + +static void blkiocg_attach(struct cgroup_subsys *subsys, struct cgroup *cgroup, + struct cgroup *prev, struct task_struct *tsk, + bool threadgroup) +{ + struct io_context *ioc; + + task_lock(tsk); + ioc = tsk->io_context; + if (ioc) + ioc->cgroup_changed = 1; + task_unlock(tsk); +} + +struct cgroup_subsys blkio_subsys = { + .name = "blkio", + .create = blkiocg_create, + .can_attach = blkiocg_can_attach, + .attach = blkiocg_attach, + .destroy = blkiocg_destroy, + .populate = blkiocg_populate, + .subsys_id = blkio_subsys_id, + .use_id = 1, +}; + +void blkio_policy_register(struct blkio_policy_type *blkiop) +{ + spin_lock(&blkio_list_lock); + list_add_tail(&blkiop->list, &blkio_list); + spin_unlock(&blkio_list_lock); +} +EXPORT_SYMBOL_GPL(blkio_policy_register); + +void blkio_policy_unregister(struct blkio_policy_type *blkiop) +{ + spin_lock(&blkio_list_lock); + list_del_init(&blkiop->list); + spin_unlock(&blkio_list_lock); +} +EXPORT_SYMBOL_GPL(blkio_policy_unregister); diff -urNp linux-2.6.32.48/block/blk-cgroup.h linux-2.6.32.48-openvz/block/blk-cgroup.h --- linux-2.6.32.48/block/blk-cgroup.h 1969-12-31 19:00:00.000000000 -0500 +++ linux-2.6.32.48-openvz/block/blk-cgroup.h 2011-11-17 20:08:53.000000000 -0500 @@ -0,0 +1,127 @@ +#ifndef _BLK_CGROUP_H +#define _BLK_CGROUP_H +/* + * Common Block IO controller cgroup interface + * + * Based on ideas and code from CFQ, CFS and BFQ: + * Copyright (C) 2003 Jens Axboe + * + * Copyright (C) 2008 Fabio Checconi + * Paolo Valente + * + * Copyright (C) 2009 Vivek Goyal + * Nauman Rafique + */ + +#include + +#ifdef CONFIG_BLK_CGROUP + +struct blkio_cgroup { + struct cgroup_subsys_state css; + unsigned int weight; + spinlock_t lock; + struct hlist_head blkg_list; +}; + +struct blkio_group { + /* An rcu protected unique identifier for the group */ + void *key; + struct hlist_node blkcg_node; + unsigned short blkcg_id; +#ifdef CONFIG_DEBUG_BLK_CGROUP + /* Store cgroup path */ + char path[128]; + /* How many times this group has been removed from service tree */ + unsigned long dequeue; +#endif + /* The device MKDEV(major, minor), this group has been created for */ + dev_t dev; + + /* total disk time and nr sectors dispatched by this group */ + unsigned long time; + unsigned long sectors; +}; + +extern bool blkiocg_css_tryget(struct blkio_cgroup *blkcg); +extern void blkiocg_css_put(struct blkio_cgroup *blkcg); + +typedef void (blkio_unlink_group_fn) (void *key, struct blkio_group *blkg); +typedef void (blkio_update_group_weight_fn) (struct blkio_group *blkg, + unsigned int weight); + +struct blkio_policy_ops { + blkio_unlink_group_fn *blkio_unlink_group_fn; + blkio_update_group_weight_fn *blkio_update_group_weight_fn; +}; + +struct blkio_policy_type { + struct list_head list; + struct blkio_policy_ops ops; +}; + +/* Blkio controller policy registration */ +extern void blkio_policy_register(struct blkio_policy_type *); +extern void blkio_policy_unregister(struct blkio_policy_type *); + +#else + +struct blkio_group { +}; + +struct blkio_policy_type { +}; + +static inline void blkio_policy_register(struct blkio_policy_type *blkiop) { } +static inline void blkio_policy_unregister(struct blkio_policy_type *blkiop) { } + +#endif + +#define BLKIO_WEIGHT_MIN 100 +#define BLKIO_WEIGHT_MAX 1000 +#define BLKIO_WEIGHT_DEFAULT 500 + +#ifdef CONFIG_DEBUG_BLK_CGROUP +static inline char *blkg_path(struct blkio_group *blkg) +{ + return blkg->path; +} +void blkiocg_update_blkio_group_dequeue_stats(struct blkio_group *blkg, + unsigned long dequeue); +#else +static inline char *blkg_path(struct blkio_group *blkg) { return NULL; } +static inline void blkiocg_update_blkio_group_dequeue_stats( + struct blkio_group *blkg, unsigned long dequeue) {} +#endif + +#ifdef CONFIG_BLK_CGROUP +extern struct blkio_cgroup blkio_root_cgroup; +extern struct blkio_cgroup *cgroup_to_blkio_cgroup(struct cgroup *cgroup); +extern void blkiocg_add_blkio_group(struct blkio_cgroup *blkcg, + struct blkio_group *blkg, void *key, dev_t dev); +extern int blkiocg_del_blkio_group(struct blkio_group *blkg); +extern struct blkio_group *blkiocg_lookup_group(struct blkio_cgroup *blkcg, + void *key); +void blkiocg_update_blkio_group_stats(struct blkio_group *blkg, + unsigned long time, unsigned long sectors); +#else +struct cgroup; +static inline struct blkio_cgroup * +cgroup_to_blkio_cgroup(struct cgroup *cgroup) { return NULL; } + +static inline void blkiocg_add_blkio_group(struct blkio_cgroup *blkcg, + struct blkio_group *blkg, void *key, dev_t dev) +{ +} + +static inline int +blkiocg_del_blkio_group(struct blkio_group *blkg) { return 0; } + +static inline struct blkio_group * +blkiocg_lookup_group(struct blkio_cgroup *blkcg, void *key) { return NULL; } +static inline void blkiocg_update_blkio_group_stats(struct blkio_group *blkg, + unsigned long time, unsigned long sectors) +{ +} +#endif +#endif /* _BLK_CGROUP_H */ diff -urNp linux-2.6.32.48/block/blk-integrity.c linux-2.6.32.48-openvz/block/blk-integrity.c --- linux-2.6.32.48/block/blk-integrity.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/block/blk-integrity.c 2011-11-17 20:08:53.000000000 -0500 @@ -278,7 +278,7 @@ static struct attribute *integrity_attrs NULL, }; -static struct sysfs_ops integrity_ops = { +static const struct sysfs_ops integrity_ops = { .show = &integrity_attr_show, .store = &integrity_attr_store, }; diff -urNp linux-2.6.32.48/block/blk-iopoll.c linux-2.6.32.48-openvz/block/blk-iopoll.c --- linux-2.6.32.48/block/blk-iopoll.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/block/blk-iopoll.c 2011-11-17 20:08:53.000000000 -0500 @@ -77,7 +77,7 @@ void blk_iopoll_complete(struct blk_iopo } EXPORT_SYMBOL(blk_iopoll_complete); -static void blk_iopoll_softirq(struct softirq_action *h) +static void blk_iopoll_softirq(void) { struct list_head *list = &__get_cpu_var(blk_cpu_iopoll); int rearm = 0, budget = blk_iopoll_budget; diff -urNp linux-2.6.32.48/block/blk-map.c linux-2.6.32.48-openvz/block/blk-map.c --- linux-2.6.32.48/block/blk-map.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/block/blk-map.c 2011-11-17 20:08:53.000000000 -0500 @@ -54,7 +54,7 @@ static int __blk_rq_map_user(struct requ * direct dma. else, set up kernel bounce buffers */ uaddr = (unsigned long) ubuf; - if (blk_rq_aligned(q, ubuf, len) && !map_data) + if (blk_rq_aligned(q, (__force void *)ubuf, len) && !map_data) bio = bio_map_user(q, NULL, uaddr, len, reading, gfp_mask); else bio = bio_copy_user(q, map_data, uaddr, len, reading, gfp_mask); @@ -201,12 +201,13 @@ int blk_rq_map_user_iov(struct request_q for (i = 0; i < iov_count; i++) { unsigned long uaddr = (unsigned long)iov[i].iov_base; + if (!iov[i].iov_len) + return -EINVAL; + if (uaddr & queue_dma_alignment(q)) { unaligned = 1; break; } - if (!iov[i].iov_len) - return -EINVAL; } if (unaligned || (q->dma_pad_mask & len) || map_data) @@ -299,7 +300,7 @@ int blk_rq_map_kern(struct request_queue if (!len || !kbuf) return -EINVAL; - do_copy = !blk_rq_aligned(q, kbuf, len) || object_is_on_stack(kbuf); + do_copy = !blk_rq_aligned(q, kbuf, len) || object_starts_on_stack(kbuf); if (do_copy) bio = bio_copy_kern(q, kbuf, len, gfp_mask, reading); else diff -urNp linux-2.6.32.48/block/blk-settings.c linux-2.6.32.48-openvz/block/blk-settings.c --- linux-2.6.32.48/block/blk-settings.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/block/blk-settings.c 2011-11-17 20:08:53.000000000 -0500 @@ -9,6 +9,7 @@ #include /* for max_pfn/max_low_pfn */ #include #include +#include #include "blk.h" @@ -142,7 +143,7 @@ void blk_queue_make_request(struct reque q->nr_batching = BLK_BATCH_REQ; q->unplug_thresh = 4; /* hmm */ - q->unplug_delay = (3 * HZ) / 1000; /* 3 milliseconds */ + q->unplug_delay = msecs_to_jiffies(3); /* 3 milliseconds */ if (q->unplug_delay == 0) q->unplug_delay = 1; diff -urNp linux-2.6.32.48/block/blk-softirq.c linux-2.6.32.48-openvz/block/blk-softirq.c --- linux-2.6.32.48/block/blk-softirq.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/block/blk-softirq.c 2011-11-17 20:08:53.000000000 -0500 @@ -17,7 +17,7 @@ static DEFINE_PER_CPU(struct list_head, * Softirq action handler - move entries to local list and loop over them * while passing them to the queue registered handler. */ -static void blk_done_softirq(struct softirq_action *h) +static void blk_done_softirq(void) { struct list_head *cpu_list, local_list; diff -urNp linux-2.6.32.48/block/blk-sysfs.c linux-2.6.32.48-openvz/block/blk-sysfs.c --- linux-2.6.32.48/block/blk-sysfs.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/block/blk-sysfs.c 2011-11-17 20:08:53.000000000 -0500 @@ -414,7 +414,7 @@ static void blk_release_queue(struct kob kmem_cache_free(blk_requestq_cachep, q); } -static struct sysfs_ops queue_sysfs_ops = { +static const struct sysfs_ops queue_sysfs_ops = { .show = queue_attr_show, .store = queue_attr_store, }; diff -urNp linux-2.6.32.48/block/bsg.c linux-2.6.32.48-openvz/block/bsg.c --- linux-2.6.32.48/block/bsg.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/block/bsg.c 2011-11-17 20:09:04.000000000 -0500 @@ -15,6 +15,7 @@ #include #include #include +#include #include #include #include @@ -175,16 +176,24 @@ static int blk_fill_sgv4_hdr_rq(struct r struct sg_io_v4 *hdr, struct bsg_device *bd, fmode_t has_write_perm) { + unsigned char tmpcmd[sizeof(rq->__cmd)]; + unsigned char *cmdptr; + if (hdr->request_len > BLK_MAX_CDB) { rq->cmd = kzalloc(hdr->request_len, GFP_KERNEL); if (!rq->cmd) return -ENOMEM; - } + cmdptr = rq->cmd; + } else + cmdptr = tmpcmd; - if (copy_from_user(rq->cmd, (void *)(unsigned long)hdr->request, + if (copy_from_user(cmdptr, (void __user *)(unsigned long)hdr->request, hdr->request_len)) return -EFAULT; + if (cmdptr != rq->cmd) + memcpy(rq->cmd, cmdptr, hdr->request_len); + if (hdr->subprotocol == BSG_SUB_PROTOCOL_SCSI_CMD) { if (blk_verify_command(rq->cmd, has_write_perm)) return -EPERM; @@ -197,7 +206,7 @@ static int blk_fill_sgv4_hdr_rq(struct r rq->cmd_len = hdr->request_len; rq->cmd_type = REQ_TYPE_BLOCK_PC; - rq->timeout = (hdr->timeout * HZ) / 1000; + rq->timeout = msecs_to_jiffies(hdr->timeout); if (!rq->timeout) rq->timeout = q->sg_timeout; if (!rq->timeout) @@ -282,7 +291,7 @@ bsg_map_hdr(struct bsg_device *bd, struc rq->next_rq = next_rq; next_rq->cmd_type = rq->cmd_type; - dxferp = (void*)(unsigned long)hdr->din_xferp; + dxferp = (void __user *)(unsigned long)hdr->din_xferp; ret = blk_rq_map_user(q, next_rq, NULL, dxferp, hdr->din_xfer_len, GFP_KERNEL); if (ret) @@ -291,10 +300,10 @@ bsg_map_hdr(struct bsg_device *bd, struc if (hdr->dout_xfer_len) { dxfer_len = hdr->dout_xfer_len; - dxferp = (void*)(unsigned long)hdr->dout_xferp; + dxferp = (void __user *)(unsigned long)hdr->dout_xferp; } else if (hdr->din_xfer_len) { dxfer_len = hdr->din_xfer_len; - dxferp = (void*)(unsigned long)hdr->din_xferp; + dxferp = (void __user *)(unsigned long)hdr->din_xferp; } else dxfer_len = 0; @@ -436,7 +445,7 @@ static int blk_complete_sgv4_hdr_rq(stru int len = min_t(unsigned int, hdr->max_response_len, rq->sense_len); - ret = copy_to_user((void*)(unsigned long)hdr->response, + ret = copy_to_user((void __user *)(unsigned long)hdr->response, rq->sense, len); if (!ret) hdr->response_len = len; diff -urNp linux-2.6.32.48/block/cfq-iosched.c linux-2.6.32.48-openvz/block/cfq-iosched.c --- linux-2.6.32.48/block/cfq-iosched.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/block/cfq-iosched.c 2011-11-17 20:58:22.000000000 -0500 @@ -9,9 +9,11 @@ #include #include #include +#include #include #include #include +#include "blk-cgroup.h" /* * tunables @@ -27,6 +29,8 @@ static const int cfq_slice_sync = HZ / 1 static int cfq_slice_async = HZ / 25; static const int cfq_slice_async_rq = 2; static int cfq_slice_idle = HZ / 125; +static const int cfq_target_latency = HZ * 3/10; /* 300 ms */ +static const int cfq_hist_divisor = 4; /* * offset from end of service tree @@ -38,14 +42,12 @@ static int cfq_slice_idle = HZ / 125; */ #define CFQ_MIN_TT (2) -/* - * Allow merged cfqqs to perform this amount of seeky I/O before - * deciding to break the queues up again. - */ -#define CFQQ_COOP_TOUT (HZ) - #define CFQ_SLICE_SCALE (5) #define CFQ_HW_QUEUE_MIN (5) +#define CFQ_SERVICE_SHIFT 12 + +#define CFQQ_SEEK_THR 8 * 1024 +#define CFQQ_SEEKY(cfqq) ((cfqq)->seek_mean > CFQQ_SEEK_THR) #define RQ_CIC(rq) \ ((struct cfq_io_context *) (rq)->elevator_private) @@ -63,6 +65,7 @@ static DEFINE_SPINLOCK(ioc_gone_lock); #define cfq_class_rt(cfqq) ((cfqq)->ioprio_class == IOPRIO_CLASS_RT) #define sample_valid(samples) ((samples) > 80) +#define rb_entry_cfqg(node) rb_entry((node), struct cfq_group, rb_node) /* * Most of our rbtree usage is for sorting with min extraction, so @@ -73,8 +76,12 @@ static DEFINE_SPINLOCK(ioc_gone_lock); struct cfq_rb_root { struct rb_root rb; struct rb_node *left; + unsigned count; + u64 min_vdisktime; + struct rb_node *active; + unsigned total_weight; }; -#define CFQ_RB_ROOT (struct cfq_rb_root) { RB_ROOT, NULL, } +#define CFQ_RB_ROOT (struct cfq_rb_root) { RB_ROOT, NULL, 0, 0, } /* * Per process-grouping structure @@ -105,6 +112,11 @@ struct cfq_queue { /* fifo list of requests in sort_list */ struct list_head fifo; + /* time when queue got scheduled in to dispatch first request. */ + unsigned long dispatch_start; + unsigned int allocated_slice; + /* time when first request from queue completed and slice started. */ + unsigned long slice_start; unsigned long slice_end; long slice_resid; unsigned int slice_dispatch; @@ -122,11 +134,66 @@ struct cfq_queue { u64 seek_total; sector_t seek_mean; sector_t last_request_pos; - unsigned long seeky_start; pid_t pid; + struct cfq_rb_root *service_tree; struct cfq_queue *new_cfqq; + struct cfq_group *cfqg; + struct cfq_group *orig_cfqg; + /* Sectors dispatched in current dispatch round */ + unsigned long nr_sectors; +}; + +/* + * First index in the service_trees. + * IDLE is handled separately, so it has negative index + */ +enum wl_prio_t { + BE_WORKLOAD = 0, + RT_WORKLOAD = 1, + IDLE_WORKLOAD = 2, +}; + +/* + * Second index in the service_trees. + */ +enum wl_type_t { + ASYNC_WORKLOAD = 0, + SYNC_NOIDLE_WORKLOAD = 1, + SYNC_WORKLOAD = 2 +}; + +/* This is per cgroup per device grouping structure */ +struct cfq_group { + /* group service_tree member */ + struct rb_node rb_node; + + /* group service_tree key */ + u64 vdisktime; + unsigned int weight; + bool on_st; + + /* number of cfqq currently on this group */ + int nr_cfqq; + + /* Per group busy queus average. Useful for workload slice calc. */ + unsigned int busy_queues_avg[2]; + /* + * rr lists of queues with requests, onle rr for each priority class. + * Counts are embedded in the cfq_rb_root + */ + struct cfq_rb_root service_trees[2][3]; + struct cfq_rb_root service_tree_idle; + + unsigned long saved_workload_slice; + enum wl_type_t saved_workload; + enum wl_prio_t saved_serving_prio; + struct blkio_group blkg; +#ifdef CONFIG_CFQ_GROUP_IOSCHED + struct hlist_node cfqd_node; + atomic_t ref; +#endif }; /* @@ -134,11 +201,18 @@ struct cfq_queue { */ struct cfq_data { struct request_queue *queue; + /* Root service tree for cfq_groups */ + struct cfq_rb_root grp_service_tree; + struct cfq_group root_group; /* - * rr list of queues with requests and the count of them - */ - struct cfq_rb_root service_tree; + * The priority currently being served + */ + enum wl_prio_t serving_prio; + enum wl_type_t serving_type; + unsigned long workload_expires; + struct cfq_group *serving_group; + bool noidle_tree_requires_idle; /* * Each priority tree is sorted by next_request position. These @@ -157,8 +231,14 @@ struct cfq_data { */ int rq_queued; int hw_tag; - int hw_tag_samples; - int rq_in_driver_peak; + /* + * hw_tag can be + * -1 => indeterminate, (cfq will behave as if NCQ is present, to allow better detection) + * 1 => NCQ is present (hw_tag_est_depth is the estimated max depth) + * 0 => no NCQ + */ + int hw_tag_est_depth; + unsigned int hw_tag_samples; /* * idle window management @@ -188,6 +268,7 @@ struct cfq_data { unsigned int cfq_slice_async_rq; unsigned int cfq_slice_idle; unsigned int cfq_latency; + unsigned int cfq_group_isolation; struct list_head cic_list; @@ -196,9 +277,28 @@ struct cfq_data { */ struct cfq_queue oom_cfqq; - unsigned long last_end_sync_rq; + unsigned long last_delayed_sync; + + /* List of cfq groups being managed on this device*/ + struct hlist_head cfqg_list; + struct rcu_head rcu; }; +static struct cfq_group *cfq_get_next_cfqg(struct cfq_data *cfqd); + +static struct cfq_rb_root *service_tree_for(struct cfq_group *cfqg, + enum wl_prio_t prio, + enum wl_type_t type) +{ + if (!cfqg) + return NULL; + + if (prio == IDLE_WORKLOAD) + return &cfqg->service_tree_idle; + + return &cfqg->service_trees[prio][type]; +} + enum cfqq_state_flags { CFQ_CFQQ_FLAG_on_rr = 0, /* on round-robin busy list */ CFQ_CFQQ_FLAG_wait_request, /* waiting for a request */ @@ -210,6 +310,9 @@ enum cfqq_state_flags { CFQ_CFQQ_FLAG_slice_new, /* no requests dispatched in slice */ CFQ_CFQQ_FLAG_sync, /* synchronous queue */ CFQ_CFQQ_FLAG_coop, /* cfqq is shared */ + CFQ_CFQQ_FLAG_split_coop, /* shared cfqq will be splitted */ + CFQ_CFQQ_FLAG_deep, /* sync cfqq experienced large depth */ + CFQ_CFQQ_FLAG_wait_busy, /* Waiting for next request */ }; #define CFQ_CFQQ_FNS(name) \ @@ -236,13 +339,78 @@ CFQ_CFQQ_FNS(prio_changed); CFQ_CFQQ_FNS(slice_new); CFQ_CFQQ_FNS(sync); CFQ_CFQQ_FNS(coop); +CFQ_CFQQ_FNS(split_coop); +CFQ_CFQQ_FNS(deep); +CFQ_CFQQ_FNS(wait_busy); #undef CFQ_CFQQ_FNS +#ifdef CONFIG_DEBUG_CFQ_IOSCHED +#define cfq_log_cfqq(cfqd, cfqq, fmt, args...) \ + blk_add_trace_msg((cfqd)->queue, "cfq%d%c %s " fmt, (cfqq)->pid, \ + cfq_cfqq_sync((cfqq)) ? 'S' : 'A', \ + blkg_path(&(cfqq)->cfqg->blkg), ##args); + +#define cfq_log_cfqg(cfqd, cfqg, fmt, args...) \ + blk_add_trace_msg((cfqd)->queue, "%s " fmt, \ + blkg_path(&(cfqg)->blkg), ##args); \ + +#else #define cfq_log_cfqq(cfqd, cfqq, fmt, args...) \ blk_add_trace_msg((cfqd)->queue, "cfq%d " fmt, (cfqq)->pid, ##args) +#define cfq_log_cfqg(cfqd, cfqg, fmt, args...) do {} while (0); +#endif #define cfq_log(cfqd, fmt, args...) \ blk_add_trace_msg((cfqd)->queue, "cfq " fmt, ##args) +/* Traverses through cfq group service trees */ +#define for_each_cfqg_st(cfqg, i, j, st) \ + for (i = 0; i <= IDLE_WORKLOAD; i++) \ + for (j = 0, st = i < IDLE_WORKLOAD ? &cfqg->service_trees[i][j]\ + : &cfqg->service_tree_idle; \ + (i < IDLE_WORKLOAD && j <= SYNC_WORKLOAD) || \ + (i == IDLE_WORKLOAD && j == 0); \ + j++, st = i < IDLE_WORKLOAD ? \ + &cfqg->service_trees[i][j]: NULL) \ + + +static inline enum wl_prio_t cfqq_prio(struct cfq_queue *cfqq) +{ + if (cfq_class_idle(cfqq)) + return IDLE_WORKLOAD; + if (cfq_class_rt(cfqq)) + return RT_WORKLOAD; + return BE_WORKLOAD; +} + + +static enum wl_type_t cfqq_type(struct cfq_queue *cfqq) +{ + if (!cfq_cfqq_sync(cfqq)) + return ASYNC_WORKLOAD; + if (!cfq_cfqq_idle_window(cfqq)) + return SYNC_NOIDLE_WORKLOAD; + return SYNC_WORKLOAD; +} + +static inline int cfq_group_busy_queues_wl(enum wl_prio_t wl, + struct cfq_data *cfqd, + struct cfq_group *cfqg) +{ + if (wl == IDLE_WORKLOAD) + return cfqg->service_tree_idle.count; + + return cfqg->service_trees[wl][ASYNC_WORKLOAD].count + + cfqg->service_trees[wl][SYNC_NOIDLE_WORKLOAD].count + + cfqg->service_trees[wl][SYNC_WORKLOAD].count; +} + +static inline int cfqg_busy_async_queues(struct cfq_data *cfqd, + struct cfq_group *cfqg) +{ + return cfqg->service_trees[RT_WORKLOAD][ASYNC_WORKLOAD].count + + cfqg->service_trees[BE_WORKLOAD][ASYNC_WORKLOAD].count; +} + static void cfq_dispatch_insert(struct request_queue *, struct request *); static struct cfq_queue *cfq_get_queue(struct cfq_data *, bool, struct io_context *, gfp_t); @@ -291,7 +459,7 @@ static int cfq_queue_empty(struct reques { struct cfq_data *cfqd = q->elevator->elevator_data; - return !cfqd->busy_queues; + return !cfqd->rq_queued; } /* @@ -315,10 +483,110 @@ cfq_prio_to_slice(struct cfq_data *cfqd, return cfq_prio_slice(cfqd, cfq_cfqq_sync(cfqq), cfqq->ioprio); } +static inline u64 cfq_scale_slice(unsigned long delta, struct cfq_group *cfqg) +{ + u64 d = delta << CFQ_SERVICE_SHIFT; + + d = d * BLKIO_WEIGHT_DEFAULT; + do_div(d, cfqg->weight); + return d; +} + +static inline u64 max_vdisktime(u64 min_vdisktime, u64 vdisktime) +{ + s64 delta = (s64)(vdisktime - min_vdisktime); + if (delta > 0) + min_vdisktime = vdisktime; + + return min_vdisktime; +} + +static inline u64 min_vdisktime(u64 min_vdisktime, u64 vdisktime) +{ + s64 delta = (s64)(vdisktime - min_vdisktime); + if (delta < 0) + min_vdisktime = vdisktime; + + return min_vdisktime; +} + +static void update_min_vdisktime(struct cfq_rb_root *st) +{ + u64 vdisktime = st->min_vdisktime; + struct cfq_group *cfqg; + + if (st->active) { + cfqg = rb_entry_cfqg(st->active); + vdisktime = cfqg->vdisktime; + } + + if (st->left) { + cfqg = rb_entry_cfqg(st->left); + vdisktime = min_vdisktime(vdisktime, cfqg->vdisktime); + } + + st->min_vdisktime = max_vdisktime(st->min_vdisktime, vdisktime); +} + +/* + * get averaged number of queues of RT/BE priority. + * average is updated, with a formula that gives more weight to higher numbers, + * to quickly follows sudden increases and decrease slowly + */ + +static inline unsigned cfq_group_get_avg_queues(struct cfq_data *cfqd, + struct cfq_group *cfqg, bool rt) +{ + unsigned min_q, max_q; + unsigned mult = cfq_hist_divisor - 1; + unsigned round = cfq_hist_divisor / 2; + unsigned busy = cfq_group_busy_queues_wl(rt, cfqd, cfqg); + + min_q = min(cfqg->busy_queues_avg[rt], busy); + max_q = max(cfqg->busy_queues_avg[rt], busy); + cfqg->busy_queues_avg[rt] = (mult * max_q + min_q + round) / + cfq_hist_divisor; + return cfqg->busy_queues_avg[rt]; +} + +static inline unsigned +cfq_group_slice(struct cfq_data *cfqd, struct cfq_group *cfqg) +{ + struct cfq_rb_root *st = &cfqd->grp_service_tree; + + return cfq_target_latency * cfqg->weight / st->total_weight; +} + static inline void cfq_set_prio_slice(struct cfq_data *cfqd, struct cfq_queue *cfqq) { - cfqq->slice_end = cfq_prio_to_slice(cfqd, cfqq) + jiffies; + unsigned slice = cfq_prio_to_slice(cfqd, cfqq); + if (cfqd->cfq_latency) { + /* + * interested queues (we consider only the ones with the same + * priority class in the cfq group) + */ + unsigned iq = cfq_group_get_avg_queues(cfqd, cfqq->cfqg, + cfq_class_rt(cfqq)); + unsigned sync_slice = cfqd->cfq_slice[1]; + unsigned expect_latency = sync_slice * iq; + unsigned group_slice = cfq_group_slice(cfqd, cfqq->cfqg); + + if (expect_latency > group_slice) { + unsigned base_low_slice = 2 * cfqd->cfq_slice_idle; + /* scale low_slice according to IO priority + * and sync vs async */ + unsigned low_slice = + min(slice, base_low_slice * slice / sync_slice); + /* the adapted slice value is scaled to fit all iqs + * into the target latency */ + slice = max(slice * group_slice / expect_latency, + low_slice); + } + } + cfqq->slice_start = jiffies; + cfqq->slice_end = jiffies + slice; + cfqq->allocated_slice = slice; cfq_log_cfqq(cfqd, cfqq, "set_slice=%lu", cfqq->slice_end - jiffies); } @@ -343,9 +611,9 @@ static inline bool cfq_slice_used(struct * behind the head is penalized and only allowed to a certain extent. */ static struct request * -cfq_choose_req(struct cfq_data *cfqd, struct request *rq1, struct request *rq2) +cfq_choose_req(struct cfq_data *cfqd, struct request *rq1, struct request *rq2, sector_t last) { - sector_t last, s1, s2, d1 = 0, d2 = 0; + sector_t s1, s2, d1 = 0, d2 = 0; unsigned long back_max; #define CFQ_RQ1_WRAP 0x01 /* request 1 wraps */ #define CFQ_RQ2_WRAP 0x02 /* request 2 wraps */ @@ -368,8 +636,6 @@ cfq_choose_req(struct cfq_data *cfqd, st s1 = blk_rq_pos(rq1); s2 = blk_rq_pos(rq2); - last = cfqd->last_position; - /* * by definition, 1KiB is 2 sectors */ @@ -437,6 +703,10 @@ cfq_choose_req(struct cfq_data *cfqd, st */ static struct cfq_queue *cfq_rb_first(struct cfq_rb_root *root) { + /* Service tree is empty */ + if (!root->count) + return NULL; + if (!root->left) root->left = rb_first(&root->rb); @@ -446,6 +716,17 @@ static struct cfq_queue *cfq_rb_first(st return NULL; } +static struct cfq_group *cfq_rb_first_group(struct cfq_rb_root *root) +{ + if (!root->left) + root->left = rb_first(&root->rb); + + if (root->left) + return rb_entry_cfqg(root->left); + + return NULL; +} + static void rb_erase_init(struct rb_node *n, struct rb_root *root) { rb_erase(n, root); @@ -457,6 +738,7 @@ static void cfq_rb_erase(struct rb_node if (root->left == n) root->left = NULL; rb_erase_init(n, &root->rb); + --root->count; } /* @@ -483,7 +765,7 @@ cfq_find_next_rq(struct cfq_data *cfqd, next = rb_entry_rq(rbnext); } - return cfq_choose_req(cfqd, next, prev); + return cfq_choose_req(cfqd, next, prev, blk_rq_pos(last)); } static unsigned long cfq_slice_offset(struct cfq_data *cfqd, @@ -492,12 +774,334 @@ static unsigned long cfq_slice_offset(st /* * just an approximation, should be ok. */ - return (cfqd->busy_queues - 1) * (cfq_prio_slice(cfqd, 1, 0) - + return (cfqq->cfqg->nr_cfqq - 1) * (cfq_prio_slice(cfqd, 1, 0) - cfq_prio_slice(cfqd, cfq_cfqq_sync(cfqq), cfqq->ioprio)); } +static inline s64 +cfqg_key(struct cfq_rb_root *st, struct cfq_group *cfqg) +{ + return cfqg->vdisktime - st->min_vdisktime; +} + +static void +__cfq_group_service_tree_add(struct cfq_rb_root *st, struct cfq_group *cfqg) +{ + struct rb_node **node = &st->rb.rb_node; + struct rb_node *parent = NULL; + struct cfq_group *__cfqg; + s64 key = cfqg_key(st, cfqg); + int left = 1; + + while (*node != NULL) { + parent = *node; + __cfqg = rb_entry_cfqg(parent); + + if (key < cfqg_key(st, __cfqg)) + node = &parent->rb_left; + else { + node = &parent->rb_right; + left = 0; + } + } + + if (left) + st->left = &cfqg->rb_node; + + rb_link_node(&cfqg->rb_node, parent, node); + rb_insert_color(&cfqg->rb_node, &st->rb); +} + +static void +cfq_group_service_tree_add(struct cfq_data *cfqd, struct cfq_group *cfqg) +{ + struct cfq_rb_root *st = &cfqd->grp_service_tree; + struct cfq_group *__cfqg; + struct rb_node *n; + + cfqg->nr_cfqq++; + if (cfqg->on_st) + return; + + /* + * Currently put the group at the end. Later implement something + * so that groups get lesser vtime based on their weights, so that + * if group does not loose all if it was not continously backlogged. + */ + n = rb_last(&st->rb); + if (n) { + __cfqg = rb_entry_cfqg(n); + cfqg->vdisktime = __cfqg->vdisktime + CFQ_IDLE_DELAY; + } else + cfqg->vdisktime = st->min_vdisktime; + + __cfq_group_service_tree_add(st, cfqg); + cfqg->on_st = true; + st->total_weight += cfqg->weight; +} + +static void +cfq_group_service_tree_del(struct cfq_data *cfqd, struct cfq_group *cfqg) +{ + struct cfq_rb_root *st = &cfqd->grp_service_tree; + + if (st->active == &cfqg->rb_node) + st->active = NULL; + + BUG_ON(cfqg->nr_cfqq < 1); + cfqg->nr_cfqq--; + + /* If there are other cfq queues under this group, don't delete it */ + if (cfqg->nr_cfqq) + return; + + cfq_log_cfqg(cfqd, cfqg, "del_from_rr group"); + cfqg->on_st = false; + st->total_weight -= cfqg->weight; + if (!RB_EMPTY_NODE(&cfqg->rb_node)) + cfq_rb_erase(&cfqg->rb_node, st); + cfqg->saved_workload_slice = 0; + blkiocg_update_blkio_group_dequeue_stats(&cfqg->blkg, 1); +} + +static inline unsigned int cfq_cfqq_slice_usage(struct cfq_queue *cfqq) +{ + unsigned int slice_used; + + /* + * Queue got expired before even a single request completed or + * got expired immediately after first request completion. + */ + if (!cfqq->slice_start || cfqq->slice_start == jiffies) { + /* + * Also charge the seek time incurred to the group, otherwise + * if there are mutiple queues in the group, each can dispatch + * a single request on seeky media and cause lots of seek time + * and group will never know it. + */ + slice_used = max_t(unsigned, (jiffies - cfqq->dispatch_start), + 1); + } else { + slice_used = jiffies - cfqq->slice_start; + if (slice_used > cfqq->allocated_slice) + slice_used = cfqq->allocated_slice; + } + + cfq_log_cfqq(cfqq->cfqd, cfqq, "sl_used=%u sect=%lu", slice_used, + cfqq->nr_sectors); + return slice_used; +} + +static void cfq_group_served(struct cfq_data *cfqd, struct cfq_group *cfqg, + struct cfq_queue *cfqq) +{ + struct cfq_rb_root *st = &cfqd->grp_service_tree; + unsigned int used_sl, charge_sl; + int nr_sync = cfqg->nr_cfqq - cfqg_busy_async_queues(cfqd, cfqg) + - cfqg->service_tree_idle.count; + + BUG_ON(nr_sync < 0); + used_sl = charge_sl = cfq_cfqq_slice_usage(cfqq); + + if (!cfq_cfqq_sync(cfqq) && !nr_sync) + charge_sl = cfqq->allocated_slice; + + /* Can't update vdisktime while group is on service tree */ + cfq_rb_erase(&cfqg->rb_node, st); + cfqg->vdisktime += cfq_scale_slice(charge_sl, cfqg); + __cfq_group_service_tree_add(st, cfqg); + + /* This group is being expired. Save the context */ + if (time_after(cfqd->workload_expires, jiffies)) { + cfqg->saved_workload_slice = cfqd->workload_expires + - jiffies; + cfqg->saved_workload = cfqd->serving_type; + cfqg->saved_serving_prio = cfqd->serving_prio; + } else + cfqg->saved_workload_slice = 0; + + cfq_log_cfqg(cfqd, cfqg, "served: vt=%llu min_vt=%llu", cfqg->vdisktime, + st->min_vdisktime); + blkiocg_update_blkio_group_stats(&cfqg->blkg, used_sl, + cfqq->nr_sectors); +} + +#ifdef CONFIG_CFQ_GROUP_IOSCHED +static inline struct cfq_group *cfqg_of_blkg(struct blkio_group *blkg) +{ + if (blkg) + return container_of(blkg, struct cfq_group, blkg); + return NULL; +} + +void +cfq_update_blkio_group_weight(struct blkio_group *blkg, unsigned int weight) +{ + cfqg_of_blkg(blkg)->weight = weight; +} + +static struct cfq_group * +cfq_find_alloc_cfqg(struct cfq_data *cfqd, struct cgroup *cgroup, int create) +{ + struct blkio_cgroup *blkcg = cgroup_to_blkio_cgroup(cgroup); + struct cfq_group *cfqg = NULL; + void *key = cfqd; + int i, j; + struct cfq_rb_root *st; + struct backing_dev_info *bdi = &cfqd->queue->backing_dev_info; + unsigned int major, minor; + + /* Do we need to take this reference */ + if (!blkiocg_css_tryget(blkcg)) + return NULL;; + + cfqg = cfqg_of_blkg(blkiocg_lookup_group(blkcg, key)); + if (cfqg || !create) + goto done; + + cfqg = kzalloc_node(sizeof(*cfqg), GFP_ATOMIC, cfqd->queue->node); + if (!cfqg) + goto done; + + cfqg->weight = blkcg->weight; + for_each_cfqg_st(cfqg, i, j, st) + *st = CFQ_RB_ROOT; + RB_CLEAR_NODE(&cfqg->rb_node); + + /* + * Take the initial reference that will be released on destroy + * This can be thought of a joint reference by cgroup and + * elevator which will be dropped by either elevator exit + * or cgroup deletion path depending on who is exiting first. + */ + atomic_set(&cfqg->ref, 1); + + /* Add group onto cgroup list */ + sscanf(dev_name(bdi->dev), "%u:%u", &major, &minor); + blkiocg_add_blkio_group(blkcg, &cfqg->blkg, (void *)cfqd, + MKDEV(major, minor)); + + /* Add group on cfqd list */ + hlist_add_head(&cfqg->cfqd_node, &cfqd->cfqg_list); + +done: + blkiocg_css_put(blkcg); + return cfqg; +} + +/* + * Search for the cfq group current task belongs to. If create = 1, then also + * create the cfq group if it does not exist. request_queue lock must be held. + */ +static struct cfq_group *cfq_get_cfqg(struct cfq_data *cfqd, int create) +{ + struct cgroup *cgroup; + struct cfq_group *cfqg = NULL; + + rcu_read_lock(); + cgroup = task_cgroup(current, blkio_subsys_id); + cfqg = cfq_find_alloc_cfqg(cfqd, cgroup, create); + if (!cfqg && create) + cfqg = &cfqd->root_group; + rcu_read_unlock(); + return cfqg; +} + +static void cfq_link_cfqq_cfqg(struct cfq_queue *cfqq, struct cfq_group *cfqg) +{ + /* Currently, all async queues are mapped to root group */ + if (!cfq_cfqq_sync(cfqq)) + cfqg = &cfqq->cfqd->root_group; + + cfqq->cfqg = cfqg; + /* cfqq reference on cfqg */ + atomic_inc(&cfqq->cfqg->ref); +} + +static void cfq_put_cfqg(struct cfq_group *cfqg) +{ + struct cfq_rb_root *st; + int i, j; + + BUG_ON(atomic_read(&cfqg->ref) <= 0); + if (!atomic_dec_and_test(&cfqg->ref)) + return; + for_each_cfqg_st(cfqg, i, j, st) + BUG_ON(!RB_EMPTY_ROOT(&st->rb) || st->active != NULL); + kfree(cfqg); +} + +static void cfq_destroy_cfqg(struct cfq_data *cfqd, struct cfq_group *cfqg) +{ + /* Something wrong if we are trying to remove same group twice */ + BUG_ON(hlist_unhashed(&cfqg->cfqd_node)); + + hlist_del_init(&cfqg->cfqd_node); + + /* + * Put the reference taken at the time of creation so that when all + * queues are gone, group can be destroyed. + */ + cfq_put_cfqg(cfqg); +} + +static void cfq_release_cfq_groups(struct cfq_data *cfqd) +{ + struct hlist_node *pos, *n; + struct cfq_group *cfqg; + + hlist_for_each_entry_safe(cfqg, pos, n, &cfqd->cfqg_list, cfqd_node) { + /* + * If cgroup removal path got to blk_group first and removed + * it from cgroup list, then it will take care of destroying + * cfqg also. + */ + if (!blkiocg_del_blkio_group(&cfqg->blkg)) + cfq_destroy_cfqg(cfqd, cfqg); + } +} + /* - * The cfqd->service_tree holds all pending cfq_queue's that have + * Blk cgroup controller notification saying that blkio_group object is being + * delinked as associated cgroup object is going away. That also means that + * no new IO will come in this group. So get rid of this group as soon as + * any pending IO in the group is finished. + * + * This function is called under rcu_read_lock(). key is the rcu protected + * pointer. That means "key" is a valid cfq_data pointer as long as we are rcu + * read lock. + * + * "key" was fetched from blkio_group under blkio_cgroup->lock. That means + * it should not be NULL as even if elevator was exiting, cgroup deltion + * path got to it first. + */ +void cfq_unlink_blkio_group(void *key, struct blkio_group *blkg) +{ + unsigned long flags; + struct cfq_data *cfqd = key; + + spin_lock_irqsave(cfqd->queue->queue_lock, flags); + cfq_destroy_cfqg(cfqd, cfqg_of_blkg(blkg)); + spin_unlock_irqrestore(cfqd->queue->queue_lock, flags); +} + +#else /* GROUP_IOSCHED */ +static struct cfq_group *cfq_get_cfqg(struct cfq_data *cfqd, int create) +{ + return &cfqd->root_group; +} +static inline void +cfq_link_cfqq_cfqg(struct cfq_queue *cfqq, struct cfq_group *cfqg) { + cfqq->cfqg = cfqg; +} + +static void cfq_release_cfq_groups(struct cfq_data *cfqd) {} +static inline void cfq_put_cfqg(struct cfq_group *cfqg) {} + +#endif /* GROUP_IOSCHED */ + +/* + * The cfqd->service_trees holds all pending cfq_queue's that have * requests waiting to be processed. It is sorted in the order that * we will service the queues. */ @@ -507,11 +1111,42 @@ static void cfq_service_tree_add(struct struct rb_node **p, *parent; struct cfq_queue *__cfqq; unsigned long rb_key; + struct cfq_rb_root *service_tree; int left; + int new_cfqq = 1; + int group_changed = 0; + +#ifdef CONFIG_CFQ_GROUP_IOSCHED + if (!cfqd->cfq_group_isolation + && cfqq_type(cfqq) == SYNC_NOIDLE_WORKLOAD + && cfqq->cfqg && cfqq->cfqg != &cfqd->root_group) { + /* Move this cfq to root group */ + cfq_log_cfqq(cfqd, cfqq, "moving to root group"); + if (!RB_EMPTY_NODE(&cfqq->rb_node)) + cfq_group_service_tree_del(cfqd, cfqq->cfqg); + cfqq->orig_cfqg = cfqq->cfqg; + cfqq->cfqg = &cfqd->root_group; + atomic_inc(&cfqd->root_group.ref); + group_changed = 1; + } else if (!cfqd->cfq_group_isolation + && cfqq_type(cfqq) == SYNC_WORKLOAD && cfqq->orig_cfqg) { + /* cfqq is sequential now needs to go to its original group */ + BUG_ON(cfqq->cfqg != &cfqd->root_group); + if (!RB_EMPTY_NODE(&cfqq->rb_node)) + cfq_group_service_tree_del(cfqd, cfqq->cfqg); + cfq_put_cfqg(cfqq->cfqg); + cfqq->cfqg = cfqq->orig_cfqg; + cfqq->orig_cfqg = NULL; + group_changed = 1; + cfq_log_cfqq(cfqd, cfqq, "moved to origin group"); + } +#endif + service_tree = service_tree_for(cfqq->cfqg, cfqq_prio(cfqq), + cfqq_type(cfqq)); if (cfq_class_idle(cfqq)) { rb_key = CFQ_IDLE_DELAY; - parent = rb_last(&cfqd->service_tree.rb); + parent = rb_last(&service_tree->rb); if (parent && parent != &cfqq->rb_node) { __cfqq = rb_entry(parent, struct cfq_queue, rb_node); rb_key += __cfqq->rb_key; @@ -529,23 +1164,27 @@ static void cfq_service_tree_add(struct cfqq->slice_resid = 0; } else { rb_key = -HZ; - __cfqq = cfq_rb_first(&cfqd->service_tree); + __cfqq = cfq_rb_first(service_tree); rb_key += __cfqq ? __cfqq->rb_key : jiffies; } if (!RB_EMPTY_NODE(&cfqq->rb_node)) { + new_cfqq = 0; /* * same position, nothing more to do */ - if (rb_key == cfqq->rb_key) + if (rb_key == cfqq->rb_key && + cfqq->service_tree == service_tree) return; - cfq_rb_erase(&cfqq->rb_node, &cfqd->service_tree); + cfq_rb_erase(&cfqq->rb_node, cfqq->service_tree); + cfqq->service_tree = NULL; } left = 1; parent = NULL; - p = &cfqd->service_tree.rb.rb_node; + cfqq->service_tree = service_tree; + p = &service_tree->rb.rb_node; while (*p) { struct rb_node **n; @@ -553,35 +1192,28 @@ static void cfq_service_tree_add(struct __cfqq = rb_entry(parent, struct cfq_queue, rb_node); /* - * sort RT queues first, we always want to give - * preference to them. IDLE queues goes to the back. - * after that, sort on the next service time. + * sort by key, that represents service time. */ - if (cfq_class_rt(cfqq) > cfq_class_rt(__cfqq)) - n = &(*p)->rb_left; - else if (cfq_class_rt(cfqq) < cfq_class_rt(__cfqq)) - n = &(*p)->rb_right; - else if (cfq_class_idle(cfqq) < cfq_class_idle(__cfqq)) - n = &(*p)->rb_left; - else if (cfq_class_idle(cfqq) > cfq_class_idle(__cfqq)) - n = &(*p)->rb_right; - else if (time_before(rb_key, __cfqq->rb_key)) + if (time_before(rb_key, __cfqq->rb_key)) n = &(*p)->rb_left; - else + else { n = &(*p)->rb_right; - - if (n == &(*p)->rb_right) left = 0; + } p = n; } if (left) - cfqd->service_tree.left = &cfqq->rb_node; + service_tree->left = &cfqq->rb_node; cfqq->rb_key = rb_key; rb_link_node(&cfqq->rb_node, parent, p); - rb_insert_color(&cfqq->rb_node, &cfqd->service_tree.rb); + rb_insert_color(&cfqq->rb_node, &service_tree->rb); + service_tree->count++; + if ((add_front || !new_cfqq) && !group_changed) + return; + cfq_group_service_tree_add(cfqd, cfqq->cfqg); } static struct cfq_queue * @@ -683,13 +1315,16 @@ static void cfq_del_cfqq_rr(struct cfq_d BUG_ON(!cfq_cfqq_on_rr(cfqq)); cfq_clear_cfqq_on_rr(cfqq); - if (!RB_EMPTY_NODE(&cfqq->rb_node)) - cfq_rb_erase(&cfqq->rb_node, &cfqd->service_tree); + if (!RB_EMPTY_NODE(&cfqq->rb_node)) { + cfq_rb_erase(&cfqq->rb_node, cfqq->service_tree); + cfqq->service_tree = NULL; + } if (cfqq->p_root) { rb_erase(&cfqq->p_node, cfqq->p_root); cfqq->p_root = NULL; } + cfq_group_service_tree_del(cfqd, cfqq->cfqg); BUG_ON(!cfqd->busy_queues); cfqd->busy_queues--; } @@ -700,7 +1335,6 @@ static void cfq_del_cfqq_rr(struct cfq_d static void cfq_del_rq_rb(struct request *rq) { struct cfq_queue *cfqq = RQ_CFQQ(rq); - struct cfq_data *cfqd = cfqq->cfqd; const int sync = rq_is_sync(rq); BUG_ON(!cfqq->queued[sync]); @@ -708,8 +1342,17 @@ static void cfq_del_rq_rb(struct request elv_rb_del(&cfqq->sort_list, rq); - if (cfq_cfqq_on_rr(cfqq) && RB_EMPTY_ROOT(&cfqq->sort_list)) - cfq_del_cfqq_rr(cfqd, cfqq); + if (cfq_cfqq_on_rr(cfqq) && RB_EMPTY_ROOT(&cfqq->sort_list)) { + /* + * Queue will be deleted from service tree when we actually + * expire it later. Right now just remove it from prio tree + * as it is empty. + */ + if (cfqq->p_root) { + rb_erase(&cfqq->p_node, cfqq->p_root); + cfqq->p_root = NULL; + } + } } static void cfq_add_rq_rb(struct request *rq) @@ -734,7 +1377,7 @@ static void cfq_add_rq_rb(struct request * check if this request is a better next-serve candidate */ prev = cfqq->next_rq; - cfqq->next_rq = cfq_choose_req(cfqd, cfqq->next_rq, rq); + cfqq->next_rq = cfq_choose_req(cfqd, cfqq->next_rq, rq, cfqd->last_position); /* * adjust priority tree position, if ->next_rq changes @@ -841,6 +1484,7 @@ static void cfq_merged_requests(struct request_queue *q, struct request *rq, struct request *next) { + struct cfq_queue *cfqq = RQ_CFQQ(rq); /* * reposition in fifo if next is older than rq */ @@ -850,6 +1494,8 @@ cfq_merged_requests(struct request_queue rq_set_fifo_time(rq, rq_fifo_time(next)); } + if (cfqq->next_rq == next) + cfqq->next_rq = rq; cfq_remove_request(next); } @@ -883,8 +1529,12 @@ static void __cfq_set_active_queue(struc { if (cfqq) { cfq_log_cfqq(cfqd, cfqq, "set_active"); + cfqq->slice_start = 0; + cfqq->dispatch_start = jiffies; + cfqq->allocated_slice = 0; cfqq->slice_end = 0; cfqq->slice_dispatch = 0; + cfqq->nr_sectors = 0; cfq_clear_cfqq_wait_request(cfqq); cfq_clear_cfqq_must_dispatch(cfqq); @@ -911,6 +1561,16 @@ __cfq_slice_expired(struct cfq_data *cfq del_timer(&cfqd->idle_slice_timer); cfq_clear_cfqq_wait_request(cfqq); + cfq_clear_cfqq_wait_busy(cfqq); + + /* + * If this cfqq is shared between multiple processes, check to + * make sure that those processes are still issuing I/Os within + * the mean seek distance. If not, it may be time to break the + * queues apart again. + */ + if (cfq_cfqq_coop(cfqq) && CFQQ_SEEKY(cfqq)) + cfq_mark_cfqq_split_coop(cfqq); /* * store what was left of this slice, if the queue idled/timed out @@ -920,11 +1580,19 @@ __cfq_slice_expired(struct cfq_data *cfq cfq_log_cfqq(cfqd, cfqq, "resid=%ld", cfqq->slice_resid); } + cfq_group_served(cfqd, cfqq->cfqg, cfqq); + + if (cfq_cfqq_on_rr(cfqq) && RB_EMPTY_ROOT(&cfqq->sort_list)) + cfq_del_cfqq_rr(cfqd, cfqq); + cfq_resort_rr_list(cfqd, cfqq); if (cfqq == cfqd->active_queue) cfqd->active_queue = NULL; + if (&cfqq->cfqg->rb_node == cfqd->grp_service_tree.active) + cfqd->grp_service_tree.active = NULL; + if (cfqd->active_cic) { put_io_context(cfqd->active_cic->ioc); cfqd->active_cic = NULL; @@ -945,10 +1613,39 @@ static inline void cfq_slice_expired(str */ static struct cfq_queue *cfq_get_next_queue(struct cfq_data *cfqd) { - if (RB_EMPTY_ROOT(&cfqd->service_tree.rb)) + struct cfq_rb_root *service_tree = + service_tree_for(cfqd->serving_group, cfqd->serving_prio, + cfqd->serving_type); + + if (!cfqd->rq_queued) + return NULL; + + /* There is nothing to dispatch */ + if (!service_tree) + return NULL; + if (RB_EMPTY_ROOT(&service_tree->rb)) + return NULL; + return cfq_rb_first(service_tree); +} + +static struct cfq_queue *cfq_get_next_queue_forced(struct cfq_data *cfqd) +{ + struct cfq_group *cfqg; + struct cfq_queue *cfqq; + int i, j; + struct cfq_rb_root *st; + + if (!cfqd->rq_queued) + return NULL; + + cfqg = cfq_get_next_cfqg(cfqd); + if (!cfqg) return NULL; - return cfq_rb_first(&cfqd->service_tree); + for_each_cfqg_st(cfqg, i, j, st) + if ((cfqq = cfq_rb_first(st)) != NULL) + return cfqq; + return NULL; } /* @@ -973,17 +1670,18 @@ static inline sector_t cfq_dist_from_las return cfqd->last_position - blk_rq_pos(rq); } -#define CFQQ_SEEK_THR 8 * 1024 -#define CFQQ_SEEKY(cfqq) ((cfqq)->seek_mean > CFQQ_SEEK_THR) - static inline int cfq_rq_close(struct cfq_data *cfqd, struct cfq_queue *cfqq, - struct request *rq) + struct request *rq, bool for_preempt) { sector_t sdist = cfqq->seek_mean; if (!sample_valid(cfqq->seek_samples)) sdist = CFQQ_SEEK_THR; + /* if seek_mean is big, using it as close criteria is meaningless */ + if (sdist > CFQQ_SEEK_THR && !for_preempt) + sdist = CFQQ_SEEK_THR; + return cfq_dist_from_last(cfqd, rq) <= sdist; } @@ -1011,7 +1709,7 @@ static struct cfq_queue *cfqq_close(stru * will contain the closest sector. */ __cfqq = rb_entry(parent, struct cfq_queue, p_node); - if (cfq_rq_close(cfqd, cur_cfqq, __cfqq->next_rq)) + if (cfq_rq_close(cfqd, cur_cfqq, __cfqq->next_rq, false)) return __cfqq; if (blk_rq_pos(__cfqq->next_rq) < sector) @@ -1022,7 +1720,7 @@ static struct cfq_queue *cfqq_close(stru return NULL; __cfqq = rb_entry(node, struct cfq_queue, p_node); - if (cfq_rq_close(cfqd, cur_cfqq, __cfqq->next_rq)) + if (cfq_rq_close(cfqd, cur_cfqq, __cfqq->next_rq, false)) return __cfqq; return NULL; @@ -1049,6 +1747,12 @@ static struct cfq_queue *cfq_close_coope return NULL; /* + * Don't search priority tree if it's the only queue in the group. + */ + if (cur_cfqq->cfqg->nr_cfqq == 1) + return NULL; + + /* * We should notice if some of the queues are cooperating, eg * working closely on the same area of the disk. In that case, * we can group them together and don't waste time idling. @@ -1057,6 +1761,10 @@ static struct cfq_queue *cfq_close_coope if (!cfqq) return NULL; + /* If new queue belongs to different cfq_group, don't choose it */ + if (cur_cfqq->cfqg != cfqq->cfqg) + return NULL; + /* * It only makes sense to merge sync queues. */ @@ -1065,9 +1773,43 @@ static struct cfq_queue *cfq_close_coope if (CFQQ_SEEKY(cfqq)) return NULL; + /* + * Do not merge queues of different priority classes + */ + if (cfq_class_rt(cfqq) != cfq_class_rt(cur_cfqq)) + return NULL; + return cfqq; } +/* + * Determine whether we should enforce idle window for this queue. + */ + +static bool cfq_should_idle(struct cfq_data *cfqd, struct cfq_queue *cfqq) +{ + enum wl_prio_t prio = cfqq_prio(cfqq); + struct cfq_rb_root *service_tree = cfqq->service_tree; + + BUG_ON(!service_tree); + BUG_ON(!service_tree->count); + + /* We never do for idle class queues. */ + if (prio == IDLE_WORKLOAD) + return false; + + /* We do for queues that were marked with idle window flag. */ + if (cfq_cfqq_idle_window(cfqq) && + !(blk_queue_nonrot(cfqd->queue) && cfqd->hw_tag)) + return true; + + /* + * Otherwise, we do only if they are the last ones + * in their service tree. + */ + return service_tree->count == 1 && cfq_cfqq_sync(cfqq); +} + static void cfq_arm_slice_timer(struct cfq_data *cfqd) { struct cfq_queue *cfqq = cfqd->active_queue; @@ -1088,13 +1830,13 @@ static void cfq_arm_slice_timer(struct c /* * idle is disabled, either manually or by past process history */ - if (!cfqd->cfq_slice_idle || !cfq_cfqq_idle_window(cfqq)) + if (!cfqd->cfq_slice_idle || !cfq_should_idle(cfqd, cfqq)) return; /* - * still requests with the driver, don't idle + * still active requests from this queue, don't idle */ - if (rq_in_driver(cfqd)) + if (cfqq->dispatched) return; /* @@ -1115,14 +1857,7 @@ static void cfq_arm_slice_timer(struct c cfq_mark_cfqq_wait_request(cfqq); - /* - * we don't want to idle for seeks, but we do want to allow - * fair distribution of slice time for a process doing back-to-back - * seeks. so allow a little bit of time for him to submit a new rq - */ sl = cfqd->cfq_slice_idle; - if (sample_valid(cfqq->seek_samples) && CFQQ_SEEKY(cfqq)) - sl = min(sl, msecs_to_jiffies(CFQ_MIN_TT)); mod_timer(&cfqd->idle_slice_timer, jiffies + sl); cfq_log_cfqq(cfqd, cfqq, "arm_idle: %lu", sl); @@ -1145,6 +1880,7 @@ static void cfq_dispatch_insert(struct r if (cfq_cfqq_sync(cfqq)) cfqd->sync_flight++; + cfqq->nr_sectors += blk_rq_sectors(rq); } /* @@ -1198,15 +1934,6 @@ static void cfq_setup_merge(struct cfq_q int process_refs, new_process_refs; struct cfq_queue *__cfqq; - /* - * If there are no process references on the new_cfqq, then it is - * unsafe to follow the ->new_cfqq chain as other cfqq's in the - * chain may have dropped their last reference (not just their - * last process reference). - */ - if (!cfqq_process_refs(new_cfqq)) - return; - /* Avoid a circular list and skip interim queue merges */ while ((__cfqq = new_cfqq->new_cfqq)) { if (__cfqq == cfqq) @@ -1215,17 +1942,17 @@ static void cfq_setup_merge(struct cfq_q } process_refs = cfqq_process_refs(cfqq); - new_process_refs = cfqq_process_refs(new_cfqq); /* * If the process for the cfqq has gone away, there is no * sense in merging the queues. */ - if (process_refs == 0 || new_process_refs == 0) + if (process_refs == 0) return; /* * Merge in the direction of the lesser amount of work. */ + new_process_refs = cfqq_process_refs(new_cfqq); if (new_process_refs >= process_refs) { cfqq->new_cfqq = new_cfqq; atomic_add(process_refs, &new_cfqq->ref); @@ -1235,6 +1962,140 @@ static void cfq_setup_merge(struct cfq_q } } +static enum wl_type_t cfq_choose_wl(struct cfq_data *cfqd, + struct cfq_group *cfqg, enum wl_prio_t prio) +{ + struct cfq_queue *queue; + int i; + bool key_valid = false; + unsigned long lowest_key = 0; + enum wl_type_t cur_best = SYNC_NOIDLE_WORKLOAD; + + for (i = 0; i <= SYNC_WORKLOAD; ++i) { + /* select the one with lowest rb_key */ + queue = cfq_rb_first(service_tree_for(cfqg, prio, i)); + if (queue && + (!key_valid || time_before(queue->rb_key, lowest_key))) { + lowest_key = queue->rb_key; + cur_best = i; + key_valid = true; + } + } + + return cur_best; +} + +static void choose_service_tree(struct cfq_data *cfqd, struct cfq_group *cfqg) +{ + unsigned slice; + unsigned count; + struct cfq_rb_root *st; + unsigned group_slice; + + if (!cfqg) { + cfqd->serving_prio = IDLE_WORKLOAD; + cfqd->workload_expires = jiffies + 1; + return; + } + + /* Choose next priority. RT > BE > IDLE */ + if (cfq_group_busy_queues_wl(RT_WORKLOAD, cfqd, cfqg)) + cfqd->serving_prio = RT_WORKLOAD; + else if (cfq_group_busy_queues_wl(BE_WORKLOAD, cfqd, cfqg)) + cfqd->serving_prio = BE_WORKLOAD; + else { + cfqd->serving_prio = IDLE_WORKLOAD; + cfqd->workload_expires = jiffies + 1; + return; + } + + /* + * For RT and BE, we have to choose also the type + * (SYNC, SYNC_NOIDLE, ASYNC), and to compute a workload + * expiration time + */ + st = service_tree_for(cfqg, cfqd->serving_prio, cfqd->serving_type); + count = st->count; + + /* + * check workload expiration, and that we still have other queues ready + */ + if (count && !time_after(jiffies, cfqd->workload_expires)) + return; + + /* otherwise select new workload type */ + cfqd->serving_type = + cfq_choose_wl(cfqd, cfqg, cfqd->serving_prio); + st = service_tree_for(cfqg, cfqd->serving_prio, cfqd->serving_type); + count = st->count; + + /* + * the workload slice is computed as a fraction of target latency + * proportional to the number of queues in that workload, over + * all the queues in the same priority class + */ + group_slice = cfq_group_slice(cfqd, cfqg); + + slice = group_slice * count / + max_t(unsigned, cfqg->busy_queues_avg[cfqd->serving_prio], + cfq_group_busy_queues_wl(cfqd->serving_prio, cfqd, cfqg)); + + if (cfqd->serving_type == ASYNC_WORKLOAD) { + unsigned int tmp; + + /* + * Async queues are currently system wide. Just taking + * proportion of queues with-in same group will lead to higher + * async ratio system wide as generally root group is going + * to have higher weight. A more accurate thing would be to + * calculate system wide asnc/sync ratio. + */ + tmp = cfq_target_latency * cfqg_busy_async_queues(cfqd, cfqg); + tmp = tmp/cfqd->busy_queues; + slice = min_t(unsigned, slice, tmp); + + /* async workload slice is scaled down according to + * the sync/async slice ratio. */ + slice = slice * cfqd->cfq_slice[0] / cfqd->cfq_slice[1]; + } else + /* sync workload slice is at least 2 * cfq_slice_idle */ + slice = max(slice, 2 * cfqd->cfq_slice_idle); + + slice = max_t(unsigned, slice, CFQ_MIN_TT); + cfqd->workload_expires = jiffies + slice; + cfqd->noidle_tree_requires_idle = false; +} + +static struct cfq_group *cfq_get_next_cfqg(struct cfq_data *cfqd) +{ + struct cfq_rb_root *st = &cfqd->grp_service_tree; + struct cfq_group *cfqg; + + if (RB_EMPTY_ROOT(&st->rb)) + return NULL; + cfqg = cfq_rb_first_group(st); + st->active = &cfqg->rb_node; + update_min_vdisktime(st); + return cfqg; +} + +static void cfq_choose_cfqg(struct cfq_data *cfqd) +{ + struct cfq_group *cfqg = cfq_get_next_cfqg(cfqd); + + cfqd->serving_group = cfqg; + + /* Restore the workload type data */ + if (cfqg->saved_workload_slice) { + cfqd->workload_expires = jiffies + cfqg->saved_workload_slice; + cfqd->serving_type = cfqg->saved_workload; + cfqd->serving_prio = cfqg->saved_serving_prio; + } else + cfqd->workload_expires = jiffies - 1; + + choose_service_tree(cfqd, cfqg); +} + /* * Select a queue for service. If we have a current active queue, * check whether to continue servicing it, or retrieve and set a new one. @@ -1247,13 +2108,37 @@ static struct cfq_queue *cfq_select_queu if (!cfqq) goto new_queue; + if (!cfqd->rq_queued) + return NULL; + /* - * The active queue has run out of time, expire it and select new. + * We were waiting for group to get backlogged. Expire the queue */ - if (cfq_slice_used(cfqq) && !cfq_cfqq_must_dispatch(cfqq)) + if (cfq_cfqq_wait_busy(cfqq) && !RB_EMPTY_ROOT(&cfqq->sort_list)) goto expire; /* + * The active queue has run out of time, expire it and select new. + */ + if (cfq_slice_used(cfqq) && !cfq_cfqq_must_dispatch(cfqq)) { + /* + * If slice had not expired at the completion of last request + * we might not have turned on wait_busy flag. Don't expire + * the queue yet. Allow the group to get backlogged. + * + * The very fact that we have used the slice, that means we + * have been idling all along on this queue and it should be + * ok to wait for this request to complete. + */ + if (cfqq->cfqg->nr_cfqq == 1 && RB_EMPTY_ROOT(&cfqq->sort_list) + && cfqq->dispatched && cfq_should_idle(cfqd, cfqq)) { + cfqq = NULL; + goto keep_queue; + } else + goto expire; + } + + /* * The active queue has requests and isn't expired, allow it to * dispatch. */ @@ -1279,7 +2164,7 @@ static struct cfq_queue *cfq_select_queu * conditions to happen (or time out) before selecting a new queue. */ if (timer_pending(&cfqd->idle_slice_timer) || - (cfqq->dispatched && cfq_cfqq_idle_window(cfqq))) { + (cfqq->dispatched && cfq_should_idle(cfqd, cfqq))) { cfqq = NULL; goto keep_queue; } @@ -1287,6 +2172,13 @@ static struct cfq_queue *cfq_select_queu expire: cfq_slice_expired(cfqd, 0); new_queue: + /* + * Current queue expired. Check if we have to switch to a new + * service tree + */ + if (!new_cfqq) + cfq_choose_cfqg(cfqd); + cfqq = cfq_set_active_queue(cfqd, new_cfqq); keep_queue: return cfqq; @@ -1302,6 +2194,9 @@ static int __cfq_forced_dispatch_cfqq(st } BUG_ON(!list_empty(&cfqq->fifo)); + + /* By default cfqq is not expired if it is empty. Do it explicitly */ + __cfq_slice_expired(cfqq->cfqd, cfqq, 0); return dispatched; } @@ -1314,11 +2209,10 @@ static int cfq_forced_dispatch(struct cf struct cfq_queue *cfqq; int dispatched = 0; - while ((cfqq = cfq_rb_first(&cfqd->service_tree)) != NULL) + while ((cfqq = cfq_get_next_queue_forced(cfqd)) != NULL) dispatched += __cfq_forced_dispatch_cfqq(cfqq); cfq_slice_expired(cfqd, 0); - BUG_ON(cfqd->busy_queues); cfq_log(cfqd, "forced_dispatch=%d", dispatched); @@ -1332,7 +2226,7 @@ static bool cfq_may_dispatch(struct cfq_ /* * Drain async requests before we start sync IO */ - if (cfq_cfqq_idle_window(cfqq) && cfqd->rq_in_driver[BLK_RW_ASYNC]) + if (cfq_should_idle(cfqd, cfqq) && cfqd->rq_in_driver[BLK_RW_ASYNC]) return false; /* @@ -1362,9 +2256,9 @@ static bool cfq_may_dispatch(struct cfq_ return false; /* - * Sole queue user, allow bigger slice + * Sole queue user, no limit */ - max_dispatch *= 4; + max_dispatch = -1; } /* @@ -1373,7 +2267,7 @@ static bool cfq_may_dispatch(struct cfq_ * based on the last sync IO we serviced */ if (!cfq_cfqq_sync(cfqq) && cfqd->cfq_latency) { - unsigned long last_sync = jiffies - cfqd->last_end_sync_rq; + unsigned long last_sync = jiffies - cfqd->last_delayed_sync; unsigned int depth; depth = last_sync / cfqd->cfq_slice[1]; @@ -1471,11 +2365,13 @@ static int cfq_dispatch_requests(struct * task holds one reference to the queue, dropped when task exits. each rq * in-flight on this queue also holds a reference, dropped when rq is freed. * + * Each cfq queue took a reference on the parent group. Drop it now. * queue lock must be held here. */ static void cfq_put_queue(struct cfq_queue *cfqq) { struct cfq_data *cfqd = cfqq->cfqd; + struct cfq_group *cfqg, *orig_cfqg; BUG_ON(atomic_read(&cfqq->ref) <= 0); @@ -1485,14 +2381,19 @@ static void cfq_put_queue(struct cfq_que cfq_log_cfqq(cfqd, cfqq, "put_queue"); BUG_ON(rb_first(&cfqq->sort_list)); BUG_ON(cfqq->allocated[READ] + cfqq->allocated[WRITE]); - BUG_ON(cfq_cfqq_on_rr(cfqq)); + cfqg = cfqq->cfqg; + orig_cfqg = cfqq->orig_cfqg; if (unlikely(cfqd->active_queue == cfqq)) { __cfq_slice_expired(cfqd, cfqq, 0); cfq_schedule_dispatch(cfqd); } + BUG_ON(cfq_cfqq_on_rr(cfqq)); kmem_cache_free(cfq_pool, cfqq); + cfq_put_cfqg(cfqg); + if (orig_cfqg) + cfq_put_cfqg(orig_cfqg); } /* @@ -1622,14 +2523,8 @@ static void __cfq_exit_single_io_context cic->dead_key = (unsigned long) cic->key; cic->key = NULL; - rcu_read_lock(); - if (rcu_dereference(ioc->ioc_data) == cic) { - rcu_read_unlock(); - spin_lock(&ioc->lock); + if (ioc->ioc_data == cic) rcu_assign_pointer(ioc->ioc_data, NULL); - spin_unlock(&ioc->lock); - } else - rcu_read_unlock(); if (cic->cfqq[BLK_RW_ASYNC]) { cfq_exit_cfqq(cfqd, cic->cfqq[BLK_RW_ASYNC]); @@ -1791,14 +2686,51 @@ static void cfq_init_cfqq(struct cfq_dat cfqq->pid = pid; } +#ifdef CONFIG_CFQ_GROUP_IOSCHED +static void changed_cgroup(struct io_context *ioc, struct cfq_io_context *cic) +{ + struct cfq_queue *sync_cfqq = cic_to_cfqq(cic, 1); + struct cfq_data *cfqd = cic->key; + unsigned long flags; + struct request_queue *q; + + if (unlikely(!cfqd)) + return; + + q = cfqd->queue; + + spin_lock_irqsave(q->queue_lock, flags); + + if (sync_cfqq) { + /* + * Drop reference to sync queue. A new sync queue will be + * assigned in new group upon arrival of a fresh request. + */ + cfq_log_cfqq(cfqd, sync_cfqq, "changed cgroup"); + cic_set_cfqq(cic, NULL, 1); + cfq_put_queue(sync_cfqq); + } + + spin_unlock_irqrestore(q->queue_lock, flags); +} + +static void cfq_ioc_set_cgroup(struct io_context *ioc) +{ + call_for_each_cic(ioc, changed_cgroup); + ioc->cgroup_changed = 0; +} +#endif /* CONFIG_CFQ_GROUP_IOSCHED */ + static struct cfq_queue * cfq_find_alloc_queue(struct cfq_data *cfqd, bool is_sync, struct io_context *ioc, gfp_t gfp_mask) { struct cfq_queue *cfqq, *new_cfqq = NULL; struct cfq_io_context *cic; + struct cfq_group *cfqg; retry: + cfqg = cfq_get_cfqg(cfqd, 1); cic = cfq_cic_lookup(cfqd, ioc); /* cic always exists here */ cfqq = cic_to_cfqq(cic, is_sync); @@ -1829,6 +2761,7 @@ retry: if (cfqq) { cfq_init_cfqq(cfqd, cfqq, current->pid, is_sync); cfq_init_prio_data(cfqq, ioc); + cfq_link_cfqq_cfqg(cfqq, cfqg); cfq_log_cfqq(cfqd, cfqq, "alloced"); } else cfqq = &cfqd->oom_cfqq; @@ -2020,6 +2953,10 @@ out: if (unlikely(ioc->ioprio_changed)) cfq_ioc_set_ioprio(ioc); +#ifdef CONFIG_CFQ_GROUP_IOSCHED + if (unlikely(ioc->cgroup_changed)) + cfq_ioc_set_cgroup(ioc); +#endif return cic; err_free: cfq_cic_free(cic); @@ -2067,19 +3004,6 @@ cfq_update_io_seektime(struct cfq_data * total = cfqq->seek_total + (cfqq->seek_samples/2); do_div(total, cfqq->seek_samples); cfqq->seek_mean = (sector_t)total; - - /* - * If this cfqq is shared between multiple processes, check to - * make sure that those processes are still issuing I/Os within - * the mean seek distance. If not, it may be time to break the - * queues apart again. - */ - if (cfq_cfqq_coop(cfqq)) { - if (CFQQ_SEEKY(cfqq) && !cfqq->seeky_start) - cfqq->seeky_start = jiffies; - else if (!CFQQ_SEEKY(cfqq)) - cfqq->seeky_start = 0; - } } /* @@ -2100,14 +3024,15 @@ cfq_update_idle_window(struct cfq_data * enable_idle = old_idle = cfq_cfqq_idle_window(cfqq); + if (cfqq->queued[0] + cfqq->queued[1] >= 4) + cfq_mark_cfqq_deep(cfqq); + if (!atomic_read(&cic->ioc->nr_tasks) || !cfqd->cfq_slice_idle || - (!cfqd->cfq_latency && cfqd->hw_tag && CFQQ_SEEKY(cfqq))) + (!cfq_cfqq_deep(cfqq) && sample_valid(cfqq->seek_samples) + && CFQQ_SEEKY(cfqq))) enable_idle = 0; else if (sample_valid(cic->ttime_samples)) { - unsigned int slice_idle = cfqd->cfq_slice_idle; - if (sample_valid(cfqq->seek_samples) && CFQQ_SEEKY(cfqq)) - slice_idle = msecs_to_jiffies(CFQ_MIN_TT); - if (cic->ttime_mean > slice_idle) + if (cic->ttime_mean > cfqd->cfq_slice_idle) enable_idle = 0; else enable_idle = 1; @@ -2136,9 +3061,6 @@ cfq_should_preempt(struct cfq_data *cfqd if (!cfqq) return false; - if (cfq_slice_used(cfqq)) - return true; - if (cfq_class_idle(new_cfqq)) return false; @@ -2146,12 +3068,31 @@ cfq_should_preempt(struct cfq_data *cfqd return true; /* + * Don't allow a non-RT request to preempt an ongoing RT cfqq timeslice. + */ + if (cfq_class_rt(cfqq) && !cfq_class_rt(new_cfqq)) + return false; + + /* * if the new request is sync, but the currently running queue is * not, let the sync request have priority. */ if (rq_is_sync(rq) && !cfq_cfqq_sync(cfqq)) return true; + if (new_cfqq->cfqg != cfqq->cfqg) + return false; + + if (cfq_slice_used(cfqq)) + return true; + + /* Allow preemption only if we are idling on sync-noidle tree */ + if (cfqd->serving_type == SYNC_NOIDLE_WORKLOAD && + cfqq_type(new_cfqq) == SYNC_NOIDLE_WORKLOAD && + new_cfqq->service_tree->count == 2 && + RB_EMPTY_ROOT(&cfqq->sort_list)) + return true; + /* * So both queues are sync. Let the new request get disk time if * it's a metadata request and the current queue is doing regular IO. @@ -2172,7 +3113,7 @@ cfq_should_preempt(struct cfq_data *cfqd * if this request is as-good as one we would expect from the * current cfqq, let it preempt */ - if (cfq_rq_close(cfqd, cfqq, rq)) + if (cfq_rq_close(cfqd, cfqq, rq, true)) return true; return false; @@ -2234,9 +3175,10 @@ cfq_rq_enqueued(struct cfq_data *cfqd, s if (blk_rq_bytes(rq) > PAGE_CACHE_SIZE || cfqd->busy_queues > 1) { del_timer(&cfqd->idle_slice_timer); - __blk_run_queue(cfqd->queue); - } - cfq_mark_cfqq_must_dispatch(cfqq); + cfq_clear_cfqq_wait_request(cfqq); + __blk_run_queue(cfqd->queue); + } else + cfq_mark_cfqq_must_dispatch(cfqq); } } else if (cfq_should_preempt(cfqd, cfqq, rq)) { /* @@ -2258,10 +3200,9 @@ static void cfq_insert_request(struct re cfq_log_cfqq(cfqd, cfqq, "insert_request"); cfq_init_prio_data(cfqq, RQ_CIC(rq)->ioc); - cfq_add_rq_rb(rq); - rq_set_fifo_time(rq, jiffies + cfqd->cfq_fifo_expire[rq_is_sync(rq)]); list_add_tail(&rq->queuelist, &cfqq->fifo); + cfq_add_rq_rb(rq); cfq_rq_enqueued(cfqd, cfqq, rq); } @@ -2272,23 +3213,64 @@ static void cfq_insert_request(struct re */ static void cfq_update_hw_tag(struct cfq_data *cfqd) { - if (rq_in_driver(cfqd) > cfqd->rq_in_driver_peak) - cfqd->rq_in_driver_peak = rq_in_driver(cfqd); + struct cfq_queue *cfqq = cfqd->active_queue; + + if (rq_in_driver(cfqd) > cfqd->hw_tag_est_depth) + cfqd->hw_tag_est_depth = rq_in_driver(cfqd); + + if (cfqd->hw_tag == 1) + return; if (cfqd->rq_queued <= CFQ_HW_QUEUE_MIN && rq_in_driver(cfqd) <= CFQ_HW_QUEUE_MIN) return; + /* + * If active queue hasn't enough requests and can idle, cfq might not + * dispatch sufficient requests to hardware. Don't zero hw_tag in this + * case + */ + if (cfqq && cfq_cfqq_idle_window(cfqq) && + cfqq->dispatched + cfqq->queued[0] + cfqq->queued[1] < + CFQ_HW_QUEUE_MIN && rq_in_driver(cfqd) < CFQ_HW_QUEUE_MIN) + return; + if (cfqd->hw_tag_samples++ < 50) return; - if (cfqd->rq_in_driver_peak >= CFQ_HW_QUEUE_MIN) + if (cfqd->hw_tag_est_depth >= CFQ_HW_QUEUE_MIN) cfqd->hw_tag = 1; else cfqd->hw_tag = 0; +} + +static bool cfq_should_wait_busy(struct cfq_data *cfqd, struct cfq_queue *cfqq) +{ + struct cfq_io_context *cic = cfqd->active_cic; + + /* If there are other queues in the group, don't wait */ + if (cfqq->cfqg->nr_cfqq > 1) + return false; + + if (cfq_slice_used(cfqq)) + return true; + + /* if slice left is less than think time, wait busy */ + if (cic && sample_valid(cic->ttime_samples) + && (cfqq->slice_end - jiffies < cic->ttime_mean)) + return true; - cfqd->hw_tag_samples = 0; - cfqd->rq_in_driver_peak = 0; + /* + * If think times is less than a jiffy than ttime_mean=0 and above + * will not be true. It might happen that slice has not expired yet + * but will expire soon (4-5 ns) during select_queue(). To cover the + * case where think time is less than a jiffy, mark the queue wait + * busy if only 1 jiffy is left in the slice. + */ + if (cfqq->slice_end - jiffies == 1) + return true; + + return false; } static void cfq_completed_request(struct request_queue *q, struct request *rq) @@ -2299,7 +3281,7 @@ static void cfq_completed_request(struct unsigned long now; now = jiffies; - cfq_log_cfqq(cfqd, cfqq, "complete"); + cfq_log_cfqq(cfqd, cfqq, "complete rqnoidle %d", !!rq_noidle(rq)); cfq_update_hw_tag(cfqd); @@ -2313,7 +3295,8 @@ static void cfq_completed_request(struct if (sync) { RQ_CIC(rq)->last_end_request = now; - cfqd->last_end_sync_rq = now; + if (!time_after(rq->start_time + cfqd->cfq_fifo_expire[1], now)) + cfqd->last_delayed_sync = now; } /* @@ -2327,18 +3310,32 @@ static void cfq_completed_request(struct cfq_set_prio_slice(cfqd, cfqq); cfq_clear_cfqq_slice_new(cfqq); } + + /* + * Should we wait for next request to come in before we expire + * the queue. + */ + if (cfq_should_wait_busy(cfqd, cfqq)) { + cfqq->slice_end = jiffies + cfqd->cfq_slice_idle; + cfq_mark_cfqq_wait_busy(cfqq); + } + /* - * If there are no requests waiting in this queue, and - * there are other queues ready to issue requests, AND - * those other queues are issuing requests within our - * mean seek distance, give them a chance to run instead - * of idling. + * Idling is not enabled on: + * - expired queues + * - idle-priority queues + * - async queues + * - queues with still some requests queued + * - when there is a close cooperator */ if (cfq_slice_used(cfqq) || cfq_class_idle(cfqq)) cfq_slice_expired(cfqd, 1); - else if (cfqq_empty && !cfq_close_cooperator(cfqd, cfqq) && - sync && !rq_noidle(rq)) - cfq_arm_slice_timer(cfqd); + else if (sync && cfqq_empty && + !cfq_close_cooperator(cfqd, cfqq)) { + cfqd->noidle_tree_requires_idle |= !rq_noidle(rq); + if (cfqd->noidle_tree_requires_idle) + cfq_arm_slice_timer(cfqd); + } } if (!rq_in_driver(cfqd)) @@ -2362,12 +3359,10 @@ static void cfq_prio_boost(struct cfq_qu cfqq->ioprio = IOPRIO_NORM; } else { /* - * check if we need to unboost the queue + * unboost the queue (if needed) */ - if (cfqq->ioprio_class != cfqq->org_ioprio_class) - cfqq->ioprio_class = cfqq->org_ioprio_class; - if (cfqq->ioprio != cfqq->org_ioprio) - cfqq->ioprio = cfqq->org_ioprio; + cfqq->ioprio_class = cfqq->org_ioprio_class; + cfqq->ioprio = cfqq->org_ioprio; } } @@ -2442,14 +3437,6 @@ cfq_merge_cfqqs(struct cfq_data *cfqd, s return cic_to_cfqq(cic, 1); } -static int should_split_cfqq(struct cfq_queue *cfqq) -{ - if (cfqq->seeky_start && - time_after(jiffies, cfqq->seeky_start + CFQQ_COOP_TOUT)) - return 1; - return 0; -} - /* * Returns NULL if a new cfqq should be allocated, or the old cfqq if this * was the last process referring to said cfqq. @@ -2458,9 +3445,9 @@ static struct cfq_queue * split_cfqq(struct cfq_io_context *cic, struct cfq_queue *cfqq) { if (cfqq_process_refs(cfqq) == 1) { - cfqq->seeky_start = 0; cfqq->pid = current->pid; cfq_clear_cfqq_coop(cfqq); + cfq_clear_cfqq_split_coop(cfqq); return cfqq; } @@ -2499,7 +3486,7 @@ new_queue: /* * If the queue was seeky for too long, break it apart. */ - if (cfq_cfqq_coop(cfqq) && should_split_cfqq(cfqq)) { + if (cfq_cfqq_coop(cfqq) && cfq_cfqq_split_coop(cfqq)) { cfq_log_cfqq(cfqd, cfqq, "breaking apart cfqq"); cfqq = split_cfqq(cic, cfqq); if (!cfqq) @@ -2588,6 +3575,11 @@ static void cfq_idle_slice_timer(unsigne */ if (!RB_EMPTY_ROOT(&cfqq->sort_list)) goto out_kick; + + /* + * Queue depth flag is reset only when the idle didn't succeed + */ + cfq_clear_cfqq_deep(cfqq); } expire: cfq_slice_expired(cfqd, timed_out); @@ -2618,6 +3610,11 @@ static void cfq_put_async_queues(struct cfq_put_queue(cfqd->async_idle_cfqq); } +static void cfq_cfqd_free(struct rcu_head *head) +{ + kfree(container_of(head, struct cfq_data, rcu)); +} + static void cfq_exit_queue(struct elevator_queue *e) { struct cfq_data *cfqd = e->elevator_data; @@ -2639,25 +3636,49 @@ static void cfq_exit_queue(struct elevat } cfq_put_async_queues(cfqd); + cfq_release_cfq_groups(cfqd); + blkiocg_del_blkio_group(&cfqd->root_group.blkg); spin_unlock_irq(q->queue_lock); cfq_shutdown_timer_wq(cfqd); - kfree(cfqd); + /* Wait for cfqg->blkg->key accessors to exit their grace periods. */ + call_rcu(&cfqd->rcu, cfq_cfqd_free); } static void *cfq_init_queue(struct request_queue *q) { struct cfq_data *cfqd; - int i; + int i, j; + struct cfq_group *cfqg; + struct cfq_rb_root *st; cfqd = kmalloc_node(sizeof(*cfqd), GFP_KERNEL | __GFP_ZERO, q->node); if (!cfqd) return NULL; - cfqd->service_tree = CFQ_RB_ROOT; + /* Init root service tree */ + cfqd->grp_service_tree = CFQ_RB_ROOT; + /* Init root group */ + cfqg = &cfqd->root_group; + for_each_cfqg_st(cfqg, i, j, st) + *st = CFQ_RB_ROOT; + RB_CLEAR_NODE(&cfqg->rb_node); + + /* Give preference to root group over other groups */ + cfqg->weight = 2*BLKIO_WEIGHT_DEFAULT; + +#ifdef CONFIG_CFQ_GROUP_IOSCHED + /* + * Take a reference to root group which we never drop. This is just + * to make sure that cfq_put_cfqg() does not try to kfree root group + */ + atomic_set(&cfqg->ref, 1); + blkiocg_add_blkio_group(&blkio_root_cgroup, &cfqg->blkg, (void *)cfqd, + 0); +#endif /* * Not strictly needed (since RB_ROOT just clears the node and we * zeroed cfqd on alloc), but better be safe in case someone decides @@ -2673,6 +3694,7 @@ static void *cfq_init_queue(struct reque */ cfq_init_cfqq(cfqd, &cfqd->oom_cfqq, 1, 0); atomic_inc(&cfqd->oom_cfqq.ref); + cfq_link_cfqq_cfqg(&cfqd->oom_cfqq, &cfqd->root_group); INIT_LIST_HEAD(&cfqd->cic_list); @@ -2694,8 +3716,14 @@ static void *cfq_init_queue(struct reque cfqd->cfq_slice_async_rq = cfq_slice_async_rq; cfqd->cfq_slice_idle = cfq_slice_idle; cfqd->cfq_latency = 1; - cfqd->hw_tag = 1; - cfqd->last_end_sync_rq = jiffies; + cfqd->cfq_group_isolation = 0; + cfqd->hw_tag = -1; + /* + * we optimistically start assuming sync ops weren't delayed in last + * second, in order to have larger depth for async operations. + */ + cfqd->last_delayed_sync = jiffies - HZ; + INIT_RCU_HEAD(&cfqd->rcu); return cfqd; } @@ -2764,6 +3792,7 @@ SHOW_FUNCTION(cfq_slice_sync_show, cfqd- SHOW_FUNCTION(cfq_slice_async_show, cfqd->cfq_slice[0], 1); SHOW_FUNCTION(cfq_slice_async_rq_show, cfqd->cfq_slice_async_rq, 0); SHOW_FUNCTION(cfq_low_latency_show, cfqd->cfq_latency, 0); +SHOW_FUNCTION(cfq_group_isolation_show, cfqd->cfq_group_isolation, 0); #undef SHOW_FUNCTION #define STORE_FUNCTION(__FUNC, __PTR, MIN, MAX, __CONV) \ @@ -2796,6 +3825,7 @@ STORE_FUNCTION(cfq_slice_async_store, &c STORE_FUNCTION(cfq_slice_async_rq_store, &cfqd->cfq_slice_async_rq, 1, UINT_MAX, 0); STORE_FUNCTION(cfq_low_latency_store, &cfqd->cfq_latency, 0, 1, 0); +STORE_FUNCTION(cfq_group_isolation_store, &cfqd->cfq_group_isolation, 0, 1, 0); #undef STORE_FUNCTION #define CFQ_ATTR(name) \ @@ -2812,6 +3842,7 @@ static struct elv_fs_entry cfq_attrs[] = CFQ_ATTR(slice_async_rq), CFQ_ATTR(slice_idle), CFQ_ATTR(low_latency), + CFQ_ATTR(group_isolation), __ATTR_NULL }; @@ -2841,6 +3872,17 @@ static struct elevator_type iosched_cfq .elevator_owner = THIS_MODULE, }; +#ifdef CONFIG_CFQ_GROUP_IOSCHED +static struct blkio_policy_type blkio_policy_cfq = { + .ops = { + .blkio_unlink_group_fn = cfq_unlink_blkio_group, + .blkio_update_group_weight_fn = cfq_update_blkio_group_weight, + }, +}; +#else +static struct blkio_policy_type blkio_policy_cfq; +#endif + static int __init cfq_init(void) { /* @@ -2855,6 +3897,7 @@ static int __init cfq_init(void) return -ENOMEM; elv_register(&iosched_cfq); + blkio_policy_register(&blkio_policy_cfq); return 0; } @@ -2862,6 +3905,7 @@ static int __init cfq_init(void) static void __exit cfq_exit(void) { DECLARE_COMPLETION_ONSTACK(all_gone); + blkio_policy_unregister(&blkio_policy_cfq); elv_unregister(&iosched_cfq); ioc_gone = &all_gone; /* ioc_gone's update must be visible before reading ioc_count */ diff -urNp linux-2.6.32.48/block/compat_ioctl.c linux-2.6.32.48-openvz/block/compat_ioctl.c --- linux-2.6.32.48/block/compat_ioctl.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/block/compat_ioctl.c 2011-11-17 20:09:04.000000000 -0500 @@ -354,7 +354,7 @@ static int compat_fd_ioctl(struct block_ err |= __get_user(f->spec1, &uf->spec1); err |= __get_user(f->fmt_gap, &uf->fmt_gap); err |= __get_user(name, &uf->name); - f->name = compat_ptr(name); + f->name = (void __force_kernel *)compat_ptr(name); if (err) { err = -EFAULT; goto out; diff -urNp linux-2.6.32.48/block/elevator.c linux-2.6.32.48-openvz/block/elevator.c --- linux-2.6.32.48/block/elevator.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/block/elevator.c 2011-11-17 20:08:53.000000000 -0500 @@ -889,7 +889,7 @@ elv_attr_store(struct kobject *kobj, str return error; } -static struct sysfs_ops elv_sysfs_ops = { +static const struct sysfs_ops elv_sysfs_ops = { .show = elv_attr_show, .store = elv_attr_store, }; @@ -959,12 +959,12 @@ void elv_unregister(struct elevator_type */ if (e->ops.trim) { read_lock(&tasklist_lock); - do_each_thread(g, p) { + do_each_thread_all(g, p) { task_lock(p); if (p->io_context) e->ops.trim(p->io_context); task_unlock(p); - } while_each_thread(g, p); + } while_each_thread_all(g, p); read_unlock(&tasklist_lock); } diff -urNp linux-2.6.32.48/block/genhd.c linux-2.6.32.48-openvz/block/genhd.c --- linux-2.6.32.48/block/genhd.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/block/genhd.c 2011-11-17 20:08:53.000000000 -0500 @@ -22,9 +22,7 @@ #include "blk.h" static DEFINE_MUTEX(block_class_lock); -#ifndef CONFIG_SYSFS_DEPRECATED struct kobject *block_depr; -#endif /* for extended dynamic devt allocation, currently only one major is used */ #define MAX_EXT_DEVT (1 << MINORBITS) @@ -793,7 +791,7 @@ static int __init genhd_device_init(void { int error; - block_class.dev_kobj = sysfs_dev_block_kobj; + block_class.dev_kobj = ve_sysfs_dev_block_kobj; error = class_register(&block_class); if (unlikely(error)) return error; @@ -802,10 +800,10 @@ static int __init genhd_device_init(void register_blkdev(BLOCK_EXT_MAJOR, "blkext"); -#ifndef CONFIG_SYSFS_DEPRECATED - /* create top-level block dir */ - block_depr = kobject_create_and_add("block", NULL); -#endif + if (!sysfs_deprecated) + /* create top-level block dir */ + block_depr = kobject_create_and_add("block", NULL); + return 0; } @@ -997,6 +995,7 @@ static void disk_release(struct device * struct class block_class = { .name = "block", }; +EXPORT_SYMBOL(block_class); static char *block_devnode(struct device *dev, mode_t *mode) { diff -urNp linux-2.6.32.48/block/Kconfig linux-2.6.32.48-openvz/block/Kconfig --- linux-2.6.32.48/block/Kconfig 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/block/Kconfig 2011-11-17 20:08:53.000000000 -0500 @@ -77,6 +77,28 @@ config BLK_DEV_INTEGRITY T10/SCSI Data Integrity Field or the T13/ATA External Path Protection. If in doubt, say N. +config BLK_CGROUP + bool + depends on CGROUPS + default n + ---help--- + Generic block IO controller cgroup interface. This is the common + cgroup interface which should be used by various IO controlling + policies. + + Currently, CFQ IO scheduler uses it to recognize task groups and + control disk bandwidth allocation (proportional time slice allocation) + to such task groups. + +config DEBUG_BLK_CGROUP + bool + depends on BLK_CGROUP + default n + ---help--- + Enable some debugging help. Currently it stores the cgroup path + in the blk group which can be used by cfq for tracing various + group related activity. + endif # BLOCK config BLOCK_COMPAT diff -urNp linux-2.6.32.48/block/Kconfig.iosched linux-2.6.32.48-openvz/block/Kconfig.iosched --- linux-2.6.32.48/block/Kconfig.iosched 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/block/Kconfig.iosched 2011-11-17 20:08:53.000000000 -0500 @@ -40,6 +40,23 @@ config IOSCHED_CFQ working environment, suitable for desktop systems. This is the default I/O scheduler. +config CFQ_GROUP_IOSCHED + bool "CFQ Group Scheduling support" + depends on IOSCHED_CFQ && CGROUPS + select BLK_CGROUP + default n + ---help--- + Enable group IO scheduling in CFQ. + +config DEBUG_CFQ_IOSCHED + bool "Debug CFQ Scheduling" + depends on CFQ_GROUP_IOSCHED + select DEBUG_BLK_CGROUP + default n + ---help--- + Enable CFQ IO scheduling debugging in CFQ. Currently it makes + blktrace output more verbose. + choice prompt "Default I/O scheduler" default DEFAULT_CFQ diff -urNp linux-2.6.32.48/block/Makefile linux-2.6.32.48-openvz/block/Makefile --- linux-2.6.32.48/block/Makefile 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/block/Makefile 2011-11-17 20:08:53.000000000 -0500 @@ -8,6 +8,7 @@ obj-$(CONFIG_BLOCK) := elevator.o blk-co blk-iopoll.o ioctl.o genhd.o scsi_ioctl.o obj-$(CONFIG_BLK_DEV_BSG) += bsg.o +obj-$(CONFIG_BLK_CGROUP) += blk-cgroup.o obj-$(CONFIG_IOSCHED_NOOP) += noop-iosched.o obj-$(CONFIG_IOSCHED_AS) += as-iosched.o obj-$(CONFIG_IOSCHED_DEADLINE) += deadline-iosched.o diff -urNp linux-2.6.32.48/block/scsi_ioctl.c linux-2.6.32.48-openvz/block/scsi_ioctl.c --- linux-2.6.32.48/block/scsi_ioctl.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/block/scsi_ioctl.c 2011-11-17 20:08:53.000000000 -0500 @@ -220,8 +220,20 @@ EXPORT_SYMBOL(blk_verify_command); static int blk_fill_sghdr_rq(struct request_queue *q, struct request *rq, struct sg_io_hdr *hdr, fmode_t mode) { - if (copy_from_user(rq->cmd, hdr->cmdp, hdr->cmd_len)) + unsigned char tmpcmd[sizeof(rq->__cmd)]; + unsigned char *cmdptr; + + if (rq->cmd != rq->__cmd) + cmdptr = rq->cmd; + else + cmdptr = tmpcmd; + + if (copy_from_user(cmdptr, hdr->cmdp, hdr->cmd_len)) return -EFAULT; + + if (cmdptr != rq->cmd) + memcpy(rq->cmd, cmdptr, hdr->cmd_len); + if (blk_verify_command(rq->cmd, mode & FMODE_WRITE)) return -EPERM; @@ -430,6 +442,8 @@ int sg_scsi_ioctl(struct request_queue * int err; unsigned int in_len, out_len, bytes, opcode, cmdlen; char *buffer = NULL, sense[SCSI_SENSE_BUFFERSIZE]; + unsigned char tmpcmd[sizeof(rq->__cmd)]; + unsigned char *cmdptr; if (!sic) return -EINVAL; @@ -463,9 +477,18 @@ int sg_scsi_ioctl(struct request_queue * */ err = -EFAULT; rq->cmd_len = cmdlen; - if (copy_from_user(rq->cmd, sic->data, cmdlen)) + + if (rq->cmd != rq->__cmd) + cmdptr = rq->cmd; + else + cmdptr = tmpcmd; + + if (copy_from_user(cmdptr, sic->data, cmdlen)) goto error; + if (rq->cmd != cmdptr) + memcpy(rq->cmd, cmdptr, cmdlen); + if (in_len && copy_from_user(buffer, sic->data + cmdlen, in_len)) goto error; diff -urNp linux-2.6.32.48/COPYING.Parallels linux-2.6.32.48-openvz/COPYING.Parallels --- linux-2.6.32.48/COPYING.Parallels 1969-12-31 19:00:00.000000000 -0500 +++ linux-2.6.32.48-openvz/COPYING.Parallels 2011-11-17 20:08:53.000000000 -0500 @@ -0,0 +1,350 @@ + +Nothing in this license should be construed as a grant by Parallels of any rights +beyond the rights specified in the GNU General Public License, and nothing in +this license should be construed as a waiver by Parallels of its patent, copyright +and/or trademark rights, beyond the waiver required by the GNU General Public +License. This license is expressly inapplicable to any product that is not +within the scope of the GNU General Public License + +---------------------------------------- + + GNU GENERAL PUBLIC LICENSE + Version 2, June 1991 + + Copyright (C) 1989, 1991 Free Software Foundation, Inc. + 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + Everyone is permitted to copy and distribute verbatim copies + of this license document, but changing it is not allowed. + + Preamble + + The licenses for most software are designed to take away your +freedom to share and change it. By contrast, the GNU General Public +License is intended to guarantee your freedom to share and change free +software--to make sure the software is free for all its users. This +General Public License applies to most of the Free Software +Foundation's software and to any other program whose authors commit to +using it. (Some other Free Software Foundation software is covered by +the GNU Library General Public License instead.) You can apply it to +your programs, too. + + When we speak of free software, we are referring to freedom, not +price. Our General Public Licenses are designed to make sure that you +have the freedom to distribute copies of free software (and charge for +this service if you wish), that you receive source code or can get it +if you want it, that you can change the software or use pieces of it +in new free programs; and that you know you can do these things. + + To protect your rights, we need to make restrictions that forbid +anyone to deny you these rights or to ask you to surrender the rights. +These restrictions translate to certain responsibilities for you if you +distribute copies of the software, or if you modify it. + + For example, if you distribute copies of such a program, whether +gratis or for a fee, you must give the recipients all the rights that +you have. You must make sure that they, too, receive or can get the +source code. And you must show them these terms so they know their +rights. + + We protect your rights with two steps: (1) copyright the software, and +(2) offer you this license which gives you legal permission to copy, +distribute and/or modify the software. + + Also, for each author's protection and ours, we want to make certain +that everyone understands that there is no warranty for this free +software. If the software is modified by someone else and passed on, we +want its recipients to know that what they have is not the original, so +that any problems introduced by others will not reflect on the original +authors' reputations. + + Finally, any free program is threatened constantly by software +patents. We wish to avoid the danger that redistributors of a free +program will individually obtain patent licenses, in effect making the +program proprietary. To prevent this, we have made it clear that any +patent must be licensed for everyone's free use or not licensed at all. + + The precise terms and conditions for copying, distribution and +modification follow. + + GNU GENERAL PUBLIC LICENSE + TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION + + 0. This License applies to any program or other work which contains +a notice placed by the copyright holder saying it may be distributed +under the terms of this General Public License. The "Program", below, +refers to any such program or work, and a "work based on the Program" +means either the Program or any derivative work under copyright law: +that is to say, a work containing the Program or a portion of it, +either verbatim or with modifications and/or translated into another +language. (Hereinafter, translation is included without limitation in +the term "modification".) Each licensee is addressed as "you". + +Activities other than copying, distribution and modification are not +covered by this License; they are outside its scope. The act of +running the Program is not restricted, and the output from the Program +is covered only if its contents constitute a work based on the +Program (independent of having been made by running the Program). +Whether that is true depends on what the Program does. + + 1. You may copy and distribute verbatim copies of the Program's +source code as you receive it, in any medium, provided that you +conspicuously and appropriately publish on each copy an appropriate +copyright notice and disclaimer of warranty; keep intact all the +notices that refer to this License and to the absence of any warranty; +and give any other recipients of the Program a copy of this License +along with the Program. + +You may charge a fee for the physical act of transferring a copy, and +you may at your option offer warranty protection in exchange for a fee. + + 2. You may modify your copy or copies of the Program or any portion +of it, thus forming a work based on the Program, and copy and +distribute such modifications or work under the terms of Section 1 +above, provided that you also meet all of these conditions: + + a) You must cause the modified files to carry prominent notices + stating that you changed the files and the date of any change. + + b) You must cause any work that you distribute or publish, that in + whole or in part contains or is derived from the Program or any + part thereof, to be licensed as a whole at no charge to all third + parties under the terms of this License. + + c) If the modified program normally reads commands interactively + when run, you must cause it, when started running for such + interactive use in the most ordinary way, to print or display an + announcement including an appropriate copyright notice and a + notice that there is no warranty (or else, saying that you provide + a warranty) and that users may redistribute the program under + these conditions, and telling the user how to view a copy of this + License. (Exception: if the Program itself is interactive but + does not normally print such an announcement, your work based on + the Program is not required to print an announcement.) + +These requirements apply to the modified work as a whole. If +identifiable sections of that work are not derived from the Program, +and can be reasonably considered independent and separate works in +themselves, then this License, and its terms, do not apply to those +sections when you distribute them as separate works. But when you +distribute the same sections as part of a whole which is a work based +on the Program, the distribution of the whole must be on the terms of +this License, whose permissions for other licensees extend to the +entire whole, and thus to each and every part regardless of who wrote it. + +Thus, it is not the intent of this section to claim rights or contest +your rights to work written entirely by you; rather, the intent is to +exercise the right to control the distribution of derivative or +collective works based on the Program. + +In addition, mere aggregation of another work not based on the Program +with the Program (or with a work based on the Program) on a volume of +a storage or distribution medium does not bring the other work under +the scope of this License. + + 3. You may copy and distribute the Program (or a work based on it, +under Section 2) in object code or executable form under the terms of +Sections 1 and 2 above provided that you also do one of the following: + + a) Accompany it with the complete corresponding machine-readable + source code, which must be distributed under the terms of Sections + 1 and 2 above on a medium customarily used for software interchange; or, + + b) Accompany it with a written offer, valid for at least three + years, to give any third party, for a charge no more than your + cost of physically performing source distribution, a complete + machine-readable copy of the corresponding source code, to be + distributed under the terms of Sections 1 and 2 above on a medium + customarily used for software interchange; or, + + c) Accompany it with the information you received as to the offer + to distribute corresponding source code. (This alternative is + allowed only for noncommercial distribution and only if you + received the program in object code or executable form with such + an offer, in accord with Subsection b above.) + +The source code for a work means the preferred form of the work for +making modifications to it. For an executable work, complete source +code means all the source code for all modules it contains, plus any +associated interface definition files, plus the scripts used to +control compilation and installation of the executable. However, as a +special exception, the source code distributed need not include +anything that is normally distributed (in either source or binary +form) with the major components (compiler, kernel, and so on) of the +operating system on which the executable runs, unless that component +itself accompanies the executable. + +If distribution of executable or object code is made by offering +access to copy from a designated place, then offering equivalent +access to copy the source code from the same place counts as +distribution of the source code, even though third parties are not +compelled to copy the source along with the object code. + + 4. You may not copy, modify, sublicense, or distribute the Program +except as expressly provided under this License. Any attempt +otherwise to copy, modify, sublicense or distribute the Program is +void, and will automatically terminate your rights under this License. +However, parties who have received copies, or rights, from you under +this License will not have their licenses terminated so long as such +parties remain in full compliance. + + 5. You are not required to accept this License, since you have not +signed it. However, nothing else grants you permission to modify or +distribute the Program or its derivative works. These actions are +prohibited by law if you do not accept this License. Therefore, by +modifying or distributing the Program (or any work based on the +Program), you indicate your acceptance of this License to do so, and +all its terms and conditions for copying, distributing or modifying +the Program or works based on it. + + 6. Each time you redistribute the Program (or any work based on the +Program), the recipient automatically receives a license from the +original licensor to copy, distribute or modify the Program subject to +these terms and conditions. You may not impose any further +restrictions on the recipients' exercise of the rights granted herein. +You are not responsible for enforcing compliance by third parties to +this License. + + 7. If, as a consequence of a court judgment or allegation of patent +infringement or for any other reason (not limited to patent issues), +conditions are imposed on you (whether by court order, agreement or +otherwise) that contradict the conditions of this License, they do not +excuse you from the conditions of this License. If you cannot +distribute so as to satisfy simultaneously your obligations under this +License and any other pertinent obligations, then as a consequence you +may not distribute the Program at all. For example, if a patent +license would not permit royalty-free redistribution of the Program by +all those who receive copies directly or indirectly through you, then +the only way you could satisfy both it and this License would be to +refrain entirely from distribution of the Program. + +If any portion of this section is held invalid or unenforceable under +any particular circumstance, the balance of the section is intended to +apply and the section as a whole is intended to apply in other +circumstances. + +It is not the purpose of this section to induce you to infringe any +patents or other property right claims or to contest validity of any +such claims; this section has the sole purpose of protecting the +integrity of the free software distribution system, which is +implemented by public license practices. Many people have made +generous contributions to the wide range of software distributed +through that system in reliance on consistent application of that +system; it is up to the author/donor to decide if he or she is willing +to distribute software through any other system and a licensee cannot +impose that choice. + +This section is intended to make thoroughly clear what is believed to +be a consequence of the rest of this License. + + 8. If the distribution and/or use of the Program is restricted in +certain countries either by patents or by copyrighted interfaces, the +original copyright holder who places the Program under this License +may add an explicit geographical distribution limitation excluding +those countries, so that distribution is permitted only in or among +countries not thus excluded. In such case, this License incorporates +the limitation as if written in the body of this License. + + 9. The Free Software Foundation may publish revised and/or new versions +of the General Public License from time to time. Such new versions will +be similar in spirit to the present version, but may differ in detail to +address new problems or concerns. + +Each version is given a distinguishing version number. If the Program +specifies a version number of this License which applies to it and "any +later version", you have the option of following the terms and conditions +either of that version or of any later version published by the Free +Software Foundation. If the Program does not specify a version number of +this License, you may choose any version ever published by the Free Software +Foundation. + + 10. If you wish to incorporate parts of the Program into other free +programs whose distribution conditions are different, write to the author +to ask for permission. For software which is copyrighted by the Free +Software Foundation, write to the Free Software Foundation; we sometimes +make exceptions for this. Our decision will be guided by the two goals +of preserving the free status of all derivatives of our free software and +of promoting the sharing and reuse of software generally. + + NO WARRANTY + + 11. BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY +FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW. EXCEPT WHEN +OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES +PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED +OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF +MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE ENTIRE RISK AS +TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU. SHOULD THE +PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING, +REPAIR OR CORRECTION. + + 12. IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING +WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR +REDISTRIBUTE THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, +INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING +OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED +TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY +YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER +PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE +POSSIBILITY OF SUCH DAMAGES. + + END OF TERMS AND CONDITIONS + + How to Apply These Terms to Your New Programs + + If you develop a new program, and you want it to be of the greatest +possible use to the public, the best way to achieve this is to make it +free software which everyone can redistribute and change under these terms. + + To do so, attach the following notices to the program. It is safest +to attach them to the start of each source file to most effectively +convey the exclusion of warranty; and each file should have at least +the "copyright" line and a pointer to where the full notice is found. + + + Copyright (C) + + This program is free software; you can redistribute it and/or modify + it under the terms of the GNU General Public License as published by + the Free Software Foundation; either version 2 of the License, or + (at your option) any later version. + + This program is distributed in the hope that it will be useful, + but WITHOUT ANY WARRANTY; without even the implied warranty of + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + GNU General Public License for more details. + + You should have received a copy of the GNU General Public License + along with this program; if not, write to the Free Software + Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + + +Also add information on how to contact you by electronic and paper mail. + +If the program is interactive, make it output a short notice like this +when it starts in an interactive mode: + + Gnomovision version 69, Copyright (C) year name of author + Gnomovision comes with ABSOLUTELY NO WARRANTY; for details type `show w'. + This is free software, and you are welcome to redistribute it + under certain conditions; type `show c' for details. + +The hypothetical commands `show w' and `show c' should show the appropriate +parts of the General Public License. Of course, the commands you use may +be called something other than `show w' and `show c'; they could even be +mouse-clicks or menu items--whatever suits your program. + +You should also get your employer (if you work as a programmer) or your +school, if any, to sign a "copyright disclaimer" for the program, if +necessary. Here is a sample; alter the names: + + Yoyodyne, Inc., hereby disclaims all copyright interest in the program + `Gnomovision' (which makes passes at compilers) written by James Hacker. + + , 1 April 1989 + Ty Coon, President of Vice + +This General Public License does not permit incorporating your program into +proprietary programs. If your program is a subroutine library, you may +consider it more useful to permit linking proprietary applications with the +library. If this is what you want to do, use the GNU Library General +Public License instead of this License. diff -urNp linux-2.6.32.48/crypto/cryptd.c linux-2.6.32.48-openvz/crypto/cryptd.c --- linux-2.6.32.48/crypto/cryptd.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/crypto/cryptd.c 2011-11-17 20:09:04.000000000 -0500 @@ -50,7 +50,7 @@ struct cryptd_blkcipher_ctx { struct cryptd_blkcipher_request_ctx { crypto_completion_t complete; -}; +} __no_const; struct cryptd_hash_ctx { struct crypto_shash *child; diff -urNp linux-2.6.32.48/crypto/gf128mul.c linux-2.6.32.48-openvz/crypto/gf128mul.c --- linux-2.6.32.48/crypto/gf128mul.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/crypto/gf128mul.c 2011-11-17 20:09:04.000000000 -0500 @@ -182,7 +182,7 @@ void gf128mul_lle(be128 *r, const be128 for (i = 0; i < 7; ++i) gf128mul_x_lle(&p[i + 1], &p[i]); - memset(r, 0, sizeof(r)); + memset(r, 0, sizeof(*r)); for (i = 0;;) { u8 ch = ((u8 *)b)[15 - i]; @@ -220,7 +220,7 @@ void gf128mul_bbe(be128 *r, const be128 for (i = 0; i < 7; ++i) gf128mul_x_bbe(&p[i + 1], &p[i]); - memset(r, 0, sizeof(r)); + memset(r, 0, sizeof(*r)); for (i = 0;;) { u8 ch = ((u8 *)b)[i]; diff -urNp linux-2.6.32.48/crypto/serpent.c linux-2.6.32.48-openvz/crypto/serpent.c --- linux-2.6.32.48/crypto/serpent.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/crypto/serpent.c 2011-11-17 20:09:04.000000000 -0500 @@ -21,6 +21,7 @@ #include #include #include +#include /* Key is padded to the maximum of 256 bits before round key generation. * Any key length <= 256 bits (32 bytes) is allowed by the algorithm. @@ -224,6 +225,8 @@ static int serpent_setkey(struct crypto_ u32 r0,r1,r2,r3,r4; int i; + pax_track_stack(); + /* Copy key, add padding */ for (i = 0; i < keylen; ++i) diff -urNp linux-2.6.32.48/Documentation/dontdiff linux-2.6.32.48-openvz/Documentation/dontdiff --- linux-2.6.32.48/Documentation/dontdiff 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/Documentation/dontdiff 2011-11-17 20:09:04.000000000 -0500 @@ -1,13 +1,16 @@ *.a *.aux *.bin +*.cis *.cpio *.csp +*.dbg *.dsp *.dvi *.elf *.eps *.fw +*.gcno *.gen.S *.gif *.grep @@ -38,8 +41,10 @@ *.tab.h *.tex *.ver +*.vim *.xml *_MODULES +*_reg_safe.h *_vga16.c *~ *.9 @@ -49,11 +54,16 @@ 53c700_d.h CVS ChangeSet +GPATH +GRTAGS +GSYMS +GTAGS Image Kerntypes Module.markers Module.symvers PENDING +PERF* SCCS System.map* TAGS @@ -76,7 +86,11 @@ btfixupprep build bvmlinux bzImage* +capability_names.h +capflags.c classlist.h* +clut_vga16.c +common-cmds.h comp*.log compile.h* conf @@ -97,19 +111,21 @@ elfconfig.h* fixdep fore200e_mkfirm fore200e_pca_fw.c* +gate.lds gconf gen-devlist gen_crc32table gen_init_cpio genksyms *_gray256.c +hash ihex2fw ikconfig.h* initramfs_data.cpio +initramfs_data.cpio.bz2 initramfs_data.cpio.gz initramfs_list kallsyms -kconfig keywords.c ksym.c* ksym.h* @@ -133,7 +149,9 @@ mkboot mkbugboot mkcpustr mkdep +mkpiggy mkprep +mkregtable mktables mktree modpost @@ -149,6 +167,7 @@ patches* pca200e.bin pca200e_ecd.bin2 piggy.gz +piggy.S piggyback pnmtologo ppc_defs.h* @@ -157,12 +176,15 @@ qconf raid6altivec*.c raid6int*.c raid6tables.c +regdb.c relocs +rlim_names.h series setup setup.bin setup.elf sImage +slabinfo sm_tbl* split-include syscalltab.h @@ -186,14 +208,20 @@ version.h* vmlinux vmlinux-* vmlinux.aout +vmlinux.bin.all +vmlinux.bin.bz2 vmlinux.lds +vmlinux.relocs +voffset.h vsyscall.lds vsyscall_32.lds wanxlfw.inc uImage unifdef +utsrelease.h wakeup.bin wakeup.elf wakeup.lds zImage* zconf.hash.c +zoffset.h diff -urNp linux-2.6.32.48/Documentation/kernel-parameters.txt linux-2.6.32.48-openvz/Documentation/kernel-parameters.txt --- linux-2.6.32.48/Documentation/kernel-parameters.txt 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/Documentation/kernel-parameters.txt 2011-11-17 20:08:53.000000000 -0500 @@ -1837,6 +1837,13 @@ and is between 256 and 4096 characters. the specified number of seconds. This is to be used if your oopses keep scrolling off the screen. + pax_nouderef [X86] disables UDEREF. Most likely needed under certain + virtualization environments that don't cope well with the + expand down segment used by UDEREF on X86-32 or the frequent + page table updates on X86-64. + + pax_softmode= 0/1 to disable/enable PaX softmode on boot already. + pcbit= [HW,ISDN] pcd. [PARIDE] diff -urNp linux-2.6.32.48/drivers/acpi/acpi_pad.c linux-2.6.32.48-openvz/drivers/acpi/acpi_pad.c --- linux-2.6.32.48/drivers/acpi/acpi_pad.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/acpi/acpi_pad.c 2011-11-17 20:08:53.000000000 -0500 @@ -30,7 +30,7 @@ #include #include -#define ACPI_PROCESSOR_AGGREGATOR_CLASS "processor_aggregator" +#define ACPI_PROCESSOR_AGGREGATOR_CLASS "acpi_pad" #define ACPI_PROCESSOR_AGGREGATOR_DEVICE_NAME "Processor Aggregator" #define ACPI_PROCESSOR_AGGREGATOR_NOTIFY 0x80 static DEFINE_MUTEX(isolated_cpus_lock); diff -urNp linux-2.6.32.48/drivers/acpi/battery.c linux-2.6.32.48-openvz/drivers/acpi/battery.c --- linux-2.6.32.48/drivers/acpi/battery.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/acpi/battery.c 2011-11-17 20:08:53.000000000 -0500 @@ -763,7 +763,7 @@ DECLARE_FILE_FUNCTIONS(alarm); } static struct battery_file { - struct file_operations ops; + const struct file_operations ops; mode_t mode; const char *name; } acpi_battery_file[] = { diff -urNp linux-2.6.32.48/drivers/acpi/dock.c linux-2.6.32.48-openvz/drivers/acpi/dock.c --- linux-2.6.32.48/drivers/acpi/dock.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/acpi/dock.c 2011-11-17 20:08:53.000000000 -0500 @@ -77,7 +77,7 @@ struct dock_dependent_device { struct list_head list; struct list_head hotplug_list; acpi_handle handle; - struct acpi_dock_ops *ops; + const struct acpi_dock_ops *ops; void *context; }; @@ -605,7 +605,7 @@ EXPORT_SYMBOL_GPL(unregister_dock_notifi * the dock driver after _DCK is executed. */ int -register_hotplug_dock_device(acpi_handle handle, struct acpi_dock_ops *ops, +register_hotplug_dock_device(acpi_handle handle, const struct acpi_dock_ops *ops, void *context) { struct dock_dependent_device *dd; diff -urNp linux-2.6.32.48/drivers/acpi/osl.c linux-2.6.32.48-openvz/drivers/acpi/osl.c --- linux-2.6.32.48/drivers/acpi/osl.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/acpi/osl.c 2011-11-17 20:08:53.000000000 -0500 @@ -523,6 +523,8 @@ acpi_os_read_memory(acpi_physical_addres void __iomem *virt_addr; virt_addr = ioremap(phys_addr, width); + if (!virt_addr) + return AE_NO_MEMORY; if (!value) value = &dummy; @@ -551,6 +553,8 @@ acpi_os_write_memory(acpi_physical_addre void __iomem *virt_addr; virt_addr = ioremap(phys_addr, width); + if (!virt_addr) + return AE_NO_MEMORY; switch (width) { case 8: diff -urNp linux-2.6.32.48/drivers/acpi/power_meter.c linux-2.6.32.48-openvz/drivers/acpi/power_meter.c --- linux-2.6.32.48/drivers/acpi/power_meter.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/acpi/power_meter.c 2011-11-17 20:08:53.000000000 -0500 @@ -315,8 +315,6 @@ static ssize_t set_trip(struct device *d return res; temp /= 1000; - if (temp < 0) - return -EINVAL; mutex_lock(&resource->lock); resource->trip[attr->index - 7] = temp; diff -urNp linux-2.6.32.48/drivers/acpi/proc.c linux-2.6.32.48-openvz/drivers/acpi/proc.c --- linux-2.6.32.48/drivers/acpi/proc.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/acpi/proc.c 2011-11-17 20:08:53.000000000 -0500 @@ -391,20 +391,15 @@ acpi_system_write_wakeup_device(struct f size_t count, loff_t * ppos) { struct list_head *node, *next; - char strbuf[5]; - char str[5] = ""; - unsigned int len = count; + char strbuf[5] = {0}; struct acpi_device *found_dev = NULL; - if (len > 4) - len = 4; - if (len < 0) - return -EFAULT; + if (count > 4) + count = 4; - if (copy_from_user(strbuf, buffer, len)) + if (copy_from_user(strbuf, buffer, count)) return -EFAULT; - strbuf[len] = '\0'; - sscanf(strbuf, "%s", str); + strbuf[count] = '\0'; mutex_lock(&acpi_device_lock); list_for_each_safe(node, next, &acpi_wakeup_device_list) { @@ -413,7 +408,7 @@ acpi_system_write_wakeup_device(struct f if (!dev->wakeup.flags.valid) continue; - if (!strncmp(dev->pnp.bus_id, str, 4)) { + if (!strncmp(dev->pnp.bus_id, strbuf, 4)) { dev->wakeup.state.enabled = dev->wakeup.state.enabled ? 0 : 1; found_dev = dev; diff -urNp linux-2.6.32.48/drivers/acpi/processor_core.c linux-2.6.32.48-openvz/drivers/acpi/processor_core.c --- linux-2.6.32.48/drivers/acpi/processor_core.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/acpi/processor_core.c 2011-11-17 20:08:53.000000000 -0500 @@ -790,7 +790,7 @@ static int __cpuinit acpi_processor_add( return 0; } - BUG_ON((pr->id >= nr_cpu_ids) || (pr->id < 0)); + BUG_ON(pr->id >= nr_cpu_ids); /* * Buggy BIOS check diff -urNp linux-2.6.32.48/drivers/acpi/sbshc.c linux-2.6.32.48-openvz/drivers/acpi/sbshc.c --- linux-2.6.32.48/drivers/acpi/sbshc.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/acpi/sbshc.c 2011-11-17 20:08:53.000000000 -0500 @@ -17,7 +17,7 @@ #define PREFIX "ACPI: " -#define ACPI_SMB_HC_CLASS "smbus_host_controller" +#define ACPI_SMB_HC_CLASS "smbus_host_ctl" #define ACPI_SMB_HC_DEVICE_NAME "ACPI SMBus HC" struct acpi_smb_hc { diff -urNp linux-2.6.32.48/drivers/acpi/sleep.c linux-2.6.32.48-openvz/drivers/acpi/sleep.c --- linux-2.6.32.48/drivers/acpi/sleep.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/acpi/sleep.c 2011-11-17 20:08:53.000000000 -0500 @@ -283,7 +283,7 @@ static int acpi_suspend_state_valid(susp } } -static struct platform_suspend_ops acpi_suspend_ops = { +static const struct platform_suspend_ops acpi_suspend_ops = { .valid = acpi_suspend_state_valid, .begin = acpi_suspend_begin, .prepare_late = acpi_pm_prepare, @@ -311,7 +311,7 @@ static int acpi_suspend_begin_old(suspen * The following callbacks are used if the pre-ACPI 2.0 suspend ordering has * been requested. */ -static struct platform_suspend_ops acpi_suspend_ops_old = { +static const struct platform_suspend_ops acpi_suspend_ops_old = { .valid = acpi_suspend_state_valid, .begin = acpi_suspend_begin_old, .prepare_late = acpi_pm_disable_gpes, @@ -460,7 +460,7 @@ static void acpi_pm_enable_gpes(void) acpi_enable_all_runtime_gpes(); } -static struct platform_hibernation_ops acpi_hibernation_ops = { +static const struct platform_hibernation_ops acpi_hibernation_ops = { .begin = acpi_hibernation_begin, .end = acpi_pm_end, .pre_snapshot = acpi_hibernation_pre_snapshot, @@ -513,7 +513,7 @@ static int acpi_hibernation_pre_snapshot * The following callbacks are used if the pre-ACPI 2.0 suspend ordering has * been requested. */ -static struct platform_hibernation_ops acpi_hibernation_ops_old = { +static const struct platform_hibernation_ops acpi_hibernation_ops_old = { .begin = acpi_hibernation_begin_old, .end = acpi_pm_end, .pre_snapshot = acpi_hibernation_pre_snapshot_old, diff -urNp linux-2.6.32.48/drivers/acpi/video.c linux-2.6.32.48-openvz/drivers/acpi/video.c --- linux-2.6.32.48/drivers/acpi/video.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/acpi/video.c 2011-11-17 20:08:53.000000000 -0500 @@ -359,7 +359,7 @@ static int acpi_video_set_brightness(str vd->brightness->levels[request_level]); } -static struct backlight_ops acpi_backlight_ops = { +static const struct backlight_ops acpi_backlight_ops = { .get_brightness = acpi_video_get_brightness, .update_status = acpi_video_set_brightness, }; diff -urNp linux-2.6.32.48/drivers/ata/ahci.c linux-2.6.32.48-openvz/drivers/ata/ahci.c --- linux-2.6.32.48/drivers/ata/ahci.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/ahci.c 2011-11-17 20:09:04.000000000 -0500 @@ -387,7 +387,7 @@ static struct scsi_host_template ahci_sh .sdev_attrs = ahci_sdev_attrs, }; -static struct ata_port_operations ahci_ops = { +static const struct ata_port_operations ahci_ops = { .inherits = &sata_pmp_port_ops, .qc_defer = sata_pmp_qc_defer_cmd_switch, @@ -424,17 +424,17 @@ static struct ata_port_operations ahci_o .port_stop = ahci_port_stop, }; -static struct ata_port_operations ahci_vt8251_ops = { +static const struct ata_port_operations ahci_vt8251_ops = { .inherits = &ahci_ops, .hardreset = ahci_vt8251_hardreset, }; -static struct ata_port_operations ahci_p5wdh_ops = { +static const struct ata_port_operations ahci_p5wdh_ops = { .inherits = &ahci_ops, .hardreset = ahci_p5wdh_hardreset, }; -static struct ata_port_operations ahci_sb600_ops = { +static const struct ata_port_operations ahci_sb600_ops = { .inherits = &ahci_ops, .softreset = ahci_sb600_softreset, .pmp_softreset = ahci_sb600_softreset, diff -urNp linux-2.6.32.48/drivers/ata/ata_generic.c linux-2.6.32.48-openvz/drivers/ata/ata_generic.c --- linux-2.6.32.48/drivers/ata/ata_generic.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/ata_generic.c 2011-11-17 20:08:53.000000000 -0500 @@ -104,7 +104,7 @@ static struct scsi_host_template generic ATA_BMDMA_SHT(DRV_NAME), }; -static struct ata_port_operations generic_port_ops = { +static const struct ata_port_operations generic_port_ops = { .inherits = &ata_bmdma_port_ops, .cable_detect = ata_cable_unknown, .set_mode = generic_set_mode, diff -urNp linux-2.6.32.48/drivers/ata/ata_piix.c linux-2.6.32.48-openvz/drivers/ata/ata_piix.c --- linux-2.6.32.48/drivers/ata/ata_piix.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/ata_piix.c 2011-11-17 20:08:53.000000000 -0500 @@ -318,7 +318,7 @@ static struct scsi_host_template piix_sh ATA_BMDMA_SHT(DRV_NAME), }; -static struct ata_port_operations piix_pata_ops = { +static const struct ata_port_operations piix_pata_ops = { .inherits = &ata_bmdma32_port_ops, .cable_detect = ata_cable_40wire, .set_piomode = piix_set_piomode, @@ -326,22 +326,22 @@ static struct ata_port_operations piix_p .prereset = piix_pata_prereset, }; -static struct ata_port_operations piix_vmw_ops = { +static const struct ata_port_operations piix_vmw_ops = { .inherits = &piix_pata_ops, .bmdma_status = piix_vmw_bmdma_status, }; -static struct ata_port_operations ich_pata_ops = { +static const struct ata_port_operations ich_pata_ops = { .inherits = &piix_pata_ops, .cable_detect = ich_pata_cable_detect, .set_dmamode = ich_set_dmamode, }; -static struct ata_port_operations piix_sata_ops = { +static const struct ata_port_operations piix_sata_ops = { .inherits = &ata_bmdma_port_ops, }; -static struct ata_port_operations piix_sidpr_sata_ops = { +static const struct ata_port_operations piix_sidpr_sata_ops = { .inherits = &piix_sata_ops, .hardreset = sata_std_hardreset, .scr_read = piix_sidpr_scr_read, diff -urNp linux-2.6.32.48/drivers/ata/libata-acpi.c linux-2.6.32.48-openvz/drivers/ata/libata-acpi.c --- linux-2.6.32.48/drivers/ata/libata-acpi.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/libata-acpi.c 2011-11-17 20:08:53.000000000 -0500 @@ -223,12 +223,12 @@ static void ata_acpi_dev_uevent(acpi_han ata_acpi_uevent(dev->link->ap, dev, event); } -static struct acpi_dock_ops ata_acpi_dev_dock_ops = { +static const struct acpi_dock_ops ata_acpi_dev_dock_ops = { .handler = ata_acpi_dev_notify_dock, .uevent = ata_acpi_dev_uevent, }; -static struct acpi_dock_ops ata_acpi_ap_dock_ops = { +static const struct acpi_dock_ops ata_acpi_ap_dock_ops = { .handler = ata_acpi_ap_notify_dock, .uevent = ata_acpi_ap_uevent, }; diff -urNp linux-2.6.32.48/drivers/ata/libata-core.c linux-2.6.32.48-openvz/drivers/ata/libata-core.c --- linux-2.6.32.48/drivers/ata/libata-core.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/libata-core.c 2011-11-17 20:09:04.000000000 -0500 @@ -4954,7 +4954,7 @@ void ata_qc_free(struct ata_queued_cmd * struct ata_port *ap; unsigned int tag; - WARN_ON_ONCE(qc == NULL); /* ata_qc_from_tag _might_ return NULL */ + BUG_ON(qc == NULL); /* ata_qc_from_tag _might_ return NULL */ ap = qc->ap; qc->flags = 0; @@ -4970,7 +4970,7 @@ void __ata_qc_complete(struct ata_queued struct ata_port *ap; struct ata_link *link; - WARN_ON_ONCE(qc == NULL); /* ata_qc_from_tag _might_ return NULL */ + BUG_ON(qc == NULL); /* ata_qc_from_tag _might_ return NULL */ WARN_ON_ONCE(!(qc->flags & ATA_QCFLAG_ACTIVE)); ap = qc->ap; link = qc->dev->link; @@ -5987,7 +5987,7 @@ static void ata_host_stop(struct device * LOCKING: * None. */ -static void ata_finalize_port_ops(struct ata_port_operations *ops) +static void ata_finalize_port_ops(const struct ata_port_operations *ops) { static DEFINE_SPINLOCK(lock); const struct ata_port_operations *cur; @@ -5999,6 +5999,7 @@ static void ata_finalize_port_ops(struct return; spin_lock(&lock); + pax_open_kernel(); for (cur = ops->inherits; cur; cur = cur->inherits) { void **inherit = (void **)cur; @@ -6012,8 +6013,9 @@ static void ata_finalize_port_ops(struct if (IS_ERR(*pp)) *pp = NULL; - ops->inherits = NULL; + *(struct ata_port_operations **)&ops->inherits = NULL; + pax_close_kernel(); spin_unlock(&lock); } @@ -6110,7 +6112,7 @@ int ata_host_start(struct ata_host *host */ /* KILLME - the only user left is ipr */ void ata_host_init(struct ata_host *host, struct device *dev, - unsigned long flags, struct ata_port_operations *ops) + unsigned long flags, const struct ata_port_operations *ops) { spin_lock_init(&host->lock); host->dev = dev; @@ -6773,7 +6775,7 @@ static void ata_dummy_error_handler(stru /* truly dummy */ } -struct ata_port_operations ata_dummy_port_ops = { +const struct ata_port_operations ata_dummy_port_ops = { .qc_prep = ata_noop_qc_prep, .qc_issue = ata_dummy_qc_issue, .error_handler = ata_dummy_error_handler, diff -urNp linux-2.6.32.48/drivers/ata/libata-eh.c linux-2.6.32.48-openvz/drivers/ata/libata-eh.c --- linux-2.6.32.48/drivers/ata/libata-eh.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/libata-eh.c 2011-11-17 20:09:04.000000000 -0500 @@ -2423,6 +2423,8 @@ void ata_eh_report(struct ata_port *ap) { struct ata_link *link; + pax_track_stack(); + ata_for_each_link(link, ap, HOST_FIRST) ata_eh_link_report(link); } @@ -3594,7 +3596,7 @@ void ata_do_eh(struct ata_port *ap, ata_ */ void ata_std_error_handler(struct ata_port *ap) { - struct ata_port_operations *ops = ap->ops; + const struct ata_port_operations *ops = ap->ops; ata_reset_fn_t hardreset = ops->hardreset; /* ignore built-in hardreset if SCR access is not available */ diff -urNp linux-2.6.32.48/drivers/ata/libata-pmp.c linux-2.6.32.48-openvz/drivers/ata/libata-pmp.c --- linux-2.6.32.48/drivers/ata/libata-pmp.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/libata-pmp.c 2011-11-17 20:08:53.000000000 -0500 @@ -841,7 +841,7 @@ static int sata_pmp_handle_link_fail(str */ static int sata_pmp_eh_recover(struct ata_port *ap) { - struct ata_port_operations *ops = ap->ops; + const struct ata_port_operations *ops = ap->ops; int pmp_tries, link_tries[SATA_PMP_MAX_PORTS]; struct ata_link *pmp_link = &ap->link; struct ata_device *pmp_dev = pmp_link->device; diff -urNp linux-2.6.32.48/drivers/ata/pata_acpi.c linux-2.6.32.48-openvz/drivers/ata/pata_acpi.c --- linux-2.6.32.48/drivers/ata/pata_acpi.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_acpi.c 2011-11-17 20:08:53.000000000 -0500 @@ -215,7 +215,7 @@ static struct scsi_host_template pacpi_s ATA_BMDMA_SHT(DRV_NAME), }; -static struct ata_port_operations pacpi_ops = { +static const struct ata_port_operations pacpi_ops = { .inherits = &ata_bmdma_port_ops, .qc_issue = pacpi_qc_issue, .cable_detect = pacpi_cable_detect, diff -urNp linux-2.6.32.48/drivers/ata/pata_ali.c linux-2.6.32.48-openvz/drivers/ata/pata_ali.c --- linux-2.6.32.48/drivers/ata/pata_ali.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_ali.c 2011-11-17 20:08:53.000000000 -0500 @@ -365,7 +365,7 @@ static struct scsi_host_template ali_sht * Port operations for PIO only ALi */ -static struct ata_port_operations ali_early_port_ops = { +static const struct ata_port_operations ali_early_port_ops = { .inherits = &ata_sff_port_ops, .cable_detect = ata_cable_40wire, .set_piomode = ali_set_piomode, @@ -382,7 +382,7 @@ static const struct ata_port_operations * Port operations for DMA capable ALi without cable * detect */ -static struct ata_port_operations ali_20_port_ops = { +static const struct ata_port_operations ali_20_port_ops = { .inherits = &ali_dma_base_ops, .cable_detect = ata_cable_40wire, .mode_filter = ali_20_filter, @@ -393,7 +393,7 @@ static struct ata_port_operations ali_20 /* * Port operations for DMA capable ALi with cable detect */ -static struct ata_port_operations ali_c2_port_ops = { +static const struct ata_port_operations ali_c2_port_ops = { .inherits = &ali_dma_base_ops, .check_atapi_dma = ali_check_atapi_dma, .cable_detect = ali_c2_cable_detect, @@ -404,7 +404,7 @@ static struct ata_port_operations ali_c2 /* * Port operations for DMA capable ALi with cable detect */ -static struct ata_port_operations ali_c4_port_ops = { +static const struct ata_port_operations ali_c4_port_ops = { .inherits = &ali_dma_base_ops, .check_atapi_dma = ali_check_atapi_dma, .cable_detect = ali_c2_cable_detect, @@ -414,7 +414,7 @@ static struct ata_port_operations ali_c4 /* * Port operations for DMA capable ALi with cable detect and LBA48 */ -static struct ata_port_operations ali_c5_port_ops = { +static const struct ata_port_operations ali_c5_port_ops = { .inherits = &ali_dma_base_ops, .check_atapi_dma = ali_check_atapi_dma, .dev_config = ali_warn_atapi_dma, diff -urNp linux-2.6.32.48/drivers/ata/pata_amd.c linux-2.6.32.48-openvz/drivers/ata/pata_amd.c --- linux-2.6.32.48/drivers/ata/pata_amd.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_amd.c 2011-11-17 20:08:53.000000000 -0500 @@ -397,28 +397,28 @@ static const struct ata_port_operations .prereset = amd_pre_reset, }; -static struct ata_port_operations amd33_port_ops = { +static const struct ata_port_operations amd33_port_ops = { .inherits = &amd_base_port_ops, .cable_detect = ata_cable_40wire, .set_piomode = amd33_set_piomode, .set_dmamode = amd33_set_dmamode, }; -static struct ata_port_operations amd66_port_ops = { +static const struct ata_port_operations amd66_port_ops = { .inherits = &amd_base_port_ops, .cable_detect = ata_cable_unknown, .set_piomode = amd66_set_piomode, .set_dmamode = amd66_set_dmamode, }; -static struct ata_port_operations amd100_port_ops = { +static const struct ata_port_operations amd100_port_ops = { .inherits = &amd_base_port_ops, .cable_detect = ata_cable_unknown, .set_piomode = amd100_set_piomode, .set_dmamode = amd100_set_dmamode, }; -static struct ata_port_operations amd133_port_ops = { +static const struct ata_port_operations amd133_port_ops = { .inherits = &amd_base_port_ops, .cable_detect = amd_cable_detect, .set_piomode = amd133_set_piomode, @@ -433,13 +433,13 @@ static const struct ata_port_operations .host_stop = nv_host_stop, }; -static struct ata_port_operations nv100_port_ops = { +static const struct ata_port_operations nv100_port_ops = { .inherits = &nv_base_port_ops, .set_piomode = nv100_set_piomode, .set_dmamode = nv100_set_dmamode, }; -static struct ata_port_operations nv133_port_ops = { +static const struct ata_port_operations nv133_port_ops = { .inherits = &nv_base_port_ops, .set_piomode = nv133_set_piomode, .set_dmamode = nv133_set_dmamode, diff -urNp linux-2.6.32.48/drivers/ata/pata_artop.c linux-2.6.32.48-openvz/drivers/ata/pata_artop.c --- linux-2.6.32.48/drivers/ata/pata_artop.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_artop.c 2011-11-17 20:08:53.000000000 -0500 @@ -311,7 +311,7 @@ static struct scsi_host_template artop_s ATA_BMDMA_SHT(DRV_NAME), }; -static struct ata_port_operations artop6210_ops = { +static const struct ata_port_operations artop6210_ops = { .inherits = &ata_bmdma_port_ops, .cable_detect = ata_cable_40wire, .set_piomode = artop6210_set_piomode, @@ -320,7 +320,7 @@ static struct ata_port_operations artop6 .qc_defer = artop6210_qc_defer, }; -static struct ata_port_operations artop6260_ops = { +static const struct ata_port_operations artop6260_ops = { .inherits = &ata_bmdma_port_ops, .cable_detect = artop6260_cable_detect, .set_piomode = artop6260_set_piomode, diff -urNp linux-2.6.32.48/drivers/ata/pata_at32.c linux-2.6.32.48-openvz/drivers/ata/pata_at32.c --- linux-2.6.32.48/drivers/ata/pata_at32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_at32.c 2011-11-17 20:08:53.000000000 -0500 @@ -172,7 +172,7 @@ static struct scsi_host_template at32_sh ATA_PIO_SHT(DRV_NAME), }; -static struct ata_port_operations at32_port_ops = { +static const struct ata_port_operations at32_port_ops = { .inherits = &ata_sff_port_ops, .cable_detect = ata_cable_40wire, .set_piomode = pata_at32_set_piomode, diff -urNp linux-2.6.32.48/drivers/ata/pata_at91.c linux-2.6.32.48-openvz/drivers/ata/pata_at91.c --- linux-2.6.32.48/drivers/ata/pata_at91.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_at91.c 2011-11-17 20:08:53.000000000 -0500 @@ -195,7 +195,7 @@ static struct scsi_host_template pata_at ATA_PIO_SHT(DRV_NAME), }; -static struct ata_port_operations pata_at91_port_ops = { +static const struct ata_port_operations pata_at91_port_ops = { .inherits = &ata_sff_port_ops, .sff_data_xfer = pata_at91_data_xfer_noirq, diff -urNp linux-2.6.32.48/drivers/ata/pata_atiixp.c linux-2.6.32.48-openvz/drivers/ata/pata_atiixp.c --- linux-2.6.32.48/drivers/ata/pata_atiixp.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_atiixp.c 2011-11-17 20:08:53.000000000 -0500 @@ -205,7 +205,7 @@ static struct scsi_host_template atiixp_ .sg_tablesize = LIBATA_DUMB_MAX_PRD, }; -static struct ata_port_operations atiixp_port_ops = { +static const struct ata_port_operations atiixp_port_ops = { .inherits = &ata_bmdma_port_ops, .qc_prep = ata_sff_dumb_qc_prep, diff -urNp linux-2.6.32.48/drivers/ata/pata_atp867x.c linux-2.6.32.48-openvz/drivers/ata/pata_atp867x.c --- linux-2.6.32.48/drivers/ata/pata_atp867x.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_atp867x.c 2011-11-17 20:08:53.000000000 -0500 @@ -274,7 +274,7 @@ static struct scsi_host_template atp867x ATA_BMDMA_SHT(DRV_NAME), }; -static struct ata_port_operations atp867x_ops = { +static const struct ata_port_operations atp867x_ops = { .inherits = &ata_bmdma_port_ops, .cable_detect = atp867x_cable_detect, .set_piomode = atp867x_set_piomode, diff -urNp linux-2.6.32.48/drivers/ata/pata_bf54x.c linux-2.6.32.48-openvz/drivers/ata/pata_bf54x.c --- linux-2.6.32.48/drivers/ata/pata_bf54x.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_bf54x.c 2011-11-17 20:08:53.000000000 -0500 @@ -1464,7 +1464,7 @@ static struct scsi_host_template bfin_sh .dma_boundary = ATA_DMA_BOUNDARY, }; -static struct ata_port_operations bfin_pata_ops = { +static const struct ata_port_operations bfin_pata_ops = { .inherits = &ata_sff_port_ops, .set_piomode = bfin_set_piomode, diff -urNp linux-2.6.32.48/drivers/ata/pata_cmd640.c linux-2.6.32.48-openvz/drivers/ata/pata_cmd640.c --- linux-2.6.32.48/drivers/ata/pata_cmd640.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_cmd640.c 2011-11-17 20:08:53.000000000 -0500 @@ -168,7 +168,7 @@ static struct scsi_host_template cmd640_ ATA_BMDMA_SHT(DRV_NAME), }; -static struct ata_port_operations cmd640_port_ops = { +static const struct ata_port_operations cmd640_port_ops = { .inherits = &ata_bmdma_port_ops, /* In theory xfer_noirq is not needed once we kill the prefetcher */ .sff_data_xfer = ata_sff_data_xfer_noirq, diff -urNp linux-2.6.32.48/drivers/ata/pata_cmd64x.c linux-2.6.32.48-openvz/drivers/ata/pata_cmd64x.c --- linux-2.6.32.48/drivers/ata/pata_cmd64x.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_cmd64x.c 2011-11-17 20:09:04.000000000 -0500 @@ -271,18 +271,18 @@ static const struct ata_port_operations .set_dmamode = cmd64x_set_dmamode, }; -static struct ata_port_operations cmd64x_port_ops = { +static const struct ata_port_operations cmd64x_port_ops = { .inherits = &cmd64x_base_ops, .cable_detect = ata_cable_40wire, }; -static struct ata_port_operations cmd646r1_port_ops = { +static const struct ata_port_operations cmd646r1_port_ops = { .inherits = &cmd64x_base_ops, .bmdma_stop = cmd646r1_bmdma_stop, .cable_detect = ata_cable_40wire, }; -static struct ata_port_operations cmd648_port_ops = { +static const struct ata_port_operations cmd648_port_ops = { .inherits = &cmd64x_base_ops, .bmdma_stop = cmd648_bmdma_stop, .cable_detect = cmd648_cable_detect, diff -urNp linux-2.6.32.48/drivers/ata/pata_cs5520.c linux-2.6.32.48-openvz/drivers/ata/pata_cs5520.c --- linux-2.6.32.48/drivers/ata/pata_cs5520.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_cs5520.c 2011-11-17 20:08:53.000000000 -0500 @@ -144,7 +144,7 @@ static struct scsi_host_template cs5520_ .sg_tablesize = LIBATA_DUMB_MAX_PRD, }; -static struct ata_port_operations cs5520_port_ops = { +static const struct ata_port_operations cs5520_port_ops = { .inherits = &ata_bmdma_port_ops, .qc_prep = ata_sff_dumb_qc_prep, .cable_detect = ata_cable_40wire, diff -urNp linux-2.6.32.48/drivers/ata/pata_cs5530.c linux-2.6.32.48-openvz/drivers/ata/pata_cs5530.c --- linux-2.6.32.48/drivers/ata/pata_cs5530.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_cs5530.c 2011-11-17 20:08:53.000000000 -0500 @@ -164,7 +164,7 @@ static struct scsi_host_template cs5530_ .sg_tablesize = LIBATA_DUMB_MAX_PRD, }; -static struct ata_port_operations cs5530_port_ops = { +static const struct ata_port_operations cs5530_port_ops = { .inherits = &ata_bmdma_port_ops, .qc_prep = ata_sff_dumb_qc_prep, diff -urNp linux-2.6.32.48/drivers/ata/pata_cs5535.c linux-2.6.32.48-openvz/drivers/ata/pata_cs5535.c --- linux-2.6.32.48/drivers/ata/pata_cs5535.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_cs5535.c 2011-11-17 20:08:53.000000000 -0500 @@ -160,7 +160,7 @@ static struct scsi_host_template cs5535_ ATA_BMDMA_SHT(DRV_NAME), }; -static struct ata_port_operations cs5535_port_ops = { +static const struct ata_port_operations cs5535_port_ops = { .inherits = &ata_bmdma_port_ops, .cable_detect = cs5535_cable_detect, .set_piomode = cs5535_set_piomode, diff -urNp linux-2.6.32.48/drivers/ata/pata_cs5536.c linux-2.6.32.48-openvz/drivers/ata/pata_cs5536.c --- linux-2.6.32.48/drivers/ata/pata_cs5536.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_cs5536.c 2011-11-17 20:08:53.000000000 -0500 @@ -223,7 +223,7 @@ static struct scsi_host_template cs5536_ ATA_BMDMA_SHT(DRV_NAME), }; -static struct ata_port_operations cs5536_port_ops = { +static const struct ata_port_operations cs5536_port_ops = { .inherits = &ata_bmdma_port_ops, .cable_detect = cs5536_cable_detect, .set_piomode = cs5536_set_piomode, diff -urNp linux-2.6.32.48/drivers/ata/pata_cypress.c linux-2.6.32.48-openvz/drivers/ata/pata_cypress.c --- linux-2.6.32.48/drivers/ata/pata_cypress.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_cypress.c 2011-11-17 20:08:53.000000000 -0500 @@ -113,7 +113,7 @@ static struct scsi_host_template cy82c69 ATA_BMDMA_SHT(DRV_NAME), }; -static struct ata_port_operations cy82c693_port_ops = { +static const struct ata_port_operations cy82c693_port_ops = { .inherits = &ata_bmdma_port_ops, .cable_detect = ata_cable_40wire, .set_piomode = cy82c693_set_piomode, diff -urNp linux-2.6.32.48/drivers/ata/pata_efar.c linux-2.6.32.48-openvz/drivers/ata/pata_efar.c --- linux-2.6.32.48/drivers/ata/pata_efar.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_efar.c 2011-11-17 20:08:53.000000000 -0500 @@ -222,7 +222,7 @@ static struct scsi_host_template efar_sh ATA_BMDMA_SHT(DRV_NAME), }; -static struct ata_port_operations efar_ops = { +static const struct ata_port_operations efar_ops = { .inherits = &ata_bmdma_port_ops, .cable_detect = efar_cable_detect, .set_piomode = efar_set_piomode, diff -urNp linux-2.6.32.48/drivers/ata/pata_hpt366.c linux-2.6.32.48-openvz/drivers/ata/pata_hpt366.c --- linux-2.6.32.48/drivers/ata/pata_hpt366.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_hpt366.c 2011-11-17 20:09:04.000000000 -0500 @@ -282,7 +282,7 @@ static struct scsi_host_template hpt36x_ * Configuration for HPT366/68 */ -static struct ata_port_operations hpt366_port_ops = { +static const struct ata_port_operations hpt366_port_ops = { .inherits = &ata_bmdma_port_ops, .cable_detect = hpt36x_cable_detect, .mode_filter = hpt366_filter, diff -urNp linux-2.6.32.48/drivers/ata/pata_hpt37x.c linux-2.6.32.48-openvz/drivers/ata/pata_hpt37x.c --- linux-2.6.32.48/drivers/ata/pata_hpt37x.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_hpt37x.c 2011-11-17 20:09:04.000000000 -0500 @@ -576,7 +576,7 @@ static struct scsi_host_template hpt37x_ * Configuration for HPT370 */ -static struct ata_port_operations hpt370_port_ops = { +static const struct ata_port_operations hpt370_port_ops = { .inherits = &ata_bmdma_port_ops, .bmdma_stop = hpt370_bmdma_stop, @@ -591,7 +591,7 @@ static struct ata_port_operations hpt370 * Configuration for HPT370A. Close to 370 but less filters */ -static struct ata_port_operations hpt370a_port_ops = { +static const struct ata_port_operations hpt370a_port_ops = { .inherits = &hpt370_port_ops, .mode_filter = hpt370a_filter, }; @@ -601,7 +601,7 @@ static struct ata_port_operations hpt370 * and DMA mode setting functionality. */ -static struct ata_port_operations hpt372_port_ops = { +static const struct ata_port_operations hpt372_port_ops = { .inherits = &ata_bmdma_port_ops, .bmdma_stop = hpt37x_bmdma_stop, @@ -616,7 +616,7 @@ static struct ata_port_operations hpt372 * but we have a different cable detection procedure for function 1. */ -static struct ata_port_operations hpt374_fn1_port_ops = { +static const struct ata_port_operations hpt374_fn1_port_ops = { .inherits = &hpt372_port_ops, .prereset = hpt374_fn1_pre_reset, }; diff -urNp linux-2.6.32.48/drivers/ata/pata_hpt3x2n.c linux-2.6.32.48-openvz/drivers/ata/pata_hpt3x2n.c --- linux-2.6.32.48/drivers/ata/pata_hpt3x2n.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_hpt3x2n.c 2011-11-17 20:09:04.000000000 -0500 @@ -337,7 +337,7 @@ static struct scsi_host_template hpt3x2n * Configuration for HPT3x2n. */ -static struct ata_port_operations hpt3x2n_port_ops = { +static const struct ata_port_operations hpt3x2n_port_ops = { .inherits = &ata_bmdma_port_ops, .bmdma_stop = hpt3x2n_bmdma_stop, diff -urNp linux-2.6.32.48/drivers/ata/pata_hpt3x3.c linux-2.6.32.48-openvz/drivers/ata/pata_hpt3x3.c --- linux-2.6.32.48/drivers/ata/pata_hpt3x3.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_hpt3x3.c 2011-11-17 20:08:53.000000000 -0500 @@ -141,7 +141,7 @@ static struct scsi_host_template hpt3x3_ ATA_BMDMA_SHT(DRV_NAME), }; -static struct ata_port_operations hpt3x3_port_ops = { +static const struct ata_port_operations hpt3x3_port_ops = { .inherits = &ata_bmdma_port_ops, .cable_detect = ata_cable_40wire, .set_piomode = hpt3x3_set_piomode, diff -urNp linux-2.6.32.48/drivers/ata/pata_icside.c linux-2.6.32.48-openvz/drivers/ata/pata_icside.c --- linux-2.6.32.48/drivers/ata/pata_icside.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_icside.c 2011-11-17 20:08:53.000000000 -0500 @@ -319,7 +319,7 @@ static void pata_icside_postreset(struct } } -static struct ata_port_operations pata_icside_port_ops = { +static const struct ata_port_operations pata_icside_port_ops = { .inherits = &ata_sff_port_ops, /* no need to build any PRD tables for DMA */ .qc_prep = ata_noop_qc_prep, diff -urNp linux-2.6.32.48/drivers/ata/pata_isapnp.c linux-2.6.32.48-openvz/drivers/ata/pata_isapnp.c --- linux-2.6.32.48/drivers/ata/pata_isapnp.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_isapnp.c 2011-11-17 20:08:53.000000000 -0500 @@ -23,12 +23,12 @@ static struct scsi_host_template isapnp_ ATA_PIO_SHT(DRV_NAME), }; -static struct ata_port_operations isapnp_port_ops = { +static const struct ata_port_operations isapnp_port_ops = { .inherits = &ata_sff_port_ops, .cable_detect = ata_cable_40wire, }; -static struct ata_port_operations isapnp_noalt_port_ops = { +static const struct ata_port_operations isapnp_noalt_port_ops = { .inherits = &ata_sff_port_ops, .cable_detect = ata_cable_40wire, /* No altstatus so we don't want to use the lost interrupt poll */ diff -urNp linux-2.6.32.48/drivers/ata/pata_it8213.c linux-2.6.32.48-openvz/drivers/ata/pata_it8213.c --- linux-2.6.32.48/drivers/ata/pata_it8213.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_it8213.c 2011-11-17 20:08:53.000000000 -0500 @@ -234,7 +234,7 @@ static struct scsi_host_template it8213_ }; -static struct ata_port_operations it8213_ops = { +static const struct ata_port_operations it8213_ops = { .inherits = &ata_bmdma_port_ops, .cable_detect = it8213_cable_detect, .set_piomode = it8213_set_piomode, diff -urNp linux-2.6.32.48/drivers/ata/pata_it821x.c linux-2.6.32.48-openvz/drivers/ata/pata_it821x.c --- linux-2.6.32.48/drivers/ata/pata_it821x.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_it821x.c 2011-11-17 20:08:53.000000000 -0500 @@ -800,7 +800,7 @@ static struct scsi_host_template it821x_ ATA_BMDMA_SHT(DRV_NAME), }; -static struct ata_port_operations it821x_smart_port_ops = { +static const struct ata_port_operations it821x_smart_port_ops = { .inherits = &ata_bmdma_port_ops, .check_atapi_dma= it821x_check_atapi_dma, @@ -814,7 +814,7 @@ static struct ata_port_operations it821x .port_start = it821x_port_start, }; -static struct ata_port_operations it821x_passthru_port_ops = { +static const struct ata_port_operations it821x_passthru_port_ops = { .inherits = &ata_bmdma_port_ops, .check_atapi_dma= it821x_check_atapi_dma, @@ -830,7 +830,7 @@ static struct ata_port_operations it821x .port_start = it821x_port_start, }; -static struct ata_port_operations it821x_rdc_port_ops = { +static const struct ata_port_operations it821x_rdc_port_ops = { .inherits = &ata_bmdma_port_ops, .check_atapi_dma= it821x_check_atapi_dma, diff -urNp linux-2.6.32.48/drivers/ata/pata_ixp4xx_cf.c linux-2.6.32.48-openvz/drivers/ata/pata_ixp4xx_cf.c --- linux-2.6.32.48/drivers/ata/pata_ixp4xx_cf.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_ixp4xx_cf.c 2011-11-17 20:08:53.000000000 -0500 @@ -89,7 +89,7 @@ static struct scsi_host_template ixp4xx_ ATA_PIO_SHT(DRV_NAME), }; -static struct ata_port_operations ixp4xx_port_ops = { +static const struct ata_port_operations ixp4xx_port_ops = { .inherits = &ata_sff_port_ops, .sff_data_xfer = ixp4xx_mmio_data_xfer, .cable_detect = ata_cable_40wire, diff -urNp linux-2.6.32.48/drivers/ata/pata_jmicron.c linux-2.6.32.48-openvz/drivers/ata/pata_jmicron.c --- linux-2.6.32.48/drivers/ata/pata_jmicron.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_jmicron.c 2011-11-17 20:08:53.000000000 -0500 @@ -111,7 +111,7 @@ static struct scsi_host_template jmicron ATA_BMDMA_SHT(DRV_NAME), }; -static struct ata_port_operations jmicron_ops = { +static const struct ata_port_operations jmicron_ops = { .inherits = &ata_bmdma_port_ops, .prereset = jmicron_pre_reset, }; diff -urNp linux-2.6.32.48/drivers/ata/pata_legacy.c linux-2.6.32.48-openvz/drivers/ata/pata_legacy.c --- linux-2.6.32.48/drivers/ata/pata_legacy.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_legacy.c 2011-11-17 20:08:53.000000000 -0500 @@ -106,7 +106,7 @@ struct legacy_probe { struct legacy_controller { const char *name; - struct ata_port_operations *ops; + const struct ata_port_operations *ops; unsigned int pio_mask; unsigned int flags; unsigned int pflags; @@ -223,12 +223,12 @@ static const struct ata_port_operations * pio_mask as well. */ -static struct ata_port_operations simple_port_ops = { +static const struct ata_port_operations simple_port_ops = { .inherits = &legacy_base_port_ops, .sff_data_xfer = ata_sff_data_xfer_noirq, }; -static struct ata_port_operations legacy_port_ops = { +static const struct ata_port_operations legacy_port_ops = { .inherits = &legacy_base_port_ops, .sff_data_xfer = ata_sff_data_xfer_noirq, .set_mode = legacy_set_mode, @@ -324,7 +324,7 @@ static unsigned int pdc_data_xfer_vlb(st return buflen; } -static struct ata_port_operations pdc20230_port_ops = { +static const struct ata_port_operations pdc20230_port_ops = { .inherits = &legacy_base_port_ops, .set_piomode = pdc20230_set_piomode, .sff_data_xfer = pdc_data_xfer_vlb, @@ -357,7 +357,7 @@ static void ht6560a_set_piomode(struct a ioread8(ap->ioaddr.status_addr); } -static struct ata_port_operations ht6560a_port_ops = { +static const struct ata_port_operations ht6560a_port_ops = { .inherits = &legacy_base_port_ops, .set_piomode = ht6560a_set_piomode, }; @@ -400,7 +400,7 @@ static void ht6560b_set_piomode(struct a ioread8(ap->ioaddr.status_addr); } -static struct ata_port_operations ht6560b_port_ops = { +static const struct ata_port_operations ht6560b_port_ops = { .inherits = &legacy_base_port_ops, .set_piomode = ht6560b_set_piomode, }; @@ -499,7 +499,7 @@ static void opti82c611a_set_piomode(stru } -static struct ata_port_operations opti82c611a_port_ops = { +static const struct ata_port_operations opti82c611a_port_ops = { .inherits = &legacy_base_port_ops, .set_piomode = opti82c611a_set_piomode, }; @@ -609,7 +609,7 @@ static unsigned int opti82c46x_qc_issue( return ata_sff_qc_issue(qc); } -static struct ata_port_operations opti82c46x_port_ops = { +static const struct ata_port_operations opti82c46x_port_ops = { .inherits = &legacy_base_port_ops, .set_piomode = opti82c46x_set_piomode, .qc_issue = opti82c46x_qc_issue, @@ -771,20 +771,20 @@ static int qdi_port(struct platform_devi return 0; } -static struct ata_port_operations qdi6500_port_ops = { +static const struct ata_port_operations qdi6500_port_ops = { .inherits = &legacy_base_port_ops, .set_piomode = qdi6500_set_piomode, .qc_issue = qdi_qc_issue, .sff_data_xfer = vlb32_data_xfer, }; -static struct ata_port_operations qdi6580_port_ops = { +static const struct ata_port_operations qdi6580_port_ops = { .inherits = &legacy_base_port_ops, .set_piomode = qdi6580_set_piomode, .sff_data_xfer = vlb32_data_xfer, }; -static struct ata_port_operations qdi6580dp_port_ops = { +static const struct ata_port_operations qdi6580dp_port_ops = { .inherits = &legacy_base_port_ops, .set_piomode = qdi6580dp_set_piomode, .sff_data_xfer = vlb32_data_xfer, @@ -855,7 +855,7 @@ static int winbond_port(struct platform_ return 0; } -static struct ata_port_operations winbond_port_ops = { +static const struct ata_port_operations winbond_port_ops = { .inherits = &legacy_base_port_ops, .set_piomode = winbond_set_piomode, .sff_data_xfer = vlb32_data_xfer, @@ -978,7 +978,7 @@ static __init int legacy_init_one(struct int pio_modes = controller->pio_mask; unsigned long io = probe->port; u32 mask = (1 << probe->slot); - struct ata_port_operations *ops = controller->ops; + const struct ata_port_operations *ops = controller->ops; struct legacy_data *ld = &legacy_data[probe->slot]; struct ata_host *host = NULL; struct ata_port *ap; diff -urNp linux-2.6.32.48/drivers/ata/pata_marvell.c linux-2.6.32.48-openvz/drivers/ata/pata_marvell.c --- linux-2.6.32.48/drivers/ata/pata_marvell.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_marvell.c 2011-11-17 20:08:53.000000000 -0500 @@ -100,7 +100,7 @@ static struct scsi_host_template marvell ATA_BMDMA_SHT(DRV_NAME), }; -static struct ata_port_operations marvell_ops = { +static const struct ata_port_operations marvell_ops = { .inherits = &ata_bmdma_port_ops, .cable_detect = marvell_cable_detect, .prereset = marvell_pre_reset, diff -urNp linux-2.6.32.48/drivers/ata/pata_mpc52xx.c linux-2.6.32.48-openvz/drivers/ata/pata_mpc52xx.c --- linux-2.6.32.48/drivers/ata/pata_mpc52xx.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_mpc52xx.c 2011-11-17 20:08:53.000000000 -0500 @@ -609,7 +609,7 @@ static struct scsi_host_template mpc52xx ATA_PIO_SHT(DRV_NAME), }; -static struct ata_port_operations mpc52xx_ata_port_ops = { +static const struct ata_port_operations mpc52xx_ata_port_ops = { .inherits = &ata_bmdma_port_ops, .sff_dev_select = mpc52xx_ata_dev_select, .set_piomode = mpc52xx_ata_set_piomode, diff -urNp linux-2.6.32.48/drivers/ata/pata_mpiix.c linux-2.6.32.48-openvz/drivers/ata/pata_mpiix.c --- linux-2.6.32.48/drivers/ata/pata_mpiix.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_mpiix.c 2011-11-17 20:08:53.000000000 -0500 @@ -140,7 +140,7 @@ static struct scsi_host_template mpiix_s ATA_PIO_SHT(DRV_NAME), }; -static struct ata_port_operations mpiix_port_ops = { +static const struct ata_port_operations mpiix_port_ops = { .inherits = &ata_sff_port_ops, .qc_issue = mpiix_qc_issue, .cable_detect = ata_cable_40wire, diff -urNp linux-2.6.32.48/drivers/ata/pata_netcell.c linux-2.6.32.48-openvz/drivers/ata/pata_netcell.c --- linux-2.6.32.48/drivers/ata/pata_netcell.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_netcell.c 2011-11-17 20:08:53.000000000 -0500 @@ -34,7 +34,7 @@ static struct scsi_host_template netcell ATA_BMDMA_SHT(DRV_NAME), }; -static struct ata_port_operations netcell_ops = { +static const struct ata_port_operations netcell_ops = { .inherits = &ata_bmdma_port_ops, .cable_detect = ata_cable_80wire, .read_id = netcell_read_id, diff -urNp linux-2.6.32.48/drivers/ata/pata_ninja32.c linux-2.6.32.48-openvz/drivers/ata/pata_ninja32.c --- linux-2.6.32.48/drivers/ata/pata_ninja32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_ninja32.c 2011-11-17 20:08:53.000000000 -0500 @@ -81,7 +81,7 @@ static struct scsi_host_template ninja32 ATA_BMDMA_SHT(DRV_NAME), }; -static struct ata_port_operations ninja32_port_ops = { +static const struct ata_port_operations ninja32_port_ops = { .inherits = &ata_bmdma_port_ops, .sff_dev_select = ninja32_dev_select, .cable_detect = ata_cable_40wire, diff -urNp linux-2.6.32.48/drivers/ata/pata_ns87410.c linux-2.6.32.48-openvz/drivers/ata/pata_ns87410.c --- linux-2.6.32.48/drivers/ata/pata_ns87410.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_ns87410.c 2011-11-17 20:08:53.000000000 -0500 @@ -132,7 +132,7 @@ static struct scsi_host_template ns87410 ATA_PIO_SHT(DRV_NAME), }; -static struct ata_port_operations ns87410_port_ops = { +static const struct ata_port_operations ns87410_port_ops = { .inherits = &ata_sff_port_ops, .qc_issue = ns87410_qc_issue, .cable_detect = ata_cable_40wire, diff -urNp linux-2.6.32.48/drivers/ata/pata_ns87415.c linux-2.6.32.48-openvz/drivers/ata/pata_ns87415.c --- linux-2.6.32.48/drivers/ata/pata_ns87415.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_ns87415.c 2011-11-17 20:08:53.000000000 -0500 @@ -299,7 +299,7 @@ static u8 ns87560_bmdma_status(struct at } #endif /* 87560 SuperIO Support */ -static struct ata_port_operations ns87415_pata_ops = { +static const struct ata_port_operations ns87415_pata_ops = { .inherits = &ata_bmdma_port_ops, .check_atapi_dma = ns87415_check_atapi_dma, @@ -313,7 +313,7 @@ static struct ata_port_operations ns8741 }; #if defined(CONFIG_SUPERIO) -static struct ata_port_operations ns87560_pata_ops = { +static const struct ata_port_operations ns87560_pata_ops = { .inherits = &ns87415_pata_ops, .sff_tf_read = ns87560_tf_read, .sff_check_status = ns87560_check_status, diff -urNp linux-2.6.32.48/drivers/ata/pata_octeon_cf.c linux-2.6.32.48-openvz/drivers/ata/pata_octeon_cf.c --- linux-2.6.32.48/drivers/ata/pata_octeon_cf.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_octeon_cf.c 2011-11-17 20:08:53.000000000 -0500 @@ -801,6 +801,7 @@ static unsigned int octeon_cf_qc_issue(s return 0; } +/* cannot be const */ static struct ata_port_operations octeon_cf_ops = { .inherits = &ata_sff_port_ops, .check_atapi_dma = octeon_cf_check_atapi_dma, diff -urNp linux-2.6.32.48/drivers/ata/pata_oldpiix.c linux-2.6.32.48-openvz/drivers/ata/pata_oldpiix.c --- linux-2.6.32.48/drivers/ata/pata_oldpiix.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_oldpiix.c 2011-11-17 20:08:53.000000000 -0500 @@ -208,7 +208,7 @@ static struct scsi_host_template oldpiix ATA_BMDMA_SHT(DRV_NAME), }; -static struct ata_port_operations oldpiix_pata_ops = { +static const struct ata_port_operations oldpiix_pata_ops = { .inherits = &ata_bmdma_port_ops, .qc_issue = oldpiix_qc_issue, .cable_detect = ata_cable_40wire, diff -urNp linux-2.6.32.48/drivers/ata/pata_opti.c linux-2.6.32.48-openvz/drivers/ata/pata_opti.c --- linux-2.6.32.48/drivers/ata/pata_opti.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_opti.c 2011-11-17 20:08:53.000000000 -0500 @@ -152,7 +152,7 @@ static struct scsi_host_template opti_sh ATA_PIO_SHT(DRV_NAME), }; -static struct ata_port_operations opti_port_ops = { +static const struct ata_port_operations opti_port_ops = { .inherits = &ata_sff_port_ops, .cable_detect = ata_cable_40wire, .set_piomode = opti_set_piomode, diff -urNp linux-2.6.32.48/drivers/ata/pata_optidma.c linux-2.6.32.48-openvz/drivers/ata/pata_optidma.c --- linux-2.6.32.48/drivers/ata/pata_optidma.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_optidma.c 2011-11-17 20:08:53.000000000 -0500 @@ -337,7 +337,7 @@ static struct scsi_host_template optidma ATA_BMDMA_SHT(DRV_NAME), }; -static struct ata_port_operations optidma_port_ops = { +static const struct ata_port_operations optidma_port_ops = { .inherits = &ata_bmdma_port_ops, .cable_detect = ata_cable_40wire, .set_piomode = optidma_set_pio_mode, @@ -346,7 +346,7 @@ static struct ata_port_operations optidm .prereset = optidma_pre_reset, }; -static struct ata_port_operations optiplus_port_ops = { +static const struct ata_port_operations optiplus_port_ops = { .inherits = &optidma_port_ops, .set_piomode = optiplus_set_pio_mode, .set_dmamode = optiplus_set_dma_mode, diff -urNp linux-2.6.32.48/drivers/ata/pata_palmld.c linux-2.6.32.48-openvz/drivers/ata/pata_palmld.c --- linux-2.6.32.48/drivers/ata/pata_palmld.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_palmld.c 2011-11-17 20:08:53.000000000 -0500 @@ -37,7 +37,7 @@ static struct scsi_host_template palmld_ ATA_PIO_SHT(DRV_NAME), }; -static struct ata_port_operations palmld_port_ops = { +static const struct ata_port_operations palmld_port_ops = { .inherits = &ata_sff_port_ops, .sff_data_xfer = ata_sff_data_xfer_noirq, .cable_detect = ata_cable_40wire, diff -urNp linux-2.6.32.48/drivers/ata/pata_pcmcia.c linux-2.6.32.48-openvz/drivers/ata/pata_pcmcia.c --- linux-2.6.32.48/drivers/ata/pata_pcmcia.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_pcmcia.c 2011-11-17 20:08:53.000000000 -0500 @@ -162,14 +162,14 @@ static struct scsi_host_template pcmcia_ ATA_PIO_SHT(DRV_NAME), }; -static struct ata_port_operations pcmcia_port_ops = { +static const struct ata_port_operations pcmcia_port_ops = { .inherits = &ata_sff_port_ops, .sff_data_xfer = ata_sff_data_xfer_noirq, .cable_detect = ata_cable_40wire, .set_mode = pcmcia_set_mode, }; -static struct ata_port_operations pcmcia_8bit_port_ops = { +static const struct ata_port_operations pcmcia_8bit_port_ops = { .inherits = &ata_sff_port_ops, .sff_data_xfer = ata_data_xfer_8bit, .cable_detect = ata_cable_40wire, @@ -256,7 +256,7 @@ static int pcmcia_init_one(struct pcmcia unsigned long io_base, ctl_base; void __iomem *io_addr, *ctl_addr; int n_ports = 1; - struct ata_port_operations *ops = &pcmcia_port_ops; + const struct ata_port_operations *ops = &pcmcia_port_ops; info = kzalloc(sizeof(*info), GFP_KERNEL); if (info == NULL) diff -urNp linux-2.6.32.48/drivers/ata/pata_pdc2027x.c linux-2.6.32.48-openvz/drivers/ata/pata_pdc2027x.c --- linux-2.6.32.48/drivers/ata/pata_pdc2027x.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_pdc2027x.c 2011-11-17 20:08:53.000000000 -0500 @@ -132,14 +132,14 @@ static struct scsi_host_template pdc2027 ATA_BMDMA_SHT(DRV_NAME), }; -static struct ata_port_operations pdc2027x_pata100_ops = { +static const struct ata_port_operations pdc2027x_pata100_ops = { .inherits = &ata_bmdma_port_ops, .check_atapi_dma = pdc2027x_check_atapi_dma, .cable_detect = pdc2027x_cable_detect, .prereset = pdc2027x_prereset, }; -static struct ata_port_operations pdc2027x_pata133_ops = { +static const struct ata_port_operations pdc2027x_pata133_ops = { .inherits = &pdc2027x_pata100_ops, .mode_filter = pdc2027x_mode_filter, .set_piomode = pdc2027x_set_piomode, diff -urNp linux-2.6.32.48/drivers/ata/pata_pdc202xx_old.c linux-2.6.32.48-openvz/drivers/ata/pata_pdc202xx_old.c --- linux-2.6.32.48/drivers/ata/pata_pdc202xx_old.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_pdc202xx_old.c 2011-11-17 20:08:53.000000000 -0500 @@ -274,7 +274,7 @@ static struct scsi_host_template pdc202x ATA_BMDMA_SHT(DRV_NAME), }; -static struct ata_port_operations pdc2024x_port_ops = { +static const struct ata_port_operations pdc2024x_port_ops = { .inherits = &ata_bmdma_port_ops, .cable_detect = ata_cable_40wire, @@ -284,7 +284,7 @@ static struct ata_port_operations pdc202 .sff_exec_command = pdc202xx_exec_command, }; -static struct ata_port_operations pdc2026x_port_ops = { +static const struct ata_port_operations pdc2026x_port_ops = { .inherits = &pdc2024x_port_ops, .check_atapi_dma = pdc2026x_check_atapi_dma, diff -urNp linux-2.6.32.48/drivers/ata/pata_platform.c linux-2.6.32.48-openvz/drivers/ata/pata_platform.c --- linux-2.6.32.48/drivers/ata/pata_platform.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_platform.c 2011-11-17 20:08:53.000000000 -0500 @@ -48,7 +48,7 @@ static struct scsi_host_template pata_pl ATA_PIO_SHT(DRV_NAME), }; -static struct ata_port_operations pata_platform_port_ops = { +static const struct ata_port_operations pata_platform_port_ops = { .inherits = &ata_sff_port_ops, .sff_data_xfer = ata_sff_data_xfer_noirq, .cable_detect = ata_cable_unknown, diff -urNp linux-2.6.32.48/drivers/ata/pata_qdi.c linux-2.6.32.48-openvz/drivers/ata/pata_qdi.c --- linux-2.6.32.48/drivers/ata/pata_qdi.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_qdi.c 2011-11-17 20:08:53.000000000 -0500 @@ -157,7 +157,7 @@ static struct scsi_host_template qdi_sht ATA_PIO_SHT(DRV_NAME), }; -static struct ata_port_operations qdi6500_port_ops = { +static const struct ata_port_operations qdi6500_port_ops = { .inherits = &ata_sff_port_ops, .qc_issue = qdi_qc_issue, .sff_data_xfer = qdi_data_xfer, @@ -165,7 +165,7 @@ static struct ata_port_operations qdi650 .set_piomode = qdi6500_set_piomode, }; -static struct ata_port_operations qdi6580_port_ops = { +static const struct ata_port_operations qdi6580_port_ops = { .inherits = &qdi6500_port_ops, .set_piomode = qdi6580_set_piomode, }; diff -urNp linux-2.6.32.48/drivers/ata/pata_radisys.c linux-2.6.32.48-openvz/drivers/ata/pata_radisys.c --- linux-2.6.32.48/drivers/ata/pata_radisys.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_radisys.c 2011-11-17 20:08:53.000000000 -0500 @@ -187,7 +187,7 @@ static struct scsi_host_template radisys ATA_BMDMA_SHT(DRV_NAME), }; -static struct ata_port_operations radisys_pata_ops = { +static const struct ata_port_operations radisys_pata_ops = { .inherits = &ata_bmdma_port_ops, .qc_issue = radisys_qc_issue, .cable_detect = ata_cable_unknown, diff -urNp linux-2.6.32.48/drivers/ata/pata_rb532_cf.c linux-2.6.32.48-openvz/drivers/ata/pata_rb532_cf.c --- linux-2.6.32.48/drivers/ata/pata_rb532_cf.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_rb532_cf.c 2011-11-17 20:08:53.000000000 -0500 @@ -68,7 +68,7 @@ static irqreturn_t rb532_pata_irq_handle return IRQ_HANDLED; } -static struct ata_port_operations rb532_pata_port_ops = { +static const struct ata_port_operations rb532_pata_port_ops = { .inherits = &ata_sff_port_ops, .sff_data_xfer = ata_sff_data_xfer32, }; diff -urNp linux-2.6.32.48/drivers/ata/pata_rdc.c linux-2.6.32.48-openvz/drivers/ata/pata_rdc.c --- linux-2.6.32.48/drivers/ata/pata_rdc.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_rdc.c 2011-11-17 20:08:53.000000000 -0500 @@ -272,7 +272,7 @@ static void rdc_set_dmamode(struct ata_p pci_write_config_byte(dev, 0x48, udma_enable); } -static struct ata_port_operations rdc_pata_ops = { +static const struct ata_port_operations rdc_pata_ops = { .inherits = &ata_bmdma32_port_ops, .cable_detect = rdc_pata_cable_detect, .set_piomode = rdc_set_piomode, diff -urNp linux-2.6.32.48/drivers/ata/pata_rz1000.c linux-2.6.32.48-openvz/drivers/ata/pata_rz1000.c --- linux-2.6.32.48/drivers/ata/pata_rz1000.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_rz1000.c 2011-11-17 20:08:53.000000000 -0500 @@ -54,7 +54,7 @@ static struct scsi_host_template rz1000_ ATA_PIO_SHT(DRV_NAME), }; -static struct ata_port_operations rz1000_port_ops = { +static const struct ata_port_operations rz1000_port_ops = { .inherits = &ata_sff_port_ops, .cable_detect = ata_cable_40wire, .set_mode = rz1000_set_mode, diff -urNp linux-2.6.32.48/drivers/ata/pata_sc1200.c linux-2.6.32.48-openvz/drivers/ata/pata_sc1200.c --- linux-2.6.32.48/drivers/ata/pata_sc1200.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_sc1200.c 2011-11-17 20:08:53.000000000 -0500 @@ -207,7 +207,7 @@ static struct scsi_host_template sc1200_ .sg_tablesize = LIBATA_DUMB_MAX_PRD, }; -static struct ata_port_operations sc1200_port_ops = { +static const struct ata_port_operations sc1200_port_ops = { .inherits = &ata_bmdma_port_ops, .qc_prep = ata_sff_dumb_qc_prep, .qc_issue = sc1200_qc_issue, diff -urNp linux-2.6.32.48/drivers/ata/pata_scc.c linux-2.6.32.48-openvz/drivers/ata/pata_scc.c --- linux-2.6.32.48/drivers/ata/pata_scc.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_scc.c 2011-11-17 20:08:53.000000000 -0500 @@ -965,7 +965,7 @@ static struct scsi_host_template scc_sht ATA_BMDMA_SHT(DRV_NAME), }; -static struct ata_port_operations scc_pata_ops = { +static const struct ata_port_operations scc_pata_ops = { .inherits = &ata_bmdma_port_ops, .set_piomode = scc_set_piomode, diff -urNp linux-2.6.32.48/drivers/ata/pata_sch.c linux-2.6.32.48-openvz/drivers/ata/pata_sch.c --- linux-2.6.32.48/drivers/ata/pata_sch.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_sch.c 2011-11-17 20:08:53.000000000 -0500 @@ -75,7 +75,7 @@ static struct scsi_host_template sch_sht ATA_BMDMA_SHT(DRV_NAME), }; -static struct ata_port_operations sch_pata_ops = { +static const struct ata_port_operations sch_pata_ops = { .inherits = &ata_bmdma_port_ops, .cable_detect = ata_cable_unknown, .set_piomode = sch_set_piomode, diff -urNp linux-2.6.32.48/drivers/ata/pata_serverworks.c linux-2.6.32.48-openvz/drivers/ata/pata_serverworks.c --- linux-2.6.32.48/drivers/ata/pata_serverworks.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_serverworks.c 2011-11-17 20:08:53.000000000 -0500 @@ -299,7 +299,7 @@ static struct scsi_host_template serverw ATA_BMDMA_SHT(DRV_NAME), }; -static struct ata_port_operations serverworks_osb4_port_ops = { +static const struct ata_port_operations serverworks_osb4_port_ops = { .inherits = &ata_bmdma_port_ops, .cable_detect = serverworks_cable_detect, .mode_filter = serverworks_osb4_filter, @@ -307,7 +307,7 @@ static struct ata_port_operations server .set_dmamode = serverworks_set_dmamode, }; -static struct ata_port_operations serverworks_csb_port_ops = { +static const struct ata_port_operations serverworks_csb_port_ops = { .inherits = &serverworks_osb4_port_ops, .mode_filter = serverworks_csb_filter, }; diff -urNp linux-2.6.32.48/drivers/ata/pata_sil680.c linux-2.6.32.48-openvz/drivers/ata/pata_sil680.c --- linux-2.6.32.48/drivers/ata/pata_sil680.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_sil680.c 2011-11-17 20:09:04.000000000 -0500 @@ -194,7 +194,7 @@ static struct scsi_host_template sil680_ ATA_BMDMA_SHT(DRV_NAME), }; -static struct ata_port_operations sil680_port_ops = { +static const struct ata_port_operations sil680_port_ops = { .inherits = &ata_bmdma32_port_ops, .cable_detect = sil680_cable_detect, .set_piomode = sil680_set_piomode, diff -urNp linux-2.6.32.48/drivers/ata/pata_sis.c linux-2.6.32.48-openvz/drivers/ata/pata_sis.c --- linux-2.6.32.48/drivers/ata/pata_sis.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_sis.c 2011-11-17 20:08:53.000000000 -0500 @@ -503,47 +503,47 @@ static struct scsi_host_template sis_sht ATA_BMDMA_SHT(DRV_NAME), }; -static struct ata_port_operations sis_133_for_sata_ops = { +static const struct ata_port_operations sis_133_for_sata_ops = { .inherits = &ata_bmdma_port_ops, .set_piomode = sis_133_set_piomode, .set_dmamode = sis_133_set_dmamode, .cable_detect = sis_133_cable_detect, }; -static struct ata_port_operations sis_base_ops = { +static const struct ata_port_operations sis_base_ops = { .inherits = &ata_bmdma_port_ops, .prereset = sis_pre_reset, }; -static struct ata_port_operations sis_133_ops = { +static const struct ata_port_operations sis_133_ops = { .inherits = &sis_base_ops, .set_piomode = sis_133_set_piomode, .set_dmamode = sis_133_set_dmamode, .cable_detect = sis_133_cable_detect, }; -static struct ata_port_operations sis_133_early_ops = { +static const struct ata_port_operations sis_133_early_ops = { .inherits = &sis_base_ops, .set_piomode = sis_100_set_piomode, .set_dmamode = sis_133_early_set_dmamode, .cable_detect = sis_66_cable_detect, }; -static struct ata_port_operations sis_100_ops = { +static const struct ata_port_operations sis_100_ops = { .inherits = &sis_base_ops, .set_piomode = sis_100_set_piomode, .set_dmamode = sis_100_set_dmamode, .cable_detect = sis_66_cable_detect, }; -static struct ata_port_operations sis_66_ops = { +static const struct ata_port_operations sis_66_ops = { .inherits = &sis_base_ops, .set_piomode = sis_old_set_piomode, .set_dmamode = sis_66_set_dmamode, .cable_detect = sis_66_cable_detect, }; -static struct ata_port_operations sis_old_ops = { +static const struct ata_port_operations sis_old_ops = { .inherits = &sis_base_ops, .set_piomode = sis_old_set_piomode, .set_dmamode = sis_old_set_dmamode, diff -urNp linux-2.6.32.48/drivers/ata/pata_sl82c105.c linux-2.6.32.48-openvz/drivers/ata/pata_sl82c105.c --- linux-2.6.32.48/drivers/ata/pata_sl82c105.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_sl82c105.c 2011-11-17 20:08:53.000000000 -0500 @@ -231,7 +231,7 @@ static struct scsi_host_template sl82c10 ATA_BMDMA_SHT(DRV_NAME), }; -static struct ata_port_operations sl82c105_port_ops = { +static const struct ata_port_operations sl82c105_port_ops = { .inherits = &ata_bmdma_port_ops, .qc_defer = sl82c105_qc_defer, .bmdma_start = sl82c105_bmdma_start, diff -urNp linux-2.6.32.48/drivers/ata/pata_triflex.c linux-2.6.32.48-openvz/drivers/ata/pata_triflex.c --- linux-2.6.32.48/drivers/ata/pata_triflex.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_triflex.c 2011-11-17 20:08:53.000000000 -0500 @@ -178,7 +178,7 @@ static struct scsi_host_template triflex ATA_BMDMA_SHT(DRV_NAME), }; -static struct ata_port_operations triflex_port_ops = { +static const struct ata_port_operations triflex_port_ops = { .inherits = &ata_bmdma_port_ops, .bmdma_start = triflex_bmdma_start, .bmdma_stop = triflex_bmdma_stop, diff -urNp linux-2.6.32.48/drivers/ata/pata_via.c linux-2.6.32.48-openvz/drivers/ata/pata_via.c --- linux-2.6.32.48/drivers/ata/pata_via.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_via.c 2011-11-17 20:08:53.000000000 -0500 @@ -419,7 +419,7 @@ static struct scsi_host_template via_sht ATA_BMDMA_SHT(DRV_NAME), }; -static struct ata_port_operations via_port_ops = { +static const struct ata_port_operations via_port_ops = { .inherits = &ata_bmdma_port_ops, .cable_detect = via_cable_detect, .set_piomode = via_set_piomode, @@ -429,7 +429,7 @@ static struct ata_port_operations via_po .port_start = via_port_start, }; -static struct ata_port_operations via_port_ops_noirq = { +static const struct ata_port_operations via_port_ops_noirq = { .inherits = &via_port_ops, .sff_data_xfer = ata_sff_data_xfer_noirq, }; diff -urNp linux-2.6.32.48/drivers/ata/pata_winbond.c linux-2.6.32.48-openvz/drivers/ata/pata_winbond.c --- linux-2.6.32.48/drivers/ata/pata_winbond.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pata_winbond.c 2011-11-17 20:08:53.000000000 -0500 @@ -125,7 +125,7 @@ static struct scsi_host_template winbond ATA_PIO_SHT(DRV_NAME), }; -static struct ata_port_operations winbond_port_ops = { +static const struct ata_port_operations winbond_port_ops = { .inherits = &ata_sff_port_ops, .sff_data_xfer = winbond_data_xfer, .cable_detect = ata_cable_40wire, diff -urNp linux-2.6.32.48/drivers/ata/pdc_adma.c linux-2.6.32.48-openvz/drivers/ata/pdc_adma.c --- linux-2.6.32.48/drivers/ata/pdc_adma.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/pdc_adma.c 2011-11-17 20:08:53.000000000 -0500 @@ -145,7 +145,7 @@ static struct scsi_host_template adma_at .dma_boundary = ADMA_DMA_BOUNDARY, }; -static struct ata_port_operations adma_ata_ops = { +static const struct ata_port_operations adma_ata_ops = { .inherits = &ata_sff_port_ops, .lost_interrupt = ATA_OP_NULL, diff -urNp linux-2.6.32.48/drivers/ata/sata_fsl.c linux-2.6.32.48-openvz/drivers/ata/sata_fsl.c --- linux-2.6.32.48/drivers/ata/sata_fsl.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/sata_fsl.c 2011-11-17 20:08:53.000000000 -0500 @@ -1258,7 +1258,7 @@ static struct scsi_host_template sata_fs .dma_boundary = ATA_DMA_BOUNDARY, }; -static struct ata_port_operations sata_fsl_ops = { +static const struct ata_port_operations sata_fsl_ops = { .inherits = &sata_pmp_port_ops, .qc_defer = ata_std_qc_defer, diff -urNp linux-2.6.32.48/drivers/ata/sata_inic162x.c linux-2.6.32.48-openvz/drivers/ata/sata_inic162x.c --- linux-2.6.32.48/drivers/ata/sata_inic162x.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/sata_inic162x.c 2011-11-17 20:08:53.000000000 -0500 @@ -721,7 +721,7 @@ static int inic_port_start(struct ata_po return 0; } -static struct ata_port_operations inic_port_ops = { +static const struct ata_port_operations inic_port_ops = { .inherits = &sata_port_ops, .check_atapi_dma = inic_check_atapi_dma, diff -urNp linux-2.6.32.48/drivers/ata/sata_mv.c linux-2.6.32.48-openvz/drivers/ata/sata_mv.c --- linux-2.6.32.48/drivers/ata/sata_mv.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/sata_mv.c 2011-11-17 20:08:53.000000000 -0500 @@ -656,7 +656,7 @@ static struct scsi_host_template mv6_sht .dma_boundary = MV_DMA_BOUNDARY, }; -static struct ata_port_operations mv5_ops = { +static const struct ata_port_operations mv5_ops = { .inherits = &ata_sff_port_ops, .lost_interrupt = ATA_OP_NULL, @@ -678,7 +678,7 @@ static struct ata_port_operations mv5_op .port_stop = mv_port_stop, }; -static struct ata_port_operations mv6_ops = { +static const struct ata_port_operations mv6_ops = { .inherits = &mv5_ops, .dev_config = mv6_dev_config, .scr_read = mv_scr_read, @@ -698,7 +698,7 @@ static struct ata_port_operations mv6_op .bmdma_status = mv_bmdma_status, }; -static struct ata_port_operations mv_iie_ops = { +static const struct ata_port_operations mv_iie_ops = { .inherits = &mv6_ops, .dev_config = ATA_OP_NULL, .qc_prep = mv_qc_prep_iie, diff -urNp linux-2.6.32.48/drivers/ata/sata_nv.c linux-2.6.32.48-openvz/drivers/ata/sata_nv.c --- linux-2.6.32.48/drivers/ata/sata_nv.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/sata_nv.c 2011-11-17 20:08:53.000000000 -0500 @@ -464,7 +464,7 @@ static struct scsi_host_template nv_swnc * cases. Define nv_hardreset() which only kicks in for post-boot * probing and use it for all variants. */ -static struct ata_port_operations nv_generic_ops = { +static const struct ata_port_operations nv_generic_ops = { .inherits = &ata_bmdma_port_ops, .lost_interrupt = ATA_OP_NULL, .scr_read = nv_scr_read, @@ -472,20 +472,20 @@ static struct ata_port_operations nv_gen .hardreset = nv_hardreset, }; -static struct ata_port_operations nv_nf2_ops = { +static const struct ata_port_operations nv_nf2_ops = { .inherits = &nv_generic_ops, .freeze = nv_nf2_freeze, .thaw = nv_nf2_thaw, }; -static struct ata_port_operations nv_ck804_ops = { +static const struct ata_port_operations nv_ck804_ops = { .inherits = &nv_generic_ops, .freeze = nv_ck804_freeze, .thaw = nv_ck804_thaw, .host_stop = nv_ck804_host_stop, }; -static struct ata_port_operations nv_adma_ops = { +static const struct ata_port_operations nv_adma_ops = { .inherits = &nv_ck804_ops, .check_atapi_dma = nv_adma_check_atapi_dma, @@ -509,7 +509,7 @@ static struct ata_port_operations nv_adm .host_stop = nv_adma_host_stop, }; -static struct ata_port_operations nv_swncq_ops = { +static const struct ata_port_operations nv_swncq_ops = { .inherits = &nv_generic_ops, .qc_defer = ata_std_qc_defer, diff -urNp linux-2.6.32.48/drivers/ata/sata_promise.c linux-2.6.32.48-openvz/drivers/ata/sata_promise.c --- linux-2.6.32.48/drivers/ata/sata_promise.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/sata_promise.c 2011-11-17 20:08:53.000000000 -0500 @@ -195,7 +195,7 @@ static const struct ata_port_operations .error_handler = pdc_error_handler, }; -static struct ata_port_operations pdc_sata_ops = { +static const struct ata_port_operations pdc_sata_ops = { .inherits = &pdc_common_ops, .cable_detect = pdc_sata_cable_detect, .freeze = pdc_sata_freeze, @@ -208,14 +208,14 @@ static struct ata_port_operations pdc_sa /* First-generation chips need a more restrictive ->check_atapi_dma op, and ->freeze/thaw that ignore the hotplug controls. */ -static struct ata_port_operations pdc_old_sata_ops = { +static const struct ata_port_operations pdc_old_sata_ops = { .inherits = &pdc_sata_ops, .freeze = pdc_freeze, .thaw = pdc_thaw, .check_atapi_dma = pdc_old_sata_check_atapi_dma, }; -static struct ata_port_operations pdc_pata_ops = { +static const struct ata_port_operations pdc_pata_ops = { .inherits = &pdc_common_ops, .cable_detect = pdc_pata_cable_detect, .freeze = pdc_freeze, diff -urNp linux-2.6.32.48/drivers/ata/sata_qstor.c linux-2.6.32.48-openvz/drivers/ata/sata_qstor.c --- linux-2.6.32.48/drivers/ata/sata_qstor.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/sata_qstor.c 2011-11-17 20:08:53.000000000 -0500 @@ -132,7 +132,7 @@ static struct scsi_host_template qs_ata_ .dma_boundary = QS_DMA_BOUNDARY, }; -static struct ata_port_operations qs_ata_ops = { +static const struct ata_port_operations qs_ata_ops = { .inherits = &ata_sff_port_ops, .check_atapi_dma = qs_check_atapi_dma, diff -urNp linux-2.6.32.48/drivers/ata/sata_sil24.c linux-2.6.32.48-openvz/drivers/ata/sata_sil24.c --- linux-2.6.32.48/drivers/ata/sata_sil24.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/sata_sil24.c 2011-11-17 20:08:53.000000000 -0500 @@ -388,7 +388,7 @@ static struct scsi_host_template sil24_s .dma_boundary = ATA_DMA_BOUNDARY, }; -static struct ata_port_operations sil24_ops = { +static const struct ata_port_operations sil24_ops = { .inherits = &sata_pmp_port_ops, .qc_defer = sil24_qc_defer, diff -urNp linux-2.6.32.48/drivers/ata/sata_sil.c linux-2.6.32.48-openvz/drivers/ata/sata_sil.c --- linux-2.6.32.48/drivers/ata/sata_sil.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/sata_sil.c 2011-11-17 20:08:53.000000000 -0500 @@ -182,7 +182,7 @@ static struct scsi_host_template sil_sht .sg_tablesize = ATA_MAX_PRD }; -static struct ata_port_operations sil_ops = { +static const struct ata_port_operations sil_ops = { .inherits = &ata_bmdma32_port_ops, .dev_config = sil_dev_config, .set_mode = sil_set_mode, diff -urNp linux-2.6.32.48/drivers/ata/sata_sis.c linux-2.6.32.48-openvz/drivers/ata/sata_sis.c --- linux-2.6.32.48/drivers/ata/sata_sis.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/sata_sis.c 2011-11-17 20:08:53.000000000 -0500 @@ -89,7 +89,7 @@ static struct scsi_host_template sis_sht ATA_BMDMA_SHT(DRV_NAME), }; -static struct ata_port_operations sis_ops = { +static const struct ata_port_operations sis_ops = { .inherits = &ata_bmdma_port_ops, .scr_read = sis_scr_read, .scr_write = sis_scr_write, diff -urNp linux-2.6.32.48/drivers/ata/sata_svw.c linux-2.6.32.48-openvz/drivers/ata/sata_svw.c --- linux-2.6.32.48/drivers/ata/sata_svw.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/sata_svw.c 2011-11-17 20:08:53.000000000 -0500 @@ -344,7 +344,7 @@ static struct scsi_host_template k2_sata }; -static struct ata_port_operations k2_sata_ops = { +static const struct ata_port_operations k2_sata_ops = { .inherits = &ata_bmdma_port_ops, .sff_tf_load = k2_sata_tf_load, .sff_tf_read = k2_sata_tf_read, diff -urNp linux-2.6.32.48/drivers/ata/sata_sx4.c linux-2.6.32.48-openvz/drivers/ata/sata_sx4.c --- linux-2.6.32.48/drivers/ata/sata_sx4.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/sata_sx4.c 2011-11-17 20:08:53.000000000 -0500 @@ -248,7 +248,7 @@ static struct scsi_host_template pdc_sat }; /* TODO: inherit from base port_ops after converting to new EH */ -static struct ata_port_operations pdc_20621_ops = { +static const struct ata_port_operations pdc_20621_ops = { .inherits = &ata_sff_port_ops, .check_atapi_dma = pdc_check_atapi_dma, diff -urNp linux-2.6.32.48/drivers/ata/sata_uli.c linux-2.6.32.48-openvz/drivers/ata/sata_uli.c --- linux-2.6.32.48/drivers/ata/sata_uli.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/sata_uli.c 2011-11-17 20:08:53.000000000 -0500 @@ -79,7 +79,7 @@ static struct scsi_host_template uli_sht ATA_BMDMA_SHT(DRV_NAME), }; -static struct ata_port_operations uli_ops = { +static const struct ata_port_operations uli_ops = { .inherits = &ata_bmdma_port_ops, .scr_read = uli_scr_read, .scr_write = uli_scr_write, diff -urNp linux-2.6.32.48/drivers/ata/sata_via.c linux-2.6.32.48-openvz/drivers/ata/sata_via.c --- linux-2.6.32.48/drivers/ata/sata_via.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/sata_via.c 2011-11-17 20:08:53.000000000 -0500 @@ -115,32 +115,32 @@ static struct scsi_host_template svia_sh ATA_BMDMA_SHT(DRV_NAME), }; -static struct ata_port_operations svia_base_ops = { +static const struct ata_port_operations svia_base_ops = { .inherits = &ata_bmdma_port_ops, .sff_tf_load = svia_tf_load, }; -static struct ata_port_operations vt6420_sata_ops = { +static const struct ata_port_operations vt6420_sata_ops = { .inherits = &svia_base_ops, .freeze = svia_noop_freeze, .prereset = vt6420_prereset, .bmdma_start = vt6420_bmdma_start, }; -static struct ata_port_operations vt6421_pata_ops = { +static const struct ata_port_operations vt6421_pata_ops = { .inherits = &svia_base_ops, .cable_detect = vt6421_pata_cable_detect, .set_piomode = vt6421_set_pio_mode, .set_dmamode = vt6421_set_dma_mode, }; -static struct ata_port_operations vt6421_sata_ops = { +static const struct ata_port_operations vt6421_sata_ops = { .inherits = &svia_base_ops, .scr_read = svia_scr_read, .scr_write = svia_scr_write, }; -static struct ata_port_operations vt8251_ops = { +static const struct ata_port_operations vt8251_ops = { .inherits = &svia_base_ops, .hardreset = sata_std_hardreset, .scr_read = vt8251_scr_read, diff -urNp linux-2.6.32.48/drivers/ata/sata_vsc.c linux-2.6.32.48-openvz/drivers/ata/sata_vsc.c --- linux-2.6.32.48/drivers/ata/sata_vsc.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ata/sata_vsc.c 2011-11-17 20:08:53.000000000 -0500 @@ -306,7 +306,7 @@ static struct scsi_host_template vsc_sat }; -static struct ata_port_operations vsc_sata_ops = { +static const struct ata_port_operations vsc_sata_ops = { .inherits = &ata_bmdma_port_ops, /* The IRQ handling is not quite standard SFF behaviour so we cannot use the default lost interrupt handler */ diff -urNp linux-2.6.32.48/drivers/atm/adummy.c linux-2.6.32.48-openvz/drivers/atm/adummy.c --- linux-2.6.32.48/drivers/atm/adummy.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/atm/adummy.c 2011-11-17 20:08:53.000000000 -0500 @@ -77,7 +77,7 @@ adummy_send(struct atm_vcc *vcc, struct vcc->pop(vcc, skb); else dev_kfree_skb_any(skb); - atomic_inc(&vcc->stats->tx); + atomic_inc_unchecked(&vcc->stats->tx); return 0; } diff -urNp linux-2.6.32.48/drivers/atm/ambassador.c linux-2.6.32.48-openvz/drivers/atm/ambassador.c --- linux-2.6.32.48/drivers/atm/ambassador.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/atm/ambassador.c 2011-11-17 20:08:53.000000000 -0500 @@ -453,7 +453,7 @@ static void tx_complete (amb_dev * dev, PRINTD (DBG_FLOW|DBG_TX, "tx_complete %p %p", dev, tx); // VC layer stats - atomic_inc(&ATM_SKB(skb)->vcc->stats->tx); + atomic_inc_unchecked(&ATM_SKB(skb)->vcc->stats->tx); // free the descriptor kfree (tx_descr); @@ -494,7 +494,7 @@ static void rx_complete (amb_dev * dev, dump_skb ("<<<", vc, skb); // VC layer stats - atomic_inc(&atm_vcc->stats->rx); + atomic_inc_unchecked(&atm_vcc->stats->rx); __net_timestamp(skb); // end of our responsability atm_vcc->push (atm_vcc, skb); @@ -509,7 +509,7 @@ static void rx_complete (amb_dev * dev, } else { PRINTK (KERN_INFO, "dropped over-size frame"); // should we count this? - atomic_inc(&atm_vcc->stats->rx_drop); + atomic_inc_unchecked(&atm_vcc->stats->rx_drop); } } else { @@ -1341,7 +1341,7 @@ static int amb_send (struct atm_vcc * at } if (check_area (skb->data, skb->len)) { - atomic_inc(&atm_vcc->stats->tx_err); + atomic_inc_unchecked(&atm_vcc->stats->tx_err); return -ENOMEM; // ? } diff -urNp linux-2.6.32.48/drivers/atm/atmtcp.c linux-2.6.32.48-openvz/drivers/atm/atmtcp.c --- linux-2.6.32.48/drivers/atm/atmtcp.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/atm/atmtcp.c 2011-11-17 20:08:53.000000000 -0500 @@ -206,7 +206,7 @@ static int atmtcp_v_send(struct atm_vcc if (vcc->pop) vcc->pop(vcc,skb); else dev_kfree_skb(skb); if (dev_data) return 0; - atomic_inc(&vcc->stats->tx_err); + atomic_inc_unchecked(&vcc->stats->tx_err); return -ENOLINK; } size = skb->len+sizeof(struct atmtcp_hdr); @@ -214,7 +214,7 @@ static int atmtcp_v_send(struct atm_vcc if (!new_skb) { if (vcc->pop) vcc->pop(vcc,skb); else dev_kfree_skb(skb); - atomic_inc(&vcc->stats->tx_err); + atomic_inc_unchecked(&vcc->stats->tx_err); return -ENOBUFS; } hdr = (void *) skb_put(new_skb,sizeof(struct atmtcp_hdr)); @@ -225,8 +225,8 @@ static int atmtcp_v_send(struct atm_vcc if (vcc->pop) vcc->pop(vcc,skb); else dev_kfree_skb(skb); out_vcc->push(out_vcc,new_skb); - atomic_inc(&vcc->stats->tx); - atomic_inc(&out_vcc->stats->rx); + atomic_inc_unchecked(&vcc->stats->tx); + atomic_inc_unchecked(&out_vcc->stats->rx); return 0; } @@ -300,7 +300,7 @@ static int atmtcp_c_send(struct atm_vcc out_vcc = find_vcc(dev, ntohs(hdr->vpi), ntohs(hdr->vci)); read_unlock(&vcc_sklist_lock); if (!out_vcc) { - atomic_inc(&vcc->stats->tx_err); + atomic_inc_unchecked(&vcc->stats->tx_err); goto done; } skb_pull(skb,sizeof(struct atmtcp_hdr)); @@ -312,8 +312,8 @@ static int atmtcp_c_send(struct atm_vcc __net_timestamp(new_skb); skb_copy_from_linear_data(skb, skb_put(new_skb, skb->len), skb->len); out_vcc->push(out_vcc,new_skb); - atomic_inc(&vcc->stats->tx); - atomic_inc(&out_vcc->stats->rx); + atomic_inc_unchecked(&vcc->stats->tx); + atomic_inc_unchecked(&out_vcc->stats->rx); done: if (vcc->pop) vcc->pop(vcc,skb); else dev_kfree_skb(skb); diff -urNp linux-2.6.32.48/drivers/atm/eni.c linux-2.6.32.48-openvz/drivers/atm/eni.c --- linux-2.6.32.48/drivers/atm/eni.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/atm/eni.c 2011-11-17 20:08:53.000000000 -0500 @@ -525,7 +525,7 @@ static int rx_aal0(struct atm_vcc *vcc) DPRINTK(DEV_LABEL "(itf %d): trashing empty cell\n", vcc->dev->number); length = 0; - atomic_inc(&vcc->stats->rx_err); + atomic_inc_unchecked(&vcc->stats->rx_err); } else { length = ATM_CELL_SIZE-1; /* no HEC */ @@ -580,7 +580,7 @@ static int rx_aal5(struct atm_vcc *vcc) size); } eff = length = 0; - atomic_inc(&vcc->stats->rx_err); + atomic_inc_unchecked(&vcc->stats->rx_err); } else { size = (descr & MID_RED_COUNT)*(ATM_CELL_PAYLOAD >> 2); @@ -597,7 +597,7 @@ static int rx_aal5(struct atm_vcc *vcc) "(VCI=%d,length=%ld,size=%ld (descr 0x%lx))\n", vcc->dev->number,vcc->vci,length,size << 2,descr); length = eff = 0; - atomic_inc(&vcc->stats->rx_err); + atomic_inc_unchecked(&vcc->stats->rx_err); } } skb = eff ? atm_alloc_charge(vcc,eff << 2,GFP_ATOMIC) : NULL; @@ -770,7 +770,7 @@ rx_dequeued++; vcc->push(vcc,skb); pushed++; } - atomic_inc(&vcc->stats->rx); + atomic_inc_unchecked(&vcc->stats->rx); } wake_up(&eni_dev->rx_wait); } @@ -1227,7 +1227,7 @@ static void dequeue_tx(struct atm_dev *d PCI_DMA_TODEVICE); if (vcc->pop) vcc->pop(vcc,skb); else dev_kfree_skb_irq(skb); - atomic_inc(&vcc->stats->tx); + atomic_inc_unchecked(&vcc->stats->tx); wake_up(&eni_dev->tx_wait); dma_complete++; } diff -urNp linux-2.6.32.48/drivers/atm/firestream.c linux-2.6.32.48-openvz/drivers/atm/firestream.c --- linux-2.6.32.48/drivers/atm/firestream.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/atm/firestream.c 2011-11-17 20:08:53.000000000 -0500 @@ -748,7 +748,7 @@ static void process_txdone_queue (struct } } - atomic_inc(&ATM_SKB(skb)->vcc->stats->tx); + atomic_inc_unchecked(&ATM_SKB(skb)->vcc->stats->tx); fs_dprintk (FS_DEBUG_TXMEM, "i"); fs_dprintk (FS_DEBUG_ALLOC, "Free t-skb: %p\n", skb); @@ -815,7 +815,7 @@ static void process_incoming (struct fs_ #endif skb_put (skb, qe->p1 & 0xffff); ATM_SKB(skb)->vcc = atm_vcc; - atomic_inc(&atm_vcc->stats->rx); + atomic_inc_unchecked(&atm_vcc->stats->rx); __net_timestamp(skb); fs_dprintk (FS_DEBUG_ALLOC, "Free rec-skb: %p (pushed)\n", skb); atm_vcc->push (atm_vcc, skb); @@ -836,12 +836,12 @@ static void process_incoming (struct fs_ kfree (pe); } if (atm_vcc) - atomic_inc(&atm_vcc->stats->rx_drop); + atomic_inc_unchecked(&atm_vcc->stats->rx_drop); break; case 0x1f: /* Reassembly abort: no buffers. */ /* Silently increment error counter. */ if (atm_vcc) - atomic_inc(&atm_vcc->stats->rx_drop); + atomic_inc_unchecked(&atm_vcc->stats->rx_drop); break; default: /* Hmm. Haven't written the code to handle the others yet... -- REW */ printk (KERN_WARNING "Don't know what to do with RX status %x: %s.\n", diff -urNp linux-2.6.32.48/drivers/atm/fore200e.c linux-2.6.32.48-openvz/drivers/atm/fore200e.c --- linux-2.6.32.48/drivers/atm/fore200e.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/atm/fore200e.c 2011-11-17 20:08:53.000000000 -0500 @@ -931,9 +931,9 @@ fore200e_tx_irq(struct fore200e* fore200 #endif /* check error condition */ if (*entry->status & STATUS_ERROR) - atomic_inc(&vcc->stats->tx_err); + atomic_inc_unchecked(&vcc->stats->tx_err); else - atomic_inc(&vcc->stats->tx); + atomic_inc_unchecked(&vcc->stats->tx); } } @@ -1082,7 +1082,7 @@ fore200e_push_rpd(struct fore200e* fore2 if (skb == NULL) { DPRINTK(2, "unable to alloc new skb, rx PDU length = %d\n", pdu_len); - atomic_inc(&vcc->stats->rx_drop); + atomic_inc_unchecked(&vcc->stats->rx_drop); return -ENOMEM; } @@ -1125,14 +1125,14 @@ fore200e_push_rpd(struct fore200e* fore2 dev_kfree_skb_any(skb); - atomic_inc(&vcc->stats->rx_drop); + atomic_inc_unchecked(&vcc->stats->rx_drop); return -ENOMEM; } ASSERT(atomic_read(&sk_atm(vcc)->sk_wmem_alloc) >= 0); vcc->push(vcc, skb); - atomic_inc(&vcc->stats->rx); + atomic_inc_unchecked(&vcc->stats->rx); ASSERT(atomic_read(&sk_atm(vcc)->sk_wmem_alloc) >= 0); @@ -1210,7 +1210,7 @@ fore200e_rx_irq(struct fore200e* fore200 DPRINTK(2, "damaged PDU on %d.%d.%d\n", fore200e->atm_dev->number, entry->rpd->atm_header.vpi, entry->rpd->atm_header.vci); - atomic_inc(&vcc->stats->rx_err); + atomic_inc_unchecked(&vcc->stats->rx_err); } } @@ -1655,7 +1655,7 @@ fore200e_send(struct atm_vcc *vcc, struc goto retry_here; } - atomic_inc(&vcc->stats->tx_err); + atomic_inc_unchecked(&vcc->stats->tx_err); fore200e->tx_sat++; DPRINTK(2, "tx queue of device %s is saturated, PDU dropped - heartbeat is %08x\n", diff -urNp linux-2.6.32.48/drivers/atm/he.c linux-2.6.32.48-openvz/drivers/atm/he.c --- linux-2.6.32.48/drivers/atm/he.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/atm/he.c 2011-11-17 20:08:53.000000000 -0500 @@ -1769,7 +1769,7 @@ he_service_rbrq(struct he_dev *he_dev, i if (RBRQ_HBUF_ERR(he_dev->rbrq_head)) { hprintk("HBUF_ERR! (cid 0x%x)\n", cid); - atomic_inc(&vcc->stats->rx_drop); + atomic_inc_unchecked(&vcc->stats->rx_drop); goto return_host_buffers; } @@ -1802,7 +1802,7 @@ he_service_rbrq(struct he_dev *he_dev, i RBRQ_LEN_ERR(he_dev->rbrq_head) ? "LEN_ERR" : "", vcc->vpi, vcc->vci); - atomic_inc(&vcc->stats->rx_err); + atomic_inc_unchecked(&vcc->stats->rx_err); goto return_host_buffers; } @@ -1861,7 +1861,7 @@ he_service_rbrq(struct he_dev *he_dev, i vcc->push(vcc, skb); spin_lock(&he_dev->global_lock); - atomic_inc(&vcc->stats->rx); + atomic_inc_unchecked(&vcc->stats->rx); return_host_buffers: ++pdus_assembled; @@ -2206,7 +2206,7 @@ __enqueue_tpd(struct he_dev *he_dev, str tpd->vcc->pop(tpd->vcc, tpd->skb); else dev_kfree_skb_any(tpd->skb); - atomic_inc(&tpd->vcc->stats->tx_err); + atomic_inc_unchecked(&tpd->vcc->stats->tx_err); } pci_pool_free(he_dev->tpd_pool, tpd, TPD_ADDR(tpd->status)); return; @@ -2618,7 +2618,7 @@ he_send(struct atm_vcc *vcc, struct sk_b vcc->pop(vcc, skb); else dev_kfree_skb_any(skb); - atomic_inc(&vcc->stats->tx_err); + atomic_inc_unchecked(&vcc->stats->tx_err); return -EINVAL; } @@ -2629,7 +2629,7 @@ he_send(struct atm_vcc *vcc, struct sk_b vcc->pop(vcc, skb); else dev_kfree_skb_any(skb); - atomic_inc(&vcc->stats->tx_err); + atomic_inc_unchecked(&vcc->stats->tx_err); return -EINVAL; } #endif @@ -2641,7 +2641,7 @@ he_send(struct atm_vcc *vcc, struct sk_b vcc->pop(vcc, skb); else dev_kfree_skb_any(skb); - atomic_inc(&vcc->stats->tx_err); + atomic_inc_unchecked(&vcc->stats->tx_err); spin_unlock_irqrestore(&he_dev->global_lock, flags); return -ENOMEM; } @@ -2683,7 +2683,7 @@ he_send(struct atm_vcc *vcc, struct sk_b vcc->pop(vcc, skb); else dev_kfree_skb_any(skb); - atomic_inc(&vcc->stats->tx_err); + atomic_inc_unchecked(&vcc->stats->tx_err); spin_unlock_irqrestore(&he_dev->global_lock, flags); return -ENOMEM; } @@ -2714,7 +2714,7 @@ he_send(struct atm_vcc *vcc, struct sk_b __enqueue_tpd(he_dev, tpd, cid); spin_unlock_irqrestore(&he_dev->global_lock, flags); - atomic_inc(&vcc->stats->tx); + atomic_inc_unchecked(&vcc->stats->tx); return 0; } diff -urNp linux-2.6.32.48/drivers/atm/horizon.c linux-2.6.32.48-openvz/drivers/atm/horizon.c --- linux-2.6.32.48/drivers/atm/horizon.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/atm/horizon.c 2011-11-17 20:08:53.000000000 -0500 @@ -1033,7 +1033,7 @@ static void rx_schedule (hrz_dev * dev, { struct atm_vcc * vcc = ATM_SKB(skb)->vcc; // VC layer stats - atomic_inc(&vcc->stats->rx); + atomic_inc_unchecked(&vcc->stats->rx); __net_timestamp(skb); // end of our responsability vcc->push (vcc, skb); @@ -1185,7 +1185,7 @@ static void tx_schedule (hrz_dev * const dev->tx_iovec = NULL; // VC layer stats - atomic_inc(&ATM_SKB(skb)->vcc->stats->tx); + atomic_inc_unchecked(&ATM_SKB(skb)->vcc->stats->tx); // free the skb hrz_kfree_skb (skb); diff -urNp linux-2.6.32.48/drivers/atm/idt77252.c linux-2.6.32.48-openvz/drivers/atm/idt77252.c --- linux-2.6.32.48/drivers/atm/idt77252.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/atm/idt77252.c 2011-11-17 20:08:53.000000000 -0500 @@ -810,7 +810,7 @@ drain_scq(struct idt77252_dev *card, str else dev_kfree_skb(skb); - atomic_inc(&vcc->stats->tx); + atomic_inc_unchecked(&vcc->stats->tx); } atomic_dec(&scq->used); @@ -1073,13 +1073,13 @@ dequeue_rx(struct idt77252_dev *card, st if ((sb = dev_alloc_skb(64)) == NULL) { printk("%s: Can't allocate buffers for aal0.\n", card->name); - atomic_add(i, &vcc->stats->rx_drop); + atomic_add_unchecked(i, &vcc->stats->rx_drop); break; } if (!atm_charge(vcc, sb->truesize)) { RXPRINTK("%s: atm_charge() dropped aal0 packets.\n", card->name); - atomic_add(i - 1, &vcc->stats->rx_drop); + atomic_add_unchecked(i - 1, &vcc->stats->rx_drop); dev_kfree_skb(sb); break; } @@ -1096,7 +1096,7 @@ dequeue_rx(struct idt77252_dev *card, st ATM_SKB(sb)->vcc = vcc; __net_timestamp(sb); vcc->push(vcc, sb); - atomic_inc(&vcc->stats->rx); + atomic_inc_unchecked(&vcc->stats->rx); cell += ATM_CELL_PAYLOAD; } @@ -1133,13 +1133,13 @@ dequeue_rx(struct idt77252_dev *card, st "(CDC: %08x)\n", card->name, len, rpp->len, readl(SAR_REG_CDC)); recycle_rx_pool_skb(card, rpp); - atomic_inc(&vcc->stats->rx_err); + atomic_inc_unchecked(&vcc->stats->rx_err); return; } if (stat & SAR_RSQE_CRC) { RXPRINTK("%s: AAL5 CRC error.\n", card->name); recycle_rx_pool_skb(card, rpp); - atomic_inc(&vcc->stats->rx_err); + atomic_inc_unchecked(&vcc->stats->rx_err); return; } if (skb_queue_len(&rpp->queue) > 1) { @@ -1150,7 +1150,7 @@ dequeue_rx(struct idt77252_dev *card, st RXPRINTK("%s: Can't alloc RX skb.\n", card->name); recycle_rx_pool_skb(card, rpp); - atomic_inc(&vcc->stats->rx_err); + atomic_inc_unchecked(&vcc->stats->rx_err); return; } if (!atm_charge(vcc, skb->truesize)) { @@ -1169,7 +1169,7 @@ dequeue_rx(struct idt77252_dev *card, st __net_timestamp(skb); vcc->push(vcc, skb); - atomic_inc(&vcc->stats->rx); + atomic_inc_unchecked(&vcc->stats->rx); return; } @@ -1191,7 +1191,7 @@ dequeue_rx(struct idt77252_dev *card, st __net_timestamp(skb); vcc->push(vcc, skb); - atomic_inc(&vcc->stats->rx); + atomic_inc_unchecked(&vcc->stats->rx); if (skb->truesize > SAR_FB_SIZE_3) add_rx_skb(card, 3, SAR_FB_SIZE_3, 1); @@ -1303,14 +1303,14 @@ idt77252_rx_raw(struct idt77252_dev *car if (vcc->qos.aal != ATM_AAL0) { RPRINTK("%s: raw cell for non AAL0 vc %u.%u\n", card->name, vpi, vci); - atomic_inc(&vcc->stats->rx_drop); + atomic_inc_unchecked(&vcc->stats->rx_drop); goto drop; } if ((sb = dev_alloc_skb(64)) == NULL) { printk("%s: Can't allocate buffers for AAL0.\n", card->name); - atomic_inc(&vcc->stats->rx_err); + atomic_inc_unchecked(&vcc->stats->rx_err); goto drop; } @@ -1329,7 +1329,7 @@ idt77252_rx_raw(struct idt77252_dev *car ATM_SKB(sb)->vcc = vcc; __net_timestamp(sb); vcc->push(vcc, sb); - atomic_inc(&vcc->stats->rx); + atomic_inc_unchecked(&vcc->stats->rx); drop: skb_pull(queue, 64); @@ -1954,13 +1954,13 @@ idt77252_send_skb(struct atm_vcc *vcc, s if (vc == NULL) { printk("%s: NULL connection in send().\n", card->name); - atomic_inc(&vcc->stats->tx_err); + atomic_inc_unchecked(&vcc->stats->tx_err); dev_kfree_skb(skb); return -EINVAL; } if (!test_bit(VCF_TX, &vc->flags)) { printk("%s: Trying to transmit on a non-tx VC.\n", card->name); - atomic_inc(&vcc->stats->tx_err); + atomic_inc_unchecked(&vcc->stats->tx_err); dev_kfree_skb(skb); return -EINVAL; } @@ -1972,14 +1972,14 @@ idt77252_send_skb(struct atm_vcc *vcc, s break; default: printk("%s: Unsupported AAL: %d\n", card->name, vcc->qos.aal); - atomic_inc(&vcc->stats->tx_err); + atomic_inc_unchecked(&vcc->stats->tx_err); dev_kfree_skb(skb); return -EINVAL; } if (skb_shinfo(skb)->nr_frags != 0) { printk("%s: No scatter-gather yet.\n", card->name); - atomic_inc(&vcc->stats->tx_err); + atomic_inc_unchecked(&vcc->stats->tx_err); dev_kfree_skb(skb); return -EINVAL; } @@ -1987,7 +1987,7 @@ idt77252_send_skb(struct atm_vcc *vcc, s err = queue_skb(card, vc, skb, oam); if (err) { - atomic_inc(&vcc->stats->tx_err); + atomic_inc_unchecked(&vcc->stats->tx_err); dev_kfree_skb(skb); return err; } @@ -2010,7 +2010,7 @@ idt77252_send_oam(struct atm_vcc *vcc, v skb = dev_alloc_skb(64); if (!skb) { printk("%s: Out of memory in send_oam().\n", card->name); - atomic_inc(&vcc->stats->tx_err); + atomic_inc_unchecked(&vcc->stats->tx_err); return -ENOMEM; } atomic_add(skb->truesize, &sk_atm(vcc)->sk_wmem_alloc); diff -urNp linux-2.6.32.48/drivers/atm/iphase.c linux-2.6.32.48-openvz/drivers/atm/iphase.c --- linux-2.6.32.48/drivers/atm/iphase.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/atm/iphase.c 2011-11-17 20:08:53.000000000 -0500 @@ -1123,7 +1123,7 @@ static int rx_pkt(struct atm_dev *dev) status = (u_short) (buf_desc_ptr->desc_mode); if (status & (RX_CER | RX_PTE | RX_OFL)) { - atomic_inc(&vcc->stats->rx_err); + atomic_inc_unchecked(&vcc->stats->rx_err); IF_ERR(printk("IA: bad packet, dropping it");) if (status & RX_CER) { IF_ERR(printk(" cause: packet CRC error\n");) @@ -1146,7 +1146,7 @@ static int rx_pkt(struct atm_dev *dev) len = dma_addr - buf_addr; if (len > iadev->rx_buf_sz) { printk("Over %d bytes sdu received, dropped!!!\n", iadev->rx_buf_sz); - atomic_inc(&vcc->stats->rx_err); + atomic_inc_unchecked(&vcc->stats->rx_err); goto out_free_desc; } @@ -1296,7 +1296,7 @@ static void rx_dle_intr(struct atm_dev * ia_vcc = INPH_IA_VCC(vcc); if (ia_vcc == NULL) { - atomic_inc(&vcc->stats->rx_err); + atomic_inc_unchecked(&vcc->stats->rx_err); dev_kfree_skb_any(skb); atm_return(vcc, atm_guess_pdu2truesize(len)); goto INCR_DLE; @@ -1308,7 +1308,7 @@ static void rx_dle_intr(struct atm_dev * if ((length > iadev->rx_buf_sz) || (length > (skb->len - sizeof(struct cpcs_trailer)))) { - atomic_inc(&vcc->stats->rx_err); + atomic_inc_unchecked(&vcc->stats->rx_err); IF_ERR(printk("rx_dle_intr: Bad AAL5 trailer %d (skb len %d)", length, skb->len);) dev_kfree_skb_any(skb); @@ -1324,7 +1324,7 @@ static void rx_dle_intr(struct atm_dev * IF_RX(printk("rx_dle_intr: skb push");) vcc->push(vcc,skb); - atomic_inc(&vcc->stats->rx); + atomic_inc_unchecked(&vcc->stats->rx); iadev->rx_pkt_cnt++; } INCR_DLE: @@ -2806,15 +2806,15 @@ static int ia_ioctl(struct atm_dev *dev, { struct k_sonet_stats *stats; stats = &PRIV(_ia_dev[board])->sonet_stats; - printk("section_bip: %d\n", atomic_read(&stats->section_bip)); - printk("line_bip : %d\n", atomic_read(&stats->line_bip)); - printk("path_bip : %d\n", atomic_read(&stats->path_bip)); - printk("line_febe : %d\n", atomic_read(&stats->line_febe)); - printk("path_febe : %d\n", atomic_read(&stats->path_febe)); - printk("corr_hcs : %d\n", atomic_read(&stats->corr_hcs)); - printk("uncorr_hcs : %d\n", atomic_read(&stats->uncorr_hcs)); - printk("tx_cells : %d\n", atomic_read(&stats->tx_cells)); - printk("rx_cells : %d\n", atomic_read(&stats->rx_cells)); + printk("section_bip: %d\n", atomic_read_unchecked(&stats->section_bip)); + printk("line_bip : %d\n", atomic_read_unchecked(&stats->line_bip)); + printk("path_bip : %d\n", atomic_read_unchecked(&stats->path_bip)); + printk("line_febe : %d\n", atomic_read_unchecked(&stats->line_febe)); + printk("path_febe : %d\n", atomic_read_unchecked(&stats->path_febe)); + printk("corr_hcs : %d\n", atomic_read_unchecked(&stats->corr_hcs)); + printk("uncorr_hcs : %d\n", atomic_read_unchecked(&stats->uncorr_hcs)); + printk("tx_cells : %d\n", atomic_read_unchecked(&stats->tx_cells)); + printk("rx_cells : %d\n", atomic_read_unchecked(&stats->rx_cells)); } ia_cmds.status = 0; break; @@ -2919,7 +2919,7 @@ static int ia_pkt_tx (struct atm_vcc *vc if ((desc == 0) || (desc > iadev->num_tx_desc)) { IF_ERR(printk(DEV_LABEL "invalid desc for send: %d\n", desc);) - atomic_inc(&vcc->stats->tx); + atomic_inc_unchecked(&vcc->stats->tx); if (vcc->pop) vcc->pop(vcc, skb); else @@ -3024,14 +3024,14 @@ static int ia_pkt_tx (struct atm_vcc *vc ATM_DESC(skb) = vcc->vci; skb_queue_tail(&iadev->tx_dma_q, skb); - atomic_inc(&vcc->stats->tx); + atomic_inc_unchecked(&vcc->stats->tx); iadev->tx_pkt_cnt++; /* Increment transaction counter */ writel(2, iadev->dma+IPHASE5575_TX_COUNTER); #if 0 /* add flow control logic */ - if (atomic_read(&vcc->stats->tx) % 20 == 0) { + if (atomic_read_unchecked(&vcc->stats->tx) % 20 == 0) { if (iavcc->vc_desc_cnt > 10) { vcc->tx_quota = vcc->tx_quota * 3 / 4; printk("Tx1: vcc->tx_quota = %d \n", (u32)vcc->tx_quota ); diff -urNp linux-2.6.32.48/drivers/atm/lanai.c linux-2.6.32.48-openvz/drivers/atm/lanai.c --- linux-2.6.32.48/drivers/atm/lanai.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/atm/lanai.c 2011-11-17 20:08:53.000000000 -0500 @@ -1305,7 +1305,7 @@ static void lanai_send_one_aal5(struct l vcc_tx_add_aal5_trailer(lvcc, skb->len, 0, 0); lanai_endtx(lanai, lvcc); lanai_free_skb(lvcc->tx.atmvcc, skb); - atomic_inc(&lvcc->tx.atmvcc->stats->tx); + atomic_inc_unchecked(&lvcc->tx.atmvcc->stats->tx); } /* Try to fill the buffer - don't call unless there is backlog */ @@ -1428,7 +1428,7 @@ static void vcc_rx_aal5(struct lanai_vcc ATM_SKB(skb)->vcc = lvcc->rx.atmvcc; __net_timestamp(skb); lvcc->rx.atmvcc->push(lvcc->rx.atmvcc, skb); - atomic_inc(&lvcc->rx.atmvcc->stats->rx); + atomic_inc_unchecked(&lvcc->rx.atmvcc->stats->rx); out: lvcc->rx.buf.ptr = end; cardvcc_write(lvcc, endptr, vcc_rxreadptr); @@ -1670,7 +1670,7 @@ static int handle_service(struct lanai_d DPRINTK("(itf %d) got RX service entry 0x%X for non-AAL5 " "vcc %d\n", lanai->number, (unsigned int) s, vci); lanai->stats.service_rxnotaal5++; - atomic_inc(&lvcc->rx.atmvcc->stats->rx_err); + atomic_inc_unchecked(&lvcc->rx.atmvcc->stats->rx_err); return 0; } if (likely(!(s & (SERVICE_TRASH | SERVICE_STREAM | SERVICE_CRCERR)))) { @@ -1682,7 +1682,7 @@ static int handle_service(struct lanai_d int bytes; read_unlock(&vcc_sklist_lock); DPRINTK("got trashed rx pdu on vci %d\n", vci); - atomic_inc(&lvcc->rx.atmvcc->stats->rx_err); + atomic_inc_unchecked(&lvcc->rx.atmvcc->stats->rx_err); lvcc->stats.x.aal5.service_trash++; bytes = (SERVICE_GET_END(s) * 16) - (((unsigned long) lvcc->rx.buf.ptr) - @@ -1694,7 +1694,7 @@ static int handle_service(struct lanai_d } if (s & SERVICE_STREAM) { read_unlock(&vcc_sklist_lock); - atomic_inc(&lvcc->rx.atmvcc->stats->rx_err); + atomic_inc_unchecked(&lvcc->rx.atmvcc->stats->rx_err); lvcc->stats.x.aal5.service_stream++; printk(KERN_ERR DEV_LABEL "(itf %d): Got AAL5 stream " "PDU on VCI %d!\n", lanai->number, vci); @@ -1702,7 +1702,7 @@ static int handle_service(struct lanai_d return 0; } DPRINTK("got rx crc error on vci %d\n", vci); - atomic_inc(&lvcc->rx.atmvcc->stats->rx_err); + atomic_inc_unchecked(&lvcc->rx.atmvcc->stats->rx_err); lvcc->stats.x.aal5.service_rxcrc++; lvcc->rx.buf.ptr = &lvcc->rx.buf.start[SERVICE_GET_END(s) * 4]; cardvcc_write(lvcc, SERVICE_GET_END(s), vcc_rxreadptr); diff -urNp linux-2.6.32.48/drivers/atm/nicstar.c linux-2.6.32.48-openvz/drivers/atm/nicstar.c --- linux-2.6.32.48/drivers/atm/nicstar.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/atm/nicstar.c 2011-11-17 20:08:53.000000000 -0500 @@ -1723,7 +1723,7 @@ static int ns_send(struct atm_vcc *vcc, if ((vc = (vc_map *) vcc->dev_data) == NULL) { printk("nicstar%d: vcc->dev_data == NULL on ns_send().\n", card->index); - atomic_inc(&vcc->stats->tx_err); + atomic_inc_unchecked(&vcc->stats->tx_err); dev_kfree_skb_any(skb); return -EINVAL; } @@ -1731,7 +1731,7 @@ static int ns_send(struct atm_vcc *vcc, if (!vc->tx) { printk("nicstar%d: Trying to transmit on a non-tx VC.\n", card->index); - atomic_inc(&vcc->stats->tx_err); + atomic_inc_unchecked(&vcc->stats->tx_err); dev_kfree_skb_any(skb); return -EINVAL; } @@ -1739,7 +1739,7 @@ static int ns_send(struct atm_vcc *vcc, if (vcc->qos.aal != ATM_AAL5 && vcc->qos.aal != ATM_AAL0) { printk("nicstar%d: Only AAL0 and AAL5 are supported.\n", card->index); - atomic_inc(&vcc->stats->tx_err); + atomic_inc_unchecked(&vcc->stats->tx_err); dev_kfree_skb_any(skb); return -EINVAL; } @@ -1747,7 +1747,7 @@ static int ns_send(struct atm_vcc *vcc, if (skb_shinfo(skb)->nr_frags != 0) { printk("nicstar%d: No scatter-gather yet.\n", card->index); - atomic_inc(&vcc->stats->tx_err); + atomic_inc_unchecked(&vcc->stats->tx_err); dev_kfree_skb_any(skb); return -EINVAL; } @@ -1792,11 +1792,11 @@ static int ns_send(struct atm_vcc *vcc, if (push_scqe(card, vc, scq, &scqe, skb) != 0) { - atomic_inc(&vcc->stats->tx_err); + atomic_inc_unchecked(&vcc->stats->tx_err); dev_kfree_skb_any(skb); return -EIO; } - atomic_inc(&vcc->stats->tx); + atomic_inc_unchecked(&vcc->stats->tx); return 0; } @@ -2111,14 +2111,14 @@ static void dequeue_rx(ns_dev *card, ns_ { printk("nicstar%d: Can't allocate buffers for aal0.\n", card->index); - atomic_add(i,&vcc->stats->rx_drop); + atomic_add_unchecked(i,&vcc->stats->rx_drop); break; } if (!atm_charge(vcc, sb->truesize)) { RXPRINTK("nicstar%d: atm_charge() dropped aal0 packets.\n", card->index); - atomic_add(i-1,&vcc->stats->rx_drop); /* already increased by 1 */ + atomic_add_unchecked(i-1,&vcc->stats->rx_drop); /* already increased by 1 */ dev_kfree_skb_any(sb); break; } @@ -2133,7 +2133,7 @@ static void dequeue_rx(ns_dev *card, ns_ ATM_SKB(sb)->vcc = vcc; __net_timestamp(sb); vcc->push(vcc, sb); - atomic_inc(&vcc->stats->rx); + atomic_inc_unchecked(&vcc->stats->rx); cell += ATM_CELL_PAYLOAD; } @@ -2152,7 +2152,7 @@ static void dequeue_rx(ns_dev *card, ns_ if (iovb == NULL) { printk("nicstar%d: Out of iovec buffers.\n", card->index); - atomic_inc(&vcc->stats->rx_drop); + atomic_inc_unchecked(&vcc->stats->rx_drop); recycle_rx_buf(card, skb); return; } @@ -2182,7 +2182,7 @@ static void dequeue_rx(ns_dev *card, ns_ else if (NS_SKB(iovb)->iovcnt >= NS_MAX_IOVECS) { printk("nicstar%d: received too big AAL5 SDU.\n", card->index); - atomic_inc(&vcc->stats->rx_err); + atomic_inc_unchecked(&vcc->stats->rx_err); recycle_iovec_rx_bufs(card, (struct iovec *) iovb->data, NS_MAX_IOVECS); NS_SKB(iovb)->iovcnt = 0; iovb->len = 0; @@ -2202,7 +2202,7 @@ static void dequeue_rx(ns_dev *card, ns_ printk("nicstar%d: Expected a small buffer, and this is not one.\n", card->index); which_list(card, skb); - atomic_inc(&vcc->stats->rx_err); + atomic_inc_unchecked(&vcc->stats->rx_err); recycle_rx_buf(card, skb); vc->rx_iov = NULL; recycle_iov_buf(card, iovb); @@ -2216,7 +2216,7 @@ static void dequeue_rx(ns_dev *card, ns_ printk("nicstar%d: Expected a large buffer, and this is not one.\n", card->index); which_list(card, skb); - atomic_inc(&vcc->stats->rx_err); + atomic_inc_unchecked(&vcc->stats->rx_err); recycle_iovec_rx_bufs(card, (struct iovec *) iovb->data, NS_SKB(iovb)->iovcnt); vc->rx_iov = NULL; @@ -2240,7 +2240,7 @@ static void dequeue_rx(ns_dev *card, ns_ printk(" - PDU size mismatch.\n"); else printk(".\n"); - atomic_inc(&vcc->stats->rx_err); + atomic_inc_unchecked(&vcc->stats->rx_err); recycle_iovec_rx_bufs(card, (struct iovec *) iovb->data, NS_SKB(iovb)->iovcnt); vc->rx_iov = NULL; @@ -2256,7 +2256,7 @@ static void dequeue_rx(ns_dev *card, ns_ if (!atm_charge(vcc, skb->truesize)) { push_rxbufs(card, skb); - atomic_inc(&vcc->stats->rx_drop); + atomic_inc_unchecked(&vcc->stats->rx_drop); } else { @@ -2268,7 +2268,7 @@ static void dequeue_rx(ns_dev *card, ns_ ATM_SKB(skb)->vcc = vcc; __net_timestamp(skb); vcc->push(vcc, skb); - atomic_inc(&vcc->stats->rx); + atomic_inc_unchecked(&vcc->stats->rx); } } else if (NS_SKB(iovb)->iovcnt == 2) /* One small plus one large buffer */ @@ -2283,7 +2283,7 @@ static void dequeue_rx(ns_dev *card, ns_ if (!atm_charge(vcc, sb->truesize)) { push_rxbufs(card, sb); - atomic_inc(&vcc->stats->rx_drop); + atomic_inc_unchecked(&vcc->stats->rx_drop); } else { @@ -2295,7 +2295,7 @@ static void dequeue_rx(ns_dev *card, ns_ ATM_SKB(sb)->vcc = vcc; __net_timestamp(sb); vcc->push(vcc, sb); - atomic_inc(&vcc->stats->rx); + atomic_inc_unchecked(&vcc->stats->rx); } push_rxbufs(card, skb); @@ -2306,7 +2306,7 @@ static void dequeue_rx(ns_dev *card, ns_ if (!atm_charge(vcc, skb->truesize)) { push_rxbufs(card, skb); - atomic_inc(&vcc->stats->rx_drop); + atomic_inc_unchecked(&vcc->stats->rx_drop); } else { @@ -2320,7 +2320,7 @@ static void dequeue_rx(ns_dev *card, ns_ ATM_SKB(skb)->vcc = vcc; __net_timestamp(skb); vcc->push(vcc, skb); - atomic_inc(&vcc->stats->rx); + atomic_inc_unchecked(&vcc->stats->rx); } push_rxbufs(card, sb); @@ -2342,7 +2342,7 @@ static void dequeue_rx(ns_dev *card, ns_ if (hb == NULL) { printk("nicstar%d: Out of huge buffers.\n", card->index); - atomic_inc(&vcc->stats->rx_drop); + atomic_inc_unchecked(&vcc->stats->rx_drop); recycle_iovec_rx_bufs(card, (struct iovec *) iovb->data, NS_SKB(iovb)->iovcnt); vc->rx_iov = NULL; @@ -2393,7 +2393,7 @@ static void dequeue_rx(ns_dev *card, ns_ } else dev_kfree_skb_any(hb); - atomic_inc(&vcc->stats->rx_drop); + atomic_inc_unchecked(&vcc->stats->rx_drop); } else { @@ -2427,7 +2427,7 @@ static void dequeue_rx(ns_dev *card, ns_ #endif /* NS_USE_DESTRUCTORS */ __net_timestamp(hb); vcc->push(vcc, hb); - atomic_inc(&vcc->stats->rx); + atomic_inc_unchecked(&vcc->stats->rx); } } diff -urNp linux-2.6.32.48/drivers/atm/solos-pci.c linux-2.6.32.48-openvz/drivers/atm/solos-pci.c --- linux-2.6.32.48/drivers/atm/solos-pci.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/atm/solos-pci.c 2011-11-17 20:08:53.000000000 -0500 @@ -708,7 +708,7 @@ void solos_bh(unsigned long card_arg) } atm_charge(vcc, skb->truesize); vcc->push(vcc, skb); - atomic_inc(&vcc->stats->rx); + atomic_inc_unchecked(&vcc->stats->rx); break; case PKT_STATUS: @@ -914,6 +914,8 @@ static int print_buffer(struct sk_buff * char msg[500]; char item[10]; + pax_track_stack(); + len = buf->len; for (i = 0; i < len; i++){ if(i % 8 == 0) @@ -1023,7 +1025,7 @@ static uint32_t fpga_tx(struct solos_car vcc = SKB_CB(oldskb)->vcc; if (vcc) { - atomic_inc(&vcc->stats->tx); + atomic_inc_unchecked(&vcc->stats->tx); solos_pop(vcc, oldskb); } else dev_kfree_skb_irq(oldskb); diff -urNp linux-2.6.32.48/drivers/atm/suni.c linux-2.6.32.48-openvz/drivers/atm/suni.c --- linux-2.6.32.48/drivers/atm/suni.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/atm/suni.c 2011-11-17 20:08:53.000000000 -0500 @@ -49,8 +49,8 @@ static DEFINE_SPINLOCK(sunis_lock); #define ADD_LIMITED(s,v) \ - atomic_add((v),&stats->s); \ - if (atomic_read(&stats->s) < 0) atomic_set(&stats->s,INT_MAX); + atomic_add_unchecked((v),&stats->s); \ + if (atomic_read_unchecked(&stats->s) < 0) atomic_set_unchecked(&stats->s,INT_MAX); static void suni_hz(unsigned long from_timer) diff -urNp linux-2.6.32.48/drivers/atm/uPD98402.c linux-2.6.32.48-openvz/drivers/atm/uPD98402.c --- linux-2.6.32.48/drivers/atm/uPD98402.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/atm/uPD98402.c 2011-11-17 20:08:53.000000000 -0500 @@ -41,7 +41,7 @@ static int fetch_stats(struct atm_dev *d struct sonet_stats tmp; int error = 0; - atomic_add(GET(HECCT),&PRIV(dev)->sonet_stats.uncorr_hcs); + atomic_add_unchecked(GET(HECCT),&PRIV(dev)->sonet_stats.uncorr_hcs); sonet_copy_stats(&PRIV(dev)->sonet_stats,&tmp); if (arg) error = copy_to_user(arg,&tmp,sizeof(tmp)); if (zero && !error) { @@ -160,9 +160,9 @@ static int uPD98402_ioctl(struct atm_dev #define ADD_LIMITED(s,v) \ - { atomic_add(GET(v),&PRIV(dev)->sonet_stats.s); \ - if (atomic_read(&PRIV(dev)->sonet_stats.s) < 0) \ - atomic_set(&PRIV(dev)->sonet_stats.s,INT_MAX); } + { atomic_add_unchecked(GET(v),&PRIV(dev)->sonet_stats.s); \ + if (atomic_read_unchecked(&PRIV(dev)->sonet_stats.s) < 0) \ + atomic_set_unchecked(&PRIV(dev)->sonet_stats.s,INT_MAX); } static void stat_event(struct atm_dev *dev) @@ -193,7 +193,7 @@ static void uPD98402_int(struct atm_dev if (reason & uPD98402_INT_PFM) stat_event(dev); if (reason & uPD98402_INT_PCO) { (void) GET(PCOCR); /* clear interrupt cause */ - atomic_add(GET(HECCT), + atomic_add_unchecked(GET(HECCT), &PRIV(dev)->sonet_stats.uncorr_hcs); } if ((reason & uPD98402_INT_RFO) && @@ -221,9 +221,9 @@ static int uPD98402_start(struct atm_dev PUT(~(uPD98402_INT_PFM | uPD98402_INT_ALM | uPD98402_INT_RFO | uPD98402_INT_LOS),PIMR); /* enable them */ (void) fetch_stats(dev,NULL,1); /* clear kernel counters */ - atomic_set(&PRIV(dev)->sonet_stats.corr_hcs,-1); - atomic_set(&PRIV(dev)->sonet_stats.tx_cells,-1); - atomic_set(&PRIV(dev)->sonet_stats.rx_cells,-1); + atomic_set_unchecked(&PRIV(dev)->sonet_stats.corr_hcs,-1); + atomic_set_unchecked(&PRIV(dev)->sonet_stats.tx_cells,-1); + atomic_set_unchecked(&PRIV(dev)->sonet_stats.rx_cells,-1); return 0; } diff -urNp linux-2.6.32.48/drivers/atm/zatm.c linux-2.6.32.48-openvz/drivers/atm/zatm.c --- linux-2.6.32.48/drivers/atm/zatm.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/atm/zatm.c 2011-11-17 20:08:53.000000000 -0500 @@ -458,7 +458,7 @@ printk("dummy: 0x%08lx, 0x%08lx\n",dummy } if (!size) { dev_kfree_skb_irq(skb); - if (vcc) atomic_inc(&vcc->stats->rx_err); + if (vcc) atomic_inc_unchecked(&vcc->stats->rx_err); continue; } if (!atm_charge(vcc,skb->truesize)) { @@ -468,7 +468,7 @@ printk("dummy: 0x%08lx, 0x%08lx\n",dummy skb->len = size; ATM_SKB(skb)->vcc = vcc; vcc->push(vcc,skb); - atomic_inc(&vcc->stats->rx); + atomic_inc_unchecked(&vcc->stats->rx); } zout(pos & 0xffff,MTA(mbx)); #if 0 /* probably a stupid idea */ @@ -732,7 +732,7 @@ if (*ZATM_PRV_DSC(skb) != (uPD98401_TXPD skb_queue_head(&zatm_vcc->backlog,skb); break; } - atomic_inc(&vcc->stats->tx); + atomic_inc_unchecked(&vcc->stats->tx); wake_up(&zatm_vcc->tx_wait); } diff -urNp linux-2.6.32.48/drivers/base/base.h linux-2.6.32.48-openvz/drivers/base/base.h --- linux-2.6.32.48/drivers/base/base.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/base/base.h 2011-11-17 20:08:53.000000000 -0500 @@ -129,7 +129,12 @@ extern char *make_class_name(const char extern int devres_release_all(struct device *dev); +#ifndef CONFIG_VE extern struct kset *devices_kset; +#define ve_devices_kset devices_kset +#else +#define ve_devices_kset (get_exec_env()->devices_kset) +#endif #if defined(CONFIG_MODULES) && defined(CONFIG_SYSFS) extern void module_add_driver(struct module *mod, struct device_driver *drv); diff -urNp linux-2.6.32.48/drivers/base/bus.c linux-2.6.32.48-openvz/drivers/base/bus.c --- linux-2.6.32.48/drivers/base/bus.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/base/bus.c 2011-11-17 20:08:53.000000000 -0500 @@ -70,7 +70,7 @@ static ssize_t drv_attr_store(struct kob return ret; } -static struct sysfs_ops driver_sysfs_ops = { +static const struct sysfs_ops driver_sysfs_ops = { .show = drv_attr_show, .store = drv_attr_store, }; @@ -115,7 +115,7 @@ static ssize_t bus_attr_store(struct kob return ret; } -static struct sysfs_ops bus_sysfs_ops = { +static const struct sysfs_ops bus_sysfs_ops = { .show = bus_attr_show, .store = bus_attr_store, }; @@ -154,7 +154,7 @@ static int bus_uevent_filter(struct kset return 0; } -static struct kset_uevent_ops bus_uevent_ops = { +static const struct kset_uevent_ops bus_uevent_ops = { .filter = bus_uevent_filter, }; @@ -439,21 +439,20 @@ static void device_remove_attrs(struct b } } -#ifdef CONFIG_SYSFS_DEPRECATED static int make_deprecated_bus_links(struct device *dev) { - return sysfs_create_link(&dev->kobj, - &dev->bus->p->subsys.kobj, "bus"); + if (sysfs_deprecated) + return sysfs_create_link(&dev->kobj, + &dev->bus->p->subsys.kobj, "bus"); + else + return 0; } static void remove_deprecated_bus_links(struct device *dev) { - sysfs_remove_link(&dev->kobj, "bus"); + if (sysfs_deprecated) + sysfs_remove_link(&dev->kobj, "bus"); } -#else -static inline int make_deprecated_bus_links(struct device *dev) { return 0; } -static inline void remove_deprecated_bus_links(struct device *dev) { } -#endif /** * bus_add_device - add device to bus diff -urNp linux-2.6.32.48/drivers/base/class.c linux-2.6.32.48-openvz/drivers/base/class.c --- linux-2.6.32.48/drivers/base/class.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/base/class.c 2011-11-17 20:08:53.000000000 -0500 @@ -19,6 +19,8 @@ #include #include #include +#include +#include #include "base.h" #define to_class_attr(_attr) container_of(_attr, struct class_attribute, attr) @@ -63,7 +65,7 @@ static void class_release(struct kobject kfree(cp); } -static struct sysfs_ops class_sysfs_ops = { +static const struct sysfs_ops class_sysfs_ops = { .show = class_attr_show, .store = class_attr_store, }; @@ -74,8 +76,14 @@ static struct kobj_type class_ktype = { }; /* Hotplug events for classes go to the class class_subsys */ -static struct kset *class_kset; +#ifndef CONFIG_VE +struct kset *class_kset; +EXPORT_SYMBOL_GPL(class_kset); +#define visible_class_kset class_kset +#else +#define visible_class_kset (get_exec_env()->class_kset) +#endif int class_create_file(struct class *cls, const struct class_attribute *attr) { @@ -173,14 +181,14 @@ int __class_register(struct class *cls, /* set the default /sys/dev directory for devices of this class */ if (!cls->dev_kobj) - cls->dev_kobj = sysfs_dev_char_kobj; + cls->dev_kobj = ve_sysfs_dev_char_kobj; -#if defined(CONFIG_SYSFS_DEPRECATED) && defined(CONFIG_BLOCK) +#if defined(CONFIG_BLOCK) /* let the block class directory show up in the root of sysfs */ - if (cls != &block_class) - cp->class_subsys.kobj.kset = class_kset; + if (!sysfs_deprecated || cls != &block_class) + cp->class_subsys.kobj.kset = visible_class_kset; #else - cp->class_subsys.kobj.kset = class_kset; + cp->class_subsys.kobj.kset = visible_class_kset; #endif cp->class_subsys.kobj.ktype = &class_ktype; cp->class = cls; @@ -265,7 +273,6 @@ void class_destroy(struct class *cls) class_unregister(cls); } -#ifdef CONFIG_SYSFS_DEPRECATED char *make_class_name(const char *name, struct kobject *kobj) { char *class_name; @@ -282,7 +289,6 @@ char *make_class_name(const char *name, strcat(class_name, kobject_name(kobj)); return class_name; } -#endif /** * class_dev_iter_init - initialize class device iterator @@ -508,7 +514,7 @@ struct class_compat *class_compat_regist cls = kmalloc(sizeof(struct class_compat), GFP_KERNEL); if (!cls) return NULL; - cls->kobj = kobject_create_and_add(name, &class_kset->kobj); + cls->kobj = kobject_create_and_add(name, &visible_class_kset->kobj); if (!cls->kobj) { kfree(cls); return NULL; @@ -577,13 +583,20 @@ void class_compat_remove_link(struct cla } EXPORT_SYMBOL_GPL(class_compat_remove_link); -int __init classes_init(void) +int classes_init(void) { - class_kset = kset_create_and_add("class", NULL, NULL); - if (!class_kset) + visible_class_kset = kset_create_and_add("class", NULL, NULL); + if (!visible_class_kset) return -ENOMEM; return 0; } +EXPORT_SYMBOL_GPL(classes_init); + +void classes_fini(void) +{ + kset_unregister(visible_class_kset); +} +EXPORT_SYMBOL_GPL(classes_fini); EXPORT_SYMBOL_GPL(class_create_file); EXPORT_SYMBOL_GPL(class_remove_file); diff -urNp linux-2.6.32.48/drivers/base/core.c linux-2.6.32.48-openvz/drivers/base/core.c --- linux-2.6.32.48/drivers/base/core.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/base/core.c 2011-11-17 20:08:53.000000000 -0500 @@ -23,15 +23,22 @@ #include #include #include +#include +#include #include "base.h" #include "power/power.h" int (*platform_notify)(struct device *dev) = NULL; int (*platform_notify_remove)(struct device *dev) = NULL; +#ifndef CONFIG_VE static struct kobject *dev_kobj; +#define ve_dev_kobj dev_kobj struct kobject *sysfs_dev_char_kobj; struct kobject *sysfs_dev_block_kobj; +#else +#define ve_dev_kobj (get_exec_env()->dev_kobj) +#endif #ifdef CONFIG_BLOCK static inline int device_is_not_partition(struct device *dev) @@ -100,7 +107,7 @@ static ssize_t dev_attr_store(struct kob return ret; } -static struct sysfs_ops dev_sysfs_ops = { +static const struct sysfs_ops dev_sysfs_ops = { .show = dev_attr_show, .store = dev_attr_store, }; @@ -192,7 +199,9 @@ static int dev_uevent(struct kset *kset, if (dev->driver) add_uevent_var(env, "DRIVER=%s", dev->driver->name); -#ifdef CONFIG_SYSFS_DEPRECATED + if (!sysfs_deprecated) + goto skip; + if (dev->class) { struct device *parent = dev->parent; @@ -221,7 +230,7 @@ static int dev_uevent(struct kset *kset, add_uevent_var(env, "PHYSDEVDRIVER=%s", dev->driver->name); } -#endif +skip: /* have the bus specific function add its stuff */ if (dev->bus && dev->bus->uevent) { @@ -252,7 +261,7 @@ static int dev_uevent(struct kset *kset, return retval; } -static struct kset_uevent_ops device_uevent_ops = { +static const struct kset_uevent_ops device_uevent_ops = { .filter = dev_uevent_filter, .name = dev_uevent_name, .uevent = dev_uevent, @@ -438,8 +447,9 @@ static ssize_t show_dev(struct device *d static struct device_attribute devt_attr = __ATTR(dev, S_IRUGO, show_dev, NULL); -/* kset to create /sys/devices/ */ +#ifndef CONFIG_VE struct kset *devices_kset; +#endif /** * device_create_file - create sysfs attribute file for device. @@ -557,7 +567,7 @@ static void klist_children_put(struct kl */ void device_initialize(struct device *dev) { - dev->kobj.kset = devices_kset; + dev->kobj.kset = ve_devices_kset; kobject_init(&dev->kobj, &device_ktype); INIT_LIST_HEAD(&dev->dma_pools); init_MUTEX(&dev->sem); @@ -568,8 +578,7 @@ void device_initialize(struct device *de set_dev_node(dev, -1); } -#ifdef CONFIG_SYSFS_DEPRECATED -static struct kobject *get_device_parent(struct device *dev, +static struct kobject *get_device_parent_dep(struct device *dev, struct device *parent) { /* class devices without a parent live in /sys/class// */ @@ -582,22 +591,25 @@ static struct kobject *get_device_parent return NULL; } -static inline void cleanup_device_parent(struct device *dev) {} -static inline void cleanup_glue_dir(struct device *dev, +static inline void cleanup_device_parent_dep(struct device *dev) {} +static inline void cleanup_glue_dir_dep(struct device *dev, struct kobject *glue_dir) {} +#ifndef CONFIG_VE +static struct kobject *virtual_dir = NULL; #else +# define virtual_dir (get_exec_env()->_virtual_dir) +#endif + static struct kobject *virtual_device_parent(struct device *dev) { - static struct kobject *virtual_dir = NULL; - if (!virtual_dir) virtual_dir = kobject_create_and_add("virtual", - &devices_kset->kobj); + &ve_devices_kset->kobj); return virtual_dir; } -static struct kobject *get_device_parent(struct device *dev, +static struct kobject *get_device_parent_nodep(struct device *dev, struct device *parent) { int retval; @@ -658,7 +670,7 @@ static struct kobject *get_device_parent return NULL; } -static void cleanup_glue_dir(struct device *dev, struct kobject *glue_dir) +static void cleanup_glue_dir_nodep(struct device *dev, struct kobject *glue_dir) { /* see if we live in a "glue" directory */ if (!glue_dir || !dev->class || @@ -668,11 +680,36 @@ static void cleanup_glue_dir(struct devi kobject_put(glue_dir); } +static void cleanup_device_parent_nodep(struct device *dev) +{ + cleanup_glue_dir_nodep(dev, dev->kobj.parent); +} + +static struct kobject *get_device_parent(struct device *dev, + struct device *parent) +{ + if (sysfs_deprecated) + return get_device_parent_dep(dev, parent); + else + return get_device_parent_nodep(dev, parent); +} + +static void cleanup_glue_dir(struct device *dev, struct kobject *glue_dir) +{ + if (sysfs_deprecated) + cleanup_glue_dir_dep(dev, glue_dir); + else + cleanup_glue_dir_nodep(dev, glue_dir); +} + static void cleanup_device_parent(struct device *dev) { - cleanup_glue_dir(dev, dev->kobj.parent); + if (sysfs_deprecated) + cleanup_device_parent_dep(dev); + else + cleanup_device_parent_nodep(dev); } -#endif + static void setup_parent(struct device *dev, struct device *parent) { @@ -695,7 +732,9 @@ static int device_add_class_symlinks(str if (error) goto out; -#ifdef CONFIG_SYSFS_DEPRECATED + if (!sysfs_deprecated) + goto nodep; + /* stacked class devices need a symlink in the class directory */ if (dev->kobj.parent != &dev->class->p->class_subsys.kobj && device_is_not_partition(dev)) { @@ -720,7 +759,7 @@ static int device_add_class_symlinks(str &parent->kobj, "device"); if (error) - goto out_busid; + goto out_busid_dep; class_name = make_class_name(dev->class->name, &dev->kobj); @@ -736,12 +775,14 @@ static int device_add_class_symlinks(str out_device: if (dev->parent && device_is_not_partition(dev)) sysfs_remove_link(&dev->kobj, "device"); -out_busid: +out_busid_dep: if (dev->kobj.parent != &dev->class->p->class_subsys.kobj && device_is_not_partition(dev)) sysfs_remove_link(&dev->class->p->class_subsys.kobj, dev_name(dev)); -#else + goto out_subsys; + +nodep: /* link in the class directory pointing to the device */ error = sysfs_create_link(&dev->class->p->class_subsys.kobj, &dev->kobj, dev_name(dev)); @@ -752,14 +793,12 @@ out_busid: error = sysfs_create_link(&dev->kobj, &dev->parent->kobj, "device"); if (error) - goto out_busid; + goto out_busid_nodep; } return 0; -out_busid: +out_busid_nodep: sysfs_remove_link(&dev->class->p->class_subsys.kobj, dev_name(dev)); -#endif - out_subsys: sysfs_remove_link(&dev->kobj, "subsystem"); out: @@ -771,7 +810,9 @@ static void device_remove_class_symlinks if (!dev->class) return; -#ifdef CONFIG_SYSFS_DEPRECATED + if (!sysfs_deprecated) + goto nodep; + if (dev->parent && device_is_not_partition(dev)) { char *class_name; @@ -787,13 +828,14 @@ static void device_remove_class_symlinks device_is_not_partition(dev)) sysfs_remove_link(&dev->class->p->class_subsys.kobj, dev_name(dev)); -#else + goto done; + +nodep: if (dev->parent && device_is_not_partition(dev)) sysfs_remove_link(&dev->kobj, "device"); sysfs_remove_link(&dev->class->p->class_subsys.kobj, dev_name(dev)); -#endif - +done: sysfs_remove_link(&dev->kobj, "subsystem"); } @@ -832,7 +874,7 @@ static struct kobject *device_to_dev_kob if (dev->class) kobj = dev->class->dev_kobj; else - kobj = sysfs_dev_char_kobj; + kobj = ve_sysfs_dev_char_kobj; return kobj; } @@ -1270,31 +1312,43 @@ struct device *device_find_child(struct return child; } -int __init devices_init(void) +int devices_init(void) { - devices_kset = kset_create_and_add("devices", &device_uevent_ops, NULL); - if (!devices_kset) - return -ENOMEM; - dev_kobj = kobject_create_and_add("dev", NULL); - if (!dev_kobj) + ve_devices_kset = kset_create_and_add("devices", &device_uevent_ops, NULL); + if (!ve_devices_kset) + goto dev_kset_err; + ve_dev_kobj = kobject_create_and_add("dev", NULL); + if (!ve_dev_kobj) goto dev_kobj_err; - sysfs_dev_block_kobj = kobject_create_and_add("block", dev_kobj); - if (!sysfs_dev_block_kobj) + ve_sysfs_dev_block_kobj = kobject_create_and_add("block", ve_dev_kobj); + if (!ve_sysfs_dev_block_kobj) goto block_kobj_err; - sysfs_dev_char_kobj = kobject_create_and_add("char", dev_kobj); - if (!sysfs_dev_char_kobj) + ve_sysfs_dev_char_kobj = kobject_create_and_add("char", ve_dev_kobj); + if (!ve_sysfs_dev_char_kobj) goto char_kobj_err; return 0; char_kobj_err: - kobject_put(sysfs_dev_block_kobj); + kobject_put(ve_sysfs_dev_block_kobj); block_kobj_err: - kobject_put(dev_kobj); + kobject_put(ve_dev_kobj); dev_kobj_err: - kset_unregister(devices_kset); + kset_unregister(ve_devices_kset); +dev_kset_err: return -ENOMEM; } +EXPORT_SYMBOL_GPL(devices_init); + +void devices_fini(void) +{ + kobject_put(ve_sysfs_dev_char_kobj); + kobject_put(ve_sysfs_dev_block_kobj); + kobject_put(ve_dev_kobj); + kset_unregister(ve_devices_kset); +} +EXPORT_SYMBOL_GPL(devices_fini); + EXPORT_SYMBOL_GPL(device_for_each_child); EXPORT_SYMBOL_GPL(device_find_child); @@ -1556,10 +1610,8 @@ int device_rename(struct device *dev, ch pr_debug("device: '%s': %s: renaming to '%s'\n", dev_name(dev), __func__, new_name); -#ifdef CONFIG_SYSFS_DEPRECATED - if ((dev->class) && (dev->parent)) + if (sysfs_deprecated && (dev->class) && (dev->parent)) old_class_name = make_class_name(dev->class->name, &dev->kobj); -#endif old_device_name = kstrdup(dev_name(dev), GFP_KERNEL); if (!old_device_name) { @@ -1571,8 +1623,7 @@ int device_rename(struct device *dev, ch if (error) goto out; -#ifdef CONFIG_SYSFS_DEPRECATED - if (old_class_name) { + if (sysfs_deprecated && old_class_name) { new_class_name = make_class_name(dev->class->name, &dev->kobj); if (new_class_name) { error = sysfs_create_link_nowarn(&dev->parent->kobj, @@ -1583,8 +1634,7 @@ int device_rename(struct device *dev, ch sysfs_remove_link(&dev->parent->kobj, old_class_name); } } -#else - if (dev->class) { + if (!sysfs_deprecated && dev->class) { error = sysfs_create_link_nowarn(&dev->class->p->class_subsys.kobj, &dev->kobj, dev_name(dev)); if (error) @@ -1592,7 +1642,6 @@ int device_rename(struct device *dev, ch sysfs_remove_link(&dev->class->p->class_subsys.kobj, old_device_name); } -#endif out: put_device(dev); @@ -1610,9 +1659,11 @@ static int device_move_class_links(struc struct device *new_parent) { int error = 0; -#ifdef CONFIG_SYSFS_DEPRECATED char *class_name; + if (!sysfs_deprecated) + goto nodep; + class_name = make_class_name(dev->class->name, &dev->kobj); if (!class_name) { error = -ENOMEM; @@ -1636,14 +1687,14 @@ static int device_move_class_links(struc out: kfree(class_name); return error; -#else + +nodep: if (old_parent) sysfs_remove_link(&dev->kobj, "device"); if (new_parent) error = sysfs_create_link(&dev->kobj, &new_parent->kobj, "device"); return error; -#endif } /** @@ -1734,7 +1785,12 @@ void device_shutdown(void) { struct device *dev, *devn; - list_for_each_entry_safe_reverse(dev, devn, &devices_kset->list, + if (!ve_is_super(get_exec_env())) { + printk("BUG: device_shutdown call from inside VE\n"); + return; + } + + list_for_each_entry_safe_reverse(dev, devn, &ve_devices_kset->list, kobj.entry) { if (dev->bus && dev->bus->shutdown) { dev_dbg(dev, "shutdown\n"); @@ -1744,8 +1800,9 @@ void device_shutdown(void) dev->driver->shutdown(dev); } } - kobject_put(sysfs_dev_char_kobj); - kobject_put(sysfs_dev_block_kobj); - kobject_put(dev_kobj); + + kobject_put(ve_sysfs_dev_char_kobj); + kobject_put(ve_sysfs_dev_block_kobj); + kobject_put(ve_dev_kobj); async_synchronize_full(); } diff -urNp linux-2.6.32.48/drivers/base/Kconfig linux-2.6.32.48-openvz/drivers/base/Kconfig --- linux-2.6.32.48/drivers/base/Kconfig 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/base/Kconfig 2011-11-17 20:08:53.000000000 -0500 @@ -10,7 +10,7 @@ config UEVENT_HELPER_PATH config DEVTMPFS bool "Create a kernel maintained /dev tmpfs (EXPERIMENTAL)" - depends on HOTPLUG && SHMEM && TMPFS + depends on HOTPLUG && SHMEM && TMPFS && !VE help This creates a tmpfs filesystem, and mounts it at bootup and mounts it at /dev. The kernel driver core creates device diff -urNp linux-2.6.32.48/drivers/base/memory.c linux-2.6.32.48-openvz/drivers/base/memory.c --- linux-2.6.32.48/drivers/base/memory.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/base/memory.c 2011-11-17 20:08:53.000000000 -0500 @@ -44,7 +44,7 @@ static int memory_uevent(struct kset *ks return retval; } -static struct kset_uevent_ops memory_uevent_ops = { +static const struct kset_uevent_ops memory_uevent_ops = { .name = memory_uevent_name, .uevent = memory_uevent, }; diff -urNp linux-2.6.32.48/drivers/base/sys.c linux-2.6.32.48-openvz/drivers/base/sys.c --- linux-2.6.32.48/drivers/base/sys.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/base/sys.c 2011-11-17 20:08:53.000000000 -0500 @@ -20,6 +20,8 @@ #include #include #include +#include +#include #include #include #include @@ -54,7 +56,7 @@ sysdev_store(struct kobject *kobj, struc return -EIO; } -static struct sysfs_ops sysfs_ops = { +static const struct sysfs_ops sysfs_ops = { .show = sysdev_show, .store = sysdev_store, }; @@ -104,7 +106,7 @@ static ssize_t sysdev_class_store(struct return -EIO; } -static struct sysfs_ops sysfs_class_ops = { +static const struct sysfs_ops sysfs_class_ops = { .show = sysdev_class_show, .store = sysdev_class_store, }; @@ -494,7 +496,7 @@ EXPORT_SYMBOL_GPL(sysdev_resume); int __init system_bus_init(void) { - system_kset = kset_create_and_add("system", NULL, &devices_kset->kobj); + system_kset = kset_create_and_add("system", NULL, &ve_devices_kset->kobj); if (!system_kset) return -ENOMEM; return 0; diff -urNp linux-2.6.32.48/drivers/block/cciss.c linux-2.6.32.48-openvz/drivers/block/cciss.c --- linux-2.6.32.48/drivers/block/cciss.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/block/cciss.c 2011-11-17 20:09:04.000000000 -0500 @@ -1011,6 +1011,8 @@ static int cciss_ioctl32_passthru(struct int err; u32 cp; + memset(&arg64, 0, sizeof(arg64)); + err = 0; err |= copy_from_user(&arg64.LUN_info, &arg32->LUN_info, @@ -2852,7 +2854,7 @@ static unsigned long pollcomplete(int ct /* Wait (up to 20 seconds) for a command to complete */ for (i = 20 * HZ; i > 0; i--) { - done = hba[ctlr]->access.command_completed(hba[ctlr]); + done = hba[ctlr]->access->command_completed(hba[ctlr]); if (done == FIFO_EMPTY) schedule_timeout_uninterruptible(1); else @@ -2876,7 +2878,7 @@ static int sendcmd_core(ctlr_info_t *h, resend_cmd1: /* Disable interrupt on the board. */ - h->access.set_intr_mask(h, CCISS_INTR_OFF); + h->access->set_intr_mask(h, CCISS_INTR_OFF); /* Make sure there is room in the command FIFO */ /* Actually it should be completely empty at this time */ @@ -2884,13 +2886,13 @@ resend_cmd1: /* tape side of the driver. */ for (i = 200000; i > 0; i--) { /* if fifo isn't full go */ - if (!(h->access.fifo_full(h))) + if (!(h->access->fifo_full(h))) break; udelay(10); printk(KERN_WARNING "cciss cciss%d: SendCmd FIFO full," " waiting!\n", h->ctlr); } - h->access.submit_command(h, c); /* Send the cmd */ + h->access->submit_command(h, c); /* Send the cmd */ do { complete = pollcomplete(h->ctlr); @@ -3023,7 +3025,7 @@ static void start_io(ctlr_info_t *h) while (!hlist_empty(&h->reqQ)) { c = hlist_entry(h->reqQ.first, CommandList_struct, list); /* can't do anything if fifo is full */ - if ((h->access.fifo_full(h))) { + if ((h->access->fifo_full(h))) { printk(KERN_WARNING "cciss: fifo full\n"); break; } @@ -3033,7 +3035,7 @@ static void start_io(ctlr_info_t *h) h->Qdepth--; /* Tell the controller execute command */ - h->access.submit_command(h, c); + h->access->submit_command(h, c); /* Put job onto the completed Q */ addQ(&h->cmpQ, c); @@ -3393,17 +3395,17 @@ startio: static inline unsigned long get_next_completion(ctlr_info_t *h) { - return h->access.command_completed(h); + return h->access->command_completed(h); } static inline int interrupt_pending(ctlr_info_t *h) { - return h->access.intr_pending(h); + return h->access->intr_pending(h); } static inline long interrupt_not_for_us(ctlr_info_t *h) { - return (((h->access.intr_pending(h) == 0) || + return (((h->access->intr_pending(h) == 0) || (h->interrupts_enabled == 0))); } @@ -3892,7 +3894,7 @@ static int __devinit cciss_pci_init(ctlr */ c->max_commands = readl(&(c->cfgtable->CmdsOutMax)); c->product_name = products[prod_index].product_name; - c->access = *(products[prod_index].access); + c->access = products[prod_index].access; c->nr_cmds = c->max_commands - 4; if ((readb(&c->cfgtable->Signature[0]) != 'C') || (readb(&c->cfgtable->Signature[1]) != 'I') || @@ -4291,7 +4293,7 @@ static int __devinit cciss_init_one(stru } /* make sure the board interrupts are off */ - hba[i]->access.set_intr_mask(hba[i], CCISS_INTR_OFF); + hba[i]->access->set_intr_mask(hba[i], CCISS_INTR_OFF); if (request_irq(hba[i]->intr[SIMPLE_MODE_INT], do_cciss_intr, IRQF_DISABLED | IRQF_SHARED, hba[i]->devname, hba[i])) { printk(KERN_ERR "cciss: Unable to get irq %d for %s\n", @@ -4341,7 +4343,7 @@ static int __devinit cciss_init_one(stru cciss_scsi_setup(i); /* Turn the interrupts on so we can service requests */ - hba[i]->access.set_intr_mask(hba[i], CCISS_INTR_ON); + hba[i]->access->set_intr_mask(hba[i], CCISS_INTR_ON); /* Get the firmware version */ inq_buff = kzalloc(sizeof(InquiryData_struct), GFP_KERNEL); diff -urNp linux-2.6.32.48/drivers/block/cciss.h linux-2.6.32.48-openvz/drivers/block/cciss.h --- linux-2.6.32.48/drivers/block/cciss.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/block/cciss.h 2011-11-17 20:09:04.000000000 -0500 @@ -90,7 +90,7 @@ struct ctlr_info // information about each logical volume drive_info_struct *drv[CISS_MAX_LUN]; - struct access_method access; + struct access_method *access; /* queue and queue Info */ struct hlist_head reqQ; diff -urNp linux-2.6.32.48/drivers/block/cpqarray.c linux-2.6.32.48-openvz/drivers/block/cpqarray.c --- linux-2.6.32.48/drivers/block/cpqarray.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/block/cpqarray.c 2011-11-17 20:09:04.000000000 -0500 @@ -402,7 +402,7 @@ static int __init cpqarray_register_ctlr if (register_blkdev(COMPAQ_SMART2_MAJOR+i, hba[i]->devname)) { goto Enomem4; } - hba[i]->access.set_intr_mask(hba[i], 0); + hba[i]->access->set_intr_mask(hba[i], 0); if (request_irq(hba[i]->intr, do_ida_intr, IRQF_DISABLED|IRQF_SHARED, hba[i]->devname, hba[i])) { @@ -460,7 +460,7 @@ static int __init cpqarray_register_ctlr add_timer(&hba[i]->timer); /* Enable IRQ now that spinlock and rate limit timer are set up */ - hba[i]->access.set_intr_mask(hba[i], FIFO_NOT_EMPTY); + hba[i]->access->set_intr_mask(hba[i], FIFO_NOT_EMPTY); for(j=0; jproduct_name = products[i].product_name; - c->access = *(products[i].access); + c->access = products[i].access; break; } } @@ -793,7 +793,7 @@ static int __init cpqarray_eisa_detect(v hba[ctlr]->intr = intr; sprintf(hba[ctlr]->devname, "ida%d", nr_ctlr); hba[ctlr]->product_name = products[j].product_name; - hba[ctlr]->access = *(products[j].access); + hba[ctlr]->access = products[j].access; hba[ctlr]->ctlr = ctlr; hba[ctlr]->board_id = board_id; hba[ctlr]->pci_dev = NULL; /* not PCI */ @@ -896,6 +896,8 @@ static void do_ida_request(struct reques struct scatterlist tmp_sg[SG_MAX]; int i, dir, seg; + pax_track_stack(); + if (blk_queue_plugged(q)) goto startio; @@ -968,7 +970,7 @@ static void start_io(ctlr_info_t *h) while((c = h->reqQ) != NULL) { /* Can't do anything if we're busy */ - if (h->access.fifo_full(h) == 0) + if (h->access->fifo_full(h) == 0) return; /* Get the first entry from the request Q */ @@ -976,7 +978,7 @@ static void start_io(ctlr_info_t *h) h->Qdepth--; /* Tell the controller to do our bidding */ - h->access.submit_command(h, c); + h->access->submit_command(h, c); /* Get onto the completion Q */ addQ(&h->cmpQ, c); @@ -1038,7 +1040,7 @@ static irqreturn_t do_ida_intr(int irq, unsigned long flags; __u32 a,a1; - istat = h->access.intr_pending(h); + istat = h->access->intr_pending(h); /* Is this interrupt for us? */ if (istat == 0) return IRQ_NONE; @@ -1049,7 +1051,7 @@ static irqreturn_t do_ida_intr(int irq, */ spin_lock_irqsave(IDA_LOCK(h->ctlr), flags); if (istat & FIFO_NOT_EMPTY) { - while((a = h->access.command_completed(h))) { + while((a = h->access->command_completed(h))) { a1 = a; a &= ~3; if ((c = h->cmpQ) == NULL) { @@ -1434,11 +1436,11 @@ static int sendcmd( /* * Disable interrupt */ - info_p->access.set_intr_mask(info_p, 0); + info_p->access->set_intr_mask(info_p, 0); /* Make sure there is room in the command FIFO */ /* Actually it should be completely empty at this time. */ for (i = 200000; i > 0; i--) { - temp = info_p->access.fifo_full(info_p); + temp = info_p->access->fifo_full(info_p); if (temp != 0) { break; } @@ -1451,7 +1453,7 @@ DBG( /* * Send the cmd */ - info_p->access.submit_command(info_p, c); + info_p->access->submit_command(info_p, c); complete = pollcomplete(ctlr); pci_unmap_single(info_p->pci_dev, (dma_addr_t) c->req.sg[0].addr, @@ -1534,9 +1536,9 @@ static int revalidate_allvol(ctlr_info_t * we check the new geometry. Then turn interrupts back on when * we're done. */ - host->access.set_intr_mask(host, 0); + host->access->set_intr_mask(host, 0); getgeometry(ctlr); - host->access.set_intr_mask(host, FIFO_NOT_EMPTY); + host->access->set_intr_mask(host, FIFO_NOT_EMPTY); for(i=0; i 0; i--) { - done = hba[ctlr]->access.command_completed(hba[ctlr]); + done = hba[ctlr]->access->command_completed(hba[ctlr]); if (done == 0) { udelay(10); /* a short fixed delay */ } else diff -urNp linux-2.6.32.48/drivers/block/cpqarray.h linux-2.6.32.48-openvz/drivers/block/cpqarray.h --- linux-2.6.32.48/drivers/block/cpqarray.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/block/cpqarray.h 2011-11-17 20:09:04.000000000 -0500 @@ -99,7 +99,7 @@ struct ctlr_info { drv_info_t drv[NWD]; struct proc_dir_entry *proc; - struct access_method access; + struct access_method *access; cmdlist_t *reqQ; cmdlist_t *cmpQ; diff -urNp linux-2.6.32.48/drivers/block/DAC960.c linux-2.6.32.48-openvz/drivers/block/DAC960.c --- linux-2.6.32.48/drivers/block/DAC960.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/block/DAC960.c 2011-11-17 20:08:53.000000000 -0500 @@ -1973,6 +1973,8 @@ static bool DAC960_V1_ReadDeviceConfigur unsigned long flags; int Channel, TargetID; + pax_track_stack(); + if (!init_dma_loaf(Controller->PCIDevice, &local_dma, DAC960_V1_MaxChannels*(sizeof(DAC960_V1_DCDB_T) + sizeof(DAC960_SCSI_Inquiry_T) + diff -urNp linux-2.6.32.48/drivers/block/loop.c linux-2.6.32.48-openvz/drivers/block/loop.c --- linux-2.6.32.48/drivers/block/loop.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/block/loop.c 2011-11-17 20:09:04.000000000 -0500 @@ -282,7 +282,7 @@ static int __do_lo_send_write(struct fil mm_segment_t old_fs = get_fs(); set_fs(get_ds()); - bw = file->f_op->write(file, buf, len, &pos); + bw = file->f_op->write(file, (const char __force_user *)buf, len, &pos); set_fs(old_fs); if (likely(bw == len)) return 0; diff -urNp linux-2.6.32.48/drivers/block/nbd.c linux-2.6.32.48-openvz/drivers/block/nbd.c --- linux-2.6.32.48/drivers/block/nbd.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/block/nbd.c 2011-11-17 20:09:04.000000000 -0500 @@ -155,6 +155,8 @@ static int sock_xmit(struct nbd_device * struct kvec iov; sigset_t blocked, oldset; + pax_track_stack(); + if (unlikely(!sock)) { printk(KERN_ERR "%s: Attempted %s on closed socket in sock_xmit\n", lo->disk->disk_name, (send ? "send" : "recv")); @@ -569,6 +571,8 @@ static void do_nbd_request(struct reques static int __nbd_ioctl(struct block_device *bdev, struct nbd_device *lo, unsigned int cmd, unsigned long arg) { + pax_track_stack(); + switch (cmd) { case NBD_DISCONNECT: { struct request sreq; diff -urNp linux-2.6.32.48/drivers/block/pktcdvd.c linux-2.6.32.48-openvz/drivers/block/pktcdvd.c --- linux-2.6.32.48/drivers/block/pktcdvd.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/block/pktcdvd.c 2011-11-17 20:08:53.000000000 -0500 @@ -284,7 +284,7 @@ static ssize_t kobj_pkt_store(struct kob return len; } -static struct sysfs_ops kobj_pkt_ops = { +static const struct sysfs_ops kobj_pkt_ops = { .show = kobj_pkt_show, .store = kobj_pkt_store }; diff -urNp linux-2.6.32.48/drivers/char/agp/frontend.c linux-2.6.32.48-openvz/drivers/char/agp/frontend.c --- linux-2.6.32.48/drivers/char/agp/frontend.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/char/agp/frontend.c 2011-11-17 20:08:53.000000000 -0500 @@ -824,7 +824,7 @@ static int agpioc_reserve_wrap(struct ag if (copy_from_user(&reserve, arg, sizeof(struct agp_region))) return -EFAULT; - if ((unsigned) reserve.seg_count >= ~0U/sizeof(struct agp_segment)) + if ((unsigned) reserve.seg_count >= ~0U/sizeof(struct agp_segment_priv)) return -EFAULT; client = agp_find_client_by_pid(reserve.pid); diff -urNp linux-2.6.32.48/drivers/char/briq_panel.c linux-2.6.32.48-openvz/drivers/char/briq_panel.c --- linux-2.6.32.48/drivers/char/briq_panel.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/char/briq_panel.c 2011-11-17 20:08:53.000000000 -0500 @@ -10,6 +10,7 @@ #include #include #include +#include #include #include #include @@ -36,6 +37,7 @@ static int vfd_is_open; static unsigned char vfd[40]; static int vfd_cursor; static unsigned char ledpb, led; +static DEFINE_MUTEX(vfd_mutex); static void update_vfd(void) { @@ -142,12 +144,15 @@ static ssize_t briq_panel_write(struct f if (!vfd_is_open) return -EBUSY; + mutex_lock(&vfd_mutex); for (;;) { char c; if (!indx) break; - if (get_user(c, buf)) + if (get_user(c, buf)) { + mutex_unlock(&vfd_mutex); return -EFAULT; + } if (esc) { set_led(c); esc = 0; @@ -177,6 +182,7 @@ static ssize_t briq_panel_write(struct f buf++; } update_vfd(); + mutex_unlock(&vfd_mutex); return len; } diff -urNp linux-2.6.32.48/drivers/char/genrtc.c linux-2.6.32.48-openvz/drivers/char/genrtc.c --- linux-2.6.32.48/drivers/char/genrtc.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/char/genrtc.c 2011-11-17 20:08:53.000000000 -0500 @@ -272,6 +272,7 @@ static int gen_rtc_ioctl(struct inode *i switch (cmd) { case RTC_PLL_GET: + memset(&pll, 0, sizeof(pll)); if (get_rtc_pll(&pll)) return -EINVAL; else diff -urNp linux-2.6.32.48/drivers/char/hpet.c linux-2.6.32.48-openvz/drivers/char/hpet.c --- linux-2.6.32.48/drivers/char/hpet.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/char/hpet.c 2011-11-17 20:08:53.000000000 -0500 @@ -430,7 +430,7 @@ static int hpet_release(struct inode *in return 0; } -static int hpet_ioctl_common(struct hpet_dev *, int, unsigned long, int); +static int hpet_ioctl_common(struct hpet_dev *, unsigned int, unsigned long, int); static int hpet_ioctl(struct inode *inode, struct file *file, unsigned int cmd, @@ -565,7 +565,7 @@ static inline unsigned long hpet_time_di } static int -hpet_ioctl_common(struct hpet_dev *devp, int cmd, unsigned long arg, int kernel) +hpet_ioctl_common(struct hpet_dev *devp, unsigned int cmd, unsigned long arg, int kernel) { struct hpet_timer __iomem *timer; struct hpet __iomem *hpet; @@ -608,11 +608,11 @@ hpet_ioctl_common(struct hpet_dev *devp, { struct hpet_info info; + memset(&info, 0, sizeof(info)); + if (devp->hd_ireqfreq) info.hi_ireqfreq = hpet_time_div(hpetp, devp->hd_ireqfreq); - else - info.hi_ireqfreq = 0; info.hi_flags = readq(&timer->hpet_config) & Tn_PER_INT_CAP_MASK; info.hi_hpet = hpetp->hp_which; diff -urNp linux-2.6.32.48/drivers/char/hvc_beat.c linux-2.6.32.48-openvz/drivers/char/hvc_beat.c --- linux-2.6.32.48/drivers/char/hvc_beat.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/char/hvc_beat.c 2011-11-17 20:08:53.000000000 -0500 @@ -84,7 +84,7 @@ static int hvc_beat_put_chars(uint32_t v return cnt; } -static struct hv_ops hvc_beat_get_put_ops = { +static const struct hv_ops hvc_beat_get_put_ops = { .get_chars = hvc_beat_get_chars, .put_chars = hvc_beat_put_chars, }; diff -urNp linux-2.6.32.48/drivers/char/hvc_console.c linux-2.6.32.48-openvz/drivers/char/hvc_console.c --- linux-2.6.32.48/drivers/char/hvc_console.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/char/hvc_console.c 2011-11-17 20:09:04.000000000 -0500 @@ -125,7 +125,7 @@ static struct hvc_struct *hvc_get_by_ind * console interfaces but can still be used as a tty device. This has to be * static because kmalloc will not work during early console init. */ -static struct hv_ops *cons_ops[MAX_NR_HVC_CONSOLES]; +static const struct hv_ops *cons_ops[MAX_NR_HVC_CONSOLES]; static uint32_t vtermnos[MAX_NR_HVC_CONSOLES] = {[0 ... MAX_NR_HVC_CONSOLES - 1] = -1}; @@ -249,7 +249,7 @@ static void destroy_hvc_struct(struct kr * vty adapters do NOT get an hvc_instantiate() callback since they * appear after early console init. */ -int hvc_instantiate(uint32_t vtermno, int index, struct hv_ops *ops) +int hvc_instantiate(uint32_t vtermno, int index, const struct hv_ops *ops) { struct hvc_struct *hp; @@ -758,7 +758,7 @@ static const struct tty_operations hvc_o }; struct hvc_struct __devinit *hvc_alloc(uint32_t vtermno, int data, - struct hv_ops *ops, int outbuf_size) + const struct hv_ops *ops, int outbuf_size) { struct hvc_struct *hp; int i; diff -urNp linux-2.6.32.48/drivers/char/hvc_console.h linux-2.6.32.48-openvz/drivers/char/hvc_console.h --- linux-2.6.32.48/drivers/char/hvc_console.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/char/hvc_console.h 2011-11-17 20:08:53.000000000 -0500 @@ -55,7 +55,7 @@ struct hvc_struct { int outbuf_size; int n_outbuf; uint32_t vtermno; - struct hv_ops *ops; + const struct hv_ops *ops; int irq_requested; int data; struct winsize ws; @@ -76,11 +76,11 @@ struct hv_ops { }; /* Register a vterm and a slot index for use as a console (console_init) */ -extern int hvc_instantiate(uint32_t vtermno, int index, struct hv_ops *ops); +extern int hvc_instantiate(uint32_t vtermno, int index, const struct hv_ops *ops); /* register a vterm for hvc tty operation (module_init or hotplug add) */ extern struct hvc_struct * __devinit hvc_alloc(uint32_t vtermno, int data, - struct hv_ops *ops, int outbuf_size); + const struct hv_ops *ops, int outbuf_size); /* remove a vterm from hvc tty operation (module_exit or hotplug remove) */ extern int hvc_remove(struct hvc_struct *hp); diff -urNp linux-2.6.32.48/drivers/char/hvc_iseries.c linux-2.6.32.48-openvz/drivers/char/hvc_iseries.c --- linux-2.6.32.48/drivers/char/hvc_iseries.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/char/hvc_iseries.c 2011-11-17 20:08:53.000000000 -0500 @@ -197,7 +197,7 @@ done: return sent; } -static struct hv_ops hvc_get_put_ops = { +static const struct hv_ops hvc_get_put_ops = { .get_chars = get_chars, .put_chars = put_chars, .notifier_add = notifier_add_irq, diff -urNp linux-2.6.32.48/drivers/char/hvc_iucv.c linux-2.6.32.48-openvz/drivers/char/hvc_iucv.c --- linux-2.6.32.48/drivers/char/hvc_iucv.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/char/hvc_iucv.c 2011-11-17 20:08:53.000000000 -0500 @@ -924,7 +924,7 @@ static int hvc_iucv_pm_restore_thaw(stru /* HVC operations */ -static struct hv_ops hvc_iucv_ops = { +static const struct hv_ops hvc_iucv_ops = { .get_chars = hvc_iucv_get_chars, .put_chars = hvc_iucv_put_chars, .notifier_add = hvc_iucv_notifier_add, diff -urNp linux-2.6.32.48/drivers/char/hvc_rtas.c linux-2.6.32.48-openvz/drivers/char/hvc_rtas.c --- linux-2.6.32.48/drivers/char/hvc_rtas.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/char/hvc_rtas.c 2011-11-17 20:08:53.000000000 -0500 @@ -71,7 +71,7 @@ static int hvc_rtas_read_console(uint32_ return i; } -static struct hv_ops hvc_rtas_get_put_ops = { +static const struct hv_ops hvc_rtas_get_put_ops = { .get_chars = hvc_rtas_read_console, .put_chars = hvc_rtas_write_console, }; diff -urNp linux-2.6.32.48/drivers/char/hvcs.c linux-2.6.32.48-openvz/drivers/char/hvcs.c --- linux-2.6.32.48/drivers/char/hvcs.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/char/hvcs.c 2011-11-17 20:08:53.000000000 -0500 @@ -82,6 +82,7 @@ #include #include #include +#include /* * 1.3.0 -> 1.3.1 In hvcs_open memset(..,0x00,..) instead of memset(..,0x3F,00). @@ -269,7 +270,7 @@ struct hvcs_struct { unsigned int index; struct tty_struct *tty; - int open_count; + local_t open_count; /* * Used to tell the driver kernel_thread what operations need to take @@ -419,7 +420,7 @@ static ssize_t hvcs_vterm_state_store(st spin_lock_irqsave(&hvcsd->lock, flags); - if (hvcsd->open_count > 0) { + if (local_read(&hvcsd->open_count) > 0) { spin_unlock_irqrestore(&hvcsd->lock, flags); printk(KERN_INFO "HVCS: vterm state unchanged. " "The hvcs device node is still in use.\n"); @@ -1135,7 +1136,7 @@ static int hvcs_open(struct tty_struct * if ((retval = hvcs_partner_connect(hvcsd))) goto error_release; - hvcsd->open_count = 1; + local_set(&hvcsd->open_count, 1); hvcsd->tty = tty; tty->driver_data = hvcsd; @@ -1169,7 +1170,7 @@ fast_open: spin_lock_irqsave(&hvcsd->lock, flags); kref_get(&hvcsd->kref); - hvcsd->open_count++; + local_inc(&hvcsd->open_count); hvcsd->todo_mask |= HVCS_SCHED_READ; spin_unlock_irqrestore(&hvcsd->lock, flags); @@ -1213,7 +1214,7 @@ static void hvcs_close(struct tty_struct hvcsd = tty->driver_data; spin_lock_irqsave(&hvcsd->lock, flags); - if (--hvcsd->open_count == 0) { + if (local_dec_and_test(&hvcsd->open_count)) { vio_disable_interrupts(hvcsd->vdev); @@ -1239,10 +1240,10 @@ static void hvcs_close(struct tty_struct free_irq(irq, hvcsd); kref_put(&hvcsd->kref, destroy_hvcs_struct); return; - } else if (hvcsd->open_count < 0) { + } else if (local_read(&hvcsd->open_count) < 0) { printk(KERN_ERR "HVCS: vty-server@%X open_count: %d" " is missmanaged.\n", - hvcsd->vdev->unit_address, hvcsd->open_count); + hvcsd->vdev->unit_address, local_read(&hvcsd->open_count)); } spin_unlock_irqrestore(&hvcsd->lock, flags); @@ -1258,7 +1259,7 @@ static void hvcs_hangup(struct tty_struc spin_lock_irqsave(&hvcsd->lock, flags); /* Preserve this so that we know how many kref refs to put */ - temp_open_count = hvcsd->open_count; + temp_open_count = local_read(&hvcsd->open_count); /* * Don't kref put inside the spinlock because the destruction @@ -1273,7 +1274,7 @@ static void hvcs_hangup(struct tty_struc hvcsd->tty->driver_data = NULL; hvcsd->tty = NULL; - hvcsd->open_count = 0; + local_set(&hvcsd->open_count, 0); /* This will drop any buffered data on the floor which is OK in a hangup * scenario. */ @@ -1344,7 +1345,7 @@ static int hvcs_write(struct tty_struct * the middle of a write operation? This is a crummy place to do this * but we want to keep it all in the spinlock. */ - if (hvcsd->open_count <= 0) { + if (local_read(&hvcsd->open_count) <= 0) { spin_unlock_irqrestore(&hvcsd->lock, flags); return -ENODEV; } @@ -1418,7 +1419,7 @@ static int hvcs_write_room(struct tty_st { struct hvcs_struct *hvcsd = tty->driver_data; - if (!hvcsd || hvcsd->open_count <= 0) + if (!hvcsd || local_read(&hvcsd->open_count) <= 0) return 0; return HVCS_BUFF_LEN - hvcsd->chars_in_buffer; diff -urNp linux-2.6.32.48/drivers/char/hvc_udbg.c linux-2.6.32.48-openvz/drivers/char/hvc_udbg.c --- linux-2.6.32.48/drivers/char/hvc_udbg.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/char/hvc_udbg.c 2011-11-17 20:08:53.000000000 -0500 @@ -58,7 +58,7 @@ static int hvc_udbg_get(uint32_t vtermno return i; } -static struct hv_ops hvc_udbg_ops = { +static const struct hv_ops hvc_udbg_ops = { .get_chars = hvc_udbg_get, .put_chars = hvc_udbg_put, }; diff -urNp linux-2.6.32.48/drivers/char/hvc_vio.c linux-2.6.32.48-openvz/drivers/char/hvc_vio.c --- linux-2.6.32.48/drivers/char/hvc_vio.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/char/hvc_vio.c 2011-11-17 20:08:53.000000000 -0500 @@ -77,7 +77,7 @@ static int filtered_get_chars(uint32_t v return got; } -static struct hv_ops hvc_get_put_ops = { +static const struct hv_ops hvc_get_put_ops = { .get_chars = filtered_get_chars, .put_chars = hvc_put_chars, .notifier_add = notifier_add_irq, diff -urNp linux-2.6.32.48/drivers/char/hvc_xen.c linux-2.6.32.48-openvz/drivers/char/hvc_xen.c --- linux-2.6.32.48/drivers/char/hvc_xen.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/char/hvc_xen.c 2011-11-17 20:08:53.000000000 -0500 @@ -120,7 +120,7 @@ static int read_console(uint32_t vtermno return recv; } -static struct hv_ops hvc_ops = { +static const struct hv_ops hvc_ops = { .get_chars = read_console, .put_chars = write_console, .notifier_add = notifier_add_irq, diff -urNp linux-2.6.32.48/drivers/char/ipmi/ipmi_msghandler.c linux-2.6.32.48-openvz/drivers/char/ipmi/ipmi_msghandler.c --- linux-2.6.32.48/drivers/char/ipmi/ipmi_msghandler.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/char/ipmi/ipmi_msghandler.c 2011-11-17 20:08:53.000000000 -0500 @@ -414,7 +414,7 @@ struct ipmi_smi { struct proc_dir_entry *proc_dir; char proc_dir_name[10]; - atomic_t stats[IPMI_NUM_STATS]; + atomic_unchecked_t stats[IPMI_NUM_STATS]; /* * run_to_completion duplicate of smb_info, smi_info @@ -447,9 +447,9 @@ static DEFINE_MUTEX(smi_watchers_mutex); #define ipmi_inc_stat(intf, stat) \ - atomic_inc(&(intf)->stats[IPMI_STAT_ ## stat]) + atomic_inc_unchecked(&(intf)->stats[IPMI_STAT_ ## stat]) #define ipmi_get_stat(intf, stat) \ - ((unsigned int) atomic_read(&(intf)->stats[IPMI_STAT_ ## stat])) + ((unsigned int) atomic_read_unchecked(&(intf)->stats[IPMI_STAT_ ## stat])) static int is_lan_addr(struct ipmi_addr *addr) { @@ -2808,7 +2808,7 @@ int ipmi_register_smi(struct ipmi_smi_ha INIT_LIST_HEAD(&intf->cmd_rcvrs); init_waitqueue_head(&intf->waitq); for (i = 0; i < IPMI_NUM_STATS; i++) - atomic_set(&intf->stats[i], 0); + atomic_set_unchecked(&intf->stats[i], 0); intf->proc_dir = NULL; @@ -4160,6 +4160,8 @@ static void send_panic_events(char *str) struct ipmi_smi_msg smi_msg; struct ipmi_recv_msg recv_msg; + pax_track_stack(); + si = (struct ipmi_system_interface_addr *) &addr; si->addr_type = IPMI_SYSTEM_INTERFACE_ADDR_TYPE; si->channel = IPMI_BMC_CHANNEL; diff -urNp linux-2.6.32.48/drivers/char/ipmi/ipmi_si_intf.c linux-2.6.32.48-openvz/drivers/char/ipmi/ipmi_si_intf.c --- linux-2.6.32.48/drivers/char/ipmi/ipmi_si_intf.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/char/ipmi/ipmi_si_intf.c 2011-11-17 20:08:53.000000000 -0500 @@ -277,7 +277,7 @@ struct smi_info { unsigned char slave_addr; /* Counters and things for the proc filesystem. */ - atomic_t stats[SI_NUM_STATS]; + atomic_unchecked_t stats[SI_NUM_STATS]; struct task_struct *thread; @@ -285,9 +285,9 @@ struct smi_info { }; #define smi_inc_stat(smi, stat) \ - atomic_inc(&(smi)->stats[SI_STAT_ ## stat]) + atomic_inc_unchecked(&(smi)->stats[SI_STAT_ ## stat]) #define smi_get_stat(smi, stat) \ - ((unsigned int) atomic_read(&(smi)->stats[SI_STAT_ ## stat])) + ((unsigned int) atomic_read_unchecked(&(smi)->stats[SI_STAT_ ## stat])) #define SI_MAX_PARMS 4 @@ -2931,7 +2931,7 @@ static int try_smi_init(struct smi_info atomic_set(&new_smi->req_events, 0); new_smi->run_to_completion = 0; for (i = 0; i < SI_NUM_STATS; i++) - atomic_set(&new_smi->stats[i], 0); + atomic_set_unchecked(&new_smi->stats[i], 0); new_smi->interrupt_disabled = 0; atomic_set(&new_smi->stop_operation, 0); diff -urNp linux-2.6.32.48/drivers/char/istallion.c linux-2.6.32.48-openvz/drivers/char/istallion.c --- linux-2.6.32.48/drivers/char/istallion.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/char/istallion.c 2011-11-17 20:08:53.000000000 -0500 @@ -187,7 +187,6 @@ static struct ktermios stli_deftermios * re-used for each stats call. */ static comstats_t stli_comstats; -static combrd_t stli_brdstats; static struct asystats stli_cdkstats; /*****************************************************************************/ @@ -4058,6 +4057,7 @@ static int stli_getbrdstats(combrd_t __u { struct stlibrd *brdp; unsigned int i; + combrd_t stli_brdstats; if (copy_from_user(&stli_brdstats, bp, sizeof(combrd_t))) return -EFAULT; @@ -4269,6 +4269,8 @@ static int stli_getportstruct(struct stl struct stliport stli_dummyport; struct stliport *portp; + pax_track_stack(); + if (copy_from_user(&stli_dummyport, arg, sizeof(struct stliport))) return -EFAULT; portp = stli_getport(stli_dummyport.brdnr, stli_dummyport.panelnr, @@ -4291,6 +4293,8 @@ static int stli_getbrdstruct(struct stli struct stlibrd stli_dummybrd; struct stlibrd *brdp; + pax_track_stack(); + if (copy_from_user(&stli_dummybrd, arg, sizeof(struct stlibrd))) return -EFAULT; if (stli_dummybrd.brdnr >= STL_MAXBRDS) diff -urNp linux-2.6.32.48/drivers/char/Kconfig linux-2.6.32.48-openvz/drivers/char/Kconfig --- linux-2.6.32.48/drivers/char/Kconfig 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/char/Kconfig 2011-11-17 20:08:53.000000000 -0500 @@ -90,7 +90,8 @@ config VT_HW_CONSOLE_BINDING config DEVKMEM bool "/dev/kmem virtual device support" - default y + default n + depends on !GRKERNSEC_KMEM help Say Y here if you want to support the /dev/kmem device. The /dev/kmem device is rarely used, but can be used for certain @@ -458,7 +459,7 @@ config UNIX98_PTYS config DEVPTS_MULTIPLE_INSTANCES bool "Support multiple instances of devpts" - depends on UNIX98_PTYS + depends on UNIX98_PTYS && !VE default n ---help--- Enable support for multiple instances of devpts filesystem. @@ -1114,6 +1115,7 @@ config DEVPORT bool depends on !M68K depends on ISA || PCI + depends on !GRKERNSEC_KMEM default y source "drivers/s390/char/Kconfig" diff -urNp linux-2.6.32.48/drivers/char/keyboard.c linux-2.6.32.48-openvz/drivers/char/keyboard.c --- linux-2.6.32.48/drivers/char/keyboard.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/char/keyboard.c 2011-11-17 20:08:53.000000000 -0500 @@ -162,6 +162,7 @@ unsigned char kbd_sysrq_xlate[KEY_MAX + static int sysrq_down; static int sysrq_alt_use; #endif +int sysrq_key_scancode = KEY_SYSRQ; static int sysrq_alt; /* @@ -635,6 +636,16 @@ static void k_spec(struct vc_data *vc, u kbd->kbdmode == VC_MEDIUMRAW) && value != KVAL(K_SAK)) return; /* SAK is allowed even in raw mode */ + +#if defined(CONFIG_GRKERNSEC_PROC) || defined(CONFIG_GRKERNSEC_PROC_MEMMAP) + { + void *func = fn_handler[value]; + if (func == fn_show_state || func == fn_show_ptregs || + func == fn_show_mem) + return; + } +#endif + fn_handler[value](vc); } @@ -1067,6 +1078,9 @@ static int emulate_raw(struct vc_data *v { int code; + if (keycode == sysrq_key_scancode && sysrq_alt) + goto sysrq; + switch (keycode) { case KEY_PAUSE: put_queue(vc, 0xe1); @@ -1085,6 +1099,7 @@ static int emulate_raw(struct vc_data *v break; case KEY_SYSRQ: +sysrq: /* * Real AT keyboards (that's what we're trying * to emulate here emit 0xe0 0x2a 0xe0 0x37 when @@ -1179,7 +1194,8 @@ static void kbd_keycode(unsigned int key printk(KERN_WARNING "keyboard.c: can't emulate rawmode for keycode %d\n", keycode); #ifdef CONFIG_MAGIC_SYSRQ /* Handle the SysRq Hack */ - if (keycode == KEY_SYSRQ && (sysrq_down || (down == 1 && sysrq_alt))) { + if ((keycode == sysrq_key_scancode || keycode == KEY_SYSRQ) && + (sysrq_down || (down == 1 && sysrq_alt))) { if (!sysrq_down) { sysrq_down = down; sysrq_alt_use = sysrq_alt; @@ -1386,7 +1402,7 @@ static const struct input_device_id kbd_ .evbit = { BIT_MASK(EV_SND) }, }, - { }, /* Terminating entry */ + { 0 }, /* Terminating entry */ }; MODULE_DEVICE_TABLE(input, kbd_ids); diff -urNp linux-2.6.32.48/drivers/char/mem.c linux-2.6.32.48-openvz/drivers/char/mem.c --- linux-2.6.32.48/drivers/char/mem.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/char/mem.c 2011-11-17 20:08:53.000000000 -0500 @@ -18,6 +18,7 @@ #include #include #include +#include #include #include #include @@ -35,6 +36,10 @@ # include #endif +#if defined(CONFIG_GRKERNSEC) && !defined(CONFIG_GRKERNSEC_NO_RBAC) +extern struct file_operations grsec_fops; +#endif + static inline unsigned long size_inside_page(unsigned long start, unsigned long size) { @@ -102,9 +107,13 @@ static inline int range_is_allowed(unsig while (cursor < to) { if (!devmem_is_allowed(pfn)) { +#ifdef CONFIG_GRKERNSEC_KMEM + gr_handle_mem_readwrite(from, to); +#else printk(KERN_INFO "Program %s tried to access /dev/mem between %Lx->%Lx.\n", current->comm, from, to); +#endif return 0; } cursor += PAGE_SIZE; @@ -112,6 +121,11 @@ static inline int range_is_allowed(unsig } return 1; } +#elif defined(CONFIG_GRKERNSEC_KMEM) +static inline int range_is_allowed(unsigned long pfn, unsigned long size) +{ + return 0; +} #else static inline int range_is_allowed(unsigned long pfn, unsigned long size) { @@ -155,6 +169,8 @@ static ssize_t read_mem(struct file * fi #endif while (count > 0) { + char *temp; + /* * Handle first page in case it's not aligned */ @@ -177,11 +193,31 @@ static ssize_t read_mem(struct file * fi if (!ptr) return -EFAULT; - if (copy_to_user(buf, ptr, sz)) { +#ifdef CONFIG_PAX_USERCOPY + temp = kmalloc(sz, GFP_KERNEL); + if (!temp) { + unxlate_dev_mem_ptr(p, ptr); + return -ENOMEM; + } + memcpy(temp, ptr, sz); +#else + temp = ptr; +#endif + + if (copy_to_user(buf, temp, sz)) { + +#ifdef CONFIG_PAX_USERCOPY + kfree(temp); +#endif + unxlate_dev_mem_ptr(p, ptr); return -EFAULT; } +#ifdef CONFIG_PAX_USERCOPY + kfree(temp); +#endif + unxlate_dev_mem_ptr(p, ptr); buf += sz; @@ -419,9 +455,8 @@ static ssize_t read_kmem(struct file *fi size_t count, loff_t *ppos) { unsigned long p = *ppos; - ssize_t low_count, read, sz; + ssize_t low_count, read, sz, err = 0; char * kbuf; /* k-addr because vread() takes vmlist_lock rwlock */ - int err = 0; read = 0; if (p < (unsigned long) high_memory) { @@ -444,6 +479,8 @@ static ssize_t read_kmem(struct file *fi } #endif while (low_count > 0) { + char *temp; + sz = size_inside_page(p, low_count); /* @@ -453,7 +490,22 @@ static ssize_t read_kmem(struct file *fi */ kbuf = xlate_dev_kmem_ptr((char *)p); - if (copy_to_user(buf, kbuf, sz)) +#ifdef CONFIG_PAX_USERCOPY + temp = kmalloc(sz, GFP_KERNEL); + if (!temp) + return -ENOMEM; + memcpy(temp, kbuf, sz); +#else + temp = kbuf; +#endif + + err = copy_to_user(buf, temp, sz); + +#ifdef CONFIG_PAX_USERCOPY + kfree(temp); +#endif + + if (err) return -EFAULT; buf += sz; p += sz; @@ -889,6 +941,9 @@ static const struct memdev { #ifdef CONFIG_CRASH_DUMP [12] = { "oldmem", 0, &oldmem_fops, NULL }, #endif +#if defined(CONFIG_GRKERNSEC) && !defined(CONFIG_GRKERNSEC_NO_RBAC) + [13] = { "grsec",S_IRUSR | S_IWUGO, &grsec_fops, NULL }, +#endif }; static int memory_open(struct inode *inode, struct file *filp) diff -urNp linux-2.6.32.48/drivers/char/pcmcia/ipwireless/tty.c linux-2.6.32.48-openvz/drivers/char/pcmcia/ipwireless/tty.c --- linux-2.6.32.48/drivers/char/pcmcia/ipwireless/tty.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/char/pcmcia/ipwireless/tty.c 2011-11-17 20:08:53.000000000 -0500 @@ -29,6 +29,7 @@ #include #include #include +#include #include "tty.h" #include "network.h" @@ -51,7 +52,7 @@ struct ipw_tty { int tty_type; struct ipw_network *network; struct tty_struct *linux_tty; - int open_count; + local_t open_count; unsigned int control_lines; struct mutex ipw_tty_mutex; int tx_bytes_queued; @@ -127,10 +128,10 @@ static int ipw_open(struct tty_struct *l mutex_unlock(&tty->ipw_tty_mutex); return -ENODEV; } - if (tty->open_count == 0) + if (local_read(&tty->open_count) == 0) tty->tx_bytes_queued = 0; - tty->open_count++; + local_inc(&tty->open_count); tty->linux_tty = linux_tty; linux_tty->driver_data = tty; @@ -146,9 +147,7 @@ static int ipw_open(struct tty_struct *l static void do_ipw_close(struct ipw_tty *tty) { - tty->open_count--; - - if (tty->open_count == 0) { + if (local_dec_return(&tty->open_count) == 0) { struct tty_struct *linux_tty = tty->linux_tty; if (linux_tty != NULL) { @@ -169,7 +168,7 @@ static void ipw_hangup(struct tty_struct return; mutex_lock(&tty->ipw_tty_mutex); - if (tty->open_count == 0) { + if (local_read(&tty->open_count) == 0) { mutex_unlock(&tty->ipw_tty_mutex); return; } @@ -198,7 +197,7 @@ void ipwireless_tty_received(struct ipw_ return; } - if (!tty->open_count) { + if (!local_read(&tty->open_count)) { mutex_unlock(&tty->ipw_tty_mutex); return; } @@ -240,7 +239,7 @@ static int ipw_write(struct tty_struct * return -ENODEV; mutex_lock(&tty->ipw_tty_mutex); - if (!tty->open_count) { + if (!local_read(&tty->open_count)) { mutex_unlock(&tty->ipw_tty_mutex); return -EINVAL; } @@ -280,7 +279,7 @@ static int ipw_write_room(struct tty_str if (!tty) return -ENODEV; - if (!tty->open_count) + if (!local_read(&tty->open_count)) return -EINVAL; room = IPWIRELESS_TX_QUEUE_SIZE - tty->tx_bytes_queued; @@ -322,7 +321,7 @@ static int ipw_chars_in_buffer(struct tt if (!tty) return 0; - if (!tty->open_count) + if (!local_read(&tty->open_count)) return 0; return tty->tx_bytes_queued; @@ -403,7 +402,7 @@ static int ipw_tiocmget(struct tty_struc if (!tty) return -ENODEV; - if (!tty->open_count) + if (!local_read(&tty->open_count)) return -EINVAL; return get_control_lines(tty); @@ -419,7 +418,7 @@ ipw_tiocmset(struct tty_struct *linux_tt if (!tty) return -ENODEV; - if (!tty->open_count) + if (!local_read(&tty->open_count)) return -EINVAL; return set_control_lines(tty, set, clear); @@ -433,7 +432,7 @@ static int ipw_ioctl(struct tty_struct * if (!tty) return -ENODEV; - if (!tty->open_count) + if (!local_read(&tty->open_count)) return -EINVAL; /* FIXME: Exactly how is the tty object locked here .. */ @@ -591,7 +590,7 @@ void ipwireless_tty_free(struct ipw_tty against a parallel ioctl etc */ mutex_lock(&ttyj->ipw_tty_mutex); } - while (ttyj->open_count) + while (local_read(&ttyj->open_count)) do_ipw_close(ttyj); ipwireless_disassociate_network_ttys(network, ttyj->channel_idx); diff -urNp linux-2.6.32.48/drivers/char/pty.c linux-2.6.32.48-openvz/drivers/char/pty.c --- linux-2.6.32.48/drivers/char/pty.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/char/pty.c 2011-11-17 20:09:04.000000000 -0500 @@ -30,16 +30,22 @@ #include #include +#include + #include #ifdef CONFIG_UNIX98_PTYS -static struct tty_driver *ptm_driver; -static struct tty_driver *pts_driver; +struct tty_driver *ptm_driver; +struct tty_driver *pts_driver; +EXPORT_SYMBOL(ptm_driver); +EXPORT_SYMBOL(pts_driver); #endif static void pty_close(struct tty_struct *tty, struct file *filp) { BUG_ON(!tty); + + ub_pty_uncharge(tty); if (tty->driver->subtype == PTY_TYPE_MASTER) WARN_ON(tty->count > 1); else { @@ -58,8 +64,12 @@ static void pty_close(struct tty_struct if (tty->driver->subtype == PTY_TYPE_MASTER) { set_bit(TTY_OTHER_CLOSED, &tty->flags); #ifdef CONFIG_UNIX98_PTYS - if (tty->driver == ptm_driver) + if (tty->driver->flags & TTY_DRIVER_DEVPTS_MEM) { + struct ve_struct *old_env; + old_env = set_exec_env(tty->owner_env); devpts_pty_kill(tty->link); + (void)set_exec_env(old_env); + } #endif tty_vhangup(tty->link); } @@ -201,6 +211,10 @@ static int pty_open(struct tty_struct *t if (tty->link->count != 1) goto out; + retval = -ENOMEM; + if (ub_pty_charge(tty)) + goto out; + clear_bit(TTY_OTHER_CLOSED, &tty->link->flags); set_bit(TTY_THROTTLED, &tty->flags); retval = 0; @@ -358,9 +372,12 @@ static const struct tty_operations slave .resize = pty_resize }; +struct tty_driver *pty_driver, *pty_slave_driver; +EXPORT_SYMBOL(pty_driver); +EXPORT_SYMBOL(pty_slave_driver); + static void __init legacy_pty_init(void) { - struct tty_driver *pty_driver, *pty_slave_driver; if (legacy_count <= 0) return; @@ -645,7 +662,7 @@ static int __ptmx_open(struct inode *ino return index; mutex_lock(&tty_mutex); - tty = tty_init_dev(ptm_driver, index, 1); + tty = tty_init_dev(get_exec_env()->ptm_driver, index, NULL, 1); mutex_unlock(&tty_mutex); if (IS_ERR(tty)) { @@ -661,7 +678,7 @@ static int __ptmx_open(struct inode *ino if (retval) goto out1; - retval = ptm_driver->ops->open(tty, filp); + retval = get_exec_env()->ptm_driver->ops->open(tty, filp); if (!retval) return 0; out1: @@ -736,14 +753,19 @@ static void __init unix98_pty_init(void) register_sysctl_table(pty_root_table); /* Now create the /dev/ptmx special device */ + pax_open_kernel(); tty_default_fops(&ptmx_fops); - ptmx_fops.open = ptmx_open; + *(void **)&ptmx_fops.open = ptmx_open; + pax_close_kernel(); cdev_init(&ptmx_cdev, &ptmx_fops); if (cdev_add(&ptmx_cdev, MKDEV(TTYAUX_MAJOR, 2), 1) || register_chrdev_region(MKDEV(TTYAUX_MAJOR, 2), 1, "/dev/ptmx") < 0) panic("Couldn't register /dev/ptmx driver\n"); device_create(tty_class, NULL, MKDEV(TTYAUX_MAJOR, 2), NULL, "ptmx"); +#ifdef CONFIG_VE + get_ve0()->ptm_driver = ptm_driver; +#endif } #else diff -urNp linux-2.6.32.48/drivers/char/random.c linux-2.6.32.48-openvz/drivers/char/random.c --- linux-2.6.32.48/drivers/char/random.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/char/random.c 2011-11-17 20:09:04.000000000 -0500 @@ -254,8 +254,13 @@ /* * Configuration information */ +#ifdef CONFIG_GRKERNSEC_RANDNET +#define INPUT_POOL_WORDS 512 +#define OUTPUT_POOL_WORDS 128 +#else #define INPUT_POOL_WORDS 128 #define OUTPUT_POOL_WORDS 32 +#endif #define SEC_XFER_SIZE 512 /* @@ -292,10 +297,17 @@ static struct poolinfo { int poolwords; int tap1, tap2, tap3, tap4, tap5; } poolinfo_table[] = { +#ifdef CONFIG_GRKERNSEC_RANDNET + /* x^512 + x^411 + x^308 + x^208 +x^104 + x + 1 -- 225 */ + { 512, 411, 308, 208, 104, 1 }, + /* x^128 + x^103 + x^76 + x^51 + x^25 + x + 1 -- 105 */ + { 128, 103, 76, 51, 25, 1 }, +#else /* x^128 + x^103 + x^76 + x^51 +x^25 + x + 1 -- 105 */ { 128, 103, 76, 51, 25, 1 }, /* x^32 + x^26 + x^20 + x^14 + x^7 + x + 1 -- 15 */ { 32, 26, 20, 14, 7, 1 }, +#endif #if 0 /* x^2048 + x^1638 + x^1231 + x^819 + x^411 + x + 1 -- 115 */ { 2048, 1638, 1231, 819, 411, 1 }, @@ -1209,7 +1221,7 @@ EXPORT_SYMBOL(generate_random_uuid); #include static int min_read_thresh = 8, min_write_thresh; -static int max_read_thresh = INPUT_POOL_WORDS * 32; +static int max_read_thresh = OUTPUT_POOL_WORDS * 32; static int max_write_thresh = INPUT_POOL_WORDS * 32; static char sysctl_bootid[16]; diff -urNp linux-2.6.32.48/drivers/char/rocket.c linux-2.6.32.48-openvz/drivers/char/rocket.c --- linux-2.6.32.48/drivers/char/rocket.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/char/rocket.c 2011-11-17 20:08:53.000000000 -0500 @@ -1266,6 +1266,8 @@ static int get_ports(struct r_port *info struct rocket_ports tmp; int board; + pax_track_stack(); + if (!retports) return -EFAULT; memset(&tmp, 0, sizeof (tmp)); diff -urNp linux-2.6.32.48/drivers/char/sonypi.c linux-2.6.32.48-openvz/drivers/char/sonypi.c --- linux-2.6.32.48/drivers/char/sonypi.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/char/sonypi.c 2011-11-17 20:08:53.000000000 -0500 @@ -55,6 +55,7 @@ #include #include #include +#include #include @@ -491,7 +492,7 @@ static struct sonypi_device { spinlock_t fifo_lock; wait_queue_head_t fifo_proc_list; struct fasync_struct *fifo_async; - int open_count; + local_t open_count; int model; struct input_dev *input_jog_dev; struct input_dev *input_key_dev; @@ -895,7 +896,7 @@ static int sonypi_misc_fasync(int fd, st static int sonypi_misc_release(struct inode *inode, struct file *file) { mutex_lock(&sonypi_device.lock); - sonypi_device.open_count--; + local_dec(&sonypi_device.open_count); mutex_unlock(&sonypi_device.lock); return 0; } @@ -905,9 +906,9 @@ static int sonypi_misc_open(struct inode lock_kernel(); mutex_lock(&sonypi_device.lock); /* Flush input queue on first open */ - if (!sonypi_device.open_count) + if (!local_read(&sonypi_device.open_count)) kfifo_reset(sonypi_device.fifo); - sonypi_device.open_count++; + local_inc(&sonypi_device.open_count); mutex_unlock(&sonypi_device.lock); unlock_kernel(); return 0; diff -urNp linux-2.6.32.48/drivers/char/stallion.c linux-2.6.32.48-openvz/drivers/char/stallion.c --- linux-2.6.32.48/drivers/char/stallion.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/char/stallion.c 2011-11-17 20:08:53.000000000 -0500 @@ -2448,6 +2448,8 @@ static int stl_getportstruct(struct stlp struct stlport stl_dummyport; struct stlport *portp; + pax_track_stack(); + if (copy_from_user(&stl_dummyport, arg, sizeof(struct stlport))) return -EFAULT; portp = stl_getport(stl_dummyport.brdnr, stl_dummyport.panelnr, diff -urNp linux-2.6.32.48/drivers/char/sysrq.c linux-2.6.32.48-openvz/drivers/char/sysrq.c --- linux-2.6.32.48/drivers/char/sysrq.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/char/sysrq.c 2011-11-17 20:08:53.000000000 -0500 @@ -37,7 +37,10 @@ #include #include #include +#include +#include #include +#include #include #include @@ -250,8 +253,8 @@ static struct sysrq_key_op sysrq_showall static void sysrq_handle_showregs(int key, struct tty_struct *tty) { struct pt_regs *regs = get_irq_regs(); - if (regs) - show_regs(regs); + + nmi_show_regs(regs, 0); perf_event_print_debug(); } static struct sysrq_key_op sysrq_showregs_op = { @@ -303,6 +306,7 @@ static struct sysrq_key_op sysrq_ftrace_ static void sysrq_handle_showmem(int key, struct tty_struct *tty) { show_mem(); + show_slab_info(); } static struct sysrq_key_op sysrq_showmem_op = { .handler = sysrq_handle_showmem, @@ -318,7 +322,7 @@ static void send_sig_all(int sig) { struct task_struct *p; - for_each_process(p) { + for_each_process_all(p) { if (p->mm && !is_global_init(p)) /* Not swapper, init nor kernel thread */ force_sig(sig, p); @@ -394,7 +398,267 @@ static struct sysrq_key_op sysrq_unrt_op /* Key Operations table and lock */ static DEFINE_SPINLOCK(sysrq_key_table_lock); -static struct sysrq_key_op *sysrq_key_table[36] = { +#define SYSRQ_KEY_TABLE_LENGTH 37 +static struct sysrq_key_op **sysrq_key_table; +static struct sysrq_key_op *sysrq_default_key_table[]; + +#ifdef CONFIG_SYSRQ_DEBUG +#define SYSRQ_NAMELEN_MAX 64 +#define SYSRQ_DUMP_LINES 32 + +static struct sysrq_key_op *sysrq_debug_key_table[]; +static struct sysrq_key_op *sysrq_input_key_table[]; +static unsigned long *dump_address; +static int orig_console_loglevel; +static void (*sysrq_input_return)(char *) = NULL; + +static void dump_mem(void) +{ + unsigned long value[4]; + mm_segment_t old_fs; + int line, err; + + old_fs = get_fs(); + set_fs(KERNEL_DS); + err = 0; + + for (line = 0; line < SYSRQ_DUMP_LINES; line++) { + err |= __get_user(value[0], dump_address++); + err |= __get_user(value[1], dump_address++); + err |= __get_user(value[2], dump_address++); + err |= __get_user(value[3], dump_address++); + if (err) { + printk("Invalid address %p\n", dump_address - 4); + break; + } +#if BITS_PER_LONG == 32 + printk("0x%p: %08lx %08lx %08lx %08lx\n", + dump_address - 4, + value[0], value[1], value[2], value[3]); +#else + printk("0x%p: %016lx %016lx %016lx %016lx\n", + dump_address - 4, + value[0], value[1], value[2], value[3]); +#endif + } + set_fs(old_fs); +} + +static void write_mem(unsigned long val) +{ + mm_segment_t old_fs; + unsigned long old_val; + + old_fs = get_fs(); + set_fs(KERNEL_DS); + if (__get_user(old_val, dump_address)) { + printk("Invalid address %p\n", dump_address); + goto out; + } + +#if BITS_PER_LONG == 32 + printk("Changing [%p] from %08lx to %08lx\n", + dump_address, old_val, val); +#else + printk("Changing [%p] from %016lx to %016lx\n", + dump_address, old_val, val); +#endif + __put_user(val, dump_address); +out: + set_fs(old_fs); +} + +static void handle_read(int key, struct tty_struct *tty) +{ + static int pos; + static int upper_case; + static char str[SYSRQ_NAMELEN_MAX]; + + if (key == 0) { + /* actually 0 is not shift only... */ + upper_case = 1; + return; + } + + if (key == 0x0d || pos == SYSRQ_NAMELEN_MAX - 1) { + /* enter */ + sysrq_key_table = sysrq_debug_key_table; + str[pos] = '\0'; + pos = upper_case = 0; + printk("\n"); + if (sysrq_input_return == NULL) + printk("No return handler!!!\n"); + else + sysrq_input_return(str); + return; + }; + + /* check for alowed symbols */ + if (key == '-') { + if (upper_case) + key = '_'; + goto correct; + }; + if (key >= 'a' && key <= 'z') { + if (upper_case) + key = key - 'a' + 'A'; + goto correct; + }; + if (key >= '0' && key <= '9') + goto correct; + + upper_case = 0; + return; + +correct: + str[pos] = key; + printk("%c", (char)key); + pos++; + upper_case = 0; +} + +static struct sysrq_key_op input_read = { + .handler = handle_read, + .help_msg = "", + .action_msg = NULL, +}; + +static struct sysrq_key_op *sysrq_input_key_table[SYSRQ_KEY_TABLE_LENGTH] = { + [0 ... SYSRQ_KEY_TABLE_LENGTH - 1] = &input_read, +}; + +static void return_dump_mem(char *str) +{ + unsigned long address; + char *end; + + address = simple_strtoul(str, &end, 0); + if (*end != '\0') { + printk("Bad address [%s]\n", str); + return; + } + + dump_address = (unsigned long *)address; + dump_mem(); +} + +static void handle_dump_mem(int key, struct tty_struct *tty) +{ + sysrq_input_return = return_dump_mem; + sysrq_key_table = sysrq_input_key_table; +} + +static struct sysrq_key_op debug_dump_mem = { + .handler = handle_dump_mem, + .help_msg = "Dump", + .action_msg = "Enter address:", +}; + +static void return_resolve(char *str) +{ + unsigned long address; + + address = kallsyms_lookup_name(str); + printk("%s : %lx\n", str, address); + if (address) { + dump_address = (unsigned long *)address; + printk("Now you can dump it via X\n"); + } +} + +static void handle_resolve(int key, struct tty_struct *tty) +{ + sysrq_input_return = return_resolve; + sysrq_key_table = sysrq_input_key_table; +} + +static struct sysrq_key_op debug_resolve = { + .handler = handle_resolve, + .help_msg = "Resolve", + .action_msg = "Enter symbol name:", +}; + +static void return_write_mem(char *str) +{ + unsigned long address; + unsigned long value; + char *end; + + address = simple_strtoul(str, &end, 0); + if (*end != '-') { + printk("Bad address in %s\n", str); + return; + } + value = simple_strtoul(end + 1, &end, 0); + if (*end != '\0') { + printk("Bad value in %s\n", str); + return; + } + + dump_address = (unsigned long *)address; + write_mem(value); +} + +static void handle_write_mem(int key, struct tty_struct *tty) +{ + sysrq_input_return = return_write_mem; + sysrq_key_table = sysrq_input_key_table; +} + +static struct sysrq_key_op debug_write_mem = { + .handler = handle_write_mem, + .help_msg = "Writemem", + .action_msg = "Enter address-value:", +}; + +static void handle_next(int key, struct tty_struct *tty) +{ + dump_mem(); +} + +static struct sysrq_key_op debug_next = { + .handler = handle_next, + .help_msg = "neXt", + .action_msg = "continuing", +}; + +static void handle_quit(int key, struct tty_struct *tty) +{ + sysrq_key_table = sysrq_default_key_table; + console_loglevel = orig_console_loglevel; +} + +static struct sysrq_key_op debug_quit = { + .handler = handle_quit, + .help_msg = "Quit", + .action_msg = "Thank you for using debugger", +}; + +static struct sysrq_key_op *sysrq_debug_key_table[SYSRQ_KEY_TABLE_LENGTH] = { + [13] = &debug_dump_mem, /* d */ + [26] = &debug_quit, /* q */ + [27] = &debug_resolve, /* r */ + [32] = &debug_write_mem, /* w */ + [33] = &debug_next, /* x */ +}; + +static void sysrq_handle_debug(int key, struct tty_struct *tty) +{ + orig_console_loglevel = console_loglevel; + console_loglevel = 8; + sysrq_key_table = sysrq_debug_key_table; + printk("Welcome sysrq debugging mode\n" + "Press H for help\n"); +} + +static struct sysrq_key_op sysrq_debug_op = { + .handler = sysrq_handle_debug, + .help_msg = "debuG", + .action_msg = "Select desired action", +}; +#endif + +static struct sysrq_key_op *sysrq_default_key_table[SYSRQ_KEY_TABLE_LENGTH] = { &sysrq_loglevel_op, /* 0 */ &sysrq_loglevel_op, /* 1 */ &sysrq_loglevel_op, /* 2 */ @@ -417,7 +681,11 @@ static struct sysrq_key_op *sysrq_key_ta &sysrq_term_op, /* e */ &sysrq_moom_op, /* f */ /* g: May be registered for the kernel debugger */ +#ifdef CONFIG_SYSRQ_DEBUG + &sysrq_debug_op, /* g */ +#else NULL, /* g */ +#endif NULL, /* h - reserved for help */ &sysrq_kill_op, /* i */ #ifdef CONFIG_BLOCK @@ -449,8 +717,11 @@ static struct sysrq_key_op *sysrq_key_ta /* y: May be registered on sparc64 for global register dump */ NULL, /* y */ &sysrq_ftrace_dump_op, /* z */ + NULL, /* for debugger */ }; +static struct sysrq_key_op **sysrq_key_table = sysrq_default_key_table; + /* key2index calculation, -1 on invalid index */ static int sysrq_key_table_key2index(int key) { @@ -460,6 +731,10 @@ static int sysrq_key_table_key2index(int retval = key - '0'; else if ((key >= 'a') && (key <= 'z')) retval = key + 10 - 'a'; +#ifdef CONFIG_SYSRQ_DEBUG + else if (key == 0 || key == 0x0d || key == '-') + retval = SYSRQ_KEY_TABLE_LENGTH - 1; +#endif else retval = -1; return retval; @@ -470,21 +745,21 @@ static int sysrq_key_table_key2index(int */ struct sysrq_key_op *__sysrq_get_key_op(int key) { - struct sysrq_key_op *op_p = NULL; - int i; + struct sysrq_key_op *op_p = NULL; + int i; i = sysrq_key_table_key2index(key); if (i != -1) - op_p = sysrq_key_table[i]; - return op_p; + op_p = sysrq_key_table[i]; + return op_p; } static void __sysrq_put_key_op(int key, struct sysrq_key_op *op_p) { - int i = sysrq_key_table_key2index(key); + int i = sysrq_key_table_key2index(key); - if (i != -1) - sysrq_key_table[i] = op_p; + if (i != -1) + sysrq_key_table[i] = op_p; } /* @@ -507,25 +782,25 @@ void __handle_sysrq(int key, struct tty_ */ orig_log_level = console_loglevel; console_loglevel = 7; - printk(KERN_INFO "SysRq : "); - op_p = __sysrq_get_key_op(key); - if (op_p) { + op_p = __sysrq_get_key_op(key); + if (op_p) { /* * Should we check for enabled operations (/proc/sysrq-trigger * should not) and is the invoked operation enabled? */ if (!check_mask || sysrq_on_mask(op_p->enable_mask)) { - printk("%s\n", op_p->action_msg); + if (op_p->action_msg) + printk("%s\n", op_p->action_msg); console_loglevel = orig_log_level; op_p->handler(key, tty); } else { printk("This sysrq operation is disabled.\n"); } } else { - printk("HELP : "); + printk("SysRq HELP : "); /* Only print the help msg once per handler */ - for (i = 0; i < ARRAY_SIZE(sysrq_key_table); i++) { + for (i = 0; i < SYSRQ_KEY_TABLE_LENGTH; i++) { if (sysrq_key_table[i]) { int j; @@ -555,7 +830,7 @@ void handle_sysrq(int key, struct tty_st EXPORT_SYMBOL(handle_sysrq); static int __sysrq_swap_key_ops(int key, struct sysrq_key_op *insert_op_p, - struct sysrq_key_op *remove_op_p) + struct sysrq_key_op *remove_op_p) { int retval; @@ -591,12 +866,29 @@ EXPORT_SYMBOL(unregister_sysrq_key); static ssize_t write_sysrq_trigger(struct file *file, const char __user *buf, size_t count, loff_t *ppos) { + struct ve_struct *cur = get_exec_env(); + static int pnum = 10; + if (count) { - char c; + int i, cnt; + char c[32]; - if (get_user(c, buf)) + cnt = min(count, sizeof(c)); + if (copy_from_user(c, buf, cnt)) return -EFAULT; - __handle_sysrq(c, NULL, 0); + + + for (i = 0; i < cnt && c[i] != '\n'; i++) { + if (!ve_is_super(cur)) { + if (!pnum) + continue; + printk("SysRq: CT#%u sent '%c' magic key.\n", + cur->veid, c[i]); + pnum--; + continue; + } + __handle_sysrq(c[i], NULL, 0); + } } return count; } @@ -607,7 +899,7 @@ static const struct file_operations proc static int __init sysrq_init(void) { - proc_create("sysrq-trigger", S_IWUSR, NULL, &proc_sysrq_trigger_operations); + proc_create("sysrq-trigger", S_IWUSR, &glob_proc_root, &proc_sysrq_trigger_operations); return 0; } module_init(sysrq_init); diff -urNp linux-2.6.32.48/drivers/char/tpm/tpm_bios.c linux-2.6.32.48-openvz/drivers/char/tpm/tpm_bios.c --- linux-2.6.32.48/drivers/char/tpm/tpm_bios.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/char/tpm/tpm_bios.c 2011-11-17 20:09:04.000000000 -0500 @@ -172,7 +172,7 @@ static void *tpm_bios_measurements_start event = addr; if ((event->event_type == 0 && event->event_size == 0) || - ((addr + sizeof(struct tcpa_event) + event->event_size) >= limit)) + (event->event_size >= limit - addr - sizeof(struct tcpa_event))) return NULL; return addr; @@ -197,7 +197,7 @@ static void *tpm_bios_measurements_next( return NULL; if ((event->event_type == 0 && event->event_size == 0) || - ((v + sizeof(struct tcpa_event) + event->event_size) >= limit)) + (event->event_size >= limit - v - sizeof(struct tcpa_event))) return NULL; (*pos)++; @@ -290,7 +290,8 @@ static int tpm_binary_bios_measurements_ int i; for (i = 0; i < sizeof(struct tcpa_event) + event->event_size; i++) - seq_putc(m, data[i]); + if (!seq_putc(m, data[i])) + return -EFAULT; return 0; } @@ -409,8 +410,13 @@ static int read_log(struct tpm_bios_log log->bios_event_log_end = log->bios_event_log + len; virt = acpi_os_map_memory(start, len); + if (!virt) { + kfree(log->bios_event_log); + log->bios_event_log = NULL; + return -EFAULT; + } - memcpy(log->bios_event_log, virt, len); + memcpy(log->bios_event_log, (const char __force_kernel *)virt, len); acpi_os_unmap_memory(virt, len); return 0; diff -urNp linux-2.6.32.48/drivers/char/tpm/tpm.c linux-2.6.32.48-openvz/drivers/char/tpm/tpm.c --- linux-2.6.32.48/drivers/char/tpm/tpm.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/char/tpm/tpm.c 2011-11-17 20:09:04.000000000 -0500 @@ -405,7 +405,7 @@ static ssize_t tpm_transmit(struct tpm_c chip->vendor.req_complete_val) goto out_recv; - if ((status == chip->vendor.req_canceled)) { + if (status == chip->vendor.req_canceled) { dev_err(chip->dev, "Operation Canceled\n"); rc = -ECANCELED; goto out; @@ -824,6 +824,8 @@ ssize_t tpm_show_pubek(struct device *de struct tpm_chip *chip = dev_get_drvdata(dev); + pax_track_stack(); + tpm_cmd.header.in = tpm_readpubek_header; err = transmit_cmd(chip, &tpm_cmd, READ_PUBEK_RESULT_SIZE, "attempting to read the PUBEK"); diff -urNp linux-2.6.32.48/drivers/char/tty_io.c linux-2.6.32.48-openvz/drivers/char/tty_io.c --- linux-2.6.32.48/drivers/char/tty_io.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/char/tty_io.c 2011-11-17 20:09:04.000000000 -0500 @@ -96,6 +96,8 @@ #include #include #include +#include +#include #include #include @@ -106,6 +108,7 @@ #include #include +#include #undef TTY_DEBUG_HANGUP @@ -130,6 +133,7 @@ EXPORT_SYMBOL(tty_std_termios); into this file */ LIST_HEAD(tty_drivers); /* linked list of tty drivers */ +EXPORT_SYMBOL(tty_drivers); /* Mutex to protect creating and releasing a tty. This is shared with vt.c for deeply disgusting hack reasons */ @@ -166,7 +170,7 @@ static void proc_set_tty(struct task_str struct tty_struct *alloc_tty_struct(void) { - return kzalloc(sizeof(struct tty_struct), GFP_KERNEL); + return kzalloc(sizeof(struct tty_struct), GFP_KERNEL_UBC); } /** @@ -274,9 +278,29 @@ static struct tty_driver *get_tty_driver if (device < base || device >= base + p->num) continue; *index = device - base; - return tty_driver_kref_get(p); +#ifdef CONFIG_VE + if (in_interrupt()) + goto found; + if (p->major!=PTY_MASTER_MAJOR && p->major!=PTY_SLAVE_MAJOR +#ifdef CONFIG_UNIX98_PTYS + && (p->majormajor>UNIX98_PTY_MASTER_MAJOR+UNIX98_PTY_MAJOR_COUNT-1) && + (p->majormajor>UNIX98_PTY_SLAVE_MAJOR+UNIX98_PTY_MAJOR_COUNT-1) +#endif + ) + goto found; + if (ve_is_super(p->owner_env) && ve_is_super(get_exec_env())) + goto found; + if (!ve_accessible_strict(p->owner_env, get_exec_env())) + continue; +#endif + goto found; } return NULL; + +found: + return tty_driver_kref_get(p); } #ifdef CONFIG_CONSOLE_POLL @@ -1169,7 +1193,7 @@ int tty_init_termios(struct tty_struct * tp = tty->driver->termios[idx]; if (tp == NULL) { - tp = kzalloc(sizeof(struct ktermios[2]), GFP_KERNEL); + tp = kzalloc(sizeof(struct ktermios[2]), GFP_KERNEL_UBC); if (tp == NULL) return -ENOMEM; memcpy(tp, &tty->driver->init_termios, @@ -1297,7 +1321,7 @@ static int tty_reopen(struct tty_struct */ struct tty_struct *tty_init_dev(struct tty_driver *driver, int idx, - int first_ok) + struct tty_struct *i_tty, int first_ok) { struct tty_struct *tty; int retval; @@ -1707,7 +1731,7 @@ void tty_release_dev(struct file *filp) static int __tty_open(struct inode *inode, struct file *filp) { - struct tty_struct *tty = NULL; + struct tty_struct *tty = NULL, *c_tty = NULL; int noctty, retval; struct tty_driver *driver; int index; @@ -1731,6 +1755,7 @@ retry_open: } driver = tty_driver_kref_get(tty->driver); index = tty->index; + c_tty = tty; filp->f_flags |= O_NONBLOCK; /* Don't let /dev/tty block */ /* noctty = 1; */ /* FIXME: Should we take a driver reference ? */ @@ -1740,6 +1765,12 @@ retry_open: #ifdef CONFIG_VT if (device == MKDEV(TTY_MAJOR, 0)) { extern struct tty_driver *console_driver; +#ifdef CONFIG_VE + if (!ve_is_super(get_exec_env())) { + mutex_unlock(&tty_mutex); + return -ENODEV; + } +#endif driver = tty_driver_kref_get(console_driver); index = fg_console; noctty = 1; @@ -1748,6 +1779,12 @@ retry_open: #endif if (device == MKDEV(TTYAUX_MAJOR, 1)) { struct tty_driver *console_driver = console_device(&index); +#ifdef CONFIG_VE + if (!ve_is_super(get_exec_env())) { + mutex_unlock(&tty_mutex); + return -ENODEV; + } +#endif if (console_driver) { driver = tty_driver_kref_get(console_driver); if (driver) { @@ -1773,6 +1810,7 @@ got_driver: if (IS_ERR(tty)) { mutex_unlock(&tty_mutex); + tty_driver_kref_put(driver); return PTR_ERR(tty); } } @@ -1782,7 +1820,7 @@ got_driver: if (retval) tty = ERR_PTR(retval); } else - tty = tty_init_dev(driver, index, 0); + tty = tty_init_dev(driver, index, c_tty, 0); mutex_unlock(&tty_mutex); tty_driver_kref_put(driver); @@ -2078,6 +2116,8 @@ static int tioccons(struct file *file) { if (!capable(CAP_SYS_ADMIN)) return -EPERM; + if (!ve_is_super(get_exec_env())) + return -EACCES; if (file->f_op->write == redirected_tty_write) { struct file *f; spin_lock(&redirect_lock); @@ -2582,8 +2622,10 @@ long tty_ioctl(struct file *file, unsign return retval; } +EXPORT_SYMBOL(tty_ioctl); + #ifdef CONFIG_COMPAT -static long tty_compat_ioctl(struct file *file, unsigned int cmd, +long tty_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg) { struct inode *inode = file->f_dentry->d_inode; @@ -2607,6 +2649,8 @@ static long tty_compat_ioctl(struct file return retval; } + +EXPORT_SYMBOL(tty_compat_ioctl); #endif /* @@ -2658,7 +2702,7 @@ void __do_SAK(struct tty_struct *tty) /* Now kill any processes that happen to have the * tty open. */ - do_each_thread(g, p) { + do_each_thread_all(g, p) { if (p->signal->tty == tty) { printk(KERN_NOTICE "SAK: killed process %d" " (%s): task_session(p)==tty->session\n", @@ -2690,7 +2734,7 @@ void __do_SAK(struct tty_struct *tty) spin_unlock(&p->files->file_lock); } task_unlock(p); - } while_each_thread(g, p); + } while_each_thread_all(g, p); read_unlock(&tasklist_lock); #endif } @@ -2757,6 +2801,7 @@ void initialize_tty_struct(struct tty_st tty->ops = driver->ops; tty->index = idx; tty_line_name(driver, idx, tty->name); + tty->owner_env = driver->owner_env; } /** @@ -2849,6 +2894,7 @@ struct tty_driver *alloc_tty_driver(int driver->magic = TTY_DRIVER_MAGIC; driver->num = lines; /* later we'll move allocation of tables here */ + driver->owner_env = get_ve(get_exec_env()); } return driver; } @@ -2883,6 +2929,7 @@ static void destruct_tty_driver(struct k kfree(p); cdev_del(&driver->cdev); } + put_ve(driver->owner_env); kfree(driver); } @@ -2957,6 +3004,7 @@ int tty_register_driver(struct tty_drive } mutex_lock(&tty_mutex); + driver->owner_env = get_exec_env(); list_add(&driver->tty_drivers, &tty_drivers); mutex_unlock(&tty_mutex); @@ -3052,7 +3100,7 @@ EXPORT_SYMBOL_GPL(get_current_tty); void tty_default_fops(struct file_operations *fops) { - *fops = tty_fops; + memcpy((void *)fops, &tty_fops, sizeof(tty_fops)); } /* @@ -3130,3 +3178,43 @@ static int __init tty_init(void) return 0; } module_init(tty_init); + +#ifdef CONFIG_UNIX98_PTYS +int init_ve_tty_class(void) +{ + struct class * ve_tty_class; + struct device * ve_ptmx_dev_class; + + ve_tty_class = class_create(THIS_MODULE, "tty"); + if (IS_ERR(ve_tty_class)) + return -ENOMEM; + + ve_ptmx_dev_class = device_create(ve_tty_class, NULL, + MKDEV(TTYAUX_MAJOR, 2), NULL, "ptmx"); + if (IS_ERR(ve_ptmx_dev_class)) { + class_destroy(ve_tty_class); + return PTR_ERR(ve_ptmx_dev_class); + } + + get_exec_env()->tty_class = ve_tty_class; + return 0; +} + +void fini_ve_tty_class(void) +{ + struct class *ve_tty_class = get_exec_env()->tty_class; + + device_destroy(ve_tty_class, MKDEV(TTYAUX_MAJOR, 2)); + class_destroy(ve_tty_class); +} +#else +int init_ve_tty_class(void) +{ + return 0; +} +void fini_ve_tty_class(void) +{ +} +#endif +EXPORT_SYMBOL(init_ve_tty_class); +EXPORT_SYMBOL(fini_ve_tty_class); diff -urNp linux-2.6.32.48/drivers/char/tty_ldisc.c linux-2.6.32.48-openvz/drivers/char/tty_ldisc.c --- linux-2.6.32.48/drivers/char/tty_ldisc.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/char/tty_ldisc.c 2011-11-17 20:09:04.000000000 -0500 @@ -74,7 +74,7 @@ static void put_ldisc(struct tty_ldisc * if (atomic_dec_and_lock(&ld->users, &tty_ldisc_lock)) { struct tty_ldisc_ops *ldo = ld->ops; - ldo->refcount--; + atomic_dec(&ldo->refcount); module_put(ldo->owner); spin_unlock_irqrestore(&tty_ldisc_lock, flags); @@ -109,7 +109,7 @@ int tty_register_ldisc(int disc, struct spin_lock_irqsave(&tty_ldisc_lock, flags); tty_ldiscs[disc] = new_ldisc; new_ldisc->num = disc; - new_ldisc->refcount = 0; + atomic_set(&new_ldisc->refcount, 0); spin_unlock_irqrestore(&tty_ldisc_lock, flags); return ret; @@ -137,7 +137,7 @@ int tty_unregister_ldisc(int disc) return -EINVAL; spin_lock_irqsave(&tty_ldisc_lock, flags); - if (tty_ldiscs[disc]->refcount) + if (atomic_read(&tty_ldiscs[disc]->refcount)) ret = -EBUSY; else tty_ldiscs[disc] = NULL; @@ -158,7 +158,7 @@ static struct tty_ldisc_ops *get_ldops(i if (ldops) { ret = ERR_PTR(-EAGAIN); if (try_module_get(ldops->owner)) { - ldops->refcount++; + atomic_inc(&ldops->refcount); ret = ldops; } } @@ -171,7 +171,7 @@ static void put_ldops(struct tty_ldisc_o unsigned long flags; spin_lock_irqsave(&tty_ldisc_lock, flags); - ldops->refcount--; + atomic_dec(&ldops->refcount); module_put(ldops->owner); spin_unlock_irqrestore(&tty_ldisc_lock, flags); } diff -urNp linux-2.6.32.48/drivers/char/vc_screen.c linux-2.6.32.48-openvz/drivers/char/vc_screen.c --- linux-2.6.32.48/drivers/char/vc_screen.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/char/vc_screen.c 2011-11-17 20:08:53.000000000 -0500 @@ -35,6 +35,8 @@ #include #include #include +#include + #include #include @@ -481,16 +483,22 @@ static struct class *vc_class; void vcs_make_sysfs(int index) { + struct ve_struct *ve = set_exec_env(get_ve0()); + device_create(vc_class, NULL, MKDEV(VCS_MAJOR, index + 1), NULL, "vcs%u", index + 1); device_create(vc_class, NULL, MKDEV(VCS_MAJOR, index + 129), NULL, "vcsa%u", index + 1); + set_exec_env(ve); } void vcs_remove_sysfs(int index) { + struct ve_struct *ve = set_exec_env(get_ve0()); + device_destroy(vc_class, MKDEV(VCS_MAJOR, index + 1)); device_destroy(vc_class, MKDEV(VCS_MAJOR, index + 129)); + set_exec_env(ve); } int __init vcs_init(void) diff -urNp linux-2.6.32.48/drivers/char/virtio_console.c linux-2.6.32.48-openvz/drivers/char/virtio_console.c --- linux-2.6.32.48/drivers/char/virtio_console.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/char/virtio_console.c 2011-11-17 20:09:04.000000000 -0500 @@ -133,7 +133,9 @@ static int get_chars(u32 vtermno, char * * virtqueue, so we let the drivers do some boutique early-output thing. */ int __init virtio_cons_early_init(int (*put_chars)(u32, const char *, int)) { - virtio_cons.put_chars = put_chars; + pax_open_kernel(); + *(void **)&virtio_cons.put_chars = put_chars; + pax_close_kernel(); return hvc_instantiate(0, 0, &virtio_cons); } @@ -213,11 +215,13 @@ static int __devinit virtcons_probe(stru out_vq = vqs[1]; /* Start using the new console output. */ - virtio_cons.get_chars = get_chars; - virtio_cons.put_chars = put_chars; - virtio_cons.notifier_add = notifier_add_vio; - virtio_cons.notifier_del = notifier_del_vio; - virtio_cons.notifier_hangup = notifier_del_vio; + pax_open_kernel(); + *(void **)&virtio_cons.get_chars = get_chars; + *(void **)&virtio_cons.put_chars = put_chars; + *(void **)&virtio_cons.notifier_add = notifier_add_vio; + *(void **)&virtio_cons.notifier_del = notifier_del_vio; + *(void **)&virtio_cons.notifier_hangup = notifier_del_vio; + pax_close_kernel(); /* The first argument of hvc_alloc() is the virtual console number, so * we use zero. The second argument is the parameter for the diff -urNp linux-2.6.32.48/drivers/char/vt.c linux-2.6.32.48-openvz/drivers/char/vt.c --- linux-2.6.32.48/drivers/char/vt.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/char/vt.c 2011-11-17 20:08:53.000000000 -0500 @@ -243,7 +243,7 @@ EXPORT_SYMBOL_GPL(unregister_vt_notifier static void notify_write(struct vc_data *vc, unsigned int unicode) { - struct vt_notifier_param param = { .vc = vc, unicode = unicode }; + struct vt_notifier_param param = { .vc = vc, .c = unicode }; atomic_notifier_call_chain(&vt_notifier_list, VT_WRITE, ¶m); } diff -urNp linux-2.6.32.48/drivers/char/vt_ioctl.c linux-2.6.32.48-openvz/drivers/char/vt_ioctl.c --- linux-2.6.32.48/drivers/char/vt_ioctl.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/char/vt_ioctl.c 2011-11-17 20:08:53.000000000 -0500 @@ -210,9 +210,6 @@ do_kdsk_ioctl(int cmd, struct kbentry __ if (copy_from_user(&tmp, user_kbe, sizeof(struct kbentry))) return -EFAULT; - if (!capable(CAP_SYS_TTY_CONFIG)) - perm = 0; - switch (cmd) { case KDGKBENT: key_map = key_maps[s]; @@ -224,8 +221,12 @@ do_kdsk_ioctl(int cmd, struct kbentry __ val = (i ? K_HOLE : K_NOSUCHMAP); return put_user(val, &user_kbe->kb_value); case KDSKBENT: + if (!capable(CAP_SYS_TTY_CONFIG)) + perm = 0; + if (!perm) return -EPERM; + if (!i && v == K_NOSUCHMAP) { /* deallocate map */ key_map = key_maps[s]; @@ -325,9 +326,6 @@ do_kdgkb_ioctl(int cmd, struct kbsentry int i, j, k; int ret; - if (!capable(CAP_SYS_TTY_CONFIG)) - perm = 0; - kbs = kmalloc(sizeof(*kbs), GFP_KERNEL); if (!kbs) { ret = -ENOMEM; @@ -361,6 +359,9 @@ do_kdgkb_ioctl(int cmd, struct kbsentry kfree(kbs); return ((p && *p) ? -EOVERFLOW : 0); case KDSKBSENT: + if (!capable(CAP_SYS_TTY_CONFIG)) + perm = 0; + if (!perm) { ret = -EPERM; goto reterr; diff -urNp linux-2.6.32.48/drivers/cpufreq/cpufreq.c linux-2.6.32.48-openvz/drivers/cpufreq/cpufreq.c --- linux-2.6.32.48/drivers/cpufreq/cpufreq.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/cpufreq/cpufreq.c 2011-11-17 20:09:04.000000000 -0500 @@ -750,7 +750,7 @@ static void cpufreq_sysfs_release(struct complete(&policy->kobj_unregister); } -static struct sysfs_ops sysfs_ops = { +static const struct sysfs_ops sysfs_ops = { .show = show, .store = store, }; diff -urNp linux-2.6.32.48/drivers/cpuidle/sysfs.c linux-2.6.32.48-openvz/drivers/cpuidle/sysfs.c --- linux-2.6.32.48/drivers/cpuidle/sysfs.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/cpuidle/sysfs.c 2011-11-17 20:08:53.000000000 -0500 @@ -191,7 +191,7 @@ static ssize_t cpuidle_store(struct kobj return ret; } -static struct sysfs_ops cpuidle_sysfs_ops = { +static const struct sysfs_ops cpuidle_sysfs_ops = { .show = cpuidle_show, .store = cpuidle_store, }; @@ -277,7 +277,7 @@ static ssize_t cpuidle_state_show(struct return ret; } -static struct sysfs_ops cpuidle_state_sysfs_ops = { +static const struct sysfs_ops cpuidle_state_sysfs_ops = { .show = cpuidle_state_show, }; @@ -294,7 +294,7 @@ static struct kobj_type ktype_state_cpui .release = cpuidle_state_sysfs_release, }; -static void inline cpuidle_free_state_kobj(struct cpuidle_device *device, int i) +static inline void cpuidle_free_state_kobj(struct cpuidle_device *device, int i) { kobject_put(&device->kobjs[i]->kobj); wait_for_completion(&device->kobjs[i]->kobj_unregister); diff -urNp linux-2.6.32.48/drivers/crypto/hifn_795x.c linux-2.6.32.48-openvz/drivers/crypto/hifn_795x.c --- linux-2.6.32.48/drivers/crypto/hifn_795x.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/crypto/hifn_795x.c 2011-11-17 20:08:53.000000000 -0500 @@ -1655,6 +1655,8 @@ static int hifn_test(struct hifn_device 0xCA, 0x34, 0x2B, 0x2E}; struct scatterlist sg; + pax_track_stack(); + memset(src, 0, sizeof(src)); memset(ctx.key, 0, sizeof(ctx.key)); diff -urNp linux-2.6.32.48/drivers/crypto/padlock-aes.c linux-2.6.32.48-openvz/drivers/crypto/padlock-aes.c --- linux-2.6.32.48/drivers/crypto/padlock-aes.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/crypto/padlock-aes.c 2011-11-17 20:08:53.000000000 -0500 @@ -108,6 +108,8 @@ static int aes_set_key(struct crypto_tfm struct crypto_aes_ctx gen_aes; int cpu; + pax_track_stack(); + if (key_len % 8) { *flags |= CRYPTO_TFM_RES_BAD_KEY_LEN; return -EINVAL; diff -urNp linux-2.6.32.48/drivers/dma/ioat/dma.c linux-2.6.32.48-openvz/drivers/dma/ioat/dma.c --- linux-2.6.32.48/drivers/dma/ioat/dma.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/dma/ioat/dma.c 2011-11-17 20:08:53.000000000 -0500 @@ -1146,7 +1146,7 @@ ioat_attr_show(struct kobject *kobj, str return entry->show(&chan->common, page); } -struct sysfs_ops ioat_sysfs_ops = { +const struct sysfs_ops ioat_sysfs_ops = { .show = ioat_attr_show, }; diff -urNp linux-2.6.32.48/drivers/dma/ioat/dma.h linux-2.6.32.48-openvz/drivers/dma/ioat/dma.h --- linux-2.6.32.48/drivers/dma/ioat/dma.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/dma/ioat/dma.h 2011-11-17 20:08:53.000000000 -0500 @@ -347,7 +347,7 @@ bool ioat_cleanup_preamble(struct ioat_c unsigned long *phys_complete); void ioat_kobject_add(struct ioatdma_device *device, struct kobj_type *type); void ioat_kobject_del(struct ioatdma_device *device); -extern struct sysfs_ops ioat_sysfs_ops; +extern const struct sysfs_ops ioat_sysfs_ops; extern struct ioat_sysfs_entry ioat_version_attr; extern struct ioat_sysfs_entry ioat_cap_attr; #endif /* IOATDMA_H */ diff -urNp linux-2.6.32.48/drivers/edac/edac_device_sysfs.c linux-2.6.32.48-openvz/drivers/edac/edac_device_sysfs.c --- linux-2.6.32.48/drivers/edac/edac_device_sysfs.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/edac/edac_device_sysfs.c 2011-11-17 20:08:53.000000000 -0500 @@ -137,7 +137,7 @@ static ssize_t edac_dev_ctl_info_store(s } /* edac_dev file operations for an 'ctl_info' */ -static struct sysfs_ops device_ctl_info_ops = { +static const struct sysfs_ops device_ctl_info_ops = { .show = edac_dev_ctl_info_show, .store = edac_dev_ctl_info_store }; @@ -373,7 +373,7 @@ static ssize_t edac_dev_instance_store(s } /* edac_dev file operations for an 'instance' */ -static struct sysfs_ops device_instance_ops = { +static const struct sysfs_ops device_instance_ops = { .show = edac_dev_instance_show, .store = edac_dev_instance_store }; @@ -476,7 +476,7 @@ static ssize_t edac_dev_block_store(stru } /* edac_dev file operations for a 'block' */ -static struct sysfs_ops device_block_ops = { +static const struct sysfs_ops device_block_ops = { .show = edac_dev_block_show, .store = edac_dev_block_store }; diff -urNp linux-2.6.32.48/drivers/edac/edac_mc_sysfs.c linux-2.6.32.48-openvz/drivers/edac/edac_mc_sysfs.c --- linux-2.6.32.48/drivers/edac/edac_mc_sysfs.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/edac/edac_mc_sysfs.c 2011-11-17 20:08:53.000000000 -0500 @@ -245,7 +245,7 @@ static ssize_t csrowdev_store(struct kob return -EIO; } -static struct sysfs_ops csrowfs_ops = { +static const struct sysfs_ops csrowfs_ops = { .show = csrowdev_show, .store = csrowdev_store }; @@ -575,7 +575,7 @@ static ssize_t mcidev_store(struct kobje } /* Intermediate show/store table */ -static struct sysfs_ops mci_ops = { +static const struct sysfs_ops mci_ops = { .show = mcidev_show, .store = mcidev_store }; diff -urNp linux-2.6.32.48/drivers/edac/edac_pci_sysfs.c linux-2.6.32.48-openvz/drivers/edac/edac_pci_sysfs.c --- linux-2.6.32.48/drivers/edac/edac_pci_sysfs.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/edac/edac_pci_sysfs.c 2011-11-17 20:08:53.000000000 -0500 @@ -25,8 +25,8 @@ static int edac_pci_log_pe = 1; /* log static int edac_pci_log_npe = 1; /* log PCI non-parity error errors */ static int edac_pci_poll_msec = 1000; /* one second workq period */ -static atomic_t pci_parity_count = ATOMIC_INIT(0); -static atomic_t pci_nonparity_count = ATOMIC_INIT(0); +static atomic_unchecked_t pci_parity_count = ATOMIC_INIT(0); +static atomic_unchecked_t pci_nonparity_count = ATOMIC_INIT(0); static struct kobject *edac_pci_top_main_kobj; static atomic_t edac_pci_sysfs_refcount = ATOMIC_INIT(0); @@ -121,7 +121,7 @@ static ssize_t edac_pci_instance_store(s } /* fs_ops table */ -static struct sysfs_ops pci_instance_ops = { +static const struct sysfs_ops pci_instance_ops = { .show = edac_pci_instance_show, .store = edac_pci_instance_store }; @@ -261,7 +261,7 @@ static ssize_t edac_pci_dev_store(struct return -EIO; } -static struct sysfs_ops edac_pci_sysfs_ops = { +static const struct sysfs_ops edac_pci_sysfs_ops = { .show = edac_pci_dev_show, .store = edac_pci_dev_store }; @@ -579,7 +579,7 @@ static void edac_pci_dev_parity_test(str edac_printk(KERN_CRIT, EDAC_PCI, "Signaled System Error on %s\n", pci_name(dev)); - atomic_inc(&pci_nonparity_count); + atomic_inc_unchecked(&pci_nonparity_count); } if (status & (PCI_STATUS_PARITY)) { @@ -587,7 +587,7 @@ static void edac_pci_dev_parity_test(str "Master Data Parity Error on %s\n", pci_name(dev)); - atomic_inc(&pci_parity_count); + atomic_inc_unchecked(&pci_parity_count); } if (status & (PCI_STATUS_DETECTED_PARITY)) { @@ -595,7 +595,7 @@ static void edac_pci_dev_parity_test(str "Detected Parity Error on %s\n", pci_name(dev)); - atomic_inc(&pci_parity_count); + atomic_inc_unchecked(&pci_parity_count); } } @@ -616,7 +616,7 @@ static void edac_pci_dev_parity_test(str edac_printk(KERN_CRIT, EDAC_PCI, "Bridge " "Signaled System Error on %s\n", pci_name(dev)); - atomic_inc(&pci_nonparity_count); + atomic_inc_unchecked(&pci_nonparity_count); } if (status & (PCI_STATUS_PARITY)) { @@ -624,7 +624,7 @@ static void edac_pci_dev_parity_test(str "Master Data Parity Error on " "%s\n", pci_name(dev)); - atomic_inc(&pci_parity_count); + atomic_inc_unchecked(&pci_parity_count); } if (status & (PCI_STATUS_DETECTED_PARITY)) { @@ -632,7 +632,7 @@ static void edac_pci_dev_parity_test(str "Detected Parity Error on %s\n", pci_name(dev)); - atomic_inc(&pci_parity_count); + atomic_inc_unchecked(&pci_parity_count); } } } @@ -674,7 +674,7 @@ void edac_pci_do_parity_check(void) if (!check_pci_errors) return; - before_count = atomic_read(&pci_parity_count); + before_count = atomic_read_unchecked(&pci_parity_count); /* scan all PCI devices looking for a Parity Error on devices and * bridges. @@ -686,7 +686,7 @@ void edac_pci_do_parity_check(void) /* Only if operator has selected panic on PCI Error */ if (edac_pci_get_panic_on_pe()) { /* If the count is different 'after' from 'before' */ - if (before_count != atomic_read(&pci_parity_count)) + if (before_count != atomic_read_unchecked(&pci_parity_count)) panic("EDAC: PCI Parity Error"); } } diff -urNp linux-2.6.32.48/drivers/firewire/core-card.c linux-2.6.32.48-openvz/drivers/firewire/core-card.c --- linux-2.6.32.48/drivers/firewire/core-card.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/firewire/core-card.c 2011-11-17 20:09:04.000000000 -0500 @@ -558,7 +558,7 @@ void fw_card_release(struct kref *kref) void fw_core_remove_card(struct fw_card *card) { - struct fw_card_driver dummy_driver = dummy_driver_template; + fw_card_driver_no_const dummy_driver = dummy_driver_template; card->driver->update_phy_reg(card, 4, PHY_LINK_ACTIVE | PHY_CONTENDER, 0); diff -urNp linux-2.6.32.48/drivers/firewire/core-cdev.c linux-2.6.32.48-openvz/drivers/firewire/core-cdev.c --- linux-2.6.32.48/drivers/firewire/core-cdev.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/firewire/core-cdev.c 2011-11-17 20:08:53.000000000 -0500 @@ -1141,8 +1141,7 @@ static int init_iso_resource(struct clie int ret; if ((request->channels == 0 && request->bandwidth == 0) || - request->bandwidth > BANDWIDTH_AVAILABLE_INITIAL || - request->bandwidth < 0) + request->bandwidth > BANDWIDTH_AVAILABLE_INITIAL) return -EINVAL; r = kmalloc(sizeof(*r), GFP_KERNEL); diff -urNp linux-2.6.32.48/drivers/firewire/core.h linux-2.6.32.48-openvz/drivers/firewire/core.h --- linux-2.6.32.48/drivers/firewire/core.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/firewire/core.h 2011-11-17 20:09:04.000000000 -0500 @@ -86,6 +86,7 @@ struct fw_card_driver { int (*stop_iso)(struct fw_iso_context *ctx); }; +typedef struct fw_card_driver __no_const fw_card_driver_no_const; void fw_card_initialize(struct fw_card *card, const struct fw_card_driver *driver, struct device *device); diff -urNp linux-2.6.32.48/drivers/firewire/core-transaction.c linux-2.6.32.48-openvz/drivers/firewire/core-transaction.c --- linux-2.6.32.48/drivers/firewire/core-transaction.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/firewire/core-transaction.c 2011-11-17 20:08:53.000000000 -0500 @@ -36,6 +36,7 @@ #include #include #include +#include #include @@ -344,6 +345,8 @@ int fw_run_transaction(struct fw_card *c struct transaction_callback_data d; struct fw_transaction t; + pax_track_stack(); + init_completion(&d.done); d.payload = payload; fw_send_request(card, &t, tcode, destination_id, generation, speed, diff -urNp linux-2.6.32.48/drivers/firmware/dmi_scan.c linux-2.6.32.48-openvz/drivers/firmware/dmi_scan.c --- linux-2.6.32.48/drivers/firmware/dmi_scan.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/firmware/dmi_scan.c 2011-11-17 20:09:04.000000000 -0500 @@ -391,11 +391,6 @@ void __init dmi_scan_machine(void) } } else { - /* - * no iounmap() for that ioremap(); it would be a no-op, but - * it's so early in setup that sucker gets confused into doing - * what it shouldn't if we actually call it. - */ p = dmi_ioremap(0xF0000, 0x10000); if (p == NULL) goto error; @@ -667,7 +662,7 @@ int dmi_walk(void (*decode)(const struct if (buf == NULL) return -1; - dmi_table(buf, dmi_len, dmi_num, decode, private_data); + dmi_table((char __force_kernel *)buf, dmi_len, dmi_num, decode, private_data); iounmap(buf); return 0; diff -urNp linux-2.6.32.48/drivers/firmware/edd.c linux-2.6.32.48-openvz/drivers/firmware/edd.c --- linux-2.6.32.48/drivers/firmware/edd.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/firmware/edd.c 2011-11-17 20:08:53.000000000 -0500 @@ -122,7 +122,7 @@ edd_attr_show(struct kobject * kobj, str return ret; } -static struct sysfs_ops edd_attr_ops = { +static const struct sysfs_ops edd_attr_ops = { .show = edd_attr_show, }; diff -urNp linux-2.6.32.48/drivers/firmware/efivars.c linux-2.6.32.48-openvz/drivers/firmware/efivars.c --- linux-2.6.32.48/drivers/firmware/efivars.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/firmware/efivars.c 2011-11-17 20:08:53.000000000 -0500 @@ -362,7 +362,7 @@ static ssize_t efivar_attr_store(struct return ret; } -static struct sysfs_ops efivar_attr_ops = { +static const struct sysfs_ops efivar_attr_ops = { .show = efivar_attr_show, .store = efivar_attr_store, }; diff -urNp linux-2.6.32.48/drivers/firmware/iscsi_ibft.c linux-2.6.32.48-openvz/drivers/firmware/iscsi_ibft.c --- linux-2.6.32.48/drivers/firmware/iscsi_ibft.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/firmware/iscsi_ibft.c 2011-11-17 20:08:53.000000000 -0500 @@ -525,7 +525,7 @@ static ssize_t ibft_show_attribute(struc return ret; } -static struct sysfs_ops ibft_attr_ops = { +static const struct sysfs_ops ibft_attr_ops = { .show = ibft_show_attribute, }; diff -urNp linux-2.6.32.48/drivers/firmware/memmap.c linux-2.6.32.48-openvz/drivers/firmware/memmap.c --- linux-2.6.32.48/drivers/firmware/memmap.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/firmware/memmap.c 2011-11-17 20:08:53.000000000 -0500 @@ -74,7 +74,7 @@ static struct attribute *def_attrs[] = { NULL }; -static struct sysfs_ops memmap_attr_ops = { +static const struct sysfs_ops memmap_attr_ops = { .show = memmap_attr_show, }; diff -urNp linux-2.6.32.48/drivers/gpio/vr41xx_giu.c linux-2.6.32.48-openvz/drivers/gpio/vr41xx_giu.c --- linux-2.6.32.48/drivers/gpio/vr41xx_giu.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpio/vr41xx_giu.c 2011-11-17 20:08:53.000000000 -0500 @@ -204,7 +204,7 @@ static int giu_get_irq(unsigned int irq) printk(KERN_ERR "spurious GIU interrupt: %04x(%04x),%04x(%04x)\n", maskl, pendl, maskh, pendh); - atomic_inc(&irq_err_count); + atomic_inc_unchecked(&irq_err_count); return -EINVAL; } diff -urNp linux-2.6.32.48/drivers/gpu/drm/drm_crtc.c linux-2.6.32.48-openvz/drivers/gpu/drm/drm_crtc.c --- linux-2.6.32.48/drivers/gpu/drm/drm_crtc.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/drm_crtc.c 2011-11-17 20:09:04.000000000 -0500 @@ -1323,7 +1323,7 @@ int drm_mode_getconnector(struct drm_dev */ if ((out_resp->count_modes >= mode_count) && mode_count) { copied = 0; - mode_ptr = (struct drm_mode_modeinfo *)(unsigned long)out_resp->modes_ptr; + mode_ptr = (struct drm_mode_modeinfo __user *)(unsigned long)out_resp->modes_ptr; list_for_each_entry(mode, &connector->modes, head) { drm_crtc_convert_to_umode(&u_mode, mode); if (copy_to_user(mode_ptr + copied, @@ -1338,8 +1338,8 @@ int drm_mode_getconnector(struct drm_dev if ((out_resp->count_props >= props_count) && props_count) { copied = 0; - prop_ptr = (uint32_t *)(unsigned long)(out_resp->props_ptr); - prop_values = (uint64_t *)(unsigned long)(out_resp->prop_values_ptr); + prop_ptr = (uint32_t __user *)(unsigned long)(out_resp->props_ptr); + prop_values = (uint64_t __user *)(unsigned long)(out_resp->prop_values_ptr); for (i = 0; i < DRM_CONNECTOR_MAX_PROPERTY; i++) { if (connector->property_ids[i] != 0) { if (put_user(connector->property_ids[i], @@ -1361,7 +1361,7 @@ int drm_mode_getconnector(struct drm_dev if ((out_resp->count_encoders >= encoders_count) && encoders_count) { copied = 0; - encoder_ptr = (uint32_t *)(unsigned long)(out_resp->encoders_ptr); + encoder_ptr = (uint32_t __user *)(unsigned long)(out_resp->encoders_ptr); for (i = 0; i < DRM_CONNECTOR_MAX_ENCODER; i++) { if (connector->encoder_ids[i] != 0) { if (put_user(connector->encoder_ids[i], @@ -1513,7 +1513,7 @@ int drm_mode_setcrtc(struct drm_device * } for (i = 0; i < crtc_req->count_connectors; i++) { - set_connectors_ptr = (uint32_t *)(unsigned long)crtc_req->set_connectors_ptr; + set_connectors_ptr = (uint32_t __user *)(unsigned long)crtc_req->set_connectors_ptr; if (get_user(out_id, &set_connectors_ptr[i])) { ret = -EFAULT; goto out; @@ -2118,7 +2118,7 @@ int drm_mode_getproperty_ioctl(struct dr out_resp->flags = property->flags; if ((out_resp->count_values >= value_count) && value_count) { - values_ptr = (uint64_t *)(unsigned long)out_resp->values_ptr; + values_ptr = (uint64_t __user *)(unsigned long)out_resp->values_ptr; for (i = 0; i < value_count; i++) { if (copy_to_user(values_ptr + i, &property->values[i], sizeof(uint64_t))) { ret = -EFAULT; @@ -2131,7 +2131,7 @@ int drm_mode_getproperty_ioctl(struct dr if (property->flags & DRM_MODE_PROP_ENUM) { if ((out_resp->count_enum_blobs >= enum_count) && enum_count) { copied = 0; - enum_ptr = (struct drm_mode_property_enum *)(unsigned long)out_resp->enum_blob_ptr; + enum_ptr = (struct drm_mode_property_enum __user *)(unsigned long)out_resp->enum_blob_ptr; list_for_each_entry(prop_enum, &property->enum_blob_list, head) { if (copy_to_user(&enum_ptr[copied].value, &prop_enum->value, sizeof(uint64_t))) { @@ -2154,7 +2154,7 @@ int drm_mode_getproperty_ioctl(struct dr if ((out_resp->count_enum_blobs >= blob_count) && blob_count) { copied = 0; blob_id_ptr = (uint32_t *)(unsigned long)out_resp->enum_blob_ptr; - blob_length_ptr = (uint32_t *)(unsigned long)out_resp->values_ptr; + blob_length_ptr = (uint32_t __user *)(unsigned long)out_resp->values_ptr; list_for_each_entry(prop_blob, &property->enum_blob_list, head) { if (put_user(prop_blob->base.id, blob_id_ptr + copied)) { @@ -2226,7 +2226,7 @@ int drm_mode_getblob_ioctl(struct drm_de blob = obj_to_blob(obj); if (out_resp->length == blob->length) { - blob_ptr = (void *)(unsigned long)out_resp->data; + blob_ptr = (void __user *)(unsigned long)out_resp->data; if (copy_to_user(blob_ptr, blob->data, blob->length)){ ret = -EFAULT; goto done; diff -urNp linux-2.6.32.48/drivers/gpu/drm/drm_crtc_helper.c linux-2.6.32.48-openvz/drivers/gpu/drm/drm_crtc_helper.c --- linux-2.6.32.48/drivers/gpu/drm/drm_crtc_helper.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/drm_crtc_helper.c 2011-11-17 20:08:53.000000000 -0500 @@ -573,7 +573,7 @@ static bool drm_encoder_crtc_ok(struct d struct drm_crtc *tmp; int crtc_mask = 1; - WARN(!crtc, "checking null crtc?"); + BUG_ON(!crtc); dev = crtc->dev; @@ -642,6 +642,8 @@ bool drm_crtc_helper_set_mode(struct drm adjusted_mode = drm_mode_duplicate(dev, mode); + pax_track_stack(); + crtc->enabled = drm_helper_crtc_in_use(crtc); if (!crtc->enabled) diff -urNp linux-2.6.32.48/drivers/gpu/drm/drm_drv.c linux-2.6.32.48-openvz/drivers/gpu/drm/drm_drv.c --- linux-2.6.32.48/drivers/gpu/drm/drm_drv.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/drm_drv.c 2011-11-17 20:08:53.000000000 -0500 @@ -417,7 +417,7 @@ int drm_ioctl(struct inode *inode, struc char *kdata = NULL; atomic_inc(&dev->ioctl_count); - atomic_inc(&dev->counts[_DRM_STAT_IOCTLS]); + atomic_inc_unchecked(&dev->counts[_DRM_STAT_IOCTLS]); ++file_priv->ioctl_count; DRM_DEBUG("pid=%d, cmd=0x%02x, nr=0x%02x, dev 0x%lx, auth=%d\n", diff -urNp linux-2.6.32.48/drivers/gpu/drm/drm_fops.c linux-2.6.32.48-openvz/drivers/gpu/drm/drm_fops.c --- linux-2.6.32.48/drivers/gpu/drm/drm_fops.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/drm_fops.c 2011-11-17 20:08:53.000000000 -0500 @@ -66,7 +66,7 @@ static int drm_setup(struct drm_device * } for (i = 0; i < ARRAY_SIZE(dev->counts); i++) - atomic_set(&dev->counts[i], 0); + atomic_set_unchecked(&dev->counts[i], 0); dev->sigdata.lock = NULL; @@ -130,9 +130,9 @@ int drm_open(struct inode *inode, struct retcode = drm_open_helper(inode, filp, dev); if (!retcode) { - atomic_inc(&dev->counts[_DRM_STAT_OPENS]); + atomic_inc_unchecked(&dev->counts[_DRM_STAT_OPENS]); spin_lock(&dev->count_lock); - if (!dev->open_count++) { + if (local_inc_return(&dev->open_count) == 1) { spin_unlock(&dev->count_lock); retcode = drm_setup(dev); goto out; @@ -435,7 +435,7 @@ int drm_release(struct inode *inode, str lock_kernel(); - DRM_DEBUG("open_count = %d\n", dev->open_count); + DRM_DEBUG("open_count = %d\n", local_read(&dev->open_count)); if (dev->driver->preclose) dev->driver->preclose(dev, file_priv); @@ -447,7 +447,7 @@ int drm_release(struct inode *inode, str DRM_DEBUG("pid = %d, device = 0x%lx, open_count = %d\n", task_pid_nr(current), (long)old_encode_dev(file_priv->minor->device), - dev->open_count); + local_read(&dev->open_count)); /* if the master has gone away we can't do anything with the lock */ if (file_priv->minor->master) @@ -524,9 +524,9 @@ int drm_release(struct inode *inode, str * End inline drm_release */ - atomic_inc(&dev->counts[_DRM_STAT_CLOSES]); + atomic_inc_unchecked(&dev->counts[_DRM_STAT_CLOSES]); spin_lock(&dev->count_lock); - if (!--dev->open_count) { + if (local_dec_and_test(&dev->open_count)) { if (atomic_read(&dev->ioctl_count)) { DRM_ERROR("Device busy: %d\n", atomic_read(&dev->ioctl_count)); diff -urNp linux-2.6.32.48/drivers/gpu/drm/drm_gem.c linux-2.6.32.48-openvz/drivers/gpu/drm/drm_gem.c --- linux-2.6.32.48/drivers/gpu/drm/drm_gem.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/drm_gem.c 2011-11-17 20:08:53.000000000 -0500 @@ -83,11 +83,11 @@ drm_gem_init(struct drm_device *dev) spin_lock_init(&dev->object_name_lock); idr_init(&dev->object_name_idr); atomic_set(&dev->object_count, 0); - atomic_set(&dev->object_memory, 0); + atomic_set_unchecked(&dev->object_memory, 0); atomic_set(&dev->pin_count, 0); - atomic_set(&dev->pin_memory, 0); + atomic_set_unchecked(&dev->pin_memory, 0); atomic_set(&dev->gtt_count, 0); - atomic_set(&dev->gtt_memory, 0); + atomic_set_unchecked(&dev->gtt_memory, 0); mm = kzalloc(sizeof(struct drm_gem_mm), GFP_KERNEL); if (!mm) { @@ -150,7 +150,7 @@ drm_gem_object_alloc(struct drm_device * goto fput; } atomic_inc(&dev->object_count); - atomic_add(obj->size, &dev->object_memory); + atomic_add_unchecked(obj->size, &dev->object_memory); return obj; fput: fput(obj->filp); @@ -429,7 +429,7 @@ drm_gem_object_free(struct kref *kref) fput(obj->filp); atomic_dec(&dev->object_count); - atomic_sub(obj->size, &dev->object_memory); + atomic_sub_unchecked(obj->size, &dev->object_memory); kfree(obj); } EXPORT_SYMBOL(drm_gem_object_free); diff -urNp linux-2.6.32.48/drivers/gpu/drm/drm_info.c linux-2.6.32.48-openvz/drivers/gpu/drm/drm_info.c --- linux-2.6.32.48/drivers/gpu/drm/drm_info.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/drm_info.c 2011-11-17 20:08:53.000000000 -0500 @@ -75,10 +75,14 @@ int drm_vm_info(struct seq_file *m, void struct drm_local_map *map; struct drm_map_list *r_list; - /* Hardcoded from _DRM_FRAME_BUFFER, - _DRM_REGISTERS, _DRM_SHM, _DRM_AGP, and - _DRM_SCATTER_GATHER and _DRM_CONSISTENT */ - const char *types[] = { "FB", "REG", "SHM", "AGP", "SG", "PCI" }; + static const char * const types[] = { + [_DRM_FRAME_BUFFER] = "FB", + [_DRM_REGISTERS] = "REG", + [_DRM_SHM] = "SHM", + [_DRM_AGP] = "AGP", + [_DRM_SCATTER_GATHER] = "SG", + [_DRM_CONSISTENT] = "PCI", + [_DRM_GEM] = "GEM" }; const char *type; int i; @@ -89,7 +93,7 @@ int drm_vm_info(struct seq_file *m, void map = r_list->map; if (!map) continue; - if (map->type < 0 || map->type > 5) + if (map->type >= ARRAY_SIZE(types)) type = "??"; else type = types[map->type]; @@ -265,10 +269,10 @@ int drm_gem_object_info(struct seq_file struct drm_device *dev = node->minor->dev; seq_printf(m, "%d objects\n", atomic_read(&dev->object_count)); - seq_printf(m, "%d object bytes\n", atomic_read(&dev->object_memory)); + seq_printf(m, "%d object bytes\n", atomic_read_unchecked(&dev->object_memory)); seq_printf(m, "%d pinned\n", atomic_read(&dev->pin_count)); - seq_printf(m, "%d pin bytes\n", atomic_read(&dev->pin_memory)); - seq_printf(m, "%d gtt bytes\n", atomic_read(&dev->gtt_memory)); + seq_printf(m, "%d pin bytes\n", atomic_read_unchecked(&dev->pin_memory)); + seq_printf(m, "%d gtt bytes\n", atomic_read_unchecked(&dev->gtt_memory)); seq_printf(m, "%d gtt total\n", dev->gtt_total); return 0; } @@ -288,7 +292,11 @@ int drm_vma_info(struct seq_file *m, voi mutex_lock(&dev->struct_mutex); seq_printf(m, "vma use count: %d, high_memory = %p, 0x%08llx\n", atomic_read(&dev->vma_count), +#ifdef CONFIG_GRKERNSEC_HIDESYM + NULL, 0); +#else high_memory, (u64)virt_to_phys(high_memory)); +#endif list_for_each_entry(pt, &dev->vmalist, head) { vma = pt->vma; @@ -296,14 +304,23 @@ int drm_vma_info(struct seq_file *m, voi continue; seq_printf(m, "\n%5d 0x%08lx-0x%08lx %c%c%c%c%c%c 0x%08lx000", - pt->pid, vma->vm_start, vma->vm_end, + pt->pid, +#ifdef CONFIG_GRKERNSEC_HIDESYM + 0, 0, +#else + vma->vm_start, vma->vm_end, +#endif vma->vm_flags & VM_READ ? 'r' : '-', vma->vm_flags & VM_WRITE ? 'w' : '-', vma->vm_flags & VM_EXEC ? 'x' : '-', vma->vm_flags & VM_MAYSHARE ? 's' : 'p', vma->vm_flags & VM_LOCKED ? 'l' : '-', vma->vm_flags & VM_IO ? 'i' : '-', +#ifdef CONFIG_GRKERNSEC_HIDESYM + 0); +#else vma->vm_pgoff); +#endif #if defined(__i386__) pgprot = pgprot_val(vma->vm_page_prot); diff -urNp linux-2.6.32.48/drivers/gpu/drm/drm_ioc32.c linux-2.6.32.48-openvz/drivers/gpu/drm/drm_ioc32.c --- linux-2.6.32.48/drivers/gpu/drm/drm_ioc32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/drm_ioc32.c 2011-11-17 20:09:04.000000000 -0500 @@ -463,7 +463,7 @@ static int compat_drm_infobufs(struct fi request = compat_alloc_user_space(nbytes); if (!access_ok(VERIFY_WRITE, request, nbytes)) return -EFAULT; - list = (struct drm_buf_desc *) (request + 1); + list = (struct drm_buf_desc __user *) (request + 1); if (__put_user(count, &request->count) || __put_user(list, &request->list)) @@ -525,7 +525,7 @@ static int compat_drm_mapbufs(struct fil request = compat_alloc_user_space(nbytes); if (!access_ok(VERIFY_WRITE, request, nbytes)) return -EFAULT; - list = (struct drm_buf_pub *) (request + 1); + list = (struct drm_buf_pub __user *) (request + 1); if (__put_user(count, &request->count) || __put_user(list, &request->list)) diff -urNp linux-2.6.32.48/drivers/gpu/drm/drm_ioctl.c linux-2.6.32.48-openvz/drivers/gpu/drm/drm_ioctl.c --- linux-2.6.32.48/drivers/gpu/drm/drm_ioctl.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/drm_ioctl.c 2011-11-17 20:08:53.000000000 -0500 @@ -283,7 +283,7 @@ int drm_getstats(struct drm_device *dev, stats->data[i].value = (file_priv->master->lock.hw_lock ? file_priv->master->lock.hw_lock->lock : 0); else - stats->data[i].value = atomic_read(&dev->counts[i]); + stats->data[i].value = atomic_read_unchecked(&dev->counts[i]); stats->data[i].type = dev->types[i]; } diff -urNp linux-2.6.32.48/drivers/gpu/drm/drm_lock.c linux-2.6.32.48-openvz/drivers/gpu/drm/drm_lock.c --- linux-2.6.32.48/drivers/gpu/drm/drm_lock.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/drm_lock.c 2011-11-17 20:08:53.000000000 -0500 @@ -87,7 +87,7 @@ int drm_lock(struct drm_device *dev, voi if (drm_lock_take(&master->lock, lock->context)) { master->lock.file_priv = file_priv; master->lock.lock_time = jiffies; - atomic_inc(&dev->counts[_DRM_STAT_LOCKS]); + atomic_inc_unchecked(&dev->counts[_DRM_STAT_LOCKS]); break; /* Got lock */ } @@ -165,7 +165,7 @@ int drm_unlock(struct drm_device *dev, v return -EINVAL; } - atomic_inc(&dev->counts[_DRM_STAT_UNLOCKS]); + atomic_inc_unchecked(&dev->counts[_DRM_STAT_UNLOCKS]); /* kernel_context_switch isn't used by any of the x86 drm * modules but is required by the Sparc driver. diff -urNp linux-2.6.32.48/drivers/gpu/drm/i810/i810_dma.c linux-2.6.32.48-openvz/drivers/gpu/drm/i810/i810_dma.c --- linux-2.6.32.48/drivers/gpu/drm/i810/i810_dma.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/i810/i810_dma.c 2011-11-17 20:08:53.000000000 -0500 @@ -952,8 +952,8 @@ static int i810_dma_vertex(struct drm_de dma->buflist[vertex->idx], vertex->discard, vertex->used); - atomic_add(vertex->used, &dev->counts[_DRM_STAT_SECONDARY]); - atomic_inc(&dev->counts[_DRM_STAT_DMA]); + atomic_add_unchecked(vertex->used, &dev->counts[_DRM_STAT_SECONDARY]); + atomic_inc_unchecked(&dev->counts[_DRM_STAT_DMA]); sarea_priv->last_enqueue = dev_priv->counter - 1; sarea_priv->last_dispatch = (int)hw_status[5]; @@ -1115,8 +1115,8 @@ static int i810_dma_mc(struct drm_device i810_dma_dispatch_mc(dev, dma->buflist[mc->idx], mc->used, mc->last_render); - atomic_add(mc->used, &dev->counts[_DRM_STAT_SECONDARY]); - atomic_inc(&dev->counts[_DRM_STAT_DMA]); + atomic_add_unchecked(mc->used, &dev->counts[_DRM_STAT_SECONDARY]); + atomic_inc_unchecked(&dev->counts[_DRM_STAT_DMA]); sarea_priv->last_enqueue = dev_priv->counter - 1; sarea_priv->last_dispatch = (int)hw_status[5]; diff -urNp linux-2.6.32.48/drivers/gpu/drm/i810/i810_drv.h linux-2.6.32.48-openvz/drivers/gpu/drm/i810/i810_drv.h --- linux-2.6.32.48/drivers/gpu/drm/i810/i810_drv.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/i810/i810_drv.h 2011-11-17 20:08:53.000000000 -0500 @@ -108,8 +108,8 @@ typedef struct drm_i810_private { int page_flipping; wait_queue_head_t irq_queue; - atomic_t irq_received; - atomic_t irq_emitted; + atomic_unchecked_t irq_received; + atomic_unchecked_t irq_emitted; int front_offset; } drm_i810_private_t; diff -urNp linux-2.6.32.48/drivers/gpu/drm/i830/i830_drv.h linux-2.6.32.48-openvz/drivers/gpu/drm/i830/i830_drv.h --- linux-2.6.32.48/drivers/gpu/drm/i830/i830_drv.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/i830/i830_drv.h 2011-11-17 20:08:53.000000000 -0500 @@ -115,8 +115,8 @@ typedef struct drm_i830_private { int page_flipping; wait_queue_head_t irq_queue; - atomic_t irq_received; - atomic_t irq_emitted; + atomic_unchecked_t irq_received; + atomic_unchecked_t irq_emitted; int use_mi_batchbuffer_start; diff -urNp linux-2.6.32.48/drivers/gpu/drm/i830/i830_irq.c linux-2.6.32.48-openvz/drivers/gpu/drm/i830/i830_irq.c --- linux-2.6.32.48/drivers/gpu/drm/i830/i830_irq.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/i830/i830_irq.c 2011-11-17 20:08:53.000000000 -0500 @@ -47,7 +47,7 @@ irqreturn_t i830_driver_irq_handler(DRM_ I830_WRITE16(I830REG_INT_IDENTITY_R, temp); - atomic_inc(&dev_priv->irq_received); + atomic_inc_unchecked(&dev_priv->irq_received); wake_up_interruptible(&dev_priv->irq_queue); return IRQ_HANDLED; @@ -60,14 +60,14 @@ static int i830_emit_irq(struct drm_devi DRM_DEBUG("%s\n", __func__); - atomic_inc(&dev_priv->irq_emitted); + atomic_inc_unchecked(&dev_priv->irq_emitted); BEGIN_LP_RING(2); OUT_RING(0); OUT_RING(GFX_OP_USER_INTERRUPT); ADVANCE_LP_RING(); - return atomic_read(&dev_priv->irq_emitted); + return atomic_read_unchecked(&dev_priv->irq_emitted); } static int i830_wait_irq(struct drm_device * dev, int irq_nr) @@ -79,7 +79,7 @@ static int i830_wait_irq(struct drm_devi DRM_DEBUG("%s\n", __func__); - if (atomic_read(&dev_priv->irq_received) >= irq_nr) + if (atomic_read_unchecked(&dev_priv->irq_received) >= irq_nr) return 0; dev_priv->sarea_priv->perf_boxes |= I830_BOX_WAIT; @@ -88,7 +88,7 @@ static int i830_wait_irq(struct drm_devi for (;;) { __set_current_state(TASK_INTERRUPTIBLE); - if (atomic_read(&dev_priv->irq_received) >= irq_nr) + if (atomic_read_unchecked(&dev_priv->irq_received) >= irq_nr) break; if ((signed)(end - jiffies) <= 0) { DRM_ERROR("timeout iir %x imr %x ier %x hwstam %x\n", @@ -163,8 +163,8 @@ void i830_driver_irq_preinstall(struct d I830_WRITE16(I830REG_HWSTAM, 0xffff); I830_WRITE16(I830REG_INT_MASK_R, 0x0); I830_WRITE16(I830REG_INT_ENABLE_R, 0x0); - atomic_set(&dev_priv->irq_received, 0); - atomic_set(&dev_priv->irq_emitted, 0); + atomic_set_unchecked(&dev_priv->irq_received, 0); + atomic_set_unchecked(&dev_priv->irq_emitted, 0); init_waitqueue_head(&dev_priv->irq_queue); } diff -urNp linux-2.6.32.48/drivers/gpu/drm/i915/dvo_ch7017.c linux-2.6.32.48-openvz/drivers/gpu/drm/i915/dvo_ch7017.c --- linux-2.6.32.48/drivers/gpu/drm/i915/dvo_ch7017.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/i915/dvo_ch7017.c 2011-11-17 20:08:53.000000000 -0500 @@ -443,7 +443,7 @@ static void ch7017_destroy(struct intel_ } } -struct intel_dvo_dev_ops ch7017_ops = { +const struct intel_dvo_dev_ops ch7017_ops = { .init = ch7017_init, .detect = ch7017_detect, .mode_valid = ch7017_mode_valid, diff -urNp linux-2.6.32.48/drivers/gpu/drm/i915/dvo_ch7xxx.c linux-2.6.32.48-openvz/drivers/gpu/drm/i915/dvo_ch7xxx.c --- linux-2.6.32.48/drivers/gpu/drm/i915/dvo_ch7xxx.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/i915/dvo_ch7xxx.c 2011-11-17 20:08:53.000000000 -0500 @@ -356,7 +356,7 @@ static void ch7xxx_destroy(struct intel_ } } -struct intel_dvo_dev_ops ch7xxx_ops = { +const struct intel_dvo_dev_ops ch7xxx_ops = { .init = ch7xxx_init, .detect = ch7xxx_detect, .mode_valid = ch7xxx_mode_valid, diff -urNp linux-2.6.32.48/drivers/gpu/drm/i915/dvo.h linux-2.6.32.48-openvz/drivers/gpu/drm/i915/dvo.h --- linux-2.6.32.48/drivers/gpu/drm/i915/dvo.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/i915/dvo.h 2011-11-17 20:08:53.000000000 -0500 @@ -135,23 +135,23 @@ struct intel_dvo_dev_ops { * * \return singly-linked list of modes or NULL if no modes found. */ - struct drm_display_mode *(*get_modes)(struct intel_dvo_device *dvo); + struct drm_display_mode *(* const get_modes)(struct intel_dvo_device *dvo); /** * Clean up driver-specific bits of the output */ - void (*destroy) (struct intel_dvo_device *dvo); + void (* const destroy) (struct intel_dvo_device *dvo); /** * Debugging hook to dump device registers to log file */ - void (*dump_regs)(struct intel_dvo_device *dvo); + void (* const dump_regs)(struct intel_dvo_device *dvo); }; -extern struct intel_dvo_dev_ops sil164_ops; -extern struct intel_dvo_dev_ops ch7xxx_ops; -extern struct intel_dvo_dev_ops ivch_ops; -extern struct intel_dvo_dev_ops tfp410_ops; -extern struct intel_dvo_dev_ops ch7017_ops; +extern const struct intel_dvo_dev_ops sil164_ops; +extern const struct intel_dvo_dev_ops ch7xxx_ops; +extern const struct intel_dvo_dev_ops ivch_ops; +extern const struct intel_dvo_dev_ops tfp410_ops; +extern const struct intel_dvo_dev_ops ch7017_ops; #endif /* _INTEL_DVO_H */ diff -urNp linux-2.6.32.48/drivers/gpu/drm/i915/dvo_ivch.c linux-2.6.32.48-openvz/drivers/gpu/drm/i915/dvo_ivch.c --- linux-2.6.32.48/drivers/gpu/drm/i915/dvo_ivch.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/i915/dvo_ivch.c 2011-11-17 20:08:53.000000000 -0500 @@ -430,7 +430,7 @@ static void ivch_destroy(struct intel_dv } } -struct intel_dvo_dev_ops ivch_ops= { +const struct intel_dvo_dev_ops ivch_ops= { .init = ivch_init, .dpms = ivch_dpms, .save = ivch_save, diff -urNp linux-2.6.32.48/drivers/gpu/drm/i915/dvo_sil164.c linux-2.6.32.48-openvz/drivers/gpu/drm/i915/dvo_sil164.c --- linux-2.6.32.48/drivers/gpu/drm/i915/dvo_sil164.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/i915/dvo_sil164.c 2011-11-17 20:08:53.000000000 -0500 @@ -290,7 +290,7 @@ static void sil164_destroy(struct intel_ } } -struct intel_dvo_dev_ops sil164_ops = { +const struct intel_dvo_dev_ops sil164_ops = { .init = sil164_init, .detect = sil164_detect, .mode_valid = sil164_mode_valid, diff -urNp linux-2.6.32.48/drivers/gpu/drm/i915/dvo_tfp410.c linux-2.6.32.48-openvz/drivers/gpu/drm/i915/dvo_tfp410.c --- linux-2.6.32.48/drivers/gpu/drm/i915/dvo_tfp410.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/i915/dvo_tfp410.c 2011-11-17 20:08:53.000000000 -0500 @@ -323,7 +323,7 @@ static void tfp410_destroy(struct intel_ } } -struct intel_dvo_dev_ops tfp410_ops = { +const struct intel_dvo_dev_ops tfp410_ops = { .init = tfp410_init, .detect = tfp410_detect, .mode_valid = tfp410_mode_valid, diff -urNp linux-2.6.32.48/drivers/gpu/drm/i915/i915_debugfs.c linux-2.6.32.48-openvz/drivers/gpu/drm/i915/i915_debugfs.c --- linux-2.6.32.48/drivers/gpu/drm/i915/i915_debugfs.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/i915/i915_debugfs.c 2011-11-17 20:08:53.000000000 -0500 @@ -192,7 +192,7 @@ static int i915_interrupt_info(struct se I915_READ(GTIMR)); } seq_printf(m, "Interrupts received: %d\n", - atomic_read(&dev_priv->irq_received)); + atomic_read_unchecked(&dev_priv->irq_received)); if (dev_priv->hw_status_page != NULL) { seq_printf(m, "Current sequence: %d\n", i915_get_gem_seqno(dev)); diff -urNp linux-2.6.32.48/drivers/gpu/drm/i915/i915_drv.c linux-2.6.32.48-openvz/drivers/gpu/drm/i915/i915_drv.c --- linux-2.6.32.48/drivers/gpu/drm/i915/i915_drv.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/i915/i915_drv.c 2011-11-17 20:08:53.000000000 -0500 @@ -285,7 +285,7 @@ i915_pci_resume(struct pci_dev *pdev) return i915_resume(dev); } -static struct vm_operations_struct i915_gem_vm_ops = { +static const struct vm_operations_struct i915_gem_vm_ops = { .fault = i915_gem_fault, .open = drm_gem_vm_open, .close = drm_gem_vm_close, diff -urNp linux-2.6.32.48/drivers/gpu/drm/i915/i915_drv.h linux-2.6.32.48-openvz/drivers/gpu/drm/i915/i915_drv.h --- linux-2.6.32.48/drivers/gpu/drm/i915/i915_drv.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/i915/i915_drv.h 2011-11-17 20:09:04.000000000 -0500 @@ -168,7 +168,7 @@ struct drm_i915_display_funcs { /* display clock increase/decrease */ /* pll clock increase/decrease */ /* clock gating init */ -}; +} __no_const; typedef struct drm_i915_private { struct drm_device *dev; @@ -197,7 +197,7 @@ typedef struct drm_i915_private { int page_flipping; wait_queue_head_t irq_queue; - atomic_t irq_received; + atomic_unchecked_t irq_received; /** Protects user_irq_refcount and irq_mask_reg */ spinlock_t user_irq_lock; /** Refcount for i915_user_irq_get() versus i915_user_irq_put(). */ diff -urNp linux-2.6.32.48/drivers/gpu/drm/i915/i915_gem.c linux-2.6.32.48-openvz/drivers/gpu/drm/i915/i915_gem.c --- linux-2.6.32.48/drivers/gpu/drm/i915/i915_gem.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/i915/i915_gem.c 2011-11-17 20:08:53.000000000 -0500 @@ -102,7 +102,7 @@ i915_gem_get_aperture_ioctl(struct drm_d args->aper_size = dev->gtt_total; args->aper_available_size = (args->aper_size - - atomic_read(&dev->pin_memory)); + atomic_read_unchecked(&dev->pin_memory)); return 0; } @@ -492,6 +492,11 @@ i915_gem_pread_ioctl(struct drm_device * return -EINVAL; } + if (!access_ok(VERIFY_WRITE, (char __user *) (uintptr_t)args->data_ptr, args->size)) { + drm_gem_object_unreference(obj); + return -EFAULT; + } + if (i915_gem_object_needs_bit17_swizzle(obj)) { ret = i915_gem_shmem_pread_slow(dev, obj, args, file_priv); } else { @@ -965,6 +970,11 @@ i915_gem_pwrite_ioctl(struct drm_device return -EINVAL; } + if (!access_ok(VERIFY_READ, (char __user *) (uintptr_t)args->data_ptr, args->size)) { + drm_gem_object_unreference(obj); + return -EFAULT; + } + /* We can only do the GTT pwrite on untiled buffers, as otherwise * it would end up going through the fenced access, and we'll get * different detiling behavior between reading and writing. @@ -2054,7 +2064,7 @@ i915_gem_object_unbind(struct drm_gem_ob if (obj_priv->gtt_space) { atomic_dec(&dev->gtt_count); - atomic_sub(obj->size, &dev->gtt_memory); + atomic_sub_unchecked(obj->size, &dev->gtt_memory); drm_mm_put_block(obj_priv->gtt_space); obj_priv->gtt_space = NULL; @@ -2697,7 +2707,7 @@ i915_gem_object_bind_to_gtt(struct drm_g goto search_free; } atomic_inc(&dev->gtt_count); - atomic_add(obj->size, &dev->gtt_memory); + atomic_add_unchecked(obj->size, &dev->gtt_memory); /* Assert that the object is not currently in any GPU domain. As it * wasn't in the GTT, there shouldn't be any way it could have been in @@ -3751,9 +3761,9 @@ i915_gem_execbuffer(struct drm_device *d "%d/%d gtt bytes\n", atomic_read(&dev->object_count), atomic_read(&dev->pin_count), - atomic_read(&dev->object_memory), - atomic_read(&dev->pin_memory), - atomic_read(&dev->gtt_memory), + atomic_read_unchecked(&dev->object_memory), + atomic_read_unchecked(&dev->pin_memory), + atomic_read_unchecked(&dev->gtt_memory), dev->gtt_total); } goto err; @@ -3985,7 +3995,7 @@ i915_gem_object_pin(struct drm_gem_objec */ if (obj_priv->pin_count == 1) { atomic_inc(&dev->pin_count); - atomic_add(obj->size, &dev->pin_memory); + atomic_add_unchecked(obj->size, &dev->pin_memory); if (!obj_priv->active && (obj->write_domain & I915_GEM_GPU_DOMAINS) == 0 && !list_empty(&obj_priv->list)) @@ -4018,7 +4028,7 @@ i915_gem_object_unpin(struct drm_gem_obj list_move_tail(&obj_priv->list, &dev_priv->mm.inactive_list); atomic_dec(&dev->pin_count); - atomic_sub(obj->size, &dev->pin_memory); + atomic_sub_unchecked(obj->size, &dev->pin_memory); } i915_verify_inactive(dev, __FILE__, __LINE__); } diff -urNp linux-2.6.32.48/drivers/gpu/drm/i915/i915_irq.c linux-2.6.32.48-openvz/drivers/gpu/drm/i915/i915_irq.c --- linux-2.6.32.48/drivers/gpu/drm/i915/i915_irq.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/i915/i915_irq.c 2011-11-17 20:08:53.000000000 -0500 @@ -528,7 +528,7 @@ irqreturn_t i915_driver_irq_handler(DRM_ int irq_received; int ret = IRQ_NONE; - atomic_inc(&dev_priv->irq_received); + atomic_inc_unchecked(&dev_priv->irq_received); if (IS_IGDNG(dev)) return igdng_irq_handler(dev); @@ -1021,7 +1021,7 @@ void i915_driver_irq_preinstall(struct d { drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private; - atomic_set(&dev_priv->irq_received, 0); + atomic_set_unchecked(&dev_priv->irq_received, 0); INIT_WORK(&dev_priv->hotplug_work, i915_hotplug_work_func); INIT_WORK(&dev_priv->error_work, i915_error_work_func); diff -urNp linux-2.6.32.48/drivers/gpu/drm/i915/intel_sdvo.c linux-2.6.32.48-openvz/drivers/gpu/drm/i915/intel_sdvo.c --- linux-2.6.32.48/drivers/gpu/drm/i915/intel_sdvo.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/i915/intel_sdvo.c 2011-11-17 20:09:04.000000000 -0500 @@ -2795,7 +2795,9 @@ bool intel_sdvo_init(struct drm_device * sdvo_priv->slave_addr = intel_sdvo_get_slave_addr(dev, output_device); /* Save the bit-banging i2c functionality for use by the DDC wrapper */ - intel_sdvo_i2c_bit_algo.functionality = intel_output->i2c_bus->algo->functionality; + pax_open_kernel(); + *(void **)&intel_sdvo_i2c_bit_algo.functionality = intel_output->i2c_bus->algo->functionality; + pax_close_kernel(); /* Read the regs to test if we can talk to the device */ for (i = 0; i < 0x40; i++) { diff -urNp linux-2.6.32.48/drivers/gpu/drm/mga/mga_drv.h linux-2.6.32.48-openvz/drivers/gpu/drm/mga/mga_drv.h --- linux-2.6.32.48/drivers/gpu/drm/mga/mga_drv.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/mga/mga_drv.h 2011-11-17 20:08:53.000000000 -0500 @@ -120,9 +120,9 @@ typedef struct drm_mga_private { u32 clear_cmd; u32 maccess; - atomic_t vbl_received; /**< Number of vblanks received. */ + atomic_unchecked_t vbl_received; /**< Number of vblanks received. */ wait_queue_head_t fence_queue; - atomic_t last_fence_retired; + atomic_unchecked_t last_fence_retired; u32 next_fence_to_post; unsigned int fb_cpp; diff -urNp linux-2.6.32.48/drivers/gpu/drm/mga/mga_irq.c linux-2.6.32.48-openvz/drivers/gpu/drm/mga/mga_irq.c --- linux-2.6.32.48/drivers/gpu/drm/mga/mga_irq.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/mga/mga_irq.c 2011-11-17 20:08:53.000000000 -0500 @@ -44,7 +44,7 @@ u32 mga_get_vblank_counter(struct drm_de if (crtc != 0) return 0; - return atomic_read(&dev_priv->vbl_received); + return atomic_read_unchecked(&dev_priv->vbl_received); } @@ -60,7 +60,7 @@ irqreturn_t mga_driver_irq_handler(DRM_I /* VBLANK interrupt */ if (status & MGA_VLINEPEN) { MGA_WRITE(MGA_ICLEAR, MGA_VLINEICLR); - atomic_inc(&dev_priv->vbl_received); + atomic_inc_unchecked(&dev_priv->vbl_received); drm_handle_vblank(dev, 0); handled = 1; } @@ -80,7 +80,7 @@ irqreturn_t mga_driver_irq_handler(DRM_I MGA_WRITE(MGA_PRIMEND, prim_end); } - atomic_inc(&dev_priv->last_fence_retired); + atomic_inc_unchecked(&dev_priv->last_fence_retired); DRM_WAKEUP(&dev_priv->fence_queue); handled = 1; } @@ -131,7 +131,7 @@ int mga_driver_fence_wait(struct drm_dev * using fences. */ DRM_WAIT_ON(ret, dev_priv->fence_queue, 3 * DRM_HZ, - (((cur_fence = atomic_read(&dev_priv->last_fence_retired)) + (((cur_fence = atomic_read_unchecked(&dev_priv->last_fence_retired)) - *sequence) <= (1 << 23))); *sequence = cur_fence; diff -urNp linux-2.6.32.48/drivers/gpu/drm/r128/r128_cce.c linux-2.6.32.48-openvz/drivers/gpu/drm/r128/r128_cce.c --- linux-2.6.32.48/drivers/gpu/drm/r128/r128_cce.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/r128/r128_cce.c 2011-11-17 20:08:53.000000000 -0500 @@ -377,7 +377,7 @@ static int r128_do_init_cce(struct drm_d /* GH: Simple idle check. */ - atomic_set(&dev_priv->idle_count, 0); + atomic_set_unchecked(&dev_priv->idle_count, 0); /* We don't support anything other than bus-mastering ring mode, * but the ring can be in either AGP or PCI space for the ring diff -urNp linux-2.6.32.48/drivers/gpu/drm/r128/r128_drv.h linux-2.6.32.48-openvz/drivers/gpu/drm/r128/r128_drv.h --- linux-2.6.32.48/drivers/gpu/drm/r128/r128_drv.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/r128/r128_drv.h 2011-11-17 20:08:53.000000000 -0500 @@ -90,14 +90,14 @@ typedef struct drm_r128_private { int is_pci; unsigned long cce_buffers_offset; - atomic_t idle_count; + atomic_unchecked_t idle_count; int page_flipping; int current_page; u32 crtc_offset; u32 crtc_offset_cntl; - atomic_t vbl_received; + atomic_unchecked_t vbl_received; u32 color_fmt; unsigned int front_offset; diff -urNp linux-2.6.32.48/drivers/gpu/drm/r128/r128_irq.c linux-2.6.32.48-openvz/drivers/gpu/drm/r128/r128_irq.c --- linux-2.6.32.48/drivers/gpu/drm/r128/r128_irq.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/r128/r128_irq.c 2011-11-17 20:08:53.000000000 -0500 @@ -42,7 +42,7 @@ u32 r128_get_vblank_counter(struct drm_d if (crtc != 0) return 0; - return atomic_read(&dev_priv->vbl_received); + return atomic_read_unchecked(&dev_priv->vbl_received); } irqreturn_t r128_driver_irq_handler(DRM_IRQ_ARGS) @@ -56,7 +56,7 @@ irqreturn_t r128_driver_irq_handler(DRM_ /* VBLANK interrupt */ if (status & R128_CRTC_VBLANK_INT) { R128_WRITE(R128_GEN_INT_STATUS, R128_CRTC_VBLANK_INT_AK); - atomic_inc(&dev_priv->vbl_received); + atomic_inc_unchecked(&dev_priv->vbl_received); drm_handle_vblank(dev, 0); return IRQ_HANDLED; } diff -urNp linux-2.6.32.48/drivers/gpu/drm/r128/r128_state.c linux-2.6.32.48-openvz/drivers/gpu/drm/r128/r128_state.c --- linux-2.6.32.48/drivers/gpu/drm/r128/r128_state.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/r128/r128_state.c 2011-11-17 20:08:53.000000000 -0500 @@ -323,10 +323,10 @@ static void r128_clear_box(drm_r128_priv static void r128_cce_performance_boxes(drm_r128_private_t * dev_priv) { - if (atomic_read(&dev_priv->idle_count) == 0) { + if (atomic_read_unchecked(&dev_priv->idle_count) == 0) { r128_clear_box(dev_priv, 64, 4, 8, 8, 0, 255, 0); } else { - atomic_set(&dev_priv->idle_count, 0); + atomic_set_unchecked(&dev_priv->idle_count, 0); } } diff -urNp linux-2.6.32.48/drivers/gpu/drm/radeon/atom.c linux-2.6.32.48-openvz/drivers/gpu/drm/radeon/atom.c --- linux-2.6.32.48/drivers/gpu/drm/radeon/atom.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/radeon/atom.c 2011-11-17 20:08:53.000000000 -0500 @@ -1115,6 +1115,8 @@ struct atom_context *atom_parse(struct c char name[512]; int i; + pax_track_stack(); + ctx->card = card; ctx->bios = bios; diff -urNp linux-2.6.32.48/drivers/gpu/drm/radeon/mkregtable.c linux-2.6.32.48-openvz/drivers/gpu/drm/radeon/mkregtable.c --- linux-2.6.32.48/drivers/gpu/drm/radeon/mkregtable.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/radeon/mkregtable.c 2011-11-17 20:08:53.000000000 -0500 @@ -637,14 +637,14 @@ static int parser_auth(struct table *t, regex_t mask_rex; regmatch_t match[4]; char buf[1024]; - size_t end; + long end; int len; int done = 0; int r; unsigned o; struct offset *offset; char last_reg_s[10]; - int last_reg; + unsigned long last_reg; if (regcomp (&mask_rex, "(0x[0-9a-fA-F]*) *([_a-zA-Z0-9]*)", REG_EXTENDED)) { diff -urNp linux-2.6.32.48/drivers/gpu/drm/radeon/radeon_atombios.c linux-2.6.32.48-openvz/drivers/gpu/drm/radeon/radeon_atombios.c --- linux-2.6.32.48/drivers/gpu/drm/radeon/radeon_atombios.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/radeon/radeon_atombios.c 2011-11-17 20:08:53.000000000 -0500 @@ -275,6 +275,8 @@ bool radeon_get_atom_connector_info_from bool linkb; struct radeon_i2c_bus_rec ddc_bus; + pax_track_stack(); + atom_parse_data_header(ctx, index, &size, &frev, &crev, &data_offset); if (data_offset == 0) @@ -520,13 +522,13 @@ static uint16_t atombios_get_connector_o } } -struct bios_connector { +static struct bios_connector { bool valid; uint16_t line_mux; uint16_t devices; int connector_type; struct radeon_i2c_bus_rec ddc_bus; -}; +} bios_connectors[ATOM_MAX_SUPPORTED_DEVICE]; bool radeon_get_atom_connector_info_from_supported_devices_table(struct drm_device @@ -542,7 +544,6 @@ bool radeon_get_atom_connector_info_from uint8_t dac; union atom_supported_devices *supported_devices; int i, j; - struct bios_connector bios_connectors[ATOM_MAX_SUPPORTED_DEVICE]; atom_parse_data_header(ctx, index, &size, &frev, &crev, &data_offset); diff -urNp linux-2.6.32.48/drivers/gpu/drm/radeon/radeon_display.c linux-2.6.32.48-openvz/drivers/gpu/drm/radeon/radeon_display.c --- linux-2.6.32.48/drivers/gpu/drm/radeon/radeon_display.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/radeon/radeon_display.c 2011-11-17 20:08:53.000000000 -0500 @@ -482,7 +482,7 @@ void radeon_compute_pll(struct radeon_pl if (flags & RADEON_PLL_PREFER_CLOSEST_LOWER) { error = freq - current_freq; - error = error < 0 ? 0xffffffff : error; + error = (int32_t)error < 0 ? 0xffffffff : error; } else error = abs(current_freq - freq); vco_diff = abs(vco - best_vco); diff -urNp linux-2.6.32.48/drivers/gpu/drm/radeon/radeon_drv.h linux-2.6.32.48-openvz/drivers/gpu/drm/radeon/radeon_drv.h --- linux-2.6.32.48/drivers/gpu/drm/radeon/radeon_drv.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/radeon/radeon_drv.h 2011-11-17 20:08:53.000000000 -0500 @@ -253,7 +253,7 @@ typedef struct drm_radeon_private { /* SW interrupt */ wait_queue_head_t swi_queue; - atomic_t swi_emitted; + atomic_unchecked_t swi_emitted; int vblank_crtc; uint32_t irq_enable_reg; uint32_t r500_disp_irq_reg; diff -urNp linux-2.6.32.48/drivers/gpu/drm/radeon/radeon_fence.c linux-2.6.32.48-openvz/drivers/gpu/drm/radeon/radeon_fence.c --- linux-2.6.32.48/drivers/gpu/drm/radeon/radeon_fence.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/radeon/radeon_fence.c 2011-11-17 20:08:53.000000000 -0500 @@ -47,7 +47,7 @@ int radeon_fence_emit(struct radeon_devi write_unlock_irqrestore(&rdev->fence_drv.lock, irq_flags); return 0; } - fence->seq = atomic_add_return(1, &rdev->fence_drv.seq); + fence->seq = atomic_add_return_unchecked(1, &rdev->fence_drv.seq); if (!rdev->cp.ready) { /* FIXME: cp is not running assume everythings is done right * away @@ -364,7 +364,7 @@ int radeon_fence_driver_init(struct rade return r; } WREG32(rdev->fence_drv.scratch_reg, 0); - atomic_set(&rdev->fence_drv.seq, 0); + atomic_set_unchecked(&rdev->fence_drv.seq, 0); INIT_LIST_HEAD(&rdev->fence_drv.created); INIT_LIST_HEAD(&rdev->fence_drv.emited); INIT_LIST_HEAD(&rdev->fence_drv.signaled); diff -urNp linux-2.6.32.48/drivers/gpu/drm/radeon/radeon.h linux-2.6.32.48-openvz/drivers/gpu/drm/radeon/radeon.h --- linux-2.6.32.48/drivers/gpu/drm/radeon/radeon.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/radeon/radeon.h 2011-11-17 20:09:04.000000000 -0500 @@ -149,7 +149,7 @@ int radeon_pm_init(struct radeon_device */ struct radeon_fence_driver { uint32_t scratch_reg; - atomic_t seq; + atomic_unchecked_t seq; uint32_t last_seq; unsigned long count_timeout; wait_queue_head_t queue; @@ -640,7 +640,7 @@ struct radeon_asic { uint32_t offset, uint32_t obj_size); int (*clear_surface_reg)(struct radeon_device *rdev, int reg); void (*bandwidth_update)(struct radeon_device *rdev); -}; +} __no_const; /* * Asic structures diff -urNp linux-2.6.32.48/drivers/gpu/drm/radeon/radeon_ioc32.c linux-2.6.32.48-openvz/drivers/gpu/drm/radeon/radeon_ioc32.c --- linux-2.6.32.48/drivers/gpu/drm/radeon/radeon_ioc32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/radeon/radeon_ioc32.c 2011-11-17 20:08:53.000000000 -0500 @@ -368,7 +368,7 @@ static int compat_radeon_cp_setparam(str request = compat_alloc_user_space(sizeof(*request)); if (!access_ok(VERIFY_WRITE, request, sizeof(*request)) || __put_user(req32.param, &request->param) - || __put_user((void __user *)(unsigned long)req32.value, + || __put_user((unsigned long)req32.value, &request->value)) return -EFAULT; diff -urNp linux-2.6.32.48/drivers/gpu/drm/radeon/radeon_irq.c linux-2.6.32.48-openvz/drivers/gpu/drm/radeon/radeon_irq.c --- linux-2.6.32.48/drivers/gpu/drm/radeon/radeon_irq.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/radeon/radeon_irq.c 2011-11-17 20:08:53.000000000 -0500 @@ -225,8 +225,8 @@ static int radeon_emit_irq(struct drm_de unsigned int ret; RING_LOCALS; - atomic_inc(&dev_priv->swi_emitted); - ret = atomic_read(&dev_priv->swi_emitted); + atomic_inc_unchecked(&dev_priv->swi_emitted); + ret = atomic_read_unchecked(&dev_priv->swi_emitted); BEGIN_RING(4); OUT_RING_REG(RADEON_LAST_SWI_REG, ret); @@ -352,7 +352,7 @@ int radeon_driver_irq_postinstall(struct drm_radeon_private_t *dev_priv = (drm_radeon_private_t *) dev->dev_private; - atomic_set(&dev_priv->swi_emitted, 0); + atomic_set_unchecked(&dev_priv->swi_emitted, 0); DRM_INIT_WAITQUEUE(&dev_priv->swi_queue); dev->max_vblank_count = 0x001fffff; diff -urNp linux-2.6.32.48/drivers/gpu/drm/radeon/radeon_state.c linux-2.6.32.48-openvz/drivers/gpu/drm/radeon/radeon_state.c --- linux-2.6.32.48/drivers/gpu/drm/radeon/radeon_state.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/radeon/radeon_state.c 2011-11-17 20:08:53.000000000 -0500 @@ -3021,7 +3021,7 @@ static int radeon_cp_getparam(struct drm { drm_radeon_private_t *dev_priv = dev->dev_private; drm_radeon_getparam_t *param = data; - int value; + int value = 0; DRM_DEBUG("pid=%d\n", DRM_CURRENTPID); diff -urNp linux-2.6.32.48/drivers/gpu/drm/radeon/radeon_ttm.c linux-2.6.32.48-openvz/drivers/gpu/drm/radeon/radeon_ttm.c --- linux-2.6.32.48/drivers/gpu/drm/radeon/radeon_ttm.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/radeon/radeon_ttm.c 2011-11-17 20:08:53.000000000 -0500 @@ -535,27 +535,10 @@ void radeon_ttm_fini(struct radeon_devic DRM_INFO("radeon: ttm finalized\n"); } -static struct vm_operations_struct radeon_ttm_vm_ops; -static const struct vm_operations_struct *ttm_vm_ops = NULL; - -static int radeon_ttm_fault(struct vm_area_struct *vma, struct vm_fault *vmf) -{ - struct ttm_buffer_object *bo; - int r; - - bo = (struct ttm_buffer_object *)vma->vm_private_data; - if (bo == NULL) { - return VM_FAULT_NOPAGE; - } - r = ttm_vm_ops->fault(vma, vmf); - return r; -} - int radeon_mmap(struct file *filp, struct vm_area_struct *vma) { struct drm_file *file_priv; struct radeon_device *rdev; - int r; if (unlikely(vma->vm_pgoff < DRM_FILE_PAGE_OFFSET)) { return drm_mmap(filp, vma); @@ -563,20 +546,9 @@ int radeon_mmap(struct file *filp, struc file_priv = (struct drm_file *)filp->private_data; rdev = file_priv->minor->dev->dev_private; - if (rdev == NULL) { + if (!rdev) return -EINVAL; - } - r = ttm_bo_mmap(filp, vma, &rdev->mman.bdev); - if (unlikely(r != 0)) { - return r; - } - if (unlikely(ttm_vm_ops == NULL)) { - ttm_vm_ops = vma->vm_ops; - radeon_ttm_vm_ops = *ttm_vm_ops; - radeon_ttm_vm_ops.fault = &radeon_ttm_fault; - } - vma->vm_ops = &radeon_ttm_vm_ops; - return 0; + return ttm_bo_mmap(filp, vma, &rdev->mman.bdev); } diff -urNp linux-2.6.32.48/drivers/gpu/drm/radeon/rs690.c linux-2.6.32.48-openvz/drivers/gpu/drm/radeon/rs690.c --- linux-2.6.32.48/drivers/gpu/drm/radeon/rs690.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/radeon/rs690.c 2011-11-17 20:08:53.000000000 -0500 @@ -302,9 +302,11 @@ void rs690_crtc_bandwidth_compute(struct if (rdev->pm.max_bandwidth.full > rdev->pm.sideport_bandwidth.full && rdev->pm.sideport_bandwidth.full) rdev->pm.max_bandwidth = rdev->pm.sideport_bandwidth; - read_delay_latency.full = rfixed_const(370 * 800 * 1000); + read_delay_latency.full = rfixed_const(800 * 1000); read_delay_latency.full = rfixed_div(read_delay_latency, rdev->pm.igp_sideport_mclk); + a.full = rfixed_const(370); + read_delay_latency.full = rfixed_mul(read_delay_latency, a); } else { if (rdev->pm.max_bandwidth.full > rdev->pm.k8_bandwidth.full && rdev->pm.k8_bandwidth.full) diff -urNp linux-2.6.32.48/drivers/gpu/drm/ttm/ttm_bo.c linux-2.6.32.48-openvz/drivers/gpu/drm/ttm/ttm_bo.c --- linux-2.6.32.48/drivers/gpu/drm/ttm/ttm_bo.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/ttm/ttm_bo.c 2011-11-17 20:09:04.000000000 -0500 @@ -67,7 +67,7 @@ static struct attribute *ttm_bo_global_a NULL }; -static struct sysfs_ops ttm_bo_global_ops = { +static const struct sysfs_ops ttm_bo_global_ops = { .show = &ttm_bo_global_show }; diff -urNp linux-2.6.32.48/drivers/gpu/drm/ttm/ttm_bo_vm.c linux-2.6.32.48-openvz/drivers/gpu/drm/ttm/ttm_bo_vm.c --- linux-2.6.32.48/drivers/gpu/drm/ttm/ttm_bo_vm.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/ttm/ttm_bo_vm.c 2011-11-17 20:08:53.000000000 -0500 @@ -73,7 +73,7 @@ static int ttm_bo_vm_fault(struct vm_are { struct ttm_buffer_object *bo = (struct ttm_buffer_object *) vma->vm_private_data; - struct ttm_bo_device *bdev = bo->bdev; + struct ttm_bo_device *bdev; unsigned long bus_base; unsigned long bus_offset; unsigned long bus_size; @@ -88,6 +88,10 @@ static int ttm_bo_vm_fault(struct vm_are unsigned long address = (unsigned long)vmf->virtual_address; int retval = VM_FAULT_NOPAGE; + if (!bo) + return VM_FAULT_NOPAGE; + bdev = bo->bdev; + /* * Work around locking order reversal in fault / nopfn * between mmap_sem and bo_reserve: Perform a trylock operation diff -urNp linux-2.6.32.48/drivers/gpu/drm/ttm/ttm_global.c linux-2.6.32.48-openvz/drivers/gpu/drm/ttm/ttm_global.c --- linux-2.6.32.48/drivers/gpu/drm/ttm/ttm_global.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/ttm/ttm_global.c 2011-11-17 20:08:53.000000000 -0500 @@ -36,7 +36,7 @@ struct ttm_global_item { struct mutex mutex; void *object; - int refcount; + atomic_t refcount; }; static struct ttm_global_item glob[TTM_GLOBAL_NUM]; @@ -49,7 +49,7 @@ void ttm_global_init(void) struct ttm_global_item *item = &glob[i]; mutex_init(&item->mutex); item->object = NULL; - item->refcount = 0; + atomic_set(&item->refcount, 0); } } @@ -59,7 +59,7 @@ void ttm_global_release(void) for (i = 0; i < TTM_GLOBAL_NUM; ++i) { struct ttm_global_item *item = &glob[i]; BUG_ON(item->object != NULL); - BUG_ON(item->refcount != 0); + BUG_ON(atomic_read(&item->refcount) != 0); } } @@ -70,7 +70,7 @@ int ttm_global_item_ref(struct ttm_globa void *object; mutex_lock(&item->mutex); - if (item->refcount == 0) { + if (atomic_read(&item->refcount) == 0) { item->object = kzalloc(ref->size, GFP_KERNEL); if (unlikely(item->object == NULL)) { ret = -ENOMEM; @@ -83,7 +83,7 @@ int ttm_global_item_ref(struct ttm_globa goto out_err; } - ++item->refcount; + atomic_inc(&item->refcount); ref->object = item->object; object = item->object; mutex_unlock(&item->mutex); @@ -100,9 +100,9 @@ void ttm_global_item_unref(struct ttm_gl struct ttm_global_item *item = &glob[ref->global_type]; mutex_lock(&item->mutex); - BUG_ON(item->refcount == 0); + BUG_ON(atomic_read(&item->refcount) == 0); BUG_ON(ref->object != item->object); - if (--item->refcount == 0) { + if (atomic_dec_and_test(&item->refcount)) { ref->release(ref); item->object = NULL; } diff -urNp linux-2.6.32.48/drivers/gpu/drm/ttm/ttm_memory.c linux-2.6.32.48-openvz/drivers/gpu/drm/ttm/ttm_memory.c --- linux-2.6.32.48/drivers/gpu/drm/ttm/ttm_memory.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/ttm/ttm_memory.c 2011-11-17 20:08:53.000000000 -0500 @@ -152,7 +152,7 @@ static struct attribute *ttm_mem_zone_at NULL }; -static struct sysfs_ops ttm_mem_zone_ops = { +static const struct sysfs_ops ttm_mem_zone_ops = { .show = &ttm_mem_zone_show, .store = &ttm_mem_zone_store }; diff -urNp linux-2.6.32.48/drivers/gpu/drm/via/via_drv.h linux-2.6.32.48-openvz/drivers/gpu/drm/via/via_drv.h --- linux-2.6.32.48/drivers/gpu/drm/via/via_drv.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/via/via_drv.h 2011-11-17 20:08:53.000000000 -0500 @@ -51,7 +51,7 @@ typedef struct drm_via_ring_buffer { typedef uint32_t maskarray_t[5]; typedef struct drm_via_irq { - atomic_t irq_received; + atomic_unchecked_t irq_received; uint32_t pending_mask; uint32_t enable_mask; wait_queue_head_t irq_queue; @@ -75,7 +75,7 @@ typedef struct drm_via_private { struct timeval last_vblank; int last_vblank_valid; unsigned usec_per_vblank; - atomic_t vbl_received; + atomic_unchecked_t vbl_received; drm_via_state_t hc_state; char pci_buf[VIA_PCI_BUF_SIZE]; const uint32_t *fire_offsets[VIA_FIRE_BUF_SIZE]; diff -urNp linux-2.6.32.48/drivers/gpu/drm/via/via_irq.c linux-2.6.32.48-openvz/drivers/gpu/drm/via/via_irq.c --- linux-2.6.32.48/drivers/gpu/drm/via/via_irq.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/gpu/drm/via/via_irq.c 2011-11-17 20:08:53.000000000 -0500 @@ -102,7 +102,7 @@ u32 via_get_vblank_counter(struct drm_de if (crtc != 0) return 0; - return atomic_read(&dev_priv->vbl_received); + return atomic_read_unchecked(&dev_priv->vbl_received); } irqreturn_t via_driver_irq_handler(DRM_IRQ_ARGS) @@ -117,8 +117,8 @@ irqreturn_t via_driver_irq_handler(DRM_I status = VIA_READ(VIA_REG_INTERRUPT); if (status & VIA_IRQ_VBLANK_PENDING) { - atomic_inc(&dev_priv->vbl_received); - if (!(atomic_read(&dev_priv->vbl_received) & 0x0F)) { + atomic_inc_unchecked(&dev_priv->vbl_received); + if (!(atomic_read_unchecked(&dev_priv->vbl_received) & 0x0F)) { do_gettimeofday(&cur_vblank); if (dev_priv->last_vblank_valid) { dev_priv->usec_per_vblank = @@ -128,7 +128,7 @@ irqreturn_t via_driver_irq_handler(DRM_I dev_priv->last_vblank = cur_vblank; dev_priv->last_vblank_valid = 1; } - if (!(atomic_read(&dev_priv->vbl_received) & 0xFF)) { + if (!(atomic_read_unchecked(&dev_priv->vbl_received) & 0xFF)) { DRM_DEBUG("US per vblank is: %u\n", dev_priv->usec_per_vblank); } @@ -138,7 +138,7 @@ irqreturn_t via_driver_irq_handler(DRM_I for (i = 0; i < dev_priv->num_irqs; ++i) { if (status & cur_irq->pending_mask) { - atomic_inc(&cur_irq->irq_received); + atomic_inc_unchecked(&cur_irq->irq_received); DRM_WAKEUP(&cur_irq->irq_queue); handled = 1; if (dev_priv->irq_map[drm_via_irq_dma0_td] == i) { @@ -244,11 +244,11 @@ via_driver_irq_wait(struct drm_device * DRM_WAIT_ON(ret, cur_irq->irq_queue, 3 * DRM_HZ, ((VIA_READ(masks[irq][2]) & masks[irq][3]) == masks[irq][4])); - cur_irq_sequence = atomic_read(&cur_irq->irq_received); + cur_irq_sequence = atomic_read_unchecked(&cur_irq->irq_received); } else { DRM_WAIT_ON(ret, cur_irq->irq_queue, 3 * DRM_HZ, (((cur_irq_sequence = - atomic_read(&cur_irq->irq_received)) - + atomic_read_unchecked(&cur_irq->irq_received)) - *sequence) <= (1 << 23))); } *sequence = cur_irq_sequence; @@ -286,7 +286,7 @@ void via_driver_irq_preinstall(struct dr } for (i = 0; i < dev_priv->num_irqs; ++i) { - atomic_set(&cur_irq->irq_received, 0); + atomic_set_unchecked(&cur_irq->irq_received, 0); cur_irq->enable_mask = dev_priv->irq_masks[i][0]; cur_irq->pending_mask = dev_priv->irq_masks[i][1]; DRM_INIT_WAITQUEUE(&cur_irq->irq_queue); @@ -368,7 +368,7 @@ int via_wait_irq(struct drm_device *dev, switch (irqwait->request.type & ~VIA_IRQ_FLAGS_MASK) { case VIA_IRQ_RELATIVE: irqwait->request.sequence += - atomic_read(&cur_irq->irq_received); + atomic_read_unchecked(&cur_irq->irq_received); irqwait->request.type &= ~_DRM_VBLANK_RELATIVE; case VIA_IRQ_ABSOLUTE: break; diff -urNp linux-2.6.32.48/drivers/hid/hid-core.c linux-2.6.32.48-openvz/drivers/hid/hid-core.c --- linux-2.6.32.48/drivers/hid/hid-core.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/hid/hid-core.c 2011-11-17 20:08:53.000000000 -0500 @@ -1752,7 +1752,7 @@ static bool hid_ignore(struct hid_device int hid_add_device(struct hid_device *hdev) { - static atomic_t id = ATOMIC_INIT(0); + static atomic_unchecked_t id = ATOMIC_INIT(0); int ret; if (WARN_ON(hdev->status & HID_STAT_ADDED)) @@ -1766,7 +1766,7 @@ int hid_add_device(struct hid_device *hd /* XXX hack, any other cleaner solution after the driver core * is converted to allow more than 20 bytes as the device name? */ dev_set_name(&hdev->dev, "%04X:%04X:%04X.%04X", hdev->bus, - hdev->vendor, hdev->product, atomic_inc_return(&id)); + hdev->vendor, hdev->product, atomic_inc_return_unchecked(&id)); ret = device_add(&hdev->dev); if (!ret) diff -urNp linux-2.6.32.48/drivers/hid/usbhid/hiddev.c linux-2.6.32.48-openvz/drivers/hid/usbhid/hiddev.c --- linux-2.6.32.48/drivers/hid/usbhid/hiddev.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/hid/usbhid/hiddev.c 2011-11-17 20:08:53.000000000 -0500 @@ -617,7 +617,7 @@ static long hiddev_ioctl(struct file *fi return put_user(HID_VERSION, (int __user *)arg); case HIDIOCAPPLICATION: - if (arg < 0 || arg >= hid->maxapplication) + if (arg >= hid->maxapplication) return -EINVAL; for (i = 0; i < hid->maxcollection; i++) diff -urNp linux-2.6.32.48/drivers/hwmon/lis3lv02d.c linux-2.6.32.48-openvz/drivers/hwmon/lis3lv02d.c --- linux-2.6.32.48/drivers/hwmon/lis3lv02d.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/hwmon/lis3lv02d.c 2011-11-17 20:08:53.000000000 -0500 @@ -146,7 +146,7 @@ static irqreturn_t lis302dl_interrupt(in * the lid is closed. This leads to interrupts as soon as a little move * is done. */ - atomic_inc(&lis3_dev.count); + atomic_inc_unchecked(&lis3_dev.count); wake_up_interruptible(&lis3_dev.misc_wait); kill_fasync(&lis3_dev.async_queue, SIGIO, POLL_IN); @@ -160,7 +160,7 @@ static int lis3lv02d_misc_open(struct in if (test_and_set_bit(0, &lis3_dev.misc_opened)) return -EBUSY; /* already open */ - atomic_set(&lis3_dev.count, 0); + atomic_set_unchecked(&lis3_dev.count, 0); /* * The sensor can generate interrupts for free-fall and direction @@ -206,7 +206,7 @@ static ssize_t lis3lv02d_misc_read(struc add_wait_queue(&lis3_dev.misc_wait, &wait); while (true) { set_current_state(TASK_INTERRUPTIBLE); - data = atomic_xchg(&lis3_dev.count, 0); + data = atomic_xchg_unchecked(&lis3_dev.count, 0); if (data) break; @@ -244,7 +244,7 @@ out: static unsigned int lis3lv02d_misc_poll(struct file *file, poll_table *wait) { poll_wait(file, &lis3_dev.misc_wait, wait); - if (atomic_read(&lis3_dev.count)) + if (atomic_read_unchecked(&lis3_dev.count)) return POLLIN | POLLRDNORM; return 0; } diff -urNp linux-2.6.32.48/drivers/hwmon/lis3lv02d.h linux-2.6.32.48-openvz/drivers/hwmon/lis3lv02d.h --- linux-2.6.32.48/drivers/hwmon/lis3lv02d.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/hwmon/lis3lv02d.h 2011-11-17 20:08:53.000000000 -0500 @@ -201,7 +201,7 @@ struct lis3lv02d { struct input_polled_dev *idev; /* input device */ struct platform_device *pdev; /* platform device */ - atomic_t count; /* interrupt count after last read */ + atomic_unchecked_t count; /* interrupt count after last read */ int xcalib; /* calibrated null value for x */ int ycalib; /* calibrated null value for y */ int zcalib; /* calibrated null value for z */ diff -urNp linux-2.6.32.48/drivers/hwmon/sht15.c linux-2.6.32.48-openvz/drivers/hwmon/sht15.c --- linux-2.6.32.48/drivers/hwmon/sht15.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/hwmon/sht15.c 2011-11-17 20:08:53.000000000 -0500 @@ -112,7 +112,7 @@ struct sht15_data { int supply_uV; int supply_uV_valid; struct work_struct update_supply_work; - atomic_t interrupt_handled; + atomic_unchecked_t interrupt_handled; }; /** @@ -245,13 +245,13 @@ static inline int sht15_update_single_va return ret; gpio_direction_input(data->pdata->gpio_data); - atomic_set(&data->interrupt_handled, 0); + atomic_set_unchecked(&data->interrupt_handled, 0); enable_irq(gpio_to_irq(data->pdata->gpio_data)); if (gpio_get_value(data->pdata->gpio_data) == 0) { disable_irq_nosync(gpio_to_irq(data->pdata->gpio_data)); /* Only relevant if the interrupt hasn't occured. */ - if (!atomic_read(&data->interrupt_handled)) + if (!atomic_read_unchecked(&data->interrupt_handled)) schedule_work(&data->read_work); } ret = wait_event_timeout(data->wait_queue, @@ -398,7 +398,7 @@ static irqreturn_t sht15_interrupt_fired struct sht15_data *data = d; /* First disable the interrupt */ disable_irq_nosync(irq); - atomic_inc(&data->interrupt_handled); + atomic_inc_unchecked(&data->interrupt_handled); /* Then schedule a reading work struct */ if (data->flag != SHT15_READING_NOTHING) schedule_work(&data->read_work); @@ -449,11 +449,11 @@ static void sht15_bh_read_data(struct wo here as could have gone low in meantime so verify it hasn't! */ - atomic_set(&data->interrupt_handled, 0); + atomic_set_unchecked(&data->interrupt_handled, 0); enable_irq(gpio_to_irq(data->pdata->gpio_data)); /* If still not occured or another handler has been scheduled */ if (gpio_get_value(data->pdata->gpio_data) - || atomic_read(&data->interrupt_handled)) + || atomic_read_unchecked(&data->interrupt_handled)) return; } /* Read the data back from the device */ diff -urNp linux-2.6.32.48/drivers/hwmon/w83791d.c linux-2.6.32.48-openvz/drivers/hwmon/w83791d.c --- linux-2.6.32.48/drivers/hwmon/w83791d.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/hwmon/w83791d.c 2011-11-17 20:08:53.000000000 -0500 @@ -330,8 +330,8 @@ static int w83791d_detect(struct i2c_cli struct i2c_board_info *info); static int w83791d_remove(struct i2c_client *client); -static int w83791d_read(struct i2c_client *client, u8 register); -static int w83791d_write(struct i2c_client *client, u8 register, u8 value); +static int w83791d_read(struct i2c_client *client, u8 reg); +static int w83791d_write(struct i2c_client *client, u8 reg, u8 value); static struct w83791d_data *w83791d_update_device(struct device *dev); #ifdef DEBUG diff -urNp linux-2.6.32.48/drivers/i2c/busses/i2c-amd756-s4882.c linux-2.6.32.48-openvz/drivers/i2c/busses/i2c-amd756-s4882.c --- linux-2.6.32.48/drivers/i2c/busses/i2c-amd756-s4882.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/i2c/busses/i2c-amd756-s4882.c 2011-11-17 20:09:04.000000000 -0500 @@ -43,7 +43,7 @@ extern struct i2c_adapter amd756_smbus; static struct i2c_adapter *s4882_adapter; -static struct i2c_algorithm *s4882_algo; +static i2c_algorithm_no_const *s4882_algo; /* Wrapper access functions for multiplexed SMBus */ static DEFINE_MUTEX(amd756_lock); diff -urNp linux-2.6.32.48/drivers/i2c/busses/i2c-nforce2-s4985.c linux-2.6.32.48-openvz/drivers/i2c/busses/i2c-nforce2-s4985.c --- linux-2.6.32.48/drivers/i2c/busses/i2c-nforce2-s4985.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/i2c/busses/i2c-nforce2-s4985.c 2011-11-17 20:09:04.000000000 -0500 @@ -41,7 +41,7 @@ extern struct i2c_adapter *nforce2_smbus; static struct i2c_adapter *s4985_adapter; -static struct i2c_algorithm *s4985_algo; +static i2c_algorithm_no_const *s4985_algo; /* Wrapper access functions for multiplexed SMBus */ static DEFINE_MUTEX(nforce2_lock); diff -urNp linux-2.6.32.48/drivers/ide/ide-cd.c linux-2.6.32.48-openvz/drivers/ide/ide-cd.c --- linux-2.6.32.48/drivers/ide/ide-cd.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ide/ide-cd.c 2011-11-17 20:08:53.000000000 -0500 @@ -774,7 +774,7 @@ static void cdrom_do_block_pc(ide_drive_ alignment = queue_dma_alignment(q) | q->dma_pad_mask; if ((unsigned long)buf & alignment || blk_rq_bytes(rq) & q->dma_pad_mask - || object_is_on_stack(buf)) + || object_starts_on_stack(buf)) drive->dma = 0; } } diff -urNp linux-2.6.32.48/drivers/ide/ide-floppy.c linux-2.6.32.48-openvz/drivers/ide/ide-floppy.c --- linux-2.6.32.48/drivers/ide/ide-floppy.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ide/ide-floppy.c 2011-11-17 20:08:53.000000000 -0500 @@ -373,6 +373,8 @@ static int ide_floppy_get_capacity(ide_d u8 pc_buf[256], header_len, desc_cnt; int i, rc = 1, blocks, length; + pax_track_stack(); + ide_debug_log(IDE_DBG_FUNC, "enter"); drive->bios_cyl = 0; diff -urNp linux-2.6.32.48/drivers/ide/setup-pci.c linux-2.6.32.48-openvz/drivers/ide/setup-pci.c --- linux-2.6.32.48/drivers/ide/setup-pci.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ide/setup-pci.c 2011-11-17 20:08:53.000000000 -0500 @@ -542,6 +542,8 @@ int ide_pci_init_two(struct pci_dev *dev int ret, i, n_ports = dev2 ? 4 : 2; struct ide_hw hw[4], *hws[] = { NULL, NULL, NULL, NULL }; + pax_track_stack(); + for (i = 0; i < n_ports / 2; i++) { ret = ide_setup_pci_controller(pdev[i], d, !i); if (ret < 0) diff -urNp linux-2.6.32.48/drivers/ieee1394/dv1394.c linux-2.6.32.48-openvz/drivers/ieee1394/dv1394.c --- linux-2.6.32.48/drivers/ieee1394/dv1394.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ieee1394/dv1394.c 2011-11-17 20:08:53.000000000 -0500 @@ -739,7 +739,7 @@ static void frame_prepare(struct video_c based upon DIF section and sequence */ -static void inline +static inline void frame_put_packet (struct frame *f, struct packet *p) { int section_type = p->data[0] >> 5; /* section type is in bits 5 - 7 */ diff -urNp linux-2.6.32.48/drivers/ieee1394/hosts.c linux-2.6.32.48-openvz/drivers/ieee1394/hosts.c --- linux-2.6.32.48/drivers/ieee1394/hosts.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ieee1394/hosts.c 2011-11-17 20:08:53.000000000 -0500 @@ -78,6 +78,7 @@ static int dummy_isoctl(struct hpsb_iso } static struct hpsb_host_driver dummy_driver = { + .name = "dummy", .transmit_packet = dummy_transmit_packet, .devctl = dummy_devctl, .isoctl = dummy_isoctl diff -urNp linux-2.6.32.48/drivers/ieee1394/init_ohci1394_dma.c linux-2.6.32.48-openvz/drivers/ieee1394/init_ohci1394_dma.c --- linux-2.6.32.48/drivers/ieee1394/init_ohci1394_dma.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ieee1394/init_ohci1394_dma.c 2011-11-17 20:08:53.000000000 -0500 @@ -257,7 +257,7 @@ void __init init_ohci1394_dma_on_all_con for (func = 0; func < 8; func++) { u32 class = read_pci_config(num,slot,func, PCI_CLASS_REVISION); - if ((class == 0xffffffff)) + if (class == 0xffffffff) continue; /* No device at this func */ if (class>>8 != PCI_CLASS_SERIAL_FIREWIRE_OHCI) diff -urNp linux-2.6.32.48/drivers/ieee1394/ohci1394.c linux-2.6.32.48-openvz/drivers/ieee1394/ohci1394.c --- linux-2.6.32.48/drivers/ieee1394/ohci1394.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ieee1394/ohci1394.c 2011-11-17 20:08:53.000000000 -0500 @@ -147,9 +147,9 @@ printk(level "%s: " fmt "\n" , OHCI1394_ printk(level "%s: fw-host%d: " fmt "\n" , OHCI1394_DRIVER_NAME, ohci->host->id , ## args) /* Module Parameters */ -static int phys_dma = 1; +static int phys_dma; module_param(phys_dma, int, 0444); -MODULE_PARM_DESC(phys_dma, "Enable physical DMA (default = 1)."); +MODULE_PARM_DESC(phys_dma, "Enable physical DMA (default = 0)."); static void dma_trm_tasklet(unsigned long data); static void dma_trm_reset(struct dma_trm_ctx *d); diff -urNp linux-2.6.32.48/drivers/ieee1394/sbp2.c linux-2.6.32.48-openvz/drivers/ieee1394/sbp2.c --- linux-2.6.32.48/drivers/ieee1394/sbp2.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/ieee1394/sbp2.c 2011-11-17 20:08:53.000000000 -0500 @@ -2111,7 +2111,7 @@ MODULE_DESCRIPTION("IEEE-1394 SBP-2 prot MODULE_SUPPORTED_DEVICE(SBP2_DEVICE_NAME); MODULE_LICENSE("GPL"); -static int sbp2_module_init(void) +static int __init sbp2_module_init(void) { int ret; diff -urNp linux-2.6.32.48/drivers/infiniband/core/cm.c linux-2.6.32.48-openvz/drivers/infiniband/core/cm.c --- linux-2.6.32.48/drivers/infiniband/core/cm.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/infiniband/core/cm.c 2011-11-17 20:08:53.000000000 -0500 @@ -112,7 +112,7 @@ static char const counter_group_names[CM struct cm_counter_group { struct kobject obj; - atomic_long_t counter[CM_ATTR_COUNT]; + atomic_long_unchecked_t counter[CM_ATTR_COUNT]; }; struct cm_counter_attribute { @@ -1386,7 +1386,7 @@ static void cm_dup_req_handler(struct cm struct ib_mad_send_buf *msg = NULL; int ret; - atomic_long_inc(&work->port->counter_group[CM_RECV_DUPLICATES]. + atomic_long_inc_unchecked(&work->port->counter_group[CM_RECV_DUPLICATES]. counter[CM_REQ_COUNTER]); /* Quick state check to discard duplicate REQs. */ @@ -1764,7 +1764,7 @@ static void cm_dup_rep_handler(struct cm if (!cm_id_priv) return; - atomic_long_inc(&work->port->counter_group[CM_RECV_DUPLICATES]. + atomic_long_inc_unchecked(&work->port->counter_group[CM_RECV_DUPLICATES]. counter[CM_REP_COUNTER]); ret = cm_alloc_response_msg(work->port, work->mad_recv_wc, &msg); if (ret) @@ -1931,7 +1931,7 @@ static int cm_rtu_handler(struct cm_work if (cm_id_priv->id.state != IB_CM_REP_SENT && cm_id_priv->id.state != IB_CM_MRA_REP_RCVD) { spin_unlock_irq(&cm_id_priv->lock); - atomic_long_inc(&work->port->counter_group[CM_RECV_DUPLICATES]. + atomic_long_inc_unchecked(&work->port->counter_group[CM_RECV_DUPLICATES]. counter[CM_RTU_COUNTER]); goto out; } @@ -2110,7 +2110,7 @@ static int cm_dreq_handler(struct cm_wor cm_id_priv = cm_acquire_id(dreq_msg->remote_comm_id, dreq_msg->local_comm_id); if (!cm_id_priv) { - atomic_long_inc(&work->port->counter_group[CM_RECV_DUPLICATES]. + atomic_long_inc_unchecked(&work->port->counter_group[CM_RECV_DUPLICATES]. counter[CM_DREQ_COUNTER]); cm_issue_drep(work->port, work->mad_recv_wc); return -EINVAL; @@ -2131,7 +2131,7 @@ static int cm_dreq_handler(struct cm_wor case IB_CM_MRA_REP_RCVD: break; case IB_CM_TIMEWAIT: - atomic_long_inc(&work->port->counter_group[CM_RECV_DUPLICATES]. + atomic_long_inc_unchecked(&work->port->counter_group[CM_RECV_DUPLICATES]. counter[CM_DREQ_COUNTER]); if (cm_alloc_response_msg(work->port, work->mad_recv_wc, &msg)) goto unlock; @@ -2145,7 +2145,7 @@ static int cm_dreq_handler(struct cm_wor cm_free_msg(msg); goto deref; case IB_CM_DREQ_RCVD: - atomic_long_inc(&work->port->counter_group[CM_RECV_DUPLICATES]. + atomic_long_inc_unchecked(&work->port->counter_group[CM_RECV_DUPLICATES]. counter[CM_DREQ_COUNTER]); goto unlock; default: @@ -2501,7 +2501,7 @@ static int cm_mra_handler(struct cm_work ib_modify_mad(cm_id_priv->av.port->mad_agent, cm_id_priv->msg, timeout)) { if (cm_id_priv->id.lap_state == IB_CM_MRA_LAP_RCVD) - atomic_long_inc(&work->port-> + atomic_long_inc_unchecked(&work->port-> counter_group[CM_RECV_DUPLICATES]. counter[CM_MRA_COUNTER]); goto out; @@ -2510,7 +2510,7 @@ static int cm_mra_handler(struct cm_work break; case IB_CM_MRA_REQ_RCVD: case IB_CM_MRA_REP_RCVD: - atomic_long_inc(&work->port->counter_group[CM_RECV_DUPLICATES]. + atomic_long_inc_unchecked(&work->port->counter_group[CM_RECV_DUPLICATES]. counter[CM_MRA_COUNTER]); /* fall through */ default: @@ -2672,7 +2672,7 @@ static int cm_lap_handler(struct cm_work case IB_CM_LAP_IDLE: break; case IB_CM_MRA_LAP_SENT: - atomic_long_inc(&work->port->counter_group[CM_RECV_DUPLICATES]. + atomic_long_inc_unchecked(&work->port->counter_group[CM_RECV_DUPLICATES]. counter[CM_LAP_COUNTER]); if (cm_alloc_response_msg(work->port, work->mad_recv_wc, &msg)) goto unlock; @@ -2688,7 +2688,7 @@ static int cm_lap_handler(struct cm_work cm_free_msg(msg); goto deref; case IB_CM_LAP_RCVD: - atomic_long_inc(&work->port->counter_group[CM_RECV_DUPLICATES]. + atomic_long_inc_unchecked(&work->port->counter_group[CM_RECV_DUPLICATES]. counter[CM_LAP_COUNTER]); goto unlock; default: @@ -2972,7 +2972,7 @@ static int cm_sidr_req_handler(struct cm cur_cm_id_priv = cm_insert_remote_sidr(cm_id_priv); if (cur_cm_id_priv) { spin_unlock_irq(&cm.lock); - atomic_long_inc(&work->port->counter_group[CM_RECV_DUPLICATES]. + atomic_long_inc_unchecked(&work->port->counter_group[CM_RECV_DUPLICATES]. counter[CM_SIDR_REQ_COUNTER]); goto out; /* Duplicate message. */ } @@ -3184,10 +3184,10 @@ static void cm_send_handler(struct ib_ma if (!msg->context[0] && (attr_index != CM_REJ_COUNTER)) msg->retries = 1; - atomic_long_add(1 + msg->retries, + atomic_long_add_unchecked(1 + msg->retries, &port->counter_group[CM_XMIT].counter[attr_index]); if (msg->retries) - atomic_long_add(msg->retries, + atomic_long_add_unchecked(msg->retries, &port->counter_group[CM_XMIT_RETRIES]. counter[attr_index]); @@ -3397,7 +3397,7 @@ static void cm_recv_handler(struct ib_ma } attr_id = be16_to_cpu(mad_recv_wc->recv_buf.mad->mad_hdr.attr_id); - atomic_long_inc(&port->counter_group[CM_RECV]. + atomic_long_inc_unchecked(&port->counter_group[CM_RECV]. counter[attr_id - CM_ATTR_ID_OFFSET]); work = kmalloc(sizeof *work + sizeof(struct ib_sa_path_rec) * paths, @@ -3595,10 +3595,10 @@ static ssize_t cm_show_counter(struct ko cm_attr = container_of(attr, struct cm_counter_attribute, attr); return sprintf(buf, "%ld\n", - atomic_long_read(&group->counter[cm_attr->index])); + atomic_long_read_unchecked(&group->counter[cm_attr->index])); } -static struct sysfs_ops cm_counter_ops = { +static const struct sysfs_ops cm_counter_ops = { .show = cm_show_counter }; diff -urNp linux-2.6.32.48/drivers/infiniband/core/fmr_pool.c linux-2.6.32.48-openvz/drivers/infiniband/core/fmr_pool.c --- linux-2.6.32.48/drivers/infiniband/core/fmr_pool.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/infiniband/core/fmr_pool.c 2011-11-17 20:08:53.000000000 -0500 @@ -97,8 +97,8 @@ struct ib_fmr_pool { struct task_struct *thread; - atomic_t req_ser; - atomic_t flush_ser; + atomic_unchecked_t req_ser; + atomic_unchecked_t flush_ser; wait_queue_head_t force_wait; }; @@ -179,10 +179,10 @@ static int ib_fmr_cleanup_thread(void *p struct ib_fmr_pool *pool = pool_ptr; do { - if (atomic_read(&pool->flush_ser) - atomic_read(&pool->req_ser) < 0) { + if (atomic_read_unchecked(&pool->flush_ser) - atomic_read_unchecked(&pool->req_ser) < 0) { ib_fmr_batch_release(pool); - atomic_inc(&pool->flush_ser); + atomic_inc_unchecked(&pool->flush_ser); wake_up_interruptible(&pool->force_wait); if (pool->flush_function) @@ -190,7 +190,7 @@ static int ib_fmr_cleanup_thread(void *p } set_current_state(TASK_INTERRUPTIBLE); - if (atomic_read(&pool->flush_ser) - atomic_read(&pool->req_ser) >= 0 && + if (atomic_read_unchecked(&pool->flush_ser) - atomic_read_unchecked(&pool->req_ser) >= 0 && !kthread_should_stop()) schedule(); __set_current_state(TASK_RUNNING); @@ -282,8 +282,8 @@ struct ib_fmr_pool *ib_create_fmr_pool(s pool->dirty_watermark = params->dirty_watermark; pool->dirty_len = 0; spin_lock_init(&pool->pool_lock); - atomic_set(&pool->req_ser, 0); - atomic_set(&pool->flush_ser, 0); + atomic_set_unchecked(&pool->req_ser, 0); + atomic_set_unchecked(&pool->flush_ser, 0); init_waitqueue_head(&pool->force_wait); pool->thread = kthread_run(ib_fmr_cleanup_thread, @@ -411,11 +411,11 @@ int ib_flush_fmr_pool(struct ib_fmr_pool } spin_unlock_irq(&pool->pool_lock); - serial = atomic_inc_return(&pool->req_ser); + serial = atomic_inc_return_unchecked(&pool->req_ser); wake_up_process(pool->thread); if (wait_event_interruptible(pool->force_wait, - atomic_read(&pool->flush_ser) - serial >= 0)) + atomic_read_unchecked(&pool->flush_ser) - serial >= 0)) return -EINTR; return 0; @@ -525,7 +525,7 @@ int ib_fmr_pool_unmap(struct ib_pool_fmr } else { list_add_tail(&fmr->list, &pool->dirty_list); if (++pool->dirty_len >= pool->dirty_watermark) { - atomic_inc(&pool->req_ser); + atomic_inc_unchecked(&pool->req_ser); wake_up_process(pool->thread); } } diff -urNp linux-2.6.32.48/drivers/infiniband/core/sysfs.c linux-2.6.32.48-openvz/drivers/infiniband/core/sysfs.c --- linux-2.6.32.48/drivers/infiniband/core/sysfs.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/infiniband/core/sysfs.c 2011-11-17 20:08:53.000000000 -0500 @@ -79,7 +79,7 @@ static ssize_t port_attr_show(struct kob return port_attr->show(p, port_attr, buf); } -static struct sysfs_ops port_sysfs_ops = { +static const struct sysfs_ops port_sysfs_ops = { .show = port_attr_show }; diff -urNp linux-2.6.32.48/drivers/infiniband/core/uverbs_marshall.c linux-2.6.32.48-openvz/drivers/infiniband/core/uverbs_marshall.c --- linux-2.6.32.48/drivers/infiniband/core/uverbs_marshall.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/infiniband/core/uverbs_marshall.c 2011-11-17 20:08:53.000000000 -0500 @@ -40,18 +40,21 @@ void ib_copy_ah_attr_to_user(struct ib_u dst->grh.sgid_index = src->grh.sgid_index; dst->grh.hop_limit = src->grh.hop_limit; dst->grh.traffic_class = src->grh.traffic_class; + memset(&dst->grh.reserved, 0, sizeof(dst->grh.reserved)); dst->dlid = src->dlid; dst->sl = src->sl; dst->src_path_bits = src->src_path_bits; dst->static_rate = src->static_rate; dst->is_global = src->ah_flags & IB_AH_GRH ? 1 : 0; dst->port_num = src->port_num; + dst->reserved = 0; } EXPORT_SYMBOL(ib_copy_ah_attr_to_user); void ib_copy_qp_attr_to_user(struct ib_uverbs_qp_attr *dst, struct ib_qp_attr *src) { + dst->qp_state = src->qp_state; dst->cur_qp_state = src->cur_qp_state; dst->path_mtu = src->path_mtu; dst->path_mig_state = src->path_mig_state; @@ -83,6 +86,7 @@ void ib_copy_qp_attr_to_user(struct ib_u dst->rnr_retry = src->rnr_retry; dst->alt_port_num = src->alt_port_num; dst->alt_timeout = src->alt_timeout; + memset(dst->reserved, 0, sizeof(dst->reserved)); } EXPORT_SYMBOL(ib_copy_qp_attr_to_user); diff -urNp linux-2.6.32.48/drivers/infiniband/hw/ipath/ipath_fs.c linux-2.6.32.48-openvz/drivers/infiniband/hw/ipath/ipath_fs.c --- linux-2.6.32.48/drivers/infiniband/hw/ipath/ipath_fs.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/infiniband/hw/ipath/ipath_fs.c 2011-11-17 20:08:53.000000000 -0500 @@ -110,6 +110,8 @@ static ssize_t atomic_counters_read(stru struct infinipath_counters counters; struct ipath_devdata *dd; + pax_track_stack(); + dd = file->f_path.dentry->d_inode->i_private; dd->ipath_f_read_counters(dd, &counters); diff -urNp linux-2.6.32.48/drivers/infiniband/hw/nes/nes.c linux-2.6.32.48-openvz/drivers/infiniband/hw/nes/nes.c --- linux-2.6.32.48/drivers/infiniband/hw/nes/nes.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/infiniband/hw/nes/nes.c 2011-11-17 20:08:53.000000000 -0500 @@ -102,7 +102,7 @@ MODULE_PARM_DESC(limit_maxrdreqsz, "Limi LIST_HEAD(nes_adapter_list); static LIST_HEAD(nes_dev_list); -atomic_t qps_destroyed; +atomic_unchecked_t qps_destroyed; static unsigned int ee_flsh_adapter; static unsigned int sysfs_nonidx_addr; @@ -259,7 +259,7 @@ static void nes_cqp_rem_ref_callback(str struct nes_adapter *nesadapter = nesdev->nesadapter; u32 qp_id; - atomic_inc(&qps_destroyed); + atomic_inc_unchecked(&qps_destroyed); /* Free the control structures */ diff -urNp linux-2.6.32.48/drivers/infiniband/hw/nes/nes_cm.c linux-2.6.32.48-openvz/drivers/infiniband/hw/nes/nes_cm.c --- linux-2.6.32.48/drivers/infiniband/hw/nes/nes_cm.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/infiniband/hw/nes/nes_cm.c 2011-11-17 20:08:53.000000000 -0500 @@ -69,11 +69,11 @@ u32 cm_packets_received; u32 cm_listens_created; u32 cm_listens_destroyed; u32 cm_backlog_drops; -atomic_t cm_loopbacks; -atomic_t cm_nodes_created; -atomic_t cm_nodes_destroyed; -atomic_t cm_accel_dropped_pkts; -atomic_t cm_resets_recvd; +atomic_unchecked_t cm_loopbacks; +atomic_unchecked_t cm_nodes_created; +atomic_unchecked_t cm_nodes_destroyed; +atomic_unchecked_t cm_accel_dropped_pkts; +atomic_unchecked_t cm_resets_recvd; static inline int mini_cm_accelerated(struct nes_cm_core *, struct nes_cm_node *); @@ -149,13 +149,13 @@ static struct nes_cm_ops nes_cm_api = { static struct nes_cm_core *g_cm_core; -atomic_t cm_connects; -atomic_t cm_accepts; -atomic_t cm_disconnects; -atomic_t cm_closes; -atomic_t cm_connecteds; -atomic_t cm_connect_reqs; -atomic_t cm_rejects; +atomic_unchecked_t cm_connects; +atomic_unchecked_t cm_accepts; +atomic_unchecked_t cm_disconnects; +atomic_unchecked_t cm_closes; +atomic_unchecked_t cm_connecteds; +atomic_unchecked_t cm_connect_reqs; +atomic_unchecked_t cm_rejects; /** @@ -1195,7 +1195,7 @@ static struct nes_cm_node *make_cm_node( cm_node->rem_mac); add_hte_node(cm_core, cm_node); - atomic_inc(&cm_nodes_created); + atomic_inc_unchecked(&cm_nodes_created); return cm_node; } @@ -1253,7 +1253,7 @@ static int rem_ref_cm_node(struct nes_cm } atomic_dec(&cm_core->node_cnt); - atomic_inc(&cm_nodes_destroyed); + atomic_inc_unchecked(&cm_nodes_destroyed); nesqp = cm_node->nesqp; if (nesqp) { nesqp->cm_node = NULL; @@ -1320,7 +1320,7 @@ static int process_options(struct nes_cm static void drop_packet(struct sk_buff *skb) { - atomic_inc(&cm_accel_dropped_pkts); + atomic_inc_unchecked(&cm_accel_dropped_pkts); dev_kfree_skb_any(skb); } @@ -1377,7 +1377,7 @@ static void handle_rst_pkt(struct nes_cm int reset = 0; /* whether to send reset in case of err.. */ int passive_state; - atomic_inc(&cm_resets_recvd); + atomic_inc_unchecked(&cm_resets_recvd); nes_debug(NES_DBG_CM, "Received Reset, cm_node = %p, state = %u." " refcnt=%d\n", cm_node, cm_node->state, atomic_read(&cm_node->ref_count)); @@ -2000,7 +2000,7 @@ static struct nes_cm_node *mini_cm_conne rem_ref_cm_node(cm_node->cm_core, cm_node); return NULL; } - atomic_inc(&cm_loopbacks); + atomic_inc_unchecked(&cm_loopbacks); loopbackremotenode->loopbackpartner = cm_node; loopbackremotenode->tcp_cntxt.rcv_wscale = NES_CM_DEFAULT_RCV_WND_SCALE; @@ -2262,7 +2262,7 @@ static int mini_cm_recv_pkt(struct nes_c add_ref_cm_node(cm_node); } else if (cm_node->state == NES_CM_STATE_TSA) { rem_ref_cm_node(cm_core, cm_node); - atomic_inc(&cm_accel_dropped_pkts); + atomic_inc_unchecked(&cm_accel_dropped_pkts); dev_kfree_skb_any(skb); break; } @@ -2568,7 +2568,7 @@ static int nes_cm_disconn_true(struct ne if ((cm_id) && (cm_id->event_handler)) { if (issue_disconn) { - atomic_inc(&cm_disconnects); + atomic_inc_unchecked(&cm_disconnects); cm_event.event = IW_CM_EVENT_DISCONNECT; cm_event.status = disconn_status; cm_event.local_addr = cm_id->local_addr; @@ -2590,7 +2590,7 @@ static int nes_cm_disconn_true(struct ne } if (issue_close) { - atomic_inc(&cm_closes); + atomic_inc_unchecked(&cm_closes); nes_disconnect(nesqp, 1); cm_id->provider_data = nesqp; @@ -2710,7 +2710,7 @@ int nes_accept(struct iw_cm_id *cm_id, s nes_debug(NES_DBG_CM, "QP%u, cm_node=%p, jiffies = %lu listener = %p\n", nesqp->hwqp.qp_id, cm_node, jiffies, cm_node->listener); - atomic_inc(&cm_accepts); + atomic_inc_unchecked(&cm_accepts); nes_debug(NES_DBG_CM, "netdev refcnt = %u.\n", atomic_read(&nesvnic->netdev->refcnt)); @@ -2919,7 +2919,7 @@ int nes_reject(struct iw_cm_id *cm_id, c struct nes_cm_core *cm_core; - atomic_inc(&cm_rejects); + atomic_inc_unchecked(&cm_rejects); cm_node = (struct nes_cm_node *) cm_id->provider_data; loopback = cm_node->loopbackpartner; cm_core = cm_node->cm_core; @@ -2982,7 +2982,7 @@ int nes_connect(struct iw_cm_id *cm_id, ntohl(cm_id->local_addr.sin_addr.s_addr), ntohs(cm_id->local_addr.sin_port)); - atomic_inc(&cm_connects); + atomic_inc_unchecked(&cm_connects); nesqp->active_conn = 1; /* cache the cm_id in the qp */ @@ -3195,7 +3195,7 @@ static void cm_event_connected(struct ne if (nesqp->destroyed) { return; } - atomic_inc(&cm_connecteds); + atomic_inc_unchecked(&cm_connecteds); nes_debug(NES_DBG_CM, "QP%u attempting to connect to 0x%08X:0x%04X on" " local port 0x%04X. jiffies = %lu.\n", nesqp->hwqp.qp_id, @@ -3403,7 +3403,7 @@ static void cm_event_reset(struct nes_cm ret = cm_id->event_handler(cm_id, &cm_event); cm_id->add_ref(cm_id); - atomic_inc(&cm_closes); + atomic_inc_unchecked(&cm_closes); cm_event.event = IW_CM_EVENT_CLOSE; cm_event.status = IW_CM_EVENT_STATUS_OK; cm_event.provider_data = cm_id->provider_data; @@ -3439,7 +3439,7 @@ static void cm_event_mpa_req(struct nes_ return; cm_id = cm_node->cm_id; - atomic_inc(&cm_connect_reqs); + atomic_inc_unchecked(&cm_connect_reqs); nes_debug(NES_DBG_CM, "cm_node = %p - cm_id = %p, jiffies = %lu\n", cm_node, cm_id, jiffies); @@ -3477,7 +3477,7 @@ static void cm_event_mpa_reject(struct n return; cm_id = cm_node->cm_id; - atomic_inc(&cm_connect_reqs); + atomic_inc_unchecked(&cm_connect_reqs); nes_debug(NES_DBG_CM, "cm_node = %p - cm_id = %p, jiffies = %lu\n", cm_node, cm_id, jiffies); diff -urNp linux-2.6.32.48/drivers/infiniband/hw/nes/nes.h linux-2.6.32.48-openvz/drivers/infiniband/hw/nes/nes.h --- linux-2.6.32.48/drivers/infiniband/hw/nes/nes.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/infiniband/hw/nes/nes.h 2011-11-17 20:08:53.000000000 -0500 @@ -174,17 +174,17 @@ extern unsigned int nes_debug_level; extern unsigned int wqm_quanta; extern struct list_head nes_adapter_list; -extern atomic_t cm_connects; -extern atomic_t cm_accepts; -extern atomic_t cm_disconnects; -extern atomic_t cm_closes; -extern atomic_t cm_connecteds; -extern atomic_t cm_connect_reqs; -extern atomic_t cm_rejects; -extern atomic_t mod_qp_timouts; -extern atomic_t qps_created; -extern atomic_t qps_destroyed; -extern atomic_t sw_qps_destroyed; +extern atomic_unchecked_t cm_connects; +extern atomic_unchecked_t cm_accepts; +extern atomic_unchecked_t cm_disconnects; +extern atomic_unchecked_t cm_closes; +extern atomic_unchecked_t cm_connecteds; +extern atomic_unchecked_t cm_connect_reqs; +extern atomic_unchecked_t cm_rejects; +extern atomic_unchecked_t mod_qp_timouts; +extern atomic_unchecked_t qps_created; +extern atomic_unchecked_t qps_destroyed; +extern atomic_unchecked_t sw_qps_destroyed; extern u32 mh_detected; extern u32 mh_pauses_sent; extern u32 cm_packets_sent; @@ -196,11 +196,11 @@ extern u32 cm_packets_retrans; extern u32 cm_listens_created; extern u32 cm_listens_destroyed; extern u32 cm_backlog_drops; -extern atomic_t cm_loopbacks; -extern atomic_t cm_nodes_created; -extern atomic_t cm_nodes_destroyed; -extern atomic_t cm_accel_dropped_pkts; -extern atomic_t cm_resets_recvd; +extern atomic_unchecked_t cm_loopbacks; +extern atomic_unchecked_t cm_nodes_created; +extern atomic_unchecked_t cm_nodes_destroyed; +extern atomic_unchecked_t cm_accel_dropped_pkts; +extern atomic_unchecked_t cm_resets_recvd; extern u32 int_mod_timer_init; extern u32 int_mod_cq_depth_256; diff -urNp linux-2.6.32.48/drivers/infiniband/hw/nes/nes_nic.c linux-2.6.32.48-openvz/drivers/infiniband/hw/nes/nes_nic.c --- linux-2.6.32.48/drivers/infiniband/hw/nes/nes_nic.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/infiniband/hw/nes/nes_nic.c 2011-11-17 20:08:53.000000000 -0500 @@ -1210,17 +1210,17 @@ static void nes_netdev_get_ethtool_stats target_stat_values[++index] = mh_detected; target_stat_values[++index] = mh_pauses_sent; target_stat_values[++index] = nesvnic->endnode_ipv4_tcp_retransmits; - target_stat_values[++index] = atomic_read(&cm_connects); - target_stat_values[++index] = atomic_read(&cm_accepts); - target_stat_values[++index] = atomic_read(&cm_disconnects); - target_stat_values[++index] = atomic_read(&cm_connecteds); - target_stat_values[++index] = atomic_read(&cm_connect_reqs); - target_stat_values[++index] = atomic_read(&cm_rejects); - target_stat_values[++index] = atomic_read(&mod_qp_timouts); - target_stat_values[++index] = atomic_read(&qps_created); - target_stat_values[++index] = atomic_read(&sw_qps_destroyed); - target_stat_values[++index] = atomic_read(&qps_destroyed); - target_stat_values[++index] = atomic_read(&cm_closes); + target_stat_values[++index] = atomic_read_unchecked(&cm_connects); + target_stat_values[++index] = atomic_read_unchecked(&cm_accepts); + target_stat_values[++index] = atomic_read_unchecked(&cm_disconnects); + target_stat_values[++index] = atomic_read_unchecked(&cm_connecteds); + target_stat_values[++index] = atomic_read_unchecked(&cm_connect_reqs); + target_stat_values[++index] = atomic_read_unchecked(&cm_rejects); + target_stat_values[++index] = atomic_read_unchecked(&mod_qp_timouts); + target_stat_values[++index] = atomic_read_unchecked(&qps_created); + target_stat_values[++index] = atomic_read_unchecked(&sw_qps_destroyed); + target_stat_values[++index] = atomic_read_unchecked(&qps_destroyed); + target_stat_values[++index] = atomic_read_unchecked(&cm_closes); target_stat_values[++index] = cm_packets_sent; target_stat_values[++index] = cm_packets_bounced; target_stat_values[++index] = cm_packets_created; @@ -1230,11 +1230,11 @@ static void nes_netdev_get_ethtool_stats target_stat_values[++index] = cm_listens_created; target_stat_values[++index] = cm_listens_destroyed; target_stat_values[++index] = cm_backlog_drops; - target_stat_values[++index] = atomic_read(&cm_loopbacks); - target_stat_values[++index] = atomic_read(&cm_nodes_created); - target_stat_values[++index] = atomic_read(&cm_nodes_destroyed); - target_stat_values[++index] = atomic_read(&cm_accel_dropped_pkts); - target_stat_values[++index] = atomic_read(&cm_resets_recvd); + target_stat_values[++index] = atomic_read_unchecked(&cm_loopbacks); + target_stat_values[++index] = atomic_read_unchecked(&cm_nodes_created); + target_stat_values[++index] = atomic_read_unchecked(&cm_nodes_destroyed); + target_stat_values[++index] = atomic_read_unchecked(&cm_accel_dropped_pkts); + target_stat_values[++index] = atomic_read_unchecked(&cm_resets_recvd); target_stat_values[++index] = int_mod_timer_init; target_stat_values[++index] = int_mod_cq_depth_1; target_stat_values[++index] = int_mod_cq_depth_4; diff -urNp linux-2.6.32.48/drivers/infiniband/hw/nes/nes_verbs.c linux-2.6.32.48-openvz/drivers/infiniband/hw/nes/nes_verbs.c --- linux-2.6.32.48/drivers/infiniband/hw/nes/nes_verbs.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/infiniband/hw/nes/nes_verbs.c 2011-11-17 20:08:53.000000000 -0500 @@ -45,9 +45,9 @@ #include -atomic_t mod_qp_timouts; -atomic_t qps_created; -atomic_t sw_qps_destroyed; +atomic_unchecked_t mod_qp_timouts; +atomic_unchecked_t qps_created; +atomic_unchecked_t sw_qps_destroyed; static void nes_unregister_ofa_device(struct nes_ib_device *nesibdev); @@ -1240,7 +1240,7 @@ static struct ib_qp *nes_create_qp(struc if (init_attr->create_flags) return ERR_PTR(-EINVAL); - atomic_inc(&qps_created); + atomic_inc_unchecked(&qps_created); switch (init_attr->qp_type) { case IB_QPT_RC: if (nes_drv_opt & NES_DRV_OPT_NO_INLINE_DATA) { @@ -1568,7 +1568,7 @@ static int nes_destroy_qp(struct ib_qp * struct iw_cm_event cm_event; int ret; - atomic_inc(&sw_qps_destroyed); + atomic_inc_unchecked(&sw_qps_destroyed); nesqp->destroyed = 1; /* Blow away the connection if it exists. */ diff -urNp linux-2.6.32.48/drivers/input/gameport/gameport.c linux-2.6.32.48-openvz/drivers/input/gameport/gameport.c --- linux-2.6.32.48/drivers/input/gameport/gameport.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/input/gameport/gameport.c 2011-11-17 20:08:53.000000000 -0500 @@ -515,13 +515,13 @@ EXPORT_SYMBOL(gameport_set_phys); */ static void gameport_init_port(struct gameport *gameport) { - static atomic_t gameport_no = ATOMIC_INIT(0); + static atomic_unchecked_t gameport_no = ATOMIC_INIT(0); __module_get(THIS_MODULE); mutex_init(&gameport->drv_mutex); device_initialize(&gameport->dev); - dev_set_name(&gameport->dev, "gameport%lu", (unsigned long)atomic_inc_return(&gameport_no) - 1); + dev_set_name(&gameport->dev, "gameport%lu", (unsigned long)atomic_inc_return_unchecked(&gameport_no) - 1); gameport->dev.bus = &gameport_bus; gameport->dev.release = gameport_release_port; if (gameport->parent) diff -urNp linux-2.6.32.48/drivers/input/input.c linux-2.6.32.48-openvz/drivers/input/input.c --- linux-2.6.32.48/drivers/input/input.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/input/input.c 2011-11-17 20:08:53.000000000 -0500 @@ -1558,7 +1558,7 @@ EXPORT_SYMBOL(input_set_capability); */ int input_register_device(struct input_dev *dev) { - static atomic_t input_no = ATOMIC_INIT(0); + static atomic_unchecked_t input_no = ATOMIC_INIT(0); struct input_handler *handler; const char *path; int error; @@ -1585,7 +1585,7 @@ int input_register_device(struct input_d dev->setkeycode = input_default_setkeycode; dev_set_name(&dev->dev, "input%ld", - (unsigned long) atomic_inc_return(&input_no) - 1); + (unsigned long) atomic_inc_return_unchecked(&input_no) - 1); error = device_add(&dev->dev); if (error) diff -urNp linux-2.6.32.48/drivers/input/joystick/sidewinder.c linux-2.6.32.48-openvz/drivers/input/joystick/sidewinder.c --- linux-2.6.32.48/drivers/input/joystick/sidewinder.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/input/joystick/sidewinder.c 2011-11-17 20:08:53.000000000 -0500 @@ -30,6 +30,7 @@ #include #include #include +#include #include #include #include @@ -428,6 +429,8 @@ static int sw_read(struct sw *sw) unsigned char buf[SW_LENGTH]; int i; + pax_track_stack(); + i = sw_read_packet(sw->gameport, buf, sw->length, 0); if (sw->type == SW_ID_3DP && sw->length == 66 && i != 66) { /* Broken packet, try to fix */ diff -urNp linux-2.6.32.48/drivers/input/joystick/xpad.c linux-2.6.32.48-openvz/drivers/input/joystick/xpad.c --- linux-2.6.32.48/drivers/input/joystick/xpad.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/input/joystick/xpad.c 2011-11-17 20:08:53.000000000 -0500 @@ -621,7 +621,7 @@ static void xpad_led_set(struct led_clas static int xpad_led_probe(struct usb_xpad *xpad) { - static atomic_t led_seq = ATOMIC_INIT(0); + static atomic_unchecked_t led_seq = ATOMIC_INIT(0); long led_no; struct xpad_led *led; struct led_classdev *led_cdev; @@ -634,7 +634,7 @@ static int xpad_led_probe(struct usb_xpa if (!led) return -ENOMEM; - led_no = (long)atomic_inc_return(&led_seq) - 1; + led_no = (long)atomic_inc_return_unchecked(&led_seq) - 1; snprintf(led->name, sizeof(led->name), "xpad%ld", led_no); led->xpad = xpad; diff -urNp linux-2.6.32.48/drivers/input/serio/serio.c linux-2.6.32.48-openvz/drivers/input/serio/serio.c --- linux-2.6.32.48/drivers/input/serio/serio.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/input/serio/serio.c 2011-11-17 20:08:53.000000000 -0500 @@ -527,7 +527,7 @@ static void serio_release_port(struct de */ static void serio_init_port(struct serio *serio) { - static atomic_t serio_no = ATOMIC_INIT(0); + static atomic_unchecked_t serio_no = ATOMIC_INIT(0); __module_get(THIS_MODULE); @@ -536,7 +536,7 @@ static void serio_init_port(struct serio mutex_init(&serio->drv_mutex); device_initialize(&serio->dev); dev_set_name(&serio->dev, "serio%ld", - (long)atomic_inc_return(&serio_no) - 1); + (long)atomic_inc_return_unchecked(&serio_no) - 1); serio->dev.bus = &serio_bus; serio->dev.release = serio_release_port; if (serio->parent) { diff -urNp linux-2.6.32.48/drivers/isdn/gigaset/common.c linux-2.6.32.48-openvz/drivers/isdn/gigaset/common.c --- linux-2.6.32.48/drivers/isdn/gigaset/common.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/isdn/gigaset/common.c 2011-11-17 20:08:53.000000000 -0500 @@ -712,7 +712,7 @@ struct cardstate *gigaset_initcs(struct cs->commands_pending = 0; cs->cur_at_seq = 0; cs->gotfwver = -1; - cs->open_count = 0; + local_set(&cs->open_count, 0); cs->dev = NULL; cs->tty = NULL; cs->tty_dev = NULL; diff -urNp linux-2.6.32.48/drivers/isdn/gigaset/gigaset.h linux-2.6.32.48-openvz/drivers/isdn/gigaset/gigaset.h --- linux-2.6.32.48/drivers/isdn/gigaset/gigaset.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/isdn/gigaset/gigaset.h 2011-11-17 20:08:53.000000000 -0500 @@ -34,6 +34,7 @@ #include #include #include +#include #define GIG_VERSION {0,5,0,0} #define GIG_COMPAT {0,4,0,0} @@ -446,7 +447,7 @@ struct cardstate { spinlock_t cmdlock; unsigned curlen, cmdbytes; - unsigned open_count; + local_t open_count; struct tty_struct *tty; struct tasklet_struct if_wake_tasklet; unsigned control_state; diff -urNp linux-2.6.32.48/drivers/isdn/gigaset/interface.c linux-2.6.32.48-openvz/drivers/isdn/gigaset/interface.c --- linux-2.6.32.48/drivers/isdn/gigaset/interface.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/isdn/gigaset/interface.c 2011-11-17 20:08:53.000000000 -0500 @@ -165,9 +165,7 @@ static int if_open(struct tty_struct *tt return -ERESTARTSYS; // FIXME -EINTR? tty->driver_data = cs; - ++cs->open_count; - - if (cs->open_count == 1) { + if (local_inc_return(&cs->open_count) == 1) { spin_lock_irqsave(&cs->lock, flags); cs->tty = tty; spin_unlock_irqrestore(&cs->lock, flags); @@ -195,10 +193,10 @@ static void if_close(struct tty_struct * if (!cs->connected) gig_dbg(DEBUG_IF, "not connected"); /* nothing to do */ - else if (!cs->open_count) + else if (!local_read(&cs->open_count)) dev_warn(cs->dev, "%s: device not opened\n", __func__); else { - if (!--cs->open_count) { + if (!local_dec_return(&cs->open_count)) { spin_lock_irqsave(&cs->lock, flags); cs->tty = NULL; spin_unlock_irqrestore(&cs->lock, flags); @@ -233,7 +231,7 @@ static int if_ioctl(struct tty_struct *t if (!cs->connected) { gig_dbg(DEBUG_IF, "not connected"); retval = -ENODEV; - } else if (!cs->open_count) + } else if (!local_read(&cs->open_count)) dev_warn(cs->dev, "%s: device not opened\n", __func__); else { retval = 0; @@ -361,7 +359,7 @@ static int if_write(struct tty_struct *t if (!cs->connected) { gig_dbg(DEBUG_IF, "not connected"); retval = -ENODEV; - } else if (!cs->open_count) + } else if (!local_read(&cs->open_count)) dev_warn(cs->dev, "%s: device not opened\n", __func__); else if (cs->mstate != MS_LOCKED) { dev_warn(cs->dev, "can't write to unlocked device\n"); @@ -395,7 +393,7 @@ static int if_write_room(struct tty_stru if (!cs->connected) { gig_dbg(DEBUG_IF, "not connected"); retval = -ENODEV; - } else if (!cs->open_count) + } else if (!local_read(&cs->open_count)) dev_warn(cs->dev, "%s: device not opened\n", __func__); else if (cs->mstate != MS_LOCKED) { dev_warn(cs->dev, "can't write to unlocked device\n"); @@ -425,7 +423,7 @@ static int if_chars_in_buffer(struct tty if (!cs->connected) gig_dbg(DEBUG_IF, "not connected"); - else if (!cs->open_count) + else if (!local_read(&cs->open_count)) dev_warn(cs->dev, "%s: device not opened\n", __func__); else if (cs->mstate != MS_LOCKED) dev_warn(cs->dev, "can't write to unlocked device\n"); @@ -453,7 +451,7 @@ static void if_throttle(struct tty_struc if (!cs->connected) gig_dbg(DEBUG_IF, "not connected"); /* nothing to do */ - else if (!cs->open_count) + else if (!local_read(&cs->open_count)) dev_warn(cs->dev, "%s: device not opened\n", __func__); else { //FIXME @@ -478,7 +476,7 @@ static void if_unthrottle(struct tty_str if (!cs->connected) gig_dbg(DEBUG_IF, "not connected"); /* nothing to do */ - else if (!cs->open_count) + else if (!local_read(&cs->open_count)) dev_warn(cs->dev, "%s: device not opened\n", __func__); else { //FIXME @@ -510,7 +508,7 @@ static void if_set_termios(struct tty_st goto out; } - if (!cs->open_count) { + if (!local_read(&cs->open_count)) { dev_warn(cs->dev, "%s: device not opened\n", __func__); goto out; } diff -urNp linux-2.6.32.48/drivers/isdn/hardware/avm/b1.c linux-2.6.32.48-openvz/drivers/isdn/hardware/avm/b1.c --- linux-2.6.32.48/drivers/isdn/hardware/avm/b1.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/isdn/hardware/avm/b1.c 2011-11-17 20:08:53.000000000 -0500 @@ -173,7 +173,7 @@ int b1_load_t4file(avmcard *card, capilo } if (left) { if (t4file->user) { - if (copy_from_user(buf, dp, left)) + if (left > sizeof buf || copy_from_user(buf, dp, left)) return -EFAULT; } else { memcpy(buf, dp, left); @@ -221,7 +221,7 @@ int b1_load_config(avmcard *card, capilo } if (left) { if (config->user) { - if (copy_from_user(buf, dp, left)) + if (left > sizeof buf || copy_from_user(buf, dp, left)) return -EFAULT; } else { memcpy(buf, dp, left); diff -urNp linux-2.6.32.48/drivers/isdn/hardware/eicon/capidtmf.c linux-2.6.32.48-openvz/drivers/isdn/hardware/eicon/capidtmf.c --- linux-2.6.32.48/drivers/isdn/hardware/eicon/capidtmf.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/isdn/hardware/eicon/capidtmf.c 2011-11-17 20:08:53.000000000 -0500 @@ -498,6 +498,7 @@ void capidtmf_recv_block (t_capidtmf_sta byte goertzel_result_buffer[CAPIDTMF_RECV_TOTAL_FREQUENCY_COUNT]; short windowed_sample_buffer[CAPIDTMF_RECV_WINDOWED_SAMPLES]; + pax_track_stack(); if (p_state->recv.state & CAPIDTMF_RECV_STATE_DTMF_ACTIVE) { diff -urNp linux-2.6.32.48/drivers/isdn/hardware/eicon/capifunc.c linux-2.6.32.48-openvz/drivers/isdn/hardware/eicon/capifunc.c --- linux-2.6.32.48/drivers/isdn/hardware/eicon/capifunc.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/isdn/hardware/eicon/capifunc.c 2011-11-17 20:08:53.000000000 -0500 @@ -1055,6 +1055,8 @@ static int divacapi_connect_didd(void) IDI_SYNC_REQ req; DESCRIPTOR DIDD_Table[MAX_DESCRIPTORS]; + pax_track_stack(); + DIVA_DIDD_Read(DIDD_Table, sizeof(DIDD_Table)); for (x = 0; x < MAX_DESCRIPTORS; x++) { diff -urNp linux-2.6.32.48/drivers/isdn/hardware/eicon/diddfunc.c linux-2.6.32.48-openvz/drivers/isdn/hardware/eicon/diddfunc.c --- linux-2.6.32.48/drivers/isdn/hardware/eicon/diddfunc.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/isdn/hardware/eicon/diddfunc.c 2011-11-17 20:08:53.000000000 -0500 @@ -54,6 +54,8 @@ static int DIVA_INIT_FUNCTION connect_di IDI_SYNC_REQ req; DESCRIPTOR DIDD_Table[MAX_DESCRIPTORS]; + pax_track_stack(); + DIVA_DIDD_Read(DIDD_Table, sizeof(DIDD_Table)); for (x = 0; x < MAX_DESCRIPTORS; x++) { diff -urNp linux-2.6.32.48/drivers/isdn/hardware/eicon/divasfunc.c linux-2.6.32.48-openvz/drivers/isdn/hardware/eicon/divasfunc.c --- linux-2.6.32.48/drivers/isdn/hardware/eicon/divasfunc.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/isdn/hardware/eicon/divasfunc.c 2011-11-17 20:08:53.000000000 -0500 @@ -161,6 +161,8 @@ static int DIVA_INIT_FUNCTION connect_di IDI_SYNC_REQ req; DESCRIPTOR DIDD_Table[MAX_DESCRIPTORS]; + pax_track_stack(); + DIVA_DIDD_Read(DIDD_Table, sizeof(DIDD_Table)); for (x = 0; x < MAX_DESCRIPTORS; x++) { diff -urNp linux-2.6.32.48/drivers/isdn/hardware/eicon/divasync.h linux-2.6.32.48-openvz/drivers/isdn/hardware/eicon/divasync.h --- linux-2.6.32.48/drivers/isdn/hardware/eicon/divasync.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/isdn/hardware/eicon/divasync.h 2011-11-17 20:09:04.000000000 -0500 @@ -146,7 +146,7 @@ typedef struct _diva_didd_add_adapter { } diva_didd_add_adapter_t; typedef struct _diva_didd_remove_adapter { IDI_CALL p_request; -} diva_didd_remove_adapter_t; +} __no_const diva_didd_remove_adapter_t; typedef struct _diva_didd_read_adapter_array { void * buffer; dword length; diff -urNp linux-2.6.32.48/drivers/isdn/hardware/eicon/idifunc.c linux-2.6.32.48-openvz/drivers/isdn/hardware/eicon/idifunc.c --- linux-2.6.32.48/drivers/isdn/hardware/eicon/idifunc.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/isdn/hardware/eicon/idifunc.c 2011-11-17 20:08:53.000000000 -0500 @@ -188,6 +188,8 @@ static int DIVA_INIT_FUNCTION connect_di IDI_SYNC_REQ req; DESCRIPTOR DIDD_Table[MAX_DESCRIPTORS]; + pax_track_stack(); + DIVA_DIDD_Read(DIDD_Table, sizeof(DIDD_Table)); for (x = 0; x < MAX_DESCRIPTORS; x++) { diff -urNp linux-2.6.32.48/drivers/isdn/hardware/eicon/message.c linux-2.6.32.48-openvz/drivers/isdn/hardware/eicon/message.c --- linux-2.6.32.48/drivers/isdn/hardware/eicon/message.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/isdn/hardware/eicon/message.c 2011-11-17 20:08:53.000000000 -0500 @@ -4889,6 +4889,8 @@ static void sig_ind(PLCI *plci) dword d; word w; + pax_track_stack(); + a = plci->adapter; Id = ((word)plci->Id<<8)|a->Id; PUT_WORD(&SS_Ind[4],0x0000); @@ -7484,6 +7486,8 @@ static word add_b1(PLCI *plci, API_PARSE word j, n, w; dword d; + pax_track_stack(); + for(i=0;i<8;i++) bp_parms[i].length = 0; for(i=0;i<2;i++) global_config[i].length = 0; @@ -7958,6 +7962,8 @@ static word add_b23(PLCI *plci, API_PARS const byte llc3[] = {4,3,2,2,6,6,0}; const byte header[] = {0,2,3,3,0,0,0}; + pax_track_stack(); + for(i=0;i<8;i++) bp_parms[i].length = 0; for(i=0;i<6;i++) b2_config_parms[i].length = 0; for(i=0;i<5;i++) b3_config_parms[i].length = 0; @@ -14761,6 +14767,8 @@ static void group_optimization(DIVA_CAPI word appl_number_group_type[MAX_APPL]; PLCI *auxplci; + pax_track_stack(); + set_group_ind_mask (plci); /* all APPLs within this inc. call are allowed to dial in */ if(!a->group_optimization_enabled) diff -urNp linux-2.6.32.48/drivers/isdn/hardware/eicon/mntfunc.c linux-2.6.32.48-openvz/drivers/isdn/hardware/eicon/mntfunc.c --- linux-2.6.32.48/drivers/isdn/hardware/eicon/mntfunc.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/isdn/hardware/eicon/mntfunc.c 2011-11-17 20:08:53.000000000 -0500 @@ -79,6 +79,8 @@ static int DIVA_INIT_FUNCTION connect_di IDI_SYNC_REQ req; DESCRIPTOR DIDD_Table[MAX_DESCRIPTORS]; + pax_track_stack(); + DIVA_DIDD_Read(DIDD_Table, sizeof(DIDD_Table)); for (x = 0; x < MAX_DESCRIPTORS; x++) { diff -urNp linux-2.6.32.48/drivers/isdn/hardware/eicon/xdi_adapter.h linux-2.6.32.48-openvz/drivers/isdn/hardware/eicon/xdi_adapter.h --- linux-2.6.32.48/drivers/isdn/hardware/eicon/xdi_adapter.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/isdn/hardware/eicon/xdi_adapter.h 2011-11-17 20:09:04.000000000 -0500 @@ -44,7 +44,7 @@ typedef struct _xdi_mbox_t { typedef struct _diva_os_idi_adapter_interface { diva_init_card_proc_t cleanup_adapter_proc; diva_cmd_card_proc_t cmd_proc; -} diva_os_idi_adapter_interface_t; +} __no_const diva_os_idi_adapter_interface_t; typedef struct _diva_os_xdi_adapter { struct list_head link; diff -urNp linux-2.6.32.48/drivers/isdn/i4l/isdn_common.c linux-2.6.32.48-openvz/drivers/isdn/i4l/isdn_common.c --- linux-2.6.32.48/drivers/isdn/i4l/isdn_common.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/isdn/i4l/isdn_common.c 2011-11-17 20:08:53.000000000 -0500 @@ -1290,6 +1290,8 @@ isdn_ioctl(struct inode *inode, struct f } iocpar; void __user *argp = (void __user *)arg; + pax_track_stack(); + #define name iocpar.name #define bname iocpar.bname #define iocts iocpar.iocts diff -urNp linux-2.6.32.48/drivers/isdn/icn/icn.c linux-2.6.32.48-openvz/drivers/isdn/icn/icn.c --- linux-2.6.32.48/drivers/isdn/icn/icn.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/isdn/icn/icn.c 2011-11-17 20:08:53.000000000 -0500 @@ -1044,7 +1044,7 @@ icn_writecmd(const u_char * buf, int len if (count > len) count = len; if (user) { - if (copy_from_user(msg, buf, count)) + if (count > sizeof msg || copy_from_user(msg, buf, count)) return -EFAULT; } else memcpy(msg, buf, count); diff -urNp linux-2.6.32.48/drivers/isdn/mISDN/socket.c linux-2.6.32.48-openvz/drivers/isdn/mISDN/socket.c --- linux-2.6.32.48/drivers/isdn/mISDN/socket.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/isdn/mISDN/socket.c 2011-11-17 20:08:53.000000000 -0500 @@ -391,6 +391,7 @@ data_sock_ioctl(struct socket *sock, uns if (dev) { struct mISDN_devinfo di; + memset(&di, 0, sizeof(di)); di.id = dev->id; di.Dprotocols = dev->Dprotocols; di.Bprotocols = dev->Bprotocols | get_all_Bprotocols(); @@ -671,6 +672,7 @@ base_sock_ioctl(struct socket *sock, uns if (dev) { struct mISDN_devinfo di; + memset(&di, 0, sizeof(di)); di.id = dev->id; di.Dprotocols = dev->Dprotocols; di.Bprotocols = dev->Bprotocols | get_all_Bprotocols(); diff -urNp linux-2.6.32.48/drivers/isdn/sc/interrupt.c linux-2.6.32.48-openvz/drivers/isdn/sc/interrupt.c --- linux-2.6.32.48/drivers/isdn/sc/interrupt.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/isdn/sc/interrupt.c 2011-11-17 20:08:53.000000000 -0500 @@ -112,11 +112,19 @@ irqreturn_t interrupt_handler(int dummy, } else if(callid>=0x0000 && callid<=0x7FFF) { + int len; + pr_debug("%s: Got Incoming Call\n", sc_adapter[card]->devicename); - strcpy(setup.phone,&(rcvmsg.msg_data.byte_array[4])); - strcpy(setup.eazmsn, - sc_adapter[card]->channel[rcvmsg.phy_link_no-1].dn); + len = strlcpy(setup.phone, &(rcvmsg.msg_data.byte_array[4]), + sizeof(setup.phone)); + if (len >= sizeof(setup.phone)) + continue; + len = strlcpy(setup.eazmsn, + sc_adapter[card]->channel[rcvmsg.phy_link_no - 1].dn, + sizeof(setup.eazmsn)); + if (len >= sizeof(setup.eazmsn)) + continue; setup.si1 = 7; setup.si2 = 0; setup.plan = 0; @@ -176,7 +184,9 @@ irqreturn_t interrupt_handler(int dummy, * Handle a GetMyNumber Rsp */ if (IS_CE_MESSAGE(rcvmsg,Call,0,GetMyNumber)){ - strcpy(sc_adapter[card]->channel[rcvmsg.phy_link_no-1].dn,rcvmsg.msg_data.byte_array); + strlcpy(sc_adapter[card]->channel[rcvmsg.phy_link_no - 1].dn, + rcvmsg.msg_data.byte_array, + sizeof(rcvmsg.msg_data.byte_array)); continue; } diff -urNp linux-2.6.32.48/drivers/lguest/core.c linux-2.6.32.48-openvz/drivers/lguest/core.c --- linux-2.6.32.48/drivers/lguest/core.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/lguest/core.c 2011-11-17 20:08:53.000000000 -0500 @@ -91,9 +91,17 @@ static __init int map_switcher(void) * it's worked so far. The end address needs +1 because __get_vm_area * allocates an extra guard page, so we need space for that. */ + +#if defined(CONFIG_MODULES) && defined(CONFIG_X86_32) && defined(CONFIG_PAX_KERNEXEC) + switcher_vma = __get_vm_area(TOTAL_SWITCHER_PAGES * PAGE_SIZE, + VM_ALLOC | VM_KERNEXEC, SWITCHER_ADDR, SWITCHER_ADDR + + (TOTAL_SWITCHER_PAGES+1) * PAGE_SIZE); +#else switcher_vma = __get_vm_area(TOTAL_SWITCHER_PAGES * PAGE_SIZE, VM_ALLOC, SWITCHER_ADDR, SWITCHER_ADDR + (TOTAL_SWITCHER_PAGES+1) * PAGE_SIZE); +#endif + if (!switcher_vma) { err = -ENOMEM; printk("lguest: could not map switcher pages high\n"); @@ -118,7 +126,7 @@ static __init int map_switcher(void) * Now the Switcher is mapped at the right address, we can't fail! * Copy in the compiled-in Switcher code (from _switcher.S). */ - memcpy(switcher_vma->addr, start_switcher_text, + memcpy(switcher_vma->addr, ktla_ktva(start_switcher_text), end_switcher_text - start_switcher_text); printk(KERN_INFO "lguest: mapped switcher at %p\n", diff -urNp linux-2.6.32.48/drivers/lguest/x86/core.c linux-2.6.32.48-openvz/drivers/lguest/x86/core.c --- linux-2.6.32.48/drivers/lguest/x86/core.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/lguest/x86/core.c 2011-11-17 20:08:53.000000000 -0500 @@ -59,7 +59,7 @@ static struct { /* Offset from where switcher.S was compiled to where we've copied it */ static unsigned long switcher_offset(void) { - return SWITCHER_ADDR - (unsigned long)start_switcher_text; + return SWITCHER_ADDR - (unsigned long)ktla_ktva(start_switcher_text); } /* This cpu's struct lguest_pages. */ @@ -100,7 +100,13 @@ static void copy_in_guest_info(struct lg * These copies are pretty cheap, so we do them unconditionally: */ /* Save the current Host top-level page directory. */ + +#ifdef CONFIG_PAX_PER_CPU_PGD + pages->state.host_cr3 = read_cr3(); +#else pages->state.host_cr3 = __pa(current->mm->pgd); +#endif + /* * Set up the Guest's page tables to see this CPU's pages (and no * other CPU's pages). @@ -535,7 +541,7 @@ void __init lguest_arch_host_init(void) * compiled-in switcher code and the high-mapped copy we just made. */ for (i = 0; i < IDT_ENTRIES; i++) - default_idt_entries[i] += switcher_offset(); + default_idt_entries[i] = ktla_ktva(default_idt_entries[i]) + switcher_offset(); /* * Set up the Switcher's per-cpu areas. @@ -618,7 +624,7 @@ void __init lguest_arch_host_init(void) * it will be undisturbed when we switch. To change %cs and jump we * need this structure to feed to Intel's "lcall" instruction. */ - lguest_entry.offset = (long)switch_to_guest + switcher_offset(); + lguest_entry.offset = (long)ktla_ktva(switch_to_guest) + switcher_offset(); lguest_entry.segment = LGUEST_CS; /* diff -urNp linux-2.6.32.48/drivers/lguest/x86/switcher_32.S linux-2.6.32.48-openvz/drivers/lguest/x86/switcher_32.S --- linux-2.6.32.48/drivers/lguest/x86/switcher_32.S 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/lguest/x86/switcher_32.S 2011-11-17 20:08:53.000000000 -0500 @@ -87,6 +87,7 @@ #include #include #include +#include // We mark the start of the code to copy // It's placed in .text tho it's never run here @@ -149,6 +150,13 @@ ENTRY(switch_to_guest) // Changes type when we load it: damn Intel! // For after we switch over our page tables // That entry will be read-only: we'd crash. + +#ifdef CONFIG_PAX_KERNEXEC + mov %cr0, %edx + xor $X86_CR0_WP, %edx + mov %edx, %cr0 +#endif + movl $(GDT_ENTRY_TSS*8), %edx ltr %dx @@ -157,9 +165,15 @@ ENTRY(switch_to_guest) // Let's clear it again for our return. // The GDT descriptor of the Host // Points to the table after two "size" bytes - movl (LGUEST_PAGES_host_gdt_desc+2)(%eax), %edx + movl (LGUEST_PAGES_host_gdt_desc+2)(%eax), %eax // Clear "used" from type field (byte 5, bit 2) - andb $0xFD, (GDT_ENTRY_TSS*8 + 5)(%edx) + andb $0xFD, (GDT_ENTRY_TSS*8 + 5)(%eax) + +#ifdef CONFIG_PAX_KERNEXEC + mov %cr0, %eax + xor $X86_CR0_WP, %eax + mov %eax, %cr0 +#endif // Once our page table's switched, the Guest is live! // The Host fades as we run this final step. @@ -295,13 +309,12 @@ deliver_to_host: // I consulted gcc, and it gave // These instructions, which I gladly credit: leal (%edx,%ebx,8), %eax - movzwl (%eax),%edx - movl 4(%eax), %eax - xorw %ax, %ax - orl %eax, %edx + movl 4(%eax), %edx + movw (%eax), %dx // Now the address of the handler's in %edx // We call it now: its "iret" drops us home. - jmp *%edx + ljmp $__KERNEL_CS, $1f +1: jmp *%edx // Every interrupt can come to us here // But we must truly tell each apart. diff -urNp linux-2.6.32.48/drivers/macintosh/via-pmu-backlight.c linux-2.6.32.48-openvz/drivers/macintosh/via-pmu-backlight.c --- linux-2.6.32.48/drivers/macintosh/via-pmu-backlight.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/macintosh/via-pmu-backlight.c 2011-11-17 20:08:53.000000000 -0500 @@ -15,7 +15,7 @@ #define MAX_PMU_LEVEL 0xFF -static struct backlight_ops pmu_backlight_data; +static const struct backlight_ops pmu_backlight_data; static DEFINE_SPINLOCK(pmu_backlight_lock); static int sleeping, uses_pmu_bl; static u8 bl_curve[FB_BACKLIGHT_LEVELS]; @@ -115,7 +115,7 @@ static int pmu_backlight_get_brightness( return bd->props.brightness; } -static struct backlight_ops pmu_backlight_data = { +static const struct backlight_ops pmu_backlight_data = { .get_brightness = pmu_backlight_get_brightness, .update_status = pmu_backlight_update_status, diff -urNp linux-2.6.32.48/drivers/macintosh/via-pmu.c linux-2.6.32.48-openvz/drivers/macintosh/via-pmu.c --- linux-2.6.32.48/drivers/macintosh/via-pmu.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/macintosh/via-pmu.c 2011-11-17 20:08:53.000000000 -0500 @@ -2232,7 +2232,7 @@ static int pmu_sleep_valid(suspend_state && (pmac_call_feature(PMAC_FTR_SLEEP_STATE, NULL, 0, -1) >= 0); } -static struct platform_suspend_ops pmu_pm_ops = { +static const struct platform_suspend_ops pmu_pm_ops = { .enter = powerbook_sleep, .valid = pmu_sleep_valid, }; diff -urNp linux-2.6.32.48/drivers/md/dm.c linux-2.6.32.48-openvz/drivers/md/dm.c --- linux-2.6.32.48/drivers/md/dm.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/md/dm.c 2011-11-17 20:09:04.000000000 -0500 @@ -165,9 +165,9 @@ struct mapped_device { /* * Event handling. */ - atomic_t event_nr; + atomic_unchecked_t event_nr; wait_queue_head_t eventq; - atomic_t uevent_seq; + atomic_unchecked_t uevent_seq; struct list_head uevent_list; spinlock_t uevent_lock; /* Protect access to uevent_list */ @@ -1776,8 +1776,8 @@ static struct mapped_device *alloc_dev(i rwlock_init(&md->map_lock); atomic_set(&md->holders, 1); atomic_set(&md->open_count, 0); - atomic_set(&md->event_nr, 0); - atomic_set(&md->uevent_seq, 0); + atomic_set_unchecked(&md->event_nr, 0); + atomic_set_unchecked(&md->uevent_seq, 0); INIT_LIST_HEAD(&md->uevent_list); spin_lock_init(&md->uevent_lock); @@ -1927,7 +1927,7 @@ static void event_callback(void *context dm_send_uevents(&uevents, &disk_to_dev(md->disk)->kobj); - atomic_inc(&md->event_nr); + atomic_inc_unchecked(&md->event_nr); wake_up(&md->eventq); } @@ -2562,18 +2562,18 @@ void dm_kobject_uevent(struct mapped_dev uint32_t dm_next_uevent_seq(struct mapped_device *md) { - return atomic_add_return(1, &md->uevent_seq); + return atomic_add_return_unchecked(1, &md->uevent_seq); } uint32_t dm_get_event_nr(struct mapped_device *md) { - return atomic_read(&md->event_nr); + return atomic_read_unchecked(&md->event_nr); } int dm_wait_event(struct mapped_device *md, int event_nr) { return wait_event_interruptible(md->eventq, - (event_nr != atomic_read(&md->event_nr))); + (event_nr != atomic_read_unchecked(&md->event_nr))); } void dm_uevent_add(struct mapped_device *md, struct list_head *elist) diff -urNp linux-2.6.32.48/drivers/md/dm-ioctl.c linux-2.6.32.48-openvz/drivers/md/dm-ioctl.c --- linux-2.6.32.48/drivers/md/dm-ioctl.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/md/dm-ioctl.c 2011-11-17 20:08:53.000000000 -0500 @@ -1437,7 +1437,7 @@ static int validate_params(uint cmd, str cmd == DM_LIST_VERSIONS_CMD) return 0; - if ((cmd == DM_DEV_CREATE_CMD)) { + if (cmd == DM_DEV_CREATE_CMD) { if (!*param->name) { DMWARN("name not supplied when creating device"); return -EINVAL; diff -urNp linux-2.6.32.48/drivers/md/dm-raid1.c linux-2.6.32.48-openvz/drivers/md/dm-raid1.c --- linux-2.6.32.48/drivers/md/dm-raid1.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/md/dm-raid1.c 2011-11-17 20:08:53.000000000 -0500 @@ -41,7 +41,7 @@ enum dm_raid1_error { struct mirror { struct mirror_set *ms; - atomic_t error_count; + atomic_unchecked_t error_count; unsigned long error_type; struct dm_dev *dev; sector_t offset; @@ -203,7 +203,7 @@ static void fail_mirror(struct mirror *m * simple way to tell if a device has encountered * errors. */ - atomic_inc(&m->error_count); + atomic_inc_unchecked(&m->error_count); if (test_and_set_bit(error_type, &m->error_type)) return; @@ -225,7 +225,7 @@ static void fail_mirror(struct mirror *m } for (new = ms->mirror; new < ms->mirror + ms->nr_mirrors; new++) - if (!atomic_read(&new->error_count)) { + if (!atomic_read_unchecked(&new->error_count)) { set_default_mirror(new); break; } @@ -363,7 +363,7 @@ static struct mirror *choose_mirror(stru struct mirror *m = get_default_mirror(ms); do { - if (likely(!atomic_read(&m->error_count))) + if (likely(!atomic_read_unchecked(&m->error_count))) return m; if (m-- == ms->mirror) @@ -377,7 +377,7 @@ static int default_ok(struct mirror *m) { struct mirror *default_mirror = get_default_mirror(m->ms); - return !atomic_read(&default_mirror->error_count); + return !atomic_read_unchecked(&default_mirror->error_count); } static int mirror_available(struct mirror_set *ms, struct bio *bio) @@ -484,7 +484,7 @@ static void do_reads(struct mirror_set * */ if (likely(region_in_sync(ms, region, 1))) m = choose_mirror(ms, bio->bi_sector); - else if (m && atomic_read(&m->error_count)) + else if (m && atomic_read_unchecked(&m->error_count)) m = NULL; if (likely(m)) @@ -855,7 +855,7 @@ static int get_mirror(struct mirror_set } ms->mirror[mirror].ms = ms; - atomic_set(&(ms->mirror[mirror].error_count), 0); + atomic_set_unchecked(&(ms->mirror[mirror].error_count), 0); ms->mirror[mirror].error_type = 0; ms->mirror[mirror].offset = offset; @@ -1241,7 +1241,7 @@ static void mirror_resume(struct dm_targ */ static char device_status_char(struct mirror *m) { - if (!atomic_read(&(m->error_count))) + if (!atomic_read_unchecked(&(m->error_count))) return 'A'; return (test_bit(DM_RAID1_WRITE_ERROR, &(m->error_type))) ? 'D' : diff -urNp linux-2.6.32.48/drivers/md/dm-stripe.c linux-2.6.32.48-openvz/drivers/md/dm-stripe.c --- linux-2.6.32.48/drivers/md/dm-stripe.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/md/dm-stripe.c 2011-11-17 20:08:53.000000000 -0500 @@ -20,7 +20,7 @@ struct stripe { struct dm_dev *dev; sector_t physical_start; - atomic_t error_count; + atomic_unchecked_t error_count; }; struct stripe_c { @@ -188,7 +188,7 @@ static int stripe_ctr(struct dm_target * kfree(sc); return r; } - atomic_set(&(sc->stripe[i].error_count), 0); + atomic_set_unchecked(&(sc->stripe[i].error_count), 0); } ti->private = sc; @@ -257,7 +257,7 @@ static int stripe_status(struct dm_targe DMEMIT("%d ", sc->stripes); for (i = 0; i < sc->stripes; i++) { DMEMIT("%s ", sc->stripe[i].dev->name); - buffer[i] = atomic_read(&(sc->stripe[i].error_count)) ? + buffer[i] = atomic_read_unchecked(&(sc->stripe[i].error_count)) ? 'D' : 'A'; } buffer[i] = '\0'; @@ -304,8 +304,8 @@ static int stripe_end_io(struct dm_targe */ for (i = 0; i < sc->stripes; i++) if (!strcmp(sc->stripe[i].dev->name, major_minor)) { - atomic_inc(&(sc->stripe[i].error_count)); - if (atomic_read(&(sc->stripe[i].error_count)) < + atomic_inc_unchecked(&(sc->stripe[i].error_count)); + if (atomic_read_unchecked(&(sc->stripe[i].error_count)) < DM_IO_ERROR_THRESHOLD) queue_work(kstriped, &sc->kstriped_ws); } diff -urNp linux-2.6.32.48/drivers/md/dm-sysfs.c linux-2.6.32.48-openvz/drivers/md/dm-sysfs.c --- linux-2.6.32.48/drivers/md/dm-sysfs.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/md/dm-sysfs.c 2011-11-17 20:08:53.000000000 -0500 @@ -75,7 +75,7 @@ static struct attribute *dm_attrs[] = { NULL, }; -static struct sysfs_ops dm_sysfs_ops = { +static const struct sysfs_ops dm_sysfs_ops = { .show = dm_attr_show, }; diff -urNp linux-2.6.32.48/drivers/md/dm-table.c linux-2.6.32.48-openvz/drivers/md/dm-table.c --- linux-2.6.32.48/drivers/md/dm-table.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/md/dm-table.c 2011-11-17 20:09:04.000000000 -0500 @@ -376,7 +376,7 @@ static int device_area_is_invalid(struct if (!dev_size) return 0; - if ((start >= dev_size) || (start + len > dev_size)) { + if ((start >= dev_size) || (len > dev_size - start)) { DMWARN("%s: %s too small for target: " "start=%llu, len=%llu, dev_size=%llu", dm_device_name(ti->table->md), bdevname(bdev, b), diff -urNp linux-2.6.32.48/drivers/md/md.c linux-2.6.32.48-openvz/drivers/md/md.c --- linux-2.6.32.48/drivers/md/md.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/md/md.c 2011-11-17 20:09:04.000000000 -0500 @@ -153,10 +153,10 @@ static int start_readonly; * start build, activate spare */ static DECLARE_WAIT_QUEUE_HEAD(md_event_waiters); -static atomic_t md_event_count; +static atomic_unchecked_t md_event_count; void md_new_event(mddev_t *mddev) { - atomic_inc(&md_event_count); + atomic_inc_unchecked(&md_event_count); wake_up(&md_event_waiters); } EXPORT_SYMBOL_GPL(md_new_event); @@ -166,7 +166,7 @@ EXPORT_SYMBOL_GPL(md_new_event); */ static void md_new_event_inintr(mddev_t *mddev) { - atomic_inc(&md_event_count); + atomic_inc_unchecked(&md_event_count); wake_up(&md_event_waiters); } @@ -1226,7 +1226,7 @@ static int super_1_load(mdk_rdev_t *rdev rdev->preferred_minor = 0xffff; rdev->data_offset = le64_to_cpu(sb->data_offset); - atomic_set(&rdev->corrected_errors, le32_to_cpu(sb->cnt_corrected_read)); + atomic_set_unchecked(&rdev->corrected_errors, le32_to_cpu(sb->cnt_corrected_read)); rdev->sb_size = le32_to_cpu(sb->max_dev) * 2 + 256; bmask = queue_logical_block_size(rdev->bdev->bd_disk->queue)-1; @@ -1400,7 +1400,7 @@ static void super_1_sync(mddev_t *mddev, else sb->resync_offset = cpu_to_le64(0); - sb->cnt_corrected_read = cpu_to_le32(atomic_read(&rdev->corrected_errors)); + sb->cnt_corrected_read = cpu_to_le32(atomic_read_unchecked(&rdev->corrected_errors)); sb->raid_disks = cpu_to_le32(mddev->raid_disks); sb->size = cpu_to_le64(mddev->dev_sectors); @@ -2222,7 +2222,7 @@ __ATTR(state, S_IRUGO|S_IWUSR, state_sho static ssize_t errors_show(mdk_rdev_t *rdev, char *page) { - return sprintf(page, "%d\n", atomic_read(&rdev->corrected_errors)); + return sprintf(page, "%d\n", atomic_read_unchecked(&rdev->corrected_errors)); } static ssize_t @@ -2231,7 +2231,7 @@ errors_store(mdk_rdev_t *rdev, const cha char *e; unsigned long n = simple_strtoul(buf, &e, 10); if (*buf && (*e == 0 || *e == '\n')) { - atomic_set(&rdev->corrected_errors, n); + atomic_set_unchecked(&rdev->corrected_errors, n); return len; } return -EINVAL; @@ -2525,7 +2525,7 @@ static void rdev_free(struct kobject *ko mdk_rdev_t *rdev = container_of(ko, mdk_rdev_t, kobj); kfree(rdev); } -static struct sysfs_ops rdev_sysfs_ops = { +static const struct sysfs_ops rdev_sysfs_ops = { .show = rdev_attr_show, .store = rdev_attr_store, }; @@ -2574,8 +2574,8 @@ static mdk_rdev_t *md_import_device(dev_ rdev->data_offset = 0; rdev->sb_events = 0; atomic_set(&rdev->nr_pending, 0); - atomic_set(&rdev->read_errors, 0); - atomic_set(&rdev->corrected_errors, 0); + atomic_set_unchecked(&rdev->read_errors, 0); + atomic_set_unchecked(&rdev->corrected_errors, 0); size = rdev->bdev->bd_inode->i_size >> BLOCK_SIZE_BITS; if (!size) { @@ -3895,7 +3895,7 @@ static void md_free(struct kobject *ko) kfree(mddev); } -static struct sysfs_ops md_sysfs_ops = { +static const struct sysfs_ops md_sysfs_ops = { .show = md_attr_show, .store = md_attr_store, }; @@ -4482,7 +4482,8 @@ out: err = 0; blk_integrity_unregister(disk); md_new_event(mddev); - sysfs_notify_dirent(mddev->sysfs_state); + if (mddev->sysfs_state) + sysfs_notify_dirent(mddev->sysfs_state); return err; } @@ -5962,7 +5963,7 @@ static int md_seq_show(struct seq_file * spin_unlock(&pers_lock); seq_printf(seq, "\n"); - mi->event = atomic_read(&md_event_count); + mi->event = atomic_read_unchecked(&md_event_count); return 0; } if (v == (void*)2) { @@ -6051,7 +6052,7 @@ static int md_seq_show(struct seq_file * chunk_kb ? "KB" : "B"); if (bitmap->file) { seq_printf(seq, ", file: "); - seq_path(seq, &bitmap->file->f_path, " \t\n"); + seq_path(seq, &bitmap->file->f_path, " \t\n\\"); } seq_printf(seq, "\n"); @@ -6085,7 +6086,7 @@ static int md_seq_open(struct inode *ino else { struct seq_file *p = file->private_data; p->private = mi; - mi->event = atomic_read(&md_event_count); + mi->event = atomic_read_unchecked(&md_event_count); } return error; } @@ -6101,7 +6102,7 @@ static unsigned int mdstat_poll(struct f /* always allow read */ mask = POLLIN | POLLRDNORM; - if (mi->event != atomic_read(&md_event_count)) + if (mi->event != atomic_read_unchecked(&md_event_count)) mask |= POLLERR | POLLPRI; return mask; } @@ -6145,7 +6146,7 @@ static int is_mddev_idle(mddev_t *mddev, struct gendisk *disk = rdev->bdev->bd_contains->bd_disk; curr_events = (int)part_stat_read(&disk->part0, sectors[0]) + (int)part_stat_read(&disk->part0, sectors[1]) - - atomic_read(&disk->sync_io); + atomic_read_unchecked(&disk->sync_io); /* sync IO will cause sync_io to increase before the disk_stats * as sync_io is counted when a request starts, and * disk_stats is counted when it completes. diff -urNp linux-2.6.32.48/drivers/md/md.h linux-2.6.32.48-openvz/drivers/md/md.h --- linux-2.6.32.48/drivers/md/md.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/md/md.h 2011-11-17 20:08:53.000000000 -0500 @@ -94,10 +94,10 @@ struct mdk_rdev_s * only maintained for arrays that * support hot removal */ - atomic_t read_errors; /* number of consecutive read errors that + atomic_unchecked_t read_errors; /* number of consecutive read errors that * we have tried to ignore. */ - atomic_t corrected_errors; /* number of corrected read errors, + atomic_unchecked_t corrected_errors; /* number of corrected read errors, * for reporting to userspace and storing * in superblock. */ @@ -304,7 +304,7 @@ static inline void rdev_dec_pending(mdk_ static inline void md_sync_acct(struct block_device *bdev, unsigned long nr_sectors) { - atomic_add(nr_sectors, &bdev->bd_contains->bd_disk->sync_io); + atomic_add_unchecked(nr_sectors, &bdev->bd_contains->bd_disk->sync_io); } struct mdk_personality diff -urNp linux-2.6.32.48/drivers/md/raid10.c linux-2.6.32.48-openvz/drivers/md/raid10.c --- linux-2.6.32.48/drivers/md/raid10.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/md/raid10.c 2011-11-17 20:08:53.000000000 -0500 @@ -1255,7 +1255,7 @@ static void end_sync_read(struct bio *bi if (test_bit(BIO_UPTODATE, &bio->bi_flags)) set_bit(R10BIO_Uptodate, &r10_bio->state); else { - atomic_add(r10_bio->sectors, + atomic_add_unchecked(r10_bio->sectors, &conf->mirrors[d].rdev->corrected_errors); if (!test_bit(MD_RECOVERY_SYNC, &conf->mddev->recovery)) md_error(r10_bio->mddev, @@ -1520,7 +1520,7 @@ static void fix_read_error(conf_t *conf, test_bit(In_sync, &rdev->flags)) { atomic_inc(&rdev->nr_pending); rcu_read_unlock(); - atomic_add(s, &rdev->corrected_errors); + atomic_add_unchecked(s, &rdev->corrected_errors); if (sync_page_io(rdev->bdev, r10_bio->devs[sl].addr + sect + rdev->data_offset, diff -urNp linux-2.6.32.48/drivers/md/raid1.c linux-2.6.32.48-openvz/drivers/md/raid1.c --- linux-2.6.32.48/drivers/md/raid1.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/md/raid1.c 2011-11-17 20:08:53.000000000 -0500 @@ -1415,7 +1415,7 @@ static void sync_request_write(mddev_t * if (r1_bio->bios[d]->bi_end_io != end_sync_read) continue; rdev = conf->mirrors[d].rdev; - atomic_add(s, &rdev->corrected_errors); + atomic_add_unchecked(s, &rdev->corrected_errors); if (sync_page_io(rdev->bdev, sect + rdev->data_offset, s<<9, @@ -1564,7 +1564,7 @@ static void fix_read_error(conf_t *conf, /* Well, this device is dead */ md_error(mddev, rdev); else { - atomic_add(s, &rdev->corrected_errors); + atomic_add_unchecked(s, &rdev->corrected_errors); printk(KERN_INFO "raid1:%s: read error corrected " "(%d sectors at %llu on %s)\n", diff -urNp linux-2.6.32.48/drivers/md/raid5.c linux-2.6.32.48-openvz/drivers/md/raid5.c --- linux-2.6.32.48/drivers/md/raid5.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/md/raid5.c 2011-11-17 20:09:04.000000000 -0500 @@ -482,7 +482,7 @@ static void ops_run_io(struct stripe_hea bi->bi_next = NULL; if ((rw & WRITE) && test_bit(R5_ReWrite, &sh->dev[i].flags)) - atomic_add(STRIPE_SECTORS, + atomic_add_unchecked(STRIPE_SECTORS, &rdev->corrected_errors); generic_make_request(bi); } else { @@ -1517,15 +1517,15 @@ static void raid5_end_read_request(struc clear_bit(R5_ReadError, &sh->dev[i].flags); clear_bit(R5_ReWrite, &sh->dev[i].flags); } - if (atomic_read(&conf->disks[i].rdev->read_errors)) - atomic_set(&conf->disks[i].rdev->read_errors, 0); + if (atomic_read_unchecked(&conf->disks[i].rdev->read_errors)) + atomic_set_unchecked(&conf->disks[i].rdev->read_errors, 0); } else { const char *bdn = bdevname(conf->disks[i].rdev->bdev, b); int retry = 0; rdev = conf->disks[i].rdev; clear_bit(R5_UPTODATE, &sh->dev[i].flags); - atomic_inc(&rdev->read_errors); + atomic_inc_unchecked(&rdev->read_errors); if (conf->mddev->degraded >= conf->max_degraded) printk_rl(KERN_WARNING "raid5:%s: read error not correctable " @@ -1543,7 +1543,7 @@ static void raid5_end_read_request(struc (unsigned long long)(sh->sector + rdev->data_offset), bdn); - else if (atomic_read(&rdev->read_errors) + else if (atomic_read_unchecked(&rdev->read_errors) > conf->max_nr_stripes) printk(KERN_WARNING "raid5:%s: Too many read errors, failing device %s.\n", @@ -1870,6 +1870,7 @@ static sector_t compute_blocknr(struct s sector_t r_sector; struct stripe_head sh2; + pax_track_stack(); chunk_offset = sector_div(new_sector, sectors_per_chunk); stripe = new_sector; diff -urNp linux-2.6.32.48/drivers/media/common/saa7146_hlp.c linux-2.6.32.48-openvz/drivers/media/common/saa7146_hlp.c --- linux-2.6.32.48/drivers/media/common/saa7146_hlp.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/media/common/saa7146_hlp.c 2011-11-17 20:08:53.000000000 -0500 @@ -353,6 +353,8 @@ static void calculate_clipping_registers int x[32], y[32], w[32], h[32]; + pax_track_stack(); + /* clear out memory */ memset(&line_list[0], 0x00, sizeof(u32)*32); memset(&pixel_list[0], 0x00, sizeof(u32)*32); diff -urNp linux-2.6.32.48/drivers/media/dvb/dvb-core/dvb_ca_en50221.c linux-2.6.32.48-openvz/drivers/media/dvb/dvb-core/dvb_ca_en50221.c --- linux-2.6.32.48/drivers/media/dvb/dvb-core/dvb_ca_en50221.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/media/dvb/dvb-core/dvb_ca_en50221.c 2011-11-17 20:08:53.000000000 -0500 @@ -590,6 +590,8 @@ static int dvb_ca_en50221_read_data(stru u8 buf[HOST_LINK_BUF_SIZE]; int i; + pax_track_stack(); + dprintk("%s\n", __func__); /* check if we have space for a link buf in the rx_buffer */ @@ -1285,6 +1287,8 @@ static ssize_t dvb_ca_en50221_io_write(s unsigned long timeout; int written; + pax_track_stack(); + dprintk("%s\n", __func__); /* Incoming packet has a 2 byte header. hdr[0] = slot_id, hdr[1] = connection_id */ diff -urNp linux-2.6.32.48/drivers/media/dvb/dvb-core/dvb_demux.h linux-2.6.32.48-openvz/drivers/media/dvb/dvb-core/dvb_demux.h --- linux-2.6.32.48/drivers/media/dvb/dvb-core/dvb_demux.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/media/dvb/dvb-core/dvb_demux.h 2011-11-17 20:09:04.000000000 -0500 @@ -71,7 +71,7 @@ struct dvb_demux_feed { union { dmx_ts_cb ts; dmx_section_cb sec; - } cb; + } __no_const cb; struct dvb_demux *demux; void *priv; diff -urNp linux-2.6.32.48/drivers/media/dvb/dvb-core/dvbdev.c linux-2.6.32.48-openvz/drivers/media/dvb/dvb-core/dvbdev.c --- linux-2.6.32.48/drivers/media/dvb/dvb-core/dvbdev.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/media/dvb/dvb-core/dvbdev.c 2011-11-17 20:09:04.000000000 -0500 @@ -191,7 +191,7 @@ int dvb_register_device(struct dvb_adapt const struct dvb_device *template, void *priv, int type) { struct dvb_device *dvbdev; - struct file_operations *dvbdevfops; + file_operations_no_const *dvbdevfops; struct device *clsdev; int minor; int id; diff -urNp linux-2.6.32.48/drivers/media/dvb/dvb-usb/cxusb.c linux-2.6.32.48-openvz/drivers/media/dvb/dvb-usb/cxusb.c --- linux-2.6.32.48/drivers/media/dvb/dvb-usb/cxusb.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/media/dvb/dvb-usb/cxusb.c 2011-11-17 20:09:04.000000000 -0500 @@ -1040,7 +1040,7 @@ static struct dib0070_config dib7070p_di struct dib0700_adapter_state { int (*set_param_save) (struct dvb_frontend *, struct dvb_frontend_parameters *); -}; +} __no_const; static int dib7070_set_param_override(struct dvb_frontend *fe, struct dvb_frontend_parameters *fep) diff -urNp linux-2.6.32.48/drivers/media/dvb/dvb-usb/dib0700_core.c linux-2.6.32.48-openvz/drivers/media/dvb/dvb-usb/dib0700_core.c --- linux-2.6.32.48/drivers/media/dvb/dvb-usb/dib0700_core.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/media/dvb/dvb-usb/dib0700_core.c 2011-11-17 20:08:53.000000000 -0500 @@ -332,6 +332,8 @@ int dib0700_download_firmware(struct usb u8 buf[260]; + pax_track_stack(); + while ((ret = dvb_usb_get_hexline(fw, &hx, &pos)) > 0) { deb_fwdata("writing to address 0x%08x (buffer: 0x%02x %02x)\n",hx.addr, hx.len, hx.chk); diff -urNp linux-2.6.32.48/drivers/media/dvb/dvb-usb/dib0700_devices.c linux-2.6.32.48-openvz/drivers/media/dvb/dvb-usb/dib0700_devices.c --- linux-2.6.32.48/drivers/media/dvb/dvb-usb/dib0700_devices.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/media/dvb/dvb-usb/dib0700_devices.c 2011-11-17 20:09:04.000000000 -0500 @@ -28,7 +28,7 @@ MODULE_PARM_DESC(force_lna_activation, " struct dib0700_adapter_state { int (*set_param_save) (struct dvb_frontend *, struct dvb_frontend_parameters *); -}; +} __no_const; /* Hauppauge Nova-T 500 (aka Bristol) * has a LNA on GPIO0 which is enabled by setting 1 */ diff -urNp linux-2.6.32.48/drivers/media/dvb/frontends/dib3000.h linux-2.6.32.48-openvz/drivers/media/dvb/frontends/dib3000.h --- linux-2.6.32.48/drivers/media/dvb/frontends/dib3000.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/media/dvb/frontends/dib3000.h 2011-11-17 20:09:04.000000000 -0500 @@ -39,7 +39,7 @@ struct dib_fe_xfer_ops int (*fifo_ctrl)(struct dvb_frontend *fe, int onoff); int (*pid_ctrl)(struct dvb_frontend *fe, int index, int pid, int onoff); int (*tuner_pass_ctrl)(struct dvb_frontend *fe, int onoff, u8 pll_ctrl); -}; +} __no_const; #if defined(CONFIG_DVB_DIB3000MB) || (defined(CONFIG_DVB_DIB3000MB_MODULE) && defined(MODULE)) extern struct dvb_frontend* dib3000mb_attach(const struct dib3000_config* config, diff -urNp linux-2.6.32.48/drivers/media/dvb/frontends/or51211.c linux-2.6.32.48-openvz/drivers/media/dvb/frontends/or51211.c --- linux-2.6.32.48/drivers/media/dvb/frontends/or51211.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/media/dvb/frontends/or51211.c 2011-11-17 20:08:53.000000000 -0500 @@ -113,6 +113,8 @@ static int or51211_load_firmware (struct u8 tudata[585]; int i; + pax_track_stack(); + dprintk("Firmware is %zd bytes\n",fw->size); /* Get eprom data */ diff -urNp linux-2.6.32.48/drivers/media/radio/radio-cadet.c linux-2.6.32.48-openvz/drivers/media/radio/radio-cadet.c --- linux-2.6.32.48/drivers/media/radio/radio-cadet.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/media/radio/radio-cadet.c 2011-11-17 20:08:53.000000000 -0500 @@ -347,7 +347,7 @@ static ssize_t cadet_read(struct file *f while (i < count && dev->rdsin != dev->rdsout) readbuf[i++] = dev->rdsbuf[dev->rdsout++]; - if (copy_to_user(data, readbuf, i)) + if (i > sizeof readbuf || copy_to_user(data, readbuf, i)) return -EFAULT; return i; } diff -urNp linux-2.6.32.48/drivers/media/video/cx18/cx18-driver.c linux-2.6.32.48-openvz/drivers/media/video/cx18/cx18-driver.c --- linux-2.6.32.48/drivers/media/video/cx18/cx18-driver.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/media/video/cx18/cx18-driver.c 2011-11-17 20:08:53.000000000 -0500 @@ -56,7 +56,7 @@ static struct pci_device_id cx18_pci_tbl MODULE_DEVICE_TABLE(pci, cx18_pci_tbl); -static atomic_t cx18_instance = ATOMIC_INIT(0); +static atomic_unchecked_t cx18_instance = ATOMIC_INIT(0); /* Parameter declarations */ static int cardtype[CX18_MAX_CARDS]; @@ -288,6 +288,8 @@ void cx18_read_eeprom(struct cx18 *cx, s struct i2c_client c; u8 eedata[256]; + pax_track_stack(); + memset(&c, 0, sizeof(c)); strlcpy(c.name, "cx18 tveeprom tmp", sizeof(c.name)); c.adapter = &cx->i2c_adap[0]; @@ -800,7 +802,7 @@ static int __devinit cx18_probe(struct p struct cx18 *cx; /* FIXME - module parameter arrays constrain max instances */ - i = atomic_inc_return(&cx18_instance) - 1; + i = atomic_inc_return_unchecked(&cx18_instance) - 1; if (i >= CX18_MAX_CARDS) { printk(KERN_ERR "cx18: cannot manage card %d, driver has a " "limit of 0 - %d\n", i, CX18_MAX_CARDS - 1); diff -urNp linux-2.6.32.48/drivers/media/video/ivtv/ivtv-driver.c linux-2.6.32.48-openvz/drivers/media/video/ivtv/ivtv-driver.c --- linux-2.6.32.48/drivers/media/video/ivtv/ivtv-driver.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/media/video/ivtv/ivtv-driver.c 2011-11-17 20:08:53.000000000 -0500 @@ -79,7 +79,7 @@ static struct pci_device_id ivtv_pci_tbl MODULE_DEVICE_TABLE(pci,ivtv_pci_tbl); /* ivtv instance counter */ -static atomic_t ivtv_instance = ATOMIC_INIT(0); +static atomic_unchecked_t ivtv_instance = ATOMIC_INIT(0); /* Parameter declarations */ static int cardtype[IVTV_MAX_CARDS]; diff -urNp linux-2.6.32.48/drivers/media/video/omap24xxcam.c linux-2.6.32.48-openvz/drivers/media/video/omap24xxcam.c --- linux-2.6.32.48/drivers/media/video/omap24xxcam.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/media/video/omap24xxcam.c 2011-11-17 20:08:53.000000000 -0500 @@ -401,7 +401,7 @@ static void omap24xxcam_vbq_complete(str spin_unlock_irqrestore(&cam->core_enable_disable_lock, flags); do_gettimeofday(&vb->ts); - vb->field_count = atomic_add_return(2, &fh->field_count); + vb->field_count = atomic_add_return_unchecked(2, &fh->field_count); if (csr & csr_error) { vb->state = VIDEOBUF_ERROR; if (!atomic_read(&fh->cam->in_reset)) { diff -urNp linux-2.6.32.48/drivers/media/video/omap24xxcam.h linux-2.6.32.48-openvz/drivers/media/video/omap24xxcam.h --- linux-2.6.32.48/drivers/media/video/omap24xxcam.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/media/video/omap24xxcam.h 2011-11-17 20:08:53.000000000 -0500 @@ -533,7 +533,7 @@ struct omap24xxcam_fh { spinlock_t vbq_lock; /* spinlock for the videobuf queue */ struct videobuf_queue vbq; struct v4l2_pix_format pix; /* serialise pix by vbq->lock */ - atomic_t field_count; /* field counter for videobuf_buffer */ + atomic_unchecked_t field_count; /* field counter for videobuf_buffer */ /* accessing cam here doesn't need serialisation: it's constant */ struct omap24xxcam_device *cam; }; diff -urNp linux-2.6.32.48/drivers/media/video/pvrusb2/pvrusb2-eeprom.c linux-2.6.32.48-openvz/drivers/media/video/pvrusb2/pvrusb2-eeprom.c --- linux-2.6.32.48/drivers/media/video/pvrusb2/pvrusb2-eeprom.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/media/video/pvrusb2/pvrusb2-eeprom.c 2011-11-17 20:08:53.000000000 -0500 @@ -119,6 +119,8 @@ int pvr2_eeprom_analyze(struct pvr2_hdw u8 *eeprom; struct tveeprom tvdata; + pax_track_stack(); + memset(&tvdata,0,sizeof(tvdata)); eeprom = pvr2_eeprom_fetch(hdw); diff -urNp linux-2.6.32.48/drivers/media/video/pvrusb2/pvrusb2-hdw-internal.h linux-2.6.32.48-openvz/drivers/media/video/pvrusb2/pvrusb2-hdw-internal.h --- linux-2.6.32.48/drivers/media/video/pvrusb2/pvrusb2-hdw-internal.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/media/video/pvrusb2/pvrusb2-hdw-internal.h 2011-11-17 20:09:04.000000000 -0500 @@ -195,7 +195,7 @@ struct pvr2_hdw { /* I2C stuff */ struct i2c_adapter i2c_adap; - struct i2c_algorithm i2c_algo; + i2c_algorithm_no_const i2c_algo; pvr2_i2c_func i2c_func[PVR2_I2C_FUNC_CNT]; int i2c_cx25840_hack_state; int i2c_linked; diff -urNp linux-2.6.32.48/drivers/media/video/saa7134/saa6752hs.c linux-2.6.32.48-openvz/drivers/media/video/saa7134/saa6752hs.c --- linux-2.6.32.48/drivers/media/video/saa7134/saa6752hs.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/media/video/saa7134/saa6752hs.c 2011-11-17 20:08:53.000000000 -0500 @@ -683,6 +683,8 @@ static int saa6752hs_init(struct v4l2_su unsigned char localPAT[256]; unsigned char localPMT[256]; + pax_track_stack(); + /* Set video format - must be done first as it resets other settings */ set_reg8(client, 0x41, h->video_format); diff -urNp linux-2.6.32.48/drivers/media/video/saa7164/saa7164-cmd.c linux-2.6.32.48-openvz/drivers/media/video/saa7164/saa7164-cmd.c --- linux-2.6.32.48/drivers/media/video/saa7164/saa7164-cmd.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/media/video/saa7164/saa7164-cmd.c 2011-11-17 20:08:53.000000000 -0500 @@ -87,6 +87,8 @@ int saa7164_irq_dequeue(struct saa7164_d wait_queue_head_t *q = 0; dprintk(DBGLVL_CMD, "%s()\n", __func__); + pax_track_stack(); + /* While any outstand message on the bus exists... */ do { @@ -126,6 +128,8 @@ int saa7164_cmd_dequeue(struct saa7164_d u8 tmp[512]; dprintk(DBGLVL_CMD, "%s()\n", __func__); + pax_track_stack(); + while (loop) { tmComResInfo_t tRsp = { 0, 0, 0, 0, 0, 0 }; diff -urNp linux-2.6.32.48/drivers/media/video/usbvideo/ibmcam.c linux-2.6.32.48-openvz/drivers/media/video/usbvideo/ibmcam.c --- linux-2.6.32.48/drivers/media/video/usbvideo/ibmcam.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/media/video/usbvideo/ibmcam.c 2011-11-17 20:09:04.000000000 -0500 @@ -3947,15 +3947,15 @@ static struct usb_device_id id_table[] = static int __init ibmcam_init(void) { struct usbvideo_cb cbTbl; - memset(&cbTbl, 0, sizeof(cbTbl)); - cbTbl.probe = ibmcam_probe; - cbTbl.setupOnOpen = ibmcam_setup_on_open; - cbTbl.videoStart = ibmcam_video_start; - cbTbl.videoStop = ibmcam_video_stop; - cbTbl.processData = ibmcam_ProcessIsocData; - cbTbl.postProcess = usbvideo_DeinterlaceFrame; - cbTbl.adjustPicture = ibmcam_adjust_picture; - cbTbl.getFPS = ibmcam_calculate_fps; + memset((void *)&cbTbl, 0, sizeof(cbTbl)); + *(void **)&cbTbl.probe = ibmcam_probe; + *(void **)&cbTbl.setupOnOpen = ibmcam_setup_on_open; + *(void **)&cbTbl.videoStart = ibmcam_video_start; + *(void **)&cbTbl.videoStop = ibmcam_video_stop; + *(void **)&cbTbl.processData = ibmcam_ProcessIsocData; + *(void **)&cbTbl.postProcess = usbvideo_DeinterlaceFrame; + *(void **)&cbTbl.adjustPicture = ibmcam_adjust_picture; + *(void **)&cbTbl.getFPS = ibmcam_calculate_fps; return usbvideo_register( &cams, MAX_IBMCAM, diff -urNp linux-2.6.32.48/drivers/media/video/usbvideo/konicawc.c linux-2.6.32.48-openvz/drivers/media/video/usbvideo/konicawc.c --- linux-2.6.32.48/drivers/media/video/usbvideo/konicawc.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/media/video/usbvideo/konicawc.c 2011-11-17 20:09:04.000000000 -0500 @@ -225,7 +225,7 @@ static void konicawc_register_input(stru int error; usb_make_path(dev, cam->input_physname, sizeof(cam->input_physname)); - strncat(cam->input_physname, "/input0", sizeof(cam->input_physname)); + strlcat(cam->input_physname, "/input0", sizeof(cam->input_physname)); cam->input = input_dev = input_allocate_device(); if (!input_dev) { @@ -935,16 +935,16 @@ static int __init konicawc_init(void) struct usbvideo_cb cbTbl; printk(KERN_INFO KBUILD_MODNAME ": " DRIVER_VERSION ":" DRIVER_DESC "\n"); - memset(&cbTbl, 0, sizeof(cbTbl)); - cbTbl.probe = konicawc_probe; - cbTbl.setupOnOpen = konicawc_setup_on_open; - cbTbl.processData = konicawc_process_isoc; - cbTbl.getFPS = konicawc_calculate_fps; - cbTbl.setVideoMode = konicawc_set_video_mode; - cbTbl.startDataPump = konicawc_start_data; - cbTbl.stopDataPump = konicawc_stop_data; - cbTbl.adjustPicture = konicawc_adjust_picture; - cbTbl.userFree = konicawc_free_uvd; + memset((void * )&cbTbl, 0, sizeof(cbTbl)); + *(void **)&cbTbl.probe = konicawc_probe; + *(void **)&cbTbl.setupOnOpen = konicawc_setup_on_open; + *(void **)&cbTbl.processData = konicawc_process_isoc; + *(void **)&cbTbl.getFPS = konicawc_calculate_fps; + *(void **)&cbTbl.setVideoMode = konicawc_set_video_mode; + *(void **)&cbTbl.startDataPump = konicawc_start_data; + *(void **)&cbTbl.stopDataPump = konicawc_stop_data; + *(void **)&cbTbl.adjustPicture = konicawc_adjust_picture; + *(void **)&cbTbl.userFree = konicawc_free_uvd; return usbvideo_register( &cams, MAX_CAMERAS, diff -urNp linux-2.6.32.48/drivers/media/video/usbvideo/quickcam_messenger.c linux-2.6.32.48-openvz/drivers/media/video/usbvideo/quickcam_messenger.c --- linux-2.6.32.48/drivers/media/video/usbvideo/quickcam_messenger.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/media/video/usbvideo/quickcam_messenger.c 2011-11-17 20:08:53.000000000 -0500 @@ -89,7 +89,7 @@ static void qcm_register_input(struct qc int error; usb_make_path(dev, cam->input_physname, sizeof(cam->input_physname)); - strncat(cam->input_physname, "/input0", sizeof(cam->input_physname)); + strlcat(cam->input_physname, "/input0", sizeof(cam->input_physname)); cam->input = input_dev = input_allocate_device(); if (!input_dev) { diff -urNp linux-2.6.32.48/drivers/media/video/usbvideo/ultracam.c linux-2.6.32.48-openvz/drivers/media/video/usbvideo/ultracam.c --- linux-2.6.32.48/drivers/media/video/usbvideo/ultracam.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/media/video/usbvideo/ultracam.c 2011-11-17 20:09:04.000000000 -0500 @@ -655,14 +655,14 @@ static int __init ultracam_init(void) { struct usbvideo_cb cbTbl; memset(&cbTbl, 0, sizeof(cbTbl)); - cbTbl.probe = ultracam_probe; - cbTbl.setupOnOpen = ultracam_setup_on_open; - cbTbl.videoStart = ultracam_video_start; - cbTbl.videoStop = ultracam_video_stop; - cbTbl.processData = ultracam_ProcessIsocData; - cbTbl.postProcess = usbvideo_DeinterlaceFrame; - cbTbl.adjustPicture = ultracam_adjust_picture; - cbTbl.getFPS = ultracam_calculate_fps; + *(void **)&cbTbl.probe = ultracam_probe; + *(void **)&cbTbl.setupOnOpen = ultracam_setup_on_open; + *(void **)&cbTbl.videoStart = ultracam_video_start; + *(void **)&cbTbl.videoStop = ultracam_video_stop; + *(void **)&cbTbl.processData = ultracam_ProcessIsocData; + *(void **)&cbTbl.postProcess = usbvideo_DeinterlaceFrame; + *(void **)&cbTbl.adjustPicture = ultracam_adjust_picture; + *(void **)&cbTbl.getFPS = ultracam_calculate_fps; return usbvideo_register( &cams, MAX_CAMERAS, diff -urNp linux-2.6.32.48/drivers/media/video/usbvideo/usbvideo.c linux-2.6.32.48-openvz/drivers/media/video/usbvideo/usbvideo.c --- linux-2.6.32.48/drivers/media/video/usbvideo/usbvideo.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/media/video/usbvideo/usbvideo.c 2011-11-17 20:09:04.000000000 -0500 @@ -697,15 +697,15 @@ int usbvideo_register( __func__, cams, base_size, num_cams); /* Copy callbacks, apply defaults for those that are not set */ - memmove(&cams->cb, cbTbl, sizeof(cams->cb)); + memmove((void *)&cams->cb, cbTbl, sizeof(cams->cb)); if (cams->cb.getFrame == NULL) - cams->cb.getFrame = usbvideo_GetFrame; + *(void **)&cams->cb.getFrame = usbvideo_GetFrame; if (cams->cb.disconnect == NULL) - cams->cb.disconnect = usbvideo_Disconnect; + *(void **)&cams->cb.disconnect = usbvideo_Disconnect; if (cams->cb.startDataPump == NULL) - cams->cb.startDataPump = usbvideo_StartDataPump; + *(void **)&cams->cb.startDataPump = usbvideo_StartDataPump; if (cams->cb.stopDataPump == NULL) - cams->cb.stopDataPump = usbvideo_StopDataPump; + *(void **)&cams->cb.stopDataPump = usbvideo_StopDataPump; cams->num_cameras = num_cams; cams->cam = (struct uvd *) &cams[1]; diff -urNp linux-2.6.32.48/drivers/media/video/usbvideo/usbvideo.h linux-2.6.32.48-openvz/drivers/media/video/usbvideo/usbvideo.h --- linux-2.6.32.48/drivers/media/video/usbvideo/usbvideo.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/media/video/usbvideo/usbvideo.h 2011-11-17 20:09:04.000000000 -0500 @@ -268,7 +268,7 @@ struct usbvideo_cb { int (*startDataPump)(struct uvd *uvd); void (*stopDataPump)(struct uvd *uvd); int (*setVideoMode)(struct uvd *uvd, struct video_window *vw); -}; +} __no_const; struct usbvideo { int num_cameras; /* As allocated */ diff -urNp linux-2.6.32.48/drivers/media/video/usbvision/usbvision-core.c linux-2.6.32.48-openvz/drivers/media/video/usbvision/usbvision-core.c --- linux-2.6.32.48/drivers/media/video/usbvision/usbvision-core.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/media/video/usbvision/usbvision-core.c 2011-11-17 20:08:53.000000000 -0500 @@ -820,6 +820,8 @@ static enum ParseState usbvision_parse_c unsigned char rv, gv, bv; static unsigned char *Y, *U, *V; + pax_track_stack(); + frame = usbvision->curFrame; imageSize = frame->frmwidth * frame->frmheight; if ( (frame->v4l2_format.format == V4L2_PIX_FMT_YUV422P) || diff -urNp linux-2.6.32.48/drivers/media/video/v4l2-device.c linux-2.6.32.48-openvz/drivers/media/video/v4l2-device.c --- linux-2.6.32.48/drivers/media/video/v4l2-device.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/media/video/v4l2-device.c 2011-11-17 20:08:53.000000000 -0500 @@ -50,9 +50,9 @@ int v4l2_device_register(struct device * EXPORT_SYMBOL_GPL(v4l2_device_register); int v4l2_device_set_name(struct v4l2_device *v4l2_dev, const char *basename, - atomic_t *instance) + atomic_unchecked_t *instance) { - int num = atomic_inc_return(instance) - 1; + int num = atomic_inc_return_unchecked(instance) - 1; int len = strlen(basename); if (basename[len - 1] >= '0' && basename[len - 1] <= '9') diff -urNp linux-2.6.32.48/drivers/media/video/videobuf-dma-sg.c linux-2.6.32.48-openvz/drivers/media/video/videobuf-dma-sg.c --- linux-2.6.32.48/drivers/media/video/videobuf-dma-sg.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/media/video/videobuf-dma-sg.c 2011-11-17 20:08:53.000000000 -0500 @@ -693,6 +693,8 @@ void *videobuf_sg_alloc(size_t size) { struct videobuf_queue q; + pax_track_stack(); + /* Required to make generic handler to call __videobuf_alloc */ q.int_ops = &sg_ops; diff -urNp linux-2.6.32.48/drivers/message/fusion/mptbase.c linux-2.6.32.48-openvz/drivers/message/fusion/mptbase.c --- linux-2.6.32.48/drivers/message/fusion/mptbase.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/message/fusion/mptbase.c 2011-11-17 20:08:53.000000000 -0500 @@ -6709,8 +6709,14 @@ procmpt_iocinfo_read(char *buf, char **s len += sprintf(buf+len, " MaxChainDepth = 0x%02x frames\n", ioc->facts.MaxChainDepth); len += sprintf(buf+len, " MinBlockSize = 0x%02x bytes\n", 4*ioc->facts.BlockSize); +#ifdef CONFIG_GRKERNSEC_HIDESYM + len += sprintf(buf+len, " RequestFrames @ 0x%p (Dma @ 0x%p)\n", + NULL, NULL); +#else len += sprintf(buf+len, " RequestFrames @ 0x%p (Dma @ 0x%p)\n", (void *)ioc->req_frames, (void *)(ulong)ioc->req_frames_dma); +#endif + /* * Rounding UP to nearest 4-kB boundary here... */ diff -urNp linux-2.6.32.48/drivers/message/fusion/mptsas.c linux-2.6.32.48-openvz/drivers/message/fusion/mptsas.c --- linux-2.6.32.48/drivers/message/fusion/mptsas.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/message/fusion/mptsas.c 2011-11-17 20:08:53.000000000 -0500 @@ -436,6 +436,23 @@ mptsas_is_end_device(struct mptsas_devin return 0; } +static inline void +mptsas_set_rphy(MPT_ADAPTER *ioc, struct mptsas_phyinfo *phy_info, struct sas_rphy *rphy) +{ + if (phy_info->port_details) { + phy_info->port_details->rphy = rphy; + dsaswideprintk(ioc, printk(MYIOC_s_DEBUG_FMT "sas_rphy_add: rphy=%p\n", + ioc->name, rphy)); + } + + if (rphy) { + dsaswideprintk(ioc, dev_printk(KERN_DEBUG, + &rphy->dev, MYIOC_s_FMT "add:", ioc->name)); + dsaswideprintk(ioc, printk(MYIOC_s_DEBUG_FMT "rphy=%p release=%p\n", + ioc->name, rphy, rphy->dev.release)); + } +} + /* no mutex */ static void mptsas_port_delete(MPT_ADAPTER *ioc, struct mptsas_portinfo_details * port_details) @@ -474,23 +491,6 @@ mptsas_get_rphy(struct mptsas_phyinfo *p return NULL; } -static inline void -mptsas_set_rphy(MPT_ADAPTER *ioc, struct mptsas_phyinfo *phy_info, struct sas_rphy *rphy) -{ - if (phy_info->port_details) { - phy_info->port_details->rphy = rphy; - dsaswideprintk(ioc, printk(MYIOC_s_DEBUG_FMT "sas_rphy_add: rphy=%p\n", - ioc->name, rphy)); - } - - if (rphy) { - dsaswideprintk(ioc, dev_printk(KERN_DEBUG, - &rphy->dev, MYIOC_s_FMT "add:", ioc->name)); - dsaswideprintk(ioc, printk(MYIOC_s_DEBUG_FMT "rphy=%p release=%p\n", - ioc->name, rphy, rphy->dev.release)); - } -} - static inline struct sas_port * mptsas_get_port(struct mptsas_phyinfo *phy_info) { diff -urNp linux-2.6.32.48/drivers/message/fusion/mptscsih.c linux-2.6.32.48-openvz/drivers/message/fusion/mptscsih.c --- linux-2.6.32.48/drivers/message/fusion/mptscsih.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/message/fusion/mptscsih.c 2011-11-17 20:08:53.000000000 -0500 @@ -1248,15 +1248,16 @@ mptscsih_info(struct Scsi_Host *SChost) h = shost_priv(SChost); - if (h) { - if (h->info_kbuf == NULL) - if ((h->info_kbuf = kmalloc(0x1000 /* 4Kb */, GFP_KERNEL)) == NULL) - return h->info_kbuf; - h->info_kbuf[0] = '\0'; + if (!h) + return NULL; - mpt_print_ioc_summary(h->ioc, h->info_kbuf, &size, 0, 0); - h->info_kbuf[size-1] = '\0'; - } + if (h->info_kbuf == NULL) + if ((h->info_kbuf = kmalloc(0x1000 /* 4Kb */, GFP_KERNEL)) == NULL) + return h->info_kbuf; + h->info_kbuf[0] = '\0'; + + mpt_print_ioc_summary(h->ioc, h->info_kbuf, &size, 0, 0); + h->info_kbuf[size-1] = '\0'; return h->info_kbuf; } diff -urNp linux-2.6.32.48/drivers/message/i2o/i2o_config.c linux-2.6.32.48-openvz/drivers/message/i2o/i2o_config.c --- linux-2.6.32.48/drivers/message/i2o/i2o_config.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/message/i2o/i2o_config.c 2011-11-17 20:08:53.000000000 -0500 @@ -787,6 +787,8 @@ static int i2o_cfg_passthru(unsigned lon struct i2o_message *msg; unsigned int iop; + pax_track_stack(); + if (get_user(iop, &cmd->iop) || get_user(user_msg, &cmd->msg)) return -EFAULT; diff -urNp linux-2.6.32.48/drivers/message/i2o/i2o_proc.c linux-2.6.32.48-openvz/drivers/message/i2o/i2o_proc.c --- linux-2.6.32.48/drivers/message/i2o/i2o_proc.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/message/i2o/i2o_proc.c 2011-11-17 20:08:53.000000000 -0500 @@ -259,13 +259,6 @@ static char *scsi_devices[] = { "Array Controller Device" }; -static char *chtostr(u8 * chars, int n) -{ - char tmp[256]; - tmp[0] = 0; - return strncat(tmp, (char *)chars, n); -} - static int i2o_report_query_status(struct seq_file *seq, int block_status, char *group) { @@ -842,8 +835,7 @@ static int i2o_seq_show_ddm_table(struct seq_printf(seq, "%-#7x", ddm_table.i2o_vendor_id); seq_printf(seq, "%-#8x", ddm_table.module_id); - seq_printf(seq, "%-29s", - chtostr(ddm_table.module_name_version, 28)); + seq_printf(seq, "%-.28s", ddm_table.module_name_version); seq_printf(seq, "%9d ", ddm_table.data_size); seq_printf(seq, "%8d", ddm_table.code_size); @@ -944,8 +936,8 @@ static int i2o_seq_show_drivers_stored(s seq_printf(seq, "%-#7x", dst->i2o_vendor_id); seq_printf(seq, "%-#8x", dst->module_id); - seq_printf(seq, "%-29s", chtostr(dst->module_name_version, 28)); - seq_printf(seq, "%-9s", chtostr(dst->date, 8)); + seq_printf(seq, "%-.28s", dst->module_name_version); + seq_printf(seq, "%-.8s", dst->date); seq_printf(seq, "%8d ", dst->module_size); seq_printf(seq, "%8d ", dst->mpb_size); seq_printf(seq, "0x%04x", dst->module_flags); @@ -1276,14 +1268,10 @@ static int i2o_seq_show_dev_identity(str seq_printf(seq, "Device Class : %s\n", i2o_get_class_name(work16[0])); seq_printf(seq, "Owner TID : %0#5x\n", work16[2]); seq_printf(seq, "Parent TID : %0#5x\n", work16[3]); - seq_printf(seq, "Vendor info : %s\n", - chtostr((u8 *) (work32 + 2), 16)); - seq_printf(seq, "Product info : %s\n", - chtostr((u8 *) (work32 + 6), 16)); - seq_printf(seq, "Description : %s\n", - chtostr((u8 *) (work32 + 10), 16)); - seq_printf(seq, "Product rev. : %s\n", - chtostr((u8 *) (work32 + 14), 8)); + seq_printf(seq, "Vendor info : %.16s\n", (u8 *) (work32 + 2)); + seq_printf(seq, "Product info : %.16s\n", (u8 *) (work32 + 6)); + seq_printf(seq, "Description : %.16s\n", (u8 *) (work32 + 10)); + seq_printf(seq, "Product rev. : %.8s\n", (u8 *) (work32 + 14)); seq_printf(seq, "Serial number : "); print_serial_number(seq, (u8 *) (work32 + 16), @@ -1328,10 +1316,8 @@ static int i2o_seq_show_ddm_identity(str } seq_printf(seq, "Registering DDM TID : 0x%03x\n", result.ddm_tid); - seq_printf(seq, "Module name : %s\n", - chtostr(result.module_name, 24)); - seq_printf(seq, "Module revision : %s\n", - chtostr(result.module_rev, 8)); + seq_printf(seq, "Module name : %.24s\n", result.module_name); + seq_printf(seq, "Module revision : %.8s\n", result.module_rev); seq_printf(seq, "Serial number : "); print_serial_number(seq, result.serial_number, sizeof(result) - 36); @@ -1362,14 +1348,10 @@ static int i2o_seq_show_uinfo(struct seq return 0; } - seq_printf(seq, "Device name : %s\n", - chtostr(result.device_name, 64)); - seq_printf(seq, "Service name : %s\n", - chtostr(result.service_name, 64)); - seq_printf(seq, "Physical name : %s\n", - chtostr(result.physical_location, 64)); - seq_printf(seq, "Instance number : %s\n", - chtostr(result.instance_number, 4)); + seq_printf(seq, "Device name : %.64s\n", result.device_name); + seq_printf(seq, "Service name : %.64s\n", result.service_name); + seq_printf(seq, "Physical name : %.64s\n", result.physical_location); + seq_printf(seq, "Instance number : %.4s\n", result.instance_number); return 0; } diff -urNp linux-2.6.32.48/drivers/message/i2o/iop.c linux-2.6.32.48-openvz/drivers/message/i2o/iop.c --- linux-2.6.32.48/drivers/message/i2o/iop.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/message/i2o/iop.c 2011-11-17 20:08:53.000000000 -0500 @@ -110,10 +110,10 @@ u32 i2o_cntxt_list_add(struct i2o_contro spin_lock_irqsave(&c->context_list_lock, flags); - if (unlikely(atomic_inc_and_test(&c->context_list_counter))) - atomic_inc(&c->context_list_counter); + if (unlikely(atomic_inc_and_test_unchecked(&c->context_list_counter))) + atomic_inc_unchecked(&c->context_list_counter); - entry->context = atomic_read(&c->context_list_counter); + entry->context = atomic_read_unchecked(&c->context_list_counter); list_add(&entry->list, &c->context_list); @@ -1076,7 +1076,7 @@ struct i2o_controller *i2o_iop_alloc(voi #if BITS_PER_LONG == 64 spin_lock_init(&c->context_list_lock); - atomic_set(&c->context_list_counter, 0); + atomic_set_unchecked(&c->context_list_counter, 0); INIT_LIST_HEAD(&c->context_list); #endif diff -urNp linux-2.6.32.48/drivers/mfd/wm8350-i2c.c linux-2.6.32.48-openvz/drivers/mfd/wm8350-i2c.c --- linux-2.6.32.48/drivers/mfd/wm8350-i2c.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/mfd/wm8350-i2c.c 2011-11-17 20:08:53.000000000 -0500 @@ -43,6 +43,8 @@ static int wm8350_i2c_write_device(struc u8 msg[(WM8350_MAX_REGISTER << 1) + 1]; int ret; + pax_track_stack(); + if (bytes > ((WM8350_MAX_REGISTER << 1) + 1)) return -EINVAL; diff -urNp linux-2.6.32.48/drivers/misc/kgdbts.c linux-2.6.32.48-openvz/drivers/misc/kgdbts.c --- linux-2.6.32.48/drivers/misc/kgdbts.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/misc/kgdbts.c 2011-11-17 20:08:53.000000000 -0500 @@ -118,7 +118,7 @@ } while (0) #define MAX_CONFIG_LEN 40 -static struct kgdb_io kgdbts_io_ops; +static const struct kgdb_io kgdbts_io_ops; static char get_buf[BUFMAX]; static int get_buf_cnt; static char put_buf[BUFMAX]; @@ -1102,7 +1102,7 @@ static void kgdbts_post_exp_handler(void module_put(THIS_MODULE); } -static struct kgdb_io kgdbts_io_ops = { +static const struct kgdb_io kgdbts_io_ops = { .name = "kgdbts", .read_char = kgdbts_get_char, .write_char = kgdbts_put_char, diff -urNp linux-2.6.32.48/drivers/misc/sgi-gru/gruhandles.c linux-2.6.32.48-openvz/drivers/misc/sgi-gru/gruhandles.c --- linux-2.6.32.48/drivers/misc/sgi-gru/gruhandles.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/misc/sgi-gru/gruhandles.c 2011-11-17 20:08:53.000000000 -0500 @@ -39,8 +39,8 @@ struct mcs_op_statistic mcs_op_statistic static void update_mcs_stats(enum mcs_op op, unsigned long clks) { - atomic_long_inc(&mcs_op_statistics[op].count); - atomic_long_add(clks, &mcs_op_statistics[op].total); + atomic_long_inc_unchecked(&mcs_op_statistics[op].count); + atomic_long_add_unchecked(clks, &mcs_op_statistics[op].total); if (mcs_op_statistics[op].max < clks) mcs_op_statistics[op].max = clks; } diff -urNp linux-2.6.32.48/drivers/misc/sgi-gru/gruprocfs.c linux-2.6.32.48-openvz/drivers/misc/sgi-gru/gruprocfs.c --- linux-2.6.32.48/drivers/misc/sgi-gru/gruprocfs.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/misc/sgi-gru/gruprocfs.c 2011-11-17 20:08:53.000000000 -0500 @@ -32,9 +32,9 @@ #define printstat(s, f) printstat_val(s, &gru_stats.f, #f) -static void printstat_val(struct seq_file *s, atomic_long_t *v, char *id) +static void printstat_val(struct seq_file *s, atomic_long_unchecked_t *v, char *id) { - unsigned long val = atomic_long_read(v); + unsigned long val = atomic_long_read_unchecked(v); if (val) seq_printf(s, "%16lu %s\n", val, id); @@ -136,8 +136,8 @@ static int mcs_statistics_show(struct se "cch_interrupt_sync", "cch_deallocate", "tgh_invalidate"}; for (op = 0; op < mcsop_last; op++) { - count = atomic_long_read(&mcs_op_statistics[op].count); - total = atomic_long_read(&mcs_op_statistics[op].total); + count = atomic_long_read_unchecked(&mcs_op_statistics[op].count); + total = atomic_long_read_unchecked(&mcs_op_statistics[op].total); max = mcs_op_statistics[op].max; seq_printf(s, "%-20s%12ld%12ld%12ld\n", id[op], count, count ? total / count : 0, max); diff -urNp linux-2.6.32.48/drivers/misc/sgi-gru/grutables.h linux-2.6.32.48-openvz/drivers/misc/sgi-gru/grutables.h --- linux-2.6.32.48/drivers/misc/sgi-gru/grutables.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/misc/sgi-gru/grutables.h 2011-11-17 20:08:53.000000000 -0500 @@ -167,84 +167,84 @@ extern unsigned int gru_max_gids; * GRU statistics. */ struct gru_stats_s { - atomic_long_t vdata_alloc; - atomic_long_t vdata_free; - atomic_long_t gts_alloc; - atomic_long_t gts_free; - atomic_long_t vdata_double_alloc; - atomic_long_t gts_double_allocate; - atomic_long_t assign_context; - atomic_long_t assign_context_failed; - atomic_long_t free_context; - atomic_long_t load_user_context; - atomic_long_t load_kernel_context; - atomic_long_t lock_kernel_context; - atomic_long_t unlock_kernel_context; - atomic_long_t steal_user_context; - atomic_long_t steal_kernel_context; - atomic_long_t steal_context_failed; - atomic_long_t nopfn; - atomic_long_t break_cow; - atomic_long_t asid_new; - atomic_long_t asid_next; - atomic_long_t asid_wrap; - atomic_long_t asid_reuse; - atomic_long_t intr; - atomic_long_t intr_mm_lock_failed; - atomic_long_t call_os; - atomic_long_t call_os_offnode_reference; - atomic_long_t call_os_check_for_bug; - atomic_long_t call_os_wait_queue; - atomic_long_t user_flush_tlb; - atomic_long_t user_unload_context; - atomic_long_t user_exception; - atomic_long_t set_context_option; - atomic_long_t migrate_check; - atomic_long_t migrated_retarget; - atomic_long_t migrated_unload; - atomic_long_t migrated_unload_delay; - atomic_long_t migrated_nopfn_retarget; - atomic_long_t migrated_nopfn_unload; - atomic_long_t tlb_dropin; - atomic_long_t tlb_dropin_fail_no_asid; - atomic_long_t tlb_dropin_fail_upm; - atomic_long_t tlb_dropin_fail_invalid; - atomic_long_t tlb_dropin_fail_range_active; - atomic_long_t tlb_dropin_fail_idle; - atomic_long_t tlb_dropin_fail_fmm; - atomic_long_t tlb_dropin_fail_no_exception; - atomic_long_t tlb_dropin_fail_no_exception_war; - atomic_long_t tfh_stale_on_fault; - atomic_long_t mmu_invalidate_range; - atomic_long_t mmu_invalidate_page; - atomic_long_t mmu_clear_flush_young; - atomic_long_t flush_tlb; - atomic_long_t flush_tlb_gru; - atomic_long_t flush_tlb_gru_tgh; - atomic_long_t flush_tlb_gru_zero_asid; - - atomic_long_t copy_gpa; - - atomic_long_t mesq_receive; - atomic_long_t mesq_receive_none; - atomic_long_t mesq_send; - atomic_long_t mesq_send_failed; - atomic_long_t mesq_noop; - atomic_long_t mesq_send_unexpected_error; - atomic_long_t mesq_send_lb_overflow; - atomic_long_t mesq_send_qlimit_reached; - atomic_long_t mesq_send_amo_nacked; - atomic_long_t mesq_send_put_nacked; - atomic_long_t mesq_qf_not_full; - atomic_long_t mesq_qf_locked; - atomic_long_t mesq_qf_noop_not_full; - atomic_long_t mesq_qf_switch_head_failed; - atomic_long_t mesq_qf_unexpected_error; - atomic_long_t mesq_noop_unexpected_error; - atomic_long_t mesq_noop_lb_overflow; - atomic_long_t mesq_noop_qlimit_reached; - atomic_long_t mesq_noop_amo_nacked; - atomic_long_t mesq_noop_put_nacked; + atomic_long_unchecked_t vdata_alloc; + atomic_long_unchecked_t vdata_free; + atomic_long_unchecked_t gts_alloc; + atomic_long_unchecked_t gts_free; + atomic_long_unchecked_t vdata_double_alloc; + atomic_long_unchecked_t gts_double_allocate; + atomic_long_unchecked_t assign_context; + atomic_long_unchecked_t assign_context_failed; + atomic_long_unchecked_t free_context; + atomic_long_unchecked_t load_user_context; + atomic_long_unchecked_t load_kernel_context; + atomic_long_unchecked_t lock_kernel_context; + atomic_long_unchecked_t unlock_kernel_context; + atomic_long_unchecked_t steal_user_context; + atomic_long_unchecked_t steal_kernel_context; + atomic_long_unchecked_t steal_context_failed; + atomic_long_unchecked_t nopfn; + atomic_long_unchecked_t break_cow; + atomic_long_unchecked_t asid_new; + atomic_long_unchecked_t asid_next; + atomic_long_unchecked_t asid_wrap; + atomic_long_unchecked_t asid_reuse; + atomic_long_unchecked_t intr; + atomic_long_unchecked_t intr_mm_lock_failed; + atomic_long_unchecked_t call_os; + atomic_long_unchecked_t call_os_offnode_reference; + atomic_long_unchecked_t call_os_check_for_bug; + atomic_long_unchecked_t call_os_wait_queue; + atomic_long_unchecked_t user_flush_tlb; + atomic_long_unchecked_t user_unload_context; + atomic_long_unchecked_t user_exception; + atomic_long_unchecked_t set_context_option; + atomic_long_unchecked_t migrate_check; + atomic_long_unchecked_t migrated_retarget; + atomic_long_unchecked_t migrated_unload; + atomic_long_unchecked_t migrated_unload_delay; + atomic_long_unchecked_t migrated_nopfn_retarget; + atomic_long_unchecked_t migrated_nopfn_unload; + atomic_long_unchecked_t tlb_dropin; + atomic_long_unchecked_t tlb_dropin_fail_no_asid; + atomic_long_unchecked_t tlb_dropin_fail_upm; + atomic_long_unchecked_t tlb_dropin_fail_invalid; + atomic_long_unchecked_t tlb_dropin_fail_range_active; + atomic_long_unchecked_t tlb_dropin_fail_idle; + atomic_long_unchecked_t tlb_dropin_fail_fmm; + atomic_long_unchecked_t tlb_dropin_fail_no_exception; + atomic_long_unchecked_t tlb_dropin_fail_no_exception_war; + atomic_long_unchecked_t tfh_stale_on_fault; + atomic_long_unchecked_t mmu_invalidate_range; + atomic_long_unchecked_t mmu_invalidate_page; + atomic_long_unchecked_t mmu_clear_flush_young; + atomic_long_unchecked_t flush_tlb; + atomic_long_unchecked_t flush_tlb_gru; + atomic_long_unchecked_t flush_tlb_gru_tgh; + atomic_long_unchecked_t flush_tlb_gru_zero_asid; + + atomic_long_unchecked_t copy_gpa; + + atomic_long_unchecked_t mesq_receive; + atomic_long_unchecked_t mesq_receive_none; + atomic_long_unchecked_t mesq_send; + atomic_long_unchecked_t mesq_send_failed; + atomic_long_unchecked_t mesq_noop; + atomic_long_unchecked_t mesq_send_unexpected_error; + atomic_long_unchecked_t mesq_send_lb_overflow; + atomic_long_unchecked_t mesq_send_qlimit_reached; + atomic_long_unchecked_t mesq_send_amo_nacked; + atomic_long_unchecked_t mesq_send_put_nacked; + atomic_long_unchecked_t mesq_qf_not_full; + atomic_long_unchecked_t mesq_qf_locked; + atomic_long_unchecked_t mesq_qf_noop_not_full; + atomic_long_unchecked_t mesq_qf_switch_head_failed; + atomic_long_unchecked_t mesq_qf_unexpected_error; + atomic_long_unchecked_t mesq_noop_unexpected_error; + atomic_long_unchecked_t mesq_noop_lb_overflow; + atomic_long_unchecked_t mesq_noop_qlimit_reached; + atomic_long_unchecked_t mesq_noop_amo_nacked; + atomic_long_unchecked_t mesq_noop_put_nacked; }; @@ -252,8 +252,8 @@ enum mcs_op {cchop_allocate, cchop_start cchop_deallocate, tghop_invalidate, mcsop_last}; struct mcs_op_statistic { - atomic_long_t count; - atomic_long_t total; + atomic_long_unchecked_t count; + atomic_long_unchecked_t total; unsigned long max; }; @@ -276,7 +276,7 @@ extern struct mcs_op_statistic mcs_op_st #define STAT(id) do { \ if (gru_options & OPT_STATS) \ - atomic_long_inc(&gru_stats.id); \ + atomic_long_inc_unchecked(&gru_stats.id); \ } while (0) #ifdef CONFIG_SGI_GRU_DEBUG diff -urNp linux-2.6.32.48/drivers/misc/sgi-xp/xpc.h linux-2.6.32.48-openvz/drivers/misc/sgi-xp/xpc.h --- linux-2.6.32.48/drivers/misc/sgi-xp/xpc.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/misc/sgi-xp/xpc.h 2011-11-17 20:09:04.000000000 -0500 @@ -876,7 +876,7 @@ extern struct xpc_registration xpc_regis /* found in xpc_main.c */ extern struct device *xpc_part; extern struct device *xpc_chan; -extern struct xpc_arch_operations xpc_arch_ops; +extern const struct xpc_arch_operations xpc_arch_ops; extern int xpc_disengage_timelimit; extern int xpc_disengage_timedout; extern int xpc_activate_IRQ_rcvd; diff -urNp linux-2.6.32.48/drivers/misc/sgi-xp/xpc_main.c linux-2.6.32.48-openvz/drivers/misc/sgi-xp/xpc_main.c --- linux-2.6.32.48/drivers/misc/sgi-xp/xpc_main.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/misc/sgi-xp/xpc_main.c 2011-11-17 20:09:04.000000000 -0500 @@ -169,7 +169,7 @@ static struct notifier_block xpc_die_not .notifier_call = xpc_system_die, }; -struct xpc_arch_operations xpc_arch_ops; +const struct xpc_arch_operations xpc_arch_ops; /* * Timer function to enforce the timelimit on the partition disengage. diff -urNp linux-2.6.32.48/drivers/misc/sgi-xp/xpc_sn2.c linux-2.6.32.48-openvz/drivers/misc/sgi-xp/xpc_sn2.c --- linux-2.6.32.48/drivers/misc/sgi-xp/xpc_sn2.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/misc/sgi-xp/xpc_sn2.c 2011-11-17 20:09:04.000000000 -0500 @@ -2350,7 +2350,7 @@ xpc_received_payload_sn2(struct xpc_chan xpc_acknowledge_msgs_sn2(ch, get, msg->flags); } -static struct xpc_arch_operations xpc_arch_ops_sn2 = { +static const struct xpc_arch_operations xpc_arch_ops_sn2 = { .setup_partitions = xpc_setup_partitions_sn2, .teardown_partitions = xpc_teardown_partitions_sn2, .process_activate_IRQ_rcvd = xpc_process_activate_IRQ_rcvd_sn2, @@ -2413,7 +2413,9 @@ xpc_init_sn2(void) int ret; size_t buf_size; - xpc_arch_ops = xpc_arch_ops_sn2; + pax_open_kernel(); + memcpy((void *)&xpc_arch_ops, &xpc_arch_ops_sn2, sizeof(xpc_arch_ops_sn2)); + pax_close_kernel(); if (offsetof(struct xpc_msg_sn2, payload) > XPC_MSG_HDR_MAX_SIZE) { dev_err(xpc_part, "header portion of struct xpc_msg_sn2 is " diff -urNp linux-2.6.32.48/drivers/misc/sgi-xp/xpc_uv.c linux-2.6.32.48-openvz/drivers/misc/sgi-xp/xpc_uv.c --- linux-2.6.32.48/drivers/misc/sgi-xp/xpc_uv.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/misc/sgi-xp/xpc_uv.c 2011-11-17 20:09:04.000000000 -0500 @@ -1669,7 +1669,7 @@ xpc_received_payload_uv(struct xpc_chann XPC_DEACTIVATE_PARTITION(&xpc_partitions[ch->partid], ret); } -static struct xpc_arch_operations xpc_arch_ops_uv = { +static const struct xpc_arch_operations xpc_arch_ops_uv = { .setup_partitions = xpc_setup_partitions_uv, .teardown_partitions = xpc_teardown_partitions_uv, .process_activate_IRQ_rcvd = xpc_process_activate_IRQ_rcvd_uv, @@ -1729,7 +1729,9 @@ static struct xpc_arch_operations xpc_ar int xpc_init_uv(void) { - xpc_arch_ops = xpc_arch_ops_uv; + pax_open_kernel(); + memcpy((void *)&xpc_arch_ops, &xpc_arch_ops_uv, sizeof(xpc_arch_ops_uv)); + pax_close_kernel(); if (sizeof(struct xpc_notify_mq_msghdr_uv) > XPC_MSG_HDR_MAX_SIZE) { dev_err(xpc_part, "xpc_notify_mq_msghdr_uv is larger than %d\n", diff -urNp linux-2.6.32.48/drivers/misc/sgi-xp/xp.h linux-2.6.32.48-openvz/drivers/misc/sgi-xp/xp.h --- linux-2.6.32.48/drivers/misc/sgi-xp/xp.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/misc/sgi-xp/xp.h 2011-11-17 20:09:04.000000000 -0500 @@ -289,7 +289,7 @@ struct xpc_interface { xpc_notify_func, void *); void (*received) (short, int, void *); enum xp_retval (*partid_to_nasids) (short, void *); -}; +} __no_const; extern struct xpc_interface xpc_interface; diff -urNp linux-2.6.32.48/drivers/mtd/chips/cfi_cmdset_0001.c linux-2.6.32.48-openvz/drivers/mtd/chips/cfi_cmdset_0001.c --- linux-2.6.32.48/drivers/mtd/chips/cfi_cmdset_0001.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/mtd/chips/cfi_cmdset_0001.c 2011-11-17 20:08:53.000000000 -0500 @@ -743,6 +743,8 @@ static int chip_ready (struct map_info * struct cfi_pri_intelext *cfip = cfi->cmdset_priv; unsigned long timeo = jiffies + HZ; + pax_track_stack(); + /* Prevent setting state FL_SYNCING for chip in suspended state. */ if (mode == FL_SYNCING && chip->oldstate != FL_READY) goto sleep; @@ -1642,6 +1644,8 @@ static int __xipram do_write_buffer(stru unsigned long initial_adr; int initial_len = len; + pax_track_stack(); + wbufsize = cfi_interleave(cfi) << cfi->cfiq->MaxBufWriteSize; adr += chip->start; initial_adr = adr; @@ -1860,6 +1864,8 @@ static int __xipram do_erase_oneblock(st int retries = 3; int ret; + pax_track_stack(); + adr += chip->start; retry: diff -urNp linux-2.6.32.48/drivers/mtd/chips/cfi_cmdset_0020.c linux-2.6.32.48-openvz/drivers/mtd/chips/cfi_cmdset_0020.c --- linux-2.6.32.48/drivers/mtd/chips/cfi_cmdset_0020.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/mtd/chips/cfi_cmdset_0020.c 2011-11-17 20:08:53.000000000 -0500 @@ -255,6 +255,8 @@ static inline int do_read_onechip(struct unsigned long cmd_addr; struct cfi_private *cfi = map->fldrv_priv; + pax_track_stack(); + adr += chip->start; /* Ensure cmd read/writes are aligned. */ @@ -428,6 +430,8 @@ static inline int do_write_buffer(struct DECLARE_WAITQUEUE(wait, current); int wbufsize, z; + pax_track_stack(); + /* M58LW064A requires bus alignment for buffer wriets -- saw */ if (adr & (map_bankwidth(map)-1)) return -EINVAL; @@ -742,6 +746,8 @@ static inline int do_erase_oneblock(stru DECLARE_WAITQUEUE(wait, current); int ret = 0; + pax_track_stack(); + adr += chip->start; /* Let's determine this according to the interleave only once */ @@ -1047,6 +1053,8 @@ static inline int do_lock_oneblock(struc unsigned long timeo = jiffies + HZ; DECLARE_WAITQUEUE(wait, current); + pax_track_stack(); + adr += chip->start; /* Let's determine this according to the interleave only once */ @@ -1196,6 +1204,8 @@ static inline int do_unlock_oneblock(str unsigned long timeo = jiffies + HZ; DECLARE_WAITQUEUE(wait, current); + pax_track_stack(); + adr += chip->start; /* Let's determine this according to the interleave only once */ diff -urNp linux-2.6.32.48/drivers/mtd/devices/doc2000.c linux-2.6.32.48-openvz/drivers/mtd/devices/doc2000.c --- linux-2.6.32.48/drivers/mtd/devices/doc2000.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/mtd/devices/doc2000.c 2011-11-17 20:08:53.000000000 -0500 @@ -776,7 +776,7 @@ static int doc_write(struct mtd_info *mt /* The ECC will not be calculated correctly if less than 512 is written */ /* DBB- - if (len != 0x200 && eccbuf) + if (len != 0x200) printk(KERN_WARNING "ECC needs a full sector write (adr: %lx size %lx)\n", (long) to, (long) len); diff -urNp linux-2.6.32.48/drivers/mtd/devices/doc2001.c linux-2.6.32.48-openvz/drivers/mtd/devices/doc2001.c --- linux-2.6.32.48/drivers/mtd/devices/doc2001.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/mtd/devices/doc2001.c 2011-11-17 20:08:53.000000000 -0500 @@ -393,7 +393,7 @@ static int doc_read (struct mtd_info *mt struct Nand *mychip = &this->chips[from >> (this->chipshift)]; /* Don't allow read past end of device */ - if (from >= this->totlen) + if (from >= this->totlen || !len) return -EINVAL; /* Don't allow a single read to cross a 512-byte block boundary */ diff -urNp linux-2.6.32.48/drivers/mtd/ftl.c linux-2.6.32.48-openvz/drivers/mtd/ftl.c --- linux-2.6.32.48/drivers/mtd/ftl.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/mtd/ftl.c 2011-11-17 20:08:53.000000000 -0500 @@ -474,6 +474,8 @@ static int copy_erase_unit(partition_t * loff_t offset; uint16_t srcunitswap = cpu_to_le16(srcunit); + pax_track_stack(); + eun = &part->EUNInfo[srcunit]; xfer = &part->XferInfo[xferunit]; DEBUG(2, "ftl_cs: copying block 0x%x to 0x%x\n", diff -urNp linux-2.6.32.48/drivers/mtd/inftlcore.c linux-2.6.32.48-openvz/drivers/mtd/inftlcore.c --- linux-2.6.32.48/drivers/mtd/inftlcore.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/mtd/inftlcore.c 2011-11-17 20:08:53.000000000 -0500 @@ -260,6 +260,8 @@ static u16 INFTL_foldchain(struct INFTLr struct inftl_oob oob; size_t retlen; + pax_track_stack(); + DEBUG(MTD_DEBUG_LEVEL3, "INFTL: INFTL_foldchain(inftl=%p,thisVUC=%d," "pending=%d)\n", inftl, thisVUC, pendingblock); diff -urNp linux-2.6.32.48/drivers/mtd/inftlmount.c linux-2.6.32.48-openvz/drivers/mtd/inftlmount.c --- linux-2.6.32.48/drivers/mtd/inftlmount.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/mtd/inftlmount.c 2011-11-17 20:08:53.000000000 -0500 @@ -54,6 +54,8 @@ static int find_boot_record(struct INFTL struct INFTLPartition *ip; size_t retlen; + pax_track_stack(); + DEBUG(MTD_DEBUG_LEVEL3, "INFTL: find_boot_record(inftl=%p)\n", inftl); /* diff -urNp linux-2.6.32.48/drivers/mtd/lpddr/qinfo_probe.c linux-2.6.32.48-openvz/drivers/mtd/lpddr/qinfo_probe.c --- linux-2.6.32.48/drivers/mtd/lpddr/qinfo_probe.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/mtd/lpddr/qinfo_probe.c 2011-11-17 20:08:53.000000000 -0500 @@ -106,6 +106,8 @@ static int lpddr_pfow_present(struct map { map_word pfow_val[4]; + pax_track_stack(); + /* Check identification string */ pfow_val[0] = map_read(map, map->pfow_base + PFOW_QUERY_STRING_P); pfow_val[1] = map_read(map, map->pfow_base + PFOW_QUERY_STRING_F); diff -urNp linux-2.6.32.48/drivers/mtd/mtdchar.c linux-2.6.32.48-openvz/drivers/mtd/mtdchar.c --- linux-2.6.32.48/drivers/mtd/mtdchar.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/mtd/mtdchar.c 2011-11-17 20:08:53.000000000 -0500 @@ -460,6 +460,8 @@ static int mtd_ioctl(struct inode *inode u_long size; struct mtd_info_user info; + pax_track_stack(); + DEBUG(MTD_DEBUG_LEVEL0, "MTD_ioctl\n"); size = (cmd & IOCSIZE_MASK) >> IOCSIZE_SHIFT; diff -urNp linux-2.6.32.48/drivers/mtd/nftlcore.c linux-2.6.32.48-openvz/drivers/mtd/nftlcore.c --- linux-2.6.32.48/drivers/mtd/nftlcore.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/mtd/nftlcore.c 2011-11-17 20:08:53.000000000 -0500 @@ -254,6 +254,8 @@ static u16 NFTL_foldchain (struct NFTLre int inplace = 1; size_t retlen; + pax_track_stack(); + memset(BlockMap, 0xff, sizeof(BlockMap)); memset(BlockFreeFound, 0, sizeof(BlockFreeFound)); diff -urNp linux-2.6.32.48/drivers/mtd/nftlmount.c linux-2.6.32.48-openvz/drivers/mtd/nftlmount.c --- linux-2.6.32.48/drivers/mtd/nftlmount.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/mtd/nftlmount.c 2011-11-17 20:08:53.000000000 -0500 @@ -23,6 +23,7 @@ #include #include #include +#include #include #include #include @@ -44,6 +45,8 @@ static int find_boot_record(struct NFTLr struct mtd_info *mtd = nftl->mbd.mtd; unsigned int i; + pax_track_stack(); + /* Assume logical EraseSize == physical erasesize for starting the scan. We'll sort it out later if we find a MediaHeader which says otherwise */ /* Actually, we won't. The new DiskOnChip driver has already scanned diff -urNp linux-2.6.32.48/drivers/mtd/ubi/build.c linux-2.6.32.48-openvz/drivers/mtd/ubi/build.c --- linux-2.6.32.48/drivers/mtd/ubi/build.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/mtd/ubi/build.c 2011-11-17 20:08:53.000000000 -0500 @@ -1255,7 +1255,7 @@ module_exit(ubi_exit); static int __init bytes_str_to_int(const char *str) { char *endp; - unsigned long result; + unsigned long result, scale = 1; result = simple_strtoul(str, &endp, 0); if (str == endp || result >= INT_MAX) { @@ -1266,11 +1266,11 @@ static int __init bytes_str_to_int(const switch (*endp) { case 'G': - result *= 1024; + scale *= 1024; case 'M': - result *= 1024; + scale *= 1024; case 'K': - result *= 1024; + scale *= 1024; if (endp[1] == 'i' && endp[2] == 'B') endp += 2; case '\0': @@ -1281,7 +1281,13 @@ static int __init bytes_str_to_int(const return -EINVAL; } - return result; + if ((intoverflow_t)result*scale >= INT_MAX) { + printk(KERN_ERR "UBI error: incorrect bytes count: \"%s\"\n", + str); + return -EINVAL; + } + + return result*scale; } /** diff -urNp linux-2.6.32.48/drivers/net/bnx2.c linux-2.6.32.48-openvz/drivers/net/bnx2.c --- linux-2.6.32.48/drivers/net/bnx2.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/bnx2.c 2011-11-17 20:08:53.000000000 -0500 @@ -5809,6 +5809,8 @@ bnx2_test_nvram(struct bnx2 *bp) int rc = 0; u32 magic, csum; + pax_track_stack(); + if ((rc = bnx2_nvram_read(bp, 0, data, 4)) != 0) goto test_nvram_done; diff -urNp linux-2.6.32.48/drivers/net/cxgb3/l2t.h linux-2.6.32.48-openvz/drivers/net/cxgb3/l2t.h --- linux-2.6.32.48/drivers/net/cxgb3/l2t.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/cxgb3/l2t.h 2011-11-17 20:09:04.000000000 -0500 @@ -86,7 +86,7 @@ typedef void (*arp_failure_handler_func) */ struct l2t_skb_cb { arp_failure_handler_func arp_failure_handler; -}; +} __no_const; #define L2T_SKB_CB(skb) ((struct l2t_skb_cb *)(skb)->cb) diff -urNp linux-2.6.32.48/drivers/net/cxgb3/t3_hw.c linux-2.6.32.48-openvz/drivers/net/cxgb3/t3_hw.c --- linux-2.6.32.48/drivers/net/cxgb3/t3_hw.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/cxgb3/t3_hw.c 2011-11-17 20:08:53.000000000 -0500 @@ -699,6 +699,8 @@ static int get_vpd_params(struct adapter int i, addr, ret; struct t3_vpd vpd; + pax_track_stack(); + /* * Card information is normally at VPD_BASE but some early cards had * it at 0. diff -urNp linux-2.6.32.48/drivers/net/e1000e/82571.c linux-2.6.32.48-openvz/drivers/net/e1000e/82571.c --- linux-2.6.32.48/drivers/net/e1000e/82571.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/e1000e/82571.c 2011-11-17 20:09:04.000000000 -0500 @@ -212,7 +212,7 @@ static s32 e1000_init_mac_params_82571(s { struct e1000_hw *hw = &adapter->hw; struct e1000_mac_info *mac = &hw->mac; - struct e1000_mac_operations *func = &mac->ops; + e1000_mac_operations_no_const *func = &mac->ops; u32 swsm = 0; u32 swsm2 = 0; bool force_clear_smbi = false; @@ -1656,7 +1656,7 @@ static void e1000_clear_hw_cntrs_82571(s temp = er32(ICRXDMTC); } -static struct e1000_mac_operations e82571_mac_ops = { +static const struct e1000_mac_operations e82571_mac_ops = { /* .check_mng_mode: mac type dependent */ /* .check_for_link: media type dependent */ .id_led_init = e1000e_id_led_init, @@ -1674,7 +1674,7 @@ static struct e1000_mac_operations e8257 .setup_led = e1000e_setup_led_generic, }; -static struct e1000_phy_operations e82_phy_ops_igp = { +static const struct e1000_phy_operations e82_phy_ops_igp = { .acquire_phy = e1000_get_hw_semaphore_82571, .check_reset_block = e1000e_check_reset_block_generic, .commit_phy = NULL, @@ -1691,7 +1691,7 @@ static struct e1000_phy_operations e82_p .cfg_on_link_up = NULL, }; -static struct e1000_phy_operations e82_phy_ops_m88 = { +static const struct e1000_phy_operations e82_phy_ops_m88 = { .acquire_phy = e1000_get_hw_semaphore_82571, .check_reset_block = e1000e_check_reset_block_generic, .commit_phy = e1000e_phy_sw_reset, @@ -1708,7 +1708,7 @@ static struct e1000_phy_operations e82_p .cfg_on_link_up = NULL, }; -static struct e1000_phy_operations e82_phy_ops_bm = { +static const struct e1000_phy_operations e82_phy_ops_bm = { .acquire_phy = e1000_get_hw_semaphore_82571, .check_reset_block = e1000e_check_reset_block_generic, .commit_phy = e1000e_phy_sw_reset, @@ -1725,7 +1725,7 @@ static struct e1000_phy_operations e82_p .cfg_on_link_up = NULL, }; -static struct e1000_nvm_operations e82571_nvm_ops = { +static const struct e1000_nvm_operations e82571_nvm_ops = { .acquire_nvm = e1000_acquire_nvm_82571, .read_nvm = e1000e_read_nvm_eerd, .release_nvm = e1000_release_nvm_82571, diff -urNp linux-2.6.32.48/drivers/net/e1000e/e1000.h linux-2.6.32.48-openvz/drivers/net/e1000e/e1000.h --- linux-2.6.32.48/drivers/net/e1000e/e1000.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/e1000e/e1000.h 2011-11-17 20:08:53.000000000 -0500 @@ -375,9 +375,9 @@ struct e1000_info { u32 pba; u32 max_hw_frame_size; s32 (*get_variants)(struct e1000_adapter *); - struct e1000_mac_operations *mac_ops; - struct e1000_phy_operations *phy_ops; - struct e1000_nvm_operations *nvm_ops; + const struct e1000_mac_operations *mac_ops; + const struct e1000_phy_operations *phy_ops; + const struct e1000_nvm_operations *nvm_ops; }; /* hardware capability, feature, and workaround flags */ diff -urNp linux-2.6.32.48/drivers/net/e1000e/es2lan.c linux-2.6.32.48-openvz/drivers/net/e1000e/es2lan.c --- linux-2.6.32.48/drivers/net/e1000e/es2lan.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/e1000e/es2lan.c 2011-11-17 20:09:04.000000000 -0500 @@ -207,7 +207,7 @@ static s32 e1000_init_mac_params_80003es { struct e1000_hw *hw = &adapter->hw; struct e1000_mac_info *mac = &hw->mac; - struct e1000_mac_operations *func = &mac->ops; + e1000_mac_operations_no_const *func = &mac->ops; /* Set media type */ switch (adapter->pdev->device) { @@ -1365,7 +1365,7 @@ static void e1000_clear_hw_cntrs_80003es temp = er32(ICRXDMTC); } -static struct e1000_mac_operations es2_mac_ops = { +static const struct e1000_mac_operations es2_mac_ops = { .id_led_init = e1000e_id_led_init, .check_mng_mode = e1000e_check_mng_mode_generic, /* check_for_link dependent on media type */ @@ -1383,7 +1383,7 @@ static struct e1000_mac_operations es2_m .setup_led = e1000e_setup_led_generic, }; -static struct e1000_phy_operations es2_phy_ops = { +static const struct e1000_phy_operations es2_phy_ops = { .acquire_phy = e1000_acquire_phy_80003es2lan, .check_reset_block = e1000e_check_reset_block_generic, .commit_phy = e1000e_phy_sw_reset, @@ -1400,7 +1400,7 @@ static struct e1000_phy_operations es2_p .cfg_on_link_up = e1000_cfg_on_link_up_80003es2lan, }; -static struct e1000_nvm_operations es2_nvm_ops = { +static const struct e1000_nvm_operations es2_nvm_ops = { .acquire_nvm = e1000_acquire_nvm_80003es2lan, .read_nvm = e1000e_read_nvm_eerd, .release_nvm = e1000_release_nvm_80003es2lan, diff -urNp linux-2.6.32.48/drivers/net/e1000e/hw.h linux-2.6.32.48-openvz/drivers/net/e1000e/hw.h --- linux-2.6.32.48/drivers/net/e1000e/hw.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/e1000e/hw.h 2011-11-17 20:09:04.000000000 -0500 @@ -753,6 +753,7 @@ struct e1000_mac_operations { s32 (*setup_physical_interface)(struct e1000_hw *); s32 (*setup_led)(struct e1000_hw *); }; +typedef struct e1000_mac_operations __no_const e1000_mac_operations_no_const; /* Function pointers for the PHY. */ struct e1000_phy_operations { @@ -774,6 +775,7 @@ struct e1000_phy_operations { s32 (*write_phy_reg_locked)(struct e1000_hw *, u32, u16); s32 (*cfg_on_link_up)(struct e1000_hw *); }; +typedef struct e1000_phy_operations __no_const e1000_phy_operations_no_const; /* Function pointers for the NVM. */ struct e1000_nvm_operations { @@ -785,9 +787,10 @@ struct e1000_nvm_operations { s32 (*validate_nvm)(struct e1000_hw *); s32 (*write_nvm)(struct e1000_hw *, u16, u16, u16 *); }; +typedef struct e1000_nvm_operations __no_const e1000_nvm_operations_no_const; struct e1000_mac_info { - struct e1000_mac_operations ops; + e1000_mac_operations_no_const ops; u8 addr[6]; u8 perm_addr[6]; @@ -823,7 +826,7 @@ struct e1000_mac_info { }; struct e1000_phy_info { - struct e1000_phy_operations ops; + e1000_phy_operations_no_const ops; enum e1000_phy_type type; @@ -857,7 +860,7 @@ struct e1000_phy_info { }; struct e1000_nvm_info { - struct e1000_nvm_operations ops; + e1000_nvm_operations_no_const ops; enum e1000_nvm_type type; enum e1000_nvm_override override; diff -urNp linux-2.6.32.48/drivers/net/e1000e/ich8lan.c linux-2.6.32.48-openvz/drivers/net/e1000e/ich8lan.c --- linux-2.6.32.48/drivers/net/e1000e/ich8lan.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/e1000e/ich8lan.c 2011-11-17 20:08:53.000000000 -0500 @@ -3463,7 +3463,7 @@ static void e1000_clear_hw_cntrs_ich8lan } } -static struct e1000_mac_operations ich8_mac_ops = { +static const struct e1000_mac_operations ich8_mac_ops = { .id_led_init = e1000e_id_led_init, .check_mng_mode = e1000_check_mng_mode_ich8lan, .check_for_link = e1000_check_for_copper_link_ich8lan, @@ -3481,7 +3481,7 @@ static struct e1000_mac_operations ich8_ /* id_led_init dependent on mac type */ }; -static struct e1000_phy_operations ich8_phy_ops = { +static const struct e1000_phy_operations ich8_phy_ops = { .acquire_phy = e1000_acquire_swflag_ich8lan, .check_reset_block = e1000_check_reset_block_ich8lan, .commit_phy = NULL, @@ -3497,7 +3497,7 @@ static struct e1000_phy_operations ich8_ .write_phy_reg = e1000e_write_phy_reg_igp, }; -static struct e1000_nvm_operations ich8_nvm_ops = { +static const struct e1000_nvm_operations ich8_nvm_ops = { .acquire_nvm = e1000_acquire_nvm_ich8lan, .read_nvm = e1000_read_nvm_ich8lan, .release_nvm = e1000_release_nvm_ich8lan, diff -urNp linux-2.6.32.48/drivers/net/hamradio/6pack.c linux-2.6.32.48-openvz/drivers/net/hamradio/6pack.c --- linux-2.6.32.48/drivers/net/hamradio/6pack.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/hamradio/6pack.c 2011-11-17 20:09:04.000000000 -0500 @@ -461,6 +461,8 @@ static void sixpack_receive_buf(struct t unsigned char buf[512]; int count1; + pax_track_stack(); + if (!count) return; diff -urNp linux-2.6.32.48/drivers/net/ibmveth.c linux-2.6.32.48-openvz/drivers/net/ibmveth.c --- linux-2.6.32.48/drivers/net/ibmveth.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/ibmveth.c 2011-11-17 20:08:53.000000000 -0500 @@ -1577,7 +1577,7 @@ static struct attribute * veth_pool_attr NULL, }; -static struct sysfs_ops veth_pool_ops = { +static const struct sysfs_ops veth_pool_ops = { .show = veth_pool_show, .store = veth_pool_store, }; diff -urNp linux-2.6.32.48/drivers/net/igb/e1000_82575.c linux-2.6.32.48-openvz/drivers/net/igb/e1000_82575.c --- linux-2.6.32.48/drivers/net/igb/e1000_82575.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/igb/e1000_82575.c 2011-11-17 20:09:04.000000000 -0500 @@ -1411,7 +1411,7 @@ void igb_vmdq_set_replication_pf(struct wr32(E1000_VT_CTL, vt_ctl); } -static struct e1000_mac_operations e1000_mac_ops_82575 = { +static const struct e1000_mac_operations e1000_mac_ops_82575 = { .reset_hw = igb_reset_hw_82575, .init_hw = igb_init_hw_82575, .check_for_link = igb_check_for_link_82575, @@ -1420,13 +1420,13 @@ static struct e1000_mac_operations e1000 .get_speed_and_duplex = igb_get_speed_and_duplex_copper, }; -static struct e1000_phy_operations e1000_phy_ops_82575 = { +static const struct e1000_phy_operations e1000_phy_ops_82575 = { .acquire = igb_acquire_phy_82575, .get_cfg_done = igb_get_cfg_done_82575, .release = igb_release_phy_82575, }; -static struct e1000_nvm_operations e1000_nvm_ops_82575 = { +static const struct e1000_nvm_operations e1000_nvm_ops_82575 = { .acquire = igb_acquire_nvm_82575, .read = igb_read_nvm_eerd, .release = igb_release_nvm_82575, diff -urNp linux-2.6.32.48/drivers/net/igb/e1000_hw.h linux-2.6.32.48-openvz/drivers/net/igb/e1000_hw.h --- linux-2.6.32.48/drivers/net/igb/e1000_hw.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/igb/e1000_hw.h 2011-11-17 20:09:04.000000000 -0500 @@ -288,6 +288,7 @@ struct e1000_mac_operations { s32 (*read_mac_addr)(struct e1000_hw *); s32 (*get_speed_and_duplex)(struct e1000_hw *, u16 *, u16 *); }; +typedef struct e1000_mac_operations __no_const e1000_mac_operations_no_const; struct e1000_phy_operations { s32 (*acquire)(struct e1000_hw *); @@ -303,6 +304,7 @@ struct e1000_phy_operations { s32 (*set_d3_lplu_state)(struct e1000_hw *, bool); s32 (*write_reg)(struct e1000_hw *, u32, u16); }; +typedef struct e1000_phy_operations __no_const e1000_phy_operations_no_const; struct e1000_nvm_operations { s32 (*acquire)(struct e1000_hw *); @@ -310,6 +312,7 @@ struct e1000_nvm_operations { void (*release)(struct e1000_hw *); s32 (*write)(struct e1000_hw *, u16, u16, u16 *); }; +typedef struct e1000_nvm_operations __no_const e1000_nvm_operations_no_const; struct e1000_info { s32 (*get_invariants)(struct e1000_hw *); @@ -321,7 +324,7 @@ struct e1000_info { extern const struct e1000_info e1000_82575_info; struct e1000_mac_info { - struct e1000_mac_operations ops; + e1000_mac_operations_no_const ops; u8 addr[6]; u8 perm_addr[6]; @@ -365,7 +368,7 @@ struct e1000_mac_info { }; struct e1000_phy_info { - struct e1000_phy_operations ops; + e1000_phy_operations_no_const ops; enum e1000_phy_type type; @@ -400,7 +403,7 @@ struct e1000_phy_info { }; struct e1000_nvm_info { - struct e1000_nvm_operations ops; + e1000_nvm_operations_no_const ops; enum e1000_nvm_type type; enum e1000_nvm_override override; @@ -446,6 +449,7 @@ struct e1000_mbx_operations { s32 (*check_for_ack)(struct e1000_hw *, u16); s32 (*check_for_rst)(struct e1000_hw *, u16); }; +typedef struct e1000_mbx_operations __no_const e1000_mbx_operations_no_const; struct e1000_mbx_stats { u32 msgs_tx; @@ -457,7 +461,7 @@ struct e1000_mbx_stats { }; struct e1000_mbx_info { - struct e1000_mbx_operations ops; + e1000_mbx_operations_no_const ops; struct e1000_mbx_stats stats; u32 timeout; u32 usec_delay; diff -urNp linux-2.6.32.48/drivers/net/igbvf/vf.h linux-2.6.32.48-openvz/drivers/net/igbvf/vf.h --- linux-2.6.32.48/drivers/net/igbvf/vf.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/igbvf/vf.h 2011-11-17 20:09:04.000000000 -0500 @@ -187,9 +187,10 @@ struct e1000_mac_operations { s32 (*read_mac_addr)(struct e1000_hw *); s32 (*set_vfta)(struct e1000_hw *, u16, bool); }; +typedef struct e1000_mac_operations __no_const e1000_mac_operations_no_const; struct e1000_mac_info { - struct e1000_mac_operations ops; + e1000_mac_operations_no_const ops; u8 addr[6]; u8 perm_addr[6]; @@ -211,6 +212,7 @@ struct e1000_mbx_operations { s32 (*check_for_ack)(struct e1000_hw *); s32 (*check_for_rst)(struct e1000_hw *); }; +typedef struct e1000_mbx_operations __no_const e1000_mbx_operations_no_const; struct e1000_mbx_stats { u32 msgs_tx; @@ -222,7 +224,7 @@ struct e1000_mbx_stats { }; struct e1000_mbx_info { - struct e1000_mbx_operations ops; + e1000_mbx_operations_no_const ops; struct e1000_mbx_stats stats; u32 timeout; u32 usec_delay; diff -urNp linux-2.6.32.48/drivers/net/iseries_veth.c linux-2.6.32.48-openvz/drivers/net/iseries_veth.c --- linux-2.6.32.48/drivers/net/iseries_veth.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/iseries_veth.c 2011-11-17 20:08:53.000000000 -0500 @@ -384,7 +384,7 @@ static struct attribute *veth_cnx_defaul NULL }; -static struct sysfs_ops veth_cnx_sysfs_ops = { +static const struct sysfs_ops veth_cnx_sysfs_ops = { .show = veth_cnx_attribute_show }; @@ -441,7 +441,7 @@ static struct attribute *veth_port_defau NULL }; -static struct sysfs_ops veth_port_sysfs_ops = { +static const struct sysfs_ops veth_port_sysfs_ops = { .show = veth_port_attribute_show }; diff -urNp linux-2.6.32.48/drivers/net/ixgb/ixgb_main.c linux-2.6.32.48-openvz/drivers/net/ixgb/ixgb_main.c --- linux-2.6.32.48/drivers/net/ixgb/ixgb_main.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/ixgb/ixgb_main.c 2011-11-17 20:08:53.000000000 -0500 @@ -1052,6 +1052,8 @@ ixgb_set_multi(struct net_device *netdev u32 rctl; int i; + pax_track_stack(); + /* Check for Promiscuous and All Multicast modes */ rctl = IXGB_READ_REG(hw, RCTL); diff -urNp linux-2.6.32.48/drivers/net/ixgb/ixgb_param.c linux-2.6.32.48-openvz/drivers/net/ixgb/ixgb_param.c --- linux-2.6.32.48/drivers/net/ixgb/ixgb_param.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/ixgb/ixgb_param.c 2011-11-17 20:08:53.000000000 -0500 @@ -260,6 +260,9 @@ void __devinit ixgb_check_options(struct ixgb_adapter *adapter) { int bd = adapter->bd_number; + + pax_track_stack(); + if (bd >= IXGB_MAX_NIC) { printk(KERN_NOTICE "Warning: no configuration for board #%i\n", bd); diff -urNp linux-2.6.32.48/drivers/net/ixgbe/ixgbe_type.h linux-2.6.32.48-openvz/drivers/net/ixgbe/ixgbe_type.h --- linux-2.6.32.48/drivers/net/ixgbe/ixgbe_type.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/ixgbe/ixgbe_type.h 2011-11-17 20:09:04.000000000 -0500 @@ -2327,6 +2327,7 @@ struct ixgbe_eeprom_operations { s32 (*validate_checksum)(struct ixgbe_hw *, u16 *); s32 (*update_checksum)(struct ixgbe_hw *); }; +typedef struct ixgbe_eeprom_operations __no_const ixgbe_eeprom_operations_no_const; struct ixgbe_mac_operations { s32 (*init_hw)(struct ixgbe_hw *); @@ -2376,6 +2377,7 @@ struct ixgbe_mac_operations { /* Flow Control */ s32 (*fc_enable)(struct ixgbe_hw *, s32); }; +typedef struct ixgbe_mac_operations __no_const ixgbe_mac_operations_no_const; struct ixgbe_phy_operations { s32 (*identify)(struct ixgbe_hw *); @@ -2394,9 +2396,10 @@ struct ixgbe_phy_operations { s32 (*read_i2c_eeprom)(struct ixgbe_hw *, u8 , u8 *); s32 (*write_i2c_eeprom)(struct ixgbe_hw *, u8, u8); }; +typedef struct ixgbe_phy_operations __no_const ixgbe_phy_operations_no_const; struct ixgbe_eeprom_info { - struct ixgbe_eeprom_operations ops; + ixgbe_eeprom_operations_no_const ops; enum ixgbe_eeprom_type type; u32 semaphore_delay; u16 word_size; @@ -2404,7 +2407,7 @@ struct ixgbe_eeprom_info { }; struct ixgbe_mac_info { - struct ixgbe_mac_operations ops; + ixgbe_mac_operations_no_const ops; enum ixgbe_mac_type type; u8 addr[IXGBE_ETH_LENGTH_OF_ADDRESS]; u8 perm_addr[IXGBE_ETH_LENGTH_OF_ADDRESS]; @@ -2423,7 +2426,7 @@ struct ixgbe_mac_info { }; struct ixgbe_phy_info { - struct ixgbe_phy_operations ops; + ixgbe_phy_operations_no_const ops; struct mdio_if_info mdio; enum ixgbe_phy_type type; u32 id; diff -urNp linux-2.6.32.48/drivers/net/loopback.c linux-2.6.32.48-openvz/drivers/net/loopback.c --- linux-2.6.32.48/drivers/net/loopback.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/loopback.c 2011-11-17 20:08:53.000000000 -0500 @@ -75,6 +75,12 @@ static netdev_tx_t loopback_xmit(struct struct pcpu_lstats *pcpu_lstats, *lb_stats; int len; +#ifdef CONFIG_VE + if (unlikely(get_exec_env()->disable_net)) { + kfree_skb(skb); + return 0; + } +#endif skb_orphan(skb); skb->protocol = eth_type_trans(skb, dev); @@ -153,10 +159,16 @@ static void loopback_dev_free(struct net free_netdev(dev); } +static void loopback_cpt(struct net_device *dev, + struct cpt_ops *ops, struct cpt_context *ctx) +{ +} + static const struct net_device_ops loopback_ops = { .ndo_init = loopback_dev_init, .ndo_start_xmit= loopback_xmit, .ndo_get_stats = loopback_get_stats, + .ndo_cpt = loopback_cpt, }; /* @@ -177,7 +189,8 @@ static void loopback_setup(struct net_de | NETIF_F_NO_CSUM | NETIF_F_HIGHDMA | NETIF_F_LLTX - | NETIF_F_NETNS_LOCAL; + | NETIF_F_NETNS_LOCAL + | NETIF_F_VIRTUAL; dev->ethtool_ops = &loopback_ethtool_ops; dev->header_ops = ð_header_ops; dev->netdev_ops = &loopback_ops; diff -urNp linux-2.6.32.48/drivers/net/Makefile linux-2.6.32.48-openvz/drivers/net/Makefile --- linux-2.6.32.48/drivers/net/Makefile 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/Makefile 2011-11-17 20:09:04.000000000 -0500 @@ -41,6 +41,10 @@ ucc_geth_driver-objs := ucc_geth.o ucc_g obj-$(CONFIG_FSL_PQ_MDIO) += fsl_pq_mdio.o +obj-$(CONFIG_VE_NETDEV) += vznetdev.o +vznetdev-objs := open_vznet.o venet_core.o +obj-$(CONFIG_VE_ETHDEV) += vzethdev.o + # # link order important here # diff -urNp linux-2.6.32.48/drivers/net/mlx4/main.c linux-2.6.32.48-openvz/drivers/net/mlx4/main.c --- linux-2.6.32.48/drivers/net/mlx4/main.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/mlx4/main.c 2011-11-17 20:08:53.000000000 -0500 @@ -38,6 +38,7 @@ #include #include #include +#include #include #include @@ -730,6 +731,8 @@ static int mlx4_init_hca(struct mlx4_dev u64 icm_size; int err; + pax_track_stack(); + err = mlx4_QUERY_FW(dev); if (err) { if (err == -EACCES) diff -urNp linux-2.6.32.48/drivers/net/niu.c linux-2.6.32.48-openvz/drivers/net/niu.c --- linux-2.6.32.48/drivers/net/niu.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/niu.c 2011-11-17 20:08:53.000000000 -0500 @@ -9128,6 +9128,8 @@ static void __devinit niu_try_msix(struc int i, num_irqs, err; u8 first_ldg; + pax_track_stack(); + first_ldg = (NIU_NUM_LDG / parent->num_ports) * np->port; for (i = 0; i < (NIU_NUM_LDG / parent->num_ports); i++) ldg_num_map[i] = first_ldg + i; diff -urNp linux-2.6.32.48/drivers/net/open_vznet.c linux-2.6.32.48-openvz/drivers/net/open_vznet.c --- linux-2.6.32.48/drivers/net/open_vznet.c 1969-12-31 19:00:00.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/open_vznet.c 2011-11-17 20:08:53.000000000 -0500 @@ -0,0 +1,244 @@ +/* + * open_vznet.c + * + * Copyright (C) 2005 SWsoft + * All rights reserved. + * + * Licensing governed by "linux/COPYING.SWsoft" file. + * + */ + +/* + * Virtual Networking device used to change VE ownership on packets + */ + +#include +#include +#include + +#include +#include +#include +#include + +void veip_stop(struct ve_struct *ve) +{ + struct list_head *p, *tmp; + + write_lock_irq(&veip_hash_lock); + if (ve->veip == NULL) + goto unlock; + list_for_each_safe(p, tmp, &ve->veip->ip_lh) { + struct ip_entry_struct *ptr; + ptr = list_entry(p, struct ip_entry_struct, ve_list); + ptr->active_env = NULL; + list_del(&ptr->ve_list); + list_del(&ptr->ip_hash); + kfree(ptr); + } + veip_put(ve->veip); + ve->veip = NULL; + if (!ve_is_super(ve)) + module_put(THIS_MODULE); +unlock: + write_unlock_irq(&veip_hash_lock); +} + +int veip_start(struct ve_struct *ve) +{ + int err, get; + + err = 0; + write_lock_irq(&veip_hash_lock); + get = ve->veip == NULL; + ve->veip = veip_findcreate(ve->veid); + if (ve->veip == NULL) + err = -ENOMEM; + write_unlock_irq(&veip_hash_lock); + if (err == 0 && get && !ve_is_super(ve)) + __module_get(THIS_MODULE); + return err; +} + +int veip_entry_add(struct ve_struct *ve, struct ve_addr_struct *addr) +{ + struct ip_entry_struct *entry, *found; + int err; + + entry = kzalloc(sizeof(struct ip_entry_struct), GFP_KERNEL); + if (entry == NULL) + return -ENOMEM; + + if (ve->veip == NULL) { + /* This can happen if we load venet AFTER ve was started */ + err = veip_start(ve); + if (err < 0) + goto out; + } + + write_lock_irq(&veip_hash_lock); + err = -EADDRINUSE; + found = venet_entry_lookup(addr); + if (found != NULL) + goto out_unlock; + + entry->active_env = ve; + entry->addr = *addr; + ip_entry_hash(entry, ve->veip); + + err = 0; + entry = NULL; +out_unlock: + write_unlock_irq(&veip_hash_lock); +out: + if (entry != NULL) + kfree(entry); + return err; +} + +int veip_entry_del(envid_t veid, struct ve_addr_struct *addr) +{ + struct ip_entry_struct *found; + int err; + + err = -EADDRNOTAVAIL; + write_lock_irq(&veip_hash_lock); + found = venet_entry_lookup(addr); + if (found == NULL) + goto out; + if (found->active_env->veid != veid) + goto out; + + err = 0; + found->active_env = NULL; + + list_del(&found->ip_hash); + list_del(&found->ve_list); + kfree(found); +out: + write_unlock_irq(&veip_hash_lock); + return err; +} + +static int skb_extract_addr(struct sk_buff *skb, + struct ve_addr_struct *addr, int dir) +{ + switch (skb->protocol) { + case __constant_htons(ETH_P_IP): + addr->family = AF_INET; + addr->key[0] = 0; + addr->key[1] = 0; + addr->key[2] = 0; + addr->key[3] = (dir ? ip_hdr(skb)->daddr : ip_hdr(skb)->saddr); + return 0; +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE) + case __constant_htons(ETH_P_IPV6): + addr->family = AF_INET6; + memcpy(&addr->key, dir ? + ipv6_hdr(skb)->daddr.s6_addr32 : + ipv6_hdr(skb)->saddr.s6_addr32, + sizeof(addr->key)); + return 0; +#endif + } + + return -EAFNOSUPPORT; +} + +static struct ve_struct *venet_find_ve(struct sk_buff *skb, int dir) +{ + struct ip_entry_struct *entry; + struct ve_addr_struct addr; + + if (skb_extract_addr(skb, &addr, dir) < 0) + return NULL; + + entry = venet_entry_lookup(&addr); + if (entry == NULL) + return NULL; + + return entry->active_env; +} + +int venet_change_skb_owner(struct sk_buff *skb) +{ + struct ve_struct *ve, *ve_old; + + ve_old = skb->owner_env; + + read_lock(&veip_hash_lock); + if (!ve_is_super(ve_old)) { + /* from VE to host */ + ve = venet_find_ve(skb, 0); + if (ve == NULL) + goto out_drop; + if (!ve_accessible_strict(ve, ve_old)) + goto out_source; + skb->owner_env = get_ve0(); + } else { + /* from host to VE */ + ve = venet_find_ve(skb, 1); + if (ve == NULL) + goto out_drop; + skb->owner_env = ve; + } + read_unlock(&veip_hash_lock); + + return 0; + +out_drop: + read_unlock(&veip_hash_lock); + return -ESRCH; + +out_source: + read_unlock(&veip_hash_lock); + if (net_ratelimit() && skb->protocol == __constant_htons(ETH_P_IP)) { + printk(KERN_WARNING "Dropped packet, source wrong " + "veid=%u src-IP=%u.%u.%u.%u " + "dst-IP=%u.%u.%u.%u\n", + skb->owner_env->veid, + NIPQUAD(ip_hdr(skb)->saddr), + NIPQUAD(ip_hdr(skb)->daddr)); + } + return -EACCES; +} + +#ifdef CONFIG_PROC_FS +int veip_seq_show(struct seq_file *m, void *v) +{ + struct list_head *p; + struct ip_entry_struct *entry; + char s[40]; + + p = (struct list_head *)v; + if (p == ip_entry_hash_table) { + seq_puts(m, "Version: 2.5\n"); + return 0; + } + entry = list_entry(p, struct ip_entry_struct, ip_hash); + veaddr_print(s, sizeof(s), &entry->addr); + seq_printf(m, "%39s %10u\n", s, 0); + return 0; +} +#endif + +__exit void veip_cleanup(void) +{ + int i; + + write_lock_irq(&veip_hash_lock); + for (i = 0; i < VEIP_HASH_SZ; i++) + while (!list_empty(ip_entry_hash_table + i)) { + struct ip_entry_struct *entry; + + entry = list_first_entry(ip_entry_hash_table + i, + struct ip_entry_struct, ip_hash); + list_del(&entry->ip_hash); + kfree(entry); + } + write_unlock_irq(&veip_hash_lock); +} + +MODULE_AUTHOR("SWsoft "); +MODULE_DESCRIPTION("Virtuozzo Virtual Network Device"); +MODULE_LICENSE("GPL v2"); diff -urNp linux-2.6.32.48/drivers/net/pcnet32.c linux-2.6.32.48-openvz/drivers/net/pcnet32.c --- linux-2.6.32.48/drivers/net/pcnet32.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/pcnet32.c 2011-11-17 20:09:04.000000000 -0500 @@ -79,7 +79,7 @@ static int cards_found; /* * VLB I/O addresses */ -static unsigned int pcnet32_portlist[] __initdata = +static unsigned int pcnet32_portlist[] __devinitdata = { 0x300, 0x320, 0x340, 0x360, 0 }; static int pcnet32_debug = 0; @@ -267,7 +267,7 @@ struct pcnet32_private { struct sk_buff **rx_skbuff; dma_addr_t *tx_dma_addr; dma_addr_t *rx_dma_addr; - struct pcnet32_access a; + struct pcnet32_access *a; spinlock_t lock; /* Guard lock */ unsigned int cur_rx, cur_tx; /* The next free ring entry */ unsigned int rx_ring_size; /* current rx ring size */ @@ -457,9 +457,9 @@ static void pcnet32_netif_start(struct n u16 val; netif_wake_queue(dev); - val = lp->a.read_csr(ioaddr, CSR3); + val = lp->a->read_csr(ioaddr, CSR3); val &= 0x00ff; - lp->a.write_csr(ioaddr, CSR3, val); + lp->a->write_csr(ioaddr, CSR3, val); napi_enable(&lp->napi); } @@ -744,7 +744,7 @@ static u32 pcnet32_get_link(struct net_d r = mii_link_ok(&lp->mii_if); } else if (lp->chip_version >= PCNET32_79C970A) { ulong ioaddr = dev->base_addr; /* card base I/O address */ - r = (lp->a.read_bcr(ioaddr, 4) != 0xc0); + r = (lp->a->read_bcr(ioaddr, 4) != 0xc0); } else { /* can not detect link on really old chips */ r = 1; } @@ -806,7 +806,7 @@ static int pcnet32_set_ringparam(struct pcnet32_netif_stop(dev); spin_lock_irqsave(&lp->lock, flags); - lp->a.write_csr(ioaddr, CSR0, CSR0_STOP); /* stop the chip */ + lp->a->write_csr(ioaddr, CSR0, CSR0_STOP); /* stop the chip */ size = min(ering->tx_pending, (unsigned int)TX_MAX_RING_SIZE); @@ -886,7 +886,7 @@ static void pcnet32_ethtool_test(struct static int pcnet32_loopback_test(struct net_device *dev, uint64_t * data1) { struct pcnet32_private *lp = netdev_priv(dev); - struct pcnet32_access *a = &lp->a; /* access to registers */ + struct pcnet32_access *a = lp->a; /* access to registers */ ulong ioaddr = dev->base_addr; /* card base I/O address */ struct sk_buff *skb; /* sk buff */ int x, i; /* counters */ @@ -906,21 +906,21 @@ static int pcnet32_loopback_test(struct pcnet32_netif_stop(dev); spin_lock_irqsave(&lp->lock, flags); - lp->a.write_csr(ioaddr, CSR0, CSR0_STOP); /* stop the chip */ + lp->a->write_csr(ioaddr, CSR0, CSR0_STOP); /* stop the chip */ numbuffs = min(numbuffs, (int)min(lp->rx_ring_size, lp->tx_ring_size)); /* Reset the PCNET32 */ - lp->a.reset(ioaddr); - lp->a.write_csr(ioaddr, CSR4, 0x0915); /* auto tx pad */ + lp->a->reset(ioaddr); + lp->a->write_csr(ioaddr, CSR4, 0x0915); /* auto tx pad */ /* switch pcnet32 to 32bit mode */ - lp->a.write_bcr(ioaddr, 20, 2); + lp->a->write_bcr(ioaddr, 20, 2); /* purge & init rings but don't actually restart */ pcnet32_restart(dev, 0x0000); - lp->a.write_csr(ioaddr, CSR0, CSR0_STOP); /* Set STOP bit */ + lp->a->write_csr(ioaddr, CSR0, CSR0_STOP); /* Set STOP bit */ /* Initialize Transmit buffers. */ size = data_len + 15; @@ -966,10 +966,10 @@ static int pcnet32_loopback_test(struct /* set int loopback in CSR15 */ x = a->read_csr(ioaddr, CSR15) & 0xfffc; - lp->a.write_csr(ioaddr, CSR15, x | 0x0044); + lp->a->write_csr(ioaddr, CSR15, x | 0x0044); teststatus = cpu_to_le16(0x8000); - lp->a.write_csr(ioaddr, CSR0, CSR0_START); /* Set STRT bit */ + lp->a->write_csr(ioaddr, CSR0, CSR0_START); /* Set STRT bit */ /* Check status of descriptors */ for (x = 0; x < numbuffs; x++) { @@ -990,7 +990,7 @@ static int pcnet32_loopback_test(struct } } - lp->a.write_csr(ioaddr, CSR0, CSR0_STOP); /* Set STOP bit */ + lp->a->write_csr(ioaddr, CSR0, CSR0_STOP); /* Set STOP bit */ wmb(); if (netif_msg_hw(lp) && netif_msg_pktdata(lp)) { printk(KERN_DEBUG "%s: RX loopback packets:\n", dev->name); @@ -1039,7 +1039,7 @@ static int pcnet32_loopback_test(struct pcnet32_restart(dev, CSR0_NORMAL); } else { pcnet32_purge_rx_ring(dev); - lp->a.write_bcr(ioaddr, 20, 4); /* return to 16bit mode */ + lp->a->write_bcr(ioaddr, 20, 4); /* return to 16bit mode */ } spin_unlock_irqrestore(&lp->lock, flags); @@ -1049,7 +1049,7 @@ static int pcnet32_loopback_test(struct static void pcnet32_led_blink_callback(struct net_device *dev) { struct pcnet32_private *lp = netdev_priv(dev); - struct pcnet32_access *a = &lp->a; + struct pcnet32_access *a = lp->a; ulong ioaddr = dev->base_addr; unsigned long flags; int i; @@ -1066,7 +1066,7 @@ static void pcnet32_led_blink_callback(s static int pcnet32_phys_id(struct net_device *dev, u32 data) { struct pcnet32_private *lp = netdev_priv(dev); - struct pcnet32_access *a = &lp->a; + struct pcnet32_access *a = lp->a; ulong ioaddr = dev->base_addr; unsigned long flags; int i, regs[4]; @@ -1112,7 +1112,7 @@ static int pcnet32_suspend(struct net_de { int csr5; struct pcnet32_private *lp = netdev_priv(dev); - struct pcnet32_access *a = &lp->a; + struct pcnet32_access *a = lp->a; ulong ioaddr = dev->base_addr; int ticks; @@ -1388,8 +1388,8 @@ static int pcnet32_poll(struct napi_stru spin_lock_irqsave(&lp->lock, flags); if (pcnet32_tx(dev)) { /* reset the chip to clear the error condition, then restart */ - lp->a.reset(ioaddr); - lp->a.write_csr(ioaddr, CSR4, 0x0915); /* auto tx pad */ + lp->a->reset(ioaddr); + lp->a->write_csr(ioaddr, CSR4, 0x0915); /* auto tx pad */ pcnet32_restart(dev, CSR0_START); netif_wake_queue(dev); } @@ -1401,12 +1401,12 @@ static int pcnet32_poll(struct napi_stru __napi_complete(napi); /* clear interrupt masks */ - val = lp->a.read_csr(ioaddr, CSR3); + val = lp->a->read_csr(ioaddr, CSR3); val &= 0x00ff; - lp->a.write_csr(ioaddr, CSR3, val); + lp->a->write_csr(ioaddr, CSR3, val); /* Set interrupt enable. */ - lp->a.write_csr(ioaddr, CSR0, CSR0_INTEN); + lp->a->write_csr(ioaddr, CSR0, CSR0_INTEN); spin_unlock_irqrestore(&lp->lock, flags); } @@ -1429,7 +1429,7 @@ static void pcnet32_get_regs(struct net_ int i, csr0; u16 *buff = ptr; struct pcnet32_private *lp = netdev_priv(dev); - struct pcnet32_access *a = &lp->a; + struct pcnet32_access *a = lp->a; ulong ioaddr = dev->base_addr; unsigned long flags; @@ -1466,9 +1466,9 @@ static void pcnet32_get_regs(struct net_ for (j = 0; j < PCNET32_MAX_PHYS; j++) { if (lp->phymask & (1 << j)) { for (i = 0; i < PCNET32_REGS_PER_PHY; i++) { - lp->a.write_bcr(ioaddr, 33, + lp->a->write_bcr(ioaddr, 33, (j << 5) | i); - *buff++ = lp->a.read_bcr(ioaddr, 34); + *buff++ = lp->a->read_bcr(ioaddr, 34); } } } @@ -1858,7 +1858,7 @@ pcnet32_probe1(unsigned long ioaddr, int ((cards_found >= MAX_UNITS) || full_duplex[cards_found])) lp->options |= PCNET32_PORT_FD; - lp->a = *a; + lp->a = a; /* prior to register_netdev, dev->name is not yet correct */ if (pcnet32_alloc_ring(dev, pci_name(lp->pci_dev))) { @@ -1917,7 +1917,7 @@ pcnet32_probe1(unsigned long ioaddr, int if (lp->mii) { /* lp->phycount and lp->phymask are set to 0 by memset above */ - lp->mii_if.phy_id = ((lp->a.read_bcr(ioaddr, 33)) >> 5) & 0x1f; + lp->mii_if.phy_id = ((lp->a->read_bcr(ioaddr, 33)) >> 5) & 0x1f; /* scan for PHYs */ for (i = 0; i < PCNET32_MAX_PHYS; i++) { unsigned short id1, id2; @@ -1938,7 +1938,7 @@ pcnet32_probe1(unsigned long ioaddr, int "Found PHY %04x:%04x at address %d.\n", id1, id2, i); } - lp->a.write_bcr(ioaddr, 33, (lp->mii_if.phy_id) << 5); + lp->a->write_bcr(ioaddr, 33, (lp->mii_if.phy_id) << 5); if (lp->phycount > 1) { lp->options |= PCNET32_PORT_MII; } @@ -2109,10 +2109,10 @@ static int pcnet32_open(struct net_devic } /* Reset the PCNET32 */ - lp->a.reset(ioaddr); + lp->a->reset(ioaddr); /* switch pcnet32 to 32bit mode */ - lp->a.write_bcr(ioaddr, 20, 2); + lp->a->write_bcr(ioaddr, 20, 2); if (netif_msg_ifup(lp)) printk(KERN_DEBUG @@ -2122,14 +2122,14 @@ static int pcnet32_open(struct net_devic (u32) (lp->init_dma_addr)); /* set/reset autoselect bit */ - val = lp->a.read_bcr(ioaddr, 2) & ~2; + val = lp->a->read_bcr(ioaddr, 2) & ~2; if (lp->options & PCNET32_PORT_ASEL) val |= 2; - lp->a.write_bcr(ioaddr, 2, val); + lp->a->write_bcr(ioaddr, 2, val); /* handle full duplex setting */ if (lp->mii_if.full_duplex) { - val = lp->a.read_bcr(ioaddr, 9) & ~3; + val = lp->a->read_bcr(ioaddr, 9) & ~3; if (lp->options & PCNET32_PORT_FD) { val |= 1; if (lp->options == (PCNET32_PORT_FD | PCNET32_PORT_AUI)) @@ -2139,14 +2139,14 @@ static int pcnet32_open(struct net_devic if (lp->chip_version == 0x2627) val |= 3; } - lp->a.write_bcr(ioaddr, 9, val); + lp->a->write_bcr(ioaddr, 9, val); } /* set/reset GPSI bit in test register */ - val = lp->a.read_csr(ioaddr, 124) & ~0x10; + val = lp->a->read_csr(ioaddr, 124) & ~0x10; if ((lp->options & PCNET32_PORT_PORTSEL) == PCNET32_PORT_GPSI) val |= 0x10; - lp->a.write_csr(ioaddr, 124, val); + lp->a->write_csr(ioaddr, 124, val); /* Allied Telesyn AT 2700/2701 FX are 100Mbit only and do not negotiate */ if (pdev && pdev->subsystem_vendor == PCI_VENDOR_ID_AT && @@ -2167,24 +2167,24 @@ static int pcnet32_open(struct net_devic * duplex, and/or enable auto negotiation, and clear DANAS */ if (lp->mii && !(lp->options & PCNET32_PORT_ASEL)) { - lp->a.write_bcr(ioaddr, 32, - lp->a.read_bcr(ioaddr, 32) | 0x0080); + lp->a->write_bcr(ioaddr, 32, + lp->a->read_bcr(ioaddr, 32) | 0x0080); /* disable Auto Negotiation, set 10Mpbs, HD */ - val = lp->a.read_bcr(ioaddr, 32) & ~0xb8; + val = lp->a->read_bcr(ioaddr, 32) & ~0xb8; if (lp->options & PCNET32_PORT_FD) val |= 0x10; if (lp->options & PCNET32_PORT_100) val |= 0x08; - lp->a.write_bcr(ioaddr, 32, val); + lp->a->write_bcr(ioaddr, 32, val); } else { if (lp->options & PCNET32_PORT_ASEL) { - lp->a.write_bcr(ioaddr, 32, - lp->a.read_bcr(ioaddr, + lp->a->write_bcr(ioaddr, 32, + lp->a->read_bcr(ioaddr, 32) | 0x0080); /* enable auto negotiate, setup, disable fd */ - val = lp->a.read_bcr(ioaddr, 32) & ~0x98; + val = lp->a->read_bcr(ioaddr, 32) & ~0x98; val |= 0x20; - lp->a.write_bcr(ioaddr, 32, val); + lp->a->write_bcr(ioaddr, 32, val); } } } else { @@ -2197,10 +2197,10 @@ static int pcnet32_open(struct net_devic * There is really no good other way to handle multiple PHYs * other than turning off all automatics */ - val = lp->a.read_bcr(ioaddr, 2); - lp->a.write_bcr(ioaddr, 2, val & ~2); - val = lp->a.read_bcr(ioaddr, 32); - lp->a.write_bcr(ioaddr, 32, val & ~(1 << 7)); /* stop MII manager */ + val = lp->a->read_bcr(ioaddr, 2); + lp->a->write_bcr(ioaddr, 2, val & ~2); + val = lp->a->read_bcr(ioaddr, 32); + lp->a->write_bcr(ioaddr, 32, val & ~(1 << 7)); /* stop MII manager */ if (!(lp->options & PCNET32_PORT_ASEL)) { /* setup ecmd */ @@ -2210,7 +2210,7 @@ static int pcnet32_open(struct net_devic ecmd.speed = lp-> options & PCNET32_PORT_100 ? SPEED_100 : SPEED_10; - bcr9 = lp->a.read_bcr(ioaddr, 9); + bcr9 = lp->a->read_bcr(ioaddr, 9); if (lp->options & PCNET32_PORT_FD) { ecmd.duplex = DUPLEX_FULL; @@ -2219,7 +2219,7 @@ static int pcnet32_open(struct net_devic ecmd.duplex = DUPLEX_HALF; bcr9 |= ~(1 << 0); } - lp->a.write_bcr(ioaddr, 9, bcr9); + lp->a->write_bcr(ioaddr, 9, bcr9); } for (i = 0; i < PCNET32_MAX_PHYS; i++) { @@ -2252,9 +2252,9 @@ static int pcnet32_open(struct net_devic #ifdef DO_DXSUFLO if (lp->dxsuflo) { /* Disable transmit stop on underflow */ - val = lp->a.read_csr(ioaddr, CSR3); + val = lp->a->read_csr(ioaddr, CSR3); val |= 0x40; - lp->a.write_csr(ioaddr, CSR3, val); + lp->a->write_csr(ioaddr, CSR3, val); } #endif @@ -2270,11 +2270,11 @@ static int pcnet32_open(struct net_devic napi_enable(&lp->napi); /* Re-initialize the PCNET32, and start it when done. */ - lp->a.write_csr(ioaddr, 1, (lp->init_dma_addr & 0xffff)); - lp->a.write_csr(ioaddr, 2, (lp->init_dma_addr >> 16)); + lp->a->write_csr(ioaddr, 1, (lp->init_dma_addr & 0xffff)); + lp->a->write_csr(ioaddr, 2, (lp->init_dma_addr >> 16)); - lp->a.write_csr(ioaddr, CSR4, 0x0915); /* auto tx pad */ - lp->a.write_csr(ioaddr, CSR0, CSR0_INIT); + lp->a->write_csr(ioaddr, CSR4, 0x0915); /* auto tx pad */ + lp->a->write_csr(ioaddr, CSR0, CSR0_INIT); netif_start_queue(dev); @@ -2286,20 +2286,20 @@ static int pcnet32_open(struct net_devic i = 0; while (i++ < 100) - if (lp->a.read_csr(ioaddr, CSR0) & CSR0_IDON) + if (lp->a->read_csr(ioaddr, CSR0) & CSR0_IDON) break; /* * We used to clear the InitDone bit, 0x0100, here but Mark Stockton * reports that doing so triggers a bug in the '974. */ - lp->a.write_csr(ioaddr, CSR0, CSR0_NORMAL); + lp->a->write_csr(ioaddr, CSR0, CSR0_NORMAL); if (netif_msg_ifup(lp)) printk(KERN_DEBUG "%s: pcnet32 open after %d ticks, init block %#x csr0 %4.4x.\n", dev->name, i, (u32) (lp->init_dma_addr), - lp->a.read_csr(ioaddr, CSR0)); + lp->a->read_csr(ioaddr, CSR0)); spin_unlock_irqrestore(&lp->lock, flags); @@ -2313,7 +2313,7 @@ static int pcnet32_open(struct net_devic * Switch back to 16bit mode to avoid problems with dumb * DOS packet driver after a warm reboot */ - lp->a.write_bcr(ioaddr, 20, 4); + lp->a->write_bcr(ioaddr, 20, 4); err_free_irq: spin_unlock_irqrestore(&lp->lock, flags); @@ -2420,7 +2420,7 @@ static void pcnet32_restart(struct net_d /* wait for stop */ for (i = 0; i < 100; i++) - if (lp->a.read_csr(ioaddr, CSR0) & CSR0_STOP) + if (lp->a->read_csr(ioaddr, CSR0) & CSR0_STOP) break; if (i >= 100 && netif_msg_drv(lp)) @@ -2433,13 +2433,13 @@ static void pcnet32_restart(struct net_d return; /* ReInit Ring */ - lp->a.write_csr(ioaddr, CSR0, CSR0_INIT); + lp->a->write_csr(ioaddr, CSR0, CSR0_INIT); i = 0; while (i++ < 1000) - if (lp->a.read_csr(ioaddr, CSR0) & CSR0_IDON) + if (lp->a->read_csr(ioaddr, CSR0) & CSR0_IDON) break; - lp->a.write_csr(ioaddr, CSR0, csr0_bits); + lp->a->write_csr(ioaddr, CSR0, csr0_bits); } static void pcnet32_tx_timeout(struct net_device *dev) @@ -2452,8 +2452,8 @@ static void pcnet32_tx_timeout(struct ne if (pcnet32_debug & NETIF_MSG_DRV) printk(KERN_ERR "%s: transmit timed out, status %4.4x, resetting.\n", - dev->name, lp->a.read_csr(ioaddr, CSR0)); - lp->a.write_csr(ioaddr, CSR0, CSR0_STOP); + dev->name, lp->a->read_csr(ioaddr, CSR0)); + lp->a->write_csr(ioaddr, CSR0, CSR0_STOP); dev->stats.tx_errors++; if (netif_msg_tx_err(lp)) { int i; @@ -2497,7 +2497,7 @@ static netdev_tx_t pcnet32_start_xmit(st if (netif_msg_tx_queued(lp)) { printk(KERN_DEBUG "%s: pcnet32_start_xmit() called, csr0 %4.4x.\n", - dev->name, lp->a.read_csr(ioaddr, CSR0)); + dev->name, lp->a->read_csr(ioaddr, CSR0)); } /* Default status -- will not enable Successful-TxDone @@ -2528,7 +2528,7 @@ static netdev_tx_t pcnet32_start_xmit(st dev->stats.tx_bytes += skb->len; /* Trigger an immediate send poll. */ - lp->a.write_csr(ioaddr, CSR0, CSR0_INTEN | CSR0_TXPOLL); + lp->a->write_csr(ioaddr, CSR0, CSR0_INTEN | CSR0_TXPOLL); dev->trans_start = jiffies; @@ -2555,18 +2555,18 @@ pcnet32_interrupt(int irq, void *dev_id) spin_lock(&lp->lock); - csr0 = lp->a.read_csr(ioaddr, CSR0); + csr0 = lp->a->read_csr(ioaddr, CSR0); while ((csr0 & 0x8f00) && --boguscnt >= 0) { if (csr0 == 0xffff) { break; /* PCMCIA remove happened */ } /* Acknowledge all of the current interrupt sources ASAP. */ - lp->a.write_csr(ioaddr, CSR0, csr0 & ~0x004f); + lp->a->write_csr(ioaddr, CSR0, csr0 & ~0x004f); if (netif_msg_intr(lp)) printk(KERN_DEBUG "%s: interrupt csr0=%#2.2x new csr=%#2.2x.\n", - dev->name, csr0, lp->a.read_csr(ioaddr, CSR0)); + dev->name, csr0, lp->a->read_csr(ioaddr, CSR0)); /* Log misc errors. */ if (csr0 & 0x4000) @@ -2595,19 +2595,19 @@ pcnet32_interrupt(int irq, void *dev_id) if (napi_schedule_prep(&lp->napi)) { u16 val; /* set interrupt masks */ - val = lp->a.read_csr(ioaddr, CSR3); + val = lp->a->read_csr(ioaddr, CSR3); val |= 0x5f00; - lp->a.write_csr(ioaddr, CSR3, val); + lp->a->write_csr(ioaddr, CSR3, val); __napi_schedule(&lp->napi); break; } - csr0 = lp->a.read_csr(ioaddr, CSR0); + csr0 = lp->a->read_csr(ioaddr, CSR0); } if (netif_msg_intr(lp)) printk(KERN_DEBUG "%s: exiting interrupt, csr0=%#4.4x.\n", - dev->name, lp->a.read_csr(ioaddr, CSR0)); + dev->name, lp->a->read_csr(ioaddr, CSR0)); spin_unlock(&lp->lock); @@ -2627,21 +2627,21 @@ static int pcnet32_close(struct net_devi spin_lock_irqsave(&lp->lock, flags); - dev->stats.rx_missed_errors = lp->a.read_csr(ioaddr, 112); + dev->stats.rx_missed_errors = lp->a->read_csr(ioaddr, 112); if (netif_msg_ifdown(lp)) printk(KERN_DEBUG "%s: Shutting down ethercard, status was %2.2x.\n", - dev->name, lp->a.read_csr(ioaddr, CSR0)); + dev->name, lp->a->read_csr(ioaddr, CSR0)); /* We stop the PCNET32 here -- it occasionally polls memory if we don't. */ - lp->a.write_csr(ioaddr, CSR0, CSR0_STOP); + lp->a->write_csr(ioaddr, CSR0, CSR0_STOP); /* * Switch back to 16bit mode to avoid problems with dumb * DOS packet driver after a warm reboot */ - lp->a.write_bcr(ioaddr, 20, 4); + lp->a->write_bcr(ioaddr, 20, 4); spin_unlock_irqrestore(&lp->lock, flags); @@ -2664,7 +2664,7 @@ static struct net_device_stats *pcnet32_ unsigned long flags; spin_lock_irqsave(&lp->lock, flags); - dev->stats.rx_missed_errors = lp->a.read_csr(ioaddr, 112); + dev->stats.rx_missed_errors = lp->a->read_csr(ioaddr, 112); spin_unlock_irqrestore(&lp->lock, flags); return &dev->stats; @@ -2686,10 +2686,10 @@ static void pcnet32_load_multicast(struc if (dev->flags & IFF_ALLMULTI) { ib->filter[0] = cpu_to_le32(~0U); ib->filter[1] = cpu_to_le32(~0U); - lp->a.write_csr(ioaddr, PCNET32_MC_FILTER, 0xffff); - lp->a.write_csr(ioaddr, PCNET32_MC_FILTER+1, 0xffff); - lp->a.write_csr(ioaddr, PCNET32_MC_FILTER+2, 0xffff); - lp->a.write_csr(ioaddr, PCNET32_MC_FILTER+3, 0xffff); + lp->a->write_csr(ioaddr, PCNET32_MC_FILTER, 0xffff); + lp->a->write_csr(ioaddr, PCNET32_MC_FILTER+1, 0xffff); + lp->a->write_csr(ioaddr, PCNET32_MC_FILTER+2, 0xffff); + lp->a->write_csr(ioaddr, PCNET32_MC_FILTER+3, 0xffff); return; } /* clear the multicast filter */ @@ -2710,7 +2710,7 @@ static void pcnet32_load_multicast(struc mcast_table[crc >> 4] |= cpu_to_le16(1 << (crc & 0xf)); } for (i = 0; i < 4; i++) - lp->a.write_csr(ioaddr, PCNET32_MC_FILTER + i, + lp->a->write_csr(ioaddr, PCNET32_MC_FILTER + i, le16_to_cpu(mcast_table[i])); return; } @@ -2726,7 +2726,7 @@ static void pcnet32_set_multicast_list(s spin_lock_irqsave(&lp->lock, flags); suspended = pcnet32_suspend(dev, &flags, 0); - csr15 = lp->a.read_csr(ioaddr, CSR15); + csr15 = lp->a->read_csr(ioaddr, CSR15); if (dev->flags & IFF_PROMISC) { /* Log any net taps. */ if (netif_msg_hw(lp)) @@ -2735,21 +2735,21 @@ static void pcnet32_set_multicast_list(s lp->init_block->mode = cpu_to_le16(0x8000 | (lp->options & PCNET32_PORT_PORTSEL) << 7); - lp->a.write_csr(ioaddr, CSR15, csr15 | 0x8000); + lp->a->write_csr(ioaddr, CSR15, csr15 | 0x8000); } else { lp->init_block->mode = cpu_to_le16((lp->options & PCNET32_PORT_PORTSEL) << 7); - lp->a.write_csr(ioaddr, CSR15, csr15 & 0x7fff); + lp->a->write_csr(ioaddr, CSR15, csr15 & 0x7fff); pcnet32_load_multicast(dev); } if (suspended) { int csr5; /* clear SUSPEND (SPND) - CSR5 bit 0 */ - csr5 = lp->a.read_csr(ioaddr, CSR5); - lp->a.write_csr(ioaddr, CSR5, csr5 & (~CSR5_SUSPEND)); + csr5 = lp->a->read_csr(ioaddr, CSR5); + lp->a->write_csr(ioaddr, CSR5, csr5 & (~CSR5_SUSPEND)); } else { - lp->a.write_csr(ioaddr, CSR0, CSR0_STOP); + lp->a->write_csr(ioaddr, CSR0, CSR0_STOP); pcnet32_restart(dev, CSR0_NORMAL); netif_wake_queue(dev); } @@ -2767,8 +2767,8 @@ static int mdio_read(struct net_device * if (!lp->mii) return 0; - lp->a.write_bcr(ioaddr, 33, ((phy_id & 0x1f) << 5) | (reg_num & 0x1f)); - val_out = lp->a.read_bcr(ioaddr, 34); + lp->a->write_bcr(ioaddr, 33, ((phy_id & 0x1f) << 5) | (reg_num & 0x1f)); + val_out = lp->a->read_bcr(ioaddr, 34); return val_out; } @@ -2782,8 +2782,8 @@ static void mdio_write(struct net_device if (!lp->mii) return; - lp->a.write_bcr(ioaddr, 33, ((phy_id & 0x1f) << 5) | (reg_num & 0x1f)); - lp->a.write_bcr(ioaddr, 34, val); + lp->a->write_bcr(ioaddr, 33, ((phy_id & 0x1f) << 5) | (reg_num & 0x1f)); + lp->a->write_bcr(ioaddr, 34, val); } static int pcnet32_ioctl(struct net_device *dev, struct ifreq *rq, int cmd) @@ -2862,7 +2862,7 @@ static void pcnet32_check_media(struct n curr_link = mii_link_ok(&lp->mii_if); } else { ulong ioaddr = dev->base_addr; /* card base I/O address */ - curr_link = (lp->a.read_bcr(ioaddr, 4) != 0xc0); + curr_link = (lp->a->read_bcr(ioaddr, 4) != 0xc0); } if (!curr_link) { if (prev_link || verbose) { @@ -2887,13 +2887,13 @@ static void pcnet32_check_media(struct n (ecmd.duplex == DUPLEX_FULL) ? "full" : "half"); } - bcr9 = lp->a.read_bcr(dev->base_addr, 9); + bcr9 = lp->a->read_bcr(dev->base_addr, 9); if ((bcr9 & (1 << 0)) != lp->mii_if.full_duplex) { if (lp->mii_if.full_duplex) bcr9 |= (1 << 0); else bcr9 &= ~(1 << 0); - lp->a.write_bcr(dev->base_addr, 9, bcr9); + lp->a->write_bcr(dev->base_addr, 9, bcr9); } } else { if (netif_msg_link(lp)) diff -urNp linux-2.6.32.48/drivers/net/ppp_generic.c linux-2.6.32.48-openvz/drivers/net/ppp_generic.c --- linux-2.6.32.48/drivers/net/ppp_generic.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/ppp_generic.c 2011-11-17 20:08:53.000000000 -0500 @@ -53,6 +53,9 @@ #include #include +#include +#include + #define PPP_VERSION "2.4.2" /* @@ -366,8 +369,10 @@ static int ppp_open(struct inode *inode, /* * This could (should?) be enforced by the permissions on /dev/ppp. */ - if (!capable(CAP_NET_ADMIN)) + if (!capable(CAP_VE_NET_ADMIN)) return -EPERM; + if (!net_generic(get_exec_env()->ve_netns, ppp_net_id)) /* no VE_FEATURE_PPP */ + return -EACCES; return 0; } @@ -867,6 +872,9 @@ static __net_init int ppp_init_net(struc struct ppp_net *pn; int err; + if (!(get_exec_env()->features & VE_FEATURE_PPP)) + return 0; + pn = kzalloc(sizeof(*pn), GFP_KERNEL); if (!pn) return -ENOMEM; @@ -893,6 +901,9 @@ static __net_exit void ppp_exit_net(stru struct ppp_net *pn; pn = net_generic(net, ppp_net_id); + if (!pn) /* no VE_FEATURE_PPP */ + return; + idr_destroy(&pn->units_idr); /* * if someone has cached our net then @@ -1053,7 +1064,7 @@ static void ppp_setup(struct net_device dev->tx_queue_len = 3; dev->type = ARPHRD_PPP; dev->flags = IFF_POINTOPOINT | IFF_NOARP | IFF_MULTICAST; - dev->features |= NETIF_F_NETNS_LOCAL; + dev->features |= NETIF_F_NETNS_LOCAL | NETIF_F_VIRTUAL; dev->priv_flags &= ~IFF_XMIT_DST_RELEASE; } @@ -2568,16 +2579,16 @@ ppp_create_interface(struct net *net, in */ dev_net_set(dev, net); - ret = -EEXIST; mutex_lock(&pn->all_ppp_mutex); if (unit < 0) { unit = unit_get(&pn->units_idr, ppp); if (unit < 0) { - *retp = unit; + ret = unit; goto out2; } } else { + ret = -EEXIST; if (unit_find(&pn->units_idr, unit)) goto out2; /* unit already exists */ /* @@ -2652,10 +2663,10 @@ static void ppp_shutdown_interface(struc ppp->closing = 1; ppp_unlock(ppp); unregister_netdev(ppp->dev); + unit_put(&pn->units_idr, ppp->file.index); } else ppp_unlock(ppp); - unit_put(&pn->units_idr, ppp->file.index); ppp->file.dead = 1; ppp->owner = NULL; wake_up_interruptible(&ppp->file.rwait); @@ -2843,8 +2854,7 @@ static void __exit ppp_cleanup(void) * by holding all_ppp_mutex */ -/* associate pointer with specified number */ -static int unit_set(struct idr *p, void *ptr, int n) +static int __unit_alloc(struct idr *p, void *ptr, int n) { int unit, err; @@ -2855,10 +2865,24 @@ again: } err = idr_get_new_above(p, ptr, n, &unit); - if (err == -EAGAIN) - goto again; + if (err < 0) { + if (err == -EAGAIN) + goto again; + return err; + } - if (unit != n) { + return unit; +} + +/* associate pointer with specified number */ +static int unit_set(struct idr *p, void *ptr, int n) +{ + int unit; + + unit = __unit_alloc(p, ptr, n); + if (unit < 0) + return unit; + else if (unit != n) { idr_remove(p, unit); return -EINVAL; } @@ -2869,19 +2893,7 @@ again: /* get new free unit number and associate pointer with it */ static int unit_get(struct idr *p, void *ptr) { - int unit, err; - -again: - if (!idr_pre_get(p, GFP_KERNEL)) { - printk(KERN_ERR "PPP: No free memory for idr\n"); - return -ENOMEM; - } - - err = idr_get_new_above(p, ptr, 0, &unit); - if (err == -EAGAIN) - goto again; - - return unit; + return __unit_alloc(p, ptr, 0); } /* put unit number back to a pool */ diff -urNp linux-2.6.32.48/drivers/net/pppoe.c linux-2.6.32.48-openvz/drivers/net/pppoe.c --- linux-2.6.32.48/drivers/net/pppoe.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/pppoe.c 2011-11-17 20:08:53.000000000 -0500 @@ -77,6 +77,7 @@ #include #include #include +#include #include #include @@ -452,6 +453,8 @@ static int pppoe_rcv(struct sk_buff *skb goto drop; pn = pppoe_pernet(dev_net(dev)); + if (!pn) /* no VE_FEATURE_PPP */ + goto drop; /* Note that get_item does a sock_hold(), so sk_pppox(po) * is known to be safe. @@ -494,6 +497,9 @@ static int pppoe_disc_rcv(struct sk_buff goto abort; pn = pppoe_pernet(dev_net(dev)); + if (!pn) /* no VE_FEATURE_PPP */ + goto abort; + po = get_item(pn, ph->sid, eth_hdr(skb)->h_source, dev->ifindex); if (po) { struct sock *sk = sk_pppox(po); @@ -547,6 +553,9 @@ static int pppoe_create(struct net *net, { struct sock *sk; + if (!(get_exec_env()->features & VE_FEATURE_PPP)) + return -EACCES; + sk = sk_alloc(net, PF_PPPOX, GFP_KERNEL, &pppoe_sk_proto); if (!sk) return -ENOMEM; @@ -1144,6 +1153,9 @@ static __net_init int pppoe_init_net(str struct proc_dir_entry *pde; int err; + if (!(get_exec_env()->features & VE_FEATURE_PPP)) + return 0; + pn = kzalloc(sizeof(*pn), GFP_KERNEL); if (!pn) return -ENOMEM; @@ -1173,8 +1185,11 @@ static __net_exit void pppoe_exit_net(st { struct pppoe_net *pn; - proc_net_remove(net, "pppoe"); pn = net_generic(net, pppoe_net_id); + if (!pn) /* no VE_FEATURE_PPP */ + return; + + proc_net_remove(net, "pppoe"); /* * if someone has cached our net then * further net_generic call will return NULL diff -urNp linux-2.6.32.48/drivers/net/pppol2tp.c linux-2.6.32.48-openvz/drivers/net/pppol2tp.c --- linux-2.6.32.48/drivers/net/pppol2tp.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/pppol2tp.c 2011-11-17 20:08:53.000000000 -0500 @@ -97,6 +97,7 @@ #include #include #include +#include #include #include @@ -1591,6 +1592,9 @@ static int pppol2tp_create(struct net *n int error = -ENOMEM; struct sock *sk; + if (!(get_exec_env()->features & VE_FEATURE_PPP)) + return -EACCES; + sk = sk_alloc(net, PF_PPPOX, GFP_KERNEL, &pppol2tp_sk_proto); if (!sk) goto out; @@ -2609,6 +2613,9 @@ static __net_init int pppol2tp_init_net( struct proc_dir_entry *pde; int err; + if (!(get_exec_env()->features & VE_FEATURE_PPP)) + return 0; + pn = kzalloc(sizeof(*pn), GFP_KERNEL); if (!pn) return -ENOMEM; @@ -2639,8 +2646,11 @@ static __net_exit void pppol2tp_exit_net { struct pppoe_net *pn; - proc_net_remove(net, "pppol2tp"); pn = net_generic(net, pppol2tp_net_id); + if (!pn) /* no VE_FEATURE_PPP */ + return; + + proc_net_remove(net, "pppol2tp"); /* * if someone has cached our net then * further net_generic call will return NULL diff -urNp linux-2.6.32.48/drivers/net/tg3.h linux-2.6.32.48-openvz/drivers/net/tg3.h --- linux-2.6.32.48/drivers/net/tg3.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/tg3.h 2011-11-17 20:08:53.000000000 -0500 @@ -95,6 +95,7 @@ #define CHIPREV_ID_5750_A0 0x4000 #define CHIPREV_ID_5750_A1 0x4001 #define CHIPREV_ID_5750_A3 0x4003 +#define CHIPREV_ID_5750_C1 0x4201 #define CHIPREV_ID_5750_C2 0x4202 #define CHIPREV_ID_5752_A0_HW 0x5000 #define CHIPREV_ID_5752_A0 0x6000 diff -urNp linux-2.6.32.48/drivers/net/tokenring/abyss.c linux-2.6.32.48-openvz/drivers/net/tokenring/abyss.c --- linux-2.6.32.48/drivers/net/tokenring/abyss.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/tokenring/abyss.c 2011-11-17 20:09:04.000000000 -0500 @@ -451,10 +451,12 @@ static struct pci_driver abyss_driver = static int __init abyss_init (void) { - abyss_netdev_ops = tms380tr_netdev_ops; + pax_open_kernel(); + memcpy((void *)&abyss_netdev_ops, &tms380tr_netdev_ops, sizeof(tms380tr_netdev_ops)); - abyss_netdev_ops.ndo_open = abyss_open; - abyss_netdev_ops.ndo_stop = abyss_close; + *(void **)&abyss_netdev_ops.ndo_open = abyss_open; + *(void **)&abyss_netdev_ops.ndo_stop = abyss_close; + pax_close_kernel(); return pci_register_driver(&abyss_driver); } diff -urNp linux-2.6.32.48/drivers/net/tokenring/madgemc.c linux-2.6.32.48-openvz/drivers/net/tokenring/madgemc.c --- linux-2.6.32.48/drivers/net/tokenring/madgemc.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/tokenring/madgemc.c 2011-11-17 20:09:04.000000000 -0500 @@ -755,9 +755,11 @@ static struct mca_driver madgemc_driver static int __init madgemc_init (void) { - madgemc_netdev_ops = tms380tr_netdev_ops; - madgemc_netdev_ops.ndo_open = madgemc_open; - madgemc_netdev_ops.ndo_stop = madgemc_close; + pax_open_kernel(); + memcpy((void *)&madgemc_netdev_ops, &tms380tr_netdev_ops, sizeof(tms380tr_netdev_ops)); + *(void **)&madgemc_netdev_ops.ndo_open = madgemc_open; + *(void **)&madgemc_netdev_ops.ndo_stop = madgemc_close; + pax_close_kernel(); return mca_register_driver (&madgemc_driver); } diff -urNp linux-2.6.32.48/drivers/net/tokenring/proteon.c linux-2.6.32.48-openvz/drivers/net/tokenring/proteon.c --- linux-2.6.32.48/drivers/net/tokenring/proteon.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/tokenring/proteon.c 2011-11-17 20:09:04.000000000 -0500 @@ -353,9 +353,11 @@ static int __init proteon_init(void) struct platform_device *pdev; int i, num = 0, err = 0; - proteon_netdev_ops = tms380tr_netdev_ops; - proteon_netdev_ops.ndo_open = proteon_open; - proteon_netdev_ops.ndo_stop = tms380tr_close; + pax_open_kernel(); + memcpy((void *)&proteon_netdev_ops, &tms380tr_netdev_ops, sizeof(tms380tr_netdev_ops)); + *(void **)&proteon_netdev_ops.ndo_open = proteon_open; + *(void **)&proteon_netdev_ops.ndo_stop = tms380tr_close; + pax_close_kernel(); err = platform_driver_register(&proteon_driver); if (err) diff -urNp linux-2.6.32.48/drivers/net/tokenring/skisa.c linux-2.6.32.48-openvz/drivers/net/tokenring/skisa.c --- linux-2.6.32.48/drivers/net/tokenring/skisa.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/tokenring/skisa.c 2011-11-17 20:09:04.000000000 -0500 @@ -363,9 +363,11 @@ static int __init sk_isa_init(void) struct platform_device *pdev; int i, num = 0, err = 0; - sk_isa_netdev_ops = tms380tr_netdev_ops; - sk_isa_netdev_ops.ndo_open = sk_isa_open; - sk_isa_netdev_ops.ndo_stop = tms380tr_close; + pax_open_kernel(); + memcpy((void *)&sk_isa_netdev_ops, &tms380tr_netdev_ops, sizeof(tms380tr_netdev_ops)); + *(void **)&sk_isa_netdev_ops.ndo_open = sk_isa_open; + *(void **)&sk_isa_netdev_ops.ndo_stop = tms380tr_close; + pax_close_kernel(); err = platform_driver_register(&sk_isa_driver); if (err) diff -urNp linux-2.6.32.48/drivers/net/tulip/de2104x.c linux-2.6.32.48-openvz/drivers/net/tulip/de2104x.c --- linux-2.6.32.48/drivers/net/tulip/de2104x.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/tulip/de2104x.c 2011-11-17 20:08:53.000000000 -0500 @@ -1785,6 +1785,8 @@ static void __devinit de21041_get_srom_i struct de_srom_info_leaf *il; void *bufp; + pax_track_stack(); + /* download entire eeprom */ for (i = 0; i < DE_EEPROM_WORDS; i++) ((__le16 *)ee_data)[i] = diff -urNp linux-2.6.32.48/drivers/net/tulip/de4x5.c linux-2.6.32.48-openvz/drivers/net/tulip/de4x5.c --- linux-2.6.32.48/drivers/net/tulip/de4x5.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/tulip/de4x5.c 2011-11-17 20:08:53.000000000 -0500 @@ -5472,7 +5472,7 @@ de4x5_ioctl(struct net_device *dev, stru for (i=0; idev_addr[i]; } - if (copy_to_user(ioc->data, tmp.addr, ioc->len)) return -EFAULT; + if (ioc->len > sizeof tmp.addr || copy_to_user(ioc->data, tmp.addr, ioc->len)) return -EFAULT; break; case DE4X5_SET_HWADDR: /* Set the hardware address */ @@ -5512,7 +5512,7 @@ de4x5_ioctl(struct net_device *dev, stru spin_lock_irqsave(&lp->lock, flags); memcpy(&statbuf, &lp->pktStats, ioc->len); spin_unlock_irqrestore(&lp->lock, flags); - if (copy_to_user(ioc->data, &statbuf, ioc->len)) + if (ioc->len > sizeof statbuf || copy_to_user(ioc->data, &statbuf, ioc->len)) return -EFAULT; break; } diff -urNp linux-2.6.32.48/drivers/net/tun.c linux-2.6.32.48-openvz/drivers/net/tun.c --- linux-2.6.32.48/drivers/net/tun.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/tun.c 2011-11-17 20:08:53.000000000 -0500 @@ -61,6 +61,7 @@ #include #include #include +#include #include #include #include @@ -69,6 +70,9 @@ #include #include +#include +#include + /* Uncomment to enable debugging */ /* #define TUN_DEBUG 1 */ @@ -93,6 +97,7 @@ struct tun_file { atomic_t count; struct tun_struct *tun; struct net *net; + struct file *file; }; struct tun_sock; @@ -124,6 +129,15 @@ static inline struct tun_sock *tun_sk(st return container_of(sk, struct tun_sock, sk); } +static void __tun_attach(struct tun_struct *tun, struct tun_file *tfile) +{ + tfile->tun = tun; + tun->tfile = tfile; + dev_hold(tun->dev); + sock_hold(tun->socket.sk); + atomic_inc(&tfile->count); +} + static int tun_attach(struct tun_struct *tun, struct file *file) { struct tun_file *tfile = file->private_data; @@ -142,12 +156,7 @@ static int tun_attach(struct tun_struct goto out; err = 0; - tfile->tun = tun; - tun->tfile = tfile; - dev_hold(tun->dev); - sock_hold(tun->socket.sk); - atomic_inc(&tfile->count); - + __tun_attach(tun, tfile); out: netif_tx_unlock_bh(tun->dev); return err; @@ -418,12 +427,16 @@ tun_net_change_mtu(struct net_device *de return 0; } +static void tun_cpt(struct net_device *dev, + struct cpt_ops *ops, struct cpt_context * ctx); + static const struct net_device_ops tun_netdev_ops = { .ndo_uninit = tun_net_uninit, .ndo_open = tun_net_open, .ndo_stop = tun_net_close, .ndo_start_xmit = tun_net_xmit, .ndo_change_mtu = tun_net_change_mtu, + .ndo_cpt = tun_cpt, }; static const struct net_device_ops tap_netdev_ops = { @@ -435,6 +448,7 @@ static const struct net_device_ops tap_n .ndo_set_multicast_list = tun_net_mclist, .ndo_set_mac_address = eth_mac_addr, .ndo_validate_addr = eth_validate_addr, + .ndo_cpt = tun_cpt, }; /* Initialize net device. */ @@ -513,12 +527,8 @@ static inline struct sk_buff *tun_alloc_ struct sk_buff *skb; int err; - /* Under a page? Don't bother with paged skb. */ - if (prepad + len < PAGE_SIZE || !linear) - linear = len; - - skb = sock_alloc_send_pskb(sk, prepad + linear, len - linear, noblock, - &err); + linear = len; + skb = sock_alloc_send_skb(sk, prepad + linear, noblock, &err); if (!skb) return ERR_PTR(err); @@ -819,6 +829,7 @@ static void tun_setup(struct net_device dev->ethtool_ops = &tun_ethtool_ops; dev->destructor = tun_free_netdev; + dev->features |= NETIF_F_VIRTUAL; } /* Trivial set of netlink ops to allow deleting tun or tap @@ -864,6 +875,29 @@ static struct proto tun_proto = { .obj_size = sizeof(struct tun_sock), }; +static int tun_sk_alloc_init(struct net *net, struct tun_struct *tun, + struct sock **psk) +{ + struct sock *sk; + + sk = sk_alloc(net, AF_UNSPEC, GFP_KERNEL, &tun_proto); + if (!sk) + return -ENOMEM; + + init_waitqueue_head(&tun->socket.wait); + sock_init_data(&tun->socket, sk); + sk->sk_write_space = tun_sock_write_space; + sk->sk_sndbuf = INT_MAX; + + container_of(sk, struct tun_sock, sk)->tun = tun; + + security_tun_dev_post_create(sk); + + *psk = sk; + return 0; + +} + static int tun_flags(struct tun_struct *tun) { int flags = 0; @@ -932,7 +966,7 @@ static int tun_set_iff(struct net *net, if (((tun->owner != -1 && cred->euid != tun->owner) || (tun->group != -1 && !in_egroup_p(tun->group))) && - !capable(CAP_NET_ADMIN)) + !capable(CAP_NET_ADMIN) && !capable(CAP_VE_NET_ADMIN)) return -EPERM; err = security_tun_dev_attach(tun->socket.sk); if (err < 0) @@ -946,7 +980,7 @@ static int tun_set_iff(struct net *net, char *name; unsigned long flags = 0; - if (!capable(CAP_NET_ADMIN)) + if (!capable(CAP_NET_ADMIN) && !capable(CAP_VE_NET_ADMIN)) return -EPERM; err = security_tun_dev_create(); if (err < 0) @@ -980,20 +1014,10 @@ static int tun_set_iff(struct net *net, tun->flags = flags; tun->txflt.count = 0; - err = -ENOMEM; - sk = sk_alloc(net, AF_UNSPEC, GFP_KERNEL, &tun_proto); - if (!sk) + err = tun_sk_alloc_init(net, tun, &sk); + if (err) goto err_free_dev; - init_waitqueue_head(&tun->socket.wait); - sock_init_data(&tun->socket, sk); - sk->sk_write_space = tun_sock_write_space; - sk->sk_sndbuf = INT_MAX; - - container_of(sk, struct tun_sock, sk)->tun = tun; - - security_tun_dev_post_create(sk); - tun_net_init(dev); if (strchr(dev->name, '%')) { @@ -1006,10 +1030,10 @@ static int tun_set_iff(struct net *net, if (err < 0) goto err_free_sk; - if (!net_eq(dev_net(tun->dev), &init_net) || - device_create_file(&tun->dev->dev, &dev_attr_tun_flags) || - device_create_file(&tun->dev->dev, &dev_attr_owner) || - device_create_file(&tun->dev->dev, &dev_attr_group)) + if ((dev_net(tun->dev) == &init_net) && + (device_create_file(&tun->dev->dev, &dev_attr_tun_flags) || + device_create_file(&tun->dev->dev, &dev_attr_owner) || + device_create_file(&tun->dev->dev, &dev_attr_group))) printk(KERN_ERR "Failed to create tun sysfs files\n"); sk->sk_destruct = tun_sock_destruct; @@ -1317,6 +1341,7 @@ static int tun_chr_open(struct inode *in tfile->tun = NULL; tfile->net = get_net(current->nsproxy->net_ns); file->private_data = tfile; + tfile->file = file; return 0; } @@ -1458,6 +1483,226 @@ static const struct ethtool_ops tun_etht .set_rx_csum = tun_set_rx_csum }; +static void cpt_dump_tap_filter(struct tap_filter *flt, + struct cpt_ops *ops, struct cpt_context *ctx) +{ + struct cpt_tap_filter_image v; + loff_t saved_obj; + + ops->push_object(&saved_obj, ctx); + + v.cpt_next = CPT_NULL; + v.cpt_object = CPT_OBJ_NET_TAP_FILTER; + v.cpt_hdrlen = sizeof(v); + v.cpt_content = CPT_CONTENT_VOID; + + v.cpt_count = flt->count; + + BUILD_BUG_ON(sizeof(flt->mask) != sizeof(v.cpt_mask)); + memcpy(v.cpt_mask, flt->mask, sizeof(v.cpt_mask)); + + BUILD_BUG_ON(sizeof(flt->addr) != sizeof(v.cpt_addr)); + memcpy(v.cpt_addr, flt->addr, sizeof(v.cpt_addr)); + + ops->write(&v, sizeof(v), ctx); + + ops->pop_object(&saved_obj, ctx); +} + +static void tun_cpt(struct net_device *dev, + struct cpt_ops *ops, struct cpt_context * ctx) +{ + struct cpt_tuntap_image v; + struct tun_struct *tun; + + tun = netdev_priv(dev); + + v.cpt_next = CPT_NULL; + v.cpt_object = CPT_OBJ_NET_TUNTAP; + v.cpt_hdrlen = sizeof(v); + v.cpt_content = CPT_CONTENT_VOID; + + v.cpt_owner = tun->owner; + v.cpt_flags = tun->flags; + + if (tun->tfile->file) + v.cpt_bindfile = ops->lookup_object(CPT_OBJ_FILE, tun->tfile->file, ctx); + + v.cpt_if_flags = 0; + memset(v.cpt_dev_addr, 0, sizeof(v.cpt_dev_addr)); + memset(v.cpt_chr_filter, 0, sizeof(v.cpt_chr_filter)); + memset(v.cpt_net_filter, 0, sizeof(v.cpt_net_filter)); + + ops->write(&v, sizeof(v), ctx); + + cpt_dump_tap_filter(&tun->txflt, ops, ctx); +} + +static int rst_restore_tap_filter(loff_t start, struct cpt_tuntap_image *ti, + struct tap_filter *flt, struct rst_ops *ops, + struct cpt_context *ctx) +{ + int err; + struct cpt_tap_filter_image fi; + loff_t pos; + + /* disable filtering */ + flt->count = 0; + + pos = start + ti->cpt_hdrlen; + + /* no tap filter image? */ + if (pos >= start + ti->cpt_next) + goto convert; + + err = ops->get_object(CPT_OBJ_NET_TAP_FILTER, pos, + &fi, sizeof(fi), ctx); + if (err) + return err; + + BUILD_BUG_ON(sizeof(flt->mask) != sizeof(fi.cpt_mask)); + memcpy(flt->mask, fi.cpt_mask, sizeof(fi.cpt_mask)); + + BUILD_BUG_ON(sizeof(flt->addr) != sizeof(fi.cpt_addr)); + memcpy(flt->addr, fi.cpt_addr, sizeof(fi.cpt_addr)); + + flt->count = fi.cpt_count; + + return 0; + +convert: + /** From OLD filtering code: + * Decide whether to accept this packet. This code is designed to + * behave identically to an Ethernet interface. Accept the packet if + * - we are promiscuous. + * - the packet is addressed to us. + * - the packet is broadcast. + * - the packet is multicast and + * - we are multicast promiscous. + * - we belong to the multicast group. + */ + + /* accept all, this is default if filter is untouched */ + if (ti->cpt_if_flags & IFF_PROMISC) + return 0; + + /* accept packets addressed to character device's hardware address */ + BUILD_BUG_ON(sizeof(flt->addr[0]) != sizeof(ti->cpt_dev_addr)); + memcpy(flt->addr[0], ti->cpt_dev_addr, sizeof(ti->cpt_dev_addr)); + + /* accept broadcast */ + memset(flt->addr[1], ~0, sizeof(flt->addr[1])); + + /* accept hashed multicast: hash function the same as in old code */ + BUILD_BUG_ON(sizeof(flt->mask) != sizeof(ti->cpt_chr_filter)); + memcpy(flt->mask, ti->cpt_chr_filter, sizeof(ti->cpt_chr_filter)); + + /* accept all multicast */ + if (ti->cpt_if_flags & IFF_ALLMULTI) + memset(flt->mask, ~0, sizeof(flt->mask)); + + /* two exact filters: hw addr and broadcast */ + flt->count = 2; + + return 0; +} + +static int tun_rst(loff_t start, struct cpt_netdev_image *di, + struct rst_ops *ops, struct cpt_context *ctx) +{ + int err = -ENODEV; + struct cpt_tuntap_image ti; + struct net_device *dev; + struct file *bind_file = NULL; + struct tun_struct *tun; + struct tun_file *tfile; + struct sock *sk; + loff_t pos; + + pos = start + di->cpt_hdrlen; + err = ops->get_object(CPT_OBJ_NET_TUNTAP, pos, + &ti, sizeof(ti), ctx); + if (err) + return err; + + if (ti.cpt_bindfile) { + bind_file = ops->rst_file(ti.cpt_bindfile, -1, ctx); + if (IS_ERR(bind_file)) + return PTR_ERR(bind_file); + } + + tfile = kmalloc(sizeof(*tfile), GFP_KERNEL); + if (!tfile) + goto out; + + atomic_set(&tfile->count, 0); + tfile->tun = NULL; + tfile->net = get_net(current->nsproxy->net_ns); + tfile->file = bind_file; + + err = -ENOMEM; + dev = alloc_netdev(sizeof(struct tun_struct), di->cpt_name, tun_setup); + if (!dev) + goto out_tf; + + tun = netdev_priv(dev); + + tun->dev = dev; + tun->owner = ti.cpt_owner; + tun->flags = ti.cpt_flags; + tun_net_init(dev); + + err = tun_sk_alloc_init(current->nsproxy->net_ns, tun, &sk); + if (err) + goto out_netdev; + + err = rst_restore_tap_filter(pos, &ti, &tun->txflt, ops, ctx); + if (err < 0) + goto out_sk; + + err = register_netdevice(dev); + if (err < 0) + goto out_sk; + + pos += ti.cpt_next; + if (pos < start + di->cpt_next) { + struct cpt_hwaddr_image hw; + /* Restore hardware address */ + err = ops->get_object(CPT_OBJ_NET_HWADDR, pos, + &hw, sizeof(hw), ctx); + if (err) + goto out_unreg; + + memcpy(dev->dev_addr, hw.cpt_dev_addr, + sizeof(hw.cpt_dev_addr)); + } + + sk->sk_destruct = tun_sock_destruct; + bind_file->private_data = tfile; + __tun_attach(tun, tfile); + + fput(bind_file); + return 0; + +out_unreg: + unregister_netdevice(dev); +out_sk: + sock_put(sk); +out_netdev: + free_netdev(dev); +out_tf: + put_net(tfile->net); + kfree(tfile); +out: + fput(bind_file); + return err; +} + +static struct netdev_rst tun_netdev_rst = { + .cpt_object = CPT_OBJ_NET_TUNTAP, + .ndo_rst = tun_rst, +}; + static int __init tun_init(void) { @@ -1477,6 +1722,8 @@ static int __init tun_init(void) printk(KERN_ERR "tun: Can't register misc device %d\n", TUN_MINOR); goto err_misc; } + + register_netdev_rst(&tun_netdev_rst); return 0; err_misc: rtnl_link_unregister(&tun_link_ops); @@ -1486,6 +1733,7 @@ err_linkops: static void tun_cleanup(void) { + unregister_netdev_rst(&tun_netdev_rst); misc_deregister(&tun_miscdev); rtnl_link_unregister(&tun_link_ops); } diff -urNp linux-2.6.32.48/drivers/net/usb/hso.c linux-2.6.32.48-openvz/drivers/net/usb/hso.c --- linux-2.6.32.48/drivers/net/usb/hso.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/usb/hso.c 2011-11-17 20:08:53.000000000 -0500 @@ -71,7 +71,7 @@ #include #include #include - +#include #define DRIVER_VERSION "1.2" #define MOD_AUTHOR "Option Wireless" @@ -258,7 +258,7 @@ struct hso_serial { /* from usb_serial_port */ struct tty_struct *tty; - int open_count; + local_t open_count; spinlock_t serial_lock; int (*write_data) (struct hso_serial *serial); @@ -1180,7 +1180,7 @@ static void put_rxbuf_data_and_resubmit_ struct urb *urb; urb = serial->rx_urb[0]; - if (serial->open_count > 0) { + if (local_read(&serial->open_count) > 0) { count = put_rxbuf_data(urb, serial); if (count == -1) return; @@ -1216,7 +1216,7 @@ static void hso_std_serial_read_bulk_cal DUMP1(urb->transfer_buffer, urb->actual_length); /* Anyone listening? */ - if (serial->open_count == 0) + if (local_read(&serial->open_count) == 0) return; if (status == 0) { @@ -1311,8 +1311,7 @@ static int hso_serial_open(struct tty_st spin_unlock_irq(&serial->serial_lock); /* check for port already opened, if not set the termios */ - serial->open_count++; - if (serial->open_count == 1) { + if (local_inc_return(&serial->open_count) == 1) { tty->low_latency = 1; serial->rx_state = RX_IDLE; /* Force default termio settings */ @@ -1325,7 +1324,7 @@ static int hso_serial_open(struct tty_st result = hso_start_serial_device(serial->parent, GFP_KERNEL); if (result) { hso_stop_serial_device(serial->parent); - serial->open_count--; + local_dec(&serial->open_count); kref_put(&serial->parent->ref, hso_serial_ref_free); } } else { @@ -1362,10 +1361,10 @@ static void hso_serial_close(struct tty_ /* reset the rts and dtr */ /* do the actual close */ - serial->open_count--; + local_dec(&serial->open_count); - if (serial->open_count <= 0) { - serial->open_count = 0; + if (local_read(&serial->open_count) <= 0) { + local_set(&serial->open_count, 0); spin_lock_irq(&serial->serial_lock); if (serial->tty == tty) { serial->tty->driver_data = NULL; @@ -1447,7 +1446,7 @@ static void hso_serial_set_termios(struc /* the actual setup */ spin_lock_irqsave(&serial->serial_lock, flags); - if (serial->open_count) + if (local_read(&serial->open_count)) _hso_serial_set_termios(tty, old); else tty->termios = old; @@ -3097,7 +3096,7 @@ static int hso_resume(struct usb_interfa /* Start all serial ports */ for (i = 0; i < HSO_SERIAL_TTY_MINORS; i++) { if (serial_table[i] && (serial_table[i]->interface == iface)) { - if (dev2ser(serial_table[i])->open_count) { + if (local_read(&dev2ser(serial_table[i])->open_count)) { result = hso_start_serial_device(serial_table[i], GFP_NOIO); hso_kick_transmit(dev2ser(serial_table[i])); diff -urNp linux-2.6.32.48/drivers/net/venet_core.c linux-2.6.32.48-openvz/drivers/net/venet_core.c --- linux-2.6.32.48/drivers/net/venet_core.c 1969-12-31 19:00:00.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/venet_core.c 2011-11-17 20:08:53.000000000 -0500 @@ -0,0 +1,864 @@ +/* + * venet_core.c + * + * Copyright (C) 2005 SWsoft + * All rights reserved. + * + * Licensing governed by "linux/COPYING.SWsoft" file. + * + */ + +/* + * Common part for Virtuozzo virtual network devices + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include +#include /* For the statistics structure. */ +#include /* For ARPHRD_ETHER */ +#include +#include +#include +#include +#include + +struct list_head ip_entry_hash_table[VEIP_HASH_SZ]; +rwlock_t veip_hash_lock = RW_LOCK_UNLOCKED; +LIST_HEAD(veip_lh); + +#define ip_entry_hash_function(ip) (ntohl(ip) & (VEIP_HASH_SZ - 1)) + +void ip_entry_hash(struct ip_entry_struct *entry, struct veip_struct *veip) +{ + list_add(&entry->ip_hash, + ip_entry_hash_table + + ip_entry_hash_function(entry->addr.key[3])); + list_add(&entry->ve_list, &veip->ip_lh); +} + +void veip_put(struct veip_struct *veip) +{ + if (!list_empty(&veip->ip_lh)) + return; + if (!list_empty(&veip->src_lh)) + return; + if (!list_empty(&veip->dst_lh)) + return; + + list_del(&veip->list); + kfree(veip); +} + +struct ip_entry_struct *venet_entry_lookup(struct ve_addr_struct *addr) +{ + struct ip_entry_struct *entry; + + list_for_each_entry (entry, ip_entry_hash_table + + ip_entry_hash_function(addr->key[3]), ip_hash) + if (memcmp(&entry->addr, addr, sizeof(*addr)) == 0) + return entry; + return NULL; +} + +struct ext_entry_struct *venet_ext_lookup(struct ve_struct *ve, + struct ve_addr_struct *addr) +{ + struct ext_entry_struct *entry; + + if (ve->veip == NULL) + return NULL; + + list_for_each_entry (entry, &ve->veip->ext_lh, list) + if (memcmp(&entry->addr, addr, sizeof(*addr)) == 0) + return entry; + return NULL; +} + +int venet_ext_add(struct ve_struct *ve, struct ve_addr_struct *addr) +{ + struct ext_entry_struct *entry, *found; + int err; + + if (ve->veip == NULL) + return -ENONET; + + entry = kzalloc(sizeof(struct ext_entry_struct), GFP_KERNEL); + if (entry == NULL) + return -ENOMEM; + + write_lock_irq(&veip_hash_lock); + err = -EADDRINUSE; + found = venet_ext_lookup(ve, addr); + if (found != NULL) + goto out_unlock; + + entry->addr = *addr; + list_add(&entry->list, &ve->veip->ext_lh); + err = 0; + entry = NULL; +out_unlock: + write_unlock_irq(&veip_hash_lock); + if (entry != NULL) + kfree(entry); + return err; +} + +int venet_ext_del(struct ve_struct *ve, struct ve_addr_struct *addr) +{ + struct ext_entry_struct *found; + int err; + + if (ve->veip == NULL) + return -ENONET; + + err = -EADDRNOTAVAIL; + write_lock_irq(&veip_hash_lock); + found = venet_ext_lookup(ve, addr); + if (found == NULL) + goto out; + + list_del(&found->list); + kfree(found); + err = 0; +out: + write_unlock_irq(&veip_hash_lock); + return err; +} + +void venet_ext_clean(struct ve_struct *ve) +{ + struct ext_entry_struct *entry, *tmp; + + if (ve->veip == NULL) + return; + + write_lock_irq(&veip_hash_lock); + list_for_each_entry_safe (entry, tmp, &ve->veip->ext_lh, list) { + list_del(&entry->list); + kfree(entry); + } + write_unlock_irq(&veip_hash_lock); +} + +struct veip_struct *veip_find(envid_t veid) +{ + struct veip_struct *ptr; + + list_for_each_entry(ptr, &veip_lh, list) { + if (ptr->veid != veid) + continue; + return ptr; + } + return NULL; +} + +struct veip_struct *veip_findcreate(envid_t veid) +{ + struct veip_struct *ptr; + + ptr = veip_find(veid); + if (ptr != NULL) + return ptr; + + ptr = kmalloc(sizeof(struct veip_struct), GFP_ATOMIC); + if (ptr == NULL) + return NULL; + memset(ptr, 0, sizeof(struct veip_struct)); + INIT_LIST_HEAD(&ptr->ip_lh); + INIT_LIST_HEAD(&ptr->src_lh); + INIT_LIST_HEAD(&ptr->dst_lh); + INIT_LIST_HEAD(&ptr->ext_lh); + ptr->veid = veid; + list_add(&ptr->list, &veip_lh); + return ptr; +} + +static int convert_sockaddr(struct sockaddr *addr, int addrlen, + struct ve_addr_struct *veaddr) +{ + int err; + + switch (addr->sa_family) { + case AF_INET: { + struct sockaddr_in *sin; + + err = -EINVAL; + if (addrlen != sizeof(struct sockaddr_in)) + break; + + err = 0; + sin = (struct sockaddr_in *)addr; + veaddr->family = AF_INET; + veaddr->key[0] = 0; + veaddr->key[1] = 0; + veaddr->key[2] = 0; + veaddr->key[3] = sin->sin_addr.s_addr; + break; + } + case AF_INET6: { + struct sockaddr_in6 *sin; + + err = -EINVAL; + if (addrlen != sizeof(struct sockaddr_in6)) + break; + + err = 0; + sin = (struct sockaddr_in6 *)addr; + veaddr->family = AF_INET6; + memcpy(veaddr->key, &sin->sin6_addr, sizeof(veaddr->key)); + break; + } + default: + err = -EAFNOSUPPORT; + } + return err; +} + +int sockaddr_to_veaddr(struct sockaddr __user *uaddr, int addrlen, + struct ve_addr_struct *veaddr) +{ + int err; + char addr[MAX_SOCK_ADDR]; + + err = move_addr_to_kernel(uaddr, addrlen, (struct sockaddr *)&addr); + if (err < 0) + goto out; + + err = convert_sockaddr((struct sockaddr *)&addr, addrlen, veaddr); +out: + return err; +} + +void veaddr_print(char *str, int len, struct ve_addr_struct *a) +{ + if (a->family == AF_INET) + snprintf(str, len, "%u.%u.%u.%u", NIPQUAD(a->key[3])); + else + snprintf(str, len, "%x:%x:%x:%x:%x:%x:%x:%x", + ntohl(a->key[0])>>16, ntohl(a->key[0])&0xFFFF, + ntohl(a->key[1])>>16, ntohl(a->key[1])&0xFFFF, + ntohl(a->key[2])>>16, ntohl(a->key[2])&0xFFFF, + ntohl(a->key[3])>>16, ntohl(a->key[3])&0xFFFF + ); +} + +/* + * Device functions + */ + +static int venet_open(struct net_device *dev) +{ + if (!ve_is_super(get_exec_env()) && !try_module_get(THIS_MODULE)) + return -EBUSY; + return 0; +} + +static int venet_close(struct net_device *master) +{ + if (!ve_is_super(get_exec_env())) + module_put(THIS_MODULE); + return 0; +} + +static void venet_destructor(struct net_device *dev) +{ + struct venet_stats *stats = (struct venet_stats *)dev->ml_priv; + if (stats == NULL) + return; + free_percpu(stats->real_stats); + kfree(stats); + dev->ml_priv = NULL; +} + +/* + * The higher levels take care of making this non-reentrant (it's + * called with bh's disabled). + */ +static int venet_xmit(struct sk_buff *skb, struct net_device *dev) +{ + struct net_device_stats *stats; + struct net_device *rcv = NULL; + int length; + + stats = venet_stats(dev, smp_processor_id()); + if (unlikely(get_exec_env()->disable_net)) + goto outf; + + if (skb->protocol == __constant_htons(ETH_P_IP)) { + struct iphdr *iph; + iph = ip_hdr(skb); + if (ipv4_is_multicast(iph->daddr)) + goto outf; + } else if (skb->protocol == __constant_htons(ETH_P_IPV6)) { + struct ipv6hdr *ip6h; + ip6h = ipv6_hdr(skb); + if (ipv6_addr_is_multicast(&ip6h->daddr)) + goto outf; + skb_orphan(skb); + } else { + goto outf; + } + + if (venet_change_skb_owner(skb) < 0) + goto outf; + + if (unlikely(skb->owner_env->disable_net)) + goto outf; + + rcv = skb->owner_env->_venet_dev; + if (!rcv) + /* VE going down */ + goto outf; + + dev_hold(rcv); + + if (!(rcv->flags & IFF_UP)) { + /* Target VE does not want to receive packets */ + dev_put(rcv); + goto outf; + } + + skb->pkt_type = PACKET_HOST; + skb->dev = rcv; + + skb_reset_mac_header(skb); + memset(skb->data - dev->hard_header_len, 0, dev->hard_header_len); + + nf_reset(skb); + length = skb->len; + + netif_rx(skb); + + stats->tx_bytes += length; + stats->tx_packets++; + if (rcv) { + struct net_device_stats *rcv_stats; + + rcv_stats = venet_stats(rcv, smp_processor_id()); + rcv_stats->rx_bytes += length; + rcv_stats->rx_packets++; + dev_put(rcv); + } + + return 0; + +outf: + kfree_skb(skb); + ++stats->tx_dropped; + return 0; +} + +static struct net_device_stats *get_stats(struct net_device *dev) +{ + int i; + struct venet_stats *stats; + + stats = (struct venet_stats *)dev->ml_priv; + memset(&stats->stats, 0, sizeof(struct net_device_stats)); + for (i=0; i < NR_CPUS; i++) { + struct net_device_stats *dev_stats; + + if (!cpu_possible(i)) + continue; + dev_stats = venet_stats(dev, i); + stats->stats.rx_bytes += dev_stats->rx_bytes; + stats->stats.tx_bytes += dev_stats->tx_bytes; + stats->stats.rx_packets += dev_stats->rx_packets; + stats->stats.tx_packets += dev_stats->tx_packets; + } + + return &stats->stats; +} + +/* Initialize the rest of the LOOPBACK device. */ +int venet_init_dev(struct net_device *dev) +{ + struct venet_stats *stats; + + stats = kzalloc(sizeof(struct venet_stats), GFP_KERNEL); + if (stats == NULL) + goto fail; + stats->real_stats = alloc_percpu(struct net_device_stats); + if (stats->real_stats == NULL) + goto fail_free; + dev->ml_priv = stats; + + /* + * Fill in the generic fields of the device structure. + */ + dev->type = ARPHRD_VOID; + dev->hard_header_len = ETH_HLEN; + dev->mtu = 1500; /* eth_mtu */ + dev->tx_queue_len = 0; + + memset(dev->broadcast, 0xFF, ETH_ALEN); + + /* New-style flags. */ + dev->flags = IFF_BROADCAST|IFF_NOARP|IFF_POINTOPOINT; + return 0; + +fail_free: + kfree(stats); +fail: + return -ENOMEM; +} + +static const struct net_device_ops venet_netdev_ops; + +static int +venet_set_op(struct net_device *dev, u32 data, + int (*fop)(struct net_device *, u32)) +{ + + struct ve_struct *ve; + int ret = 0; + + read_lock(&ve_list_lock); + for_each_ve(ve) { + struct ve_struct *ve_old; + + ve_old = set_exec_env(ve); + read_lock(&dev_base_lock); + for_each_netdev(ve->ve_netns, dev) { + if (dev->netdev_ops == &venet_netdev_ops) + ret = fop(dev, data); + } + read_unlock(&dev_base_lock); + set_exec_env(ve_old); + + if (ret < 0) + break; + } + read_unlock(&ve_list_lock); + return ret; +} + +static unsigned long common_features; + +static int venet_op_set_sg(struct net_device *dev, u32 data) +{ + if (!ve_is_super(get_exec_env())) + return -EPERM; + + if (data) + common_features |= NETIF_F_SG; + else + common_features &= ~NETIF_F_SG; + + return venet_set_op(dev, data, ethtool_op_set_sg); +} + +static int venet_op_set_tx_csum(struct net_device *dev, u32 data) +{ + if (!ve_is_super(get_exec_env())) + return -EPERM; + + if (data) + common_features |= NETIF_F_IP_CSUM; + else + common_features &= ~NETIF_F_IP_CSUM; + + return venet_set_op(dev, data, ethtool_op_set_tx_csum); +} + +static int +venet_op_set_tso(struct net_device *dev, u32 data) +{ + if (!ve_is_super(get_exec_env())) + return -EPERM; + + if (data) + common_features |= NETIF_F_TSO; + else + common_features &= ~NETIF_F_TSO; + + return venet_set_op(dev, data, ethtool_op_set_tso); +} + +#define venet_op_set_rx_csum venet_op_set_tx_csum + +static struct ethtool_ops venet_ethtool_ops = { + .get_sg = ethtool_op_get_sg, + .set_sg = venet_op_set_sg, + .get_tx_csum = ethtool_op_get_tx_csum, + .set_tx_csum = venet_op_set_tx_csum, + .get_rx_csum = ethtool_op_get_tx_csum, + .set_rx_csum = venet_op_set_rx_csum, + .get_tso = ethtool_op_get_tso, + .set_tso = venet_op_set_tso, +}; + +static void venet_cpt(struct net_device *dev, + struct cpt_ops *ops, struct cpt_context *ctx) +{ +} + +static const struct net_device_ops venet_netdev_ops = { + .ndo_start_xmit = venet_xmit, + .ndo_get_stats = get_stats, + .ndo_open = venet_open, + .ndo_stop = venet_close, + .ndo_init = venet_init_dev, + .ndo_cpt = venet_cpt, +}; + +static void venet_setup(struct net_device *dev) +{ + /* + * No other features, as they are: + * - checksumming is required, and nobody else will done our job + */ + dev->features |= NETIF_F_VENET | NETIF_F_VIRTUAL | NETIF_F_LLTX | + NETIF_F_HIGHDMA | NETIF_F_VLAN_CHALLENGED; + + dev->netdev_ops = &venet_netdev_ops; + dev->destructor = venet_destructor; + + dev->features |= common_features; + + SET_ETHTOOL_OPS(dev, &venet_ethtool_ops); +} + +#ifdef CONFIG_PROC_FS +static void veaddr_seq_print(struct seq_file *m, struct ve_struct *ve) +{ + struct ip_entry_struct *entry; + + read_lock(&veip_hash_lock); + if (ve->veip == NULL) + goto unlock; + list_for_each_entry (entry, &ve->veip->ip_lh, ve_list) { + char addr[40]; + + if (entry->active_env == NULL) + continue; + + veaddr_print(addr, sizeof(addr), &entry->addr); + if (entry->addr.family == AF_INET) + seq_printf(m, " %15s", addr); + else + seq_printf(m, " %39s", addr); + } +unlock: + read_unlock(&veip_hash_lock); +} + +static void *veip_seq_start(struct seq_file *m, loff_t *pos) +{ + loff_t l; + struct list_head *p; + int i; + + l = *pos; + write_lock_irq(&veip_hash_lock); + if (l == 0) + return ip_entry_hash_table; + for (i = 0; i < VEIP_HASH_SZ; i++) { + list_for_each(p, ip_entry_hash_table + i) { + if (--l == 0) + return p; + } + } + return NULL; +} + +static void *veip_seq_next(struct seq_file *m, void *v, loff_t *pos) +{ + struct list_head *p; + + p = (struct list_head *)v; + while (1) { + p = p->next; + if (p < ip_entry_hash_table || + p >= ip_entry_hash_table + VEIP_HASH_SZ) { + (*pos)++; + return p; + } + if (++p >= ip_entry_hash_table + VEIP_HASH_SZ) + return NULL; + } + return NULL; +} + +static void veip_seq_stop(struct seq_file *m, void *v) +{ + write_unlock_irq(&veip_hash_lock); +} + +static struct seq_operations veip_seq_op = { + .start = veip_seq_start, + .next = veip_seq_next, + .stop = veip_seq_stop, + .show = veip_seq_show, +}; + +static int veip_open(struct inode *inode, struct file *file) +{ + return seq_open(file, &veip_seq_op); +} + +static struct file_operations proc_veip_operations = { + .open = veip_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, +}; +#endif + +static int real_ve_ip_map(envid_t veid, int op, struct sockaddr __user *uaddr, + int addrlen) +{ + int err; + struct ve_struct *ve; + struct ve_addr_struct addr; + + err = -EPERM; + if (!capable_setveid()) + goto out; + + err = sockaddr_to_veaddr(uaddr, addrlen, &addr); + if (err < 0) + goto out; + + switch (op) + { + case VE_IP_ADD: + ve = get_ve_by_id(veid); + err = -ESRCH; + if (!ve) + goto out; + + down_read(&ve->op_sem); + if (ve->is_running) + err = veip_entry_add(ve, &addr); + up_read(&ve->op_sem); + put_ve(ve); + break; + + case VE_IP_DEL: + err = veip_entry_del(veid, &addr); + break; + case VE_IP_EXT_ADD: + ve = get_ve_by_id(veid); + err = -ESRCH; + if (!ve) + goto out; + + down_read(&ve->op_sem); + err = venet_ext_add(ve, &addr); + up_read(&ve->op_sem); + put_ve(ve); + break; + case VE_IP_EXT_DEL: + ve = get_ve_by_id(veid); + err = -ESRCH; + if (!ve) + goto out; + + down_read(&ve->op_sem); + err = venet_ext_del(ve, &addr); + up_read(&ve->op_sem); + put_ve(ve); + break; + default: + err = -EINVAL; + } + +out: + return err; +} + +int venet_ioctl(struct file *file, unsigned int cmd, unsigned long arg) +{ + int err; + + err = -ENOTTY; + switch(cmd) { + case VENETCTL_VE_IP_MAP: { + struct vzctl_ve_ip_map s; + err = -EFAULT; + if (copy_from_user(&s, (void __user *)arg, sizeof(s))) + break; + err = real_ve_ip_map(s.veid, s.op, s.addr, s.addrlen); + break; + } + } + return err; +} + +#ifdef CONFIG_COMPAT +int compat_venet_ioctl(struct file *file, unsigned int cmd, unsigned long arg) +{ + int err; + + switch(cmd) { + case VENETCTL_COMPAT_VE_IP_MAP: { + struct compat_vzctl_ve_ip_map cs; + + err = -EFAULT; + if (copy_from_user(&cs, (void *)arg, sizeof(cs))) + break; + + err = real_ve_ip_map(cs.veid, cs.op, compat_ptr(cs.addr), + cs.addrlen); + break; + } + default: + err = venet_ioctl(file, cmd, arg); + break; + } + return err; +} +#endif + +static struct vzioctlinfo venetcalls = { + .type = VENETCTLTYPE, + .ioctl = venet_ioctl, +#ifdef CONFIG_COMPAT + .compat_ioctl = compat_venet_ioctl, +#endif + .owner = THIS_MODULE, +}; + +int venet_dev_start(struct ve_struct *ve) +{ + struct net_device *dev_venet; + int err; + + dev_venet = alloc_netdev(0, "venet%d", venet_setup); + if (!dev_venet) + return -ENOMEM; + dev_net_set(dev_venet, ve->ve_netns); + err = dev_alloc_name(dev_venet, dev_venet->name); + if (err<0) + goto err; + if ((err = register_netdev(dev_venet)) != 0) + goto err; + ve->_venet_dev = dev_venet; + return 0; +err: + free_netdev(dev_venet); + printk(KERN_ERR "VENET initialization error err=%d\n", err); + return err; +} + +static int venet_start(void *data) +{ + struct ve_struct *env; + int err; + + env = (struct ve_struct *)data; + if (env->veip) + return -EEXIST; + + err = veip_start(env); + if (err != 0) + return err; + + err = venet_dev_start(env); + if (err) + goto err_free; + return 0; + +err_free: + veip_stop(env); + return err; +} + +static void venet_stop(void *data) +{ + struct ve_struct *env; + struct net_device *dev; + + env = (struct ve_struct *)data; + venet_ext_clean(env); + veip_stop(env); + + dev = env->_venet_dev; + if (dev == NULL) + return; + + unregister_netdev(dev); + env->_venet_dev = NULL; + free_netdev(dev); +} + +static struct ve_hook venet_ve_hook = { + .init = venet_start, + .fini = venet_stop, + .owner = THIS_MODULE, + .priority = HOOK_PRIO_NET, +}; + +__init int venet_init(void) +{ +#ifdef CONFIG_PROC_FS + struct proc_dir_entry *de; +#endif + int i, err; + + if (get_ve0()->_venet_dev != NULL) + return -EEXIST; + + for (i = 0; i < VEIP_HASH_SZ; i++) + INIT_LIST_HEAD(ip_entry_hash_table + i); + + err = venet_start(get_ve0()); + if (err) + return err; + +#ifdef CONFIG_PROC_FS + de = proc_create("veip", S_IFREG | S_IRUSR, proc_vz_dir, + &proc_veip_operations); + if (de == NULL) + printk(KERN_WARNING "venet: can't make veip proc entry\n"); +#endif + + ve_hook_register(VE_SS_CHAIN, &venet_ve_hook); + vzioctl_register(&venetcalls); + vzmon_register_veaddr_print_cb(veaddr_seq_print); + return 0; +} + +__exit void venet_exit(void) +{ + vzmon_unregister_veaddr_print_cb(veaddr_seq_print); + vzioctl_unregister(&venetcalls); + ve_hook_unregister(&venet_ve_hook); + +#ifdef CONFIG_PROC_FS + remove_proc_entry("veip", proc_vz_dir); +#endif + venet_stop(get_ve0()); + veip_cleanup(); +} + +module_init(venet_init); +module_exit(venet_exit); diff -urNp linux-2.6.32.48/drivers/net/veth.c linux-2.6.32.48-openvz/drivers/net/veth.c --- linux-2.6.32.48/drivers/net/veth.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/veth.c 2011-11-17 20:08:53.000000000 -0500 @@ -180,6 +180,7 @@ static netdev_tx_t veth_xmit(struct sk_b skb->mark = 0; secpath_reset(skb); nf_reset(skb); + skb_init_brmark(skb); length = skb->len; diff -urNp linux-2.6.32.48/drivers/net/vxge/vxge-config.h linux-2.6.32.48-openvz/drivers/net/vxge/vxge-config.h --- linux-2.6.32.48/drivers/net/vxge/vxge-config.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/vxge/vxge-config.h 2011-11-17 20:09:04.000000000 -0500 @@ -474,7 +474,7 @@ struct vxge_hw_uld_cbs { void (*link_down)(struct __vxge_hw_device *devh); void (*crit_err)(struct __vxge_hw_device *devh, enum vxge_hw_event type, u64 ext_data); -}; +} __no_const; /* * struct __vxge_hw_blockpool_entry - Block private data structure diff -urNp linux-2.6.32.48/drivers/net/vxge/vxge-main.c linux-2.6.32.48-openvz/drivers/net/vxge/vxge-main.c --- linux-2.6.32.48/drivers/net/vxge/vxge-main.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/vxge/vxge-main.c 2011-11-17 20:08:53.000000000 -0500 @@ -93,6 +93,8 @@ static inline void VXGE_COMPLETE_VPATH_T struct sk_buff *completed[NR_SKB_COMPLETED]; int more; + pax_track_stack(); + do { more = 0; skb_ptr = completed; @@ -1779,6 +1781,8 @@ static enum vxge_hw_status vxge_rth_conf u8 mtable[256] = {0}; /* CPU to vpath mapping */ int index; + pax_track_stack(); + /* * Filling * - itable with bucket numbers diff -urNp linux-2.6.32.48/drivers/net/vxge/vxge-traffic.h linux-2.6.32.48-openvz/drivers/net/vxge/vxge-traffic.h --- linux-2.6.32.48/drivers/net/vxge/vxge-traffic.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/vxge/vxge-traffic.h 2011-11-17 20:09:04.000000000 -0500 @@ -2123,7 +2123,7 @@ struct vxge_hw_mempool_cbs { struct vxge_hw_mempool_dma *dma_object, u32 index, u32 is_last); -}; +} __no_const; void __vxge_hw_mempool_destroy( diff -urNp linux-2.6.32.48/drivers/net/vzethdev.c linux-2.6.32.48-openvz/drivers/net/vzethdev.c --- linux-2.6.32.48/drivers/net/vzethdev.c 1969-12-31 19:00:00.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/vzethdev.c 2011-11-17 20:08:53.000000000 -0500 @@ -0,0 +1,749 @@ +/* + * veth.c + * + * Copyright (C) 2006 SWsoft + * All rights reserved. + * + * Licensing governed by "linux/COPYING.SWsoft" file. + * + */ + +/* + * Virtual ethernet device used to change VE ownership on packets + */ + +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include +#include + +#include +#include +#include +#include + +#include +#include +#include +#include +#include +#include +#include /* For the statistics structure. */ +#include /* For ARPHRD_ETHER */ +#include +#include +#include +#include +#include +#include + +#include +#include +#include + +#include +#include + +static LIST_HEAD(veth_hwaddr_list); +static DEFINE_RWLOCK(ve_hwaddr_lock); +static DECLARE_MUTEX(hwaddr_sem); + +struct net_device * veth_dev_start(char *dev_addr, char *name); + +struct veth_struct *hwaddr_entry_lookup(char *name) +{ + struct veth_struct *entry; + + list_for_each_entry(entry, &veth_hwaddr_list, hwaddr_list) { + BUG_ON(entry->pair == NULL); + if (strncmp(name, entry->pair->name, IFNAMSIZ) == 0) + return entry; + } + return NULL; +} + +int veth_entry_add(struct ve_struct *ve, char *dev_addr, char *name, + char *dev_addr_ve, char *name_ve) +{ + struct net_device *dev_ve; + struct net_device *dev_ve0; + struct ve_struct *old_env; + char dev_name[IFNAMSIZ]; + int err; + + down(&hwaddr_sem); + + if (name[0] == '\0') + snprintf(dev_name, sizeof(dev_name), "vz%d.%%d", ve->veid); + else { + memcpy(dev_name, name, IFNAMSIZ - 1); + dev_name[IFNAMSIZ - 1] = '\0'; + } + dev_ve0 = veth_dev_start(dev_addr, dev_name); + if (IS_ERR(dev_ve0)) { + err = PTR_ERR(dev_ve0); + goto err; + } + + old_env = set_exec_env(ve); + if (name_ve[0] == '\0') + sprintf(dev_name, "eth%%d"); + else { + memcpy(dev_name, name_ve, IFNAMSIZ - 1); + dev_name[IFNAMSIZ - 1] = '\0'; + } + dev_ve = veth_dev_start(dev_addr_ve, dev_name); + if (IS_ERR(dev_ve)) { + err = PTR_ERR(dev_ve); + goto err_ve; + } + set_exec_env(old_env); + veth_from_netdev(dev_ve)->pair = dev_ve0; + veth_from_netdev(dev_ve)->me = dev_ve; + veth_from_netdev(dev_ve0)->pair = dev_ve; + veth_from_netdev(dev_ve0)->me = dev_ve0; + + write_lock(&ve_hwaddr_lock); + list_add(&(veth_from_netdev(dev_ve)->hwaddr_list), &veth_hwaddr_list); + write_unlock(&ve_hwaddr_lock); + + up(&hwaddr_sem); + return 0; + +err_ve: + set_exec_env(old_env); + unregister_netdev(dev_ve0); +err: + up(&hwaddr_sem); + return err; +} + +void veth_pair_del(struct ve_struct *env, struct veth_struct *entry) +{ + struct net_device *dev; + struct ve_struct *old_env; + + write_lock(&ve_hwaddr_lock); + list_del(&entry->hwaddr_list); + write_unlock(&ve_hwaddr_lock); + + dev = entry->pair; + BUG_ON(entry->pair == NULL); + + veth_from_netdev(dev)->pair = NULL; + entry->pair = NULL; + rtnl_lock(); + old_env = set_exec_env(dev->owner_env); + dev_close(dev); + + /* + * Now device from VE0 does not send or receive anything, + * i.e. dev->hard_start_xmit won't be called. + */ + set_exec_env(env); + unregister_netdevice(veth_to_netdev(entry)); + set_exec_env(dev->owner_env); + unregister_netdevice(dev); + set_exec_env(old_env); + rtnl_unlock(); +} + +int veth_entry_del(struct ve_struct *ve, char *name) +{ + struct veth_struct *found; + int err; + + err = -ENODEV; + down(&hwaddr_sem); + found = hwaddr_entry_lookup(name); + if (found == NULL) + goto out; + if (veth_to_netdev(found)->owner_env != ve) + goto out; + + err = 0; + veth_pair_del(ve, found); + +out: + up(&hwaddr_sem); + return err; +} + +int veth_allow_change_mac(envid_t veid, char *name, int allow) +{ + struct ve_struct *ve; + struct veth_struct *found; + int err; + + err = -ESRCH; + ve = get_ve_by_id(veid); + if (!ve) + return err; + + down_read(&ve->op_sem); + if (!ve->is_running) + goto out_ve; + err = -ENODEV; + down(&hwaddr_sem); + found = hwaddr_entry_lookup(name); + if (found == NULL) + goto out_sem; + if (veth_to_netdev(found)->owner_env != ve) + goto out_sem; + + err = 0; + found->allow_mac_change = allow; + +out_sem: + up(&hwaddr_sem); +out_ve: + up_read(&ve->op_sem); + put_ve(ve); + return err; +} + +/* + * Device functions + */ + +static int veth_open(struct net_device *dev) +{ + return 0; +} + +static int veth_close(struct net_device *master) +{ + return 0; +} + +static void veth_destructor(struct net_device *dev) +{ + free_percpu(veth_from_netdev(dev)->real_stats); + free_netdev(dev); +} + +static struct net_device_stats *get_stats(struct net_device *dev) +{ + int i; + struct net_device_stats *stats; + + stats = &veth_from_netdev(dev)->stats; + memset(stats, 0, sizeof(struct net_device_stats)); + for (i = 0; i < NR_CPUS; i++) { + struct net_device_stats *dev_stats; + + if (!cpu_possible(i)) + continue; + dev_stats = veth_stats(dev, i); + stats->rx_bytes += dev_stats->rx_bytes; + stats->tx_bytes += dev_stats->tx_bytes; + stats->rx_packets += dev_stats->rx_packets; + stats->tx_packets += dev_stats->tx_packets; + stats->tx_dropped += dev_stats->tx_dropped; + } + + return stats; +} + +/* + * The higher levels take care of making this non-reentrant (it's + * called with bh's disabled). + */ +static int veth_xmit(struct sk_buff *skb, struct net_device *dev) +{ + struct net_device_stats *stats; + struct net_device *rcv = NULL; + struct veth_struct *entry; + int length; + + stats = veth_stats(dev, smp_processor_id()); + if (unlikely(get_exec_env()->disable_net)) + goto outf; + + entry = veth_from_netdev(dev); + rcv = entry->pair; + if (!rcv) + /* VE going down */ + goto outf; + + if (!(rcv->flags & IFF_UP)) { + /* Target VE does not want to receive packets */ + goto outf; + } + + if (unlikely(rcv->owner_env->disable_net)) + goto outf; + /* Filtering */ + if (ve_is_super(dev->owner_env) && + !veth_from_netdev(rcv)->allow_mac_change) { + /* from VE0 to VEX */ + if (ve_is_super(rcv->owner_env)) + goto out; + if (is_multicast_ether_addr( + ((struct ethhdr *)skb->data)->h_dest)) + goto out; + if (!rcv->br_port && + compare_ether_addr(((struct ethhdr *)skb->data)->h_dest, rcv->dev_addr)) + goto outf; + } else if (!ve_is_super(dev->owner_env) && + !entry->allow_mac_change) { + /* from VEX to VE0 */ + if (!skb->dev->br_port && + compare_ether_addr(((struct ethhdr *)skb->data)->h_source, dev->dev_addr)) + goto outf; + } + +out: + skb->owner_env = rcv->owner_env; + + skb->pkt_type = PACKET_HOST; + skb->protocol = eth_type_trans(skb, rcv); + + if (skb->protocol != __constant_htons(ETH_P_IP)) + skb_orphan(skb); + + nf_reset(skb); + length = skb->len; + skb_init_brmark(skb); + + netif_rx(skb); + + stats->tx_bytes += length; + stats->tx_packets++; + if (rcv) { + struct net_device_stats *rcv_stats; + rcv_stats = veth_stats(rcv, smp_processor_id()); + rcv_stats->rx_bytes += length; + rcv_stats->rx_packets++; + } + + return 0; + +outf: + kfree_skb(skb); + stats->tx_dropped++; + return 0; +} + +static int veth_set_mac(struct net_device *dev, void *p) +{ + struct sockaddr *addr = p; + + if (!ve_is_super(dev->owner_env) && + !veth_from_netdev(dev)->allow_mac_change) + return -EPERM; + if (netif_running(dev)) + return -EBUSY; + if (!is_valid_ether_addr(addr->sa_data)) + return -EADDRNOTAVAIL; + + memcpy(dev->dev_addr, addr->sa_data, dev->addr_len); + + return 0; +} + +int veth_init_dev(struct net_device *dev) +{ + veth_from_netdev(dev)->real_stats = + alloc_percpu(struct net_device_stats); + if (veth_from_netdev(dev)->real_stats == NULL) + return -ENOMEM; + + return 0; +} + +static int +veth_set_op(struct net_device *dev, u32 data, + int (*fop)(struct net_device *, u32)) +{ + struct net_device *pair; + int ret = 0; + + ret = fop(dev, data); + if (ret < 0) + goto out; + + pair = veth_from_netdev(dev)->pair; + if (pair) + ret = fop(pair, data); +out: + return ret; +} + +static int veth_op_set_sg(struct net_device *dev, u32 data) +{ + return veth_set_op(dev, data, ethtool_op_set_sg); +} + +static int veth_op_set_tx_csum(struct net_device *dev, u32 data) +{ + return veth_set_op(dev, data, ethtool_op_set_tx_csum); +} + +static int +veth_op_set_tso(struct net_device *dev, u32 data) +{ + return veth_set_op(dev, data, ethtool_op_set_tso); +} + +#define veth_op_set_rx_csum veth_op_set_tx_csum + +static struct ethtool_ops veth_ethtool_ops = { + .get_sg = ethtool_op_get_sg, + .set_sg = veth_op_set_sg, + .get_tx_csum = ethtool_op_get_tx_csum, + .set_tx_csum = veth_op_set_tx_csum, + .get_rx_csum = ethtool_op_get_tx_csum, + .set_rx_csum = veth_op_set_rx_csum, + .get_tso = ethtool_op_get_tso, + .set_tso = veth_op_set_tso, +}; + +static void veth_cpt(struct net_device *dev, + struct cpt_ops *ops, struct cpt_context *ctx) +{ + struct cpt_veth_image v; + struct veth_struct *veth; + + veth = veth_from_netdev(dev); + + v.cpt_next = CPT_NULL; + v.cpt_object = CPT_OBJ_NET_VETH; + v.cpt_hdrlen = sizeof(v); + v.cpt_content = CPT_CONTENT_VOID; + + v.cpt_allow_mac_change = veth->allow_mac_change; + + ops->write(&v, sizeof(v), ctx); +} + +static int veth_rst(loff_t pos, struct cpt_netdev_image *di, + struct rst_ops *ops, + struct cpt_context *ctx) + +{ + int err; + struct cpt_veth_image vi; + struct veth_struct *veth; + struct net_device *dev; + + pos = pos + di->cpt_hdrlen; + err = ops->get_object(CPT_OBJ_NET_VETH, pos, + &vi, sizeof(vi), ctx); + if (err) + return err; + + dev = __dev_get_by_name(get_exec_env()->ve_ns->net_ns, di->cpt_name); + if (dev == NULL) + return -ENODEV; + + veth = veth_from_netdev(dev); + veth->allow_mac_change = vi.cpt_allow_mac_change; + + return 0; +} + +static struct netdev_rst veth_netdev_rst = { + .cpt_object = CPT_OBJ_NET_VETH, + .ndo_rst = veth_rst, +}; + +static const struct net_device_ops veth_ops = { + .ndo_init = veth_init_dev, + .ndo_start_xmit = veth_xmit, + .ndo_get_stats = get_stats, + .ndo_open = veth_open, + .ndo_stop = veth_close, + .ndo_set_mac_address = veth_set_mac, + .ndo_cpt = veth_cpt, +}; + +static void veth_setup(struct net_device *dev) +{ + ether_setup(dev); + + dev->netdev_ops = &veth_ops; + dev->destructor = veth_destructor; + dev->tx_queue_len = 0; + + /* + * No other features, as they are: + * - checksumming is required, and nobody else will done our job + */ + dev->features |= NETIF_F_VENET | NETIF_F_VIRTUAL | NETIF_F_LLTX | + NETIF_F_HIGHDMA; + + SET_ETHTOOL_OPS(dev, &veth_ethtool_ops); +} + +#ifdef CONFIG_PROC_FS +#define ADDR_FMT "%02x:%02x:%02x:%02x:%02x:%02x" +#define ADDR_ARG(x) (x)[0],(x)[1],(x)[2],(x)[3],(x)[4],(x)[5] +static int vehwaddr_seq_show(struct seq_file *m, void *v) +{ + struct list_head *p; + struct veth_struct *entry; + + p = (struct list_head *)v; + if (p == &veth_hwaddr_list) { + seq_puts(m, "Version: 1.0\n"); + return 0; + } + entry = list_entry(p, struct veth_struct, hwaddr_list); + seq_printf(m, ADDR_FMT " %16s ", + ADDR_ARG(entry->pair->dev_addr), entry->pair->name); + seq_printf(m, ADDR_FMT " %16s %10u %5s\n", + ADDR_ARG(veth_to_netdev(entry)->dev_addr), + veth_to_netdev(entry)->name, + VEID(veth_to_netdev(entry)->owner_env), + entry->allow_mac_change ? "allow" : "deny"); + return 0; +} + +static void *vehwaddr_seq_start(struct seq_file *m, loff_t *pos) +{ + read_lock(&ve_hwaddr_lock); + return seq_list_start_head(&veth_hwaddr_list, *pos); +} + +static void *vehwaddr_seq_next(struct seq_file *m, void *v, loff_t *pos) +{ + return seq_list_next(v, &veth_hwaddr_list, pos); +} + +static void vehwaddr_seq_stop(struct seq_file *m, void *v) +{ + read_unlock(&ve_hwaddr_lock); +} + +static struct seq_operations vehwaddr_seq_op = { + .start = vehwaddr_seq_start, + .next = vehwaddr_seq_next, + .stop = vehwaddr_seq_stop, + .show = vehwaddr_seq_show, +}; + +static int vehwaddr_open(struct inode *inode, struct file *file) +{ + return seq_open(file, &vehwaddr_seq_op); +} + +static struct file_operations proc_vehwaddr_operations = { + .open = vehwaddr_open, + .read = seq_read, + .llseek = seq_lseek, + .release = seq_release, +}; +#endif + +int real_ve_hwaddr(envid_t veid, int op, + unsigned char *dev_addr, int addrlen, char *name, + unsigned char *dev_addr_ve, int addrlen_ve, char *name_ve) +{ + int err; + struct ve_struct *ve; + char ve_addr[ETH_ALEN]; + + err = -EPERM; + if (!capable(CAP_NET_ADMIN)) + goto out; + + err = -EINVAL; + switch (op) { + case VE_ETH_ADD: + if (addrlen != ETH_ALEN) + goto out; + if (addrlen_ve != ETH_ALEN && addrlen_ve != 0) + goto out; + /* If ve addr is not set then we use dev_addr[3] & 0x80 for it */ + if (addrlen_ve == 0 && (dev_addr[3] & 0x80)) + goto out; + if (addrlen_ve == 0) { + memcpy(ve_addr, dev_addr, ETH_ALEN); + ve_addr[3] |= 0x80; + } else { + memcpy(ve_addr, dev_addr_ve, ETH_ALEN); + } + + ve = get_ve_by_id(veid); + err = -ESRCH; + if (!ve) + goto out; + + down_read(&ve->op_sem); + if (ve->is_running) + err = veth_entry_add(ve, dev_addr, name, ve_addr, name_ve); + up_read(&ve->op_sem); + put_ve(ve); + break; + + case VE_ETH_DEL: + if (name[0] == '\0') + goto out; + ve = get_ve_by_id(veid); + err = -ESRCH; + if (!ve) + goto out; + + down_read(&ve->op_sem); + if (ve->is_running) + err = veth_entry_del(ve, name); + up_read(&ve->op_sem); + put_ve(ve); + break; + case VE_ETH_ALLOW_MAC_CHANGE: + case VE_ETH_DENY_MAC_CHANGE: + err = veth_allow_change_mac(veid, name, + op == VE_ETH_ALLOW_MAC_CHANGE); + break; + } + +out: + return err; +} + +int veth_ioctl(struct file *file, unsigned int cmd, unsigned long arg) +{ + int err; + + err = -ENOTTY; + switch(cmd) { + case VETHCTL_VE_HWADDR: { + struct vzctl_ve_hwaddr s; + + err = -EFAULT; + if (copy_from_user(&s, (void __user *)arg, sizeof(s))) + break; + err = real_ve_hwaddr(s.veid, s.op, s.dev_addr, s.addrlen, + s.dev_name, s.dev_addr_ve, s.addrlen_ve, + s.dev_name_ve); + } + break; + } + return err; +} + +static struct vzioctlinfo vethcalls = { + .type = VETHCTLTYPE, + .ioctl = veth_ioctl, + .compat_ioctl = veth_ioctl, + .owner = THIS_MODULE, +}; + +struct net_device * veth_dev_start(char *dev_addr, char *name) +{ + struct net_device *dev; + int err; + + if (!is_valid_ether_addr(dev_addr)) + return ERR_PTR(-EADDRNOTAVAIL); + + dev = alloc_netdev(sizeof(struct veth_struct), name, veth_setup); + if (!dev) + return ERR_PTR(-ENOMEM); + dev->nd_net = get_exec_env()->ve_netns; + if (strchr(dev->name, '%')) { + err = dev_alloc_name(dev, dev->name); + if (err < 0) + goto err; + } + if ((err = register_netdev(dev)) != 0) + goto err; + + memcpy(dev->dev_addr, dev_addr, ETH_ALEN); + dev->addr_len = ETH_ALEN; + + return dev; +err: + free_netdev(dev); + printk(KERN_ERR "%s initialization error err=%d\n", name, err); + return ERR_PTR(err); +} + +static int veth_start(void *data) +{ + return 0; +} + +static void veth_stop(void *data) +{ + struct ve_struct *env; + struct veth_struct *entry, *tmp; + + env = (struct ve_struct *)data; + down(&hwaddr_sem); + list_for_each_entry_safe(entry, tmp, &veth_hwaddr_list, hwaddr_list) + if (VEID(env) == VEID(veth_to_netdev(entry)->owner_env)) + veth_pair_del(env, entry); + up(&hwaddr_sem); +} + +static struct ve_hook veth_ve_hook = { + .init = veth_start, + .fini = veth_stop, + .owner = THIS_MODULE, + .priority = HOOK_PRIO_NET, +}; + +__init int veth_init(void) +{ +#ifdef CONFIG_PROC_FS + struct proc_dir_entry *de; + + de = proc_create("veth", S_IFREG|S_IRUSR, proc_vz_dir, + &proc_vehwaddr_operations); + if (de == NULL) + printk(KERN_WARNING "veth: can't make vehwaddr proc entry\n"); +#endif + + register_netdev_rst(&veth_netdev_rst); + ve_hook_register(VE_SS_CHAIN, &veth_ve_hook); + vzioctl_register(&vethcalls); + return 0; +} + +__exit void veth_exit(void) +{ + struct veth_struct *entry; + struct list_head *tmp, *n; + struct ve_struct *ve; + + vzioctl_unregister(&vethcalls); + ve_hook_unregister(&veth_ve_hook); + unregister_netdev_rst(&veth_netdev_rst); + +#ifdef CONFIG_PROC_FS + remove_proc_entry("veth", proc_vz_dir); +#endif + + down(&hwaddr_sem); + list_for_each_safe(tmp, n, &veth_hwaddr_list) { + entry = list_entry(tmp, struct veth_struct, hwaddr_list); + ve = get_ve(veth_to_netdev(entry)->owner_env); + + veth_pair_del(ve, entry); + + put_ve(ve); + } + up(&hwaddr_sem); +} + +module_init(veth_init); +module_exit(veth_exit); + +MODULE_AUTHOR("Andrey Mirkin "); +MODULE_DESCRIPTION("Virtuozzo Virtual Ethernet Device"); +MODULE_LICENSE("GPL v2"); + diff -urNp linux-2.6.32.48/drivers/net/wan/cycx_x25.c linux-2.6.32.48-openvz/drivers/net/wan/cycx_x25.c --- linux-2.6.32.48/drivers/net/wan/cycx_x25.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/wan/cycx_x25.c 2011-11-17 20:08:53.000000000 -0500 @@ -1017,6 +1017,8 @@ static void hex_dump(char *msg, unsigned unsigned char hex[1024], * phex = hex; + pax_track_stack(); + if (len >= (sizeof(hex) / 2)) len = (sizeof(hex) / 2) - 1; diff -urNp linux-2.6.32.48/drivers/net/wan/hdlc_x25.c linux-2.6.32.48-openvz/drivers/net/wan/hdlc_x25.c --- linux-2.6.32.48/drivers/net/wan/hdlc_x25.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/wan/hdlc_x25.c 2011-11-17 20:09:04.000000000 -0500 @@ -136,16 +136,16 @@ static netdev_tx_t x25_xmit(struct sk_bu static int x25_open(struct net_device *dev) { - struct lapb_register_struct cb; + static struct lapb_register_struct cb = { + .connect_confirmation = x25_connected, + .connect_indication = x25_connected, + .disconnect_confirmation = x25_disconnected, + .disconnect_indication = x25_disconnected, + .data_indication = x25_data_indication, + .data_transmit = x25_data_transmit + }; int result; - cb.connect_confirmation = x25_connected; - cb.connect_indication = x25_connected; - cb.disconnect_confirmation = x25_disconnected; - cb.disconnect_indication = x25_disconnected; - cb.data_indication = x25_data_indication; - cb.data_transmit = x25_data_transmit; - result = lapb_register(dev, &cb); if (result != LAPB_OK) return result; diff -urNp linux-2.6.32.48/drivers/net/wimax/i2400m/usb-fw.c linux-2.6.32.48-openvz/drivers/net/wimax/i2400m/usb-fw.c --- linux-2.6.32.48/drivers/net/wimax/i2400m/usb-fw.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/wimax/i2400m/usb-fw.c 2011-11-17 20:08:53.000000000 -0500 @@ -263,6 +263,8 @@ ssize_t i2400mu_bus_bm_wait_for_ack(stru int do_autopm = 1; DECLARE_COMPLETION_ONSTACK(notif_completion); + pax_track_stack(); + d_fnstart(8, dev, "(i2400m %p ack %p size %zu)\n", i2400m, ack, ack_size); BUG_ON(_ack == i2400m->bm_ack_buf); diff -urNp linux-2.6.32.48/drivers/net/wireless/airo.c linux-2.6.32.48-openvz/drivers/net/wireless/airo.c --- linux-2.6.32.48/drivers/net/wireless/airo.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/wireless/airo.c 2011-11-17 20:08:53.000000000 -0500 @@ -3003,6 +3003,8 @@ static void airo_process_scan_results (s BSSListElement * loop_net; BSSListElement * tmp_net; + pax_track_stack(); + /* Blow away current list of scan results */ list_for_each_entry_safe (loop_net, tmp_net, &ai->network_list, list) { list_move_tail (&loop_net->list, &ai->network_free_list); @@ -3783,6 +3785,8 @@ static u16 setup_card(struct airo_info * WepKeyRid wkr; int rc; + pax_track_stack(); + memset( &mySsid, 0, sizeof( mySsid ) ); kfree (ai->flash); ai->flash = NULL; @@ -4758,6 +4762,8 @@ static int proc_stats_rid_open( struct i __le32 *vals = stats.vals; int len; + pax_track_stack(); + if ((file->private_data = kzalloc(sizeof(struct proc_data ), GFP_KERNEL)) == NULL) return -ENOMEM; data = (struct proc_data *)file->private_data; @@ -5487,6 +5493,8 @@ static int proc_BSSList_open( struct ino /* If doLoseSync is not 1, we won't do a Lose Sync */ int doLoseSync = -1; + pax_track_stack(); + if ((file->private_data = kzalloc(sizeof(struct proc_data ), GFP_KERNEL)) == NULL) return -ENOMEM; data = (struct proc_data *)file->private_data; @@ -7193,6 +7201,8 @@ static int airo_get_aplist(struct net_de int i; int loseSync = capable(CAP_NET_ADMIN) ? 1: -1; + pax_track_stack(); + qual = kmalloc(IW_MAX_AP * sizeof(*qual), GFP_KERNEL); if (!qual) return -ENOMEM; @@ -7753,6 +7763,8 @@ static void airo_read_wireless_stats(str CapabilityRid cap_rid; __le32 *vals = stats_rid.vals; + pax_track_stack(); + /* Get stats out of the card */ clear_bit(JOB_WSTATS, &local->jobs); if (local->power.event) { diff -urNp linux-2.6.32.48/drivers/net/wireless/ath/ath5k/debug.c linux-2.6.32.48-openvz/drivers/net/wireless/ath/ath5k/debug.c --- linux-2.6.32.48/drivers/net/wireless/ath/ath5k/debug.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/wireless/ath/ath5k/debug.c 2011-11-17 20:08:53.000000000 -0500 @@ -205,6 +205,8 @@ static ssize_t read_file_beacon(struct f unsigned int v; u64 tsf; + pax_track_stack(); + v = ath5k_hw_reg_read(sc->ah, AR5K_BEACON); len += snprintf(buf+len, sizeof(buf)-len, "%-24s0x%08x\tintval: %d\tTIM: 0x%x\n", @@ -318,6 +320,8 @@ static ssize_t read_file_debug(struct fi unsigned int len = 0; unsigned int i; + pax_track_stack(); + len += snprintf(buf+len, sizeof(buf)-len, "DEBUG LEVEL: 0x%08x\n\n", sc->debug.level); diff -urNp linux-2.6.32.48/drivers/net/wireless/ath/ath9k/debug.c linux-2.6.32.48-openvz/drivers/net/wireless/ath/ath9k/debug.c --- linux-2.6.32.48/drivers/net/wireless/ath/ath9k/debug.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/wireless/ath/ath9k/debug.c 2011-11-17 20:08:53.000000000 -0500 @@ -220,6 +220,8 @@ static ssize_t read_file_interrupt(struc char buf[512]; unsigned int len = 0; + pax_track_stack(); + len += snprintf(buf + len, sizeof(buf) - len, "%8s: %10u\n", "RX", sc->debug.stats.istats.rxok); len += snprintf(buf + len, sizeof(buf) - len, @@ -360,6 +362,8 @@ static ssize_t read_file_wiphy(struct fi int i; u8 addr[ETH_ALEN]; + pax_track_stack(); + len += snprintf(buf + len, sizeof(buf) - len, "primary: %s (%s chan=%d ht=%d)\n", wiphy_name(sc->pri_wiphy->hw->wiphy), diff -urNp linux-2.6.32.48/drivers/net/wireless/b43/debugfs.c linux-2.6.32.48-openvz/drivers/net/wireless/b43/debugfs.c --- linux-2.6.32.48/drivers/net/wireless/b43/debugfs.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/wireless/b43/debugfs.c 2011-11-17 20:08:53.000000000 -0500 @@ -43,7 +43,7 @@ static struct dentry *rootdir; struct b43_debugfs_fops { ssize_t (*read)(struct b43_wldev *dev, char *buf, size_t bufsize); int (*write)(struct b43_wldev *dev, const char *buf, size_t count); - struct file_operations fops; + const struct file_operations fops; /* Offset of struct b43_dfs_file in struct b43_dfsentry */ size_t file_struct_offset; }; diff -urNp linux-2.6.32.48/drivers/net/wireless/b43legacy/debugfs.c linux-2.6.32.48-openvz/drivers/net/wireless/b43legacy/debugfs.c --- linux-2.6.32.48/drivers/net/wireless/b43legacy/debugfs.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/wireless/b43legacy/debugfs.c 2011-11-17 20:08:53.000000000 -0500 @@ -44,7 +44,7 @@ static struct dentry *rootdir; struct b43legacy_debugfs_fops { ssize_t (*read)(struct b43legacy_wldev *dev, char *buf, size_t bufsize); int (*write)(struct b43legacy_wldev *dev, const char *buf, size_t count); - struct file_operations fops; + const struct file_operations fops; /* Offset of struct b43legacy_dfs_file in struct b43legacy_dfsentry */ size_t file_struct_offset; /* Take wl->irq_lock before calling read/write? */ diff -urNp linux-2.6.32.48/drivers/net/wireless/ipw2x00/ipw2100.c linux-2.6.32.48-openvz/drivers/net/wireless/ipw2x00/ipw2100.c --- linux-2.6.32.48/drivers/net/wireless/ipw2x00/ipw2100.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/wireless/ipw2x00/ipw2100.c 2011-11-17 20:08:53.000000000 -0500 @@ -2014,6 +2014,8 @@ static int ipw2100_set_essid(struct ipw2 int err; DECLARE_SSID_BUF(ssid); + pax_track_stack(); + IPW_DEBUG_HC("SSID: '%s'\n", print_ssid(ssid, essid, ssid_len)); if (ssid_len) @@ -5380,6 +5382,8 @@ static int ipw2100_set_key(struct ipw210 struct ipw2100_wep_key *wep_key = (void *)cmd.host_command_parameters; int err; + pax_track_stack(); + IPW_DEBUG_HC("WEP_KEY_INFO: index = %d, len = %d/%d\n", idx, keylen, len); diff -urNp linux-2.6.32.48/drivers/net/wireless/ipw2x00/libipw_rx.c linux-2.6.32.48-openvz/drivers/net/wireless/ipw2x00/libipw_rx.c --- linux-2.6.32.48/drivers/net/wireless/ipw2x00/libipw_rx.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/wireless/ipw2x00/libipw_rx.c 2011-11-17 20:08:53.000000000 -0500 @@ -1566,6 +1566,8 @@ static void libipw_process_probe_respons unsigned long flags; DECLARE_SSID_BUF(ssid); + pax_track_stack(); + LIBIPW_DEBUG_SCAN("'%s' (%pM" "): %c%c%c%c %c%c%c%c-%c%c%c%c %c%c%c%c\n", print_ssid(ssid, info_element->data, info_element->len), diff -urNp linux-2.6.32.48/drivers/net/wireless/iwlwifi/iwl-1000.c linux-2.6.32.48-openvz/drivers/net/wireless/iwlwifi/iwl-1000.c --- linux-2.6.32.48/drivers/net/wireless/iwlwifi/iwl-1000.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/wireless/iwlwifi/iwl-1000.c 2011-11-17 20:08:53.000000000 -0500 @@ -137,7 +137,7 @@ static struct iwl_lib_ops iwl1000_lib = }, }; -static struct iwl_ops iwl1000_ops = { +static const struct iwl_ops iwl1000_ops = { .ucode = &iwl5000_ucode, .lib = &iwl1000_lib, .hcmd = &iwl5000_hcmd, diff -urNp linux-2.6.32.48/drivers/net/wireless/iwlwifi/iwl3945-base.c linux-2.6.32.48-openvz/drivers/net/wireless/iwlwifi/iwl3945-base.c --- linux-2.6.32.48/drivers/net/wireless/iwlwifi/iwl3945-base.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/wireless/iwlwifi/iwl3945-base.c 2011-11-17 20:09:04.000000000 -0500 @@ -3927,7 +3927,9 @@ static int iwl3945_pci_probe(struct pci_ */ if (iwl3945_mod_params.disable_hw_scan) { IWL_DEBUG_INFO(priv, "Disabling hw_scan\n"); - iwl3945_hw_ops.hw_scan = NULL; + pax_open_kernel(); + *(void **)&iwl3945_hw_ops.hw_scan = NULL; + pax_close_kernel(); } diff -urNp linux-2.6.32.48/drivers/net/wireless/iwlwifi/iwl-3945.c linux-2.6.32.48-openvz/drivers/net/wireless/iwlwifi/iwl-3945.c --- linux-2.6.32.48/drivers/net/wireless/iwlwifi/iwl-3945.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/wireless/iwlwifi/iwl-3945.c 2011-11-17 20:08:53.000000000 -0500 @@ -2874,7 +2874,7 @@ static struct iwl_hcmd_utils_ops iwl3945 .build_addsta_hcmd = iwl3945_build_addsta_hcmd, }; -static struct iwl_ops iwl3945_ops = { +static const struct iwl_ops iwl3945_ops = { .ucode = &iwl3945_ucode, .lib = &iwl3945_lib, .hcmd = &iwl3945_hcmd, diff -urNp linux-2.6.32.48/drivers/net/wireless/iwlwifi/iwl-4965.c linux-2.6.32.48-openvz/drivers/net/wireless/iwlwifi/iwl-4965.c --- linux-2.6.32.48/drivers/net/wireless/iwlwifi/iwl-4965.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/wireless/iwlwifi/iwl-4965.c 2011-11-17 20:08:53.000000000 -0500 @@ -2345,7 +2345,7 @@ static struct iwl_lib_ops iwl4965_lib = }, }; -static struct iwl_ops iwl4965_ops = { +static const struct iwl_ops iwl4965_ops = { .ucode = &iwl4965_ucode, .lib = &iwl4965_lib, .hcmd = &iwl4965_hcmd, diff -urNp linux-2.6.32.48/drivers/net/wireless/iwlwifi/iwl-5000.c linux-2.6.32.48-openvz/drivers/net/wireless/iwlwifi/iwl-5000.c --- linux-2.6.32.48/drivers/net/wireless/iwlwifi/iwl-5000.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/wireless/iwlwifi/iwl-5000.c 2011-11-17 20:09:04.000000000 -0500 @@ -1633,14 +1633,14 @@ static struct iwl_lib_ops iwl5150_lib = }, }; -struct iwl_ops iwl5000_ops = { +const struct iwl_ops iwl5000_ops = { .ucode = &iwl5000_ucode, .lib = &iwl5000_lib, .hcmd = &iwl5000_hcmd, .utils = &iwl5000_hcmd_utils, }; -static struct iwl_ops iwl5150_ops = { +static const struct iwl_ops iwl5150_ops = { .ucode = &iwl5000_ucode, .lib = &iwl5150_lib, .hcmd = &iwl5000_hcmd, diff -urNp linux-2.6.32.48/drivers/net/wireless/iwlwifi/iwl-6000.c linux-2.6.32.48-openvz/drivers/net/wireless/iwlwifi/iwl-6000.c --- linux-2.6.32.48/drivers/net/wireless/iwlwifi/iwl-6000.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/wireless/iwlwifi/iwl-6000.c 2011-11-17 20:08:53.000000000 -0500 @@ -146,7 +146,7 @@ static struct iwl_hcmd_utils_ops iwl6000 .calc_rssi = iwl5000_calc_rssi, }; -static struct iwl_ops iwl6000_ops = { +static const struct iwl_ops iwl6000_ops = { .ucode = &iwl5000_ucode, .lib = &iwl6000_lib, .hcmd = &iwl5000_hcmd, diff -urNp linux-2.6.32.48/drivers/net/wireless/iwlwifi/iwl-agn.c linux-2.6.32.48-openvz/drivers/net/wireless/iwlwifi/iwl-agn.c --- linux-2.6.32.48/drivers/net/wireless/iwlwifi/iwl-agn.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/wireless/iwlwifi/iwl-agn.c 2011-11-17 20:09:04.000000000 -0500 @@ -2911,7 +2911,9 @@ static int iwl_pci_probe(struct pci_dev if (iwl_debug_level & IWL_DL_INFO) dev_printk(KERN_DEBUG, &(pdev->dev), "Disabling hw_scan\n"); - iwl_hw_ops.hw_scan = NULL; + pax_open_kernel(); + *(void **)&iwl_hw_ops.hw_scan = NULL; + pax_close_kernel(); } hw = iwl_alloc_all(cfg, &iwl_hw_ops); diff -urNp linux-2.6.32.48/drivers/net/wireless/iwlwifi/iwl-agn-rs.c linux-2.6.32.48-openvz/drivers/net/wireless/iwlwifi/iwl-agn-rs.c --- linux-2.6.32.48/drivers/net/wireless/iwlwifi/iwl-agn-rs.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/wireless/iwlwifi/iwl-agn-rs.c 2011-11-17 20:08:53.000000000 -0500 @@ -857,6 +857,8 @@ static void rs_tx_status(void *priv_r, s u8 active_index = 0; s32 tpt = 0; + pax_track_stack(); + IWL_DEBUG_RATE_LIMIT(priv, "get frame ack response, update rate scale window\n"); if (!ieee80211_is_data(hdr->frame_control) || @@ -2722,6 +2724,8 @@ static void rs_fill_link_cmd(struct iwl_ u8 valid_tx_ant = 0; struct iwl_link_quality_cmd *lq_cmd = &lq_sta->lq; + pax_track_stack(); + /* Override starting rate (index 0) if needed for debug purposes */ rs_dbgfs_set_mcs(lq_sta, &new_rate, index); diff -urNp linux-2.6.32.48/drivers/net/wireless/iwlwifi/iwl-debugfs.c linux-2.6.32.48-openvz/drivers/net/wireless/iwlwifi/iwl-debugfs.c --- linux-2.6.32.48/drivers/net/wireless/iwlwifi/iwl-debugfs.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/wireless/iwlwifi/iwl-debugfs.c 2011-11-17 20:08:53.000000000 -0500 @@ -524,6 +524,8 @@ static ssize_t iwl_dbgfs_status_read(str int pos = 0; const size_t bufsz = sizeof(buf); + pax_track_stack(); + pos += scnprintf(buf + pos, bufsz - pos, "STATUS_HCMD_ACTIVE:\t %d\n", test_bit(STATUS_HCMD_ACTIVE, &priv->status)); pos += scnprintf(buf + pos, bufsz - pos, "STATUS_HCMD_SYNC_ACTIVE: %d\n", @@ -658,6 +660,8 @@ static ssize_t iwl_dbgfs_qos_read(struct const size_t bufsz = sizeof(buf); ssize_t ret; + pax_track_stack(); + for (i = 0; i < AC_NUM; i++) { pos += scnprintf(buf + pos, bufsz - pos, "\tcw_min\tcw_max\taifsn\ttxop\n"); diff -urNp linux-2.6.32.48/drivers/net/wireless/iwlwifi/iwl-debug.h linux-2.6.32.48-openvz/drivers/net/wireless/iwlwifi/iwl-debug.h --- linux-2.6.32.48/drivers/net/wireless/iwlwifi/iwl-debug.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/wireless/iwlwifi/iwl-debug.h 2011-11-17 20:08:53.000000000 -0500 @@ -118,8 +118,8 @@ void iwl_dbgfs_unregister(struct iwl_pri #endif #else -#define IWL_DEBUG(__priv, level, fmt, args...) -#define IWL_DEBUG_LIMIT(__priv, level, fmt, args...) +#define IWL_DEBUG(__priv, level, fmt, args...) do {} while (0) +#define IWL_DEBUG_LIMIT(__priv, level, fmt, args...) do {} while (0) static inline void iwl_print_hex_dump(struct iwl_priv *priv, int level, void *p, u32 len) {} diff -urNp linux-2.6.32.48/drivers/net/wireless/iwlwifi/iwl-dev.h linux-2.6.32.48-openvz/drivers/net/wireless/iwlwifi/iwl-dev.h --- linux-2.6.32.48/drivers/net/wireless/iwlwifi/iwl-dev.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/wireless/iwlwifi/iwl-dev.h 2011-11-17 20:08:53.000000000 -0500 @@ -68,7 +68,7 @@ struct iwl_tx_queue; /* shared structures from iwl-5000.c */ extern struct iwl_mod_params iwl50_mod_params; -extern struct iwl_ops iwl5000_ops; +extern const struct iwl_ops iwl5000_ops; extern struct iwl_ucode_ops iwl5000_ucode; extern struct iwl_lib_ops iwl5000_lib; extern struct iwl_hcmd_ops iwl5000_hcmd; diff -urNp linux-2.6.32.48/drivers/net/wireless/iwmc3200wifi/debugfs.c linux-2.6.32.48-openvz/drivers/net/wireless/iwmc3200wifi/debugfs.c --- linux-2.6.32.48/drivers/net/wireless/iwmc3200wifi/debugfs.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/wireless/iwmc3200wifi/debugfs.c 2011-11-17 20:08:53.000000000 -0500 @@ -299,6 +299,8 @@ static ssize_t iwm_debugfs_fw_err_read(s int buf_len = 512; size_t len = 0; + pax_track_stack(); + if (*ppos != 0) return 0; if (count < sizeof(buf)) diff -urNp linux-2.6.32.48/drivers/net/wireless/libertas/debugfs.c linux-2.6.32.48-openvz/drivers/net/wireless/libertas/debugfs.c --- linux-2.6.32.48/drivers/net/wireless/libertas/debugfs.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/wireless/libertas/debugfs.c 2011-11-17 20:08:53.000000000 -0500 @@ -708,7 +708,7 @@ out_unlock: struct lbs_debugfs_files { const char *name; int perm; - struct file_operations fops; + const struct file_operations fops; }; static const struct lbs_debugfs_files debugfs_files[] = { diff -urNp linux-2.6.32.48/drivers/net/wireless/rndis_wlan.c linux-2.6.32.48-openvz/drivers/net/wireless/rndis_wlan.c --- linux-2.6.32.48/drivers/net/wireless/rndis_wlan.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/net/wireless/rndis_wlan.c 2011-11-17 20:08:53.000000000 -0500 @@ -1176,7 +1176,7 @@ static int set_rts_threshold(struct usbn devdbg(usbdev, "set_rts_threshold %i", rts_threshold); - if (rts_threshold < 0 || rts_threshold > 2347) + if (rts_threshold > 2347) rts_threshold = 2347; tmp = cpu_to_le32(rts_threshold); diff -urNp linux-2.6.32.48/drivers/oprofile/buffer_sync.c linux-2.6.32.48-openvz/drivers/oprofile/buffer_sync.c --- linux-2.6.32.48/drivers/oprofile/buffer_sync.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/oprofile/buffer_sync.c 2011-11-17 20:08:53.000000000 -0500 @@ -341,7 +341,7 @@ static void add_data(struct op_entry *en if (cookie == NO_COOKIE) offset = pc; if (cookie == INVALID_COOKIE) { - atomic_inc(&oprofile_stats.sample_lost_no_mapping); + atomic_inc_unchecked(&oprofile_stats.sample_lost_no_mapping); offset = pc; } if (cookie != last_cookie) { @@ -385,14 +385,14 @@ add_sample(struct mm_struct *mm, struct /* add userspace sample */ if (!mm) { - atomic_inc(&oprofile_stats.sample_lost_no_mm); + atomic_inc_unchecked(&oprofile_stats.sample_lost_no_mm); return 0; } cookie = lookup_dcookie(mm, s->eip, &offset); if (cookie == INVALID_COOKIE) { - atomic_inc(&oprofile_stats.sample_lost_no_mapping); + atomic_inc_unchecked(&oprofile_stats.sample_lost_no_mapping); return 0; } @@ -561,7 +561,7 @@ void sync_buffer(int cpu) /* ignore backtraces if failed to add a sample */ if (state == sb_bt_start) { state = sb_bt_ignore; - atomic_inc(&oprofile_stats.bt_lost_no_mapping); + atomic_inc_unchecked(&oprofile_stats.bt_lost_no_mapping); } } release_mm(mm); diff -urNp linux-2.6.32.48/drivers/oprofile/event_buffer.c linux-2.6.32.48-openvz/drivers/oprofile/event_buffer.c --- linux-2.6.32.48/drivers/oprofile/event_buffer.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/oprofile/event_buffer.c 2011-11-17 20:08:53.000000000 -0500 @@ -53,7 +53,7 @@ void add_event_entry(unsigned long value } if (buffer_pos == buffer_size) { - atomic_inc(&oprofile_stats.event_lost_overflow); + atomic_inc_unchecked(&oprofile_stats.event_lost_overflow); return; } diff -urNp linux-2.6.32.48/drivers/oprofile/oprof.c linux-2.6.32.48-openvz/drivers/oprofile/oprof.c --- linux-2.6.32.48/drivers/oprofile/oprof.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/oprofile/oprof.c 2011-11-17 20:08:53.000000000 -0500 @@ -110,7 +110,7 @@ static void switch_worker(struct work_st if (oprofile_ops.switch_events()) return; - atomic_inc(&oprofile_stats.multiplex_counter); + atomic_inc_unchecked(&oprofile_stats.multiplex_counter); start_switch_worker(); } diff -urNp linux-2.6.32.48/drivers/oprofile/oprofilefs.c linux-2.6.32.48-openvz/drivers/oprofile/oprofilefs.c --- linux-2.6.32.48/drivers/oprofile/oprofilefs.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/oprofile/oprofilefs.c 2011-11-17 20:08:53.000000000 -0500 @@ -187,7 +187,7 @@ static const struct file_operations atom int oprofilefs_create_ro_atomic(struct super_block *sb, struct dentry *root, - char const *name, atomic_t *val) + char const *name, atomic_unchecked_t *val) { struct dentry *d = __oprofilefs_create_file(sb, root, name, &atomic_ro_fops, 0444); diff -urNp linux-2.6.32.48/drivers/oprofile/oprofile_stats.c linux-2.6.32.48-openvz/drivers/oprofile/oprofile_stats.c --- linux-2.6.32.48/drivers/oprofile/oprofile_stats.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/oprofile/oprofile_stats.c 2011-11-17 20:08:53.000000000 -0500 @@ -30,11 +30,11 @@ void oprofile_reset_stats(void) cpu_buf->sample_invalid_eip = 0; } - atomic_set(&oprofile_stats.sample_lost_no_mm, 0); - atomic_set(&oprofile_stats.sample_lost_no_mapping, 0); - atomic_set(&oprofile_stats.event_lost_overflow, 0); - atomic_set(&oprofile_stats.bt_lost_no_mapping, 0); - atomic_set(&oprofile_stats.multiplex_counter, 0); + atomic_set_unchecked(&oprofile_stats.sample_lost_no_mm, 0); + atomic_set_unchecked(&oprofile_stats.sample_lost_no_mapping, 0); + atomic_set_unchecked(&oprofile_stats.event_lost_overflow, 0); + atomic_set_unchecked(&oprofile_stats.bt_lost_no_mapping, 0); + atomic_set_unchecked(&oprofile_stats.multiplex_counter, 0); } diff -urNp linux-2.6.32.48/drivers/oprofile/oprofile_stats.h linux-2.6.32.48-openvz/drivers/oprofile/oprofile_stats.h --- linux-2.6.32.48/drivers/oprofile/oprofile_stats.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/oprofile/oprofile_stats.h 2011-11-17 20:08:53.000000000 -0500 @@ -13,11 +13,11 @@ #include struct oprofile_stat_struct { - atomic_t sample_lost_no_mm; - atomic_t sample_lost_no_mapping; - atomic_t bt_lost_no_mapping; - atomic_t event_lost_overflow; - atomic_t multiplex_counter; + atomic_unchecked_t sample_lost_no_mm; + atomic_unchecked_t sample_lost_no_mapping; + atomic_unchecked_t bt_lost_no_mapping; + atomic_unchecked_t event_lost_overflow; + atomic_unchecked_t multiplex_counter; }; extern struct oprofile_stat_struct oprofile_stats; diff -urNp linux-2.6.32.48/drivers/parisc/pdc_stable.c linux-2.6.32.48-openvz/drivers/parisc/pdc_stable.c --- linux-2.6.32.48/drivers/parisc/pdc_stable.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/parisc/pdc_stable.c 2011-11-17 20:08:53.000000000 -0500 @@ -481,7 +481,7 @@ pdcspath_attr_store(struct kobject *kobj return ret; } -static struct sysfs_ops pdcspath_attr_ops = { +static const struct sysfs_ops pdcspath_attr_ops = { .show = pdcspath_attr_show, .store = pdcspath_attr_store, }; diff -urNp linux-2.6.32.48/drivers/parport/procfs.c linux-2.6.32.48-openvz/drivers/parport/procfs.c --- linux-2.6.32.48/drivers/parport/procfs.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/parport/procfs.c 2011-11-17 20:08:53.000000000 -0500 @@ -64,7 +64,7 @@ static int do_active_device(ctl_table *t *ppos += len; - return copy_to_user(result, buffer, len) ? -EFAULT : 0; + return (len > sizeof buffer || copy_to_user(result, buffer, len)) ? -EFAULT : 0; } #ifdef CONFIG_PARPORT_1284 @@ -106,7 +106,7 @@ static int do_autoprobe(ctl_table *table *ppos += len; - return copy_to_user (result, buffer, len) ? -EFAULT : 0; + return (len > sizeof buffer || copy_to_user (result, buffer, len)) ? -EFAULT : 0; } #endif /* IEEE1284.3 support. */ diff -urNp linux-2.6.32.48/drivers/pci/hotplug/acpiphp_glue.c linux-2.6.32.48-openvz/drivers/pci/hotplug/acpiphp_glue.c --- linux-2.6.32.48/drivers/pci/hotplug/acpiphp_glue.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/pci/hotplug/acpiphp_glue.c 2011-11-17 20:08:53.000000000 -0500 @@ -111,7 +111,7 @@ static int post_dock_fixups(struct notif } -static struct acpi_dock_ops acpiphp_dock_ops = { +static const struct acpi_dock_ops acpiphp_dock_ops = { .handler = handle_hotplug_event_func, }; diff -urNp linux-2.6.32.48/drivers/pci/hotplug/cpci_hotplug.h linux-2.6.32.48-openvz/drivers/pci/hotplug/cpci_hotplug.h --- linux-2.6.32.48/drivers/pci/hotplug/cpci_hotplug.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/pci/hotplug/cpci_hotplug.h 2011-11-17 20:09:04.000000000 -0500 @@ -59,7 +59,7 @@ struct cpci_hp_controller_ops { int (*hardware_test) (struct slot* slot, u32 value); u8 (*get_power) (struct slot* slot); int (*set_power) (struct slot* slot, int value); -}; +} __no_const; struct cpci_hp_controller { unsigned int irq; diff -urNp linux-2.6.32.48/drivers/pci/hotplug/cpqphp_nvram.c linux-2.6.32.48-openvz/drivers/pci/hotplug/cpqphp_nvram.c --- linux-2.6.32.48/drivers/pci/hotplug/cpqphp_nvram.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/pci/hotplug/cpqphp_nvram.c 2011-11-17 20:08:53.000000000 -0500 @@ -428,9 +428,13 @@ static u32 store_HRT (void __iomem *rom_ void compaq_nvram_init (void __iomem *rom_start) { + +#ifndef CONFIG_PAX_KERNEXEC if (rom_start) { compaq_int15_entry_point = (rom_start + ROM_INT15_PHY_ADDR - ROM_PHY_ADDR); } +#endif + dbg("int15 entry = %p\n", compaq_int15_entry_point); /* initialize our int15 lock */ diff -urNp linux-2.6.32.48/drivers/pci/hotplug/fakephp.c linux-2.6.32.48-openvz/drivers/pci/hotplug/fakephp.c --- linux-2.6.32.48/drivers/pci/hotplug/fakephp.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/pci/hotplug/fakephp.c 2011-11-17 20:08:53.000000000 -0500 @@ -73,7 +73,7 @@ static void legacy_release(struct kobjec } static struct kobj_type legacy_ktype = { - .sysfs_ops = &(struct sysfs_ops){ + .sysfs_ops = &(const struct sysfs_ops){ .store = legacy_store, .show = legacy_show }, .release = &legacy_release, diff -urNp linux-2.6.32.48/drivers/pci/intel-iommu.c linux-2.6.32.48-openvz/drivers/pci/intel-iommu.c --- linux-2.6.32.48/drivers/pci/intel-iommu.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/pci/intel-iommu.c 2011-11-17 20:08:53.000000000 -0500 @@ -2643,7 +2643,7 @@ error: return 0; } -static dma_addr_t intel_map_page(struct device *dev, struct page *page, +dma_addr_t intel_map_page(struct device *dev, struct page *page, unsigned long offset, size_t size, enum dma_data_direction dir, struct dma_attrs *attrs) @@ -2719,7 +2719,7 @@ static void add_unmap(struct dmar_domain spin_unlock_irqrestore(&async_umap_flush_lock, flags); } -static void intel_unmap_page(struct device *dev, dma_addr_t dev_addr, +void intel_unmap_page(struct device *dev, dma_addr_t dev_addr, size_t size, enum dma_data_direction dir, struct dma_attrs *attrs) { @@ -2768,7 +2768,7 @@ static void intel_unmap_page(struct devi } } -static void *intel_alloc_coherent(struct device *hwdev, size_t size, +void *intel_alloc_coherent(struct device *hwdev, size_t size, dma_addr_t *dma_handle, gfp_t flags) { void *vaddr; @@ -2800,7 +2800,7 @@ static void *intel_alloc_coherent(struct return NULL; } -static void intel_free_coherent(struct device *hwdev, size_t size, void *vaddr, +void intel_free_coherent(struct device *hwdev, size_t size, void *vaddr, dma_addr_t dma_handle) { int order; @@ -2812,7 +2812,7 @@ static void intel_free_coherent(struct d free_pages((unsigned long)vaddr, order); } -static void intel_unmap_sg(struct device *hwdev, struct scatterlist *sglist, +void intel_unmap_sg(struct device *hwdev, struct scatterlist *sglist, int nelems, enum dma_data_direction dir, struct dma_attrs *attrs) { @@ -2872,7 +2872,7 @@ static int intel_nontranslate_map_sg(str return nelems; } -static int intel_map_sg(struct device *hwdev, struct scatterlist *sglist, int nelems, +int intel_map_sg(struct device *hwdev, struct scatterlist *sglist, int nelems, enum dma_data_direction dir, struct dma_attrs *attrs) { int i; @@ -2941,12 +2941,12 @@ static int intel_map_sg(struct device *h return nelems; } -static int intel_mapping_error(struct device *dev, dma_addr_t dma_addr) +int intel_mapping_error(struct device *dev, dma_addr_t dma_addr) { return !dma_addr; } -struct dma_map_ops intel_dma_ops = { +const struct dma_map_ops intel_dma_ops = { .alloc_coherent = intel_alloc_coherent, .free_coherent = intel_free_coherent, .map_sg = intel_map_sg, diff -urNp linux-2.6.32.48/drivers/pci/pcie/aspm.c linux-2.6.32.48-openvz/drivers/pci/pcie/aspm.c --- linux-2.6.32.48/drivers/pci/pcie/aspm.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/pci/pcie/aspm.c 2011-11-17 20:08:53.000000000 -0500 @@ -27,9 +27,9 @@ #define MODULE_PARAM_PREFIX "pcie_aspm." /* Note: those are not register definitions */ -#define ASPM_STATE_L0S_UP (1) /* Upstream direction L0s state */ -#define ASPM_STATE_L0S_DW (2) /* Downstream direction L0s state */ -#define ASPM_STATE_L1 (4) /* L1 state */ +#define ASPM_STATE_L0S_UP (1U) /* Upstream direction L0s state */ +#define ASPM_STATE_L0S_DW (2U) /* Downstream direction L0s state */ +#define ASPM_STATE_L1 (4U) /* L1 state */ #define ASPM_STATE_L0S (ASPM_STATE_L0S_UP | ASPM_STATE_L0S_DW) #define ASPM_STATE_ALL (ASPM_STATE_L0S | ASPM_STATE_L1) diff -urNp linux-2.6.32.48/drivers/pci/probe.c linux-2.6.32.48-openvz/drivers/pci/probe.c --- linux-2.6.32.48/drivers/pci/probe.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/pci/probe.c 2011-11-17 20:08:53.000000000 -0500 @@ -62,14 +62,14 @@ static ssize_t pci_bus_show_cpuaffinity( return ret; } -static ssize_t inline pci_bus_show_cpumaskaffinity(struct device *dev, +static inline ssize_t pci_bus_show_cpumaskaffinity(struct device *dev, struct device_attribute *attr, char *buf) { return pci_bus_show_cpuaffinity(dev, 0, attr, buf); } -static ssize_t inline pci_bus_show_cpulistaffinity(struct device *dev, +static inline ssize_t pci_bus_show_cpulistaffinity(struct device *dev, struct device_attribute *attr, char *buf) { diff -urNp linux-2.6.32.48/drivers/pci/proc.c linux-2.6.32.48-openvz/drivers/pci/proc.c --- linux-2.6.32.48/drivers/pci/proc.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/pci/proc.c 2011-11-17 20:08:53.000000000 -0500 @@ -480,7 +480,16 @@ static const struct file_operations proc static int __init pci_proc_init(void) { struct pci_dev *dev = NULL; + +#ifdef CONFIG_GRKERNSEC_PROC_ADD +#ifdef CONFIG_GRKERNSEC_PROC_USER + proc_bus_pci_dir = proc_mkdir_mode("bus/pci", S_IRUSR | S_IXUSR, NULL); +#elif defined(CONFIG_GRKERNSEC_PROC_USERGROUP) + proc_bus_pci_dir = proc_mkdir_mode("bus/pci", S_IRUSR | S_IXUSR | S_IRGRP | S_IXGRP, NULL); +#endif +#else proc_bus_pci_dir = proc_mkdir("bus/pci", NULL); +#endif proc_create("devices", 0, proc_bus_pci_dir, &proc_bus_pci_dev_operations); proc_initialized = 1; diff -urNp linux-2.6.32.48/drivers/pci/slot.c linux-2.6.32.48-openvz/drivers/pci/slot.c --- linux-2.6.32.48/drivers/pci/slot.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/pci/slot.c 2011-11-17 20:08:53.000000000 -0500 @@ -29,7 +29,7 @@ static ssize_t pci_slot_attr_store(struc return attribute->store ? attribute->store(slot, buf, len) : -EIO; } -static struct sysfs_ops pci_slot_sysfs_ops = { +static const struct sysfs_ops pci_slot_sysfs_ops = { .show = pci_slot_attr_show, .store = pci_slot_attr_store, }; diff -urNp linux-2.6.32.48/drivers/pcmcia/pcmcia_ioctl.c linux-2.6.32.48-openvz/drivers/pcmcia/pcmcia_ioctl.c --- linux-2.6.32.48/drivers/pcmcia/pcmcia_ioctl.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/pcmcia/pcmcia_ioctl.c 2011-11-17 20:08:53.000000000 -0500 @@ -819,7 +819,7 @@ static int ds_ioctl(struct inode * inode return -EFAULT; } } - buf = kmalloc(sizeof(ds_ioctl_arg_t), GFP_KERNEL); + buf = kzalloc(sizeof(ds_ioctl_arg_t), GFP_KERNEL); if (!buf) return -ENOMEM; diff -urNp linux-2.6.32.48/drivers/platform/x86/acer-wmi.c linux-2.6.32.48-openvz/drivers/platform/x86/acer-wmi.c --- linux-2.6.32.48/drivers/platform/x86/acer-wmi.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/platform/x86/acer-wmi.c 2011-11-17 20:08:53.000000000 -0500 @@ -918,7 +918,7 @@ static int update_bl_status(struct backl return 0; } -static struct backlight_ops acer_bl_ops = { +static const struct backlight_ops acer_bl_ops = { .get_brightness = read_brightness, .update_status = update_bl_status, }; diff -urNp linux-2.6.32.48/drivers/platform/x86/asus_acpi.c linux-2.6.32.48-openvz/drivers/platform/x86/asus_acpi.c --- linux-2.6.32.48/drivers/platform/x86/asus_acpi.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/platform/x86/asus_acpi.c 2011-11-17 20:08:53.000000000 -0500 @@ -1396,7 +1396,7 @@ static int asus_hotk_remove(struct acpi_ return 0; } -static struct backlight_ops asus_backlight_data = { +static const struct backlight_ops asus_backlight_data = { .get_brightness = read_brightness, .update_status = set_brightness_status, }; diff -urNp linux-2.6.32.48/drivers/platform/x86/asus-laptop.c linux-2.6.32.48-openvz/drivers/platform/x86/asus-laptop.c --- linux-2.6.32.48/drivers/platform/x86/asus-laptop.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/platform/x86/asus-laptop.c 2011-11-17 20:08:53.000000000 -0500 @@ -250,7 +250,7 @@ static struct backlight_device *asus_bac */ static int read_brightness(struct backlight_device *bd); static int update_bl_status(struct backlight_device *bd); -static struct backlight_ops asusbl_ops = { +static const struct backlight_ops asusbl_ops = { .get_brightness = read_brightness, .update_status = update_bl_status, }; diff -urNp linux-2.6.32.48/drivers/platform/x86/compal-laptop.c linux-2.6.32.48-openvz/drivers/platform/x86/compal-laptop.c --- linux-2.6.32.48/drivers/platform/x86/compal-laptop.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/platform/x86/compal-laptop.c 2011-11-17 20:08:53.000000000 -0500 @@ -163,7 +163,7 @@ static int bl_update_status(struct backl return set_lcd_level(b->props.brightness); } -static struct backlight_ops compalbl_ops = { +static const struct backlight_ops compalbl_ops = { .get_brightness = bl_get_brightness, .update_status = bl_update_status, }; diff -urNp linux-2.6.32.48/drivers/platform/x86/dell-laptop.c linux-2.6.32.48-openvz/drivers/platform/x86/dell-laptop.c --- linux-2.6.32.48/drivers/platform/x86/dell-laptop.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/platform/x86/dell-laptop.c 2011-11-17 20:08:53.000000000 -0500 @@ -318,7 +318,7 @@ static int dell_get_intensity(struct bac return buffer.output[1]; } -static struct backlight_ops dell_ops = { +static const struct backlight_ops dell_ops = { .get_brightness = dell_get_intensity, .update_status = dell_send_intensity, }; diff -urNp linux-2.6.32.48/drivers/platform/x86/eeepc-laptop.c linux-2.6.32.48-openvz/drivers/platform/x86/eeepc-laptop.c --- linux-2.6.32.48/drivers/platform/x86/eeepc-laptop.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/platform/x86/eeepc-laptop.c 2011-11-17 20:08:53.000000000 -0500 @@ -245,7 +245,7 @@ static struct device *eeepc_hwmon_device */ static int read_brightness(struct backlight_device *bd); static int update_bl_status(struct backlight_device *bd); -static struct backlight_ops eeepcbl_ops = { +static const struct backlight_ops eeepcbl_ops = { .get_brightness = read_brightness, .update_status = update_bl_status, }; diff -urNp linux-2.6.32.48/drivers/platform/x86/fujitsu-laptop.c linux-2.6.32.48-openvz/drivers/platform/x86/fujitsu-laptop.c --- linux-2.6.32.48/drivers/platform/x86/fujitsu-laptop.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/platform/x86/fujitsu-laptop.c 2011-11-17 20:08:53.000000000 -0500 @@ -436,7 +436,7 @@ static int bl_update_status(struct backl return ret; } -static struct backlight_ops fujitsubl_ops = { +static const struct backlight_ops fujitsubl_ops = { .get_brightness = bl_get_brightness, .update_status = bl_update_status, }; diff -urNp linux-2.6.32.48/drivers/platform/x86/msi-laptop.c linux-2.6.32.48-openvz/drivers/platform/x86/msi-laptop.c --- linux-2.6.32.48/drivers/platform/x86/msi-laptop.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/platform/x86/msi-laptop.c 2011-11-17 20:08:53.000000000 -0500 @@ -161,7 +161,7 @@ static int bl_update_status(struct backl return set_lcd_level(b->props.brightness); } -static struct backlight_ops msibl_ops = { +static const struct backlight_ops msibl_ops = { .get_brightness = bl_get_brightness, .update_status = bl_update_status, }; diff -urNp linux-2.6.32.48/drivers/platform/x86/panasonic-laptop.c linux-2.6.32.48-openvz/drivers/platform/x86/panasonic-laptop.c --- linux-2.6.32.48/drivers/platform/x86/panasonic-laptop.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/platform/x86/panasonic-laptop.c 2011-11-17 20:08:53.000000000 -0500 @@ -352,7 +352,7 @@ static int bl_set_status(struct backligh return acpi_pcc_write_sset(pcc, SINF_DC_CUR_BRIGHT, bright); } -static struct backlight_ops pcc_backlight_ops = { +static const struct backlight_ops pcc_backlight_ops = { .get_brightness = bl_get, .update_status = bl_set_status, }; diff -urNp linux-2.6.32.48/drivers/platform/x86/sony-laptop.c linux-2.6.32.48-openvz/drivers/platform/x86/sony-laptop.c --- linux-2.6.32.48/drivers/platform/x86/sony-laptop.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/platform/x86/sony-laptop.c 2011-11-17 20:08:53.000000000 -0500 @@ -850,7 +850,7 @@ static int sony_backlight_get_brightness } static struct backlight_device *sony_backlight_device; -static struct backlight_ops sony_backlight_ops = { +static const struct backlight_ops sony_backlight_ops = { .update_status = sony_backlight_update_status, .get_brightness = sony_backlight_get_brightness, }; diff -urNp linux-2.6.32.48/drivers/platform/x86/thinkpad_acpi.c linux-2.6.32.48-openvz/drivers/platform/x86/thinkpad_acpi.c --- linux-2.6.32.48/drivers/platform/x86/thinkpad_acpi.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/platform/x86/thinkpad_acpi.c 2011-11-17 20:09:04.000000000 -0500 @@ -2139,7 +2139,7 @@ static int hotkey_mask_get(void) return 0; } -void static hotkey_mask_warn_incomplete_mask(void) +static void hotkey_mask_warn_incomplete_mask(void) { /* log only what the user can fix... */ const u32 wantedmask = hotkey_driver_mask & @@ -6125,7 +6125,7 @@ static void tpacpi_brightness_notify_cha BACKLIGHT_UPDATE_HOTKEY); } -static struct backlight_ops ibm_backlight_data = { +static const struct backlight_ops ibm_backlight_data = { .get_brightness = brightness_get, .update_status = brightness_update_status, }; diff -urNp linux-2.6.32.48/drivers/platform/x86/toshiba_acpi.c linux-2.6.32.48-openvz/drivers/platform/x86/toshiba_acpi.c --- linux-2.6.32.48/drivers/platform/x86/toshiba_acpi.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/platform/x86/toshiba_acpi.c 2011-11-17 20:08:53.000000000 -0500 @@ -671,7 +671,7 @@ static acpi_status remove_device(void) return AE_OK; } -static struct backlight_ops toshiba_backlight_data = { +static const struct backlight_ops toshiba_backlight_data = { .get_brightness = get_lcd, .update_status = set_lcd_status, }; diff -urNp linux-2.6.32.48/drivers/pnp/pnpbios/bioscalls.c linux-2.6.32.48-openvz/drivers/pnp/pnpbios/bioscalls.c --- linux-2.6.32.48/drivers/pnp/pnpbios/bioscalls.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/pnp/pnpbios/bioscalls.c 2011-11-17 20:08:53.000000000 -0500 @@ -60,7 +60,7 @@ do { \ set_desc_limit(&gdt[(selname) >> 3], (size) - 1); \ } while(0) -static struct desc_struct bad_bios_desc = GDT_ENTRY_INIT(0x4092, +static const struct desc_struct bad_bios_desc = GDT_ENTRY_INIT(0x4093, (unsigned long)__va(0x400UL), PAGE_SIZE - 0x400 - 1); /* @@ -97,7 +97,10 @@ static inline u16 call_pnp_bios(u16 func cpu = get_cpu(); save_desc_40 = get_cpu_gdt_table(cpu)[0x40 / 8]; + + pax_open_kernel(); get_cpu_gdt_table(cpu)[0x40 / 8] = bad_bios_desc; + pax_close_kernel(); /* On some boxes IRQ's during PnP BIOS calls are deadly. */ spin_lock_irqsave(&pnp_bios_lock, flags); @@ -135,7 +138,10 @@ static inline u16 call_pnp_bios(u16 func :"memory"); spin_unlock_irqrestore(&pnp_bios_lock, flags); + pax_open_kernel(); get_cpu_gdt_table(cpu)[0x40 / 8] = save_desc_40; + pax_close_kernel(); + put_cpu(); /* If we get here and this is set then the PnP BIOS faulted on us. */ @@ -469,7 +475,7 @@ int pnp_bios_read_escd(char *data, u32 n return status; } -void pnpbios_calls_init(union pnp_bios_install_struct *header) +void __init pnpbios_calls_init(union pnp_bios_install_struct *header) { int i; @@ -477,6 +483,8 @@ void pnpbios_calls_init(union pnp_bios_i pnp_bios_callpoint.offset = header->fields.pm16offset; pnp_bios_callpoint.segment = PNP_CS16; + pax_open_kernel(); + for_each_possible_cpu(i) { struct desc_struct *gdt = get_cpu_gdt_table(i); if (!gdt) @@ -488,4 +496,6 @@ void pnpbios_calls_init(union pnp_bios_i set_desc_base(&gdt[GDT_ENTRY_PNPBIOS_DS], (unsigned long)__va(header->fields.pm16dseg)); } + + pax_close_kernel(); } diff -urNp linux-2.6.32.48/drivers/pnp/resource.c linux-2.6.32.48-openvz/drivers/pnp/resource.c --- linux-2.6.32.48/drivers/pnp/resource.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/pnp/resource.c 2011-11-17 20:08:53.000000000 -0500 @@ -355,7 +355,7 @@ int pnp_check_irq(struct pnp_dev *dev, s return 1; /* check if the resource is valid */ - if (*irq < 0 || *irq > 15) + if (*irq > 15) return 0; /* check if the resource is reserved */ @@ -419,7 +419,7 @@ int pnp_check_dma(struct pnp_dev *dev, s return 1; /* check if the resource is valid */ - if (*dma < 0 || *dma == 4 || *dma > 7) + if (*dma == 4 || *dma > 7) return 0; /* check if the resource is reserved */ diff -urNp linux-2.6.32.48/drivers/power/bq27x00_battery.c linux-2.6.32.48-openvz/drivers/power/bq27x00_battery.c --- linux-2.6.32.48/drivers/power/bq27x00_battery.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/power/bq27x00_battery.c 2011-11-17 20:09:04.000000000 -0500 @@ -44,7 +44,7 @@ struct bq27x00_device_info; struct bq27x00_access_methods { int (*read)(u8 reg, int *rt_value, int b_single, struct bq27x00_device_info *di); -}; +} __no_const; struct bq27x00_device_info { struct device *dev; diff -urNp linux-2.6.32.48/drivers/rtc/rtc-dev.c linux-2.6.32.48-openvz/drivers/rtc/rtc-dev.c --- linux-2.6.32.48/drivers/rtc/rtc-dev.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/rtc/rtc-dev.c 2011-11-17 20:08:53.000000000 -0500 @@ -14,6 +14,7 @@ #include #include #include +#include #include "rtc-core.h" static dev_t rtc_devt; @@ -357,6 +358,8 @@ static long rtc_dev_ioctl(struct file *f if (copy_from_user(&tm, uarg, sizeof(tm))) return -EFAULT; + gr_log_timechange(); + return rtc_set_time(rtc, &tm); case RTC_PIE_ON: diff -urNp linux-2.6.32.48/drivers/s390/cio/qdio_perf.c linux-2.6.32.48-openvz/drivers/s390/cio/qdio_perf.c --- linux-2.6.32.48/drivers/s390/cio/qdio_perf.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/s390/cio/qdio_perf.c 2011-11-17 20:08:53.000000000 -0500 @@ -31,51 +31,51 @@ static struct proc_dir_entry *qdio_perf_ static int qdio_perf_proc_show(struct seq_file *m, void *v) { seq_printf(m, "Number of qdio interrupts\t\t\t: %li\n", - (long)atomic_long_read(&perf_stats.qdio_int)); + (long)atomic_long_read_unchecked(&perf_stats.qdio_int)); seq_printf(m, "Number of PCI interrupts\t\t\t: %li\n", - (long)atomic_long_read(&perf_stats.pci_int)); + (long)atomic_long_read_unchecked(&perf_stats.pci_int)); seq_printf(m, "Number of adapter interrupts\t\t\t: %li\n", - (long)atomic_long_read(&perf_stats.thin_int)); + (long)atomic_long_read_unchecked(&perf_stats.thin_int)); seq_printf(m, "\n"); seq_printf(m, "Inbound tasklet runs\t\t\t\t: %li\n", - (long)atomic_long_read(&perf_stats.tasklet_inbound)); + (long)atomic_long_read_unchecked(&perf_stats.tasklet_inbound)); seq_printf(m, "Outbound tasklet runs\t\t\t\t: %li\n", - (long)atomic_long_read(&perf_stats.tasklet_outbound)); + (long)atomic_long_read_unchecked(&perf_stats.tasklet_outbound)); seq_printf(m, "Adapter interrupt tasklet runs/loops\t\t: %li/%li\n", - (long)atomic_long_read(&perf_stats.tasklet_thinint), - (long)atomic_long_read(&perf_stats.tasklet_thinint_loop)); + (long)atomic_long_read_unchecked(&perf_stats.tasklet_thinint), + (long)atomic_long_read_unchecked(&perf_stats.tasklet_thinint_loop)); seq_printf(m, "Adapter interrupt inbound tasklet runs/loops\t: %li/%li\n", - (long)atomic_long_read(&perf_stats.thinint_inbound), - (long)atomic_long_read(&perf_stats.thinint_inbound_loop)); + (long)atomic_long_read_unchecked(&perf_stats.thinint_inbound), + (long)atomic_long_read_unchecked(&perf_stats.thinint_inbound_loop)); seq_printf(m, "\n"); seq_printf(m, "Number of SIGA In issued\t\t\t: %li\n", - (long)atomic_long_read(&perf_stats.siga_in)); + (long)atomic_long_read_unchecked(&perf_stats.siga_in)); seq_printf(m, "Number of SIGA Out issued\t\t\t: %li\n", - (long)atomic_long_read(&perf_stats.siga_out)); + (long)atomic_long_read_unchecked(&perf_stats.siga_out)); seq_printf(m, "Number of SIGA Sync issued\t\t\t: %li\n", - (long)atomic_long_read(&perf_stats.siga_sync)); + (long)atomic_long_read_unchecked(&perf_stats.siga_sync)); seq_printf(m, "\n"); seq_printf(m, "Number of inbound transfers\t\t\t: %li\n", - (long)atomic_long_read(&perf_stats.inbound_handler)); + (long)atomic_long_read_unchecked(&perf_stats.inbound_handler)); seq_printf(m, "Number of outbound transfers\t\t\t: %li\n", - (long)atomic_long_read(&perf_stats.outbound_handler)); + (long)atomic_long_read_unchecked(&perf_stats.outbound_handler)); seq_printf(m, "\n"); seq_printf(m, "Number of fast requeues (outg. SBAL w/o SIGA)\t: %li\n", - (long)atomic_long_read(&perf_stats.fast_requeue)); + (long)atomic_long_read_unchecked(&perf_stats.fast_requeue)); seq_printf(m, "Number of outbound target full condition\t: %li\n", - (long)atomic_long_read(&perf_stats.outbound_target_full)); + (long)atomic_long_read_unchecked(&perf_stats.outbound_target_full)); seq_printf(m, "Number of outbound tasklet mod_timer calls\t: %li\n", - (long)atomic_long_read(&perf_stats.debug_tl_out_timer)); + (long)atomic_long_read_unchecked(&perf_stats.debug_tl_out_timer)); seq_printf(m, "Number of stop polling calls\t\t\t: %li\n", - (long)atomic_long_read(&perf_stats.debug_stop_polling)); + (long)atomic_long_read_unchecked(&perf_stats.debug_stop_polling)); seq_printf(m, "AI inbound tasklet loops after stop polling\t: %li\n", - (long)atomic_long_read(&perf_stats.thinint_inbound_loop2)); + (long)atomic_long_read_unchecked(&perf_stats.thinint_inbound_loop2)); seq_printf(m, "QEBSM EQBS total/incomplete\t\t\t: %li/%li\n", - (long)atomic_long_read(&perf_stats.debug_eqbs_all), - (long)atomic_long_read(&perf_stats.debug_eqbs_incomplete)); + (long)atomic_long_read_unchecked(&perf_stats.debug_eqbs_all), + (long)atomic_long_read_unchecked(&perf_stats.debug_eqbs_incomplete)); seq_printf(m, "QEBSM SQBS total/incomplete\t\t\t: %li/%li\n", - (long)atomic_long_read(&perf_stats.debug_sqbs_all), - (long)atomic_long_read(&perf_stats.debug_sqbs_incomplete)); + (long)atomic_long_read_unchecked(&perf_stats.debug_sqbs_all), + (long)atomic_long_read_unchecked(&perf_stats.debug_sqbs_incomplete)); seq_printf(m, "\n"); return 0; } diff -urNp linux-2.6.32.48/drivers/s390/cio/qdio_perf.h linux-2.6.32.48-openvz/drivers/s390/cio/qdio_perf.h --- linux-2.6.32.48/drivers/s390/cio/qdio_perf.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/s390/cio/qdio_perf.h 2011-11-17 20:08:53.000000000 -0500 @@ -13,46 +13,46 @@ struct qdio_perf_stats { /* interrupt handler calls */ - atomic_long_t qdio_int; - atomic_long_t pci_int; - atomic_long_t thin_int; + atomic_long_unchecked_t qdio_int; + atomic_long_unchecked_t pci_int; + atomic_long_unchecked_t thin_int; /* tasklet runs */ - atomic_long_t tasklet_inbound; - atomic_long_t tasklet_outbound; - atomic_long_t tasklet_thinint; - atomic_long_t tasklet_thinint_loop; - atomic_long_t thinint_inbound; - atomic_long_t thinint_inbound_loop; - atomic_long_t thinint_inbound_loop2; + atomic_long_unchecked_t tasklet_inbound; + atomic_long_unchecked_t tasklet_outbound; + atomic_long_unchecked_t tasklet_thinint; + atomic_long_unchecked_t tasklet_thinint_loop; + atomic_long_unchecked_t thinint_inbound; + atomic_long_unchecked_t thinint_inbound_loop; + atomic_long_unchecked_t thinint_inbound_loop2; /* signal adapter calls */ - atomic_long_t siga_out; - atomic_long_t siga_in; - atomic_long_t siga_sync; + atomic_long_unchecked_t siga_out; + atomic_long_unchecked_t siga_in; + atomic_long_unchecked_t siga_sync; /* misc */ - atomic_long_t inbound_handler; - atomic_long_t outbound_handler; - atomic_long_t fast_requeue; - atomic_long_t outbound_target_full; + atomic_long_unchecked_t inbound_handler; + atomic_long_unchecked_t outbound_handler; + atomic_long_unchecked_t fast_requeue; + atomic_long_unchecked_t outbound_target_full; /* for debugging */ - atomic_long_t debug_tl_out_timer; - atomic_long_t debug_stop_polling; - atomic_long_t debug_eqbs_all; - atomic_long_t debug_eqbs_incomplete; - atomic_long_t debug_sqbs_all; - atomic_long_t debug_sqbs_incomplete; + atomic_long_unchecked_t debug_tl_out_timer; + atomic_long_unchecked_t debug_stop_polling; + atomic_long_unchecked_t debug_eqbs_all; + atomic_long_unchecked_t debug_eqbs_incomplete; + atomic_long_unchecked_t debug_sqbs_all; + atomic_long_unchecked_t debug_sqbs_incomplete; }; extern struct qdio_perf_stats perf_stats; extern int qdio_performance_stats; -static inline void qdio_perf_stat_inc(atomic_long_t *count) +static inline void qdio_perf_stat_inc(atomic_long_unchecked_t *count) { if (qdio_performance_stats) - atomic_long_inc(count); + atomic_long_inc_unchecked(count); } int qdio_setup_perf_stats(void); diff -urNp linux-2.6.32.48/drivers/scsi/aacraid/aacraid.h linux-2.6.32.48-openvz/drivers/scsi/aacraid/aacraid.h --- linux-2.6.32.48/drivers/scsi/aacraid/aacraid.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/aacraid/aacraid.h 2011-11-17 20:09:04.000000000 -0500 @@ -471,7 +471,7 @@ struct adapter_ops int (*adapter_scsi)(struct fib * fib, struct scsi_cmnd * cmd); /* Administrative operations */ int (*adapter_comm)(struct aac_dev * dev, int comm); -}; +} __no_const; /* * Define which interrupt handler needs to be installed diff -urNp linux-2.6.32.48/drivers/scsi/aacraid/commctrl.c linux-2.6.32.48-openvz/drivers/scsi/aacraid/commctrl.c --- linux-2.6.32.48/drivers/scsi/aacraid/commctrl.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/aacraid/commctrl.c 2011-11-17 20:08:53.000000000 -0500 @@ -481,6 +481,7 @@ static int aac_send_raw_srb(struct aac_d u32 actual_fibsize64, actual_fibsize = 0; int i; + pax_track_stack(); if (dev->in_reset) { dprintk((KERN_DEBUG"aacraid: send raw srb -EBUSY\n")); diff -urNp linux-2.6.32.48/drivers/scsi/aic94xx/aic94xx_init.c linux-2.6.32.48-openvz/drivers/scsi/aic94xx/aic94xx_init.c --- linux-2.6.32.48/drivers/scsi/aic94xx/aic94xx_init.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/aic94xx/aic94xx_init.c 2011-11-17 20:08:53.000000000 -0500 @@ -485,7 +485,7 @@ static ssize_t asd_show_update_bios(stru flash_error_table[i].reason); } -static DEVICE_ATTR(update_bios, S_IRUGO|S_IWUGO, +static DEVICE_ATTR(update_bios, S_IRUGO|S_IWUSR, asd_show_update_bios, asd_store_update_bios); static int asd_create_dev_attrs(struct asd_ha_struct *asd_ha) diff -urNp linux-2.6.32.48/drivers/scsi/bfa/bfa_iocfc.h linux-2.6.32.48-openvz/drivers/scsi/bfa/bfa_iocfc.h --- linux-2.6.32.48/drivers/scsi/bfa/bfa_iocfc.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/bfa/bfa_iocfc.h 2011-11-17 20:09:04.000000000 -0500 @@ -61,7 +61,7 @@ struct bfa_hwif_s { void (*hw_isr_mode_set)(struct bfa_s *bfa, bfa_boolean_t msix); void (*hw_msix_getvecs)(struct bfa_s *bfa, u32 *vecmap, u32 *nvecs, u32 *maxvec); -}; +} __no_const; typedef void (*bfa_cb_iocfc_t) (void *cbarg, enum bfa_status status); struct bfa_iocfc_s { diff -urNp linux-2.6.32.48/drivers/scsi/bfa/bfa_ioc.h linux-2.6.32.48-openvz/drivers/scsi/bfa/bfa_ioc.h --- linux-2.6.32.48/drivers/scsi/bfa/bfa_ioc.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/bfa/bfa_ioc.h 2011-11-17 20:09:04.000000000 -0500 @@ -127,7 +127,7 @@ struct bfa_ioc_cbfn_s { bfa_ioc_disable_cbfn_t disable_cbfn; bfa_ioc_hbfail_cbfn_t hbfail_cbfn; bfa_ioc_reset_cbfn_t reset_cbfn; -}; +} __no_const; /** * Heartbeat failure notification queue element. diff -urNp linux-2.6.32.48/drivers/scsi/BusLogic.c linux-2.6.32.48-openvz/drivers/scsi/BusLogic.c --- linux-2.6.32.48/drivers/scsi/BusLogic.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/BusLogic.c 2011-11-17 20:08:53.000000000 -0500 @@ -961,6 +961,8 @@ static int __init BusLogic_InitializeFla static void __init BusLogic_InitializeProbeInfoList(struct BusLogic_HostAdapter *PrototypeHostAdapter) { + pax_track_stack(); + /* If a PCI BIOS is present, interrogate it for MultiMaster and FlashPoint Host Adapters; otherwise, default to the standard ISA MultiMaster probe. diff -urNp linux-2.6.32.48/drivers/scsi/dpt_i2o.c linux-2.6.32.48-openvz/drivers/scsi/dpt_i2o.c --- linux-2.6.32.48/drivers/scsi/dpt_i2o.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/dpt_i2o.c 2011-11-17 20:08:53.000000000 -0500 @@ -1804,6 +1804,8 @@ static int adpt_i2o_passthru(adpt_hba* p dma_addr_t addr; ulong flags = 0; + pax_track_stack(); + memset(&msg, 0, MAX_MESSAGE_SIZE*4); // get user msg size in u32s if(get_user(size, &user_msg[0])){ @@ -2297,6 +2299,8 @@ static s32 adpt_scsi_to_i2o(adpt_hba* pH s32 rcode; dma_addr_t addr; + pax_track_stack(); + memset(msg, 0 , sizeof(msg)); len = scsi_bufflen(cmd); direction = 0x00000000; diff -urNp linux-2.6.32.48/drivers/scsi/eata.c linux-2.6.32.48-openvz/drivers/scsi/eata.c --- linux-2.6.32.48/drivers/scsi/eata.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/eata.c 2011-11-17 20:08:53.000000000 -0500 @@ -1087,6 +1087,8 @@ static int port_detect(unsigned long por struct hostdata *ha; char name[16]; + pax_track_stack(); + sprintf(name, "%s%d", driver_name, j); if (!request_region(port_base, REGION_SIZE, driver_name)) { diff -urNp linux-2.6.32.48/drivers/scsi/fcoe/libfcoe.c linux-2.6.32.48-openvz/drivers/scsi/fcoe/libfcoe.c --- linux-2.6.32.48/drivers/scsi/fcoe/libfcoe.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/fcoe/libfcoe.c 2011-11-17 20:08:53.000000000 -0500 @@ -809,6 +809,8 @@ static void fcoe_ctlr_recv_els(struct fc size_t rlen; size_t dlen; + pax_track_stack(); + fiph = (struct fip_header *)skb->data; sub = fiph->fip_subcode; if (sub != FIP_SC_REQ && sub != FIP_SC_REP) diff -urNp linux-2.6.32.48/drivers/scsi/fnic/fnic_main.c linux-2.6.32.48-openvz/drivers/scsi/fnic/fnic_main.c --- linux-2.6.32.48/drivers/scsi/fnic/fnic_main.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/fnic/fnic_main.c 2011-11-17 20:09:04.000000000 -0500 @@ -669,7 +669,7 @@ static int __devinit fnic_probe(struct p /* Start local port initiatialization */ lp->link_up = 0; - lp->tt = fnic_transport_template; + memcpy((void *)&lp->tt, &fnic_transport_template, sizeof(fnic_transport_template)); lp->max_retry_count = fnic->config.flogi_retries; lp->max_rport_retry_count = fnic->config.plogi_retries; diff -urNp linux-2.6.32.48/drivers/scsi/gdth.c linux-2.6.32.48-openvz/drivers/scsi/gdth.c --- linux-2.6.32.48/drivers/scsi/gdth.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/gdth.c 2011-11-17 20:08:53.000000000 -0500 @@ -4102,6 +4102,8 @@ static int ioc_lockdrv(void __user *arg) ulong flags; gdth_ha_str *ha; + pax_track_stack(); + if (copy_from_user(&ldrv, arg, sizeof(gdth_ioctl_lockdrv))) return -EFAULT; ha = gdth_find_ha(ldrv.ionode); @@ -4134,6 +4136,8 @@ static int ioc_resetdrv(void __user *arg gdth_ha_str *ha; int rval; + pax_track_stack(); + if (copy_from_user(&res, arg, sizeof(gdth_ioctl_reset)) || res.number >= MAX_HDRIVES) return -EFAULT; @@ -4169,6 +4173,8 @@ static int ioc_general(void __user *arg, gdth_ha_str *ha; int rval; + pax_track_stack(); + if (copy_from_user(&gen, arg, sizeof(gdth_ioctl_general))) return -EFAULT; ha = gdth_find_ha(gen.ionode); @@ -4625,6 +4631,9 @@ static void gdth_flush(gdth_ha_str *ha) int i; gdth_cmd_str gdtcmd; char cmnd[MAX_COMMAND_SIZE]; + + pax_track_stack(); + memset(cmnd, 0xff, MAX_COMMAND_SIZE); TRACE2(("gdth_flush() hanum %d\n", ha->hanum)); diff -urNp linux-2.6.32.48/drivers/scsi/gdth_proc.c linux-2.6.32.48-openvz/drivers/scsi/gdth_proc.c --- linux-2.6.32.48/drivers/scsi/gdth_proc.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/gdth_proc.c 2011-11-17 20:08:53.000000000 -0500 @@ -46,6 +46,9 @@ static int gdth_set_asc_info(struct Scsi ulong64 paddr; char cmnd[MAX_COMMAND_SIZE]; + + pax_track_stack(); + memset(cmnd, 0xff, 12); memset(&gdtcmd, 0, sizeof(gdth_cmd_str)); @@ -174,6 +177,8 @@ static int gdth_get_info(char *buffer,ch gdth_hget_str *phg; char cmnd[MAX_COMMAND_SIZE]; + pax_track_stack(); + gdtcmd = kmalloc(sizeof(*gdtcmd), GFP_KERNEL); estr = kmalloc(sizeof(*estr), GFP_KERNEL); if (!gdtcmd || !estr) diff -urNp linux-2.6.32.48/drivers/scsi/hosts.c linux-2.6.32.48-openvz/drivers/scsi/hosts.c --- linux-2.6.32.48/drivers/scsi/hosts.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/hosts.c 2011-11-17 20:08:53.000000000 -0500 @@ -40,7 +40,7 @@ #include "scsi_logging.h" -static atomic_t scsi_host_next_hn; /* host_no for next new host */ +static atomic_unchecked_t scsi_host_next_hn; /* host_no for next new host */ static void scsi_host_cls_release(struct device *dev) @@ -344,7 +344,7 @@ struct Scsi_Host *scsi_host_alloc(struct * subtract one because we increment first then return, but we need to * know what the next host number was before increment */ - shost->host_no = atomic_inc_return(&scsi_host_next_hn) - 1; + shost->host_no = atomic_inc_return_unchecked(&scsi_host_next_hn) - 1; shost->dma_channel = 0xff; /* These three are default values which can be overridden */ @@ -401,9 +401,8 @@ struct Scsi_Host *scsi_host_alloc(struct device_initialize(&shost->shost_gendev); dev_set_name(&shost->shost_gendev, "host%d", shost->host_no); -#ifndef CONFIG_SYSFS_DEPRECATED - shost->shost_gendev.bus = &scsi_bus_type; -#endif + if (!sysfs_deprecated) + shost->shost_gendev.bus = &scsi_bus_type; shost->shost_gendev.type = &scsi_host_type; device_initialize(&shost->shost_dev); diff -urNp linux-2.6.32.48/drivers/scsi/ipr.c linux-2.6.32.48-openvz/drivers/scsi/ipr.c --- linux-2.6.32.48/drivers/scsi/ipr.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/ipr.c 2011-11-17 20:09:04.000000000 -0500 @@ -5286,7 +5286,7 @@ static bool ipr_qc_fill_rtf(struct ata_q return true; } -static struct ata_port_operations ipr_sata_ops = { +static const struct ata_port_operations ipr_sata_ops = { .phy_reset = ipr_ata_phy_reset, .hardreset = ipr_sata_reset, .post_internal_cmd = ipr_ata_post_internal, diff -urNp linux-2.6.32.48/drivers/scsi/ips.h linux-2.6.32.48-openvz/drivers/scsi/ips.h --- linux-2.6.32.48/drivers/scsi/ips.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/ips.h 2011-11-17 20:09:04.000000000 -0500 @@ -1027,7 +1027,7 @@ typedef struct { int (*intr)(struct ips_ha *); void (*enableint)(struct ips_ha *); uint32_t (*statupd)(struct ips_ha *); -} ips_hw_func_t; +} __no_const ips_hw_func_t; typedef struct ips_ha { uint8_t ha_id[IPS_MAX_CHANNELS+1]; diff -urNp linux-2.6.32.48/drivers/scsi/libfc/fc_exch.c linux-2.6.32.48-openvz/drivers/scsi/libfc/fc_exch.c --- linux-2.6.32.48/drivers/scsi/libfc/fc_exch.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/libfc/fc_exch.c 2011-11-17 20:08:53.000000000 -0500 @@ -86,12 +86,12 @@ struct fc_exch_mgr { * all together if not used XXX */ struct { - atomic_t no_free_exch; - atomic_t no_free_exch_xid; - atomic_t xid_not_found; - atomic_t xid_busy; - atomic_t seq_not_found; - atomic_t non_bls_resp; + atomic_unchecked_t no_free_exch; + atomic_unchecked_t no_free_exch_xid; + atomic_unchecked_t xid_not_found; + atomic_unchecked_t xid_busy; + atomic_unchecked_t seq_not_found; + atomic_unchecked_t non_bls_resp; } stats; }; #define fc_seq_exch(sp) container_of(sp, struct fc_exch, seq) @@ -510,7 +510,7 @@ static struct fc_exch *fc_exch_em_alloc( /* allocate memory for exchange */ ep = mempool_alloc(mp->ep_pool, GFP_ATOMIC); if (!ep) { - atomic_inc(&mp->stats.no_free_exch); + atomic_inc_unchecked(&mp->stats.no_free_exch); goto out; } memset(ep, 0, sizeof(*ep)); @@ -557,7 +557,7 @@ out: return ep; err: spin_unlock_bh(&pool->lock); - atomic_inc(&mp->stats.no_free_exch_xid); + atomic_inc_unchecked(&mp->stats.no_free_exch_xid); mempool_free(ep, mp->ep_pool); return NULL; } @@ -690,7 +690,7 @@ static enum fc_pf_rjt_reason fc_seq_look xid = ntohs(fh->fh_ox_id); /* we originated exch */ ep = fc_exch_find(mp, xid); if (!ep) { - atomic_inc(&mp->stats.xid_not_found); + atomic_inc_unchecked(&mp->stats.xid_not_found); reject = FC_RJT_OX_ID; goto out; } @@ -720,7 +720,7 @@ static enum fc_pf_rjt_reason fc_seq_look ep = fc_exch_find(mp, xid); if ((f_ctl & FC_FC_FIRST_SEQ) && fc_sof_is_init(fr_sof(fp))) { if (ep) { - atomic_inc(&mp->stats.xid_busy); + atomic_inc_unchecked(&mp->stats.xid_busy); reject = FC_RJT_RX_ID; goto rel; } @@ -731,7 +731,7 @@ static enum fc_pf_rjt_reason fc_seq_look } xid = ep->xid; /* get our XID */ } else if (!ep) { - atomic_inc(&mp->stats.xid_not_found); + atomic_inc_unchecked(&mp->stats.xid_not_found); reject = FC_RJT_RX_ID; /* XID not found */ goto out; } @@ -752,7 +752,7 @@ static enum fc_pf_rjt_reason fc_seq_look } else { sp = &ep->seq; if (sp->id != fh->fh_seq_id) { - atomic_inc(&mp->stats.seq_not_found); + atomic_inc_unchecked(&mp->stats.seq_not_found); reject = FC_RJT_SEQ_ID; /* sequence/exch should exist */ goto rel; } @@ -1163,22 +1163,22 @@ static void fc_exch_recv_seq_resp(struct ep = fc_exch_find(mp, ntohs(fh->fh_ox_id)); if (!ep) { - atomic_inc(&mp->stats.xid_not_found); + atomic_inc_unchecked(&mp->stats.xid_not_found); goto out; } if (ep->esb_stat & ESB_ST_COMPLETE) { - atomic_inc(&mp->stats.xid_not_found); + atomic_inc_unchecked(&mp->stats.xid_not_found); goto out; } if (ep->rxid == FC_XID_UNKNOWN) ep->rxid = ntohs(fh->fh_rx_id); if (ep->sid != 0 && ep->sid != ntoh24(fh->fh_d_id)) { - atomic_inc(&mp->stats.xid_not_found); + atomic_inc_unchecked(&mp->stats.xid_not_found); goto rel; } if (ep->did != ntoh24(fh->fh_s_id) && ep->did != FC_FID_FLOGI) { - atomic_inc(&mp->stats.xid_not_found); + atomic_inc_unchecked(&mp->stats.xid_not_found); goto rel; } sof = fr_sof(fp); @@ -1189,7 +1189,7 @@ static void fc_exch_recv_seq_resp(struct } else { sp = &ep->seq; if (sp->id != fh->fh_seq_id) { - atomic_inc(&mp->stats.seq_not_found); + atomic_inc_unchecked(&mp->stats.seq_not_found); goto rel; } } @@ -1249,9 +1249,9 @@ static void fc_exch_recv_resp(struct fc_ sp = fc_seq_lookup_orig(mp, fp); /* doesn't hold sequence */ if (!sp) - atomic_inc(&mp->stats.xid_not_found); + atomic_inc_unchecked(&mp->stats.xid_not_found); else - atomic_inc(&mp->stats.non_bls_resp); + atomic_inc_unchecked(&mp->stats.non_bls_resp); fc_frame_free(fp); } diff -urNp linux-2.6.32.48/drivers/scsi/libsas/sas_ata.c linux-2.6.32.48-openvz/drivers/scsi/libsas/sas_ata.c --- linux-2.6.32.48/drivers/scsi/libsas/sas_ata.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/libsas/sas_ata.c 2011-11-17 20:08:53.000000000 -0500 @@ -343,7 +343,7 @@ static int sas_ata_scr_read(struct ata_l } } -static struct ata_port_operations sas_sata_ops = { +static const struct ata_port_operations sas_sata_ops = { .phy_reset = sas_ata_phy_reset, .post_internal_cmd = sas_ata_post_internal, .qc_defer = ata_std_qc_defer, diff -urNp linux-2.6.32.48/drivers/scsi/lpfc/lpfc_debugfs.c linux-2.6.32.48-openvz/drivers/scsi/lpfc/lpfc_debugfs.c --- linux-2.6.32.48/drivers/scsi/lpfc/lpfc_debugfs.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/lpfc/lpfc_debugfs.c 2011-11-17 20:08:53.000000000 -0500 @@ -124,7 +124,7 @@ struct lpfc_debug { int len; }; -static atomic_t lpfc_debugfs_seq_trc_cnt = ATOMIC_INIT(0); +static atomic_unchecked_t lpfc_debugfs_seq_trc_cnt = ATOMIC_INIT(0); static unsigned long lpfc_debugfs_start_time = 0L; /** @@ -158,7 +158,7 @@ lpfc_debugfs_disc_trc_data(struct lpfc_v lpfc_debugfs_enable = 0; len = 0; - index = (atomic_read(&vport->disc_trc_cnt) + 1) & + index = (atomic_read_unchecked(&vport->disc_trc_cnt) + 1) & (lpfc_debugfs_max_disc_trc - 1); for (i = index; i < lpfc_debugfs_max_disc_trc; i++) { dtp = vport->disc_trc + i; @@ -219,7 +219,7 @@ lpfc_debugfs_slow_ring_trc_data(struct l lpfc_debugfs_enable = 0; len = 0; - index = (atomic_read(&phba->slow_ring_trc_cnt) + 1) & + index = (atomic_read_unchecked(&phba->slow_ring_trc_cnt) + 1) & (lpfc_debugfs_max_slow_ring_trc - 1); for (i = index; i < lpfc_debugfs_max_slow_ring_trc; i++) { dtp = phba->slow_ring_trc + i; @@ -397,6 +397,8 @@ lpfc_debugfs_dumpHBASlim_data(struct lpf uint32_t *ptr; char buffer[1024]; + pax_track_stack(); + off = 0; spin_lock_irq(&phba->hbalock); @@ -634,14 +636,14 @@ lpfc_debugfs_disc_trc(struct lpfc_vport !vport || !vport->disc_trc) return; - index = atomic_inc_return(&vport->disc_trc_cnt) & + index = atomic_inc_return_unchecked(&vport->disc_trc_cnt) & (lpfc_debugfs_max_disc_trc - 1); dtp = vport->disc_trc + index; dtp->fmt = fmt; dtp->data1 = data1; dtp->data2 = data2; dtp->data3 = data3; - dtp->seq_cnt = atomic_inc_return(&lpfc_debugfs_seq_trc_cnt); + dtp->seq_cnt = atomic_inc_return_unchecked(&lpfc_debugfs_seq_trc_cnt); dtp->jif = jiffies; #endif return; @@ -672,14 +674,14 @@ lpfc_debugfs_slow_ring_trc(struct lpfc_h !phba || !phba->slow_ring_trc) return; - index = atomic_inc_return(&phba->slow_ring_trc_cnt) & + index = atomic_inc_return_unchecked(&phba->slow_ring_trc_cnt) & (lpfc_debugfs_max_slow_ring_trc - 1); dtp = phba->slow_ring_trc + index; dtp->fmt = fmt; dtp->data1 = data1; dtp->data2 = data2; dtp->data3 = data3; - dtp->seq_cnt = atomic_inc_return(&lpfc_debugfs_seq_trc_cnt); + dtp->seq_cnt = atomic_inc_return_unchecked(&lpfc_debugfs_seq_trc_cnt); dtp->jif = jiffies; #endif return; @@ -1364,7 +1366,7 @@ lpfc_debugfs_initialize(struct lpfc_vpor "slow_ring buffer\n"); goto debug_failed; } - atomic_set(&phba->slow_ring_trc_cnt, 0); + atomic_set_unchecked(&phba->slow_ring_trc_cnt, 0); memset(phba->slow_ring_trc, 0, (sizeof(struct lpfc_debugfs_trc) * lpfc_debugfs_max_slow_ring_trc)); @@ -1410,7 +1412,7 @@ lpfc_debugfs_initialize(struct lpfc_vpor "buffer\n"); goto debug_failed; } - atomic_set(&vport->disc_trc_cnt, 0); + atomic_set_unchecked(&vport->disc_trc_cnt, 0); snprintf(name, sizeof(name), "discovery_trace"); vport->debug_disc_trc = diff -urNp linux-2.6.32.48/drivers/scsi/lpfc/lpfc.h linux-2.6.32.48-openvz/drivers/scsi/lpfc/lpfc.h --- linux-2.6.32.48/drivers/scsi/lpfc/lpfc.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/lpfc/lpfc.h 2011-11-17 20:08:53.000000000 -0500 @@ -400,7 +400,7 @@ struct lpfc_vport { struct dentry *debug_nodelist; struct dentry *vport_debugfs_root; struct lpfc_debugfs_trc *disc_trc; - atomic_t disc_trc_cnt; + atomic_unchecked_t disc_trc_cnt; #endif uint8_t stat_data_enabled; uint8_t stat_data_blocked; @@ -725,8 +725,8 @@ struct lpfc_hba { struct timer_list fabric_block_timer; unsigned long bit_flags; #define FABRIC_COMANDS_BLOCKED 0 - atomic_t num_rsrc_err; - atomic_t num_cmd_success; + atomic_unchecked_t num_rsrc_err; + atomic_unchecked_t num_cmd_success; unsigned long last_rsrc_error_time; unsigned long last_ramp_down_time; unsigned long last_ramp_up_time; @@ -740,7 +740,7 @@ struct lpfc_hba { struct dentry *debug_dumpDif; /* BlockGuard BPL*/ struct dentry *debug_slow_ring_trc; struct lpfc_debugfs_trc *slow_ring_trc; - atomic_t slow_ring_trc_cnt; + atomic_unchecked_t slow_ring_trc_cnt; #endif /* Used for deferred freeing of ELS data buffers */ diff -urNp linux-2.6.32.48/drivers/scsi/lpfc/lpfc_init.c linux-2.6.32.48-openvz/drivers/scsi/lpfc/lpfc_init.c --- linux-2.6.32.48/drivers/scsi/lpfc/lpfc_init.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/lpfc/lpfc_init.c 2011-11-17 20:09:04.000000000 -0500 @@ -8021,8 +8021,10 @@ lpfc_init(void) printk(LPFC_COPYRIGHT "\n"); if (lpfc_enable_npiv) { - lpfc_transport_functions.vport_create = lpfc_vport_create; - lpfc_transport_functions.vport_delete = lpfc_vport_delete; + pax_open_kernel(); + *(void **)&lpfc_transport_functions.vport_create = lpfc_vport_create; + *(void **)&lpfc_transport_functions.vport_delete = lpfc_vport_delete; + pax_close_kernel(); } lpfc_transport_template = fc_attach_transport(&lpfc_transport_functions); diff -urNp linux-2.6.32.48/drivers/scsi/lpfc/lpfc_scsi.c linux-2.6.32.48-openvz/drivers/scsi/lpfc/lpfc_scsi.c --- linux-2.6.32.48/drivers/scsi/lpfc/lpfc_scsi.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/lpfc/lpfc_scsi.c 2011-11-17 20:08:53.000000000 -0500 @@ -259,7 +259,7 @@ lpfc_rampdown_queue_depth(struct lpfc_hb uint32_t evt_posted; spin_lock_irqsave(&phba->hbalock, flags); - atomic_inc(&phba->num_rsrc_err); + atomic_inc_unchecked(&phba->num_rsrc_err); phba->last_rsrc_error_time = jiffies; if ((phba->last_ramp_down_time + QUEUE_RAMP_DOWN_INTERVAL) > jiffies) { @@ -300,7 +300,7 @@ lpfc_rampup_queue_depth(struct lpfc_vpor unsigned long flags; struct lpfc_hba *phba = vport->phba; uint32_t evt_posted; - atomic_inc(&phba->num_cmd_success); + atomic_inc_unchecked(&phba->num_cmd_success); if (vport->cfg_lun_queue_depth <= queue_depth) return; @@ -343,8 +343,8 @@ lpfc_ramp_down_queue_handler(struct lpfc int i; struct lpfc_rport_data *rdata; - num_rsrc_err = atomic_read(&phba->num_rsrc_err); - num_cmd_success = atomic_read(&phba->num_cmd_success); + num_rsrc_err = atomic_read_unchecked(&phba->num_rsrc_err); + num_cmd_success = atomic_read_unchecked(&phba->num_cmd_success); vports = lpfc_create_vport_work_array(phba); if (vports != NULL) @@ -378,8 +378,8 @@ lpfc_ramp_down_queue_handler(struct lpfc } } lpfc_destroy_vport_work_array(phba, vports); - atomic_set(&phba->num_rsrc_err, 0); - atomic_set(&phba->num_cmd_success, 0); + atomic_set_unchecked(&phba->num_rsrc_err, 0); + atomic_set_unchecked(&phba->num_cmd_success, 0); } /** @@ -427,8 +427,8 @@ lpfc_ramp_up_queue_handler(struct lpfc_h } } lpfc_destroy_vport_work_array(phba, vports); - atomic_set(&phba->num_rsrc_err, 0); - atomic_set(&phba->num_cmd_success, 0); + atomic_set_unchecked(&phba->num_rsrc_err, 0); + atomic_set_unchecked(&phba->num_cmd_success, 0); } /** diff -urNp linux-2.6.32.48/drivers/scsi/megaraid/megaraid_mbox.c linux-2.6.32.48-openvz/drivers/scsi/megaraid/megaraid_mbox.c --- linux-2.6.32.48/drivers/scsi/megaraid/megaraid_mbox.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/megaraid/megaraid_mbox.c 2011-11-17 20:08:53.000000000 -0500 @@ -3503,6 +3503,8 @@ megaraid_cmm_register(adapter_t *adapter int rval; int i; + pax_track_stack(); + // Allocate memory for the base list of scb for management module. adapter->uscb_list = kcalloc(MBOX_MAX_USER_CMDS, sizeof(scb_t), GFP_KERNEL); diff -urNp linux-2.6.32.48/drivers/scsi/osd/osd_initiator.c linux-2.6.32.48-openvz/drivers/scsi/osd/osd_initiator.c --- linux-2.6.32.48/drivers/scsi/osd/osd_initiator.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/osd/osd_initiator.c 2011-11-17 20:08:53.000000000 -0500 @@ -94,6 +94,8 @@ static int _osd_print_system_info(struct int nelem = ARRAY_SIZE(get_attrs), a = 0; int ret; + pax_track_stack(); + or = osd_start_request(od, GFP_KERNEL); if (!or) return -ENOMEM; diff -urNp linux-2.6.32.48/drivers/scsi/pmcraid.c linux-2.6.32.48-openvz/drivers/scsi/pmcraid.c --- linux-2.6.32.48/drivers/scsi/pmcraid.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/pmcraid.c 2011-11-17 20:09:04.000000000 -0500 @@ -189,8 +189,8 @@ static int pmcraid_slave_alloc(struct sc res->scsi_dev = scsi_dev; scsi_dev->hostdata = res; res->change_detected = 0; - atomic_set(&res->read_failures, 0); - atomic_set(&res->write_failures, 0); + atomic_set_unchecked(&res->read_failures, 0); + atomic_set_unchecked(&res->write_failures, 0); rc = 0; } spin_unlock_irqrestore(&pinstance->resource_lock, lock_flags); @@ -2396,9 +2396,9 @@ static int pmcraid_error_handler(struct /* If this was a SCSI read/write command keep count of errors */ if (SCSI_CMD_TYPE(scsi_cmd->cmnd[0]) == SCSI_READ_CMD) - atomic_inc(&res->read_failures); + atomic_inc_unchecked(&res->read_failures); else if (SCSI_CMD_TYPE(scsi_cmd->cmnd[0]) == SCSI_WRITE_CMD) - atomic_inc(&res->write_failures); + atomic_inc_unchecked(&res->write_failures); if (!RES_IS_GSCSI(res->cfg_entry) && masked_ioasc != PMCRAID_IOASC_HW_DEVICE_BUS_STATUS_ERROR) { @@ -4116,7 +4116,7 @@ static void pmcraid_worker_function(stru pinstance = container_of(workp, struct pmcraid_instance, worker_q); /* add resources only after host is added into system */ - if (!atomic_read(&pinstance->expose_resources)) + if (!atomic_read_unchecked(&pinstance->expose_resources)) return; spin_lock_irqsave(&pinstance->resource_lock, lock_flags); @@ -4850,7 +4850,7 @@ static int __devinit pmcraid_init_instan init_waitqueue_head(&pinstance->reset_wait_q); atomic_set(&pinstance->outstanding_cmds, 0); - atomic_set(&pinstance->expose_resources, 0); + atomic_set_unchecked(&pinstance->expose_resources, 0); INIT_LIST_HEAD(&pinstance->free_res_q); INIT_LIST_HEAD(&pinstance->used_res_q); @@ -5502,7 +5502,7 @@ static int __devinit pmcraid_probe( /* Schedule worker thread to handle CCN and take care of adding and * removing devices to OS */ - atomic_set(&pinstance->expose_resources, 1); + atomic_set_unchecked(&pinstance->expose_resources, 1); schedule_work(&pinstance->worker_q); return rc; diff -urNp linux-2.6.32.48/drivers/scsi/pmcraid.h linux-2.6.32.48-openvz/drivers/scsi/pmcraid.h --- linux-2.6.32.48/drivers/scsi/pmcraid.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/pmcraid.h 2011-11-17 20:08:53.000000000 -0500 @@ -690,7 +690,7 @@ struct pmcraid_instance { atomic_t outstanding_cmds; /* should add/delete resources to mid-layer now ?*/ - atomic_t expose_resources; + atomic_unchecked_t expose_resources; /* Tasklet to handle deferred processing */ struct tasklet_struct isr_tasklet[PMCRAID_NUM_MSIX_VECTORS]; @@ -727,8 +727,8 @@ struct pmcraid_resource_entry { struct list_head queue; /* link to "to be exposed" resources */ struct pmcraid_config_table_entry cfg_entry; struct scsi_device *scsi_dev; /* Link scsi_device structure */ - atomic_t read_failures; /* count of failed READ commands */ - atomic_t write_failures; /* count of failed WRITE commands */ + atomic_unchecked_t read_failures; /* count of failed READ commands */ + atomic_unchecked_t write_failures; /* count of failed WRITE commands */ /* To indicate add/delete/modify during CCN */ u8 change_detected; diff -urNp linux-2.6.32.48/drivers/scsi/qla2xxx/qla_def.h linux-2.6.32.48-openvz/drivers/scsi/qla2xxx/qla_def.h --- linux-2.6.32.48/drivers/scsi/qla2xxx/qla_def.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/qla2xxx/qla_def.h 2011-11-17 20:09:04.000000000 -0500 @@ -2089,7 +2089,7 @@ struct isp_operations { int (*get_flash_version) (struct scsi_qla_host *, void *); int (*start_scsi) (srb_t *); -}; +} __no_const; /* MSI-X Support *************************************************************/ diff -urNp linux-2.6.32.48/drivers/scsi/qla4xxx/ql4_def.h linux-2.6.32.48-openvz/drivers/scsi/qla4xxx/ql4_def.h --- linux-2.6.32.48/drivers/scsi/qla4xxx/ql4_def.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/qla4xxx/ql4_def.h 2011-11-17 20:08:53.000000000 -0500 @@ -240,7 +240,7 @@ struct ddb_entry { atomic_t retry_relogin_timer; /* Min Time between relogins * (4000 only) */ atomic_t relogin_timer; /* Max Time to wait for relogin to complete */ - atomic_t relogin_retry_count; /* Num of times relogin has been + atomic_unchecked_t relogin_retry_count; /* Num of times relogin has been * retried */ uint16_t port; diff -urNp linux-2.6.32.48/drivers/scsi/qla4xxx/ql4_init.c linux-2.6.32.48-openvz/drivers/scsi/qla4xxx/ql4_init.c --- linux-2.6.32.48/drivers/scsi/qla4xxx/ql4_init.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/qla4xxx/ql4_init.c 2011-11-17 20:08:53.000000000 -0500 @@ -482,7 +482,7 @@ static struct ddb_entry * qla4xxx_alloc_ atomic_set(&ddb_entry->port_down_timer, ha->port_down_retry_count); atomic_set(&ddb_entry->retry_relogin_timer, INVALID_ENTRY); atomic_set(&ddb_entry->relogin_timer, 0); - atomic_set(&ddb_entry->relogin_retry_count, 0); + atomic_set_unchecked(&ddb_entry->relogin_retry_count, 0); atomic_set(&ddb_entry->state, DDB_STATE_ONLINE); list_add_tail(&ddb_entry->list, &ha->ddb_list); ha->fw_ddb_index_map[fw_ddb_index] = ddb_entry; @@ -1308,7 +1308,7 @@ int qla4xxx_process_ddb_changed(struct s atomic_set(&ddb_entry->state, DDB_STATE_ONLINE); atomic_set(&ddb_entry->port_down_timer, ha->port_down_retry_count); - atomic_set(&ddb_entry->relogin_retry_count, 0); + atomic_set_unchecked(&ddb_entry->relogin_retry_count, 0); atomic_set(&ddb_entry->relogin_timer, 0); clear_bit(DF_RELOGIN, &ddb_entry->flags); clear_bit(DF_NO_RELOGIN, &ddb_entry->flags); diff -urNp linux-2.6.32.48/drivers/scsi/qla4xxx/ql4_os.c linux-2.6.32.48-openvz/drivers/scsi/qla4xxx/ql4_os.c --- linux-2.6.32.48/drivers/scsi/qla4xxx/ql4_os.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/qla4xxx/ql4_os.c 2011-11-17 20:08:53.000000000 -0500 @@ -641,13 +641,13 @@ static void qla4xxx_timer(struct scsi_ql ddb_entry->fw_ddb_device_state == DDB_DS_SESSION_FAILED) { /* Reset retry relogin timer */ - atomic_inc(&ddb_entry->relogin_retry_count); + atomic_inc_unchecked(&ddb_entry->relogin_retry_count); DEBUG2(printk("scsi%ld: index[%d] relogin" " timed out-retrying" " relogin (%d)\n", ha->host_no, ddb_entry->fw_ddb_index, - atomic_read(&ddb_entry-> + atomic_read_unchecked(&ddb_entry-> relogin_retry_count)) ); start_dpc++; diff -urNp linux-2.6.32.48/drivers/scsi/scsi.c linux-2.6.32.48-openvz/drivers/scsi/scsi.c --- linux-2.6.32.48/drivers/scsi/scsi.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/scsi.c 2011-11-17 20:08:53.000000000 -0500 @@ -652,7 +652,7 @@ int scsi_dispatch_cmd(struct scsi_cmnd * unsigned long timeout; int rtn = 0; - atomic_inc(&cmd->device->iorequest_cnt); + atomic_inc_unchecked(&cmd->device->iorequest_cnt); /* check if the device is still usable */ if (unlikely(cmd->device->sdev_state == SDEV_DEL)) { diff -urNp linux-2.6.32.48/drivers/scsi/scsi_debug.c linux-2.6.32.48-openvz/drivers/scsi/scsi_debug.c --- linux-2.6.32.48/drivers/scsi/scsi_debug.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/scsi_debug.c 2011-11-17 20:08:53.000000000 -0500 @@ -1395,6 +1395,8 @@ static int resp_mode_select(struct scsi_ unsigned char arr[SDEBUG_MAX_MSELECT_SZ]; unsigned char *cmd = (unsigned char *)scp->cmnd; + pax_track_stack(); + if ((errsts = check_readiness(scp, 1, devip))) return errsts; memset(arr, 0, sizeof(arr)); @@ -1492,6 +1494,8 @@ static int resp_log_sense(struct scsi_cm unsigned char arr[SDEBUG_MAX_LSENSE_SZ]; unsigned char *cmd = (unsigned char *)scp->cmnd; + pax_track_stack(); + if ((errsts = check_readiness(scp, 1, devip))) return errsts; memset(arr, 0, sizeof(arr)); diff -urNp linux-2.6.32.48/drivers/scsi/scsi_lib.c linux-2.6.32.48-openvz/drivers/scsi/scsi_lib.c --- linux-2.6.32.48/drivers/scsi/scsi_lib.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/scsi_lib.c 2011-11-17 20:08:53.000000000 -0500 @@ -1384,7 +1384,7 @@ static void scsi_kill_request(struct req scsi_init_cmd_errh(cmd); cmd->result = DID_NO_CONNECT << 16; - atomic_inc(&cmd->device->iorequest_cnt); + atomic_inc_unchecked(&cmd->device->iorequest_cnt); /* * SCSI request completion path will do scsi_device_unbusy(), @@ -1415,9 +1415,9 @@ static void scsi_softirq_done(struct req */ cmd->serial_number = 0; - atomic_inc(&cmd->device->iodone_cnt); + atomic_inc_unchecked(&cmd->device->iodone_cnt); if (cmd->result) - atomic_inc(&cmd->device->ioerr_cnt); + atomic_inc_unchecked(&cmd->device->ioerr_cnt); disposition = scsi_decide_disposition(cmd); if (disposition != SUCCESS && diff -urNp linux-2.6.32.48/drivers/scsi/scsi_scan.c linux-2.6.32.48-openvz/drivers/scsi/scsi_scan.c --- linux-2.6.32.48/drivers/scsi/scsi_scan.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/scsi_scan.c 2011-11-17 20:09:04.000000000 -0500 @@ -415,9 +415,8 @@ static struct scsi_target *scsi_alloc_ta starget->reap_ref = 1; dev->parent = get_device(parent); dev_set_name(dev, "target%d:%d:%d", shost->host_no, channel, id); -#ifndef CONFIG_SYSFS_DEPRECATED - dev->bus = &scsi_bus_type; -#endif + if (!sysfs_deprecated) + dev->bus = &scsi_bus_type; dev->type = &scsi_target_type; starget->id = id; starget->channel = channel; diff -urNp linux-2.6.32.48/drivers/scsi/scsi_sysfs.c linux-2.6.32.48-openvz/drivers/scsi/scsi_sysfs.c --- linux-2.6.32.48/drivers/scsi/scsi_sysfs.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/scsi_sysfs.c 2011-11-17 20:09:04.000000000 -0500 @@ -662,7 +662,7 @@ show_iostat_##field(struct device *dev, char *buf) \ { \ struct scsi_device *sdev = to_scsi_device(dev); \ - unsigned long long count = atomic_read(&sdev->field); \ + unsigned long long count = atomic_read_unchecked(&sdev->field); \ return snprintf(buf, 20, "0x%llx\n", count); \ } \ static DEVICE_ATTR(field, S_IRUGO, show_iostat_##field, NULL) diff -urNp linux-2.6.32.48/drivers/scsi/scsi_tgt_lib.c linux-2.6.32.48-openvz/drivers/scsi/scsi_tgt_lib.c --- linux-2.6.32.48/drivers/scsi/scsi_tgt_lib.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/scsi_tgt_lib.c 2011-11-17 20:09:04.000000000 -0500 @@ -362,7 +362,7 @@ static int scsi_map_user_pages(struct sc int err; dprintk("%lx %u\n", uaddr, len); - err = blk_rq_map_user(q, rq, NULL, (void *)uaddr, len, GFP_KERNEL); + err = blk_rq_map_user(q, rq, NULL, (void __user *)uaddr, len, GFP_KERNEL); if (err) { /* * TODO: need to fixup sg_tablesize, max_segment_size, diff -urNp linux-2.6.32.48/drivers/scsi/scsi_transport_fc.c linux-2.6.32.48-openvz/drivers/scsi/scsi_transport_fc.c --- linux-2.6.32.48/drivers/scsi/scsi_transport_fc.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/scsi_transport_fc.c 2011-11-17 20:08:53.000000000 -0500 @@ -480,7 +480,7 @@ MODULE_PARM_DESC(dev_loss_tmo, * Netlink Infrastructure */ -static atomic_t fc_event_seq; +static atomic_unchecked_t fc_event_seq; /** * fc_get_event_number - Obtain the next sequential FC event number @@ -493,7 +493,7 @@ static atomic_t fc_event_seq; u32 fc_get_event_number(void) { - return atomic_add_return(1, &fc_event_seq); + return atomic_add_return_unchecked(1, &fc_event_seq); } EXPORT_SYMBOL(fc_get_event_number); @@ -641,7 +641,7 @@ static __init int fc_transport_init(void { int error; - atomic_set(&fc_event_seq, 0); + atomic_set_unchecked(&fc_event_seq, 0); error = transport_class_register(&fc_host_class); if (error) diff -urNp linux-2.6.32.48/drivers/scsi/scsi_transport_iscsi.c linux-2.6.32.48-openvz/drivers/scsi/scsi_transport_iscsi.c --- linux-2.6.32.48/drivers/scsi/scsi_transport_iscsi.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/scsi_transport_iscsi.c 2011-11-17 20:08:53.000000000 -0500 @@ -81,7 +81,7 @@ struct iscsi_internal { struct device_attribute *session_attrs[ISCSI_SESSION_ATTRS + 1]; }; -static atomic_t iscsi_session_nr; /* sysfs session id for next new session */ +static atomic_unchecked_t iscsi_session_nr; /* sysfs session id for next new session */ static struct workqueue_struct *iscsi_eh_timer_workq; /* @@ -728,7 +728,7 @@ int iscsi_add_session(struct iscsi_cls_s int err; ihost = shost->shost_data; - session->sid = atomic_add_return(1, &iscsi_session_nr); + session->sid = atomic_add_return_unchecked(1, &iscsi_session_nr); if (id == ISCSI_MAX_TARGET) { for (id = 0; id < ISCSI_MAX_TARGET; id++) { @@ -2060,7 +2060,7 @@ static __init int iscsi_transport_init(v printk(KERN_INFO "Loading iSCSI transport class v%s.\n", ISCSI_TRANSPORT_VERSION); - atomic_set(&iscsi_session_nr, 0); + atomic_set_unchecked(&iscsi_session_nr, 0); err = class_register(&iscsi_transport_class); if (err) diff -urNp linux-2.6.32.48/drivers/scsi/scsi_transport_srp.c linux-2.6.32.48-openvz/drivers/scsi/scsi_transport_srp.c --- linux-2.6.32.48/drivers/scsi/scsi_transport_srp.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/scsi_transport_srp.c 2011-11-17 20:08:53.000000000 -0500 @@ -33,7 +33,7 @@ #include "scsi_transport_srp_internal.h" struct srp_host_attrs { - atomic_t next_port_id; + atomic_unchecked_t next_port_id; }; #define to_srp_host_attrs(host) ((struct srp_host_attrs *)(host)->shost_data) @@ -62,7 +62,7 @@ static int srp_host_setup(struct transpo struct Scsi_Host *shost = dev_to_shost(dev); struct srp_host_attrs *srp_host = to_srp_host_attrs(shost); - atomic_set(&srp_host->next_port_id, 0); + atomic_set_unchecked(&srp_host->next_port_id, 0); return 0; } @@ -211,7 +211,7 @@ struct srp_rport *srp_rport_add(struct S memcpy(rport->port_id, ids->port_id, sizeof(rport->port_id)); rport->roles = ids->roles; - id = atomic_inc_return(&to_srp_host_attrs(shost)->next_port_id); + id = atomic_inc_return_unchecked(&to_srp_host_attrs(shost)->next_port_id); dev_set_name(&rport->dev, "port-%d:%d", shost->host_no, id); transport_setup_device(&rport->dev); diff -urNp linux-2.6.32.48/drivers/scsi/sg.c linux-2.6.32.48-openvz/drivers/scsi/sg.c --- linux-2.6.32.48/drivers/scsi/sg.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/sg.c 2011-11-17 20:09:04.000000000 -0500 @@ -1064,7 +1064,7 @@ sg_ioctl(struct inode *inode, struct fil sdp->disk->disk_name, MKDEV(SCSI_GENERIC_MAJOR, sdp->index), NULL, - (char *)arg); + (char __user *)arg); case BLKTRACESTART: return blk_trace_startstop(sdp->device->request_queue, 1); case BLKTRACESTOP: @@ -2292,7 +2292,7 @@ struct sg_proc_leaf { const struct file_operations * fops; }; -static struct sg_proc_leaf sg_proc_leaf_arr[] = { +static const struct sg_proc_leaf sg_proc_leaf_arr[] = { {"allow_dio", &adio_fops}, {"debug", &debug_fops}, {"def_reserved_size", &dressz_fops}, @@ -2307,7 +2307,7 @@ sg_proc_init(void) { int k, mask; int num_leaves = ARRAY_SIZE(sg_proc_leaf_arr); - struct sg_proc_leaf * leaf; + const struct sg_proc_leaf * leaf; sg_proc_sgp = proc_mkdir(sg_proc_sg_dirname, NULL); if (!sg_proc_sgp) diff -urNp linux-2.6.32.48/drivers/scsi/sym53c8xx_2/sym_glue.c linux-2.6.32.48-openvz/drivers/scsi/sym53c8xx_2/sym_glue.c --- linux-2.6.32.48/drivers/scsi/sym53c8xx_2/sym_glue.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/scsi/sym53c8xx_2/sym_glue.c 2011-11-17 20:08:53.000000000 -0500 @@ -1754,6 +1754,8 @@ static int __devinit sym2_probe(struct p int do_iounmap = 0; int do_disable_device = 1; + pax_track_stack(); + memset(&sym_dev, 0, sizeof(sym_dev)); memset(&nvram, 0, sizeof(nvram)); sym_dev.pdev = pdev; diff -urNp linux-2.6.32.48/drivers/serial/kgdboc.c linux-2.6.32.48-openvz/drivers/serial/kgdboc.c --- linux-2.6.32.48/drivers/serial/kgdboc.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/serial/kgdboc.c 2011-11-17 20:08:53.000000000 -0500 @@ -18,7 +18,7 @@ #define MAX_CONFIG_LEN 40 -static struct kgdb_io kgdboc_io_ops; +static const struct kgdb_io kgdboc_io_ops; /* -1 = init not run yet, 0 = unconfigured, 1 = configured. */ static int configured = -1; @@ -154,7 +154,7 @@ static void kgdboc_post_exp_handler(void module_put(THIS_MODULE); } -static struct kgdb_io kgdboc_io_ops = { +static const struct kgdb_io kgdboc_io_ops = { .name = "kgdboc", .read_char = kgdboc_get_char, .write_char = kgdboc_put_char, diff -urNp linux-2.6.32.48/drivers/spi/spi.c linux-2.6.32.48-openvz/drivers/spi/spi.c --- linux-2.6.32.48/drivers/spi/spi.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/spi/spi.c 2011-11-17 20:08:53.000000000 -0500 @@ -774,7 +774,7 @@ int spi_sync(struct spi_device *spi, str EXPORT_SYMBOL_GPL(spi_sync); /* portable code must never pass more than 32 bytes */ -#define SPI_BUFSIZ max(32,SMP_CACHE_BYTES) +#define SPI_BUFSIZ max(32U,SMP_CACHE_BYTES) static u8 *buf; diff -urNp linux-2.6.32.48/drivers/staging/android/binder.c linux-2.6.32.48-openvz/drivers/staging/android/binder.c --- linux-2.6.32.48/drivers/staging/android/binder.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/android/binder.c 2011-11-17 20:08:53.000000000 -0500 @@ -2756,7 +2756,7 @@ static void binder_vma_close(struct vm_a binder_defer_work(proc, BINDER_DEFERRED_PUT_FILES); } -static struct vm_operations_struct binder_vm_ops = { +static const struct vm_operations_struct binder_vm_ops = { .open = binder_vma_open, .close = binder_vma_close, }; diff -urNp linux-2.6.32.48/drivers/staging/b3dfg/b3dfg.c linux-2.6.32.48-openvz/drivers/staging/b3dfg/b3dfg.c --- linux-2.6.32.48/drivers/staging/b3dfg/b3dfg.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/b3dfg/b3dfg.c 2011-11-17 20:08:53.000000000 -0500 @@ -455,7 +455,7 @@ static int b3dfg_vma_fault(struct vm_are return VM_FAULT_NOPAGE; } -static struct vm_operations_struct b3dfg_vm_ops = { +static const struct vm_operations_struct b3dfg_vm_ops = { .fault = b3dfg_vma_fault, }; @@ -848,7 +848,7 @@ static int b3dfg_mmap(struct file *filp, return r; } -static struct file_operations b3dfg_fops = { +static const struct file_operations b3dfg_fops = { .owner = THIS_MODULE, .open = b3dfg_open, .release = b3dfg_release, diff -urNp linux-2.6.32.48/drivers/staging/comedi/comedi_fops.c linux-2.6.32.48-openvz/drivers/staging/comedi/comedi_fops.c --- linux-2.6.32.48/drivers/staging/comedi/comedi_fops.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/comedi/comedi_fops.c 2011-11-17 20:09:04.000000000 -0500 @@ -1389,7 +1389,7 @@ void comedi_unmap(struct vm_area_struct mutex_unlock(&dev->mutex); } -static struct vm_operations_struct comedi_vm_ops = { +static const struct vm_operations_struct comedi_vm_ops = { .close = comedi_unmap, }; diff -urNp linux-2.6.32.48/drivers/staging/dream/qdsp5/adsp_driver.c linux-2.6.32.48-openvz/drivers/staging/dream/qdsp5/adsp_driver.c --- linux-2.6.32.48/drivers/staging/dream/qdsp5/adsp_driver.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/dream/qdsp5/adsp_driver.c 2011-11-17 20:08:53.000000000 -0500 @@ -576,7 +576,7 @@ static struct adsp_device *inode_to_devi static dev_t adsp_devno; static struct class *adsp_class; -static struct file_operations adsp_fops = { +static const struct file_operations adsp_fops = { .owner = THIS_MODULE, .open = adsp_open, .unlocked_ioctl = adsp_ioctl, diff -urNp linux-2.6.32.48/drivers/staging/dream/qdsp5/audio_aac.c linux-2.6.32.48-openvz/drivers/staging/dream/qdsp5/audio_aac.c --- linux-2.6.32.48/drivers/staging/dream/qdsp5/audio_aac.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/dream/qdsp5/audio_aac.c 2011-11-17 20:08:53.000000000 -0500 @@ -1022,7 +1022,7 @@ done: return rc; } -static struct file_operations audio_aac_fops = { +static const struct file_operations audio_aac_fops = { .owner = THIS_MODULE, .open = audio_open, .release = audio_release, diff -urNp linux-2.6.32.48/drivers/staging/dream/qdsp5/audio_amrnb.c linux-2.6.32.48-openvz/drivers/staging/dream/qdsp5/audio_amrnb.c --- linux-2.6.32.48/drivers/staging/dream/qdsp5/audio_amrnb.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/dream/qdsp5/audio_amrnb.c 2011-11-17 20:08:53.000000000 -0500 @@ -833,7 +833,7 @@ done: return rc; } -static struct file_operations audio_amrnb_fops = { +static const struct file_operations audio_amrnb_fops = { .owner = THIS_MODULE, .open = audamrnb_open, .release = audamrnb_release, diff -urNp linux-2.6.32.48/drivers/staging/dream/qdsp5/audio_evrc.c linux-2.6.32.48-openvz/drivers/staging/dream/qdsp5/audio_evrc.c --- linux-2.6.32.48/drivers/staging/dream/qdsp5/audio_evrc.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/dream/qdsp5/audio_evrc.c 2011-11-17 20:08:53.000000000 -0500 @@ -805,7 +805,7 @@ dma_fail: return rc; } -static struct file_operations audio_evrc_fops = { +static const struct file_operations audio_evrc_fops = { .owner = THIS_MODULE, .open = audevrc_open, .release = audevrc_release, diff -urNp linux-2.6.32.48/drivers/staging/dream/qdsp5/audio_in.c linux-2.6.32.48-openvz/drivers/staging/dream/qdsp5/audio_in.c --- linux-2.6.32.48/drivers/staging/dream/qdsp5/audio_in.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/dream/qdsp5/audio_in.c 2011-11-17 20:08:53.000000000 -0500 @@ -913,7 +913,7 @@ static int audpre_open(struct inode *ino return 0; } -static struct file_operations audio_fops = { +static const struct file_operations audio_fops = { .owner = THIS_MODULE, .open = audio_in_open, .release = audio_in_release, @@ -922,7 +922,7 @@ static struct file_operations audio_fops .unlocked_ioctl = audio_in_ioctl, }; -static struct file_operations audpre_fops = { +static const struct file_operations audpre_fops = { .owner = THIS_MODULE, .open = audpre_open, .unlocked_ioctl = audpre_ioctl, diff -urNp linux-2.6.32.48/drivers/staging/dream/qdsp5/audio_mp3.c linux-2.6.32.48-openvz/drivers/staging/dream/qdsp5/audio_mp3.c --- linux-2.6.32.48/drivers/staging/dream/qdsp5/audio_mp3.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/dream/qdsp5/audio_mp3.c 2011-11-17 20:08:53.000000000 -0500 @@ -941,7 +941,7 @@ done: return rc; } -static struct file_operations audio_mp3_fops = { +static const struct file_operations audio_mp3_fops = { .owner = THIS_MODULE, .open = audio_open, .release = audio_release, diff -urNp linux-2.6.32.48/drivers/staging/dream/qdsp5/audio_out.c linux-2.6.32.48-openvz/drivers/staging/dream/qdsp5/audio_out.c --- linux-2.6.32.48/drivers/staging/dream/qdsp5/audio_out.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/dream/qdsp5/audio_out.c 2011-11-17 20:08:53.000000000 -0500 @@ -810,7 +810,7 @@ static int audpp_open(struct inode *inod return 0; } -static struct file_operations audio_fops = { +static const struct file_operations audio_fops = { .owner = THIS_MODULE, .open = audio_open, .release = audio_release, @@ -819,7 +819,7 @@ static struct file_operations audio_fops .unlocked_ioctl = audio_ioctl, }; -static struct file_operations audpp_fops = { +static const struct file_operations audpp_fops = { .owner = THIS_MODULE, .open = audpp_open, .unlocked_ioctl = audpp_ioctl, diff -urNp linux-2.6.32.48/drivers/staging/dream/qdsp5/audio_qcelp.c linux-2.6.32.48-openvz/drivers/staging/dream/qdsp5/audio_qcelp.c --- linux-2.6.32.48/drivers/staging/dream/qdsp5/audio_qcelp.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/dream/qdsp5/audio_qcelp.c 2011-11-17 20:08:53.000000000 -0500 @@ -816,7 +816,7 @@ err: return rc; } -static struct file_operations audio_qcelp_fops = { +static const struct file_operations audio_qcelp_fops = { .owner = THIS_MODULE, .open = audqcelp_open, .release = audqcelp_release, diff -urNp linux-2.6.32.48/drivers/staging/dream/qdsp5/snd.c linux-2.6.32.48-openvz/drivers/staging/dream/qdsp5/snd.c --- linux-2.6.32.48/drivers/staging/dream/qdsp5/snd.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/dream/qdsp5/snd.c 2011-11-17 20:08:53.000000000 -0500 @@ -242,7 +242,7 @@ err: return rc; } -static struct file_operations snd_fops = { +static const struct file_operations snd_fops = { .owner = THIS_MODULE, .open = snd_open, .release = snd_release, diff -urNp linux-2.6.32.48/drivers/staging/dream/smd/smd_qmi.c linux-2.6.32.48-openvz/drivers/staging/dream/smd/smd_qmi.c --- linux-2.6.32.48/drivers/staging/dream/smd/smd_qmi.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/dream/smd/smd_qmi.c 2011-11-17 20:08:53.000000000 -0500 @@ -793,7 +793,7 @@ static int qmi_release(struct inode *ip, return 0; } -static struct file_operations qmi_fops = { +static const struct file_operations qmi_fops = { .owner = THIS_MODULE, .read = qmi_read, .write = qmi_write, diff -urNp linux-2.6.32.48/drivers/staging/dream/smd/smd_rpcrouter_device.c linux-2.6.32.48-openvz/drivers/staging/dream/smd/smd_rpcrouter_device.c --- linux-2.6.32.48/drivers/staging/dream/smd/smd_rpcrouter_device.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/dream/smd/smd_rpcrouter_device.c 2011-11-17 20:08:53.000000000 -0500 @@ -214,7 +214,7 @@ static long rpcrouter_ioctl(struct file return rc; } -static struct file_operations rpcrouter_server_fops = { +static const struct file_operations rpcrouter_server_fops = { .owner = THIS_MODULE, .open = rpcrouter_open, .release = rpcrouter_release, @@ -224,7 +224,7 @@ static struct file_operations rpcrouter_ .unlocked_ioctl = rpcrouter_ioctl, }; -static struct file_operations rpcrouter_router_fops = { +static const struct file_operations rpcrouter_router_fops = { .owner = THIS_MODULE, .open = rpcrouter_open, .release = rpcrouter_release, diff -urNp linux-2.6.32.48/drivers/staging/dst/dcore.c linux-2.6.32.48-openvz/drivers/staging/dst/dcore.c --- linux-2.6.32.48/drivers/staging/dst/dcore.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/dst/dcore.c 2011-11-17 20:08:53.000000000 -0500 @@ -149,7 +149,7 @@ static int dst_bdev_release(struct gendi return 0; } -static struct block_device_operations dst_blk_ops = { +static const struct block_device_operations dst_blk_ops = { .open = dst_bdev_open, .release = dst_bdev_release, .owner = THIS_MODULE, @@ -588,7 +588,7 @@ static struct dst_node *dst_alloc_node(s n->size = ctl->size; atomic_set(&n->refcnt, 1); - atomic_long_set(&n->gen, 0); + atomic_long_set_unchecked(&n->gen, 0); snprintf(n->name, sizeof(n->name), "%s", ctl->name); err = dst_node_sysfs_init(n); diff -urNp linux-2.6.32.48/drivers/staging/dst/trans.c linux-2.6.32.48-openvz/drivers/staging/dst/trans.c --- linux-2.6.32.48/drivers/staging/dst/trans.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/dst/trans.c 2011-11-17 20:08:53.000000000 -0500 @@ -169,7 +169,7 @@ int dst_process_bio(struct dst_node *n, t->error = 0; t->retries = 0; atomic_set(&t->refcnt, 1); - t->gen = atomic_long_inc_return(&n->gen); + t->gen = atomic_long_inc_return_unchecked(&n->gen); t->enc = bio_data_dir(bio); dst_bio_to_cmd(bio, &t->cmd, DST_IO, t->gen); diff -urNp linux-2.6.32.48/drivers/staging/et131x/et1310_tx.c linux-2.6.32.48-openvz/drivers/staging/et131x/et1310_tx.c --- linux-2.6.32.48/drivers/staging/et131x/et1310_tx.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/et131x/et1310_tx.c 2011-11-17 20:08:53.000000000 -0500 @@ -710,11 +710,11 @@ inline void et131x_free_send_packet(stru struct net_device_stats *stats = &etdev->net_stats; if (pMpTcb->Flags & fMP_DEST_BROAD) - atomic_inc(&etdev->Stats.brdcstxmt); + atomic_inc_unchecked(&etdev->Stats.brdcstxmt); else if (pMpTcb->Flags & fMP_DEST_MULTI) - atomic_inc(&etdev->Stats.multixmt); + atomic_inc_unchecked(&etdev->Stats.multixmt); else - atomic_inc(&etdev->Stats.unixmt); + atomic_inc_unchecked(&etdev->Stats.unixmt); if (pMpTcb->Packet) { stats->tx_bytes += pMpTcb->Packet->len; diff -urNp linux-2.6.32.48/drivers/staging/et131x/et131x_adapter.h linux-2.6.32.48-openvz/drivers/staging/et131x/et131x_adapter.h --- linux-2.6.32.48/drivers/staging/et131x/et131x_adapter.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/et131x/et131x_adapter.h 2011-11-17 20:08:53.000000000 -0500 @@ -145,11 +145,11 @@ typedef struct _ce_stats_t { * operations */ u32 unircv; /* # multicast packets received */ - atomic_t unixmt; /* # multicast packets for Tx */ + atomic_unchecked_t unixmt; /* # multicast packets for Tx */ u32 multircv; /* # multicast packets received */ - atomic_t multixmt; /* # multicast packets for Tx */ + atomic_unchecked_t multixmt; /* # multicast packets for Tx */ u32 brdcstrcv; /* # broadcast packets received */ - atomic_t brdcstxmt; /* # broadcast packets for Tx */ + atomic_unchecked_t brdcstxmt; /* # broadcast packets for Tx */ u32 norcvbuf; /* # Rx packets discarded */ u32 noxmtbuf; /* # Tx packets discarded */ diff -urNp linux-2.6.32.48/drivers/staging/go7007/go7007-v4l2.c linux-2.6.32.48-openvz/drivers/staging/go7007/go7007-v4l2.c --- linux-2.6.32.48/drivers/staging/go7007/go7007-v4l2.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/go7007/go7007-v4l2.c 2011-11-17 20:08:53.000000000 -0500 @@ -1700,7 +1700,7 @@ static int go7007_vm_fault(struct vm_are return 0; } -static struct vm_operations_struct go7007_vm_ops = { +static const struct vm_operations_struct go7007_vm_ops = { .open = go7007_vm_open, .close = go7007_vm_close, .fault = go7007_vm_fault, diff -urNp linux-2.6.32.48/drivers/staging/hv/blkvsc_drv.c linux-2.6.32.48-openvz/drivers/staging/hv/blkvsc_drv.c --- linux-2.6.32.48/drivers/staging/hv/blkvsc_drv.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/hv/blkvsc_drv.c 2011-11-17 20:08:53.000000000 -0500 @@ -153,7 +153,7 @@ static int blkvsc_ringbuffer_size = BLKV /* The one and only one */ static struct blkvsc_driver_context g_blkvsc_drv; -static struct block_device_operations block_ops = { +static const struct block_device_operations block_ops = { .owner = THIS_MODULE, .open = blkvsc_open, .release = blkvsc_release, diff -urNp linux-2.6.32.48/drivers/staging/hv/Channel.c linux-2.6.32.48-openvz/drivers/staging/hv/Channel.c --- linux-2.6.32.48/drivers/staging/hv/Channel.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/hv/Channel.c 2011-11-17 20:08:53.000000000 -0500 @@ -464,8 +464,8 @@ int VmbusChannelEstablishGpadl(struct vm DPRINT_ENTER(VMBUS); - nextGpadlHandle = atomic_read(&gVmbusConnection.NextGpadlHandle); - atomic_inc(&gVmbusConnection.NextGpadlHandle); + nextGpadlHandle = atomic_read_unchecked(&gVmbusConnection.NextGpadlHandle); + atomic_inc_unchecked(&gVmbusConnection.NextGpadlHandle); VmbusChannelCreateGpadlHeader(Kbuffer, Size, &msgInfo, &msgCount); ASSERT(msgInfo != NULL); diff -urNp linux-2.6.32.48/drivers/staging/hv/Hv.c linux-2.6.32.48-openvz/drivers/staging/hv/Hv.c --- linux-2.6.32.48/drivers/staging/hv/Hv.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/hv/Hv.c 2011-11-17 20:08:53.000000000 -0500 @@ -161,7 +161,7 @@ static u64 HvDoHypercall(u64 Control, vo u64 outputAddress = (Output) ? virt_to_phys(Output) : 0; u32 outputAddressHi = outputAddress >> 32; u32 outputAddressLo = outputAddress & 0xFFFFFFFF; - volatile void *hypercallPage = gHvContext.HypercallPage; + volatile void *hypercallPage = ktva_ktla(gHvContext.HypercallPage); DPRINT_DBG(VMBUS, "Hypercall ", Control, Input, Output); diff -urNp linux-2.6.32.48/drivers/staging/hv/VmbusApi.h linux-2.6.32.48-openvz/drivers/staging/hv/VmbusApi.h --- linux-2.6.32.48/drivers/staging/hv/VmbusApi.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/hv/VmbusApi.h 2011-11-17 20:09:04.000000000 -0500 @@ -109,7 +109,7 @@ struct vmbus_channel_interface { u32 *GpadlHandle); int (*TeardownGpadl)(struct hv_device *device, u32 GpadlHandle); void (*GetInfo)(struct hv_device *dev, struct hv_device_info *devinfo); -}; +} __no_const; /* Base driver object */ struct hv_driver { diff -urNp linux-2.6.32.48/drivers/staging/hv/vmbus_drv.c linux-2.6.32.48-openvz/drivers/staging/hv/vmbus_drv.c --- linux-2.6.32.48/drivers/staging/hv/vmbus_drv.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/hv/vmbus_drv.c 2011-11-17 20:08:53.000000000 -0500 @@ -532,7 +532,7 @@ static int vmbus_child_device_register(s to_device_context(root_device_obj); struct device_context *child_device_ctx = to_device_context(child_device_obj); - static atomic_t device_num = ATOMIC_INIT(0); + static atomic_unchecked_t device_num = ATOMIC_INIT(0); DPRINT_ENTER(VMBUS_DRV); @@ -541,7 +541,7 @@ static int vmbus_child_device_register(s /* Set the device name. Otherwise, device_register() will fail. */ dev_set_name(&child_device_ctx->device, "vmbus_0_%d", - atomic_inc_return(&device_num)); + atomic_inc_return_unchecked(&device_num)); /* The new device belongs to this bus */ child_device_ctx->device.bus = &g_vmbus_drv.bus; /* device->dev.bus; */ diff -urNp linux-2.6.32.48/drivers/staging/hv/VmbusPrivate.h linux-2.6.32.48-openvz/drivers/staging/hv/VmbusPrivate.h --- linux-2.6.32.48/drivers/staging/hv/VmbusPrivate.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/hv/VmbusPrivate.h 2011-11-17 20:08:53.000000000 -0500 @@ -59,7 +59,7 @@ enum VMBUS_CONNECT_STATE { struct VMBUS_CONNECTION { enum VMBUS_CONNECT_STATE ConnectState; - atomic_t NextGpadlHandle; + atomic_unchecked_t NextGpadlHandle; /* * Represents channel interrupts. Each bit position represents a diff -urNp linux-2.6.32.48/drivers/staging/iio/ring_generic.h linux-2.6.32.48-openvz/drivers/staging/iio/ring_generic.h --- linux-2.6.32.48/drivers/staging/iio/ring_generic.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/iio/ring_generic.h 2011-11-17 20:09:04.000000000 -0500 @@ -87,7 +87,7 @@ struct iio_ring_access_funcs { int (*is_enabled)(struct iio_ring_buffer *ring); int (*enable)(struct iio_ring_buffer *ring); -}; +} __no_const; /** * struct iio_ring_buffer - general ring buffer structure diff -urNp linux-2.6.32.48/drivers/staging/octeon/ethernet.c linux-2.6.32.48-openvz/drivers/staging/octeon/ethernet.c --- linux-2.6.32.48/drivers/staging/octeon/ethernet.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/octeon/ethernet.c 2011-11-17 20:08:53.000000000 -0500 @@ -294,11 +294,11 @@ static struct net_device_stats *cvm_oct_ * since the RX tasklet also increments it. */ #ifdef CONFIG_64BIT - atomic64_add(rx_status.dropped_packets, - (atomic64_t *)&priv->stats.rx_dropped); + atomic64_add_unchecked(rx_status.dropped_packets, + (atomic64_unchecked_t *)&priv->stats.rx_dropped); #else - atomic_add(rx_status.dropped_packets, - (atomic_t *)&priv->stats.rx_dropped); + atomic_add_unchecked(rx_status.dropped_packets, + (atomic_unchecked_t *)&priv->stats.rx_dropped); #endif } diff -urNp linux-2.6.32.48/drivers/staging/octeon/ethernet-rx.c linux-2.6.32.48-openvz/drivers/staging/octeon/ethernet-rx.c --- linux-2.6.32.48/drivers/staging/octeon/ethernet-rx.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/octeon/ethernet-rx.c 2011-11-17 20:08:53.000000000 -0500 @@ -406,11 +406,11 @@ void cvm_oct_tasklet_rx(unsigned long un /* Increment RX stats for virtual ports */ if (work->ipprt >= CVMX_PIP_NUM_INPUT_PORTS) { #ifdef CONFIG_64BIT - atomic64_add(1, (atomic64_t *)&priv->stats.rx_packets); - atomic64_add(skb->len, (atomic64_t *)&priv->stats.rx_bytes); + atomic64_add_unchecked(1, (atomic64_unchecked_t *)&priv->stats.rx_packets); + atomic64_add_unchecked(skb->len, (atomic64_unchecked_t *)&priv->stats.rx_bytes); #else - atomic_add(1, (atomic_t *)&priv->stats.rx_packets); - atomic_add(skb->len, (atomic_t *)&priv->stats.rx_bytes); + atomic_add_unchecked(1, (atomic_unchecked_t *)&priv->stats.rx_packets); + atomic_add_unchecked(skb->len, (atomic_unchecked_t *)&priv->stats.rx_bytes); #endif } netif_receive_skb(skb); @@ -424,9 +424,9 @@ void cvm_oct_tasklet_rx(unsigned long un dev->name); */ #ifdef CONFIG_64BIT - atomic64_add(1, (atomic64_t *)&priv->stats.rx_dropped); + atomic64_add_unchecked(1, (atomic64_t *)&priv->stats.rx_dropped); #else - atomic_add(1, (atomic_t *)&priv->stats.rx_dropped); + atomic_add_unchecked(1, (atomic_t *)&priv->stats.rx_dropped); #endif dev_kfree_skb_irq(skb); } diff -urNp linux-2.6.32.48/drivers/staging/panel/panel.c linux-2.6.32.48-openvz/drivers/staging/panel/panel.c --- linux-2.6.32.48/drivers/staging/panel/panel.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/panel/panel.c 2011-11-17 20:08:53.000000000 -0500 @@ -1305,7 +1305,7 @@ static int lcd_release(struct inode *ino return 0; } -static struct file_operations lcd_fops = { +static const struct file_operations lcd_fops = { .write = lcd_write, .open = lcd_open, .release = lcd_release, @@ -1565,7 +1565,7 @@ static int keypad_release(struct inode * return 0; } -static struct file_operations keypad_fops = { +static const struct file_operations keypad_fops = { .read = keypad_read, /* read */ .open = keypad_open, /* open */ .release = keypad_release, /* close */ diff -urNp linux-2.6.32.48/drivers/staging/phison/phison.c linux-2.6.32.48-openvz/drivers/staging/phison/phison.c --- linux-2.6.32.48/drivers/staging/phison/phison.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/phison/phison.c 2011-11-17 20:08:53.000000000 -0500 @@ -43,7 +43,7 @@ static struct scsi_host_template phison_ ATA_BMDMA_SHT(DRV_NAME), }; -static struct ata_port_operations phison_ops = { +static const struct ata_port_operations phison_ops = { .inherits = &ata_bmdma_port_ops, .prereset = phison_pre_reset, }; diff -urNp linux-2.6.32.48/drivers/staging/poch/poch.c linux-2.6.32.48-openvz/drivers/staging/poch/poch.c --- linux-2.6.32.48/drivers/staging/poch/poch.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/poch/poch.c 2011-11-17 20:08:53.000000000 -0500 @@ -1057,7 +1057,7 @@ static int poch_ioctl(struct inode *inod return 0; } -static struct file_operations poch_fops = { +static const struct file_operations poch_fops = { .owner = THIS_MODULE, .open = poch_open, .release = poch_release, diff -urNp linux-2.6.32.48/drivers/staging/pohmelfs/inode.c linux-2.6.32.48-openvz/drivers/staging/pohmelfs/inode.c --- linux-2.6.32.48/drivers/staging/pohmelfs/inode.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/pohmelfs/inode.c 2011-11-17 20:08:53.000000000 -0500 @@ -1850,7 +1850,7 @@ static int pohmelfs_fill_super(struct su mutex_init(&psb->mcache_lock); psb->mcache_root = RB_ROOT; psb->mcache_timeout = msecs_to_jiffies(5000); - atomic_long_set(&psb->mcache_gen, 0); + atomic_long_set_unchecked(&psb->mcache_gen, 0); psb->trans_max_pages = 100; @@ -1865,7 +1865,7 @@ static int pohmelfs_fill_super(struct su INIT_LIST_HEAD(&psb->crypto_ready_list); INIT_LIST_HEAD(&psb->crypto_active_list); - atomic_set(&psb->trans_gen, 1); + atomic_set_unchecked(&psb->trans_gen, 1); atomic_long_set(&psb->total_inodes, 0); mutex_init(&psb->state_lock); diff -urNp linux-2.6.32.48/drivers/staging/pohmelfs/mcache.c linux-2.6.32.48-openvz/drivers/staging/pohmelfs/mcache.c --- linux-2.6.32.48/drivers/staging/pohmelfs/mcache.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/pohmelfs/mcache.c 2011-11-17 20:08:54.000000000 -0500 @@ -121,7 +121,7 @@ struct pohmelfs_mcache *pohmelfs_mcache_ m->data = data; m->start = start; m->size = size; - m->gen = atomic_long_inc_return(&psb->mcache_gen); + m->gen = atomic_long_inc_return_unchecked(&psb->mcache_gen); mutex_lock(&psb->mcache_lock); err = pohmelfs_mcache_insert(psb, m); diff -urNp linux-2.6.32.48/drivers/staging/pohmelfs/netfs.h linux-2.6.32.48-openvz/drivers/staging/pohmelfs/netfs.h --- linux-2.6.32.48/drivers/staging/pohmelfs/netfs.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/pohmelfs/netfs.h 2011-11-17 20:08:54.000000000 -0500 @@ -570,14 +570,14 @@ struct pohmelfs_config; struct pohmelfs_sb { struct rb_root mcache_root; struct mutex mcache_lock; - atomic_long_t mcache_gen; + atomic_long_unchecked_t mcache_gen; unsigned long mcache_timeout; unsigned int idx; unsigned int trans_retries; - atomic_t trans_gen; + atomic_unchecked_t trans_gen; unsigned int crypto_attached_size; unsigned int crypto_align_size; diff -urNp linux-2.6.32.48/drivers/staging/pohmelfs/trans.c linux-2.6.32.48-openvz/drivers/staging/pohmelfs/trans.c --- linux-2.6.32.48/drivers/staging/pohmelfs/trans.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/pohmelfs/trans.c 2011-11-17 20:08:54.000000000 -0500 @@ -492,7 +492,7 @@ int netfs_trans_finish(struct netfs_tran int err; struct netfs_cmd *cmd = t->iovec.iov_base; - t->gen = atomic_inc_return(&psb->trans_gen); + t->gen = atomic_inc_return_unchecked(&psb->trans_gen); cmd->size = t->iovec.iov_len - sizeof(struct netfs_cmd) + t->attached_size + t->attached_pages * sizeof(struct netfs_cmd); diff -urNp linux-2.6.32.48/drivers/staging/sep/sep_driver.c linux-2.6.32.48-openvz/drivers/staging/sep/sep_driver.c --- linux-2.6.32.48/drivers/staging/sep/sep_driver.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/sep/sep_driver.c 2011-11-17 20:08:54.000000000 -0500 @@ -2603,7 +2603,7 @@ static struct pci_driver sep_pci_driver static dev_t sep_devno; /* the files operations structure of the driver */ -static struct file_operations sep_file_operations = { +static const struct file_operations sep_file_operations = { .owner = THIS_MODULE, .ioctl = sep_ioctl, .poll = sep_poll, diff -urNp linux-2.6.32.48/drivers/staging/usbip/usbip_common.h linux-2.6.32.48-openvz/drivers/staging/usbip/usbip_common.h --- linux-2.6.32.48/drivers/staging/usbip/usbip_common.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/usbip/usbip_common.h 2011-11-17 20:09:04.000000000 -0500 @@ -374,7 +374,7 @@ struct usbip_device { void (*shutdown)(struct usbip_device *); void (*reset)(struct usbip_device *); void (*unusable)(struct usbip_device *); - } eh_ops; + } __no_const eh_ops; }; diff -urNp linux-2.6.32.48/drivers/staging/usbip/vhci.h linux-2.6.32.48-openvz/drivers/staging/usbip/vhci.h --- linux-2.6.32.48/drivers/staging/usbip/vhci.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/usbip/vhci.h 2011-11-17 20:08:54.000000000 -0500 @@ -92,7 +92,7 @@ struct vhci_hcd { unsigned resuming:1; unsigned long re_timeout; - atomic_t seqnum; + atomic_unchecked_t seqnum; /* * NOTE: diff -urNp linux-2.6.32.48/drivers/staging/usbip/vhci_hcd.c linux-2.6.32.48-openvz/drivers/staging/usbip/vhci_hcd.c --- linux-2.6.32.48/drivers/staging/usbip/vhci_hcd.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/usbip/vhci_hcd.c 2011-11-17 20:08:54.000000000 -0500 @@ -534,7 +534,7 @@ static void vhci_tx_urb(struct urb *urb) return; } - priv->seqnum = atomic_inc_return(&the_controller->seqnum); + priv->seqnum = atomic_inc_return_unchecked(&the_controller->seqnum); if (priv->seqnum == 0xffff) usbip_uinfo("seqnum max\n"); @@ -793,7 +793,7 @@ static int vhci_urb_dequeue(struct usb_h return -ENOMEM; } - unlink->seqnum = atomic_inc_return(&the_controller->seqnum); + unlink->seqnum = atomic_inc_return_unchecked(&the_controller->seqnum); if (unlink->seqnum == 0xffff) usbip_uinfo("seqnum max\n"); @@ -988,7 +988,7 @@ static int vhci_start(struct usb_hcd *hc vdev->rhport = rhport; } - atomic_set(&vhci->seqnum, 0); + atomic_set_unchecked(&vhci->seqnum, 0); spin_lock_init(&vhci->lock); diff -urNp linux-2.6.32.48/drivers/staging/usbip/vhci_rx.c linux-2.6.32.48-openvz/drivers/staging/usbip/vhci_rx.c --- linux-2.6.32.48/drivers/staging/usbip/vhci_rx.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/usbip/vhci_rx.c 2011-11-17 20:08:54.000000000 -0500 @@ -78,7 +78,7 @@ static void vhci_recv_ret_submit(struct usbip_uerr("cannot find a urb of seqnum %u\n", pdu->base.seqnum); usbip_uinfo("max seqnum %d\n", - atomic_read(&the_controller->seqnum)); + atomic_read_unchecked(&the_controller->seqnum)); usbip_event_add(ud, VDEV_EVENT_ERROR_TCP); return; } diff -urNp linux-2.6.32.48/drivers/staging/vme/devices/vme_user.c linux-2.6.32.48-openvz/drivers/staging/vme/devices/vme_user.c --- linux-2.6.32.48/drivers/staging/vme/devices/vme_user.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/vme/devices/vme_user.c 2011-11-17 20:08:54.000000000 -0500 @@ -136,7 +136,7 @@ static int vme_user_ioctl(struct inode * static int __init vme_user_probe(struct device *, int, int); static int __exit vme_user_remove(struct device *, int, int); -static struct file_operations vme_user_fops = { +static const struct file_operations vme_user_fops = { .open = vme_user_open, .release = vme_user_release, .read = vme_user_read, diff -urNp linux-2.6.32.48/drivers/staging/vt6655/hostap.c linux-2.6.32.48-openvz/drivers/staging/vt6655/hostap.c --- linux-2.6.32.48/drivers/staging/vt6655/hostap.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/vt6655/hostap.c 2011-11-17 20:09:04.000000000 -0500 @@ -84,7 +84,7 @@ static int hostap_enable_hostapd(PSDevic PSDevice apdev_priv; struct net_device *dev = pDevice->dev; int ret; - const struct net_device_ops apdev_netdev_ops = { + net_device_ops_no_const apdev_netdev_ops = { .ndo_start_xmit = pDevice->tx_80211, }; diff -urNp linux-2.6.32.48/drivers/staging/vt6656/hostap.c linux-2.6.32.48-openvz/drivers/staging/vt6656/hostap.c --- linux-2.6.32.48/drivers/staging/vt6656/hostap.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/vt6656/hostap.c 2011-11-17 20:09:04.000000000 -0500 @@ -86,7 +86,7 @@ static int hostap_enable_hostapd(PSDevic PSDevice apdev_priv; struct net_device *dev = pDevice->dev; int ret; - const struct net_device_ops apdev_netdev_ops = { + net_device_ops_no_const apdev_netdev_ops = { .ndo_start_xmit = pDevice->tx_80211, }; diff -urNp linux-2.6.32.48/drivers/staging/wlan-ng/hfa384x_usb.c linux-2.6.32.48-openvz/drivers/staging/wlan-ng/hfa384x_usb.c --- linux-2.6.32.48/drivers/staging/wlan-ng/hfa384x_usb.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/staging/wlan-ng/hfa384x_usb.c 2011-11-17 20:09:04.000000000 -0500 @@ -205,7 +205,7 @@ static void unlocked_usbctlx_complete(hf struct usbctlx_completor { int (*complete) (struct usbctlx_completor *); -}; +} __no_const; typedef struct usbctlx_completor usbctlx_completor_t; static int diff -urNp linux-2.6.32.48/drivers/telephony/ixj.c linux-2.6.32.48-openvz/drivers/telephony/ixj.c --- linux-2.6.32.48/drivers/telephony/ixj.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/telephony/ixj.c 2011-11-17 20:08:54.000000000 -0500 @@ -4976,6 +4976,8 @@ static int ixj_daa_cid_read(IXJ *j) bool mContinue; char *pIn, *pOut; + pax_track_stack(); + if (!SCI_Prepare(j)) return 0; diff -urNp linux-2.6.32.48/drivers/uio/uio.c linux-2.6.32.48-openvz/drivers/uio/uio.c --- linux-2.6.32.48/drivers/uio/uio.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/uio/uio.c 2011-11-17 20:08:54.000000000 -0500 @@ -23,6 +23,7 @@ #include #include #include +#include #define UIO_MAX_DEVICES 255 @@ -30,10 +31,10 @@ struct uio_device { struct module *owner; struct device *dev; int minor; - atomic_t event; + atomic_unchecked_t event; struct fasync_struct *async_queue; wait_queue_head_t wait; - int vma_count; + local_t vma_count; struct uio_info *info; struct kobject *map_dir; struct kobject *portio_dir; @@ -129,7 +130,7 @@ static ssize_t map_type_show(struct kobj return entry->show(mem, buf); } -static struct sysfs_ops map_sysfs_ops = { +static const struct sysfs_ops map_sysfs_ops = { .show = map_type_show, }; @@ -217,7 +218,7 @@ static ssize_t portio_type_show(struct k return entry->show(port, buf); } -static struct sysfs_ops portio_sysfs_ops = { +static const struct sysfs_ops portio_sysfs_ops = { .show = portio_type_show, }; @@ -255,7 +256,7 @@ static ssize_t show_event(struct device struct uio_device *idev = dev_get_drvdata(dev); if (idev) return sprintf(buf, "%u\n", - (unsigned int)atomic_read(&idev->event)); + (unsigned int)atomic_read_unchecked(&idev->event)); else return -ENODEV; } @@ -424,7 +425,7 @@ void uio_event_notify(struct uio_info *i { struct uio_device *idev = info->uio_dev; - atomic_inc(&idev->event); + atomic_inc_unchecked(&idev->event); wake_up_interruptible(&idev->wait); kill_fasync(&idev->async_queue, SIGIO, POLL_IN); } @@ -477,7 +478,7 @@ static int uio_open(struct inode *inode, } listener->dev = idev; - listener->event_count = atomic_read(&idev->event); + listener->event_count = atomic_read_unchecked(&idev->event); filep->private_data = listener; if (idev->info->open) { @@ -528,7 +529,7 @@ static unsigned int uio_poll(struct file return -EIO; poll_wait(filep, &idev->wait, wait); - if (listener->event_count != atomic_read(&idev->event)) + if (listener->event_count != atomic_read_unchecked(&idev->event)) return POLLIN | POLLRDNORM; return 0; } @@ -553,7 +554,7 @@ static ssize_t uio_read(struct file *fil do { set_current_state(TASK_INTERRUPTIBLE); - event_count = atomic_read(&idev->event); + event_count = atomic_read_unchecked(&idev->event); if (event_count != listener->event_count) { if (copy_to_user(buf, &event_count, count)) retval = -EFAULT; @@ -624,13 +625,13 @@ static int uio_find_mem_index(struct vm_ static void uio_vma_open(struct vm_area_struct *vma) { struct uio_device *idev = vma->vm_private_data; - idev->vma_count++; + local_inc(&idev->vma_count); } static void uio_vma_close(struct vm_area_struct *vma) { struct uio_device *idev = vma->vm_private_data; - idev->vma_count--; + local_dec(&idev->vma_count); } static int uio_vma_fault(struct vm_area_struct *vma, struct vm_fault *vmf) @@ -840,7 +841,7 @@ int __uio_register_device(struct module idev->owner = owner; idev->info = info; init_waitqueue_head(&idev->wait); - atomic_set(&idev->event, 0); + atomic_set_unchecked(&idev->event, 0); ret = uio_get_minor(idev); if (ret) diff -urNp linux-2.6.32.48/drivers/usb/atm/usbatm.c linux-2.6.32.48-openvz/drivers/usb/atm/usbatm.c --- linux-2.6.32.48/drivers/usb/atm/usbatm.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/usb/atm/usbatm.c 2011-11-17 20:08:54.000000000 -0500 @@ -333,7 +333,7 @@ static void usbatm_extract_one_cell(stru if (printk_ratelimit()) atm_warn(instance, "%s: OAM not supported (vpi %d, vci %d)!\n", __func__, vpi, vci); - atomic_inc(&vcc->stats->rx_err); + atomic_inc_unchecked(&vcc->stats->rx_err); return; } @@ -361,7 +361,7 @@ static void usbatm_extract_one_cell(stru if (length > ATM_MAX_AAL5_PDU) { atm_rldbg(instance, "%s: bogus length %u (vcc: 0x%p)!\n", __func__, length, vcc); - atomic_inc(&vcc->stats->rx_err); + atomic_inc_unchecked(&vcc->stats->rx_err); goto out; } @@ -370,14 +370,14 @@ static void usbatm_extract_one_cell(stru if (sarb->len < pdu_length) { atm_rldbg(instance, "%s: bogus pdu_length %u (sarb->len: %u, vcc: 0x%p)!\n", __func__, pdu_length, sarb->len, vcc); - atomic_inc(&vcc->stats->rx_err); + atomic_inc_unchecked(&vcc->stats->rx_err); goto out; } if (crc32_be(~0, skb_tail_pointer(sarb) - pdu_length, pdu_length) != 0xc704dd7b) { atm_rldbg(instance, "%s: packet failed crc check (vcc: 0x%p)!\n", __func__, vcc); - atomic_inc(&vcc->stats->rx_err); + atomic_inc_unchecked(&vcc->stats->rx_err); goto out; } @@ -387,7 +387,7 @@ static void usbatm_extract_one_cell(stru if (printk_ratelimit()) atm_err(instance, "%s: no memory for skb (length: %u)!\n", __func__, length); - atomic_inc(&vcc->stats->rx_drop); + atomic_inc_unchecked(&vcc->stats->rx_drop); goto out; } @@ -412,7 +412,7 @@ static void usbatm_extract_one_cell(stru vcc->push(vcc, skb); - atomic_inc(&vcc->stats->rx); + atomic_inc_unchecked(&vcc->stats->rx); out: skb_trim(sarb, 0); } @@ -616,7 +616,7 @@ static void usbatm_tx_process(unsigned l struct atm_vcc *vcc = UDSL_SKB(skb)->atm.vcc; usbatm_pop(vcc, skb); - atomic_inc(&vcc->stats->tx); + atomic_inc_unchecked(&vcc->stats->tx); skb = skb_dequeue(&instance->sndqueue); } @@ -775,11 +775,11 @@ static int usbatm_atm_proc_read(struct a if (!left--) return sprintf(page, "AAL5: tx %d ( %d err ), rx %d ( %d err, %d drop )\n", - atomic_read(&atm_dev->stats.aal5.tx), - atomic_read(&atm_dev->stats.aal5.tx_err), - atomic_read(&atm_dev->stats.aal5.rx), - atomic_read(&atm_dev->stats.aal5.rx_err), - atomic_read(&atm_dev->stats.aal5.rx_drop)); + atomic_read_unchecked(&atm_dev->stats.aal5.tx), + atomic_read_unchecked(&atm_dev->stats.aal5.tx_err), + atomic_read_unchecked(&atm_dev->stats.aal5.rx), + atomic_read_unchecked(&atm_dev->stats.aal5.rx_err), + atomic_read_unchecked(&atm_dev->stats.aal5.rx_drop)); if (!left--) { if (instance->disconnected) diff -urNp linux-2.6.32.48/drivers/usb/class/cdc-wdm.c linux-2.6.32.48-openvz/drivers/usb/class/cdc-wdm.c --- linux-2.6.32.48/drivers/usb/class/cdc-wdm.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/usb/class/cdc-wdm.c 2011-11-17 20:08:54.000000000 -0500 @@ -314,7 +314,7 @@ static ssize_t wdm_write if (r < 0) goto outnp; - if (!file->f_flags && O_NONBLOCK) + if (!(file->f_flags & O_NONBLOCK)) r = wait_event_interruptible(desc->wait, !test_bit(WDM_IN_USE, &desc->flags)); else diff -urNp linux-2.6.32.48/drivers/usb/core/hcd.c linux-2.6.32.48-openvz/drivers/usb/core/hcd.c --- linux-2.6.32.48/drivers/usb/core/hcd.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/usb/core/hcd.c 2011-11-17 20:08:54.000000000 -0500 @@ -2216,7 +2216,7 @@ EXPORT_SYMBOL_GPL(usb_hcd_platform_shutd #if defined(CONFIG_USB_MON) || defined(CONFIG_USB_MON_MODULE) -struct usb_mon_operations *mon_ops; +const struct usb_mon_operations *mon_ops; /* * The registration is unlocked. @@ -2226,7 +2226,7 @@ struct usb_mon_operations *mon_ops; * symbols from usbcore, usbcore gets referenced and cannot be unloaded first. */ -int usb_mon_register (struct usb_mon_operations *ops) +int usb_mon_register (const struct usb_mon_operations *ops) { if (mon_ops) diff -urNp linux-2.6.32.48/drivers/usb/core/hcd.h linux-2.6.32.48-openvz/drivers/usb/core/hcd.h --- linux-2.6.32.48/drivers/usb/core/hcd.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/usb/core/hcd.h 2011-11-17 20:08:54.000000000 -0500 @@ -486,13 +486,13 @@ static inline void usbfs_cleanup(void) { #if defined(CONFIG_USB_MON) || defined(CONFIG_USB_MON_MODULE) struct usb_mon_operations { - void (*urb_submit)(struct usb_bus *bus, struct urb *urb); - void (*urb_submit_error)(struct usb_bus *bus, struct urb *urb, int err); - void (*urb_complete)(struct usb_bus *bus, struct urb *urb, int status); + void (* const urb_submit)(struct usb_bus *bus, struct urb *urb); + void (* const urb_submit_error)(struct usb_bus *bus, struct urb *urb, int err); + void (* const urb_complete)(struct usb_bus *bus, struct urb *urb, int status); /* void (*urb_unlink)(struct usb_bus *bus, struct urb *urb); */ }; -extern struct usb_mon_operations *mon_ops; +extern const struct usb_mon_operations *mon_ops; static inline void usbmon_urb_submit(struct usb_bus *bus, struct urb *urb) { @@ -514,7 +514,7 @@ static inline void usbmon_urb_complete(s (*mon_ops->urb_complete)(bus, urb, status); } -int usb_mon_register(struct usb_mon_operations *ops); +int usb_mon_register(const struct usb_mon_operations *ops); void usb_mon_deregister(void); #else diff -urNp linux-2.6.32.48/drivers/usb/core/message.c linux-2.6.32.48-openvz/drivers/usb/core/message.c --- linux-2.6.32.48/drivers/usb/core/message.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/usb/core/message.c 2011-11-17 20:08:54.000000000 -0500 @@ -914,8 +914,8 @@ char *usb_cache_string(struct usb_device buf = kmalloc(MAX_USB_STRING_SIZE, GFP_NOIO); if (buf) { len = usb_string(udev, index, buf, MAX_USB_STRING_SIZE); - if (len > 0) { - smallbuf = kmalloc(++len, GFP_NOIO); + if (len++ > 0) { + smallbuf = kmalloc(len, GFP_NOIO); if (!smallbuf) return buf; memcpy(smallbuf, buf, len); diff -urNp linux-2.6.32.48/drivers/usb/misc/appledisplay.c linux-2.6.32.48-openvz/drivers/usb/misc/appledisplay.c --- linux-2.6.32.48/drivers/usb/misc/appledisplay.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/usb/misc/appledisplay.c 2011-11-17 20:08:54.000000000 -0500 @@ -178,7 +178,7 @@ static int appledisplay_bl_get_brightnes return pdata->msgdata[1]; } -static struct backlight_ops appledisplay_bl_data = { +static const struct backlight_ops appledisplay_bl_data = { .get_brightness = appledisplay_bl_get_brightness, .update_status = appledisplay_bl_update_status, }; diff -urNp linux-2.6.32.48/drivers/usb/mon/mon_main.c linux-2.6.32.48-openvz/drivers/usb/mon/mon_main.c --- linux-2.6.32.48/drivers/usb/mon/mon_main.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/usb/mon/mon_main.c 2011-11-17 20:08:54.000000000 -0500 @@ -238,7 +238,7 @@ static struct notifier_block mon_nb = { /* * Ops */ -static struct usb_mon_operations mon_ops_0 = { +static const struct usb_mon_operations mon_ops_0 = { .urb_submit = mon_submit, .urb_submit_error = mon_submit_error, .urb_complete = mon_complete, diff -urNp linux-2.6.32.48/drivers/usb/wusbcore/wa-hc.h linux-2.6.32.48-openvz/drivers/usb/wusbcore/wa-hc.h --- linux-2.6.32.48/drivers/usb/wusbcore/wa-hc.h 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/usb/wusbcore/wa-hc.h 2011-11-17 20:08:54.000000000 -0500 @@ -192,7 +192,7 @@ struct wahc { struct list_head xfer_delayed_list; spinlock_t xfer_list_lock; struct work_struct xfer_work; - atomic_t xfer_id_count; + atomic_unchecked_t xfer_id_count; }; @@ -246,7 +246,7 @@ static inline void wa_init(struct wahc * INIT_LIST_HEAD(&wa->xfer_delayed_list); spin_lock_init(&wa->xfer_list_lock); INIT_WORK(&wa->xfer_work, wa_urb_enqueue_run); - atomic_set(&wa->xfer_id_count, 1); + atomic_set_unchecked(&wa->xfer_id_count, 1); } /** diff -urNp linux-2.6.32.48/drivers/usb/wusbcore/wa-xfer.c linux-2.6.32.48-openvz/drivers/usb/wusbcore/wa-xfer.c --- linux-2.6.32.48/drivers/usb/wusbcore/wa-xfer.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/usb/wusbcore/wa-xfer.c 2011-11-17 20:08:54.000000000 -0500 @@ -293,7 +293,7 @@ out: */ static void wa_xfer_id_init(struct wa_xfer *xfer) { - xfer->id = atomic_add_return(1, &xfer->wa->xfer_id_count); + xfer->id = atomic_add_return_unchecked(1, &xfer->wa->xfer_id_count); } /* diff -urNp linux-2.6.32.48/drivers/uwb/wlp/messages.c linux-2.6.32.48-openvz/drivers/uwb/wlp/messages.c --- linux-2.6.32.48/drivers/uwb/wlp/messages.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/uwb/wlp/messages.c 2011-11-17 20:08:54.000000000 -0500 @@ -903,7 +903,7 @@ int wlp_parse_f0(struct wlp *wlp, struct size_t len = skb->len; size_t used; ssize_t result; - struct wlp_nonce enonce, rnonce; + struct wlp_nonce enonce = {{0}}, rnonce = {{0}}; enum wlp_assc_error assc_err; char enonce_buf[WLP_WSS_NONCE_STRSIZE]; char rnonce_buf[WLP_WSS_NONCE_STRSIZE]; diff -urNp linux-2.6.32.48/drivers/uwb/wlp/sysfs.c linux-2.6.32.48-openvz/drivers/uwb/wlp/sysfs.c --- linux-2.6.32.48/drivers/uwb/wlp/sysfs.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/uwb/wlp/sysfs.c 2011-11-17 20:08:54.000000000 -0500 @@ -615,8 +615,7 @@ ssize_t wlp_wss_attr_store(struct kobjec return ret; } -static -struct sysfs_ops wss_sysfs_ops = { +static const struct sysfs_ops wss_sysfs_ops = { .show = wlp_wss_attr_show, .store = wlp_wss_attr_store, }; diff -urNp linux-2.6.32.48/drivers/video/atmel_lcdfb.c linux-2.6.32.48-openvz/drivers/video/atmel_lcdfb.c --- linux-2.6.32.48/drivers/video/atmel_lcdfb.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/video/atmel_lcdfb.c 2011-11-17 20:08:54.000000000 -0500 @@ -110,7 +110,7 @@ static int atmel_bl_get_brightness(struc return lcdc_readl(sinfo, ATMEL_LCDC_CONTRAST_VAL); } -static struct backlight_ops atmel_lcdc_bl_ops = { +static const struct backlight_ops atmel_lcdc_bl_ops = { .update_status = atmel_bl_update_status, .get_brightness = atmel_bl_get_brightness, }; diff -urNp linux-2.6.32.48/drivers/video/aty/aty128fb.c linux-2.6.32.48-openvz/drivers/video/aty/aty128fb.c --- linux-2.6.32.48/drivers/video/aty/aty128fb.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/video/aty/aty128fb.c 2011-11-17 20:08:54.000000000 -0500 @@ -1787,7 +1787,7 @@ static int aty128_bl_get_brightness(stru return bd->props.brightness; } -static struct backlight_ops aty128_bl_data = { +static const struct backlight_ops aty128_bl_data = { .get_brightness = aty128_bl_get_brightness, .update_status = aty128_bl_update_status, }; diff -urNp linux-2.6.32.48/drivers/video/aty/atyfb_base.c linux-2.6.32.48-openvz/drivers/video/aty/atyfb_base.c --- linux-2.6.32.48/drivers/video/aty/atyfb_base.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/video/aty/atyfb_base.c 2011-11-17 20:08:54.000000000 -0500 @@ -2225,7 +2225,7 @@ static int aty_bl_get_brightness(struct return bd->props.brightness; } -static struct backlight_ops aty_bl_data = { +static const struct backlight_ops aty_bl_data = { .get_brightness = aty_bl_get_brightness, .update_status = aty_bl_update_status, }; diff -urNp linux-2.6.32.48/drivers/video/aty/radeon_backlight.c linux-2.6.32.48-openvz/drivers/video/aty/radeon_backlight.c --- linux-2.6.32.48/drivers/video/aty/radeon_backlight.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/video/aty/radeon_backlight.c 2011-11-17 20:08:54.000000000 -0500 @@ -127,7 +127,7 @@ static int radeon_bl_get_brightness(stru return bd->props.brightness; } -static struct backlight_ops radeon_bl_data = { +static const struct backlight_ops radeon_bl_data = { .get_brightness = radeon_bl_get_brightness, .update_status = radeon_bl_update_status, }; diff -urNp linux-2.6.32.48/drivers/video/backlight/adp5520_bl.c linux-2.6.32.48-openvz/drivers/video/backlight/adp5520_bl.c --- linux-2.6.32.48/drivers/video/backlight/adp5520_bl.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/video/backlight/adp5520_bl.c 2011-11-17 20:08:54.000000000 -0500 @@ -84,7 +84,7 @@ static int adp5520_bl_get_brightness(str return error ? data->current_brightness : reg_val; } -static struct backlight_ops adp5520_bl_ops = { +static const struct backlight_ops adp5520_bl_ops = { .update_status = adp5520_bl_update_status, .get_brightness = adp5520_bl_get_brightness, }; diff -urNp linux-2.6.32.48/drivers/video/backlight/adx_bl.c linux-2.6.32.48-openvz/drivers/video/backlight/adx_bl.c --- linux-2.6.32.48/drivers/video/backlight/adx_bl.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/video/backlight/adx_bl.c 2011-11-17 20:08:54.000000000 -0500 @@ -61,7 +61,7 @@ static int adx_backlight_check_fb(struct return 1; } -static struct backlight_ops adx_backlight_ops = { +static const struct backlight_ops adx_backlight_ops = { .options = 0, .update_status = adx_backlight_update_status, .get_brightness = adx_backlight_get_brightness, diff -urNp linux-2.6.32.48/drivers/video/backlight/atmel-pwm-bl.c linux-2.6.32.48-openvz/drivers/video/backlight/atmel-pwm-bl.c --- linux-2.6.32.48/drivers/video/backlight/atmel-pwm-bl.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/video/backlight/atmel-pwm-bl.c 2011-11-17 20:08:54.000000000 -0500 @@ -113,7 +113,7 @@ static int atmel_pwm_bl_init_pwm(struct return pwm_channel_enable(&pwmbl->pwmc); } -static struct backlight_ops atmel_pwm_bl_ops = { +static const struct backlight_ops atmel_pwm_bl_ops = { .get_brightness = atmel_pwm_bl_get_intensity, .update_status = atmel_pwm_bl_set_intensity, }; diff -urNp linux-2.6.32.48/drivers/video/backlight/backlight.c linux-2.6.32.48-openvz/drivers/video/backlight/backlight.c --- linux-2.6.32.48/drivers/video/backlight/backlight.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/video/backlight/backlight.c 2011-11-17 20:08:54.000000000 -0500 @@ -269,7 +269,7 @@ EXPORT_SYMBOL(backlight_force_update); * ERR_PTR() or a pointer to the newly allocated device. */ struct backlight_device *backlight_device_register(const char *name, - struct device *parent, void *devdata, struct backlight_ops *ops) + struct device *parent, void *devdata, const struct backlight_ops *ops) { struct backlight_device *new_bd; int rc; diff -urNp linux-2.6.32.48/drivers/video/backlight/corgi_lcd.c linux-2.6.32.48-openvz/drivers/video/backlight/corgi_lcd.c --- linux-2.6.32.48/drivers/video/backlight/corgi_lcd.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/video/backlight/corgi_lcd.c 2011-11-17 20:08:54.000000000 -0500 @@ -451,7 +451,7 @@ void corgi_lcd_limit_intensity(int limit } EXPORT_SYMBOL(corgi_lcd_limit_intensity); -static struct backlight_ops corgi_bl_ops = { +static const struct backlight_ops corgi_bl_ops = { .get_brightness = corgi_bl_get_intensity, .update_status = corgi_bl_update_status, }; diff -urNp linux-2.6.32.48/drivers/video/backlight/cr_bllcd.c linux-2.6.32.48-openvz/drivers/video/backlight/cr_bllcd.c --- linux-2.6.32.48/drivers/video/backlight/cr_bllcd.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/video/backlight/cr_bllcd.c 2011-11-17 20:08:54.000000000 -0500 @@ -108,7 +108,7 @@ static int cr_backlight_get_intensity(st return intensity; } -static struct backlight_ops cr_backlight_ops = { +static const struct backlight_ops cr_backlight_ops = { .get_brightness = cr_backlight_get_intensity, .update_status = cr_backlight_set_intensity, }; diff -urNp linux-2.6.32.48/drivers/video/backlight/da903x_bl.c linux-2.6.32.48-openvz/drivers/video/backlight/da903x_bl.c --- linux-2.6.32.48/drivers/video/backlight/da903x_bl.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/video/backlight/da903x_bl.c 2011-11-17 20:08:54.000000000 -0500 @@ -94,7 +94,7 @@ static int da903x_backlight_get_brightne return data->current_brightness; } -static struct backlight_ops da903x_backlight_ops = { +static const struct backlight_ops da903x_backlight_ops = { .update_status = da903x_backlight_update_status, .get_brightness = da903x_backlight_get_brightness, }; diff -urNp linux-2.6.32.48/drivers/video/backlight/generic_bl.c linux-2.6.32.48-openvz/drivers/video/backlight/generic_bl.c --- linux-2.6.32.48/drivers/video/backlight/generic_bl.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/video/backlight/generic_bl.c 2011-11-17 20:08:54.000000000 -0500 @@ -70,7 +70,7 @@ void corgibl_limit_intensity(int limit) } EXPORT_SYMBOL(corgibl_limit_intensity); -static struct backlight_ops genericbl_ops = { +static const struct backlight_ops genericbl_ops = { .options = BL_CORE_SUSPENDRESUME, .get_brightness = genericbl_get_intensity, .update_status = genericbl_send_intensity, diff -urNp linux-2.6.32.48/drivers/video/backlight/hp680_bl.c linux-2.6.32.48-openvz/drivers/video/backlight/hp680_bl.c --- linux-2.6.32.48/drivers/video/backlight/hp680_bl.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/video/backlight/hp680_bl.c 2011-11-17 20:08:54.000000000 -0500 @@ -98,7 +98,7 @@ static int hp680bl_get_intensity(struct return current_intensity; } -static struct backlight_ops hp680bl_ops = { +static const struct backlight_ops hp680bl_ops = { .get_brightness = hp680bl_get_intensity, .update_status = hp680bl_set_intensity, }; diff -urNp linux-2.6.32.48/drivers/video/backlight/jornada720_bl.c linux-2.6.32.48-openvz/drivers/video/backlight/jornada720_bl.c --- linux-2.6.32.48/drivers/video/backlight/jornada720_bl.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/video/backlight/jornada720_bl.c 2011-11-17 20:08:54.000000000 -0500 @@ -93,7 +93,7 @@ out: return ret; } -static struct backlight_ops jornada_bl_ops = { +static const struct backlight_ops jornada_bl_ops = { .get_brightness = jornada_bl_get_brightness, .update_status = jornada_bl_update_status, .options = BL_CORE_SUSPENDRESUME, diff -urNp linux-2.6.32.48/drivers/video/backlight/kb3886_bl.c linux-2.6.32.48-openvz/drivers/video/backlight/kb3886_bl.c --- linux-2.6.32.48/drivers/video/backlight/kb3886_bl.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/video/backlight/kb3886_bl.c 2011-11-17 20:08:54.000000000 -0500 @@ -134,7 +134,7 @@ static int kb3886bl_get_intensity(struct return kb3886bl_intensity; } -static struct backlight_ops kb3886bl_ops = { +static const struct backlight_ops kb3886bl_ops = { .get_brightness = kb3886bl_get_intensity, .update_status = kb3886bl_send_intensity, }; diff -urNp linux-2.6.32.48/drivers/video/backlight/locomolcd.c linux-2.6.32.48-openvz/drivers/video/backlight/locomolcd.c --- linux-2.6.32.48/drivers/video/backlight/locomolcd.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/video/backlight/locomolcd.c 2011-11-17 20:08:54.000000000 -0500 @@ -141,7 +141,7 @@ static int locomolcd_get_intensity(struc return current_intensity; } -static struct backlight_ops locomobl_data = { +static const struct backlight_ops locomobl_data = { .get_brightness = locomolcd_get_intensity, .update_status = locomolcd_set_intensity, }; diff -urNp linux-2.6.32.48/drivers/video/backlight/mbp_nvidia_bl.c linux-2.6.32.48-openvz/drivers/video/backlight/mbp_nvidia_bl.c --- linux-2.6.32.48/drivers/video/backlight/mbp_nvidia_bl.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/video/backlight/mbp_nvidia_bl.c 2011-11-17 20:08:54.000000000 -0500 @@ -33,7 +33,7 @@ struct dmi_match_data { unsigned long iostart; unsigned long iolen; /* Backlight operations structure. */ - struct backlight_ops backlight_ops; + const struct backlight_ops backlight_ops; }; /* Module parameters. */ diff -urNp linux-2.6.32.48/drivers/video/backlight/omap1_bl.c linux-2.6.32.48-openvz/drivers/video/backlight/omap1_bl.c --- linux-2.6.32.48/drivers/video/backlight/omap1_bl.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/video/backlight/omap1_bl.c 2011-11-17 20:08:54.000000000 -0500 @@ -125,7 +125,7 @@ static int omapbl_get_intensity(struct b return bl->current_intensity; } -static struct backlight_ops omapbl_ops = { +static const struct backlight_ops omapbl_ops = { .get_brightness = omapbl_get_intensity, .update_status = omapbl_update_status, }; diff -urNp linux-2.6.32.48/drivers/video/backlight/progear_bl.c linux-2.6.32.48-openvz/drivers/video/backlight/progear_bl.c --- linux-2.6.32.48/drivers/video/backlight/progear_bl.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/video/backlight/progear_bl.c 2011-11-17 20:08:54.000000000 -0500 @@ -54,7 +54,7 @@ static int progearbl_get_intensity(struc return intensity - HW_LEVEL_MIN; } -static struct backlight_ops progearbl_ops = { +static const struct backlight_ops progearbl_ops = { .get_brightness = progearbl_get_intensity, .update_status = progearbl_set_intensity, }; diff -urNp linux-2.6.32.48/drivers/video/backlight/pwm_bl.c linux-2.6.32.48-openvz/drivers/video/backlight/pwm_bl.c --- linux-2.6.32.48/drivers/video/backlight/pwm_bl.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/video/backlight/pwm_bl.c 2011-11-17 20:08:54.000000000 -0500 @@ -56,7 +56,7 @@ static int pwm_backlight_get_brightness( return bl->props.brightness; } -static struct backlight_ops pwm_backlight_ops = { +static const struct backlight_ops pwm_backlight_ops = { .update_status = pwm_backlight_update_status, .get_brightness = pwm_backlight_get_brightness, }; diff -urNp linux-2.6.32.48/drivers/video/backlight/tosa_bl.c linux-2.6.32.48-openvz/drivers/video/backlight/tosa_bl.c --- linux-2.6.32.48/drivers/video/backlight/tosa_bl.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/video/backlight/tosa_bl.c 2011-11-17 20:08:54.000000000 -0500 @@ -72,7 +72,7 @@ static int tosa_bl_get_brightness(struct return props->brightness; } -static struct backlight_ops bl_ops = { +static const struct backlight_ops bl_ops = { .get_brightness = tosa_bl_get_brightness, .update_status = tosa_bl_update_status, }; diff -urNp linux-2.6.32.48/drivers/video/backlight/wm831x_bl.c linux-2.6.32.48-openvz/drivers/video/backlight/wm831x_bl.c --- linux-2.6.32.48/drivers/video/backlight/wm831x_bl.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/video/backlight/wm831x_bl.c 2011-11-17 20:08:54.000000000 -0500 @@ -112,7 +112,7 @@ static int wm831x_backlight_get_brightne return data->current_brightness; } -static struct backlight_ops wm831x_backlight_ops = { +static const struct backlight_ops wm831x_backlight_ops = { .options = BL_CORE_SUSPENDRESUME, .update_status = wm831x_backlight_update_status, .get_brightness = wm831x_backlight_get_brightness, diff -urNp linux-2.6.32.48/drivers/video/bf54x-lq043fb.c linux-2.6.32.48-openvz/drivers/video/bf54x-lq043fb.c --- linux-2.6.32.48/drivers/video/bf54x-lq043fb.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/video/bf54x-lq043fb.c 2011-11-17 20:08:54.000000000 -0500 @@ -463,7 +463,7 @@ static int bl_get_brightness(struct back return 0; } -static struct backlight_ops bfin_lq043fb_bl_ops = { +static const struct backlight_ops bfin_lq043fb_bl_ops = { .get_brightness = bl_get_brightness, }; diff -urNp linux-2.6.32.48/drivers/video/bfin-t350mcqb-fb.c linux-2.6.32.48-openvz/drivers/video/bfin-t350mcqb-fb.c --- linux-2.6.32.48/drivers/video/bfin-t350mcqb-fb.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/video/bfin-t350mcqb-fb.c 2011-11-17 20:08:54.000000000 -0500 @@ -381,7 +381,7 @@ static int bl_get_brightness(struct back return 0; } -static struct backlight_ops bfin_lq043fb_bl_ops = { +static const struct backlight_ops bfin_lq043fb_bl_ops = { .get_brightness = bl_get_brightness, }; diff -urNp linux-2.6.32.48/drivers/video/fbcmap.c linux-2.6.32.48-openvz/drivers/video/fbcmap.c --- linux-2.6.32.48/drivers/video/fbcmap.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/video/fbcmap.c 2011-11-17 20:08:54.000000000 -0500 @@ -266,8 +266,7 @@ int fb_set_user_cmap(struct fb_cmap_user rc = -ENODEV; goto out; } - if (cmap->start < 0 || (!info->fbops->fb_setcolreg && - !info->fbops->fb_setcmap)) { + if (!info->fbops->fb_setcolreg && !info->fbops->fb_setcmap) { rc = -EINVAL; goto out1; } diff -urNp linux-2.6.32.48/drivers/video/fbmem.c linux-2.6.32.48-openvz/drivers/video/fbmem.c --- linux-2.6.32.48/drivers/video/fbmem.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/video/fbmem.c 2011-11-17 20:08:54.000000000 -0500 @@ -403,7 +403,7 @@ static void fb_do_show_logo(struct fb_in image->dx += image->width + 8; } } else if (rotate == FB_ROTATE_UD) { - for (x = 0; x < num && image->dx >= 0; x++) { + for (x = 0; x < num && (__s32)image->dx >= 0; x++) { info->fbops->fb_imageblit(info, image); image->dx -= image->width + 8; } @@ -415,7 +415,7 @@ static void fb_do_show_logo(struct fb_in image->dy += image->height + 8; } } else if (rotate == FB_ROTATE_CCW) { - for (x = 0; x < num && image->dy >= 0; x++) { + for (x = 0; x < num && (__s32)image->dy >= 0; x++) { info->fbops->fb_imageblit(info, image); image->dy -= image->height + 8; } @@ -915,6 +915,8 @@ fb_set_var(struct fb_info *info, struct int flags = info->flags; int ret = 0; + pax_track_stack(); + if (var->activate & FB_ACTIVATE_INV_MODE) { struct fb_videomode mode1, mode2; @@ -1040,6 +1042,8 @@ static long do_fb_ioctl(struct fb_info * void __user *argp = (void __user *)arg; long ret = 0; + pax_track_stack(); + switch (cmd) { case FBIOGET_VSCREENINFO: if (!lock_fb_info(info)) @@ -1119,7 +1123,7 @@ static long do_fb_ioctl(struct fb_info * return -EFAULT; if (con2fb.console < 1 || con2fb.console > MAX_NR_CONSOLES) return -EINVAL; - if (con2fb.framebuffer < 0 || con2fb.framebuffer >= FB_MAX) + if (con2fb.framebuffer >= FB_MAX) return -EINVAL; if (!registered_fb[con2fb.framebuffer]) request_module("fb%d", con2fb.framebuffer); diff -urNp linux-2.6.32.48/drivers/video/i810/i810_accel.c linux-2.6.32.48-openvz/drivers/video/i810/i810_accel.c --- linux-2.6.32.48/drivers/video/i810/i810_accel.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/video/i810/i810_accel.c 2011-11-17 20:08:54.000000000 -0500 @@ -73,6 +73,7 @@ static inline int wait_for_space(struct } } printk("ringbuffer lockup!!!\n"); + printk("head:%u tail:%u iring.size:%u space:%u\n", head, tail, par->iring.size, space); i810_report_error(mmio); par->dev_flags |= LOCKUP; info->pixmap.scan_align = 1; diff -urNp linux-2.6.32.48/drivers/video/logo/logo_linux_clut224.ppm linux-2.6.32.48-openvz/drivers/video/logo/logo_linux_clut224.ppm --- linux-2.6.32.48/drivers/video/logo/logo_linux_clut224.ppm 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/video/logo/logo_linux_clut224.ppm 2011-11-17 20:09:04.000000000 -0500 @@ -1,1604 +1,1123 @@ P3 -# Standard 224-color Linux logo 80 80 255 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 6 6 6 6 6 6 10 10 10 10 10 10 - 10 10 10 6 6 6 6 6 6 6 6 6 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 6 6 6 10 10 10 14 14 14 - 22 22 22 26 26 26 30 30 30 34 34 34 - 30 30 30 30 30 30 26 26 26 18 18 18 - 14 14 14 10 10 10 6 6 6 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 1 0 0 1 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 6 6 6 14 14 14 26 26 26 42 42 42 - 54 54 54 66 66 66 78 78 78 78 78 78 - 78 78 78 74 74 74 66 66 66 54 54 54 - 42 42 42 26 26 26 18 18 18 10 10 10 - 6 6 6 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 1 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 10 10 10 - 22 22 22 42 42 42 66 66 66 86 86 86 - 66 66 66 38 38 38 38 38 38 22 22 22 - 26 26 26 34 34 34 54 54 54 66 66 66 - 86 86 86 70 70 70 46 46 46 26 26 26 - 14 14 14 6 6 6 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 1 0 0 1 0 0 1 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 10 10 10 26 26 26 - 50 50 50 82 82 82 58 58 58 6 6 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 6 6 6 54 54 54 86 86 86 66 66 66 - 38 38 38 18 18 18 6 6 6 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 6 6 6 22 22 22 50 50 50 - 78 78 78 34 34 34 2 2 6 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 2 2 6 6 6 6 70 70 70 - 78 78 78 46 46 46 22 22 22 6 6 6 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 1 0 0 1 0 0 1 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 6 6 6 18 18 18 42 42 42 82 82 82 - 26 26 26 2 2 6 2 2 6 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 2 2 6 2 2 6 14 14 14 - 46 46 46 34 34 34 6 6 6 2 2 6 - 42 42 42 78 78 78 42 42 42 18 18 18 - 6 6 6 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 1 0 0 0 0 0 1 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 10 10 10 30 30 30 66 66 66 58 58 58 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 2 2 6 2 2 6 26 26 26 - 86 86 86 101 101 101 46 46 46 10 10 10 - 2 2 6 58 58 58 70 70 70 34 34 34 - 10 10 10 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 1 0 0 1 0 0 1 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 14 14 14 42 42 42 86 86 86 10 10 10 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 2 2 6 2 2 6 30 30 30 - 94 94 94 94 94 94 58 58 58 26 26 26 - 2 2 6 6 6 6 78 78 78 54 54 54 - 22 22 22 6 6 6 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 6 6 6 - 22 22 22 62 62 62 62 62 62 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 2 2 6 2 2 6 26 26 26 - 54 54 54 38 38 38 18 18 18 10 10 10 - 2 2 6 2 2 6 34 34 34 82 82 82 - 38 38 38 14 14 14 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 1 0 0 1 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 6 6 6 - 30 30 30 78 78 78 30 30 30 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 2 2 6 2 2 6 10 10 10 - 10 10 10 2 2 6 2 2 6 2 2 6 - 2 2 6 2 2 6 2 2 6 78 78 78 - 50 50 50 18 18 18 6 6 6 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 1 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 10 10 10 - 38 38 38 86 86 86 14 14 14 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 2 2 6 2 2 6 54 54 54 - 66 66 66 26 26 26 6 6 6 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 1 0 0 1 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 14 14 14 - 42 42 42 82 82 82 2 2 6 2 2 6 - 2 2 6 6 6 6 10 10 10 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 2 2 6 2 2 6 6 6 6 - 14 14 14 10 10 10 2 2 6 2 2 6 - 2 2 6 2 2 6 2 2 6 18 18 18 - 82 82 82 34 34 34 10 10 10 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 1 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 14 14 14 - 46 46 46 86 86 86 2 2 6 2 2 6 - 6 6 6 6 6 6 22 22 22 34 34 34 - 6 6 6 2 2 6 2 2 6 2 2 6 - 2 2 6 2 2 6 18 18 18 34 34 34 - 10 10 10 50 50 50 22 22 22 2 2 6 - 2 2 6 2 2 6 2 2 6 10 10 10 - 86 86 86 42 42 42 14 14 14 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 1 0 0 1 0 0 1 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 14 14 14 - 46 46 46 86 86 86 2 2 6 2 2 6 - 38 38 38 116 116 116 94 94 94 22 22 22 - 22 22 22 2 2 6 2 2 6 2 2 6 - 14 14 14 86 86 86 138 138 138 162 162 162 -154 154 154 38 38 38 26 26 26 6 6 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 86 86 86 46 46 46 14 14 14 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 14 14 14 - 46 46 46 86 86 86 2 2 6 14 14 14 -134 134 134 198 198 198 195 195 195 116 116 116 - 10 10 10 2 2 6 2 2 6 6 6 6 -101 98 89 187 187 187 210 210 210 218 218 218 -214 214 214 134 134 134 14 14 14 6 6 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 86 86 86 50 50 50 18 18 18 6 6 6 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 1 0 0 0 - 0 0 1 0 0 1 0 0 1 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 14 14 14 - 46 46 46 86 86 86 2 2 6 54 54 54 -218 218 218 195 195 195 226 226 226 246 246 246 - 58 58 58 2 2 6 2 2 6 30 30 30 -210 210 210 253 253 253 174 174 174 123 123 123 -221 221 221 234 234 234 74 74 74 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 70 70 70 58 58 58 22 22 22 6 6 6 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 14 14 14 - 46 46 46 82 82 82 2 2 6 106 106 106 -170 170 170 26 26 26 86 86 86 226 226 226 -123 123 123 10 10 10 14 14 14 46 46 46 -231 231 231 190 190 190 6 6 6 70 70 70 - 90 90 90 238 238 238 158 158 158 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 70 70 70 58 58 58 22 22 22 6 6 6 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 1 0 0 0 - 0 0 1 0 0 1 0 0 1 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 14 14 14 - 42 42 42 86 86 86 6 6 6 116 116 116 -106 106 106 6 6 6 70 70 70 149 149 149 -128 128 128 18 18 18 38 38 38 54 54 54 -221 221 221 106 106 106 2 2 6 14 14 14 - 46 46 46 190 190 190 198 198 198 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 74 74 74 62 62 62 22 22 22 6 6 6 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 1 0 0 0 - 0 0 1 0 0 0 0 0 1 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 14 14 14 - 42 42 42 94 94 94 14 14 14 101 101 101 -128 128 128 2 2 6 18 18 18 116 116 116 -118 98 46 121 92 8 121 92 8 98 78 10 -162 162 162 106 106 106 2 2 6 2 2 6 - 2 2 6 195 195 195 195 195 195 6 6 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 74 74 74 62 62 62 22 22 22 6 6 6 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 1 0 0 1 - 0 0 1 0 0 0 0 0 1 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 10 10 10 - 38 38 38 90 90 90 14 14 14 58 58 58 -210 210 210 26 26 26 54 38 6 154 114 10 -226 170 11 236 186 11 225 175 15 184 144 12 -215 174 15 175 146 61 37 26 9 2 2 6 - 70 70 70 246 246 246 138 138 138 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 70 70 70 66 66 66 26 26 26 6 6 6 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 10 10 10 - 38 38 38 86 86 86 14 14 14 10 10 10 -195 195 195 188 164 115 192 133 9 225 175 15 -239 182 13 234 190 10 232 195 16 232 200 30 -245 207 45 241 208 19 232 195 16 184 144 12 -218 194 134 211 206 186 42 42 42 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 50 50 50 74 74 74 30 30 30 6 6 6 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 10 10 10 - 34 34 34 86 86 86 14 14 14 2 2 6 -121 87 25 192 133 9 219 162 10 239 182 13 -236 186 11 232 195 16 241 208 19 244 214 54 -246 218 60 246 218 38 246 215 20 241 208 19 -241 208 19 226 184 13 121 87 25 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 50 50 50 82 82 82 34 34 34 10 10 10 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 10 10 10 - 34 34 34 82 82 82 30 30 30 61 42 6 -180 123 7 206 145 10 230 174 11 239 182 13 -234 190 10 238 202 15 241 208 19 246 218 74 -246 218 38 246 215 20 246 215 20 246 215 20 -226 184 13 215 174 15 184 144 12 6 6 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 26 26 26 94 94 94 42 42 42 14 14 14 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 10 10 10 - 30 30 30 78 78 78 50 50 50 104 69 6 -192 133 9 216 158 10 236 178 12 236 186 11 -232 195 16 241 208 19 244 214 54 245 215 43 -246 215 20 246 215 20 241 208 19 198 155 10 -200 144 11 216 158 10 156 118 10 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 6 6 6 90 90 90 54 54 54 18 18 18 - 6 6 6 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 10 10 10 - 30 30 30 78 78 78 46 46 46 22 22 22 -137 92 6 210 162 10 239 182 13 238 190 10 -238 202 15 241 208 19 246 215 20 246 215 20 -241 208 19 203 166 17 185 133 11 210 150 10 -216 158 10 210 150 10 102 78 10 2 2 6 - 6 6 6 54 54 54 14 14 14 2 2 6 - 2 2 6 62 62 62 74 74 74 30 30 30 - 10 10 10 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 10 10 10 - 34 34 34 78 78 78 50 50 50 6 6 6 - 94 70 30 139 102 15 190 146 13 226 184 13 -232 200 30 232 195 16 215 174 15 190 146 13 -168 122 10 192 133 9 210 150 10 213 154 11 -202 150 34 182 157 106 101 98 89 2 2 6 - 2 2 6 78 78 78 116 116 116 58 58 58 - 2 2 6 22 22 22 90 90 90 46 46 46 - 18 18 18 6 6 6 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 10 10 10 - 38 38 38 86 86 86 50 50 50 6 6 6 -128 128 128 174 154 114 156 107 11 168 122 10 -198 155 10 184 144 12 197 138 11 200 144 11 -206 145 10 206 145 10 197 138 11 188 164 115 -195 195 195 198 198 198 174 174 174 14 14 14 - 2 2 6 22 22 22 116 116 116 116 116 116 - 22 22 22 2 2 6 74 74 74 70 70 70 - 30 30 30 10 10 10 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 6 6 6 18 18 18 - 50 50 50 101 101 101 26 26 26 10 10 10 -138 138 138 190 190 190 174 154 114 156 107 11 -197 138 11 200 144 11 197 138 11 192 133 9 -180 123 7 190 142 34 190 178 144 187 187 187 -202 202 202 221 221 221 214 214 214 66 66 66 - 2 2 6 2 2 6 50 50 50 62 62 62 - 6 6 6 2 2 6 10 10 10 90 90 90 - 50 50 50 18 18 18 6 6 6 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 10 10 10 34 34 34 - 74 74 74 74 74 74 2 2 6 6 6 6 -144 144 144 198 198 198 190 190 190 178 166 146 -154 121 60 156 107 11 156 107 11 168 124 44 -174 154 114 187 187 187 190 190 190 210 210 210 -246 246 246 253 253 253 253 253 253 182 182 182 - 6 6 6 2 2 6 2 2 6 2 2 6 - 2 2 6 2 2 6 2 2 6 62 62 62 - 74 74 74 34 34 34 14 14 14 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 10 10 10 22 22 22 54 54 54 - 94 94 94 18 18 18 2 2 6 46 46 46 -234 234 234 221 221 221 190 190 190 190 190 190 -190 190 190 187 187 187 187 187 187 190 190 190 -190 190 190 195 195 195 214 214 214 242 242 242 -253 253 253 253 253 253 253 253 253 253 253 253 - 82 82 82 2 2 6 2 2 6 2 2 6 - 2 2 6 2 2 6 2 2 6 14 14 14 - 86 86 86 54 54 54 22 22 22 6 6 6 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 6 6 6 18 18 18 46 46 46 90 90 90 - 46 46 46 18 18 18 6 6 6 182 182 182 -253 253 253 246 246 246 206 206 206 190 190 190 -190 190 190 190 190 190 190 190 190 190 190 190 -206 206 206 231 231 231 250 250 250 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -202 202 202 14 14 14 2 2 6 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 42 42 42 86 86 86 42 42 42 18 18 18 - 6 6 6 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 6 6 6 - 14 14 14 38 38 38 74 74 74 66 66 66 - 2 2 6 6 6 6 90 90 90 250 250 250 -253 253 253 253 253 253 238 238 238 198 198 198 -190 190 190 190 190 190 195 195 195 221 221 221 -246 246 246 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 82 82 82 2 2 6 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 78 78 78 70 70 70 34 34 34 - 14 14 14 6 6 6 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 14 14 14 - 34 34 34 66 66 66 78 78 78 6 6 6 - 2 2 6 18 18 18 218 218 218 253 253 253 -253 253 253 253 253 253 253 253 253 246 246 246 -226 226 226 231 231 231 246 246 246 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 178 178 178 2 2 6 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 18 18 18 90 90 90 62 62 62 - 30 30 30 10 10 10 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 10 10 10 26 26 26 - 58 58 58 90 90 90 18 18 18 2 2 6 - 2 2 6 110 110 110 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -250 250 250 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 231 231 231 18 18 18 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 2 2 6 18 18 18 94 94 94 - 54 54 54 26 26 26 10 10 10 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 6 6 6 22 22 22 50 50 50 - 90 90 90 26 26 26 2 2 6 2 2 6 - 14 14 14 195 195 195 250 250 250 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -250 250 250 242 242 242 54 54 54 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 2 2 6 2 2 6 38 38 38 - 86 86 86 50 50 50 22 22 22 6 6 6 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 6 6 6 14 14 14 38 38 38 82 82 82 - 34 34 34 2 2 6 2 2 6 2 2 6 - 42 42 42 195 195 195 246 246 246 253 253 253 -253 253 253 253 253 253 253 253 253 250 250 250 -242 242 242 242 242 242 250 250 250 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 250 250 250 246 246 246 238 238 238 -226 226 226 231 231 231 101 101 101 6 6 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 38 38 38 82 82 82 42 42 42 14 14 14 - 6 6 6 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 10 10 10 26 26 26 62 62 62 66 66 66 - 2 2 6 2 2 6 2 2 6 6 6 6 - 70 70 70 170 170 170 206 206 206 234 234 234 -246 246 246 250 250 250 250 250 250 238 238 238 -226 226 226 231 231 231 238 238 238 250 250 250 -250 250 250 250 250 250 246 246 246 231 231 231 -214 214 214 206 206 206 202 202 202 202 202 202 -198 198 198 202 202 202 182 182 182 18 18 18 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 62 62 62 66 66 66 30 30 30 - 10 10 10 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 14 14 14 42 42 42 82 82 82 18 18 18 - 2 2 6 2 2 6 2 2 6 10 10 10 - 94 94 94 182 182 182 218 218 218 242 242 242 -250 250 250 253 253 253 253 253 253 250 250 250 -234 234 234 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 246 246 246 -238 238 238 226 226 226 210 210 210 202 202 202 -195 195 195 195 195 195 210 210 210 158 158 158 - 6 6 6 14 14 14 50 50 50 14 14 14 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 6 6 6 86 86 86 46 46 46 - 18 18 18 6 6 6 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 6 6 6 - 22 22 22 54 54 54 70 70 70 2 2 6 - 2 2 6 10 10 10 2 2 6 22 22 22 -166 166 166 231 231 231 250 250 250 253 253 253 -253 253 253 253 253 253 253 253 253 250 250 250 -242 242 242 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 246 246 246 -231 231 231 206 206 206 198 198 198 226 226 226 - 94 94 94 2 2 6 6 6 6 38 38 38 - 30 30 30 2 2 6 2 2 6 2 2 6 - 2 2 6 2 2 6 62 62 62 66 66 66 - 26 26 26 10 10 10 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 10 10 10 - 30 30 30 74 74 74 50 50 50 2 2 6 - 26 26 26 26 26 26 2 2 6 106 106 106 -238 238 238 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 246 246 246 218 218 218 202 202 202 -210 210 210 14 14 14 2 2 6 2 2 6 - 30 30 30 22 22 22 2 2 6 2 2 6 - 2 2 6 2 2 6 18 18 18 86 86 86 - 42 42 42 14 14 14 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 14 14 14 - 42 42 42 90 90 90 22 22 22 2 2 6 - 42 42 42 2 2 6 18 18 18 218 218 218 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 250 250 250 221 221 221 -218 218 218 101 101 101 2 2 6 14 14 14 - 18 18 18 38 38 38 10 10 10 2 2 6 - 2 2 6 2 2 6 2 2 6 78 78 78 - 58 58 58 22 22 22 6 6 6 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 6 6 6 18 18 18 - 54 54 54 82 82 82 2 2 6 26 26 26 - 22 22 22 2 2 6 123 123 123 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 250 250 250 -238 238 238 198 198 198 6 6 6 38 38 38 - 58 58 58 26 26 26 38 38 38 2 2 6 - 2 2 6 2 2 6 2 2 6 46 46 46 - 78 78 78 30 30 30 10 10 10 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 10 10 10 30 30 30 - 74 74 74 58 58 58 2 2 6 42 42 42 - 2 2 6 22 22 22 231 231 231 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 250 250 250 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 246 246 246 46 46 46 38 38 38 - 42 42 42 14 14 14 38 38 38 14 14 14 - 2 2 6 2 2 6 2 2 6 6 6 6 - 86 86 86 46 46 46 14 14 14 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 6 6 6 14 14 14 42 42 42 - 90 90 90 18 18 18 18 18 18 26 26 26 - 2 2 6 116 116 116 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 250 250 250 238 238 238 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 94 94 94 6 6 6 - 2 2 6 2 2 6 10 10 10 34 34 34 - 2 2 6 2 2 6 2 2 6 2 2 6 - 74 74 74 58 58 58 22 22 22 6 6 6 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 10 10 10 26 26 26 66 66 66 - 82 82 82 2 2 6 38 38 38 6 6 6 - 14 14 14 210 210 210 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 246 246 246 242 242 242 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 144 144 144 2 2 6 - 2 2 6 2 2 6 2 2 6 46 46 46 - 2 2 6 2 2 6 2 2 6 2 2 6 - 42 42 42 74 74 74 30 30 30 10 10 10 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 6 6 6 14 14 14 42 42 42 90 90 90 - 26 26 26 6 6 6 42 42 42 2 2 6 - 74 74 74 250 250 250 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 242 242 242 242 242 242 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 182 182 182 2 2 6 - 2 2 6 2 2 6 2 2 6 46 46 46 - 2 2 6 2 2 6 2 2 6 2 2 6 - 10 10 10 86 86 86 38 38 38 10 10 10 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 10 10 10 26 26 26 66 66 66 82 82 82 - 2 2 6 22 22 22 18 18 18 2 2 6 -149 149 149 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 234 234 234 242 242 242 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 206 206 206 2 2 6 - 2 2 6 2 2 6 2 2 6 38 38 38 - 2 2 6 2 2 6 2 2 6 2 2 6 - 6 6 6 86 86 86 46 46 46 14 14 14 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 6 6 6 - 18 18 18 46 46 46 86 86 86 18 18 18 - 2 2 6 34 34 34 10 10 10 6 6 6 -210 210 210 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 234 234 234 242 242 242 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 221 221 221 6 6 6 - 2 2 6 2 2 6 6 6 6 30 30 30 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 82 82 82 54 54 54 18 18 18 - 6 6 6 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 10 10 10 - 26 26 26 66 66 66 62 62 62 2 2 6 - 2 2 6 38 38 38 10 10 10 26 26 26 -238 238 238 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 231 231 231 238 238 238 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 231 231 231 6 6 6 - 2 2 6 2 2 6 10 10 10 30 30 30 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 66 66 66 58 58 58 22 22 22 - 6 6 6 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 10 10 10 - 38 38 38 78 78 78 6 6 6 2 2 6 - 2 2 6 46 46 46 14 14 14 42 42 42 -246 246 246 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 231 231 231 242 242 242 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 234 234 234 10 10 10 - 2 2 6 2 2 6 22 22 22 14 14 14 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 66 66 66 62 62 62 22 22 22 - 6 6 6 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 6 6 6 18 18 18 - 50 50 50 74 74 74 2 2 6 2 2 6 - 14 14 14 70 70 70 34 34 34 62 62 62 -250 250 250 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 231 231 231 246 246 246 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 234 234 234 14 14 14 - 2 2 6 2 2 6 30 30 30 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 66 66 66 62 62 62 22 22 22 - 6 6 6 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 6 6 6 18 18 18 - 54 54 54 62 62 62 2 2 6 2 2 6 - 2 2 6 30 30 30 46 46 46 70 70 70 -250 250 250 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 231 231 231 246 246 246 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 226 226 226 10 10 10 - 2 2 6 6 6 6 30 30 30 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 66 66 66 58 58 58 22 22 22 - 6 6 6 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 6 6 6 22 22 22 - 58 58 58 62 62 62 2 2 6 2 2 6 - 2 2 6 2 2 6 30 30 30 78 78 78 -250 250 250 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 231 231 231 246 246 246 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 206 206 206 2 2 6 - 22 22 22 34 34 34 18 14 6 22 22 22 - 26 26 26 18 18 18 6 6 6 2 2 6 - 2 2 6 82 82 82 54 54 54 18 18 18 - 6 6 6 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 6 6 6 26 26 26 - 62 62 62 106 106 106 74 54 14 185 133 11 -210 162 10 121 92 8 6 6 6 62 62 62 -238 238 238 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 231 231 231 246 246 246 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 158 158 158 18 18 18 - 14 14 14 2 2 6 2 2 6 2 2 6 - 6 6 6 18 18 18 66 66 66 38 38 38 - 6 6 6 94 94 94 50 50 50 18 18 18 - 6 6 6 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 6 6 6 - 10 10 10 10 10 10 18 18 18 38 38 38 - 78 78 78 142 134 106 216 158 10 242 186 14 -246 190 14 246 190 14 156 118 10 10 10 10 - 90 90 90 238 238 238 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 231 231 231 250 250 250 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 246 230 190 -238 204 91 238 204 91 181 142 44 37 26 9 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 2 2 6 38 38 38 46 46 46 - 26 26 26 106 106 106 54 54 54 18 18 18 - 6 6 6 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 6 6 6 14 14 14 22 22 22 - 30 30 30 38 38 38 50 50 50 70 70 70 -106 106 106 190 142 34 226 170 11 242 186 14 -246 190 14 246 190 14 246 190 14 154 114 10 - 6 6 6 74 74 74 226 226 226 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 231 231 231 250 250 250 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 228 184 62 -241 196 14 241 208 19 232 195 16 38 30 10 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 6 6 6 30 30 30 26 26 26 -203 166 17 154 142 90 66 66 66 26 26 26 - 6 6 6 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 6 6 6 18 18 18 38 38 38 58 58 58 - 78 78 78 86 86 86 101 101 101 123 123 123 -175 146 61 210 150 10 234 174 13 246 186 14 -246 190 14 246 190 14 246 190 14 238 190 10 -102 78 10 2 2 6 46 46 46 198 198 198 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 234 234 234 242 242 242 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 224 178 62 -242 186 14 241 196 14 210 166 10 22 18 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 2 2 6 6 6 6 121 92 8 -238 202 15 232 195 16 82 82 82 34 34 34 - 10 10 10 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 14 14 14 38 38 38 70 70 70 154 122 46 -190 142 34 200 144 11 197 138 11 197 138 11 -213 154 11 226 170 11 242 186 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -225 175 15 46 32 6 2 2 6 22 22 22 -158 158 158 250 250 250 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 250 250 250 242 242 242 224 178 62 -239 182 13 236 186 11 213 154 11 46 32 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 2 2 6 61 42 6 225 175 15 -238 190 10 236 186 11 112 100 78 42 42 42 - 14 14 14 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 6 6 6 - 22 22 22 54 54 54 154 122 46 213 154 11 -226 170 11 230 174 11 226 170 11 226 170 11 -236 178 12 242 186 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -241 196 14 184 144 12 10 10 10 2 2 6 - 6 6 6 116 116 116 242 242 242 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 231 231 231 198 198 198 214 170 54 -236 178 12 236 178 12 210 150 10 137 92 6 - 18 14 6 2 2 6 2 2 6 2 2 6 - 6 6 6 70 47 6 200 144 11 236 178 12 -239 182 13 239 182 13 124 112 88 58 58 58 - 22 22 22 6 6 6 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 10 10 10 - 30 30 30 70 70 70 180 133 36 226 170 11 -239 182 13 242 186 14 242 186 14 246 186 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 190 14 232 195 16 98 70 6 2 2 6 - 2 2 6 2 2 6 66 66 66 221 221 221 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 206 206 206 198 198 198 214 166 58 -230 174 11 230 174 11 216 158 10 192 133 9 -163 110 8 116 81 8 102 78 10 116 81 8 -167 114 7 197 138 11 226 170 11 239 182 13 -242 186 14 242 186 14 162 146 94 78 78 78 - 34 34 34 14 14 14 6 6 6 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 6 6 6 - 30 30 30 78 78 78 190 142 34 226 170 11 -239 182 13 246 190 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 190 14 241 196 14 203 166 17 22 18 6 - 2 2 6 2 2 6 2 2 6 38 38 38 -218 218 218 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -250 250 250 206 206 206 198 198 198 202 162 69 -226 170 11 236 178 12 224 166 10 210 150 10 -200 144 11 197 138 11 192 133 9 197 138 11 -210 150 10 226 170 11 242 186 14 246 190 14 -246 190 14 246 186 14 225 175 15 124 112 88 - 62 62 62 30 30 30 14 14 14 6 6 6 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 10 10 10 - 30 30 30 78 78 78 174 135 50 224 166 10 -239 182 13 246 190 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 190 14 246 190 14 241 196 14 139 102 15 - 2 2 6 2 2 6 2 2 6 2 2 6 - 78 78 78 250 250 250 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -250 250 250 214 214 214 198 198 198 190 150 46 -219 162 10 236 178 12 234 174 13 224 166 10 -216 158 10 213 154 11 213 154 11 216 158 10 -226 170 11 239 182 13 246 190 14 246 190 14 -246 190 14 246 190 14 242 186 14 206 162 42 -101 101 101 58 58 58 30 30 30 14 14 14 - 6 6 6 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 10 10 10 - 30 30 30 74 74 74 174 135 50 216 158 10 -236 178 12 246 190 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 190 14 246 190 14 241 196 14 226 184 13 - 61 42 6 2 2 6 2 2 6 2 2 6 - 22 22 22 238 238 238 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 226 226 226 187 187 187 180 133 36 -216 158 10 236 178 12 239 182 13 236 178 12 -230 174 11 226 170 11 226 170 11 230 174 11 -236 178 12 242 186 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 186 14 239 182 13 -206 162 42 106 106 106 66 66 66 34 34 34 - 14 14 14 6 6 6 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 6 6 6 - 26 26 26 70 70 70 163 133 67 213 154 11 -236 178 12 246 190 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 241 196 14 -190 146 13 18 14 6 2 2 6 2 2 6 - 46 46 46 246 246 246 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 221 221 221 86 86 86 156 107 11 -216 158 10 236 178 12 242 186 14 246 186 14 -242 186 14 239 182 13 239 182 13 242 186 14 -242 186 14 246 186 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -242 186 14 225 175 15 142 122 72 66 66 66 - 30 30 30 10 10 10 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 6 6 6 - 26 26 26 70 70 70 163 133 67 210 150 10 -236 178 12 246 190 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -232 195 16 121 92 8 34 34 34 106 106 106 -221 221 221 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -242 242 242 82 82 82 18 14 6 163 110 8 -216 158 10 236 178 12 242 186 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 190 14 246 190 14 242 186 14 163 133 67 - 46 46 46 18 18 18 6 6 6 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 10 10 10 - 30 30 30 78 78 78 163 133 67 210 150 10 -236 178 12 246 186 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -241 196 14 215 174 15 190 178 144 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 218 218 218 - 58 58 58 2 2 6 22 18 6 167 114 7 -216 158 10 236 178 12 246 186 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 190 14 246 186 14 242 186 14 190 150 46 - 54 54 54 22 22 22 6 6 6 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 14 14 14 - 38 38 38 86 86 86 180 133 36 213 154 11 -236 178 12 246 186 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 190 14 232 195 16 190 146 13 214 214 214 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 250 250 250 170 170 170 26 26 26 - 2 2 6 2 2 6 37 26 9 163 110 8 -219 162 10 239 182 13 246 186 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 186 14 236 178 12 224 166 10 142 122 72 - 46 46 46 18 18 18 6 6 6 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 6 6 6 18 18 18 - 50 50 50 109 106 95 192 133 9 224 166 10 -242 186 14 246 190 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -242 186 14 226 184 13 210 162 10 142 110 46 -226 226 226 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -253 253 253 253 253 253 253 253 253 253 253 253 -198 198 198 66 66 66 2 2 6 2 2 6 - 2 2 6 2 2 6 50 34 6 156 107 11 -219 162 10 239 182 13 246 186 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 242 186 14 -234 174 13 213 154 11 154 122 46 66 66 66 - 30 30 30 10 10 10 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 6 6 6 22 22 22 - 58 58 58 154 121 60 206 145 10 234 174 13 -242 186 14 246 186 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 186 14 236 178 12 210 162 10 163 110 8 - 61 42 6 138 138 138 218 218 218 250 250 250 -253 253 253 253 253 253 253 253 253 250 250 250 -242 242 242 210 210 210 144 144 144 66 66 66 - 6 6 6 2 2 6 2 2 6 2 2 6 - 2 2 6 2 2 6 61 42 6 163 110 8 -216 158 10 236 178 12 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 190 14 239 182 13 230 174 11 216 158 10 -190 142 34 124 112 88 70 70 70 38 38 38 - 18 18 18 6 6 6 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 6 6 6 22 22 22 - 62 62 62 168 124 44 206 145 10 224 166 10 -236 178 12 239 182 13 242 186 14 242 186 14 -246 186 14 246 190 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 190 14 236 178 12 216 158 10 175 118 6 - 80 54 7 2 2 6 6 6 6 30 30 30 - 54 54 54 62 62 62 50 50 50 38 38 38 - 14 14 14 2 2 6 2 2 6 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 6 6 6 80 54 7 167 114 7 -213 154 11 236 178 12 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 190 14 242 186 14 239 182 13 239 182 13 -230 174 11 210 150 10 174 135 50 124 112 88 - 82 82 82 54 54 54 34 34 34 18 18 18 - 6 6 6 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 6 6 6 18 18 18 - 50 50 50 158 118 36 192 133 9 200 144 11 -216 158 10 219 162 10 224 166 10 226 170 11 -230 174 11 236 178 12 239 182 13 239 182 13 -242 186 14 246 186 14 246 190 14 246 190 14 -246 190 14 246 190 14 246 190 14 246 190 14 -246 186 14 230 174 11 210 150 10 163 110 8 -104 69 6 10 10 10 2 2 6 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 6 6 6 91 60 6 167 114 7 -206 145 10 230 174 11 242 186 14 246 190 14 -246 190 14 246 190 14 246 186 14 242 186 14 -239 182 13 230 174 11 224 166 10 213 154 11 -180 133 36 124 112 88 86 86 86 58 58 58 - 38 38 38 22 22 22 10 10 10 6 6 6 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 14 14 14 - 34 34 34 70 70 70 138 110 50 158 118 36 -167 114 7 180 123 7 192 133 9 197 138 11 -200 144 11 206 145 10 213 154 11 219 162 10 -224 166 10 230 174 11 239 182 13 242 186 14 -246 186 14 246 186 14 246 186 14 246 186 14 -239 182 13 216 158 10 185 133 11 152 99 6 -104 69 6 18 14 6 2 2 6 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 2 2 6 2 2 6 2 2 6 - 2 2 6 6 6 6 80 54 7 152 99 6 -192 133 9 219 162 10 236 178 12 239 182 13 -246 186 14 242 186 14 239 182 13 236 178 12 -224 166 10 206 145 10 192 133 9 154 121 60 - 94 94 94 62 62 62 42 42 42 22 22 22 - 14 14 14 6 6 6 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 6 6 6 - 18 18 18 34 34 34 58 58 58 78 78 78 -101 98 89 124 112 88 142 110 46 156 107 11 -163 110 8 167 114 7 175 118 6 180 123 7 -185 133 11 197 138 11 210 150 10 219 162 10 -226 170 11 236 178 12 236 178 12 234 174 13 -219 162 10 197 138 11 163 110 8 130 83 6 - 91 60 6 10 10 10 2 2 6 2 2 6 - 18 18 18 38 38 38 38 38 38 38 38 38 - 38 38 38 38 38 38 38 38 38 38 38 38 - 38 38 38 38 38 38 26 26 26 2 2 6 - 2 2 6 6 6 6 70 47 6 137 92 6 -175 118 6 200 144 11 219 162 10 230 174 11 -234 174 13 230 174 11 219 162 10 210 150 10 -192 133 9 163 110 8 124 112 88 82 82 82 - 50 50 50 30 30 30 14 14 14 6 6 6 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 6 6 6 14 14 14 22 22 22 34 34 34 - 42 42 42 58 58 58 74 74 74 86 86 86 -101 98 89 122 102 70 130 98 46 121 87 25 -137 92 6 152 99 6 163 110 8 180 123 7 -185 133 11 197 138 11 206 145 10 200 144 11 -180 123 7 156 107 11 130 83 6 104 69 6 - 50 34 6 54 54 54 110 110 110 101 98 89 - 86 86 86 82 82 82 78 78 78 78 78 78 - 78 78 78 78 78 78 78 78 78 78 78 78 - 78 78 78 82 82 82 86 86 86 94 94 94 -106 106 106 101 101 101 86 66 34 124 80 6 -156 107 11 180 123 7 192 133 9 200 144 11 -206 145 10 200 144 11 192 133 9 175 118 6 -139 102 15 109 106 95 70 70 70 42 42 42 - 22 22 22 10 10 10 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 6 6 6 10 10 10 - 14 14 14 22 22 22 30 30 30 38 38 38 - 50 50 50 62 62 62 74 74 74 90 90 90 -101 98 89 112 100 78 121 87 25 124 80 6 -137 92 6 152 99 6 152 99 6 152 99 6 -138 86 6 124 80 6 98 70 6 86 66 30 -101 98 89 82 82 82 58 58 58 46 46 46 - 38 38 38 34 34 34 34 34 34 34 34 34 - 34 34 34 34 34 34 34 34 34 34 34 34 - 34 34 34 34 34 34 38 38 38 42 42 42 - 54 54 54 82 82 82 94 86 76 91 60 6 -134 86 6 156 107 11 167 114 7 175 118 6 -175 118 6 167 114 7 152 99 6 121 87 25 -101 98 89 62 62 62 34 34 34 18 18 18 - 6 6 6 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 6 6 6 6 6 6 10 10 10 - 18 18 18 22 22 22 30 30 30 42 42 42 - 50 50 50 66 66 66 86 86 86 101 98 89 -106 86 58 98 70 6 104 69 6 104 69 6 -104 69 6 91 60 6 82 62 34 90 90 90 - 62 62 62 38 38 38 22 22 22 14 14 14 - 10 10 10 10 10 10 10 10 10 10 10 10 - 10 10 10 10 10 10 6 6 6 10 10 10 - 10 10 10 10 10 10 10 10 10 14 14 14 - 22 22 22 42 42 42 70 70 70 89 81 66 - 80 54 7 104 69 6 124 80 6 137 92 6 -134 86 6 116 81 8 100 82 52 86 86 86 - 58 58 58 30 30 30 14 14 14 6 6 6 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 6 6 6 10 10 10 14 14 14 - 18 18 18 26 26 26 38 38 38 54 54 54 - 70 70 70 86 86 86 94 86 76 89 81 66 - 89 81 66 86 86 86 74 74 74 50 50 50 - 30 30 30 14 14 14 6 6 6 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 6 6 6 18 18 18 34 34 34 58 58 58 - 82 82 82 89 81 66 89 81 66 89 81 66 - 94 86 66 94 86 76 74 74 74 50 50 50 - 26 26 26 14 14 14 6 6 6 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 6 6 6 6 6 6 14 14 14 18 18 18 - 30 30 30 38 38 38 46 46 46 54 54 54 - 50 50 50 42 42 42 30 30 30 18 18 18 - 10 10 10 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 6 6 6 14 14 14 26 26 26 - 38 38 38 50 50 50 58 58 58 58 58 58 - 54 54 54 42 42 42 30 30 30 18 18 18 - 10 10 10 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 6 6 6 - 6 6 6 10 10 10 14 14 14 18 18 18 - 18 18 18 14 14 14 10 10 10 6 6 6 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 6 6 6 - 14 14 14 18 18 18 22 22 22 22 22 22 - 18 18 18 14 14 14 10 10 10 6 6 6 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 - 0 0 0 0 0 0 0 0 0 0 0 0 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 3 3 3 0 0 0 0 0 0 +0 0 0 0 0 0 0 0 0 0 0 0 3 3 3 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 1 1 1 0 0 0 +0 0 0 3 3 3 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 2 1 0 2 1 0 3 2 2 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 2 2 2 0 0 0 3 4 3 26 28 28 +37 38 37 37 38 37 14 17 19 2 2 2 0 0 0 2 2 2 +5 5 5 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 3 3 3 0 0 0 1 1 1 6 6 6 +2 2 2 0 0 0 3 3 3 4 4 4 4 4 4 4 4 4 +4 4 5 3 3 3 1 0 0 0 0 0 1 0 0 0 0 0 +1 1 1 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +2 2 2 0 0 0 0 0 0 14 17 19 60 74 84 137 136 137 +153 152 153 137 136 137 125 124 125 60 73 81 6 6 6 3 1 0 +0 0 0 3 3 3 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 0 0 0 4 4 4 41 54 63 125 124 125 +60 73 81 6 6 6 4 0 0 3 3 3 4 4 4 4 4 4 +4 4 4 0 0 0 6 9 11 41 54 63 41 65 82 22 30 35 +2 2 2 2 1 0 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 5 5 5 5 5 5 2 2 2 0 0 0 +4 0 0 6 6 6 41 54 63 137 136 137 174 174 174 167 166 167 +165 164 165 165 164 165 163 162 163 163 162 163 125 124 125 41 54 63 +1 1 1 0 0 0 0 0 0 3 3 3 5 5 5 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 5 5 5 5 5 5 +3 3 3 2 0 0 4 0 0 60 73 81 156 155 156 167 166 167 +163 162 163 85 115 134 5 7 8 0 0 0 4 4 4 5 5 5 +0 0 0 2 5 5 55 98 126 90 154 193 90 154 193 72 125 159 +37 51 59 2 0 0 1 1 1 4 5 5 4 4 4 4 4 4 +4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 5 5 5 4 4 4 1 1 1 0 0 0 3 3 3 +37 38 37 125 124 125 163 162 163 174 174 174 158 157 158 158 157 158 +156 155 156 156 155 156 158 157 158 165 164 165 174 174 174 166 165 166 +125 124 125 16 19 21 1 0 0 0 0 0 0 0 0 4 4 4 +5 5 5 5 5 5 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 5 5 5 5 5 5 1 1 1 +0 0 0 0 0 0 37 38 37 153 152 153 174 174 174 158 157 158 +174 174 174 163 162 163 37 38 37 4 3 3 4 0 0 1 1 1 +0 0 0 22 40 52 101 161 196 101 161 196 90 154 193 101 161 196 +64 123 161 14 17 19 0 0 0 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 5 5 5 +5 5 5 2 2 2 0 0 0 4 0 0 24 26 27 85 115 134 +156 155 156 174 174 174 167 166 167 156 155 156 154 153 154 157 156 157 +156 155 156 156 155 156 155 154 155 153 152 153 158 157 158 167 166 167 +174 174 174 156 155 156 60 74 84 16 19 21 0 0 0 0 0 0 +1 1 1 5 5 5 5 5 5 4 4 4 4 4 4 4 4 4 +4 4 4 5 5 5 6 6 6 3 3 3 0 0 0 4 0 0 +13 16 17 60 73 81 137 136 137 165 164 165 156 155 156 153 152 153 +174 174 174 177 184 187 60 73 81 3 1 0 0 0 0 1 1 2 +22 30 35 64 123 161 136 185 209 90 154 193 90 154 193 90 154 193 +90 154 193 21 29 34 0 0 0 3 2 2 4 4 5 4 4 4 +4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 5 5 5 3 3 3 +0 0 0 0 0 0 10 13 16 60 74 84 157 156 157 174 174 174 +174 174 174 158 157 158 153 152 153 154 153 154 156 155 156 155 154 155 +156 155 156 155 154 155 154 153 154 157 156 157 154 153 154 153 152 153 +163 162 163 174 174 174 177 184 187 137 136 137 60 73 81 13 16 17 +4 0 0 0 0 0 3 3 3 5 5 5 4 4 4 4 4 4 +5 5 5 4 4 4 1 1 1 0 0 0 3 3 3 41 54 63 +131 129 131 174 174 174 174 174 174 174 174 174 167 166 167 174 174 174 +190 197 201 137 136 137 24 26 27 4 0 0 16 21 25 50 82 103 +90 154 193 136 185 209 90 154 193 101 161 196 101 161 196 101 161 196 +31 91 132 3 6 7 0 0 0 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 2 2 2 0 0 0 4 0 0 +4 0 0 43 57 68 137 136 137 177 184 187 174 174 174 163 162 163 +155 154 155 155 154 155 156 155 156 155 154 155 158 157 158 165 164 165 +167 166 167 166 165 166 163 162 163 157 156 157 155 154 155 155 154 155 +153 152 153 156 155 156 167 166 167 174 174 174 174 174 174 131 129 131 +41 54 63 5 5 5 0 0 0 0 0 0 3 3 3 4 4 4 +1 1 1 0 0 0 1 0 0 26 28 28 125 124 125 174 174 174 +177 184 187 174 174 174 174 174 174 156 155 156 131 129 131 137 136 137 +125 124 125 24 26 27 4 0 0 41 65 82 90 154 193 136 185 209 +136 185 209 101 161 196 53 118 160 37 112 160 90 154 193 34 86 122 +7 12 15 0 0 0 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 3 3 3 0 0 0 0 0 0 5 5 5 37 38 37 +125 124 125 167 166 167 174 174 174 167 166 167 158 157 158 155 154 155 +156 155 156 156 155 156 156 155 156 163 162 163 167 166 167 155 154 155 +137 136 137 153 152 153 156 155 156 165 164 165 163 162 163 156 155 156 +156 155 156 156 155 156 155 154 155 158 157 158 166 165 166 174 174 174 +167 166 167 125 124 125 37 38 37 1 0 0 0 0 0 0 0 0 +0 0 0 24 26 27 60 74 84 158 157 158 174 174 174 174 174 174 +166 165 166 158 157 158 125 124 125 41 54 63 13 16 17 6 6 6 +6 6 6 37 38 37 80 127 157 136 185 209 101 161 196 101 161 196 +90 154 193 28 67 93 6 10 14 13 20 25 13 20 25 6 10 14 +1 1 2 4 3 3 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +1 1 1 1 0 0 4 3 3 37 38 37 60 74 84 153 152 153 +167 166 167 167 166 167 158 157 158 154 153 154 155 154 155 156 155 156 +157 156 157 158 157 158 167 166 167 167 166 167 131 129 131 43 57 68 +26 28 28 37 38 37 60 73 81 131 129 131 165 164 165 166 165 166 +158 157 158 155 154 155 156 155 156 156 155 156 156 155 156 158 157 158 +165 164 165 174 174 174 163 162 163 60 74 84 16 19 21 13 16 17 +60 73 81 131 129 131 174 174 174 174 174 174 167 166 167 165 164 165 +137 136 137 60 73 81 24 26 27 4 0 0 4 0 0 16 19 21 +52 104 138 101 161 196 136 185 209 136 185 209 90 154 193 27 99 146 +13 20 25 4 5 7 2 5 5 4 5 7 1 1 2 0 0 0 +4 4 4 4 4 4 3 3 3 2 2 2 2 2 2 4 4 4 +4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 3 3 3 0 0 0 +0 0 0 13 16 17 60 73 81 137 136 137 174 174 174 166 165 166 +158 157 158 156 155 156 157 156 157 156 155 156 155 154 155 158 157 158 +167 166 167 174 174 174 153 152 153 60 73 81 16 19 21 4 0 0 +4 0 0 4 0 0 6 6 6 26 28 28 60 74 84 158 157 158 +174 174 174 166 165 166 157 156 157 155 154 155 156 155 156 156 155 156 +155 154 155 158 157 158 167 166 167 167 166 167 131 129 131 125 124 125 +137 136 137 167 166 167 167 166 167 174 174 174 158 157 158 125 124 125 +16 19 21 4 0 0 4 0 0 10 13 16 49 76 92 107 159 188 +136 185 209 136 185 209 90 154 193 26 108 161 22 40 52 6 10 14 +2 3 3 1 1 2 1 1 2 4 4 5 4 4 5 4 4 5 +4 4 5 2 2 1 0 0 0 0 0 0 0 0 0 2 2 2 +4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 5 5 5 3 3 3 0 0 0 1 0 0 4 0 0 +37 51 59 131 129 131 167 166 167 167 166 167 163 162 163 157 156 157 +157 156 157 155 154 155 153 152 153 157 156 157 167 166 167 174 174 174 +153 152 153 125 124 125 37 38 37 4 0 0 4 0 0 4 0 0 +4 3 3 4 3 3 4 0 0 6 6 6 4 0 0 37 38 37 +125 124 125 174 174 174 174 174 174 165 164 165 156 155 156 154 153 154 +156 155 156 156 155 156 155 154 155 163 162 163 158 157 158 163 162 163 +174 174 174 174 174 174 174 174 174 125 124 125 37 38 37 0 0 0 +4 0 0 6 9 11 41 54 63 90 154 193 136 185 209 146 190 211 +136 185 209 37 112 160 22 40 52 6 10 14 3 6 7 1 1 2 +1 1 2 3 3 3 1 1 2 3 3 3 4 4 4 4 4 4 +2 2 2 2 0 0 16 19 21 37 38 37 24 26 27 0 0 0 +0 0 0 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 5 5 5 5 5 5 +4 4 4 0 0 0 0 0 0 0 0 0 26 28 28 120 125 127 +158 157 158 174 174 174 165 164 165 157 156 157 155 154 155 156 155 156 +153 152 153 153 152 153 167 166 167 174 174 174 174 174 174 125 124 125 +37 38 37 4 0 0 0 0 0 4 0 0 4 3 3 4 4 4 +4 4 4 4 4 4 5 5 5 4 0 0 4 0 0 4 0 0 +4 3 3 43 57 68 137 136 137 174 174 174 174 174 174 165 164 165 +154 153 154 153 152 153 153 152 153 153 152 153 163 162 163 174 174 174 +174 174 174 153 152 153 60 73 81 6 6 6 4 0 0 4 3 3 +32 43 50 80 127 157 136 185 209 146 190 211 146 190 211 90 154 193 +28 67 93 28 67 93 40 71 93 3 6 7 1 1 2 2 5 5 +50 82 103 79 117 143 26 37 45 0 0 0 3 3 3 1 1 1 +0 0 0 41 54 63 137 136 137 174 174 174 153 152 153 60 73 81 +2 0 0 0 0 0 +4 4 4 4 4 4 4 4 4 4 4 4 6 6 6 2 2 2 +0 0 0 2 0 0 24 26 27 60 74 84 153 152 153 174 174 174 +174 174 174 157 156 157 154 153 154 156 155 156 154 153 154 153 152 153 +165 164 165 174 174 174 177 184 187 137 136 137 43 57 68 6 6 6 +4 0 0 2 0 0 3 3 3 5 5 5 5 5 5 4 4 4 +4 4 4 4 4 4 4 4 4 5 5 5 6 6 6 4 3 3 +4 0 0 4 0 0 24 26 27 60 73 81 153 152 153 174 174 174 +174 174 174 158 157 158 158 157 158 174 174 174 174 174 174 158 157 158 +60 74 84 24 26 27 4 0 0 4 0 0 17 23 27 59 113 148 +136 185 209 191 222 234 146 190 211 136 185 209 31 91 132 7 11 13 +22 40 52 101 161 196 90 154 193 6 9 11 3 4 4 43 95 132 +136 185 209 172 205 220 55 98 126 0 0 0 0 0 0 2 0 0 +26 28 28 153 152 153 177 184 187 167 166 167 177 184 187 165 164 165 +37 38 37 0 0 0 +4 4 4 4 4 4 5 5 5 5 5 5 1 1 1 0 0 0 +13 16 17 60 73 81 137 136 137 174 174 174 174 174 174 165 164 165 +153 152 153 153 152 153 155 154 155 154 153 154 158 157 158 174 174 174 +177 184 187 163 162 163 60 73 81 16 19 21 4 0 0 4 0 0 +4 3 3 4 4 4 5 5 5 5 5 5 4 4 4 5 5 5 +5 5 5 5 5 5 5 5 5 4 4 4 4 4 4 5 5 5 +6 6 6 4 0 0 4 0 0 4 0 0 24 26 27 60 74 84 +166 165 166 174 174 174 177 184 187 165 164 165 125 124 125 24 26 27 +4 0 0 4 0 0 5 5 5 50 82 103 136 185 209 172 205 220 +146 190 211 136 185 209 26 108 161 22 40 52 7 12 15 44 81 103 +71 116 144 28 67 93 37 51 59 41 65 82 100 139 164 101 161 196 +90 154 193 90 154 193 28 67 93 0 0 0 0 0 0 26 28 28 +125 124 125 167 166 167 163 162 163 153 152 153 163 162 163 174 174 174 +85 115 134 4 0 0 +4 4 4 5 5 5 4 4 4 1 0 0 4 0 0 34 47 55 +125 124 125 174 174 174 174 174 174 167 166 167 157 156 157 153 152 153 +155 154 155 155 154 155 158 157 158 166 165 166 167 166 167 154 153 154 +125 124 125 26 28 28 4 0 0 4 0 0 4 0 0 5 5 5 +5 5 5 4 4 4 4 4 4 4 4 4 4 4 4 1 1 1 +0 0 0 0 0 0 1 1 1 4 4 4 4 4 4 4 4 4 +5 5 5 5 5 5 4 3 3 4 0 0 4 0 0 6 6 6 +37 38 37 131 129 131 137 136 137 37 38 37 0 0 0 4 0 0 +4 5 5 43 61 72 90 154 193 172 205 220 146 190 211 136 185 209 +90 154 193 28 67 93 13 20 25 43 61 72 71 116 144 44 81 103 +2 5 5 7 11 13 59 113 148 101 161 196 90 154 193 28 67 93 +13 20 25 6 10 14 0 0 0 13 16 17 60 73 81 137 136 137 +166 165 166 158 157 158 156 155 156 154 153 154 167 166 167 174 174 174 +60 73 81 4 0 0 +4 4 4 4 4 4 0 0 0 3 3 3 60 74 84 174 174 174 +174 174 174 167 166 167 163 162 163 155 154 155 157 156 157 155 154 155 +156 155 156 163 162 163 167 166 167 158 157 158 125 124 125 37 38 37 +4 3 3 4 0 0 4 0 0 6 6 6 6 6 6 5 5 5 +4 4 4 4 4 4 4 4 4 1 1 1 0 0 0 2 3 3 +10 13 16 7 11 13 1 0 0 0 0 0 2 2 1 4 4 4 +4 4 4 4 4 4 4 4 4 5 5 5 4 3 3 4 0 0 +4 0 0 7 11 13 13 16 17 4 0 0 3 3 3 34 47 55 +80 127 157 146 190 211 172 205 220 136 185 209 136 185 209 136 185 209 +28 67 93 22 40 52 55 98 126 55 98 126 21 29 34 7 11 13 +50 82 103 101 161 196 101 161 196 35 83 115 13 20 25 2 2 1 +1 1 2 1 1 2 37 51 59 131 129 131 174 174 174 174 174 174 +167 166 167 163 162 163 163 162 163 167 166 167 174 174 174 125 124 125 +16 19 21 4 0 0 +4 4 4 4 0 0 4 0 0 60 74 84 174 174 174 174 174 174 +158 157 158 155 154 155 155 154 155 156 155 156 155 154 155 158 157 158 +167 166 167 165 164 165 131 129 131 60 73 81 13 16 17 4 0 0 +4 0 0 4 3 3 6 6 6 4 3 3 5 5 5 4 4 4 +4 4 4 3 2 2 0 0 0 0 0 0 7 11 13 45 69 86 +80 127 157 71 116 144 43 61 72 7 11 13 0 0 0 1 1 1 +4 3 3 4 4 4 4 4 4 4 4 4 6 6 6 5 5 5 +3 2 2 4 0 0 1 0 0 21 29 34 59 113 148 136 185 209 +146 190 211 136 185 209 136 185 209 136 185 209 136 185 209 136 185 209 +68 124 159 44 81 103 22 40 52 13 16 17 43 61 72 90 154 193 +136 185 209 59 113 148 21 29 34 3 4 3 1 1 1 0 0 0 +24 26 27 125 124 125 163 162 163 174 174 174 166 165 166 165 164 165 +163 162 163 125 124 125 125 124 125 125 124 125 125 124 125 26 28 28 +4 0 0 4 3 3 +3 3 3 0 0 0 24 26 27 153 152 153 177 184 187 158 157 158 +156 155 156 156 155 156 155 154 155 155 154 155 165 164 165 174 174 174 +155 154 155 60 74 84 26 28 28 4 0 0 4 0 0 3 1 0 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 3 3 +2 0 0 0 0 0 0 0 0 32 43 50 72 125 159 101 161 196 +136 185 209 101 161 196 101 161 196 79 117 143 32 43 50 0 0 0 +0 0 0 2 2 2 4 4 4 4 4 4 3 3 3 1 0 0 +0 0 0 4 5 5 49 76 92 101 161 196 146 190 211 146 190 211 +136 185 209 136 185 209 136 185 209 136 185 209 136 185 209 90 154 193 +28 67 93 13 16 17 37 51 59 80 127 157 136 185 209 90 154 193 +22 40 52 6 9 11 3 4 3 2 2 1 16 19 21 60 73 81 +137 136 137 163 162 163 158 157 158 166 165 166 167 166 167 153 152 153 +60 74 84 37 38 37 6 6 6 13 16 17 4 0 0 1 0 0 +3 2 2 4 4 4 +3 2 2 4 0 0 37 38 37 137 136 137 167 166 167 158 157 158 +157 156 157 154 153 154 157 156 157 167 166 167 174 174 174 125 124 125 +37 38 37 4 0 0 4 0 0 4 0 0 4 3 3 4 4 4 +4 4 4 4 4 4 5 5 5 5 5 5 1 1 1 0 0 0 +0 0 0 16 21 25 55 98 126 90 154 193 136 185 209 101 161 196 +101 161 196 101 161 196 136 185 209 136 185 209 101 161 196 55 98 126 +14 17 19 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 +22 40 52 90 154 193 146 190 211 146 190 211 136 185 209 136 185 209 +136 185 209 136 185 209 136 185 209 101 161 196 35 83 115 7 11 13 +17 23 27 59 113 148 136 185 209 101 161 196 34 86 122 7 12 15 +2 5 5 3 4 3 6 6 6 60 73 81 131 129 131 163 162 163 +166 165 166 174 174 174 174 174 174 163 162 163 125 124 125 41 54 63 +13 16 17 4 0 0 4 0 0 4 0 0 1 0 0 2 2 2 +4 4 4 4 4 4 +1 1 1 2 1 0 43 57 68 137 136 137 153 152 153 153 152 153 +163 162 163 156 155 156 165 164 165 167 166 167 60 74 84 6 6 6 +4 0 0 4 0 0 5 5 5 4 4 4 4 4 4 4 4 4 +4 5 5 6 6 6 4 3 3 0 0 0 0 0 0 11 15 18 +40 71 93 100 139 164 101 161 196 101 161 196 101 161 196 101 161 196 +101 161 196 101 161 196 101 161 196 101 161 196 136 185 209 136 185 209 +101 161 196 45 69 86 6 6 6 0 0 0 17 23 27 55 98 126 +136 185 209 146 190 211 136 185 209 136 185 209 136 185 209 136 185 209 +136 185 209 136 185 209 90 154 193 22 40 52 7 11 13 50 82 103 +136 185 209 136 185 209 53 118 160 22 40 52 7 11 13 2 5 5 +3 4 3 37 38 37 125 124 125 157 156 157 166 165 166 167 166 167 +174 174 174 174 174 174 137 136 137 60 73 81 4 0 0 4 0 0 +4 0 0 4 0 0 5 5 5 3 3 3 3 3 3 4 4 4 +4 4 4 4 4 4 +4 0 0 4 0 0 41 54 63 137 136 137 125 124 125 131 129 131 +155 154 155 167 166 167 174 174 174 60 74 84 6 6 6 4 0 0 +4 3 3 6 6 6 4 4 4 4 4 4 4 4 4 5 5 5 +4 4 4 1 1 1 0 0 0 3 6 7 41 65 82 72 125 159 +101 161 196 101 161 196 101 161 196 90 154 193 90 154 193 101 161 196 +101 161 196 101 161 196 101 161 196 101 161 196 101 161 196 136 185 209 +136 185 209 136 185 209 80 127 157 55 98 126 101 161 196 146 190 211 +136 185 209 136 185 209 136 185 209 101 161 196 136 185 209 101 161 196 +136 185 209 101 161 196 35 83 115 22 30 35 101 161 196 172 205 220 +90 154 193 28 67 93 7 11 13 2 5 5 3 4 3 13 16 17 +85 115 134 167 166 167 174 174 174 174 174 174 174 174 174 174 174 174 +167 166 167 60 74 84 13 16 17 4 0 0 4 0 0 4 3 3 +6 6 6 5 5 5 4 4 4 5 5 5 4 4 4 5 5 5 +5 5 5 5 5 5 +1 1 1 4 0 0 41 54 63 137 136 137 137 136 137 125 124 125 +131 129 131 167 166 167 157 156 157 37 38 37 6 6 6 4 0 0 +6 6 6 5 5 5 4 4 4 4 4 4 4 5 5 2 2 1 +0 0 0 0 0 0 26 37 45 58 111 146 101 161 196 101 161 196 +101 161 196 90 154 193 90 154 193 90 154 193 101 161 196 101 161 196 +101 161 196 101 161 196 101 161 196 101 161 196 101 161 196 101 161 196 +101 161 196 136 185 209 136 185 209 136 185 209 146 190 211 136 185 209 +136 185 209 101 161 196 136 185 209 136 185 209 101 161 196 136 185 209 +101 161 196 136 185 209 136 185 209 136 185 209 136 185 209 16 89 141 +7 11 13 2 5 5 2 5 5 13 16 17 60 73 81 154 154 154 +174 174 174 174 174 174 174 174 174 174 174 174 163 162 163 125 124 125 +24 26 27 4 0 0 4 0 0 4 0 0 5 5 5 5 5 5 +4 4 4 4 4 4 4 4 4 5 5 5 5 5 5 5 5 5 +5 5 5 4 4 4 +4 0 0 6 6 6 37 38 37 137 136 137 137 136 137 131 129 131 +131 129 131 153 152 153 131 129 131 26 28 28 4 0 0 4 3 3 +6 6 6 4 4 4 4 4 4 4 4 4 0 0 0 0 0 0 +13 20 25 51 88 114 90 154 193 101 161 196 101 161 196 90 154 193 +90 154 193 90 154 193 90 154 193 90 154 193 90 154 193 101 161 196 +101 161 196 101 161 196 101 161 196 101 161 196 136 185 209 101 161 196 +101 161 196 136 185 209 101 161 196 136 185 209 136 185 209 101 161 196 +136 185 209 101 161 196 136 185 209 101 161 196 101 161 196 101 161 196 +136 185 209 136 185 209 136 185 209 37 112 160 21 29 34 5 7 8 +2 5 5 13 16 17 43 57 68 131 129 131 174 174 174 174 174 174 +174 174 174 167 166 167 157 156 157 125 124 125 37 38 37 4 0 0 +4 0 0 4 0 0 5 5 5 5 5 5 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +1 1 1 4 0 0 41 54 63 153 152 153 137 136 137 137 136 137 +137 136 137 153 152 153 125 124 125 24 26 27 4 0 0 3 2 2 +4 4 4 4 4 4 4 3 3 4 0 0 3 6 7 43 61 72 +64 123 161 101 161 196 90 154 193 90 154 193 90 154 193 90 154 193 +90 154 193 90 154 193 90 154 193 90 154 193 101 161 196 90 154 193 +101 161 196 101 161 196 101 161 196 101 161 196 101 161 196 101 161 196 +101 161 196 101 161 196 101 161 196 101 161 196 101 161 196 101 161 196 +136 185 209 101 161 196 101 161 196 136 185 209 136 185 209 101 161 196 +101 161 196 90 154 193 28 67 93 13 16 17 7 11 13 3 6 7 +37 51 59 125 124 125 163 162 163 174 174 174 167 166 167 166 165 166 +167 166 167 131 129 131 60 73 81 4 0 0 4 0 0 4 0 0 +3 3 3 5 5 5 6 6 6 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 0 0 4 0 0 41 54 63 137 136 137 153 152 153 137 136 137 +153 152 153 157 156 157 125 124 125 24 26 27 0 0 0 2 2 2 +4 4 4 4 4 4 2 0 0 0 0 0 28 67 93 90 154 193 +90 154 193 90 154 193 90 154 193 90 154 193 64 123 161 90 154 193 +90 154 193 90 154 193 90 154 193 90 154 193 90 154 193 101 161 196 +90 154 193 101 161 196 101 161 196 101 161 196 90 154 193 136 185 209 +101 161 196 101 161 196 136 185 209 101 161 196 136 185 209 101 161 196 +101 161 196 101 161 196 136 185 209 101 161 196 101 161 196 90 154 193 +35 83 115 13 16 17 3 6 7 2 5 5 13 16 17 60 74 84 +154 154 154 166 165 166 165 164 165 158 157 158 163 162 163 157 156 157 +60 74 84 13 16 17 4 0 0 4 0 0 3 2 2 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +1 1 1 4 0 0 41 54 63 157 156 157 155 154 155 137 136 137 +153 152 153 158 157 158 137 136 137 26 28 28 2 0 0 2 2 2 +4 4 4 4 4 4 1 0 0 6 10 14 34 86 122 90 154 193 +64 123 161 90 154 193 64 123 161 90 154 193 90 154 193 90 154 193 +64 123 161 90 154 193 90 154 193 90 154 193 90 154 193 90 154 193 +101 161 196 101 161 196 101 161 196 101 161 196 101 161 196 101 161 196 +101 161 196 101 161 196 101 161 196 101 161 196 101 161 196 101 161 196 +136 185 209 101 161 196 136 185 209 90 154 193 26 108 161 22 40 52 +13 16 17 5 7 8 2 5 5 2 5 5 37 38 37 165 164 165 +174 174 174 163 162 163 154 154 154 165 164 165 167 166 167 60 73 81 +6 6 6 4 0 0 4 0 0 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 0 0 6 6 6 41 54 63 156 155 156 158 157 158 153 152 153 +156 155 156 165 164 165 137 136 137 26 28 28 0 0 0 2 2 2 +4 4 5 4 4 4 2 0 0 7 12 15 31 96 139 64 123 161 +90 154 193 64 123 161 90 154 193 90 154 193 64 123 161 90 154 193 +90 154 193 90 154 193 90 154 193 90 154 193 90 154 193 90 154 193 +90 154 193 90 154 193 90 154 193 101 161 196 101 161 196 101 161 196 +101 161 196 101 161 196 101 161 196 101 161 196 101 161 196 136 185 209 +101 161 196 136 185 209 26 108 161 22 40 52 7 11 13 5 7 8 +2 5 5 2 5 5 2 5 5 2 2 1 37 38 37 158 157 158 +174 174 174 154 154 154 156 155 156 167 166 167 165 164 165 37 38 37 +4 0 0 4 3 3 5 5 5 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +3 1 0 4 0 0 60 73 81 157 156 157 163 162 163 153 152 153 +158 157 158 167 166 167 137 136 137 26 28 28 2 0 0 2 2 2 +4 5 5 4 4 4 4 0 0 7 12 15 24 86 132 26 108 161 +37 112 160 64 123 161 90 154 193 64 123 161 90 154 193 90 154 193 +90 154 193 90 154 193 90 154 193 90 154 193 90 154 193 90 154 193 +90 154 193 101 161 196 90 154 193 101 161 196 101 161 196 101 161 196 +101 161 196 101 161 196 101 161 196 136 185 209 101 161 196 136 185 209 +90 154 193 35 83 115 13 16 17 13 16 17 7 11 13 3 6 7 +5 7 8 6 6 6 3 4 3 2 2 1 30 32 34 154 154 154 +167 166 167 154 154 154 154 154 154 174 174 174 165 164 165 37 38 37 +6 6 6 4 0 0 6 6 6 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 0 0 4 0 0 41 54 63 163 162 163 166 165 166 154 154 154 +163 162 163 174 174 174 137 136 137 26 28 28 0 0 0 2 2 2 +4 5 5 4 4 5 1 1 2 6 10 14 28 67 93 18 97 151 +18 97 151 18 97 151 26 108 161 37 112 160 37 112 160 90 154 193 +64 123 161 90 154 193 90 154 193 90 154 193 90 154 193 101 161 196 +90 154 193 101 161 196 101 161 196 90 154 193 101 161 196 101 161 196 +101 161 196 101 161 196 101 161 196 136 185 209 90 154 193 16 89 141 +13 20 25 7 11 13 5 7 8 5 7 8 2 5 5 4 5 5 +3 4 3 4 5 5 3 4 3 0 0 0 37 38 37 158 157 158 +174 174 174 158 157 158 158 157 158 167 166 167 174 174 174 41 54 63 +4 0 0 3 2 2 5 5 5 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +1 1 1 4 0 0 60 73 81 165 164 165 174 174 174 158 157 158 +167 166 167 174 174 174 153 152 153 26 28 28 2 0 0 2 2 2 +4 5 5 4 4 4 4 0 0 7 12 15 10 87 144 10 87 144 +18 97 151 18 97 151 18 97 151 26 108 161 26 108 161 26 108 161 +26 108 161 37 112 160 53 118 160 90 154 193 90 154 193 90 154 193 +90 154 193 90 154 193 101 161 196 101 161 196 101 161 196 101 161 196 +101 161 196 136 185 209 90 154 193 26 108 161 22 40 52 13 16 17 +7 11 13 3 6 7 5 7 8 5 7 8 2 5 5 4 5 5 +4 5 5 6 6 6 3 4 3 0 0 0 30 32 34 158 157 158 +174 174 174 156 155 156 155 154 155 165 164 165 154 153 154 37 38 37 +4 0 0 4 3 3 5 5 5 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 0 0 4 0 0 60 73 81 167 166 167 174 174 174 163 162 163 +174 174 174 174 174 174 153 152 153 26 28 28 0 0 0 3 3 3 +5 5 5 4 4 4 1 1 2 7 12 15 28 67 93 18 97 151 +18 97 151 18 97 151 18 97 151 18 97 151 18 97 151 26 108 161 +26 108 161 26 108 161 26 108 161 26 108 161 26 108 161 26 108 161 +90 154 193 26 108 161 90 154 193 90 154 193 90 154 193 101 161 196 +101 161 196 26 108 161 22 40 52 13 16 17 7 11 13 2 5 5 +2 5 5 6 6 6 2 5 5 4 5 5 4 5 5 4 5 5 +3 4 3 5 5 5 3 4 3 2 0 0 30 32 34 137 136 137 +153 152 153 137 136 137 131 129 131 137 136 137 131 129 131 37 38 37 +4 0 0 4 3 3 5 5 5 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +1 1 1 4 0 0 60 73 81 167 166 167 174 174 174 166 165 166 +174 174 174 177 184 187 153 152 153 30 32 34 1 0 0 3 3 3 +5 5 5 4 3 3 4 0 0 7 12 15 10 87 144 10 87 144 +18 97 151 18 97 151 18 97 151 26 108 161 26 108 161 26 108 161 +26 108 161 26 108 161 26 108 161 26 108 161 26 108 161 26 108 161 +26 108 161 26 108 161 26 108 161 90 154 193 90 154 193 26 108 161 +35 83 115 13 16 17 7 11 13 5 7 8 3 6 7 5 7 8 +2 5 5 6 6 6 4 5 5 4 5 5 3 4 3 4 5 5 +3 4 3 6 6 6 3 4 3 0 0 0 26 28 28 125 124 125 +131 129 131 125 124 125 125 124 125 131 129 131 131 129 131 37 38 37 +4 0 0 3 3 3 5 5 5 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +3 1 0 4 0 0 60 73 81 174 174 174 177 184 187 167 166 167 +174 174 174 177 184 187 153 152 153 30 32 34 0 0 0 3 3 3 +5 5 5 4 4 4 1 1 2 6 10 14 28 67 93 18 97 151 +18 97 151 18 97 151 18 97 151 18 97 151 18 97 151 26 108 161 +26 108 161 26 108 161 26 108 161 26 108 161 26 108 161 26 108 161 +26 108 161 90 154 193 26 108 161 26 108 161 24 86 132 13 20 25 +7 11 13 13 20 25 22 40 52 5 7 8 3 4 3 3 4 3 +4 5 5 3 4 3 4 5 5 3 4 3 4 5 5 3 4 3 +4 4 4 5 5 5 3 3 3 2 0 0 26 28 28 125 124 125 +137 136 137 125 124 125 125 124 125 137 136 137 131 129 131 37 38 37 +0 0 0 3 3 3 5 5 5 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +1 1 1 4 0 0 60 73 81 174 174 174 177 184 187 174 174 174 +174 174 174 190 197 201 157 156 157 30 32 34 1 0 0 3 3 3 +5 5 5 4 3 3 4 0 0 7 12 15 10 87 144 10 87 144 +18 97 151 19 95 150 19 95 150 18 97 151 18 97 151 26 108 161 +18 97 151 26 108 161 26 108 161 26 108 161 26 108 161 90 154 193 +26 108 161 26 108 161 26 108 161 22 40 52 2 5 5 3 4 3 +28 67 93 37 112 160 34 86 122 2 5 5 3 4 3 3 4 3 +3 4 3 3 4 3 3 4 3 2 2 1 3 4 3 4 4 4 +4 5 5 5 5 5 3 3 3 0 0 0 26 28 28 131 129 131 +137 136 137 125 124 125 125 124 125 137 136 137 131 129 131 37 38 37 +0 0 0 3 3 3 5 5 5 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 0 0 4 0 0 60 73 81 174 174 174 177 184 187 174 174 174 +174 174 174 190 197 201 158 157 158 30 32 34 0 0 0 2 2 2 +5 5 5 4 4 4 1 1 2 6 10 14 28 67 93 18 97 151 +10 87 144 19 95 150 19 95 150 18 97 151 18 97 151 18 97 151 +26 108 161 26 108 161 26 108 161 26 108 161 26 108 161 26 108 161 +18 97 151 22 40 52 2 5 5 2 2 1 22 40 52 26 108 161 +90 154 193 37 112 160 22 40 52 3 4 3 13 20 25 22 30 35 +3 6 7 1 1 1 2 2 2 6 9 11 5 5 5 4 3 3 +4 4 4 5 5 5 3 3 3 2 0 0 26 28 28 131 129 131 +137 136 137 125 124 125 125 124 125 137 136 137 131 129 131 37 38 37 +0 0 0 3 3 3 5 5 5 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +1 1 1 4 0 0 60 73 81 177 184 187 193 200 203 174 174 174 +177 184 187 193 200 203 163 162 163 30 32 34 4 0 0 2 2 2 +5 5 5 4 3 3 4 0 0 6 10 14 24 86 132 10 87 144 +10 87 144 10 87 144 19 95 150 19 95 150 19 95 150 18 97 151 +26 108 161 26 108 161 26 108 161 90 154 193 26 108 161 28 67 93 +6 10 14 2 5 5 13 20 25 24 86 132 37 112 160 90 154 193 +10 87 144 7 12 15 2 5 5 28 67 93 37 112 160 28 67 93 +2 2 1 7 12 15 35 83 115 28 67 93 3 6 7 1 0 0 +4 4 4 5 5 5 3 3 3 0 0 0 26 28 28 131 129 131 +137 136 137 125 124 125 125 124 125 137 136 137 131 129 131 37 38 37 +0 0 0 3 3 3 5 5 5 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 0 0 4 0 0 60 73 81 174 174 174 190 197 201 174 174 174 +177 184 187 193 200 203 163 162 163 30 32 34 0 0 0 2 2 2 +5 5 5 4 4 4 1 1 2 6 10 14 28 67 93 10 87 144 +10 87 144 16 89 141 19 95 150 10 87 144 26 108 161 26 108 161 +26 108 161 26 108 161 26 108 161 28 67 93 6 10 14 1 1 2 +7 12 15 28 67 93 26 108 161 16 89 141 24 86 132 21 29 34 +3 4 3 21 29 34 37 112 160 37 112 160 27 99 146 21 29 34 +21 29 34 26 108 161 90 154 193 35 83 115 1 1 2 2 0 0 +4 4 4 5 5 5 3 3 3 2 0 0 26 28 28 125 124 125 +137 136 137 125 124 125 125 124 125 137 136 137 131 129 131 37 38 37 +0 0 0 3 3 3 5 5 5 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +3 1 0 4 0 0 60 73 81 193 200 203 193 200 203 174 174 174 +190 197 201 193 200 203 165 164 165 37 38 37 4 0 0 2 2 2 +5 5 5 4 3 3 4 0 0 6 10 14 24 86 132 10 87 144 +10 87 144 10 87 144 16 89 141 18 97 151 18 97 151 10 87 144 +24 86 132 24 86 132 13 20 25 4 5 7 4 5 7 22 40 52 +18 97 151 37 112 160 26 108 161 7 12 15 1 1 1 0 0 0 +28 67 93 37 112 160 26 108 161 28 67 93 22 40 52 28 67 93 +26 108 161 90 154 193 26 108 161 10 87 144 0 0 0 2 0 0 +4 4 4 5 5 5 3 3 3 0 0 0 26 28 28 131 129 131 +137 136 137 125 124 125 125 124 125 137 136 137 131 129 131 37 38 37 +0 0 0 3 3 3 5 5 5 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 0 0 6 6 6 60 73 81 174 174 174 193 200 203 174 174 174 +190 197 201 193 200 203 165 164 165 30 32 34 0 0 0 2 2 2 +5 5 5 4 4 4 1 1 2 6 10 14 28 67 93 10 87 144 +10 87 144 10 87 144 10 87 144 18 97 151 28 67 93 6 10 14 +0 0 0 1 1 2 4 5 7 13 20 25 16 89 141 26 108 161 +26 108 161 26 108 161 24 86 132 6 9 11 2 3 3 22 40 52 +37 112 160 16 89 141 22 40 52 28 67 93 26 108 161 26 108 161 +90 154 193 26 108 161 26 108 161 28 67 93 1 1 1 4 0 0 +4 4 4 5 5 5 3 3 3 4 0 0 26 28 28 124 126 130 +137 136 137 125 124 125 125 124 125 137 136 137 131 129 131 37 38 37 +0 0 0 3 3 3 5 5 5 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 0 0 4 0 0 60 73 81 193 200 203 193 200 203 174 174 174 +193 200 203 193 200 203 167 166 167 37 38 37 4 0 0 2 2 2 +5 5 5 4 4 4 4 0 0 6 10 14 28 67 93 10 87 144 +10 87 144 10 87 144 18 97 151 10 87 144 13 20 25 4 5 7 +1 1 2 1 1 1 22 40 52 26 108 161 26 108 161 26 108 161 +26 108 161 26 108 161 26 108 161 24 86 132 22 40 52 22 40 52 +22 40 52 22 40 52 10 87 144 26 108 161 26 108 161 26 108 161 +26 108 161 26 108 161 90 154 193 10 87 144 0 0 0 4 0 0 +4 4 4 5 5 5 3 3 3 0 0 0 26 28 28 131 129 131 +137 136 137 125 124 125 125 124 125 137 136 137 131 129 131 37 38 37 +0 0 0 3 3 3 5 5 5 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 0 0 6 6 6 60 73 81 174 174 174 220 221 221 174 174 174 +190 197 201 205 212 215 167 166 167 30 32 34 0 0 0 2 2 2 +5 5 5 4 4 4 1 1 2 6 10 14 28 67 93 10 87 144 +10 87 144 10 87 144 10 87 144 10 87 144 22 40 52 1 1 2 +2 0 0 1 1 2 24 86 132 26 108 161 26 108 161 26 108 161 +26 108 161 19 95 150 16 89 141 10 87 144 22 40 52 22 40 52 +10 87 144 26 108 161 37 112 160 26 108 161 26 108 161 26 108 161 +26 108 161 26 108 161 26 108 161 28 67 93 2 0 0 3 1 0 +4 4 4 5 5 5 3 3 3 2 0 0 26 28 28 131 129 131 +137 136 137 125 124 125 125 124 125 137 136 137 131 129 131 37 38 37 +0 0 0 3 3 3 5 5 5 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 0 0 4 0 0 60 73 81 220 221 221 190 197 201 174 174 174 +193 200 203 193 200 203 174 174 174 37 38 37 4 0 0 2 2 2 +5 5 5 4 4 4 3 2 2 1 1 2 13 20 25 10 87 144 +10 87 144 10 87 144 10 87 144 10 87 144 10 87 144 13 20 25 +13 20 25 22 40 52 10 87 144 18 97 151 18 97 151 26 108 161 +10 87 144 13 20 25 6 10 14 21 29 34 24 86 132 18 97 151 +26 108 161 26 108 161 26 108 161 26 108 161 26 108 161 26 108 161 +26 108 161 90 154 193 18 97 151 13 20 25 0 0 0 4 3 3 +4 4 4 5 5 5 3 3 3 0 0 0 26 28 28 131 129 131 +137 136 137 125 124 125 125 124 125 137 136 137 131 129 131 37 38 37 +0 0 0 3 3 3 5 5 5 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 0 0 6 6 6 60 73 81 174 174 174 220 221 221 174 174 174 +190 197 201 220 221 221 167 166 167 30 32 34 1 0 0 2 2 2 +5 5 5 4 4 4 4 4 5 2 5 5 4 5 7 13 20 25 +28 67 93 10 87 144 10 87 144 10 87 144 10 87 144 10 87 144 +10 87 144 10 87 144 18 97 151 10 87 144 18 97 151 18 97 151 +28 67 93 2 3 3 0 0 0 28 67 93 26 108 161 26 108 161 +26 108 161 26 108 161 26 108 161 26 108 161 26 108 161 26 108 161 +26 108 161 10 87 144 13 20 25 1 1 2 3 2 2 4 4 4 +4 4 4 5 5 5 3 3 3 2 0 0 26 28 28 131 129 131 +137 136 137 125 124 125 125 124 125 137 136 137 131 129 131 37 38 37 +0 0 0 3 3 3 5 5 5 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 0 0 4 0 0 60 73 81 220 221 221 190 197 201 174 174 174 +193 200 203 193 200 203 174 174 174 26 28 28 4 0 0 4 3 3 +5 5 5 4 4 4 4 4 4 4 4 5 1 1 2 2 5 5 +4 5 7 22 40 52 10 87 144 10 87 144 18 97 151 10 87 144 +10 87 144 10 87 144 10 87 144 10 87 144 10 87 144 18 97 151 +10 87 144 28 67 93 22 40 52 10 87 144 26 108 161 18 97 151 +18 97 151 18 97 151 26 108 161 26 108 161 26 108 161 26 108 161 +22 40 52 1 1 2 0 0 0 2 3 3 4 4 4 4 4 4 +4 4 4 5 5 5 4 4 4 0 0 0 26 28 28 131 129 131 +137 136 137 125 124 125 125 124 125 137 136 137 131 129 131 37 38 37 +0 0 0 3 3 3 5 5 5 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 0 0 6 6 6 60 73 81 174 174 174 220 221 221 174 174 174 +190 197 201 220 221 221 190 197 201 41 54 63 4 0 0 2 2 2 +6 6 6 4 4 4 4 4 4 4 4 5 4 4 5 3 3 3 +1 1 2 1 1 2 6 10 14 22 40 52 10 87 144 18 97 151 +18 97 151 10 87 144 10 87 144 10 87 144 18 97 151 10 87 144 +10 87 144 18 97 151 26 108 161 18 97 151 18 97 151 10 87 144 +26 108 161 26 108 161 26 108 161 10 87 144 28 67 93 6 10 14 +1 1 2 1 1 2 4 3 3 4 4 5 4 4 4 4 4 4 +5 5 5 5 5 5 1 1 1 4 0 0 37 51 59 137 136 137 +137 136 137 125 124 125 125 124 125 137 136 137 131 129 131 37 38 37 +0 0 0 3 3 3 5 5 5 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 0 0 4 0 0 60 73 81 220 221 221 193 200 203 174 174 174 +193 200 203 193 200 203 220 221 221 137 136 137 13 16 17 4 0 0 +2 2 2 4 4 4 4 4 4 4 4 4 4 4 4 4 4 5 +4 4 5 4 3 3 1 1 2 4 5 7 13 20 25 28 67 93 +10 87 144 10 87 144 10 87 144 10 87 144 10 87 144 10 87 144 +10 87 144 18 97 151 18 97 151 10 87 144 18 97 151 26 108 161 +26 108 161 18 97 151 28 67 93 6 10 14 0 0 0 0 0 0 +2 3 3 4 5 5 4 4 5 4 4 4 4 4 4 5 5 5 +3 3 3 1 1 1 0 0 0 16 19 21 125 124 125 137 136 137 +131 129 131 125 124 125 125 124 125 137 136 137 131 129 131 37 38 37 +0 0 0 3 3 3 5 5 5 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 0 0 6 6 6 60 73 81 174 174 174 220 221 221 174 174 174 +193 200 203 190 197 201 220 221 221 220 221 221 153 152 153 30 32 34 +0 0 0 0 0 0 2 2 2 4 4 4 4 4 4 4 4 4 +4 4 4 4 5 5 4 5 7 1 1 2 1 1 2 4 5 7 +13 20 25 28 67 93 10 87 144 18 97 151 10 87 144 10 87 144 +10 87 144 10 87 144 10 87 144 18 97 151 26 108 161 18 97 151 +28 67 93 7 12 15 0 0 0 0 0 0 2 2 1 4 4 4 +4 5 5 4 5 5 4 4 4 4 4 4 3 3 3 0 0 0 +0 0 0 0 0 0 37 38 37 125 124 125 158 157 158 131 129 131 +125 124 125 125 124 125 125 124 125 137 136 137 131 129 131 37 38 37 +0 0 0 3 3 3 5 5 5 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 3 3 4 0 0 41 54 63 193 200 203 220 221 221 174 174 174 +193 200 203 193 200 203 193 200 203 220 221 221 244 246 246 193 200 203 +120 125 127 5 5 5 1 0 0 0 0 0 1 1 1 4 4 4 +4 4 4 4 4 4 4 5 5 4 5 5 4 4 5 1 1 2 +4 5 7 4 5 7 22 40 52 10 87 144 10 87 144 10 87 144 +10 87 144 10 87 144 18 97 151 10 87 144 10 87 144 13 20 25 +4 5 7 2 3 3 1 1 2 4 4 4 4 5 5 4 4 4 +4 4 4 4 4 4 4 4 4 1 1 1 0 0 0 1 1 2 +24 26 27 60 74 84 153 152 153 163 162 163 137 136 137 125 124 125 +125 124 125 125 124 125 125 124 125 137 136 137 125 124 125 26 28 28 +0 0 0 3 3 3 5 5 5 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 0 0 6 6 6 26 28 28 156 155 156 220 221 221 220 221 221 +174 174 174 193 200 203 193 200 203 193 200 203 205 212 215 220 221 221 +220 221 221 167 166 167 60 73 81 7 11 13 0 0 0 0 0 0 +3 3 3 4 4 4 4 4 4 4 4 4 4 4 5 4 4 5 +4 4 5 1 1 2 1 1 2 4 5 7 22 40 52 10 87 144 +10 87 144 10 87 144 10 87 144 22 40 52 4 5 7 1 1 2 +1 1 2 4 4 5 4 4 4 4 4 4 4 4 4 4 4 4 +5 5 5 2 2 2 0 0 0 4 0 0 16 19 21 60 73 81 +137 136 137 167 166 167 158 157 158 137 136 137 131 129 131 131 129 131 +125 124 125 125 124 125 131 129 131 155 154 155 60 74 84 5 7 8 +0 0 0 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +5 5 5 4 0 0 4 0 0 60 73 81 193 200 203 220 221 221 +193 200 203 193 200 203 193 200 203 193 200 203 205 212 215 220 221 221 +220 221 221 220 221 221 220 221 221 137 136 137 43 57 68 6 6 6 +4 0 0 1 1 1 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 5 4 4 5 3 2 2 1 1 2 2 5 5 13 20 25 +22 40 52 22 40 52 13 20 25 2 3 3 1 1 2 3 3 3 +4 5 7 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +1 1 1 0 0 0 2 3 3 41 54 63 131 129 131 166 165 166 +166 165 166 155 154 155 153 152 153 137 136 137 137 136 137 125 124 125 +125 124 125 137 136 137 137 136 137 125 124 125 37 38 37 4 3 3 +4 3 3 5 5 5 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 3 3 6 6 6 6 6 6 13 16 17 60 73 81 167 166 167 +220 221 221 220 221 221 220 221 221 193 200 203 193 200 203 193 200 203 +205 212 215 220 221 221 220 221 221 244 246 246 205 212 215 125 124 125 +24 26 27 0 0 0 0 0 0 2 2 2 5 5 5 5 5 5 +4 4 4 4 4 4 4 4 4 4 4 5 1 1 2 4 5 7 +4 5 7 4 5 7 1 1 2 3 2 2 4 4 5 4 4 4 +4 4 4 4 4 4 5 5 5 4 4 4 0 0 0 0 0 0 +2 0 0 26 28 28 125 124 125 174 174 174 174 174 174 166 165 166 +156 155 156 153 152 153 137 136 137 137 136 137 131 129 131 137 136 137 +137 136 137 137 136 137 60 74 84 30 32 34 4 0 0 4 0 0 +5 5 5 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +5 5 5 6 6 6 4 0 0 4 0 0 6 6 6 26 28 28 +125 124 125 174 174 174 220 221 221 220 221 221 220 221 221 193 200 203 +205 212 215 220 221 221 205 212 215 220 221 221 220 221 221 244 246 246 +193 200 203 60 74 84 13 16 17 4 0 0 0 0 0 3 3 3 +5 5 5 5 5 5 4 4 4 4 4 4 4 4 5 3 3 3 +1 1 2 3 3 3 4 4 5 4 4 5 4 4 4 4 4 4 +5 5 5 5 5 5 2 2 2 0 0 0 0 0 0 13 16 17 +60 74 84 174 174 174 193 200 203 174 174 174 167 166 167 163 162 163 +153 152 153 153 152 153 137 136 137 137 136 137 153 152 153 137 136 137 +125 124 125 41 54 63 24 26 27 4 0 0 4 0 0 5 5 5 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 3 3 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 +6 6 6 37 38 37 131 129 131 220 221 221 220 221 221 220 221 221 +193 200 203 193 200 203 220 221 221 205 212 215 220 221 221 244 246 246 +244 246 246 244 246 246 174 174 174 41 54 63 0 0 0 0 0 0 +0 0 0 4 4 4 5 5 5 5 5 5 4 4 4 4 4 5 +4 4 5 4 4 5 4 4 4 4 4 4 6 6 6 6 6 6 +3 3 3 0 0 0 2 0 0 13 16 17 60 73 81 156 155 156 +220 221 221 193 200 203 174 174 174 165 164 165 163 162 163 154 153 154 +153 152 153 153 152 153 158 157 158 163 162 163 137 136 137 60 73 81 +13 16 17 4 0 0 4 0 0 4 3 3 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +5 5 5 4 3 3 4 3 3 6 6 6 6 6 6 6 6 6 +6 6 6 6 6 6 6 6 6 37 38 37 167 166 167 244 246 246 +244 246 246 220 221 221 205 212 215 205 212 215 220 221 221 193 200 203 +220 221 221 244 246 246 244 246 246 244 246 246 137 136 137 37 38 37 +3 2 2 0 0 0 1 1 1 5 5 5 5 5 5 4 4 4 +4 4 4 4 4 4 4 4 4 5 5 5 4 4 4 1 1 1 +0 0 0 5 5 5 43 57 68 153 152 153 193 200 203 220 221 221 +177 184 187 174 174 174 167 166 167 166 165 166 158 157 158 157 156 157 +158 157 158 166 165 166 156 155 156 85 115 134 13 16 17 4 0 0 +4 0 0 4 0 0 5 5 5 5 5 5 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +5 5 5 4 3 3 6 6 6 6 6 6 4 0 0 6 6 6 +6 6 6 6 6 6 6 6 6 6 6 6 13 16 17 60 73 81 +177 184 187 220 221 221 220 221 221 220 221 221 205 212 215 220 221 221 +220 221 221 205 212 215 220 221 221 244 246 246 244 246 246 205 212 215 +125 124 125 30 32 34 0 0 0 0 0 0 2 2 2 5 5 5 +4 4 4 4 4 4 4 4 4 1 1 1 0 0 0 1 0 0 +37 38 37 131 129 131 205 212 215 220 221 221 193 200 203 174 174 174 +174 174 174 174 174 174 167 166 167 165 164 165 166 165 166 167 166 167 +158 157 158 125 124 125 37 38 37 4 0 0 4 0 0 4 0 0 +4 3 3 5 5 5 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 4 4 5 5 5 4 3 3 4 3 3 6 6 6 6 6 6 +4 0 0 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 +26 28 28 125 124 125 205 212 215 220 221 221 220 221 221 220 221 221 +205 212 215 220 221 221 205 212 215 220 221 221 220 221 221 244 246 246 +244 246 246 190 197 201 60 74 84 16 19 21 4 0 0 0 0 0 +0 0 0 0 0 0 0 0 0 0 0 0 16 19 21 120 125 127 +177 184 187 220 221 221 205 212 215 177 184 187 174 174 174 177 184 187 +174 174 174 174 174 174 167 166 167 174 174 174 166 165 166 137 136 137 +60 73 81 13 16 17 4 0 0 4 0 0 4 3 3 6 6 6 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +5 5 5 4 3 3 5 5 5 4 3 3 6 6 6 4 0 0 +6 6 6 6 6 6 4 0 0 6 6 6 4 0 0 6 6 6 +6 6 6 6 6 6 37 38 37 137 136 137 193 200 203 220 221 221 +220 221 221 205 212 215 220 221 221 205 212 215 205 212 215 220 221 221 +220 221 221 220 221 221 244 246 246 166 165 166 43 57 68 2 2 2 +0 0 0 4 0 0 16 19 21 60 73 81 157 156 157 202 210 214 +220 221 221 193 200 203 177 184 187 177 184 187 177 184 187 174 174 174 +174 174 174 174 174 174 174 174 174 157 156 157 60 74 84 24 26 27 +4 0 0 4 0 0 4 0 0 6 6 6 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 4 4 4 4 4 5 5 5 4 3 3 5 5 5 6 6 6 +6 6 6 4 0 0 6 6 6 6 6 6 6 6 6 4 0 0 +4 0 0 4 0 0 6 6 6 24 26 27 60 73 81 167 166 167 +220 221 221 220 221 221 220 221 221 205 212 215 205 212 215 205 212 215 +205 212 215 220 221 221 220 221 221 220 221 221 205 212 215 137 136 137 +60 74 84 125 124 125 137 136 137 190 197 201 220 221 221 193 200 203 +177 184 187 177 184 187 177 184 187 174 174 174 174 174 174 177 184 187 +190 197 201 174 174 174 125 124 125 37 38 37 6 6 6 4 0 0 +4 0 0 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 4 4 4 4 4 5 5 5 5 5 5 4 3 3 6 6 6 +4 0 0 6 6 6 6 6 6 6 6 6 4 0 0 6 6 6 +6 6 6 6 6 6 4 0 0 4 0 0 6 6 6 6 6 6 +125 124 125 193 200 203 244 246 246 220 221 221 205 212 215 205 212 215 +205 212 215 193 200 203 205 212 215 205 212 215 220 221 221 220 221 221 +193 200 203 193 200 203 205 212 215 193 200 203 193 200 203 177 184 187 +190 197 201 190 197 201 174 174 174 190 197 201 193 200 203 190 197 201 +153 152 153 60 73 81 4 0 0 4 0 0 4 0 0 3 2 2 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 5 5 5 4 3 3 +6 6 6 4 3 3 4 3 3 4 3 3 6 6 6 6 6 6 +4 0 0 6 6 6 6 6 6 6 6 6 4 0 0 4 0 0 +4 0 0 26 28 28 131 129 131 220 221 221 244 246 246 220 221 221 +205 212 215 193 200 203 205 212 215 193 200 203 193 200 203 205 212 215 +220 221 221 193 200 203 193 200 203 193 200 203 190 197 201 174 174 174 +174 174 174 190 197 201 193 200 203 193 200 203 167 166 167 125 124 125 +6 6 6 4 0 0 4 0 0 4 3 3 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 5 5 5 +5 5 5 4 3 3 5 5 5 6 6 6 4 3 3 5 5 5 +6 6 6 6 6 6 4 0 0 6 6 6 6 6 6 6 6 6 +4 0 0 4 0 0 6 6 6 41 54 63 158 157 158 220 221 221 +220 221 221 220 221 221 193 200 203 193 200 203 193 200 203 190 197 201 +190 197 201 190 197 201 190 197 201 190 197 201 174 174 174 193 200 203 +193 200 203 220 221 221 174 174 174 125 124 125 37 38 37 4 0 0 +4 0 0 4 3 3 6 6 6 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 5 5 5 4 3 3 4 3 3 4 3 3 5 5 5 +4 3 3 6 6 6 5 5 5 4 3 3 6 6 6 6 6 6 +6 6 6 6 6 6 4 0 0 4 0 0 13 16 17 60 73 81 +174 174 174 220 221 221 220 221 221 205 212 215 190 197 201 174 174 174 +193 200 203 174 174 174 190 197 201 174 174 174 193 200 203 220 221 221 +193 200 203 131 129 131 37 38 37 6 6 6 4 0 0 4 0 0 +6 6 6 6 6 6 4 3 3 5 5 5 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 5 5 5 5 5 5 5 5 5 +5 5 5 4 3 3 4 3 3 5 5 5 4 3 3 4 3 3 +5 5 5 6 6 6 6 6 6 4 0 0 6 6 6 6 6 6 +6 6 6 125 124 125 174 174 174 220 221 221 220 221 221 193 200 203 +193 200 203 193 200 203 193 200 203 193 200 203 220 221 221 158 157 158 +60 73 81 6 6 6 4 0 0 4 0 0 5 5 5 6 6 6 +5 5 5 5 5 5 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 5 5 5 5 5 5 4 3 3 5 5 5 4 3 3 +5 5 5 5 5 5 6 6 6 6 6 6 4 0 0 4 0 0 +4 0 0 4 0 0 26 28 28 125 124 125 174 174 174 193 200 203 +193 200 203 174 174 174 193 200 203 167 166 167 125 124 125 6 6 6 +6 6 6 6 6 6 4 0 0 6 6 6 6 6 6 5 5 5 +4 3 3 5 5 5 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 5 5 5 +4 3 3 6 6 6 4 0 0 6 6 6 6 6 6 6 6 6 +6 6 6 4 0 0 4 0 0 6 6 6 37 38 37 125 124 125 +153 152 153 131 129 131 125 124 125 37 38 37 6 6 6 6 6 6 +6 6 6 4 0 0 6 6 6 6 6 6 4 3 3 5 5 5 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 5 5 5 5 5 5 4 3 3 5 5 5 4 3 3 +6 6 6 6 6 6 4 0 0 4 0 0 6 6 6 6 6 6 +24 26 27 24 26 27 6 6 6 6 6 6 6 6 6 4 0 0 +6 6 6 6 6 6 4 0 0 6 6 6 5 5 5 4 3 3 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 5 5 5 4 3 3 5 5 5 6 6 6 +4 0 0 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 +6 6 6 6 6 6 6 6 6 4 0 0 6 6 6 6 6 6 +4 0 0 6 6 6 6 6 6 4 3 3 5 5 5 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 5 5 5 4 3 3 5 5 5 +5 5 5 5 5 5 4 0 0 6 6 6 4 0 0 6 6 6 +6 6 6 6 6 6 6 6 6 4 0 0 6 6 6 4 0 0 +6 6 6 4 3 3 5 5 5 4 3 3 5 5 5 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 5 5 5 +4 3 3 6 6 6 4 3 3 6 6 6 6 6 6 6 6 6 +4 0 0 6 6 6 4 0 0 6 6 6 6 6 6 6 6 6 +6 6 6 4 3 3 5 5 5 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 5 5 5 4 3 3 5 5 5 4 0 0 6 6 6 +6 6 6 4 0 0 6 6 6 6 6 6 4 0 0 6 6 6 +4 3 3 5 5 5 5 5 5 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 5 5 5 4 3 3 5 5 5 6 6 6 4 3 3 +4 3 3 6 6 6 6 6 6 4 3 3 6 6 6 4 3 3 +5 5 5 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 5 5 5 4 3 3 6 6 6 +5 5 5 4 3 3 4 3 3 4 3 3 5 5 5 5 5 5 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 5 5 5 4 3 3 +5 5 5 4 3 3 5 5 5 5 5 5 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 +4 4 4 4 4 4 diff -urNp linux-2.6.32.48/drivers/video/nvidia/nv_backlight.c linux-2.6.32.48-openvz/drivers/video/nvidia/nv_backlight.c --- linux-2.6.32.48/drivers/video/nvidia/nv_backlight.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/video/nvidia/nv_backlight.c 2011-11-17 20:08:54.000000000 -0500 @@ -87,7 +87,7 @@ static int nvidia_bl_get_brightness(stru return bd->props.brightness; } -static struct backlight_ops nvidia_bl_ops = { +static const struct backlight_ops nvidia_bl_ops = { .get_brightness = nvidia_bl_get_brightness, .update_status = nvidia_bl_update_status, }; diff -urNp linux-2.6.32.48/drivers/video/riva/fbdev.c linux-2.6.32.48-openvz/drivers/video/riva/fbdev.c --- linux-2.6.32.48/drivers/video/riva/fbdev.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/video/riva/fbdev.c 2011-11-17 20:08:54.000000000 -0500 @@ -331,7 +331,7 @@ static int riva_bl_get_brightness(struct return bd->props.brightness; } -static struct backlight_ops riva_bl_ops = { +static const struct backlight_ops riva_bl_ops = { .get_brightness = riva_bl_get_brightness, .update_status = riva_bl_update_status, }; diff -urNp linux-2.6.32.48/drivers/video/uvesafb.c linux-2.6.32.48-openvz/drivers/video/uvesafb.c --- linux-2.6.32.48/drivers/video/uvesafb.c 2011-11-08 19:02:43.000000000 -0500 +++ linux-2.6.32.48-openvz/drivers/video/uvesafb.c 2011-11-17 20:08:54.000000000 -0500 @@ -18,6 +18,7 @@ #include #include #include +#include #include