From b534816b552d35bbd3c60702139ed5c7da2f55c2 Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Wed, 4 Feb 2009 18:33:38 -0800 Subject: x86: don't apply __supported_pte_mask to non-present ptes On an x86 system which doesn't support global mappings, __supported_pte_mask has _PAGE_GLOBAL clear, to make sure it never appears in the PTE. pfn_pte() and so on will enforce it with: static inline pte_t pfn_pte(unsigned long page_nr, pgprot_t pgprot) { return __pte((((phys_addr_t)page_nr << PAGE_SHIFT) | pgprot_val(pgprot)) & __supported_pte_mask); } However, we overload _PAGE_GLOBAL with _PAGE_PROTNONE on non-present ptes to distinguish them from swap entries. However, applying __supported_pte_mask indiscriminately will clear the bit and corrupt the pte. I guess the best fix is to only apply __supported_pte_mask to present ptes. This seems like the right solution to me, as it means we can completely ignore the issue of overlaps between the present pte bits and the non-present pte-as-swap entry use of the bits. __supported_pte_mask contains the set of flags we support on the current hardware. We also use bits in the pte for things like logically present ptes with no permissions, and swap entries for swapped out pages. We should only apply __supported_pte_mask to present ptes, because otherwise we may destroy other information being stored in the ptes. Signed-off-by: Jeremy Fitzhardinge Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/pgtable.h | 26 ++++++++++++++++++++------ arch/x86/include/asm/xen/page.h | 2 +- 2 files changed, 21 insertions(+), 7 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/pgtable.h b/arch/x86/include/asm/pgtable.h index 06bbcbd66e9c..4f5af8447d54 100644 --- a/arch/x86/include/asm/pgtable.h +++ b/arch/x86/include/asm/pgtable.h @@ -302,16 +302,30 @@ static inline pte_t pte_mkspecial(pte_t pte) extern pteval_t __supported_pte_mask; +/* + * Mask out unsupported bits in a present pgprot. Non-present pgprots + * can use those bits for other purposes, so leave them be. + */ +static inline pgprotval_t massage_pgprot(pgprot_t pgprot) +{ + pgprotval_t protval = pgprot_val(pgprot); + + if (protval & _PAGE_PRESENT) + protval &= __supported_pte_mask; + + return protval; +} + static inline pte_t pfn_pte(unsigned long page_nr, pgprot_t pgprot) { - return __pte((((phys_addr_t)page_nr << PAGE_SHIFT) | - pgprot_val(pgprot)) & __supported_pte_mask); + return __pte(((phys_addr_t)page_nr << PAGE_SHIFT) | + massage_pgprot(pgprot)); } static inline pmd_t pfn_pmd(unsigned long page_nr, pgprot_t pgprot) { - return __pmd((((phys_addr_t)page_nr << PAGE_SHIFT) | - pgprot_val(pgprot)) & __supported_pte_mask); + return __pmd(((phys_addr_t)page_nr << PAGE_SHIFT) | + massage_pgprot(pgprot)); } static inline pte_t pte_modify(pte_t pte, pgprot_t newprot) @@ -323,7 +337,7 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot) * the newprot (if present): */ val &= _PAGE_CHG_MASK; - val |= pgprot_val(newprot) & (~_PAGE_CHG_MASK) & __supported_pte_mask; + val |= massage_pgprot(newprot) & ~_PAGE_CHG_MASK; return __pte(val); } @@ -339,7 +353,7 @@ static inline pgprot_t pgprot_modify(pgprot_t oldprot, pgprot_t newprot) #define pte_pgprot(x) __pgprot(pte_flags(x) & PTE_FLAGS_MASK) -#define canon_pgprot(p) __pgprot(pgprot_val(p) & __supported_pte_mask) +#define canon_pgprot(p) __pgprot(massage_pgprot(p)) static inline int is_new_memtype_allowed(unsigned long flags, unsigned long new_flags) diff --git a/arch/x86/include/asm/xen/page.h b/arch/x86/include/asm/xen/page.h index 7ef617ef1df3..4bd990ee43df 100644 --- a/arch/x86/include/asm/xen/page.h +++ b/arch/x86/include/asm/xen/page.h @@ -137,7 +137,7 @@ static inline pte_t mfn_pte(unsigned long page_nr, pgprot_t pgprot) pte_t pte; pte.pte = ((phys_addr_t)page_nr << PAGE_SHIFT) | - (pgprot_val(pgprot) & __supported_pte_mask); + massage_pgprot(pgprot); return pte; } -- cgit v1.2.3 From e736ad548db152776de61d7a26805cfae77ce5ce Mon Sep 17 00:00:00 2001 From: "Pallipadi, Venkatesh" Date: Fri, 6 Feb 2009 16:52:05 -0800 Subject: x86: add clflush before monitor for Intel 7400 series For Intel 7400 series CPUs, the recommendation is to use a clflush on the monitored address just before monitor and mwait pair [1]. This clflush makes sure that there are no false wakeups from mwait when the monitored address was recently written to. [1] "MONITOR/MWAIT Recommendations for Intel Xeon Processor 7400 series" section in specification update document of 7400 series http://download.intel.com/design/xeon/specupdt/32033601.pdf Signed-off-by: Venkatesh Pallipadi Signed-off-by: Ingo Molnar --- arch/x86/include/asm/cpufeature.h | 1 + arch/x86/kernel/cpu/intel.c | 3 +++ arch/x86/kernel/process.c | 6 ++++++ 3 files changed, 10 insertions(+) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/cpufeature.h b/arch/x86/include/asm/cpufeature.h index ea408dcba513..7301e60dc4a8 100644 --- a/arch/x86/include/asm/cpufeature.h +++ b/arch/x86/include/asm/cpufeature.h @@ -93,6 +93,7 @@ #define X86_FEATURE_XTOPOLOGY (3*32+22) /* cpu topology enum extensions */ #define X86_FEATURE_TSC_RELIABLE (3*32+23) /* TSC is known to be reliable */ #define X86_FEATURE_NONSTOP_TSC (3*32+24) /* TSC does not stop in C states */ +#define X86_FEATURE_CLFLUSH_MONITOR (3*32+25) /* "" clflush reqd with monitor */ /* Intel-defined CPU features, CPUID level 0x00000001 (ecx), word 4 */ #define X86_FEATURE_XMM3 (4*32+ 0) /* "pni" SSE-3 */ diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c index 430e5c38a544..24ff26a38ade 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -291,6 +291,9 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c) ds_init_intel(c); } + if (c->x86 == 6 && c->x86_model == 29 && cpu_has_clflush) + set_cpu_cap(c, X86_FEATURE_CLFLUSH_MONITOR); + #ifdef CONFIG_X86_64 if (c->x86 == 15) c->x86_cache_alignment = c->x86_clflush_size * 2; diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index e68bb9e30864..6d12f7e37f8c 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -180,6 +180,9 @@ void mwait_idle_with_hints(unsigned long ax, unsigned long cx) trace_power_start(&it, POWER_CSTATE, (ax>>4)+1); if (!need_resched()) { + if (cpu_has(¤t_cpu_data, X86_FEATURE_CLFLUSH_MONITOR)) + clflush((void *)¤t_thread_info()->flags); + __monitor((void *)¤t_thread_info()->flags, 0, 0); smp_mb(); if (!need_resched()) @@ -194,6 +197,9 @@ static void mwait_idle(void) struct power_trace it; if (!need_resched()) { trace_power_start(&it, POWER_CSTATE, 1); + if (cpu_has(¤t_cpu_data, X86_FEATURE_CLFLUSH_MONITOR)) + clflush((void *)¤t_thread_info()->flags); + __monitor((void *)¤t_thread_info()->flags, 0, 0); smp_mb(); if (!need_resched()) -- cgit v1.2.3 From 3f4a739c6accd651a11fcf3c7a20ec8147c42660 Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Sun, 8 Feb 2009 16:18:03 -0800 Subject: x86: find nr_irqs_gsi with mp_ioapic_routing Impact: find right nr_irqs_gsi on some systems. One test-system has gap between gsi's: [ 0.000000] ACPI: IOAPIC (id[0x04] address[0xfec00000] gsi_base[0]) [ 0.000000] IOAPIC[0]: apic_id 4, version 0, address 0xfec00000, GSI 0-23 [ 0.000000] ACPI: IOAPIC (id[0x05] address[0xfeafd000] gsi_base[48]) [ 0.000000] IOAPIC[1]: apic_id 5, version 0, address 0xfeafd000, GSI 48-54 [ 0.000000] ACPI: IOAPIC (id[0x06] address[0xfeafc000] gsi_base[56]) [ 0.000000] IOAPIC[2]: apic_id 6, version 0, address 0xfeafc000, GSI 56-62 ... [ 0.000000] nr_irqs_gsi: 38 So nr_irqs_gsi is not right. some irq for MSI will overwrite with io_apic. need to get that with acpi_probe_gsi when acpi io_apic is used Signed-off-by: Yinghai Lu Signed-off-by: Ingo Molnar --- arch/x86/include/asm/mpspec.h | 6 ++++++ arch/x86/kernel/acpi/boot.c | 23 +++++++++++++++++++++++ arch/x86/kernel/io_apic.c | 20 +++++++++++++++----- 3 files changed, 44 insertions(+), 5 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/mpspec.h b/arch/x86/include/asm/mpspec.h index 62d14ce3cd00..bd22f2a3713f 100644 --- a/arch/x86/include/asm/mpspec.h +++ b/arch/x86/include/asm/mpspec.h @@ -60,6 +60,7 @@ extern void mp_override_legacy_irq(u8 bus_irq, u8 polarity, u8 trigger, u32 gsi); extern void mp_config_acpi_legacy_irqs(void); extern int mp_register_gsi(u32 gsi, int edge_level, int active_high_low); +extern int acpi_probe_gsi(void); #ifdef CONFIG_X86_IO_APIC extern int mp_config_acpi_gsi(unsigned char number, unsigned int devfn, u8 pin, u32 gsi, int triggering, int polarity); @@ -71,6 +72,11 @@ mp_config_acpi_gsi(unsigned char number, unsigned int devfn, u8 pin, return 0; } #endif +#else /* !CONFIG_ACPI: */ +static inline int acpi_probe_gsi(void) +{ + return 0; +} #endif /* CONFIG_ACPI */ #define PHYSID_ARRAY_SIZE BITS_TO_LONGS(MAX_APICS) diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c index d37593c2f438..7678f10c4568 100644 --- a/arch/x86/kernel/acpi/boot.c +++ b/arch/x86/kernel/acpi/boot.c @@ -973,6 +973,29 @@ void __init mp_register_ioapic(int id, u32 address, u32 gsi_base) nr_ioapics++; } +int __init acpi_probe_gsi(void) +{ + int idx; + int gsi; + int max_gsi = 0; + + if (acpi_disabled) + return 0; + + if (!acpi_ioapic) + return 0; + + max_gsi = 0; + for (idx = 0; idx < nr_ioapics; idx++) { + gsi = mp_ioapic_routing[idx].gsi_end; + + if (gsi > max_gsi) + max_gsi = gsi; + } + + return max_gsi + 1; +} + static void assign_to_mp_irq(struct mp_config_intsrc *m, struct mp_config_intsrc *mp_irq) { diff --git a/arch/x86/kernel/io_apic.c b/arch/x86/kernel/io_apic.c index 9b0c480c383b..bc7ac4da90d7 100644 --- a/arch/x86/kernel/io_apic.c +++ b/arch/x86/kernel/io_apic.c @@ -3841,14 +3841,24 @@ int __init io_apic_get_redir_entries (int ioapic) void __init probe_nr_irqs_gsi(void) { - int idx; int nr = 0; - for (idx = 0; idx < nr_ioapics; idx++) - nr += io_apic_get_redir_entries(idx) + 1; - - if (nr > nr_irqs_gsi) + nr = acpi_probe_gsi(); + if (nr > nr_irqs_gsi) { nr_irqs_gsi = nr; + } else { + /* for acpi=off or acpi is not compiled in */ + int idx; + + nr = 0; + for (idx = 0; idx < nr_ioapics; idx++) + nr += io_apic_get_redir_entries(idx) + 1; + + if (nr > nr_irqs_gsi) + nr_irqs_gsi = nr; + } + + printk(KERN_DEBUG "nr_irqs_gsi: %d\n", nr_irqs_gsi); } /* -------------------------------------------------------------------------- -- cgit v1.2.3 From 914c3d630b29b07d04908eab1b246812dadd5bd6 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 9 Feb 2009 22:17:39 +0900 Subject: x86: include correct %gs in a.out core dump Impact: dump the correct %gs into a.out core dump aout_dump_thread() read %gs but didn't include it in core dump. Fix it. Signed-off-by: Tejun Heo Signed-off-by: Ingo Molnar --- arch/x86/include/asm/a.out-core.h | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/a.out-core.h b/arch/x86/include/asm/a.out-core.h index 37822206083e..3c601f8224be 100644 --- a/arch/x86/include/asm/a.out-core.h +++ b/arch/x86/include/asm/a.out-core.h @@ -23,8 +23,6 @@ */ static inline void aout_dump_thread(struct pt_regs *regs, struct user *dump) { - u16 gs; - /* changed the size calculations - should hopefully work better. lbt */ dump->magic = CMAGIC; dump->start_code = 0; @@ -57,7 +55,7 @@ static inline void aout_dump_thread(struct pt_regs *regs, struct user *dump) dump->regs.ds = (u16)regs->ds; dump->regs.es = (u16)regs->es; dump->regs.fs = (u16)regs->fs; - savesegment(gs, gs); + savesegment(gs, dump->regs.gs); dump->regs.orig_ax = regs->orig_ax; dump->regs.ip = regs->ip; dump->regs.cs = (u16)regs->cs; -- cgit v1.2.3 From ae6af41f5a4841f06eb92bc86ad020ad44ae2a30 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 9 Feb 2009 22:17:39 +0900 Subject: x86: math_emu info cleanup Impact: cleanup * Come on, struct info? s/struct info/struct math_emu_info/ * Use struct pt_regs and kernel_vm86_regs instead of defining its own register frame structure. Signed-off-by: Tejun Heo Signed-off-by: Ingo Molnar --- arch/x86/include/asm/math_emu.h | 29 +++++------------- arch/x86/include/asm/processor.h | 2 +- arch/x86/math-emu/fpu_entry.c | 2 +- arch/x86/math-emu/fpu_proto.h | 2 +- arch/x86/math-emu/fpu_system.h | 14 ++++----- arch/x86/math-emu/get_address.c | 63 +++++++++++++++++++--------------------- 6 files changed, 48 insertions(+), 64 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/math_emu.h b/arch/x86/include/asm/math_emu.h index 5a65b107ad58..302492c77956 100644 --- a/arch/x86/include/asm/math_emu.h +++ b/arch/x86/include/asm/math_emu.h @@ -1,31 +1,18 @@ #ifndef _ASM_X86_MATH_EMU_H #define _ASM_X86_MATH_EMU_H +#include +#include + /* This structure matches the layout of the data saved to the stack following a device-not-present interrupt, part of it saved automatically by the 80386/80486. */ -struct info { +struct math_emu_info { long ___orig_eip; - long ___ebx; - long ___ecx; - long ___edx; - long ___esi; - long ___edi; - long ___ebp; - long ___eax; - long ___ds; - long ___es; - long ___fs; - long ___orig_eax; - long ___eip; - long ___cs; - long ___eflags; - long ___esp; - long ___ss; - long ___vm86_es; /* This and the following only in vm86 mode */ - long ___vm86_ds; - long ___vm86_fs; - long ___vm86_gs; + union { + struct pt_regs regs; + struct kernel_vm86_regs vm86; + }; }; #endif /* _ASM_X86_MATH_EMU_H */ diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h index 091cd8855f2e..3bfd5235a9eb 100644 --- a/arch/x86/include/asm/processor.h +++ b/arch/x86/include/asm/processor.h @@ -353,7 +353,7 @@ struct i387_soft_struct { u8 no_update; u8 rm; u8 alimit; - struct info *info; + struct math_emu_info *info; u32 entry_eip; }; diff --git a/arch/x86/math-emu/fpu_entry.c b/arch/x86/math-emu/fpu_entry.c index c7b06feb139b..c268abe72253 100644 --- a/arch/x86/math-emu/fpu_entry.c +++ b/arch/x86/math-emu/fpu_entry.c @@ -659,7 +659,7 @@ static int valid_prefix(u_char *Byte, u_char __user **fpu_eip, } } -void math_abort(struct info *info, unsigned int signal) +void math_abort(struct math_emu_info *info, unsigned int signal) { FPU_EIP = FPU_ORIG_EIP; current->thread.trap_no = 16; diff --git a/arch/x86/math-emu/fpu_proto.h b/arch/x86/math-emu/fpu_proto.h index aa49b6a0d850..51bfbb61c5b1 100644 --- a/arch/x86/math-emu/fpu_proto.h +++ b/arch/x86/math-emu/fpu_proto.h @@ -52,7 +52,7 @@ extern void fst_i_(void); extern void fstp_i(void); /* fpu_entry.c */ asmlinkage extern void math_emulate(long arg); -extern void math_abort(struct info *info, unsigned int signal); +extern void math_abort(struct math_emu_info *info, unsigned int signal); /* fpu_etc.c */ extern void FPU_etc(void); /* fpu_tags.c */ diff --git a/arch/x86/math-emu/fpu_system.h b/arch/x86/math-emu/fpu_system.h index 13488fa153e0..6729c6a31348 100644 --- a/arch/x86/math-emu/fpu_system.h +++ b/arch/x86/math-emu/fpu_system.h @@ -18,7 +18,7 @@ /* This sets the pointer FPU_info to point to the argument part of the stack frame of math_emulate() */ -#define SETUP_DATA_AREA(arg) FPU_info = (struct info *) &arg +#define SETUP_DATA_AREA(arg) FPU_info = (struct math_emu_info *) &arg /* s is always from a cpu register, and the cpu does bounds checking * during register load --> no further bounds checks needed */ @@ -38,12 +38,12 @@ #define I387 (current->thread.xstate) #define FPU_info (I387->soft.info) -#define FPU_CS (*(unsigned short *) &(FPU_info->___cs)) -#define FPU_SS (*(unsigned short *) &(FPU_info->___ss)) -#define FPU_DS (*(unsigned short *) &(FPU_info->___ds)) -#define FPU_EAX (FPU_info->___eax) -#define FPU_EFLAGS (FPU_info->___eflags) -#define FPU_EIP (FPU_info->___eip) +#define FPU_CS (*(unsigned short *) &(FPU_info->regs.cs)) +#define FPU_SS (*(unsigned short *) &(FPU_info->regs.ss)) +#define FPU_DS (*(unsigned short *) &(FPU_info->regs.ds)) +#define FPU_EAX (FPU_info->regs.ax) +#define FPU_EFLAGS (FPU_info->regs.flags) +#define FPU_EIP (FPU_info->regs.ip) #define FPU_ORIG_EIP (FPU_info->___orig_eip) #define FPU_lookahead (I387->soft.lookahead) diff --git a/arch/x86/math-emu/get_address.c b/arch/x86/math-emu/get_address.c index d701e2b39e44..62daa7fcc44c 100644 --- a/arch/x86/math-emu/get_address.c +++ b/arch/x86/math-emu/get_address.c @@ -29,42 +29,39 @@ #define FPU_WRITE_BIT 0x10 static int reg_offset[] = { - offsetof(struct info, ___eax), - offsetof(struct info, ___ecx), - offsetof(struct info, ___edx), - offsetof(struct info, ___ebx), - offsetof(struct info, ___esp), - offsetof(struct info, ___ebp), - offsetof(struct info, ___esi), - offsetof(struct info, ___edi) + offsetof(struct math_emu_info, regs.ax), + offsetof(struct math_emu_info, regs.cx), + offsetof(struct math_emu_info, regs.dx), + offsetof(struct math_emu_info, regs.bx), + offsetof(struct math_emu_info, regs.sp), + offsetof(struct math_emu_info, regs.bp), + offsetof(struct math_emu_info, regs.si), + offsetof(struct math_emu_info, regs.di) }; #define REG_(x) (*(long *)(reg_offset[(x)]+(u_char *) FPU_info)) static int reg_offset_vm86[] = { - offsetof(struct info, ___cs), - offsetof(struct info, ___vm86_ds), - offsetof(struct info, ___vm86_es), - offsetof(struct info, ___vm86_fs), - offsetof(struct info, ___vm86_gs), - offsetof(struct info, ___ss), - offsetof(struct info, ___vm86_ds) + offsetof(struct math_emu_info, regs.cs), + offsetof(struct math_emu_info, vm86.ds), + offsetof(struct math_emu_info, vm86.es), + offsetof(struct math_emu_info, vm86.fs), + offsetof(struct math_emu_info, vm86.gs), + offsetof(struct math_emu_info, regs.ss), + offsetof(struct math_emu_info, vm86.ds) }; #define VM86_REG_(x) (*(unsigned short *) \ (reg_offset_vm86[((unsigned)x)]+(u_char *) FPU_info)) -/* This dummy, gs is not saved on the stack. */ -#define ___GS ___ds - static int reg_offset_pm[] = { - offsetof(struct info, ___cs), - offsetof(struct info, ___ds), - offsetof(struct info, ___es), - offsetof(struct info, ___fs), - offsetof(struct info, ___GS), - offsetof(struct info, ___ss), - offsetof(struct info, ___ds) + offsetof(struct math_emu_info, regs.cs), + offsetof(struct math_emu_info, regs.ds), + offsetof(struct math_emu_info, regs.es), + offsetof(struct math_emu_info, regs.fs), + offsetof(struct math_emu_info, regs.ds), /* dummy, not saved on stack */ + offsetof(struct math_emu_info, regs.ss), + offsetof(struct math_emu_info, regs.ds) }; #define PM_REG_(x) (*(unsigned short *) \ @@ -349,34 +346,34 @@ void __user *FPU_get_address_16(u_char FPU_modrm, unsigned long *fpu_eip, } switch (rm) { case 0: - address += FPU_info->___ebx + FPU_info->___esi; + address += FPU_info->regs.bx + FPU_info->regs.si; break; case 1: - address += FPU_info->___ebx + FPU_info->___edi; + address += FPU_info->regs.bx + FPU_info->regs.di; break; case 2: - address += FPU_info->___ebp + FPU_info->___esi; + address += FPU_info->regs.bp + FPU_info->regs.si; if (addr_modes.override.segment == PREFIX_DEFAULT) addr_modes.override.segment = PREFIX_SS_; break; case 3: - address += FPU_info->___ebp + FPU_info->___edi; + address += FPU_info->regs.bp + FPU_info->regs.di; if (addr_modes.override.segment == PREFIX_DEFAULT) addr_modes.override.segment = PREFIX_SS_; break; case 4: - address += FPU_info->___esi; + address += FPU_info->regs.si; break; case 5: - address += FPU_info->___edi; + address += FPU_info->regs.di; break; case 6: - address += FPU_info->___ebp; + address += FPU_info->regs.bp; if (addr_modes.override.segment == PREFIX_DEFAULT) addr_modes.override.segment = PREFIX_SS_; break; case 7: - address += FPU_info->___ebx; + address += FPU_info->regs.bx; break; } -- cgit v1.2.3 From a5ef7ca0e2636bad0ccd07b996d775348ae2b65e Mon Sep 17 00:00:00 2001 From: Kyle McMartin Date: Sun, 8 Feb 2009 17:39:58 -0500 Subject: x86: spinlocks: define dummy __raw_spin_is_contended Architectures other than mips and x86 are not using ticket spinlocks. Therefore, the contention on the lock is meaningless, since there is nobody known to be waiting on it (arguably /fairly/ unfair locks). Dummy it out to return 0 on other architectures. Signed-off-by: Kyle McMartin Acked-by: Ralf Baechle Acked-by: Ingo Molnar Signed-off-by: Linus Torvalds --- arch/mips/include/asm/spinlock.h | 1 + arch/x86/include/asm/paravirt.h | 1 + arch/x86/include/asm/spinlock.h | 1 + include/linux/spinlock.h | 5 +++++ 4 files changed, 8 insertions(+) (limited to 'arch/x86/include') diff --git a/arch/mips/include/asm/spinlock.h b/arch/mips/include/asm/spinlock.h index 1a1f320c30d8..0884947ebe27 100644 --- a/arch/mips/include/asm/spinlock.h +++ b/arch/mips/include/asm/spinlock.h @@ -51,6 +51,7 @@ static inline int __raw_spin_is_contended(raw_spinlock_t *lock) return (((counters >> 14) - counters) & 0x1fff) > 1; } +#define __raw_spin_is_contended __raw_spin_is_contended static inline void __raw_spin_lock(raw_spinlock_t *lock) { diff --git a/arch/x86/include/asm/paravirt.h b/arch/x86/include/asm/paravirt.h index ba3e2ff6aedc..c09a14127584 100644 --- a/arch/x86/include/asm/paravirt.h +++ b/arch/x86/include/asm/paravirt.h @@ -1402,6 +1402,7 @@ static inline int __raw_spin_is_contended(struct raw_spinlock *lock) { return PVOP_CALL1(int, pv_lock_ops.spin_is_contended, lock); } +#define __raw_spin_is_contended __raw_spin_is_contended static __always_inline void __raw_spin_lock(struct raw_spinlock *lock) { diff --git a/arch/x86/include/asm/spinlock.h b/arch/x86/include/asm/spinlock.h index d17c91981da2..8247e94ac6b1 100644 --- a/arch/x86/include/asm/spinlock.h +++ b/arch/x86/include/asm/spinlock.h @@ -245,6 +245,7 @@ static inline int __raw_spin_is_contended(raw_spinlock_t *lock) { return __ticket_spin_is_contended(lock); } +#define __raw_spin_is_contended __raw_spin_is_contended static __always_inline void __raw_spin_lock(raw_spinlock_t *lock) { diff --git a/include/linux/spinlock.h b/include/linux/spinlock.h index e0c0fccced46..a0c66a2e00ad 100644 --- a/include/linux/spinlock.h +++ b/include/linux/spinlock.h @@ -124,7 +124,12 @@ do { \ #ifdef CONFIG_GENERIC_LOCKBREAK #define spin_is_contended(lock) ((lock)->break_lock) #else + +#ifdef __raw_spin_is_contended #define spin_is_contended(lock) __raw_spin_is_contended(&(lock)->raw_lock) +#else +#define spin_is_contended(lock) (((void)(lock), 0)) +#endif /*__raw_spin_is_contended*/ #endif /** -- cgit v1.2.3 From d315760ffa261c15ff92699ac6f514112543d7ca Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Mon, 9 Feb 2009 22:17:39 +0900 Subject: x86: fix math_emu register frame access do_device_not_available() is the handler for #NM and it declares that it takes a unsigned long and calls math_emu(), which takes a long argument and surprisingly expects the stack frame starting at the zero argument would match struct math_emu_info, which isn't true regardless of configuration in the current code. This patch makes do_device_not_available() take struct pt_regs like other exception handlers and initialize struct math_emu_info with pointer to it and pass pointer to the math_emu_info to math_emulate() like normal C functions do. This way, unless gcc makes a copy of struct pt_regs in do_device_not_available(), the register frame is correctly accessed regardless of kernel configuration or compiler used. This doesn't fix all math_emu problems but it at least gets it somewhat working. Signed-off-by: Tejun Heo Signed-off-by: Ingo Molnar --- arch/x86/include/asm/math_emu.h | 4 +-- arch/x86/include/asm/traps.h | 4 +-- arch/x86/kernel/traps.c | 15 ++++++---- arch/x86/math-emu/fpu_entry.c | 4 +-- arch/x86/math-emu/fpu_proto.h | 2 +- arch/x86/math-emu/fpu_system.h | 16 ++++------ arch/x86/math-emu/get_address.c | 66 ++++++++++++++++++++--------------------- 7 files changed, 55 insertions(+), 56 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/math_emu.h b/arch/x86/include/asm/math_emu.h index 302492c77956..031f6266f425 100644 --- a/arch/x86/include/asm/math_emu.h +++ b/arch/x86/include/asm/math_emu.h @@ -11,8 +11,8 @@ struct math_emu_info { long ___orig_eip; union { - struct pt_regs regs; - struct kernel_vm86_regs vm86; + struct pt_regs *regs; + struct kernel_vm86_regs *vm86; }; }; #endif /* _ASM_X86_MATH_EMU_H */ diff --git a/arch/x86/include/asm/traps.h b/arch/x86/include/asm/traps.h index 2ee0a3bceedf..cf3bb053da0b 100644 --- a/arch/x86/include/asm/traps.h +++ b/arch/x86/include/asm/traps.h @@ -41,7 +41,7 @@ dotraplinkage void do_int3(struct pt_regs *, long); dotraplinkage void do_overflow(struct pt_regs *, long); dotraplinkage void do_bounds(struct pt_regs *, long); dotraplinkage void do_invalid_op(struct pt_regs *, long); -dotraplinkage void do_device_not_available(struct pt_regs *, long); +dotraplinkage void do_device_not_available(struct pt_regs); dotraplinkage void do_coprocessor_segment_overrun(struct pt_regs *, long); dotraplinkage void do_invalid_TSS(struct pt_regs *, long); dotraplinkage void do_segment_not_present(struct pt_regs *, long); @@ -77,7 +77,7 @@ extern int panic_on_unrecovered_nmi; extern int kstack_depth_to_print; void math_error(void __user *); -asmlinkage void math_emulate(long); +void math_emulate(struct math_emu_info *); #ifdef CONFIG_X86_32 unsigned long patch_espfix_desc(unsigned long, unsigned long); #else diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c index 98c2d055284b..7932338d7cb3 100644 --- a/arch/x86/kernel/traps.c +++ b/arch/x86/kernel/traps.c @@ -896,7 +896,7 @@ asmlinkage void math_state_restore(void) EXPORT_SYMBOL_GPL(math_state_restore); #ifndef CONFIG_MATH_EMULATION -asmlinkage void math_emulate(long arg) +void math_emulate(struct math_emu_info *info) { printk(KERN_EMERG "math-emulation not enabled and no coprocessor found.\n"); @@ -906,16 +906,19 @@ asmlinkage void math_emulate(long arg) } #endif /* CONFIG_MATH_EMULATION */ -dotraplinkage void __kprobes -do_device_not_available(struct pt_regs *regs, long error) +dotraplinkage void __kprobes do_device_not_available(struct pt_regs regs) { #ifdef CONFIG_X86_32 if (read_cr0() & X86_CR0_EM) { - conditional_sti(regs); - math_emulate(0); + struct math_emu_info info = { }; + + conditional_sti(®s); + + info.regs = ®s; + math_emulate(&info); } else { math_state_restore(); /* interrupts still off */ - conditional_sti(regs); + conditional_sti(®s); } #else math_state_restore(); diff --git a/arch/x86/math-emu/fpu_entry.c b/arch/x86/math-emu/fpu_entry.c index c268abe72253..5d87f586f8d7 100644 --- a/arch/x86/math-emu/fpu_entry.c +++ b/arch/x86/math-emu/fpu_entry.c @@ -131,7 +131,7 @@ u_char emulating = 0; static int valid_prefix(u_char *Byte, u_char __user ** fpu_eip, overrides * override); -asmlinkage void math_emulate(long arg) +void math_emulate(struct math_emu_info *info) { u_char FPU_modrm, byte1; unsigned short code; @@ -161,7 +161,7 @@ asmlinkage void math_emulate(long arg) RE_ENTRANT_CHECK_ON; #endif /* RE_ENTRANT_CHECKING */ - SETUP_DATA_AREA(arg); + FPU_info = info; FPU_ORIG_EIP = FPU_EIP; diff --git a/arch/x86/math-emu/fpu_proto.h b/arch/x86/math-emu/fpu_proto.h index 51bfbb61c5b1..9779df436b7d 100644 --- a/arch/x86/math-emu/fpu_proto.h +++ b/arch/x86/math-emu/fpu_proto.h @@ -51,7 +51,7 @@ extern void ffreep(void); extern void fst_i_(void); extern void fstp_i(void); /* fpu_entry.c */ -asmlinkage extern void math_emulate(long arg); +extern void math_emulate(struct math_emu_info *info); extern void math_abort(struct math_emu_info *info, unsigned int signal); /* fpu_etc.c */ extern void FPU_etc(void); diff --git a/arch/x86/math-emu/fpu_system.h b/arch/x86/math-emu/fpu_system.h index 6729c6a31348..50fa0ec2c8a5 100644 --- a/arch/x86/math-emu/fpu_system.h +++ b/arch/x86/math-emu/fpu_system.h @@ -16,10 +16,6 @@ #include #include -/* This sets the pointer FPU_info to point to the argument part - of the stack frame of math_emulate() */ -#define SETUP_DATA_AREA(arg) FPU_info = (struct math_emu_info *) &arg - /* s is always from a cpu register, and the cpu does bounds checking * during register load --> no further bounds checks needed */ #define LDT_DESCRIPTOR(s) (((struct desc_struct *)current->mm->context.ldt)[(s) >> 3]) @@ -38,12 +34,12 @@ #define I387 (current->thread.xstate) #define FPU_info (I387->soft.info) -#define FPU_CS (*(unsigned short *) &(FPU_info->regs.cs)) -#define FPU_SS (*(unsigned short *) &(FPU_info->regs.ss)) -#define FPU_DS (*(unsigned short *) &(FPU_info->regs.ds)) -#define FPU_EAX (FPU_info->regs.ax) -#define FPU_EFLAGS (FPU_info->regs.flags) -#define FPU_EIP (FPU_info->regs.ip) +#define FPU_CS (*(unsigned short *) &(FPU_info->regs->cs)) +#define FPU_SS (*(unsigned short *) &(FPU_info->regs->ss)) +#define FPU_DS (*(unsigned short *) &(FPU_info->regs->ds)) +#define FPU_EAX (FPU_info->regs->ax) +#define FPU_EFLAGS (FPU_info->regs->flags) +#define FPU_EIP (FPU_info->regs->ip) #define FPU_ORIG_EIP (FPU_info->___orig_eip) #define FPU_lookahead (I387->soft.lookahead) diff --git a/arch/x86/math-emu/get_address.c b/arch/x86/math-emu/get_address.c index 62daa7fcc44c..420b3b6e3915 100644 --- a/arch/x86/math-emu/get_address.c +++ b/arch/x86/math-emu/get_address.c @@ -29,43 +29,43 @@ #define FPU_WRITE_BIT 0x10 static int reg_offset[] = { - offsetof(struct math_emu_info, regs.ax), - offsetof(struct math_emu_info, regs.cx), - offsetof(struct math_emu_info, regs.dx), - offsetof(struct math_emu_info, regs.bx), - offsetof(struct math_emu_info, regs.sp), - offsetof(struct math_emu_info, regs.bp), - offsetof(struct math_emu_info, regs.si), - offsetof(struct math_emu_info, regs.di) + offsetof(struct pt_regs, ax), + offsetof(struct pt_regs, cx), + offsetof(struct pt_regs, dx), + offsetof(struct pt_regs, bx), + offsetof(struct pt_regs, sp), + offsetof(struct pt_regs, bp), + offsetof(struct pt_regs, si), + offsetof(struct pt_regs, di) }; -#define REG_(x) (*(long *)(reg_offset[(x)]+(u_char *) FPU_info)) +#define REG_(x) (*(long *)(reg_offset[(x)] + (u_char *)FPU_info->regs)) static int reg_offset_vm86[] = { - offsetof(struct math_emu_info, regs.cs), - offsetof(struct math_emu_info, vm86.ds), - offsetof(struct math_emu_info, vm86.es), - offsetof(struct math_emu_info, vm86.fs), - offsetof(struct math_emu_info, vm86.gs), - offsetof(struct math_emu_info, regs.ss), - offsetof(struct math_emu_info, vm86.ds) + offsetof(struct pt_regs, cs), + offsetof(struct kernel_vm86_regs, ds), + offsetof(struct kernel_vm86_regs, es), + offsetof(struct kernel_vm86_regs, fs), + offsetof(struct kernel_vm86_regs, gs), + offsetof(struct pt_regs, ss), + offsetof(struct kernel_vm86_regs, ds) }; #define VM86_REG_(x) (*(unsigned short *) \ - (reg_offset_vm86[((unsigned)x)]+(u_char *) FPU_info)) + (reg_offset_vm86[((unsigned)x)] + (u_char *)FPU_info->regs)) static int reg_offset_pm[] = { - offsetof(struct math_emu_info, regs.cs), - offsetof(struct math_emu_info, regs.ds), - offsetof(struct math_emu_info, regs.es), - offsetof(struct math_emu_info, regs.fs), - offsetof(struct math_emu_info, regs.ds), /* dummy, not saved on stack */ - offsetof(struct math_emu_info, regs.ss), - offsetof(struct math_emu_info, regs.ds) + offsetof(struct pt_regs, cs), + offsetof(struct pt_regs, ds), + offsetof(struct pt_regs, es), + offsetof(struct pt_regs, fs), + offsetof(struct pt_regs, ds), /* dummy, not saved on stack */ + offsetof(struct pt_regs, ss), + offsetof(struct pt_regs, ds) }; #define PM_REG_(x) (*(unsigned short *) \ - (reg_offset_pm[((unsigned)x)]+(u_char *) FPU_info)) + (reg_offset_pm[((unsigned)x)] + (u_char *)FPU_info->regs)) /* Decode the SIB byte. This function assumes mod != 0 */ static int sib(int mod, unsigned long *fpu_eip) @@ -346,34 +346,34 @@ void __user *FPU_get_address_16(u_char FPU_modrm, unsigned long *fpu_eip, } switch (rm) { case 0: - address += FPU_info->regs.bx + FPU_info->regs.si; + address += FPU_info->regs->bx + FPU_info->regs->si; break; case 1: - address += FPU_info->regs.bx + FPU_info->regs.di; + address += FPU_info->regs->bx + FPU_info->regs->di; break; case 2: - address += FPU_info->regs.bp + FPU_info->regs.si; + address += FPU_info->regs->bp + FPU_info->regs->si; if (addr_modes.override.segment == PREFIX_DEFAULT) addr_modes.override.segment = PREFIX_SS_; break; case 3: - address += FPU_info->regs.bp + FPU_info->regs.di; + address += FPU_info->regs->bp + FPU_info->regs->di; if (addr_modes.override.segment == PREFIX_DEFAULT) addr_modes.override.segment = PREFIX_SS_; break; case 4: - address += FPU_info->regs.si; + address += FPU_info->regs->si; break; case 5: - address += FPU_info->regs.di; + address += FPU_info->regs->di; break; case 6: - address += FPU_info->regs.bp; + address += FPU_info->regs->bp; if (addr_modes.override.segment == PREFIX_DEFAULT) addr_modes.override.segment = PREFIX_SS_; break; case 7: - address += FPU_info->regs.bx; + address += FPU_info->regs->bx; break; } -- cgit v1.2.3 From be03d9e8022030c16abf534e33e185bfc3d40eef Mon Sep 17 00:00:00 2001 From: Suresh Siddha Date: Wed, 11 Feb 2009 11:20:23 -0800 Subject: x86, pat: fix warn_on_once() while mapping 0-1MB range with /dev/mem Jeff Mahoney reported: > With Suse's hwinfo tool, on -tip: > WARNING: at arch/x86/mm/pat.c:637 reserve_pfn_range+0x5b/0x26d() reserve_pfn_range() is not tracking the memory range below 1MB as non-RAM and as such is inconsistent with similar checks in reserve_memtype() and free_memtype() Rename the pagerange_is_ram() to pat_pagerange_is_ram() and add the "track legacy 1MB region as non RAM" condition. And also, fix reserve_pfn_range() to return -EINVAL, when the pfn range is RAM. This is to be consistent with this API design. Reported-and-tested-by: Jeff Mahoney Signed-off-by: Suresh Siddha Signed-off-by: Venkatesh Pallipadi Signed-off-by: Ingo Molnar --- arch/x86/include/asm/page.h | 1 - arch/x86/mm/ioremap.c | 19 ----------- arch/x86/mm/pat.c | 83 ++++++++++++++++++++++++--------------------- 3 files changed, 45 insertions(+), 58 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/page.h b/arch/x86/include/asm/page.h index e9873a2e8695..776579119a00 100644 --- a/arch/x86/include/asm/page.h +++ b/arch/x86/include/asm/page.h @@ -57,7 +57,6 @@ typedef struct { pgdval_t pgd; } pgd_t; typedef struct { pgprotval_t pgprot; } pgprot_t; extern int page_is_ram(unsigned long pagenr); -extern int pagerange_is_ram(unsigned long start, unsigned long end); extern int devmem_is_allowed(unsigned long pagenr); extern void map_devmem(unsigned long pfn, unsigned long size, pgprot_t vma_prot); diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c index af750ab973b6..f45d5e29a72e 100644 --- a/arch/x86/mm/ioremap.c +++ b/arch/x86/mm/ioremap.c @@ -134,25 +134,6 @@ int page_is_ram(unsigned long pagenr) return 0; } -int pagerange_is_ram(unsigned long start, unsigned long end) -{ - int ram_page = 0, not_rampage = 0; - unsigned long page_nr; - - for (page_nr = (start >> PAGE_SHIFT); page_nr < (end >> PAGE_SHIFT); - ++page_nr) { - if (page_is_ram(page_nr)) - ram_page = 1; - else - not_rampage = 1; - - if (ram_page == not_rampage) - return -1; - } - - return ram_page; -} - /* * Fix up the linear direct mapping of the kernel to avoid cache attribute * conflicts. diff --git a/arch/x86/mm/pat.c b/arch/x86/mm/pat.c index 7b61036427df..aebbf67a79d0 100644 --- a/arch/x86/mm/pat.c +++ b/arch/x86/mm/pat.c @@ -211,6 +211,33 @@ chk_conflict(struct memtype *new, struct memtype *entry, unsigned long *type) static struct memtype *cached_entry; static u64 cached_start; +static int pat_pagerange_is_ram(unsigned long start, unsigned long end) +{ + int ram_page = 0, not_rampage = 0; + unsigned long page_nr; + + for (page_nr = (start >> PAGE_SHIFT); page_nr < (end >> PAGE_SHIFT); + ++page_nr) { + /* + * For legacy reasons, physical address range in the legacy ISA + * region is tracked as non-RAM. This will allow users of + * /dev/mem to map portions of legacy ISA region, even when + * some of those portions are listed(or not even listed) with + * different e820 types(RAM/reserved/..) + */ + if (page_nr >= (ISA_END_ADDRESS >> PAGE_SHIFT) && + page_is_ram(page_nr)) + ram_page = 1; + else + not_rampage = 1; + + if (ram_page == not_rampage) + return -1; + } + + return ram_page; +} + /* * For RAM pages, mark the pages as non WB memory type using * PageNonWB (PG_arch_1). We allow only one set_memory_uc() or @@ -336,20 +363,12 @@ int reserve_memtype(u64 start, u64 end, unsigned long req_type, if (new_type) *new_type = actual_type; - /* - * For legacy reasons, some parts of the physical address range in the - * legacy 1MB region is treated as non-RAM (even when listed as RAM in - * the e820 tables). So we will track the memory attributes of this - * legacy 1MB region using the linear memtype_list always. - */ - if (end >= ISA_END_ADDRESS) { - is_range_ram = pagerange_is_ram(start, end); - if (is_range_ram == 1) - return reserve_ram_pages_type(start, end, req_type, - new_type); - else if (is_range_ram < 0) - return -EINVAL; - } + is_range_ram = pat_pagerange_is_ram(start, end); + if (is_range_ram == 1) + return reserve_ram_pages_type(start, end, req_type, + new_type); + else if (is_range_ram < 0) + return -EINVAL; new = kmalloc(sizeof(struct memtype), GFP_KERNEL); if (!new) @@ -446,19 +465,11 @@ int free_memtype(u64 start, u64 end) if (is_ISA_range(start, end - 1)) return 0; - /* - * For legacy reasons, some parts of the physical address range in the - * legacy 1MB region is treated as non-RAM (even when listed as RAM in - * the e820 tables). So we will track the memory attributes of this - * legacy 1MB region using the linear memtype_list always. - */ - if (end >= ISA_END_ADDRESS) { - is_range_ram = pagerange_is_ram(start, end); - if (is_range_ram == 1) - return free_ram_pages_type(start, end); - else if (is_range_ram < 0) - return -EINVAL; - } + is_range_ram = pat_pagerange_is_ram(start, end); + if (is_range_ram == 1) + return free_ram_pages_type(start, end); + else if (is_range_ram < 0) + return -EINVAL; spin_lock(&memtype_lock); list_for_each_entry(entry, &memtype_list, nd) { @@ -626,17 +637,13 @@ static int reserve_pfn_range(u64 paddr, unsigned long size, pgprot_t *vma_prot, unsigned long flags; unsigned long want_flags = (pgprot_val(*vma_prot) & _PAGE_CACHE_MASK); - is_ram = pagerange_is_ram(paddr, paddr + size); + is_ram = pat_pagerange_is_ram(paddr, paddr + size); - if (is_ram != 0) { - /* - * For mapping RAM pages, drivers need to call - * set_memory_[uc|wc|wb] directly, for reserve and free, before - * setting up the PTE. - */ - WARN_ON_ONCE(1); - return 0; - } + /* + * reserve_pfn_range() doesn't support RAM pages. + */ + if (is_ram != 0) + return -EINVAL; ret = reserve_memtype(paddr, paddr + size, want_flags, &flags); if (ret) @@ -693,7 +700,7 @@ static void free_pfn_range(u64 paddr, unsigned long size) { int is_ram; - is_ram = pagerange_is_ram(paddr, paddr + size); + is_ram = pat_pagerange_is_ram(paddr, paddr + size); if (is_ram == 0) free_memtype(paddr, paddr + size); } -- cgit v1.2.3 From d85cf93da66977dbc645352be1b2084a659d8a0b Mon Sep 17 00:00:00 2001 From: Jeremy Fitzhardinge Date: Thu, 12 Feb 2009 10:02:56 -0800 Subject: x86/paravirt: make arch_flush_lazy_mmu/cpu disable preemption Impact: avoid access to percpu vars in preempible context They are intended to be used whenever there's the possibility that there's some stale state which is going to be overwritten with a queued update, or to force a state change when we may be in lazy mode. Either way, we could end up calling it with preemption enabled, so wrap the functions in their own little preempt-disable section so they can be safely called in any context (though preemption should never be enabled if we're actually in a lazy state). (Move out of line to avoid #include dependencies.) Signed-off-by: Jeremy Fitzhardinge Signed-off-by: Thomas Gleixner --- arch/x86/include/asm/paravirt.h | 17 ++--------------- arch/x86/kernel/paravirt.c | 24 ++++++++++++++++++++++++ 2 files changed, 26 insertions(+), 15 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/paravirt.h b/arch/x86/include/asm/paravirt.h index ba3e2ff6aedc..a660eceaa273 100644 --- a/arch/x86/include/asm/paravirt.h +++ b/arch/x86/include/asm/paravirt.h @@ -1352,14 +1352,7 @@ static inline void arch_leave_lazy_cpu_mode(void) PVOP_VCALL0(pv_cpu_ops.lazy_mode.leave); } -static inline void arch_flush_lazy_cpu_mode(void) -{ - if (unlikely(paravirt_get_lazy_mode() == PARAVIRT_LAZY_CPU)) { - arch_leave_lazy_cpu_mode(); - arch_enter_lazy_cpu_mode(); - } -} - +void arch_flush_lazy_cpu_mode(void); #define __HAVE_ARCH_ENTER_LAZY_MMU_MODE static inline void arch_enter_lazy_mmu_mode(void) @@ -1372,13 +1365,7 @@ static inline void arch_leave_lazy_mmu_mode(void) PVOP_VCALL0(pv_mmu_ops.lazy_mode.leave); } -static inline void arch_flush_lazy_mmu_mode(void) -{ - if (unlikely(paravirt_get_lazy_mode() == PARAVIRT_LAZY_MMU)) { - arch_leave_lazy_mmu_mode(); - arch_enter_lazy_mmu_mode(); - } -} +void arch_flush_lazy_mmu_mode(void); static inline void __set_fixmap(unsigned /* enum fixed_addresses */ idx, unsigned long phys, pgprot_t flags) diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c index e4c8fb608873..dcba6c567a2a 100644 --- a/arch/x86/kernel/paravirt.c +++ b/arch/x86/kernel/paravirt.c @@ -268,6 +268,30 @@ enum paravirt_lazy_mode paravirt_get_lazy_mode(void) return __get_cpu_var(paravirt_lazy_mode); } +void arch_flush_lazy_mmu_mode(void) +{ + preempt_disable(); + + if (paravirt_get_lazy_mode() == PARAVIRT_LAZY_MMU) { + arch_leave_lazy_mmu_mode(); + arch_enter_lazy_mmu_mode(); + } + + preempt_enable(); +} + +void arch_flush_lazy_cpu_mode(void) +{ + preempt_disable(); + + if (paravirt_get_lazy_mode() == PARAVIRT_LAZY_CPU) { + arch_leave_lazy_cpu_mode(); + arch_enter_lazy_cpu_mode(); + } + + preempt_enable(); +} + struct pv_info pv_info = { .name = "bare hardware", .paravirt_enabled = 0, -- cgit v1.2.3 From 7a0eb1960e8ddcb68ea631caf16815485af0e228 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Mon, 19 Jan 2009 14:57:52 +0200 Subject: KVM: Avoid using CONFIG_ in userspace visible headers Kconfig symbols are not available in userspace, and are not stripped by headers-install. Avoid their use by adding #defines in to suit each architecture. Signed-off-by: Avi Kivity --- arch/ia64/include/asm/kvm.h | 4 ++++ arch/x86/include/asm/kvm.h | 7 +++++++ include/linux/kvm.h | 10 +++++----- 3 files changed, 16 insertions(+), 5 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/ia64/include/asm/kvm.h b/arch/ia64/include/asm/kvm.h index 68aa6da807c1..bfa86b6af7cd 100644 --- a/arch/ia64/include/asm/kvm.h +++ b/arch/ia64/include/asm/kvm.h @@ -25,6 +25,10 @@ #include +/* Select x86 specific features in */ +#define __KVM_HAVE_IOAPIC +#define __KVM_HAVE_DEVICE_ASSIGNMENT + /* Architectural interrupt line count. */ #define KVM_NR_INTERRUPTS 256 diff --git a/arch/x86/include/asm/kvm.h b/arch/x86/include/asm/kvm.h index d2e3bf3608af..886c9402ec45 100644 --- a/arch/x86/include/asm/kvm.h +++ b/arch/x86/include/asm/kvm.h @@ -9,6 +9,13 @@ #include #include +/* Select x86 specific features in */ +#define __KVM_HAVE_PIT +#define __KVM_HAVE_IOAPIC +#define __KVM_HAVE_DEVICE_ASSIGNMENT +#define __KVM_HAVE_MSI +#define __KVM_HAVE_USER_NMI + /* Architectural interrupt line count. */ #define KVM_NR_INTERRUPTS 256 diff --git a/include/linux/kvm.h b/include/linux/kvm.h index 5715f1907601..0424326f1679 100644 --- a/include/linux/kvm.h +++ b/include/linux/kvm.h @@ -58,10 +58,10 @@ struct kvm_irqchip { __u32 pad; union { char dummy[512]; /* reserving space */ -#ifdef CONFIG_X86 +#ifdef __KVM_HAVE_PIT struct kvm_pic_state pic; #endif -#if defined(CONFIG_X86) || defined(CONFIG_IA64) +#ifdef __KVM_HAVE_IOAPIC struct kvm_ioapic_state ioapic; #endif } chip; @@ -384,16 +384,16 @@ struct kvm_trace_rec { #define KVM_CAP_MP_STATE 14 #define KVM_CAP_COALESCED_MMIO 15 #define KVM_CAP_SYNC_MMU 16 /* Changes to host mmap are reflected in guest */ -#if defined(CONFIG_X86)||defined(CONFIG_IA64) +#ifdef __KVM_HAVE_DEVICE_ASSIGNMENT #define KVM_CAP_DEVICE_ASSIGNMENT 17 #endif #define KVM_CAP_IOMMU 18 -#if defined(CONFIG_X86) +#ifdef __KVM_HAVE_MSI #define KVM_CAP_DEVICE_MSI 20 #endif /* Bug in KVM_SET_USER_MEMORY_REGION fixed: */ #define KVM_CAP_DESTROY_MEMORY_REGION_WORKS 21 -#if defined(CONFIG_X86) +#ifdef __KVM_HAVE_USER_NMI #define KVM_CAP_USER_NMI 22 #endif -- cgit v1.2.3 From f2dbcfa738368c8a40d4a5f0b65dc9879577cb21 Mon Sep 17 00:00:00 2001 From: KAMEZAWA Hiroyuki Date: Wed, 18 Feb 2009 14:48:32 -0800 Subject: mm: clean up for early_pfn_to_nid() What's happening is that the assertion in mm/page_alloc.c:move_freepages() is triggering: BUG_ON(page_zone(start_page) != page_zone(end_page)); Once I knew this is what was happening, I added some annotations: if (unlikely(page_zone(start_page) != page_zone(end_page))) { printk(KERN_ERR "move_freepages: Bogus zones: " "start_page[%p] end_page[%p] zone[%p]\n", start_page, end_page, zone); printk(KERN_ERR "move_freepages: " "start_zone[%p] end_zone[%p]\n", page_zone(start_page), page_zone(end_page)); printk(KERN_ERR "move_freepages: " "start_pfn[0x%lx] end_pfn[0x%lx]\n", page_to_pfn(start_page), page_to_pfn(end_page)); printk(KERN_ERR "move_freepages: " "start_nid[%d] end_nid[%d]\n", page_to_nid(start_page), page_to_nid(end_page)); ... And here's what I got: move_freepages: Bogus zones: start_page[2207d0000] end_page[2207dffc0] zone[fffff8103effcb00] move_freepages: start_zone[fffff8103effcb00] end_zone[fffff8003fffeb00] move_freepages: start_pfn[0x81f600] end_pfn[0x81f7ff] move_freepages: start_nid[1] end_nid[0] My memory layout on this box is: [ 0.000000] Zone PFN ranges: [ 0.000000] Normal 0x00000000 -> 0x0081ff5d [ 0.000000] Movable zone start PFN for each node [ 0.000000] early_node_map[8] active PFN ranges [ 0.000000] 0: 0x00000000 -> 0x00020000 [ 0.000000] 1: 0x00800000 -> 0x0081f7ff [ 0.000000] 1: 0x0081f800 -> 0x0081fe50 [ 0.000000] 1: 0x0081fed1 -> 0x0081fed8 [ 0.000000] 1: 0x0081feda -> 0x0081fedb [ 0.000000] 1: 0x0081fedd -> 0x0081fee5 [ 0.000000] 1: 0x0081fee7 -> 0x0081ff51 [ 0.000000] 1: 0x0081ff59 -> 0x0081ff5d So it's a block move in that 0x81f600-->0x81f7ff region which triggers the problem. This patch: Declaration of early_pfn_to_nid() is scattered over per-arch include files, and it seems it's complicated to know when the declaration is used. I think it makes fix-for-memmap-init not easy. This patch moves all declaration to include/linux/mm.h After this, if !CONFIG_NODES_POPULATES_NODE_MAP && !CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID -> Use static definition in include/linux/mm.h else if !CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID -> Use generic definition in mm/page_alloc.c else -> per-arch back end function will be called. Signed-off-by: KAMEZAWA Hiroyuki Tested-by: KOSAKI Motohiro Reported-by: David Miller Cc: Mel Gorman Cc: Heiko Carstens Cc: [2.6.25.x, 2.6.26.x, 2.6.27.x, 2.6.28.x] Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- arch/ia64/include/asm/mmzone.h | 4 ---- arch/ia64/mm/numa.c | 2 +- arch/x86/include/asm/mmzone_32.h | 2 -- arch/x86/include/asm/mmzone_64.h | 2 -- arch/x86/mm/numa_64.c | 2 +- include/linux/mm.h | 19 ++++++++++++++++--- mm/page_alloc.c | 8 +++++++- 7 files changed, 25 insertions(+), 14 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/ia64/include/asm/mmzone.h b/arch/ia64/include/asm/mmzone.h index 34efe88eb849..f2ca32069b3f 100644 --- a/arch/ia64/include/asm/mmzone.h +++ b/arch/ia64/include/asm/mmzone.h @@ -31,10 +31,6 @@ static inline int pfn_to_nid(unsigned long pfn) #endif } -#ifdef CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID -extern int early_pfn_to_nid(unsigned long pfn); -#endif - #ifdef CONFIG_IA64_DIG /* DIG systems are small */ # define MAX_PHYSNODE_ID 8 # define NR_NODE_MEMBLKS (MAX_NUMNODES * 8) diff --git a/arch/ia64/mm/numa.c b/arch/ia64/mm/numa.c index b73bf1838e57..5061c3fb6796 100644 --- a/arch/ia64/mm/numa.c +++ b/arch/ia64/mm/numa.c @@ -58,7 +58,7 @@ paddr_to_nid(unsigned long paddr) * SPARSEMEM to allocate the SPARSEMEM sectionmap on the NUMA node where * the section resides. */ -int early_pfn_to_nid(unsigned long pfn) +int __meminit __early_pfn_to_nid(unsigned long pfn) { int i, section = pfn >> PFN_SECTION_SHIFT, ssec, esec; diff --git a/arch/x86/include/asm/mmzone_32.h b/arch/x86/include/asm/mmzone_32.h index 07f1af494ca5..105fb90a0635 100644 --- a/arch/x86/include/asm/mmzone_32.h +++ b/arch/x86/include/asm/mmzone_32.h @@ -32,8 +32,6 @@ static inline void get_memcfg_numa(void) get_memcfg_numa_flat(); } -extern int early_pfn_to_nid(unsigned long pfn); - extern void resume_map_numa_kva(pgd_t *pgd); #else /* !CONFIG_NUMA */ diff --git a/arch/x86/include/asm/mmzone_64.h b/arch/x86/include/asm/mmzone_64.h index a5b3817d4b9e..a29f48c2a322 100644 --- a/arch/x86/include/asm/mmzone_64.h +++ b/arch/x86/include/asm/mmzone_64.h @@ -40,8 +40,6 @@ static inline __attribute__((pure)) int phys_to_nid(unsigned long addr) #define node_end_pfn(nid) (NODE_DATA(nid)->node_start_pfn + \ NODE_DATA(nid)->node_spanned_pages) -extern int early_pfn_to_nid(unsigned long pfn); - #ifdef CONFIG_NUMA_EMU #define FAKE_NODE_MIN_SIZE (64 * 1024 * 1024) #define FAKE_NODE_MIN_HASH_MASK (~(FAKE_NODE_MIN_SIZE - 1UL)) diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c index 71a14f89f89e..f3516da035d1 100644 --- a/arch/x86/mm/numa_64.c +++ b/arch/x86/mm/numa_64.c @@ -145,7 +145,7 @@ int __init compute_hash_shift(struct bootnode *nodes, int numnodes, return shift; } -int early_pfn_to_nid(unsigned long pfn) +int __meminit __early_pfn_to_nid(unsigned long pfn) { return phys_to_nid(pfn << PAGE_SHIFT); } diff --git a/include/linux/mm.h b/include/linux/mm.h index 10074212a35b..065cdf8c09fb 100644 --- a/include/linux/mm.h +++ b/include/linux/mm.h @@ -1041,10 +1041,23 @@ extern void free_bootmem_with_active_regions(int nid, typedef int (*work_fn_t)(unsigned long, unsigned long, void *); extern void work_with_active_regions(int nid, work_fn_t work_fn, void *data); extern void sparse_memory_present_with_active_regions(int nid); -#ifndef CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID -extern int early_pfn_to_nid(unsigned long pfn); -#endif /* CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID */ #endif /* CONFIG_ARCH_POPULATES_NODE_MAP */ + +#if !defined(CONFIG_ARCH_POPULATES_NODE_MAP) && \ + !defined(CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID) +static inline int __early_pfn_to_nid(unsigned long pfn) +{ + return 0; +} +#else +/* please see mm/page_alloc.c */ +extern int __meminit early_pfn_to_nid(unsigned long pfn); +#ifdef CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID +/* there is a per-arch backend function. */ +extern int __meminit __early_pfn_to_nid(unsigned long pfn); +#endif /* CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID */ +#endif + extern void set_dma_reserve(unsigned long new_dma_reserve); extern void memmap_init_zone(unsigned long, int, unsigned long, unsigned long, enum memmap_context); diff --git a/mm/page_alloc.c b/mm/page_alloc.c index 5675b3073854..c5dd74602efc 100644 --- a/mm/page_alloc.c +++ b/mm/page_alloc.c @@ -2989,7 +2989,7 @@ static int __meminit next_active_region_index_in_nid(int index, int nid) * was used and there are no special requirements, this is a convenient * alternative */ -int __meminit early_pfn_to_nid(unsigned long pfn) +int __meminit __early_pfn_to_nid(unsigned long pfn) { int i; @@ -3005,6 +3005,12 @@ int __meminit early_pfn_to_nid(unsigned long pfn) } #endif /* CONFIG_HAVE_ARCH_EARLY_PFN_TO_NID */ +int __meminit early_pfn_to_nid(unsigned long pfn) +{ + return __early_pfn_to_nid(pfn); +} + + /* Basic iterator support to walk early_node_map[] */ #define for_each_active_range_index_in_nid(i, nid) \ for (i = first_active_region_index_in_nid(nid); i != -1; \ -- cgit v1.2.3 From 4ab0d47d0ab311eb181532c1ecb6d02905685071 Mon Sep 17 00:00:00 2001 From: Venkatesh Pallipadi Date: Tue, 24 Feb 2009 17:35:12 -0800 Subject: gpu/drm, x86, PAT: io_mapping_create_wc and resource_size_t io_mapping_create_wc should take a resource_size_t parameter in place of unsigned long. With unsigned long, there will be no way to map greater than 4GB address in i386/32 bit. On x86, greater than 4GB addresses cannot be mapped on i386 without PAE. Return error for such a case. Patch also adds a structure for io_mapping, that saves the base, size and type on HAVE_ATOMIC_IOMAP archs, that can be used to verify the offset on io_mapping_map calls. Signed-off-by: Venkatesh Pallipadi Signed-off-by: Suresh Siddha Cc: Dave Airlie Cc: Jesse Barnes Cc: Eric Anholt Cc: Keith Packard Signed-off-by: Ingo Molnar --- arch/x86/include/asm/iomap.h | 3 +++ arch/x86/mm/iomap_32.c | 18 +++++++++++++++++ include/linux/io-mapping.h | 46 +++++++++++++++++++++++++++++++++----------- 3 files changed, 56 insertions(+), 11 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/x86/include/asm/iomap.h b/arch/x86/include/asm/iomap.h index c1f06289b14b..86af26091d6c 100644 --- a/arch/x86/include/asm/iomap.h +++ b/arch/x86/include/asm/iomap.h @@ -23,6 +23,9 @@ #include #include +int +is_io_mapping_possible(resource_size_t base, unsigned long size); + void * iomap_atomic_prot_pfn(unsigned long pfn, enum km_type type, pgprot_t prot); diff --git a/arch/x86/mm/iomap_32.c b/arch/x86/mm/iomap_32.c index ca53224fc56c..6c2b1af16926 100644 --- a/arch/x86/mm/iomap_32.c +++ b/arch/x86/mm/iomap_32.c @@ -20,6 +20,24 @@ #include #include +#ifdef CONFIG_X86_PAE +int +is_io_mapping_possible(resource_size_t base, unsigned long size) +{ + return 1; +} +#else +int +is_io_mapping_possible(resource_size_t base, unsigned long size) +{ + /* There is no way to map greater than 1 << 32 address without PAE */ + if (base + size > 0x100000000ULL) + return 0; + + return 1; +} +#endif + /* Map 'pfn' using fixed map 'type' and protections 'prot' */ void * diff --git a/include/linux/io-mapping.h b/include/linux/io-mapping.h index 82df31726a54..cbc2f0cd631b 100644 --- a/include/linux/io-mapping.h +++ b/include/linux/io-mapping.h @@ -30,11 +30,14 @@ * See Documentation/io_mapping.txt */ -/* this struct isn't actually defined anywhere */ -struct io_mapping; - #ifdef CONFIG_HAVE_ATOMIC_IOMAP +struct io_mapping { + resource_size_t base; + unsigned long size; + pgprot_t prot; +}; + /* * For small address space machines, mapping large objects * into the kernel virtual space isn't practical. Where @@ -43,23 +46,40 @@ struct io_mapping; */ static inline struct io_mapping * -io_mapping_create_wc(unsigned long base, unsigned long size) +io_mapping_create_wc(resource_size_t base, unsigned long size) { - return (struct io_mapping *) base; + struct io_mapping *iomap; + + if (!is_io_mapping_possible(base, size)) + return NULL; + + iomap = kmalloc(sizeof(*iomap), GFP_KERNEL); + if (!iomap) + return NULL; + + iomap->base = base; + iomap->size = size; + iomap->prot = pgprot_writecombine(__pgprot(__PAGE_KERNEL)); + return iomap; } static inline void io_mapping_free(struct io_mapping *mapping) { + kfree(mapping); } /* Atomic map/unmap */ static inline void * io_mapping_map_atomic_wc(struct io_mapping *mapping, unsigned long offset) { - offset += (unsigned long) mapping; - return iomap_atomic_prot_pfn(offset >> PAGE_SHIFT, KM_USER0, - __pgprot(__PAGE_KERNEL_WC)); + resource_size_t phys_addr; + unsigned long pfn; + + BUG_ON(offset >= mapping->size); + phys_addr = mapping->base + offset; + pfn = (unsigned long) (phys_addr >> PAGE_SHIFT); + return iomap_atomic_prot_pfn(pfn, KM_USER0, mapping->prot); } static inline void @@ -71,8 +91,9 @@ io_mapping_unmap_atomic(void *vaddr) static inline void * io_mapping_map_wc(struct io_mapping *mapping, unsigned long offset) { - offset += (unsigned long) mapping; - return ioremap_wc(offset, PAGE_SIZE); + BUG_ON(offset >= mapping->size); + resource_size_t phys_addr = mapping->base + offset; + return ioremap_wc(phys_addr, PAGE_SIZE); } static inline void @@ -83,9 +104,12 @@ io_mapping_unmap(void *vaddr) #else +/* this struct isn't actually defined anywhere */ +struct io_mapping; + /* Create the io_mapping object*/ static inline struct io_mapping * -io_mapping_create_wc(unsigned long base, unsigned long size) +io_mapping_create_wc(resource_size_t base, unsigned long size) { return (struct io_mapping *) ioremap_wc(base, size); } -- cgit v1.2.3 From 5b1017404aea6d2e552e991b3fd814d839e9cd67 Mon Sep 17 00:00:00 2001 From: Roland McGrath Date: Fri, 27 Feb 2009 23:25:54 -0800 Subject: x86-64: seccomp: fix 32/64 syscall hole On x86-64, a 32-bit process (TIF_IA32) can switch to 64-bit mode with ljmp, and then use the "syscall" instruction to make a 64-bit system call. A 64-bit process make a 32-bit system call with int $0x80. In both these cases under CONFIG_SECCOMP=y, secure_computing() will use the wrong system call number table. The fix is simple: test TS_COMPAT instead of TIF_IA32. Here is an example exploit: /* test case for seccomp circumvention on x86-64 There are two failure modes: compile with -m64 or compile with -m32. The -m64 case is the worst one, because it does "chmod 777 ." (could be any chmod call). The -m32 case demonstrates it was able to do stat(), which can glean information but not harm anything directly. A buggy kernel will let the test do something, print, and exit 1; a fixed kernel will make it exit with SIGKILL before it does anything. */ #define _GNU_SOURCE #include #include #include #include #include #include #include int main (int argc, char **argv) { char buf[100]; static const char dot[] = "."; long ret; unsigned st[24]; if (prctl (PR_SET_SECCOMP, 1, 0, 0, 0) != 0) perror ("prctl(PR_SET_SECCOMP) -- not compiled into kernel?"); #ifdef __x86_64__ assert ((uintptr_t) dot < (1UL << 32)); asm ("int $0x80 # %0 <- %1(%2 %3)" : "=a" (ret) : "0" (15), "b" (dot), "c" (0777)); ret = snprintf (buf, sizeof buf, "result %ld (check mode on .!)\n", ret); #elif defined __i386__ asm (".code32\n" "pushl %%cs\n" "pushl $2f\n" "ljmpl $0x33, $1f\n" ".code64\n" "1: syscall # %0 <- %1(%2 %3)\n" "lretl\n" ".code32\n" "2:" : "=a" (ret) : "0" (4), "D" (dot), "S" (&st)); if (ret == 0) ret = snprintf (buf, sizeof buf, "stat . -> st_uid=%u\n", st[7]); else ret = snprintf (buf, sizeof buf, "result %ld\n", ret); #else # error "not this one" #endif write (1, buf, ret); syscall (__NR_exit, 1); return 2; } Signed-off-by: Roland McGrath [ I don't know if anybody actually uses seccomp, but it's enabled in at least both Fedora and SuSE kernels, so maybe somebody is. - Linus ] Signed-off-by: Linus Torvalds --- arch/mips/include/asm/seccomp.h | 1 - arch/powerpc/include/asm/compat.h | 5 +++++ arch/powerpc/include/asm/seccomp.h | 4 ---- arch/sparc/include/asm/compat.h | 5 +++++ arch/sparc/include/asm/seccomp.h | 6 ------ arch/x86/include/asm/seccomp_32.h | 6 ------ arch/x86/include/asm/seccomp_64.h | 8 -------- kernel/seccomp.c | 7 ++++--- 8 files changed, 14 insertions(+), 28 deletions(-) (limited to 'arch/x86/include') diff --git a/arch/mips/include/asm/seccomp.h b/arch/mips/include/asm/seccomp.h index 36ed44070256..a6772e9507f5 100644 --- a/arch/mips/include/asm/seccomp.h +++ b/arch/mips/include/asm/seccomp.h @@ -1,6 +1,5 @@ #ifndef __ASM_SECCOMP_H -#include #include #define __NR_seccomp_read __NR_read diff --git a/arch/powerpc/include/asm/compat.h b/arch/powerpc/include/asm/compat.h index d811a8cd7b58..4774c2f92232 100644 --- a/arch/powerpc/include/asm/compat.h +++ b/arch/powerpc/include/asm/compat.h @@ -210,5 +210,10 @@ struct compat_shmid64_ds { compat_ulong_t __unused6; }; +static inline int is_compat_task(void) +{ + return test_thread_flag(TIF_32BIT); +} + #endif /* __KERNEL__ */ #endif /* _ASM_POWERPC_COMPAT_H */ diff --git a/arch/powerpc/include/asm/seccomp.h b/arch/powerpc/include/asm/seccomp.h index 853765eb1f65..00c1d9133cfe 100644 --- a/arch/powerpc/include/asm/seccomp.h +++ b/arch/powerpc/include/asm/seccomp.h @@ -1,10 +1,6 @@ #ifndef _ASM_POWERPC_SECCOMP_H #define _ASM_POWERPC_SECCOMP_H -#ifdef __KERNEL__ -#include -#endif - #include #define __NR_seccomp_read __NR_read diff --git a/arch/sparc/include/asm/compat.h b/arch/sparc/include/asm/compat.h index f260b58f5ce9..0e706257918f 100644 --- a/arch/sparc/include/asm/compat.h +++ b/arch/sparc/include/asm/compat.h @@ -240,4 +240,9 @@ struct compat_shmid64_ds { unsigned int __unused2; }; +static inline int is_compat_task(void) +{ + return test_thread_flag(TIF_32BIT); +} + #endif /* _ASM_SPARC64_COMPAT_H */ diff --git a/arch/sparc/include/asm/seccomp.h b/arch/sparc/include/asm/seccomp.h index 7fcd9968192b..adca1bce41d4 100644 --- a/arch/sparc/include/asm/seccomp.h +++ b/arch/sparc/include/asm/seccomp.h @@ -1,11 +1,5 @@ #ifndef _ASM_SECCOMP_H -#include /* already defines TIF_32BIT */ - -#ifndef TIF_32BIT -#error "unexpected TIF_32BIT on sparc64" -#endif - #include #define __NR_seccomp_read __NR_read diff --git a/arch/x86/include/asm/seccomp_32.h b/arch/x86/include/asm/seccomp_32.h index a6ad87b352c4..b811d6f5780c 100644 --- a/arch/x86/include/asm/seccomp_32.h +++ b/arch/x86/include/asm/seccomp_32.h @@ -1,12 +1,6 @@ #ifndef _ASM_X86_SECCOMP_32_H #define _ASM_X86_SECCOMP_32_H -#include - -#ifdef TIF_32BIT -#error "unexpected TIF_32BIT on i386" -#endif - #include #define __NR_seccomp_read __NR_read diff --git a/arch/x86/include/asm/seccomp_64.h b/arch/x86/include/asm/seccomp_64.h index 4171bb794e9e..84ec1bd161a5 100644 --- a/arch/x86/include/asm/seccomp_64.h +++ b/arch/x86/include/asm/seccomp_64.h @@ -1,14 +1,6 @@ #ifndef _ASM_X86_SECCOMP_64_H #define _ASM_X86_SECCOMP_64_H -#include - -#ifdef TIF_32BIT -#error "unexpected TIF_32BIT on x86_64" -#else -#define TIF_32BIT TIF_IA32 -#endif - #include #include diff --git a/kernel/seccomp.c b/kernel/seccomp.c index ad64fcb731f2..57d4b13b631d 100644 --- a/kernel/seccomp.c +++ b/kernel/seccomp.c @@ -8,6 +8,7 @@ #include #include +#include /* #define SECCOMP_DEBUG 1 */ #define NR_SECCOMP_MODES 1 @@ -22,7 +23,7 @@ static int mode1_syscalls[] = { 0, /* null terminated */ }; -#ifdef TIF_32BIT +#ifdef CONFIG_COMPAT static int mode1_syscalls_32[] = { __NR_seccomp_read_32, __NR_seccomp_write_32, __NR_seccomp_exit_32, __NR_seccomp_sigreturn_32, 0, /* null terminated */ @@ -37,8 +38,8 @@ void __secure_computing(int this_syscall) switch (mode) { case 1: syscall = mode1_syscalls; -#ifdef TIF_32BIT - if (test_thread_flag(TIF_32BIT)) +#ifdef CONFIG_COMPAT + if (is_compat_task()) syscall = mode1_syscalls_32; #endif do { -- cgit v1.2.3