diff options
Diffstat (limited to 'arch/x86')
162 files changed, 3181 insertions, 2139 deletions
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 52421d5..f46f30d 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -24,18 +24,21 @@ config X86 select HAVE_UNSTABLE_SCHED_CLOCK select HAVE_IDE select HAVE_OPROFILE + select HAVE_PERF_COUNTERS if (!M386 && !M486) select HAVE_IOREMAP_PROT select HAVE_KPROBES select ARCH_WANT_OPTIONAL_GPIOLIB select ARCH_WANT_FRAME_POINTERS + select HAVE_DMA_ATTRS select HAVE_KRETPROBES select HAVE_FTRACE_MCOUNT_RECORD select HAVE_DYNAMIC_FTRACE select HAVE_FUNCTION_TRACER select HAVE_FUNCTION_GRAPH_TRACER + select HAVE_FUNCTION_GRAPH_FP_TEST select HAVE_FUNCTION_TRACE_MCOUNT_TEST select HAVE_FTRACE_NMI_ENTER if DYNAMIC_FTRACE - select HAVE_FTRACE_SYSCALLS + select HAVE_SYSCALL_TRACEPOINTS select HAVE_KVM select HAVE_ARCH_KGDB select HAVE_ARCH_TRACEHOOK @@ -741,7 +744,6 @@ config X86_UP_IOAPIC config X86_LOCAL_APIC def_bool y depends on X86_64 || SMP || X86_32_NON_STANDARD || X86_UP_APIC - select HAVE_PERF_COUNTERS if (!M386 && !M486) config X86_IO_APIC def_bool y @@ -1912,25 +1914,26 @@ config DMAR_DEFAULT_ON recommended you say N here while the DMAR code remains experimental. -config DMAR_GFX_WA - def_bool y - prompt "Support for Graphics workaround" +config DMAR_BROKEN_GFX_WA + def_bool n + prompt "Workaround broken graphics drivers (going away soon)" depends on DMAR ---help--- Current Graphics drivers tend to use physical address for DMA and avoid using DMA APIs. Setting this config option permits the IOMMU driver to set a unity map for all the OS-visible memory. Hence the driver can continue - to use physical addresses for DMA. + to use physical addresses for DMA, at least until this + option is removed in the 2.6.32 kernel. config DMAR_FLOPPY_WA def_bool y depends on DMAR ---help--- - Floppy disk drivers are know to bypass DMA API calls + Floppy disk drivers are known to bypass DMA API calls thereby failing to work when IOMMU is enabled. This workaround will setup a 1:1 mapping for the first - 16M to make floppy (an ISA device) work. + 16MiB to make floppy (an ISA device) work. config INTR_REMAP bool "Support for Interrupt Remapping (EXPERIMENTAL)" diff --git a/arch/x86/boot/Makefile b/arch/x86/boot/Makefile index 8d16ada..ec749c2 100644 --- a/arch/x86/boot/Makefile +++ b/arch/x86/boot/Makefile @@ -70,6 +70,7 @@ KBUILD_CFLAGS := $(LINUXINCLUDE) -g -Os -D_SETUP -D__KERNEL__ \ $(call cc-option, -mpreferred-stack-boundary=2) KBUILD_CFLAGS += $(call cc-option, -m32) KBUILD_AFLAGS := $(KBUILD_CFLAGS) -D__ASSEMBLY__ +GCOV_PROFILE := n $(obj)/bzImage: asflags-y := $(SVGA_MODE) diff --git a/arch/x86/boot/bioscall.S b/arch/x86/boot/bioscall.S index 5077937..1dfbf64 100644 --- a/arch/x86/boot/bioscall.S +++ b/arch/x86/boot/bioscall.S @@ -13,7 +13,7 @@ * touching registers they shouldn't be. */ - .code16 + .code16gcc .text .globl intcall .type intcall, @function diff --git a/arch/x86/boot/compressed/Makefile b/arch/x86/boot/compressed/Makefile index 49c8a4c..f8ed065 100644 --- a/arch/x86/boot/compressed/Makefile +++ b/arch/x86/boot/compressed/Makefile @@ -4,7 +4,7 @@ # create a compressed vmlinux image from the original vmlinux # -targets := vmlinux vmlinux.bin vmlinux.bin.gz vmlinux.bin.bz2 vmlinux.bin.lzma head_$(BITS).o misc.o piggy.o +targets := vmlinux.lds vmlinux vmlinux.bin vmlinux.bin.gz vmlinux.bin.bz2 vmlinux.bin.lzma head_$(BITS).o misc.o piggy.o KBUILD_CFLAGS := -m$(BITS) -D__KERNEL__ $(LINUX_INCLUDE) -O2 KBUILD_CFLAGS += -fno-strict-aliasing -fPIC @@ -15,6 +15,7 @@ KBUILD_CFLAGS += $(call cc-option,-ffreestanding) KBUILD_CFLAGS += $(call cc-option,-fno-stack-protector) KBUILD_AFLAGS := $(KBUILD_CFLAGS) -D__ASSEMBLY__ +GCOV_PROFILE := n LDFLAGS := -m elf_$(UTS_MACHINE) LDFLAGS_vmlinux := -T diff --git a/arch/x86/boot/video-bios.c b/arch/x86/boot/video-bios.c index d660be4..49e0c18 100644 --- a/arch/x86/boot/video-bios.c +++ b/arch/x86/boot/video-bios.c @@ -37,14 +37,13 @@ static int set_bios_mode(u8 mode) ireg.al = mode; /* AH=0x00 Set Video Mode */ intcall(0x10, &ireg, NULL); - ireg.ah = 0x0f; /* Get Current Video Mode */ intcall(0x10, &ireg, &oreg); do_restore = 1; /* Assume video contents were lost */ /* Not all BIOSes are clean with the top bit */ - new_mode = ireg.al & 0x7f; + new_mode = oreg.al & 0x7f; if (new_mode == mode) return 0; /* Mode change OK */ diff --git a/arch/x86/boot/video-vesa.c b/arch/x86/boot/video-vesa.c index c700147..275dd17 100644 --- a/arch/x86/boot/video-vesa.c +++ b/arch/x86/boot/video-vesa.c @@ -45,7 +45,7 @@ static int vesa_probe(void) ireg.di = (size_t)&vginfo; intcall(0x10, &ireg, &oreg); - if (ireg.ax != 0x004f || + if (oreg.ax != 0x004f || vginfo.signature != VESA_MAGIC || vginfo.version < 0x0102) return 0; /* Not present */ @@ -70,7 +70,7 @@ static int vesa_probe(void) ireg.di = (size_t)&vminfo; intcall(0x10, &ireg, &oreg); - if (ireg.ax != 0x004f) + if (oreg.ax != 0x004f) continue; if ((vminfo.mode_attr & 0x15) == 0x05) { diff --git a/arch/x86/configs/i386_defconfig b/arch/x86/configs/i386_defconfig index edb992e..d28fad1 100644 --- a/arch/x86/configs/i386_defconfig +++ b/arch/x86/configs/i386_defconfig @@ -2355,7 +2355,7 @@ CONFIG_HAVE_FUNCTION_TRACE_MCOUNT_TEST=y CONFIG_HAVE_DYNAMIC_FTRACE=y CONFIG_HAVE_FTRACE_MCOUNT_RECORD=y CONFIG_HAVE_HW_BRANCH_TRACER=y -CONFIG_HAVE_FTRACE_SYSCALLS=y +CONFIG_HAVE_SYSCALL_TRACEPOINTS=y CONFIG_RING_BUFFER=y CONFIG_TRACING=y CONFIG_TRACING_SUPPORT=y diff --git a/arch/x86/configs/x86_64_defconfig b/arch/x86/configs/x86_64_defconfig index cee1dd2..6c86acd 100644 --- a/arch/x86/configs/x86_64_defconfig +++ b/arch/x86/configs/x86_64_defconfig @@ -2329,7 +2329,7 @@ CONFIG_HAVE_FUNCTION_TRACE_MCOUNT_TEST=y CONFIG_HAVE_DYNAMIC_FTRACE=y CONFIG_HAVE_FTRACE_MCOUNT_RECORD=y CONFIG_HAVE_HW_BRANCH_TRACER=y -CONFIG_HAVE_FTRACE_SYSCALLS=y +CONFIG_HAVE_SYSCALL_TRACEPOINTS=y CONFIG_RING_BUFFER=y CONFIG_TRACING=y CONFIG_TRACING_SUPPORT=y diff --git a/arch/x86/crypto/aesni-intel_asm.S b/arch/x86/crypto/aesni-intel_asm.S index caba996..eb0566e 100644 --- a/arch/x86/crypto/aesni-intel_asm.S +++ b/arch/x86/crypto/aesni-intel_asm.S @@ -845,7 +845,7 @@ ENTRY(aesni_cbc_enc) */ ENTRY(aesni_cbc_dec) cmp $16, LEN - jb .Lcbc_dec_ret + jb .Lcbc_dec_just_ret mov 480(KEYP), KLEN add $240, KEYP movups (IVP), IV @@ -891,6 +891,7 @@ ENTRY(aesni_cbc_dec) add $16, OUTP cmp $16, LEN jge .Lcbc_dec_loop1 - movups IV, (IVP) .Lcbc_dec_ret: + movups IV, (IVP) +.Lcbc_dec_just_ret: ret diff --git a/arch/x86/crypto/aesni-intel_glue.c b/arch/x86/crypto/aesni-intel_glue.c index 4e66339..c580c5e 100644 --- a/arch/x86/crypto/aesni-intel_glue.c +++ b/arch/x86/crypto/aesni-intel_glue.c @@ -198,6 +198,7 @@ static int ecb_encrypt(struct blkcipher_desc *desc, blkcipher_walk_init(&walk, dst, src, nbytes); err = blkcipher_walk_virt(desc, &walk); + desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; kernel_fpu_begin(); while ((nbytes = walk.nbytes)) { @@ -221,6 +222,7 @@ static int ecb_decrypt(struct blkcipher_desc *desc, blkcipher_walk_init(&walk, dst, src, nbytes); err = blkcipher_walk_virt(desc, &walk); + desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; kernel_fpu_begin(); while ((nbytes = walk.nbytes)) { @@ -266,6 +268,7 @@ static int cbc_encrypt(struct blkcipher_desc *desc, blkcipher_walk_init(&walk, dst, src, nbytes); err = blkcipher_walk_virt(desc, &walk); + desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; kernel_fpu_begin(); while ((nbytes = walk.nbytes)) { @@ -289,6 +292,7 @@ static int cbc_decrypt(struct blkcipher_desc *desc, blkcipher_walk_init(&walk, dst, src, nbytes); err = blkcipher_walk_virt(desc, &walk); + desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP; kernel_fpu_begin(); while ((nbytes = walk.nbytes)) { diff --git a/arch/x86/crypto/fpu.c b/arch/x86/crypto/fpu.c index 5f9781a..daef6cd 100644 --- a/arch/x86/crypto/fpu.c +++ b/arch/x86/crypto/fpu.c @@ -48,7 +48,7 @@ static int crypto_fpu_encrypt(struct blkcipher_desc *desc_in, struct blkcipher_desc desc = { .tfm = child, .info = desc_in->info, - .flags = desc_in->flags, + .flags = desc_in->flags & ~CRYPTO_TFM_REQ_MAY_SLEEP, }; kernel_fpu_begin(); @@ -67,7 +67,7 @@ static int crypto_fpu_decrypt(struct blkcipher_desc *desc_in, struct blkcipher_desc desc = { .tfm = child, .info = desc_in->info, - .flags = desc_in->flags, + .flags = desc_in->flags & ~CRYPTO_TFM_REQ_MAY_SLEEP, }; kernel_fpu_begin(); diff --git a/arch/x86/include/asm/acpi.h b/arch/x86/include/asm/acpi.h index 4518dc5..20d1465 100644 --- a/arch/x86/include/asm/acpi.h +++ b/arch/x86/include/asm/acpi.h @@ -144,6 +144,7 @@ static inline unsigned int acpi_processor_cstate_check(unsigned int max_cstate) #else /* !CONFIG_ACPI */ +#define acpi_disabled 1 #define acpi_lapic 0 #define acpi_ioapic 0 static inline void acpi_noirq_set(void) { } diff --git a/arch/x86/include/asm/amd_iommu.h b/arch/x86/include/asm/amd_iommu.h index 262e028..bdf96f1 100644 --- a/arch/x86/include/asm/amd_iommu.h +++ b/arch/x86/include/asm/amd_iommu.h @@ -29,9 +29,11 @@ extern void amd_iommu_detect(void); extern irqreturn_t amd_iommu_int_handler(int irq, void *data); extern void amd_iommu_flush_all_domains(void); extern void amd_iommu_flush_all_devices(void); +extern void amd_iommu_shutdown(void); #else static inline int amd_iommu_init(void) { return -ENODEV; } static inline void amd_iommu_detect(void) { } +static inline void amd_iommu_shutdown(void) { } #endif #endif /* _ASM_X86_AMD_IOMMU_H */ diff --git a/arch/x86/include/asm/atomic_32.h b/arch/x86/include/asm/atomic_32.h index 8cb9c81..dc5a667 100644 --- a/arch/x86/include/asm/atomic_32.h +++ b/arch/x86/include/asm/atomic_32.h @@ -19,7 +19,10 @@ * * Atomically reads the value of @v. */ -#define atomic_read(v) ((v)->counter) +static inline int atomic_read(const atomic_t *v) +{ + return v->counter; +} /** * atomic_set - set atomic variable @@ -28,7 +31,10 @@ * * Atomically sets the value of @v to @i. */ -#define atomic_set(v, i) (((v)->counter) = (i)) +static inline void atomic_set(atomic_t *v, int i) +{ + v->counter = i; +} /** * atomic_add - add integer to atomic variable @@ -200,8 +206,15 @@ static inline int atomic_sub_return(int i, atomic_t *v) return atomic_add_return(-i, v); } -#define atomic_cmpxchg(v, old, new) (cmpxchg(&((v)->counter), (old), (new))) -#define atomic_xchg(v, new) (xchg(&((v)->counter), (new))) +static inline int atomic_cmpxchg(atomic_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); +} /** * atomic_add_unless - add unless the number is already a given value @@ -250,67 +263,22 @@ static inline int atomic_add_unless(atomic_t *v, int a, int u) /* An 64bit atomic type */ typedef struct { - unsigned long long counter; + u64 __aligned(8) counter; } atomic64_t; #define ATOMIC64_INIT(val) { (val) } -/** - * atomic64_read - read atomic64 variable - * @v: pointer of type atomic64_t - * - * Atomically reads the value of @v. - * Doesn't imply a read memory barrier. - */ -#define __atomic64_read(ptr) ((ptr)->counter) - -static inline unsigned long long -cmpxchg8b(unsigned long long *ptr, unsigned long long old, unsigned long long new) -{ - asm volatile( - - LOCK_PREFIX "cmpxchg8b (%[ptr])\n" - - : "=A" (old) - - : [ptr] "D" (ptr), - "A" (old), - "b" (ll_low(new)), - "c" (ll_high(new)) - - : "memory"); - - return old; -} - -static inline unsigned long long -atomic64_cmpxchg(atomic64_t *ptr, unsigned long long old_val, - unsigned long long new_val) -{ - return cmpxchg8b(&ptr->counter, old_val, new_val); -} +extern u64 atomic64_cmpxchg(atomic64_t *ptr, u64 old_val, u64 new_val); /** * atomic64_xchg - xchg atomic64 variable * @ptr: pointer to type atomic64_t * @new_val: value to assign - * @old_val: old value that was there * * Atomically xchgs the value of @ptr to @new_val and returns * the old value. */ - -static inline unsigned long long -atomic64_xchg(atomic64_t *ptr, unsigned long long new_val) -{ - unsigned long long old_val; - - do { - old_val = atomic_read(ptr); - } while (atomic64_cmpxchg(ptr, old_val, new_val) != old_val); - - return old_val; -} +extern u64 atomic64_xchg(atomic64_t *ptr, u64 new_val); /** * atomic64_set - set atomic64 variable @@ -319,10 +287,7 @@ atomic64_xchg(atomic64_t *ptr, unsigned long long new_val) * * Atomically sets the value of @ptr to @new_val. */ -static inline void atomic64_set(atomic64_t *ptr, unsigned long long new_val) -{ - atomic64_xchg(ptr, new_val); -} +extern void atomic64_set(atomic64_t *ptr, u64 new_val); /** * atomic64_read - read atomic64 variable @@ -330,17 +295,30 @@ static inline void atomic64_set(atomic64_t *ptr, unsigned long long new_val) * * Atomically reads the value of @ptr and returns it. */ -static inline unsigned long long atomic64_read(atomic64_t *ptr) +static inline u64 atomic64_read(atomic64_t *ptr) { - unsigned long long curr_val; - - do { - curr_val = __atomic64_read(ptr); - } while (atomic64_cmpxchg(ptr, curr_val, curr_val) != curr_val); - - return curr_val; + u64 res; + + /* + * Note, we inline this atomic64_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; } +extern u64 atomic64_read(atomic64_t *ptr); + /** * atomic64_add_return - add and return * @delta: integer value to add @@ -348,34 +326,14 @@ static inline unsigned long long atomic64_read(atomic64_t *ptr) * * Atomically adds @delta to @ptr and returns @delta + *@ptr */ -static inline unsigned long long -atomic64_add_return(unsigned long long delta, atomic64_t *ptr) -{ - unsigned long long old_val, new_val; - - do { - old_val = atomic_read(ptr); - new_val = old_val + delta; - - } while (atomic64_cmpxchg(ptr, old_val, new_val) != old_val); - - return new_val; -} - -static inline long atomic64_sub_return(unsigned long long delta, atomic64_t *ptr) -{ - return atomic64_add_return(-delta, ptr); -} +extern u64 atomic64_add_return(u64 delta, atomic64_t *ptr); -static inline long atomic64_inc_return(atomic64_t *ptr) -{ - return atomic64_add_return(1, ptr); -} - -static inline long atomic64_dec_return(atomic64_t *ptr) -{ - return atomic64_sub_return(1, ptr); -} +/* + * Other variants with different arithmetic operators: + */ +extern u64 atomic64_sub_return(u64 delta, atomic64_t *ptr); +extern u64 atomic64_inc_return(atomic64_t *ptr); +extern u64 atomic64_dec_return(atomic64_t *ptr); /** * atomic64_add - add integer to atomic64 variable @@ -384,10 +342,7 @@ static inline long atomic64_dec_return(atomic64_t *ptr) * * Atomically adds @delta to @ptr. */ -static inline void atomic64_add(unsigned long long delta, atomic64_t *ptr) -{ - atomic64_add_return(delta, ptr); -} +extern void atomic64_add(u64 delta, atomic64_t *ptr); /** * atomic64_sub - subtract the atomic64 variable @@ -396,10 +351,7 @@ static inline void atomic64_add(unsigned long long delta, atomic64_t *ptr) * * Atomically subtracts @delta from @ptr. */ -static inline void atomic64_sub(unsigned long long delta, atomic64_t *ptr) -{ - atomic64_add(-delta, ptr); -} +extern void atomic64_sub(u64 delta, atomic64_t *ptr); /** * atomic64_sub_and_test - subtract value from variable and test result @@ -410,13 +362,7 @@ static inline void atomic64_sub(unsigned long long delta, atomic64_t *ptr) * true if the result is zero, or false for all * other cases. */ -static inline int -atomic64_sub_and_test(unsigned long long delta, atomic64_t *ptr) -{ - unsigned long long old_val = atomic64_sub_return(delta, ptr); - - return old_val == 0; -} +extern int atomic64_sub_and_test(u64 delta, atomic64_t *ptr); /** * atomic64_inc - increment atomic64 variable @@ -424,10 +370,7 @@ atomic64_sub_and_test(unsigned long long delta, atomic64_t *ptr) * * Atomically increments @ptr by 1. */ -static inline void atomic64_inc(atomic64_t *ptr) -{ - atomic64_add(1, ptr); -} +extern void atomic64_inc(atomic64_t *ptr); /** * atomic64_dec - decrement atomic64 variable @@ -435,10 +378,7 @@ static inline void atomic64_inc(atomic64_t *ptr) * * Atomically decrements @ptr by 1. */ -static inline void atomic64_dec(atomic64_t *ptr) -{ - atomic64_sub(1, ptr); -} +extern void atomic64_dec(atomic64_t *ptr); /** * atomic64_dec_and_test - decrement and test @@ -448,10 +388,7 @@ static inline void atomic64_dec(atomic64_t *ptr) * returns true if the result is 0, or false for all other * cases. */ -static inline int atomic64_dec_and_test(atomic64_t *ptr) -{ - return atomic64_sub_and_test(1, ptr); -} +extern int atomic64_dec_and_test(atomic64_t *ptr); /** * atomic64_inc_and_test - increment and test @@ -461,10 +398,7 @@ static inline int atomic64_dec_and_test(atomic64_t *ptr) * and returns true if the result is zero, or false for all * other cases. */ -static inline int atomic64_inc_and_test(atomic64_t *ptr) -{ - return atomic64_sub_and_test(-1, ptr); -} +extern int atomic64_inc_and_test(atomic64_t *ptr); /** * atomic64_add_negative - add and test if negative @@ -475,13 +409,7 @@ static inline int atomic64_inc_and_test(atomic64_t *ptr) * if the result is negative, or false when * result is greater than or equal to zero. */ -static inline int -atomic64_add_negative(unsigned long long delta, atomic64_t *ptr) -{ - long long old_val = atomic64_add_return(delta, ptr); - - return old_val < 0; -} +extern int atomic64_add_negative(u64 delta, atomic64_t *ptr); #include <asm-generic/atomic-long.h> #endif /* _ASM_X86_ATOMIC_32_H */ diff --git a/arch/x86/include/asm/atomic_64.h b/arch/x86/include/asm/atomic_64.h index 0d63602..d605dc2 100644 --- a/arch/x86/include/asm/atomic_64.h +++ b/arch/x86/include/asm/atomic_64.h @@ -18,7 +18,10 @@ * * Atomically reads the value of @v. */ -#define atomic_read(v) ((v)->counter) +static inline int atomic_read(const atomic_t *v) +{ + return v->counter; +} /** * atomic_set - set atomic variable @@ -27,7 +30,10 @@ * * Atomically sets the value of @v to @i. */ -#define atomic_set(v, i) (((v)->counter) = (i)) +static inline void atomic_set(atomic_t *v, int i) +{ + v->counter = i; +} /** * atomic_add - add integer to atomic variable @@ -192,7 +198,10 @@ static inline int atomic_sub_return(int i, atomic_t *v) * Atomically reads the value of @v. * Doesn't imply a read memory barrier. */ -#define atomic64_read(v) ((v)->counter) +static inline long atomic64_read(const atomic64_t *v) +{ + return v->counter; +} /** * atomic64_set - set atomic64 variable @@ -201,7 +210,10 @@ static inline int atomic_sub_return(int i, atomic_t *v) * * Atomically sets the value of @v to @i. */ -#define atomic64_set(v, i) (((v)->counter) = (i)) +static inline void atomic64_set(atomic64_t *v, long i) +{ + v->counter = i; +} /** * atomic64_add - add integer to atomic64 variable @@ -355,11 +367,25 @@ static inline long atomic64_sub_return(long i, atomic64_t *v) #define atomic64_inc_return(v) (atomic64_add_return(1, (v))) #define atomic64_dec_return(v) (atomic64_sub_return(1, (v))) -#define atomic64_cmpxchg(v, old, new) (cmpxchg(&((v)->counter), (old), (new))) -#define atomic64_xchg(v, new) (xchg(&((v)->counter), new)) +static inline long atomic64_cmpxchg(atomic64_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); +} -#define atomic_cmpxchg(v, old, new) (cmpxchg(&((v)->counter), (old), (new))) -#define atomic_xchg(v, new) (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_xchg(atomic_t *v, int new) +{ + return xchg(&v->counter, new); +} /** * atomic_add_unless - add unless the number is a given value diff --git a/arch/x86/include/asm/boot.h b/arch/x86/include/asm/boot.h index 418e632..7a10659 100644 --- a/arch/x86/include/asm/boot.h +++ b/arch/x86/include/asm/boot.h @@ -8,7 +8,7 @@ #ifdef __KERNEL__ -#include <asm/page_types.h> +#include <asm/pgtable_types.h> /* Physical address where kernel should be loaded. */ #define LOAD_PHYSICAL_ADDR ((CONFIG_PHYSICAL_START \ @@ -16,10 +16,10 @@ & ~(CONFIG_PHYSICAL_ALIGN - 1)) /* Minimum kernel alignment, as a power of two */ -#ifdef CONFIG_x86_64 +#ifdef CONFIG_X86_64 #define MIN_KERNEL_ALIGN_LG2 PMD_SHIFT #else -#define MIN_KERNEL_ALIGN_LG2 (PAGE_SHIFT+1) +#define MIN_KERNEL_ALIGN_LG2 (PAGE_SHIFT + THREAD_ORDER) #endif #define MIN_KERNEL_ALIGN (_AC(1, UL) << MIN_KERNEL_ALIGN_LG2) diff --git a/arch/x86/include/asm/desc.h b/arch/x86/include/asm/desc.h index c45f415..c993e9e 100644 --- a/arch/x86/include/asm/desc.h +++ b/arch/x86/include/asm/desc.h @@ -1,7 +1,6 @@ #ifndef _ASM_X86_DESC_H #define _ASM_X86_DESC_H -#ifndef __ASSEMBLY__ #include <asm/desc_defs.h> #include <asm/ldt.h> #include <asm/mmu.h> @@ -380,29 +379,4 @@ static inline void set_system_intr_gate_ist(int n, void *addr, unsigned ist) _set_gate(n, GATE_INTERRUPT, addr, 0x3, ist, __KERNEL_CS); } -#else -/* - * GET_DESC_BASE reads the descriptor base of the specified segment. - * - * Args: - * idx - descriptor index - * gdt - GDT pointer - * base - 32bit register to which the base will be written - * lo_w - lo word of the "base" register - * lo_b - lo byte of the "base" register - * hi_b - hi byte of the low word of the "base" register - * - * Example: - * GET_DESC_BASE(GDT_ENTRY_ESPFIX_SS, %ebx, %eax, %ax, %al, %ah) - * Will read the base address of GDT_ENTRY_ESPFIX_SS and put it into %eax. - */ -#define GET_DESC_BASE(idx, gdt, base, lo_w, lo_b, hi_b) \ - movb idx * 8 + 4(gdt), lo_b; \ - movb idx * 8 + 7(gdt), hi_b; \ - shll $16, base; \ - movw idx * 8 + 2(gdt), lo_w; - - -#endif /* __ASSEMBLY__ */ - #endif /* _ASM_X86_DESC_H */ diff --git a/arch/x86/include/asm/dma-mapping.h b/arch/x86/include/asm/dma-mapping.h index b93405b..1c3f943 100644 --- a/arch/x86/include/asm/dma-mapping.h +++ b/arch/x86/include/asm/dma-mapping.h @@ -33,6 +33,8 @@ static inline struct dma_map_ops *get_dma_ops(struct device *dev) #endif } +#include <asm-generic/dma-mapping-common.h> + /* Make sure we keep the same behaviour */ static inline int dma_mapping_error(struct device *dev, dma_addr_t dma_addr) { @@ -53,177 +55,6 @@ extern int dma_set_mask(struct device *dev, u64 mask); extern void *dma_generic_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_addr, gfp_t flag); -static inline dma_addr_t -dma_map_single(struct device *hwdev, void *ptr, size_t size, - enum dma_data_direction dir) -{ - struct dma_map_ops *ops = get_dma_ops(hwdev); - dma_addr_t addr; - - BUG_ON(!valid_dma_direction(dir)); - kmemcheck_mark_initialized(ptr, size); - addr = ops->map_page(hwdev, virt_to_page(ptr), - (unsigned long)ptr & ~PAGE_MASK, size, - dir, NULL); - debug_dma_map_page(hwdev, virt_to_page(ptr), - (unsigned long)ptr & ~PAGE_MASK, size, - dir, addr, true); - return addr; -} - -static inline void -dma_unmap_single(struct device *dev, dma_addr_t addr, size_t size, - enum dma_data_direction dir) -{ - struct dma_map_ops *ops = get_dma_ops(dev); - - BUG_ON(!valid_dma_direction(dir)); - if (ops->unmap_page) - ops->unmap_page(dev, addr, size, dir, NULL); - debug_dma_unmap_page(dev, addr, size, dir, true); -} - -static inline int -dma_map_sg(struct device *hwdev, struct scatterlist *sg, - int nents, enum dma_data_direction dir) -{ - struct dma_map_ops *ops = get_dma_ops(hwdev); - int ents; - struct scatterlist *s; - int i; - - BUG_ON(!valid_dma_direction(dir)); - for_each_sg(sg, s, nents, i) - kmemcheck_mark_initialized(sg_virt(s), s->length); - ents = ops->map_sg(hwdev, sg, nents, dir, NULL); - debug_dma_map_sg(hwdev, sg, nents, ents, dir); - - return ents; -} - -static inline void -dma_unmap_sg(struct device *hwdev, struct scatterlist *sg, int nents, - enum dma_data_direction dir) -{ - struct dma_map_ops *ops = get_dma_ops(hwdev); - - BUG_ON(!valid_dma_direction(dir)); - debug_dma_unmap_sg(hwdev, sg, nents, dir); - if (ops->unmap_sg) - ops->unmap_sg(hwdev, sg, nents, dir, NULL); -} - -static inline void -dma_sync_single_for_cpu(struct device *hwdev, dma_addr_t dma_handle, - size_t size, enum dma_data_direction dir) -{ - struct dma_map_ops *ops = get_dma_ops(hwdev); - - BUG_ON(!valid_dma_direction(dir)); - if (ops->sync_single_for_cpu) - ops->sync_single_for_cpu(hwdev, dma_handle, size, dir); - debug_dma_sync_single_for_cpu(hwdev, dma_handle, size, dir); - flush_write_buffers(); -} - -static inline void -dma_sync_single_for_device(struct device *hwdev, dma_addr_t dma_handle, - size_t size, enum dma_data_direction dir) -{ - struct dma_map_ops *ops = get_dma_ops(hwdev); - - BUG_ON(!valid_dma_direction(dir)); - if (ops->sync_single_for_device) - ops->sync_single_for_device(hwdev, dma_handle, size, dir); - debug_dma_sync_single_for_device(hwdev, dma_handle, size, dir); - flush_write_buffers(); -} - -static inline void -dma_sync_single_range_for_cpu(struct device *hwdev, dma_addr_t dma_handle, - unsigned long offset, size_t size, - enum dma_data_direction dir) -{ - struct dma_map_ops *ops = get_dma_ops(hwdev); - - BUG_ON(!valid_dma_direction(dir)); - if (ops->sync_single_range_for_cpu) - ops->sync_single_range_for_cpu(hwdev, dma_handle, offset, - size, dir); - debug_dma_sync_single_range_for_cpu(hwdev, dma_handle, - offset, size, dir); - flush_write_buffers(); -} - -static inline void -dma_sync_single_range_for_device(struct device *hwdev, dma_addr_t dma_handle, - unsigned long offset, size_t size, - enum dma_data_direction dir) -{ - struct dma_map_ops *ops = get_dma_ops(hwdev); - - BUG_ON(!valid_dma_direction(dir)); - if (ops->sync_single_range_for_device) - ops->sync_single_range_for_device(hwdev, dma_handle, - offset, size, dir); - debug_dma_sync_single_range_for_device(hwdev, dma_handle, - offset, size, dir); - flush_write_buffers(); -} - -static inline void -dma_sync_sg_for_cpu(struct device *hwdev, struct scatterlist *sg, - int nelems, enum dma_data_direction dir) -{ - struct dma_map_ops *ops = get_dma_ops(hwdev); - - BUG_ON(!valid_dma_direction(dir)); - if (ops->sync_sg_for_cpu) - ops->sync_sg_for_cpu(hwdev, sg, nelems, dir); - debug_dma_sync_sg_for_cpu(hwdev, sg, nelems, dir); - flush_write_buffers(); -} - -static inline void -dma_sync_sg_for_device(struct device *hwdev, struct scatterlist *sg, - int nelems, enum dma_data_direction dir) -{ - struct dma_map_ops *ops = get_dma_ops(hwdev); - - BUG_ON(!valid_dma_direction(dir)); - if (ops->sync_sg_for_device) - ops->sync_sg_for_device(hwdev, sg, nelems, dir); - debug_dma_sync_sg_for_device(hwdev, sg, nelems, dir); - - flush_write_buffers(); -} - -static inline dma_addr_t dma_map_page(struct device *dev, struct page *page, - size_t offset, size_t size, - enum dma_data_direction dir) -{ - struct dma_map_ops *ops = get_dma_ops(dev); - dma_addr_t addr; - - BUG_ON(!valid_dma_direction(dir)); - kmemcheck_mark_initialized(page_address(page) + offset, size); - addr = ops->map_page(dev, page, offset, size, dir, NULL); - debug_dma_map_page(dev, page, offset, size, dir, addr, false); - - return addr; -} - -static inline void dma_unmap_page(struct device *dev, dma_addr_t addr, - size_t size, enum dma_data_direction dir) -{ - struct dma_map_ops *ops = get_dma_ops(dev); - - BUG_ON(!valid_dma_direction(dir)); - if (ops->unmap_page) - ops->unmap_page(dev, addr, size, dir, NULL); - debug_dma_unmap_page(dev, addr, size, dir, false); -} - static inline void dma_cache_sync(struct device *dev, void *vaddr, size_t size, enum dma_data_direction dir) diff --git a/arch/x86/include/asm/efi.h b/arch/x86/include/asm/efi.h index edc90f2..8406ed7 100644 --- a/arch/x86/include/asm/efi.h +++ b/arch/x86/include/asm/efi.h @@ -33,7 +33,7 @@ extern unsigned long asmlinkage efi_call_phys(void *, ...); #define efi_call_virt6(f, a1, a2, a3, a4, a5, a6) \ efi_call_virt(f, a1, a2, a3, a4, a5, a6) -#define efi_ioremap(addr, size) ioremap_cache(addr, size) +#define efi_ioremap(addr, size, type) ioremap_cache(addr, size) #else /* !CONFIG_X86_32 */ @@ -84,7 +84,8 @@ extern u64 efi_call6(void *fp, u64 arg1, u64 arg2, u64 arg3, efi_call6((void *)(efi.systab->runtime->f), (u64)(a1), (u64)(a2), \ (u64)(a3), (u64)(a4), (u64)(a5), (u64)(a6)) -extern void __iomem *efi_ioremap(unsigned long addr, unsigned long size); +extern void __iomem *efi_ioremap(unsigned long addr, unsigned long size, + u32 type); #endif /* CONFIG_X86_32 */ diff --git a/arch/x86/include/asm/fixmap.h b/arch/x86/include/asm/fixmap.h index 2d81af3..7b2d71d 100644 --- a/arch/x86/include/asm/fixmap.h +++ b/arch/x86/include/asm/fixmap.h @@ -111,12 +111,9 @@ enum fixed_addresses { #ifdef CONFIG_PARAVIRT FIX_PARAVIRT_BOOTMAP, #endif - FIX_TEXT_POKE0, /* reserve 2 pages for text_poke() */ - FIX_TEXT_POKE1, + FIX_TEXT_POKE1, /* reserve 2 pages for text_poke() */ + FIX_TEXT_POKE0, /* first page is last, because allocation is backward */ __end_of_permanent_fixed_addresses, -#ifdef CONFIG_PROVIDE_OHCI1394_DMA_INIT - FIX_OHCI1394_BASE, -#endif /* * 256 temporary boot-time mappings, used by early_ioremap(), * before ioremap() is functional. @@ -129,6 +126,9 @@ enum fixed_addresses { FIX_BTMAP_END = __end_of_permanent_fixed_addresses + 256 - (__end_of_permanent_fixed_addresses & 255), FIX_BTMAP_BEGIN = FIX_BTMAP_END + NR_FIX_BTMAPS*FIX_BTMAPS_SLOTS - 1, +#ifdef CONFIG_PROVIDE_OHCI1394_DMA_INIT + FIX_OHCI1394_BASE, +#endif #ifdef CONFIG_X86_32 FIX_WP_TEST, #endif diff --git a/arch/x86/include/asm/ftrace.h b/arch/x86/include/asm/ftrace.h index bd2c651..db24c22 100644 --- a/arch/x86/include/asm/ftrace.h +++ b/arch/x86/include/asm/ftrace.h @@ -28,13 +28,6 @@ #endif -/* FIXME: I don't want to stay hardcoded */ -#ifdef CONFIG_X86_64 -# define FTRACE_SYSCALL_MAX 296 -#else -# define FTRACE_SYSCALL_MAX 333 -#endif - #ifdef CONFIG_FUNCTION_TRACER #define MCOUNT_ADDR ((long)(mcount)) #define MCOUNT_INSN_SIZE 5 /* sizeof mcount call */ diff --git a/arch/x86/include/asm/io_apic.h b/arch/x86/include/asm/io_apic.h index daf866e..330ee80 100644 --- a/arch/x86/include/asm/io_apic.h +++ b/arch/x86/include/asm/io_apic.h @@ -161,6 +161,7 @@ extern int io_apic_set_pci_routing(struct device *dev, int irq, struct io_apic_irq_attr *irq_attr); extern int (*ioapic_renumber_irq)(int ioapic, int irq); extern void ioapic_init_mappings(void); +extern void ioapic_insert_resources(void); extern struct IO_APIC_route_entry **alloc_ioapic_entries(void); extern void free_ioapic_entries(struct IO_APIC_route_entry **ioapic_entries); @@ -180,6 +181,7 @@ extern void ioapic_write_entry(int apic, int pin, #define io_apic_assign_pci_irqs 0 static const int timer_through_8259 = 0; static inline void ioapic_init_mappings(void) { } +static inline void ioapic_insert_resources(void) { } static inline void probe_nr_irqs_gsi(void) { } #endif diff --git a/arch/x86/include/asm/iommu.h b/arch/x86/include/asm/iommu.h index af326a2..fd6d21b 100644 --- a/arch/x86/include/asm/iommu.h +++ b/arch/x86/include/asm/iommu.h @@ -6,6 +6,7 @@ extern void no_iommu_init(void); extern struct dma_map_ops nommu_dma_ops; extern int force_iommu, no_iommu; extern int iommu_detected; +extern int iommu_pass_through; /* 10 seconds */ #define DMAR_OPERATION_TIMEOUT ((cycles_t) tsc_khz*10*1000) diff --git a/arch/x86/include/asm/irqflags.h b/arch/x86/include/asm/irqflags.h index 2bdab21..c6ccbe7 100644 --- a/arch/x86/include/asm/irqflags.h +++ b/arch/x86/include/asm/irqflags.h @@ -12,9 +12,15 @@ static inline unsigned long native_save_fl(void) { unsigned long flags; + /* + * Note: this needs to be "=r" not "=rm", because we have the + * stack offset from what gcc expects at the time the "pop" is + * executed, and so a memory reference with respect to the stack + * would end up using the wrong address. + */ asm volatile("# __raw_save_flags\n\t" "pushf ; pop %0" - : "=g" (flags) + : "=r" (flags) : /* no input */ : "memory"); diff --git a/arch/x86/include/asm/lguest.h b/arch/x86/include/asm/lguest.h index 313389c..5136dad 100644 --- a/arch/x86/include/asm/lguest.h +++ b/arch/x86/include/asm/lguest.h @@ -17,8 +17,7 @@ /* Pages for switcher itself, then two pages per cpu */ #define TOTAL_SWITCHER_PAGES (SHARED_SWITCHER_PAGES + 2 * nr_cpu_ids) -/* We map at -4M (-2M when PAE is activated) for ease of mapping - * into the guest (one PTE page). */ +/* We map at -4M (-2M for PAE) for ease of mapping (one PTE page). */ #ifdef CONFIG_X86_PAE #define SWITCHER_ADDR 0xFFE00000 #else diff --git a/arch/x86/include/asm/lguest_hcall.h b/arch/x86/include/asm/lguest_hcall.h index d31c4a6..ba0eed8 100644 --- a/arch/x86/include/asm/lguest_hcall.h +++ b/arch/x86/include/asm/lguest_hcall.h @@ -30,27 +30,27 @@ #include <asm/hw_irq.h> #include <asm/kvm_para.h> -/*G:031 But first, how does our Guest contact the Host to ask for privileged +/*G:030 + * But first, how does our Guest contact the Host to ask for privileged * operations? There are two ways: the direct way is to make a "hypercall", * to make requests of the Host Itself. * - * We use the KVM hypercall mechanism. Seventeen hypercalls are - * available: the hypercall number is put in the %eax register, and the - * arguments (when required) are placed in %ebx, %ecx, %edx and %esi. - * If a return value makes sense, it's returned in %eax. + * We use the KVM hypercall mechanism, though completely different hypercall + * numbers. Seventeen hypercalls are available: the hypercall number is put in + * the %eax register, and the arguments (when required) are placed in %ebx, + * %ecx, %edx and %esi. If a return value makes sense, it's returned in %eax. * * Grossly invalid calls result in Sudden Death at the hands of the vengeful * Host, rather than returning failure. This reflects Winston Churchill's - * definition of a gentleman: "someone who is only rude intentionally". */ -/*:*/ + * definition of a gentleman: "someone who is only rude intentionally". +:*/ /* Can't use our min() macro here: needs to be a constant */ #define LGUEST_IRQS (NR_IRQS < 32 ? NR_IRQS: 32) #define LHCALL_RING_SIZE 64 struct hcall_args { - /* These map directly onto eax, ebx, ecx, edx and esi - * in struct lguest_regs */ + /* These map directly onto eax/ebx/ecx/edx/esi in struct lguest_regs */ unsigned long arg0, arg1, arg2, arg3, arg4; }; diff --git a/arch/x86/include/asm/mce.h b/arch/x86/include/asm/mce.h index 540a466..5cdd8d1 100644 --- a/arch/x86/include/asm/mce.h +++ b/arch/x86/include/asm/mce.h @@ -102,15 +102,39 @@ struct mce_log { #ifdef __KERNEL__ +#include <linux/percpu.h> +#include <linux/init.h> +#include <asm/atomic.h> + extern int mce_disabled; +extern int mce_p5_enabled; -#include <asm/atomic.h> -#include <linux/percpu.h> +#ifdef CONFIG_X86_MCE +void mcheck_init(struct cpuinfo_x86 *c); +#else +static inline void mcheck_init(struct cpuinfo_x86 *c) {} +#endif + +#ifdef CONFIG_X86_OLD_MCE +extern int nr_mce_banks; +void amd_mcheck_init(struct cpuinfo_x86 *c); +void intel_p4_mcheck_init(struct cpuinfo_x86 *c); +void intel_p6_mcheck_init(struct cpuinfo_x86 *c); +#endif + +#ifdef CONFIG_X86_ANCIENT_MCE +void intel_p5_mcheck_init(struct cpuinfo_x86 *c); +void winchip_mcheck_init(struct cpuinfo_x86 *c); +static inline void enable_p5_mce(void) { mce_p5_enabled = 1; } +#else +static inline void intel_p5_mcheck_init(struct cpuinfo_x86 *c) {} +static inline void winchip_mcheck_init(struct cpuinfo_x86 *c) {} +static inline void enable_p5_mce(void) {} +#endif void mce_setup(struct mce *m); void mce_log(struct mce *m); DECLARE_PER_CPU(struct sys_device, mce_dev); -extern void (*threshold_cpu_callback)(unsigned long action, unsigned int cpu); /* * To support more than 128 would need to escape the predefined @@ -145,12 +169,8 @@ int mce_available(struct cpuinfo_x86 *c); DECLARE_PER_CPU(unsigned, mce_exception_count); DECLARE_PER_CPU(unsigned, mce_poll_count); -void mce_log_therm_throt_event(__u64 status); - extern atomic_t mce_entry; -void do_machine_check(struct pt_regs *, long); - typedef DECLARE_BITMAP(mce_banks_t, MAX_NR_BANKS); DECLARE_PER_CPU(mce_banks_t, mce_poll_banks); @@ -167,13 +187,32 @@ void mce_notify_process(void); DECLARE_PER_CPU(struct mce, injectm); extern struct file_operations mce_chrdev_ops; -#ifdef CONFIG_X86_MCE -void mcheck_init(struct cpuinfo_x86 *c); -#else -#define mcheck_init(c) do { } while (0) -#endif +/* + * Exception handler + */ + +/* Call the installed machine check handler for this CPU setup. */ +extern void (*machine_check_vector)(struct pt_regs *, long error_code); +void do_machine_check(struct pt_regs *, long); + +/* + * Threshold handler + */ extern void (*mce_threshold_vector)(void); +extern void (*threshold_cpu_callback)(unsigned long action, unsigned int cpu); + +/* + * Thermal handler + */ + +void intel_init_thermal(struct cpuinfo_x86 *c); + +#ifdef CONFIG_X86_NEW_MCE +void mce_log_therm_throt_event(__u64 status); +#else +static inline void mce_log_therm_throt_event(__u64 status) {} +#endif #endif /* __KERNEL__ */ #endif /* _ASM_X86_MCE_H */ diff --git a/arch/x86/include/asm/msr-index.h b/arch/x86/include/asm/msr-index.h index 1692fb5..6be7fc2 100644 --- a/arch/x86/include/asm/msr-index.h +++ b/arch/x86/include/asm/msr-index.h @@ -246,10 +246,6 @@ #define MSR_IA32_MISC_ENABLE_TURBO_DISABLE (1ULL << 38) #define MSR_IA32_MISC_ENABLE_IP_PREF_DISABLE (1ULL << 39) -/* Intel Model 6 */ -#define MSR_P6_EVNTSEL0 0x00000186 -#define MSR_P6_EVNTSEL1 0x00000187 - /* P4/Xeon+ specific */ #define MSR_IA32_MCG_EAX 0x00000180 #define MSR_IA32_MCG_EBX 0x00000181 diff --git a/arch/x86/include/asm/msr.h b/arch/x86/include/asm/msr.h index 2260376..48ad9d2 100644 --- a/arch/x86/include/asm/msr.h +++ b/arch/x86/include/asm/msr.h @@ -3,13 +3,10 @@ #include <asm/msr-index.h> -#ifndef __ASSEMBLY__ -# include <linux/types.h> -#endif - #ifdef __KERNEL__ #ifndef __ASSEMBLY__ +#include <linux/types.h> #include <asm/asm.h> #include <asm/errno.h> #include <asm/cpumask.h> @@ -264,6 +261,4 @@ static inline int wrmsr_safe_on_cpu(unsigned int cpu, u32 msr_no, u32 l, u32 h) #endif /* CONFIG_SMP */ #endif /* __ASSEMBLY__ */ #endif /* __KERNEL__ */ - - #endif /* _ASM_X86_MSR_H */ diff --git a/arch/x86/include/asm/nmi.h b/arch/x86/include/asm/nmi.h index c972644..c86e5ed4 100644 --- a/arch/x86/include/asm/nmi.h +++ b/arch/x86/include/asm/nmi.h @@ -72,7 +72,6 @@ void lapic_watchdog_stop(void); int lapic_watchdog_init(unsigned nmi_hz); int lapic_wd_event(unsigned nmi_hz); unsigned lapic_adjust_nmi_hz(unsigned hz); -int lapic_watchdog_ok(void); void disable_lapic_nmi_watchdog(void); void enable_lapic_nmi_watchdog(void); void stop_nmi(void); diff --git a/arch/x86/include/asm/page_64_types.h b/arch/x86/include/asm/page_64_types.h index 8d382d3..7639dbf 100644 --- a/arch/x86/include/asm/page_64_types.h +++ b/arch/x86/include/asm/page_64_types.h @@ -41,7 +41,7 @@ /* See Documentation/x86/x86_64/mm.txt for a description of the memory map. */ #define __PHYSICAL_MASK_SHIFT 46 -#define __VIRTUAL_MASK_SHIFT 48 +#define __VIRTUAL_MASK_SHIFT 47 /* * Kernel image size is limited to 512 MB (see level2_kernel_pgt in diff --git a/arch/x86/include/asm/pci.h b/arch/x86/include/asm/pci.h index b51a1e8..1ff685c 100644 --- a/arch/x86/include/asm/pci.h +++ b/arch/x86/include/asm/pci.h @@ -91,7 +91,7 @@ extern void pci_iommu_alloc(void); #define PCI_DMA_BUS_IS_PHYS (dma_ops->is_phys) -#if defined(CONFIG_X86_64) || defined(CONFIG_DMA_API_DEBUG) +#if defined(CONFIG_X86_64) || defined(CONFIG_DMAR) || defined(CONFIG_DMA_API_DEBUG) #define DECLARE_PCI_UNMAP_ADDR(ADDR_NAME) \ dma_addr_t ADDR_NAME; @@ -130,6 +130,7 @@ extern void pci_iommu_alloc(void); /* generic pci stuff */ #include <asm-generic/pci.h> +#define PCIBIOS_MAX_MEM_32 0xffffffff #ifdef CONFIG_NUMA /* Returns the node based on pci bus */ diff --git a/arch/x86/include/asm/pci_x86.h b/arch/x86/include/asm/pci_x86.h index e60fd3e..b399988 100644 --- a/arch/x86/include/asm/pci_x86.h +++ b/arch/x86/include/asm/pci_x86.h @@ -121,6 +121,9 @@ extern int __init pcibios_init(void); extern int __init pci_mmcfg_arch_init(void); extern void __init pci_mmcfg_arch_free(void); +extern struct acpi_mcfg_allocation *pci_mmcfg_config; +extern int pci_mmcfg_config_num; + /* * AMD Fam10h CPUs are buggy, and cannot access MMIO config space * on their northbrige except through the * %eax register. As such, you MUST diff --git a/arch/x86/include/asm/percpu.h b/arch/x86/include/asm/percpu.h index 02ecb30..103f1dd 100644 --- a/arch/x86/include/asm/percpu.h +++ b/arch/x86/include/asm/percpu.h @@ -42,6 +42,7 @@ #else /* ...!ASSEMBLY */ +#include <linux/kernel.h> #include <linux/stringify.h> #ifdef CONFIG_SMP @@ -155,6 +156,15 @@ do { \ /* We can use this directly for local CPU (faster). */ DECLARE_PER_CPU(unsigned long, this_cpu_off); +#ifdef CONFIG_NEED_MULTIPLE_NODES +void *pcpu_lpage_remapped(void *kaddr); +#else +static inline void *pcpu_lpage_remapped(void *kaddr) +{ + return NULL; +} +#endif + #endif /* !__ASSEMBLY__ */ #ifdef CONFIG_SMP diff --git a/arch/x86/include/asm/perf_counter.h b/arch/x86/include/asm/perf_counter.h index 876ed97..fa64e40 100644 --- a/arch/x86/include/asm/perf_counter.h +++ b/arch/x86/include/asm/perf_counter.h @@ -84,14 +84,12 @@ union cpuid10_edx { #define MSR_ARCH_PERFMON_FIXED_CTR2 0x30b #define X86_PMC_IDX_FIXED_BUS_CYCLES (X86_PMC_IDX_FIXED + 2) -extern void set_perf_counter_pending(void); - -#define clear_perf_counter_pending() do { } while (0) -#define test_perf_counter_pending() (0) - #ifdef CONFIG_PERF_COUNTERS extern void init_hw_perf_counters(void); extern void perf_counters_lapic_init(void); + +#define PERF_COUNTER_INDEX_OFFSET 0 + #else static inline void init_hw_perf_counters(void) { } static inline void perf_counters_lapic_init(void) { } diff --git a/arch/x86/include/asm/pgalloc.h b/arch/x86/include/asm/pgalloc.h index dd14c54..0e8c2a0 100644 --- a/arch/x86/include/asm/pgalloc.h +++ b/arch/x86/include/asm/pgalloc.h @@ -46,7 +46,13 @@ static inline void pte_free(struct mm_struct *mm, struct page *pte) __free_page(pte); } -extern void __pte_free_tlb(struct mmu_gather *tlb, struct page *pte); +extern void ___pte_free_tlb(struct mmu_gather *tlb, struct page *pte); + +static inline void __pte_free_tlb(struct mmu_gather *tlb, struct page *pte, + unsigned long address) +{ + ___pte_free_tlb(tlb, pte); +} static inline void pmd_populate_kernel(struct mm_struct *mm, pmd_t *pmd, pte_t *pte) @@ -78,7 +84,13 @@ static inline void pmd_free(struct mm_struct *mm, pmd_t *pmd) free_page((unsigned long)pmd); } -extern void __pmd_free_tlb(struct mmu_gather *tlb, pmd_t *pmd); +extern void ___pmd_free_tlb(struct mmu_gather *tlb, pmd_t *pmd); + +static inline void __pmd_free_tlb(struct mmu_gather *tlb, pmd_t *pmd, + unsigned long adddress) +{ + ___pmd_free_tlb(tlb, pmd); +} #ifdef CONFIG_X86_PAE extern void pud_populate(struct mm_struct *mm, pud_t *pudp, pmd_t *pmd); @@ -108,7 +120,14 @@ static inline void pud_free(struct mm_struct *mm, pud_t *pud) free_page((unsigned long)pud); } -extern void __pud_free_tlb(struct mmu_gather *tlb, pud_t *pud); +extern void ___pud_free_tlb(struct mmu_gather *tlb, pud_t *pud); + +static inline void __pud_free_tlb(struct mmu_gather *tlb, pud_t *pud, + unsigned long address) +{ + ___pud_free_tlb(tlb, pud); +} + #endif /* PAGETABLE_LEVELS > 3 */ #endif /* PAGETABLE_LEVELS > 2 */ diff --git a/arch/x86/include/asm/pgtable.h b/arch/x86/include/asm/pgtable.h index 3cc06e3..1674807 100644 --- a/arch/x86/include/asm/pgtable.h +++ b/arch/x86/include/asm/pgtable.h @@ -2,6 +2,7 @@ #define _ASM_X86_PGTABLE_H #include <asm/page.h> +#include <asm/e820.h> #include <asm/pgtable_types.h> @@ -269,10 +270,17 @@ static inline pgprot_t pgprot_modify(pgprot_t oldprot, pgprot_t newprot) #define canon_pgprot(p) __pgprot(massage_pgprot(p)) -static inline int is_new_memtype_allowed(unsigned long flags, - unsigned long new_flags) +static inline int is_new_memtype_allowed(u64 paddr, unsigned long size, + unsigned long flags, + unsigned long new_flags) { /* + * PAT type is always WB for ISA. So no need to check. + */ + if (is_ISA_range(paddr, paddr + size - 1)) + return 1; + + /* * Certain new memtypes are not allowed with certain * requested memtype: * - request is uncached, return cannot be write-back diff --git a/arch/x86/include/asm/pgtable_32.h b/arch/x86/include/asm/pgtable_32.h index 31bd120..01fd946 100644 --- a/arch/x86/include/asm/pgtable_32.h +++ b/arch/x86/include/asm/pgtable_32.h @@ -49,13 +49,17 @@ extern void set_pmd_pfn(unsigned long, unsigned long, pgprot_t); #endif #if defined(CONFIG_HIGHPTE) +#define __KM_PTE \ + (in_nmi() ? KM_NMI_PTE : \ + in_irq() ? KM_IRQ_PTE : \ + KM_PTE0) #define pte_offset_map(dir, address) \ - ((pte_t *)kmap_atomic_pte(pmd_page(*(dir)), KM_PTE0) + \ + ((pte_t *)kmap_atomic_pte(pmd_page(*(dir)), __KM_PTE) + \ pte_index((address))) #define pte_offset_map_nested(dir, address) \ ((pte_t *)kmap_atomic_pte(pmd_page(*(dir)), KM_PTE1) + \ pte_index((address))) -#define pte_unmap(pte) kunmap_atomic((pte), KM_PTE0) +#define pte_unmap(pte) kunmap_atomic((pte), __KM_PTE) #define pte_unmap_nested(pte) kunmap_atomic((pte), KM_PTE1) #else #define pte_offset_map(dir, address) \ diff --git a/arch/x86/include/asm/pgtable_64.h b/arch/x86/include/asm/pgtable_64.h index abde308..c57a301 100644 --- a/arch/x86/include/asm/pgtable_64.h +++ b/arch/x86/include/asm/pgtable_64.h @@ -165,10 +165,7 @@ extern void cleanup_highmap(void); /* fs/proc/kcore.c */ #define kc_vaddr_to_offset(v) ((v) & __VIRTUAL_MASK) -#define kc_offset_to_vaddr(o) \ - (((o) & (1UL << (__VIRTUAL_MASK_SHIFT - 1))) \ - ? ((o) | ~__VIRTUAL_MASK) \ - : (o)) +#define kc_offset_to_vaddr(o) ((o) | ~__VIRTUAL_MASK) #define __HAVE_ARCH_PTE_SAME #endif /* !__ASSEMBLY__ */ diff --git a/arch/x86/include/asm/proto.h b/arch/x86/include/asm/proto.h index 49fb3ec..621f56d 100644 --- a/arch/x86/include/asm/proto.h +++ b/arch/x86/include/asm/proto.h @@ -22,7 +22,14 @@ extern int reboot_force; long do_arch_prctl(struct task_struct *task, int code, unsigned long addr); -#define round_up(x, y) (((x) + (y) - 1) & ~((y) - 1)) -#define round_down(x, y) ((x) & ~((y) - 1)) +/* + * This looks more complex than it should be. But we need to + * get the type for the ~ right in round_down (it needs to be + * as wide as the result!), and we want to evaluate the macro + * arguments just once each. + */ +#define __round_mask(x,y) ((__typeof__(x))((y)-1)) +#define round_up(x,y) ((((x)-1) | __round_mask(x,y))+1) +#define round_down(x,y) ((x) & ~__round_mask(x,y)) #endif /* _ASM_X86_PROTO_H */ diff --git a/arch/x86/include/asm/spinlock.h b/arch/x86/include/asm/spinlock.h index b7e5db8..4e77853 100644 --- a/arch/x86/include/asm/spinlock.h +++ b/arch/x86/include/asm/spinlock.h @@ -302,4 +302,8 @@ static inline void __raw_write_unlock(raw_rwlock_t *rw) #define _raw_read_relax(lock) cpu_relax() #define _raw_write_relax(lock) cpu_relax() +/* The {read|write|spin}_lock() on x86 are full memory barriers. */ +static inline void smp_mb__after_lock(void) { } +#define ARCH_HAS_SMP_MB_AFTER_LOCK + #endif /* _ASM_X86_SPINLOCK_H */ diff --git a/arch/x86/include/asm/stacktrace.h b/arch/x86/include/asm/stacktrace.h index f517944..cf86a5e 100644 --- a/arch/x86/include/asm/stacktrace.h +++ b/arch/x86/include/asm/stacktrace.h @@ -3,6 +3,8 @@ extern int kstack_depth_to_print; +int x86_is_stack_id(int id, char *name); + /* Generic stack tracer with callbacks */ struct stacktrace_ops { diff --git a/arch/x86/include/asm/therm_throt.h b/arch/x86/include/asm/therm_throt.h deleted file mode 100644 index c62349e..0000000 --- a/arch/x86/include/asm/therm_throt.h +++ /dev/null @@ -1,9 +0,0 @@ -#ifndef _ASM_X86_THERM_THROT_H -#define _ASM_X86_THERM_THROT_H - -#include <asm/atomic.h> - -extern atomic_t therm_throt_en; -int therm_throt_process(int curr); - -#endif /* _ASM_X86_THERM_THROT_H */ diff --git a/arch/x86/include/asm/thread_info.h b/arch/x86/include/asm/thread_info.h index b078352..6f7786a 100644 --- a/arch/x86/include/asm/thread_info.h +++ b/arch/x86/include/asm/thread_info.h @@ -49,7 +49,7 @@ struct thread_info { .exec_domain = &default_exec_domain, \ .flags = 0, \ .cpu = 0, \ - .preempt_count = 1, \ + .preempt_count = INIT_PREEMPT_COUNT, \ .addr_limit = KERNEL_DS, \ .restart_block = { \ .fn = do_no_restart_syscall, \ @@ -95,7 +95,7 @@ struct thread_info { #define TIF_DEBUGCTLMSR 25 /* uses thread_struct.debugctlmsr */ #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_FTRACE 28 /* for ftrace syscall instrumentation */ +#define TIF_SYSCALL_TRACEPOINT 28 /* syscall tracepoint instrumentation */ #define _TIF_SYSCALL_TRACE (1 << TIF_SYSCALL_TRACE) #define _TIF_NOTIFY_RESUME (1 << TIF_NOTIFY_RESUME) @@ -118,17 +118,17 @@ struct thread_info { #define _TIF_DEBUGCTLMSR (1 << TIF_DEBUGCTLMSR) #define _TIF_DS_AREA_MSR (1 << TIF_DS_AREA_MSR) #define _TIF_LAZY_MMU_UPDATES (1 << TIF_LAZY_MMU_UPDATES) -#define _TIF_SYSCALL_FTRACE (1 << TIF_SYSCALL_FTRACE) +#define _TIF_SYSCALL_TRACEPOINT (1 << TIF_SYSCALL_TRACEPOINT) /* work to do in syscall_trace_enter() */ #define _TIF_WORK_SYSCALL_ENTRY \ - (_TIF_SYSCALL_TRACE | _TIF_SYSCALL_EMU | _TIF_SYSCALL_FTRACE | \ - _TIF_SYSCALL_AUDIT | _TIF_SECCOMP | _TIF_SINGLESTEP) + (_TIF_SYSCALL_TRACE | _TIF_SYSCALL_EMU | _TIF_SYSCALL_AUDIT | \ + _TIF_SECCOMP | _TIF_SINGLESTEP | _TIF_SYSCALL_TRACEPOINT) /* work to do in syscall_trace_leave() */ #define _TIF_WORK_SYSCALL_EXIT \ (_TIF_SYSCALL_TRACE | _TIF_SYSCALL_AUDIT | _TIF_SINGLESTEP | \ - _TIF_SYSCALL_FTRACE) + _TIF_SYSCALL_TRACEPOINT) /* work to do on interrupt/exception return */ #define _TIF_WORK_MASK \ @@ -137,7 +137,8 @@ struct thread_info { _TIF_SINGLESTEP|_TIF_SECCOMP|_TIF_SYSCALL_EMU)) /* work to do on any return to user space */ -#define _TIF_ALLWORK_MASK ((0x0000FFFF & ~_TIF_SECCOMP) | _TIF_SYSCALL_FTRACE) +#define _TIF_ALLWORK_MASK \ + ((0x0000FFFF & ~_TIF_SECCOMP) | _TIF_SYSCALL_TRACEPOINT) /* Only used for 64 bit */ #define _TIF_DO_NOTIFY_MASK \ diff --git a/arch/x86/include/asm/timer.h b/arch/x86/include/asm/timer.h index bd37ed4..20ca9c4 100644 --- a/arch/x86/include/asm/timer.h +++ b/arch/x86/include/asm/timer.h @@ -45,12 +45,16 @@ extern int no_timer_check; */ DECLARE_PER_CPU(unsigned long, cyc2ns); +DECLARE_PER_CPU(unsigned long long, cyc2ns_offset); #define CYC2NS_SCALE_FACTOR 10 /* 2^10, carefully chosen */ static inline unsigned long long __cycles_2_ns(unsigned long long cyc) { - return cyc * per_cpu(cyc2ns, smp_processor_id()) >> CYC2NS_SCALE_FACTOR; + int cpu = smp_processor_id(); + unsigned long long ns = per_cpu(cyc2ns_offset, cpu); + ns += cyc * per_cpu(cyc2ns, cpu) >> CYC2NS_SCALE_FACTOR; + return ns; } static inline unsigned long long cycles_2_ns(unsigned long long cyc) diff --git a/arch/x86/include/asm/uaccess.h b/arch/x86/include/asm/uaccess.h index b685ece..d2c6c93 100644 --- a/arch/x86/include/asm/uaccess.h +++ b/arch/x86/include/asm/uaccess.h @@ -25,7 +25,7 @@ #define MAKE_MM_SEG(s) ((mm_segment_t) { (s) }) #define KERNEL_DS MAKE_MM_SEG(-1UL) -#define USER_DS MAKE_MM_SEG(PAGE_OFFSET) +#define USER_DS MAKE_MM_SEG(TASK_SIZE_MAX) #define get_ds() (KERNEL_DS) #define get_fs() (current_thread_info()->addr_limit) @@ -212,9 +212,9 @@ extern int __get_user_bad(void); : "A" ((typeof(*(ptr)))(x)), "c" (ptr) : "ebx") #else #define __put_user_asm_u64(x, ptr, retval, errret) \ - __put_user_asm(x, ptr, retval, "q", "", "Zr", errret) + __put_user_asm(x, ptr, retval, "q", "", "er", errret) #define __put_user_asm_ex_u64(x, addr) \ - __put_user_asm_ex(x, addr, "q", "", "Zr") + __put_user_asm_ex(x, addr, "q", "", "er") #define __put_user_x8(x, ptr, __ret_pu) __put_user_x(8, x, ptr, __ret_pu) #endif diff --git a/arch/x86/include/asm/uaccess_64.h b/arch/x86/include/asm/uaccess_64.h index 8cc6873..db24b21 100644 --- a/arch/x86/include/asm/uaccess_64.h +++ b/arch/x86/include/asm/uaccess_64.h @@ -88,11 +88,11 @@ int __copy_to_user(void __user *dst, const void *src, unsigned size) ret, "l", "k", "ir", 4); return ret; case 8:__put_user_asm(*(u64 *)src, (u64 __user *)dst, - ret, "q", "", "ir", 8); + ret, "q", "", "er", 8); return ret; case 10: __put_user_asm(*(u64 *)src, (u64 __user *)dst, - ret, "q", "", "ir", 10); + ret, "q", "", "er", 10); if (unlikely(ret)) return ret; asm("":::"memory"); @@ -101,12 +101,12 @@ int __copy_to_user(void __user *dst, const void *src, unsigned size) return ret; case 16: __put_user_asm(*(u64 *)src, (u64 __user *)dst, - ret, "q", "", "ir", 16); + ret, "q", "", "er", 16); if (unlikely(ret)) return ret; asm("":::"memory"); __put_user_asm(1[(u64 *)src], 1 + (u64 __user *)dst, - ret, "q", "", "ir", 8); + ret, "q", "", "er", 8); return ret; default: return copy_user_generic((__force void *)dst, src, size); @@ -157,7 +157,7 @@ int __copy_in_user(void __user *dst, const void __user *src, unsigned size) ret, "q", "", "=r", 8); if (likely(!ret)) __put_user_asm(tmp, (u64 __user *)dst, - ret, "q", "", "ir", 8); + ret, "q", "", "er", 8); return ret; } default: diff --git a/arch/x86/include/asm/unistd_32.h b/arch/x86/include/asm/unistd_32.h index 732a3070..8deaada 100644 --- a/arch/x86/include/asm/unistd_32.h +++ b/arch/x86/include/asm/unistd_32.h @@ -345,6 +345,8 @@ #ifdef __KERNEL__ +#define NR_syscalls 337 + #define __ARCH_WANT_IPC_PARSE_VERSION #define __ARCH_WANT_OLD_READDIR #define __ARCH_WANT_OLD_STAT diff --git a/arch/x86/include/asm/unistd_64.h b/arch/x86/include/asm/unistd_64.h index 900e161..b9f3c60 100644 --- a/arch/x86/include/asm/unistd_64.h +++ b/arch/x86/include/asm/unistd_64.h @@ -688,6 +688,12 @@ __SYSCALL(__NR_perf_counter_open, sys_perf_counter_open) #endif /* __NO_STUBS */ #ifdef __KERNEL__ + +#ifndef COMPILE_OFFSETS +#include <asm/asm-offsets.h> +#define NR_syscalls (__NR_syscall_max + 1) +#endif + /* * "Conditional" syscalls * diff --git a/arch/x86/include/asm/uv/uv_bau.h b/arch/x86/include/asm/uv/uv_bau.h index bddd44f..80e2984 100644 --- a/arch/x86/include/asm/uv/uv_bau.h +++ b/arch/x86/include/asm/uv/uv_bau.h @@ -133,7 +133,7 @@ struct bau_msg_payload { * see table 4.2.3.0.1 in broacast_assist spec. */ struct bau_msg_header { - unsigned int dest_subnodeid:6; /* must be zero */ + unsigned int dest_subnodeid:6; /* must be 0x10, for the LB */ /* bits 5:0 */ unsigned int base_dest_nodeid:15; /* nasid>>1 (pnode) of */ /* bits 20:6 */ /* first bit in node_map */ diff --git a/arch/x86/include/asm/uv/uv_hub.h b/arch/x86/include/asm/uv/uv_hub.h index 341070f..77a6850 100644 --- a/arch/x86/include/asm/uv/uv_hub.h +++ b/arch/x86/include/asm/uv/uv_hub.h @@ -175,7 +175,7 @@ DECLARE_PER_CPU(struct uv_hub_info_s, __uv_hub_info); #define UV_GLOBAL_MMR32_PNODE_BITS(p) ((p) << (UV_GLOBAL_MMR32_PNODE_SHIFT)) #define UV_GLOBAL_MMR64_PNODE_BITS(p) \ - ((unsigned long)(UV_PNODE_TO_GNODE(p)) << UV_GLOBAL_MMR64_PNODE_SHIFT) + (((unsigned long)(p)) << UV_GLOBAL_MMR64_PNODE_SHIFT) #define UV_APIC_PNODE_SHIFT 6 @@ -327,6 +327,7 @@ struct uv_blade_info { unsigned short nr_possible_cpus; unsigned short nr_online_cpus; unsigned short pnode; + short memory_nid; }; extern struct uv_blade_info *uv_blade_info; extern short *uv_node_to_blade; @@ -363,6 +364,12 @@ static inline int uv_blade_to_pnode(int bid) return uv_blade_info[bid].pnode; } +/* Nid of memory node on blade. -1 if no blade-local memory */ +static inline int uv_blade_to_memory_nid(int bid) +{ + return uv_blade_info[bid].memory_nid; +} + /* Determine the number of possible cpus on a blade */ static inline int uv_blade_nr_possible_cpus(int bid) { diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index b67efd1..bf04201 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -24,6 +24,10 @@ CFLAGS_vsyscall_64.o := $(PROFILING) -g0 $(nostackp) CFLAGS_hpet.o := $(nostackp) CFLAGS_tsc.o := $(nostackp) CFLAGS_paravirt.o := $(nostackp) +GCOV_PROFILE_vsyscall_64.o := n +GCOV_PROFILE_hpet.o := n +GCOV_PROFILE_tsc.o := n +GCOV_PROFILE_paravirt.o := n obj-y := process_$(BITS).o signal.o entry_$(BITS).o obj-y += traps.o irq.o irq_$(BITS).o dumpstack_$(BITS).o diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c index 6310861..6b8ca3a 100644 --- a/arch/x86/kernel/acpi/boot.c +++ b/arch/x86/kernel/acpi/boot.c @@ -44,11 +44,7 @@ static int __initdata acpi_force = 0; u32 acpi_rsdt_forced; -#ifdef CONFIG_ACPI -int acpi_disabled = 0; -#else -int acpi_disabled = 1; -#endif +int acpi_disabled; EXPORT_SYMBOL(acpi_disabled); #ifdef CONFIG_X86_64 @@ -122,72 +118,6 @@ void __init __acpi_unmap_table(char *map, unsigned long size) early_iounmap(map, size); } -#ifdef CONFIG_PCI_MMCONFIG - -static int acpi_mcfg_64bit_base_addr __initdata = FALSE; - -/* The physical address of the MMCONFIG aperture. Set from ACPI tables. */ -struct acpi_mcfg_allocation *pci_mmcfg_config; -int pci_mmcfg_config_num; - -static int __init acpi_mcfg_oem_check(struct acpi_table_mcfg *mcfg) -{ - if (!strcmp(mcfg->header.oem_id, "SGI")) - acpi_mcfg_64bit_base_addr = TRUE; - - return 0; -} - -int __init acpi_parse_mcfg(struct acpi_table_header *header) -{ - struct acpi_table_mcfg *mcfg; - unsigned long i; - int config_size; - - if (!header) - return -EINVAL; - - mcfg = (struct acpi_table_mcfg *)header; - - /* how many config structures do we have */ - pci_mmcfg_config_num = 0; - i = header->length - sizeof(struct acpi_table_mcfg); - while (i >= sizeof(struct acpi_mcfg_allocation)) { - ++pci_mmcfg_config_num; - i -= sizeof(struct acpi_mcfg_allocation); - }; - if (pci_mmcfg_config_num == 0) { - printk(KERN_ERR PREFIX "MMCONFIG has no entries\n"); - return -ENODEV; - } - - config_size = pci_mmcfg_config_num * sizeof(*pci_mmcfg_config); - pci_mmcfg_config = kmalloc(config_size, GFP_KERNEL); - if (!pci_mmcfg_config) { - printk(KERN_WARNING PREFIX - "No memory for MCFG config tables\n"); - return -ENOMEM; - } - - memcpy(pci_mmcfg_config, &mcfg[1], config_size); - - acpi_mcfg_oem_check(mcfg); - - for (i = 0; i < pci_mmcfg_config_num; ++i) { - if ((pci_mmcfg_config[i].address > 0xFFFFFFFF) && - !acpi_mcfg_64bit_base_addr) { - printk(KERN_ERR PREFIX - "MMCONFIG not in low 4GB of memory\n"); - kfree(pci_mmcfg_config); - pci_mmcfg_config_num = 0; - return -ENODEV; - } - } - - return 0; -} -#endif /* CONFIG_PCI_MMCONFIG */ - #ifdef CONFIG_X86_LOCAL_APIC static int __init acpi_parse_madt(struct acpi_table_header *table) { @@ -1519,14 +1449,6 @@ static struct dmi_system_id __initdata acpi_dmi_table[] = { }, { .callback = force_acpi_ht, - .ident = "ASUS P4B266", - .matches = { - DMI_MATCH(DMI_BOARD_VENDOR, "ASUSTeK Computer INC."), - DMI_MATCH(DMI_BOARD_NAME, "P4B266"), - }, - }, - { - .callback = force_acpi_ht, .ident = "ASUS P2B-DS", .matches = { DMI_MATCH(DMI_BOARD_VENDOR, "ASUSTeK Computer INC."), diff --git a/arch/x86/kernel/acpi/cstate.c b/arch/x86/kernel/acpi/cstate.c index bbbe4bb..8c44c23 100644 --- a/arch/x86/kernel/acpi/cstate.c +++ b/arch/x86/kernel/acpi/cstate.c @@ -34,12 +34,22 @@ void acpi_processor_power_init_bm_check(struct acpi_processor_flags *flags, flags->bm_check = 1; else if (c->x86_vendor == X86_VENDOR_INTEL) { /* - * Today all CPUs that support C3 share cache. - * TBD: This needs to look at cache shared map, once - * multi-core detection patch makes to the base. + * Today all MP CPUs that support C3 share cache. + * And caches should not be flushed by software while + * entering C3 type state. */ flags->bm_check = 1; } + + /* + * On all recent Intel platforms, ARB_DISABLE is a nop. + * So, set bm_control to zero to indicate that ARB_DISABLE + * is not required while entering C3 type state on + * P4, Core and beyond CPUs + */ + if (c->x86_vendor == X86_VENDOR_INTEL && + (c->x86 > 0x6 || (c->x86 == 6 && c->x86_model >= 14))) + flags->bm_control = 0; } EXPORT_SYMBOL(acpi_processor_power_init_bm_check); diff --git a/arch/x86/kernel/acpi/processor.c b/arch/x86/kernel/acpi/processor.c index 7c074ee..d296f4a 100644 --- a/arch/x86/kernel/acpi/processor.c +++ b/arch/x86/kernel/acpi/processor.c @@ -72,6 +72,7 @@ static void init_intel_pdc(struct acpi_processor *pr, struct cpuinfo_x86 *c) return; } + /* Initialize _PDC data based on the CPU vendor */ void arch_acpi_processor_init_pdc(struct acpi_processor *pr) { @@ -85,3 +86,15 @@ void arch_acpi_processor_init_pdc(struct acpi_processor *pr) } EXPORT_SYMBOL(arch_acpi_processor_init_pdc); + +void arch_acpi_processor_cleanup_pdc(struct acpi_processor *pr) +{ + if (pr->pdc) { + kfree(pr->pdc->pointer->buffer.pointer); + kfree(pr->pdc->pointer); + kfree(pr->pdc); + pr->pdc = NULL; + } +} + +EXPORT_SYMBOL(arch_acpi_processor_cleanup_pdc); diff --git a/arch/x86/kernel/acpi/realmode/Makefile b/arch/x86/kernel/acpi/realmode/Makefile index 167bc16..6a564ac 100644 --- a/arch/x86/kernel/acpi/realmode/Makefile +++ b/arch/x86/kernel/acpi/realmode/Makefile @@ -42,6 +42,7 @@ KBUILD_CFLAGS := $(LINUXINCLUDE) -g -Os -D_SETUP -D_WAKEUP -D__KERNEL__ \ $(call cc-option, -mpreferred-stack-boundary=2) KBUILD_CFLAGS += $(call cc-option, -m32) KBUILD_AFLAGS := $(KBUILD_CFLAGS) -D__ASSEMBLY__ +GCOV_PROFILE := n WAKEUP_OBJS = $(addprefix $(obj)/,$(wakeup-y)) diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c index 1c60554..6c99f50 100644 --- a/arch/x86/kernel/amd_iommu.c +++ b/arch/x86/kernel/amd_iommu.c @@ -434,6 +434,16 @@ static void iommu_flush_tlb(struct amd_iommu *iommu, u16 domid) iommu_queue_inv_iommu_pages(iommu, address, domid, 0, 1); } +/* Flush the whole IO/TLB for a given protection domain - including PDE */ +static void iommu_flush_tlb_pde(struct amd_iommu *iommu, u16 domid) +{ + u64 address = CMD_INV_IOMMU_ALL_PAGES_ADDRESS; + + INC_STATS_COUNTER(domain_flush_single); + + iommu_queue_inv_iommu_pages(iommu, address, domid, 1, 1); +} + /* * This function is used to flush the IO/TLB for a given protection domain * on every IOMMU in the system @@ -1078,7 +1088,13 @@ static void attach_device(struct amd_iommu *iommu, amd_iommu_pd_table[devid] = domain; write_unlock_irqrestore(&amd_iommu_devtable_lock, flags); + /* + * We might boot into a crash-kernel here. The crashed kernel + * left the caches in the IOMMU dirty. So we have to flush + * here to evict all dirty stuff. + */ iommu_queue_inv_dev_entry(iommu, devid); + iommu_flush_tlb_pde(iommu, domain->id); } /* @@ -1176,7 +1192,7 @@ out: return 0; } -struct notifier_block device_nb = { +static struct notifier_block device_nb = { .notifier_call = device_change_notifier, }; @@ -1747,7 +1763,7 @@ static void *alloc_coherent(struct device *dev, size_t size, flag |= __GFP_ZERO; virt_addr = (void *)__get_free_pages(flag, get_order(size)); if (!virt_addr) - return 0; + return NULL; paddr = virt_to_phys(virt_addr); diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c index 238989e..c1b17e9 100644 --- a/arch/x86/kernel/amd_iommu_init.c +++ b/arch/x86/kernel/amd_iommu_init.c @@ -260,6 +260,14 @@ static void iommu_enable(struct amd_iommu *iommu) static void iommu_disable(struct amd_iommu *iommu) { + /* Disable command buffer */ + iommu_feature_disable(iommu, CONTROL_CMDBUF_EN); + + /* Disable event logging and event interrupts */ + iommu_feature_disable(iommu, CONTROL_EVT_INT_EN); + iommu_feature_disable(iommu, CONTROL_EVT_LOG_EN); + + /* Disable IOMMU hardware itself */ iommu_feature_disable(iommu, CONTROL_IOMMU_EN); } @@ -464,6 +472,8 @@ static u8 * __init alloc_event_buffer(struct amd_iommu *iommu) if (iommu->evt_buf == NULL) return NULL; + iommu->evt_buf_size = EVT_BUFFER_SIZE; + return iommu->evt_buf; } @@ -478,6 +488,10 @@ static void iommu_enable_event_buffer(struct amd_iommu *iommu) memcpy_toio(iommu->mmio_base + MMIO_EVT_BUF_OFFSET, &entry, sizeof(entry)); + /* set head and tail to zero manually */ + writel(0x00, iommu->mmio_base + MMIO_EVT_HEAD_OFFSET); + writel(0x00, iommu->mmio_base + MMIO_EVT_TAIL_OFFSET); + iommu_feature_enable(iommu, CONTROL_EVT_LOG_EN); } @@ -679,6 +693,7 @@ static void __init init_iommu_from_acpi(struct amd_iommu *iommu, devid = e->devid; devid_to = e->ext >> 8; + set_dev_entry_from_acpi(iommu, devid , e->flags, 0); set_dev_entry_from_acpi(iommu, devid_to, e->flags, 0); amd_iommu_alias_table[devid] = devid_to; break; @@ -737,11 +752,13 @@ static void __init init_iommu_from_acpi(struct amd_iommu *iommu, devid = e->devid; for (dev_i = devid_start; dev_i <= devid; ++dev_i) { - if (alias) + if (alias) { amd_iommu_alias_table[dev_i] = devid_to; - set_dev_entry_from_acpi(iommu, - amd_iommu_alias_table[dev_i], - flags, ext_flags); + set_dev_entry_from_acpi(iommu, + devid_to, flags, ext_flags); + } + set_dev_entry_from_acpi(iommu, dev_i, + flags, ext_flags); } break; default: @@ -1042,6 +1059,7 @@ static void enable_iommus(void) struct amd_iommu *iommu; for_each_iommu(iommu) { + iommu_disable(iommu); iommu_set_device_table(iommu); iommu_enable_command_buffer(iommu); iommu_enable_event_buffer(iommu); @@ -1066,12 +1084,6 @@ static void disable_iommus(void) static int amd_iommu_resume(struct sys_device *dev) { - /* - * Disable IOMMUs before reprogramming the hardware registers. - * IOMMU is still enabled from the resume kernel. - */ - disable_iommus(); - /* re-load the hardware */ enable_iommus(); @@ -1079,8 +1091,8 @@ static int amd_iommu_resume(struct sys_device *dev) * we have to flush after the IOMMUs are enabled because a * disabled IOMMU will never execute the commands we send */ - amd_iommu_flush_all_domains(); amd_iommu_flush_all_devices(); + amd_iommu_flush_all_domains(); return 0; } @@ -1273,6 +1285,11 @@ free: goto out; } +void amd_iommu_shutdown(void) +{ + disable_iommus(); +} + /**************************************************************************** * * Early detect code. This code runs at IOMMU detection time in the DMA diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c index 8c7c042..0a1c283 100644 --- a/arch/x86/kernel/apic/apic.c +++ b/arch/x86/kernel/apic/apic.c @@ -140,7 +140,6 @@ int x2apic_mode; #ifdef CONFIG_X86_X2APIC /* x2apic enabled before OS handover */ static int x2apic_preenabled; -static int disable_x2apic; static __init int setup_nox2apic(char *str) { if (x2apic_enabled()) { @@ -149,7 +148,6 @@ static __init int setup_nox2apic(char *str) return 0; } - disable_x2apic = 1; setup_clear_cpu_cap(X86_FEATURE_X2APIC); return 0; } diff --git a/arch/x86/kernel/apic/es7000_32.c b/arch/x86/kernel/apic/es7000_32.c index 69328ac..8952a58 100644 --- a/arch/x86/kernel/apic/es7000_32.c +++ b/arch/x86/kernel/apic/es7000_32.c @@ -652,7 +652,8 @@ static int es7000_mps_oem_check_cluster(struct mpc_table *mpc, char *oem, return ret && es7000_apic_is_cluster(); } -struct apic apic_es7000_cluster = { +/* We've been warned by a false positive warning.Use __refdata to keep calm. */ +struct apic __refdata apic_es7000_cluster = { .name = "es7000", .probe = probe_es7000, diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index ef8d929..d2ed6c5 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -462,7 +462,8 @@ static struct IO_APIC_route_entry ioapic_read_entry(int apic, int pin) static void __ioapic_write_entry(int apic, int pin, struct IO_APIC_route_entry e) { - union entry_union eu; + union entry_union eu = {{0, 0}}; + eu.entry = e; io_apic_write(apic, 0x11 + 2*pin, eu.w2); io_apic_write(apic, 0x10 + 2*pin, eu.w1); @@ -1413,6 +1414,9 @@ int setup_ioapic_entry(int apic_id, int irq, irte.vector = vector; irte.dest_id = IRTE_DEST(destination); + /* Set source-id of interrupt request */ + set_ioapic_sid(&irte, apic_id); + modify_irte(irq, &irte); ir_entry->index2 = (index >> 15) & 0x1; @@ -1712,25 +1716,19 @@ __apicdebuginit(void) print_IO_APIC(void) return; } -__apicdebuginit(void) print_APIC_bitfield(int base) +__apicdebuginit(void) print_APIC_field(int base) { - unsigned int v; - int i, j; + int i; if (apic_verbosity == APIC_QUIET) return; - printk(KERN_DEBUG "0123456789abcdef0123456789abcdef\n" KERN_DEBUG); - for (i = 0; i < 8; i++) { - v = apic_read(base + i*0x10); - for (j = 0; j < 32; j++) { - if (v & (1<<j)) - printk("1"); - else - printk("0"); - } - printk("\n"); - } + printk(KERN_DEBUG); + + for (i = 0; i < 8; i++) + printk(KERN_CONT "%08x", apic_read(base + i*0x10)); + + printk(KERN_CONT "\n"); } __apicdebuginit(void) print_local_APIC(void *dummy) @@ -1741,7 +1739,7 @@ __apicdebuginit(void) print_local_APIC(void *dummy) if (apic_verbosity == APIC_QUIET) return; - printk("\n" KERN_DEBUG "printing local APIC contents on CPU#%d/%d:\n", + printk(KERN_DEBUG "printing local APIC contents on CPU#%d/%d:\n", smp_processor_id(), hard_smp_processor_id()); v = apic_read(APIC_ID); printk(KERN_INFO "... APIC ID: %08x (%01x)\n", v, read_apic_id()); @@ -1782,11 +1780,11 @@ __apicdebuginit(void) print_local_APIC(void *dummy) printk(KERN_DEBUG "... APIC SPIV: %08x\n", v); printk(KERN_DEBUG "... APIC ISR field:\n"); - print_APIC_bitfield(APIC_ISR); + print_APIC_field(APIC_ISR); printk(KERN_DEBUG "... APIC TMR field:\n"); - print_APIC_bitfield(APIC_TMR); + print_APIC_field(APIC_TMR); printk(KERN_DEBUG "... APIC IRR field:\n"); - print_APIC_bitfield(APIC_IRR); + print_APIC_field(APIC_IRR); if (APIC_INTEGRATED(ver)) { /* !82489DX */ if (maxlvt > 3) /* Due to the Pentium erratum 3AP. */ @@ -2003,7 +2001,9 @@ void disable_IO_APIC(void) /* * Use virtual wire A mode when interrupt remapping is enabled. */ - disconnect_bsp_APIC(!intr_remapping_enabled && ioapic_i8259.pin != -1); + if (cpu_has_apic) + disconnect_bsp_APIC(!intr_remapping_enabled && + ioapic_i8259.pin != -1); } #ifdef CONFIG_X86_32 @@ -3287,6 +3287,9 @@ static int msi_compose_msg(struct pci_dev *pdev, unsigned int irq, struct msi_ms irte.vector = cfg->vector; irte.dest_id = IRTE_DEST(dest); + /* Set source-id of interrupt request */ + set_msi_sid(&irte, pdev); + modify_irte(irq, &irte); msg->address_hi = MSI_ADDR_BASE_HI; @@ -3567,7 +3570,7 @@ static int dmar_msi_set_affinity(unsigned int irq, const struct cpumask *mask) #endif /* CONFIG_SMP */ -struct irq_chip dmar_msi_type = { +static struct irq_chip dmar_msi_type = { .name = "DMAR_MSI", .unmask = dmar_msi_unmask, .mask = dmar_msi_mask, @@ -3790,6 +3793,9 @@ int arch_enable_uv_irq(char *irq_name, unsigned int irq, int cpu, int mmr_blade, mmr_pnode = uv_blade_to_pnode(mmr_blade); uv_write_global_mmr64(mmr_pnode, mmr_offset, mmr_value); + if (cfg->move_in_progress) + send_cleanup_vector(cfg); + return irq; } @@ -4178,28 +4184,20 @@ fake_ioapic_page: } } -static int __init ioapic_insert_resources(void) +void __init ioapic_insert_resources(void) { int i; struct resource *r = ioapic_resources; if (!r) { - if (nr_ioapics > 0) { + if (nr_ioapics > 0) printk(KERN_ERR "IO APIC resources couldn't be allocated.\n"); - return -1; - } - return 0; + return; } for (i = 0; i < nr_ioapics; i++) { insert_resource(&iomem_resource, r); r++; } - - return 0; } - -/* Insert the IO APIC resources after PCI initialization has occured to handle - * IO APICS that are mapped in on a BAR in PCI space. */ -late_initcall(ioapic_insert_resources); diff --git a/arch/x86/kernel/apic/ipi.c b/arch/x86/kernel/apic/ipi.c index dbf5445..6ef00ba 100644 --- a/arch/x86/kernel/apic/ipi.c +++ b/arch/x86/kernel/apic/ipi.c @@ -106,6 +106,9 @@ void default_send_IPI_mask_logical(const struct cpumask *cpumask, int vector) unsigned long mask = cpumask_bits(cpumask)[0]; unsigned long flags; + if (WARN_ONCE(!mask, "empty IPI mask")) + return; + local_irq_save(flags); WARN_ON(mask & ~cpumask_bits(cpu_online_mask)[0]); __default_send_IPI_dest_field(mask, vector, apic->dest_logical); diff --git a/arch/x86/kernel/apic/numaq_32.c b/arch/x86/kernel/apic/numaq_32.c index 533e59c..ca96e68 100644 --- a/arch/x86/kernel/apic/numaq_32.c +++ b/arch/x86/kernel/apic/numaq_32.c @@ -493,7 +493,8 @@ static void numaq_setup_portio_remap(void) (u_long) xquad_portio, (u_long) num_quads*XQUAD_PORTIO_QUAD); } -struct apic apic_numaq = { +/* Use __refdata to keep false positive warning calm. */ +struct apic __refdata apic_numaq = { .name = "NUMAQ", .probe = probe_numaq, diff --git a/arch/x86/kernel/apic/probe_32.c b/arch/x86/kernel/apic/probe_32.c index 440a8bc..0c0182c 100644 --- a/arch/x86/kernel/apic/probe_32.c +++ b/arch/x86/kernel/apic/probe_32.c @@ -20,23 +20,12 @@ #include <asm/apic.h> #include <asm/setup.h> -#include <linux/threads.h> -#include <linux/cpumask.h> -#include <asm/mpspec.h> -#include <asm/fixmap.h> -#include <asm/apicdef.h> -#include <linux/kernel.h> -#include <linux/string.h> #include <linux/smp.h> -#include <linux/init.h> #include <asm/ipi.h> -#include <linux/smp.h> -#include <linux/init.h> #include <linux/interrupt.h> #include <asm/acpi.h> #include <asm/e820.h> -#include <asm/setup.h> #ifdef CONFIG_HOTPLUG_CPU #define DEFAULT_SEND_IPI (1) diff --git a/arch/x86/kernel/apic/probe_64.c b/arch/x86/kernel/apic/probe_64.c index bc3e880..fcec2f1 100644 --- a/arch/x86/kernel/apic/probe_64.c +++ b/arch/x86/kernel/apic/probe_64.c @@ -44,6 +44,11 @@ static struct apic *apic_probe[] __initdata = { NULL, }; +static int apicid_phys_pkg_id(int initial_apic_id, int index_msb) +{ + return hard_smp_processor_id() >> index_msb; +} + /* * Check the APIC IDs in bios_cpu_apicid and choose the APIC mode. */ @@ -69,6 +74,11 @@ void __init default_setup_apic_routing(void) printk(KERN_INFO "Setting APIC routing to %s\n", apic->name); } + if (is_vsmp_box()) { + /* need to update phys_pkg_id */ + apic->phys_pkg_id = apicid_phys_pkg_id; + } + /* * Now that apic routing model is selected, configure the * fault handling for intr remapping. diff --git a/arch/x86/kernel/apic/summit_32.c b/arch/x86/kernel/apic/summit_32.c index 344eee4..eafdfbd1 100644 --- a/arch/x86/kernel/apic/summit_32.c +++ b/arch/x86/kernel/apic/summit_32.c @@ -44,7 +44,6 @@ #include <asm/ipi.h> #include <linux/kernel.h> #include <linux/string.h> -#include <linux/init.h> #include <linux/gfp.h> #include <linux/smp.h> diff --git a/arch/x86/kernel/apic/x2apic_cluster.c b/arch/x86/kernel/apic/x2apic_cluster.c index 8e4cbb2..a5371ec 100644 --- a/arch/x86/kernel/apic/x2apic_cluster.c +++ b/arch/x86/kernel/apic/x2apic_cluster.c @@ -17,11 +17,13 @@ static int x2apic_acpi_madt_oem_check(char *oem_id, char *oem_table_id) return x2apic_enabled(); } -/* Start with all IRQs pointing to boot CPU. IRQ balancing will shift them. */ - +/* + * need to use more than cpu 0, because we need more vectors when + * MSI-X are used. + */ static const struct cpumask *x2apic_target_cpus(void) { - return cpumask_of(0); + return cpu_online_mask; } /* @@ -170,7 +172,7 @@ static unsigned long set_apic_id(unsigned int id) static int x2apic_cluster_phys_pkg_id(int initial_apicid, int index_msb) { - return current_cpu_data.initial_apicid >> index_msb; + return initial_apicid >> index_msb; } static void x2apic_send_IPI_self(int vector) diff --git a/arch/x86/kernel/apic/x2apic_phys.c b/arch/x86/kernel/apic/x2apic_phys.c index a284359..a8989aa 100644 --- a/arch/x86/kernel/apic/x2apic_phys.c +++ b/arch/x86/kernel/apic/x2apic_phys.c @@ -27,11 +27,13 @@ static int x2apic_acpi_madt_oem_check(char *oem_id, char *oem_table_id) return 0; } -/* Start with all IRQs pointing to boot CPU. IRQ balancing will shift them. */ - +/* + * need to use more than cpu 0, because we need more vectors when + * MSI-X are used. + */ static const struct cpumask *x2apic_target_cpus(void) { - return cpumask_of(0); + return cpu_online_mask; } static void x2apic_vector_allocation_domain(int cpu, struct cpumask *retmask) @@ -162,7 +164,7 @@ static unsigned long set_apic_id(unsigned int id) static int x2apic_phys_pkg_id(int initial_apicid, int index_msb) { - return current_cpu_data.initial_apicid >> index_msb; + return initial_apicid >> index_msb; } static void x2apic_send_IPI_self(int vector) diff --git a/arch/x86/kernel/apic/x2apic_uv_x.c b/arch/x86/kernel/apic/x2apic_uv_x.c index 096d19a..6011593 100644 --- a/arch/x86/kernel/apic/x2apic_uv_x.c +++ b/arch/x86/kernel/apic/x2apic_uv_x.c @@ -46,7 +46,7 @@ static int early_get_nodeid(void) return node_id.s.node_id; } -static int uv_acpi_madt_oem_check(char *oem_id, char *oem_table_id) +static int __init uv_acpi_madt_oem_check(char *oem_id, char *oem_table_id) { if (!strcmp(oem_id, "SGI")) { if (!strcmp(oem_table_id, "UVL")) @@ -253,7 +253,7 @@ static void uv_send_IPI_self(int vector) apic_write(APIC_SELF_IPI, vector); } -struct apic apic_x2apic_uv_x = { +struct apic __refdata apic_x2apic_uv_x = { .name = "UV large system", .probe = NULL, @@ -261,7 +261,7 @@ struct apic apic_x2apic_uv_x = { .apic_id_registered = uv_apic_id_registered, .irq_delivery_mode = dest_Fixed, - .irq_dest_mode = 1, /* logical */ + .irq_dest_mode = 0, /* physical */ .target_cpus = uv_target_cpus, .disable_esr = 0, @@ -362,12 +362,6 @@ static __init void get_lowmem_redirect(unsigned long *base, unsigned long *size) BUG(); } -static __init void map_low_mmrs(void) -{ - init_extra_mapping_uc(UV_GLOBAL_MMR32_BASE, UV_GLOBAL_MMR32_SIZE); - init_extra_mapping_uc(UV_LOCAL_MMR_BASE, UV_LOCAL_MMR_SIZE); -} - enum map_type {map_wb, map_uc}; static __init void map_high(char *id, unsigned long base, int shift, @@ -395,26 +389,6 @@ static __init void map_gru_high(int max_pnode) map_high("GRU", gru.s.base, shift, max_pnode, map_wb); } -static __init void map_config_high(int max_pnode) -{ - union uvh_rh_gam_cfg_overlay_config_mmr_u cfg; - int shift = UVH_RH_GAM_CFG_OVERLAY_CONFIG_MMR_BASE_SHFT; - - cfg.v = uv_read_local_mmr(UVH_RH_GAM_CFG_OVERLAY_CONFIG_MMR); - if (cfg.s.enable) - map_high("CONFIG", cfg.s.base, shift, max_pnode, map_uc); -} - -static __init void map_mmr_high(int max_pnode) -{ - union uvh_rh_gam_mmr_overlay_config_mmr_u mmr; - int shift = UVH_RH_GAM_MMR_OVERLAY_CONFIG_MMR_BASE_SHFT; - - mmr.v = uv_read_local_mmr(UVH_RH_GAM_MMR_OVERLAY_CONFIG_MMR); - if (mmr.s.enable) - map_high("MMR", mmr.s.base, shift, max_pnode, map_uc); -} - static __init void map_mmioh_high(int max_pnode) { union uvh_rh_gam_mmioh_overlay_config_mmr_u mmioh; @@ -566,8 +540,6 @@ void __init uv_system_init(void) unsigned long mmr_base, present, paddr; unsigned short pnode_mask; - map_low_mmrs(); - m_n_config.v = uv_read_local_mmr(UVH_SI_ADDR_MAP_CONFIG); m_val = m_n_config.s.m_skt; n_val = m_n_config.s.n_skt; @@ -591,6 +563,8 @@ void __init uv_system_init(void) bytes = sizeof(struct uv_blade_info) * uv_num_possible_blades(); uv_blade_info = kmalloc(bytes, GFP_KERNEL); BUG_ON(!uv_blade_info); + for (blade = 0; blade < uv_num_possible_blades(); blade++) + uv_blade_info[blade].memory_nid = -1; get_lowmem_redirect(&lowmem_redir_base, &lowmem_redir_size); @@ -629,6 +603,9 @@ void __init uv_system_init(void) lcpu = uv_blade_info[blade].nr_possible_cpus; uv_blade_info[blade].nr_possible_cpus++; + /* Any node on the blade, else will contain -1. */ + uv_blade_info[blade].memory_nid = nid; + uv_cpu_hub_info(cpu)->lowmem_remap_base = lowmem_redir_base; uv_cpu_hub_info(cpu)->lowmem_remap_top = lowmem_redir_size; uv_cpu_hub_info(cpu)->m_val = m_val; @@ -662,11 +639,10 @@ void __init uv_system_init(void) pnode = (paddr >> m_val) & pnode_mask; blade = boot_pnode_to_blade(pnode); uv_node_to_blade[nid] = blade; + max_pnode = max(pnode, max_pnode); } map_gru_high(max_pnode); - map_mmr_high(max_pnode); - map_config_high(max_pnode); map_mmioh_high(max_pnode); uv_cpu_init(); diff --git a/arch/x86/kernel/apm_32.c b/arch/x86/kernel/apm_32.c index 79302e9..442b550 100644 --- a/arch/x86/kernel/apm_32.c +++ b/arch/x86/kernel/apm_32.c @@ -811,7 +811,7 @@ static int apm_do_idle(void) u8 ret = 0; int idled = 0; int polling; - int err; + int err = 0; polling = !!(current_thread_info()->status & TS_POLLING); if (polling) { diff --git a/arch/x86/kernel/asm-offsets_64.c b/arch/x86/kernel/asm-offsets_64.c index 898ecc4..4a6aeed 100644 --- a/arch/x86/kernel/asm-offsets_64.c +++ b/arch/x86/kernel/asm-offsets_64.c @@ -3,6 +3,7 @@ * This code generates raw asm output which is post-processed to extract * and format the required data. */ +#define COMPILE_OFFSETS #include <linux/crypto.h> #include <linux/sched.h> diff --git a/arch/x86/kernel/cpu/Makefile b/arch/x86/kernel/cpu/Makefile index 3efcb2b..c1f253d 100644 --- a/arch/x86/kernel/cpu/Makefile +++ b/arch/x86/kernel/cpu/Makefile @@ -7,6 +7,10 @@ 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 diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index e5b27d8..63fddcd0 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -258,13 +258,15 @@ static void __cpuinit amd_detect_cmp(struct cpuinfo_x86 *c) { #ifdef CONFIG_X86_HT unsigned bits; + int cpu = smp_processor_id(); bits = c->x86_coreid_bits; - /* Low order bits define the core id (index of core in socket) */ c->cpu_core_id = c->initial_apicid & ((1 << bits)-1); /* Convert the initial APIC ID into the socket ID */ c->phys_proc_id = c->initial_apicid >> bits; + /* use socket ID also for last level cache */ + per_cpu(cpu_llc_id, cpu) = c->phys_proc_id; #endif } @@ -354,7 +356,7 @@ static void __cpuinit early_init_amd(struct cpuinfo_x86 *c) #endif #if defined(CONFIG_X86_LOCAL_APIC) && defined(CONFIG_PCI) /* check CPU config space for extended APIC ID */ - if (c->x86 >= 0xf) { + if (cpu_has_apic && c->x86 >= 0xf) { unsigned int val; val = read_pci_config(0, 24, 0, 0x68); if ((val & ((1 << 17) | (1 << 18))) == ((1 << 17) | (1 << 18))) @@ -398,6 +400,13 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c) level = cpuid_eax(1); if((level >= 0x0f48 && level < 0x0f50) || level >= 0x0f58) set_cpu_cap(c, X86_FEATURE_REP_GOOD); + + /* + * Some BIOSes incorrectly force this feature, but only K8 + * revision D (model = 0x14) and later actually support it. + */ + if (c->x86_model < 0x14) + clear_cpu_cap(c, X86_FEATURE_LAHF_LM); } if (c->x86 == 0x10 || c->x86 == 0x11) set_cpu_cap(c, X86_FEATURE_REP_GOOD); diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 9fa3388..5ce60a8 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -59,7 +59,30 @@ void __init setup_cpu_local_masks(void) alloc_bootmem_cpumask_var(&cpu_sibling_setup_mask); } -static const struct cpu_dev *this_cpu __cpuinitdata; +static void __cpuinit default_init(struct cpuinfo_x86 *c) +{ +#ifdef CONFIG_X86_64 + display_cacheinfo(c); +#else + /* Not much we can do here... */ + /* Check if at least it has cpuid */ + if (c->cpuid_level == -1) { + /* No cpuid. It must be an ancient CPU */ + if (c->x86 == 4) + strcpy(c->x86_model_id, "486"); + else if (c->x86 == 3) + strcpy(c->x86_model_id, "386"); + } +#endif +} + +static const struct cpu_dev __cpuinitconst default_cpu = { + .c_init = default_init, + .c_vendor = "Unknown", + .c_x86_vendor = X86_VENDOR_UNKNOWN, +}; + +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 @@ -108,7 +131,7 @@ DEFINE_PER_CPU_PAGE_ALIGNED(struct gdt_page, gdt_page) = { .gdt = { /* data */ [GDT_ENTRY_APMBIOS_BASE+2] = { { { 0x0000ffff, 0x00409200 } } }, - [GDT_ENTRY_ESPFIX_SS] = { { { 0x00000000, 0x00c09200 } } }, + [GDT_ENTRY_ESPFIX_SS] = { { { 0x0000ffff, 0x00cf9200 } } }, [GDT_ENTRY_PERCPU] = { { { 0x0000ffff, 0x00cf9200 } } }, GDT_STACK_CANARY_INIT #endif @@ -332,29 +355,6 @@ void switch_to_new_gdt(int cpu) static const struct cpu_dev *__cpuinitdata cpu_devs[X86_VENDOR_NUM] = {}; -static void __cpuinit default_init(struct cpuinfo_x86 *c) -{ -#ifdef CONFIG_X86_64 - display_cacheinfo(c); -#else - /* Not much we can do here... */ - /* Check if at least it has cpuid */ - if (c->cpuid_level == -1) { - /* No cpuid. It must be an ancient CPU */ - if (c->x86 == 4) - strcpy(c->x86_model_id, "486"); - else if (c->x86 == 3) - strcpy(c->x86_model_id, "386"); - } -#endif -} - -static const struct cpu_dev __cpuinitconst default_cpu = { - .c_init = default_init, - .c_vendor = "Unknown", - .c_x86_vendor = X86_VENDOR_UNKNOWN, -}; - static void __cpuinit get_model_name(struct cpuinfo_x86 *c) { unsigned int *v; diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c index cf52215..2a50ef8 100644 --- a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c +++ b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c @@ -1,3 +1,4 @@ + /* * (c) 2003-2006 Advanced Micro Devices, Inc. * Your use of this code is subject to the terms and conditions of the @@ -117,20 +118,17 @@ static int query_current_values_with_pending_wait(struct powernow_k8_data *data) u32 i = 0; if (cpu_family == CPU_HW_PSTATE) { - if (data->currpstate == HW_PSTATE_INVALID) { - /* read (initial) hw pstate if not yet set */ - rdmsr(MSR_PSTATE_STATUS, lo, hi); - i = lo & HW_PSTATE_MASK; - - /* - * a workaround for family 11h erratum 311 might cause - * an "out-of-range Pstate if the core is in Pstate-0 - */ - if (i >= data->numps) - data->currpstate = HW_PSTATE_0; - else - data->currpstate = i; - } + rdmsr(MSR_PSTATE_STATUS, lo, hi); + i = lo & HW_PSTATE_MASK; + data->currpstate = i; + + /* + * a workaround for family 11h erratum 311 might cause + * an "out-of-range Pstate if the core is in Pstate-0 + */ + if ((boot_cpu_data.x86 == 0x11) && (i >= data->numps)) + data->currpstate = HW_PSTATE_0; + return 0; } do { @@ -301,7 +299,7 @@ static int transition_pstate(struct powernow_k8_data *data, u32 pstate) static int transition_fid_vid(struct powernow_k8_data *data, u32 reqfid, u32 reqvid) { - if (core_voltage_pre_transition(data, reqvid)) + if (core_voltage_pre_transition(data, reqvid, reqfid)) return 1; if (core_frequency_transition(data, reqfid)) @@ -329,17 +327,20 @@ static int transition_fid_vid(struct powernow_k8_data *data, /* Phase 1 - core voltage transition ... setup voltage */ static int core_voltage_pre_transition(struct powernow_k8_data *data, - u32 reqvid) + u32 reqvid, u32 reqfid) { u32 rvosteps = data->rvo; u32 savefid = data->currfid; - u32 maxvid, lo; + u32 maxvid, lo, rvomult = 1; dprintk("ph1 (cpu%d): start, currfid 0x%x, currvid 0x%x, " "reqvid 0x%x, rvo 0x%x\n", smp_processor_id(), data->currfid, data->currvid, reqvid, data->rvo); + if ((savefid < LO_FID_TABLE_TOP) && (reqfid < LO_FID_TABLE_TOP)) + rvomult = 2; + rvosteps *= rvomult; rdmsr(MSR_FIDVID_STATUS, lo, maxvid); maxvid = 0x1f & (maxvid >> 16); dprintk("ph1 maxvid=0x%x\n", maxvid); @@ -353,7 +354,8 @@ static int core_voltage_pre_transition(struct powernow_k8_data *data, return 1; } - while ((rvosteps > 0) && ((data->rvo + data->currvid) > reqvid)) { + while ((rvosteps > 0) && + ((rvomult * data->rvo + data->currvid) > reqvid)) { if (data->currvid == maxvid) { rvosteps = 0; } else { @@ -386,13 +388,6 @@ static int core_frequency_transition(struct powernow_k8_data *data, u32 reqfid) u32 vcoreqfid, vcocurrfid, vcofiddiff; u32 fid_interval, savevid = data->currvid; - if ((reqfid < HI_FID_TABLE_BOTTOM) && - (data->currfid < HI_FID_TABLE_BOTTOM)) { - printk(KERN_ERR PFX "ph2: illegal lo-lo transition " - "0x%x 0x%x\n", reqfid, data->currfid); - return 1; - } - if (data->currfid == reqfid) { printk(KERN_ERR PFX "ph2 null fid transition 0x%x\n", data->currfid); @@ -409,6 +404,9 @@ static int core_frequency_transition(struct powernow_k8_data *data, u32 reqfid) vcofiddiff = vcocurrfid > vcoreqfid ? vcocurrfid - vcoreqfid : vcoreqfid - vcocurrfid; + if ((reqfid <= LO_FID_TABLE_TOP) && (data->currfid <= LO_FID_TABLE_TOP)) + vcofiddiff = 0; + while (vcofiddiff > 2) { (data->currfid & 1) ? (fid_interval = 1) : (fid_interval = 2); @@ -510,41 +508,34 @@ static int core_voltage_post_transition(struct powernow_k8_data *data, return 0; } -static int check_supported_cpu(unsigned int cpu) +static void check_supported_cpu(void *_rc) { - cpumask_t oldmask; u32 eax, ebx, ecx, edx; - unsigned int rc = 0; + int *rc = _rc; - oldmask = current->cpus_allowed; - set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu)); - - if (smp_processor_id() != cpu) { - printk(KERN_ERR PFX "limiting to cpu %u failed\n", cpu); - goto out; - } + *rc = -ENODEV; if (current_cpu_data.x86_vendor != X86_VENDOR_AMD) - goto out; + return; eax = cpuid_eax(CPUID_PROCESSOR_SIGNATURE); if (((eax & CPUID_XFAM) != CPUID_XFAM_K8) && ((eax & CPUID_XFAM) < CPUID_XFAM_10H)) - goto out; + return; if ((eax & CPUID_XFAM) == CPUID_XFAM_K8) { if (((eax & CPUID_USE_XFAM_XMOD) != CPUID_USE_XFAM_XMOD) || ((eax & CPUID_XMOD) > CPUID_XMOD_REV_MASK)) { printk(KERN_INFO PFX "Processor cpuid %x not supported\n", eax); - goto out; + return; } eax = cpuid_eax(CPUID_GET_MAX_CAPABILITIES); if (eax < CPUID_FREQ_VOLT_CAPABILITIES) { printk(KERN_INFO PFX "No frequency change capabilities detected\n"); - goto out; + return; } cpuid(CPUID_FREQ_VOLT_CAPABILITIES, &eax, &ebx, &ecx, &edx); @@ -552,21 +543,17 @@ static int check_supported_cpu(unsigned int cpu) != P_STATE_TRANSITION_CAPABLE) { printk(KERN_INFO PFX "Power state transitions not supported\n"); - goto out; + return; } } else { /* must be a HW Pstate capable processor */ cpuid(CPUID_FREQ_VOLT_CAPABILITIES, &eax, &ebx, &ecx, &edx); if ((edx & USE_HW_PSTATE) == USE_HW_PSTATE) cpu_family = CPU_HW_PSTATE; else - goto out; + return; } - rc = 1; - -out: - set_cpus_allowed_ptr(current, &oldmask); - return rc; + *rc = 0; } static int check_pst_table(struct powernow_k8_data *data, struct pst_s *pst, @@ -823,13 +810,14 @@ static void powernow_k8_acpi_pst_values(struct powernow_k8_data *data, if (!data->acpi_data.state_count || (cpu_family == CPU_HW_PSTATE)) return; - control = data->acpi_data.states[index].control; data->irt = (control - >> IRT_SHIFT) & IRT_MASK; data->rvo = (control >> - RVO_SHIFT) & RVO_MASK; data->exttype = (control - >> EXT_TYPE_SHIFT) & EXT_TYPE_MASK; - data->plllock = (control >> PLL_L_SHIFT) & PLL_L_MASK; data->vidmvs = 1 - << ((control >> MVS_SHIFT) & MVS_MASK); data->vstable = - (control >> VST_SHIFT) & VST_MASK; } + control = data->acpi_data.states[index].control; + data->irt = (control >> IRT_SHIFT) & IRT_MASK; + data->rvo = (control >> RVO_SHIFT) & RVO_MASK; + data->exttype = (control >> EXT_TYPE_SHIFT) & EXT_TYPE_MASK; + data->plllock = (control >> PLL_L_SHIFT) & PLL_L_MASK; + data->vidmvs = 1 << ((control >> MVS_SHIFT) & MVS_MASK); + data->vstable = (control >> VST_SHIFT) & VST_MASK; +} static int powernow_k8_cpu_init_acpi(struct powernow_k8_data *data) { @@ -1046,6 +1034,19 @@ static int get_transition_latency(struct powernow_k8_data *data) if (cur_latency > max_latency) max_latency = cur_latency; } + if (max_latency == 0) { + /* + * Fam 11h always returns 0 as transition latency. + * This is intended and means "very fast". While cpufreq core + * and governors currently can handle that gracefully, better + * set it to 1 to avoid problems in the future. + * For all others it's a BIOS bug. + */ + if (!boot_cpu_data.x86 == 0x11) + printk(KERN_ERR FW_WARN PFX "Invalid zero transition " + "latency\n"); + max_latency = 1; + } /* value in usecs, needs to be in nanoseconds */ return 1000 * max_latency; } @@ -1080,20 +1081,12 @@ static int transition_frequency_fidvid(struct powernow_k8_data *data, return 0; } - if ((fid < HI_FID_TABLE_BOTTOM) && - (data->currfid < HI_FID_TABLE_BOTTOM)) { - printk(KERN_ERR PFX - "ignoring illegal change in lo freq table-%x to 0x%x\n", - data->currfid, fid); - return 1; - } - dprintk("cpu %d, changing to fid 0x%x, vid 0x%x\n", smp_processor_id(), fid, vid); freqs.old = find_khz_freq_from_fid(data->currfid); freqs.new = find_khz_freq_from_fid(fid); - for_each_cpu_mask_nr(i, *(data->available_cores)) { + for_each_cpu(i, data->available_cores) { freqs.cpu = i; cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); } @@ -1101,7 +1094,7 @@ static int transition_frequency_fidvid(struct powernow_k8_data *data, res = transition_fid_vid(data, fid, vid); freqs.new = find_khz_freq_from_fid(data->currfid); - for_each_cpu_mask_nr(i, *(data->available_cores)) { + for_each_cpu(i, data->available_cores) { freqs.cpu = i; cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); } @@ -1126,7 +1119,7 @@ static int transition_frequency_pstate(struct powernow_k8_data *data, data->currpstate); freqs.new = find_khz_freq_from_pstate(data->powernow_table, pstate); - for_each_cpu_mask_nr(i, *(data->available_cores)) { + for_each_cpu(i, data->available_cores) { freqs.cpu = i; cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); } @@ -1134,7 +1127,7 @@ static int transition_frequency_pstate(struct powernow_k8_data *data, res = transition_pstate(data, pstate); freqs.new = find_khz_freq_from_pstate(data->powernow_table, pstate); - for_each_cpu_mask_nr(i, *(data->available_cores)) { + for_each_cpu(i, data->available_cores) { freqs.cpu = i; cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); } @@ -1235,21 +1228,47 @@ static int powernowk8_verify(struct cpufreq_policy *pol) return cpufreq_frequency_table_verify(pol, data->powernow_table); } -static const char ACPI_PSS_BIOS_BUG_MSG[] = - KERN_ERR FW_BUG PFX "No compatible ACPI _PSS objects found.\n" - KERN_ERR FW_BUG PFX "Try again with latest BIOS.\n"; +struct init_on_cpu { + struct powernow_k8_data *data; + int rc; +}; + +static void __cpuinit powernowk8_cpu_init_on_cpu(void *_init_on_cpu) +{ + struct init_on_cpu *init_on_cpu = _init_on_cpu; + + if (pending_bit_stuck()) { + printk(KERN_ERR PFX "failing init, change pending bit set\n"); + init_on_cpu->rc = -ENODEV; + return; + } + + if (query_current_values_with_pending_wait(init_on_cpu->data)) { + init_on_cpu->rc = -ENODEV; + return; + } + + if (cpu_family == CPU_OPTERON) + fidvid_msr_init(); + + init_on_cpu->rc = 0; +} /* per CPU init entry point to the driver */ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol) { + static const char ACPI_PSS_BIOS_BUG_MSG[] = + KERN_ERR FW_BUG PFX "No compatible ACPI _PSS objects found.\n" + FW_BUG PFX "Try again with latest BIOS.\n"; struct powernow_k8_data *data; - cpumask_t oldmask; + struct init_on_cpu init_on_cpu; int rc; if (!cpu_online(pol->cpu)) return -ENODEV; - if (!check_supported_cpu(pol->cpu)) + smp_call_function_single(pol->cpu, check_supported_cpu, &rc, 1); + if (rc) return -ENODEV; data = kzalloc(sizeof(struct powernow_k8_data), GFP_KERNEL); @@ -1289,27 +1308,12 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol) pol->cpuinfo.transition_latency = get_transition_latency(data); /* only run on specific CPU from here on */ - oldmask = current->cpus_allowed; - set_cpus_allowed_ptr(current, &cpumask_of_cpu(pol->cpu)); - - if (smp_processor_id() != pol->cpu) { - printk(KERN_ERR PFX "limiting to cpu %u failed\n", pol->cpu); - goto err_out_unmask; - } - - if (pending_bit_stuck()) { - printk(KERN_ERR PFX "failing init, change pending bit set\n"); - goto err_out_unmask; - } - - if (query_current_values_with_pending_wait(data)) - goto err_out_unmask; - - if (cpu_family == CPU_OPTERON) - fidvid_msr_init(); - - /* run on any CPU again */ - set_cpus_allowed_ptr(current, &oldmask); + init_on_cpu.data = data; + smp_call_function_single(data->cpu, powernowk8_cpu_init_on_cpu, + &init_on_cpu, 1); + rc = init_on_cpu.rc; + if (rc != 0) + goto err_out_exit_acpi; if (cpu_family == CPU_HW_PSTATE) cpumask_copy(pol->cpus, cpumask_of(pol->cpu)); @@ -1346,8 +1350,7 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol) return 0; -err_out_unmask: - set_cpus_allowed_ptr(current, &oldmask); +err_out_exit_acpi: powernow_k8_cpu_exit_acpi(data); err_out: @@ -1372,28 +1375,25 @@ static int __devexit powernowk8_cpu_exit(struct cpufreq_policy *pol) return 0; } +static void query_values_on_cpu(void *_err) +{ + int *err = _err; + struct powernow_k8_data *data = __get_cpu_var(powernow_data); + + *err = query_current_values_with_pending_wait(data); +} + static unsigned int powernowk8_get(unsigned int cpu) { - struct powernow_k8_data *data; - cpumask_t oldmask = current->cpus_allowed; + struct powernow_k8_data *data = per_cpu(powernow_data, cpu); unsigned int khz = 0; - unsigned int first; - - first = cpumask_first(cpu_core_mask(cpu)); - data = per_cpu(powernow_data, first); + int err; if (!data) return -EINVAL; - set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu)); - if (smp_processor_id() != cpu) { - printk(KERN_ERR PFX - "limiting to CPU %d failed in powernowk8_get\n", cpu); - set_cpus_allowed_ptr(current, &oldmask); - return 0; - } - - if (query_current_values_with_pending_wait(data)) + smp_call_function_single(cpu, query_values_on_cpu, &err, true); + if (err) goto out; if (cpu_family == CPU_HW_PSTATE) @@ -1404,7 +1404,6 @@ static unsigned int powernowk8_get(unsigned int cpu) out: - set_cpus_allowed_ptr(current, &oldmask); return khz; } @@ -1430,7 +1429,9 @@ static int __cpuinit powernowk8_init(void) unsigned int i, supported_cpus = 0; for_each_online_cpu(i) { - if (check_supported_cpu(i)) + int rc; + smp_call_function_single(i, check_supported_cpu, &rc, 1); + if (rc == 0) supported_cpus++; } diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k8.h b/arch/x86/kernel/cpu/cpufreq/powernow-k8.h index 6c6698f..02ce824 100644 --- a/arch/x86/kernel/cpu/cpufreq/powernow-k8.h +++ b/arch/x86/kernel/cpu/cpufreq/powernow-k8.h @@ -215,7 +215,8 @@ struct pst_s { #define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, "powernow-k8", msg) -static int core_voltage_pre_transition(struct powernow_k8_data *data, u32 reqvid); +static int core_voltage_pre_transition(struct powernow_k8_data *data, + u32 reqvid, u32 regfid); static int core_voltage_post_transition(struct powernow_k8_data *data, u32 reqvid); static int core_frequency_transition(struct powernow_k8_data *data, u32 reqfid); @@ -223,14 +224,3 @@ static void powernow_k8_acpi_pst_values(struct powernow_k8_data *data, unsigned static int fill_powernow_table_pstate(struct powernow_k8_data *data, struct cpufreq_frequency_table *powernow_table); static int fill_powernow_table_fidvid(struct powernow_k8_data *data, struct cpufreq_frequency_table *powernow_table); - -#ifdef CONFIG_SMP -static inline void define_siblings(int cpu, cpumask_t cpu_sharedcore_mask[]) -{ -} -#else -static inline void define_siblings(int cpu, cpumask_t cpu_sharedcore_mask[]) -{ - cpu_set(0, cpu_sharedcore_mask[0]); -} -#endif diff --git a/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c b/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c index 55c831e..8d672ef 100644 --- a/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c +++ b/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c @@ -323,14 +323,8 @@ static unsigned int get_cur_freq(unsigned int cpu) { unsigned l, h; unsigned clock_freq; - cpumask_t saved_mask; - saved_mask = current->cpus_allowed; - set_cpus_allowed_ptr(current, &cpumask_of_cpu(cpu)); - if (smp_processor_id() != cpu) - return 0; - - rdmsr(MSR_IA32_PERF_STATUS, l, h); + rdmsr_on_cpu(cpu, MSR_IA32_PERF_STATUS, &l, &h); clock_freq = extract_clock(l, cpu, 0); if (unlikely(clock_freq == 0)) { @@ -340,11 +334,9 @@ static unsigned int get_cur_freq(unsigned int cpu) * P-state transition (like TM2). Get the last freq set * in PERF_CTL. */ - rdmsr(MSR_IA32_PERF_CTL, l, h); + rdmsr_on_cpu(cpu, MSR_IA32_PERF_CTL, &l, &h); clock_freq = extract_clock(l, cpu, 1); } - - set_cpus_allowed_ptr(current, &saved_mask); return clock_freq; } @@ -467,15 +459,10 @@ static int centrino_target (struct cpufreq_policy *policy, struct cpufreq_freqs freqs; int retval = 0; unsigned int j, k, first_cpu, tmp; - cpumask_var_t saved_mask, covered_cpus; + cpumask_var_t covered_cpus; - if (unlikely(!alloc_cpumask_var(&saved_mask, GFP_KERNEL))) - return -ENOMEM; - if (unlikely(!zalloc_cpumask_var(&covered_cpus, GFP_KERNEL))) { - free_cpumask_var(saved_mask); + if (unlikely(!zalloc_cpumask_var(&covered_cpus, GFP_KERNEL))) return -ENOMEM; - } - cpumask_copy(saved_mask, ¤t->cpus_allowed); if (unlikely(per_cpu(centrino_model, cpu) == NULL)) { retval = -ENODEV; @@ -493,7 +480,7 @@ static int centrino_target (struct cpufreq_policy *policy, first_cpu = 1; for_each_cpu(j, policy->cpus) { - const struct cpumask *mask; + int good_cpu; /* cpufreq holds the hotplug lock, so we are safe here */ if (!cpu_online(j)) @@ -504,32 +491,30 @@ static int centrino_target (struct cpufreq_policy *policy, * Make sure we are running on CPU that wants to change freq */ if (policy->shared_type == CPUFREQ_SHARED_TYPE_ANY) - mask = policy->cpus; + good_cpu = cpumask_any_and(policy->cpus, + cpu_online_mask); else - mask = cpumask_of(j); + good_cpu = j; - set_cpus_allowed_ptr(current, mask); - preempt_disable(); - if (unlikely(!cpu_isset(smp_processor_id(), *mask))) { + if (good_cpu >= nr_cpu_ids) { dprintk("couldn't limit to CPUs in this domain\n"); retval = -EAGAIN; if (first_cpu) { /* We haven't started the transition yet. */ - goto migrate_end; + goto out; } - preempt_enable(); break; } msr = per_cpu(centrino_model, cpu)->op_points[newstate].index; if (first_cpu) { - rdmsr(MSR_IA32_PERF_CTL, oldmsr, h); + rdmsr_on_cpu(good_cpu, MSR_IA32_PERF_CTL, &oldmsr, &h); if (msr == (oldmsr & 0xffff)) { dprintk("no change needed - msr was and needs " "to be %x\n", oldmsr); retval = 0; - goto migrate_end; + goto out; } freqs.old = extract_clock(oldmsr, cpu, 0); @@ -553,14 +538,11 @@ static int centrino_target (struct cpufreq_policy *policy, oldmsr |= msr; } - wrmsr(MSR_IA32_PERF_CTL, oldmsr, h); - if (policy->shared_type == CPUFREQ_SHARED_TYPE_ANY) { - preempt_enable(); + wrmsr_on_cpu(good_cpu, MSR_IA32_PERF_CTL, oldmsr, h); + if (policy->shared_type == CPUFREQ_SHARED_TYPE_ANY) break; - } - cpu_set(j, *covered_cpus); - preempt_enable(); + cpumask_set_cpu(j, covered_cpus); } for_each_cpu(k, policy->cpus) { @@ -578,10 +560,8 @@ static int centrino_target (struct cpufreq_policy *policy, * Best effort undo.. */ - for_each_cpu_mask_nr(j, *covered_cpus) { - set_cpus_allowed_ptr(current, &cpumask_of_cpu(j)); - wrmsr(MSR_IA32_PERF_CTL, oldmsr, h); - } + for_each_cpu(j, covered_cpus) + wrmsr_on_cpu(j, MSR_IA32_PERF_CTL, oldmsr, h); tmp = freqs.new; freqs.new = freqs.old; @@ -593,15 +573,9 @@ static int centrino_target (struct cpufreq_policy *policy, cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); } } - set_cpus_allowed_ptr(current, saved_mask); retval = 0; - goto out; -migrate_end: - preempt_enable(); - set_cpus_allowed_ptr(current, saved_mask); out: - free_cpumask_var(saved_mask); free_cpumask_var(covered_cpus); return retval; } diff --git a/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c b/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c index 016c1a4..6911e91 100644 --- a/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c +++ b/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c @@ -89,7 +89,8 @@ static int speedstep_find_register(void) * speedstep_set_state - set the SpeedStep state * @state: new processor frequency state (SPEEDSTEP_LOW or SPEEDSTEP_HIGH) * - * Tries to change the SpeedStep state. + * Tries to change the SpeedStep state. Can be called from + * smp_call_function_single. */ static void speedstep_set_state(unsigned int state) { @@ -143,6 +144,11 @@ static void speedstep_set_state(unsigned int state) return; } +/* Wrapper for smp_call_function_single. */ +static void _speedstep_set_state(void *_state) +{ + speedstep_set_state(*(unsigned int *)_state); +} /** * speedstep_activate - activate SpeedStep control in the chipset @@ -226,22 +232,28 @@ static unsigned int speedstep_detect_chipset(void) return 0; } -static unsigned int _speedstep_get(const struct cpumask *cpus) -{ +struct get_freq_data { unsigned int speed; - cpumask_t cpus_allowed; - - cpus_allowed = current->cpus_allowed; - set_cpus_allowed_ptr(current, cpus); - speed = speedstep_get_frequency(speedstep_processor); - set_cpus_allowed_ptr(current, &cpus_allowed); - dprintk("detected %u kHz as current frequency\n", speed); - return speed; + unsigned int processor; +}; + +static void get_freq_data(void *_data) +{ + struct get_freq_data *data = _data; + + data->speed = speedstep_get_frequency(data->processor); } static unsigned int speedstep_get(unsigned int cpu) { - return _speedstep_get(cpumask_of(cpu)); + struct get_freq_data data = { .processor = cpu }; + + /* You're supposed to ensure CPU is online. */ + if (smp_call_function_single(cpu, get_freq_data, &data, 1) != 0) + BUG(); + + dprintk("detected %u kHz as current frequency\n", data.speed); + return data.speed; } /** @@ -257,16 +269,16 @@ static int speedstep_target(struct cpufreq_policy *policy, unsigned int target_freq, unsigned int relation) { - unsigned int newstate = 0; + unsigned int newstate = 0, policy_cpu; struct cpufreq_freqs freqs; - cpumask_t cpus_allowed; int i; if (cpufreq_frequency_table_target(policy, &speedstep_freqs[0], target_freq, relation, &newstate)) return -EINVAL; - freqs.old = _speedstep_get(policy->cpus); + policy_cpu = cpumask_any_and(policy->cpus, cpu_online_mask); + freqs.old = speedstep_get(policy_cpu); freqs.new = speedstep_freqs[newstate].frequency; freqs.cpu = policy->cpu; @@ -276,20 +288,13 @@ static int speedstep_target(struct cpufreq_policy *policy, if (freqs.old == freqs.new) return 0; - cpus_allowed = current->cpus_allowed; - for_each_cpu(i, policy->cpus) { freqs.cpu = i; cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); } - /* switch to physical CPU where state is to be changed */ - set_cpus_allowed_ptr(current, policy->cpus); - - speedstep_set_state(newstate); - - /* allow to be run on all CPUs */ - set_cpus_allowed_ptr(current, &cpus_allowed); + smp_call_function_single(policy_cpu, _speedstep_set_state, &newstate, + true); for_each_cpu(i, policy->cpus) { freqs.cpu = i; @@ -312,33 +317,43 @@ static int speedstep_verify(struct cpufreq_policy *policy) return cpufreq_frequency_table_verify(policy, &speedstep_freqs[0]); } +struct get_freqs { + struct cpufreq_policy *policy; + int ret; +}; + +static void get_freqs_on_cpu(void *_get_freqs) +{ + struct get_freqs *get_freqs = _get_freqs; + + get_freqs->ret = + speedstep_get_freqs(speedstep_processor, + &speedstep_freqs[SPEEDSTEP_LOW].frequency, + &speedstep_freqs[SPEEDSTEP_HIGH].frequency, + &get_freqs->policy->cpuinfo.transition_latency, + &speedstep_set_state); +} static int speedstep_cpu_init(struct cpufreq_policy *policy) { - int result = 0; - unsigned int speed; - cpumask_t cpus_allowed; + int result; + unsigned int policy_cpu, speed; + struct get_freqs gf; /* only run on CPU to be set, or on its sibling */ #ifdef CONFIG_SMP cpumask_copy(policy->cpus, cpu_sibling_mask(policy->cpu)); #endif - - cpus_allowed = current->cpus_allowed; - set_cpus_allowed_ptr(current, policy->cpus); + policy_cpu = cpumask_any_and(policy->cpus, cpu_online_mask); /* detect low and high frequency and transition latency */ - result = speedstep_get_freqs(speedstep_processor, - &speedstep_freqs[SPEEDSTEP_LOW].frequency, - &speedstep_freqs[SPEEDSTEP_HIGH].frequency, - &policy->cpuinfo.transition_latency, - &speedstep_set_state); - set_cpus_allowed_ptr(current, &cpus_allowed); - if (result) - return result; + gf.policy = policy; + smp_call_function_single(policy_cpu, get_freqs_on_cpu, &gf, 1); + if (gf.ret) + return gf.ret; /* get current speed setting */ - speed = _speedstep_get(policy->cpus); + speed = speedstep_get(policy_cpu); if (!speed) return -EIO; diff --git a/arch/x86/kernel/cpu/cpufreq/speedstep-lib.c b/arch/x86/kernel/cpu/cpufreq/speedstep-lib.c index 2e3c686..f4c290b 100644 --- a/arch/x86/kernel/cpu/cpufreq/speedstep-lib.c +++ b/arch/x86/kernel/cpu/cpufreq/speedstep-lib.c @@ -226,6 +226,7 @@ static unsigned int pentium4_get_frequency(void) } +/* Warning: may get called from smp_call_function_single. */ unsigned int speedstep_get_frequency(unsigned int processor) { switch (processor) { diff --git a/arch/x86/kernel/cpu/mcheck/Makefile b/arch/x86/kernel/cpu/mcheck/Makefile index 45004fa..188a1ca 100644 --- a/arch/x86/kernel/cpu/mcheck/Makefile +++ b/arch/x86/kernel/cpu/mcheck/Makefile @@ -1,11 +1,12 @@ -obj-y = mce.o therm_throt.o +obj-y = mce.o obj-$(CONFIG_X86_NEW_MCE) += mce-severity.o obj-$(CONFIG_X86_OLD_MCE) += k7.o p4.o p6.o obj-$(CONFIG_X86_ANCIENT_MCE) += winchip.o p5.o -obj-$(CONFIG_X86_MCE_P4THERMAL) += mce_intel.o -obj-$(CONFIG_X86_MCE_INTEL) += mce_intel_64.o mce_intel.o -obj-$(CONFIG_X86_MCE_AMD) += mce_amd_64.o +obj-$(CONFIG_X86_MCE_INTEL) += mce_intel.o +obj-$(CONFIG_X86_MCE_AMD) += mce_amd.o obj-$(CONFIG_X86_MCE_NONFATAL) += non-fatal.o obj-$(CONFIG_X86_MCE_THRESHOLD) += threshold.o obj-$(CONFIG_X86_MCE_INJECT) += mce-inject.o + +obj-$(CONFIG_X86_THERMAL_VECTOR) += therm_throt.o diff --git a/arch/x86/kernel/cpu/mcheck/k7.c b/arch/x86/kernel/cpu/mcheck/k7.c index 89e5104..b945d5d 100644 --- a/arch/x86/kernel/cpu/mcheck/k7.c +++ b/arch/x86/kernel/cpu/mcheck/k7.c @@ -10,10 +10,9 @@ #include <asm/processor.h> #include <asm/system.h> +#include <asm/mce.h> #include <asm/msr.h> -#include "mce.h" - /* Machine Check Handler For AMD Athlon/Duron: */ static void k7_machine_check(struct pt_regs *regs, long error_code) { diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c index fabba15..0121304 100644 --- a/arch/x86/kernel/cpu/mcheck/mce.c +++ b/arch/x86/kernel/cpu/mcheck/mce.c @@ -44,7 +44,6 @@ #include <asm/msr.h> #include "mce-internal.h" -#include "mce.h" /* Handle unconfigured int18 (should never happen) */ static void unexpected_machine_check(struct pt_regs *regs, long error_code) @@ -57,7 +56,7 @@ static void unexpected_machine_check(struct pt_regs *regs, long error_code) void (*machine_check_vector)(struct pt_regs *, long error_code) = unexpected_machine_check; -int mce_disabled; +int mce_disabled __read_mostly; #ifdef CONFIG_X86_NEW_MCE @@ -76,21 +75,22 @@ DEFINE_PER_CPU(unsigned, mce_exception_count); * 2: SIGBUS or log uncorrected errors (if possible), log corrected errors * 3: never panic or SIGBUS, log all errors (for testing only) */ -static int tolerant = 1; -static int banks; -static u64 *bank; -static unsigned long notify_user; -static int rip_msr; -static int mce_bootlog = -1; -static int monarch_timeout = -1; -static int mce_panic_timeout; -static int mce_dont_log_ce; -int mce_cmci_disabled; -int mce_ignore_ce; -int mce_ser; - -static char trigger[128]; -static char *trigger_argv[2] = { trigger, NULL }; +static int tolerant __read_mostly = 1; +static int banks __read_mostly; +static u64 *bank __read_mostly; +static int rip_msr __read_mostly; +static int mce_bootlog __read_mostly = -1; +static int monarch_timeout __read_mostly = -1; +static int mce_panic_timeout __read_mostly; +static int mce_dont_log_ce __read_mostly; +int mce_cmci_disabled __read_mostly; +int mce_ignore_ce __read_mostly; +int mce_ser __read_mostly; + +/* User mode helper program triggered by machine check event */ +static unsigned long mce_need_notify; +static char mce_helper[128]; +static char *mce_helper_argv[2] = { mce_helper, NULL }; static unsigned long dont_init_banks; @@ -180,7 +180,7 @@ void mce_log(struct mce *mce) wmb(); mce->finished = 1; - set_bit(0, ¬ify_user); + set_bit(0, &mce_need_notify); } static void print_mce(struct mce *m) @@ -194,14 +194,14 @@ static void print_mce(struct mce *m) m->cs, m->ip); if (m->cs == __KERNEL_CS) print_symbol("{%s}", m->ip); - printk("\n"); + printk(KERN_CONT "\n"); } printk(KERN_EMERG "TSC %llx ", m->tsc); if (m->addr) - printk("ADDR %llx ", m->addr); + printk(KERN_CONT "ADDR %llx ", m->addr); if (m->misc) - printk("MISC %llx ", m->misc); - printk("\n"); + printk(KERN_CONT "MISC %llx ", m->misc); + printk(KERN_CONT "\n"); printk(KERN_EMERG "PROCESSOR %u:%x TIME %llu SOCKET %u APIC %x\n", m->cpuvendor, m->cpuid, m->time, m->socketid, m->apicid); @@ -209,13 +209,13 @@ static void print_mce(struct mce *m) static void print_mce_head(void) { - printk(KERN_EMERG "\n" KERN_EMERG "HARDWARE ERROR\n"); + printk(KERN_EMERG "\nHARDWARE ERROR\n"); } static void print_mce_tail(void) { printk(KERN_EMERG "This is not a software problem!\n" - KERN_EMERG "Run through mcelog --ascii to decode and contact your hardware vendor\n"); + "Run through mcelog --ascii to decode and contact your hardware vendor\n"); } #define PANIC_TIMEOUT 5 /* 5 seconds */ @@ -691,18 +691,21 @@ static atomic_t global_nwo; * in the entry order. * TBD double check parallel CPU hotunplug */ -static int mce_start(int no_way_out, int *order) +static int mce_start(int *no_way_out) { - int nwo; + int order; int cpus = num_online_cpus(); u64 timeout = (u64)monarch_timeout * NSEC_PER_USEC; - if (!timeout) { - *order = -1; - return no_way_out; - } + if (!timeout) + return -1; - atomic_add(no_way_out, &global_nwo); + atomic_add(*no_way_out, &global_nwo); + /* + * global_nwo should be updated before mce_callin + */ + smp_wmb(); + order = atomic_add_return(1, &mce_callin); /* * Wait for everyone. @@ -710,40 +713,43 @@ static int mce_start(int no_way_out, int *order) while (atomic_read(&mce_callin) != cpus) { if (mce_timed_out(&timeout)) { atomic_set(&global_nwo, 0); - *order = -1; - return no_way_out; + return -1; } ndelay(SPINUNIT); } /* - * Cache the global no_way_out state. + * mce_callin should be read before global_nwo */ - nwo = atomic_read(&global_nwo); + smp_rmb(); - /* - * Monarch starts executing now, the others wait. - */ - if (*order == 1) { + if (order == 1) { + /* + * Monarch: Starts executing now, the others wait. + */ atomic_set(&mce_executing, 1); - return nwo; + } else { + /* + * Subject: Now start the scanning loop one by one in + * the original callin order. + * This way when there are any shared banks it will be + * only seen by one CPU before cleared, avoiding duplicates. + */ + while (atomic_read(&mce_executing) < order) { + if (mce_timed_out(&timeout)) { + atomic_set(&global_nwo, 0); + return -1; + } + ndelay(SPINUNIT); + } } /* - * Now start the scanning loop one by one - * in the original callin order. - * This way when there are any shared banks it will - * be only seen by one CPU before cleared, avoiding duplicates. + * Cache the global no_way_out state. */ - while (atomic_read(&mce_executing) < *order) { - if (mce_timed_out(&timeout)) { - atomic_set(&global_nwo, 0); - *order = -1; - return no_way_out; - } - ndelay(SPINUNIT); - } - return nwo; + *no_way_out = atomic_read(&global_nwo); + + return order; } /* @@ -863,7 +869,6 @@ void do_machine_check(struct pt_regs *regs, long error_code) * check handler. */ int order; - /* * If no_way_out gets set, there is no safe way to recover from this * MCE. If tolerant is cranked up, we'll try anyway. @@ -887,7 +892,6 @@ void do_machine_check(struct pt_regs *regs, long error_code) if (!banks) goto out; - order = atomic_add_return(1, &mce_callin); mce_setup(&m); m.mcgstatus = mce_rdmsrl(MSR_IA32_MCG_STATUS); @@ -909,7 +913,7 @@ void do_machine_check(struct pt_regs *regs, long error_code) * This way we don't report duplicated events on shared banks * because the first one to see it will clear it. */ - no_way_out = mce_start(no_way_out, &order); + order = mce_start(&no_way_out); for (i = 0; i < banks; i++) { __clear_bit(i, toclear); if (!bank[i]) @@ -1113,12 +1117,12 @@ static void mcheck_timer(unsigned long data) *n = min(*n*2, (int)round_jiffies_relative(check_interval*HZ)); t->expires = jiffies + *n; - add_timer(t); + add_timer_on(t, smp_processor_id()); } static void mce_do_trigger(struct work_struct *work) { - call_usermodehelper(trigger, trigger_argv, NULL, UMH_NO_WAIT); + call_usermodehelper(mce_helper, mce_helper_argv, NULL, UMH_NO_WAIT); } static DECLARE_WORK(mce_trigger_work, mce_do_trigger); @@ -1135,7 +1139,7 @@ int mce_notify_irq(void) clear_thread_flag(TIF_MCE_NOTIFY); - if (test_and_clear_bit(0, ¬ify_user)) { + if (test_and_clear_bit(0, &mce_need_notify)) { wake_up_interruptible(&mce_wait); /* @@ -1143,7 +1147,7 @@ int mce_notify_irq(void) * work_pending is always cleared before the function is * executed. */ - if (trigger[0] && !work_pending(&mce_trigger_work)) + if (mce_helper[0] && !work_pending(&mce_trigger_work)) schedule_work(&mce_trigger_work); if (__ratelimit(&ratelimit)) @@ -1222,8 +1226,13 @@ static void mce_init(void) } /* Add per CPU specific workarounds here */ -static void mce_cpu_quirks(struct cpuinfo_x86 *c) +static int mce_cpu_quirks(struct cpuinfo_x86 *c) { + if (c->x86_vendor == X86_VENDOR_UNKNOWN) { + pr_info("MCE: unknown CPU type - not enabling MCE support.\n"); + return -EOPNOTSUPP; + } + /* This should be disabled by the BIOS, but isn't always */ if (c->x86_vendor == X86_VENDOR_AMD) { if (c->x86 == 15 && banks > 4) { @@ -1245,7 +1254,7 @@ static void mce_cpu_quirks(struct cpuinfo_x86 *c) * Various K7s with broken bank 0 around. Always disable * by default. */ - if (c->x86 == 6) + if (c->x86 == 6 && banks > 0) bank[0] = 0; } @@ -1269,11 +1278,20 @@ static void mce_cpu_quirks(struct cpuinfo_x86 *c) if ((c->x86 > 6 || (c->x86 == 6 && c->x86_model >= 0xe)) && monarch_timeout < 0) monarch_timeout = USEC_PER_SEC; + + /* + * There are also broken BIOSes on some Pentium M and + * earlier systems: + */ + if (c->x86 == 6 && c->x86_model <= 13 && mce_bootlog < 0) + mce_bootlog = 0; } if (monarch_timeout < 0) monarch_timeout = 0; if (mce_bootlog != 0) mce_panic_timeout = 30; + + return 0; } static void __cpuinit mce_ancient_init(struct cpuinfo_x86 *c) @@ -1282,8 +1300,7 @@ static void __cpuinit mce_ancient_init(struct cpuinfo_x86 *c) return; switch (c->x86_vendor) { case X86_VENDOR_INTEL: - if (mce_p5_enabled()) - intel_p5_mcheck_init(c); + intel_p5_mcheck_init(c); break; case X86_VENDOR_CENTAUR: winchip_mcheck_init(c); @@ -1318,7 +1335,7 @@ static void mce_init_timer(void) return; setup_timer(t, mcheck_timer, smp_processor_id()); t->expires = round_jiffies(jiffies + *n); - add_timer(t); + add_timer_on(t, smp_processor_id()); } /* @@ -1335,11 +1352,10 @@ void __cpuinit mcheck_init(struct cpuinfo_x86 *c) if (!mce_available(c)) return; - if (mce_cap_init() < 0) { + if (mce_cap_init() < 0 || mce_cpu_quirks(c) < 0) { mce_disabled = 1; return; } - mce_cpu_quirks(c); machine_check_vector = do_machine_check; @@ -1609,8 +1625,9 @@ static int mce_resume(struct sys_device *dev) static void mce_cpu_restart(void *data) { del_timer_sync(&__get_cpu_var(mce_timer)); - if (mce_available(¤t_cpu_data)) - mce_init(); + if (!mce_available(¤t_cpu_data)) + return; + mce_init(); mce_init_timer(); } @@ -1620,6 +1637,26 @@ static void mce_restart(void) on_each_cpu(mce_cpu_restart, NULL, 1); } +/* Toggle features for corrected errors */ +static void mce_disable_ce(void *all) +{ + if (!mce_available(¤t_cpu_data)) + return; + if (all) + del_timer_sync(&__get_cpu_var(mce_timer)); + cmci_clear(); +} + +static void mce_enable_ce(void *all) +{ + if (!mce_available(¤t_cpu_data)) + return; + cmci_reenable(); + cmci_recheck(); + if (all) + mce_init_timer(); +} + static struct sysdev_class mce_sysclass = { .suspend = mce_suspend, .shutdown = mce_shutdown, @@ -1659,26 +1696,70 @@ static ssize_t set_bank(struct sys_device *s, struct sysdev_attribute *attr, static ssize_t show_trigger(struct sys_device *s, struct sysdev_attribute *attr, char *buf) { - strcpy(buf, trigger); + strcpy(buf, mce_helper); strcat(buf, "\n"); - return strlen(trigger) + 1; + return strlen(mce_helper) + 1; } static ssize_t set_trigger(struct sys_device *s, struct sysdev_attribute *attr, const char *buf, size_t siz) { char *p; - int len; - strncpy(trigger, buf, sizeof(trigger)); - trigger[sizeof(trigger)-1] = 0; - len = strlen(trigger); - p = strchr(trigger, '\n'); + strncpy(mce_helper, buf, sizeof(mce_helper)); + mce_helper[sizeof(mce_helper)-1] = 0; + p = strchr(mce_helper, '\n'); - if (*p) + if (p) *p = 0; - return len; + return strlen(mce_helper) + !!p; +} + +static ssize_t set_ignore_ce(struct sys_device *s, + struct sysdev_attribute *attr, + const char *buf, size_t size) +{ + u64 new; + + if (strict_strtoull(buf, 0, &new) < 0) + return -EINVAL; + + if (mce_ignore_ce ^ !!new) { + if (new) { + /* disable ce features */ + on_each_cpu(mce_disable_ce, (void *)1, 1); + mce_ignore_ce = 1; + } else { + /* enable ce features */ + mce_ignore_ce = 0; + on_each_cpu(mce_enable_ce, (void *)1, 1); + } + } + return size; +} + +static ssize_t set_cmci_disabled(struct sys_device *s, + struct sysdev_attribute *attr, + const char *buf, size_t size) +{ + u64 new; + + if (strict_strtoull(buf, 0, &new) < 0) + return -EINVAL; + + if (mce_cmci_disabled ^ !!new) { + if (new) { + /* disable cmci */ + on_each_cpu(mce_disable_ce, NULL, 1); + mce_cmci_disabled = 1; + } else { + /* enable cmci */ + mce_cmci_disabled = 0; + on_each_cpu(mce_enable_ce, NULL, 1); + } + } + return size; } static ssize_t store_int_with_restart(struct sys_device *s, @@ -1693,6 +1774,7 @@ static ssize_t store_int_with_restart(struct sys_device *s, static SYSDEV_ATTR(trigger, 0644, show_trigger, set_trigger); static SYSDEV_INT_ATTR(tolerant, 0644, tolerant); static SYSDEV_INT_ATTR(monarch_timeout, 0644, monarch_timeout); +static SYSDEV_INT_ATTR(dont_log_ce, 0644, mce_dont_log_ce); static struct sysdev_ext_attribute attr_check_interval = { _SYSDEV_ATTR(check_interval, 0644, sysdev_show_int, @@ -1700,9 +1782,24 @@ static struct sysdev_ext_attribute attr_check_interval = { &check_interval }; +static struct sysdev_ext_attribute attr_ignore_ce = { + _SYSDEV_ATTR(ignore_ce, 0644, sysdev_show_int, set_ignore_ce), + &mce_ignore_ce +}; + +static struct sysdev_ext_attribute attr_cmci_disabled = { + _SYSDEV_ATTR(cmci_disabled, 0644, sysdev_show_int, set_cmci_disabled), + &mce_cmci_disabled +}; + static struct sysdev_attribute *mce_attrs[] = { - &attr_tolerant.attr, &attr_check_interval.attr, &attr_trigger, + &attr_tolerant.attr, + &attr_check_interval.attr, + &attr_trigger, &attr_monarch_timeout.attr, + &attr_dont_log_ce.attr, + &attr_ignore_ce.attr, + &attr_cmci_disabled.attr, NULL }; @@ -1712,7 +1809,7 @@ static cpumask_var_t mce_dev_initialized; static __cpuinit int mce_create_device(unsigned int cpu) { int err; - int i; + int i, j; if (!mce_available(&boot_cpu_data)) return -EIO; @@ -1730,9 +1827,9 @@ static __cpuinit int mce_create_device(unsigned int cpu) if (err) goto error; } - for (i = 0; i < banks; i++) { + for (j = 0; j < banks; j++) { err = sysdev_create_file(&per_cpu(mce_dev, cpu), - &bank_attrs[i]); + &bank_attrs[j]); if (err) goto error2; } @@ -1740,8 +1837,8 @@ static __cpuinit int mce_create_device(unsigned int cpu) return 0; error2: - while (--i >= 0) - sysdev_remove_file(&per_cpu(mce_dev, cpu), &bank_attrs[i]); + while (--j >= 0) + sysdev_remove_file(&per_cpu(mce_dev, cpu), &bank_attrs[j]); error: while (--i >= 0) sysdev_remove_file(&per_cpu(mce_dev, cpu), mce_attrs[i]); @@ -1883,7 +1980,7 @@ static __init int mce_init_device(void) if (!mce_available(&boot_cpu_data)) return -EIO; - alloc_cpumask_var(&mce_dev_initialized, GFP_KERNEL); + zalloc_cpumask_var(&mce_dev_initialized, GFP_KERNEL); err = mce_init_banks(); if (err) @@ -1915,7 +2012,7 @@ EXPORT_SYMBOL_GPL(nr_mce_banks); /* non-fatal.o */ /* This has to be run for each processor */ void mcheck_init(struct cpuinfo_x86 *c) { - if (mce_disabled == 1) + if (mce_disabled) return; switch (c->x86_vendor) { @@ -1945,10 +2042,9 @@ void mcheck_init(struct cpuinfo_x86 *c) static int __init mcheck_enable(char *str) { - mce_disabled = -1; + mce_p5_enabled = 1; return 1; } - __setup("mce", mcheck_enable); #endif /* CONFIG_X86_OLD_MCE */ diff --git a/arch/x86/kernel/cpu/mcheck/mce.h b/arch/x86/kernel/cpu/mcheck/mce.h deleted file mode 100644 index 84a552b..0000000 --- a/arch/x86/kernel/cpu/mcheck/mce.h +++ /dev/null @@ -1,38 +0,0 @@ -#include <linux/init.h> -#include <asm/mce.h> - -#ifdef CONFIG_X86_OLD_MCE -void amd_mcheck_init(struct cpuinfo_x86 *c); -void intel_p4_mcheck_init(struct cpuinfo_x86 *c); -void intel_p6_mcheck_init(struct cpuinfo_x86 *c); -#endif - -#ifdef CONFIG_X86_ANCIENT_MCE -void intel_p5_mcheck_init(struct cpuinfo_x86 *c); -void winchip_mcheck_init(struct cpuinfo_x86 *c); -extern int mce_p5_enable; -static inline int mce_p5_enabled(void) { return mce_p5_enable; } -static inline void enable_p5_mce(void) { mce_p5_enable = 1; } -#else -static inline void intel_p5_mcheck_init(struct cpuinfo_x86 *c) {} -static inline void winchip_mcheck_init(struct cpuinfo_x86 *c) {} -static inline int mce_p5_enabled(void) { return 0; } -static inline void enable_p5_mce(void) { } -#endif - -/* Call the installed machine check handler for this CPU setup. */ -extern void (*machine_check_vector)(struct pt_regs *, long error_code); - -#ifdef CONFIG_X86_OLD_MCE - -extern int nr_mce_banks; - -void intel_set_thermal_handler(void); - -#else - -static inline void intel_set_thermal_handler(void) { } - -#endif - -void intel_init_thermal(struct cpuinfo_x86 *c); diff --git a/arch/x86/kernel/cpu/mcheck/mce_amd_64.c b/arch/x86/kernel/cpu/mcheck/mce_amd.c index ddae216..ddae216 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_amd_64.c +++ b/arch/x86/kernel/cpu/mcheck/mce_amd.c diff --git a/arch/x86/kernel/cpu/mcheck/mce_intel.c b/arch/x86/kernel/cpu/mcheck/mce_intel.c index 2b011d2..e1acec0f 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_intel.c +++ b/arch/x86/kernel/cpu/mcheck/mce_intel.c @@ -1,74 +1,226 @@ /* - * Common code for Intel machine checks + * Intel specific MCE features. + * Copyright 2004 Zwane Mwaikambo <zwane@linuxpower.ca> + * Copyright (C) 2008, 2009 Intel Corporation + * Author: Andi Kleen */ -#include <linux/interrupt.h> -#include <linux/kernel.h> -#include <linux/types.h> -#include <linux/init.h> -#include <linux/smp.h> -#include <asm/therm_throt.h> -#include <asm/processor.h> -#include <asm/system.h> +#include <linux/init.h> +#include <linux/interrupt.h> +#include <linux/percpu.h> #include <asm/apic.h> +#include <asm/processor.h> #include <asm/msr.h> +#include <asm/mce.h> + +/* + * Support for Intel Correct Machine Check Interrupts. This allows + * the CPU to raise an interrupt when a corrected machine check happened. + * Normally we pick those up using a regular polling timer. + * Also supports reliable discovery of shared banks. + */ -#include "mce.h" +static DEFINE_PER_CPU(mce_banks_t, mce_banks_owned); -void intel_init_thermal(struct cpuinfo_x86 *c) +/* + * cmci_discover_lock protects against parallel discovery attempts + * which could race against each other. + */ +static DEFINE_SPINLOCK(cmci_discover_lock); + +#define CMCI_THRESHOLD 1 + +static int cmci_supported(int *banks) { - unsigned int cpu = smp_processor_id(); - int tm2 = 0; - u32 l, h; + u64 cap; - /* Thermal monitoring depends on ACPI and clock modulation*/ - if (!cpu_has(c, X86_FEATURE_ACPI) || !cpu_has(c, X86_FEATURE_ACC)) - return; + if (mce_cmci_disabled || mce_ignore_ce) + return 0; /* - * First check if its enabled already, in which case there might - * be some SMM goo which handles it, so we can't even put a handler - * since it might be delivered via SMI already: + * Vendor check is not strictly needed, but the initial + * initialization is vendor keyed and this + * makes sure none of the backdoors are entered otherwise. */ - rdmsr(MSR_IA32_MISC_ENABLE, l, h); - h = apic_read(APIC_LVTTHMR); - if ((l & MSR_IA32_MISC_ENABLE_TM1) && (h & APIC_DM_SMI)) { - printk(KERN_DEBUG - "CPU%d: Thermal monitoring handled by SMI\n", cpu); - return; + if (boot_cpu_data.x86_vendor != X86_VENDOR_INTEL) + return 0; + if (!cpu_has_apic || lapic_get_maxlvt() < 6) + return 0; + rdmsrl(MSR_IA32_MCG_CAP, cap); + *banks = min_t(unsigned, MAX_NR_BANKS, cap & 0xff); + return !!(cap & MCG_CMCI_P); +} + +/* + * The interrupt handler. This is called on every event. + * Just call the poller directly to log any events. + * This could in theory increase the threshold under high load, + * but doesn't for now. + */ +static void intel_threshold_interrupt(void) +{ + machine_check_poll(MCP_TIMESTAMP, &__get_cpu_var(mce_banks_owned)); + mce_notify_irq(); +} + +static void print_update(char *type, int *hdr, int num) +{ + if (*hdr == 0) + printk(KERN_INFO "CPU %d MCA banks", smp_processor_id()); + *hdr = 1; + printk(KERN_CONT " %s:%d", type, num); +} + +/* + * Enable CMCI (Corrected Machine Check Interrupt) for available MCE banks + * on this CPU. Use the algorithm recommended in the SDM to discover shared + * banks. + */ +static void cmci_discover(int banks, int boot) +{ + unsigned long *owned = (void *)&__get_cpu_var(mce_banks_owned); + unsigned long flags; + int hdr = 0; + int i; + + spin_lock_irqsave(&cmci_discover_lock, flags); + for (i = 0; i < banks; i++) { + u64 val; + + if (test_bit(i, owned)) + continue; + + rdmsrl(MSR_IA32_MC0_CTL2 + i, val); + + /* Already owned by someone else? */ + if (val & CMCI_EN) { + if (test_and_clear_bit(i, owned) || boot) + print_update("SHD", &hdr, i); + __clear_bit(i, __get_cpu_var(mce_poll_banks)); + continue; + } + + val |= CMCI_EN | CMCI_THRESHOLD; + wrmsrl(MSR_IA32_MC0_CTL2 + i, val); + rdmsrl(MSR_IA32_MC0_CTL2 + i, val); + + /* Did the enable bit stick? -- the bank supports CMCI */ + if (val & CMCI_EN) { + if (!test_and_set_bit(i, owned) || boot) + print_update("CMCI", &hdr, i); + __clear_bit(i, __get_cpu_var(mce_poll_banks)); + } else { + WARN_ON(!test_bit(i, __get_cpu_var(mce_poll_banks))); + } } + spin_unlock_irqrestore(&cmci_discover_lock, flags); + if (hdr) + printk(KERN_CONT "\n"); +} + +/* + * Just in case we missed an event during initialization check + * all the CMCI owned banks. + */ +void cmci_recheck(void) +{ + unsigned long flags; + int banks; + + if (!mce_available(¤t_cpu_data) || !cmci_supported(&banks)) + return; + local_irq_save(flags); + machine_check_poll(MCP_TIMESTAMP, &__get_cpu_var(mce_banks_owned)); + local_irq_restore(flags); +} - if (cpu_has(c, X86_FEATURE_TM2) && (l & MSR_IA32_MISC_ENABLE_TM2)) - tm2 = 1; +/* + * Disable CMCI on this CPU for all banks it owns when it goes down. + * This allows other CPUs to claim the banks on rediscovery. + */ +void cmci_clear(void) +{ + unsigned long flags; + int i; + int banks; + u64 val; - /* Check whether a vector already exists */ - if (h & APIC_VECTOR_MASK) { - printk(KERN_DEBUG - "CPU%d: Thermal LVT vector (%#x) already installed\n", - cpu, (h & APIC_VECTOR_MASK)); + if (!cmci_supported(&banks)) return; + spin_lock_irqsave(&cmci_discover_lock, flags); + for (i = 0; i < banks; i++) { + if (!test_bit(i, __get_cpu_var(mce_banks_owned))) + continue; + /* Disable CMCI */ + rdmsrl(MSR_IA32_MC0_CTL2 + i, val); + val &= ~(CMCI_EN|CMCI_THRESHOLD_MASK); + wrmsrl(MSR_IA32_MC0_CTL2 + i, val); + __clear_bit(i, __get_cpu_var(mce_banks_owned)); } + spin_unlock_irqrestore(&cmci_discover_lock, flags); +} + +/* + * After a CPU went down cycle through all the others and rediscover + * Must run in process context. + */ +void cmci_rediscover(int dying) +{ + int banks; + int cpu; + cpumask_var_t old; + + if (!cmci_supported(&banks)) + return; + if (!alloc_cpumask_var(&old, GFP_KERNEL)) + return; + cpumask_copy(old, ¤t->cpus_allowed); - /* We'll mask the thermal vector in the lapic till we're ready: */ - h = THERMAL_APIC_VECTOR | APIC_DM_FIXED | APIC_LVT_MASKED; - apic_write(APIC_LVTTHMR, h); + for_each_online_cpu(cpu) { + if (cpu == dying) + continue; + if (set_cpus_allowed_ptr(current, cpumask_of(cpu))) + continue; + /* Recheck banks in case CPUs don't all have the same */ + if (cmci_supported(&banks)) + cmci_discover(banks, 0); + } - rdmsr(MSR_IA32_THERM_INTERRUPT, l, h); - wrmsr(MSR_IA32_THERM_INTERRUPT, - l | (THERM_INT_LOW_ENABLE | THERM_INT_HIGH_ENABLE), h); + set_cpus_allowed_ptr(current, old); + free_cpumask_var(old); +} - intel_set_thermal_handler(); +/* + * Reenable CMCI on this CPU in case a CPU down failed. + */ +void cmci_reenable(void) +{ + int banks; + if (cmci_supported(&banks)) + cmci_discover(banks, 0); +} - rdmsr(MSR_IA32_MISC_ENABLE, l, h); - wrmsr(MSR_IA32_MISC_ENABLE, l | MSR_IA32_MISC_ENABLE_TM1, h); +static void intel_init_cmci(void) +{ + int banks; - /* Unmask the thermal vector: */ - l = apic_read(APIC_LVTTHMR); - apic_write(APIC_LVTTHMR, l & ~APIC_LVT_MASKED); + if (!cmci_supported(&banks)) + return; - printk(KERN_INFO "CPU%d: Thermal monitoring enabled (%s)\n", - cpu, tm2 ? "TM2" : "TM1"); + mce_threshold_vector = intel_threshold_interrupt; + cmci_discover(banks, 1); + /* + * For CPU #0 this runs with still disabled APIC, but that's + * ok because only the vector is set up. We still do another + * check for the banks later for CPU #0 just to make sure + * to not miss any events. + */ + apic_write(APIC_LVTCMCI, THRESHOLD_APIC_VECTOR|APIC_DM_FIXED); + cmci_recheck(); +} - /* enable thermal throttle processing */ - atomic_set(&therm_throt_en, 1); +void mce_intel_feature_init(struct cpuinfo_x86 *c) +{ + intel_init_thermal(c); + intel_init_cmci(); } diff --git a/arch/x86/kernel/cpu/mcheck/mce_intel_64.c b/arch/x86/kernel/cpu/mcheck/mce_intel_64.c deleted file mode 100644 index f2ef695..0000000 --- a/arch/x86/kernel/cpu/mcheck/mce_intel_64.c +++ /dev/null @@ -1,248 +0,0 @@ -/* - * Intel specific MCE features. - * Copyright 2004 Zwane Mwaikambo <zwane@linuxpower.ca> - * Copyright (C) 2008, 2009 Intel Corporation - * Author: Andi Kleen - */ - -#include <linux/init.h> -#include <linux/interrupt.h> -#include <linux/percpu.h> -#include <asm/processor.h> -#include <asm/apic.h> -#include <asm/msr.h> -#include <asm/mce.h> -#include <asm/hw_irq.h> -#include <asm/idle.h> -#include <asm/therm_throt.h> - -#include "mce.h" - -asmlinkage void smp_thermal_interrupt(void) -{ - __u64 msr_val; - - ack_APIC_irq(); - - exit_idle(); - irq_enter(); - - rdmsrl(MSR_IA32_THERM_STATUS, msr_val); - if (therm_throt_process(msr_val & THERM_STATUS_PROCHOT)) - mce_log_therm_throt_event(msr_val); - - inc_irq_stat(irq_thermal_count); - irq_exit(); -} - -/* - * Support for Intel Correct Machine Check Interrupts. This allows - * the CPU to raise an interrupt when a corrected machine check happened. - * Normally we pick those up using a regular polling timer. - * Also supports reliable discovery of shared banks. - */ - -static DEFINE_PER_CPU(mce_banks_t, mce_banks_owned); - -/* - * cmci_discover_lock protects against parallel discovery attempts - * which could race against each other. - */ -static DEFINE_SPINLOCK(cmci_discover_lock); - -#define CMCI_THRESHOLD 1 - -static int cmci_supported(int *banks) -{ - u64 cap; - - if (mce_cmci_disabled || mce_ignore_ce) - return 0; - - /* - * Vendor check is not strictly needed, but the initial - * initialization is vendor keyed and this - * makes sure none of the backdoors are entered otherwise. - */ - if (boot_cpu_data.x86_vendor != X86_VENDOR_INTEL) - return 0; - if (!cpu_has_apic || lapic_get_maxlvt() < 6) - return 0; - rdmsrl(MSR_IA32_MCG_CAP, cap); - *banks = min_t(unsigned, MAX_NR_BANKS, cap & 0xff); - return !!(cap & MCG_CMCI_P); -} - -/* - * The interrupt handler. This is called on every event. - * Just call the poller directly to log any events. - * This could in theory increase the threshold under high load, - * but doesn't for now. - */ -static void intel_threshold_interrupt(void) -{ - machine_check_poll(MCP_TIMESTAMP, &__get_cpu_var(mce_banks_owned)); - mce_notify_irq(); -} - -static void print_update(char *type, int *hdr, int num) -{ - if (*hdr == 0) - printk(KERN_INFO "CPU %d MCA banks", smp_processor_id()); - *hdr = 1; - printk(KERN_CONT " %s:%d", type, num); -} - -/* - * Enable CMCI (Corrected Machine Check Interrupt) for available MCE banks - * on this CPU. Use the algorithm recommended in the SDM to discover shared - * banks. - */ -static void cmci_discover(int banks, int boot) -{ - unsigned long *owned = (void *)&__get_cpu_var(mce_banks_owned); - unsigned long flags; - int hdr = 0; - int i; - - spin_lock_irqsave(&cmci_discover_lock, flags); - for (i = 0; i < banks; i++) { - u64 val; - - if (test_bit(i, owned)) - continue; - - rdmsrl(MSR_IA32_MC0_CTL2 + i, val); - - /* Already owned by someone else? */ - if (val & CMCI_EN) { - if (test_and_clear_bit(i, owned) || boot) - print_update("SHD", &hdr, i); - __clear_bit(i, __get_cpu_var(mce_poll_banks)); - continue; - } - - val |= CMCI_EN | CMCI_THRESHOLD; - wrmsrl(MSR_IA32_MC0_CTL2 + i, val); - rdmsrl(MSR_IA32_MC0_CTL2 + i, val); - - /* Did the enable bit stick? -- the bank supports CMCI */ - if (val & CMCI_EN) { - if (!test_and_set_bit(i, owned) || boot) - print_update("CMCI", &hdr, i); - __clear_bit(i, __get_cpu_var(mce_poll_banks)); - } else { - WARN_ON(!test_bit(i, __get_cpu_var(mce_poll_banks))); - } - } - spin_unlock_irqrestore(&cmci_discover_lock, flags); - if (hdr) - printk(KERN_CONT "\n"); -} - -/* - * Just in case we missed an event during initialization check - * all the CMCI owned banks. - */ -void cmci_recheck(void) -{ - unsigned long flags; - int banks; - - if (!mce_available(¤t_cpu_data) || !cmci_supported(&banks)) - return; - local_irq_save(flags); - machine_check_poll(MCP_TIMESTAMP, &__get_cpu_var(mce_banks_owned)); - local_irq_restore(flags); -} - -/* - * Disable CMCI on this CPU for all banks it owns when it goes down. - * This allows other CPUs to claim the banks on rediscovery. - */ -void cmci_clear(void) -{ - unsigned long flags; - int i; - int banks; - u64 val; - - if (!cmci_supported(&banks)) - return; - spin_lock_irqsave(&cmci_discover_lock, flags); - for (i = 0; i < banks; i++) { - if (!test_bit(i, __get_cpu_var(mce_banks_owned))) - continue; - /* Disable CMCI */ - rdmsrl(MSR_IA32_MC0_CTL2 + i, val); - val &= ~(CMCI_EN|CMCI_THRESHOLD_MASK); - wrmsrl(MSR_IA32_MC0_CTL2 + i, val); - __clear_bit(i, __get_cpu_var(mce_banks_owned)); - } - spin_unlock_irqrestore(&cmci_discover_lock, flags); -} - -/* - * After a CPU went down cycle through all the others and rediscover - * Must run in process context. - */ -void cmci_rediscover(int dying) -{ - int banks; - int cpu; - cpumask_var_t old; - - if (!cmci_supported(&banks)) - return; - if (!alloc_cpumask_var(&old, GFP_KERNEL)) - return; - cpumask_copy(old, ¤t->cpus_allowed); - - for_each_online_cpu(cpu) { - if (cpu == dying) - continue; - if (set_cpus_allowed_ptr(current, cpumask_of(cpu))) - continue; - /* Recheck banks in case CPUs don't all have the same */ - if (cmci_supported(&banks)) - cmci_discover(banks, 0); - } - - set_cpus_allowed_ptr(current, old); - free_cpumask_var(old); -} - -/* - * Reenable CMCI on this CPU in case a CPU down failed. - */ -void cmci_reenable(void) -{ - int banks; - if (cmci_supported(&banks)) - cmci_discover(banks, 0); -} - -static void intel_init_cmci(void) -{ - int banks; - - if (!cmci_supported(&banks)) - return; - - mce_threshold_vector = intel_threshold_interrupt; - cmci_discover(banks, 1); - /* - * For CPU #0 this runs with still disabled APIC, but that's - * ok because only the vector is set up. We still do another - * check for the banks later for CPU #0 just to make sure - * to not miss any events. - */ - apic_write(APIC_LVTCMCI, THRESHOLD_APIC_VECTOR|APIC_DM_FIXED); - cmci_recheck(); -} - -void mce_intel_feature_init(struct cpuinfo_x86 *c) -{ - intel_init_thermal(c); - intel_init_cmci(); -} diff --git a/arch/x86/kernel/cpu/mcheck/non-fatal.c b/arch/x86/kernel/cpu/mcheck/non-fatal.c index 70b7104..f5f2d6f 100644 --- a/arch/x86/kernel/cpu/mcheck/non-fatal.c +++ b/arch/x86/kernel/cpu/mcheck/non-fatal.c @@ -17,10 +17,9 @@ #include <asm/processor.h> #include <asm/system.h> +#include <asm/mce.h> #include <asm/msr.h> -#include "mce.h" - static int firstbank; #define MCE_RATE (15*HZ) /* timer rate is 15s */ diff --git a/arch/x86/kernel/cpu/mcheck/p4.c b/arch/x86/kernel/cpu/mcheck/p4.c index 82cee10..4482aea 100644 --- a/arch/x86/kernel/cpu/mcheck/p4.c +++ b/arch/x86/kernel/cpu/mcheck/p4.c @@ -1,21 +1,15 @@ /* * P4 specific Machine Check Exception Reporting */ - -#include <linux/interrupt.h> #include <linux/kernel.h> #include <linux/types.h> #include <linux/init.h> #include <linux/smp.h> -#include <asm/therm_throt.h> #include <asm/processor.h> -#include <asm/system.h> -#include <asm/apic.h> +#include <asm/mce.h> #include <asm/msr.h> -#include "mce.h" - /* as supported by the P4/Xeon family */ struct intel_mce_extended_msrs { u32 eax; @@ -33,46 +27,6 @@ struct intel_mce_extended_msrs { static int mce_num_extended_msrs; - -#ifdef CONFIG_X86_MCE_P4THERMAL - -static void unexpected_thermal_interrupt(struct pt_regs *regs) -{ - printk(KERN_ERR "CPU%d: Unexpected LVT TMR interrupt!\n", - smp_processor_id()); - add_taint(TAINT_MACHINE_CHECK); -} - -/* P4/Xeon Thermal transition interrupt handler: */ -static void intel_thermal_interrupt(struct pt_regs *regs) -{ - __u64 msr_val; - - ack_APIC_irq(); - - rdmsrl(MSR_IA32_THERM_STATUS, msr_val); - therm_throt_process(msr_val & THERM_STATUS_PROCHOT); -} - -/* Thermal interrupt handler for this CPU setup: */ -static void (*vendor_thermal_interrupt)(struct pt_regs *regs) = - unexpected_thermal_interrupt; - -void smp_thermal_interrupt(struct pt_regs *regs) -{ - irq_enter(); - vendor_thermal_interrupt(regs); - __get_cpu_var(irq_stat).irq_thermal_count++; - irq_exit(); -} - -void intel_set_thermal_handler(void) -{ - vendor_thermal_interrupt = intel_thermal_interrupt; -} - -#endif /* CONFIG_X86_MCE_P4THERMAL */ - /* P4/Xeon Extended MCE MSR retrieval, return 0 if unsupported */ static void intel_get_extended_msrs(struct intel_mce_extended_msrs *r) { diff --git a/arch/x86/kernel/cpu/mcheck/p5.c b/arch/x86/kernel/cpu/mcheck/p5.c index 015f481..5c0e653 100644 --- a/arch/x86/kernel/cpu/mcheck/p5.c +++ b/arch/x86/kernel/cpu/mcheck/p5.c @@ -10,12 +10,11 @@ #include <asm/processor.h> #include <asm/system.h> +#include <asm/mce.h> #include <asm/msr.h> -#include "mce.h" - /* By default disabled */ -int mce_p5_enable; +int mce_p5_enabled __read_mostly; /* Machine check handler for Pentium class Intel CPUs: */ static void pentium_machine_check(struct pt_regs *regs, long error_code) @@ -43,15 +42,13 @@ void intel_p5_mcheck_init(struct cpuinfo_x86 *c) { u32 l, h; - /* Check for MCE support: */ - if (!cpu_has(c, X86_FEATURE_MCE)) + /* Default P5 to off as its often misconnected: */ + if (!mce_p5_enabled) return; -#ifdef CONFIG_X86_OLD_MCE - /* Default P5 to off as its often misconnected: */ - if (mce_disabled != -1) + /* Check for MCE support: */ + if (!cpu_has(c, X86_FEATURE_MCE)) return; -#endif machine_check_vector = pentium_machine_check; /* Make sure the vector pointer is visible before we enable MCEs: */ diff --git a/arch/x86/kernel/cpu/mcheck/p6.c b/arch/x86/kernel/cpu/mcheck/p6.c index 43c24e6..01e4f81 100644 --- a/arch/x86/kernel/cpu/mcheck/p6.c +++ b/arch/x86/kernel/cpu/mcheck/p6.c @@ -10,10 +10,9 @@ #include <asm/processor.h> #include <asm/system.h> +#include <asm/mce.h> #include <asm/msr.h> -#include "mce.h" - /* Machine Check Handler For PII/PIII */ static void intel_machine_check(struct pt_regs *regs, long error_code) { diff --git a/arch/x86/kernel/cpu/mcheck/therm_throt.c b/arch/x86/kernel/cpu/mcheck/therm_throt.c index 7b1ae2e..5957a93 100644 --- a/arch/x86/kernel/cpu/mcheck/therm_throt.c +++ b/arch/x86/kernel/cpu/mcheck/therm_throt.c @@ -13,21 +13,32 @@ * Credits: Adapted from Zwane Mwaikambo's original code in mce_intel.c. * Inspired by Ross Biro's and Al Borchers' counter code. */ +#include <linux/interrupt.h> #include <linux/notifier.h> #include <linux/jiffies.h> +#include <linux/kernel.h> #include <linux/percpu.h> #include <linux/sysdev.h> +#include <linux/types.h> +#include <linux/init.h> +#include <linux/smp.h> #include <linux/cpu.h> -#include <asm/therm_throt.h> +#include <asm/processor.h> +#include <asm/system.h> +#include <asm/apic.h> +#include <asm/idle.h> +#include <asm/mce.h> +#include <asm/msr.h> /* How long to wait between reporting thermal events */ #define CHECK_INTERVAL (300 * HZ) static DEFINE_PER_CPU(__u64, next_check) = INITIAL_JIFFIES; static DEFINE_PER_CPU(unsigned long, thermal_throttle_count); +static DEFINE_PER_CPU(bool, thermal_throttle_active); -atomic_t therm_throt_en = ATOMIC_INIT(0); +static atomic_t therm_throt_en = ATOMIC_INIT(0); #ifdef CONFIG_SYSFS #define define_therm_throt_sysdev_one_ro(_name) \ @@ -82,31 +93,37 @@ static struct attribute_group thermal_throttle_attr_group = { * 1 : Event should be logged further, and a message has been * printed to the syslog. */ -int therm_throt_process(int curr) +static int therm_throt_process(int curr) { unsigned int cpu = smp_processor_id(); __u64 tmp_jiffs = get_jiffies_64(); + bool was_throttled = __get_cpu_var(thermal_throttle_active); + bool is_throttled = __get_cpu_var(thermal_throttle_active) = curr; - if (curr) + if (is_throttled) __get_cpu_var(thermal_throttle_count)++; - if (time_before64(tmp_jiffs, __get_cpu_var(next_check))) + if (!(was_throttled ^ is_throttled) && + time_before64(tmp_jiffs, __get_cpu_var(next_check))) return 0; __get_cpu_var(next_check) = tmp_jiffs + CHECK_INTERVAL; /* if we just entered the thermal event */ - if (curr) { + if (is_throttled) { printk(KERN_CRIT "CPU%d: Temperature above threshold, " - "cpu clock throttled (total events = %lu)\n", cpu, - __get_cpu_var(thermal_throttle_count)); + "cpu clock throttled (total events = %lu)\n", + cpu, __get_cpu_var(thermal_throttle_count)); add_taint(TAINT_MACHINE_CHECK); - } else { - printk(KERN_CRIT "CPU%d: Temperature/speed normal\n", cpu); + return 1; + } + if (was_throttled) { + printk(KERN_INFO "CPU%d: Temperature/speed normal\n", cpu); + return 1; } - return 1; + return 0; } #ifdef CONFIG_SYSFS @@ -186,6 +203,94 @@ static __init int thermal_throttle_init_device(void) return 0; } - device_initcall(thermal_throttle_init_device); + #endif /* CONFIG_SYSFS */ + +/* Thermal transition interrupt handler */ +static void intel_thermal_interrupt(void) +{ + __u64 msr_val; + + rdmsrl(MSR_IA32_THERM_STATUS, msr_val); + if (therm_throt_process(msr_val & THERM_STATUS_PROCHOT)) + mce_log_therm_throt_event(msr_val); +} + +static void unexpected_thermal_interrupt(void) +{ + printk(KERN_ERR "CPU%d: Unexpected LVT TMR interrupt!\n", + smp_processor_id()); + add_taint(TAINT_MACHINE_CHECK); +} + +static void (*smp_thermal_vector)(void) = unexpected_thermal_interrupt; + +asmlinkage void smp_thermal_interrupt(struct pt_regs *regs) +{ + exit_idle(); + irq_enter(); + inc_irq_stat(irq_thermal_count); + smp_thermal_vector(); + irq_exit(); + /* Ack only at the end to avoid potential reentry */ + ack_APIC_irq(); +} + +void intel_init_thermal(struct cpuinfo_x86 *c) +{ + unsigned int cpu = smp_processor_id(); + int tm2 = 0; + u32 l, h; + + /* Thermal monitoring depends on ACPI and clock modulation*/ + if (!cpu_has(c, X86_FEATURE_ACPI) || !cpu_has(c, X86_FEATURE_ACC)) + return; + + /* + * First check if its enabled already, in which case there might + * be some SMM goo which handles it, so we can't even put a handler + * since it might be delivered via SMI already: + */ + rdmsr(MSR_IA32_MISC_ENABLE, l, h); + h = apic_read(APIC_LVTTHMR); + if ((l & MSR_IA32_MISC_ENABLE_TM1) && (h & APIC_DM_SMI)) { + printk(KERN_DEBUG + "CPU%d: Thermal monitoring handled by SMI\n", cpu); + return; + } + + if (cpu_has(c, X86_FEATURE_TM2) && (l & MSR_IA32_MISC_ENABLE_TM2)) + tm2 = 1; + + /* Check whether a vector already exists */ + if (h & APIC_VECTOR_MASK) { + printk(KERN_DEBUG + "CPU%d: Thermal LVT vector (%#x) already installed\n", + cpu, (h & APIC_VECTOR_MASK)); + return; + } + + /* We'll mask the thermal vector in the lapic till we're ready: */ + h = THERMAL_APIC_VECTOR | APIC_DM_FIXED | APIC_LVT_MASKED; + apic_write(APIC_LVTTHMR, h); + + rdmsr(MSR_IA32_THERM_INTERRUPT, l, h); + wrmsr(MSR_IA32_THERM_INTERRUPT, + l | (THERM_INT_LOW_ENABLE | THERM_INT_HIGH_ENABLE), h); + + smp_thermal_vector = intel_thermal_interrupt; + + rdmsr(MSR_IA32_MISC_ENABLE, l, h); + wrmsr(MSR_IA32_MISC_ENABLE, l | MSR_IA32_MISC_ENABLE_TM1, h); + + /* Unmask the thermal vector: */ + l = apic_read(APIC_LVTTHMR); + apic_write(APIC_LVTTHMR, l & ~APIC_LVT_MASKED); + + printk(KERN_INFO "CPU%d: Thermal monitoring enabled (%s)\n", + cpu, tm2 ? "TM2" : "TM1"); + + /* enable thermal throttle processing */ + atomic_set(&therm_throt_en, 1); +} diff --git a/arch/x86/kernel/cpu/mcheck/winchip.c b/arch/x86/kernel/cpu/mcheck/winchip.c index 81b0248..54060f5 100644 --- a/arch/x86/kernel/cpu/mcheck/winchip.c +++ b/arch/x86/kernel/cpu/mcheck/winchip.c @@ -9,10 +9,9 @@ #include <asm/processor.h> #include <asm/system.h> +#include <asm/mce.h> #include <asm/msr.h> -#include "mce.h" - /* Machine check handler for WinChip C6: */ static void winchip_machine_check(struct pt_regs *regs, long error_code) { diff --git a/arch/x86/kernel/cpu/perf_counter.c b/arch/x86/kernel/cpu/perf_counter.c index 275bc14..900332b 100644 --- a/arch/x86/kernel/cpu/perf_counter.c +++ b/arch/x86/kernel/cpu/perf_counter.c @@ -19,6 +19,7 @@ #include <linux/kdebug.h> #include <linux/sched.h> #include <linux/uaccess.h> +#include <linux/highmem.h> #include <asm/apic.h> #include <asm/stacktrace.h> @@ -54,6 +55,7 @@ struct x86_pmu { int num_counters_fixed; int counter_bits; u64 counter_mask; + int apic; u64 max_period; u64 intel_ctrl; }; @@ -65,6 +67,52 @@ static DEFINE_PER_CPU(struct cpu_hw_counters, cpu_hw_counters) = { }; /* + * Not sure about some of these + */ +static const u64 p6_perfmon_event_map[] = +{ + [PERF_COUNT_HW_CPU_CYCLES] = 0x0079, + [PERF_COUNT_HW_INSTRUCTIONS] = 0x00c0, + [PERF_COUNT_HW_CACHE_REFERENCES] = 0x0f2e, + [PERF_COUNT_HW_CACHE_MISSES] = 0x012e, + [PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = 0x00c4, + [PERF_COUNT_HW_BRANCH_MISSES] = 0x00c5, + [PERF_COUNT_HW_BUS_CYCLES] = 0x0062, +}; + +static u64 p6_pmu_event_map(int event) +{ + return p6_perfmon_event_map[event]; +} + +/* + * Counter setting that is specified not to count anything. + * We use this to effectively disable a counter. + * + * L2_RQSTS with 0 MESI unit mask. + */ +#define P6_NOP_COUNTER 0x0000002EULL + +static u64 p6_pmu_raw_event(u64 event) +{ +#define P6_EVNTSEL_EVENT_MASK 0x000000FFULL +#define P6_EVNTSEL_UNIT_MASK 0x0000FF00ULL +#define P6_EVNTSEL_EDGE_MASK 0x00040000ULL +#define P6_EVNTSEL_INV_MASK 0x00800000ULL +#define P6_EVNTSEL_COUNTER_MASK 0xFF000000ULL + +#define P6_EVNTSEL_MASK \ + (P6_EVNTSEL_EVENT_MASK | \ + P6_EVNTSEL_UNIT_MASK | \ + P6_EVNTSEL_EDGE_MASK | \ + P6_EVNTSEL_INV_MASK | \ + P6_EVNTSEL_COUNTER_MASK) + + return event & P6_EVNTSEL_MASK; +} + + +/* * Intel PerfMon v3. Used on Core2 and later. */ static const u64 intel_perfmon_event_map[] = @@ -389,23 +437,23 @@ static u64 intel_pmu_raw_event(u64 event) return event & CORE_EVNTSEL_MASK; } -static const u64 amd_0f_hw_cache_event_ids +static const u64 amd_hw_cache_event_ids [PERF_COUNT_HW_CACHE_MAX] [PERF_COUNT_HW_CACHE_OP_MAX] [PERF_COUNT_HW_CACHE_RESULT_MAX] = { [ C(L1D) ] = { [ C(OP_READ) ] = { - [ C(RESULT_ACCESS) ] = 0, - [ C(RESULT_MISS) ] = 0, + [ C(RESULT_ACCESS) ] = 0x0040, /* Data Cache Accesses */ + [ C(RESULT_MISS) ] = 0x0041, /* Data Cache Misses */ }, [ C(OP_WRITE) ] = { - [ C(RESULT_ACCESS) ] = 0, + [ C(RESULT_ACCESS) ] = 0x0142, /* Data Cache Refills :system */ [ C(RESULT_MISS) ] = 0, }, [ C(OP_PREFETCH) ] = { - [ C(RESULT_ACCESS) ] = 0, - [ C(RESULT_MISS) ] = 0, + [ C(RESULT_ACCESS) ] = 0x0267, /* Data Prefetcher :attempts */ + [ C(RESULT_MISS) ] = 0x0167, /* Data Prefetcher :cancelled */ }, }, [ C(L1I ) ] = { @@ -418,17 +466,17 @@ static const u64 amd_0f_hw_cache_event_ids [ C(RESULT_MISS) ] = -1, }, [ C(OP_PREFETCH) ] = { - [ C(RESULT_ACCESS) ] = 0, + [ C(RESULT_ACCESS) ] = 0x014B, /* Prefetch Instructions :Load */ [ C(RESULT_MISS) ] = 0, }, }, [ C(LL ) ] = { [ C(OP_READ) ] = { - [ C(RESULT_ACCESS) ] = 0, - [ C(RESULT_MISS) ] = 0, + [ C(RESULT_ACCESS) ] = 0x037D, /* Requests to L2 Cache :IC+DC */ + [ C(RESULT_MISS) ] = 0x037E, /* L2 Cache Misses : IC+DC */ }, [ C(OP_WRITE) ] = { - [ C(RESULT_ACCESS) ] = 0, + [ C(RESULT_ACCESS) ] = 0x017F, /* L2 Fill/Writeback */ [ C(RESULT_MISS) ] = 0, }, [ C(OP_PREFETCH) ] = { @@ -438,8 +486,8 @@ static const u64 amd_0f_hw_cache_event_ids }, [ C(DTLB) ] = { [ C(OP_READ) ] = { - [ C(RESULT_ACCESS) ] = 0, - [ C(RESULT_MISS) ] = 0, + [ C(RESULT_ACCESS) ] = 0x0040, /* Data Cache Accesses */ + [ C(RESULT_MISS) ] = 0x0046, /* L1 DTLB and L2 DLTB Miss */ }, [ C(OP_WRITE) ] = { [ C(RESULT_ACCESS) ] = 0, @@ -566,6 +614,7 @@ static DEFINE_MUTEX(pmc_reserve_mutex); static bool reserve_pmc_hardware(void) { +#ifdef CONFIG_X86_LOCAL_APIC int i; if (nmi_watchdog == NMI_LOCAL_APIC) @@ -580,9 +629,11 @@ static bool reserve_pmc_hardware(void) if (!reserve_evntsel_nmi(x86_pmu.eventsel + i)) goto eventsel_fail; } +#endif return true; +#ifdef CONFIG_X86_LOCAL_APIC eventsel_fail: for (i--; i >= 0; i--) release_evntsel_nmi(x86_pmu.eventsel + i); @@ -597,10 +648,12 @@ perfctr_fail: enable_lapic_nmi_watchdog(); return false; +#endif } static void release_pmc_hardware(void) { +#ifdef CONFIG_X86_LOCAL_APIC int i; for (i = 0; i < x86_pmu.num_counters; i++) { @@ -610,6 +663,7 @@ static void release_pmc_hardware(void) if (nmi_watchdog == NMI_LOCAL_APIC) enable_lapic_nmi_watchdog(); +#endif } static void hw_perf_counter_destroy(struct perf_counter *counter) @@ -665,6 +719,7 @@ static int __hw_perf_counter_init(struct perf_counter *counter) { struct perf_counter_attr *attr = &counter->attr; struct hw_perf_counter *hwc = &counter->hw; + u64 config; int err; if (!x86_pmu_initialized()) @@ -700,6 +755,15 @@ static int __hw_perf_counter_init(struct perf_counter *counter) hwc->sample_period = x86_pmu.max_period; hwc->last_period = hwc->sample_period; atomic64_set(&hwc->period_left, hwc->sample_period); + } else { + /* + * If we have a PMU initialized but no APIC + * interrupts, we cannot sample hardware + * counters (user-space has to fall back and + * sample via a hrtimer based software counter): + */ + if (!x86_pmu.apic) + return -EOPNOTSUPP; } counter->destroy = hw_perf_counter_destroy; @@ -717,14 +781,40 @@ static int __hw_perf_counter_init(struct perf_counter *counter) if (attr->config >= x86_pmu.max_events) return -EINVAL; + /* * The generic map: */ - hwc->config |= x86_pmu.event_map(attr->config); + config = x86_pmu.event_map(attr->config); + + if (config == 0) + return -ENOENT; + + if (config == -1LL) + return -EINVAL; + + hwc->config |= config; return 0; } +static void p6_pmu_disable_all(void) +{ + struct cpu_hw_counters *cpuc = &__get_cpu_var(cpu_hw_counters); + u64 val; + + if (!cpuc->enabled) + return; + + cpuc->enabled = 0; + barrier(); + + /* p6 only has one enable register */ + rdmsrl(MSR_P6_EVNTSEL0, val); + val &= ~ARCH_PERFMON_EVENTSEL0_ENABLE; + wrmsrl(MSR_P6_EVNTSEL0, val); +} + static void intel_pmu_disable_all(void) { wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, 0); @@ -766,6 +856,23 @@ void hw_perf_disable(void) return x86_pmu.disable_all(); } +static void p6_pmu_enable_all(void) +{ + struct cpu_hw_counters *cpuc = &__get_cpu_var(cpu_hw_counters); + unsigned long val; + + if (cpuc->enabled) + return; + + cpuc->enabled = 1; + barrier(); + + /* p6 only has one enable register */ + rdmsrl(MSR_P6_EVNTSEL0, val); + val |= ARCH_PERFMON_EVENTSEL0_ENABLE; + wrmsrl(MSR_P6_EVNTSEL0, val); +} + static void intel_pmu_enable_all(void) { wrmsrl(MSR_CORE_PERF_GLOBAL_CTRL, x86_pmu.intel_ctrl); @@ -783,13 +890,13 @@ static void amd_pmu_enable_all(void) barrier(); for (idx = 0; idx < x86_pmu.num_counters; idx++) { + struct perf_counter *counter = cpuc->counters[idx]; u64 val; if (!test_bit(idx, cpuc->active_mask)) continue; - rdmsrl(MSR_K7_EVNTSEL0 + idx, val); - if (val & ARCH_PERFMON_EVENTSEL0_ENABLE) - continue; + + val = counter->hw.config; val |= ARCH_PERFMON_EVENTSEL0_ENABLE; wrmsrl(MSR_K7_EVNTSEL0 + idx, val); } @@ -818,16 +925,13 @@ static inline void intel_pmu_ack_status(u64 ack) static inline void x86_pmu_enable_counter(struct hw_perf_counter *hwc, int idx) { - int err; - err = checking_wrmsrl(hwc->config_base + idx, + (void)checking_wrmsrl(hwc->config_base + idx, hwc->config | ARCH_PERFMON_EVENTSEL0_ENABLE); } static inline void x86_pmu_disable_counter(struct hw_perf_counter *hwc, int idx) { - int err; - err = checking_wrmsrl(hwc->config_base + idx, - hwc->config); + (void)checking_wrmsrl(hwc->config_base + idx, hwc->config); } static inline void @@ -835,13 +939,24 @@ intel_pmu_disable_fixed(struct hw_perf_counter *hwc, int __idx) { int idx = __idx - X86_PMC_IDX_FIXED; u64 ctrl_val, mask; - int err; mask = 0xfULL << (idx * 4); rdmsrl(hwc->config_base, ctrl_val); ctrl_val &= ~mask; - err = checking_wrmsrl(hwc->config_base, ctrl_val); + (void)checking_wrmsrl(hwc->config_base, ctrl_val); +} + +static inline void +p6_pmu_disable_counter(struct hw_perf_counter *hwc, int idx) +{ + struct cpu_hw_counters *cpuc = &__get_cpu_var(cpu_hw_counters); + u64 val = P6_NOP_COUNTER; + + if (cpuc->enabled) + val |= ARCH_PERFMON_EVENTSEL0_ENABLE; + + (void)checking_wrmsrl(hwc->config_base + idx, val); } static inline void @@ -911,6 +1026,8 @@ x86_perf_counter_set_period(struct perf_counter *counter, err = checking_wrmsrl(hwc->counter_base + idx, (u64)(-left) & x86_pmu.counter_mask); + perf_counter_update_userpage(counter); + return ret; } @@ -940,6 +1057,19 @@ intel_pmu_enable_fixed(struct hw_perf_counter *hwc, int __idx) err = checking_wrmsrl(hwc->config_base, ctrl_val); } +static void p6_pmu_enable_counter(struct hw_perf_counter *hwc, int idx) +{ + struct cpu_hw_counters *cpuc = &__get_cpu_var(cpu_hw_counters); + u64 val; + + val = hwc->config; + if (cpuc->enabled) + val |= ARCH_PERFMON_EVENTSEL0_ENABLE; + + (void)checking_wrmsrl(hwc->config_base + idx, val); +} + + static void intel_pmu_enable_counter(struct hw_perf_counter *hwc, int idx) { if (unlikely(hwc->config_base == MSR_ARCH_PERFMON_FIXED_CTR_CTRL)) { @@ -956,8 +1086,6 @@ static void amd_pmu_enable_counter(struct hw_perf_counter *hwc, int idx) if (cpuc->enabled) x86_pmu_enable_counter(hwc, idx); - else - x86_pmu_disable_counter(hwc, idx); } static int @@ -968,13 +1096,6 @@ fixed_mode_idx(struct perf_counter *counter, struct hw_perf_counter *hwc) if (!x86_pmu.num_counters_fixed) return -1; - /* - * Quirk, IA32_FIXED_CTRs do not work on current Atom processors: - */ - if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL && - boot_cpu_data.x86_model == 28) - return -1; - event = hwc->config & ARCH_PERFMON_EVENT_MASK; if (unlikely(event == x86_pmu.event_map(PERF_COUNT_HW_INSTRUCTIONS))) @@ -1040,6 +1161,8 @@ try_generic: x86_perf_counter_set_period(counter, hwc, idx); x86_pmu.enable(hwc, idx); + perf_counter_update_userpage(counter); + return 0; } @@ -1132,6 +1255,8 @@ static void x86_pmu_disable(struct perf_counter *counter) x86_perf_counter_update(counter, hwc, idx); cpuc->counters[idx] = NULL; clear_bit(idx, cpuc->used_mask); + + perf_counter_update_userpage(counter); } /* @@ -1176,6 +1301,49 @@ static void intel_pmu_reset(void) local_irq_restore(flags); } +static int p6_pmu_handle_irq(struct pt_regs *regs) +{ + struct perf_sample_data data; + struct cpu_hw_counters *cpuc; + struct perf_counter *counter; + struct hw_perf_counter *hwc; + int idx, handled = 0; + u64 val; + + data.regs = regs; + data.addr = 0; + + cpuc = &__get_cpu_var(cpu_hw_counters); + + for (idx = 0; idx < x86_pmu.num_counters; idx++) { + if (!test_bit(idx, cpuc->active_mask)) + continue; + + counter = cpuc->counters[idx]; + hwc = &counter->hw; + + val = x86_perf_counter_update(counter, hwc, idx); + if (val & (1ULL << (x86_pmu.counter_bits - 1))) + continue; + + /* + * counter overflow + */ + handled = 1; + data.period = counter->hw.last_period; + + if (!x86_perf_counter_set_period(counter, hwc, idx)) + continue; + + if (perf_counter_overflow(counter, 1, &data)) + p6_pmu_disable_counter(hwc, idx); + } + + if (handled) + inc_irq_stat(apic_perf_irqs); + + return handled; +} /* * This handler is triggered by the local APIC, so the APIC IRQ handling @@ -1185,14 +1353,13 @@ static int intel_pmu_handle_irq(struct pt_regs *regs) { struct perf_sample_data data; struct cpu_hw_counters *cpuc; - int bit, cpu, loops; + int bit, loops; u64 ack, status; data.regs = regs; data.addr = 0; - cpu = smp_processor_id(); - cpuc = &per_cpu(cpu_hw_counters, cpu); + cpuc = &__get_cpu_var(cpu_hw_counters); perf_disable(); status = intel_pmu_get_status(); @@ -1223,6 +1390,8 @@ again: if (!intel_pmu_save_and_restart(counter)) continue; + data.period = counter->hw.last_period; + if (perf_counter_overflow(counter, 1, &data)) intel_pmu_disable_counter(&counter->hw, bit); } @@ -1247,14 +1416,13 @@ static int amd_pmu_handle_irq(struct pt_regs *regs) struct cpu_hw_counters *cpuc; struct perf_counter *counter; struct hw_perf_counter *hwc; - int cpu, idx, handled = 0; + int idx, handled = 0; u64 val; data.regs = regs; data.addr = 0; - cpu = smp_processor_id(); - cpuc = &per_cpu(cpu_hw_counters, cpu); + cpuc = &__get_cpu_var(cpu_hw_counters); for (idx = 0; idx < x86_pmu.num_counters; idx++) { if (!test_bit(idx, cpuc->active_mask)) @@ -1297,18 +1465,22 @@ void smp_perf_pending_interrupt(struct pt_regs *regs) void set_perf_counter_pending(void) { +#ifdef CONFIG_X86_LOCAL_APIC apic->send_IPI_self(LOCAL_PENDING_VECTOR); +#endif } void perf_counters_lapic_init(void) { - if (!x86_pmu_initialized()) +#ifdef CONFIG_X86_LOCAL_APIC + if (!x86_pmu.apic || !x86_pmu_initialized()) return; /* * Always use NMI for PMU */ apic_write(APIC_LVTPC, APIC_DM_NMI); +#endif } static int __kprobes @@ -1332,7 +1504,9 @@ perf_counter_nmi_handler(struct notifier_block *self, regs = args->regs; +#ifdef CONFIG_X86_LOCAL_APIC apic_write(APIC_LVTPC, APIC_DM_NMI); +#endif /* * Can't rely on the handled return value to say it was our NMI, two * counters could trigger 'simultaneously' raising two back-to-back NMIs. @@ -1351,6 +1525,33 @@ static __read_mostly struct notifier_block perf_counter_nmi_notifier = { .priority = 1 }; +static struct x86_pmu p6_pmu = { + .name = "p6", + .handle_irq = p6_pmu_handle_irq, + .disable_all = p6_pmu_disable_all, + .enable_all = p6_pmu_enable_all, + .enable = p6_pmu_enable_counter, + .disable = p6_pmu_disable_counter, + .eventsel = MSR_P6_EVNTSEL0, + .perfctr = MSR_P6_PERFCTR0, + .event_map = p6_pmu_event_map, + .raw_event = p6_pmu_raw_event, + .max_events = ARRAY_SIZE(p6_perfmon_event_map), + .apic = 1, + .max_period = (1ULL << 31) - 1, + .version = 0, + .num_counters = 2, + /* + * Counters have 40 bits implemented. However they are designed such + * that bits [32-39] are sign extensions of bit 31. As such the + * effective width of a counter for P6-like PMU is 32 bits only. + * + * See IA-32 Intel Architecture Software developer manual Vol 3B + */ + .counter_bits = 32, + .counter_mask = (1ULL << 32) - 1, +}; + static struct x86_pmu intel_pmu = { .name = "Intel", .handle_irq = intel_pmu_handle_irq, @@ -1363,6 +1564,7 @@ static struct x86_pmu intel_pmu = { .event_map = intel_pmu_event_map, .raw_event = intel_pmu_raw_event, .max_events = ARRAY_SIZE(intel_perfmon_event_map), + .apic = 1, /* * Intel PMCs cannot be accessed sanely above 32 bit width, * so we install an artificial 1<<31 period regardless of @@ -1386,10 +1588,43 @@ static struct x86_pmu amd_pmu = { .num_counters = 4, .counter_bits = 48, .counter_mask = (1ULL << 48) - 1, + .apic = 1, /* use highest bit to detect overflow */ .max_period = (1ULL << 47) - 1, }; +static int p6_pmu_init(void) +{ + switch (boot_cpu_data.x86_model) { + case 1: + case 3: /* Pentium Pro */ + case 5: + case 6: /* Pentium II */ + case 7: + case 8: + case 11: /* Pentium III */ + break; + case 9: + case 13: + /* Pentium M */ + break; + default: + pr_cont("unsupported p6 CPU model %d ", + boot_cpu_data.x86_model); + return -ENODEV; + } + + x86_pmu = p6_pmu; + + if (!cpu_has_apic) { + pr_info("no APIC, boot with the \"lapic\" boot parameter to force-enable it.\n"); + pr_info("no hardware sampling interrupt available.\n"); + x86_pmu.apic = 0; + } + + return 0; +} + static int intel_pmu_init(void) { union cpuid10_edx edx; @@ -1398,8 +1633,14 @@ static int intel_pmu_init(void) unsigned int ebx; int version; - if (!cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON)) + if (!cpu_has(&boot_cpu_data, X86_FEATURE_ARCH_PERFMON)) { + /* check for P6 processor family */ + if (boot_cpu_data.x86 == 6) { + return p6_pmu_init(); + } else { return -ENODEV; + } + } /* * Check whether the Architectural PerfMon supports @@ -1425,8 +1666,6 @@ static int intel_pmu_init(void) */ x86_pmu.num_counters_fixed = max((int)edx.split.num_counters_fixed, 3); - rdmsrl(MSR_CORE_PERF_GLOBAL_CTRL, x86_pmu.intel_ctrl); - /* * Install the hw-cache-events table: */ @@ -1459,18 +1698,16 @@ static int intel_pmu_init(void) static int amd_pmu_init(void) { + /* Performance-monitoring supported from K7 and later: */ + if (boot_cpu_data.x86 < 6) + return -ENODEV; + x86_pmu = amd_pmu; - switch (boot_cpu_data.x86) { - case 0x0f: - case 0x10: - case 0x11: - memcpy(hw_cache_event_ids, amd_0f_hw_cache_event_ids, - sizeof(hw_cache_event_ids)); + /* Events are common for all AMDs */ + memcpy(hw_cache_event_ids, amd_hw_cache_event_ids, + sizeof(hw_cache_event_ids)); - pr_cont("AMD Family 0f/10/11 events, "); - break; - } return 0; } @@ -1498,21 +1735,22 @@ void __init init_hw_perf_counters(void) pr_cont("%s PMU driver.\n", x86_pmu.name); if (x86_pmu.num_counters > X86_PMC_MAX_GENERIC) { - x86_pmu.num_counters = X86_PMC_MAX_GENERIC; WARN(1, KERN_ERR "hw perf counters %d > max(%d), clipping!", x86_pmu.num_counters, X86_PMC_MAX_GENERIC); + x86_pmu.num_counters = X86_PMC_MAX_GENERIC; } perf_counter_mask = (1 << x86_pmu.num_counters) - 1; perf_max_counters = x86_pmu.num_counters; if (x86_pmu.num_counters_fixed > X86_PMC_MAX_FIXED) { - x86_pmu.num_counters_fixed = X86_PMC_MAX_FIXED; WARN(1, KERN_ERR "hw perf counters fixed %d > max(%d), clipping!", x86_pmu.num_counters_fixed, X86_PMC_MAX_FIXED); + x86_pmu.num_counters_fixed = X86_PMC_MAX_FIXED; } perf_counter_mask |= ((1LL << x86_pmu.num_counters_fixed)-1) << X86_PMC_IDX_FIXED; + x86_pmu.intel_ctrl = perf_counter_mask; perf_counters_lapic_init(); register_die_notifier(&perf_counter_nmi_notifier); @@ -1554,14 +1792,15 @@ const struct pmu *hw_perf_counter_init(struct perf_counter *counter) */ static inline -void callchain_store(struct perf_callchain_entry *entry, unsigned long ip) +void callchain_store(struct perf_callchain_entry *entry, u64 ip) { - if (entry->nr < MAX_STACK_DEPTH) + if (entry->nr < PERF_MAX_STACK_DEPTH) entry->ip[entry->nr++] = ip; } static DEFINE_PER_CPU(struct perf_callchain_entry, irq_entry); static DEFINE_PER_CPU(struct perf_callchain_entry, nmi_entry); +static DEFINE_PER_CPU(int, in_nmi_frame); static void @@ -1577,14 +1816,19 @@ static void backtrace_warning(void *data, char *msg) static int backtrace_stack(void *data, char *name) { - /* Don't bother with IRQ stacks for now */ - return -1; + per_cpu(in_nmi_frame, smp_processor_id()) = + x86_is_stack_id(NMI_STACK, name); + + return 0; } static void backtrace_address(void *data, unsigned long addr, int reliable) { struct perf_callchain_entry *entry = data; + if (per_cpu(in_nmi_frame, smp_processor_id())) + return; + if (reliable) callchain_store(entry, addr); } @@ -1596,47 +1840,59 @@ static const struct stacktrace_ops backtrace_ops = { .address = backtrace_address, }; +#include "../dumpstack.h" + static void perf_callchain_kernel(struct pt_regs *regs, struct perf_callchain_entry *entry) { - unsigned long bp; - char *stack; - int nr = entry->nr; + callchain_store(entry, PERF_CONTEXT_KERNEL); + callchain_store(entry, regs->ip); - callchain_store(entry, instruction_pointer(regs)); + dump_trace(NULL, regs, NULL, 0, &backtrace_ops, entry); +} - stack = ((char *)regs + sizeof(struct pt_regs)); -#ifdef CONFIG_FRAME_POINTER - bp = frame_pointer(regs); -#else - bp = 0; -#endif +/* + * best effort, GUP based copy_from_user() that assumes IRQ or NMI context + */ +static unsigned long +copy_from_user_nmi(void *to, const void __user *from, unsigned long n) +{ + unsigned long offset, addr = (unsigned long)from; + int type = in_nmi() ? KM_NMI : KM_IRQ0; + unsigned long size, len = 0; + struct page *page; + void *map; + int ret; - dump_trace(NULL, regs, (void *)stack, bp, &backtrace_ops, entry); + do { + ret = __get_user_pages_fast(addr, 1, 0, &page); + if (!ret) + break; - entry->kernel = entry->nr - nr; -} + offset = addr & (PAGE_SIZE - 1); + size = min(PAGE_SIZE - offset, n - len); + map = kmap_atomic(page, type); + memcpy(to, map+offset, size); + kunmap_atomic(map, type); + put_page(page); -struct stack_frame { - const void __user *next_fp; - unsigned long return_address; -}; + len += size; + to += size; + addr += size; + + } while (len < n); + + return len; +} static int copy_stack_frame(const void __user *fp, struct stack_frame *frame) { - int ret; - - if (!access_ok(VERIFY_READ, fp, sizeof(*frame))) - return 0; + unsigned long bytes; - ret = 1; - pagefault_disable(); - if (__copy_from_user_inatomic(frame, fp, sizeof(*frame))) - ret = 0; - pagefault_enable(); + bytes = copy_from_user_nmi(frame, fp, sizeof(*frame)); - return ret; + return bytes == sizeof(*frame); } static void @@ -1644,28 +1900,28 @@ perf_callchain_user(struct pt_regs *regs, struct perf_callchain_entry *entry) { struct stack_frame frame; const void __user *fp; - int nr = entry->nr; - regs = (struct pt_regs *)current->thread.sp0 - 1; - fp = (void __user *)regs->bp; + if (!user_mode(regs)) + regs = task_pt_regs(current); + + fp = (void __user *)regs->bp; + callchain_store(entry, PERF_CONTEXT_USER); callchain_store(entry, regs->ip); - while (entry->nr < MAX_STACK_DEPTH) { - frame.next_fp = NULL; + while (entry->nr < PERF_MAX_STACK_DEPTH) { + frame.next_frame = NULL; frame.return_address = 0; if (!copy_stack_frame(fp, &frame)) break; - if ((unsigned long)fp < user_stack_pointer(regs)) + if ((unsigned long)fp < regs->sp) break; callchain_store(entry, frame.return_address); - fp = frame.next_fp; + fp = frame.next_frame; } - - entry->user = entry->nr - nr; } static void @@ -1701,9 +1957,6 @@ struct perf_callchain_entry *perf_callchain(struct pt_regs *regs) entry = &__get_cpu_var(irq_entry); entry->nr = 0; - entry->hv = 0; - entry->kernel = 0; - entry->user = 0; perf_do_callchain(regs, entry); diff --git a/arch/x86/kernel/cpu/perfctr-watchdog.c b/arch/x86/kernel/cpu/perfctr-watchdog.c index d6f5b9f..e60ed74 100644 --- a/arch/x86/kernel/cpu/perfctr-watchdog.c +++ b/arch/x86/kernel/cpu/perfctr-watchdog.c @@ -716,11 +716,15 @@ static void probe_nmi_watchdog(void) wd_ops = &k7_wd_ops; break; case X86_VENDOR_INTEL: - /* - * Work around Core Duo (Yonah) errata AE49 where perfctr1 - * doesn't have a working enable bit. + /* Work around where perfctr1 doesn't have a working enable + * bit as described in the following errata: + * AE49 Core Duo and Intel Core Solo 65 nm + * AN49 Intel Pentium Dual-Core + * AF49 Dual-Core Intel Xeon Processor LV */ - if (boot_cpu_data.x86 == 6 && boot_cpu_data.x86_model == 14) { + if ((boot_cpu_data.x86 == 6 && boot_cpu_data.x86_model == 14) || + ((boot_cpu_data.x86 == 6 && boot_cpu_data.x86_model == 15 && + boot_cpu_data.x86_mask == 4))) { intel_arch_wd_ops.perfctr = MSR_ARCH_PERFMON_PERFCTR0; intel_arch_wd_ops.evntsel = MSR_ARCH_PERFMON_EVENTSEL0; } @@ -799,8 +803,3 @@ int __kprobes lapic_wd_event(unsigned nmi_hz) wd_ops->rearm(wd, nmi_hz); return 1; } - -int lapic_watchdog_ok(void) -{ - return wd_ops != NULL; -} diff --git a/arch/x86/kernel/crash.c b/arch/x86/kernel/crash.c index ff95824..5e409dc 100644 --- a/arch/x86/kernel/crash.c +++ b/arch/x86/kernel/crash.c @@ -27,6 +27,7 @@ #include <asm/cpu.h> #include <asm/reboot.h> #include <asm/virtext.h> +#include <asm/iommu.h> #if defined(CONFIG_SMP) && defined(CONFIG_X86_LOCAL_APIC) @@ -103,5 +104,10 @@ void native_machine_crash_shutdown(struct pt_regs *regs) #ifdef CONFIG_HPET_TIMER hpet_disable(); #endif + +#ifdef CONFIG_X86_64 + pci_iommu_shutdown(); +#endif + crash_save_cpu(regs, safe_smp_processor_id()); } diff --git a/arch/x86/kernel/dumpstack.c b/arch/x86/kernel/dumpstack.c index 95ea5fa..c8405718 100644 --- a/arch/x86/kernel/dumpstack.c +++ b/arch/x86/kernel/dumpstack.c @@ -22,6 +22,7 @@ #include "dumpstack.h" int panic_on_unrecovered_nmi; +int panic_on_io_nmi; unsigned int code_bytes = 64; int kstack_depth_to_print = 3 * STACKSLOTS_PER_LINE; static int die_counter; diff --git a/arch/x86/kernel/dumpstack_32.c b/arch/x86/kernel/dumpstack_32.c index d593cd1..bca5fba 100644 --- a/arch/x86/kernel/dumpstack_32.c +++ b/arch/x86/kernel/dumpstack_32.c @@ -19,6 +19,12 @@ #include "dumpstack.h" +/* Just a stub for now */ +int x86_is_stack_id(int id, char *name) +{ + return 0; +} + void dump_trace(struct task_struct *task, struct pt_regs *regs, unsigned long *stack, unsigned long bp, const struct stacktrace_ops *ops, void *data) diff --git a/arch/x86/kernel/dumpstack_64.c b/arch/x86/kernel/dumpstack_64.c index d35db59..54b0a32 100644 --- a/arch/x86/kernel/dumpstack_64.c +++ b/arch/x86/kernel/dumpstack_64.c @@ -19,10 +19,8 @@ #include "dumpstack.h" -static unsigned long *in_exception_stack(unsigned cpu, unsigned long stack, - unsigned *usedp, char **idp) -{ - static char ids[][8] = { + +static char x86_stack_ids[][8] = { [DEBUG_STACK - 1] = "#DB", [NMI_STACK - 1] = "NMI", [DOUBLEFAULT_STACK - 1] = "#DF", @@ -33,6 +31,15 @@ static unsigned long *in_exception_stack(unsigned cpu, unsigned long stack, N_EXCEPTION_STACKS + DEBUG_STKSZ / EXCEPTION_STKSZ - 2] = "#DB[?]" #endif }; + +int x86_is_stack_id(int id, char *name) +{ + return x86_stack_ids[id - 1] == name; +} + +static unsigned long *in_exception_stack(unsigned cpu, unsigned long stack, + unsigned *usedp, char **idp) +{ unsigned k; /* @@ -61,7 +68,7 @@ static unsigned long *in_exception_stack(unsigned cpu, unsigned long stack, if (*usedp & (1U << k)) break; *usedp |= 1U << k; - *idp = ids[k]; + *idp = x86_stack_ids[k]; return (unsigned long *)end; } /* @@ -81,12 +88,13 @@ static unsigned long *in_exception_stack(unsigned cpu, unsigned long stack, do { ++j; end -= EXCEPTION_STKSZ; - ids[j][4] = '1' + (j - N_EXCEPTION_STACKS); + x86_stack_ids[j][4] = '1' + + (j - N_EXCEPTION_STACKS); } while (stack < end - EXCEPTION_STKSZ); if (*usedp & (1U << j)) break; *usedp |= 1U << j; - *idp = ids[j]; + *idp = x86_stack_ids[j]; return (unsigned long *)end; } #endif diff --git a/arch/x86/kernel/e820.c b/arch/x86/kernel/e820.c index 7271fa3..5cb5725 100644 --- a/arch/x86/kernel/e820.c +++ b/arch/x86/kernel/e820.c @@ -627,10 +627,9 @@ __init void e820_setup_gap(void) #ifdef CONFIG_X86_64 if (!found) { gapstart = (max_pfn << PAGE_SHIFT) + 1024*1024; - printk(KERN_ERR "PCI: Warning: Cannot find a gap in the 32bit " - "address range\n" - KERN_ERR "PCI: Unassigned devices with 32bit resource " - "registers may break!\n"); + printk(KERN_ERR + "PCI: Warning: Cannot find a gap in the 32bit address range\n" + "PCI: Unassigned devices with 32bit resource registers may break!\n"); } #endif @@ -1383,6 +1382,8 @@ static unsigned long ram_alignment(resource_size_t pos) return 32*1024*1024; } +#define MAX_RESOURCE_SIZE ((resource_size_t)-1) + void __init e820_reserve_resources_late(void) { int i; @@ -1400,17 +1401,19 @@ void __init e820_reserve_resources_late(void) * avoid stolen RAM: */ for (i = 0; i < e820.nr_map; i++) { - struct e820entry *entry = &e820_saved.map[i]; - resource_size_t start, end; + struct e820entry *entry = &e820.map[i]; + u64 start, end; if (entry->type != E820_RAM) continue; start = entry->addr + entry->size; - end = round_up(start, ram_alignment(start)); - if (start == end) + end = round_up(start, ram_alignment(start)) - 1; + if (end > MAX_RESOURCE_SIZE) + end = MAX_RESOURCE_SIZE; + if (start >= end) continue; - reserve_region_with_split(&iomem_resource, start, - end - 1, "RAM buffer"); + reserve_region_with_split(&iomem_resource, start, end, + "RAM buffer"); } } diff --git a/arch/x86/kernel/efi.c b/arch/x86/kernel/efi.c index 1736acc..fe26ba3 100644 --- a/arch/x86/kernel/efi.c +++ b/arch/x86/kernel/efi.c @@ -240,10 +240,35 @@ static void __init do_add_efi_memmap(void) unsigned long long size = md->num_pages << EFI_PAGE_SHIFT; int e820_type; - if (md->attribute & EFI_MEMORY_WB) - e820_type = E820_RAM; - else + switch (md->type) { + case EFI_LOADER_CODE: + case EFI_LOADER_DATA: + case EFI_BOOT_SERVICES_CODE: + case EFI_BOOT_SERVICES_DATA: + case EFI_CONVENTIONAL_MEMORY: + if (md->attribute & EFI_MEMORY_WB) + e820_type = E820_RAM; + else + e820_type = E820_RESERVED; + break; + case EFI_ACPI_RECLAIM_MEMORY: + e820_type = E820_ACPI; + break; + case EFI_ACPI_MEMORY_NVS: + e820_type = E820_NVS; + break; + case EFI_UNUSABLE_MEMORY: + e820_type = E820_UNUSABLE; + break; + default: + /* + * EFI_RESERVED_TYPE EFI_RUNTIME_SERVICES_CODE + * EFI_RUNTIME_SERVICES_DATA EFI_MEMORY_MAPPED_IO + * EFI_MEMORY_MAPPED_IO_PORT_SPACE EFI_PAL_CODE + */ e820_type = E820_RESERVED; + break; + } e820_add_region(start, size, e820_type); } sanitize_e820_map(e820.map, ARRAY_SIZE(e820.map), &e820.nr_map); @@ -329,7 +354,7 @@ void __init efi_init(void) */ c16 = tmp = early_ioremap(efi.systab->fw_vendor, 2); if (c16) { - for (i = 0; i < sizeof(vendor) && *c16; ++i) + for (i = 0; i < sizeof(vendor) - 1 && *c16; ++i) vendor[i] = *c16++; vendor[i] = '\0'; } else @@ -487,7 +512,7 @@ void __init efi_enter_virtual_mode(void) && end_pfn <= max_pfn_mapped)) va = __va(md->phys_addr); else - va = efi_ioremap(md->phys_addr, size); + va = efi_ioremap(md->phys_addr, size, md->type); md->virt_addr = (u64) (unsigned long) va; diff --git a/arch/x86/kernel/efi_64.c b/arch/x86/kernel/efi_64.c index 22c3b78..ac0621a 100644 --- a/arch/x86/kernel/efi_64.c +++ b/arch/x86/kernel/efi_64.c @@ -98,10 +98,14 @@ void __init efi_call_phys_epilog(void) early_runtime_code_mapping_set_exec(0); } -void __iomem *__init efi_ioremap(unsigned long phys_addr, unsigned long size) +void __iomem *__init efi_ioremap(unsigned long phys_addr, unsigned long size, + u32 type) { unsigned long last_map_pfn; + if (type == EFI_MEMORY_MAPPED_IO) + return ioremap(phys_addr, size); + last_map_pfn = init_memory_mapping(phys_addr, phys_addr + size); if ((last_map_pfn << PAGE_SHIFT) < phys_addr + size) return NULL; diff --git a/arch/x86/kernel/entry_32.S b/arch/x86/kernel/entry_32.S index c929add..c097e7d 100644 --- a/arch/x86/kernel/entry_32.S +++ b/arch/x86/kernel/entry_32.S @@ -48,7 +48,6 @@ #include <asm/segment.h> #include <asm/smp.h> #include <asm/page_types.h> -#include <asm/desc.h> #include <asm/percpu.h> #include <asm/dwarf2.h> #include <asm/processor-flags.h> @@ -84,7 +83,7 @@ #define preempt_stop(clobbers) DISABLE_INTERRUPTS(clobbers); TRACE_IRQS_OFF #else #define preempt_stop(clobbers) -#define resume_kernel restore_nocheck +#define resume_kernel restore_all #endif .macro TRACE_IRQS_IRET @@ -372,7 +371,7 @@ END(ret_from_exception) ENTRY(resume_kernel) DISABLE_INTERRUPTS(CLBR_ANY) cmpl $0,TI_preempt_count(%ebp) # non-zero preempt_count ? - jnz restore_nocheck + jnz restore_all need_resched: movl TI_flags(%ebp), %ecx # need_resched set ? testb $_TIF_NEED_RESCHED, %cl @@ -540,6 +539,8 @@ syscall_exit: jne syscall_exit_work restore_all: + TRACE_IRQS_IRET +restore_all_notrace: movl PT_EFLAGS(%esp), %eax # mix EFLAGS, SS and CS # Warning: PT_OLDSS(%esp) contains the wrong/random values if we # are returning to the kernel. @@ -551,8 +552,6 @@ restore_all: CFI_REMEMBER_STATE je ldt_ss # returning to user-space with LDT SS restore_nocheck: - TRACE_IRQS_IRET -restore_nocheck_notrace: RESTORE_REGS 4 # skip orig_eax/error_code CFI_ADJUST_CFA_OFFSET -4 irq_return: @@ -588,22 +587,34 @@ ldt_ss: jne restore_nocheck #endif - /* If returning to userspace with 16bit stack, - * try to fix the higher word of ESP, as the CPU - * won't restore it. - * This is an "official" bug of all the x86-compatible - * CPUs, which we can try to work around to make - * dosemu and wine happy. */ - movl PT_OLDESP(%esp), %eax - movl %esp, %edx - call patch_espfix_desc +/* + * Setup and switch to ESPFIX stack + * + * We're returning to userspace with a 16 bit stack. The CPU will not + * restore the high word of ESP for us on executing iret... This is an + * "official" bug of all the x86-compatible CPUs, which we can work + * around to make dosemu and wine happy. We do this by preloading the + * high word of ESP with the high word of the userspace ESP while + * compensating for the offset by changing to the ESPFIX segment with + * a base address that matches for the difference. + */ + mov %esp, %edx /* load kernel esp */ + 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) + shr $16, %edx + mov %dl, GDT_ENTRY_ESPFIX_SS * 8 + 4(%ebx) /* bits 16..23 */ + mov %dh, GDT_ENTRY_ESPFIX_SS * 8 + 7(%ebx) /* bits 24..31 */ pushl $__ESPFIX_SS CFI_ADJUST_CFA_OFFSET 4 - pushl %eax + push %eax /* new kernel esp */ CFI_ADJUST_CFA_OFFSET 4 + /* Disable interrupts, but do not irqtrace this section: we + * will soon execute iret and the tracer was already set to + * the irqstate after the iret */ DISABLE_INTERRUPTS(CLBR_EAX) - TRACE_IRQS_OFF - lss (%esp), %esp + lss (%esp), %esp /* switch to espfix segment */ CFI_ADJUST_CFA_OFFSET -8 jmp restore_nocheck CFI_ENDPROC @@ -716,15 +727,24 @@ PTREGSCALL(vm86) PTREGSCALL(vm86old) .macro FIXUP_ESPFIX_STACK - /* since we are on a wrong stack, we cant make it a C code :( */ +/* + * Switch back for ESPFIX stack to the normal zerobased stack + * + * We can't call C functions using the ESPFIX stack. This code reads + * the high word of the segment base from the GDT and swiches to the + * normal stack and adjusts ESP with the matching offset. + */ + /* fixup the stack */ PER_CPU(gdt_page, %ebx) - GET_DESC_BASE(GDT_ENTRY_ESPFIX_SS, %ebx, %eax, %ax, %al, %ah) - addl %esp, %eax + 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 + addl %esp, %eax /* the adjusted stack pointer */ pushl $__KERNEL_DS CFI_ADJUST_CFA_OFFSET 4 pushl %eax CFI_ADJUST_CFA_OFFSET 4 - lss (%esp), %esp + lss (%esp), %esp /* switch to the normal stack segment */ CFI_ADJUST_CFA_OFFSET -8 .endm .macro UNWIND_ESPFIX_STACK @@ -1154,6 +1174,7 @@ ENTRY(ftrace_graph_caller) pushl %edx movl 0xc(%esp), %edx lea 0x4(%ebp), %eax + movl (%ebp), %ecx subl $MCOUNT_INSN_SIZE, %edx call prepare_ftrace_return popl %edx @@ -1168,6 +1189,7 @@ return_to_handler: pushl %eax pushl %ecx pushl %edx + movl %ebp, %eax call ftrace_return_to_handler movl %eax, 0xc(%esp) popl %edx @@ -1329,7 +1351,7 @@ nmi_stack_correct: xorl %edx,%edx # zero error code movl %esp,%eax # pt_regs pointer call do_nmi - jmp restore_nocheck_notrace + jmp restore_all_notrace CFI_ENDPROC nmi_stack_fixup: diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S index de74f0a..c251be7 100644 --- a/arch/x86/kernel/entry_64.S +++ b/arch/x86/kernel/entry_64.S @@ -135,6 +135,7 @@ ENTRY(ftrace_graph_caller) leaq 8(%rbp), %rdi movq 0x38(%rsp), %rsi + movq (%rbp), %rdx subq $MCOUNT_INSN_SIZE, %rsi call prepare_ftrace_return @@ -150,6 +151,7 @@ GLOBAL(return_to_handler) /* Save the return values */ movq %rax, (%rsp) movq %rdx, 8(%rsp) + movq %rbp, %rdi call ftrace_return_to_handler diff --git a/arch/x86/kernel/ftrace.c b/arch/x86/kernel/ftrace.c index b79c553..9dbb527 100644 --- a/arch/x86/kernel/ftrace.c +++ b/arch/x86/kernel/ftrace.c @@ -408,7 +408,8 @@ int ftrace_disable_ftrace_graph_caller(void) * Hook the return address and push it in the stack of return addrs * in current thread info. */ -void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr) +void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr, + unsigned long frame_pointer) { unsigned long old; int faulted; @@ -416,10 +417,6 @@ void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr) unsigned long return_hooker = (unsigned long) &return_to_handler; - /* Nmi's are currently unsupported */ - if (unlikely(in_nmi())) - return; - if (unlikely(atomic_read(¤t->tracing_graph_pause))) return; @@ -453,7 +450,8 @@ void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr) return; } - if (ftrace_push_return_trace(old, self_addr, &trace.depth) == -EBUSY) { + if (ftrace_push_return_trace(old, self_addr, &trace.depth, + frame_pointer) == -EBUSY) { *parent = old; return; } @@ -496,37 +494,56 @@ static struct syscall_metadata *find_syscall_meta(unsigned long *syscall) struct syscall_metadata *syscall_nr_to_meta(int nr) { - if (!syscalls_metadata || nr >= FTRACE_SYSCALL_MAX || nr < 0) + if (!syscalls_metadata || nr >= NR_syscalls || nr < 0) return NULL; return syscalls_metadata[nr]; } -void arch_init_ftrace_syscalls(void) +int syscall_name_to_nr(char *name) +{ + int i; + + if (!syscalls_metadata) + return -1; + + for (i = 0; i < NR_syscalls; i++) { + if (syscalls_metadata[i]) { + if (!strcmp(syscalls_metadata[i]->name, name)) + return i; + } + } + return -1; +} + +void set_syscall_enter_id(int num, int id) +{ + syscalls_metadata[num]->enter_id = id; +} + +void set_syscall_exit_id(int num, int id) +{ + syscalls_metadata[num]->exit_id = id; +} + +static int __init arch_init_ftrace_syscalls(void) { int i; struct syscall_metadata *meta; unsigned long **psys_syscall_table = &sys_call_table; - static atomic_t refs; - - if (atomic_inc_return(&refs) != 1) - goto end; syscalls_metadata = kzalloc(sizeof(*syscalls_metadata) * - FTRACE_SYSCALL_MAX, GFP_KERNEL); + NR_syscalls, GFP_KERNEL); if (!syscalls_metadata) { WARN_ON(1); - return; + return -ENOMEM; } - for (i = 0; i < FTRACE_SYSCALL_MAX; i++) { + for (i = 0; i < NR_syscalls; i++) { meta = find_syscall_meta(psys_syscall_table[i]); syscalls_metadata[i] = meta; } - return; - - /* Paranoid: avoid overflow */ -end: - atomic_dec(&refs); + return 0; } +arch_initcall(arch_init_ftrace_syscalls); #endif diff --git a/arch/x86/kernel/head_32.S b/arch/x86/kernel/head_32.S index dc5ed4b..cc827ac 100644 --- a/arch/x86/kernel/head_32.S +++ b/arch/x86/kernel/head_32.S @@ -13,7 +13,6 @@ #include <asm/segment.h> #include <asm/page_types.h> #include <asm/pgtable_types.h> -#include <asm/desc.h> #include <asm/cache.h> #include <asm/thread_info.h> #include <asm/asm-offsets.h> @@ -262,9 +261,7 @@ page_pde_offset = (__PAGE_OFFSET >> 20); * which will be freed later */ -#ifndef CONFIG_HOTPLUG_CPU -.section .init.text,"ax",@progbits -#endif +__CPUINIT #ifdef CONFIG_SMP ENTRY(startup_32_smp) @@ -603,7 +600,7 @@ ignore_int: #endif iret -.section .cpuinit.data,"wa" + __REFDATA .align 4 ENTRY(initial_code) .long i386_start_kernel diff --git a/arch/x86/kernel/head_64.S b/arch/x86/kernel/head_64.S index 54b29bb..fa54f78 100644 --- a/arch/x86/kernel/head_64.S +++ b/arch/x86/kernel/head_64.S @@ -12,7 +12,6 @@ #include <linux/linkage.h> #include <linux/threads.h> #include <linux/init.h> -#include <asm/desc.h> #include <asm/segment.h> #include <asm/pgtable.h> #include <asm/page.h> diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c index 81408b9..dedc2bd 100644 --- a/arch/x86/kernel/hpet.c +++ b/arch/x86/kernel/hpet.c @@ -510,7 +510,8 @@ static int hpet_setup_irq(struct hpet_dev *dev) { if (request_irq(dev->irq, hpet_interrupt_handler, - IRQF_DISABLED|IRQF_NOBALANCING, dev->name, dev)) + IRQF_TIMER | IRQF_DISABLED | IRQF_NOBALANCING, + dev->name, dev)) return -1; disable_irq(dev->irq); diff --git a/arch/x86/kernel/irqinit.c b/arch/x86/kernel/irqinit.c index 696f0e4..92b7703 100644 --- a/arch/x86/kernel/irqinit.c +++ b/arch/x86/kernel/irqinit.c @@ -187,7 +187,7 @@ static void __init apic_intr_init(void) #ifdef CONFIG_X86_THERMAL_VECTOR alloc_intr_gate(THERMAL_APIC_VECTOR, thermal_interrupt); #endif -#ifdef CONFIG_X86_THRESHOLD +#ifdef CONFIG_X86_MCE_THRESHOLD alloc_intr_gate(THRESHOLD_APIC_VECTOR, threshold_interrupt); #endif #if defined(CONFIG_X86_NEW_MCE) && defined(CONFIG_X86_LOCAL_APIC) diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c index a78ecad..c664d51 100644 --- a/arch/x86/kernel/kvm.c +++ b/arch/x86/kernel/kvm.c @@ -200,7 +200,7 @@ static void kvm_leave_lazy_mmu(void) state->mode = paravirt_get_lazy_mode(); } -static void paravirt_ops_setup(void) +static void __init paravirt_ops_setup(void) { pv_info.name = "KVM"; pv_info.paravirt_enabled = 1; diff --git a/arch/x86/kernel/mfgpt_32.c b/arch/x86/kernel/mfgpt_32.c index 846510b..2a62d84 100644 --- a/arch/x86/kernel/mfgpt_32.c +++ b/arch/x86/kernel/mfgpt_32.c @@ -347,7 +347,7 @@ static irqreturn_t mfgpt_tick(int irq, void *dev_id) static struct irqaction mfgptirq = { .handler = mfgpt_tick, - .flags = IRQF_DISABLED | IRQF_NOBALANCING, + .flags = IRQF_DISABLED | IRQF_NOBALANCING | IRQF_TIMER, .name = "mfgpt-timer" }; diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c index 745579b..1a041bc 100644 --- a/arch/x86/kernel/pci-dma.c +++ b/arch/x86/kernel/pci-dma.c @@ -32,6 +32,8 @@ int no_iommu __read_mostly; /* Set this to 1 if there is a HW IOMMU in the system */ int iommu_detected __read_mostly = 0; +int iommu_pass_through; + dma_addr_t bad_dma_address __read_mostly = 0; EXPORT_SYMBOL(bad_dma_address); @@ -210,6 +212,10 @@ static __init int iommu_setup(char *p) if (!strncmp(p, "soft", 4)) swiotlb = 1; #endif + if (!strncmp(p, "pt", 2)) { + iommu_pass_through = 1; + return 1; + } gart_parse_options(p); @@ -290,6 +296,8 @@ static int __init pci_iommu_init(void) void pci_iommu_shutdown(void) { gart_iommu_shutdown(); + + amd_iommu_shutdown(); } /* Must execute after PCI subsystem */ fs_initcall(pci_iommu_init); diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c index cfd9f90..d2e56b8 100644 --- a/arch/x86/kernel/pci-gart_64.c +++ b/arch/x86/kernel/pci-gart_64.c @@ -675,7 +675,7 @@ static __init int init_k8_gatt(struct agp_kern_info *info) nommu: /* Should not happen anymore */ printk(KERN_WARNING "PCI-DMA: More than 4GB of RAM and no IOMMU\n" - KERN_WARNING "falling back to iommu=soft.\n"); + "falling back to iommu=soft.\n"); return -1; } diff --git a/arch/x86/kernel/pci-swiotlb.c b/arch/x86/kernel/pci-swiotlb.c index a1712f2..6af96ee 100644 --- a/arch/x86/kernel/pci-swiotlb.c +++ b/arch/x86/kernel/pci-swiotlb.c @@ -71,7 +71,8 @@ void __init pci_swiotlb_init(void) { /* don't initialize swiotlb if iommu=off (no_iommu=1) */ #ifdef CONFIG_X86_64 - if (!iommu_detected && !no_iommu && max_pfn > MAX_DMA32_PFN) + if ((!iommu_detected && !no_iommu && max_pfn > MAX_DMA32_PFN) || + iommu_pass_through) swiotlb = 1; #endif if (swiotlb_force) diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index fc6e4b7..1092a1a 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -509,16 +509,12 @@ static void c1e_idle(void) if (!cpumask_test_cpu(cpu, c1e_mask)) { cpumask_set_cpu(cpu, c1e_mask); /* - * Force broadcast so ACPI can not interfere. Needs - * to run with interrupts enabled as it uses - * smp_function_call. + * Force broadcast so ACPI can not interfere. */ - local_irq_enable(); clockevents_notify(CLOCK_EVT_NOTIFY_BROADCAST_FORCE, &cpu); printk(KERN_INFO "Switch to broadcast mode on CPU%d\n", cpu); - local_irq_disable(); } clockevents_notify(CLOCK_EVT_NOTIFY_BROADCAST_ENTER, &cpu); diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c index cabdabc..113b892 100644 --- a/arch/x86/kernel/ptrace.c +++ b/arch/x86/kernel/ptrace.c @@ -36,10 +36,11 @@ #include <asm/ds.h> #include <asm/hw_breakpoint.h> -#include <trace/syscall.h> - #include "tls.h" +#define CREATE_TRACE_POINTS +#include <trace/events/syscalls.h> + enum x86_regset { REGSET_GENERAL, REGSET_FP, @@ -1548,8 +1549,8 @@ asmregparm long syscall_trace_enter(struct pt_regs *regs) tracehook_report_syscall_entry(regs)) ret = -1L; - if (unlikely(test_thread_flag(TIF_SYSCALL_FTRACE))) - ftrace_syscall_enter(regs); + if (unlikely(test_thread_flag(TIF_SYSCALL_TRACEPOINT))) + trace_sys_enter(regs, regs->orig_ax); if (unlikely(current->audit_context)) { if (IS_IA32) @@ -1574,8 +1575,8 @@ asmregparm void syscall_trace_leave(struct pt_regs *regs) if (unlikely(current->audit_context)) audit_syscall_exit(AUDITSC_RESULT(regs->ax), regs->ax); - if (unlikely(test_thread_flag(TIF_SYSCALL_FTRACE))) - ftrace_syscall_exit(regs); + if (unlikely(test_thread_flag(TIF_SYSCALL_TRACEPOINT))) + trace_sys_exit(regs, regs->ax); if (test_thread_flag(TIF_SYSCALL_TRACE)) tracehook_report_syscall_exit(regs, 0); diff --git a/arch/x86/kernel/pvclock.c b/arch/x86/kernel/pvclock.c index 4f9c55f..03801f2 100644 --- a/arch/x86/kernel/pvclock.c +++ b/arch/x86/kernel/pvclock.c @@ -60,7 +60,7 @@ static inline u64 scale_delta(u64 delta, u32 mul_frac, int shift) "adc %5,%%edx ; " : "=A" (product), "=r" (tmp1), "=r" (tmp2) : "a" ((u32)delta), "1" ((u32)(delta >> 32)), "2" (mul_frac) ); -#elif __x86_64__ +#elif defined(__x86_64__) __asm__ ( "mul %%rdx ; shrd $32,%%rdx,%%rax" : "=a" (product) : "0" (delta), "d" ((u64)mul_frac) ); diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c index d2d1ce8..a06e8d1 100644 --- a/arch/x86/kernel/reboot.c +++ b/arch/x86/kernel/reboot.c @@ -3,6 +3,7 @@ #include <linux/init.h> #include <linux/pm.h> #include <linux/efi.h> +#include <linux/dmi.h> #include <acpi/reboot.h> #include <asm/io.h> #include <asm/apic.h> @@ -17,7 +18,6 @@ #include <asm/cpu.h> #ifdef CONFIG_X86_32 -# include <linux/dmi.h> # include <linux/ctype.h> # include <linux/mc146818rtc.h> #else @@ -249,6 +249,14 @@ static struct dmi_system_id __initdata reboot_dmi_table[] = { DMI_MATCH(DMI_PRODUCT_NAME, "VGN-Z540N"), }, }, + { /* Handle problems with rebooting on CompuLab SBC-FITPC2 */ + .callback = set_bios_reboot, + .ident = "CompuLab SBC-FITPC2", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "CompuLab"), + DMI_MATCH(DMI_PRODUCT_NAME, "SBC-FITPC2"), + }, + }, { } }; @@ -396,6 +404,46 @@ EXPORT_SYMBOL(machine_real_restart); #endif /* CONFIG_X86_32 */ +/* + * Some Apple MacBook and MacBookPro's needs reboot=p to be able to reboot + */ +static int __init set_pci_reboot(const struct dmi_system_id *d) +{ + if (reboot_type != BOOT_CF9) { + reboot_type = BOOT_CF9; + printk(KERN_INFO "%s series board detected. " + "Selecting PCI-method for reboots.\n", d->ident); + } + return 0; +} + +static struct dmi_system_id __initdata pci_reboot_dmi_table[] = { + { /* Handle problems with rebooting on Apple MacBook5 */ + .callback = set_pci_reboot, + .ident = "Apple MacBook5", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Apple Inc."), + DMI_MATCH(DMI_PRODUCT_NAME, "MacBook5"), + }, + }, + { /* Handle problems with rebooting on Apple MacBookPro5 */ + .callback = set_pci_reboot, + .ident = "Apple MacBookPro5", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Apple Inc."), + DMI_MATCH(DMI_PRODUCT_NAME, "MacBookPro5"), + }, + }, + { } +}; + +static int __init pci_reboot_init(void) +{ + dmi_check_system(pci_reboot_dmi_table); + return 0; +} +core_initcall(pci_reboot_init); + static inline void kb_wait(void) { int i; diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index be5ae80..63f32d2 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -289,6 +289,20 @@ void * __init extend_brk(size_t size, size_t align) return ret; } +#ifdef CONFIG_X86_64 +static void __init init_gbpages(void) +{ + if (direct_gbpages && cpu_has_gbpages) + printk(KERN_INFO "Using GB pages for direct mapping\n"); + else + direct_gbpages = 0; +} +#else +static inline void init_gbpages(void) +{ +} +#endif + static void __init reserve_brk(void) { if (_brk_end > _brk_start) @@ -658,6 +672,19 @@ static struct dmi_system_id __initdata bad_bios_dmi_table[] = { DMI_MATCH(DMI_BIOS_VENDOR, "Phoenix Technologies"), }, }, + { + /* + * AMI BIOS with low memory corruption was found on Intel DG45ID board. + * It hase different DMI_BIOS_VENDOR = "Intel Corp.", for now we will + * match only DMI_BOARD_NAME and see if there is more bad products + * with this vendor. + */ + .callback = dmi_low_memory_corruption, + .ident = "AMI BIOS", + .matches = { + DMI_MATCH(DMI_BOARD_NAME, "DG45ID"), + }, + }, #endif {} }; @@ -871,6 +898,8 @@ void __init setup_arch(char **cmdline_p) reserve_brk(); + init_gbpages(); + /* max_pfn_mapped is updated here */ max_low_pfn_mapped = init_memory_mapping(0, max_low_pfn<<PAGE_SHIFT); max_pfn_mapped = max_low_pfn_mapped; diff --git a/arch/x86/kernel/setup_percpu.c b/arch/x86/kernel/setup_percpu.c index 9c3f082..07d8191 100644 --- a/arch/x86/kernel/setup_percpu.c +++ b/arch/x86/kernel/setup_percpu.c @@ -124,7 +124,7 @@ static void * __init pcpu_alloc_bootmem(unsigned int cpu, unsigned long size, } /* - * Remap allocator + * Large page remap allocator * * This allocator uses PMD page as unit. A PMD page is allocated for * each cpu and each is remapped into vmalloc area using PMD mapping. @@ -137,105 +137,185 @@ static void * __init pcpu_alloc_bootmem(unsigned int cpu, unsigned long size, * better than only using 4k mappings while still being NUMA friendly. */ #ifdef CONFIG_NEED_MULTIPLE_NODES -static size_t pcpur_size __initdata; -static void **pcpur_ptrs __initdata; +struct pcpul_ent { + unsigned int cpu; + void *ptr; +}; + +static size_t pcpul_size; +static struct pcpul_ent *pcpul_map; +static struct vm_struct pcpul_vm; -static struct page * __init pcpur_get_page(unsigned int cpu, int pageno) +static struct page * __init pcpul_get_page(unsigned int cpu, int pageno) { size_t off = (size_t)pageno << PAGE_SHIFT; - if (off >= pcpur_size) + if (off >= pcpul_size) return NULL; - return virt_to_page(pcpur_ptrs[cpu] + off); + return virt_to_page(pcpul_map[cpu].ptr + off); } -static ssize_t __init setup_pcpu_remap(size_t static_size) +static ssize_t __init setup_pcpu_lpage(size_t static_size, bool chosen) { - static struct vm_struct vm; - size_t ptrs_size, dyn_size; + size_t map_size, dyn_size; unsigned int cpu; + int i, j; ssize_t ret; - /* - * If large page isn't supported, there's no benefit in doing - * this. Also, on non-NUMA, embedding is better. - * - * NOTE: disabled for now. - */ - if (true || !cpu_has_pse || !pcpu_need_numa()) + if (!chosen) { + size_t vm_size = VMALLOC_END - VMALLOC_START; + size_t tot_size = nr_cpu_ids * PMD_SIZE; + + /* on non-NUMA, embedding is better */ + if (!pcpu_need_numa()) + return -EINVAL; + + /* don't consume more than 20% of vmalloc area */ + if (tot_size > vm_size / 5) { + pr_info("PERCPU: too large chunk size %zuMB for " + "large page remap\n", tot_size >> 20); + return -EINVAL; + } + } + + /* need PSE */ + if (!cpu_has_pse) { + pr_warning("PERCPU: lpage allocator requires PSE\n"); return -EINVAL; + } /* * Currently supports only single page. Supporting multiple * pages won't be too difficult if it ever becomes necessary. */ - pcpur_size = PFN_ALIGN(static_size + PERCPU_MODULE_RESERVE + + pcpul_size = PFN_ALIGN(static_size + PERCPU_MODULE_RESERVE + PERCPU_DYNAMIC_RESERVE); - if (pcpur_size > PMD_SIZE) { + if (pcpul_size > PMD_SIZE) { pr_warning("PERCPU: static data is larger than large page, " "can't use large page\n"); return -EINVAL; } - dyn_size = pcpur_size - static_size - PERCPU_FIRST_CHUNK_RESERVE; + dyn_size = pcpul_size - static_size - PERCPU_FIRST_CHUNK_RESERVE; /* allocate pointer array and alloc large pages */ - ptrs_size = PFN_ALIGN(num_possible_cpus() * sizeof(pcpur_ptrs[0])); - pcpur_ptrs = alloc_bootmem(ptrs_size); + map_size = PFN_ALIGN(nr_cpu_ids * sizeof(pcpul_map[0])); + pcpul_map = alloc_bootmem(map_size); for_each_possible_cpu(cpu) { - pcpur_ptrs[cpu] = pcpu_alloc_bootmem(cpu, PMD_SIZE, PMD_SIZE); - if (!pcpur_ptrs[cpu]) + pcpul_map[cpu].cpu = cpu; + pcpul_map[cpu].ptr = pcpu_alloc_bootmem(cpu, PMD_SIZE, + PMD_SIZE); + if (!pcpul_map[cpu].ptr) { + pr_warning("PERCPU: failed to allocate large page " + "for cpu%u\n", cpu); goto enomem; + } /* - * Only use pcpur_size bytes and give back the rest. + * Only use pcpul_size bytes and give back the rest. * * Ingo: The 2MB up-rounding bootmem is needed to make * sure the partial 2MB page is still fully RAM - it's * not well-specified to have a PAT-incompatible area * (unmapped RAM, device memory, etc.) in that hole. */ - free_bootmem(__pa(pcpur_ptrs[cpu] + pcpur_size), - PMD_SIZE - pcpur_size); + free_bootmem(__pa(pcpul_map[cpu].ptr + pcpul_size), + PMD_SIZE - pcpul_size); - memcpy(pcpur_ptrs[cpu], __per_cpu_load, static_size); + memcpy(pcpul_map[cpu].ptr, __per_cpu_load, static_size); } /* allocate address and map */ - vm.flags = VM_ALLOC; - vm.size = num_possible_cpus() * PMD_SIZE; - vm_area_register_early(&vm, PMD_SIZE); + pcpul_vm.flags = VM_ALLOC; + pcpul_vm.size = nr_cpu_ids * PMD_SIZE; + vm_area_register_early(&pcpul_vm, PMD_SIZE); for_each_possible_cpu(cpu) { - pmd_t *pmd; + pmd_t *pmd, pmd_v; - pmd = populate_extra_pmd((unsigned long)vm.addr - + cpu * PMD_SIZE); - set_pmd(pmd, pfn_pmd(page_to_pfn(virt_to_page(pcpur_ptrs[cpu])), - PAGE_KERNEL_LARGE)); + pmd = populate_extra_pmd((unsigned long)pcpul_vm.addr + + cpu * PMD_SIZE); + pmd_v = pfn_pmd(page_to_pfn(virt_to_page(pcpul_map[cpu].ptr)), + PAGE_KERNEL_LARGE); + set_pmd(pmd, pmd_v); } /* we're ready, commit */ pr_info("PERCPU: Remapped at %p with large pages, static data " - "%zu bytes\n", vm.addr, static_size); + "%zu bytes\n", pcpul_vm.addr, static_size); - ret = pcpu_setup_first_chunk(pcpur_get_page, static_size, + ret = pcpu_setup_first_chunk(pcpul_get_page, static_size, PERCPU_FIRST_CHUNK_RESERVE, dyn_size, - PMD_SIZE, vm.addr, NULL); - goto out_free_ar; + PMD_SIZE, pcpul_vm.addr, NULL); + + /* sort pcpul_map array for pcpu_lpage_remapped() */ + for (i = 0; i < nr_cpu_ids - 1; i++) + for (j = i + 1; j < nr_cpu_ids; j++) + if (pcpul_map[i].ptr > pcpul_map[j].ptr) { + struct pcpul_ent tmp = pcpul_map[i]; + pcpul_map[i] = pcpul_map[j]; + pcpul_map[j] = tmp; + } + + return ret; enomem: for_each_possible_cpu(cpu) - if (pcpur_ptrs[cpu]) - free_bootmem(__pa(pcpur_ptrs[cpu]), PMD_SIZE); - ret = -ENOMEM; -out_free_ar: - free_bootmem(__pa(pcpur_ptrs), ptrs_size); - return ret; + if (pcpul_map[cpu].ptr) + free_bootmem(__pa(pcpul_map[cpu].ptr), pcpul_size); + free_bootmem(__pa(pcpul_map), map_size); + return -ENOMEM; +} + +/** + * pcpu_lpage_remapped - determine whether a kaddr is in pcpul recycled area + * @kaddr: the kernel address in question + * + * Determine whether @kaddr falls in the pcpul recycled area. This is + * used by pageattr to detect VM aliases and break up the pcpu PMD + * mapping such that the same physical page is not mapped under + * different attributes. + * + * The recycled area is always at the tail of a partially used PMD + * page. + * + * RETURNS: + * Address of corresponding remapped pcpu address if match is found; + * otherwise, NULL. + */ +void *pcpu_lpage_remapped(void *kaddr) +{ + void *pmd_addr = (void *)((unsigned long)kaddr & PMD_MASK); + unsigned long offset = (unsigned long)kaddr & ~PMD_MASK; + int left = 0, right = nr_cpu_ids - 1; + int pos; + + /* pcpul in use at all? */ + if (!pcpul_map) + return NULL; + + /* okay, perform binary search */ + while (left <= right) { + pos = (left + right) / 2; + + if (pcpul_map[pos].ptr < pmd_addr) + left = pos + 1; + else if (pcpul_map[pos].ptr > pmd_addr) + right = pos - 1; + else { + /* it shouldn't be in the area for the first chunk */ + WARN_ON(offset < pcpul_size); + + return pcpul_vm.addr + + pcpul_map[pos].cpu * PMD_SIZE + offset; + } + } + + return NULL; } #else -static ssize_t __init setup_pcpu_remap(size_t static_size) +static ssize_t __init setup_pcpu_lpage(size_t static_size, bool chosen) { return -EINVAL; } @@ -249,7 +329,7 @@ static ssize_t __init setup_pcpu_remap(size_t static_size) * mapping so that it can use PMD mapping without additional TLB * pressure. */ -static ssize_t __init setup_pcpu_embed(size_t static_size) +static ssize_t __init setup_pcpu_embed(size_t static_size, bool chosen) { size_t reserve = PERCPU_MODULE_RESERVE + PERCPU_DYNAMIC_RESERVE; @@ -258,7 +338,7 @@ static ssize_t __init setup_pcpu_embed(size_t static_size) * this. Also, embedding allocation doesn't play well with * NUMA. */ - if (!cpu_has_pse || pcpu_need_numa()) + if (!chosen && (!cpu_has_pse || pcpu_need_numa())) return -EINVAL; return pcpu_embed_first_chunk(static_size, PERCPU_FIRST_CHUNK_RESERVE, @@ -297,7 +377,7 @@ static ssize_t __init setup_pcpu_4k(size_t static_size) pcpu4k_nr_static_pages = PFN_UP(static_size); /* unaligned allocations can't be freed, round up to page size */ - pages_size = PFN_ALIGN(pcpu4k_nr_static_pages * num_possible_cpus() + pages_size = PFN_ALIGN(pcpu4k_nr_static_pages * nr_cpu_ids * sizeof(pcpu4k_pages[0])); pcpu4k_pages = alloc_bootmem(pages_size); @@ -308,8 +388,11 @@ static ssize_t __init setup_pcpu_4k(size_t static_size) void *ptr; ptr = pcpu_alloc_bootmem(cpu, PAGE_SIZE, PAGE_SIZE); - if (!ptr) + if (!ptr) { + pr_warning("PERCPU: failed to allocate " + "4k page for cpu%u\n", cpu); goto enomem; + } memcpy(ptr, __per_cpu_load + i * PAGE_SIZE, PAGE_SIZE); pcpu4k_pages[j++] = virt_to_page(ptr); @@ -333,6 +416,16 @@ out_free_ar: return ret; } +/* for explicit first chunk allocator selection */ +static char pcpu_chosen_alloc[16] __initdata; + +static int __init percpu_alloc_setup(char *str) +{ + strncpy(pcpu_chosen_alloc, str, sizeof(pcpu_chosen_alloc) - 1); + return 0; +} +early_param("percpu_alloc", percpu_alloc_setup); + static inline void setup_percpu_segment(int cpu) { #ifdef CONFIG_X86_32 @@ -346,11 +439,6 @@ static inline void setup_percpu_segment(int cpu) #endif } -/* - * Great future plan: - * Declare PDA itself and support (irqstack,tss,pgd) as per cpu data. - * Always point %gs to its beginning - */ void __init setup_per_cpu_areas(void) { size_t static_size = __per_cpu_end - __per_cpu_start; @@ -367,9 +455,26 @@ void __init setup_per_cpu_areas(void) * of large page mappings. Please read comments on top of * each allocator for details. */ - ret = setup_pcpu_remap(static_size); - if (ret < 0) - ret = setup_pcpu_embed(static_size); + ret = -EINVAL; + if (strlen(pcpu_chosen_alloc)) { + if (strcmp(pcpu_chosen_alloc, "4k")) { + if (!strcmp(pcpu_chosen_alloc, "lpage")) + ret = setup_pcpu_lpage(static_size, true); + else if (!strcmp(pcpu_chosen_alloc, "embed")) + ret = setup_pcpu_embed(static_size, true); + else + pr_warning("PERCPU: unknown allocator %s " + "specified\n", pcpu_chosen_alloc); + if (ret < 0) + pr_warning("PERCPU: %s allocator failed (%zd), " + "falling back to 4k\n", + pcpu_chosen_alloc, ret); + } + } else { + ret = setup_pcpu_lpage(static_size, false); + if (ret < 0) + ret = setup_pcpu_embed(static_size, false); + } if (ret < 0) ret = setup_pcpu_4k(static_size); if (ret < 0) diff --git a/arch/x86/kernel/sys_x86_64.c b/arch/x86/kernel/sys_x86_64.c index 6bc211a..45e00eb 100644 --- a/arch/x86/kernel/sys_x86_64.c +++ b/arch/x86/kernel/sys_x86_64.c @@ -18,9 +18,9 @@ #include <asm/ia32.h> #include <asm/syscalls.h> -asmlinkage long sys_mmap(unsigned long addr, unsigned long len, - unsigned long prot, unsigned long flags, - unsigned long fd, unsigned long off) +SYSCALL_DEFINE6(mmap, unsigned long, addr, unsigned long, len, + unsigned long, prot, unsigned long, flags, + unsigned long, fd, unsigned long, off) { long error; struct file *file; @@ -226,7 +226,7 @@ bottomup: } -asmlinkage long sys_uname(struct new_utsname __user *name) +SYSCALL_DEFINE1(uname, struct new_utsname __user *, name) { int err; down_read(&uts_sem); diff --git a/arch/x86/kernel/tlb_uv.c b/arch/x86/kernel/tlb_uv.c index 124d40c..77b9689 100644 --- a/arch/x86/kernel/tlb_uv.c +++ b/arch/x86/kernel/tlb_uv.c @@ -711,7 +711,6 @@ uv_activation_descriptor_init(int node, int pnode) unsigned long pa; unsigned long m; unsigned long n; - unsigned long mmr_image; struct bau_desc *adp; struct bau_desc *ad2; @@ -727,12 +726,8 @@ uv_activation_descriptor_init(int node, int pnode) n = pa >> uv_nshift; m = pa & uv_mmask; - mmr_image = uv_read_global_mmr64(pnode, UVH_LB_BAU_SB_DESCRIPTOR_BASE); - if (mmr_image) { - uv_write_global_mmr64(pnode, (unsigned long) - UVH_LB_BAU_SB_DESCRIPTOR_BASE, - (n << UV_DESC_BASE_PNODE_SHIFT | m)); - } + uv_write_global_mmr64(pnode, UVH_LB_BAU_SB_DESCRIPTOR_BASE, + (n << UV_DESC_BASE_PNODE_SHIFT | m)); /* * initializing all 8 (UV_ITEMS_PER_DESCRIPTOR) descriptors for each @@ -749,6 +744,7 @@ uv_activation_descriptor_init(int node, int pnode) * note that base_dest_nodeid is actually a nasid. */ ad2->header.base_dest_nodeid = uv_partition_base_pnode << 1; + ad2->header.dest_subnodeid = 0x10; /* the LB */ ad2->header.command = UV_NET_ENDPOINT_INTD; ad2->header.int_both = 1; /* diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c index 286d64e..ae04589 100644 --- a/arch/x86/kernel/traps.c +++ b/arch/x86/kernel/traps.c @@ -54,6 +54,7 @@ #include <asm/traps.h> #include <asm/desc.h> #include <asm/i387.h> +#include <asm/mce.h> #include <asm/mach_traps.h> @@ -65,8 +66,6 @@ #include <asm/setup.h> #include <asm/traps.h> -#include "cpu/mcheck/mce.h" - asmlinkage int system_call(void); /* Do we ignore FPU interrupts ? */ @@ -347,6 +346,9 @@ io_check_error(unsigned char reason, struct pt_regs *regs) printk(KERN_EMERG "NMI: IOCK error (debug interrupt?)\n"); show_registers(regs); + if (panic_on_io_nmi) + panic("NMI IOCK error: Not continuing"); + /* Re-enable the IOCK line, wait for a few seconds */ reason = (reason & 0xf) | 8; outb(reason, 0x61); diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c index ae3180c..71f4368 100644 --- a/arch/x86/kernel/tsc.c +++ b/arch/x86/kernel/tsc.c @@ -275,15 +275,20 @@ static unsigned long pit_calibrate_tsc(u32 latch, unsigned long ms, int loopmin) * use the TSC value at the transitions to calculate a pretty * good value for the TSC frequencty. */ +static inline int pit_verify_msb(unsigned char val) +{ + /* Ignore LSB */ + inb(0x42); + return inb(0x42) == val; +} + static inline int pit_expect_msb(unsigned char val, u64 *tscp, unsigned long *deltap) { int count; u64 tsc = 0; for (count = 0; count < 50000; count++) { - /* Ignore LSB */ - inb(0x42); - if (inb(0x42) != val) + if (!pit_verify_msb(val)) break; tsc = get_cycles(); } @@ -336,8 +341,7 @@ static unsigned long quick_pit_calibrate(void) * to do that is to just read back the 16-bit counter * once from the PIT. */ - inb(0x42); - inb(0x42); + pit_verify_msb(0); if (pit_expect_msb(0xff, &tsc, &d1)) { for (i = 1; i <= MAX_QUICK_PIT_ITERATIONS; i++) { @@ -348,8 +352,19 @@ static unsigned long quick_pit_calibrate(void) * Iterate until the error is less than 500 ppm */ delta -= tsc; - if (d1+d2 < delta >> 11) - goto success; + if (d1+d2 >= delta >> 11) + continue; + + /* + * Check the PIT one more time to verify that + * all TSC reads were stable wrt the PIT. + * + * This also guarantees serialization of the + * last cycle read ('d2') in pit_expect_msb. + */ + if (!pit_verify_msb(0xfe - i)) + break; + goto success; } } printk("Fast TSC calibration failed\n"); @@ -590,22 +605,26 @@ EXPORT_SYMBOL(recalibrate_cpu_khz); */ DEFINE_PER_CPU(unsigned long, cyc2ns); +DEFINE_PER_CPU(unsigned long long, cyc2ns_offset); static void set_cyc2ns_scale(unsigned long cpu_khz, int cpu) { - unsigned long long tsc_now, ns_now; + unsigned long long tsc_now, ns_now, *offset; unsigned long flags, *scale; local_irq_save(flags); sched_clock_idle_sleep_event(); scale = &per_cpu(cyc2ns, cpu); + offset = &per_cpu(cyc2ns_offset, cpu); rdtscll(tsc_now); ns_now = __cycles_2_ns(tsc_now); - if (cpu_khz) + if (cpu_khz) { *scale = (NSEC_PER_MSEC << CYC2NS_SCALE_FACTOR)/cpu_khz; + *offset = ns_now - (tsc_now * *scale >> CYC2NS_SCALE_FACTOR); + } sched_clock_idle_wakeup_event(0); local_irq_restore(flags); @@ -632,17 +651,15 @@ static int time_cpufreq_notifier(struct notifier_block *nb, unsigned long val, void *data) { struct cpufreq_freqs *freq = data; - unsigned long *lpj, dummy; + unsigned long *lpj; if (cpu_has(&cpu_data(freq->cpu), X86_FEATURE_CONSTANT_TSC)) return 0; - lpj = &dummy; - if (!(freq->flags & CPUFREQ_CONST_LOOPS)) + lpj = &boot_cpu_data.loops_per_jiffy; #ifdef CONFIG_SMP + if (!(freq->flags & CPUFREQ_CONST_LOOPS)) lpj = &cpu_data(freq->cpu).loops_per_jiffy; -#else - lpj = &boot_cpu_data.loops_per_jiffy; #endif if (!ref_freq) { diff --git a/arch/x86/kernel/vmi_32.c b/arch/x86/kernel/vmi_32.c index b263423..95a7289 100644 --- a/arch/x86/kernel/vmi_32.c +++ b/arch/x86/kernel/vmi_32.c @@ -441,7 +441,7 @@ vmi_startup_ipi_hook(int phys_apicid, unsigned long start_eip, ap.ds = __USER_DS; ap.es = __USER_DS; ap.fs = __KERNEL_PERCPU; - ap.gs = 0; + ap.gs = __KERNEL_STACK_CANARY; ap.eflags = 0; diff --git a/arch/x86/kernel/vmlinux.lds.S b/arch/x86/kernel/vmlinux.lds.S index 367e878..9fc1782 100644 --- a/arch/x86/kernel/vmlinux.lds.S +++ b/arch/x86/kernel/vmlinux.lds.S @@ -46,11 +46,10 @@ PHDRS { data PT_LOAD FLAGS(7); /* RWE */ #ifdef CONFIG_X86_64 user PT_LOAD FLAGS(7); /* RWE */ - data.init PT_LOAD FLAGS(7); /* RWE */ #ifdef CONFIG_SMP percpu PT_LOAD FLAGS(7); /* RWE */ #endif - data.init2 PT_LOAD FLAGS(7); /* RWE */ + init PT_LOAD FLAGS(7); /* RWE */ #endif note PT_NOTE FLAGS(0); /* ___ */ } @@ -103,72 +102,43 @@ SECTIONS __stop___ex_table = .; } :text = 0x9090 - RODATA + RO_DATA(PAGE_SIZE) /* Data */ - . = ALIGN(PAGE_SIZE); .data : AT(ADDR(.data) - LOAD_OFFSET) { /* Start of data section */ _sdata = .; - DATA_DATA - CONSTRUCTORS -#ifdef CONFIG_X86_64 - /* End of data section */ - _edata = .; -#endif - } :data + /* init_task */ + INIT_TASK_DATA(THREAD_SIZE) #ifdef CONFIG_X86_32 - /* 32 bit has nosave before _edata */ - . = ALIGN(PAGE_SIZE); - .data_nosave : AT(ADDR(.data_nosave) - LOAD_OFFSET) { - __nosave_begin = .; - *(.data.nosave) - . = ALIGN(PAGE_SIZE); - __nosave_end = .; - } + /* 32 bit has nosave before _edata */ + NOSAVE_DATA #endif - . = ALIGN(PAGE_SIZE); - .data.page_aligned : AT(ADDR(.data.page_aligned) - LOAD_OFFSET) { - *(.data.page_aligned) + PAGE_ALIGNED_DATA(PAGE_SIZE) *(.data.idt) - } -#ifdef CONFIG_X86_32 - . = ALIGN(32); -#else - . = ALIGN(PAGE_SIZE); - . = ALIGN(CONFIG_X86_L1_CACHE_BYTES); -#endif - .data.cacheline_aligned : - AT(ADDR(.data.cacheline_aligned) - LOAD_OFFSET) { - *(.data.cacheline_aligned) - } + CACHELINE_ALIGNED_DATA(CONFIG_X86_L1_CACHE_BYTES) - /* rarely changed data like cpu maps */ -#ifdef CONFIG_X86_32 - . = ALIGN(32); -#else - . = ALIGN(CONFIG_X86_INTERNODE_CACHE_BYTES); -#endif - .data.read_mostly : AT(ADDR(.data.read_mostly) - LOAD_OFFSET) { - *(.data.read_mostly) + DATA_DATA + CONSTRUCTORS + + /* rarely changed data like cpu maps */ + READ_MOSTLY_DATA(CONFIG_X86_INTERNODE_CACHE_BYTES) -#ifdef CONFIG_X86_32 /* End of data section */ _edata = .; -#endif - } + } :data #ifdef CONFIG_X86_64 #define VSYSCALL_ADDR (-10*1024*1024) -#define VSYSCALL_PHYS_ADDR ((LOADADDR(.data.read_mostly) + \ - SIZEOF(.data.read_mostly) + 4095) & ~(4095)) -#define VSYSCALL_VIRT_ADDR ((ADDR(.data.read_mostly) + \ - SIZEOF(.data.read_mostly) + 4095) & ~(4095)) +#define VSYSCALL_PHYS_ADDR ((LOADADDR(.data) + SIZEOF(.data) + \ + PAGE_SIZE - 1) & ~(PAGE_SIZE - 1)) +#define VSYSCALL_VIRT_ADDR ((ADDR(.data) + SIZEOF(.data) + \ + PAGE_SIZE - 1) & ~(PAGE_SIZE - 1)) #define VLOAD_OFFSET (VSYSCALL_ADDR - VSYSCALL_PHYS_ADDR) #define VLOAD(x) (ADDR(x) - VLOAD_OFFSET) @@ -234,35 +204,29 @@ SECTIONS #endif /* CONFIG_X86_64 */ - /* init_task */ - . = ALIGN(THREAD_SIZE); - .data.init_task : AT(ADDR(.data.init_task) - LOAD_OFFSET) { - *(.data.init_task) + /* Init code and data - will be freed after init */ + . = ALIGN(PAGE_SIZE); + .init.begin : AT(ADDR(.init.begin) - LOAD_OFFSET) { + __init_begin = .; /* paired with __init_end */ } -#ifdef CONFIG_X86_64 - :data.init -#endif +#if defined(CONFIG_X86_64) && defined(CONFIG_SMP) /* - * smp_locks might be freed after init - * start/end must be page aligned + * percpu offsets are zero-based on SMP. PERCPU_VADDR() changes the + * output PHDR, so the next output section - .init.text - should + * start another segment - init. */ - . = ALIGN(PAGE_SIZE); - .smp_locks : AT(ADDR(.smp_locks) - LOAD_OFFSET) { - __smp_locks = .; - *(.smp_locks) - __smp_locks_end = .; - . = ALIGN(PAGE_SIZE); - } + PERCPU_VADDR(0, :percpu) +#endif - /* Init code and data - will be freed after init */ - . = ALIGN(PAGE_SIZE); .init.text : AT(ADDR(.init.text) - LOAD_OFFSET) { - __init_begin = .; /* paired with __init_end */ _sinittext = .; INIT_TEXT _einittext = .; } +#ifdef CONFIG_X86_64 + :init +#endif .init.data : AT(ADDR(.init.data) - LOAD_OFFSET) { INIT_DATA @@ -333,17 +297,7 @@ SECTIONS } #endif -#if defined(CONFIG_X86_64) && defined(CONFIG_SMP) - /* - * percpu offsets are zero-based on SMP. PERCPU_VADDR() changes the - * output PHDR, so the next output section - __data_nosave - should - * start another section data.init2. Also, pda should be at the head of - * percpu area. Preallocate it and define the percpu offset symbol - * so that it can be accessed as a percpu variable. - */ - . = ALIGN(PAGE_SIZE); - PERCPU_VADDR(0, :percpu) -#else +#if !defined(CONFIG_X86_64) || !defined(CONFIG_SMP) PERCPU(PAGE_SIZE) #endif @@ -354,15 +308,22 @@ SECTIONS __init_end = .; } + /* + * smp_locks might be freed after init + * start/end must be page aligned + */ + . = ALIGN(PAGE_SIZE); + .smp_locks : AT(ADDR(.smp_locks) - LOAD_OFFSET) { + __smp_locks = .; + *(.smp_locks) + __smp_locks_end = .; + . = ALIGN(PAGE_SIZE); + } + #ifdef CONFIG_X86_64 .data_nosave : AT(ADDR(.data_nosave) - LOAD_OFFSET) { - . = ALIGN(PAGE_SIZE); - __nosave_begin = .; - *(.data.nosave) - . = ALIGN(PAGE_SIZE); - __nosave_end = .; - } :data.init2 - /* use another section data.init2, see PERCPU_VADDR() above */ + NOSAVE_DATA + } #endif /* BSS */ @@ -400,8 +361,8 @@ SECTIONS #ifdef CONFIG_X86_32 -ASSERT((_end - LOAD_OFFSET <= KERNEL_IMAGE_SIZE), - "kernel image bigger than KERNEL_IMAGE_SIZE") +. = ASSERT((_end - LOAD_OFFSET <= KERNEL_IMAGE_SIZE), + "kernel image bigger than KERNEL_IMAGE_SIZE"); #else /* * Per-cpu symbols which need to be offset from __per_cpu_load @@ -414,12 +375,12 @@ INIT_PER_CPU(irq_stack_union); /* * Build-time check on the image size: */ -ASSERT((_end - _text <= KERNEL_IMAGE_SIZE), - "kernel image bigger than KERNEL_IMAGE_SIZE") +. = ASSERT((_end - _text <= KERNEL_IMAGE_SIZE), + "kernel image bigger than KERNEL_IMAGE_SIZE"); #ifdef CONFIG_SMP -ASSERT((per_cpu__irq_stack_union == 0), - "irq_stack_union is not at start of per-cpu area"); +. = ASSERT((per_cpu__irq_stack_union == 0), + "irq_stack_union is not at start of per-cpu area"); #endif #endif /* CONFIG_X86_32 */ @@ -427,7 +388,7 @@ ASSERT((per_cpu__irq_stack_union == 0), #ifdef CONFIG_KEXEC #include <asm/kexec.h> -ASSERT(kexec_control_code_size <= KEXEC_CONTROL_CODE_MAX_SIZE, - "kexec control code size is too big") +. = ASSERT(kexec_control_code_size <= KEXEC_CONTROL_CODE_MAX_SIZE, + "kexec control code size is too big"); #endif diff --git a/arch/x86/kvm/i8254.c b/arch/x86/kvm/i8254.c index 4d6f0d2..21f68e0 100644 --- a/arch/x86/kvm/i8254.c +++ b/arch/x86/kvm/i8254.c @@ -104,6 +104,9 @@ static s64 __kpit_elapsed(struct kvm *kvm) ktime_t remaining; struct kvm_kpit_state *ps = &kvm->arch.vpit->pit_state; + if (!ps->pit_timer.period) + return 0; + /* * The Counter does not stop when it reaches zero. In * Modes 0, 1, 4, and 5 the Counter ``wraps around'' to diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index 5c3d6e8..0ef5bb2 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -489,16 +489,20 @@ static unsigned long *gfn_to_rmap(struct kvm *kvm, gfn_t gfn, int lpage) * * If rmapp bit zero is one, (then rmap & ~1) points to a struct kvm_rmap_desc * containing more mappings. + * + * Returns the number of rmap entries before the spte was added or zero if + * the spte was not added. + * */ -static void rmap_add(struct kvm_vcpu *vcpu, u64 *spte, gfn_t gfn, int lpage) +static int rmap_add(struct kvm_vcpu *vcpu, u64 *spte, gfn_t gfn, int lpage) { struct kvm_mmu_page *sp; struct kvm_rmap_desc *desc; unsigned long *rmapp; - int i; + int i, count = 0; if (!is_rmap_pte(*spte)) - return; + return count; gfn = unalias_gfn(vcpu->kvm, gfn); sp = page_header(__pa(spte)); sp->gfns[spte - sp->spt] = gfn; @@ -515,8 +519,10 @@ static void rmap_add(struct kvm_vcpu *vcpu, u64 *spte, gfn_t gfn, int lpage) } else { rmap_printk("rmap_add: %p %llx many->many\n", spte, *spte); desc = (struct kvm_rmap_desc *)(*rmapp & ~1ul); - while (desc->shadow_ptes[RMAP_EXT-1] && desc->more) + while (desc->shadow_ptes[RMAP_EXT-1] && desc->more) { desc = desc->more; + count += RMAP_EXT; + } if (desc->shadow_ptes[RMAP_EXT-1]) { desc->more = mmu_alloc_rmap_desc(vcpu); desc = desc->more; @@ -525,6 +531,7 @@ static void rmap_add(struct kvm_vcpu *vcpu, u64 *spte, gfn_t gfn, int lpage) ; desc->shadow_ptes[i] = spte; } + return count; } static void rmap_desc_remove_entry(unsigned long *rmapp, @@ -754,6 +761,19 @@ static int kvm_age_rmapp(struct kvm *kvm, unsigned long *rmapp) return young; } +#define RMAP_RECYCLE_THRESHOLD 1000 + +static void rmap_recycle(struct kvm_vcpu *vcpu, gfn_t gfn, int lpage) +{ + unsigned long *rmapp; + + gfn = unalias_gfn(vcpu->kvm, gfn); + rmapp = gfn_to_rmap(vcpu->kvm, gfn, lpage); + + kvm_unmap_rmapp(vcpu->kvm, rmapp); + kvm_flush_remote_tlbs(vcpu->kvm); +} + int kvm_age_hva(struct kvm *kvm, unsigned long hva) { return kvm_handle_hva(kvm, hva, kvm_age_rmapp); @@ -1407,24 +1427,25 @@ static int kvm_mmu_zap_page(struct kvm *kvm, struct kvm_mmu_page *sp) */ void kvm_mmu_change_mmu_pages(struct kvm *kvm, unsigned int kvm_nr_mmu_pages) { + int used_pages; + + used_pages = kvm->arch.n_alloc_mmu_pages - kvm->arch.n_free_mmu_pages; + used_pages = max(0, used_pages); + /* * If we set the number of mmu pages to be smaller be than the * number of actived pages , we must to free some mmu pages before we * change the value */ - if ((kvm->arch.n_alloc_mmu_pages - kvm->arch.n_free_mmu_pages) > - kvm_nr_mmu_pages) { - int n_used_mmu_pages = kvm->arch.n_alloc_mmu_pages - - kvm->arch.n_free_mmu_pages; - - while (n_used_mmu_pages > kvm_nr_mmu_pages) { + if (used_pages > kvm_nr_mmu_pages) { + while (used_pages > kvm_nr_mmu_pages) { struct kvm_mmu_page *page; page = container_of(kvm->arch.active_mmu_pages.prev, struct kvm_mmu_page, link); kvm_mmu_zap_page(kvm, page); - n_used_mmu_pages--; + used_pages--; } kvm->arch.n_free_mmu_pages = 0; } @@ -1740,6 +1761,7 @@ static void mmu_set_spte(struct kvm_vcpu *vcpu, u64 *shadow_pte, { int was_rmapped = 0; int was_writeble = is_writeble_pte(*shadow_pte); + int rmap_count; pgprintk("%s: spte %llx access %x write_fault %d" " user_fault %d gfn %lx\n", @@ -1781,9 +1803,11 @@ static void mmu_set_spte(struct kvm_vcpu *vcpu, u64 *shadow_pte, page_header_update_slot(vcpu->kvm, shadow_pte, gfn); if (!was_rmapped) { - rmap_add(vcpu, shadow_pte, gfn, largepage); + rmap_count = rmap_add(vcpu, shadow_pte, gfn, largepage); if (!is_rmap_pte(*shadow_pte)) kvm_release_pfn_clean(pfn); + if (rmap_count > RMAP_RECYCLE_THRESHOLD) + rmap_recycle(vcpu, gfn, largepage); } else { if (was_writeble) kvm_release_pfn_dirty(pfn); @@ -2157,7 +2181,7 @@ static void reset_rsvds_bits_mask(struct kvm_vcpu *vcpu, int level) else /* 32 bits PSE 4MB page */ context->rsvd_bits_mask[1][1] = rsvd_bits(13, 21); - context->rsvd_bits_mask[1][0] = ~0ull; + context->rsvd_bits_mask[1][0] = context->rsvd_bits_mask[1][0]; break; case PT32E_ROOT_LEVEL: context->rsvd_bits_mask[0][2] = @@ -2170,7 +2194,7 @@ static void reset_rsvds_bits_mask(struct kvm_vcpu *vcpu, int level) context->rsvd_bits_mask[1][1] = exb_bit_rsvd | rsvd_bits(maxphyaddr, 62) | rsvd_bits(13, 20); /* large page */ - context->rsvd_bits_mask[1][0] = ~0ull; + context->rsvd_bits_mask[1][0] = context->rsvd_bits_mask[1][0]; break; case PT64_ROOT_LEVEL: context->rsvd_bits_mask[0][3] = exb_bit_rsvd | @@ -2186,7 +2210,7 @@ static void reset_rsvds_bits_mask(struct kvm_vcpu *vcpu, int level) context->rsvd_bits_mask[1][1] = exb_bit_rsvd | rsvd_bits(maxphyaddr, 51) | rsvd_bits(13, 20); /* large page */ - context->rsvd_bits_mask[1][0] = ~0ull; + context->rsvd_bits_mask[1][0] = context->rsvd_bits_mask[1][0]; break; } } diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h index 258e459..67785f6 100644 --- a/arch/x86/kvm/paging_tmpl.h +++ b/arch/x86/kvm/paging_tmpl.h @@ -281,7 +281,7 @@ static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr, { unsigned access = gw->pt_access; struct kvm_mmu_page *shadow_page; - u64 spte, *sptep; + u64 spte, *sptep = NULL; int direct; gfn_t table_gfn; int r; diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 71510e0..b1f658a 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -711,6 +711,7 @@ static void svm_vcpu_load(struct kvm_vcpu *vcpu, int cpu) svm->vmcb->control.tsc_offset += delta; vcpu->cpu = cpu; kvm_migrate_timers(vcpu); + svm->asid_generation = 0; } for (i = 0; i < NR_HOST_SAVE_USER_MSRS; i++) @@ -1031,7 +1032,6 @@ static void new_asid(struct vcpu_svm *svm, struct svm_cpu_data *svm_data) svm->vmcb->control.tlb_ctl = TLB_CONTROL_FLUSH_ALL_ASID; } - svm->vcpu.cpu = svm_data->cpu; svm->asid_generation = svm_data->asid_generation; svm->vmcb->control.asid = svm_data->next_asid++; } @@ -2300,8 +2300,8 @@ static void pre_svm_run(struct vcpu_svm *svm) struct svm_cpu_data *svm_data = per_cpu(svm_data, cpu); svm->vmcb->control.tlb_ctl = TLB_CONTROL_DO_NOTHING; - if (svm->vcpu.cpu != cpu || - svm->asid_generation != svm_data->asid_generation) + /* FIXME: handle wraparound of asid_generation */ + if (svm->asid_generation != svm_data->asid_generation) new_asid(svm, svm_data); } diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index e770bf3..29f9129 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -3012,6 +3012,12 @@ static int handle_vmcall(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) return 1; } +static int handle_vmx_insn(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) +{ + kvm_queue_exception(vcpu, UD_VECTOR); + return 1; +} + static int handle_invlpg(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) { unsigned long exit_qualification = vmcs_readl(EXIT_QUALIFICATION); @@ -3151,8 +3157,8 @@ static void handle_invalid_guest_state(struct kvm_vcpu *vcpu, struct vcpu_vmx *vmx = to_vmx(vcpu); enum emulation_result err = EMULATE_DONE; - preempt_enable(); local_irq_enable(); + preempt_enable(); while (!guest_state_valid(vcpu)) { err = emulate_instruction(vcpu, kvm_run, 0, 0, 0); @@ -3162,7 +3168,7 @@ static void handle_invalid_guest_state(struct kvm_vcpu *vcpu, if (err != EMULATE_DONE) { kvm_report_emulation_failure(vcpu, "emulation failure"); - return; + break; } if (signal_pending(current)) @@ -3171,8 +3177,8 @@ static void handle_invalid_guest_state(struct kvm_vcpu *vcpu, schedule(); } - local_irq_disable(); preempt_disable(); + local_irq_disable(); vmx->invalid_state_emulation_result = err; } @@ -3198,6 +3204,15 @@ static int (*kvm_vmx_exit_handlers[])(struct kvm_vcpu *vcpu, [EXIT_REASON_HLT] = handle_halt, [EXIT_REASON_INVLPG] = handle_invlpg, [EXIT_REASON_VMCALL] = handle_vmcall, + [EXIT_REASON_VMCLEAR] = handle_vmx_insn, + [EXIT_REASON_VMLAUNCH] = handle_vmx_insn, + [EXIT_REASON_VMPTRLD] = handle_vmx_insn, + [EXIT_REASON_VMPTRST] = handle_vmx_insn, + [EXIT_REASON_VMREAD] = handle_vmx_insn, + [EXIT_REASON_VMRESUME] = handle_vmx_insn, + [EXIT_REASON_VMWRITE] = handle_vmx_insn, + [EXIT_REASON_VMOFF] = handle_vmx_insn, + [EXIT_REASON_VMON] = handle_vmx_insn, [EXIT_REASON_TPR_BELOW_THRESHOLD] = handle_tpr_below_threshold, [EXIT_REASON_APIC_ACCESS] = handle_apic_access, [EXIT_REASON_WBINVD] = handle_wbinvd, diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 249540f..3d45290 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -704,11 +704,48 @@ static bool msr_mtrr_valid(unsigned msr) return false; } +static bool valid_pat_type(unsigned t) +{ + return t < 8 && (1 << t) & 0xf3; /* 0, 1, 4, 5, 6, 7 */ +} + +static bool valid_mtrr_type(unsigned t) +{ + return t < 8 && (1 << t) & 0x73; /* 0, 1, 4, 5, 6 */ +} + +static bool mtrr_valid(struct kvm_vcpu *vcpu, u32 msr, u64 data) +{ + int i; + + if (!msr_mtrr_valid(msr)) + return false; + + if (msr == MSR_IA32_CR_PAT) { + for (i = 0; i < 8; i++) + if (!valid_pat_type((data >> (i * 8)) & 0xff)) + return false; + return true; + } else if (msr == MSR_MTRRdefType) { + if (data & ~0xcff) + return false; + return valid_mtrr_type(data & 0xff); + } else if (msr >= MSR_MTRRfix64K_00000 && msr <= MSR_MTRRfix4K_F8000) { + for (i = 0; i < 8 ; i++) + if (!valid_mtrr_type((data >> (i * 8)) & 0xff)) + return false; + return true; + } + + /* variable MTRRs */ + return valid_mtrr_type(data & 0xff); +} + static int set_msr_mtrr(struct kvm_vcpu *vcpu, u32 msr, u64 data) { u64 *p = (u64 *)&vcpu->arch.mtrr_state.fixed_ranges; - if (!msr_mtrr_valid(msr)) + if (!mtrr_valid(vcpu, msr, data)) return 1; if (msr == MSR_MTRRdefType) { @@ -898,6 +935,7 @@ int kvm_get_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata) case MSR_VM_HSAVE_PA: case MSR_P6_EVNTSEL0: case MSR_P6_EVNTSEL1: + case MSR_K7_EVNTSEL0: data = 0; break; case MSR_MTRRcap: @@ -1078,14 +1116,13 @@ long kvm_arch_dev_ioctl(struct file *filp, if (copy_to_user(user_msr_list, &msr_list, sizeof msr_list)) goto out; r = -E2BIG; - if (n < num_msrs_to_save) + if (n < msr_list.nmsrs) goto out; r = -EFAULT; if (copy_to_user(user_msr_list->indices, &msrs_to_save, num_msrs_to_save * sizeof(u32))) goto out; - if (copy_to_user(user_msr_list->indices - + num_msrs_to_save * sizeof(u32), + if (copy_to_user(user_msr_list->indices + num_msrs_to_save, &emulated_msrs, ARRAY_SIZE(emulated_msrs) * sizeof(u32))) goto out; diff --git a/arch/x86/kvm/x86_emulate.c b/arch/x86/kvm/x86_emulate.c index c1b6c23..616de46 100644 --- a/arch/x86/kvm/x86_emulate.c +++ b/arch/x86/kvm/x86_emulate.c @@ -1361,7 +1361,7 @@ static inline int writeback(struct x86_emulate_ctxt *ctxt, return 0; } -void toggle_interruptibility(struct x86_emulate_ctxt *ctxt, u32 mask) +static void toggle_interruptibility(struct x86_emulate_ctxt *ctxt, u32 mask) { u32 int_shadow = kvm_x86_ops->get_interrupt_shadow(ctxt->vcpu, mask); /* diff --git a/arch/x86/lguest/boot.c b/arch/x86/lguest/boot.c index 7bc65f0..d677fa9 100644 --- a/arch/x86/lguest/boot.c +++ b/arch/x86/lguest/boot.c @@ -22,7 +22,8 @@ * * So how does the kernel know it's a Guest? We'll see that later, but let's * just say that we end up here where we replace the native functions various - * "paravirt" structures with our Guest versions, then boot like normal. :*/ + * "paravirt" structures with our Guest versions, then boot like normal. +:*/ /* * Copyright (C) 2006, Rusty Russell <rusty@rustcorp.com.au> IBM Corporation. @@ -74,7 +75,8 @@ * * The Guest in our tale is a simple creature: identical to the Host but * behaving in simplified but equivalent ways. In particular, the Guest is the - * same kernel as the Host (or at least, built from the same source code). :*/ + * same kernel as the Host (or at least, built from the same source code). +:*/ struct lguest_data lguest_data = { .hcall_status = { [0 ... LHCALL_RING_SIZE-1] = 0xFF }, @@ -85,7 +87,8 @@ struct lguest_data lguest_data = { .syscall_vec = SYSCALL_VECTOR, }; -/*G:037 async_hcall() is pretty simple: I'm quite proud of it really. We have a +/*G:037 + * async_hcall() is pretty simple: I'm quite proud of it really. We have a * ring buffer of stored hypercalls which the Host will run though next time we * do a normal hypercall. Each entry in the ring has 5 slots for the hypercall * arguments, and a "hcall_status" word which is 0 if the call is ready to go, @@ -94,7 +97,8 @@ struct lguest_data lguest_data = { * If we come around to a slot which hasn't been finished, then the table is * full and we just make the hypercall directly. This has the nice side * effect of causing the Host to run all the stored calls in the ring buffer - * which empties it for next time! */ + * which empties it for next time! + */ static void async_hcall(unsigned long call, unsigned long arg1, unsigned long arg2, unsigned long arg3, unsigned long arg4) @@ -103,9 +107,11 @@ static void async_hcall(unsigned long call, unsigned long arg1, static unsigned int next_call; unsigned long flags; - /* Disable interrupts if not already disabled: we don't want an + /* + * Disable interrupts if not already disabled: we don't want an * interrupt handler making a hypercall while we're already doing - * one! */ + * one! + */ local_irq_save(flags); if (lguest_data.hcall_status[next_call] != 0xFF) { /* Table full, so do normal hcall which will flush table. */ @@ -125,8 +131,9 @@ static void async_hcall(unsigned long call, unsigned long arg1, local_irq_restore(flags); } -/*G:035 Notice the lazy_hcall() above, rather than hcall(). This is our first - * real optimization trick! +/*G:035 + * Notice the lazy_hcall() above, rather than hcall(). This is our first real + * optimization trick! * * When lazy_mode is set, it means we're allowed to defer all hypercalls and do * them as a batch when lazy_mode is eventually turned off. Because hypercalls @@ -136,7 +143,8 @@ static void async_hcall(unsigned long call, unsigned long arg1, * lguest_leave_lazy_mode(). * * So, when we're in lazy mode, we call async_hcall() to store the call for - * future processing: */ + * future processing: + */ static void lazy_hcall1(unsigned long call, unsigned long arg1) { @@ -146,6 +154,7 @@ static void lazy_hcall1(unsigned long call, async_hcall(call, arg1, 0, 0, 0); } +/* You can imagine what lazy_hcall2, 3 and 4 look like. :*/ static void lazy_hcall2(unsigned long call, unsigned long arg1, unsigned long arg2) @@ -181,8 +190,10 @@ static void lazy_hcall4(unsigned long call, } #endif -/* When lazy mode is turned off reset the per-cpu lazy mode variable and then - * issue the do-nothing hypercall to flush any stored calls. */ +/*G:036 + * When lazy mode is turned off reset the per-cpu lazy mode variable and then + * issue the do-nothing hypercall to flush any stored calls. +:*/ static void lguest_leave_lazy_mmu_mode(void) { kvm_hypercall0(LHCALL_FLUSH_ASYNC); @@ -208,9 +219,11 @@ static void lguest_end_context_switch(struct task_struct *next) * check there before it tries to deliver an interrupt. */ -/* save_flags() is expected to return the processor state (ie. "flags"). The +/* + * save_flags() is expected to return the processor state (ie. "flags"). The * flags word contains all kind of stuff, but in practice Linux only cares - * about the interrupt flag. Our "save_flags()" just returns that. */ + * about the interrupt flag. Our "save_flags()" just returns that. + */ static unsigned long save_fl(void) { return lguest_data.irq_enabled; @@ -222,13 +235,15 @@ static void irq_disable(void) lguest_data.irq_enabled = 0; } -/* Let's pause a moment. Remember how I said these are called so often? +/* + * Let's pause a moment. Remember how I said these are called so often? * Jeremy Fitzhardinge optimized them so hard early in 2009 that he had to * break some rules. In particular, these functions are assumed to save their * own registers if they need to: normal C functions assume they can trash the * eax register. To use normal C functions, we use * PV_CALLEE_SAVE_REGS_THUNK(), which pushes %eax onto the stack, calls the - * C function, then restores it. */ + * C function, then restores it. + */ PV_CALLEE_SAVE_REGS_THUNK(save_fl); PV_CALLEE_SAVE_REGS_THUNK(irq_disable); /*:*/ @@ -237,18 +252,18 @@ PV_CALLEE_SAVE_REGS_THUNK(irq_disable); extern void lg_irq_enable(void); extern void lg_restore_fl(unsigned long flags); -/*M:003 Note that we don't check for outstanding interrupts when we re-enable - * them (or when we unmask an interrupt). This seems to work for the moment, - * since interrupts are rare and we'll just get the interrupt on the next timer - * tick, but now we can run with CONFIG_NO_HZ, we should revisit this. One way - * would be to put the "irq_enabled" field in a page by itself, and have the - * Host write-protect it when an interrupt comes in when irqs are disabled. - * There will then be a page fault as soon as interrupts are re-enabled. +/*M:003 + * We could be more efficient in our checking of outstanding interrupts, rather + * than using a branch. One way would be to put the "irq_enabled" field in a + * page by itself, and have the Host write-protect it when an interrupt comes + * in when irqs are disabled. There will then be a page fault as soon as + * interrupts are re-enabled. * * A better method is to implement soft interrupt disable generally for x86: * instead of disabling interrupts, we set a flag. If an interrupt does come * in, we then disable them for real. This is uncommon, so we could simply use - * a hypercall for interrupt control and not worry about efficiency. :*/ + * a hypercall for interrupt control and not worry about efficiency. +:*/ /*G:034 * The Interrupt Descriptor Table (IDT). @@ -261,10 +276,12 @@ extern void lg_restore_fl(unsigned long flags); static void lguest_write_idt_entry(gate_desc *dt, int entrynum, const gate_desc *g) { - /* The gate_desc structure is 8 bytes long: we hand it to the Host in + /* + * The gate_desc structure is 8 bytes long: we hand it to the Host in * two 32-bit chunks. The whole 32-bit kernel used to hand descriptors * around like this; typesafety wasn't a big concern in Linux's early - * years. */ + * years. + */ u32 *desc = (u32 *)g; /* Keep the local copy up to date. */ native_write_idt_entry(dt, entrynum, g); @@ -272,9 +289,11 @@ static void lguest_write_idt_entry(gate_desc *dt, kvm_hypercall3(LHCALL_LOAD_IDT_ENTRY, entrynum, desc[0], desc[1]); } -/* Changing to a different IDT is very rare: we keep the IDT up-to-date every +/* + * Changing to a different IDT is very rare: we keep the IDT up-to-date every * time it is written, so we can simply loop through all entries and tell the - * Host about them. */ + * Host about them. + */ static void lguest_load_idt(const struct desc_ptr *desc) { unsigned int i; @@ -305,9 +324,11 @@ static void lguest_load_gdt(const struct desc_ptr *desc) kvm_hypercall3(LHCALL_LOAD_GDT_ENTRY, i, gdt[i].a, gdt[i].b); } -/* For a single GDT entry which changes, we do the lazy thing: alter our GDT, +/* + * For a single GDT entry which changes, we do the lazy thing: alter our GDT, * then tell the Host to reload the entire thing. This operation is so rare - * that this naive implementation is reasonable. */ + * that this naive implementation is reasonable. + */ static void lguest_write_gdt_entry(struct desc_struct *dt, int entrynum, const void *desc, int type) { @@ -317,29 +338,36 @@ static void lguest_write_gdt_entry(struct desc_struct *dt, int entrynum, dt[entrynum].a, dt[entrynum].b); } -/* OK, I lied. There are three "thread local storage" GDT entries which change +/* + * OK, I lied. There are three "thread local storage" GDT entries which change * on every context switch (these three entries are how glibc implements - * __thread variables). So we have a hypercall specifically for this case. */ + * __thread variables). So we have a hypercall specifically for this case. + */ static void lguest_load_tls(struct thread_struct *t, unsigned int cpu) { - /* There's one problem which normal hardware doesn't have: the Host + /* + * There's one problem which normal hardware doesn't have: the Host * can't handle us removing entries we're currently using. So we clear - * the GS register here: if it's needed it'll be reloaded anyway. */ + * the GS register here: if it's needed it'll be reloaded anyway. + */ lazy_load_gs(0); lazy_hcall2(LHCALL_LOAD_TLS, __pa(&t->tls_array), cpu); } -/*G:038 That's enough excitement for now, back to ploughing through each of - * the different pv_ops structures (we're about 1/3 of the way through). +/*G:038 + * That's enough excitement for now, back to ploughing through each of the + * different pv_ops structures (we're about 1/3 of the way through). * * This is the Local Descriptor Table, another weird Intel thingy. Linux only * uses this for some strange applications like Wine. We don't do anything - * here, so they'll get an informative and friendly Segmentation Fault. */ + * here, so they'll get an informative and friendly Segmentation Fault. + */ static void lguest_set_ldt(const void *addr, unsigned entries) { } -/* This loads a GDT entry into the "Task Register": that entry points to a +/* + * This loads a GDT entry into the "Task Register": that entry points to a * structure called the Task State Segment. Some comments scattered though the * kernel code indicate that this used for task switching in ages past, along * with blood sacrifice and astrology. @@ -347,19 +375,21 @@ static void lguest_set_ldt(const void *addr, unsigned entries) * Now there's nothing interesting in here that we don't get told elsewhere. * But the native version uses the "ltr" instruction, which makes the Host * complain to the Guest about a Segmentation Fault and it'll oops. So we - * override the native version with a do-nothing version. */ + * override the native version with a do-nothing version. + */ static void lguest_load_tr_desc(void) { } -/* The "cpuid" instruction is a way of querying both the CPU identity +/* + * The "cpuid" instruction is a way of querying both the CPU identity * (manufacturer, model, etc) and its features. It was introduced before the * Pentium in 1993 and keeps getting extended by both Intel, AMD and others. * As you might imagine, after a decade and a half this treatment, it is now a * giant ball of hair. Its entry in the current Intel manual runs to 28 pages. * * This instruction even it has its own Wikipedia entry. The Wikipedia entry - * has been translated into 4 languages. I am not making this up! + * has been translated into 5 languages. I am not making this up! * * We could get funky here and identify ourselves as "GenuineLguest", but * instead we just use the real "cpuid" instruction. Then I pretty much turned @@ -371,7 +401,8 @@ static void lguest_load_tr_desc(void) * Replacing the cpuid so we can turn features off is great for the kernel, but * anyone (including userspace) can just use the raw "cpuid" instruction and * the Host won't even notice since it isn't privileged. So we try not to get - * too worked up about it. */ + * too worked up about it. + */ static void lguest_cpuid(unsigned int *ax, unsigned int *bx, unsigned int *cx, unsigned int *dx) { @@ -379,38 +410,63 @@ static void lguest_cpuid(unsigned int *ax, unsigned int *bx, native_cpuid(ax, bx, cx, dx); switch (function) { - case 1: /* Basic feature request. */ - /* We only allow kernel to see SSE3, CMPXCHG16B and SSSE3 */ + /* + * CPUID 0 gives the highest legal CPUID number (and the ID string). + * We futureproof our code a little by sticking to known CPUID values. + */ + case 0: + if (*ax > 5) + *ax = 5; + break; + + /* + * CPUID 1 is a basic feature request. + * + * CX: we only allow kernel to see SSE3, CMPXCHG16B and SSSE3 + * DX: SSE, SSE2, FXSR, MMX, CMOV, CMPXCHG8B, TSC, FPU and PAE. + */ + case 1: *cx &= 0x00002201; - /* SSE, SSE2, FXSR, MMX, CMOV, CMPXCHG8B, TSC, FPU, PAE. */ *dx &= 0x07808151; - /* The Host can do a nice optimization if it knows that the + /* + * The Host can do a nice optimization if it knows that the * kernel mappings (addresses above 0xC0000000 or whatever * PAGE_OFFSET is set to) haven't changed. But Linux calls * flush_tlb_user() for both user and kernel mappings unless - * the Page Global Enable (PGE) feature bit is set. */ + * the Page Global Enable (PGE) feature bit is set. + */ *dx |= 0x00002000; - /* We also lie, and say we're family id 5. 6 or greater + /* + * We also lie, and say we're family id 5. 6 or greater * leads to a rdmsr in early_init_intel which we can't handle. - * Family ID is returned as bits 8-12 in ax. */ + * Family ID is returned as bits 8-12 in ax. + */ *ax &= 0xFFFFF0FF; *ax |= 0x00000500; break; + /* + * 0x80000000 returns the highest Extended Function, so we futureproof + * like we do above by limiting it to known fields. + */ case 0x80000000: - /* Futureproof this a little: if they ask how much extended - * processor information there is, limit it to known fields. */ if (*ax > 0x80000008) *ax = 0x80000008; break; + + /* + * PAE systems can mark pages as non-executable. Linux calls this the + * NX bit. Intel calls it XD (eXecute Disable), AMD EVP (Enhanced + * Virus Protection). We just switch turn if off here, since we don't + * support it. + */ case 0x80000001: - /* Here we should fix nx cap depending on host. */ - /* For this version of PAE, we just clear NX bit. */ *dx &= ~(1 << 20); break; } } -/* Intel has four control registers, imaginatively named cr0, cr2, cr3 and cr4. +/* + * Intel has four control registers, imaginatively named cr0, cr2, cr3 and cr4. * I assume there's a cr1, but it hasn't bothered us yet, so we'll not bother * it. The Host needs to know when the Guest wants to change them, so we have * a whole series of functions like read_cr0() and write_cr0(). @@ -425,7 +481,8 @@ static void lguest_cpuid(unsigned int *ax, unsigned int *bx, * name like "FPUTRAP bit" be a little less cryptic? * * We store cr0 locally because the Host never changes it. The Guest sometimes - * wants to read it and we'd prefer not to bother the Host unnecessarily. */ + * wants to read it and we'd prefer not to bother the Host unnecessarily. + */ static unsigned long current_cr0; static void lguest_write_cr0(unsigned long val) { @@ -438,18 +495,22 @@ static unsigned long lguest_read_cr0(void) return current_cr0; } -/* Intel provided a special instruction to clear the TS bit for people too cool +/* + * Intel provided a special instruction to clear the TS bit for people too cool * to use write_cr0() to do it. This "clts" instruction is faster, because all - * the vowels have been optimized out. */ + * the vowels have been optimized out. + */ static void lguest_clts(void) { lazy_hcall1(LHCALL_TS, 0); current_cr0 &= ~X86_CR0_TS; } -/* cr2 is the virtual address of the last page fault, which the Guest only ever +/* + * cr2 is the virtual address of the last page fault, which the Guest only ever * reads. The Host kindly writes this into our "struct lguest_data", so we - * just read it out of there. */ + * just read it out of there. + */ static unsigned long lguest_read_cr2(void) { return lguest_data.cr2; @@ -458,10 +519,12 @@ static unsigned long lguest_read_cr2(void) /* See lguest_set_pte() below. */ static bool cr3_changed = false; -/* cr3 is the current toplevel pagetable page: the principle is the same as +/* + * cr3 is the current toplevel pagetable page: the principle is the same as * cr0. Keep a local copy, and tell the Host when it changes. The only * difference is that our local copy is in lguest_data because the Host needs - * to set it upon our initial hypercall. */ + * to set it upon our initial hypercall. + */ static void lguest_write_cr3(unsigned long cr3) { lguest_data.pgdir = cr3; @@ -506,7 +569,7 @@ static void lguest_write_cr4(unsigned long val) * cr3 ---> +---------+ * | --------->+---------+ * | | | PADDR1 | - * Top-level | | PADDR2 | + * Mid-level | | PADDR2 | * (PMD) page | | | * | | Lower-level | * | | (PTE) page | @@ -526,21 +589,62 @@ static void lguest_write_cr4(unsigned long val) * Index into top Index into second Offset within page * page directory page pagetable page * - * The kernel spends a lot of time changing both the top-level page directory - * and lower-level pagetable pages. The Guest doesn't know physical addresses, - * so while it maintains these page tables exactly like normal, it also needs - * to keep the Host informed whenever it makes a change: the Host will create - * the real page tables based on the Guests'. + * Now, unfortunately, this isn't the whole story: Intel added Physical Address + * Extension (PAE) to allow 32 bit systems to use 64GB of memory (ie. 36 bits). + * These are held in 64-bit page table entries, so we can now only fit 512 + * entries in a page, and the neat three-level tree breaks down. + * + * The result is a four level page table: + * + * cr3 --> [ 4 Upper ] + * [ Level ] + * [ Entries ] + * [(PUD Page)]---> +---------+ + * | --------->+---------+ + * | | | PADDR1 | + * Mid-level | | PADDR2 | + * (PMD) page | | | + * | | Lower-level | + * | | (PTE) page | + * | | | | + * .... .... + * + * + * And the virtual address is decoded as: + * + * 1 1 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 + * |<-2->|<--- 9 bits ---->|<---- 9 bits --->|<------ 12 bits ------>| + * Index into Index into mid Index into lower Offset within page + * top entries directory page pagetable page + * + * It's too hard to switch between these two formats at runtime, so Linux only + * supports one or the other depending on whether CONFIG_X86_PAE is set. Many + * distributions turn it on, and not just for people with silly amounts of + * memory: the larger PTE entries allow room for the NX bit, which lets the + * kernel disable execution of pages and increase security. + * + * This was a problem for lguest, which couldn't run on these distributions; + * then Matias Zabaljauregui figured it all out and implemented it, and only a + * handful of puppies were crushed in the process! + * + * Back to our point: the kernel spends a lot of time changing both the + * top-level page directory and lower-level pagetable pages. The Guest doesn't + * know physical addresses, so while it maintains these page tables exactly + * like normal, it also needs to keep the Host informed whenever it makes a + * change: the Host will create the real page tables based on the Guests'. */ -/* The Guest calls this to set a second-level entry (pte), ie. to map a page - * into a process' address space. We set the entry then tell the Host the - * toplevel and address this corresponds to. The Guest uses one pagetable per - * process, so we need to tell the Host which one we're changing (mm->pgd). */ +/* + * The Guest calls this after it has set a second-level entry (pte), ie. to map + * a page into a process' address space. Wetell the Host the toplevel and + * address this corresponds to. The Guest uses one pagetable per process, so + * we need to tell the Host which one we're changing (mm->pgd). + */ static void lguest_pte_update(struct mm_struct *mm, unsigned long addr, pte_t *ptep) { #ifdef CONFIG_X86_PAE + /* PAE needs to hand a 64 bit page table entry, so it uses two args. */ lazy_hcall4(LHCALL_SET_PTE, __pa(mm->pgd), addr, ptep->pte_low, ptep->pte_high); #else @@ -548,6 +652,7 @@ static void lguest_pte_update(struct mm_struct *mm, unsigned long addr, #endif } +/* This is the "set and update" combo-meal-deal version. */ static void lguest_set_pte_at(struct mm_struct *mm, unsigned long addr, pte_t *ptep, pte_t pteval) { @@ -555,10 +660,13 @@ static void lguest_set_pte_at(struct mm_struct *mm, unsigned long addr, lguest_pte_update(mm, addr, ptep); } -/* The Guest calls lguest_set_pud to set a top-level entry and lguest_set_pmd +/* + * The Guest calls lguest_set_pud to set a top-level entry and lguest_set_pmd * to set a middle-level entry when PAE is activated. + * * Again, we set the entry then tell the Host which page we changed, - * and the index of the entry we changed. */ + * and the index of the entry we changed. + */ #ifdef CONFIG_X86_PAE static void lguest_set_pud(pud_t *pudp, pud_t pudval) { @@ -577,8 +685,7 @@ static void lguest_set_pmd(pmd_t *pmdp, pmd_t pmdval) } #else -/* The Guest calls lguest_set_pmd to set a top-level entry when PAE is not - * activated. */ +/* The Guest calls lguest_set_pmd to set a top-level entry when !PAE. */ static void lguest_set_pmd(pmd_t *pmdp, pmd_t pmdval) { native_set_pmd(pmdp, pmdval); @@ -587,7 +694,8 @@ static void lguest_set_pmd(pmd_t *pmdp, pmd_t pmdval) } #endif -/* There are a couple of legacy places where the kernel sets a PTE, but we +/* + * There are a couple of legacy places where the kernel sets a PTE, but we * don't know the top level any more. This is useless for us, since we don't * know which pagetable is changing or what address, so we just tell the Host * to forget all of them. Fortunately, this is very rare. @@ -595,7 +703,8 @@ static void lguest_set_pmd(pmd_t *pmdp, pmd_t pmdval) * ... except in early boot when the kernel sets up the initial pagetables, * which makes booting astonishingly slow: 1.83 seconds! So we don't even tell * the Host anything changed until we've done the first page table switch, - * which brings boot back to 0.25 seconds. */ + * which brings boot back to 0.25 seconds. + */ static void lguest_set_pte(pte_t *ptep, pte_t pteval) { native_set_pte(ptep, pteval); @@ -604,6 +713,11 @@ static void lguest_set_pte(pte_t *ptep, pte_t pteval) } #ifdef CONFIG_X86_PAE +/* + * With 64-bit PTE values, we need to be careful setting them: if we set 32 + * bits at a time, the hardware could see a weird half-set entry. These + * versions ensure we update all 64 bits at once. + */ static void lguest_set_pte_atomic(pte_t *ptep, pte_t pte) { native_set_pte_atomic(ptep, pte); @@ -611,19 +725,21 @@ static void lguest_set_pte_atomic(pte_t *ptep, pte_t pte) lazy_hcall1(LHCALL_FLUSH_TLB, 1); } -void lguest_pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep) +static void lguest_pte_clear(struct mm_struct *mm, unsigned long addr, + pte_t *ptep) { native_pte_clear(mm, addr, ptep); lguest_pte_update(mm, addr, ptep); } -void lguest_pmd_clear(pmd_t *pmdp) +static void lguest_pmd_clear(pmd_t *pmdp) { lguest_set_pmd(pmdp, __pmd(0)); } #endif -/* Unfortunately for Lguest, the pv_mmu_ops for page tables were based on +/* + * Unfortunately for Lguest, the pv_mmu_ops for page tables were based on * native page table operations. On native hardware you can set a new page * table entry whenever you want, but if you want to remove one you have to do * a TLB flush (a TLB is a little cache of page table entries kept by the CPU). @@ -632,24 +748,29 @@ void lguest_pmd_clear(pmd_t *pmdp) * called when a valid entry is written, not when it's removed (ie. marked not * present). Instead, this is where we come when the Guest wants to remove a * page table entry: we tell the Host to set that entry to 0 (ie. the present - * bit is zero). */ + * bit is zero). + */ static void lguest_flush_tlb_single(unsigned long addr) { /* Simply set it to zero: if it was not, it will fault back in. */ lazy_hcall3(LHCALL_SET_PTE, lguest_data.pgdir, addr, 0); } -/* This is what happens after the Guest has removed a large number of entries. +/* + * This is what happens after the Guest has removed a large number of entries. * This tells the Host that any of the page table entries for userspace might - * have changed, ie. virtual addresses below PAGE_OFFSET. */ + * have changed, ie. virtual addresses below PAGE_OFFSET. + */ static void lguest_flush_tlb_user(void) { lazy_hcall1(LHCALL_FLUSH_TLB, 0); } -/* This is called when the kernel page tables have changed. That's not very +/* + * This is called when the kernel page tables have changed. That's not very * common (unless the Guest is using highmem, which makes the Guest extremely - * slow), so it's worth separating this from the user flushing above. */ + * slow), so it's worth separating this from the user flushing above. + */ static void lguest_flush_tlb_kernel(void) { lazy_hcall1(LHCALL_FLUSH_TLB, 1); @@ -686,26 +807,38 @@ static struct irq_chip lguest_irq_controller = { .unmask = enable_lguest_irq, }; -/* This sets up the Interrupt Descriptor Table (IDT) entry for each hardware +/* + * This sets up the Interrupt Descriptor Table (IDT) entry for each hardware * interrupt (except 128, which is used for system calls), and then tells the * Linux infrastructure that each interrupt is controlled by our level-based - * lguest interrupt controller. */ + * lguest interrupt controller. + */ static void __init lguest_init_IRQ(void) { unsigned int i; for (i = FIRST_EXTERNAL_VECTOR; i < NR_VECTORS; i++) { - /* Some systems map "vectors" to interrupts weirdly. Lguest has - * a straightforward 1 to 1 mapping, so force that here. */ + /* Some systems map "vectors" to interrupts weirdly. Not us! */ __get_cpu_var(vector_irq)[i] = i - FIRST_EXTERNAL_VECTOR; if (i != SYSCALL_VECTOR) set_intr_gate(i, interrupt[i - FIRST_EXTERNAL_VECTOR]); } - /* This call is required to set up for 4k stacks, where we have - * separate stacks for hard and soft interrupts. */ + + /* + * This call is required to set up for 4k stacks, where we have + * separate stacks for hard and soft interrupts. + */ irq_ctx_init(smp_processor_id()); } +/* + * With CONFIG_SPARSE_IRQ, interrupt descriptors are allocated as-needed, so + * rather than set them in lguest_init_IRQ we are called here every time an + * lguest device needs an interrupt. + * + * FIXME: irq_to_desc_alloc_node() can fail due to lack of memory, we should + * pass that up! + */ void lguest_setup_irq(unsigned int irq) { irq_to_desc_alloc_node(irq, 0); @@ -724,31 +857,39 @@ static unsigned long lguest_get_wallclock(void) return lguest_data.time.tv_sec; } -/* The TSC is an Intel thing called the Time Stamp Counter. The Host tells us +/* + * The TSC is an Intel thing called the Time Stamp Counter. The Host tells us * what speed it runs at, or 0 if it's unusable as a reliable clock source. * This matches what we want here: if we return 0 from this function, the x86 - * TSC clock will give up and not register itself. */ + * TSC clock will give up and not register itself. + */ static unsigned long lguest_tsc_khz(void) { return lguest_data.tsc_khz; } -/* If we can't use the TSC, the kernel falls back to our lower-priority - * "lguest_clock", where we read the time value given to us by the Host. */ +/* + * If we can't use the TSC, the kernel falls back to our lower-priority + * "lguest_clock", where we read the time value given to us by the Host. + */ static cycle_t lguest_clock_read(struct clocksource *cs) { unsigned long sec, nsec; - /* Since the time is in two parts (seconds and nanoseconds), we risk + /* + * Since the time is in two parts (seconds and nanoseconds), we risk * reading it just as it's changing from 99 & 0.999999999 to 100 and 0, * and getting 99 and 0. As Linux tends to come apart under the stress - * of time travel, we must be careful: */ + * of time travel, we must be careful: + */ do { /* First we read the seconds part. */ sec = lguest_data.time.tv_sec; - /* This read memory barrier tells the compiler and the CPU that + /* + * This read memory barrier tells the compiler and the CPU that * this can't be reordered: we have to complete the above - * before going on. */ + * before going on. + */ rmb(); /* Now we read the nanoseconds part. */ nsec = lguest_data.time.tv_nsec; @@ -772,9 +913,11 @@ static struct clocksource lguest_clock = { .flags = CLOCK_SOURCE_IS_CONTINUOUS, }; -/* We also need a "struct clock_event_device": Linux asks us to set it to go +/* + * We also need a "struct clock_event_device": Linux asks us to set it to go * off some time in the future. Actually, James Morris figured all this out, I - * just applied the patch. */ + * just applied the patch. + */ static int lguest_clockevent_set_next_event(unsigned long delta, struct clock_event_device *evt) { @@ -824,8 +967,10 @@ static struct clock_event_device lguest_clockevent = { .max_delta_ns = LG_CLOCK_MAX_DELTA, }; -/* This is the Guest timer interrupt handler (hardware interrupt 0). We just - * call the clockevent infrastructure and it does whatever needs doing. */ +/* + * This is the Guest timer interrupt handler (hardware interrupt 0). We just + * call the clockevent infrastructure and it does whatever needs doing. + */ static void lguest_time_irq(unsigned int irq, struct irq_desc *desc) { unsigned long flags; @@ -836,10 +981,12 @@ static void lguest_time_irq(unsigned int irq, struct irq_desc *desc) local_irq_restore(flags); } -/* At some point in the boot process, we get asked to set up our timing +/* + * At some point in the boot process, we get asked to set up our timing * infrastructure. The kernel doesn't expect timer interrupts before this, but * we cleverly initialized the "blocked_interrupts" field of "struct - * lguest_data" so that timer interrupts were blocked until now. */ + * lguest_data" so that timer interrupts were blocked until now. + */ static void lguest_time_init(void) { /* Set up the timer interrupt (0) to go to our simple timer routine */ @@ -863,14 +1010,16 @@ static void lguest_time_init(void) * to work. They're pretty simple. */ -/* The Guest needs to tell the Host what stack it expects traps to use. For +/* + * The Guest needs to tell the Host what stack it expects traps to use. For * native hardware, this is part of the Task State Segment mentioned above in * lguest_load_tr_desc(), but to help hypervisors there's this special call. * * We tell the Host the segment we want to use (__KERNEL_DS is the kernel data * segment), the privilege level (we're privilege level 1, the Host is 0 and * will not tolerate us trying to use that), the stack pointer, and the number - * of pages in the stack. */ + * of pages in the stack. + */ static void lguest_load_sp0(struct tss_struct *tss, struct thread_struct *thread) { @@ -884,7 +1033,8 @@ static void lguest_set_debugreg(int regno, unsigned long value) /* FIXME: Implement */ } -/* There are times when the kernel wants to make sure that no memory writes are +/* + * There are times when the kernel wants to make sure that no memory writes are * caught in the cache (that they've all reached real hardware devices). This * doesn't matter for the Guest which has virtual hardware. * @@ -898,11 +1048,13 @@ static void lguest_wbinvd(void) { } -/* If the Guest expects to have an Advanced Programmable Interrupt Controller, +/* + * If the Guest expects to have an Advanced Programmable Interrupt Controller, * we play dumb by ignoring writes and returning 0 for reads. So it's no * longer Programmable nor Controlling anything, and I don't think 8 lines of * code qualifies for Advanced. It will also never interrupt anything. It - * does, however, allow us to get through the Linux boot code. */ + * does, however, allow us to get through the Linux boot code. + */ #ifdef CONFIG_X86_LOCAL_APIC static void lguest_apic_write(u32 reg, u32 v) { @@ -951,11 +1103,13 @@ static void lguest_safe_halt(void) kvm_hypercall0(LHCALL_HALT); } -/* The SHUTDOWN hypercall takes a string to describe what's happening, and +/* + * The SHUTDOWN hypercall takes a string to describe what's happening, and * an argument which says whether this to restart (reboot) the Guest or not. * * Note that the Host always prefers that the Guest speak in physical addresses - * rather than virtual addresses, so we use __pa() here. */ + * rather than virtual addresses, so we use __pa() here. + */ static void lguest_power_off(void) { kvm_hypercall2(LHCALL_SHUTDOWN, __pa("Power down"), @@ -986,8 +1140,10 @@ static __init char *lguest_memory_setup(void) * nice to move it back to lguest_init. Patch welcome... */ atomic_notifier_chain_register(&panic_notifier_list, &paniced); - /* The Linux bootloader header contains an "e820" memory map: the - * Launcher populated the first entry with our memory limit. */ + /* + *The Linux bootloader header contains an "e820" memory map: the + * Launcher populated the first entry with our memory limit. + */ e820_add_region(boot_params.e820_map[0].addr, boot_params.e820_map[0].size, boot_params.e820_map[0].type); @@ -996,16 +1152,17 @@ static __init char *lguest_memory_setup(void) return "LGUEST"; } -/* We will eventually use the virtio console device to produce console output, +/* + * We will eventually use the virtio console device to produce console output, * but before that is set up we use LHCALL_NOTIFY on normal memory to produce - * console output. */ + * console output. + */ static __init int early_put_chars(u32 vtermno, const char *buf, int count) { char scratch[17]; unsigned int len = count; - /* We use a nul-terminated string, so we have to make a copy. Icky, - * huh? */ + /* We use a nul-terminated string, so we make a copy. Icky, huh? */ if (len > sizeof(scratch) - 1) len = sizeof(scratch) - 1; scratch[len] = '\0'; @@ -1016,8 +1173,10 @@ static __init int early_put_chars(u32 vtermno, const char *buf, int count) return len; } -/* Rebooting also tells the Host we're finished, but the RESTART flag tells the - * Launcher to reboot us. */ +/* + * Rebooting also tells the Host we're finished, but the RESTART flag tells the + * Launcher to reboot us. + */ static void lguest_restart(char *reason) { kvm_hypercall2(LHCALL_SHUTDOWN, __pa(reason), LGUEST_SHUTDOWN_RESTART); @@ -1044,7 +1203,8 @@ static void lguest_restart(char *reason) * fit comfortably. * * First we need assembly templates of each of the patchable Guest operations, - * and these are in i386_head.S. */ + * and these are in i386_head.S. + */ /*G:060 We construct a table from the assembler templates: */ static const struct lguest_insns @@ -1055,9 +1215,11 @@ static const struct lguest_insns [PARAVIRT_PATCH(pv_irq_ops.save_fl)] = { lgstart_pushf, lgend_pushf }, }; -/* Now our patch routine is fairly simple (based on the native one in +/* + * Now our patch routine is fairly simple (based on the native one in * paravirt.c). If we have a replacement, we copy it in and return how much of - * the available space we used. */ + * the available space we used. + */ static unsigned lguest_patch(u8 type, u16 clobber, void *ibuf, unsigned long addr, unsigned len) { @@ -1069,8 +1231,7 @@ static unsigned lguest_patch(u8 type, u16 clobber, void *ibuf, insn_len = lguest_insns[type].end - lguest_insns[type].start; - /* Similarly if we can't fit replacement (shouldn't happen, but let's - * be thorough). */ + /* Similarly if it can't fit (doesn't happen, but let's be thorough). */ if (len < insn_len) return paravirt_patch_default(type, clobber, ibuf, addr, len); @@ -1079,22 +1240,28 @@ static unsigned lguest_patch(u8 type, u16 clobber, void *ibuf, return insn_len; } -/*G:030 Once we get to lguest_init(), we know we're a Guest. The various +/*G:029 + * Once we get to lguest_init(), we know we're a Guest. The various * pv_ops structures in the kernel provide points for (almost) every routine we - * have to override to avoid privileged instructions. */ + * have to override to avoid privileged instructions. + */ __init void lguest_init(void) { - /* We're under lguest, paravirt is enabled, and we're running at - * privilege level 1, not 0 as normal. */ + /* We're under lguest. */ pv_info.name = "lguest"; + /* Paravirt is enabled. */ pv_info.paravirt_enabled = 1; + /* We're running at privilege level 1, not 0 as normal. */ pv_info.kernel_rpl = 1; + /* Everyone except Xen runs with this set. */ pv_info.shared_kernel_pmd = 1; - /* We set up all the lguest overrides for sensitive operations. These - * are detailed with the operations themselves. */ + /* + * We set up all the lguest overrides for sensitive operations. These + * are detailed with the operations themselves. + */ - /* interrupt-related operations */ + /* Interrupt-related operations */ pv_irq_ops.init_IRQ = lguest_init_IRQ; pv_irq_ops.save_fl = PV_CALLEE_SAVE(save_fl); pv_irq_ops.restore_fl = __PV_IS_CALLEE_SAVE(lg_restore_fl); @@ -1102,11 +1269,11 @@ __init void lguest_init(void) pv_irq_ops.irq_enable = __PV_IS_CALLEE_SAVE(lg_irq_enable); pv_irq_ops.safe_halt = lguest_safe_halt; - /* init-time operations */ + /* Setup operations */ pv_init_ops.memory_setup = lguest_memory_setup; pv_init_ops.patch = lguest_patch; - /* Intercepts of various cpu instructions */ + /* Intercepts of various CPU instructions */ pv_cpu_ops.load_gdt = lguest_load_gdt; pv_cpu_ops.cpuid = lguest_cpuid; pv_cpu_ops.load_idt = lguest_load_idt; @@ -1127,7 +1294,7 @@ __init void lguest_init(void) pv_cpu_ops.start_context_switch = paravirt_start_context_switch; pv_cpu_ops.end_context_switch = lguest_end_context_switch; - /* pagetable management */ + /* Pagetable management */ pv_mmu_ops.write_cr3 = lguest_write_cr3; pv_mmu_ops.flush_tlb_user = lguest_flush_tlb_user; pv_mmu_ops.flush_tlb_single = lguest_flush_tlb_single; @@ -1149,54 +1316,71 @@ __init void lguest_init(void) pv_mmu_ops.pte_update_defer = lguest_pte_update; #ifdef CONFIG_X86_LOCAL_APIC - /* apic read/write intercepts */ + /* APIC read/write intercepts */ set_lguest_basic_apic_ops(); #endif - /* time operations */ + /* Time operations */ pv_time_ops.get_wallclock = lguest_get_wallclock; pv_time_ops.time_init = lguest_time_init; pv_time_ops.get_tsc_khz = lguest_tsc_khz; - /* Now is a good time to look at the implementations of these functions - * before returning to the rest of lguest_init(). */ + /* + * Now is a good time to look at the implementations of these functions + * before returning to the rest of lguest_init(). + */ - /*G:070 Now we've seen all the paravirt_ops, we return to + /*G:070 + * Now we've seen all the paravirt_ops, we return to * lguest_init() where the rest of the fairly chaotic boot setup - * occurs. */ + * occurs. + */ - /* The stack protector is a weird thing where gcc places a canary + /* + * The stack protector is a weird thing where gcc places a canary * value on the stack and then checks it on return. This file is * compiled with -fno-stack-protector it, so we got this far without * problems. The value of the canary is kept at offset 20 from the * %gs register, so we need to set that up before calling C functions - * in other files. */ + * in other files. + */ setup_stack_canary_segment(0); - /* We could just call load_stack_canary_segment(), but we might as - * call switch_to_new_gdt() which loads the whole table and sets up - * the per-cpu segment descriptor register %fs as well. */ + + /* + * We could just call load_stack_canary_segment(), but we might as well + * call switch_to_new_gdt() which loads the whole table and sets up the + * per-cpu segment descriptor register %fs as well. + */ switch_to_new_gdt(0); - /* As described in head_32.S, we map the first 128M of memory. */ + /* We actually boot with all memory mapped, but let's say 128MB. */ max_pfn_mapped = (128*1024*1024) >> PAGE_SHIFT; - /* The Host<->Guest Switcher lives at the top of our address space, and + /* + * The Host<->Guest Switcher lives at the top of our address space, and * the Host told us how big it is when we made LGUEST_INIT hypercall: - * it put the answer in lguest_data.reserve_mem */ + * it put the answer in lguest_data.reserve_mem + */ reserve_top_address(lguest_data.reserve_mem); - /* If we don't initialize the lock dependency checker now, it crashes - * paravirt_disable_iospace. */ + /* + * If we don't initialize the lock dependency checker now, it crashes + * paravirt_disable_iospace. + */ lockdep_init(); - /* The IDE code spends about 3 seconds probing for disks: if we reserve + /* + * The IDE code spends about 3 seconds probing for disks: if we reserve * all the I/O ports up front it can't get them and so doesn't probe. * Other device drivers are similar (but less severe). This cuts the - * kernel boot time on my machine from 4.1 seconds to 0.45 seconds. */ + * kernel boot time on my machine from 4.1 seconds to 0.45 seconds. + */ paravirt_disable_iospace(); - /* This is messy CPU setup stuff which the native boot code does before - * start_kernel, so we have to do, too: */ + /* + * This is messy CPU setup stuff which the native boot code does before + * start_kernel, so we have to do, too: + */ cpu_detect(&new_cpu_data); /* head.S usually sets up the first capability word, so do it here. */ new_cpu_data.x86_capability[0] = cpuid_edx(1); @@ -1213,22 +1397,28 @@ __init void lguest_init(void) acpi_ht = 0; #endif - /* We set the preferred console to "hvc". This is the "hypervisor + /* + * We set the preferred console to "hvc". This is the "hypervisor * virtual console" driver written by the PowerPC people, which we also - * adapted for lguest's use. */ + * adapted for lguest's use. + */ add_preferred_console("hvc", 0, NULL); /* Register our very early console. */ virtio_cons_early_init(early_put_chars); - /* Last of all, we set the power management poweroff hook to point to + /* + * Last of all, we set the power management poweroff hook to point to * the Guest routine to power off, and the reboot hook to our restart - * routine. */ + * routine. + */ pm_power_off = lguest_power_off; machine_ops.restart = lguest_restart; - /* Now we're set up, call i386_start_kernel() in head32.c and we proceed - * to boot as normal. It never returns. */ + /* + * Now we're set up, call i386_start_kernel() in head32.c and we proceed + * to boot as normal. It never returns. + */ i386_start_kernel(); } /* diff --git a/arch/x86/lguest/i386_head.S b/arch/x86/lguest/i386_head.S index a9c8cfe..27eac0f 100644 --- a/arch/x86/lguest/i386_head.S +++ b/arch/x86/lguest/i386_head.S @@ -5,7 +5,8 @@ #include <asm/thread_info.h> #include <asm/processor-flags.h> -/*G:020 Our story starts with the kernel booting into startup_32 in +/*G:020 + * Our story starts with the kernel booting into startup_32 in * arch/x86/kernel/head_32.S. It expects a boot header, which is created by * the bootloader (the Launcher in our case). * @@ -21,11 +22,14 @@ * data without remembering to subtract __PAGE_OFFSET! * * The .section line puts this code in .init.text so it will be discarded after - * boot. */ + * boot. + */ .section .init.text, "ax", @progbits ENTRY(lguest_entry) - /* We make the "initialization" hypercall now to tell the Host about - * us, and also find out where it put our page tables. */ + /* + * We make the "initialization" hypercall now to tell the Host about + * us, and also find out where it put our page tables. + */ movl $LHCALL_LGUEST_INIT, %eax movl $lguest_data - __PAGE_OFFSET, %ebx .byte 0x0f,0x01,0xc1 /* KVM_HYPERCALL */ @@ -33,13 +37,14 @@ ENTRY(lguest_entry) /* Set up the initial stack so we can run C code. */ movl $(init_thread_union+THREAD_SIZE),%esp - /* Jumps are relative, and we're running __PAGE_OFFSET too low at the - * moment. */ + /* Jumps are relative: we're running __PAGE_OFFSET too low. */ jmp lguest_init+__PAGE_OFFSET -/*G:055 We create a macro which puts the assembler code between lgstart_ and - * lgend_ markers. These templates are put in the .text section: they can't be - * discarded after boot as we may need to patch modules, too. */ +/*G:055 + * We create a macro which puts the assembler code between lgstart_ and lgend_ + * markers. These templates are put in the .text section: they can't be + * discarded after boot as we may need to patch modules, too. + */ .text #define LGUEST_PATCH(name, insns...) \ lgstart_##name: insns; lgend_##name:; \ @@ -48,83 +53,103 @@ ENTRY(lguest_entry) LGUEST_PATCH(cli, movl $0, lguest_data+LGUEST_DATA_irq_enabled) LGUEST_PATCH(pushf, movl lguest_data+LGUEST_DATA_irq_enabled, %eax) -/*G:033 But using those wrappers is inefficient (we'll see why that doesn't - * matter for save_fl and irq_disable later). If we write our routines - * carefully in assembler, we can avoid clobbering any registers and avoid - * jumping through the wrapper functions. +/*G:033 + * But using those wrappers is inefficient (we'll see why that doesn't matter + * for save_fl and irq_disable later). If we write our routines carefully in + * assembler, we can avoid clobbering any registers and avoid jumping through + * the wrapper functions. * * I skipped over our first piece of assembler, but this one is worth studying - * in a bit more detail so I'll describe in easy stages. First, the routine - * to enable interrupts: */ + * in a bit more detail so I'll describe in easy stages. First, the routine to + * enable interrupts: + */ ENTRY(lg_irq_enable) - /* The reverse of irq_disable, this sets lguest_data.irq_enabled to - * X86_EFLAGS_IF (ie. "Interrupts enabled"). */ + /* + * The reverse of irq_disable, this sets lguest_data.irq_enabled to + * X86_EFLAGS_IF (ie. "Interrupts enabled"). + */ movl $X86_EFLAGS_IF, lguest_data+LGUEST_DATA_irq_enabled - /* But now we need to check if the Host wants to know: there might have + /* + * But now we need to check if the Host wants to know: there might have * been interrupts waiting to be delivered, in which case it will have * set lguest_data.irq_pending to X86_EFLAGS_IF. If it's not zero, we - * jump to send_interrupts, otherwise we're done. */ + * jump to send_interrupts, otherwise we're done. + */ testl $0, lguest_data+LGUEST_DATA_irq_pending jnz send_interrupts - /* One cool thing about x86 is that you can do many things without using + /* + * One cool thing about x86 is that you can do many things without using * a register. In this case, the normal path hasn't needed to save or - * restore any registers at all! */ + * restore any registers at all! + */ ret send_interrupts: - /* OK, now we need a register: eax is used for the hypercall number, + /* + * OK, now we need a register: eax is used for the hypercall number, * which is LHCALL_SEND_INTERRUPTS. * * We used not to bother with this pending detection at all, which was * much simpler. Sooner or later the Host would realize it had to * send us an interrupt. But that turns out to make performance 7 * times worse on a simple tcp benchmark. So now we do this the hard - * way. */ + * way. + */ pushl %eax movl $LHCALL_SEND_INTERRUPTS, %eax - /* This is a vmcall instruction (same thing that KVM uses). Older + /* + * This is a vmcall instruction (same thing that KVM uses). Older * assembler versions might not know the "vmcall" instruction, so we - * create one manually here. */ + * create one manually here. + */ .byte 0x0f,0x01,0xc1 /* KVM_HYPERCALL */ + /* Put eax back the way we found it. */ popl %eax ret -/* Finally, the "popf" or "restore flags" routine. The %eax register holds the +/* + * Finally, the "popf" or "restore flags" routine. The %eax register holds the * flags (in practice, either X86_EFLAGS_IF or 0): if it's X86_EFLAGS_IF we're - * enabling interrupts again, if it's 0 we're leaving them off. */ + * enabling interrupts again, if it's 0 we're leaving them off. + */ ENTRY(lg_restore_fl) /* This is just "lguest_data.irq_enabled = flags;" */ movl %eax, lguest_data+LGUEST_DATA_irq_enabled - /* Now, if the %eax value has enabled interrupts and + /* + * Now, if the %eax value has enabled interrupts and * lguest_data.irq_pending is set, we want to tell the Host so it can * deliver any outstanding interrupts. Fortunately, both values will * be X86_EFLAGS_IF (ie. 512) in that case, and the "testl" * instruction will AND them together for us. If both are set, we - * jump to send_interrupts. */ + * jump to send_interrupts. + */ testl lguest_data+LGUEST_DATA_irq_pending, %eax jnz send_interrupts /* Again, the normal path has used no extra registers. Clever, huh? */ ret +/*:*/ /* These demark the EIP range where host should never deliver interrupts. */ .global lguest_noirq_start .global lguest_noirq_end -/*M:004 When the Host reflects a trap or injects an interrupt into the Guest, - * it sets the eflags interrupt bit on the stack based on - * lguest_data.irq_enabled, so the Guest iret logic does the right thing when - * restoring it. However, when the Host sets the Guest up for direct traps, - * such as system calls, the processor is the one to push eflags onto the - * stack, and the interrupt bit will be 1 (in reality, interrupts are always - * enabled in the Guest). +/*M:004 + * When the Host reflects a trap or injects an interrupt into the Guest, it + * sets the eflags interrupt bit on the stack based on lguest_data.irq_enabled, + * so the Guest iret logic does the right thing when restoring it. However, + * when the Host sets the Guest up for direct traps, such as system calls, the + * processor is the one to push eflags onto the stack, and the interrupt bit + * will be 1 (in reality, interrupts are always enabled in the Guest). * * This turns out to be harmless: the only trap which should happen under Linux * with interrupts disabled is Page Fault (due to our lazy mapping of vmalloc * regions), which has to be reflected through the Host anyway. If another * trap *does* go off when interrupts are disabled, the Guest will panic, and - * we'll never get to this iret! :*/ + * we'll never get to this iret! +:*/ -/*G:045 There is one final paravirt_op that the Guest implements, and glancing - * at it you can see why I left it to last. It's *cool*! It's in *assembler*! +/*G:045 + * There is one final paravirt_op that the Guest implements, and glancing at it + * you can see why I left it to last. It's *cool*! It's in *assembler*! * * The "iret" instruction is used to return from an interrupt or trap. The * stack looks like this: @@ -148,15 +173,18 @@ ENTRY(lg_restore_fl) * return to userspace or wherever. Our solution to this is to surround the * code with lguest_noirq_start: and lguest_noirq_end: labels. We tell the * Host that it is *never* to interrupt us there, even if interrupts seem to be - * enabled. */ + * enabled. + */ ENTRY(lguest_iret) pushl %eax movl 12(%esp), %eax lguest_noirq_start: - /* Note the %ss: segment prefix here. Normal data accesses use the + /* + * Note the %ss: segment prefix here. Normal data accesses use the * "ds" segment, but that will have already been restored for whatever * we're returning to (such as userspace): we can't trust it. The %ss: - * prefix makes sure we use the stack segment, which is still valid. */ + * prefix makes sure we use the stack segment, which is still valid. + */ movl %eax,%ss:lguest_data+LGUEST_DATA_irq_enabled popl %eax iret diff --git a/arch/x86/lib/Makefile b/arch/x86/lib/Makefile index f9d3563..07c3189 100644 --- a/arch/x86/lib/Makefile +++ b/arch/x86/lib/Makefile @@ -10,6 +10,7 @@ lib-y += usercopy_$(BITS).o getuser.o putuser.o lib-y += memcpy_$(BITS).o ifeq ($(CONFIG_X86_32),y) + obj-y += atomic64_32.o lib-y += checksum_32.o lib-y += strstr_32.o lib-y += semaphore_32.o string_32.o diff --git a/arch/x86/lib/atomic64_32.c b/arch/x86/lib/atomic64_32.c new file mode 100644 index 0000000..824fa0b --- /dev/null +++ b/arch/x86/lib/atomic64_32.c @@ -0,0 +1,230 @@ +#include <linux/compiler.h> +#include <linux/module.h> +#include <linux/types.h> + +#include <asm/processor.h> +#include <asm/cmpxchg.h> +#include <asm/atomic.h> + +static noinline u64 cmpxchg8b(u64 *ptr, u64 old, u64 new) +{ + u32 low = new; + u32 high = new >> 32; + + asm volatile( + LOCK_PREFIX "cmpxchg8b %1\n" + : "+A" (old), "+m" (*ptr) + : "b" (low), "c" (high) + ); + return old; +} + +u64 atomic64_cmpxchg(atomic64_t *ptr, u64 old_val, u64 new_val) +{ + return cmpxchg8b(&ptr->counter, old_val, new_val); +} +EXPORT_SYMBOL(atomic64_cmpxchg); + +/** + * atomic64_xchg - xchg atomic64 variable + * @ptr: pointer to type atomic64_t + * @new_val: value to assign + * + * Atomically xchgs the value of @ptr to @new_val and returns + * the old value. + */ +u64 atomic64_xchg(atomic64_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(ptr, old_val, new_val); + + } while (real_val != old_val); + + return old_val; +} +EXPORT_SYMBOL(atomic64_xchg); + +/** + * atomic64_set - set atomic64 variable + * @ptr: pointer to type atomic64_t + * @new_val: value to assign + * + * Atomically sets the value of @ptr to @new_val. + */ +void atomic64_set(atomic64_t *ptr, u64 new_val) +{ + atomic64_xchg(ptr, new_val); +} +EXPORT_SYMBOL(atomic64_set); + +/** +EXPORT_SYMBOL(atomic64_read); + * atomic64_add_return - add and return + * @delta: integer value to add + * @ptr: pointer to type atomic64_t + * + * Atomically adds @delta to @ptr and returns @delta + *@ptr + */ +noinline u64 atomic64_add_return(u64 delta, atomic64_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(ptr, old_val, new_val); + + } while (real_val != old_val); + + return new_val; +} +EXPORT_SYMBOL(atomic64_add_return); + +u64 atomic64_sub_return(u64 delta, atomic64_t *ptr) +{ + return atomic64_add_return(-delta, ptr); +} +EXPORT_SYMBOL(atomic64_sub_return); + +u64 atomic64_inc_return(atomic64_t *ptr) +{ + return atomic64_add_return(1, ptr); +} +EXPORT_SYMBOL(atomic64_inc_return); + +u64 atomic64_dec_return(atomic64_t *ptr) +{ + return atomic64_sub_return(1, ptr); +} +EXPORT_SYMBOL(atomic64_dec_return); + +/** + * atomic64_add - add integer to atomic64 variable + * @delta: integer value to add + * @ptr: pointer to type atomic64_t + * + * Atomically adds @delta to @ptr. + */ +void atomic64_add(u64 delta, atomic64_t *ptr) +{ + atomic64_add_return(delta, ptr); +} +EXPORT_SYMBOL(atomic64_add); + +/** + * atomic64_sub - subtract the atomic64 variable + * @delta: integer value to subtract + * @ptr: pointer to type atomic64_t + * + * Atomically subtracts @delta from @ptr. + */ +void atomic64_sub(u64 delta, atomic64_t *ptr) +{ + atomic64_add(-delta, ptr); +} +EXPORT_SYMBOL(atomic64_sub); + +/** + * atomic64_sub_and_test - subtract value from variable and test result + * @delta: integer value to subtract + * @ptr: pointer to type atomic64_t + * + * Atomically subtracts @delta from @ptr and returns + * true if the result is zero, or false for all + * other cases. + */ +int atomic64_sub_and_test(u64 delta, atomic64_t *ptr) +{ + u64 new_val = atomic64_sub_return(delta, ptr); + + return new_val == 0; +} +EXPORT_SYMBOL(atomic64_sub_and_test); + +/** + * atomic64_inc - increment atomic64 variable + * @ptr: pointer to type atomic64_t + * + * Atomically increments @ptr by 1. + */ +void atomic64_inc(atomic64_t *ptr) +{ + atomic64_add(1, ptr); +} +EXPORT_SYMBOL(atomic64_inc); + +/** + * atomic64_dec - decrement atomic64 variable + * @ptr: pointer to type atomic64_t + * + * Atomically decrements @ptr by 1. + */ +void atomic64_dec(atomic64_t *ptr) +{ + atomic64_sub(1, ptr); +} +EXPORT_SYMBOL(atomic64_dec); + +/** + * atomic64_dec_and_test - decrement and test + * @ptr: pointer to type atomic64_t + * + * Atomically decrements @ptr by 1 and + * returns true if the result is 0, or false for all other + * cases. + */ +int atomic64_dec_and_test(atomic64_t *ptr) +{ + return atomic64_sub_and_test(1, ptr); +} +EXPORT_SYMBOL(atomic64_dec_and_test); + +/** + * atomic64_inc_and_test - increment and test + * @ptr: pointer to type atomic64_t + * + * Atomically increments @ptr by 1 + * and returns true if the result is zero, or false for all + * other cases. + */ +int atomic64_inc_and_test(atomic64_t *ptr) +{ + return atomic64_sub_and_test(-1, ptr); +} +EXPORT_SYMBOL(atomic64_inc_and_test); + +/** + * atomic64_add_negative - add and test if negative + * @delta: integer value to add + * @ptr: pointer to type atomic64_t + * + * Atomically adds @delta to @ptr and returns true + * if the result is negative, or false when + * result is greater than or equal to zero. + */ +int atomic64_add_negative(u64 delta, atomic64_t *ptr) +{ + s64 new_val = atomic64_add_return(delta, ptr); + + return new_val < 0; +} +EXPORT_SYMBOL(atomic64_add_negative); diff --git a/arch/x86/lib/clear_page_64.S b/arch/x86/lib/clear_page_64.S index 9a10a78..ebeafcc 100644 --- a/arch/x86/lib/clear_page_64.S +++ b/arch/x86/lib/clear_page_64.S @@ -5,15 +5,14 @@ * Zero a page. * rdi page */ - ALIGN -clear_page_c: +ENTRY(clear_page_c) CFI_STARTPROC movl $4096/8,%ecx xorl %eax,%eax rep stosq ret CFI_ENDPROC -ENDPROC(clear_page) +ENDPROC(clear_page_c) ENTRY(clear_page) CFI_STARTPROC diff --git a/arch/x86/lib/copy_user_64.S b/arch/x86/lib/copy_user_64.S index f118c11..6ba0f7b 100644 --- a/arch/x86/lib/copy_user_64.S +++ b/arch/x86/lib/copy_user_64.S @@ -75,6 +75,7 @@ ENTRY(copy_to_user) jae 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) diff --git a/arch/x86/lib/delay.c b/arch/x86/lib/delay.c index f456860..ff485d3 100644 --- a/arch/x86/lib/delay.c +++ b/arch/x86/lib/delay.c @@ -55,8 +55,10 @@ static void delay_tsc(unsigned long loops) preempt_disable(); cpu = smp_processor_id(); + rdtsc_barrier(); rdtscl(bclock); for (;;) { + rdtsc_barrier(); rdtscl(now); if ((now - bclock) >= loops) break; @@ -78,6 +80,7 @@ static void delay_tsc(unsigned long loops) if (unlikely(cpu != smp_processor_id())) { loops -= (now - bclock); cpu = smp_processor_id(); + rdtsc_barrier(); rdtscl(bclock); } } diff --git a/arch/x86/lib/msr.c b/arch/x86/lib/msr.c index 1440b9c..caa24ac 100644 --- a/arch/x86/lib/msr.c +++ b/arch/x86/lib/msr.c @@ -89,16 +89,13 @@ void rdmsr_on_cpus(const cpumask_t *mask, u32 msr_no, struct msr *msrs) rv.msrs = msrs; rv.msr_no = msr_no; - preempt_disable(); - /* - * FIXME: handle the CPU we're executing on separately for now until - * smp_call_function_many has been fixed to not skip it. - */ - this_cpu = raw_smp_processor_id(); - smp_call_function_single(this_cpu, __rdmsr_on_cpu, &rv, 1); + this_cpu = get_cpu(); + + if (cpumask_test_cpu(this_cpu, mask)) + __rdmsr_on_cpu(&rv); smp_call_function_many(mask, __rdmsr_on_cpu, &rv, 1); - preempt_enable(); + put_cpu(); } EXPORT_SYMBOL(rdmsr_on_cpus); @@ -121,16 +118,13 @@ void wrmsr_on_cpus(const cpumask_t *mask, u32 msr_no, struct msr *msrs) rv.msrs = msrs; rv.msr_no = msr_no; - preempt_disable(); - /* - * FIXME: handle the CPU we're executing on separately for now until - * smp_call_function_many has been fixed to not skip it. - */ - this_cpu = raw_smp_processor_id(); - smp_call_function_single(this_cpu, __wrmsr_on_cpu, &rv, 1); + this_cpu = get_cpu(); + + if (cpumask_test_cpu(this_cpu, mask)) + __wrmsr_on_cpu(&rv); smp_call_function_many(mask, __wrmsr_on_cpu, &rv, 1); - preempt_enable(); + put_cpu(); } EXPORT_SYMBOL(wrmsr_on_cpus); diff --git a/arch/x86/lib/usercopy_32.c b/arch/x86/lib/usercopy_32.c index 7c8ca91..1f118d4 100644 --- a/arch/x86/lib/usercopy_32.c +++ b/arch/x86/lib/usercopy_32.c @@ -751,7 +751,7 @@ survive: if (retval == -ENOMEM && is_global_init(current)) { up_read(¤t->mm->mmap_sem); - congestion_wait(WRITE, HZ/50); + congestion_wait(BLK_RW_ASYNC, HZ/50); goto survive; } diff --git a/arch/x86/lib/usercopy_64.c b/arch/x86/lib/usercopy_64.c index ec13cb5..b7c2849 100644 --- a/arch/x86/lib/usercopy_64.c +++ b/arch/x86/lib/usercopy_64.c @@ -127,7 +127,7 @@ EXPORT_SYMBOL(__strnlen_user); long strnlen_user(const char __user *s, long n) { - if (!access_ok(VERIFY_READ, s, n)) + if (!access_ok(VERIFY_READ, s, 1)) return 0; return __strnlen_user(s, n); } diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c index baa0e86..bfae139 100644 --- a/arch/x86/mm/fault.c +++ b/arch/x86/mm/fault.c @@ -426,10 +426,11 @@ static noinline int vmalloc_fault(unsigned long address) } static const char errata93_warning[] = -KERN_ERR "******* Your BIOS seems to not contain a fix for K8 errata #93\n" -KERN_ERR "******* Working around it, but it may cause SEGVs or burn power.\n" -KERN_ERR "******* Please consider a BIOS update.\n" -KERN_ERR "******* Disabling USB legacy in the BIOS may also help.\n"; +KERN_ERR +"******* Your BIOS seems to not contain a fix for K8 errata #93\n" +"******* Working around it, but it may cause SEGVs or burn power.\n" +"******* Please consider a BIOS update.\n" +"******* Disabling USB legacy in the BIOS may also help.\n"; /* * No vm86 mode in 64-bit mode: @@ -696,7 +697,7 @@ show_signal_msg(struct pt_regs *regs, unsigned long error_code, if (!printk_ratelimit()) return; - printk(KERN_CONT "%s%s[%d]: segfault at %lx ip %p sp %p error %lx", + printk("%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); @@ -952,8 +953,6 @@ do_page_fault(struct pt_regs *regs, unsigned long error_code) tsk = current; mm = tsk->mm; - prefetchw(&mm->mmap_sem); - /* Get the faulting address: */ address = read_cr2(); @@ -963,6 +962,7 @@ do_page_fault(struct pt_regs *regs, unsigned long error_code) */ if (kmemcheck_active(regs)) kmemcheck_hide(regs); + prefetchw(&mm->mmap_sem); if (unlikely(kmmio_fault(regs, address))) return; @@ -1114,7 +1114,7 @@ good_area: * make sure we exit gracefully rather than endlessly redo * the fault: */ - fault = handle_mm_fault(mm, vma, address, write); + fault = handle_mm_fault(mm, vma, address, write ? FAULT_FLAG_WRITE : 0); if (unlikely(fault & VM_FAULT_ERROR)) { mm_fault_error(regs, error_code, address, fault); diff --git a/arch/x86/mm/gup.c b/arch/x86/mm/gup.c index 6340cef..71da1bc 100644 --- a/arch/x86/mm/gup.c +++ b/arch/x86/mm/gup.c @@ -14,7 +14,7 @@ static inline pte_t gup_get_pte(pte_t *ptep) { #ifndef CONFIG_X86_PAE - return *ptep; + return ACCESS_ONCE(*ptep); #else /* * With get_user_pages_fast, we walk down the pagetables without taking @@ -219,6 +219,62 @@ static int gup_pud_range(pgd_t pgd, unsigned long addr, unsigned long end, return 1; } +/* + * Like get_user_pages_fast() except its IRQ-safe in that it won't fall + * back to the regular GUP. + */ +int __get_user_pages_fast(unsigned long start, int nr_pages, int write, + struct page **pages) +{ + struct mm_struct *mm = current->mm; + unsigned long addr, len, end; + unsigned long next; + unsigned long flags; + pgd_t *pgdp; + int nr = 0; + + start &= PAGE_MASK; + addr = start; + len = (unsigned long) nr_pages << PAGE_SHIFT; + end = start + len; + if (unlikely(!access_ok(write ? VERIFY_WRITE : VERIFY_READ, + (void __user *)start, len))) + return 0; + + /* + * XXX: batch / limit 'nr', to avoid large irq off latency + * needs some instrumenting to determine the common sizes used by + * important workloads (eg. DB2), and whether limiting the batch size + * will decrease performance. + * + * It seems like we're in the clear for the moment. Direct-IO is + * the main guy that batches up lots of get_user_pages, and even + * they are limited to 64-at-a-time which is not so many. + */ + /* + * This doesn't prevent pagetable teardown, but does prevent + * the pagetables and pages from being freed on x86. + * + * So long as we atomically load page table pointers versus teardown + * (which we do on x86, with the above PAE exception), we can follow the + * address down to the the page and take a ref on it. + */ + local_irq_save(flags); + pgdp = pgd_offset(mm, addr); + do { + pgd_t pgd = *pgdp; + + next = pgd_addr_end(addr, end); + if (pgd_none(pgd)) + break; + if (!gup_pud_range(pgd, addr, next, write, pages, &nr)) + break; + } while (pgdp++, addr = next, addr != end); + local_irq_restore(flags); + + return nr; +} + /** * get_user_pages_fast() - pin user pages in memory * @start: starting user address @@ -247,11 +303,16 @@ int get_user_pages_fast(unsigned long start, int nr_pages, int write, start &= PAGE_MASK; addr = start; len = (unsigned long) nr_pages << PAGE_SHIFT; + end = start + len; - if (unlikely(!access_ok(write ? VERIFY_WRITE : VERIFY_READ, - (void __user *)start, len))) + if (end < start) goto slow_irqon; +#ifdef CONFIG_X86_64 + if (end >> __VIRTUAL_MASK_SHIFT) + goto slow_irqon; +#endif + /* * XXX: batch / limit 'nr', to avoid large irq off latency * needs some instrumenting to determine the common sizes used by diff --git a/arch/x86/mm/highmem_32.c b/arch/x86/mm/highmem_32.c index 58f621e..2112ed5 100644 --- a/arch/x86/mm/highmem_32.c +++ b/arch/x86/mm/highmem_32.c @@ -103,6 +103,7 @@ EXPORT_SYMBOL(kmap); EXPORT_SYMBOL(kunmap); EXPORT_SYMBOL(kmap_atomic); EXPORT_SYMBOL(kunmap_atomic); +EXPORT_SYMBOL(kmap_atomic_prot); void __init set_highmem_pages_init(void) { diff --git a/arch/x86/mm/init.c b/arch/x86/mm/init.c index f53b57e..0607119 100644 --- a/arch/x86/mm/init.c +++ b/arch/x86/mm/init.c @@ -12,6 +12,7 @@ #include <asm/system.h> #include <asm/tlbflush.h> #include <asm/tlb.h> +#include <asm/proto.h> DEFINE_PER_CPU(struct mmu_gather, mmu_gathers); @@ -177,20 +178,6 @@ static int __meminit save_mr(struct map_range *mr, int nr_range, return nr_range; } -#ifdef CONFIG_X86_64 -static void __init init_gbpages(void) -{ - if (direct_gbpages && cpu_has_gbpages) - printk(KERN_INFO "Using GB pages for direct mapping\n"); - else - direct_gbpages = 0; -} -#else -static inline void init_gbpages(void) -{ -} -#endif - /* * Setup the direct mapping of the physical memory at PAGE_OFFSET. * This runs before bootmem is initialized and gets pages directly from @@ -210,9 +197,6 @@ unsigned long __init_refok init_memory_mapping(unsigned long start, printk(KERN_INFO "init_memory_mapping: %016lx-%016lx\n", start, end); - if (!after_bootmem) - init_gbpages(); - #if defined(CONFIG_DEBUG_PAGEALLOC) || defined(CONFIG_KMEMCHECK) /* * For CONFIG_DEBUG_PAGEALLOC, identity mapping will use small pages. diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c index 9c54329..ea56b8c 100644 --- a/arch/x86/mm/init_64.c +++ b/arch/x86/mm/init_64.c @@ -527,7 +527,7 @@ phys_pud_update(pgd_t *pgd, unsigned long addr, unsigned long end, return phys_pud_init(pud, addr, end, page_size_mask); } -unsigned long __init +unsigned long __meminit kernel_physical_mapping_init(unsigned long start, unsigned long end, unsigned long page_size_mask) @@ -598,6 +598,15 @@ void __init paging_init(void) sparse_memory_present_with_active_regions(MAX_NUMNODES); sparse_init(); + + /* + * clear the default setting with node 0 + * note: don't use nodes_clear here, that is really clearing when + * numa support is not compiled in, and later node_set_state + * will not set it back. + */ + node_clear_state(0, N_NORMAL_MEMORY); + free_area_init_nodes(max_zone_pfns); } @@ -787,7 +796,7 @@ int __init reserve_bootmem_generic(unsigned long phys, unsigned long len, return ret; #else - reserve_bootmem(phys, len, BOOTMEM_DEFAULT); + reserve_bootmem(phys, len, flags); #endif if (phys+len <= MAX_DMA_PFN*PAGE_SIZE) { diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c index 3cfe9ce..7e600c1 100644 --- a/arch/x86/mm/pageattr.c +++ b/arch/x86/mm/pageattr.c @@ -11,6 +11,7 @@ #include <linux/interrupt.h> #include <linux/seq_file.h> #include <linux/debugfs.h> +#include <linux/pfn.h> #include <asm/e820.h> #include <asm/processor.h> @@ -590,9 +591,12 @@ static int __change_page_attr(struct cpa_data *cpa, int primary) unsigned int level; pte_t *kpte, old_pte; - if (cpa->flags & CPA_PAGES_ARRAY) - address = (unsigned long)page_address(cpa->pages[cpa->curpage]); - else if (cpa->flags & CPA_ARRAY) + if (cpa->flags & CPA_PAGES_ARRAY) { + struct page *page = cpa->pages[cpa->curpage]; + if (unlikely(PageHighMem(page))) + return 0; + address = (unsigned long)page_address(page); + } else if (cpa->flags & CPA_ARRAY) address = cpa->vaddr[cpa->curpage]; else address = *cpa->vaddr; @@ -681,8 +685,9 @@ static int __change_page_attr_set_clr(struct cpa_data *cpa, int checkalias); static int cpa_process_alias(struct cpa_data *cpa) { struct cpa_data alias_cpa; - int ret = 0; - unsigned long temp_cpa_vaddr, vaddr; + unsigned long laddr = (unsigned long)__va(cpa->pfn << PAGE_SHIFT); + unsigned long vaddr, remapped; + int ret; if (cpa->pfn >= max_pfn_mapped) return 0; @@ -695,9 +700,12 @@ static int cpa_process_alias(struct cpa_data *cpa) * No need to redo, when the primary call touched the direct * mapping already: */ - if (cpa->flags & CPA_PAGES_ARRAY) - vaddr = (unsigned long)page_address(cpa->pages[cpa->curpage]); - else if (cpa->flags & CPA_ARRAY) + if (cpa->flags & CPA_PAGES_ARRAY) { + struct page *page = cpa->pages[cpa->curpage]; + if (unlikely(PageHighMem(page))) + return 0; + vaddr = (unsigned long)page_address(page); + } else if (cpa->flags & CPA_ARRAY) vaddr = cpa->vaddr[cpa->curpage]; else vaddr = *cpa->vaddr; @@ -706,42 +714,55 @@ static int cpa_process_alias(struct cpa_data *cpa) PAGE_OFFSET + (max_pfn_mapped << PAGE_SHIFT)))) { alias_cpa = *cpa; - temp_cpa_vaddr = (unsigned long) __va(cpa->pfn << PAGE_SHIFT); - alias_cpa.vaddr = &temp_cpa_vaddr; + alias_cpa.vaddr = &laddr; alias_cpa.flags &= ~(CPA_PAGES_ARRAY | CPA_ARRAY); - ret = __change_page_attr_set_clr(&alias_cpa, 0); + if (ret) + return ret; } #ifdef CONFIG_X86_64 - if (ret) - return ret; - /* - * No need to redo, when the primary call touched the high - * mapping already: - */ - if (within(vaddr, (unsigned long) _text, _brk_end)) - return 0; - /* - * If the physical address is inside the kernel map, we need + * If the primary call didn't touch the high mapping already + * and the physical address is inside the kernel map, we need * to touch the high mapped kernel as well: */ - if (!within(cpa->pfn, highmap_start_pfn(), highmap_end_pfn())) - return 0; + if (!within(vaddr, (unsigned long)_text, _brk_end) && + within(cpa->pfn, highmap_start_pfn(), highmap_end_pfn())) { + unsigned long temp_cpa_vaddr = (cpa->pfn << PAGE_SHIFT) + + __START_KERNEL_map - phys_base; + alias_cpa = *cpa; + alias_cpa.vaddr = &temp_cpa_vaddr; + alias_cpa.flags &= ~(CPA_PAGES_ARRAY | CPA_ARRAY); - alias_cpa = *cpa; - temp_cpa_vaddr = (cpa->pfn << PAGE_SHIFT) + __START_KERNEL_map - phys_base; - alias_cpa.vaddr = &temp_cpa_vaddr; - alias_cpa.flags &= ~(CPA_PAGES_ARRAY | CPA_ARRAY); + /* + * The high mapping range is imprecise, so ignore the + * return value. + */ + __change_page_attr_set_clr(&alias_cpa, 0); + } +#endif /* - * The high mapping range is imprecise, so ignore the return value. + * If the PMD page was partially used for per-cpu remapping, + * the recycled area needs to be split and modified. Because + * the area is always proper subset of a PMD page + * cpa->numpages is guaranteed to be 1 for these areas, so + * there's no need to loop over and check for further remaps. */ - __change_page_attr_set_clr(&alias_cpa, 0); -#endif - return ret; + remapped = (unsigned long)pcpu_lpage_remapped((void *)laddr); + if (remapped) { + WARN_ON(cpa->numpages > 1); + alias_cpa = *cpa; + alias_cpa.vaddr = &remapped; + alias_cpa.flags &= ~(CPA_PAGES_ARRAY | CPA_ARRAY); + ret = __change_page_attr_set_clr(&alias_cpa, 0); + if (ret) + return ret; + } + + return 0; } static int __change_page_attr_set_clr(struct cpa_data *cpa, int checkalias) @@ -982,12 +1003,15 @@ EXPORT_SYMBOL(set_memory_array_uc); int _set_memory_wc(unsigned long addr, int numpages) { int ret; + unsigned long addr_copy = addr; + ret = change_page_attr_set(&addr, numpages, __pgprot(_PAGE_CACHE_UC_MINUS), 0); - if (!ret) { - ret = change_page_attr_set(&addr, numpages, - __pgprot(_PAGE_CACHE_WC), 0); + ret = change_page_attr_set_clr(&addr_copy, numpages, + __pgprot(_PAGE_CACHE_WC), + __pgprot(_PAGE_CACHE_MASK), + 0, 0, NULL); } return ret; } @@ -1104,7 +1128,9 @@ int set_pages_array_uc(struct page **pages, int addrinarray) int free_idx; for (i = 0; i < addrinarray; i++) { - start = (unsigned long)page_address(pages[i]); + if (PageHighMem(pages[i])) + continue; + start = page_to_pfn(pages[i]) << PAGE_SHIFT; end = start + PAGE_SIZE; if (reserve_memtype(start, end, _PAGE_CACHE_UC_MINUS, NULL)) goto err_out; @@ -1117,7 +1143,9 @@ int set_pages_array_uc(struct page **pages, int addrinarray) err_out: free_idx = i; for (i = 0; i < free_idx; i++) { - start = (unsigned long)page_address(pages[i]); + if (PageHighMem(pages[i])) + continue; + start = page_to_pfn(pages[i]) << PAGE_SHIFT; end = start + PAGE_SIZE; free_memtype(start, end); } @@ -1146,7 +1174,9 @@ int set_pages_array_wb(struct page **pages, int addrinarray) return retval; for (i = 0; i < addrinarray; i++) { - start = (unsigned long)page_address(pages[i]); + if (PageHighMem(pages[i])) + continue; + start = page_to_pfn(pages[i]) << PAGE_SHIFT; end = start + PAGE_SIZE; free_memtype(start, end); } diff --git a/arch/x86/mm/pat.c b/arch/x86/mm/pat.c index e6718bb..352aa9e 100644 --- a/arch/x86/mm/pat.c +++ b/arch/x86/mm/pat.c @@ -623,7 +623,8 @@ static int reserve_pfn_range(u64 paddr, unsigned long size, pgprot_t *vma_prot, return ret; if (flags != want_flags) { - if (strict_prot || !is_new_memtype_allowed(want_flags, flags)) { + if (strict_prot || + !is_new_memtype_allowed(paddr, size, want_flags, flags)) { free_memtype(paddr, paddr + size); printk(KERN_ERR "%s:%d map pfn expected mapping type %s" " for %Lx-%Lx, got %s\n", diff --git a/arch/x86/mm/pgtable.c b/arch/x86/mm/pgtable.c index 8e43bdd..ed34f5e 100644 --- a/arch/x86/mm/pgtable.c +++ b/arch/x86/mm/pgtable.c @@ -25,7 +25,7 @@ pgtable_t pte_alloc_one(struct mm_struct *mm, unsigned long address) return pte; } -void __pte_free_tlb(struct mmu_gather *tlb, struct page *pte) +void ___pte_free_tlb(struct mmu_gather *tlb, struct page *pte) { pgtable_page_dtor(pte); paravirt_release_pte(page_to_pfn(pte)); @@ -33,14 +33,14 @@ void __pte_free_tlb(struct mmu_gather *tlb, struct page *pte) } #if PAGETABLE_LEVELS > 2 -void __pmd_free_tlb(struct mmu_gather *tlb, pmd_t *pmd) +void ___pmd_free_tlb(struct mmu_gather *tlb, pmd_t *pmd) { paravirt_release_pmd(__pa(pmd) >> PAGE_SHIFT); tlb_remove_page(tlb, virt_to_page(pmd)); } #if PAGETABLE_LEVELS > 3 -void __pud_free_tlb(struct mmu_gather *tlb, pud_t *pud) +void ___pud_free_tlb(struct mmu_gather *tlb, pud_t *pud) { paravirt_release_pud(__pa(pud) >> PAGE_SHIFT); tlb_remove_page(tlb, virt_to_page(pud)); @@ -329,7 +329,6 @@ void __init reserve_top_address(unsigned long reserve) printk(KERN_INFO "Reserving virtual address space above 0x%08x\n", (int)-reserve); __FIXADDR_TOP = -reserve - PAGE_SIZE; - __VMALLOC_RESERVE += reserve; #endif } diff --git a/arch/x86/mm/srat_64.c b/arch/x86/mm/srat_64.c index 2dfcbf9..dbb5381 100644 --- a/arch/x86/mm/srat_64.c +++ b/arch/x86/mm/srat_64.c @@ -79,8 +79,10 @@ static __init void bad_srat(void) acpi_numa = -1; for (i = 0; i < MAX_LOCAL_APIC; i++) apicid_to_node[i] = NUMA_NO_NODE; - for (i = 0; i < MAX_NUMNODES; i++) - nodes_add[i].start = nodes[i].end = 0; + for (i = 0; i < MAX_NUMNODES; i++) { + nodes[i].start = nodes[i].end = 0; + nodes_add[i].start = nodes_add[i].end = 0; + } remove_all_active_ranges(); } diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c index 821e970..c814e14 100644 --- a/arch/x86/mm/tlb.c +++ b/arch/x86/mm/tlb.c @@ -183,18 +183,17 @@ static void flush_tlb_others_ipi(const struct cpumask *cpumask, f->flush_mm = mm; f->flush_va = va; - cpumask_andnot(to_cpumask(f->flush_cpumask), - cpumask, cpumask_of(smp_processor_id())); - - /* - * We have to send the IPI only to - * CPUs affected. - */ - apic->send_IPI_mask(to_cpumask(f->flush_cpumask), - INVALIDATE_TLB_VECTOR_START + sender); + if (cpumask_andnot(to_cpumask(f->flush_cpumask), cpumask, cpumask_of(smp_processor_id()))) { + /* + * We have to send the IPI only to + * CPUs affected. + */ + apic->send_IPI_mask(to_cpumask(f->flush_cpumask), + INVALIDATE_TLB_VECTOR_START + sender); - while (!cpumask_empty(to_cpumask(f->flush_cpumask))) - cpu_relax(); + while (!cpumask_empty(to_cpumask(f->flush_cpumask))) + cpu_relax(); + } f->flush_mm = NULL; f->flush_va = 0; diff --git a/arch/x86/oprofile/nmi_int.c b/arch/x86/oprofile/nmi_int.c index b07dd8d..89b9a5c 100644 --- a/arch/x86/oprofile/nmi_int.c +++ b/arch/x86/oprofile/nmi_int.c @@ -390,7 +390,7 @@ static int __init p4_init(char **cpu_type) static int force_arch_perfmon; static int force_cpu_type(const char *str, struct kernel_param *kp) { - if (!strcmp(str, "archperfmon")) { + if (!strcmp(str, "arch_perfmon")) { force_arch_perfmon = 1; printk(KERN_INFO "oprofile: forcing architectural perfmon\n"); } diff --git a/arch/x86/pci/acpi.c b/arch/x86/pci/acpi.c index c0ecf25..1014eb4 100644 --- a/arch/x86/pci/acpi.c +++ b/arch/x86/pci/acpi.c @@ -38,15 +38,26 @@ count_resource(struct acpi_resource *acpi_res, void *data) struct acpi_resource_address64 addr; acpi_status status; - if (info->res_num >= PCI_BUS_NUM_RESOURCES) - return AE_OK; - status = resource_to_addr(acpi_res, &addr); if (ACPI_SUCCESS(status)) info->res_num++; return AE_OK; } +static int +bus_has_transparent_bridge(struct pci_bus *bus) +{ + struct pci_dev *dev; + + list_for_each_entry(dev, &bus->devices, bus_list) { + u16 class = dev->class >> 8; + + if (class == PCI_CLASS_BRIDGE_PCI && dev->transparent) + return true; + } + return false; +} + static acpi_status setup_resource(struct acpi_resource *acpi_res, void *data) { @@ -56,9 +67,11 @@ setup_resource(struct acpi_resource *acpi_res, void *data) acpi_status status; unsigned long flags; struct resource *root; + int max_root_bus_resources = PCI_BUS_NUM_RESOURCES; + u64 start, end; - if (info->res_num >= PCI_BUS_NUM_RESOURCES) - return AE_OK; + if (bus_has_transparent_bridge(info->bus)) + max_root_bus_resources -= 3; status = resource_to_addr(acpi_res, &addr); if (!ACPI_SUCCESS(status)) @@ -75,11 +88,22 @@ setup_resource(struct acpi_resource *acpi_res, void *data) } else return AE_OK; + start = addr.minimum + addr.translation_offset; + end = start + addr.address_length - 1; + if (info->res_num >= max_root_bus_resources) { + printk(KERN_WARNING "PCI: Failed to allocate 0x%lx-0x%lx " + "from %s for %s due to _CRS returning more than " + "%d resource descriptors\n", (unsigned long) start, + (unsigned long) end, root->name, info->name, + max_root_bus_resources); + return AE_OK; + } + res = &info->res[info->res_num]; res->name = info->name; res->flags = flags; - res->start = addr.minimum + addr.translation_offset; - res->end = res->start + addr.address_length - 1; + res->start = start; + res->end = end; res->child = NULL; if (insert_resource(root, res)) { @@ -94,23 +118,6 @@ setup_resource(struct acpi_resource *acpi_res, void *data) } static void -adjust_transparent_bridge_resources(struct pci_bus *bus) -{ - struct pci_dev *dev; - - list_for_each_entry(dev, &bus->devices, bus_list) { - int i; - u16 class = dev->class >> 8; - - if (class == PCI_CLASS_BRIDGE_PCI && dev->transparent) { - for(i = 3; i < PCI_BUS_NUM_RESOURCES; i++) - dev->subordinate->resource[i] = - dev->bus->resource[i - 3]; - } - } -} - -static void get_current_resources(struct acpi_device *device, int busnum, int domain, struct pci_bus *bus) { @@ -137,8 +144,6 @@ get_current_resources(struct acpi_device *device, int busnum, info.res_num = 0; acpi_walk_resources(device->handle, METHOD_NAME__CRS, setup_resource, &info); - if (info.res_num) - adjust_transparent_bridge_resources(bus); return; @@ -201,8 +206,15 @@ struct pci_bus * __devinit pci_acpi_scan_root(struct acpi_device *device, int do */ memcpy(bus->sysdata, sd, sizeof(*sd)); kfree(sd); - } else - bus = pci_scan_bus_parented(NULL, busnum, &pci_root_ops, sd); + } else { + bus = pci_create_bus(NULL, busnum, &pci_root_ops, sd); + if (bus) { + if (pci_probe & PCI_USE__CRS) + get_current_resources(device, busnum, domain, + bus); + bus->subordinate = pci_scan_child_bus(bus); + } + } if (!bus) kfree(sd); @@ -217,8 +229,6 @@ struct pci_bus * __devinit pci_acpi_scan_root(struct acpi_device *device, int do #endif } - if (bus && (pci_probe & PCI_USE__CRS)) - get_current_resources(device, busnum, domain, bus); return bus; } diff --git a/arch/x86/pci/amd_bus.c b/arch/x86/pci/amd_bus.c index f893d6a..3ffa10d 100644 --- a/arch/x86/pci/amd_bus.c +++ b/arch/x86/pci/amd_bus.c @@ -100,8 +100,9 @@ void x86_pci_root_bus_res_quirks(struct pci_bus *b) int j; struct pci_root_info *info; - /* don't go for it if _CRS is used */ - if (pci_probe & PCI_USE__CRS) + /* don't go for it if _CRS is used already */ + if (b->resource[0] != &ioport_resource || + b->resource[1] != &iomem_resource) return; /* if only one root bus, don't need to anything */ @@ -116,6 +117,9 @@ void x86_pci_root_bus_res_quirks(struct pci_bus *b) if (i == pci_root_num) return; + printk(KERN_DEBUG "PCI: peer root bus %02x res updated from pci conf\n", + b->number); + info = &pci_root_info[i]; for (j = 0; j < info->res_num; j++) { struct resource *res; diff --git a/arch/x86/pci/i386.c b/arch/x86/pci/i386.c index a85bef2..52e62e5 100644 --- a/arch/x86/pci/i386.c +++ b/arch/x86/pci/i386.c @@ -35,6 +35,7 @@ #include <asm/pat.h> #include <asm/e820.h> #include <asm/pci_x86.h> +#include <asm/io_apic.h> static int @@ -116,7 +117,7 @@ static void __init pcibios_allocate_bus_resources(struct list_head *bus_list) struct pci_bus *bus; struct pci_dev *dev; int idx; - struct resource *r, *pr; + struct resource *r; /* Depth-First Search on bus tree */ list_for_each_entry(bus, bus_list, node) { @@ -126,9 +127,8 @@ static void __init pcibios_allocate_bus_resources(struct list_head *bus_list) r = &dev->resource[idx]; if (!r->flags) continue; - pr = pci_find_parent_resource(dev, r); - if (!r->start || !pr || - request_resource(pr, r) < 0) { + if (!r->start || + pci_claim_resource(dev, idx) < 0) { dev_info(&dev->dev, "BAR %d: can't allocate resource\n", idx); /* * Something is wrong with the region. @@ -149,7 +149,7 @@ static void __init pcibios_allocate_resources(int pass) struct pci_dev *dev = NULL; int idx, disabled; u16 command; - struct resource *r, *pr; + struct resource *r; for_each_pci_dev(dev) { pci_read_config_word(dev, PCI_COMMAND, &command); @@ -168,8 +168,7 @@ static void __init pcibios_allocate_resources(int pass) (unsigned long long) r->start, (unsigned long long) r->end, r->flags, disabled, pass); - pr = pci_find_parent_resource(dev, r); - if (!pr || request_resource(pr, r) < 0) { + if (pci_claim_resource(dev, idx) < 0) { dev_info(&dev->dev, "BAR %d: can't allocate resource\n", idx); /* We'll assign a new address later */ r->end -= r->start; @@ -197,7 +196,7 @@ static void __init pcibios_allocate_resources(int pass) static int __init pcibios_assign_resources(void) { struct pci_dev *dev = NULL; - struct resource *r, *pr; + struct resource *r; if (!(pci_probe & PCI_ASSIGN_ROMS)) { /* @@ -209,8 +208,7 @@ static int __init pcibios_assign_resources(void) r = &dev->resource[PCI_ROM_RESOURCE]; if (!r->flags || !r->start) continue; - pr = pci_find_parent_resource(dev, r); - if (!pr || request_resource(pr, r) < 0) { + if (pci_claim_resource(dev, PCI_ROM_RESOURCE) < 0) { r->end -= r->start; r->start = 0; } @@ -230,6 +228,12 @@ void __init pcibios_resource_survey(void) pcibios_allocate_resources(1); e820_reserve_resources_late(); + /* + * Insert the IO APIC resources after PCI initialization has + * occured to handle IO APICS that are mapped in on a BAR in + * PCI space, but before trying to assign unassigned pci res. + */ + ioapic_insert_resources(); } /** diff --git a/arch/x86/pci/mmconfig-shared.c b/arch/x86/pci/mmconfig-shared.c index 8766b0e..712443e 100644 --- a/arch/x86/pci/mmconfig-shared.c +++ b/arch/x86/pci/mmconfig-shared.c @@ -523,6 +523,69 @@ reject: static int __initdata known_bridge; +static int acpi_mcfg_64bit_base_addr __initdata = FALSE; + +/* The physical address of the MMCONFIG aperture. Set from ACPI tables. */ +struct acpi_mcfg_allocation *pci_mmcfg_config; +int pci_mmcfg_config_num; + +static int __init acpi_mcfg_oem_check(struct acpi_table_mcfg *mcfg) +{ + if (!strcmp(mcfg->header.oem_id, "SGI")) + acpi_mcfg_64bit_base_addr = TRUE; + + return 0; +} + +static int __init pci_parse_mcfg(struct acpi_table_header *header) +{ + struct acpi_table_mcfg *mcfg; + unsigned long i; + int config_size; + + if (!header) + return -EINVAL; + + mcfg = (struct acpi_table_mcfg *)header; + + /* how many config structures do we have */ + pci_mmcfg_config_num = 0; + i = header->length - sizeof(struct acpi_table_mcfg); + while (i >= sizeof(struct acpi_mcfg_allocation)) { + ++pci_mmcfg_config_num; + i -= sizeof(struct acpi_mcfg_allocation); + }; + if (pci_mmcfg_config_num == 0) { + printk(KERN_ERR PREFIX "MMCONFIG has no entries\n"); + return -ENODEV; + } + + config_size = pci_mmcfg_config_num * sizeof(*pci_mmcfg_config); + pci_mmcfg_config = kmalloc(config_size, GFP_KERNEL); + if (!pci_mmcfg_config) { + printk(KERN_WARNING PREFIX + "No memory for MCFG config tables\n"); + return -ENOMEM; + } + + memcpy(pci_mmcfg_config, &mcfg[1], config_size); + + acpi_mcfg_oem_check(mcfg); + + for (i = 0; i < pci_mmcfg_config_num; ++i) { + if ((pci_mmcfg_config[i].address > 0xFFFFFFFF) && + !acpi_mcfg_64bit_base_addr) { + printk(KERN_ERR PREFIX + "MMCONFIG not in low 4GB of memory\n"); + kfree(pci_mmcfg_config); + pci_mmcfg_config_num = 0; + return -ENODEV; + } + } + + return 0; +} + static void __init __pci_mmcfg_init(int early) { /* MMCONFIG disabled */ @@ -543,7 +606,7 @@ static void __init __pci_mmcfg_init(int early) } if (!known_bridge) - acpi_table_parse(ACPI_SIG_MCFG, acpi_parse_mcfg); + acpi_table_parse(ACPI_SIG_MCFG, pci_parse_mcfg); pci_mmcfg_reject_broken(early); diff --git a/arch/x86/power/Makefile b/arch/x86/power/Makefile index de2abbd..a6a198c 100644 --- a/arch/x86/power/Makefile +++ b/arch/x86/power/Makefile @@ -1,7 +1,7 @@ # __restore_processor_state() restores %gs after S3 resume and so should not # itself be stack-protected nostackp := $(call cc-option, -fno-stack-protector) -CFLAGS_cpu_$(BITS).o := $(nostackp) +CFLAGS_cpu.o := $(nostackp) obj-$(CONFIG_PM_SLEEP) += cpu.o obj-$(CONFIG_HIBERNATION) += hibernate_$(BITS).o hibernate_asm_$(BITS).o diff --git a/arch/x86/power/cpu.c b/arch/x86/power/cpu.c index 394cbb8..9e63db8 100644 --- a/arch/x86/power/cpu.c +++ b/arch/x86/power/cpu.c @@ -226,7 +226,7 @@ static void __restore_processor_state(struct saved_context *ctxt) do_fpu_end(); mtrr_ap_init(); -#ifdef CONFIG_X86_32 +#ifdef CONFIG_X86_OLD_MCE mcheck_init(&boot_cpu_data); #endif } diff --git a/arch/x86/vdso/Makefile b/arch/x86/vdso/Makefile index 16a9020..88112b49 100644 --- a/arch/x86/vdso/Makefile +++ b/arch/x86/vdso/Makefile @@ -123,6 +123,7 @@ quiet_cmd_vdso = VDSO $@ -Wl,-T,$(filter %.lds,$^) $(filter %.o,$^) VDSO_LDFLAGS = -fPIC -shared $(call ld-option, -Wl$(comma)--hash-style=sysv) +GCOV_PROFILE := n # # Install the unstripped copy of vdso*.so listed in $(vdso-install-y). diff --git a/arch/x86/xen/Makefile b/arch/x86/xen/Makefile index 172438f..7410640 100644 --- a/arch/x86/xen/Makefile +++ b/arch/x86/xen/Makefile @@ -5,6 +5,10 @@ CFLAGS_REMOVE_time.o = -pg CFLAGS_REMOVE_irq.o = -pg endif +# Make sure early boot has no stackprotector +nostackp := $(call cc-option, -fno-stack-protector) +CFLAGS_enlighten.o := $(nostackp) + obj-y := enlighten.o setup.o multicalls.o mmu.o irq.o \ time.o xen-asm.o xen-asm_$(BITS).o \ grant-table.o suspend.o diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c index 0a1700a..eb33aaa 100644 --- a/arch/x86/xen/enlighten.c +++ b/arch/x86/xen/enlighten.c @@ -215,6 +215,7 @@ static __init void xen_init_cpuid_mask(void) (1 << X86_FEATURE_ACPI)); /* disable ACPI */ ax = 1; + cx = 0; xen_cpuid(&ax, &bx, &cx, &dx); /* cpuid claims we support xsave; try enabling it to see what happens */ @@ -974,10 +975,6 @@ asmlinkage void __init xen_start_kernel(void) xen_domain_type = XEN_PV_DOMAIN; - BUG_ON(memcmp(xen_start_info->magic, "xen-3", 5) != 0); - - xen_setup_features(); - /* Install Xen paravirt ops */ pv_info = xen_info; pv_init_ops = xen_init_ops; @@ -986,8 +983,15 @@ asmlinkage void __init xen_start_kernel(void) pv_apic_ops = xen_apic_ops; pv_mmu_ops = xen_mmu_ops; - xen_init_irq_ops(); +#ifdef CONFIG_X86_64 + /* + * Setup percpu state. We only need to do this for 64-bit + * because 32-bit already has %fs set properly. + */ + load_percpu_segment(0); +#endif + xen_init_irq_ops(); xen_init_cpuid_mask(); #ifdef CONFIG_X86_LOCAL_APIC @@ -997,6 +1001,8 @@ asmlinkage void __init xen_start_kernel(void) set_xen_basic_apic_ops(); #endif + xen_setup_features(); + if (xen_feature(XENFEAT_mmu_pt_update_preserve_ad)) { pv_mmu_ops.ptep_modify_prot_start = xen_ptep_modify_prot_start; pv_mmu_ops.ptep_modify_prot_commit = xen_ptep_modify_prot_commit; @@ -1004,13 +1010,6 @@ asmlinkage void __init xen_start_kernel(void) machine_ops = xen_machine_ops; -#ifdef CONFIG_X86_64 - /* - * Setup percpu state. We only need to do this for 64-bit - * because 32-bit already has %fs set properly. - */ - load_percpu_segment(0); -#endif /* * The only reliable way to retain the initial address of the * percpu gdt_page is to remember it here, so we can go and @@ -1061,6 +1060,7 @@ asmlinkage void __init xen_start_kernel(void) /* set up basic CPUID stuff */ cpu_detect(&new_cpu_data); new_cpu_data.hard_math = 1; + new_cpu_data.wp_works_ok = 1; new_cpu_data.x86_capability[0] = cpuid_edx(1); #endif |
