diff options
Diffstat (limited to 'arch/x86')
211 files changed, 6616 insertions, 15089 deletions
diff --git a/arch/x86/Kbuild b/arch/x86/Kbuild index 0e103236b754..0e9dec6cadd1 100644 --- a/arch/x86/Kbuild +++ b/arch/x86/Kbuild @@ -15,3 +15,4 @@ obj-y += vdso/ obj-$(CONFIG_IA32_EMULATION) += ia32/ obj-y += platform/ +obj-y += net/ diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index cc6c53a95bfd..fa2cc8c5d01c 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -8,6 +8,7 @@ config 64BIT config X86_32 def_bool !64BIT + select CLKSRC_I8253 config X86_64 def_bool 64BIT @@ -16,8 +17,6 @@ config X86_64 config X86 def_bool y select HAVE_AOUT if X86_32 - select HAVE_READQ - select HAVE_WRITEQ select HAVE_UNSTABLE_SCHED_CLOCK select HAVE_IDE select HAVE_OPROFILE @@ -71,7 +70,7 @@ config X86 select GENERIC_IRQ_SHOW select IRQ_FORCED_THREADING select USE_GENERIC_SMP_HELPERS if SMP - select ARCH_NO_SYSDEV_OPS + select HAVE_BPF_JIT if (X86_64 && NET) config INSTRUCTION_DECODER def_bool (KPROBES || PERF_EVENTS) @@ -112,7 +111,14 @@ config MMU def_bool y config ZONE_DMA - def_bool y + bool "DMA memory allocation support" if EXPERT + default y + help + DMA memory allocation support allows devices with less than 32-bit + addressing to allocate within the first 16MB of address space. + Disable if no such devices will be used. + + If unsure, say Y. config SBUS bool @@ -365,17 +371,6 @@ config X86_UV # Following is an alphabetically sorted list of 32 bit extended platforms # Please maintain the alphabetic order if and when there are additions -config X86_ELAN - bool "AMD Elan" - depends on X86_32 - depends on X86_EXTENDED_PLATFORM - ---help--- - Select this for an AMD Elan processor. - - Do not use this option for K6/Athlon/Opteron processors! - - If unsure, choose "PC-compatible" instead. - config X86_INTEL_CE bool "CE4100 TV platform" depends on PCI @@ -690,6 +685,7 @@ config AMD_IOMMU bool "AMD IOMMU support" select SWIOTLB select PCI_MSI + select PCI_IOV depends on X86_64 && PCI && ACPI ---help--- With this option you can enable support for AMD IOMMU hardware in @@ -1174,7 +1170,7 @@ comment "NUMA (Summit) requires SMP, 64GB highmem support, ACPI" config AMD_NUMA def_bool y prompt "Old style AMD Opteron NUMA detection" - depends on X86_64 && NUMA && PCI + depends on NUMA && PCI ---help--- Enable AMD NUMA node topology detection. You should say Y here if you have a multi processor AMD system. This uses an old method to @@ -1201,7 +1197,7 @@ config NODES_SPAN_OTHER_NODES config NUMA_EMU bool "NUMA emulation" - depends on X86_64 && NUMA + depends on NUMA ---help--- Enable NUMA emulation. A flat machine will be split into virtual nodes when booted with "numa=fake=N", where N is the @@ -1223,6 +1219,10 @@ config HAVE_ARCH_BOOTMEM def_bool y depends on X86_32 && NUMA +config HAVE_ARCH_ALLOC_REMAP + def_bool y + depends on X86_32 && NUMA + config ARCH_HAVE_MEMORY_PRESENT def_bool y depends on X86_32 && DISCONTIGMEM @@ -1231,13 +1231,9 @@ config NEED_NODE_MEMMAP_SIZE def_bool y depends on X86_32 && (DISCONTIGMEM || SPARSEMEM) -config HAVE_ARCH_ALLOC_REMAP - def_bool y - depends on X86_32 && NUMA - config ARCH_FLATMEM_ENABLE def_bool y - depends on X86_32 && ARCH_SELECT_MEMORY_MODEL && !NUMA + depends on X86_32 && !NUMA config ARCH_DISCONTIGMEM_ENABLE def_bool y @@ -1247,20 +1243,16 @@ config ARCH_DISCONTIGMEM_DEFAULT def_bool y depends on NUMA && X86_32 -config ARCH_PROC_KCORE_TEXT - def_bool y - depends on X86_64 && PROC_KCORE - -config ARCH_SPARSEMEM_DEFAULT - def_bool y - depends on X86_64 - config ARCH_SPARSEMEM_ENABLE def_bool y depends on X86_64 || NUMA || (EXPERIMENTAL && X86_32) || X86_32_NON_STANDARD select SPARSEMEM_STATIC if X86_32 select SPARSEMEM_VMEMMAP_ENABLE if X86_64 +config ARCH_SPARSEMEM_DEFAULT + def_bool y + depends on X86_64 + config ARCH_SELECT_MEMORY_MODEL def_bool y depends on ARCH_SPARSEMEM_ENABLE @@ -1269,6 +1261,10 @@ config ARCH_MEMORY_PROBE def_bool X86_64 depends on MEMORY_HOTPLUG +config ARCH_PROC_KCORE_TEXT + def_bool y + depends on X86_64 && PROC_KCORE + config ILLEGAL_POINTER_VALUE hex default 0 if X86_32 @@ -1703,10 +1699,6 @@ config ARCH_ENABLE_MEMORY_HOTREMOVE def_bool y depends on MEMORY_HOTPLUG -config HAVE_ARCH_EARLY_PFN_TO_NID - def_bool X86_64 - depends on NUMA - config USE_PERCPU_NUMA_NODE_ID def_bool y depends on NUMA @@ -1848,7 +1840,7 @@ config APM_ALLOW_INTS endif # APM -source "arch/x86/kernel/cpu/cpufreq/Kconfig" +source "drivers/cpufreq/Kconfig" source "drivers/cpuidle/Kconfig" @@ -2076,7 +2068,7 @@ config OLPC depends on !X86_PAE select GPIOLIB select OF - select OF_PROMTREE if PROC_DEVICETREE + select OF_PROMTREE ---help--- Add support for detecting the unique features of the OLPC XO hardware. diff --git a/arch/x86/Kconfig.cpu b/arch/x86/Kconfig.cpu index d161e939df62..6a7cfdf8ff69 100644 --- a/arch/x86/Kconfig.cpu +++ b/arch/x86/Kconfig.cpu @@ -1,6 +1,4 @@ # Put here option for CPU selection and depending optimization -if !X86_ELAN - choice prompt "Processor family" default M686 if X86_32 @@ -203,6 +201,14 @@ config MWINCHIP3D stores for this CPU, which can increase performance of some operations. +config MELAN + bool "AMD Elan" + depends on X86_32 + ---help--- + Select this for an AMD Elan processor. + + Do not use this option for K6/Athlon/Opteron processors! + config MGEODEGX1 bool "GeodeGX1" depends on X86_32 @@ -292,8 +298,6 @@ config X86_GENERIC This is really intended for distributors who need more generic optimizations. -endif - # # Define implied options from the CPU selection here config X86_INTERNODE_CACHE_SHIFT @@ -312,7 +316,7 @@ config X86_L1_CACHE_SHIFT int default "7" if MPENTIUM4 || MPSC default "6" if MK7 || MK8 || MPENTIUMM || MCORE2 || MATOM || MVIAC7 || X86_GENERIC || GENERIC_CPU - default "4" if X86_ELAN || M486 || M386 || MGEODEGX1 + default "4" if MELAN || M486 || M386 || MGEODEGX1 default "5" if MWINCHIP3D || MWINCHIPC6 || MCRUSOE || MEFFICEON || MCYRIXIII || MK6 || MPENTIUMIII || MPENTIUMII || M686 || M586MMX || M586TSC || M586 || MVIAC3_2 || MGEODE_LX config X86_XADD @@ -358,7 +362,7 @@ config X86_POPAD_OK config X86_ALIGNMENT_16 def_bool y - depends on MWINCHIP3D || MWINCHIPC6 || MCYRIXIII || X86_ELAN || MK6 || M586MMX || M586TSC || M586 || M486 || MVIAC3_2 || MGEODEGX1 + depends on MWINCHIP3D || MWINCHIPC6 || MCYRIXIII || MELAN || MK6 || M586MMX || M586TSC || M586 || M486 || MVIAC3_2 || MGEODEGX1 config X86_INTEL_USERCOPY def_bool y diff --git a/arch/x86/Kconfig.debug b/arch/x86/Kconfig.debug index 615e18810f48..c0f8a5c88910 100644 --- a/arch/x86/Kconfig.debug +++ b/arch/x86/Kconfig.debug @@ -66,26 +66,6 @@ config DEBUG_STACKOVERFLOW This option will cause messages to be printed if free stack space drops below a certain limit. -config DEBUG_STACK_USAGE - bool "Stack utilization instrumentation" - depends on DEBUG_KERNEL - ---help--- - Enables the display of the minimum amount of free stack which each - task has ever had available in the sysrq-T and sysrq-P debug output. - - This option will slow down process creation somewhat. - -config DEBUG_PER_CPU_MAPS - bool "Debug access to per_cpu maps" - depends on DEBUG_KERNEL - depends on SMP - ---help--- - Say Y to verify that the per_cpu map being accessed has - been setup. Adds a fair amount of code to kernel memory - and decreases performance. - - Say N if unsure. - config X86_PTDUMP bool "Export kernel pagetable layout to userspace via debugfs" depends on DEBUG_KERNEL diff --git a/arch/x86/Makefile_32.cpu b/arch/x86/Makefile_32.cpu index f2ee1abb1df9..86cee7b749e1 100644 --- a/arch/x86/Makefile_32.cpu +++ b/arch/x86/Makefile_32.cpu @@ -37,7 +37,7 @@ cflags-$(CONFIG_MATOM) += $(call cc-option,-march=atom,$(call cc-option,-march= $(call cc-option,-mtune=atom,$(call cc-option,-mtune=generic)) # AMD Elan support -cflags-$(CONFIG_X86_ELAN) += -march=i486 +cflags-$(CONFIG_MELAN) += -march=i486 # Geode GX1 support cflags-$(CONFIG_MGEODEGX1) += -march=pentium-mmx diff --git a/arch/x86/boot/memory.c b/arch/x86/boot/memory.c index cae3feb1035e..db75d07c3645 100644 --- a/arch/x86/boot/memory.c +++ b/arch/x86/boot/memory.c @@ -91,7 +91,7 @@ static int detect_memory_e801(void) if (oreg.ax > 15*1024) { return -1; /* Bogus! */ } else if (oreg.ax == 15*1024) { - boot_params.alt_mem_k = (oreg.dx << 6) + oreg.ax; + boot_params.alt_mem_k = (oreg.bx << 6) + oreg.ax; } else { /* * This ignores memory above 16MB if we have a memory diff --git a/arch/x86/crypto/Makefile b/arch/x86/crypto/Makefile index 1a58ad89fdf7..c04f1b7a9139 100644 --- a/arch/x86/crypto/Makefile +++ b/arch/x86/crypto/Makefile @@ -2,8 +2,6 @@ # Arch-specific CryptoAPI modules. # -obj-$(CONFIG_CRYPTO_FPU) += fpu.o - obj-$(CONFIG_CRYPTO_AES_586) += aes-i586.o obj-$(CONFIG_CRYPTO_TWOFISH_586) += twofish-i586.o obj-$(CONFIG_CRYPTO_SALSA20_586) += salsa20-i586.o @@ -24,6 +22,6 @@ aes-x86_64-y := aes-x86_64-asm_64.o aes_glue.o twofish-x86_64-y := twofish-x86_64-asm_64.o twofish_glue.o salsa20-x86_64-y := salsa20-x86_64-asm_64.o salsa20_glue.o -aesni-intel-y := aesni-intel_asm.o aesni-intel_glue.o +aesni-intel-y := aesni-intel_asm.o aesni-intel_glue.o fpu.o ghash-clmulni-intel-y := ghash-clmulni-intel_asm.o ghash-clmulni-intel_glue.o diff --git a/arch/x86/crypto/aesni-intel_glue.c b/arch/x86/crypto/aesni-intel_glue.c index 2577613fb32b..feee8ff1d05e 100644 --- a/arch/x86/crypto/aesni-intel_glue.c +++ b/arch/x86/crypto/aesni-intel_glue.c @@ -94,6 +94,10 @@ asmlinkage void aesni_cbc_enc(struct crypto_aes_ctx *ctx, u8 *out, const u8 *in, unsigned int len, u8 *iv); asmlinkage void aesni_cbc_dec(struct crypto_aes_ctx *ctx, u8 *out, const u8 *in, unsigned int len, u8 *iv); + +int crypto_fpu_init(void); +void crypto_fpu_exit(void); + #ifdef CONFIG_X86_64 asmlinkage void aesni_ctr_enc(struct crypto_aes_ctx *ctx, u8 *out, const u8 *in, unsigned int len, u8 *iv); @@ -1257,6 +1261,8 @@ static int __init aesni_init(void) return -ENODEV; } + if ((err = crypto_fpu_init())) + goto fpu_err; if ((err = crypto_register_alg(&aesni_alg))) goto aes_err; if ((err = crypto_register_alg(&__aesni_alg))) @@ -1334,6 +1340,7 @@ blk_ecb_err: __aes_err: crypto_unregister_alg(&aesni_alg); aes_err: +fpu_err: return err; } @@ -1363,6 +1370,8 @@ static void __exit aesni_exit(void) crypto_unregister_alg(&blk_ecb_alg); crypto_unregister_alg(&__aesni_alg); crypto_unregister_alg(&aesni_alg); + + crypto_fpu_exit(); } module_init(aesni_init); diff --git a/arch/x86/crypto/fpu.c b/arch/x86/crypto/fpu.c index 1a8f8649c035..98d7a188f46b 100644 --- a/arch/x86/crypto/fpu.c +++ b/arch/x86/crypto/fpu.c @@ -150,18 +150,12 @@ static struct crypto_template crypto_fpu_tmpl = { .module = THIS_MODULE, }; -static int __init crypto_fpu_module_init(void) +int __init crypto_fpu_init(void) { return crypto_register_template(&crypto_fpu_tmpl); } -static void __exit crypto_fpu_module_exit(void) +void __exit crypto_fpu_exit(void) { crypto_unregister_template(&crypto_fpu_tmpl); } - -module_init(crypto_fpu_module_init); -module_exit(crypto_fpu_module_exit); - -MODULE_LICENSE("GPL"); -MODULE_DESCRIPTION("FPU block cipher wrapper"); diff --git a/arch/x86/ia32/ia32entry.S b/arch/x86/ia32/ia32entry.S index 849a9d23c71d..95f5826be458 100644 --- a/arch/x86/ia32/ia32entry.S +++ b/arch/x86/ia32/ia32entry.S @@ -848,4 +848,5 @@ ia32_sys_call_table: .quad compat_sys_open_by_handle_at .quad compat_sys_clock_adjtime .quad sys_syncfs + .quad compat_sys_sendmmsg /* 345 */ ia32_syscall_end: diff --git a/arch/x86/include/asm/acpi.h b/arch/x86/include/asm/acpi.h index 12e0e7dd869c..416d865eae39 100644 --- a/arch/x86/include/asm/acpi.h +++ b/arch/x86/include/asm/acpi.h @@ -183,8 +183,6 @@ static inline void disable_acpi(void) { } #define ARCH_HAS_POWER_INIT 1 -struct bootnode; - #ifdef CONFIG_ACPI_NUMA extern int acpi_numa; extern int x86_acpi_numa_init(void); diff --git a/arch/x86/include/asm/alternative-asm.h b/arch/x86/include/asm/alternative-asm.h index a63a68be1cce..94d420b360d1 100644 --- a/arch/x86/include/asm/alternative-asm.h +++ b/arch/x86/include/asm/alternative-asm.h @@ -15,4 +15,13 @@ .endm #endif +.macro altinstruction_entry orig alt feature orig_len alt_len + .align 8 + .quad \orig + .quad \alt + .word \feature + .byte \orig_len + .byte \alt_len +.endm + #endif /* __ASSEMBLY__ */ diff --git a/arch/x86/include/asm/alternative.h b/arch/x86/include/asm/alternative.h index 13009d1af99a..bf535f947e8c 100644 --- a/arch/x86/include/asm/alternative.h +++ b/arch/x86/include/asm/alternative.h @@ -4,7 +4,6 @@ #include <linux/types.h> #include <linux/stddef.h> #include <linux/stringify.h> -#include <linux/jump_label.h> #include <asm/asm.h> /* @@ -191,12 +190,4 @@ extern void *text_poke(void *addr, const void *opcode, size_t len); extern void *text_poke_smp(void *addr, const void *opcode, size_t len); extern void text_poke_smp_batch(struct text_poke_param *params, int n); -#if defined(CONFIG_DYNAMIC_FTRACE) || defined(HAVE_JUMP_LABEL) -#define IDEAL_NOP_SIZE_5 5 -extern unsigned char ideal_nop5[IDEAL_NOP_SIZE_5]; -extern void arch_init_ideal_nop5(void); -#else -static inline void arch_init_ideal_nop5(void) {} -#endif - #endif /* _ASM_X86_ALTERNATIVE_H */ diff --git a/arch/x86/include/asm/amd_iommu_proto.h b/arch/x86/include/asm/amd_iommu_proto.h index 916bc8111a01..55d95eb789b3 100644 --- a/arch/x86/include/asm/amd_iommu_proto.h +++ b/arch/x86/include/asm/amd_iommu_proto.h @@ -19,13 +19,12 @@ #ifndef _ASM_X86_AMD_IOMMU_PROTO_H #define _ASM_X86_AMD_IOMMU_PROTO_H -struct amd_iommu; +#include <asm/amd_iommu_types.h> extern int amd_iommu_init_dma_ops(void); extern int amd_iommu_init_passthrough(void); +extern irqreturn_t amd_iommu_int_thread(int irq, void *data); 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_apply_erratum_63(u16 devid); extern void amd_iommu_reset_cmd_buffer(struct amd_iommu *iommu); extern int amd_iommu_init_devices(void); @@ -44,4 +43,12 @@ static inline bool is_rd890_iommu(struct pci_dev *pdev) (pdev->device == PCI_DEVICE_ID_RD890_IOMMU); } +static inline bool iommu_feature(struct amd_iommu *iommu, u64 f) +{ + if (!(iommu->cap & (1 << IOMMU_CAP_EFR))) + return false; + + return !!(iommu->features & f); +} + #endif /* _ASM_X86_AMD_IOMMU_PROTO_H */ diff --git a/arch/x86/include/asm/amd_iommu_types.h b/arch/x86/include/asm/amd_iommu_types.h index e3509fc303bf..4c9982995414 100644 --- a/arch/x86/include/asm/amd_iommu_types.h +++ b/arch/x86/include/asm/amd_iommu_types.h @@ -68,12 +68,25 @@ #define MMIO_CONTROL_OFFSET 0x0018 #define MMIO_EXCL_BASE_OFFSET 0x0020 #define MMIO_EXCL_LIMIT_OFFSET 0x0028 +#define MMIO_EXT_FEATURES 0x0030 #define MMIO_CMD_HEAD_OFFSET 0x2000 #define MMIO_CMD_TAIL_OFFSET 0x2008 #define MMIO_EVT_HEAD_OFFSET 0x2010 #define MMIO_EVT_TAIL_OFFSET 0x2018 #define MMIO_STATUS_OFFSET 0x2020 + +/* Extended Feature Bits */ +#define FEATURE_PREFETCH (1ULL<<0) +#define FEATURE_PPR (1ULL<<1) +#define FEATURE_X2APIC (1ULL<<2) +#define FEATURE_NX (1ULL<<3) +#define FEATURE_GT (1ULL<<4) +#define FEATURE_IA (1ULL<<6) +#define FEATURE_GA (1ULL<<7) +#define FEATURE_HE (1ULL<<8) +#define FEATURE_PC (1ULL<<9) + /* MMIO status bits */ #define MMIO_STATUS_COM_WAIT_INT_MASK 0x04 @@ -113,7 +126,9 @@ /* command specific defines */ #define CMD_COMPL_WAIT 0x01 #define CMD_INV_DEV_ENTRY 0x02 -#define CMD_INV_IOMMU_PAGES 0x03 +#define CMD_INV_IOMMU_PAGES 0x03 +#define CMD_INV_IOTLB_PAGES 0x04 +#define CMD_INV_ALL 0x08 #define CMD_COMPL_WAIT_STORE_MASK 0x01 #define CMD_COMPL_WAIT_INT_MASK 0x02 @@ -215,6 +230,8 @@ #define IOMMU_PTE_IR (1ULL << 61) #define IOMMU_PTE_IW (1ULL << 62) +#define DTE_FLAG_IOTLB 0x01 + #define IOMMU_PAGE_MASK (((1ULL << 52) - 1) & ~0xfffULL) #define IOMMU_PTE_PRESENT(pte) ((pte) & IOMMU_PTE_P) #define IOMMU_PTE_PAGE(pte) (phys_to_virt((pte) & IOMMU_PAGE_MASK)) @@ -227,6 +244,7 @@ /* IOMMU capabilities */ #define IOMMU_CAP_IOTLB 24 #define IOMMU_CAP_NPCACHE 26 +#define IOMMU_CAP_EFR 27 #define MAX_DOMAIN_ID 65536 @@ -249,6 +267,8 @@ extern bool amd_iommu_dump; /* global flag if IOMMUs cache non-present entries */ extern bool amd_iommu_np_cache; +/* Only true if all IOMMUs support device IOTLBs */ +extern bool amd_iommu_iotlb_sup; /* * Make iterating over all IOMMUs easier @@ -371,6 +391,9 @@ struct amd_iommu { /* flags read from acpi table */ u8 acpi_flags; + /* Extended features */ + u64 features; + /* * Capability pointer. There could be more than one IOMMU per PCI * device function if there are more than one AMD IOMMU capability @@ -409,9 +432,6 @@ struct amd_iommu { /* if one, we need to send a completion wait command */ bool need_sync; - /* becomes true if a command buffer reset is running */ - bool reset_in_progress; - /* default dma_ops domain for that IOMMU */ struct dma_ops_domain *default_dom; diff --git a/arch/x86/include/asm/amd_nb.h b/arch/x86/include/asm/amd_nb.h index 331682231bb4..67f87f257611 100644 --- a/arch/x86/include/asm/amd_nb.h +++ b/arch/x86/include/asm/amd_nb.h @@ -11,7 +11,6 @@ struct amd_nb_bus_dev_range { extern const struct pci_device_id amd_nb_misc_ids[]; extern const struct amd_nb_bus_dev_range amd_nb_bus_dev_ranges[]; -struct bootnode; extern bool early_is_amd_nb(u32 value); extern int amd_cache_northbridges(void); diff --git a/arch/x86/include/asm/apic.h b/arch/x86/include/asm/apic.h index 2b7d573be549..4a0b7c7e2cce 100644 --- a/arch/x86/include/asm/apic.h +++ b/arch/x86/include/asm/apic.h @@ -363,7 +363,12 @@ struct apic { */ int (*x86_32_early_logical_apicid)(int cpu); - /* determine CPU -> NUMA node mapping */ + /* + * Optional method called from setup_local_APIC() after logical + * apicid is guaranteed to be known to initialize apicid -> node + * mapping if NUMA initialization hasn't done so already. Don't + * add new users. + */ int (*x86_32_numa_cpu_node)(int cpu); #endif }; @@ -376,6 +381,26 @@ struct apic { extern struct apic *apic; /* + * APIC drivers are probed based on how they are listed in the .apicdrivers + * section. So the order is important and enforced by the ordering + * of different apic driver files in the Makefile. + * + * For the files having two apic drivers, we use apic_drivers() + * to enforce the order with in them. + */ +#define apic_driver(sym) \ + static struct apic *__apicdrivers_##sym __used \ + __aligned(sizeof(struct apic *)) \ + __section(.apicdrivers) = { &sym } + +#define apic_drivers(sym1, sym2) \ + static struct apic *__apicdrivers_##sym1##sym2[2] __used \ + __aligned(sizeof(struct apic *)) \ + __section(.apicdrivers) = { &sym1, &sym2 } + +extern struct apic *__apicdrivers[], *__apicdrivers_end[]; + +/* * APIC functionality to boot other CPUs - only used on SMP: */ #ifdef CONFIG_SMP @@ -453,15 +478,10 @@ static inline unsigned default_get_apic_id(unsigned long x) #define DEFAULT_TRAMPOLINE_PHYS_HIGH 0x469 #ifdef CONFIG_X86_64 -extern struct apic apic_flat; -extern struct apic apic_physflat; -extern struct apic apic_x2apic_cluster; -extern struct apic apic_x2apic_phys; extern int default_acpi_madt_oem_check(char *, char *); extern void apic_send_IPI_self(int vector); -extern struct apic apic_x2apic_uv_x; DECLARE_PER_CPU(int, x2apic_extra_bits); extern int default_cpu_present_to_apicid(int mps_cpu); @@ -475,7 +495,7 @@ static inline void default_wait_for_init_deassert(atomic_t *deassert) return; } -extern void generic_bigsmp_probe(void); +extern struct apic *generic_bigsmp_probe(void); #ifdef CONFIG_X86_LOCAL_APIC @@ -511,8 +531,6 @@ extern struct apic apic_noop; #ifdef CONFIG_X86_32 -extern struct apic apic_default; - static inline int noop_x86_32_early_logical_apicid(int cpu) { return BAD_APICID; @@ -537,8 +555,6 @@ static inline int default_phys_pkg_id(int cpuid_apic, int index_msb) return cpuid_apic >> index_msb; } -extern int default_x86_32_numa_cpu_node(int cpu); - #endif static inline unsigned int diff --git a/arch/x86/include/asm/apicdef.h b/arch/x86/include/asm/apicdef.h index d87988bacf3e..34595d5e1038 100644 --- a/arch/x86/include/asm/apicdef.h +++ b/arch/x86/include/asm/apicdef.h @@ -78,6 +78,7 @@ #define APIC_DEST_LOGICAL 0x00800 #define APIC_DEST_PHYSICAL 0x00000 #define APIC_DM_FIXED 0x00000 +#define APIC_DM_FIXED_MASK 0x00700 #define APIC_DM_LOWEST 0x00100 #define APIC_DM_SMI 0x00200 #define APIC_DM_REMRD 0x00300 diff --git a/arch/x86/include/asm/bios_ebda.h b/arch/x86/include/asm/bios_ebda.h index 3c7521063d3f..aa6a3170ab5a 100644 --- a/arch/x86/include/asm/bios_ebda.h +++ b/arch/x86/include/asm/bios_ebda.h @@ -4,16 +4,40 @@ #include <asm/io.h> /* - * there is a real-mode segmented pointer pointing to the - * 4K EBDA area at 0x40E. + * Returns physical address of EBDA. Returns 0 if there is no EBDA. */ static inline unsigned int get_bios_ebda(void) { + /* + * There is a real-mode segmented pointer pointing to the + * 4K EBDA area at 0x40E. + */ unsigned int address = *(unsigned short *)phys_to_virt(0x40E); address <<= 4; return address; /* 0 means none */ } +/* + * Return the sanitized length of the EBDA in bytes, if it exists. + */ +static inline unsigned int get_bios_ebda_length(void) +{ + unsigned int address; + unsigned int length; + + address = get_bios_ebda(); + if (!address) + return 0; + + /* EBDA length is byte 0 of the EBDA (stored in KiB) */ + length = *(unsigned char *)phys_to_virt(address); + length <<= 10; + + /* Trim the length if it extends beyond 640KiB */ + length = min_t(unsigned int, (640 * 1024) - address, length); + return length; +} + void reserve_ebda_region(void); #ifdef CONFIG_X86_CHECK_BIOS_CORRUPTION diff --git a/arch/x86/include/asm/cpufeature.h b/arch/x86/include/asm/cpufeature.h index 91f3e087cf21..5dc6acc98dbd 100644 --- a/arch/x86/include/asm/cpufeature.h +++ b/arch/x86/include/asm/cpufeature.h @@ -195,6 +195,8 @@ /* Intel-defined CPU features, CPUID level 0x00000007:0 (ebx), word 9 */ #define X86_FEATURE_FSGSBASE (9*32+ 0) /* {RD/WR}{FS/GS}BASE instructions*/ +#define X86_FEATURE_SMEP (9*32+ 7) /* Supervisor Mode Execution Protection */ +#define X86_FEATURE_ERMS (9*32+ 9) /* Enhanced REP MOVSB/STOSB */ #if defined(__KERNEL__) && !defined(__ASSEMBLY__) @@ -207,8 +209,7 @@ extern const char * const x86_power_flags[32]; #define test_cpu_cap(c, bit) \ test_bit(bit, (unsigned long *)((c)->x86_capability)) -#define cpu_has(c, bit) \ - (__builtin_constant_p(bit) && \ +#define REQUIRED_MASK_BIT_SET(bit) \ ( (((bit)>>5)==0 && (1UL<<((bit)&31) & REQUIRED_MASK0)) || \ (((bit)>>5)==1 && (1UL<<((bit)&31) & REQUIRED_MASK1)) || \ (((bit)>>5)==2 && (1UL<<((bit)&31) & REQUIRED_MASK2)) || \ @@ -218,10 +219,16 @@ extern const char * const x86_power_flags[32]; (((bit)>>5)==6 && (1UL<<((bit)&31) & REQUIRED_MASK6)) || \ (((bit)>>5)==7 && (1UL<<((bit)&31) & REQUIRED_MASK7)) || \ (((bit)>>5)==8 && (1UL<<((bit)&31) & REQUIRED_MASK8)) || \ - (((bit)>>5)==9 && (1UL<<((bit)&31) & REQUIRED_MASK9)) ) \ - ? 1 : \ + (((bit)>>5)==9 && (1UL<<((bit)&31) & REQUIRED_MASK9)) ) + +#define cpu_has(c, bit) \ + (__builtin_constant_p(bit) && REQUIRED_MASK_BIT_SET(bit) ? 1 : \ test_cpu_cap(c, bit)) +#define this_cpu_has(bit) \ + (__builtin_constant_p(bit) && REQUIRED_MASK_BIT_SET(bit) ? 1 : \ + x86_this_cpu_test_bit(bit, (unsigned long *)&cpu_info.x86_capability)) + #define boot_cpu_has(bit) cpu_has(&boot_cpu_data, bit) #define set_cpu_cap(c, bit) set_bit(bit, (unsigned long *)((c)->x86_capability)) diff --git a/arch/x86/include/asm/dma.h b/arch/x86/include/asm/dma.h index 057099e5faba..0bdb0c54d9a1 100644 --- a/arch/x86/include/asm/dma.h +++ b/arch/x86/include/asm/dma.h @@ -69,22 +69,18 @@ #define MAX_DMA_CHANNELS 8 -#ifdef CONFIG_X86_32 - -/* The maximum address that we can perform a DMA transfer to on this platform */ -#define MAX_DMA_ADDRESS (PAGE_OFFSET + 0x1000000) - -#else - /* 16MB ISA DMA zone */ #define MAX_DMA_PFN ((16 * 1024 * 1024) >> PAGE_SHIFT) /* 4GB broken PCI/AGP hardware bus master zone */ #define MAX_DMA32_PFN ((4UL * 1024 * 1024 * 1024) >> PAGE_SHIFT) +#ifdef CONFIG_X86_32 +/* The maximum address that we can perform a DMA transfer to on this platform */ +#define MAX_DMA_ADDRESS (PAGE_OFFSET + 0x1000000) +#else /* Compat define for old dma zone */ #define MAX_DMA_ADDRESS ((unsigned long)__va(MAX_DMA_PFN << PAGE_SHIFT)) - #endif /* 8237 DMA controllers */ diff --git a/arch/x86/include/asm/efi.h b/arch/x86/include/asm/efi.h index 8e4a16508d4e..7093e4a6a0bc 100644 --- a/arch/x86/include/asm/efi.h +++ b/arch/x86/include/asm/efi.h @@ -90,6 +90,7 @@ extern void __iomem *efi_ioremap(unsigned long addr, unsigned long size, #endif /* CONFIG_X86_32 */ extern int add_efi_memmap; +extern void efi_set_executable(efi_memory_desc_t *md, bool executable); extern void efi_memblock_x86_reserve_range(void); extern void efi_call_phys_prelog(void); extern void efi_call_phys_epilog(void); diff --git a/arch/x86/include/asm/ftrace.h b/arch/x86/include/asm/ftrace.h index db24c2278be0..268c783ab1c0 100644 --- a/arch/x86/include/asm/ftrace.h +++ b/arch/x86/include/asm/ftrace.h @@ -38,11 +38,10 @@ extern void mcount(void); static inline unsigned long ftrace_call_adjust(unsigned long addr) { /* - * call mcount is "e8 <4 byte offset>" - * The addr points to the 4 byte offset and the caller of this - * function wants the pointer to e8. Simply subtract one. + * addr is the address of the mcount call instruction. + * recordmcount does the necessary offset calculation. */ - return addr - 1; + return addr; } #ifdef CONFIG_DYNAMIC_FTRACE diff --git a/arch/x86/include/asm/i8253.h b/arch/x86/include/asm/i8253.h index fc1f579fb965..65aaa91d5850 100644 --- a/arch/x86/include/asm/i8253.h +++ b/arch/x86/include/asm/i8253.h @@ -6,6 +6,8 @@ #define PIT_CH0 0x40 #define PIT_CH2 0x42 +#define PIT_LATCH LATCH + extern raw_spinlock_t i8253_lock; extern struct clock_event_device *global_clock_event; diff --git a/arch/x86/include/asm/io.h b/arch/x86/include/asm/io.h index 072273082528..d02804d650c4 100644 --- a/arch/x86/include/asm/io.h +++ b/arch/x86/include/asm/io.h @@ -38,7 +38,6 @@ #include <linux/string.h> #include <linux/compiler.h> -#include <asm-generic/int-ll64.h> #include <asm/page.h> #include <xen/xen.h> @@ -87,27 +86,6 @@ build_mmio_write(__writel, "l", unsigned int, "r", ) build_mmio_read(readq, "q", unsigned long, "=r", :"memory") build_mmio_write(writeq, "q", unsigned long, "r", :"memory") -#else - -static inline __u64 readq(const volatile void __iomem *addr) -{ - const volatile u32 __iomem *p = addr; - u32 low, high; - - low = readl(p); - high = readl(p + 1); - - return low + ((u64)high << 32); -} - -static inline void writeq(__u64 val, volatile void __iomem *addr) -{ - writel(val, addr); - writel(val >> 32, addr+4); -} - -#endif - #define readq_relaxed(a) readq(a) #define __raw_readq(a) readq(a) @@ -117,6 +95,8 @@ static inline void writeq(__u64 val, volatile void __iomem *addr) #define readq readq #define writeq writeq +#endif + /** * virt_to_phys - map virtual addresses to physical * @address: address to remap diff --git a/arch/x86/include/asm/io_apic.h b/arch/x86/include/asm/io_apic.h index c4bd267dfc50..690d1cc9a877 100644 --- a/arch/x86/include/asm/io_apic.h +++ b/arch/x86/include/asm/io_apic.h @@ -105,12 +105,12 @@ struct IR_IO_APIC_route_entry { * # of IO-APICs and # of IRQ routing registers */ extern int nr_ioapics; -extern int nr_ioapic_registers[MAX_IO_APICS]; -#define MP_MAX_IOAPIC_PIN 127 +extern int mpc_ioapic_id(int ioapic); +extern unsigned int mpc_ioapic_addr(int ioapic); +extern struct mp_ioapic_gsi *mp_ioapic_gsi_routing(int ioapic); -/* I/O APIC entries */ -extern struct mpc_ioapic mp_ioapics[MAX_IO_APICS]; +#define MP_MAX_IOAPIC_PIN 127 /* # of MP IRQ source entries */ extern int mp_irq_entries; @@ -150,13 +150,11 @@ void setup_IO_APIC_irq_extra(u32 gsi); extern void ioapic_and_gsi_init(void); extern void ioapic_insert_resources(void); -int io_apic_setup_irq_pin(unsigned int irq, int node, struct io_apic_irq_attr *attr); +int io_apic_setup_irq_pin_once(unsigned int irq, int node, struct io_apic_irq_attr *attr); -extern struct IO_APIC_route_entry **alloc_ioapic_entries(void); -extern void free_ioapic_entries(struct IO_APIC_route_entry **ioapic_entries); -extern int save_IO_APIC_setup(struct IO_APIC_route_entry **ioapic_entries); -extern void mask_IO_APIC_setup(struct IO_APIC_route_entry **ioapic_entries); -extern int restore_IO_APIC_setup(struct IO_APIC_route_entry **ioapic_entries); +extern int save_ioapic_entries(void); +extern void mask_ioapic_entries(void); +extern int restore_ioapic_entries(void); extern int get_nr_irqs_gsi(void); @@ -192,19 +190,13 @@ struct io_apic_irq_attr; static inline int io_apic_set_pci_routing(struct device *dev, int irq, struct io_apic_irq_attr *irq_attr) { return 0; } -static inline struct IO_APIC_route_entry **alloc_ioapic_entries(void) -{ - return NULL; -} - -static inline void free_ioapic_entries(struct IO_APIC_route_entry **ent) { } -static inline int save_IO_APIC_setup(struct IO_APIC_route_entry **ent) +static inline int save_ioapic_entries(void) { return -ENOMEM; } -static inline void mask_IO_APIC_setup(struct IO_APIC_route_entry **ent) { } -static inline int restore_IO_APIC_setup(struct IO_APIC_route_entry **ent) +static inline void mask_ioapic_entries(void) { } +static inline int restore_ioapic_entries(void) { return -ENOMEM; } diff --git a/arch/x86/include/asm/jump_label.h b/arch/x86/include/asm/jump_label.h index 574dbc22893a..a32b18ce6ead 100644 --- a/arch/x86/include/asm/jump_label.h +++ b/arch/x86/include/asm/jump_label.h @@ -5,20 +5,25 @@ #include <linux/types.h> #include <asm/nops.h> +#include <asm/asm.h> #define JUMP_LABEL_NOP_SIZE 5 -# define JUMP_LABEL_INITIAL_NOP ".byte 0xe9 \n\t .long 0\n\t" - -# define JUMP_LABEL(key, label) \ - do { \ - asm goto("1:" \ - JUMP_LABEL_INITIAL_NOP \ - ".pushsection __jump_table, \"aw\" \n\t"\ - _ASM_PTR "1b, %l[" #label "], %c0 \n\t" \ - ".popsection \n\t" \ - : : "i" (key) : : label); \ - } while (0) +#define JUMP_LABEL_INITIAL_NOP ".byte 0xe9 \n\t .long 0\n\t" + +static __always_inline bool arch_static_branch(struct jump_label_key *key) +{ + asm goto("1:" + JUMP_LABEL_INITIAL_NOP + ".pushsection __jump_table, \"aw\" \n\t" + _ASM_ALIGN "\n\t" + _ASM_PTR "1b, %l[l_yes], %c0 \n\t" + ".popsection \n\t" + : : "i" (key) : : l_yes); + return false; +l_yes: + return true; +} #endif /* __KERNEL__ */ diff --git a/arch/x86/include/asm/kvm_emulate.h b/arch/x86/include/asm/kvm_emulate.h index 0f5213564326..0049211959c0 100644 --- a/arch/x86/include/asm/kvm_emulate.h +++ b/arch/x86/include/asm/kvm_emulate.h @@ -14,6 +14,8 @@ #include <asm/desc_defs.h> struct x86_emulate_ctxt; +enum x86_intercept; +enum x86_intercept_stage; struct x86_exception { u8 vector; @@ -24,6 +26,24 @@ struct x86_exception { }; /* + * This struct is used to carry enough information from the instruction + * decoder to main KVM so that a decision can be made whether the + * instruction needs to be intercepted or not. + */ +struct x86_instruction_info { + u8 intercept; /* which intercept */ + u8 rep_prefix; /* rep prefix? */ + u8 modrm_mod; /* mod part of modrm */ + u8 modrm_reg; /* index of register used */ + u8 modrm_rm; /* rm part of modrm */ + u64 src_val; /* value of source operand */ + u8 src_bytes; /* size of source operand */ + u8 dst_bytes; /* size of destination operand */ + u8 ad_bytes; /* size of src/dst address */ + u64 next_rip; /* rip following the instruction */ +}; + +/* * x86_emulate_ops: * * These operations represent the instruction emulator's interface to memory. @@ -62,6 +82,7 @@ struct x86_exception { #define X86EMUL_RETRY_INSTR 3 /* retry the instruction for some reason */ #define X86EMUL_CMPXCHG_FAILED 4 /* cmpxchg did not see expected value */ #define X86EMUL_IO_NEEDED 5 /* IO is needed to complete emulation */ +#define X86EMUL_INTERCEPTED 6 /* Intercepted by nested VMCB/VMCS */ struct x86_emulate_ops { /* @@ -71,8 +92,9 @@ struct x86_emulate_ops { * @val: [OUT] Value read from memory, zero-extended to 'u_long'. * @bytes: [IN ] Number of bytes to read from memory. */ - int (*read_std)(unsigned long addr, void *val, - unsigned int bytes, struct kvm_vcpu *vcpu, + int (*read_std)(struct x86_emulate_ctxt *ctxt, + unsigned long addr, void *val, + unsigned int bytes, struct x86_exception *fault); /* @@ -82,8 +104,8 @@ struct x86_emulate_ops { * @val: [OUT] Value write to memory, zero-extended to 'u_long'. * @bytes: [IN ] Number of bytes to write to memory. */ - int (*write_std)(unsigned long addr, void *val, - unsigned int bytes, struct kvm_vcpu *vcpu, + int (*write_std)(struct x86_emulate_ctxt *ctxt, + unsigned long addr, void *val, unsigned int bytes, struct x86_exception *fault); /* * fetch: Read bytes of standard (non-emulated/special) memory. @@ -92,8 +114,8 @@ struct x86_emulate_ops { * @val: [OUT] Value read from memory, zero-extended to 'u_long'. * @bytes: [IN ] Number of bytes to read from memory. */ - int (*fetch)(unsigned long addr, void *val, - unsigned int bytes, struct kvm_vcpu *vcpu, + int (*fetch)(struct x86_emulate_ctxt *ctxt, + unsigned long addr, void *val, unsigned int bytes, struct x86_exception *fault); /* @@ -102,11 +124,9 @@ struct x86_emulate_ops { * @val: [OUT] Value read from memory, zero-extended to 'u_long'. * @bytes: [IN ] Number of bytes to read from memory. */ - int (*read_emulated)(unsigned long addr, - void *val, - unsigned int bytes, - struct x86_exception *fault, - struct kvm_vcpu *vcpu); + int (*read_emulated)(struct x86_emulate_ctxt *ctxt, + unsigned long addr, void *val, unsigned int bytes, + struct x86_exception *fault); /* * write_emulated: Write bytes to emulated/special memory area. @@ -115,11 +135,10 @@ struct x86_emulate_ops { * required). * @bytes: [IN ] Number of bytes to write to memory. */ - int (*write_emulated)(unsigned long addr, - const void *val, + int (*write_emulated)(struct x86_emulate_ctxt *ctxt, + unsigned long addr, const void *val, unsigned int bytes, - struct x86_exception *fault, - struct kvm_vcpu *vcpu); + struct x86_exception *fault); /* * cmpxchg_emulated: Emulate an atomic (LOCKed) CMPXCHG operation on an @@ -129,40 +148,54 @@ struct x86_emulate_ops { * @new: [IN ] Value to write to @addr. * @bytes: [IN ] Number of bytes to access using CMPXCHG. */ - int (*cmpxchg_emulated)(unsigned long addr, + int (*cmpxchg_emulated)(struct x86_emulate_ctxt *ctxt, + unsigned long addr, const void *old, const void *new, unsigned int bytes, - struct x86_exception *fault, - struct kvm_vcpu *vcpu); - - int (*pio_in_emulated)(int size, unsigned short port, void *val, - unsigned int count, struct kvm_vcpu *vcpu); - - int (*pio_out_emulated)(int size, unsigned short port, const void *val, - unsigned int count, struct kvm_vcpu *vcpu); - - bool (*get_cached_descriptor)(struct desc_struct *desc, u32 *base3, - int seg, struct kvm_vcpu *vcpu); - void (*set_cached_descriptor)(struct desc_struct *desc, u32 base3, - int seg, struct kvm_vcpu *vcpu); - u16 (*get_segment_selector)(int seg, struct kvm_vcpu *vcpu); - void (*set_segment_selector)(u16 sel, int seg, struct kvm_vcpu *vcpu); - unsigned long (*get_cached_segment_base)(int seg, struct kvm_vcpu *vcpu); - void (*get_gdt)(struct desc_ptr *dt, struct kvm_vcpu *vcpu); - void (*get_idt)(struct desc_ptr *dt, struct kvm_vcpu *vcpu); - ulong (*get_cr)(int cr, struct kvm_vcpu *vcpu); - int (*set_cr)(int cr, ulong val, struct kvm_vcpu *vcpu); - int (*cpl)(struct kvm_vcpu *vcpu); - int (*get_dr)(int dr, unsigned long *dest, struct kvm_vcpu *vcpu); - int (*set_dr)(int dr, unsigned long value, struct kvm_vcpu *vcpu); - int (*set_msr)(struct kvm_vcpu *vcpu, u32 msr_index, u64 data); - int (*get_msr)(struct kvm_vcpu *vcpu, u32 msr_index, u64 *pdata); + struct x86_exception *fault); + void (*invlpg)(struct x86_emulate_ctxt *ctxt, ulong addr); + + int (*pio_in_emulated)(struct x86_emulate_ctxt *ctxt, + int size, unsigned short port, void *val, + unsigned int count); + + int (*pio_out_emulated)(struct x86_emulate_ctxt *ctxt, + int size, unsigned short port, const void *val, + unsigned int count); + + bool (*get_segment)(struct x86_emulate_ctxt *ctxt, u16 *selector, + struct desc_struct *desc, u32 *base3, int seg); + void (*set_segment)(struct x86_emulate_ctxt *ctxt, u16 selector, + struct desc_struct *desc, u32 base3, int seg); + unsigned long (*get_cached_segment_base)(struct x86_emulate_ctxt *ctxt, + int seg); + void (*get_gdt)(struct x86_emulate_ctxt *ctxt, struct desc_ptr *dt); + void (*get_idt)(struct x86_emulate_ctxt *ctxt, struct desc_ptr *dt); + void (*set_gdt)(struct x86_emulate_ctxt *ctxt, struct desc_ptr *dt); + void (*set_idt)(struct x86_emulate_ctxt *ctxt, struct desc_ptr *dt); + ulong (*get_cr)(struct x86_emulate_ctxt *ctxt, int cr); + int (*set_cr)(struct x86_emulate_ctxt *ctxt, int cr, ulong val); + int (*cpl)(struct x86_emulate_ctxt *ctxt); + int (*get_dr)(struct x86_emulate_ctxt *ctxt, int dr, ulong *dest); + int (*set_dr)(struct x86_emulate_ctxt *ctxt, int dr, ulong value); + int (*set_msr)(struct x86_emulate_ctxt *ctxt, u32 msr_index, u64 data); + int (*get_msr)(struct x86_emulate_ctxt *ctxt, u32 msr_index, u64 *pdata); + void (*halt)(struct x86_emulate_ctxt *ctxt); + void (*wbinvd)(struct x86_emulate_ctxt *ctxt); + int (*fix_hypercall)(struct x86_emulate_ctxt *ctxt); + void (*get_fpu)(struct x86_emulate_ctxt *ctxt); /* disables preempt */ + void (*put_fpu)(struct x86_emulate_ctxt *ctxt); /* reenables preempt */ + int (*intercept)(struct x86_emulate_ctxt *ctxt, + struct x86_instruction_info *info, + enum x86_intercept_stage stage); }; +typedef u32 __attribute__((vector_size(16))) sse128_t; + /* Type, address-of, and value of an instruction's operand. */ struct operand { - enum { OP_REG, OP_MEM, OP_IMM, OP_NONE } type; + enum { OP_REG, OP_MEM, OP_IMM, OP_XMM, OP_NONE } type; unsigned int bytes; union { unsigned long orig_val; @@ -174,11 +207,13 @@ struct operand { ulong ea; unsigned seg; } mem; + unsigned xmm; } addr; union { unsigned long val; u64 val64; char valptr[sizeof(unsigned long) + 2]; + sse128_t vec_val; }; }; @@ -197,6 +232,7 @@ struct read_cache { struct decode_cache { u8 twobyte; u8 b; + u8 intercept; u8 lock_prefix; u8 rep_prefix; u8 op_bytes; @@ -209,6 +245,7 @@ struct decode_cache { u8 seg_override; unsigned int d; int (*execute)(struct x86_emulate_ctxt *ctxt); + int (*check_perm)(struct x86_emulate_ctxt *ctxt); unsigned long regs[NR_VCPU_REGS]; unsigned long eip; /* modrm */ @@ -227,17 +264,15 @@ struct x86_emulate_ctxt { struct x86_emulate_ops *ops; /* Register state before/after emulation. */ - struct kvm_vcpu *vcpu; - unsigned long eflags; unsigned long eip; /* eip before instruction emulation */ /* Emulated execution mode, represented by an X86EMUL_MODE value. */ int mode; - u32 cs_base; /* interruptibility state, as a result of execution of STI or MOV SS */ int interruptibility; + bool guest_mode; /* guest running a nested guest */ bool perm_ok; /* do not check permissions if true */ bool only_vendor_specific_insn; @@ -249,8 +284,8 @@ struct x86_emulate_ctxt { }; /* Repeat String Operation Prefix */ -#define REPE_PREFIX 1 -#define REPNE_PREFIX 2 +#define REPE_PREFIX 0xf3 +#define REPNE_PREFIX 0xf2 /* Execution mode, passed to the emulator. */ #define X86EMUL_MODE_REAL 0 /* Real mode. */ @@ -259,6 +294,69 @@ struct x86_emulate_ctxt { #define X86EMUL_MODE_PROT32 4 /* 32-bit protected mode. */ #define X86EMUL_MODE_PROT64 8 /* 64-bit (long) mode. */ +/* any protected mode */ +#define X86EMUL_MODE_PROT (X86EMUL_MODE_PROT16|X86EMUL_MODE_PROT32| \ + X86EMUL_MODE_PROT64) + +enum x86_intercept_stage { + X86_ICTP_NONE = 0, /* Allow zero-init to not match anything */ + X86_ICPT_PRE_EXCEPT, + X86_ICPT_POST_EXCEPT, + X86_ICPT_POST_MEMACCESS, +}; + +enum x86_intercept { + x86_intercept_none, + x86_intercept_cr_read, + x86_intercept_cr_write, + x86_intercept_clts, + x86_intercept_lmsw, + x86_intercept_smsw, + x86_intercept_dr_read, + x86_intercept_dr_write, + x86_intercept_lidt, + x86_intercept_sidt, + x86_intercept_lgdt, + x86_intercept_sgdt, + x86_intercept_lldt, + x86_intercept_sldt, + x86_intercept_ltr, + x86_intercept_str, + x86_intercept_rdtsc, + x86_intercept_rdpmc, + x86_intercept_pushf, + x86_intercept_popf, + x86_intercept_cpuid, + x86_intercept_rsm, + x86_intercept_iret, + x86_intercept_intn, + x86_intercept_invd, + x86_intercept_pause, + x86_intercept_hlt, + x86_intercept_invlpg, + x86_intercept_invlpga, + x86_intercept_vmrun, + x86_intercept_vmload, + x86_intercept_vmsave, + x86_intercept_vmmcall, + x86_intercept_stgi, + x86_intercept_clgi, + x86_intercept_skinit, + x86_intercept_rdtscp, + x86_intercept_icebp, + x86_intercept_wbinvd, + x86_intercept_monitor, + x86_intercept_mwait, + x86_intercept_rdmsr, + x86_intercept_wrmsr, + x86_intercept_in, + x86_intercept_ins, + x86_intercept_out, + x86_intercept_outs, + + nr_x86_intercepts +}; + /* Host execution mode. */ #if defined(CONFIG_X86_32) #define X86EMUL_MODE_HOST X86EMUL_MODE_PROT32 @@ -270,6 +368,7 @@ int x86_decode_insn(struct x86_emulate_ctxt *ctxt, void *insn, int insn_len); #define EMULATION_FAILED -1 #define EMULATION_OK 0 #define EMULATION_RESTART 1 +#define EMULATION_INTERCEPTED 2 int x86_emulate_insn(struct x86_emulate_ctxt *ctxt); int emulator_task_switch(struct x86_emulate_ctxt *ctxt, u16 tss_selector, int reason, diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index c8af0991fdf0..d2ac8e2ee897 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -30,14 +30,30 @@ #define KVM_MEMORY_SLOTS 32 /* memory slots that does not exposed to userspace */ #define KVM_PRIVATE_MEM_SLOTS 4 +#define KVM_MMIO_SIZE 16 #define KVM_PIO_PAGE_OFFSET 1 #define KVM_COALESCED_MMIO_PAGE_OFFSET 2 +#define CR0_RESERVED_BITS \ + (~(unsigned long)(X86_CR0_PE | X86_CR0_MP | X86_CR0_EM | X86_CR0_TS \ + | X86_CR0_ET | X86_CR0_NE | X86_CR0_WP | X86_CR0_AM \ + | X86_CR0_NW | X86_CR0_CD | X86_CR0_PG)) + #define CR3_PAE_RESERVED_BITS ((X86_CR3_PWT | X86_CR3_PCD) - 1) #define CR3_NONPAE_RESERVED_BITS ((PAGE_SIZE-1) & ~(X86_CR3_PWT | X86_CR3_PCD)) #define CR3_L_MODE_RESERVED_BITS (CR3_NONPAE_RESERVED_BITS | \ 0xFFFFFF0000000000ULL) +#define CR4_RESERVED_BITS \ + (~(unsigned long)(X86_CR4_VME | X86_CR4_PVI | X86_CR4_TSD | X86_CR4_DE\ + | X86_CR4_PSE | X86_CR4_PAE | X86_CR4_MCE \ + | X86_CR4_PGE | X86_CR4_PCE | X86_CR4_OSFXSR \ + | X86_CR4_OSXSAVE \ + | X86_CR4_OSXMMEXCPT | X86_CR4_VMXE)) + +#define CR8_RESERVED_BITS (~(unsigned long)X86_CR8_TPR) + + #define INVALID_PAGE (~(hpa_t)0) #define VALID_PAGE(x) ((x) != INVALID_PAGE) @@ -118,6 +134,9 @@ enum kvm_reg { enum kvm_reg_ex { VCPU_EXREG_PDPTR = NR_VCPU_REGS, VCPU_EXREG_CR3, + VCPU_EXREG_RFLAGS, + VCPU_EXREG_CPL, + VCPU_EXREG_SEGMENTS, }; enum { @@ -256,7 +275,7 @@ struct kvm_mmu { struct kvm_mmu_page *sp); void (*invlpg)(struct kvm_vcpu *vcpu, gva_t gva); void (*update_pte)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp, - u64 *spte, const void *pte, unsigned long mmu_seq); + u64 *spte, const void *pte); hpa_t root_hpa; int root_level; int shadow_root_level; @@ -340,7 +359,6 @@ struct kvm_vcpu_arch { struct fpu guest_fpu; u64 xcr0; - gva_t mmio_fault_cr2; struct kvm_pio_request pio; void *pio_data; @@ -367,18 +385,22 @@ struct kvm_vcpu_arch { /* emulate context */ struct x86_emulate_ctxt emulate_ctxt; + bool emulate_regs_need_sync_to_vcpu; + bool emulate_regs_need_sync_from_vcpu; gpa_t time; struct pvclock_vcpu_time_info hv_clock; unsigned int hw_tsc_khz; unsigned int time_offset; struct page *time_page; - u64 last_host_tsc; u64 last_guest_tsc; u64 last_kernel_ns; u64 last_tsc_nsec; u64 last_tsc_write; + u32 virtual_tsc_khz; bool tsc_catchup; + u32 tsc_catchup_mult; + s8 tsc_catchup_shift; bool nmi_pending; bool nmi_injected; @@ -448,9 +470,6 @@ struct kvm_arch { u64 last_tsc_nsec; u64 last_tsc_offset; u64 last_tsc_write; - u32 virtual_tsc_khz; - u32 virtual_tsc_mult; - s8 virtual_tsc_shift; struct kvm_xen_hvm_config xen_hvm_config; @@ -502,6 +521,8 @@ struct kvm_vcpu_stat { u32 nmi_injections; }; +struct x86_instruction_info; + struct kvm_x86_ops { int (*cpu_has_kvm_support)(void); /* __init */ int (*disabled_by_bios)(void); /* __init */ @@ -586,9 +607,17 @@ struct kvm_x86_ops { bool (*has_wbinvd_exit)(void); + void (*set_tsc_khz)(struct kvm_vcpu *vcpu, u32 user_tsc_khz); void (*write_tsc_offset)(struct kvm_vcpu *vcpu, u64 offset); + u64 (*compute_tsc_offset)(struct kvm_vcpu *vcpu, u64 target_tsc); + void (*get_exit_info)(struct kvm_vcpu *vcpu, u64 *info1, u64 *info2); + + int (*check_intercept)(struct kvm_vcpu *vcpu, + struct x86_instruction_info *info, + enum x86_intercept_stage stage); + const struct trace_print_flags *exit_reasons_str; }; @@ -627,6 +656,13 @@ u8 kvm_get_guest_memory_type(struct kvm_vcpu *vcpu, gfn_t gfn); extern bool tdp_enabled; +/* control of guest tsc rate supported? */ +extern bool kvm_has_tsc_control; +/* minimum supported tsc_khz for guests */ +extern u32 kvm_min_guest_tsc_khz; +/* maximum supported tsc_khz for guests */ +extern u32 kvm_max_guest_tsc_khz; + enum emulation_result { EMULATE_DONE, /* no further processing */ EMULATE_DO_MMIO, /* kvm_run filled with mmio request */ @@ -645,9 +681,6 @@ static inline int emulate_instruction(struct kvm_vcpu *vcpu, return x86_emulate_instruction(vcpu, 0, emulation_type, NULL, 0); } -void realmode_lgdt(struct kvm_vcpu *vcpu, u16 size, unsigned long address); -void realmode_lidt(struct kvm_vcpu *vcpu, u16 size, unsigned long address); - void kvm_enable_efer_bits(u64); int kvm_get_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 *data); int kvm_set_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 data); @@ -657,8 +690,6 @@ struct x86_emulate_ctxt; int kvm_fast_pio_out(struct kvm_vcpu *vcpu, int size, unsigned short port); void kvm_emulate_cpuid(struct kvm_vcpu *vcpu); int kvm_emulate_halt(struct kvm_vcpu *vcpu); -int emulate_invlpg(struct kvm_vcpu *vcpu, gva_t address); -int emulate_clts(struct kvm_vcpu *vcpu); int kvm_emulate_wbinvd(struct kvm_vcpu *vcpu); void kvm_get_segment(struct kvm_vcpu *vcpu, struct kvm_segment *var, int seg); @@ -721,8 +752,6 @@ gpa_t kvm_mmu_gva_to_gpa_system(struct kvm_vcpu *vcpu, gva_t gva, int kvm_emulate_hypercall(struct kvm_vcpu *vcpu); -int kvm_fix_hypercall(struct kvm_vcpu *vcpu); - int kvm_mmu_page_fault(struct kvm_vcpu *vcpu, gva_t gva, u32 error_code, void *insn, int insn_len); void kvm_mmu_invlpg(struct kvm_vcpu *vcpu, gva_t gva); diff --git a/arch/x86/include/asm/linkage.h b/arch/x86/include/asm/linkage.h index 12d55e773eb6..48142971b25d 100644 --- a/arch/x86/include/asm/linkage.h +++ b/arch/x86/include/asm/linkage.h @@ -8,11 +8,6 @@ #ifdef CONFIG_X86_32 #define asmlinkage CPP_ASMLINKAGE __attribute__((regparm(0))) -/* - * For 32-bit UML - mark functions implemented in assembly that use - * regparm input parameters: - */ -#define asmregparm __attribute__((regparm(3))) /* * Make sure the compiler doesn't do anything stupid with the diff --git a/arch/x86/include/asm/mce.h b/arch/x86/include/asm/mce.h index eb16e94ae04f..021979a6e23f 100644 --- a/arch/x86/include/asm/mce.h +++ b/arch/x86/include/asm/mce.h @@ -142,8 +142,6 @@ static inline void winchip_mcheck_init(struct cpuinfo_x86 *c) {} static inline void enable_p5_mce(void) {} #endif -extern void (*x86_mce_decode_callback)(struct mce *m); - void mce_setup(struct mce *m); void mce_log(struct mce *m); DECLARE_PER_CPU(struct sys_device, mce_dev); diff --git a/arch/x86/include/asm/mmzone_32.h b/arch/x86/include/asm/mmzone_32.h index 91df7c51806c..5e83a416eca8 100644 --- a/arch/x86/include/asm/mmzone_32.h +++ b/arch/x86/include/asm/mmzone_32.h @@ -13,31 +13,11 @@ extern struct pglist_data *node_data[]; #define NODE_DATA(nid) (node_data[nid]) #include <asm/numaq.h> -/* summit or generic arch */ -#include <asm/srat.h> - -extern int get_memcfg_numa_flat(void); -/* - * This allows any one NUMA architecture to be compiled - * for, and still fall back to the flat function if it - * fails. - */ -static inline void get_memcfg_numa(void) -{ - - if (get_memcfg_numaq()) - return; - if (get_memcfg_from_srat()) - return; - get_memcfg_numa_flat(); -} extern void resume_map_numa_kva(pgd_t *pgd); #else /* !CONFIG_NUMA */ -#define get_memcfg_numa get_memcfg_numa_flat - static inline void resume_map_numa_kva(pgd_t *pgd) {} #endif /* CONFIG_NUMA */ diff --git a/arch/x86/include/asm/mmzone_64.h b/arch/x86/include/asm/mmzone_64.h index 288b96f815a6..b3f88d7867c7 100644 --- a/arch/x86/include/asm/mmzone_64.h +++ b/arch/x86/include/asm/mmzone_64.h @@ -4,36 +4,13 @@ #ifndef _ASM_X86_MMZONE_64_H #define _ASM_X86_MMZONE_64_H - #ifdef CONFIG_NUMA #include <linux/mmdebug.h> - #include <asm/smp.h> -/* Simple perfect hash to map physical addresses to node numbers */ -struct memnode { - int shift; - unsigned int mapsize; - s16 *map; - s16 embedded_map[64 - 8]; -} ____cacheline_aligned; /* total size = 128 bytes */ -extern struct memnode memnode; -#define memnode_shift memnode.shift -#define memnodemap memnode.map -#define memnodemapsize memnode.mapsize - extern struct pglist_data *node_data[]; -static inline __attribute__((pure)) int phys_to_nid(unsigned long addr) -{ - unsigned nid; - VIRTUAL_BUG_ON(!memnodemap); - nid = memnodemap[addr >> memnode_shift]; - VIRTUAL_BUG_ON(nid >= MAX_NUMNODES || !node_data[nid]); - return nid; -} - #define NODE_DATA(nid) (node_data[nid]) #define node_start_pfn(nid) (NODE_DATA(nid)->node_start_pfn) diff --git a/arch/x86/include/asm/module.h b/arch/x86/include/asm/module.h index 67763c5d8b4e..9eae7752ae9b 100644 --- a/arch/x86/include/asm/module.h +++ b/arch/x86/include/asm/module.h @@ -35,7 +35,7 @@ #define MODULE_PROC_FAMILY "K7 " #elif defined CONFIG_MK8 #define MODULE_PROC_FAMILY "K8 " -#elif defined CONFIG_X86_ELAN +#elif defined CONFIG_MELAN #define MODULE_PROC_FAMILY "ELAN " #elif defined CONFIG_MCRUSOE #define MODULE_PROC_FAMILY "CRUSOE " diff --git a/arch/x86/include/asm/msr-index.h b/arch/x86/include/asm/msr-index.h index 3cce71413d0b..485b4f1f079b 100644 --- a/arch/x86/include/asm/msr-index.h +++ b/arch/x86/include/asm/msr-index.h @@ -118,6 +118,7 @@ complete list. */ #define MSR_AMD64_PATCH_LEVEL 0x0000008b +#define MSR_AMD64_TSC_RATIO 0xc0000104 #define MSR_AMD64_NB_CFG 0xc001001f #define MSR_AMD64_PATCH_LOADER 0xc0010020 #define MSR_AMD64_OSVW_ID_LENGTH 0xc0010140 diff --git a/arch/x86/include/asm/nops.h b/arch/x86/include/asm/nops.h index af788496020b..405b4032a60b 100644 --- a/arch/x86/include/asm/nops.h +++ b/arch/x86/include/asm/nops.h @@ -1,7 +1,13 @@ #ifndef _ASM_X86_NOPS_H #define _ASM_X86_NOPS_H -/* Define nops for use with alternative() */ +/* + * Define nops for use with alternative() and for tracing. + * + * *_NOP5_ATOMIC must be a single instruction. + */ + +#define NOP_DS_PREFIX 0x3e /* generic versions from gas 1: nop @@ -13,14 +19,15 @@ 6: leal 0x00000000(%esi),%esi 7: leal 0x00000000(,%esi,1),%esi */ -#define GENERIC_NOP1 ".byte 0x90\n" -#define GENERIC_NOP2 ".byte 0x89,0xf6\n" -#define GENERIC_NOP3 ".byte 0x8d,0x76,0x00\n" -#define GENERIC_NOP4 ".byte 0x8d,0x74,0x26,0x00\n" -#define GENERIC_NOP5 GENERIC_NOP1 GENERIC_NOP4 -#define GENERIC_NOP6 ".byte 0x8d,0xb6,0x00,0x00,0x00,0x00\n" -#define GENERIC_NOP7 ".byte 0x8d,0xb4,0x26,0x00,0x00,0x00,0x00\n" -#define GENERIC_NOP8 GENERIC_NOP1 GENERIC_NOP7 +#define GENERIC_NOP1 0x90 +#define GENERIC_NOP2 0x89,0xf6 +#define GENERIC_NOP3 0x8d,0x76,0x00 +#define GENERIC_NOP4 0x8d,0x74,0x26,0x00 +#define GENERIC_NOP5 GENERIC_NOP1,GENERIC_NOP4 +#define GENERIC_NOP6 0x8d,0xb6,0x00,0x00,0x00,0x00 +#define GENERIC_NOP7 0x8d,0xb4,0x26,0x00,0x00,0x00,0x00 +#define GENERIC_NOP8 GENERIC_NOP1,GENERIC_NOP7 +#define GENERIC_NOP5_ATOMIC NOP_DS_PREFIX,GENERIC_NOP4 /* Opteron 64bit nops 1: nop @@ -29,13 +36,14 @@ 4: osp osp osp nop */ #define K8_NOP1 GENERIC_NOP1 -#define K8_NOP2 ".byte 0x66,0x90\n" -#define K8_NOP3 ".byte 0x66,0x66,0x90\n" -#define K8_NOP4 ".byte 0x66,0x66,0x66,0x90\n" -#define K8_NOP5 K8_NOP3 K8_NOP2 -#define K8_NOP6 K8_NOP3 K8_NOP3 -#define K8_NOP7 K8_NOP4 K8_NOP3 -#define K8_NOP8 K8_NOP4 K8_NOP4 +#define K8_NOP2 0x66,K8_NOP1 +#define K8_NOP3 0x66,K8_NOP2 +#define K8_NOP4 0x66,K8_NOP3 +#define K8_NOP5 K8_NOP3,K8_NOP2 +#define K8_NOP6 K8_NOP3,K8_NOP3 +#define K8_NOP7 K8_NOP4,K8_NOP3 +#define K8_NOP8 K8_NOP4,K8_NOP4 +#define K8_NOP5_ATOMIC 0x66,K8_NOP4 /* K7 nops uses eax dependencies (arbitrary choice) @@ -47,13 +55,14 @@ 7: leal 0x00000000(,%eax,1),%eax */ #define K7_NOP1 GENERIC_NOP1 -#define K7_NOP2 ".byte 0x8b,0xc0\n" -#define K7_NOP3 ".byte 0x8d,0x04,0x20\n" -#define K7_NOP4 ".byte 0x8d,0x44,0x20,0x00\n" -#define K7_NOP5 K7_NOP4 ASM_NOP1 -#define K7_NOP6 ".byte 0x8d,0x80,0,0,0,0\n" -#define K7_NOP7 ".byte 0x8D,0x04,0x05,0,0,0,0\n" -#define K7_NOP8 K7_NOP7 ASM_NOP1 +#define K7_NOP2 0x8b,0xc0 +#define K7_NOP3 0x8d,0x04,0x20 +#define K7_NOP4 0x8d,0x44,0x20,0x00 +#define K7_NOP5 K7_NOP4,K7_NOP1 +#define K7_NOP6 0x8d,0x80,0,0,0,0 +#define K7_NOP7 0x8D,0x04,0x05,0,0,0,0 +#define K7_NOP8 K7_NOP7,K7_NOP1 +#define K7_NOP5_ATOMIC NOP_DS_PREFIX,K7_NOP4 /* P6 nops uses eax dependencies (Intel-recommended choice) @@ -69,52 +78,65 @@ There is kernel code that depends on this. */ #define P6_NOP1 GENERIC_NOP1 -#define P6_NOP2 ".byte 0x66,0x90\n" -#define P6_NOP3 ".byte 0x0f,0x1f,0x00\n" -#define P6_NOP4 ".byte 0x0f,0x1f,0x40,0\n" -#define P6_NOP5 ".byte 0x0f,0x1f,0x44,0x00,0\n" -#define P6_NOP6 ".byte 0x66,0x0f,0x1f,0x44,0x00,0\n" -#define P6_NOP7 ".byte 0x0f,0x1f,0x80,0,0,0,0\n" -#define P6_NOP8 ".byte 0x0f,0x1f,0x84,0x00,0,0,0,0\n" +#define P6_NOP2 0x66,0x90 +#define P6_NOP3 0x0f,0x1f,0x00 +#define P6_NOP4 0x0f,0x1f,0x40,0 +#define P6_NOP5 0x0f,0x1f,0x44,0x00,0 +#define P6_NOP6 0x66,0x0f,0x1f,0x44,0x00,0 +#define P6_NOP7 0x0f,0x1f,0x80,0,0,0,0 +#define P6_NOP8 0x0f,0x1f,0x84,0x00,0,0,0,0 +#define P6_NOP5_ATOMIC P6_NOP5 + +#define _ASM_MK_NOP(x) ".byte " __stringify(x) "\n" #if defined(CONFIG_MK7) -#define ASM_NOP1 K7_NOP1 -#define ASM_NOP2 K7_NOP2 -#define ASM_NOP3 K7_NOP3 -#define ASM_NOP4 K7_NOP4 -#define ASM_NOP5 K7_NOP5 -#define ASM_NOP6 K7_NOP6 -#define ASM_NOP7 K7_NOP7 -#define ASM_NOP8 K7_NOP8 +#define ASM_NOP1 _ASM_MK_NOP(K7_NOP1) +#define ASM_NOP2 _ASM_MK_NOP(K7_NOP2) +#define ASM_NOP3 _ASM_MK_NOP(K7_NOP3) +#define ASM_NOP4 _ASM_MK_NOP(K7_NOP4) +#define ASM_NOP5 _ASM_MK_NOP(K7_NOP5) +#define ASM_NOP6 _ASM_MK_NOP(K7_NOP6) +#define ASM_NOP7 _ASM_MK_NOP(K7_NOP7) +#define ASM_NOP8 _ASM_MK_NOP(K7_NOP8) +#define ASM_NOP5_ATOMIC _ASM_MK_NOP(K7_NOP5_ATOMIC) #elif defined(CONFIG_X86_P6_NOP) -#define ASM_NOP1 P6_NOP1 -#define ASM_NOP2 P6_NOP2 -#define ASM_NOP3 P6_NOP3 -#define ASM_NOP4 P6_NOP4 -#define ASM_NOP5 P6_NOP5 -#define ASM_NOP6 P6_NOP6 -#define ASM_NOP7 P6_NOP7 -#define ASM_NOP8 P6_NOP8 +#define ASM_NOP1 _ASM_MK_NOP(P6_NOP1) +#define ASM_NOP2 _ASM_MK_NOP(P6_NOP2) +#define ASM_NOP3 _ASM_MK_NOP(P6_NOP3) +#define ASM_NOP4 _ASM_MK_NOP(P6_NOP4) +#define ASM_NOP5 _ASM_MK_NOP(P6_NOP5) +#define ASM_NOP6 _ASM_MK_NOP(P6_NOP6) +#define ASM_NOP7 _ASM_MK_NOP(P6_NOP7) +#define ASM_NOP8 _ASM_MK_NOP(P6_NOP8) +#define ASM_NOP5_ATOMIC _ASM_MK_NOP(P6_NOP5_ATOMIC) #elif defined(CONFIG_X86_64) -#define ASM_NOP1 K8_NOP1 -#define ASM_NOP2 K8_NOP2 -#define ASM_NOP3 K8_NOP3 -#define ASM_NOP4 K8_NOP4 -#define ASM_NOP5 K8_NOP5 -#define ASM_NOP6 K8_NOP6 -#define ASM_NOP7 K8_NOP7 -#define ASM_NOP8 K8_NOP8 +#define ASM_NOP1 _ASM_MK_NOP(K8_NOP1) +#define ASM_NOP2 _ASM_MK_NOP(K8_NOP2) +#define ASM_NOP3 _ASM_MK_NOP(K8_NOP3) +#define ASM_NOP4 _ASM_MK_NOP(K8_NOP4) +#define ASM_NOP5 _ASM_MK_NOP(K8_NOP5) +#define ASM_NOP6 _ASM_MK_NOP(K8_NOP6) +#define ASM_NOP7 _ASM_MK_NOP(K8_NOP7) +#define ASM_NOP8 _ASM_MK_NOP(K8_NOP8) +#define ASM_NOP5_ATOMIC _ASM_MK_NOP(K8_NOP5_ATOMIC) #else -#define ASM_NOP1 GENERIC_NOP1 -#define ASM_NOP2 GENERIC_NOP2 -#define ASM_NOP3 GENERIC_NOP3 -#define ASM_NOP4 GENERIC_NOP4 -#define ASM_NOP5 GENERIC_NOP5 -#define ASM_NOP6 GENERIC_NOP6 -#define ASM_NOP7 GENERIC_NOP7 -#define ASM_NOP8 GENERIC_NOP8 +#define ASM_NOP1 _ASM_MK_NOP(GENERIC_NOP1) +#define ASM_NOP2 _ASM_MK_NOP(GENERIC_NOP2) +#define ASM_NOP3 _ASM_MK_NOP(GENERIC_NOP3) +#define ASM_NOP4 _ASM_MK_NOP(GENERIC_NOP4) +#define ASM_NOP5 _ASM_MK_NOP(GENERIC_NOP5) +#define ASM_NOP6 _ASM_MK_NOP(GENERIC_NOP6) +#define ASM_NOP7 _ASM_MK_NOP(GENERIC_NOP7) +#define ASM_NOP8 _ASM_MK_NOP(GENERIC_NOP8) +#define ASM_NOP5_ATOMIC _ASM_MK_NOP(GENERIC_NOP5_ATOMIC) #endif #define ASM_NOP_MAX 8 +#define NOP_ATOMIC5 (ASM_NOP_MAX+1) /* Entry for the 5-byte atomic NOP */ + +#ifndef __ASSEMBLY__ +extern const unsigned char * const *ideal_nops; +extern void arch_init_ideal_nops(void); +#endif #endif /* _ASM_X86_NOPS_H */ diff --git a/arch/x86/include/asm/numa.h b/arch/x86/include/asm/numa.h index a50fc9f493b3..bfacd2ccf651 100644 --- a/arch/x86/include/asm/numa.h +++ b/arch/x86/include/asm/numa.h @@ -1,12 +1,24 @@ #ifndef _ASM_X86_NUMA_H #define _ASM_X86_NUMA_H +#include <linux/nodemask.h> + #include <asm/topology.h> #include <asm/apicdef.h> #ifdef CONFIG_NUMA #define NR_NODE_MEMBLKS (MAX_NUMNODES*2) +#define ZONE_ALIGN (1UL << (MAX_ORDER+PAGE_SHIFT)) + +/* + * Too small node sizes may confuse the VM badly. Usually they + * result from BIOS bugs. So dont recognize nodes as standalone + * NUMA entities that have less than this amount of RAM listed: + */ +#define NODE_MIN_SIZE (4*1024*1024) + +extern int numa_off; /* * __apicid_to_node[] stores the raw mapping between physical apicid and @@ -17,15 +29,27 @@ * numa_cpu_node(). */ extern s16 __apicid_to_node[MAX_LOCAL_APIC]; +extern nodemask_t numa_nodes_parsed __initdata; + +extern int __init numa_add_memblk(int nodeid, u64 start, u64 end); +extern void __init numa_set_distance(int from, int to, int distance); static inline void set_apicid_to_node(int apicid, s16 node) { __apicid_to_node[apicid] = node; } + +extern int __cpuinit numa_cpu_node(int cpu); + #else /* CONFIG_NUMA */ static inline void set_apicid_to_node(int apicid, s16 node) { } + +static inline int numa_cpu_node(int cpu) +{ + return NUMA_NO_NODE; +} #endif /* CONFIG_NUMA */ #ifdef CONFIG_X86_32 @@ -37,14 +61,12 @@ static inline void set_apicid_to_node(int apicid, s16 node) #ifdef CONFIG_NUMA extern void __cpuinit numa_set_node(int cpu, int node); extern void __cpuinit numa_clear_node(int cpu); -extern void __init numa_init_array(void); extern void __init init_cpu_to_node(void); extern void __cpuinit numa_add_cpu(int cpu); extern void __cpuinit numa_remove_cpu(int cpu); #else /* CONFIG_NUMA */ static inline void numa_set_node(int cpu, int node) { } static inline void numa_clear_node(int cpu) { } -static inline void numa_init_array(void) { } static inline void init_cpu_to_node(void) { } static inline void numa_add_cpu(int cpu) { } static inline void numa_remove_cpu(int cpu) { } @@ -54,4 +76,10 @@ static inline void numa_remove_cpu(int cpu) { } void debug_cpumask_set_cpu(int cpu, int node, bool enable); #endif +#ifdef CONFIG_NUMA_EMU +#define FAKE_NODE_MIN_SIZE ((u64)32 << 20) +#define FAKE_NODE_MIN_HASH_MASK (~(FAKE_NODE_MIN_SIZE - 1UL)) +void numa_emu_cmdline(char *); +#endif /* CONFIG_NUMA_EMU */ + #endif /* _ASM_X86_NUMA_H */ diff --git a/arch/x86/include/asm/numa_32.h b/arch/x86/include/asm/numa_32.h index c6beed1ef103..e7d6b8254742 100644 --- a/arch/x86/include/asm/numa_32.h +++ b/arch/x86/include/asm/numa_32.h @@ -1,16 +1,6 @@ #ifndef _ASM_X86_NUMA_32_H #define _ASM_X86_NUMA_32_H -extern int numa_off; - -extern int pxm_to_nid(int pxm); - -#ifdef CONFIG_NUMA -extern int __cpuinit numa_cpu_node(int cpu); -#else /* CONFIG_NUMA */ -static inline int numa_cpu_node(int cpu) { return NUMA_NO_NODE; } -#endif /* CONFIG_NUMA */ - #ifdef CONFIG_HIGHMEM extern void set_highmem_pages_init(void); #else diff --git a/arch/x86/include/asm/numa_64.h b/arch/x86/include/asm/numa_64.h index 344eb1790b46..0c05f7ae46e8 100644 --- a/arch/x86/include/asm/numa_64.h +++ b/arch/x86/include/asm/numa_64.h @@ -1,42 +1,6 @@ #ifndef _ASM_X86_NUMA_64_H #define _ASM_X86_NUMA_64_H -#include <linux/nodemask.h> - -struct bootnode { - u64 start; - u64 end; -}; - -#define ZONE_ALIGN (1UL << (MAX_ORDER+PAGE_SHIFT)) - -extern int numa_off; - extern unsigned long numa_free_all_bootmem(void); -extern void setup_node_bootmem(int nodeid, unsigned long start, - unsigned long end); - -#ifdef CONFIG_NUMA -/* - * Too small node sizes may confuse the VM badly. Usually they - * result from BIOS bugs. So dont recognize nodes as standalone - * NUMA entities that have less than this amount of RAM listed: - */ -#define NODE_MIN_SIZE (4*1024*1024) - -extern nodemask_t numa_nodes_parsed __initdata; - -extern int __cpuinit numa_cpu_node(int cpu); -extern int __init numa_add_memblk(int nodeid, u64 start, u64 end); -extern void __init numa_set_distance(int from, int to, int distance); - -#ifdef CONFIG_NUMA_EMU -#define FAKE_NODE_MIN_SIZE ((u64)32 << 20) -#define FAKE_NODE_MIN_HASH_MASK (~(FAKE_NODE_MIN_SIZE - 1UL)) -void numa_emu_cmdline(char *); -#endif /* CONFIG_NUMA_EMU */ -#else -static inline int numa_cpu_node(int cpu) { return NUMA_NO_NODE; } -#endif #endif /* _ASM_X86_NUMA_64_H */ diff --git a/arch/x86/include/asm/numaq.h b/arch/x86/include/asm/numaq.h index 37c516545ec8..c3b3c322fd87 100644 --- a/arch/x86/include/asm/numaq.h +++ b/arch/x86/include/asm/numaq.h @@ -29,7 +29,7 @@ #ifdef CONFIG_X86_NUMAQ extern int found_numaq; -extern int get_memcfg_numaq(void); +extern int numaq_numa_init(void); extern int pci_numaq_init(void); extern void *xquad_portio; @@ -166,11 +166,6 @@ struct sys_cfg_data { void numaq_tsc_disable(void); -#else -static inline int get_memcfg_numaq(void) -{ - return 0; -} #endif /* CONFIG_X86_NUMAQ */ #endif /* _ASM_X86_NUMAQ_H */ diff --git a/arch/x86/include/asm/olpc_ofw.h b/arch/x86/include/asm/olpc_ofw.h index c5d3a5abbb9f..24487712e0b1 100644 --- a/arch/x86/include/asm/olpc_ofw.h +++ b/arch/x86/include/asm/olpc_ofw.h @@ -26,15 +26,12 @@ extern void setup_olpc_ofw_pgd(void); /* check if OFW was detected during boot */ extern bool olpc_ofw_present(void); +extern void olpc_dt_build_devicetree(void); + #else /* !CONFIG_OLPC */ static inline void olpc_ofw_detect(void) { } static inline void setup_olpc_ofw_pgd(void) { } -#endif /* !CONFIG_OLPC */ - -#ifdef CONFIG_OF_PROMTREE -extern void olpc_dt_build_devicetree(void); -#else static inline void olpc_dt_build_devicetree(void) { } -#endif +#endif /* !CONFIG_OLPC */ #endif /* _ASM_X86_OLPC_OFW_H */ diff --git a/arch/x86/include/asm/pci.h b/arch/x86/include/asm/pci.h index 676129229630..d498943b906c 100644 --- a/arch/x86/include/asm/pci.h +++ b/arch/x86/include/asm/pci.h @@ -135,8 +135,6 @@ void default_teardown_msi_irqs(struct pci_dev *dev); #include "pci_64.h" #endif -void dma32_reserve_bootmem(void); - /* implement the pci_ DMA API in terms of the generic device dma_ one */ #include <asm-generic/pci-dma-compat.h> diff --git a/arch/x86/include/asm/percpu.h b/arch/x86/include/asm/percpu.h index d475b4398d8b..a0a9779084d1 100644 --- a/arch/x86/include/asm/percpu.h +++ b/arch/x86/include/asm/percpu.h @@ -509,6 +509,11 @@ do { \ * it in software. The address used in the cmpxchg16 instruction must be * aligned to a 16 byte boundary. */ +#ifdef CONFIG_SMP +#define CMPXCHG16B_EMU_CALL "call this_cpu_cmpxchg16b_emu\n\t" ASM_NOP3 +#else +#define CMPXCHG16B_EMU_CALL "call this_cpu_cmpxchg16b_emu\n\t" ASM_NOP2 +#endif #define percpu_cmpxchg16b_double(pcp1, o1, o2, n1, n2) \ ({ \ char __ret; \ @@ -517,7 +522,7 @@ do { \ typeof(o2) __o2 = o2; \ typeof(o2) __n2 = n2; \ typeof(o2) __dummy; \ - alternative_io("call this_cpu_cmpxchg16b_emu\n\t" P6_NOP4, \ + alternative_io(CMPXCHG16B_EMU_CALL, \ "cmpxchg16b " __percpu_prefix "(%%rsi)\n\tsetz %0\n\t", \ X86_FEATURE_CX16, \ ASM_OUTPUT2("=a"(__ret), "=d"(__dummy)), \ @@ -542,6 +547,33 @@ do { \ old__; \ }) +static __always_inline int x86_this_cpu_constant_test_bit(unsigned int nr, + const unsigned long __percpu *addr) +{ + unsigned long __percpu *a = (unsigned long *)addr + nr / BITS_PER_LONG; + + return ((1UL << (nr % BITS_PER_LONG)) & percpu_read(*a)) != 0; +} + +static inline int x86_this_cpu_variable_test_bit(int nr, + const unsigned long __percpu *addr) +{ + int oldbit; + + asm volatile("bt "__percpu_arg(2)",%1\n\t" + "sbb %0,%0" + : "=r" (oldbit) + : "m" (*(unsigned long *)addr), "Ir" (nr)); + + return oldbit; +} + +#define x86_this_cpu_test_bit(nr, addr) \ + (__builtin_constant_p((nr)) \ + ? x86_this_cpu_constant_test_bit((nr), (addr)) \ + : x86_this_cpu_variable_test_bit((nr), (addr))) + + #include <asm-generic/percpu.h> /* We can use this directly for local CPU (faster). */ diff --git a/arch/x86/include/asm/pgtable_types.h b/arch/x86/include/asm/pgtable_types.h index 7db7723d1f32..d56187c6b838 100644 --- a/arch/x86/include/asm/pgtable_types.h +++ b/arch/x86/include/asm/pgtable_types.h @@ -299,6 +299,7 @@ int phys_mem_access_prot_allowed(struct file *file, unsigned long pfn, /* Install a pte for a particular vaddr in kernel space. */ void set_pte_vaddr(unsigned long vaddr, pte_t pte); +extern void native_pagetable_reserve(u64 start, u64 end); #ifdef CONFIG_X86_32 extern void native_pagetable_setup_start(pgd_t *base); extern void native_pagetable_setup_done(pgd_t *base); diff --git a/arch/x86/include/asm/probe_roms.h b/arch/x86/include/asm/probe_roms.h new file mode 100644 index 000000000000..4950a0b1d09c --- /dev/null +++ b/arch/x86/include/asm/probe_roms.h @@ -0,0 +1,8 @@ +#ifndef _PROBE_ROMS_H_ +#define _PROBE_ROMS_H_ +struct pci_dev; + +extern void __iomem *pci_map_biosrom(struct pci_dev *pdev); +extern void pci_unmap_biosrom(void __iomem *rom); +extern size_t pci_biosrom_size(struct pci_dev *pdev); +#endif diff --git a/arch/x86/include/asm/processor-flags.h b/arch/x86/include/asm/processor-flags.h index a898a2b6e10c..59ab4dffa377 100644 --- a/arch/x86/include/asm/processor-flags.h +++ b/arch/x86/include/asm/processor-flags.h @@ -60,6 +60,7 @@ #define X86_CR4_OSXMMEXCPT 0x00000400 /* enable unmasked SSE exceptions */ #define X86_CR4_VMXE 0x00002000 /* enable VMX virtualization */ #define X86_CR4_OSXSAVE 0x00040000 /* enable xsave and xrestore */ +#define X86_CR4_SMEP 0x00100000 /* enable SMEP support */ /* * x86-64 Task Priority Register, CR8 diff --git a/arch/x86/include/asm/setup.h b/arch/x86/include/asm/setup.h index db8aa19a08a2..9756551ec760 100644 --- a/arch/x86/include/asm/setup.h +++ b/arch/x86/include/asm/setup.h @@ -88,7 +88,7 @@ void *extend_brk(size_t size, size_t align); * executable.) */ #define RESERVE_BRK(name,sz) \ - static void __section(.discard.text) __used \ + static void __section(.discard.text) __used notrace \ __brk_reservation_fn_##name##__(void) { \ asm volatile ( \ ".pushsection .brk_reservation,\"aw\",@nobits;" \ @@ -104,10 +104,10 @@ void *extend_brk(size_t size, size_t align); type *name; \ RESERVE_BRK(name, sizeof(type) * entries) +extern void probe_roms(void); #ifdef __i386__ void __init i386_start_kernel(void); -extern void probe_roms(void); #else void __init x86_64_start_kernel(char *real_mode); diff --git a/arch/x86/include/asm/srat.h b/arch/x86/include/asm/srat.h deleted file mode 100644 index b508d639d1a7..000000000000 --- a/arch/x86/include/asm/srat.h +++ /dev/null @@ -1,39 +0,0 @@ -/* - * Some of the code in this file has been gleaned from the 64 bit - * discontigmem support code base. - * - * Copyright (C) 2002, IBM Corp. - * - * All rights reserved. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or - * NON INFRINGEMENT. See the GNU General Public License for more - * details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. - * - * Send feedback to Pat Gaughen <gone@us.ibm.com> - */ - -#ifndef _ASM_X86_SRAT_H -#define _ASM_X86_SRAT_H - -#ifdef CONFIG_ACPI_NUMA -extern int get_memcfg_from_srat(void); -#else -static inline int get_memcfg_from_srat(void) -{ - return 0; -} -#endif - -#endif /* _ASM_X86_SRAT_H */ diff --git a/arch/x86/include/asm/stacktrace.h b/arch/x86/include/asm/stacktrace.h index d7e89c83645d..70bbe39043a9 100644 --- a/arch/x86/include/asm/stacktrace.h +++ b/arch/x86/include/asm/stacktrace.h @@ -37,9 +37,6 @@ print_context_stack_bp(struct thread_info *tinfo, /* Generic stack tracer with callbacks */ struct stacktrace_ops { - void (*warning)(void *data, char *msg); - /* msg must contain %s for the symbol */ - void (*warning_symbol)(void *data, char *msg, unsigned long symbol); void (*address)(void *data, unsigned long address, int reliable); /* On negative return stop dumping */ int (*stack)(void *data, char *name); diff --git a/arch/x86/include/asm/system.h b/arch/x86/include/asm/system.h index 12569e691ce3..c2ff2a1d845e 100644 --- a/arch/x86/include/asm/system.h +++ b/arch/x86/include/asm/system.h @@ -303,24 +303,81 @@ static inline void native_wbinvd(void) #ifdef CONFIG_PARAVIRT #include <asm/paravirt.h> #else -#define read_cr0() (native_read_cr0()) -#define write_cr0(x) (native_write_cr0(x)) -#define read_cr2() (native_read_cr2()) -#define write_cr2(x) (native_write_cr2(x)) -#define read_cr3() (native_read_cr3()) -#define write_cr3(x) (native_write_cr3(x)) -#define read_cr4() (native_read_cr4()) -#define read_cr4_safe() (native_read_cr4_safe()) -#define write_cr4(x) (native_write_cr4(x)) -#define wbinvd() (native_wbinvd()) + +static inline unsigned long read_cr0(void) +{ + return native_read_cr0(); +} + +static inline void write_cr0(unsigned long x) +{ + native_write_cr0(x); +} + +static inline unsigned long read_cr2(void) +{ + return native_read_cr2(); +} + +static inline void write_cr2(unsigned long x) +{ + native_write_cr2(x); +} + +static inline unsigned long read_cr3(void) +{ + return native_read_cr3(); +} + +static inline void write_cr3(unsigned long x) +{ + native_write_cr3(x); +} + +static inline unsigned long read_cr4(void) +{ + return native_read_cr4(); +} + +static inline unsigned long read_cr4_safe(void) +{ + return native_read_cr4_safe(); +} + +static inline void write_cr4(unsigned long x) +{ + native_write_cr4(x); +} + +static inline void wbinvd(void) +{ + native_wbinvd(); +} + #ifdef CONFIG_X86_64 -#define read_cr8() (native_read_cr8()) -#define write_cr8(x) (native_write_cr8(x)) -#define load_gs_index native_load_gs_index + +static inline unsigned long read_cr8(void) +{ + return native_read_cr8(); +} + +static inline void write_cr8(unsigned long x) +{ + native_write_cr8(x); +} + +static inline void load_gs_index(unsigned selector) +{ + native_load_gs_index(selector); +} + #endif /* Clear the 'TS' bit */ -#define clts() (native_clts()) +static inline void clts(void) +{ + native_clts(); +} #endif/* CONFIG_PARAVIRT */ diff --git a/arch/x86/include/asm/topology.h b/arch/x86/include/asm/topology.h index 910a7084f7f2..c00692476e9f 100644 --- a/arch/x86/include/asm/topology.h +++ b/arch/x86/include/asm/topology.h @@ -93,19 +93,11 @@ extern void setup_node_to_cpumask_map(void); #define pcibus_to_node(bus) __pcibus_to_node(bus) #ifdef CONFIG_X86_32 -extern unsigned long node_start_pfn[]; -extern unsigned long node_end_pfn[]; -extern unsigned long node_remap_size[]; -#define node_has_online_mem(nid) (node_start_pfn[nid] != node_end_pfn[nid]) - # define SD_CACHE_NICE_TRIES 1 # define SD_IDLE_IDX 1 - #else - # define SD_CACHE_NICE_TRIES 2 # define SD_IDLE_IDX 2 - #endif /* sched_domains SD_NODE_INIT for NUMA machines */ diff --git a/arch/x86/include/asm/uaccess.h b/arch/x86/include/asm/uaccess.h index abd3e0ea762a..99ddd148a760 100644 --- a/arch/x86/include/asm/uaccess.h +++ b/arch/x86/include/asm/uaccess.h @@ -6,7 +6,6 @@ #include <linux/errno.h> #include <linux/compiler.h> #include <linux/thread_info.h> -#include <linux/prefetch.h> #include <linux/string.h> #include <asm/asm.h> #include <asm/page.h> @@ -42,7 +41,7 @@ * Returns 0 if the range is valid, nonzero otherwise. * * This is equivalent to the following test: - * (u33)addr + (u33)size >= (u33)current->addr_limit.seg (u65 for x86_64) + * (u33)addr + (u33)size > (u33)current->addr_limit.seg (u65 for x86_64) * * This needs 33-bit (65-bit for x86_64) arithmetic. We have a carry... */ diff --git a/arch/x86/include/asm/uaccess_32.h b/arch/x86/include/asm/uaccess_32.h index 088d09fb1615..566e803cc602 100644 --- a/arch/x86/include/asm/uaccess_32.h +++ b/arch/x86/include/asm/uaccess_32.h @@ -6,7 +6,6 @@ */ #include <linux/errno.h> #include <linux/thread_info.h> -#include <linux/prefetch.h> #include <linux/string.h> #include <asm/asm.h> #include <asm/page.h> diff --git a/arch/x86/include/asm/uaccess_64.h b/arch/x86/include/asm/uaccess_64.h index 316708d5af92..1c66d30971ad 100644 --- a/arch/x86/include/asm/uaccess_64.h +++ b/arch/x86/include/asm/uaccess_64.h @@ -6,7 +6,6 @@ */ #include <linux/compiler.h> #include <linux/errno.h> -#include <linux/prefetch.h> #include <linux/lockdep.h> #include <asm/alternative.h> #include <asm/cpufeature.h> diff --git a/arch/x86/include/asm/unistd_32.h b/arch/x86/include/asm/unistd_32.h index a755ef5e5977..fb6a625c99bf 100644 --- a/arch/x86/include/asm/unistd_32.h +++ b/arch/x86/include/asm/unistd_32.h @@ -350,10 +350,11 @@ #define __NR_open_by_handle_at 342 #define __NR_clock_adjtime 343 #define __NR_syncfs 344 +#define __NR_sendmmsg 345 #ifdef __KERNEL__ -#define NR_syscalls 345 +#define NR_syscalls 346 #define __ARCH_WANT_IPC_PARSE_VERSION #define __ARCH_WANT_OLD_READDIR diff --git a/arch/x86/include/asm/unistd_64.h b/arch/x86/include/asm/unistd_64.h index 160fa76bd578..79f90eb15aad 100644 --- a/arch/x86/include/asm/unistd_64.h +++ b/arch/x86/include/asm/unistd_64.h @@ -677,6 +677,8 @@ __SYSCALL(__NR_open_by_handle_at, sys_open_by_handle_at) __SYSCALL(__NR_clock_adjtime, sys_clock_adjtime) #define __NR_syncfs 306 __SYSCALL(__NR_syncfs, sys_syncfs) +#define __NR_sendmmsg 307 +__SYSCALL(__NR_sendmmsg, sys_sendmmsg) #ifndef __NO_STUBS #define __ARCH_WANT_OLD_READDIR diff --git a/arch/x86/include/asm/uv/uv_bau.h b/arch/x86/include/asm/uv/uv_bau.h index 3e094af443c3..130f1eeee5fe 100644 --- a/arch/x86/include/asm/uv/uv_bau.h +++ b/arch/x86/include/asm/uv/uv_bau.h @@ -94,6 +94,8 @@ /* after this # consecutive successes, bump up the throttle if it was lowered */ #define COMPLETE_THRESHOLD 5 +#define UV_LB_SUBNODEID 0x10 + /* * number of entries in the destination side payload queue */ @@ -124,7 +126,7 @@ * The distribution specification (32 bytes) is interpreted as a 256-bit * distribution vector. Adjacent bits correspond to consecutive even numbered * nodeIDs. The result of adding the index of a given bit to the 15-bit - * 'base_dest_nodeid' field of the header corresponds to the + * 'base_dest_nasid' field of the header corresponds to the * destination nodeID associated with that specified bit. */ struct bau_target_uvhubmask { @@ -176,7 +178,7 @@ struct bau_msg_payload { struct bau_msg_header { unsigned int dest_subnodeid:6; /* must be 0x10, for the LB */ /* bits 5:0 */ - unsigned int base_dest_nodeid:15; /* nasid of the */ + unsigned int base_dest_nasid:15; /* nasid of the */ /* bits 20:6 */ /* first bit in uvhub map */ unsigned int command:8; /* message type */ /* bits 28:21 */ @@ -378,6 +380,10 @@ struct ptc_stats { unsigned long d_rcanceled; /* number of messages canceled by resets */ }; +struct hub_and_pnode { + short uvhub; + short pnode; +}; /* * one per-cpu; to locate the software tables */ @@ -399,10 +405,12 @@ struct bau_control { int baudisabled; int set_bau_off; short cpu; + short osnode; short uvhub_cpu; short uvhub; short cpus_in_socket; short cpus_in_uvhub; + short partition_base_pnode; unsigned short message_number; unsigned short uvhub_quiesce; short socket_acknowledge_count[DEST_Q_SIZE]; @@ -422,15 +430,16 @@ struct bau_control { int congested_period; cycles_t period_time; long period_requests; + struct hub_and_pnode *target_hub_and_pnode; }; static inline int bau_uvhub_isset(int uvhub, struct bau_target_uvhubmask *dstp) { return constant_test_bit(uvhub, &dstp->bits[0]); } -static inline void bau_uvhub_set(int uvhub, struct bau_target_uvhubmask *dstp) +static inline void bau_uvhub_set(int pnode, struct bau_target_uvhubmask *dstp) { - __set_bit(uvhub, &dstp->bits[0]); + __set_bit(pnode, &dstp->bits[0]); } static inline void bau_uvhubs_clear(struct bau_target_uvhubmask *dstp, int nbits) diff --git a/arch/x86/include/asm/uv/uv_hub.h b/arch/x86/include/asm/uv/uv_hub.h index a501741c2335..4298002d0c83 100644 --- a/arch/x86/include/asm/uv/uv_hub.h +++ b/arch/x86/include/asm/uv/uv_hub.h @@ -398,6 +398,8 @@ struct uv_blade_info { unsigned short nr_online_cpus; unsigned short pnode; short memory_nid; + spinlock_t nmi_lock; + unsigned long nmi_count; }; extern struct uv_blade_info *uv_blade_info; extern short *uv_node_to_blade; diff --git a/arch/x86/include/asm/uv/uv_mmrs.h b/arch/x86/include/asm/uv/uv_mmrs.h index 20cafeac7455..f5bb64a823d7 100644 --- a/arch/x86/include/asm/uv/uv_mmrs.h +++ b/arch/x86/include/asm/uv/uv_mmrs.h @@ -5,7 +5,7 @@ * * SGI UV MMR definitions * - * Copyright (C) 2007-2010 Silicon Graphics, Inc. All rights reserved. + * Copyright (C) 2007-2011 Silicon Graphics, Inc. All rights reserved. */ #ifndef _ASM_X86_UV_UV_MMRS_H @@ -1099,5 +1099,19 @@ union uvh_rtc1_int_config_u { } s; }; +/* ========================================================================= */ +/* UVH_SCRATCH5 */ +/* ========================================================================= */ +#define UVH_SCRATCH5 0x2d0200UL +#define UVH_SCRATCH5_32 0x00778 + +#define UVH_SCRATCH5_SCRATCH5_SHFT 0 +#define UVH_SCRATCH5_SCRATCH5_MASK 0xffffffffffffffffUL +union uvh_scratch5_u { + unsigned long v; + struct uvh_scratch5_s { + unsigned long scratch5 : 64; /* RW, W1CS */ + } s; +}; #endif /* __ASM_UV_MMRS_X86_H__ */ diff --git a/arch/x86/include/asm/x2apic.h b/arch/x86/include/asm/x2apic.h new file mode 100644 index 000000000000..6bf5b8e478c0 --- /dev/null +++ b/arch/x86/include/asm/x2apic.h @@ -0,0 +1,62 @@ +/* + * Common bits for X2APIC cluster/physical modes. + */ + +#ifndef _ASM_X86_X2APIC_H +#define _ASM_X86_X2APIC_H + +#include <asm/apic.h> +#include <asm/ipi.h> +#include <linux/cpumask.h> + +/* + * 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 cpu_online_mask; +} + +static int x2apic_apic_id_registered(void) +{ + return 1; +} + +/* + * For now each logical cpu is in its own vector allocation domain. + */ +static void x2apic_vector_allocation_domain(int cpu, struct cpumask *retmask) +{ + cpumask_clear(retmask); + cpumask_set_cpu(cpu, retmask); +} + +static void +__x2apic_send_IPI_dest(unsigned int apicid, int vector, unsigned int dest) +{ + unsigned long cfg = __prepare_ICR(0, vector, dest); + native_x2apic_icr_write(cfg, apicid); +} + +static unsigned int x2apic_get_apic_id(unsigned long id) +{ + return id; +} + +static unsigned long x2apic_set_apic_id(unsigned int id) +{ + return id; +} + +static int x2apic_phys_pkg_id(int initial_apicid, int index_msb) +{ + return initial_apicid >> index_msb; +} + +static void x2apic_send_IPI_self(int vector) +{ + apic_write(APIC_SELF_IPI, vector); +} + +#endif /* _ASM_X86_X2APIC_H */ diff --git a/arch/x86/include/asm/x86_init.h b/arch/x86/include/asm/x86_init.h index 643ebf2e2ad8..d3d859035af9 100644 --- a/arch/x86/include/asm/x86_init.h +++ b/arch/x86/include/asm/x86_init.h @@ -68,6 +68,17 @@ struct x86_init_oem { }; /** + * struct x86_init_mapping - platform specific initial kernel pagetable setup + * @pagetable_reserve: reserve a range of addresses for kernel pagetable usage + * + * For more details on the purpose of this hook, look in + * init_memory_mapping and the commit that added it. + */ +struct x86_init_mapping { + void (*pagetable_reserve)(u64 start, u64 end); +}; + +/** * struct x86_init_paging - platform specific paging functions * @pagetable_setup_start: platform specific pre paging_init() call * @pagetable_setup_done: platform specific post paging_init() call @@ -123,6 +134,7 @@ struct x86_init_ops { struct x86_init_mpparse mpparse; struct x86_init_irqs irqs; struct x86_init_oem oem; + struct x86_init_mapping mapping; struct x86_init_paging paging; struct x86_init_timers timers; struct x86_init_iommu iommu; diff --git a/arch/x86/include/asm/xen/page.h b/arch/x86/include/asm/xen/page.h index c61934fbf22a..64a619d47d34 100644 --- a/arch/x86/include/asm/xen/page.h +++ b/arch/x86/include/asm/xen/page.h @@ -47,8 +47,9 @@ extern bool __set_phys_to_machine(unsigned long pfn, unsigned long mfn); extern unsigned long set_phys_range_identity(unsigned long pfn_s, unsigned long pfn_e); -extern int m2p_add_override(unsigned long mfn, struct page *page); -extern int m2p_remove_override(struct page *page); +extern int m2p_add_override(unsigned long mfn, struct page *page, + bool clear_pte); +extern int m2p_remove_override(struct page *page, bool clear_pte); extern struct page *m2p_find_override(unsigned long mfn); extern unsigned long m2p_find_override_pfn(unsigned long mfn, unsigned long pfn); diff --git a/arch/x86/include/asm/xen/pci.h b/arch/x86/include/asm/xen/pci.h index aa8620989162..4fbda9a3f339 100644 --- a/arch/x86/include/asm/xen/pci.h +++ b/arch/x86/include/asm/xen/pci.h @@ -15,10 +15,26 @@ static inline int pci_xen_hvm_init(void) #endif #if defined(CONFIG_XEN_DOM0) void __init xen_setup_pirqs(void); +int xen_find_device_domain_owner(struct pci_dev *dev); +int xen_register_device_domain_owner(struct pci_dev *dev, uint16_t domain); +int xen_unregister_device_domain_owner(struct pci_dev *dev); #else static inline void __init xen_setup_pirqs(void) { } +static inline int xen_find_device_domain_owner(struct pci_dev *dev) +{ + return -1; +} +static inline int xen_register_device_domain_owner(struct pci_dev *dev, + uint16_t domain) +{ + return -1; +} +static inline int xen_unregister_device_domain_owner(struct pci_dev *dev) +{ + return -1; +} #endif #if defined(CONFIG_PCI_MSI) diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile index 7338ef2218bc..250806472a7e 100644 --- a/arch/x86/kernel/Makefile +++ b/arch/x86/kernel/Makefile @@ -36,7 +36,7 @@ obj-y += traps.o irq.o irq_$(BITS).o dumpstack_$(BITS).o obj-y += time.o ioport.o ldt.o dumpstack.o obj-y += setup.o x86_init.o i8259.o irqinit.o jump_label.o obj-$(CONFIG_IRQ_WORK) += irq_work.o -obj-$(CONFIG_X86_32) += probe_roms_32.o +obj-y += probe_roms.o obj-$(CONFIG_X86_32) += sys_i386_32.o i386_ksyms_32.o obj-$(CONFIG_X86_64) += sys_x86_64.o x8664_ksyms_64.o obj-$(CONFIG_X86_64) += syscall_64.o vsyscall_64.o @@ -117,7 +117,7 @@ obj-$(CONFIG_OF) += devicetree.o ifeq ($(CONFIG_X86_64),y) obj-$(CONFIG_AUDIT) += audit_64.o - obj-$(CONFIG_GART_IOMMU) += pci-gart_64.o aperture_64.o + obj-$(CONFIG_GART_IOMMU) += amd_gart_64.o aperture_64.o obj-$(CONFIG_CALGARY_IOMMU) += pci-calgary_64.o tce_64.o obj-$(CONFIG_AMD_IOMMU) += amd_iommu_init.o amd_iommu.o diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c index 9a966c579af5..4558f0d0822d 100644 --- a/arch/x86/kernel/acpi/boot.c +++ b/arch/x86/kernel/acpi/boot.c @@ -970,7 +970,7 @@ void __init mp_override_legacy_irq(u8 bus_irq, u8 polarity, u8 trigger, u32 gsi) mp_irq.irqflag = (trigger << 2) | polarity; mp_irq.srcbus = MP_ISA_BUS; mp_irq.srcbusirq = bus_irq; /* IRQ */ - mp_irq.dstapic = mp_ioapics[ioapic].apicid; /* APIC ID */ + mp_irq.dstapic = mpc_ioapic_id(ioapic); /* APIC ID */ mp_irq.dstirq = pin; /* INTIN# */ mp_save_irq(&mp_irq); @@ -1021,7 +1021,7 @@ void __init mp_config_acpi_legacy_irqs(void) if (ioapic < 0) continue; pin = mp_find_ioapic_pin(ioapic, gsi); - dstapic = mp_ioapics[ioapic].apicid; + dstapic = mpc_ioapic_id(ioapic); for (idx = 0; idx < mp_irq_entries; idx++) { struct mpc_intsrc *irq = mp_irqs + idx; @@ -1082,7 +1082,7 @@ static int mp_config_acpi_gsi(struct device *dev, u32 gsi, int trigger, mp_irq.srcbus = number; mp_irq.srcbusirq = (((devfn >> 3) & 0x1f) << 2) | ((pin - 1) & 3); ioapic = mp_find_ioapic(gsi); - mp_irq.dstapic = mp_ioapics[ioapic].apicid; + mp_irq.dstapic = mpc_ioapic_id(ioapic); mp_irq.dstirq = mp_find_ioapic_pin(ioapic, gsi); mp_save_irq(&mp_irq); @@ -1113,7 +1113,7 @@ int mp_register_gsi(struct device *dev, u32 gsi, int trigger, int polarity) if (ioapic_pin > MP_MAX_IOAPIC_PIN) { printk(KERN_ERR "Invalid reference to IOAPIC pin " - "%d-%d\n", mp_ioapics[ioapic].apicid, + "%d-%d\n", mpc_ioapic_id(ioapic), ioapic_pin); return gsi; } diff --git a/arch/x86/kernel/acpi/sleep.c b/arch/x86/kernel/acpi/sleep.c index ff93bc1b09c3..18a857ba7a25 100644 --- a/arch/x86/kernel/acpi/sleep.c +++ b/arch/x86/kernel/acpi/sleep.c @@ -112,11 +112,6 @@ static int __init acpi_sleep_setup(char *str) #ifdef CONFIG_HIBERNATION if (strncmp(str, "s4_nohwsig", 10) == 0) acpi_no_s4_hw_signature(); - if (strncmp(str, "s4_nonvs", 8) == 0) { - pr_warning("ACPI: acpi_sleep=s4_nonvs is deprecated, " - "please use acpi_sleep=nonvs instead"); - acpi_nvs_nosave(); - } #endif if (strncmp(str, "nonvs", 5) == 0) acpi_nvs_nosave(); diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c index 4a234677e213..a81f2d52f869 100644 --- a/arch/x86/kernel/alternative.c +++ b/arch/x86/kernel/alternative.c @@ -67,17 +67,30 @@ __setup("noreplace-paravirt", setup_noreplace_paravirt); #define DPRINTK(fmt, args...) if (debug_alternative) \ printk(KERN_DEBUG fmt, args) +/* + * Each GENERIC_NOPX is of X bytes, and defined as an array of bytes + * that correspond to that nop. Getting from one nop to the next, we + * add to the array the offset that is equal to the sum of all sizes of + * nops preceding the one we are after. + * + * Note: The GENERIC_NOP5_ATOMIC is at the end, as it breaks the + * nice symmetry of sizes of the previous nops. + */ #if defined(GENERIC_NOP1) && !defined(CONFIG_X86_64) -/* Use inline assembly to define this because the nops are defined - as inline assembly strings in the include files and we cannot - get them easily into strings. */ -asm("\t" __stringify(__INITRODATA_OR_MODULE) "\nintelnops: " - GENERIC_NOP1 GENERIC_NOP2 GENERIC_NOP3 GENERIC_NOP4 GENERIC_NOP5 GENERIC_NOP6 - GENERIC_NOP7 GENERIC_NOP8 - "\t.previous"); -extern const unsigned char intelnops[]; -static const unsigned char *const __initconst_or_module -intel_nops[ASM_NOP_MAX+1] = { +static const unsigned char intelnops[] = +{ + GENERIC_NOP1, + GENERIC_NOP2, + GENERIC_NOP3, + GENERIC_NOP4, + GENERIC_NOP5, + GENERIC_NOP6, + GENERIC_NOP7, + GENERIC_NOP8, + GENERIC_NOP5_ATOMIC +}; +static const unsigned char * const intel_nops[ASM_NOP_MAX+2] = +{ NULL, intelnops, intelnops + 1, @@ -87,17 +100,25 @@ intel_nops[ASM_NOP_MAX+1] = { intelnops + 1 + 2 + 3 + 4 + 5, intelnops + 1 + 2 + 3 + 4 + 5 + 6, intelnops + 1 + 2 + 3 + 4 + 5 + 6 + 7, + intelnops + 1 + 2 + 3 + 4 + 5 + 6 + 7 + 8, }; #endif #ifdef K8_NOP1 -asm("\t" __stringify(__INITRODATA_OR_MODULE) "\nk8nops: " - K8_NOP1 K8_NOP2 K8_NOP3 K8_NOP4 K8_NOP5 K8_NOP6 - K8_NOP7 K8_NOP8 - "\t.previous"); -extern const unsigned char k8nops[]; -static const unsigned char *const __initconst_or_module -k8_nops[ASM_NOP_MAX+1] = { +static const unsigned char k8nops[] = +{ + K8_NOP1, + K8_NOP2, + K8_NOP3, + K8_NOP4, + K8_NOP5, + K8_NOP6, + K8_NOP7, + K8_NOP8, + K8_NOP5_ATOMIC +}; +static const unsigned char * const k8_nops[ASM_NOP_MAX+2] = +{ NULL, k8nops, k8nops + 1, @@ -107,17 +128,25 @@ k8_nops[ASM_NOP_MAX+1] = { k8nops + 1 + 2 + 3 + 4 + 5, k8nops + 1 + 2 + 3 + 4 + 5 + 6, k8nops + 1 + 2 + 3 + 4 + 5 + 6 + 7, + k8nops + 1 + 2 + 3 + 4 + 5 + 6 + 7 + 8, }; #endif #if defined(K7_NOP1) && !defined(CONFIG_X86_64) -asm("\t" __stringify(__INITRODATA_OR_MODULE) "\nk7nops: " - K7_NOP1 K7_NOP2 K7_NOP3 K7_NOP4 K7_NOP5 K7_NOP6 - K7_NOP7 K7_NOP8 - "\t.previous"); -extern const unsigned char k7nops[]; -static const unsigned char *const __initconst_or_module -k7_nops[ASM_NOP_MAX+1] = { +static const unsigned char k7nops[] = +{ + K7_NOP1, + K7_NOP2, + K7_NOP3, + K7_NOP4, + K7_NOP5, + K7_NOP6, + K7_NOP7, + K7_NOP8, + K7_NOP5_ATOMIC +}; +static const unsigned char * const k7_nops[ASM_NOP_MAX+2] = +{ NULL, k7nops, k7nops + 1, @@ -127,17 +156,25 @@ k7_nops[ASM_NOP_MAX+1] = { k7nops + 1 + 2 + 3 + 4 + 5, k7nops + 1 + 2 + 3 + 4 + 5 + 6, k7nops + 1 + 2 + 3 + 4 + 5 + 6 + 7, + k7nops + 1 + 2 + 3 + 4 + 5 + 6 + 7 + 8, }; #endif #ifdef P6_NOP1 -asm("\t" __stringify(__INITRODATA_OR_MODULE) "\np6nops: " - P6_NOP1 P6_NOP2 P6_NOP3 P6_NOP4 P6_NOP5 P6_NOP6 - P6_NOP7 P6_NOP8 - "\t.previous"); -extern const unsigned char p6nops[]; -static const unsigned char *const __initconst_or_module -p6_nops[ASM_NOP_MAX+1] = { +static const unsigned char __initconst_or_module p6nops[] = +{ + P6_NOP1, + P6_NOP2, + P6_NOP3, + P6_NOP4, + P6_NOP5, + P6_NOP6, + P6_NOP7, + P6_NOP8, + P6_NOP5_ATOMIC +}; +static const unsigned char * const p6_nops[ASM_NOP_MAX+2] = +{ NULL, p6nops, p6nops + 1, @@ -147,47 +184,65 @@ p6_nops[ASM_NOP_MAX+1] = { p6nops + 1 + 2 + 3 + 4 + 5, p6nops + 1 + 2 + 3 + 4 + 5 + 6, p6nops + 1 + 2 + 3 + 4 + 5 + 6 + 7, + p6nops + 1 + 2 + 3 + 4 + 5 + 6 + 7 + 8, }; #endif +/* Initialize these to a safe default */ #ifdef CONFIG_X86_64 +const unsigned char * const *ideal_nops = p6_nops; +#else +const unsigned char * const *ideal_nops = intel_nops; +#endif -extern char __vsyscall_0; -static const unsigned char *const *__init_or_module find_nop_table(void) +void __init arch_init_ideal_nops(void) { - if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL && - boot_cpu_has(X86_FEATURE_NOPL)) - return p6_nops; - else - return k8_nops; -} - -#else /* CONFIG_X86_64 */ + switch (boot_cpu_data.x86_vendor) { + case X86_VENDOR_INTEL: + /* + * Due to a decoder implementation quirk, some + * specific Intel CPUs actually perform better with + * the "k8_nops" than with the SDM-recommended NOPs. + */ + if (boot_cpu_data.x86 == 6 && + boot_cpu_data.x86_model >= 0x0f && + boot_cpu_data.x86_model != 0x1c && + boot_cpu_data.x86_model != 0x26 && + boot_cpu_data.x86_model != 0x27 && + boot_cpu_data.x86_model < 0x30) { + ideal_nops = k8_nops; + } else if (boot_cpu_has(X86_FEATURE_NOPL)) { + ideal_nops = p6_nops; + } else { +#ifdef CONFIG_X86_64 + ideal_nops = k8_nops; +#else + ideal_nops = intel_nops; +#endif + } -static const unsigned char *const *__init_or_module find_nop_table(void) -{ - if (boot_cpu_has(X86_FEATURE_K8)) - return k8_nops; - else if (boot_cpu_has(X86_FEATURE_K7)) - return k7_nops; - else if (boot_cpu_has(X86_FEATURE_NOPL)) - return p6_nops; - else - return intel_nops; + default: +#ifdef CONFIG_X86_64 + ideal_nops = k8_nops; +#else + if (boot_cpu_has(X86_FEATURE_K8)) + ideal_nops = k8_nops; + else if (boot_cpu_has(X86_FEATURE_K7)) + ideal_nops = k7_nops; + else + ideal_nops = intel_nops; +#endif + } } -#endif /* CONFIG_X86_64 */ - /* Use this to add nops to a buffer, then text_poke the whole buffer. */ static void __init_or_module add_nops(void *insns, unsigned int len) { - const unsigned char *const *noptable = find_nop_table(); - while (len > 0) { unsigned int noplen = len; if (noplen > ASM_NOP_MAX) noplen = ASM_NOP_MAX; - memcpy(insns, noptable[noplen], noplen); + memcpy(insns, ideal_nops[noplen], noplen); insns += noplen; len -= noplen; } @@ -195,6 +250,7 @@ static void __init_or_module add_nops(void *insns, unsigned int len) extern struct alt_instr __alt_instructions[], __alt_instructions_end[]; extern s32 __smp_locks[], __smp_locks_end[]; +extern char __vsyscall_0; void *text_poke_early(void *addr, const void *opcode, size_t len); /* Replace instructions with better alternatives for this CPU type. @@ -210,6 +266,15 @@ void __init_or_module apply_alternatives(struct alt_instr *start, u8 insnbuf[MAX_PATCH_LEN]; DPRINTK("%s: alt table %p -> %p\n", __func__, start, end); + /* + * The scan order should be from start to end. A later scanned + * alternative code can overwrite a previous scanned alternative code. + * Some kernel functions (e.g. memcpy, memset, etc) use this order to + * patch code. + * + * So be careful if you want to change the scan order to any other + * order. + */ for (a = start; a < end; a++) { u8 *instr = a->instr; BUG_ON(a->replacementlen > a->instrlen); @@ -678,29 +743,3 @@ void __kprobes text_poke_smp_batch(struct text_poke_param *params, int n) wrote_text = 0; __stop_machine(stop_machine_text_poke, (void *)&tpp, NULL); } - -#if defined(CONFIG_DYNAMIC_FTRACE) || defined(HAVE_JUMP_LABEL) - -#ifdef CONFIG_X86_64 -unsigned char ideal_nop5[5] = { 0x66, 0x66, 0x66, 0x66, 0x90 }; -#else -unsigned char ideal_nop5[5] = { 0x3e, 0x8d, 0x74, 0x26, 0x00 }; -#endif - -void __init arch_init_ideal_nop5(void) -{ - /* - * There is no good nop for all x86 archs. This selection - * algorithm should be unified with the one in find_nop_table(), - * but this should be good enough for now. - * - * For cases other than the ones below, use the safe (as in - * always functional) defaults above. - */ -#ifdef CONFIG_X86_64 - /* Don't use these on 32 bits due to broken virtualizers */ - if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) - memcpy(ideal_nop5, p6_nops[5], 5); -#endif -} -#endif diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/amd_gart_64.c index b117efd24f71..b117efd24f71 100644 --- a/arch/x86/kernel/pci-gart_64.c +++ b/arch/x86/kernel/amd_gart_64.c diff --git a/arch/x86/kernel/amd_iommu.c b/arch/x86/kernel/amd_iommu.c index 57ca77787220..cd8cbeb5fa34 100644 --- a/arch/x86/kernel/amd_iommu.c +++ b/arch/x86/kernel/amd_iommu.c @@ -18,6 +18,7 @@ */ #include <linux/pci.h> +#include <linux/pci-ats.h> #include <linux/bitmap.h> #include <linux/slab.h> #include <linux/debugfs.h> @@ -25,6 +26,7 @@ #include <linux/dma-mapping.h> #include <linux/iommu-helper.h> #include <linux/iommu.h> +#include <linux/delay.h> #include <asm/proto.h> #include <asm/iommu.h> #include <asm/gart.h> @@ -34,7 +36,7 @@ #define CMD_SET_TYPE(cmd, t) ((cmd)->data[1] |= ((t) << 28)) -#define EXIT_LOOP_COUNT 10000000 +#define LOOP_TIMEOUT 100000 static DEFINE_RWLOCK(amd_iommu_devtable_lock); @@ -57,7 +59,6 @@ struct iommu_cmd { u32 data[4]; }; -static void reset_iommu_command_buffer(struct amd_iommu *iommu); static void update_domain(struct protection_domain *domain); /**************************************************************************** @@ -322,8 +323,6 @@ static void iommu_print_event(struct amd_iommu *iommu, void *__evt) break; case EVENT_TYPE_ILL_CMD: printk("ILLEGAL_COMMAND_ERROR address=0x%016llx]\n", address); - iommu->reset_in_progress = true; - reset_iommu_command_buffer(iommu); dump_command(address); break; case EVENT_TYPE_CMD_HARD_ERR: @@ -367,7 +366,7 @@ static void iommu_poll_events(struct amd_iommu *iommu) spin_unlock_irqrestore(&iommu->lock, flags); } -irqreturn_t amd_iommu_int_handler(int irq, void *data) +irqreturn_t amd_iommu_int_thread(int irq, void *data) { struct amd_iommu *iommu; @@ -377,192 +376,300 @@ irqreturn_t amd_iommu_int_handler(int irq, void *data) return IRQ_HANDLED; } +irqreturn_t amd_iommu_int_handler(int irq, void *data) +{ + return IRQ_WAKE_THREAD; +} + /**************************************************************************** * * IOMMU command queuing functions * ****************************************************************************/ -/* - * Writes the command to the IOMMUs command buffer and informs the - * hardware about the new command. Must be called with iommu->lock held. - */ -static int __iommu_queue_command(struct amd_iommu *iommu, struct iommu_cmd *cmd) +static int wait_on_sem(volatile u64 *sem) +{ + int i = 0; + + while (*sem == 0 && i < LOOP_TIMEOUT) { + udelay(1); + i += 1; + } + + if (i == LOOP_TIMEOUT) { + pr_alert("AMD-Vi: Completion-Wait loop timed out\n"); + return -EIO; + } + + return 0; +} + +static void copy_cmd_to_buffer(struct amd_iommu *iommu, + struct iommu_cmd *cmd, + u32 tail) { - u32 tail, head; u8 *target; - WARN_ON(iommu->cmd_buf_size & CMD_BUFFER_UNINITIALIZED); - tail = readl(iommu->mmio_base + MMIO_CMD_TAIL_OFFSET); target = iommu->cmd_buf + tail; - memcpy_toio(target, cmd, sizeof(*cmd)); - tail = (tail + sizeof(*cmd)) % iommu->cmd_buf_size; - head = readl(iommu->mmio_base + MMIO_CMD_HEAD_OFFSET); - if (tail == head) - return -ENOMEM; + tail = (tail + sizeof(*cmd)) % iommu->cmd_buf_size; + + /* Copy command to buffer */ + memcpy(target, cmd, sizeof(*cmd)); + + /* Tell the IOMMU about it */ writel(tail, iommu->mmio_base + MMIO_CMD_TAIL_OFFSET); +} - return 0; +static void build_completion_wait(struct iommu_cmd *cmd, u64 address) +{ + WARN_ON(address & 0x7ULL); + + memset(cmd, 0, sizeof(*cmd)); + cmd->data[0] = lower_32_bits(__pa(address)) | CMD_COMPL_WAIT_STORE_MASK; + cmd->data[1] = upper_32_bits(__pa(address)); + cmd->data[2] = 1; + CMD_SET_TYPE(cmd, CMD_COMPL_WAIT); +} + +static void build_inv_dte(struct iommu_cmd *cmd, u16 devid) +{ + memset(cmd, 0, sizeof(*cmd)); + cmd->data[0] = devid; + CMD_SET_TYPE(cmd, CMD_INV_DEV_ENTRY); +} + +static void build_inv_iommu_pages(struct iommu_cmd *cmd, u64 address, + size_t size, u16 domid, int pde) +{ + u64 pages; + int s; + + pages = iommu_num_pages(address, size, PAGE_SIZE); + s = 0; + + if (pages > 1) { + /* + * If we have to flush more than one page, flush all + * TLB entries for this domain + */ + address = CMD_INV_IOMMU_ALL_PAGES_ADDRESS; + s = 1; + } + + address &= PAGE_MASK; + + memset(cmd, 0, sizeof(*cmd)); + cmd->data[1] |= domid; + cmd->data[2] = lower_32_bits(address); + cmd->data[3] = upper_32_bits(address); + CMD_SET_TYPE(cmd, CMD_INV_IOMMU_PAGES); + if (s) /* size bit - we flush more than one 4kb page */ + cmd->data[2] |= CMD_INV_IOMMU_PAGES_SIZE_MASK; + if (pde) /* PDE bit - we wan't flush everything not only the PTEs */ + cmd->data[2] |= CMD_INV_IOMMU_PAGES_PDE_MASK; +} + +static void build_inv_iotlb_pages(struct iommu_cmd *cmd, u16 devid, int qdep, + u64 address, size_t size) +{ + u64 pages; + int s; + + pages = iommu_num_pages(address, size, PAGE_SIZE); + s = 0; + + if (pages > 1) { + /* + * If we have to flush more than one page, flush all + * TLB entries for this domain + */ + address = CMD_INV_IOMMU_ALL_PAGES_ADDRESS; + s = 1; + } + + address &= PAGE_MASK; + + memset(cmd, 0, sizeof(*cmd)); + cmd->data[0] = devid; + cmd->data[0] |= (qdep & 0xff) << 24; + cmd->data[1] = devid; + cmd->data[2] = lower_32_bits(address); + cmd->data[3] = upper_32_bits(address); + CMD_SET_TYPE(cmd, CMD_INV_IOTLB_PAGES); + if (s) + cmd->data[2] |= CMD_INV_IOMMU_PAGES_SIZE_MASK; +} + +static void build_inv_all(struct iommu_cmd *cmd) +{ + memset(cmd, 0, sizeof(*cmd)); + CMD_SET_TYPE(cmd, CMD_INV_ALL); } /* - * General queuing function for commands. Takes iommu->lock and calls - * __iommu_queue_command(). + * Writes the command to the IOMMUs command buffer and informs the + * hardware about the new command. */ static int iommu_queue_command(struct amd_iommu *iommu, struct iommu_cmd *cmd) { + u32 left, tail, head, next_tail; unsigned long flags; - int ret; + WARN_ON(iommu->cmd_buf_size & CMD_BUFFER_UNINITIALIZED); + +again: spin_lock_irqsave(&iommu->lock, flags); - ret = __iommu_queue_command(iommu, cmd); - if (!ret) - iommu->need_sync = true; - spin_unlock_irqrestore(&iommu->lock, flags); - return ret; -} + head = readl(iommu->mmio_base + MMIO_CMD_HEAD_OFFSET); + tail = readl(iommu->mmio_base + MMIO_CMD_TAIL_OFFSET); + next_tail = (tail + sizeof(*cmd)) % iommu->cmd_buf_size; + left = (head - next_tail) % iommu->cmd_buf_size; -/* - * This function waits until an IOMMU has completed a completion - * wait command - */ -static void __iommu_wait_for_completion(struct amd_iommu *iommu) -{ - int ready = 0; - unsigned status = 0; - unsigned long i = 0; + if (left <= 2) { + struct iommu_cmd sync_cmd; + volatile u64 sem = 0; + int ret; + + build_completion_wait(&sync_cmd, (u64)&sem); + copy_cmd_to_buffer(iommu, &sync_cmd, tail); - INC_STATS_COUNTER(compl_wait); + spin_unlock_irqrestore(&iommu->lock, flags); + + if ((ret = wait_on_sem(&sem)) != 0) + return ret; - while (!ready && (i < EXIT_LOOP_COUNT)) { - ++i; - /* wait for the bit to become one */ - status = readl(iommu->mmio_base + MMIO_STATUS_OFFSET); - ready = status & MMIO_STATUS_COM_WAIT_INT_MASK; + goto again; } - /* set bit back to zero */ - status &= ~MMIO_STATUS_COM_WAIT_INT_MASK; - writel(status, iommu->mmio_base + MMIO_STATUS_OFFSET); + copy_cmd_to_buffer(iommu, cmd, tail); + + /* We need to sync now to make sure all commands are processed */ + iommu->need_sync = true; + + spin_unlock_irqrestore(&iommu->lock, flags); - if (unlikely(i == EXIT_LOOP_COUNT)) - iommu->reset_in_progress = true; + return 0; } /* * This function queues a completion wait command into the command * buffer of an IOMMU */ -static int __iommu_completion_wait(struct amd_iommu *iommu) +static int iommu_completion_wait(struct amd_iommu *iommu) { struct iommu_cmd cmd; + volatile u64 sem = 0; + int ret; + + if (!iommu->need_sync) + return 0; - memset(&cmd, 0, sizeof(cmd)); - cmd.data[0] = CMD_COMPL_WAIT_INT_MASK; - CMD_SET_TYPE(&cmd, CMD_COMPL_WAIT); + build_completion_wait(&cmd, (u64)&sem); - return __iommu_queue_command(iommu, &cmd); + ret = iommu_queue_command(iommu, &cmd); + if (ret) + return ret; + + return wait_on_sem(&sem); } -/* - * This function is called whenever we need to ensure that the IOMMU has - * completed execution of all commands we sent. It sends a - * COMPLETION_WAIT command and waits for it to finish. The IOMMU informs - * us about that by writing a value to a physical address we pass with - * the command. - */ -static int iommu_completion_wait(struct amd_iommu *iommu) +static int iommu_flush_dte(struct amd_iommu *iommu, u16 devid) { - int ret = 0; - unsigned long flags; - - spin_lock_irqsave(&iommu->lock, flags); + struct iommu_cmd cmd; - if (!iommu->need_sync) - goto out; + build_inv_dte(&cmd, devid); - ret = __iommu_completion_wait(iommu); + return iommu_queue_command(iommu, &cmd); +} - iommu->need_sync = false; +static void iommu_flush_dte_all(struct amd_iommu *iommu) +{ + u32 devid; - if (ret) - goto out; + for (devid = 0; devid <= 0xffff; ++devid) + iommu_flush_dte(iommu, devid); - __iommu_wait_for_completion(iommu); + iommu_completion_wait(iommu); +} -out: - spin_unlock_irqrestore(&iommu->lock, flags); +/* + * This function uses heavy locking and may disable irqs for some time. But + * this is no issue because it is only called during resume. + */ +static void iommu_flush_tlb_all(struct amd_iommu *iommu) +{ + u32 dom_id; - if (iommu->reset_in_progress) - reset_iommu_command_buffer(iommu); + for (dom_id = 0; dom_id <= 0xffff; ++dom_id) { + struct iommu_cmd cmd; + build_inv_iommu_pages(&cmd, 0, CMD_INV_IOMMU_ALL_PAGES_ADDRESS, + dom_id, 1); + iommu_queue_command(iommu, &cmd); + } - return 0; + iommu_completion_wait(iommu); } -static void iommu_flush_complete(struct protection_domain *domain) +static void iommu_flush_all(struct amd_iommu *iommu) { - int i; + struct iommu_cmd cmd; - for (i = 0; i < amd_iommus_present; ++i) { - if (!domain->dev_iommu[i]) - continue; + build_inv_all(&cmd); - /* - * Devices of this domain are behind this IOMMU - * We need to wait for completion of all commands. - */ - iommu_completion_wait(amd_iommus[i]); + iommu_queue_command(iommu, &cmd); + iommu_completion_wait(iommu); +} + +void iommu_flush_all_caches(struct amd_iommu *iommu) +{ + if (iommu_feature(iommu, FEATURE_IA)) { + iommu_flush_all(iommu); + } else { + iommu_flush_dte_all(iommu); + iommu_flush_tlb_all(iommu); } } /* - * Command send function for invalidating a device table entry + * Command send function for flushing on-device TLB */ -static int iommu_flush_device(struct device *dev) +static int device_flush_iotlb(struct device *dev, u64 address, size_t size) { + struct pci_dev *pdev = to_pci_dev(dev); struct amd_iommu *iommu; struct iommu_cmd cmd; u16 devid; + int qdep; + qdep = pci_ats_queue_depth(pdev); devid = get_device_id(dev); iommu = amd_iommu_rlookup_table[devid]; - /* Build command */ - memset(&cmd, 0, sizeof(cmd)); - CMD_SET_TYPE(&cmd, CMD_INV_DEV_ENTRY); - cmd.data[0] = devid; + build_inv_iotlb_pages(&cmd, devid, qdep, address, size); return iommu_queue_command(iommu, &cmd); } -static void __iommu_build_inv_iommu_pages(struct iommu_cmd *cmd, u64 address, - u16 domid, int pde, int s) -{ - memset(cmd, 0, sizeof(*cmd)); - address &= PAGE_MASK; - CMD_SET_TYPE(cmd, CMD_INV_IOMMU_PAGES); - cmd->data[1] |= domid; - cmd->data[2] = lower_32_bits(address); - cmd->data[3] = upper_32_bits(address); - if (s) /* size bit - we flush more than one 4kb page */ - cmd->data[2] |= CMD_INV_IOMMU_PAGES_SIZE_MASK; - if (pde) /* PDE bit - we wan't flush everything not only the PTEs */ - cmd->data[2] |= CMD_INV_IOMMU_PAGES_PDE_MASK; -} - /* - * Generic command send function for invalidaing TLB entries + * Command send function for invalidating a device table entry */ -static int iommu_queue_inv_iommu_pages(struct amd_iommu *iommu, - u64 address, u16 domid, int pde, int s) +static int device_flush_dte(struct device *dev) { - struct iommu_cmd cmd; + struct amd_iommu *iommu; + struct pci_dev *pdev; + u16 devid; int ret; - __iommu_build_inv_iommu_pages(&cmd, address, domid, pde, s); + pdev = to_pci_dev(dev); + devid = get_device_id(dev); + iommu = amd_iommu_rlookup_table[devid]; - ret = iommu_queue_command(iommu, &cmd); + ret = iommu_flush_dte(iommu, devid); + if (ret) + return ret; + + if (pci_ats_enabled(pdev)) + ret = device_flush_iotlb(dev, 0, ~0UL); return ret; } @@ -572,23 +679,14 @@ static int iommu_queue_inv_iommu_pages(struct amd_iommu *iommu, * It invalidates a single PTE if the range to flush is within a single * page. Otherwise it flushes the whole TLB of the IOMMU. */ -static void __iommu_flush_pages(struct protection_domain *domain, - u64 address, size_t size, int pde) +static void __domain_flush_pages(struct protection_domain *domain, + u64 address, size_t size, int pde) { - int s = 0, i; - unsigned long pages = iommu_num_pages(address, size, PAGE_SIZE); - - address &= PAGE_MASK; - - if (pages > 1) { - /* - * If we have to flush more than one page, flush all - * TLB entries for this domain - */ - address = CMD_INV_IOMMU_ALL_PAGES_ADDRESS; - s = 1; - } + struct iommu_dev_data *dev_data; + struct iommu_cmd cmd; + int ret = 0, i; + build_inv_iommu_pages(&cmd, address, size, domain->id, pde); for (i = 0; i < amd_iommus_present; ++i) { if (!domain->dev_iommu[i]) @@ -598,101 +696,70 @@ static void __iommu_flush_pages(struct protection_domain *domain, * Devices of this domain are behind this IOMMU * We need a TLB flush */ - iommu_queue_inv_iommu_pages(amd_iommus[i], address, - domain->id, pde, s); + ret |= iommu_queue_command(amd_iommus[i], &cmd); + } + + list_for_each_entry(dev_data, &domain->dev_list, list) { + struct pci_dev *pdev = to_pci_dev(dev_data->dev); + + if (!pci_ats_enabled(pdev)) + continue; + + ret |= device_flush_iotlb(dev_data->dev, address, size); } - return; + WARN_ON(ret); } -static void iommu_flush_pages(struct protection_domain *domain, - u64 address, size_t size) +static void domain_flush_pages(struct protection_domain *domain, + u64 address, size_t size) { - __iommu_flush_pages(domain, address, size, 0); + __domain_flush_pages(domain, address, size, 0); } /* Flush the whole IO/TLB for a given protection domain */ -static void iommu_flush_tlb(struct protection_domain *domain) +static void domain_flush_tlb(struct protection_domain *domain) { - __iommu_flush_pages(domain, 0, CMD_INV_IOMMU_ALL_PAGES_ADDRESS, 0); + __domain_flush_pages(domain, 0, CMD_INV_IOMMU_ALL_PAGES_ADDRESS, 0); } /* Flush the whole IO/TLB for a given protection domain - including PDE */ -static void iommu_flush_tlb_pde(struct protection_domain *domain) +static void domain_flush_tlb_pde(struct protection_domain *domain) { - __iommu_flush_pages(domain, 0, CMD_INV_IOMMU_ALL_PAGES_ADDRESS, 1); -} - - -/* - * This function flushes the DTEs for all devices in domain - */ -static void iommu_flush_domain_devices(struct protection_domain *domain) -{ - struct iommu_dev_data *dev_data; - unsigned long flags; - - spin_lock_irqsave(&domain->lock, flags); - - list_for_each_entry(dev_data, &domain->dev_list, list) - iommu_flush_device(dev_data->dev); - - spin_unlock_irqrestore(&domain->lock, flags); + __domain_flush_pages(domain, 0, CMD_INV_IOMMU_ALL_PAGES_ADDRESS, 1); } -static void iommu_flush_all_domain_devices(void) +static void domain_flush_complete(struct protection_domain *domain) { - struct protection_domain *domain; - unsigned long flags; + int i; - spin_lock_irqsave(&amd_iommu_pd_lock, flags); + for (i = 0; i < amd_iommus_present; ++i) { + if (!domain->dev_iommu[i]) + continue; - list_for_each_entry(domain, &amd_iommu_pd_list, list) { - iommu_flush_domain_devices(domain); - iommu_flush_complete(domain); + /* + * Devices of this domain are behind this IOMMU + * We need to wait for completion of all commands. + */ + iommu_completion_wait(amd_iommus[i]); } - - spin_unlock_irqrestore(&amd_iommu_pd_lock, flags); } -void amd_iommu_flush_all_devices(void) -{ - iommu_flush_all_domain_devices(); -} /* - * This function uses heavy locking and may disable irqs for some time. But - * this is no issue because it is only called during resume. + * This function flushes the DTEs for all devices in domain */ -void amd_iommu_flush_all_domains(void) +static void domain_flush_devices(struct protection_domain *domain) { - struct protection_domain *domain; + struct iommu_dev_data *dev_data; unsigned long flags; - spin_lock_irqsave(&amd_iommu_pd_lock, flags); - - list_for_each_entry(domain, &amd_iommu_pd_list, list) { - spin_lock(&domain->lock); - iommu_flush_tlb_pde(domain); - iommu_flush_complete(domain); - spin_unlock(&domain->lock); - } - - spin_unlock_irqrestore(&amd_iommu_pd_lock, flags); -} - -static void reset_iommu_command_buffer(struct amd_iommu *iommu) -{ - pr_err("AMD-Vi: Resetting IOMMU command buffer\n"); - - if (iommu->reset_in_progress) - panic("AMD-Vi: ILLEGAL_COMMAND_ERROR while resetting command buffer\n"); + spin_lock_irqsave(&domain->lock, flags); - amd_iommu_reset_cmd_buffer(iommu); - amd_iommu_flush_all_devices(); - amd_iommu_flush_all_domains(); + list_for_each_entry(dev_data, &domain->dev_list, list) + device_flush_dte(dev_data->dev); - iommu->reset_in_progress = false; + spin_unlock_irqrestore(&domain->lock, flags); } /**************************************************************************** @@ -1410,17 +1477,22 @@ static bool dma_ops_domain(struct protection_domain *domain) return domain->flags & PD_DMA_OPS_MASK; } -static void set_dte_entry(u16 devid, struct protection_domain *domain) +static void set_dte_entry(u16 devid, struct protection_domain *domain, bool ats) { u64 pte_root = virt_to_phys(domain->pt_root); + u32 flags = 0; pte_root |= (domain->mode & DEV_ENTRY_MODE_MASK) << DEV_ENTRY_MODE_SHIFT; pte_root |= IOMMU_PTE_IR | IOMMU_PTE_IW | IOMMU_PTE_P | IOMMU_PTE_TV; - amd_iommu_dev_table[devid].data[2] = domain->id; - amd_iommu_dev_table[devid].data[1] = upper_32_bits(pte_root); - amd_iommu_dev_table[devid].data[0] = lower_32_bits(pte_root); + if (ats) + flags |= DTE_FLAG_IOTLB; + + amd_iommu_dev_table[devid].data[3] |= flags; + amd_iommu_dev_table[devid].data[2] = domain->id; + amd_iommu_dev_table[devid].data[1] = upper_32_bits(pte_root); + amd_iommu_dev_table[devid].data[0] = lower_32_bits(pte_root); } static void clear_dte_entry(u16 devid) @@ -1437,23 +1509,29 @@ static void do_attach(struct device *dev, struct protection_domain *domain) { struct iommu_dev_data *dev_data; struct amd_iommu *iommu; + struct pci_dev *pdev; + bool ats = false; u16 devid; devid = get_device_id(dev); iommu = amd_iommu_rlookup_table[devid]; dev_data = get_dev_data(dev); + pdev = to_pci_dev(dev); + + if (amd_iommu_iotlb_sup) + ats = pci_ats_enabled(pdev); /* Update data structures */ dev_data->domain = domain; list_add(&dev_data->list, &domain->dev_list); - set_dte_entry(devid, domain); + set_dte_entry(devid, domain, ats); /* Do reference counting */ domain->dev_iommu[iommu->index] += 1; domain->dev_cnt += 1; /* Flush the DTE entry */ - iommu_flush_device(dev); + device_flush_dte(dev); } static void do_detach(struct device *dev) @@ -1476,7 +1554,7 @@ static void do_detach(struct device *dev) clear_dte_entry(devid); /* Flush the DTE entry */ - iommu_flush_device(dev); + device_flush_dte(dev); } /* @@ -1539,9 +1617,13 @@ out_unlock: static int attach_device(struct device *dev, struct protection_domain *domain) { + struct pci_dev *pdev = to_pci_dev(dev); unsigned long flags; int ret; + if (amd_iommu_iotlb_sup) + pci_enable_ats(pdev, PAGE_SHIFT); + write_lock_irqsave(&amd_iommu_devtable_lock, flags); ret = __attach_device(dev, domain); write_unlock_irqrestore(&amd_iommu_devtable_lock, flags); @@ -1551,7 +1633,7 @@ static int attach_device(struct device *dev, * left the caches in the IOMMU dirty. So we have to flush * here to evict all dirty stuff. */ - iommu_flush_tlb_pde(domain); + domain_flush_tlb_pde(domain); return ret; } @@ -1598,12 +1680,16 @@ static void __detach_device(struct device *dev) */ static void detach_device(struct device *dev) { + struct pci_dev *pdev = to_pci_dev(dev); unsigned long flags; /* lock device table */ write_lock_irqsave(&amd_iommu_devtable_lock, flags); __detach_device(dev); write_unlock_irqrestore(&amd_iommu_devtable_lock, flags); + + if (amd_iommu_iotlb_sup && pci_ats_enabled(pdev)) + pci_disable_ats(pdev); } /* @@ -1615,10 +1701,9 @@ static struct protection_domain *domain_for_device(struct device *dev) struct protection_domain *dom; struct iommu_dev_data *dev_data, *alias_data; unsigned long flags; - u16 devid, alias; + u16 devid; devid = get_device_id(dev); - alias = amd_iommu_alias_table[devid]; dev_data = get_dev_data(dev); alias_data = get_dev_data(dev_data->alias); if (!alias_data) @@ -1692,7 +1777,7 @@ static int device_change_notifier(struct notifier_block *nb, goto out; } - iommu_flush_device(dev); + device_flush_dte(dev); iommu_completion_wait(iommu); out: @@ -1753,8 +1838,9 @@ static void update_device_table(struct protection_domain *domain) struct iommu_dev_data *dev_data; list_for_each_entry(dev_data, &domain->dev_list, list) { + struct pci_dev *pdev = to_pci_dev(dev_data->dev); u16 devid = get_device_id(dev_data->dev); - set_dte_entry(devid, domain); + set_dte_entry(devid, domain, pci_ats_enabled(pdev)); } } @@ -1764,8 +1850,9 @@ static void update_domain(struct protection_domain *domain) return; update_device_table(domain); - iommu_flush_domain_devices(domain); - iommu_flush_tlb_pde(domain); + + domain_flush_devices(domain); + domain_flush_tlb_pde(domain); domain->updated = false; } @@ -1924,10 +2011,10 @@ retry: ADD_STATS_COUNTER(alloced_io_mem, size); if (unlikely(dma_dom->need_flush && !amd_iommu_unmap_flush)) { - iommu_flush_tlb(&dma_dom->domain); + domain_flush_tlb(&dma_dom->domain); dma_dom->need_flush = false; } else if (unlikely(amd_iommu_np_cache)) - iommu_flush_pages(&dma_dom->domain, address, size); + domain_flush_pages(&dma_dom->domain, address, size); out: return address; @@ -1976,7 +2063,7 @@ static void __unmap_single(struct dma_ops_domain *dma_dom, dma_ops_free_addresses(dma_dom, dma_addr, pages); if (amd_iommu_unmap_flush || dma_dom->need_flush) { - iommu_flush_pages(&dma_dom->domain, flush_addr, size); + domain_flush_pages(&dma_dom->domain, flush_addr, size); dma_dom->need_flush = false; } } @@ -2012,7 +2099,7 @@ static dma_addr_t map_page(struct device *dev, struct page *page, if (addr == DMA_ERROR_CODE) goto out; - iommu_flush_complete(domain); + domain_flush_complete(domain); out: spin_unlock_irqrestore(&domain->lock, flags); @@ -2039,7 +2126,7 @@ static void unmap_page(struct device *dev, dma_addr_t dma_addr, size_t size, __unmap_single(domain->priv, dma_addr, size, dir); - iommu_flush_complete(domain); + domain_flush_complete(domain); spin_unlock_irqrestore(&domain->lock, flags); } @@ -2104,7 +2191,7 @@ static int map_sg(struct device *dev, struct scatterlist *sglist, goto unmap; } - iommu_flush_complete(domain); + domain_flush_complete(domain); out: spin_unlock_irqrestore(&domain->lock, flags); @@ -2150,7 +2237,7 @@ static void unmap_sg(struct device *dev, struct scatterlist *sglist, s->dma_address = s->dma_length = 0; } - iommu_flush_complete(domain); + domain_flush_complete(domain); spin_unlock_irqrestore(&domain->lock, flags); } @@ -2200,7 +2287,7 @@ static void *alloc_coherent(struct device *dev, size_t size, goto out_free; } - iommu_flush_complete(domain); + domain_flush_complete(domain); spin_unlock_irqrestore(&domain->lock, flags); @@ -2232,7 +2319,7 @@ static void free_coherent(struct device *dev, size_t size, __unmap_single(domain->priv, dma_addr, size, DMA_BIDIRECTIONAL); - iommu_flush_complete(domain); + domain_flush_complete(domain); spin_unlock_irqrestore(&domain->lock, flags); @@ -2476,7 +2563,7 @@ static void amd_iommu_detach_device(struct iommu_domain *dom, if (!iommu) return; - iommu_flush_device(dev); + device_flush_dte(dev); iommu_completion_wait(iommu); } @@ -2542,7 +2629,7 @@ static int amd_iommu_unmap(struct iommu_domain *dom, unsigned long iova, unmap_size = iommu_unmap_page(domain, iova, page_size); mutex_unlock(&domain->api_lock); - iommu_flush_tlb_pde(domain); + domain_flush_tlb_pde(domain); return get_order(unmap_size); } diff --git a/arch/x86/kernel/amd_iommu_init.c b/arch/x86/kernel/amd_iommu_init.c index 246d727b65b7..9179c21120a8 100644 --- a/arch/x86/kernel/amd_iommu_init.c +++ b/arch/x86/kernel/amd_iommu_init.c @@ -137,6 +137,7 @@ int amd_iommus_present; /* IOMMUs have a non-present cache? */ bool amd_iommu_np_cache __read_mostly; +bool amd_iommu_iotlb_sup __read_mostly = true; /* * The ACPI table parsing functions set this variable on an error @@ -180,6 +181,12 @@ static u32 dev_table_size; /* size of the device table */ static u32 alias_table_size; /* size of the alias table */ static u32 rlookup_table_size; /* size if the rlookup table */ +/* + * This function flushes all internal caches of + * the IOMMU used by this driver. + */ +extern void iommu_flush_all_caches(struct amd_iommu *iommu); + static inline void update_last_devid(u16 devid) { if (devid > amd_iommu_last_bdf) @@ -293,9 +300,23 @@ static void iommu_feature_disable(struct amd_iommu *iommu, u8 bit) /* Function to enable the hardware */ static void iommu_enable(struct amd_iommu *iommu) { - printk(KERN_INFO "AMD-Vi: Enabling IOMMU at %s cap 0x%hx\n", + static const char * const feat_str[] = { + "PreF", "PPR", "X2APIC", "NX", "GT", "[5]", + "IA", "GA", "HE", "PC", NULL + }; + int i; + + printk(KERN_INFO "AMD-Vi: Enabling IOMMU at %s cap 0x%hx", dev_name(&iommu->dev->dev), iommu->cap_ptr); + if (iommu->cap & (1 << IOMMU_CAP_EFR)) { + printk(KERN_CONT " extended features: "); + for (i = 0; feat_str[i]; ++i) + if (iommu_feature(iommu, (1ULL << i))) + printk(KERN_CONT " %s", feat_str[i]); + } + printk(KERN_CONT "\n"); + iommu_feature_enable(iommu, CONTROL_IOMMU_EN); } @@ -651,7 +672,7 @@ static void __init set_device_exclusion_range(u16 devid, struct ivmd_header *m) static void __init init_iommu_from_pci(struct amd_iommu *iommu) { int cap_ptr = iommu->cap_ptr; - u32 range, misc; + u32 range, misc, low, high; int i, j; pci_read_config_dword(iommu->dev, cap_ptr + MMIO_CAP_HDR_OFFSET, @@ -667,6 +688,15 @@ static void __init init_iommu_from_pci(struct amd_iommu *iommu) MMIO_GET_LD(range)); iommu->evt_msi_num = MMIO_MSI_NUM(misc); + if (!(iommu->cap & (1 << IOMMU_CAP_IOTLB))) + amd_iommu_iotlb_sup = false; + + /* read extended feature bits */ + low = readl(iommu->mmio_base + MMIO_EXT_FEATURES); + high = readl(iommu->mmio_base + MMIO_EXT_FEATURES + 4); + + iommu->features = ((u64)high << 32) | low; + if (!is_rd890_iommu(iommu->dev)) return; @@ -1004,10 +1034,11 @@ static int iommu_setup_msi(struct amd_iommu *iommu) if (pci_enable_msi(iommu->dev)) return 1; - r = request_irq(iommu->dev->irq, amd_iommu_int_handler, - IRQF_SAMPLE_RANDOM, - "AMD-Vi", - NULL); + r = request_threaded_irq(iommu->dev->irq, + amd_iommu_int_handler, + amd_iommu_int_thread, + 0, "AMD-Vi", + iommu->dev); if (r) { pci_disable_msi(iommu->dev); @@ -1244,6 +1275,7 @@ static void enable_iommus(void) iommu_set_exclusion_range(iommu); iommu_init_msi(iommu); iommu_enable(iommu); + iommu_flush_all_caches(iommu); } } @@ -1274,8 +1306,8 @@ static void amd_iommu_resume(void) * we have to flush after the IOMMUs are enabled because a * disabled IOMMU will never execute the commands we send */ - amd_iommu_flush_all_devices(); - amd_iommu_flush_all_domains(); + for_each_iommu(iommu) + iommu_flush_all_caches(iommu); } static int amd_iommu_suspend(void) diff --git a/arch/x86/kernel/apb_timer.c b/arch/x86/kernel/apb_timer.c index cd1ffed4ee22..289e92862fd9 100644 --- a/arch/x86/kernel/apb_timer.c +++ b/arch/x86/kernel/apb_timer.c @@ -177,7 +177,6 @@ static struct clocksource clocksource_apbt = { .rating = APBT_CLOCKSOURCE_RATING, .read = apbt_read_clocksource, .mask = APBT_MASK, - .shift = APBT_SHIFT, .flags = CLOCK_SOURCE_IS_CONTINUOUS, .resume = apbt_restart_clocksource, }; @@ -543,14 +542,7 @@ static int apbt_clocksource_register(void) if (t1 == apbt_read_clocksource(&clocksource_apbt)) panic("APBT counter not counting. APBT disabled\n"); - /* - * initialize and register APBT clocksource - * convert that to ns/clock cycle - * mult = (ns/c) * 2^APBT_SHIFT - */ - clocksource_apbt.mult = div_sc(MSEC_PER_SEC, - (unsigned long) apbt_freq, APBT_SHIFT); - clocksource_register(&clocksource_apbt); + clocksource_register_khz(&clocksource_apbt, (u32)apbt_freq*1000); return 0; } diff --git a/arch/x86/kernel/aperture_64.c b/arch/x86/kernel/aperture_64.c index 73fb469908c6..3d2661ca6542 100644 --- a/arch/x86/kernel/aperture_64.c +++ b/arch/x86/kernel/aperture_64.c @@ -30,6 +30,22 @@ #include <asm/amd_nb.h> #include <asm/x86_init.h> +/* + * Using 512M as goal, in case kexec will load kernel_big + * that will do the on-position decompress, and could overlap with + * with the gart aperture that is used. + * Sequence: + * kernel_small + * ==> kexec (with kdump trigger path or gart still enabled) + * ==> kernel_small (gart area become e820_reserved) + * ==> kexec (with kdump trigger path or gart still enabled) + * ==> kerne_big (uncompressed size will be big than 64M or 128M) + * So don't use 512M below as gart iommu, leave the space for kernel + * code for safe. + */ +#define GART_MIN_ADDR (512ULL << 20) +#define GART_MAX_ADDR (1ULL << 32) + int gart_iommu_aperture; int gart_iommu_aperture_disabled __initdata; int gart_iommu_aperture_allowed __initdata; @@ -70,21 +86,9 @@ static u32 __init allocate_aperture(void) * memory. Unfortunately we cannot move it up because that would * make the IOMMU useless. */ - /* - * using 512M as goal, in case kexec will load kernel_big - * that will do the on position decompress, and could overlap with - * that position with gart that is used. - * sequende: - * kernel_small - * ==> kexec (with kdump trigger path or previous doesn't shutdown gart) - * ==> kernel_small(gart area become e820_reserved) - * ==> kexec (with kdump trigger path or previous doesn't shutdown gart) - * ==> kerne_big (uncompressed size will be big than 64M or 128M) - * so don't use 512M below as gart iommu, leave the space for kernel - * code for safe - */ - addr = memblock_find_in_range(0, 1ULL<<32, aper_size, 512ULL<<20); - if (addr == MEMBLOCK_ERROR || addr + aper_size > 0xffffffff) { + addr = memblock_find_in_range(GART_MIN_ADDR, GART_MAX_ADDR, + aper_size, aper_size); + if (addr == MEMBLOCK_ERROR || addr + aper_size > GART_MAX_ADDR) { printk(KERN_ERR "Cannot allocate aperture memory hole (%lx,%uK)\n", addr, aper_size>>10); diff --git a/arch/x86/kernel/apic/Makefile b/arch/x86/kernel/apic/Makefile index 3966b564ea47..767fd04f2843 100644 --- a/arch/x86/kernel/apic/Makefile +++ b/arch/x86/kernel/apic/Makefile @@ -2,20 +2,25 @@ # Makefile for local APIC drivers and for the IO-APIC code # -obj-$(CONFIG_X86_LOCAL_APIC) += apic.o apic_noop.o probe_$(BITS).o ipi.o +obj-$(CONFIG_X86_LOCAL_APIC) += apic.o apic_noop.o ipi.o obj-y += hw_nmi.o obj-$(CONFIG_X86_IO_APIC) += io_apic.o obj-$(CONFIG_SMP) += ipi.o ifeq ($(CONFIG_X86_64),y) -obj-y += apic_flat_64.o -obj-$(CONFIG_X86_X2APIC) += x2apic_cluster.o -obj-$(CONFIG_X86_X2APIC) += x2apic_phys.o +# APIC probe will depend on the listing order here obj-$(CONFIG_X86_UV) += x2apic_uv_x.o +obj-$(CONFIG_X86_X2APIC) += x2apic_phys.o +obj-$(CONFIG_X86_X2APIC) += x2apic_cluster.o +obj-y += apic_flat_64.o endif -obj-$(CONFIG_X86_BIGSMP) += bigsmp_32.o +# APIC probe will depend on the listing order here obj-$(CONFIG_X86_NUMAQ) += numaq_32.o -obj-$(CONFIG_X86_ES7000) += es7000_32.o obj-$(CONFIG_X86_SUMMIT) += summit_32.o +obj-$(CONFIG_X86_BIGSMP) += bigsmp_32.o +obj-$(CONFIG_X86_ES7000) += es7000_32.o + +# For 32bit, probe_32 need to be listed last +obj-$(CONFIG_X86_LOCAL_APIC) += probe_$(BITS).o diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c index fabf01eff771..b961af86bfea 100644 --- a/arch/x86/kernel/apic/apic.c +++ b/arch/x86/kernel/apic/apic.c @@ -505,7 +505,7 @@ static void __cpuinit setup_APIC_timer(void) { struct clock_event_device *levt = &__get_cpu_var(lapic_events); - if (cpu_has(__this_cpu_ptr(&cpu_info), X86_FEATURE_ARAT)) { + if (this_cpu_has(X86_FEATURE_ARAT)) { lapic_clockevent.features &= ~CLOCK_EVT_FEAT_C3STOP; /* Make LAPIC timer preferrable over percpu HPET */ lapic_clockevent.rating = 150; @@ -1237,6 +1237,17 @@ void __cpuinit setup_local_APIC(void) /* always use the value from LDR */ early_per_cpu(x86_cpu_to_logical_apicid, cpu) = logical_smp_processor_id(); + + /* + * Some NUMA implementations (NUMAQ) don't initialize apicid to + * node mapping during NUMA init. Now that logical apicid is + * guaranteed to be known, give it another chance. This is already + * a bit too late - percpu allocation has already happened without + * proper NUMA affinity. + */ + if (apic->x86_32_numa_cpu_node) + set_apicid_to_node(early_per_cpu(x86_cpu_to_apicid, cpu), + apic->x86_32_numa_cpu_node(cpu)); #endif /* @@ -1450,7 +1461,6 @@ int __init enable_IR(void) void __init enable_IR_x2apic(void) { unsigned long flags; - struct IO_APIC_route_entry **ioapic_entries; int ret, x2apic_enabled = 0; int dmar_table_init_ret; @@ -1458,13 +1468,7 @@ void __init enable_IR_x2apic(void) if (dmar_table_init_ret && !x2apic_supported()) return; - ioapic_entries = alloc_ioapic_entries(); - if (!ioapic_entries) { - pr_err("Allocate ioapic_entries failed\n"); - goto out; - } - - ret = save_IO_APIC_setup(ioapic_entries); + ret = save_ioapic_entries(); if (ret) { pr_info("Saving IO-APIC state failed: %d\n", ret); goto out; @@ -1472,7 +1476,7 @@ void __init enable_IR_x2apic(void) local_irq_save(flags); legacy_pic->mask_all(); - mask_IO_APIC_setup(ioapic_entries); + mask_ioapic_entries(); if (dmar_table_init_ret) ret = 0; @@ -1503,14 +1507,11 @@ void __init enable_IR_x2apic(void) nox2apic: if (!ret) /* IR enabling failed */ - restore_IO_APIC_setup(ioapic_entries); + restore_ioapic_entries(); legacy_pic->restore_mask(); local_irq_restore(flags); out: - if (ioapic_entries) - free_ioapic_entries(ioapic_entries); - if (x2apic_enabled) return; @@ -1812,30 +1813,41 @@ void smp_spurious_interrupt(struct pt_regs *regs) */ void smp_error_interrupt(struct pt_regs *regs) { - u32 v, v1; + u32 v0, v1; + u32 i = 0; + static const char * const error_interrupt_reason[] = { + "Send CS error", /* APIC Error Bit 0 */ + "Receive CS error", /* APIC Error Bit 1 */ + "Send accept error", /* APIC Error Bit 2 */ + "Receive accept error", /* APIC Error Bit 3 */ + "Redirectable IPI", /* APIC Error Bit 4 */ + "Send illegal vector", /* APIC Error Bit 5 */ + "Received illegal vector", /* APIC Error Bit 6 */ + "Illegal register address", /* APIC Error Bit 7 */ + }; exit_idle(); irq_enter(); /* First tickle the hardware, only then report what went on. -- REW */ - v = apic_read(APIC_ESR); + v0 = apic_read(APIC_ESR); apic_write(APIC_ESR, 0); v1 = apic_read(APIC_ESR); ack_APIC_irq(); atomic_inc(&irq_err_count); - /* - * Here is what the APIC error bits mean: - * 0: Send CS error - * 1: Receive CS error - * 2: Send accept error - * 3: Receive accept error - * 4: Reserved - * 5: Send illegal vector - * 6: Received illegal vector - * 7: Illegal register address - */ - pr_debug("APIC error on CPU%d: %02x(%02x)\n", - smp_processor_id(), v , v1); + apic_printk(APIC_DEBUG, KERN_DEBUG "APIC error on CPU%d: %02x(%02x)", + smp_processor_id(), v0 , v1); + + v1 = v1 & 0xff; + while (v1) { + if (v1 & 0x1) + apic_printk(APIC_DEBUG, KERN_CONT " : %s", error_interrupt_reason[i]); + i++; + v1 >>= 1; + }; + + apic_printk(APIC_DEBUG, KERN_CONT "\n"); + irq_exit(); } @@ -2003,21 +2015,6 @@ void default_init_apic_ldr(void) apic_write(APIC_LDR, val); } -#ifdef CONFIG_X86_32 -int default_x86_32_numa_cpu_node(int cpu) -{ -#ifdef CONFIG_NUMA - int apicid = early_per_cpu(x86_cpu_to_apicid, cpu); - - if (apicid != BAD_APICID) - return __apicid_to_node[apicid]; - return NUMA_NO_NODE; -#else - return 0; -#endif -} -#endif - /* * Power management */ @@ -2088,28 +2085,20 @@ static void lapic_resume(void) { unsigned int l, h; unsigned long flags; - int maxlvt, ret; - struct IO_APIC_route_entry **ioapic_entries = NULL; + int maxlvt; if (!apic_pm_state.active) return; local_irq_save(flags); if (intr_remapping_enabled) { - ioapic_entries = alloc_ioapic_entries(); - if (!ioapic_entries) { - WARN(1, "Alloc ioapic_entries in lapic resume failed."); - goto restore; - } - - ret = save_IO_APIC_setup(ioapic_entries); - if (ret) { - WARN(1, "Saving IO-APIC state failed: %d\n", ret); - free_ioapic_entries(ioapic_entries); - goto restore; - } - - mask_IO_APIC_setup(ioapic_entries); + /* + * IO-APIC and PIC have their own resume routines. + * We just mask them here to make sure the interrupt + * subsystem is completely quiet while we enable x2apic + * and interrupt-remapping. + */ + mask_ioapic_entries(); legacy_pic->mask_all(); } @@ -2152,13 +2141,9 @@ static void lapic_resume(void) apic_write(APIC_ESR, 0); apic_read(APIC_ESR); - if (intr_remapping_enabled) { + if (intr_remapping_enabled) reenable_intr_remapping(x2apic_mode); - legacy_pic->restore_mask(); - restore_IO_APIC_setup(ioapic_entries); - free_ioapic_entries(ioapic_entries); - } -restore: + local_irq_restore(flags); } diff --git a/arch/x86/kernel/apic/apic_flat_64.c b/arch/x86/kernel/apic/apic_flat_64.c index 5652d31fe108..f7a41e4cae47 100644 --- a/arch/x86/kernel/apic/apic_flat_64.c +++ b/arch/x86/kernel/apic/apic_flat_64.c @@ -16,6 +16,7 @@ #include <linux/ctype.h> #include <linux/init.h> #include <linux/hardirq.h> +#include <linux/module.h> #include <asm/smp.h> #include <asm/apic.h> #include <asm/ipi.h> @@ -24,6 +25,12 @@ #include <acpi/acpi_bus.h> #endif +static struct apic apic_physflat; +static struct apic apic_flat; + +struct apic __read_mostly *apic = &apic_flat; +EXPORT_SYMBOL_GPL(apic); + static int flat_acpi_madt_oem_check(char *oem_id, char *oem_table_id) { return 1; @@ -164,7 +171,7 @@ static int flat_phys_pkg_id(int initial_apic_id, int index_msb) return initial_apic_id >> index_msb; } -struct apic apic_flat = { +static struct apic apic_flat = { .name = "flat", .probe = NULL, .acpi_madt_oem_check = flat_acpi_madt_oem_check, @@ -312,10 +319,18 @@ physflat_cpu_mask_to_apicid_and(const struct cpumask *cpumask, return per_cpu(x86_cpu_to_apicid, cpu); } -struct apic apic_physflat = { +static int physflat_probe(void) +{ + if (apic == &apic_physflat || num_possible_cpus() > 8) + return 1; + + return 0; +} + +static struct apic apic_physflat = { .name = "physical flat", - .probe = NULL, + .probe = physflat_probe, .acpi_madt_oem_check = physflat_acpi_madt_oem_check, .apic_id_registered = flat_apic_id_registered, @@ -369,3 +384,8 @@ struct apic apic_physflat = { .wait_icr_idle = native_apic_wait_icr_idle, .safe_wait_icr_idle = native_safe_apic_wait_icr_idle, }; + +/* + * We need to check for physflat first, so this order is important. + */ +apic_drivers(apic_physflat, apic_flat); diff --git a/arch/x86/kernel/apic/apic_noop.c b/arch/x86/kernel/apic/apic_noop.c index f1baa2dc087a..775b82bc655c 100644 --- a/arch/x86/kernel/apic/apic_noop.c +++ b/arch/x86/kernel/apic/apic_noop.c @@ -119,14 +119,6 @@ static void noop_apic_write(u32 reg, u32 v) WARN_ON_ONCE(cpu_has_apic && !disable_apic); } -#ifdef CONFIG_X86_32 -static int noop_x86_32_numa_cpu_node(int cpu) -{ - /* we're always on node 0 */ - return 0; -} -#endif - struct apic apic_noop = { .name = "noop", .probe = noop_probe, @@ -195,6 +187,5 @@ struct apic apic_noop = { #ifdef CONFIG_X86_32 .x86_32_early_logical_apicid = noop_x86_32_early_logical_apicid, - .x86_32_numa_cpu_node = noop_x86_32_numa_cpu_node, #endif }; diff --git a/arch/x86/kernel/apic/bigsmp_32.c b/arch/x86/kernel/apic/bigsmp_32.c index 541a2e431659..efd737e827f4 100644 --- a/arch/x86/kernel/apic/bigsmp_32.c +++ b/arch/x86/kernel/apic/bigsmp_32.c @@ -193,7 +193,7 @@ static int probe_bigsmp(void) return dmi_bigsmp; } -struct apic apic_bigsmp = { +static struct apic apic_bigsmp = { .name = "bigsmp", .probe = probe_bigsmp, @@ -253,5 +253,14 @@ struct apic apic_bigsmp = { .safe_wait_icr_idle = native_safe_apic_wait_icr_idle, .x86_32_early_logical_apicid = bigsmp_early_logical_apicid, - .x86_32_numa_cpu_node = default_x86_32_numa_cpu_node, }; + +struct apic * __init generic_bigsmp_probe(void) +{ + if (probe_bigsmp()) + return &apic_bigsmp; + + return NULL; +} + +apic_driver(apic_bigsmp); diff --git a/arch/x86/kernel/apic/es7000_32.c b/arch/x86/kernel/apic/es7000_32.c index 3e9de4854c5b..9536b3fe43f8 100644 --- a/arch/x86/kernel/apic/es7000_32.c +++ b/arch/x86/kernel/apic/es7000_32.c @@ -510,11 +510,6 @@ static void es7000_setup_apic_routing(void) nr_ioapics, cpumask_bits(es7000_target_cpus())[0]); } -static int es7000_numa_cpu_node(int cpu) -{ - return 0; -} - static int es7000_cpu_present_to_apicid(int mps_cpu) { if (!mps_cpu) @@ -625,7 +620,7 @@ static int es7000_mps_oem_check_cluster(struct mpc_table *mpc, char *oem, } /* We've been warned by a false positive warning.Use __refdata to keep calm. */ -struct apic __refdata apic_es7000_cluster = { +static struct apic __refdata apic_es7000_cluster = { .name = "es7000", .probe = probe_es7000, @@ -688,10 +683,9 @@ struct apic __refdata apic_es7000_cluster = { .safe_wait_icr_idle = native_safe_apic_wait_icr_idle, .x86_32_early_logical_apicid = es7000_early_logical_apicid, - .x86_32_numa_cpu_node = es7000_numa_cpu_node, }; -struct apic __refdata apic_es7000 = { +static struct apic __refdata apic_es7000 = { .name = "es7000", .probe = probe_es7000, @@ -752,5 +746,10 @@ struct apic __refdata apic_es7000 = { .safe_wait_icr_idle = native_safe_apic_wait_icr_idle, .x86_32_early_logical_apicid = es7000_early_logical_apicid, - .x86_32_numa_cpu_node = es7000_numa_cpu_node, }; + +/* + * Need to check for es7000 followed by es7000_cluster, so this order + * in apic_drivers is important. + */ +apic_drivers(apic_es7000, apic_es7000_cluster); diff --git a/arch/x86/kernel/apic/hw_nmi.c b/arch/x86/kernel/apic/hw_nmi.c index 5260fe91bcb6..d5e57db0f7be 100644 --- a/arch/x86/kernel/apic/hw_nmi.c +++ b/arch/x86/kernel/apic/hw_nmi.c @@ -19,9 +19,9 @@ #include <linux/delay.h> #ifdef CONFIG_HARDLOCKUP_DETECTOR -u64 hw_nmi_get_sample_period(void) +u64 hw_nmi_get_sample_period(int watchdog_thresh) { - return (u64)(cpu_khz) * 1000 * 60; + return (u64)(cpu_khz) * 1000 * watchdog_thresh; } #endif diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c index 68df09bba92e..e5293394b548 100644 --- a/arch/x86/kernel/apic/io_apic.c +++ b/arch/x86/kernel/apic/io_apic.c @@ -76,17 +76,40 @@ int sis_apic_bug = -1; static DEFINE_RAW_SPINLOCK(ioapic_lock); static DEFINE_RAW_SPINLOCK(vector_lock); -/* - * # of IRQ routing registers - */ -int nr_ioapic_registers[MAX_IO_APICS]; +static struct ioapic { + /* + * # of IRQ routing registers + */ + int nr_registers; + /* + * Saved state during suspend/resume, or while enabling intr-remap. + */ + struct IO_APIC_route_entry *saved_registers; + /* I/O APIC config */ + struct mpc_ioapic mp_config; + /* IO APIC gsi routing info */ + struct mp_ioapic_gsi gsi_config; + DECLARE_BITMAP(pin_programmed, MP_MAX_IOAPIC_PIN + 1); +} ioapics[MAX_IO_APICS]; -/* I/O APIC entries */ -struct mpc_ioapic mp_ioapics[MAX_IO_APICS]; -int nr_ioapics; +#define mpc_ioapic_ver(id) ioapics[id].mp_config.apicver + +int mpc_ioapic_id(int id) +{ + return ioapics[id].mp_config.apicid; +} -/* IO APIC gsi routing info */ -struct mp_ioapic_gsi mp_gsi_routing[MAX_IO_APICS]; +unsigned int mpc_ioapic_addr(int id) +{ + return ioapics[id].mp_config.apicaddr; +} + +struct mp_ioapic_gsi *mp_ioapic_gsi_routing(int id) +{ + return &ioapics[id].gsi_config; +} + +int nr_ioapics; /* The one past the highest gsi number used */ u32 gsi_top; @@ -128,8 +151,8 @@ static int __init parse_noapic(char *str) } early_param("noapic", parse_noapic); -static int io_apic_setup_irq_pin_once(unsigned int irq, int node, - struct io_apic_irq_attr *attr); +static int io_apic_setup_irq_pin(unsigned int irq, int node, + struct io_apic_irq_attr *attr); /* Will be called in mpparse/acpi/sfi codes for saving IRQ info */ void mp_save_irq(struct mpc_intsrc *m) @@ -179,6 +202,14 @@ int __init arch_early_irq_init(void) io_apic_irqs = ~0UL; } + for (i = 0; i < nr_ioapics; i++) { + ioapics[i].saved_registers = + kzalloc(sizeof(struct IO_APIC_route_entry) * + ioapics[i].nr_registers, GFP_KERNEL); + if (!ioapics[i].saved_registers) + pr_err("IOAPIC %d: suspend/resume impossible!\n", i); + } + cfg = irq_cfgx; count = ARRAY_SIZE(irq_cfgx); node = cpu_to_node(0); @@ -297,7 +328,7 @@ struct io_apic { static __attribute_const__ struct io_apic __iomem *io_apic_base(int idx) { return (void __iomem *) __fix_to_virt(FIX_IO_APIC_BASE_0 + idx) - + (mp_ioapics[idx].apicaddr & ~PAGE_MASK); + + (mpc_ioapic_addr(idx) & ~PAGE_MASK); } static inline void io_apic_eoi(unsigned int apic, unsigned int vector) @@ -573,7 +604,7 @@ static void clear_IO_APIC (void) int apic, pin; for (apic = 0; apic < nr_ioapics; apic++) - for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) + for (pin = 0; pin < ioapics[apic].nr_registers; pin++) clear_IO_APIC_pin(apic, pin); } @@ -615,74 +646,43 @@ static int __init ioapic_pirq_setup(char *str) __setup("pirq=", ioapic_pirq_setup); #endif /* CONFIG_X86_32 */ -struct IO_APIC_route_entry **alloc_ioapic_entries(void) -{ - int apic; - struct IO_APIC_route_entry **ioapic_entries; - - ioapic_entries = kzalloc(sizeof(*ioapic_entries) * nr_ioapics, - GFP_KERNEL); - if (!ioapic_entries) - return 0; - - for (apic = 0; apic < nr_ioapics; apic++) { - ioapic_entries[apic] = - kzalloc(sizeof(struct IO_APIC_route_entry) * - nr_ioapic_registers[apic], GFP_KERNEL); - if (!ioapic_entries[apic]) - goto nomem; - } - - return ioapic_entries; - -nomem: - while (--apic >= 0) - kfree(ioapic_entries[apic]); - kfree(ioapic_entries); - - return 0; -} - /* * Saves all the IO-APIC RTE's */ -int save_IO_APIC_setup(struct IO_APIC_route_entry **ioapic_entries) +int save_ioapic_entries(void) { int apic, pin; - - if (!ioapic_entries) - return -ENOMEM; + int err = 0; for (apic = 0; apic < nr_ioapics; apic++) { - if (!ioapic_entries[apic]) - return -ENOMEM; + if (!ioapics[apic].saved_registers) { + err = -ENOMEM; + continue; + } - for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) - ioapic_entries[apic][pin] = + for (pin = 0; pin < ioapics[apic].nr_registers; pin++) + ioapics[apic].saved_registers[pin] = ioapic_read_entry(apic, pin); } - return 0; + return err; } /* * Mask all IO APIC entries. */ -void mask_IO_APIC_setup(struct IO_APIC_route_entry **ioapic_entries) +void mask_ioapic_entries(void) { int apic, pin; - if (!ioapic_entries) - return; - for (apic = 0; apic < nr_ioapics; apic++) { - if (!ioapic_entries[apic]) - break; + if (!ioapics[apic].saved_registers) + continue; - for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) { + for (pin = 0; pin < ioapics[apic].nr_registers; pin++) { struct IO_APIC_route_entry entry; - entry = ioapic_entries[apic][pin]; + entry = ioapics[apic].saved_registers[pin]; if (!entry.mask) { entry.mask = 1; ioapic_write_entry(apic, pin, entry); @@ -692,36 +692,23 @@ void mask_IO_APIC_setup(struct IO_APIC_route_entry **ioapic_entries) } /* - * Restore IO APIC entries which was saved in ioapic_entries. + * Restore IO APIC entries which was saved in the ioapic structure. */ -int restore_IO_APIC_setup(struct IO_APIC_route_entry **ioapic_entries) +int restore_ioapic_entries(void) { int apic, pin; - if (!ioapic_entries) - return -ENOMEM; - for (apic = 0; apic < nr_ioapics; apic++) { - if (!ioapic_entries[apic]) - return -ENOMEM; + if (!ioapics[apic].saved_registers) + continue; - for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) + for (pin = 0; pin < ioapics[apic].nr_registers; pin++) ioapic_write_entry(apic, pin, - ioapic_entries[apic][pin]); + ioapics[apic].saved_registers[pin]); } return 0; } -void free_ioapic_entries(struct IO_APIC_route_entry **ioapic_entries) -{ - int apic; - - for (apic = 0; apic < nr_ioapics; apic++) - kfree(ioapic_entries[apic]); - - kfree(ioapic_entries); -} - /* * Find the IRQ entry number of a certain pin. */ @@ -731,7 +718,7 @@ static int find_irq_entry(int apic, int pin, int type) for (i = 0; i < mp_irq_entries; i++) if (mp_irqs[i].irqtype == type && - (mp_irqs[i].dstapic == mp_ioapics[apic].apicid || + (mp_irqs[i].dstapic == mpc_ioapic_id(apic) || mp_irqs[i].dstapic == MP_APIC_ALL) && mp_irqs[i].dstirq == pin) return i; @@ -773,7 +760,7 @@ static int __init find_isa_irq_apic(int irq, int type) if (i < mp_irq_entries) { int apic; for(apic = 0; apic < nr_ioapics; apic++) { - if (mp_ioapics[apic].apicid == mp_irqs[i].dstapic) + if (mpc_ioapic_id(apic) == mp_irqs[i].dstapic) return apic; } } @@ -942,6 +929,7 @@ static int pin_2_irq(int idx, int apic, int pin) { int irq; int bus = mp_irqs[idx].srcbus; + struct mp_ioapic_gsi *gsi_cfg = mp_ioapic_gsi_routing(apic); /* * Debugging check, we are in big trouble if this message pops up! @@ -952,7 +940,7 @@ static int pin_2_irq(int idx, int apic, int pin) if (test_bit(bus, mp_bus_not_pci)) { irq = mp_irqs[idx].srcbusirq; } else { - u32 gsi = mp_gsi_routing[apic].gsi_base + pin; + u32 gsi = gsi_cfg->gsi_base + pin; if (gsi >= NR_IRQS_LEGACY) irq = gsi; @@ -1003,7 +991,7 @@ int IO_APIC_get_PCI_irq_vector(int bus, int slot, int pin, int lbus = mp_irqs[i].srcbus; for (apic = 0; apic < nr_ioapics; apic++) - if (mp_ioapics[apic].apicid == mp_irqs[i].dstapic || + if (mpc_ioapic_id(apic) == mp_irqs[i].dstapic || mp_irqs[i].dstapic == MP_APIC_ALL) break; @@ -1222,7 +1210,7 @@ static inline int IO_APIC_irq_trigger(int irq) int apic, idx, pin; for (apic = 0; apic < nr_ioapics; apic++) { - for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) { + for (pin = 0; pin < ioapics[apic].nr_registers; pin++) { idx = find_irq_entry(apic, pin, mp_INT); if ((idx != -1) && (irq == pin_2_irq(idx, apic, pin))) return irq_trigger(idx); @@ -1350,14 +1338,14 @@ static void setup_ioapic_irq(int apic_id, int pin, unsigned int irq, apic_printk(APIC_VERBOSE,KERN_DEBUG "IOAPIC[%d]: Set routing entry (%d-%d -> 0x%x -> " "IRQ %d Mode:%i Active:%i)\n", - apic_id, mp_ioapics[apic_id].apicid, pin, cfg->vector, + apic_id, mpc_ioapic_id(apic_id), pin, cfg->vector, irq, trigger, polarity); - if (setup_ioapic_entry(mp_ioapics[apic_id].apicid, irq, &entry, + if (setup_ioapic_entry(mpc_ioapic_id(apic_id), irq, &entry, dest, trigger, polarity, cfg->vector, pin)) { printk("Failed to setup ioapic entry for ioapic %d, pin %d\n", - mp_ioapics[apic_id].apicid, pin); + mpc_ioapic_id(apic_id), pin); __clear_irq_vector(irq, cfg); return; } @@ -1369,17 +1357,13 @@ static void setup_ioapic_irq(int apic_id, int pin, unsigned int irq, ioapic_write_entry(apic_id, pin, entry); } -static struct { - DECLARE_BITMAP(pin_programmed, MP_MAX_IOAPIC_PIN + 1); -} mp_ioapic_routing[MAX_IO_APICS]; - static bool __init io_apic_pin_not_connected(int idx, int apic_id, int pin) { if (idx != -1) return false; apic_printk(APIC_VERBOSE, KERN_DEBUG " apic %d pin %d not connected\n", - mp_ioapics[apic_id].apicid, pin); + mpc_ioapic_id(apic_id), pin); return true; } @@ -1389,7 +1373,7 @@ static void __init __io_apic_setup_irqs(unsigned int apic_id) struct io_apic_irq_attr attr; unsigned int pin, irq; - for (pin = 0; pin < nr_ioapic_registers[apic_id]; pin++) { + for (pin = 0; pin < ioapics[apic_id].nr_registers; pin++) { idx = find_irq_entry(apic_id, pin, mp_INT); if (io_apic_pin_not_connected(idx, apic_id, pin)) continue; @@ -1511,7 +1495,7 @@ __apicdebuginit(void) print_IO_APIC(void) printk(KERN_DEBUG "number of MP IRQ sources: %d.\n", mp_irq_entries); for (i = 0; i < nr_ioapics; i++) printk(KERN_DEBUG "number of IO-APIC #%d registers: %d.\n", - mp_ioapics[i].apicid, nr_ioapic_registers[i]); + mpc_ioapic_id(i), ioapics[i].nr_registers); /* * We are a bit conservative about what we expect. We have to @@ -1531,7 +1515,7 @@ __apicdebuginit(void) print_IO_APIC(void) raw_spin_unlock_irqrestore(&ioapic_lock, flags); printk("\n"); - printk(KERN_DEBUG "IO APIC #%d......\n", mp_ioapics[apic].apicid); + printk(KERN_DEBUG "IO APIC #%d......\n", mpc_ioapic_id(apic)); printk(KERN_DEBUG ".... register #00: %08X\n", reg_00.raw); printk(KERN_DEBUG "....... : physical APIC id: %02X\n", reg_00.bits.ID); printk(KERN_DEBUG "....... : Delivery Type: %X\n", reg_00.bits.delivery_type); @@ -1825,7 +1809,7 @@ void __init enable_IO_APIC(void) for(apic = 0; apic < nr_ioapics; apic++) { int pin; /* See if any of the pins is in ExtINT mode */ - for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) { + for (pin = 0; pin < ioapics[apic].nr_registers; pin++) { struct IO_APIC_route_entry entry; entry = ioapic_read_entry(apic, pin); @@ -1949,14 +1933,14 @@ void __init setup_ioapic_ids_from_mpc_nocheck(void) reg_00.raw = io_apic_read(apic_id, 0); raw_spin_unlock_irqrestore(&ioapic_lock, flags); - old_id = mp_ioapics[apic_id].apicid; + old_id = mpc_ioapic_id(apic_id); - if (mp_ioapics[apic_id].apicid >= get_physical_broadcast()) { + if (mpc_ioapic_id(apic_id) >= get_physical_broadcast()) { printk(KERN_ERR "BIOS bug, IO-APIC#%d ID is %d in the MPC table!...\n", - apic_id, mp_ioapics[apic_id].apicid); + apic_id, mpc_ioapic_id(apic_id)); printk(KERN_ERR "... fixing up to %d. (tell your hw vendor)\n", reg_00.bits.ID); - mp_ioapics[apic_id].apicid = reg_00.bits.ID; + ioapics[apic_id].mp_config.apicid = reg_00.bits.ID; } /* @@ -1965,9 +1949,9 @@ void __init setup_ioapic_ids_from_mpc_nocheck(void) * 'stuck on smp_invalidate_needed IPI wait' messages. */ if (apic->check_apicid_used(&phys_id_present_map, - mp_ioapics[apic_id].apicid)) { + mpc_ioapic_id(apic_id))) { printk(KERN_ERR "BIOS bug, IO-APIC#%d ID %d is already used!...\n", - apic_id, mp_ioapics[apic_id].apicid); + apic_id, mpc_ioapic_id(apic_id)); for (i = 0; i < get_physical_broadcast(); i++) if (!physid_isset(i, phys_id_present_map)) break; @@ -1976,13 +1960,14 @@ void __init setup_ioapic_ids_from_mpc_nocheck(void) printk(KERN_ERR "... fixing up to %d. (tell your hw vendor)\n", i); physid_set(i, phys_id_present_map); - mp_ioapics[apic_id].apicid = i; + ioapics[apic_id].mp_config.apicid = i; } else { physid_mask_t tmp; - apic->apicid_to_cpu_present(mp_ioapics[apic_id].apicid, &tmp); + apic->apicid_to_cpu_present(mpc_ioapic_id(apic_id), + &tmp); apic_printk(APIC_VERBOSE, "Setting %d in the " "phys_id_present_map\n", - mp_ioapics[apic_id].apicid); + mpc_ioapic_id(apic_id)); physids_or(phys_id_present_map, phys_id_present_map, tmp); } @@ -1990,24 +1975,24 @@ void __init setup_ioapic_ids_from_mpc_nocheck(void) * We need to adjust the IRQ routing table * if the ID changed. */ - if (old_id != mp_ioapics[apic_id].apicid) + if (old_id != mpc_ioapic_id(apic_id)) for (i = 0; i < mp_irq_entries; i++) if (mp_irqs[i].dstapic == old_id) mp_irqs[i].dstapic - = mp_ioapics[apic_id].apicid; + = mpc_ioapic_id(apic_id); /* * Update the ID register according to the right value * from the MPC table if they are different. */ - if (mp_ioapics[apic_id].apicid == reg_00.bits.ID) + if (mpc_ioapic_id(apic_id) == reg_00.bits.ID) continue; apic_printk(APIC_VERBOSE, KERN_INFO "...changing IO-APIC physical APIC ID to %d ...", - mp_ioapics[apic_id].apicid); + mpc_ioapic_id(apic_id)); - reg_00.bits.ID = mp_ioapics[apic_id].apicid; + reg_00.bits.ID = mpc_ioapic_id(apic_id); raw_spin_lock_irqsave(&ioapic_lock, flags); io_apic_write(apic_id, 0, reg_00.raw); raw_spin_unlock_irqrestore(&ioapic_lock, flags); @@ -2018,7 +2003,7 @@ void __init setup_ioapic_ids_from_mpc_nocheck(void) raw_spin_lock_irqsave(&ioapic_lock, flags); reg_00.raw = io_apic_read(apic_id, 0); raw_spin_unlock_irqrestore(&ioapic_lock, flags); - if (reg_00.bits.ID != mp_ioapics[apic_id].apicid) + if (reg_00.bits.ID != mpc_ioapic_id(apic_id)) printk("could not set ID!\n"); else apic_printk(APIC_VERBOSE, " ok.\n"); @@ -2404,7 +2389,7 @@ static void eoi_ioapic_irq(unsigned int irq, struct irq_cfg *cfg) raw_spin_lock_irqsave(&ioapic_lock, flags); for_each_irq_pin(entry, cfg->irq_2_pin) { - if (mp_ioapics[entry->apic].apicver >= 0x20) { + if (mpc_ioapic_ver(entry->apic) >= 0x20) { /* * Intr-remapping uses pin number as the virtual vector * in the RTE. Actual vector is programmed in @@ -2918,49 +2903,19 @@ static int __init io_apic_bug_finalize(void) late_initcall(io_apic_bug_finalize); -static struct IO_APIC_route_entry *ioapic_saved_data[MAX_IO_APICS]; - -static void suspend_ioapic(int ioapic_id) +static void resume_ioapic_id(int ioapic_id) { - struct IO_APIC_route_entry *saved_data = ioapic_saved_data[ioapic_id]; - int i; - - if (!saved_data) - return; - - for (i = 0; i < nr_ioapic_registers[ioapic_id]; i++) - saved_data[i] = ioapic_read_entry(ioapic_id, i); -} - -static int ioapic_suspend(void) -{ - int ioapic_id; - - for (ioapic_id = 0; ioapic_id < nr_ioapics; ioapic_id++) - suspend_ioapic(ioapic_id); - - return 0; -} - -static void resume_ioapic(int ioapic_id) -{ - struct IO_APIC_route_entry *saved_data = ioapic_saved_data[ioapic_id]; unsigned long flags; union IO_APIC_reg_00 reg_00; - int i; - if (!saved_data) - return; raw_spin_lock_irqsave(&ioapic_lock, flags); reg_00.raw = io_apic_read(ioapic_id, 0); - if (reg_00.bits.ID != mp_ioapics[ioapic_id].apicid) { - reg_00.bits.ID = mp_ioapics[ioapic_id].apicid; + if (reg_00.bits.ID != mpc_ioapic_id(ioapic_id)) { + reg_00.bits.ID = mpc_ioapic_id(ioapic_id); io_apic_write(ioapic_id, 0, reg_00.raw); } raw_spin_unlock_irqrestore(&ioapic_lock, flags); - for (i = 0; i < nr_ioapic_registers[ioapic_id]; i++) - ioapic_write_entry(ioapic_id, i, saved_data[i]); } static void ioapic_resume(void) @@ -2968,28 +2923,18 @@ static void ioapic_resume(void) int ioapic_id; for (ioapic_id = nr_ioapics - 1; ioapic_id >= 0; ioapic_id--) - resume_ioapic(ioapic_id); + resume_ioapic_id(ioapic_id); + + restore_ioapic_entries(); } static struct syscore_ops ioapic_syscore_ops = { - .suspend = ioapic_suspend, + .suspend = save_ioapic_entries, .resume = ioapic_resume, }; static int __init ioapic_init_ops(void) { - int i; - - for (i = 0; i < nr_ioapics; i++) { - unsigned int size; - - size = nr_ioapic_registers[i] - * sizeof(struct IO_APIC_route_entry); - ioapic_saved_data[i] = kzalloc(size, GFP_KERNEL); - if (!ioapic_saved_data[i]) - pr_err("IOAPIC %d: suspend/resume impossible!\n", i); - } - register_syscore_ops(&ioapic_syscore_ops); return 0; @@ -3570,7 +3515,7 @@ int arch_setup_ht_irq(unsigned int irq, struct pci_dev *dev) } #endif /* CONFIG_HT_IRQ */ -int +static int io_apic_setup_irq_pin(unsigned int irq, int node, struct io_apic_irq_attr *attr) { struct irq_cfg *cfg = alloc_irq_and_cfg_at(irq, node); @@ -3585,21 +3530,21 @@ io_apic_setup_irq_pin(unsigned int irq, int node, struct io_apic_irq_attr *attr) return ret; } -static int io_apic_setup_irq_pin_once(unsigned int irq, int node, - struct io_apic_irq_attr *attr) +int io_apic_setup_irq_pin_once(unsigned int irq, int node, + struct io_apic_irq_attr *attr) { unsigned int id = attr->ioapic, pin = attr->ioapic_pin; int ret; /* Avoid redundant programming */ - if (test_bit(pin, mp_ioapic_routing[id].pin_programmed)) { + if (test_bit(pin, ioapics[id].pin_programmed)) { pr_debug("Pin %d-%d already programmed\n", - mp_ioapics[id].apicid, pin); + mpc_ioapic_id(id), pin); return 0; } ret = io_apic_setup_irq_pin(irq, node, attr); if (!ret) - set_bit(pin, mp_ioapic_routing[id].pin_programmed); + set_bit(pin, ioapics[id].pin_programmed); return ret; } @@ -3764,8 +3709,7 @@ static u8 __init io_apic_unique_id(u8 id) bitmap_zero(used, 256); for (i = 0; i < nr_ioapics; i++) { - struct mpc_ioapic *ia = &mp_ioapics[i]; - __set_bit(ia->apicid, used); + __set_bit(mpc_ioapic_id(i), used); } if (!test_bit(id, used)) return id; @@ -3825,7 +3769,7 @@ void __init setup_ioapic_dest(void) return; for (ioapic = 0; ioapic < nr_ioapics; ioapic++) - for (pin = 0; pin < nr_ioapic_registers[ioapic]; pin++) { + for (pin = 0; pin < ioapics[ioapic].nr_registers; pin++) { irq_entry = find_irq_entry(ioapic, pin, mp_INT); if (irq_entry == -1) continue; @@ -3896,7 +3840,7 @@ void __init ioapic_and_gsi_init(void) ioapic_res = ioapic_setup_resources(nr_ioapics); for (i = 0; i < nr_ioapics; i++) { if (smp_found_config) { - ioapic_phys = mp_ioapics[i].apicaddr; + ioapic_phys = mpc_ioapic_addr(i); #ifdef CONFIG_X86_32 if (!ioapic_phys) { printk(KERN_ERR @@ -3956,8 +3900,9 @@ int mp_find_ioapic(u32 gsi) /* Find the IOAPIC that manages this GSI. */ for (i = 0; i < nr_ioapics; i++) { - if ((gsi >= mp_gsi_routing[i].gsi_base) - && (gsi <= mp_gsi_routing[i].gsi_end)) + struct mp_ioapic_gsi *gsi_cfg = mp_ioapic_gsi_routing(i); + if ((gsi >= gsi_cfg->gsi_base) + && (gsi <= gsi_cfg->gsi_end)) return i; } @@ -3967,12 +3912,16 @@ int mp_find_ioapic(u32 gsi) int mp_find_ioapic_pin(int ioapic, u32 gsi) { + struct mp_ioapic_gsi *gsi_cfg; + if (WARN_ON(ioapic == -1)) return -1; - if (WARN_ON(gsi > mp_gsi_routing[ioapic].gsi_end)) + + gsi_cfg = mp_ioapic_gsi_routing(ioapic); + if (WARN_ON(gsi > gsi_cfg->gsi_end)) return -1; - return gsi - mp_gsi_routing[ioapic].gsi_base; + return gsi - gsi_cfg->gsi_base; } static __init int bad_ioapic(unsigned long address) @@ -3994,40 +3943,42 @@ void __init mp_register_ioapic(int id, u32 address, u32 gsi_base) { int idx = 0; int entries; + struct mp_ioapic_gsi *gsi_cfg; if (bad_ioapic(address)) return; idx = nr_ioapics; - mp_ioapics[idx].type = MP_IOAPIC; - mp_ioapics[idx].flags = MPC_APIC_USABLE; - mp_ioapics[idx].apicaddr = address; + ioapics[idx].mp_config.type = MP_IOAPIC; + ioapics[idx].mp_config.flags = MPC_APIC_USABLE; + ioapics[idx].mp_config.apicaddr = address; set_fixmap_nocache(FIX_IO_APIC_BASE_0 + idx, address); - mp_ioapics[idx].apicid = io_apic_unique_id(id); - mp_ioapics[idx].apicver = io_apic_get_version(idx); + ioapics[idx].mp_config.apicid = io_apic_unique_id(id); + ioapics[idx].mp_config.apicver = io_apic_get_version(idx); /* * Build basic GSI lookup table to facilitate gsi->io_apic lookups * and to prevent reprogramming of IOAPIC pins (PCI GSIs). */ entries = io_apic_get_redir_entries(idx); - mp_gsi_routing[idx].gsi_base = gsi_base; - mp_gsi_routing[idx].gsi_end = gsi_base + entries - 1; + gsi_cfg = mp_ioapic_gsi_routing(idx); + gsi_cfg->gsi_base = gsi_base; + gsi_cfg->gsi_end = gsi_base + entries - 1; /* * The number of IO-APIC IRQ registers (== #pins): */ - nr_ioapic_registers[idx] = entries; + ioapics[idx].nr_registers = entries; - if (mp_gsi_routing[idx].gsi_end >= gsi_top) - gsi_top = mp_gsi_routing[idx].gsi_end + 1; + if (gsi_cfg->gsi_end >= gsi_top) + gsi_top = gsi_cfg->gsi_end + 1; printk(KERN_INFO "IOAPIC[%d]: apic_id %d, version %d, address 0x%x, " - "GSI %d-%d\n", idx, mp_ioapics[idx].apicid, - mp_ioapics[idx].apicver, mp_ioapics[idx].apicaddr, - mp_gsi_routing[idx].gsi_base, mp_gsi_routing[idx].gsi_end); + "GSI %d-%d\n", idx, mpc_ioapic_id(idx), + mpc_ioapic_ver(idx), mpc_ioapic_addr(idx), + gsi_cfg->gsi_base, gsi_cfg->gsi_end); nr_ioapics++; } diff --git a/arch/x86/kernel/apic/numaq_32.c b/arch/x86/kernel/apic/numaq_32.c index 6273eee5134b..c4a61ca1349a 100644 --- a/arch/x86/kernel/apic/numaq_32.c +++ b/arch/x86/kernel/apic/numaq_32.c @@ -48,8 +48,6 @@ #include <asm/e820.h> #include <asm/ipi.h> -#define MB_TO_PAGES(addr) ((addr) << (20 - PAGE_SHIFT)) - int found_numaq; /* @@ -79,31 +77,20 @@ int quad_local_to_mp_bus_id[NR_CPUS/4][4]; static inline void numaq_register_node(int node, struct sys_cfg_data *scd) { struct eachquadmem *eq = scd->eq + node; + u64 start = (u64)(eq->hi_shrd_mem_start - eq->priv_mem_size) << 20; + u64 end = (u64)(eq->hi_shrd_mem_start + eq->hi_shrd_mem_size) << 20; + int ret; - node_set_online(node); - - /* Convert to pages */ - node_start_pfn[node] = - MB_TO_PAGES(eq->hi_shrd_mem_start - eq->priv_mem_size); - - node_end_pfn[node] = - MB_TO_PAGES(eq->hi_shrd_mem_start + eq->hi_shrd_mem_size); - - memblock_x86_register_active_regions(node, node_start_pfn[node], - node_end_pfn[node]); - - memory_present(node, node_start_pfn[node], node_end_pfn[node]); - - node_remap_size[node] = node_memmap_size_bytes(node, - node_start_pfn[node], - node_end_pfn[node]); + node_set(node, numa_nodes_parsed); + ret = numa_add_memblk(node, start, end); + BUG_ON(ret < 0); } /* * Function: smp_dump_qct() * * Description: gets memory layout from the quad config table. This - * function also updates node_online_map with the nodes (quads) present. + * function also updates numa_nodes_parsed with the nodes (quads) present. */ static void __init smp_dump_qct(void) { @@ -112,7 +99,6 @@ static void __init smp_dump_qct(void) scd = (void *)__va(SYS_CFG_DATA_PRIV_ADDR); - nodes_clear(node_online_map); for_each_node(node) { if (scd->quads_present31_0 & (1 << node)) numaq_register_node(node, scd); @@ -282,14 +268,14 @@ static __init void early_check_numaq(void) } } -int __init get_memcfg_numaq(void) +int __init numaq_numa_init(void) { early_check_numaq(); if (!found_numaq) - return 0; + return -ENOENT; smp_dump_qct(); - return 1; + return 0; } #define NUMAQ_APIC_DFR_VALUE (APIC_DFR_CLUSTER) @@ -486,8 +472,8 @@ static void numaq_setup_portio_remap(void) (u_long) xquad_portio, (u_long) num_quads*XQUAD_PORTIO_QUAD); } -/* Use __refdata to keep false positive warning calm. */ -struct apic __refdata apic_numaq = { +/* Use __refdata to keep false positive warning calm. */ +static struct apic __refdata apic_numaq = { .name = "NUMAQ", .probe = probe_numaq, @@ -551,3 +537,5 @@ struct apic __refdata apic_numaq = { .x86_32_early_logical_apicid = noop_x86_32_early_logical_apicid, .x86_32_numa_cpu_node = numaq_numa_cpu_node, }; + +apic_driver(apic_numaq); diff --git a/arch/x86/kernel/apic/probe_32.c b/arch/x86/kernel/apic/probe_32.c index fc84c7b61108..b5254ad044ab 100644 --- a/arch/x86/kernel/apic/probe_32.c +++ b/arch/x86/kernel/apic/probe_32.c @@ -52,31 +52,6 @@ static int __init print_ipi_mode(void) } late_initcall(print_ipi_mode); -void __init default_setup_apic_routing(void) -{ - int version = apic_version[boot_cpu_physical_apicid]; - - if (num_possible_cpus() > 8) { - switch (boot_cpu_data.x86_vendor) { - case X86_VENDOR_INTEL: - if (!APIC_XAPIC(version)) { - def_to_bigsmp = 0; - break; - } - /* If P4 and above fall through */ - case X86_VENDOR_AMD: - def_to_bigsmp = 1; - } - } - -#ifdef CONFIG_X86_BIGSMP - generic_bigsmp_probe(); -#endif - - if (apic->setup_apic_routing) - apic->setup_apic_routing(); -} - static int default_x86_32_early_logical_apicid(int cpu) { return 1 << cpu; @@ -112,7 +87,7 @@ static int probe_default(void) return 1; } -struct apic apic_default = { +static struct apic apic_default = { .name = "default", .probe = probe_default, @@ -172,47 +147,24 @@ struct apic apic_default = { .safe_wait_icr_idle = native_safe_apic_wait_icr_idle, .x86_32_early_logical_apicid = default_x86_32_early_logical_apicid, - .x86_32_numa_cpu_node = default_x86_32_numa_cpu_node, }; -extern struct apic apic_numaq; -extern struct apic apic_summit; -extern struct apic apic_bigsmp; -extern struct apic apic_es7000; -extern struct apic apic_es7000_cluster; +apic_driver(apic_default); struct apic *apic = &apic_default; EXPORT_SYMBOL_GPL(apic); -static struct apic *apic_probe[] __initdata = { -#ifdef CONFIG_X86_NUMAQ - &apic_numaq, -#endif -#ifdef CONFIG_X86_SUMMIT - &apic_summit, -#endif -#ifdef CONFIG_X86_BIGSMP - &apic_bigsmp, -#endif -#ifdef CONFIG_X86_ES7000 - &apic_es7000, - &apic_es7000_cluster, -#endif - &apic_default, /* must be last */ - NULL, -}; - static int cmdline_apic __initdata; static int __init parse_apic(char *arg) { - int i; + struct apic **drv; if (!arg) return -EINVAL; - for (i = 0; apic_probe[i]; i++) { - if (!strcmp(apic_probe[i]->name, arg)) { - apic = apic_probe[i]; + for (drv = __apicdrivers; drv < __apicdrivers_end; drv++) { + if (!strcmp((*drv)->name, arg)) { + apic = *drv; cmdline_apic = 1; return 0; } @@ -223,38 +175,58 @@ static int __init parse_apic(char *arg) } early_param("apic", parse_apic); -void __init generic_bigsmp_probe(void) +void __init default_setup_apic_routing(void) { + int version = apic_version[boot_cpu_physical_apicid]; + + if (num_possible_cpus() > 8) { + switch (boot_cpu_data.x86_vendor) { + case X86_VENDOR_INTEL: + if (!APIC_XAPIC(version)) { + def_to_bigsmp = 0; + break; + } + /* If P4 and above fall through */ + case X86_VENDOR_AMD: + def_to_bigsmp = 1; + } + } + #ifdef CONFIG_X86_BIGSMP /* - * This routine is used to switch to bigsmp mode when + * This is used to switch to bigsmp mode when * - There is no apic= option specified by the user * - generic_apic_probe() has chosen apic_default as the sub_arch * - we find more than 8 CPUs in acpi LAPIC listing with xAPIC support */ if (!cmdline_apic && apic == &apic_default) { - if (apic_bigsmp.probe()) { - apic = &apic_bigsmp; + struct apic *bigsmp = generic_bigsmp_probe(); + if (bigsmp) { + apic = bigsmp; printk(KERN_INFO "Overriding APIC driver with %s\n", apic->name); } } #endif + + if (apic->setup_apic_routing) + apic->setup_apic_routing(); } void __init generic_apic_probe(void) { if (!cmdline_apic) { - int i; - for (i = 0; apic_probe[i]; i++) { - if (apic_probe[i]->probe()) { - apic = apic_probe[i]; + struct apic **drv; + + for (drv = __apicdrivers; drv < __apicdrivers_end; drv++) { + if ((*drv)->probe()) { + apic = *drv; break; } } /* Not visible without early console */ - if (!apic_probe[i]) + if (drv == __apicdrivers_end) panic("Didn't find an APIC driver"); } printk(KERN_INFO "Using APIC driver %s\n", apic->name); @@ -265,16 +237,16 @@ void __init generic_apic_probe(void) int __init generic_mps_oem_check(struct mpc_table *mpc, char *oem, char *productid) { - int i; + struct apic **drv; - for (i = 0; apic_probe[i]; ++i) { - if (!apic_probe[i]->mps_oem_check) + for (drv = __apicdrivers; drv < __apicdrivers_end; drv++) { + if (!((*drv)->mps_oem_check)) continue; - if (!apic_probe[i]->mps_oem_check(mpc, oem, productid)) + if (!(*drv)->mps_oem_check(mpc, oem, productid)) continue; if (!cmdline_apic) { - apic = apic_probe[i]; + apic = *drv; printk(KERN_INFO "Switched to APIC driver `%s'.\n", apic->name); } @@ -285,16 +257,16 @@ generic_mps_oem_check(struct mpc_table *mpc, char *oem, char *productid) int __init default_acpi_madt_oem_check(char *oem_id, char *oem_table_id) { - int i; + struct apic **drv; - for (i = 0; apic_probe[i]; ++i) { - if (!apic_probe[i]->acpi_madt_oem_check) + for (drv = __apicdrivers; drv < __apicdrivers_end; drv++) { + if (!(*drv)->acpi_madt_oem_check) continue; - if (!apic_probe[i]->acpi_madt_oem_check(oem_id, oem_table_id)) + if (!(*drv)->acpi_madt_oem_check(oem_id, oem_table_id)) continue; if (!cmdline_apic) { - apic = apic_probe[i]; + apic = *drv; printk(KERN_INFO "Switched to APIC driver `%s'.\n", apic->name); } diff --git a/arch/x86/kernel/apic/probe_64.c b/arch/x86/kernel/apic/probe_64.c index d8c4a6feb286..3fe986698929 100644 --- a/arch/x86/kernel/apic/probe_64.c +++ b/arch/x86/kernel/apic/probe_64.c @@ -23,27 +23,6 @@ #include <asm/ipi.h> #include <asm/setup.h> -extern struct apic apic_flat; -extern struct apic apic_physflat; -extern struct apic apic_x2xpic_uv_x; -extern struct apic apic_x2apic_phys; -extern struct apic apic_x2apic_cluster; - -struct apic __read_mostly *apic = &apic_flat; -EXPORT_SYMBOL_GPL(apic); - -static struct apic *apic_probe[] __initdata = { -#ifdef CONFIG_X86_UV - &apic_x2apic_uv_x, -#endif -#ifdef CONFIG_X86_X2APIC - &apic_x2apic_phys, - &apic_x2apic_cluster, -#endif - &apic_physflat, - NULL, -}; - static int apicid_phys_pkg_id(int initial_apic_id, int index_msb) { return hard_smp_processor_id() >> index_msb; @@ -54,26 +33,20 @@ static int apicid_phys_pkg_id(int initial_apic_id, int index_msb) */ void __init default_setup_apic_routing(void) { + struct apic **drv; enable_IR_x2apic(); -#ifdef CONFIG_X86_X2APIC - if (x2apic_mode -#ifdef CONFIG_X86_UV - && apic != &apic_x2apic_uv_x -#endif - ) { - if (x2apic_phys) - apic = &apic_x2apic_phys; - else - apic = &apic_x2apic_cluster; + for (drv = __apicdrivers; drv < __apicdrivers_end; drv++) { + if ((*drv)->probe && (*drv)->probe()) { + if (apic != *drv) { + apic = *drv; + pr_info("Switched APIC routing to %s.\n", + apic->name); + } + break; + } } -#endif - - if (apic == &apic_flat && num_possible_cpus() > 8) - apic = &apic_physflat; - - printk(KERN_INFO "Setting APIC routing to %s\n", apic->name); if (is_vsmp_box()) { /* need to update phys_pkg_id */ @@ -90,13 +63,15 @@ void apic_send_IPI_self(int vector) int __init default_acpi_madt_oem_check(char *oem_id, char *oem_table_id) { - int i; + struct apic **drv; - for (i = 0; apic_probe[i]; ++i) { - if (apic_probe[i]->acpi_madt_oem_check(oem_id, oem_table_id)) { - apic = apic_probe[i]; - printk(KERN_INFO "Setting APIC routing to %s.\n", - apic->name); + for (drv = __apicdrivers; drv < __apicdrivers_end; drv++) { + if ((*drv)->acpi_madt_oem_check(oem_id, oem_table_id)) { + if (apic != *drv) { + apic = *drv; + pr_info("Setting APIC routing to %s.\n", + apic->name); + } return 1; } } diff --git a/arch/x86/kernel/apic/summit_32.c b/arch/x86/kernel/apic/summit_32.c index e4b8059b414a..19114423c58c 100644 --- a/arch/x86/kernel/apic/summit_32.c +++ b/arch/x86/kernel/apic/summit_32.c @@ -491,7 +491,7 @@ void setup_summit(void) } #endif -struct apic apic_summit = { +static struct apic apic_summit = { .name = "summit", .probe = probe_summit, @@ -551,5 +551,6 @@ struct apic apic_summit = { .safe_wait_icr_idle = native_safe_apic_wait_icr_idle, .x86_32_early_logical_apicid = summit_early_logical_apicid, - .x86_32_numa_cpu_node = default_x86_32_numa_cpu_node, }; + +apic_driver(apic_summit); diff --git a/arch/x86/kernel/apic/x2apic_cluster.c b/arch/x86/kernel/apic/x2apic_cluster.c index 90949bbd566d..500795875827 100644 --- a/arch/x86/kernel/apic/x2apic_cluster.c +++ b/arch/x86/kernel/apic/x2apic_cluster.c @@ -5,118 +5,95 @@ #include <linux/ctype.h> #include <linux/init.h> #include <linux/dmar.h> +#include <linux/cpu.h> #include <asm/smp.h> -#include <asm/apic.h> -#include <asm/ipi.h> +#include <asm/x2apic.h> static DEFINE_PER_CPU(u32, x86_cpu_to_logical_apicid); +static DEFINE_PER_CPU(cpumask_var_t, cpus_in_cluster); +static DEFINE_PER_CPU(cpumask_var_t, ipi_mask); static int x2apic_acpi_madt_oem_check(char *oem_id, char *oem_table_id) { return x2apic_enabled(); } -/* - * 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) +static inline u32 x2apic_cluster(int cpu) { - return cpu_online_mask; -} - -/* - * for now each logical cpu is in its own vector allocation domain. - */ -static void x2apic_vector_allocation_domain(int cpu, struct cpumask *retmask) -{ - cpumask_clear(retmask); - cpumask_set_cpu(cpu, retmask); + return per_cpu(x86_cpu_to_logical_apicid, cpu) >> 16; } static void - __x2apic_send_IPI_dest(unsigned int apicid, int vector, unsigned int dest) +__x2apic_send_IPI_mask(const struct cpumask *mask, int vector, int apic_dest) { - unsigned long cfg; + struct cpumask *cpus_in_cluster_ptr; + struct cpumask *ipi_mask_ptr; + unsigned int cpu, this_cpu; + unsigned long flags; + u32 dest; + + x2apic_wrmsr_fence(); + + local_irq_save(flags); - cfg = __prepare_ICR(0, vector, dest); + this_cpu = smp_processor_id(); /* - * send the IPI. + * We are to modify mask, so we need an own copy + * and be sure it's manipulated with irq off. */ - native_x2apic_icr_write(cfg, apicid); -} + ipi_mask_ptr = __raw_get_cpu_var(ipi_mask); + cpumask_copy(ipi_mask_ptr, mask); -/* - * for now, we send the IPI's one by one in the cpumask. - * TBD: Based on the cpu mask, we can send the IPI's to the cluster group - * at once. We have 16 cpu's in a cluster. This will minimize IPI register - * writes. - */ -static void x2apic_send_IPI_mask(const struct cpumask *mask, int vector) -{ - unsigned long query_cpu; - unsigned long flags; + /* + * The idea is to send one IPI per cluster. + */ + for_each_cpu(cpu, ipi_mask_ptr) { + unsigned long i; - x2apic_wrmsr_fence(); + cpus_in_cluster_ptr = per_cpu(cpus_in_cluster, cpu); + dest = 0; - local_irq_save(flags); - for_each_cpu(query_cpu, mask) { - __x2apic_send_IPI_dest( - per_cpu(x86_cpu_to_logical_apicid, query_cpu), - vector, apic->dest_logical); + /* Collect cpus in cluster. */ + for_each_cpu_and(i, ipi_mask_ptr, cpus_in_cluster_ptr) { + if (apic_dest == APIC_DEST_ALLINC || i != this_cpu) + dest |= per_cpu(x86_cpu_to_logical_apicid, i); + } + + if (!dest) + continue; + + __x2apic_send_IPI_dest(dest, vector, apic->dest_logical); + /* + * Cluster sibling cpus should be discared now so + * we would not send IPI them second time. + */ + cpumask_andnot(ipi_mask_ptr, ipi_mask_ptr, cpus_in_cluster_ptr); } + local_irq_restore(flags); } +static void x2apic_send_IPI_mask(const struct cpumask *mask, int vector) +{ + __x2apic_send_IPI_mask(mask, vector, APIC_DEST_ALLINC); +} + static void x2apic_send_IPI_mask_allbutself(const struct cpumask *mask, int vector) { - unsigned long this_cpu = smp_processor_id(); - unsigned long query_cpu; - unsigned long flags; - - x2apic_wrmsr_fence(); - - local_irq_save(flags); - for_each_cpu(query_cpu, mask) { - if (query_cpu == this_cpu) - continue; - __x2apic_send_IPI_dest( - per_cpu(x86_cpu_to_logical_apicid, query_cpu), - vector, apic->dest_logical); - } - local_irq_restore(flags); + __x2apic_send_IPI_mask(mask, vector, APIC_DEST_ALLBUT); } static void x2apic_send_IPI_allbutself(int vector) { - unsigned long this_cpu = smp_processor_id(); - unsigned long query_cpu; - unsigned long flags; - - x2apic_wrmsr_fence(); - - local_irq_save(flags); - for_each_online_cpu(query_cpu) { - if (query_cpu == this_cpu) - continue; - __x2apic_send_IPI_dest( - per_cpu(x86_cpu_to_logical_apicid, query_cpu), - vector, apic->dest_logical); - } - local_irq_restore(flags); + __x2apic_send_IPI_mask(cpu_online_mask, vector, APIC_DEST_ALLBUT); } static void x2apic_send_IPI_all(int vector) { - x2apic_send_IPI_mask(cpu_online_mask, vector); -} - -static int x2apic_apic_id_registered(void) -{ - return 1; + __x2apic_send_IPI_mask(cpu_online_mask, vector, APIC_DEST_ALLINC); } static unsigned int x2apic_cpu_mask_to_apicid(const struct cpumask *cpumask) @@ -151,43 +128,90 @@ x2apic_cpu_mask_to_apicid_and(const struct cpumask *cpumask, return per_cpu(x86_cpu_to_logical_apicid, cpu); } -static unsigned int x2apic_cluster_phys_get_apic_id(unsigned long x) +static void init_x2apic_ldr(void) { - unsigned int id; + unsigned int this_cpu = smp_processor_id(); + unsigned int cpu; - id = x; - return id; + per_cpu(x86_cpu_to_logical_apicid, this_cpu) = apic_read(APIC_LDR); + + __cpu_set(this_cpu, per_cpu(cpus_in_cluster, this_cpu)); + for_each_online_cpu(cpu) { + if (x2apic_cluster(this_cpu) != x2apic_cluster(cpu)) + continue; + __cpu_set(this_cpu, per_cpu(cpus_in_cluster, cpu)); + __cpu_set(cpu, per_cpu(cpus_in_cluster, this_cpu)); + } } -static unsigned long set_apic_id(unsigned int id) + /* + * At CPU state changes, update the x2apic cluster sibling info. + */ +static int __cpuinit +update_clusterinfo(struct notifier_block *nfb, unsigned long action, void *hcpu) { - unsigned long x; + unsigned int this_cpu = (unsigned long)hcpu; + unsigned int cpu; + int err = 0; + + switch (action) { + case CPU_UP_PREPARE: + if (!zalloc_cpumask_var(&per_cpu(cpus_in_cluster, this_cpu), + GFP_KERNEL)) { + err = -ENOMEM; + } else if (!zalloc_cpumask_var(&per_cpu(ipi_mask, this_cpu), + GFP_KERNEL)) { + free_cpumask_var(per_cpu(cpus_in_cluster, this_cpu)); + err = -ENOMEM; + } + break; + case CPU_UP_CANCELED: + case CPU_UP_CANCELED_FROZEN: + case CPU_DEAD: + for_each_online_cpu(cpu) { + if (x2apic_cluster(this_cpu) != x2apic_cluster(cpu)) + continue; + __cpu_clear(this_cpu, per_cpu(cpus_in_cluster, cpu)); + __cpu_clear(cpu, per_cpu(cpus_in_cluster, this_cpu)); + } + free_cpumask_var(per_cpu(cpus_in_cluster, this_cpu)); + free_cpumask_var(per_cpu(ipi_mask, this_cpu)); + break; + } - x = id; - return x; + return notifier_from_errno(err); } -static int x2apic_cluster_phys_pkg_id(int initial_apicid, int index_msb) -{ - return initial_apicid >> index_msb; -} +static struct notifier_block __refdata x2apic_cpu_notifier = { + .notifier_call = update_clusterinfo, +}; -static void x2apic_send_IPI_self(int vector) +static int x2apic_init_cpu_notifier(void) { - apic_write(APIC_SELF_IPI, vector); + int cpu = smp_processor_id(); + + zalloc_cpumask_var(&per_cpu(cpus_in_cluster, cpu), GFP_KERNEL); + zalloc_cpumask_var(&per_cpu(ipi_mask, cpu), GFP_KERNEL); + + BUG_ON(!per_cpu(cpus_in_cluster, cpu) || !per_cpu(ipi_mask, cpu)); + + __cpu_set(cpu, per_cpu(cpus_in_cluster, cpu)); + register_hotcpu_notifier(&x2apic_cpu_notifier); + return 1; } -static void init_x2apic_ldr(void) +static int x2apic_cluster_probe(void) { - int cpu = smp_processor_id(); - - per_cpu(x86_cpu_to_logical_apicid, cpu) = apic_read(APIC_LDR); + if (x2apic_mode) + return x2apic_init_cpu_notifier(); + else + return 0; } -struct apic apic_x2apic_cluster = { +static struct apic apic_x2apic_cluster = { .name = "cluster x2apic", - .probe = NULL, + .probe = x2apic_cluster_probe, .acpi_madt_oem_check = x2apic_acpi_madt_oem_check, .apic_id_registered = x2apic_apic_id_registered, @@ -211,11 +235,11 @@ struct apic apic_x2apic_cluster = { .setup_portio_remap = NULL, .check_phys_apicid_present = default_check_phys_apicid_present, .enable_apic_mode = NULL, - .phys_pkg_id = x2apic_cluster_phys_pkg_id, + .phys_pkg_id = x2apic_phys_pkg_id, .mps_oem_check = NULL, - .get_apic_id = x2apic_cluster_phys_get_apic_id, - .set_apic_id = set_apic_id, + .get_apic_id = x2apic_get_apic_id, + .set_apic_id = x2apic_set_apic_id, .apic_id_mask = 0xFFFFFFFFu, .cpu_mask_to_apicid = x2apic_cpu_mask_to_apicid, @@ -240,3 +264,5 @@ struct apic apic_x2apic_cluster = { .wait_icr_idle = native_x2apic_wait_icr_idle, .safe_wait_icr_idle = native_safe_x2apic_wait_icr_idle, }; + +apic_driver(apic_x2apic_cluster); diff --git a/arch/x86/kernel/apic/x2apic_phys.c b/arch/x86/kernel/apic/x2apic_phys.c index c7e6d6645bf4..f5373dfde21e 100644 --- a/arch/x86/kernel/apic/x2apic_phys.c +++ b/arch/x86/kernel/apic/x2apic_phys.c @@ -7,11 +7,12 @@ #include <linux/dmar.h> #include <asm/smp.h> -#include <asm/apic.h> -#include <asm/ipi.h> +#include <asm/x2apic.h> int x2apic_phys; +static struct apic apic_x2apic_phys; + static int set_x2apic_phys_mode(char *arg) { x2apic_phys = 1; @@ -27,94 +28,46 @@ static int x2apic_acpi_madt_oem_check(char *oem_id, char *oem_table_id) return 0; } -/* - * 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 cpu_online_mask; -} - -static void x2apic_vector_allocation_domain(int cpu, struct cpumask *retmask) -{ - cpumask_clear(retmask); - cpumask_set_cpu(cpu, retmask); -} - -static void __x2apic_send_IPI_dest(unsigned int apicid, int vector, - unsigned int dest) -{ - unsigned long cfg; - - cfg = __prepare_ICR(0, vector, dest); - - /* - * send the IPI. - */ - native_x2apic_icr_write(cfg, apicid); -} - -static void x2apic_send_IPI_mask(const struct cpumask *mask, int vector) +static void +__x2apic_send_IPI_mask(const struct cpumask *mask, int vector, int apic_dest) { unsigned long query_cpu; + unsigned long this_cpu; unsigned long flags; x2apic_wrmsr_fence(); local_irq_save(flags); + + this_cpu = smp_processor_id(); for_each_cpu(query_cpu, mask) { + if (apic_dest == APIC_DEST_ALLBUT && this_cpu == query_cpu) + continue; __x2apic_send_IPI_dest(per_cpu(x86_cpu_to_apicid, query_cpu), vector, APIC_DEST_PHYSICAL); } local_irq_restore(flags); } +static void x2apic_send_IPI_mask(const struct cpumask *mask, int vector) +{ + __x2apic_send_IPI_mask(mask, vector, APIC_DEST_ALLINC); +} + static void x2apic_send_IPI_mask_allbutself(const struct cpumask *mask, int vector) { - unsigned long this_cpu = smp_processor_id(); - unsigned long query_cpu; - unsigned long flags; - - x2apic_wrmsr_fence(); - - local_irq_save(flags); - for_each_cpu(query_cpu, mask) { - if (query_cpu != this_cpu) - __x2apic_send_IPI_dest( - per_cpu(x86_cpu_to_apicid, query_cpu), - vector, APIC_DEST_PHYSICAL); - } - local_irq_restore(flags); + __x2apic_send_IPI_mask(mask, vector, APIC_DEST_ALLBUT); } static void x2apic_send_IPI_allbutself(int vector) { - unsigned long this_cpu = smp_processor_id(); - unsigned long query_cpu; - unsigned long flags; - - x2apic_wrmsr_fence(); - - local_irq_save(flags); - for_each_online_cpu(query_cpu) { - if (query_cpu == this_cpu) - continue; - __x2apic_send_IPI_dest(per_cpu(x86_cpu_to_apicid, query_cpu), - vector, APIC_DEST_PHYSICAL); - } - local_irq_restore(flags); + __x2apic_send_IPI_mask(cpu_online_mask, vector, APIC_DEST_ALLBUT); } static void x2apic_send_IPI_all(int vector) { - x2apic_send_IPI_mask(cpu_online_mask, vector); -} - -static int x2apic_apic_id_registered(void) -{ - return 1; + __x2apic_send_IPI_mask(cpu_online_mask, vector, APIC_DEST_ALLINC); } static unsigned int x2apic_cpu_mask_to_apicid(const struct cpumask *cpumask) @@ -149,34 +102,22 @@ x2apic_cpu_mask_to_apicid_and(const struct cpumask *cpumask, return per_cpu(x86_cpu_to_apicid, cpu); } -static unsigned int x2apic_phys_get_apic_id(unsigned long x) -{ - return x; -} - -static unsigned long set_apic_id(unsigned int id) -{ - return id; -} - -static int x2apic_phys_pkg_id(int initial_apicid, int index_msb) +static void init_x2apic_ldr(void) { - return initial_apicid >> index_msb; } -static void x2apic_send_IPI_self(int vector) +static int x2apic_phys_probe(void) { - apic_write(APIC_SELF_IPI, vector); -} + if (x2apic_mode && x2apic_phys) + return 1; -static void init_x2apic_ldr(void) -{ + return apic == &apic_x2apic_phys; } -struct apic apic_x2apic_phys = { +static struct apic apic_x2apic_phys = { .name = "physical x2apic", - .probe = NULL, + .probe = x2apic_phys_probe, .acpi_madt_oem_check = x2apic_acpi_madt_oem_check, .apic_id_registered = x2apic_apic_id_registered, @@ -203,8 +144,8 @@ struct apic apic_x2apic_phys = { .phys_pkg_id = x2apic_phys_pkg_id, .mps_oem_check = NULL, - .get_apic_id = x2apic_phys_get_apic_id, - .set_apic_id = set_apic_id, + .get_apic_id = x2apic_get_apic_id, + .set_apic_id = x2apic_set_apic_id, .apic_id_mask = 0xFFFFFFFFu, .cpu_mask_to_apicid = x2apic_cpu_mask_to_apicid, @@ -229,3 +170,5 @@ struct apic apic_x2apic_phys = { .wait_icr_idle = native_x2apic_wait_icr_idle, .safe_wait_icr_idle = native_safe_x2apic_wait_icr_idle, }; + +apic_driver(apic_x2apic_phys); diff --git a/arch/x86/kernel/apic/x2apic_uv_x.c b/arch/x86/kernel/apic/x2apic_uv_x.c index 33b10a0fc095..f450b683dfcf 100644 --- a/arch/x86/kernel/apic/x2apic_uv_x.c +++ b/arch/x86/kernel/apic/x2apic_uv_x.c @@ -37,6 +37,13 @@ #include <asm/smp.h> #include <asm/x86_init.h> #include <asm/emergency-restart.h> +#include <asm/nmi.h> + +/* BMC sets a bit this MMR non-zero before sending an NMI */ +#define UVH_NMI_MMR UVH_SCRATCH5 +#define UVH_NMI_MMR_CLEAR (UVH_NMI_MMR + 8) +#define UV_NMI_PENDING_MASK (1UL << 63) +DEFINE_PER_CPU(unsigned long, cpu_last_nmi_count); DEFINE_PER_CPU(int, x2apic_extra_bits); @@ -51,6 +58,8 @@ unsigned int uv_apicid_hibits; EXPORT_SYMBOL_GPL(uv_apicid_hibits); static DEFINE_SPINLOCK(uv_nmi_lock); +static struct apic apic_x2apic_uv_x; + static unsigned long __init uv_early_read_mmr(unsigned long addr) { unsigned long val, *mmr; @@ -319,10 +328,15 @@ static void uv_send_IPI_self(int vector) apic_write(APIC_SELF_IPI, vector); } -struct apic __refdata apic_x2apic_uv_x = { +static int uv_probe(void) +{ + return apic == &apic_x2apic_uv_x; +} + +static struct apic __refdata apic_x2apic_uv_x = { .name = "UV large system", - .probe = NULL, + .probe = uv_probe, .acpi_madt_oem_check = uv_acpi_madt_oem_check, .apic_id_registered = uv_apic_id_registered, @@ -642,18 +656,46 @@ void __cpuinit uv_cpu_init(void) */ int uv_handle_nmi(struct notifier_block *self, unsigned long reason, void *data) { + unsigned long real_uv_nmi; + int bid; + if (reason != DIE_NMIUNKNOWN) return NOTIFY_OK; if (in_crash_kexec) /* do nothing if entering the crash kernel */ return NOTIFY_OK; + + /* + * Each blade has an MMR that indicates when an NMI has been sent + * to cpus on the blade. If an NMI is detected, atomically + * clear the MMR and update a per-blade NMI count used to + * cause each cpu on the blade to notice a new NMI. + */ + bid = uv_numa_blade_id(); + real_uv_nmi = (uv_read_local_mmr(UVH_NMI_MMR) & UV_NMI_PENDING_MASK); + + if (unlikely(real_uv_nmi)) { + spin_lock(&uv_blade_info[bid].nmi_lock); + real_uv_nmi = (uv_read_local_mmr(UVH_NMI_MMR) & UV_NMI_PENDING_MASK); + if (real_uv_nmi) { + uv_blade_info[bid].nmi_count++; + uv_write_local_mmr(UVH_NMI_MMR_CLEAR, UV_NMI_PENDING_MASK); + } + spin_unlock(&uv_blade_info[bid].nmi_lock); + } + + if (likely(__get_cpu_var(cpu_last_nmi_count) == uv_blade_info[bid].nmi_count)) + return NOTIFY_DONE; + + __get_cpu_var(cpu_last_nmi_count) = uv_blade_info[bid].nmi_count; + /* - * Use a lock so only one cpu prints at a time - * to prevent intermixed output. + * Use a lock so only one cpu prints at a time. + * This prevents intermixed output. */ spin_lock(&uv_nmi_lock); - pr_info("NMI stack dump cpu %u:\n", smp_processor_id()); + pr_info("UV NMI stack dump cpu %u:\n", smp_processor_id()); dump_stack(); spin_unlock(&uv_nmi_lock); @@ -661,7 +703,8 @@ int uv_handle_nmi(struct notifier_block *self, unsigned long reason, void *data) } static struct notifier_block uv_dump_stack_nmi_nb = { - .notifier_call = uv_handle_nmi + .notifier_call = uv_handle_nmi, + .priority = NMI_LOCAL_LOW_PRIOR - 1, }; void uv_register_nmi_notifier(void) @@ -720,8 +763,9 @@ void __init uv_system_init(void) printk(KERN_DEBUG "UV: Found %d blades\n", uv_num_possible_blades()); bytes = sizeof(struct uv_blade_info) * uv_num_possible_blades(); - uv_blade_info = kmalloc(bytes, GFP_KERNEL); + uv_blade_info = kzalloc(bytes, GFP_KERNEL); BUG_ON(!uv_blade_info); + for (blade = 0; blade < uv_num_possible_blades(); blade++) uv_blade_info[blade].memory_nid = -1; @@ -747,6 +791,7 @@ void __init uv_system_init(void) uv_blade_info[blade].pnode = pnode; uv_blade_info[blade].nr_possible_cpus = 0; uv_blade_info[blade].nr_online_cpus = 0; + spin_lock_init(&uv_blade_info[blade].nmi_lock); max_pnode = max(pnode, max_pnode); blade++; } @@ -821,3 +866,5 @@ void __init uv_system_init(void) if (is_kdump_kernel()) reboot_type = BOOT_ACPI; } + +apic_driver(apic_x2apic_uv_x); diff --git a/arch/x86/kernel/apm_32.c b/arch/x86/kernel/apm_32.c index adee12e0da1f..3bfa02235965 100644 --- a/arch/x86/kernel/apm_32.c +++ b/arch/x86/kernel/apm_32.c @@ -1238,7 +1238,6 @@ static int suspend(int vetoable) dpm_suspend_noirq(PMSG_SUSPEND); local_irq_disable(); - sysdev_suspend(PMSG_SUSPEND); syscore_suspend(); local_irq_enable(); @@ -1258,7 +1257,6 @@ static int suspend(int vetoable) err = (err == APM_SUCCESS) ? 0 : -EIO; syscore_resume(); - sysdev_resume(); local_irq_enable(); dpm_resume_noirq(PMSG_RESUME); @@ -1282,7 +1280,6 @@ static void standby(void) dpm_suspend_noirq(PMSG_SUSPEND); local_irq_disable(); - sysdev_suspend(PMSG_SUSPEND); syscore_suspend(); local_irq_enable(); @@ -1292,7 +1289,6 @@ static void standby(void) local_irq_disable(); syscore_resume(); - sysdev_resume(); local_irq_enable(); dpm_resume_noirq(PMSG_RESUME); diff --git a/arch/x86/kernel/cpu/Makefile b/arch/x86/kernel/cpu/Makefile index 3f0ebe429a01..6042981d0309 100644 --- a/arch/x86/kernel/cpu/Makefile +++ b/arch/x86/kernel/cpu/Makefile @@ -30,7 +30,6 @@ obj-$(CONFIG_PERF_EVENTS) += perf_event.o obj-$(CONFIG_X86_MCE) += mcheck/ obj-$(CONFIG_MTRR) += mtrr/ -obj-$(CONFIG_CPU_FREQ) += cpufreq/ obj-$(CONFIG_X86_LOCAL_APIC) += perfctr-watchdog.o diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index 3532d3bf8105..8f5cabb3c5b0 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -613,7 +613,7 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c) #endif /* As a rule processors have APIC timer running in deep C states */ - if (c->x86 >= 0xf && !cpu_has_amd_erratum(amd_erratum_400)) + if (c->x86 > 0xf && !cpu_has_amd_erratum(amd_erratum_400)) set_cpu_cap(c, X86_FEATURE_ARAT); /* @@ -629,10 +629,13 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c) * Fixes: https://bugzilla.kernel.org/show_bug.cgi?id=33012 */ u64 mask; + int err; - rdmsrl(MSR_AMD64_MCx_MASK(4), mask); - mask |= (1 << 10); - wrmsrl(MSR_AMD64_MCx_MASK(4), mask); + err = rdmsrl_safe(MSR_AMD64_MCx_MASK(4), &mask); + if (err == 0) { + mask |= (1 << 10); + checking_wrmsrl(MSR_AMD64_MCx_MASK(4), mask); + } } } diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index e2ced0074a45..c8b41623377f 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -254,6 +254,25 @@ static inline void squash_the_stupid_serial_number(struct cpuinfo_x86 *c) } #endif +static int disable_smep __cpuinitdata; +static __init int setup_disable_smep(char *arg) +{ + disable_smep = 1; + return 1; +} +__setup("nosmep", setup_disable_smep); + +static __cpuinit void setup_smep(struct cpuinfo_x86 *c) +{ + if (cpu_has(c, X86_FEATURE_SMEP)) { + if (unlikely(disable_smep)) { + setup_clear_cpu_cap(X86_FEATURE_SMEP); + clear_in_cr4(X86_CR4_SMEP); + } else + set_in_cr4(X86_CR4_SMEP); + } +} + /* * Some CPU features depend on higher CPUID levels, which may not always * be available due to CPUID level capping or broken virtualization @@ -565,8 +584,7 @@ void __cpuinit get_cpu_cap(struct cpuinfo_x86 *c) cpuid_count(0x00000007, 0, &eax, &ebx, &ecx, &edx); - if (eax > 0) - c->x86_capability[9] = ebx; + c->x86_capability[9] = ebx; } /* AMD-defined flags: level 0x80000001 */ @@ -668,6 +686,8 @@ static void __init early_identify_cpu(struct cpuinfo_x86 *c) c->cpu_index = 0; #endif filter_cpuid_features(c, false); + + setup_smep(c); } void __init early_cpu_init(void) @@ -753,6 +773,8 @@ static void __cpuinit generic_identify(struct cpuinfo_x86 *c) #endif } + setup_smep(c); + get_model_name(c); /* Default name */ detect_nopl(c); diff --git a/arch/x86/kernel/cpu/cpufreq/Kconfig b/arch/x86/kernel/cpu/cpufreq/Kconfig deleted file mode 100644 index 870e6cc6ad28..000000000000 --- a/arch/x86/kernel/cpu/cpufreq/Kconfig +++ /dev/null @@ -1,266 +0,0 @@ -# -# CPU Frequency scaling -# - -menu "CPU Frequency scaling" - -source "drivers/cpufreq/Kconfig" - -if CPU_FREQ - -comment "CPUFreq processor drivers" - -config X86_PCC_CPUFREQ - tristate "Processor Clocking Control interface driver" - depends on ACPI && ACPI_PROCESSOR - help - This driver adds support for the PCC interface. - - For details, take a look at: - <file:Documentation/cpu-freq/pcc-cpufreq.txt>. - - To compile this driver as a module, choose M here: the - module will be called pcc-cpufreq. - - If in doubt, say N. - -config X86_ACPI_CPUFREQ - tristate "ACPI Processor P-States driver" - select CPU_FREQ_TABLE - depends on ACPI_PROCESSOR - help - This driver adds a CPUFreq driver which utilizes the ACPI - Processor Performance States. - This driver also supports Intel Enhanced Speedstep. - - To compile this driver as a module, choose M here: the - module will be called acpi-cpufreq. - - For details, take a look at <file:Documentation/cpu-freq/>. - - If in doubt, say N. - -config ELAN_CPUFREQ - tristate "AMD Elan SC400 and SC410" - select CPU_FREQ_TABLE - depends on X86_ELAN - ---help--- - This adds the CPUFreq driver for AMD Elan SC400 and SC410 - processors. - - You need to specify the processor maximum speed as boot - parameter: elanfreq=maxspeed (in kHz) or as module - parameter "max_freq". - - For details, take a look at <file:Documentation/cpu-freq/>. - - If in doubt, say N. - -config SC520_CPUFREQ - tristate "AMD Elan SC520" - select CPU_FREQ_TABLE - depends on X86_ELAN - ---help--- - This adds the CPUFreq driver for AMD Elan SC520 processor. - - For details, take a look at <file:Documentation/cpu-freq/>. - - If in doubt, say N. - - -config X86_POWERNOW_K6 - tristate "AMD Mobile K6-2/K6-3 PowerNow!" - select CPU_FREQ_TABLE - depends on X86_32 - help - This adds the CPUFreq driver for mobile AMD K6-2+ and mobile - AMD K6-3+ processors. - - For details, take a look at <file:Documentation/cpu-freq/>. - - If in doubt, say N. - -config X86_POWERNOW_K7 - tristate "AMD Mobile Athlon/Duron PowerNow!" - select CPU_FREQ_TABLE - depends on X86_32 - help - This adds the CPUFreq driver for mobile AMD K7 mobile processors. - - For details, take a look at <file:Documentation/cpu-freq/>. - - If in doubt, say N. - -config X86_POWERNOW_K7_ACPI - bool - depends on X86_POWERNOW_K7 && ACPI_PROCESSOR - depends on !(X86_POWERNOW_K7 = y && ACPI_PROCESSOR = m) - depends on X86_32 - default y - -config X86_POWERNOW_K8 - tristate "AMD Opteron/Athlon64 PowerNow!" - select CPU_FREQ_TABLE - depends on ACPI && ACPI_PROCESSOR - help - This adds the CPUFreq driver for K8/K10 Opteron/Athlon64 processors. - - To compile this driver as a module, choose M here: the - module will be called powernow-k8. - - For details, take a look at <file:Documentation/cpu-freq/>. - -config X86_GX_SUSPMOD - tristate "Cyrix MediaGX/NatSemi Geode Suspend Modulation" - depends on X86_32 && PCI - help - This add the CPUFreq driver for NatSemi Geode processors which - support suspend modulation. - - For details, take a look at <file:Documentation/cpu-freq/>. - - If in doubt, say N. - -config X86_SPEEDSTEP_CENTRINO - tristate "Intel Enhanced SpeedStep (deprecated)" - select CPU_FREQ_TABLE - select X86_SPEEDSTEP_CENTRINO_TABLE if X86_32 - depends on X86_32 || (X86_64 && ACPI_PROCESSOR) - help - This is deprecated and this functionality is now merged into - acpi_cpufreq (X86_ACPI_CPUFREQ). Use that driver instead of - speedstep_centrino. - This adds the CPUFreq driver for Enhanced SpeedStep enabled - mobile CPUs. This means Intel Pentium M (Centrino) CPUs - or 64bit enabled Intel Xeons. - - To compile this driver as a module, choose M here: the - module will be called speedstep-centrino. - - For details, take a look at <file:Documentation/cpu-freq/>. - - If in doubt, say N. - -config X86_SPEEDSTEP_CENTRINO_TABLE - bool "Built-in tables for Banias CPUs" - depends on X86_32 && X86_SPEEDSTEP_CENTRINO - default y - help - Use built-in tables for Banias CPUs if ACPI encoding - is not available. - - If in doubt, say N. - -config X86_SPEEDSTEP_ICH - tristate "Intel Speedstep on ICH-M chipsets (ioport interface)" - select CPU_FREQ_TABLE - depends on X86_32 - help - This adds the CPUFreq driver for certain mobile Intel Pentium III - (Coppermine), all mobile Intel Pentium III-M (Tualatin) and all - mobile Intel Pentium 4 P4-M on systems which have an Intel ICH2, - ICH3 or ICH4 southbridge. - - For details, take a look at <file:Documentation/cpu-freq/>. - - If in doubt, say N. - -config X86_SPEEDSTEP_SMI - tristate "Intel SpeedStep on 440BX/ZX/MX chipsets (SMI interface)" - select CPU_FREQ_TABLE - depends on X86_32 && EXPERIMENTAL - help - This adds the CPUFreq driver for certain mobile Intel Pentium III - (Coppermine), all mobile Intel Pentium III-M (Tualatin) - on systems which have an Intel 440BX/ZX/MX southbridge. - - For details, take a look at <file:Documentation/cpu-freq/>. - - If in doubt, say N. - -config X86_P4_CLOCKMOD - tristate "Intel Pentium 4 clock modulation" - select CPU_FREQ_TABLE - help - This adds the CPUFreq driver for Intel Pentium 4 / XEON - processors. When enabled it will lower CPU temperature by skipping - clocks. - - This driver should be only used in exceptional - circumstances when very low power is needed because it causes severe - slowdowns and noticeable latencies. Normally Speedstep should be used - instead. - - To compile this driver as a module, choose M here: the - module will be called p4-clockmod. - - For details, take a look at <file:Documentation/cpu-freq/>. - - Unless you are absolutely sure say N. - -config X86_CPUFREQ_NFORCE2 - tristate "nVidia nForce2 FSB changing" - depends on X86_32 && EXPERIMENTAL - help - This adds the CPUFreq driver for FSB changing on nVidia nForce2 - platforms. - - For details, take a look at <file:Documentation/cpu-freq/>. - - If in doubt, say N. - -config X86_LONGRUN - tristate "Transmeta LongRun" - depends on X86_32 - help - This adds the CPUFreq driver for Transmeta Crusoe and Efficeon processors - which support LongRun. - - For details, take a look at <file:Documentation/cpu-freq/>. - - If in doubt, say N. - -config X86_LONGHAUL - tristate "VIA Cyrix III Longhaul" - select CPU_FREQ_TABLE - depends on X86_32 && ACPI_PROCESSOR - help - This adds the CPUFreq driver for VIA Samuel/CyrixIII, - VIA Cyrix Samuel/C3, VIA Cyrix Ezra and VIA Cyrix Ezra-T - processors. - - For details, take a look at <file:Documentation/cpu-freq/>. - - If in doubt, say N. - -config X86_E_POWERSAVER - tristate "VIA C7 Enhanced PowerSaver (DANGEROUS)" - select CPU_FREQ_TABLE - depends on X86_32 && EXPERIMENTAL - help - This adds the CPUFreq driver for VIA C7 processors. However, this driver - does not have any safeguards to prevent operating the CPU out of spec - and is thus considered dangerous. Please use the regular ACPI cpufreq - driver, enabled by CONFIG_X86_ACPI_CPUFREQ. - - If in doubt, say N. - -comment "shared options" - -config X86_SPEEDSTEP_LIB - tristate - default (X86_SPEEDSTEP_ICH || X86_SPEEDSTEP_SMI || X86_P4_CLOCKMOD) - -config X86_SPEEDSTEP_RELAXED_CAP_CHECK - bool "Relaxed speedstep capability checks" - depends on X86_32 && (X86_SPEEDSTEP_SMI || X86_SPEEDSTEP_ICH) - help - Don't perform all checks for a speedstep capable system which would - normally be done. Some ancient or strange systems, though speedstep - capable, don't always indicate that they are speedstep capable. This - option lets the probing code bypass some of those checks if the - parameter "relaxed_check=1" is passed to the module. - -endif # CPU_FREQ - -endmenu diff --git a/arch/x86/kernel/cpu/cpufreq/Makefile b/arch/x86/kernel/cpu/cpufreq/Makefile deleted file mode 100644 index bd54bf67e6fb..000000000000 --- a/arch/x86/kernel/cpu/cpufreq/Makefile +++ /dev/null @@ -1,21 +0,0 @@ -# Link order matters. K8 is preferred to ACPI because of firmware bugs in early -# K8 systems. ACPI is preferred to all other hardware-specific drivers. -# speedstep-* is preferred over p4-clockmod. - -obj-$(CONFIG_X86_POWERNOW_K8) += powernow-k8.o mperf.o -obj-$(CONFIG_X86_ACPI_CPUFREQ) += acpi-cpufreq.o mperf.o -obj-$(CONFIG_X86_PCC_CPUFREQ) += pcc-cpufreq.o -obj-$(CONFIG_X86_POWERNOW_K6) += powernow-k6.o -obj-$(CONFIG_X86_POWERNOW_K7) += powernow-k7.o -obj-$(CONFIG_X86_LONGHAUL) += longhaul.o -obj-$(CONFIG_X86_E_POWERSAVER) += e_powersaver.o -obj-$(CONFIG_ELAN_CPUFREQ) += elanfreq.o -obj-$(CONFIG_SC520_CPUFREQ) += sc520_freq.o -obj-$(CONFIG_X86_LONGRUN) += longrun.o -obj-$(CONFIG_X86_GX_SUSPMOD) += gx-suspmod.o -obj-$(CONFIG_X86_SPEEDSTEP_ICH) += speedstep-ich.o -obj-$(CONFIG_X86_SPEEDSTEP_LIB) += speedstep-lib.o -obj-$(CONFIG_X86_SPEEDSTEP_SMI) += speedstep-smi.o -obj-$(CONFIG_X86_SPEEDSTEP_CENTRINO) += speedstep-centrino.o -obj-$(CONFIG_X86_P4_CLOCKMOD) += p4-clockmod.o -obj-$(CONFIG_X86_CPUFREQ_NFORCE2) += cpufreq-nforce2.o diff --git a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c deleted file mode 100644 index a2baafb2fe6d..000000000000 --- a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c +++ /dev/null @@ -1,776 +0,0 @@ -/* - * acpi-cpufreq.c - ACPI Processor P-States Driver - * - * Copyright (C) 2001, 2002 Andy Grover <andrew.grover@intel.com> - * Copyright (C) 2001, 2002 Paul Diefenbaugh <paul.s.diefenbaugh@intel.com> - * Copyright (C) 2002 - 2004 Dominik Brodowski <linux@brodo.de> - * Copyright (C) 2006 Denis Sadykov <denis.m.sadykov@intel.com> - * - * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or (at - * your option) any later version. - * - * This program is distributed in the hope that it will be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU - * General Public License for more details. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write to the Free Software Foundation, Inc., - * 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA. - * - * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ - */ - -#include <linux/kernel.h> -#include <linux/module.h> -#include <linux/init.h> -#include <linux/smp.h> -#include <linux/sched.h> -#include <linux/cpufreq.h> -#include <linux/compiler.h> -#include <linux/dmi.h> -#include <linux/slab.h> - -#include <linux/acpi.h> -#include <linux/io.h> -#include <linux/delay.h> -#include <linux/uaccess.h> - -#include <acpi/processor.h> - -#include <asm/msr.h> -#include <asm/processor.h> -#include <asm/cpufeature.h> -#include "mperf.h" - -#define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, \ - "acpi-cpufreq", msg) - -MODULE_AUTHOR("Paul Diefenbaugh, Dominik Brodowski"); -MODULE_DESCRIPTION("ACPI Processor P-States Driver"); -MODULE_LICENSE("GPL"); - -enum { - UNDEFINED_CAPABLE = 0, - SYSTEM_INTEL_MSR_CAPABLE, - SYSTEM_IO_CAPABLE, -}; - -#define INTEL_MSR_RANGE (0xffff) - -struct acpi_cpufreq_data { - struct acpi_processor_performance *acpi_data; - struct cpufreq_frequency_table *freq_table; - unsigned int resume; - unsigned int cpu_feature; -}; - -static DEFINE_PER_CPU(struct acpi_cpufreq_data *, acfreq_data); - -/* acpi_perf_data is a pointer to percpu data. */ -static struct acpi_processor_performance __percpu *acpi_perf_data; - -static struct cpufreq_driver acpi_cpufreq_driver; - -static unsigned int acpi_pstate_strict; - -static int check_est_cpu(unsigned int cpuid) -{ - struct cpuinfo_x86 *cpu = &cpu_data(cpuid); - - return cpu_has(cpu, X86_FEATURE_EST); -} - -static unsigned extract_io(u32 value, struct acpi_cpufreq_data *data) -{ - struct acpi_processor_performance *perf; - int i; - - perf = data->acpi_data; - - for (i = 0; i < perf->state_count; i++) { - if (value == perf->states[i].status) - return data->freq_table[i].frequency; - } - return 0; -} - -static unsigned extract_msr(u32 msr, struct acpi_cpufreq_data *data) -{ - int i; - struct acpi_processor_performance *perf; - - msr &= INTEL_MSR_RANGE; - perf = data->acpi_data; - - for (i = 0; data->freq_table[i].frequency != CPUFREQ_TABLE_END; i++) { - if (msr == perf->states[data->freq_table[i].index].status) - return data->freq_table[i].frequency; - } - return data->freq_table[0].frequency; -} - -static unsigned extract_freq(u32 val, struct acpi_cpufreq_data *data) -{ - switch (data->cpu_feature) { - case SYSTEM_INTEL_MSR_CAPABLE: - return extract_msr(val, data); - case SYSTEM_IO_CAPABLE: - return extract_io(val, data); - default: - return 0; - } -} - -struct msr_addr { - u32 reg; -}; - -struct io_addr { - u16 port; - u8 bit_width; -}; - -struct drv_cmd { - unsigned int type; - const struct cpumask *mask; - union { - struct msr_addr msr; - struct io_addr io; - } addr; - u32 val; -}; - -/* Called via smp_call_function_single(), on the target CPU */ -static void do_drv_read(void *_cmd) -{ - struct drv_cmd *cmd = _cmd; - u32 h; - - switch (cmd->type) { - case SYSTEM_INTEL_MSR_CAPABLE: - rdmsr(cmd->addr.msr.reg, cmd->val, h); - break; - case SYSTEM_IO_CAPABLE: - acpi_os_read_port((acpi_io_address)cmd->addr.io.port, - &cmd->val, - (u32)cmd->addr.io.bit_width); - break; - default: - break; - } -} - -/* Called via smp_call_function_many(), on the target CPUs */ -static void do_drv_write(void *_cmd) -{ - struct drv_cmd *cmd = _cmd; - u32 lo, hi; - - switch (cmd->type) { - case SYSTEM_INTEL_MSR_CAPABLE: - rdmsr(cmd->addr.msr.reg, lo, hi); - lo = (lo & ~INTEL_MSR_RANGE) | (cmd->val & INTEL_MSR_RANGE); - wrmsr(cmd->addr.msr.reg, lo, hi); - break; - case SYSTEM_IO_CAPABLE: - acpi_os_write_port((acpi_io_address)cmd->addr.io.port, - cmd->val, - (u32)cmd->addr.io.bit_width); - break; - default: - break; - } -} - -static void drv_read(struct drv_cmd *cmd) -{ - int err; - cmd->val = 0; - - err = smp_call_function_any(cmd->mask, do_drv_read, cmd, 1); - WARN_ON_ONCE(err); /* smp_call_function_any() was buggy? */ -} - -static void drv_write(struct drv_cmd *cmd) -{ - int this_cpu; - - this_cpu = get_cpu(); - if (cpumask_test_cpu(this_cpu, cmd->mask)) - do_drv_write(cmd); - smp_call_function_many(cmd->mask, do_drv_write, cmd, 1); - put_cpu(); -} - -static u32 get_cur_val(const struct cpumask *mask) -{ - struct acpi_processor_performance *perf; - struct drv_cmd cmd; - - if (unlikely(cpumask_empty(mask))) - return 0; - - switch (per_cpu(acfreq_data, cpumask_first(mask))->cpu_feature) { - case SYSTEM_INTEL_MSR_CAPABLE: - cmd.type = SYSTEM_INTEL_MSR_CAPABLE; - cmd.addr.msr.reg = MSR_IA32_PERF_STATUS; - break; - case SYSTEM_IO_CAPABLE: - cmd.type = SYSTEM_IO_CAPABLE; - perf = per_cpu(acfreq_data, cpumask_first(mask))->acpi_data; - cmd.addr.io.port = perf->control_register.address; - cmd.addr.io.bit_width = perf->control_register.bit_width; - break; - default: - return 0; - } - - cmd.mask = mask; - drv_read(&cmd); - - dprintk("get_cur_val = %u\n", cmd.val); - - return cmd.val; -} - -static unsigned int get_cur_freq_on_cpu(unsigned int cpu) -{ - struct acpi_cpufreq_data *data = per_cpu(acfreq_data, cpu); - unsigned int freq; - unsigned int cached_freq; - - dprintk("get_cur_freq_on_cpu (%d)\n", cpu); - - if (unlikely(data == NULL || - data->acpi_data == NULL || data->freq_table == NULL)) { - return 0; - } - - cached_freq = data->freq_table[data->acpi_data->state].frequency; - freq = extract_freq(get_cur_val(cpumask_of(cpu)), data); - if (freq != cached_freq) { - /* - * The dreaded BIOS frequency change behind our back. - * Force set the frequency on next target call. - */ - data->resume = 1; - } - - dprintk("cur freq = %u\n", freq); - - return freq; -} - -static unsigned int check_freqs(const struct cpumask *mask, unsigned int freq, - struct acpi_cpufreq_data *data) -{ - unsigned int cur_freq; - unsigned int i; - - for (i = 0; i < 100; i++) { - cur_freq = extract_freq(get_cur_val(mask), data); - if (cur_freq == freq) - return 1; - udelay(10); - } - return 0; -} - -static int acpi_cpufreq_target(struct cpufreq_policy *policy, - unsigned int target_freq, unsigned int relation) -{ - struct acpi_cpufreq_data *data = per_cpu(acfreq_data, policy->cpu); - struct acpi_processor_performance *perf; - struct cpufreq_freqs freqs; - struct drv_cmd cmd; - unsigned int next_state = 0; /* Index into freq_table */ - unsigned int next_perf_state = 0; /* Index into perf table */ - unsigned int i; - int result = 0; - - dprintk("acpi_cpufreq_target %d (%d)\n", target_freq, policy->cpu); - - if (unlikely(data == NULL || - data->acpi_data == NULL || data->freq_table == NULL)) { - return -ENODEV; - } - - perf = data->acpi_data; - result = cpufreq_frequency_table_target(policy, - data->freq_table, - target_freq, - relation, &next_state); - if (unlikely(result)) { - result = -ENODEV; - goto out; - } - - next_perf_state = data->freq_table[next_state].index; - if (perf->state == next_perf_state) { - if (unlikely(data->resume)) { - dprintk("Called after resume, resetting to P%d\n", - next_perf_state); - data->resume = 0; - } else { - dprintk("Already at target state (P%d)\n", - next_perf_state); - goto out; - } - } - - switch (data->cpu_feature) { - case SYSTEM_INTEL_MSR_CAPABLE: - cmd.type = SYSTEM_INTEL_MSR_CAPABLE; - cmd.addr.msr.reg = MSR_IA32_PERF_CTL; - cmd.val = (u32) perf->states[next_perf_state].control; - break; - case SYSTEM_IO_CAPABLE: - cmd.type = SYSTEM_IO_CAPABLE; - cmd.addr.io.port = perf->control_register.address; - cmd.addr.io.bit_width = perf->control_register.bit_width; - cmd.val = (u32) perf->states[next_perf_state].control; - break; - default: - result = -ENODEV; - goto out; - } - - /* cpufreq holds the hotplug lock, so we are safe from here on */ - if (policy->shared_type != CPUFREQ_SHARED_TYPE_ANY) - cmd.mask = policy->cpus; - else - cmd.mask = cpumask_of(policy->cpu); - - freqs.old = perf->states[perf->state].core_frequency * 1000; - freqs.new = data->freq_table[next_state].frequency; - for_each_cpu(i, policy->cpus) { - freqs.cpu = i; - cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); - } - - drv_write(&cmd); - - if (acpi_pstate_strict) { - if (!check_freqs(cmd.mask, freqs.new, data)) { - dprintk("acpi_cpufreq_target failed (%d)\n", - policy->cpu); - result = -EAGAIN; - goto out; - } - } - - for_each_cpu(i, policy->cpus) { - freqs.cpu = i; - cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); - } - perf->state = next_perf_state; - -out: - return result; -} - -static int acpi_cpufreq_verify(struct cpufreq_policy *policy) -{ - struct acpi_cpufreq_data *data = per_cpu(acfreq_data, policy->cpu); - - dprintk("acpi_cpufreq_verify\n"); - - return cpufreq_frequency_table_verify(policy, data->freq_table); -} - -static unsigned long -acpi_cpufreq_guess_freq(struct acpi_cpufreq_data *data, unsigned int cpu) -{ - struct acpi_processor_performance *perf = data->acpi_data; - - if (cpu_khz) { - /* search the closest match to cpu_khz */ - unsigned int i; - unsigned long freq; - unsigned long freqn = perf->states[0].core_frequency * 1000; - - for (i = 0; i < (perf->state_count-1); i++) { - freq = freqn; - freqn = perf->states[i+1].core_frequency * 1000; - if ((2 * cpu_khz) > (freqn + freq)) { - perf->state = i; - return freq; - } - } - perf->state = perf->state_count-1; - return freqn; - } else { - /* assume CPU is at P0... */ - perf->state = 0; - return perf->states[0].core_frequency * 1000; - } -} - -static void free_acpi_perf_data(void) -{ - unsigned int i; - - /* Freeing a NULL pointer is OK, and alloc_percpu zeroes. */ - for_each_possible_cpu(i) - free_cpumask_var(per_cpu_ptr(acpi_perf_data, i) - ->shared_cpu_map); - free_percpu(acpi_perf_data); -} - -/* - * acpi_cpufreq_early_init - initialize ACPI P-States library - * - * Initialize the ACPI P-States library (drivers/acpi/processor_perflib.c) - * in order to determine correct frequency and voltage pairings. We can - * do _PDC and _PSD and find out the processor dependency for the - * actual init that will happen later... - */ -static int __init acpi_cpufreq_early_init(void) -{ - unsigned int i; - dprintk("acpi_cpufreq_early_init\n"); - - acpi_perf_data = alloc_percpu(struct acpi_processor_performance); - if (!acpi_perf_data) { - dprintk("Memory allocation error for acpi_perf_data.\n"); - return -ENOMEM; - } - for_each_possible_cpu(i) { - if (!zalloc_cpumask_var_node( - &per_cpu_ptr(acpi_perf_data, i)->shared_cpu_map, - GFP_KERNEL, cpu_to_node(i))) { - - /* Freeing a NULL pointer is OK: alloc_percpu zeroes. */ - free_acpi_perf_data(); - return -ENOMEM; - } - } - - /* Do initialization in ACPI core */ - acpi_processor_preregister_performance(acpi_perf_data); - return 0; -} - -#ifdef CONFIG_SMP -/* - * Some BIOSes do SW_ANY coordination internally, either set it up in hw - * or do it in BIOS firmware and won't inform about it to OS. If not - * detected, this has a side effect of making CPU run at a different speed - * than OS intended it to run at. Detect it and handle it cleanly. - */ -static int bios_with_sw_any_bug; - -static int sw_any_bug_found(const struct dmi_system_id *d) -{ - bios_with_sw_any_bug = 1; - return 0; -} - -static const struct dmi_system_id sw_any_bug_dmi_table[] = { - { - .callback = sw_any_bug_found, - .ident = "Supermicro Server X6DLP", - .matches = { - DMI_MATCH(DMI_SYS_VENDOR, "Supermicro"), - DMI_MATCH(DMI_BIOS_VERSION, "080010"), - DMI_MATCH(DMI_PRODUCT_NAME, "X6DLP"), - }, - }, - { } -}; - -static int acpi_cpufreq_blacklist(struct cpuinfo_x86 *c) -{ - /* Intel Xeon Processor 7100 Series Specification Update - * http://www.intel.com/Assets/PDF/specupdate/314554.pdf - * AL30: A Machine Check Exception (MCE) Occurring during an - * Enhanced Intel SpeedStep Technology Ratio Change May Cause - * Both Processor Cores to Lock Up. */ - if (c->x86_vendor == X86_VENDOR_INTEL) { - if ((c->x86 == 15) && - (c->x86_model == 6) && - (c->x86_mask == 8)) { - printk(KERN_INFO "acpi-cpufreq: Intel(R) " - "Xeon(R) 7100 Errata AL30, processors may " - "lock up on frequency changes: disabling " - "acpi-cpufreq.\n"); - return -ENODEV; - } - } - return 0; -} -#endif - -static int acpi_cpufreq_cpu_init(struct cpufreq_policy *policy) -{ - unsigned int i; - unsigned int valid_states = 0; - unsigned int cpu = policy->cpu; - struct acpi_cpufreq_data *data; - unsigned int result = 0; - struct cpuinfo_x86 *c = &cpu_data(policy->cpu); - struct acpi_processor_performance *perf; -#ifdef CONFIG_SMP - static int blacklisted; -#endif - - dprintk("acpi_cpufreq_cpu_init\n"); - -#ifdef CONFIG_SMP - if (blacklisted) - return blacklisted; - blacklisted = acpi_cpufreq_blacklist(c); - if (blacklisted) - return blacklisted; -#endif - - data = kzalloc(sizeof(struct acpi_cpufreq_data), GFP_KERNEL); - if (!data) - return -ENOMEM; - - data->acpi_data = per_cpu_ptr(acpi_perf_data, cpu); - per_cpu(acfreq_data, cpu) = data; - - if (cpu_has(c, X86_FEATURE_CONSTANT_TSC)) - acpi_cpufreq_driver.flags |= CPUFREQ_CONST_LOOPS; - - result = acpi_processor_register_performance(data->acpi_data, cpu); - if (result) - goto err_free; - - perf = data->acpi_data; - policy->shared_type = perf->shared_type; - - /* - * Will let policy->cpus know about dependency only when software - * coordination is required. - */ - if (policy->shared_type == CPUFREQ_SHARED_TYPE_ALL || - policy->shared_type == CPUFREQ_SHARED_TYPE_ANY) { - cpumask_copy(policy->cpus, perf->shared_cpu_map); - } - cpumask_copy(policy->related_cpus, perf->shared_cpu_map); - -#ifdef CONFIG_SMP - dmi_check_system(sw_any_bug_dmi_table); - if (bios_with_sw_any_bug && cpumask_weight(policy->cpus) == 1) { - policy->shared_type = CPUFREQ_SHARED_TYPE_ALL; - cpumask_copy(policy->cpus, cpu_core_mask(cpu)); - } -#endif - - /* capability check */ - if (perf->state_count <= 1) { - dprintk("No P-States\n"); - result = -ENODEV; - goto err_unreg; - } - - if (perf->control_register.space_id != perf->status_register.space_id) { - result = -ENODEV; - goto err_unreg; - } - - switch (perf->control_register.space_id) { - case ACPI_ADR_SPACE_SYSTEM_IO: - dprintk("SYSTEM IO addr space\n"); - data->cpu_feature = SYSTEM_IO_CAPABLE; - break; - case ACPI_ADR_SPACE_FIXED_HARDWARE: - dprintk("HARDWARE addr space\n"); - if (!check_est_cpu(cpu)) { - result = -ENODEV; - goto err_unreg; - } - data->cpu_feature = SYSTEM_INTEL_MSR_CAPABLE; - break; - default: - dprintk("Unknown addr space %d\n", - (u32) (perf->control_register.space_id)); - result = -ENODEV; - goto err_unreg; - } - - data->freq_table = kmalloc(sizeof(struct cpufreq_frequency_table) * - (perf->state_count+1), GFP_KERNEL); - if (!data->freq_table) { - result = -ENOMEM; - goto err_unreg; - } - - /* detect transition latency */ - policy->cpuinfo.transition_latency = 0; - for (i = 0; i < perf->state_count; i++) { - if ((perf->states[i].transition_latency * 1000) > - policy->cpuinfo.transition_latency) - policy->cpuinfo.transition_latency = - perf->states[i].transition_latency * 1000; - } - - /* Check for high latency (>20uS) from buggy BIOSes, like on T42 */ - if (perf->control_register.space_id == ACPI_ADR_SPACE_FIXED_HARDWARE && - policy->cpuinfo.transition_latency > 20 * 1000) { - policy->cpuinfo.transition_latency = 20 * 1000; - printk_once(KERN_INFO - "P-state transition latency capped at 20 uS\n"); - } - - /* table init */ - for (i = 0; i < perf->state_count; i++) { - if (i > 0 && perf->states[i].core_frequency >= - data->freq_table[valid_states-1].frequency / 1000) - continue; - - data->freq_table[valid_states].index = i; - data->freq_table[valid_states].frequency = - perf->states[i].core_frequency * 1000; - valid_states++; - } - data->freq_table[valid_states].frequency = CPUFREQ_TABLE_END; - perf->state = 0; - - result = cpufreq_frequency_table_cpuinfo(policy, data->freq_table); - if (result) - goto err_freqfree; - - if (perf->states[0].core_frequency * 1000 != policy->cpuinfo.max_freq) - printk(KERN_WARNING FW_WARN "P-state 0 is not max freq\n"); - - switch (perf->control_register.space_id) { - case ACPI_ADR_SPACE_SYSTEM_IO: - /* Current speed is unknown and not detectable by IO port */ - policy->cur = acpi_cpufreq_guess_freq(data, policy->cpu); - break; - case ACPI_ADR_SPACE_FIXED_HARDWARE: - acpi_cpufreq_driver.get = get_cur_freq_on_cpu; - policy->cur = get_cur_freq_on_cpu(cpu); - break; - default: - break; - } - - /* notify BIOS that we exist */ - acpi_processor_notify_smm(THIS_MODULE); - - /* Check for APERF/MPERF support in hardware */ - if (cpu_has(c, X86_FEATURE_APERFMPERF)) - acpi_cpufreq_driver.getavg = cpufreq_get_measured_perf; - - dprintk("CPU%u - ACPI performance management activated.\n", cpu); - for (i = 0; i < perf->state_count; i++) - dprintk(" %cP%d: %d MHz, %d mW, %d uS\n", - (i == perf->state ? '*' : ' '), i, - (u32) perf->states[i].core_frequency, - (u32) perf->states[i].power, - (u32) perf->states[i].transition_latency); - - cpufreq_frequency_table_get_attr(data->freq_table, policy->cpu); - - /* - * the first call to ->target() should result in us actually - * writing something to the appropriate registers. - */ - data->resume = 1; - - return result; - -err_freqfree: - kfree(data->freq_table); -err_unreg: - acpi_processor_unregister_performance(perf, cpu); -err_free: - kfree(data); - per_cpu(acfreq_data, cpu) = NULL; - - return result; -} - -static int acpi_cpufreq_cpu_exit(struct cpufreq_policy *policy) -{ - struct acpi_cpufreq_data *data = per_cpu(acfreq_data, policy->cpu); - - dprintk("acpi_cpufreq_cpu_exit\n"); - - if (data) { - cpufreq_frequency_table_put_attr(policy->cpu); - per_cpu(acfreq_data, policy->cpu) = NULL; - acpi_processor_unregister_performance(data->acpi_data, - policy->cpu); - kfree(data->freq_table); - kfree(data); - } - - return 0; -} - -static int acpi_cpufreq_resume(struct cpufreq_policy *policy) -{ - struct acpi_cpufreq_data *data = per_cpu(acfreq_data, policy->cpu); - - dprintk("acpi_cpufreq_resume\n"); - - data->resume = 1; - - return 0; -} - -static struct freq_attr *acpi_cpufreq_attr[] = { - &cpufreq_freq_attr_scaling_available_freqs, - NULL, -}; - -static struct cpufreq_driver acpi_cpufreq_driver = { - .verify = acpi_cpufreq_verify, - .target = acpi_cpufreq_target, - .bios_limit = acpi_processor_get_bios_limit, - .init = acpi_cpufreq_cpu_init, - .exit = acpi_cpufreq_cpu_exit, - .resume = acpi_cpufreq_resume, - .name = "acpi-cpufreq", - .owner = THIS_MODULE, - .attr = acpi_cpufreq_attr, -}; - -static int __init acpi_cpufreq_init(void) -{ - int ret; - - if (acpi_disabled) - return 0; - - dprintk("acpi_cpufreq_init\n"); - - ret = acpi_cpufreq_early_init(); - if (ret) - return ret; - - ret = cpufreq_register_driver(&acpi_cpufreq_driver); - if (ret) - free_acpi_perf_data(); - - return ret; -} - -static void __exit acpi_cpufreq_exit(void) -{ - dprintk("acpi_cpufreq_exit\n"); - - cpufreq_unregister_driver(&acpi_cpufreq_driver); - - free_percpu(acpi_perf_data); -} - -module_param(acpi_pstate_strict, uint, 0644); -MODULE_PARM_DESC(acpi_pstate_strict, - "value 0 or non-zero. non-zero -> strict ACPI checks are " - "performed during frequency changes."); - -late_initcall(acpi_cpufreq_init); -module_exit(acpi_cpufreq_exit); - -MODULE_ALIAS("acpi"); diff --git a/arch/x86/kernel/cpu/cpufreq/cpufreq-nforce2.c b/arch/x86/kernel/cpu/cpufreq/cpufreq-nforce2.c deleted file mode 100644 index 141abebc4516..000000000000 --- a/arch/x86/kernel/cpu/cpufreq/cpufreq-nforce2.c +++ /dev/null @@ -1,446 +0,0 @@ -/* - * (C) 2004-2006 Sebastian Witt <se.witt@gmx.net> - * - * Licensed under the terms of the GNU GPL License version 2. - * Based upon reverse engineered information - * - * BIG FAT DISCLAIMER: Work in progress code. Possibly *dangerous* - */ - -#include <linux/kernel.h> -#include <linux/module.h> -#include <linux/moduleparam.h> -#include <linux/init.h> -#include <linux/cpufreq.h> -#include <linux/pci.h> -#include <linux/delay.h> - -#define NFORCE2_XTAL 25 -#define NFORCE2_BOOTFSB 0x48 -#define NFORCE2_PLLENABLE 0xa8 -#define NFORCE2_PLLREG 0xa4 -#define NFORCE2_PLLADR 0xa0 -#define NFORCE2_PLL(mul, div) (0x100000 | (mul << 8) | div) - -#define NFORCE2_MIN_FSB 50 -#define NFORCE2_SAFE_DISTANCE 50 - -/* Delay in ms between FSB changes */ -/* #define NFORCE2_DELAY 10 */ - -/* - * nforce2_chipset: - * FSB is changed using the chipset - */ -static struct pci_dev *nforce2_dev; - -/* fid: - * multiplier * 10 - */ -static int fid; - -/* min_fsb, max_fsb: - * minimum and maximum FSB (= FSB at boot time) - */ -static int min_fsb; -static int max_fsb; - -MODULE_AUTHOR("Sebastian Witt <se.witt@gmx.net>"); -MODULE_DESCRIPTION("nForce2 FSB changing cpufreq driver"); -MODULE_LICENSE("GPL"); - -module_param(fid, int, 0444); -module_param(min_fsb, int, 0444); - -MODULE_PARM_DESC(fid, "CPU multiplier to use (11.5 = 115)"); -MODULE_PARM_DESC(min_fsb, - "Minimum FSB to use, if not defined: current FSB - 50"); - -#define PFX "cpufreq-nforce2: " -#define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, \ - "cpufreq-nforce2", msg) - -/** - * nforce2_calc_fsb - calculate FSB - * @pll: PLL value - * - * Calculates FSB from PLL value - */ -static int nforce2_calc_fsb(int pll) -{ - unsigned char mul, div; - - mul = (pll >> 8) & 0xff; - div = pll & 0xff; - - if (div > 0) - return NFORCE2_XTAL * mul / div; - - return 0; -} - -/** - * nforce2_calc_pll - calculate PLL value - * @fsb: FSB - * - * Calculate PLL value for given FSB - */ -static int nforce2_calc_pll(unsigned int fsb) -{ - unsigned char xmul, xdiv; - unsigned char mul = 0, div = 0; - int tried = 0; - - /* Try to calculate multiplier and divider up to 4 times */ - while (((mul == 0) || (div == 0)) && (tried <= 3)) { - for (xdiv = 2; xdiv <= 0x80; xdiv++) - for (xmul = 1; xmul <= 0xfe; xmul++) - if (nforce2_calc_fsb(NFORCE2_PLL(xmul, xdiv)) == - fsb + tried) { - mul = xmul; - div = xdiv; - } - tried++; - } - - if ((mul == 0) || (div == 0)) - return -1; - - return NFORCE2_PLL(mul, div); -} - -/** - * nforce2_write_pll - write PLL value to chipset - * @pll: PLL value - * - * Writes new FSB PLL value to chipset - */ -static void nforce2_write_pll(int pll) -{ - int temp; - - /* Set the pll addr. to 0x00 */ - pci_write_config_dword(nforce2_dev, NFORCE2_PLLADR, 0); - - /* Now write the value in all 64 registers */ - for (temp = 0; temp <= 0x3f; temp++) - pci_write_config_dword(nforce2_dev, NFORCE2_PLLREG, pll); - - return; -} - -/** - * nforce2_fsb_read - Read FSB - * - * Read FSB from chipset - * If bootfsb != 0, return FSB at boot-time - */ -static unsigned int nforce2_fsb_read(int bootfsb) -{ - struct pci_dev *nforce2_sub5; - u32 fsb, temp = 0; - - /* Get chipset boot FSB from subdevice 5 (FSB at boot-time) */ - nforce2_sub5 = pci_get_subsys(PCI_VENDOR_ID_NVIDIA, 0x01EF, - PCI_ANY_ID, PCI_ANY_ID, NULL); - if (!nforce2_sub5) - return 0; - - pci_read_config_dword(nforce2_sub5, NFORCE2_BOOTFSB, &fsb); - fsb /= 1000000; - - /* Check if PLL register is already set */ - pci_read_config_byte(nforce2_dev, NFORCE2_PLLENABLE, (u8 *)&temp); - - if (bootfsb || !temp) - return fsb; - - /* Use PLL register FSB value */ - pci_read_config_dword(nforce2_dev, NFORCE2_PLLREG, &temp); - fsb = nforce2_calc_fsb(temp); - - return fsb; -} - -/** - * nforce2_set_fsb - set new FSB - * @fsb: New FSB - * - * Sets new FSB - */ -static int nforce2_set_fsb(unsigned int fsb) -{ - u32 temp = 0; - unsigned int tfsb; - int diff; - int pll = 0; - - if ((fsb > max_fsb) || (fsb < NFORCE2_MIN_FSB)) { - printk(KERN_ERR PFX "FSB %d is out of range!\n", fsb); - return -EINVAL; - } - - tfsb = nforce2_fsb_read(0); - if (!tfsb) { - printk(KERN_ERR PFX "Error while reading the FSB\n"); - return -EINVAL; - } - - /* First write? Then set actual value */ - pci_read_config_byte(nforce2_dev, NFORCE2_PLLENABLE, (u8 *)&temp); - if (!temp) { - pll = nforce2_calc_pll(tfsb); - - if (pll < 0) - return -EINVAL; - - nforce2_write_pll(pll); - } - - /* Enable write access */ - temp = 0x01; - pci_write_config_byte(nforce2_dev, NFORCE2_PLLENABLE, (u8)temp); - - diff = tfsb - fsb; - - if (!diff) - return 0; - - while ((tfsb != fsb) && (tfsb <= max_fsb) && (tfsb >= min_fsb)) { - if (diff < 0) - tfsb++; - else - tfsb--; - - /* Calculate the PLL reg. value */ - pll = nforce2_calc_pll(tfsb); - if (pll == -1) - return -EINVAL; - - nforce2_write_pll(pll); -#ifdef NFORCE2_DELAY - mdelay(NFORCE2_DELAY); -#endif - } - - temp = 0x40; - pci_write_config_byte(nforce2_dev, NFORCE2_PLLADR, (u8)temp); - - return 0; -} - -/** - * nforce2_get - get the CPU frequency - * @cpu: CPU number - * - * Returns the CPU frequency - */ -static unsigned int nforce2_get(unsigned int cpu) -{ - if (cpu) - return 0; - return nforce2_fsb_read(0) * fid * 100; -} - -/** - * nforce2_target - set a new CPUFreq policy - * @policy: new policy - * @target_freq: the target frequency - * @relation: how that frequency relates to achieved frequency - * (CPUFREQ_RELATION_L or CPUFREQ_RELATION_H) - * - * Sets a new CPUFreq policy. - */ -static int nforce2_target(struct cpufreq_policy *policy, - unsigned int target_freq, unsigned int relation) -{ -/* unsigned long flags; */ - struct cpufreq_freqs freqs; - unsigned int target_fsb; - - if ((target_freq > policy->max) || (target_freq < policy->min)) - return -EINVAL; - - target_fsb = target_freq / (fid * 100); - - freqs.old = nforce2_get(policy->cpu); - freqs.new = target_fsb * fid * 100; - freqs.cpu = 0; /* Only one CPU on nForce2 platforms */ - - if (freqs.old == freqs.new) - return 0; - - dprintk("Old CPU frequency %d kHz, new %d kHz\n", - freqs.old, freqs.new); - - cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); - - /* Disable IRQs */ - /* local_irq_save(flags); */ - - if (nforce2_set_fsb(target_fsb) < 0) - printk(KERN_ERR PFX "Changing FSB to %d failed\n", - target_fsb); - else - dprintk("Changed FSB successfully to %d\n", - target_fsb); - - /* Enable IRQs */ - /* local_irq_restore(flags); */ - - cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); - - return 0; -} - -/** - * nforce2_verify - verifies a new CPUFreq policy - * @policy: new policy - */ -static int nforce2_verify(struct cpufreq_policy *policy) -{ - unsigned int fsb_pol_max; - - fsb_pol_max = policy->max / (fid * 100); - - if (policy->min < (fsb_pol_max * fid * 100)) - policy->max = (fsb_pol_max + 1) * fid * 100; - - cpufreq_verify_within_limits(policy, - policy->cpuinfo.min_freq, - policy->cpuinfo.max_freq); - return 0; -} - -static int nforce2_cpu_init(struct cpufreq_policy *policy) -{ - unsigned int fsb; - unsigned int rfid; - - /* capability check */ - if (policy->cpu != 0) - return -ENODEV; - - /* Get current FSB */ - fsb = nforce2_fsb_read(0); - - if (!fsb) - return -EIO; - - /* FIX: Get FID from CPU */ - if (!fid) { - if (!cpu_khz) { - printk(KERN_WARNING PFX - "cpu_khz not set, can't calculate multiplier!\n"); - return -ENODEV; - } - - fid = cpu_khz / (fsb * 100); - rfid = fid % 5; - - if (rfid) { - if (rfid > 2) - fid += 5 - rfid; - else - fid -= rfid; - } - } - - printk(KERN_INFO PFX "FSB currently at %i MHz, FID %d.%d\n", fsb, - fid / 10, fid % 10); - - /* Set maximum FSB to FSB at boot time */ - max_fsb = nforce2_fsb_read(1); - - if (!max_fsb) - return -EIO; - - if (!min_fsb) - min_fsb = max_fsb - NFORCE2_SAFE_DISTANCE; - - if (min_fsb < NFORCE2_MIN_FSB) - min_fsb = NFORCE2_MIN_FSB; - - /* cpuinfo and default policy values */ - policy->cpuinfo.min_freq = min_fsb * fid * 100; - policy->cpuinfo.max_freq = max_fsb * fid * 100; - policy->cpuinfo.transition_latency = CPUFREQ_ETERNAL; - policy->cur = nforce2_get(policy->cpu); - policy->min = policy->cpuinfo.min_freq; - policy->max = policy->cpuinfo.max_freq; - - return 0; -} - -static int nforce2_cpu_exit(struct cpufreq_policy *policy) -{ - return 0; -} - -static struct cpufreq_driver nforce2_driver = { - .name = "nforce2", - .verify = nforce2_verify, - .target = nforce2_target, - .get = nforce2_get, - .init = nforce2_cpu_init, - .exit = nforce2_cpu_exit, - .owner = THIS_MODULE, -}; - -/** - * nforce2_detect_chipset - detect the Southbridge which contains FSB PLL logic - * - * Detects nForce2 A2 and C1 stepping - * - */ -static int nforce2_detect_chipset(void) -{ - nforce2_dev = pci_get_subsys(PCI_VENDOR_ID_NVIDIA, - PCI_DEVICE_ID_NVIDIA_NFORCE2, - PCI_ANY_ID, PCI_ANY_ID, NULL); - - if (nforce2_dev == NULL) - return -ENODEV; - - printk(KERN_INFO PFX "Detected nForce2 chipset revision %X\n", - nforce2_dev->revision); - printk(KERN_INFO PFX - "FSB changing is maybe unstable and can lead to " - "crashes and data loss.\n"); - - return 0; -} - -/** - * nforce2_init - initializes the nForce2 CPUFreq driver - * - * Initializes the nForce2 FSB support. Returns -ENODEV on unsupported - * devices, -EINVAL on problems during initiatization, and zero on - * success. - */ -static int __init nforce2_init(void) -{ - /* TODO: do we need to detect the processor? */ - - /* detect chipset */ - if (nforce2_detect_chipset()) { - printk(KERN_INFO PFX "No nForce2 chipset.\n"); - return -ENODEV; - } - - return cpufreq_register_driver(&nforce2_driver); -} - -/** - * nforce2_exit - unregisters cpufreq module - * - * Unregisters nForce2 FSB change support. - */ -static void __exit nforce2_exit(void) -{ - cpufreq_unregister_driver(&nforce2_driver); -} - -module_init(nforce2_init); -module_exit(nforce2_exit); - diff --git a/arch/x86/kernel/cpu/cpufreq/e_powersaver.c b/arch/x86/kernel/cpu/cpufreq/e_powersaver.c deleted file mode 100644 index 35a257dd4bb7..000000000000 --- a/arch/x86/kernel/cpu/cpufreq/e_powersaver.c +++ /dev/null @@ -1,367 +0,0 @@ -/* - * Based on documentation provided by Dave Jones. Thanks! - * - * Licensed under the terms of the GNU GPL License version 2. - * - * BIG FAT DISCLAIMER: Work in progress code. Possibly *dangerous* - */ - -#include <linux/kernel.h> -#include <linux/module.h> -#include <linux/init.h> -#include <linux/cpufreq.h> -#include <linux/ioport.h> -#include <linux/slab.h> -#include <linux/timex.h> -#include <linux/io.h> -#include <linux/delay.h> - -#include <asm/msr.h> -#include <asm/tsc.h> - -#define EPS_BRAND_C7M 0 -#define EPS_BRAND_C7 1 -#define EPS_BRAND_EDEN 2 -#define EPS_BRAND_C3 3 -#define EPS_BRAND_C7D 4 - -struct eps_cpu_data { - u32 fsb; - struct cpufreq_frequency_table freq_table[]; -}; - -static struct eps_cpu_data *eps_cpu[NR_CPUS]; - - -static unsigned int eps_get(unsigned int cpu) -{ - struct eps_cpu_data *centaur; - u32 lo, hi; - - if (cpu) - return 0; - centaur = eps_cpu[cpu]; - if (centaur == NULL) - return 0; - - /* Return current frequency */ - rdmsr(MSR_IA32_PERF_STATUS, lo, hi); - return centaur->fsb * ((lo >> 8) & 0xff); -} - -static int eps_set_state(struct eps_cpu_data *centaur, - unsigned int cpu, - u32 dest_state) -{ - struct cpufreq_freqs freqs; - u32 lo, hi; - int err = 0; - int i; - - freqs.old = eps_get(cpu); - freqs.new = centaur->fsb * ((dest_state >> 8) & 0xff); - freqs.cpu = cpu; - cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); - - /* Wait while CPU is busy */ - rdmsr(MSR_IA32_PERF_STATUS, lo, hi); - i = 0; - while (lo & ((1 << 16) | (1 << 17))) { - udelay(16); - rdmsr(MSR_IA32_PERF_STATUS, lo, hi); - i++; - if (unlikely(i > 64)) { - err = -ENODEV; - goto postchange; - } - } - /* Set new multiplier and voltage */ - wrmsr(MSR_IA32_PERF_CTL, dest_state & 0xffff, 0); - /* Wait until transition end */ - i = 0; - do { - udelay(16); - rdmsr(MSR_IA32_PERF_STATUS, lo, hi); - i++; - if (unlikely(i > 64)) { - err = -ENODEV; - goto postchange; - } - } while (lo & ((1 << 16) | (1 << 17))); - - /* Return current frequency */ -postchange: - rdmsr(MSR_IA32_PERF_STATUS, lo, hi); - freqs.new = centaur->fsb * ((lo >> 8) & 0xff); - -#ifdef DEBUG - { - u8 current_multiplier, current_voltage; - - /* Print voltage and multiplier */ - rdmsr(MSR_IA32_PERF_STATUS, lo, hi); - current_voltage = lo & 0xff; - printk(KERN_INFO "eps: Current voltage = %dmV\n", - current_voltage * 16 + 700); - current_multiplier = (lo >> 8) & 0xff; - printk(KERN_INFO "eps: Current multiplier = %d\n", - current_multiplier); - } -#endif - cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); - return err; -} - -static int eps_target(struct cpufreq_policy *policy, - unsigned int target_freq, - unsigned int relation) -{ - struct eps_cpu_data *centaur; - unsigned int newstate = 0; - unsigned int cpu = policy->cpu; - unsigned int dest_state; - int ret; - - if (unlikely(eps_cpu[cpu] == NULL)) - return -ENODEV; - centaur = eps_cpu[cpu]; - - if (unlikely(cpufreq_frequency_table_target(policy, - &eps_cpu[cpu]->freq_table[0], - target_freq, - relation, - &newstate))) { - return -EINVAL; - } - - /* Make frequency transition */ - dest_state = centaur->freq_table[newstate].index & 0xffff; - ret = eps_set_state(centaur, cpu, dest_state); - if (ret) - printk(KERN_ERR "eps: Timeout!\n"); - return ret; -} - -static int eps_verify(struct cpufreq_policy *policy) -{ - return cpufreq_frequency_table_verify(policy, - &eps_cpu[policy->cpu]->freq_table[0]); -} - -static int eps_cpu_init(struct cpufreq_policy *policy) -{ - unsigned int i; - u32 lo, hi; - u64 val; - u8 current_multiplier, current_voltage; - u8 max_multiplier, max_voltage; - u8 min_multiplier, min_voltage; - u8 brand = 0; - u32 fsb; - struct eps_cpu_data *centaur; - struct cpuinfo_x86 *c = &cpu_data(0); - struct cpufreq_frequency_table *f_table; - int k, step, voltage; - int ret; - int states; - - if (policy->cpu != 0) - return -ENODEV; - - /* Check brand */ - printk(KERN_INFO "eps: Detected VIA "); - - switch (c->x86_model) { - case 10: - rdmsr(0x1153, lo, hi); - brand = (((lo >> 2) ^ lo) >> 18) & 3; - printk(KERN_CONT "Model A "); - break; - case 13: - rdmsr(0x1154, lo, hi); - brand = (((lo >> 4) ^ (lo >> 2))) & 0x000000ff; - printk(KERN_CONT "Model D "); - break; - } - - switch (brand) { - case EPS_BRAND_C7M: - printk(KERN_CONT "C7-M\n"); - break; - case EPS_BRAND_C7: - printk(KERN_CONT "C7\n"); - break; - case EPS_BRAND_EDEN: - printk(KERN_CONT "Eden\n"); - break; - case EPS_BRAND_C7D: - printk(KERN_CONT "C7-D\n"); - break; - case EPS_BRAND_C3: - printk(KERN_CONT "C3\n"); - return -ENODEV; - break; - } - /* Enable Enhanced PowerSaver */ - rdmsrl(MSR_IA32_MISC_ENABLE, val); - if (!(val & MSR_IA32_MISC_ENABLE_ENHANCED_SPEEDSTEP)) { - val |= MSR_IA32_MISC_ENABLE_ENHANCED_SPEEDSTEP; - wrmsrl(MSR_IA32_MISC_ENABLE, val); - /* Can be locked at 0 */ - rdmsrl(MSR_IA32_MISC_ENABLE, val); - if (!(val & MSR_IA32_MISC_ENABLE_ENHANCED_SPEEDSTEP)) { - printk(KERN_INFO "eps: Can't enable Enhanced PowerSaver\n"); - return -ENODEV; - } - } - - /* Print voltage and multiplier */ - rdmsr(MSR_IA32_PERF_STATUS, lo, hi); - current_voltage = lo & 0xff; - printk(KERN_INFO "eps: Current voltage = %dmV\n", - current_voltage * 16 + 700); - current_multiplier = (lo >> 8) & 0xff; - printk(KERN_INFO "eps: Current multiplier = %d\n", current_multiplier); - - /* Print limits */ - max_voltage = hi & 0xff; - printk(KERN_INFO "eps: Highest voltage = %dmV\n", - max_voltage * 16 + 700); - max_multiplier = (hi >> 8) & 0xff; - printk(KERN_INFO "eps: Highest multiplier = %d\n", max_multiplier); - min_voltage = (hi >> 16) & 0xff; - printk(KERN_INFO "eps: Lowest voltage = %dmV\n", - min_voltage * 16 + 700); - min_multiplier = (hi >> 24) & 0xff; - printk(KERN_INFO "eps: Lowest multiplier = %d\n", min_multiplier); - - /* Sanity checks */ - if (current_multiplier == 0 || max_multiplier == 0 - || min_multiplier == 0) - return -EINVAL; - if (current_multiplier > max_multiplier - || max_multiplier <= min_multiplier) - return -EINVAL; - if (current_voltage > 0x1f || max_voltage > 0x1f) - return -EINVAL; - if (max_voltage < min_voltage) - return -EINVAL; - - /* Calc FSB speed */ - fsb = cpu_khz / current_multiplier; - /* Calc number of p-states supported */ - if (brand == EPS_BRAND_C7M) - states = max_multiplier - min_multiplier + 1; - else - states = 2; - - /* Allocate private data and frequency table for current cpu */ - centaur = kzalloc(sizeof(struct eps_cpu_data) - + (states + 1) * sizeof(struct cpufreq_frequency_table), - GFP_KERNEL); - if (!centaur) - return -ENOMEM; - eps_cpu[0] = centaur; - - /* Copy basic values */ - centaur->fsb = fsb; - - /* Fill frequency and MSR value table */ - f_table = ¢aur->freq_table[0]; - if (brand != EPS_BRAND_C7M) { - f_table[0].frequency = fsb * min_multiplier; - f_table[0].index = (min_multiplier << 8) | min_voltage; - f_table[1].frequency = fsb * max_multiplier; - f_table[1].index = (max_multiplier << 8) | max_voltage; - f_table[2].frequency = CPUFREQ_TABLE_END; - } else { - k = 0; - step = ((max_voltage - min_voltage) * 256) - / (max_multiplier - min_multiplier); - for (i = min_multiplier; i <= max_multiplier; i++) { - voltage = (k * step) / 256 + min_voltage; - f_table[k].frequency = fsb * i; - f_table[k].index = (i << 8) | voltage; - k++; - } - f_table[k].frequency = CPUFREQ_TABLE_END; - } - - policy->cpuinfo.transition_latency = 140000; /* 844mV -> 700mV in ns */ - policy->cur = fsb * current_multiplier; - - ret = cpufreq_frequency_table_cpuinfo(policy, ¢aur->freq_table[0]); - if (ret) { - kfree(centaur); - return ret; - } - - cpufreq_frequency_table_get_attr(¢aur->freq_table[0], policy->cpu); - return 0; -} - -static int eps_cpu_exit(struct cpufreq_policy *policy) -{ - unsigned int cpu = policy->cpu; - struct eps_cpu_data *centaur; - u32 lo, hi; - - if (eps_cpu[cpu] == NULL) - return -ENODEV; - centaur = eps_cpu[cpu]; - - /* Get max frequency */ - rdmsr(MSR_IA32_PERF_STATUS, lo, hi); - /* Set max frequency */ - eps_set_state(centaur, cpu, hi & 0xffff); - /* Bye */ - cpufreq_frequency_table_put_attr(policy->cpu); - kfree(eps_cpu[cpu]); - eps_cpu[cpu] = NULL; - return 0; -} - -static struct freq_attr *eps_attr[] = { - &cpufreq_freq_attr_scaling_available_freqs, - NULL, -}; - -static struct cpufreq_driver eps_driver = { - .verify = eps_verify, - .target = eps_target, - .init = eps_cpu_init, - .exit = eps_cpu_exit, - .get = eps_get, - .name = "e_powersaver", - .owner = THIS_MODULE, - .attr = eps_attr, -}; - -static int __init eps_init(void) -{ - struct cpuinfo_x86 *c = &cpu_data(0); - - /* This driver will work only on Centaur C7 processors with - * Enhanced SpeedStep/PowerSaver registers */ - if (c->x86_vendor != X86_VENDOR_CENTAUR - || c->x86 != 6 || c->x86_model < 10) - return -ENODEV; - if (!cpu_has(c, X86_FEATURE_EST)) - return -ENODEV; - - if (cpufreq_register_driver(&eps_driver)) - return -EINVAL; - return 0; -} - -static void __exit eps_exit(void) -{ - cpufreq_unregister_driver(&eps_driver); -} - -MODULE_AUTHOR("Rafal Bilski <rafalbilski@interia.pl>"); -MODULE_DESCRIPTION("Enhanced PowerSaver driver for VIA C7 CPU's."); -MODULE_LICENSE("GPL"); - -module_init(eps_init); -module_exit(eps_exit); diff --git a/arch/x86/kernel/cpu/cpufreq/elanfreq.c b/arch/x86/kernel/cpu/cpufreq/elanfreq.c deleted file mode 100644 index c587db472a75..000000000000 --- a/arch/x86/kernel/cpu/cpufreq/elanfreq.c +++ /dev/null @@ -1,309 +0,0 @@ -/* - * elanfreq: cpufreq driver for the AMD ELAN family - * - * (c) Copyright 2002 Robert Schwebel <r.schwebel@pengutronix.de> - * - * Parts of this code are (c) Sven Geggus <sven@geggus.net> - * - * All Rights Reserved. - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - * - * 2002-02-13: - initial revision for 2.4.18-pre9 by Robert Schwebel - * - */ - -#include <linux/kernel.h> -#include <linux/module.h> -#include <linux/init.h> - -#include <linux/delay.h> -#include <linux/cpufreq.h> - -#include <asm/msr.h> -#include <linux/timex.h> -#include <linux/io.h> - -#define REG_CSCIR 0x22 /* Chip Setup and Control Index Register */ -#define REG_CSCDR 0x23 /* Chip Setup and Control Data Register */ - -/* Module parameter */ -static int max_freq; - -struct s_elan_multiplier { - int clock; /* frequency in kHz */ - int val40h; /* PMU Force Mode register */ - int val80h; /* CPU Clock Speed Register */ -}; - -/* - * It is important that the frequencies - * are listed in ascending order here! - */ -static struct s_elan_multiplier elan_multiplier[] = { - {1000, 0x02, 0x18}, - {2000, 0x02, 0x10}, - {4000, 0x02, 0x08}, - {8000, 0x00, 0x00}, - {16000, 0x00, 0x02}, - {33000, 0x00, 0x04}, - {66000, 0x01, 0x04}, - {99000, 0x01, 0x05} -}; - -static struct cpufreq_frequency_table elanfreq_table[] = { - {0, 1000}, - {1, 2000}, - {2, 4000}, - {3, 8000}, - {4, 16000}, - {5, 33000}, - {6, 66000}, - {7, 99000}, - {0, CPUFREQ_TABLE_END}, -}; - - -/** - * elanfreq_get_cpu_frequency: determine current cpu speed - * - * Finds out at which frequency the CPU of the Elan SOC runs - * at the moment. Frequencies from 1 to 33 MHz are generated - * the normal way, 66 and 99 MHz are called "Hyperspeed Mode" - * and have the rest of the chip running with 33 MHz. - */ - -static unsigned int elanfreq_get_cpu_frequency(unsigned int cpu) -{ - u8 clockspeed_reg; /* Clock Speed Register */ - - local_irq_disable(); - outb_p(0x80, REG_CSCIR); - clockspeed_reg = inb_p(REG_CSCDR); - local_irq_enable(); - - if ((clockspeed_reg & 0xE0) == 0xE0) - return 0; - - /* Are we in CPU clock multiplied mode (66/99 MHz)? */ - if ((clockspeed_reg & 0xE0) == 0xC0) { - if ((clockspeed_reg & 0x01) == 0) - return 66000; - else - return 99000; - } - - /* 33 MHz is not 32 MHz... */ - if ((clockspeed_reg & 0xE0) == 0xA0) - return 33000; - - return (1<<((clockspeed_reg & 0xE0) >> 5)) * 1000; -} - - -/** - * elanfreq_set_cpu_frequency: Change the CPU core frequency - * @cpu: cpu number - * @freq: frequency in kHz - * - * This function takes a frequency value and changes the CPU frequency - * according to this. Note that the frequency has to be checked by - * elanfreq_validatespeed() for correctness! - * - * There is no return value. - */ - -static void elanfreq_set_cpu_state(unsigned int state) -{ - struct cpufreq_freqs freqs; - - freqs.old = elanfreq_get_cpu_frequency(0); - freqs.new = elan_multiplier[state].clock; - freqs.cpu = 0; /* elanfreq.c is UP only driver */ - - cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); - - printk(KERN_INFO "elanfreq: attempting to set frequency to %i kHz\n", - elan_multiplier[state].clock); - - - /* - * Access to the Elan's internal registers is indexed via - * 0x22: Chip Setup & Control Register Index Register (CSCI) - * 0x23: Chip Setup & Control Register Data Register (CSCD) - * - */ - - /* - * 0x40 is the Power Management Unit's Force Mode Register. - * Bit 6 enables Hyperspeed Mode (66/100 MHz core frequency) - */ - - local_irq_disable(); - outb_p(0x40, REG_CSCIR); /* Disable hyperspeed mode */ - outb_p(0x00, REG_CSCDR); - local_irq_enable(); /* wait till internal pipelines and */ - udelay(1000); /* buffers have cleaned up */ - - local_irq_disable(); - - /* now, set the CPU clock speed register (0x80) */ - outb_p(0x80, REG_CSCIR); - outb_p(elan_multiplier[state].val80h, REG_CSCDR); - - /* now, the hyperspeed bit in PMU Force Mode Register (0x40) */ - outb_p(0x40, REG_CSCIR); - outb_p(elan_multiplier[state].val40h, REG_CSCDR); - udelay(10000); - local_irq_enable(); - - cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); -}; - - -/** - * elanfreq_validatespeed: test if frequency range is valid - * @policy: the policy to validate - * - * This function checks if a given frequency range in kHz is valid - * for the hardware supported by the driver. - */ - -static int elanfreq_verify(struct cpufreq_policy *policy) -{ - return cpufreq_frequency_table_verify(policy, &elanfreq_table[0]); -} - -static int elanfreq_target(struct cpufreq_policy *policy, - unsigned int target_freq, - unsigned int relation) -{ - unsigned int newstate = 0; - - if (cpufreq_frequency_table_target(policy, &elanfreq_table[0], - target_freq, relation, &newstate)) - return -EINVAL; - - elanfreq_set_cpu_state(newstate); - - return 0; -} - - -/* - * Module init and exit code - */ - -static int elanfreq_cpu_init(struct cpufreq_policy *policy) -{ - struct cpuinfo_x86 *c = &cpu_data(0); - unsigned int i; - int result; - - /* capability check */ - if ((c->x86_vendor != X86_VENDOR_AMD) || - (c->x86 != 4) || (c->x86_model != 10)) - return -ENODEV; - - /* max freq */ - if (!max_freq) - max_freq = elanfreq_get_cpu_frequency(0); - - /* table init */ - for (i = 0; (elanfreq_table[i].frequency != CPUFREQ_TABLE_END); i++) { - if (elanfreq_table[i].frequency > max_freq) - elanfreq_table[i].frequency = CPUFREQ_ENTRY_INVALID; - } - - /* cpuinfo and default policy values */ - policy->cpuinfo.transition_latency = CPUFREQ_ETERNAL; - policy->cur = elanfreq_get_cpu_frequency(0); - - result = cpufreq_frequency_table_cpuinfo(policy, elanfreq_table); - if (result) - return result; - - cpufreq_frequency_table_get_attr(elanfreq_table, policy->cpu); - return 0; -} - - -static int elanfreq_cpu_exit(struct cpufreq_policy *policy) -{ - cpufreq_frequency_table_put_attr(policy->cpu); - return 0; -} - - -#ifndef MODULE -/** - * elanfreq_setup - elanfreq command line parameter parsing - * - * elanfreq command line parameter. Use: - * elanfreq=66000 - * to set the maximum CPU frequency to 66 MHz. Note that in - * case you do not give this boot parameter, the maximum - * frequency will fall back to _current_ CPU frequency which - * might be lower. If you build this as a module, use the - * max_freq module parameter instead. - */ -static int __init elanfreq_setup(char *str) -{ - max_freq = simple_strtoul(str, &str, 0); - printk(KERN_WARNING "You're using the deprecated elanfreq command line option. Use elanfreq.max_freq instead, please!\n"); - return 1; -} -__setup("elanfreq=", elanfreq_setup); -#endif - - -static struct freq_attr *elanfreq_attr[] = { - &cpufreq_freq_attr_scaling_available_freqs, - NULL, -}; - - -static struct cpufreq_driver elanfreq_driver = { - .get = elanfreq_get_cpu_frequency, - .verify = elanfreq_verify, - .target = elanfreq_target, - .init = elanfreq_cpu_init, - .exit = elanfreq_cpu_exit, - .name = "elanfreq", - .owner = THIS_MODULE, - .attr = elanfreq_attr, -}; - - -static int __init elanfreq_init(void) -{ - struct cpuinfo_x86 *c = &cpu_data(0); - - /* Test if we have the right hardware */ - if ((c->x86_vendor != X86_VENDOR_AMD) || - (c->x86 != 4) || (c->x86_model != 10)) { - printk(KERN_INFO "elanfreq: error: no Elan processor found!\n"); - return -ENODEV; - } - return cpufreq_register_driver(&elanfreq_driver); -} - - -static void __exit elanfreq_exit(void) -{ - cpufreq_unregister_driver(&elanfreq_driver); -} - - -module_param(max_freq, int, 0444); - -MODULE_LICENSE("GPL"); -MODULE_AUTHOR("Robert Schwebel <r.schwebel@pengutronix.de>, " - "Sven Geggus <sven@geggus.net>"); -MODULE_DESCRIPTION("cpufreq driver for AMD's Elan CPUs"); - -module_init(elanfreq_init); -module_exit(elanfreq_exit); diff --git a/arch/x86/kernel/cpu/cpufreq/gx-suspmod.c b/arch/x86/kernel/cpu/cpufreq/gx-suspmod.c deleted file mode 100644 index 32974cf84232..000000000000 --- a/arch/x86/kernel/cpu/cpufreq/gx-suspmod.c +++ /dev/null @@ -1,517 +0,0 @@ -/* - * Cyrix MediaGX and NatSemi Geode Suspend Modulation - * (C) 2002 Zwane Mwaikambo <zwane@commfireservices.com> - * (C) 2002 Hiroshi Miura <miura@da-cha.org> - * All Rights Reserved - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * version 2 as published by the Free Software Foundation - * - * The author(s) of this software shall not be held liable for damages - * of any nature resulting due to the use of this software. This - * software is provided AS-IS with no warranties. - * - * Theoretical note: - * - * (see Geode(tm) CS5530 manual (rev.4.1) page.56) - * - * CPU frequency control on NatSemi Geode GX1/GXLV processor and CS55x0 - * are based on Suspend Modulation. - * - * Suspend Modulation works by asserting and de-asserting the SUSP# pin - * to CPU(GX1/GXLV) for configurable durations. When asserting SUSP# - * the CPU enters an idle state. GX1 stops its core clock when SUSP# is - * asserted then power consumption is reduced. - * - * Suspend Modulation's OFF/ON duration are configurable - * with 'Suspend Modulation OFF Count Register' - * and 'Suspend Modulation ON Count Register'. - * These registers are 8bit counters that represent the number of - * 32us intervals which the SUSP# pin is asserted(ON)/de-asserted(OFF) - * to the processor. - * - * These counters define a ratio which is the effective frequency - * of operation of the system. - * - * OFF Count - * F_eff = Fgx * ---------------------- - * OFF Count + ON Count - * - * 0 <= On Count, Off Count <= 255 - * - * From these limits, we can get register values - * - * off_duration + on_duration <= MAX_DURATION - * on_duration = off_duration * (stock_freq - freq) / freq - * - * off_duration = (freq * DURATION) / stock_freq - * on_duration = DURATION - off_duration - * - * - *--------------------------------------------------------------------------- - * - * ChangeLog: - * Dec. 12, 2003 Hiroshi Miura <miura@da-cha.org> - * - fix on/off register mistake - * - fix cpu_khz calc when it stops cpu modulation. - * - * Dec. 11, 2002 Hiroshi Miura <miura@da-cha.org> - * - rewrite for Cyrix MediaGX Cx5510/5520 and - * NatSemi Geode Cs5530(A). - * - * Jul. ??, 2002 Zwane Mwaikambo <zwane@commfireservices.com> - * - cs5530_mod patch for 2.4.19-rc1. - * - *--------------------------------------------------------------------------- - * - * Todo - * Test on machines with 5510, 5530, 5530A - */ - -/************************************************************************ - * Suspend Modulation - Definitions * - ************************************************************************/ - -#include <linux/kernel.h> -#include <linux/module.h> -#include <linux/init.h> -#include <linux/smp.h> -#include <linux/cpufreq.h> -#include <linux/pci.h> -#include <linux/errno.h> -#include <linux/slab.h> - -#include <asm/processor-cyrix.h> - -/* PCI config registers, all at F0 */ -#define PCI_PMER1 0x80 /* power management enable register 1 */ -#define PCI_PMER2 0x81 /* power management enable register 2 */ -#define PCI_PMER3 0x82 /* power management enable register 3 */ -#define PCI_IRQTC 0x8c /* irq speedup timer counter register:typical 2 to 4ms */ -#define PCI_VIDTC 0x8d /* video speedup timer counter register: typical 50 to 100ms */ -#define PCI_MODOFF 0x94 /* suspend modulation OFF counter register, 1 = 32us */ -#define PCI_MODON 0x95 /* suspend modulation ON counter register */ -#define PCI_SUSCFG 0x96 /* suspend configuration register */ - -/* PMER1 bits */ -#define GPM (1<<0) /* global power management */ -#define GIT (1<<1) /* globally enable PM device idle timers */ -#define GTR (1<<2) /* globally enable IO traps */ -#define IRQ_SPDUP (1<<3) /* disable clock throttle during interrupt handling */ -#define VID_SPDUP (1<<4) /* disable clock throttle during vga video handling */ - -/* SUSCFG bits */ -#define SUSMOD (1<<0) /* enable/disable suspend modulation */ -/* the below is supported only with cs5530 (after rev.1.2)/cs5530A */ -#define SMISPDUP (1<<1) /* select how SMI re-enable suspend modulation: */ - /* IRQTC timer or read SMI speedup disable reg.(F1BAR[08-09h]) */ -#define SUSCFG (1<<2) /* enable powering down a GXLV processor. "Special 3Volt Suspend" mode */ -/* the below is supported only with cs5530A */ -#define PWRSVE_ISA (1<<3) /* stop ISA clock */ -#define PWRSVE (1<<4) /* active idle */ - -struct gxfreq_params { - u8 on_duration; - u8 off_duration; - u8 pci_suscfg; - u8 pci_pmer1; - u8 pci_pmer2; - struct pci_dev *cs55x0; -}; - -static struct gxfreq_params *gx_params; -static int stock_freq; - -/* PCI bus clock - defaults to 30.000 if cpu_khz is not available */ -static int pci_busclk; -module_param(pci_busclk, int, 0444); - -/* maximum duration for which the cpu may be suspended - * (32us * MAX_DURATION). If no parameter is given, this defaults - * to 255. - * Note that this leads to a maximum of 8 ms(!) where the CPU clock - * is suspended -- processing power is just 0.39% of what it used to be, - * though. 781.25 kHz(!) for a 200 MHz processor -- wow. */ -static int max_duration = 255; -module_param(max_duration, int, 0444); - -/* For the default policy, we want at least some processing power - * - let's say 5%. (min = maxfreq / POLICY_MIN_DIV) - */ -#define POLICY_MIN_DIV 20 - - -#define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, \ - "gx-suspmod", msg) - -/** - * we can detect a core multipiler from dir0_lsb - * from GX1 datasheet p.56, - * MULT[3:0]: - * 0000 = SYSCLK multiplied by 4 (test only) - * 0001 = SYSCLK multiplied by 10 - * 0010 = SYSCLK multiplied by 4 - * 0011 = SYSCLK multiplied by 6 - * 0100 = SYSCLK multiplied by 9 - * 0101 = SYSCLK multiplied by 5 - * 0110 = SYSCLK multiplied by 7 - * 0111 = SYSCLK multiplied by 8 - * of 33.3MHz - **/ -static int gx_freq_mult[16] = { - 4, 10, 4, 6, 9, 5, 7, 8, - 0, 0, 0, 0, 0, 0, 0, 0 -}; - - -/**************************************************************** - * Low Level chipset interface * - ****************************************************************/ -static struct pci_device_id gx_chipset_tbl[] __initdata = { - { PCI_VDEVICE(CYRIX, PCI_DEVICE_ID_CYRIX_5530_LEGACY), }, - { PCI_VDEVICE(CYRIX, PCI_DEVICE_ID_CYRIX_5520), }, - { PCI_VDEVICE(CYRIX, PCI_DEVICE_ID_CYRIX_5510), }, - { 0, }, -}; - -static void gx_write_byte(int reg, int value) -{ - pci_write_config_byte(gx_params->cs55x0, reg, value); -} - -/** - * gx_detect_chipset: - * - **/ -static __init struct pci_dev *gx_detect_chipset(void) -{ - struct pci_dev *gx_pci = NULL; - - /* check if CPU is a MediaGX or a Geode. */ - if ((boot_cpu_data.x86_vendor != X86_VENDOR_NSC) && - (boot_cpu_data.x86_vendor != X86_VENDOR_CYRIX)) { - dprintk("error: no MediaGX/Geode processor found!\n"); - return NULL; - } - - /* detect which companion chip is used */ - for_each_pci_dev(gx_pci) { - if ((pci_match_id(gx_chipset_tbl, gx_pci)) != NULL) - return gx_pci; - } - - dprintk("error: no supported chipset found!\n"); - return NULL; -} - -/** - * gx_get_cpuspeed: - * - * Finds out at which efficient frequency the Cyrix MediaGX/NatSemi - * Geode CPU runs. - */ -static unsigned int gx_get_cpuspeed(unsigned int cpu) -{ - if ((gx_params->pci_suscfg & SUSMOD) == 0) - return stock_freq; - - return (stock_freq * gx_params->off_duration) - / (gx_params->on_duration + gx_params->off_duration); -} - -/** - * gx_validate_speed: - * determine current cpu speed - * - **/ - -static unsigned int gx_validate_speed(unsigned int khz, u8 *on_duration, - u8 *off_duration) -{ - unsigned int i; - u8 tmp_on, tmp_off; - int old_tmp_freq = stock_freq; - int tmp_freq; - - *off_duration = 1; - *on_duration = 0; - - for (i = max_duration; i > 0; i--) { - tmp_off = ((khz * i) / stock_freq) & 0xff; - tmp_on = i - tmp_off; - tmp_freq = (stock_freq * tmp_off) / i; - /* if this relation is closer to khz, use this. If it's equal, - * prefer it, too - lower latency */ - if (abs(tmp_freq - khz) <= abs(old_tmp_freq - khz)) { - *on_duration = tmp_on; - *off_duration = tmp_off; - old_tmp_freq = tmp_freq; - } - } - - return old_tmp_freq; -} - - -/** - * gx_set_cpuspeed: - * set cpu speed in khz. - **/ - -static void gx_set_cpuspeed(unsigned int khz) -{ - u8 suscfg, pmer1; - unsigned int new_khz; - unsigned long flags; - struct cpufreq_freqs freqs; - - freqs.cpu = 0; - freqs.old = gx_get_cpuspeed(0); - - new_khz = gx_validate_speed(khz, &gx_params->on_duration, - &gx_params->off_duration); - - freqs.new = new_khz; - - cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); - local_irq_save(flags); - - - - if (new_khz != stock_freq) { - /* if new khz == 100% of CPU speed, it is special case */ - switch (gx_params->cs55x0->device) { - case PCI_DEVICE_ID_CYRIX_5530_LEGACY: - pmer1 = gx_params->pci_pmer1 | IRQ_SPDUP | VID_SPDUP; - /* FIXME: need to test other values -- Zwane,Miura */ - /* typical 2 to 4ms */ - gx_write_byte(PCI_IRQTC, 4); - /* typical 50 to 100ms */ - gx_write_byte(PCI_VIDTC, 100); - gx_write_byte(PCI_PMER1, pmer1); - - if (gx_params->cs55x0->revision < 0x10) { - /* CS5530(rev 1.2, 1.3) */ - suscfg = gx_params->pci_suscfg|SUSMOD; - } else { - /* CS5530A,B.. */ - suscfg = gx_params->pci_suscfg|SUSMOD|PWRSVE; - } - break; - case PCI_DEVICE_ID_CYRIX_5520: - case PCI_DEVICE_ID_CYRIX_5510: - suscfg = gx_params->pci_suscfg | SUSMOD; - break; - default: - local_irq_restore(flags); - dprintk("fatal: try to set unknown chipset.\n"); - return; - } - } else { - suscfg = gx_params->pci_suscfg & ~(SUSMOD); - gx_params->off_duration = 0; - gx_params->on_duration = 0; - dprintk("suspend modulation disabled: cpu runs 100%% speed.\n"); - } - - gx_write_byte(PCI_MODOFF, gx_params->off_duration); - gx_write_byte(PCI_MODON, gx_params->on_duration); - - gx_write_byte(PCI_SUSCFG, suscfg); - pci_read_config_byte(gx_params->cs55x0, PCI_SUSCFG, &suscfg); - - local_irq_restore(flags); - - gx_params->pci_suscfg = suscfg; - - cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); - - dprintk("suspend modulation w/ duration of ON:%d us, OFF:%d us\n", - gx_params->on_duration * 32, gx_params->off_duration * 32); - dprintk("suspend modulation w/ clock speed: %d kHz.\n", freqs.new); -} - -/**************************************************************** - * High level functions * - ****************************************************************/ - -/* - * cpufreq_gx_verify: test if frequency range is valid - * - * This function checks if a given frequency range in kHz is valid - * for the hardware supported by the driver. - */ - -static int cpufreq_gx_verify(struct cpufreq_policy *policy) -{ - unsigned int tmp_freq = 0; - u8 tmp1, tmp2; - - if (!stock_freq || !policy) - return -EINVAL; - - policy->cpu = 0; - cpufreq_verify_within_limits(policy, (stock_freq / max_duration), - stock_freq); - - /* it needs to be assured that at least one supported frequency is - * within policy->min and policy->max. If it is not, policy->max - * needs to be increased until one freuqency is supported. - * policy->min may not be decreased, though. This way we guarantee a - * specific processing capacity. - */ - tmp_freq = gx_validate_speed(policy->min, &tmp1, &tmp2); - if (tmp_freq < policy->min) - tmp_freq += stock_freq / max_duration; - policy->min = tmp_freq; - if (policy->min > policy->max) - policy->max = tmp_freq; - tmp_freq = gx_validate_speed(policy->max, &tmp1, &tmp2); - if (tmp_freq > policy->max) - tmp_freq -= stock_freq / max_duration; - policy->max = tmp_freq; - if (policy->max < policy->min) - policy->max = policy->min; - cpufreq_verify_within_limits(policy, (stock_freq / max_duration), - stock_freq); - - return 0; -} - -/* - * cpufreq_gx_target: - * - */ -static int cpufreq_gx_target(struct cpufreq_policy *policy, - unsigned int target_freq, - unsigned int relation) -{ - u8 tmp1, tmp2; - unsigned int tmp_freq; - - if (!stock_freq || !policy) - return -EINVAL; - - policy->cpu = 0; - - tmp_freq = gx_validate_speed(target_freq, &tmp1, &tmp2); - while (tmp_freq < policy->min) { - tmp_freq += stock_freq / max_duration; - tmp_freq = gx_validate_speed(tmp_freq, &tmp1, &tmp2); - } - while (tmp_freq > policy->max) { - tmp_freq -= stock_freq / max_duration; - tmp_freq = gx_validate_speed(tmp_freq, &tmp1, &tmp2); - } - - gx_set_cpuspeed(tmp_freq); - - return 0; -} - -static int cpufreq_gx_cpu_init(struct cpufreq_policy *policy) -{ - unsigned int maxfreq, curfreq; - - if (!policy || policy->cpu != 0) - return -ENODEV; - - /* determine maximum frequency */ - if (pci_busclk) - maxfreq = pci_busclk * gx_freq_mult[getCx86(CX86_DIR1) & 0x0f]; - else if (cpu_khz) - maxfreq = cpu_khz; - else - maxfreq = 30000 * gx_freq_mult[getCx86(CX86_DIR1) & 0x0f]; - - stock_freq = maxfreq; - curfreq = gx_get_cpuspeed(0); - - dprintk("cpu max frequency is %d.\n", maxfreq); - dprintk("cpu current frequency is %dkHz.\n", curfreq); - - /* setup basic struct for cpufreq API */ - policy->cpu = 0; - - if (max_duration < POLICY_MIN_DIV) - policy->min = maxfreq / max_duration; - else - policy->min = maxfreq / POLICY_MIN_DIV; - policy->max = maxfreq; - policy->cur = curfreq; - policy->cpuinfo.min_freq = maxfreq / max_duration; - policy->cpuinfo.max_freq = maxfreq; - policy->cpuinfo.transition_latency = CPUFREQ_ETERNAL; - - return 0; -} - -/* - * cpufreq_gx_init: - * MediaGX/Geode GX initialize cpufreq driver - */ -static struct cpufreq_driver gx_suspmod_driver = { - .get = gx_get_cpuspeed, - .verify = cpufreq_gx_verify, - .target = cpufreq_gx_target, - .init = cpufreq_gx_cpu_init, - .name = "gx-suspmod", - .owner = THIS_MODULE, -}; - -static int __init cpufreq_gx_init(void) -{ - int ret; - struct gxfreq_params *params; - struct pci_dev *gx_pci; - - /* Test if we have the right hardware */ - gx_pci = gx_detect_chipset(); - if (gx_pci == NULL) - return -ENODEV; - - /* check whether module parameters are sane */ - if (max_duration > 0xff) - max_duration = 0xff; - - dprintk("geode suspend modulation available.\n"); - - params = kzalloc(sizeof(struct gxfreq_params), GFP_KERNEL); - if (params == NULL) - return -ENOMEM; - - params->cs55x0 = gx_pci; - gx_params = params; - - /* keep cs55x0 configurations */ - pci_read_config_byte(params->cs55x0, PCI_SUSCFG, &(params->pci_suscfg)); - pci_read_config_byte(params->cs55x0, PCI_PMER1, &(params->pci_pmer1)); - pci_read_config_byte(params->cs55x0, PCI_PMER2, &(params->pci_pmer2)); - pci_read_config_byte(params->cs55x0, PCI_MODON, &(params->on_duration)); - pci_read_config_byte(params->cs55x0, PCI_MODOFF, - &(params->off_duration)); - - ret = cpufreq_register_driver(&gx_suspmod_driver); - if (ret) { - kfree(params); - return ret; /* register error! */ - } - - return 0; -} - -static void __exit cpufreq_gx_exit(void) -{ - cpufreq_unregister_driver(&gx_suspmod_driver); - pci_dev_put(gx_params->cs55x0); - kfree(gx_params); -} - -MODULE_AUTHOR("Hiroshi Miura <miura@da-cha.org>"); -MODULE_DESCRIPTION("Cpufreq driver for Cyrix MediaGX and NatSemi Geode"); -MODULE_LICENSE("GPL"); - -module_init(cpufreq_gx_init); -module_exit(cpufreq_gx_exit); - diff --git a/arch/x86/kernel/cpu/cpufreq/longhaul.c b/arch/x86/kernel/cpu/cpufreq/longhaul.c deleted file mode 100644 index cf48cdd6907d..000000000000 --- a/arch/x86/kernel/cpu/cpufreq/longhaul.c +++ /dev/null @@ -1,1029 +0,0 @@ -/* - * (C) 2001-2004 Dave Jones. <davej@redhat.com> - * (C) 2002 Padraig Brady. <padraig@antefacto.com> - * - * Licensed under the terms of the GNU GPL License version 2. - * Based upon datasheets & sample CPUs kindly provided by VIA. - * - * VIA have currently 3 different versions of Longhaul. - * Version 1 (Longhaul) uses the BCR2 MSR at 0x1147. - * It is present only in Samuel 1 (C5A), Samuel 2 (C5B) stepping 0. - * Version 2 of longhaul is backward compatible with v1, but adds - * LONGHAUL MSR for purpose of both frequency and voltage scaling. - * Present in Samuel 2 (steppings 1-7 only) (C5B), and Ezra (C5C). - * Version 3 of longhaul got renamed to Powersaver and redesigned - * to use only the POWERSAVER MSR at 0x110a. - * It is present in Ezra-T (C5M), Nehemiah (C5X) and above. - * It's pretty much the same feature wise to longhaul v2, though - * there is provision for scaling FSB too, but this doesn't work - * too well in practice so we don't even try to use this. - * - * BIG FAT DISCLAIMER: Work in progress code. Possibly *dangerous* - */ - -#include <linux/kernel.h> -#include <linux/module.h> -#include <linux/moduleparam.h> -#include <linux/init.h> -#include <linux/cpufreq.h> -#include <linux/pci.h> -#include <linux/slab.h> -#include <linux/string.h> -#include <linux/delay.h> -#include <linux/timex.h> -#include <linux/io.h> -#include <linux/acpi.h> - -#include <asm/msr.h> -#include <acpi/processor.h> - -#include "longhaul.h" - -#define PFX "longhaul: " - -#define TYPE_LONGHAUL_V1 1 -#define TYPE_LONGHAUL_V2 2 -#define TYPE_POWERSAVER 3 - -#define CPU_SAMUEL 1 -#define CPU_SAMUEL2 2 -#define CPU_EZRA 3 -#define CPU_EZRA_T 4 -#define CPU_NEHEMIAH 5 -#define CPU_NEHEMIAH_C 6 - -/* Flags */ -#define USE_ACPI_C3 (1 << 1) -#define USE_NORTHBRIDGE (1 << 2) - -static int cpu_model; -static unsigned int numscales = 16; -static unsigned int fsb; - -static const struct mV_pos *vrm_mV_table; -static const unsigned char *mV_vrm_table; - -static unsigned int highest_speed, lowest_speed; /* kHz */ -static unsigned int minmult, maxmult; -static int can_scale_voltage; -static struct acpi_processor *pr; -static struct acpi_processor_cx *cx; -static u32 acpi_regs_addr; -static u8 longhaul_flags; -static unsigned int longhaul_index; - -/* Module parameters */ -static int scale_voltage; -static int disable_acpi_c3; -static int revid_errata; - -#define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, \ - "longhaul", msg) - - -/* Clock ratios multiplied by 10 */ -static int mults[32]; -static int eblcr[32]; -static int longhaul_version; -static struct cpufreq_frequency_table *longhaul_table; - -#ifdef CONFIG_CPU_FREQ_DEBUG -static char speedbuffer[8]; - -static char *print_speed(int speed) -{ - if (speed < 1000) { - snprintf(speedbuffer, sizeof(speedbuffer), "%dMHz", speed); - return speedbuffer; - } - - if (speed%1000 == 0) - snprintf(speedbuffer, sizeof(speedbuffer), - "%dGHz", speed/1000); - else - snprintf(speedbuffer, sizeof(speedbuffer), - "%d.%dGHz", speed/1000, (speed%1000)/100); - - return speedbuffer; -} -#endif - - -static unsigned int calc_speed(int mult) -{ - int khz; - khz = (mult/10)*fsb; - if (mult%10) - khz += fsb/2; - khz *= 1000; - return khz; -} - - -static int longhaul_get_cpu_mult(void) -{ - unsigned long invalue = 0, lo, hi; - - rdmsr(MSR_IA32_EBL_CR_POWERON, lo, hi); - invalue = (lo & (1<<22|1<<23|1<<24|1<<25))>>22; - if (longhaul_version == TYPE_LONGHAUL_V2 || - longhaul_version == TYPE_POWERSAVER) { - if (lo & (1<<27)) - invalue += 16; - } - return eblcr[invalue]; -} - -/* For processor with BCR2 MSR */ - -static void do_longhaul1(unsigned int mults_index) -{ - union msr_bcr2 bcr2; - - rdmsrl(MSR_VIA_BCR2, bcr2.val); - /* Enable software clock multiplier */ - bcr2.bits.ESOFTBF = 1; - bcr2.bits.CLOCKMUL = mults_index & 0xff; - - /* Sync to timer tick */ - safe_halt(); - /* Change frequency on next halt or sleep */ - wrmsrl(MSR_VIA_BCR2, bcr2.val); - /* Invoke transition */ - ACPI_FLUSH_CPU_CACHE(); - halt(); - - /* Disable software clock multiplier */ - local_irq_disable(); - rdmsrl(MSR_VIA_BCR2, bcr2.val); - bcr2.bits.ESOFTBF = 0; - wrmsrl(MSR_VIA_BCR2, bcr2.val); -} - -/* For processor with Longhaul MSR */ - -static void do_powersaver(int cx_address, unsigned int mults_index, - unsigned int dir) -{ - union msr_longhaul longhaul; - u32 t; - - rdmsrl(MSR_VIA_LONGHAUL, longhaul.val); - /* Setup new frequency */ - if (!revid_errata) - longhaul.bits.RevisionKey = longhaul.bits.RevisionID; - else - longhaul.bits.RevisionKey = 0; - longhaul.bits.SoftBusRatio = mults_index & 0xf; - longhaul.bits.SoftBusRatio4 = (mults_index & 0x10) >> 4; - /* Setup new voltage */ - if (can_scale_voltage) - longhaul.bits.SoftVID = (mults_index >> 8) & 0x1f; - /* Sync to timer tick */ - safe_halt(); - /* Raise voltage if necessary */ - if (can_scale_voltage && dir) { - longhaul.bits.EnableSoftVID = 1; - wrmsrl(MSR_VIA_LONGHAUL, longhaul.val); - /* Change voltage */ - if (!cx_address) { - ACPI_FLUSH_CPU_CACHE(); - halt(); - } else { - ACPI_FLUSH_CPU_CACHE(); - /* Invoke C3 */ - inb(cx_address); - /* Dummy op - must do something useless after P_LVL3 - * read */ - t = inl(acpi_gbl_FADT.xpm_timer_block.address); - } - longhaul.bits.EnableSoftVID = 0; - wrmsrl(MSR_VIA_LONGHAUL, longhaul.val); - } - - /* Change frequency on next halt or sleep */ - longhaul.bits.EnableSoftBusRatio = 1; - wrmsrl(MSR_VIA_LONGHAUL, longhaul.val); - if (!cx_address) { - ACPI_FLUSH_CPU_CACHE(); - halt(); - } else { - ACPI_FLUSH_CPU_CACHE(); - /* Invoke C3 */ - inb(cx_address); - /* Dummy op - must do something useless after P_LVL3 read */ - t = inl(acpi_gbl_FADT.xpm_timer_block.address); - } - /* Disable bus ratio bit */ - longhaul.bits.EnableSoftBusRatio = 0; - wrmsrl(MSR_VIA_LONGHAUL, longhaul.val); - - /* Reduce voltage if necessary */ - if (can_scale_voltage && !dir) { - longhaul.bits.EnableSoftVID = 1; - wrmsrl(MSR_VIA_LONGHAUL, longhaul.val); - /* Change voltage */ - if (!cx_address) { - ACPI_FLUSH_CPU_CACHE(); - halt(); - } else { - ACPI_FLUSH_CPU_CACHE(); - /* Invoke C3 */ - inb(cx_address); - /* Dummy op - must do something useless after P_LVL3 - * read */ - t = inl(acpi_gbl_FADT.xpm_timer_block.address); - } - longhaul.bits.EnableSoftVID = 0; - wrmsrl(MSR_VIA_LONGHAUL, longhaul.val); - } -} - -/** - * longhaul_set_cpu_frequency() - * @mults_index : bitpattern of the new multiplier. - * - * Sets a new clock ratio. - */ - -static void longhaul_setstate(unsigned int table_index) -{ - unsigned int mults_index; - int speed, mult; - struct cpufreq_freqs freqs; - unsigned long flags; - unsigned int pic1_mask, pic2_mask; - u16 bm_status = 0; - u32 bm_timeout = 1000; - unsigned int dir = 0; - - mults_index = longhaul_table[table_index].index; - /* Safety precautions */ - mult = mults[mults_index & 0x1f]; - if (mult == -1) - return; - speed = calc_speed(mult); - if ((speed > highest_speed) || (speed < lowest_speed)) - return; - /* Voltage transition before frequency transition? */ - if (can_scale_voltage && longhaul_index < table_index) - dir = 1; - - freqs.old = calc_speed(longhaul_get_cpu_mult()); - freqs.new = speed; - freqs.cpu = 0; /* longhaul.c is UP only driver */ - - cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); - - dprintk("Setting to FSB:%dMHz Mult:%d.%dx (%s)\n", - fsb, mult/10, mult%10, print_speed(speed/1000)); -retry_loop: - preempt_disable(); - local_irq_save(flags); - - pic2_mask = inb(0xA1); - pic1_mask = inb(0x21); /* works on C3. save mask. */ - outb(0xFF, 0xA1); /* Overkill */ - outb(0xFE, 0x21); /* TMR0 only */ - - /* Wait while PCI bus is busy. */ - if (acpi_regs_addr && (longhaul_flags & USE_NORTHBRIDGE - || ((pr != NULL) && pr->flags.bm_control))) { - bm_status = inw(acpi_regs_addr); - bm_status &= 1 << 4; - while (bm_status && bm_timeout) { - outw(1 << 4, acpi_regs_addr); - bm_timeout--; - bm_status = inw(acpi_regs_addr); - bm_status &= 1 << 4; - } - } - - if (longhaul_flags & USE_NORTHBRIDGE) { - /* Disable AGP and PCI arbiters */ - outb(3, 0x22); - } else if ((pr != NULL) && pr->flags.bm_control) { - /* Disable bus master arbitration */ - acpi_write_bit_register(ACPI_BITREG_ARB_DISABLE, 1); - } - switch (longhaul_version) { - - /* - * Longhaul v1. (Samuel[C5A] and Samuel2 stepping 0[C5B]) - * Software controlled multipliers only. - */ - case TYPE_LONGHAUL_V1: - do_longhaul1(mults_index); - break; - - /* - * Longhaul v2 appears in Samuel2 Steppings 1->7 [C5B] and Ezra [C5C] - * - * Longhaul v3 (aka Powersaver). (Ezra-T [C5M] & Nehemiah [C5N]) - * Nehemiah can do FSB scaling too, but this has never been proven - * to work in practice. - */ - case TYPE_LONGHAUL_V2: - case TYPE_POWERSAVER: - if (longhaul_flags & USE_ACPI_C3) { - /* Don't allow wakeup */ - acpi_write_bit_register(ACPI_BITREG_BUS_MASTER_RLD, 0); - do_powersaver(cx->address, mults_index, dir); - } else { - do_powersaver(0, mults_index, dir); - } - break; - } - - if (longhaul_flags & USE_NORTHBRIDGE) { - /* Enable arbiters */ - outb(0, 0x22); - } else if ((pr != NULL) && pr->flags.bm_control) { - /* Enable bus master arbitration */ - acpi_write_bit_register(ACPI_BITREG_ARB_DISABLE, 0); - } - outb(pic2_mask, 0xA1); /* restore mask */ - outb(pic1_mask, 0x21); - - local_irq_restore(flags); - preempt_enable(); - - freqs.new = calc_speed(longhaul_get_cpu_mult()); - /* Check if requested frequency is set. */ - if (unlikely(freqs.new != speed)) { - printk(KERN_INFO PFX "Failed to set requested frequency!\n"); - /* Revision ID = 1 but processor is expecting revision key - * equal to 0. Jumpers at the bottom of processor will change - * multiplier and FSB, but will not change bits in Longhaul - * MSR nor enable voltage scaling. */ - if (!revid_errata) { - printk(KERN_INFO PFX "Enabling \"Ignore Revision ID\" " - "option.\n"); - revid_errata = 1; - msleep(200); - goto retry_loop; - } - /* Why ACPI C3 sometimes doesn't work is a mystery for me. - * But it does happen. Processor is entering ACPI C3 state, - * but it doesn't change frequency. I tried poking various - * bits in northbridge registers, but without success. */ - if (longhaul_flags & USE_ACPI_C3) { - printk(KERN_INFO PFX "Disabling ACPI C3 support.\n"); - longhaul_flags &= ~USE_ACPI_C3; - if (revid_errata) { - printk(KERN_INFO PFX "Disabling \"Ignore " - "Revision ID\" option.\n"); - revid_errata = 0; - } - msleep(200); - goto retry_loop; - } - /* This shouldn't happen. Longhaul ver. 2 was reported not - * working on processors without voltage scaling, but with - * RevID = 1. RevID errata will make things right. Just - * to be 100% sure. */ - if (longhaul_version == TYPE_LONGHAUL_V2) { - printk(KERN_INFO PFX "Switching to Longhaul ver. 1\n"); - longhaul_version = TYPE_LONGHAUL_V1; - msleep(200); - goto retry_loop; - } - } - /* Report true CPU frequency */ - cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); - - if (!bm_timeout) - printk(KERN_INFO PFX "Warning: Timeout while waiting for " - "idle PCI bus.\n"); -} - -/* - * Centaur decided to make life a little more tricky. - * Only longhaul v1 is allowed to read EBLCR BSEL[0:1]. - * Samuel2 and above have to try and guess what the FSB is. - * We do this by assuming we booted at maximum multiplier, and interpolate - * between that value multiplied by possible FSBs and cpu_mhz which - * was calculated at boot time. Really ugly, but no other way to do this. - */ - -#define ROUNDING 0xf - -static int guess_fsb(int mult) -{ - int speed = cpu_khz / 1000; - int i; - int speeds[] = { 666, 1000, 1333, 2000 }; - int f_max, f_min; - - for (i = 0; i < 4; i++) { - f_max = ((speeds[i] * mult) + 50) / 100; - f_max += (ROUNDING / 2); - f_min = f_max - ROUNDING; - if ((speed <= f_max) && (speed >= f_min)) - return speeds[i] / 10; - } - return 0; -} - - -static int __cpuinit longhaul_get_ranges(void) -{ - unsigned int i, j, k = 0; - unsigned int ratio; - int mult; - - /* Get current frequency */ - mult = longhaul_get_cpu_mult(); - if (mult == -1) { - printk(KERN_INFO PFX "Invalid (reserved) multiplier!\n"); - return -EINVAL; - } - fsb = guess_fsb(mult); - if (fsb == 0) { - printk(KERN_INFO PFX "Invalid (reserved) FSB!\n"); - return -EINVAL; - } - /* Get max multiplier - as we always did. - * Longhaul MSR is useful only when voltage scaling is enabled. - * C3 is booting at max anyway. */ - maxmult = mult; - /* Get min multiplier */ - switch (cpu_model) { - case CPU_NEHEMIAH: - minmult = 50; - break; - case CPU_NEHEMIAH_C: - minmult = 40; - break; - default: - minmult = 30; - break; - } - - dprintk("MinMult:%d.%dx MaxMult:%d.%dx\n", - minmult/10, minmult%10, maxmult/10, maxmult%10); - - highest_speed = calc_speed(maxmult); - lowest_speed = calc_speed(minmult); - dprintk("FSB:%dMHz Lowest speed: %s Highest speed:%s\n", fsb, - print_speed(lowest_speed/1000), - print_speed(highest_speed/1000)); - - if (lowest_speed == highest_speed) { - printk(KERN_INFO PFX "highestspeed == lowest, aborting.\n"); - return -EINVAL; - } - if (lowest_speed > highest_speed) { - printk(KERN_INFO PFX "nonsense! lowest (%d > %d) !\n", - lowest_speed, highest_speed); - return -EINVAL; - } - - longhaul_table = kmalloc((numscales + 1) * sizeof(*longhaul_table), - GFP_KERNEL); - if (!longhaul_table) - return -ENOMEM; - - for (j = 0; j < numscales; j++) { - ratio = mults[j]; - if (ratio == -1) - continue; - if (ratio > maxmult || ratio < minmult) - continue; - longhaul_table[k].frequency = calc_speed(ratio); - longhaul_table[k].index = j; - k++; - } - if (k <= 1) { - kfree(longhaul_table); - return -ENODEV; - } - /* Sort */ - for (j = 0; j < k - 1; j++) { - unsigned int min_f, min_i; - min_f = longhaul_table[j].frequency; - min_i = j; - for (i = j + 1; i < k; i++) { - if (longhaul_table[i].frequency < min_f) { - min_f = longhaul_table[i].frequency; - min_i = i; - } - } - if (min_i != j) { - swap(longhaul_table[j].frequency, - longhaul_table[min_i].frequency); - swap(longhaul_table[j].index, - longhaul_table[min_i].index); - } - } - - longhaul_table[k].frequency = CPUFREQ_TABLE_END; - - /* Find index we are running on */ - for (j = 0; j < k; j++) { - if (mults[longhaul_table[j].index & 0x1f] == mult) { - longhaul_index = j; - break; - } - } - return 0; -} - - -static void __cpuinit longhaul_setup_voltagescaling(void) -{ - union msr_longhaul longhaul; - struct mV_pos minvid, maxvid, vid; - unsigned int j, speed, pos, kHz_step, numvscales; - int min_vid_speed; - - rdmsrl(MSR_VIA_LONGHAUL, longhaul.val); - if (!(longhaul.bits.RevisionID & 1)) { - printk(KERN_INFO PFX "Voltage scaling not supported by CPU.\n"); - return; - } - - if (!longhaul.bits.VRMRev) { - printk(KERN_INFO PFX "VRM 8.5\n"); - vrm_mV_table = &vrm85_mV[0]; - mV_vrm_table = &mV_vrm85[0]; - } else { - printk(KERN_INFO PFX "Mobile VRM\n"); - if (cpu_model < CPU_NEHEMIAH) - return; - vrm_mV_table = &mobilevrm_mV[0]; - mV_vrm_table = &mV_mobilevrm[0]; - } - - minvid = vrm_mV_table[longhaul.bits.MinimumVID]; - maxvid = vrm_mV_table[longhaul.bits.MaximumVID]; - - if (minvid.mV == 0 || maxvid.mV == 0 || minvid.mV > maxvid.mV) { - printk(KERN_INFO PFX "Bogus values Min:%d.%03d Max:%d.%03d. " - "Voltage scaling disabled.\n", - minvid.mV/1000, minvid.mV%1000, - maxvid.mV/1000, maxvid.mV%1000); - return; - } - - if (minvid.mV == maxvid.mV) { - printk(KERN_INFO PFX "Claims to support voltage scaling but " - "min & max are both %d.%03d. " - "Voltage scaling disabled\n", - maxvid.mV/1000, maxvid.mV%1000); - return; - } - - /* How many voltage steps*/ - numvscales = maxvid.pos - minvid.pos + 1; - printk(KERN_INFO PFX - "Max VID=%d.%03d " - "Min VID=%d.%03d, " - "%d possible voltage scales\n", - maxvid.mV/1000, maxvid.mV%1000, - minvid.mV/1000, minvid.mV%1000, - numvscales); - - /* Calculate max frequency at min voltage */ - j = longhaul.bits.MinMHzBR; - if (longhaul.bits.MinMHzBR4) - j += 16; - min_vid_speed = eblcr[j]; - if (min_vid_speed == -1) - return; - switch (longhaul.bits.MinMHzFSB) { - case 0: - min_vid_speed *= 13333; - break; - case 1: - min_vid_speed *= 10000; - break; - case 3: - min_vid_speed *= 6666; - break; - default: - return; - break; - } - if (min_vid_speed >= highest_speed) - return; - /* Calculate kHz for one voltage step */ - kHz_step = (highest_speed - min_vid_speed) / numvscales; - - j = 0; - while (longhaul_table[j].frequency != CPUFREQ_TABLE_END) { - speed = longhaul_table[j].frequency; - if (speed > min_vid_speed) - pos = (speed - min_vid_speed) / kHz_step + minvid.pos; - else - pos = minvid.pos; - longhaul_table[j].index |= mV_vrm_table[pos] << 8; - vid = vrm_mV_table[mV_vrm_table[pos]]; - printk(KERN_INFO PFX "f: %d kHz, index: %d, vid: %d mV\n", - speed, j, vid.mV); - j++; - } - - can_scale_voltage = 1; - printk(KERN_INFO PFX "Voltage scaling enabled.\n"); -} - - -static int longhaul_verify(struct cpufreq_policy *policy) -{ - return cpufreq_frequency_table_verify(policy, longhaul_table); -} - - -static int longhaul_target(struct cpufreq_policy *policy, - unsigned int target_freq, unsigned int relation) -{ - unsigned int table_index = 0; - unsigned int i; - unsigned int dir = 0; - u8 vid, current_vid; - - if (cpufreq_frequency_table_target(policy, longhaul_table, target_freq, - relation, &table_index)) - return -EINVAL; - - /* Don't set same frequency again */ - if (longhaul_index == table_index) - return 0; - - if (!can_scale_voltage) - longhaul_setstate(table_index); - else { - /* On test system voltage transitions exceeding single - * step up or down were turning motherboard off. Both - * "ondemand" and "userspace" are unsafe. C7 is doing - * this in hardware, C3 is old and we need to do this - * in software. */ - i = longhaul_index; - current_vid = (longhaul_table[longhaul_index].index >> 8); - current_vid &= 0x1f; - if (table_index > longhaul_index) - dir = 1; - while (i != table_index) { - vid = (longhaul_table[i].index >> 8) & 0x1f; - if (vid != current_vid) { - longhaul_setstate(i); - current_vid = vid; - msleep(200); - } - if (dir) - i++; - else - i--; - } - longhaul_setstate(table_index); - } - longhaul_index = table_index; - return 0; -} - - -static unsigned int longhaul_get(unsigned int cpu) -{ - if (cpu) - return 0; - return calc_speed(longhaul_get_cpu_mult()); -} - -static acpi_status longhaul_walk_callback(acpi_handle obj_handle, - u32 nesting_level, - void *context, void **return_value) -{ - struct acpi_device *d; - - if (acpi_bus_get_device(obj_handle, &d)) - return 0; - - *return_value = acpi_driver_data(d); - return 1; -} - -/* VIA don't support PM2 reg, but have something similar */ -static int enable_arbiter_disable(void) -{ - struct pci_dev *dev; - int status = 1; - int reg; - u8 pci_cmd; - - /* Find PLE133 host bridge */ - reg = 0x78; - dev = pci_get_device(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8601_0, - NULL); - /* Find PM133/VT8605 host bridge */ - if (dev == NULL) - dev = pci_get_device(PCI_VENDOR_ID_VIA, - PCI_DEVICE_ID_VIA_8605_0, NULL); - /* Find CLE266 host bridge */ - if (dev == NULL) { - reg = 0x76; - dev = pci_get_device(PCI_VENDOR_ID_VIA, - PCI_DEVICE_ID_VIA_862X_0, NULL); - /* Find CN400 V-Link host bridge */ - if (dev == NULL) - dev = pci_get_device(PCI_VENDOR_ID_VIA, 0x7259, NULL); - } - if (dev != NULL) { - /* Enable access to port 0x22 */ - pci_read_config_byte(dev, reg, &pci_cmd); - if (!(pci_cmd & 1<<7)) { - pci_cmd |= 1<<7; - pci_write_config_byte(dev, reg, pci_cmd); - pci_read_config_byte(dev, reg, &pci_cmd); - if (!(pci_cmd & 1<<7)) { - printk(KERN_ERR PFX - "Can't enable access to port 0x22.\n"); - status = 0; - } - } - pci_dev_put(dev); - return status; - } - return 0; -} - -static int longhaul_setup_southbridge(void) -{ - struct pci_dev *dev; - u8 pci_cmd; - - /* Find VT8235 southbridge */ - dev = pci_get_device(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8235, NULL); - if (dev == NULL) - /* Find VT8237 southbridge */ - dev = pci_get_device(PCI_VENDOR_ID_VIA, - PCI_DEVICE_ID_VIA_8237, NULL); - if (dev != NULL) { - /* Set transition time to max */ - pci_read_config_byte(dev, 0xec, &pci_cmd); - pci_cmd &= ~(1 << 2); - pci_write_config_byte(dev, 0xec, pci_cmd); - pci_read_config_byte(dev, 0xe4, &pci_cmd); - pci_cmd &= ~(1 << 7); - pci_write_config_byte(dev, 0xe4, pci_cmd); - pci_read_config_byte(dev, 0xe5, &pci_cmd); - pci_cmd |= 1 << 7; - pci_write_config_byte(dev, 0xe5, pci_cmd); - /* Get address of ACPI registers block*/ - pci_read_config_byte(dev, 0x81, &pci_cmd); - if (pci_cmd & 1 << 7) { - pci_read_config_dword(dev, 0x88, &acpi_regs_addr); - acpi_regs_addr &= 0xff00; - printk(KERN_INFO PFX "ACPI I/O at 0x%x\n", - acpi_regs_addr); - } - - pci_dev_put(dev); - return 1; - } - return 0; -} - -static int __cpuinit longhaul_cpu_init(struct cpufreq_policy *policy) -{ - struct cpuinfo_x86 *c = &cpu_data(0); - char *cpuname = NULL; - int ret; - u32 lo, hi; - - /* Check what we have on this motherboard */ - switch (c->x86_model) { - case 6: - cpu_model = CPU_SAMUEL; - cpuname = "C3 'Samuel' [C5A]"; - longhaul_version = TYPE_LONGHAUL_V1; - memcpy(mults, samuel1_mults, sizeof(samuel1_mults)); - memcpy(eblcr, samuel1_eblcr, sizeof(samuel1_eblcr)); - break; - - case 7: - switch (c->x86_mask) { - case 0: - longhaul_version = TYPE_LONGHAUL_V1; - cpu_model = CPU_SAMUEL2; - cpuname = "C3 'Samuel 2' [C5B]"; - /* Note, this is not a typo, early Samuel2's had - * Samuel1 ratios. */ - memcpy(mults, samuel1_mults, sizeof(samuel1_mults)); - memcpy(eblcr, samuel2_eblcr, sizeof(samuel2_eblcr)); - break; - case 1 ... 15: - longhaul_version = TYPE_LONGHAUL_V2; - if (c->x86_mask < 8) { - cpu_model = CPU_SAMUEL2; - cpuname = "C3 'Samuel 2' [C5B]"; - } else { - cpu_model = CPU_EZRA; - cpuname = "C3 'Ezra' [C5C]"; - } - memcpy(mults, ezra_mults, sizeof(ezra_mults)); - memcpy(eblcr, ezra_eblcr, sizeof(ezra_eblcr)); - break; - } - break; - - case 8: - cpu_model = CPU_EZRA_T; - cpuname = "C3 'Ezra-T' [C5M]"; - longhaul_version = TYPE_POWERSAVER; - numscales = 32; - memcpy(mults, ezrat_mults, sizeof(ezrat_mults)); - memcpy(eblcr, ezrat_eblcr, sizeof(ezrat_eblcr)); - break; - - case 9: - longhaul_version = TYPE_POWERSAVER; - numscales = 32; - memcpy(mults, nehemiah_mults, sizeof(nehemiah_mults)); - memcpy(eblcr, nehemiah_eblcr, sizeof(nehemiah_eblcr)); - switch (c->x86_mask) { - case 0 ... 1: - cpu_model = CPU_NEHEMIAH; - cpuname = "C3 'Nehemiah A' [C5XLOE]"; - break; - case 2 ... 4: - cpu_model = CPU_NEHEMIAH; - cpuname = "C3 'Nehemiah B' [C5XLOH]"; - break; - case 5 ... 15: - cpu_model = CPU_NEHEMIAH_C; - cpuname = "C3 'Nehemiah C' [C5P]"; - break; - } - break; - - default: - cpuname = "Unknown"; - break; - } - /* Check Longhaul ver. 2 */ - if (longhaul_version == TYPE_LONGHAUL_V2) { - rdmsr(MSR_VIA_LONGHAUL, lo, hi); - if (lo == 0 && hi == 0) - /* Looks like MSR isn't present */ - longhaul_version = TYPE_LONGHAUL_V1; - } - - printk(KERN_INFO PFX "VIA %s CPU detected. ", cpuname); - switch (longhaul_version) { - case TYPE_LONGHAUL_V1: - case TYPE_LONGHAUL_V2: - printk(KERN_CONT "Longhaul v%d supported.\n", longhaul_version); - break; - case TYPE_POWERSAVER: - printk(KERN_CONT "Powersaver supported.\n"); - break; - }; - - /* Doesn't hurt */ - longhaul_setup_southbridge(); - - /* Find ACPI data for processor */ - acpi_walk_namespace(ACPI_TYPE_PROCESSOR, ACPI_ROOT_OBJECT, - ACPI_UINT32_MAX, &longhaul_walk_callback, NULL, - NULL, (void *)&pr); - - /* Check ACPI support for C3 state */ - if (pr != NULL && longhaul_version == TYPE_POWERSAVER) { - cx = &pr->power.states[ACPI_STATE_C3]; - if (cx->address > 0 && cx->latency <= 1000) - longhaul_flags |= USE_ACPI_C3; - } - /* Disable if it isn't working */ - if (disable_acpi_c3) - longhaul_flags &= ~USE_ACPI_C3; - /* Check if northbridge is friendly */ - if (enable_arbiter_disable()) - longhaul_flags |= USE_NORTHBRIDGE; - - /* Check ACPI support for bus master arbiter disable */ - if (!(longhaul_flags & USE_ACPI_C3 - || longhaul_flags & USE_NORTHBRIDGE) - && ((pr == NULL) || !(pr->flags.bm_control))) { - printk(KERN_ERR PFX - "No ACPI support. Unsupported northbridge.\n"); - return -ENODEV; - } - - if (longhaul_flags & USE_NORTHBRIDGE) - printk(KERN_INFO PFX "Using northbridge support.\n"); - if (longhaul_flags & USE_ACPI_C3) - printk(KERN_INFO PFX "Using ACPI support.\n"); - - ret = longhaul_get_ranges(); - if (ret != 0) - return ret; - - if ((longhaul_version != TYPE_LONGHAUL_V1) && (scale_voltage != 0)) - longhaul_setup_voltagescaling(); - - policy->cpuinfo.transition_latency = 200000; /* nsec */ - policy->cur = calc_speed(longhaul_get_cpu_mult()); - - ret = cpufreq_frequency_table_cpuinfo(policy, longhaul_table); - if (ret) - return ret; - - cpufreq_frequency_table_get_attr(longhaul_table, policy->cpu); - - return 0; -} - -static int __devexit longhaul_cpu_exit(struct cpufreq_policy *policy) -{ - cpufreq_frequency_table_put_attr(policy->cpu); - return 0; -} - -static struct freq_attr *longhaul_attr[] = { - &cpufreq_freq_attr_scaling_available_freqs, - NULL, -}; - -static struct cpufreq_driver longhaul_driver = { - .verify = longhaul_verify, - .target = longhaul_target, - .get = longhaul_get, - .init = longhaul_cpu_init, - .exit = __devexit_p(longhaul_cpu_exit), - .name = "longhaul", - .owner = THIS_MODULE, - .attr = longhaul_attr, -}; - - -static int __init longhaul_init(void) -{ - struct cpuinfo_x86 *c = &cpu_data(0); - - if (c->x86_vendor != X86_VENDOR_CENTAUR || c->x86 != 6) - return -ENODEV; - -#ifdef CONFIG_SMP - if (num_online_cpus() > 1) { - printk(KERN_ERR PFX "More than 1 CPU detected, " - "longhaul disabled.\n"); - return -ENODEV; - } -#endif -#ifdef CONFIG_X86_IO_APIC - if (cpu_has_apic) { - printk(KERN_ERR PFX "APIC detected. Longhaul is currently " - "broken in this configuration.\n"); - return -ENODEV; - } -#endif - switch (c->x86_model) { - case 6 ... 9: - return cpufreq_register_driver(&longhaul_driver); - case 10: - printk(KERN_ERR PFX "Use acpi-cpufreq driver for VIA C7\n"); - default: - ; - } - - return -ENODEV; -} - - -static void __exit longhaul_exit(void) -{ - int i; - - for (i = 0; i < numscales; i++) { - if (mults[i] == maxmult) { - longhaul_setstate(i); - break; - } - } - - cpufreq_unregister_driver(&longhaul_driver); - kfree(longhaul_table); -} - -/* Even if BIOS is exporting ACPI C3 state, and it is used - * with success when CPU is idle, this state doesn't - * trigger frequency transition in some cases. */ -module_param(disable_acpi_c3, int, 0644); -MODULE_PARM_DESC(disable_acpi_c3, "Don't use ACPI C3 support"); -/* Change CPU voltage with frequency. Very useful to save - * power, but most VIA C3 processors aren't supporting it. */ -module_param(scale_voltage, int, 0644); -MODULE_PARM_DESC(scale_voltage, "Scale voltage of processor"); -/* Force revision key to 0 for processors which doesn't - * support voltage scaling, but are introducing itself as - * such. */ -module_param(revid_errata, int, 0644); -MODULE_PARM_DESC(revid_errata, "Ignore CPU Revision ID"); - -MODULE_AUTHOR("Dave Jones <davej@redhat.com>"); -MODULE_DESCRIPTION("Longhaul driver for VIA Cyrix processors."); -MODULE_LICENSE("GPL"); - -late_initcall(longhaul_init); -module_exit(longhaul_exit); diff --git a/arch/x86/kernel/cpu/cpufreq/longhaul.h b/arch/x86/kernel/cpu/cpufreq/longhaul.h deleted file mode 100644 index cbf48fbca881..000000000000 --- a/arch/x86/kernel/cpu/cpufreq/longhaul.h +++ /dev/null @@ -1,353 +0,0 @@ -/* - * longhaul.h - * (C) 2003 Dave Jones. - * - * Licensed under the terms of the GNU GPL License version 2. - * - * VIA-specific information - */ - -union msr_bcr2 { - struct { - unsigned Reseved:19, // 18:0 - ESOFTBF:1, // 19 - Reserved2:3, // 22:20 - CLOCKMUL:4, // 26:23 - Reserved3:5; // 31:27 - } bits; - unsigned long val; -}; - -union msr_longhaul { - struct { - unsigned RevisionID:4, // 3:0 - RevisionKey:4, // 7:4 - EnableSoftBusRatio:1, // 8 - EnableSoftVID:1, // 9 - EnableSoftBSEL:1, // 10 - Reserved:3, // 11:13 - SoftBusRatio4:1, // 14 - VRMRev:1, // 15 - SoftBusRatio:4, // 19:16 - SoftVID:5, // 24:20 - Reserved2:3, // 27:25 - SoftBSEL:2, // 29:28 - Reserved3:2, // 31:30 - MaxMHzBR:4, // 35:32 - MaximumVID:5, // 40:36 - MaxMHzFSB:2, // 42:41 - MaxMHzBR4:1, // 43 - Reserved4:4, // 47:44 - MinMHzBR:4, // 51:48 - MinimumVID:5, // 56:52 - MinMHzFSB:2, // 58:57 - MinMHzBR4:1, // 59 - Reserved5:4; // 63:60 - } bits; - unsigned long long val; -}; - -/* - * Clock ratio tables. Div/Mod by 10 to get ratio. - * The eblcr values specify the ratio read from the CPU. - * The mults values specify what to write to the CPU. - */ - -/* - * VIA C3 Samuel 1 & Samuel 2 (stepping 0) - */ -static const int __cpuinitdata samuel1_mults[16] = { - -1, /* 0000 -> RESERVED */ - 30, /* 0001 -> 3.0x */ - 40, /* 0010 -> 4.0x */ - -1, /* 0011 -> RESERVED */ - -1, /* 0100 -> RESERVED */ - 35, /* 0101 -> 3.5x */ - 45, /* 0110 -> 4.5x */ - 55, /* 0111 -> 5.5x */ - 60, /* 1000 -> 6.0x */ - 70, /* 1001 -> 7.0x */ - 80, /* 1010 -> 8.0x */ - 50, /* 1011 -> 5.0x */ - 65, /* 1100 -> 6.5x */ - 75, /* 1101 -> 7.5x */ - -1, /* 1110 -> RESERVED */ - -1, /* 1111 -> RESERVED */ -}; - -static const int __cpuinitdata samuel1_eblcr[16] = { - 50, /* 0000 -> RESERVED */ - 30, /* 0001 -> 3.0x */ - 40, /* 0010 -> 4.0x */ - -1, /* 0011 -> RESERVED */ - 55, /* 0100 -> 5.5x */ - 35, /* 0101 -> 3.5x */ - 45, /* 0110 -> 4.5x */ - -1, /* 0111 -> RESERVED */ - -1, /* 1000 -> RESERVED */ - 70, /* 1001 -> 7.0x */ - 80, /* 1010 -> 8.0x */ - 60, /* 1011 -> 6.0x */ - -1, /* 1100 -> RESERVED */ - 75, /* 1101 -> 7.5x */ - -1, /* 1110 -> RESERVED */ - 65, /* 1111 -> 6.5x */ -}; - -/* - * VIA C3 Samuel2 Stepping 1->15 - */ -static const int __cpuinitdata samuel2_eblcr[16] = { - 50, /* 0000 -> 5.0x */ - 30, /* 0001 -> 3.0x */ - 40, /* 0010 -> 4.0x */ - 100, /* 0011 -> 10.0x */ - 55, /* 0100 -> 5.5x */ - 35, /* 0101 -> 3.5x */ - 45, /* 0110 -> 4.5x */ - 110, /* 0111 -> 11.0x */ - 90, /* 1000 -> 9.0x */ - 70, /* 1001 -> 7.0x */ - 80, /* 1010 -> 8.0x */ - 60, /* 1011 -> 6.0x */ - 120, /* 1100 -> 12.0x */ - 75, /* 1101 -> 7.5x */ - 130, /* 1110 -> 13.0x */ - 65, /* 1111 -> 6.5x */ -}; - -/* - * VIA C3 Ezra - */ -static const int __cpuinitdata ezra_mults[16] = { - 100, /* 0000 -> 10.0x */ - 30, /* 0001 -> 3.0x */ - 40, /* 0010 -> 4.0x */ - 90, /* 0011 -> 9.0x */ - 95, /* 0100 -> 9.5x */ - 35, /* 0101 -> 3.5x */ - 45, /* 0110 -> 4.5x */ - 55, /* 0111 -> 5.5x */ - 60, /* 1000 -> 6.0x */ - 70, /* 1001 -> 7.0x */ - 80, /* 1010 -> 8.0x */ - 50, /* 1011 -> 5.0x */ - 65, /* 1100 -> 6.5x */ - 75, /* 1101 -> 7.5x */ - 85, /* 1110 -> 8.5x */ - 120, /* 1111 -> 12.0x */ -}; - -static const int __cpuinitdata ezra_eblcr[16] = { - 50, /* 0000 -> 5.0x */ - 30, /* 0001 -> 3.0x */ - 40, /* 0010 -> 4.0x */ - 100, /* 0011 -> 10.0x */ - 55, /* 0100 -> 5.5x */ - 35, /* 0101 -> 3.5x */ - 45, /* 0110 -> 4.5x */ - 95, /* 0111 -> 9.5x */ - 90, /* 1000 -> 9.0x */ - 70, /* 1001 -> 7.0x */ - 80, /* 1010 -> 8.0x */ - 60, /* 1011 -> 6.0x */ - 120, /* 1100 -> 12.0x */ - 75, /* 1101 -> 7.5x */ - 85, /* 1110 -> 8.5x */ - 65, /* 1111 -> 6.5x */ -}; - -/* - * VIA C3 (Ezra-T) [C5M]. - */ -static const int __cpuinitdata ezrat_mults[32] = { - 100, /* 0000 -> 10.0x */ - 30, /* 0001 -> 3.0x */ - 40, /* 0010 -> 4.0x */ - 90, /* 0011 -> 9.0x */ - 95, /* 0100 -> 9.5x */ - 35, /* 0101 -> 3.5x */ - 45, /* 0110 -> 4.5x */ - 55, /* 0111 -> 5.5x */ - 60, /* 1000 -> 6.0x */ - 70, /* 1001 -> 7.0x */ - 80, /* 1010 -> 8.0x */ - 50, /* 1011 -> 5.0x */ - 65, /* 1100 -> 6.5x */ - 75, /* 1101 -> 7.5x */ - 85, /* 1110 -> 8.5x */ - 120, /* 1111 -> 12.0x */ - - -1, /* 0000 -> RESERVED (10.0x) */ - 110, /* 0001 -> 11.0x */ - -1, /* 0010 -> 12.0x */ - -1, /* 0011 -> RESERVED (9.0x)*/ - 105, /* 0100 -> 10.5x */ - 115, /* 0101 -> 11.5x */ - 125, /* 0110 -> 12.5x */ - 135, /* 0111 -> 13.5x */ - 140, /* 1000 -> 14.0x */ - 150, /* 1001 -> 15.0x */ - 160, /* 1010 -> 16.0x */ - 130, /* 1011 -> 13.0x */ - 145, /* 1100 -> 14.5x */ - 155, /* 1101 -> 15.5x */ - -1, /* 1110 -> RESERVED (13.0x) */ - -1, /* 1111 -> RESERVED (12.0x) */ -}; - -static const int __cpuinitdata ezrat_eblcr[32] = { - 50, /* 0000 -> 5.0x */ - 30, /* 0001 -> 3.0x */ - 40, /* 0010 -> 4.0x */ - 100, /* 0011 -> 10.0x */ - 55, /* 0100 -> 5.5x */ - 35, /* 0101 -> 3.5x */ - 45, /* 0110 -> 4.5x */ - 95, /* 0111 -> 9.5x */ - 90, /* 1000 -> 9.0x */ - 70, /* 1001 -> 7.0x */ - 80, /* 1010 -> 8.0x */ - 60, /* 1011 -> 6.0x */ - 120, /* 1100 -> 12.0x */ - 75, /* 1101 -> 7.5x */ - 85, /* 1110 -> 8.5x */ - 65, /* 1111 -> 6.5x */ - - -1, /* 0000 -> RESERVED (9.0x) */ - 110, /* 0001 -> 11.0x */ - 120, /* 0010 -> 12.0x */ - -1, /* 0011 -> RESERVED (10.0x)*/ - 135, /* 0100 -> 13.5x */ - 115, /* 0101 -> 11.5x */ - 125, /* 0110 -> 12.5x */ - 105, /* 0111 -> 10.5x */ - 130, /* 1000 -> 13.0x */ - 150, /* 1001 -> 15.0x */ - 160, /* 1010 -> 16.0x */ - 140, /* 1011 -> 14.0x */ - -1, /* 1100 -> RESERVED (12.0x) */ - 155, /* 1101 -> 15.5x */ - -1, /* 1110 -> RESERVED (13.0x) */ - 145, /* 1111 -> 14.5x */ -}; - -/* - * VIA C3 Nehemiah */ - -static const int __cpuinitdata nehemiah_mults[32] = { - 100, /* 0000 -> 10.0x */ - -1, /* 0001 -> 16.0x */ - 40, /* 0010 -> 4.0x */ - 90, /* 0011 -> 9.0x */ - 95, /* 0100 -> 9.5x */ - -1, /* 0101 -> RESERVED */ - 45, /* 0110 -> 4.5x */ - 55, /* 0111 -> 5.5x */ - 60, /* 1000 -> 6.0x */ - 70, /* 1001 -> 7.0x */ - 80, /* 1010 -> 8.0x */ - 50, /* 1011 -> 5.0x */ - 65, /* 1100 -> 6.5x */ - 75, /* 1101 -> 7.5x */ - 85, /* 1110 -> 8.5x */ - 120, /* 1111 -> 12.0x */ - -1, /* 0000 -> 10.0x */ - 110, /* 0001 -> 11.0x */ - -1, /* 0010 -> 12.0x */ - -1, /* 0011 -> 9.0x */ - 105, /* 0100 -> 10.5x */ - 115, /* 0101 -> 11.5x */ - 125, /* 0110 -> 12.5x */ - 135, /* 0111 -> 13.5x */ - 140, /* 1000 -> 14.0x */ - 150, /* 1001 -> 15.0x */ - 160, /* 1010 -> 16.0x */ - 130, /* 1011 -> 13.0x */ - 145, /* 1100 -> 14.5x */ - 155, /* 1101 -> 15.5x */ - -1, /* 1110 -> RESERVED (13.0x) */ - -1, /* 1111 -> 12.0x */ -}; - -static const int __cpuinitdata nehemiah_eblcr[32] = { - 50, /* 0000 -> 5.0x */ - 160, /* 0001 -> 16.0x */ - 40, /* 0010 -> 4.0x */ - 100, /* 0011 -> 10.0x */ - 55, /* 0100 -> 5.5x */ - -1, /* 0101 -> RESERVED */ - 45, /* 0110 -> 4.5x */ - 95, /* 0111 -> 9.5x */ - 90, /* 1000 -> 9.0x */ - 70, /* 1001 -> 7.0x */ - 80, /* 1010 -> 8.0x */ - 60, /* 1011 -> 6.0x */ - 120, /* 1100 -> 12.0x */ - 75, /* 1101 -> 7.5x */ - 85, /* 1110 -> 8.5x */ - 65, /* 1111 -> 6.5x */ - 90, /* 0000 -> 9.0x */ - 110, /* 0001 -> 11.0x */ - 120, /* 0010 -> 12.0x */ - 100, /* 0011 -> 10.0x */ - 135, /* 0100 -> 13.5x */ - 115, /* 0101 -> 11.5x */ - 125, /* 0110 -> 12.5x */ - 105, /* 0111 -> 10.5x */ - 130, /* 1000 -> 13.0x */ - 150, /* 1001 -> 15.0x */ - 160, /* 1010 -> 16.0x */ - 140, /* 1011 -> 14.0x */ - 120, /* 1100 -> 12.0x */ - 155, /* 1101 -> 15.5x */ - -1, /* 1110 -> RESERVED (13.0x) */ - 145 /* 1111 -> 14.5x */ -}; - -/* - * Voltage scales. Div/Mod by 1000 to get actual voltage. - * Which scale to use depends on the VRM type in use. - */ - -struct mV_pos { - unsigned short mV; - unsigned short pos; -}; - -static const struct mV_pos __cpuinitdata vrm85_mV[32] = { - {1250, 8}, {1200, 6}, {1150, 4}, {1100, 2}, - {1050, 0}, {1800, 30}, {1750, 28}, {1700, 26}, - {1650, 24}, {1600, 22}, {1550, 20}, {1500, 18}, - {1450, 16}, {1400, 14}, {1350, 12}, {1300, 10}, - {1275, 9}, {1225, 7}, {1175, 5}, {1125, 3}, - {1075, 1}, {1825, 31}, {1775, 29}, {1725, 27}, - {1675, 25}, {1625, 23}, {1575, 21}, {1525, 19}, - {1475, 17}, {1425, 15}, {1375, 13}, {1325, 11} -}; - -static const unsigned char __cpuinitdata mV_vrm85[32] = { - 0x04, 0x14, 0x03, 0x13, 0x02, 0x12, 0x01, 0x11, - 0x00, 0x10, 0x0f, 0x1f, 0x0e, 0x1e, 0x0d, 0x1d, - 0x0c, 0x1c, 0x0b, 0x1b, 0x0a, 0x1a, 0x09, 0x19, - 0x08, 0x18, 0x07, 0x17, 0x06, 0x16, 0x05, 0x15 -}; - -static const struct mV_pos __cpuinitdata mobilevrm_mV[32] = { - {1750, 31}, {1700, 30}, {1650, 29}, {1600, 28}, - {1550, 27}, {1500, 26}, {1450, 25}, {1400, 24}, - {1350, 23}, {1300, 22}, {1250, 21}, {1200, 20}, - {1150, 19}, {1100, 18}, {1050, 17}, {1000, 16}, - {975, 15}, {950, 14}, {925, 13}, {900, 12}, - {875, 11}, {850, 10}, {825, 9}, {800, 8}, - {775, 7}, {750, 6}, {725, 5}, {700, 4}, - {675, 3}, {650, 2}, {625, 1}, {600, 0} -}; - -static const unsigned char __cpuinitdata mV_mobilevrm[32] = { - 0x1f, 0x1e, 0x1d, 0x1c, 0x1b, 0x1a, 0x19, 0x18, - 0x17, 0x16, 0x15, 0x14, 0x13, 0x12, 0x11, 0x10, - 0x0f, 0x0e, 0x0d, 0x0c, 0x0b, 0x0a, 0x09, 0x08, - 0x07, 0x06, 0x05, 0x04, 0x03, 0x02, 0x01, 0x00 -}; - diff --git a/arch/x86/kernel/cpu/cpufreq/longrun.c b/arch/x86/kernel/cpu/cpufreq/longrun.c deleted file mode 100644 index d9f51367666b..000000000000 --- a/arch/x86/kernel/cpu/cpufreq/longrun.c +++ /dev/null @@ -1,327 +0,0 @@ -/* - * (C) 2002 - 2003 Dominik Brodowski <linux@brodo.de> - * - * Licensed under the terms of the GNU GPL License version 2. - * - * BIG FAT DISCLAIMER: Work in progress code. Possibly *dangerous* - */ - -#include <linux/kernel.h> -#include <linux/module.h> -#include <linux/init.h> -#include <linux/cpufreq.h> -#include <linux/timex.h> - -#include <asm/msr.h> -#include <asm/processor.h> - -#define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, \ - "longrun", msg) - -static struct cpufreq_driver longrun_driver; - -/** - * longrun_{low,high}_freq is needed for the conversion of cpufreq kHz - * values into per cent values. In TMTA microcode, the following is valid: - * performance_pctg = (current_freq - low_freq)/(high_freq - low_freq) - */ -static unsigned int longrun_low_freq, longrun_high_freq; - - -/** - * longrun_get_policy - get the current LongRun policy - * @policy: struct cpufreq_policy where current policy is written into - * - * Reads the current LongRun policy by access to MSR_TMTA_LONGRUN_FLAGS - * and MSR_TMTA_LONGRUN_CTRL - */ -static void __cpuinit longrun_get_policy(struct cpufreq_policy *policy) -{ - u32 msr_lo, msr_hi; - - rdmsr(MSR_TMTA_LONGRUN_FLAGS, msr_lo, msr_hi); - dprintk("longrun flags are %x - %x\n", msr_lo, msr_hi); - if (msr_lo & 0x01) - policy->policy = CPUFREQ_POLICY_PERFORMANCE; - else - policy->policy = CPUFREQ_POLICY_POWERSAVE; - - rdmsr(MSR_TMTA_LONGRUN_CTRL, msr_lo, msr_hi); - dprintk("longrun ctrl is %x - %x\n", msr_lo, msr_hi); - msr_lo &= 0x0000007F; - msr_hi &= 0x0000007F; - - if (longrun_high_freq <= longrun_low_freq) { - /* Assume degenerate Longrun table */ - policy->min = policy->max = longrun_high_freq; - } else { - policy->min = longrun_low_freq + msr_lo * - ((longrun_high_freq - longrun_low_freq) / 100); - policy->max = longrun_low_freq + msr_hi * - ((longrun_high_freq - longrun_low_freq) / 100); - } - policy->cpu = 0; -} - - -/** - * longrun_set_policy - sets a new CPUFreq policy - * @policy: new policy - * - * Sets a new CPUFreq policy on LongRun-capable processors. This function - * has to be called with cpufreq_driver locked. - */ -static int longrun_set_policy(struct cpufreq_policy *policy) -{ - u32 msr_lo, msr_hi; - u32 pctg_lo, pctg_hi; - - if (!policy) - return -EINVAL; - - if (longrun_high_freq <= longrun_low_freq) { - /* Assume degenerate Longrun table */ - pctg_lo = pctg_hi = 100; - } else { - pctg_lo = (policy->min - longrun_low_freq) / - ((longrun_high_freq - longrun_low_freq) / 100); - pctg_hi = (policy->max - longrun_low_freq) / - ((longrun_high_freq - longrun_low_freq) / 100); - } - - if (pctg_hi > 100) - pctg_hi = 100; - if (pctg_lo > pctg_hi) - pctg_lo = pctg_hi; - - /* performance or economy mode */ - rdmsr(MSR_TMTA_LONGRUN_FLAGS, msr_lo, msr_hi); - msr_lo &= 0xFFFFFFFE; - switch (policy->policy) { - case CPUFREQ_POLICY_PERFORMANCE: - msr_lo |= 0x00000001; - break; - case CPUFREQ_POLICY_POWERSAVE: - break; - } - wrmsr(MSR_TMTA_LONGRUN_FLAGS, msr_lo, msr_hi); - - /* lower and upper boundary */ - rdmsr(MSR_TMTA_LONGRUN_CTRL, msr_lo, msr_hi); - msr_lo &= 0xFFFFFF80; - msr_hi &= 0xFFFFFF80; - msr_lo |= pctg_lo; - msr_hi |= pctg_hi; - wrmsr(MSR_TMTA_LONGRUN_CTRL, msr_lo, msr_hi); - - return 0; -} - - -/** - * longrun_verify_poliy - verifies a new CPUFreq policy - * @policy: the policy to verify - * - * Validates a new CPUFreq policy. This function has to be called with - * cpufreq_driver locked. - */ -static int longrun_verify_policy(struct cpufreq_policy *policy) -{ - if (!policy) - return -EINVAL; - - policy->cpu = 0; - cpufreq_verify_within_limits(policy, - policy->cpuinfo.min_freq, - policy->cpuinfo.max_freq); - - if ((policy->policy != CPUFREQ_POLICY_POWERSAVE) && - (policy->policy != CPUFREQ_POLICY_PERFORMANCE)) - return -EINVAL; - - return 0; -} - -static unsigned int longrun_get(unsigned int cpu) -{ - u32 eax, ebx, ecx, edx; - - if (cpu) - return 0; - - cpuid(0x80860007, &eax, &ebx, &ecx, &edx); - dprintk("cpuid eax is %u\n", eax); - - return eax * 1000; -} - -/** - * longrun_determine_freqs - determines the lowest and highest possible core frequency - * @low_freq: an int to put the lowest frequency into - * @high_freq: an int to put the highest frequency into - * - * Determines the lowest and highest possible core frequencies on this CPU. - * This is necessary to calculate the performance percentage according to - * TMTA rules: - * performance_pctg = (target_freq - low_freq)/(high_freq - low_freq) - */ -static int __cpuinit longrun_determine_freqs(unsigned int *low_freq, - unsigned int *high_freq) -{ - u32 msr_lo, msr_hi; - u32 save_lo, save_hi; - u32 eax, ebx, ecx, edx; - u32 try_hi; - struct cpuinfo_x86 *c = &cpu_data(0); - - if (!low_freq || !high_freq) - return -EINVAL; - - if (cpu_has(c, X86_FEATURE_LRTI)) { - /* if the LongRun Table Interface is present, the - * detection is a bit easier: - * For minimum frequency, read out the maximum - * level (msr_hi), write that into "currently - * selected level", and read out the frequency. - * For maximum frequency, read out level zero. - */ - /* minimum */ - rdmsr(MSR_TMTA_LRTI_READOUT, msr_lo, msr_hi); - wrmsr(MSR_TMTA_LRTI_READOUT, msr_hi, msr_hi); - rdmsr(MSR_TMTA_LRTI_VOLT_MHZ, msr_lo, msr_hi); - *low_freq = msr_lo * 1000; /* to kHz */ - - /* maximum */ - wrmsr(MSR_TMTA_LRTI_READOUT, 0, msr_hi); - rdmsr(MSR_TMTA_LRTI_VOLT_MHZ, msr_lo, msr_hi); - *high_freq = msr_lo * 1000; /* to kHz */ - - dprintk("longrun table interface told %u - %u kHz\n", - *low_freq, *high_freq); - - if (*low_freq > *high_freq) - *low_freq = *high_freq; - return 0; - } - - /* set the upper border to the value determined during TSC init */ - *high_freq = (cpu_khz / 1000); - *high_freq = *high_freq * 1000; - dprintk("high frequency is %u kHz\n", *high_freq); - - /* get current borders */ - rdmsr(MSR_TMTA_LONGRUN_CTRL, msr_lo, msr_hi); - save_lo = msr_lo & 0x0000007F; - save_hi = msr_hi & 0x0000007F; - - /* if current perf_pctg is larger than 90%, we need to decrease the - * upper limit to make the calculation more accurate. - */ - cpuid(0x80860007, &eax, &ebx, &ecx, &edx); - /* try decreasing in 10% steps, some processors react only - * on some barrier values */ - for (try_hi = 80; try_hi > 0 && ecx > 90; try_hi -= 10) { - /* set to 0 to try_hi perf_pctg */ - msr_lo &= 0xFFFFFF80; - msr_hi &= 0xFFFFFF80; - msr_hi |= try_hi; - wrmsr(MSR_TMTA_LONGRUN_CTRL, msr_lo, msr_hi); - - /* read out current core MHz and current perf_pctg */ - cpuid(0x80860007, &eax, &ebx, &ecx, &edx); - - /* restore values */ - wrmsr(MSR_TMTA_LONGRUN_CTRL, save_lo, save_hi); - } - dprintk("percentage is %u %%, freq is %u MHz\n", ecx, eax); - - /* performance_pctg = (current_freq - low_freq)/(high_freq - low_freq) - * eqals - * low_freq * (1 - perf_pctg) = (cur_freq - high_freq * perf_pctg) - * - * high_freq * perf_pctg is stored tempoarily into "ebx". - */ - ebx = (((cpu_khz / 1000) * ecx) / 100); /* to MHz */ - - if ((ecx > 95) || (ecx == 0) || (eax < ebx)) - return -EIO; - - edx = ((eax - ebx) * 100) / (100 - ecx); - *low_freq = edx * 1000; /* back to kHz */ - - dprintk("low frequency is %u kHz\n", *low_freq); - - if (*low_freq > *high_freq) - *low_freq = *high_freq; - - return 0; -} - - -static int __cpuinit longrun_cpu_init(struct cpufreq_policy *policy) -{ - int result = 0; - - /* capability check */ - if (policy->cpu != 0) - return -ENODEV; - - /* detect low and high frequency */ - result = longrun_determine_freqs(&longrun_low_freq, &longrun_high_freq); - if (result) - return result; - - /* cpuinfo and default policy values */ - policy->cpuinfo.min_freq = longrun_low_freq; - policy->cpuinfo.max_freq = longrun_high_freq; - policy->cpuinfo.transition_latency = CPUFREQ_ETERNAL; - longrun_get_policy(policy); - - return 0; -} - - -static struct cpufreq_driver longrun_driver = { - .flags = CPUFREQ_CONST_LOOPS, - .verify = longrun_verify_policy, - .setpolicy = longrun_set_policy, - .get = longrun_get, - .init = longrun_cpu_init, - .name = "longrun", - .owner = THIS_MODULE, -}; - - -/** - * longrun_init - initializes the Transmeta Crusoe LongRun CPUFreq driver - * - * Initializes the LongRun support. - */ -static int __init longrun_init(void) -{ - struct cpuinfo_x86 *c = &cpu_data(0); - - if (c->x86_vendor != X86_VENDOR_TRANSMETA || - !cpu_has(c, X86_FEATURE_LONGRUN)) - return -ENODEV; - - return cpufreq_register_driver(&longrun_driver); -} - - -/** - * longrun_exit - unregisters LongRun support - */ -static void __exit longrun_exit(void) -{ - cpufreq_unregister_driver(&longrun_driver); -} - - -MODULE_AUTHOR("Dominik Brodowski <linux@brodo.de>"); -MODULE_DESCRIPTION("LongRun driver for Transmeta Crusoe and " - "Efficeon processors."); -MODULE_LICENSE("GPL"); - -module_init(longrun_init); -module_exit(longrun_exit); diff --git a/arch/x86/kernel/cpu/cpufreq/mperf.c b/arch/x86/kernel/cpu/cpufreq/mperf.c deleted file mode 100644 index 911e193018ae..000000000000 --- a/arch/x86/kernel/cpu/cpufreq/mperf.c +++ /dev/null @@ -1,51 +0,0 @@ -#include <linux/kernel.h> -#include <linux/smp.h> -#include <linux/module.h> -#include <linux/init.h> -#include <linux/cpufreq.h> -#include <linux/slab.h> - -#include "mperf.h" - -static DEFINE_PER_CPU(struct aperfmperf, acfreq_old_perf); - -/* Called via smp_call_function_single(), on the target CPU */ -static void read_measured_perf_ctrs(void *_cur) -{ - struct aperfmperf *am = _cur; - - get_aperfmperf(am); -} - -/* - * Return the measured active (C0) frequency on this CPU since last call - * to this function. - * Input: cpu number - * Return: Average CPU frequency in terms of max frequency (zero on error) - * - * We use IA32_MPERF and IA32_APERF MSRs to get the measured performance - * over a period of time, while CPU is in C0 state. - * IA32_MPERF counts at the rate of max advertised frequency - * IA32_APERF counts at the rate of actual CPU frequency - * Only IA32_APERF/IA32_MPERF ratio is architecturally defined and - * no meaning should be associated with absolute values of these MSRs. - */ -unsigned int cpufreq_get_measured_perf(struct cpufreq_policy *policy, - unsigned int cpu) -{ - struct aperfmperf perf; - unsigned long ratio; - unsigned int retval; - - if (smp_call_function_single(cpu, read_measured_perf_ctrs, &perf, 1)) - return 0; - - ratio = calc_aperfmperf_ratio(&per_cpu(acfreq_old_perf, cpu), &perf); - per_cpu(acfreq_old_perf, cpu) = perf; - - retval = (policy->cpuinfo.max_freq * ratio) >> APERFMPERF_SHIFT; - - return retval; -} -EXPORT_SYMBOL_GPL(cpufreq_get_measured_perf); -MODULE_LICENSE("GPL"); diff --git a/arch/x86/kernel/cpu/cpufreq/mperf.h b/arch/x86/kernel/cpu/cpufreq/mperf.h deleted file mode 100644 index 5dbf2950dc22..000000000000 --- a/arch/x86/kernel/cpu/cpufreq/mperf.h +++ /dev/null @@ -1,9 +0,0 @@ -/* - * (c) 2010 Advanced Micro Devices, Inc. - * Your use of this code is subject to the terms and conditions of the - * GNU general public license version 2. See "COPYING" or - * http://www.gnu.org/licenses/gpl.html - */ - -unsigned int cpufreq_get_measured_perf(struct cpufreq_policy *policy, - unsigned int cpu); diff --git a/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c b/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c deleted file mode 100644 index 52c93648e492..000000000000 --- a/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c +++ /dev/null @@ -1,331 +0,0 @@ -/* - * Pentium 4/Xeon CPU on demand clock modulation/speed scaling - * (C) 2002 - 2003 Dominik Brodowski <linux@brodo.de> - * (C) 2002 Zwane Mwaikambo <zwane@commfireservices.com> - * (C) 2002 Arjan van de Ven <arjanv@redhat.com> - * (C) 2002 Tora T. Engstad - * All Rights Reserved - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - * - * The author(s) of this software shall not be held liable for damages - * of any nature resulting due to the use of this software. This - * software is provided AS-IS with no warranties. - * - * Date Errata Description - * 20020525 N44, O17 12.5% or 25% DC causes lockup - * - */ - -#include <linux/kernel.h> -#include <linux/module.h> -#include <linux/init.h> -#include <linux/smp.h> -#include <linux/cpufreq.h> -#include <linux/cpumask.h> -#include <linux/timex.h> - -#include <asm/processor.h> -#include <asm/msr.h> -#include <asm/timer.h> - -#include "speedstep-lib.h" - -#define PFX "p4-clockmod: " -#define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, \ - "p4-clockmod", msg) - -/* - * Duty Cycle (3bits), note DC_DISABLE is not specified in - * intel docs i just use it to mean disable - */ -enum { - DC_RESV, DC_DFLT, DC_25PT, DC_38PT, DC_50PT, - DC_64PT, DC_75PT, DC_88PT, DC_DISABLE -}; - -#define DC_ENTRIES 8 - - -static int has_N44_O17_errata[NR_CPUS]; -static unsigned int stock_freq; -static struct cpufreq_driver p4clockmod_driver; -static unsigned int cpufreq_p4_get(unsigned int cpu); - -static int cpufreq_p4_setdc(unsigned int cpu, unsigned int newstate) -{ - u32 l, h; - - if (!cpu_online(cpu) || - (newstate > DC_DISABLE) || (newstate == DC_RESV)) - return -EINVAL; - - rdmsr_on_cpu(cpu, MSR_IA32_THERM_STATUS, &l, &h); - - if (l & 0x01) - dprintk("CPU#%d currently thermal throttled\n", cpu); - - if (has_N44_O17_errata[cpu] && - (newstate == DC_25PT || newstate == DC_DFLT)) - newstate = DC_38PT; - - rdmsr_on_cpu(cpu, MSR_IA32_THERM_CONTROL, &l, &h); - if (newstate == DC_DISABLE) { - dprintk("CPU#%d disabling modulation\n", cpu); - wrmsr_on_cpu(cpu, MSR_IA32_THERM_CONTROL, l & ~(1<<4), h); - } else { - dprintk("CPU#%d setting duty cycle to %d%%\n", - cpu, ((125 * newstate) / 10)); - /* bits 63 - 5 : reserved - * bit 4 : enable/disable - * bits 3-1 : duty cycle - * bit 0 : reserved - */ - l = (l & ~14); - l = l | (1<<4) | ((newstate & 0x7)<<1); - wrmsr_on_cpu(cpu, MSR_IA32_THERM_CONTROL, l, h); - } - - return 0; -} - - -static struct cpufreq_frequency_table p4clockmod_table[] = { - {DC_RESV, CPUFREQ_ENTRY_INVALID}, - {DC_DFLT, 0}, - {DC_25PT, 0}, - {DC_38PT, 0}, - {DC_50PT, 0}, - {DC_64PT, 0}, - {DC_75PT, 0}, - {DC_88PT, 0}, - {DC_DISABLE, 0}, - {DC_RESV, CPUFREQ_TABLE_END}, -}; - - -static int cpufreq_p4_target(struct cpufreq_policy *policy, - unsigned int target_freq, - unsigned int relation) -{ - unsigned int newstate = DC_RESV; - struct cpufreq_freqs freqs; - int i; - - if (cpufreq_frequency_table_target(policy, &p4clockmod_table[0], - target_freq, relation, &newstate)) - return -EINVAL; - - freqs.old = cpufreq_p4_get(policy->cpu); - freqs.new = stock_freq * p4clockmod_table[newstate].index / 8; - - if (freqs.new == freqs.old) - return 0; - - /* notifiers */ - for_each_cpu(i, policy->cpus) { - freqs.cpu = i; - cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); - } - - /* run on each logical CPU, - * see section 13.15.3 of IA32 Intel Architecture Software - * Developer's Manual, Volume 3 - */ - for_each_cpu(i, policy->cpus) - cpufreq_p4_setdc(i, p4clockmod_table[newstate].index); - - /* notifiers */ - for_each_cpu(i, policy->cpus) { - freqs.cpu = i; - cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); - } - - return 0; -} - - -static int cpufreq_p4_verify(struct cpufreq_policy *policy) -{ - return cpufreq_frequency_table_verify(policy, &p4clockmod_table[0]); -} - - -static unsigned int cpufreq_p4_get_frequency(struct cpuinfo_x86 *c) -{ - if (c->x86 == 0x06) { - if (cpu_has(c, X86_FEATURE_EST)) - printk_once(KERN_WARNING PFX "Warning: EST-capable " - "CPU detected. The acpi-cpufreq module offers " - "voltage scaling in addition to frequency " - "scaling. You should use that instead of " - "p4-clockmod, if possible.\n"); - switch (c->x86_model) { - case 0x0E: /* Core */ - case 0x0F: /* Core Duo */ - case 0x16: /* Celeron Core */ - case 0x1C: /* Atom */ - p4clockmod_driver.flags |= CPUFREQ_CONST_LOOPS; - return speedstep_get_frequency(SPEEDSTEP_CPU_PCORE); - case 0x0D: /* Pentium M (Dothan) */ - p4clockmod_driver.flags |= CPUFREQ_CONST_LOOPS; - /* fall through */ - case 0x09: /* Pentium M (Banias) */ - return speedstep_get_frequency(SPEEDSTEP_CPU_PM); - } - } - - if (c->x86 != 0xF) - return 0; - - /* on P-4s, the TSC runs with constant frequency independent whether - * throttling is active or not. */ - p4clockmod_driver.flags |= CPUFREQ_CONST_LOOPS; - - if (speedstep_detect_processor() == SPEEDSTEP_CPU_P4M) { - printk(KERN_WARNING PFX "Warning: Pentium 4-M detected. " - "The speedstep-ich or acpi cpufreq modules offer " - "voltage scaling in addition of frequency scaling. " - "You should use either one instead of p4-clockmod, " - "if possible.\n"); - return speedstep_get_frequency(SPEEDSTEP_CPU_P4M); - } - - return speedstep_get_frequency(SPEEDSTEP_CPU_P4D); -} - - - -static int cpufreq_p4_cpu_init(struct cpufreq_policy *policy) -{ - struct cpuinfo_x86 *c = &cpu_data(policy->cpu); - int cpuid = 0; - unsigned int i; - -#ifdef CONFIG_SMP - cpumask_copy(policy->cpus, cpu_sibling_mask(policy->cpu)); -#endif - - /* Errata workaround */ - cpuid = (c->x86 << 8) | (c->x86_model << 4) | c->x86_mask; - switch (cpuid) { - case 0x0f07: - case 0x0f0a: - case 0x0f11: - case 0x0f12: - has_N44_O17_errata[policy->cpu] = 1; - dprintk("has errata -- disabling low frequencies\n"); - } - - if (speedstep_detect_processor() == SPEEDSTEP_CPU_P4D && - c->x86_model < 2) { - /* switch to maximum frequency and measure result */ - cpufreq_p4_setdc(policy->cpu, DC_DISABLE); - recalibrate_cpu_khz(); - } - /* get max frequency */ - stock_freq = cpufreq_p4_get_frequency(c); - if (!stock_freq) - return -EINVAL; - - /* table init */ - for (i = 1; (p4clockmod_table[i].frequency != CPUFREQ_TABLE_END); i++) { - if ((i < 2) && (has_N44_O17_errata[policy->cpu])) - p4clockmod_table[i].frequency = CPUFREQ_ENTRY_INVALID; - else - p4clockmod_table[i].frequency = (stock_freq * i)/8; - } - cpufreq_frequency_table_get_attr(p4clockmod_table, policy->cpu); - - /* cpuinfo and default policy values */ - - /* the transition latency is set to be 1 higher than the maximum - * transition latency of the ondemand governor */ - policy->cpuinfo.transition_latency = 10000001; - policy->cur = stock_freq; - - return cpufreq_frequency_table_cpuinfo(policy, &p4clockmod_table[0]); -} - - -static int cpufreq_p4_cpu_exit(struct cpufreq_policy *policy) -{ - cpufreq_frequency_table_put_attr(policy->cpu); - return 0; -} - -static unsigned int cpufreq_p4_get(unsigned int cpu) -{ - u32 l, h; - - rdmsr_on_cpu(cpu, MSR_IA32_THERM_CONTROL, &l, &h); - - if (l & 0x10) { - l = l >> 1; - l &= 0x7; - } else - l = DC_DISABLE; - - if (l != DC_DISABLE) - return stock_freq * l / 8; - - return stock_freq; -} - -static struct freq_attr *p4clockmod_attr[] = { - &cpufreq_freq_attr_scaling_available_freqs, - NULL, -}; - -static struct cpufreq_driver p4clockmod_driver = { - .verify = cpufreq_p4_verify, - .target = cpufreq_p4_target, - .init = cpufreq_p4_cpu_init, - .exit = cpufreq_p4_cpu_exit, - .get = cpufreq_p4_get, - .name = "p4-clockmod", - .owner = THIS_MODULE, - .attr = p4clockmod_attr, -}; - - -static int __init cpufreq_p4_init(void) -{ - struct cpuinfo_x86 *c = &cpu_data(0); - int ret; - - /* - * THERM_CONTROL is architectural for IA32 now, so - * we can rely on the capability checks - */ - if (c->x86_vendor != X86_VENDOR_INTEL) - return -ENODEV; - - if (!test_cpu_cap(c, X86_FEATURE_ACPI) || - !test_cpu_cap(c, X86_FEATURE_ACC)) - return -ENODEV; - - ret = cpufreq_register_driver(&p4clockmod_driver); - if (!ret) - printk(KERN_INFO PFX "P4/Xeon(TM) CPU On-Demand Clock " - "Modulation available\n"); - - return ret; -} - - -static void __exit cpufreq_p4_exit(void) -{ - cpufreq_unregister_driver(&p4clockmod_driver); -} - - -MODULE_AUTHOR("Zwane Mwaikambo <zwane@commfireservices.com>"); -MODULE_DESCRIPTION("cpufreq driver for Pentium(TM) 4/Xeon(TM)"); -MODULE_LICENSE("GPL"); - -late_initcall(cpufreq_p4_init); -module_exit(cpufreq_p4_exit); diff --git a/arch/x86/kernel/cpu/cpufreq/pcc-cpufreq.c b/arch/x86/kernel/cpu/cpufreq/pcc-cpufreq.c deleted file mode 100644 index 755a31e0f5b0..000000000000 --- a/arch/x86/kernel/cpu/cpufreq/pcc-cpufreq.c +++ /dev/null @@ -1,624 +0,0 @@ -/* - * pcc-cpufreq.c - Processor Clocking Control firmware cpufreq interface - * - * Copyright (C) 2009 Red Hat, Matthew Garrett <mjg@redhat.com> - * Copyright (C) 2009 Hewlett-Packard Development Company, L.P. - * Nagananda Chumbalkar <nagananda.chumbalkar@hp.com> - * - * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; version 2 of the License. - * - * This program is distributed in the hope that it will be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or NON - * INFRINGEMENT. See the GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License along - * with this program; if not, write to the Free Software Foundation, Inc., - * 675 Mass Ave, Cambridge, MA 02139, USA. - * - * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ - */ - -#include <linux/kernel.h> -#include <linux/module.h> -#include <linux/init.h> -#include <linux/smp.h> -#include <linux/sched.h> -#include <linux/cpufreq.h> -#include <linux/compiler.h> -#include <linux/slab.h> - -#include <linux/acpi.h> -#include <linux/io.h> -#include <linux/spinlock.h> -#include <linux/uaccess.h> - -#include <acpi/processor.h> - -#define PCC_VERSION "1.00.00" -#define POLL_LOOPS 300 - -#define CMD_COMPLETE 0x1 -#define CMD_GET_FREQ 0x0 -#define CMD_SET_FREQ 0x1 - -#define BUF_SZ 4 - -#define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, \ - "pcc-cpufreq", msg) - -struct pcc_register_resource { - u8 descriptor; - u16 length; - u8 space_id; - u8 bit_width; - u8 bit_offset; - u8 access_size; - u64 address; -} __attribute__ ((packed)); - -struct pcc_memory_resource { - u8 descriptor; - u16 length; - u8 space_id; - u8 resource_usage; - u8 type_specific; - u64 granularity; - u64 minimum; - u64 maximum; - u64 translation_offset; - u64 address_length; -} __attribute__ ((packed)); - -static struct cpufreq_driver pcc_cpufreq_driver; - -struct pcc_header { - u32 signature; - u16 length; - u8 major; - u8 minor; - u32 features; - u16 command; - u16 status; - u32 latency; - u32 minimum_time; - u32 maximum_time; - u32 nominal; - u32 throttled_frequency; - u32 minimum_frequency; -}; - -static void __iomem *pcch_virt_addr; -static struct pcc_header __iomem *pcch_hdr; - -static DEFINE_SPINLOCK(pcc_lock); - -static struct acpi_generic_address doorbell; - -static u64 doorbell_preserve; -static u64 doorbell_write; - -static u8 OSC_UUID[16] = {0x63, 0x9B, 0x2C, 0x9F, 0x70, 0x91, 0x49, 0x1f, - 0xBB, 0x4F, 0xA5, 0x98, 0x2F, 0xA1, 0xB5, 0x46}; - -struct pcc_cpu { - u32 input_offset; - u32 output_offset; -}; - -static struct pcc_cpu __percpu *pcc_cpu_info; - -static int pcc_cpufreq_verify(struct cpufreq_policy *policy) -{ - cpufreq_verify_within_limits(policy, policy->cpuinfo.min_freq, - policy->cpuinfo.max_freq); - return 0; -} - -static inline void pcc_cmd(void) -{ - u64 doorbell_value; - int i; - - acpi_read(&doorbell_value, &doorbell); - acpi_write((doorbell_value & doorbell_preserve) | doorbell_write, - &doorbell); - - for (i = 0; i < POLL_LOOPS; i++) { - if (ioread16(&pcch_hdr->status) & CMD_COMPLETE) - break; - } -} - -static inline void pcc_clear_mapping(void) -{ - if (pcch_virt_addr) - iounmap(pcch_virt_addr); - pcch_virt_addr = NULL; -} - -static unsigned int pcc_get_freq(unsigned int cpu) -{ - struct pcc_cpu *pcc_cpu_data; - unsigned int curr_freq; - unsigned int freq_limit; - u16 status; - u32 input_buffer; - u32 output_buffer; - - spin_lock(&pcc_lock); - - dprintk("get: get_freq for CPU %d\n", cpu); - pcc_cpu_data = per_cpu_ptr(pcc_cpu_info, cpu); - - input_buffer = 0x1; - iowrite32(input_buffer, - (pcch_virt_addr + pcc_cpu_data->input_offset)); - iowrite16(CMD_GET_FREQ, &pcch_hdr->command); - - pcc_cmd(); - - output_buffer = - ioread32(pcch_virt_addr + pcc_cpu_data->output_offset); - - /* Clear the input buffer - we are done with the current command */ - memset_io((pcch_virt_addr + pcc_cpu_data->input_offset), 0, BUF_SZ); - - status = ioread16(&pcch_hdr->status); - if (status != CMD_COMPLETE) { - dprintk("get: FAILED: for CPU %d, status is %d\n", - cpu, status); - goto cmd_incomplete; - } - iowrite16(0, &pcch_hdr->status); - curr_freq = (((ioread32(&pcch_hdr->nominal) * (output_buffer & 0xff)) - / 100) * 1000); - - dprintk("get: SUCCESS: (virtual) output_offset for cpu %d is " - "0x%x, contains a value of: 0x%x. Speed is: %d MHz\n", - cpu, (pcch_virt_addr + pcc_cpu_data->output_offset), - output_buffer, curr_freq); - - freq_limit = (output_buffer >> 8) & 0xff; - if (freq_limit != 0xff) { - dprintk("get: frequency for cpu %d is being temporarily" - " capped at %d\n", cpu, curr_freq); - } - - spin_unlock(&pcc_lock); - return curr_freq; - -cmd_incomplete: - iowrite16(0, &pcch_hdr->status); - spin_unlock(&pcc_lock); - return 0; -} - -static int pcc_cpufreq_target(struct cpufreq_policy *policy, - unsigned int target_freq, - unsigned int relation) -{ - struct pcc_cpu *pcc_cpu_data; - struct cpufreq_freqs freqs; - u16 status; - u32 input_buffer; - int cpu; - - spin_lock(&pcc_lock); - cpu = policy->cpu; - pcc_cpu_data = per_cpu_ptr(pcc_cpu_info, cpu); - - dprintk("target: CPU %d should go to target freq: %d " - "(virtual) input_offset is 0x%x\n", - cpu, target_freq, - (pcch_virt_addr + pcc_cpu_data->input_offset)); - - freqs.new = target_freq; - freqs.cpu = cpu; - cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); - - input_buffer = 0x1 | (((target_freq * 100) - / (ioread32(&pcch_hdr->nominal) * 1000)) << 8); - iowrite32(input_buffer, - (pcch_virt_addr + pcc_cpu_data->input_offset)); - iowrite16(CMD_SET_FREQ, &pcch_hdr->command); - - pcc_cmd(); - - /* Clear the input buffer - we are done with the current command */ - memset_io((pcch_virt_addr + pcc_cpu_data->input_offset), 0, BUF_SZ); - - status = ioread16(&pcch_hdr->status); - if (status != CMD_COMPLETE) { - dprintk("target: FAILED for cpu %d, with status: 0x%x\n", - cpu, status); - goto cmd_incomplete; - } - iowrite16(0, &pcch_hdr->status); - - cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); - dprintk("target: was SUCCESSFUL for cpu %d\n", cpu); - spin_unlock(&pcc_lock); - - return 0; - -cmd_incomplete: - iowrite16(0, &pcch_hdr->status); - spin_unlock(&pcc_lock); - return -EINVAL; -} - -static int pcc_get_offset(int cpu) -{ - acpi_status status; - struct acpi_buffer buffer = {ACPI_ALLOCATE_BUFFER, NULL}; - union acpi_object *pccp, *offset; - struct pcc_cpu *pcc_cpu_data; - struct acpi_processor *pr; - int ret = 0; - - pr = per_cpu(processors, cpu); - pcc_cpu_data = per_cpu_ptr(pcc_cpu_info, cpu); - - status = acpi_evaluate_object(pr->handle, "PCCP", NULL, &buffer); - if (ACPI_FAILURE(status)) - return -ENODEV; - - pccp = buffer.pointer; - if (!pccp || pccp->type != ACPI_TYPE_PACKAGE) { - ret = -ENODEV; - goto out_free; - }; - - offset = &(pccp->package.elements[0]); - if (!offset || offset->type != ACPI_TYPE_INTEGER) { - ret = -ENODEV; - goto out_free; - } - - pcc_cpu_data->input_offset = offset->integer.value; - - offset = &(pccp->package.elements[1]); - if (!offset || offset->type != ACPI_TYPE_INTEGER) { - ret = -ENODEV; - goto out_free; - } - - pcc_cpu_data->output_offset = offset->integer.value; - - memset_io((pcch_virt_addr + pcc_cpu_data->input_offset), 0, BUF_SZ); - memset_io((pcch_virt_addr + pcc_cpu_data->output_offset), 0, BUF_SZ); - - dprintk("pcc_get_offset: for CPU %d: pcc_cpu_data " - "input_offset: 0x%x, pcc_cpu_data output_offset: 0x%x\n", - cpu, pcc_cpu_data->input_offset, pcc_cpu_data->output_offset); -out_free: - kfree(buffer.pointer); - return ret; -} - -static int __init pcc_cpufreq_do_osc(acpi_handle *handle) -{ - acpi_status status; - struct acpi_object_list input; - struct acpi_buffer output = {ACPI_ALLOCATE_BUFFER, NULL}; - union acpi_object in_params[4]; - union acpi_object *out_obj; - u32 capabilities[2]; - u32 errors; - u32 supported; - int ret = 0; - - input.count = 4; - input.pointer = in_params; - in_params[0].type = ACPI_TYPE_BUFFER; - in_params[0].buffer.length = 16; - in_params[0].buffer.pointer = OSC_UUID; - in_params[1].type = ACPI_TYPE_INTEGER; - in_params[1].integer.value = 1; - in_params[2].type = ACPI_TYPE_INTEGER; - in_params[2].integer.value = 2; - in_params[3].type = ACPI_TYPE_BUFFER; - in_params[3].buffer.length = 8; - in_params[3].buffer.pointer = (u8 *)&capabilities; - - capabilities[0] = OSC_QUERY_ENABLE; - capabilities[1] = 0x1; - - status = acpi_evaluate_object(*handle, "_OSC", &input, &output); - if (ACPI_FAILURE(status)) - return -ENODEV; - - if (!output.length) - return -ENODEV; - - out_obj = output.pointer; - if (out_obj->type != ACPI_TYPE_BUFFER) { - ret = -ENODEV; - goto out_free; - } - - errors = *((u32 *)out_obj->buffer.pointer) & ~(1 << 0); - if (errors) { - ret = -ENODEV; - goto out_free; - } - - supported = *((u32 *)(out_obj->buffer.pointer + 4)); - if (!(supported & 0x1)) { - ret = -ENODEV; - goto out_free; - } - - kfree(output.pointer); - capabilities[0] = 0x0; - capabilities[1] = 0x1; - - status = acpi_evaluate_object(*handle, "_OSC", &input, &output); - if (ACPI_FAILURE(status)) - return -ENODEV; - - if (!output.length) - return -ENODEV; - - out_obj = output.pointer; - if (out_obj->type != ACPI_TYPE_BUFFER) { - ret = -ENODEV; - goto out_free; - } - - errors = *((u32 *)out_obj->buffer.pointer) & ~(1 << 0); - if (errors) { - ret = -ENODEV; - goto out_free; - } - - supported = *((u32 *)(out_obj->buffer.pointer + 4)); - if (!(supported & 0x1)) { - ret = -ENODEV; - goto out_free; - } - -out_free: - kfree(output.pointer); - return ret; -} - -static int __init pcc_cpufreq_probe(void) -{ - acpi_status status; - struct acpi_buffer output = {ACPI_ALLOCATE_BUFFER, NULL}; - struct pcc_memory_resource *mem_resource; - struct pcc_register_resource *reg_resource; - union acpi_object *out_obj, *member; - acpi_handle handle, osc_handle, pcch_handle; - int ret = 0; - - status = acpi_get_handle(NULL, "\\_SB", &handle); - if (ACPI_FAILURE(status)) - return -ENODEV; - - status = acpi_get_handle(handle, "PCCH", &pcch_handle); - if (ACPI_FAILURE(status)) - return -ENODEV; - - status = acpi_get_handle(handle, "_OSC", &osc_handle); - if (ACPI_SUCCESS(status)) { - ret = pcc_cpufreq_do_osc(&osc_handle); - if (ret) - dprintk("probe: _OSC evaluation did not succeed\n"); - /* Firmware's use of _OSC is optional */ - ret = 0; - } - - status = acpi_evaluate_object(handle, "PCCH", NULL, &output); - if (ACPI_FAILURE(status)) - return -ENODEV; - - out_obj = output.pointer; - if (out_obj->type != ACPI_TYPE_PACKAGE) { - ret = -ENODEV; - goto out_free; - } - - member = &out_obj->package.elements[0]; - if (member->type != ACPI_TYPE_BUFFER) { - ret = -ENODEV; - goto out_free; - } - - mem_resource = (struct pcc_memory_resource *)member->buffer.pointer; - - dprintk("probe: mem_resource descriptor: 0x%x," - " length: %d, space_id: %d, resource_usage: %d," - " type_specific: %d, granularity: 0x%llx," - " minimum: 0x%llx, maximum: 0x%llx," - " translation_offset: 0x%llx, address_length: 0x%llx\n", - mem_resource->descriptor, mem_resource->length, - mem_resource->space_id, mem_resource->resource_usage, - mem_resource->type_specific, mem_resource->granularity, - mem_resource->minimum, mem_resource->maximum, - mem_resource->translation_offset, - mem_resource->address_length); - - if (mem_resource->space_id != ACPI_ADR_SPACE_SYSTEM_MEMORY) { - ret = -ENODEV; - goto out_free; - } - - pcch_virt_addr = ioremap_nocache(mem_resource->minimum, - mem_resource->address_length); - if (pcch_virt_addr == NULL) { - dprintk("probe: could not map shared mem region\n"); - goto out_free; - } - pcch_hdr = pcch_virt_addr; - - dprintk("probe: PCCH header (virtual) addr: 0x%p\n", pcch_hdr); - dprintk("probe: PCCH header is at physical address: 0x%llx," - " signature: 0x%x, length: %d bytes, major: %d, minor: %d," - " supported features: 0x%x, command field: 0x%x," - " status field: 0x%x, nominal latency: %d us\n", - mem_resource->minimum, ioread32(&pcch_hdr->signature), - ioread16(&pcch_hdr->length), ioread8(&pcch_hdr->major), - ioread8(&pcch_hdr->minor), ioread32(&pcch_hdr->features), - ioread16(&pcch_hdr->command), ioread16(&pcch_hdr->status), - ioread32(&pcch_hdr->latency)); - - dprintk("probe: min time between commands: %d us," - " max time between commands: %d us," - " nominal CPU frequency: %d MHz," - " minimum CPU frequency: %d MHz," - " minimum CPU frequency without throttling: %d MHz\n", - ioread32(&pcch_hdr->minimum_time), - ioread32(&pcch_hdr->maximum_time), - ioread32(&pcch_hdr->nominal), - ioread32(&pcch_hdr->throttled_frequency), - ioread32(&pcch_hdr->minimum_frequency)); - - member = &out_obj->package.elements[1]; - if (member->type != ACPI_TYPE_BUFFER) { - ret = -ENODEV; - goto pcch_free; - } - - reg_resource = (struct pcc_register_resource *)member->buffer.pointer; - - doorbell.space_id = reg_resource->space_id; - doorbell.bit_width = reg_resource->bit_width; - doorbell.bit_offset = reg_resource->bit_offset; - doorbell.access_width = 64; - doorbell.address = reg_resource->address; - - dprintk("probe: doorbell: space_id is %d, bit_width is %d, " - "bit_offset is %d, access_width is %d, address is 0x%llx\n", - doorbell.space_id, doorbell.bit_width, doorbell.bit_offset, - doorbell.access_width, reg_resource->address); - - member = &out_obj->package.elements[2]; - if (member->type != ACPI_TYPE_INTEGER) { - ret = -ENODEV; - goto pcch_free; - } - - doorbell_preserve = member->integer.value; - - member = &out_obj->package.elements[3]; - if (member->type != ACPI_TYPE_INTEGER) { - ret = -ENODEV; - goto pcch_free; - } - - doorbell_write = member->integer.value; - - dprintk("probe: doorbell_preserve: 0x%llx," - " doorbell_write: 0x%llx\n", - doorbell_preserve, doorbell_write); - - pcc_cpu_info = alloc_percpu(struct pcc_cpu); - if (!pcc_cpu_info) { - ret = -ENOMEM; - goto pcch_free; - } - - printk(KERN_DEBUG "pcc-cpufreq: (v%s) driver loaded with frequency" - " limits: %d MHz, %d MHz\n", PCC_VERSION, - ioread32(&pcch_hdr->minimum_frequency), - ioread32(&pcch_hdr->nominal)); - kfree(output.pointer); - return ret; -pcch_free: - pcc_clear_mapping(); -out_free: - kfree(output.pointer); - return ret; -} - -static int pcc_cpufreq_cpu_init(struct cpufreq_policy *policy) -{ - unsigned int cpu = policy->cpu; - unsigned int result = 0; - - if (!pcch_virt_addr) { - result = -1; - goto out; - } - - result = pcc_get_offset(cpu); - if (result) { - dprintk("init: PCCP evaluation failed\n"); - goto out; - } - - policy->max = policy->cpuinfo.max_freq = - ioread32(&pcch_hdr->nominal) * 1000; - policy->min = policy->cpuinfo.min_freq = - ioread32(&pcch_hdr->minimum_frequency) * 1000; - policy->cur = pcc_get_freq(cpu); - - if (!policy->cur) { - dprintk("init: Unable to get current CPU frequency\n"); - result = -EINVAL; - goto out; - } - - dprintk("init: policy->max is %d, policy->min is %d\n", - policy->max, policy->min); -out: - return result; -} - -static int pcc_cpufreq_cpu_exit(struct cpufreq_policy *policy) -{ - return 0; -} - -static struct cpufreq_driver pcc_cpufreq_driver = { - .flags = CPUFREQ_CONST_LOOPS, - .get = pcc_get_freq, - .verify = pcc_cpufreq_verify, - .target = pcc_cpufreq_target, - .init = pcc_cpufreq_cpu_init, - .exit = pcc_cpufreq_cpu_exit, - .name = "pcc-cpufreq", - .owner = THIS_MODULE, -}; - -static int __init pcc_cpufreq_init(void) -{ - int ret; - - if (acpi_disabled) - return 0; - - ret = pcc_cpufreq_probe(); - if (ret) { - dprintk("pcc_cpufreq_init: PCCH evaluation failed\n"); - return ret; - } - - ret = cpufreq_register_driver(&pcc_cpufreq_driver); - - return ret; -} - -static void __exit pcc_cpufreq_exit(void) -{ - cpufreq_unregister_driver(&pcc_cpufreq_driver); - - pcc_clear_mapping(); - - free_percpu(pcc_cpu_info); -} - -MODULE_AUTHOR("Matthew Garrett, Naga Chumbalkar"); -MODULE_VERSION(PCC_VERSION); -MODULE_DESCRIPTION("Processor Clocking Control interface driver"); -MODULE_LICENSE("GPL"); - -late_initcall(pcc_cpufreq_init); -module_exit(pcc_cpufreq_exit); diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k6.c b/arch/x86/kernel/cpu/cpufreq/powernow-k6.c deleted file mode 100644 index b3379d6a5c57..000000000000 --- a/arch/x86/kernel/cpu/cpufreq/powernow-k6.c +++ /dev/null @@ -1,261 +0,0 @@ -/* - * This file was based upon code in Powertweak Linux (http://powertweak.sf.net) - * (C) 2000-2003 Dave Jones, Arjan van de Ven, Janne Pänkälä, - * Dominik Brodowski. - * - * Licensed under the terms of the GNU GPL License version 2. - * - * BIG FAT DISCLAIMER: Work in progress code. Possibly *dangerous* - */ - -#include <linux/kernel.h> -#include <linux/module.h> -#include <linux/init.h> -#include <linux/cpufreq.h> -#include <linux/ioport.h> -#include <linux/timex.h> -#include <linux/io.h> - -#include <asm/msr.h> - -#define POWERNOW_IOPORT 0xfff0 /* it doesn't matter where, as long - as it is unused */ - -#define PFX "powernow-k6: " -static unsigned int busfreq; /* FSB, in 10 kHz */ -static unsigned int max_multiplier; - - -/* Clock ratio multiplied by 10 - see table 27 in AMD#23446 */ -static struct cpufreq_frequency_table clock_ratio[] = { - {45, /* 000 -> 4.5x */ 0}, - {50, /* 001 -> 5.0x */ 0}, - {40, /* 010 -> 4.0x */ 0}, - {55, /* 011 -> 5.5x */ 0}, - {20, /* 100 -> 2.0x */ 0}, - {30, /* 101 -> 3.0x */ 0}, - {60, /* 110 -> 6.0x */ 0}, - {35, /* 111 -> 3.5x */ 0}, - {0, CPUFREQ_TABLE_END} -}; - - -/** - * powernow_k6_get_cpu_multiplier - returns the current FSB multiplier - * - * Returns the current setting of the frequency multiplier. Core clock - * speed is frequency of the Front-Side Bus multiplied with this value. - */ -static int powernow_k6_get_cpu_multiplier(void) -{ - u64 invalue = 0; - u32 msrval; - - msrval = POWERNOW_IOPORT + 0x1; - wrmsr(MSR_K6_EPMR, msrval, 0); /* enable the PowerNow port */ - invalue = inl(POWERNOW_IOPORT + 0x8); - msrval = POWERNOW_IOPORT + 0x0; - wrmsr(MSR_K6_EPMR, msrval, 0); /* disable it again */ - - return clock_ratio[(invalue >> 5)&7].index; -} - - -/** - * powernow_k6_set_state - set the PowerNow! multiplier - * @best_i: clock_ratio[best_i] is the target multiplier - * - * Tries to change the PowerNow! multiplier - */ -static void powernow_k6_set_state(unsigned int best_i) -{ - unsigned long outvalue = 0, invalue = 0; - unsigned long msrval; - struct cpufreq_freqs freqs; - - if (clock_ratio[best_i].index > max_multiplier) { - printk(KERN_ERR PFX "invalid target frequency\n"); - return; - } - - freqs.old = busfreq * powernow_k6_get_cpu_multiplier(); - freqs.new = busfreq * clock_ratio[best_i].index; - freqs.cpu = 0; /* powernow-k6.c is UP only driver */ - - cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); - - /* we now need to transform best_i to the BVC format, see AMD#23446 */ - - outvalue = (1<<12) | (1<<10) | (1<<9) | (best_i<<5); - - msrval = POWERNOW_IOPORT + 0x1; - wrmsr(MSR_K6_EPMR, msrval, 0); /* enable the PowerNow port */ - invalue = inl(POWERNOW_IOPORT + 0x8); - invalue = invalue & 0xf; - outvalue = outvalue | invalue; - outl(outvalue , (POWERNOW_IOPORT + 0x8)); - msrval = POWERNOW_IOPORT + 0x0; - wrmsr(MSR_K6_EPMR, msrval, 0); /* disable it again */ - - cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); - - return; -} - - -/** - * powernow_k6_verify - verifies a new CPUfreq policy - * @policy: new policy - * - * Policy must be within lowest and highest possible CPU Frequency, - * and at least one possible state must be within min and max. - */ -static int powernow_k6_verify(struct cpufreq_policy *policy) -{ - return cpufreq_frequency_table_verify(policy, &clock_ratio[0]); -} - - -/** - * powernow_k6_setpolicy - sets a new CPUFreq policy - * @policy: new policy - * @target_freq: the target frequency - * @relation: how that frequency relates to achieved frequency - * (CPUFREQ_RELATION_L or CPUFREQ_RELATION_H) - * - * sets a new CPUFreq policy - */ -static int powernow_k6_target(struct cpufreq_policy *policy, - unsigned int target_freq, - unsigned int relation) -{ - unsigned int newstate = 0; - - if (cpufreq_frequency_table_target(policy, &clock_ratio[0], - target_freq, relation, &newstate)) - return -EINVAL; - - powernow_k6_set_state(newstate); - - return 0; -} - - -static int powernow_k6_cpu_init(struct cpufreq_policy *policy) -{ - unsigned int i, f; - int result; - - if (policy->cpu != 0) - return -ENODEV; - - /* get frequencies */ - max_multiplier = powernow_k6_get_cpu_multiplier(); - busfreq = cpu_khz / max_multiplier; - - /* table init */ - for (i = 0; (clock_ratio[i].frequency != CPUFREQ_TABLE_END); i++) { - f = clock_ratio[i].index; - if (f > max_multiplier) - clock_ratio[i].frequency = CPUFREQ_ENTRY_INVALID; - else - clock_ratio[i].frequency = busfreq * f; - } - - /* cpuinfo and default policy values */ - policy->cpuinfo.transition_latency = 200000; - policy->cur = busfreq * max_multiplier; - - result = cpufreq_frequency_table_cpuinfo(policy, clock_ratio); - if (result) - return result; - - cpufreq_frequency_table_get_attr(clock_ratio, policy->cpu); - - return 0; -} - - -static int powernow_k6_cpu_exit(struct cpufreq_policy *policy) -{ - unsigned int i; - for (i = 0; i < 8; i++) { - if (i == max_multiplier) - powernow_k6_set_state(i); - } - cpufreq_frequency_table_put_attr(policy->cpu); - return 0; -} - -static unsigned int powernow_k6_get(unsigned int cpu) -{ - unsigned int ret; - ret = (busfreq * powernow_k6_get_cpu_multiplier()); - return ret; -} - -static struct freq_attr *powernow_k6_attr[] = { - &cpufreq_freq_attr_scaling_available_freqs, - NULL, -}; - -static struct cpufreq_driver powernow_k6_driver = { - .verify = powernow_k6_verify, - .target = powernow_k6_target, - .init = powernow_k6_cpu_init, - .exit = powernow_k6_cpu_exit, - .get = powernow_k6_get, - .name = "powernow-k6", - .owner = THIS_MODULE, - .attr = powernow_k6_attr, -}; - - -/** - * powernow_k6_init - initializes the k6 PowerNow! CPUFreq driver - * - * Initializes the K6 PowerNow! support. Returns -ENODEV on unsupported - * devices, -EINVAL or -ENOMEM on problems during initiatization, and zero - * on success. - */ -static int __init powernow_k6_init(void) -{ - struct cpuinfo_x86 *c = &cpu_data(0); - - if ((c->x86_vendor != X86_VENDOR_AMD) || (c->x86 != 5) || - ((c->x86_model != 12) && (c->x86_model != 13))) - return -ENODEV; - - if (!request_region(POWERNOW_IOPORT, 16, "PowerNow!")) { - printk(KERN_INFO PFX "PowerNow IOPORT region already used.\n"); - return -EIO; - } - - if (cpufreq_register_driver(&powernow_k6_driver)) { - release_region(POWERNOW_IOPORT, 16); - return -EINVAL; - } - - return 0; -} - - -/** - * powernow_k6_exit - unregisters AMD K6-2+/3+ PowerNow! support - * - * Unregisters AMD K6-2+ / K6-3+ PowerNow! support. - */ -static void __exit powernow_k6_exit(void) -{ - cpufreq_unregister_driver(&powernow_k6_driver); - release_region(POWERNOW_IOPORT, 16); -} - - -MODULE_AUTHOR("Arjan van de Ven, Dave Jones <davej@redhat.com>, " - "Dominik Brodowski <linux@brodo.de>"); -MODULE_DESCRIPTION("PowerNow! driver for AMD K6-2+ / K6-3+ processors."); -MODULE_LICENSE("GPL"); - -module_init(powernow_k6_init); -module_exit(powernow_k6_exit); diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k7.c b/arch/x86/kernel/cpu/cpufreq/powernow-k7.c deleted file mode 100644 index 4a45fd6e41ba..000000000000 --- a/arch/x86/kernel/cpu/cpufreq/powernow-k7.c +++ /dev/null @@ -1,752 +0,0 @@ -/* - * AMD K7 Powernow driver. - * (C) 2003 Dave Jones on behalf of SuSE Labs. - * (C) 2003-2004 Dave Jones <davej@redhat.com> - * - * Licensed under the terms of the GNU GPL License version 2. - * Based upon datasheets & sample CPUs kindly provided by AMD. - * - * Errata 5: - * CPU may fail to execute a FID/VID change in presence of interrupt. - * - We cli/sti on stepping A0 CPUs around the FID/VID transition. - * Errata 15: - * CPU with half frequency multipliers may hang upon wakeup from disconnect. - * - We disable half multipliers if ACPI is used on A0 stepping CPUs. - */ - -#include <linux/kernel.h> -#include <linux/module.h> -#include <linux/moduleparam.h> -#include <linux/init.h> -#include <linux/cpufreq.h> -#include <linux/slab.h> -#include <linux/string.h> -#include <linux/dmi.h> -#include <linux/timex.h> -#include <linux/io.h> - -#include <asm/timer.h> /* Needed for recalibrate_cpu_khz() */ -#include <asm/msr.h> -#include <asm/system.h> - -#ifdef CONFIG_X86_POWERNOW_K7_ACPI -#include <linux/acpi.h> -#include <acpi/processor.h> -#endif - -#include "powernow-k7.h" - -#define PFX "powernow: " - - -struct psb_s { - u8 signature[10]; - u8 tableversion; - u8 flags; - u16 settlingtime; - u8 reserved1; - u8 numpst; -}; - -struct pst_s { - u32 cpuid; - u8 fsbspeed; - u8 maxfid; - u8 startvid; - u8 numpstates; -}; - -#ifdef CONFIG_X86_POWERNOW_K7_ACPI -union powernow_acpi_control_t { - struct { - unsigned long fid:5, - vid:5, - sgtc:20, - res1:2; - } bits; - unsigned long val; -}; -#endif - -#ifdef CONFIG_CPU_FREQ_DEBUG -/* divide by 1000 to get VCore voltage in V. */ -static const int mobile_vid_table[32] = { - 2000, 1950, 1900, 1850, 1800, 1750, 1700, 1650, - 1600, 1550, 1500, 1450, 1400, 1350, 1300, 0, - 1275, 1250, 1225, 1200, 1175, 1150, 1125, 1100, - 1075, 1050, 1025, 1000, 975, 950, 925, 0, -}; -#endif - -/* divide by 10 to get FID. */ -static const int fid_codes[32] = { - 110, 115, 120, 125, 50, 55, 60, 65, - 70, 75, 80, 85, 90, 95, 100, 105, - 30, 190, 40, 200, 130, 135, 140, 210, - 150, 225, 160, 165, 170, 180, -1, -1, -}; - -/* This parameter is used in order to force ACPI instead of legacy method for - * configuration purpose. - */ - -static int acpi_force; - -static struct cpufreq_frequency_table *powernow_table; - -static unsigned int can_scale_bus; -static unsigned int can_scale_vid; -static unsigned int minimum_speed = -1; -static unsigned int maximum_speed; -static unsigned int number_scales; -static unsigned int fsb; -static unsigned int latency; -static char have_a0; - -#define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, \ - "powernow-k7", msg) - -static int check_fsb(unsigned int fsbspeed) -{ - int delta; - unsigned int f = fsb / 1000; - - delta = (fsbspeed > f) ? fsbspeed - f : f - fsbspeed; - return delta < 5; -} - -static int check_powernow(void) -{ - struct cpuinfo_x86 *c = &cpu_data(0); - unsigned int maxei, eax, ebx, ecx, edx; - - if ((c->x86_vendor != X86_VENDOR_AMD) || (c->x86 != 6)) { -#ifdef MODULE - printk(KERN_INFO PFX "This module only works with " - "AMD K7 CPUs\n"); -#endif - return 0; - } - - /* Get maximum capabilities */ - maxei = cpuid_eax(0x80000000); - if (maxei < 0x80000007) { /* Any powernow info ? */ -#ifdef MODULE - printk(KERN_INFO PFX "No powernow capabilities detected\n"); -#endif - return 0; - } - - if ((c->x86_model == 6) && (c->x86_mask == 0)) { - printk(KERN_INFO PFX "K7 660[A0] core detected, " - "enabling errata workarounds\n"); - have_a0 = 1; - } - - cpuid(0x80000007, &eax, &ebx, &ecx, &edx); - - /* Check we can actually do something before we say anything.*/ - if (!(edx & (1 << 1 | 1 << 2))) - return 0; - - printk(KERN_INFO PFX "PowerNOW! Technology present. Can scale: "); - - if (edx & 1 << 1) { - printk("frequency"); - can_scale_bus = 1; - } - - if ((edx & (1 << 1 | 1 << 2)) == 0x6) - printk(" and "); - - if (edx & 1 << 2) { - printk("voltage"); - can_scale_vid = 1; - } - - printk(".\n"); - return 1; -} - -#ifdef CONFIG_X86_POWERNOW_K7_ACPI -static void invalidate_entry(unsigned int entry) -{ - powernow_table[entry].frequency = CPUFREQ_ENTRY_INVALID; -} -#endif - -static int get_ranges(unsigned char *pst) -{ - unsigned int j; - unsigned int speed; - u8 fid, vid; - - powernow_table = kzalloc((sizeof(struct cpufreq_frequency_table) * - (number_scales + 1)), GFP_KERNEL); - if (!powernow_table) - return -ENOMEM; - - for (j = 0 ; j < number_scales; j++) { - fid = *pst++; - - powernow_table[j].frequency = (fsb * fid_codes[fid]) / 10; - powernow_table[j].index = fid; /* lower 8 bits */ - - speed = powernow_table[j].frequency; - - if ((fid_codes[fid] % 10) == 5) { -#ifdef CONFIG_X86_POWERNOW_K7_ACPI - if (have_a0 == 1) - invalidate_entry(j); -#endif - } - - if (speed < minimum_speed) - minimum_speed = speed; - if (speed > maximum_speed) - maximum_speed = speed; - - vid = *pst++; - powernow_table[j].index |= (vid << 8); /* upper 8 bits */ - - dprintk(" FID: 0x%x (%d.%dx [%dMHz]) " - "VID: 0x%x (%d.%03dV)\n", fid, fid_codes[fid] / 10, - fid_codes[fid] % 10, speed/1000, vid, - mobile_vid_table[vid]/1000, - mobile_vid_table[vid]%1000); - } - powernow_table[number_scales].frequency = CPUFREQ_TABLE_END; - powernow_table[number_scales].index = 0; - - return 0; -} - - -static void change_FID(int fid) -{ - union msr_fidvidctl fidvidctl; - - rdmsrl(MSR_K7_FID_VID_CTL, fidvidctl.val); - if (fidvidctl.bits.FID != fid) { - fidvidctl.bits.SGTC = latency; - fidvidctl.bits.FID = fid; - fidvidctl.bits.VIDC = 0; - fidvidctl.bits.FIDC = 1; - wrmsrl(MSR_K7_FID_VID_CTL, fidvidctl.val); - } -} - - -static void change_VID(int vid) -{ - union msr_fidvidctl fidvidctl; - - rdmsrl(MSR_K7_FID_VID_CTL, fidvidctl.val); - if (fidvidctl.bits.VID != vid) { - fidvidctl.bits.SGTC = latency; - fidvidctl.bits.VID = vid; - fidvidctl.bits.FIDC = 0; - fidvidctl.bits.VIDC = 1; - wrmsrl(MSR_K7_FID_VID_CTL, fidvidctl.val); - } -} - - -static void change_speed(unsigned int index) -{ - u8 fid, vid; - struct cpufreq_freqs freqs; - union msr_fidvidstatus fidvidstatus; - int cfid; - - /* fid are the lower 8 bits of the index we stored into - * the cpufreq frequency table in powernow_decode_bios, - * vid are the upper 8 bits. - */ - - fid = powernow_table[index].index & 0xFF; - vid = (powernow_table[index].index & 0xFF00) >> 8; - - freqs.cpu = 0; - - rdmsrl(MSR_K7_FID_VID_STATUS, fidvidstatus.val); - cfid = fidvidstatus.bits.CFID; - freqs.old = fsb * fid_codes[cfid] / 10; - - freqs.new = powernow_table[index].frequency; - - cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); - - /* Now do the magic poking into the MSRs. */ - - if (have_a0 == 1) /* A0 errata 5 */ - local_irq_disable(); - - if (freqs.old > freqs.new) { - /* Going down, so change FID first */ - change_FID(fid); - change_VID(vid); - } else { - /* Going up, so change VID first */ - change_VID(vid); - change_FID(fid); - } - - - if (have_a0 == 1) - local_irq_enable(); - - cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); -} - - -#ifdef CONFIG_X86_POWERNOW_K7_ACPI - -static struct acpi_processor_performance *acpi_processor_perf; - -static int powernow_acpi_init(void) -{ - int i; - int retval = 0; - union powernow_acpi_control_t pc; - - if (acpi_processor_perf != NULL && powernow_table != NULL) { - retval = -EINVAL; - goto err0; - } - - acpi_processor_perf = kzalloc(sizeof(struct acpi_processor_performance), - GFP_KERNEL); - if (!acpi_processor_perf) { - retval = -ENOMEM; - goto err0; - } - - if (!zalloc_cpumask_var(&acpi_processor_perf->shared_cpu_map, - GFP_KERNEL)) { - retval = -ENOMEM; - goto err05; - } - - if (acpi_processor_register_performance(acpi_processor_perf, 0)) { - retval = -EIO; - goto err1; - } - - if (acpi_processor_perf->control_register.space_id != - ACPI_ADR_SPACE_FIXED_HARDWARE) { - retval = -ENODEV; - goto err2; - } - - if (acpi_processor_perf->status_register.space_id != - ACPI_ADR_SPACE_FIXED_HARDWARE) { - retval = -ENODEV; - goto err2; - } - - number_scales = acpi_processor_perf->state_count; - - if (number_scales < 2) { - retval = -ENODEV; - goto err2; - } - - powernow_table = kzalloc((sizeof(struct cpufreq_frequency_table) * - (number_scales + 1)), GFP_KERNEL); - if (!powernow_table) { - retval = -ENOMEM; - goto err2; - } - - pc.val = (unsigned long) acpi_processor_perf->states[0].control; - for (i = 0; i < number_scales; i++) { - u8 fid, vid; - struct acpi_processor_px *state = - &acpi_processor_perf->states[i]; - unsigned int speed, speed_mhz; - - pc.val = (unsigned long) state->control; - dprintk("acpi: P%d: %d MHz %d mW %d uS control %08x SGTC %d\n", - i, - (u32) state->core_frequency, - (u32) state->power, - (u32) state->transition_latency, - (u32) state->control, - pc.bits.sgtc); - - vid = pc.bits.vid; - fid = pc.bits.fid; - - powernow_table[i].frequency = fsb * fid_codes[fid] / 10; - powernow_table[i].index = fid; /* lower 8 bits */ - powernow_table[i].index |= (vid << 8); /* upper 8 bits */ - - speed = powernow_table[i].frequency; - speed_mhz = speed / 1000; - - /* processor_perflib will multiply the MHz value by 1000 to - * get a KHz value (e.g. 1266000). However, powernow-k7 works - * with true KHz values (e.g. 1266768). To ensure that all - * powernow frequencies are available, we must ensure that - * ACPI doesn't restrict them, so we round up the MHz value - * to ensure that perflib's computed KHz value is greater than - * or equal to powernow's KHz value. - */ - if (speed % 1000 > 0) - speed_mhz++; - - if ((fid_codes[fid] % 10) == 5) { - if (have_a0 == 1) - invalidate_entry(i); - } - - dprintk(" FID: 0x%x (%d.%dx [%dMHz]) " - "VID: 0x%x (%d.%03dV)\n", fid, fid_codes[fid] / 10, - fid_codes[fid] % 10, speed_mhz, vid, - mobile_vid_table[vid]/1000, - mobile_vid_table[vid]%1000); - - if (state->core_frequency != speed_mhz) { - state->core_frequency = speed_mhz; - dprintk(" Corrected ACPI frequency to %d\n", - speed_mhz); - } - - if (latency < pc.bits.sgtc) - latency = pc.bits.sgtc; - - if (speed < minimum_speed) - minimum_speed = speed; - if (speed > maximum_speed) - maximum_speed = speed; - } - - powernow_table[i].frequency = CPUFREQ_TABLE_END; - powernow_table[i].index = 0; - - /* notify BIOS that we exist */ - acpi_processor_notify_smm(THIS_MODULE); - - return 0; - -err2: - acpi_processor_unregister_performance(acpi_processor_perf, 0); -err1: - free_cpumask_var(acpi_processor_perf->shared_cpu_map); -err05: - kfree(acpi_processor_perf); -err0: - printk(KERN_WARNING PFX "ACPI perflib can not be used on " - "this platform\n"); - acpi_processor_perf = NULL; - return retval; -} -#else -static int powernow_acpi_init(void) -{ - printk(KERN_INFO PFX "no support for ACPI processor found." - " Please recompile your kernel with ACPI processor\n"); - return -EINVAL; -} -#endif - -static void print_pst_entry(struct pst_s *pst, unsigned int j) -{ - dprintk("PST:%d (@%p)\n", j, pst); - dprintk(" cpuid: 0x%x fsb: %d maxFID: 0x%x startvid: 0x%x\n", - pst->cpuid, pst->fsbspeed, pst->maxfid, pst->startvid); -} - -static int powernow_decode_bios(int maxfid, int startvid) -{ - struct psb_s *psb; - struct pst_s *pst; - unsigned int i, j; - unsigned char *p; - unsigned int etuple; - unsigned int ret; - - etuple = cpuid_eax(0x80000001); - - for (i = 0xC0000; i < 0xffff0 ; i += 16) { - - p = phys_to_virt(i); - - if (memcmp(p, "AMDK7PNOW!", 10) == 0) { - dprintk("Found PSB header at %p\n", p); - psb = (struct psb_s *) p; - dprintk("Table version: 0x%x\n", psb->tableversion); - if (psb->tableversion != 0x12) { - printk(KERN_INFO PFX "Sorry, only v1.2 tables" - " supported right now\n"); - return -ENODEV; - } - - dprintk("Flags: 0x%x\n", psb->flags); - if ((psb->flags & 1) == 0) - dprintk("Mobile voltage regulator\n"); - else - dprintk("Desktop voltage regulator\n"); - - latency = psb->settlingtime; - if (latency < 100) { - printk(KERN_INFO PFX "BIOS set settling time " - "to %d microseconds. " - "Should be at least 100. " - "Correcting.\n", latency); - latency = 100; - } - dprintk("Settling Time: %d microseconds.\n", - psb->settlingtime); - dprintk("Has %d PST tables. (Only dumping ones " - "relevant to this CPU).\n", - psb->numpst); - - p += sizeof(struct psb_s); - - pst = (struct pst_s *) p; - - for (j = 0; j < psb->numpst; j++) { - pst = (struct pst_s *) p; - number_scales = pst->numpstates; - - if ((etuple == pst->cpuid) && - check_fsb(pst->fsbspeed) && - (maxfid == pst->maxfid) && - (startvid == pst->startvid)) { - print_pst_entry(pst, j); - p = (char *)pst + sizeof(struct pst_s); - ret = get_ranges(p); - return ret; - } else { - unsigned int k; - p = (char *)pst + sizeof(struct pst_s); - for (k = 0; k < number_scales; k++) - p += 2; - } - } - printk(KERN_INFO PFX "No PST tables match this cpuid " - "(0x%x)\n", etuple); - printk(KERN_INFO PFX "This is indicative of a broken " - "BIOS.\n"); - - return -EINVAL; - } - p++; - } - - return -ENODEV; -} - - -static int powernow_target(struct cpufreq_policy *policy, - unsigned int target_freq, - unsigned int relation) -{ - unsigned int newstate; - - if (cpufreq_frequency_table_target(policy, powernow_table, target_freq, - relation, &newstate)) - return -EINVAL; - - change_speed(newstate); - - return 0; -} - - -static int powernow_verify(struct cpufreq_policy *policy) -{ - return cpufreq_frequency_table_verify(policy, powernow_table); -} - -/* - * We use the fact that the bus frequency is somehow - * a multiple of 100000/3 khz, then we compute sgtc according - * to this multiple. - * That way, we match more how AMD thinks all of that work. - * We will then get the same kind of behaviour already tested under - * the "well-known" other OS. - */ -static int __cpuinit fixup_sgtc(void) -{ - unsigned int sgtc; - unsigned int m; - - m = fsb / 3333; - if ((m % 10) >= 5) - m += 5; - - m /= 10; - - sgtc = 100 * m * latency; - sgtc = sgtc / 3; - if (sgtc > 0xfffff) { - printk(KERN_WARNING PFX "SGTC too large %d\n", sgtc); - sgtc = 0xfffff; - } - return sgtc; -} - -static unsigned int powernow_get(unsigned int cpu) -{ - union msr_fidvidstatus fidvidstatus; - unsigned int cfid; - - if (cpu) - return 0; - rdmsrl(MSR_K7_FID_VID_STATUS, fidvidstatus.val); - cfid = fidvidstatus.bits.CFID; - - return fsb * fid_codes[cfid] / 10; -} - - -static int __cpuinit acer_cpufreq_pst(const struct dmi_system_id *d) -{ - printk(KERN_WARNING PFX - "%s laptop with broken PST tables in BIOS detected.\n", - d->ident); - printk(KERN_WARNING PFX - "You need to downgrade to 3A21 (09/09/2002), or try a newer " - "BIOS than 3A71 (01/20/2003)\n"); - printk(KERN_WARNING PFX - "cpufreq scaling has been disabled as a result of this.\n"); - return 0; -} - -/* - * Some Athlon laptops have really fucked PST tables. - * A BIOS update is all that can save them. - * Mention this, and disable cpufreq. - */ -static struct dmi_system_id __cpuinitdata powernow_dmi_table[] = { - { - .callback = acer_cpufreq_pst, - .ident = "Acer Aspire", - .matches = { - DMI_MATCH(DMI_SYS_VENDOR, "Insyde Software"), - DMI_MATCH(DMI_BIOS_VERSION, "3A71"), - }, - }, - { } -}; - -static int __cpuinit powernow_cpu_init(struct cpufreq_policy *policy) -{ - union msr_fidvidstatus fidvidstatus; - int result; - - if (policy->cpu != 0) - return -ENODEV; - - rdmsrl(MSR_K7_FID_VID_STATUS, fidvidstatus.val); - - recalibrate_cpu_khz(); - - fsb = (10 * cpu_khz) / fid_codes[fidvidstatus.bits.CFID]; - if (!fsb) { - printk(KERN_WARNING PFX "can not determine bus frequency\n"); - return -EINVAL; - } - dprintk("FSB: %3dMHz\n", fsb/1000); - - if (dmi_check_system(powernow_dmi_table) || acpi_force) { - printk(KERN_INFO PFX "PSB/PST known to be broken. " - "Trying ACPI instead\n"); - result = powernow_acpi_init(); - } else { - result = powernow_decode_bios(fidvidstatus.bits.MFID, - fidvidstatus.bits.SVID); - if (result) { - printk(KERN_INFO PFX "Trying ACPI perflib\n"); - maximum_speed = 0; - minimum_speed = -1; - latency = 0; - result = powernow_acpi_init(); - if (result) { - printk(KERN_INFO PFX - "ACPI and legacy methods failed\n"); - } - } else { - /* SGTC use the bus clock as timer */ - latency = fixup_sgtc(); - printk(KERN_INFO PFX "SGTC: %d\n", latency); - } - } - - if (result) - return result; - - printk(KERN_INFO PFX "Minimum speed %d MHz. Maximum speed %d MHz.\n", - minimum_speed/1000, maximum_speed/1000); - - policy->cpuinfo.transition_latency = - cpufreq_scale(2000000UL, fsb, latency); - - policy->cur = powernow_get(0); - - cpufreq_frequency_table_get_attr(powernow_table, policy->cpu); - - return cpufreq_frequency_table_cpuinfo(policy, powernow_table); -} - -static int powernow_cpu_exit(struct cpufreq_policy *policy) -{ - cpufreq_frequency_table_put_attr(policy->cpu); - -#ifdef CONFIG_X86_POWERNOW_K7_ACPI - if (acpi_processor_perf) { - acpi_processor_unregister_performance(acpi_processor_perf, 0); - free_cpumask_var(acpi_processor_perf->shared_cpu_map); - kfree(acpi_processor_perf); - } -#endif - - kfree(powernow_table); - return 0; -} - -static struct freq_attr *powernow_table_attr[] = { - &cpufreq_freq_attr_scaling_available_freqs, - NULL, -}; - -static struct cpufreq_driver powernow_driver = { - .verify = powernow_verify, - .target = powernow_target, - .get = powernow_get, -#ifdef CONFIG_X86_POWERNOW_K7_ACPI - .bios_limit = acpi_processor_get_bios_limit, -#endif - .init = powernow_cpu_init, - .exit = powernow_cpu_exit, - .name = "powernow-k7", - .owner = THIS_MODULE, - .attr = powernow_table_attr, -}; - -static int __init powernow_init(void) -{ - if (check_powernow() == 0) - return -ENODEV; - return cpufreq_register_driver(&powernow_driver); -} - - -static void __exit powernow_exit(void) -{ - cpufreq_unregister_driver(&powernow_driver); -} - -module_param(acpi_force, int, 0444); -MODULE_PARM_DESC(acpi_force, "Force ACPI to be used."); - -MODULE_AUTHOR("Dave Jones <davej@redhat.com>"); -MODULE_DESCRIPTION("Powernow driver for AMD K7 processors."); -MODULE_LICENSE("GPL"); - -late_initcall(powernow_init); -module_exit(powernow_exit); - diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k7.h b/arch/x86/kernel/cpu/cpufreq/powernow-k7.h deleted file mode 100644 index 35fb4eaf6e1c..000000000000 --- a/arch/x86/kernel/cpu/cpufreq/powernow-k7.h +++ /dev/null @@ -1,43 +0,0 @@ -/* - * (C) 2003 Dave Jones. - * - * Licensed under the terms of the GNU GPL License version 2. - * - * AMD-specific information - * - */ - -union msr_fidvidctl { - struct { - unsigned FID:5, // 4:0 - reserved1:3, // 7:5 - VID:5, // 12:8 - reserved2:3, // 15:13 - FIDC:1, // 16 - VIDC:1, // 17 - reserved3:2, // 19:18 - FIDCHGRATIO:1, // 20 - reserved4:11, // 31-21 - SGTC:20, // 32:51 - reserved5:12; // 63:52 - } bits; - unsigned long long val; -}; - -union msr_fidvidstatus { - struct { - unsigned CFID:5, // 4:0 - reserved1:3, // 7:5 - SFID:5, // 12:8 - reserved2:3, // 15:13 - MFID:5, // 20:16 - reserved3:11, // 31:21 - CVID:5, // 36:32 - reserved4:3, // 39:37 - SVID:5, // 44:40 - reserved5:3, // 47:45 - MVID:5, // 52:48 - reserved6:11; // 63:53 - } bits; - unsigned long long val; -}; diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c deleted file mode 100644 index 2368e38327b3..000000000000 --- a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c +++ /dev/null @@ -1,1607 +0,0 @@ -/* - * (c) 2003-2010 Advanced Micro Devices, Inc. - * Your use of this code is subject to the terms and conditions of the - * GNU general public license version 2. See "COPYING" or - * http://www.gnu.org/licenses/gpl.html - * - * Support : mark.langsdorf@amd.com - * - * Based on the powernow-k7.c module written by Dave Jones. - * (C) 2003 Dave Jones on behalf of SuSE Labs - * (C) 2004 Dominik Brodowski <linux@brodo.de> - * (C) 2004 Pavel Machek <pavel@ucw.cz> - * Licensed under the terms of the GNU GPL License version 2. - * Based upon datasheets & sample CPUs kindly provided by AMD. - * - * Valuable input gratefully received from Dave Jones, Pavel Machek, - * Dominik Brodowski, Jacob Shin, and others. - * Originally developed by Paul Devriendt. - * Processor information obtained from Chapter 9 (Power and Thermal Management) - * of the "BIOS and Kernel Developer's Guide for the AMD Athlon 64 and AMD - * Opteron Processors" available for download from www.amd.com - * - * Tables for specific CPUs can be inferred from - * http://www.amd.com/us-en/assets/content_type/white_papers_and_tech_docs/30430.pdf - */ - -#include <linux/kernel.h> -#include <linux/smp.h> -#include <linux/module.h> -#include <linux/init.h> -#include <linux/cpufreq.h> -#include <linux/slab.h> -#include <linux/string.h> -#include <linux/cpumask.h> -#include <linux/sched.h> /* for current / set_cpus_allowed() */ -#include <linux/io.h> -#include <linux/delay.h> - -#include <asm/msr.h> - -#include <linux/acpi.h> -#include <linux/mutex.h> -#include <acpi/processor.h> - -#define PFX "powernow-k8: " -#define VERSION "version 2.20.00" -#include "powernow-k8.h" -#include "mperf.h" - -/* serialize freq changes */ -static DEFINE_MUTEX(fidvid_mutex); - -static DEFINE_PER_CPU(struct powernow_k8_data *, powernow_data); - -static int cpu_family = CPU_OPTERON; - -/* core performance boost */ -static bool cpb_capable, cpb_enabled; -static struct msr __percpu *msrs; - -static struct cpufreq_driver cpufreq_amd64_driver; - -#ifndef CONFIG_SMP -static inline const struct cpumask *cpu_core_mask(int cpu) -{ - return cpumask_of(0); -} -#endif - -/* Return a frequency in MHz, given an input fid */ -static u32 find_freq_from_fid(u32 fid) -{ - return 800 + (fid * 100); -} - -/* Return a frequency in KHz, given an input fid */ -static u32 find_khz_freq_from_fid(u32 fid) -{ - return 1000 * find_freq_from_fid(fid); -} - -static u32 find_khz_freq_from_pstate(struct cpufreq_frequency_table *data, - u32 pstate) -{ - return data[pstate].frequency; -} - -/* Return the vco fid for an input fid - * - * Each "low" fid has corresponding "high" fid, and you can get to "low" fids - * only from corresponding high fids. This returns "high" fid corresponding to - * "low" one. - */ -static u32 convert_fid_to_vco_fid(u32 fid) -{ - if (fid < HI_FID_TABLE_BOTTOM) - return 8 + (2 * fid); - else - return fid; -} - -/* - * Return 1 if the pending bit is set. Unless we just instructed the processor - * to transition to a new state, seeing this bit set is really bad news. - */ -static int pending_bit_stuck(void) -{ - u32 lo, hi; - - if (cpu_family == CPU_HW_PSTATE) - return 0; - - rdmsr(MSR_FIDVID_STATUS, lo, hi); - return lo & MSR_S_LO_CHANGE_PENDING ? 1 : 0; -} - -/* - * Update the global current fid / vid values from the status msr. - * Returns 1 on error. - */ -static int query_current_values_with_pending_wait(struct powernow_k8_data *data) -{ - u32 lo, hi; - u32 i = 0; - - if (cpu_family == CPU_HW_PSTATE) { - 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 { - if (i++ > 10000) { - dprintk("detected change pending stuck\n"); - return 1; - } - rdmsr(MSR_FIDVID_STATUS, lo, hi); - } while (lo & MSR_S_LO_CHANGE_PENDING); - - data->currvid = hi & MSR_S_HI_CURRENT_VID; - data->currfid = lo & MSR_S_LO_CURRENT_FID; - - return 0; -} - -/* the isochronous relief time */ -static void count_off_irt(struct powernow_k8_data *data) -{ - udelay((1 << data->irt) * 10); - return; -} - -/* the voltage stabilization time */ -static void count_off_vst(struct powernow_k8_data *data) -{ - udelay(data->vstable * VST_UNITS_20US); - return; -} - -/* need to init the control msr to a safe value (for each cpu) */ -static void fidvid_msr_init(void) -{ - u32 lo, hi; - u8 fid, vid; - - rdmsr(MSR_FIDVID_STATUS, lo, hi); - vid = hi & MSR_S_HI_CURRENT_VID; - fid = lo & MSR_S_LO_CURRENT_FID; - lo = fid | (vid << MSR_C_LO_VID_SHIFT); - hi = MSR_C_HI_STP_GNT_BENIGN; - dprintk("cpu%d, init lo 0x%x, hi 0x%x\n", smp_processor_id(), lo, hi); - wrmsr(MSR_FIDVID_CTL, lo, hi); -} - -/* write the new fid value along with the other control fields to the msr */ -static int write_new_fid(struct powernow_k8_data *data, u32 fid) -{ - u32 lo; - u32 savevid = data->currvid; - u32 i = 0; - - if ((fid & INVALID_FID_MASK) || (data->currvid & INVALID_VID_MASK)) { - printk(KERN_ERR PFX "internal error - overflow on fid write\n"); - return 1; - } - - lo = fid; - lo |= (data->currvid << MSR_C_LO_VID_SHIFT); - lo |= MSR_C_LO_INIT_FID_VID; - - dprintk("writing fid 0x%x, lo 0x%x, hi 0x%x\n", - fid, lo, data->plllock * PLL_LOCK_CONVERSION); - - do { - wrmsr(MSR_FIDVID_CTL, lo, data->plllock * PLL_LOCK_CONVERSION); - if (i++ > 100) { - printk(KERN_ERR PFX - "Hardware error - pending bit very stuck - " - "no further pstate changes possible\n"); - return 1; - } - } while (query_current_values_with_pending_wait(data)); - - count_off_irt(data); - - if (savevid != data->currvid) { - printk(KERN_ERR PFX - "vid change on fid trans, old 0x%x, new 0x%x\n", - savevid, data->currvid); - return 1; - } - - if (fid != data->currfid) { - printk(KERN_ERR PFX - "fid trans failed, fid 0x%x, curr 0x%x\n", fid, - data->currfid); - return 1; - } - - return 0; -} - -/* Write a new vid to the hardware */ -static int write_new_vid(struct powernow_k8_data *data, u32 vid) -{ - u32 lo; - u32 savefid = data->currfid; - int i = 0; - - if ((data->currfid & INVALID_FID_MASK) || (vid & INVALID_VID_MASK)) { - printk(KERN_ERR PFX "internal error - overflow on vid write\n"); - return 1; - } - - lo = data->currfid; - lo |= (vid << MSR_C_LO_VID_SHIFT); - lo |= MSR_C_LO_INIT_FID_VID; - - dprintk("writing vid 0x%x, lo 0x%x, hi 0x%x\n", - vid, lo, STOP_GRANT_5NS); - - do { - wrmsr(MSR_FIDVID_CTL, lo, STOP_GRANT_5NS); - if (i++ > 100) { - printk(KERN_ERR PFX "internal error - pending bit " - "very stuck - no further pstate " - "changes possible\n"); - return 1; - } - } while (query_current_values_with_pending_wait(data)); - - if (savefid != data->currfid) { - printk(KERN_ERR PFX "fid changed on vid trans, old " - "0x%x new 0x%x\n", - savefid, data->currfid); - return 1; - } - - if (vid != data->currvid) { - printk(KERN_ERR PFX "vid trans failed, vid 0x%x, " - "curr 0x%x\n", - vid, data->currvid); - return 1; - } - - return 0; -} - -/* - * Reduce the vid by the max of step or reqvid. - * Decreasing vid codes represent increasing voltages: - * vid of 0 is 1.550V, vid of 0x1e is 0.800V, vid of VID_OFF is off. - */ -static int decrease_vid_code_by_step(struct powernow_k8_data *data, - u32 reqvid, u32 step) -{ - if ((data->currvid - reqvid) > step) - reqvid = data->currvid - step; - - if (write_new_vid(data, reqvid)) - return 1; - - count_off_vst(data); - - return 0; -} - -/* Change hardware pstate by single MSR write */ -static int transition_pstate(struct powernow_k8_data *data, u32 pstate) -{ - wrmsr(MSR_PSTATE_CTRL, pstate, 0); - data->currpstate = pstate; - return 0; -} - -/* Change Opteron/Athlon64 fid and vid, by the 3 phases. */ -static int transition_fid_vid(struct powernow_k8_data *data, - u32 reqfid, u32 reqvid) -{ - if (core_voltage_pre_transition(data, reqvid, reqfid)) - return 1; - - if (core_frequency_transition(data, reqfid)) - return 1; - - if (core_voltage_post_transition(data, reqvid)) - return 1; - - if (query_current_values_with_pending_wait(data)) - return 1; - - if ((reqfid != data->currfid) || (reqvid != data->currvid)) { - printk(KERN_ERR PFX "failed (cpu%d): req 0x%x 0x%x, " - "curr 0x%x 0x%x\n", - smp_processor_id(), - reqfid, reqvid, data->currfid, data->currvid); - return 1; - } - - dprintk("transitioned (cpu%d): new fid 0x%x, vid 0x%x\n", - smp_processor_id(), data->currfid, data->currvid); - - return 0; -} - -/* Phase 1 - core voltage transition ... setup voltage */ -static int core_voltage_pre_transition(struct powernow_k8_data *data, - u32 reqvid, u32 reqfid) -{ - u32 rvosteps = data->rvo; - u32 savefid = data->currfid; - 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); - if (reqvid < maxvid) /* lower numbers are higher voltages */ - reqvid = maxvid; - - while (data->currvid > reqvid) { - dprintk("ph1: curr 0x%x, req vid 0x%x\n", - data->currvid, reqvid); - if (decrease_vid_code_by_step(data, reqvid, data->vidmvs)) - return 1; - } - - while ((rvosteps > 0) && - ((rvomult * data->rvo + data->currvid) > reqvid)) { - if (data->currvid == maxvid) { - rvosteps = 0; - } else { - dprintk("ph1: changing vid for rvo, req 0x%x\n", - data->currvid - 1); - if (decrease_vid_code_by_step(data, data->currvid-1, 1)) - return 1; - rvosteps--; - } - } - - if (query_current_values_with_pending_wait(data)) - return 1; - - if (savefid != data->currfid) { - printk(KERN_ERR PFX "ph1 err, currfid changed 0x%x\n", - data->currfid); - return 1; - } - - dprintk("ph1 complete, currfid 0x%x, currvid 0x%x\n", - data->currfid, data->currvid); - - return 0; -} - -/* Phase 2 - core frequency transition */ -static int core_frequency_transition(struct powernow_k8_data *data, u32 reqfid) -{ - u32 vcoreqfid, vcocurrfid, vcofiddiff; - u32 fid_interval, savevid = data->currvid; - - if (data->currfid == reqfid) { - printk(KERN_ERR PFX "ph2 null fid transition 0x%x\n", - data->currfid); - return 0; - } - - dprintk("ph2 (cpu%d): starting, currfid 0x%x, currvid 0x%x, " - "reqfid 0x%x\n", - smp_processor_id(), - data->currfid, data->currvid, reqfid); - - vcoreqfid = convert_fid_to_vco_fid(reqfid); - vcocurrfid = convert_fid_to_vco_fid(data->currfid); - 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); - - if (reqfid > data->currfid) { - if (data->currfid > LO_FID_TABLE_TOP) { - if (write_new_fid(data, - data->currfid + fid_interval)) - return 1; - } else { - if (write_new_fid - (data, - 2 + convert_fid_to_vco_fid(data->currfid))) - return 1; - } - } else { - if (write_new_fid(data, data->currfid - fid_interval)) - return 1; - } - - vcocurrfid = convert_fid_to_vco_fid(data->currfid); - vcofiddiff = vcocurrfid > vcoreqfid ? vcocurrfid - vcoreqfid - : vcoreqfid - vcocurrfid; - } - - if (write_new_fid(data, reqfid)) - return 1; - - if (query_current_values_with_pending_wait(data)) - return 1; - - if (data->currfid != reqfid) { - printk(KERN_ERR PFX - "ph2: mismatch, failed fid transition, " - "curr 0x%x, req 0x%x\n", - data->currfid, reqfid); - return 1; - } - - if (savevid != data->currvid) { - printk(KERN_ERR PFX "ph2: vid changed, save 0x%x, curr 0x%x\n", - savevid, data->currvid); - return 1; - } - - dprintk("ph2 complete, currfid 0x%x, currvid 0x%x\n", - data->currfid, data->currvid); - - return 0; -} - -/* Phase 3 - core voltage transition flow ... jump to the final vid. */ -static int core_voltage_post_transition(struct powernow_k8_data *data, - u32 reqvid) -{ - u32 savefid = data->currfid; - u32 savereqvid = reqvid; - - dprintk("ph3 (cpu%d): starting, currfid 0x%x, currvid 0x%x\n", - smp_processor_id(), - data->currfid, data->currvid); - - if (reqvid != data->currvid) { - if (write_new_vid(data, reqvid)) - return 1; - - if (savefid != data->currfid) { - printk(KERN_ERR PFX - "ph3: bad fid change, save 0x%x, curr 0x%x\n", - savefid, data->currfid); - return 1; - } - - if (data->currvid != reqvid) { - printk(KERN_ERR PFX - "ph3: failed vid transition\n, " - "req 0x%x, curr 0x%x", - reqvid, data->currvid); - return 1; - } - } - - if (query_current_values_with_pending_wait(data)) - return 1; - - if (savereqvid != data->currvid) { - dprintk("ph3 failed, currvid 0x%x\n", data->currvid); - return 1; - } - - if (savefid != data->currfid) { - dprintk("ph3 failed, currfid changed 0x%x\n", - data->currfid); - return 1; - } - - dprintk("ph3 complete, currfid 0x%x, currvid 0x%x\n", - data->currfid, data->currvid); - - return 0; -} - -static void check_supported_cpu(void *_rc) -{ - u32 eax, ebx, ecx, edx; - int *rc = _rc; - - *rc = -ENODEV; - - if (__this_cpu_read(cpu_info.x86_vendor) != X86_VENDOR_AMD) - return; - - eax = cpuid_eax(CPUID_PROCESSOR_SIGNATURE); - if (((eax & CPUID_XFAM) != CPUID_XFAM_K8) && - ((eax & CPUID_XFAM) < CPUID_XFAM_10H)) - 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); - return; - } - - eax = cpuid_eax(CPUID_GET_MAX_CAPABILITIES); - if (eax < CPUID_FREQ_VOLT_CAPABILITIES) { - printk(KERN_INFO PFX - "No frequency change capabilities detected\n"); - return; - } - - cpuid(CPUID_FREQ_VOLT_CAPABILITIES, &eax, &ebx, &ecx, &edx); - if ((edx & P_STATE_TRANSITION_CAPABLE) - != P_STATE_TRANSITION_CAPABLE) { - printk(KERN_INFO PFX - "Power state transitions not supported\n"); - 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 - return; - } - - *rc = 0; -} - -static int check_pst_table(struct powernow_k8_data *data, struct pst_s *pst, - u8 maxvid) -{ - unsigned int j; - u8 lastfid = 0xff; - - for (j = 0; j < data->numps; j++) { - if (pst[j].vid > LEAST_VID) { - printk(KERN_ERR FW_BUG PFX "vid %d invalid : 0x%x\n", - j, pst[j].vid); - return -EINVAL; - } - if (pst[j].vid < data->rvo) { - /* vid + rvo >= 0 */ - printk(KERN_ERR FW_BUG PFX "0 vid exceeded with pstate" - " %d\n", j); - return -ENODEV; - } - if (pst[j].vid < maxvid + data->rvo) { - /* vid + rvo >= maxvid */ - printk(KERN_ERR FW_BUG PFX "maxvid exceeded with pstate" - " %d\n", j); - return -ENODEV; - } - if (pst[j].fid > MAX_FID) { - printk(KERN_ERR FW_BUG PFX "maxfid exceeded with pstate" - " %d\n", j); - return -ENODEV; - } - if (j && (pst[j].fid < HI_FID_TABLE_BOTTOM)) { - /* Only first fid is allowed to be in "low" range */ - printk(KERN_ERR FW_BUG PFX "two low fids - %d : " - "0x%x\n", j, pst[j].fid); - return -EINVAL; - } - if (pst[j].fid < lastfid) - lastfid = pst[j].fid; - } - if (lastfid & 1) { - printk(KERN_ERR FW_BUG PFX "lastfid invalid\n"); - return -EINVAL; - } - if (lastfid > LO_FID_TABLE_TOP) - printk(KERN_INFO FW_BUG PFX - "first fid not from lo freq table\n"); - - return 0; -} - -static void invalidate_entry(struct cpufreq_frequency_table *powernow_table, - unsigned int entry) -{ - powernow_table[entry].frequency = CPUFREQ_ENTRY_INVALID; -} - -static void print_basics(struct powernow_k8_data *data) -{ - int j; - for (j = 0; j < data->numps; j++) { - if (data->powernow_table[j].frequency != - CPUFREQ_ENTRY_INVALID) { - if (cpu_family == CPU_HW_PSTATE) { - printk(KERN_INFO PFX - " %d : pstate %d (%d MHz)\n", j, - data->powernow_table[j].index, - data->powernow_table[j].frequency/1000); - } else { - printk(KERN_INFO PFX - "fid 0x%x (%d MHz), vid 0x%x\n", - data->powernow_table[j].index & 0xff, - data->powernow_table[j].frequency/1000, - data->powernow_table[j].index >> 8); - } - } - } - if (data->batps) - printk(KERN_INFO PFX "Only %d pstates on battery\n", - data->batps); -} - -static u32 freq_from_fid_did(u32 fid, u32 did) -{ - u32 mhz = 0; - - if (boot_cpu_data.x86 == 0x10) - mhz = (100 * (fid + 0x10)) >> did; - else if (boot_cpu_data.x86 == 0x11) - mhz = (100 * (fid + 8)) >> did; - else - BUG(); - - return mhz * 1000; -} - -static int fill_powernow_table(struct powernow_k8_data *data, - struct pst_s *pst, u8 maxvid) -{ - struct cpufreq_frequency_table *powernow_table; - unsigned int j; - - if (data->batps) { - /* use ACPI support to get full speed on mains power */ - printk(KERN_WARNING PFX - "Only %d pstates usable (use ACPI driver for full " - "range\n", data->batps); - data->numps = data->batps; - } - - for (j = 1; j < data->numps; j++) { - if (pst[j-1].fid >= pst[j].fid) { - printk(KERN_ERR PFX "PST out of sequence\n"); - return -EINVAL; - } - } - - if (data->numps < 2) { - printk(KERN_ERR PFX "no p states to transition\n"); - return -ENODEV; - } - - if (check_pst_table(data, pst, maxvid)) - return -EINVAL; - - powernow_table = kmalloc((sizeof(struct cpufreq_frequency_table) - * (data->numps + 1)), GFP_KERNEL); - if (!powernow_table) { - printk(KERN_ERR PFX "powernow_table memory alloc failure\n"); - return -ENOMEM; - } - - for (j = 0; j < data->numps; j++) { - int freq; - powernow_table[j].index = pst[j].fid; /* lower 8 bits */ - powernow_table[j].index |= (pst[j].vid << 8); /* upper 8 bits */ - freq = find_khz_freq_from_fid(pst[j].fid); - powernow_table[j].frequency = freq; - } - powernow_table[data->numps].frequency = CPUFREQ_TABLE_END; - powernow_table[data->numps].index = 0; - - if (query_current_values_with_pending_wait(data)) { - kfree(powernow_table); - return -EIO; - } - - dprintk("cfid 0x%x, cvid 0x%x\n", data->currfid, data->currvid); - data->powernow_table = powernow_table; - if (cpumask_first(cpu_core_mask(data->cpu)) == data->cpu) - print_basics(data); - - for (j = 0; j < data->numps; j++) - if ((pst[j].fid == data->currfid) && - (pst[j].vid == data->currvid)) - return 0; - - dprintk("currfid/vid do not match PST, ignoring\n"); - return 0; -} - -/* Find and validate the PSB/PST table in BIOS. */ -static int find_psb_table(struct powernow_k8_data *data) -{ - struct psb_s *psb; - unsigned int i; - u32 mvs; - u8 maxvid; - u32 cpst = 0; - u32 thiscpuid; - - for (i = 0xc0000; i < 0xffff0; i += 0x10) { - /* Scan BIOS looking for the signature. */ - /* It can not be at ffff0 - it is too big. */ - - psb = phys_to_virt(i); - if (memcmp(psb, PSB_ID_STRING, PSB_ID_STRING_LEN) != 0) - continue; - - dprintk("found PSB header at 0x%p\n", psb); - - dprintk("table vers: 0x%x\n", psb->tableversion); - if (psb->tableversion != PSB_VERSION_1_4) { - printk(KERN_ERR FW_BUG PFX "PSB table is not v1.4\n"); - return -ENODEV; - } - - dprintk("flags: 0x%x\n", psb->flags1); - if (psb->flags1) { - printk(KERN_ERR FW_BUG PFX "unknown flags\n"); - return -ENODEV; - } - - data->vstable = psb->vstable; - dprintk("voltage stabilization time: %d(*20us)\n", - data->vstable); - - dprintk("flags2: 0x%x\n", psb->flags2); - data->rvo = psb->flags2 & 3; - data->irt = ((psb->flags2) >> 2) & 3; - mvs = ((psb->flags2) >> 4) & 3; - data->vidmvs = 1 << mvs; - data->batps = ((psb->flags2) >> 6) & 3; - - dprintk("ramp voltage offset: %d\n", data->rvo); - dprintk("isochronous relief time: %d\n", data->irt); - dprintk("maximum voltage step: %d - 0x%x\n", mvs, data->vidmvs); - - dprintk("numpst: 0x%x\n", psb->num_tables); - cpst = psb->num_tables; - if ((psb->cpuid == 0x00000fc0) || - (psb->cpuid == 0x00000fe0)) { - thiscpuid = cpuid_eax(CPUID_PROCESSOR_SIGNATURE); - if ((thiscpuid == 0x00000fc0) || - (thiscpuid == 0x00000fe0)) - cpst = 1; - } - if (cpst != 1) { - printk(KERN_ERR FW_BUG PFX "numpst must be 1\n"); - return -ENODEV; - } - - data->plllock = psb->plllocktime; - dprintk("plllocktime: 0x%x (units 1us)\n", psb->plllocktime); - dprintk("maxfid: 0x%x\n", psb->maxfid); - dprintk("maxvid: 0x%x\n", psb->maxvid); - maxvid = psb->maxvid; - - data->numps = psb->numps; - dprintk("numpstates: 0x%x\n", data->numps); - return fill_powernow_table(data, - (struct pst_s *)(psb+1), maxvid); - } - /* - * If you see this message, complain to BIOS manufacturer. If - * he tells you "we do not support Linux" or some similar - * nonsense, remember that Windows 2000 uses the same legacy - * mechanism that the old Linux PSB driver uses. Tell them it - * is broken with Windows 2000. - * - * The reference to the AMD documentation is chapter 9 in the - * BIOS and Kernel Developer's Guide, which is available on - * www.amd.com - */ - printk(KERN_ERR FW_BUG PFX "No PSB or ACPI _PSS objects\n"); - printk(KERN_ERR PFX "Make sure that your BIOS is up to date" - " and Cool'N'Quiet support is enabled in BIOS setup\n"); - return -ENODEV; -} - -static void powernow_k8_acpi_pst_values(struct powernow_k8_data *data, - unsigned int index) -{ - u64 control; - - 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; -} - -static int powernow_k8_cpu_init_acpi(struct powernow_k8_data *data) -{ - struct cpufreq_frequency_table *powernow_table; - int ret_val = -ENODEV; - u64 control, status; - - if (acpi_processor_register_performance(&data->acpi_data, data->cpu)) { - dprintk("register performance failed: bad ACPI data\n"); - return -EIO; - } - - /* verify the data contained in the ACPI structures */ - if (data->acpi_data.state_count <= 1) { - dprintk("No ACPI P-States\n"); - goto err_out; - } - - control = data->acpi_data.control_register.space_id; - status = data->acpi_data.status_register.space_id; - - if ((control != ACPI_ADR_SPACE_FIXED_HARDWARE) || - (status != ACPI_ADR_SPACE_FIXED_HARDWARE)) { - dprintk("Invalid control/status registers (%x - %x)\n", - control, status); - goto err_out; - } - - /* fill in data->powernow_table */ - powernow_table = kmalloc((sizeof(struct cpufreq_frequency_table) - * (data->acpi_data.state_count + 1)), GFP_KERNEL); - if (!powernow_table) { - dprintk("powernow_table memory alloc failure\n"); - goto err_out; - } - - /* fill in data */ - data->numps = data->acpi_data.state_count; - powernow_k8_acpi_pst_values(data, 0); - - if (cpu_family == CPU_HW_PSTATE) - ret_val = fill_powernow_table_pstate(data, powernow_table); - else - ret_val = fill_powernow_table_fidvid(data, powernow_table); - if (ret_val) - goto err_out_mem; - - powernow_table[data->acpi_data.state_count].frequency = - CPUFREQ_TABLE_END; - powernow_table[data->acpi_data.state_count].index = 0; - data->powernow_table = powernow_table; - - if (cpumask_first(cpu_core_mask(data->cpu)) == data->cpu) - print_basics(data); - - /* notify BIOS that we exist */ - acpi_processor_notify_smm(THIS_MODULE); - - if (!zalloc_cpumask_var(&data->acpi_data.shared_cpu_map, GFP_KERNEL)) { - printk(KERN_ERR PFX - "unable to alloc powernow_k8_data cpumask\n"); - ret_val = -ENOMEM; - goto err_out_mem; - } - - return 0; - -err_out_mem: - kfree(powernow_table); - -err_out: - acpi_processor_unregister_performance(&data->acpi_data, data->cpu); - - /* data->acpi_data.state_count informs us at ->exit() - * whether ACPI was used */ - data->acpi_data.state_count = 0; - - return ret_val; -} - -static int fill_powernow_table_pstate(struct powernow_k8_data *data, - struct cpufreq_frequency_table *powernow_table) -{ - int i; - u32 hi = 0, lo = 0; - rdmsr(MSR_PSTATE_CUR_LIMIT, lo, hi); - data->max_hw_pstate = (lo & HW_PSTATE_MAX_MASK) >> HW_PSTATE_MAX_SHIFT; - - for (i = 0; i < data->acpi_data.state_count; i++) { - u32 index; - - index = data->acpi_data.states[i].control & HW_PSTATE_MASK; - if (index > data->max_hw_pstate) { - printk(KERN_ERR PFX "invalid pstate %d - " - "bad value %d.\n", i, index); - printk(KERN_ERR PFX "Please report to BIOS " - "manufacturer\n"); - invalidate_entry(powernow_table, i); - continue; - } - rdmsr(MSR_PSTATE_DEF_BASE + index, lo, hi); - if (!(hi & HW_PSTATE_VALID_MASK)) { - dprintk("invalid pstate %d, ignoring\n", index); - invalidate_entry(powernow_table, i); - continue; - } - - powernow_table[i].index = index; - - /* Frequency may be rounded for these */ - if ((boot_cpu_data.x86 == 0x10 && boot_cpu_data.x86_model < 10) - || boot_cpu_data.x86 == 0x11) { - powernow_table[i].frequency = - freq_from_fid_did(lo & 0x3f, (lo >> 6) & 7); - } else - powernow_table[i].frequency = - data->acpi_data.states[i].core_frequency * 1000; - } - return 0; -} - -static int fill_powernow_table_fidvid(struct powernow_k8_data *data, - struct cpufreq_frequency_table *powernow_table) -{ - int i; - - for (i = 0; i < data->acpi_data.state_count; i++) { - u32 fid; - u32 vid; - u32 freq, index; - u64 status, control; - - if (data->exttype) { - status = data->acpi_data.states[i].status; - fid = status & EXT_FID_MASK; - vid = (status >> VID_SHIFT) & EXT_VID_MASK; - } else { - control = data->acpi_data.states[i].control; - fid = control & FID_MASK; - vid = (control >> VID_SHIFT) & VID_MASK; - } - - dprintk(" %d : fid 0x%x, vid 0x%x\n", i, fid, vid); - - index = fid | (vid<<8); - powernow_table[i].index = index; - - freq = find_khz_freq_from_fid(fid); - powernow_table[i].frequency = freq; - - /* verify frequency is OK */ - if ((freq > (MAX_FREQ * 1000)) || (freq < (MIN_FREQ * 1000))) { - dprintk("invalid freq %u kHz, ignoring\n", freq); - invalidate_entry(powernow_table, i); - continue; - } - - /* verify voltage is OK - - * BIOSs are using "off" to indicate invalid */ - if (vid == VID_OFF) { - dprintk("invalid vid %u, ignoring\n", vid); - invalidate_entry(powernow_table, i); - continue; - } - - if (freq != (data->acpi_data.states[i].core_frequency * 1000)) { - printk(KERN_INFO PFX "invalid freq entries " - "%u kHz vs. %u kHz\n", freq, - (unsigned int) - (data->acpi_data.states[i].core_frequency - * 1000)); - invalidate_entry(powernow_table, i); - continue; - } - } - return 0; -} - -static void powernow_k8_cpu_exit_acpi(struct powernow_k8_data *data) -{ - if (data->acpi_data.state_count) - acpi_processor_unregister_performance(&data->acpi_data, - data->cpu); - free_cpumask_var(data->acpi_data.shared_cpu_map); -} - -static int get_transition_latency(struct powernow_k8_data *data) -{ - int max_latency = 0; - int i; - for (i = 0; i < data->acpi_data.state_count; i++) { - int cur_latency = data->acpi_data.states[i].transition_latency - + data->acpi_data.states[i].bus_master_latency; - if (cur_latency > max_latency) - max_latency = cur_latency; - } - if (max_latency == 0) { - /* - * Fam 11h and later may return 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. - */ - 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; -} - -/* Take a frequency, and issue the fid/vid transition command */ -static int transition_frequency_fidvid(struct powernow_k8_data *data, - unsigned int index) -{ - u32 fid = 0; - u32 vid = 0; - int res, i; - struct cpufreq_freqs freqs; - - dprintk("cpu %d transition to index %u\n", smp_processor_id(), index); - - /* fid/vid correctness check for k8 */ - /* fid are the lower 8 bits of the index we stored into - * the cpufreq frequency table in find_psb_table, vid - * are the upper 8 bits. - */ - fid = data->powernow_table[index].index & 0xFF; - vid = (data->powernow_table[index].index & 0xFF00) >> 8; - - dprintk("table matched fid 0x%x, giving vid 0x%x\n", fid, vid); - - if (query_current_values_with_pending_wait(data)) - return 1; - - if ((data->currvid == vid) && (data->currfid == fid)) { - dprintk("target matches current values (fid 0x%x, vid 0x%x)\n", - fid, vid); - return 0; - } - - 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(i, data->available_cores) { - freqs.cpu = i; - cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); - } - - res = transition_fid_vid(data, fid, vid); - freqs.new = find_khz_freq_from_fid(data->currfid); - - for_each_cpu(i, data->available_cores) { - freqs.cpu = i; - cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); - } - return res; -} - -/* Take a frequency, and issue the hardware pstate transition command */ -static int transition_frequency_pstate(struct powernow_k8_data *data, - unsigned int index) -{ - u32 pstate = 0; - int res, i; - struct cpufreq_freqs freqs; - - dprintk("cpu %d transition to index %u\n", smp_processor_id(), index); - - /* get MSR index for hardware pstate transition */ - pstate = index & HW_PSTATE_MASK; - if (pstate > data->max_hw_pstate) - return 0; - freqs.old = find_khz_freq_from_pstate(data->powernow_table, - data->currpstate); - freqs.new = find_khz_freq_from_pstate(data->powernow_table, pstate); - - for_each_cpu(i, data->available_cores) { - freqs.cpu = i; - cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); - } - - res = transition_pstate(data, pstate); - freqs.new = find_khz_freq_from_pstate(data->powernow_table, pstate); - - for_each_cpu(i, data->available_cores) { - freqs.cpu = i; - cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); - } - return res; -} - -/* Driver entry point to switch to the target frequency */ -static int powernowk8_target(struct cpufreq_policy *pol, - unsigned targfreq, unsigned relation) -{ - cpumask_var_t oldmask; - struct powernow_k8_data *data = per_cpu(powernow_data, pol->cpu); - u32 checkfid; - u32 checkvid; - unsigned int newstate; - int ret = -EIO; - - if (!data) - return -EINVAL; - - checkfid = data->currfid; - checkvid = data->currvid; - - /* only run on specific CPU from here on. */ - /* This is poor form: use a workqueue or smp_call_function_single */ - if (!alloc_cpumask_var(&oldmask, GFP_KERNEL)) - return -ENOMEM; - - cpumask_copy(oldmask, tsk_cpus_allowed(current)); - set_cpus_allowed_ptr(current, cpumask_of(pol->cpu)); - - if (smp_processor_id() != pol->cpu) { - printk(KERN_ERR PFX "limiting to cpu %u failed\n", pol->cpu); - goto err_out; - } - - if (pending_bit_stuck()) { - printk(KERN_ERR PFX "failing targ, change pending bit set\n"); - goto err_out; - } - - dprintk("targ: cpu %d, %d kHz, min %d, max %d, relation %d\n", - pol->cpu, targfreq, pol->min, pol->max, relation); - - if (query_current_values_with_pending_wait(data)) - goto err_out; - - if (cpu_family != CPU_HW_PSTATE) { - dprintk("targ: curr fid 0x%x, vid 0x%x\n", - data->currfid, data->currvid); - - if ((checkvid != data->currvid) || - (checkfid != data->currfid)) { - printk(KERN_INFO PFX - "error - out of sync, fix 0x%x 0x%x, " - "vid 0x%x 0x%x\n", - checkfid, data->currfid, - checkvid, data->currvid); - } - } - - if (cpufreq_frequency_table_target(pol, data->powernow_table, - targfreq, relation, &newstate)) - goto err_out; - - mutex_lock(&fidvid_mutex); - - powernow_k8_acpi_pst_values(data, newstate); - - if (cpu_family == CPU_HW_PSTATE) - ret = transition_frequency_pstate(data, newstate); - else - ret = transition_frequency_fidvid(data, newstate); - if (ret) { - printk(KERN_ERR PFX "transition frequency failed\n"); - ret = 1; - mutex_unlock(&fidvid_mutex); - goto err_out; - } - mutex_unlock(&fidvid_mutex); - - if (cpu_family == CPU_HW_PSTATE) - pol->cur = find_khz_freq_from_pstate(data->powernow_table, - newstate); - else - pol->cur = find_khz_freq_from_fid(data->currfid); - ret = 0; - -err_out: - set_cpus_allowed_ptr(current, oldmask); - free_cpumask_var(oldmask); - return ret; -} - -/* Driver entry point to verify the policy and range of frequencies */ -static int powernowk8_verify(struct cpufreq_policy *pol) -{ - struct powernow_k8_data *data = per_cpu(powernow_data, pol->cpu); - - if (!data) - return -EINVAL; - - return cpufreq_frequency_table_verify(pol, data->powernow_table); -} - -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; - struct init_on_cpu init_on_cpu; - int rc; - struct cpuinfo_x86 *c = &cpu_data(pol->cpu); - - if (!cpu_online(pol->cpu)) - return -ENODEV; - - smp_call_function_single(pol->cpu, check_supported_cpu, &rc, 1); - if (rc) - return -ENODEV; - - data = kzalloc(sizeof(struct powernow_k8_data), GFP_KERNEL); - if (!data) { - printk(KERN_ERR PFX "unable to alloc powernow_k8_data"); - return -ENOMEM; - } - - data->cpu = pol->cpu; - data->currpstate = HW_PSTATE_INVALID; - - if (powernow_k8_cpu_init_acpi(data)) { - /* - * Use the PSB BIOS structure. This is only available on - * an UP version, and is deprecated by AMD. - */ - if (num_online_cpus() != 1) { - printk_once(ACPI_PSS_BIOS_BUG_MSG); - goto err_out; - } - if (pol->cpu != 0) { - printk(KERN_ERR FW_BUG PFX "No ACPI _PSS objects for " - "CPU other than CPU0. Complain to your BIOS " - "vendor.\n"); - goto err_out; - } - rc = find_psb_table(data); - if (rc) - goto err_out; - - /* Take a crude guess here. - * That guess was in microseconds, so multiply with 1000 */ - pol->cpuinfo.transition_latency = ( - ((data->rvo + 8) * data->vstable * VST_UNITS_20US) + - ((1 << data->irt) * 30)) * 1000; - } else /* ACPI _PSS objects available */ - pol->cpuinfo.transition_latency = get_transition_latency(data); - - /* only run on specific CPU from here on */ - 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)); - else - cpumask_copy(pol->cpus, cpu_core_mask(pol->cpu)); - data->available_cores = pol->cpus; - - if (cpu_family == CPU_HW_PSTATE) - pol->cur = find_khz_freq_from_pstate(data->powernow_table, - data->currpstate); - else - pol->cur = find_khz_freq_from_fid(data->currfid); - dprintk("policy current frequency %d kHz\n", pol->cur); - - /* min/max the cpu is capable of */ - if (cpufreq_frequency_table_cpuinfo(pol, data->powernow_table)) { - printk(KERN_ERR FW_BUG PFX "invalid powernow_table\n"); - powernow_k8_cpu_exit_acpi(data); - kfree(data->powernow_table); - kfree(data); - return -EINVAL; - } - - /* Check for APERF/MPERF support in hardware */ - if (cpu_has(c, X86_FEATURE_APERFMPERF)) - cpufreq_amd64_driver.getavg = cpufreq_get_measured_perf; - - cpufreq_frequency_table_get_attr(data->powernow_table, pol->cpu); - - if (cpu_family == CPU_HW_PSTATE) - dprintk("cpu_init done, current pstate 0x%x\n", - data->currpstate); - else - dprintk("cpu_init done, current fid 0x%x, vid 0x%x\n", - data->currfid, data->currvid); - - per_cpu(powernow_data, pol->cpu) = data; - - return 0; - -err_out_exit_acpi: - powernow_k8_cpu_exit_acpi(data); - -err_out: - kfree(data); - return -ENODEV; -} - -static int __devexit powernowk8_cpu_exit(struct cpufreq_policy *pol) -{ - struct powernow_k8_data *data = per_cpu(powernow_data, pol->cpu); - - if (!data) - return -EINVAL; - - powernow_k8_cpu_exit_acpi(data); - - cpufreq_frequency_table_put_attr(pol->cpu); - - kfree(data->powernow_table); - kfree(data); - per_cpu(powernow_data, pol->cpu) = NULL; - - return 0; -} - -static void query_values_on_cpu(void *_err) -{ - int *err = _err; - struct powernow_k8_data *data = __this_cpu_read(powernow_data); - - *err = query_current_values_with_pending_wait(data); -} - -static unsigned int powernowk8_get(unsigned int cpu) -{ - struct powernow_k8_data *data = per_cpu(powernow_data, cpu); - unsigned int khz = 0; - int err; - - if (!data) - return 0; - - smp_call_function_single(cpu, query_values_on_cpu, &err, true); - if (err) - goto out; - - if (cpu_family == CPU_HW_PSTATE) - khz = find_khz_freq_from_pstate(data->powernow_table, - data->currpstate); - else - khz = find_khz_freq_from_fid(data->currfid); - - -out: - return khz; -} - -static void _cpb_toggle_msrs(bool t) -{ - int cpu; - - get_online_cpus(); - - rdmsr_on_cpus(cpu_online_mask, MSR_K7_HWCR, msrs); - - for_each_cpu(cpu, cpu_online_mask) { - struct msr *reg = per_cpu_ptr(msrs, cpu); - if (t) - reg->l &= ~BIT(25); - else - reg->l |= BIT(25); - } - wrmsr_on_cpus(cpu_online_mask, MSR_K7_HWCR, msrs); - - put_online_cpus(); -} - -/* - * Switch on/off core performance boosting. - * - * 0=disable - * 1=enable. - */ -static void cpb_toggle(bool t) -{ - if (!cpb_capable) - return; - - if (t && !cpb_enabled) { - cpb_enabled = true; - _cpb_toggle_msrs(t); - printk(KERN_INFO PFX "Core Boosting enabled.\n"); - } else if (!t && cpb_enabled) { - cpb_enabled = false; - _cpb_toggle_msrs(t); - printk(KERN_INFO PFX "Core Boosting disabled.\n"); - } -} - -static ssize_t store_cpb(struct cpufreq_policy *policy, const char *buf, - size_t count) -{ - int ret = -EINVAL; - unsigned long val = 0; - - ret = strict_strtoul(buf, 10, &val); - if (!ret && (val == 0 || val == 1) && cpb_capable) - cpb_toggle(val); - else - return -EINVAL; - - return count; -} - -static ssize_t show_cpb(struct cpufreq_policy *policy, char *buf) -{ - return sprintf(buf, "%u\n", cpb_enabled); -} - -#define define_one_rw(_name) \ -static struct freq_attr _name = \ -__ATTR(_name, 0644, show_##_name, store_##_name) - -define_one_rw(cpb); - -static struct freq_attr *powernow_k8_attr[] = { - &cpufreq_freq_attr_scaling_available_freqs, - &cpb, - NULL, -}; - -static struct cpufreq_driver cpufreq_amd64_driver = { - .verify = powernowk8_verify, - .target = powernowk8_target, - .bios_limit = acpi_processor_get_bios_limit, - .init = powernowk8_cpu_init, - .exit = __devexit_p(powernowk8_cpu_exit), - .get = powernowk8_get, - .name = "powernow-k8", - .owner = THIS_MODULE, - .attr = powernow_k8_attr, -}; - -/* - * Clear the boost-disable flag on the CPU_DOWN path so that this cpu - * cannot block the remaining ones from boosting. On the CPU_UP path we - * simply keep the boost-disable flag in sync with the current global - * state. - */ -static int cpb_notify(struct notifier_block *nb, unsigned long action, - void *hcpu) -{ - unsigned cpu = (long)hcpu; - u32 lo, hi; - - switch (action) { - case CPU_UP_PREPARE: - case CPU_UP_PREPARE_FROZEN: - - if (!cpb_enabled) { - rdmsr_on_cpu(cpu, MSR_K7_HWCR, &lo, &hi); - lo |= BIT(25); - wrmsr_on_cpu(cpu, MSR_K7_HWCR, lo, hi); - } - break; - - case CPU_DOWN_PREPARE: - case CPU_DOWN_PREPARE_FROZEN: - rdmsr_on_cpu(cpu, MSR_K7_HWCR, &lo, &hi); - lo &= ~BIT(25); - wrmsr_on_cpu(cpu, MSR_K7_HWCR, lo, hi); - break; - - default: - break; - } - - return NOTIFY_OK; -} - -static struct notifier_block cpb_nb = { - .notifier_call = cpb_notify, -}; - -/* driver entry point for init */ -static int __cpuinit powernowk8_init(void) -{ - unsigned int i, supported_cpus = 0, cpu; - int rv; - - for_each_online_cpu(i) { - int rc; - smp_call_function_single(i, check_supported_cpu, &rc, 1); - if (rc == 0) - supported_cpus++; - } - - if (supported_cpus != num_online_cpus()) - return -ENODEV; - - printk(KERN_INFO PFX "Found %d %s (%d cpu cores) (" VERSION ")\n", - num_online_nodes(), boot_cpu_data.x86_model_id, supported_cpus); - - if (boot_cpu_has(X86_FEATURE_CPB)) { - - cpb_capable = true; - - msrs = msrs_alloc(); - if (!msrs) { - printk(KERN_ERR "%s: Error allocating msrs!\n", __func__); - return -ENOMEM; - } - - register_cpu_notifier(&cpb_nb); - - rdmsr_on_cpus(cpu_online_mask, MSR_K7_HWCR, msrs); - - for_each_cpu(cpu, cpu_online_mask) { - struct msr *reg = per_cpu_ptr(msrs, cpu); - cpb_enabled |= !(!!(reg->l & BIT(25))); - } - - printk(KERN_INFO PFX "Core Performance Boosting: %s.\n", - (cpb_enabled ? "on" : "off")); - } - - rv = cpufreq_register_driver(&cpufreq_amd64_driver); - if (rv < 0 && boot_cpu_has(X86_FEATURE_CPB)) { - unregister_cpu_notifier(&cpb_nb); - msrs_free(msrs); - msrs = NULL; - } - return rv; -} - -/* driver entry point for term */ -static void __exit powernowk8_exit(void) -{ - dprintk("exit\n"); - - if (boot_cpu_has(X86_FEATURE_CPB)) { - msrs_free(msrs); - msrs = NULL; - - unregister_cpu_notifier(&cpb_nb); - } - - cpufreq_unregister_driver(&cpufreq_amd64_driver); -} - -MODULE_AUTHOR("Paul Devriendt <paul.devriendt@amd.com> and " - "Mark Langsdorf <mark.langsdorf@amd.com>"); -MODULE_DESCRIPTION("AMD Athlon 64 and Opteron processor frequency driver."); -MODULE_LICENSE("GPL"); - -late_initcall(powernowk8_init); -module_exit(powernowk8_exit); diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k8.h b/arch/x86/kernel/cpu/cpufreq/powernow-k8.h deleted file mode 100644 index df3529b1c02d..000000000000 --- a/arch/x86/kernel/cpu/cpufreq/powernow-k8.h +++ /dev/null @@ -1,224 +0,0 @@ -/* - * (c) 2003-2006 Advanced Micro Devices, Inc. - * Your use of this code is subject to the terms and conditions of the - * GNU general public license version 2. See "COPYING" or - * http://www.gnu.org/licenses/gpl.html - */ - -enum pstate { - HW_PSTATE_INVALID = 0xff, - HW_PSTATE_0 = 0, - HW_PSTATE_1 = 1, - HW_PSTATE_2 = 2, - HW_PSTATE_3 = 3, - HW_PSTATE_4 = 4, - HW_PSTATE_5 = 5, - HW_PSTATE_6 = 6, - HW_PSTATE_7 = 7, -}; - -struct powernow_k8_data { - unsigned int cpu; - - u32 numps; /* number of p-states */ - u32 batps; /* number of p-states supported on battery */ - u32 max_hw_pstate; /* maximum legal hardware pstate */ - - /* these values are constant when the PSB is used to determine - * vid/fid pairings, but are modified during the ->target() call - * when ACPI is used */ - u32 rvo; /* ramp voltage offset */ - u32 irt; /* isochronous relief time */ - u32 vidmvs; /* usable value calculated from mvs */ - u32 vstable; /* voltage stabilization time, units 20 us */ - u32 plllock; /* pll lock time, units 1 us */ - u32 exttype; /* extended interface = 1 */ - - /* keep track of the current fid / vid or pstate */ - u32 currvid; - u32 currfid; - enum pstate currpstate; - - /* the powernow_table includes all frequency and vid/fid pairings: - * fid are the lower 8 bits of the index, vid are the upper 8 bits. - * frequency is in kHz */ - struct cpufreq_frequency_table *powernow_table; - - /* the acpi table needs to be kept. it's only available if ACPI was - * used to determine valid frequency/vid/fid states */ - struct acpi_processor_performance acpi_data; - - /* we need to keep track of associated cores, but let cpufreq - * handle hotplug events - so just point at cpufreq pol->cpus - * structure */ - struct cpumask *available_cores; -}; - -/* processor's cpuid instruction support */ -#define CPUID_PROCESSOR_SIGNATURE 1 /* function 1 */ -#define CPUID_XFAM 0x0ff00000 /* extended family */ -#define CPUID_XFAM_K8 0 -#define CPUID_XMOD 0x000f0000 /* extended model */ -#define CPUID_XMOD_REV_MASK 0x000c0000 -#define CPUID_XFAM_10H 0x00100000 /* family 0x10 */ -#define CPUID_USE_XFAM_XMOD 0x00000f00 -#define CPUID_GET_MAX_CAPABILITIES 0x80000000 -#define CPUID_FREQ_VOLT_CAPABILITIES 0x80000007 -#define P_STATE_TRANSITION_CAPABLE 6 - -/* Model Specific Registers for p-state transitions. MSRs are 64-bit. For */ -/* writes (wrmsr - opcode 0f 30), the register number is placed in ecx, and */ -/* the value to write is placed in edx:eax. For reads (rdmsr - opcode 0f 32), */ -/* the register number is placed in ecx, and the data is returned in edx:eax. */ - -#define MSR_FIDVID_CTL 0xc0010041 -#define MSR_FIDVID_STATUS 0xc0010042 - -/* Field definitions within the FID VID Low Control MSR : */ -#define MSR_C_LO_INIT_FID_VID 0x00010000 -#define MSR_C_LO_NEW_VID 0x00003f00 -#define MSR_C_LO_NEW_FID 0x0000003f -#define MSR_C_LO_VID_SHIFT 8 - -/* Field definitions within the FID VID High Control MSR : */ -#define MSR_C_HI_STP_GNT_TO 0x000fffff - -/* Field definitions within the FID VID Low Status MSR : */ -#define MSR_S_LO_CHANGE_PENDING 0x80000000 /* cleared when completed */ -#define MSR_S_LO_MAX_RAMP_VID 0x3f000000 -#define MSR_S_LO_MAX_FID 0x003f0000 -#define MSR_S_LO_START_FID 0x00003f00 -#define MSR_S_LO_CURRENT_FID 0x0000003f - -/* Field definitions within the FID VID High Status MSR : */ -#define MSR_S_HI_MIN_WORKING_VID 0x3f000000 -#define MSR_S_HI_MAX_WORKING_VID 0x003f0000 -#define MSR_S_HI_START_VID 0x00003f00 -#define MSR_S_HI_CURRENT_VID 0x0000003f -#define MSR_C_HI_STP_GNT_BENIGN 0x00000001 - - -/* Hardware Pstate _PSS and MSR definitions */ -#define USE_HW_PSTATE 0x00000080 -#define HW_PSTATE_MASK 0x00000007 -#define HW_PSTATE_VALID_MASK 0x80000000 -#define HW_PSTATE_MAX_MASK 0x000000f0 -#define HW_PSTATE_MAX_SHIFT 4 -#define MSR_PSTATE_DEF_BASE 0xc0010064 /* base of Pstate MSRs */ -#define MSR_PSTATE_STATUS 0xc0010063 /* Pstate Status MSR */ -#define MSR_PSTATE_CTRL 0xc0010062 /* Pstate control MSR */ -#define MSR_PSTATE_CUR_LIMIT 0xc0010061 /* pstate current limit MSR */ - -/* define the two driver architectures */ -#define CPU_OPTERON 0 -#define CPU_HW_PSTATE 1 - - -/* - * There are restrictions frequencies have to follow: - * - only 1 entry in the low fid table ( <=1.4GHz ) - * - lowest entry in the high fid table must be >= 2 * the entry in the - * low fid table - * - lowest entry in the high fid table must be a <= 200MHz + 2 * the entry - * in the low fid table - * - the parts can only step at <= 200 MHz intervals, odd fid values are - * supported in revision G and later revisions. - * - lowest frequency must be >= interprocessor hypertransport link speed - * (only applies to MP systems obviously) - */ - -/* fids (frequency identifiers) are arranged in 2 tables - lo and hi */ -#define LO_FID_TABLE_TOP 7 /* fid values marking the boundary */ -#define HI_FID_TABLE_BOTTOM 8 /* between the low and high tables */ - -#define LO_VCOFREQ_TABLE_TOP 1400 /* corresponding vco frequency values */ -#define HI_VCOFREQ_TABLE_BOTTOM 1600 - -#define MIN_FREQ_RESOLUTION 200 /* fids jump by 2 matching freq jumps by 200 */ - -#define MAX_FID 0x2a /* Spec only gives FID values as far as 5 GHz */ -#define LEAST_VID 0x3e /* Lowest (numerically highest) useful vid value */ - -#define MIN_FREQ 800 /* Min and max freqs, per spec */ -#define MAX_FREQ 5000 - -#define INVALID_FID_MASK 0xffffffc0 /* not a valid fid if these bits are set */ -#define INVALID_VID_MASK 0xffffffc0 /* not a valid vid if these bits are set */ - -#define VID_OFF 0x3f - -#define STOP_GRANT_5NS 1 /* min poss memory access latency for voltage change */ - -#define PLL_LOCK_CONVERSION (1000/5) /* ms to ns, then divide by clock period */ - -#define MAXIMUM_VID_STEPS 1 /* Current cpus only allow a single step of 25mV */ -#define VST_UNITS_20US 20 /* Voltage Stabilization Time is in units of 20us */ - -/* - * Most values of interest are encoded in a single field of the _PSS - * entries: the "control" value. - */ - -#define IRT_SHIFT 30 -#define RVO_SHIFT 28 -#define EXT_TYPE_SHIFT 27 -#define PLL_L_SHIFT 20 -#define MVS_SHIFT 18 -#define VST_SHIFT 11 -#define VID_SHIFT 6 -#define IRT_MASK 3 -#define RVO_MASK 3 -#define EXT_TYPE_MASK 1 -#define PLL_L_MASK 0x7f -#define MVS_MASK 3 -#define VST_MASK 0x7f -#define VID_MASK 0x1f -#define FID_MASK 0x1f -#define EXT_VID_MASK 0x3f -#define EXT_FID_MASK 0x3f - - -/* - * Version 1.4 of the PSB table. This table is constructed by BIOS and is - * to tell the OS's power management driver which VIDs and FIDs are - * supported by this particular processor. - * If the data in the PSB / PST is wrong, then this driver will program the - * wrong values into hardware, which is very likely to lead to a crash. - */ - -#define PSB_ID_STRING "AMDK7PNOW!" -#define PSB_ID_STRING_LEN 10 - -#define PSB_VERSION_1_4 0x14 - -struct psb_s { - u8 signature[10]; - u8 tableversion; - u8 flags1; - u16 vstable; - u8 flags2; - u8 num_tables; - u32 cpuid; - u8 plllocktime; - u8 maxfid; - u8 maxvid; - u8 numps; -}; - -/* Pairs of fid/vid values are appended to the version 1.4 PSB table. */ -struct pst_s { - u8 fid; - u8 vid; -}; - -#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, 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); - -static void powernow_k8_acpi_pst_values(struct powernow_k8_data *data, unsigned int index); - -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); diff --git a/arch/x86/kernel/cpu/cpufreq/sc520_freq.c b/arch/x86/kernel/cpu/cpufreq/sc520_freq.c deleted file mode 100644 index 435a996a613a..000000000000 --- a/arch/x86/kernel/cpu/cpufreq/sc520_freq.c +++ /dev/null @@ -1,194 +0,0 @@ -/* - * sc520_freq.c: cpufreq driver for the AMD Elan sc520 - * - * Copyright (C) 2005 Sean Young <sean@mess.org> - * - * This program is free software; you can redistribute it and/or - * modify it under the terms of the GNU General Public License - * as published by the Free Software Foundation; either version - * 2 of the License, or (at your option) any later version. - * - * Based on elanfreq.c - * - * 2005-03-30: - initial revision - */ - -#include <linux/kernel.h> -#include <linux/module.h> -#include <linux/init.h> - -#include <linux/delay.h> -#include <linux/cpufreq.h> -#include <linux/timex.h> -#include <linux/io.h> - -#include <asm/msr.h> - -#define MMCR_BASE 0xfffef000 /* The default base address */ -#define OFFS_CPUCTL 0x2 /* CPU Control Register */ - -static __u8 __iomem *cpuctl; - -#define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, \ - "sc520_freq", msg) -#define PFX "sc520_freq: " - -static struct cpufreq_frequency_table sc520_freq_table[] = { - {0x01, 100000}, - {0x02, 133000}, - {0, CPUFREQ_TABLE_END}, -}; - -static unsigned int sc520_freq_get_cpu_frequency(unsigned int cpu) -{ - u8 clockspeed_reg = *cpuctl; - - switch (clockspeed_reg & 0x03) { - default: - printk(KERN_ERR PFX "error: cpuctl register has unexpected " - "value %02x\n", clockspeed_reg); - case 0x01: - return 100000; - case 0x02: - return 133000; - } -} - -static void sc520_freq_set_cpu_state(unsigned int state) -{ - - struct cpufreq_freqs freqs; - u8 clockspeed_reg; - - freqs.old = sc520_freq_get_cpu_frequency(0); - freqs.new = sc520_freq_table[state].frequency; - freqs.cpu = 0; /* AMD Elan is UP */ - - cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); - - dprintk("attempting to set frequency to %i kHz\n", - sc520_freq_table[state].frequency); - - local_irq_disable(); - - clockspeed_reg = *cpuctl & ~0x03; - *cpuctl = clockspeed_reg | sc520_freq_table[state].index; - - local_irq_enable(); - - cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); -}; - -static int sc520_freq_verify(struct cpufreq_policy *policy) -{ - return cpufreq_frequency_table_verify(policy, &sc520_freq_table[0]); -} - -static int sc520_freq_target(struct cpufreq_policy *policy, - unsigned int target_freq, - unsigned int relation) -{ - unsigned int newstate = 0; - - if (cpufreq_frequency_table_target(policy, sc520_freq_table, - target_freq, relation, &newstate)) - return -EINVAL; - - sc520_freq_set_cpu_state(newstate); - - return 0; -} - - -/* - * Module init and exit code - */ - -static int sc520_freq_cpu_init(struct cpufreq_policy *policy) -{ - struct cpuinfo_x86 *c = &cpu_data(0); - int result; - - /* capability check */ - if (c->x86_vendor != X86_VENDOR_AMD || - c->x86 != 4 || c->x86_model != 9) - return -ENODEV; - - /* cpuinfo and default policy values */ - policy->cpuinfo.transition_latency = 1000000; /* 1ms */ - policy->cur = sc520_freq_get_cpu_frequency(0); - - result = cpufreq_frequency_table_cpuinfo(policy, sc520_freq_table); - if (result) - return result; - - cpufreq_frequency_table_get_attr(sc520_freq_table, policy->cpu); - - return 0; -} - - -static int sc520_freq_cpu_exit(struct cpufreq_policy *policy) -{ - cpufreq_frequency_table_put_attr(policy->cpu); - return 0; -} - - -static struct freq_attr *sc520_freq_attr[] = { - &cpufreq_freq_attr_scaling_available_freqs, - NULL, -}; - - -static struct cpufreq_driver sc520_freq_driver = { - .get = sc520_freq_get_cpu_frequency, - .verify = sc520_freq_verify, - .target = sc520_freq_target, - .init = sc520_freq_cpu_init, - .exit = sc520_freq_cpu_exit, - .name = "sc520_freq", - .owner = THIS_MODULE, - .attr = sc520_freq_attr, -}; - - -static int __init sc520_freq_init(void) -{ - struct cpuinfo_x86 *c = &cpu_data(0); - int err; - - /* Test if we have the right hardware */ - if (c->x86_vendor != X86_VENDOR_AMD || - c->x86 != 4 || c->x86_model != 9) { - dprintk("no Elan SC520 processor found!\n"); - return -ENODEV; - } - cpuctl = ioremap((unsigned long)(MMCR_BASE + OFFS_CPUCTL), 1); - if (!cpuctl) { - printk(KERN_ERR "sc520_freq: error: failed to remap memory\n"); - return -ENOMEM; - } - - err = cpufreq_register_driver(&sc520_freq_driver); - if (err) - iounmap(cpuctl); - - return err; -} - - -static void __exit sc520_freq_exit(void) -{ - cpufreq_unregister_driver(&sc520_freq_driver); - iounmap(cpuctl); -} - - -MODULE_LICENSE("GPL"); -MODULE_AUTHOR("Sean Young <sean@mess.org>"); -MODULE_DESCRIPTION("cpufreq driver for AMD's Elan sc520 CPU"); - -module_init(sc520_freq_init); -module_exit(sc520_freq_exit); - diff --git a/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c b/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c deleted file mode 100644 index 9b1ff37de46a..000000000000 --- a/arch/x86/kernel/cpu/cpufreq/speedstep-centrino.c +++ /dev/null @@ -1,636 +0,0 @@ -/* - * cpufreq driver for Enhanced SpeedStep, as found in Intel's Pentium - * M (part of the Centrino chipset). - * - * Since the original Pentium M, most new Intel CPUs support Enhanced - * SpeedStep. - * - * Despite the "SpeedStep" in the name, this is almost entirely unlike - * traditional SpeedStep. - * - * Modelled on speedstep.c - * - * Copyright (C) 2003 Jeremy Fitzhardinge <jeremy@goop.org> - */ - -#include <linux/kernel.h> -#include <linux/module.h> -#include <linux/init.h> -#include <linux/cpufreq.h> -#include <linux/sched.h> /* current */ -#include <linux/delay.h> -#include <linux/compiler.h> -#include <linux/gfp.h> - -#include <asm/msr.h> -#include <asm/processor.h> -#include <asm/cpufeature.h> - -#define PFX "speedstep-centrino: " -#define MAINTAINER "cpufreq@vger.kernel.org" - -#define dprintk(msg...) \ - cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, "speedstep-centrino", msg) - -#define INTEL_MSR_RANGE (0xffff) - -struct cpu_id -{ - __u8 x86; /* CPU family */ - __u8 x86_model; /* model */ - __u8 x86_mask; /* stepping */ -}; - -enum { - CPU_BANIAS, - CPU_DOTHAN_A1, - CPU_DOTHAN_A2, - CPU_DOTHAN_B0, - CPU_MP4HT_D0, - CPU_MP4HT_E0, -}; - -static const struct cpu_id cpu_ids[] = { - [CPU_BANIAS] = { 6, 9, 5 }, - [CPU_DOTHAN_A1] = { 6, 13, 1 }, - [CPU_DOTHAN_A2] = { 6, 13, 2 }, - [CPU_DOTHAN_B0] = { 6, 13, 6 }, - [CPU_MP4HT_D0] = {15, 3, 4 }, - [CPU_MP4HT_E0] = {15, 4, 1 }, -}; -#define N_IDS ARRAY_SIZE(cpu_ids) - -struct cpu_model -{ - const struct cpu_id *cpu_id; - const char *model_name; - unsigned max_freq; /* max clock in kHz */ - - struct cpufreq_frequency_table *op_points; /* clock/voltage pairs */ -}; -static int centrino_verify_cpu_id(const struct cpuinfo_x86 *c, - const struct cpu_id *x); - -/* Operating points for current CPU */ -static DEFINE_PER_CPU(struct cpu_model *, centrino_model); -static DEFINE_PER_CPU(const struct cpu_id *, centrino_cpu); - -static struct cpufreq_driver centrino_driver; - -#ifdef CONFIG_X86_SPEEDSTEP_CENTRINO_TABLE - -/* Computes the correct form for IA32_PERF_CTL MSR for a particular - frequency/voltage operating point; frequency in MHz, volts in mV. - This is stored as "index" in the structure. */ -#define OP(mhz, mv) \ - { \ - .frequency = (mhz) * 1000, \ - .index = (((mhz)/100) << 8) | ((mv - 700) / 16) \ - } - -/* - * These voltage tables were derived from the Intel Pentium M - * datasheet, document 25261202.pdf, Table 5. I have verified they - * are consistent with my IBM ThinkPad X31, which has a 1.3GHz Pentium - * M. - */ - -/* Ultra Low Voltage Intel Pentium M processor 900MHz (Banias) */ -static struct cpufreq_frequency_table banias_900[] = -{ - OP(600, 844), - OP(800, 988), - OP(900, 1004), - { .frequency = CPUFREQ_TABLE_END } -}; - -/* Ultra Low Voltage Intel Pentium M processor 1000MHz (Banias) */ -static struct cpufreq_frequency_table banias_1000[] = -{ - OP(600, 844), - OP(800, 972), - OP(900, 988), - OP(1000, 1004), - { .frequency = CPUFREQ_TABLE_END } -}; - -/* Low Voltage Intel Pentium M processor 1.10GHz (Banias) */ -static struct cpufreq_frequency_table banias_1100[] = -{ - OP( 600, 956), - OP( 800, 1020), - OP( 900, 1100), - OP(1000, 1164), - OP(1100, 1180), - { .frequency = CPUFREQ_TABLE_END } -}; - - -/* Low Voltage Intel Pentium M processor 1.20GHz (Banias) */ -static struct cpufreq_frequency_table banias_1200[] = -{ - OP( 600, 956), - OP( 800, 1004), - OP( 900, 1020), - OP(1000, 1100), - OP(1100, 1164), - OP(1200, 1180), - { .frequency = CPUFREQ_TABLE_END } -}; - -/* Intel Pentium M processor 1.30GHz (Banias) */ -static struct cpufreq_frequency_table banias_1300[] = -{ - OP( 600, 956), - OP( 800, 1260), - OP(1000, 1292), - OP(1200, 1356), - OP(1300, 1388), - { .frequency = CPUFREQ_TABLE_END } -}; - -/* Intel Pentium M processor 1.40GHz (Banias) */ -static struct cpufreq_frequency_table banias_1400[] = -{ - OP( 600, 956), - OP( 800, 1180), - OP(1000, 1308), - OP(1200, 1436), - OP(1400, 1484), - { .frequency = CPUFREQ_TABLE_END } -}; - -/* Intel Pentium M processor 1.50GHz (Banias) */ -static struct cpufreq_frequency_table banias_1500[] = -{ - OP( 600, 956), - OP( 800, 1116), - OP(1000, 1228), - OP(1200, 1356), - OP(1400, 1452), - OP(1500, 1484), - { .frequency = CPUFREQ_TABLE_END } -}; - -/* Intel Pentium M processor 1.60GHz (Banias) */ -static struct cpufreq_frequency_table banias_1600[] = -{ - OP( 600, 956), - OP( 800, 1036), - OP(1000, 1164), - OP(1200, 1276), - OP(1400, 1420), - OP(1600, 1484), - { .frequency = CPUFREQ_TABLE_END } -}; - -/* Intel Pentium M processor 1.70GHz (Banias) */ -static struct cpufreq_frequency_table banias_1700[] = -{ - OP( 600, 956), - OP( 800, 1004), - OP(1000, 1116), - OP(1200, 1228), - OP(1400, 1308), - OP(1700, 1484), - { .frequency = CPUFREQ_TABLE_END } -}; -#undef OP - -#define _BANIAS(cpuid, max, name) \ -{ .cpu_id = cpuid, \ - .model_name = "Intel(R) Pentium(R) M processor " name "MHz", \ - .max_freq = (max)*1000, \ - .op_points = banias_##max, \ -} -#define BANIAS(max) _BANIAS(&cpu_ids[CPU_BANIAS], max, #max) - -/* CPU models, their operating frequency range, and freq/voltage - operating points */ -static struct cpu_model models[] = -{ - _BANIAS(&cpu_ids[CPU_BANIAS], 900, " 900"), - BANIAS(1000), - BANIAS(1100), - BANIAS(1200), - BANIAS(1300), - BANIAS(1400), - BANIAS(1500), - BANIAS(1600), - BANIAS(1700), - - /* NULL model_name is a wildcard */ - { &cpu_ids[CPU_DOTHAN_A1], NULL, 0, NULL }, - { &cpu_ids[CPU_DOTHAN_A2], NULL, 0, NULL }, - { &cpu_ids[CPU_DOTHAN_B0], NULL, 0, NULL }, - { &cpu_ids[CPU_MP4HT_D0], NULL, 0, NULL }, - { &cpu_ids[CPU_MP4HT_E0], NULL, 0, NULL }, - - { NULL, } -}; -#undef _BANIAS -#undef BANIAS - -static int centrino_cpu_init_table(struct cpufreq_policy *policy) -{ - struct cpuinfo_x86 *cpu = &cpu_data(policy->cpu); - struct cpu_model *model; - - for(model = models; model->cpu_id != NULL; model++) - if (centrino_verify_cpu_id(cpu, model->cpu_id) && - (model->model_name == NULL || - strcmp(cpu->x86_model_id, model->model_name) == 0)) - break; - - if (model->cpu_id == NULL) { - /* No match at all */ - dprintk("no support for CPU model \"%s\": " - "send /proc/cpuinfo to " MAINTAINER "\n", - cpu->x86_model_id); - return -ENOENT; - } - - if (model->op_points == NULL) { - /* Matched a non-match */ - dprintk("no table support for CPU model \"%s\"\n", - cpu->x86_model_id); - dprintk("try using the acpi-cpufreq driver\n"); - return -ENOENT; - } - - per_cpu(centrino_model, policy->cpu) = model; - - dprintk("found \"%s\": max frequency: %dkHz\n", - model->model_name, model->max_freq); - - return 0; -} - -#else -static inline int centrino_cpu_init_table(struct cpufreq_policy *policy) -{ - return -ENODEV; -} -#endif /* CONFIG_X86_SPEEDSTEP_CENTRINO_TABLE */ - -static int centrino_verify_cpu_id(const struct cpuinfo_x86 *c, - const struct cpu_id *x) -{ - if ((c->x86 == x->x86) && - (c->x86_model == x->x86_model) && - (c->x86_mask == x->x86_mask)) - return 1; - return 0; -} - -/* To be called only after centrino_model is initialized */ -static unsigned extract_clock(unsigned msr, unsigned int cpu, int failsafe) -{ - int i; - - /* - * Extract clock in kHz from PERF_CTL value - * for centrino, as some DSDTs are buggy. - * Ideally, this can be done using the acpi_data structure. - */ - if ((per_cpu(centrino_cpu, cpu) == &cpu_ids[CPU_BANIAS]) || - (per_cpu(centrino_cpu, cpu) == &cpu_ids[CPU_DOTHAN_A1]) || - (per_cpu(centrino_cpu, cpu) == &cpu_ids[CPU_DOTHAN_B0])) { - msr = (msr >> 8) & 0xff; - return msr * 100000; - } - - if ((!per_cpu(centrino_model, cpu)) || - (!per_cpu(centrino_model, cpu)->op_points)) - return 0; - - msr &= 0xffff; - for (i = 0; - per_cpu(centrino_model, cpu)->op_points[i].frequency - != CPUFREQ_TABLE_END; - i++) { - if (msr == per_cpu(centrino_model, cpu)->op_points[i].index) - return per_cpu(centrino_model, cpu)-> - op_points[i].frequency; - } - if (failsafe) - return per_cpu(centrino_model, cpu)->op_points[i-1].frequency; - else - return 0; -} - -/* Return the current CPU frequency in kHz */ -static unsigned int get_cur_freq(unsigned int cpu) -{ - unsigned l, h; - unsigned clock_freq; - - rdmsr_on_cpu(cpu, MSR_IA32_PERF_STATUS, &l, &h); - clock_freq = extract_clock(l, cpu, 0); - - if (unlikely(clock_freq == 0)) { - /* - * On some CPUs, we can see transient MSR values (which are - * not present in _PSS), while CPU is doing some automatic - * P-state transition (like TM2). Get the last freq set - * in PERF_CTL. - */ - rdmsr_on_cpu(cpu, MSR_IA32_PERF_CTL, &l, &h); - clock_freq = extract_clock(l, cpu, 1); - } - return clock_freq; -} - - -static int centrino_cpu_init(struct cpufreq_policy *policy) -{ - struct cpuinfo_x86 *cpu = &cpu_data(policy->cpu); - unsigned freq; - unsigned l, h; - int ret; - int i; - - /* Only Intel makes Enhanced Speedstep-capable CPUs */ - if (cpu->x86_vendor != X86_VENDOR_INTEL || - !cpu_has(cpu, X86_FEATURE_EST)) - return -ENODEV; - - if (cpu_has(cpu, X86_FEATURE_CONSTANT_TSC)) - centrino_driver.flags |= CPUFREQ_CONST_LOOPS; - - if (policy->cpu != 0) - return -ENODEV; - - for (i = 0; i < N_IDS; i++) - if (centrino_verify_cpu_id(cpu, &cpu_ids[i])) - break; - - if (i != N_IDS) - per_cpu(centrino_cpu, policy->cpu) = &cpu_ids[i]; - - if (!per_cpu(centrino_cpu, policy->cpu)) { - dprintk("found unsupported CPU with " - "Enhanced SpeedStep: send /proc/cpuinfo to " - MAINTAINER "\n"); - return -ENODEV; - } - - if (centrino_cpu_init_table(policy)) { - return -ENODEV; - } - - /* Check to see if Enhanced SpeedStep is enabled, and try to - enable it if not. */ - rdmsr(MSR_IA32_MISC_ENABLE, l, h); - - if (!(l & MSR_IA32_MISC_ENABLE_ENHANCED_SPEEDSTEP)) { - l |= MSR_IA32_MISC_ENABLE_ENHANCED_SPEEDSTEP; - dprintk("trying to enable Enhanced SpeedStep (%x)\n", l); - wrmsr(MSR_IA32_MISC_ENABLE, l, h); - - /* check to see if it stuck */ - rdmsr(MSR_IA32_MISC_ENABLE, l, h); - if (!(l & MSR_IA32_MISC_ENABLE_ENHANCED_SPEEDSTEP)) { - printk(KERN_INFO PFX - "couldn't enable Enhanced SpeedStep\n"); - return -ENODEV; - } - } - - freq = get_cur_freq(policy->cpu); - policy->cpuinfo.transition_latency = 10000; - /* 10uS transition latency */ - policy->cur = freq; - - dprintk("centrino_cpu_init: cur=%dkHz\n", policy->cur); - - ret = cpufreq_frequency_table_cpuinfo(policy, - per_cpu(centrino_model, policy->cpu)->op_points); - if (ret) - return (ret); - - cpufreq_frequency_table_get_attr( - per_cpu(centrino_model, policy->cpu)->op_points, policy->cpu); - - return 0; -} - -static int centrino_cpu_exit(struct cpufreq_policy *policy) -{ - unsigned int cpu = policy->cpu; - - if (!per_cpu(centrino_model, cpu)) - return -ENODEV; - - cpufreq_frequency_table_put_attr(cpu); - - per_cpu(centrino_model, cpu) = NULL; - - return 0; -} - -/** - * centrino_verify - verifies a new CPUFreq policy - * @policy: new policy - * - * Limit must be within this model's frequency range at least one - * border included. - */ -static int centrino_verify (struct cpufreq_policy *policy) -{ - return cpufreq_frequency_table_verify(policy, - per_cpu(centrino_model, policy->cpu)->op_points); -} - -/** - * centrino_setpolicy - set a new CPUFreq policy - * @policy: new policy - * @target_freq: the target frequency - * @relation: how that frequency relates to achieved frequency - * (CPUFREQ_RELATION_L or CPUFREQ_RELATION_H) - * - * Sets a new CPUFreq policy. - */ -static int centrino_target (struct cpufreq_policy *policy, - unsigned int target_freq, - unsigned int relation) -{ - unsigned int newstate = 0; - unsigned int msr, oldmsr = 0, h = 0, cpu = policy->cpu; - struct cpufreq_freqs freqs; - int retval = 0; - unsigned int j, k, first_cpu, tmp; - cpumask_var_t covered_cpus; - - if (unlikely(!zalloc_cpumask_var(&covered_cpus, GFP_KERNEL))) - return -ENOMEM; - - if (unlikely(per_cpu(centrino_model, cpu) == NULL)) { - retval = -ENODEV; - goto out; - } - - if (unlikely(cpufreq_frequency_table_target(policy, - per_cpu(centrino_model, cpu)->op_points, - target_freq, - relation, - &newstate))) { - retval = -EINVAL; - goto out; - } - - first_cpu = 1; - for_each_cpu(j, policy->cpus) { - int good_cpu; - - /* cpufreq holds the hotplug lock, so we are safe here */ - if (!cpu_online(j)) - continue; - - /* - * Support for SMP systems. - * Make sure we are running on CPU that wants to change freq - */ - if (policy->shared_type == CPUFREQ_SHARED_TYPE_ANY) - good_cpu = cpumask_any_and(policy->cpus, - cpu_online_mask); - else - good_cpu = j; - - 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 out; - } - break; - } - - msr = per_cpu(centrino_model, cpu)->op_points[newstate].index; - - if (first_cpu) { - 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 out; - } - - freqs.old = extract_clock(oldmsr, cpu, 0); - freqs.new = extract_clock(msr, cpu, 0); - - dprintk("target=%dkHz old=%d new=%d msr=%04x\n", - target_freq, freqs.old, freqs.new, msr); - - for_each_cpu(k, policy->cpus) { - if (!cpu_online(k)) - continue; - freqs.cpu = k; - cpufreq_notify_transition(&freqs, - CPUFREQ_PRECHANGE); - } - - first_cpu = 0; - /* all but 16 LSB are reserved, treat them with care */ - oldmsr &= ~0xffff; - msr &= 0xffff; - oldmsr |= msr; - } - - wrmsr_on_cpu(good_cpu, MSR_IA32_PERF_CTL, oldmsr, h); - if (policy->shared_type == CPUFREQ_SHARED_TYPE_ANY) - break; - - cpumask_set_cpu(j, covered_cpus); - } - - for_each_cpu(k, policy->cpus) { - if (!cpu_online(k)) - continue; - freqs.cpu = k; - cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); - } - - if (unlikely(retval)) { - /* - * We have failed halfway through the frequency change. - * We have sent callbacks to policy->cpus and - * MSRs have already been written on coverd_cpus. - * Best effort undo.. - */ - - for_each_cpu(j, covered_cpus) - wrmsr_on_cpu(j, MSR_IA32_PERF_CTL, oldmsr, h); - - tmp = freqs.new; - freqs.new = freqs.old; - freqs.old = tmp; - for_each_cpu(j, policy->cpus) { - if (!cpu_online(j)) - continue; - cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); - cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); - } - } - retval = 0; - -out: - free_cpumask_var(covered_cpus); - return retval; -} - -static struct freq_attr* centrino_attr[] = { - &cpufreq_freq_attr_scaling_available_freqs, - NULL, -}; - -static struct cpufreq_driver centrino_driver = { - .name = "centrino", /* should be speedstep-centrino, - but there's a 16 char limit */ - .init = centrino_cpu_init, - .exit = centrino_cpu_exit, - .verify = centrino_verify, - .target = centrino_target, - .get = get_cur_freq, - .attr = centrino_attr, - .owner = THIS_MODULE, -}; - - -/** - * centrino_init - initializes the Enhanced SpeedStep CPUFreq driver - * - * Initializes the Enhanced SpeedStep support. Returns -ENODEV on - * unsupported devices, -ENOENT if there's no voltage table for this - * particular CPU model, -EINVAL on problems during initiatization, - * and zero on success. - * - * This is quite picky. Not only does the CPU have to advertise the - * "est" flag in the cpuid capability flags, we look for a specific - * CPU model and stepping, and we need to have the exact model name in - * our voltage tables. That is, be paranoid about not releasing - * someone's valuable magic smoke. - */ -static int __init centrino_init(void) -{ - struct cpuinfo_x86 *cpu = &cpu_data(0); - - if (!cpu_has(cpu, X86_FEATURE_EST)) - return -ENODEV; - - return cpufreq_register_driver(¢rino_driver); -} - -static void __exit centrino_exit(void) -{ - cpufreq_unregister_driver(¢rino_driver); -} - -MODULE_AUTHOR ("Jeremy Fitzhardinge <jeremy@goop.org>"); -MODULE_DESCRIPTION ("Enhanced SpeedStep driver for Intel Pentium M processors."); -MODULE_LICENSE ("GPL"); - -late_initcall(centrino_init); -module_exit(centrino_exit); diff --git a/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c b/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c deleted file mode 100644 index 561758e95180..000000000000 --- a/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c +++ /dev/null @@ -1,452 +0,0 @@ -/* - * (C) 2001 Dave Jones, Arjan van de ven. - * (C) 2002 - 2003 Dominik Brodowski <linux@brodo.de> - * - * Licensed under the terms of the GNU GPL License version 2. - * Based upon reverse engineered information, and on Intel documentation - * for chipsets ICH2-M and ICH3-M. - * - * Many thanks to Ducrot Bruno for finding and fixing the last - * "missing link" for ICH2-M/ICH3-M support, and to Thomas Winkler - * for extensive testing. - * - * BIG FAT DISCLAIMER: Work in progress code. Possibly *dangerous* - */ - - -/********************************************************************* - * SPEEDSTEP - DEFINITIONS * - *********************************************************************/ - -#include <linux/kernel.h> -#include <linux/module.h> -#include <linux/init.h> -#include <linux/cpufreq.h> -#include <linux/pci.h> -#include <linux/sched.h> - -#include "speedstep-lib.h" - - -/* speedstep_chipset: - * It is necessary to know which chipset is used. As accesses to - * this device occur at various places in this module, we need a - * static struct pci_dev * pointing to that device. - */ -static struct pci_dev *speedstep_chipset_dev; - - -/* speedstep_processor - */ -static enum speedstep_processor speedstep_processor; - -static u32 pmbase; - -/* - * There are only two frequency states for each processor. Values - * are in kHz for the time being. - */ -static struct cpufreq_frequency_table speedstep_freqs[] = { - {SPEEDSTEP_HIGH, 0}, - {SPEEDSTEP_LOW, 0}, - {0, CPUFREQ_TABLE_END}, -}; - - -#define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, \ - "speedstep-ich", msg) - - -/** - * speedstep_find_register - read the PMBASE address - * - * Returns: -ENODEV if no register could be found - */ -static int speedstep_find_register(void) -{ - if (!speedstep_chipset_dev) - return -ENODEV; - - /* get PMBASE */ - pci_read_config_dword(speedstep_chipset_dev, 0x40, &pmbase); - if (!(pmbase & 0x01)) { - printk(KERN_ERR "speedstep-ich: could not find speedstep register\n"); - return -ENODEV; - } - - pmbase &= 0xFFFFFFFE; - if (!pmbase) { - printk(KERN_ERR "speedstep-ich: could not find speedstep register\n"); - return -ENODEV; - } - - dprintk("pmbase is 0x%x\n", pmbase); - return 0; -} - -/** - * speedstep_set_state - set the SpeedStep state - * @state: new processor frequency state (SPEEDSTEP_LOW or SPEEDSTEP_HIGH) - * - * Tries to change the SpeedStep state. Can be called from - * smp_call_function_single. - */ -static void speedstep_set_state(unsigned int state) -{ - u8 pm2_blk; - u8 value; - unsigned long flags; - - if (state > 0x1) - return; - - /* Disable IRQs */ - local_irq_save(flags); - - /* read state */ - value = inb(pmbase + 0x50); - - dprintk("read at pmbase 0x%x + 0x50 returned 0x%x\n", pmbase, value); - - /* write new state */ - value &= 0xFE; - value |= state; - - dprintk("writing 0x%x to pmbase 0x%x + 0x50\n", value, pmbase); - - /* Disable bus master arbitration */ - pm2_blk = inb(pmbase + 0x20); - pm2_blk |= 0x01; - outb(pm2_blk, (pmbase + 0x20)); - - /* Actual transition */ - outb(value, (pmbase + 0x50)); - - /* Restore bus master arbitration */ - pm2_blk &= 0xfe; - outb(pm2_blk, (pmbase + 0x20)); - - /* check if transition was successful */ - value = inb(pmbase + 0x50); - - /* Enable IRQs */ - local_irq_restore(flags); - - dprintk("read at pmbase 0x%x + 0x50 returned 0x%x\n", pmbase, value); - - if (state == (value & 0x1)) - dprintk("change to %u MHz succeeded\n", - speedstep_get_frequency(speedstep_processor) / 1000); - else - printk(KERN_ERR "cpufreq: change failed - I/O error\n"); - - 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 - * - * Tries to activate the SpeedStep status and control registers. - * Returns -EINVAL on an unsupported chipset, and zero on success. - */ -static int speedstep_activate(void) -{ - u16 value = 0; - - if (!speedstep_chipset_dev) - return -EINVAL; - - pci_read_config_word(speedstep_chipset_dev, 0x00A0, &value); - if (!(value & 0x08)) { - value |= 0x08; - dprintk("activating SpeedStep (TM) registers\n"); - pci_write_config_word(speedstep_chipset_dev, 0x00A0, value); - } - - return 0; -} - - -/** - * speedstep_detect_chipset - detect the Southbridge which contains SpeedStep logic - * - * Detects ICH2-M, ICH3-M and ICH4-M so far. The pci_dev points to - * the LPC bridge / PM module which contains all power-management - * functions. Returns the SPEEDSTEP_CHIPSET_-number for the detected - * chipset, or zero on failure. - */ -static unsigned int speedstep_detect_chipset(void) -{ - speedstep_chipset_dev = pci_get_subsys(PCI_VENDOR_ID_INTEL, - PCI_DEVICE_ID_INTEL_82801DB_12, - PCI_ANY_ID, PCI_ANY_ID, - NULL); - if (speedstep_chipset_dev) - return 4; /* 4-M */ - - speedstep_chipset_dev = pci_get_subsys(PCI_VENDOR_ID_INTEL, - PCI_DEVICE_ID_INTEL_82801CA_12, - PCI_ANY_ID, PCI_ANY_ID, - NULL); - if (speedstep_chipset_dev) - return 3; /* 3-M */ - - - speedstep_chipset_dev = pci_get_subsys(PCI_VENDOR_ID_INTEL, - PCI_DEVICE_ID_INTEL_82801BA_10, - PCI_ANY_ID, PCI_ANY_ID, - NULL); - if (speedstep_chipset_dev) { - /* speedstep.c causes lockups on Dell Inspirons 8000 and - * 8100 which use a pretty old revision of the 82815 - * host brige. Abort on these systems. - */ - static struct pci_dev *hostbridge; - - hostbridge = pci_get_subsys(PCI_VENDOR_ID_INTEL, - PCI_DEVICE_ID_INTEL_82815_MC, - PCI_ANY_ID, PCI_ANY_ID, - NULL); - - if (!hostbridge) - return 2; /* 2-M */ - - if (hostbridge->revision < 5) { - dprintk("hostbridge does not support speedstep\n"); - speedstep_chipset_dev = NULL; - pci_dev_put(hostbridge); - return 0; - } - - pci_dev_put(hostbridge); - return 2; /* 2-M */ - } - - return 0; -} - -static void get_freq_data(void *_speed) -{ - unsigned int *speed = _speed; - - *speed = speedstep_get_frequency(speedstep_processor); -} - -static unsigned int speedstep_get(unsigned int cpu) -{ - unsigned int speed; - - /* You're supposed to ensure CPU is online. */ - if (smp_call_function_single(cpu, get_freq_data, &speed, 1) != 0) - BUG(); - - dprintk("detected %u kHz as current frequency\n", speed); - return speed; -} - -/** - * speedstep_target - set a new CPUFreq policy - * @policy: new policy - * @target_freq: the target frequency - * @relation: how that frequency relates to achieved frequency - * (CPUFREQ_RELATION_L or CPUFREQ_RELATION_H) - * - * Sets a new CPUFreq policy. - */ -static int speedstep_target(struct cpufreq_policy *policy, - unsigned int target_freq, - unsigned int relation) -{ - unsigned int newstate = 0, policy_cpu; - struct cpufreq_freqs freqs; - int i; - - if (cpufreq_frequency_table_target(policy, &speedstep_freqs[0], - target_freq, relation, &newstate)) - return -EINVAL; - - 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; - - dprintk("transiting from %u to %u kHz\n", freqs.old, freqs.new); - - /* no transition necessary */ - if (freqs.old == freqs.new) - return 0; - - for_each_cpu(i, policy->cpus) { - freqs.cpu = i; - cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); - } - - smp_call_function_single(policy_cpu, _speedstep_set_state, &newstate, - true); - - for_each_cpu(i, policy->cpus) { - freqs.cpu = i; - cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); - } - - return 0; -} - - -/** - * speedstep_verify - verifies a new CPUFreq policy - * @policy: new policy - * - * Limit must be within speedstep_low_freq and speedstep_high_freq, with - * at least one border included. - */ -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; - 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 - policy_cpu = cpumask_any_and(policy->cpus, cpu_online_mask); - - /* detect low and high frequency and transition latency */ - 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_cpu); - if (!speed) - return -EIO; - - dprintk("currently at %s speed setting - %i MHz\n", - (speed == speedstep_freqs[SPEEDSTEP_LOW].frequency) - ? "low" : "high", - (speed / 1000)); - - /* cpuinfo and default policy values */ - policy->cur = speed; - - result = cpufreq_frequency_table_cpuinfo(policy, speedstep_freqs); - if (result) - return result; - - cpufreq_frequency_table_get_attr(speedstep_freqs, policy->cpu); - - return 0; -} - - -static int speedstep_cpu_exit(struct cpufreq_policy *policy) -{ - cpufreq_frequency_table_put_attr(policy->cpu); - return 0; -} - -static struct freq_attr *speedstep_attr[] = { - &cpufreq_freq_attr_scaling_available_freqs, - NULL, -}; - - -static struct cpufreq_driver speedstep_driver = { - .name = "speedstep-ich", - .verify = speedstep_verify, - .target = speedstep_target, - .init = speedstep_cpu_init, - .exit = speedstep_cpu_exit, - .get = speedstep_get, - .owner = THIS_MODULE, - .attr = speedstep_attr, -}; - - -/** - * speedstep_init - initializes the SpeedStep CPUFreq driver - * - * Initializes the SpeedStep support. Returns -ENODEV on unsupported - * devices, -EINVAL on problems during initiatization, and zero on - * success. - */ -static int __init speedstep_init(void) -{ - /* detect processor */ - speedstep_processor = speedstep_detect_processor(); - if (!speedstep_processor) { - dprintk("Intel(R) SpeedStep(TM) capable processor " - "not found\n"); - return -ENODEV; - } - - /* detect chipset */ - if (!speedstep_detect_chipset()) { - dprintk("Intel(R) SpeedStep(TM) for this chipset not " - "(yet) available.\n"); - return -ENODEV; - } - - /* activate speedstep support */ - if (speedstep_activate()) { - pci_dev_put(speedstep_chipset_dev); - return -EINVAL; - } - - if (speedstep_find_register()) - return -ENODEV; - - return cpufreq_register_driver(&speedstep_driver); -} - - -/** - * speedstep_exit - unregisters SpeedStep support - * - * Unregisters SpeedStep support. - */ -static void __exit speedstep_exit(void) -{ - pci_dev_put(speedstep_chipset_dev); - cpufreq_unregister_driver(&speedstep_driver); -} - - -MODULE_AUTHOR("Dave Jones <davej@redhat.com>, " - "Dominik Brodowski <linux@brodo.de>"); -MODULE_DESCRIPTION("Speedstep driver for Intel mobile processors on chipsets " - "with ICH-M southbridges."); -MODULE_LICENSE("GPL"); - -module_init(speedstep_init); -module_exit(speedstep_exit); diff --git a/arch/x86/kernel/cpu/cpufreq/speedstep-lib.c b/arch/x86/kernel/cpu/cpufreq/speedstep-lib.c deleted file mode 100644 index a94ec6be69fa..000000000000 --- a/arch/x86/kernel/cpu/cpufreq/speedstep-lib.c +++ /dev/null @@ -1,481 +0,0 @@ -/* - * (C) 2002 - 2003 Dominik Brodowski <linux@brodo.de> - * - * Licensed under the terms of the GNU GPL License version 2. - * - * Library for common functions for Intel SpeedStep v.1 and v.2 support - * - * BIG FAT DISCLAIMER: Work in progress code. Possibly *dangerous* - */ - -#include <linux/kernel.h> -#include <linux/module.h> -#include <linux/moduleparam.h> -#include <linux/init.h> -#include <linux/cpufreq.h> - -#include <asm/msr.h> -#include <asm/tsc.h> -#include "speedstep-lib.h" - -#define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, \ - "speedstep-lib", msg) - -#define PFX "speedstep-lib: " - -#ifdef CONFIG_X86_SPEEDSTEP_RELAXED_CAP_CHECK -static int relaxed_check; -#else -#define relaxed_check 0 -#endif - -/********************************************************************* - * GET PROCESSOR CORE SPEED IN KHZ * - *********************************************************************/ - -static unsigned int pentium3_get_frequency(enum speedstep_processor processor) -{ - /* See table 14 of p3_ds.pdf and table 22 of 29834003.pdf */ - struct { - unsigned int ratio; /* Frequency Multiplier (x10) */ - u8 bitmap; /* power on configuration bits - [27, 25:22] (in MSR 0x2a) */ - } msr_decode_mult[] = { - { 30, 0x01 }, - { 35, 0x05 }, - { 40, 0x02 }, - { 45, 0x06 }, - { 50, 0x00 }, - { 55, 0x04 }, - { 60, 0x0b }, - { 65, 0x0f }, - { 70, 0x09 }, - { 75, 0x0d }, - { 80, 0x0a }, - { 85, 0x26 }, - { 90, 0x20 }, - { 100, 0x2b }, - { 0, 0xff } /* error or unknown value */ - }; - - /* PIII(-M) FSB settings: see table b1-b of 24547206.pdf */ - struct { - unsigned int value; /* Front Side Bus speed in MHz */ - u8 bitmap; /* power on configuration bits [18: 19] - (in MSR 0x2a) */ - } msr_decode_fsb[] = { - { 66, 0x0 }, - { 100, 0x2 }, - { 133, 0x1 }, - { 0, 0xff} - }; - - u32 msr_lo, msr_tmp; - int i = 0, j = 0; - - /* read MSR 0x2a - we only need the low 32 bits */ - rdmsr(MSR_IA32_EBL_CR_POWERON, msr_lo, msr_tmp); - dprintk("P3 - MSR_IA32_EBL_CR_POWERON: 0x%x 0x%x\n", msr_lo, msr_tmp); - msr_tmp = msr_lo; - - /* decode the FSB */ - msr_tmp &= 0x00c0000; - msr_tmp >>= 18; - while (msr_tmp != msr_decode_fsb[i].bitmap) { - if (msr_decode_fsb[i].bitmap == 0xff) - return 0; - i++; - } - - /* decode the multiplier */ - if (processor == SPEEDSTEP_CPU_PIII_C_EARLY) { - dprintk("workaround for early PIIIs\n"); - msr_lo &= 0x03c00000; - } else - msr_lo &= 0x0bc00000; - msr_lo >>= 22; - while (msr_lo != msr_decode_mult[j].bitmap) { - if (msr_decode_mult[j].bitmap == 0xff) - return 0; - j++; - } - - dprintk("speed is %u\n", - (msr_decode_mult[j].ratio * msr_decode_fsb[i].value * 100)); - - return msr_decode_mult[j].ratio * msr_decode_fsb[i].value * 100; -} - - -static unsigned int pentiumM_get_frequency(void) -{ - u32 msr_lo, msr_tmp; - - rdmsr(MSR_IA32_EBL_CR_POWERON, msr_lo, msr_tmp); - dprintk("PM - MSR_IA32_EBL_CR_POWERON: 0x%x 0x%x\n", msr_lo, msr_tmp); - - /* see table B-2 of 24547212.pdf */ - if (msr_lo & 0x00040000) { - printk(KERN_DEBUG PFX "PM - invalid FSB: 0x%x 0x%x\n", - msr_lo, msr_tmp); - return 0; - } - - msr_tmp = (msr_lo >> 22) & 0x1f; - dprintk("bits 22-26 are 0x%x, speed is %u\n", - msr_tmp, (msr_tmp * 100 * 1000)); - - return msr_tmp * 100 * 1000; -} - -static unsigned int pentium_core_get_frequency(void) -{ - u32 fsb = 0; - u32 msr_lo, msr_tmp; - int ret; - - rdmsr(MSR_FSB_FREQ, msr_lo, msr_tmp); - /* see table B-2 of 25366920.pdf */ - switch (msr_lo & 0x07) { - case 5: - fsb = 100000; - break; - case 1: - fsb = 133333; - break; - case 3: - fsb = 166667; - break; - case 2: - fsb = 200000; - break; - case 0: - fsb = 266667; - break; - case 4: - fsb = 333333; - break; - default: - printk(KERN_ERR "PCORE - MSR_FSB_FREQ undefined value"); - } - - rdmsr(MSR_IA32_EBL_CR_POWERON, msr_lo, msr_tmp); - dprintk("PCORE - MSR_IA32_EBL_CR_POWERON: 0x%x 0x%x\n", - msr_lo, msr_tmp); - - msr_tmp = (msr_lo >> 22) & 0x1f; - dprintk("bits 22-26 are 0x%x, speed is %u\n", - msr_tmp, (msr_tmp * fsb)); - - ret = (msr_tmp * fsb); - return ret; -} - - -static unsigned int pentium4_get_frequency(void) -{ - struct cpuinfo_x86 *c = &boot_cpu_data; - u32 msr_lo, msr_hi, mult; - unsigned int fsb = 0; - unsigned int ret; - u8 fsb_code; - - /* Pentium 4 Model 0 and 1 do not have the Core Clock Frequency - * to System Bus Frequency Ratio Field in the Processor Frequency - * Configuration Register of the MSR. Therefore the current - * frequency cannot be calculated and has to be measured. - */ - if (c->x86_model < 2) - return cpu_khz; - - rdmsr(0x2c, msr_lo, msr_hi); - - dprintk("P4 - MSR_EBC_FREQUENCY_ID: 0x%x 0x%x\n", msr_lo, msr_hi); - - /* decode the FSB: see IA-32 Intel (C) Architecture Software - * Developer's Manual, Volume 3: System Prgramming Guide, - * revision #12 in Table B-1: MSRs in the Pentium 4 and - * Intel Xeon Processors, on page B-4 and B-5. - */ - fsb_code = (msr_lo >> 16) & 0x7; - switch (fsb_code) { - case 0: - fsb = 100 * 1000; - break; - case 1: - fsb = 13333 * 10; - break; - case 2: - fsb = 200 * 1000; - break; - } - - if (!fsb) - printk(KERN_DEBUG PFX "couldn't detect FSB speed. " - "Please send an e-mail to <linux@brodo.de>\n"); - - /* Multiplier. */ - mult = msr_lo >> 24; - - dprintk("P4 - FSB %u kHz; Multiplier %u; Speed %u kHz\n", - fsb, mult, (fsb * mult)); - - ret = (fsb * mult); - return ret; -} - - -/* Warning: may get called from smp_call_function_single. */ -unsigned int speedstep_get_frequency(enum speedstep_processor processor) -{ - switch (processor) { - case SPEEDSTEP_CPU_PCORE: - return pentium_core_get_frequency(); - case SPEEDSTEP_CPU_PM: - return pentiumM_get_frequency(); - case SPEEDSTEP_CPU_P4D: - case SPEEDSTEP_CPU_P4M: - return pentium4_get_frequency(); - case SPEEDSTEP_CPU_PIII_T: - case SPEEDSTEP_CPU_PIII_C: - case SPEEDSTEP_CPU_PIII_C_EARLY: - return pentium3_get_frequency(processor); - default: - return 0; - }; - return 0; -} -EXPORT_SYMBOL_GPL(speedstep_get_frequency); - - -/********************************************************************* - * DETECT SPEEDSTEP-CAPABLE PROCESSOR * - *********************************************************************/ - -unsigned int speedstep_detect_processor(void) -{ - struct cpuinfo_x86 *c = &cpu_data(0); - u32 ebx, msr_lo, msr_hi; - - dprintk("x86: %x, model: %x\n", c->x86, c->x86_model); - - if ((c->x86_vendor != X86_VENDOR_INTEL) || - ((c->x86 != 6) && (c->x86 != 0xF))) - return 0; - - if (c->x86 == 0xF) { - /* Intel Mobile Pentium 4-M - * or Intel Mobile Pentium 4 with 533 MHz FSB */ - if (c->x86_model != 2) - return 0; - - ebx = cpuid_ebx(0x00000001); - ebx &= 0x000000FF; - - dprintk("ebx value is %x, x86_mask is %x\n", ebx, c->x86_mask); - - switch (c->x86_mask) { - case 4: - /* - * B-stepping [M-P4-M] - * sample has ebx = 0x0f, production has 0x0e. - */ - if ((ebx == 0x0e) || (ebx == 0x0f)) - return SPEEDSTEP_CPU_P4M; - break; - case 7: - /* - * C-stepping [M-P4-M] - * needs to have ebx=0x0e, else it's a celeron: - * cf. 25130917.pdf / page 7, footnote 5 even - * though 25072120.pdf / page 7 doesn't say - * samples are only of B-stepping... - */ - if (ebx == 0x0e) - return SPEEDSTEP_CPU_P4M; - break; - case 9: - /* - * D-stepping [M-P4-M or M-P4/533] - * - * this is totally strange: CPUID 0x0F29 is - * used by M-P4-M, M-P4/533 and(!) Celeron CPUs. - * The latter need to be sorted out as they don't - * support speedstep. - * Celerons with CPUID 0x0F29 may have either - * ebx=0x8 or 0xf -- 25130917.pdf doesn't say anything - * specific. - * M-P4-Ms may have either ebx=0xe or 0xf [see above] - * M-P4/533 have either ebx=0xe or 0xf. [25317607.pdf] - * also, M-P4M HTs have ebx=0x8, too - * For now, they are distinguished by the model_id - * string - */ - if ((ebx == 0x0e) || - (strstr(c->x86_model_id, - "Mobile Intel(R) Pentium(R) 4") != NULL)) - return SPEEDSTEP_CPU_P4M; - break; - default: - break; - } - return 0; - } - - switch (c->x86_model) { - case 0x0B: /* Intel PIII [Tualatin] */ - /* cpuid_ebx(1) is 0x04 for desktop PIII, - * 0x06 for mobile PIII-M */ - ebx = cpuid_ebx(0x00000001); - dprintk("ebx is %x\n", ebx); - - ebx &= 0x000000FF; - - if (ebx != 0x06) - return 0; - - /* So far all PIII-M processors support SpeedStep. See - * Intel's 24540640.pdf of June 2003 - */ - return SPEEDSTEP_CPU_PIII_T; - - case 0x08: /* Intel PIII [Coppermine] */ - - /* all mobile PIII Coppermines have FSB 100 MHz - * ==> sort out a few desktop PIIIs. */ - rdmsr(MSR_IA32_EBL_CR_POWERON, msr_lo, msr_hi); - dprintk("Coppermine: MSR_IA32_EBL_CR_POWERON is 0x%x, 0x%x\n", - msr_lo, msr_hi); - msr_lo &= 0x00c0000; - if (msr_lo != 0x0080000) - return 0; - - /* - * If the processor is a mobile version, - * platform ID has bit 50 set - * it has SpeedStep technology if either - * bit 56 or 57 is set - */ - rdmsr(MSR_IA32_PLATFORM_ID, msr_lo, msr_hi); - dprintk("Coppermine: MSR_IA32_PLATFORM ID is 0x%x, 0x%x\n", - msr_lo, msr_hi); - if ((msr_hi & (1<<18)) && - (relaxed_check ? 1 : (msr_hi & (3<<24)))) { - if (c->x86_mask == 0x01) { - dprintk("early PIII version\n"); - return SPEEDSTEP_CPU_PIII_C_EARLY; - } else - return SPEEDSTEP_CPU_PIII_C; - } - - default: - return 0; - } -} -EXPORT_SYMBOL_GPL(speedstep_detect_processor); - - -/********************************************************************* - * DETECT SPEEDSTEP SPEEDS * - *********************************************************************/ - -unsigned int speedstep_get_freqs(enum speedstep_processor processor, - unsigned int *low_speed, - unsigned int *high_speed, - unsigned int *transition_latency, - void (*set_state) (unsigned int state)) -{ - unsigned int prev_speed; - unsigned int ret = 0; - unsigned long flags; - struct timeval tv1, tv2; - - if ((!processor) || (!low_speed) || (!high_speed) || (!set_state)) - return -EINVAL; - - dprintk("trying to determine both speeds\n"); - - /* get current speed */ - prev_speed = speedstep_get_frequency(processor); - if (!prev_speed) - return -EIO; - - dprintk("previous speed is %u\n", prev_speed); - - local_irq_save(flags); - - /* switch to low state */ - set_state(SPEEDSTEP_LOW); - *low_speed = speedstep_get_frequency(processor); - if (!*low_speed) { - ret = -EIO; - goto out; - } - - dprintk("low speed is %u\n", *low_speed); - - /* start latency measurement */ - if (transition_latency) - do_gettimeofday(&tv1); - - /* switch to high state */ - set_state(SPEEDSTEP_HIGH); - - /* end latency measurement */ - if (transition_latency) - do_gettimeofday(&tv2); - - *high_speed = speedstep_get_frequency(processor); - if (!*high_speed) { - ret = -EIO; - goto out; - } - - dprintk("high speed is %u\n", *high_speed); - - if (*low_speed == *high_speed) { - ret = -ENODEV; - goto out; - } - - /* switch to previous state, if necessary */ - if (*high_speed != prev_speed) - set_state(SPEEDSTEP_LOW); - - if (transition_latency) { - *transition_latency = (tv2.tv_sec - tv1.tv_sec) * USEC_PER_SEC + - tv2.tv_usec - tv1.tv_usec; - dprintk("transition latency is %u uSec\n", *transition_latency); - - /* convert uSec to nSec and add 20% for safety reasons */ - *transition_latency *= 1200; - - /* check if the latency measurement is too high or too low - * and set it to a safe value (500uSec) in that case - */ - if (*transition_latency > 10000000 || - *transition_latency < 50000) { - printk(KERN_WARNING PFX "frequency transition " - "measured seems out of range (%u " - "nSec), falling back to a safe one of" - "%u nSec.\n", - *transition_latency, 500000); - *transition_latency = 500000; - } - } - -out: - local_irq_restore(flags); - return ret; -} -EXPORT_SYMBOL_GPL(speedstep_get_freqs); - -#ifdef CONFIG_X86_SPEEDSTEP_RELAXED_CAP_CHECK -module_param(relaxed_check, int, 0444); -MODULE_PARM_DESC(relaxed_check, - "Don't do all checks for speedstep capability."); -#endif - -MODULE_AUTHOR("Dominik Brodowski <linux@brodo.de>"); -MODULE_DESCRIPTION("Library for Intel SpeedStep 1 or 2 cpufreq drivers."); -MODULE_LICENSE("GPL"); diff --git a/arch/x86/kernel/cpu/cpufreq/speedstep-lib.h b/arch/x86/kernel/cpu/cpufreq/speedstep-lib.h deleted file mode 100644 index 70d9cea1219d..000000000000 --- a/arch/x86/kernel/cpu/cpufreq/speedstep-lib.h +++ /dev/null @@ -1,49 +0,0 @@ -/* - * (C) 2002 - 2003 Dominik Brodowski <linux@brodo.de> - * - * Licensed under the terms of the GNU GPL License version 2. - * - * Library for common functions for Intel SpeedStep v.1 and v.2 support - * - * BIG FAT DISCLAIMER: Work in progress code. Possibly *dangerous* - */ - - - -/* processors */ -enum speedstep_processor { - SPEEDSTEP_CPU_PIII_C_EARLY = 0x00000001, /* Coppermine core */ - SPEEDSTEP_CPU_PIII_C = 0x00000002, /* Coppermine core */ - SPEEDSTEP_CPU_PIII_T = 0x00000003, /* Tualatin core */ - SPEEDSTEP_CPU_P4M = 0x00000004, /* P4-M */ -/* the following processors are not speedstep-capable and are not auto-detected - * in speedstep_detect_processor(). However, their speed can be detected using - * the speedstep_get_frequency() call. */ - SPEEDSTEP_CPU_PM = 0xFFFFFF03, /* Pentium M */ - SPEEDSTEP_CPU_P4D = 0xFFFFFF04, /* desktop P4 */ - SPEEDSTEP_CPU_PCORE = 0xFFFFFF05, /* Core */ -}; - -/* speedstep states -- only two of them */ - -#define SPEEDSTEP_HIGH 0x00000000 -#define SPEEDSTEP_LOW 0x00000001 - - -/* detect a speedstep-capable processor */ -extern enum speedstep_processor speedstep_detect_processor(void); - -/* detect the current speed (in khz) of the processor */ -extern unsigned int speedstep_get_frequency(enum speedstep_processor processor); - - -/* detect the low and high speeds of the processor. The callback - * set_state"'s first argument is either SPEEDSTEP_HIGH or - * SPEEDSTEP_LOW; the second argument is zero so that no - * cpufreq_notify_transition calls are initiated. - */ -extern unsigned int speedstep_get_freqs(enum speedstep_processor processor, - unsigned int *low_speed, - unsigned int *high_speed, - unsigned int *transition_latency, - void (*set_state) (unsigned int state)); diff --git a/arch/x86/kernel/cpu/cpufreq/speedstep-smi.c b/arch/x86/kernel/cpu/cpufreq/speedstep-smi.c deleted file mode 100644 index 91bc25b67bc1..000000000000 --- a/arch/x86/kernel/cpu/cpufreq/speedstep-smi.c +++ /dev/null @@ -1,467 +0,0 @@ -/* - * Intel SpeedStep SMI driver. - * - * (C) 2003 Hiroshi Miura <miura@da-cha.org> - * - * Licensed under the terms of the GNU GPL License version 2. - * - */ - - -/********************************************************************* - * SPEEDSTEP - DEFINITIONS * - *********************************************************************/ - -#include <linux/kernel.h> -#include <linux/module.h> -#include <linux/moduleparam.h> -#include <linux/init.h> -#include <linux/cpufreq.h> -#include <linux/delay.h> -#include <linux/io.h> -#include <asm/ist.h> - -#include "speedstep-lib.h" - -/* speedstep system management interface port/command. - * - * These parameters are got from IST-SMI BIOS call. - * If user gives it, these are used. - * - */ -static int smi_port; -static int smi_cmd; -static unsigned int smi_sig; - -/* info about the processor */ -static enum speedstep_processor speedstep_processor; - -/* - * There are only two frequency states for each processor. Values - * are in kHz for the time being. - */ -static struct cpufreq_frequency_table speedstep_freqs[] = { - {SPEEDSTEP_HIGH, 0}, - {SPEEDSTEP_LOW, 0}, - {0, CPUFREQ_TABLE_END}, -}; - -#define GET_SPEEDSTEP_OWNER 0 -#define GET_SPEEDSTEP_STATE 1 -#define SET_SPEEDSTEP_STATE 2 -#define GET_SPEEDSTEP_FREQS 4 - -/* how often shall the SMI call be tried if it failed, e.g. because - * of DMA activity going on? */ -#define SMI_TRIES 5 - -#define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, \ - "speedstep-smi", msg) - -/** - * speedstep_smi_ownership - */ -static int speedstep_smi_ownership(void) -{ - u32 command, result, magic, dummy; - u32 function = GET_SPEEDSTEP_OWNER; - unsigned char magic_data[] = "Copyright (c) 1999 Intel Corporation"; - - command = (smi_sig & 0xffffff00) | (smi_cmd & 0xff); - magic = virt_to_phys(magic_data); - - dprintk("trying to obtain ownership with command %x at port %x\n", - command, smi_port); - - __asm__ __volatile__( - "push %%ebp\n" - "out %%al, (%%dx)\n" - "pop %%ebp\n" - : "=D" (result), - "=a" (dummy), "=b" (dummy), "=c" (dummy), "=d" (dummy), - "=S" (dummy) - : "a" (command), "b" (function), "c" (0), "d" (smi_port), - "D" (0), "S" (magic) - : "memory" - ); - - dprintk("result is %x\n", result); - - return result; -} - -/** - * speedstep_smi_get_freqs - get SpeedStep preferred & current freq. - * @low: the low frequency value is placed here - * @high: the high frequency value is placed here - * - * Only available on later SpeedStep-enabled systems, returns false results or - * even hangs [cf. bugme.osdl.org # 1422] on earlier systems. Empirical testing - * shows that the latter occurs if !(ist_info.event & 0xFFFF). - */ -static int speedstep_smi_get_freqs(unsigned int *low, unsigned int *high) -{ - u32 command, result = 0, edi, high_mhz, low_mhz, dummy; - u32 state = 0; - u32 function = GET_SPEEDSTEP_FREQS; - - if (!(ist_info.event & 0xFFFF)) { - dprintk("bug #1422 -- can't read freqs from BIOS\n"); - return -ENODEV; - } - - command = (smi_sig & 0xffffff00) | (smi_cmd & 0xff); - - dprintk("trying to determine frequencies with command %x at port %x\n", - command, smi_port); - - __asm__ __volatile__( - "push %%ebp\n" - "out %%al, (%%dx)\n" - "pop %%ebp" - : "=a" (result), - "=b" (high_mhz), - "=c" (low_mhz), - "=d" (state), "=D" (edi), "=S" (dummy) - : "a" (command), - "b" (function), - "c" (state), - "d" (smi_port), "S" (0), "D" (0) - ); - - dprintk("result %x, low_freq %u, high_freq %u\n", - result, low_mhz, high_mhz); - - /* abort if results are obviously incorrect... */ - if ((high_mhz + low_mhz) < 600) - return -EINVAL; - - *high = high_mhz * 1000; - *low = low_mhz * 1000; - - return result; -} - -/** - * speedstep_get_state - set the SpeedStep state - * @state: processor frequency state (SPEEDSTEP_LOW or SPEEDSTEP_HIGH) - * - */ -static int speedstep_get_state(void) -{ - u32 function = GET_SPEEDSTEP_STATE; - u32 result, state, edi, command, dummy; - - command = (smi_sig & 0xffffff00) | (smi_cmd & 0xff); - - dprintk("trying to determine current setting with command %x " - "at port %x\n", command, smi_port); - - __asm__ __volatile__( - "push %%ebp\n" - "out %%al, (%%dx)\n" - "pop %%ebp\n" - : "=a" (result), - "=b" (state), "=D" (edi), - "=c" (dummy), "=d" (dummy), "=S" (dummy) - : "a" (command), "b" (function), "c" (0), - "d" (smi_port), "S" (0), "D" (0) - ); - - dprintk("state is %x, result is %x\n", state, result); - - return state & 1; -} - - -/** - * speedstep_set_state - set the SpeedStep state - * @state: new processor frequency state (SPEEDSTEP_LOW or SPEEDSTEP_HIGH) - * - */ -static void speedstep_set_state(unsigned int state) -{ - unsigned int result = 0, command, new_state, dummy; - unsigned long flags; - unsigned int function = SET_SPEEDSTEP_STATE; - unsigned int retry = 0; - - if (state > 0x1) - return; - - /* Disable IRQs */ - local_irq_save(flags); - - command = (smi_sig & 0xffffff00) | (smi_cmd & 0xff); - - dprintk("trying to set frequency to state %u " - "with command %x at port %x\n", - state, command, smi_port); - - do { - if (retry) { - dprintk("retry %u, previous result %u, waiting...\n", - retry, result); - mdelay(retry * 50); - } - retry++; - __asm__ __volatile__( - "push %%ebp\n" - "out %%al, (%%dx)\n" - "pop %%ebp" - : "=b" (new_state), "=D" (result), - "=c" (dummy), "=a" (dummy), - "=d" (dummy), "=S" (dummy) - : "a" (command), "b" (function), "c" (state), - "d" (smi_port), "S" (0), "D" (0) - ); - } while ((new_state != state) && (retry <= SMI_TRIES)); - - /* enable IRQs */ - local_irq_restore(flags); - - if (new_state == state) - dprintk("change to %u MHz succeeded after %u tries " - "with result %u\n", - (speedstep_freqs[new_state].frequency / 1000), - retry, result); - else - printk(KERN_ERR "cpufreq: change to state %u " - "failed with new_state %u and result %u\n", - state, new_state, result); - - return; -} - - -/** - * speedstep_target - set a new CPUFreq policy - * @policy: new policy - * @target_freq: new freq - * @relation: - * - * Sets a new CPUFreq policy/freq. - */ -static int speedstep_target(struct cpufreq_policy *policy, - unsigned int target_freq, unsigned int relation) -{ - unsigned int newstate = 0; - struct cpufreq_freqs freqs; - - if (cpufreq_frequency_table_target(policy, &speedstep_freqs[0], - target_freq, relation, &newstate)) - return -EINVAL; - - freqs.old = speedstep_freqs[speedstep_get_state()].frequency; - freqs.new = speedstep_freqs[newstate].frequency; - freqs.cpu = 0; /* speedstep.c is UP only driver */ - - if (freqs.old == freqs.new) - return 0; - - cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); - speedstep_set_state(newstate); - cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); - - return 0; -} - - -/** - * speedstep_verify - verifies a new CPUFreq policy - * @policy: new policy - * - * Limit must be within speedstep_low_freq and speedstep_high_freq, with - * at least one border included. - */ -static int speedstep_verify(struct cpufreq_policy *policy) -{ - return cpufreq_frequency_table_verify(policy, &speedstep_freqs[0]); -} - - -static int speedstep_cpu_init(struct cpufreq_policy *policy) -{ - int result; - unsigned int speed, state; - unsigned int *low, *high; - - /* capability check */ - if (policy->cpu != 0) - return -ENODEV; - - result = speedstep_smi_ownership(); - if (result) { - dprintk("fails in acquiring ownership of a SMI interface.\n"); - return -EINVAL; - } - - /* detect low and high frequency */ - low = &speedstep_freqs[SPEEDSTEP_LOW].frequency; - high = &speedstep_freqs[SPEEDSTEP_HIGH].frequency; - - result = speedstep_smi_get_freqs(low, high); - if (result) { - /* fall back to speedstep_lib.c dection mechanism: - * try both states out */ - dprintk("could not detect low and high frequencies " - "by SMI call.\n"); - result = speedstep_get_freqs(speedstep_processor, - low, high, - NULL, - &speedstep_set_state); - - if (result) { - dprintk("could not detect two different speeds" - " -- aborting.\n"); - return result; - } else - dprintk("workaround worked.\n"); - } - - /* get current speed setting */ - state = speedstep_get_state(); - speed = speedstep_freqs[state].frequency; - - dprintk("currently at %s speed setting - %i MHz\n", - (speed == speedstep_freqs[SPEEDSTEP_LOW].frequency) - ? "low" : "high", - (speed / 1000)); - - /* cpuinfo and default policy values */ - policy->cpuinfo.transition_latency = CPUFREQ_ETERNAL; - policy->cur = speed; - - result = cpufreq_frequency_table_cpuinfo(policy, speedstep_freqs); - if (result) - return result; - - cpufreq_frequency_table_get_attr(speedstep_freqs, policy->cpu); - - return 0; -} - -static int speedstep_cpu_exit(struct cpufreq_policy *policy) -{ - cpufreq_frequency_table_put_attr(policy->cpu); - return 0; -} - -static unsigned int speedstep_get(unsigned int cpu) -{ - if (cpu) - return -ENODEV; - return speedstep_get_frequency(speedstep_processor); -} - - -static int speedstep_resume(struct cpufreq_policy *policy) -{ - int result = speedstep_smi_ownership(); - - if (result) - dprintk("fails in re-acquiring ownership of a SMI interface.\n"); - - return result; -} - -static struct freq_attr *speedstep_attr[] = { - &cpufreq_freq_attr_scaling_available_freqs, - NULL, -}; - -static struct cpufreq_driver speedstep_driver = { - .name = "speedstep-smi", - .verify = speedstep_verify, - .target = speedstep_target, - .init = speedstep_cpu_init, - .exit = speedstep_cpu_exit, - .get = speedstep_get, - .resume = speedstep_resume, - .owner = THIS_MODULE, - .attr = speedstep_attr, -}; - -/** - * speedstep_init - initializes the SpeedStep CPUFreq driver - * - * Initializes the SpeedStep support. Returns -ENODEV on unsupported - * BIOS, -EINVAL on problems during initiatization, and zero on - * success. - */ -static int __init speedstep_init(void) -{ - speedstep_processor = speedstep_detect_processor(); - - switch (speedstep_processor) { - case SPEEDSTEP_CPU_PIII_T: - case SPEEDSTEP_CPU_PIII_C: - case SPEEDSTEP_CPU_PIII_C_EARLY: - break; - default: - speedstep_processor = 0; - } - - if (!speedstep_processor) { - dprintk("No supported Intel CPU detected.\n"); - return -ENODEV; - } - - dprintk("signature:0x%.8lx, command:0x%.8lx, " - "event:0x%.8lx, perf_level:0x%.8lx.\n", - ist_info.signature, ist_info.command, - ist_info.event, ist_info.perf_level); - - /* Error if no IST-SMI BIOS or no PARM - sig= 'ISGE' aka 'Intel Speedstep Gate E' */ - if ((ist_info.signature != 0x47534943) && ( - (smi_port == 0) || (smi_cmd == 0))) - return -ENODEV; - - if (smi_sig == 1) - smi_sig = 0x47534943; - else - smi_sig = ist_info.signature; - - /* setup smi_port from MODLULE_PARM or BIOS */ - if ((smi_port > 0xff) || (smi_port < 0)) - return -EINVAL; - else if (smi_port == 0) - smi_port = ist_info.command & 0xff; - - if ((smi_cmd > 0xff) || (smi_cmd < 0)) - return -EINVAL; - else if (smi_cmd == 0) - smi_cmd = (ist_info.command >> 16) & 0xff; - - return cpufreq_register_driver(&speedstep_driver); -} - - -/** - * speedstep_exit - unregisters SpeedStep support - * - * Unregisters SpeedStep support. - */ -static void __exit speedstep_exit(void) -{ - cpufreq_unregister_driver(&speedstep_driver); -} - -module_param(smi_port, int, 0444); -module_param(smi_cmd, int, 0444); -module_param(smi_sig, uint, 0444); - -MODULE_PARM_DESC(smi_port, "Override the BIOS-given IST port with this value " - "-- Intel's default setting is 0xb2"); -MODULE_PARM_DESC(smi_cmd, "Override the BIOS-given IST command with this value " - "-- Intel's default setting is 0x82"); -MODULE_PARM_DESC(smi_sig, "Set to 1 to fake the IST signature when using the " - "SMI interface."); - -MODULE_AUTHOR("Hiroshi Miura"); -MODULE_DESCRIPTION("Speedstep driver for IST applet SMI interface."); -MODULE_LICENSE("GPL"); - -module_init(speedstep_init); -module_exit(speedstep_exit); diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c index df86bc8c859d..1edf5ba4fb2b 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -29,10 +29,10 @@ static void __cpuinit early_init_intel(struct cpuinfo_x86 *c) { + u64 misc_enable; + /* Unmask CPUID levels if masked: */ if (c->x86 > 6 || (c->x86 == 6 && c->x86_model >= 0xd)) { - u64 misc_enable; - rdmsrl(MSR_IA32_MISC_ENABLE, misc_enable); if (misc_enable & MSR_IA32_MISC_ENABLE_LIMIT_CPUID) { @@ -118,8 +118,6 @@ static void __cpuinit early_init_intel(struct cpuinfo_x86 *c) * (model 2) with the same problem. */ if (c->x86 == 15) { - u64 misc_enable; - rdmsrl(MSR_IA32_MISC_ENABLE, misc_enable); if (misc_enable & MSR_IA32_MISC_ENABLE_FAST_STRING) { @@ -130,6 +128,19 @@ static void __cpuinit early_init_intel(struct cpuinfo_x86 *c) } } #endif + + /* + * If fast string is not enabled in IA32_MISC_ENABLE for any reason, + * clear the fast string and enhanced fast string CPU capabilities. + */ + if (c->x86 > 6 || (c->x86 == 6 && c->x86_model >= 0xd)) { + rdmsrl(MSR_IA32_MISC_ENABLE, misc_enable); + if (!(misc_enable & MSR_IA32_MISC_ENABLE_FAST_STRING)) { + printk(KERN_INFO "Disabled fast string operations\n"); + setup_clear_cpu_cap(X86_FEATURE_REP_GOOD); + setup_clear_cpu_cap(X86_FEATURE_ERMS); + } + } } #ifdef CONFIG_X86_32 @@ -400,12 +411,10 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c) switch (c->x86_model) { case 5: - if (c->x86_mask == 0) { - if (l2 == 0) - p = "Celeron (Covington)"; - else if (l2 == 256) - p = "Mobile Pentium II (Dixon)"; - } + if (l2 == 0) + p = "Celeron (Covington)"; + else if (l2 == 256) + p = "Mobile Pentium II (Dixon)"; break; case 6: diff --git a/arch/x86/kernel/cpu/intel_cacheinfo.c b/arch/x86/kernel/cpu/intel_cacheinfo.c index 1ce1af2899df..c105c533ed94 100644 --- a/arch/x86/kernel/cpu/intel_cacheinfo.c +++ b/arch/x86/kernel/cpu/intel_cacheinfo.c @@ -327,7 +327,6 @@ static void __cpuinit amd_calc_l3_indices(struct amd_l3_cache *l3) l3->subcaches[2] = sc2 = !(val & BIT(8)) + !(val & BIT(9)); l3->subcaches[3] = sc3 = !(val & BIT(12)) + !(val & BIT(13)); - l3->indices = (max(max(max(sc0, sc1), sc2), sc3) << 10) - 1; l3->indices = (max(max3(sc0, sc1, sc2), sc3) << 10) - 1; } @@ -454,27 +453,16 @@ int amd_set_l3_disable_slot(struct amd_l3_cache *l3, int cpu, unsigned slot, { int ret = 0; -#define SUBCACHE_MASK (3UL << 20) -#define SUBCACHE_INDEX 0xfff - - /* - * check whether this slot is already used or - * the index is already disabled - */ + /* check if @slot is already used or the index is already disabled */ ret = amd_get_l3_disable_slot(l3, slot); if (ret >= 0) return -EINVAL; - /* - * check whether the other slot has disabled the - * same index already - */ - if (index == amd_get_l3_disable_slot(l3, !slot)) + if (index > l3->indices) return -EINVAL; - /* do not allow writes outside of allowed bits */ - if ((index & ~(SUBCACHE_MASK | SUBCACHE_INDEX)) || - ((index & SUBCACHE_INDEX) > l3->indices)) + /* check whether the other slot has disabled the same index already */ + if (index == amd_get_l3_disable_slot(l3, !slot)) return -EINVAL; amd_l3_disable_index(l3, cpu, slot, index); diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c index 3385ea26f684..ff1ae9b6464d 100644 --- a/arch/x86/kernel/cpu/mcheck/mce.c +++ b/arch/x86/kernel/cpu/mcheck/mce.c @@ -105,20 +105,6 @@ static int cpu_missing; ATOMIC_NOTIFIER_HEAD(x86_mce_decoder_chain); EXPORT_SYMBOL_GPL(x86_mce_decoder_chain); -static int default_decode_mce(struct notifier_block *nb, unsigned long val, - void *data) -{ - pr_emerg(HW_ERR "No human readable MCE decoding support on this CPU type.\n"); - pr_emerg(HW_ERR "Run the message through 'mcelog --ascii' to decode.\n"); - - return NOTIFY_STOP; -} - -static struct notifier_block mce_dec_nb = { - .notifier_call = default_decode_mce, - .priority = -1, -}; - /* MCA banks polled by the period polling timer for corrected events */ DEFINE_PER_CPU(mce_banks_t, mce_poll_banks) = { [0 ... BITS_TO_LONGS(MAX_NR_BANKS)-1] = ~0UL @@ -212,6 +198,8 @@ void mce_log(struct mce *mce) static void print_mce(struct mce *m) { + int ret = 0; + pr_emerg(HW_ERR "CPU %d: Machine Check Exception: %Lx Bank %d: %016Lx\n", m->extcpu, m->mcgstatus, m->bank, m->status); @@ -239,7 +227,11 @@ static void print_mce(struct mce *m) * Print out human-readable details about the MCE error, * (if the CPU has an implementation for that) */ - atomic_notifier_call_chain(&x86_mce_decoder_chain, 0, m); + ret = atomic_notifier_call_chain(&x86_mce_decoder_chain, 0, m); + if (ret == NOTIFY_STOP) + return; + + pr_emerg_ratelimited(HW_ERR "Run the above through 'mcelog --ascii'\n"); } #define PANIC_TIMEOUT 5 /* 5 seconds */ @@ -590,7 +582,6 @@ void machine_check_poll(enum mcp_flags flags, mce_banks_t *b) if (!(flags & MCP_DONTLOG) && !mce_dont_log_ce) { mce_log(&m); atomic_notifier_call_chain(&x86_mce_decoder_chain, 0, &m); - add_taint(TAINT_MACHINE_CHECK); } /* @@ -1722,8 +1713,6 @@ __setup("mce", mcheck_enable); int __init mcheck_init(void) { - atomic_notifier_chain_register(&x86_mce_decoder_chain, &mce_dec_nb); - mcheck_intel_therm_init(); return 0; diff --git a/arch/x86/kernel/cpu/mcheck/mce_amd.c b/arch/x86/kernel/cpu/mcheck/mce_amd.c index 167f97b5596e..bb0adad35143 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_amd.c +++ b/arch/x86/kernel/cpu/mcheck/mce_amd.c @@ -509,6 +509,7 @@ recurse: out_free: if (b) { kobject_put(&b->kobj); + list_del(&b->miscj); kfree(b); } return err; diff --git a/arch/x86/kernel/cpu/mcheck/therm_throt.c b/arch/x86/kernel/cpu/mcheck/therm_throt.c index 6f8c5e9da97f..27c625178bf1 100644 --- a/arch/x86/kernel/cpu/mcheck/therm_throt.c +++ b/arch/x86/kernel/cpu/mcheck/therm_throt.c @@ -187,8 +187,6 @@ static int therm_throt_process(bool new_event, int event, int level) this_cpu, level == CORE_LEVEL ? "Core" : "Package", state->count); - - add_taint(TAINT_MACHINE_CHECK); return 1; } if (old_event) { @@ -355,7 +353,6 @@ static void notify_thresholds(__u64 msr_val) static void intel_thermal_interrupt(void) { __u64 msr_val; - struct cpuinfo_x86 *c = &cpu_data(smp_processor_id()); rdmsrl(MSR_IA32_THERM_STATUS, msr_val); @@ -367,19 +364,19 @@ static void intel_thermal_interrupt(void) CORE_LEVEL) != 0) mce_log_therm_throt_event(CORE_THROTTLED | msr_val); - if (cpu_has(c, X86_FEATURE_PLN)) + if (this_cpu_has(X86_FEATURE_PLN)) if (therm_throt_process(msr_val & THERM_STATUS_POWER_LIMIT, POWER_LIMIT_EVENT, CORE_LEVEL) != 0) mce_log_therm_throt_event(CORE_POWER_LIMIT | msr_val); - if (cpu_has(c, X86_FEATURE_PTS)) { + if (this_cpu_has(X86_FEATURE_PTS)) { rdmsrl(MSR_IA32_PACKAGE_THERM_STATUS, msr_val); if (therm_throt_process(msr_val & PACKAGE_THERM_STATUS_PROCHOT, THERMAL_THROTTLING_EVENT, PACKAGE_LEVEL) != 0) mce_log_therm_throt_event(PACKAGE_THROTTLED | msr_val); - if (cpu_has(c, X86_FEATURE_PLN)) + if (this_cpu_has(X86_FEATURE_PLN)) if (therm_throt_process(msr_val & PACKAGE_THERM_STATUS_POWER_LIMIT, POWER_LIMIT_EVENT, @@ -393,7 +390,6 @@ static void unexpected_thermal_interrupt(void) { printk(KERN_ERR "CPU%d: Unexpected LVT thermal interrupt!\n", smp_processor_id()); - add_taint(TAINT_MACHINE_CHECK); } static void (*smp_thermal_vector)(void) = unexpected_thermal_interrupt; @@ -446,18 +442,20 @@ void intel_init_thermal(struct cpuinfo_x86 *c) */ rdmsr(MSR_IA32_MISC_ENABLE, l, h); + h = lvtthmr_init; /* * The initial value of thermal LVT entries on all APs always reads * 0x10000 because APs are woken up by BSP issuing INIT-SIPI-SIPI * sequence to them and LVT registers are reset to 0s except for * the mask bits which are set to 1s when APs receive INIT IPI. - * Always restore the value that BIOS has programmed on AP based on - * BSP's info we saved since BIOS is always setting the same value - * for all threads/cores + * If BIOS takes over the thermal interrupt and sets its interrupt + * delivery mode to SMI (not fixed), it restores the value that the + * BIOS has programmed on AP based on BSP's info we saved since BIOS + * is always setting the same value for all threads/cores. */ - apic_write(APIC_LVTTHMR, lvtthmr_init); + if ((h & APIC_DM_FIXED_MASK) != APIC_DM_FIXED) + apic_write(APIC_LVTTHMR, lvtthmr_init); - h = lvtthmr_init; if ((l & MSR_IA32_MISC_ENABLE_TM1) && (h & APIC_DM_SMI)) { printk(KERN_DEBUG diff --git a/arch/x86/kernel/cpu/perf_event.c b/arch/x86/kernel/cpu/perf_event.c index 632e5dc9c9c0..3a0338b4b179 100644 --- a/arch/x86/kernel/cpu/perf_event.c +++ b/arch/x86/kernel/cpu/perf_event.c @@ -31,6 +31,7 @@ #include <asm/nmi.h> #include <asm/compat.h> #include <asm/smp.h> +#include <asm/alternative.h> #if 0 #undef wrmsrl @@ -363,12 +364,18 @@ again: return new_raw_count; } -/* using X86_FEATURE_PERFCTR_CORE to later implement ALTERNATIVE() here */ static inline int x86_pmu_addr_offset(int index) { - if (boot_cpu_has(X86_FEATURE_PERFCTR_CORE)) - return index << 1; - return index; + int offset; + + /* offset = X86_FEATURE_PERFCTR_CORE ? index << 1 : index */ + alternative_io(ASM_NOP2, + "shll $1, %%eax", + X86_FEATURE_PERFCTR_CORE, + "=a" (offset), + "a" (index)); + + return offset; } static inline unsigned int x86_pmu_config_addr(int index) @@ -613,8 +620,8 @@ static int x86_setup_perfctr(struct perf_event *event) /* * Branch tracing: */ - if ((attr->config == PERF_COUNT_HW_BRANCH_INSTRUCTIONS) && - (hwc->sample_period == 1)) { + if (attr->config == PERF_COUNT_HW_BRANCH_INSTRUCTIONS && + !attr->freq && hwc->sample_period == 1) { /* BTS is not supported by this architecture. */ if (!x86_pmu.bts_active) return -EOPNOTSUPP; @@ -1288,6 +1295,16 @@ static int x86_pmu_handle_irq(struct pt_regs *regs) cpuc = &__get_cpu_var(cpu_hw_events); + /* + * Some chipsets need to unmask the LVTPC in a particular spot + * inside the nmi handler. As a result, the unmasking was pushed + * into all the nmi handlers. + * + * This generic handler doesn't seem to have any issues where the + * unmasking occurs so it was left at the top. + */ + apic_write(APIC_LVTPC, APIC_DM_NMI); + for (idx = 0; idx < x86_pmu.num_counters; idx++) { if (!test_bit(idx, cpuc->active_mask)) { /* @@ -1374,8 +1391,6 @@ perf_event_nmi_handler(struct notifier_block *self, return NOTIFY_DONE; } - apic_write(APIC_LVTPC, APIC_DM_NMI); - handled = x86_pmu.handle_irq(args->regs); if (!handled) return NOTIFY_DONE; @@ -1758,17 +1773,6 @@ static struct pmu pmu = { * callchain support */ -static void -backtrace_warning_symbol(void *data, char *msg, unsigned long symbol) -{ - /* Ignore warnings */ -} - -static void backtrace_warning(void *data, char *msg) -{ - /* Ignore warnings */ -} - static int backtrace_stack(void *data, char *name) { return 0; @@ -1782,8 +1786,6 @@ static void backtrace_address(void *data, unsigned long addr, int reliable) } static const struct stacktrace_ops backtrace_ops = { - .warning = backtrace_warning, - .warning_symbol = backtrace_warning_symbol, .stack = backtrace_stack, .address = backtrace_address, .walk_stack = print_context_stack_bp, diff --git a/arch/x86/kernel/cpu/perf_event_amd.c b/arch/x86/kernel/cpu/perf_event_amd.c index cf4e369cea67..fe29c1d2219e 100644 --- a/arch/x86/kernel/cpu/perf_event_amd.c +++ b/arch/x86/kernel/cpu/perf_event_amd.c @@ -96,12 +96,14 @@ static __initconst const u64 amd_hw_cache_event_ids */ static const u64 amd_perfmon_event_map[] = { - [PERF_COUNT_HW_CPU_CYCLES] = 0x0076, - [PERF_COUNT_HW_INSTRUCTIONS] = 0x00c0, - [PERF_COUNT_HW_CACHE_REFERENCES] = 0x0080, - [PERF_COUNT_HW_CACHE_MISSES] = 0x0081, - [PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = 0x00c2, - [PERF_COUNT_HW_BRANCH_MISSES] = 0x00c3, + [PERF_COUNT_HW_CPU_CYCLES] = 0x0076, + [PERF_COUNT_HW_INSTRUCTIONS] = 0x00c0, + [PERF_COUNT_HW_CACHE_REFERENCES] = 0x0080, + [PERF_COUNT_HW_CACHE_MISSES] = 0x0081, + [PERF_COUNT_HW_BRANCH_INSTRUCTIONS] = 0x00c2, + [PERF_COUNT_HW_BRANCH_MISSES] = 0x00c3, + [PERF_COUNT_HW_STALLED_CYCLES_FRONTEND] = 0x00d0, /* "Decoder empty" event */ + [PERF_COUNT_HW_STALLED_CYCLES_BACKEND] = 0x00d1, /* "Dispatch stalls" event */ }; static u64 amd_pmu_event_map(int hw_event) diff --git a/arch/x86/kernel/cpu/perf_event_intel.c b/arch/x86/kernel/cpu/perf_event_intel.c index 43fa20b13817..41178c826c48 100644 --- a/arch/x86/kernel/cpu/perf_event_intel.c +++ b/arch/x86/kernel/cpu/perf_event_intel.c @@ -25,7 +25,7 @@ struct intel_percore { /* * Intel PerfMon, used on Core and later. */ -static const u64 intel_perfmon_event_map[] = +static u64 intel_perfmon_event_map[PERF_COUNT_HW_MAX] __read_mostly = { [PERF_COUNT_HW_CPU_CYCLES] = 0x003c, [PERF_COUNT_HW_INSTRUCTIONS] = 0x00c0, @@ -36,7 +36,7 @@ static const u64 intel_perfmon_event_map[] = [PERF_COUNT_HW_BUS_CYCLES] = 0x013c, }; -static struct event_constraint intel_core_event_constraints[] = +static struct event_constraint intel_core_event_constraints[] __read_mostly = { INTEL_EVENT_CONSTRAINT(0x11, 0x2), /* FP_ASSIST */ INTEL_EVENT_CONSTRAINT(0x12, 0x2), /* MUL */ @@ -47,7 +47,7 @@ static struct event_constraint intel_core_event_constraints[] = EVENT_CONSTRAINT_END }; -static struct event_constraint intel_core2_event_constraints[] = +static struct event_constraint intel_core2_event_constraints[] __read_mostly = { FIXED_EVENT_CONSTRAINT(0x00c0, 0), /* INST_RETIRED.ANY */ FIXED_EVENT_CONSTRAINT(0x003c, 1), /* CPU_CLK_UNHALTED.CORE */ @@ -70,7 +70,7 @@ static struct event_constraint intel_core2_event_constraints[] = EVENT_CONSTRAINT_END }; -static struct event_constraint intel_nehalem_event_constraints[] = +static struct event_constraint intel_nehalem_event_constraints[] __read_mostly = { FIXED_EVENT_CONSTRAINT(0x00c0, 0), /* INST_RETIRED.ANY */ FIXED_EVENT_CONSTRAINT(0x003c, 1), /* CPU_CLK_UNHALTED.CORE */ @@ -86,19 +86,19 @@ static struct event_constraint intel_nehalem_event_constraints[] = EVENT_CONSTRAINT_END }; -static struct extra_reg intel_nehalem_extra_regs[] = +static struct extra_reg intel_nehalem_extra_regs[] __read_mostly = { INTEL_EVENT_EXTRA_REG(0xb7, MSR_OFFCORE_RSP_0, 0xffff), EVENT_EXTRA_END }; -static struct event_constraint intel_nehalem_percore_constraints[] = +static struct event_constraint intel_nehalem_percore_constraints[] __read_mostly = { INTEL_EVENT_CONSTRAINT(0xb7, 0), EVENT_CONSTRAINT_END }; -static struct event_constraint intel_westmere_event_constraints[] = +static struct event_constraint intel_westmere_event_constraints[] __read_mostly = { FIXED_EVENT_CONSTRAINT(0x00c0, 0), /* INST_RETIRED.ANY */ FIXED_EVENT_CONSTRAINT(0x003c, 1), /* CPU_CLK_UNHALTED.CORE */ @@ -110,7 +110,7 @@ static struct event_constraint intel_westmere_event_constraints[] = EVENT_CONSTRAINT_END }; -static struct event_constraint intel_snb_event_constraints[] = +static struct event_constraint intel_snb_event_constraints[] __read_mostly = { FIXED_EVENT_CONSTRAINT(0x00c0, 0), /* INST_RETIRED.ANY */ FIXED_EVENT_CONSTRAINT(0x003c, 1), /* CPU_CLK_UNHALTED.CORE */ @@ -123,21 +123,21 @@ static struct event_constraint intel_snb_event_constraints[] = EVENT_CONSTRAINT_END }; -static struct extra_reg intel_westmere_extra_regs[] = +static struct extra_reg intel_westmere_extra_regs[] __read_mostly = { INTEL_EVENT_EXTRA_REG(0xb7, MSR_OFFCORE_RSP_0, 0xffff), INTEL_EVENT_EXTRA_REG(0xbb, MSR_OFFCORE_RSP_1, 0xffff), EVENT_EXTRA_END }; -static struct event_constraint intel_westmere_percore_constraints[] = +static struct event_constraint intel_westmere_percore_constraints[] __read_mostly = { INTEL_EVENT_CONSTRAINT(0xb7, 0), INTEL_EVENT_CONSTRAINT(0xbb, 0), EVENT_CONSTRAINT_END }; -static struct event_constraint intel_gen_event_constraints[] = +static struct event_constraint intel_gen_event_constraints[] __read_mostly = { FIXED_EVENT_CONSTRAINT(0x00c0, 0), /* INST_RETIRED.ANY */ FIXED_EVENT_CONSTRAINT(0x003c, 1), /* CPU_CLK_UNHALTED.CORE */ @@ -184,26 +184,23 @@ static __initconst const u64 snb_hw_cache_event_ids }, }, [ C(LL ) ] = { - /* - * TBD: Need Off-core Response Performance Monitoring support - */ [ C(OP_READ) ] = { - /* OFFCORE_RESPONSE_0.ANY_DATA.LOCAL_CACHE */ + /* OFFCORE_RESPONSE.ANY_DATA.LOCAL_CACHE */ [ C(RESULT_ACCESS) ] = 0x01b7, - /* OFFCORE_RESPONSE_1.ANY_DATA.ANY_LLC_MISS */ - [ C(RESULT_MISS) ] = 0x01bb, + /* OFFCORE_RESPONSE.ANY_DATA.ANY_LLC_MISS */ + [ C(RESULT_MISS) ] = 0x01b7, }, [ C(OP_WRITE) ] = { - /* OFFCORE_RESPONSE_0.ANY_RFO.LOCAL_CACHE */ + /* OFFCORE_RESPONSE.ANY_RFO.LOCAL_CACHE */ [ C(RESULT_ACCESS) ] = 0x01b7, - /* OFFCORE_RESPONSE_1.ANY_RFO.ANY_LLC_MISS */ - [ C(RESULT_MISS) ] = 0x01bb, + /* OFFCORE_RESPONSE.ANY_RFO.ANY_LLC_MISS */ + [ C(RESULT_MISS) ] = 0x01b7, }, [ C(OP_PREFETCH) ] = { - /* OFFCORE_RESPONSE_0.PREFETCH.LOCAL_CACHE */ + /* OFFCORE_RESPONSE.PREFETCH.LOCAL_CACHE */ [ C(RESULT_ACCESS) ] = 0x01b7, - /* OFFCORE_RESPONSE_1.PREFETCH.ANY_LLC_MISS */ - [ C(RESULT_MISS) ] = 0x01bb, + /* OFFCORE_RESPONSE.PREFETCH.ANY_LLC_MISS */ + [ C(RESULT_MISS) ] = 0x01b7, }, }, [ C(DTLB) ] = { @@ -285,26 +282,26 @@ static __initconst const u64 westmere_hw_cache_event_ids }, [ C(LL ) ] = { [ C(OP_READ) ] = { - /* OFFCORE_RESPONSE_0.ANY_DATA.LOCAL_CACHE */ + /* OFFCORE_RESPONSE.ANY_DATA.LOCAL_CACHE */ [ C(RESULT_ACCESS) ] = 0x01b7, - /* OFFCORE_RESPONSE_1.ANY_DATA.ANY_LLC_MISS */ - [ C(RESULT_MISS) ] = 0x01bb, + /* OFFCORE_RESPONSE.ANY_DATA.ANY_LLC_MISS */ + [ C(RESULT_MISS) ] = 0x01b7, }, /* * Use RFO, not WRITEBACK, because a write miss would typically occur * on RFO. */ [ C(OP_WRITE) ] = { - /* OFFCORE_RESPONSE_1.ANY_RFO.LOCAL_CACHE */ - [ C(RESULT_ACCESS) ] = 0x01bb, - /* OFFCORE_RESPONSE_0.ANY_RFO.ANY_LLC_MISS */ + /* OFFCORE_RESPONSE.ANY_RFO.LOCAL_CACHE */ + [ C(RESULT_ACCESS) ] = 0x01b7, + /* OFFCORE_RESPONSE.ANY_RFO.ANY_LLC_MISS */ [ C(RESULT_MISS) ] = 0x01b7, }, [ C(OP_PREFETCH) ] = { - /* OFFCORE_RESPONSE_0.PREFETCH.LOCAL_CACHE */ + /* OFFCORE_RESPONSE.PREFETCH.LOCAL_CACHE */ [ C(RESULT_ACCESS) ] = 0x01b7, - /* OFFCORE_RESPONSE_1.PREFETCH.ANY_LLC_MISS */ - [ C(RESULT_MISS) ] = 0x01bb, + /* OFFCORE_RESPONSE.PREFETCH.ANY_LLC_MISS */ + [ C(RESULT_MISS) ] = 0x01b7, }, }, [ C(DTLB) ] = { @@ -352,16 +349,36 @@ static __initconst const u64 westmere_hw_cache_event_ids }; /* - * OFFCORE_RESPONSE MSR bits (subset), See IA32 SDM Vol 3 30.6.1.3 + * Nehalem/Westmere MSR_OFFCORE_RESPONSE bits; + * See IA32 SDM Vol 3B 30.6.1.3 */ -#define DMND_DATA_RD (1 << 0) -#define DMND_RFO (1 << 1) -#define DMND_WB (1 << 3) -#define PF_DATA_RD (1 << 4) -#define PF_DATA_RFO (1 << 5) -#define RESP_UNCORE_HIT (1 << 8) -#define RESP_MISS (0xf600) /* non uncore hit */ +#define NHM_DMND_DATA_RD (1 << 0) +#define NHM_DMND_RFO (1 << 1) +#define NHM_DMND_IFETCH (1 << 2) +#define NHM_DMND_WB (1 << 3) +#define NHM_PF_DATA_RD (1 << 4) +#define NHM_PF_DATA_RFO (1 << 5) +#define NHM_PF_IFETCH (1 << 6) +#define NHM_OFFCORE_OTHER (1 << 7) +#define NHM_UNCORE_HIT (1 << 8) +#define NHM_OTHER_CORE_HIT_SNP (1 << 9) +#define NHM_OTHER_CORE_HITM (1 << 10) + /* reserved */ +#define NHM_REMOTE_CACHE_FWD (1 << 12) +#define NHM_REMOTE_DRAM (1 << 13) +#define NHM_LOCAL_DRAM (1 << 14) +#define NHM_NON_DRAM (1 << 15) + +#define NHM_ALL_DRAM (NHM_REMOTE_DRAM|NHM_LOCAL_DRAM) + +#define NHM_DMND_READ (NHM_DMND_DATA_RD) +#define NHM_DMND_WRITE (NHM_DMND_RFO|NHM_DMND_WB) +#define NHM_DMND_PREFETCH (NHM_PF_DATA_RD|NHM_PF_DATA_RFO) + +#define NHM_L3_HIT (NHM_UNCORE_HIT|NHM_OTHER_CORE_HIT_SNP|NHM_OTHER_CORE_HITM) +#define NHM_L3_MISS (NHM_NON_DRAM|NHM_ALL_DRAM|NHM_REMOTE_CACHE_FWD) +#define NHM_L3_ACCESS (NHM_L3_HIT|NHM_L3_MISS) static __initconst const u64 nehalem_hw_cache_extra_regs [PERF_COUNT_HW_CACHE_MAX] @@ -370,16 +387,16 @@ static __initconst const u64 nehalem_hw_cache_extra_regs { [ C(LL ) ] = { [ C(OP_READ) ] = { - [ C(RESULT_ACCESS) ] = DMND_DATA_RD|RESP_UNCORE_HIT, - [ C(RESULT_MISS) ] = DMND_DATA_RD|RESP_MISS, + [ C(RESULT_ACCESS) ] = NHM_DMND_READ|NHM_L3_ACCESS, + [ C(RESULT_MISS) ] = NHM_DMND_READ|NHM_L3_MISS, }, [ C(OP_WRITE) ] = { - [ C(RESULT_ACCESS) ] = DMND_RFO|DMND_WB|RESP_UNCORE_HIT, - [ C(RESULT_MISS) ] = DMND_RFO|DMND_WB|RESP_MISS, + [ C(RESULT_ACCESS) ] = NHM_DMND_WRITE|NHM_L3_ACCESS, + [ C(RESULT_MISS) ] = NHM_DMND_WRITE|NHM_L3_MISS, }, [ C(OP_PREFETCH) ] = { - [ C(RESULT_ACCESS) ] = PF_DATA_RD|PF_DATA_RFO|RESP_UNCORE_HIT, - [ C(RESULT_MISS) ] = PF_DATA_RD|PF_DATA_RFO|RESP_MISS, + [ C(RESULT_ACCESS) ] = NHM_DMND_PREFETCH|NHM_L3_ACCESS, + [ C(RESULT_MISS) ] = NHM_DMND_PREFETCH|NHM_L3_MISS, }, } }; @@ -933,6 +950,16 @@ static int intel_pmu_handle_irq(struct pt_regs *regs) cpuc = &__get_cpu_var(cpu_hw_events); + /* + * Some chipsets need to unmask the LVTPC in a particular spot + * inside the nmi handler. As a result, the unmasking was pushed + * into all the nmi handlers. + * + * This handler doesn't seem to have any issues with the unmasking + * so it was left at the top. + */ + apic_write(APIC_LVTPC, APIC_DM_NMI); + intel_pmu_disable_all(); handled = intel_pmu_drain_bts_buffer(); status = intel_pmu_get_status(); @@ -998,6 +1025,9 @@ intel_bts_constraints(struct perf_event *event) struct hw_perf_event *hwc = &event->hw; unsigned int hw_event, bts_event; + if (event->attr.freq) + return NULL; + hw_event = hwc->config & INTEL_ARCH_EVENT_MASK; bts_event = x86_pmu.event_map(PERF_COUNT_HW_BRANCH_INSTRUCTIONS); @@ -1305,7 +1335,7 @@ static void intel_clovertown_quirks(void) * AJ106 could possibly be worked around by not allowing LBR * usage from PEBS, including the fixup. * AJ68 could possibly be worked around by always programming - * a pebs_event_reset[0] value and coping with the lost events. + * a pebs_event_reset[0] value and coping with the lost events. * * But taken together it might just make sense to not enable PEBS on * these chips. @@ -1409,6 +1439,23 @@ static __init int intel_pmu_init(void) x86_pmu.percore_constraints = intel_nehalem_percore_constraints; x86_pmu.enable_all = intel_pmu_nhm_enable_all; x86_pmu.extra_regs = intel_nehalem_extra_regs; + + /* UOPS_ISSUED.STALLED_CYCLES */ + intel_perfmon_event_map[PERF_COUNT_HW_STALLED_CYCLES_FRONTEND] = 0x180010e; + /* UOPS_EXECUTED.CORE_ACTIVE_CYCLES,c=1,i=1 */ + intel_perfmon_event_map[PERF_COUNT_HW_STALLED_CYCLES_BACKEND] = 0x1803fb1; + + if (ebx & 0x40) { + /* + * Erratum AAJ80 detected, we work it around by using + * the BR_MISP_EXEC.ANY event. This will over-count + * branch-misses, but it's still much better than the + * architectural event which is often completely bogus: + */ + intel_perfmon_event_map[PERF_COUNT_HW_BRANCH_MISSES] = 0x7f89; + + pr_cont("erratum AAJ80 worked around, "); + } pr_cont("Nehalem events, "); break; @@ -1438,6 +1485,12 @@ static __init int intel_pmu_init(void) x86_pmu.enable_all = intel_pmu_nhm_enable_all; x86_pmu.pebs_constraints = intel_westmere_pebs_event_constraints; x86_pmu.extra_regs = intel_westmere_extra_regs; + + /* UOPS_ISSUED.STALLED_CYCLES */ + intel_perfmon_event_map[PERF_COUNT_HW_STALLED_CYCLES_FRONTEND] = 0x180010e; + /* UOPS_EXECUTED.CORE_ACTIVE_CYCLES,c=1,i=1 */ + intel_perfmon_event_map[PERF_COUNT_HW_STALLED_CYCLES_BACKEND] = 0x1803fb1; + pr_cont("Westmere events, "); break; @@ -1449,6 +1502,12 @@ static __init int intel_pmu_init(void) x86_pmu.event_constraints = intel_snb_event_constraints; x86_pmu.pebs_constraints = intel_snb_pebs_events; + + /* UOPS_ISSUED.ANY,c=1,i=1 to count stall cycles */ + intel_perfmon_event_map[PERF_COUNT_HW_STALLED_CYCLES_FRONTEND] = 0x180010e; + /* UOPS_DISPATCHED.THREAD,c=1,i=1 to count stall cycles*/ + intel_perfmon_event_map[PERF_COUNT_HW_STALLED_CYCLES_BACKEND] = 0x18001b1; + pr_cont("SandyBridge events, "); break; diff --git a/arch/x86/kernel/cpu/perf_event_p4.c b/arch/x86/kernel/cpu/perf_event_p4.c index d1f77e2934a1..ead584fb6a7d 100644 --- a/arch/x86/kernel/cpu/perf_event_p4.c +++ b/arch/x86/kernel/cpu/perf_event_p4.c @@ -468,7 +468,7 @@ static struct p4_event_bind p4_event_bind_map[] = { .opcode = P4_OPCODE(P4_EVENT_MISPRED_BRANCH_RETIRED), .escr_msr = { MSR_P4_CRU_ESCR0, MSR_P4_CRU_ESCR1 }, .escr_emask = - P4_ESCR_EMASK_BIT(P4_EVENT_MISPRED_BRANCH_RETIRED, NBOGUS), + P4_ESCR_EMASK_BIT(P4_EVENT_MISPRED_BRANCH_RETIRED, NBOGUS), .cntr = { {12, 13, 16}, {14, 15, 17} }, }, [P4_EVENT_X87_ASSIST] = { @@ -912,8 +912,7 @@ static int p4_pmu_handle_irq(struct pt_regs *regs) int idx, handled = 0; u64 val; - data.addr = 0; - data.raw = NULL; + perf_sample_data_init(&data, 0); cpuc = &__get_cpu_var(cpu_hw_events); @@ -950,11 +949,20 @@ static int p4_pmu_handle_irq(struct pt_regs *regs) x86_pmu_stop(event, 0); } - if (handled) { - /* p4 quirk: unmask it again */ - apic_write(APIC_LVTPC, apic_read(APIC_LVTPC) & ~APIC_LVT_MASKED); + if (handled) inc_irq_stat(apic_perf_irqs); - } + + /* + * When dealing with the unmasking of the LVTPC on P4 perf hw, it has + * been observed that the OVF bit flag has to be cleared first _before_ + * the LVTPC can be unmasked. + * + * The reason is the NMI line will continue to be asserted while the OVF + * bit is set. This causes a second NMI to generate if the LVTPC is + * unmasked before the OVF bit is cleared, leading to unknown NMI + * messages. + */ + apic_write(APIC_LVTPC, APIC_DM_NMI); return handled; } @@ -1188,7 +1196,7 @@ static __init int p4_pmu_init(void) { unsigned int low, high; - /* If we get stripped -- indexig fails */ + /* If we get stripped -- indexing fails */ BUILD_BUG_ON(ARCH_P4_MAX_CCCR > X86_PMC_MAX_GENERIC); rdmsr(MSR_IA32_MISC_ENABLE, low, high); diff --git a/arch/x86/kernel/devicetree.c b/arch/x86/kernel/devicetree.c index 706a9fb46a58..690bc8461835 100644 --- a/arch/x86/kernel/devicetree.c +++ b/arch/x86/kernel/devicetree.c @@ -369,6 +369,7 @@ static struct of_ioapic_type of_ioapic_type[] = static int ioapic_xlate(struct irq_domain *id, const u32 *intspec, u32 intsize, u32 *out_hwirq, u32 *out_type) { + struct mp_ioapic_gsi *gsi_cfg; struct io_apic_irq_attr attr; struct of_ioapic_type *it; u32 line, idx, type; @@ -378,7 +379,8 @@ static int ioapic_xlate(struct irq_domain *id, const u32 *intspec, u32 intsize, line = *intspec; idx = (u32) id->priv; - *out_hwirq = line + mp_gsi_routing[idx].gsi_base; + gsi_cfg = mp_ioapic_gsi_routing(idx); + *out_hwirq = line + gsi_cfg->gsi_base; intspec++; type = *intspec; @@ -391,7 +393,7 @@ static int ioapic_xlate(struct irq_domain *id, const u32 *intspec, u32 intsize, set_io_apic_irq_attr(&attr, idx, line, it->trigger, it->polarity); - return io_apic_setup_irq_pin(*out_hwirq, cpu_to_node(0), &attr); + return io_apic_setup_irq_pin_once(*out_hwirq, cpu_to_node(0), &attr); } static void __init ioapic_add_ofnode(struct device_node *np) @@ -407,7 +409,7 @@ static void __init ioapic_add_ofnode(struct device_node *np) } for (i = 0; i < nr_ioapics; i++) { - if (r.start == mp_ioapics[i].apicaddr) { + if (r.start == mpc_ioapic_addr(i)) { struct irq_domain *id; id = kzalloc(sizeof(*id), GFP_KERNEL); diff --git a/arch/x86/kernel/dumpstack.c b/arch/x86/kernel/dumpstack.c index e2a3f0606da4..1aae78f775fc 100644 --- a/arch/x86/kernel/dumpstack.c +++ b/arch/x86/kernel/dumpstack.c @@ -135,20 +135,6 @@ print_context_stack_bp(struct thread_info *tinfo, } EXPORT_SYMBOL_GPL(print_context_stack_bp); - -static void -print_trace_warning_symbol(void *data, char *msg, unsigned long symbol) -{ - printk(data); - print_symbol(msg, symbol); - printk("\n"); -} - -static void print_trace_warning(void *data, char *msg) -{ - printk("%s%s\n", (char *)data, msg); -} - static int print_trace_stack(void *data, char *name) { printk("%s <%s> ", (char *)data, name); @@ -166,8 +152,6 @@ static void print_trace_address(void *data, unsigned long addr, int reliable) } static const struct stacktrace_ops print_trace_ops = { - .warning = print_trace_warning, - .warning_symbol = print_trace_warning_symbol, .stack = print_trace_stack, .address = print_trace_address, .walk_stack = print_context_stack, @@ -279,7 +263,6 @@ int __kprobes __die(const char *str, struct pt_regs *regs, long err) printk("DEBUG_PAGEALLOC"); #endif printk("\n"); - sysfs_printk_last_file(); if (notify_die(DIE_OOPS, str, regs, err, current->thread.trap_no, SIGSEGV) == NOTIFY_STOP) return 1; diff --git a/arch/x86/kernel/ftrace.c b/arch/x86/kernel/ftrace.c index a93742a57468..0ba15a6cc57e 100644 --- a/arch/x86/kernel/ftrace.c +++ b/arch/x86/kernel/ftrace.c @@ -260,9 +260,9 @@ do_ftrace_mod_code(unsigned long ip, void *new_code) return mod_code_status; } -static unsigned char *ftrace_nop_replace(void) +static const unsigned char *ftrace_nop_replace(void) { - return ideal_nop5; + return ideal_nops[NOP_ATOMIC5]; } static int diff --git a/arch/x86/kernel/head32.c b/arch/x86/kernel/head32.c index d6d6bb361931..3bb08509a7a1 100644 --- a/arch/x86/kernel/head32.c +++ b/arch/x86/kernel/head32.c @@ -23,7 +23,6 @@ static void __init i386_default_early_setup(void) { /* Initialize 32bit specific setup functions */ - x86_init.resources.probe_roms = probe_roms; x86_init.resources.reserve_resources = i386_reserve_resources; x86_init.mpparse.setup_ioapic_ids = setup_ioapic_ids_from_mpc; diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c index bfe8f729e086..6781765b3a0d 100644 --- a/arch/x86/kernel/hpet.c +++ b/arch/x86/kernel/hpet.c @@ -217,7 +217,7 @@ static void hpet_reserve_platform_timers(unsigned int id) { } /* * Common hpet info */ -static unsigned long hpet_period; +static unsigned long hpet_freq; static void hpet_legacy_set_mode(enum clock_event_mode mode, struct clock_event_device *evt); @@ -232,7 +232,6 @@ static struct clock_event_device hpet_clockevent = { .features = CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT, .set_mode = hpet_legacy_set_mode, .set_next_event = hpet_legacy_next_event, - .shift = 32, .irq = 0, .rating = 50, }; @@ -290,28 +289,12 @@ static void hpet_legacy_clockevent_register(void) hpet_enable_legacy_int(); /* - * The mult factor is defined as (include/linux/clockchips.h) - * mult/2^shift = cyc/ns (in contrast to ns/cyc in clocksource.h) - * hpet_period is in units of femtoseconds (per cycle), so - * mult/2^shift = cyc/ns = 10^6/hpet_period - * mult = (10^6 * 2^shift)/hpet_period - * mult = (FSEC_PER_NSEC << hpet_clockevent.shift)/hpet_period - */ - hpet_clockevent.mult = div_sc((unsigned long) FSEC_PER_NSEC, - hpet_period, hpet_clockevent.shift); - /* Calculate the min / max delta */ - hpet_clockevent.max_delta_ns = clockevent_delta2ns(0x7FFFFFFF, - &hpet_clockevent); - /* Setup minimum reprogramming delta. */ - hpet_clockevent.min_delta_ns = clockevent_delta2ns(HPET_MIN_PROG_DELTA, - &hpet_clockevent); - - /* * Start hpet with the boot cpu mask and make it * global after the IO_APIC has been initialized. */ hpet_clockevent.cpumask = cpumask_of(smp_processor_id()); - clockevents_register_device(&hpet_clockevent); + clockevents_config_and_register(&hpet_clockevent, hpet_freq, + HPET_MIN_PROG_DELTA, 0x7FFFFFFF); global_clock_event = &hpet_clockevent; printk(KERN_DEBUG "hpet clockevent registered\n"); } @@ -549,7 +532,6 @@ static int hpet_setup_irq(struct hpet_dev *dev) static void init_one_hpet_msi_clockevent(struct hpet_dev *hdev, int cpu) { struct clock_event_device *evt = &hdev->evt; - uint64_t hpet_freq; WARN_ON(cpu != smp_processor_id()); if (!(hdev->flags & HPET_DEV_VALID)) @@ -571,24 +553,10 @@ static void init_one_hpet_msi_clockevent(struct hpet_dev *hdev, int cpu) evt->set_mode = hpet_msi_set_mode; evt->set_next_event = hpet_msi_next_event; - evt->shift = 32; - - /* - * The period is a femto seconds value. We need to calculate the - * scaled math multiplication factor for nanosecond to hpet tick - * conversion. - */ - hpet_freq = FSEC_PER_SEC; - do_div(hpet_freq, hpet_period); - evt->mult = div_sc((unsigned long) hpet_freq, - NSEC_PER_SEC, evt->shift); - /* Calculate the max delta */ - evt->max_delta_ns = clockevent_delta2ns(0x7FFFFFFF, evt); - /* 5 usec minimum reprogramming delta. */ - evt->min_delta_ns = 5000; - evt->cpumask = cpumask_of(hdev->cpu); - clockevents_register_device(evt); + + clockevents_config_and_register(evt, hpet_freq, HPET_MIN_PROG_DELTA, + 0x7FFFFFFF); } #ifdef CONFIG_HPET @@ -792,7 +760,6 @@ static struct clocksource clocksource_hpet = { static int hpet_clocksource_register(void) { u64 start, now; - u64 hpet_freq; cycle_t t1; /* Start the counter */ @@ -819,24 +786,7 @@ static int hpet_clocksource_register(void) return -ENODEV; } - /* - * The definition of mult is (include/linux/clocksource.h) - * mult/2^shift = ns/cyc and hpet_period is in units of fsec/cyc - * so we first need to convert hpet_period to ns/cyc units: - * mult/2^shift = ns/cyc = hpet_period/10^6 - * mult = (hpet_period * 2^shift)/10^6 - * mult = (hpet_period << shift)/FSEC_PER_NSEC - */ - - /* Need to convert hpet_period (fsec/cyc) to cyc/sec: - * - * cyc/sec = FSEC_PER_SEC/hpet_period(fsec/cyc) - * cyc/sec = (FSEC_PER_NSEC * NSEC_PER_SEC)/hpet_period - */ - hpet_freq = FSEC_PER_SEC; - do_div(hpet_freq, hpet_period); clocksource_register_hz(&clocksource_hpet, (u32)hpet_freq); - return 0; } @@ -845,7 +795,9 @@ static int hpet_clocksource_register(void) */ int __init hpet_enable(void) { + unsigned long hpet_period; unsigned int id; + u64 freq; int i; if (!is_hpet_capable()) @@ -884,6 +836,14 @@ int __init hpet_enable(void) goto out_nohpet; /* + * The period is a femto seconds value. Convert it to a + * frequency. + */ + freq = FSEC_PER_SEC; + do_div(freq, hpet_period); + hpet_freq = freq; + + /* * Read the HPET ID register to retrieve the IRQ routing * information and the number of channels */ diff --git a/arch/x86/kernel/i8253.c b/arch/x86/kernel/i8253.c index 2dfd31597443..fb66dc9e36cb 100644 --- a/arch/x86/kernel/i8253.c +++ b/arch/x86/kernel/i8253.c @@ -93,7 +93,6 @@ static struct clock_event_device pit_ce = { .features = CLOCK_EVT_FEAT_PERIODIC | CLOCK_EVT_FEAT_ONESHOT, .set_mode = init_pit_timer, .set_next_event = pit_next_event, - .shift = 32, .irq = 0, }; @@ -108,90 +107,12 @@ void __init setup_pit_timer(void) * IO_APIC has been initialized. */ pit_ce.cpumask = cpumask_of(smp_processor_id()); - pit_ce.mult = div_sc(CLOCK_TICK_RATE, NSEC_PER_SEC, pit_ce.shift); - pit_ce.max_delta_ns = clockevent_delta2ns(0x7FFF, &pit_ce); - pit_ce.min_delta_ns = clockevent_delta2ns(0xF, &pit_ce); - clockevents_register_device(&pit_ce); + clockevents_config_and_register(&pit_ce, CLOCK_TICK_RATE, 0xF, 0x7FFF); global_clock_event = &pit_ce; } #ifndef CONFIG_X86_64 -/* - * Since the PIT overflows every tick, its not very useful - * to just read by itself. So use jiffies to emulate a free - * running counter: - */ -static cycle_t pit_read(struct clocksource *cs) -{ - static int old_count; - static u32 old_jifs; - unsigned long flags; - int count; - u32 jifs; - - raw_spin_lock_irqsave(&i8253_lock, flags); - /* - * Although our caller may have the read side of xtime_lock, - * this is now a seqlock, and we are cheating in this routine - * by having side effects on state that we cannot undo if - * there is a collision on the seqlock and our caller has to - * retry. (Namely, old_jifs and old_count.) So we must treat - * jiffies as volatile despite the lock. We read jiffies - * before latching the timer count to guarantee that although - * the jiffies value might be older than the count (that is, - * the counter may underflow between the last point where - * jiffies was incremented and the point where we latch the - * count), it cannot be newer. - */ - jifs = jiffies; - outb_pit(0x00, PIT_MODE); /* latch the count ASAP */ - count = inb_pit(PIT_CH0); /* read the latched count */ - count |= inb_pit(PIT_CH0) << 8; - - /* VIA686a test code... reset the latch if count > max + 1 */ - if (count > LATCH) { - outb_pit(0x34, PIT_MODE); - outb_pit(LATCH & 0xff, PIT_CH0); - outb_pit(LATCH >> 8, PIT_CH0); - count = LATCH - 1; - } - - /* - * It's possible for count to appear to go the wrong way for a - * couple of reasons: - * - * 1. The timer counter underflows, but we haven't handled the - * resulting interrupt and incremented jiffies yet. - * 2. Hardware problem with the timer, not giving us continuous time, - * the counter does small "jumps" upwards on some Pentium systems, - * (see c't 95/10 page 335 for Neptun bug.) - * - * Previous attempts to handle these cases intelligently were - * buggy, so we just do the simple thing now. - */ - if (count > old_count && jifs == old_jifs) - count = old_count; - - old_count = count; - old_jifs = jifs; - - raw_spin_unlock_irqrestore(&i8253_lock, flags); - - count = (LATCH - 1) - count; - - return (cycle_t)(jifs * LATCH) + count; -} - -static struct clocksource pit_cs = { - .name = "pit", - .rating = 110, - .read = pit_read, - .mask = CLOCKSOURCE_MASK(32), - .mult = 0, - .shift = 20, -}; - static int __init init_pit_clocksource(void) { /* @@ -205,10 +126,7 @@ static int __init init_pit_clocksource(void) pit_ce.mode != CLOCK_EVT_MODE_PERIODIC) return 0; - pit_cs.mult = clocksource_hz2mult(CLOCK_TICK_RATE, pit_cs.shift); - - return clocksource_register(&pit_cs); + return clocksource_i8253_init(); } arch_initcall(init_pit_clocksource); - #endif /* !CONFIG_X86_64 */ diff --git a/arch/x86/kernel/irq.c b/arch/x86/kernel/irq.c index 1cb0b9fc78dc..6c0802eb2f7f 100644 --- a/arch/x86/kernel/irq.c +++ b/arch/x86/kernel/irq.c @@ -249,7 +249,7 @@ void fixup_irqs(void) data = irq_desc_get_irq_data(desc); affinity = data->affinity; - if (!irq_has_action(irq) || + if (!irq_has_action(irq) || irqd_is_per_cpu(data) || cpumask_subset(affinity, cpu_online_mask)) { raw_spin_unlock(&desc->lock); continue; @@ -276,7 +276,8 @@ void fixup_irqs(void) else if (!(warned++)) set_affinity = 0; - if (!irqd_can_move_in_process_context(data) && chip->irq_unmask) + if (!irqd_can_move_in_process_context(data) && + !irqd_irq_disabled(data) && chip->irq_unmask) chip->irq_unmask(data); raw_spin_unlock(&desc->lock); diff --git a/arch/x86/kernel/jump_label.c b/arch/x86/kernel/jump_label.c index 961b6b30ba90..3fee346ef545 100644 --- a/arch/x86/kernel/jump_label.c +++ b/arch/x86/kernel/jump_label.c @@ -34,7 +34,7 @@ void arch_jump_label_transform(struct jump_entry *entry, code.offset = entry->target - (entry->code + JUMP_LABEL_NOP_SIZE); } else - memcpy(&code, ideal_nop5, JUMP_LABEL_NOP_SIZE); + memcpy(&code, ideal_nops[NOP_ATOMIC5], JUMP_LABEL_NOP_SIZE); get_online_cpus(); mutex_lock(&text_mutex); text_poke_smp((void *)entry->code, &code, JUMP_LABEL_NOP_SIZE); @@ -44,7 +44,8 @@ void arch_jump_label_transform(struct jump_entry *entry, void arch_jump_label_text_poke_early(jump_label_t addr) { - text_poke_early((void *)addr, ideal_nop5, JUMP_LABEL_NOP_SIZE); + text_poke_early((void *)addr, ideal_nops[NOP_ATOMIC5], + JUMP_LABEL_NOP_SIZE); } #endif diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c index c969fd9d1566..f1a6244d7d93 100644 --- a/arch/x86/kernel/kprobes.c +++ b/arch/x86/kernel/kprobes.c @@ -1183,12 +1183,13 @@ static void __kprobes optimized_callback(struct optimized_kprobe *op, struct pt_regs *regs) { struct kprobe_ctlblk *kcb = get_kprobe_ctlblk(); + unsigned long flags; /* This is possible if op is under delayed unoptimizing */ if (kprobe_disabled(&op->kp)) return; - preempt_disable(); + local_irq_save(flags); if (kprobe_running()) { kprobes_inc_nmissed_count(&op->kp); } else { @@ -1207,7 +1208,7 @@ static void __kprobes optimized_callback(struct optimized_kprobe *op, opt_pre_handler(&op->kp, regs); __this_cpu_write(current_kprobe, NULL); } - preempt_enable_no_resched(); + local_irq_restore(flags); } static int __kprobes copy_optimized_instructions(u8 *dest, u8 *src) diff --git a/arch/x86/kernel/kvmclock.c b/arch/x86/kernel/kvmclock.c index f98d3eafe07a..6389a6bca11b 100644 --- a/arch/x86/kernel/kvmclock.c +++ b/arch/x86/kernel/kvmclock.c @@ -26,8 +26,6 @@ #include <asm/x86_init.h> #include <asm/reboot.h> -#define KVM_SCALE 22 - static int kvmclock = 1; static int msr_kvm_system_time = MSR_KVM_SYSTEM_TIME; static int msr_kvm_wall_clock = MSR_KVM_WALL_CLOCK; @@ -120,8 +118,6 @@ static struct clocksource kvm_clock = { .read = kvm_clock_get_cycles, .rating = 400, .mask = CLOCKSOURCE_MASK(64), - .mult = 1 << KVM_SCALE, - .shift = KVM_SCALE, .flags = CLOCK_SOURCE_IS_CONTINUOUS, }; @@ -203,7 +199,7 @@ void __init kvmclock_init(void) machine_ops.crash_shutdown = kvm_crash_shutdown; #endif kvm_get_preset_lpj(); - clocksource_register(&kvm_clock); + clocksource_register_hz(&kvm_clock, NSEC_PER_SEC); pv_info.paravirt_enabled = 1; pv_info.name = "KVM"; diff --git a/arch/x86/kernel/module.c b/arch/x86/kernel/module.c index ab23f1ad4bf1..52f256f2cc81 100644 --- a/arch/x86/kernel/module.c +++ b/arch/x86/kernel/module.c @@ -24,6 +24,7 @@ #include <linux/bug.h> #include <linux/mm.h> #include <linux/gfp.h> +#include <linux/jump_label.h> #include <asm/system.h> #include <asm/page.h> diff --git a/arch/x86/kernel/mpparse.c b/arch/x86/kernel/mpparse.c index 5a532ce646bf..9103b89c145a 100644 --- a/arch/x86/kernel/mpparse.c +++ b/arch/x86/kernel/mpparse.c @@ -285,7 +285,7 @@ static void __init construct_default_ioirq_mptable(int mpc_default_type) intsrc.type = MP_INTSRC; intsrc.irqflag = 0; /* conforming */ intsrc.srcbus = 0; - intsrc.dstapic = mp_ioapics[0].apicid; + intsrc.dstapic = mpc_ioapic_id(0); intsrc.irqtype = mp_INT; @@ -715,17 +715,15 @@ static void __init check_irq_src(struct mpc_intsrc *m, int *nr_m_spare) } } -static int +static int __init check_slot(unsigned long mpc_new_phys, unsigned long mpc_new_length, int count) { - int ret = 0; - if (!mpc_new_phys || count <= mpc_new_length) { WARN(1, "update_mptable: No spare slots (length: %x)\n", count); return -1; } - return ret; + return 0; } #else /* CONFIG_X86_IO_APIC */ static diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c index 9ea999a4dcc1..b49d00da2aed 100644 --- a/arch/x86/kernel/pci-dma.c +++ b/arch/x86/kernel/pci-dma.c @@ -68,74 +68,10 @@ int dma_set_mask(struct device *dev, u64 mask) } EXPORT_SYMBOL(dma_set_mask); -#if defined(CONFIG_X86_64) && !defined(CONFIG_NUMA) -static __initdata void *dma32_bootmem_ptr; -static unsigned long dma32_bootmem_size __initdata = (128ULL<<20); - -static int __init parse_dma32_size_opt(char *p) -{ - if (!p) - return -EINVAL; - dma32_bootmem_size = memparse(p, &p); - return 0; -} -early_param("dma32_size", parse_dma32_size_opt); - -void __init dma32_reserve_bootmem(void) -{ - unsigned long size, align; - if (max_pfn <= MAX_DMA32_PFN) - return; - - /* - * check aperture_64.c allocate_aperture() for reason about - * using 512M as goal - */ - align = 64ULL<<20; - size = roundup(dma32_bootmem_size, align); - dma32_bootmem_ptr = __alloc_bootmem_nopanic(size, align, - 512ULL<<20); - /* - * Kmemleak should not scan this block as it may not be mapped via the - * kernel direct mapping. - */ - kmemleak_ignore(dma32_bootmem_ptr); - if (dma32_bootmem_ptr) - dma32_bootmem_size = size; - else - dma32_bootmem_size = 0; -} -static void __init dma32_free_bootmem(void) -{ - - if (max_pfn <= MAX_DMA32_PFN) - return; - - if (!dma32_bootmem_ptr) - return; - - free_bootmem(__pa(dma32_bootmem_ptr), dma32_bootmem_size); - - dma32_bootmem_ptr = NULL; - dma32_bootmem_size = 0; -} -#else -void __init dma32_reserve_bootmem(void) -{ -} -static void __init dma32_free_bootmem(void) -{ -} - -#endif - void __init pci_iommu_alloc(void) { struct iommu_table_entry *p; - /* free the range so iommu could get some range less than 4G */ - dma32_free_bootmem(); - sort_iommu_table(__iommu_table, __iommu_table_end); check_iommu_entries(__iommu_table, __iommu_table_end); diff --git a/arch/x86/kernel/pci-iommu_table.c b/arch/x86/kernel/pci-iommu_table.c index 55d745ec1181..35ccf75696eb 100644 --- a/arch/x86/kernel/pci-iommu_table.c +++ b/arch/x86/kernel/pci-iommu_table.c @@ -50,20 +50,14 @@ void __init check_iommu_entries(struct iommu_table_entry *start, struct iommu_table_entry *finish) { struct iommu_table_entry *p, *q, *x; - char sym_p[KSYM_SYMBOL_LEN]; - char sym_q[KSYM_SYMBOL_LEN]; /* Simple cyclic dependency checker. */ for (p = start; p < finish; p++) { q = find_dependents_of(start, finish, p); x = find_dependents_of(start, finish, q); if (p == x) { - sprint_symbol(sym_p, (unsigned long)p->detect); - sprint_symbol(sym_q, (unsigned long)q->detect); - - printk(KERN_ERR "CYCLIC DEPENDENCY FOUND! %s depends" \ - " on %s and vice-versa. BREAKING IT.\n", - sym_p, sym_q); + printk(KERN_ERR "CYCLIC DEPENDENCY FOUND! %pS depends on %pS and vice-versa. BREAKING IT.\n", + p->detect, q->detect); /* Heavy handed way..*/ x->depend = 0; } @@ -72,12 +66,8 @@ void __init check_iommu_entries(struct iommu_table_entry *start, for (p = start; p < finish; p++) { q = find_dependents_of(p, finish, p); if (q && q > p) { - sprint_symbol(sym_p, (unsigned long)p->detect); - sprint_symbol(sym_q, (unsigned long)q->detect); - - printk(KERN_ERR "EXECUTION ORDER INVALID! %s "\ - "should be called before %s!\n", - sym_p, sym_q); + printk(KERN_ERR "EXECUTION ORDER INVALID! %pS should be called before %pS!\n", + p->detect, q->detect); } } } diff --git a/arch/x86/kernel/probe_roms_32.c b/arch/x86/kernel/probe_roms.c index 071e7fea42e5..ba0a4cce53be 100644 --- a/arch/x86/kernel/probe_roms_32.c +++ b/arch/x86/kernel/probe_roms.c @@ -73,6 +73,107 @@ static struct resource video_rom_resource = { .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM }; +/* does this oprom support the given pci device, or any of the devices + * that the driver supports? + */ +static bool match_id(struct pci_dev *pdev, unsigned short vendor, unsigned short device) +{ + struct pci_driver *drv = pdev->driver; + const struct pci_device_id *id; + + if (pdev->vendor == vendor && pdev->device == device) + return true; + + for (id = drv ? drv->id_table : NULL; id && id->vendor; id++) + if (id->vendor == vendor && id->device == device) + break; + + return id && id->vendor; +} + +static bool probe_list(struct pci_dev *pdev, unsigned short vendor, + const unsigned char *rom_list) +{ + unsigned short device; + + do { + if (probe_kernel_address(rom_list, device) != 0) + device = 0; + + if (device && match_id(pdev, vendor, device)) + break; + + rom_list += 2; + } while (device); + + return !!device; +} + +static struct resource *find_oprom(struct pci_dev *pdev) +{ + struct resource *oprom = NULL; + int i; + + for (i = 0; i < ARRAY_SIZE(adapter_rom_resources); i++) { + struct resource *res = &adapter_rom_resources[i]; + unsigned short offset, vendor, device, list, rev; + const unsigned char *rom; + + if (res->end == 0) + break; + + rom = isa_bus_to_virt(res->start); + if (probe_kernel_address(rom + 0x18, offset) != 0) + continue; + + if (probe_kernel_address(rom + offset + 0x4, vendor) != 0) + continue; + + if (probe_kernel_address(rom + offset + 0x6, device) != 0) + continue; + + if (match_id(pdev, vendor, device)) { + oprom = res; + break; + } + + if (probe_kernel_address(rom + offset + 0x8, list) == 0 && + probe_kernel_address(rom + offset + 0xc, rev) == 0 && + rev >= 3 && list && + probe_list(pdev, vendor, rom + offset + list)) { + oprom = res; + break; + } + } + + return oprom; +} + +void *pci_map_biosrom(struct pci_dev *pdev) +{ + struct resource *oprom = find_oprom(pdev); + + if (!oprom) + return NULL; + + return ioremap(oprom->start, resource_size(oprom)); +} +EXPORT_SYMBOL(pci_map_biosrom); + +void pci_unmap_biosrom(void __iomem *image) +{ + iounmap(image); +} +EXPORT_SYMBOL(pci_unmap_biosrom); + +size_t pci_biosrom_size(struct pci_dev *pdev) +{ + struct resource *oprom = find_oprom(pdev); + + return oprom ? resource_size(oprom) : 0; +} +EXPORT_SYMBOL(pci_biosrom_size); + #define ROMSIGNATURE 0xaa55 static int __init romsignature(const unsigned char *rom) diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index d46cbe46b7ab..88a90a977f8e 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -449,7 +449,7 @@ EXPORT_SYMBOL_GPL(cpu_idle_wait); void mwait_idle_with_hints(unsigned long ax, unsigned long cx) { if (!need_resched()) { - if (cpu_has(__this_cpu_ptr(&cpu_info), X86_FEATURE_CLFLUSH_MONITOR)) + if (this_cpu_has(X86_FEATURE_CLFLUSH_MONITOR)) clflush((void *)¤t_thread_info()->flags); __monitor((void *)¤t_thread_info()->flags, 0, 0); @@ -465,7 +465,7 @@ static void mwait_idle(void) if (!need_resched()) { trace_power_start(POWER_CSTATE, 1, smp_processor_id()); trace_cpu_idle(1, smp_processor_id()); - if (cpu_has(__this_cpu_ptr(&cpu_info), X86_FEATURE_CLFLUSH_MONITOR)) + if (this_cpu_has(X86_FEATURE_CLFLUSH_MONITOR)) clflush((void *)¤t_thread_info()->flags); __monitor((void *)¤t_thread_info()->flags, 0, 0); diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c index 45892dc4b72a..807c2a2b80f1 100644 --- a/arch/x86/kernel/ptrace.c +++ b/arch/x86/kernel/ptrace.c @@ -608,6 +608,9 @@ static int ptrace_write_dr7(struct task_struct *tsk, unsigned long data) unsigned len, type; struct perf_event *bp; + if (ptrace_get_breakpoints(tsk) < 0) + return -ESRCH; + data &= ~DR_CONTROL_RESERVED; old_dr7 = ptrace_get_dr7(thread->ptrace_bps); restore: @@ -655,6 +658,9 @@ restore: } goto restore; } + + ptrace_put_breakpoints(tsk); + return ((orig_ret < 0) ? orig_ret : rc); } @@ -668,10 +674,17 @@ static unsigned long ptrace_get_debugreg(struct task_struct *tsk, int n) if (n < HBP_NUM) { struct perf_event *bp; + + if (ptrace_get_breakpoints(tsk) < 0) + return -ESRCH; + bp = thread->ptrace_bps[n]; if (!bp) - return 0; - val = bp->hw.info.address; + val = 0; + else + val = bp->hw.info.address; + + ptrace_put_breakpoints(tsk); } else if (n == 6) { val = thread->debugreg6; } else if (n == 7) { @@ -686,6 +699,10 @@ static int ptrace_set_breakpoint_addr(struct task_struct *tsk, int nr, struct perf_event *bp; struct thread_struct *t = &tsk->thread; struct perf_event_attr attr; + int err = 0; + + if (ptrace_get_breakpoints(tsk) < 0) + return -ESRCH; if (!t->ptrace_bps[nr]) { ptrace_breakpoint_init(&attr); @@ -709,24 +726,23 @@ static int ptrace_set_breakpoint_addr(struct task_struct *tsk, int nr, * writing for the user. And anyway this is the previous * behaviour. */ - if (IS_ERR(bp)) - return PTR_ERR(bp); + if (IS_ERR(bp)) { + err = PTR_ERR(bp); + goto put; + } t->ptrace_bps[nr] = bp; } else { - int err; - bp = t->ptrace_bps[nr]; attr = bp->attr; attr.bp_addr = addr; err = modify_user_hw_breakpoint(bp, &attr); - if (err) - return err; } - - return 0; +put: + ptrace_put_breakpoints(tsk); + return err; } /* @@ -1347,7 +1363,7 @@ void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs, * We must return the syscall number to actually look up in the table. * This can be -1L to skip running any syscall at all. */ -asmregparm long syscall_trace_enter(struct pt_regs *regs) +long syscall_trace_enter(struct pt_regs *regs) { long ret = 0; @@ -1392,7 +1408,7 @@ asmregparm long syscall_trace_enter(struct pt_regs *regs) return ret ?: regs->orig_ax; } -asmregparm void syscall_trace_leave(struct pt_regs *regs) +void syscall_trace_leave(struct pt_regs *regs) { bool step; diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c index 08c44b08bf5b..0c016f727695 100644 --- a/arch/x86/kernel/reboot.c +++ b/arch/x86/kernel/reboot.c @@ -36,7 +36,7 @@ EXPORT_SYMBOL(pm_power_off); static const struct desc_ptr no_idt = {}; static int reboot_mode; -enum reboot_type reboot_type = BOOT_KBD; +enum reboot_type reboot_type = BOOT_ACPI; int reboot_force; #if defined(CONFIG_X86_32) && defined(CONFIG_SMP) @@ -478,9 +478,24 @@ void __attribute__((weak)) mach_reboot_fixups(void) { } +/* + * Windows compatible x86 hardware expects the following on reboot: + * + * 1) If the FADT has the ACPI reboot register flag set, try it + * 2) If still alive, write to the keyboard controller + * 3) If still alive, write to the ACPI reboot register again + * 4) If still alive, write to the keyboard controller again + * + * If the machine is still alive at this stage, it gives up. We default to + * following the same pattern, except that if we're still alive after (4) we'll + * try to force a triple fault and then cycle between hitting the keyboard + * controller and doing that + */ static void native_machine_emergency_restart(void) { int i; + int attempt = 0; + int orig_reboot_type = reboot_type; if (reboot_emergency) emergency_vmx_disable_all(); @@ -502,6 +517,13 @@ static void native_machine_emergency_restart(void) outb(0xfe, 0x64); /* pulse reset low */ udelay(50); } + if (attempt == 0 && orig_reboot_type == BOOT_ACPI) { + attempt = 1; + reboot_type = BOOT_ACPI; + } else { + reboot_type = BOOT_TRIPLE; + } + break; case BOOT_TRIPLE: load_idt(&no_idt); diff --git a/arch/x86/kernel/reboot_32.S b/arch/x86/kernel/reboot_32.S index 29092b38d816..1d5c46df0d78 100644 --- a/arch/x86/kernel/reboot_32.S +++ b/arch/x86/kernel/reboot_32.S @@ -21,26 +21,26 @@ r_base = . /* Get our own relocated address */ call 1f 1: popl %ebx - subl $1b, %ebx + subl $(1b - r_base), %ebx /* Compute the equivalent real-mode segment */ movl %ebx, %ecx shrl $4, %ecx /* Patch post-real-mode segment jump */ - movw dispatch_table(%ebx,%eax,2),%ax - movw %ax, 101f(%ebx) - movw %cx, 102f(%ebx) + movw (dispatch_table - r_base)(%ebx,%eax,2),%ax + movw %ax, (101f - r_base)(%ebx) + movw %cx, (102f - r_base)(%ebx) /* Set up the IDT for real mode. */ - lidtl machine_real_restart_idt(%ebx) + lidtl (machine_real_restart_idt - r_base)(%ebx) /* * Set up a GDT from which we can load segment descriptors for real * mode. The GDT is not used in real mode; it is just needed here to * prepare the descriptors. */ - lgdtl machine_real_restart_gdt(%ebx) + lgdtl (machine_real_restart_gdt - r_base)(%ebx) /* * Load the data segment registers with 16-bit compatible values diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index 4be9b398470e..a3e5948670c2 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -691,8 +691,6 @@ early_param("reservelow", parse_reservelow); void __init setup_arch(char **cmdline_p) { - unsigned long flags; - #ifdef CONFIG_X86_32 memcpy(&boot_cpu_data, &new_cpu_data, sizeof(new_cpu_data)); visws_early_detect(); @@ -948,6 +946,8 @@ void __init setup_arch(char **cmdline_p) if (init_ohci1394_dma_early) init_ohci1394_dma_on_all_controllers(); #endif + /* Allocate bigger log buffer */ + setup_log_buf(1); reserve_initrd(); @@ -966,7 +966,6 @@ void __init setup_arch(char **cmdline_p) initmem_init(); memblock_find_dma_reserve(); - dma32_reserve_bootmem(); #ifdef CONFIG_KVM_CLOCK kvmclock_init(); @@ -1041,9 +1040,7 @@ void __init setup_arch(char **cmdline_p) mcheck_init(); - local_irq_save(flags); - arch_init_ideal_nop5(); - local_irq_restore(flags); + arch_init_ideal_nops(); } #ifdef CONFIG_X86_32 diff --git a/arch/x86/kernel/signal.c b/arch/x86/kernel/signal.c index 4fd173cd8e57..40a24932a8a1 100644 --- a/arch/x86/kernel/signal.c +++ b/arch/x86/kernel/signal.c @@ -601,10 +601,7 @@ long sys_rt_sigreturn(struct pt_regs *regs) goto badframe; sigdelsetmask(&set, ~_BLOCKABLE); - spin_lock_irq(¤t->sighand->siglock); - current->blocked = set; - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); + set_current_blocked(&set); if (restore_sigcontext(regs, &frame->uc.uc_mcontext, &ax)) goto badframe; @@ -682,6 +679,7 @@ static int handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka, sigset_t *oldset, struct pt_regs *regs) { + sigset_t blocked; int ret; /* Are we from a system call? */ @@ -741,12 +739,10 @@ handle_signal(unsigned long sig, siginfo_t *info, struct k_sigaction *ka, */ regs->flags &= ~X86_EFLAGS_TF; - spin_lock_irq(¤t->sighand->siglock); - sigorsets(¤t->blocked, ¤t->blocked, &ka->sa.sa_mask); + sigorsets(&blocked, ¤t->blocked, &ka->sa.sa_mask); if (!(ka->sa.sa_flags & SA_NODEFER)) - sigaddset(¤t->blocked, sig); - recalc_sigpending(); - spin_unlock_irq(¤t->sighand->siglock); + sigaddset(&blocked, sig); + set_current_blocked(&blocked); tracehook_signal_handler(sig, info, ka, regs, test_thread_flag(TIF_SINGLESTEP)); diff --git a/arch/x86/kernel/smp.c b/arch/x86/kernel/smp.c index 513deac7228d..013e7eba83bb 100644 --- a/arch/x86/kernel/smp.c +++ b/arch/x86/kernel/smp.c @@ -194,14 +194,13 @@ static void native_stop_other_cpus(int wait) } /* - * Reschedule call back. Nothing to do, - * all the work is done automatically when - * we return from the interrupt. + * Reschedule call back. */ void smp_reschedule_interrupt(struct pt_regs *regs) { ack_APIC_irq(); inc_irq_stat(irq_resched_count); + scheduler_ipi(); /* * KVM uses this interrupt to force a cpu out of guest mode */ diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index c2871d3c71b6..a3c430bdfb60 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -1332,9 +1332,9 @@ static inline void mwait_play_dead(void) void *mwait_ptr; struct cpuinfo_x86 *c = __this_cpu_ptr(&cpu_info); - if (!(cpu_has(c, X86_FEATURE_MWAIT) && mwait_usable(c))) + if (!this_cpu_has(X86_FEATURE_MWAIT) && mwait_usable(c)) return; - if (!cpu_has(__this_cpu_ptr(&cpu_info), X86_FEATURE_CLFLSH)) + if (!this_cpu_has(X86_FEATURE_CLFLSH)) return; if (__this_cpu_read(cpu_info.cpuid_level) < CPUID_MWAIT_LEAF) return; diff --git a/arch/x86/kernel/stacktrace.c b/arch/x86/kernel/stacktrace.c index 6515733a289d..55d9bc03f696 100644 --- a/arch/x86/kernel/stacktrace.c +++ b/arch/x86/kernel/stacktrace.c @@ -9,15 +9,6 @@ #include <linux/uaccess.h> #include <asm/stacktrace.h> -static void save_stack_warning(void *data, char *msg) -{ -} - -static void -save_stack_warning_symbol(void *data, char *msg, unsigned long symbol) -{ -} - static int save_stack_stack(void *data, char *name) { return 0; @@ -53,16 +44,12 @@ save_stack_address_nosched(void *data, unsigned long addr, int reliable) } static const struct stacktrace_ops save_stack_ops = { - .warning = save_stack_warning, - .warning_symbol = save_stack_warning_symbol, .stack = save_stack_stack, .address = save_stack_address, .walk_stack = print_context_stack, }; static const struct stacktrace_ops save_stack_ops_nosched = { - .warning = save_stack_warning, - .warning_symbol = save_stack_warning_symbol, .stack = save_stack_stack, .address = save_stack_address_nosched, .walk_stack = print_context_stack, diff --git a/arch/x86/kernel/syscall_table_32.S b/arch/x86/kernel/syscall_table_32.S index abce34d5c79d..32cbffb0c494 100644 --- a/arch/x86/kernel/syscall_table_32.S +++ b/arch/x86/kernel/syscall_table_32.S @@ -344,3 +344,4 @@ ENTRY(sys_call_table) .long sys_open_by_handle_at .long sys_clock_adjtime .long sys_syncfs + .long sys_sendmmsg /* 345 */ diff --git a/arch/x86/kernel/tboot.c b/arch/x86/kernel/tboot.c index 998e972f3b1a..30ac65df7d4e 100644 --- a/arch/x86/kernel/tboot.c +++ b/arch/x86/kernel/tboot.c @@ -110,7 +110,6 @@ static struct mm_struct tboot_mm = { .mmap_sem = __RWSEM_INITIALIZER(init_mm.mmap_sem), .page_table_lock = __SPIN_LOCK_UNLOCKED(init_mm.page_table_lock), .mmlist = LIST_HEAD_INIT(init_mm.mmlist), - .cpu_vm_mask = CPU_MASK_ALL, }; static inline void switch_to_tboot_pt(void) diff --git a/arch/x86/kernel/test_nx.c b/arch/x86/kernel/test_nx.c index 787a5e499dd1..3f92ce07e525 100644 --- a/arch/x86/kernel/test_nx.c +++ b/arch/x86/kernel/test_nx.c @@ -161,7 +161,7 @@ static int test_NX(void) } #endif - return 0; + return ret; } static void test_exit(void) diff --git a/arch/x86/kernel/vmlinux.lds.S b/arch/x86/kernel/vmlinux.lds.S index 624a2016198e..61682f0ac264 100644 --- a/arch/x86/kernel/vmlinux.lds.S +++ b/arch/x86/kernel/vmlinux.lds.S @@ -306,6 +306,13 @@ SECTIONS } . = ALIGN(8); + .apicdrivers : AT(ADDR(.apicdrivers) - LOAD_OFFSET) { + __apicdrivers = .; + *(.apicdrivers); + __apicdrivers_end = .; + } + + . = ALIGN(8); /* * .exit.text is discard at runtime, not link time, to deal with * references from .altinstructions and .eh_frame @@ -319,7 +326,7 @@ SECTIONS } #if !defined(CONFIG_X86_64) || !defined(CONFIG_SMP) - PERCPU(INTERNODE_CACHE_BYTES, PAGE_SIZE) + PERCPU_SECTION(INTERNODE_CACHE_BYTES) #endif . = ALIGN(PAGE_SIZE); diff --git a/arch/x86/kernel/x86_init.c b/arch/x86/kernel/x86_init.c index c11514e9128b..6f164bd5e14d 100644 --- a/arch/x86/kernel/x86_init.c +++ b/arch/x86/kernel/x86_init.c @@ -35,7 +35,7 @@ void iommu_shutdown_noop(void) { } struct x86_init_ops x86_init __initdata = { .resources = { - .probe_roms = x86_init_noop, + .probe_roms = probe_roms, .reserve_resources = reserve_standard_io_resources, .memory_setup = default_machine_specific_memory_setup, }, @@ -61,6 +61,10 @@ struct x86_init_ops x86_init __initdata = { .banner = default_banner, }, + .mapping = { + .pagetable_reserve = native_pagetable_reserve, + }, + .paging = { .pagetable_setup_start = native_pagetable_setup_start, .pagetable_setup_done = native_pagetable_setup_done, diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c index 0ad47b819a8b..d6e2477feb18 100644 --- a/arch/x86/kvm/emulate.c +++ b/arch/x86/kvm/emulate.c @@ -73,9 +73,14 @@ #define MemAbs (1<<11) /* Memory operand is absolute displacement */ #define String (1<<12) /* String instruction (rep capable) */ #define Stack (1<<13) /* Stack instruction (push/pop) */ +#define GroupMask (7<<14) /* Opcode uses one of the group mechanisms */ #define Group (1<<14) /* Bits 3:5 of modrm byte extend opcode */ -#define GroupDual (1<<15) /* Alternate decoding of mod == 3 */ +#define GroupDual (2<<14) /* Alternate decoding of mod == 3 */ +#define Prefix (3<<14) /* Instruction varies with 66/f2/f3 prefix */ +#define RMExt (4<<14) /* Opcode extension in ModRM r/m if mod == 3 */ +#define Sse (1<<17) /* SSE Vector instruction */ /* Misc flags */ +#define Prot (1<<21) /* instruction generates #UD if not in prot-mode */ #define VendorSpecific (1<<22) /* Vendor specific instruction */ #define NoAccess (1<<23) /* Don't access memory (lea/invlpg/verr etc) */ #define Op3264 (1<<24) /* Operand is 64b in long mode, 32b otherwise */ @@ -102,11 +107,14 @@ struct opcode { u32 flags; + u8 intercept; union { int (*execute)(struct x86_emulate_ctxt *ctxt); struct opcode *group; struct group_dual *gdual; + struct gprefix *gprefix; } u; + int (*check_perm)(struct x86_emulate_ctxt *ctxt); }; struct group_dual { @@ -114,6 +122,13 @@ struct group_dual { struct opcode mod3[8]; }; +struct gprefix { + struct opcode pfx_no; + struct opcode pfx_66; + struct opcode pfx_f2; + struct opcode pfx_f3; +}; + /* EFLAGS bit definitions. */ #define EFLG_ID (1<<21) #define EFLG_VIP (1<<20) @@ -248,42 +263,42 @@ struct group_dual { "w", "r", _LO32, "r", "", "r") /* Instruction has three operands and one operand is stored in ECX register */ -#define __emulate_2op_cl(_op, _cl, _src, _dst, _eflags, _suffix, _type) \ - do { \ - unsigned long _tmp; \ - _type _clv = (_cl).val; \ - _type _srcv = (_src).val; \ - _type _dstv = (_dst).val; \ - \ - __asm__ __volatile__ ( \ - _PRE_EFLAGS("0", "5", "2") \ - _op _suffix " %4,%1 \n" \ - _POST_EFLAGS("0", "5", "2") \ - : "=m" (_eflags), "+r" (_dstv), "=&r" (_tmp) \ - : "c" (_clv) , "r" (_srcv), "i" (EFLAGS_MASK) \ - ); \ - \ - (_cl).val = (unsigned long) _clv; \ - (_src).val = (unsigned long) _srcv; \ - (_dst).val = (unsigned long) _dstv; \ +#define __emulate_2op_cl(_op, _cl, _src, _dst, _eflags, _suffix, _type) \ + do { \ + unsigned long _tmp; \ + _type _clv = (_cl).val; \ + _type _srcv = (_src).val; \ + _type _dstv = (_dst).val; \ + \ + __asm__ __volatile__ ( \ + _PRE_EFLAGS("0", "5", "2") \ + _op _suffix " %4,%1 \n" \ + _POST_EFLAGS("0", "5", "2") \ + : "=m" (_eflags), "+r" (_dstv), "=&r" (_tmp) \ + : "c" (_clv) , "r" (_srcv), "i" (EFLAGS_MASK) \ + ); \ + \ + (_cl).val = (unsigned long) _clv; \ + (_src).val = (unsigned long) _srcv; \ + (_dst).val = (unsigned long) _dstv; \ } while (0) -#define emulate_2op_cl(_op, _cl, _src, _dst, _eflags) \ - do { \ - switch ((_dst).bytes) { \ - case 2: \ - __emulate_2op_cl(_op, _cl, _src, _dst, _eflags, \ - "w", unsigned short); \ - break; \ - case 4: \ - __emulate_2op_cl(_op, _cl, _src, _dst, _eflags, \ - "l", unsigned int); \ - break; \ - case 8: \ - ON64(__emulate_2op_cl(_op, _cl, _src, _dst, _eflags, \ - "q", unsigned long)); \ - break; \ - } \ +#define emulate_2op_cl(_op, _cl, _src, _dst, _eflags) \ + do { \ + switch ((_dst).bytes) { \ + case 2: \ + __emulate_2op_cl(_op, _cl, _src, _dst, _eflags, \ + "w", unsigned short); \ + break; \ + case 4: \ + __emulate_2op_cl(_op, _cl, _src, _dst, _eflags, \ + "l", unsigned int); \ + break; \ + case 8: \ + ON64(__emulate_2op_cl(_op, _cl, _src, _dst, _eflags, \ + "q", unsigned long)); \ + break; \ + } \ } while (0) #define __emulate_1op(_op, _dst, _eflags, _suffix) \ @@ -346,13 +361,25 @@ struct group_dual { } while (0) /* instruction has only one source operand, destination is implicit (e.g. mul, div, imul, idiv) */ -#define emulate_1op_rax_rdx(_op, _src, _rax, _rdx, _eflags) \ - do { \ - switch((_src).bytes) { \ - case 1: __emulate_1op_rax_rdx(_op, _src, _rax, _rdx, _eflags, "b"); break; \ - case 2: __emulate_1op_rax_rdx(_op, _src, _rax, _rdx, _eflags, "w"); break; \ - case 4: __emulate_1op_rax_rdx(_op, _src, _rax, _rdx, _eflags, "l"); break; \ - case 8: ON64(__emulate_1op_rax_rdx(_op, _src, _rax, _rdx, _eflags, "q")); break; \ +#define emulate_1op_rax_rdx(_op, _src, _rax, _rdx, _eflags) \ + do { \ + switch((_src).bytes) { \ + case 1: \ + __emulate_1op_rax_rdx(_op, _src, _rax, _rdx, \ + _eflags, "b"); \ + break; \ + case 2: \ + __emulate_1op_rax_rdx(_op, _src, _rax, _rdx, \ + _eflags, "w"); \ + break; \ + case 4: \ + __emulate_1op_rax_rdx(_op, _src, _rax, _rdx, \ + _eflags, "l"); \ + break; \ + case 8: \ + ON64(__emulate_1op_rax_rdx(_op, _src, _rax, _rdx, \ + _eflags, "q")); \ + break; \ } \ } while (0) @@ -388,13 +415,33 @@ struct group_dual { (_type)_x; \ }) -#define insn_fetch_arr(_arr, _size, _eip) \ +#define insn_fetch_arr(_arr, _size, _eip) \ ({ rc = do_insn_fetch(ctxt, ops, (_eip), _arr, (_size)); \ if (rc != X86EMUL_CONTINUE) \ goto done; \ (_eip) += (_size); \ }) +static int emulator_check_intercept(struct x86_emulate_ctxt *ctxt, + enum x86_intercept intercept, + enum x86_intercept_stage stage) +{ + struct x86_instruction_info info = { + .intercept = intercept, + .rep_prefix = ctxt->decode.rep_prefix, + .modrm_mod = ctxt->decode.modrm_mod, + .modrm_reg = ctxt->decode.modrm_reg, + .modrm_rm = ctxt->decode.modrm_rm, + .src_val = ctxt->decode.src.val64, + .src_bytes = ctxt->decode.src.bytes, + .dst_bytes = ctxt->decode.dst.bytes, + .ad_bytes = ctxt->decode.ad_bytes, + .next_rip = ctxt->eip, + }; + + return ctxt->ops->intercept(ctxt, &info, stage); +} + static inline unsigned long ad_mask(struct decode_cache *c) { return (1UL << (c->ad_bytes << 3)) - 1; @@ -430,6 +477,13 @@ static inline void jmp_rel(struct decode_cache *c, int rel) register_address_increment(c, &c->eip, rel); } +static u32 desc_limit_scaled(struct desc_struct *desc) +{ + u32 limit = get_desc_limit(desc); + + return desc->g ? (limit << 12) | 0xfff : limit; +} + static void set_seg_override(struct decode_cache *c, int seg) { c->has_seg_override = true; @@ -442,11 +496,10 @@ static unsigned long seg_base(struct x86_emulate_ctxt *ctxt, if (ctxt->mode == X86EMUL_MODE_PROT64 && seg < VCPU_SREG_FS) return 0; - return ops->get_cached_segment_base(seg, ctxt->vcpu); + return ops->get_cached_segment_base(ctxt, seg); } static unsigned seg_override(struct x86_emulate_ctxt *ctxt, - struct x86_emulate_ops *ops, struct decode_cache *c) { if (!c->has_seg_override) @@ -455,18 +508,6 @@ static unsigned seg_override(struct x86_emulate_ctxt *ctxt, return c->seg_override; } -static ulong linear(struct x86_emulate_ctxt *ctxt, - struct segmented_address addr) -{ - struct decode_cache *c = &ctxt->decode; - ulong la; - - la = seg_base(ctxt, ctxt->ops, addr.seg) + addr.ea; - if (c->ad_bytes != 8) - la &= (u32)-1; - return la; -} - static int emulate_exception(struct x86_emulate_ctxt *ctxt, int vec, u32 error, bool valid) { @@ -476,11 +517,21 @@ static int emulate_exception(struct x86_emulate_ctxt *ctxt, int vec, return X86EMUL_PROPAGATE_FAULT; } +static int emulate_db(struct x86_emulate_ctxt *ctxt) +{ + return emulate_exception(ctxt, DB_VECTOR, 0, false); +} + static int emulate_gp(struct x86_emulate_ctxt *ctxt, int err) { return emulate_exception(ctxt, GP_VECTOR, err, true); } +static int emulate_ss(struct x86_emulate_ctxt *ctxt, int err) +{ + return emulate_exception(ctxt, SS_VECTOR, err, true); +} + static int emulate_ud(struct x86_emulate_ctxt *ctxt) { return emulate_exception(ctxt, UD_VECTOR, 0, false); @@ -496,6 +547,128 @@ static int emulate_de(struct x86_emulate_ctxt *ctxt) return emulate_exception(ctxt, DE_VECTOR, 0, false); } +static int emulate_nm(struct x86_emulate_ctxt *ctxt) +{ + return emulate_exception(ctxt, NM_VECTOR, 0, false); +} + +static u16 get_segment_selector(struct x86_emulate_ctxt *ctxt, unsigned seg) +{ + u16 selector; + struct desc_struct desc; + + ctxt->ops->get_segment(ctxt, &selector, &desc, NULL, seg); + return selector; +} + +static void set_segment_selector(struct x86_emulate_ctxt *ctxt, u16 selector, + unsigned seg) +{ + u16 dummy; + u32 base3; + struct desc_struct desc; + + ctxt->ops->get_segment(ctxt, &dummy, &desc, &base3, seg); + ctxt->ops->set_segment(ctxt, selector, &desc, base3, seg); +} + +static int __linearize(struct x86_emulate_ctxt *ctxt, + struct segmented_address addr, + unsigned size, bool write, bool fetch, + ulong *linear) +{ + struct decode_cache *c = &ctxt->decode; + struct desc_struct desc; + bool usable; + ulong la; + u32 lim; + u16 sel; + unsigned cpl, rpl; + + la = seg_base(ctxt, ctxt->ops, addr.seg) + addr.ea; + switch (ctxt->mode) { + case X86EMUL_MODE_REAL: + break; + case X86EMUL_MODE_PROT64: + if (((signed long)la << 16) >> 16 != la) + return emulate_gp(ctxt, 0); + break; + default: + usable = ctxt->ops->get_segment(ctxt, &sel, &desc, NULL, + addr.seg); + if (!usable) + goto bad; + /* code segment or read-only data segment */ + if (((desc.type & 8) || !(desc.type & 2)) && write) + goto bad; + /* unreadable code segment */ + if (!fetch && (desc.type & 8) && !(desc.type & 2)) + goto bad; + lim = desc_limit_scaled(&desc); + if ((desc.type & 8) || !(desc.type & 4)) { + /* expand-up segment */ + if (addr.ea > lim || (u32)(addr.ea + size - 1) > lim) + goto bad; + } else { + /* exapand-down segment */ + if (addr.ea <= lim || (u32)(addr.ea + size - 1) <= lim) + goto bad; + lim = desc.d ? 0xffffffff : 0xffff; + if (addr.ea > lim || (u32)(addr.ea + size - 1) > lim) + goto bad; + } + cpl = ctxt->ops->cpl(ctxt); + rpl = sel & 3; + cpl = max(cpl, rpl); + if (!(desc.type & 8)) { + /* data segment */ + if (cpl > desc.dpl) + goto bad; + } else if ((desc.type & 8) && !(desc.type & 4)) { + /* nonconforming code segment */ + if (cpl != desc.dpl) + goto bad; + } else if ((desc.type & 8) && (desc.type & 4)) { + /* conforming code segment */ + if (cpl < desc.dpl) + goto bad; + } + break; + } + if (fetch ? ctxt->mode != X86EMUL_MODE_PROT64 : c->ad_bytes != 8) + la &= (u32)-1; + *linear = la; + return X86EMUL_CONTINUE; +bad: + if (addr.seg == VCPU_SREG_SS) + return emulate_ss(ctxt, addr.seg); + else + return emulate_gp(ctxt, addr.seg); +} + +static int linearize(struct x86_emulate_ctxt *ctxt, + struct segmented_address addr, + unsigned size, bool write, + ulong *linear) +{ + return __linearize(ctxt, addr, size, write, false, linear); +} + + +static int segmented_read_std(struct x86_emulate_ctxt *ctxt, + struct segmented_address addr, + void *data, + unsigned size) +{ + int rc; + ulong linear; + + rc = linearize(ctxt, addr, size, false, &linear); + if (rc != X86EMUL_CONTINUE) + return rc; + return ctxt->ops->read_std(ctxt, linear, data, size, &ctxt->exception); +} + static int do_fetch_insn_byte(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops, unsigned long eip, u8 *dest) @@ -505,10 +678,15 @@ static int do_fetch_insn_byte(struct x86_emulate_ctxt *ctxt, int size, cur_size; if (eip == fc->end) { + unsigned long linear; + struct segmented_address addr = { .seg=VCPU_SREG_CS, .ea=eip}; cur_size = fc->end - fc->start; size = min(15UL - cur_size, PAGE_SIZE - offset_in_page(eip)); - rc = ops->fetch(ctxt->cs_base + eip, fc->data + cur_size, - size, ctxt->vcpu, &ctxt->exception); + rc = __linearize(ctxt, addr, size, false, true, &linear); + if (rc != X86EMUL_CONTINUE) + return rc; + rc = ops->fetch(ctxt, linear, fc->data + cur_size, + size, &ctxt->exception); if (rc != X86EMUL_CONTINUE) return rc; fc->end += size; @@ -551,7 +729,6 @@ static void *decode_register(u8 modrm_reg, unsigned long *regs, } static int read_descriptor(struct x86_emulate_ctxt *ctxt, - struct x86_emulate_ops *ops, struct segmented_address addr, u16 *size, unsigned long *address, int op_bytes) { @@ -560,13 +737,11 @@ static int read_descriptor(struct x86_emulate_ctxt *ctxt, if (op_bytes == 2) op_bytes = 3; *address = 0; - rc = ops->read_std(linear(ctxt, addr), (unsigned long *)size, 2, - ctxt->vcpu, &ctxt->exception); + rc = segmented_read_std(ctxt, addr, size, 2); if (rc != X86EMUL_CONTINUE) return rc; addr.ea += 2; - rc = ops->read_std(linear(ctxt, addr), address, op_bytes, - ctxt->vcpu, &ctxt->exception); + rc = segmented_read_std(ctxt, addr, address, op_bytes); return rc; } @@ -623,7 +798,63 @@ static void fetch_register_operand(struct operand *op) } } -static void decode_register_operand(struct operand *op, +static void read_sse_reg(struct x86_emulate_ctxt *ctxt, sse128_t *data, int reg) +{ + ctxt->ops->get_fpu(ctxt); + switch (reg) { + case 0: asm("movdqu %%xmm0, %0" : "=m"(*data)); break; + case 1: asm("movdqu %%xmm1, %0" : "=m"(*data)); break; + case 2: asm("movdqu %%xmm2, %0" : "=m"(*data)); break; + case 3: asm("movdqu %%xmm3, %0" : "=m"(*data)); break; + case 4: asm("movdqu %%xmm4, %0" : "=m"(*data)); break; + case 5: asm("movdqu %%xmm5, %0" : "=m"(*data)); break; + case 6: asm("movdqu %%xmm6, %0" : "=m"(*data)); break; + case 7: asm("movdqu %%xmm7, %0" : "=m"(*data)); break; +#ifdef CONFIG_X86_64 + case 8: asm("movdqu %%xmm8, %0" : "=m"(*data)); break; + case 9: asm("movdqu %%xmm9, %0" : "=m"(*data)); break; + case 10: asm("movdqu %%xmm10, %0" : "=m"(*data)); break; + case 11: asm("movdqu %%xmm11, %0" : "=m"(*data)); break; + case 12: asm("movdqu %%xmm12, %0" : "=m"(*data)); break; + case 13: asm("movdqu %%xmm13, %0" : "=m"(*data)); break; + case 14: asm("movdqu %%xmm14, %0" : "=m"(*data)); break; + case 15: asm("movdqu %%xmm15, %0" : "=m"(*data)); break; +#endif + default: BUG(); + } + ctxt->ops->put_fpu(ctxt); +} + +static void write_sse_reg(struct x86_emulate_ctxt *ctxt, sse128_t *data, + int reg) +{ + ctxt->ops->get_fpu(ctxt); + switch (reg) { + case 0: asm("movdqu %0, %%xmm0" : : "m"(*data)); break; + case 1: asm("movdqu %0, %%xmm1" : : "m"(*data)); break; + case 2: asm("movdqu %0, %%xmm2" : : "m"(*data)); break; + case 3: asm("movdqu %0, %%xmm3" : : "m"(*data)); break; + case 4: asm("movdqu %0, %%xmm4" : : "m"(*data)); break; + case 5: asm("movdqu %0, %%xmm5" : : "m"(*data)); break; + case 6: asm("movdqu %0, %%xmm6" : : "m"(*data)); break; + case 7: asm("movdqu %0, %%xmm7" : : "m"(*data)); break; +#ifdef CONFIG_X86_64 + case 8: asm("movdqu %0, %%xmm8" : : "m"(*data)); break; + case 9: asm("movdqu %0, %%xmm9" : : "m"(*data)); break; + case 10: asm("movdqu %0, %%xmm10" : : "m"(*data)); break; + case 11: asm("movdqu %0, %%xmm11" : : "m"(*data)); break; + case 12: asm("movdqu %0, %%xmm12" : : "m"(*data)); break; + case 13: asm("movdqu %0, %%xmm13" : : "m"(*data)); break; + case 14: asm("movdqu %0, %%xmm14" : : "m"(*data)); break; + case 15: asm("movdqu %0, %%xmm15" : : "m"(*data)); break; +#endif + default: BUG(); + } + ctxt->ops->put_fpu(ctxt); +} + +static void decode_register_operand(struct x86_emulate_ctxt *ctxt, + struct operand *op, struct decode_cache *c, int inhibit_bytereg) { @@ -632,6 +863,15 @@ static void decode_register_operand(struct operand *op, if (!(c->d & ModRM)) reg = (c->b & 7) | ((c->rex_prefix & 1) << 3); + + if (c->d & Sse) { + op->type = OP_XMM; + op->bytes = 16; + op->addr.xmm = reg; + read_sse_reg(ctxt, &op->vec_val, reg); + return; + } + op->type = OP_REG; if ((c->d & ByteOp) && !inhibit_bytereg) { op->addr.reg = decode_register(reg, c->regs, highbyte_regs); @@ -671,6 +911,13 @@ static int decode_modrm(struct x86_emulate_ctxt *ctxt, op->bytes = (c->d & ByteOp) ? 1 : c->op_bytes; op->addr.reg = decode_register(c->modrm_rm, c->regs, c->d & ByteOp); + if (c->d & Sse) { + op->type = OP_XMM; + op->bytes = 16; + op->addr.xmm = c->modrm_rm; + read_sse_reg(ctxt, &op->vec_val, c->modrm_rm); + return rc; + } fetch_register_operand(op); return rc; } @@ -819,8 +1066,8 @@ static int read_emulated(struct x86_emulate_ctxt *ctxt, if (mc->pos < mc->end) goto read_cached; - rc = ops->read_emulated(addr, mc->data + mc->end, n, - &ctxt->exception, ctxt->vcpu); + rc = ops->read_emulated(ctxt, addr, mc->data + mc->end, n, + &ctxt->exception); if (rc != X86EMUL_CONTINUE) return rc; mc->end += n; @@ -834,6 +1081,50 @@ static int read_emulated(struct x86_emulate_ctxt *ctxt, return X86EMUL_CONTINUE; } +static int segmented_read(struct x86_emulate_ctxt *ctxt, + struct segmented_address addr, + void *data, + unsigned size) +{ + int rc; + ulong linear; + + rc = linearize(ctxt, addr, size, false, &linear); + if (rc != X86EMUL_CONTINUE) + return rc; + return read_emulated(ctxt, ctxt->ops, linear, data, size); +} + +static int segmented_write(struct x86_emulate_ctxt *ctxt, + struct segmented_address addr, + const void *data, + unsigned size) +{ + int rc; + ulong linear; + + rc = linearize(ctxt, addr, size, true, &linear); + if (rc != X86EMUL_CONTINUE) + return rc; + return ctxt->ops->write_emulated(ctxt, linear, data, size, + &ctxt->exception); +} + +static int segmented_cmpxchg(struct x86_emulate_ctxt *ctxt, + struct segmented_address addr, + const void *orig_data, const void *data, + unsigned size) +{ + int rc; + ulong linear; + + rc = linearize(ctxt, addr, size, true, &linear); + if (rc != X86EMUL_CONTINUE) + return rc; + return ctxt->ops->cmpxchg_emulated(ctxt, linear, orig_data, data, + size, &ctxt->exception); +} + static int pio_in_emulated(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops, unsigned int size, unsigned short port, @@ -854,7 +1145,7 @@ static int pio_in_emulated(struct x86_emulate_ctxt *ctxt, if (n == 0) n = 1; rc->pos = rc->end = 0; - if (!ops->pio_in_emulated(size, port, rc->data, n, ctxt->vcpu)) + if (!ops->pio_in_emulated(ctxt, size, port, rc->data, n)) return 0; rc->end = n * size; } @@ -864,28 +1155,22 @@ static int pio_in_emulated(struct x86_emulate_ctxt *ctxt, return 1; } -static u32 desc_limit_scaled(struct desc_struct *desc) -{ - u32 limit = get_desc_limit(desc); - - return desc->g ? (limit << 12) | 0xfff : limit; -} - static void get_descriptor_table_ptr(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops, u16 selector, struct desc_ptr *dt) { if (selector & 1 << 2) { struct desc_struct desc; + u16 sel; + memset (dt, 0, sizeof *dt); - if (!ops->get_cached_descriptor(&desc, NULL, VCPU_SREG_LDTR, - ctxt->vcpu)) + if (!ops->get_segment(ctxt, &sel, &desc, NULL, VCPU_SREG_LDTR)) return; dt->size = desc_limit_scaled(&desc); /* what if limit > 65535? */ dt->address = get_desc_base(&desc); } else - ops->get_gdt(dt, ctxt->vcpu); + ops->get_gdt(ctxt, dt); } /* allowed just for 8 bytes segments */ @@ -903,8 +1188,7 @@ static int read_segment_descriptor(struct x86_emulate_ctxt *ctxt, if (dt.size < index * 8 + 7) return emulate_gp(ctxt, selector & 0xfffc); addr = dt.address + index * 8; - ret = ops->read_std(addr, desc, sizeof *desc, ctxt->vcpu, - &ctxt->exception); + ret = ops->read_std(ctxt, addr, desc, sizeof *desc, &ctxt->exception); return ret; } @@ -925,8 +1209,7 @@ static int write_segment_descriptor(struct x86_emulate_ctxt *ctxt, return emulate_gp(ctxt, selector & 0xfffc); addr = dt.address + index * 8; - ret = ops->write_std(addr, desc, sizeof *desc, ctxt->vcpu, - &ctxt->exception); + ret = ops->write_std(ctxt, addr, desc, sizeof *desc, &ctxt->exception); return ret; } @@ -986,7 +1269,7 @@ static int load_segment_descriptor(struct x86_emulate_ctxt *ctxt, rpl = selector & 3; dpl = seg_desc.dpl; - cpl = ops->cpl(ctxt->vcpu); + cpl = ops->cpl(ctxt); switch (seg) { case VCPU_SREG_SS: @@ -1042,8 +1325,7 @@ static int load_segment_descriptor(struct x86_emulate_ctxt *ctxt, return ret; } load: - ops->set_segment_selector(selector, seg, ctxt->vcpu); - ops->set_cached_descriptor(&seg_desc, 0, seg, ctxt->vcpu); + ops->set_segment(ctxt, selector, &seg_desc, 0, seg); return X86EMUL_CONTINUE; exception: emulate_exception(ctxt, err_vec, err_code, true); @@ -1069,8 +1351,7 @@ static void write_register_operand(struct operand *op) } } -static inline int writeback(struct x86_emulate_ctxt *ctxt, - struct x86_emulate_ops *ops) +static int writeback(struct x86_emulate_ctxt *ctxt) { int rc; struct decode_cache *c = &ctxt->decode; @@ -1081,23 +1362,22 @@ static inline int writeback(struct x86_emulate_ctxt *ctxt, break; case OP_MEM: if (c->lock_prefix) - rc = ops->cmpxchg_emulated( - linear(ctxt, c->dst.addr.mem), - &c->dst.orig_val, - &c->dst.val, - c->dst.bytes, - &ctxt->exception, - ctxt->vcpu); + rc = segmented_cmpxchg(ctxt, + c->dst.addr.mem, + &c->dst.orig_val, + &c->dst.val, + c->dst.bytes); else - rc = ops->write_emulated( - linear(ctxt, c->dst.addr.mem), - &c->dst.val, - c->dst.bytes, - &ctxt->exception, - ctxt->vcpu); + rc = segmented_write(ctxt, + c->dst.addr.mem, + &c->dst.val, + c->dst.bytes); if (rc != X86EMUL_CONTINUE) return rc; break; + case OP_XMM: + write_sse_reg(ctxt, &c->dst.vec_val, c->dst.addr.xmm); + break; case OP_NONE: /* no writeback */ break; @@ -1107,21 +1387,21 @@ static inline int writeback(struct x86_emulate_ctxt *ctxt, return X86EMUL_CONTINUE; } -static inline void emulate_push(struct x86_emulate_ctxt *ctxt, - struct x86_emulate_ops *ops) +static int em_push(struct x86_emulate_ctxt *ctxt) { struct decode_cache *c = &ctxt->decode; + struct segmented_address addr; - c->dst.type = OP_MEM; - c->dst.bytes = c->op_bytes; - c->dst.val = c->src.val; register_address_increment(c, &c->regs[VCPU_REGS_RSP], -c->op_bytes); - c->dst.addr.mem.ea = register_address(c, c->regs[VCPU_REGS_RSP]); - c->dst.addr.mem.seg = VCPU_SREG_SS; + addr.ea = register_address(c, c->regs[VCPU_REGS_RSP]); + addr.seg = VCPU_SREG_SS; + + /* Disable writeback. */ + c->dst.type = OP_NONE; + return segmented_write(ctxt, addr, &c->src.val, c->op_bytes); } static int emulate_pop(struct x86_emulate_ctxt *ctxt, - struct x86_emulate_ops *ops, void *dest, int len) { struct decode_cache *c = &ctxt->decode; @@ -1130,7 +1410,7 @@ static int emulate_pop(struct x86_emulate_ctxt *ctxt, addr.ea = register_address(c, c->regs[VCPU_REGS_RSP]); addr.seg = VCPU_SREG_SS; - rc = read_emulated(ctxt, ops, linear(ctxt, addr), dest, len); + rc = segmented_read(ctxt, addr, dest, len); if (rc != X86EMUL_CONTINUE) return rc; @@ -1138,6 +1418,13 @@ static int emulate_pop(struct x86_emulate_ctxt *ctxt, return rc; } +static int em_pop(struct x86_emulate_ctxt *ctxt) +{ + struct decode_cache *c = &ctxt->decode; + + return emulate_pop(ctxt, &c->dst.val, c->op_bytes); +} + static int emulate_popf(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops, void *dest, int len) @@ -1145,9 +1432,9 @@ static int emulate_popf(struct x86_emulate_ctxt *ctxt, int rc; unsigned long val, change_mask; int iopl = (ctxt->eflags & X86_EFLAGS_IOPL) >> IOPL_SHIFT; - int cpl = ops->cpl(ctxt->vcpu); + int cpl = ops->cpl(ctxt); - rc = emulate_pop(ctxt, ops, &val, len); + rc = emulate_pop(ctxt, &val, len); if (rc != X86EMUL_CONTINUE) return rc; @@ -1179,14 +1466,24 @@ static int emulate_popf(struct x86_emulate_ctxt *ctxt, return rc; } -static void emulate_push_sreg(struct x86_emulate_ctxt *ctxt, - struct x86_emulate_ops *ops, int seg) +static int em_popf(struct x86_emulate_ctxt *ctxt) { struct decode_cache *c = &ctxt->decode; - c->src.val = ops->get_segment_selector(seg, ctxt->vcpu); + c->dst.type = OP_REG; + c->dst.addr.reg = &ctxt->eflags; + c->dst.bytes = c->op_bytes; + return emulate_popf(ctxt, ctxt->ops, &c->dst.val, c->op_bytes); +} - emulate_push(ctxt, ops); +static int emulate_push_sreg(struct x86_emulate_ctxt *ctxt, + struct x86_emulate_ops *ops, int seg) +{ + struct decode_cache *c = &ctxt->decode; + + c->src.val = get_segment_selector(ctxt, seg); + + return em_push(ctxt); } static int emulate_pop_sreg(struct x86_emulate_ctxt *ctxt, @@ -1196,7 +1493,7 @@ static int emulate_pop_sreg(struct x86_emulate_ctxt *ctxt, unsigned long selector; int rc; - rc = emulate_pop(ctxt, ops, &selector, c->op_bytes); + rc = emulate_pop(ctxt, &selector, c->op_bytes); if (rc != X86EMUL_CONTINUE) return rc; @@ -1204,8 +1501,7 @@ static int emulate_pop_sreg(struct x86_emulate_ctxt *ctxt, return rc; } -static int emulate_pusha(struct x86_emulate_ctxt *ctxt, - struct x86_emulate_ops *ops) +static int em_pusha(struct x86_emulate_ctxt *ctxt) { struct decode_cache *c = &ctxt->decode; unsigned long old_esp = c->regs[VCPU_REGS_RSP]; @@ -1216,23 +1512,25 @@ static int emulate_pusha(struct x86_emulate_ctxt *ctxt, (reg == VCPU_REGS_RSP) ? (c->src.val = old_esp) : (c->src.val = c->regs[reg]); - emulate_push(ctxt, ops); - - rc = writeback(ctxt, ops); + rc = em_push(ctxt); if (rc != X86EMUL_CONTINUE) return rc; ++reg; } - /* Disable writeback. */ - c->dst.type = OP_NONE; - return rc; } -static int emulate_popa(struct x86_emulate_ctxt *ctxt, - struct x86_emulate_ops *ops) +static int em_pushf(struct x86_emulate_ctxt *ctxt) +{ + struct decode_cache *c = &ctxt->decode; + + c->src.val = (unsigned long)ctxt->eflags; + return em_push(ctxt); +} + +static int em_popa(struct x86_emulate_ctxt *ctxt) { struct decode_cache *c = &ctxt->decode; int rc = X86EMUL_CONTINUE; @@ -1245,7 +1543,7 @@ static int emulate_popa(struct x86_emulate_ctxt *ctxt, --reg; } - rc = emulate_pop(ctxt, ops, &c->regs[reg], c->op_bytes); + rc = emulate_pop(ctxt, &c->regs[reg], c->op_bytes); if (rc != X86EMUL_CONTINUE) break; --reg; @@ -1265,37 +1563,32 @@ int emulate_int_real(struct x86_emulate_ctxt *ctxt, /* TODO: Add limit checks */ c->src.val = ctxt->eflags; - emulate_push(ctxt, ops); - rc = writeback(ctxt, ops); + rc = em_push(ctxt); if (rc != X86EMUL_CONTINUE) return rc; ctxt->eflags &= ~(EFLG_IF | EFLG_TF | EFLG_AC); - c->src.val = ops->get_segment_selector(VCPU_SREG_CS, ctxt->vcpu); - emulate_push(ctxt, ops); - rc = writeback(ctxt, ops); + c->src.val = get_segment_selector(ctxt, VCPU_SREG_CS); + rc = em_push(ctxt); if (rc != X86EMUL_CONTINUE) return rc; c->src.val = c->eip; - emulate_push(ctxt, ops); - rc = writeback(ctxt, ops); + rc = em_push(ctxt); if (rc != X86EMUL_CONTINUE) return rc; - c->dst.type = OP_NONE; - - ops->get_idt(&dt, ctxt->vcpu); + ops->get_idt(ctxt, &dt); eip_addr = dt.address + (irq << 2); cs_addr = dt.address + (irq << 2) + 2; - rc = ops->read_std(cs_addr, &cs, 2, ctxt->vcpu, &ctxt->exception); + rc = ops->read_std(ctxt, cs_addr, &cs, 2, &ctxt->exception); if (rc != X86EMUL_CONTINUE) return rc; - rc = ops->read_std(eip_addr, &eip, 2, ctxt->vcpu, &ctxt->exception); + rc = ops->read_std(ctxt, eip_addr, &eip, 2, &ctxt->exception); if (rc != X86EMUL_CONTINUE) return rc; @@ -1339,7 +1632,7 @@ static int emulate_iret_real(struct x86_emulate_ctxt *ctxt, /* TODO: Add stack limit check */ - rc = emulate_pop(ctxt, ops, &temp_eip, c->op_bytes); + rc = emulate_pop(ctxt, &temp_eip, c->op_bytes); if (rc != X86EMUL_CONTINUE) return rc; @@ -1347,12 +1640,12 @@ static int emulate_iret_real(struct x86_emulate_ctxt *ctxt, if (temp_eip & ~0xffff) return emulate_gp(ctxt, 0); - rc = emulate_pop(ctxt, ops, &cs, c->op_bytes); + rc = emulate_pop(ctxt, &cs, c->op_bytes); if (rc != X86EMUL_CONTINUE) return rc; - rc = emulate_pop(ctxt, ops, &temp_eflags, c->op_bytes); + rc = emulate_pop(ctxt, &temp_eflags, c->op_bytes); if (rc != X86EMUL_CONTINUE) return rc; @@ -1394,15 +1687,31 @@ static inline int emulate_iret(struct x86_emulate_ctxt *ctxt, } } -static inline int emulate_grp1a(struct x86_emulate_ctxt *ctxt, - struct x86_emulate_ops *ops) +static int em_jmp_far(struct x86_emulate_ctxt *ctxt) +{ + struct decode_cache *c = &ctxt->decode; + int rc; + unsigned short sel; + + memcpy(&sel, c->src.valptr + c->op_bytes, 2); + + rc = load_segment_descriptor(ctxt, ctxt->ops, sel, VCPU_SREG_CS); + if (rc != X86EMUL_CONTINUE) + return rc; + + c->eip = 0; + memcpy(&c->eip, c->src.valptr, c->op_bytes); + return X86EMUL_CONTINUE; +} + +static int em_grp1a(struct x86_emulate_ctxt *ctxt) { struct decode_cache *c = &ctxt->decode; - return emulate_pop(ctxt, ops, &c->dst.val, c->dst.bytes); + return emulate_pop(ctxt, &c->dst.val, c->dst.bytes); } -static inline void emulate_grp2(struct x86_emulate_ctxt *ctxt) +static int em_grp2(struct x86_emulate_ctxt *ctxt) { struct decode_cache *c = &ctxt->decode; switch (c->modrm_reg) { @@ -1429,10 +1738,10 @@ static inline void emulate_grp2(struct x86_emulate_ctxt *ctxt) emulate_2op_SrcB("sar", c->src, c->dst, ctxt->eflags); break; } + return X86EMUL_CONTINUE; } -static inline int emulate_grp3(struct x86_emulate_ctxt *ctxt, - struct x86_emulate_ops *ops) +static int em_grp3(struct x86_emulate_ctxt *ctxt) { struct decode_cache *c = &ctxt->decode; unsigned long *rax = &c->regs[VCPU_REGS_RAX]; @@ -1471,10 +1780,10 @@ static inline int emulate_grp3(struct x86_emulate_ctxt *ctxt, return X86EMUL_CONTINUE; } -static inline int emulate_grp45(struct x86_emulate_ctxt *ctxt, - struct x86_emulate_ops *ops) +static int em_grp45(struct x86_emulate_ctxt *ctxt) { struct decode_cache *c = &ctxt->decode; + int rc = X86EMUL_CONTINUE; switch (c->modrm_reg) { case 0: /* inc */ @@ -1488,21 +1797,23 @@ static inline int emulate_grp45(struct x86_emulate_ctxt *ctxt, old_eip = c->eip; c->eip = c->src.val; c->src.val = old_eip; - emulate_push(ctxt, ops); + rc = em_push(ctxt); break; } case 4: /* jmp abs */ c->eip = c->src.val; break; + case 5: /* jmp far */ + rc = em_jmp_far(ctxt); + break; case 6: /* push */ - emulate_push(ctxt, ops); + rc = em_push(ctxt); break; } - return X86EMUL_CONTINUE; + return rc; } -static inline int emulate_grp9(struct x86_emulate_ctxt *ctxt, - struct x86_emulate_ops *ops) +static int em_grp9(struct x86_emulate_ctxt *ctxt) { struct decode_cache *c = &ctxt->decode; u64 old = c->dst.orig_val64; @@ -1528,12 +1839,12 @@ static int emulate_ret_far(struct x86_emulate_ctxt *ctxt, int rc; unsigned long cs; - rc = emulate_pop(ctxt, ops, &c->eip, c->op_bytes); + rc = emulate_pop(ctxt, &c->eip, c->op_bytes); if (rc != X86EMUL_CONTINUE) return rc; if (c->op_bytes == 4) c->eip = (u32)c->eip; - rc = emulate_pop(ctxt, ops, &cs, c->op_bytes); + rc = emulate_pop(ctxt, &cs, c->op_bytes); if (rc != X86EMUL_CONTINUE) return rc; rc = load_segment_descriptor(ctxt, ops, (u16)cs, VCPU_SREG_CS); @@ -1562,8 +1873,10 @@ setup_syscalls_segments(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops, struct desc_struct *cs, struct desc_struct *ss) { + u16 selector; + memset(cs, 0, sizeof(struct desc_struct)); - ops->get_cached_descriptor(cs, NULL, VCPU_SREG_CS, ctxt->vcpu); + ops->get_segment(ctxt, &selector, cs, NULL, VCPU_SREG_CS); memset(ss, 0, sizeof(struct desc_struct)); cs->l = 0; /* will be adjusted later */ @@ -1593,44 +1906,44 @@ emulate_syscall(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) struct desc_struct cs, ss; u64 msr_data; u16 cs_sel, ss_sel; + u64 efer = 0; /* syscall is not available in real mode */ if (ctxt->mode == X86EMUL_MODE_REAL || ctxt->mode == X86EMUL_MODE_VM86) return emulate_ud(ctxt); + ops->get_msr(ctxt, MSR_EFER, &efer); setup_syscalls_segments(ctxt, ops, &cs, &ss); - ops->get_msr(ctxt->vcpu, MSR_STAR, &msr_data); + ops->get_msr(ctxt, MSR_STAR, &msr_data); msr_data >>= 32; cs_sel = (u16)(msr_data & 0xfffc); ss_sel = (u16)(msr_data + 8); - if (is_long_mode(ctxt->vcpu)) { + if (efer & EFER_LMA) { cs.d = 0; cs.l = 1; } - ops->set_cached_descriptor(&cs, 0, VCPU_SREG_CS, ctxt->vcpu); - ops->set_segment_selector(cs_sel, VCPU_SREG_CS, ctxt->vcpu); - ops->set_cached_descriptor(&ss, 0, VCPU_SREG_SS, ctxt->vcpu); - ops->set_segment_selector(ss_sel, VCPU_SREG_SS, ctxt->vcpu); + ops->set_segment(ctxt, cs_sel, &cs, 0, VCPU_SREG_CS); + ops->set_segment(ctxt, ss_sel, &ss, 0, VCPU_SREG_SS); c->regs[VCPU_REGS_RCX] = c->eip; - if (is_long_mode(ctxt->vcpu)) { + if (efer & EFER_LMA) { #ifdef CONFIG_X86_64 c->regs[VCPU_REGS_R11] = ctxt->eflags & ~EFLG_RF; - ops->get_msr(ctxt->vcpu, + ops->get_msr(ctxt, ctxt->mode == X86EMUL_MODE_PROT64 ? MSR_LSTAR : MSR_CSTAR, &msr_data); c->eip = msr_data; - ops->get_msr(ctxt->vcpu, MSR_SYSCALL_MASK, &msr_data); + ops->get_msr(ctxt, MSR_SYSCALL_MASK, &msr_data); ctxt->eflags &= ~(msr_data | EFLG_RF); #endif } else { /* legacy mode */ - ops->get_msr(ctxt->vcpu, MSR_STAR, &msr_data); + ops->get_msr(ctxt, MSR_STAR, &msr_data); c->eip = (u32)msr_data; ctxt->eflags &= ~(EFLG_VM | EFLG_IF | EFLG_RF); @@ -1646,7 +1959,9 @@ emulate_sysenter(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) struct desc_struct cs, ss; u64 msr_data; u16 cs_sel, ss_sel; + u64 efer = 0; + ctxt->ops->get_msr(ctxt, MSR_EFER, &efer); /* inject #GP if in real mode */ if (ctxt->mode == X86EMUL_MODE_REAL) return emulate_gp(ctxt, 0); @@ -1659,7 +1974,7 @@ emulate_sysenter(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) setup_syscalls_segments(ctxt, ops, &cs, &ss); - ops->get_msr(ctxt->vcpu, MSR_IA32_SYSENTER_CS, &msr_data); + ops->get_msr(ctxt, MSR_IA32_SYSENTER_CS, &msr_data); switch (ctxt->mode) { case X86EMUL_MODE_PROT32: if ((msr_data & 0xfffc) == 0x0) @@ -1676,21 +1991,18 @@ emulate_sysenter(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) cs_sel &= ~SELECTOR_RPL_MASK; ss_sel = cs_sel + 8; ss_sel &= ~SELECTOR_RPL_MASK; - if (ctxt->mode == X86EMUL_MODE_PROT64 - || is_long_mode(ctxt->vcpu)) { + if (ctxt->mode == X86EMUL_MODE_PROT64 || (efer & EFER_LMA)) { cs.d = 0; cs.l = 1; } - ops->set_cached_descriptor(&cs, 0, VCPU_SREG_CS, ctxt->vcpu); - ops->set_segment_selector(cs_sel, VCPU_SREG_CS, ctxt->vcpu); - ops->set_cached_descriptor(&ss, 0, VCPU_SREG_SS, ctxt->vcpu); - ops->set_segment_selector(ss_sel, VCPU_SREG_SS, ctxt->vcpu); + ops->set_segment(ctxt, cs_sel, &cs, 0, VCPU_SREG_CS); + ops->set_segment(ctxt, ss_sel, &ss, 0, VCPU_SREG_SS); - ops->get_msr(ctxt->vcpu, MSR_IA32_SYSENTER_EIP, &msr_data); + ops->get_msr(ctxt, MSR_IA32_SYSENTER_EIP, &msr_data); c->eip = msr_data; - ops->get_msr(ctxt->vcpu, MSR_IA32_SYSENTER_ESP, &msr_data); + ops->get_msr(ctxt, MSR_IA32_SYSENTER_ESP, &msr_data); c->regs[VCPU_REGS_RSP] = msr_data; return X86EMUL_CONTINUE; @@ -1719,7 +2031,7 @@ emulate_sysexit(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) cs.dpl = 3; ss.dpl = 3; - ops->get_msr(ctxt->vcpu, MSR_IA32_SYSENTER_CS, &msr_data); + ops->get_msr(ctxt, MSR_IA32_SYSENTER_CS, &msr_data); switch (usermode) { case X86EMUL_MODE_PROT32: cs_sel = (u16)(msr_data + 16); @@ -1739,10 +2051,8 @@ emulate_sysexit(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) cs_sel |= SELECTOR_RPL_MASK; ss_sel |= SELECTOR_RPL_MASK; - ops->set_cached_descriptor(&cs, 0, VCPU_SREG_CS, ctxt->vcpu); - ops->set_segment_selector(cs_sel, VCPU_SREG_CS, ctxt->vcpu); - ops->set_cached_descriptor(&ss, 0, VCPU_SREG_SS, ctxt->vcpu); - ops->set_segment_selector(ss_sel, VCPU_SREG_SS, ctxt->vcpu); + ops->set_segment(ctxt, cs_sel, &cs, 0, VCPU_SREG_CS); + ops->set_segment(ctxt, ss_sel, &ss, 0, VCPU_SREG_SS); c->eip = c->regs[VCPU_REGS_RDX]; c->regs[VCPU_REGS_RSP] = c->regs[VCPU_REGS_RCX]; @@ -1759,7 +2069,7 @@ static bool emulator_bad_iopl(struct x86_emulate_ctxt *ctxt, if (ctxt->mode == X86EMUL_MODE_VM86) return true; iopl = (ctxt->eflags & X86_EFLAGS_IOPL) >> IOPL_SHIFT; - return ops->cpl(ctxt->vcpu) > iopl; + return ops->cpl(ctxt) > iopl; } static bool emulator_io_port_access_allowed(struct x86_emulate_ctxt *ctxt, @@ -1769,11 +2079,11 @@ static bool emulator_io_port_access_allowed(struct x86_emulate_ctxt *ctxt, struct desc_struct tr_seg; u32 base3; int r; - u16 io_bitmap_ptr, perm, bit_idx = port & 0x7; + u16 tr, io_bitmap_ptr, perm, bit_idx = port & 0x7; unsigned mask = (1 << len) - 1; unsigned long base; - ops->get_cached_descriptor(&tr_seg, &base3, VCPU_SREG_TR, ctxt->vcpu); + ops->get_segment(ctxt, &tr, &tr_seg, &base3, VCPU_SREG_TR); if (!tr_seg.p) return false; if (desc_limit_scaled(&tr_seg) < 103) @@ -1782,13 +2092,12 @@ static bool emulator_io_port_access_allowed(struct x86_emulate_ctxt *ctxt, #ifdef CONFIG_X86_64 base |= ((u64)base3) << 32; #endif - r = ops->read_std(base + 102, &io_bitmap_ptr, 2, ctxt->vcpu, NULL); + r = ops->read_std(ctxt, base + 102, &io_bitmap_ptr, 2, NULL); if (r != X86EMUL_CONTINUE) return false; if (io_bitmap_ptr + port/8 > desc_limit_scaled(&tr_seg)) return false; - r = ops->read_std(base + io_bitmap_ptr + port/8, &perm, 2, ctxt->vcpu, - NULL); + r = ops->read_std(ctxt, base + io_bitmap_ptr + port/8, &perm, 2, NULL); if (r != X86EMUL_CONTINUE) return false; if ((perm >> bit_idx) & mask) @@ -1829,11 +2138,11 @@ static void save_state_to_tss16(struct x86_emulate_ctxt *ctxt, tss->si = c->regs[VCPU_REGS_RSI]; tss->di = c->regs[VCPU_REGS_RDI]; - tss->es = ops->get_segment_selector(VCPU_SREG_ES, ctxt->vcpu); - tss->cs = ops->get_segment_selector(VCPU_SREG_CS, ctxt->vcpu); - tss->ss = ops->get_segment_selector(VCPU_SREG_SS, ctxt->vcpu); - tss->ds = ops->get_segment_selector(VCPU_SREG_DS, ctxt->vcpu); - tss->ldt = ops->get_segment_selector(VCPU_SREG_LDTR, ctxt->vcpu); + tss->es = get_segment_selector(ctxt, VCPU_SREG_ES); + tss->cs = get_segment_selector(ctxt, VCPU_SREG_CS); + tss->ss = get_segment_selector(ctxt, VCPU_SREG_SS); + tss->ds = get_segment_selector(ctxt, VCPU_SREG_DS); + tss->ldt = get_segment_selector(ctxt, VCPU_SREG_LDTR); } static int load_state_from_tss16(struct x86_emulate_ctxt *ctxt, @@ -1858,11 +2167,11 @@ static int load_state_from_tss16(struct x86_emulate_ctxt *ctxt, * SDM says that segment selectors are loaded before segment * descriptors */ - ops->set_segment_selector(tss->ldt, VCPU_SREG_LDTR, ctxt->vcpu); - ops->set_segment_selector(tss->es, VCPU_SREG_ES, ctxt->vcpu); - ops->set_segment_selector(tss->cs, VCPU_SREG_CS, ctxt->vcpu); - ops->set_segment_selector(tss->ss, VCPU_SREG_SS, ctxt->vcpu); - ops->set_segment_selector(tss->ds, VCPU_SREG_DS, ctxt->vcpu); + set_segment_selector(ctxt, tss->ldt, VCPU_SREG_LDTR); + set_segment_selector(ctxt, tss->es, VCPU_SREG_ES); + set_segment_selector(ctxt, tss->cs, VCPU_SREG_CS); + set_segment_selector(ctxt, tss->ss, VCPU_SREG_SS); + set_segment_selector(ctxt, tss->ds, VCPU_SREG_DS); /* * Now load segment descriptors. If fault happenes at this stage @@ -1896,7 +2205,7 @@ static int task_switch_16(struct x86_emulate_ctxt *ctxt, int ret; u32 new_tss_base = get_desc_base(new_desc); - ret = ops->read_std(old_tss_base, &tss_seg, sizeof tss_seg, ctxt->vcpu, + ret = ops->read_std(ctxt, old_tss_base, &tss_seg, sizeof tss_seg, &ctxt->exception); if (ret != X86EMUL_CONTINUE) /* FIXME: need to provide precise fault address */ @@ -1904,13 +2213,13 @@ static int task_switch_16(struct x86_emulate_ctxt *ctxt, save_state_to_tss16(ctxt, ops, &tss_seg); - ret = ops->write_std(old_tss_base, &tss_seg, sizeof tss_seg, ctxt->vcpu, + ret = ops->write_std(ctxt, old_tss_base, &tss_seg, sizeof tss_seg, &ctxt->exception); if (ret != X86EMUL_CONTINUE) /* FIXME: need to provide precise fault address */ return ret; - ret = ops->read_std(new_tss_base, &tss_seg, sizeof tss_seg, ctxt->vcpu, + ret = ops->read_std(ctxt, new_tss_base, &tss_seg, sizeof tss_seg, &ctxt->exception); if (ret != X86EMUL_CONTINUE) /* FIXME: need to provide precise fault address */ @@ -1919,10 +2228,10 @@ static int task_switch_16(struct x86_emulate_ctxt *ctxt, if (old_tss_sel != 0xffff) { tss_seg.prev_task_link = old_tss_sel; - ret = ops->write_std(new_tss_base, + ret = ops->write_std(ctxt, new_tss_base, &tss_seg.prev_task_link, sizeof tss_seg.prev_task_link, - ctxt->vcpu, &ctxt->exception); + &ctxt->exception); if (ret != X86EMUL_CONTINUE) /* FIXME: need to provide precise fault address */ return ret; @@ -1937,7 +2246,7 @@ static void save_state_to_tss32(struct x86_emulate_ctxt *ctxt, { struct decode_cache *c = &ctxt->decode; - tss->cr3 = ops->get_cr(3, ctxt->vcpu); + tss->cr3 = ops->get_cr(ctxt, 3); tss->eip = c->eip; tss->eflags = ctxt->eflags; tss->eax = c->regs[VCPU_REGS_RAX]; @@ -1949,13 +2258,13 @@ static void save_state_to_tss32(struct x86_emulate_ctxt *ctxt, tss->esi = c->regs[VCPU_REGS_RSI]; tss->edi = c->regs[VCPU_REGS_RDI]; - tss->es = ops->get_segment_selector(VCPU_SREG_ES, ctxt->vcpu); - tss->cs = ops->get_segment_selector(VCPU_SREG_CS, ctxt->vcpu); - tss->ss = ops->get_segment_selector(VCPU_SREG_SS, ctxt->vcpu); - tss->ds = ops->get_segment_selector(VCPU_SREG_DS, ctxt->vcpu); - tss->fs = ops->get_segment_selector(VCPU_SREG_FS, ctxt->vcpu); - tss->gs = ops->get_segment_selector(VCPU_SREG_GS, ctxt->vcpu); - tss->ldt_selector = ops->get_segment_selector(VCPU_SREG_LDTR, ctxt->vcpu); + tss->es = get_segment_selector(ctxt, VCPU_SREG_ES); + tss->cs = get_segment_selector(ctxt, VCPU_SREG_CS); + tss->ss = get_segment_selector(ctxt, VCPU_SREG_SS); + tss->ds = get_segment_selector(ctxt, VCPU_SREG_DS); + tss->fs = get_segment_selector(ctxt, VCPU_SREG_FS); + tss->gs = get_segment_selector(ctxt, VCPU_SREG_GS); + tss->ldt_selector = get_segment_selector(ctxt, VCPU_SREG_LDTR); } static int load_state_from_tss32(struct x86_emulate_ctxt *ctxt, @@ -1965,7 +2274,7 @@ static int load_state_from_tss32(struct x86_emulate_ctxt *ctxt, struct decode_cache *c = &ctxt->decode; int ret; - if (ops->set_cr(3, tss->cr3, ctxt->vcpu)) + if (ops->set_cr(ctxt, 3, tss->cr3)) return emulate_gp(ctxt, 0); c->eip = tss->eip; ctxt->eflags = tss->eflags | 2; @@ -1982,13 +2291,13 @@ static int load_state_from_tss32(struct x86_emulate_ctxt *ctxt, * SDM says that segment selectors are loaded before segment * descriptors */ - ops->set_segment_selector(tss->ldt_selector, VCPU_SREG_LDTR, ctxt->vcpu); - ops->set_segment_selector(tss->es, VCPU_SREG_ES, ctxt->vcpu); - ops->set_segment_selector(tss->cs, VCPU_SREG_CS, ctxt->vcpu); - ops->set_segment_selector(tss->ss, VCPU_SREG_SS, ctxt->vcpu); - ops->set_segment_selector(tss->ds, VCPU_SREG_DS, ctxt->vcpu); - ops->set_segment_selector(tss->fs, VCPU_SREG_FS, ctxt->vcpu); - ops->set_segment_selector(tss->gs, VCPU_SREG_GS, ctxt->vcpu); + set_segment_selector(ctxt, tss->ldt_selector, VCPU_SREG_LDTR); + set_segment_selector(ctxt, tss->es, VCPU_SREG_ES); + set_segment_selector(ctxt, tss->cs, VCPU_SREG_CS); + set_segment_selector(ctxt, tss->ss, VCPU_SREG_SS); + set_segment_selector(ctxt, tss->ds, VCPU_SREG_DS); + set_segment_selector(ctxt, tss->fs, VCPU_SREG_FS); + set_segment_selector(ctxt, tss->gs, VCPU_SREG_GS); /* * Now load segment descriptors. If fault happenes at this stage @@ -2028,7 +2337,7 @@ static int task_switch_32(struct x86_emulate_ctxt *ctxt, int ret; u32 new_tss_base = get_desc_base(new_desc); - ret = ops->read_std(old_tss_base, &tss_seg, sizeof tss_seg, ctxt->vcpu, + ret = ops->read_std(ctxt, old_tss_base, &tss_seg, sizeof tss_seg, &ctxt->exception); if (ret != X86EMUL_CONTINUE) /* FIXME: need to provide precise fault address */ @@ -2036,13 +2345,13 @@ static int task_switch_32(struct x86_emulate_ctxt *ctxt, save_state_to_tss32(ctxt, ops, &tss_seg); - ret = ops->write_std(old_tss_base, &tss_seg, sizeof tss_seg, ctxt->vcpu, + ret = ops->write_std(ctxt, old_tss_base, &tss_seg, sizeof tss_seg, &ctxt->exception); if (ret != X86EMUL_CONTINUE) /* FIXME: need to provide precise fault address */ return ret; - ret = ops->read_std(new_tss_base, &tss_seg, sizeof tss_seg, ctxt->vcpu, + ret = ops->read_std(ctxt, new_tss_base, &tss_seg, sizeof tss_seg, &ctxt->exception); if (ret != X86EMUL_CONTINUE) /* FIXME: need to provide precise fault address */ @@ -2051,10 +2360,10 @@ static int task_switch_32(struct x86_emulate_ctxt *ctxt, if (old_tss_sel != 0xffff) { tss_seg.prev_task_link = old_tss_sel; - ret = ops->write_std(new_tss_base, + ret = ops->write_std(ctxt, new_tss_base, &tss_seg.prev_task_link, sizeof tss_seg.prev_task_link, - ctxt->vcpu, &ctxt->exception); + &ctxt->exception); if (ret != X86EMUL_CONTINUE) /* FIXME: need to provide precise fault address */ return ret; @@ -2070,9 +2379,9 @@ static int emulator_do_task_switch(struct x86_emulate_ctxt *ctxt, { struct desc_struct curr_tss_desc, next_tss_desc; int ret; - u16 old_tss_sel = ops->get_segment_selector(VCPU_SREG_TR, ctxt->vcpu); + u16 old_tss_sel = get_segment_selector(ctxt, VCPU_SREG_TR); ulong old_tss_base = - ops->get_cached_segment_base(VCPU_SREG_TR, ctxt->vcpu); + ops->get_cached_segment_base(ctxt, VCPU_SREG_TR); u32 desc_limit; /* FIXME: old_tss_base == ~0 ? */ @@ -2088,7 +2397,7 @@ static int emulator_do_task_switch(struct x86_emulate_ctxt *ctxt, if (reason != TASK_SWITCH_IRET) { if ((tss_selector & 3) > next_tss_desc.dpl || - ops->cpl(ctxt->vcpu) > next_tss_desc.dpl) + ops->cpl(ctxt) > next_tss_desc.dpl) return emulate_gp(ctxt, 0); } @@ -2132,9 +2441,8 @@ static int emulator_do_task_switch(struct x86_emulate_ctxt *ctxt, &next_tss_desc); } - ops->set_cr(0, ops->get_cr(0, ctxt->vcpu) | X86_CR0_TS, ctxt->vcpu); - ops->set_cached_descriptor(&next_tss_desc, 0, VCPU_SREG_TR, ctxt->vcpu); - ops->set_segment_selector(tss_selector, VCPU_SREG_TR, ctxt->vcpu); + ops->set_cr(ctxt, 0, ops->get_cr(ctxt, 0) | X86_CR0_TS); + ops->set_segment(ctxt, tss_selector, &next_tss_desc, 0, VCPU_SREG_TR); if (has_error_code) { struct decode_cache *c = &ctxt->decode; @@ -2142,7 +2450,7 @@ static int emulator_do_task_switch(struct x86_emulate_ctxt *ctxt, c->op_bytes = c->ad_bytes = (next_tss_desc.type & 8) ? 4 : 2; c->lock_prefix = 0; c->src.val = (unsigned long) error_code; - emulate_push(ctxt, ops); + ret = em_push(ctxt); } return ret; @@ -2162,13 +2470,10 @@ int emulator_task_switch(struct x86_emulate_ctxt *ctxt, rc = emulator_do_task_switch(ctxt, ops, tss_selector, reason, has_error_code, error_code); - if (rc == X86EMUL_CONTINUE) { - rc = writeback(ctxt, ops); - if (rc == X86EMUL_CONTINUE) - ctxt->eip = c->eip; - } + if (rc == X86EMUL_CONTINUE) + ctxt->eip = c->eip; - return (rc == X86EMUL_UNHANDLEABLE) ? -1 : 0; + return (rc == X86EMUL_UNHANDLEABLE) ? EMULATION_FAILED : EMULATION_OK; } static void string_addr_inc(struct x86_emulate_ctxt *ctxt, unsigned seg, @@ -2182,12 +2487,6 @@ static void string_addr_inc(struct x86_emulate_ctxt *ctxt, unsigned seg, op->addr.mem.seg = seg; } -static int em_push(struct x86_emulate_ctxt *ctxt) -{ - emulate_push(ctxt, ctxt->ops); - return X86EMUL_CONTINUE; -} - static int em_das(struct x86_emulate_ctxt *ctxt) { struct decode_cache *c = &ctxt->decode; @@ -2234,7 +2533,7 @@ static int em_call_far(struct x86_emulate_ctxt *ctxt) ulong old_eip; int rc; - old_cs = ctxt->ops->get_segment_selector(VCPU_SREG_CS, ctxt->vcpu); + old_cs = get_segment_selector(ctxt, VCPU_SREG_CS); old_eip = c->eip; memcpy(&sel, c->src.valptr + c->op_bytes, 2); @@ -2245,20 +2544,12 @@ static int em_call_far(struct x86_emulate_ctxt *ctxt) memcpy(&c->eip, c->src.valptr, c->op_bytes); c->src.val = old_cs; - emulate_push(ctxt, ctxt->ops); - rc = writeback(ctxt, ctxt->ops); + rc = em_push(ctxt); if (rc != X86EMUL_CONTINUE) return rc; c->src.val = old_eip; - emulate_push(ctxt, ctxt->ops); - rc = writeback(ctxt, ctxt->ops); - if (rc != X86EMUL_CONTINUE) - return rc; - - c->dst.type = OP_NONE; - - return X86EMUL_CONTINUE; + return em_push(ctxt); } static int em_ret_near_imm(struct x86_emulate_ctxt *ctxt) @@ -2269,13 +2560,79 @@ static int em_ret_near_imm(struct x86_emulate_ctxt *ctxt) c->dst.type = OP_REG; c->dst.addr.reg = &c->eip; c->dst.bytes = c->op_bytes; - rc = emulate_pop(ctxt, ctxt->ops, &c->dst.val, c->op_bytes); + rc = emulate_pop(ctxt, &c->dst.val, c->op_bytes); if (rc != X86EMUL_CONTINUE) return rc; register_address_increment(c, &c->regs[VCPU_REGS_RSP], c->src.val); return X86EMUL_CONTINUE; } +static int em_add(struct x86_emulate_ctxt *ctxt) +{ + struct decode_cache *c = &ctxt->decode; + + emulate_2op_SrcV("add", c->src, c->dst, ctxt->eflags); + return X86EMUL_CONTINUE; +} + +static int em_or(struct x86_emulate_ctxt *ctxt) +{ + struct decode_cache *c = &ctxt->decode; + + emulate_2op_SrcV("or", c->src, c->dst, ctxt->eflags); + return X86EMUL_CONTINUE; +} + +static int em_adc(struct x86_emulate_ctxt *ctxt) +{ + struct decode_cache *c = &ctxt->decode; + + emulate_2op_SrcV("adc", c->src, c->dst, ctxt->eflags); + return X86EMUL_CONTINUE; +} + +static int em_sbb(struct x86_emulate_ctxt *ctxt) +{ + struct decode_cache *c = &ctxt->decode; + + emulate_2op_SrcV("sbb", c->src, c->dst, ctxt->eflags); + return X86EMUL_CONTINUE; +} + +static int em_and(struct x86_emulate_ctxt *ctxt) +{ + struct decode_cache *c = &ctxt->decode; + + emulate_2op_SrcV("and", c->src, c->dst, ctxt->eflags); + return X86EMUL_CONTINUE; +} + +static int em_sub(struct x86_emulate_ctxt *ctxt) +{ + struct decode_cache *c = &ctxt->decode; + + emulate_2op_SrcV("sub", c->src, c->dst, ctxt->eflags); + return X86EMUL_CONTINUE; +} + +static int em_xor(struct x86_emulate_ctxt *ctxt) +{ + struct decode_cache *c = &ctxt->decode; + + emulate_2op_SrcV("xor", c->src, c->dst, ctxt->eflags); + return X86EMUL_CONTINUE; +} + +static int em_cmp(struct x86_emulate_ctxt *ctxt) +{ + struct decode_cache *c = &ctxt->decode; + + emulate_2op_SrcV("cmp", c->src, c->dst, ctxt->eflags); + /* Disable writeback. */ + c->dst.type = OP_NONE; + return X86EMUL_CONTINUE; +} + static int em_imul(struct x86_emulate_ctxt *ctxt) { struct decode_cache *c = &ctxt->decode; @@ -2306,13 +2663,10 @@ static int em_cwd(struct x86_emulate_ctxt *ctxt) static int em_rdtsc(struct x86_emulate_ctxt *ctxt) { - unsigned cpl = ctxt->ops->cpl(ctxt->vcpu); struct decode_cache *c = &ctxt->decode; u64 tsc = 0; - if (cpl > 0 && (ctxt->ops->get_cr(4, ctxt->vcpu) & X86_CR4_TSD)) - return emulate_gp(ctxt, 0); - ctxt->ops->get_msr(ctxt->vcpu, MSR_IA32_TSC, &tsc); + ctxt->ops->get_msr(ctxt, MSR_IA32_TSC, &tsc); c->regs[VCPU_REGS_RAX] = (u32)tsc; c->regs[VCPU_REGS_RDX] = tsc >> 32; return X86EMUL_CONTINUE; @@ -2325,22 +2679,375 @@ static int em_mov(struct x86_emulate_ctxt *ctxt) return X86EMUL_CONTINUE; } +static int em_movdqu(struct x86_emulate_ctxt *ctxt) +{ + struct decode_cache *c = &ctxt->decode; + memcpy(&c->dst.vec_val, &c->src.vec_val, c->op_bytes); + return X86EMUL_CONTINUE; +} + +static int em_invlpg(struct x86_emulate_ctxt *ctxt) +{ + struct decode_cache *c = &ctxt->decode; + int rc; + ulong linear; + + rc = linearize(ctxt, c->src.addr.mem, 1, false, &linear); + if (rc == X86EMUL_CONTINUE) + ctxt->ops->invlpg(ctxt, linear); + /* Disable writeback. */ + c->dst.type = OP_NONE; + return X86EMUL_CONTINUE; +} + +static int em_clts(struct x86_emulate_ctxt *ctxt) +{ + ulong cr0; + + cr0 = ctxt->ops->get_cr(ctxt, 0); + cr0 &= ~X86_CR0_TS; + ctxt->ops->set_cr(ctxt, 0, cr0); + return X86EMUL_CONTINUE; +} + +static int em_vmcall(struct x86_emulate_ctxt *ctxt) +{ + struct decode_cache *c = &ctxt->decode; + int rc; + + if (c->modrm_mod != 3 || c->modrm_rm != 1) + return X86EMUL_UNHANDLEABLE; + + rc = ctxt->ops->fix_hypercall(ctxt); + if (rc != X86EMUL_CONTINUE) + return rc; + + /* Let the processor re-execute the fixed hypercall */ + c->eip = ctxt->eip; + /* Disable writeback. */ + c->dst.type = OP_NONE; + return X86EMUL_CONTINUE; +} + +static int em_lgdt(struct x86_emulate_ctxt *ctxt) +{ + struct decode_cache *c = &ctxt->decode; + struct desc_ptr desc_ptr; + int rc; + + rc = read_descriptor(ctxt, c->src.addr.mem, + &desc_ptr.size, &desc_ptr.address, + c->op_bytes); + if (rc != X86EMUL_CONTINUE) + return rc; + ctxt->ops->set_gdt(ctxt, &desc_ptr); + /* Disable writeback. */ + c->dst.type = OP_NONE; + return X86EMUL_CONTINUE; +} + +static int em_vmmcall(struct x86_emulate_ctxt *ctxt) +{ + struct decode_cache *c = &ctxt->decode; + int rc; + + rc = ctxt->ops->fix_hypercall(ctxt); + + /* Disable writeback. */ + c->dst.type = OP_NONE; + return rc; +} + +static int em_lidt(struct x86_emulate_ctxt *ctxt) +{ + struct decode_cache *c = &ctxt->decode; + struct desc_ptr desc_ptr; + int rc; + + rc = read_descriptor(ctxt, c->src.addr.mem, + &desc_ptr.size, &desc_ptr.address, + c->op_bytes); + if (rc != X86EMUL_CONTINUE) + return rc; + ctxt->ops->set_idt(ctxt, &desc_ptr); + /* Disable writeback. */ + c->dst.type = OP_NONE; + return X86EMUL_CONTINUE; +} + +static int em_smsw(struct x86_emulate_ctxt *ctxt) +{ + struct decode_cache *c = &ctxt->decode; + + c->dst.bytes = 2; + c->dst.val = ctxt->ops->get_cr(ctxt, 0); + return X86EMUL_CONTINUE; +} + +static int em_lmsw(struct x86_emulate_ctxt *ctxt) +{ + struct decode_cache *c = &ctxt->decode; + ctxt->ops->set_cr(ctxt, 0, (ctxt->ops->get_cr(ctxt, 0) & ~0x0eul) + | (c->src.val & 0x0f)); + c->dst.type = OP_NONE; + return X86EMUL_CONTINUE; +} + +static bool valid_cr(int nr) +{ + switch (nr) { + case 0: + case 2 ... 4: + case 8: + return true; + default: + return false; + } +} + +static int check_cr_read(struct x86_emulate_ctxt *ctxt) +{ + struct decode_cache *c = &ctxt->decode; + + if (!valid_cr(c->modrm_reg)) + return emulate_ud(ctxt); + + return X86EMUL_CONTINUE; +} + +static int check_cr_write(struct x86_emulate_ctxt *ctxt) +{ + struct decode_cache *c = &ctxt->decode; + u64 new_val = c->src.val64; + int cr = c->modrm_reg; + u64 efer = 0; + + static u64 cr_reserved_bits[] = { + 0xffffffff00000000ULL, + 0, 0, 0, /* CR3 checked later */ + CR4_RESERVED_BITS, + 0, 0, 0, + CR8_RESERVED_BITS, + }; + + if (!valid_cr(cr)) + return emulate_ud(ctxt); + + if (new_val & cr_reserved_bits[cr]) + return emulate_gp(ctxt, 0); + + switch (cr) { + case 0: { + u64 cr4; + if (((new_val & X86_CR0_PG) && !(new_val & X86_CR0_PE)) || + ((new_val & X86_CR0_NW) && !(new_val & X86_CR0_CD))) + return emulate_gp(ctxt, 0); + + cr4 = ctxt->ops->get_cr(ctxt, 4); + ctxt->ops->get_msr(ctxt, MSR_EFER, &efer); + + if ((new_val & X86_CR0_PG) && (efer & EFER_LME) && + !(cr4 & X86_CR4_PAE)) + return emulate_gp(ctxt, 0); + + break; + } + case 3: { + u64 rsvd = 0; + + ctxt->ops->get_msr(ctxt, MSR_EFER, &efer); + if (efer & EFER_LMA) + rsvd = CR3_L_MODE_RESERVED_BITS; + else if (ctxt->ops->get_cr(ctxt, 4) & X86_CR4_PAE) + rsvd = CR3_PAE_RESERVED_BITS; + else if (ctxt->ops->get_cr(ctxt, 0) & X86_CR0_PG) + rsvd = CR3_NONPAE_RESERVED_BITS; + + if (new_val & rsvd) + return emulate_gp(ctxt, 0); + + break; + } + case 4: { + u64 cr4; + + cr4 = ctxt->ops->get_cr(ctxt, 4); + ctxt->ops->get_msr(ctxt, MSR_EFER, &efer); + + if ((efer & EFER_LMA) && !(new_val & X86_CR4_PAE)) + return emulate_gp(ctxt, 0); + + break; + } + } + + return X86EMUL_CONTINUE; +} + +static int check_dr7_gd(struct x86_emulate_ctxt *ctxt) +{ + unsigned long dr7; + + ctxt->ops->get_dr(ctxt, 7, &dr7); + + /* Check if DR7.Global_Enable is set */ + return dr7 & (1 << 13); +} + +static int check_dr_read(struct x86_emulate_ctxt *ctxt) +{ + struct decode_cache *c = &ctxt->decode; + int dr = c->modrm_reg; + u64 cr4; + + if (dr > 7) + return emulate_ud(ctxt); + + cr4 = ctxt->ops->get_cr(ctxt, 4); + if ((cr4 & X86_CR4_DE) && (dr == 4 || dr == 5)) + return emulate_ud(ctxt); + + if (check_dr7_gd(ctxt)) + return emulate_db(ctxt); + + return X86EMUL_CONTINUE; +} + +static int check_dr_write(struct x86_emulate_ctxt *ctxt) +{ + struct decode_cache *c = &ctxt->decode; + u64 new_val = c->src.val64; + int dr = c->modrm_reg; + + if ((dr == 6 || dr == 7) && (new_val & 0xffffffff00000000ULL)) + return emulate_gp(ctxt, 0); + + return check_dr_read(ctxt); +} + +static int check_svme(struct x86_emulate_ctxt *ctxt) +{ + u64 efer; + + ctxt->ops->get_msr(ctxt, MSR_EFER, &efer); + + if (!(efer & EFER_SVME)) + return emulate_ud(ctxt); + + return X86EMUL_CONTINUE; +} + +static int check_svme_pa(struct x86_emulate_ctxt *ctxt) +{ + u64 rax = ctxt->decode.regs[VCPU_REGS_RAX]; + + /* Valid physical address? */ + if (rax & 0xffff000000000000ULL) + return emulate_gp(ctxt, 0); + + return check_svme(ctxt); +} + +static int check_rdtsc(struct x86_emulate_ctxt *ctxt) +{ + u64 cr4 = ctxt->ops->get_cr(ctxt, 4); + + if (cr4 & X86_CR4_TSD && ctxt->ops->cpl(ctxt)) + return emulate_ud(ctxt); + + return X86EMUL_CONTINUE; +} + +static int check_rdpmc(struct x86_emulate_ctxt *ctxt) +{ + u64 cr4 = ctxt->ops->get_cr(ctxt, 4); + u64 rcx = ctxt->decode.regs[VCPU_REGS_RCX]; + + if ((!(cr4 & X86_CR4_PCE) && ctxt->ops->cpl(ctxt)) || + (rcx > 3)) + return emulate_gp(ctxt, 0); + + return X86EMUL_CONTINUE; +} + +static int check_perm_in(struct x86_emulate_ctxt *ctxt) +{ + struct decode_cache *c = &ctxt->decode; + + c->dst.bytes = min(c->dst.bytes, 4u); + if (!emulator_io_permited(ctxt, ctxt->ops, c->src.val, c->dst.bytes)) + return emulate_gp(ctxt, 0); + + return X86EMUL_CONTINUE; +} + +static int check_perm_out(struct x86_emulate_ctxt *ctxt) +{ + struct decode_cache *c = &ctxt->decode; + + c->src.bytes = min(c->src.bytes, 4u); + if (!emulator_io_permited(ctxt, ctxt->ops, c->dst.val, c->src.bytes)) + return emulate_gp(ctxt, 0); + + return X86EMUL_CONTINUE; +} + #define D(_y) { .flags = (_y) } +#define DI(_y, _i) { .flags = (_y), .intercept = x86_intercept_##_i } +#define DIP(_y, _i, _p) { .flags = (_y), .intercept = x86_intercept_##_i, \ + .check_perm = (_p) } #define N D(0) +#define EXT(_f, _e) { .flags = ((_f) | RMExt), .u.group = (_e) } #define G(_f, _g) { .flags = ((_f) | Group), .u.group = (_g) } -#define GD(_f, _g) { .flags = ((_f) | Group | GroupDual), .u.gdual = (_g) } +#define GD(_f, _g) { .flags = ((_f) | GroupDual), .u.gdual = (_g) } #define I(_f, _e) { .flags = (_f), .u.execute = (_e) } +#define II(_f, _e, _i) \ + { .flags = (_f), .u.execute = (_e), .intercept = x86_intercept_##_i } +#define IIP(_f, _e, _i, _p) \ + { .flags = (_f), .u.execute = (_e), .intercept = x86_intercept_##_i, \ + .check_perm = (_p) } +#define GP(_f, _g) { .flags = ((_f) | Prefix), .u.gprefix = (_g) } #define D2bv(_f) D((_f) | ByteOp), D(_f) +#define D2bvIP(_f, _i, _p) DIP((_f) | ByteOp, _i, _p), DIP(_f, _i, _p) #define I2bv(_f, _e) I((_f) | ByteOp, _e), I(_f, _e) -#define D6ALU(_f) D2bv((_f) | DstMem | SrcReg | ModRM), \ - D2bv(((_f) | DstReg | SrcMem | ModRM) & ~Lock), \ - D2bv(((_f) & ~Lock) | DstAcc | SrcImm) +#define I6ALU(_f, _e) I2bv((_f) | DstMem | SrcReg | ModRM, _e), \ + I2bv(((_f) | DstReg | SrcMem | ModRM) & ~Lock, _e), \ + I2bv(((_f) & ~Lock) | DstAcc | SrcImm, _e) +static struct opcode group7_rm1[] = { + DI(SrcNone | ModRM | Priv, monitor), + DI(SrcNone | ModRM | Priv, mwait), + N, N, N, N, N, N, +}; + +static struct opcode group7_rm3[] = { + DIP(SrcNone | ModRM | Prot | Priv, vmrun, check_svme_pa), + II(SrcNone | ModRM | Prot | VendorSpecific, em_vmmcall, vmmcall), + DIP(SrcNone | ModRM | Prot | Priv, vmload, check_svme_pa), + DIP(SrcNone | ModRM | Prot | Priv, vmsave, check_svme_pa), + DIP(SrcNone | ModRM | Prot | Priv, stgi, check_svme), + DIP(SrcNone | ModRM | Prot | Priv, clgi, check_svme), + DIP(SrcNone | ModRM | Prot | Priv, skinit, check_svme), + DIP(SrcNone | ModRM | Prot | Priv, invlpga, check_svme), +}; + +static struct opcode group7_rm7[] = { + N, + DIP(SrcNone | ModRM, rdtscp, check_rdtsc), + N, N, N, N, N, N, +}; static struct opcode group1[] = { - X7(D(Lock)), N + I(Lock, em_add), + I(Lock, em_or), + I(Lock, em_adc), + I(Lock, em_sbb), + I(Lock, em_and), + I(Lock, em_sub), + I(Lock, em_xor), + I(0, em_cmp), }; static struct opcode group1A[] = { @@ -2366,16 +3073,28 @@ static struct opcode group5[] = { D(SrcMem | ModRM | Stack), N, }; +static struct opcode group6[] = { + DI(ModRM | Prot, sldt), + DI(ModRM | Prot, str), + DI(ModRM | Prot | Priv, lldt), + DI(ModRM | Prot | Priv, ltr), + N, N, N, N, +}; + static struct group_dual group7 = { { - N, N, D(ModRM | SrcMem | Priv), D(ModRM | SrcMem | Priv), - D(SrcNone | ModRM | DstMem | Mov), N, - D(SrcMem16 | ModRM | Mov | Priv), - D(SrcMem | ModRM | ByteOp | Priv | NoAccess), + DI(ModRM | Mov | DstMem | Priv, sgdt), + DI(ModRM | Mov | DstMem | Priv, sidt), + II(ModRM | SrcMem | Priv, em_lgdt, lgdt), + II(ModRM | SrcMem | Priv, em_lidt, lidt), + II(SrcNone | ModRM | DstMem | Mov, em_smsw, smsw), N, + II(SrcMem16 | ModRM | Mov | Priv, em_lmsw, lmsw), + II(SrcMem | ModRM | ByteOp | Priv | NoAccess, em_invlpg, invlpg), }, { - D(SrcNone | ModRM | Priv | VendorSpecific), N, - N, D(SrcNone | ModRM | Priv | VendorSpecific), - D(SrcNone | ModRM | DstMem | Mov), N, - D(SrcMem16 | ModRM | Mov | Priv), N, + I(SrcNone | ModRM | Priv | VendorSpecific, em_vmcall), + EXT(0, group7_rm1), + N, EXT(0, group7_rm3), + II(SrcNone | ModRM | DstMem | Mov, em_smsw, smsw), N, + II(SrcMem16 | ModRM | Mov | Priv, em_lmsw, lmsw), EXT(0, group7_rm7), } }; static struct opcode group8[] = { @@ -2394,35 +3113,40 @@ static struct opcode group11[] = { I(DstMem | SrcImm | ModRM | Mov, em_mov), X7(D(Undefined)), }; +static struct gprefix pfx_0f_6f_0f_7f = { + N, N, N, I(Sse, em_movdqu), +}; + static struct opcode opcode_table[256] = { /* 0x00 - 0x07 */ - D6ALU(Lock), + I6ALU(Lock, em_add), D(ImplicitOps | Stack | No64), D(ImplicitOps | Stack | No64), /* 0x08 - 0x0F */ - D6ALU(Lock), + I6ALU(Lock, em_or), D(ImplicitOps | Stack | No64), N, /* 0x10 - 0x17 */ - D6ALU(Lock), + I6ALU(Lock, em_adc), D(ImplicitOps | Stack | No64), D(ImplicitOps | Stack | No64), /* 0x18 - 0x1F */ - D6ALU(Lock), + I6ALU(Lock, em_sbb), D(ImplicitOps | Stack | No64), D(ImplicitOps | Stack | No64), /* 0x20 - 0x27 */ - D6ALU(Lock), N, N, + I6ALU(Lock, em_and), N, N, /* 0x28 - 0x2F */ - D6ALU(Lock), N, I(ByteOp | DstAcc | No64, em_das), + I6ALU(Lock, em_sub), N, I(ByteOp | DstAcc | No64, em_das), /* 0x30 - 0x37 */ - D6ALU(Lock), N, N, + I6ALU(Lock, em_xor), N, N, /* 0x38 - 0x3F */ - D6ALU(0), N, N, + I6ALU(0, em_cmp), N, N, /* 0x40 - 0x4F */ X16(D(DstReg)), /* 0x50 - 0x57 */ X8(I(SrcReg | Stack, em_push)), /* 0x58 - 0x5F */ - X8(D(DstReg | Stack)), + X8(I(DstReg | Stack, em_pop)), /* 0x60 - 0x67 */ - D(ImplicitOps | Stack | No64), D(ImplicitOps | Stack | No64), + I(ImplicitOps | Stack | No64, em_pusha), + I(ImplicitOps | Stack | No64, em_popa), N, D(DstReg | SrcMem32 | ModRM | Mov) /* movsxd (x86/64) */ , N, N, N, N, /* 0x68 - 0x6F */ @@ -2430,8 +3154,8 @@ static struct opcode opcode_table[256] = { I(DstReg | SrcMem | ModRM | Src2Imm, em_imul_3op), I(SrcImmByte | Mov | Stack, em_push), I(DstReg | SrcMem | ModRM | Src2ImmByte, em_imul_3op), - D2bv(DstDI | Mov | String), /* insb, insw/insd */ - D2bv(SrcSI | ImplicitOps | String), /* outsb, outsw/outsd */ + D2bvIP(DstDI | Mov | String, ins, check_perm_in), /* insb, insw/insd */ + D2bvIP(SrcSI | ImplicitOps | String, outs, check_perm_out), /* outsb, outsw/outsd */ /* 0x70 - 0x7F */ X16(D(SrcImmByte)), /* 0x80 - 0x87 */ @@ -2446,21 +3170,22 @@ static struct opcode opcode_table[256] = { D(DstMem | SrcNone | ModRM | Mov), D(ModRM | SrcMem | NoAccess | DstReg), D(ImplicitOps | SrcMem16 | ModRM), G(0, group1A), /* 0x90 - 0x97 */ - X8(D(SrcAcc | DstReg)), + DI(SrcAcc | DstReg, pause), X7(D(SrcAcc | DstReg)), /* 0x98 - 0x9F */ D(DstAcc | SrcNone), I(ImplicitOps | SrcAcc, em_cwd), I(SrcImmFAddr | No64, em_call_far), N, - D(ImplicitOps | Stack), D(ImplicitOps | Stack), N, N, + II(ImplicitOps | Stack, em_pushf, pushf), + II(ImplicitOps | Stack, em_popf, popf), N, N, /* 0xA0 - 0xA7 */ I2bv(DstAcc | SrcMem | Mov | MemAbs, em_mov), I2bv(DstMem | SrcAcc | Mov | MemAbs, em_mov), I2bv(SrcSI | DstDI | Mov | String, em_mov), - D2bv(SrcSI | DstDI | String), + I2bv(SrcSI | DstDI | String, em_cmp), /* 0xA8 - 0xAF */ D2bv(DstAcc | SrcImm), I2bv(SrcAcc | DstDI | Mov | String, em_mov), I2bv(SrcSI | DstAcc | Mov | String, em_mov), - D2bv(SrcAcc | DstDI | String), + I2bv(SrcAcc | DstDI | String, em_cmp), /* 0xB0 - 0xB7 */ X8(I(ByteOp | DstReg | SrcImm | Mov, em_mov)), /* 0xB8 - 0xBF */ @@ -2473,7 +3198,8 @@ static struct opcode opcode_table[256] = { G(ByteOp, group11), G(0, group11), /* 0xC8 - 0xCF */ N, N, N, D(ImplicitOps | Stack), - D(ImplicitOps), D(SrcImmByte), D(ImplicitOps | No64), D(ImplicitOps), + D(ImplicitOps), DI(SrcImmByte, intn), + D(ImplicitOps | No64), DI(ImplicitOps, iret), /* 0xD0 - 0xD7 */ D2bv(DstMem | SrcOne | ModRM), D2bv(DstMem | ModRM), N, N, N, N, @@ -2481,14 +3207,17 @@ static struct opcode opcode_table[256] = { N, N, N, N, N, N, N, N, /* 0xE0 - 0xE7 */ X4(D(SrcImmByte)), - D2bv(SrcImmUByte | DstAcc), D2bv(SrcAcc | DstImmUByte), + D2bvIP(SrcImmUByte | DstAcc, in, check_perm_in), + D2bvIP(SrcAcc | DstImmUByte, out, check_perm_out), /* 0xE8 - 0xEF */ D(SrcImm | Stack), D(SrcImm | ImplicitOps), D(SrcImmFAddr | No64), D(SrcImmByte | ImplicitOps), - D2bv(SrcNone | DstAcc), D2bv(SrcAcc | ImplicitOps), + D2bvIP(SrcNone | DstAcc, in, check_perm_in), + D2bvIP(SrcAcc | ImplicitOps, out, check_perm_out), /* 0xF0 - 0xF7 */ - N, N, N, N, - D(ImplicitOps | Priv), D(ImplicitOps), G(ByteOp, group3), G(0, group3), + N, DI(ImplicitOps, icebp), N, N, + DI(ImplicitOps | Priv, hlt), D(ImplicitOps), + G(ByteOp, group3), G(0, group3), /* 0xF8 - 0xFF */ D(ImplicitOps), D(ImplicitOps), D(ImplicitOps), D(ImplicitOps), D(ImplicitOps), D(ImplicitOps), G(0, group4), G(0, group5), @@ -2496,20 +3225,24 @@ static struct opcode opcode_table[256] = { static struct opcode twobyte_table[256] = { /* 0x00 - 0x0F */ - N, GD(0, &group7), N, N, - N, D(ImplicitOps | VendorSpecific), D(ImplicitOps | Priv), N, - D(ImplicitOps | Priv), D(ImplicitOps | Priv), N, N, + G(0, group6), GD(0, &group7), N, N, + N, D(ImplicitOps | VendorSpecific), DI(ImplicitOps | Priv, clts), N, + DI(ImplicitOps | Priv, invd), DI(ImplicitOps | Priv, wbinvd), N, N, N, D(ImplicitOps | ModRM), N, N, /* 0x10 - 0x1F */ N, N, N, N, N, N, N, N, D(ImplicitOps | ModRM), N, N, N, N, N, N, N, /* 0x20 - 0x2F */ - D(ModRM | DstMem | Priv | Op3264), D(ModRM | DstMem | Priv | Op3264), - D(ModRM | SrcMem | Priv | Op3264), D(ModRM | SrcMem | Priv | Op3264), + DIP(ModRM | DstMem | Priv | Op3264, cr_read, check_cr_read), + DIP(ModRM | DstMem | Priv | Op3264, dr_read, check_dr_read), + DIP(ModRM | SrcMem | Priv | Op3264, cr_write, check_cr_write), + DIP(ModRM | SrcMem | Priv | Op3264, dr_write, check_dr_write), N, N, N, N, N, N, N, N, N, N, N, N, /* 0x30 - 0x3F */ - D(ImplicitOps | Priv), I(ImplicitOps, em_rdtsc), - D(ImplicitOps | Priv), N, + DI(ImplicitOps | Priv, wrmsr), + IIP(ImplicitOps, em_rdtsc, rdtsc, check_rdtsc), + DI(ImplicitOps | Priv, rdmsr), + DIP(ImplicitOps | Priv, rdpmc, check_rdpmc), D(ImplicitOps | VendorSpecific), D(ImplicitOps | Priv | VendorSpecific), N, N, N, N, N, N, N, N, N, N, @@ -2518,21 +3251,27 @@ static struct opcode twobyte_table[256] = { /* 0x50 - 0x5F */ N, N, N, N, N, N, N, N, N, N, N, N, N, N, N, N, /* 0x60 - 0x6F */ - N, N, N, N, N, N, N, N, N, N, N, N, N, N, N, N, + N, N, N, N, + N, N, N, N, + N, N, N, N, + N, N, N, GP(SrcMem | DstReg | ModRM | Mov, &pfx_0f_6f_0f_7f), /* 0x70 - 0x7F */ - N, N, N, N, N, N, N, N, N, N, N, N, N, N, N, N, + N, N, N, N, + N, N, N, N, + N, N, N, N, + N, N, N, GP(SrcReg | DstMem | ModRM | Mov, &pfx_0f_6f_0f_7f), /* 0x80 - 0x8F */ X16(D(SrcImm)), /* 0x90 - 0x9F */ X16(D(ByteOp | DstMem | SrcNone | ModRM| Mov)), /* 0xA0 - 0xA7 */ D(ImplicitOps | Stack), D(ImplicitOps | Stack), - N, D(DstMem | SrcReg | ModRM | BitOp), + DI(ImplicitOps, cpuid), D(DstMem | SrcReg | ModRM | BitOp), D(DstMem | SrcReg | Src2ImmByte | ModRM), D(DstMem | SrcReg | Src2CL | ModRM), N, N, /* 0xA8 - 0xAF */ D(ImplicitOps | Stack), D(ImplicitOps | Stack), - N, D(DstMem | SrcReg | ModRM | BitOp | Lock), + DI(ImplicitOps, rsm), D(DstMem | SrcReg | ModRM | BitOp | Lock), D(DstMem | SrcReg | Src2ImmByte | ModRM), D(DstMem | SrcReg | Src2CL | ModRM), D(ModRM), I(DstReg | SrcMem | ModRM, em_imul), @@ -2564,10 +3303,13 @@ static struct opcode twobyte_table[256] = { #undef G #undef GD #undef I +#undef GP +#undef EXT #undef D2bv +#undef D2bvIP #undef I2bv -#undef D6ALU +#undef I6ALU static unsigned imm_size(struct decode_cache *c) { @@ -2625,8 +3367,9 @@ x86_decode_insn(struct x86_emulate_ctxt *ctxt, void *insn, int insn_len) struct decode_cache *c = &ctxt->decode; int rc = X86EMUL_CONTINUE; int mode = ctxt->mode; - int def_op_bytes, def_ad_bytes, dual, goffset; - struct opcode opcode, *g_mod012, *g_mod3; + int def_op_bytes, def_ad_bytes, goffset, simd_prefix; + bool op_prefix = false; + struct opcode opcode; struct operand memop = { .type = OP_NONE }; c->eip = ctxt->eip; @@ -2634,7 +3377,6 @@ x86_decode_insn(struct x86_emulate_ctxt *ctxt, void *insn, int insn_len) c->fetch.end = c->fetch.start + insn_len; if (insn_len > 0) memcpy(c->fetch.data, insn, insn_len); - ctxt->cs_base = seg_base(ctxt, ops, VCPU_SREG_CS); switch (mode) { case X86EMUL_MODE_REAL: @@ -2662,6 +3404,7 @@ x86_decode_insn(struct x86_emulate_ctxt *ctxt, void *insn, int insn_len) for (;;) { switch (c->b = insn_fetch(u8, 1, c->eip)) { case 0x66: /* operand-size override */ + op_prefix = true; /* switch between 2/4 bytes */ c->op_bytes = def_op_bytes ^ 6; break; @@ -2692,10 +3435,8 @@ x86_decode_insn(struct x86_emulate_ctxt *ctxt, void *insn, int insn_len) c->lock_prefix = 1; break; case 0xf2: /* REPNE/REPNZ */ - c->rep_prefix = REPNE_PREFIX; - break; case 0xf3: /* REP/REPE/REPZ */ - c->rep_prefix = REPE_PREFIX; + c->rep_prefix = c->b; break; default: goto done_prefixes; @@ -2722,29 +3463,49 @@ done_prefixes: } c->d = opcode.flags; - if (c->d & Group) { - dual = c->d & GroupDual; - c->modrm = insn_fetch(u8, 1, c->eip); - --c->eip; - - if (c->d & GroupDual) { - g_mod012 = opcode.u.gdual->mod012; - g_mod3 = opcode.u.gdual->mod3; - } else - g_mod012 = g_mod3 = opcode.u.group; - - c->d &= ~(Group | GroupDual); - - goffset = (c->modrm >> 3) & 7; + while (c->d & GroupMask) { + switch (c->d & GroupMask) { + case Group: + c->modrm = insn_fetch(u8, 1, c->eip); + --c->eip; + goffset = (c->modrm >> 3) & 7; + opcode = opcode.u.group[goffset]; + break; + case GroupDual: + c->modrm = insn_fetch(u8, 1, c->eip); + --c->eip; + goffset = (c->modrm >> 3) & 7; + if ((c->modrm >> 6) == 3) + opcode = opcode.u.gdual->mod3[goffset]; + else + opcode = opcode.u.gdual->mod012[goffset]; + break; + case RMExt: + goffset = c->modrm & 7; + opcode = opcode.u.group[goffset]; + break; + case Prefix: + if (c->rep_prefix && op_prefix) + return X86EMUL_UNHANDLEABLE; + simd_prefix = op_prefix ? 0x66 : c->rep_prefix; + switch (simd_prefix) { + case 0x00: opcode = opcode.u.gprefix->pfx_no; break; + case 0x66: opcode = opcode.u.gprefix->pfx_66; break; + case 0xf2: opcode = opcode.u.gprefix->pfx_f2; break; + case 0xf3: opcode = opcode.u.gprefix->pfx_f3; break; + } + break; + default: + return X86EMUL_UNHANDLEABLE; + } - if ((c->modrm >> 6) == 3) - opcode = g_mod3[goffset]; - else - opcode = g_mod012[goffset]; + c->d &= ~GroupMask; c->d |= opcode.flags; } c->execute = opcode.u.execute; + c->check_perm = opcode.check_perm; + c->intercept = opcode.intercept; /* Unrecognised? */ if (c->d == 0 || (c->d & Undefined)) @@ -2763,6 +3524,9 @@ done_prefixes: c->op_bytes = 4; } + if (c->d & Sse) + c->op_bytes = 16; + /* ModRM and SIB bytes. */ if (c->d & ModRM) { rc = decode_modrm(ctxt, ops, &memop); @@ -2776,7 +3540,7 @@ done_prefixes: if (!c->has_seg_override) set_seg_override(c, VCPU_SREG_DS); - memop.addr.mem.seg = seg_override(ctxt, ops, c); + memop.addr.mem.seg = seg_override(ctxt, c); if (memop.type == OP_MEM && c->ad_bytes != 8) memop.addr.mem.ea = (u32)memop.addr.mem.ea; @@ -2792,7 +3556,7 @@ done_prefixes: case SrcNone: break; case SrcReg: - decode_register_operand(&c->src, c, 0); + decode_register_operand(ctxt, &c->src, c, 0); break; case SrcMem16: memop.bytes = 2; @@ -2836,7 +3600,7 @@ done_prefixes: c->src.bytes = (c->d & ByteOp) ? 1 : c->op_bytes; c->src.addr.mem.ea = register_address(c, c->regs[VCPU_REGS_RSI]); - c->src.addr.mem.seg = seg_override(ctxt, ops, c), + c->src.addr.mem.seg = seg_override(ctxt, c); c->src.val = 0; break; case SrcImmFAddr: @@ -2883,7 +3647,7 @@ done_prefixes: /* Decode and fetch the destination operand: register or memory. */ switch (c->d & DstMask) { case DstReg: - decode_register_operand(&c->dst, c, + decode_register_operand(ctxt, &c->dst, c, c->twobyte && (c->b == 0xb6 || c->b == 0xb7)); break; case DstImmUByte: @@ -2926,7 +3690,7 @@ done_prefixes: } done: - return (rc == X86EMUL_UNHANDLEABLE) ? -1 : 0; + return (rc == X86EMUL_UNHANDLEABLE) ? EMULATION_FAILED : EMULATION_OK; } static bool string_insn_completed(struct x86_emulate_ctxt *ctxt) @@ -2979,12 +3743,51 @@ x86_emulate_insn(struct x86_emulate_ctxt *ctxt) goto done; } + if ((c->d & Sse) + && ((ops->get_cr(ctxt, 0) & X86_CR0_EM) + || !(ops->get_cr(ctxt, 4) & X86_CR4_OSFXSR))) { + rc = emulate_ud(ctxt); + goto done; + } + + if ((c->d & Sse) && (ops->get_cr(ctxt, 0) & X86_CR0_TS)) { + rc = emulate_nm(ctxt); + goto done; + } + + if (unlikely(ctxt->guest_mode) && c->intercept) { + rc = emulator_check_intercept(ctxt, c->intercept, + X86_ICPT_PRE_EXCEPT); + if (rc != X86EMUL_CONTINUE) + goto done; + } + /* Privileged instruction can be executed only in CPL=0 */ - if ((c->d & Priv) && ops->cpl(ctxt->vcpu)) { + if ((c->d & Priv) && ops->cpl(ctxt)) { rc = emulate_gp(ctxt, 0); goto done; } + /* Instruction can only be executed in protected mode */ + if ((c->d & Prot) && !(ctxt->mode & X86EMUL_MODE_PROT)) { + rc = emulate_ud(ctxt); + goto done; + } + + /* Do instruction specific permission checks */ + if (c->check_perm) { + rc = c->check_perm(ctxt); + if (rc != X86EMUL_CONTINUE) + goto done; + } + + if (unlikely(ctxt->guest_mode) && c->intercept) { + rc = emulator_check_intercept(ctxt, c->intercept, + X86_ICPT_POST_EXCEPT); + if (rc != X86EMUL_CONTINUE) + goto done; + } + if (c->rep_prefix && (c->d & String)) { /* All REP prefixes have the same first termination condition */ if (address_mask(c, c->regs[VCPU_REGS_RCX]) == 0) { @@ -2994,16 +3797,16 @@ x86_emulate_insn(struct x86_emulate_ctxt *ctxt) } if ((c->src.type == OP_MEM) && !(c->d & NoAccess)) { - rc = read_emulated(ctxt, ops, linear(ctxt, c->src.addr.mem), - c->src.valptr, c->src.bytes); + rc = segmented_read(ctxt, c->src.addr.mem, + c->src.valptr, c->src.bytes); if (rc != X86EMUL_CONTINUE) goto done; c->src.orig_val64 = c->src.val64; } if (c->src2.type == OP_MEM) { - rc = read_emulated(ctxt, ops, linear(ctxt, c->src2.addr.mem), - &c->src2.val, c->src2.bytes); + rc = segmented_read(ctxt, c->src2.addr.mem, + &c->src2.val, c->src2.bytes); if (rc != X86EMUL_CONTINUE) goto done; } @@ -3014,7 +3817,7 @@ x86_emulate_insn(struct x86_emulate_ctxt *ctxt) if ((c->dst.type == OP_MEM) && !(c->d & Mov)) { /* optimisation - avoid slow emulated read if Mov */ - rc = read_emulated(ctxt, ops, linear(ctxt, c->dst.addr.mem), + rc = segmented_read(ctxt, c->dst.addr.mem, &c->dst.val, c->dst.bytes); if (rc != X86EMUL_CONTINUE) goto done; @@ -3023,6 +3826,13 @@ x86_emulate_insn(struct x86_emulate_ctxt *ctxt) special_insn: + if (unlikely(ctxt->guest_mode) && c->intercept) { + rc = emulator_check_intercept(ctxt, c->intercept, + X86_ICPT_POST_MEMACCESS); + if (rc != X86EMUL_CONTINUE) + goto done; + } + if (c->execute) { rc = c->execute(ctxt); if (rc != X86EMUL_CONTINUE) @@ -3034,75 +3844,33 @@ special_insn: goto twobyte_insn; switch (c->b) { - case 0x00 ... 0x05: - add: /* add */ - emulate_2op_SrcV("add", c->src, c->dst, ctxt->eflags); - break; case 0x06: /* push es */ - emulate_push_sreg(ctxt, ops, VCPU_SREG_ES); + rc = emulate_push_sreg(ctxt, ops, VCPU_SREG_ES); break; case 0x07: /* pop es */ rc = emulate_pop_sreg(ctxt, ops, VCPU_SREG_ES); break; - case 0x08 ... 0x0d: - or: /* or */ - emulate_2op_SrcV("or", c->src, c->dst, ctxt->eflags); - break; case 0x0e: /* push cs */ - emulate_push_sreg(ctxt, ops, VCPU_SREG_CS); - break; - case 0x10 ... 0x15: - adc: /* adc */ - emulate_2op_SrcV("adc", c->src, c->dst, ctxt->eflags); + rc = emulate_push_sreg(ctxt, ops, VCPU_SREG_CS); break; case 0x16: /* push ss */ - emulate_push_sreg(ctxt, ops, VCPU_SREG_SS); + rc = emulate_push_sreg(ctxt, ops, VCPU_SREG_SS); break; case 0x17: /* pop ss */ rc = emulate_pop_sreg(ctxt, ops, VCPU_SREG_SS); break; - case 0x18 ... 0x1d: - sbb: /* sbb */ - emulate_2op_SrcV("sbb", c->src, c->dst, ctxt->eflags); - break; case 0x1e: /* push ds */ - emulate_push_sreg(ctxt, ops, VCPU_SREG_DS); + rc = emulate_push_sreg(ctxt, ops, VCPU_SREG_DS); break; case 0x1f: /* pop ds */ rc = emulate_pop_sreg(ctxt, ops, VCPU_SREG_DS); break; - case 0x20 ... 0x25: - and: /* and */ - emulate_2op_SrcV("and", c->src, c->dst, ctxt->eflags); - break; - case 0x28 ... 0x2d: - sub: /* sub */ - emulate_2op_SrcV("sub", c->src, c->dst, ctxt->eflags); - break; - case 0x30 ... 0x35: - xor: /* xor */ - emulate_2op_SrcV("xor", c->src, c->dst, ctxt->eflags); - break; - case 0x38 ... 0x3d: - cmp: /* cmp */ - emulate_2op_SrcV("cmp", c->src, c->dst, ctxt->eflags); - break; case 0x40 ... 0x47: /* inc r16/r32 */ emulate_1op("inc", c->dst, ctxt->eflags); break; case 0x48 ... 0x4f: /* dec r16/r32 */ emulate_1op("dec", c->dst, ctxt->eflags); break; - case 0x58 ... 0x5f: /* pop reg */ - pop_instruction: - rc = emulate_pop(ctxt, ops, &c->dst.val, c->op_bytes); - break; - case 0x60: /* pusha */ - rc = emulate_pusha(ctxt, ops); - break; - case 0x61: /* popa */ - rc = emulate_popa(ctxt, ops); - break; case 0x63: /* movsxd */ if (ctxt->mode != X86EMUL_MODE_PROT64) goto cannot_emulate; @@ -3121,26 +3889,6 @@ special_insn: if (test_cc(c->b, ctxt->eflags)) jmp_rel(c, c->src.val); break; - case 0x80 ... 0x83: /* Grp1 */ - switch (c->modrm_reg) { - case 0: - goto add; - case 1: - goto or; - case 2: - goto adc; - case 3: - goto sbb; - case 4: - goto and; - case 5: - goto sub; - case 6: - goto xor; - case 7: - goto cmp; - } - break; case 0x84 ... 0x85: test: emulate_2op_SrcV("test", c->src, c->dst, ctxt->eflags); @@ -3162,7 +3910,7 @@ special_insn: rc = emulate_ud(ctxt); goto done; } - c->dst.val = ops->get_segment_selector(c->modrm_reg, ctxt->vcpu); + c->dst.val = get_segment_selector(ctxt, c->modrm_reg); break; case 0x8d: /* lea r16/r32, m */ c->dst.val = c->src.addr.mem.ea; @@ -3187,7 +3935,7 @@ special_insn: break; } case 0x8f: /* pop (sole member of Grp1a) */ - rc = emulate_grp1a(ctxt, ops); + rc = em_grp1a(ctxt); break; case 0x90 ... 0x97: /* nop / xchg reg, rax */ if (c->dst.addr.reg == &c->regs[VCPU_REGS_RAX]) @@ -3200,31 +3948,17 @@ special_insn: case 8: c->dst.val = (s32)c->dst.val; break; } break; - case 0x9c: /* pushf */ - c->src.val = (unsigned long) ctxt->eflags; - emulate_push(ctxt, ops); - break; - case 0x9d: /* popf */ - c->dst.type = OP_REG; - c->dst.addr.reg = &ctxt->eflags; - c->dst.bytes = c->op_bytes; - rc = emulate_popf(ctxt, ops, &c->dst.val, c->op_bytes); - break; - case 0xa6 ... 0xa7: /* cmps */ - c->dst.type = OP_NONE; /* Disable writeback. */ - goto cmp; case 0xa8 ... 0xa9: /* test ax, imm */ goto test; - case 0xae ... 0xaf: /* scas */ - goto cmp; case 0xc0 ... 0xc1: - emulate_grp2(ctxt); + rc = em_grp2(ctxt); break; case 0xc3: /* ret */ c->dst.type = OP_REG; c->dst.addr.reg = &c->eip; c->dst.bytes = c->op_bytes; - goto pop_instruction; + rc = em_pop(ctxt); + break; case 0xc4: /* les */ rc = emulate_load_segment(ctxt, ops, VCPU_SREG_ES); break; @@ -3252,11 +3986,11 @@ special_insn: rc = emulate_iret(ctxt, ops); break; case 0xd0 ... 0xd1: /* Grp2 */ - emulate_grp2(ctxt); + rc = em_grp2(ctxt); break; case 0xd2 ... 0xd3: /* Grp2 */ c->src.val = c->regs[VCPU_REGS_RCX]; - emulate_grp2(ctxt); + rc = em_grp2(ctxt); break; case 0xe0 ... 0xe2: /* loop/loopz/loopnz */ register_address_increment(c, &c->regs[VCPU_REGS_RCX], -1); @@ -3278,23 +4012,14 @@ special_insn: long int rel = c->src.val; c->src.val = (unsigned long) c->eip; jmp_rel(c, rel); - emulate_push(ctxt, ops); + rc = em_push(ctxt); break; } case 0xe9: /* jmp rel */ goto jmp; - case 0xea: { /* jmp far */ - unsigned short sel; - jump_far: - memcpy(&sel, c->src.valptr + c->op_bytes, 2); - - if (load_segment_descriptor(ctxt, ops, sel, VCPU_SREG_CS)) - goto done; - - c->eip = 0; - memcpy(&c->eip, c->src.valptr, c->op_bytes); + case 0xea: /* jmp far */ + rc = em_jmp_far(ctxt); break; - } case 0xeb: jmp: /* jmp rel short */ jmp_rel(c, c->src.val); @@ -3304,11 +4029,6 @@ special_insn: case 0xed: /* in (e/r)ax,dx */ c->src.val = c->regs[VCPU_REGS_RDX]; do_io_in: - c->dst.bytes = min(c->dst.bytes, 4u); - if (!emulator_io_permited(ctxt, ops, c->src.val, c->dst.bytes)) { - rc = emulate_gp(ctxt, 0); - goto done; - } if (!pio_in_emulated(ctxt, ops, c->dst.bytes, c->src.val, &c->dst.val)) goto done; /* IO is needed */ @@ -3317,25 +4037,19 @@ special_insn: case 0xef: /* out dx,(e/r)ax */ c->dst.val = c->regs[VCPU_REGS_RDX]; do_io_out: - c->src.bytes = min(c->src.bytes, 4u); - if (!emulator_io_permited(ctxt, ops, c->dst.val, - c->src.bytes)) { - rc = emulate_gp(ctxt, 0); - goto done; - } - ops->pio_out_emulated(c->src.bytes, c->dst.val, - &c->src.val, 1, ctxt->vcpu); + ops->pio_out_emulated(ctxt, c->src.bytes, c->dst.val, + &c->src.val, 1); c->dst.type = OP_NONE; /* Disable writeback. */ break; case 0xf4: /* hlt */ - ctxt->vcpu->arch.halt_request = 1; + ctxt->ops->halt(ctxt); break; case 0xf5: /* cmc */ /* complement carry flag from eflags reg */ ctxt->eflags ^= EFLG_CF; break; case 0xf6 ... 0xf7: /* Grp3 */ - rc = emulate_grp3(ctxt, ops); + rc = em_grp3(ctxt); break; case 0xf8: /* clc */ ctxt->eflags &= ~EFLG_CF; @@ -3366,13 +4080,11 @@ special_insn: ctxt->eflags |= EFLG_DF; break; case 0xfe: /* Grp4 */ - grp45: - rc = emulate_grp45(ctxt, ops); + rc = em_grp45(ctxt); break; case 0xff: /* Grp5 */ - if (c->modrm_reg == 5) - goto jump_far; - goto grp45; + rc = em_grp45(ctxt); + break; default: goto cannot_emulate; } @@ -3381,7 +4093,7 @@ special_insn: goto done; writeback: - rc = writeback(ctxt, ops); + rc = writeback(ctxt); if (rc != X86EMUL_CONTINUE) goto done; @@ -3392,7 +4104,7 @@ writeback: c->dst.type = saved_dst_type; if ((c->d & SrcMask) == SrcSI) - string_addr_inc(ctxt, seg_override(ctxt, ops, c), + string_addr_inc(ctxt, seg_override(ctxt, c), VCPU_REGS_RSI, &c->src); if ((c->d & DstMask) == DstDI) @@ -3427,115 +4139,34 @@ writeback: done: if (rc == X86EMUL_PROPAGATE_FAULT) ctxt->have_exception = true; + if (rc == X86EMUL_INTERCEPTED) + return EMULATION_INTERCEPTED; + return (rc == X86EMUL_UNHANDLEABLE) ? EMULATION_FAILED : EMULATION_OK; twobyte_insn: switch (c->b) { - case 0x01: /* lgdt, lidt, lmsw */ - switch (c->modrm_reg) { - u16 size; - unsigned long address; - - case 0: /* vmcall */ - if (c->modrm_mod != 3 || c->modrm_rm != 1) - goto cannot_emulate; - - rc = kvm_fix_hypercall(ctxt->vcpu); - if (rc != X86EMUL_CONTINUE) - goto done; - - /* Let the processor re-execute the fixed hypercall */ - c->eip = ctxt->eip; - /* Disable writeback. */ - c->dst.type = OP_NONE; - break; - case 2: /* lgdt */ - rc = read_descriptor(ctxt, ops, c->src.addr.mem, - &size, &address, c->op_bytes); - if (rc != X86EMUL_CONTINUE) - goto done; - realmode_lgdt(ctxt->vcpu, size, address); - /* Disable writeback. */ - c->dst.type = OP_NONE; - break; - case 3: /* lidt/vmmcall */ - if (c->modrm_mod == 3) { - switch (c->modrm_rm) { - case 1: - rc = kvm_fix_hypercall(ctxt->vcpu); - break; - default: - goto cannot_emulate; - } - } else { - rc = read_descriptor(ctxt, ops, c->src.addr.mem, - &size, &address, - c->op_bytes); - if (rc != X86EMUL_CONTINUE) - goto done; - realmode_lidt(ctxt->vcpu, size, address); - } - /* Disable writeback. */ - c->dst.type = OP_NONE; - break; - case 4: /* smsw */ - c->dst.bytes = 2; - c->dst.val = ops->get_cr(0, ctxt->vcpu); - break; - case 6: /* lmsw */ - ops->set_cr(0, (ops->get_cr(0, ctxt->vcpu) & ~0x0eul) | - (c->src.val & 0x0f), ctxt->vcpu); - c->dst.type = OP_NONE; - break; - case 5: /* not defined */ - emulate_ud(ctxt); - rc = X86EMUL_PROPAGATE_FAULT; - goto done; - case 7: /* invlpg*/ - emulate_invlpg(ctxt->vcpu, - linear(ctxt, c->src.addr.mem)); - /* Disable writeback. */ - c->dst.type = OP_NONE; - break; - default: - goto cannot_emulate; - } - break; case 0x05: /* syscall */ rc = emulate_syscall(ctxt, ops); break; case 0x06: - emulate_clts(ctxt->vcpu); + rc = em_clts(ctxt); break; case 0x09: /* wbinvd */ - kvm_emulate_wbinvd(ctxt->vcpu); + (ctxt->ops->wbinvd)(ctxt); break; case 0x08: /* invd */ case 0x0d: /* GrpP (prefetch) */ case 0x18: /* Grp16 (prefetch/nop) */ break; case 0x20: /* mov cr, reg */ - switch (c->modrm_reg) { - case 1: - case 5 ... 7: - case 9 ... 15: - emulate_ud(ctxt); - rc = X86EMUL_PROPAGATE_FAULT; - goto done; - } - c->dst.val = ops->get_cr(c->modrm_reg, ctxt->vcpu); + c->dst.val = ops->get_cr(ctxt, c->modrm_reg); break; case 0x21: /* mov from dr to reg */ - if ((ops->get_cr(4, ctxt->vcpu) & X86_CR4_DE) && - (c->modrm_reg == 4 || c->modrm_reg == 5)) { - emulate_ud(ctxt); - rc = X86EMUL_PROPAGATE_FAULT; - goto done; - } - ops->get_dr(c->modrm_reg, &c->dst.val, ctxt->vcpu); + ops->get_dr(ctxt, c->modrm_reg, &c->dst.val); break; case 0x22: /* mov reg, cr */ - if (ops->set_cr(c->modrm_reg, c->src.val, ctxt->vcpu)) { + if (ops->set_cr(ctxt, c->modrm_reg, c->src.val)) { emulate_gp(ctxt, 0); rc = X86EMUL_PROPAGATE_FAULT; goto done; @@ -3543,16 +4174,9 @@ twobyte_insn: c->dst.type = OP_NONE; break; case 0x23: /* mov from reg to dr */ - if ((ops->get_cr(4, ctxt->vcpu) & X86_CR4_DE) && - (c->modrm_reg == 4 || c->modrm_reg == 5)) { - emulate_ud(ctxt); - rc = X86EMUL_PROPAGATE_FAULT; - goto done; - } - - if (ops->set_dr(c->modrm_reg, c->src.val & + if (ops->set_dr(ctxt, c->modrm_reg, c->src.val & ((ctxt->mode == X86EMUL_MODE_PROT64) ? - ~0ULL : ~0U), ctxt->vcpu) < 0) { + ~0ULL : ~0U)) < 0) { /* #UD condition is already handled by the code above */ emulate_gp(ctxt, 0); rc = X86EMUL_PROPAGATE_FAULT; @@ -3565,7 +4189,7 @@ twobyte_insn: /* wrmsr */ msr_data = (u32)c->regs[VCPU_REGS_RAX] | ((u64)c->regs[VCPU_REGS_RDX] << 32); - if (ops->set_msr(ctxt->vcpu, c->regs[VCPU_REGS_RCX], msr_data)) { + if (ops->set_msr(ctxt, c->regs[VCPU_REGS_RCX], msr_data)) { emulate_gp(ctxt, 0); rc = X86EMUL_PROPAGATE_FAULT; goto done; @@ -3574,7 +4198,7 @@ twobyte_insn: break; case 0x32: /* rdmsr */ - if (ops->get_msr(ctxt->vcpu, c->regs[VCPU_REGS_RCX], &msr_data)) { + if (ops->get_msr(ctxt, c->regs[VCPU_REGS_RCX], &msr_data)) { emulate_gp(ctxt, 0); rc = X86EMUL_PROPAGATE_FAULT; goto done; @@ -3603,7 +4227,7 @@ twobyte_insn: c->dst.val = test_cc(c->b, ctxt->eflags); break; case 0xa0: /* push fs */ - emulate_push_sreg(ctxt, ops, VCPU_SREG_FS); + rc = emulate_push_sreg(ctxt, ops, VCPU_SREG_FS); break; case 0xa1: /* pop fs */ rc = emulate_pop_sreg(ctxt, ops, VCPU_SREG_FS); @@ -3620,7 +4244,7 @@ twobyte_insn: emulate_2op_cl("shld", c->src2, c->src, c->dst, ctxt->eflags); break; case 0xa8: /* push gs */ - emulate_push_sreg(ctxt, ops, VCPU_SREG_GS); + rc = emulate_push_sreg(ctxt, ops, VCPU_SREG_GS); break; case 0xa9: /* pop gs */ rc = emulate_pop_sreg(ctxt, ops, VCPU_SREG_GS); @@ -3727,7 +4351,7 @@ twobyte_insn: (u64) c->src.val; break; case 0xc7: /* Grp9 (cmpxchg8b) */ - rc = emulate_grp9(ctxt, ops); + rc = em_grp9(ctxt); break; default: goto cannot_emulate; @@ -3739,5 +4363,5 @@ twobyte_insn: goto writeback; cannot_emulate: - return -1; + return EMULATION_FAILED; } diff --git a/arch/x86/kvm/i8254.h b/arch/x86/kvm/i8254.h index 46d08ca0b48f..51a97426e791 100644 --- a/arch/x86/kvm/i8254.h +++ b/arch/x86/kvm/i8254.h @@ -33,7 +33,6 @@ struct kvm_kpit_state { }; struct kvm_pit { - unsigned long base_addresss; struct kvm_io_device dev; struct kvm_io_device speaker_dev; struct kvm *kvm; @@ -51,7 +50,6 @@ struct kvm_pit { #define KVM_MAX_PIT_INTR_INTERVAL HZ / 100 #define KVM_PIT_CHANNEL_MASK 0x3 -void kvm_inject_pit_timer_irqs(struct kvm_vcpu *vcpu); void kvm_pit_load_count(struct kvm *kvm, int channel, u32 val, int hpet_legacy_start); struct kvm_pit *kvm_create_pit(struct kvm *kvm, u32 flags); void kvm_free_pit(struct kvm *kvm); diff --git a/arch/x86/kvm/irq.h b/arch/x86/kvm/irq.h index ba910d149410..53e2d084bffb 100644 --- a/arch/x86/kvm/irq.h +++ b/arch/x86/kvm/irq.h @@ -75,7 +75,6 @@ struct kvm_pic *kvm_create_pic(struct kvm *kvm); void kvm_destroy_pic(struct kvm *kvm); int kvm_pic_read_irq(struct kvm *kvm); void kvm_pic_update_irq(struct kvm_pic *s); -void kvm_pic_clear_isr_ack(struct kvm *kvm); static inline struct kvm_pic *pic_irqchip(struct kvm *kvm) { @@ -100,7 +99,6 @@ void __kvm_migrate_apic_timer(struct kvm_vcpu *vcpu); void __kvm_migrate_pit_timer(struct kvm_vcpu *vcpu); void __kvm_migrate_timers(struct kvm_vcpu *vcpu); -int pit_has_pending_timer(struct kvm_vcpu *vcpu); int apic_has_pending_timer(struct kvm_vcpu *vcpu); #endif diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index 22fae7593ee7..bd14bb4c8594 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -1206,7 +1206,7 @@ static void nonpaging_invlpg(struct kvm_vcpu *vcpu, gva_t gva) static void nonpaging_update_pte(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp, u64 *spte, - const void *pte, unsigned long mmu_seq) + const void *pte) { WARN_ON(1); } @@ -3163,9 +3163,8 @@ static void mmu_pte_write_zap_pte(struct kvm_vcpu *vcpu, } static void mmu_pte_write_new_pte(struct kvm_vcpu *vcpu, - struct kvm_mmu_page *sp, - u64 *spte, - const void *new, unsigned long mmu_seq) + struct kvm_mmu_page *sp, u64 *spte, + const void *new) { if (sp->role.level != PT_PAGE_TABLE_LEVEL) { ++vcpu->kvm->stat.mmu_pde_zapped; @@ -3173,7 +3172,7 @@ static void mmu_pte_write_new_pte(struct kvm_vcpu *vcpu, } ++vcpu->kvm->stat.mmu_pte_updated; - vcpu->arch.mmu.update_pte(vcpu, sp, spte, new, mmu_seq); + vcpu->arch.mmu.update_pte(vcpu, sp, spte, new); } static bool need_remote_flush(u64 old, u64 new) @@ -3229,7 +3228,6 @@ void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa, struct kvm_mmu_page *sp; struct hlist_node *node; LIST_HEAD(invalid_list); - unsigned long mmu_seq; u64 entry, gentry, *spte; unsigned pte_size, page_offset, misaligned, quadrant, offset; int level, npte, invlpg_counter, r, flooded = 0; @@ -3271,9 +3269,6 @@ void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa, break; } - mmu_seq = vcpu->kvm->mmu_notifier_seq; - smp_rmb(); - spin_lock(&vcpu->kvm->mmu_lock); if (atomic_read(&vcpu->kvm->arch.invlpg_counter) != invlpg_counter) gentry = 0; @@ -3345,8 +3340,7 @@ void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa, if (gentry && !((sp->role.word ^ vcpu->arch.mmu.base_role.word) & mask.word)) - mmu_pte_write_new_pte(vcpu, sp, spte, &gentry, - mmu_seq); + mmu_pte_write_new_pte(vcpu, sp, spte, &gentry); if (!remote_flush && need_remote_flush(entry, *spte)) remote_flush = true; ++spte; @@ -3551,10 +3545,11 @@ static int kvm_mmu_remove_some_alloc_mmu_pages(struct kvm *kvm, return kvm_mmu_prepare_zap_page(kvm, page, invalid_list); } -static int mmu_shrink(struct shrinker *shrink, int nr_to_scan, gfp_t gfp_mask) +static int mmu_shrink(struct shrinker *shrink, struct shrink_control *sc) { struct kvm *kvm; struct kvm *kvm_freed = NULL; + int nr_to_scan = sc->nr_to_scan; if (nr_to_scan == 0) goto out; diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h index c6397795d865..6c4dc010c4cb 100644 --- a/arch/x86/kvm/paging_tmpl.h +++ b/arch/x86/kvm/paging_tmpl.h @@ -78,15 +78,19 @@ static gfn_t gpte_to_gfn_lvl(pt_element_t gpte, int lvl) return (gpte & PT_LVL_ADDR_MASK(lvl)) >> PAGE_SHIFT; } -static bool FNAME(cmpxchg_gpte)(struct kvm *kvm, - gfn_t table_gfn, unsigned index, - pt_element_t orig_pte, pt_element_t new_pte) +static int FNAME(cmpxchg_gpte)(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu, + pt_element_t __user *ptep_user, unsigned index, + pt_element_t orig_pte, pt_element_t new_pte) { + int npages; pt_element_t ret; pt_element_t *table; struct page *page; - page = gfn_to_page(kvm, table_gfn); + npages = get_user_pages_fast((unsigned long)ptep_user, 1, 1, &page); + /* Check if the user is doing something meaningless. */ + if (unlikely(npages != 1)) + return -EFAULT; table = kmap_atomic(page, KM_USER0); ret = CMPXCHG(&table[index], orig_pte, new_pte); @@ -117,6 +121,7 @@ static int FNAME(walk_addr_generic)(struct guest_walker *walker, gva_t addr, u32 access) { pt_element_t pte; + pt_element_t __user *ptep_user; gfn_t table_gfn; unsigned index, pt_access, uninitialized_var(pte_access); gpa_t pte_gpa; @@ -152,6 +157,9 @@ walk: pt_access = ACC_ALL; for (;;) { + gfn_t real_gfn; + unsigned long host_addr; + index = PT_INDEX(addr, walker->level); table_gfn = gpte_to_gfn(pte); @@ -160,43 +168,64 @@ walk: walker->table_gfn[walker->level - 1] = table_gfn; walker->pte_gpa[walker->level - 1] = pte_gpa; - if (kvm_read_guest_page_mmu(vcpu, mmu, table_gfn, &pte, - offset, sizeof(pte), - PFERR_USER_MASK|PFERR_WRITE_MASK)) { + real_gfn = mmu->translate_gpa(vcpu, gfn_to_gpa(table_gfn), + PFERR_USER_MASK|PFERR_WRITE_MASK); + if (unlikely(real_gfn == UNMAPPED_GVA)) { + present = false; + break; + } + real_gfn = gpa_to_gfn(real_gfn); + + host_addr = gfn_to_hva(vcpu->kvm, real_gfn); + if (unlikely(kvm_is_error_hva(host_addr))) { + present = false; + break; + } + + ptep_user = (pt_element_t __user *)((void *)host_addr + offset); + if (unlikely(__copy_from_user(&pte, ptep_user, sizeof(pte)))) { present = false; break; } trace_kvm_mmu_paging_element(pte, walker->level); - if (!is_present_gpte(pte)) { + if (unlikely(!is_present_gpte(pte))) { present = false; break; } - if (is_rsvd_bits_set(&vcpu->arch.mmu, pte, walker->level)) { + if (unlikely(is_rsvd_bits_set(&vcpu->arch.mmu, pte, + walker->level))) { rsvd_fault = true; break; } - if (write_fault && !is_writable_pte(pte)) - if (user_fault || is_write_protection(vcpu)) - eperm = true; + if (unlikely(write_fault && !is_writable_pte(pte) + && (user_fault || is_write_protection(vcpu)))) + eperm = true; - if (user_fault && !(pte & PT_USER_MASK)) + if (unlikely(user_fault && !(pte & PT_USER_MASK))) eperm = true; #if PTTYPE == 64 - if (fetch_fault && (pte & PT64_NX_MASK)) + if (unlikely(fetch_fault && (pte & PT64_NX_MASK))) eperm = true; #endif - if (!eperm && !rsvd_fault && !(pte & PT_ACCESSED_MASK)) { + if (!eperm && !rsvd_fault + && unlikely(!(pte & PT_ACCESSED_MASK))) { + int ret; trace_kvm_mmu_set_accessed_bit(table_gfn, index, sizeof(pte)); - if (FNAME(cmpxchg_gpte)(vcpu->kvm, table_gfn, - index, pte, pte|PT_ACCESSED_MASK)) + ret = FNAME(cmpxchg_gpte)(vcpu, mmu, ptep_user, index, + pte, pte|PT_ACCESSED_MASK); + if (unlikely(ret < 0)) { + present = false; + break; + } else if (ret) goto walk; + mark_page_dirty(vcpu->kvm, table_gfn); pte |= PT_ACCESSED_MASK; } @@ -241,17 +270,21 @@ walk: --walker->level; } - if (!present || eperm || rsvd_fault) + if (unlikely(!present || eperm || rsvd_fault)) goto error; - if (write_fault && !is_dirty_gpte(pte)) { - bool ret; + if (write_fault && unlikely(!is_dirty_gpte(pte))) { + int ret; trace_kvm_mmu_set_dirty_bit(table_gfn, index, sizeof(pte)); - ret = FNAME(cmpxchg_gpte)(vcpu->kvm, table_gfn, index, pte, - pte|PT_DIRTY_MASK); - if (ret) + ret = FNAME(cmpxchg_gpte)(vcpu, mmu, ptep_user, index, + pte, pte|PT_DIRTY_MASK); + if (unlikely(ret < 0)) { + present = false; + goto error; + } else if (ret) goto walk; + mark_page_dirty(vcpu->kvm, table_gfn); pte |= PT_DIRTY_MASK; walker->ptes[walker->level - 1] = pte; @@ -325,7 +358,7 @@ no_present: } static void FNAME(update_pte)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp, - u64 *spte, const void *pte, unsigned long mmu_seq) + u64 *spte, const void *pte) { pt_element_t gpte; unsigned pte_access; @@ -342,8 +375,6 @@ static void FNAME(update_pte)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp, kvm_release_pfn_clean(pfn); return; } - if (mmu_notifier_retry(vcpu, mmu_seq)) - return; /* * we call mmu_set_spte() with host_writable = true because that diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 6bb15d583e47..506e4fe23adc 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -63,6 +63,10 @@ MODULE_LICENSE("GPL"); #define DEBUGCTL_RESERVED_BITS (~(0x3fULL)) +#define TSC_RATIO_RSVD 0xffffff0000000000ULL +#define TSC_RATIO_MIN 0x0000000000000001ULL +#define TSC_RATIO_MAX 0x000000ffffffffffULL + static bool erratum_383_found __read_mostly; static const u32 host_save_user_msrs[] = { @@ -93,14 +97,6 @@ struct nested_state { /* A VMEXIT is required but not yet emulated */ bool exit_required; - /* - * If we vmexit during an instruction emulation we need this to restore - * the l1 guest rip after the emulation - */ - unsigned long vmexit_rip; - unsigned long vmexit_rsp; - unsigned long vmexit_rax; - /* cache for intercepts of the guest */ u32 intercept_cr; u32 intercept_dr; @@ -144,8 +140,13 @@ struct vcpu_svm { unsigned int3_injected; unsigned long int3_rip; u32 apf_reason; + + u64 tsc_ratio; }; +static DEFINE_PER_CPU(u64, current_tsc_ratio); +#define TSC_RATIO_DEFAULT 0x0100000000ULL + #define MSR_INVALID 0xffffffffU static struct svm_direct_access_msrs { @@ -190,6 +191,7 @@ static int nested_svm_intercept(struct vcpu_svm *svm); static int nested_svm_vmexit(struct vcpu_svm *svm); static int nested_svm_check_exception(struct vcpu_svm *svm, unsigned nr, bool has_error_code, u32 error_code); +static u64 __scale_tsc(u64 ratio, u64 tsc); enum { VMCB_INTERCEPTS, /* Intercept vectors, TSC offset, @@ -376,7 +378,6 @@ struct svm_cpu_data { }; static DEFINE_PER_CPU(struct svm_cpu_data *, svm_data); -static uint32_t svm_features; struct svm_init_data { int cpu; @@ -569,6 +570,10 @@ static int has_svm(void) static void svm_hardware_disable(void *garbage) { + /* Make sure we clean up behind us */ + if (static_cpu_has(X86_FEATURE_TSCRATEMSR)) + wrmsrl(MSR_AMD64_TSC_RATIO, TSC_RATIO_DEFAULT); + cpu_svm_disable(); } @@ -610,6 +615,11 @@ static int svm_hardware_enable(void *garbage) wrmsrl(MSR_VM_HSAVE_PA, page_to_pfn(sd->save_area) << PAGE_SHIFT); + if (static_cpu_has(X86_FEATURE_TSCRATEMSR)) { + wrmsrl(MSR_AMD64_TSC_RATIO, TSC_RATIO_DEFAULT); + __get_cpu_var(current_tsc_ratio) = TSC_RATIO_DEFAULT; + } + svm_init_erratum_383(); return 0; @@ -791,6 +801,23 @@ static __init int svm_hardware_setup(void) if (boot_cpu_has(X86_FEATURE_FXSR_OPT)) kvm_enable_efer_bits(EFER_FFXSR); + if (boot_cpu_has(X86_FEATURE_TSCRATEMSR)) { + u64 max; + + kvm_has_tsc_control = true; + + /* + * Make sure the user can only configure tsc_khz values that + * fit into a signed integer. + * A min value is not calculated needed because it will always + * be 1 on all machines and a value of 0 is used to disable + * tsc-scaling for the vcpu. + */ + max = min(0x7fffffffULL, __scale_tsc(tsc_khz, TSC_RATIO_MAX)); + + kvm_max_guest_tsc_khz = max; + } + if (nested) { printk(KERN_INFO "kvm: Nested Virtualization enabled\n"); kvm_enable_efer_bits(EFER_SVME | EFER_LMSLE); @@ -802,8 +829,6 @@ static __init int svm_hardware_setup(void) goto err; } - svm_features = cpuid_edx(SVM_CPUID_FUNC); - if (!boot_cpu_has(X86_FEATURE_NPT)) npt_enabled = false; @@ -854,6 +879,64 @@ static void init_sys_seg(struct vmcb_seg *seg, uint32_t type) seg->base = 0; } +static u64 __scale_tsc(u64 ratio, u64 tsc) +{ + u64 mult, frac, _tsc; + + mult = ratio >> 32; + frac = ratio & ((1ULL << 32) - 1); + + _tsc = tsc; + _tsc *= mult; + _tsc += (tsc >> 32) * frac; + _tsc += ((tsc & ((1ULL << 32) - 1)) * frac) >> 32; + + return _tsc; +} + +static u64 svm_scale_tsc(struct kvm_vcpu *vcpu, u64 tsc) +{ + struct vcpu_svm *svm = to_svm(vcpu); + u64 _tsc = tsc; + + if (svm->tsc_ratio != TSC_RATIO_DEFAULT) + _tsc = __scale_tsc(svm->tsc_ratio, tsc); + + return _tsc; +} + +static void svm_set_tsc_khz(struct kvm_vcpu *vcpu, u32 user_tsc_khz) +{ + struct vcpu_svm *svm = to_svm(vcpu); + u64 ratio; + u64 khz; + + /* TSC scaling supported? */ + if (!boot_cpu_has(X86_FEATURE_TSCRATEMSR)) + return; + + /* TSC-Scaling disabled or guest TSC same frequency as host TSC? */ + if (user_tsc_khz == 0) { + vcpu->arch.virtual_tsc_khz = 0; + svm->tsc_ratio = TSC_RATIO_DEFAULT; + return; + } + + khz = user_tsc_khz; + + /* TSC scaling required - calculate ratio */ + ratio = khz << 32; + do_div(ratio, tsc_khz); + + if (ratio == 0 || ratio & TSC_RATIO_RSVD) { + WARN_ONCE(1, "Invalid TSC ratio - virtual-tsc-khz=%u\n", + user_tsc_khz); + return; + } + vcpu->arch.virtual_tsc_khz = user_tsc_khz; + svm->tsc_ratio = ratio; +} + static void svm_write_tsc_offset(struct kvm_vcpu *vcpu, u64 offset) { struct vcpu_svm *svm = to_svm(vcpu); @@ -880,6 +963,15 @@ static void svm_adjust_tsc_offset(struct kvm_vcpu *vcpu, s64 adjustment) mark_dirty(svm->vmcb, VMCB_INTERCEPTS); } +static u64 svm_compute_tsc_offset(struct kvm_vcpu *vcpu, u64 target_tsc) +{ + u64 tsc; + + tsc = svm_scale_tsc(vcpu, native_read_tsc()); + + return target_tsc - tsc; +} + static void init_vmcb(struct vcpu_svm *svm) { struct vmcb_control_area *control = &svm->vmcb->control; @@ -975,7 +1067,7 @@ static void init_vmcb(struct vcpu_svm *svm) svm_set_efer(&svm->vcpu, 0); save->dr6 = 0xffff0ff0; save->dr7 = 0x400; - save->rflags = 2; + kvm_set_rflags(&svm->vcpu, 2); save->rip = 0x0000fff0; svm->vcpu.arch.regs[VCPU_REGS_RIP] = save->rip; @@ -1048,6 +1140,8 @@ static struct kvm_vcpu *svm_create_vcpu(struct kvm *kvm, unsigned int id) goto out; } + svm->tsc_ratio = TSC_RATIO_DEFAULT; + err = kvm_vcpu_init(&svm->vcpu, kvm, id); if (err) goto free_svm; @@ -1141,6 +1235,12 @@ static void svm_vcpu_load(struct kvm_vcpu *vcpu, int cpu) for (i = 0; i < NR_HOST_SAVE_USER_MSRS; i++) rdmsrl(host_save_user_msrs[i], svm->host_user_msrs[i]); + + if (static_cpu_has(X86_FEATURE_TSCRATEMSR) && + svm->tsc_ratio != __get_cpu_var(current_tsc_ratio)) { + __get_cpu_var(current_tsc_ratio) = svm->tsc_ratio; + wrmsrl(MSR_AMD64_TSC_RATIO, svm->tsc_ratio); + } } static void svm_vcpu_put(struct kvm_vcpu *vcpu) @@ -1365,31 +1465,6 @@ static void svm_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0) { struct vcpu_svm *svm = to_svm(vcpu); - if (is_guest_mode(vcpu)) { - /* - * We are here because we run in nested mode, the host kvm - * intercepts cr0 writes but the l1 hypervisor does not. - * But the L1 hypervisor may intercept selective cr0 writes. - * This needs to be checked here. - */ - unsigned long old, new; - - /* Remove bits that would trigger a real cr0 write intercept */ - old = vcpu->arch.cr0 & SVM_CR0_SELECTIVE_MASK; - new = cr0 & SVM_CR0_SELECTIVE_MASK; - - if (old == new) { - /* cr0 write with ts and mp unchanged */ - svm->vmcb->control.exit_code = SVM_EXIT_CR0_SEL_WRITE; - if (nested_svm_exit_handled(svm) == NESTED_EXIT_DONE) { - svm->nested.vmexit_rip = kvm_rip_read(vcpu); - svm->nested.vmexit_rsp = kvm_register_read(vcpu, VCPU_REGS_RSP); - svm->nested.vmexit_rax = kvm_register_read(vcpu, VCPU_REGS_RAX); - return; - } - } - } - #ifdef CONFIG_X86_64 if (vcpu->arch.efer & EFER_LME) { if (!is_paging(vcpu) && (cr0 & X86_CR0_PG)) { @@ -2127,7 +2202,7 @@ static int nested_svm_vmexit(struct vcpu_svm *svm) nested_vmcb->save.cr3 = kvm_read_cr3(&svm->vcpu); nested_vmcb->save.cr2 = vmcb->save.cr2; nested_vmcb->save.cr4 = svm->vcpu.arch.cr4; - nested_vmcb->save.rflags = vmcb->save.rflags; + nested_vmcb->save.rflags = kvm_get_rflags(&svm->vcpu); nested_vmcb->save.rip = vmcb->save.rip; nested_vmcb->save.rsp = vmcb->save.rsp; nested_vmcb->save.rax = vmcb->save.rax; @@ -2184,7 +2259,7 @@ static int nested_svm_vmexit(struct vcpu_svm *svm) svm->vmcb->save.ds = hsave->save.ds; svm->vmcb->save.gdtr = hsave->save.gdtr; svm->vmcb->save.idtr = hsave->save.idtr; - svm->vmcb->save.rflags = hsave->save.rflags; + kvm_set_rflags(&svm->vcpu, hsave->save.rflags); svm_set_efer(&svm->vcpu, hsave->save.efer); svm_set_cr0(&svm->vcpu, hsave->save.cr0 | X86_CR0_PE); svm_set_cr4(&svm->vcpu, hsave->save.cr4); @@ -2312,7 +2387,7 @@ static bool nested_svm_vmrun(struct vcpu_svm *svm) hsave->save.efer = svm->vcpu.arch.efer; hsave->save.cr0 = kvm_read_cr0(&svm->vcpu); hsave->save.cr4 = svm->vcpu.arch.cr4; - hsave->save.rflags = vmcb->save.rflags; + hsave->save.rflags = kvm_get_rflags(&svm->vcpu); hsave->save.rip = kvm_rip_read(&svm->vcpu); hsave->save.rsp = vmcb->save.rsp; hsave->save.rax = vmcb->save.rax; @@ -2323,7 +2398,7 @@ static bool nested_svm_vmrun(struct vcpu_svm *svm) copy_vmcb_control_area(hsave, vmcb); - if (svm->vmcb->save.rflags & X86_EFLAGS_IF) + if (kvm_get_rflags(&svm->vcpu) & X86_EFLAGS_IF) svm->vcpu.arch.hflags |= HF_HIF_MASK; else svm->vcpu.arch.hflags &= ~HF_HIF_MASK; @@ -2341,7 +2416,7 @@ static bool nested_svm_vmrun(struct vcpu_svm *svm) svm->vmcb->save.ds = nested_vmcb->save.ds; svm->vmcb->save.gdtr = nested_vmcb->save.gdtr; svm->vmcb->save.idtr = nested_vmcb->save.idtr; - svm->vmcb->save.rflags = nested_vmcb->save.rflags; + kvm_set_rflags(&svm->vcpu, nested_vmcb->save.rflags); svm_set_efer(&svm->vcpu, nested_vmcb->save.efer); svm_set_cr0(&svm->vcpu, nested_vmcb->save.cr0); svm_set_cr4(&svm->vcpu, nested_vmcb->save.cr4); @@ -2443,13 +2518,13 @@ static int vmload_interception(struct vcpu_svm *svm) if (nested_svm_check_permissions(svm)) return 1; - svm->next_rip = kvm_rip_read(&svm->vcpu) + 3; - skip_emulated_instruction(&svm->vcpu); - nested_vmcb = nested_svm_map(svm, svm->vmcb->save.rax, &page); if (!nested_vmcb) return 1; + svm->next_rip = kvm_rip_read(&svm->vcpu) + 3; + skip_emulated_instruction(&svm->vcpu); + nested_svm_vmloadsave(nested_vmcb, svm->vmcb); nested_svm_unmap(page); @@ -2464,13 +2539,13 @@ static int vmsave_interception(struct vcpu_svm *svm) if (nested_svm_check_permissions(svm)) return 1; - svm->next_rip = kvm_rip_read(&svm->vcpu) + 3; - skip_emulated_instruction(&svm->vcpu); - nested_vmcb = nested_svm_map(svm, svm->vmcb->save.rax, &page); if (!nested_vmcb) return 1; + svm->next_rip = kvm_rip_read(&svm->vcpu) + 3; + skip_emulated_instruction(&svm->vcpu); + nested_svm_vmloadsave(svm->vmcb, nested_vmcb); nested_svm_unmap(page); @@ -2676,6 +2751,29 @@ static int emulate_on_interception(struct vcpu_svm *svm) return emulate_instruction(&svm->vcpu, 0) == EMULATE_DONE; } +bool check_selective_cr0_intercepted(struct vcpu_svm *svm, unsigned long val) +{ + unsigned long cr0 = svm->vcpu.arch.cr0; + bool ret = false; + u64 intercept; + + intercept = svm->nested.intercept; + + if (!is_guest_mode(&svm->vcpu) || + (!(intercept & (1ULL << INTERCEPT_SELECTIVE_CR0)))) + return false; + + cr0 &= ~SVM_CR0_SELECTIVE_MASK; + val &= ~SVM_CR0_SELECTIVE_MASK; + + if (cr0 ^ val) { + svm->vmcb->control.exit_code = SVM_EXIT_CR0_SEL_WRITE; + ret = (nested_svm_exit_handled(svm) == NESTED_EXIT_DONE); + } + + return ret; +} + #define CR_VALID (1ULL << 63) static int cr_interception(struct vcpu_svm *svm) @@ -2699,7 +2797,11 @@ static int cr_interception(struct vcpu_svm *svm) val = kvm_register_read(&svm->vcpu, reg); switch (cr) { case 0: - err = kvm_set_cr0(&svm->vcpu, val); + if (!check_selective_cr0_intercepted(svm, val)) + err = kvm_set_cr0(&svm->vcpu, val); + else + return 1; + break; case 3: err = kvm_set_cr3(&svm->vcpu, val); @@ -2744,23 +2846,6 @@ static int cr_interception(struct vcpu_svm *svm) return 1; } -static int cr0_write_interception(struct vcpu_svm *svm) -{ - struct kvm_vcpu *vcpu = &svm->vcpu; - int r; - - r = cr_interception(svm); - - if (svm->nested.vmexit_rip) { - kvm_register_write(vcpu, VCPU_REGS_RIP, svm->nested.vmexit_rip); - kvm_register_write(vcpu, VCPU_REGS_RSP, svm->nested.vmexit_rsp); - kvm_register_write(vcpu, VCPU_REGS_RAX, svm->nested.vmexit_rax); - svm->nested.vmexit_rip = 0; - } - - return r; -} - static int dr_interception(struct vcpu_svm *svm) { int reg, dr; @@ -2813,7 +2898,9 @@ static int svm_get_msr(struct kvm_vcpu *vcpu, unsigned ecx, u64 *data) case MSR_IA32_TSC: { struct vmcb *vmcb = get_host_vmcb(svm); - *data = vmcb->control.tsc_offset + native_read_tsc(); + *data = vmcb->control.tsc_offset + + svm_scale_tsc(vcpu, native_read_tsc()); + break; } case MSR_STAR: @@ -3048,7 +3135,7 @@ static int (*svm_exit_handlers[])(struct vcpu_svm *svm) = { [SVM_EXIT_READ_CR4] = cr_interception, [SVM_EXIT_READ_CR8] = cr_interception, [SVM_EXIT_CR0_SEL_WRITE] = emulate_on_interception, - [SVM_EXIT_WRITE_CR0] = cr0_write_interception, + [SVM_EXIT_WRITE_CR0] = cr_interception, [SVM_EXIT_WRITE_CR3] = cr_interception, [SVM_EXIT_WRITE_CR4] = cr_interception, [SVM_EXIT_WRITE_CR8] = cr8_write_interception, @@ -3104,97 +3191,109 @@ static int (*svm_exit_handlers[])(struct vcpu_svm *svm) = { [SVM_EXIT_NPF] = pf_interception, }; -void dump_vmcb(struct kvm_vcpu *vcpu) +static void dump_vmcb(struct kvm_vcpu *vcpu) { struct vcpu_svm *svm = to_svm(vcpu); struct vmcb_control_area *control = &svm->vmcb->control; struct vmcb_save_area *save = &svm->vmcb->save; pr_err("VMCB Control Area:\n"); - pr_err("cr_read: %04x\n", control->intercept_cr & 0xffff); - pr_err("cr_write: %04x\n", control->intercept_cr >> 16); - pr_err("dr_read: %04x\n", control->intercept_dr & 0xffff); - pr_err("dr_write: %04x\n", control->intercept_dr >> 16); - pr_err("exceptions: %08x\n", control->intercept_exceptions); - pr_err("intercepts: %016llx\n", control->intercept); - pr_err("pause filter count: %d\n", control->pause_filter_count); - pr_err("iopm_base_pa: %016llx\n", control->iopm_base_pa); - pr_err("msrpm_base_pa: %016llx\n", control->msrpm_base_pa); - pr_err("tsc_offset: %016llx\n", control->tsc_offset); - pr_err("asid: %d\n", control->asid); - pr_err("tlb_ctl: %d\n", control->tlb_ctl); - pr_err("int_ctl: %08x\n", control->int_ctl); - pr_err("int_vector: %08x\n", control->int_vector); - pr_err("int_state: %08x\n", control->int_state); - pr_err("exit_code: %08x\n", control->exit_code); - pr_err("exit_info1: %016llx\n", control->exit_info_1); - pr_err("exit_info2: %016llx\n", control->exit_info_2); - pr_err("exit_int_info: %08x\n", control->exit_int_info); - pr_err("exit_int_info_err: %08x\n", control->exit_int_info_err); - pr_err("nested_ctl: %lld\n", control->nested_ctl); - pr_err("nested_cr3: %016llx\n", control->nested_cr3); - pr_err("event_inj: %08x\n", control->event_inj); - pr_err("event_inj_err: %08x\n", control->event_inj_err); - pr_err("lbr_ctl: %lld\n", control->lbr_ctl); - pr_err("next_rip: %016llx\n", control->next_rip); + pr_err("%-20s%04x\n", "cr_read:", control->intercept_cr & 0xffff); + pr_err("%-20s%04x\n", "cr_write:", control->intercept_cr >> 16); + pr_err("%-20s%04x\n", "dr_read:", control->intercept_dr & 0xffff); + pr_err("%-20s%04x\n", "dr_write:", control->intercept_dr >> 16); + pr_err("%-20s%08x\n", "exceptions:", control->intercept_exceptions); + pr_err("%-20s%016llx\n", "intercepts:", control->intercept); + pr_err("%-20s%d\n", "pause filter count:", control->pause_filter_count); + pr_err("%-20s%016llx\n", "iopm_base_pa:", control->iopm_base_pa); + pr_err("%-20s%016llx\n", "msrpm_base_pa:", control->msrpm_base_pa); + pr_err("%-20s%016llx\n", "tsc_offset:", control->tsc_offset); + pr_err("%-20s%d\n", "asid:", control->asid); + pr_err("%-20s%d\n", "tlb_ctl:", control->tlb_ctl); + pr_err("%-20s%08x\n", "int_ctl:", control->int_ctl); + pr_err("%-20s%08x\n", "int_vector:", control->int_vector); + pr_err("%-20s%08x\n", "int_state:", control->int_state); + pr_err("%-20s%08x\n", "exit_code:", control->exit_code); + pr_err("%-20s%016llx\n", "exit_info1:", control->exit_info_1); + pr_err("%-20s%016llx\n", "exit_info2:", control->exit_info_2); + pr_err("%-20s%08x\n", "exit_int_info:", control->exit_int_info); + pr_err("%-20s%08x\n", "exit_int_info_err:", control->exit_int_info_err); + pr_err("%-20s%lld\n", "nested_ctl:", control->nested_ctl); + pr_err("%-20s%016llx\n", "nested_cr3:", control->nested_cr3); + pr_err("%-20s%08x\n", "event_inj:", control->event_inj); + pr_err("%-20s%08x\n", "event_inj_err:", control->event_inj_err); + pr_err("%-20s%lld\n", "lbr_ctl:", control->lbr_ctl); + pr_err("%-20s%016llx\n", "next_rip:", control->next_rip); pr_err("VMCB State Save Area:\n"); - pr_err("es: s: %04x a: %04x l: %08x b: %016llx\n", - save->es.selector, save->es.attrib, - save->es.limit, save->es.base); - pr_err("cs: s: %04x a: %04x l: %08x b: %016llx\n", - save->cs.selector, save->cs.attrib, - save->cs.limit, save->cs.base); - pr_err("ss: s: %04x a: %04x l: %08x b: %016llx\n", - save->ss.selector, save->ss.attrib, - save->ss.limit, save->ss.base); - pr_err("ds: s: %04x a: %04x l: %08x b: %016llx\n", - save->ds.selector, save->ds.attrib, - save->ds.limit, save->ds.base); - pr_err("fs: s: %04x a: %04x l: %08x b: %016llx\n", - save->fs.selector, save->fs.attrib, - save->fs.limit, save->fs.base); - pr_err("gs: s: %04x a: %04x l: %08x b: %016llx\n", - save->gs.selector, save->gs.attrib, - save->gs.limit, save->gs.base); - pr_err("gdtr: s: %04x a: %04x l: %08x b: %016llx\n", - save->gdtr.selector, save->gdtr.attrib, - save->gdtr.limit, save->gdtr.base); - pr_err("ldtr: s: %04x a: %04x l: %08x b: %016llx\n", - save->ldtr.selector, save->ldtr.attrib, - save->ldtr.limit, save->ldtr.base); - pr_err("idtr: s: %04x a: %04x l: %08x b: %016llx\n", - save->idtr.selector, save->idtr.attrib, - save->idtr.limit, save->idtr.base); - pr_err("tr: s: %04x a: %04x l: %08x b: %016llx\n", - save->tr.selector, save->tr.attrib, - save->tr.limit, save->tr.base); + pr_err("%-5s s: %04x a: %04x l: %08x b: %016llx\n", + "es:", + save->es.selector, save->es.attrib, + save->es.limit, save->es.base); + pr_err("%-5s s: %04x a: %04x l: %08x b: %016llx\n", + "cs:", + save->cs.selector, save->cs.attrib, + save->cs.limit, save->cs.base); + pr_err("%-5s s: %04x a: %04x l: %08x b: %016llx\n", + "ss:", + save->ss.selector, save->ss.attrib, + save->ss.limit, save->ss.base); + pr_err("%-5s s: %04x a: %04x l: %08x b: %016llx\n", + "ds:", + save->ds.selector, save->ds.attrib, + save->ds.limit, save->ds.base); + pr_err("%-5s s: %04x a: %04x l: %08x b: %016llx\n", + "fs:", + save->fs.selector, save->fs.attrib, + save->fs.limit, save->fs.base); + pr_err("%-5s s: %04x a: %04x l: %08x b: %016llx\n", + "gs:", + save->gs.selector, save->gs.attrib, + save->gs.limit, save->gs.base); + pr_err("%-5s s: %04x a: %04x l: %08x b: %016llx\n", + "gdtr:", + save->gdtr.selector, save->gdtr.attrib, + save->gdtr.limit, save->gdtr.base); + pr_err("%-5s s: %04x a: %04x l: %08x b: %016llx\n", + "ldtr:", + save->ldtr.selector, save->ldtr.attrib, + save->ldtr.limit, save->ldtr.base); + pr_err("%-5s s: %04x a: %04x l: %08x b: %016llx\n", + "idtr:", + save->idtr.selector, save->idtr.attrib, + save->idtr.limit, save->idtr.base); + pr_err("%-5s s: %04x a: %04x l: %08x b: %016llx\n", + "tr:", + save->tr.selector, save->tr.attrib, + save->tr.limit, save->tr.base); pr_err("cpl: %d efer: %016llx\n", save->cpl, save->efer); - pr_err("cr0: %016llx cr2: %016llx\n", - save->cr0, save->cr2); - pr_err("cr3: %016llx cr4: %016llx\n", - save->cr3, save->cr4); - pr_err("dr6: %016llx dr7: %016llx\n", - save->dr6, save->dr7); - pr_err("rip: %016llx rflags: %016llx\n", - save->rip, save->rflags); - pr_err("rsp: %016llx rax: %016llx\n", - save->rsp, save->rax); - pr_err("star: %016llx lstar: %016llx\n", - save->star, save->lstar); - pr_err("cstar: %016llx sfmask: %016llx\n", - save->cstar, save->sfmask); - pr_err("kernel_gs_base: %016llx sysenter_cs: %016llx\n", - save->kernel_gs_base, save->sysenter_cs); - pr_err("sysenter_esp: %016llx sysenter_eip: %016llx\n", - save->sysenter_esp, save->sysenter_eip); - pr_err("gpat: %016llx dbgctl: %016llx\n", - save->g_pat, save->dbgctl); - pr_err("br_from: %016llx br_to: %016llx\n", - save->br_from, save->br_to); - pr_err("excp_from: %016llx excp_to: %016llx\n", - save->last_excp_from, save->last_excp_to); - + pr_err("%-15s %016llx %-13s %016llx\n", + "cr0:", save->cr0, "cr2:", save->cr2); + pr_err("%-15s %016llx %-13s %016llx\n", + "cr3:", save->cr3, "cr4:", save->cr4); + pr_err("%-15s %016llx %-13s %016llx\n", + "dr6:", save->dr6, "dr7:", save->dr7); + pr_err("%-15s %016llx %-13s %016llx\n", + "rip:", save->rip, "rflags:", save->rflags); + pr_err("%-15s %016llx %-13s %016llx\n", + "rsp:", save->rsp, "rax:", save->rax); + pr_err("%-15s %016llx %-13s %016llx\n", + "star:", save->star, "lstar:", save->lstar); + pr_err("%-15s %016llx %-13s %016llx\n", + "cstar:", save->cstar, "sfmask:", save->sfmask); + pr_err("%-15s %016llx %-13s %016llx\n", + "kernel_gs_base:", save->kernel_gs_base, + "sysenter_cs:", save->sysenter_cs); + pr_err("%-15s %016llx %-13s %016llx\n", + "sysenter_esp:", save->sysenter_esp, + "sysenter_eip:", save->sysenter_eip); + pr_err("%-15s %016llx %-13s %016llx\n", + "gpat:", save->g_pat, "dbgctl:", save->dbgctl); + pr_err("%-15s %016llx %-13s %016llx\n", + "br_from:", save->br_from, "br_to:", save->br_to); + pr_err("%-15s %016llx %-13s %016llx\n", + "excp_from:", save->last_excp_from, + "excp_to:", save->last_excp_to); } static void svm_get_exit_info(struct kvm_vcpu *vcpu, u64 *info1, u64 *info2) @@ -3384,7 +3483,7 @@ static int svm_interrupt_allowed(struct kvm_vcpu *vcpu) (vmcb->control.int_state & SVM_INTERRUPT_SHADOW_MASK)) return 0; - ret = !!(vmcb->save.rflags & X86_EFLAGS_IF); + ret = !!(kvm_get_rflags(vcpu) & X86_EFLAGS_IF); if (is_guest_mode(vcpu)) return ret && !(svm->vcpu.arch.hflags & HF_VINTR_MASK); @@ -3871,6 +3970,186 @@ static void svm_fpu_deactivate(struct kvm_vcpu *vcpu) update_cr0_intercept(svm); } +#define PRE_EX(exit) { .exit_code = (exit), \ + .stage = X86_ICPT_PRE_EXCEPT, } +#define POST_EX(exit) { .exit_code = (exit), \ + .stage = X86_ICPT_POST_EXCEPT, } +#define POST_MEM(exit) { .exit_code = (exit), \ + .stage = X86_ICPT_POST_MEMACCESS, } + +static struct __x86_intercept { + u32 exit_code; + enum x86_intercept_stage stage; +} x86_intercept_map[] = { + [x86_intercept_cr_read] = POST_EX(SVM_EXIT_READ_CR0), + [x86_intercept_cr_write] = POST_EX(SVM_EXIT_WRITE_CR0), + [x86_intercept_clts] = POST_EX(SVM_EXIT_WRITE_CR0), + [x86_intercept_lmsw] = POST_EX(SVM_EXIT_WRITE_CR0), + [x86_intercept_smsw] = POST_EX(SVM_EXIT_READ_CR0), + [x86_intercept_dr_read] = POST_EX(SVM_EXIT_READ_DR0), + [x86_intercept_dr_write] = POST_EX(SVM_EXIT_WRITE_DR0), + [x86_intercept_sldt] = POST_EX(SVM_EXIT_LDTR_READ), + [x86_intercept_str] = POST_EX(SVM_EXIT_TR_READ), + [x86_intercept_lldt] = POST_EX(SVM_EXIT_LDTR_WRITE), + [x86_intercept_ltr] = POST_EX(SVM_EXIT_TR_WRITE), + [x86_intercept_sgdt] = POST_EX(SVM_EXIT_GDTR_READ), + [x86_intercept_sidt] = POST_EX(SVM_EXIT_IDTR_READ), + [x86_intercept_lgdt] = POST_EX(SVM_EXIT_GDTR_WRITE), + [x86_intercept_lidt] = POST_EX(SVM_EXIT_IDTR_WRITE), + [x86_intercept_vmrun] = POST_EX(SVM_EXIT_VMRUN), + [x86_intercept_vmmcall] = POST_EX(SVM_EXIT_VMMCALL), + [x86_intercept_vmload] = POST_EX(SVM_EXIT_VMLOAD), + [x86_intercept_vmsave] = POST_EX(SVM_EXIT_VMSAVE), + [x86_intercept_stgi] = POST_EX(SVM_EXIT_STGI), + [x86_intercept_clgi] = POST_EX(SVM_EXIT_CLGI), + [x86_intercept_skinit] = POST_EX(SVM_EXIT_SKINIT), + [x86_intercept_invlpga] = POST_EX(SVM_EXIT_INVLPGA), + [x86_intercept_rdtscp] = POST_EX(SVM_EXIT_RDTSCP), + [x86_intercept_monitor] = POST_MEM(SVM_EXIT_MONITOR), + [x86_intercept_mwait] = POST_EX(SVM_EXIT_MWAIT), + [x86_intercept_invlpg] = POST_EX(SVM_EXIT_INVLPG), + [x86_intercept_invd] = POST_EX(SVM_EXIT_INVD), + [x86_intercept_wbinvd] = POST_EX(SVM_EXIT_WBINVD), + [x86_intercept_wrmsr] = POST_EX(SVM_EXIT_MSR), + [x86_intercept_rdtsc] = POST_EX(SVM_EXIT_RDTSC), + [x86_intercept_rdmsr] = POST_EX(SVM_EXIT_MSR), + [x86_intercept_rdpmc] = POST_EX(SVM_EXIT_RDPMC), + [x86_intercept_cpuid] = PRE_EX(SVM_EXIT_CPUID), + [x86_intercept_rsm] = PRE_EX(SVM_EXIT_RSM), + [x86_intercept_pause] = PRE_EX(SVM_EXIT_PAUSE), + [x86_intercept_pushf] = PRE_EX(SVM_EXIT_PUSHF), + [x86_intercept_popf] = PRE_EX(SVM_EXIT_POPF), + [x86_intercept_intn] = PRE_EX(SVM_EXIT_SWINT), + [x86_intercept_iret] = PRE_EX(SVM_EXIT_IRET), + [x86_intercept_icebp] = PRE_EX(SVM_EXIT_ICEBP), + [x86_intercept_hlt] = POST_EX(SVM_EXIT_HLT), + [x86_intercept_in] = POST_EX(SVM_EXIT_IOIO), + [x86_intercept_ins] = POST_EX(SVM_EXIT_IOIO), + [x86_intercept_out] = POST_EX(SVM_EXIT_IOIO), + [x86_intercept_outs] = POST_EX(SVM_EXIT_IOIO), +}; + +#undef PRE_EX +#undef POST_EX +#undef POST_MEM + +static int svm_check_intercept(struct kvm_vcpu *vcpu, + struct x86_instruction_info *info, + enum x86_intercept_stage stage) +{ + struct vcpu_svm *svm = to_svm(vcpu); + int vmexit, ret = X86EMUL_CONTINUE; + struct __x86_intercept icpt_info; + struct vmcb *vmcb = svm->vmcb; + + if (info->intercept >= ARRAY_SIZE(x86_intercept_map)) + goto out; + + icpt_info = x86_intercept_map[info->intercept]; + + if (stage != icpt_info.stage) + goto out; + + switch (icpt_info.exit_code) { + case SVM_EXIT_READ_CR0: + if (info->intercept == x86_intercept_cr_read) + icpt_info.exit_code += info->modrm_reg; + break; + case SVM_EXIT_WRITE_CR0: { + unsigned long cr0, val; + u64 intercept; + + if (info->intercept == x86_intercept_cr_write) + icpt_info.exit_code += info->modrm_reg; + + if (icpt_info.exit_code != SVM_EXIT_WRITE_CR0) + break; + + intercept = svm->nested.intercept; + + if (!(intercept & (1ULL << INTERCEPT_SELECTIVE_CR0))) + break; + + cr0 = vcpu->arch.cr0 & ~SVM_CR0_SELECTIVE_MASK; + val = info->src_val & ~SVM_CR0_SELECTIVE_MASK; + + if (info->intercept == x86_intercept_lmsw) { + cr0 &= 0xfUL; + val &= 0xfUL; + /* lmsw can't clear PE - catch this here */ + if (cr0 & X86_CR0_PE) + val |= X86_CR0_PE; + } + + if (cr0 ^ val) + icpt_info.exit_code = SVM_EXIT_CR0_SEL_WRITE; + + break; + } + case SVM_EXIT_READ_DR0: + case SVM_EXIT_WRITE_DR0: + icpt_info.exit_code += info->modrm_reg; + break; + case SVM_EXIT_MSR: + if (info->intercept == x86_intercept_wrmsr) + vmcb->control.exit_info_1 = 1; + else + vmcb->control.exit_info_1 = 0; + break; + case SVM_EXIT_PAUSE: + /* + * We get this for NOP only, but pause + * is rep not, check this here + */ + if (info->rep_prefix != REPE_PREFIX) + goto out; + case SVM_EXIT_IOIO: { + u64 exit_info; + u32 bytes; + + exit_info = (vcpu->arch.regs[VCPU_REGS_RDX] & 0xffff) << 16; + + if (info->intercept == x86_intercept_in || + info->intercept == x86_intercept_ins) { + exit_info |= SVM_IOIO_TYPE_MASK; + bytes = info->src_bytes; + } else { + bytes = info->dst_bytes; + } + + if (info->intercept == x86_intercept_outs || + info->intercept == x86_intercept_ins) + exit_info |= SVM_IOIO_STR_MASK; + + if (info->rep_prefix) + exit_info |= SVM_IOIO_REP_MASK; + + bytes = min(bytes, 4u); + + exit_info |= bytes << SVM_IOIO_SIZE_SHIFT; + + exit_info |= (u32)info->ad_bytes << (SVM_IOIO_ASIZE_SHIFT - 1); + + vmcb->control.exit_info_1 = exit_info; + vmcb->control.exit_info_2 = info->next_rip; + + break; + } + default: + break; + } + + vmcb->control.next_rip = info->next_rip; + vmcb->control.exit_code = icpt_info.exit_code; + vmexit = nested_svm_exit_handled(svm); + + ret = (vmexit == NESTED_EXIT_DONE) ? X86EMUL_INTERCEPTED + : X86EMUL_CONTINUE; + +out: + return ret; +} + static struct kvm_x86_ops svm_x86_ops = { .cpu_has_kvm_support = has_svm, .disabled_by_bios = is_disabled, @@ -3952,10 +4231,14 @@ static struct kvm_x86_ops svm_x86_ops = { .has_wbinvd_exit = svm_has_wbinvd_exit, + .set_tsc_khz = svm_set_tsc_khz, .write_tsc_offset = svm_write_tsc_offset, .adjust_tsc_offset = svm_adjust_tsc_offset, + .compute_tsc_offset = svm_compute_tsc_offset, .set_tdp_cr3 = set_tdp_cr3, + + .check_intercept = svm_check_intercept, }; static int __init svm_init(void) diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 5b4cdcbd154c..4c3fa0f67469 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -128,8 +128,11 @@ struct vcpu_vmx { unsigned long host_rsp; int launched; u8 fail; + u8 cpl; + bool nmi_known_unmasked; u32 exit_intr_info; u32 idt_vectoring_info; + ulong rflags; struct shared_msr_entry *guest_msrs; int nmsrs; int save_nmsrs; @@ -159,6 +162,10 @@ struct vcpu_vmx { u32 ar; } tr, es, ds, fs, gs; } rmode; + struct { + u32 bitmask; /* 4 bits per segment (1 bit per field) */ + struct kvm_save_segment seg[8]; + } segment_cache; int vpid; bool emulation_required; @@ -171,6 +178,15 @@ struct vcpu_vmx { bool rdtscp_enabled; }; +enum segment_cache_field { + SEG_FIELD_SEL = 0, + SEG_FIELD_BASE = 1, + SEG_FIELD_LIMIT = 2, + SEG_FIELD_AR = 3, + + SEG_FIELD_NR = 4 +}; + static inline struct vcpu_vmx *to_vmx(struct kvm_vcpu *vcpu) { return container_of(vcpu, struct vcpu_vmx, vcpu); @@ -643,6 +659,62 @@ static void vmcs_set_bits(unsigned long field, u32 mask) vmcs_writel(field, vmcs_readl(field) | mask); } +static void vmx_segment_cache_clear(struct vcpu_vmx *vmx) +{ + vmx->segment_cache.bitmask = 0; +} + +static bool vmx_segment_cache_test_set(struct vcpu_vmx *vmx, unsigned seg, + unsigned field) +{ + bool ret; + u32 mask = 1 << (seg * SEG_FIELD_NR + field); + + if (!(vmx->vcpu.arch.regs_avail & (1 << VCPU_EXREG_SEGMENTS))) { + vmx->vcpu.arch.regs_avail |= (1 << VCPU_EXREG_SEGMENTS); + vmx->segment_cache.bitmask = 0; + } + ret = vmx->segment_cache.bitmask & mask; + vmx->segment_cache.bitmask |= mask; + return ret; +} + +static u16 vmx_read_guest_seg_selector(struct vcpu_vmx *vmx, unsigned seg) +{ + u16 *p = &vmx->segment_cache.seg[seg].selector; + + if (!vmx_segment_cache_test_set(vmx, seg, SEG_FIELD_SEL)) + *p = vmcs_read16(kvm_vmx_segment_fields[seg].selector); + return *p; +} + +static ulong vmx_read_guest_seg_base(struct vcpu_vmx *vmx, unsigned seg) +{ + ulong *p = &vmx->segment_cache.seg[seg].base; + + if (!vmx_segment_cache_test_set(vmx, seg, SEG_FIELD_BASE)) + *p = vmcs_readl(kvm_vmx_segment_fields[seg].base); + return *p; +} + +static u32 vmx_read_guest_seg_limit(struct vcpu_vmx *vmx, unsigned seg) +{ + u32 *p = &vmx->segment_cache.seg[seg].limit; + + if (!vmx_segment_cache_test_set(vmx, seg, SEG_FIELD_LIMIT)) + *p = vmcs_read32(kvm_vmx_segment_fields[seg].limit); + return *p; +} + +static u32 vmx_read_guest_seg_ar(struct vcpu_vmx *vmx, unsigned seg) +{ + u32 *p = &vmx->segment_cache.seg[seg].ar; + + if (!vmx_segment_cache_test_set(vmx, seg, SEG_FIELD_AR)) + *p = vmcs_read32(kvm_vmx_segment_fields[seg].ar_bytes); + return *p; +} + static void update_exception_bitmap(struct kvm_vcpu *vcpu) { u32 eb; @@ -970,17 +1042,24 @@ static unsigned long vmx_get_rflags(struct kvm_vcpu *vcpu) { unsigned long rflags, save_rflags; - rflags = vmcs_readl(GUEST_RFLAGS); - if (to_vmx(vcpu)->rmode.vm86_active) { - rflags &= RMODE_GUEST_OWNED_EFLAGS_BITS; - save_rflags = to_vmx(vcpu)->rmode.save_rflags; - rflags |= save_rflags & ~RMODE_GUEST_OWNED_EFLAGS_BITS; + if (!test_bit(VCPU_EXREG_RFLAGS, (ulong *)&vcpu->arch.regs_avail)) { + __set_bit(VCPU_EXREG_RFLAGS, (ulong *)&vcpu->arch.regs_avail); + rflags = vmcs_readl(GUEST_RFLAGS); + if (to_vmx(vcpu)->rmode.vm86_active) { + rflags &= RMODE_GUEST_OWNED_EFLAGS_BITS; + save_rflags = to_vmx(vcpu)->rmode.save_rflags; + rflags |= save_rflags & ~RMODE_GUEST_OWNED_EFLAGS_BITS; + } + to_vmx(vcpu)->rflags = rflags; } - return rflags; + return to_vmx(vcpu)->rflags; } static void vmx_set_rflags(struct kvm_vcpu *vcpu, unsigned long rflags) { + __set_bit(VCPU_EXREG_RFLAGS, (ulong *)&vcpu->arch.regs_avail); + __clear_bit(VCPU_EXREG_CPL, (ulong *)&vcpu->arch.regs_avail); + to_vmx(vcpu)->rflags = rflags; if (to_vmx(vcpu)->rmode.vm86_active) { to_vmx(vcpu)->rmode.save_rflags = rflags; rflags |= X86_EFLAGS_IOPL | X86_EFLAGS_VM; @@ -1053,7 +1132,10 @@ static void vmx_queue_exception(struct kvm_vcpu *vcpu, unsigned nr, } if (vmx->rmode.vm86_active) { - if (kvm_inject_realmode_interrupt(vcpu, nr) != EMULATE_DONE) + int inc_eip = 0; + if (kvm_exception_is_soft(nr)) + inc_eip = vcpu->arch.event_exit_inst_len; + if (kvm_inject_realmode_interrupt(vcpu, nr, inc_eip) != EMULATE_DONE) kvm_make_request(KVM_REQ_TRIPLE_FAULT, vcpu); return; } @@ -1151,6 +1233,16 @@ static u64 guest_read_tsc(void) } /* + * Empty call-back. Needs to be implemented when VMX enables the SET_TSC_KHZ + * ioctl. In this case the call-back should update internal vmx state to make + * the changes effective. + */ +static void vmx_set_tsc_khz(struct kvm_vcpu *vcpu, u32 user_tsc_khz) +{ + /* Nothing to do here */ +} + +/* * writes 'offset' into guest's timestamp counter offset register */ static void vmx_write_tsc_offset(struct kvm_vcpu *vcpu, u64 offset) @@ -1164,6 +1256,11 @@ static void vmx_adjust_tsc_offset(struct kvm_vcpu *vcpu, s64 adjustment) vmcs_write64(TSC_OFFSET, offset + adjustment); } +static u64 vmx_compute_tsc_offset(struct kvm_vcpu *vcpu, u64 target_tsc) +{ + return target_tsc - native_read_tsc(); +} + /* * Reads an msr value (of 'msr_index') into 'pdata'. * Returns 0 on success, non-0 otherwise. @@ -1243,9 +1340,11 @@ static int vmx_set_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 data) break; #ifdef CONFIG_X86_64 case MSR_FS_BASE: + vmx_segment_cache_clear(vmx); vmcs_writel(GUEST_FS_BASE, data); break; case MSR_GS_BASE: + vmx_segment_cache_clear(vmx); vmcs_writel(GUEST_GS_BASE, data); break; case MSR_KERNEL_GS_BASE: @@ -1689,6 +1788,8 @@ static void enter_pmode(struct kvm_vcpu *vcpu) vmx->emulation_required = 1; vmx->rmode.vm86_active = 0; + vmx_segment_cache_clear(vmx); + vmcs_write16(GUEST_TR_SELECTOR, vmx->rmode.tr.selector); vmcs_writel(GUEST_TR_BASE, vmx->rmode.tr.base); vmcs_write32(GUEST_TR_LIMIT, vmx->rmode.tr.limit); @@ -1712,6 +1813,8 @@ static void enter_pmode(struct kvm_vcpu *vcpu) fix_pmode_dataseg(VCPU_SREG_GS, &vmx->rmode.gs); fix_pmode_dataseg(VCPU_SREG_FS, &vmx->rmode.fs); + vmx_segment_cache_clear(vmx); + vmcs_write16(GUEST_SS_SELECTOR, 0); vmcs_write32(GUEST_SS_AR_BYTES, 0x93); @@ -1775,6 +1878,8 @@ static void enter_rmode(struct kvm_vcpu *vcpu) vcpu->srcu_idx = srcu_read_lock(&vcpu->kvm->srcu); } + vmx_segment_cache_clear(vmx); + vmx->rmode.tr.selector = vmcs_read16(GUEST_TR_SELECTOR); vmx->rmode.tr.base = vmcs_readl(GUEST_TR_BASE); vmcs_writel(GUEST_TR_BASE, rmode_tss_base(vcpu->kvm)); @@ -1851,6 +1956,8 @@ static void enter_lmode(struct kvm_vcpu *vcpu) { u32 guest_tr_ar; + vmx_segment_cache_clear(to_vmx(vcpu)); + guest_tr_ar = vmcs_read32(GUEST_TR_AR_BYTES); if ((guest_tr_ar & AR_TYPE_MASK) != AR_TYPE_BUSY_64_TSS) { printk(KERN_DEBUG "%s: tss fixup for long mode. \n", @@ -1998,6 +2105,7 @@ static void vmx_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0) vmcs_writel(CR0_READ_SHADOW, cr0); vmcs_writel(GUEST_CR0, hw_cr0); vcpu->arch.cr0 = cr0; + __clear_bit(VCPU_EXREG_CPL, (ulong *)&vcpu->arch.regs_avail); } static u64 construct_eptp(unsigned long root_hpa) @@ -2053,7 +2161,6 @@ static void vmx_get_segment(struct kvm_vcpu *vcpu, struct kvm_segment *var, int seg) { struct vcpu_vmx *vmx = to_vmx(vcpu); - struct kvm_vmx_segment_field *sf = &kvm_vmx_segment_fields[seg]; struct kvm_save_segment *save; u32 ar; @@ -2075,13 +2182,13 @@ static void vmx_get_segment(struct kvm_vcpu *vcpu, var->limit = save->limit; ar = save->ar; if (seg == VCPU_SREG_TR - || var->selector == vmcs_read16(sf->selector)) + || var->selector == vmx_read_guest_seg_selector(vmx, seg)) goto use_saved_rmode_seg; } - var->base = vmcs_readl(sf->base); - var->limit = vmcs_read32(sf->limit); - var->selector = vmcs_read16(sf->selector); - ar = vmcs_read32(sf->ar_bytes); + var->base = vmx_read_guest_seg_base(vmx, seg); + var->limit = vmx_read_guest_seg_limit(vmx, seg); + var->selector = vmx_read_guest_seg_selector(vmx, seg); + ar = vmx_read_guest_seg_ar(vmx, seg); use_saved_rmode_seg: if ((ar & AR_UNUSABLE_MASK) && !emulate_invalid_guest_state) ar = 0; @@ -2098,27 +2205,37 @@ use_saved_rmode_seg: static u64 vmx_get_segment_base(struct kvm_vcpu *vcpu, int seg) { - struct kvm_vmx_segment_field *sf = &kvm_vmx_segment_fields[seg]; struct kvm_segment s; if (to_vmx(vcpu)->rmode.vm86_active) { vmx_get_segment(vcpu, &s, seg); return s.base; } - return vmcs_readl(sf->base); + return vmx_read_guest_seg_base(to_vmx(vcpu), seg); } -static int vmx_get_cpl(struct kvm_vcpu *vcpu) +static int __vmx_get_cpl(struct kvm_vcpu *vcpu) { if (!is_protmode(vcpu)) return 0; - if (vmx_get_rflags(vcpu) & X86_EFLAGS_VM) /* if virtual 8086 */ + if (!is_long_mode(vcpu) + && (kvm_get_rflags(vcpu) & X86_EFLAGS_VM)) /* if virtual 8086 */ return 3; - return vmcs_read16(GUEST_CS_SELECTOR) & 3; + return vmx_read_guest_seg_selector(to_vmx(vcpu), VCPU_SREG_CS) & 3; } +static int vmx_get_cpl(struct kvm_vcpu *vcpu) +{ + if (!test_bit(VCPU_EXREG_CPL, (ulong *)&vcpu->arch.regs_avail)) { + __set_bit(VCPU_EXREG_CPL, (ulong *)&vcpu->arch.regs_avail); + to_vmx(vcpu)->cpl = __vmx_get_cpl(vcpu); + } + return to_vmx(vcpu)->cpl; +} + + static u32 vmx_segment_access_rights(struct kvm_segment *var) { u32 ar; @@ -2148,6 +2265,8 @@ static void vmx_set_segment(struct kvm_vcpu *vcpu, struct kvm_vmx_segment_field *sf = &kvm_vmx_segment_fields[seg]; u32 ar; + vmx_segment_cache_clear(vmx); + if (vmx->rmode.vm86_active && seg == VCPU_SREG_TR) { vmcs_write16(sf->selector, var->selector); vmx->rmode.tr.selector = var->selector; @@ -2184,11 +2303,12 @@ static void vmx_set_segment(struct kvm_vcpu *vcpu, ar |= 0x1; /* Accessed */ vmcs_write32(sf->ar_bytes, ar); + __clear_bit(VCPU_EXREG_CPL, (ulong *)&vcpu->arch.regs_avail); } static void vmx_get_cs_db_l_bits(struct kvm_vcpu *vcpu, int *db, int *l) { - u32 ar = vmcs_read32(GUEST_CS_AR_BYTES); + u32 ar = vmx_read_guest_seg_ar(to_vmx(vcpu), VCPU_SREG_CS); *db = (ar >> 14) & 1; *l = (ar >> 13) & 1; @@ -2775,6 +2895,8 @@ static int vmx_vcpu_reset(struct kvm_vcpu *vcpu) if (ret != 0) goto out; + vmx_segment_cache_clear(vmx); + seg_setup(VCPU_SREG_CS); /* * GUEST_CS_BASE should really be 0xffff0000, but VT vm86 mode @@ -2904,7 +3026,10 @@ static void vmx_inject_irq(struct kvm_vcpu *vcpu) ++vcpu->stat.irq_injections; if (vmx->rmode.vm86_active) { - if (kvm_inject_realmode_interrupt(vcpu, irq) != EMULATE_DONE) + int inc_eip = 0; + if (vcpu->arch.interrupt.soft) + inc_eip = vcpu->arch.event_exit_inst_len; + if (kvm_inject_realmode_interrupt(vcpu, irq, inc_eip) != EMULATE_DONE) kvm_make_request(KVM_REQ_TRIPLE_FAULT, vcpu); return; } @@ -2937,8 +3062,9 @@ static void vmx_inject_nmi(struct kvm_vcpu *vcpu) } ++vcpu->stat.nmi_injections; + vmx->nmi_known_unmasked = false; if (vmx->rmode.vm86_active) { - if (kvm_inject_realmode_interrupt(vcpu, NMI_VECTOR) != EMULATE_DONE) + if (kvm_inject_realmode_interrupt(vcpu, NMI_VECTOR, 0) != EMULATE_DONE) kvm_make_request(KVM_REQ_TRIPLE_FAULT, vcpu); return; } @@ -2961,6 +3087,8 @@ static bool vmx_get_nmi_mask(struct kvm_vcpu *vcpu) { if (!cpu_has_virtual_nmis()) return to_vmx(vcpu)->soft_vnmi_blocked; + if (to_vmx(vcpu)->nmi_known_unmasked) + return false; return vmcs_read32(GUEST_INTERRUPTIBILITY_INFO) & GUEST_INTR_STATE_NMI; } @@ -2974,6 +3102,7 @@ static void vmx_set_nmi_mask(struct kvm_vcpu *vcpu, bool masked) vmx->vnmi_blocked_time = 0; } } else { + vmx->nmi_known_unmasked = !masked; if (masked) vmcs_set_bits(GUEST_INTERRUPTIBILITY_INFO, GUEST_INTR_STATE_NMI); @@ -3091,7 +3220,7 @@ static int handle_exception(struct kvm_vcpu *vcpu) enum emulation_result er; vect_info = vmx->idt_vectoring_info; - intr_info = vmcs_read32(VM_EXIT_INTR_INFO); + intr_info = vmx->exit_intr_info; if (is_machine_check(intr_info)) return handle_machine_check(vcpu); @@ -3122,7 +3251,6 @@ static int handle_exception(struct kvm_vcpu *vcpu) } error_code = 0; - rip = kvm_rip_read(vcpu); if (intr_info & INTR_INFO_DELIVER_CODE_MASK) error_code = vmcs_read32(VM_EXIT_INTR_ERROR_CODE); if (is_page_fault(intr_info)) { @@ -3169,6 +3297,7 @@ static int handle_exception(struct kvm_vcpu *vcpu) vmx->vcpu.arch.event_exit_inst_len = vmcs_read32(VM_EXIT_INSTRUCTION_LEN); kvm_run->exit_reason = KVM_EXIT_DEBUG; + rip = kvm_rip_read(vcpu); kvm_run->debug.arch.pc = vmcs_readl(GUEST_CS_BASE) + rip; kvm_run->debug.arch.exception = ex_no; break; @@ -3505,9 +3634,7 @@ static int handle_task_switch(struct kvm_vcpu *vcpu) switch (type) { case INTR_TYPE_NMI_INTR: vcpu->arch.nmi_injected = false; - if (cpu_has_virtual_nmis()) - vmcs_set_bits(GUEST_INTERRUPTIBILITY_INFO, - GUEST_INTR_STATE_NMI); + vmx_set_nmi_mask(vcpu, true); break; case INTR_TYPE_EXT_INTR: case INTR_TYPE_SOFT_INTR: @@ -3867,12 +3994,17 @@ static void update_cr8_intercept(struct kvm_vcpu *vcpu, int tpr, int irr) static void vmx_complete_atomic_exit(struct vcpu_vmx *vmx) { - u32 exit_intr_info = vmx->exit_intr_info; + u32 exit_intr_info; + + if (!(vmx->exit_reason == EXIT_REASON_MCE_DURING_VMENTRY + || vmx->exit_reason == EXIT_REASON_EXCEPTION_NMI)) + return; + + vmx->exit_intr_info = vmcs_read32(VM_EXIT_INTR_INFO); + exit_intr_info = vmx->exit_intr_info; /* Handle machine checks before interrupts are enabled */ - if ((vmx->exit_reason == EXIT_REASON_MCE_DURING_VMENTRY) - || (vmx->exit_reason == EXIT_REASON_EXCEPTION_NMI - && is_machine_check(exit_intr_info))) + if (is_machine_check(exit_intr_info)) kvm_machine_check(); /* We need to handle NMIs before interrupts are enabled */ @@ -3886,7 +4018,7 @@ static void vmx_complete_atomic_exit(struct vcpu_vmx *vmx) static void vmx_recover_nmi_blocking(struct vcpu_vmx *vmx) { - u32 exit_intr_info = vmx->exit_intr_info; + u32 exit_intr_info; bool unblock_nmi; u8 vector; bool idtv_info_valid; @@ -3894,6 +4026,13 @@ static void vmx_recover_nmi_blocking(struct vcpu_vmx *vmx) idtv_info_valid = vmx->idt_vectoring_info & VECTORING_INFO_VALID_MASK; if (cpu_has_virtual_nmis()) { + if (vmx->nmi_known_unmasked) + return; + /* + * Can't use vmx->exit_intr_info since we're not sure what + * the exit reason is. + */ + exit_intr_info = vmcs_read32(VM_EXIT_INTR_INFO); unblock_nmi = (exit_intr_info & INTR_INFO_UNBLOCK_NMI) != 0; vector = exit_intr_info & INTR_INFO_VECTOR_MASK; /* @@ -3910,6 +4049,10 @@ static void vmx_recover_nmi_blocking(struct vcpu_vmx *vmx) vector != DF_VECTOR && !idtv_info_valid) vmcs_set_bits(GUEST_INTERRUPTIBILITY_INFO, GUEST_INTR_STATE_NMI); + else + vmx->nmi_known_unmasked = + !(vmcs_read32(GUEST_INTERRUPTIBILITY_INFO) + & GUEST_INTR_STATE_NMI); } else if (unlikely(vmx->soft_vnmi_blocked)) vmx->vnmi_blocked_time += ktime_to_ns(ktime_sub(ktime_get(), vmx->entry_time)); @@ -3946,8 +4089,7 @@ static void __vmx_complete_interrupts(struct vcpu_vmx *vmx, * Clear bit "block by NMI" before VM entry if a NMI * delivery faulted. */ - vmcs_clear_bits(GUEST_INTERRUPTIBILITY_INFO, - GUEST_INTR_STATE_NMI); + vmx_set_nmi_mask(&vmx->vcpu, false); break; case INTR_TYPE_SOFT_EXCEPTION: vmx->vcpu.arch.event_exit_inst_len = @@ -4124,7 +4266,10 @@ static void __noclone vmx_vcpu_run(struct kvm_vcpu *vcpu) ); vcpu->arch.regs_avail = ~((1 << VCPU_REGS_RIP) | (1 << VCPU_REGS_RSP) + | (1 << VCPU_EXREG_RFLAGS) + | (1 << VCPU_EXREG_CPL) | (1 << VCPU_EXREG_PDPTR) + | (1 << VCPU_EXREG_SEGMENTS) | (1 << VCPU_EXREG_CR3)); vcpu->arch.regs_dirty = 0; @@ -4134,7 +4279,6 @@ static void __noclone vmx_vcpu_run(struct kvm_vcpu *vcpu) vmx->launched = 1; vmx->exit_reason = vmcs_read32(VM_EXIT_REASON); - vmx->exit_intr_info = vmcs_read32(VM_EXIT_INTR_INFO); vmx_complete_atomic_exit(vmx); vmx_recover_nmi_blocking(vmx); @@ -4195,8 +4339,8 @@ static struct kvm_vcpu *vmx_create_vcpu(struct kvm *kvm, unsigned int id) goto free_vcpu; vmx->guest_msrs = kmalloc(PAGE_SIZE, GFP_KERNEL); + err = -ENOMEM; if (!vmx->guest_msrs) { - err = -ENOMEM; goto uninit_vcpu; } @@ -4215,7 +4359,8 @@ static struct kvm_vcpu *vmx_create_vcpu(struct kvm *kvm, unsigned int id) if (err) goto free_vmcs; if (vm_need_virtualize_apic_accesses(kvm)) - if (alloc_apic_access_page(kvm) != 0) + err = alloc_apic_access_page(kvm); + if (err) goto free_vmcs; if (enable_ept) { @@ -4368,6 +4513,13 @@ static void vmx_set_supported_cpuid(u32 func, struct kvm_cpuid_entry2 *entry) { } +static int vmx_check_intercept(struct kvm_vcpu *vcpu, + struct x86_instruction_info *info, + enum x86_intercept_stage stage) +{ + return X86EMUL_CONTINUE; +} + static struct kvm_x86_ops vmx_x86_ops = { .cpu_has_kvm_support = cpu_has_kvm_support, .disabled_by_bios = vmx_disabled_by_bios, @@ -4449,10 +4601,14 @@ static struct kvm_x86_ops vmx_x86_ops = { .has_wbinvd_exit = cpu_has_vmx_wbinvd_exit, + .set_tsc_khz = vmx_set_tsc_khz, .write_tsc_offset = vmx_write_tsc_offset, .adjust_tsc_offset = vmx_adjust_tsc_offset, + .compute_tsc_offset = vmx_compute_tsc_offset, .set_tdp_cr3 = vmx_set_cr3, + + .check_intercept = vmx_check_intercept, }; static int __init vmx_init(void) diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 934b4c6b0bf9..77c9d8673dc4 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -60,22 +60,12 @@ #include <asm/div64.h> #define MAX_IO_MSRS 256 -#define CR0_RESERVED_BITS \ - (~(unsigned long)(X86_CR0_PE | X86_CR0_MP | X86_CR0_EM | X86_CR0_TS \ - | X86_CR0_ET | X86_CR0_NE | X86_CR0_WP | X86_CR0_AM \ - | X86_CR0_NW | X86_CR0_CD | X86_CR0_PG)) -#define CR4_RESERVED_BITS \ - (~(unsigned long)(X86_CR4_VME | X86_CR4_PVI | X86_CR4_TSD | X86_CR4_DE\ - | X86_CR4_PSE | X86_CR4_PAE | X86_CR4_MCE \ - | X86_CR4_PGE | X86_CR4_PCE | X86_CR4_OSFXSR \ - | X86_CR4_OSXSAVE \ - | X86_CR4_OSXMMEXCPT | X86_CR4_VMXE)) - -#define CR8_RESERVED_BITS (~(unsigned long)X86_CR8_TPR) - #define KVM_MAX_MCE_BANKS 32 #define KVM_MCE_CAP_SUPPORTED (MCG_CTL_P | MCG_SER_P) +#define emul_to_vcpu(ctxt) \ + container_of(ctxt, struct kvm_vcpu, arch.emulate_ctxt) + /* EFER defaults: * - enable syscall per default because its emulated by KVM * - enable LME and LMA per default on 64 bit KVM @@ -100,6 +90,11 @@ EXPORT_SYMBOL_GPL(kvm_x86_ops); int ignore_msrs = 0; module_param_named(ignore_msrs, ignore_msrs, bool, S_IRUGO | S_IWUSR); +bool kvm_has_tsc_control; +EXPORT_SYMBOL_GPL(kvm_has_tsc_control); +u32 kvm_max_guest_tsc_khz; +EXPORT_SYMBOL_GPL(kvm_max_guest_tsc_khz); + #define KVM_NR_SHARED_MSRS 16 struct kvm_shared_msrs_global { @@ -157,6 +152,8 @@ struct kvm_stats_debugfs_item debugfs_entries[] = { u64 __read_mostly host_xcr0; +int emulator_fix_hypercall(struct x86_emulate_ctxt *ctxt); + static inline void kvm_async_pf_hash_reset(struct kvm_vcpu *vcpu) { int i; @@ -361,8 +358,8 @@ void kvm_propagate_fault(struct kvm_vcpu *vcpu, struct x86_exception *fault) void kvm_inject_nmi(struct kvm_vcpu *vcpu) { - kvm_make_request(KVM_REQ_NMI, vcpu); kvm_make_request(KVM_REQ_EVENT, vcpu); + vcpu->arch.nmi_pending = 1; } EXPORT_SYMBOL_GPL(kvm_inject_nmi); @@ -982,7 +979,15 @@ static inline int kvm_tsc_changes_freq(void) return ret; } -static inline u64 nsec_to_cycles(u64 nsec) +static u64 vcpu_tsc_khz(struct kvm_vcpu *vcpu) +{ + if (vcpu->arch.virtual_tsc_khz) + return vcpu->arch.virtual_tsc_khz; + else + return __this_cpu_read(cpu_tsc_khz); +} + +static inline u64 nsec_to_cycles(struct kvm_vcpu *vcpu, u64 nsec) { u64 ret; @@ -990,25 +995,24 @@ static inline u64 nsec_to_cycles(u64 nsec) if (kvm_tsc_changes_freq()) printk_once(KERN_WARNING "kvm: unreliable cycle conversion on adjustable rate TSC\n"); - ret = nsec * __this_cpu_read(cpu_tsc_khz); + ret = nsec * vcpu_tsc_khz(vcpu); do_div(ret, USEC_PER_SEC); return ret; } -static void kvm_arch_set_tsc_khz(struct kvm *kvm, u32 this_tsc_khz) +static void kvm_init_tsc_catchup(struct kvm_vcpu *vcpu, u32 this_tsc_khz) { /* Compute a scale to convert nanoseconds in TSC cycles */ kvm_get_time_scale(this_tsc_khz, NSEC_PER_SEC / 1000, - &kvm->arch.virtual_tsc_shift, - &kvm->arch.virtual_tsc_mult); - kvm->arch.virtual_tsc_khz = this_tsc_khz; + &vcpu->arch.tsc_catchup_shift, + &vcpu->arch.tsc_catchup_mult); } static u64 compute_guest_tsc(struct kvm_vcpu *vcpu, s64 kernel_ns) { u64 tsc = pvclock_scale_delta(kernel_ns-vcpu->arch.last_tsc_nsec, - vcpu->kvm->arch.virtual_tsc_mult, - vcpu->kvm->arch.virtual_tsc_shift); + vcpu->arch.tsc_catchup_mult, + vcpu->arch.tsc_catchup_shift); tsc += vcpu->arch.last_tsc_write; return tsc; } @@ -1021,7 +1025,7 @@ void kvm_write_tsc(struct kvm_vcpu *vcpu, u64 data) s64 sdiff; raw_spin_lock_irqsave(&kvm->arch.tsc_write_lock, flags); - offset = data - native_read_tsc(); + offset = kvm_x86_ops->compute_tsc_offset(vcpu, data); ns = get_kernel_ns(); elapsed = ns - kvm->arch.last_tsc_nsec; sdiff = data - kvm->arch.last_tsc_write; @@ -1037,13 +1041,13 @@ void kvm_write_tsc(struct kvm_vcpu *vcpu, u64 data) * In that case, for a reliable TSC, we can match TSC offsets, * or make a best guest using elapsed value. */ - if (sdiff < nsec_to_cycles(5ULL * NSEC_PER_SEC) && + if (sdiff < nsec_to_cycles(vcpu, 5ULL * NSEC_PER_SEC) && elapsed < 5ULL * NSEC_PER_SEC) { if (!check_tsc_unstable()) { offset = kvm->arch.last_tsc_offset; pr_debug("kvm: matched tsc offset for %llu\n", data); } else { - u64 delta = nsec_to_cycles(elapsed); + u64 delta = nsec_to_cycles(vcpu, elapsed); offset += delta; pr_debug("kvm: adjusted tsc offset by %llu\n", delta); } @@ -1075,8 +1079,7 @@ static int kvm_guest_time_update(struct kvm_vcpu *v) local_irq_save(flags); kvm_get_msr(v, MSR_IA32_TSC, &tsc_timestamp); kernel_ns = get_kernel_ns(); - this_tsc_khz = __this_cpu_read(cpu_tsc_khz); - + this_tsc_khz = vcpu_tsc_khz(v); if (unlikely(this_tsc_khz == 0)) { local_irq_restore(flags); kvm_make_request(KVM_REQ_CLOCK_UPDATE, v); @@ -1993,6 +1996,7 @@ int kvm_dev_ioctl_check_extension(long ext) case KVM_CAP_X86_ROBUST_SINGLESTEP: case KVM_CAP_XSAVE: case KVM_CAP_ASYNC_PF: + case KVM_CAP_GET_TSC_KHZ: r = 1; break; case KVM_CAP_COALESCED_MMIO: @@ -2019,6 +2023,9 @@ int kvm_dev_ioctl_check_extension(long ext) case KVM_CAP_XCRS: r = cpu_has_xsave; break; + case KVM_CAP_TSC_CONTROL: + r = kvm_has_tsc_control; + break; default: r = 0; break; @@ -2120,8 +2127,13 @@ void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu) kvm_x86_ops->vcpu_load(vcpu, cpu); if (unlikely(vcpu->cpu != cpu) || check_tsc_unstable()) { /* Make sure TSC doesn't go backwards */ - s64 tsc_delta = !vcpu->arch.last_host_tsc ? 0 : - native_read_tsc() - vcpu->arch.last_host_tsc; + s64 tsc_delta; + u64 tsc; + + kvm_get_msr(vcpu, MSR_IA32_TSC, &tsc); + tsc_delta = !vcpu->arch.last_guest_tsc ? 0 : + tsc - vcpu->arch.last_guest_tsc; + if (tsc_delta < 0) mark_tsc_unstable("KVM discovered backwards TSC"); if (check_tsc_unstable()) { @@ -2139,7 +2151,7 @@ void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu) { kvm_x86_ops->vcpu_put(vcpu); kvm_put_guest_fpu(vcpu); - vcpu->arch.last_host_tsc = native_read_tsc(); + kvm_get_msr(vcpu, MSR_IA32_TSC, &vcpu->arch.last_guest_tsc); } static int is_efer_nx(void) @@ -2324,6 +2336,12 @@ static void do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function, F(3DNOWPREFETCH) | 0 /* OSVW */ | 0 /* IBS */ | F(XOP) | 0 /* SKINIT, WDT, LWP */ | F(FMA4) | F(TBM); + /* cpuid 0xC0000001.edx */ + const u32 kvm_supported_word5_x86_features = + F(XSTORE) | F(XSTORE_EN) | F(XCRYPT) | F(XCRYPT_EN) | + F(ACE2) | F(ACE2_EN) | F(PHE) | F(PHE_EN) | + F(PMM) | F(PMM_EN); + /* all calls to cpuid_count() should be made on the same cpu */ get_cpu(); do_cpuid_1_ent(entry, function, index); @@ -2418,6 +2436,7 @@ static void do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function, entry->eax = (1 << KVM_FEATURE_CLOCKSOURCE) | (1 << KVM_FEATURE_NOP_IO_DELAY) | (1 << KVM_FEATURE_CLOCKSOURCE2) | + (1 << KVM_FEATURE_ASYNC_PF) | (1 << KVM_FEATURE_CLOCKSOURCE_STABLE_BIT); entry->ebx = 0; entry->ecx = 0; @@ -2432,6 +2451,20 @@ static void do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function, entry->ecx &= kvm_supported_word6_x86_features; cpuid_mask(&entry->ecx, 6); break; + /*Add support for Centaur's CPUID instruction*/ + case 0xC0000000: + /*Just support up to 0xC0000004 now*/ + entry->eax = min(entry->eax, 0xC0000004); + break; + case 0xC0000001: + entry->edx &= kvm_supported_word5_x86_features; + cpuid_mask(&entry->edx, 5); + break; + case 0xC0000002: + case 0xC0000003: + case 0xC0000004: + /*Now nothing to do, reserved for the future*/ + break; } kvm_x86_ops->set_supported_cpuid(function, entry); @@ -2478,6 +2511,26 @@ static int kvm_dev_ioctl_get_supported_cpuid(struct kvm_cpuid2 *cpuid, if (nent >= cpuid->nent) goto out_free; + /* Add support for Centaur's CPUID instruction. */ + if (boot_cpu_data.x86_vendor == X86_VENDOR_CENTAUR) { + do_cpuid_ent(&cpuid_entries[nent], 0xC0000000, 0, + &nent, cpuid->nent); + + r = -E2BIG; + if (nent >= cpuid->nent) + goto out_free; + + limit = cpuid_entries[nent - 1].eax; + for (func = 0xC0000001; + func <= limit && nent < cpuid->nent; ++func) + do_cpuid_ent(&cpuid_entries[nent], func, 0, + &nent, cpuid->nent); + + r = -E2BIG; + if (nent >= cpuid->nent) + goto out_free; + } + do_cpuid_ent(&cpuid_entries[nent], KVM_CPUID_SIGNATURE, 0, &nent, cpuid->nent); @@ -3046,6 +3099,32 @@ long kvm_arch_vcpu_ioctl(struct file *filp, r = kvm_vcpu_ioctl_x86_set_xcrs(vcpu, u.xcrs); break; } + case KVM_SET_TSC_KHZ: { + u32 user_tsc_khz; + + r = -EINVAL; + if (!kvm_has_tsc_control) + break; + + user_tsc_khz = (u32)arg; + + if (user_tsc_khz >= kvm_max_guest_tsc_khz) + goto out; + + kvm_x86_ops->set_tsc_khz(vcpu, user_tsc_khz); + + r = 0; + goto out; + } + case KVM_GET_TSC_KHZ: { + r = -EIO; + if (check_tsc_unstable()) + goto out; + + r = vcpu_tsc_khz(vcpu); + + goto out; + } default: r = -EINVAL; } @@ -3595,20 +3674,43 @@ static void kvm_init_msr_list(void) static int vcpu_mmio_write(struct kvm_vcpu *vcpu, gpa_t addr, int len, const void *v) { - if (vcpu->arch.apic && - !kvm_iodevice_write(&vcpu->arch.apic->dev, addr, len, v)) - return 0; + int handled = 0; + int n; - return kvm_io_bus_write(vcpu->kvm, KVM_MMIO_BUS, addr, len, v); + do { + n = min(len, 8); + if (!(vcpu->arch.apic && + !kvm_iodevice_write(&vcpu->arch.apic->dev, addr, n, v)) + && kvm_io_bus_write(vcpu->kvm, KVM_MMIO_BUS, addr, n, v)) + break; + handled += n; + addr += n; + len -= n; + v += n; + } while (len); + + return handled; } static int vcpu_mmio_read(struct kvm_vcpu *vcpu, gpa_t addr, int len, void *v) { - if (vcpu->arch.apic && - !kvm_iodevice_read(&vcpu->arch.apic->dev, addr, len, v)) - return 0; + int handled = 0; + int n; + + do { + n = min(len, 8); + if (!(vcpu->arch.apic && + !kvm_iodevice_read(&vcpu->arch.apic->dev, addr, n, v)) + && kvm_io_bus_read(vcpu->kvm, KVM_MMIO_BUS, addr, n, v)) + break; + trace_kvm_mmio(KVM_TRACE_MMIO_READ, n, addr, *(u64 *)v); + handled += n; + addr += n; + len -= n; + v += n; + } while (len); - return kvm_io_bus_read(vcpu->kvm, KVM_MMIO_BUS, addr, len, v); + return handled; } static void kvm_set_segment(struct kvm_vcpu *vcpu, @@ -3703,37 +3805,43 @@ out: } /* used for instruction fetching */ -static int kvm_fetch_guest_virt(gva_t addr, void *val, unsigned int bytes, - struct kvm_vcpu *vcpu, +static int kvm_fetch_guest_virt(struct x86_emulate_ctxt *ctxt, + gva_t addr, void *val, unsigned int bytes, struct x86_exception *exception) { + struct kvm_vcpu *vcpu = emul_to_vcpu(ctxt); u32 access = (kvm_x86_ops->get_cpl(vcpu) == 3) ? PFERR_USER_MASK : 0; + return kvm_read_guest_virt_helper(addr, val, bytes, vcpu, access | PFERR_FETCH_MASK, exception); } -static int kvm_read_guest_virt(gva_t addr, void *val, unsigned int bytes, - struct kvm_vcpu *vcpu, +static int kvm_read_guest_virt(struct x86_emulate_ctxt *ctxt, + gva_t addr, void *val, unsigned int bytes, struct x86_exception *exception) { + struct kvm_vcpu *vcpu = emul_to_vcpu(ctxt); u32 access = (kvm_x86_ops->get_cpl(vcpu) == 3) ? PFERR_USER_MASK : 0; + return kvm_read_guest_virt_helper(addr, val, bytes, vcpu, access, exception); } -static int kvm_read_guest_virt_system(gva_t addr, void *val, unsigned int bytes, - struct kvm_vcpu *vcpu, +static int kvm_read_guest_virt_system(struct x86_emulate_ctxt *ctxt, + gva_t addr, void *val, unsigned int bytes, struct x86_exception *exception) { + struct kvm_vcpu *vcpu = emul_to_vcpu(ctxt); return kvm_read_guest_virt_helper(addr, val, bytes, vcpu, 0, exception); } -static int kvm_write_guest_virt_system(gva_t addr, void *val, +static int kvm_write_guest_virt_system(struct x86_emulate_ctxt *ctxt, + gva_t addr, void *val, unsigned int bytes, - struct kvm_vcpu *vcpu, struct x86_exception *exception) { + struct kvm_vcpu *vcpu = emul_to_vcpu(ctxt); void *data = val; int r = X86EMUL_CONTINUE; @@ -3761,13 +3869,15 @@ out: return r; } -static int emulator_read_emulated(unsigned long addr, +static int emulator_read_emulated(struct x86_emulate_ctxt *ctxt, + unsigned long addr, void *val, unsigned int bytes, - struct x86_exception *exception, - struct kvm_vcpu *vcpu) + struct x86_exception *exception) { + struct kvm_vcpu *vcpu = emul_to_vcpu(ctxt); gpa_t gpa; + int handled; if (vcpu->mmio_read_completed) { memcpy(val, vcpu->mmio_data, bytes); @@ -3786,7 +3896,7 @@ static int emulator_read_emulated(unsigned long addr, if ((gpa & PAGE_MASK) == APIC_DEFAULT_PHYS_BASE) goto mmio; - if (kvm_read_guest_virt(addr, val, bytes, vcpu, exception) + if (kvm_read_guest_virt(ctxt, addr, val, bytes, exception) == X86EMUL_CONTINUE) return X86EMUL_CONTINUE; @@ -3794,18 +3904,24 @@ mmio: /* * Is this MMIO handled locally? */ - if (!vcpu_mmio_read(vcpu, gpa, bytes, val)) { - trace_kvm_mmio(KVM_TRACE_MMIO_READ, bytes, gpa, *(u64 *)val); + handled = vcpu_mmio_read(vcpu, gpa, bytes, val); + + if (handled == bytes) return X86EMUL_CONTINUE; - } + + gpa += handled; + bytes -= handled; + val += handled; trace_kvm_mmio(KVM_TRACE_MMIO_READ_UNSATISFIED, bytes, gpa, 0); vcpu->mmio_needed = 1; vcpu->run->exit_reason = KVM_EXIT_MMIO; vcpu->run->mmio.phys_addr = vcpu->mmio_phys_addr = gpa; - vcpu->run->mmio.len = vcpu->mmio_size = bytes; + vcpu->mmio_size = bytes; + vcpu->run->mmio.len = min(vcpu->mmio_size, 8); vcpu->run->mmio.is_write = vcpu->mmio_is_write = 0; + vcpu->mmio_index = 0; return X86EMUL_IO_NEEDED; } @@ -3829,6 +3945,7 @@ static int emulator_write_emulated_onepage(unsigned long addr, struct kvm_vcpu *vcpu) { gpa_t gpa; + int handled; gpa = kvm_mmu_gva_to_gpa_write(vcpu, addr, exception); @@ -3847,25 +3964,35 @@ mmio: /* * Is this MMIO handled locally? */ - if (!vcpu_mmio_write(vcpu, gpa, bytes, val)) + handled = vcpu_mmio_write(vcpu, gpa, bytes, val); + if (handled == bytes) return X86EMUL_CONTINUE; + gpa += handled; + bytes -= handled; + val += handled; + vcpu->mmio_needed = 1; + memcpy(vcpu->mmio_data, val, bytes); vcpu->run->exit_reason = KVM_EXIT_MMIO; vcpu->run->mmio.phys_addr = vcpu->mmio_phys_addr = gpa; - vcpu->run->mmio.len = vcpu->mmio_size = bytes; + vcpu->mmio_size = bytes; + vcpu->run->mmio.len = min(vcpu->mmio_size, 8); vcpu->run->mmio.is_write = vcpu->mmio_is_write = 1; - memcpy(vcpu->run->mmio.data, val, bytes); + memcpy(vcpu->run->mmio.data, vcpu->mmio_data, 8); + vcpu->mmio_index = 0; return X86EMUL_CONTINUE; } -int emulator_write_emulated(unsigned long addr, +int emulator_write_emulated(struct x86_emulate_ctxt *ctxt, + unsigned long addr, const void *val, unsigned int bytes, - struct x86_exception *exception, - struct kvm_vcpu *vcpu) + struct x86_exception *exception) { + struct kvm_vcpu *vcpu = emul_to_vcpu(ctxt); + /* Crossing a page boundary? */ if (((addr + bytes - 1) ^ addr) & PAGE_MASK) { int rc, now; @@ -3893,13 +4020,14 @@ int emulator_write_emulated(unsigned long addr, (cmpxchg64((u64 *)(ptr), *(u64 *)(old), *(u64 *)(new)) == *(u64 *)(old)) #endif -static int emulator_cmpxchg_emulated(unsigned long addr, +static int emulator_cmpxchg_emulated(struct x86_emulate_ctxt *ctxt, + unsigned long addr, const void *old, const void *new, unsigned int bytes, - struct x86_exception *exception, - struct kvm_vcpu *vcpu) + struct x86_exception *exception) { + struct kvm_vcpu *vcpu = emul_to_vcpu(ctxt); gpa_t gpa; struct page *page; char *kaddr; @@ -3955,7 +4083,7 @@ static int emulator_cmpxchg_emulated(unsigned long addr, emul_write: printk_once(KERN_WARNING "kvm: emulating exchange as write\n"); - return emulator_write_emulated(addr, new, bytes, exception, vcpu); + return emulator_write_emulated(ctxt, addr, new, bytes, exception); } static int kernel_pio(struct kvm_vcpu *vcpu, void *pd) @@ -3974,9 +4102,12 @@ static int kernel_pio(struct kvm_vcpu *vcpu, void *pd) } -static int emulator_pio_in_emulated(int size, unsigned short port, void *val, - unsigned int count, struct kvm_vcpu *vcpu) +static int emulator_pio_in_emulated(struct x86_emulate_ctxt *ctxt, + int size, unsigned short port, void *val, + unsigned int count) { + struct kvm_vcpu *vcpu = emul_to_vcpu(ctxt); + if (vcpu->arch.pio.count) goto data_avail; @@ -4004,10 +4135,12 @@ static int emulator_pio_in_emulated(int size, unsigned short port, void *val, return 0; } -static int emulator_pio_out_emulated(int size, unsigned short port, - const void *val, unsigned int count, - struct kvm_vcpu *vcpu) +static int emulator_pio_out_emulated(struct x86_emulate_ctxt *ctxt, + int size, unsigned short port, + const void *val, unsigned int count) { + struct kvm_vcpu *vcpu = emul_to_vcpu(ctxt); + trace_kvm_pio(1, port, size, count); vcpu->arch.pio.port = port; @@ -4037,10 +4170,9 @@ static unsigned long get_segment_base(struct kvm_vcpu *vcpu, int seg) return kvm_x86_ops->get_segment_base(vcpu, seg); } -int emulate_invlpg(struct kvm_vcpu *vcpu, gva_t address) +static void emulator_invlpg(struct x86_emulate_ctxt *ctxt, ulong address) { - kvm_mmu_invlpg(vcpu, address); - return X86EMUL_CONTINUE; + kvm_mmu_invlpg(emul_to_vcpu(ctxt), address); } int kvm_emulate_wbinvd(struct kvm_vcpu *vcpu) @@ -4062,22 +4194,20 @@ int kvm_emulate_wbinvd(struct kvm_vcpu *vcpu) } EXPORT_SYMBOL_GPL(kvm_emulate_wbinvd); -int emulate_clts(struct kvm_vcpu *vcpu) +static void emulator_wbinvd(struct x86_emulate_ctxt *ctxt) { - kvm_x86_ops->set_cr0(vcpu, kvm_read_cr0_bits(vcpu, ~X86_CR0_TS)); - kvm_x86_ops->fpu_activate(vcpu); - return X86EMUL_CONTINUE; + kvm_emulate_wbinvd(emul_to_vcpu(ctxt)); } -int emulator_get_dr(int dr, unsigned long *dest, struct kvm_vcpu *vcpu) +int emulator_get_dr(struct x86_emulate_ctxt *ctxt, int dr, unsigned long *dest) { - return _kvm_get_dr(vcpu, dr, dest); + return _kvm_get_dr(emul_to_vcpu(ctxt), dr, dest); } -int emulator_set_dr(int dr, unsigned long value, struct kvm_vcpu *vcpu) +int emulator_set_dr(struct x86_emulate_ctxt *ctxt, int dr, unsigned long value) { - return __kvm_set_dr(vcpu, dr, value); + return __kvm_set_dr(emul_to_vcpu(ctxt), dr, value); } static u64 mk_cr_64(u64 curr_cr, u32 new_val) @@ -4085,8 +4215,9 @@ static u64 mk_cr_64(u64 curr_cr, u32 new_val) return (curr_cr & ~((1ULL << 32) - 1)) | new_val; } -static unsigned long emulator_get_cr(int cr, struct kvm_vcpu *vcpu) +static unsigned long emulator_get_cr(struct x86_emulate_ctxt *ctxt, int cr) { + struct kvm_vcpu *vcpu = emul_to_vcpu(ctxt); unsigned long value; switch (cr) { @@ -4113,8 +4244,9 @@ static unsigned long emulator_get_cr(int cr, struct kvm_vcpu *vcpu) return value; } -static int emulator_set_cr(int cr, unsigned long val, struct kvm_vcpu *vcpu) +static int emulator_set_cr(struct x86_emulate_ctxt *ctxt, int cr, ulong val) { + struct kvm_vcpu *vcpu = emul_to_vcpu(ctxt); int res = 0; switch (cr) { @@ -4141,33 +4273,45 @@ static int emulator_set_cr(int cr, unsigned long val, struct kvm_vcpu *vcpu) return res; } -static int emulator_get_cpl(struct kvm_vcpu *vcpu) +static int emulator_get_cpl(struct x86_emulate_ctxt *ctxt) +{ + return kvm_x86_ops->get_cpl(emul_to_vcpu(ctxt)); +} + +static void emulator_get_gdt(struct x86_emulate_ctxt *ctxt, struct desc_ptr *dt) +{ + kvm_x86_ops->get_gdt(emul_to_vcpu(ctxt), dt); +} + +static void emulator_get_idt(struct x86_emulate_ctxt *ctxt, struct desc_ptr *dt) { - return kvm_x86_ops->get_cpl(vcpu); + kvm_x86_ops->get_idt(emul_to_vcpu(ctxt), dt); } -static void emulator_get_gdt(struct desc_ptr *dt, struct kvm_vcpu *vcpu) +static void emulator_set_gdt(struct x86_emulate_ctxt *ctxt, struct desc_ptr *dt) { - kvm_x86_ops->get_gdt(vcpu, dt); + kvm_x86_ops->set_gdt(emul_to_vcpu(ctxt), dt); } -static void emulator_get_idt(struct desc_ptr *dt, struct kvm_vcpu *vcpu) +static void emulator_set_idt(struct x86_emulate_ctxt *ctxt, struct desc_ptr *dt) { - kvm_x86_ops->get_idt(vcpu, dt); + kvm_x86_ops->set_idt(emul_to_vcpu(ctxt), dt); } -static unsigned long emulator_get_cached_segment_base(int seg, - struct kvm_vcpu *vcpu) +static unsigned long emulator_get_cached_segment_base( + struct x86_emulate_ctxt *ctxt, int seg) { - return get_segment_base(vcpu, seg); + return get_segment_base(emul_to_vcpu(ctxt), seg); } -static bool emulator_get_cached_descriptor(struct desc_struct *desc, u32 *base3, - int seg, struct kvm_vcpu *vcpu) +static bool emulator_get_segment(struct x86_emulate_ctxt *ctxt, u16 *selector, + struct desc_struct *desc, u32 *base3, + int seg) { struct kvm_segment var; - kvm_get_segment(vcpu, &var, seg); + kvm_get_segment(emul_to_vcpu(ctxt), &var, seg); + *selector = var.selector; if (var.unusable) return false; @@ -4192,14 +4336,14 @@ static bool emulator_get_cached_descriptor(struct desc_struct *desc, u32 *base3, return true; } -static void emulator_set_cached_descriptor(struct desc_struct *desc, u32 base3, - int seg, struct kvm_vcpu *vcpu) +static void emulator_set_segment(struct x86_emulate_ctxt *ctxt, u16 selector, + struct desc_struct *desc, u32 base3, + int seg) { + struct kvm_vcpu *vcpu = emul_to_vcpu(ctxt); struct kvm_segment var; - /* needed to preserve selector */ - kvm_get_segment(vcpu, &var, seg); - + var.selector = selector; var.base = get_desc_base(desc); #ifdef CONFIG_X86_64 var.base |= ((u64)base3) << 32; @@ -4223,22 +4367,44 @@ static void emulator_set_cached_descriptor(struct desc_struct *desc, u32 base3, return; } -static u16 emulator_get_segment_selector(int seg, struct kvm_vcpu *vcpu) +static int emulator_get_msr(struct x86_emulate_ctxt *ctxt, + u32 msr_index, u64 *pdata) { - struct kvm_segment kvm_seg; + return kvm_get_msr(emul_to_vcpu(ctxt), msr_index, pdata); +} - kvm_get_segment(vcpu, &kvm_seg, seg); - return kvm_seg.selector; +static int emulator_set_msr(struct x86_emulate_ctxt *ctxt, + u32 msr_index, u64 data) +{ + return kvm_set_msr(emul_to_vcpu(ctxt), msr_index, data); } -static void emulator_set_segment_selector(u16 sel, int seg, - struct kvm_vcpu *vcpu) +static void emulator_halt(struct x86_emulate_ctxt *ctxt) { - struct kvm_segment kvm_seg; + emul_to_vcpu(ctxt)->arch.halt_request = 1; +} - kvm_get_segment(vcpu, &kvm_seg, seg); - kvm_seg.selector = sel; - kvm_set_segment(vcpu, &kvm_seg, seg); +static void emulator_get_fpu(struct x86_emulate_ctxt *ctxt) +{ + preempt_disable(); + kvm_load_guest_fpu(emul_to_vcpu(ctxt)); + /* + * CR0.TS may reference the host fpu state, not the guest fpu state, + * so it may be clear at this point. + */ + clts(); +} + +static void emulator_put_fpu(struct x86_emulate_ctxt *ctxt) +{ + preempt_enable(); +} + +static int emulator_intercept(struct x86_emulate_ctxt *ctxt, + struct x86_instruction_info *info, + enum x86_intercept_stage stage) +{ + return kvm_x86_ops->check_intercept(emul_to_vcpu(ctxt), info, stage); } static struct x86_emulate_ops emulate_ops = { @@ -4248,22 +4414,29 @@ static struct x86_emulate_ops emulate_ops = { .read_emulated = emulator_read_emulated, .write_emulated = emulator_write_emulated, .cmpxchg_emulated = emulator_cmpxchg_emulated, + .invlpg = emulator_invlpg, .pio_in_emulated = emulator_pio_in_emulated, .pio_out_emulated = emulator_pio_out_emulated, - .get_cached_descriptor = emulator_get_cached_descriptor, - .set_cached_descriptor = emulator_set_cached_descriptor, - .get_segment_selector = emulator_get_segment_selector, - .set_segment_selector = emulator_set_segment_selector, + .get_segment = emulator_get_segment, + .set_segment = emulator_set_segment, .get_cached_segment_base = emulator_get_cached_segment_base, .get_gdt = emulator_get_gdt, .get_idt = emulator_get_idt, + .set_gdt = emulator_set_gdt, + .set_idt = emulator_set_idt, .get_cr = emulator_get_cr, .set_cr = emulator_set_cr, .cpl = emulator_get_cpl, .get_dr = emulator_get_dr, .set_dr = emulator_set_dr, - .set_msr = kvm_set_msr, - .get_msr = kvm_get_msr, + .set_msr = emulator_set_msr, + .get_msr = emulator_get_msr, + .halt = emulator_halt, + .wbinvd = emulator_wbinvd, + .fix_hypercall = emulator_fix_hypercall, + .get_fpu = emulator_get_fpu, + .put_fpu = emulator_put_fpu, + .intercept = emulator_intercept, }; static void cache_all_regs(struct kvm_vcpu *vcpu) @@ -4305,12 +4478,17 @@ static void init_emulate_ctxt(struct kvm_vcpu *vcpu) struct decode_cache *c = &vcpu->arch.emulate_ctxt.decode; int cs_db, cs_l; + /* + * TODO: fix emulate.c to use guest_read/write_register + * instead of direct ->regs accesses, can save hundred cycles + * on Intel for instructions that don't read/change RSP, for + * for example. + */ cache_all_regs(vcpu); kvm_x86_ops->get_cs_db_l_bits(vcpu, &cs_db, &cs_l); - vcpu->arch.emulate_ctxt.vcpu = vcpu; - vcpu->arch.emulate_ctxt.eflags = kvm_x86_ops->get_rflags(vcpu); + vcpu->arch.emulate_ctxt.eflags = kvm_get_rflags(vcpu); vcpu->arch.emulate_ctxt.eip = kvm_rip_read(vcpu); vcpu->arch.emulate_ctxt.mode = (!is_protmode(vcpu)) ? X86EMUL_MODE_REAL : @@ -4318,11 +4496,13 @@ static void init_emulate_ctxt(struct kvm_vcpu *vcpu) ? X86EMUL_MODE_VM86 : cs_l ? X86EMUL_MODE_PROT64 : cs_db ? X86EMUL_MODE_PROT32 : X86EMUL_MODE_PROT16; + vcpu->arch.emulate_ctxt.guest_mode = is_guest_mode(vcpu); memset(c, 0, sizeof(struct decode_cache)); memcpy(c->regs, vcpu->arch.regs, sizeof c->regs); + vcpu->arch.emulate_regs_need_sync_from_vcpu = false; } -int kvm_inject_realmode_interrupt(struct kvm_vcpu *vcpu, int irq) +int kvm_inject_realmode_interrupt(struct kvm_vcpu *vcpu, int irq, int inc_eip) { struct decode_cache *c = &vcpu->arch.emulate_ctxt.decode; int ret; @@ -4331,7 +4511,8 @@ int kvm_inject_realmode_interrupt(struct kvm_vcpu *vcpu, int irq) vcpu->arch.emulate_ctxt.decode.op_bytes = 2; vcpu->arch.emulate_ctxt.decode.ad_bytes = 2; - vcpu->arch.emulate_ctxt.decode.eip = vcpu->arch.emulate_ctxt.eip; + vcpu->arch.emulate_ctxt.decode.eip = vcpu->arch.emulate_ctxt.eip + + inc_eip; ret = emulate_int_real(&vcpu->arch.emulate_ctxt, &emulate_ops, irq); if (ret != X86EMUL_CONTINUE) @@ -4340,7 +4521,7 @@ int kvm_inject_realmode_interrupt(struct kvm_vcpu *vcpu, int irq) vcpu->arch.emulate_ctxt.eip = c->eip; memcpy(vcpu->arch.regs, c->regs, sizeof c->regs); kvm_rip_write(vcpu, vcpu->arch.emulate_ctxt.eip); - kvm_x86_ops->set_rflags(vcpu, vcpu->arch.emulate_ctxt.eflags); + kvm_set_rflags(vcpu, vcpu->arch.emulate_ctxt.eflags); if (irq == NMI_VECTOR) vcpu->arch.nmi_pending = false; @@ -4402,16 +4583,9 @@ int x86_emulate_instruction(struct kvm_vcpu *vcpu, { int r; struct decode_cache *c = &vcpu->arch.emulate_ctxt.decode; + bool writeback = true; kvm_clear_exception_queue(vcpu); - vcpu->arch.mmio_fault_cr2 = cr2; - /* - * TODO: fix emulate.c to use guest_read/write_register - * instead of direct ->regs accesses, can save hundred cycles - * on Intel for instructions that don't read/change RSP, for - * for example. - */ - cache_all_regs(vcpu); if (!(emulation_type & EMULTYPE_NO_DECODE)) { init_emulate_ctxt(vcpu); @@ -4442,13 +4616,19 @@ int x86_emulate_instruction(struct kvm_vcpu *vcpu, return EMULATE_DONE; } - /* this is needed for vmware backdor interface to work since it + /* this is needed for vmware backdoor interface to work since it changes registers values during IO operation */ - memcpy(c->regs, vcpu->arch.regs, sizeof c->regs); + if (vcpu->arch.emulate_regs_need_sync_from_vcpu) { + vcpu->arch.emulate_regs_need_sync_from_vcpu = false; + memcpy(c->regs, vcpu->arch.regs, sizeof c->regs); + } restart: r = x86_emulate_insn(&vcpu->arch.emulate_ctxt); + if (r == EMULATION_INTERCEPTED) + return EMULATE_DONE; + if (r == EMULATION_FAILED) { if (reexecute_instruction(vcpu, cr2)) return EMULATE_DONE; @@ -4462,21 +4642,28 @@ restart: } else if (vcpu->arch.pio.count) { if (!vcpu->arch.pio.in) vcpu->arch.pio.count = 0; + else + writeback = false; r = EMULATE_DO_MMIO; } else if (vcpu->mmio_needed) { - if (vcpu->mmio_is_write) - vcpu->mmio_needed = 0; + if (!vcpu->mmio_is_write) + writeback = false; r = EMULATE_DO_MMIO; } else if (r == EMULATION_RESTART) goto restart; else r = EMULATE_DONE; - toggle_interruptibility(vcpu, vcpu->arch.emulate_ctxt.interruptibility); - kvm_x86_ops->set_rflags(vcpu, vcpu->arch.emulate_ctxt.eflags); - kvm_make_request(KVM_REQ_EVENT, vcpu); - memcpy(vcpu->arch.regs, c->regs, sizeof c->regs); - kvm_rip_write(vcpu, vcpu->arch.emulate_ctxt.eip); + if (writeback) { + toggle_interruptibility(vcpu, + vcpu->arch.emulate_ctxt.interruptibility); + kvm_set_rflags(vcpu, vcpu->arch.emulate_ctxt.eflags); + kvm_make_request(KVM_REQ_EVENT, vcpu); + memcpy(vcpu->arch.regs, c->regs, sizeof c->regs); + vcpu->arch.emulate_regs_need_sync_to_vcpu = false; + kvm_rip_write(vcpu, vcpu->arch.emulate_ctxt.eip); + } else + vcpu->arch.emulate_regs_need_sync_to_vcpu = true; return r; } @@ -4485,7 +4672,8 @@ EXPORT_SYMBOL_GPL(x86_emulate_instruction); int kvm_fast_pio_out(struct kvm_vcpu *vcpu, int size, unsigned short port) { unsigned long val = kvm_register_read(vcpu, VCPU_REGS_RAX); - int ret = emulator_pio_out_emulated(size, port, &val, 1, vcpu); + int ret = emulator_pio_out_emulated(&vcpu->arch.emulate_ctxt, + size, port, &val, 1); /* do not return to emulator after return from userspace */ vcpu->arch.pio.count = 0; return ret; @@ -4879,8 +5067,9 @@ out: } EXPORT_SYMBOL_GPL(kvm_emulate_hypercall); -int kvm_fix_hypercall(struct kvm_vcpu *vcpu) +int emulator_fix_hypercall(struct x86_emulate_ctxt *ctxt) { + struct kvm_vcpu *vcpu = emul_to_vcpu(ctxt); char instruction[3]; unsigned long rip = kvm_rip_read(vcpu); @@ -4893,21 +5082,8 @@ int kvm_fix_hypercall(struct kvm_vcpu *vcpu) kvm_x86_ops->patch_hypercall(vcpu, instruction); - return emulator_write_emulated(rip, instruction, 3, NULL, vcpu); -} - -void realmode_lgdt(struct kvm_vcpu *vcpu, u16 limit, unsigned long base) -{ - struct desc_ptr dt = { limit, base }; - - kvm_x86_ops->set_gdt(vcpu, &dt); -} - -void realmode_lidt(struct kvm_vcpu *vcpu, u16 limit, unsigned long base) -{ - struct desc_ptr dt = { limit, base }; - - kvm_x86_ops->set_idt(vcpu, &dt); + return emulator_write_emulated(&vcpu->arch.emulate_ctxt, + rip, instruction, 3, NULL); } static int move_to_next_stateful_cpuid_entry(struct kvm_vcpu *vcpu, int i) @@ -5170,6 +5346,7 @@ static void kvm_put_guest_xcr0(struct kvm_vcpu *vcpu) static int vcpu_enter_guest(struct kvm_vcpu *vcpu) { int r; + bool nmi_pending; bool req_int_win = !irqchip_in_kernel(vcpu->kvm) && vcpu->run->request_interrupt_window; @@ -5207,19 +5384,25 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu) r = 1; goto out; } - if (kvm_check_request(KVM_REQ_NMI, vcpu)) - vcpu->arch.nmi_pending = true; } r = kvm_mmu_reload(vcpu); if (unlikely(r)) goto out; + /* + * An NMI can be injected between local nmi_pending read and + * vcpu->arch.nmi_pending read inside inject_pending_event(). + * But in that case, KVM_REQ_EVENT will be set, which makes + * the race described above benign. + */ + nmi_pending = ACCESS_ONCE(vcpu->arch.nmi_pending); + if (kvm_check_request(KVM_REQ_EVENT, vcpu) || req_int_win) { inject_pending_event(vcpu); /* enable NMI/IRQ window open exits if needed */ - if (vcpu->arch.nmi_pending) + if (nmi_pending) kvm_x86_ops->enable_nmi_window(vcpu); else if (kvm_cpu_has_interrupt(vcpu) || req_int_win) kvm_x86_ops->enable_irq_window(vcpu); @@ -5399,6 +5582,41 @@ static int __vcpu_run(struct kvm_vcpu *vcpu) return r; } +static int complete_mmio(struct kvm_vcpu *vcpu) +{ + struct kvm_run *run = vcpu->run; + int r; + + if (!(vcpu->arch.pio.count || vcpu->mmio_needed)) + return 1; + + if (vcpu->mmio_needed) { + vcpu->mmio_needed = 0; + if (!vcpu->mmio_is_write) + memcpy(vcpu->mmio_data + vcpu->mmio_index, + run->mmio.data, 8); + vcpu->mmio_index += 8; + if (vcpu->mmio_index < vcpu->mmio_size) { + run->exit_reason = KVM_EXIT_MMIO; + run->mmio.phys_addr = vcpu->mmio_phys_addr + vcpu->mmio_index; + memcpy(run->mmio.data, vcpu->mmio_data + vcpu->mmio_index, 8); + run->mmio.len = min(vcpu->mmio_size - vcpu->mmio_index, 8); + run->mmio.is_write = vcpu->mmio_is_write; + vcpu->mmio_needed = 1; + return 0; + } + if (vcpu->mmio_is_write) + return 1; + vcpu->mmio_read_completed = 1; + } + vcpu->srcu_idx = srcu_read_lock(&vcpu->kvm->srcu); + r = emulate_instruction(vcpu, EMULTYPE_NO_DECODE); + srcu_read_unlock(&vcpu->kvm->srcu, vcpu->srcu_idx); + if (r != EMULATE_DONE) + return 0; + return 1; +} + int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) { int r; @@ -5425,20 +5643,10 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) } } - if (vcpu->arch.pio.count || vcpu->mmio_needed) { - if (vcpu->mmio_needed) { - memcpy(vcpu->mmio_data, kvm_run->mmio.data, 8); - vcpu->mmio_read_completed = 1; - vcpu->mmio_needed = 0; - } - vcpu->srcu_idx = srcu_read_lock(&vcpu->kvm->srcu); - r = emulate_instruction(vcpu, EMULTYPE_NO_DECODE); - srcu_read_unlock(&vcpu->kvm->srcu, vcpu->srcu_idx); - if (r != EMULATE_DONE) { - r = 0; - goto out; - } - } + r = complete_mmio(vcpu); + if (r <= 0) + goto out; + if (kvm_run->exit_reason == KVM_EXIT_HYPERCALL) kvm_register_write(vcpu, VCPU_REGS_RAX, kvm_run->hypercall.ret); @@ -5455,6 +5663,18 @@ out: int kvm_arch_vcpu_ioctl_get_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs) { + if (vcpu->arch.emulate_regs_need_sync_to_vcpu) { + /* + * We are here if userspace calls get_regs() in the middle of + * instruction emulation. Registers state needs to be copied + * back from emulation context to vcpu. Usrapace shouldn't do + * that usually, but some bad designed PV devices (vmware + * backdoor interface) need this to work + */ + struct decode_cache *c = &vcpu->arch.emulate_ctxt.decode; + memcpy(vcpu->arch.regs, c->regs, sizeof c->regs); + vcpu->arch.emulate_regs_need_sync_to_vcpu = false; + } regs->rax = kvm_register_read(vcpu, VCPU_REGS_RAX); regs->rbx = kvm_register_read(vcpu, VCPU_REGS_RBX); regs->rcx = kvm_register_read(vcpu, VCPU_REGS_RCX); @@ -5482,6 +5702,9 @@ int kvm_arch_vcpu_ioctl_get_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs) int kvm_arch_vcpu_ioctl_set_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs) { + vcpu->arch.emulate_regs_need_sync_from_vcpu = true; + vcpu->arch.emulate_regs_need_sync_to_vcpu = false; + kvm_register_write(vcpu, VCPU_REGS_RAX, regs->rax); kvm_register_write(vcpu, VCPU_REGS_RBX, regs->rbx); kvm_register_write(vcpu, VCPU_REGS_RCX, regs->rcx); @@ -5592,7 +5815,7 @@ int kvm_task_switch(struct kvm_vcpu *vcpu, u16 tss_selector, int reason, memcpy(vcpu->arch.regs, c->regs, sizeof c->regs); kvm_rip_write(vcpu, vcpu->arch.emulate_ctxt.eip); - kvm_x86_ops->set_rflags(vcpu, vcpu->arch.emulate_ctxt.eflags); + kvm_set_rflags(vcpu, vcpu->arch.emulate_ctxt.eflags); kvm_make_request(KVM_REQ_EVENT, vcpu); return EMULATE_DONE; } @@ -5974,8 +6197,7 @@ int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu) } vcpu->arch.pio_data = page_address(page); - if (!kvm->arch.virtual_tsc_khz) - kvm_arch_set_tsc_khz(kvm, max_tsc_khz); + kvm_init_tsc_catchup(vcpu, max_tsc_khz); r = kvm_mmu_create(vcpu); if (r < 0) diff --git a/arch/x86/kvm/x86.h b/arch/x86/kvm/x86.h index c600da830ce0..e407ed3df817 100644 --- a/arch/x86/kvm/x86.h +++ b/arch/x86/kvm/x86.h @@ -77,7 +77,7 @@ static inline u32 bit(int bitno) void kvm_before_handle_nmi(struct kvm_vcpu *vcpu); void kvm_after_handle_nmi(struct kvm_vcpu *vcpu); -int kvm_inject_realmode_interrupt(struct kvm_vcpu *vcpu, int irq); +int kvm_inject_realmode_interrupt(struct kvm_vcpu *vcpu, int irq, int inc_eip); void kvm_write_tsc(struct kvm_vcpu *vcpu, u64 data); diff --git a/arch/x86/lguest/boot.c b/arch/x86/lguest/boot.c index 1cd608973ce5..e191c096ab90 100644 --- a/arch/x86/lguest/boot.c +++ b/arch/x86/lguest/boot.c @@ -7,7 +7,7 @@ * kernel and insert a module (lg.ko) which allows us to run other Linux * kernels the same way we'd run processes. We call the first kernel the Host, * and the others the Guests. The program which sets up and configures Guests - * (such as the example in Documentation/lguest/lguest.c) is called the + * (such as the example in Documentation/virtual/lguest/lguest.c) is called the * Launcher. * * Secondly, we only run specially modified Guests, not normal kernels: setting @@ -913,8 +913,6 @@ static struct clocksource lguest_clock = { .rating = 200, .read = lguest_clock_read, .mask = CLOCKSOURCE_MASK(64), - .mult = 1 << 22, - .shift = 22, .flags = CLOCK_SOURCE_IS_CONTINUOUS, }; @@ -997,7 +995,7 @@ static void lguest_time_init(void) /* Set up the timer interrupt (0) to go to our simple timer routine */ irq_set_handler(0, lguest_time_irq); - clocksource_register(&lguest_clock); + clocksource_register_hz(&lguest_clock, NSEC_PER_SEC); /* We can't set cpumask in the initializer: damn C limitations! Set it * here and register our timer device. */ diff --git a/arch/x86/lib/clear_page_64.S b/arch/x86/lib/clear_page_64.S index aa4326bfb24a..f2145cfa12a6 100644 --- a/arch/x86/lib/clear_page_64.S +++ b/arch/x86/lib/clear_page_64.S @@ -1,5 +1,6 @@ #include <linux/linkage.h> #include <asm/dwarf2.h> +#include <asm/alternative-asm.h> /* * Zero a page. @@ -14,6 +15,15 @@ ENTRY(clear_page_c) CFI_ENDPROC ENDPROC(clear_page_c) +ENTRY(clear_page_c_e) + CFI_STARTPROC + movl $4096,%ecx + xorl %eax,%eax + rep stosb + ret + CFI_ENDPROC +ENDPROC(clear_page_c_e) + ENTRY(clear_page) CFI_STARTPROC xorl %eax,%eax @@ -38,21 +48,26 @@ ENTRY(clear_page) .Lclear_page_end: ENDPROC(clear_page) - /* Some CPUs run faster using the string instructions. - It is also a lot simpler. Use this when possible */ + /* + * Some CPUs support enhanced REP MOVSB/STOSB instructions. + * It is recommended to use this when possible. + * If enhanced REP MOVSB/STOSB is not available, try to use fast string. + * Otherwise, use original function. + * + */ #include <asm/cpufeature.h> .section .altinstr_replacement,"ax" 1: .byte 0xeb /* jmp <disp8> */ .byte (clear_page_c - clear_page) - (2f - 1b) /* offset */ -2: +2: .byte 0xeb /* jmp <disp8> */ + .byte (clear_page_c_e - clear_page) - (3f - 2b) /* offset */ +3: .previous .section .altinstructions,"a" - .align 8 - .quad clear_page - .quad 1b - .word X86_FEATURE_REP_GOOD - .byte .Lclear_page_end - clear_page - .byte 2b - 1b + altinstruction_entry clear_page,1b,X86_FEATURE_REP_GOOD,\ + .Lclear_page_end-clear_page, 2b-1b + altinstruction_entry clear_page,2b,X86_FEATURE_ERMS, \ + .Lclear_page_end-clear_page,3b-2b .previous diff --git a/arch/x86/lib/copy_user_64.S b/arch/x86/lib/copy_user_64.S index 99e482615195..024840266ba0 100644 --- a/arch/x86/lib/copy_user_64.S +++ b/arch/x86/lib/copy_user_64.S @@ -15,23 +15,30 @@ #include <asm/asm-offsets.h> #include <asm/thread_info.h> #include <asm/cpufeature.h> +#include <asm/alternative-asm.h> - .macro ALTERNATIVE_JUMP feature,orig,alt +/* + * By placing feature2 after feature1 in altinstructions section, we logically + * implement: + * If CPU has feature2, jmp to alt2 is used + * else if CPU has feature1, jmp to alt1 is used + * else jmp to orig is used. + */ + .macro ALTERNATIVE_JUMP feature1,feature2,orig,alt1,alt2 0: .byte 0xe9 /* 32bit jump */ .long \orig-1f /* by default jump to orig */ 1: .section .altinstr_replacement,"ax" 2: .byte 0xe9 /* near jump with 32bit immediate */ - .long \alt-1b /* offset */ /* or alternatively to alt */ + .long \alt1-1b /* offset */ /* or alternatively to alt1 */ +3: .byte 0xe9 /* near jump with 32bit immediate */ + .long \alt2-1b /* offset */ /* or alternatively to alt2 */ .previous + .section .altinstructions,"a" - .align 8 - .quad 0b - .quad 2b - .word \feature /* when feature is set */ - .byte 5 - .byte 5 + altinstruction_entry 0b,2b,\feature1,5,5 + altinstruction_entry 0b,3b,\feature2,5,5 .previous .endm @@ -72,8 +79,10 @@ ENTRY(_copy_to_user) addq %rdx,%rcx jc bad_to_user cmpq TI_addr_limit(%rax),%rcx - jae bad_to_user - ALTERNATIVE_JUMP X86_FEATURE_REP_GOOD,copy_user_generic_unrolled,copy_user_generic_string + ja bad_to_user + ALTERNATIVE_JUMP X86_FEATURE_REP_GOOD,X86_FEATURE_ERMS, \ + copy_user_generic_unrolled,copy_user_generic_string, \ + copy_user_enhanced_fast_string CFI_ENDPROC ENDPROC(_copy_to_user) @@ -85,8 +94,10 @@ ENTRY(_copy_from_user) addq %rdx,%rcx jc bad_from_user cmpq TI_addr_limit(%rax),%rcx - jae bad_from_user - ALTERNATIVE_JUMP X86_FEATURE_REP_GOOD,copy_user_generic_unrolled,copy_user_generic_string + ja bad_from_user + ALTERNATIVE_JUMP X86_FEATURE_REP_GOOD,X86_FEATURE_ERMS, \ + copy_user_generic_unrolled,copy_user_generic_string, \ + copy_user_enhanced_fast_string CFI_ENDPROC ENDPROC(_copy_from_user) @@ -255,3 +266,37 @@ ENTRY(copy_user_generic_string) .previous CFI_ENDPROC ENDPROC(copy_user_generic_string) + +/* + * Some CPUs are adding enhanced REP MOVSB/STOSB instructions. + * It's recommended to use enhanced REP MOVSB/STOSB if it's enabled. + * + * Input: + * rdi destination + * rsi source + * rdx count + * + * Output: + * eax uncopied bytes or 0 if successful. + */ +ENTRY(copy_user_enhanced_fast_string) + CFI_STARTPROC + andl %edx,%edx + jz 2f + movl %edx,%ecx +1: rep + movsb +2: xorl %eax,%eax + ret + + .section .fixup,"ax" +12: movl %ecx,%edx /* ecx is zerorest also */ + jmp copy_user_handle_tail + .previous + + .section __ex_table,"a" + .align 8 + .quad 1b,12b + .previous + CFI_ENDPROC +ENDPROC(copy_user_enhanced_fast_string) diff --git a/arch/x86/lib/memcpy_64.S b/arch/x86/lib/memcpy_64.S index 75ef61e35e38..efbf2a0ecdea 100644 --- a/arch/x86/lib/memcpy_64.S +++ b/arch/x86/lib/memcpy_64.S @@ -4,6 +4,7 @@ #include <asm/cpufeature.h> #include <asm/dwarf2.h> +#include <asm/alternative-asm.h> /* * memcpy - Copy a memory block. @@ -37,6 +38,23 @@ .Lmemcpy_e: .previous +/* + * memcpy_c_e() - enhanced fast string memcpy. This is faster and simpler than + * memcpy_c. Use memcpy_c_e when possible. + * + * This gets patched over the unrolled variant (below) via the + * alternative instructions framework: + */ + .section .altinstr_replacement, "ax", @progbits +.Lmemcpy_c_e: + movq %rdi, %rax + + movl %edx, %ecx + rep movsb + ret +.Lmemcpy_e_e: + .previous + ENTRY(__memcpy) ENTRY(memcpy) CFI_STARTPROC @@ -49,7 +67,7 @@ ENTRY(memcpy) jb .Lhandle_tail /* - * We check whether memory false dependece could occur, + * We check whether memory false dependence could occur, * then jump to corresponding copy mode. */ cmp %dil, %sil @@ -171,21 +189,22 @@ ENDPROC(memcpy) ENDPROC(__memcpy) /* - * Some CPUs run faster using the string copy instructions. - * It is also a lot simpler. Use this when possible: - */ - - .section .altinstructions, "a" - .align 8 - .quad memcpy - .quad .Lmemcpy_c - .word X86_FEATURE_REP_GOOD - - /* + * Some CPUs are adding enhanced REP MOVSB/STOSB feature + * If the feature is supported, memcpy_c_e() is the first choice. + * If enhanced rep movsb copy is not available, use fast string copy + * memcpy_c() when possible. This is faster and code is simpler than + * original memcpy(). + * Otherwise, original memcpy() is used. + * In .altinstructions section, ERMS feature is placed after REG_GOOD + * feature to implement the right patch order. + * * Replace only beginning, memcpy is used to apply alternatives, * so it is silly to overwrite itself with nops - reboot is the * only outcome... */ - .byte .Lmemcpy_e - .Lmemcpy_c - .byte .Lmemcpy_e - .Lmemcpy_c + .section .altinstructions, "a" + altinstruction_entry memcpy,.Lmemcpy_c,X86_FEATURE_REP_GOOD,\ + .Lmemcpy_e-.Lmemcpy_c,.Lmemcpy_e-.Lmemcpy_c + altinstruction_entry memcpy,.Lmemcpy_c_e,X86_FEATURE_ERMS, \ + .Lmemcpy_e_e-.Lmemcpy_c_e,.Lmemcpy_e_e-.Lmemcpy_c_e .previous diff --git a/arch/x86/lib/memmove_64.S b/arch/x86/lib/memmove_64.S index 0ecb8433e5a8..d0ec9c2936d7 100644 --- a/arch/x86/lib/memmove_64.S +++ b/arch/x86/lib/memmove_64.S @@ -8,6 +8,7 @@ #define _STRING_C #include <linux/linkage.h> #include <asm/dwarf2.h> +#include <asm/cpufeature.h> #undef memmove @@ -24,6 +25,7 @@ */ ENTRY(memmove) CFI_STARTPROC + /* Handle more 32bytes in loop */ mov %rdi, %rax cmp $0x20, %rdx @@ -31,8 +33,13 @@ ENTRY(memmove) /* Decide forward/backward copy mode */ cmp %rdi, %rsi - jb 2f + jge .Lmemmove_begin_forward + mov %rsi, %r8 + add %rdx, %r8 + cmp %rdi, %r8 + jg 2f +.Lmemmove_begin_forward: /* * movsq instruction have many startup latency * so we handle small size by general register. @@ -78,6 +85,8 @@ ENTRY(memmove) rep movsq movq %r11, (%r10) jmp 13f +.Lmemmove_end_forward: + /* * Handle data backward by movsq. */ @@ -194,4 +203,22 @@ ENTRY(memmove) 13: retq CFI_ENDPROC + + .section .altinstr_replacement,"ax" +.Lmemmove_begin_forward_efs: + /* Forward moving data. */ + movq %rdx, %rcx + rep movsb + retq +.Lmemmove_end_forward_efs: + .previous + + .section .altinstructions,"a" + .align 8 + .quad .Lmemmove_begin_forward + .quad .Lmemmove_begin_forward_efs + .word X86_FEATURE_ERMS + .byte .Lmemmove_end_forward-.Lmemmove_begin_forward + .byte .Lmemmove_end_forward_efs-.Lmemmove_begin_forward_efs + .previous ENDPROC(memmove) diff --git a/arch/x86/lib/memset_64.S b/arch/x86/lib/memset_64.S index 09d344269652..79bd454b78a3 100644 --- a/arch/x86/lib/memset_64.S +++ b/arch/x86/lib/memset_64.S @@ -2,9 +2,13 @@ #include <linux/linkage.h> #include <asm/dwarf2.h> +#include <asm/cpufeature.h> +#include <asm/alternative-asm.h> /* - * ISO C memset - set a memory block to a byte value. + * ISO C memset - set a memory block to a byte value. This function uses fast + * string to get better performance than the original function. The code is + * simpler and shorter than the orignal function as well. * * rdi destination * rsi value (char) @@ -31,6 +35,28 @@ .Lmemset_e: .previous +/* + * ISO C memset - set a memory block to a byte value. This function uses + * enhanced rep stosb to override the fast string function. + * The code is simpler and shorter than the fast string function as well. + * + * rdi destination + * rsi value (char) + * rdx count (bytes) + * + * rax original destination + */ + .section .altinstr_replacement, "ax", @progbits +.Lmemset_c_e: + movq %rdi,%r9 + movb %sil,%al + movl %edx,%ecx + rep stosb + movq %r9,%rax + ret +.Lmemset_e_e: + .previous + ENTRY(memset) ENTRY(__memset) CFI_STARTPROC @@ -112,16 +138,20 @@ ENTRY(__memset) ENDPROC(memset) ENDPROC(__memset) - /* Some CPUs run faster using the string instructions. - It is also a lot simpler. Use this when possible */ - -#include <asm/cpufeature.h> - + /* Some CPUs support enhanced REP MOVSB/STOSB feature. + * It is recommended to use this when possible. + * + * If enhanced REP MOVSB/STOSB feature is not available, use fast string + * instructions. + * + * Otherwise, use original memset function. + * + * In .altinstructions section, ERMS feature is placed after REG_GOOD + * feature to implement the right patch order. + */ .section .altinstructions,"a" - .align 8 - .quad memset - .quad .Lmemset_c - .word X86_FEATURE_REP_GOOD - .byte .Lfinal - memset - .byte .Lmemset_e - .Lmemset_c + altinstruction_entry memset,.Lmemset_c,X86_FEATURE_REP_GOOD,\ + .Lfinal-memset,.Lmemset_e-.Lmemset_c + altinstruction_entry memset,.Lmemset_c_e,X86_FEATURE_ERMS, \ + .Lfinal-memset,.Lmemset_e_e-.Lmemset_c_e .previous diff --git a/arch/x86/mm/Makefile b/arch/x86/mm/Makefile index 3e608edf9958..3d11327c9ab4 100644 --- a/arch/x86/mm/Makefile +++ b/arch/x86/mm/Makefile @@ -23,8 +23,8 @@ mmiotrace-y := kmmio.o pf_in.o mmio-mod.o obj-$(CONFIG_MMIOTRACE_TEST) += testmmiotrace.o obj-$(CONFIG_NUMA) += numa.o numa_$(BITS).o -obj-$(CONFIG_AMD_NUMA) += amdtopology_64.o -obj-$(CONFIG_ACPI_NUMA) += srat_$(BITS).o +obj-$(CONFIG_AMD_NUMA) += amdtopology.o +obj-$(CONFIG_ACPI_NUMA) += srat.o obj-$(CONFIG_NUMA_EMU) += numa_emulation.o obj-$(CONFIG_HAVE_MEMBLOCK) += memblock.o diff --git a/arch/x86/mm/amdtopology_64.c b/arch/x86/mm/amdtopology.c index 0919c26820d4..5247d01329ca 100644 --- a/arch/x86/mm/amdtopology_64.c +++ b/arch/x86/mm/amdtopology.c @@ -12,6 +12,7 @@ #include <linux/module.h> #include <linux/nodemask.h> #include <linux/memblock.h> +#include <linux/bootmem.h> #include <asm/io.h> #include <linux/pci_ids.h> @@ -69,10 +70,10 @@ static __init void early_get_boot_cpu_id(void) int __init amd_numa_init(void) { - unsigned long start = PFN_PHYS(0); - unsigned long end = PFN_PHYS(max_pfn); + u64 start = PFN_PHYS(0); + u64 end = PFN_PHYS(max_pfn); unsigned numnodes; - unsigned long prevbase; + u64 prevbase; int i, j, nb; u32 nodeid, reg; unsigned int bits, cores, apicid_base; @@ -95,7 +96,7 @@ int __init amd_numa_init(void) prevbase = 0; for (i = 0; i < 8; i++) { - unsigned long base, limit; + u64 base, limit; base = read_pci_config(0, nb, 1, 0x40 + i*8); limit = read_pci_config(0, nb, 1, 0x44 + i*8); @@ -107,18 +108,18 @@ int __init amd_numa_init(void) continue; } if (nodeid >= numnodes) { - pr_info("Ignoring excess node %d (%lx:%lx)\n", nodeid, + pr_info("Ignoring excess node %d (%Lx:%Lx)\n", nodeid, base, limit); continue; } if (!limit) { - pr_info("Skipping node entry %d (base %lx)\n", + pr_info("Skipping node entry %d (base %Lx)\n", i, base); continue; } if ((base >> 8) & 3 || (limit >> 8) & 3) { - pr_err("Node %d using interleaving mode %lx/%lx\n", + pr_err("Node %d using interleaving mode %Lx/%Lx\n", nodeid, (base >> 8) & 3, (limit >> 8) & 3); return -EINVAL; } @@ -150,19 +151,19 @@ int __init amd_numa_init(void) continue; } if (limit < base) { - pr_err("Node %d bogus settings %lx-%lx.\n", + pr_err("Node %d bogus settings %Lx-%Lx.\n", nodeid, base, limit); continue; } /* Could sort here, but pun for now. Should not happen anyroads. */ if (prevbase > base) { - pr_err("Node map not sorted %lx,%lx\n", + pr_err("Node map not sorted %Lx,%Lx\n", prevbase, base); return -EINVAL; } - pr_info("Node %d MemBase %016lx Limit %016lx\n", + pr_info("Node %d MemBase %016Lx Limit %016Lx\n", nodeid, base, limit); prevbase = base; diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c index 20e3f8702d1e..f7a2a054a3c0 100644 --- a/arch/x86/mm/fault.c +++ b/arch/x86/mm/fault.c @@ -12,6 +12,7 @@ #include <linux/mmiotrace.h> /* kmmio_handler, ... */ #include <linux/perf_event.h> /* perf_sw_event */ #include <linux/hugetlb.h> /* hstate_index_to_shift */ +#include <linux/prefetch.h> /* prefetchw */ #include <asm/traps.h> /* dotraplinkage, ... */ #include <asm/pgalloc.h> /* pgd_*(), ... */ @@ -964,7 +965,7 @@ do_page_fault(struct pt_regs *regs, unsigned long error_code) struct mm_struct *mm; int fault; int write = error_code & PF_WRITE; - unsigned int flags = FAULT_FLAG_ALLOW_RETRY | + unsigned int flags = FAULT_FLAG_ALLOW_RETRY | FAULT_FLAG_KILLABLE | (write ? FAULT_FLAG_WRITE : 0); tsk = current; @@ -1138,6 +1139,16 @@ good_area: } /* + * Pagefault was interrupted by SIGKILL. We have no reason to + * continue pagefault. + */ + if ((fault & VM_FAULT_RETRY) && fatal_signal_pending(current)) { + if (!(error_code & PF_USER)) + no_context(regs, error_code, address); + return; + } + + /* * Major/minor page fault accounting is only done on the * initial attempt. If we go through a retry, it is extremely * likely that the page will be found in page cache at that point. diff --git a/arch/x86/mm/hugetlbpage.c b/arch/x86/mm/hugetlbpage.c index d4203988504a..f581a18c0d4d 100644 --- a/arch/x86/mm/hugetlbpage.c +++ b/arch/x86/mm/hugetlbpage.c @@ -72,7 +72,7 @@ static void huge_pmd_share(struct mm_struct *mm, unsigned long addr, pud_t *pud) if (!vma_shareable(vma, addr)) return; - spin_lock(&mapping->i_mmap_lock); + mutex_lock(&mapping->i_mmap_mutex); vma_prio_tree_foreach(svma, &iter, &mapping->i_mmap, idx, idx) { if (svma == vma) continue; @@ -97,7 +97,7 @@ static void huge_pmd_share(struct mm_struct *mm, unsigned long addr, pud_t *pud) put_page(virt_to_page(spte)); spin_unlock(&mm->page_table_lock); out: - spin_unlock(&mapping->i_mmap_lock); + mutex_unlock(&mapping->i_mmap_mutex); } /* diff --git a/arch/x86/mm/init.c b/arch/x86/mm/init.c index 286d289b039b..30326443ab81 100644 --- a/arch/x86/mm/init.c +++ b/arch/x86/mm/init.c @@ -16,8 +16,6 @@ #include <asm/tlb.h> #include <asm/proto.h> -DEFINE_PER_CPU(struct mmu_gather, mmu_gathers); - unsigned long __initdata pgt_buf_start; unsigned long __meminitdata pgt_buf_end; unsigned long __meminitdata pgt_buf_top; @@ -81,6 +79,11 @@ static void __init find_early_table_space(unsigned long end, int use_pse, end, pgt_buf_start << PAGE_SHIFT, pgt_buf_top << PAGE_SHIFT); } +void __init native_pagetable_reserve(u64 start, u64 end) +{ + memblock_x86_reserve_range(start, end, "PGTABLE"); +} + struct map_range { unsigned long start; unsigned long end; @@ -272,9 +275,24 @@ unsigned long __init_refok init_memory_mapping(unsigned long start, __flush_tlb_all(); + /* + * Reserve the kernel pagetable pages we used (pgt_buf_start - + * pgt_buf_end) and free the other ones (pgt_buf_end - pgt_buf_top) + * so that they can be reused for other purposes. + * + * On native it just means calling memblock_x86_reserve_range, on Xen it + * also means marking RW the pagetable pages that we allocated before + * but that haven't been used. + * + * In fact on xen we mark RO the whole range pgt_buf_start - + * pgt_buf_top, because we have to make sure that when + * init_memory_mapping reaches the pagetable pages area, it maps + * RO all the pagetable pages, including the ones that are beyond + * pgt_buf_end at that time. + */ if (!after_bootmem && pgt_buf_end > pgt_buf_start) - memblock_x86_reserve_range(pgt_buf_start << PAGE_SHIFT, - pgt_buf_end << PAGE_SHIFT, "PGTABLE"); + x86_init.mapping.pagetable_reserve(PFN_PHYS(pgt_buf_start), + PFN_PHYS(pgt_buf_end)); if (!after_bootmem) early_memtest(start, end); diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c index 80088f994193..29f7c6d98179 100644 --- a/arch/x86/mm/init_32.c +++ b/arch/x86/mm/init_32.c @@ -678,8 +678,10 @@ static void __init zone_sizes_init(void) { unsigned long max_zone_pfns[MAX_NR_ZONES]; memset(max_zone_pfns, 0, sizeof(max_zone_pfns)); +#ifdef CONFIG_ZONE_DMA max_zone_pfns[ZONE_DMA] = virt_to_phys((char *)MAX_DMA_ADDRESS) >> PAGE_SHIFT; +#endif max_zone_pfns[ZONE_NORMAL] = max_low_pfn; #ifdef CONFIG_HIGHMEM max_zone_pfns[ZONE_HIGHMEM] = highend_pfn; @@ -716,6 +718,7 @@ void __init paging_init(void) * NOTE: at this point the bootmem allocator is fully available. */ olpc_dt_build_devicetree(); + sparse_memory_present_with_active_regions(MAX_NUMNODES); sparse_init(); zone_sizes_init(); } diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c index 794233587287..d865c4aeec55 100644 --- a/arch/x86/mm/init_64.c +++ b/arch/x86/mm/init_64.c @@ -616,7 +616,9 @@ void __init paging_init(void) unsigned long max_zone_pfns[MAX_NR_ZONES]; memset(max_zone_pfns, 0, sizeof(max_zone_pfns)); +#ifdef CONFIG_ZONE_DMA max_zone_pfns[ZONE_DMA] = MAX_DMA_PFN; +#endif max_zone_pfns[ZONE_DMA32] = MAX_DMA32_PFN; max_zone_pfns[ZONE_NORMAL] = max_pfn; @@ -679,14 +681,6 @@ int arch_add_memory(int nid, u64 start, u64 size) } EXPORT_SYMBOL_GPL(arch_add_memory); -#if !defined(CONFIG_ACPI_NUMA) && defined(CONFIG_NUMA) -int memory_add_physaddr_to_nid(u64 start) -{ - return 0; -} -EXPORT_SYMBOL_GPL(memory_add_physaddr_to_nid); -#endif - #endif /* CONFIG_MEMORY_HOTPLUG */ static struct kcore_list kcore_vsyscall; diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c index 0369843511dc..be1ef574ce9a 100644 --- a/arch/x86/mm/ioremap.c +++ b/arch/x86/mm/ioremap.c @@ -91,13 +91,6 @@ static void __iomem *__ioremap_caller(resource_size_t phys_addr, return (__force void __iomem *)phys_to_virt(phys_addr); /* - * Check if the request spans more than any BAR in the iomem resource - * tree. - */ - WARN_ONCE(iomem_map_sanity_check(phys_addr, size), - KERN_INFO "Info: mapping multiple BARs. Your kernel is fine."); - - /* * Don't allow anybody to remap normal RAM that we're using.. */ last_pfn = last_addr >> PAGE_SHIFT; @@ -170,6 +163,13 @@ static void __iomem *__ioremap_caller(resource_size_t phys_addr, ret_addr = (void __iomem *) (vaddr + offset); mmiotrace_ioremap(unaligned_phys_addr, unaligned_size, ret_addr); + /* + * Check if the request spans more than any BAR in the iomem resource + * tree. + */ + WARN_ONCE(iomem_map_sanity_check(unaligned_phys_addr, unaligned_size), + KERN_INFO "Info: mapping multiple BARs. Your kernel is fine."); + return ret_addr; err_free_area: free_vm_area(area); diff --git a/arch/x86/mm/numa.c b/arch/x86/mm/numa.c index 745258dfc4dc..f5510d889a22 100644 --- a/arch/x86/mm/numa.c +++ b/arch/x86/mm/numa.c @@ -1,11 +1,39 @@ /* Common code for 32 and 64-bit NUMA */ -#include <linux/topology.h> -#include <linux/module.h> +#include <linux/kernel.h> +#include <linux/mm.h> +#include <linux/string.h> +#include <linux/init.h> #include <linux/bootmem.h> -#include <asm/numa.h> +#include <linux/memblock.h> +#include <linux/mmzone.h> +#include <linux/ctype.h> +#include <linux/module.h> +#include <linux/nodemask.h> +#include <linux/sched.h> +#include <linux/topology.h> + +#include <asm/e820.h> +#include <asm/proto.h> +#include <asm/dma.h> #include <asm/acpi.h> +#include <asm/amd_nb.h> + +#include "numa_internal.h" int __initdata numa_off; +nodemask_t numa_nodes_parsed __initdata; + +struct pglist_data *node_data[MAX_NUMNODES] __read_mostly; +EXPORT_SYMBOL(node_data); + +static struct numa_meminfo numa_meminfo +#ifndef CONFIG_MEMORY_HOTPLUG +__initdata +#endif +; + +static int numa_distance_cnt; +static u8 *numa_distance; static __init int numa_setup(char *opt) { @@ -32,6 +60,15 @@ s16 __apicid_to_node[MAX_LOCAL_APIC] __cpuinitdata = { [0 ... MAX_LOCAL_APIC-1] = NUMA_NO_NODE }; +int __cpuinit numa_cpu_node(int cpu) +{ + int apicid = early_per_cpu(x86_cpu_to_apicid, cpu); + + if (apicid != BAD_APICID) + return __apicid_to_node[apicid]; + return NUMA_NO_NODE; +} + cpumask_var_t node_to_cpumask_map[MAX_NUMNODES]; EXPORT_SYMBOL(node_to_cpumask_map); @@ -95,6 +132,407 @@ void __init setup_node_to_cpumask_map(void) pr_debug("Node to cpumask map for %d nodes\n", nr_node_ids); } +static int __init numa_add_memblk_to(int nid, u64 start, u64 end, + struct numa_meminfo *mi) +{ + /* ignore zero length blks */ + if (start == end) + return 0; + + /* whine about and ignore invalid blks */ + if (start > end || nid < 0 || nid >= MAX_NUMNODES) { + pr_warning("NUMA: Warning: invalid memblk node %d (%Lx-%Lx)\n", + nid, start, end); + return 0; + } + + if (mi->nr_blks >= NR_NODE_MEMBLKS) { + pr_err("NUMA: too many memblk ranges\n"); + return -EINVAL; + } + + mi->blk[mi->nr_blks].start = start; + mi->blk[mi->nr_blks].end = end; + mi->blk[mi->nr_blks].nid = nid; + mi->nr_blks++; + return 0; +} + +/** + * numa_remove_memblk_from - Remove one numa_memblk from a numa_meminfo + * @idx: Index of memblk to remove + * @mi: numa_meminfo to remove memblk from + * + * Remove @idx'th numa_memblk from @mi by shifting @mi->blk[] and + * decrementing @mi->nr_blks. + */ +void __init numa_remove_memblk_from(int idx, struct numa_meminfo *mi) +{ + mi->nr_blks--; + memmove(&mi->blk[idx], &mi->blk[idx + 1], + (mi->nr_blks - idx) * sizeof(mi->blk[0])); +} + +/** + * numa_add_memblk - Add one numa_memblk to numa_meminfo + * @nid: NUMA node ID of the new memblk + * @start: Start address of the new memblk + * @end: End address of the new memblk + * + * Add a new memblk to the default numa_meminfo. + * + * RETURNS: + * 0 on success, -errno on failure. + */ +int __init numa_add_memblk(int nid, u64 start, u64 end) +{ + return numa_add_memblk_to(nid, start, end, &numa_meminfo); +} + +/* Initialize NODE_DATA for a node on the local memory */ +static void __init setup_node_data(int nid, u64 start, u64 end) +{ + const u64 nd_low = PFN_PHYS(MAX_DMA_PFN); + const u64 nd_high = PFN_PHYS(max_pfn_mapped); + const size_t nd_size = roundup(sizeof(pg_data_t), PAGE_SIZE); + bool remapped = false; + u64 nd_pa; + void *nd; + int tnid; + + /* + * Don't confuse VM with a node that doesn't have the + * minimum amount of memory: + */ + if (end && (end - start) < NODE_MIN_SIZE) + return; + + /* initialize remap allocator before aligning to ZONE_ALIGN */ + init_alloc_remap(nid, start, end); + + start = roundup(start, ZONE_ALIGN); + + printk(KERN_INFO "Initmem setup node %d %016Lx-%016Lx\n", + nid, start, end); + + /* + * Allocate node data. Try remap allocator first, node-local + * memory and then any node. Never allocate in DMA zone. + */ + nd = alloc_remap(nid, nd_size); + if (nd) { + nd_pa = __pa(nd); + remapped = true; + } else { + nd_pa = memblock_x86_find_in_range_node(nid, nd_low, nd_high, + nd_size, SMP_CACHE_BYTES); + if (nd_pa == MEMBLOCK_ERROR) + nd_pa = memblock_find_in_range(nd_low, nd_high, + nd_size, SMP_CACHE_BYTES); + if (nd_pa == MEMBLOCK_ERROR) { + pr_err("Cannot find %zu bytes in node %d\n", + nd_size, nid); + return; + } + memblock_x86_reserve_range(nd_pa, nd_pa + nd_size, "NODE_DATA"); + nd = __va(nd_pa); + } + + /* report and initialize */ + printk(KERN_INFO " NODE_DATA [%016Lx - %016Lx]%s\n", + nd_pa, nd_pa + nd_size - 1, remapped ? " (remapped)" : ""); + tnid = early_pfn_to_nid(nd_pa >> PAGE_SHIFT); + if (!remapped && tnid != nid) + printk(KERN_INFO " NODE_DATA(%d) on node %d\n", nid, tnid); + + node_data[nid] = nd; + memset(NODE_DATA(nid), 0, sizeof(pg_data_t)); + NODE_DATA(nid)->node_id = nid; + NODE_DATA(nid)->node_start_pfn = start >> PAGE_SHIFT; + NODE_DATA(nid)->node_spanned_pages = (end - start) >> PAGE_SHIFT; + + node_set_online(nid); +} + +/** + * numa_cleanup_meminfo - Cleanup a numa_meminfo + * @mi: numa_meminfo to clean up + * + * Sanitize @mi by merging and removing unncessary memblks. Also check for + * conflicts and clear unused memblks. + * + * RETURNS: + * 0 on success, -errno on failure. + */ +int __init numa_cleanup_meminfo(struct numa_meminfo *mi) +{ + const u64 low = 0; + const u64 high = PFN_PHYS(max_pfn); + int i, j, k; + + /* first, trim all entries */ + for (i = 0; i < mi->nr_blks; i++) { + struct numa_memblk *bi = &mi->blk[i]; + + /* make sure all blocks are inside the limits */ + bi->start = max(bi->start, low); + bi->end = min(bi->end, high); + + /* and there's no empty block */ + if (bi->start >= bi->end) + numa_remove_memblk_from(i--, mi); + } + + /* merge neighboring / overlapping entries */ + for (i = 0; i < mi->nr_blks; i++) { + struct numa_memblk *bi = &mi->blk[i]; + + for (j = i + 1; j < mi->nr_blks; j++) { + struct numa_memblk *bj = &mi->blk[j]; + u64 start, end; + + /* + * See whether there are overlapping blocks. Whine + * about but allow overlaps of the same nid. They + * will be merged below. + */ + if (bi->end > bj->start && bi->start < bj->end) { + if (bi->nid != bj->nid) { + pr_err("NUMA: node %d (%Lx-%Lx) overlaps with node %d (%Lx-%Lx)\n", + bi->nid, bi->start, bi->end, + bj->nid, bj->start, bj->end); + return -EINVAL; + } + pr_warning("NUMA: Warning: node %d (%Lx-%Lx) overlaps with itself (%Lx-%Lx)\n", + bi->nid, bi->start, bi->end, + bj->start, bj->end); + } + + /* + * Join together blocks on the same node, holes + * between which don't overlap with memory on other + * nodes. + */ + if (bi->nid != bj->nid) + continue; + start = min(bi->start, bj->start); + end = max(bi->end, bj->end); + for (k = 0; k < mi->nr_blks; k++) { + struct numa_memblk *bk = &mi->blk[k]; + + if (bi->nid == bk->nid) + continue; + if (start < bk->end && end > bk->start) + break; + } + if (k < mi->nr_blks) + continue; + printk(KERN_INFO "NUMA: Node %d [%Lx,%Lx) + [%Lx,%Lx) -> [%Lx,%Lx)\n", + bi->nid, bi->start, bi->end, bj->start, bj->end, + start, end); + bi->start = start; + bi->end = end; + numa_remove_memblk_from(j--, mi); + } + } + + /* clear unused ones */ + for (i = mi->nr_blks; i < ARRAY_SIZE(mi->blk); i++) { + mi->blk[i].start = mi->blk[i].end = 0; + mi->blk[i].nid = NUMA_NO_NODE; + } + + return 0; +} + +/* + * Set nodes, which have memory in @mi, in *@nodemask. + */ +static void __init numa_nodemask_from_meminfo(nodemask_t *nodemask, + const struct numa_meminfo *mi) +{ + int i; + + for (i = 0; i < ARRAY_SIZE(mi->blk); i++) + if (mi->blk[i].start != mi->blk[i].end && + mi->blk[i].nid != NUMA_NO_NODE) + node_set(mi->blk[i].nid, *nodemask); +} + +/** + * numa_reset_distance - Reset NUMA distance table + * + * The current table is freed. The next numa_set_distance() call will + * create a new one. + */ +void __init numa_reset_distance(void) +{ + size_t size = numa_distance_cnt * numa_distance_cnt * sizeof(numa_distance[0]); + + /* numa_distance could be 1LU marking allocation failure, test cnt */ + if (numa_distance_cnt) + memblock_x86_free_range(__pa(numa_distance), + __pa(numa_distance) + size); + numa_distance_cnt = 0; + numa_distance = NULL; /* enable table creation */ +} + +static int __init numa_alloc_distance(void) +{ + nodemask_t nodes_parsed; + size_t size; + int i, j, cnt = 0; + u64 phys; + + /* size the new table and allocate it */ + nodes_parsed = numa_nodes_parsed; + numa_nodemask_from_meminfo(&nodes_parsed, &numa_meminfo); + + for_each_node_mask(i, nodes_parsed) + cnt = i; + cnt++; + size = cnt * cnt * sizeof(numa_distance[0]); + + phys = memblock_find_in_range(0, PFN_PHYS(max_pfn_mapped), + size, PAGE_SIZE); + if (phys == MEMBLOCK_ERROR) { + pr_warning("NUMA: Warning: can't allocate distance table!\n"); + /* don't retry until explicitly reset */ + numa_distance = (void *)1LU; + return -ENOMEM; + } + memblock_x86_reserve_range(phys, phys + size, "NUMA DIST"); + + numa_distance = __va(phys); + numa_distance_cnt = cnt; + + /* fill with the default distances */ + for (i = 0; i < cnt; i++) + for (j = 0; j < cnt; j++) + numa_distance[i * cnt + j] = i == j ? + LOCAL_DISTANCE : REMOTE_DISTANCE; + printk(KERN_DEBUG "NUMA: Initialized distance table, cnt=%d\n", cnt); + + return 0; +} + +/** + * numa_set_distance - Set NUMA distance from one NUMA to another + * @from: the 'from' node to set distance + * @to: the 'to' node to set distance + * @distance: NUMA distance + * + * Set the distance from node @from to @to to @distance. If distance table + * doesn't exist, one which is large enough to accommodate all the currently + * known nodes will be created. + * + * If such table cannot be allocated, a warning is printed and further + * calls are ignored until the distance table is reset with + * numa_reset_distance(). + * + * If @from or @to is higher than the highest known node at the time of + * table creation or @distance doesn't make sense, the call is ignored. + * This is to allow simplification of specific NUMA config implementations. + */ +void __init numa_set_distance(int from, int to, int distance) +{ + if (!numa_distance && numa_alloc_distance() < 0) + return; + + if (from >= numa_distance_cnt || to >= numa_distance_cnt) { + printk_once(KERN_DEBUG "NUMA: Debug: distance out of bound, from=%d to=%d distance=%d\n", + from, to, distance); + return; + } + + if ((u8)distance != distance || + (from == to && distance != LOCAL_DISTANCE)) { + pr_warn_once("NUMA: Warning: invalid distance parameter, from=%d to=%d distance=%d\n", + from, to, distance); + return; + } + + numa_distance[from * numa_distance_cnt + to] = distance; +} + +int __node_distance(int from, int to) +{ + if (from >= numa_distance_cnt || to >= numa_distance_cnt) + return from == to ? LOCAL_DISTANCE : REMOTE_DISTANCE; + return numa_distance[from * numa_distance_cnt + to]; +} +EXPORT_SYMBOL(__node_distance); + +/* + * Sanity check to catch more bad NUMA configurations (they are amazingly + * common). Make sure the nodes cover all memory. + */ +static bool __init numa_meminfo_cover_memory(const struct numa_meminfo *mi) +{ + u64 numaram, e820ram; + int i; + + numaram = 0; + for (i = 0; i < mi->nr_blks; i++) { + u64 s = mi->blk[i].start >> PAGE_SHIFT; + u64 e = mi->blk[i].end >> PAGE_SHIFT; + numaram += e - s; + numaram -= __absent_pages_in_range(mi->blk[i].nid, s, e); + if ((s64)numaram < 0) + numaram = 0; + } + + e820ram = max_pfn - (memblock_x86_hole_size(0, + PFN_PHYS(max_pfn)) >> PAGE_SHIFT); + /* We seem to lose 3 pages somewhere. Allow 1M of slack. */ + if ((s64)(e820ram - numaram) >= (1 << (20 - PAGE_SHIFT))) { + printk(KERN_ERR "NUMA: nodes only cover %LuMB of your %LuMB e820 RAM. Not used.\n", + (numaram << PAGE_SHIFT) >> 20, + (e820ram << PAGE_SHIFT) >> 20); + return false; + } + return true; +} + +static int __init numa_register_memblks(struct numa_meminfo *mi) +{ + int i, nid; + + /* Account for nodes with cpus and no memory */ + node_possible_map = numa_nodes_parsed; + numa_nodemask_from_meminfo(&node_possible_map, mi); + if (WARN_ON(nodes_empty(node_possible_map))) + return -EINVAL; + + for (i = 0; i < mi->nr_blks; i++) + memblock_x86_register_active_regions(mi->blk[i].nid, + mi->blk[i].start >> PAGE_SHIFT, + mi->blk[i].end >> PAGE_SHIFT); + + /* for out of order entries */ + sort_node_map(); + if (!numa_meminfo_cover_memory(mi)) + return -EINVAL; + + /* Finally register nodes. */ + for_each_node_mask(nid, node_possible_map) { + u64 start = PFN_PHYS(max_pfn); + u64 end = 0; + + for (i = 0; i < mi->nr_blks; i++) { + if (nid != mi->blk[i].nid) + continue; + start = min(mi->blk[i].start, start); + end = max(mi->blk[i].end, end); + } + + if (start < end) + setup_node_data(nid, start, end); + } + + return 0; +} + /* * There are unfortunately some poorly designed mainboards around that * only connect memory to a single CPU. This breaks the 1:1 cpu->node @@ -102,7 +540,7 @@ void __init setup_node_to_cpumask_map(void) * as the number of CPUs is not known yet. We round robin the existing * nodes. */ -void __init numa_init_array(void) +static void __init numa_init_array(void) { int rr, i; @@ -117,6 +555,95 @@ void __init numa_init_array(void) } } +static int __init numa_init(int (*init_func)(void)) +{ + int i; + int ret; + + for (i = 0; i < MAX_LOCAL_APIC; i++) + set_apicid_to_node(i, NUMA_NO_NODE); + + nodes_clear(numa_nodes_parsed); + nodes_clear(node_possible_map); + nodes_clear(node_online_map); + memset(&numa_meminfo, 0, sizeof(numa_meminfo)); + remove_all_active_ranges(); + numa_reset_distance(); + + ret = init_func(); + if (ret < 0) + return ret; + ret = numa_cleanup_meminfo(&numa_meminfo); + if (ret < 0) + return ret; + + numa_emulation(&numa_meminfo, numa_distance_cnt); + + ret = numa_register_memblks(&numa_meminfo); + if (ret < 0) + return ret; + + for (i = 0; i < nr_cpu_ids; i++) { + int nid = early_cpu_to_node(i); + + if (nid == NUMA_NO_NODE) + continue; + if (!node_online(nid)) + numa_clear_node(i); + } + numa_init_array(); + return 0; +} + +/** + * dummy_numa_init - Fallback dummy NUMA init + * + * Used if there's no underlying NUMA architecture, NUMA initialization + * fails, or NUMA is disabled on the command line. + * + * Must online at least one node and add memory blocks that cover all + * allowed memory. This function must not fail. + */ +static int __init dummy_numa_init(void) +{ + printk(KERN_INFO "%s\n", + numa_off ? "NUMA turned off" : "No NUMA configuration found"); + printk(KERN_INFO "Faking a node at %016Lx-%016Lx\n", + 0LLU, PFN_PHYS(max_pfn)); + + node_set(0, numa_nodes_parsed); + numa_add_memblk(0, 0, PFN_PHYS(max_pfn)); + + return 0; +} + +/** + * x86_numa_init - Initialize NUMA + * + * Try each configured NUMA initialization method until one succeeds. The + * last fallback is dummy single node config encomapssing whole memory and + * never fails. + */ +void __init x86_numa_init(void) +{ + if (!numa_off) { +#ifdef CONFIG_X86_NUMAQ + if (!numa_init(numaq_numa_init)) + return; +#endif +#ifdef CONFIG_ACPI_NUMA + if (!numa_init(x86_acpi_numa_init)) + return; +#endif +#ifdef CONFIG_AMD_NUMA + if (!numa_init(amd_numa_init)) + return; +#endif + } + + numa_init(dummy_numa_init); +} + static __init int find_near_online_node(int node) { int n, val; @@ -282,3 +809,18 @@ const struct cpumask *cpumask_of_node(int node) EXPORT_SYMBOL(cpumask_of_node); #endif /* !CONFIG_DEBUG_PER_CPU_MAPS */ + +#ifdef CONFIG_MEMORY_HOTPLUG +int memory_add_physaddr_to_nid(u64 start) +{ + struct numa_meminfo *mi = &numa_meminfo; + int nid = mi->blk[0].nid; + int i; + + for (i = 0; i < mi->nr_blks; i++) + if (mi->blk[i].start <= start && mi->blk[i].end > start) + nid = mi->blk[i].nid; + return nid; +} +EXPORT_SYMBOL_GPL(memory_add_physaddr_to_nid); +#endif diff --git a/arch/x86/mm/numa_32.c b/arch/x86/mm/numa_32.c index bde3906420df..849a975d3fa0 100644 --- a/arch/x86/mm/numa_32.c +++ b/arch/x86/mm/numa_32.c @@ -22,39 +22,11 @@ * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. */ -#include <linux/mm.h> #include <linux/bootmem.h> #include <linux/memblock.h> -#include <linux/mmzone.h> -#include <linux/highmem.h> -#include <linux/initrd.h> -#include <linux/nodemask.h> #include <linux/module.h> -#include <linux/kexec.h> -#include <linux/pfn.h> -#include <linux/swap.h> -#include <linux/acpi.h> - -#include <asm/e820.h> -#include <asm/setup.h> -#include <asm/mmzone.h> -#include <asm/bios_ebda.h> -#include <asm/proto.h> - -struct pglist_data *node_data[MAX_NUMNODES] __read_mostly; -EXPORT_SYMBOL(node_data); - -/* - * numa interface - we expect the numa architecture specific code to have - * populated the following initialisation. - * - * 1) node_online_map - the map of all nodes configured (online) in the system - * 2) node_start_pfn - the starting page frame number for a node - * 3) node_end_pfn - the ending page fram number for a node - */ -unsigned long node_start_pfn[MAX_NUMNODES] __read_mostly; -unsigned long node_end_pfn[MAX_NUMNODES] __read_mostly; +#include "numa_internal.h" #ifdef CONFIG_DISCONTIGMEM /* @@ -99,108 +71,46 @@ unsigned long node_memmap_size_bytes(int nid, unsigned long start_pfn, } #endif -extern unsigned long find_max_low_pfn(void); extern unsigned long highend_pfn, highstart_pfn; #define LARGE_PAGE_BYTES (PTRS_PER_PTE * PAGE_SIZE) -unsigned long node_remap_size[MAX_NUMNODES]; static void *node_remap_start_vaddr[MAX_NUMNODES]; void set_pmd_pfn(unsigned long vaddr, unsigned long pfn, pgprot_t flags); -static unsigned long kva_start_pfn; -static unsigned long kva_pages; - -int __cpuinit numa_cpu_node(int cpu) -{ - return apic->x86_32_numa_cpu_node(cpu); -} - -/* - * FLAT - support for basic PC memory model with discontig enabled, essentially - * a single node with all available processors in it with a flat - * memory map. - */ -int __init get_memcfg_numa_flat(void) -{ - printk(KERN_DEBUG "NUMA - single node, flat memory mode\n"); - - node_start_pfn[0] = 0; - node_end_pfn[0] = max_pfn; - memblock_x86_register_active_regions(0, 0, max_pfn); - memory_present(0, 0, max_pfn); - node_remap_size[0] = node_memmap_size_bytes(0, 0, max_pfn); - - /* Indicate there is one node available. */ - nodes_clear(node_online_map); - node_set_online(0); - return 1; -} - -/* - * Find the highest page frame number we have available for the node - */ -static void __init propagate_e820_map_node(int nid) -{ - if (node_end_pfn[nid] > max_pfn) - node_end_pfn[nid] = max_pfn; - /* - * if a user has given mem=XXXX, then we need to make sure - * that the node _starts_ before that, too, not just ends - */ - if (node_start_pfn[nid] > max_pfn) - node_start_pfn[nid] = max_pfn; - BUG_ON(node_start_pfn[nid] > node_end_pfn[nid]); -} - -/* - * Allocate memory for the pg_data_t for this node via a crude pre-bootmem - * method. For node zero take this from the bottom of memory, for - * subsequent nodes place them at node_remap_start_vaddr which contains - * node local data in physically node local memory. See setup_memory() - * for details. - */ -static void __init allocate_pgdat(int nid) -{ - char buf[16]; - - if (node_has_online_mem(nid) && node_remap_start_vaddr[nid]) - NODE_DATA(nid) = (pg_data_t *)node_remap_start_vaddr[nid]; - else { - unsigned long pgdat_phys; - pgdat_phys = memblock_find_in_range(min_low_pfn<<PAGE_SHIFT, - max_pfn_mapped<<PAGE_SHIFT, - sizeof(pg_data_t), - PAGE_SIZE); - NODE_DATA(nid) = (pg_data_t *)(pfn_to_kaddr(pgdat_phys>>PAGE_SHIFT)); - memset(buf, 0, sizeof(buf)); - sprintf(buf, "NODE_DATA %d", nid); - memblock_x86_reserve_range(pgdat_phys, pgdat_phys + sizeof(pg_data_t), buf); - } - printk(KERN_DEBUG "allocate_pgdat: node %d NODE_DATA %08lx\n", - nid, (unsigned long)NODE_DATA(nid)); -} - /* - * In the DISCONTIGMEM and SPARSEMEM memory model, a portion of the kernel - * virtual address space (KVA) is reserved and portions of nodes are mapped - * using it. This is to allow node-local memory to be allocated for - * structures that would normally require ZONE_NORMAL. The memory is - * allocated with alloc_remap() and callers should be prepared to allocate - * from the bootmem allocator instead. + * Remap memory allocator */ static unsigned long node_remap_start_pfn[MAX_NUMNODES]; static void *node_remap_end_vaddr[MAX_NUMNODES]; static void *node_remap_alloc_vaddr[MAX_NUMNODES]; -static unsigned long node_remap_offset[MAX_NUMNODES]; +/** + * alloc_remap - Allocate remapped memory + * @nid: NUMA node to allocate memory from + * @size: The size of allocation + * + * Allocate @size bytes from the remap area of NUMA node @nid. The + * size of the remap area is predetermined by init_alloc_remap() and + * only the callers considered there should call this function. For + * more info, please read the comment on top of init_alloc_remap(). + * + * The caller must be ready to handle allocation failure from this + * function and fall back to regular memory allocator in such cases. + * + * CONTEXT: + * Single CPU early boot context. + * + * RETURNS: + * Pointer to the allocated memory on success, %NULL on failure. + */ void *alloc_remap(int nid, unsigned long size) { void *allocation = node_remap_alloc_vaddr[nid]; size = ALIGN(size, L1_CACHE_BYTES); - if (!allocation || (allocation + size) >= node_remap_end_vaddr[nid]) + if (!allocation || (allocation + size) > node_remap_end_vaddr[nid]) return NULL; node_remap_alloc_vaddr[nid] += size; @@ -209,26 +119,6 @@ void *alloc_remap(int nid, unsigned long size) return allocation; } -static void __init remap_numa_kva(void) -{ - void *vaddr; - unsigned long pfn; - int node; - - for_each_online_node(node) { - printk(KERN_DEBUG "remap_numa_kva: node %d\n", node); - for (pfn=0; pfn < node_remap_size[node]; pfn += PTRS_PER_PTE) { - vaddr = node_remap_start_vaddr[node]+(pfn<<PAGE_SHIFT); - printk(KERN_DEBUG "remap_numa_kva: %08lx to pfn %08lx\n", - (unsigned long)vaddr, - node_remap_start_pfn[node] + pfn); - set_pmd_pfn((ulong) vaddr, - node_remap_start_pfn[node] + pfn, - PAGE_KERNEL_LARGE); - } - } -} - #ifdef CONFIG_HIBERNATION /** * resume_map_numa_kva - add KVA mapping to the temporary page tables created @@ -240,15 +130,16 @@ void resume_map_numa_kva(pgd_t *pgd_base) int node; for_each_online_node(node) { - unsigned long start_va, start_pfn, size, pfn; + unsigned long start_va, start_pfn, nr_pages, pfn; start_va = (unsigned long)node_remap_start_vaddr[node]; start_pfn = node_remap_start_pfn[node]; - size = node_remap_size[node]; + nr_pages = (node_remap_end_vaddr[node] - + node_remap_start_vaddr[node]) >> PAGE_SHIFT; printk(KERN_DEBUG "%s: node %d\n", __func__, node); - for (pfn = 0; pfn < size; pfn += PTRS_PER_PTE) { + for (pfn = 0; pfn < nr_pages; pfn += PTRS_PER_PTE) { unsigned long vaddr = start_va + (pfn << PAGE_SHIFT); pgd_t *pgd = pgd_base + pgd_index(vaddr); pud_t *pud = pud_offset(pgd, vaddr); @@ -264,132 +155,89 @@ void resume_map_numa_kva(pgd_t *pgd_base) } #endif -static __init unsigned long calculate_numa_remap_pages(void) +/** + * init_alloc_remap - Initialize remap allocator for a NUMA node + * @nid: NUMA node to initizlie remap allocator for + * + * NUMA nodes may end up without any lowmem. As allocating pgdat and + * memmap on a different node with lowmem is inefficient, a special + * remap allocator is implemented which can be used by alloc_remap(). + * + * For each node, the amount of memory which will be necessary for + * pgdat and memmap is calculated and two memory areas of the size are + * allocated - one in the node and the other in lowmem; then, the area + * in the node is remapped to the lowmem area. + * + * As pgdat and memmap must be allocated in lowmem anyway, this + * doesn't waste lowmem address space; however, the actual lowmem + * which gets remapped over is wasted. The amount shouldn't be + * problematic on machines this feature will be used. + * + * Initialization failure isn't fatal. alloc_remap() is used + * opportunistically and the callers will fall back to other memory + * allocation mechanisms on failure. + */ +void __init init_alloc_remap(int nid, u64 start, u64 end) { - int nid; - unsigned long size, reserve_pages = 0; - - for_each_online_node(nid) { - u64 node_kva_target; - u64 node_kva_final; - - /* - * The acpi/srat node info can show hot-add memroy zones - * where memory could be added but not currently present. - */ - printk(KERN_DEBUG "node %d pfn: [%lx - %lx]\n", - nid, node_start_pfn[nid], node_end_pfn[nid]); - if (node_start_pfn[nid] > max_pfn) - continue; - if (!node_end_pfn[nid]) - continue; - if (node_end_pfn[nid] > max_pfn) - node_end_pfn[nid] = max_pfn; - - /* ensure the remap includes space for the pgdat. */ - size = node_remap_size[nid] + sizeof(pg_data_t); - - /* convert size to large (pmd size) pages, rounding up */ - size = (size + LARGE_PAGE_BYTES - 1) / LARGE_PAGE_BYTES; - /* now the roundup is correct, convert to PAGE_SIZE pages */ - size = size * PTRS_PER_PTE; - - node_kva_target = round_down(node_end_pfn[nid] - size, - PTRS_PER_PTE); - node_kva_target <<= PAGE_SHIFT; - do { - node_kva_final = memblock_find_in_range(node_kva_target, - ((u64)node_end_pfn[nid])<<PAGE_SHIFT, - ((u64)size)<<PAGE_SHIFT, - LARGE_PAGE_BYTES); - node_kva_target -= LARGE_PAGE_BYTES; - } while (node_kva_final == MEMBLOCK_ERROR && - (node_kva_target>>PAGE_SHIFT) > (node_start_pfn[nid])); - - if (node_kva_final == MEMBLOCK_ERROR) - panic("Can not get kva ram\n"); - - node_remap_size[nid] = size; - node_remap_offset[nid] = reserve_pages; - reserve_pages += size; - printk(KERN_DEBUG "Reserving %ld pages of KVA for lmem_map of" - " node %d at %llx\n", - size, nid, node_kva_final>>PAGE_SHIFT); - - /* - * prevent kva address below max_low_pfn want it on system - * with less memory later. - * layout will be: KVA address , KVA RAM - * - * we are supposed to only record the one less then max_low_pfn - * but we could have some hole in high memory, and it will only - * check page_is_ram(pfn) && !page_is_reserved_early(pfn) to decide - * to use it as free. - * So memblock_x86_reserve_range here, hope we don't run out of that array - */ - memblock_x86_reserve_range(node_kva_final, - node_kva_final+(((u64)size)<<PAGE_SHIFT), - "KVA RAM"); - - node_remap_start_pfn[nid] = node_kva_final>>PAGE_SHIFT; - } - printk(KERN_INFO "Reserving total of %lx pages for numa KVA remap\n", - reserve_pages); - return reserve_pages; -} + unsigned long start_pfn = start >> PAGE_SHIFT; + unsigned long end_pfn = end >> PAGE_SHIFT; + unsigned long size, pfn; + u64 node_pa, remap_pa; + void *remap_va; -static void init_remap_allocator(int nid) -{ - node_remap_start_vaddr[nid] = pfn_to_kaddr( - kva_start_pfn + node_remap_offset[nid]); - node_remap_end_vaddr[nid] = node_remap_start_vaddr[nid] + - (node_remap_size[nid] * PAGE_SIZE); - node_remap_alloc_vaddr[nid] = node_remap_start_vaddr[nid] + - ALIGN(sizeof(pg_data_t), PAGE_SIZE); - - printk(KERN_DEBUG "node %d will remap to vaddr %08lx - %08lx\n", nid, - (ulong) node_remap_start_vaddr[nid], - (ulong) node_remap_end_vaddr[nid]); + /* + * The acpi/srat node info can show hot-add memroy zones where + * memory could be added but not currently present. + */ + printk(KERN_DEBUG "node %d pfn: [%lx - %lx]\n", + nid, start_pfn, end_pfn); + + /* calculate the necessary space aligned to large page size */ + size = node_memmap_size_bytes(nid, start_pfn, end_pfn); + size += ALIGN(sizeof(pg_data_t), PAGE_SIZE); + size = ALIGN(size, LARGE_PAGE_BYTES); + + /* allocate node memory and the lowmem remap area */ + node_pa = memblock_find_in_range(start, end, size, LARGE_PAGE_BYTES); + if (node_pa == MEMBLOCK_ERROR) { + pr_warning("remap_alloc: failed to allocate %lu bytes for node %d\n", + size, nid); + return; + } + memblock_x86_reserve_range(node_pa, node_pa + size, "KVA RAM"); + + remap_pa = memblock_find_in_range(min_low_pfn << PAGE_SHIFT, + max_low_pfn << PAGE_SHIFT, + size, LARGE_PAGE_BYTES); + if (remap_pa == MEMBLOCK_ERROR) { + pr_warning("remap_alloc: failed to allocate %lu bytes remap area for node %d\n", + size, nid); + memblock_x86_free_range(node_pa, node_pa + size); + return; + } + memblock_x86_reserve_range(remap_pa, remap_pa + size, "KVA PG"); + remap_va = phys_to_virt(remap_pa); + + /* perform actual remap */ + for (pfn = 0; pfn < size >> PAGE_SHIFT; pfn += PTRS_PER_PTE) + set_pmd_pfn((unsigned long)remap_va + (pfn << PAGE_SHIFT), + (node_pa >> PAGE_SHIFT) + pfn, + PAGE_KERNEL_LARGE); + + /* initialize remap allocator parameters */ + node_remap_start_pfn[nid] = node_pa >> PAGE_SHIFT; + node_remap_start_vaddr[nid] = remap_va; + node_remap_end_vaddr[nid] = remap_va + size; + node_remap_alloc_vaddr[nid] = remap_va; + + printk(KERN_DEBUG "remap_alloc: node %d [%08llx-%08llx) -> [%p-%p)\n", + nid, node_pa, node_pa + size, remap_va, remap_va + size); } void __init initmem_init(void) { - int nid; - long kva_target_pfn; - - /* - * When mapping a NUMA machine we allocate the node_mem_map arrays - * from node local memory. They are then mapped directly into KVA - * between zone normal and vmalloc space. Calculate the size of - * this space and use it to adjust the boundary between ZONE_NORMAL - * and ZONE_HIGHMEM. - */ - - get_memcfg_numa(); - numa_init_array(); - - kva_pages = roundup(calculate_numa_remap_pages(), PTRS_PER_PTE); + x86_numa_init(); - kva_target_pfn = round_down(max_low_pfn - kva_pages, PTRS_PER_PTE); - do { - kva_start_pfn = memblock_find_in_range(kva_target_pfn<<PAGE_SHIFT, - max_low_pfn<<PAGE_SHIFT, - kva_pages<<PAGE_SHIFT, - PTRS_PER_PTE<<PAGE_SHIFT) >> PAGE_SHIFT; - kva_target_pfn -= PTRS_PER_PTE; - } while (kva_start_pfn == MEMBLOCK_ERROR && kva_target_pfn > min_low_pfn); - - if (kva_start_pfn == MEMBLOCK_ERROR) - panic("Can not get kva space\n"); - - printk(KERN_INFO "kva_start_pfn ~ %lx max_low_pfn ~ %lx\n", - kva_start_pfn, max_low_pfn); - printk(KERN_INFO "max_pfn = %lx\n", max_pfn); - - /* avoid clash with initrd */ - memblock_x86_reserve_range(kva_start_pfn<<PAGE_SHIFT, - (kva_start_pfn + kva_pages)<<PAGE_SHIFT, - "KVA PG"); #ifdef CONFIG_HIGHMEM highstart_pfn = highend_pfn = max_pfn; if (max_pfn > max_low_pfn) @@ -409,51 +257,9 @@ void __init initmem_init(void) printk(KERN_DEBUG "Low memory ends at vaddr %08lx\n", (ulong) pfn_to_kaddr(max_low_pfn)); - for_each_online_node(nid) { - init_remap_allocator(nid); - - allocate_pgdat(nid); - } - remap_numa_kva(); printk(KERN_DEBUG "High memory starts at vaddr %08lx\n", (ulong) pfn_to_kaddr(highstart_pfn)); - for_each_online_node(nid) - propagate_e820_map_node(nid); - - for_each_online_node(nid) { - memset(NODE_DATA(nid), 0, sizeof(struct pglist_data)); - NODE_DATA(nid)->node_id = nid; - } setup_bootmem_allocator(); } - -#ifdef CONFIG_MEMORY_HOTPLUG -static int paddr_to_nid(u64 addr) -{ - int nid; - unsigned long pfn = PFN_DOWN(addr); - - for_each_node(nid) - if (node_start_pfn[nid] <= pfn && - pfn < node_end_pfn[nid]) - return nid; - - return -1; -} - -/* - * This function is used to ask node id BEFORE memmap and mem_section's - * initialization (pfn_to_nid() can't be used yet). - * If _PXM is not defined on ACPI's DSDT, node id must be found by this. - */ -int memory_add_physaddr_to_nid(u64 addr) -{ - int nid = paddr_to_nid(addr); - return (nid >= 0) ? nid : 0; -} - -EXPORT_SYMBOL_GPL(memory_add_physaddr_to_nid); -#endif - diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c index e8c00cc72033..dd27f401f0a0 100644 --- a/arch/x86/mm/numa_64.c +++ b/arch/x86/mm/numa_64.c @@ -2,646 +2,13 @@ * Generic VM initialization for x86-64 NUMA setups. * Copyright 2002,2003 Andi Kleen, SuSE Labs. */ -#include <linux/kernel.h> -#include <linux/mm.h> -#include <linux/string.h> -#include <linux/init.h> #include <linux/bootmem.h> -#include <linux/memblock.h> -#include <linux/mmzone.h> -#include <linux/ctype.h> -#include <linux/module.h> -#include <linux/nodemask.h> -#include <linux/sched.h> -#include <linux/acpi.h> - -#include <asm/e820.h> -#include <asm/proto.h> -#include <asm/dma.h> -#include <asm/acpi.h> -#include <asm/amd_nb.h> #include "numa_internal.h" -struct pglist_data *node_data[MAX_NUMNODES] __read_mostly; -EXPORT_SYMBOL(node_data); - -nodemask_t numa_nodes_parsed __initdata; - -struct memnode memnode; - -static unsigned long __initdata nodemap_addr; -static unsigned long __initdata nodemap_size; - -static struct numa_meminfo numa_meminfo __initdata; - -static int numa_distance_cnt; -static u8 *numa_distance; - -/* - * Given a shift value, try to populate memnodemap[] - * Returns : - * 1 if OK - * 0 if memnodmap[] too small (of shift too small) - * -1 if node overlap or lost ram (shift too big) - */ -static int __init populate_memnodemap(const struct numa_meminfo *mi, int shift) -{ - unsigned long addr, end; - int i, res = -1; - - memset(memnodemap, 0xff, sizeof(s16)*memnodemapsize); - for (i = 0; i < mi->nr_blks; i++) { - addr = mi->blk[i].start; - end = mi->blk[i].end; - if (addr >= end) - continue; - if ((end >> shift) >= memnodemapsize) - return 0; - do { - if (memnodemap[addr >> shift] != NUMA_NO_NODE) - return -1; - memnodemap[addr >> shift] = mi->blk[i].nid; - addr += (1UL << shift); - } while (addr < end); - res = 1; - } - return res; -} - -static int __init allocate_cachealigned_memnodemap(void) -{ - unsigned long addr; - - memnodemap = memnode.embedded_map; - if (memnodemapsize <= ARRAY_SIZE(memnode.embedded_map)) - return 0; - - addr = 0x8000; - nodemap_size = roundup(sizeof(s16) * memnodemapsize, L1_CACHE_BYTES); - nodemap_addr = memblock_find_in_range(addr, get_max_mapped(), - nodemap_size, L1_CACHE_BYTES); - if (nodemap_addr == MEMBLOCK_ERROR) { - printk(KERN_ERR - "NUMA: Unable to allocate Memory to Node hash map\n"); - nodemap_addr = nodemap_size = 0; - return -1; - } - memnodemap = phys_to_virt(nodemap_addr); - memblock_x86_reserve_range(nodemap_addr, nodemap_addr + nodemap_size, "MEMNODEMAP"); - - printk(KERN_DEBUG "NUMA: Allocated memnodemap from %lx - %lx\n", - nodemap_addr, nodemap_addr + nodemap_size); - return 0; -} - -/* - * The LSB of all start and end addresses in the node map is the value of the - * maximum possible shift. - */ -static int __init extract_lsb_from_nodes(const struct numa_meminfo *mi) -{ - int i, nodes_used = 0; - unsigned long start, end; - unsigned long bitfield = 0, memtop = 0; - - for (i = 0; i < mi->nr_blks; i++) { - start = mi->blk[i].start; - end = mi->blk[i].end; - if (start >= end) - continue; - bitfield |= start; - nodes_used++; - if (end > memtop) - memtop = end; - } - if (nodes_used <= 1) - i = 63; - else - i = find_first_bit(&bitfield, sizeof(unsigned long)*8); - memnodemapsize = (memtop >> i)+1; - return i; -} - -static int __init compute_hash_shift(const struct numa_meminfo *mi) -{ - int shift; - - shift = extract_lsb_from_nodes(mi); - if (allocate_cachealigned_memnodemap()) - return -1; - printk(KERN_DEBUG "NUMA: Using %d for the hash shift.\n", - shift); - - if (populate_memnodemap(mi, shift) != 1) { - printk(KERN_INFO "Your memory is not aligned you need to " - "rebuild your kernel with a bigger NODEMAPSIZE " - "shift=%d\n", shift); - return -1; - } - return shift; -} - -int __meminit __early_pfn_to_nid(unsigned long pfn) -{ - return phys_to_nid(pfn << PAGE_SHIFT); -} - -static void * __init early_node_mem(int nodeid, unsigned long start, - unsigned long end, unsigned long size, - unsigned long align) -{ - unsigned long mem; - - /* - * put it on high as possible - * something will go with NODE_DATA - */ - if (start < (MAX_DMA_PFN<<PAGE_SHIFT)) - start = MAX_DMA_PFN<<PAGE_SHIFT; - if (start < (MAX_DMA32_PFN<<PAGE_SHIFT) && - end > (MAX_DMA32_PFN<<PAGE_SHIFT)) - start = MAX_DMA32_PFN<<PAGE_SHIFT; - mem = memblock_x86_find_in_range_node(nodeid, start, end, size, align); - if (mem != MEMBLOCK_ERROR) - return __va(mem); - - /* extend the search scope */ - end = max_pfn_mapped << PAGE_SHIFT; - start = MAX_DMA_PFN << PAGE_SHIFT; - mem = memblock_find_in_range(start, end, size, align); - if (mem != MEMBLOCK_ERROR) - return __va(mem); - - printk(KERN_ERR "Cannot find %lu bytes in node %d\n", - size, nodeid); - - return NULL; -} - -static int __init numa_add_memblk_to(int nid, u64 start, u64 end, - struct numa_meminfo *mi) -{ - /* ignore zero length blks */ - if (start == end) - return 0; - - /* whine about and ignore invalid blks */ - if (start > end || nid < 0 || nid >= MAX_NUMNODES) { - pr_warning("NUMA: Warning: invalid memblk node %d (%Lx-%Lx)\n", - nid, start, end); - return 0; - } - - if (mi->nr_blks >= NR_NODE_MEMBLKS) { - pr_err("NUMA: too many memblk ranges\n"); - return -EINVAL; - } - - mi->blk[mi->nr_blks].start = start; - mi->blk[mi->nr_blks].end = end; - mi->blk[mi->nr_blks].nid = nid; - mi->nr_blks++; - return 0; -} - -/** - * numa_remove_memblk_from - Remove one numa_memblk from a numa_meminfo - * @idx: Index of memblk to remove - * @mi: numa_meminfo to remove memblk from - * - * Remove @idx'th numa_memblk from @mi by shifting @mi->blk[] and - * decrementing @mi->nr_blks. - */ -void __init numa_remove_memblk_from(int idx, struct numa_meminfo *mi) -{ - mi->nr_blks--; - memmove(&mi->blk[idx], &mi->blk[idx + 1], - (mi->nr_blks - idx) * sizeof(mi->blk[0])); -} - -/** - * numa_add_memblk - Add one numa_memblk to numa_meminfo - * @nid: NUMA node ID of the new memblk - * @start: Start address of the new memblk - * @end: End address of the new memblk - * - * Add a new memblk to the default numa_meminfo. - * - * RETURNS: - * 0 on success, -errno on failure. - */ -int __init numa_add_memblk(int nid, u64 start, u64 end) -{ - return numa_add_memblk_to(nid, start, end, &numa_meminfo); -} - -/* Initialize bootmem allocator for a node */ -void __init -setup_node_bootmem(int nodeid, unsigned long start, unsigned long end) -{ - unsigned long start_pfn, last_pfn, nodedata_phys; - const int pgdat_size = roundup(sizeof(pg_data_t), PAGE_SIZE); - int nid; - - if (!end) - return; - - /* - * Don't confuse VM with a node that doesn't have the - * minimum amount of memory: - */ - if (end && (end - start) < NODE_MIN_SIZE) - return; - - start = roundup(start, ZONE_ALIGN); - - printk(KERN_INFO "Initmem setup node %d %016lx-%016lx\n", nodeid, - start, end); - - start_pfn = start >> PAGE_SHIFT; - last_pfn = end >> PAGE_SHIFT; - - node_data[nodeid] = early_node_mem(nodeid, start, end, pgdat_size, - SMP_CACHE_BYTES); - if (node_data[nodeid] == NULL) - return; - nodedata_phys = __pa(node_data[nodeid]); - memblock_x86_reserve_range(nodedata_phys, nodedata_phys + pgdat_size, "NODE_DATA"); - printk(KERN_INFO " NODE_DATA [%016lx - %016lx]\n", nodedata_phys, - nodedata_phys + pgdat_size - 1); - nid = phys_to_nid(nodedata_phys); - if (nid != nodeid) - printk(KERN_INFO " NODE_DATA(%d) on node %d\n", nodeid, nid); - - memset(NODE_DATA(nodeid), 0, sizeof(pg_data_t)); - NODE_DATA(nodeid)->node_id = nodeid; - NODE_DATA(nodeid)->node_start_pfn = start_pfn; - NODE_DATA(nodeid)->node_spanned_pages = last_pfn - start_pfn; - - node_set_online(nodeid); -} - -/** - * numa_cleanup_meminfo - Cleanup a numa_meminfo - * @mi: numa_meminfo to clean up - * - * Sanitize @mi by merging and removing unncessary memblks. Also check for - * conflicts and clear unused memblks. - * - * RETURNS: - * 0 on success, -errno on failure. - */ -int __init numa_cleanup_meminfo(struct numa_meminfo *mi) -{ - const u64 low = 0; - const u64 high = (u64)max_pfn << PAGE_SHIFT; - int i, j, k; - - for (i = 0; i < mi->nr_blks; i++) { - struct numa_memblk *bi = &mi->blk[i]; - - /* make sure all blocks are inside the limits */ - bi->start = max(bi->start, low); - bi->end = min(bi->end, high); - - /* and there's no empty block */ - if (bi->start == bi->end) { - numa_remove_memblk_from(i--, mi); - continue; - } - - for (j = i + 1; j < mi->nr_blks; j++) { - struct numa_memblk *bj = &mi->blk[j]; - unsigned long start, end; - - /* - * See whether there are overlapping blocks. Whine - * about but allow overlaps of the same nid. They - * will be merged below. - */ - if (bi->end > bj->start && bi->start < bj->end) { - if (bi->nid != bj->nid) { - pr_err("NUMA: node %d (%Lx-%Lx) overlaps with node %d (%Lx-%Lx)\n", - bi->nid, bi->start, bi->end, - bj->nid, bj->start, bj->end); - return -EINVAL; - } - pr_warning("NUMA: Warning: node %d (%Lx-%Lx) overlaps with itself (%Lx-%Lx)\n", - bi->nid, bi->start, bi->end, - bj->start, bj->end); - } - - /* - * Join together blocks on the same node, holes - * between which don't overlap with memory on other - * nodes. - */ - if (bi->nid != bj->nid) - continue; - start = max(min(bi->start, bj->start), low); - end = min(max(bi->end, bj->end), high); - for (k = 0; k < mi->nr_blks; k++) { - struct numa_memblk *bk = &mi->blk[k]; - - if (bi->nid == bk->nid) - continue; - if (start < bk->end && end > bk->start) - break; - } - if (k < mi->nr_blks) - continue; - printk(KERN_INFO "NUMA: Node %d [%Lx,%Lx) + [%Lx,%Lx) -> [%lx,%lx)\n", - bi->nid, bi->start, bi->end, bj->start, bj->end, - start, end); - bi->start = start; - bi->end = end; - numa_remove_memblk_from(j--, mi); - } - } - - for (i = mi->nr_blks; i < ARRAY_SIZE(mi->blk); i++) { - mi->blk[i].start = mi->blk[i].end = 0; - mi->blk[i].nid = NUMA_NO_NODE; - } - - return 0; -} - -/* - * Set nodes, which have memory in @mi, in *@nodemask. - */ -static void __init numa_nodemask_from_meminfo(nodemask_t *nodemask, - const struct numa_meminfo *mi) -{ - int i; - - for (i = 0; i < ARRAY_SIZE(mi->blk); i++) - if (mi->blk[i].start != mi->blk[i].end && - mi->blk[i].nid != NUMA_NO_NODE) - node_set(mi->blk[i].nid, *nodemask); -} - -/** - * numa_reset_distance - Reset NUMA distance table - * - * The current table is freed. The next numa_set_distance() call will - * create a new one. - */ -void __init numa_reset_distance(void) -{ - size_t size = numa_distance_cnt * numa_distance_cnt * sizeof(numa_distance[0]); - - /* numa_distance could be 1LU marking allocation failure, test cnt */ - if (numa_distance_cnt) - memblock_x86_free_range(__pa(numa_distance), - __pa(numa_distance) + size); - numa_distance_cnt = 0; - numa_distance = NULL; /* enable table creation */ -} - -static int __init numa_alloc_distance(void) -{ - nodemask_t nodes_parsed; - size_t size; - int i, j, cnt = 0; - u64 phys; - - /* size the new table and allocate it */ - nodes_parsed = numa_nodes_parsed; - numa_nodemask_from_meminfo(&nodes_parsed, &numa_meminfo); - - for_each_node_mask(i, nodes_parsed) - cnt = i; - cnt++; - size = cnt * cnt * sizeof(numa_distance[0]); - - phys = memblock_find_in_range(0, (u64)max_pfn_mapped << PAGE_SHIFT, - size, PAGE_SIZE); - if (phys == MEMBLOCK_ERROR) { - pr_warning("NUMA: Warning: can't allocate distance table!\n"); - /* don't retry until explicitly reset */ - numa_distance = (void *)1LU; - return -ENOMEM; - } - memblock_x86_reserve_range(phys, phys + size, "NUMA DIST"); - - numa_distance = __va(phys); - numa_distance_cnt = cnt; - - /* fill with the default distances */ - for (i = 0; i < cnt; i++) - for (j = 0; j < cnt; j++) - numa_distance[i * cnt + j] = i == j ? - LOCAL_DISTANCE : REMOTE_DISTANCE; - printk(KERN_DEBUG "NUMA: Initialized distance table, cnt=%d\n", cnt); - - return 0; -} - -/** - * numa_set_distance - Set NUMA distance from one NUMA to another - * @from: the 'from' node to set distance - * @to: the 'to' node to set distance - * @distance: NUMA distance - * - * Set the distance from node @from to @to to @distance. If distance table - * doesn't exist, one which is large enough to accommodate all the currently - * known nodes will be created. - * - * If such table cannot be allocated, a warning is printed and further - * calls are ignored until the distance table is reset with - * numa_reset_distance(). - * - * If @from or @to is higher than the highest known node at the time of - * table creation or @distance doesn't make sense, the call is ignored. - * This is to allow simplification of specific NUMA config implementations. - */ -void __init numa_set_distance(int from, int to, int distance) -{ - if (!numa_distance && numa_alloc_distance() < 0) - return; - - if (from >= numa_distance_cnt || to >= numa_distance_cnt) { - printk_once(KERN_DEBUG "NUMA: Debug: distance out of bound, from=%d to=%d distance=%d\n", - from, to, distance); - return; - } - - if ((u8)distance != distance || - (from == to && distance != LOCAL_DISTANCE)) { - pr_warn_once("NUMA: Warning: invalid distance parameter, from=%d to=%d distance=%d\n", - from, to, distance); - return; - } - - numa_distance[from * numa_distance_cnt + to] = distance; -} - -int __node_distance(int from, int to) -{ - if (from >= numa_distance_cnt || to >= numa_distance_cnt) - return from == to ? LOCAL_DISTANCE : REMOTE_DISTANCE; - return numa_distance[from * numa_distance_cnt + to]; -} -EXPORT_SYMBOL(__node_distance); - -/* - * Sanity check to catch more bad NUMA configurations (they are amazingly - * common). Make sure the nodes cover all memory. - */ -static bool __init numa_meminfo_cover_memory(const struct numa_meminfo *mi) -{ - unsigned long numaram, e820ram; - int i; - - numaram = 0; - for (i = 0; i < mi->nr_blks; i++) { - unsigned long s = mi->blk[i].start >> PAGE_SHIFT; - unsigned long e = mi->blk[i].end >> PAGE_SHIFT; - numaram += e - s; - numaram -= __absent_pages_in_range(mi->blk[i].nid, s, e); - if ((long)numaram < 0) - numaram = 0; - } - - e820ram = max_pfn - (memblock_x86_hole_size(0, - max_pfn << PAGE_SHIFT) >> PAGE_SHIFT); - /* We seem to lose 3 pages somewhere. Allow 1M of slack. */ - if ((long)(e820ram - numaram) >= (1 << (20 - PAGE_SHIFT))) { - printk(KERN_ERR "NUMA: nodes only cover %luMB of your %luMB e820 RAM. Not used.\n", - (numaram << PAGE_SHIFT) >> 20, - (e820ram << PAGE_SHIFT) >> 20); - return false; - } - return true; -} - -static int __init numa_register_memblks(struct numa_meminfo *mi) -{ - int i, nid; - - /* Account for nodes with cpus and no memory */ - node_possible_map = numa_nodes_parsed; - numa_nodemask_from_meminfo(&node_possible_map, mi); - if (WARN_ON(nodes_empty(node_possible_map))) - return -EINVAL; - - memnode_shift = compute_hash_shift(mi); - if (memnode_shift < 0) { - printk(KERN_ERR "NUMA: No NUMA node hash function found. Contact maintainer\n"); - return -EINVAL; - } - - for (i = 0; i < mi->nr_blks; i++) - memblock_x86_register_active_regions(mi->blk[i].nid, - mi->blk[i].start >> PAGE_SHIFT, - mi->blk[i].end >> PAGE_SHIFT); - - /* for out of order entries */ - sort_node_map(); - if (!numa_meminfo_cover_memory(mi)) - return -EINVAL; - - /* Finally register nodes. */ - for_each_node_mask(nid, node_possible_map) { - u64 start = (u64)max_pfn << PAGE_SHIFT; - u64 end = 0; - - for (i = 0; i < mi->nr_blks; i++) { - if (nid != mi->blk[i].nid) - continue; - start = min(mi->blk[i].start, start); - end = max(mi->blk[i].end, end); - } - - if (start < end) - setup_node_bootmem(nid, start, end); - } - - return 0; -} - -/** - * dummy_numma_init - Fallback dummy NUMA init - * - * Used if there's no underlying NUMA architecture, NUMA initialization - * fails, or NUMA is disabled on the command line. - * - * Must online at least one node and add memory blocks that cover all - * allowed memory. This function must not fail. - */ -static int __init dummy_numa_init(void) -{ - printk(KERN_INFO "%s\n", - numa_off ? "NUMA turned off" : "No NUMA configuration found"); - printk(KERN_INFO "Faking a node at %016lx-%016lx\n", - 0LU, max_pfn << PAGE_SHIFT); - - node_set(0, numa_nodes_parsed); - numa_add_memblk(0, 0, (u64)max_pfn << PAGE_SHIFT); - - return 0; -} - -static int __init numa_init(int (*init_func)(void)) -{ - int i; - int ret; - - for (i = 0; i < MAX_LOCAL_APIC; i++) - set_apicid_to_node(i, NUMA_NO_NODE); - - nodes_clear(numa_nodes_parsed); - nodes_clear(node_possible_map); - nodes_clear(node_online_map); - memset(&numa_meminfo, 0, sizeof(numa_meminfo)); - remove_all_active_ranges(); - numa_reset_distance(); - - ret = init_func(); - if (ret < 0) - return ret; - ret = numa_cleanup_meminfo(&numa_meminfo); - if (ret < 0) - return ret; - - numa_emulation(&numa_meminfo, numa_distance_cnt); - - ret = numa_register_memblks(&numa_meminfo); - if (ret < 0) - return ret; - - for (i = 0; i < nr_cpu_ids; i++) { - int nid = early_cpu_to_node(i); - - if (nid == NUMA_NO_NODE) - continue; - if (!node_online(nid)) - numa_clear_node(i); - } - numa_init_array(); - return 0; -} - void __init initmem_init(void) { - int ret; - - if (!numa_off) { -#ifdef CONFIG_ACPI_NUMA - ret = numa_init(x86_acpi_numa_init); - if (!ret) - return; -#endif -#ifdef CONFIG_AMD_NUMA - ret = numa_init(amd_numa_init); - if (!ret) - return; -#endif - } - - numa_init(dummy_numa_init); + x86_numa_init(); } unsigned long __init numa_free_all_bootmem(void) @@ -656,12 +23,3 @@ unsigned long __init numa_free_all_bootmem(void) return pages; } - -int __cpuinit numa_cpu_node(int cpu) -{ - int apicid = early_per_cpu(x86_cpu_to_apicid, cpu); - - if (apicid != BAD_APICID) - return __apicid_to_node[apicid]; - return NUMA_NO_NODE; -} diff --git a/arch/x86/mm/numa_emulation.c b/arch/x86/mm/numa_emulation.c index de84cc140379..d0ed086b6247 100644 --- a/arch/x86/mm/numa_emulation.c +++ b/arch/x86/mm/numa_emulation.c @@ -5,6 +5,7 @@ #include <linux/errno.h> #include <linux/topology.h> #include <linux/memblock.h> +#include <linux/bootmem.h> #include <asm/dma.h> #include "numa_internal.h" @@ -84,7 +85,13 @@ static int __init split_nodes_interleave(struct numa_meminfo *ei, nr_nodes = MAX_NUMNODES; } - size = (max_addr - addr - memblock_x86_hole_size(addr, max_addr)) / nr_nodes; + /* + * Calculate target node size. x86_32 freaks on __udivdi3() so do + * the division in ulong number of pages and convert back. + */ + size = max_addr - addr - memblock_x86_hole_size(addr, max_addr); + size = PFN_PHYS((unsigned long)(size >> PAGE_SHIFT) / nr_nodes); + /* * Calculate the number of big nodes that can be allocated as a result * of consolidating the remainder. @@ -226,7 +233,7 @@ static int __init split_nodes_size_interleave(struct numa_meminfo *ei, */ while (nodes_weight(physnode_mask)) { for_each_node_mask(i, physnode_mask) { - u64 dma32_end = MAX_DMA32_PFN << PAGE_SHIFT; + u64 dma32_end = PFN_PHYS(MAX_DMA32_PFN); u64 start, limit, end; int phys_blk; @@ -298,7 +305,7 @@ void __init numa_emulation(struct numa_meminfo *numa_meminfo, int numa_dist_cnt) { static struct numa_meminfo ei __initdata; static struct numa_meminfo pi __initdata; - const u64 max_addr = max_pfn << PAGE_SHIFT; + const u64 max_addr = PFN_PHYS(max_pfn); u8 *phys_dist = NULL; size_t phys_size = numa_dist_cnt * numa_dist_cnt * sizeof(phys_dist[0]); int max_emu_nid, dfl_phys_nid; @@ -342,8 +349,7 @@ void __init numa_emulation(struct numa_meminfo *numa_meminfo, int numa_dist_cnt) if (numa_dist_cnt) { u64 phys; - phys = memblock_find_in_range(0, - (u64)max_pfn_mapped << PAGE_SHIFT, + phys = memblock_find_in_range(0, PFN_PHYS(max_pfn_mapped), phys_size, PAGE_SIZE); if (phys == MEMBLOCK_ERROR) { pr_warning("NUMA: Warning: can't allocate copy of distance table, disabling emulation\n"); diff --git a/arch/x86/mm/numa_internal.h b/arch/x86/mm/numa_internal.h index ef2d97377d7c..7178c3afe05e 100644 --- a/arch/x86/mm/numa_internal.h +++ b/arch/x86/mm/numa_internal.h @@ -19,6 +19,14 @@ void __init numa_remove_memblk_from(int idx, struct numa_meminfo *mi); int __init numa_cleanup_meminfo(struct numa_meminfo *mi); void __init numa_reset_distance(void); +void __init x86_numa_init(void); + +#ifdef CONFIG_X86_64 +static inline void init_alloc_remap(int nid, u64 start, u64 end) { } +#else +void __init init_alloc_remap(int nid, u64 start, u64 end); +#endif + #ifdef CONFIG_NUMA_EMU void __init numa_emulation(struct numa_meminfo *numa_meminfo, int numa_dist_cnt); diff --git a/arch/x86/mm/pf_in.c b/arch/x86/mm/pf_in.c index 38e6d174c497..9f0614daea85 100644 --- a/arch/x86/mm/pf_in.c +++ b/arch/x86/mm/pf_in.c @@ -414,22 +414,17 @@ unsigned long get_ins_reg_val(unsigned long ins_addr, struct pt_regs *regs) unsigned char *p; struct prefix_bits prf; int i; - unsigned long rv; p = (unsigned char *)ins_addr; p += skip_prefix(p, &prf); p += get_opcode(p, &opcode); for (i = 0; i < ARRAY_SIZE(reg_rop); i++) - if (reg_rop[i] == opcode) { - rv = REG_READ; + if (reg_rop[i] == opcode) goto do_work; - } for (i = 0; i < ARRAY_SIZE(reg_wop); i++) - if (reg_wop[i] == opcode) { - rv = REG_WRITE; + if (reg_wop[i] == opcode) goto do_work; - } printk(KERN_ERR "mmiotrace: Not a register instruction, opcode " "0x%02x\n", opcode); @@ -474,16 +469,13 @@ unsigned long get_ins_imm_val(unsigned long ins_addr) unsigned char *p; struct prefix_bits prf; int i; - unsigned long rv; p = (unsigned char *)ins_addr; p += skip_prefix(p, &prf); p += get_opcode(p, &opcode); for (i = 0; i < ARRAY_SIZE(imm_wop); i++) - if (imm_wop[i] == opcode) { - rv = IMM_WRITE; + if (imm_wop[i] == opcode) goto do_work; - } printk(KERN_ERR "mmiotrace: Not an immediate instruction, opcode " "0x%02x\n", opcode); diff --git a/arch/x86/mm/srat_64.c b/arch/x86/mm/srat.c index 8e9d3394f6d4..81dbfdeb080d 100644 --- a/arch/x86/mm/srat_64.c +++ b/arch/x86/mm/srat.c @@ -26,8 +26,6 @@ int acpi_numa __initdata; -static struct bootnode nodes_add[MAX_NUMNODES]; - static __init int setup_node(int pxm) { return acpi_map_pxm_to_node(pxm); @@ -37,7 +35,6 @@ static __init void bad_srat(void) { printk(KERN_ERR "SRAT: SRAT not used.\n"); acpi_numa = -1; - memset(nodes_add, 0, sizeof(nodes_add)); } static __init inline int srat_disabled(void) @@ -131,73 +128,17 @@ acpi_numa_processor_affinity_init(struct acpi_srat_cpu_affinity *pa) pxm, apic_id, node); } -#ifdef CONFIG_MEMORY_HOTPLUG_SPARSE +#ifdef CONFIG_MEMORY_HOTPLUG static inline int save_add_info(void) {return 1;} #else static inline int save_add_info(void) {return 0;} #endif -/* - * Update nodes_add[] - * This code supports one contiguous hot add area per node - */ -static void __init -update_nodes_add(int node, unsigned long start, unsigned long end) -{ - unsigned long s_pfn = start >> PAGE_SHIFT; - unsigned long e_pfn = end >> PAGE_SHIFT; - int changed = 0; - struct bootnode *nd = &nodes_add[node]; - - /* I had some trouble with strange memory hotadd regions breaking - the boot. Be very strict here and reject anything unexpected. - If you want working memory hotadd write correct SRATs. - - The node size check is a basic sanity check to guard against - mistakes */ - if ((signed long)(end - start) < NODE_MIN_SIZE) { - printk(KERN_ERR "SRAT: Hotplug area too small\n"); - return; - } - - /* This check might be a bit too strict, but I'm keeping it for now. */ - if (absent_pages_in_range(s_pfn, e_pfn) != e_pfn - s_pfn) { - printk(KERN_ERR - "SRAT: Hotplug area %lu -> %lu has existing memory\n", - s_pfn, e_pfn); - return; - } - - /* Looks good */ - - if (nd->start == nd->end) { - nd->start = start; - nd->end = end; - changed = 1; - } else { - if (nd->start == end) { - nd->start = start; - changed = 1; - } - if (nd->end == start) { - nd->end = end; - changed = 1; - } - if (!changed) - printk(KERN_ERR "SRAT: Hotplug zone not continuous. Partly ignored\n"); - } - - if (changed) { - node_set(node, numa_nodes_parsed); - printk(KERN_INFO "SRAT: hot plug zone found %Lx - %Lx\n", - nd->start, nd->end); - } -} /* Callback for parsing of the Proximity Domain <-> Memory Area mappings */ void __init acpi_numa_memory_affinity_init(struct acpi_srat_mem_affinity *ma) { - unsigned long start, end; + u64 start, end; int node, pxm; if (srat_disabled()) @@ -226,11 +167,8 @@ acpi_numa_memory_affinity_init(struct acpi_srat_mem_affinity *ma) return; } - printk(KERN_INFO "SRAT: Node %u PXM %u %lx-%lx\n", node, pxm, + printk(KERN_INFO "SRAT: Node %u PXM %u %Lx-%Lx\n", node, pxm, start, end); - - if (ma->flags & ACPI_SRAT_MEM_HOT_PLUGGABLE) - update_nodes_add(node, start, end); } void __init acpi_numa_arch_fixup(void) {} @@ -244,17 +182,3 @@ int __init x86_acpi_numa_init(void) return ret; return srat_disabled() ? -EINVAL : 0; } - -#if defined(CONFIG_MEMORY_HOTPLUG_SPARSE) || defined(CONFIG_ACPI_HOTPLUG_MEMORY) -int memory_add_physaddr_to_nid(u64 start) -{ - int i, ret = 0; - - for_each_node(i) - if (nodes_add[i].start <= start && nodes_add[i].end > start) - ret = i; - - return ret; -} -EXPORT_SYMBOL_GPL(memory_add_physaddr_to_nid); -#endif diff --git a/arch/x86/mm/srat_32.c b/arch/x86/mm/srat_32.c deleted file mode 100644 index 364f36bdfad8..000000000000 --- a/arch/x86/mm/srat_32.c +++ /dev/null @@ -1,288 +0,0 @@ -/* - * Some of the code in this file has been gleaned from the 64 bit - * discontigmem support code base. - * - * Copyright (C) 2002, IBM Corp. - * - * All rights reserved. - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, but - * WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE, GOOD TITLE or - * NON INFRINGEMENT. See the GNU General Public License for more - * details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. - * - * Send feedback to Pat Gaughen <gone@us.ibm.com> - */ -#include <linux/mm.h> -#include <linux/bootmem.h> -#include <linux/memblock.h> -#include <linux/mmzone.h> -#include <linux/acpi.h> -#include <linux/nodemask.h> -#include <asm/srat.h> -#include <asm/topology.h> -#include <asm/smp.h> -#include <asm/e820.h> - -/* - * proximity macros and definitions - */ -#define NODE_ARRAY_INDEX(x) ((x) / 8) /* 8 bits/char */ -#define NODE_ARRAY_OFFSET(x) ((x) % 8) /* 8 bits/char */ -#define BMAP_SET(bmap, bit) ((bmap)[NODE_ARRAY_INDEX(bit)] |= 1 << NODE_ARRAY_OFFSET(bit)) -#define BMAP_TEST(bmap, bit) ((bmap)[NODE_ARRAY_INDEX(bit)] & (1 << NODE_ARRAY_OFFSET(bit))) -/* bitmap length; _PXM is at most 255 */ -#define PXM_BITMAP_LEN (MAX_PXM_DOMAINS / 8) -static u8 __initdata pxm_bitmap[PXM_BITMAP_LEN]; /* bitmap of proximity domains */ - -#define MAX_CHUNKS_PER_NODE 3 -#define MAXCHUNKS (MAX_CHUNKS_PER_NODE * MAX_NUMNODES) -struct node_memory_chunk_s { - unsigned long start_pfn; - unsigned long end_pfn; - u8 pxm; // proximity domain of node - u8 nid; // which cnode contains this chunk? - u8 bank; // which mem bank on this node -}; -static struct node_memory_chunk_s __initdata node_memory_chunk[MAXCHUNKS]; - -static int __initdata num_memory_chunks; /* total number of memory chunks */ -static u8 __initdata apicid_to_pxm[MAX_LOCAL_APIC]; - -int acpi_numa __initdata; - -static __init void bad_srat(void) -{ - printk(KERN_ERR "SRAT: SRAT not used.\n"); - acpi_numa = -1; - num_memory_chunks = 0; -} - -static __init inline int srat_disabled(void) -{ - return numa_off || acpi_numa < 0; -} - -/* Identify CPU proximity domains */ -void __init -acpi_numa_processor_affinity_init(struct acpi_srat_cpu_affinity *cpu_affinity) -{ - if (srat_disabled()) - return; - if (cpu_affinity->header.length != - sizeof(struct acpi_srat_cpu_affinity)) { - bad_srat(); - return; - } - - if ((cpu_affinity->flags & ACPI_SRAT_CPU_ENABLED) == 0) - return; /* empty entry */ - - /* mark this node as "seen" in node bitmap */ - BMAP_SET(pxm_bitmap, cpu_affinity->proximity_domain_lo); - - /* don't need to check apic_id here, because it is always 8 bits */ - apicid_to_pxm[cpu_affinity->apic_id] = cpu_affinity->proximity_domain_lo; - - printk(KERN_DEBUG "CPU %02x in proximity domain %02x\n", - cpu_affinity->apic_id, cpu_affinity->proximity_domain_lo); -} - -/* - * Identify memory proximity domains and hot-remove capabilities. - * Fill node memory chunk list structure. - */ -void __init -acpi_numa_memory_affinity_init(struct acpi_srat_mem_affinity *memory_affinity) -{ - unsigned long long paddr, size; - unsigned long start_pfn, end_pfn; - u8 pxm; - struct node_memory_chunk_s *p, *q, *pend; - - if (srat_disabled()) - return; - if (memory_affinity->header.length != - sizeof(struct acpi_srat_mem_affinity)) { - bad_srat(); - return; - } - - if ((memory_affinity->flags & ACPI_SRAT_MEM_ENABLED) == 0) - return; /* empty entry */ - - pxm = memory_affinity->proximity_domain & 0xff; - - /* mark this node as "seen" in node bitmap */ - BMAP_SET(pxm_bitmap, pxm); - - /* calculate info for memory chunk structure */ - paddr = memory_affinity->base_address; - size = memory_affinity->length; - - start_pfn = paddr >> PAGE_SHIFT; - end_pfn = (paddr + size) >> PAGE_SHIFT; - - - if (num_memory_chunks >= MAXCHUNKS) { - printk(KERN_WARNING "Too many mem chunks in SRAT." - " Ignoring %lld MBytes at %llx\n", - size/(1024*1024), paddr); - return; - } - - /* Insertion sort based on base address */ - pend = &node_memory_chunk[num_memory_chunks]; - for (p = &node_memory_chunk[0]; p < pend; p++) { - if (start_pfn < p->start_pfn) - break; - } - if (p < pend) { - for (q = pend; q >= p; q--) - *(q + 1) = *q; - } - p->start_pfn = start_pfn; - p->end_pfn = end_pfn; - p->pxm = pxm; - - num_memory_chunks++; - - printk(KERN_DEBUG "Memory range %08lx to %08lx" - " in proximity domain %02x %s\n", - start_pfn, end_pfn, - pxm, - ((memory_affinity->flags & ACPI_SRAT_MEM_HOT_PLUGGABLE) ? - "enabled and removable" : "enabled" ) ); -} - -/* Callback for SLIT parsing */ -void __init acpi_numa_slit_init(struct acpi_table_slit *slit) -{ -} - -void acpi_numa_arch_fixup(void) -{ -} -/* - * The SRAT table always lists ascending addresses, so can always - * assume that the first "start" address that you see is the real - * start of the node, and that the current "end" address is after - * the previous one. - */ -static __init int node_read_chunk(int nid, struct node_memory_chunk_s *memory_chunk) -{ - /* - * Only add present memory as told by the e820. - * There is no guarantee from the SRAT that the memory it - * enumerates is present at boot time because it represents - * *possible* memory hotplug areas the same as normal RAM. - */ - if (memory_chunk->start_pfn >= max_pfn) { - printk(KERN_INFO "Ignoring SRAT pfns: %08lx - %08lx\n", - memory_chunk->start_pfn, memory_chunk->end_pfn); - return -1; - } - if (memory_chunk->nid != nid) - return -1; - - if (!node_has_online_mem(nid)) - node_start_pfn[nid] = memory_chunk->start_pfn; - - if (node_start_pfn[nid] > memory_chunk->start_pfn) - node_start_pfn[nid] = memory_chunk->start_pfn; - - if (node_end_pfn[nid] < memory_chunk->end_pfn) - node_end_pfn[nid] = memory_chunk->end_pfn; - - return 0; -} - -int __init get_memcfg_from_srat(void) -{ - int i, j, nid; - - if (srat_disabled()) - goto out_fail; - - if (acpi_numa_init() < 0) - goto out_fail; - - if (num_memory_chunks == 0) { - printk(KERN_DEBUG - "could not find any ACPI SRAT memory areas.\n"); - goto out_fail; - } - - /* Calculate total number of nodes in system from PXM bitmap and create - * a set of sequential node IDs starting at zero. (ACPI doesn't seem - * to specify the range of _PXM values.) - */ - /* - * MCD - we no longer HAVE to number nodes sequentially. PXM domain - * numbers could go as high as 256, and MAX_NUMNODES for i386 is typically - * 32, so we will continue numbering them in this manner until MAX_NUMNODES - * approaches MAX_PXM_DOMAINS for i386. - */ - nodes_clear(node_online_map); - for (i = 0; i < MAX_PXM_DOMAINS; i++) { - if (BMAP_TEST(pxm_bitmap, i)) { - int nid = acpi_map_pxm_to_node(i); - node_set_online(nid); - } - } - BUG_ON(num_online_nodes() == 0); - - /* set cnode id in memory chunk structure */ - for (i = 0; i < num_memory_chunks; i++) - node_memory_chunk[i].nid = pxm_to_node(node_memory_chunk[i].pxm); - - printk(KERN_DEBUG "pxm bitmap: "); - for (i = 0; i < sizeof(pxm_bitmap); i++) { - printk(KERN_CONT "%02x ", pxm_bitmap[i]); - } - printk(KERN_CONT "\n"); - printk(KERN_DEBUG "Number of logical nodes in system = %d\n", - num_online_nodes()); - printk(KERN_DEBUG "Number of memory chunks in system = %d\n", - num_memory_chunks); - - for (i = 0; i < MAX_LOCAL_APIC; i++) - set_apicid_to_node(i, pxm_to_node(apicid_to_pxm[i])); - - for (j = 0; j < num_memory_chunks; j++){ - struct node_memory_chunk_s * chunk = &node_memory_chunk[j]; - printk(KERN_DEBUG - "chunk %d nid %d start_pfn %08lx end_pfn %08lx\n", - j, chunk->nid, chunk->start_pfn, chunk->end_pfn); - if (node_read_chunk(chunk->nid, chunk)) - continue; - - memblock_x86_register_active_regions(chunk->nid, chunk->start_pfn, - min(chunk->end_pfn, max_pfn)); - } - /* for out of order entries in SRAT */ - sort_node_map(); - - for_each_online_node(nid) { - unsigned long start = node_start_pfn[nid]; - unsigned long end = min(node_end_pfn[nid], max_pfn); - - memory_present(nid, start, end); - node_remap_size[nid] = node_memmap_size_bytes(nid, start, end); - } - return 1; -out_fail: - printk(KERN_DEBUG "failed to get NUMA memory information from SRAT" - " table\n"); - return 0; -} diff --git a/arch/x86/net/Makefile b/arch/x86/net/Makefile new file mode 100644 index 000000000000..90568c33ddb0 --- /dev/null +++ b/arch/x86/net/Makefile @@ -0,0 +1,4 @@ +# +# Arch-specific network modules +# +obj-$(CONFIG_BPF_JIT) += bpf_jit.o bpf_jit_comp.o diff --git a/arch/x86/net/bpf_jit.S b/arch/x86/net/bpf_jit.S new file mode 100644 index 000000000000..66870223f8c5 --- /dev/null +++ b/arch/x86/net/bpf_jit.S @@ -0,0 +1,140 @@ +/* bpf_jit.S : BPF JIT helper functions + * + * Copyright (C) 2011 Eric Dumazet (eric.dumazet@gmail.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; version 2 + * of the License. + */ +#include <linux/linkage.h> +#include <asm/dwarf2.h> + +/* + * Calling convention : + * rdi : skb pointer + * esi : offset of byte(s) to fetch in skb (can be scratched) + * r8 : copy of skb->data + * r9d : hlen = skb->len - skb->data_len + */ +#define SKBDATA %r8 + +sk_load_word_ind: + .globl sk_load_word_ind + + add %ebx,%esi /* offset += X */ +# test %esi,%esi /* if (offset < 0) goto bpf_error; */ + js bpf_error + +sk_load_word: + .globl sk_load_word + + mov %r9d,%eax # hlen + sub %esi,%eax # hlen - offset + cmp $3,%eax + jle bpf_slow_path_word + mov (SKBDATA,%rsi),%eax + bswap %eax /* ntohl() */ + ret + + +sk_load_half_ind: + .globl sk_load_half_ind + + add %ebx,%esi /* offset += X */ + js bpf_error + +sk_load_half: + .globl sk_load_half + + mov %r9d,%eax + sub %esi,%eax # hlen - offset + cmp $1,%eax + jle bpf_slow_path_half + movzwl (SKBDATA,%rsi),%eax + rol $8,%ax # ntohs() + ret + +sk_load_byte_ind: + .globl sk_load_byte_ind + add %ebx,%esi /* offset += X */ + js bpf_error + +sk_load_byte: + .globl sk_load_byte + + cmp %esi,%r9d /* if (offset >= hlen) goto bpf_slow_path_byte */ + jle bpf_slow_path_byte + movzbl (SKBDATA,%rsi),%eax + ret + +/** + * sk_load_byte_msh - BPF_S_LDX_B_MSH helper + * + * Implements BPF_S_LDX_B_MSH : ldxb 4*([offset]&0xf) + * Must preserve A accumulator (%eax) + * Inputs : %esi is the offset value, already known positive + */ +ENTRY(sk_load_byte_msh) + CFI_STARTPROC + cmp %esi,%r9d /* if (offset >= hlen) goto bpf_slow_path_byte_msh */ + jle bpf_slow_path_byte_msh + movzbl (SKBDATA,%rsi),%ebx + and $15,%bl + shl $2,%bl + ret + CFI_ENDPROC +ENDPROC(sk_load_byte_msh) + +bpf_error: +# force a return 0 from jit handler + xor %eax,%eax + mov -8(%rbp),%rbx + leaveq + ret + +/* rsi contains offset and can be scratched */ +#define bpf_slow_path_common(LEN) \ + push %rdi; /* save skb */ \ + push %r9; \ + push SKBDATA; \ +/* rsi already has offset */ \ + mov $LEN,%ecx; /* len */ \ + lea -12(%rbp),%rdx; \ + call skb_copy_bits; \ + test %eax,%eax; \ + pop SKBDATA; \ + pop %r9; \ + pop %rdi + + +bpf_slow_path_word: + bpf_slow_path_common(4) + js bpf_error + mov -12(%rbp),%eax + bswap %eax + ret + +bpf_slow_path_half: + bpf_slow_path_common(2) + js bpf_error + mov -12(%rbp),%ax + rol $8,%ax + movzwl %ax,%eax + ret + +bpf_slow_path_byte: + bpf_slow_path_common(1) + js bpf_error + movzbl -12(%rbp),%eax + ret + +bpf_slow_path_byte_msh: + xchg %eax,%ebx /* dont lose A , X is about to be scratched */ + bpf_slow_path_common(1) + js bpf_error + movzbl -12(%rbp),%eax + and $15,%al + shl $2,%al + xchg %eax,%ebx + ret diff --git a/arch/x86/net/bpf_jit_comp.c b/arch/x86/net/bpf_jit_comp.c new file mode 100644 index 000000000000..bfab3fa10edc --- /dev/null +++ b/arch/x86/net/bpf_jit_comp.c @@ -0,0 +1,654 @@ +/* bpf_jit_comp.c : BPF JIT compiler + * + * Copyright (C) 2011 Eric Dumazet (eric.dumazet@gmail.com) + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; version 2 + * of the License. + */ +#include <linux/moduleloader.h> +#include <asm/cacheflush.h> +#include <linux/netdevice.h> +#include <linux/filter.h> + +/* + * Conventions : + * EAX : BPF A accumulator + * EBX : BPF X accumulator + * RDI : pointer to skb (first argument given to JIT function) + * RBP : frame pointer (even if CONFIG_FRAME_POINTER=n) + * ECX,EDX,ESI : scratch registers + * r9d : skb->len - skb->data_len (headlen) + * r8 : skb->data + * -8(RBP) : saved RBX value + * -16(RBP)..-80(RBP) : BPF_MEMWORDS values + */ +int bpf_jit_enable __read_mostly; + +/* + * assembly code in arch/x86/net/bpf_jit.S + */ +extern u8 sk_load_word[], sk_load_half[], sk_load_byte[], sk_load_byte_msh[]; +extern u8 sk_load_word_ind[], sk_load_half_ind[], sk_load_byte_ind[]; + +static inline u8 *emit_code(u8 *ptr, u32 bytes, unsigned int len) +{ + if (len == 1) + *ptr = bytes; + else if (len == 2) + *(u16 *)ptr = bytes; + else { + *(u32 *)ptr = bytes; + barrier(); + } + return ptr + len; +} + +#define EMIT(bytes, len) do { prog = emit_code(prog, bytes, len); } while (0) + +#define EMIT1(b1) EMIT(b1, 1) +#define EMIT2(b1, b2) EMIT((b1) + ((b2) << 8), 2) +#define EMIT3(b1, b2, b3) EMIT((b1) + ((b2) << 8) + ((b3) << 16), 3) +#define EMIT4(b1, b2, b3, b4) EMIT((b1) + ((b2) << 8) + ((b3) << 16) + ((b4) << 24), 4) +#define EMIT1_off32(b1, off) do { EMIT1(b1); EMIT(off, 4);} while (0) + +#define CLEAR_A() EMIT2(0x31, 0xc0) /* xor %eax,%eax */ +#define CLEAR_X() EMIT2(0x31, 0xdb) /* xor %ebx,%ebx */ + +static inline bool is_imm8(int value) +{ + return value <= 127 && value >= -128; +} + +static inline bool is_near(int offset) +{ + return offset <= 127 && offset >= -128; +} + +#define EMIT_JMP(offset) \ +do { \ + if (offset) { \ + if (is_near(offset)) \ + EMIT2(0xeb, offset); /* jmp .+off8 */ \ + else \ + EMIT1_off32(0xe9, offset); /* jmp .+off32 */ \ + } \ +} while (0) + +/* list of x86 cond jumps opcodes (. + s8) + * Add 0x10 (and an extra 0x0f) to generate far jumps (. + s32) + */ +#define X86_JB 0x72 +#define X86_JAE 0x73 +#define X86_JE 0x74 +#define X86_JNE 0x75 +#define X86_JBE 0x76 +#define X86_JA 0x77 + +#define EMIT_COND_JMP(op, offset) \ +do { \ + if (is_near(offset)) \ + EMIT2(op, offset); /* jxx .+off8 */ \ + else { \ + EMIT2(0x0f, op + 0x10); \ + EMIT(offset, 4); /* jxx .+off32 */ \ + } \ +} while (0) + +#define COND_SEL(CODE, TOP, FOP) \ + case CODE: \ + t_op = TOP; \ + f_op = FOP; \ + goto cond_branch + + +#define SEEN_DATAREF 1 /* might call external helpers */ +#define SEEN_XREG 2 /* ebx is used */ +#define SEEN_MEM 4 /* use mem[] for temporary storage */ + +static inline void bpf_flush_icache(void *start, void *end) +{ + mm_segment_t old_fs = get_fs(); + + set_fs(KERNEL_DS); + smp_wmb(); + flush_icache_range((unsigned long)start, (unsigned long)end); + set_fs(old_fs); +} + + +void bpf_jit_compile(struct sk_filter *fp) +{ + u8 temp[64]; + u8 *prog; + unsigned int proglen, oldproglen = 0; + int ilen, i; + int t_offset, f_offset; + u8 t_op, f_op, seen = 0, pass; + u8 *image = NULL; + u8 *func; + int pc_ret0 = -1; /* bpf index of first RET #0 instruction (if any) */ + unsigned int cleanup_addr; /* epilogue code offset */ + unsigned int *addrs; + const struct sock_filter *filter = fp->insns; + int flen = fp->len; + + if (!bpf_jit_enable) + return; + + addrs = kmalloc(flen * sizeof(*addrs), GFP_KERNEL); + if (addrs == NULL) + return; + + /* Before first pass, make a rough estimation of addrs[] + * each bpf instruction is translated to less than 64 bytes + */ + for (proglen = 0, i = 0; i < flen; i++) { + proglen += 64; + addrs[i] = proglen; + } + cleanup_addr = proglen; /* epilogue address */ + + for (pass = 0; pass < 10; pass++) { + /* no prologue/epilogue for trivial filters (RET something) */ + proglen = 0; + prog = temp; + + if (seen) { + EMIT4(0x55, 0x48, 0x89, 0xe5); /* push %rbp; mov %rsp,%rbp */ + EMIT4(0x48, 0x83, 0xec, 96); /* subq $96,%rsp */ + /* note : must save %rbx in case bpf_error is hit */ + if (seen & (SEEN_XREG | SEEN_DATAREF)) + EMIT4(0x48, 0x89, 0x5d, 0xf8); /* mov %rbx, -8(%rbp) */ + if (seen & SEEN_XREG) + CLEAR_X(); /* make sure we dont leek kernel memory */ + + /* + * If this filter needs to access skb data, + * loads r9 and r8 with : + * r9 = skb->len - skb->data_len + * r8 = skb->data + */ + if (seen & SEEN_DATAREF) { + if (offsetof(struct sk_buff, len) <= 127) + /* mov off8(%rdi),%r9d */ + EMIT4(0x44, 0x8b, 0x4f, offsetof(struct sk_buff, len)); + else { + /* mov off32(%rdi),%r9d */ + EMIT3(0x44, 0x8b, 0x8f); + EMIT(offsetof(struct sk_buff, len), 4); + } + if (is_imm8(offsetof(struct sk_buff, data_len))) + /* sub off8(%rdi),%r9d */ + EMIT4(0x44, 0x2b, 0x4f, offsetof(struct sk_buff, data_len)); + else { + EMIT3(0x44, 0x2b, 0x8f); + EMIT(offsetof(struct sk_buff, data_len), 4); + } + + if (is_imm8(offsetof(struct sk_buff, data))) + /* mov off8(%rdi),%r8 */ + EMIT4(0x4c, 0x8b, 0x47, offsetof(struct sk_buff, data)); + else { + /* mov off32(%rdi),%r8 */ + EMIT3(0x4c, 0x8b, 0x87); + EMIT(offsetof(struct sk_buff, data), 4); + } + } + } + + switch (filter[0].code) { + case BPF_S_RET_K: + case BPF_S_LD_W_LEN: + case BPF_S_ANC_PROTOCOL: + case BPF_S_ANC_IFINDEX: + case BPF_S_ANC_MARK: + case BPF_S_ANC_RXHASH: + case BPF_S_ANC_CPU: + case BPF_S_ANC_QUEUE: + case BPF_S_LD_W_ABS: + case BPF_S_LD_H_ABS: + case BPF_S_LD_B_ABS: + /* first instruction sets A register (or is RET 'constant') */ + break; + default: + /* make sure we dont leak kernel information to user */ + CLEAR_A(); /* A = 0 */ + } + + for (i = 0; i < flen; i++) { + unsigned int K = filter[i].k; + + switch (filter[i].code) { + case BPF_S_ALU_ADD_X: /* A += X; */ + seen |= SEEN_XREG; + EMIT2(0x01, 0xd8); /* add %ebx,%eax */ + break; + case BPF_S_ALU_ADD_K: /* A += K; */ + if (!K) + break; + if (is_imm8(K)) + EMIT3(0x83, 0xc0, K); /* add imm8,%eax */ + else + EMIT1_off32(0x05, K); /* add imm32,%eax */ + break; + case BPF_S_ALU_SUB_X: /* A -= X; */ + seen |= SEEN_XREG; + EMIT2(0x29, 0xd8); /* sub %ebx,%eax */ + break; + case BPF_S_ALU_SUB_K: /* A -= K */ + if (!K) + break; + if (is_imm8(K)) + EMIT3(0x83, 0xe8, K); /* sub imm8,%eax */ + else + EMIT1_off32(0x2d, K); /* sub imm32,%eax */ + break; + case BPF_S_ALU_MUL_X: /* A *= X; */ + seen |= SEEN_XREG; + EMIT3(0x0f, 0xaf, 0xc3); /* imul %ebx,%eax */ + break; + case BPF_S_ALU_MUL_K: /* A *= K */ + if (is_imm8(K)) + EMIT3(0x6b, 0xc0, K); /* imul imm8,%eax,%eax */ + else { + EMIT2(0x69, 0xc0); /* imul imm32,%eax */ + EMIT(K, 4); + } + break; + case BPF_S_ALU_DIV_X: /* A /= X; */ + seen |= SEEN_XREG; + EMIT2(0x85, 0xdb); /* test %ebx,%ebx */ + if (pc_ret0 != -1) + EMIT_COND_JMP(X86_JE, addrs[pc_ret0] - (addrs[i] - 4)); + else { + EMIT_COND_JMP(X86_JNE, 2 + 5); + CLEAR_A(); + EMIT1_off32(0xe9, cleanup_addr - (addrs[i] - 4)); /* jmp .+off32 */ + } + EMIT4(0x31, 0xd2, 0xf7, 0xf3); /* xor %edx,%edx; div %ebx */ + break; + case BPF_S_ALU_DIV_K: /* A = reciprocal_divide(A, K); */ + EMIT3(0x48, 0x69, 0xc0); /* imul imm32,%rax,%rax */ + EMIT(K, 4); + EMIT4(0x48, 0xc1, 0xe8, 0x20); /* shr $0x20,%rax */ + break; + case BPF_S_ALU_AND_X: + seen |= SEEN_XREG; + EMIT2(0x21, 0xd8); /* and %ebx,%eax */ + break; + case BPF_S_ALU_AND_K: + if (K >= 0xFFFFFF00) { + EMIT2(0x24, K & 0xFF); /* and imm8,%al */ + } else if (K >= 0xFFFF0000) { + EMIT2(0x66, 0x25); /* and imm16,%ax */ + EMIT2(K, 2); + } else { + EMIT1_off32(0x25, K); /* and imm32,%eax */ + } + break; + case BPF_S_ALU_OR_X: + seen |= SEEN_XREG; + EMIT2(0x09, 0xd8); /* or %ebx,%eax */ + break; + case BPF_S_ALU_OR_K: + if (is_imm8(K)) + EMIT3(0x83, 0xc8, K); /* or imm8,%eax */ + else + EMIT1_off32(0x0d, K); /* or imm32,%eax */ + break; + case BPF_S_ALU_LSH_X: /* A <<= X; */ + seen |= SEEN_XREG; + EMIT4(0x89, 0xd9, 0xd3, 0xe0); /* mov %ebx,%ecx; shl %cl,%eax */ + break; + case BPF_S_ALU_LSH_K: + if (K == 0) + break; + else if (K == 1) + EMIT2(0xd1, 0xe0); /* shl %eax */ + else + EMIT3(0xc1, 0xe0, K); + break; + case BPF_S_ALU_RSH_X: /* A >>= X; */ + seen |= SEEN_XREG; + EMIT4(0x89, 0xd9, 0xd3, 0xe8); /* mov %ebx,%ecx; shr %cl,%eax */ + break; + case BPF_S_ALU_RSH_K: /* A >>= K; */ + if (K == 0) + break; + else if (K == 1) + EMIT2(0xd1, 0xe8); /* shr %eax */ + else + EMIT3(0xc1, 0xe8, K); + break; + case BPF_S_ALU_NEG: + EMIT2(0xf7, 0xd8); /* neg %eax */ + break; + case BPF_S_RET_K: + if (!K) { + if (pc_ret0 == -1) + pc_ret0 = i; + CLEAR_A(); + } else { + EMIT1_off32(0xb8, K); /* mov $imm32,%eax */ + } + /* fallinto */ + case BPF_S_RET_A: + if (seen) { + if (i != flen - 1) { + EMIT_JMP(cleanup_addr - addrs[i]); + break; + } + if (seen & SEEN_XREG) + EMIT4(0x48, 0x8b, 0x5d, 0xf8); /* mov -8(%rbp),%rbx */ + EMIT1(0xc9); /* leaveq */ + } + EMIT1(0xc3); /* ret */ + break; + case BPF_S_MISC_TAX: /* X = A */ + seen |= SEEN_XREG; + EMIT2(0x89, 0xc3); /* mov %eax,%ebx */ + break; + case BPF_S_MISC_TXA: /* A = X */ + seen |= SEEN_XREG; + EMIT2(0x89, 0xd8); /* mov %ebx,%eax */ + break; + case BPF_S_LD_IMM: /* A = K */ + if (!K) + CLEAR_A(); + else + EMIT1_off32(0xb8, K); /* mov $imm32,%eax */ + break; + case BPF_S_LDX_IMM: /* X = K */ + seen |= SEEN_XREG; + if (!K) + CLEAR_X(); + else + EMIT1_off32(0xbb, K); /* mov $imm32,%ebx */ + break; + case BPF_S_LD_MEM: /* A = mem[K] : mov off8(%rbp),%eax */ + seen |= SEEN_MEM; + EMIT3(0x8b, 0x45, 0xf0 - K*4); + break; + case BPF_S_LDX_MEM: /* X = mem[K] : mov off8(%rbp),%ebx */ + seen |= SEEN_XREG | SEEN_MEM; + EMIT3(0x8b, 0x5d, 0xf0 - K*4); + break; + case BPF_S_ST: /* mem[K] = A : mov %eax,off8(%rbp) */ + seen |= SEEN_MEM; + EMIT3(0x89, 0x45, 0xf0 - K*4); + break; + case BPF_S_STX: /* mem[K] = X : mov %ebx,off8(%rbp) */ + seen |= SEEN_XREG | SEEN_MEM; + EMIT3(0x89, 0x5d, 0xf0 - K*4); + break; + case BPF_S_LD_W_LEN: /* A = skb->len; */ + BUILD_BUG_ON(FIELD_SIZEOF(struct sk_buff, len) != 4); + if (is_imm8(offsetof(struct sk_buff, len))) + /* mov off8(%rdi),%eax */ + EMIT3(0x8b, 0x47, offsetof(struct sk_buff, len)); + else { + EMIT2(0x8b, 0x87); + EMIT(offsetof(struct sk_buff, len), 4); + } + break; + case BPF_S_LDX_W_LEN: /* X = skb->len; */ + seen |= SEEN_XREG; + if (is_imm8(offsetof(struct sk_buff, len))) + /* mov off8(%rdi),%ebx */ + EMIT3(0x8b, 0x5f, offsetof(struct sk_buff, len)); + else { + EMIT2(0x8b, 0x9f); + EMIT(offsetof(struct sk_buff, len), 4); + } + break; + case BPF_S_ANC_PROTOCOL: /* A = ntohs(skb->protocol); */ + BUILD_BUG_ON(FIELD_SIZEOF(struct sk_buff, protocol) != 2); + if (is_imm8(offsetof(struct sk_buff, protocol))) { + /* movzwl off8(%rdi),%eax */ + EMIT4(0x0f, 0xb7, 0x47, offsetof(struct sk_buff, protocol)); + } else { + EMIT3(0x0f, 0xb7, 0x87); /* movzwl off32(%rdi),%eax */ + EMIT(offsetof(struct sk_buff, protocol), 4); + } + EMIT2(0x86, 0xc4); /* ntohs() : xchg %al,%ah */ + break; + case BPF_S_ANC_IFINDEX: + if (is_imm8(offsetof(struct sk_buff, dev))) { + /* movq off8(%rdi),%rax */ + EMIT4(0x48, 0x8b, 0x47, offsetof(struct sk_buff, dev)); + } else { + EMIT3(0x48, 0x8b, 0x87); /* movq off32(%rdi),%rax */ + EMIT(offsetof(struct sk_buff, dev), 4); + } + EMIT3(0x48, 0x85, 0xc0); /* test %rax,%rax */ + EMIT_COND_JMP(X86_JE, cleanup_addr - (addrs[i] - 6)); + BUILD_BUG_ON(FIELD_SIZEOF(struct net_device, ifindex) != 4); + EMIT2(0x8b, 0x80); /* mov off32(%rax),%eax */ + EMIT(offsetof(struct net_device, ifindex), 4); + break; + case BPF_S_ANC_MARK: + BUILD_BUG_ON(FIELD_SIZEOF(struct sk_buff, mark) != 4); + if (is_imm8(offsetof(struct sk_buff, mark))) { + /* mov off8(%rdi),%eax */ + EMIT3(0x8b, 0x47, offsetof(struct sk_buff, mark)); + } else { + EMIT2(0x8b, 0x87); + EMIT(offsetof(struct sk_buff, mark), 4); + } + break; + case BPF_S_ANC_RXHASH: + BUILD_BUG_ON(FIELD_SIZEOF(struct sk_buff, rxhash) != 4); + if (is_imm8(offsetof(struct sk_buff, rxhash))) { + /* mov off8(%rdi),%eax */ + EMIT3(0x8b, 0x47, offsetof(struct sk_buff, rxhash)); + } else { + EMIT2(0x8b, 0x87); + EMIT(offsetof(struct sk_buff, rxhash), 4); + } + break; + case BPF_S_ANC_QUEUE: + BUILD_BUG_ON(FIELD_SIZEOF(struct sk_buff, queue_mapping) != 2); + if (is_imm8(offsetof(struct sk_buff, queue_mapping))) { + /* movzwl off8(%rdi),%eax */ + EMIT4(0x0f, 0xb7, 0x47, offsetof(struct sk_buff, queue_mapping)); + } else { + EMIT3(0x0f, 0xb7, 0x87); /* movzwl off32(%rdi),%eax */ + EMIT(offsetof(struct sk_buff, queue_mapping), 4); + } + break; + case BPF_S_ANC_CPU: +#ifdef CONFIG_SMP + EMIT4(0x65, 0x8b, 0x04, 0x25); /* mov %gs:off32,%eax */ + EMIT((u32)(unsigned long)&cpu_number, 4); /* A = smp_processor_id(); */ +#else + CLEAR_A(); +#endif + break; + case BPF_S_LD_W_ABS: + func = sk_load_word; +common_load: seen |= SEEN_DATAREF; + if ((int)K < 0) + goto out; + t_offset = func - (image + addrs[i]); + EMIT1_off32(0xbe, K); /* mov imm32,%esi */ + EMIT1_off32(0xe8, t_offset); /* call */ + break; + case BPF_S_LD_H_ABS: + func = sk_load_half; + goto common_load; + case BPF_S_LD_B_ABS: + func = sk_load_byte; + goto common_load; + case BPF_S_LDX_B_MSH: + if ((int)K < 0) { + if (pc_ret0 != -1) { + EMIT_JMP(addrs[pc_ret0] - addrs[i]); + break; + } + CLEAR_A(); + EMIT_JMP(cleanup_addr - addrs[i]); + break; + } + seen |= SEEN_DATAREF | SEEN_XREG; + t_offset = sk_load_byte_msh - (image + addrs[i]); + EMIT1_off32(0xbe, K); /* mov imm32,%esi */ + EMIT1_off32(0xe8, t_offset); /* call sk_load_byte_msh */ + break; + case BPF_S_LD_W_IND: + func = sk_load_word_ind; +common_load_ind: seen |= SEEN_DATAREF | SEEN_XREG; + t_offset = func - (image + addrs[i]); + EMIT1_off32(0xbe, K); /* mov imm32,%esi */ + EMIT1_off32(0xe8, t_offset); /* call sk_load_xxx_ind */ + break; + case BPF_S_LD_H_IND: + func = sk_load_half_ind; + goto common_load_ind; + case BPF_S_LD_B_IND: + func = sk_load_byte_ind; + goto common_load_ind; + case BPF_S_JMP_JA: + t_offset = addrs[i + K] - addrs[i]; + EMIT_JMP(t_offset); + break; + COND_SEL(BPF_S_JMP_JGT_K, X86_JA, X86_JBE); + COND_SEL(BPF_S_JMP_JGE_K, X86_JAE, X86_JB); + COND_SEL(BPF_S_JMP_JEQ_K, X86_JE, X86_JNE); + COND_SEL(BPF_S_JMP_JSET_K,X86_JNE, X86_JE); + COND_SEL(BPF_S_JMP_JGT_X, X86_JA, X86_JBE); + COND_SEL(BPF_S_JMP_JGE_X, X86_JAE, X86_JB); + COND_SEL(BPF_S_JMP_JEQ_X, X86_JE, X86_JNE); + COND_SEL(BPF_S_JMP_JSET_X,X86_JNE, X86_JE); + +cond_branch: f_offset = addrs[i + filter[i].jf] - addrs[i]; + t_offset = addrs[i + filter[i].jt] - addrs[i]; + + /* same targets, can avoid doing the test :) */ + if (filter[i].jt == filter[i].jf) { + EMIT_JMP(t_offset); + break; + } + + switch (filter[i].code) { + case BPF_S_JMP_JGT_X: + case BPF_S_JMP_JGE_X: + case BPF_S_JMP_JEQ_X: + seen |= SEEN_XREG; + EMIT2(0x39, 0xd8); /* cmp %ebx,%eax */ + break; + case BPF_S_JMP_JSET_X: + seen |= SEEN_XREG; + EMIT2(0x85, 0xd8); /* test %ebx,%eax */ + break; + case BPF_S_JMP_JEQ_K: + if (K == 0) { + EMIT2(0x85, 0xc0); /* test %eax,%eax */ + break; + } + case BPF_S_JMP_JGT_K: + case BPF_S_JMP_JGE_K: + if (K <= 127) + EMIT3(0x83, 0xf8, K); /* cmp imm8,%eax */ + else + EMIT1_off32(0x3d, K); /* cmp imm32,%eax */ + break; + case BPF_S_JMP_JSET_K: + if (K <= 0xFF) + EMIT2(0xa8, K); /* test imm8,%al */ + else if (!(K & 0xFFFF00FF)) + EMIT3(0xf6, 0xc4, K >> 8); /* test imm8,%ah */ + else if (K <= 0xFFFF) { + EMIT2(0x66, 0xa9); /* test imm16,%ax */ + EMIT(K, 2); + } else { + EMIT1_off32(0xa9, K); /* test imm32,%eax */ + } + break; + } + if (filter[i].jt != 0) { + if (filter[i].jf) + t_offset += is_near(f_offset) ? 2 : 6; + EMIT_COND_JMP(t_op, t_offset); + if (filter[i].jf) + EMIT_JMP(f_offset); + break; + } + EMIT_COND_JMP(f_op, f_offset); + break; + default: + /* hmm, too complex filter, give up with jit compiler */ + goto out; + } + ilen = prog - temp; + if (image) { + if (unlikely(proglen + ilen > oldproglen)) { + pr_err("bpb_jit_compile fatal error\n"); + kfree(addrs); + module_free(NULL, image); + return; + } + memcpy(image + proglen, temp, ilen); + } + proglen += ilen; + addrs[i] = proglen; + prog = temp; + } + /* last bpf instruction is always a RET : + * use it to give the cleanup instruction(s) addr + */ + cleanup_addr = proglen - 1; /* ret */ + if (seen) + cleanup_addr -= 1; /* leaveq */ + if (seen & SEEN_XREG) + cleanup_addr -= 4; /* mov -8(%rbp),%rbx */ + + if (image) { + WARN_ON(proglen != oldproglen); + break; + } + if (proglen == oldproglen) { + image = module_alloc(max_t(unsigned int, + proglen, + sizeof(struct work_struct))); + if (!image) + goto out; + } + oldproglen = proglen; + } + if (bpf_jit_enable > 1) + pr_err("flen=%d proglen=%u pass=%d image=%p\n", + flen, proglen, pass, image); + + if (image) { + if (bpf_jit_enable > 1) + print_hex_dump(KERN_ERR, "JIT code: ", DUMP_PREFIX_ADDRESS, + 16, 1, image, proglen, false); + + bpf_flush_icache(image, image + proglen); + + fp->bpf_func = (void *)image; + } +out: + kfree(addrs); + return; +} + +static void jit_free_defer(struct work_struct *arg) +{ + module_free(NULL, arg); +} + +/* run from softirq, we must use a work_struct to call + * module_free() from process context + */ +void bpf_jit_free(struct sk_filter *fp) +{ + if (fp->bpf_func != sk_run_filter) { + struct work_struct *work = (struct work_struct *)fp->bpf_func; + + INIT_WORK(work, jit_free_defer); + schedule_work(work); + } +} diff --git a/arch/x86/oprofile/backtrace.c b/arch/x86/oprofile/backtrace.c index 2d49d4e19a36..a5b64ab4cd6e 100644 --- a/arch/x86/oprofile/backtrace.c +++ b/arch/x86/oprofile/backtrace.c @@ -16,17 +16,6 @@ #include <asm/stacktrace.h> #include <linux/compat.h> -static void backtrace_warning_symbol(void *data, char *msg, - unsigned long symbol) -{ - /* Ignore warnings */ -} - -static void backtrace_warning(void *data, char *msg) -{ - /* Ignore warnings */ -} - static int backtrace_stack(void *data, char *name) { /* Yes, we want all stacks */ @@ -42,8 +31,6 @@ static void backtrace_address(void *data, unsigned long addr, int reliable) } static struct stacktrace_ops backtrace_ops = { - .warning = backtrace_warning, - .warning_symbol = backtrace_warning_symbol, .stack = backtrace_stack, .address = backtrace_address, .walk_stack = print_context_stack, diff --git a/arch/x86/pci/direct.c b/arch/x86/pci/direct.c index bd33620b0071..e6fd8473fb7b 100644 --- a/arch/x86/pci/direct.c +++ b/arch/x86/pci/direct.c @@ -280,12 +280,9 @@ void __init pci_direct_init(int type) int __init pci_direct_probe(void) { - struct resource *region, *region2; - if ((pci_probe & PCI_PROBE_CONF1) == 0) goto type2; - region = request_region(0xCF8, 8, "PCI conf1"); - if (!region) + if (!request_region(0xCF8, 8, "PCI conf1")) goto type2; if (pci_check_type1()) { @@ -293,16 +290,14 @@ int __init pci_direct_probe(void) port_cf9_safe = true; return 1; } - release_resource(region); + release_region(0xCF8, 8); type2: if ((pci_probe & PCI_PROBE_CONF2) == 0) return 0; - region = request_region(0xCF8, 4, "PCI conf2"); - if (!region) + if (!request_region(0xCF8, 4, "PCI conf2")) return 0; - region2 = request_region(0xC000, 0x1000, "PCI conf2"); - if (!region2) + if (!request_region(0xC000, 0x1000, "PCI conf2")) goto fail2; if (pci_check_type2()) { @@ -311,8 +306,8 @@ int __init pci_direct_probe(void) return 2; } - release_resource(region2); + release_region(0xC000, 0x1000); fail2: - release_resource(region); + release_region(0xCF8, 4); return 0; } diff --git a/arch/x86/pci/irq.c b/arch/x86/pci/irq.c index 8201165bae28..372e9b8989b3 100644 --- a/arch/x86/pci/irq.c +++ b/arch/x86/pci/irq.c @@ -602,7 +602,9 @@ static __init int intel_router_probe(struct irq_router *r, struct pci_dev *route || (device >= PCI_DEVICE_ID_INTEL_COUGARPOINT_LPC_MIN && device <= PCI_DEVICE_ID_INTEL_COUGARPOINT_LPC_MAX) || (device >= PCI_DEVICE_ID_INTEL_DH89XXCC_LPC_MIN && - device <= PCI_DEVICE_ID_INTEL_DH89XXCC_LPC_MAX)) { + device <= PCI_DEVICE_ID_INTEL_DH89XXCC_LPC_MAX) + || (device >= PCI_DEVICE_ID_INTEL_PANTHERPOINT_LPC_MIN && + device <= PCI_DEVICE_ID_INTEL_PANTHERPOINT_LPC_MAX)) { r->name = "PIIX/ICH"; r->get = pirq_piix_get; r->set = pirq_piix_set; diff --git a/arch/x86/pci/mmconfig-shared.c b/arch/x86/pci/mmconfig-shared.c index e282886616a0..750c346ef50a 100644 --- a/arch/x86/pci/mmconfig-shared.c +++ b/arch/x86/pci/mmconfig-shared.c @@ -606,6 +606,16 @@ static void __init __pci_mmcfg_init(int early) if (list_empty(&pci_mmcfg_list)) return; + if (pcibios_last_bus < 0) { + const struct pci_mmcfg_region *cfg; + + list_for_each_entry(cfg, &pci_mmcfg_list, list) { + if (cfg->segment) + break; + pcibios_last_bus = cfg->end_bus; + } + } + if (pci_mmcfg_arch_init()) pci_probe = (pci_probe & ~PCI_PROBE_MASK) | PCI_PROBE_MMCONF; else { diff --git a/arch/x86/pci/xen.c b/arch/x86/pci/xen.c index e37b407a0ee8..8214724ce54d 100644 --- a/arch/x86/pci/xen.c +++ b/arch/x86/pci/xen.c @@ -108,7 +108,8 @@ static int xen_hvm_setup_msi_irqs(struct pci_dev *dev, int nvec, int type) } irq = xen_bind_pirq_msi_to_irq(dev, msidesc, pirq, 0, (type == PCI_CAP_ID_MSIX) ? - "msi-x" : "msi"); + "msi-x" : "msi", + DOMID_SELF); if (irq < 0) goto error; dev_dbg(&dev->dev, @@ -148,7 +149,8 @@ static int xen_setup_msi_irqs(struct pci_dev *dev, int nvec, int type) irq = xen_bind_pirq_msi_to_irq(dev, msidesc, v[i], 0, (type == PCI_CAP_ID_MSIX) ? "pcifront-msi-x" : - "pcifront-msi"); + "pcifront-msi", + DOMID_SELF); if (irq < 0) goto free; i++; @@ -190,9 +192,16 @@ static int xen_initdom_setup_msi_irqs(struct pci_dev *dev, int nvec, int type) list_for_each_entry(msidesc, &dev->msi_list, list) { struct physdev_map_pirq map_irq; + domid_t domid; + + domid = ret = xen_find_device_domain_owner(dev); + /* N.B. Casting int's -ENODEV to uint16_t results in 0xFFED, + * hence check ret value for < 0. */ + if (ret < 0) + domid = DOMID_SELF; memset(&map_irq, 0, sizeof(map_irq)); - map_irq.domid = DOMID_SELF; + map_irq.domid = domid; map_irq.type = MAP_PIRQ_TYPE_MSI; map_irq.index = -1; map_irq.pirq = -1; @@ -215,14 +224,16 @@ static int xen_initdom_setup_msi_irqs(struct pci_dev *dev, int nvec, int type) ret = HYPERVISOR_physdev_op(PHYSDEVOP_map_pirq, &map_irq); if (ret) { - dev_warn(&dev->dev, "xen map irq failed %d\n", ret); + dev_warn(&dev->dev, "xen map irq failed %d for %d domain\n", + ret, domid); goto out; } ret = xen_bind_pirq_msi_to_irq(dev, msidesc, map_irq.pirq, map_irq.index, (type == PCI_CAP_ID_MSIX) ? - "msi-x" : "msi"); + "msi-x" : "msi", + domid); if (ret < 0) goto out; } @@ -461,3 +472,78 @@ void __init xen_setup_pirqs(void) } } #endif + +#ifdef CONFIG_XEN_DOM0 +struct xen_device_domain_owner { + domid_t domain; + struct pci_dev *dev; + struct list_head list; +}; + +static DEFINE_SPINLOCK(dev_domain_list_spinlock); +static struct list_head dev_domain_list = LIST_HEAD_INIT(dev_domain_list); + +static struct xen_device_domain_owner *find_device(struct pci_dev *dev) +{ + struct xen_device_domain_owner *owner; + + list_for_each_entry(owner, &dev_domain_list, list) { + if (owner->dev == dev) + return owner; + } + return NULL; +} + +int xen_find_device_domain_owner(struct pci_dev *dev) +{ + struct xen_device_domain_owner *owner; + int domain = -ENODEV; + + spin_lock(&dev_domain_list_spinlock); + owner = find_device(dev); + if (owner) + domain = owner->domain; + spin_unlock(&dev_domain_list_spinlock); + return domain; +} +EXPORT_SYMBOL_GPL(xen_find_device_domain_owner); + +int xen_register_device_domain_owner(struct pci_dev *dev, uint16_t domain) +{ + struct xen_device_domain_owner *owner; + + owner = kzalloc(sizeof(struct xen_device_domain_owner), GFP_KERNEL); + if (!owner) + return -ENODEV; + + spin_lock(&dev_domain_list_spinlock); + if (find_device(dev)) { + spin_unlock(&dev_domain_list_spinlock); + kfree(owner); + return -EEXIST; + } + owner->domain = domain; + owner->dev = dev; + list_add_tail(&owner->list, &dev_domain_list); + spin_unlock(&dev_domain_list_spinlock); + return 0; +} +EXPORT_SYMBOL_GPL(xen_register_device_domain_owner); + +int xen_unregister_device_domain_owner(struct pci_dev *dev) +{ + struct xen_device_domain_owner *owner; + + spin_lock(&dev_domain_list_spinlock); + owner = find_device(dev); + if (!owner) { + spin_unlock(&dev_domain_list_spinlock); + return -ENODEV; + } + list_del(&owner->list); + spin_unlock(&dev_domain_list_spinlock); + kfree(owner); + return 0; +} +EXPORT_SYMBOL_GPL(xen_unregister_device_domain_owner); +#endif diff --git a/arch/x86/platform/ce4100/falconfalls.dts b/arch/x86/platform/ce4100/falconfalls.dts index 2d6d226f2b10..e70be38ce039 100644 --- a/arch/x86/platform/ce4100/falconfalls.dts +++ b/arch/x86/platform/ce4100/falconfalls.dts @@ -347,7 +347,7 @@ "pciclass0c03"; reg = <0x16800 0x0 0x0 0x0 0x0>; - interrupts = <22 3>; + interrupts = <22 1>; }; usb@d,1 { @@ -357,7 +357,7 @@ "pciclass0c03"; reg = <0x16900 0x0 0x0 0x0 0x0>; - interrupts = <22 3>; + interrupts = <22 1>; }; sata@e,0 { @@ -367,7 +367,7 @@ "pciclass0106"; reg = <0x17000 0x0 0x0 0x0 0x0>; - interrupts = <23 3>; + interrupts = <23 1>; }; flash@f,0 { diff --git a/arch/x86/platform/efi/efi.c b/arch/x86/platform/efi/efi.c index 0fe27d7c6258..b30aa26a8df2 100644 --- a/arch/x86/platform/efi/efi.c +++ b/arch/x86/platform/efi/efi.c @@ -145,17 +145,6 @@ static void virt_efi_reset_system(int reset_type, data_size, data); } -static efi_status_t virt_efi_set_virtual_address_map( - unsigned long memory_map_size, - unsigned long descriptor_size, - u32 descriptor_version, - efi_memory_desc_t *virtual_map) -{ - return efi_call_virt4(set_virtual_address_map, - memory_map_size, descriptor_size, - descriptor_version, virtual_map); -} - static efi_status_t __init phys_efi_set_virtual_address_map( unsigned long memory_map_size, unsigned long descriptor_size, @@ -468,11 +457,25 @@ void __init efi_init(void) #endif } +void __init efi_set_executable(efi_memory_desc_t *md, bool executable) +{ + u64 addr, npages; + + addr = md->virt_addr; + npages = md->num_pages; + + memrange_efi_to_native(&addr, &npages); + + if (executable) + set_memory_x(addr, npages); + else + set_memory_nx(addr, npages); +} + static void __init runtime_code_page_mkexec(void) { efi_memory_desc_t *md; void *p; - u64 addr, npages; /* Make EFI runtime service code area executable */ for (p = memmap.map; p < memmap.map_end; p += memmap.desc_size) { @@ -481,10 +484,7 @@ static void __init runtime_code_page_mkexec(void) if (md->type != EFI_RUNTIME_SERVICES_CODE) continue; - addr = md->virt_addr; - npages = md->num_pages; - memrange_efi_to_native(&addr, &npages); - set_memory_x(addr, npages); + efi_set_executable(md, true); } } @@ -498,13 +498,42 @@ static void __init runtime_code_page_mkexec(void) */ void __init efi_enter_virtual_mode(void) { - efi_memory_desc_t *md; + efi_memory_desc_t *md, *prev_md = NULL; efi_status_t status; unsigned long size; u64 end, systab, addr, npages, end_pfn; - void *p, *va; + void *p, *va, *new_memmap = NULL; + int count = 0; efi.systab = NULL; + + /* Merge contiguous regions of the same type and attribute */ + for (p = memmap.map; p < memmap.map_end; p += memmap.desc_size) { + u64 prev_size; + md = p; + + if (!prev_md) { + prev_md = md; + continue; + } + + if (prev_md->type != md->type || + prev_md->attribute != md->attribute) { + prev_md = md; + continue; + } + + prev_size = prev_md->num_pages << EFI_PAGE_SHIFT; + + if (md->phys_addr == (prev_md->phys_addr + prev_size)) { + prev_md->num_pages += md->num_pages; + md->type = EFI_RESERVED_TYPE; + md->attribute = 0; + continue; + } + prev_md = md; + } + for (p = memmap.map; p < memmap.map_end; p += memmap.desc_size) { md = p; if (!(md->attribute & EFI_MEMORY_RUNTIME)) @@ -541,15 +570,21 @@ void __init efi_enter_virtual_mode(void) systab += md->virt_addr - md->phys_addr; efi.systab = (efi_system_table_t *) (unsigned long) systab; } + new_memmap = krealloc(new_memmap, + (count + 1) * memmap.desc_size, + GFP_KERNEL); + memcpy(new_memmap + (count * memmap.desc_size), md, + memmap.desc_size); + count++; } BUG_ON(!efi.systab); status = phys_efi_set_virtual_address_map( - memmap.desc_size * memmap.nr_map, + memmap.desc_size * count, memmap.desc_size, memmap.desc_version, - memmap.phys_map); + (efi_memory_desc_t *)__pa(new_memmap)); if (status != EFI_SUCCESS) { printk(KERN_ALERT "Unable to switch EFI into virtual mode " @@ -572,11 +607,12 @@ void __init efi_enter_virtual_mode(void) efi.set_variable = virt_efi_set_variable; efi.get_next_high_mono_count = virt_efi_get_next_high_mono_count; efi.reset_system = virt_efi_reset_system; - efi.set_virtual_address_map = virt_efi_set_virtual_address_map; + efi.set_virtual_address_map = NULL; if (__supported_pte_mask & _PAGE_NX) runtime_code_page_mkexec(); early_iounmap(memmap.map, memmap.nr_map * memmap.desc_size); memmap.map = NULL; + kfree(new_memmap); } /* diff --git a/arch/x86/platform/efi/efi_64.c b/arch/x86/platform/efi/efi_64.c index ac0621a7ac3d..2649426a7905 100644 --- a/arch/x86/platform/efi/efi_64.c +++ b/arch/x86/platform/efi/efi_64.c @@ -41,22 +41,7 @@ static pgd_t save_pgd __initdata; static unsigned long efi_flags __initdata; -static void __init early_mapping_set_exec(unsigned long start, - unsigned long end, - int executable) -{ - unsigned long num_pages; - - start &= PMD_MASK; - end = (end + PMD_SIZE - 1) & PMD_MASK; - num_pages = (end - start) >> PAGE_SHIFT; - if (executable) - set_memory_x((unsigned long)__va(start), num_pages); - else - set_memory_nx((unsigned long)__va(start), num_pages); -} - -static void __init early_runtime_code_mapping_set_exec(int executable) +static void __init early_code_mapping_set_exec(int executable) { efi_memory_desc_t *md; void *p; @@ -67,11 +52,8 @@ static void __init early_runtime_code_mapping_set_exec(int executable) /* Make EFI runtime service code area executable */ for (p = memmap.map; p < memmap.map_end; p += memmap.desc_size) { md = p; - if (md->type == EFI_RUNTIME_SERVICES_CODE) { - unsigned long end; - end = md->phys_addr + (md->num_pages << EFI_PAGE_SHIFT); - early_mapping_set_exec(md->phys_addr, end, executable); - } + if (md->type == EFI_RUNTIME_SERVICES_CODE) + efi_set_executable(md, executable); } } @@ -79,7 +61,7 @@ void __init efi_call_phys_prelog(void) { unsigned long vaddress; - early_runtime_code_mapping_set_exec(1); + early_code_mapping_set_exec(1); local_irq_save(efi_flags); vaddress = (unsigned long)__va(0x0UL); save_pgd = *pgd_offset_k(0x0UL); @@ -95,7 +77,7 @@ void __init efi_call_phys_epilog(void) set_pgd(pgd_offset_k(0x0UL), save_pgd); __flush_tlb_all(); local_irq_restore(efi_flags); - early_runtime_code_mapping_set_exec(0); + early_code_mapping_set_exec(0); } void __iomem *__init efi_ioremap(unsigned long phys_addr, unsigned long size, @@ -107,8 +89,10 @@ void __iomem *__init efi_ioremap(unsigned long phys_addr, unsigned long size, 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; + if ((last_map_pfn << PAGE_SHIFT) < phys_addr + size) { + unsigned long top = last_map_pfn << PAGE_SHIFT; + efi_ioremap(top, size - (top - phys_addr), type); + } return (void __iomem *)__va(phys_addr); } diff --git a/arch/x86/platform/mrst/mrst.c b/arch/x86/platform/mrst/mrst.c index 275dbc19e2cf..7000e74b3087 100644 --- a/arch/x86/platform/mrst/mrst.c +++ b/arch/x86/platform/mrst/mrst.c @@ -194,7 +194,7 @@ static unsigned long __init mrst_calibrate_tsc(void) return 0; } -void __init mrst_time_init(void) +static void __init mrst_time_init(void) { sfi_table_parse(SFI_SIG_MTMR, NULL, NULL, sfi_parse_mtmr); switch (mrst_timer_options) { @@ -216,7 +216,7 @@ void __init mrst_time_init(void) apbt_time_init(); } -void __cpuinit mrst_arch_setup(void) +static void __cpuinit mrst_arch_setup(void) { if (boot_cpu_data.x86 == 6 && boot_cpu_data.x86_model == 0x27) __mrst_cpu_chip = MRST_CPU_CHIP_PENWELL; diff --git a/arch/x86/platform/olpc/Makefile b/arch/x86/platform/olpc/Makefile index c2a8cab65e5d..81c5e2165c24 100644 --- a/arch/x86/platform/olpc/Makefile +++ b/arch/x86/platform/olpc/Makefile @@ -1,4 +1,2 @@ -obj-$(CONFIG_OLPC) += olpc.o +obj-$(CONFIG_OLPC) += olpc.o olpc_ofw.o olpc_dt.o obj-$(CONFIG_OLPC_XO1) += olpc-xo1.o -obj-$(CONFIG_OLPC) += olpc_ofw.o -obj-$(CONFIG_OF_PROMTREE) += olpc_dt.o diff --git a/arch/x86/platform/olpc/olpc.c b/arch/x86/platform/olpc/olpc.c index edaf3fe8dc5e..0060fd59ea00 100644 --- a/arch/x86/platform/olpc/olpc.c +++ b/arch/x86/platform/olpc/olpc.c @@ -18,6 +18,7 @@ #include <linux/io.h> #include <linux/string.h> #include <linux/platform_device.h> +#include <linux/of.h> #include <asm/geode.h> #include <asm/setup.h> @@ -187,41 +188,43 @@ err: } EXPORT_SYMBOL_GPL(olpc_ec_cmd); -static bool __init check_ofw_architecture(void) +static bool __init check_ofw_architecture(struct device_node *root) { - size_t propsize; - char olpc_arch[5]; - const void *args[] = { NULL, "architecture", olpc_arch, (void *)5 }; - void *res[] = { &propsize }; + const char *olpc_arch; + int propsize; - if (olpc_ofw("getprop", args, res)) { - printk(KERN_ERR "ofw: getprop call failed!\n"); - return false; - } + olpc_arch = of_get_property(root, "architecture", &propsize); return propsize == 5 && strncmp("OLPC", olpc_arch, 5) == 0; } -static u32 __init get_board_revision(void) +static u32 __init get_board_revision(struct device_node *root) { - size_t propsize; - __be32 rev; - const void *args[] = { NULL, "board-revision-int", &rev, (void *)4 }; - void *res[] = { &propsize }; - - if (olpc_ofw("getprop", args, res) || propsize != 4) { - printk(KERN_ERR "ofw: getprop call failed!\n"); - return cpu_to_be32(0); - } - return be32_to_cpu(rev); + int propsize; + const __be32 *rev; + + rev = of_get_property(root, "board-revision-int", &propsize); + if (propsize != 4) + return 0; + + return be32_to_cpu(*rev); } static bool __init platform_detect(void) { - if (!check_ofw_architecture()) + struct device_node *root = of_find_node_by_path("/"); + bool success; + + if (!root) return false; - olpc_platform_info.flags |= OLPC_F_PRESENT; - olpc_platform_info.boardrev = get_board_revision(); - return true; + + success = check_ofw_architecture(root); + if (success) { + olpc_platform_info.boardrev = get_board_revision(root); + olpc_platform_info.flags |= OLPC_F_PRESENT; + } + + of_node_put(root); + return success; } static int __init add_xo1_platform_devices(void) diff --git a/arch/x86/platform/olpc/olpc_dt.c b/arch/x86/platform/olpc/olpc_dt.c index 044bda5b3174..d39f63d017d2 100644 --- a/arch/x86/platform/olpc/olpc_dt.c +++ b/arch/x86/platform/olpc/olpc_dt.c @@ -19,7 +19,9 @@ #include <linux/kernel.h> #include <linux/bootmem.h> #include <linux/of.h> +#include <linux/of_platform.h> #include <linux/of_pdt.h> +#include <asm/olpc.h> #include <asm/olpc_ofw.h> static phandle __init olpc_dt_getsibling(phandle node) @@ -180,3 +182,20 @@ void __init olpc_dt_build_devicetree(void) pr_info("PROM DT: Built device tree with %u bytes of memory.\n", prom_early_allocated); } + +/* A list of DT node/bus matches that we want to expose as platform devices */ +static struct of_device_id __initdata of_ids[] = { + { .compatible = "olpc,xo1-battery" }, + { .compatible = "olpc,xo1-dcon" }, + { .compatible = "olpc,xo1-rtc" }, + {}, +}; + +static int __init olpc_create_platform_devices(void) +{ + if (machine_is_olpc()) + return of_platform_bus_probe(NULL, of_ids, NULL); + else + return 0; +} +device_initcall(olpc_create_platform_devices); diff --git a/arch/x86/platform/uv/tlb_uv.c b/arch/x86/platform/uv/tlb_uv.c index 7cb6424317f6..c58e0ea39ef5 100644 --- a/arch/x86/platform/uv/tlb_uv.c +++ b/arch/x86/platform/uv/tlb_uv.c @@ -699,16 +699,17 @@ const struct cpumask *uv_flush_tlb_others(const struct cpumask *cpumask, struct mm_struct *mm, unsigned long va, unsigned int cpu) { - int tcpu; - int uvhub; int locals = 0; int remotes = 0; int hubs = 0; + int tcpu; + int tpnode; struct bau_desc *bau_desc; struct cpumask *flush_mask; struct ptc_stats *stat; struct bau_control *bcp; struct bau_control *tbcp; + struct hub_and_pnode *hpp; /* kernel was booted 'nobau' */ if (nobau) @@ -750,11 +751,18 @@ const struct cpumask *uv_flush_tlb_others(const struct cpumask *cpumask, bau_desc += UV_ITEMS_PER_DESCRIPTOR * bcp->uvhub_cpu; bau_uvhubs_clear(&bau_desc->distribution, UV_DISTRIBUTION_SIZE); - /* cpu statistics */ for_each_cpu(tcpu, flush_mask) { - uvhub = uv_cpu_to_blade_id(tcpu); - bau_uvhub_set(uvhub, &bau_desc->distribution); - if (uvhub == bcp->uvhub) + /* + * The distribution vector is a bit map of pnodes, relative + * to the partition base pnode (and the partition base nasid + * in the header). + * Translate cpu to pnode and hub using an array stored + * in local memory. + */ + hpp = &bcp->socket_master->target_hub_and_pnode[tcpu]; + tpnode = hpp->pnode - bcp->partition_base_pnode; + bau_uvhub_set(tpnode, &bau_desc->distribution); + if (hpp->uvhub == bcp->uvhub) locals++; else remotes++; @@ -855,7 +863,7 @@ void uv_bau_message_interrupt(struct pt_regs *regs) * an interrupt, but causes an error message to be returned to * the sender. */ -static void uv_enable_timeouts(void) +static void __init uv_enable_timeouts(void) { int uvhub; int nuvhubs; @@ -1326,10 +1334,10 @@ static int __init uv_ptc_init(void) } /* - * initialize the sending side's sending buffers + * Initialize the sending side's sending buffers. */ static void -uv_activation_descriptor_init(int node, int pnode) +uv_activation_descriptor_init(int node, int pnode, int base_pnode) { int i; int cpu; @@ -1352,11 +1360,11 @@ uv_activation_descriptor_init(int node, int pnode) n = pa >> uv_nshift; m = pa & uv_mmask; + /* the 14-bit pnode */ 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 + * Initializing all 8 (UV_ITEMS_PER_DESCRIPTOR) descriptors for each * cpu even though we only use the first one; one descriptor can * describe a broadcast to 256 uv hubs. */ @@ -1365,12 +1373,13 @@ uv_activation_descriptor_init(int node, int pnode) memset(bd2, 0, sizeof(struct bau_desc)); bd2->header.sw_ack_flag = 1; /* - * base_dest_nodeid is the nasid of the first uvhub - * in the partition. The bit map will indicate uvhub numbers, - * which are 0-N in a partition. Pnodes are unique system-wide. + * The base_dest_nasid set in the message header is the nasid + * of the first uvhub in the partition. The bit map will + * indicate destination pnode numbers relative to that base. + * They may not be consecutive if nasid striding is being used. */ - bd2->header.base_dest_nodeid = UV_PNODE_TO_NASID(uv_partition_base_pnode); - bd2->header.dest_subnodeid = 0x10; /* the LB */ + bd2->header.base_dest_nasid = UV_PNODE_TO_NASID(base_pnode); + bd2->header.dest_subnodeid = UV_LB_SUBNODEID; bd2->header.command = UV_NET_ENDPOINT_INTD; bd2->header.int_both = 1; /* @@ -1442,7 +1451,7 @@ uv_payload_queue_init(int node, int pnode) /* * Initialization of each UV hub's structures */ -static void __init uv_init_uvhub(int uvhub, int vector) +static void __init uv_init_uvhub(int uvhub, int vector, int base_pnode) { int node; int pnode; @@ -1450,11 +1459,11 @@ static void __init uv_init_uvhub(int uvhub, int vector) node = uvhub_to_first_node(uvhub); pnode = uv_blade_to_pnode(uvhub); - uv_activation_descriptor_init(node, pnode); + uv_activation_descriptor_init(node, pnode, base_pnode); uv_payload_queue_init(node, pnode); /* - * the below initialization can't be in firmware because the - * messaging IRQ will be determined by the OS + * The below initialization can't be in firmware because the + * messaging IRQ will be determined by the OS. */ apicid = uvhub_to_first_apicid(uvhub) | uv_apicid_hibits; uv_write_global_mmr64(pnode, UVH_BAU_DATA_CONFIG, @@ -1491,10 +1500,11 @@ calculate_destination_timeout(void) /* * initialize the bau_control structure for each cpu */ -static int __init uv_init_per_cpu(int nuvhubs) +static int __init uv_init_per_cpu(int nuvhubs, int base_part_pnode) { int i; int cpu; + int tcpu; int pnode; int uvhub; int have_hmaster; @@ -1528,6 +1538,15 @@ static int __init uv_init_per_cpu(int nuvhubs) bcp = &per_cpu(bau_control, cpu); memset(bcp, 0, sizeof(struct bau_control)); pnode = uv_cpu_hub_info(cpu)->pnode; + if ((pnode - base_part_pnode) >= UV_DISTRIBUTION_SIZE) { + printk(KERN_EMERG + "cpu %d pnode %d-%d beyond %d; BAU disabled\n", + cpu, pnode, base_part_pnode, + UV_DISTRIBUTION_SIZE); + return 1; + } + bcp->osnode = cpu_to_node(cpu); + bcp->partition_base_pnode = uv_partition_base_pnode; uvhub = uv_cpu_hub_info(cpu)->numa_blade_id; *(uvhub_mask + (uvhub/8)) |= (1 << (uvhub%8)); bdp = &uvhub_descs[uvhub]; @@ -1536,7 +1555,7 @@ static int __init uv_init_per_cpu(int nuvhubs) bdp->pnode = pnode; /* kludge: 'assuming' one node per socket, and assuming that disabling a socket just leaves a gap in node numbers */ - socket = (cpu_to_node(cpu) & 1); + socket = bcp->osnode & 1; bdp->socket_mask |= (1 << socket); sdp = &bdp->socket[socket]; sdp->cpu_number[sdp->num_cpus] = cpu; @@ -1585,6 +1604,20 @@ static int __init uv_init_per_cpu(int nuvhubs) nextsocket: socket++; socket_mask = (socket_mask >> 1); + /* each socket gets a local array of pnodes/hubs */ + bcp = smaster; + bcp->target_hub_and_pnode = kmalloc_node( + sizeof(struct hub_and_pnode) * + num_possible_cpus(), GFP_KERNEL, bcp->osnode); + memset(bcp->target_hub_and_pnode, 0, + sizeof(struct hub_and_pnode) * + num_possible_cpus()); + for_each_present_cpu(tcpu) { + bcp->target_hub_and_pnode[tcpu].pnode = + uv_cpu_hub_info(tcpu)->pnode; + bcp->target_hub_and_pnode[tcpu].uvhub = + uv_cpu_hub_info(tcpu)->numa_blade_id; + } } } kfree(uvhub_descs); @@ -1637,21 +1670,22 @@ static int __init uv_bau_init(void) spin_lock_init(&disable_lock); congested_cycles = microsec_2_cycles(congested_response_us); - if (uv_init_per_cpu(nuvhubs)) { - nobau = 1; - return 0; - } - uv_partition_base_pnode = 0x7fffffff; - for (uvhub = 0; uvhub < nuvhubs; uvhub++) + for (uvhub = 0; uvhub < nuvhubs; uvhub++) { if (uv_blade_nr_possible_cpus(uvhub) && (uv_blade_to_pnode(uvhub) < uv_partition_base_pnode)) uv_partition_base_pnode = uv_blade_to_pnode(uvhub); + } + + if (uv_init_per_cpu(nuvhubs, uv_partition_base_pnode)) { + nobau = 1; + return 0; + } vector = UV_BAU_MESSAGE; for_each_possible_blade(uvhub) if (uv_blade_nr_possible_cpus(uvhub)) - uv_init_uvhub(uvhub, vector); + uv_init_uvhub(uvhub, vector, uv_partition_base_pnode); uv_enable_timeouts(); alloc_intr_gate(vector, uv_bau_message_intr1); diff --git a/arch/x86/platform/uv/uv_time.c b/arch/x86/platform/uv/uv_time.c index 9daf5d1af9f1..0eb90184515f 100644 --- a/arch/x86/platform/uv/uv_time.c +++ b/arch/x86/platform/uv/uv_time.c @@ -40,7 +40,6 @@ static struct clocksource clocksource_uv = { .rating = 400, .read = uv_read_rtc, .mask = (cycle_t)UVH_RTC_REAL_TIME_CLOCK_MASK, - .shift = 10, .flags = CLOCK_SOURCE_IS_CONTINUOUS, }; @@ -372,14 +371,11 @@ static __init int uv_rtc_setup_clock(void) if (!is_uv_system()) return -ENODEV; - clocksource_uv.mult = clocksource_hz2mult(sn_rtc_cycles_per_second, - clocksource_uv.shift); - /* If single blade, prefer tsc */ if (uv_num_possible_blades() == 1) clocksource_uv.rating = 250; - rc = clocksource_register(&clocksource_uv); + rc = clocksource_register_hz(&clocksource_uv, sn_rtc_cycles_per_second); if (rc) printk(KERN_INFO "UV RTC clocksource failed rc %d\n", rc); else diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c index e3c6a06cf725..dd7b88f2ec7a 100644 --- a/arch/x86/xen/enlighten.c +++ b/arch/x86/xen/enlighten.c @@ -235,7 +235,7 @@ static void xen_cpuid(unsigned int *ax, unsigned int *bx, *dx &= maskedx; } -static __init void xen_init_cpuid_mask(void) +static void __init xen_init_cpuid_mask(void) { unsigned int ax, bx, cx, dx; unsigned int xsave_mask; @@ -400,7 +400,7 @@ static void xen_load_gdt(const struct desc_ptr *dtr) /* * load_gdt for early boot, when the gdt is only mapped once */ -static __init void xen_load_gdt_boot(const struct desc_ptr *dtr) +static void __init xen_load_gdt_boot(const struct desc_ptr *dtr) { unsigned long va = dtr->address; unsigned int size = dtr->size + 1; @@ -662,7 +662,7 @@ static void xen_write_gdt_entry(struct desc_struct *dt, int entry, * Version of write_gdt_entry for use at early boot-time needed to * update an entry as simply as possible. */ -static __init void xen_write_gdt_entry_boot(struct desc_struct *dt, int entry, +static void __init xen_write_gdt_entry_boot(struct desc_struct *dt, int entry, const void *desc, int type) { switch (type) { @@ -933,18 +933,18 @@ static unsigned xen_patch(u8 type, u16 clobbers, void *insnbuf, return ret; } -static const struct pv_info xen_info __initdata = { +static const struct pv_info xen_info __initconst = { .paravirt_enabled = 1, .shared_kernel_pmd = 0, .name = "Xen", }; -static const struct pv_init_ops xen_init_ops __initdata = { +static const struct pv_init_ops xen_init_ops __initconst = { .patch = xen_patch, }; -static const struct pv_cpu_ops xen_cpu_ops __initdata = { +static const struct pv_cpu_ops xen_cpu_ops __initconst = { .cpuid = xen_cpuid, .set_debugreg = xen_set_debugreg, @@ -1004,7 +1004,7 @@ static const struct pv_cpu_ops xen_cpu_ops __initdata = { .end_context_switch = xen_end_context_switch, }; -static const struct pv_apic_ops xen_apic_ops __initdata = { +static const struct pv_apic_ops xen_apic_ops __initconst = { #ifdef CONFIG_X86_LOCAL_APIC .startup_ipi_hook = paravirt_nop, #endif @@ -1055,7 +1055,7 @@ int xen_panic_handler_init(void) return 0; } -static const struct machine_ops __initdata xen_machine_ops = { +static const struct machine_ops xen_machine_ops __initconst = { .restart = xen_restart, .halt = xen_machine_halt, .power_off = xen_machine_halt, @@ -1332,7 +1332,7 @@ static int __cpuinit xen_hvm_cpu_notify(struct notifier_block *self, return NOTIFY_OK; } -static struct notifier_block __cpuinitdata xen_hvm_cpu_notifier = { +static struct notifier_block xen_hvm_cpu_notifier __cpuinitdata = { .notifier_call = xen_hvm_cpu_notify, }; @@ -1381,7 +1381,7 @@ bool xen_hvm_need_lapic(void) } EXPORT_SYMBOL_GPL(xen_hvm_need_lapic); -const __refconst struct hypervisor_x86 x86_hyper_xen_hvm = { +const struct hypervisor_x86 x86_hyper_xen_hvm __refconst = { .name = "Xen HVM", .detect = xen_hvm_platform, .init_platform = xen_hvm_guest_init, diff --git a/arch/x86/xen/irq.c b/arch/x86/xen/irq.c index 6a6fe8939645..8bbb465b6f0a 100644 --- a/arch/x86/xen/irq.c +++ b/arch/x86/xen/irq.c @@ -113,7 +113,7 @@ static void xen_halt(void) xen_safe_halt(); } -static const struct pv_irq_ops xen_irq_ops __initdata = { +static const struct pv_irq_ops xen_irq_ops __initconst = { .save_fl = PV_CALLEE_SAVE(xen_save_fl), .restore_fl = PV_CALLEE_SAVE(xen_restore_fl), .irq_disable = PV_CALLEE_SAVE(xen_irq_disable), diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c index aef7af92b28b..02d752460371 100644 --- a/arch/x86/xen/mmu.c +++ b/arch/x86/xen/mmu.c @@ -1054,7 +1054,7 @@ void xen_mm_pin_all(void) * that's before we have page structures to store the bits. So do all * the book-keeping now. */ -static __init int xen_mark_pinned(struct mm_struct *mm, struct page *page, +static int __init xen_mark_pinned(struct mm_struct *mm, struct page *page, enum pt_level level) { SetPagePinned(page); @@ -1187,7 +1187,7 @@ static void drop_other_mm_ref(void *info) active_mm = percpu_read(cpu_tlbstate.active_mm); - if (active_mm == mm) + if (active_mm == mm && percpu_read(cpu_tlbstate.state) != TLBSTATE_OK) leave_mm(smp_processor_id()); /* If this cpu still has a stale cr3 reference, then make sure @@ -1271,13 +1271,27 @@ void xen_exit_mmap(struct mm_struct *mm) spin_unlock(&mm->page_table_lock); } -static __init void xen_pagetable_setup_start(pgd_t *base) +static void __init xen_pagetable_setup_start(pgd_t *base) { } +static __init void xen_mapping_pagetable_reserve(u64 start, u64 end) +{ + /* reserve the range used */ + native_pagetable_reserve(start, end); + + /* set as RW the rest */ + printk(KERN_DEBUG "xen: setting RW the range %llx - %llx\n", end, + PFN_PHYS(pgt_buf_top)); + while (end < PFN_PHYS(pgt_buf_top)) { + make_lowmem_page_readwrite(__va(end)); + end += PAGE_SIZE; + } +} + static void xen_post_allocator_init(void); -static __init void xen_pagetable_setup_done(pgd_t *base) +static void __init xen_pagetable_setup_done(pgd_t *base) { xen_setup_shared_info(); xen_post_allocator_init(); @@ -1474,7 +1488,7 @@ static void xen_pgd_free(struct mm_struct *mm, pgd_t *pgd) } #ifdef CONFIG_X86_32 -static __init pte_t mask_rw_pte(pte_t *ptep, pte_t pte) +static pte_t __init mask_rw_pte(pte_t *ptep, pte_t pte) { /* If there's an existing pte, then don't allow _PAGE_RW to be set */ if (pte_val_ma(*ptep) & _PAGE_PRESENT) @@ -1484,7 +1498,7 @@ static __init pte_t mask_rw_pte(pte_t *ptep, pte_t pte) return pte; } #else /* CONFIG_X86_64 */ -static __init pte_t mask_rw_pte(pte_t *ptep, pte_t pte) +static pte_t __init mask_rw_pte(pte_t *ptep, pte_t pte) { unsigned long pfn = pte_pfn(pte); @@ -1495,7 +1509,7 @@ static __init pte_t mask_rw_pte(pte_t *ptep, pte_t pte) * it is RO. */ if (((!is_early_ioremap_ptep(ptep) && - pfn >= pgt_buf_start && pfn < pgt_buf_end)) || + pfn >= pgt_buf_start && pfn < pgt_buf_top)) || (is_early_ioremap_ptep(ptep) && pfn != (pgt_buf_end - 1))) pte = pte_wrprotect(pte); @@ -1505,7 +1519,7 @@ static __init pte_t mask_rw_pte(pte_t *ptep, pte_t pte) /* Init-time set_pte while constructing initial pagetables, which doesn't allow RO pagetable pages to be remapped RW */ -static __init void xen_set_pte_init(pte_t *ptep, pte_t pte) +static void __init xen_set_pte_init(pte_t *ptep, pte_t pte) { pte = mask_rw_pte(ptep, pte); @@ -1523,7 +1537,7 @@ static void pin_pagetable_pfn(unsigned cmd, unsigned long pfn) /* Early in boot, while setting up the initial pagetable, assume everything is pinned. */ -static __init void xen_alloc_pte_init(struct mm_struct *mm, unsigned long pfn) +static void __init xen_alloc_pte_init(struct mm_struct *mm, unsigned long pfn) { #ifdef CONFIG_FLATMEM BUG_ON(mem_map); /* should only be used early */ @@ -1533,7 +1547,7 @@ static __init void xen_alloc_pte_init(struct mm_struct *mm, unsigned long pfn) } /* Used for pmd and pud */ -static __init void xen_alloc_pmd_init(struct mm_struct *mm, unsigned long pfn) +static void __init xen_alloc_pmd_init(struct mm_struct *mm, unsigned long pfn) { #ifdef CONFIG_FLATMEM BUG_ON(mem_map); /* should only be used early */ @@ -1543,13 +1557,13 @@ static __init void xen_alloc_pmd_init(struct mm_struct *mm, unsigned long pfn) /* Early release_pte assumes that all pts are pinned, since there's only init_mm and anything attached to that is pinned. */ -static __init void xen_release_pte_init(unsigned long pfn) +static void __init xen_release_pte_init(unsigned long pfn) { pin_pagetable_pfn(MMUEXT_UNPIN_TABLE, pfn); make_lowmem_page_readwrite(__va(PFN_PHYS(pfn))); } -static __init void xen_release_pmd_init(unsigned long pfn) +static void __init xen_release_pmd_init(unsigned long pfn) { make_lowmem_page_readwrite(__va(PFN_PHYS(pfn))); } @@ -1675,7 +1689,7 @@ static void set_page_prot(void *addr, pgprot_t prot) BUG(); } -static __init void xen_map_identity_early(pmd_t *pmd, unsigned long max_pfn) +static void __init xen_map_identity_early(pmd_t *pmd, unsigned long max_pfn) { unsigned pmdidx, pteidx; unsigned ident_pte; @@ -1758,7 +1772,7 @@ static void convert_pfn_mfn(void *v) * of the physical mapping once some sort of allocator has been set * up. */ -__init pgd_t *xen_setup_kernel_pagetable(pgd_t *pgd, +pgd_t * __init xen_setup_kernel_pagetable(pgd_t *pgd, unsigned long max_pfn) { pud_t *l3; @@ -1829,7 +1843,7 @@ __init pgd_t *xen_setup_kernel_pagetable(pgd_t *pgd, static RESERVE_BRK_ARRAY(pmd_t, initial_kernel_pmd, PTRS_PER_PMD); static RESERVE_BRK_ARRAY(pmd_t, swapper_kernel_pmd, PTRS_PER_PMD); -static __init void xen_write_cr3_init(unsigned long cr3) +static void __init xen_write_cr3_init(unsigned long cr3) { unsigned long pfn = PFN_DOWN(__pa(swapper_pg_dir)); @@ -1866,7 +1880,7 @@ static __init void xen_write_cr3_init(unsigned long cr3) pv_mmu_ops.write_cr3 = &xen_write_cr3; } -__init pgd_t *xen_setup_kernel_pagetable(pgd_t *pgd, +pgd_t * __init xen_setup_kernel_pagetable(pgd_t *pgd, unsigned long max_pfn) { pmd_t *kernel_pmd; @@ -1972,7 +1986,7 @@ static void xen_set_fixmap(unsigned idx, phys_addr_t phys, pgprot_t prot) #endif } -__init void xen_ident_map_ISA(void) +void __init xen_ident_map_ISA(void) { unsigned long pa; @@ -1995,7 +2009,7 @@ __init void xen_ident_map_ISA(void) xen_flush_tlb(); } -static __init void xen_post_allocator_init(void) +static void __init xen_post_allocator_init(void) { #ifdef CONFIG_XEN_DEBUG pv_mmu_ops.make_pte = PV_CALLEE_SAVE(xen_make_pte_debug); @@ -2032,7 +2046,7 @@ static void xen_leave_lazy_mmu(void) preempt_enable(); } -static const struct pv_mmu_ops xen_mmu_ops __initdata = { +static const struct pv_mmu_ops xen_mmu_ops __initconst = { .read_cr2 = xen_read_cr2, .write_cr2 = xen_write_cr2, @@ -2105,6 +2119,7 @@ static const struct pv_mmu_ops xen_mmu_ops __initdata = { void __init xen_init_mmu_ops(void) { + x86_init.mapping.pagetable_reserve = xen_mapping_pagetable_reserve; x86_init.paging.pagetable_setup_start = xen_pagetable_setup_start; x86_init.paging.pagetable_setup_done = xen_pagetable_setup_done; pv_mmu_ops = xen_mmu_ops; diff --git a/arch/x86/xen/p2m.c b/arch/x86/xen/p2m.c index 141eb0de8b06..58efeb9d5440 100644 --- a/arch/x86/xen/p2m.c +++ b/arch/x86/xen/p2m.c @@ -522,11 +522,20 @@ static bool __init __early_alloc_p2m(unsigned long pfn) /* Boundary cross-over for the edges: */ if (idx) { unsigned long *p2m = extend_brk(PAGE_SIZE, PAGE_SIZE); + unsigned long *mid_mfn_p; p2m_init(p2m); p2m_top[topidx][mididx] = p2m; + /* For save/restore we need to MFN of the P2M saved */ + + mid_mfn_p = p2m_top_mfn_p[topidx]; + WARN(mid_mfn_p[mididx] != virt_to_mfn(p2m_missing), + "P2M_TOP_P[%d][%d] != MFN of p2m_missing!\n", + topidx, mididx); + mid_mfn_p[mididx] = virt_to_mfn(p2m); + } return idx != 0; } @@ -549,12 +558,29 @@ unsigned long __init set_phys_range_identity(unsigned long pfn_s, pfn += P2M_MID_PER_PAGE * P2M_PER_PAGE) { unsigned topidx = p2m_top_index(pfn); - if (p2m_top[topidx] == p2m_mid_missing) { - unsigned long **mid = extend_brk(PAGE_SIZE, PAGE_SIZE); + unsigned long *mid_mfn_p; + unsigned long **mid; + + mid = p2m_top[topidx]; + mid_mfn_p = p2m_top_mfn_p[topidx]; + if (mid == p2m_mid_missing) { + mid = extend_brk(PAGE_SIZE, PAGE_SIZE); p2m_mid_init(mid); p2m_top[topidx] = mid; + + BUG_ON(mid_mfn_p != p2m_mid_missing_mfn); + } + /* And the save/restore P2M tables.. */ + if (mid_mfn_p == p2m_mid_missing_mfn) { + mid_mfn_p = extend_brk(PAGE_SIZE, PAGE_SIZE); + p2m_mid_mfn_init(mid_mfn_p); + + p2m_top_mfn_p[topidx] = mid_mfn_p; + p2m_top_mfn[topidx] = virt_to_mfn(mid_mfn_p); + /* Note: we don't set mid_mfn_p[midix] here, + * look in __early_alloc_p2m */ } } @@ -650,7 +676,7 @@ static unsigned long mfn_hash(unsigned long mfn) } /* Add an MFN override for a particular page */ -int m2p_add_override(unsigned long mfn, struct page *page) +int m2p_add_override(unsigned long mfn, struct page *page, bool clear_pte) { unsigned long flags; unsigned long pfn; @@ -662,7 +688,6 @@ int m2p_add_override(unsigned long mfn, struct page *page) if (!PageHighMem(page)) { address = (unsigned long)__va(pfn << PAGE_SHIFT); ptep = lookup_address(address, &level); - if (WARN(ptep == NULL || level != PG_LEVEL_4K, "m2p_add_override: pfn %lx not mapped", pfn)) return -EINVAL; @@ -674,18 +699,17 @@ int m2p_add_override(unsigned long mfn, struct page *page) if (unlikely(!set_phys_to_machine(pfn, FOREIGN_FRAME(mfn)))) return -ENOMEM; - if (!PageHighMem(page)) + if (clear_pte && !PageHighMem(page)) /* Just zap old mapping for now */ pte_clear(&init_mm, address, ptep); - spin_lock_irqsave(&m2p_override_lock, flags); list_add(&page->lru, &m2p_overrides[mfn_hash(mfn)]); spin_unlock_irqrestore(&m2p_override_lock, flags); return 0; } - -int m2p_remove_override(struct page *page) +EXPORT_SYMBOL_GPL(m2p_add_override); +int m2p_remove_override(struct page *page, bool clear_pte) { unsigned long flags; unsigned long mfn; @@ -713,7 +737,7 @@ int m2p_remove_override(struct page *page) spin_unlock_irqrestore(&m2p_override_lock, flags); set_phys_to_machine(pfn, page->index); - if (!PageHighMem(page)) + if (clear_pte && !PageHighMem(page)) set_pte_at(&init_mm, address, ptep, pfn_pte(pfn, PAGE_KERNEL)); /* No tlb flush necessary because the caller already @@ -721,6 +745,7 @@ int m2p_remove_override(struct page *page) return 0; } +EXPORT_SYMBOL_GPL(m2p_remove_override); struct page *m2p_find_override(unsigned long mfn) { diff --git a/arch/x86/xen/pci-swiotlb-xen.c b/arch/x86/xen/pci-swiotlb-xen.c index bfd0632fe65e..b480d4207a4c 100644 --- a/arch/x86/xen/pci-swiotlb-xen.c +++ b/arch/x86/xen/pci-swiotlb-xen.c @@ -36,7 +36,7 @@ int __init pci_xen_swiotlb_detect(void) /* If running as PV guest, either iommu=soft, or swiotlb=force will * activate this IOMMU. If running as PV privileged, activate it - * irregardlesss. + * irregardless. */ if ((xen_initial_domain() || swiotlb || swiotlb_force) && (xen_pv_domain())) diff --git a/arch/x86/xen/setup.c b/arch/x86/xen/setup.c index 90bac0aac3a5..be1a464f6d66 100644 --- a/arch/x86/xen/setup.c +++ b/arch/x86/xen/setup.c @@ -50,7 +50,7 @@ phys_addr_t xen_extra_mem_start, xen_extra_mem_size; */ #define EXTRA_MEM_RATIO (10) -static __init void xen_add_extra_mem(unsigned long pages) +static void __init xen_add_extra_mem(unsigned long pages) { unsigned long pfn; @@ -166,7 +166,7 @@ static unsigned long __init xen_set_identity(const struct e820entry *list, if (last > end) continue; - if (entry->type == E820_RAM) { + if ((entry->type == E820_RAM) || (entry->type == E820_UNUSABLE)) { if (start > start_pci) identity += set_phys_range_identity( PFN_UP(start_pci), PFN_DOWN(start)); @@ -227,7 +227,11 @@ char * __init xen_memory_setup(void) memcpy(map_raw, map, sizeof(map)); e820.nr_map = 0; +#ifdef CONFIG_X86_32 + xen_extra_mem_start = mem_end; +#else xen_extra_mem_start = max((1ULL << 32), mem_end); +#endif for (i = 0; i < memmap.nr_entries; i++) { unsigned long long end; @@ -336,7 +340,7 @@ static void __init fiddle_vdso(void) #endif } -static __cpuinit int register_callback(unsigned type, const void *func) +static int __cpuinit register_callback(unsigned type, const void *func) { struct callback_register callback = { .type = type, diff --git a/arch/x86/xen/smp.c b/arch/x86/xen/smp.c index 30612441ed99..41038c01de40 100644 --- a/arch/x86/xen/smp.c +++ b/arch/x86/xen/smp.c @@ -46,18 +46,17 @@ static irqreturn_t xen_call_function_interrupt(int irq, void *dev_id); static irqreturn_t xen_call_function_single_interrupt(int irq, void *dev_id); /* - * Reschedule call back. Nothing to do, - * all the work is done automatically when - * we return from the interrupt. + * Reschedule call back. */ static irqreturn_t xen_reschedule_interrupt(int irq, void *dev_id) { inc_irq_stat(irq_resched_count); + scheduler_ipi(); return IRQ_HANDLED; } -static __cpuinit void cpu_bringup(void) +static void __cpuinit cpu_bringup(void) { int cpu = smp_processor_id(); @@ -85,7 +84,7 @@ static __cpuinit void cpu_bringup(void) wmb(); /* make sure everything is out */ } -static __cpuinit void cpu_bringup_and_idle(void) +static void __cpuinit cpu_bringup_and_idle(void) { cpu_bringup(); cpu_idle(); @@ -242,7 +241,7 @@ static void __init xen_smp_prepare_cpus(unsigned int max_cpus) } } -static __cpuinit int +static int __cpuinit cpu_initialize_context(unsigned int cpu, struct task_struct *idle) { struct vcpu_guest_context *ctxt; @@ -486,7 +485,7 @@ static irqreturn_t xen_call_function_single_interrupt(int irq, void *dev_id) return IRQ_HANDLED; } -static const struct smp_ops xen_smp_ops __initdata = { +static const struct smp_ops xen_smp_ops __initconst = { .smp_prepare_boot_cpu = xen_smp_prepare_boot_cpu, .smp_prepare_cpus = xen_smp_prepare_cpus, .smp_cpus_done = xen_smp_cpus_done, diff --git a/arch/x86/xen/time.c b/arch/x86/xen/time.c index 2e2d370a47b1..5158c505bef9 100644 --- a/arch/x86/xen/time.c +++ b/arch/x86/xen/time.c @@ -26,8 +26,6 @@ #include "xen-ops.h" -#define XEN_SHIFT 22 - /* Xen may fire a timer up to this many ns early */ #define TIMER_SLOP 100000 #define NS_PER_TICK (1000000000LL / HZ) @@ -211,8 +209,6 @@ static struct clocksource xen_clocksource __read_mostly = { .rating = 400, .read = xen_clocksource_get_cycles, .mask = ~0, - .mult = 1<<XEN_SHIFT, /* time directly in nanoseconds */ - .shift = XEN_SHIFT, .flags = CLOCK_SOURCE_IS_CONTINUOUS, }; @@ -439,16 +435,16 @@ void xen_timer_resume(void) } } -static const struct pv_time_ops xen_time_ops __initdata = { +static const struct pv_time_ops xen_time_ops __initconst = { .sched_clock = xen_clocksource_read, }; -static __init void xen_time_init(void) +static void __init xen_time_init(void) { int cpu = smp_processor_id(); struct timespec tp; - clocksource_register(&xen_clocksource); + clocksource_register_hz(&xen_clocksource, NSEC_PER_SEC); if (HYPERVISOR_vcpu_op(VCPUOP_stop_periodic_timer, cpu, NULL) == 0) { /* Successfully turned off 100Hz tick, so we have the @@ -468,7 +464,7 @@ static __init void xen_time_init(void) xen_setup_cpu_clockevents(); } -__init void xen_init_time_ops(void) +void __init xen_init_time_ops(void) { pv_time_ops = xen_time_ops; @@ -490,7 +486,7 @@ static void xen_hvm_setup_cpu_clockevents(void) xen_setup_cpu_clockevents(); } -__init void xen_hvm_init_time_ops(void) +void __init xen_hvm_init_time_ops(void) { /* vector callback is needed otherwise we cannot receive interrupts * on cpu > 0 and at this point we don't know how many cpus are diff --git a/arch/x86/xen/xen-ops.h b/arch/x86/xen/xen-ops.h index 3112f55638c4..97dfdc8757b3 100644 --- a/arch/x86/xen/xen-ops.h +++ b/arch/x86/xen/xen-ops.h @@ -74,7 +74,7 @@ static inline void xen_hvm_smp_init(void) {} #ifdef CONFIG_PARAVIRT_SPINLOCKS void __init xen_init_spinlocks(void); -__cpuinit void xen_init_lock_cpu(int cpu); +void __cpuinit xen_init_lock_cpu(int cpu); void xen_uninit_lock_cpu(int cpu); #else static inline void xen_init_spinlocks(void) |