diff options
Diffstat (limited to 'arch/x86')
159 files changed, 5278 insertions, 2181 deletions
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 73f7fe8fd4d..3a330a437c6 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -165,6 +165,9 @@ config GENERIC_HARDIRQS bool default y +config GENERIC_HARDIRQS_NO__DO_IRQ + def_bool y + config GENERIC_IRQ_PROBE bool default y @@ -1802,6 +1805,17 @@ config DMAR and include PCI device scope covered by these DMA remapping devices. +config DMAR_DEFAULT_ON + def_bool y + prompt "Enable DMA Remapping Devices by default" + depends on DMAR + help + Selecting this option will enable a DMAR device at boot time if + one is found. If this option is not selected, DMAR support can + be enabled by passing intel_iommu=on to the kernel. It is + recommended you say N here while the DMAR code remains + experimental. + config DMAR_GFX_WA def_bool y prompt "Support for Graphics workaround" diff --git a/arch/x86/Kconfig.cpu b/arch/x86/Kconfig.cpu index 8078955845a..c98d52e8296 100644 --- a/arch/x86/Kconfig.cpu +++ b/arch/x86/Kconfig.cpu @@ -167,9 +167,9 @@ config MK7 config MK8 bool "Opteron/Athlon64/Hammer/K8" help - Select this for an AMD Opteron or Athlon64 Hammer-family processor. Enables - use of some extended instructions, and passes appropriate optimization - flags to GCC. + Select this for an AMD Opteron or Athlon64 Hammer-family processor. + Enables use of some extended instructions, and passes appropriate + optimization flags to GCC. config MCRUSOE bool "Crusoe" @@ -256,9 +256,11 @@ config MPSC config MCORE2 bool "Core 2/newer Xeon" help - Select this for Intel Core 2 and newer Core 2 Xeons (Xeon 51xx and 53xx) - CPUs. You can distinguish newer from older Xeons by the CPU family - in /proc/cpuinfo. Newer ones have 6 and older ones 15 (not a typo) + + Select this for Intel Core 2 and newer Core 2 Xeons (Xeon 51xx and + 53xx) CPUs. You can distinguish newer from older Xeons by the CPU + family in /proc/cpuinfo. Newer ones have 6 and older ones 15 + (not a typo) config GENERIC_CPU bool "Generic-x86-64" @@ -320,14 +322,14 @@ config X86_PPRO_FENCE bool "PentiumPro memory ordering errata workaround" depends on M686 || M586MMX || M586TSC || M586 || M486 || M386 || MGEODEGX1 help - Old PentiumPro multiprocessor systems had errata that could cause memory - operations to violate the x86 ordering standard in rare cases. Enabling this - option will attempt to work around some (but not all) occurances of - this problem, at the cost of much heavier spinlock and memory barrier - operations. + Old PentiumPro multiprocessor systems had errata that could cause + memory operations to violate the x86 ordering standard in rare cases. + Enabling this option will attempt to work around some (but not all) + occurances of this problem, at the cost of much heavier spinlock and + memory barrier operations. - If unsure, say n here. Even distro kernels should think twice before enabling - this: there are few systems, and an unlikely bug. + If unsure, say n here. Even distro kernels should think twice before + enabling this: there are few systems, and an unlikely bug. config X86_F00F_BUG def_bool y diff --git a/arch/x86/Kconfig.debug b/arch/x86/Kconfig.debug index 10d6cc3fd05..e1983fa025d 100644 --- a/arch/x86/Kconfig.debug +++ b/arch/x86/Kconfig.debug @@ -174,28 +174,8 @@ config IOMMU_LEAK Add a simple leak tracer to the IOMMU code. This is useful when you are debugging a buggy device driver that leaks IOMMU mappings. -config MMIOTRACE - bool "Memory mapped IO tracing" - depends on DEBUG_KERNEL && PCI - select TRACING - help - Mmiotrace traces Memory Mapped I/O access and is meant for - debugging and reverse engineering. It is called from the ioremap - implementation and works via page faults. Tracing is disabled by - default and can be enabled at run-time. - - See Documentation/tracers/mmiotrace.txt. - If you are not helping to develop drivers, say N. - -config MMIOTRACE_TEST - tristate "Test module for mmiotrace" - depends on MMIOTRACE && m - help - This is a dumb module for testing mmiotrace. It is very dangerous - as it will write garbage to IO memory starting at a given address. - However, it should be safe to use on e.g. unused portion of VRAM. - - Say N, unless you absolutely know what you are doing. +config HAVE_MMIOTRACE_SUPPORT + def_bool y # # IO delay types: diff --git a/arch/x86/boot/video-vesa.c b/arch/x86/boot/video-vesa.c index 75115849af3..4a58c8ce3f6 100644 --- a/arch/x86/boot/video-vesa.c +++ b/arch/x86/boot/video-vesa.c @@ -269,9 +269,8 @@ void vesa_store_edid(void) we genuinely have to assume all registers are destroyed here. */ asm("pushw %%es; movw %2,%%es; "INT10"; popw %%es" - : "+a" (ax), "+b" (bx) - : "c" (cx), "D" (di) - : "esi"); + : "+a" (ax), "+b" (bx), "+c" (cx), "+D" (di) + : : "esi", "edx"); if (ax != 0x004f) return; /* No EDID */ @@ -285,9 +284,9 @@ void vesa_store_edid(void) dx = 0; /* EDID block number */ di =(size_t) &boot_params.edid_info; /* (ES:)Pointer to block */ asm(INT10 - : "+a" (ax), "+b" (bx), "+d" (dx), "=m" (boot_params.edid_info) - : "c" (cx), "D" (di) - : "esi"); + : "+a" (ax), "+b" (bx), "+d" (dx), "=m" (boot_params.edid_info), + "+c" (cx), "+D" (di) + : : "esi"); #endif /* CONFIG_FIRMWARE_EDID */ } diff --git a/arch/x86/configs/i386_defconfig b/arch/x86/configs/i386_defconfig index b30a08ed8eb..edba00d98ac 100644 --- a/arch/x86/configs/i386_defconfig +++ b/arch/x86/configs/i386_defconfig @@ -1331,8 +1331,8 @@ CONFIG_I2C_I801=y # Miscellaneous I2C Chip support # # CONFIG_DS1682 is not set -# CONFIG_AT24 is not set -# CONFIG_SENSORS_EEPROM is not set +# CONFIG_EEPROM_AT24 is not set +# CONFIG_EEPROM_LEGACY is not set # CONFIG_SENSORS_PCF8574 is not set # CONFIG_PCF8575 is not set # CONFIG_SENSORS_PCA9539 is not set diff --git a/arch/x86/configs/x86_64_defconfig b/arch/x86/configs/x86_64_defconfig index 0e7dbc0a3e4..322dd2748fc 100644 --- a/arch/x86/configs/x86_64_defconfig +++ b/arch/x86/configs/x86_64_defconfig @@ -1311,8 +1311,8 @@ CONFIG_I2C_I801=y # Miscellaneous I2C Chip support # # CONFIG_DS1682 is not set -# CONFIG_AT24 is not set -# CONFIG_SENSORS_EEPROM is not set +# CONFIG_EEPROM_AT24 is not set +# CONFIG_EEPROM_LEGACY is not set # CONFIG_SENSORS_PCF8574 is not set # CONFIG_PCF8575 is not set # CONFIG_SENSORS_PCA9539 is not set diff --git a/arch/x86/crypto/Makefile b/arch/x86/crypto/Makefile index 903de4aa509..ebe7deedd5b 100644 --- a/arch/x86/crypto/Makefile +++ b/arch/x86/crypto/Makefile @@ -9,6 +9,7 @@ obj-$(CONFIG_CRYPTO_SALSA20_586) += salsa20-i586.o obj-$(CONFIG_CRYPTO_AES_X86_64) += aes-x86_64.o obj-$(CONFIG_CRYPTO_TWOFISH_X86_64) += twofish-x86_64.o obj-$(CONFIG_CRYPTO_SALSA20_X86_64) += salsa20-x86_64.o +obj-$(CONFIG_CRYPTO_AES_NI_INTEL) += aesni-intel.o obj-$(CONFIG_CRYPTO_CRC32C_INTEL) += crc32c-intel.o @@ -19,3 +20,5 @@ salsa20-i586-y := salsa20-i586-asm_32.o salsa20_glue.o 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 diff --git a/arch/x86/crypto/aes-i586-asm_32.S b/arch/x86/crypto/aes-i586-asm_32.S index e41b147f450..b949ec2f9af 100644 --- a/arch/x86/crypto/aes-i586-asm_32.S +++ b/arch/x86/crypto/aes-i586-asm_32.S @@ -41,14 +41,14 @@ #define tlen 1024 // length of each of 4 'xor' arrays (256 32-bit words) /* offsets to parameters with one register pushed onto stack */ -#define tfm 8 +#define ctx 8 #define out_blk 12 #define in_blk 16 -/* offsets in crypto_tfm structure */ -#define klen (crypto_tfm_ctx_offset + 0) -#define ekey (crypto_tfm_ctx_offset + 4) -#define dkey (crypto_tfm_ctx_offset + 244) +/* offsets in crypto_aes_ctx structure */ +#define klen (480) +#define ekey (0) +#define dkey (240) // register mapping for encrypt and decrypt subroutines @@ -217,7 +217,7 @@ do_col (table, r5,r0,r1,r4, r2,r3); /* idx=r5 */ // AES (Rijndael) Encryption Subroutine -/* void aes_enc_blk(struct crypto_tfm *tfm, u8 *out_blk, const u8 *in_blk) */ +/* void aes_enc_blk(struct crypto_aes_ctx *ctx, u8 *out_blk, const u8 *in_blk) */ .global aes_enc_blk @@ -228,7 +228,7 @@ aes_enc_blk: push %ebp - mov tfm(%esp),%ebp + mov ctx(%esp),%ebp // CAUTION: the order and the values used in these assigns // rely on the register mappings @@ -292,7 +292,7 @@ aes_enc_blk: ret // AES (Rijndael) Decryption Subroutine -/* void aes_dec_blk(struct crypto_tfm *tfm, u8 *out_blk, const u8 *in_blk) */ +/* void aes_dec_blk(struct crypto_aes_ctx *ctx, u8 *out_blk, const u8 *in_blk) */ .global aes_dec_blk @@ -303,7 +303,7 @@ aes_enc_blk: aes_dec_blk: push %ebp - mov tfm(%esp),%ebp + mov ctx(%esp),%ebp // CAUTION: the order and the values used in these assigns // rely on the register mappings diff --git a/arch/x86/crypto/aes-x86_64-asm_64.S b/arch/x86/crypto/aes-x86_64-asm_64.S index a120f526c3d..5b577d5a059 100644 --- a/arch/x86/crypto/aes-x86_64-asm_64.S +++ b/arch/x86/crypto/aes-x86_64-asm_64.S @@ -17,8 +17,6 @@ #include <asm/asm-offsets.h> -#define BASE crypto_tfm_ctx_offset - #define R1 %rax #define R1E %eax #define R1X %ax @@ -56,13 +54,13 @@ .align 8; \ FUNC: movq r1,r2; \ movq r3,r4; \ - leaq BASE+KEY+48+4(r8),r9; \ + leaq KEY+48(r8),r9; \ movq r10,r11; \ movl (r7),r5 ## E; \ movl 4(r7),r1 ## E; \ movl 8(r7),r6 ## E; \ movl 12(r7),r7 ## E; \ - movl BASE+0(r8),r10 ## E; \ + movl 480(r8),r10 ## E; \ xorl -48(r9),r5 ## E; \ xorl -44(r9),r1 ## E; \ xorl -40(r9),r6 ## E; \ diff --git a/arch/x86/crypto/aes_glue.c b/arch/x86/crypto/aes_glue.c index 71f45782711..49ae9fe32b2 100644 --- a/arch/x86/crypto/aes_glue.c +++ b/arch/x86/crypto/aes_glue.c @@ -5,17 +5,29 @@ #include <crypto/aes.h> -asmlinkage void aes_enc_blk(struct crypto_tfm *tfm, u8 *out, const u8 *in); -asmlinkage void aes_dec_blk(struct crypto_tfm *tfm, u8 *out, const u8 *in); +asmlinkage void aes_enc_blk(struct crypto_aes_ctx *ctx, u8 *out, const u8 *in); +asmlinkage void aes_dec_blk(struct crypto_aes_ctx *ctx, u8 *out, const u8 *in); + +void crypto_aes_encrypt_x86(struct crypto_aes_ctx *ctx, u8 *dst, const u8 *src) +{ + aes_enc_blk(ctx, dst, src); +} +EXPORT_SYMBOL_GPL(crypto_aes_encrypt_x86); + +void crypto_aes_decrypt_x86(struct crypto_aes_ctx *ctx, u8 *dst, const u8 *src) +{ + aes_dec_blk(ctx, dst, src); +} +EXPORT_SYMBOL_GPL(crypto_aes_decrypt_x86); static void aes_encrypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src) { - aes_enc_blk(tfm, dst, src); + aes_enc_blk(crypto_tfm_ctx(tfm), dst, src); } static void aes_decrypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src) { - aes_dec_blk(tfm, dst, src); + aes_dec_blk(crypto_tfm_ctx(tfm), dst, src); } static struct crypto_alg aes_alg = { diff --git a/arch/x86/crypto/aesni-intel_asm.S b/arch/x86/crypto/aesni-intel_asm.S new file mode 100644 index 00000000000..caba9960170 --- /dev/null +++ b/arch/x86/crypto/aesni-intel_asm.S @@ -0,0 +1,896 @@ +/* + * Implement AES algorithm in Intel AES-NI instructions. + * + * The white paper of AES-NI instructions can be downloaded from: + * http://softwarecommunity.intel.com/isn/downloads/intelavx/AES-Instructions-Set_WP.pdf + * + * Copyright (C) 2008, Intel Corp. + * Author: Huang Ying <ying.huang@intel.com> + * Vinodh Gopal <vinodh.gopal@intel.com> + * Kahraman Akdemir + * + * 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. + */ + +#include <linux/linkage.h> + +.text + +#define STATE1 %xmm0 +#define STATE2 %xmm4 +#define STATE3 %xmm5 +#define STATE4 %xmm6 +#define STATE STATE1 +#define IN1 %xmm1 +#define IN2 %xmm7 +#define IN3 %xmm8 +#define IN4 %xmm9 +#define IN IN1 +#define KEY %xmm2 +#define IV %xmm3 + +#define KEYP %rdi +#define OUTP %rsi +#define INP %rdx +#define LEN %rcx +#define IVP %r8 +#define KLEN %r9d +#define T1 %r10 +#define TKEYP T1 +#define T2 %r11 + +_key_expansion_128: +_key_expansion_256a: + pshufd $0b11111111, %xmm1, %xmm1 + shufps $0b00010000, %xmm0, %xmm4 + pxor %xmm4, %xmm0 + shufps $0b10001100, %xmm0, %xmm4 + pxor %xmm4, %xmm0 + pxor %xmm1, %xmm0 + movaps %xmm0, (%rcx) + add $0x10, %rcx + ret + +_key_expansion_192a: + pshufd $0b01010101, %xmm1, %xmm1 + shufps $0b00010000, %xmm0, %xmm4 + pxor %xmm4, %xmm0 + shufps $0b10001100, %xmm0, %xmm4 + pxor %xmm4, %xmm0 + pxor %xmm1, %xmm0 + + movaps %xmm2, %xmm5 + movaps %xmm2, %xmm6 + pslldq $4, %xmm5 + pshufd $0b11111111, %xmm0, %xmm3 + pxor %xmm3, %xmm2 + pxor %xmm5, %xmm2 + + movaps %xmm0, %xmm1 + shufps $0b01000100, %xmm0, %xmm6 + movaps %xmm6, (%rcx) + shufps $0b01001110, %xmm2, %xmm1 + movaps %xmm1, 16(%rcx) + add $0x20, %rcx + ret + +_key_expansion_192b: + pshufd $0b01010101, %xmm1, %xmm1 + shufps $0b00010000, %xmm0, %xmm4 + pxor %xmm4, %xmm0 + shufps $0b10001100, %xmm0, %xmm4 + pxor %xmm4, %xmm0 + pxor %xmm1, %xmm0 + + movaps %xmm2, %xmm5 + pslldq $4, %xmm5 + pshufd $0b11111111, %xmm0, %xmm3 + pxor %xmm3, %xmm2 + pxor %xmm5, %xmm2 + + movaps %xmm0, (%rcx) + add $0x10, %rcx + ret + +_key_expansion_256b: + pshufd $0b10101010, %xmm1, %xmm1 + shufps $0b00010000, %xmm2, %xmm4 + pxor %xmm4, %xmm2 + shufps $0b10001100, %xmm2, %xmm4 + pxor %xmm4, %xmm2 + pxor %xmm1, %xmm2 + movaps %xmm2, (%rcx) + add $0x10, %rcx + ret + +/* + * int aesni_set_key(struct crypto_aes_ctx *ctx, const u8 *in_key, + * unsigned int key_len) + */ +ENTRY(aesni_set_key) + movups (%rsi), %xmm0 # user key (first 16 bytes) + movaps %xmm0, (%rdi) + lea 0x10(%rdi), %rcx # key addr + movl %edx, 480(%rdi) + pxor %xmm4, %xmm4 # xmm4 is assumed 0 in _key_expansion_x + cmp $24, %dl + jb .Lenc_key128 + je .Lenc_key192 + movups 0x10(%rsi), %xmm2 # other user key + movaps %xmm2, (%rcx) + add $0x10, %rcx + # aeskeygenassist $0x1, %xmm2, %xmm1 # round 1 + .byte 0x66, 0x0f, 0x3a, 0xdf, 0xca, 0x01 + call _key_expansion_256a + # aeskeygenassist $0x1, %xmm0, %xmm1 + .byte 0x66, 0x0f, 0x3a, 0xdf, 0xc8, 0x01 + call _key_expansion_256b + # aeskeygenassist $0x2, %xmm2, %xmm1 # round 2 + .byte 0x66, 0x0f, 0x3a, 0xdf, 0xca, 0x02 + call _key_expansion_256a + # aeskeygenassist $0x2, %xmm0, %xmm1 + .byte 0x66, 0x0f, 0x3a, 0xdf, 0xc8, 0x02 + call _key_expansion_256b + # aeskeygenassist $0x4, %xmm2, %xmm1 # round 3 + .byte 0x66, 0x0f, 0x3a, 0xdf, 0xca, 0x04 + call _key_expansion_256a + # aeskeygenassist $0x4, %xmm0, %xmm1 + .byte 0x66, 0x0f, 0x3a, 0xdf, 0xc8, 0x04 + call _key_expansion_256b + # aeskeygenassist $0x8, %xmm2, %xmm1 # round 4 + .byte 0x66, 0x0f, 0x3a, 0xdf, 0xca, 0x08 + call _key_expansion_256a + # aeskeygenassist $0x8, %xmm0, %xmm1 + .byte 0x66, 0x0f, 0x3a, 0xdf, 0xc8, 0x08 + call _key_expansion_256b + # aeskeygenassist $0x10, %xmm2, %xmm1 # round 5 + .byte 0x66, 0x0f, 0x3a, 0xdf, 0xca, 0x10 + call _key_expansion_256a + # aeskeygenassist $0x10, %xmm0, %xmm1 + .byte 0x66, 0x0f, 0x3a, 0xdf, 0xc8, 0x10 + call _key_expansion_256b + # aeskeygenassist $0x20, %xmm2, %xmm1 # round 6 + .byte 0x66, 0x0f, 0x3a, 0xdf, 0xca, 0x20 + call _key_expansion_256a + # aeskeygenassist $0x20, %xmm0, %xmm1 + .byte 0x66, 0x0f, 0x3a, 0xdf, 0xc8, 0x20 + call _key_expansion_256b + # aeskeygenassist $0x40, %xmm2, %xmm1 # round 7 + .byte 0x66, 0x0f, 0x3a, 0xdf, 0xca, 0x40 + call _key_expansion_256a + jmp .Ldec_key +.Lenc_key192: + movq 0x10(%rsi), %xmm2 # other user key + # aeskeygenassist $0x1, %xmm2, %xmm1 # round 1 + .byte 0x66, 0x0f, 0x3a, 0xdf, 0xca, 0x01 + call _key_expansion_192a + # aeskeygenassist $0x2, %xmm2, %xmm1 # round 2 + .byte 0x66, 0x0f, 0x3a, 0xdf, 0xca, 0x02 + call _key_expansion_192b + # aeskeygenassist $0x4, %xmm2, %xmm1 # round 3 + .byte 0x66, 0x0f, 0x3a, 0xdf, 0xca, 0x04 + call _key_expansion_192a + # aeskeygenassist $0x8, %xmm2, %xmm1 # round 4 + .byte 0x66, 0x0f, 0x3a, 0xdf, 0xca, 0x08 + call _key_expansion_192b + # aeskeygenassist $0x10, %xmm2, %xmm1 # round 5 + .byte 0x66, 0x0f, 0x3a, 0xdf, 0xca, 0x10 + call _key_expansion_192a + # aeskeygenassist $0x20, %xmm2, %xmm1 # round 6 + .byte 0x66, 0x0f, 0x3a, 0xdf, 0xca, 0x20 + call _key_expansion_192b + # aeskeygenassist $0x40, %xmm2, %xmm1 # round 7 + .byte 0x66, 0x0f, 0x3a, 0xdf, 0xca, 0x40 + call _key_expansion_192a + # aeskeygenassist $0x80, %xmm2, %xmm1 # round 8 + .byte 0x66, 0x0f, 0x3a, 0xdf, 0xca, 0x80 + call _key_expansion_192b + jmp .Ldec_key +.Lenc_key128: + # aeskeygenassist $0x1, %xmm0, %xmm1 # round 1 + .byte 0x66, 0x0f, 0x3a, 0xdf, 0xc8, 0x01 + call _key_expansion_128 + # aeskeygenassist $0x2, %xmm0, %xmm1 # round 2 + .byte 0x66, 0x0f, 0x3a, 0xdf, 0xc8, 0x02 + call _key_expansion_128 + # aeskeygenassist $0x4, %xmm0, %xmm1 # round 3 + .byte 0x66, 0x0f, 0x3a, 0xdf, 0xc8, 0x04 + call _key_expansion_128 + # aeskeygenassist $0x8, %xmm0, %xmm1 # round 4 + .byte 0x66, 0x0f, 0x3a, 0xdf, 0xc8, 0x08 + call _key_expansion_128 + # aeskeygenassist $0x10, %xmm0, %xmm1 # round 5 + .byte 0x66, 0x0f, 0x3a, 0xdf, 0xc8, 0x10 + call _key_expansion_128 + # aeskeygenassist $0x20, %xmm0, %xmm1 # round 6 + .byte 0x66, 0x0f, 0x3a, 0xdf, 0xc8, 0x20 + call _key_expansion_128 + # aeskeygenassist $0x40, %xmm0, %xmm1 # round 7 + .byte 0x66, 0x0f, 0x3a, 0xdf, 0xc8, 0x40 + call _key_expansion_128 + # aeskeygenassist $0x80, %xmm0, %xmm1 # round 8 + .byte 0x66, 0x0f, 0x3a, 0xdf, 0xc8, 0x80 + call _key_expansion_128 + # aeskeygenassist $0x1b, %xmm0, %xmm1 # round 9 + .byte 0x66, 0x0f, 0x3a, 0xdf, 0xc8, 0x1b + call _key_expansion_128 + # aeskeygenassist $0x36, %xmm0, %xmm1 # round 10 + .byte 0x66, 0x0f, 0x3a, 0xdf, 0xc8, 0x36 + call _key_expansion_128 +.Ldec_key: + sub $0x10, %rcx + movaps (%rdi), %xmm0 + movaps (%rcx), %xmm1 + movaps %xmm0, 240(%rcx) + movaps %xmm1, 240(%rdi) + add $0x10, %rdi + lea 240-16(%rcx), %rsi +.align 4 +.Ldec_key_loop: + movaps (%rdi), %xmm0 + # aesimc %xmm0, %xmm1 + .byte 0x66, 0x0f, 0x38, 0xdb, 0xc8 + movaps %xmm1, (%rsi) + add $0x10, %rdi + sub $0x10, %rsi + cmp %rcx, %rdi + jb .Ldec_key_loop + xor %rax, %rax + ret + +/* + * void aesni_enc(struct crypto_aes_ctx *ctx, u8 *dst, const u8 *src) + */ +ENTRY(aesni_enc) + movl 480(KEYP), KLEN # key length + movups (INP), STATE # input + call _aesni_enc1 + movups STATE, (OUTP) # output + ret + +/* + * _aesni_enc1: internal ABI + * input: + * KEYP: key struct pointer + * KLEN: round count + * STATE: initial state (input) + * output: + * STATE: finial state (output) + * changed: + * KEY + * TKEYP (T1) + */ +_aesni_enc1: + movaps (KEYP), KEY # key + mov KEYP, TKEYP + pxor KEY, STATE # round 0 + add $0x30, TKEYP + cmp $24, KLEN + jb .Lenc128 + lea 0x20(TKEYP), TKEYP + je .Lenc192 + add $0x20, TKEYP + movaps -0x60(TKEYP), KEY + # aesenc KEY, STATE + .byte 0x66, 0x0f, 0x38, 0xdc, 0xc2 + movaps -0x50(TKEYP), KEY + # aesenc KEY, STATE + .byte 0x66, 0x0f, 0x38, 0xdc, 0xc2 +.align 4 +.Lenc192: + movaps -0x40(TKEYP), KEY + # aesenc KEY, STATE + .byte 0x66, 0x0f, 0x38, 0xdc, 0xc2 + movaps -0x30(TKEYP), KEY + # aesenc KEY, STATE + .byte 0x66, 0x0f, 0x38, 0xdc, 0xc2 +.align 4 +.Lenc128: + movaps -0x20(TKEYP), KEY + # aesenc KEY, STATE + .byte 0x66, 0x0f, 0x38, 0xdc, 0xc2 + movaps -0x10(TKEYP), KEY + # aesenc KEY, STATE + .byte 0x66, 0x0f, 0x38, 0xdc, 0xc2 + movaps (TKEYP), KEY + # aesenc KEY, STATE + .byte 0x66, 0x0f, 0x38, 0xdc, 0xc2 + movaps 0x10(TKEYP), KEY + # aesenc KEY, STATE + .byte 0x66, 0x0f, 0x38, 0xdc, 0xc2 + movaps 0x20(TKEYP), KEY + # aesenc KEY, STATE + .byte 0x66, 0x0f, 0x38, 0xdc, 0xc2 + movaps 0x30(TKEYP), KEY + # aesenc KEY, STATE + .byte 0x66, 0x0f, 0x38, 0xdc, 0xc2 + movaps 0x40(TKEYP), KEY + # aesenc KEY, STATE + .byte 0x66, 0x0f, 0x38, 0xdc, 0xc2 + movaps 0x50(TKEYP), KEY + # aesenc KEY, STATE + .byte 0x66, 0x0f, 0x38, 0xdc, 0xc2 + movaps 0x60(TKEYP), KEY + # aesenc KEY, STATE + .byte 0x66, 0x0f, 0x38, 0xdc, 0xc2 + movaps 0x70(TKEYP), KEY + # aesenclast KEY, STATE # last round + .byte 0x66, 0x0f, 0x38, 0xdd, 0xc2 + ret + +/* + * _aesni_enc4: internal ABI + * input: + * KEYP: key struct pointer + * KLEN: round count + * STATE1: initial state (input) + * STATE2 + * STATE3 + * STATE4 + * output: + * STATE1: finial state (output) + * STATE2 + * STATE3 + * STATE4 + * changed: + * KEY + * TKEYP (T1) + */ +_aesni_enc4: + movaps (KEYP), KEY # key + mov KEYP, TKEYP + pxor KEY, STATE1 # round 0 + pxor KEY, STATE2 + pxor KEY, STATE3 + pxor KEY, STATE4 + add $0x30, TKEYP + cmp $24, KLEN + jb .L4enc128 + lea 0x20(TKEYP), TKEYP + je .L4enc192 + add $0x20, TKEYP + movaps -0x60(TKEYP), KEY + # aesenc KEY, STATE1 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xc2 + # aesenc KEY, STATE2 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xe2 + # aesenc KEY, STATE3 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xea + # aesenc KEY, STATE4 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xf2 + movaps -0x50(TKEYP), KEY + # aesenc KEY, STATE1 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xc2 + # aesenc KEY, STATE2 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xe2 + # aesenc KEY, STATE3 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xea + # aesenc KEY, STATE4 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xf2 +#.align 4 +.L4enc192: + movaps -0x40(TKEYP), KEY + # aesenc KEY, STATE1 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xc2 + # aesenc KEY, STATE2 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xe2 + # aesenc KEY, STATE3 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xea + # aesenc KEY, STATE4 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xf2 + movaps -0x30(TKEYP), KEY + # aesenc KEY, STATE1 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xc2 + # aesenc KEY, STATE2 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xe2 + # aesenc KEY, STATE3 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xea + # aesenc KEY, STATE4 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xf2 +#.align 4 +.L4enc128: + movaps -0x20(TKEYP), KEY + # aesenc KEY, STATE1 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xc2 + # aesenc KEY, STATE2 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xe2 + # aesenc KEY, STATE3 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xea + # aesenc KEY, STATE4 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xf2 + movaps -0x10(TKEYP), KEY + # aesenc KEY, STATE1 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xc2 + # aesenc KEY, STATE2 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xe2 + # aesenc KEY, STATE3 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xea + # aesenc KEY, STATE4 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xf2 + movaps (TKEYP), KEY + # aesenc KEY, STATE1 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xc2 + # aesenc KEY, STATE2 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xe2 + # aesenc KEY, STATE3 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xea + # aesenc KEY, STATE4 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xf2 + movaps 0x10(TKEYP), KEY + # aesenc KEY, STATE1 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xc2 + # aesenc KEY, STATE2 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xe2 + # aesenc KEY, STATE3 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xea + # aesenc KEY, STATE4 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xf2 + movaps 0x20(TKEYP), KEY + # aesenc KEY, STATE1 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xc2 + # aesenc KEY, STATE2 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xe2 + # aesenc KEY, STATE3 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xea + # aesenc KEY, STATE4 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xf2 + movaps 0x30(TKEYP), KEY + # aesenc KEY, STATE1 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xc2 + # aesenc KEY, STATE2 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xe2 + # aesenc KEY, STATE3 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xea + # aesenc KEY, STATE4 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xf2 + movaps 0x40(TKEYP), KEY + # aesenc KEY, STATE1 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xc2 + # aesenc KEY, STATE2 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xe2 + # aesenc KEY, STATE3 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xea + # aesenc KEY, STATE4 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xf2 + movaps 0x50(TKEYP), KEY + # aesenc KEY, STATE1 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xc2 + # aesenc KEY, STATE2 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xe2 + # aesenc KEY, STATE3 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xea + # aesenc KEY, STATE4 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xf2 + movaps 0x60(TKEYP), KEY + # aesenc KEY, STATE1 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xc2 + # aesenc KEY, STATE2 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xe2 + # aesenc KEY, STATE3 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xea + # aesenc KEY, STATE4 + .byte 0x66, 0x0f, 0x38, 0xdc, 0xf2 + movaps 0x70(TKEYP), KEY + # aesenclast KEY, STATE1 # last round + .byte 0x66, 0x0f, 0x38, 0xdd, 0xc2 + # aesenclast KEY, STATE2 + .byte 0x66, 0x0f, 0x38, 0xdd, 0xe2 + # aesenclast KEY, STATE3 + .byte 0x66, 0x0f, 0x38, 0xdd, 0xea + # aesenclast KEY, STATE4 + .byte 0x66, 0x0f, 0x38, 0xdd, 0xf2 + ret + +/* + * void aesni_dec (struct crypto_aes_ctx *ctx, u8 *dst, const u8 *src) + */ +ENTRY(aesni_dec) + mov 480(KEYP), KLEN # key length + add $240, KEYP + movups (INP), STATE # input + call _aesni_dec1 + movups STATE, (OUTP) #output + ret + +/* + * _aesni_dec1: internal ABI + * input: + * KEYP: key struct pointer + * KLEN: key length + * STATE: initial state (input) + * output: + * STATE: finial state (output) + * changed: + * KEY + * TKEYP (T1) + */ +_aesni_dec1: + movaps (KEYP), KEY # key + mov KEYP, TKEYP + pxor KEY, STATE # round 0 + add $0x30, TKEYP + cmp $24, KLEN + jb .Ldec128 + lea 0x20(TKEYP), TKEYP + je .Ldec192 + add $0x20, TKEYP + movaps -0x60(TKEYP), KEY + # aesdec KEY, STATE + .byte 0x66, 0x0f, 0x38, 0xde, 0xc2 + movaps -0x50(TKEYP), KEY + # aesdec KEY, STATE + .byte 0x66, 0x0f, 0x38, 0xde, 0xc2 +.align 4 +.Ldec192: + movaps -0x40(TKEYP), KEY + # aesdec KEY, STATE + .byte 0x66, 0x0f, 0x38, 0xde, 0xc2 + movaps -0x30(TKEYP), KEY + # aesdec KEY, STATE + .byte 0x66, 0x0f, 0x38, 0xde, 0xc2 +.align 4 +.Ldec128: + movaps -0x20(TKEYP), KEY + # aesdec KEY, STATE + .byte 0x66, 0x0f, 0x38, 0xde, 0xc2 + movaps -0x10(TKEYP), KEY + # aesdec KEY, STATE + .byte 0x66, 0x0f, 0x38, 0xde, 0xc2 + movaps (TKEYP), KEY + # aesdec KEY, STATE + .byte 0x66, 0x0f, 0x38, 0xde, 0xc2 + movaps 0x10(TKEYP), KEY + # aesdec KEY, STATE + .byte 0x66, 0x0f, 0x38, 0xde, 0xc2 + movaps 0x20(TKEYP), KEY + # aesdec KEY, STATE + .byte 0x66, 0x0f, 0x38, 0xde, 0xc2 + movaps 0x30(TKEYP), KEY + # aesdec KEY, STATE + .byte 0x66, 0x0f, 0x38, 0xde, 0xc2 + movaps 0x40(TKEYP), KEY + # aesdec KEY, STATE + .byte 0x66, 0x0f, 0x38, 0xde, 0xc2 + movaps 0x50(TKEYP), KEY + # aesdec KEY, STATE + .byte 0x66, 0x0f, 0x38, 0xde, 0xc2 + movaps 0x60(TKEYP), KEY + # aesdec KEY, STATE + .byte 0x66, 0x0f, 0x38, 0xde, 0xc2 + movaps 0x70(TKEYP), KEY + # aesdeclast KEY, STATE # last round + .byte 0x66, 0x0f, 0x38, 0xdf, 0xc2 + ret + +/* + * _aesni_dec4: internal ABI + * input: + * KEYP: key struct pointer + * KLEN: key length + * STATE1: initial state (input) + * STATE2 + * STATE3 + * STATE4 + * output: + * STATE1: finial state (output) + * STATE2 + * STATE3 + * STATE4 + * changed: + * KEY + * TKEYP (T1) + */ +_aesni_dec4: + movaps (KEYP), KEY # key + mov KEYP, TKEYP + pxor KEY, STATE1 # round 0 + pxor KEY, STATE2 + pxor KEY, STATE3 + pxor KEY, STATE4 + add $0x30, TKEYP + cmp $24, KLEN + jb .L4dec128 + lea 0x20(TKEYP), TKEYP + je .L4dec192 + add $0x20, TKEYP + movaps -0x60(TKEYP), KEY + # aesdec KEY, STATE1 + .byte 0x66, 0x0f, 0x38, 0xde, 0xc2 + # aesdec KEY, STATE2 + .byte 0x66, 0x0f, 0x38, 0xde, 0xe2 + # aesdec KEY, STATE3 + .byte 0x66, 0x0f, 0x38, 0xde, 0xea + # aesdec KEY, STATE4 + .byte 0x66, 0x0f, 0x38, 0xde, 0xf2 + movaps -0x50(TKEYP), KEY + # aesdec KEY, STATE1 + .byte 0x66, 0x0f, 0x38, 0xde, 0xc2 + # aesdec KEY, STATE2 + .byte 0x66, 0x0f, 0x38, 0xde, 0xe2 + # aesdec KEY, STATE3 + .byte 0x66, 0x0f, 0x38, 0xde, 0xea + # aesdec KEY, STATE4 + .byte 0x66, 0x0f, 0x38, 0xde, 0xf2 +.align 4 +.L4dec192: + movaps -0x40(TKEYP), KEY + # aesdec KEY, STATE1 + .byte 0x66, 0x0f, 0x38, 0xde, 0xc2 + # aesdec KEY, STATE2 + .byte 0x66, 0x0f, 0x38, 0xde, 0xe2 + # aesdec KEY, STATE3 + .byte 0x66, 0x0f, 0x38, 0xde, 0xea + # aesdec KEY, STATE4 + .byte 0x66, 0x0f, 0x38, 0xde, 0xf2 + movaps -0x30(TKEYP), KEY + # aesdec KEY, STATE1 + .byte 0x66, 0x0f, 0x38, 0xde, 0xc2 + # aesdec KEY, STATE2 + .byte 0x66, 0x0f, 0x38, 0xde, 0xe2 + # aesdec KEY, STATE3 + .byte 0x66, 0x0f, 0x38, 0xde, 0xea + # aesdec KEY, STATE4 + .byte 0x66, 0x0f, 0x38, 0xde, 0xf2 +.align 4 +.L4dec128: + movaps -0x20(TKEYP), KEY + # aesdec KEY, STATE1 + .byte 0x66, 0x0f, 0x38, 0xde, 0xc2 + # aesdec KEY, STATE2 + .byte 0x66, 0x0f, 0x38, 0xde, 0xe2 + # aesdec KEY, STATE3 + .byte 0x66, 0x0f, 0x38, 0xde, 0xea + # aesdec KEY, STATE4 + .byte 0x66, 0x0f, 0x38, 0xde, 0xf2 + movaps -0x10(TKEYP), KEY + # aesdec KEY, STATE1 + .byte 0x66, 0x0f, 0x38, 0xde, 0xc2 + # aesdec KEY, STATE2 + .byte 0x66, 0x0f, 0x38, 0xde, 0xe2 + # aesdec KEY, STATE3 + .byte 0x66, 0x0f, 0x38, 0xde, 0xea + # aesdec KEY, STATE4 + .byte 0x66, 0x0f, 0x38, 0xde, 0xf2 + movaps (TKEYP), KEY + # aesdec KEY, STATE1 + .byte 0x66, 0x0f, 0x38, 0xde, 0xc2 + # aesdec KEY, STATE2 + .byte 0x66, 0x0f, 0x38, 0xde, 0xe2 + # aesdec KEY, STATE3 + .byte 0x66, 0x0f, 0x38, 0xde, 0xea + # aesdec KEY, STATE4 + .byte 0x66, 0x0f, 0x38, 0xde, 0xf2 + movaps 0x10(TKEYP), KEY + # aesdec KEY, STATE1 + .byte 0x66, 0x0f, 0x38, 0xde, 0xc2 + # aesdec KEY, STATE2 + .byte 0x66, 0x0f, 0x38, 0xde, 0xe2 + # aesdec KEY, STATE3 + .byte 0x66, 0x0f, 0x38, 0xde, 0xea + # aesdec KEY, STATE4 + .byte 0x66, 0x0f, 0x38, 0xde, 0xf2 + movaps 0x20(TKEYP), KEY + # aesdec KEY, STATE1 + .byte 0x66, 0x0f, 0x38, 0xde, 0xc2 + # aesdec KEY, STATE2 + .byte 0x66, 0x0f, 0x38, 0xde, 0xe2 + # aesdec KEY, STATE3 + .byte 0x66, 0x0f, 0x38, 0xde, 0xea + # aesdec KEY, STATE4 + .byte 0x66, 0x0f, 0x38, 0xde, 0xf2 + movaps 0x30(TKEYP), KEY + # aesdec KEY, STATE1 + .byte 0x66, 0x0f, 0x38, 0xde, 0xc2 + # aesdec KEY, STATE2 + .byte 0x66, 0x0f, 0x38, 0xde, 0xe2 + # aesdec KEY, STATE3 + .byte 0x66, 0x0f, 0x38, 0xde, 0xea + # aesdec KEY, STATE4 + .byte 0x66, 0x0f, 0x38, 0xde, 0xf2 + movaps 0x40(TKEYP), KEY + # aesdec KEY, STATE1 + .byte 0x66, 0x0f, 0x38, 0xde, 0xc2 + # aesdec KEY, STATE2 + .byte 0x66, 0x0f, 0x38, 0xde, 0xe2 + # aesdec KEY, STATE3 + .byte 0x66, 0x0f, 0x38, 0xde, 0xea + # aesdec KEY, STATE4 + .byte 0x66, 0x0f, 0x38, 0xde, 0xf2 + movaps 0x50(TKEYP), KEY + # aesdec KEY, STATE1 + .byte 0x66, 0x0f, 0x38, 0xde, 0xc2 + # aesdec KEY, STATE2 + .byte 0x66, 0x0f, 0x38, 0xde, 0xe2 + # aesdec KEY, STATE3 + .byte 0x66, 0x0f, 0x38, 0xde, 0xea + # aesdec KEY, STATE4 + .byte 0x66, 0x0f, 0x38, 0xde, 0xf2 + movaps 0x60(TKEYP), KEY + # aesdec KEY, STATE1 + .byte 0x66, 0x0f, 0x38, 0xde, 0xc2 + # aesdec KEY, STATE2 + .byte 0x66, 0x0f, 0x38, 0xde, 0xe2 + # aesdec KEY, STATE3 + .byte 0x66, 0x0f, 0x38, 0xde, 0xea + # aesdec KEY, STATE4 + .byte 0x66, 0x0f, 0x38, 0xde, 0xf2 + movaps 0x70(TKEYP), KEY + # aesdeclast KEY, STATE1 # last round + .byte 0x66, 0x0f, 0x38, 0xdf, 0xc2 + # aesdeclast KEY, STATE2 + .byte 0x66, 0x0f, 0x38, 0xdf, 0xe2 + # aesdeclast KEY, STATE3 + .byte 0x66, 0x0f, 0x38, 0xdf, 0xea + # aesdeclast KEY, STATE4 + .byte 0x66, 0x0f, 0x38, 0xdf, 0xf2 + ret + +/* + * void aesni_ecb_enc(struct crypto_aes_ctx *ctx, const u8 *dst, u8 *src, + * size_t len) + */ +ENTRY(aesni_ecb_enc) + test LEN, LEN # check length + jz .Lecb_enc_ret + mov 480(KEYP), KLEN + cmp $16, LEN + jb .Lecb_enc_ret + cmp $64, LEN + jb .Lecb_enc_loop1 +.align 4 +.Lecb_enc_loop4: + movups (INP), STATE1 + movups 0x10(INP), STATE2 + movups 0x20(INP), STATE3 + movups 0x30(INP), STATE4 + call _aesni_enc4 + movups STATE1, (OUTP) + movups STATE2, 0x10(OUTP) + movups STATE3, 0x20(OUTP) + movups STATE4, 0x30(OUTP) + sub $64, LEN + add $64, INP + add $64, OUTP + cmp $64, LEN + jge .Lecb_enc_loop4 + cmp $16, LEN + jb .Lecb_enc_ret +.align 4 +.Lecb_enc_loop1: + movups (INP), STATE1 + call _aesni_enc1 + movups STATE1, (OUTP) + sub $16, LEN + add $16, INP + add $16, OUTP + cmp $16, LEN + jge .Lecb_enc_loop1 +.Lecb_enc_ret: + ret + +/* + * void aesni_ecb_dec(struct crypto_aes_ctx *ctx, const u8 *dst, u8 *src, + * size_t len); + */ +ENTRY(aesni_ecb_dec) + test LEN, LEN + jz .Lecb_dec_ret + mov 480(KEYP), KLEN + add $240, KEYP + cmp $16, LEN + jb .Lecb_dec_ret + cmp $64, LEN + jb .Lecb_dec_loop1 +.align 4 +.Lecb_dec_loop4: + movups (INP), STATE1 + movups 0x10(INP), STATE2 + movups 0x20(INP), STATE3 + movups 0x30(INP), STATE4 + call _aesni_dec4 + movups STATE1, (OUTP) + movups STATE2, 0x10(OUTP) + movups STATE3, 0x20(OUTP) + movups STATE4, 0x30(OUTP) + sub $64, LEN + add $64, INP + add $64, OUTP + cmp $64, LEN + jge .Lecb_dec_loop4 + cmp $16, LEN + jb .Lecb_dec_ret +.align 4 +.Lecb_dec_loop1: + movups (INP), STATE1 + call _aesni_dec1 + movups STATE1, (OUTP) + sub $16, LEN + add $16, INP + add $16, OUTP + cmp $16, LEN + jge .Lecb_dec_loop1 +.Lecb_dec_ret: + ret + +/* + * void aesni_cbc_enc(struct crypto_aes_ctx *ctx, const u8 *dst, u8 *src, + * size_t len, u8 *iv) + */ +ENTRY(aesni_cbc_enc) + cmp $16, LEN + jb .Lcbc_enc_ret + mov 480(KEYP), KLEN + movups (IVP), STATE # load iv as initial state +.align 4 +.Lcbc_enc_loop: + movups (INP), IN # load input + pxor IN, STATE + call _aesni_enc1 + movups STATE, (OUTP) # store output + sub $16, LEN + add $16, INP + add $16, OUTP + cmp $16, LEN + jge .Lcbc_enc_loop + movups STATE, (IVP) +.Lcbc_enc_ret: + ret + +/* + * void aesni_cbc_dec(struct crypto_aes_ctx *ctx, const u8 *dst, u8 *src, + * size_t len, u8 *iv) + */ +ENTRY(aesni_cbc_dec) + cmp $16, LEN + jb .Lcbc_dec_ret + mov 480(KEYP), KLEN + add $240, KEYP + movups (IVP), IV + cmp $64, LEN + jb .Lcbc_dec_loop1 +.align 4 +.Lcbc_dec_loop4: + movups (INP), IN1 + movaps IN1, STATE1 + movups 0x10(INP), IN2 + movaps IN2, STATE2 + movups 0x20(INP), IN3 + movaps IN3, STATE3 + movups 0x30(INP), IN4 + movaps IN4, STATE4 + call _aesni_dec4 + pxor IV, STATE1 + pxor IN1, STATE2 + pxor IN2, STATE3 + pxor IN3, STATE4 + movaps IN4, IV + movups STATE1, (OUTP) + movups STATE2, 0x10(OUTP) + movups STATE3, 0x20(OUTP) + movups STATE4, 0x30(OUTP) + sub $64, LEN + add $64, INP + add $64, OUTP + cmp $64, LEN + jge .Lcbc_dec_loop4 + cmp $16, LEN + jb .Lcbc_dec_ret +.align 4 +.Lcbc_dec_loop1: + movups (INP), IN + movaps IN, STATE + call _aesni_dec1 + pxor IV, STATE + movups STATE, (OUTP) + movaps IN, IV + sub $16, LEN + add $16, INP + add $16, OUTP + cmp $16, LEN + jge .Lcbc_dec_loop1 + movups IV, (IVP) +.Lcbc_dec_ret: + ret diff --git a/arch/x86/crypto/aesni-intel_glue.c b/arch/x86/crypto/aesni-intel_glue.c new file mode 100644 index 00000000000..02af0af6549 --- /dev/null +++ b/arch/x86/crypto/aesni-intel_glue.c @@ -0,0 +1,461 @@ +/* + * Support for Intel AES-NI instructions. This file contains glue + * code, the real AES implementation is in intel-aes_asm.S. + * + * Copyright (C) 2008, Intel Corp. + * Author: Huang Ying <ying.huang@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. + */ + +#include <linux/hardirq.h> +#include <linux/types.h> +#include <linux/crypto.h> +#include <linux/err.h> +#include <crypto/algapi.h> +#include <crypto/aes.h> +#include <crypto/cryptd.h> +#include <asm/i387.h> +#include <asm/aes.h> + +struct async_aes_ctx { + struct cryptd_ablkcipher *cryptd_tfm; +}; + +#define AESNI_ALIGN 16 +#define AES_BLOCK_MASK (~(AES_BLOCK_SIZE-1)) + +asmlinkage int aesni_set_key(struct crypto_aes_ctx *ctx, const u8 *in_key, + unsigned int key_len); +asmlinkage void aesni_enc(struct crypto_aes_ctx *ctx, u8 *out, + const u8 *in); +asmlinkage void aesni_dec(struct crypto_aes_ctx *ctx, u8 *out, + const u8 *in); +asmlinkage void aesni_ecb_enc(struct crypto_aes_ctx *ctx, u8 *out, + const u8 *in, unsigned int len); +asmlinkage void aesni_ecb_dec(struct crypto_aes_ctx *ctx, u8 *out, + const u8 *in, unsigned int len); +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); + +static inline int kernel_fpu_using(void) +{ + if (in_interrupt() && !(read_cr0() & X86_CR0_TS)) + return 1; + return 0; +} + +static inline struct crypto_aes_ctx *aes_ctx(void *raw_ctx) +{ + unsigned long addr = (unsigned long)raw_ctx; + unsigned long align = AESNI_ALIGN; + + if (align <= crypto_tfm_ctx_alignment()) + align = 1; + return (struct crypto_aes_ctx *)ALIGN(addr, align); +} + +static int aes_set_key_common(struct crypto_tfm *tfm, void *raw_ctx, + const u8 *in_key, unsigned int key_len) +{ + struct crypto_aes_ctx *ctx = aes_ctx(raw_ctx); + u32 *flags = &tfm->crt_flags; + int err; + + if (key_len != AES_KEYSIZE_128 && key_len != AES_KEYSIZE_192 && + key_len != AES_KEYSIZE_256) { + *flags |= CRYPTO_TFM_RES_BAD_KEY_LEN; + return -EINVAL; + } + + if (kernel_fpu_using()) + err = crypto_aes_expand_key(ctx, in_key, key_len); + else { + kernel_fpu_begin(); + err = aesni_set_key(ctx, in_key, key_len); + kernel_fpu_end(); + } + + return err; +} + +static int aes_set_key(struct crypto_tfm *tfm, const u8 *in_key, + unsigned int key_len) +{ + return aes_set_key_common(tfm, crypto_tfm_ctx(tfm), in_key, key_len); +} + +static void aes_encrypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src) +{ + struct crypto_aes_ctx *ctx = aes_ctx(crypto_tfm_ctx(tfm)); + + if (kernel_fpu_using()) + crypto_aes_encrypt_x86(ctx, dst, src); + else { + kernel_fpu_begin(); + aesni_enc(ctx, dst, src); + kernel_fpu_end(); + } +} + +static void aes_decrypt(struct crypto_tfm *tfm, u8 *dst, const u8 *src) +{ + struct crypto_aes_ctx *ctx = aes_ctx(crypto_tfm_ctx(tfm)); + + if (kernel_fpu_using()) + crypto_aes_decrypt_x86(ctx, dst, src); + else { + kernel_fpu_begin(); + aesni_dec(ctx, dst, src); + kernel_fpu_end(); + } +} + +static struct crypto_alg aesni_alg = { + .cra_name = "aes", + .cra_driver_name = "aes-aesni", + .cra_priority = 300, + .cra_flags = CRYPTO_ALG_TYPE_CIPHER, + .cra_blocksize = AES_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct crypto_aes_ctx)+AESNI_ALIGN-1, + .cra_alignmask = 0, + .cra_module = THIS_MODULE, + .cra_list = LIST_HEAD_INIT(aesni_alg.cra_list), + .cra_u = { + .cipher = { + .cia_min_keysize = AES_MIN_KEY_SIZE, + .cia_max_keysize = AES_MAX_KEY_SIZE, + .cia_setkey = aes_set_key, + .cia_encrypt = aes_encrypt, + .cia_decrypt = aes_decrypt + } + } +}; + +static int ecb_encrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct crypto_aes_ctx *ctx = aes_ctx(crypto_blkcipher_ctx(desc->tfm)); + struct blkcipher_walk walk; + int err; + + blkcipher_walk_init(&walk, dst, src, nbytes); + err = blkcipher_walk_virt(desc, &walk); + + kernel_fpu_begin(); + while ((nbytes = walk.nbytes)) { + aesni_ecb_enc(ctx, walk.dst.virt.addr, walk.src.virt.addr, + nbytes & AES_BLOCK_MASK); + nbytes &= AES_BLOCK_SIZE - 1; + err = blkcipher_walk_done(desc, &walk, nbytes); + } + kernel_fpu_end(); + + return err; +} + +static int ecb_decrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct crypto_aes_ctx *ctx = aes_ctx(crypto_blkcipher_ctx(desc->tfm)); + struct blkcipher_walk walk; + int err; + + blkcipher_walk_init(&walk, dst, src, nbytes); + err = blkcipher_walk_virt(desc, &walk); + + kernel_fpu_begin(); + while ((nbytes = walk.nbytes)) { + aesni_ecb_dec(ctx, walk.dst.virt.addr, walk.src.virt.addr, + nbytes & AES_BLOCK_MASK); + nbytes &= AES_BLOCK_SIZE - 1; + err = blkcipher_walk_done(desc, &walk, nbytes); + } + kernel_fpu_end(); + + return err; +} + +static struct crypto_alg blk_ecb_alg = { + .cra_name = "__ecb-aes-aesni", + .cra_driver_name = "__driver-ecb-aes-aesni", + .cra_priority = 0, + .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER, + .cra_blocksize = AES_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct crypto_aes_ctx)+AESNI_ALIGN-1, + .cra_alignmask = 0, + .cra_type = &crypto_blkcipher_type, + .cra_module = THIS_MODULE, + .cra_list = LIST_HEAD_INIT(blk_ecb_alg.cra_list), + .cra_u = { + .blkcipher = { + .min_keysize = AES_MIN_KEY_SIZE, + .max_keysize = AES_MAX_KEY_SIZE, + .setkey = aes_set_key, + .encrypt = ecb_encrypt, + .decrypt = ecb_decrypt, + }, + }, +}; + +static int cbc_encrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct crypto_aes_ctx *ctx = aes_ctx(crypto_blkcipher_ctx(desc->tfm)); + struct blkcipher_walk walk; + int err; + + blkcipher_walk_init(&walk, dst, src, nbytes); + err = blkcipher_walk_virt(desc, &walk); + + kernel_fpu_begin(); + while ((nbytes = walk.nbytes)) { + aesni_cbc_enc(ctx, walk.dst.virt.addr, walk.src.virt.addr, + nbytes & AES_BLOCK_MASK, walk.iv); + nbytes &= AES_BLOCK_SIZE - 1; + err = blkcipher_walk_done(desc, &walk, nbytes); + } + kernel_fpu_end(); + + return err; +} + +static int cbc_decrypt(struct blkcipher_desc *desc, + struct scatterlist *dst, struct scatterlist *src, + unsigned int nbytes) +{ + struct crypto_aes_ctx *ctx = aes_ctx(crypto_blkcipher_ctx(desc->tfm)); + struct blkcipher_walk walk; + int err; + + blkcipher_walk_init(&walk, dst, src, nbytes); + err = blkcipher_walk_virt(desc, &walk); + + kernel_fpu_begin(); + while ((nbytes = walk.nbytes)) { + aesni_cbc_dec(ctx, walk.dst.virt.addr, walk.src.virt.addr, + nbytes & AES_BLOCK_MASK, walk.iv); + nbytes &= AES_BLOCK_SIZE - 1; + err = blkcipher_walk_done(desc, &walk, nbytes); + } + kernel_fpu_end(); + + return err; +} + +static struct crypto_alg blk_cbc_alg = { + .cra_name = "__cbc-aes-aesni", + .cra_driver_name = "__driver-cbc-aes-aesni", + .cra_priority = 0, + .cra_flags = CRYPTO_ALG_TYPE_BLKCIPHER, + .cra_blocksize = AES_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct crypto_aes_ctx)+AESNI_ALIGN-1, + .cra_alignmask = 0, + .cra_type = &crypto_blkcipher_type, + .cra_module = THIS_MODULE, + .cra_list = LIST_HEAD_INIT(blk_cbc_alg.cra_list), + .cra_u = { + .blkcipher = { + .min_keysize = AES_MIN_KEY_SIZE, + .max_keysize = AES_MAX_KEY_SIZE, + .setkey = aes_set_key, + .encrypt = cbc_encrypt, + .decrypt = cbc_decrypt, + }, + }, +}; + +static int ablk_set_key(struct crypto_ablkcipher *tfm, const u8 *key, + unsigned int key_len) +{ + struct async_aes_ctx *ctx = crypto_ablkcipher_ctx(tfm); + + return crypto_ablkcipher_setkey(&ctx->cryptd_tfm->base, key, key_len); +} + +static int ablk_encrypt(struct ablkcipher_request *req) +{ + struct crypto_ablkcipher *tfm = crypto_ablkcipher_reqtfm(req); + struct async_aes_ctx *ctx = crypto_ablkcipher_ctx(tfm); + + if (kernel_fpu_using()) { + struct ablkcipher_request *cryptd_req = + ablkcipher_request_ctx(req); + memcpy(cryptd_req, req, sizeof(*req)); + ablkcipher_request_set_tfm(cryptd_req, &ctx->cryptd_tfm->base); + return crypto_ablkcipher_encrypt(cryptd_req); + } else { + struct blkcipher_desc desc; + desc.tfm = cryptd_ablkcipher_child(ctx->cryptd_tfm); + desc.info = req->info; + desc.flags = 0; + return crypto_blkcipher_crt(desc.tfm)->encrypt( + &desc, req->dst, req->src, req->nbytes); + } +} + +static int ablk_decrypt(struct ablkcipher_request *req) +{ + struct crypto_ablkcipher *tfm = crypto_ablkcipher_reqtfm(req); + struct async_aes_ctx *ctx = crypto_ablkcipher_ctx(tfm); + + if (kernel_fpu_using()) { + struct ablkcipher_request *cryptd_req = + ablkcipher_request_ctx(req); + memcpy(cryptd_req, req, sizeof(*req)); + ablkcipher_request_set_tfm(cryptd_req, &ctx->cryptd_tfm->base); + return crypto_ablkcipher_decrypt(cryptd_req); + } else { + struct blkcipher_desc desc; + desc.tfm = cryptd_ablkcipher_child(ctx->cryptd_tfm); + desc.info = req->info; + desc.flags = 0; + return crypto_blkcipher_crt(desc.tfm)->decrypt( + &desc, req->dst, req->src, req->nbytes); + } +} + +static void ablk_exit(struct crypto_tfm *tfm) +{ + struct async_aes_ctx *ctx = crypto_tfm_ctx(tfm); + + cryptd_free_ablkcipher(ctx->cryptd_tfm); +} + +static void ablk_init_common(struct crypto_tfm *tfm, + struct cryptd_ablkcipher *cryptd_tfm) +{ + struct async_aes_ctx *ctx = crypto_tfm_ctx(tfm); + + ctx->cryptd_tfm = cryptd_tfm; + tfm->crt_ablkcipher.reqsize = sizeof(struct ablkcipher_request) + + crypto_ablkcipher_reqsize(&cryptd_tfm->base); +} + +static int ablk_ecb_init(struct crypto_tfm *tfm) +{ + struct cryptd_ablkcipher *cryptd_tfm; + + cryptd_tfm = cryptd_alloc_ablkcipher("__driver-ecb-aes-aesni", 0, 0); + if (IS_ERR(cryptd_tfm)) + return PTR_ERR(cryptd_tfm); + ablk_init_common(tfm, cryptd_tfm); + return 0; +} + +static struct crypto_alg ablk_ecb_alg = { + .cra_name = "ecb(aes)", + .cra_driver_name = "ecb-aes-aesni", + .cra_priority = 400, + .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER|CRYPTO_ALG_ASYNC, + .cra_blocksize = AES_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct async_aes_ctx), + .cra_alignmask = 0, + .cra_type = &crypto_ablkcipher_type, + .cra_module = THIS_MODULE, + .cra_list = LIST_HEAD_INIT(ablk_ecb_alg.cra_list), + .cra_init = ablk_ecb_init, + .cra_exit = ablk_exit, + .cra_u = { + .ablkcipher = { + .min_keysize = AES_MIN_KEY_SIZE, + .max_keysize = AES_MAX_KEY_SIZE, + .setkey = ablk_set_key, + .encrypt = ablk_encrypt, + .decrypt = ablk_decrypt, + }, + }, +}; + +static int ablk_cbc_init(struct crypto_tfm *tfm) +{ + struct cryptd_ablkcipher *cryptd_tfm; + + cryptd_tfm = cryptd_alloc_ablkcipher("__driver-cbc-aes-aesni", 0, 0); + if (IS_ERR(cryptd_tfm)) + return PTR_ERR(cryptd_tfm); + ablk_init_common(tfm, cryptd_tfm); + return 0; +} + +static struct crypto_alg ablk_cbc_alg = { + .cra_name = "cbc(aes)", + .cra_driver_name = "cbc-aes-aesni", + .cra_priority = 400, + .cra_flags = CRYPTO_ALG_TYPE_ABLKCIPHER|CRYPTO_ALG_ASYNC, + .cra_blocksize = AES_BLOCK_SIZE, + .cra_ctxsize = sizeof(struct async_aes_ctx), + .cra_alignmask = 0, + .cra_type = &crypto_ablkcipher_type, + .cra_module = THIS_MODULE, + .cra_list = LIST_HEAD_INIT(ablk_cbc_alg.cra_list), + .cra_init = ablk_cbc_init, + .cra_exit = ablk_exit, + .cra_u = { + .ablkcipher = { + .min_keysize = AES_MIN_KEY_SIZE, + .max_keysize = AES_MAX_KEY_SIZE, + .ivsize = AES_BLOCK_SIZE, + .setkey = ablk_set_key, + .encrypt = ablk_encrypt, + .decrypt = ablk_decrypt, + }, + }, +}; + +static int __init aesni_init(void) +{ + int err; + + if (!cpu_has_aes) { + printk(KERN_ERR "Intel AES-NI instructions are not detected.\n"); + return -ENODEV; + } + if ((err = crypto_register_alg(&aesni_alg))) + goto aes_err; + if ((err = crypto_register_alg(&blk_ecb_alg))) + goto blk_ecb_err; + if ((err = crypto_register_alg(&blk_cbc_alg))) + goto blk_cbc_err; + if ((err = crypto_register_alg(&ablk_ecb_alg))) + goto ablk_ecb_err; + if ((err = crypto_register_alg(&ablk_cbc_alg))) + goto ablk_cbc_err; + + return err; + +ablk_cbc_err: + crypto_unregister_alg(&ablk_ecb_alg); +ablk_ecb_err: + crypto_unregister_alg(&blk_cbc_alg); +blk_cbc_err: + crypto_unregister_alg(&blk_ecb_alg); +blk_ecb_err: + crypto_unregister_alg(&aesni_alg); +aes_err: + return err; +} + +static void __exit aesni_exit(void) +{ + crypto_unregister_alg(&ablk_cbc_alg); + crypto_unregister_alg(&ablk_ecb_alg); + crypto_unregister_alg(&blk_cbc_alg); + crypto_unregister_alg(&blk_ecb_alg); + crypto_unregister_alg(&aesni_alg); +} + +module_init(aesni_init); +module_exit(aesni_exit); + +MODULE_DESCRIPTION("Rijndael (AES) Cipher Algorithm, Intel AES-NI instructions optimized"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("aes"); diff --git a/arch/x86/ia32/ia32entry.S b/arch/x86/ia32/ia32entry.S index 256b00b6189..5a0d76dc56a 100644 --- a/arch/x86/ia32/ia32entry.S +++ b/arch/x86/ia32/ia32entry.S @@ -418,9 +418,9 @@ ENTRY(ia32_syscall) orl $TS_COMPAT,TI_status(%r10) testl $_TIF_WORK_SYSCALL_ENTRY,TI_flags(%r10) jnz ia32_tracesys -ia32_do_syscall: cmpl $(IA32_NR_syscalls-1),%eax - ja int_ret_from_sys_call /* ia32_tracesys has set RAX(%rsp) */ + ja ia32_badsys +ia32_do_call: IA32_ARG_FIXUP call *ia32_sys_call_table(,%rax,8) # xxx: rip relative ia32_sysret: @@ -435,7 +435,9 @@ ia32_tracesys: call syscall_trace_enter LOAD_ARGS32 ARGOFFSET /* reload args from stack in case ptrace changed it */ RESTORE_REST - jmp ia32_do_syscall + cmpl $(IA32_NR_syscalls-1),%eax + ja int_ret_from_sys_call /* ia32_tracesys has set RAX(%rsp) */ + jmp ia32_do_call END(ia32_syscall) ia32_badsys: diff --git a/arch/x86/include/asm/a.out-core.h b/arch/x86/include/asm/a.out-core.h index 37822206083..3c601f8224b 100644 --- a/arch/x86/include/asm/a.out-core.h +++ b/arch/x86/include/asm/a.out-core.h @@ -23,8 +23,6 @@ */ static inline void aout_dump_thread(struct pt_regs *regs, struct user *dump) { - u16 gs; - /* changed the size calculations - should hopefully work better. lbt */ dump->magic = CMAGIC; dump->start_code = 0; @@ -57,7 +55,7 @@ static inline void aout_dump_thread(struct pt_regs *regs, struct user *dump) dump->regs.ds = (u16)regs->ds; dump->regs.es = (u16)regs->es; dump->regs.fs = (u16)regs->fs; - savesegment(gs, gs); + savesegment(gs, dump->regs.gs); dump->regs.orig_ax = regs->orig_ax; dump->regs.ip = regs->ip; dump->regs.cs = (u16)regs->cs; diff --git a/arch/x86/include/asm/aes.h b/arch/x86/include/asm/aes.h new file mode 100644 index 00000000000..80545a1cbe3 --- /dev/null +++ b/arch/x86/include/asm/aes.h @@ -0,0 +1,11 @@ +#ifndef ASM_X86_AES_H +#define ASM_X86_AES_H + +#include <linux/crypto.h> +#include <crypto/aes.h> + +void crypto_aes_encrypt_x86(struct crypto_aes_ctx *ctx, u8 *dst, + const u8 *src); +void crypto_aes_decrypt_x86(struct crypto_aes_ctx *ctx, u8 *dst, + const u8 *src); +#endif diff --git a/arch/x86/include/asm/bitops.h b/arch/x86/include/asm/bitops.h index e02a359d2aa..02b47a603fc 100644 --- a/arch/x86/include/asm/bitops.h +++ b/arch/x86/include/asm/bitops.h @@ -3,6 +3,9 @@ /* * Copyright 1992, Linus Torvalds. + * + * Note: inlines with more than a single statement should be marked + * __always_inline to avoid problems with older gcc's inlining heuristics. */ #ifndef _LINUX_BITOPS_H @@ -53,7 +56,8 @@ * Note that @nr may be almost arbitrarily large; this function is not * restricted to acting on a single-word quantity. */ -static inline void set_bit(unsigned int nr, volatile unsigned long *addr) +static __always_inline void +set_bit(unsigned int nr, volatile unsigned long *addr) { if (IS_IMMEDIATE(nr)) { asm volatile(LOCK_PREFIX "orb %1,%0" @@ -90,7 +94,8 @@ static inline void __set_bit(int nr, volatile unsigned long *addr) * you should call smp_mb__before_clear_bit() and/or smp_mb__after_clear_bit() * in order to ensure changes are visible on other processors. */ -static inline void clear_bit(int nr, volatile unsigned long *addr) +static __always_inline void +clear_bit(int nr, volatile unsigned long *addr) { if (IS_IMMEDIATE(nr)) { asm volatile(LOCK_PREFIX "andb %1,%0" @@ -204,7 +209,8 @@ static inline int test_and_set_bit(int nr, volatile unsigned long *addr) * * This is the same as test_and_set_bit on x86. */ -static inline int test_and_set_bit_lock(int nr, volatile unsigned long *addr) +static __always_inline int +test_and_set_bit_lock(int nr, volatile unsigned long *addr) { return test_and_set_bit(nr, addr); } @@ -300,7 +306,7 @@ static inline int test_and_change_bit(int nr, volatile unsigned long *addr) return oldbit; } -static inline int constant_test_bit(unsigned int nr, const volatile unsigned long *addr) +static __always_inline int constant_test_bit(unsigned int nr, const volatile unsigned long *addr) { return ((1UL << (nr % BITS_PER_LONG)) & (((unsigned long *)addr)[nr / BITS_PER_LONG])) != 0; diff --git a/arch/x86/include/asm/cpufeature.h b/arch/x86/include/asm/cpufeature.h index ea408dcba51..0beba0d1468 100644 --- a/arch/x86/include/asm/cpufeature.h +++ b/arch/x86/include/asm/cpufeature.h @@ -93,6 +93,7 @@ #define X86_FEATURE_XTOPOLOGY (3*32+22) /* cpu topology enum extensions */ #define X86_FEATURE_TSC_RELIABLE (3*32+23) /* TSC is known to be reliable */ #define X86_FEATURE_NONSTOP_TSC (3*32+24) /* TSC does not stop in C states */ +#define X86_FEATURE_CLFLUSH_MONITOR (3*32+25) /* "" clflush reqd with monitor */ /* Intel-defined CPU features, CPUID level 0x00000001 (ecx), word 4 */ #define X86_FEATURE_XMM3 (4*32+ 0) /* "pni" SSE-3 */ @@ -212,6 +213,7 @@ extern const char * const x86_power_flags[32]; #define cpu_has_xmm boot_cpu_has(X86_FEATURE_XMM) #define cpu_has_xmm2 boot_cpu_has(X86_FEATURE_XMM2) #define cpu_has_xmm3 boot_cpu_has(X86_FEATURE_XMM3) +#define cpu_has_aes boot_cpu_has(X86_FEATURE_AES) #define cpu_has_ht boot_cpu_has(X86_FEATURE_HT) #define cpu_has_mp boot_cpu_has(X86_FEATURE_MP) #define cpu_has_nx boot_cpu_has(X86_FEATURE_NX) diff --git a/arch/x86/include/asm/dma-mapping.h b/arch/x86/include/asm/dma-mapping.h index 4035357f5b9..132a134d12f 100644 --- a/arch/x86/include/asm/dma-mapping.h +++ b/arch/x86/include/asm/dma-mapping.h @@ -2,8 +2,8 @@ #define _ASM_X86_DMA_MAPPING_H /* - * IOMMU interface. See Documentation/DMA-mapping.txt and DMA-API.txt for - * documentation. + * IOMMU interface. See Documentation/PCI/PCI-DMA-mapping.txt and + * Documentation/DMA-API.txt for documentation. */ #include <linux/scatterlist.h> diff --git a/arch/x86/include/asm/e820.h b/arch/x86/include/asm/e820.h index 3d8ceddbd40..00d41ce4c84 100644 --- a/arch/x86/include/asm/e820.h +++ b/arch/x86/include/asm/e820.h @@ -49,6 +49,7 @@ #define E820_RESERVED_KERN 128 #ifndef __ASSEMBLY__ +#include <linux/types.h> struct e820entry { __u64 addr; /* start of memory segment */ __u64 size; /* size of memory segment */ diff --git a/arch/x86/include/asm/efi.h b/arch/x86/include/asm/efi.h index ca5ffb2856b..edc90f23e70 100644 --- a/arch/x86/include/asm/efi.h +++ b/arch/x86/include/asm/efi.h @@ -37,8 +37,6 @@ extern unsigned long asmlinkage efi_call_phys(void *, ...); #else /* !CONFIG_X86_32 */ -#define MAX_EFI_IO_PAGES 100 - extern u64 efi_call0(void *fp); extern u64 efi_call1(void *fp, u64 arg1); extern u64 efi_call2(void *fp, u64 arg1, u64 arg2); diff --git a/arch/x86/include/asm/fixmap_64.h b/arch/x86/include/asm/fixmap_64.h index 00a30ab9b1a..8be740977db 100644 --- a/arch/x86/include/asm/fixmap_64.h +++ b/arch/x86/include/asm/fixmap_64.h @@ -16,7 +16,6 @@ #include <asm/apicdef.h> #include <asm/page.h> #include <asm/vsyscall.h> -#include <asm/efi.h> /* * Here we define all the compile-time 'special' virtual @@ -43,9 +42,6 @@ enum fixed_addresses { FIX_APIC_BASE, /* local (CPU) APIC) -- required for SMP or not */ FIX_IO_APIC_BASE_0, FIX_IO_APIC_BASE_END = FIX_IO_APIC_BASE_0 + MAX_IO_APICS - 1, - FIX_EFI_IO_MAP_LAST_PAGE, - FIX_EFI_IO_MAP_FIRST_PAGE = FIX_EFI_IO_MAP_LAST_PAGE - + MAX_EFI_IO_PAGES - 1, #ifdef CONFIG_PARAVIRT FIX_PARAVIRT_BOOTMAP, #endif diff --git a/arch/x86/include/asm/i387.h b/arch/x86/include/asm/i387.h index 48f0004db8c..71c9e518398 100644 --- a/arch/x86/include/asm/i387.h +++ b/arch/x86/include/asm/i387.h @@ -172,7 +172,13 @@ static inline void __save_init_fpu(struct task_struct *tsk) #else /* CONFIG_X86_32 */ -extern void finit(void); +#ifdef CONFIG_MATH_EMULATION +extern void finit_task(struct task_struct *tsk); +#else +static inline void finit_task(struct task_struct *tsk) +{ +} +#endif static inline void tolerant_fwait(void) { diff --git a/arch/x86/include/asm/io.h b/arch/x86/include/asm/io.h index 05cfed4485f..1dbbdf4be9b 100644 --- a/arch/x86/include/asm/io.h +++ b/arch/x86/include/asm/io.h @@ -99,7 +99,6 @@ extern void __iomem *ioremap_wc(unsigned long offset, unsigned long size); * A boot-time mapping is currently limited to at most 16 pages. */ extern void early_ioremap_init(void); -extern void early_ioremap_clear(void); extern void early_ioremap_reset(void); extern void __iomem *early_ioremap(unsigned long offset, unsigned long size); extern void __iomem *early_memremap(unsigned long offset, unsigned long size); diff --git a/arch/x86/include/asm/iomap.h b/arch/x86/include/asm/iomap.h index c1f06289b14..86af26091d6 100644 --- a/arch/x86/include/asm/iomap.h +++ b/arch/x86/include/asm/iomap.h @@ -23,6 +23,9 @@ #include <asm/pgtable.h> #include <asm/tlbflush.h> +int +is_io_mapping_possible(resource_size_t base, unsigned long size); + void * iomap_atomic_prot_pfn(unsigned long pfn, enum km_type type, pgprot_t prot); diff --git a/arch/x86/include/asm/kvm.h b/arch/x86/include/asm/kvm.h index b95162af0bf..dc3f6cf1170 100644 --- a/arch/x86/include/asm/kvm.h +++ b/arch/x86/include/asm/kvm.h @@ -6,9 +6,17 @@ * */ -#include <asm/types.h> +#include <linux/types.h> #include <linux/ioctl.h> +/* Select x86 specific features in <linux/kvm.h> */ +#define __KVM_HAVE_PIT +#define __KVM_HAVE_IOAPIC +#define __KVM_HAVE_DEVICE_ASSIGNMENT +#define __KVM_HAVE_MSI +#define __KVM_HAVE_USER_NMI +#define __KVM_HAVE_GUEST_DEBUG + /* Architectural interrupt line count. */ #define KVM_NR_INTERRUPTS 256 @@ -205,7 +213,30 @@ struct kvm_pit_channel_state { __s64 count_load_time; }; +struct kvm_debug_exit_arch { + __u32 exception; + __u32 pad; + __u64 pc; + __u64 dr6; + __u64 dr7; +}; + +#define KVM_GUESTDBG_USE_SW_BP 0x00010000 +#define KVM_GUESTDBG_USE_HW_BP 0x00020000 +#define KVM_GUESTDBG_INJECT_DB 0x00040000 +#define KVM_GUESTDBG_INJECT_BP 0x00080000 + +/* for KVM_SET_GUEST_DEBUG */ +struct kvm_guest_debug_arch { + __u64 debugreg[8]; +}; + struct kvm_pit_state { struct kvm_pit_channel_state channels[3]; }; + +struct kvm_reinject_control { + __u8 pit_reinject; + __u8 reserved[31]; +}; #endif /* _ASM_X86_KVM_H */ diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 730843d1d2f..f0faf58044f 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -22,6 +22,7 @@ #include <asm/pvclock-abi.h> #include <asm/desc.h> #include <asm/mtrr.h> +#include <asm/msr-index.h> #define KVM_MAX_VCPUS 16 #define KVM_MEMORY_SLOTS 32 @@ -134,11 +135,18 @@ enum { #define KVM_NR_MEM_OBJS 40 -struct kvm_guest_debug { - int enabled; - unsigned long bp[4]; - int singlestep; -}; +#define KVM_NR_DB_REGS 4 + +#define DR6_BD (1 << 13) +#define DR6_BS (1 << 14) +#define DR6_FIXED_1 0xffff0ff0 +#define DR6_VOLATILE 0x0000e00f + +#define DR7_BP_EN_MASK 0x000000ff +#define DR7_GE (1 << 9) +#define DR7_GD (1 << 13) +#define DR7_FIXED_1 0x00000400 +#define DR7_VOLATILE 0xffff23ff /* * We don't want allocation failures within the mmu code, so we preallocate @@ -162,7 +170,8 @@ struct kvm_pte_chain { * bits 0:3 - total guest paging levels (2-4, or zero for real mode) * bits 4:7 - page table level for this shadow (1-4) * bits 8:9 - page table quadrant for 2-level guests - * bit 16 - "metaphysical" - gfn is not a real page (huge page/real mode) + * bit 16 - direct mapping of virtual to physical mapping at gfn + * used for real mode and two-dimensional paging * bits 17:19 - common access permissions for all ptes in this shadow page */ union kvm_mmu_page_role { @@ -172,9 +181,10 @@ union kvm_mmu_page_role { unsigned level:4; unsigned quadrant:2; unsigned pad_for_nice_hex_output:6; - unsigned metaphysical:1; + unsigned direct:1; unsigned access:3; unsigned invalid:1; + unsigned cr4_pge:1; }; }; @@ -218,6 +228,18 @@ struct kvm_pv_mmu_op_buffer { char buf[512] __aligned(sizeof(long)); }; +struct kvm_pio_request { + unsigned long count; + int cur_count; + gva_t guest_gva; + int in; + int port; + int size; + int string; + int down; + int rep; +}; + /* * x86 supports 3 paging modes (4-level 64-bit, 3-level 64-bit, and 2-level * 32-bit). The kvm_mmu structure abstracts the details of the current mmu @@ -236,6 +258,7 @@ struct kvm_mmu { hpa_t root_hpa; int root_level; int shadow_root_level; + union kvm_mmu_page_role base_role; u64 *pae_root; }; @@ -258,6 +281,7 @@ struct kvm_vcpu_arch { unsigned long cr3; unsigned long cr4; unsigned long cr8; + u32 hflags; u64 pdptrs[4]; /* pae */ u64 shadow_efer; u64 apic_base; @@ -338,6 +362,15 @@ struct kvm_vcpu_arch { struct mtrr_state_type mtrr_state; u32 pat; + + int switch_db_regs; + unsigned long host_db[KVM_NR_DB_REGS]; + unsigned long host_dr6; + unsigned long host_dr7; + unsigned long db[KVM_NR_DB_REGS]; + unsigned long dr6; + unsigned long dr7; + unsigned long eff_db[KVM_NR_DB_REGS]; }; struct kvm_mem_alias { @@ -378,6 +411,7 @@ struct kvm_arch{ unsigned long irq_sources_bitmap; unsigned long irq_states[KVM_IOAPIC_NUM_PINS]; + u64 vm_init_tsc; }; struct kvm_vm_stat { @@ -446,8 +480,7 @@ struct kvm_x86_ops { void (*vcpu_put)(struct kvm_vcpu *vcpu); int (*set_guest_debug)(struct kvm_vcpu *vcpu, - struct kvm_debug_guest *dbg); - void (*guest_debug_pre)(struct kvm_vcpu *vcpu); + struct kvm_guest_debug *dbg); int (*get_msr)(struct kvm_vcpu *vcpu, u32 msr_index, u64 *pdata); int (*set_msr)(struct kvm_vcpu *vcpu, u32 msr_index, u64 data); u64 (*get_segment_base)(struct kvm_vcpu *vcpu, int seg); @@ -583,16 +616,12 @@ void kvm_queue_exception_e(struct kvm_vcpu *vcpu, unsigned nr, u32 error_code); void kvm_inject_page_fault(struct kvm_vcpu *vcpu, unsigned long cr2, u32 error_code); -void kvm_pic_set_irq(void *opaque, int irq, int level); +int kvm_pic_set_irq(void *opaque, int irq, int level); void kvm_inject_nmi(struct kvm_vcpu *vcpu); void fx_init(struct kvm_vcpu *vcpu); -int emulator_read_std(unsigned long addr, - void *val, - unsigned int bytes, - struct kvm_vcpu *vcpu); int emulator_write_emulated(unsigned long addr, const void *val, unsigned int bytes, @@ -737,6 +766,10 @@ enum { TASK_SWITCH_GATE = 3, }; +#define HF_GIF_MASK (1 << 0) +#define HF_HIF_MASK (1 << 1) +#define HF_VINTR_MASK (1 << 2) + /* * Hardware virtualization extension instructions may fault if a * reboot turns off virtualization while processes are running. diff --git a/arch/x86/include/asm/math_emu.h b/arch/x86/include/asm/math_emu.h index 5a65b107ad5..031f6266f42 100644 --- a/arch/x86/include/asm/math_emu.h +++ b/arch/x86/include/asm/math_emu.h @@ -1,31 +1,18 @@ #ifndef _ASM_X86_MATH_EMU_H #define _ASM_X86_MATH_EMU_H +#include <asm/ptrace.h> +#include <asm/vm86.h> + /* This structure matches the layout of the data saved to the stack following a device-not-present interrupt, part of it saved automatically by the 80386/80486. */ -struct info { +struct math_emu_info { long ___orig_eip; - long ___ebx; - long ___ecx; - long ___edx; - long ___esi; - long ___edi; - long ___ebp; - long ___eax; - long ___ds; - long ___es; - long ___fs; - long ___orig_eax; - long ___eip; - long ___cs; - long ___eflags; - long ___esp; - long ___ss; - long ___vm86_es; /* This and the following only in vm86 mode */ - long ___vm86_ds; - long ___vm86_fs; - long ___vm86_gs; + union { + struct pt_regs *regs; + struct kernel_vm86_regs *vm86; + }; }; #endif /* _ASM_X86_MATH_EMU_H */ diff --git a/arch/x86/include/asm/mce.h b/arch/x86/include/asm/mce.h index 1d6e17c2f23..32c6e17b960 100644 --- a/arch/x86/include/asm/mce.h +++ b/arch/x86/include/asm/mce.h @@ -3,8 +3,8 @@ #ifdef __x86_64__ +#include <linux/types.h> #include <asm/ioctls.h> -#include <asm/types.h> /* * Machine Check support for x86 @@ -115,8 +115,6 @@ extern int mce_notify_user(void); #endif /* !CONFIG_X86_32 */ - - #ifdef CONFIG_X86_MCE extern void mcheck_init(struct cpuinfo_x86 *c); #else @@ -126,5 +124,4 @@ extern void stop_mce(void); extern void restart_mce(void); #endif /* __KERNEL__ */ - #endif /* _ASM_X86_MCE_H */ diff --git a/arch/x86/include/asm/mmzone_32.h b/arch/x86/include/asm/mmzone_32.h index 07f1af494ca..105fb90a063 100644 --- a/arch/x86/include/asm/mmzone_32.h +++ b/arch/x86/include/asm/mmzone_32.h @@ -32,8 +32,6 @@ static inline void get_memcfg_numa(void) get_memcfg_numa_flat(); } -extern int early_pfn_to_nid(unsigned long pfn); - extern void resume_map_numa_kva(pgd_t *pgd); #else /* !CONFIG_NUMA */ diff --git a/arch/x86/include/asm/mmzone_64.h b/arch/x86/include/asm/mmzone_64.h index a5b3817d4b9..a29f48c2a32 100644 --- a/arch/x86/include/asm/mmzone_64.h +++ b/arch/x86/include/asm/mmzone_64.h @@ -40,8 +40,6 @@ static inline __attribute__((pure)) int phys_to_nid(unsigned long addr) #define node_end_pfn(nid) (NODE_DATA(nid)->node_start_pfn + \ NODE_DATA(nid)->node_spanned_pages) -extern int early_pfn_to_nid(unsigned long pfn); - #ifdef CONFIG_NUMA_EMU #define FAKE_NODE_MIN_SIZE (64 * 1024 * 1024) #define FAKE_NODE_MIN_HASH_MASK (~(FAKE_NODE_MIN_SIZE - 1UL)) diff --git a/arch/x86/include/asm/mpspec.h b/arch/x86/include/asm/mpspec.h index 62d14ce3cd0..bd22f2a3713 100644 --- a/arch/x86/include/asm/mpspec.h +++ b/arch/x86/include/asm/mpspec.h @@ -60,6 +60,7 @@ extern void mp_override_legacy_irq(u8 bus_irq, u8 polarity, u8 trigger, u32 gsi); extern void mp_config_acpi_legacy_irqs(void); extern int mp_register_gsi(u32 gsi, int edge_level, int active_high_low); +extern int acpi_probe_gsi(void); #ifdef CONFIG_X86_IO_APIC extern int mp_config_acpi_gsi(unsigned char number, unsigned int devfn, u8 pin, u32 gsi, int triggering, int polarity); @@ -71,6 +72,11 @@ mp_config_acpi_gsi(unsigned char number, unsigned int devfn, u8 pin, return 0; } #endif +#else /* !CONFIG_ACPI: */ +static inline int acpi_probe_gsi(void) +{ + return 0; +} #endif /* CONFIG_ACPI */ #define PHYSID_ARRAY_SIZE BITS_TO_LONGS(MAX_APICS) diff --git a/arch/x86/include/asm/msr-index.h b/arch/x86/include/asm/msr-index.h index cb58643947b..f4e505f286b 100644 --- a/arch/x86/include/asm/msr-index.h +++ b/arch/x86/include/asm/msr-index.h @@ -18,11 +18,15 @@ #define _EFER_LME 8 /* Long mode enable */ #define _EFER_LMA 10 /* Long mode active (read-only) */ #define _EFER_NX 11 /* No execute enable */ +#define _EFER_SVME 12 /* Enable virtualization */ +#define _EFER_FFXSR 14 /* Enable Fast FXSAVE/FXRSTOR */ #define EFER_SCE (1<<_EFER_SCE) #define EFER_LME (1<<_EFER_LME) #define EFER_LMA (1<<_EFER_LMA) #define EFER_NX (1<<_EFER_NX) +#define EFER_SVME (1<<_EFER_SVME) +#define EFER_FFXSR (1<<_EFER_FFXSR) /* Intel MSRs. Some also available on other CPUs */ #define MSR_IA32_PERFCTR0 0x000000c1 @@ -202,6 +206,35 @@ #define MSR_IA32_THERM_STATUS 0x0000019c #define MSR_IA32_MISC_ENABLE 0x000001a0 +/* MISC_ENABLE bits: architectural */ +#define MSR_IA32_MISC_ENABLE_FAST_STRING (1ULL << 0) +#define MSR_IA32_MISC_ENABLE_TCC (1ULL << 1) +#define MSR_IA32_MISC_ENABLE_EMON (1ULL << 7) +#define MSR_IA32_MISC_ENABLE_BTS_UNAVAIL (1ULL << 11) +#define MSR_IA32_MISC_ENABLE_PEBS_UNAVAIL (1ULL << 12) +#define MSR_IA32_MISC_ENABLE_ENHANCED_SPEEDSTEP (1ULL << 16) +#define MSR_IA32_MISC_ENABLE_MWAIT (1ULL << 18) +#define MSR_IA32_MISC_ENABLE_LIMIT_CPUID (1ULL << 22) +#define MSR_IA32_MISC_ENABLE_XTPR_DISABLE (1ULL << 23) +#define MSR_IA32_MISC_ENABLE_XD_DISABLE (1ULL << 34) + +/* MISC_ENABLE bits: model-specific, meaning may vary from core to core */ +#define MSR_IA32_MISC_ENABLE_X87_COMPAT (1ULL << 2) +#define MSR_IA32_MISC_ENABLE_TM1 (1ULL << 3) +#define MSR_IA32_MISC_ENABLE_SPLIT_LOCK_DISABLE (1ULL << 4) +#define MSR_IA32_MISC_ENABLE_L3CACHE_DISABLE (1ULL << 6) +#define MSR_IA32_MISC_ENABLE_SUPPRESS_LOCK (1ULL << 8) +#define MSR_IA32_MISC_ENABLE_PREFETCH_DISABLE (1ULL << 9) +#define MSR_IA32_MISC_ENABLE_FERR (1ULL << 10) +#define MSR_IA32_MISC_ENABLE_FERR_MULTIPLEX (1ULL << 10) +#define MSR_IA32_MISC_ENABLE_TM2 (1ULL << 13) +#define MSR_IA32_MISC_ENABLE_ADJ_PREF_DISABLE (1ULL << 19) +#define MSR_IA32_MISC_ENABLE_SPEEDSTEP_LOCK (1ULL << 20) +#define MSR_IA32_MISC_ENABLE_L1D_CONTEXT (1ULL << 24) +#define MSR_IA32_MISC_ENABLE_DCU_PREF_DISABLE (1ULL << 37) +#define MSR_IA32_MISC_ENABLE_TURBO_DISABLE (1ULL << 38) +#define MSR_IA32_MISC_ENABLE_IP_PREF_DISABLE (1ULL << 39) + /* Intel Model 6 */ #define MSR_P6_EVNTSEL0 0x00000186 #define MSR_P6_EVNTSEL1 0x00000187 @@ -331,4 +364,9 @@ #define MSR_IA32_VMX_PROCBASED_CTLS2 0x0000048b #define MSR_IA32_VMX_EPT_VPID_CAP 0x0000048c +/* AMD-V MSRs */ + +#define MSR_VM_CR 0xc0010114 +#define MSR_VM_HSAVE_PA 0xc0010117 + #endif /* _ASM_X86_MSR_INDEX_H */ diff --git a/arch/x86/include/asm/mtrr.h b/arch/x86/include/asm/mtrr.h index 14080d22edb..a51ada8467d 100644 --- a/arch/x86/include/asm/mtrr.h +++ b/arch/x86/include/asm/mtrr.h @@ -23,6 +23,7 @@ #ifndef _ASM_X86_MTRR_H #define _ASM_X86_MTRR_H +#include <linux/types.h> #include <linux/ioctl.h> #include <linux/errno.h> diff --git a/arch/x86/include/asm/page.h b/arch/x86/include/asm/page.h index e9873a2e869..776579119a0 100644 --- a/arch/x86/include/asm/page.h +++ b/arch/x86/include/asm/page.h @@ -57,7 +57,6 @@ typedef struct { pgdval_t pgd; } pgd_t; typedef struct { pgprotval_t pgprot; } pgprot_t; extern int page_is_ram(unsigned long pagenr); -extern int pagerange_is_ram(unsigned long start, unsigned long end); extern int devmem_is_allowed(unsigned long pagenr); extern void map_devmem(unsigned long pfn, unsigned long size, pgprot_t vma_prot); diff --git a/arch/x86/include/asm/paravirt.h b/arch/x86/include/asm/paravirt.h index ba3e2ff6aed..e299287e8e3 100644 --- a/arch/x86/include/asm/paravirt.h +++ b/arch/x86/include/asm/paravirt.h @@ -1352,14 +1352,7 @@ static inline void arch_leave_lazy_cpu_mode(void) PVOP_VCALL0(pv_cpu_ops.lazy_mode.leave); } -static inline void arch_flush_lazy_cpu_mode(void) -{ - if (unlikely(paravirt_get_lazy_mode() == PARAVIRT_LAZY_CPU)) { - arch_leave_lazy_cpu_mode(); - arch_enter_lazy_cpu_mode(); - } -} - +void arch_flush_lazy_cpu_mode(void); #define __HAVE_ARCH_ENTER_LAZY_MMU_MODE static inline void arch_enter_lazy_mmu_mode(void) @@ -1372,13 +1365,7 @@ static inline void arch_leave_lazy_mmu_mode(void) PVOP_VCALL0(pv_mmu_ops.lazy_mode.leave); } -static inline void arch_flush_lazy_mmu_mode(void) -{ - if (unlikely(paravirt_get_lazy_mode() == PARAVIRT_LAZY_MMU)) { - arch_leave_lazy_mmu_mode(); - arch_enter_lazy_mmu_mode(); - } -} +void arch_flush_lazy_mmu_mode(void); static inline void __set_fixmap(unsigned /* enum fixed_addresses */ idx, unsigned long phys, pgprot_t flags) @@ -1402,6 +1389,7 @@ static inline int __raw_spin_is_contended(struct raw_spinlock *lock) { return PVOP_CALL1(int, pv_lock_ops.spin_is_contended, lock); } +#define __raw_spin_is_contended __raw_spin_is_contended static __always_inline void __raw_spin_lock(struct raw_spinlock *lock) { diff --git a/arch/x86/include/asm/pgalloc.h b/arch/x86/include/asm/pgalloc.h index cb7c151a8bf..dd14c54ac71 100644 --- a/arch/x86/include/asm/pgalloc.h +++ b/arch/x86/include/asm/pgalloc.h @@ -42,6 +42,7 @@ static inline void pte_free_kernel(struct mm_struct *mm, pte_t *pte) static inline void pte_free(struct mm_struct *mm, struct page *pte) { + pgtable_page_dtor(pte); __free_page(pte); } diff --git a/arch/x86/include/asm/pgtable.h b/arch/x86/include/asm/pgtable.h index 06bbcbd66e9..4f5af8447d5 100644 --- a/arch/x86/include/asm/pgtable.h +++ b/arch/x86/include/asm/pgtable.h @@ -302,16 +302,30 @@ static inline pte_t pte_mkspecial(pte_t pte) extern pteval_t __supported_pte_mask; +/* + * Mask out unsupported bits in a present pgprot. Non-present pgprots + * can use those bits for other purposes, so leave them be. + */ +static inline pgprotval_t massage_pgprot(pgprot_t pgprot) +{ + pgprotval_t protval = pgprot_val(pgprot); + + if (protval & _PAGE_PRESENT) + protval &= __supported_pte_mask; + + return protval; +} + static inline pte_t pfn_pte(unsigned long page_nr, pgprot_t pgprot) { - return __pte((((phys_addr_t)page_nr << PAGE_SHIFT) | - pgprot_val(pgprot)) & __supported_pte_mask); + return __pte(((phys_addr_t)page_nr << PAGE_SHIFT) | + massage_pgprot(pgprot)); } static inline pmd_t pfn_pmd(unsigned long page_nr, pgprot_t pgprot) { - return __pmd((((phys_addr_t)page_nr << PAGE_SHIFT) | - pgprot_val(pgprot)) & __supported_pte_mask); + return __pmd(((phys_addr_t)page_nr << PAGE_SHIFT) | + massage_pgprot(pgprot)); } static inline pte_t pte_modify(pte_t pte, pgprot_t newprot) @@ -323,7 +337,7 @@ static inline pte_t pte_modify(pte_t pte, pgprot_t newprot) * the newprot (if present): */ val &= _PAGE_CHG_MASK; - val |= pgprot_val(newprot) & (~_PAGE_CHG_MASK) & __supported_pte_mask; + val |= massage_pgprot(newprot) & ~_PAGE_CHG_MASK; return __pte(val); } @@ -339,7 +353,7 @@ static inline pgprot_t pgprot_modify(pgprot_t oldprot, pgprot_t newprot) #define pte_pgprot(x) __pgprot(pte_flags(x) & PTE_FLAGS_MASK) -#define canon_pgprot(p) __pgprot(pgprot_val(p) & __supported_pte_mask) +#define canon_pgprot(p) __pgprot(massage_pgprot(p)) static inline int is_new_memtype_allowed(unsigned long flags, unsigned long new_flags) diff --git a/arch/x86/include/asm/prctl.h b/arch/x86/include/asm/prctl.h index a8894647dd9..3ac5032fae0 100644 --- a/arch/x86/include/asm/prctl.h +++ b/arch/x86/include/asm/prctl.h @@ -6,8 +6,4 @@ #define ARCH_GET_FS 0x1003 #define ARCH_GET_GS 0x1004 -#ifdef CONFIG_X86_64 -extern long sys_arch_prctl(int, unsigned long); -#endif /* CONFIG_X86_64 */ - #endif /* _ASM_X86_PRCTL_H */ diff --git a/arch/x86/include/asm/processor.h b/arch/x86/include/asm/processor.h index 091cd8855f2..3bfd5235a9e 100644 --- a/arch/x86/include/asm/processor.h +++ b/arch/x86/include/asm/processor.h @@ -353,7 +353,7 @@ struct i387_soft_struct { u8 no_update; u8 rm; u8 alimit; - struct info *info; + struct math_emu_info *info; u32 entry_eip; }; diff --git a/arch/x86/include/asm/ptrace-abi.h b/arch/x86/include/asm/ptrace-abi.h index 25f1bb8fc62..8e0f8d199e0 100644 --- a/arch/x86/include/asm/ptrace-abi.h +++ b/arch/x86/include/asm/ptrace-abi.h @@ -83,7 +83,7 @@ #ifdef CONFIG_X86_PTRACE_BTS #ifndef __ASSEMBLY__ -#include <asm/types.h> +#include <linux/types.h> /* configuration/status structure used in PTRACE_BTS_CONFIG and PTRACE_BTS_STATUS commands. diff --git a/arch/x86/include/asm/seccomp_32.h b/arch/x86/include/asm/seccomp_32.h index a6ad87b352c..b811d6f5780 100644 --- a/arch/x86/include/asm/seccomp_32.h +++ b/arch/x86/include/asm/seccomp_32.h @@ -1,12 +1,6 @@ #ifndef _ASM_X86_SECCOMP_32_H #define _ASM_X86_SECCOMP_32_H -#include <linux/thread_info.h> - -#ifdef TIF_32BIT -#error "unexpected TIF_32BIT on i386" -#endif - #include <linux/unistd.h> #define __NR_seccomp_read __NR_read diff --git a/arch/x86/include/asm/seccomp_64.h b/arch/x86/include/asm/seccomp_64.h index 4171bb794e9..84ec1bd161a 100644 --- a/arch/x86/include/asm/seccomp_64.h +++ b/arch/x86/include/asm/seccomp_64.h @@ -1,14 +1,6 @@ #ifndef _ASM_X86_SECCOMP_64_H #define _ASM_X86_SECCOMP_64_H -#include <linux/thread_info.h> - -#ifdef TIF_32BIT -#error "unexpected TIF_32BIT on x86_64" -#else -#define TIF_32BIT TIF_IA32 -#endif - #include <linux/unistd.h> #include <asm/ia32_unistd.h> diff --git a/arch/x86/include/asm/setup.h b/arch/x86/include/asm/setup.h index ebe858cdc8a..c2308f5250f 100644 --- a/arch/x86/include/asm/setup.h +++ b/arch/x86/include/asm/setup.h @@ -1,27 +1,12 @@ #ifndef _ASM_X86_SETUP_H #define _ASM_X86_SETUP_H +#ifdef __KERNEL__ + #define COMMAND_LINE_SIZE 2048 #ifndef __ASSEMBLY__ -/* Interrupt control for vSMPowered x86_64 systems */ -void vsmp_init(void); - - -void setup_bios_corruption_check(void); - - -#ifdef CONFIG_X86_VISWS -extern void visws_early_detect(void); -extern int is_visws_box(void); -#else -static inline void visws_early_detect(void) { } -static inline int is_visws_box(void) { return 0; } -#endif - -extern int wakeup_secondary_cpu_via_nmi(int apicid, unsigned long start_eip); -extern int wakeup_secondary_cpu_via_init(int apicid, unsigned long start_eip); /* * Any setup quirks to be performed? */ @@ -48,16 +33,8 @@ struct x86_quirks { int (*update_genapic)(void); }; -extern struct x86_quirks *x86_quirks; -extern unsigned long saved_video_mode; - -#ifndef CONFIG_PARAVIRT -#define paravirt_post_allocator_init() do {} while (0) -#endif #endif /* __ASSEMBLY__ */ -#ifdef __KERNEL__ - #ifdef __i386__ #include <linux/pfn.h> @@ -78,6 +55,28 @@ extern unsigned long saved_video_mode; #ifndef __ASSEMBLY__ #include <asm/bootparam.h> +/* Interrupt control for vSMPowered x86_64 systems */ +void vsmp_init(void); + +void setup_bios_corruption_check(void); + +#ifdef CONFIG_X86_VISWS +extern void visws_early_detect(void); +extern int is_visws_box(void); +#else +static inline void visws_early_detect(void) { } +static inline int is_visws_box(void) { return 0; } +#endif + +extern int wakeup_secondary_cpu_via_nmi(int apicid, unsigned long start_eip); +extern int wakeup_secondary_cpu_via_init(int apicid, unsigned long start_eip); +extern struct x86_quirks *x86_quirks; +extern unsigned long saved_video_mode; + +#ifndef CONFIG_PARAVIRT +#define paravirt_post_allocator_init() do {} while (0) +#endif + #ifndef _SETUP /* diff --git a/arch/x86/include/asm/sigcontext.h b/arch/x86/include/asm/sigcontext.h index 0afcb5e58ac..ec666491aaa 100644 --- a/arch/x86/include/asm/sigcontext.h +++ b/arch/x86/include/asm/sigcontext.h @@ -2,7 +2,7 @@ #define _ASM_X86_SIGCONTEXT_H #include <linux/compiler.h> -#include <asm/types.h> +#include <linux/types.h> #define FP_XSTATE_MAGIC1 0x46505853U #define FP_XSTATE_MAGIC2 0x46505845U diff --git a/arch/x86/include/asm/sigcontext32.h b/arch/x86/include/asm/sigcontext32.h index 6126188cf3a..ad1478c4ae1 100644 --- a/arch/x86/include/asm/sigcontext32.h +++ b/arch/x86/include/asm/sigcontext32.h @@ -1,6 +1,8 @@ #ifndef _ASM_X86_SIGCONTEXT32_H #define _ASM_X86_SIGCONTEXT32_H +#include <linux/types.h> + /* signal context for 32bit programs. */ #define X86_FXSR_MAGIC 0x0000 diff --git a/arch/x86/include/asm/socket.h b/arch/x86/include/asm/socket.h index 8ab9cc8b2ec..ca8bf2cd0ba 100644 --- a/arch/x86/include/asm/socket.h +++ b/arch/x86/include/asm/socket.h @@ -54,4 +54,7 @@ #define SO_MARK 36 +#define SO_TIMESTAMPING 37 +#define SCM_TIMESTAMPING SO_TIMESTAMPING + #endif /* _ASM_X86_SOCKET_H */ diff --git a/arch/x86/include/asm/spinlock.h b/arch/x86/include/asm/spinlock.h index d17c91981da..8247e94ac6b 100644 --- a/arch/x86/include/asm/spinlock.h +++ b/arch/x86/include/asm/spinlock.h @@ -245,6 +245,7 @@ static inline int __raw_spin_is_contended(raw_spinlock_t *lock) { return __ticket_spin_is_contended(lock); } +#define __raw_spin_is_contended __raw_spin_is_contended static __always_inline void __raw_spin_lock(raw_spinlock_t *lock) { diff --git a/arch/x86/include/asm/svm.h b/arch/x86/include/asm/svm.h index 1b8afa78e86..82ada75f3eb 100644 --- a/arch/x86/include/asm/svm.h +++ b/arch/x86/include/asm/svm.h @@ -174,10 +174,6 @@ struct __attribute__ ((__packed__)) vmcb { #define SVM_CPUID_FEATURE_SHIFT 2 #define SVM_CPUID_FUNC 0x8000000a -#define MSR_EFER_SVME_MASK (1ULL << 12) -#define MSR_VM_CR 0xc0010114 -#define MSR_VM_HSAVE_PA 0xc0010117ULL - #define SVM_VM_CR_SVM_DISABLE 4 #define SVM_SELECTOR_S_SHIFT 4 diff --git a/arch/x86/include/asm/swab.h b/arch/x86/include/asm/swab.h index 306d4178ffc..557cd9f0066 100644 --- a/arch/x86/include/asm/swab.h +++ b/arch/x86/include/asm/swab.h @@ -1,7 +1,7 @@ #ifndef _ASM_X86_SWAB_H #define _ASM_X86_SWAB_H -#include <asm/types.h> +#include <linux/types.h> #include <linux/compiler.h> static inline __attribute_const__ __u32 __arch_swab32(__u32 val) diff --git a/arch/x86/include/asm/syscalls.h b/arch/x86/include/asm/syscalls.h index 9c6797c3e56..e26d34b0bc7 100644 --- a/arch/x86/include/asm/syscalls.h +++ b/arch/x86/include/asm/syscalls.h @@ -40,7 +40,7 @@ asmlinkage int sys_sigaction(int, const struct old_sigaction __user *, struct old_sigaction __user *); asmlinkage int sys_sigaltstack(unsigned long); asmlinkage unsigned long sys_sigreturn(unsigned long); -asmlinkage int sys_rt_sigreturn(struct pt_regs); +asmlinkage int sys_rt_sigreturn(unsigned long); /* kernel/ioport.c */ asmlinkage long sys_iopl(unsigned long); @@ -74,6 +74,7 @@ asmlinkage long sys_vfork(struct pt_regs *); asmlinkage long sys_execve(char __user *, char __user * __user *, char __user * __user *, struct pt_regs *); +long sys_arch_prctl(int, unsigned long); /* kernel/ioport.c */ asmlinkage long sys_iopl(unsigned int, struct pt_regs *); diff --git a/arch/x86/include/asm/timer.h b/arch/x86/include/asm/timer.h index 2bb6a835c45..4f5c2472485 100644 --- a/arch/x86/include/asm/timer.h +++ b/arch/x86/include/asm/timer.h @@ -11,8 +11,8 @@ unsigned long native_calibrate_tsc(void); #ifdef CONFIG_X86_32 extern int timer_ack; +#endif extern int recalibrate_cpu_khz(void); -#endif /* CONFIG_X86_32 */ extern int no_timer_check; diff --git a/arch/x86/include/asm/timex.h b/arch/x86/include/asm/timex.h index 1287dc1347d..b5c9d45c981 100644 --- a/arch/x86/include/asm/timex.h +++ b/arch/x86/include/asm/timex.h @@ -1,18 +1,13 @@ -/* x86 architecture timex specifications */ #ifndef _ASM_X86_TIMEX_H #define _ASM_X86_TIMEX_H #include <asm/processor.h> #include <asm/tsc.h> -#ifdef CONFIG_X86_ELAN -# define PIT_TICK_RATE 1189200 /* AMD Elan has different frequency! */ -#elif defined(CONFIG_X86_RDC321X) -# define PIT_TICK_RATE 1041667 /* Underlying HZ for R8610 */ -#else -# define PIT_TICK_RATE 1193182 /* Underlying HZ */ -#endif -#define CLOCK_TICK_RATE PIT_TICK_RATE +/* The PIT ticks at this frequency (in HZ): */ +#define PIT_TICK_RATE 1193182 + +#define CLOCK_TICK_RATE PIT_TICK_RATE #define ARCH_HAS_READ_CURRENT_TIMER diff --git a/arch/x86/include/asm/traps.h b/arch/x86/include/asm/traps.h index 2ee0a3bceed..cf3bb053da0 100644 --- a/arch/x86/include/asm/traps.h +++ b/arch/x86/include/asm/traps.h @@ -41,7 +41,7 @@ dotraplinkage void do_int3(struct pt_regs *, long); dotraplinkage void do_overflow(struct pt_regs *, long); dotraplinkage void do_bounds(struct pt_regs *, long); dotraplinkage void do_invalid_op(struct pt_regs *, long); -dotraplinkage void do_device_not_available(struct pt_regs *, long); +dotraplinkage void do_device_not_available(struct pt_regs); dotraplinkage void do_coprocessor_segment_overrun(struct pt_regs *, long); dotraplinkage void do_invalid_TSS(struct pt_regs *, long); dotraplinkage void do_segment_not_present(struct pt_regs *, long); @@ -77,7 +77,7 @@ extern int panic_on_unrecovered_nmi; extern int kstack_depth_to_print; void math_error(void __user *); -asmlinkage void math_emulate(long); +void math_emulate(struct math_emu_info *); #ifdef CONFIG_X86_32 unsigned long patch_espfix_desc(unsigned long, unsigned long); #else diff --git a/arch/x86/include/asm/virtext.h b/arch/x86/include/asm/virtext.h index 59363627523..e0f9aa16358 100644 --- a/arch/x86/include/asm/virtext.h +++ b/arch/x86/include/asm/virtext.h @@ -118,7 +118,7 @@ static inline void cpu_svm_disable(void) wrmsrl(MSR_VM_HSAVE_PA, 0); rdmsrl(MSR_EFER, efer); - wrmsrl(MSR_EFER, efer & ~MSR_EFER_SVME_MASK); + wrmsrl(MSR_EFER, efer & ~EFER_SVME); } /** Makes sure SVM is disabled, if it is supported on the CPU diff --git a/arch/x86/include/asm/vmx.h b/arch/x86/include/asm/vmx.h index d0238e6151d..498f944010b 100644 --- a/arch/x86/include/asm/vmx.h +++ b/arch/x86/include/asm/vmx.h @@ -270,8 +270,9 @@ enum vmcs_field { #define INTR_TYPE_EXT_INTR (0 << 8) /* external interrupt */ #define INTR_TYPE_NMI_INTR (2 << 8) /* NMI */ -#define INTR_TYPE_EXCEPTION (3 << 8) /* processor exception */ +#define INTR_TYPE_HARD_EXCEPTION (3 << 8) /* processor exception */ #define INTR_TYPE_SOFT_INTR (4 << 8) /* software interrupt */ +#define INTR_TYPE_SOFT_EXCEPTION (6 << 8) /* software exception */ /* GUEST_INTERRUPTIBILITY_INFO flags. */ #define GUEST_INTR_STATE_STI 0x00000001 @@ -311,7 +312,7 @@ enum vmcs_field { #define DEBUG_REG_ACCESS_TYPE 0x10 /* 4, direction of access */ #define TYPE_MOV_TO_DR (0 << 4) #define TYPE_MOV_FROM_DR (1 << 4) -#define DEBUG_REG_ACCESS_REG 0xf00 /* 11:8, general purpose reg. */ +#define DEBUG_REG_ACCESS_REG(eq) (((eq) >> 8) & 0xf) /* 11:8, general purpose reg. */ /* segment AR */ diff --git a/arch/x86/include/asm/xen/page.h b/arch/x86/include/asm/xen/page.h index 7ef617ef1df..4bd990ee43d 100644 --- a/arch/x86/include/asm/xen/page.h +++ b/arch/x86/include/asm/xen/page.h @@ -137,7 +137,7 @@ static inline pte_t mfn_pte(unsigned long page_nr, pgprot_t pgprot) pte_t pte; pte.pte = ((phys_addr_t)page_nr << PAGE_SHIFT) | - (pgprot_val(pgprot) & __supported_pte_mask); + massage_pgprot(pgprot); return pte; } diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c index d37593c2f43..7678f10c456 100644 --- a/arch/x86/kernel/acpi/boot.c +++ b/arch/x86/kernel/acpi/boot.c @@ -973,6 +973,29 @@ void __init mp_register_ioapic(int id, u32 address, u32 gsi_base) nr_ioapics++; } +int __init acpi_probe_gsi(void) +{ + int idx; + int gsi; + int max_gsi = 0; + + if (acpi_disabled) + return 0; + + if (!acpi_ioapic) + return 0; + + max_gsi = 0; + for (idx = 0; idx < nr_ioapics; idx++) { + gsi = mp_ioapic_routing[idx].gsi_end; + + if (gsi > max_gsi) + max_gsi = gsi; + } + + return max_gsi + 1; +} + static void assign_to_mp_irq(struct mp_config_intsrc *m, struct mp_config_intsrc *mp_irq) { diff --git a/arch/x86/kernel/acpi/sleep.c b/arch/x86/kernel/acpi/sleep.c index 707c1f6f95f..a60c1f3bcb8 100644 --- a/arch/x86/kernel/acpi/sleep.c +++ b/arch/x86/kernel/acpi/sleep.c @@ -156,11 +156,11 @@ 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) + acpi_s4_no_nvs(); #endif if (strncmp(str, "old_ordering", 12) == 0) acpi_old_suspend_ordering(); - if (strncmp(str, "s4_nonvs", 8) == 0) - acpi_s4_no_nvs(); str = strchr(str, ','); if (str != NULL) str += strspn(str, ", \t"); diff --git a/arch/x86/kernel/acpi/wakeup_64.S b/arch/x86/kernel/acpi/wakeup_64.S index bcc293423a7..96258d9dc97 100644 --- a/arch/x86/kernel/acpi/wakeup_64.S +++ b/arch/x86/kernel/acpi/wakeup_64.S @@ -13,7 +13,6 @@ * Hooray, we are in Long 64-bit mode (but still running in low memory) */ ENTRY(wakeup_long64) -wakeup_long64: movq saved_magic, %rax movq $0x123456789abcdef0, %rdx cmpq %rdx, %rax @@ -34,16 +33,12 @@ wakeup_long64: movq saved_rip, %rax jmp *%rax +ENDPROC(wakeup_long64) bogus_64_magic: jmp bogus_64_magic - .align 2 - .p2align 4,,15 -.globl do_suspend_lowlevel - .type do_suspend_lowlevel,@function -do_suspend_lowlevel: -.LFB5: +ENTRY(do_suspend_lowlevel) subq $8, %rsp xorl %eax, %eax call save_processor_state @@ -67,7 +62,7 @@ do_suspend_lowlevel: pushfq popq pt_regs_flags(%rax) - movq $.L97, saved_rip(%rip) + movq $resume_point, saved_rip(%rip) movq %rsp, saved_rsp movq %rbp, saved_rbp @@ -78,14 +73,12 @@ do_suspend_lowlevel: addq $8, %rsp movl $3, %edi xorl %eax, %eax - jmp acpi_enter_sleep_state -.L97: - .p2align 4,,7 -.L99: - .align 4 - movl $24, %eax - movw %ax, %ds + call acpi_enter_sleep_state + /* in case something went wrong, restore the machine status and go on */ + jmp resume_point + .align 4 +resume_point: /* We don't restore %rax, it must be 0 anyway */ movq $saved_context, %rax movq saved_context_cr4(%rax), %rbx @@ -117,12 +110,9 @@ do_suspend_lowlevel: xorl %eax, %eax addq $8, %rsp jmp restore_processor_state -.LFE5: -.Lfe5: - .size do_suspend_lowlevel, .Lfe5-do_suspend_lowlevel - +ENDPROC(do_suspend_lowlevel) + .data -ALIGN ENTRY(saved_rbp) .quad 0 ENTRY(saved_rsi) .quad 0 ENTRY(saved_rdi) .quad 0 diff --git a/arch/x86/kernel/apic.c b/arch/x86/kernel/apic.c index 0f830e4f567..570f36e44e5 100644 --- a/arch/x86/kernel/apic.c +++ b/arch/x86/kernel/apic.c @@ -862,7 +862,7 @@ void clear_local_APIC(void) } /* lets not touch this if we didn't frob it */ -#if defined(CONFIG_X86_MCE_P4THERMAL) || defined(X86_MCE_INTEL) +#if defined(CONFIG_X86_MCE_P4THERMAL) || defined(CONFIG_X86_MCE_INTEL) if (maxlvt >= 5) { v = apic_read(APIC_LVTTHMR); apic_write(APIC_LVTTHMR, v | APIC_LVT_MASKED); @@ -895,6 +895,10 @@ void disable_local_APIC(void) { unsigned int value; + /* APIC hasn't been mapped yet */ + if (!apic_phys) + return; + clear_local_APIC(); /* @@ -1432,7 +1436,7 @@ static int __init detect_init_APIC(void) switch (boot_cpu_data.x86_vendor) { case X86_VENDOR_AMD: if ((boot_cpu_data.x86 == 6 && boot_cpu_data.x86_model > 1) || - (boot_cpu_data.x86 == 15)) + (boot_cpu_data.x86 >= 15)) break; goto no_apic; case X86_VENDOR_INTEL: @@ -1833,6 +1837,11 @@ void __cpuinit generic_processor_info(int apicid, int version) num_processors++; cpu = cpumask_next_zero(-1, cpu_present_mask); + if (version != apic_version[boot_cpu_physical_apicid]) + WARN_ONCE(1, + "ACPI: apic version mismatch, bootcpu: %x cpu %d: %x\n", + apic_version[boot_cpu_physical_apicid], cpu, version); + physid_set(apicid, phys_cpu_present_map); if (apicid == boot_cpu_physical_apicid) { /* diff --git a/arch/x86/kernel/apm_32.c b/arch/x86/kernel/apm_32.c index 98807bb095a..266ec6c18b6 100644 --- a/arch/x86/kernel/apm_32.c +++ b/arch/x86/kernel/apm_32.c @@ -1192,6 +1192,7 @@ static int suspend(int vetoable) device_suspend(PMSG_SUSPEND); local_irq_disable(); device_power_down(PMSG_SUSPEND); + sysdev_suspend(PMSG_SUSPEND); local_irq_enable(); @@ -1208,6 +1209,7 @@ static int suspend(int vetoable) if (err != APM_SUCCESS) apm_error("suspend", err); err = (err == APM_SUCCESS) ? 0 : -EIO; + sysdev_resume(); device_power_up(PMSG_RESUME); local_irq_enable(); device_resume(PMSG_RESUME); @@ -1228,6 +1230,7 @@ static void standby(void) local_irq_disable(); device_power_down(PMSG_SUSPEND); + sysdev_suspend(PMSG_SUSPEND); local_irq_enable(); err = set_system_power_state(APM_STATE_STANDBY); @@ -1235,6 +1238,7 @@ static void standby(void) apm_error("standby", err); local_irq_disable(); + sysdev_resume(); device_power_up(PMSG_RESUME); local_irq_enable(); } diff --git a/arch/x86/kernel/cpu/cpufreq/Kconfig b/arch/x86/kernel/cpu/cpufreq/Kconfig index efae3b22a0f..52c83987547 100644 --- a/arch/x86/kernel/cpu/cpufreq/Kconfig +++ b/arch/x86/kernel/cpu/cpufreq/Kconfig @@ -87,30 +87,15 @@ config X86_POWERNOW_K7_ACPI 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 mobile AMD Opteron/Athlon64 processors. + 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/>. - If in doubt, say N. - -config X86_POWERNOW_K8_ACPI - bool - prompt "ACPI Support" if X86_32 - depends on ACPI && X86_POWERNOW_K8 && ACPI_PROCESSOR - depends on !(X86_POWERNOW_K8 = y && ACPI_PROCESSOR = m) - default y - help - This provides access to the K8s Processor Performance States via ACPI. - This driver is probably required for CPUFreq to work with multi-socket and - SMP systems. It is not required on at least some single-socket yet - multi-core systems, even if SMP is enabled. - - It is safe to say Y here. - config X86_GX_SUSPMOD tristate "Cyrix MediaGX/NatSemi Geode Suspend Modulation" depends on X86_32 && PCI @@ -245,17 +230,6 @@ config X86_E_POWERSAVER comment "shared options" -config X86_ACPI_CPUFREQ_PROC_INTF - bool "/proc/acpi/processor/../performance interface (deprecated)" - depends on PROC_FS - depends on X86_ACPI_CPUFREQ || X86_POWERNOW_K7_ACPI || X86_POWERNOW_K8_ACPI - help - This enables the deprecated /proc/acpi/processor/../performance - interface. While it is helpful for debugging, the generic, - cross-architecture cpufreq interfaces should be used. - - If in doubt, say N. - config X86_SPEEDSTEP_LIB tristate default (X86_SPEEDSTEP_ICH || X86_SPEEDSTEP_SMI || X86_P4_CLOCKMOD) diff --git a/arch/x86/kernel/cpu/cpufreq/Makefile b/arch/x86/kernel/cpu/cpufreq/Makefile index 560f7760dae..509296df294 100644 --- a/arch/x86/kernel/cpu/cpufreq/Makefile +++ b/arch/x86/kernel/cpu/cpufreq/Makefile @@ -1,6 +1,11 @@ +# 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 +obj-$(CONFIG_X86_ACPI_CPUFREQ) += acpi-cpufreq.o obj-$(CONFIG_X86_POWERNOW_K6) += powernow-k6.o obj-$(CONFIG_X86_POWERNOW_K7) += powernow-k7.o -obj-$(CONFIG_X86_POWERNOW_K8) += powernow-k8.o obj-$(CONFIG_X86_LONGHAUL) += longhaul.o obj-$(CONFIG_X86_E_POWERSAVER) += e_powersaver.o obj-$(CONFIG_ELAN_CPUFREQ) += elanfreq.o @@ -10,7 +15,6 @@ obj-$(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_ACPI_CPUFREQ) += acpi-cpufreq.o obj-$(CONFIG_X86_SPEEDSTEP_CENTRINO) += speedstep-centrino.o obj-$(CONFIG_X86_P4_CLOCKMOD) += p4-clockmod.o obj-$(CONFIG_X86_CPUFREQ_NFORCE2) += cpufreq-nforce2.o diff --git a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c index 6f11e029e8c..3babe1f1e91 100644 --- a/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c +++ b/arch/x86/kernel/cpu/cpufreq/acpi-cpufreq.c @@ -1,5 +1,5 @@ /* - * acpi-cpufreq.c - ACPI Processor P-States Driver ($Revision: 1.4 $) + * 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> @@ -36,16 +36,18 @@ #include <linux/ftrace.h> #include <linux/acpi.h> +#include <linux/io.h> +#include <linux/delay.h> +#include <linux/uaccess.h> + #include <acpi/processor.h> -#include <asm/io.h> #include <asm/msr.h> #include <asm/processor.h> #include <asm/cpufeature.h> -#include <asm/delay.h> -#include <asm/uaccess.h> -#define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, "acpi-cpufreq", msg) +#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"); @@ -95,7 +97,7 @@ static unsigned extract_io(u32 value, struct acpi_cpufreq_data *data) perf = data->acpi_data; - for (i=0; i<perf->state_count; i++) { + for (i = 0; i < perf->state_count; i++) { if (value == perf->states[i].status) return data->freq_table[i].frequency; } @@ -110,7 +112,7 @@ static unsigned extract_msr(u32 msr, struct acpi_cpufreq_data *data) msr &= INTEL_MSR_RANGE; perf = data->acpi_data; - for (i=0; data->freq_table[i].frequency != CPUFREQ_TABLE_END; i++) { + 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; } @@ -138,20 +140,19 @@ struct io_addr { u8 bit_width; }; -typedef union { - struct msr_addr msr; - struct io_addr io; -} drv_addr_union; - struct drv_cmd { unsigned int type; - cpumask_var_t mask; - drv_addr_union addr; + const struct cpumask *mask; + union { + struct msr_addr msr; + struct io_addr io; + } addr; u32 val; }; -static void do_drv_read(struct drv_cmd *cmd) +static long do_drv_read(void *_cmd) { + struct drv_cmd *cmd = _cmd; u32 h; switch (cmd->type) { @@ -166,10 +167,12 @@ static void do_drv_read(struct drv_cmd *cmd) default: break; } + return 0; } -static void do_drv_write(struct drv_cmd *cmd) +static long do_drv_write(void *_cmd) { + struct drv_cmd *cmd = _cmd; u32 lo, hi; switch (cmd->type) { @@ -186,30 +189,23 @@ static void do_drv_write(struct drv_cmd *cmd) default: break; } + return 0; } static void drv_read(struct drv_cmd *cmd) { - cpumask_t saved_mask = current->cpus_allowed; cmd->val = 0; - set_cpus_allowed_ptr(current, cmd->mask); - do_drv_read(cmd); - set_cpus_allowed_ptr(current, &saved_mask); + work_on_cpu(cpumask_any(cmd->mask), do_drv_read, cmd); } static void drv_write(struct drv_cmd *cmd) { - cpumask_t saved_mask = current->cpus_allowed; unsigned int i; for_each_cpu(i, cmd->mask) { - set_cpus_allowed_ptr(current, cpumask_of(i)); - do_drv_write(cmd); + work_on_cpu(i, do_drv_write, cmd); } - - set_cpus_allowed_ptr(current, &saved_mask); - return; } static u32 get_cur_val(const struct cpumask *mask) @@ -235,8 +231,7 @@ static u32 get_cur_val(const struct cpumask *mask) return 0; } - cpumask_copy(cmd.mask, mask); - + cmd.mask = mask; drv_read(&cmd); dprintk("get_cur_val = %u\n", cmd.val); @@ -368,13 +363,13 @@ static unsigned int get_cur_freq_on_cpu(unsigned int cpu) return freq; } -static unsigned int check_freqs(const cpumask_t *mask, unsigned int 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++) { + for (i = 0; i < 100; i++) { cur_freq = extract_freq(get_cur_val(mask), data); if (cur_freq == freq) return 1; @@ -403,9 +398,6 @@ static int acpi_cpufreq_target(struct cpufreq_policy *policy, return -ENODEV; } - if (unlikely(!alloc_cpumask_var(&cmd.mask, GFP_KERNEL))) - return -ENOMEM; - perf = data->acpi_data; result = cpufreq_frequency_table_target(policy, data->freq_table, @@ -450,9 +442,9 @@ static int acpi_cpufreq_target(struct cpufreq_policy *policy, /* cpufreq holds the hotplug lock, so we are safe from here on */ if (policy->shared_type != CPUFREQ_SHARED_TYPE_ANY) - cpumask_and(cmd.mask, cpu_online_mask, policy->cpus); + cmd.mask = policy->cpus; else - cpumask_copy(cmd.mask, cpumask_of(policy->cpu)); + cmd.mask = cpumask_of(policy->cpu); freqs.old = perf->states[perf->state].core_frequency * 1000; freqs.new = data->freq_table[next_state].frequency; @@ -479,7 +471,6 @@ static int acpi_cpufreq_target(struct cpufreq_policy *policy, perf->state = next_perf_state; out: - free_cpumask_var(cmd.mask); return result; } @@ -503,7 +494,7 @@ acpi_cpufreq_guess_freq(struct acpi_cpufreq_data *data, unsigned int cpu) unsigned long freq; unsigned long freqn = perf->states[0].core_frequency * 1000; - for (i=0; i<(perf->state_count-1); i++) { + 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)) { @@ -682,7 +673,7 @@ static int acpi_cpufreq_cpu_init(struct cpufreq_policy *policy) /* detect transition latency */ policy->cpuinfo.transition_latency = 0; - for (i=0; i<perf->state_count; i++) { + for (i = 0; i < perf->state_count; i++) { if ((perf->states[i].transition_latency * 1000) > policy->cpuinfo.transition_latency) policy->cpuinfo.transition_latency = @@ -691,8 +682,8 @@ static int acpi_cpufreq_cpu_init(struct cpufreq_policy *policy) data->max_freq = perf->states[0].core_frequency * 1000; /* table init */ - for (i=0; i<perf->state_count; i++) { - if (i>0 && perf->states[i].core_frequency >= + 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; diff --git a/arch/x86/kernel/cpu/cpufreq/cpufreq-nforce2.c b/arch/x86/kernel/cpu/cpufreq/cpufreq-nforce2.c index 965ea52767a..733093d6043 100644 --- a/arch/x86/kernel/cpu/cpufreq/cpufreq-nforce2.c +++ b/arch/x86/kernel/cpu/cpufreq/cpufreq-nforce2.c @@ -32,7 +32,7 @@ * nforce2_chipset: * FSB is changed using the chipset */ -static struct pci_dev *nforce2_chipset_dev; +static struct pci_dev *nforce2_dev; /* fid: * multiplier * 10 @@ -56,7 +56,9 @@ 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 dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, "cpufreq-nforce2", msg) +#define PFX "cpufreq-nforce2: " +#define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, \ + "cpufreq-nforce2", msg) /** * nforce2_calc_fsb - calculate FSB @@ -118,11 +120,11 @@ static void nforce2_write_pll(int pll) int temp; /* Set the pll addr. to 0x00 */ - pci_write_config_dword(nforce2_chipset_dev, NFORCE2_PLLADR, 0); + 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_chipset_dev, NFORCE2_PLLREG, pll); + pci_write_config_dword(nforce2_dev, NFORCE2_PLLREG, pll); return; } @@ -139,8 +141,8 @@ static unsigned int nforce2_fsb_read(int bootfsb) 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); + nforce2_sub5 = pci_get_subsys(PCI_VENDOR_ID_NVIDIA, 0x01EF, + PCI_ANY_ID, PCI_ANY_ID, NULL); if (!nforce2_sub5) return 0; @@ -148,13 +150,13 @@ static unsigned int nforce2_fsb_read(int bootfsb) fsb /= 1000000; /* Check if PLL register is already set */ - pci_read_config_byte(nforce2_chipset_dev, NFORCE2_PLLENABLE, (u8 *)&temp); + 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_chipset_dev, NFORCE2_PLLREG, &temp); + pci_read_config_dword(nforce2_dev, NFORCE2_PLLREG, &temp); fsb = nforce2_calc_fsb(temp); return fsb; @@ -174,18 +176,18 @@ static int nforce2_set_fsb(unsigned int fsb) int pll = 0; if ((fsb > max_fsb) || (fsb < NFORCE2_MIN_FSB)) { - printk(KERN_ERR "cpufreq: FSB %d is out of range!\n", 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 "cpufreq: Error while reading the FSB\n"); + printk(KERN_ERR PFX "Error while reading the FSB\n"); return -EINVAL; } /* First write? Then set actual value */ - pci_read_config_byte(nforce2_chipset_dev, NFORCE2_PLLENABLE, (u8 *)&temp); + pci_read_config_byte(nforce2_dev, NFORCE2_PLLENABLE, (u8 *)&temp); if (!temp) { pll = nforce2_calc_pll(tfsb); @@ -197,7 +199,7 @@ static int nforce2_set_fsb(unsigned int fsb) /* Enable write access */ temp = 0x01; - pci_write_config_byte(nforce2_chipset_dev, NFORCE2_PLLENABLE, (u8)temp); + pci_write_config_byte(nforce2_dev, NFORCE2_PLLENABLE, (u8)temp); diff = tfsb - fsb; @@ -222,7 +224,7 @@ static int nforce2_set_fsb(unsigned int fsb) } temp = 0x40; - pci_write_config_byte(nforce2_chipset_dev, NFORCE2_PLLADR, (u8)temp); + pci_write_config_byte(nforce2_dev, NFORCE2_PLLADR, (u8)temp); return 0; } @@ -244,7 +246,8 @@ static unsigned int nforce2_get(unsigned int cpu) * 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) + * @relation: how that frequency relates to achieved frequency + * (CPUFREQ_RELATION_L or CPUFREQ_RELATION_H) * * Sets a new CPUFreq policy. */ @@ -276,7 +279,7 @@ static int nforce2_target(struct cpufreq_policy *policy, /* local_irq_save(flags); */ if (nforce2_set_fsb(target_fsb) < 0) - printk(KERN_ERR "cpufreq: Changing FSB to %d failed\n", + printk(KERN_ERR PFX "Changing FSB to %d failed\n", target_fsb); else dprintk("Changed FSB successfully to %d\n", @@ -327,8 +330,8 @@ static int nforce2_cpu_init(struct cpufreq_policy *policy) /* FIX: Get FID from CPU */ if (!fid) { if (!cpu_khz) { - printk(KERN_WARNING - "cpufreq: cpu_khz not set, can't calculate multiplier!\n"); + printk(KERN_WARNING PFX + "cpu_khz not set, can't calculate multiplier!\n"); return -ENODEV; } @@ -343,7 +346,7 @@ static int nforce2_cpu_init(struct cpufreq_policy *policy) } } - printk(KERN_INFO "cpufreq: FSB currently at %i MHz, FID %d.%d\n", fsb, + 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 */ @@ -392,17 +395,18 @@ static struct cpufreq_driver nforce2_driver = { */ static unsigned int nforce2_detect_chipset(void) { - nforce2_chipset_dev = pci_get_subsys(PCI_VENDOR_ID_NVIDIA, + nforce2_dev = pci_get_subsys(PCI_VENDOR_ID_NVIDIA, PCI_DEVICE_ID_NVIDIA_NFORCE2, PCI_ANY_ID, PCI_ANY_ID, NULL); - if (nforce2_chipset_dev == NULL) + if (nforce2_dev == NULL) return -ENODEV; - printk(KERN_INFO "cpufreq: Detected nForce2 chipset revision %X\n", - nforce2_chipset_dev->revision); - printk(KERN_INFO - "cpufreq: FSB changing is maybe unstable and can lead to crashes and data loss.\n"); + 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; } @@ -420,7 +424,7 @@ static int __init nforce2_init(void) /* detect chipset */ if (nforce2_detect_chipset()) { - printk(KERN_ERR "cpufreq: No nForce2 chipset.\n"); + printk(KERN_INFO PFX "No nForce2 chipset.\n"); return -ENODEV; } diff --git a/arch/x86/kernel/cpu/cpufreq/e_powersaver.c b/arch/x86/kernel/cpu/cpufreq/e_powersaver.c index c2f930d8664..3f83ea12c47 100644 --- a/arch/x86/kernel/cpu/cpufreq/e_powersaver.c +++ b/arch/x86/kernel/cpu/cpufreq/e_powersaver.c @@ -12,12 +12,12 @@ #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> -#include <asm/timex.h> -#include <asm/io.h> -#include <asm/delay.h> #define EPS_BRAND_C7M 0 #define EPS_BRAND_C7 1 @@ -184,7 +184,7 @@ static int eps_cpu_init(struct cpufreq_policy *policy) break; } - switch(brand) { + switch (brand) { case EPS_BRAND_C7M: printk(KERN_CONT "C7-M\n"); break; @@ -218,17 +218,20 @@ static int eps_cpu_init(struct cpufreq_policy *policy) /* 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); + 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); + 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); + 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); @@ -318,7 +321,7 @@ static int eps_cpu_exit(struct cpufreq_policy *policy) return 0; } -static struct freq_attr* eps_attr[] = { +static struct freq_attr *eps_attr[] = { &cpufreq_freq_attr_scaling_available_freqs, NULL, }; @@ -356,7 +359,7 @@ static void __exit eps_exit(void) cpufreq_unregister_driver(&eps_driver); } -MODULE_AUTHOR("Rafa³ Bilski <rafalbilski@interia.pl>"); +MODULE_AUTHOR("Rafal Bilski <rafalbilski@interia.pl>"); MODULE_DESCRIPTION("Enhanced PowerSaver driver for VIA C7 CPU's."); MODULE_LICENSE("GPL"); diff --git a/arch/x86/kernel/cpu/cpufreq/elanfreq.c b/arch/x86/kernel/cpu/cpufreq/elanfreq.c index fe613c93b36..006b278b0d5 100644 --- a/arch/x86/kernel/cpu/cpufreq/elanfreq.c +++ b/arch/x86/kernel/cpu/cpufreq/elanfreq.c @@ -184,7 +184,8 @@ static int elanfreq_target(struct cpufreq_policy *policy, { unsigned int newstate = 0; - if (cpufreq_frequency_table_target(policy, &elanfreq_table[0], target_freq, relation, &newstate)) + if (cpufreq_frequency_table_target(policy, &elanfreq_table[0], + target_freq, relation, &newstate)) return -EINVAL; elanfreq_set_cpu_state(newstate); @@ -301,7 +302,8 @@ static void __exit elanfreq_exit(void) module_param(max_freq, int, 0444); MODULE_LICENSE("GPL"); -MODULE_AUTHOR("Robert Schwebel <r.schwebel@pengutronix.de>, Sven Geggus <sven@geggus.net>"); +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); diff --git a/arch/x86/kernel/cpu/cpufreq/gx-suspmod.c b/arch/x86/kernel/cpu/cpufreq/gx-suspmod.c index 9d9eae82e60..ac27ec2264d 100644 --- a/arch/x86/kernel/cpu/cpufreq/gx-suspmod.c +++ b/arch/x86/kernel/cpu/cpufreq/gx-suspmod.c @@ -79,8 +79,9 @@ #include <linux/smp.h> #include <linux/cpufreq.h> #include <linux/pci.h> +#include <linux/errno.h> + #include <asm/processor-cyrix.h> -#include <asm/errno.h> /* PCI config registers, all at F0 */ #define PCI_PMER1 0x80 /* power management enable register 1 */ @@ -122,8 +123,8 @@ 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 = 0; -module_param (pci_busclk, int, 0444); +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 @@ -132,7 +133,7 @@ module_param (pci_busclk, int, 0444); * 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); +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) @@ -140,7 +141,8 @@ module_param (max_duration, int, 0444); #define POLICY_MIN_DIV 20 -#define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, "gx-suspmod", msg) +#define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, \ + "gx-suspmod", msg) /** * we can detect a core multipiler from dir0_lsb @@ -166,12 +168,20 @@ static int gx_freq_mult[16] = { * Low Level chipset interface * ****************************************************************/ static struct pci_device_id gx_chipset_tbl[] __initdata = { - { PCI_VENDOR_ID_CYRIX, PCI_DEVICE_ID_CYRIX_5530_LEGACY, PCI_ANY_ID, PCI_ANY_ID }, - { PCI_VENDOR_ID_CYRIX, PCI_DEVICE_ID_CYRIX_5520, PCI_ANY_ID, PCI_ANY_ID }, - { PCI_VENDOR_ID_CYRIX, PCI_DEVICE_ID_CYRIX_5510, PCI_ANY_ID, PCI_ANY_ID }, + { PCI_VENDOR_ID_CYRIX, PCI_DEVICE_ID_CYRIX_5530_LEGACY, + PCI_ANY_ID, PCI_ANY_ID }, + { PCI_VENDOR_ID_CYRIX, PCI_DEVICE_ID_CYRIX_5520, + PCI_ANY_ID, PCI_ANY_ID }, + { PCI_VENDOR_ID_CYRIX, PCI_DEVICE_ID_CYRIX_5510, + PCI_ANY_ID, PCI_ANY_ID }, { 0, }, }; +static void gx_write_byte(int reg, int value) +{ + pci_write_config_byte(gx_params->cs55x0, reg, value); +} + /** * gx_detect_chipset: * @@ -200,7 +210,8 @@ static __init struct pci_dev *gx_detect_chipset(void) /** * gx_get_cpuspeed: * - * Finds out at which efficient frequency the Cyrix MediaGX/NatSemi Geode CPU runs. + * Finds out at which efficient frequency the Cyrix MediaGX/NatSemi + * Geode CPU runs. */ static unsigned int gx_get_cpuspeed(unsigned int cpu) { @@ -217,17 +228,18 @@ static unsigned int gx_get_cpuspeed(unsigned int cpu) * **/ -static unsigned int gx_validate_speed(unsigned int khz, u8 *on_duration, u8 *off_duration) +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; + *off_duration = 1; + *on_duration = 0; - for (i=max_duration; i>0; i--) { + 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; @@ -259,26 +271,34 @@ static void gx_set_cpuspeed(unsigned int khz) freqs.cpu = 0; freqs.old = gx_get_cpuspeed(0); - new_khz = gx_validate_speed(khz, &gx_params->on_duration, &gx_params->off_duration); + 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 */ + + + 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 */ - pci_write_config_byte(gx_params->cs55x0, PCI_IRQTC, 4); /* typical 2 to 4ms */ - pci_write_config_byte(gx_params->cs55x0, PCI_VIDTC, 100);/* typical 50 to 100ms */ - pci_write_config_byte(gx_params->cs55x0, 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; + /* 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: @@ -294,13 +314,13 @@ static void gx_set_cpuspeed(unsigned int khz) suscfg = gx_params->pci_suscfg & ~(SUSMOD); gx_params->off_duration = 0; gx_params->on_duration = 0; - dprintk("suspend modulation disabled: cpu runs 100 percent speed.\n"); + dprintk("suspend modulation disabled: cpu runs 100%% speed.\n"); } - pci_write_config_byte(gx_params->cs55x0, PCI_MODOFF, gx_params->off_duration); - pci_write_config_byte(gx_params->cs55x0, PCI_MODON, gx_params->on_duration); + gx_write_byte(PCI_MODOFF, gx_params->off_duration); + gx_write_byte(PCI_MODON, gx_params->on_duration); - pci_write_config_byte(gx_params->cs55x0, PCI_SUSCFG, suscfg); + gx_write_byte(PCI_SUSCFG, suscfg); pci_read_config_byte(gx_params->cs55x0, PCI_SUSCFG, &suscfg); local_irq_restore(flags); @@ -334,7 +354,8 @@ static int cpufreq_gx_verify(struct cpufreq_policy *policy) return -EINVAL; policy->cpu = 0; - cpufreq_verify_within_limits(policy, (stock_freq / max_duration), stock_freq); + 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 @@ -354,7 +375,8 @@ static int cpufreq_gx_verify(struct cpufreq_policy *policy) policy->max = tmp_freq; if (policy->max < policy->min) policy->max = policy->min; - cpufreq_verify_within_limits(policy, (stock_freq / max_duration), stock_freq); + cpufreq_verify_within_limits(policy, (stock_freq / max_duration), + stock_freq); return 0; } @@ -398,18 +420,18 @@ static int cpufreq_gx_cpu_init(struct cpufreq_policy *policy) return -ENODEV; /* determine maximum frequency */ - if (pci_busclk) { + if (pci_busclk) maxfreq = pci_busclk * gx_freq_mult[getCx86(CX86_DIR1) & 0x0f]; - } else if (cpu_khz) { + else if (cpu_khz) maxfreq = cpu_khz; - } else { + 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); + dprintk("cpu current frequency is %dkHz.\n", curfreq); /* setup basic struct for cpufreq API */ policy->cpu = 0; @@ -447,7 +469,8 @@ static int __init cpufreq_gx_init(void) struct pci_dev *gx_pci; /* Test if we have the right hardware */ - if ((gx_pci = gx_detect_chipset()) == NULL) + gx_pci = gx_detect_chipset(); + if (gx_pci == NULL) return -ENODEV; /* check whether module parameters are sane */ @@ -468,9 +491,11 @@ static int __init cpufreq_gx_init(void) 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)); + pci_read_config_byte(params->cs55x0, PCI_MODOFF, + &(params->off_duration)); - if ((ret = cpufreq_register_driver(&gx_suspmod_driver))) { + ret = cpufreq_register_driver(&gx_suspmod_driver); + if (ret) { kfree(params); return ret; /* register error! */ } @@ -485,9 +510,9 @@ static void __exit cpufreq_gx_exit(void) 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_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 index a4cff5d6e38..f1c51aea064 100644 --- a/arch/x86/kernel/cpu/cpufreq/longhaul.c +++ b/arch/x86/kernel/cpu/cpufreq/longhaul.c @@ -30,12 +30,12 @@ #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 <linux/kernel.h> #include <asm/msr.h> -#include <asm/timex.h> -#include <asm/io.h> -#include <asm/acpi.h> -#include <linux/acpi.h> #include <acpi/processor.h> #include "longhaul.h" @@ -58,7 +58,7 @@ #define USE_NORTHBRIDGE (1 << 2) static int cpu_model; -static unsigned int numscales=16; +static unsigned int numscales = 16; static unsigned int fsb; static const struct mV_pos *vrm_mV_table; @@ -67,8 +67,8 @@ 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 = NULL; -static struct acpi_processor_cx *cx = NULL; +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; @@ -78,12 +78,13 @@ static int scale_voltage; static int disable_acpi_c3; static int revid_errata; -#define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, "longhaul", msg) +#define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, \ + "longhaul", msg) /* Clock ratios multiplied by 10 */ -static int clock_ratio[32]; -static int eblcr_table[32]; +static int mults[32]; +static int eblcr[32]; static int longhaul_version; static struct cpufreq_frequency_table *longhaul_table; @@ -93,7 +94,7 @@ static char speedbuffer[8]; static char *print_speed(int speed) { if (speed < 1000) { - snprintf(speedbuffer, sizeof(speedbuffer),"%dMHz", speed); + snprintf(speedbuffer, sizeof(speedbuffer), "%dMHz", speed); return speedbuffer; } @@ -122,27 +123,28 @@ static unsigned int calc_speed(int mult) static int longhaul_get_cpu_mult(void) { - unsigned long invalue=0,lo, hi; + 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) { + 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; + invalue += 16; } - return eblcr_table[invalue]; + return eblcr[invalue]; } /* For processor with BCR2 MSR */ -static void do_longhaul1(unsigned int clock_ratio_index) +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 = clock_ratio_index & 0xff; + bcr2.bits.CLOCKMUL = mults_index & 0xff; /* Sync to timer tick */ safe_halt(); @@ -161,7 +163,7 @@ static void do_longhaul1(unsigned int clock_ratio_index) /* For processor with Longhaul MSR */ -static void do_powersaver(int cx_address, unsigned int clock_ratio_index, +static void do_powersaver(int cx_address, unsigned int mults_index, unsigned int dir) { union msr_longhaul longhaul; @@ -173,11 +175,11 @@ static void do_powersaver(int cx_address, unsigned int clock_ratio_index, longhaul.bits.RevisionKey = longhaul.bits.RevisionID; else longhaul.bits.RevisionKey = 0; - longhaul.bits.SoftBusRatio = clock_ratio_index & 0xf; - longhaul.bits.SoftBusRatio4 = (clock_ratio_index & 0x10) >> 4; + longhaul.bits.SoftBusRatio = mults_index & 0xf; + longhaul.bits.SoftBusRatio4 = (mults_index & 0x10) >> 4; /* Setup new voltage */ if (can_scale_voltage) - longhaul.bits.SoftVID = (clock_ratio_index >> 8) & 0x1f; + longhaul.bits.SoftVID = (mults_index >> 8) & 0x1f; /* Sync to timer tick */ safe_halt(); /* Raise voltage if necessary */ @@ -240,14 +242,14 @@ static void do_powersaver(int cx_address, unsigned int clock_ratio_index, /** * longhaul_set_cpu_frequency() - * @clock_ratio_index : bitpattern of the new multiplier. + * @mults_index : bitpattern of the new multiplier. * * Sets a new clock ratio. */ static void longhaul_setstate(unsigned int table_index) { - unsigned int clock_ratio_index; + unsigned int mults_index; int speed, mult; struct cpufreq_freqs freqs; unsigned long flags; @@ -256,9 +258,9 @@ static void longhaul_setstate(unsigned int table_index) u32 bm_timeout = 1000; unsigned int dir = 0; - clock_ratio_index = longhaul_table[table_index].index; + mults_index = longhaul_table[table_index].index; /* Safety precautions */ - mult = clock_ratio[clock_ratio_index & 0x1f]; + mult = mults[mults_index & 0x1f]; if (mult == -1) return; speed = calc_speed(mult); @@ -274,7 +276,7 @@ static void longhaul_setstate(unsigned int table_index) cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); - dprintk ("Setting to FSB:%dMHz Mult:%d.%dx (%s)\n", + dprintk("Setting to FSB:%dMHz Mult:%d.%dx (%s)\n", fsb, mult/10, mult%10, print_speed(speed/1000)); retry_loop: preempt_disable(); @@ -282,8 +284,8 @@ retry_loop: pic2_mask = inb(0xA1); pic1_mask = inb(0x21); /* works on C3. save mask. */ - outb(0xFF,0xA1); /* Overkill */ - outb(0xFE,0x21); /* TMR0 only */ + outb(0xFF, 0xA1); /* Overkill */ + outb(0xFE, 0x21); /* TMR0 only */ /* Wait while PCI bus is busy. */ if (acpi_regs_addr && (longhaul_flags & USE_NORTHBRIDGE @@ -312,7 +314,7 @@ retry_loop: * Software controlled multipliers only. */ case TYPE_LONGHAUL_V1: - do_longhaul1(clock_ratio_index); + do_longhaul1(mults_index); break; /* @@ -327,9 +329,9 @@ retry_loop: if (longhaul_flags & USE_ACPI_C3) { /* Don't allow wakeup */ acpi_set_register(ACPI_BITREG_BUS_MASTER_RLD, 0); - do_powersaver(cx->address, clock_ratio_index, dir); + do_powersaver(cx->address, mults_index, dir); } else { - do_powersaver(0, clock_ratio_index, dir); + do_powersaver(0, mults_index, dir); } break; } @@ -341,8 +343,8 @@ retry_loop: /* Enable bus master arbitration */ acpi_set_register(ACPI_BITREG_ARB_DISABLE, 0); } - outb(pic2_mask,0xA1); /* restore mask */ - outb(pic1_mask,0x21); + outb(pic2_mask, 0xA1); /* restore mask */ + outb(pic1_mask, 0x21); local_irq_restore(flags); preempt_enable(); @@ -392,7 +394,8 @@ retry_loop: cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); if (!bm_timeout) - printk(KERN_INFO PFX "Warning: Timeout while waiting for idle PCI bus.\n"); + printk(KERN_INFO PFX "Warning: Timeout while waiting for " + "idle PCI bus.\n"); } /* @@ -458,31 +461,32 @@ static int __init longhaul_get_ranges(void) break; } - dprintk ("MinMult:%d.%dx MaxMult:%d.%dx\n", + 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, + 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"); + printk(KERN_INFO PFX "highestspeed == lowest, aborting.\n"); return -EINVAL; } if (lowest_speed > highest_speed) { - printk (KERN_INFO PFX "nonsense! lowest (%d > %d) !\n", + printk(KERN_INFO PFX "nonsense! lowest (%d > %d) !\n", lowest_speed, highest_speed); return -EINVAL; } - longhaul_table = kmalloc((numscales + 1) * sizeof(struct cpufreq_frequency_table), GFP_KERNEL); - if(!longhaul_table) + longhaul_table = kmalloc((numscales + 1) * sizeof(*longhaul_table), + GFP_KERNEL); + if (!longhaul_table) return -ENOMEM; for (j = 0; j < numscales; j++) { - ratio = clock_ratio[j]; + ratio = mults[j]; if (ratio == -1) continue; if (ratio > maxmult || ratio < minmult) @@ -507,13 +511,10 @@ static int __init longhaul_get_ranges(void) } } if (min_i != j) { - unsigned int temp; - temp = longhaul_table[j].frequency; - longhaul_table[j].frequency = longhaul_table[min_i].frequency; - longhaul_table[min_i].frequency = temp; - temp = longhaul_table[j].index; - longhaul_table[j].index = longhaul_table[min_i].index; - longhaul_table[min_i].index = temp; + swap(longhaul_table[j].frequency, + longhaul_table[min_i].frequency); + swap(longhaul_table[j].index, + longhaul_table[min_i].index); } } @@ -521,7 +522,7 @@ static int __init longhaul_get_ranges(void) /* Find index we are running on */ for (j = 0; j < k; j++) { - if (clock_ratio[longhaul_table[j].index & 0x1f] == mult) { + if (mults[longhaul_table[j].index & 0x1f] == mult) { longhaul_index = j; break; } @@ -559,20 +560,22 @@ static void __init longhaul_setup_voltagescaling(void) 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. " + 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); + 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", + 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 */ + /* How many voltage steps*/ numvscales = maxvid.pos - minvid.pos + 1; printk(KERN_INFO PFX "Max VID=%d.%03d " @@ -586,7 +589,7 @@ static void __init longhaul_setup_voltagescaling(void) j = longhaul.bits.MinMHzBR; if (longhaul.bits.MinMHzBR4) j += 16; - min_vid_speed = eblcr_table[j]; + min_vid_speed = eblcr[j]; if (min_vid_speed == -1) return; switch (longhaul.bits.MinMHzFSB) { @@ -617,7 +620,8 @@ static void __init longhaul_setup_voltagescaling(void) 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); + printk(KERN_INFO PFX "f: %d kHz, index: %d, vid: %d mV\n", + speed, j, vid.mV); j++; } @@ -640,7 +644,8 @@ static int longhaul_target(struct cpufreq_policy *policy, unsigned int dir = 0; u8 vid, current_vid; - if (cpufreq_frequency_table_target(policy, longhaul_table, target_freq, relation, &table_index)) + if (cpufreq_frequency_table_target(policy, longhaul_table, target_freq, + relation, &table_index)) return -EINVAL; /* Don't set same frequency again */ @@ -656,7 +661,8 @@ static int longhaul_target(struct cpufreq_policy *policy, * 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) & 0x1f; + current_vid = (longhaul_table[longhaul_index].index >> 8); + current_vid &= 0x1f; if (table_index > longhaul_index) dir = 1; while (i != table_index) { @@ -691,9 +697,9 @@ static acpi_status longhaul_walk_callback(acpi_handle obj_handle, { struct acpi_device *d; - if ( acpi_bus_get_device(obj_handle, &d) ) { + if (acpi_bus_get_device(obj_handle, &d)) return 0; - } + *return_value = acpi_driver_data(d); return 1; } @@ -750,7 +756,7 @@ static int longhaul_setup_southbridge(void) /* Find VT8235 southbridge */ dev = pci_get_device(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8235, NULL); if (dev == NULL) - /* Find VT8237 southbridge */ + /* Find VT8237 southbridge */ dev = pci_get_device(PCI_VENDOR_ID_VIA, PCI_DEVICE_ID_VIA_8237, NULL); if (dev != NULL) { @@ -769,7 +775,8 @@ static int longhaul_setup_southbridge(void) 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); + printk(KERN_INFO PFX "ACPI I/O at 0x%x\n", + acpi_regs_addr); } pci_dev_put(dev); @@ -781,7 +788,7 @@ static int longhaul_setup_southbridge(void) static int __init longhaul_cpu_init(struct cpufreq_policy *policy) { struct cpuinfo_x86 *c = &cpu_data(0); - char *cpuname=NULL; + char *cpuname = NULL; int ret; u32 lo, hi; @@ -791,8 +798,8 @@ static int __init longhaul_cpu_init(struct cpufreq_policy *policy) cpu_model = CPU_SAMUEL; cpuname = "C3 'Samuel' [C5A]"; longhaul_version = TYPE_LONGHAUL_V1; - memcpy (clock_ratio, samuel1_clock_ratio, sizeof(samuel1_clock_ratio)); - memcpy (eblcr_table, samuel1_eblcr, sizeof(samuel1_eblcr)); + memcpy(mults, samuel1_mults, sizeof(samuel1_mults)); + memcpy(eblcr, samuel1_eblcr, sizeof(samuel1_eblcr)); break; case 7: @@ -803,10 +810,8 @@ static int __init longhaul_cpu_init(struct cpufreq_policy *policy) cpuname = "C3 'Samuel 2' [C5B]"; /* Note, this is not a typo, early Samuel2's had * Samuel1 ratios. */ - memcpy(clock_ratio, samuel1_clock_ratio, - sizeof(samuel1_clock_ratio)); - memcpy(eblcr_table, samuel2_eblcr, - sizeof(samuel2_eblcr)); + memcpy(mults, samuel1_mults, sizeof(samuel1_mults)); + memcpy(eblcr, samuel2_eblcr, sizeof(samuel2_eblcr)); break; case 1 ... 15: longhaul_version = TYPE_LONGHAUL_V1; @@ -817,10 +822,8 @@ static int __init longhaul_cpu_init(struct cpufreq_policy *policy) cpu_model = CPU_EZRA; cpuname = "C3 'Ezra' [C5C]"; } - memcpy(clock_ratio, ezra_clock_ratio, - sizeof(ezra_clock_ratio)); - memcpy(eblcr_table, ezra_eblcr, - sizeof(ezra_eblcr)); + memcpy(mults, ezra_mults, sizeof(ezra_mults)); + memcpy(eblcr, ezra_eblcr, sizeof(ezra_eblcr)); break; } break; @@ -829,18 +832,16 @@ static int __init longhaul_cpu_init(struct cpufreq_policy *policy) cpu_model = CPU_EZRA_T; cpuname = "C3 'Ezra-T' [C5M]"; longhaul_version = TYPE_POWERSAVER; - numscales=32; - memcpy (clock_ratio, ezrat_clock_ratio, sizeof(ezrat_clock_ratio)); - memcpy (eblcr_table, ezrat_eblcr, sizeof(ezrat_eblcr)); + 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(clock_ratio, - nehemiah_clock_ratio, - sizeof(nehemiah_clock_ratio)); - memcpy(eblcr_table, nehemiah_eblcr, sizeof(nehemiah_eblcr)); + 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; @@ -869,14 +870,14 @@ static int __init longhaul_cpu_init(struct cpufreq_policy *policy) longhaul_version = TYPE_LONGHAUL_V1; } - printk (KERN_INFO PFX "VIA %s CPU detected. ", cpuname); + printk(KERN_INFO PFX "VIA %s CPU detected. ", cpuname); switch (longhaul_version) { case TYPE_LONGHAUL_V1: case TYPE_LONGHAUL_V2: - printk ("Longhaul v%d supported.\n", longhaul_version); + printk(KERN_CONT "Longhaul v%d supported.\n", longhaul_version); break; case TYPE_POWERSAVER: - printk ("Powersaver supported.\n"); + printk(KERN_CONT "Powersaver supported.\n"); break; }; @@ -940,7 +941,7 @@ static int __devexit longhaul_cpu_exit(struct cpufreq_policy *policy) return 0; } -static struct freq_attr* longhaul_attr[] = { +static struct freq_attr *longhaul_attr[] = { &cpufreq_freq_attr_scaling_available_freqs, NULL, }; @@ -966,13 +967,15 @@ static int __init longhaul_init(void) #ifdef CONFIG_SMP if (num_online_cpus() > 1) { - printk(KERN_ERR PFX "More than 1 CPU detected, longhaul disabled.\n"); + 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"); + printk(KERN_ERR PFX "APIC detected. Longhaul is currently " + "broken in this configuration.\n"); return -ENODEV; } #endif @@ -993,8 +996,8 @@ static void __exit longhaul_exit(void) { int i; - for (i=0; i < numscales; i++) { - if (clock_ratio[i] == maxmult) { + for (i = 0; i < numscales; i++) { + if (mults[i] == maxmult) { longhaul_setstate(i); break; } @@ -1007,11 +1010,11 @@ static void __exit longhaul_exit(void) /* 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_param(disable_acpi_c3, int, 0644); MODULE_PARM_DESC(disable_acpi_c3, "Don't use ACPI C3 support"); /* Change CPU voltage with frequency. Very usefull to save * power, but most VIA C3 processors aren't supporting it. */ -module_param (scale_voltage, int, 0644); +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 @@ -1019,9 +1022,9 @@ MODULE_PARM_DESC(scale_voltage, "Scale voltage of processor"); 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"); +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 index 4fcc320997d..e2360a469f7 100644 --- a/arch/x86/kernel/cpu/cpufreq/longhaul.h +++ b/arch/x86/kernel/cpu/cpufreq/longhaul.h @@ -49,14 +49,14 @@ union msr_longhaul { /* * Clock ratio tables. Div/Mod by 10 to get ratio. - * The eblcr ones specify the ratio read from the CPU. - * The clock_ratio ones specify what to write to the CPU. + * 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 __initdata samuel1_clock_ratio[16] = { +static const int __initdata samuel1_mults[16] = { -1, /* 0000 -> RESERVED */ 30, /* 0001 -> 3.0x */ 40, /* 0010 -> 4.0x */ @@ -119,7 +119,7 @@ static const int __initdata samuel2_eblcr[16] = { /* * VIA C3 Ezra */ -static const int __initdata ezra_clock_ratio[16] = { +static const int __initdata ezra_mults[16] = { 100, /* 0000 -> 10.0x */ 30, /* 0001 -> 3.0x */ 40, /* 0010 -> 4.0x */ @@ -160,7 +160,7 @@ static const int __initdata ezra_eblcr[16] = { /* * VIA C3 (Ezra-T) [C5M]. */ -static const int __initdata ezrat_clock_ratio[32] = { +static const int __initdata ezrat_mults[32] = { 100, /* 0000 -> 10.0x */ 30, /* 0001 -> 3.0x */ 40, /* 0010 -> 4.0x */ @@ -235,7 +235,7 @@ static const int __initdata ezrat_eblcr[32] = { /* * VIA C3 Nehemiah */ -static const int __initdata nehemiah_clock_ratio[32] = { +static const int __initdata nehemiah_mults[32] = { 100, /* 0000 -> 10.0x */ -1, /* 0001 -> 16.0x */ 40, /* 0010 -> 4.0x */ diff --git a/arch/x86/kernel/cpu/cpufreq/longrun.c b/arch/x86/kernel/cpu/cpufreq/longrun.c index 777a7ff075d..da5f70fcb76 100644 --- a/arch/x86/kernel/cpu/cpufreq/longrun.c +++ b/arch/x86/kernel/cpu/cpufreq/longrun.c @@ -11,12 +11,13 @@ #include <linux/init.h> #include <linux/slab.h> #include <linux/cpufreq.h> +#include <linux/timex.h> #include <asm/msr.h> #include <asm/processor.h> -#include <asm/timex.h> -#define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, "longrun", msg) +#define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, \ + "longrun", msg) static struct cpufreq_driver longrun_driver; @@ -51,7 +52,7 @@ static void __init longrun_get_policy(struct cpufreq_policy *policy) msr_lo &= 0x0000007F; msr_hi &= 0x0000007F; - if ( longrun_high_freq <= longrun_low_freq ) { + if (longrun_high_freq <= longrun_low_freq) { /* Assume degenerate Longrun table */ policy->min = policy->max = longrun_high_freq; } else { @@ -79,7 +80,7 @@ static int longrun_set_policy(struct cpufreq_policy *policy) if (!policy) return -EINVAL; - if ( longrun_high_freq <= longrun_low_freq ) { + if (longrun_high_freq <= longrun_low_freq) { /* Assume degenerate Longrun table */ pctg_lo = pctg_hi = 100; } else { @@ -152,7 +153,7 @@ static unsigned int longrun_get(unsigned int cpu) cpuid(0x80860007, &eax, &ebx, &ecx, &edx); dprintk("cpuid eax is %u\n", eax); - return (eax * 1000); + return eax * 1000; } /** @@ -196,7 +197,8 @@ static unsigned int __init longrun_determine_freqs(unsigned int *low_freq, 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); + dprintk("longrun table interface told %u - %u kHz\n", + *low_freq, *high_freq); if (*low_freq > *high_freq) *low_freq = *high_freq; @@ -219,7 +221,7 @@ static unsigned int __init longrun_determine_freqs(unsigned int *low_freq, 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) { + 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; @@ -236,7 +238,7 @@ static unsigned int __init longrun_determine_freqs(unsigned int *low_freq, /* performance_pctg = (current_freq - low_freq)/(high_freq - low_freq) * eqals - * low_freq * ( 1 - perf_pctg) = (cur_freq - high_freq * perf_pctg) + * low_freq * (1 - perf_pctg) = (cur_freq - high_freq * perf_pctg) * * high_freq * perf_pctg is stored tempoarily into "ebx". */ @@ -317,9 +319,10 @@ static void __exit longrun_exit(void) } -MODULE_AUTHOR ("Dominik Brodowski <linux@brodo.de>"); -MODULE_DESCRIPTION ("LongRun driver for Transmeta Crusoe and Efficeon processors."); -MODULE_LICENSE ("GPL"); +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/p4-clockmod.c b/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c index b585e04cbc9..41ed94915f9 100644 --- a/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c +++ b/arch/x86/kernel/cpu/cpufreq/p4-clockmod.c @@ -27,15 +27,17 @@ #include <linux/cpufreq.h> #include <linux/slab.h> #include <linux/cpumask.h> +#include <linux/timex.h> #include <asm/processor.h> #include <asm/msr.h> -#include <asm/timex.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) +#define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, \ + "p4-clockmod", msg) /* * Duty Cycle (3bits), note DC_DISABLE is not specified in @@ -58,7 +60,8 @@ static int cpufreq_p4_setdc(unsigned int cpu, unsigned int newstate) { u32 l, h; - if (!cpu_online(cpu) || (newstate > DC_DISABLE) || (newstate == DC_RESV)) + if (!cpu_online(cpu) || + (newstate > DC_DISABLE) || (newstate == DC_RESV)) return -EINVAL; rdmsr_on_cpu(cpu, MSR_IA32_THERM_STATUS, &l, &h); @@ -66,7 +69,8 @@ static int cpufreq_p4_setdc(unsigned int cpu, unsigned int newstate) if (l & 0x01) dprintk("CPU#%d currently thermal throttled\n", cpu); - if (has_N44_O17_errata[cpu] && (newstate == DC_25PT || newstate == DC_DFLT)) + 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); @@ -112,7 +116,8 @@ static int cpufreq_p4_target(struct cpufreq_policy *policy, struct cpufreq_freqs freqs; int i; - if (cpufreq_frequency_table_target(policy, &p4clockmod_table[0], target_freq, relation, &newstate)) + if (cpufreq_frequency_table_target(policy, &p4clockmod_table[0], + target_freq, relation, &newstate)) return -EINVAL; freqs.old = cpufreq_p4_get(policy->cpu); @@ -127,7 +132,8 @@ static int cpufreq_p4_target(struct cpufreq_policy *policy, cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); } - /* run on each logical CPU, see section 13.15.3 of IA32 Intel Architecture Software + /* 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) @@ -153,28 +159,30 @@ static unsigned int cpufreq_p4_get_frequency(struct cpuinfo_x86 *c) { if (c->x86 == 0x06) { if (cpu_has(c, X86_FEATURE_EST)) - printk(KERN_WARNING PFX "Warning: EST-capable CPU detected. " - "The acpi-cpufreq module offers voltage scaling" - " in addition of frequency scaling. You should use " - "that instead of p4-clockmod, if possible.\n"); + printk(KERN_WARNING PFX "Warning: EST-capable CPU " + "detected. The acpi-cpufreq module offers " + "voltage scaling in addition of 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 */ p4clockmod_driver.flags |= CPUFREQ_CONST_LOOPS; - return speedstep_get_processor_frequency(SPEEDSTEP_PROCESSOR_PCORE); + 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_processor_frequency(SPEEDSTEP_PROCESSOR_PM); + return speedstep_get_frequency(SPEEDSTEP_CPU_PM); } } if (c->x86 != 0xF) { if (!cpu_has(c, X86_FEATURE_EST)) - printk(KERN_WARNING PFX "Unknown p4-clockmod-capable CPU. " - "Please send an e-mail to <cpufreq@vger.kernel.org>\n"); + printk(KERN_WARNING PFX "Unknown CPU. " + "Please send an e-mail to " + "<cpufreq@vger.kernel.org>\n"); return 0; } @@ -182,16 +190,16 @@ static unsigned int cpufreq_p4_get_frequency(struct cpuinfo_x86 *c) * throttling is active or not. */ p4clockmod_driver.flags |= CPUFREQ_CONST_LOOPS; - if (speedstep_detect_processor() == SPEEDSTEP_PROCESSOR_P4M) { + 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_processor_frequency(SPEEDSTEP_PROCESSOR_P4M); + return speedstep_get_frequency(SPEEDSTEP_CPU_P4M); } - return speedstep_get_processor_frequency(SPEEDSTEP_PROCESSOR_P4D); + return speedstep_get_frequency(SPEEDSTEP_CPU_P4D); } @@ -217,14 +225,20 @@ static int cpufreq_p4_cpu_init(struct cpufreq_policy *policy) 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])) + 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; @@ -232,7 +246,10 @@ static int cpufreq_p4_cpu_init(struct cpufreq_policy *policy) cpufreq_frequency_table_get_attr(p4clockmod_table, policy->cpu); /* cpuinfo and default policy values */ - policy->cpuinfo.transition_latency = 1000000; /* assumed */ + + /* 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]); @@ -258,12 +275,12 @@ static unsigned int cpufreq_p4_get(unsigned int cpu) l = DC_DISABLE; if (l != DC_DISABLE) - return (stock_freq * l / 8); + return stock_freq * l / 8; return stock_freq; } -static struct freq_attr* p4clockmod_attr[] = { +static struct freq_attr *p4clockmod_attr[] = { &cpufreq_freq_attr_scaling_available_freqs, NULL, }; @@ -277,7 +294,6 @@ static struct cpufreq_driver p4clockmod_driver = { .name = "p4-clockmod", .owner = THIS_MODULE, .attr = p4clockmod_attr, - .hide_interface = 1, }; @@ -299,9 +315,10 @@ static int __init cpufreq_p4_init(void) ret = cpufreq_register_driver(&p4clockmod_driver); if (!ret) - printk(KERN_INFO PFX "P4/Xeon(TM) CPU On-Demand Clock Modulation available\n"); + printk(KERN_INFO PFX "P4/Xeon(TM) CPU On-Demand Clock " + "Modulation available\n"); - return (ret); + return ret; } @@ -311,9 +328,9 @@ static void __exit cpufreq_p4_exit(void) } -MODULE_AUTHOR ("Zwane Mwaikambo <zwane@commfireservices.com>"); -MODULE_DESCRIPTION ("cpufreq driver for Pentium(TM) 4/Xeon(TM)"); -MODULE_LICENSE ("GPL"); +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/powernow-k6.c b/arch/x86/kernel/cpu/cpufreq/powernow-k6.c index c1ac5790c63..f10dea409f4 100644 --- a/arch/x86/kernel/cpu/cpufreq/powernow-k6.c +++ b/arch/x86/kernel/cpu/cpufreq/powernow-k6.c @@ -1,6 +1,7 @@ /* * 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. + * (C) 2000-2003 Dave Jones, Arjan van de Ven, Janne Pänkälä, + * Dominik Brodowski. * * Licensed under the terms of the GNU GPL License version 2. * @@ -13,14 +14,15 @@ #include <linux/cpufreq.h> #include <linux/ioport.h> #include <linux/slab.h> - -#include <asm/msr.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; @@ -47,8 +49,8 @@ static struct cpufreq_frequency_table clock_ratio[] = { */ static int powernow_k6_get_cpu_multiplier(void) { - u64 invalue = 0; - u32 msrval; + u64 invalue = 0; + u32 msrval; msrval = POWERNOW_IOPORT + 0x1; wrmsr(MSR_K6_EPMR, msrval, 0); /* enable the PowerNow port */ @@ -68,12 +70,12 @@ static int powernow_k6_get_cpu_multiplier(void) */ static void powernow_k6_set_state(unsigned int best_i) { - unsigned long outvalue = 0, invalue = 0; - unsigned long msrval; - struct cpufreq_freqs freqs; + unsigned long outvalue = 0, invalue = 0; + unsigned long msrval; + struct cpufreq_freqs freqs; if (clock_ratio[best_i].index > max_multiplier) { - printk(KERN_ERR "cpufreq: invalid target frequency\n"); + printk(KERN_ERR PFX "invalid target frequency\n"); return; } @@ -119,7 +121,8 @@ static int powernow_k6_verify(struct cpufreq_policy *policy) * 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) + * @relation: how that frequency relates to achieved frequency + * (CPUFREQ_RELATION_L or CPUFREQ_RELATION_H) * * sets a new CPUFreq policy */ @@ -127,9 +130,10 @@ static int powernow_k6_target(struct cpufreq_policy *policy, unsigned int target_freq, unsigned int relation) { - unsigned int newstate = 0; + unsigned int newstate = 0; - if (cpufreq_frequency_table_target(policy, &clock_ratio[0], target_freq, relation, &newstate)) + if (cpufreq_frequency_table_target(policy, &clock_ratio[0], + target_freq, relation, &newstate)) return -EINVAL; powernow_k6_set_state(newstate); @@ -140,7 +144,7 @@ static int powernow_k6_target(struct cpufreq_policy *policy, static int powernow_k6_cpu_init(struct cpufreq_policy *policy) { - unsigned int i; + unsigned int i, f; int result; if (policy->cpu != 0) @@ -152,10 +156,11 @@ static int powernow_k6_cpu_init(struct cpufreq_policy *policy) /* table init */ for (i = 0; (clock_ratio[i].frequency != CPUFREQ_TABLE_END); i++) { - if (clock_ratio[i].index > max_multiplier) + f = clock_ratio[i].index; + if (f > max_multiplier) clock_ratio[i].frequency = CPUFREQ_ENTRY_INVALID; else - clock_ratio[i].frequency = busfreq * clock_ratio[i].index; + clock_ratio[i].frequency = busfreq * f; } /* cpuinfo and default policy values */ @@ -185,7 +190,9 @@ static int powernow_k6_cpu_exit(struct cpufreq_policy *policy) static unsigned int powernow_k6_get(unsigned int cpu) { - return busfreq * powernow_k6_get_cpu_multiplier(); + unsigned int ret; + ret = (busfreq * powernow_k6_get_cpu_multiplier()); + return ret; } static struct freq_attr *powernow_k6_attr[] = { @@ -221,7 +228,7 @@ static int __init powernow_k6_init(void) return -ENODEV; if (!request_region(POWERNOW_IOPORT, 16, "PowerNow!")) { - printk("cpufreq: PowerNow IOPORT region already used.\n"); + printk(KERN_INFO PFX "PowerNow IOPORT region already used.\n"); return -EIO; } @@ -246,7 +253,8 @@ static void __exit powernow_k6_exit(void) } -MODULE_AUTHOR("Arjan van de Ven, Dave Jones <davej@redhat.com>, Dominik Brodowski <linux@brodo.de>"); +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"); diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k7.c b/arch/x86/kernel/cpu/cpufreq/powernow-k7.c index 1b446d79a8f..3c28ccd4974 100644 --- a/arch/x86/kernel/cpu/cpufreq/powernow-k7.c +++ b/arch/x86/kernel/cpu/cpufreq/powernow-k7.c @@ -6,10 +6,12 @@ * Licensed under the terms of the GNU GPL License version 2. * Based upon datasheets & sample CPUs kindly provided by AMD. * - * Errata 5: Processor 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: Processors with half frequency multipliers may hang upon wakeup from disconnect. - * - We disable half multipliers if ACPI is used on A0 stepping CPUs. + * 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> @@ -20,11 +22,11 @@ #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/timer.h> -#include <asm/timex.h> -#include <asm/io.h> #include <asm/system.h> #ifdef CONFIG_X86_POWERNOW_K7_ACPI @@ -58,9 +60,9 @@ struct pst_s { union powernow_acpi_control_t { struct { unsigned long fid:5, - vid:5, - sgtc:20, - res1:2; + vid:5, + sgtc:20, + res1:2; } bits; unsigned long val; }; @@ -94,14 +96,15 @@ 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 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) +#define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, \ + "powernow-k7", msg) static int check_fsb(unsigned int fsbspeed) { @@ -109,7 +112,7 @@ static int check_fsb(unsigned int fsbspeed) unsigned int f = fsb / 1000; delta = (fsbspeed > f) ? fsbspeed - f : f - fsbspeed; - return (delta < 5); + return delta < 5; } static int check_powernow(void) @@ -117,24 +120,26 @@ 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)) { + 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"); + printk(KERN_INFO PFX "This module only works with " + "AMD K7 CPUs\n"); #endif return 0; } /* Get maximum capabilities */ - maxei = cpuid_eax (0x80000000); + maxei = cpuid_eax(0x80000000); if (maxei < 0x80000007) { /* Any powernow info ? */ #ifdef MODULE - printk (KERN_INFO PFX "No powernow capabilities detected\n"); + 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"); + printk(KERN_INFO PFX "K7 660[A0] core detected, " + "enabling errata workarounds\n"); have_a0 = 1; } @@ -144,37 +149,42 @@ static int check_powernow(void) if (!(edx & (1 << 1 | 1 << 2))) return 0; - printk (KERN_INFO PFX "PowerNOW! Technology present. Can scale: "); + printk(KERN_INFO PFX "PowerNOW! Technology present. Can scale: "); if (edx & 1 << 1) { - printk ("frequency"); - can_scale_bus=1; + printk("frequency"); + can_scale_bus = 1; } if ((edx & (1 << 1 | 1 << 2)) == 0x6) - printk (" and "); + printk(" and "); if (edx & 1 << 2) { - printk ("voltage"); - can_scale_vid=1; + printk("voltage"); + can_scale_vid = 1; } - printk (".\n"); + printk(".\n"); return 1; } +static void invalidate_entry(unsigned int entry) +{ + powernow_table[entry].frequency = CPUFREQ_ENTRY_INVALID; +} -static int get_ranges (unsigned char *pst) +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); + 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++) { + for (j = 0 ; j < number_scales; j++) { fid = *pst++; powernow_table[j].frequency = (fsb * fid_codes[fid]) / 10; @@ -182,10 +192,10 @@ static int get_ranges (unsigned char *pst) speed = powernow_table[j].frequency; - if ((fid_codes[fid] % 10)==5) { + if ((fid_codes[fid] % 10) == 5) { #ifdef CONFIG_X86_POWERNOW_K7_ACPI if (have_a0 == 1) - powernow_table[j].frequency = CPUFREQ_ENTRY_INVALID; + invalidate_entry(j); #endif } @@ -197,7 +207,7 @@ static int get_ranges (unsigned char *pst) vid = *pst++; powernow_table[j].index |= (vid << 8); /* upper 8 bits */ - dprintk (" FID: 0x%x (%d.%dx [%dMHz]) " + 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, @@ -214,13 +224,13 @@ static void change_FID(int fid) { union msr_fidvidctl fidvidctl; - rdmsrl (MSR_K7_FID_VID_CTL, fidvidctl.val); + 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); + wrmsrl(MSR_K7_FID_VID_CTL, fidvidctl.val); } } @@ -229,18 +239,18 @@ static void change_VID(int vid) { union msr_fidvidctl fidvidctl; - rdmsrl (MSR_K7_FID_VID_CTL, fidvidctl.val); + 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); + wrmsrl(MSR_K7_FID_VID_CTL, fidvidctl.val); } } -static void change_speed (unsigned int index) +static void change_speed(unsigned int index) { u8 fid, vid; struct cpufreq_freqs freqs; @@ -257,7 +267,7 @@ static void change_speed (unsigned int index) freqs.cpu = 0; - rdmsrl (MSR_K7_FID_VID_STATUS, fidvidstatus.val); + rdmsrl(MSR_K7_FID_VID_STATUS, fidvidstatus.val); cfid = fidvidstatus.bits.CFID; freqs.old = fsb * fid_codes[cfid] / 10; @@ -321,12 +331,14 @@ static int powernow_acpi_init(void) goto err1; } - if (acpi_processor_perf->control_register.space_id != ACPI_ADR_SPACE_FIXED_HARDWARE) { + 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) { + if (acpi_processor_perf->status_register.space_id != + ACPI_ADR_SPACE_FIXED_HARDWARE) { retval = -ENODEV; goto err2; } @@ -338,7 +350,8 @@ static int powernow_acpi_init(void) goto err2; } - powernow_table = kzalloc((number_scales + 1) * (sizeof(struct cpufreq_frequency_table)), GFP_KERNEL); + powernow_table = kzalloc((sizeof(struct cpufreq_frequency_table) * + (number_scales + 1)), GFP_KERNEL); if (!powernow_table) { retval = -ENOMEM; goto err2; @@ -352,7 +365,7 @@ static int powernow_acpi_init(void) 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", + dprintk("acpi: P%d: %d MHz %d mW %d uS control %08x SGTC %d\n", i, (u32) state->core_frequency, (u32) state->power, @@ -381,12 +394,12 @@ static int powernow_acpi_init(void) if (speed % 1000 > 0) speed_mhz++; - if ((fid_codes[fid] % 10)==5) { + if ((fid_codes[fid] % 10) == 5) { if (have_a0 == 1) - powernow_table[i].frequency = CPUFREQ_ENTRY_INVALID; + invalidate_entry(i); } - dprintk (" FID: 0x%x (%d.%dx [%dMHz]) " + 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, @@ -422,7 +435,8 @@ err1: err05: kfree(acpi_processor_perf); err0: - printk(KERN_WARNING PFX "ACPI perflib can not be used in this platform\n"); + printk(KERN_WARNING PFX "ACPI perflib can not be used on " + "this platform\n"); acpi_processor_perf = NULL; return retval; } @@ -435,7 +449,14 @@ static int powernow_acpi_init(void) } #endif -static int powernow_decode_bios (int maxfid, int startvid) +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; @@ -446,61 +467,67 @@ static int powernow_decode_bios (int maxfid, int startvid) etuple = cpuid_eax(0x80000001); - for (i=0xC0000; i < 0xffff0 ; i+=16) { + 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); + 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); + 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"); + 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"); - } + 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); + 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); + 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); + p += sizeof(struct psb_s); pst = (struct pst_s *) p; - for (j=0; j<psb->numpst; j++) { + 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)) - { - 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); - - ret = get_ranges ((char *) pst + sizeof (struct pst_s)); + 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; + 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"); + 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; } @@ -511,13 +538,14 @@ static int powernow_decode_bios (int maxfid, int startvid) } -static int powernow_target (struct cpufreq_policy *policy, +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)) + if (cpufreq_frequency_table_target(policy, powernow_table, target_freq, + relation, &newstate)) return -EINVAL; change_speed(newstate); @@ -526,7 +554,7 @@ static int powernow_target (struct cpufreq_policy *policy, } -static int powernow_verify (struct cpufreq_policy *policy) +static int powernow_verify(struct cpufreq_policy *policy) { return cpufreq_frequency_table_verify(policy, powernow_table); } @@ -566,18 +594,23 @@ static unsigned int powernow_get(unsigned int cpu) if (cpu) return 0; - rdmsrl (MSR_K7_FID_VID_STATUS, fidvidstatus.val); + rdmsrl(MSR_K7_FID_VID_STATUS, fidvidstatus.val); cfid = fidvidstatus.bits.CFID; - return (fsb * fid_codes[cfid] / 10); + return fsb * fid_codes[cfid] / 10; } static int __init acer_cpufreq_pst(const struct dmi_system_id *d) { - printk(KERN_WARNING "%s laptop with broken PST tables in BIOS detected.\n", d->ident); - printk(KERN_WARNING "You need to downgrade to 3A21 (09/09/2002), or try a newer BIOS than 3A71 (01/20/2003)\n"); - printk(KERN_WARNING "cpufreq scaling has been disabled as a result of this.\n"); + 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; } @@ -598,7 +631,7 @@ static struct dmi_system_id __initdata powernow_dmi_table[] = { { } }; -static int __init powernow_cpu_init (struct cpufreq_policy *policy) +static int __init powernow_cpu_init(struct cpufreq_policy *policy) { union msr_fidvidstatus fidvidstatus; int result; @@ -606,7 +639,7 @@ static int __init powernow_cpu_init (struct cpufreq_policy *policy) if (policy->cpu != 0) return -ENODEV; - rdmsrl (MSR_K7_FID_VID_STATUS, fidvidstatus.val); + rdmsrl(MSR_K7_FID_VID_STATUS, fidvidstatus.val); recalibrate_cpu_khz(); @@ -618,19 +651,21 @@ static int __init powernow_cpu_init (struct cpufreq_policy *policy) 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"); + 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); + result = powernow_decode_bios(fidvidstatus.bits.MFID, + fidvidstatus.bits.SVID); if (result) { - printk (KERN_INFO PFX "Trying ACPI perflib\n"); + 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"); - printk (KERN_INFO PFX "See http://www.codemonkey.org.uk/projects/cpufreq/powernow-k7.html\n"); + printk(KERN_INFO PFX + "ACPI and legacy methods failed\n"); } } else { /* SGTC use the bus clock as timer */ @@ -642,10 +677,11 @@ static int __init powernow_cpu_init (struct cpufreq_policy *policy) if (result) return result; - printk (KERN_INFO PFX "Minimum speed %d MHz. Maximum speed %d MHz.\n", + 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->cpuinfo.transition_latency = + cpufreq_scale(2000000UL, fsb, latency); policy->cur = powernow_get(0); @@ -654,7 +690,8 @@ static int __init powernow_cpu_init (struct cpufreq_policy *policy) return cpufreq_frequency_table_cpuinfo(policy, powernow_table); } -static int powernow_cpu_exit (struct cpufreq_policy *policy) { +static int powernow_cpu_exit(struct cpufreq_policy *policy) +{ cpufreq_frequency_table_put_attr(policy->cpu); #ifdef CONFIG_X86_POWERNOW_K7_ACPI @@ -669,7 +706,7 @@ static int powernow_cpu_exit (struct cpufreq_policy *policy) { return 0; } -static struct freq_attr* powernow_table_attr[] = { +static struct freq_attr *powernow_table_attr[] = { &cpufreq_freq_attr_scaling_available_freqs, NULL, }; @@ -685,15 +722,15 @@ static struct cpufreq_driver powernow_driver = { .attr = powernow_table_attr, }; -static int __init powernow_init (void) +static int __init powernow_init(void) { - if (check_powernow()==0) + if (check_powernow() == 0) return -ENODEV; return cpufreq_register_driver(&powernow_driver); } -static void __exit powernow_exit (void) +static void __exit powernow_exit(void) { cpufreq_unregister_driver(&powernow_driver); } @@ -701,9 +738,9 @@ static void __exit powernow_exit (void) 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"); +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-k8.c b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c index 5c28b37dea1..a15ac94e0b9 100644 --- a/arch/x86/kernel/cpu/cpufreq/powernow-k8.c +++ b/arch/x86/kernel/cpu/cpufreq/powernow-k8.c @@ -33,16 +33,14 @@ #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 <asm/io.h> -#include <asm/delay.h> -#ifdef CONFIG_X86_POWERNOW_K8_ACPI #include <linux/acpi.h> #include <linux/mutex.h> #include <acpi/processor.h> -#endif #define PFX "powernow-k8: " #define VERSION "version 2.20.00" @@ -71,7 +69,8 @@ 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) +static u32 find_khz_freq_from_pstate(struct cpufreq_frequency_table *data, + u32 pstate) { return data[pstate].frequency; } @@ -186,7 +185,9 @@ static int write_new_fid(struct powernow_k8_data *data, u32 fid) return 1; } - lo = fid | (data->currvid << MSR_C_LO_VID_SHIFT) | MSR_C_LO_INIT_FID_VID; + 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); @@ -194,7 +195,9 @@ static int write_new_fid(struct powernow_k8_data *data, u32 fid) 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"); + 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)); @@ -202,14 +205,16 @@ static int write_new_fid(struct powernow_k8_data *data, u32 fid) 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); + 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); + printk(KERN_ERR PFX + "fid trans failed, fid 0x%x, curr 0x%x\n", fid, + data->currfid); return 1; } @@ -228,7 +233,9 @@ static int write_new_vid(struct powernow_k8_data *data, u32 vid) return 1; } - lo = data->currfid | (vid << MSR_C_LO_VID_SHIFT) | MSR_C_LO_INIT_FID_VID; + 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); @@ -236,20 +243,24 @@ static int write_new_vid(struct powernow_k8_data *data, u32 vid) 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"); + 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", + 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); + printk(KERN_ERR PFX "vid trans failed, vid 0x%x, " + "curr 0x%x\n", + vid, data->currvid); return 1; } @@ -261,7 +272,8 @@ static int write_new_vid(struct powernow_k8_data *data, u32 vid) * 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) +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; @@ -283,7 +295,8 @@ static int transition_pstate(struct powernow_k8_data *data, u32 pstate) } /* Change Opteron/Athlon64 fid and vid, by the 3 phases. */ -static int transition_fid_vid(struct powernow_k8_data *data, u32 reqfid, u32 reqvid) +static int transition_fid_vid(struct powernow_k8_data *data, + u32 reqfid, u32 reqvid) { if (core_voltage_pre_transition(data, reqvid)) return 1; @@ -298,7 +311,8 @@ static int transition_fid_vid(struct powernow_k8_data *data, u32 reqfid, u32 req 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", + 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; @@ -311,13 +325,15 @@ static int transition_fid_vid(struct powernow_k8_data *data, u32 reqfid, u32 req } /* Phase 1 - core voltage transition ... setup voltage */ -static int core_voltage_pre_transition(struct powernow_k8_data *data, u32 reqvid) +static int core_voltage_pre_transition(struct powernow_k8_data *data, + u32 reqvid) { u32 rvosteps = data->rvo; u32 savefid = data->currfid; u32 maxvid, lo; - dprintk("ph1 (cpu%d): start, currfid 0x%x, currvid 0x%x, reqvid 0x%x, rvo 0x%x\n", + 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); @@ -340,7 +356,7 @@ static int core_voltage_pre_transition(struct powernow_k8_data *data, u32 reqvid } 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)) + if (decrease_vid_code_by_step(data, data->currvid-1, 1)) return 1; rvosteps--; } @@ -350,7 +366,8 @@ static int core_voltage_pre_transition(struct powernow_k8_data *data, u32 reqvid return 1; if (savefid != data->currfid) { - printk(KERN_ERR PFX "ph1 err, currfid changed 0x%x\n", data->currfid); + printk(KERN_ERR PFX "ph1 err, currfid changed 0x%x\n", + data->currfid); return 1; } @@ -363,20 +380,24 @@ static int core_voltage_pre_transition(struct powernow_k8_data *data, u32 reqvid /* Phase 2 - core frequency transition */ static int core_frequency_transition(struct powernow_k8_data *data, u32 reqfid) { - u32 vcoreqfid, vcocurrfid, vcofiddiff, fid_interval, savevid = data->currvid; + u32 vcoreqfid, vcocurrfid, vcofiddiff; + u32 fid_interval, savevid = data->currvid; - if ((reqfid < HI_FID_TABLE_BOTTOM) && (data->currfid < HI_FID_TABLE_BOTTOM)) { - printk(KERN_ERR PFX "ph2: illegal lo-lo transition 0x%x 0x%x\n", - reqfid, data->currfid); + if ((reqfid < HI_FID_TABLE_BOTTOM) && + (data->currfid < HI_FID_TABLE_BOTTOM)) { + printk(KERN_ERR PFX "ph2: illegal lo-lo transition " + "0x%x 0x%x\n", reqfid, data->currfid); return 1; } if (data->currfid == reqfid) { - printk(KERN_ERR PFX "ph2 null fid transition 0x%x\n", data->currfid); + 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", + dprintk("ph2 (cpu%d): starting, currfid 0x%x, currvid 0x%x, " + "reqfid 0x%x\n", smp_processor_id(), data->currfid, data->currvid, reqfid); @@ -390,14 +411,14 @@ static int core_frequency_transition(struct powernow_k8_data *data, u32 reqfid) if (reqfid > data->currfid) { if (data->currfid > LO_FID_TABLE_TOP) { - if (write_new_fid(data, data->currfid + fid_interval)) { + 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))) { + (data, + 2 + convert_fid_to_vco_fid(data->currfid))) return 1; - } } } else { if (write_new_fid(data, data->currfid - fid_interval)) @@ -417,7 +438,8 @@ static int core_frequency_transition(struct powernow_k8_data *data, u32 reqfid) if (data->currfid != reqfid) { printk(KERN_ERR PFX - "ph2: mismatch, failed fid transition, curr 0x%x, req 0x%x\n", + "ph2: mismatch, failed fid transition, " + "curr 0x%x, req 0x%x\n", data->currfid, reqfid); return 1; } @@ -435,7 +457,8 @@ static int core_frequency_transition(struct powernow_k8_data *data, u32 reqfid) } /* Phase 3 - core voltage transition flow ... jump to the final vid. */ -static int core_voltage_post_transition(struct powernow_k8_data *data, u32 reqvid) +static int core_voltage_post_transition(struct powernow_k8_data *data, + u32 reqvid) { u32 savefid = data->currfid; u32 savereqvid = reqvid; @@ -457,7 +480,8 @@ static int core_voltage_post_transition(struct powernow_k8_data *data, u32 reqvi if (data->currvid != reqvid) { printk(KERN_ERR PFX - "ph3: failed vid transition\n, req 0x%x, curr 0x%x", + "ph3: failed vid transition\n, " + "req 0x%x, curr 0x%x", reqvid, data->currvid); return 1; } @@ -508,7 +532,8 @@ static int check_supported_cpu(unsigned int cpu) 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); + printk(KERN_INFO PFX + "Processor cpuid %x not supported\n", eax); goto out; } @@ -520,8 +545,10 @@ static int check_supported_cpu(unsigned int cpu) } 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"); + if ((edx & P_STATE_TRANSITION_CAPABLE) + != P_STATE_TRANSITION_CAPABLE) { + printk(KERN_INFO PFX + "Power state transitions not supported\n"); goto out; } } else { /* must be a HW Pstate capable processor */ @@ -539,7 +566,8 @@ out: return rc; } -static int check_pst_table(struct powernow_k8_data *data, struct pst_s *pst, u8 maxvid) +static int check_pst_table(struct powernow_k8_data *data, struct pst_s *pst, + u8 maxvid) { unsigned int j; u8 lastfid = 0xff; @@ -550,12 +578,14 @@ static int check_pst_table(struct powernow_k8_data *data, struct pst_s *pst, u8 j, pst[j].vid); return -EINVAL; } - if (pst[j].vid < data->rvo) { /* vid + rvo >= 0 */ + 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 */ + 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; @@ -579,23 +609,31 @@ static int check_pst_table(struct powernow_k8_data *data, struct pst_s *pst, u8 return -EINVAL; } if (lastfid > LO_FID_TABLE_TOP) - printk(KERN_INFO FW_BUG PFX "first fid not from lo freq table\n"); + printk(KERN_INFO FW_BUG PFX + "first fid not from lo freq table\n"); return 0; } +static void invalidate_entry(struct powernow_k8_data *data, unsigned int entry) +{ + data->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 (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, + 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 " %d : fid 0x%x (%d MHz), vid 0x%x\n", + printk(KERN_INFO PFX + " %d : fid 0x%x (%d MHz), vid 0x%x\n", j, data->powernow_table[j].index & 0xff, data->powernow_table[j].frequency/1000, @@ -604,20 +642,25 @@ static void print_basics(struct powernow_k8_data *data) } } if (data->batps) - printk(KERN_INFO PFX "Only %d pstates on battery\n", data->batps); + printk(KERN_INFO PFX "Only %d pstates on battery\n", + data->batps); } -static int fill_powernow_table(struct powernow_k8_data *data, struct pst_s *pst, u8 maxvid) +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); + 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++ ) { + 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; @@ -640,9 +683,11 @@ static int fill_powernow_table(struct powernow_k8_data *data, struct pst_s *pst, } 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 */ - powernow_table[j].frequency = find_khz_freq_from_fid(pst[j].fid); + 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; @@ -658,7 +703,8 @@ static int fill_powernow_table(struct powernow_k8_data *data, struct pst_s *pst, print_basics(data); for (j = 0; j < data->numps; j++) - if ((pst[j].fid==data->currfid) && (pst[j].vid==data->currvid)) + if ((pst[j].fid == data->currfid) && + (pst[j].vid == data->currvid)) return 0; dprintk("currfid/vid do not match PST, ignoring\n"); @@ -698,7 +744,8 @@ static int find_psb_table(struct powernow_k8_data *data) } data->vstable = psb->vstable; - dprintk("voltage stabilization time: %d(*20us)\n", data->vstable); + dprintk("voltage stabilization time: %d(*20us)\n", + data->vstable); dprintk("flags2: 0x%x\n", psb->flags2); data->rvo = psb->flags2 & 3; @@ -713,11 +760,12 @@ static int find_psb_table(struct powernow_k8_data *data) dprintk("numpst: 0x%x\n", psb->num_tables); cpst = psb->num_tables; - if ((psb->cpuid == 0x00000fc0) || (psb->cpuid == 0x00000fe0) ){ + if ((psb->cpuid == 0x00000fc0) || + (psb->cpuid == 0x00000fe0)) { thiscpuid = cpuid_eax(CPUID_PROCESSOR_SIGNATURE); - if ((thiscpuid == 0x00000fc0) || (thiscpuid == 0x00000fe0) ) { + if ((thiscpuid == 0x00000fc0) || + (thiscpuid == 0x00000fe0)) cpst = 1; - } } if (cpst != 1) { printk(KERN_ERR FW_BUG PFX "numpst must be 1\n"); @@ -732,7 +780,8 @@ static int find_psb_table(struct powernow_k8_data *data) data->numps = psb->numps; dprintk("numpstates: 0x%x\n", data->numps); - return fill_powernow_table(data, (struct pst_s *)(psb+1), maxvid); + return fill_powernow_table(data, + (struct pst_s *)(psb+1), maxvid); } /* * If you see this message, complain to BIOS manufacturer. If @@ -745,28 +794,31 @@ static int find_psb_table(struct powernow_k8_data *data) * BIOS and Kernel Developer's Guide, which is available on * www.amd.com */ - printk(KERN_ERR PFX "BIOS error - no PSB or ACPI _PSS objects\n"); + printk(KERN_ERR FW_BUG PFX "No PSB or ACPI _PSS objects\n"); return -ENODEV; } -#ifdef CONFIG_X86_POWERNOW_K8_ACPI -static void powernow_k8_acpi_pst_values(struct powernow_k8_data *data, unsigned int index) +static void powernow_k8_acpi_pst_values(struct powernow_k8_data *data, + unsigned int index) { + acpi_integer control; + if (!data->acpi_data.state_count || (cpu_family == CPU_HW_PSTATE)) return; - data->irt = (data->acpi_data.states[index].control >> IRT_SHIFT) & IRT_MASK; - data->rvo = (data->acpi_data.states[index].control >> RVO_SHIFT) & RVO_MASK; - data->exttype = (data->acpi_data.states[index].control >> EXT_TYPE_SHIFT) & EXT_TYPE_MASK; - data->plllock = (data->acpi_data.states[index].control >> PLL_L_SHIFT) & PLL_L_MASK; - data->vidmvs = 1 << ((data->acpi_data.states[index].control >> MVS_SHIFT) & MVS_MASK); - data->vstable = (data->acpi_data.states[index].control >> VST_SHIFT) & VST_MASK; -} + control = data->acpi_data.states[index].control; data->irt = (control + >> IRT_SHIFT) & IRT_MASK; data->rvo = (control >> + RVO_SHIFT) & RVO_MASK; data->exttype = (control + >> EXT_TYPE_SHIFT) & EXT_TYPE_MASK; + data->plllock = (control >> PLL_L_SHIFT) & PLL_L_MASK; data->vidmvs = 1 + << ((control >> MVS_SHIFT) & MVS_MASK); data->vstable = + (control >> VST_SHIFT) & VST_MASK; } static int powernow_k8_cpu_init_acpi(struct powernow_k8_data *data) { struct cpufreq_frequency_table *powernow_table; int ret_val = -ENODEV; + acpi_integer space_id; if (acpi_processor_register_performance(&data->acpi_data, data->cpu)) { dprintk("register performance failed: bad ACPI data\n"); @@ -779,11 +831,12 @@ static int powernow_k8_cpu_init_acpi(struct powernow_k8_data *data) goto err_out; } - if ((data->acpi_data.control_register.space_id != ACPI_ADR_SPACE_FIXED_HARDWARE) || - (data->acpi_data.status_register.space_id != ACPI_ADR_SPACE_FIXED_HARDWARE)) { + space_id = data->acpi_data.control_register.space_id; + if ((space_id != ACPI_ADR_SPACE_FIXED_HARDWARE) || + (space_id != ACPI_ADR_SPACE_FIXED_HARDWARE)) { dprintk("Invalid control/status registers (%x - %x)\n", data->acpi_data.control_register.space_id, - data->acpi_data.status_register.space_id); + space_id); goto err_out; } @@ -802,7 +855,8 @@ static int powernow_k8_cpu_init_acpi(struct powernow_k8_data *data) 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].frequency = + CPUFREQ_TABLE_END; powernow_table[data->acpi_data.state_count].index = 0; data->powernow_table = powernow_table; @@ -830,13 +884,15 @@ err_out_mem: 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 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) +static int fill_powernow_table_pstate(struct powernow_k8_data *data, + struct cpufreq_frequency_table *powernow_table) { int i; u32 hi = 0, lo = 0; @@ -848,84 +904,101 @@ static int fill_powernow_table_pstate(struct powernow_k8_data *data, struct cpuf 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"); - powernow_table[i].frequency = CPUFREQ_ENTRY_INVALID; + 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(data, i); continue; } rdmsr(MSR_PSTATE_DEF_BASE + index, lo, hi); if (!(hi & HW_PSTATE_VALID_MASK)) { dprintk("invalid pstate %d, ignoring\n", index); - powernow_table[i].frequency = CPUFREQ_ENTRY_INVALID; + invalidate_entry(data, i); continue; } powernow_table[i].index = index; - powernow_table[i].frequency = data->acpi_data.states[i].core_frequency * 1000; + 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) +static int fill_powernow_table_fidvid(struct powernow_k8_data *data, + struct cpufreq_frequency_table *powernow_table) { int i; int cntlofreq = 0; + for (i = 0; i < data->acpi_data.state_count; i++) { u32 fid; u32 vid; + u32 freq, index; + acpi_integer status, control; if (data->exttype) { - fid = data->acpi_data.states[i].status & EXT_FID_MASK; - vid = (data->acpi_data.states[i].status >> VID_SHIFT) & EXT_VID_MASK; + status = data->acpi_data.states[i].status; + fid = status & EXT_FID_MASK; + vid = (status >> VID_SHIFT) & EXT_VID_MASK; } else { - fid = data->acpi_data.states[i].control & FID_MASK; - vid = (data->acpi_data.states[i].control >> VID_SHIFT) & VID_MASK; + 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); - powernow_table[i].index = fid; /* lower 8 bits */ - powernow_table[i].index |= (vid << 8); /* upper 8 bits */ - powernow_table[i].frequency = find_khz_freq_from_fid(fid); + 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 ((powernow_table[i].frequency > (MAX_FREQ * 1000)) || - (powernow_table[i].frequency < (MIN_FREQ * 1000))) { - dprintk("invalid freq %u kHz, ignoring\n", powernow_table[i].frequency); - powernow_table[i].frequency = CPUFREQ_ENTRY_INVALID; + if ((freq > (MAX_FREQ * 1000)) || (freq < (MIN_FREQ * 1000))) { + dprintk("invalid freq %u kHz, ignoring\n", freq); + invalidate_entry(data, i); continue; } - /* verify voltage is OK - BIOSs are using "off" to indicate invalid */ + /* verify voltage is OK - + * BIOSs are using "off" to indicate invalid */ if (vid == VID_OFF) { dprintk("invalid vid %u, ignoring\n", vid); - powernow_table[i].frequency = CPUFREQ_ENTRY_INVALID; + invalidate_entry(data, i); continue; } /* verify only 1 entry from the lo frequency table */ if (fid < HI_FID_TABLE_BOTTOM) { if (cntlofreq) { - /* if both entries are the same, ignore this one ... */ - if ((powernow_table[i].frequency != powernow_table[cntlofreq].frequency) || - (powernow_table[i].index != powernow_table[cntlofreq].index)) { - printk(KERN_ERR PFX "Too many lo freq table entries\n"); + /* if both entries are the same, + * ignore this one ... */ + if ((freq != powernow_table[cntlofreq].frequency) || + (index != powernow_table[cntlofreq].index)) { + printk(KERN_ERR PFX + "Too many lo freq table " + "entries\n"); return 1; } - dprintk("double low frequency table entry, ignoring it.\n"); - powernow_table[i].frequency = CPUFREQ_ENTRY_INVALID; + dprintk("double low frequency table entry, " + "ignoring it.\n"); + invalidate_entry(data, i); continue; } else cntlofreq = i; } - if (powernow_table[i].frequency != (data->acpi_data.states[i].core_frequency * 1000)) { - printk(KERN_INFO PFX "invalid freq entries %u kHz vs. %u kHz\n", - powernow_table[i].frequency, - (unsigned int) (data->acpi_data.states[i].core_frequency * 1000)); - powernow_table[i].frequency = CPUFREQ_ENTRY_INVALID; + 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(data, i); continue; } } @@ -935,18 +1008,28 @@ static int fill_powernow_table_fidvid(struct powernow_k8_data *data, struct cpuf 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); + acpi_processor_unregister_performance(&data->acpi_data, + data->cpu); free_cpumask_var(data->acpi_data.shared_cpu_map); } -#else -static int powernow_k8_cpu_init_acpi(struct powernow_k8_data *data) { return -ENODEV; } -static void powernow_k8_cpu_exit_acpi(struct powernow_k8_data *data) { return; } -static void powernow_k8_acpi_pst_values(struct powernow_k8_data *data, unsigned int index) { return; } -#endif /* CONFIG_X86_POWERNOW_K8_ACPI */ +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; + } + /* 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) +static int transition_frequency_fidvid(struct powernow_k8_data *data, + unsigned int index) { u32 fid = 0; u32 vid = 0; @@ -974,7 +1057,8 @@ static int transition_frequency_fidvid(struct powernow_k8_data *data, unsigned i return 0; } - if ((fid < HI_FID_TABLE_BOTTOM) && (data->currfid < HI_FID_TABLE_BOTTOM)) { + if ((fid < HI_FID_TABLE_BOTTOM) && + (data->currfid < HI_FID_TABLE_BOTTOM)) { printk(KERN_ERR PFX "ignoring illegal change in lo freq table-%x to 0x%x\n", data->currfid, fid); @@ -1002,7 +1086,8 @@ static int transition_frequency_fidvid(struct powernow_k8_data *data, unsigned i } /* Take a frequency, and issue the hardware pstate transition command */ -static int transition_frequency_pstate(struct powernow_k8_data *data, unsigned int index) +static int transition_frequency_pstate(struct powernow_k8_data *data, + unsigned int index) { u32 pstate = 0; int res, i; @@ -1014,7 +1099,8 @@ static int transition_frequency_pstate(struct powernow_k8_data *data, unsigned i 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.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_mask_nr(i, *(data->available_cores)) { @@ -1033,7 +1119,8 @@ static int transition_frequency_pstate(struct powernow_k8_data *data, unsigned i } /* Driver entry point to switch to the target frequency */ -static int powernowk8_target(struct cpufreq_policy *pol, unsigned targfreq, unsigned relation) +static int powernowk8_target(struct cpufreq_policy *pol, + unsigned targfreq, unsigned relation) { cpumask_t oldmask; struct powernow_k8_data *data = per_cpu(powernow_data, pol->cpu); @@ -1072,14 +1159,18 @@ static int powernowk8_target(struct cpufreq_policy *pol, unsigned targfreq, unsi dprintk("targ: curr fid 0x%x, vid 0x%x\n", data->currfid, data->currvid); - if ((checkvid != data->currvid) || (checkfid != data->currfid)) { + 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); + "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)) + if (cpufreq_frequency_table_target(pol, data->powernow_table, + targfreq, relation, &newstate)) goto err_out; mutex_lock(&fidvid_mutex); @@ -1099,7 +1190,8 @@ static int powernowk8_target(struct cpufreq_policy *pol, unsigned targfreq, unsi mutex_unlock(&fidvid_mutex); if (cpu_family == CPU_HW_PSTATE) - pol->cur = find_khz_freq_from_pstate(data->powernow_table, newstate); + pol->cur = find_khz_freq_from_pstate(data->powernow_table, + newstate); else pol->cur = find_khz_freq_from_fid(data->currfid); ret = 0; @@ -1126,6 +1218,7 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol) struct powernow_k8_data *data; cpumask_t oldmask; int rc; + static int print_once; if (!cpu_online(pol->cpu)) return -ENODEV; @@ -1142,25 +1235,25 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol) data->cpu = pol->cpu; data->currpstate = HW_PSTATE_INVALID; - rc = powernow_k8_cpu_init_acpi(data); - if (rc) { + if (powernow_k8_cpu_init_acpi(data)) { /* * Use the PSB BIOS structure. This is only availabe on * an UP version, and is deprecated by AMD. */ if (num_online_cpus() != 1) { -#ifndef CONFIG_ACPI_PROCESSOR - printk(KERN_ERR PFX "ACPI Processor support is required " - "for SMP systems but is absent. Please load the " - "ACPI Processor module before starting this " - "driver.\n"); -#else - printk(KERN_ERR FW_BUG PFX "Your BIOS does not provide" - " ACPI _PSS objects in a way that Linux " - "understands. Please report this to the Linux " - "ACPI maintainers and complain to your BIOS " - "vendor.\n"); -#endif + /* + * Replace this one with print_once as soon as such a + * thing gets introduced + */ + if (!print_once) { + WARN_ONCE(1, KERN_ERR FW_BUG PFX "Your BIOS " + "does not provide ACPI _PSS objects " + "in a way that Linux understands. " + "Please report this to the Linux ACPI" + " maintainers and complain to your " + "BIOS vendor.\n"); + print_once++; + } goto err_out; } if (pol->cpu != 0) { @@ -1170,10 +1263,16 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol) goto err_out; } rc = find_psb_table(data); - if (rc) { + 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 */ oldmask = current->cpus_allowed; @@ -1181,16 +1280,16 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol) if (smp_processor_id() != pol->cpu) { printk(KERN_ERR PFX "limiting to cpu %u failed\n", pol->cpu); - goto err_out; + goto err_out_unmask; } if (pending_bit_stuck()) { printk(KERN_ERR PFX "failing init, change pending bit set\n"); - goto err_out; + goto err_out_unmask; } if (query_current_values_with_pending_wait(data)) - goto err_out; + goto err_out_unmask; if (cpu_family == CPU_OPTERON) fidvid_msr_init(); @@ -1204,13 +1303,9 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol) cpumask_copy(pol->cpus, &per_cpu(cpu_core_map, pol->cpu)); data->available_cores = pol->cpus; - /* 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) - + (3 * (1 << data->irt) * 10)) * 1000; - if (cpu_family == CPU_HW_PSTATE) - pol->cur = find_khz_freq_from_pstate(data->powernow_table, data->currpstate); + 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); @@ -1227,7 +1322,8 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol) 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); + 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); @@ -1236,15 +1332,16 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol) return 0; -err_out: +err_out_unmask: set_cpus_allowed_ptr(current, &oldmask); powernow_k8_cpu_exit_acpi(data); +err_out: kfree(data); return -ENODEV; } -static int __devexit powernowk8_cpu_exit (struct cpufreq_policy *pol) +static int __devexit powernowk8_cpu_exit(struct cpufreq_policy *pol) { struct powernow_k8_data *data = per_cpu(powernow_data, pol->cpu); @@ -1261,7 +1358,7 @@ static int __devexit powernowk8_cpu_exit (struct cpufreq_policy *pol) return 0; } -static unsigned int powernowk8_get (unsigned int cpu) +static unsigned int powernowk8_get(unsigned int cpu) { struct powernow_k8_data *data; cpumask_t oldmask = current->cpus_allowed; @@ -1297,7 +1394,7 @@ out: return khz; } -static struct freq_attr* powernow_k8_attr[] = { +static struct freq_attr *powernow_k8_attr[] = { &cpufreq_freq_attr_scaling_available_freqs, NULL, }; @@ -1342,7 +1439,8 @@ static void __exit powernowk8_exit(void) cpufreq_unregister_driver(&cpufreq_amd64_driver); } -MODULE_AUTHOR("Paul Devriendt <paul.devriendt@amd.com> and Mark Langsdorf <mark.langsdorf@amd.com>"); +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"); diff --git a/arch/x86/kernel/cpu/cpufreq/powernow-k8.h b/arch/x86/kernel/cpu/cpufreq/powernow-k8.h index 8ecc75b6c7c..6c6698feade 100644 --- a/arch/x86/kernel/cpu/cpufreq/powernow-k8.h +++ b/arch/x86/kernel/cpu/cpufreq/powernow-k8.h @@ -45,11 +45,10 @@ struct powernow_k8_data { * frequency is in kHz */ struct cpufreq_frequency_table *powernow_table; -#ifdef CONFIG_X86_POWERNOW_K8_ACPI /* 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; -#endif + /* we need to keep track of associated cores, but let cpufreq * handle hotplug events - so just point at cpufreq pol->cpus * structure */ @@ -222,10 +221,8 @@ 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); -#ifdef CONFIG_X86_POWERNOW_K8_ACPI 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); -#endif #ifdef CONFIG_SMP static inline void define_siblings(int cpu, cpumask_t cpu_sharedcore_mask[]) diff --git a/arch/x86/kernel/cpu/cpufreq/sc520_freq.c b/arch/x86/kernel/cpu/cpufreq/sc520_freq.c index 42da9bd677d..435a996a613 100644 --- a/arch/x86/kernel/cpu/cpufreq/sc520_freq.c +++ b/arch/x86/kernel/cpu/cpufreq/sc520_freq.c @@ -19,17 +19,19 @@ #include <linux/delay.h> #include <linux/cpufreq.h> +#include <linux/timex.h> +#include <linux/io.h> #include <asm/msr.h> -#include <asm/timex.h> -#include <asm/io.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 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}, @@ -43,7 +45,8 @@ static unsigned int sc520_freq_get_cpu_frequency(unsigned int cpu) switch (clockspeed_reg & 0x03) { default: - printk(KERN_ERR "sc520_freq: error: cpuctl register has unexpected value %02x\n", clockspeed_reg); + printk(KERN_ERR PFX "error: cpuctl register has unexpected " + "value %02x\n", clockspeed_reg); case 0x01: return 100000; case 0x02: @@ -51,7 +54,7 @@ static unsigned int sc520_freq_get_cpu_frequency(unsigned int cpu) } } -static void sc520_freq_set_cpu_state (unsigned int state) +static void sc520_freq_set_cpu_state(unsigned int state) { struct cpufreq_freqs freqs; @@ -76,18 +79,19 @@ static void sc520_freq_set_cpu_state (unsigned int state) cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); }; -static int sc520_freq_verify (struct cpufreq_policy *policy) +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, +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)) + if (cpufreq_frequency_table_target(policy, sc520_freq_table, + target_freq, relation, &newstate)) return -EINVAL; sc520_freq_set_cpu_state(newstate); @@ -116,7 +120,7 @@ static int sc520_freq_cpu_init(struct cpufreq_policy *policy) result = cpufreq_frequency_table_cpuinfo(policy, sc520_freq_table); if (result) - return (result); + return result; cpufreq_frequency_table_get_attr(sc520_freq_table, policy->cpu); @@ -131,7 +135,7 @@ static int sc520_freq_cpu_exit(struct cpufreq_policy *policy) } -static struct freq_attr* sc520_freq_attr[] = { +static struct freq_attr *sc520_freq_attr[] = { &cpufreq_freq_attr_scaling_available_freqs, NULL, }; @@ -155,13 +159,13 @@ static int __init sc520_freq_init(void) int err; /* Test if we have the right hardware */ - if(c->x86_vendor != X86_VENDOR_AMD || - c->x86 != 4 || c->x86_model != 9) { + 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) { + if (!cpuctl) { printk(KERN_ERR "sc520_freq: error: failed to remap memory\n"); return -ENOMEM; } diff --git a/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c b/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c index dedc1e98f16..8bbb11adb31 100644 --- a/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c +++ b/arch/x86/kernel/cpu/cpufreq/speedstep-ich.c @@ -39,7 +39,7 @@ static struct pci_dev *speedstep_chipset_dev; /* speedstep_processor */ -static unsigned int speedstep_processor = 0; +static unsigned int speedstep_processor; static u32 pmbase; @@ -54,7 +54,8 @@ static struct cpufreq_frequency_table speedstep_freqs[] = { }; -#define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, "speedstep-ich", msg) +#define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, \ + "speedstep-ich", msg) /** @@ -62,7 +63,7 @@ static struct cpufreq_frequency_table speedstep_freqs[] = { * * Returns: -ENODEV if no register could be found */ -static int speedstep_find_register (void) +static int speedstep_find_register(void) { if (!speedstep_chipset_dev) return -ENODEV; @@ -90,7 +91,7 @@ static int speedstep_find_register (void) * * Tries to change the SpeedStep state. */ -static void speedstep_set_state (unsigned int state) +static void speedstep_set_state(unsigned int state) { u8 pm2_blk; u8 value; @@ -133,11 +134,11 @@ static void speedstep_set_state (unsigned int state) 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_processor_frequency(speedstep_processor) / 1000)); - } else { - printk (KERN_ERR "cpufreq: change failed - I/O error\n"); - } + 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; } @@ -149,7 +150,7 @@ static void speedstep_set_state (unsigned int state) * Tries to activate the SpeedStep status and control registers. * Returns -EINVAL on an unsupported chipset, and zero on success. */ -static int speedstep_activate (void) +static int speedstep_activate(void) { u16 value = 0; @@ -175,20 +176,18 @@ static int speedstep_activate (void) * functions. Returns the SPEEDSTEP_CHIPSET_-number for the detected * chipset, or zero on failure. */ -static unsigned int speedstep_detect_chipset (void) +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, + 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, + PCI_ANY_ID, PCI_ANY_ID, NULL); if (speedstep_chipset_dev) return 3; /* 3-M */ @@ -196,8 +195,7 @@ static unsigned int speedstep_detect_chipset (void) speedstep_chipset_dev = pci_get_subsys(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82801BA_10, - PCI_ANY_ID, - PCI_ANY_ID, + PCI_ANY_ID, PCI_ANY_ID, NULL); if (speedstep_chipset_dev) { /* speedstep.c causes lockups on Dell Inspirons 8000 and @@ -208,8 +206,7 @@ static unsigned int speedstep_detect_chipset (void) hostbridge = pci_get_subsys(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82815_MC, - PCI_ANY_ID, - PCI_ANY_ID, + PCI_ANY_ID, PCI_ANY_ID, NULL); if (!hostbridge) @@ -236,7 +233,7 @@ static unsigned int _speedstep_get(const struct cpumask *cpus) cpus_allowed = current->cpus_allowed; set_cpus_allowed_ptr(current, cpus); - speed = speedstep_get_processor_frequency(speedstep_processor); + speed = speedstep_get_frequency(speedstep_processor); set_cpus_allowed_ptr(current, &cpus_allowed); dprintk("detected %u kHz as current frequency\n", speed); return speed; @@ -251,11 +248,12 @@ static unsigned int speedstep_get(unsigned int cpu) * 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) + * @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, +static int speedstep_target(struct cpufreq_policy *policy, unsigned int target_freq, unsigned int relation) { @@ -264,7 +262,8 @@ static int speedstep_target (struct cpufreq_policy *policy, cpumask_t cpus_allowed; int i; - if (cpufreq_frequency_table_target(policy, &speedstep_freqs[0], target_freq, relation, &newstate)) + if (cpufreq_frequency_table_target(policy, &speedstep_freqs[0], + target_freq, relation, &newstate)) return -EINVAL; freqs.old = _speedstep_get(policy->cpus); @@ -308,7 +307,7 @@ static int speedstep_target (struct cpufreq_policy *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) +static int speedstep_verify(struct cpufreq_policy *policy) { return cpufreq_frequency_table_verify(policy, &speedstep_freqs[0]); } @@ -344,7 +343,8 @@ static int speedstep_cpu_init(struct cpufreq_policy *policy) return -EIO; dprintk("currently at %s speed setting - %i MHz\n", - (speed == speedstep_freqs[SPEEDSTEP_LOW].frequency) ? "low" : "high", + (speed == speedstep_freqs[SPEEDSTEP_LOW].frequency) + ? "low" : "high", (speed / 1000)); /* cpuinfo and default policy values */ @@ -352,9 +352,9 @@ static int speedstep_cpu_init(struct cpufreq_policy *policy) result = cpufreq_frequency_table_cpuinfo(policy, speedstep_freqs); if (result) - return (result); + return result; - cpufreq_frequency_table_get_attr(speedstep_freqs, policy->cpu); + cpufreq_frequency_table_get_attr(speedstep_freqs, policy->cpu); return 0; } @@ -366,7 +366,7 @@ static int speedstep_cpu_exit(struct cpufreq_policy *policy) return 0; } -static struct freq_attr* speedstep_attr[] = { +static struct freq_attr *speedstep_attr[] = { &cpufreq_freq_attr_scaling_available_freqs, NULL, }; @@ -396,13 +396,15 @@ 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"); + 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"); + dprintk("Intel(R) SpeedStep(TM) for this chipset not " + "(yet) available.\n"); return -ENODEV; } @@ -431,9 +433,11 @@ static void __exit speedstep_exit(void) } -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_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 index cdac7d62369..2e3c6862657 100644 --- a/arch/x86/kernel/cpu/cpufreq/speedstep-lib.c +++ b/arch/x86/kernel/cpu/cpufreq/speedstep-lib.c @@ -16,12 +16,16 @@ #include <linux/slab.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 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 = 0; +static int relaxed_check; #else #define relaxed_check 0 #endif @@ -30,14 +34,14 @@ static int relaxed_check = 0; * GET PROCESSOR CORE SPEED IN KHZ * *********************************************************************/ -static unsigned int pentium3_get_frequency (unsigned int processor) +static unsigned int pentium3_get_frequency(unsigned int processor) { - /* See table 14 of p3_ds.pdf and table 22 of 29834003.pdf */ + /* 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 [] = { + } msr_decode_mult[] = { { 30, 0x01 }, { 35, 0x05 }, { 40, 0x02 }, @@ -52,7 +56,7 @@ static unsigned int pentium3_get_frequency (unsigned int processor) { 85, 0x26 }, { 90, 0x20 }, { 100, 0x2b }, - { 0, 0xff } /* error or unknown value */ + { 0, 0xff } /* error or unknown value */ }; /* PIII(-M) FSB settings: see table b1-b of 24547206.pdf */ @@ -60,7 +64,7 @@ static unsigned int pentium3_get_frequency (unsigned int processor) unsigned int value; /* Front Side Bus speed in MHz */ u8 bitmap; /* power on configuration bits [18: 19] (in MSR 0x2a) */ - } msr_decode_fsb [] = { + } msr_decode_fsb[] = { { 66, 0x0 }, { 100, 0x2 }, { 133, 0x1 }, @@ -85,7 +89,7 @@ static unsigned int pentium3_get_frequency (unsigned int processor) } /* decode the multiplier */ - if (processor == SPEEDSTEP_PROCESSOR_PIII_C_EARLY) { + if (processor == SPEEDSTEP_CPU_PIII_C_EARLY) { dprintk("workaround for early PIIIs\n"); msr_lo &= 0x03c00000; } else @@ -97,9 +101,10 @@ static unsigned int pentium3_get_frequency (unsigned int processor) j++; } - dprintk("speed is %u\n", (msr_decode_mult[j].ratio * msr_decode_fsb[i].value * 100)); + 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); + return msr_decode_mult[j].ratio * msr_decode_fsb[i].value * 100; } @@ -112,20 +117,23 @@ static unsigned int pentiumM_get_frequency(void) /* see table B-2 of 24547212.pdf */ if (msr_lo & 0x00040000) { - printk(KERN_DEBUG "speedstep-lib: PM - invalid FSB: 0x%x 0x%x\n", msr_lo, msr_tmp); + 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)); + dprintk("bits 22-26 are 0x%x, speed is %u\n", + msr_tmp, (msr_tmp * 100 * 1000)); - return (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 */ @@ -153,12 +161,15 @@ static unsigned int pentium_core_get_frequency(void) } 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); + 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)); + dprintk("bits 22-26 are 0x%x, speed is %u\n", + msr_tmp, (msr_tmp * fsb)); - return (msr_tmp * fsb); + ret = (msr_tmp * fsb); + return ret; } @@ -167,6 +178,16 @@ 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); @@ -177,62 +198,61 @@ static unsigned int pentium4_get_frequency(void) * revision #12 in Table B-1: MSRs in the Pentium 4 and * Intel Xeon Processors, on page B-4 and B-5. */ - if (c->x86_model < 2) + fsb_code = (msr_lo >> 16) & 0x7; + switch (fsb_code) { + case 0: fsb = 100 * 1000; - else { - u8 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; - } + break; + case 1: + fsb = 13333 * 10; + break; + case 2: + fsb = 200 * 1000; + break; } if (!fsb) - printk(KERN_DEBUG "speedstep-lib: couldn't detect FSB speed. Please send an e-mail to <linux@brodo.de>\n"); + 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)); + dprintk("P4 - FSB %u kHz; Multiplier %u; Speed %u kHz\n", + fsb, mult, (fsb * mult)); - return (fsb * mult); + ret = (fsb * mult); + return ret; } -unsigned int speedstep_get_processor_frequency(unsigned int processor) +unsigned int speedstep_get_frequency(unsigned int processor) { switch (processor) { - case SPEEDSTEP_PROCESSOR_PCORE: + case SPEEDSTEP_CPU_PCORE: return pentium_core_get_frequency(); - case SPEEDSTEP_PROCESSOR_PM: + case SPEEDSTEP_CPU_PM: return pentiumM_get_frequency(); - case SPEEDSTEP_PROCESSOR_P4D: - case SPEEDSTEP_PROCESSOR_P4M: + case SPEEDSTEP_CPU_P4D: + case SPEEDSTEP_CPU_P4M: return pentium4_get_frequency(); - case SPEEDSTEP_PROCESSOR_PIII_T: - case SPEEDSTEP_PROCESSOR_PIII_C: - case SPEEDSTEP_PROCESSOR_PIII_C_EARLY: + 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_processor_frequency); +EXPORT_SYMBOL_GPL(speedstep_get_frequency); /********************************************************************* * DETECT SPEEDSTEP-CAPABLE PROCESSOR * *********************************************************************/ -unsigned int speedstep_detect_processor (void) +unsigned int speedstep_detect_processor(void) { struct cpuinfo_x86 *c = &cpu_data(0); u32 ebx, msr_lo, msr_hi; @@ -261,7 +281,7 @@ unsigned int speedstep_detect_processor (void) * sample has ebx = 0x0f, production has 0x0e. */ if ((ebx == 0x0e) || (ebx == 0x0f)) - return SPEEDSTEP_PROCESSOR_P4M; + return SPEEDSTEP_CPU_P4M; break; case 7: /* @@ -272,7 +292,7 @@ unsigned int speedstep_detect_processor (void) * samples are only of B-stepping... */ if (ebx == 0x0e) - return SPEEDSTEP_PROCESSOR_P4M; + return SPEEDSTEP_CPU_P4M; break; case 9: /* @@ -288,10 +308,13 @@ unsigned int speedstep_detect_processor (void) * 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 + * 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_PROCESSOR_P4M; + if ((ebx == 0x0e) || + (strstr(c->x86_model_id, + "Mobile Intel(R) Pentium(R) 4") != NULL)) + return SPEEDSTEP_CPU_P4M; break; default: break; @@ -301,7 +324,8 @@ unsigned int speedstep_detect_processor (void) switch (c->x86_model) { case 0x0B: /* Intel PIII [Tualatin] */ - /* cpuid_ebx(1) is 0x04 for desktop PIII, 0x06 for mobile PIII-M */ + /* cpuid_ebx(1) is 0x04 for desktop PIII, + * 0x06 for mobile PIII-M */ ebx = cpuid_ebx(0x00000001); dprintk("ebx is %x\n", ebx); @@ -313,14 +337,15 @@ unsigned int speedstep_detect_processor (void) /* So far all PIII-M processors support SpeedStep. See * Intel's 24540640.pdf of June 2003 */ - return SPEEDSTEP_PROCESSOR_PIII_T; + 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); + 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; @@ -332,13 +357,15 @@ unsigned int speedstep_detect_processor (void) * 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)))) { + 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_PROCESSOR_PIII_C_EARLY; + return SPEEDSTEP_CPU_PIII_C_EARLY; } else - return SPEEDSTEP_PROCESSOR_PIII_C; + return SPEEDSTEP_CPU_PIII_C; } default: @@ -369,7 +396,7 @@ unsigned int speedstep_get_freqs(unsigned int processor, dprintk("trying to determine both speeds\n"); /* get current speed */ - prev_speed = speedstep_get_processor_frequency(processor); + prev_speed = speedstep_get_frequency(processor); if (!prev_speed) return -EIO; @@ -379,7 +406,7 @@ unsigned int speedstep_get_freqs(unsigned int processor, /* switch to low state */ set_state(SPEEDSTEP_LOW); - *low_speed = speedstep_get_processor_frequency(processor); + *low_speed = speedstep_get_frequency(processor); if (!*low_speed) { ret = -EIO; goto out; @@ -398,7 +425,7 @@ unsigned int speedstep_get_freqs(unsigned int processor, if (transition_latency) do_gettimeofday(&tv2); - *high_speed = speedstep_get_processor_frequency(processor); + *high_speed = speedstep_get_frequency(processor); if (!*high_speed) { ret = -EIO; goto out; @@ -426,9 +453,12 @@ unsigned int speedstep_get_freqs(unsigned int processor, /* 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 "speedstep: frequency transition measured seems out of " - "range (%u nSec), falling back to a safe one of %u nSec.\n", + 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; } @@ -436,15 +466,16 @@ unsigned int speedstep_get_freqs(unsigned int processor, out: local_irq_restore(flags); - return (ret); + 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."); +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"); +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 index b11bcc608ca..2b6c04e5a30 100644 --- a/arch/x86/kernel/cpu/cpufreq/speedstep-lib.h +++ b/arch/x86/kernel/cpu/cpufreq/speedstep-lib.h @@ -12,17 +12,17 @@ /* processors */ -#define SPEEDSTEP_PROCESSOR_PIII_C_EARLY 0x00000001 /* Coppermine core */ -#define SPEEDSTEP_PROCESSOR_PIII_C 0x00000002 /* Coppermine core */ -#define SPEEDSTEP_PROCESSOR_PIII_T 0x00000003 /* Tualatin core */ -#define SPEEDSTEP_PROCESSOR_P4M 0x00000004 /* P4-M */ +#define SPEEDSTEP_CPU_PIII_C_EARLY 0x00000001 /* Coppermine core */ +#define SPEEDSTEP_CPU_PIII_C 0x00000002 /* Coppermine core */ +#define SPEEDSTEP_CPU_PIII_T 0x00000003 /* Tualatin core */ +#define 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_processor_frequency() call. */ -#define SPEEDSTEP_PROCESSOR_PM 0xFFFFFF03 /* Pentium M */ -#define SPEEDSTEP_PROCESSOR_P4D 0xFFFFFF04 /* desktop P4 */ -#define SPEEDSTEP_PROCESSOR_PCORE 0xFFFFFF05 /* Core */ + * the speedstep_get_frequency() call. */ +#define SPEEDSTEP_CPU_PM 0xFFFFFF03 /* Pentium M */ +#define SPEEDSTEP_CPU_P4D 0xFFFFFF04 /* desktop P4 */ +#define SPEEDSTEP_CPU_PCORE 0xFFFFFF05 /* Core */ /* speedstep states -- only two of them */ @@ -34,7 +34,7 @@ extern unsigned int speedstep_detect_processor (void); /* detect the current speed (in khz) of the processor */ -extern unsigned int speedstep_get_processor_frequency(unsigned int processor); +extern unsigned int speedstep_get_frequency(unsigned int processor); /* detect the low and high speeds of the processor. The callback diff --git a/arch/x86/kernel/cpu/cpufreq/speedstep-smi.c b/arch/x86/kernel/cpu/cpufreq/speedstep-smi.c index 8a85c93bd62..befea088e4f 100644 --- a/arch/x86/kernel/cpu/cpufreq/speedstep-smi.c +++ b/arch/x86/kernel/cpu/cpufreq/speedstep-smi.c @@ -19,8 +19,8 @@ #include <linux/cpufreq.h> #include <linux/slab.h> #include <linux/delay.h> +#include <linux/io.h> #include <asm/ist.h> -#include <asm/io.h> #include "speedstep-lib.h" @@ -30,12 +30,12 @@ * If user gives it, these are used. * */ -static int smi_port = 0; -static int smi_cmd = 0; -static unsigned int smi_sig = 0; +static int smi_port; +static int smi_cmd; +static unsigned int smi_sig; /* info about the processor */ -static unsigned int speedstep_processor = 0; +static unsigned int speedstep_processor; /* * There are only two frequency states for each processor. Values @@ -56,12 +56,13 @@ static struct cpufreq_frequency_table speedstep_freqs[] = { * of DMA activity going on? */ #define SMI_TRIES 5 -#define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, "speedstep-smi", msg) +#define dprintk(msg...) cpufreq_debug_printk(CPUFREQ_DEBUG_DRIVER, \ + "speedstep-smi", msg) /** * speedstep_smi_ownership */ -static int speedstep_smi_ownership (void) +static int speedstep_smi_ownership(void) { u32 command, result, magic, dummy; u32 function = GET_SPEEDSTEP_OWNER; @@ -70,16 +71,18 @@ static int speedstep_smi_ownership (void) 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); + 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) + : "=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) + "D" (0), "S" (magic) : "memory" ); @@ -97,10 +100,10 @@ static int speedstep_smi_ownership (void) * 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) +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 state = 0; u32 function = GET_SPEEDSTEP_FREQS; if (!(ist_info.event & 0xFFFF)) { @@ -110,17 +113,25 @@ static int speedstep_smi_get_freqs (unsigned int *low, unsigned int *high) command = (smi_sig & 0xffffff00) | (smi_cmd & 0xff); - dprintk("trying to determine frequencies with command %x at port %x\n", command, smi_port); + 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) + : "=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); + 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) @@ -137,26 +148,30 @@ static int speedstep_smi_get_freqs (unsigned int *low, unsigned int *high) * @state: processor frequency state (SPEEDSTEP_LOW or SPEEDSTEP_HIGH) * */ -static int speedstep_get_state (void) +static int speedstep_get_state(void) { - u32 function=GET_SPEEDSTEP_STATE; + 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); + 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) + : "=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); + return state & 1; } @@ -165,11 +180,11 @@ static int speedstep_get_state (void) * @state: new processor frequency state (SPEEDSTEP_LOW or SPEEDSTEP_HIGH) * */ -static void speedstep_set_state (unsigned int state) +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 function = SET_SPEEDSTEP_STATE; unsigned int retry = 0; if (state > 0x1) @@ -180,11 +195,14 @@ static void speedstep_set_state (unsigned int state) 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); + 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); + dprintk("retry %u, previous result %u, waiting...\n", + retry, result); mdelay(retry * 50); } retry++; @@ -192,20 +210,26 @@ static void speedstep_set_state (unsigned int state) "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) + : "=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); - } + 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; } @@ -219,13 +243,14 @@ static void speedstep_set_state (unsigned int state) * * Sets a new CPUFreq policy/freq. */ -static int speedstep_target (struct cpufreq_policy *policy, +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)) + if (cpufreq_frequency_table_target(policy, &speedstep_freqs[0], + target_freq, relation, &newstate)) return -EINVAL; freqs.old = speedstep_freqs[speedstep_get_state()].frequency; @@ -250,7 +275,7 @@ static int speedstep_target (struct cpufreq_policy *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) +static int speedstep_verify(struct cpufreq_policy *policy) { return cpufreq_frequency_table_verify(policy, &speedstep_freqs[0]); } @@ -259,7 +284,8 @@ static int speedstep_verify (struct cpufreq_policy *policy) static int speedstep_cpu_init(struct cpufreq_policy *policy) { int result; - unsigned int speed,state; + unsigned int speed, state; + unsigned int *low, *high; /* capability check */ if (policy->cpu != 0) @@ -272,19 +298,23 @@ static int speedstep_cpu_init(struct cpufreq_policy *policy) } /* detect low and high frequency */ - result = speedstep_smi_get_freqs(&speedstep_freqs[SPEEDSTEP_LOW].frequency, - &speedstep_freqs[SPEEDSTEP_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"); + /* 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, - &speedstep_freqs[SPEEDSTEP_LOW].frequency, - &speedstep_freqs[SPEEDSTEP_HIGH].frequency, + low, high, NULL, &speedstep_set_state); if (result) { - dprintk("could not detect two different speeds -- aborting.\n"); + dprintk("could not detect two different speeds" + " -- aborting.\n"); return result; } else dprintk("workaround worked.\n"); @@ -295,7 +325,8 @@ static int speedstep_cpu_init(struct cpufreq_policy *policy) speed = speedstep_freqs[state].frequency; dprintk("currently at %s speed setting - %i MHz\n", - (speed == speedstep_freqs[SPEEDSTEP_LOW].frequency) ? "low" : "high", + (speed == speedstep_freqs[SPEEDSTEP_LOW].frequency) + ? "low" : "high", (speed / 1000)); /* cpuinfo and default policy values */ @@ -304,7 +335,7 @@ static int speedstep_cpu_init(struct cpufreq_policy *policy) result = cpufreq_frequency_table_cpuinfo(policy, speedstep_freqs); if (result) - return (result); + return result; cpufreq_frequency_table_get_attr(speedstep_freqs, policy->cpu); @@ -321,7 +352,7 @@ static unsigned int speedstep_get(unsigned int cpu) { if (cpu) return -ENODEV; - return speedstep_get_processor_frequency(speedstep_processor); + return speedstep_get_frequency(speedstep_processor); } @@ -335,7 +366,7 @@ static int speedstep_resume(struct cpufreq_policy *policy) return result; } -static struct freq_attr* speedstep_attr[] = { +static struct freq_attr *speedstep_attr[] = { &cpufreq_freq_attr_scaling_available_freqs, NULL, }; @@ -364,21 +395,23 @@ static int __init speedstep_init(void) speedstep_processor = speedstep_detect_processor(); switch (speedstep_processor) { - case SPEEDSTEP_PROCESSOR_PIII_T: - case SPEEDSTEP_PROCESSOR_PIII_C: - case SPEEDSTEP_PROCESSOR_PIII_C_EARLY: + 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"); + 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); + 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' */ @@ -416,17 +449,20 @@ 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_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_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_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 8ea6929e974..5fff00c70de 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -4,6 +4,7 @@ #include <linux/string.h> #include <linux/bitops.h> #include <linux/smp.h> +#include <linux/sched.h> #include <linux/thread_info.h> #include <linux/module.h> @@ -29,6 +30,19 @@ static void __cpuinit early_init_intel(struct cpuinfo_x86 *c) { + /* 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) { + misc_enable &= ~MSR_IA32_MISC_ENABLE_LIMIT_CPUID; + wrmsrl(MSR_IA32_MISC_ENABLE, misc_enable); + c->cpuid_level = cpuid_eax(0); + } + } + if ((c->x86 == 0xf && c->x86_model >= 0x03) || (c->x86 == 0x6 && c->x86_model >= 0x0e)) set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC); @@ -43,11 +57,16 @@ static void __cpuinit early_init_intel(struct cpuinfo_x86 *c) /* * c->x86_power is 8000_0007 edx. Bit 8 is TSC runs at constant rate - * with P/T states and does not stop in deep C-states + * with P/T states and does not stop in deep C-states. + * + * It is also reliable across cores and sockets. (but not across + * cabinets - we turn it off in that case explicitly.) */ if (c->x86_power & (1 << 8)) { set_cpu_cap(c, X86_FEATURE_CONSTANT_TSC); set_cpu_cap(c, X86_FEATURE_NONSTOP_TSC); + set_cpu_cap(c, X86_FEATURE_TSC_RELIABLE); + sched_clock_stable = 1; } } @@ -278,6 +297,9 @@ static void __cpuinit init_intel(struct cpuinfo_x86 *c) ds_init_intel(c); } + if (c->x86 == 6 && c->x86_model == 29 && cpu_has_clflush) + set_cpu_cap(c, X86_FEATURE_CLFLUSH_MONITOR); + #ifdef CONFIG_X86_64 if (c->x86 == 15) c->x86_cache_alignment = c->x86_clflush_size * 2; diff --git a/arch/x86/kernel/cpu/intel_cacheinfo.c b/arch/x86/kernel/cpu/intel_cacheinfo.c index 48533d77be7..da299eb85fc 100644 --- a/arch/x86/kernel/cpu/intel_cacheinfo.c +++ b/arch/x86/kernel/cpu/intel_cacheinfo.c @@ -36,8 +36,11 @@ static struct _cache_table cache_table[] __cpuinitdata = { { 0x06, LVL_1_INST, 8 }, /* 4-way set assoc, 32 byte line size */ { 0x08, LVL_1_INST, 16 }, /* 4-way set assoc, 32 byte line size */ + { 0x09, LVL_1_INST, 32 }, /* 4-way set assoc, 64 byte line size */ { 0x0a, LVL_1_DATA, 8 }, /* 2 way set assoc, 32 byte line size */ { 0x0c, LVL_1_DATA, 16 }, /* 4-way set assoc, 32 byte line size */ + { 0x0d, LVL_1_DATA, 16 }, /* 4-way set assoc, 64 byte line size */ + { 0x21, LVL_2, 256 }, /* 8-way set assoc, 64 byte line size */ { 0x22, LVL_3, 512 }, /* 4-way set assoc, sectored cache, 64 byte line size */ { 0x23, LVL_3, 1024 }, /* 8-way set assoc, sectored cache, 64 byte line size */ { 0x25, LVL_3, 2048 }, /* 8-way set assoc, sectored cache, 64 byte line size */ @@ -85,6 +88,18 @@ static struct _cache_table cache_table[] __cpuinitdata = { 0x85, LVL_2, 2048 }, /* 8-way set assoc, 32 byte line size */ { 0x86, LVL_2, 512 }, /* 4-way set assoc, 64 byte line size */ { 0x87, LVL_2, 1024 }, /* 8-way set assoc, 64 byte line size */ + { 0xd0, LVL_3, 512 }, /* 4-way set assoc, 64 byte line size */ + { 0xd1, LVL_3, 1024 }, /* 4-way set assoc, 64 byte line size */ + { 0xd2, LVL_3, 2048 }, /* 4-way set assoc, 64 byte line size */ + { 0xd6, LVL_3, 1024 }, /* 8-way set assoc, 64 byte line size */ + { 0xd7, LVL_3, 2038 }, /* 8-way set assoc, 64 byte line size */ + { 0xd8, LVL_3, 4096 }, /* 12-way set assoc, 64 byte line size */ + { 0xdc, LVL_3, 2048 }, /* 12-way set assoc, 64 byte line size */ + { 0xdd, LVL_3, 4096 }, /* 12-way set assoc, 64 byte line size */ + { 0xde, LVL_3, 8192 }, /* 12-way set assoc, 64 byte line size */ + { 0xe2, LVL_3, 2048 }, /* 16-way set assoc, 64 byte line size */ + { 0xe3, LVL_3, 4096 }, /* 16-way set assoc, 64 byte line size */ + { 0xe4, LVL_3, 8192 }, /* 16-way set assoc, 64 byte line size */ { 0x00, 0, 0} }; diff --git a/arch/x86/kernel/cpu/mcheck/mce_64.c b/arch/x86/kernel/cpu/mcheck/mce_64.c index 1c838032fd3..fe79985ce0f 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_64.c +++ b/arch/x86/kernel/cpu/mcheck/mce_64.c @@ -295,11 +295,11 @@ void do_machine_check(struct pt_regs * regs, long error_code) * If we know that the error was in user space, send a * SIGBUS. Otherwise, panic if tolerance is low. * - * do_exit() takes an awful lot of locks and has a slight + * force_sig() takes an awful lot of locks and has a slight * risk of deadlocking. */ if (user_space) { - do_exit(SIGBUS); + force_sig(SIGBUS, current); } else if (panic_on_oops || tolerant < 2) { mce_panic("Uncorrected machine check", &panicm, mcestart); @@ -490,7 +490,7 @@ static void __cpuinit mce_cpu_quirks(struct cpuinfo_x86 *c) } -static void __cpuinit mce_cpu_features(struct cpuinfo_x86 *c) +static void mce_cpu_features(struct cpuinfo_x86 *c) { switch (c->x86_vendor) { case X86_VENDOR_INTEL: @@ -734,6 +734,7 @@ __setup("mce=", mcheck_enable); static int mce_resume(struct sys_device *dev) { mce_init(NULL); + mce_cpu_features(¤t_cpu_data); return 0; } diff --git a/arch/x86/kernel/cpu/mcheck/mce_amd_64.c b/arch/x86/kernel/cpu/mcheck/mce_amd_64.c index 8ae8c4ff094..f2ee0ae29bd 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_amd_64.c +++ b/arch/x86/kernel/cpu/mcheck/mce_amd_64.c @@ -121,7 +121,7 @@ static long threshold_restart_bank(void *_tr) } /* cpu init entry point, called from mce.c with preempt off */ -void __cpuinit mce_amd_feature_init(struct cpuinfo_x86 *c) +void mce_amd_feature_init(struct cpuinfo_x86 *c) { unsigned int bank, block; unsigned int cpu = smp_processor_id(); diff --git a/arch/x86/kernel/cpu/mcheck/mce_intel_64.c b/arch/x86/kernel/cpu/mcheck/mce_intel_64.c index 4b48f251fd3..f44c3662436 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_intel_64.c +++ b/arch/x86/kernel/cpu/mcheck/mce_intel_64.c @@ -30,7 +30,7 @@ asmlinkage void smp_thermal_interrupt(void) irq_exit(); } -static void __cpuinit intel_init_thermal(struct cpuinfo_x86 *c) +static void intel_init_thermal(struct cpuinfo_x86 *c) { u32 l, h; int tm2 = 0; @@ -84,7 +84,7 @@ static void __cpuinit intel_init_thermal(struct cpuinfo_x86 *c) return; } -void __cpuinit mce_intel_feature_init(struct cpuinfo_x86 *c) +void mce_intel_feature_init(struct cpuinfo_x86 *c) { intel_init_thermal(c); } diff --git a/arch/x86/kernel/cpu/mtrr/generic.c b/arch/x86/kernel/cpu/mtrr/generic.c index b59ddcc88cd..0c0a455fe95 100644 --- a/arch/x86/kernel/cpu/mtrr/generic.c +++ b/arch/x86/kernel/cpu/mtrr/generic.c @@ -33,11 +33,13 @@ u64 mtrr_tom2; struct mtrr_state_type mtrr_state = {}; EXPORT_SYMBOL_GPL(mtrr_state); -#undef MODULE_PARAM_PREFIX -#define MODULE_PARAM_PREFIX "mtrr." - -static int mtrr_show; -module_param_named(show, mtrr_show, bool, 0); +static int __initdata mtrr_show; +static int __init mtrr_debug(char *opt) +{ + mtrr_show = 1; + return 0; +} +early_param("mtrr.show", mtrr_debug); /* * Returns the effective MTRR type for the region diff --git a/arch/x86/kernel/cpu/mtrr/main.c b/arch/x86/kernel/cpu/mtrr/main.c index d259e5d2e05..236a401b825 100644 --- a/arch/x86/kernel/cpu/mtrr/main.c +++ b/arch/x86/kernel/cpu/mtrr/main.c @@ -1594,8 +1594,7 @@ int __init mtrr_trim_uncached_memory(unsigned long end_pfn) /* kvm/qemu doesn't have mtrr set right, don't trim them all */ if (!highest_pfn) { - WARN(!kvm_para_available(), KERN_WARNING - "WARNING: strange, CPU MTRRs all blank?\n"); + printk(KERN_INFO "CPU MTRRs all blank - virtualized system.\n"); return 0; } diff --git a/arch/x86/kernel/ds.c b/arch/x86/kernel/ds.c index da91701a234..87b67e3a765 100644 --- a/arch/x86/kernel/ds.c +++ b/arch/x86/kernel/ds.c @@ -15,8 +15,8 @@ * - buffer allocation (memory accounting) * * - * Copyright (C) 2007-2008 Intel Corporation. - * Markus Metzger <markus.t.metzger@intel.com>, 2007-2008 + * Copyright (C) 2007-2009 Intel Corporation. + * Markus Metzger <markus.t.metzger@intel.com>, 2007-2009 */ @@ -729,7 +729,7 @@ struct pebs_tracer *ds_request_pebs(struct task_struct *task, spin_unlock_irqrestore(&ds_lock, irq); - ds_write_config(tracer->ds.context, &tracer->trace.ds, ds_bts); + ds_write_config(tracer->ds.context, &tracer->trace.ds, ds_pebs); ds_resume_pebs(tracer); return tracer; @@ -890,7 +890,7 @@ int ds_set_pebs_reset(struct pebs_tracer *tracer, u64 value) } static const struct ds_configuration ds_cfg_netburst = { - .name = "netburst", + .name = "Netburst", .ctl[dsf_bts] = (1 << 2) | (1 << 3), .ctl[dsf_bts_kernel] = (1 << 5), .ctl[dsf_bts_user] = (1 << 6), @@ -904,7 +904,7 @@ static const struct ds_configuration ds_cfg_netburst = { #endif }; static const struct ds_configuration ds_cfg_pentium_m = { - .name = "pentium m", + .name = "Pentium M", .ctl[dsf_bts] = (1 << 6) | (1 << 7), .sizeof_field = sizeof(long), @@ -915,8 +915,8 @@ static const struct ds_configuration ds_cfg_pentium_m = { .sizeof_rec[ds_pebs] = sizeof(long) * 18, #endif }; -static const struct ds_configuration ds_cfg_core2 = { - .name = "core 2", +static const struct ds_configuration ds_cfg_core2_atom = { + .name = "Core 2/Atom", .ctl[dsf_bts] = (1 << 6) | (1 << 7), .ctl[dsf_bts_kernel] = (1 << 9), .ctl[dsf_bts_user] = (1 << 10), @@ -949,19 +949,22 @@ void __cpuinit ds_init_intel(struct cpuinfo_x86 *c) switch (c->x86) { case 0x6: switch (c->x86_model) { - case 0 ... 0xC: - /* sorry, don't know about them */ - break; - case 0xD: - case 0xE: /* Pentium M */ + case 0x9: + case 0xd: /* Pentium M */ ds_configure(&ds_cfg_pentium_m); break; - default: /* Core2, Atom, ... */ - ds_configure(&ds_cfg_core2); + case 0xf: + case 0x17: /* Core2 */ + case 0x1c: /* Atom */ + ds_configure(&ds_cfg_core2_atom); + break; + case 0x1a: /* i7 */ + default: + /* sorry, don't know about them */ break; } break; - case 0xF: + case 0xf: switch (c->x86_model) { case 0x0: case 0x1: @@ -1026,5 +1029,4 @@ void ds_copy_thread(struct task_struct *tsk, struct task_struct *father) void ds_exit_thread(struct task_struct *tsk) { - WARN_ON(tsk->thread.ds_ctx); } diff --git a/arch/x86/kernel/efi.c b/arch/x86/kernel/efi.c index 1119d247fe1..eb1ef3b67dd 100644 --- a/arch/x86/kernel/efi.c +++ b/arch/x86/kernel/efi.c @@ -467,7 +467,7 @@ void __init efi_enter_virtual_mode(void) efi_memory_desc_t *md; efi_status_t status; unsigned long size; - u64 end, systab, addr, npages; + u64 end, systab, addr, npages, end_pfn; void *p, *va; efi.systab = NULL; @@ -479,7 +479,10 @@ void __init efi_enter_virtual_mode(void) size = md->num_pages << EFI_PAGE_SHIFT; end = md->phys_addr + size; - if (PFN_UP(end) <= max_low_pfn_mapped) + end_pfn = PFN_UP(end); + if (end_pfn <= max_low_pfn_mapped + || (end_pfn > (1UL << (32 - PAGE_SHIFT)) + && end_pfn <= max_pfn_mapped)) va = __va(md->phys_addr); else va = efi_ioremap(md->phys_addr, size); diff --git a/arch/x86/kernel/efi_64.c b/arch/x86/kernel/efi_64.c index 652c5287215..cb783b92c50 100644 --- a/arch/x86/kernel/efi_64.c +++ b/arch/x86/kernel/efi_64.c @@ -99,24 +99,11 @@ void __init efi_call_phys_epilog(void) void __iomem *__init efi_ioremap(unsigned long phys_addr, unsigned long size) { - static unsigned pages_mapped __initdata; - unsigned i, pages; - unsigned long offset; + unsigned long last_map_pfn; - pages = PFN_UP(phys_addr + size) - PFN_DOWN(phys_addr); - offset = phys_addr & ~PAGE_MASK; - phys_addr &= PAGE_MASK; - - if (pages_mapped + pages > MAX_EFI_IO_PAGES) + last_map_pfn = init_memory_mapping(phys_addr, phys_addr + size); + if ((last_map_pfn << PAGE_SHIFT) < phys_addr + size) return NULL; - for (i = 0; i < pages; i++) { - __set_fixmap(FIX_EFI_IO_MAP_FIRST_PAGE - pages_mapped, - phys_addr, PAGE_KERNEL); - phys_addr += PAGE_SIZE; - pages_mapped++; - } - - return (void __iomem *)__fix_to_virt(FIX_EFI_IO_MAP_FIRST_PAGE - \ - (pages_mapped - pages)) + offset; + return (void __iomem *)__va(phys_addr); } diff --git a/arch/x86/kernel/entry_64.S b/arch/x86/kernel/entry_64.S index e28c7a98779..a1346217e43 100644 --- a/arch/x86/kernel/entry_64.S +++ b/arch/x86/kernel/entry_64.S @@ -346,6 +346,7 @@ ENTRY(save_args) popq_cfi %rax /* move return address... */ mov %gs:pda_irqstackptr,%rsp EMPTY_FRAME 0 + pushq_cfi %rbp /* backlink for unwinder */ pushq_cfi %rax /* ... to the new stack */ /* * We entered an interrupt context - irqs are off: diff --git a/arch/x86/kernel/ftrace.c b/arch/x86/kernel/ftrace.c index 1b43086b097..231bdd3c5b1 100644 --- a/arch/x86/kernel/ftrace.c +++ b/arch/x86/kernel/ftrace.c @@ -488,20 +488,21 @@ void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr) * ignore such a protection. */ asm volatile( - "1: " _ASM_MOV " (%[parent_old]), %[old]\n" - "2: " _ASM_MOV " %[return_hooker], (%[parent_replaced])\n" + "1: " _ASM_MOV " (%[parent]), %[old]\n" + "2: " _ASM_MOV " %[return_hooker], (%[parent])\n" " movl $0, %[faulted]\n" + "3:\n" ".section .fixup, \"ax\"\n" - "3: movl $1, %[faulted]\n" + "4: movl $1, %[faulted]\n" + " jmp 3b\n" ".previous\n" - _ASM_EXTABLE(1b, 3b) - _ASM_EXTABLE(2b, 3b) + _ASM_EXTABLE(1b, 4b) + _ASM_EXTABLE(2b, 4b) - : [parent_replaced] "=r" (parent), [old] "=r" (old), - [faulted] "=r" (faulted) - : [parent_old] "0" (parent), [return_hooker] "r" (return_hooker) + : [old] "=r" (old), [faulted] "=r" (faulted) + : [parent] "r" (parent), [return_hooker] "r" (return_hooker) : "memory" ); diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c index cd759ad9069..648b3a2a3a4 100644 --- a/arch/x86/kernel/hpet.c +++ b/arch/x86/kernel/hpet.c @@ -80,6 +80,7 @@ static inline void hpet_clear_mapping(void) */ static int boot_hpet_disable; int hpet_force_user; +static int hpet_verbose; static int __init hpet_setup(char *str) { @@ -88,6 +89,8 @@ static int __init hpet_setup(char *str) boot_hpet_disable = 1; if (!strncmp("force", str, 5)) hpet_force_user = 1; + if (!strncmp("verbose", str, 7)) + hpet_verbose = 1; } return 1; } @@ -119,6 +122,43 @@ int is_hpet_enabled(void) } EXPORT_SYMBOL_GPL(is_hpet_enabled); +static void _hpet_print_config(const char *function, int line) +{ + u32 i, timers, l, h; + printk(KERN_INFO "hpet: %s(%d):\n", function, line); + l = hpet_readl(HPET_ID); + h = hpet_readl(HPET_PERIOD); + timers = ((l & HPET_ID_NUMBER) >> HPET_ID_NUMBER_SHIFT) + 1; + printk(KERN_INFO "hpet: ID: 0x%x, PERIOD: 0x%x\n", l, h); + l = hpet_readl(HPET_CFG); + h = hpet_readl(HPET_STATUS); + printk(KERN_INFO "hpet: CFG: 0x%x, STATUS: 0x%x\n", l, h); + l = hpet_readl(HPET_COUNTER); + h = hpet_readl(HPET_COUNTER+4); + printk(KERN_INFO "hpet: COUNTER_l: 0x%x, COUNTER_h: 0x%x\n", l, h); + + for (i = 0; i < timers; i++) { + l = hpet_readl(HPET_Tn_CFG(i)); + h = hpet_readl(HPET_Tn_CFG(i)+4); + printk(KERN_INFO "hpet: T%d: CFG_l: 0x%x, CFG_h: 0x%x\n", + i, l, h); + l = hpet_readl(HPET_Tn_CMP(i)); + h = hpet_readl(HPET_Tn_CMP(i)+4); + printk(KERN_INFO "hpet: T%d: CMP_l: 0x%x, CMP_h: 0x%x\n", + i, l, h); + l = hpet_readl(HPET_Tn_ROUTE(i)); + h = hpet_readl(HPET_Tn_ROUTE(i)+4); + printk(KERN_INFO "hpet: T%d ROUTE_l: 0x%x, ROUTE_h: 0x%x\n", + i, l, h); + } +} + +#define hpet_print_config() \ +do { \ + if (hpet_verbose) \ + _hpet_print_config(__FUNCTION__, __LINE__); \ +} while (0) + /* * When the hpet driver (/dev/hpet) is enabled, we need to reserve * timer 0 and timer 1 in case of RTC emulation. @@ -191,27 +231,37 @@ static struct clock_event_device hpet_clockevent = { .rating = 50, }; -static void hpet_start_counter(void) +static void hpet_stop_counter(void) { unsigned long cfg = hpet_readl(HPET_CFG); - cfg &= ~HPET_CFG_ENABLE; hpet_writel(cfg, HPET_CFG); hpet_writel(0, HPET_COUNTER); hpet_writel(0, HPET_COUNTER + 4); +} + +static void hpet_start_counter(void) +{ + unsigned long cfg = hpet_readl(HPET_CFG); cfg |= HPET_CFG_ENABLE; hpet_writel(cfg, HPET_CFG); } +static void hpet_restart_counter(void) +{ + hpet_stop_counter(); + hpet_start_counter(); +} + static void hpet_resume_device(void) { force_hpet_resume(); } -static void hpet_restart_counter(void) +static void hpet_resume_counter(void) { hpet_resume_device(); - hpet_start_counter(); + hpet_restart_counter(); } static void hpet_enable_legacy_int(void) @@ -259,27 +309,23 @@ static int hpet_setup_msi_irq(unsigned int irq); static void hpet_set_mode(enum clock_event_mode mode, struct clock_event_device *evt, int timer) { - unsigned long cfg, cmp, now; + unsigned long cfg; uint64_t delta; switch (mode) { case CLOCK_EVT_MODE_PERIODIC: + hpet_stop_counter(); delta = ((uint64_t)(NSEC_PER_SEC/HZ)) * evt->mult; delta >>= evt->shift; - now = hpet_readl(HPET_COUNTER); - cmp = now + (unsigned long) delta; cfg = hpet_readl(HPET_Tn_CFG(timer)); + /* Make sure we use edge triggered interrupts */ + cfg &= ~HPET_TN_LEVEL; cfg |= HPET_TN_ENABLE | HPET_TN_PERIODIC | HPET_TN_SETVAL | HPET_TN_32BIT; hpet_writel(cfg, HPET_Tn_CFG(timer)); - /* - * The first write after writing TN_SETVAL to the - * config register sets the counter value, the second - * write sets the period. - */ - hpet_writel(cmp, HPET_Tn_CMP(timer)); - udelay(1); hpet_writel((unsigned long) delta, HPET_Tn_CMP(timer)); + hpet_start_counter(); + hpet_print_config(); break; case CLOCK_EVT_MODE_ONESHOT: @@ -306,6 +352,7 @@ static void hpet_set_mode(enum clock_event_mode mode, irq_set_affinity(hdev->irq, cpumask_of(hdev->cpu)); enable_irq(hdev->irq); } + hpet_print_config(); break; } } @@ -524,6 +571,7 @@ static void hpet_msi_capability_lookup(unsigned int start_timer) num_timers = ((id & HPET_ID_NUMBER) >> HPET_ID_NUMBER_SHIFT); num_timers++; /* Value read out starts from 0 */ + hpet_print_config(); hpet_devs = kzalloc(sizeof(struct hpet_dev) * num_timers, GFP_KERNEL); if (!hpet_devs) @@ -628,11 +676,12 @@ static int hpet_cpuhp_notify(struct notifier_block *n, switch (action & 0xf) { case CPU_ONLINE: - INIT_DELAYED_WORK(&work.work, hpet_work); + INIT_DELAYED_WORK_ON_STACK(&work.work, hpet_work); init_completion(&work.complete); /* FIXME: add schedule_work_on() */ schedule_delayed_work_on(cpu, &work.work, 0); wait_for_completion(&work.complete); + destroy_timer_on_stack(&work.work.timer); break; case CPU_DEAD: if (hdev) { @@ -692,7 +741,7 @@ static struct clocksource clocksource_hpet = { .mask = HPET_MASK, .shift = HPET_SHIFT, .flags = CLOCK_SOURCE_IS_CONTINUOUS, - .resume = hpet_restart_counter, + .resume = hpet_resume_counter, #ifdef CONFIG_X86_64 .vread = vread_hpet, #endif @@ -704,7 +753,7 @@ static int hpet_clocksource_register(void) cycle_t t1; /* Start the counter */ - hpet_start_counter(); + hpet_restart_counter(); /* Verify whether hpet counter works */ t1 = read_hpet(); @@ -790,6 +839,7 @@ int __init hpet_enable(void) * information and the number of channels */ id = hpet_readl(HPET_ID); + hpet_print_config(); #ifdef CONFIG_HPET_EMULATE_RTC /* @@ -842,6 +892,7 @@ static __init int hpet_late_init(void) return -ENODEV; hpet_reserve_platform_timers(hpet_readl(HPET_ID)); + hpet_print_config(); for_each_online_cpu(cpu) { hpet_cpuhp_notify(NULL, CPU_ONLINE, (void *)(long)cpu); @@ -896,7 +947,7 @@ static unsigned long hpet_rtc_flags; static int hpet_prev_update_sec; static struct rtc_time hpet_alarm_time; static unsigned long hpet_pie_count; -static unsigned long hpet_t1_cmp; +static u32 hpet_t1_cmp; static unsigned long hpet_default_delta; static unsigned long hpet_pie_delta; static unsigned long hpet_pie_limit; @@ -904,6 +955,14 @@ static unsigned long hpet_pie_limit; static rtc_irq_handler irq_handler; /* + * Check that the hpet counter c1 is ahead of the c2 + */ +static inline int hpet_cnt_ahead(u32 c1, u32 c2) +{ + return (s32)(c2 - c1) < 0; +} + +/* * Registers a IRQ handler. */ int hpet_register_irq_handler(rtc_irq_handler handler) @@ -1074,7 +1133,7 @@ static void hpet_rtc_timer_reinit(void) hpet_t1_cmp += delta; hpet_writel(hpet_t1_cmp, HPET_T1_CMP); lost_ints++; - } while ((long)(hpet_readl(HPET_COUNTER) - hpet_t1_cmp) > 0); + } while (!hpet_cnt_ahead(hpet_t1_cmp, hpet_readl(HPET_COUNTER))); if (lost_ints) { if (hpet_rtc_flags & RTC_PIE) diff --git a/arch/x86/kernel/i387.c b/arch/x86/kernel/i387.c index b0f61f0dcd0..f2f8540a7f3 100644 --- a/arch/x86/kernel/i387.c +++ b/arch/x86/kernel/i387.c @@ -136,7 +136,7 @@ int init_fpu(struct task_struct *tsk) #ifdef CONFIG_X86_32 if (!HAVE_HWFP) { memset(tsk->thread.xstate, 0, xstate_size); - finit(); + finit_task(tsk); set_stopped_child_used_math(tsk); return 0; } diff --git a/arch/x86/kernel/i8237.c b/arch/x86/kernel/i8237.c index dbd6c1d1b63..b42ca694dc6 100644 --- a/arch/x86/kernel/i8237.c +++ b/arch/x86/kernel/i8237.c @@ -28,10 +28,10 @@ static int i8237A_resume(struct sys_device *dev) flags = claim_dma_lock(); - dma_outb(DMA1_RESET_REG, 0); - dma_outb(DMA2_RESET_REG, 0); + dma_outb(0, DMA1_RESET_REG); + dma_outb(0, DMA2_RESET_REG); - for (i = 0;i < 8;i++) { + for (i = 0; i < 8; i++) { set_dma_addr(i, 0x000000); /* DMA count is a bit weird so this is not 0 */ set_dma_count(i, 1); @@ -51,14 +51,14 @@ static int i8237A_suspend(struct sys_device *dev, pm_message_t state) } static struct sysdev_class i8237_sysdev_class = { - .name = "i8237", - .suspend = i8237A_suspend, - .resume = i8237A_resume, + .name = "i8237", + .suspend = i8237A_suspend, + .resume = i8237A_resume, }; static struct sys_device device_i8237A = { - .id = 0, - .cls = &i8237_sysdev_class, + .id = 0, + .cls = &i8237_sysdev_class, }; static int __init i8237A_init_sysfs(void) @@ -68,5 +68,4 @@ static int __init i8237A_init_sysfs(void) error = sysdev_register(&device_i8237A); return error; } - device_initcall(i8237A_init_sysfs); diff --git a/arch/x86/kernel/io_apic.c b/arch/x86/kernel/io_apic.c index 1c4a1302536..bc7ac4da90d 100644 --- a/arch/x86/kernel/io_apic.c +++ b/arch/x86/kernel/io_apic.c @@ -2528,14 +2528,15 @@ static void irq_complete_move(struct irq_desc **descp) vector = ~get_irq_regs()->orig_ax; me = smp_processor_id(); + + if (vector == cfg->vector && cpumask_test_cpu(me, cfg->domain)) { #ifdef CONFIG_NUMA_MIGRATE_IRQ_DESC *descp = desc = move_irq_desc(desc, me); /* get the new one */ cfg = desc->chip_data; #endif - - if (vector == cfg->vector && cpumask_test_cpu(me, cfg->domain)) send_cleanup_vector(cfg); + } } #else static inline void irq_complete_move(struct irq_desc **descp) {} @@ -3840,14 +3841,24 @@ int __init io_apic_get_redir_entries (int ioapic) void __init probe_nr_irqs_gsi(void) { - int idx; int nr = 0; - for (idx = 0; idx < nr_ioapics; idx++) - nr += io_apic_get_redir_entries(idx) + 1; - - if (nr > nr_irqs_gsi) + nr = acpi_probe_gsi(); + if (nr > nr_irqs_gsi) { nr_irqs_gsi = nr; + } else { + /* for acpi=off or acpi is not compiled in */ + int idx; + + nr = 0; + for (idx = 0; idx < nr_ioapics; idx++) + nr += io_apic_get_redir_entries(idx) + 1; + + if (nr > nr_irqs_gsi) + nr_irqs_gsi = nr; + } + + printk(KERN_DEBUG "nr_irqs_gsi: %d\n", nr_irqs_gsi); } /* -------------------------------------------------------------------------- diff --git a/arch/x86/kernel/irqinit_32.c b/arch/x86/kernel/irqinit_32.c index 1507ad4e674..10a09c2f182 100644 --- a/arch/x86/kernel/irqinit_32.c +++ b/arch/x86/kernel/irqinit_32.c @@ -78,15 +78,6 @@ void __init init_ISA_irqs(void) } } -/* - * IRQ2 is cascade interrupt to second interrupt controller - */ -static struct irqaction irq2 = { - .handler = no_action, - .mask = CPU_MASK_NONE, - .name = "cascade", -}; - DEFINE_PER_CPU(vector_irq_t, vector_irq) = { [0 ... IRQ0_VECTOR - 1] = -1, [IRQ0_VECTOR] = 0, @@ -178,9 +169,6 @@ void __init native_init_IRQ(void) alloc_intr_gate(THERMAL_APIC_VECTOR, thermal_interrupt); #endif - if (!acpi_ioapic) - setup_irq(2, &irq2); - /* setup after call gates are initialised (usually add in * the architecture specific gates) */ diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c index 884d985b8b8..4558dd3918c 100644 --- a/arch/x86/kernel/kprobes.c +++ b/arch/x86/kernel/kprobes.c @@ -193,6 +193,9 @@ static int __kprobes can_boost(kprobe_opcode_t *opcodes) kprobe_opcode_t opcode; kprobe_opcode_t *orig_opcodes = opcodes; + if (search_exception_tables(opcodes)) + return 0; /* Page fault may occur on this address. */ + retry: if (opcodes - orig_opcodes > MAX_INSN_SIZE - 1) return 0; @@ -446,7 +449,7 @@ void __kprobes arch_prepare_kretprobe(struct kretprobe_instance *ri, static void __kprobes setup_singlestep(struct kprobe *p, struct pt_regs *regs, struct kprobe_ctlblk *kcb) { -#if !defined(CONFIG_PREEMPT) || defined(CONFIG_PM) +#if !defined(CONFIG_PREEMPT) || defined(CONFIG_FREEZER) if (p->ainsn.boostable == 1 && !p->post_handler) { /* Boost up -- we can execute copied instructions directly */ reset_current_kprobe(); diff --git a/arch/x86/kernel/olpc.c b/arch/x86/kernel/olpc.c index 7a13fac63a1..4006c522adc 100644 --- a/arch/x86/kernel/olpc.c +++ b/arch/x86/kernel/olpc.c @@ -203,7 +203,7 @@ static void __init platform_detect(void) static void __init platform_detect(void) { /* stopgap until OFW support is added to the kernel */ - olpc_platform_info.boardrev = 0xc2; + olpc_platform_info.boardrev = olpc_board(0xc2); } #endif diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c index e4c8fb60887..c6520a4e85d 100644 --- a/arch/x86/kernel/paravirt.c +++ b/arch/x86/kernel/paravirt.c @@ -268,6 +268,32 @@ enum paravirt_lazy_mode paravirt_get_lazy_mode(void) return __get_cpu_var(paravirt_lazy_mode); } +void arch_flush_lazy_mmu_mode(void) +{ + preempt_disable(); + + if (paravirt_get_lazy_mode() == PARAVIRT_LAZY_MMU) { + WARN_ON(preempt_count() == 1); + arch_leave_lazy_mmu_mode(); + arch_enter_lazy_mmu_mode(); + } + + preempt_enable(); +} + +void arch_flush_lazy_cpu_mode(void) +{ + preempt_disable(); + + if (paravirt_get_lazy_mode() == PARAVIRT_LAZY_CPU) { + WARN_ON(preempt_count() == 1); + arch_leave_lazy_cpu_mode(); + arch_enter_lazy_cpu_mode(); + } + + preempt_enable(); +} + struct pv_info pv_info = { .name = "bare hardware", .paravirt_enabled = 0, diff --git a/arch/x86/kernel/pci-gart_64.c b/arch/x86/kernel/pci-gart_64.c index 00c2bcd4146..d5768b1af08 100644 --- a/arch/x86/kernel/pci-gart_64.c +++ b/arch/x86/kernel/pci-gart_64.c @@ -5,7 +5,7 @@ * This allows to use PCI devices that only support 32bit addresses on systems * with more than 4GB. * - * See Documentation/DMA-mapping.txt for the interface specification. + * See Documentation/PCI/PCI-DMA-mapping.txt for the interface specification. * * Copyright 2002 Andi Kleen, SuSE Labs. * Subject to the GNU General Public License v2 only. diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index e68bb9e3086..6d12f7e37f8 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -180,6 +180,9 @@ void mwait_idle_with_hints(unsigned long ax, unsigned long cx) trace_power_start(&it, POWER_CSTATE, (ax>>4)+1); if (!need_resched()) { + if (cpu_has(¤t_cpu_data, X86_FEATURE_CLFLUSH_MONITOR)) + clflush((void *)¤t_thread_info()->flags); + __monitor((void *)¤t_thread_info()->flags, 0, 0); smp_mb(); if (!need_resched()) @@ -194,6 +197,9 @@ static void mwait_idle(void) struct power_trace it; if (!need_resched()) { trace_power_start(&it, POWER_CSTATE, 1); + if (cpu_has(¤t_cpu_data, X86_FEATURE_CLFLUSH_MONITOR)) + clflush((void *)¤t_thread_info()->flags); + __monitor((void *)¤t_thread_info()->flags, 0, 0); smp_mb(); if (!need_resched()) diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c index a546f55c77b..bd4da2af08a 100644 --- a/arch/x86/kernel/process_32.c +++ b/arch/x86/kernel/process_32.c @@ -104,9 +104,6 @@ void cpu_idle(void) check_pgt_cache(); rmb(); - if (rcu_pending(cpu)) - rcu_check_callbacks(cpu, 0); - if (cpu_is_offline(cpu)) play_dead(); diff --git a/arch/x86/kernel/process_64.c b/arch/x86/kernel/process_64.c index 416fb9282f4..85b4cb5c198 100644 --- a/arch/x86/kernel/process_64.c +++ b/arch/x86/kernel/process_64.c @@ -40,6 +40,7 @@ #include <linux/uaccess.h> #include <linux/io.h> #include <linux/ftrace.h> +#include <linux/dmi.h> #include <asm/pgtable.h> #include <asm/system.h> @@ -151,14 +152,18 @@ void __show_regs(struct pt_regs *regs, int all) unsigned long d0, d1, d2, d3, d6, d7; unsigned int fsindex, gsindex; unsigned int ds, cs, es; + const char *board; printk("\n"); print_modules(); - printk(KERN_INFO "Pid: %d, comm: %.20s %s %s %.*s\n", + board = dmi_get_system_info(DMI_PRODUCT_NAME); + if (!board) + board = ""; + printk(KERN_INFO "Pid: %d, comm: %.20s %s %s %.*s %s\n", current->pid, current->comm, print_tainted(), init_utsname()->release, (int)strcspn(init_utsname()->version, " "), - init_utsname()->version); + init_utsname()->version, board); printk(KERN_INFO "RIP: %04lx:[<%016lx>] ", regs->cs & 0xffff, regs->ip); printk_address(regs->ip, 1); printk(KERN_INFO "RSP: %04lx:%016lx EFLAGS: %08lx\n", regs->ss, diff --git a/arch/x86/kernel/ptrace.c b/arch/x86/kernel/ptrace.c index 0a5df5f82fb..06ca07f6ad8 100644 --- a/arch/x86/kernel/ptrace.c +++ b/arch/x86/kernel/ptrace.c @@ -810,12 +810,16 @@ static void ptrace_bts_untrace(struct task_struct *child) static void ptrace_bts_detach(struct task_struct *child) { - if (unlikely(child->bts)) { - ds_release_bts(child->bts); - child->bts = NULL; - - ptrace_bts_free_buffer(child); - } + /* + * Ptrace_detach() races with ptrace_untrace() in case + * the child dies and is reaped by another thread. + * + * We only do the memory accounting at this point and + * leave the buffer deallocation and the bts tracer + * release to ptrace_bts_untrace() which will be called + * later on with tasklist_lock held. + */ + release_locked_buffer(child->bts_buffer, child->bts_size); } #else static inline void ptrace_bts_fork(struct task_struct *tsk) {} @@ -1384,7 +1388,7 @@ void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs, #ifdef CONFIG_X86_32 # define IS_IA32 1 #elif defined CONFIG_IA32_EMULATION -# define IS_IA32 test_thread_flag(TIF_IA32) +# define IS_IA32 is_compat_task() #else # define IS_IA32 0 #endif diff --git a/arch/x86/kernel/quirks.c b/arch/x86/kernel/quirks.c index 309949e9e1c..697d1b78cfb 100644 --- a/arch/x86/kernel/quirks.c +++ b/arch/x86/kernel/quirks.c @@ -172,7 +172,8 @@ DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH8_4, ich_force_enable_hpet); DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_ICH9_7, ich_force_enable_hpet); - +DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x3a16, /* ICH10 */ + ich_force_enable_hpet); static struct pci_dev *cached_dev; diff --git a/arch/x86/kernel/reboot.c b/arch/x86/kernel/reboot.c index 2b46eb41643..4526b3a75ed 100644 --- a/arch/x86/kernel/reboot.c +++ b/arch/x86/kernel/reboot.c @@ -217,6 +217,14 @@ static struct dmi_system_id __initdata reboot_dmi_table[] = { DMI_MATCH(DMI_PRODUCT_NAME, "HP Compaq"), }, }, + { /* Handle problems with rebooting on Dell XPS710 */ + .callback = set_bios_reboot, + .ident = "Dell XPS710", + .matches = { + DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."), + DMI_MATCH(DMI_PRODUCT_NAME, "Dell XPS710"), + }, + }, { } }; diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c index ae0d8042cf6..6a8811a6932 100644 --- a/arch/x86/kernel/setup.c +++ b/arch/x86/kernel/setup.c @@ -607,7 +607,7 @@ struct x86_quirks *x86_quirks __initdata = &default_x86_quirks; static int __init dmi_low_memory_corruption(const struct dmi_system_id *d) { printk(KERN_NOTICE - "%s detected: BIOS may corrupt low RAM, working it around.\n", + "%s detected: BIOS may corrupt low RAM, working around it.\n", d->ident); e820_update_range(0, 0x10000, E820_RAM, E820_RESERVED); @@ -770,6 +770,9 @@ void __init setup_arch(char **cmdline_p) finish_e820_parsing(); + if (efi_enabled) + efi_init(); + dmi_scan_machine(); dmi_check_system(bad_bios_dmi_table); @@ -789,8 +792,6 @@ void __init setup_arch(char **cmdline_p) insert_resource(&iomem_resource, &data_resource); insert_resource(&iomem_resource, &bss_resource); - if (efi_enabled) - efi_init(); #ifdef CONFIG_X86_32 if (ppro_with_ram_bug()) { diff --git a/arch/x86/kernel/setup_percpu.c b/arch/x86/kernel/setup_percpu.c index 55c46074eba..01161077a49 100644 --- a/arch/x86/kernel/setup_percpu.c +++ b/arch/x86/kernel/setup_percpu.c @@ -136,7 +136,7 @@ static void __init setup_cpu_pda_map(void) #ifdef CONFIG_X86_64 /* correctly size the local cpu masks */ -static void setup_cpu_local_masks(void) +static void __init setup_cpu_local_masks(void) { alloc_bootmem_cpumask_var(&cpu_initialized_mask); alloc_bootmem_cpumask_var(&cpu_callin_mask); diff --git a/arch/x86/kernel/signal.c b/arch/x86/kernel/signal.c index 89bb7668041..df0587f24c5 100644 --- a/arch/x86/kernel/signal.c +++ b/arch/x86/kernel/signal.c @@ -632,9 +632,16 @@ badframe: } #ifdef CONFIG_X86_32 -asmlinkage int sys_rt_sigreturn(struct pt_regs regs) +/* + * Note: do not pass in pt_regs directly as with tail-call optimization + * GCC will incorrectly stomp on the caller's frame and corrupt user-space + * register state: + */ +asmlinkage int sys_rt_sigreturn(unsigned long __unused) { - return do_rt_sigreturn(®s); + struct pt_regs *regs = (struct pt_regs *)&__unused; + + return do_rt_sigreturn(regs); } #else /* !CONFIG_X86_32 */ asmlinkage long sys_rt_sigreturn(struct pt_regs *regs) diff --git a/arch/x86/kernel/time_64.c b/arch/x86/kernel/time_64.c index e6e695acd72..241ec3923f6 100644 --- a/arch/x86/kernel/time_64.c +++ b/arch/x86/kernel/time_64.c @@ -115,7 +115,7 @@ unsigned long __init calibrate_cpu(void) static struct irqaction irq0 = { .handler = timer_interrupt, - .flags = IRQF_DISABLED | IRQF_IRQPOLL | IRQF_NOBALANCING, + .flags = IRQF_DISABLED | IRQF_IRQPOLL | IRQF_NOBALANCING | IRQF_TIMER, .mask = CPU_MASK_NONE, .name = "timer" }; diff --git a/arch/x86/kernel/tlb_uv.c b/arch/x86/kernel/tlb_uv.c index f885023167e..6812b829ed8 100644 --- a/arch/x86/kernel/tlb_uv.c +++ b/arch/x86/kernel/tlb_uv.c @@ -200,6 +200,7 @@ static int uv_wait_completion(struct bau_desc *bau_desc, destination_timeouts = 0; } } + cpu_relax(); } return FLUSH_COMPLETE; } diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c index 98c2d055284..a9e7548e179 100644 --- a/arch/x86/kernel/traps.c +++ b/arch/x86/kernel/traps.c @@ -99,6 +99,12 @@ static inline void preempt_conditional_sti(struct pt_regs *regs) local_irq_enable(); } +static inline void conditional_cli(struct pt_regs *regs) +{ + if (regs->flags & X86_EFLAGS_IF) + local_irq_disable(); +} + static inline void preempt_conditional_cli(struct pt_regs *regs) { if (regs->flags & X86_EFLAGS_IF) @@ -626,8 +632,10 @@ clear_dr7: #ifdef CONFIG_X86_32 debug_vm86: + /* reenable preemption: handle_vm86_trap() might sleep */ + dec_preempt_count(); handle_vm86_trap((struct kernel_vm86_regs *) regs, error_code, 1); - preempt_conditional_cli(regs); + conditional_cli(regs); return; #endif @@ -896,7 +904,7 @@ asmlinkage void math_state_restore(void) EXPORT_SYMBOL_GPL(math_state_restore); #ifndef CONFIG_MATH_EMULATION -asmlinkage void math_emulate(long arg) +void math_emulate(struct math_emu_info *info) { printk(KERN_EMERG "math-emulation not enabled and no coprocessor found.\n"); @@ -906,16 +914,19 @@ asmlinkage void math_emulate(long arg) } #endif /* CONFIG_MATH_EMULATION */ -dotraplinkage void __kprobes -do_device_not_available(struct pt_regs *regs, long error) +dotraplinkage void __kprobes do_device_not_available(struct pt_regs regs) { #ifdef CONFIG_X86_32 if (read_cr0() & X86_CR0_EM) { - conditional_sti(regs); - math_emulate(0); + struct math_emu_info info = { }; + + conditional_sti(®s); + + info.regs = ®s; + math_emulate(&info); } else { math_state_restore(); /* interrupts still off */ - conditional_sti(regs); + conditional_sti(®s); } #else math_state_restore(); diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c index 599e5816863..08afa1579e6 100644 --- a/arch/x86/kernel/tsc.c +++ b/arch/x86/kernel/tsc.c @@ -17,20 +17,21 @@ #include <asm/delay.h> #include <asm/hypervisor.h> -unsigned int cpu_khz; /* TSC clocks / usec, not used here */ +unsigned int __read_mostly cpu_khz; /* TSC clocks / usec, not used here */ EXPORT_SYMBOL(cpu_khz); -unsigned int tsc_khz; + +unsigned int __read_mostly tsc_khz; EXPORT_SYMBOL(tsc_khz); /* * TSC can be unstable due to cpufreq or due to unsynced TSCs */ -static int tsc_unstable; +static int __read_mostly tsc_unstable; /* native_sched_clock() is called before tsc_init(), so we must start with the TSC soft disabled to prevent erroneous rdtsc usage on !cpu_has_tsc processors */ -static int tsc_disabled = -1; +static int __read_mostly tsc_disabled = -1; static int tsc_clocksource_reliable; /* @@ -273,30 +274,43 @@ static unsigned long pit_calibrate_tsc(u32 latch, unsigned long ms, int loopmin) * use the TSC value at the transitions to calculate a pretty * good value for the TSC frequencty. */ -static inline int pit_expect_msb(unsigned char val) +static inline int pit_expect_msb(unsigned char val, u64 *tscp, unsigned long *deltap) { - int count = 0; + int count; + u64 tsc = 0; for (count = 0; count < 50000; count++) { /* Ignore LSB */ inb(0x42); if (inb(0x42) != val) break; + tsc = get_cycles(); } - return count > 50; + *deltap = get_cycles() - tsc; + *tscp = tsc; + + /* + * We require _some_ success, but the quality control + * will be based on the error terms on the TSC values. + */ + return count > 5; } /* - * How many MSB values do we want to see? We aim for a - * 15ms calibration, which assuming a 2us counter read - * error should give us roughly 150 ppm precision for - * the calibration. + * How many MSB values do we want to see? We aim for + * a maximum error rate of 500ppm (in practice the + * real error is much smaller), but refuse to spend + * more than 25ms on it. */ -#define QUICK_PIT_MS 15 -#define QUICK_PIT_ITERATIONS (QUICK_PIT_MS * PIT_TICK_RATE / 1000 / 256) +#define MAX_QUICK_PIT_MS 25 +#define MAX_QUICK_PIT_ITERATIONS (MAX_QUICK_PIT_MS * PIT_TICK_RATE / 1000 / 256) static unsigned long quick_pit_calibrate(void) { + int i; + u64 tsc, delta; + unsigned long d1, d2; + /* Set the Gate high, disable speaker */ outb((inb(0x61) & ~0x02) | 0x01, 0x61); @@ -315,45 +329,52 @@ static unsigned long quick_pit_calibrate(void) outb(0xff, 0x42); outb(0xff, 0x42); - if (pit_expect_msb(0xff)) { - int i; - u64 t1, t2, delta; - unsigned char expect = 0xfe; - - t1 = get_cycles(); - for (i = 0; i < QUICK_PIT_ITERATIONS; i++, expect--) { - if (!pit_expect_msb(expect)) - goto failed; + /* + * The PIT starts counting at the next edge, so we + * need to delay for a microsecond. The easiest way + * to do that is to just read back the 16-bit counter + * once from the PIT. + */ + inb(0x42); + inb(0x42); + + if (pit_expect_msb(0xff, &tsc, &d1)) { + for (i = 1; i <= MAX_QUICK_PIT_ITERATIONS; i++) { + if (!pit_expect_msb(0xff-i, &delta, &d2)) + break; + + /* + * Iterate until the error is less than 500 ppm + */ + delta -= tsc; + if (d1+d2 < delta >> 11) + goto success; } - t2 = get_cycles(); - - /* - * Make sure we can rely on the second TSC timestamp: - */ - if (!pit_expect_msb(expect)) - goto failed; - - /* - * Ok, if we get here, then we've seen the - * MSB of the PIT decrement QUICK_PIT_ITERATIONS - * times, and each MSB had many hits, so we never - * had any sudden jumps. - * - * As a result, we can depend on there not being - * any odd delays anywhere, and the TSC reads are - * reliable. - * - * kHz = ticks / time-in-seconds / 1000; - * kHz = (t2 - t1) / (QPI * 256 / PIT_TICK_RATE) / 1000 - * kHz = ((t2 - t1) * PIT_TICK_RATE) / (QPI * 256 * 1000) - */ - delta = (t2 - t1)*PIT_TICK_RATE; - do_div(delta, QUICK_PIT_ITERATIONS*256*1000); - printk("Fast TSC calibration using PIT\n"); - return delta; } -failed: + printk("Fast TSC calibration failed\n"); return 0; + +success: + /* + * Ok, if we get here, then we've seen the + * MSB of the PIT decrement 'i' times, and the + * error has shrunk to less than 500 ppm. + * + * As a result, we can depend on there not being + * any odd delays anywhere, and the TSC reads are + * reliable (within the error). We also adjust the + * delta to the middle of the error bars, just + * because it looks nicer. + * + * kHz = ticks / time-in-seconds / 1000; + * kHz = (t2 - t1) / (I * 256 / PIT_TICK_RATE) / 1000 + * kHz = ((t2 - t1) * PIT_TICK_RATE) / (I * 256 * 1000) + */ + delta += (long)(d2 - d1)/2; + delta *= PIT_TICK_RATE; + do_div(delta, i*256*1000); + printk("Fast TSC calibration using PIT\n"); + return delta; } /** @@ -523,8 +544,6 @@ unsigned long native_calibrate_tsc(void) return tsc_pit_min; } -#ifdef CONFIG_X86_32 -/* Only called from the Powernow K7 cpu freq driver */ int recalibrate_cpu_khz(void) { #ifndef CONFIG_SMP @@ -546,7 +565,6 @@ int recalibrate_cpu_khz(void) EXPORT_SYMBOL(recalibrate_cpu_khz); -#endif /* CONFIG_X86_32 */ /* Accelerators for sched_clock() * convert from cycles(64bits) => nanoseconds (64bits) diff --git a/arch/x86/kernel/vmi_32.c b/arch/x86/kernel/vmi_32.c index 23206ba1687..bef58b4982d 100644 --- a/arch/x86/kernel/vmi_32.c +++ b/arch/x86/kernel/vmi_32.c @@ -321,6 +321,16 @@ static void vmi_release_pmd(unsigned long pfn) } /* + * We use the pgd_free hook for releasing the pgd page: + */ +static void vmi_pgd_free(struct mm_struct *mm, pgd_t *pgd) +{ + unsigned long pfn = __pa(pgd) >> PAGE_SHIFT; + + vmi_ops.release_page(pfn, VMI_PAGE_L2); +} + +/* * Helper macros for MMU update flags. We can defer updates until a flush * or page invalidation only if the update is to the current address space * (otherwise, there is no flush). We must check against init_mm, since @@ -762,6 +772,7 @@ static inline int __init activate_vmi(void) if (vmi_ops.release_page) { pv_mmu_ops.release_pte = vmi_release_pte; pv_mmu_ops.release_pmd = vmi_release_pmd; + pv_mmu_ops.pgd_free = vmi_pgd_free; } /* Set linear is needed in all cases */ @@ -858,7 +869,7 @@ void __init vmi_init(void) #endif } -void vmi_activate(void) +void __init vmi_activate(void) { unsigned long flags; diff --git a/arch/x86/kernel/vmiclock_32.c b/arch/x86/kernel/vmiclock_32.c index c4c1f9e0940..e5b088fffa4 100644 --- a/arch/x86/kernel/vmiclock_32.c +++ b/arch/x86/kernel/vmiclock_32.c @@ -202,7 +202,7 @@ static irqreturn_t vmi_timer_interrupt(int irq, void *dev_id) static struct irqaction vmi_clock_action = { .name = "vmi-timer", .handler = vmi_timer_interrupt, - .flags = IRQF_DISABLED | IRQF_NOBALANCING, + .flags = IRQF_DISABLED | IRQF_NOBALANCING | IRQF_TIMER, .mask = CPU_MASK_ALL, }; @@ -283,10 +283,13 @@ void __devinit vmi_time_ap_init(void) #endif /** vmi clocksource */ +static struct clocksource clocksource_vmi; static cycle_t read_real_cycles(void) { - return vmi_timer_ops.get_cycle_counter(VMI_CYCLES_REAL); + cycle_t ret = (cycle_t)vmi_timer_ops.get_cycle_counter(VMI_CYCLES_REAL); + return ret >= clocksource_vmi.cycle_last ? + ret : clocksource_vmi.cycle_last; } static struct clocksource clocksource_vmi = { diff --git a/arch/x86/kvm/Kconfig b/arch/x86/kvm/Kconfig index b81125f0bde..0a303c3ed11 100644 --- a/arch/x86/kvm/Kconfig +++ b/arch/x86/kvm/Kconfig @@ -4,6 +4,10 @@ config HAVE_KVM bool +config HAVE_KVM_IRQCHIP + bool + default y + menuconfig VIRTUALIZATION bool "Virtualization" depends on HAVE_KVM || X86 diff --git a/arch/x86/kvm/i8254.c b/arch/x86/kvm/i8254.c index e665d1c623c..c13bb92d315 100644 --- a/arch/x86/kvm/i8254.c +++ b/arch/x86/kvm/i8254.c @@ -201,13 +201,16 @@ static int __pit_timer_fn(struct kvm_kpit_state *ps) if (!atomic_inc_and_test(&pt->pending)) set_bit(KVM_REQ_PENDING_TIMER, &vcpu0->requests); + if (!pt->reinject) + atomic_set(&pt->pending, 1); + if (vcpu0 && waitqueue_active(&vcpu0->wq)) wake_up_interruptible(&vcpu0->wq); hrtimer_add_expires_ns(&pt->timer, pt->period); pt->scheduled = hrtimer_get_expires_ns(&pt->timer); if (pt->period) - ps->channels[0].count_load_time = hrtimer_get_expires(&pt->timer); + ps->channels[0].count_load_time = ktime_get(); return (pt->period == 0 ? 0 : 1); } @@ -536,6 +539,16 @@ void kvm_pit_reset(struct kvm_pit *pit) pit->pit_state.irq_ack = 1; } +static void pit_mask_notifer(struct kvm_irq_mask_notifier *kimn, bool mask) +{ + struct kvm_pit *pit = container_of(kimn, struct kvm_pit, mask_notifier); + + if (!mask) { + atomic_set(&pit->pit_state.pit_timer.pending, 0); + pit->pit_state.irq_ack = 1; + } +} + struct kvm_pit *kvm_create_pit(struct kvm *kvm) { struct kvm_pit *pit; @@ -545,9 +558,7 @@ struct kvm_pit *kvm_create_pit(struct kvm *kvm) if (!pit) return NULL; - mutex_lock(&kvm->lock); pit->irq_source_id = kvm_request_irq_source_id(kvm); - mutex_unlock(&kvm->lock); if (pit->irq_source_id < 0) { kfree(pit); return NULL; @@ -580,10 +591,14 @@ struct kvm_pit *kvm_create_pit(struct kvm *kvm) pit_state->irq_ack_notifier.gsi = 0; pit_state->irq_ack_notifier.irq_acked = kvm_pit_ack_irq; kvm_register_irq_ack_notifier(kvm, &pit_state->irq_ack_notifier); + pit_state->pit_timer.reinject = true; mutex_unlock(&pit->pit_state.lock); kvm_pit_reset(pit); + pit->mask_notifier.func = pit_mask_notifer; + kvm_register_irq_mask_notifier(kvm, 0, &pit->mask_notifier); + return pit; } @@ -592,6 +607,8 @@ void kvm_free_pit(struct kvm *kvm) struct hrtimer *timer; if (kvm->arch.vpit) { + kvm_unregister_irq_mask_notifier(kvm, 0, + &kvm->arch.vpit->mask_notifier); mutex_lock(&kvm->arch.vpit->pit_state.lock); timer = &kvm->arch.vpit->pit_state.pit_timer.timer; hrtimer_cancel(timer); diff --git a/arch/x86/kvm/i8254.h b/arch/x86/kvm/i8254.h index 4178022b97a..6acbe4b505d 100644 --- a/arch/x86/kvm/i8254.h +++ b/arch/x86/kvm/i8254.h @@ -9,6 +9,7 @@ struct kvm_kpit_timer { s64 period; /* unit: ns */ s64 scheduled; atomic_t pending; + bool reinject; }; struct kvm_kpit_channel_state { @@ -45,6 +46,7 @@ struct kvm_pit { struct kvm *kvm; struct kvm_kpit_state pit_state; int irq_source_id; + struct kvm_irq_mask_notifier mask_notifier; }; #define KVM_PIT_BASE_ADDRESS 0x40 diff --git a/arch/x86/kvm/i8259.c b/arch/x86/kvm/i8259.c index 179dcb0103f..1ccb50c74f1 100644 --- a/arch/x86/kvm/i8259.c +++ b/arch/x86/kvm/i8259.c @@ -32,11 +32,13 @@ #include <linux/kvm_host.h> static void pic_lock(struct kvm_pic *s) + __acquires(&s->lock) { spin_lock(&s->lock); } static void pic_unlock(struct kvm_pic *s) + __releases(&s->lock) { struct kvm *kvm = s->kvm; unsigned acks = s->pending_acks; @@ -49,7 +51,8 @@ static void pic_unlock(struct kvm_pic *s) spin_unlock(&s->lock); while (acks) { - kvm_notify_acked_irq(kvm, __ffs(acks)); + kvm_notify_acked_irq(kvm, SELECT_PIC(__ffs(acks)), + __ffs(acks)); acks &= acks - 1; } @@ -76,12 +79,13 @@ void kvm_pic_clear_isr_ack(struct kvm *kvm) /* * set irq level. If an edge is detected, then the IRR is set to 1 */ -static inline void pic_set_irq1(struct kvm_kpic_state *s, int irq, int level) +static inline int pic_set_irq1(struct kvm_kpic_state *s, int irq, int level) { - int mask; + int mask, ret = 1; mask = 1 << irq; if (s->elcr & mask) /* level triggered */ if (level) { + ret = !(s->irr & mask); s->irr |= mask; s->last_irr |= mask; } else { @@ -90,11 +94,15 @@ static inline void pic_set_irq1(struct kvm_kpic_state *s, int irq, int level) } else /* edge triggered */ if (level) { - if ((s->last_irr & mask) == 0) + if ((s->last_irr & mask) == 0) { + ret = !(s->irr & mask); s->irr |= mask; + } s->last_irr |= mask; } else s->last_irr &= ~mask; + + return (s->imr & mask) ? -1 : ret; } /* @@ -171,16 +179,19 @@ void kvm_pic_update_irq(struct kvm_pic *s) pic_unlock(s); } -void kvm_pic_set_irq(void *opaque, int irq, int level) +int kvm_pic_set_irq(void *opaque, int irq, int level) { struct kvm_pic *s = opaque; + int ret = -1; pic_lock(s); if (irq >= 0 && irq < PIC_NUM_PINS) { - pic_set_irq1(&s->pics[irq >> 3], irq & 7, level); + ret = pic_set_irq1(&s->pics[irq >> 3], irq & 7, level); pic_update_irq(s); } pic_unlock(s); + + return ret; } /* @@ -232,7 +243,7 @@ int kvm_pic_read_irq(struct kvm *kvm) } pic_update_irq(s); pic_unlock(s); - kvm_notify_acked_irq(kvm, irq); + kvm_notify_acked_irq(kvm, SELECT_PIC(irq), irq); return intno; } diff --git a/arch/x86/kvm/irq.c b/arch/x86/kvm/irq.c index c019b8edcdb..cf17ed52f6f 100644 --- a/arch/x86/kvm/irq.c +++ b/arch/x86/kvm/irq.c @@ -87,13 +87,6 @@ void kvm_inject_pending_timer_irqs(struct kvm_vcpu *vcpu) } EXPORT_SYMBOL_GPL(kvm_inject_pending_timer_irqs); -void kvm_timer_intr_post(struct kvm_vcpu *vcpu, int vec) -{ - kvm_apic_timer_intr_post(vcpu, vec); - /* TODO: PIT, RTC etc. */ -} -EXPORT_SYMBOL_GPL(kvm_timer_intr_post); - void __kvm_migrate_timers(struct kvm_vcpu *vcpu) { __kvm_migrate_apic_timer(vcpu); diff --git a/arch/x86/kvm/irq.h b/arch/x86/kvm/irq.h index 2bf32a03cee..9f593188129 100644 --- a/arch/x86/kvm/irq.h +++ b/arch/x86/kvm/irq.h @@ -32,6 +32,8 @@ #include "lapic.h" #define PIC_NUM_PINS 16 +#define SELECT_PIC(irq) \ + ((irq) < 8 ? KVM_IRQCHIP_PIC_MASTER : KVM_IRQCHIP_PIC_SLAVE) struct kvm; struct kvm_vcpu; @@ -89,7 +91,6 @@ static inline int irqchip_in_kernel(struct kvm *kvm) void kvm_pic_reset(struct kvm_kpic_state *s); -void kvm_timer_intr_post(struct kvm_vcpu *vcpu, int vec); void kvm_inject_pending_timer_irqs(struct kvm_vcpu *vcpu); void kvm_inject_apic_timer_irqs(struct kvm_vcpu *vcpu); void kvm_apic_nmi_wd_deliver(struct kvm_vcpu *vcpu); diff --git a/arch/x86/kvm/kvm_svm.h b/arch/x86/kvm/kvm_svm.h index 8e5ee99551f..ed66e4c078d 100644 --- a/arch/x86/kvm/kvm_svm.h +++ b/arch/x86/kvm/kvm_svm.h @@ -18,7 +18,6 @@ static const u32 host_save_user_msrs[] = { }; #define NR_HOST_SAVE_USER_MSRS ARRAY_SIZE(host_save_user_msrs) -#define NUM_DB_REGS 4 struct kvm_vcpu; @@ -29,18 +28,23 @@ struct vcpu_svm { struct svm_cpu_data *svm_data; uint64_t asid_generation; - unsigned long db_regs[NUM_DB_REGS]; - u64 next_rip; u64 host_user_msrs[NR_HOST_SAVE_USER_MSRS]; u64 host_gs_base; unsigned long host_cr2; - unsigned long host_db_regs[NUM_DB_REGS]; - unsigned long host_dr6; - unsigned long host_dr7; u32 *msrpm; + struct vmcb *hsave; + u64 hsave_msr; + + u64 nested_vmcb; + + /* These are the merged vectors */ + u32 *nested_msrpm; + + /* gpa pointers to the real vectors */ + u64 nested_vmcb_msrpm; }; #endif diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c index afac68c0815..f0b67f2cdd6 100644 --- a/arch/x86/kvm/lapic.c +++ b/arch/x86/kvm/lapic.c @@ -35,6 +35,12 @@ #include "kvm_cache_regs.h" #include "irq.h" +#ifndef CONFIG_X86_64 +#define mod_64(x, y) ((x) - (y) * div64_u64(x, y)) +#else +#define mod_64(x, y) ((x) % (y)) +#endif + #define PRId64 "d" #define PRIx64 "llx" #define PRIu64 "u" @@ -511,52 +517,22 @@ static void apic_send_ipi(struct kvm_lapic *apic) static u32 apic_get_tmcct(struct kvm_lapic *apic) { - u64 counter_passed; - ktime_t passed, now; + ktime_t remaining; + s64 ns; u32 tmcct; ASSERT(apic != NULL); - now = apic->timer.dev.base->get_time(); - tmcct = apic_get_reg(apic, APIC_TMICT); - /* if initial count is 0, current count should also be 0 */ - if (tmcct == 0) + if (apic_get_reg(apic, APIC_TMICT) == 0) return 0; - if (unlikely(ktime_to_ns(now) <= - ktime_to_ns(apic->timer.last_update))) { - /* Wrap around */ - passed = ktime_add(( { - (ktime_t) { - .tv64 = KTIME_MAX - - (apic->timer.last_update).tv64}; } - ), now); - apic_debug("time elapsed\n"); - } else - passed = ktime_sub(now, apic->timer.last_update); - - counter_passed = div64_u64(ktime_to_ns(passed), - (APIC_BUS_CYCLE_NS * apic->timer.divide_count)); - - if (counter_passed > tmcct) { - if (unlikely(!apic_lvtt_period(apic))) { - /* one-shot timers stick at 0 until reset */ - tmcct = 0; - } else { - /* - * periodic timers reset to APIC_TMICT when they - * hit 0. The while loop simulates this happening N - * times. (counter_passed %= tmcct) would also work, - * but might be slower or not work on 32-bit?? - */ - while (counter_passed > tmcct) - counter_passed -= tmcct; - tmcct -= counter_passed; - } - } else { - tmcct -= counter_passed; - } + remaining = hrtimer_expires_remaining(&apic->timer.dev); + if (ktime_to_ns(remaining) < 0) + remaining = ktime_set(0, 0); + + ns = mod_64(ktime_to_ns(remaining), apic->timer.period); + tmcct = div64_u64(ns, (APIC_BUS_CYCLE_NS * apic->timer.divide_count)); return tmcct; } @@ -653,8 +629,6 @@ static void start_apic_timer(struct kvm_lapic *apic) { ktime_t now = apic->timer.dev.base->get_time(); - apic->timer.last_update = now; - apic->timer.period = apic_get_reg(apic, APIC_TMICT) * APIC_BUS_CYCLE_NS * apic->timer.divide_count; atomic_set(&apic->timer.pending, 0); @@ -1110,16 +1084,6 @@ void kvm_inject_apic_timer_irqs(struct kvm_vcpu *vcpu) } } -void kvm_apic_timer_intr_post(struct kvm_vcpu *vcpu, int vec) -{ - struct kvm_lapic *apic = vcpu->arch.apic; - - if (apic && apic_lvt_vector(apic, APIC_LVTT) == vec) - apic->timer.last_update = ktime_add_ns( - apic->timer.last_update, - apic->timer.period); -} - int kvm_get_apic_interrupt(struct kvm_vcpu *vcpu) { int vector = kvm_apic_has_interrupt(vcpu); diff --git a/arch/x86/kvm/lapic.h b/arch/x86/kvm/lapic.h index 81858881287..45ab6ee7120 100644 --- a/arch/x86/kvm/lapic.h +++ b/arch/x86/kvm/lapic.h @@ -12,7 +12,6 @@ struct kvm_lapic { atomic_t pending; s64 period; /* unit: ns */ u32 divide_count; - ktime_t last_update; struct hrtimer dev; } timer; struct kvm_vcpu *vcpu; @@ -42,7 +41,6 @@ void kvm_set_apic_base(struct kvm_vcpu *vcpu, u64 data); void kvm_apic_post_state_restore(struct kvm_vcpu *vcpu); int kvm_lapic_enabled(struct kvm_vcpu *vcpu); int kvm_lapic_find_highest_irr(struct kvm_vcpu *vcpu); -void kvm_apic_timer_intr_post(struct kvm_vcpu *vcpu, int vec); void kvm_lapic_set_vapic_addr(struct kvm_vcpu *vcpu, gpa_t vapic_addr); void kvm_lapic_sync_from_vapic(struct kvm_vcpu *vcpu); diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index 83f11c7474a..2a36f7f7c4c 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -145,11 +145,20 @@ struct kvm_rmap_desc { struct kvm_rmap_desc *more; }; -struct kvm_shadow_walk { - int (*entry)(struct kvm_shadow_walk *walk, struct kvm_vcpu *vcpu, - u64 addr, u64 *spte, int level); +struct kvm_shadow_walk_iterator { + u64 addr; + hpa_t shadow_addr; + int level; + u64 *sptep; + unsigned index; }; +#define for_each_shadow_entry(_vcpu, _addr, _walker) \ + for (shadow_walk_init(&(_walker), _vcpu, _addr); \ + shadow_walk_okay(&(_walker)); \ + shadow_walk_next(&(_walker))) + + struct kvm_unsync_walk { int (*entry) (struct kvm_mmu_page *sp, struct kvm_unsync_walk *walk); }; @@ -343,7 +352,6 @@ static void *mmu_memory_cache_alloc(struct kvm_mmu_memory_cache *mc, BUG_ON(!mc->nobjs); p = mc->objects[--mc->nobjs]; - memset(p, 0, size); return p; } @@ -794,10 +802,8 @@ static struct kvm_mmu_page *kvm_mmu_alloc_page(struct kvm_vcpu *vcpu, set_page_private(virt_to_page(sp->spt), (unsigned long)sp); list_add(&sp->link, &vcpu->kvm->arch.active_mmu_pages); INIT_LIST_HEAD(&sp->oos_link); - ASSERT(is_empty_shadow_page(sp->spt)); bitmap_zero(sp->slot_bitmap, KVM_MEMORY_SLOTS + KVM_PRIVATE_MEM_SLOTS); sp->multimapped = 0; - sp->global = 1; sp->parent_pte = parent_pte; --vcpu->kvm->arch.n_free_mmu_pages; return sp; @@ -983,8 +989,8 @@ struct kvm_mmu_pages { idx < 512; \ idx = find_next_bit(bitmap, 512, idx+1)) -int mmu_pages_add(struct kvm_mmu_pages *pvec, struct kvm_mmu_page *sp, - int idx) +static int mmu_pages_add(struct kvm_mmu_pages *pvec, struct kvm_mmu_page *sp, + int idx) { int i; @@ -1059,7 +1065,7 @@ static struct kvm_mmu_page *kvm_mmu_lookup_page(struct kvm *kvm, gfn_t gfn) index = kvm_page_table_hashfn(gfn); bucket = &kvm->arch.mmu_page_hash[index]; hlist_for_each_entry(sp, node, bucket, hash_link) - if (sp->gfn == gfn && !sp->role.metaphysical + if (sp->gfn == gfn && !sp->role.direct && !sp->role.invalid) { pgprintk("%s: found role %x\n", __func__, sp->role.word); @@ -1115,8 +1121,9 @@ struct mmu_page_path { i < pvec.nr && ({ sp = pvec.page[i].sp; 1;}); \ i = mmu_pages_next(&pvec, &parents, i)) -int mmu_pages_next(struct kvm_mmu_pages *pvec, struct mmu_page_path *parents, - int i) +static int mmu_pages_next(struct kvm_mmu_pages *pvec, + struct mmu_page_path *parents, + int i) { int n; @@ -1135,7 +1142,7 @@ int mmu_pages_next(struct kvm_mmu_pages *pvec, struct mmu_page_path *parents, return n; } -void mmu_pages_clear_parents(struct mmu_page_path *parents) +static void mmu_pages_clear_parents(struct mmu_page_path *parents) { struct kvm_mmu_page *sp; unsigned int level = 0; @@ -1193,7 +1200,7 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu, gfn_t gfn, gva_t gaddr, unsigned level, - int metaphysical, + int direct, unsigned access, u64 *parent_pte) { @@ -1204,10 +1211,9 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp; struct hlist_node *node, *tmp; - role.word = 0; - role.glevels = vcpu->arch.mmu.root_level; + role = vcpu->arch.mmu.base_role; role.level = level; - role.metaphysical = metaphysical; + role.direct = direct; role.access = access; if (vcpu->arch.mmu.root_level <= PT32_ROOT_LEVEL) { quadrant = gaddr >> (PAGE_SHIFT + (PT64_PT_BITS * level)); @@ -1242,8 +1248,9 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu, pgprintk("%s: adding gfn %lx role %x\n", __func__, gfn, role.word); sp->gfn = gfn; sp->role = role; + sp->global = role.cr4_pge; hlist_add_head(&sp->hash_link, bucket); - if (!metaphysical) { + if (!direct) { if (rmap_write_protect(vcpu->kvm, gfn)) kvm_flush_remote_tlbs(vcpu->kvm); account_shadowed(vcpu->kvm, gfn); @@ -1255,35 +1262,35 @@ static struct kvm_mmu_page *kvm_mmu_get_page(struct kvm_vcpu *vcpu, return sp; } -static int walk_shadow(struct kvm_shadow_walk *walker, - struct kvm_vcpu *vcpu, u64 addr) +static void shadow_walk_init(struct kvm_shadow_walk_iterator *iterator, + struct kvm_vcpu *vcpu, u64 addr) { - hpa_t shadow_addr; - int level; - int r; - u64 *sptep; - unsigned index; - - shadow_addr = vcpu->arch.mmu.root_hpa; - level = vcpu->arch.mmu.shadow_root_level; - if (level == PT32E_ROOT_LEVEL) { - shadow_addr = vcpu->arch.mmu.pae_root[(addr >> 30) & 3]; - shadow_addr &= PT64_BASE_ADDR_MASK; - if (!shadow_addr) - return 1; - --level; + iterator->addr = addr; + iterator->shadow_addr = vcpu->arch.mmu.root_hpa; + iterator->level = vcpu->arch.mmu.shadow_root_level; + if (iterator->level == PT32E_ROOT_LEVEL) { + iterator->shadow_addr + = vcpu->arch.mmu.pae_root[(addr >> 30) & 3]; + iterator->shadow_addr &= PT64_BASE_ADDR_MASK; + --iterator->level; + if (!iterator->shadow_addr) + iterator->level = 0; } +} - while (level >= PT_PAGE_TABLE_LEVEL) { - index = SHADOW_PT_INDEX(addr, level); - sptep = ((u64 *)__va(shadow_addr)) + index; - r = walker->entry(walker, vcpu, addr, sptep, level); - if (r) - return r; - shadow_addr = *sptep & PT64_BASE_ADDR_MASK; - --level; - } - return 0; +static bool shadow_walk_okay(struct kvm_shadow_walk_iterator *iterator) +{ + if (iterator->level < PT_PAGE_TABLE_LEVEL) + return false; + iterator->index = SHADOW_PT_INDEX(iterator->addr, iterator->level); + iterator->sptep = ((u64 *)__va(iterator->shadow_addr)) + iterator->index; + return true; +} + +static void shadow_walk_next(struct kvm_shadow_walk_iterator *iterator) +{ + iterator->shadow_addr = *iterator->sptep & PT64_BASE_ADDR_MASK; + --iterator->level; } static void kvm_mmu_page_unlink_children(struct kvm *kvm, @@ -1388,7 +1395,7 @@ static int kvm_mmu_zap_page(struct kvm *kvm, struct kvm_mmu_page *sp) kvm_mmu_page_unlink_children(kvm, sp); kvm_mmu_unlink_parents(kvm, sp); kvm_flush_remote_tlbs(kvm); - if (!sp->role.invalid && !sp->role.metaphysical) + if (!sp->role.invalid && !sp->role.direct) unaccount_shadowed(kvm, sp->gfn); if (sp->unsync) kvm_unlink_unsync_page(kvm, sp); @@ -1451,7 +1458,7 @@ static int kvm_mmu_unprotect_page(struct kvm *kvm, gfn_t gfn) index = kvm_page_table_hashfn(gfn); bucket = &kvm->arch.mmu_page_hash[index]; hlist_for_each_entry_safe(sp, node, n, bucket, hash_link) - if (sp->gfn == gfn && !sp->role.metaphysical) { + if (sp->gfn == gfn && !sp->role.direct) { pgprintk("%s: gfn %lx role %x\n", __func__, gfn, sp->role.word); r = 1; @@ -1463,11 +1470,20 @@ static int kvm_mmu_unprotect_page(struct kvm *kvm, gfn_t gfn) static void mmu_unshadow(struct kvm *kvm, gfn_t gfn) { + unsigned index; + struct hlist_head *bucket; struct kvm_mmu_page *sp; + struct hlist_node *node, *nn; - while ((sp = kvm_mmu_lookup_page(kvm, gfn)) != NULL) { - pgprintk("%s: zap %lx %x\n", __func__, gfn, sp->role.word); - kvm_mmu_zap_page(kvm, sp); + index = kvm_page_table_hashfn(gfn); + bucket = &kvm->arch.mmu_page_hash[index]; + hlist_for_each_entry_safe(sp, node, nn, bucket, hash_link) { + if (sp->gfn == gfn && !sp->role.direct + && !sp->role.invalid) { + pgprintk("%s: zap %lx %x\n", + __func__, gfn, sp->role.word); + kvm_mmu_zap_page(kvm, sp); + } } } @@ -1622,7 +1638,7 @@ static int kvm_unsync_page(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp) bucket = &vcpu->kvm->arch.mmu_page_hash[index]; /* don't unsync if pagetable is shadowed with multiple roles */ hlist_for_each_entry_safe(s, node, n, bucket, hash_link) { - if (s->gfn != sp->gfn || s->role.metaphysical) + if (s->gfn != sp->gfn || s->role.direct) continue; if (s->role.word != sp->role.word) return 1; @@ -1669,8 +1685,6 @@ static int set_spte(struct kvm_vcpu *vcpu, u64 *shadow_pte, u64 mt_mask = shadow_mt_mask; struct kvm_mmu_page *sp = page_header(__pa(shadow_pte)); - if (!(vcpu->arch.cr4 & X86_CR4_PGE)) - global = 0; if (!global && sp->global) { sp->global = 0; if (sp->unsync) { @@ -1698,8 +1712,13 @@ static int set_spte(struct kvm_vcpu *vcpu, u64 *shadow_pte, if (largepage) spte |= PT_PAGE_SIZE_MASK; if (mt_mask) { - mt_mask = get_memory_type(vcpu, gfn) << - kvm_x86_ops->get_mt_mask_shift(); + if (!kvm_is_mmio_pfn(pfn)) { + mt_mask = get_memory_type(vcpu, gfn) << + kvm_x86_ops->get_mt_mask_shift(); + mt_mask |= VMX_EPT_IGMT_BIT; + } else + mt_mask = MTRR_TYPE_UNCACHABLE << + kvm_x86_ops->get_mt_mask_shift(); spte |= mt_mask; } @@ -1772,12 +1791,8 @@ static void mmu_set_spte(struct kvm_vcpu *vcpu, u64 *shadow_pte, pgprintk("hfn old %lx new %lx\n", spte_to_pfn(*shadow_pte), pfn); rmap_remove(vcpu->kvm, shadow_pte); - } else { - if (largepage) - was_rmapped = is_large_pte(*shadow_pte); - else - was_rmapped = 1; - } + } else + was_rmapped = 1; } if (set_spte(vcpu, shadow_pte, pte_access, user_fault, write_fault, dirty, largepage, global, gfn, pfn, speculative, true)) { @@ -1815,67 +1830,42 @@ static void nonpaging_new_cr3(struct kvm_vcpu *vcpu) { } -struct direct_shadow_walk { - struct kvm_shadow_walk walker; - pfn_t pfn; - int write; - int largepage; - int pt_write; -}; - -static int direct_map_entry(struct kvm_shadow_walk *_walk, - struct kvm_vcpu *vcpu, - u64 addr, u64 *sptep, int level) +static int __direct_map(struct kvm_vcpu *vcpu, gpa_t v, int write, + int largepage, gfn_t gfn, pfn_t pfn) { - struct direct_shadow_walk *walk = - container_of(_walk, struct direct_shadow_walk, walker); + struct kvm_shadow_walk_iterator iterator; struct kvm_mmu_page *sp; + int pt_write = 0; gfn_t pseudo_gfn; - gfn_t gfn = addr >> PAGE_SHIFT; - - if (level == PT_PAGE_TABLE_LEVEL - || (walk->largepage && level == PT_DIRECTORY_LEVEL)) { - mmu_set_spte(vcpu, sptep, ACC_ALL, ACC_ALL, - 0, walk->write, 1, &walk->pt_write, - walk->largepage, 0, gfn, walk->pfn, false); - ++vcpu->stat.pf_fixed; - return 1; - } - if (*sptep == shadow_trap_nonpresent_pte) { - pseudo_gfn = (addr & PT64_DIR_BASE_ADDR_MASK) >> PAGE_SHIFT; - sp = kvm_mmu_get_page(vcpu, pseudo_gfn, (gva_t)addr, level - 1, - 1, ACC_ALL, sptep); - if (!sp) { - pgprintk("nonpaging_map: ENOMEM\n"); - kvm_release_pfn_clean(walk->pfn); - return -ENOMEM; + for_each_shadow_entry(vcpu, (u64)gfn << PAGE_SHIFT, iterator) { + if (iterator.level == PT_PAGE_TABLE_LEVEL + || (largepage && iterator.level == PT_DIRECTORY_LEVEL)) { + mmu_set_spte(vcpu, iterator.sptep, ACC_ALL, ACC_ALL, + 0, write, 1, &pt_write, + largepage, 0, gfn, pfn, false); + ++vcpu->stat.pf_fixed; + break; } - set_shadow_pte(sptep, - __pa(sp->spt) - | PT_PRESENT_MASK | PT_WRITABLE_MASK - | shadow_user_mask | shadow_x_mask); - } - return 0; -} + if (*iterator.sptep == shadow_trap_nonpresent_pte) { + pseudo_gfn = (iterator.addr & PT64_DIR_BASE_ADDR_MASK) >> PAGE_SHIFT; + sp = kvm_mmu_get_page(vcpu, pseudo_gfn, iterator.addr, + iterator.level - 1, + 1, ACC_ALL, iterator.sptep); + if (!sp) { + pgprintk("nonpaging_map: ENOMEM\n"); + kvm_release_pfn_clean(pfn); + return -ENOMEM; + } -static int __direct_map(struct kvm_vcpu *vcpu, gpa_t v, int write, - int largepage, gfn_t gfn, pfn_t pfn) -{ - int r; - struct direct_shadow_walk walker = { - .walker = { .entry = direct_map_entry, }, - .pfn = pfn, - .largepage = largepage, - .write = write, - .pt_write = 0, - }; - - r = walk_shadow(&walker.walker, vcpu, gfn << PAGE_SHIFT); - if (r < 0) - return r; - return walker.pt_write; + set_shadow_pte(iterator.sptep, + __pa(sp->spt) + | PT_PRESENT_MASK | PT_WRITABLE_MASK + | shadow_user_mask | shadow_x_mask); + } + } + return pt_write; } static int nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write, gfn_t gfn) @@ -1957,7 +1947,7 @@ static void mmu_alloc_roots(struct kvm_vcpu *vcpu) int i; gfn_t root_gfn; struct kvm_mmu_page *sp; - int metaphysical = 0; + int direct = 0; root_gfn = vcpu->arch.cr3 >> PAGE_SHIFT; @@ -1966,18 +1956,18 @@ static void mmu_alloc_roots(struct kvm_vcpu *vcpu) ASSERT(!VALID_PAGE(root)); if (tdp_enabled) - metaphysical = 1; + direct = 1; sp = kvm_mmu_get_page(vcpu, root_gfn, 0, - PT64_ROOT_LEVEL, metaphysical, + PT64_ROOT_LEVEL, direct, ACC_ALL, NULL); root = __pa(sp->spt); ++sp->root_count; vcpu->arch.mmu.root_hpa = root; return; } - metaphysical = !is_paging(vcpu); + direct = !is_paging(vcpu); if (tdp_enabled) - metaphysical = 1; + direct = 1; for (i = 0; i < 4; ++i) { hpa_t root = vcpu->arch.mmu.pae_root[i]; @@ -1991,7 +1981,7 @@ static void mmu_alloc_roots(struct kvm_vcpu *vcpu) } else if (vcpu->arch.mmu.root_level == 0) root_gfn = 0; sp = kvm_mmu_get_page(vcpu, root_gfn, i << 30, - PT32_ROOT_LEVEL, metaphysical, + PT32_ROOT_LEVEL, direct, ACC_ALL, NULL); root = __pa(sp->spt); ++sp->root_count; @@ -2246,17 +2236,23 @@ static int init_kvm_tdp_mmu(struct kvm_vcpu *vcpu) static int init_kvm_softmmu(struct kvm_vcpu *vcpu) { + int r; + ASSERT(vcpu); ASSERT(!VALID_PAGE(vcpu->arch.mmu.root_hpa)); if (!is_paging(vcpu)) - return nonpaging_init_context(vcpu); + r = nonpaging_init_context(vcpu); else if (is_long_mode(vcpu)) - return paging64_init_context(vcpu); + r = paging64_init_context(vcpu); else if (is_pae(vcpu)) - return paging32E_init_context(vcpu); + r = paging32E_init_context(vcpu); else - return paging32_init_context(vcpu); + r = paging32_init_context(vcpu); + + vcpu->arch.mmu.base_role.glevels = vcpu->arch.mmu.root_level; + + return r; } static int init_kvm_mmu(struct kvm_vcpu *vcpu) @@ -2487,7 +2483,7 @@ void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa, index = kvm_page_table_hashfn(gfn); bucket = &vcpu->kvm->arch.mmu_page_hash[index]; hlist_for_each_entry_safe(sp, node, n, bucket, hash_link) { - if (sp->gfn != gfn || sp->role.metaphysical || sp->role.invalid) + if (sp->gfn != gfn || sp->role.direct || sp->role.invalid) continue; pte_size = sp->role.glevels == PT32_ROOT_LEVEL ? 4 : 8; misaligned = (offset ^ (offset + bytes - 1)) & ~(pte_size - 1); @@ -3125,7 +3121,7 @@ static void audit_write_protection(struct kvm_vcpu *vcpu) gfn_t gfn; list_for_each_entry(sp, &vcpu->kvm->arch.active_mmu_pages, link) { - if (sp->role.metaphysical) + if (sp->role.direct) continue; gfn = unalias_gfn(vcpu->kvm, sp->gfn); diff --git a/arch/x86/kvm/mmu.h b/arch/x86/kvm/mmu.h index 258e5d56298..eaab2145f62 100644 --- a/arch/x86/kvm/mmu.h +++ b/arch/x86/kvm/mmu.h @@ -54,7 +54,7 @@ static inline int kvm_mmu_reload(struct kvm_vcpu *vcpu) static inline int is_long_mode(struct kvm_vcpu *vcpu) { #ifdef CONFIG_X86_64 - return vcpu->arch.shadow_efer & EFER_LME; + return vcpu->arch.shadow_efer & EFER_LMA; #else return 0; #endif diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h index 9fd78b6e17a..6bd70206c56 100644 --- a/arch/x86/kvm/paging_tmpl.h +++ b/arch/x86/kvm/paging_tmpl.h @@ -25,7 +25,6 @@ #if PTTYPE == 64 #define pt_element_t u64 #define guest_walker guest_walker64 - #define shadow_walker shadow_walker64 #define FNAME(name) paging##64_##name #define PT_BASE_ADDR_MASK PT64_BASE_ADDR_MASK #define PT_DIR_BASE_ADDR_MASK PT64_DIR_BASE_ADDR_MASK @@ -42,7 +41,6 @@ #elif PTTYPE == 32 #define pt_element_t u32 #define guest_walker guest_walker32 - #define shadow_walker shadow_walker32 #define FNAME(name) paging##32_##name #define PT_BASE_ADDR_MASK PT32_BASE_ADDR_MASK #define PT_DIR_BASE_ADDR_MASK PT32_DIR_BASE_ADDR_MASK @@ -73,18 +71,6 @@ struct guest_walker { u32 error_code; }; -struct shadow_walker { - struct kvm_shadow_walk walker; - struct guest_walker *guest_walker; - int user_fault; - int write_fault; - int largepage; - int *ptwrite; - pfn_t pfn; - u64 *sptep; - gpa_t pte_gpa; -}; - static gfn_t gpte_to_gfn(pt_element_t gpte) { return (gpte & PT_BASE_ADDR_MASK) >> PAGE_SHIFT; @@ -283,91 +269,79 @@ static void FNAME(update_pte)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *page, /* * Fetch a shadow pte for a specific level in the paging hierarchy. */ -static int FNAME(shadow_walk_entry)(struct kvm_shadow_walk *_sw, - struct kvm_vcpu *vcpu, u64 addr, - u64 *sptep, int level) +static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr, + struct guest_walker *gw, + int user_fault, int write_fault, int largepage, + int *ptwrite, pfn_t pfn) { - struct shadow_walker *sw = - container_of(_sw, struct shadow_walker, walker); - struct guest_walker *gw = sw->guest_walker; unsigned access = gw->pt_access; struct kvm_mmu_page *shadow_page; - u64 spte; - int metaphysical; + u64 spte, *sptep; + int direct; gfn_t table_gfn; int r; + int level; pt_element_t curr_pte; + struct kvm_shadow_walk_iterator iterator; - if (level == PT_PAGE_TABLE_LEVEL - || (sw->largepage && level == PT_DIRECTORY_LEVEL)) { - mmu_set_spte(vcpu, sptep, access, gw->pte_access & access, - sw->user_fault, sw->write_fault, - gw->ptes[gw->level-1] & PT_DIRTY_MASK, - sw->ptwrite, sw->largepage, - gw->ptes[gw->level-1] & PT_GLOBAL_MASK, - gw->gfn, sw->pfn, false); - sw->sptep = sptep; - return 1; - } + if (!is_present_pte(gw->ptes[gw->level - 1])) + return NULL; - if (is_shadow_present_pte(*sptep) && !is_large_pte(*sptep)) - return 0; + for_each_shadow_entry(vcpu, addr, iterator) { + level = iterator.level; + sptep = iterator.sptep; + if (level == PT_PAGE_TABLE_LEVEL + || (largepage && level == PT_DIRECTORY_LEVEL)) { + mmu_set_spte(vcpu, sptep, access, + gw->pte_access & access, + user_fault, write_fault, + gw->ptes[gw->level-1] & PT_DIRTY_MASK, + ptwrite, largepage, + gw->ptes[gw->level-1] & PT_GLOBAL_MASK, + gw->gfn, pfn, false); + break; + } - if (is_large_pte(*sptep)) { - set_shadow_pte(sptep, shadow_trap_nonpresent_pte); - kvm_flush_remote_tlbs(vcpu->kvm); - rmap_remove(vcpu->kvm, sptep); - } + if (is_shadow_present_pte(*sptep) && !is_large_pte(*sptep)) + continue; - if (level == PT_DIRECTORY_LEVEL && gw->level == PT_DIRECTORY_LEVEL) { - metaphysical = 1; - if (!is_dirty_pte(gw->ptes[level - 1])) - access &= ~ACC_WRITE_MASK; - table_gfn = gpte_to_gfn(gw->ptes[level - 1]); - } else { - metaphysical = 0; - table_gfn = gw->table_gfn[level - 2]; - } - shadow_page = kvm_mmu_get_page(vcpu, table_gfn, (gva_t)addr, level-1, - metaphysical, access, sptep); - if (!metaphysical) { - r = kvm_read_guest_atomic(vcpu->kvm, gw->pte_gpa[level - 2], - &curr_pte, sizeof(curr_pte)); - if (r || curr_pte != gw->ptes[level - 2]) { - kvm_mmu_put_page(shadow_page, sptep); - kvm_release_pfn_clean(sw->pfn); - sw->sptep = NULL; - return 1; + if (is_large_pte(*sptep)) { + rmap_remove(vcpu->kvm, sptep); + set_shadow_pte(sptep, shadow_trap_nonpresent_pte); + kvm_flush_remote_tlbs(vcpu->kvm); } - } - spte = __pa(shadow_page->spt) | PT_PRESENT_MASK | PT_ACCESSED_MASK - | PT_WRITABLE_MASK | PT_USER_MASK; - *sptep = spte; - return 0; -} - -static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr, - struct guest_walker *guest_walker, - int user_fault, int write_fault, int largepage, - int *ptwrite, pfn_t pfn) -{ - struct shadow_walker walker = { - .walker = { .entry = FNAME(shadow_walk_entry), }, - .guest_walker = guest_walker, - .user_fault = user_fault, - .write_fault = write_fault, - .largepage = largepage, - .ptwrite = ptwrite, - .pfn = pfn, - }; - - if (!is_present_pte(guest_walker->ptes[guest_walker->level - 1])) - return NULL; + if (level == PT_DIRECTORY_LEVEL + && gw->level == PT_DIRECTORY_LEVEL) { + direct = 1; + if (!is_dirty_pte(gw->ptes[level - 1])) + access &= ~ACC_WRITE_MASK; + table_gfn = gpte_to_gfn(gw->ptes[level - 1]); + } else { + direct = 0; + table_gfn = gw->table_gfn[level - 2]; + } + shadow_page = kvm_mmu_get_page(vcpu, table_gfn, addr, level-1, + direct, access, sptep); + if (!direct) { + r = kvm_read_guest_atomic(vcpu->kvm, + gw->pte_gpa[level - 2], + &curr_pte, sizeof(curr_pte)); + if (r || curr_pte != gw->ptes[level - 2]) { + kvm_mmu_put_page(shadow_page, sptep); + kvm_release_pfn_clean(pfn); + sptep = NULL; + break; + } + } - walk_shadow(&walker.walker, vcpu, addr); + spte = __pa(shadow_page->spt) + | PT_PRESENT_MASK | PT_ACCESSED_MASK + | PT_WRITABLE_MASK | PT_USER_MASK; + *sptep = spte; + } - return walker.sptep; + return sptep; } /* @@ -465,54 +439,56 @@ out_unlock: return 0; } -static int FNAME(shadow_invlpg_entry)(struct kvm_shadow_walk *_sw, - struct kvm_vcpu *vcpu, u64 addr, - u64 *sptep, int level) +static void FNAME(invlpg)(struct kvm_vcpu *vcpu, gva_t gva) { - struct shadow_walker *sw = - container_of(_sw, struct shadow_walker, walker); - - /* FIXME: properly handle invlpg on large guest pages */ - if (level == PT_PAGE_TABLE_LEVEL || - ((level == PT_DIRECTORY_LEVEL) && is_large_pte(*sptep))) { - struct kvm_mmu_page *sp = page_header(__pa(sptep)); + struct kvm_shadow_walk_iterator iterator; + pt_element_t gpte; + gpa_t pte_gpa = -1; + int level; + u64 *sptep; + int need_flush = 0; - sw->pte_gpa = (sp->gfn << PAGE_SHIFT); - sw->pte_gpa += (sptep - sp->spt) * sizeof(pt_element_t); + spin_lock(&vcpu->kvm->mmu_lock); - if (is_shadow_present_pte(*sptep)) { - rmap_remove(vcpu->kvm, sptep); - if (is_large_pte(*sptep)) - --vcpu->kvm->stat.lpages; + for_each_shadow_entry(vcpu, gva, iterator) { + level = iterator.level; + sptep = iterator.sptep; + + /* FIXME: properly handle invlpg on large guest pages */ + if (level == PT_PAGE_TABLE_LEVEL || + ((level == PT_DIRECTORY_LEVEL) && is_large_pte(*sptep))) { + struct kvm_mmu_page *sp = page_header(__pa(sptep)); + + pte_gpa = (sp->gfn << PAGE_SHIFT); + pte_gpa += (sptep - sp->spt) * sizeof(pt_element_t); + + if (is_shadow_present_pte(*sptep)) { + rmap_remove(vcpu->kvm, sptep); + if (is_large_pte(*sptep)) + --vcpu->kvm->stat.lpages; + need_flush = 1; + } + set_shadow_pte(sptep, shadow_trap_nonpresent_pte); + break; } - set_shadow_pte(sptep, shadow_trap_nonpresent_pte); - return 1; - } - if (!is_shadow_present_pte(*sptep)) - return 1; - return 0; -} -static void FNAME(invlpg)(struct kvm_vcpu *vcpu, gva_t gva) -{ - pt_element_t gpte; - struct shadow_walker walker = { - .walker = { .entry = FNAME(shadow_invlpg_entry), }, - .pte_gpa = -1, - }; + if (!is_shadow_present_pte(*sptep)) + break; + } - spin_lock(&vcpu->kvm->mmu_lock); - walk_shadow(&walker.walker, vcpu, gva); + if (need_flush) + kvm_flush_remote_tlbs(vcpu->kvm); spin_unlock(&vcpu->kvm->mmu_lock); - if (walker.pte_gpa == -1) + + if (pte_gpa == -1) return; - if (kvm_read_guest_atomic(vcpu->kvm, walker.pte_gpa, &gpte, + if (kvm_read_guest_atomic(vcpu->kvm, pte_gpa, &gpte, sizeof(pt_element_t))) return; if (is_present_pte(gpte) && (gpte & PT_ACCESSED_MASK)) { if (mmu_topup_memory_caches(vcpu)) return; - kvm_mmu_pte_write(vcpu, walker.pte_gpa, (const u8 *)&gpte, + kvm_mmu_pte_write(vcpu, pte_gpa, (const u8 *)&gpte, sizeof(pt_element_t), 0); } } @@ -540,7 +516,7 @@ static void FNAME(prefetch_page)(struct kvm_vcpu *vcpu, pt_element_t pt[256 / sizeof(pt_element_t)]; gpa_t pte_gpa; - if (sp->role.metaphysical + if (sp->role.direct || (PTTYPE == 32 && sp->role.level > PT_PAGE_TABLE_LEVEL)) { nonpaging_prefetch_page(vcpu, sp); return; @@ -619,7 +595,6 @@ static int FNAME(sync_page)(struct kvm_vcpu *vcpu, struct kvm_mmu_page *sp) #undef pt_element_t #undef guest_walker -#undef shadow_walker #undef FNAME #undef PT_BASE_ADDR_MASK #undef PT_INDEX diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 1452851ae25..1821c207819 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -38,9 +38,6 @@ MODULE_LICENSE("GPL"); #define IOPM_ALLOC_ORDER 2 #define MSRPM_ALLOC_ORDER 1 -#define DR7_GD_MASK (1 << 13) -#define DR6_BD_MASK (1 << 13) - #define SEG_TYPE_LDT 2 #define SEG_TYPE_BUSY_TSS16 3 @@ -50,6 +47,15 @@ MODULE_LICENSE("GPL"); #define DEBUGCTL_RESERVED_BITS (~(0x3fULL)) +/* Turn on to get debugging output*/ +/* #define NESTED_DEBUG */ + +#ifdef NESTED_DEBUG +#define nsvm_printk(fmt, args...) printk(KERN_INFO fmt, ## args) +#else +#define nsvm_printk(fmt, args...) do {} while(0) +#endif + /* enable NPT for AMD64 and X86 with PAE */ #if defined(CONFIG_X86_64) || defined(CONFIG_X86_PAE) static bool npt_enabled = true; @@ -60,14 +66,29 @@ static int npt = 1; module_param(npt, int, S_IRUGO); +static int nested = 0; +module_param(nested, int, S_IRUGO); + static void kvm_reput_irq(struct vcpu_svm *svm); static void svm_flush_tlb(struct kvm_vcpu *vcpu); +static int nested_svm_exit_handled(struct vcpu_svm *svm, bool kvm_override); +static int nested_svm_vmexit(struct vcpu_svm *svm); +static int nested_svm_vmsave(struct vcpu_svm *svm, void *nested_vmcb, + void *arg2, void *opaque); +static int nested_svm_check_exception(struct vcpu_svm *svm, unsigned nr, + bool has_error_code, u32 error_code); + static inline struct vcpu_svm *to_svm(struct kvm_vcpu *vcpu) { return container_of(vcpu, struct vcpu_svm, vcpu); } +static inline bool is_nested(struct vcpu_svm *svm) +{ + return svm->nested_vmcb; +} + static unsigned long iopm_base; struct kvm_ldttss_desc { @@ -157,32 +178,6 @@ static inline void kvm_write_cr2(unsigned long val) asm volatile ("mov %0, %%cr2" :: "r" (val)); } -static inline unsigned long read_dr6(void) -{ - unsigned long dr6; - - asm volatile ("mov %%dr6, %0" : "=r" (dr6)); - return dr6; -} - -static inline void write_dr6(unsigned long val) -{ - asm volatile ("mov %0, %%dr6" :: "r" (val)); -} - -static inline unsigned long read_dr7(void) -{ - unsigned long dr7; - - asm volatile ("mov %%dr7, %0" : "=r" (dr7)); - return dr7; -} - -static inline void write_dr7(unsigned long val) -{ - asm volatile ("mov %0, %%dr7" :: "r" (val)); -} - static inline void force_new_asid(struct kvm_vcpu *vcpu) { to_svm(vcpu)->asid_generation--; @@ -198,7 +193,7 @@ static void svm_set_efer(struct kvm_vcpu *vcpu, u64 efer) if (!npt_enabled && !(efer & EFER_LMA)) efer &= ~EFER_LME; - to_svm(vcpu)->vmcb->save.efer = efer | MSR_EFER_SVME_MASK; + to_svm(vcpu)->vmcb->save.efer = efer | EFER_SVME; vcpu->arch.shadow_efer = efer; } @@ -207,6 +202,11 @@ static void svm_queue_exception(struct kvm_vcpu *vcpu, unsigned nr, { struct vcpu_svm *svm = to_svm(vcpu); + /* If we are within a nested VM we'd better #VMEXIT and let the + guest handle the exception */ + if (nested_svm_check_exception(svm, nr, has_error_code, error_code)) + return; + svm->vmcb->control.event_inj = nr | SVM_EVTINJ_VALID | (has_error_code ? SVM_EVTINJ_VALID_ERR : 0) @@ -242,7 +242,7 @@ static void skip_emulated_instruction(struct kvm_vcpu *vcpu) kvm_rip_write(vcpu, svm->next_rip); svm->vmcb->control.int_state &= ~SVM_INTERRUPT_SHADOW_MASK; - vcpu->arch.interrupt_window_open = 1; + vcpu->arch.interrupt_window_open = (svm->vcpu.arch.hflags & HF_GIF_MASK); } static int has_svm(void) @@ -250,7 +250,7 @@ static int has_svm(void) const char *msg; if (!cpu_has_svm(&msg)) { - printk(KERN_INFO "has_svn: %s\n", msg); + printk(KERN_INFO "has_svm: %s\n", msg); return 0; } @@ -292,7 +292,7 @@ static void svm_hardware_enable(void *garbage) svm_data->tss_desc = (struct kvm_ldttss_desc *)(gdt + GDT_ENTRY_TSS); rdmsrl(MSR_EFER, efer); - wrmsrl(MSR_EFER, efer | MSR_EFER_SVME_MASK); + wrmsrl(MSR_EFER, efer | EFER_SVME); wrmsrl(MSR_VM_HSAVE_PA, page_to_pfn(svm_data->save_area) << PAGE_SHIFT); @@ -417,6 +417,14 @@ static __init int svm_hardware_setup(void) if (boot_cpu_has(X86_FEATURE_NX)) kvm_enable_efer_bits(EFER_NX); + if (boot_cpu_has(X86_FEATURE_FXSR_OPT)) + kvm_enable_efer_bits(EFER_FFXSR); + + if (nested) { + printk(KERN_INFO "kvm: Nested Virtualization enabled\n"); + kvm_enable_efer_bits(EFER_SVME); + } + for_each_online_cpu(cpu) { r = svm_cpu_init(cpu); if (r) @@ -559,7 +567,7 @@ static void init_vmcb(struct vcpu_svm *svm) init_sys_seg(&save->ldtr, SEG_TYPE_LDT); init_sys_seg(&save->tr, SEG_TYPE_BUSY_TSS16); - save->efer = MSR_EFER_SVME_MASK; + save->efer = EFER_SVME; save->dr6 = 0xffff0ff0; save->dr7 = 0x400; save->rflags = 2; @@ -591,6 +599,9 @@ static void init_vmcb(struct vcpu_svm *svm) save->cr4 = 0; } force_new_asid(&svm->vcpu); + + svm->nested_vmcb = 0; + svm->vcpu.arch.hflags = HF_GIF_MASK; } static int svm_vcpu_reset(struct kvm_vcpu *vcpu) @@ -615,6 +626,8 @@ static struct kvm_vcpu *svm_create_vcpu(struct kvm *kvm, unsigned int id) struct vcpu_svm *svm; struct page *page; struct page *msrpm_pages; + struct page *hsave_page; + struct page *nested_msrpm_pages; int err; svm = kmem_cache_zalloc(kvm_vcpu_cache, GFP_KERNEL); @@ -637,14 +650,25 @@ static struct kvm_vcpu *svm_create_vcpu(struct kvm *kvm, unsigned int id) msrpm_pages = alloc_pages(GFP_KERNEL, MSRPM_ALLOC_ORDER); if (!msrpm_pages) goto uninit; + + nested_msrpm_pages = alloc_pages(GFP_KERNEL, MSRPM_ALLOC_ORDER); + if (!nested_msrpm_pages) + goto uninit; + svm->msrpm = page_address(msrpm_pages); svm_vcpu_init_msrpm(svm->msrpm); + hsave_page = alloc_page(GFP_KERNEL); + if (!hsave_page) + goto uninit; + svm->hsave = page_address(hsave_page); + + svm->nested_msrpm = page_address(nested_msrpm_pages); + svm->vmcb = page_address(page); clear_page(svm->vmcb); svm->vmcb_pa = page_to_pfn(page) << PAGE_SHIFT; svm->asid_generation = 0; - memset(svm->db_regs, 0, sizeof(svm->db_regs)); init_vmcb(svm); fx_init(&svm->vcpu); @@ -669,6 +693,8 @@ static void svm_free_vcpu(struct kvm_vcpu *vcpu) __free_page(pfn_to_page(svm->vmcb_pa >> PAGE_SHIFT)); __free_pages(virt_to_page(svm->msrpm), MSRPM_ALLOC_ORDER); + __free_page(virt_to_page(svm->hsave)); + __free_pages(virt_to_page(svm->nested_msrpm), MSRPM_ALLOC_ORDER); kvm_vcpu_uninit(vcpu); kmem_cache_free(kvm_vcpu_cache, svm); } @@ -718,6 +744,16 @@ static void svm_set_rflags(struct kvm_vcpu *vcpu, unsigned long rflags) to_svm(vcpu)->vmcb->save.rflags = rflags; } +static void svm_set_vintr(struct vcpu_svm *svm) +{ + svm->vmcb->control.intercept |= 1ULL << INTERCEPT_VINTR; +} + +static void svm_clear_vintr(struct vcpu_svm *svm) +{ + svm->vmcb->control.intercept &= ~(1ULL << INTERCEPT_VINTR); +} + static struct vmcb_seg *svm_seg(struct kvm_vcpu *vcpu, int seg) { struct vmcb_save_area *save = &to_svm(vcpu)->vmcb->save; @@ -760,20 +796,37 @@ static void svm_get_segment(struct kvm_vcpu *vcpu, var->db = (s->attrib >> SVM_SELECTOR_DB_SHIFT) & 1; var->g = (s->attrib >> SVM_SELECTOR_G_SHIFT) & 1; - /* - * SVM always stores 0 for the 'G' bit in the CS selector in - * the VMCB on a VMEXIT. This hurts cross-vendor migration: - * Intel's VMENTRY has a check on the 'G' bit. - */ - if (seg == VCPU_SREG_CS) + switch (seg) { + case VCPU_SREG_CS: + /* + * SVM always stores 0 for the 'G' bit in the CS selector in + * the VMCB on a VMEXIT. This hurts cross-vendor migration: + * Intel's VMENTRY has a check on the 'G' bit. + */ var->g = s->limit > 0xfffff; - - /* - * Work around a bug where the busy flag in the tr selector - * isn't exposed - */ - if (seg == VCPU_SREG_TR) + break; + case VCPU_SREG_TR: + /* + * Work around a bug where the busy flag in the tr selector + * isn't exposed + */ var->type |= 0x2; + break; + case VCPU_SREG_DS: + case VCPU_SREG_ES: + case VCPU_SREG_FS: + case VCPU_SREG_GS: + /* + * The accessed bit must always be set in the segment + * descriptor cache, although it can be cleared in the + * descriptor, the cached bit always remains at 1. Since + * Intel has a check on this, set it here to support + * cross-vendor migration. + */ + if (!var->unusable) + var->type |= 0x1; + break; + } var->unusable = !var->present; } @@ -905,9 +958,37 @@ static void svm_set_segment(struct kvm_vcpu *vcpu, } -static int svm_guest_debug(struct kvm_vcpu *vcpu, struct kvm_debug_guest *dbg) +static int svm_guest_debug(struct kvm_vcpu *vcpu, struct kvm_guest_debug *dbg) { - return -EOPNOTSUPP; + int old_debug = vcpu->guest_debug; + struct vcpu_svm *svm = to_svm(vcpu); + + vcpu->guest_debug = dbg->control; + + svm->vmcb->control.intercept_exceptions &= + ~((1 << DB_VECTOR) | (1 << BP_VECTOR)); + if (vcpu->guest_debug & KVM_GUESTDBG_ENABLE) { + if (vcpu->guest_debug & + (KVM_GUESTDBG_SINGLESTEP | KVM_GUESTDBG_USE_HW_BP)) + svm->vmcb->control.intercept_exceptions |= + 1 << DB_VECTOR; + if (vcpu->guest_debug & KVM_GUESTDBG_USE_SW_BP) + svm->vmcb->control.intercept_exceptions |= + 1 << BP_VECTOR; + } else + vcpu->guest_debug = 0; + + if (vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP) + svm->vmcb->save.dr7 = dbg->arch.debugreg[7]; + else + svm->vmcb->save.dr7 = vcpu->arch.dr7; + + if (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP) + svm->vmcb->save.rflags |= X86_EFLAGS_TF | X86_EFLAGS_RF; + else if (old_debug & KVM_GUESTDBG_SINGLESTEP) + svm->vmcb->save.rflags &= ~(X86_EFLAGS_TF | X86_EFLAGS_RF); + + return 0; } static int svm_get_irq(struct kvm_vcpu *vcpu) @@ -949,7 +1030,29 @@ static void new_asid(struct vcpu_svm *svm, struct svm_cpu_data *svm_data) static unsigned long svm_get_dr(struct kvm_vcpu *vcpu, int dr) { - unsigned long val = to_svm(vcpu)->db_regs[dr]; + struct vcpu_svm *svm = to_svm(vcpu); + unsigned long val; + + switch (dr) { + case 0 ... 3: + val = vcpu->arch.db[dr]; + break; + case 6: + if (vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP) + val = vcpu->arch.dr6; + else + val = svm->vmcb->save.dr6; + break; + case 7: + if (vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP) + val = vcpu->arch.dr7; + else + val = svm->vmcb->save.dr7; + break; + default: + val = 0; + } + KVMTRACE_2D(DR_READ, vcpu, (u32)dr, (u32)val, handler); return val; } @@ -959,33 +1062,40 @@ static void svm_set_dr(struct kvm_vcpu *vcpu, int dr, unsigned long value, { struct vcpu_svm *svm = to_svm(vcpu); - *exception = 0; + KVMTRACE_2D(DR_WRITE, vcpu, (u32)dr, (u32)value, handler); - if (svm->vmcb->save.dr7 & DR7_GD_MASK) { - svm->vmcb->save.dr7 &= ~DR7_GD_MASK; - svm->vmcb->save.dr6 |= DR6_BD_MASK; - *exception = DB_VECTOR; - return; - } + *exception = 0; switch (dr) { case 0 ... 3: - svm->db_regs[dr] = value; + vcpu->arch.db[dr] = value; + if (!(vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP)) + vcpu->arch.eff_db[dr] = value; return; case 4 ... 5: - if (vcpu->arch.cr4 & X86_CR4_DE) { + if (vcpu->arch.cr4 & X86_CR4_DE) *exception = UD_VECTOR; + return; + case 6: + if (value & 0xffffffff00000000ULL) { + *exception = GP_VECTOR; return; } - case 7: { - if (value & ~((1ULL << 32) - 1)) { + vcpu->arch.dr6 = (value & DR6_VOLATILE) | DR6_FIXED_1; + return; + case 7: + if (value & 0xffffffff00000000ULL) { *exception = GP_VECTOR; return; } - svm->vmcb->save.dr7 = value; + vcpu->arch.dr7 = (value & DR7_VOLATILE) | DR7_FIXED_1; + if (!(vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP)) { + svm->vmcb->save.dr7 = vcpu->arch.dr7; + vcpu->arch.switch_db_regs = (value & DR7_BP_EN_MASK); + } return; - } default: + /* FIXME: Possible case? */ printk(KERN_DEBUG "%s: unexpected dr %u\n", __func__, dr); *exception = UD_VECTOR; @@ -1031,6 +1141,27 @@ static int pf_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run) return kvm_mmu_page_fault(&svm->vcpu, fault_address, error_code); } +static int db_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run) +{ + if (!(svm->vcpu.guest_debug & + (KVM_GUESTDBG_SINGLESTEP | KVM_GUESTDBG_USE_HW_BP))) { + kvm_queue_exception(&svm->vcpu, DB_VECTOR); + return 1; + } + kvm_run->exit_reason = KVM_EXIT_DEBUG; + kvm_run->debug.arch.pc = svm->vmcb->save.cs.base + svm->vmcb->save.rip; + kvm_run->debug.arch.exception = DB_VECTOR; + return 0; +} + +static int bp_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run) +{ + kvm_run->exit_reason = KVM_EXIT_DEBUG; + kvm_run->debug.arch.pc = svm->vmcb->save.cs.base + svm->vmcb->save.rip; + kvm_run->debug.arch.exception = BP_VECTOR; + return 0; +} + static int ud_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run) { int er; @@ -1080,7 +1211,7 @@ static int shutdown_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run) static int io_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run) { u32 io_info = svm->vmcb->control.exit_info_1; /* address size bug? */ - int size, down, in, string, rep; + int size, in, string; unsigned port; ++svm->vcpu.stat.io_exits; @@ -1099,8 +1230,6 @@ static int io_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run) in = (io_info & SVM_IOIO_TYPE_MASK) != 0; port = io_info >> 16; size = (io_info & SVM_IOIO_SIZE_MASK) >> SVM_IOIO_SIZE_SHIFT; - rep = (io_info & SVM_IOIO_REP_MASK) != 0; - down = (svm->vmcb->save.rflags & X86_EFLAGS_DF) != 0; skip_emulated_instruction(&svm->vcpu); return kvm_emulate_pio(&svm->vcpu, kvm_run, in, size, port); @@ -1139,6 +1268,567 @@ static int vmmcall_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run) return 1; } +static int nested_svm_check_permissions(struct vcpu_svm *svm) +{ + if (!(svm->vcpu.arch.shadow_efer & EFER_SVME) + || !is_paging(&svm->vcpu)) { + kvm_queue_exception(&svm->vcpu, UD_VECTOR); + return 1; + } + + if (svm->vmcb->save.cpl) { + kvm_inject_gp(&svm->vcpu, 0); + return 1; + } + + return 0; +} + +static int nested_svm_check_exception(struct vcpu_svm *svm, unsigned nr, + bool has_error_code, u32 error_code) +{ + if (is_nested(svm)) { + svm->vmcb->control.exit_code = SVM_EXIT_EXCP_BASE + nr; + svm->vmcb->control.exit_code_hi = 0; + svm->vmcb->control.exit_info_1 = error_code; + svm->vmcb->control.exit_info_2 = svm->vcpu.arch.cr2; + if (nested_svm_exit_handled(svm, false)) { + nsvm_printk("VMexit -> EXCP 0x%x\n", nr); + + nested_svm_vmexit(svm); + return 1; + } + } + + return 0; +} + +static inline int nested_svm_intr(struct vcpu_svm *svm) +{ + if (is_nested(svm)) { + if (!(svm->vcpu.arch.hflags & HF_VINTR_MASK)) + return 0; + + if (!(svm->vcpu.arch.hflags & HF_HIF_MASK)) + return 0; + + svm->vmcb->control.exit_code = SVM_EXIT_INTR; + + if (nested_svm_exit_handled(svm, false)) { + nsvm_printk("VMexit -> INTR\n"); + nested_svm_vmexit(svm); + return 1; + } + } + + return 0; +} + +static struct page *nested_svm_get_page(struct vcpu_svm *svm, u64 gpa) +{ + struct page *page; + + down_read(¤t->mm->mmap_sem); + page = gfn_to_page(svm->vcpu.kvm, gpa >> PAGE_SHIFT); + up_read(¤t->mm->mmap_sem); + + if (is_error_page(page)) { + printk(KERN_INFO "%s: could not find page at 0x%llx\n", + __func__, gpa); + kvm_release_page_clean(page); + kvm_inject_gp(&svm->vcpu, 0); + return NULL; + } + return page; +} + +static int nested_svm_do(struct vcpu_svm *svm, + u64 arg1_gpa, u64 arg2_gpa, void *opaque, + int (*handler)(struct vcpu_svm *svm, + void *arg1, + void *arg2, + void *opaque)) +{ + struct page *arg1_page; + struct page *arg2_page = NULL; + void *arg1; + void *arg2 = NULL; + int retval; + + arg1_page = nested_svm_get_page(svm, arg1_gpa); + if(arg1_page == NULL) + return 1; + + if (arg2_gpa) { + arg2_page = nested_svm_get_page(svm, arg2_gpa); + if(arg2_page == NULL) { + kvm_release_page_clean(arg1_page); + return 1; + } + } + + arg1 = kmap_atomic(arg1_page, KM_USER0); + if (arg2_gpa) + arg2 = kmap_atomic(arg2_page, KM_USER1); + + retval = handler(svm, arg1, arg2, opaque); + + kunmap_atomic(arg1, KM_USER0); + if (arg2_gpa) + kunmap_atomic(arg2, KM_USER1); + + kvm_release_page_dirty(arg1_page); + if (arg2_gpa) + kvm_release_page_dirty(arg2_page); + + return retval; +} + +static int nested_svm_exit_handled_real(struct vcpu_svm *svm, + void *arg1, + void *arg2, + void *opaque) +{ + struct vmcb *nested_vmcb = (struct vmcb *)arg1; + bool kvm_overrides = *(bool *)opaque; + u32 exit_code = svm->vmcb->control.exit_code; + + if (kvm_overrides) { + switch (exit_code) { + case SVM_EXIT_INTR: + case SVM_EXIT_NMI: + return 0; + /* For now we are always handling NPFs when using them */ + case SVM_EXIT_NPF: + if (npt_enabled) + return 0; + break; + /* When we're shadowing, trap PFs */ + case SVM_EXIT_EXCP_BASE + PF_VECTOR: + if (!npt_enabled) + return 0; + break; + default: + break; + } + } + + switch (exit_code) { + case SVM_EXIT_READ_CR0 ... SVM_EXIT_READ_CR8: { + u32 cr_bits = 1 << (exit_code - SVM_EXIT_READ_CR0); + if (nested_vmcb->control.intercept_cr_read & cr_bits) + return 1; + break; + } + case SVM_EXIT_WRITE_CR0 ... SVM_EXIT_WRITE_CR8: { + u32 cr_bits = 1 << (exit_code - SVM_EXIT_WRITE_CR0); + if (nested_vmcb->control.intercept_cr_write & cr_bits) + return 1; + break; + } + case SVM_EXIT_READ_DR0 ... SVM_EXIT_READ_DR7: { + u32 dr_bits = 1 << (exit_code - SVM_EXIT_READ_DR0); + if (nested_vmcb->control.intercept_dr_read & dr_bits) + return 1; + break; + } + case SVM_EXIT_WRITE_DR0 ... SVM_EXIT_WRITE_DR7: { + u32 dr_bits = 1 << (exit_code - SVM_EXIT_WRITE_DR0); + if (nested_vmcb->control.intercept_dr_write & dr_bits) + return 1; + break; + } + case SVM_EXIT_EXCP_BASE ... SVM_EXIT_EXCP_BASE + 0x1f: { + u32 excp_bits = 1 << (exit_code - SVM_EXIT_EXCP_BASE); + if (nested_vmcb->control.intercept_exceptions & excp_bits) + return 1; + break; + } + default: { + u64 exit_bits = 1ULL << (exit_code - SVM_EXIT_INTR); + nsvm_printk("exit code: 0x%x\n", exit_code); + if (nested_vmcb->control.intercept & exit_bits) + return 1; + } + } + + return 0; +} + +static int nested_svm_exit_handled_msr(struct vcpu_svm *svm, + void *arg1, void *arg2, + void *opaque) +{ + struct vmcb *nested_vmcb = (struct vmcb *)arg1; + u8 *msrpm = (u8 *)arg2; + u32 t0, t1; + u32 msr = svm->vcpu.arch.regs[VCPU_REGS_RCX]; + u32 param = svm->vmcb->control.exit_info_1 & 1; + + if (!(nested_vmcb->control.intercept & (1ULL << INTERCEPT_MSR_PROT))) + return 0; + + switch(msr) { + case 0 ... 0x1fff: + t0 = (msr * 2) % 8; + t1 = msr / 8; + break; + case 0xc0000000 ... 0xc0001fff: + t0 = (8192 + msr - 0xc0000000) * 2; + t1 = (t0 / 8); + t0 %= 8; + break; + case 0xc0010000 ... 0xc0011fff: + t0 = (16384 + msr - 0xc0010000) * 2; + t1 = (t0 / 8); + t0 %= 8; + break; + default: + return 1; + break; + } + if (msrpm[t1] & ((1 << param) << t0)) + return 1; + + return 0; +} + +static int nested_svm_exit_handled(struct vcpu_svm *svm, bool kvm_override) +{ + bool k = kvm_override; + + switch (svm->vmcb->control.exit_code) { + case SVM_EXIT_MSR: + return nested_svm_do(svm, svm->nested_vmcb, + svm->nested_vmcb_msrpm, NULL, + nested_svm_exit_handled_msr); + default: break; + } + + return nested_svm_do(svm, svm->nested_vmcb, 0, &k, + nested_svm_exit_handled_real); +} + +static int nested_svm_vmexit_real(struct vcpu_svm *svm, void *arg1, + void *arg2, void *opaque) +{ + struct vmcb *nested_vmcb = (struct vmcb *)arg1; + struct vmcb *hsave = svm->hsave; + u64 nested_save[] = { nested_vmcb->save.cr0, + nested_vmcb->save.cr3, + nested_vmcb->save.cr4, + nested_vmcb->save.efer, + nested_vmcb->control.intercept_cr_read, + nested_vmcb->control.intercept_cr_write, + nested_vmcb->control.intercept_dr_read, + nested_vmcb->control.intercept_dr_write, + nested_vmcb->control.intercept_exceptions, + nested_vmcb->control.intercept, + nested_vmcb->control.msrpm_base_pa, + nested_vmcb->control.iopm_base_pa, + nested_vmcb->control.tsc_offset }; + + /* Give the current vmcb to the guest */ + memcpy(nested_vmcb, svm->vmcb, sizeof(struct vmcb)); + nested_vmcb->save.cr0 = nested_save[0]; + if (!npt_enabled) + nested_vmcb->save.cr3 = nested_save[1]; + nested_vmcb->save.cr4 = nested_save[2]; + nested_vmcb->save.efer = nested_save[3]; + nested_vmcb->control.intercept_cr_read = nested_save[4]; + nested_vmcb->control.intercept_cr_write = nested_save[5]; + nested_vmcb->control.intercept_dr_read = nested_save[6]; + nested_vmcb->control.intercept_dr_write = nested_save[7]; + nested_vmcb->control.intercept_exceptions = nested_save[8]; + nested_vmcb->control.intercept = nested_save[9]; + nested_vmcb->control.msrpm_base_pa = nested_save[10]; + nested_vmcb->control.iopm_base_pa = nested_save[11]; + nested_vmcb->control.tsc_offset = nested_save[12]; + + /* We always set V_INTR_MASKING and remember the old value in hflags */ + if (!(svm->vcpu.arch.hflags & HF_VINTR_MASK)) + nested_vmcb->control.int_ctl &= ~V_INTR_MASKING_MASK; + + if ((nested_vmcb->control.int_ctl & V_IRQ_MASK) && + (nested_vmcb->control.int_vector)) { + nsvm_printk("WARNING: IRQ 0x%x still enabled on #VMEXIT\n", + nested_vmcb->control.int_vector); + } + + /* Restore the original control entries */ + svm->vmcb->control = hsave->control; + + /* Kill any pending exceptions */ + if (svm->vcpu.arch.exception.pending == true) + nsvm_printk("WARNING: Pending Exception\n"); + svm->vcpu.arch.exception.pending = false; + + /* Restore selected save entries */ + svm->vmcb->save.es = hsave->save.es; + svm->vmcb->save.cs = hsave->save.cs; + svm->vmcb->save.ss = hsave->save.ss; + 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; + 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); + if (npt_enabled) { + svm->vmcb->save.cr3 = hsave->save.cr3; + svm->vcpu.arch.cr3 = hsave->save.cr3; + } else { + kvm_set_cr3(&svm->vcpu, hsave->save.cr3); + } + kvm_register_write(&svm->vcpu, VCPU_REGS_RAX, hsave->save.rax); + kvm_register_write(&svm->vcpu, VCPU_REGS_RSP, hsave->save.rsp); + kvm_register_write(&svm->vcpu, VCPU_REGS_RIP, hsave->save.rip); + svm->vmcb->save.dr7 = 0; + svm->vmcb->save.cpl = 0; + svm->vmcb->control.exit_int_info = 0; + + svm->vcpu.arch.hflags &= ~HF_GIF_MASK; + /* Exit nested SVM mode */ + svm->nested_vmcb = 0; + + return 0; +} + +static int nested_svm_vmexit(struct vcpu_svm *svm) +{ + nsvm_printk("VMexit\n"); + if (nested_svm_do(svm, svm->nested_vmcb, 0, + NULL, nested_svm_vmexit_real)) + return 1; + + kvm_mmu_reset_context(&svm->vcpu); + kvm_mmu_load(&svm->vcpu); + + return 0; +} + +static int nested_svm_vmrun_msrpm(struct vcpu_svm *svm, void *arg1, + void *arg2, void *opaque) +{ + int i; + u32 *nested_msrpm = (u32*)arg1; + for (i=0; i< PAGE_SIZE * (1 << MSRPM_ALLOC_ORDER) / 4; i++) + svm->nested_msrpm[i] = svm->msrpm[i] | nested_msrpm[i]; + svm->vmcb->control.msrpm_base_pa = __pa(svm->nested_msrpm); + + return 0; +} + +static int nested_svm_vmrun(struct vcpu_svm *svm, void *arg1, + void *arg2, void *opaque) +{ + struct vmcb *nested_vmcb = (struct vmcb *)arg1; + struct vmcb *hsave = svm->hsave; + + /* nested_vmcb is our indicator if nested SVM is activated */ + svm->nested_vmcb = svm->vmcb->save.rax; + + /* Clear internal status */ + svm->vcpu.arch.exception.pending = false; + + /* Save the old vmcb, so we don't need to pick what we save, but + can restore everything when a VMEXIT occurs */ + memcpy(hsave, svm->vmcb, sizeof(struct vmcb)); + /* We need to remember the original CR3 in the SPT case */ + if (!npt_enabled) + hsave->save.cr3 = svm->vcpu.arch.cr3; + hsave->save.cr4 = svm->vcpu.arch.cr4; + hsave->save.rip = svm->next_rip; + + if (svm->vmcb->save.rflags & X86_EFLAGS_IF) + svm->vcpu.arch.hflags |= HF_HIF_MASK; + else + svm->vcpu.arch.hflags &= ~HF_HIF_MASK; + + /* Load the nested guest state */ + svm->vmcb->save.es = nested_vmcb->save.es; + svm->vmcb->save.cs = nested_vmcb->save.cs; + svm->vmcb->save.ss = nested_vmcb->save.ss; + 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; + 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); + if (npt_enabled) { + svm->vmcb->save.cr3 = nested_vmcb->save.cr3; + svm->vcpu.arch.cr3 = nested_vmcb->save.cr3; + } else { + kvm_set_cr3(&svm->vcpu, nested_vmcb->save.cr3); + kvm_mmu_reset_context(&svm->vcpu); + } + svm->vmcb->save.cr2 = nested_vmcb->save.cr2; + kvm_register_write(&svm->vcpu, VCPU_REGS_RAX, nested_vmcb->save.rax); + kvm_register_write(&svm->vcpu, VCPU_REGS_RSP, nested_vmcb->save.rsp); + kvm_register_write(&svm->vcpu, VCPU_REGS_RIP, nested_vmcb->save.rip); + /* In case we don't even reach vcpu_run, the fields are not updated */ + svm->vmcb->save.rax = nested_vmcb->save.rax; + svm->vmcb->save.rsp = nested_vmcb->save.rsp; + svm->vmcb->save.rip = nested_vmcb->save.rip; + svm->vmcb->save.dr7 = nested_vmcb->save.dr7; + svm->vmcb->save.dr6 = nested_vmcb->save.dr6; + svm->vmcb->save.cpl = nested_vmcb->save.cpl; + + /* We don't want a nested guest to be more powerful than the guest, + so all intercepts are ORed */ + svm->vmcb->control.intercept_cr_read |= + nested_vmcb->control.intercept_cr_read; + svm->vmcb->control.intercept_cr_write |= + nested_vmcb->control.intercept_cr_write; + svm->vmcb->control.intercept_dr_read |= + nested_vmcb->control.intercept_dr_read; + svm->vmcb->control.intercept_dr_write |= + nested_vmcb->control.intercept_dr_write; + svm->vmcb->control.intercept_exceptions |= + nested_vmcb->control.intercept_exceptions; + + svm->vmcb->control.intercept |= nested_vmcb->control.intercept; + + svm->nested_vmcb_msrpm = nested_vmcb->control.msrpm_base_pa; + + force_new_asid(&svm->vcpu); + svm->vmcb->control.exit_int_info = nested_vmcb->control.exit_int_info; + svm->vmcb->control.exit_int_info_err = nested_vmcb->control.exit_int_info_err; + svm->vmcb->control.int_ctl = nested_vmcb->control.int_ctl | V_INTR_MASKING_MASK; + if (nested_vmcb->control.int_ctl & V_IRQ_MASK) { + nsvm_printk("nSVM Injecting Interrupt: 0x%x\n", + nested_vmcb->control.int_ctl); + } + if (nested_vmcb->control.int_ctl & V_INTR_MASKING_MASK) + svm->vcpu.arch.hflags |= HF_VINTR_MASK; + else + svm->vcpu.arch.hflags &= ~HF_VINTR_MASK; + + nsvm_printk("nSVM exit_int_info: 0x%x | int_state: 0x%x\n", + nested_vmcb->control.exit_int_info, + nested_vmcb->control.int_state); + + svm->vmcb->control.int_vector = nested_vmcb->control.int_vector; + svm->vmcb->control.int_state = nested_vmcb->control.int_state; + svm->vmcb->control.tsc_offset += nested_vmcb->control.tsc_offset; + if (nested_vmcb->control.event_inj & SVM_EVTINJ_VALID) + nsvm_printk("Injecting Event: 0x%x\n", + nested_vmcb->control.event_inj); + svm->vmcb->control.event_inj = nested_vmcb->control.event_inj; + svm->vmcb->control.event_inj_err = nested_vmcb->control.event_inj_err; + + svm->vcpu.arch.hflags |= HF_GIF_MASK; + + return 0; +} + +static int nested_svm_vmloadsave(struct vmcb *from_vmcb, struct vmcb *to_vmcb) +{ + to_vmcb->save.fs = from_vmcb->save.fs; + to_vmcb->save.gs = from_vmcb->save.gs; + to_vmcb->save.tr = from_vmcb->save.tr; + to_vmcb->save.ldtr = from_vmcb->save.ldtr; + to_vmcb->save.kernel_gs_base = from_vmcb->save.kernel_gs_base; + to_vmcb->save.star = from_vmcb->save.star; + to_vmcb->save.lstar = from_vmcb->save.lstar; + to_vmcb->save.cstar = from_vmcb->save.cstar; + to_vmcb->save.sfmask = from_vmcb->save.sfmask; + to_vmcb->save.sysenter_cs = from_vmcb->save.sysenter_cs; + to_vmcb->save.sysenter_esp = from_vmcb->save.sysenter_esp; + to_vmcb->save.sysenter_eip = from_vmcb->save.sysenter_eip; + + return 1; +} + +static int nested_svm_vmload(struct vcpu_svm *svm, void *nested_vmcb, + void *arg2, void *opaque) +{ + return nested_svm_vmloadsave((struct vmcb *)nested_vmcb, svm->vmcb); +} + +static int nested_svm_vmsave(struct vcpu_svm *svm, void *nested_vmcb, + void *arg2, void *opaque) +{ + return nested_svm_vmloadsave(svm->vmcb, (struct vmcb *)nested_vmcb); +} + +static int vmload_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run) +{ + if (nested_svm_check_permissions(svm)) + return 1; + + svm->next_rip = kvm_rip_read(&svm->vcpu) + 3; + skip_emulated_instruction(&svm->vcpu); + + nested_svm_do(svm, svm->vmcb->save.rax, 0, NULL, nested_svm_vmload); + + return 1; +} + +static int vmsave_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run) +{ + if (nested_svm_check_permissions(svm)) + return 1; + + svm->next_rip = kvm_rip_read(&svm->vcpu) + 3; + skip_emulated_instruction(&svm->vcpu); + + nested_svm_do(svm, svm->vmcb->save.rax, 0, NULL, nested_svm_vmsave); + + return 1; +} + +static int vmrun_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run) +{ + nsvm_printk("VMrun\n"); + if (nested_svm_check_permissions(svm)) + return 1; + + svm->next_rip = kvm_rip_read(&svm->vcpu) + 3; + skip_emulated_instruction(&svm->vcpu); + + if (nested_svm_do(svm, svm->vmcb->save.rax, 0, + NULL, nested_svm_vmrun)) + return 1; + + if (nested_svm_do(svm, svm->nested_vmcb_msrpm, 0, + NULL, nested_svm_vmrun_msrpm)) + return 1; + + return 1; +} + +static int stgi_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run) +{ + if (nested_svm_check_permissions(svm)) + return 1; + + svm->next_rip = kvm_rip_read(&svm->vcpu) + 3; + skip_emulated_instruction(&svm->vcpu); + + svm->vcpu.arch.hflags |= HF_GIF_MASK; + + return 1; +} + +static int clgi_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run) +{ + if (nested_svm_check_permissions(svm)) + return 1; + + svm->next_rip = kvm_rip_read(&svm->vcpu) + 3; + skip_emulated_instruction(&svm->vcpu); + + svm->vcpu.arch.hflags &= ~HF_GIF_MASK; + + /* After a CLGI no interrupts should come */ + svm_clear_vintr(svm); + svm->vmcb->control.int_ctl &= ~V_IRQ_MASK; + + return 1; +} + static int invalid_op_interception(struct vcpu_svm *svm, struct kvm_run *kvm_run) { @@ -1250,6 +1940,15 @@ static int svm_get_msr(struct kvm_vcpu *vcpu, unsigned ecx, u64 *data) case MSR_IA32_LASTINTTOIP: *data = svm->vmcb->save.last_excp_to; break; + case MSR_VM_HSAVE_PA: + *data = svm->hsave_msr; + break; + case MSR_VM_CR: + *data = 0; + break; + case MSR_IA32_UCODE_REV: + *data = 0x01000065; + break; default: return kvm_get_msr_common(vcpu, ecx, data); } @@ -1344,6 +2043,9 @@ static int svm_set_msr(struct kvm_vcpu *vcpu, unsigned ecx, u64 data) pr_unimpl(vcpu, "unimplemented perfctr wrmsr: 0x%x data 0x%llx\n", ecx, data); break; + case MSR_VM_HSAVE_PA: + svm->hsave_msr = data; + break; default: return kvm_set_msr_common(vcpu, ecx, data); } @@ -1380,7 +2082,7 @@ static int interrupt_window_interception(struct vcpu_svm *svm, { KVMTRACE_0D(PEND_INTR, &svm->vcpu, handler); - svm->vmcb->control.intercept &= ~(1ULL << INTERCEPT_VINTR); + svm_clear_vintr(svm); svm->vmcb->control.int_ctl &= ~V_IRQ_MASK; /* * If the user space waits to inject interrupts, exit as soon as @@ -1417,6 +2119,8 @@ static int (*svm_exit_handlers[])(struct vcpu_svm *svm, [SVM_EXIT_WRITE_DR3] = emulate_on_interception, [SVM_EXIT_WRITE_DR5] = emulate_on_interception, [SVM_EXIT_WRITE_DR7] = emulate_on_interception, + [SVM_EXIT_EXCP_BASE + DB_VECTOR] = db_interception, + [SVM_EXIT_EXCP_BASE + BP_VECTOR] = bp_interception, [SVM_EXIT_EXCP_BASE + UD_VECTOR] = ud_interception, [SVM_EXIT_EXCP_BASE + PF_VECTOR] = pf_interception, [SVM_EXIT_EXCP_BASE + NM_VECTOR] = nm_interception, @@ -1436,12 +2140,12 @@ static int (*svm_exit_handlers[])(struct vcpu_svm *svm, [SVM_EXIT_MSR] = msr_interception, [SVM_EXIT_TASK_SWITCH] = task_switch_interception, [SVM_EXIT_SHUTDOWN] = shutdown_interception, - [SVM_EXIT_VMRUN] = invalid_op_interception, + [SVM_EXIT_VMRUN] = vmrun_interception, [SVM_EXIT_VMMCALL] = vmmcall_interception, - [SVM_EXIT_VMLOAD] = invalid_op_interception, - [SVM_EXIT_VMSAVE] = invalid_op_interception, - [SVM_EXIT_STGI] = invalid_op_interception, - [SVM_EXIT_CLGI] = invalid_op_interception, + [SVM_EXIT_VMLOAD] = vmload_interception, + [SVM_EXIT_VMSAVE] = vmsave_interception, + [SVM_EXIT_STGI] = stgi_interception, + [SVM_EXIT_CLGI] = clgi_interception, [SVM_EXIT_SKINIT] = invalid_op_interception, [SVM_EXIT_WBINVD] = emulate_on_interception, [SVM_EXIT_MONITOR] = invalid_op_interception, @@ -1457,6 +2161,17 @@ static int handle_exit(struct kvm_run *kvm_run, struct kvm_vcpu *vcpu) KVMTRACE_3D(VMEXIT, vcpu, exit_code, (u32)svm->vmcb->save.rip, (u32)((u64)svm->vmcb->save.rip >> 32), entryexit); + if (is_nested(svm)) { + nsvm_printk("nested handle_exit: 0x%x | 0x%lx | 0x%lx | 0x%lx\n", + exit_code, svm->vmcb->control.exit_info_1, + svm->vmcb->control.exit_info_2, svm->vmcb->save.rip); + if (nested_svm_exit_handled(svm, true)) { + nested_svm_vmexit(svm); + nsvm_printk("-> #VMEXIT\n"); + return 1; + } + } + if (npt_enabled) { int mmu_reload = 0; if ((vcpu->arch.cr0 ^ svm->vmcb->save.cr0) & X86_CR0_PG) { @@ -1544,6 +2259,8 @@ static void svm_set_irq(struct kvm_vcpu *vcpu, int irq) { struct vcpu_svm *svm = to_svm(vcpu); + nested_svm_intr(svm); + svm_inject_irq(svm, irq); } @@ -1589,18 +2306,23 @@ static void svm_intr_assist(struct kvm_vcpu *vcpu) if (!kvm_cpu_has_interrupt(vcpu)) goto out; + if (nested_svm_intr(svm)) + goto out; + + if (!(svm->vcpu.arch.hflags & HF_GIF_MASK)) + goto out; + if (!(vmcb->save.rflags & X86_EFLAGS_IF) || (vmcb->control.int_state & SVM_INTERRUPT_SHADOW_MASK) || (vmcb->control.event_inj & SVM_EVTINJ_VALID)) { /* unable to deliver irq, set pending irq */ - vmcb->control.intercept |= (1ULL << INTERCEPT_VINTR); + svm_set_vintr(svm); svm_inject_irq(svm, 0x0); goto out; } /* Okay, we can deliver the interrupt: grab it and update PIC state. */ intr_vector = kvm_cpu_get_interrupt(vcpu); svm_inject_irq(svm, intr_vector); - kvm_timer_intr_post(vcpu, intr_vector); out: update_cr8_intercept(vcpu); } @@ -1616,7 +2338,8 @@ static void kvm_reput_irq(struct vcpu_svm *svm) } svm->vcpu.arch.interrupt_window_open = - !(control->int_state & SVM_INTERRUPT_SHADOW_MASK); + !(control->int_state & SVM_INTERRUPT_SHADOW_MASK) && + (svm->vcpu.arch.hflags & HF_GIF_MASK); } static void svm_do_inject_vector(struct vcpu_svm *svm) @@ -1638,9 +2361,13 @@ static void do_interrupt_requests(struct kvm_vcpu *vcpu, struct vcpu_svm *svm = to_svm(vcpu); struct vmcb_control_area *control = &svm->vmcb->control; + if (nested_svm_intr(svm)) + return; + svm->vcpu.arch.interrupt_window_open = (!(control->int_state & SVM_INTERRUPT_SHADOW_MASK) && - (svm->vmcb->save.rflags & X86_EFLAGS_IF)); + (svm->vmcb->save.rflags & X86_EFLAGS_IF) && + (svm->vcpu.arch.hflags & HF_GIF_MASK)); if (svm->vcpu.arch.interrupt_window_open && svm->vcpu.arch.irq_summary) /* @@ -1653,9 +2380,9 @@ static void do_interrupt_requests(struct kvm_vcpu *vcpu, */ if (!svm->vcpu.arch.interrupt_window_open && (svm->vcpu.arch.irq_summary || kvm_run->request_interrupt_window)) - control->intercept |= 1ULL << INTERCEPT_VINTR; - else - control->intercept &= ~(1ULL << INTERCEPT_VINTR); + svm_set_vintr(svm); + else + svm_clear_vintr(svm); } static int svm_set_tss_addr(struct kvm *kvm, unsigned int addr) @@ -1663,22 +2390,6 @@ static int svm_set_tss_addr(struct kvm *kvm, unsigned int addr) return 0; } -static void save_db_regs(unsigned long *db_regs) -{ - asm volatile ("mov %%dr0, %0" : "=r"(db_regs[0])); - asm volatile ("mov %%dr1, %0" : "=r"(db_regs[1])); - asm volatile ("mov %%dr2, %0" : "=r"(db_regs[2])); - asm volatile ("mov %%dr3, %0" : "=r"(db_regs[3])); -} - -static void load_db_regs(unsigned long *db_regs) -{ - asm volatile ("mov %0, %%dr0" : : "r"(db_regs[0])); - asm volatile ("mov %0, %%dr1" : : "r"(db_regs[1])); - asm volatile ("mov %0, %%dr2" : : "r"(db_regs[2])); - asm volatile ("mov %0, %%dr3" : : "r"(db_regs[3])); -} - static void svm_flush_tlb(struct kvm_vcpu *vcpu) { force_new_asid(vcpu); @@ -1737,19 +2448,12 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) gs_selector = kvm_read_gs(); ldt_selector = kvm_read_ldt(); svm->host_cr2 = kvm_read_cr2(); - svm->host_dr6 = read_dr6(); - svm->host_dr7 = read_dr7(); - svm->vmcb->save.cr2 = vcpu->arch.cr2; + if (!is_nested(svm)) + svm->vmcb->save.cr2 = vcpu->arch.cr2; /* required for live migration with NPT */ if (npt_enabled) svm->vmcb->save.cr3 = vcpu->arch.cr3; - if (svm->vmcb->save.dr7 & 0xff) { - write_dr7(0); - save_db_regs(svm->host_db_regs); - load_db_regs(svm->db_regs); - } - clgi(); local_irq_enable(); @@ -1825,16 +2529,11 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) #endif ); - if ((svm->vmcb->save.dr7 & 0xff)) - load_db_regs(svm->host_db_regs); - vcpu->arch.cr2 = svm->vmcb->save.cr2; vcpu->arch.regs[VCPU_REGS_RAX] = svm->vmcb->save.rax; vcpu->arch.regs[VCPU_REGS_RSP] = svm->vmcb->save.rsp; vcpu->arch.regs[VCPU_REGS_RIP] = svm->vmcb->save.rip; - write_dr6(svm->host_dr6); - write_dr7(svm->host_dr7); kvm_write_cr2(svm->host_cr2); kvm_load_fs(fs_selector); diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 6259d746764..bb481330716 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -91,6 +91,7 @@ struct vcpu_vmx { } rmode; int vpid; bool emulation_required; + enum emulation_result invalid_state_emulation_result; /* Support for vnmi-less CPUs */ int soft_vnmi_blocked; @@ -189,21 +190,21 @@ static inline int is_page_fault(u32 intr_info) { return (intr_info & (INTR_INFO_INTR_TYPE_MASK | INTR_INFO_VECTOR_MASK | INTR_INFO_VALID_MASK)) == - (INTR_TYPE_EXCEPTION | PF_VECTOR | INTR_INFO_VALID_MASK); + (INTR_TYPE_HARD_EXCEPTION | PF_VECTOR | INTR_INFO_VALID_MASK); } static inline int is_no_device(u32 intr_info) { return (intr_info & (INTR_INFO_INTR_TYPE_MASK | INTR_INFO_VECTOR_MASK | INTR_INFO_VALID_MASK)) == - (INTR_TYPE_EXCEPTION | NM_VECTOR | INTR_INFO_VALID_MASK); + (INTR_TYPE_HARD_EXCEPTION | NM_VECTOR | INTR_INFO_VALID_MASK); } static inline int is_invalid_opcode(u32 intr_info) { return (intr_info & (INTR_INFO_INTR_TYPE_MASK | INTR_INFO_VECTOR_MASK | INTR_INFO_VALID_MASK)) == - (INTR_TYPE_EXCEPTION | UD_VECTOR | INTR_INFO_VALID_MASK); + (INTR_TYPE_HARD_EXCEPTION | UD_VECTOR | INTR_INFO_VALID_MASK); } static inline int is_external_interrupt(u32 intr_info) @@ -480,8 +481,13 @@ static void update_exception_bitmap(struct kvm_vcpu *vcpu) eb = (1u << PF_VECTOR) | (1u << UD_VECTOR); if (!vcpu->fpu_active) eb |= 1u << NM_VECTOR; - if (vcpu->guest_debug.enabled) - eb |= 1u << DB_VECTOR; + if (vcpu->guest_debug & KVM_GUESTDBG_ENABLE) { + if (vcpu->guest_debug & + (KVM_GUESTDBG_SINGLESTEP | KVM_GUESTDBG_USE_HW_BP)) + eb |= 1u << DB_VECTOR; + if (vcpu->guest_debug & KVM_GUESTDBG_USE_SW_BP) + eb |= 1u << BP_VECTOR; + } if (vcpu->arch.rmode.active) eb = ~0; if (vm_need_ept()) @@ -747,29 +753,33 @@ static void vmx_queue_exception(struct kvm_vcpu *vcpu, unsigned nr, bool has_error_code, u32 error_code) { struct vcpu_vmx *vmx = to_vmx(vcpu); + u32 intr_info = nr | INTR_INFO_VALID_MASK; - if (has_error_code) + if (has_error_code) { vmcs_write32(VM_ENTRY_EXCEPTION_ERROR_CODE, error_code); + intr_info |= INTR_INFO_DELIVER_CODE_MASK; + } if (vcpu->arch.rmode.active) { vmx->rmode.irq.pending = true; vmx->rmode.irq.vector = nr; vmx->rmode.irq.rip = kvm_rip_read(vcpu); - if (nr == BP_VECTOR) + if (nr == BP_VECTOR || nr == OF_VECTOR) vmx->rmode.irq.rip++; - vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, - nr | INTR_TYPE_SOFT_INTR - | (has_error_code ? INTR_INFO_DELIVER_CODE_MASK : 0) - | INTR_INFO_VALID_MASK); + intr_info |= INTR_TYPE_SOFT_INTR; + vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, intr_info); vmcs_write32(VM_ENTRY_INSTRUCTION_LEN, 1); kvm_rip_write(vcpu, vmx->rmode.irq.rip - 1); return; } - vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, - nr | INTR_TYPE_EXCEPTION - | (has_error_code ? INTR_INFO_DELIVER_CODE_MASK : 0) - | INTR_INFO_VALID_MASK); + if (nr == BP_VECTOR || nr == OF_VECTOR) { + vmcs_write32(VM_ENTRY_INSTRUCTION_LEN, 1); + intr_info |= INTR_TYPE_SOFT_EXCEPTION; + } else + intr_info |= INTR_TYPE_HARD_EXCEPTION; + + vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, intr_info); } static bool vmx_exception_injected(struct kvm_vcpu *vcpu) @@ -856,11 +866,8 @@ static u64 guest_read_tsc(void) * writes 'guest_tsc' into guest's timestamp counter "register" * guest_tsc = host_tsc + tsc_offset ==> tsc_offset = guest_tsc - host_tsc */ -static void guest_write_tsc(u64 guest_tsc) +static void guest_write_tsc(u64 guest_tsc, u64 host_tsc) { - u64 host_tsc; - - rdtscll(host_tsc); vmcs_write64(TSC_OFFSET, guest_tsc - host_tsc); } @@ -903,6 +910,7 @@ static int vmx_get_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 *pdata) data = vmcs_readl(GUEST_SYSENTER_ESP); break; default: + vmx_load_host_state(to_vmx(vcpu)); msr = find_msr_entry(to_vmx(vcpu), msr_index); if (msr) { data = msr->data; @@ -924,14 +932,15 @@ static int vmx_set_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 data) { struct vcpu_vmx *vmx = to_vmx(vcpu); struct kvm_msr_entry *msr; + u64 host_tsc; int ret = 0; switch (msr_index) { -#ifdef CONFIG_X86_64 case MSR_EFER: vmx_load_host_state(vmx); ret = kvm_set_msr_common(vcpu, msr_index, data); break; +#ifdef CONFIG_X86_64 case MSR_FS_BASE: vmcs_writel(GUEST_FS_BASE, data); break; @@ -949,7 +958,8 @@ static int vmx_set_msr(struct kvm_vcpu *vcpu, u32 msr_index, u64 data) vmcs_writel(GUEST_SYSENTER_ESP, data); break; case MSR_IA32_TIME_STAMP_COUNTER: - guest_write_tsc(data); + rdtscll(host_tsc); + guest_write_tsc(data, host_tsc); break; case MSR_P6_PERFCTR0: case MSR_P6_PERFCTR1: @@ -998,40 +1008,28 @@ static void vmx_cache_reg(struct kvm_vcpu *vcpu, enum kvm_reg reg) } } -static int set_guest_debug(struct kvm_vcpu *vcpu, struct kvm_debug_guest *dbg) +static int set_guest_debug(struct kvm_vcpu *vcpu, struct kvm_guest_debug *dbg) { - unsigned long dr7 = 0x400; - int old_singlestep; - - old_singlestep = vcpu->guest_debug.singlestep; - - vcpu->guest_debug.enabled = dbg->enabled; - if (vcpu->guest_debug.enabled) { - int i; + int old_debug = vcpu->guest_debug; + unsigned long flags; - dr7 |= 0x200; /* exact */ - for (i = 0; i < 4; ++i) { - if (!dbg->breakpoints[i].enabled) - continue; - vcpu->guest_debug.bp[i] = dbg->breakpoints[i].address; - dr7 |= 2 << (i*2); /* global enable */ - dr7 |= 0 << (i*4+16); /* execution breakpoint */ - } + vcpu->guest_debug = dbg->control; + if (!(vcpu->guest_debug & KVM_GUESTDBG_ENABLE)) + vcpu->guest_debug = 0; - vcpu->guest_debug.singlestep = dbg->singlestep; - } else - vcpu->guest_debug.singlestep = 0; - - if (old_singlestep && !vcpu->guest_debug.singlestep) { - unsigned long flags; + if (vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP) + vmcs_writel(GUEST_DR7, dbg->arch.debugreg[7]); + else + vmcs_writel(GUEST_DR7, vcpu->arch.dr7); - flags = vmcs_readl(GUEST_RFLAGS); + flags = vmcs_readl(GUEST_RFLAGS); + if (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP) + flags |= X86_EFLAGS_TF | X86_EFLAGS_RF; + else if (old_debug & KVM_GUESTDBG_SINGLESTEP) flags &= ~(X86_EFLAGS_TF | X86_EFLAGS_RF); - vmcs_writel(GUEST_RFLAGS, flags); - } + vmcs_writel(GUEST_RFLAGS, flags); update_exception_bitmap(vcpu); - vmcs_writel(GUEST_DR7, dr7); return 0; } @@ -1432,6 +1430,29 @@ continue_rmode: init_rmode(vcpu->kvm); } +static void vmx_set_efer(struct kvm_vcpu *vcpu, u64 efer) +{ + struct vcpu_vmx *vmx = to_vmx(vcpu); + struct kvm_msr_entry *msr = find_msr_entry(vmx, MSR_EFER); + + vcpu->arch.shadow_efer = efer; + if (!msr) + return; + if (efer & EFER_LMA) { + vmcs_write32(VM_ENTRY_CONTROLS, + vmcs_read32(VM_ENTRY_CONTROLS) | + VM_ENTRY_IA32E_MODE); + msr->data = efer; + } else { + vmcs_write32(VM_ENTRY_CONTROLS, + vmcs_read32(VM_ENTRY_CONTROLS) & + ~VM_ENTRY_IA32E_MODE); + + msr->data = efer & ~EFER_LME; + } + setup_msrs(vmx); +} + #ifdef CONFIG_X86_64 static void enter_lmode(struct kvm_vcpu *vcpu) @@ -1446,13 +1467,8 @@ static void enter_lmode(struct kvm_vcpu *vcpu) (guest_tr_ar & ~AR_TYPE_MASK) | AR_TYPE_BUSY_64_TSS); } - vcpu->arch.shadow_efer |= EFER_LMA; - - find_msr_entry(to_vmx(vcpu), MSR_EFER)->data |= EFER_LMA | EFER_LME; - vmcs_write32(VM_ENTRY_CONTROLS, - vmcs_read32(VM_ENTRY_CONTROLS) - | VM_ENTRY_IA32E_MODE); + vmx_set_efer(vcpu, vcpu->arch.shadow_efer); } static void exit_lmode(struct kvm_vcpu *vcpu) @@ -1611,30 +1627,6 @@ static void vmx_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4) vmcs_writel(GUEST_CR4, hw_cr4); } -static void vmx_set_efer(struct kvm_vcpu *vcpu, u64 efer) -{ - struct vcpu_vmx *vmx = to_vmx(vcpu); - struct kvm_msr_entry *msr = find_msr_entry(vmx, MSR_EFER); - - vcpu->arch.shadow_efer = efer; - if (!msr) - return; - if (efer & EFER_LMA) { - vmcs_write32(VM_ENTRY_CONTROLS, - vmcs_read32(VM_ENTRY_CONTROLS) | - VM_ENTRY_IA32E_MODE); - msr->data = efer; - - } else { - vmcs_write32(VM_ENTRY_CONTROLS, - vmcs_read32(VM_ENTRY_CONTROLS) & - ~VM_ENTRY_IA32E_MODE); - - msr->data = efer & ~EFER_LME; - } - setup_msrs(vmx); -} - static u64 vmx_get_segment_base(struct kvm_vcpu *vcpu, int seg) { struct kvm_vmx_segment_field *sf = &kvm_vmx_segment_fields[seg]; @@ -1652,7 +1644,7 @@ static void vmx_get_segment(struct kvm_vcpu *vcpu, var->limit = vmcs_read32(sf->limit); var->selector = vmcs_read16(sf->selector); ar = vmcs_read32(sf->ar_bytes); - if (ar & AR_UNUSABLE_MASK) + if ((ar & AR_UNUSABLE_MASK) && !emulate_invalid_guest_state) ar = 0; var->type = ar & 15; var->s = (ar >> 4) & 1; @@ -1787,14 +1779,16 @@ static bool code_segment_valid(struct kvm_vcpu *vcpu) vmx_get_segment(vcpu, &cs, VCPU_SREG_CS); cs_rpl = cs.selector & SELECTOR_RPL_MASK; + if (cs.unusable) + return false; if (~cs.type & (AR_TYPE_CODE_MASK|AR_TYPE_ACCESSES_MASK)) return false; if (!cs.s) return false; - if (!(~cs.type & (AR_TYPE_CODE_MASK|AR_TYPE_WRITEABLE_MASK))) { + if (cs.type & AR_TYPE_WRITEABLE_MASK) { if (cs.dpl > cs_rpl) return false; - } else if (cs.type & AR_TYPE_CODE_MASK) { + } else { if (cs.dpl != cs_rpl) return false; } @@ -1813,7 +1807,9 @@ static bool stack_segment_valid(struct kvm_vcpu *vcpu) vmx_get_segment(vcpu, &ss, VCPU_SREG_SS); ss_rpl = ss.selector & SELECTOR_RPL_MASK; - if ((ss.type != 3) || (ss.type != 7)) + if (ss.unusable) + return true; + if (ss.type != 3 && ss.type != 7) return false; if (!ss.s) return false; @@ -1833,6 +1829,8 @@ static bool data_segment_valid(struct kvm_vcpu *vcpu, int seg) vmx_get_segment(vcpu, &var, seg); rpl = var.selector & SELECTOR_RPL_MASK; + if (var.unusable) + return true; if (!var.s) return false; if (!var.present) @@ -1854,9 +1852,11 @@ static bool tr_valid(struct kvm_vcpu *vcpu) vmx_get_segment(vcpu, &tr, VCPU_SREG_TR); + if (tr.unusable) + return false; if (tr.selector & SELECTOR_TI_MASK) /* TI = 1 */ return false; - if ((tr.type != 3) || (tr.type != 11)) /* TODO: Check if guest is in IA32e mode */ + if (tr.type != 3 && tr.type != 11) /* TODO: Check if guest is in IA32e mode */ return false; if (!tr.present) return false; @@ -1870,6 +1870,8 @@ static bool ldtr_valid(struct kvm_vcpu *vcpu) vmx_get_segment(vcpu, &ldtr, VCPU_SREG_LDTR); + if (ldtr.unusable) + return true; if (ldtr.selector & SELECTOR_TI_MASK) /* TI = 1 */ return false; if (ldtr.type != 2) @@ -2111,7 +2113,7 @@ static int vmx_vcpu_setup(struct vcpu_vmx *vmx) { u32 host_sysenter_cs, msr_low, msr_high; u32 junk; - u64 host_pat; + u64 host_pat, tsc_this, tsc_base; unsigned long a; struct descriptor_table dt; int i; @@ -2239,6 +2241,12 @@ static int vmx_vcpu_setup(struct vcpu_vmx *vmx) vmcs_writel(CR0_GUEST_HOST_MASK, ~0UL); vmcs_writel(CR4_GUEST_HOST_MASK, KVM_GUEST_CR4_MASK); + tsc_base = vmx->vcpu.kvm->arch.vm_init_tsc; + rdtscll(tsc_this); + if (tsc_this < vmx->vcpu.kvm->arch.vm_init_tsc) + tsc_base = tsc_this; + + guest_write_tsc(0, tsc_base); return 0; } @@ -2318,7 +2326,6 @@ static int vmx_vcpu_reset(struct kvm_vcpu *vcpu) kvm_rip_write(vcpu, 0); kvm_register_write(vcpu, VCPU_REGS_RSP, 0); - /* todo: dr0 = dr1 = dr2 = dr3 = 0; dr6 = 0xffff0ff0 */ vmcs_writel(GUEST_DR7, 0x400); vmcs_writel(GUEST_GDTR_BASE, 0); @@ -2331,8 +2338,6 @@ static int vmx_vcpu_reset(struct kvm_vcpu *vcpu) vmcs_write32(GUEST_INTERRUPTIBILITY_INFO, 0); vmcs_write32(GUEST_PENDING_DBG_EXCEPTIONS, 0); - guest_write_tsc(0); - /* Special registers */ vmcs_write64(GUEST_IA32_DEBUGCTL, 0); @@ -2485,6 +2490,11 @@ static void do_interrupt_requests(struct kvm_vcpu *vcpu, { vmx_update_window_states(vcpu); + if (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP) + vmcs_clear_bits(GUEST_INTERRUPTIBILITY_INFO, + GUEST_INTR_STATE_STI | + GUEST_INTR_STATE_MOV_SS); + if (vcpu->arch.nmi_pending && !vcpu->arch.nmi_injected) { if (vcpu->arch.interrupt.pending) { enable_nmi_window(vcpu); @@ -2535,24 +2545,6 @@ static int vmx_set_tss_addr(struct kvm *kvm, unsigned int addr) return 0; } -static void kvm_guest_debug_pre(struct kvm_vcpu *vcpu) -{ - struct kvm_guest_debug *dbg = &vcpu->guest_debug; - - set_debugreg(dbg->bp[0], 0); - set_debugreg(dbg->bp[1], 1); - set_debugreg(dbg->bp[2], 2); - set_debugreg(dbg->bp[3], 3); - - if (dbg->singlestep) { - unsigned long flags; - - flags = vmcs_readl(GUEST_RFLAGS); - flags |= X86_EFLAGS_TF | X86_EFLAGS_RF; - vmcs_writel(GUEST_RFLAGS, flags); - } -} - static int handle_rmode_exception(struct kvm_vcpu *vcpu, int vec, u32 err_code) { @@ -2569,9 +2561,17 @@ static int handle_rmode_exception(struct kvm_vcpu *vcpu, * the required debugging infrastructure rework. */ switch (vec) { - case DE_VECTOR: case DB_VECTOR: + if (vcpu->guest_debug & + (KVM_GUESTDBG_SINGLESTEP | KVM_GUESTDBG_USE_HW_BP)) + return 0; + kvm_queue_exception(vcpu, vec); + return 1; case BP_VECTOR: + if (vcpu->guest_debug & KVM_GUESTDBG_USE_SW_BP) + return 0; + /* fall through */ + case DE_VECTOR: case OF_VECTOR: case BR_VECTOR: case UD_VECTOR: @@ -2588,8 +2588,8 @@ static int handle_rmode_exception(struct kvm_vcpu *vcpu, static int handle_exception(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) { struct vcpu_vmx *vmx = to_vmx(vcpu); - u32 intr_info, error_code; - unsigned long cr2, rip; + u32 intr_info, ex_no, error_code; + unsigned long cr2, rip, dr6; u32 vect_info; enum emulation_result er; @@ -2648,14 +2648,30 @@ static int handle_exception(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) return 1; } - if ((intr_info & (INTR_INFO_INTR_TYPE_MASK | INTR_INFO_VECTOR_MASK)) == - (INTR_TYPE_EXCEPTION | 1)) { + ex_no = intr_info & INTR_INFO_VECTOR_MASK; + switch (ex_no) { + case DB_VECTOR: + dr6 = vmcs_readl(EXIT_QUALIFICATION); + if (!(vcpu->guest_debug & + (KVM_GUESTDBG_SINGLESTEP | KVM_GUESTDBG_USE_HW_BP))) { + vcpu->arch.dr6 = dr6 | DR6_FIXED_1; + kvm_queue_exception(vcpu, DB_VECTOR); + return 1; + } + kvm_run->debug.arch.dr6 = dr6 | DR6_FIXED_1; + kvm_run->debug.arch.dr7 = vmcs_readl(GUEST_DR7); + /* fall through */ + case BP_VECTOR: kvm_run->exit_reason = KVM_EXIT_DEBUG; - return 0; + kvm_run->debug.arch.pc = vmcs_readl(GUEST_CS_BASE) + rip; + kvm_run->debug.arch.exception = ex_no; + break; + default: + kvm_run->exit_reason = KVM_EXIT_EXCEPTION; + kvm_run->ex.exception = ex_no; + kvm_run->ex.error_code = error_code; + break; } - kvm_run->exit_reason = KVM_EXIT_EXCEPTION; - kvm_run->ex.exception = intr_info & INTR_INFO_VECTOR_MASK; - kvm_run->ex.error_code = error_code; return 0; } @@ -2676,7 +2692,7 @@ static int handle_triple_fault(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) static int handle_io(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) { unsigned long exit_qualification; - int size, down, in, string, rep; + int size, in, string; unsigned port; ++vcpu->stat.io_exits; @@ -2692,8 +2708,6 @@ static int handle_io(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) size = (exit_qualification & 7) + 1; in = (exit_qualification & 8) != 0; - down = (vmcs_readl(GUEST_RFLAGS) & X86_EFLAGS_DF) != 0; - rep = (exit_qualification & 32) != 0; port = exit_qualification >> 16; skip_emulated_instruction(vcpu); @@ -2794,21 +2808,44 @@ static int handle_dr(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) unsigned long val; int dr, reg; - /* - * FIXME: this code assumes the host is debugging the guest. - * need to deal with guest debugging itself too. - */ + dr = vmcs_readl(GUEST_DR7); + if (dr & DR7_GD) { + /* + * As the vm-exit takes precedence over the debug trap, we + * need to emulate the latter, either for the host or the + * guest debugging itself. + */ + if (vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP) { + kvm_run->debug.arch.dr6 = vcpu->arch.dr6; + kvm_run->debug.arch.dr7 = dr; + kvm_run->debug.arch.pc = + vmcs_readl(GUEST_CS_BASE) + + vmcs_readl(GUEST_RIP); + kvm_run->debug.arch.exception = DB_VECTOR; + kvm_run->exit_reason = KVM_EXIT_DEBUG; + return 0; + } else { + vcpu->arch.dr7 &= ~DR7_GD; + vcpu->arch.dr6 |= DR6_BD; + vmcs_writel(GUEST_DR7, vcpu->arch.dr7); + kvm_queue_exception(vcpu, DB_VECTOR); + return 1; + } + } + exit_qualification = vmcs_readl(EXIT_QUALIFICATION); - dr = exit_qualification & 7; - reg = (exit_qualification >> 8) & 15; - if (exit_qualification & 16) { - /* mov from dr */ + dr = exit_qualification & DEBUG_REG_ACCESS_NUM; + reg = DEBUG_REG_ACCESS_REG(exit_qualification); + if (exit_qualification & TYPE_MOV_FROM_DR) { switch (dr) { + case 0 ... 3: + val = vcpu->arch.db[dr]; + break; case 6: - val = 0xffff0ff0; + val = vcpu->arch.dr6; break; case 7: - val = 0x400; + val = vcpu->arch.dr7; break; default: val = 0; @@ -2816,7 +2853,38 @@ static int handle_dr(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) kvm_register_write(vcpu, reg, val); KVMTRACE_2D(DR_READ, vcpu, (u32)dr, (u32)val, handler); } else { - /* mov to dr */ + val = vcpu->arch.regs[reg]; + switch (dr) { + case 0 ... 3: + vcpu->arch.db[dr] = val; + if (!(vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP)) + vcpu->arch.eff_db[dr] = val; + break; + case 4 ... 5: + if (vcpu->arch.cr4 & X86_CR4_DE) + kvm_queue_exception(vcpu, UD_VECTOR); + break; + case 6: + if (val & 0xffffffff00000000ULL) { + kvm_queue_exception(vcpu, GP_VECTOR); + break; + } + vcpu->arch.dr6 = (val & DR6_VOLATILE) | DR6_FIXED_1; + break; + case 7: + if (val & 0xffffffff00000000ULL) { + kvm_queue_exception(vcpu, GP_VECTOR); + break; + } + vcpu->arch.dr7 = (val & DR7_VOLATILE) | DR7_FIXED_1; + if (!(vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP)) { + vmcs_writel(GUEST_DR7, vcpu->arch.dr7); + vcpu->arch.switch_db_regs = + (val & DR7_BP_EN_MASK); + } + break; + } + KVMTRACE_2D(DR_WRITE, vcpu, (u32)dr, (u32)val, handler); } skip_emulated_instruction(vcpu); return 1; @@ -2967,17 +3035,25 @@ static int handle_task_switch(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) } tss_selector = exit_qualification; - return kvm_task_switch(vcpu, tss_selector, reason); + if (!kvm_task_switch(vcpu, tss_selector, reason)) + return 0; + + /* clear all local breakpoint enable flags */ + vmcs_writel(GUEST_DR7, vmcs_readl(GUEST_DR7) & ~55); + + /* + * TODO: What about debug traps on tss switch? + * Are we supposed to inject them and update dr6? + */ + + return 1; } static int handle_ept_violation(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) { u64 exit_qualification; - enum emulation_result er; gpa_t gpa; - unsigned long hva; int gla_validity; - int r; exit_qualification = vmcs_read64(EXIT_QUALIFICATION); @@ -3000,32 +3076,7 @@ static int handle_ept_violation(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) } gpa = vmcs_read64(GUEST_PHYSICAL_ADDRESS); - hva = gfn_to_hva(vcpu->kvm, gpa >> PAGE_SHIFT); - if (!kvm_is_error_hva(hva)) { - r = kvm_mmu_page_fault(vcpu, gpa & PAGE_MASK, 0); - if (r < 0) { - printk(KERN_ERR "EPT: Not enough memory!\n"); - return -ENOMEM; - } - return 1; - } else { - /* must be MMIO */ - er = emulate_instruction(vcpu, kvm_run, 0, 0, 0); - - if (er == EMULATE_FAIL) { - printk(KERN_ERR - "EPT: Fail to handle EPT violation vmexit!er is %d\n", - er); - printk(KERN_ERR "EPT: GPA: 0x%lx, GVA: 0x%lx\n", - (long unsigned int)vmcs_read64(GUEST_PHYSICAL_ADDRESS), - (long unsigned int)vmcs_read64(GUEST_LINEAR_ADDRESS)); - printk(KERN_ERR "EPT: Exit qualification is 0x%lx\n", - (long unsigned int)exit_qualification); - return -ENOTSUPP; - } else if (er == EMULATE_DO_MMIO) - return 0; - } - return 1; + return kvm_mmu_page_fault(vcpu, gpa & PAGE_MASK, 0); } static int handle_nmi_window(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) @@ -3045,7 +3096,7 @@ static void handle_invalid_guest_state(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) { struct vcpu_vmx *vmx = to_vmx(vcpu); - int err; + enum emulation_result err = EMULATE_DONE; preempt_enable(); local_irq_enable(); @@ -3070,10 +3121,7 @@ static void handle_invalid_guest_state(struct kvm_vcpu *vcpu, local_irq_disable(); preempt_disable(); - /* Guest state should be valid now except if we need to - * emulate an MMIO */ - if (guest_state_valid(vcpu)) - vmx->emulation_required = 0; + vmx->invalid_state_emulation_result = err; } /* @@ -3122,8 +3170,11 @@ static int kvm_handle_exit(struct kvm_run *kvm_run, struct kvm_vcpu *vcpu) /* If we need to emulate an MMIO from handle_invalid_guest_state * we just return 0 */ - if (vmx->emulation_required && emulate_invalid_guest_state) - return 0; + if (vmx->emulation_required && emulate_invalid_guest_state) { + if (guest_state_valid(vcpu)) + vmx->emulation_required = 0; + return vmx->invalid_state_emulation_result != EMULATE_DO_MMIO; + } /* Access CR3 don't cause VMExit in paging mode, so we need * to sync with guest real CR3. */ @@ -3237,7 +3288,8 @@ static void vmx_complete_interrupts(struct vcpu_vmx *vmx) vmx->vcpu.arch.nmi_injected = false; } kvm_clear_exception_queue(&vmx->vcpu); - if (idtv_info_valid && type == INTR_TYPE_EXCEPTION) { + if (idtv_info_valid && (type == INTR_TYPE_HARD_EXCEPTION || + type == INTR_TYPE_SOFT_EXCEPTION)) { if (idt_vectoring_info & VECTORING_INFO_DELIVER_CODE_MASK) { error = vmcs_read32(IDT_VECTORING_ERROR_CODE); kvm_queue_exception_e(&vmx->vcpu, vector, error); @@ -3258,6 +3310,11 @@ static void vmx_intr_assist(struct kvm_vcpu *vcpu) vmx_update_window_states(vcpu); + if (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP) + vmcs_clear_bits(GUEST_INTERRUPTIBILITY_INFO, + GUEST_INTR_STATE_STI | + GUEST_INTR_STATE_MOV_SS); + if (vcpu->arch.nmi_pending && !vcpu->arch.nmi_injected) { if (vcpu->arch.interrupt.pending) { enable_nmi_window(vcpu); @@ -3285,7 +3342,6 @@ static void vmx_intr_assist(struct kvm_vcpu *vcpu) } if (vcpu->arch.interrupt.pending) { vmx_inject_irq(vcpu, vcpu->arch.interrupt.nr); - kvm_timer_intr_post(vcpu, vcpu->arch.interrupt.nr); if (kvm_cpu_has_interrupt(vcpu)) enable_irq_window(vcpu); } @@ -3347,6 +3403,8 @@ static void vmx_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) */ vmcs_writel(HOST_CR0, read_cr0()); + set_debugreg(vcpu->arch.dr6, 6); + asm( /* Store host registers */ "push %%"R"dx; push %%"R"bp;" @@ -3441,6 +3499,8 @@ static void vmx_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) vcpu->arch.regs_avail = ~((1 << VCPU_REGS_RIP) | (1 << VCPU_REGS_RSP)); vcpu->arch.regs_dirty = 0; + get_debugreg(vcpu->arch.dr6, 6); + vmx->idt_vectoring_info = vmcs_read32(IDT_VECTORING_INFO_FIELD); if (vmx->rmode.irq.pending) fixup_rmode_irq(vmx); @@ -3595,7 +3655,6 @@ static struct kvm_x86_ops vmx_x86_ops = { .vcpu_put = vmx_vcpu_put, .set_guest_debug = set_guest_debug, - .guest_debug_pre = kvm_guest_debug_pre, .get_msr = vmx_get_msr, .set_msr = vmx_set_msr, .get_segment_base = vmx_get_segment_base, @@ -3687,8 +3746,7 @@ static int __init vmx_init(void) if (vm_need_ept()) { bypass_guest_pf = 0; kvm_mmu_set_base_ptes(VMX_EPT_READABLE_MASK | - VMX_EPT_WRITABLE_MASK | - VMX_EPT_IGMT_BIT); + VMX_EPT_WRITABLE_MASK); kvm_mmu_set_mask_ptes(0ull, 0ull, 0ull, 0ull, VMX_EPT_EXECUTABLE_MASK, VMX_EPT_DEFAULT_MT << VMX_EPT_MT_EPTE_SHIFT); diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index cc17546a240..8ca100a9eca 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -36,6 +36,7 @@ #include <linux/highmem.h> #include <linux/iommu.h> #include <linux/intel-iommu.h> +#include <linux/cpufreq.h> #include <asm/uaccess.h> #include <asm/msr.h> @@ -69,6 +70,8 @@ static u64 __read_mostly efer_reserved_bits = 0xfffffffffffffffeULL; static int kvm_dev_ioctl_get_supported_cpuid(struct kvm_cpuid2 *cpuid, struct kvm_cpuid_entry2 __user *entries); +struct kvm_cpuid_entry2 *kvm_find_cpuid_entry(struct kvm_vcpu *vcpu, + u32 function, u32 index); struct kvm_x86_ops *kvm_x86_ops; EXPORT_SYMBOL_GPL(kvm_x86_ops); @@ -173,6 +176,7 @@ void kvm_inject_page_fault(struct kvm_vcpu *vcpu, unsigned long addr, u32 error_code) { ++vcpu->stat.pf_guest; + if (vcpu->arch.exception.pending) { if (vcpu->arch.exception.nr == PF_VECTOR) { printk(KERN_DEBUG "kvm: inject_page_fault:" @@ -361,6 +365,7 @@ void kvm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4) } kvm_x86_ops->set_cr4(vcpu, cr4); vcpu->arch.cr4 = cr4; + vcpu->arch.mmu.base_role.cr4_pge = (cr4 & X86_CR4_PGE) && !tdp_enabled; kvm_mmu_sync_global(vcpu); kvm_mmu_reset_context(vcpu); } @@ -442,6 +447,11 @@ unsigned long kvm_get_cr8(struct kvm_vcpu *vcpu) } EXPORT_SYMBOL_GPL(kvm_get_cr8); +static inline u32 bit(int bitno) +{ + return 1 << (bitno & 31); +} + /* * List of msr numbers which we expose to userspace through KVM_GET_MSRS * and KVM_SET_MSRS, and KVM_GET_MSR_INDEX_LIST. @@ -456,7 +466,7 @@ static u32 msrs_to_save[] = { MSR_CSTAR, MSR_KERNEL_GS_BASE, MSR_SYSCALL_MASK, MSR_LSTAR, #endif MSR_IA32_TIME_STAMP_COUNTER, MSR_KVM_SYSTEM_TIME, MSR_KVM_WALL_CLOCK, - MSR_IA32_PERF_STATUS, MSR_IA32_CR_PAT + MSR_IA32_PERF_STATUS, MSR_IA32_CR_PAT, MSR_VM_HSAVE_PA }; static unsigned num_msrs_to_save; @@ -481,6 +491,28 @@ static void set_efer(struct kvm_vcpu *vcpu, u64 efer) return; } + if (efer & EFER_FFXSR) { + struct kvm_cpuid_entry2 *feat; + + feat = kvm_find_cpuid_entry(vcpu, 0x80000001, 0); + if (!feat || !(feat->edx & bit(X86_FEATURE_FXSR_OPT))) { + printk(KERN_DEBUG "set_efer: #GP, enable FFXSR w/o CPUID capability\n"); + kvm_inject_gp(vcpu, 0); + return; + } + } + + if (efer & EFER_SVME) { + struct kvm_cpuid_entry2 *feat; + + feat = kvm_find_cpuid_entry(vcpu, 0x80000001, 0); + if (!feat || !(feat->ecx & bit(X86_FEATURE_SVM))) { + printk(KERN_DEBUG "set_efer: #GP, enable SVM w/o SVM\n"); + kvm_inject_gp(vcpu, 0); + return; + } + } + kvm_x86_ops->set_efer(vcpu, efer); efer &= ~EFER_LMA; @@ -586,6 +618,8 @@ static void kvm_set_time_scale(uint32_t tsc_khz, struct pvclock_vcpu_time_info * hv_clock->tsc_to_system_mul); } +static DEFINE_PER_CPU(unsigned long, cpu_tsc_khz); + static void kvm_write_guest_time(struct kvm_vcpu *v) { struct timespec ts; @@ -596,9 +630,9 @@ static void kvm_write_guest_time(struct kvm_vcpu *v) if ((!vcpu->time_page)) return; - if (unlikely(vcpu->hv_clock_tsc_khz != tsc_khz)) { - kvm_set_time_scale(tsc_khz, &vcpu->hv_clock); - vcpu->hv_clock_tsc_khz = tsc_khz; + if (unlikely(vcpu->hv_clock_tsc_khz != __get_cpu_var(cpu_tsc_khz))) { + kvm_set_time_scale(__get_cpu_var(cpu_tsc_khz), &vcpu->hv_clock); + vcpu->hv_clock_tsc_khz = __get_cpu_var(cpu_tsc_khz); } /* Keep irq disabled to prevent changes to the clock */ @@ -629,6 +663,16 @@ static void kvm_write_guest_time(struct kvm_vcpu *v) mark_page_dirty(v->kvm, vcpu->time >> PAGE_SHIFT); } +static int kvm_request_guest_time_update(struct kvm_vcpu *v) +{ + struct kvm_vcpu_arch *vcpu = &v->arch; + + if (!vcpu->time_page) + return 0; + set_bit(KVM_REQ_KVMCLOCK_UPDATE, &v->requests); + return 1; +} + static bool msr_mtrr_valid(unsigned msr) { switch (msr) { @@ -722,6 +766,7 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data) break; case MSR_IA32_UCODE_REV: case MSR_IA32_UCODE_WRITE: + case MSR_VM_HSAVE_PA: break; case 0x200 ... 0x2ff: return set_msr_mtrr(vcpu, msr, data); @@ -758,7 +803,7 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data) vcpu->arch.time_page = NULL; } - kvm_write_guest_time(vcpu); + kvm_request_guest_time_update(vcpu); break; } default: @@ -843,6 +888,7 @@ int kvm_get_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata) case MSR_IA32_LASTBRANCHTOIP: case MSR_IA32_LASTINTFROMIP: case MSR_IA32_LASTINTTOIP: + case MSR_VM_HSAVE_PA: data = 0; break; case MSR_MTRRcap: @@ -972,6 +1018,8 @@ int kvm_dev_ioctl_check_extension(long ext) case KVM_CAP_NOP_IO_DELAY: case KVM_CAP_MP_STATE: case KVM_CAP_SYNC_MMU: + case KVM_CAP_REINJECT_CONTROL: + case KVM_CAP_IRQ_INJECT_STATUS: r = 1; break; case KVM_CAP_COALESCED_MMIO: @@ -1042,7 +1090,7 @@ long kvm_arch_dev_ioctl(struct file *filp, if (copy_from_user(&cpuid, cpuid_arg, sizeof cpuid)) goto out; r = kvm_dev_ioctl_get_supported_cpuid(&cpuid, - cpuid_arg->entries); + cpuid_arg->entries); if (r) goto out; @@ -1062,7 +1110,7 @@ out: void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu) { kvm_x86_ops->vcpu_load(vcpu, cpu); - kvm_write_guest_time(vcpu); + kvm_request_guest_time_update(vcpu); } void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu) @@ -1140,8 +1188,8 @@ out: } static int kvm_vcpu_ioctl_set_cpuid2(struct kvm_vcpu *vcpu, - struct kvm_cpuid2 *cpuid, - struct kvm_cpuid_entry2 __user *entries) + struct kvm_cpuid2 *cpuid, + struct kvm_cpuid_entry2 __user *entries) { int r; @@ -1160,8 +1208,8 @@ out: } static int kvm_vcpu_ioctl_get_cpuid2(struct kvm_vcpu *vcpu, - struct kvm_cpuid2 *cpuid, - struct kvm_cpuid_entry2 __user *entries) + struct kvm_cpuid2 *cpuid, + struct kvm_cpuid_entry2 __user *entries) { int r; @@ -1170,7 +1218,7 @@ static int kvm_vcpu_ioctl_get_cpuid2(struct kvm_vcpu *vcpu, goto out; r = -EFAULT; if (copy_to_user(entries, &vcpu->arch.cpuid_entries, - vcpu->arch.cpuid_nent * sizeof(struct kvm_cpuid_entry2))) + vcpu->arch.cpuid_nent * sizeof(struct kvm_cpuid_entry2))) goto out; return 0; @@ -1179,18 +1227,13 @@ out: return r; } -static inline u32 bit(int bitno) -{ - return 1 << (bitno & 31); -} - static void do_cpuid_1_ent(struct kvm_cpuid_entry2 *entry, u32 function, - u32 index) + u32 index) { entry->function = function; entry->index = index; cpuid_count(entry->function, entry->index, - &entry->eax, &entry->ebx, &entry->ecx, &entry->edx); + &entry->eax, &entry->ebx, &entry->ecx, &entry->edx); entry->flags = 0; } @@ -1220,15 +1263,17 @@ static void do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function, #ifdef CONFIG_X86_64 bit(X86_FEATURE_LM) | #endif + bit(X86_FEATURE_FXSR_OPT) | bit(X86_FEATURE_MMXEXT) | bit(X86_FEATURE_3DNOWEXT) | bit(X86_FEATURE_3DNOW); const u32 kvm_supported_word3_x86_features = bit(X86_FEATURE_XMM3) | bit(X86_FEATURE_CX16); const u32 kvm_supported_word6_x86_features = - bit(X86_FEATURE_LAHF_LM) | bit(X86_FEATURE_CMP_LEGACY); + bit(X86_FEATURE_LAHF_LM) | bit(X86_FEATURE_CMP_LEGACY) | + bit(X86_FEATURE_SVM); - /* all func 2 cpuid_count() should be called on the same cpu */ + /* all calls to cpuid_count() should be made on the same cpu */ get_cpu(); do_cpuid_1_ent(entry, function, index); ++*nent; @@ -1302,7 +1347,7 @@ static void do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function, } static int kvm_dev_ioctl_get_supported_cpuid(struct kvm_cpuid2 *cpuid, - struct kvm_cpuid_entry2 __user *entries) + struct kvm_cpuid_entry2 __user *entries) { struct kvm_cpuid_entry2 *cpuid_entries; int limit, nent = 0, r = -E2BIG; @@ -1319,7 +1364,7 @@ static int kvm_dev_ioctl_get_supported_cpuid(struct kvm_cpuid2 *cpuid, limit = cpuid_entries[0].eax; for (func = 1; func <= limit && nent < cpuid->nent; ++func) do_cpuid_ent(&cpuid_entries[nent], func, 0, - &nent, cpuid->nent); + &nent, cpuid->nent); r = -E2BIG; if (nent >= cpuid->nent) goto out_free; @@ -1328,10 +1373,10 @@ static int kvm_dev_ioctl_get_supported_cpuid(struct kvm_cpuid2 *cpuid, limit = cpuid_entries[nent - 1].eax; for (func = 0x80000001; func <= limit && nent < cpuid->nent; ++func) do_cpuid_ent(&cpuid_entries[nent], func, 0, - &nent, cpuid->nent); + &nent, cpuid->nent); r = -EFAULT; if (copy_to_user(entries, cpuid_entries, - nent * sizeof(struct kvm_cpuid_entry2))) + nent * sizeof(struct kvm_cpuid_entry2))) goto out_free; cpuid->nent = nent; r = 0; @@ -1475,7 +1520,7 @@ long kvm_arch_vcpu_ioctl(struct file *filp, if (copy_from_user(&cpuid, cpuid_arg, sizeof cpuid)) goto out; r = kvm_vcpu_ioctl_set_cpuid2(vcpu, &cpuid, - cpuid_arg->entries); + cpuid_arg->entries); if (r) goto out; break; @@ -1488,7 +1533,7 @@ long kvm_arch_vcpu_ioctl(struct file *filp, if (copy_from_user(&cpuid, cpuid_arg, sizeof cpuid)) goto out; r = kvm_vcpu_ioctl_get_cpuid2(vcpu, &cpuid, - cpuid_arg->entries); + cpuid_arg->entries); if (r) goto out; r = -EFAULT; @@ -1708,6 +1753,15 @@ static int kvm_vm_ioctl_set_pit(struct kvm *kvm, struct kvm_pit_state *ps) return r; } +static int kvm_vm_ioctl_reinject(struct kvm *kvm, + struct kvm_reinject_control *control) +{ + if (!kvm->arch.vpit) + return -ENXIO; + kvm->arch.vpit->pit_state.pit_timer.reinject = control->pit_reinject; + return 0; +} + /* * Get (and clear) the dirty memory log for a memory slot. */ @@ -1805,13 +1859,26 @@ long kvm_arch_vm_ioctl(struct file *filp, } } else goto out; + r = kvm_setup_default_irq_routing(kvm); + if (r) { + kfree(kvm->arch.vpic); + kfree(kvm->arch.vioapic); + goto out; + } break; case KVM_CREATE_PIT: + mutex_lock(&kvm->lock); + r = -EEXIST; + if (kvm->arch.vpit) + goto create_pit_unlock; r = -ENOMEM; kvm->arch.vpit = kvm_create_pit(kvm); if (kvm->arch.vpit) r = 0; + create_pit_unlock: + mutex_unlock(&kvm->lock); break; + case KVM_IRQ_LINE_STATUS: case KVM_IRQ_LINE: { struct kvm_irq_level irq_event; @@ -1819,10 +1886,17 @@ long kvm_arch_vm_ioctl(struct file *filp, if (copy_from_user(&irq_event, argp, sizeof irq_event)) goto out; if (irqchip_in_kernel(kvm)) { + __s32 status; mutex_lock(&kvm->lock); - kvm_set_irq(kvm, KVM_USERSPACE_IRQ_SOURCE_ID, - irq_event.irq, irq_event.level); + status = kvm_set_irq(kvm, KVM_USERSPACE_IRQ_SOURCE_ID, + irq_event.irq, irq_event.level); mutex_unlock(&kvm->lock); + if (ioctl == KVM_IRQ_LINE_STATUS) { + irq_event.status = status; + if (copy_to_user(argp, &irq_event, + sizeof irq_event)) + goto out; + } r = 0; } break; @@ -1905,6 +1979,17 @@ long kvm_arch_vm_ioctl(struct file *filp, r = 0; break; } + case KVM_REINJECT_CONTROL: { + struct kvm_reinject_control control; + r = -EFAULT; + if (copy_from_user(&control, argp, sizeof(control))) + goto out; + r = kvm_vm_ioctl_reinject(kvm, &control); + if (r) + goto out; + r = 0; + break; + } default: ; } @@ -1958,10 +2043,8 @@ static struct kvm_io_device *vcpu_find_mmio_dev(struct kvm_vcpu *vcpu, return dev; } -int emulator_read_std(unsigned long addr, - void *val, - unsigned int bytes, - struct kvm_vcpu *vcpu) +static int kvm_read_guest_virt(gva_t addr, void *val, unsigned int bytes, + struct kvm_vcpu *vcpu) { void *data = val; int r = X86EMUL_CONTINUE; @@ -1969,27 +2052,57 @@ int emulator_read_std(unsigned long addr, while (bytes) { gpa_t gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, addr); unsigned offset = addr & (PAGE_SIZE-1); - unsigned tocopy = min(bytes, (unsigned)PAGE_SIZE - offset); + unsigned toread = min(bytes, (unsigned)PAGE_SIZE - offset); int ret; if (gpa == UNMAPPED_GVA) { r = X86EMUL_PROPAGATE_FAULT; goto out; } - ret = kvm_read_guest(vcpu->kvm, gpa, data, tocopy); + ret = kvm_read_guest(vcpu->kvm, gpa, data, toread); if (ret < 0) { r = X86EMUL_UNHANDLEABLE; goto out; } - bytes -= tocopy; - data += tocopy; - addr += tocopy; + bytes -= toread; + data += toread; + addr += toread; } out: return r; } -EXPORT_SYMBOL_GPL(emulator_read_std); + +static int kvm_write_guest_virt(gva_t addr, void *val, unsigned int bytes, + struct kvm_vcpu *vcpu) +{ + void *data = val; + int r = X86EMUL_CONTINUE; + + while (bytes) { + gpa_t gpa = vcpu->arch.mmu.gva_to_gpa(vcpu, addr); + unsigned offset = addr & (PAGE_SIZE-1); + unsigned towrite = min(bytes, (unsigned)PAGE_SIZE - offset); + int ret; + + if (gpa == UNMAPPED_GVA) { + r = X86EMUL_PROPAGATE_FAULT; + goto out; + } + ret = kvm_write_guest(vcpu->kvm, gpa, data, towrite); + if (ret < 0) { + r = X86EMUL_UNHANDLEABLE; + goto out; + } + + bytes -= towrite; + data += towrite; + addr += towrite; + } +out: + return r; +} + static int emulator_read_emulated(unsigned long addr, void *val, @@ -2011,8 +2124,8 @@ static int emulator_read_emulated(unsigned long addr, if ((gpa & PAGE_MASK) == APIC_DEFAULT_PHYS_BASE) goto mmio; - if (emulator_read_std(addr, val, bytes, vcpu) - == X86EMUL_CONTINUE) + if (kvm_read_guest_virt(addr, val, bytes, vcpu) + == X86EMUL_CONTINUE) return X86EMUL_CONTINUE; if (gpa == UNMAPPED_GVA) return X86EMUL_PROPAGATE_FAULT; @@ -2215,7 +2328,7 @@ void kvm_report_emulation_failure(struct kvm_vcpu *vcpu, const char *context) rip_linear = rip + get_segment_base(vcpu, VCPU_SREG_CS); - emulator_read_std(rip_linear, (void *)opcodes, 4, vcpu); + kvm_read_guest_virt(rip_linear, (void *)opcodes, 4, vcpu); printk(KERN_ERR "emulation failed (%s) rip %lx %02x %02x %02x %02x\n", context, rip, opcodes[0], opcodes[1], opcodes[2], opcodes[3]); @@ -2223,7 +2336,7 @@ void kvm_report_emulation_failure(struct kvm_vcpu *vcpu, const char *context) EXPORT_SYMBOL_GPL(kvm_report_emulation_failure); static struct x86_emulate_ops emulate_ops = { - .read_std = emulator_read_std, + .read_std = kvm_read_guest_virt, .read_emulated = emulator_read_emulated, .write_emulated = emulator_write_emulated, .cmpxchg_emulated = emulator_cmpxchg_emulated, @@ -2325,40 +2438,19 @@ int emulate_instruction(struct kvm_vcpu *vcpu, } EXPORT_SYMBOL_GPL(emulate_instruction); -static void free_pio_guest_pages(struct kvm_vcpu *vcpu) -{ - int i; - - for (i = 0; i < ARRAY_SIZE(vcpu->arch.pio.guest_pages); ++i) - if (vcpu->arch.pio.guest_pages[i]) { - kvm_release_page_dirty(vcpu->arch.pio.guest_pages[i]); - vcpu->arch.pio.guest_pages[i] = NULL; - } -} - static int pio_copy_data(struct kvm_vcpu *vcpu) { void *p = vcpu->arch.pio_data; - void *q; + gva_t q = vcpu->arch.pio.guest_gva; unsigned bytes; - int nr_pages = vcpu->arch.pio.guest_pages[1] ? 2 : 1; + int ret; - q = vmap(vcpu->arch.pio.guest_pages, nr_pages, VM_READ|VM_WRITE, - PAGE_KERNEL); - if (!q) { - free_pio_guest_pages(vcpu); - return -ENOMEM; - } - q += vcpu->arch.pio.guest_page_offset; bytes = vcpu->arch.pio.size * vcpu->arch.pio.cur_count; if (vcpu->arch.pio.in) - memcpy(q, p, bytes); + ret = kvm_write_guest_virt(q, p, bytes, vcpu); else - memcpy(p, q, bytes); - q -= vcpu->arch.pio.guest_page_offset; - vunmap(q); - free_pio_guest_pages(vcpu); - return 0; + ret = kvm_read_guest_virt(q, p, bytes, vcpu); + return ret; } int complete_pio(struct kvm_vcpu *vcpu) @@ -2469,7 +2561,6 @@ int kvm_emulate_pio(struct kvm_vcpu *vcpu, struct kvm_run *run, int in, vcpu->arch.pio.in = in; vcpu->arch.pio.string = 0; vcpu->arch.pio.down = 0; - vcpu->arch.pio.guest_page_offset = 0; vcpu->arch.pio.rep = 0; if (vcpu->run->io.direction == KVM_EXIT_IO_IN) @@ -2497,9 +2588,7 @@ int kvm_emulate_pio_string(struct kvm_vcpu *vcpu, struct kvm_run *run, int in, gva_t address, int rep, unsigned port) { unsigned now, in_page; - int i, ret = 0; - int nr_pages = 1; - struct page *page; + int ret = 0; struct kvm_io_device *pio_dev; vcpu->run->exit_reason = KVM_EXIT_IO; @@ -2511,7 +2600,6 @@ int kvm_emulate_pio_string(struct kvm_vcpu *vcpu, struct kvm_run *run, int in, vcpu->arch.pio.in = in; vcpu->arch.pio.string = 1; vcpu->arch.pio.down = down; - vcpu->arch.pio.guest_page_offset = offset_in_page(address); vcpu->arch.pio.rep = rep; if (vcpu->run->io.direction == KVM_EXIT_IO_IN) @@ -2531,15 +2619,8 @@ int kvm_emulate_pio_string(struct kvm_vcpu *vcpu, struct kvm_run *run, int in, else in_page = offset_in_page(address) + size; now = min(count, (unsigned long)in_page / size); - if (!now) { - /* - * String I/O straddles page boundary. Pin two guest pages - * so that we satisfy atomicity constraints. Do just one - * transaction to avoid complexity. - */ - nr_pages = 2; + if (!now) now = 1; - } if (down) { /* * String I/O in reverse. Yuck. Kill the guest, fix later. @@ -2554,15 +2635,7 @@ int kvm_emulate_pio_string(struct kvm_vcpu *vcpu, struct kvm_run *run, int in, if (vcpu->arch.pio.cur_count == vcpu->arch.pio.count) kvm_x86_ops->skip_emulated_instruction(vcpu); - for (i = 0; i < nr_pages; ++i) { - page = gva_to_page(vcpu, address + i * PAGE_SIZE); - vcpu->arch.pio.guest_pages[i] = page; - if (!page) { - kvm_inject_gp(vcpu, 0); - free_pio_guest_pages(vcpu); - return 1; - } - } + vcpu->arch.pio.guest_gva = address; pio_dev = vcpu_find_pio_dev(vcpu, port, vcpu->arch.pio.cur_count, @@ -2570,7 +2643,11 @@ int kvm_emulate_pio_string(struct kvm_vcpu *vcpu, struct kvm_run *run, int in, if (!vcpu->arch.pio.in) { /* string PIO write */ ret = pio_copy_data(vcpu); - if (ret >= 0 && pio_dev) { + if (ret == X86EMUL_PROPAGATE_FAULT) { + kvm_inject_gp(vcpu, 0); + return 1; + } + if (ret == 0 && pio_dev) { pio_string_write(pio_dev, vcpu); complete_pio(vcpu); if (vcpu->arch.pio.count == 0) @@ -2585,9 +2662,72 @@ int kvm_emulate_pio_string(struct kvm_vcpu *vcpu, struct kvm_run *run, int in, } EXPORT_SYMBOL_GPL(kvm_emulate_pio_string); +static void bounce_off(void *info) +{ + /* nothing */ +} + +static unsigned int ref_freq; +static unsigned long tsc_khz_ref; + +static int kvmclock_cpufreq_notifier(struct notifier_block *nb, unsigned long val, + void *data) +{ + struct cpufreq_freqs *freq = data; + struct kvm *kvm; + struct kvm_vcpu *vcpu; + int i, send_ipi = 0; + + if (!ref_freq) + ref_freq = freq->old; + + if (val == CPUFREQ_PRECHANGE && freq->old > freq->new) + return 0; + if (val == CPUFREQ_POSTCHANGE && freq->old < freq->new) + return 0; + per_cpu(cpu_tsc_khz, freq->cpu) = cpufreq_scale(tsc_khz_ref, ref_freq, freq->new); + + spin_lock(&kvm_lock); + list_for_each_entry(kvm, &vm_list, vm_list) { + for (i = 0; i < KVM_MAX_VCPUS; ++i) { + vcpu = kvm->vcpus[i]; + if (!vcpu) + continue; + if (vcpu->cpu != freq->cpu) + continue; + if (!kvm_request_guest_time_update(vcpu)) + continue; + if (vcpu->cpu != smp_processor_id()) + send_ipi++; + } + } + spin_unlock(&kvm_lock); + + if (freq->old < freq->new && send_ipi) { + /* + * We upscale the frequency. Must make the guest + * doesn't see old kvmclock values while running with + * the new frequency, otherwise we risk the guest sees + * time go backwards. + * + * In case we update the frequency for another cpu + * (which might be in guest context) send an interrupt + * to kick the cpu out of guest context. Next time + * guest context is entered kvmclock will be updated, + * so the guest will not see stale values. + */ + smp_call_function_single(freq->cpu, bounce_off, NULL, 1); + } + return 0; +} + +static struct notifier_block kvmclock_cpufreq_notifier_block = { + .notifier_call = kvmclock_cpufreq_notifier +}; + int kvm_arch_init(void *opaque) { - int r; + int r, cpu; struct kvm_x86_ops *ops = (struct kvm_x86_ops *)opaque; if (kvm_x86_ops) { @@ -2618,6 +2758,15 @@ int kvm_arch_init(void *opaque) kvm_mmu_set_base_ptes(PT_PRESENT_MASK); kvm_mmu_set_mask_ptes(PT_USER_MASK, PT_ACCESSED_MASK, PT_DIRTY_MASK, PT64_NX_MASK, 0, 0); + + for_each_possible_cpu(cpu) + per_cpu(cpu_tsc_khz, cpu) = tsc_khz; + if (!boot_cpu_has(X86_FEATURE_CONSTANT_TSC)) { + tsc_khz_ref = tsc_khz; + cpufreq_register_notifier(&kvmclock_cpufreq_notifier_block, + CPUFREQ_TRANSITION_NOTIFIER); + } + return 0; out: @@ -2825,25 +2974,20 @@ static int is_matching_cpuid_entry(struct kvm_cpuid_entry2 *e, if ((e->flags & KVM_CPUID_FLAG_SIGNIFCANT_INDEX) && e->index != index) return 0; if ((e->flags & KVM_CPUID_FLAG_STATEFUL_FUNC) && - !(e->flags & KVM_CPUID_FLAG_STATE_READ_NEXT)) + !(e->flags & KVM_CPUID_FLAG_STATE_READ_NEXT)) return 0; return 1; } -void kvm_emulate_cpuid(struct kvm_vcpu *vcpu) +struct kvm_cpuid_entry2 *kvm_find_cpuid_entry(struct kvm_vcpu *vcpu, + u32 function, u32 index) { int i; - u32 function, index; - struct kvm_cpuid_entry2 *e, *best; + struct kvm_cpuid_entry2 *best = NULL; - function = kvm_register_read(vcpu, VCPU_REGS_RAX); - index = kvm_register_read(vcpu, VCPU_REGS_RCX); - kvm_register_write(vcpu, VCPU_REGS_RAX, 0); - kvm_register_write(vcpu, VCPU_REGS_RBX, 0); - kvm_register_write(vcpu, VCPU_REGS_RCX, 0); - kvm_register_write(vcpu, VCPU_REGS_RDX, 0); - best = NULL; for (i = 0; i < vcpu->arch.cpuid_nent; ++i) { + struct kvm_cpuid_entry2 *e; + e = &vcpu->arch.cpuid_entries[i]; if (is_matching_cpuid_entry(e, function, index)) { if (e->flags & KVM_CPUID_FLAG_STATEFUL_FUNC) @@ -2858,6 +3002,21 @@ void kvm_emulate_cpuid(struct kvm_vcpu *vcpu) if (!best || e->function > best->function) best = e; } + return best; +} + +void kvm_emulate_cpuid(struct kvm_vcpu *vcpu) +{ + u32 function, index; + struct kvm_cpuid_entry2 *best; + + function = kvm_register_read(vcpu, VCPU_REGS_RAX); + index = kvm_register_read(vcpu, VCPU_REGS_RCX); + kvm_register_write(vcpu, VCPU_REGS_RAX, 0); + kvm_register_write(vcpu, VCPU_REGS_RBX, 0); + kvm_register_write(vcpu, VCPU_REGS_RCX, 0); + kvm_register_write(vcpu, VCPU_REGS_RDX, 0); + best = kvm_find_cpuid_entry(vcpu, function, index); if (best) { kvm_register_write(vcpu, VCPU_REGS_RAX, best->eax); kvm_register_write(vcpu, VCPU_REGS_RBX, best->ebx); @@ -2943,6 +3102,8 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) if (vcpu->requests) { if (test_and_clear_bit(KVM_REQ_MIGRATE_TIMER, &vcpu->requests)) __kvm_migrate_timers(vcpu); + if (test_and_clear_bit(KVM_REQ_KVMCLOCK_UPDATE, &vcpu->requests)) + kvm_write_guest_time(vcpu); if (test_and_clear_bit(KVM_REQ_MMU_SYNC, &vcpu->requests)) kvm_mmu_sync_roots(vcpu); if (test_and_clear_bit(KVM_REQ_TLB_FLUSH, &vcpu->requests)) @@ -2977,9 +3138,6 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) goto out; } - if (vcpu->guest_debug.enabled) - kvm_x86_ops->guest_debug_pre(vcpu); - vcpu->guest_mode = 1; /* * Make sure that guest_mode assignment won't happen after @@ -3000,10 +3158,34 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) kvm_guest_enter(); + get_debugreg(vcpu->arch.host_dr6, 6); + get_debugreg(vcpu->arch.host_dr7, 7); + if (unlikely(vcpu->arch.switch_db_regs)) { + get_debugreg(vcpu->arch.host_db[0], 0); + get_debugreg(vcpu->arch.host_db[1], 1); + get_debugreg(vcpu->arch.host_db[2], 2); + get_debugreg(vcpu->arch.host_db[3], 3); + + set_debugreg(0, 7); + set_debugreg(vcpu->arch.eff_db[0], 0); + set_debugreg(vcpu->arch.eff_db[1], 1); + set_debugreg(vcpu->arch.eff_db[2], 2); + set_debugreg(vcpu->arch.eff_db[3], 3); + } KVMTRACE_0D(VMENTRY, vcpu, entryexit); kvm_x86_ops->run(vcpu, kvm_run); + if (unlikely(vcpu->arch.switch_db_regs)) { + set_debugreg(0, 7); + set_debugreg(vcpu->arch.host_db[0], 0); + set_debugreg(vcpu->arch.host_db[1], 1); + set_debugreg(vcpu->arch.host_db[2], 2); + set_debugreg(vcpu->arch.host_db[3], 3); + } + set_debugreg(vcpu->arch.host_dr6, 6); + set_debugreg(vcpu->arch.host_dr7, 7); + vcpu->guest_mode = 0; local_irq_enable(); @@ -3190,7 +3372,7 @@ int kvm_arch_vcpu_ioctl_get_regs(struct kvm_vcpu *vcpu, struct kvm_regs *regs) /* * Don't leak debug flags in case they were set for guest debugging */ - if (vcpu->guest_debug.enabled && vcpu->guest_debug.singlestep) + if (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP) regs->rflags &= ~(X86_EFLAGS_TF | X86_EFLAGS_RF); vcpu_put(vcpu); @@ -3809,15 +3991,32 @@ int kvm_arch_vcpu_ioctl_set_sregs(struct kvm_vcpu *vcpu, return 0; } -int kvm_arch_vcpu_ioctl_debug_guest(struct kvm_vcpu *vcpu, - struct kvm_debug_guest *dbg) +int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu, + struct kvm_guest_debug *dbg) { - int r; + int i, r; vcpu_load(vcpu); + if ((dbg->control & (KVM_GUESTDBG_ENABLE | KVM_GUESTDBG_USE_HW_BP)) == + (KVM_GUESTDBG_ENABLE | KVM_GUESTDBG_USE_HW_BP)) { + for (i = 0; i < KVM_NR_DB_REGS; ++i) + vcpu->arch.eff_db[i] = dbg->arch.debugreg[i]; + vcpu->arch.switch_db_regs = + (dbg->arch.debugreg[7] & DR7_BP_EN_MASK); + } else { + for (i = 0; i < KVM_NR_DB_REGS; i++) + vcpu->arch.eff_db[i] = vcpu->arch.db[i]; + vcpu->arch.switch_db_regs = (vcpu->arch.dr7 & DR7_BP_EN_MASK); + } + r = kvm_x86_ops->set_guest_debug(vcpu, dbg); + if (dbg->control & KVM_GUESTDBG_INJECT_DB) + kvm_queue_exception(vcpu, DB_VECTOR); + else if (dbg->control & KVM_GUESTDBG_INJECT_BP) + kvm_queue_exception(vcpu, BP_VECTOR); + vcpu_put(vcpu); return r; @@ -4005,6 +4204,11 @@ int kvm_arch_vcpu_reset(struct kvm_vcpu *vcpu) vcpu->arch.nmi_pending = false; vcpu->arch.nmi_injected = false; + vcpu->arch.switch_db_regs = 0; + memset(vcpu->arch.db, 0, sizeof(vcpu->arch.db)); + vcpu->arch.dr6 = DR6_FIXED_1; + vcpu->arch.dr7 = DR7_FIXED_1; + return kvm_x86_ops->vcpu_reset(vcpu); } @@ -4098,6 +4302,8 @@ struct kvm *kvm_arch_create_vm(void) /* Reserve bit 0 of irq_sources_bitmap for userspace irq source */ set_bit(KVM_USERSPACE_IRQ_SOURCE_ID, &kvm->arch.irq_sources_bitmap); + rdtscll(kvm->arch.vm_init_tsc); + return kvm; } @@ -4127,9 +4333,13 @@ static void kvm_free_vcpus(struct kvm *kvm) } -void kvm_arch_destroy_vm(struct kvm *kvm) +void kvm_arch_sync_events(struct kvm *kvm) { kvm_free_all_assigned_devices(kvm); +} + +void kvm_arch_destroy_vm(struct kvm *kvm) +{ kvm_iommu_unmap_guest(kvm); kvm_free_pit(kvm); kfree(kvm->arch.vpic); diff --git a/arch/x86/kvm/x86_emulate.c b/arch/x86/kvm/x86_emulate.c index d174db7a337..ca91749d208 100644 --- a/arch/x86/kvm/x86_emulate.c +++ b/arch/x86/kvm/x86_emulate.c @@ -178,7 +178,7 @@ static u32 opcode_table[256] = { 0, ImplicitOps | Stack, 0, 0, ByteOp | DstMem | SrcImm | ModRM | Mov, DstMem | SrcImm | ModRM | Mov, /* 0xC8 - 0xCF */ - 0, 0, 0, 0, 0, 0, 0, 0, + 0, 0, 0, ImplicitOps | Stack, 0, 0, 0, 0, /* 0xD0 - 0xD7 */ ByteOp | DstMem | SrcImplicit | ModRM, DstMem | SrcImplicit | ModRM, ByteOp | DstMem | SrcImplicit | ModRM, DstMem | SrcImplicit | ModRM, @@ -1136,18 +1136,19 @@ static inline void emulate_push(struct x86_emulate_ctxt *ctxt) } static int emulate_pop(struct x86_emulate_ctxt *ctxt, - struct x86_emulate_ops *ops) + struct x86_emulate_ops *ops, + void *dest, int len) { struct decode_cache *c = &ctxt->decode; int rc; rc = ops->read_emulated(register_address(c, ss_base(ctxt), c->regs[VCPU_REGS_RSP]), - &c->src.val, c->src.bytes, ctxt->vcpu); + dest, len, ctxt->vcpu); if (rc != 0) return rc; - register_address_increment(c, &c->regs[VCPU_REGS_RSP], c->src.bytes); + register_address_increment(c, &c->regs[VCPU_REGS_RSP], len); return rc; } @@ -1157,11 +1158,9 @@ static inline int emulate_grp1a(struct x86_emulate_ctxt *ctxt, struct decode_cache *c = &ctxt->decode; int rc; - c->src.bytes = c->dst.bytes; - rc = emulate_pop(ctxt, ops); + rc = emulate_pop(ctxt, ops, &c->dst.val, c->dst.bytes); if (rc != 0) return rc; - c->dst.val = c->src.val; return 0; } @@ -1279,6 +1278,25 @@ static inline int emulate_grp9(struct x86_emulate_ctxt *ctxt, return 0; } +static int emulate_ret_far(struct x86_emulate_ctxt *ctxt, + struct x86_emulate_ops *ops) +{ + struct decode_cache *c = &ctxt->decode; + int rc; + unsigned long cs; + + rc = emulate_pop(ctxt, ops, &c->eip, c->op_bytes); + if (rc) + return rc; + if (c->op_bytes == 4) + c->eip = (u32)c->eip; + rc = emulate_pop(ctxt, ops, &cs, c->op_bytes); + if (rc) + return rc; + rc = kvm_load_segment_descriptor(ctxt->vcpu, (u16)cs, 1, VCPU_SREG_CS); + return rc; +} + static inline int writeback(struct x86_emulate_ctxt *ctxt, struct x86_emulate_ops *ops) { @@ -1467,11 +1485,9 @@ special_insn: break; case 0x58 ... 0x5f: /* pop reg */ pop_instruction: - c->src.bytes = c->op_bytes; - rc = emulate_pop(ctxt, ops); + rc = emulate_pop(ctxt, ops, &c->dst.val, c->op_bytes); if (rc != 0) goto done; - c->dst.val = c->src.val; break; case 0x63: /* movsxd */ if (ctxt->mode != X86EMUL_MODE_PROT64) @@ -1738,6 +1754,11 @@ special_insn: mov: c->dst.val = c->src.val; break; + case 0xcb: /* ret far */ + rc = emulate_ret_far(ctxt, ops); + if (rc) + goto done; + break; case 0xd0 ... 0xd1: /* Grp2 */ c->src.val = 1; emulate_grp2(ctxt); @@ -1908,11 +1929,16 @@ twobyte_insn: c->dst.type = OP_NONE; break; case 3: /* lidt/vmmcall */ - if (c->modrm_mod == 3 && c->modrm_rm == 1) { - rc = kvm_fix_hypercall(ctxt->vcpu); - if (rc) - goto done; - kvm_emulate_hypercall(ctxt->vcpu); + if (c->modrm_mod == 3) { + switch (c->modrm_rm) { + case 1: + rc = kvm_fix_hypercall(ctxt->vcpu); + if (rc) + goto done; + break; + default: + goto cannot_emulate; + } } else { rc = read_descriptor(ctxt, ops, c->src.ptr, &size, &address, diff --git a/arch/x86/lguest/boot.c b/arch/x86/lguest/boot.c index a7ed208f81e..960a8d9c049 100644 --- a/arch/x86/lguest/boot.c +++ b/arch/x86/lguest/boot.c @@ -343,6 +343,11 @@ static void lguest_cpuid(unsigned int *ax, unsigned int *bx, * flush_tlb_user() for both user and kernel mappings unless * the Page Global Enable (PGE) feature bit is set. */ *dx |= 0x00002000; + /* We also lie, and say we're family id 5. 6 or greater + * leads to a rdmsr in early_init_intel which we can't handle. + * Family ID is returned as bits 8-12 in ax. */ + *ax &= 0xFFFFF0FF; + *ax |= 0x00000500; break; case 0x80000000: /* Futureproof this a little: if they ask how much extended @@ -589,19 +594,21 @@ static void __init lguest_init_IRQ(void) /* Some systems map "vectors" to interrupts weirdly. Lguest has * a straightforward 1 to 1 mapping, so force that here. */ __get_cpu_var(vector_irq)[vector] = i; - if (vector != SYSCALL_VECTOR) { - set_intr_gate(vector, - interrupt[vector-FIRST_EXTERNAL_VECTOR]); - set_irq_chip_and_handler_name(i, &lguest_irq_controller, - handle_level_irq, - "level"); - } + if (vector != SYSCALL_VECTOR) + set_intr_gate(vector, interrupt[i]); } /* This call is required to set up for 4k stacks, where we have * separate stacks for hard and soft interrupts. */ irq_ctx_init(smp_processor_id()); } +void lguest_setup_irq(unsigned int irq) +{ + irq_to_desc_alloc_cpu(irq, 0); + set_irq_chip_and_handler_name(irq, &lguest_irq_controller, + handle_level_irq, "level"); +} + /* * Time. * @@ -931,7 +938,7 @@ static void lguest_restart(char *reason) * that we can fit comfortably. * * First we need assembly templates of each of the patchable Guest operations, - * and these are in lguest_asm.S. */ + * and these are in i386_head.S. */ /*G:060 We construct a table from the assembler templates: */ static const struct lguest_insns @@ -1093,7 +1100,7 @@ __init void lguest_init(void) acpi_ht = 0; #endif - /* We set the perferred console to "hvc". This is the "hypervisor + /* We set the preferred console to "hvc". This is the "hypervisor * virtual console" driver written by the PowerPC people, which we also * adapted for lguest's use. */ add_preferred_console("hvc", 0, NULL); diff --git a/arch/x86/lib/usercopy_32.c b/arch/x86/lib/usercopy_32.c index 4a20b2f9a38..7c8ca91bb9e 100644 --- a/arch/x86/lib/usercopy_32.c +++ b/arch/x86/lib/usercopy_32.c @@ -56,7 +56,7 @@ do { \ " jmp 2b\n" \ ".previous\n" \ _ASM_EXTABLE(0b,3b) \ - : "=d"(res), "=c"(count), "=&a" (__d0), "=&S" (__d1), \ + : "=&d"(res), "=&c"(count), "=&a" (__d0), "=&S" (__d1), \ "=&D" (__d2) \ : "i"(-EFAULT), "0"(count), "1"(count), "3"(src), "4"(dst) \ : "memory"); \ @@ -218,7 +218,7 @@ long strnlen_user(const char __user *s, long n) " .align 4\n" " .long 0b,2b\n" ".previous" - :"=r" (n), "=D" (s), "=a" (res), "=c" (tmp) + :"=&r" (n), "=&D" (s), "=&a" (res), "=&c" (tmp) :"0" (n), "1" (s), "2" (0), "3" (mask) :"cc"); return res & mask; diff --git a/arch/x86/lib/usercopy_64.c b/arch/x86/lib/usercopy_64.c index 64d6c84e635..ec13cb5f17e 100644 --- a/arch/x86/lib/usercopy_64.c +++ b/arch/x86/lib/usercopy_64.c @@ -32,7 +32,7 @@ do { \ " jmp 2b\n" \ ".previous\n" \ _ASM_EXTABLE(0b,3b) \ - : "=r"(res), "=c"(count), "=&a" (__d0), "=&S" (__d1), \ + : "=&r"(res), "=&c"(count), "=&a" (__d0), "=&S" (__d1), \ "=&D" (__d2) \ : "i"(-EFAULT), "0"(count), "1"(count), "3"(src), "4"(dst) \ : "memory"); \ @@ -86,7 +86,7 @@ unsigned long __clear_user(void __user *addr, unsigned long size) ".previous\n" _ASM_EXTABLE(0b,3b) _ASM_EXTABLE(1b,2b) - : [size8] "=c"(size), [dst] "=&D" (__d0) + : [size8] "=&c"(size), [dst] "=&D" (__d0) : [size1] "r"(size & 7), "[size8]" (size / 8), "[dst]"(addr), [zero] "r" (0UL), [eight] "r" (8UL)); return size; diff --git a/arch/x86/mach-default/setup.c b/arch/x86/mach-default/setup.c index df167f26562..50b59187112 100644 --- a/arch/x86/mach-default/setup.c +++ b/arch/x86/mach-default/setup.c @@ -38,6 +38,15 @@ void __init pre_intr_init_hook(void) init_ISA_irqs(); } +/* + * IRQ2 is cascade interrupt to second interrupt controller + */ +static struct irqaction irq2 = { + .handler = no_action, + .mask = CPU_MASK_NONE, + .name = "cascade", +}; + /** * intr_init_hook - post gate setup interrupt initialisation * @@ -53,6 +62,9 @@ void __init intr_init_hook(void) if (x86_quirks->arch_intr_init()) return; } + if (!acpi_ioapic) + setup_irq(2, &irq2); + } /** @@ -84,7 +96,7 @@ void __init trap_init_hook(void) static struct irqaction irq0 = { .handler = timer_interrupt, - .flags = IRQF_DISABLED | IRQF_NOBALANCING | IRQF_IRQPOLL, + .flags = IRQF_DISABLED | IRQF_NOBALANCING | IRQF_IRQPOLL | IRQF_TIMER, .mask = CPU_MASK_NONE, .name = "timer" }; diff --git a/arch/x86/mach-voyager/setup.c b/arch/x86/mach-voyager/setup.c index a580b9562e7..8e5118371f0 100644 --- a/arch/x86/mach-voyager/setup.c +++ b/arch/x86/mach-voyager/setup.c @@ -33,20 +33,30 @@ void __init intr_init_hook(void) setup_irq(2, &irq2); } -void __init pre_setup_arch_hook(void) +static void voyager_disable_tsc(void) { /* Voyagers run their CPUs from independent clocks, so disable * the TSC code because we can't sync them */ setup_clear_cpu_cap(X86_FEATURE_TSC); } +void __init pre_setup_arch_hook(void) +{ + voyager_disable_tsc(); +} + +void __init pre_time_init_hook(void) +{ + voyager_disable_tsc(); +} + void __init trap_init_hook(void) { } static struct irqaction irq0 = { .handler = timer_interrupt, - .flags = IRQF_DISABLED | IRQF_NOBALANCING | IRQF_IRQPOLL, + .flags = IRQF_DISABLED | IRQF_NOBALANCING | IRQF_IRQPOLL | IRQF_TIMER, .mask = CPU_MASK_NONE, .name = "timer" }; diff --git a/arch/x86/mach-voyager/voyager_smp.c b/arch/x86/mach-voyager/voyager_smp.c index 9840b7ec749..b9cc84a2a4f 100644 --- a/arch/x86/mach-voyager/voyager_smp.c +++ b/arch/x86/mach-voyager/voyager_smp.c @@ -65,7 +65,7 @@ static volatile unsigned long smp_invalidate_needed; /* Bitmask of CPUs present in the system - exported by i386_syms.c, used * by scheduler but indexed physically */ -cpumask_t phys_cpu_present_map = CPU_MASK_NONE; +static cpumask_t voyager_phys_cpu_present_map = CPU_MASK_NONE; /* The internal functions */ static void send_CPI(__u32 cpuset, __u8 cpi); @@ -81,7 +81,7 @@ static void enable_local_vic_irq(unsigned int irq); static void disable_local_vic_irq(unsigned int irq); static void before_handle_vic_irq(unsigned int irq); static void after_handle_vic_irq(unsigned int irq); -static void set_vic_irq_affinity(unsigned int irq, cpumask_t mask); +static void set_vic_irq_affinity(unsigned int irq, const struct cpumask *mask); static void ack_vic_irq(unsigned int irq); static void vic_enable_cpi(void); static void do_boot_cpu(__u8 cpuid); @@ -211,8 +211,6 @@ static __u32 cpu_booted_map; static cpumask_t smp_commenced_mask = CPU_MASK_NONE; /* This is for the new dynamic CPU boot code */ -cpumask_t cpu_callin_map = CPU_MASK_NONE; -cpumask_t cpu_callout_map = CPU_MASK_NONE; /* The per processor IRQ masks (these are usually kept in sync) */ static __u16 vic_irq_mask[NR_CPUS] __cacheline_aligned; @@ -368,19 +366,19 @@ void __init find_smp_config(void) /* set up everything for just this CPU, we can alter * this as we start the other CPUs later */ /* now get the CPU disposition from the extended CMOS */ - cpus_addr(phys_cpu_present_map)[0] = + cpus_addr(voyager_phys_cpu_present_map)[0] = voyager_extended_cmos_read(VOYAGER_PROCESSOR_PRESENT_MASK); - cpus_addr(phys_cpu_present_map)[0] |= + cpus_addr(voyager_phys_cpu_present_map)[0] |= voyager_extended_cmos_read(VOYAGER_PROCESSOR_PRESENT_MASK + 1) << 8; - cpus_addr(phys_cpu_present_map)[0] |= + cpus_addr(voyager_phys_cpu_present_map)[0] |= voyager_extended_cmos_read(VOYAGER_PROCESSOR_PRESENT_MASK + 2) << 16; - cpus_addr(phys_cpu_present_map)[0] |= + cpus_addr(voyager_phys_cpu_present_map)[0] |= voyager_extended_cmos_read(VOYAGER_PROCESSOR_PRESENT_MASK + 3) << 24; - cpu_possible_map = phys_cpu_present_map; - printk("VOYAGER SMP: phys_cpu_present_map = 0x%lx\n", - cpus_addr(phys_cpu_present_map)[0]); + init_cpu_possible(&voyager_phys_cpu_present_map); + printk("VOYAGER SMP: voyager_phys_cpu_present_map = 0x%lx\n", + cpus_addr(voyager_phys_cpu_present_map)[0]); /* Here we set up the VIC to enable SMP */ /* enable the CPIs by writing the base vector to their register */ outb(VIC_DEFAULT_CPI_BASE, VIC_CPI_BASE_REGISTER); @@ -630,15 +628,15 @@ void __init smp_boot_cpus(void) /* now that the cat has probed the Voyager System Bus, sanity * check the cpu map */ if (((voyager_quad_processors | voyager_extended_vic_processors) - & cpus_addr(phys_cpu_present_map)[0]) != - cpus_addr(phys_cpu_present_map)[0]) { + & cpus_addr(voyager_phys_cpu_present_map)[0]) != + cpus_addr(voyager_phys_cpu_present_map)[0]) { /* should panic */ printk("\n\n***WARNING*** " "Sanity check of CPU present map FAILED\n"); } } else if (voyager_level == 4) voyager_extended_vic_processors = - cpus_addr(phys_cpu_present_map)[0]; + cpus_addr(voyager_phys_cpu_present_map)[0]; /* this sets up the idle task to run on the current cpu */ voyager_extended_cpus = 1; @@ -672,7 +670,7 @@ void __init smp_boot_cpus(void) /* loop over all the extended VIC CPUs and boot them. The * Quad CPUs must be bootstrapped by their extended VIC cpu */ for (i = 0; i < nr_cpu_ids; i++) { - if (i == boot_cpu_id || !cpu_isset(i, phys_cpu_present_map)) + if (i == boot_cpu_id || !cpu_isset(i, voyager_phys_cpu_present_map)) continue; do_boot_cpu(i); /* This udelay seems to be needed for the Quad boots @@ -1599,16 +1597,16 @@ static void after_handle_vic_irq(unsigned int irq) * change the mask and then do an interrupt enable CPI to re-enable on * the selected processors */ -void set_vic_irq_affinity(unsigned int irq, cpumask_t mask) +void set_vic_irq_affinity(unsigned int irq, const struct cpumask *mask) { /* Only extended processors handle interrupts */ unsigned long real_mask; unsigned long irq_mask = 1 << irq; int cpu; - real_mask = cpus_addr(mask)[0] & voyager_extended_vic_processors; + real_mask = cpus_addr(*mask)[0] & voyager_extended_vic_processors; - if (cpus_addr(mask)[0] == 0) + if (cpus_addr(*mask)[0] == 0) /* can't have no CPUs to accept the interrupt -- extremely * bad things will happen */ return; @@ -1750,10 +1748,11 @@ static void __cpuinit voyager_smp_prepare_boot_cpu(void) init_gdt(smp_processor_id()); switch_to_new_gdt(); - cpu_set(smp_processor_id(), cpu_online_map); - cpu_set(smp_processor_id(), cpu_callout_map); - cpu_set(smp_processor_id(), cpu_possible_map); - cpu_set(smp_processor_id(), cpu_present_map); + cpu_online_map = cpumask_of_cpu(smp_processor_id()); + cpu_callout_map = cpumask_of_cpu(smp_processor_id()); + cpu_callin_map = CPU_MASK_NONE; + cpu_present_map = cpumask_of_cpu(smp_processor_id()); + } static int __cpuinit voyager_cpu_up(unsigned int cpu) @@ -1783,9 +1782,9 @@ void __init smp_setup_processor_id(void) x86_write_percpu(cpu_number, hard_smp_processor_id()); } -static void voyager_send_call_func(cpumask_t callmask) +static void voyager_send_call_func(const struct cpumask *callmask) { - __u32 mask = cpus_addr(callmask)[0] & ~(1 << smp_processor_id()); + __u32 mask = cpus_addr(*callmask)[0] & ~(1 << smp_processor_id()); send_CPI(mask, VIC_CALL_FUNCTION_CPI); } diff --git a/arch/x86/math-emu/fpu_aux.c b/arch/x86/math-emu/fpu_aux.c index 491e737ce54..aa098708877 100644 --- a/arch/x86/math-emu/fpu_aux.c +++ b/arch/x86/math-emu/fpu_aux.c @@ -30,20 +30,29 @@ static void fclex(void) } /* Needs to be externally visible */ -void finit(void) +void finit_task(struct task_struct *tsk) { - control_word = 0x037f; - partial_status = 0; - top = 0; /* We don't keep top in the status word internally. */ - fpu_tag_word = 0xffff; + struct i387_soft_struct *soft = &tsk->thread.xstate->soft; + struct address *oaddr, *iaddr; + soft->cwd = 0x037f; + soft->swd = 0; + soft->ftop = 0; /* We don't keep top in the status word internally. */ + soft->twd = 0xffff; /* The behaviour is different from that detailed in Section 15.1.6 of the Intel manual */ - operand_address.offset = 0; - operand_address.selector = 0; - instruction_address.offset = 0; - instruction_address.selector = 0; - instruction_address.opcode = 0; - no_ip_update = 1; + oaddr = (struct address *)&soft->foo; + oaddr->offset = 0; + oaddr->selector = 0; + iaddr = (struct address *)&soft->fip; + iaddr->offset = 0; + iaddr->selector = 0; + iaddr->opcode = 0; + soft->no_update = 1; +} + +void finit(void) +{ + finit_task(current); } /* diff --git a/arch/x86/math-emu/fpu_entry.c b/arch/x86/math-emu/fpu_entry.c index c7b06feb139..5d87f586f8d 100644 --- a/arch/x86/math-emu/fpu_entry.c +++ b/arch/x86/math-emu/fpu_entry.c @@ -131,7 +131,7 @@ u_char emulating = 0; static int valid_prefix(u_char *Byte, u_char __user ** fpu_eip, overrides * override); -asmlinkage void math_emulate(long arg) +void math_emulate(struct math_emu_info *info) { u_char FPU_modrm, byte1; unsigned short code; @@ -161,7 +161,7 @@ asmlinkage void math_emulate(long arg) RE_ENTRANT_CHECK_ON; #endif /* RE_ENTRANT_CHECKING */ - SETUP_DATA_AREA(arg); + FPU_info = info; FPU_ORIG_EIP = FPU_EIP; @@ -659,7 +659,7 @@ static int valid_prefix(u_char *Byte, u_char __user **fpu_eip, } } -void math_abort(struct info *info, unsigned int signal) +void math_abort(struct math_emu_info *info, unsigned int signal) { FPU_EIP = FPU_ORIG_EIP; current->thread.trap_no = 16; diff --git a/arch/x86/math-emu/fpu_proto.h b/arch/x86/math-emu/fpu_proto.h index aa49b6a0d85..9779df436b7 100644 --- a/arch/x86/math-emu/fpu_proto.h +++ b/arch/x86/math-emu/fpu_proto.h @@ -51,8 +51,8 @@ extern void ffreep(void); extern void fst_i_(void); extern void fstp_i(void); /* fpu_entry.c */ -asmlinkage extern void math_emulate(long arg); -extern void math_abort(struct info *info, unsigned int signal); +extern void math_emulate(struct math_emu_info *info); +extern void math_abort(struct math_emu_info *info, unsigned int signal); /* fpu_etc.c */ extern void FPU_etc(void); /* fpu_tags.c */ diff --git a/arch/x86/math-emu/fpu_system.h b/arch/x86/math-emu/fpu_system.h index 13488fa153e..50fa0ec2c8a 100644 --- a/arch/x86/math-emu/fpu_system.h +++ b/arch/x86/math-emu/fpu_system.h @@ -16,10 +16,6 @@ #include <linux/kernel.h> #include <linux/mm.h> -/* This sets the pointer FPU_info to point to the argument part - of the stack frame of math_emulate() */ -#define SETUP_DATA_AREA(arg) FPU_info = (struct info *) &arg - /* s is always from a cpu register, and the cpu does bounds checking * during register load --> no further bounds checks needed */ #define LDT_DESCRIPTOR(s) (((struct desc_struct *)current->mm->context.ldt)[(s) >> 3]) @@ -38,12 +34,12 @@ #define I387 (current->thread.xstate) #define FPU_info (I387->soft.info) -#define FPU_CS (*(unsigned short *) &(FPU_info->___cs)) -#define FPU_SS (*(unsigned short *) &(FPU_info->___ss)) -#define FPU_DS (*(unsigned short *) &(FPU_info->___ds)) -#define FPU_EAX (FPU_info->___eax) -#define FPU_EFLAGS (FPU_info->___eflags) -#define FPU_EIP (FPU_info->___eip) +#define FPU_CS (*(unsigned short *) &(FPU_info->regs->cs)) +#define FPU_SS (*(unsigned short *) &(FPU_info->regs->ss)) +#define FPU_DS (*(unsigned short *) &(FPU_info->regs->ds)) +#define FPU_EAX (FPU_info->regs->ax) +#define FPU_EFLAGS (FPU_info->regs->flags) +#define FPU_EIP (FPU_info->regs->ip) #define FPU_ORIG_EIP (FPU_info->___orig_eip) #define FPU_lookahead (I387->soft.lookahead) diff --git a/arch/x86/math-emu/get_address.c b/arch/x86/math-emu/get_address.c index d701e2b39e4..420b3b6e391 100644 --- a/arch/x86/math-emu/get_address.c +++ b/arch/x86/math-emu/get_address.c @@ -29,46 +29,43 @@ #define FPU_WRITE_BIT 0x10 static int reg_offset[] = { - offsetof(struct info, ___eax), - offsetof(struct info, ___ecx), - offsetof(struct info, ___edx), - offsetof(struct info, ___ebx), - offsetof(struct info, ___esp), - offsetof(struct info, ___ebp), - offsetof(struct info, ___esi), - offsetof(struct info, ___edi) + offsetof(struct pt_regs, ax), + offsetof(struct pt_regs, cx), + offsetof(struct pt_regs, dx), + offsetof(struct pt_regs, bx), + offsetof(struct pt_regs, sp), + offsetof(struct pt_regs, bp), + offsetof(struct pt_regs, si), + offsetof(struct pt_regs, di) }; -#define REG_(x) (*(long *)(reg_offset[(x)]+(u_char *) FPU_info)) +#define REG_(x) (*(long *)(reg_offset[(x)] + (u_char *)FPU_info->regs)) static int reg_offset_vm86[] = { - offsetof(struct info, ___cs), - offsetof(struct info, ___vm86_ds), - offsetof(struct info, ___vm86_es), - offsetof(struct info, ___vm86_fs), - offsetof(struct info, ___vm86_gs), - offsetof(struct info, ___ss), - offsetof(struct info, ___vm86_ds) + offsetof(struct pt_regs, cs), + offsetof(struct kernel_vm86_regs, ds), + offsetof(struct kernel_vm86_regs, es), + offsetof(struct kernel_vm86_regs, fs), + offsetof(struct kernel_vm86_regs, gs), + offsetof(struct pt_regs, ss), + offsetof(struct kernel_vm86_regs, ds) }; #define VM86_REG_(x) (*(unsigned short *) \ - (reg_offset_vm86[((unsigned)x)]+(u_char *) FPU_info)) - -/* This dummy, gs is not saved on the stack. */ -#define ___GS ___ds + (reg_offset_vm86[((unsigned)x)] + (u_char *)FPU_info->regs)) static int reg_offset_pm[] = { - offsetof(struct info, ___cs), - offsetof(struct info, ___ds), - offsetof(struct info, ___es), - offsetof(struct info, ___fs), - offsetof(struct info, ___GS), - offsetof(struct info, ___ss), - offsetof(struct info, ___ds) + offsetof(struct pt_regs, cs), + offsetof(struct pt_regs, ds), + offsetof(struct pt_regs, es), + offsetof(struct pt_regs, fs), + offsetof(struct pt_regs, ds), /* dummy, not saved on stack */ + offsetof(struct pt_regs, ss), + offsetof(struct pt_regs, ds) }; #define PM_REG_(x) (*(unsigned short *) \ - (reg_offset_pm[((unsigned)x)]+(u_char *) FPU_info)) + (reg_offset_pm[((unsigned)x)] + (u_char *)FPU_info->regs)) /* Decode the SIB byte. This function assumes mod != 0 */ static int sib(int mod, unsigned long *fpu_eip) @@ -349,34 +346,34 @@ void __user *FPU_get_address_16(u_char FPU_modrm, unsigned long *fpu_eip, } switch (rm) { case 0: - address += FPU_info->___ebx + FPU_info->___esi; + address += FPU_info->regs->bx + FPU_info->regs->si; break; case 1: - address += FPU_info->___ebx + FPU_info->___edi; + address += FPU_info->regs->bx + FPU_info->regs->di; break; case 2: - address += FPU_info->___ebp + FPU_info->___esi; + address += FPU_info->regs->bp + FPU_info->regs->si; if (addr_modes.override.segment == PREFIX_DEFAULT) addr_modes.override.segment = PREFIX_SS_; break; case 3: - address += FPU_info->___ebp + FPU_info->___edi; + address += FPU_info->regs->bp + FPU_info->regs->di; if (addr_modes.override.segment == PREFIX_DEFAULT) addr_modes.override.segment = PREFIX_SS_; break; case 4: - address += FPU_info->___esi; + address += FPU_info->regs->si; break; case 5: - address += FPU_info->___edi; + address += FPU_info->regs->di; break; case 6: - address += FPU_info->___ebp; + address += FPU_info->regs->bp; if (addr_modes.override.segment == PREFIX_DEFAULT) addr_modes.override.segment = PREFIX_SS_; break; case 7: - address += FPU_info->___ebx; + address += FPU_info->regs->bx; break; } diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c index 90dfae511a4..c76ef1d701c 100644 --- a/arch/x86/mm/fault.c +++ b/arch/x86/mm/fault.c @@ -603,8 +603,6 @@ void __kprobes do_page_fault(struct pt_regs *regs, unsigned long error_code) si_code = SEGV_MAPERR; - if (notify_page_fault(regs)) - return; if (unlikely(kmmio_fault(regs, address))) return; @@ -634,6 +632,9 @@ void __kprobes do_page_fault(struct pt_regs *regs, unsigned long error_code) if (spurious_fault(address, error_code)) return; + /* kprobes don't want to hook the spurious faults. */ + if (notify_page_fault(regs)) + return; /* * Don't take the mm semaphore here. If we fixup a prefetch * fault we could otherwise deadlock. @@ -641,6 +642,9 @@ void __kprobes do_page_fault(struct pt_regs *regs, unsigned long error_code) goto bad_area_nosemaphore; } + /* kprobes don't want to hook the spurious faults. */ + if (notify_page_fault(regs)) + return; /* * It's safe to allow irq's after cr2 has been saved and the diff --git a/arch/x86/mm/init_32.c b/arch/x86/mm/init_32.c index 88f1b10de3b..2cef0507441 100644 --- a/arch/x86/mm/init_32.c +++ b/arch/x86/mm/init_32.c @@ -138,6 +138,47 @@ static pte_t * __init one_page_table_init(pmd_t *pmd) return pte_offset_kernel(pmd, 0); } +static pte_t *__init page_table_kmap_check(pte_t *pte, pmd_t *pmd, + unsigned long vaddr, pte_t *lastpte) +{ +#ifdef CONFIG_HIGHMEM + /* + * Something (early fixmap) may already have put a pte + * page here, which causes the page table allocation + * to become nonlinear. Attempt to fix it, and if it + * is still nonlinear then we have to bug. + */ + int pmd_idx_kmap_begin = fix_to_virt(FIX_KMAP_END) >> PMD_SHIFT; + int pmd_idx_kmap_end = fix_to_virt(FIX_KMAP_BEGIN) >> PMD_SHIFT; + + if (pmd_idx_kmap_begin != pmd_idx_kmap_end + && (vaddr >> PMD_SHIFT) >= pmd_idx_kmap_begin + && (vaddr >> PMD_SHIFT) <= pmd_idx_kmap_end + && ((__pa(pte) >> PAGE_SHIFT) < table_start + || (__pa(pte) >> PAGE_SHIFT) >= table_end)) { + pte_t *newpte; + int i; + + BUG_ON(after_init_bootmem); + newpte = alloc_low_page(); + for (i = 0; i < PTRS_PER_PTE; i++) + set_pte(newpte + i, pte[i]); + + paravirt_alloc_pte(&init_mm, __pa(newpte) >> PAGE_SHIFT); + set_pmd(pmd, __pmd(__pa(newpte)|_PAGE_TABLE)); + BUG_ON(newpte != pte_offset_kernel(pmd, 0)); + __flush_tlb_all(); + + paravirt_release_pte(__pa(pte) >> PAGE_SHIFT); + pte = newpte; + } + BUG_ON(vaddr < fix_to_virt(FIX_KMAP_BEGIN - 1) + && vaddr > fix_to_virt(FIX_KMAP_END) + && lastpte && lastpte + PTRS_PER_PTE != pte); +#endif + return pte; +} + /* * This function initializes a certain range of kernel virtual memory * with new bootmem page tables, everywhere page tables are missing in @@ -154,6 +195,7 @@ page_table_range_init(unsigned long start, unsigned long end, pgd_t *pgd_base) unsigned long vaddr; pgd_t *pgd; pmd_t *pmd; + pte_t *pte = NULL; vaddr = start; pgd_idx = pgd_index(vaddr); @@ -165,7 +207,8 @@ page_table_range_init(unsigned long start, unsigned long end, pgd_t *pgd_base) pmd = pmd + pmd_index(vaddr); for (; (pmd_idx < PTRS_PER_PMD) && (vaddr != end); pmd++, pmd_idx++) { - one_page_table_init(pmd); + pte = page_table_kmap_check(one_page_table_init(pmd), + pmd, vaddr, pte); vaddr += PMD_SIZE; } @@ -508,7 +551,6 @@ static void __init early_ioremap_page_table_range_init(pgd_t *pgd_base) * Fixed mappings, only the page table structure has to be * created - mappings will be set by set_fixmap(): */ - early_ioremap_clear(); vaddr = __fix_to_virt(__end_of_fixed_addresses - 1) & PMD_MASK; end = (FIXADDR_TOP + PMD_SIZE - 1) & PMD_MASK; page_table_range_init(vaddr, end, pgd_base); @@ -801,7 +843,7 @@ static void __init find_early_table_space(unsigned long end, int use_pse) tables += PAGE_ALIGN(ptes * sizeof(pte_t)); /* for fixmap */ - tables += PAGE_SIZE * 2; + tables += PAGE_ALIGN(__end_of_fixed_addresses * sizeof(pte_t)); /* * RED-PEN putting page tables only on node 0 could diff --git a/arch/x86/mm/init_64.c b/arch/x86/mm/init_64.c index 23f68e77ad1..b1352250096 100644 --- a/arch/x86/mm/init_64.c +++ b/arch/x86/mm/init_64.c @@ -596,7 +596,7 @@ static void __init init_gbpages(void) direct_gbpages = 0; } -static unsigned long __init kernel_physical_mapping_init(unsigned long start, +static unsigned long __meminit kernel_physical_mapping_init(unsigned long start, unsigned long end, unsigned long page_size_mask) { @@ -714,6 +714,8 @@ unsigned long __init_refok init_memory_mapping(unsigned long start, pos = start_pfn << PAGE_SHIFT; end_pfn = ((pos + (PMD_SIZE - 1)) >> PMD_SHIFT) << (PMD_SHIFT - PAGE_SHIFT); + if (end_pfn > (end >> PAGE_SHIFT)) + end_pfn = end >> PAGE_SHIFT; if (start_pfn < end_pfn) { nr_range = save_mr(mr, nr_range, start_pfn, end_pfn, 0); pos = end_pfn << PAGE_SHIFT; diff --git a/arch/x86/mm/iomap_32.c b/arch/x86/mm/iomap_32.c index d0151d8ce45..04102d42ff4 100644 --- a/arch/x86/mm/iomap_32.c +++ b/arch/x86/mm/iomap_32.c @@ -17,8 +17,20 @@ */ #include <asm/iomap.h> +#include <asm/pat.h> #include <linux/module.h> +int is_io_mapping_possible(resource_size_t base, unsigned long size) +{ +#ifndef CONFIG_X86_PAE + /* There is no way to map greater than 1 << 32 address without PAE */ + if (base + size > 0x100000000ULL) + return 0; +#endif + return 1; +} +EXPORT_SYMBOL_GPL(is_io_mapping_possible); + /* Map 'pfn' using fixed map 'type' and protections 'prot' */ void * @@ -29,6 +41,15 @@ iomap_atomic_prot_pfn(unsigned long pfn, enum km_type type, pgprot_t prot) pagefault_disable(); + /* + * For non-PAT systems, promote PAGE_KERNEL_WC to PAGE_KERNEL_UC_MINUS. + * PAGE_KERNEL_WC maps to PWT, which translates to uncached if the + * MTRR is UC or WC. UC_MINUS gets the real intention, of the + * user, which is "WC if the MTRR is WC, UC if you can't do that." + */ + if (!pat_enabled && pgprot_val(prot) == pgprot_val(PAGE_KERNEL_WC)) + prot = PAGE_KERNEL_UC_MINUS; + idx = type + KM_TYPE_NR*smp_processor_id(); vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx); set_pte(kmap_pte-idx, pfn_pte(pfn, prot)); diff --git a/arch/x86/mm/ioremap.c b/arch/x86/mm/ioremap.c index bd85d42819e..f45d5e29a72 100644 --- a/arch/x86/mm/ioremap.c +++ b/arch/x86/mm/ioremap.c @@ -134,25 +134,6 @@ int page_is_ram(unsigned long pagenr) return 0; } -int pagerange_is_ram(unsigned long start, unsigned long end) -{ - int ram_page = 0, not_rampage = 0; - unsigned long page_nr; - - for (page_nr = (start >> PAGE_SHIFT); page_nr < (end >> PAGE_SHIFT); - ++page_nr) { - if (page_is_ram(page_nr)) - ram_page = 1; - else - not_rampage = 1; - - if (ram_page == not_rampage) - return -1; - } - - return ram_page; -} - /* * Fix up the linear direct mapping of the kernel to avoid cache attribute * conflicts. @@ -557,34 +538,9 @@ void __init early_ioremap_init(void) } } -void __init early_ioremap_clear(void) -{ - pmd_t *pmd; - - if (early_ioremap_debug) - printk(KERN_INFO "early_ioremap_clear()\n"); - - pmd = early_ioremap_pmd(fix_to_virt(FIX_BTMAP_BEGIN)); - pmd_clear(pmd); - paravirt_release_pte(__pa(bm_pte) >> PAGE_SHIFT); - __flush_tlb_all(); -} - void __init early_ioremap_reset(void) { - enum fixed_addresses idx; - unsigned long addr, phys; - pte_t *pte; - after_paging_init = 1; - for (idx = FIX_BTMAP_BEGIN; idx >= FIX_BTMAP_END; idx--) { - addr = fix_to_virt(idx); - pte = early_ioremap_pte(addr); - if (pte_present(*pte)) { - phys = pte_val(*pte) & PAGE_MASK; - set_fixmap(idx, phys); - } - } } static void __init __early_set_fixmap(enum fixed_addresses idx, diff --git a/arch/x86/mm/kmmio.c b/arch/x86/mm/kmmio.c index 93d82038af4..6a518dd08a3 100644 --- a/arch/x86/mm/kmmio.c +++ b/arch/x86/mm/kmmio.c @@ -32,11 +32,14 @@ struct kmmio_fault_page { struct list_head list; struct kmmio_fault_page *release_next; unsigned long page; /* location of the fault page */ + bool old_presence; /* page presence prior to arming */ + bool armed; /* * Number of times this page has been registered as a part * of a probe. If zero, page is disarmed and this may be freed. - * Used only by writers (RCU). + * Used only by writers (RCU) and post_kmmio_handler(). + * Protected by kmmio_lock, when linked into kmmio_page_table. */ int count; }; @@ -105,57 +108,85 @@ static struct kmmio_fault_page *get_kmmio_fault_page(unsigned long page) return NULL; } -static void set_page_present(unsigned long addr, bool present, - unsigned int *pglevel) +static void set_pmd_presence(pmd_t *pmd, bool present, bool *old) +{ + pmdval_t v = pmd_val(*pmd); + *old = !!(v & _PAGE_PRESENT); + v &= ~_PAGE_PRESENT; + if (present) + v |= _PAGE_PRESENT; + set_pmd(pmd, __pmd(v)); +} + +static void set_pte_presence(pte_t *pte, bool present, bool *old) +{ + pteval_t v = pte_val(*pte); + *old = !!(v & _PAGE_PRESENT); + v &= ~_PAGE_PRESENT; + if (present) + v |= _PAGE_PRESENT; + set_pte_atomic(pte, __pte(v)); +} + +static int set_page_presence(unsigned long addr, bool present, bool *old) { - pteval_t pteval; - pmdval_t pmdval; unsigned int level; - pmd_t *pmd; pte_t *pte = lookup_address(addr, &level); if (!pte) { pr_err("kmmio: no pte for page 0x%08lx\n", addr); - return; + return -1; } - if (pglevel) - *pglevel = level; - switch (level) { case PG_LEVEL_2M: - pmd = (pmd_t *)pte; - pmdval = pmd_val(*pmd) & ~_PAGE_PRESENT; - if (present) - pmdval |= _PAGE_PRESENT; - set_pmd(pmd, __pmd(pmdval)); + set_pmd_presence((pmd_t *)pte, present, old); break; - case PG_LEVEL_4K: - pteval = pte_val(*pte) & ~_PAGE_PRESENT; - if (present) - pteval |= _PAGE_PRESENT; - set_pte_atomic(pte, __pte(pteval)); + set_pte_presence(pte, present, old); break; - default: pr_err("kmmio: unexpected page level 0x%x.\n", level); - return; + return -1; } __flush_tlb_one(addr); + return 0; } -/** Mark the given page as not present. Access to it will trigger a fault. */ -static void arm_kmmio_fault_page(unsigned long page, unsigned int *pglevel) +/* + * Mark the given page as not present. Access to it will trigger a fault. + * + * Struct kmmio_fault_page is protected by RCU and kmmio_lock, but the + * protection is ignored here. RCU read lock is assumed held, so the struct + * will not disappear unexpectedly. Furthermore, the caller must guarantee, + * that double arming the same virtual address (page) cannot occur. + * + * Double disarming on the other hand is allowed, and may occur when a fault + * and mmiotrace shutdown happen simultaneously. + */ +static int arm_kmmio_fault_page(struct kmmio_fault_page *f) { - set_page_present(page & PAGE_MASK, false, pglevel); + int ret; + WARN_ONCE(f->armed, KERN_ERR "kmmio page already armed.\n"); + if (f->armed) { + pr_warning("kmmio double-arm: page 0x%08lx, ref %d, old %d\n", + f->page, f->count, f->old_presence); + } + ret = set_page_presence(f->page, false, &f->old_presence); + WARN_ONCE(ret < 0, KERN_ERR "kmmio arming 0x%08lx failed.\n", f->page); + f->armed = true; + return ret; } -/** Mark the given page as present. */ -static void disarm_kmmio_fault_page(unsigned long page, unsigned int *pglevel) +/** Restore the given page to saved presence state. */ +static void disarm_kmmio_fault_page(struct kmmio_fault_page *f) { - set_page_present(page & PAGE_MASK, true, pglevel); + bool tmp; + int ret = set_page_presence(f->page, f->old_presence, &tmp); + WARN_ONCE(ret < 0, + KERN_ERR "kmmio disarming 0x%08lx failed.\n", f->page); + f->armed = false; } /* @@ -202,28 +233,32 @@ int kmmio_handler(struct pt_regs *regs, unsigned long addr) ctx = &get_cpu_var(kmmio_ctx); if (ctx->active) { - disarm_kmmio_fault_page(faultpage->page, NULL); if (addr == ctx->addr) { /* - * On SMP we sometimes get recursive probe hits on the - * same address. Context is already saved, fall out. + * A second fault on the same page means some other + * condition needs handling by do_page_fault(), the + * page really not being present is the most common. */ - pr_debug("kmmio: duplicate probe hit on CPU %d, for " - "address 0x%08lx.\n", - smp_processor_id(), addr); - ret = 1; - goto no_kmmio_ctx; - } - /* - * Prevent overwriting already in-flight context. - * This should not happen, let's hope disarming at least - * prevents a panic. - */ - pr_emerg("kmmio: recursive probe hit on CPU %d, " + pr_debug("kmmio: secondary hit for 0x%08lx CPU %d.\n", + addr, smp_processor_id()); + + if (!faultpage->old_presence) + pr_info("kmmio: unexpected secondary hit for " + "address 0x%08lx on CPU %d.\n", addr, + smp_processor_id()); + } else { + /* + * Prevent overwriting already in-flight context. + * This should not happen, let's hope disarming at + * least prevents a panic. + */ + pr_emerg("kmmio: recursive probe hit on CPU %d, " "for address 0x%08lx. Ignoring.\n", smp_processor_id(), addr); - pr_emerg("kmmio: previous hit was at 0x%08lx.\n", - ctx->addr); + pr_emerg("kmmio: previous hit was at 0x%08lx.\n", + ctx->addr); + disarm_kmmio_fault_page(faultpage); + } goto no_kmmio_ctx; } ctx->active++; @@ -244,7 +279,7 @@ int kmmio_handler(struct pt_regs *regs, unsigned long addr) regs->flags &= ~X86_EFLAGS_IF; /* Now we set present bit in PTE and single step. */ - disarm_kmmio_fault_page(ctx->fpage->page, NULL); + disarm_kmmio_fault_page(ctx->fpage); /* * If another cpu accesses the same page while we are stepping, @@ -275,7 +310,7 @@ static int post_kmmio_handler(unsigned long condition, struct pt_regs *regs) struct kmmio_context *ctx = &get_cpu_var(kmmio_ctx); if (!ctx->active) { - pr_debug("kmmio: spurious debug trap on CPU %d.\n", + pr_warning("kmmio: spurious debug trap on CPU %d.\n", smp_processor_id()); goto out; } @@ -283,7 +318,11 @@ static int post_kmmio_handler(unsigned long condition, struct pt_regs *regs) if (ctx->probe && ctx->probe->post_handler) ctx->probe->post_handler(ctx->probe, condition, regs); - arm_kmmio_fault_page(ctx->fpage->page, NULL); + /* Prevent racing against release_kmmio_fault_page(). */ + spin_lock(&kmmio_lock); + if (ctx->fpage->count) + arm_kmmio_fault_page(ctx->fpage); + spin_unlock(&kmmio_lock); regs->flags &= ~X86_EFLAGS_TF; regs->flags |= ctx->saved_flags; @@ -315,20 +354,24 @@ static int add_kmmio_fault_page(unsigned long page) f = get_kmmio_fault_page(page); if (f) { if (!f->count) - arm_kmmio_fault_page(f->page, NULL); + arm_kmmio_fault_page(f); f->count++; return 0; } - f = kmalloc(sizeof(*f), GFP_ATOMIC); + f = kzalloc(sizeof(*f), GFP_ATOMIC); if (!f) return -1; f->count = 1; f->page = page; - list_add_rcu(&f->list, kmmio_page_list(f->page)); - arm_kmmio_fault_page(f->page, NULL); + if (arm_kmmio_fault_page(f)) { + kfree(f); + return -1; + } + + list_add_rcu(&f->list, kmmio_page_list(f->page)); return 0; } @@ -347,7 +390,7 @@ static void release_kmmio_fault_page(unsigned long page, f->count--; BUG_ON(f->count < 0); if (!f->count) { - disarm_kmmio_fault_page(f->page, NULL); + disarm_kmmio_fault_page(f); f->release_next = *release_list; *release_list = f; } @@ -408,23 +451,24 @@ static void rcu_free_kmmio_fault_pages(struct rcu_head *head) static void remove_kmmio_fault_pages(struct rcu_head *head) { - struct kmmio_delayed_release *dr = container_of( - head, - struct kmmio_delayed_release, - rcu); + struct kmmio_delayed_release *dr = + container_of(head, struct kmmio_delayed_release, rcu); struct kmmio_fault_page *p = dr->release_list; struct kmmio_fault_page **prevp = &dr->release_list; unsigned long flags; + spin_lock_irqsave(&kmmio_lock, flags); while (p) { - if (!p->count) + if (!p->count) { list_del_rcu(&p->list); - else + prevp = &p->release_next; + } else { *prevp = p->release_next; - prevp = &p->release_next; + } p = p->release_next; } spin_unlock_irqrestore(&kmmio_lock, flags); + /* This is the real RCU destroy call. */ call_rcu(&dr->rcu, rcu_free_kmmio_fault_pages); } diff --git a/arch/x86/mm/numa_64.c b/arch/x86/mm/numa_64.c index 71a14f89f89..f3516da035d 100644 --- a/arch/x86/mm/numa_64.c +++ b/arch/x86/mm/numa_64.c @@ -145,7 +145,7 @@ int __init compute_hash_shift(struct bootnode *nodes, int numnodes, return shift; } -int early_pfn_to_nid(unsigned long pfn) +int __meminit __early_pfn_to_nid(unsigned long pfn) { return phys_to_nid(pfn << PAGE_SHIFT); } diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c index e89d24815f2..7233bd7e357 100644 --- a/arch/x86/mm/pageattr.c +++ b/arch/x86/mm/pageattr.c @@ -508,18 +508,24 @@ static int split_large_page(pte_t *kpte, unsigned long address) #endif /* - * Install the new, split up pagetable. Important details here: + * Install the new, split up pagetable. * - * On Intel the NX bit of all levels must be cleared to make a - * page executable. See section 4.13.2 of Intel 64 and IA-32 - * Architectures Software Developer's Manual). + * We use the standard kernel pagetable protections for the new + * pagetable protections, the actual ptes set above control the + * primary protection behavior: + */ + __set_pmd_pte(kpte, address, mk_pte(base, __pgprot(_KERNPG_TABLE))); + + /* + * Intel Atom errata AAH41 workaround. * - * Mark the entry present. The current mapping might be - * set to not present, which we preserved above. + * The real fix should be in hw or in a microcode update, but + * we also probabilistically try to reduce the window of having + * a large TLB mixed with 4K TLBs while instruction fetches are + * going on. */ - ref_prot = pte_pgprot(pte_mkexec(pte_clrhuge(*kpte))); - pgprot_val(ref_prot) |= _PAGE_PRESENT; - __set_pmd_pte(kpte, address, mk_pte(base, ref_prot)); + __flush_tlb_all(); + base = NULL; out_unlock: @@ -534,6 +540,36 @@ out_unlock: return 0; } +static int __cpa_process_fault(struct cpa_data *cpa, unsigned long vaddr, + int primary) +{ + /* + * Ignore all non primary paths. + */ + if (!primary) + return 0; + + /* + * Ignore the NULL PTE for kernel identity mapping, as it is expected + * to have holes. + * Also set numpages to '1' indicating that we processed cpa req for + * one virtual address page and its pfn. TBD: numpages can be set based + * on the initial value and the level returned by lookup_address(). + */ + if (within(vaddr, PAGE_OFFSET, + PAGE_OFFSET + (max_pfn_mapped << PAGE_SHIFT))) { + cpa->numpages = 1; + cpa->pfn = __pa(vaddr) >> PAGE_SHIFT; + return 0; + } else { + WARN(1, KERN_WARNING "CPA: called for zero pte. " + "vaddr = %lx cpa->vaddr = %lx\n", vaddr, + *cpa->vaddr); + + return -EFAULT; + } +} + static int __change_page_attr(struct cpa_data *cpa, int primary) { unsigned long address; @@ -545,21 +581,14 @@ static int __change_page_attr(struct cpa_data *cpa, int primary) address = cpa->vaddr[cpa->curpage]; else address = *cpa->vaddr; - repeat: kpte = lookup_address(address, &level); if (!kpte) - return 0; + return __cpa_process_fault(cpa, address, primary); old_pte = *kpte; - if (!pte_val(old_pte)) { - if (!primary) - return 0; - WARN(1, KERN_WARNING "CPA: called for zero pte. " - "vaddr = %lx cpa->vaddr = %lx\n", address, - *cpa->vaddr); - return -EINVAL; - } + if (!pte_val(old_pte)) + return __cpa_process_fault(cpa, address, primary); if (level == PG_LEVEL_4K) { pte_t new_pte; @@ -657,12 +686,7 @@ static int cpa_process_alias(struct cpa_data *cpa) vaddr = *cpa->vaddr; if (!(within(vaddr, PAGE_OFFSET, - PAGE_OFFSET + (max_low_pfn_mapped << PAGE_SHIFT)) -#ifdef CONFIG_X86_64 - || within(vaddr, PAGE_OFFSET + (1UL<<32), - PAGE_OFFSET + (max_pfn_mapped << PAGE_SHIFT)) -#endif - )) { + PAGE_OFFSET + (max_pfn_mapped << PAGE_SHIFT)))) { alias_cpa = *cpa; temp_cpa_vaddr = (unsigned long) __va(cpa->pfn << PAGE_SHIFT); @@ -793,6 +817,13 @@ static int change_page_attr_set_clr(unsigned long *addr, int numpages, vm_unmap_aliases(); + /* + * If we're called with lazy mmu updates enabled, the + * in-memory pte state may be stale. Flush pending updates to + * bring them up to date. + */ + arch_flush_lazy_mmu_mode(); + cpa.vaddr = addr; cpa.numpages = numpages; cpa.mask_set = mask_set; @@ -835,6 +866,13 @@ static int change_page_attr_set_clr(unsigned long *addr, int numpages, } else cpa_flush_all(cache); + /* + * If we've been called with lazy mmu updates enabled, then + * make sure that everything gets flushed out before we + * return. + */ + arch_flush_lazy_mmu_mode(); + out: return ret; } diff --git a/arch/x86/mm/pat.c b/arch/x86/mm/pat.c index 8b08fb95527..e0ab173b697 100644 --- a/arch/x86/mm/pat.c +++ b/arch/x86/mm/pat.c @@ -11,6 +11,7 @@ #include <linux/bootmem.h> #include <linux/debugfs.h> #include <linux/kernel.h> +#include <linux/module.h> #include <linux/gfp.h> #include <linux/mm.h> #include <linux/fs.h> @@ -211,6 +212,33 @@ chk_conflict(struct memtype *new, struct memtype *entry, unsigned long *type) static struct memtype *cached_entry; static u64 cached_start; +static int pat_pagerange_is_ram(unsigned long start, unsigned long end) +{ + int ram_page = 0, not_rampage = 0; + unsigned long page_nr; + + for (page_nr = (start >> PAGE_SHIFT); page_nr < (end >> PAGE_SHIFT); + ++page_nr) { + /* + * For legacy reasons, physical address range in the legacy ISA + * region is tracked as non-RAM. This will allow users of + * /dev/mem to map portions of legacy ISA region, even when + * some of those portions are listed(or not even listed) with + * different e820 types(RAM/reserved/..) + */ + if (page_nr >= (ISA_END_ADDRESS >> PAGE_SHIFT) && + page_is_ram(page_nr)) + ram_page = 1; + else + not_rampage = 1; + + if (ram_page == not_rampage) + return -1; + } + + return ram_page; +} + /* * For RAM pages, mark the pages as non WB memory type using * PageNonWB (PG_arch_1). We allow only one set_memory_uc() or @@ -333,9 +361,13 @@ int reserve_memtype(u64 start, u64 end, unsigned long req_type, req_type & _PAGE_CACHE_MASK); } - is_range_ram = pagerange_is_ram(start, end); + if (new_type) + *new_type = actual_type; + + is_range_ram = pat_pagerange_is_ram(start, end); if (is_range_ram == 1) - return reserve_ram_pages_type(start, end, req_type, new_type); + return reserve_ram_pages_type(start, end, req_type, + new_type); else if (is_range_ram < 0) return -EINVAL; @@ -347,9 +379,6 @@ int reserve_memtype(u64 start, u64 end, unsigned long req_type, new->end = end; new->type = actual_type; - if (new_type) - *new_type = actual_type; - spin_lock(&memtype_lock); if (cached_entry && start >= cached_start) @@ -437,7 +466,7 @@ int free_memtype(u64 start, u64 end) if (is_ISA_range(start, end - 1)) return 0; - is_range_ram = pagerange_is_ram(start, end); + is_range_ram = pat_pagerange_is_ram(start, end); if (is_range_ram == 1) return free_ram_pages_type(start, end); else if (is_range_ram < 0) @@ -609,17 +638,13 @@ static int reserve_pfn_range(u64 paddr, unsigned long size, pgprot_t *vma_prot, unsigned long flags; unsigned long want_flags = (pgprot_val(*vma_prot) & _PAGE_CACHE_MASK); - is_ram = pagerange_is_ram(paddr, paddr + size); + is_ram = pat_pagerange_is_ram(paddr, paddr + size); - if (is_ram != 0) { - /* - * For mapping RAM pages, drivers need to call - * set_memory_[uc|wc|wb] directly, for reserve and free, before - * setting up the PTE. - */ - WARN_ON_ONCE(1); - return 0; - } + /* + * reserve_pfn_range() doesn't support RAM pages. + */ + if (is_ram != 0) + return -EINVAL; ret = reserve_memtype(paddr, paddr + size, want_flags, &flags); if (ret) @@ -676,7 +701,7 @@ static void free_pfn_range(u64 paddr, unsigned long size) { int is_ram; - is_ram = pagerange_is_ram(paddr, paddr + size); + is_ram = pat_pagerange_is_ram(paddr, paddr + size); if (is_ram == 0) free_memtype(paddr, paddr + size); } @@ -844,6 +869,7 @@ pgprot_t pgprot_writecombine(pgprot_t prot) else return pgprot_noncached(prot); } +EXPORT_SYMBOL_GPL(pgprot_writecombine); #if defined(CONFIG_DEBUG_FS) && defined(CONFIG_X86_PAT) diff --git a/arch/x86/mm/testmmiotrace.c b/arch/x86/mm/testmmiotrace.c index ab50a8d7402..427fd1b56df 100644 --- a/arch/x86/mm/testmmiotrace.c +++ b/arch/x86/mm/testmmiotrace.c @@ -1,5 +1,5 @@ /* - * Written by Pekka Paalanen, 2008 <pq@iki.fi> + * Written by Pekka Paalanen, 2008-2009 <pq@iki.fi> */ #include <linux/module.h> #include <linux/io.h> @@ -9,35 +9,74 @@ static unsigned long mmio_address; module_param(mmio_address, ulong, 0); -MODULE_PARM_DESC(mmio_address, "Start address of the mapping of 16 kB."); +MODULE_PARM_DESC(mmio_address, " Start address of the mapping of 16 kB " + "(or 8 MB if read_far is non-zero)."); + +static unsigned long read_far = 0x400100; +module_param(read_far, ulong, 0); +MODULE_PARM_DESC(read_far, " Offset of a 32-bit read within 8 MB " + "(default: 0x400100)."); + +static unsigned v16(unsigned i) +{ + return i * 12 + 7; +} + +static unsigned v32(unsigned i) +{ + return i * 212371 + 13; +} static void do_write_test(void __iomem *p) { unsigned int i; + pr_info(MODULE_NAME ": write test.\n"); mmiotrace_printk("Write test.\n"); + for (i = 0; i < 256; i++) iowrite8(i, p + i); + for (i = 1024; i < (5 * 1024); i += 2) - iowrite16(i * 12 + 7, p + i); + iowrite16(v16(i), p + i); + for (i = (5 * 1024); i < (16 * 1024); i += 4) - iowrite32(i * 212371 + 13, p + i); + iowrite32(v32(i), p + i); } static void do_read_test(void __iomem *p) { unsigned int i; + unsigned errs[3] = { 0 }; + pr_info(MODULE_NAME ": read test.\n"); mmiotrace_printk("Read test.\n"); + for (i = 0; i < 256; i++) - ioread8(p + i); + if (ioread8(p + i) != i) + ++errs[0]; + for (i = 1024; i < (5 * 1024); i += 2) - ioread16(p + i); + if (ioread16(p + i) != v16(i)) + ++errs[1]; + for (i = (5 * 1024); i < (16 * 1024); i += 4) - ioread32(p + i); + if (ioread32(p + i) != v32(i)) + ++errs[2]; + + mmiotrace_printk("Read errors: 8-bit %d, 16-bit %d, 32-bit %d.\n", + errs[0], errs[1], errs[2]); } -static void do_test(void) +static void do_read_far_test(void __iomem *p) { - void __iomem *p = ioremap_nocache(mmio_address, 0x4000); + pr_info(MODULE_NAME ": read far test.\n"); + mmiotrace_printk("Read far test.\n"); + + ioread32(p + read_far); +} + +static void do_test(unsigned long size) +{ + void __iomem *p = ioremap_nocache(mmio_address, size); if (!p) { pr_err(MODULE_NAME ": could not ioremap, aborting.\n"); return; @@ -45,11 +84,15 @@ static void do_test(void) mmiotrace_printk("ioremap returned %p.\n", p); do_write_test(p); do_read_test(p); + if (read_far && read_far < size - 4) + do_read_far_test(p); iounmap(p); } static int __init init(void) { + unsigned long size = (read_far) ? (8 << 20) : (16 << 10); + if (mmio_address == 0) { pr_err(MODULE_NAME ": you have to use the module argument " "mmio_address.\n"); @@ -58,10 +101,11 @@ static int __init init(void) return -ENXIO; } - pr_warning(MODULE_NAME ": WARNING: mapping 16 kB @ 0x%08lx " - "in PCI address space, and writing " - "rubbish in there.\n", mmio_address); - do_test(); + pr_warning(MODULE_NAME ": WARNING: mapping %lu kB @ 0x%08lx in PCI " + "address space, and writing 16 kB of rubbish in there.\n", + size >> 10, mmio_address); + do_test(size); + pr_info(MODULE_NAME ": All done.\n"); return 0; } diff --git a/arch/x86/oprofile/op_model_ppro.c b/arch/x86/oprofile/op_model_ppro.c index e9f80c744cf..10131fbdaad 100644 --- a/arch/x86/oprofile/op_model_ppro.c +++ b/arch/x86/oprofile/op_model_ppro.c @@ -78,8 +78,18 @@ static void ppro_setup_ctrs(struct op_msrs const * const msrs) if (cpu_has_arch_perfmon) { union cpuid10_eax eax; eax.full = cpuid_eax(0xa); - if (counter_width < eax.split.bit_width) - counter_width = eax.split.bit_width; + + /* + * For Core2 (family 6, model 15), don't reset the + * counter width: + */ + if (!(eax.split.version_id == 0 && + current_cpu_data.x86 == 6 && + current_cpu_data.x86_model == 15)) { + + if (counter_width < eax.split.bit_width) + counter_width = eax.split.bit_width; + } } /* clear all counters */ diff --git a/arch/x86/pci/irq.c b/arch/x86/pci/irq.c index 4064345cf14..fecbce6e7d7 100644 --- a/arch/x86/pci/irq.c +++ b/arch/x86/pci/irq.c @@ -572,6 +572,7 @@ static __init int intel_router_probe(struct irq_router *r, struct pci_dev *route case PCI_DEVICE_ID_INTEL_ICH7_1: case PCI_DEVICE_ID_INTEL_ICH7_30: case PCI_DEVICE_ID_INTEL_ICH7_31: + case PCI_DEVICE_ID_INTEL_TGP_LPC: case PCI_DEVICE_ID_INTEL_ESB2_0: case PCI_DEVICE_ID_INTEL_ICH8_0: case PCI_DEVICE_ID_INTEL_ICH8_1: diff --git a/arch/x86/scripts/strip-symbols b/arch/x86/scripts/strip-symbols deleted file mode 100644 index a2f1ccb827c..00000000000 --- a/arch/x86/scripts/strip-symbols +++ /dev/null @@ -1 +0,0 @@ -__cpu_vendor_dev_X86_VENDOR_* diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c index bea215230b2..b58e9633814 100644 --- a/arch/x86/xen/enlighten.c +++ b/arch/x86/xen/enlighten.c @@ -1672,6 +1672,9 @@ asmlinkage void __init xen_start_kernel(void) possible map and a non-dummy shared_info. */ per_cpu(xen_vcpu, 0) = &HYPERVISOR_shared_info->vcpu_info[0]; + local_irq_disable(); + early_boot_irqs_off(); + xen_raw_console_write("mapping kernel into physical memory\n"); pgd = xen_setup_kernel_pagetable(pgd, xen_start_info->nr_pages); diff --git a/arch/x86/xen/multicalls.h b/arch/x86/xen/multicalls.h index 85893824161..fa3e10725d9 100644 --- a/arch/x86/xen/multicalls.h +++ b/arch/x86/xen/multicalls.h @@ -19,8 +19,10 @@ DECLARE_PER_CPU(unsigned long, xen_mc_irq_flags); paired with xen_mc_issue() */ static inline void xen_mc_batch(void) { + unsigned long flags; /* need to disable interrupts until this entry is complete */ - local_irq_save(__get_cpu_var(xen_mc_irq_flags)); + local_irq_save(flags); + __get_cpu_var(xen_mc_irq_flags) = flags; } static inline struct multicall_space xen_mc_entry(size_t args) |