From 7a470e826d7521bec6af789deab31cfa4fd05af3 Mon Sep 17 00:00:00 2001 From: Sohil Mehta Date: Thu, 7 Nov 2024 23:30:00 +0000 Subject: [PATCH 01/16] x86/cpufeatures: Free up unused feature bits Linux defined feature bits X86_FEATURE_P3 and X86_FEATURE_P4 are not used anywhere. Commit f31d731e4467 ("x86: use X86_FEATURE_NOPL in alternatives") got rid of the last usage in 2008. Remove the related mappings and code. Just like all X86_FEATURE bits, the raw bit numbers can be exposed to userspace via MODULE_DEVICE_TABLE(). There is a very small theoretical chance of userspace getting confused if these bits got reassigned and changed logical meaning. But these bits were never used for a device table, so it's highly unlikely this will ever happen in practice. [ dhansen: clarify userspace visibility of these bits ] Signed-off-by: Sohil Mehta Signed-off-by: Dave Hansen Signed-off-by: Ingo Molnar Link: https://lore.kernel.org/all/20241107233000.2742619-1-sohil.mehta%40intel.com --- arch/x86/include/asm/cpufeatures.h | 4 ++-- arch/x86/kernel/cpu/intel.c | 5 ----- 2 files changed, 2 insertions(+), 7 deletions(-) diff --git a/arch/x86/include/asm/cpufeatures.h b/arch/x86/include/asm/cpufeatures.h index 17b6590748c0..f725ccc77b01 100644 --- a/arch/x86/include/asm/cpufeatures.h +++ b/arch/x86/include/asm/cpufeatures.h @@ -83,8 +83,8 @@ #define X86_FEATURE_CENTAUR_MCR ( 3*32+ 3) /* "centaur_mcr" Centaur MCRs (= MTRRs) */ #define X86_FEATURE_K8 ( 3*32+ 4) /* Opteron, Athlon64 */ #define X86_FEATURE_ZEN5 ( 3*32+ 5) /* CPU based on Zen5 microarchitecture */ -#define X86_FEATURE_P3 ( 3*32+ 6) /* P3 */ -#define X86_FEATURE_P4 ( 3*32+ 7) /* P4 */ +/* Free ( 3*32+ 6) */ +/* Free ( 3*32+ 7) */ #define X86_FEATURE_CONSTANT_TSC ( 3*32+ 8) /* "constant_tsc" TSC ticks at a constant rate */ #define X86_FEATURE_UP ( 3*32+ 9) /* "up" SMP kernel running on UP */ #define X86_FEATURE_ART ( 3*32+10) /* "art" Always running timer (ART) */ diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c index d1de300af173..5a9fbe962135 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -597,11 +597,6 @@ static void init_intel(struct cpuinfo_x86 *c) if (p) strcpy(c->x86_model_id, p); } - - if (c->x86 == 15) - set_cpu_cap(c, X86_FEATURE_P4); - if (c->x86 == 6) - set_cpu_cap(c, X86_FEATURE_P3); #endif /* Work around errata */ -- 2.51.0 From 29188c16006176caee6cb6729103be51a29c1a93 Mon Sep 17 00:00:00 2001 From: Juergen Gross Date: Tue, 3 Dec 2024 08:15:50 +0100 Subject: [PATCH 02/16] x86/paravirt: Remove the WBINVD callback The pv_ops::cpu.wbinvd paravirt callback is a leftover of lguest times. Today it is no longer needed, as all users use the native WBINVD implementation. Remove the callback and rename native_wbinvd() to wbinvd(). Signed-off-by: Juergen Gross Signed-off-by: Ingo Molnar Link: https://lore.kernel.org/r/20241203071550.26487-1-jgross@suse.com --- arch/x86/include/asm/paravirt.h | 7 ------- arch/x86/include/asm/paravirt_types.h | 2 -- arch/x86/include/asm/special_insns.h | 8 +------- arch/x86/kernel/cpu/resctrl/pseudo_lock.c | 2 +- arch/x86/kernel/paravirt.c | 6 ------ arch/x86/kernel/process.c | 4 ++-- arch/x86/xen/enlighten_pv.c | 2 -- 7 files changed, 4 insertions(+), 27 deletions(-) diff --git a/arch/x86/include/asm/paravirt.h b/arch/x86/include/asm/paravirt.h index d4eb9e1d61b8..041aff51eb50 100644 --- a/arch/x86/include/asm/paravirt.h +++ b/arch/x86/include/asm/paravirt.h @@ -180,13 +180,6 @@ static inline void halt(void) PVOP_VCALL0(irq.halt); } -extern noinstr void pv_native_wbinvd(void); - -static __always_inline void wbinvd(void) -{ - PVOP_ALT_VCALL0(cpu.wbinvd, "wbinvd", ALT_NOT_XEN); -} - static inline u64 paravirt_read_msr(unsigned msr) { return PVOP_CALL1(u64, cpu.read_msr, msr); diff --git a/arch/x86/include/asm/paravirt_types.h b/arch/x86/include/asm/paravirt_types.h index 8d4fbe1be489..fea56b04f436 100644 --- a/arch/x86/include/asm/paravirt_types.h +++ b/arch/x86/include/asm/paravirt_types.h @@ -86,8 +86,6 @@ struct pv_cpu_ops { void (*update_io_bitmap)(void); #endif - void (*wbinvd)(void); - /* cpuid emulation, mostly so that caps bits can be disabled */ void (*cpuid)(unsigned int *eax, unsigned int *ebx, unsigned int *ecx, unsigned int *edx); diff --git a/arch/x86/include/asm/special_insns.h b/arch/x86/include/asm/special_insns.h index aec6e2d3aa1d..fab7c8af27a4 100644 --- a/arch/x86/include/asm/special_insns.h +++ b/arch/x86/include/asm/special_insns.h @@ -115,7 +115,7 @@ static inline void wrpkru(u32 pkru) } #endif -static __always_inline void native_wbinvd(void) +static __always_inline void wbinvd(void) { asm volatile("wbinvd": : :"memory"); } @@ -167,12 +167,6 @@ static inline void __write_cr4(unsigned long x) { native_write_cr4(x); } - -static __always_inline void wbinvd(void) -{ - native_wbinvd(); -} - #endif /* CONFIG_PARAVIRT_XXL */ static __always_inline void clflush(volatile void *__p) diff --git a/arch/x86/kernel/cpu/resctrl/pseudo_lock.c b/arch/x86/kernel/cpu/resctrl/pseudo_lock.c index 972e6b6b0481..b72f7e91387e 100644 --- a/arch/x86/kernel/cpu/resctrl/pseudo_lock.c +++ b/arch/x86/kernel/cpu/resctrl/pseudo_lock.c @@ -459,7 +459,7 @@ static int pseudo_lock_fn(void *_rdtgrp) * increase likelihood that allocated cache portion will be filled * with associated memory. */ - native_wbinvd(); + wbinvd(); /* * Always called with interrupts enabled. By disabling interrupts diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c index fec381533555..927e33e6843a 100644 --- a/arch/x86/kernel/paravirt.c +++ b/arch/x86/kernel/paravirt.c @@ -116,11 +116,6 @@ static noinstr void pv_native_set_debugreg(int regno, unsigned long val) native_set_debugreg(regno, val); } -noinstr void pv_native_wbinvd(void) -{ - native_wbinvd(); -} - static noinstr void pv_native_safe_halt(void) { native_safe_halt(); @@ -148,7 +143,6 @@ struct paravirt_patch_template pv_ops = { .cpu.read_cr0 = native_read_cr0, .cpu.write_cr0 = native_write_cr0, .cpu.write_cr4 = native_write_cr4, - .cpu.wbinvd = pv_native_wbinvd, .cpu.read_msr = native_read_msr, .cpu.write_msr = native_write_msr, .cpu.read_msr_safe = native_read_msr_safe, diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index f63f8fd00a91..58ead05a1c29 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -825,7 +825,7 @@ void __noreturn stop_this_cpu(void *dummy) * X86_FEATURE_SME due to cmdline options. */ if (c->extended_cpuid_level >= 0x8000001f && (cpuid_eax(0x8000001f) & BIT(0))) - native_wbinvd(); + wbinvd(); /* * This brings a cache line back and dirties it, but @@ -846,7 +846,7 @@ void __noreturn stop_this_cpu(void *dummy) /* * Use native_halt() so that memory contents don't change * (stack usage and variables) after possibly issuing the - * native_wbinvd() above. + * wbinvd() above. */ native_halt(); } diff --git a/arch/x86/xen/enlighten_pv.c b/arch/x86/xen/enlighten_pv.c index d6818c6cafda..fd2169063480 100644 --- a/arch/x86/xen/enlighten_pv.c +++ b/arch/x86/xen/enlighten_pv.c @@ -1161,8 +1161,6 @@ static const typeof(pv_ops) xen_cpu_ops __initconst = { .write_cr4 = xen_write_cr4, - .wbinvd = pv_native_wbinvd, - .read_msr = xen_read_msr, .write_msr = xen_write_msr, -- 2.51.0 From 4bf610499c429fa0bfb3fa94be450f01016224c5 Mon Sep 17 00:00:00 2001 From: "Borislav Petkov (AMD)" Date: Thu, 31 Oct 2024 11:34:01 +0100 Subject: [PATCH 03/16] x86/cpufeature: Document cpu_feature_enabled() as the default to use cpu_feature_enabled() should be used in most cases when CPU feature support needs to be tested in code. Document that. Reported-by: Sohil Mehta Signed-off-by: Borislav Petkov (AMD) Reviewed-by: Sohil Mehta Acked-by: Dave Hansen Link: https://lore.kernel.org/r/20241031103401.GBZyNdGQ-ZyXKyzC_z@fat_crate.local --- arch/x86/include/asm/cpufeature.h | 18 ++++++------------ 1 file changed, 6 insertions(+), 12 deletions(-) diff --git a/arch/x86/include/asm/cpufeature.h b/arch/x86/include/asm/cpufeature.h index 0b9611da6c53..de1ad09fe8d7 100644 --- a/arch/x86/include/asm/cpufeature.h +++ b/arch/x86/include/asm/cpufeature.h @@ -132,11 +132,12 @@ extern const char * const x86_bug_flags[NBUGINTS*32]; x86_this_cpu_test_bit(bit, cpu_info.x86_capability)) /* - * This macro is for detection of features which need kernel - * infrastructure to be used. It may *not* directly test the CPU - * itself. Use the cpu_has() family if you want true runtime - * testing of CPU features, like in hypervisor code where you are - * supporting a possible guest feature where host support for it + * This is the default CPU features testing macro to use in code. + * + * It is for detection of features which need kernel infrastructure to be + * used. It may *not* directly test the CPU itself. Use the cpu_has() family + * if you want true runtime testing of CPU features, like in hypervisor code + * where you are supporting a possible guest feature where host support for it * is not relevant. */ #define cpu_feature_enabled(bit) \ @@ -161,13 +162,6 @@ extern void clear_cpu_cap(struct cpuinfo_x86 *c, unsigned int bit); #define setup_force_cpu_bug(bit) setup_force_cpu_cap(bit) /* - * Static testing of CPU features. Used the same as boot_cpu_has(). It - * statically patches the target code for additional performance. Use - * static_cpu_has() only in fast paths, where every cycle counts. Which - * means that the boot_cpu_has() variant is already fast enough for the - * majority of cases and you should stick to using it as it is generally - * only two instructions: a RIP-relative MOV and a TEST. - * * Do not use an "m" constraint for [cap_byte] here: gcc doesn't know * that this is only used on a fallback path and will sometimes cause * it to manifest the address of boot_cpu_data in a register, fouling -- 2.51.0 From b8e10c86e674eb19e0e53dcf4fa3e71cba1e0c1c Mon Sep 17 00:00:00 2001 From: Dave Hansen Date: Fri, 13 Dec 2024 10:51:28 -0800 Subject: [PATCH 04/16] x86/cpu: Introduce new microcode matching helper The 'x86_cpu_id' and 'x86_cpu_desc' structures are very similar and need to be consolidated. There is a microcode version matching function for 'x86_cpu_desc' but not 'x86_cpu_id'. Create one for 'x86_cpu_id'. This essentially just leverages the x86_cpu_id->driver_data field to replace the less generic x86_cpu_desc->x86_microcode_rev field. Signed-off-by: Dave Hansen Link: https://lore.kernel.org/all/20241213185128.8F24EEFC%40davehans-spike.ostc.intel.com --- arch/x86/include/asm/cpu_device_id.h | 1 + arch/x86/kernel/cpu/match.c | 11 +++++++++++ 2 files changed, 12 insertions(+) diff --git a/arch/x86/include/asm/cpu_device_id.h b/arch/x86/include/asm/cpu_device_id.h index e4121d9aa9e1..9c77dbee96a7 100644 --- a/arch/x86/include/asm/cpu_device_id.h +++ b/arch/x86/include/asm/cpu_device_id.h @@ -278,5 +278,6 @@ struct x86_cpu_desc { extern const struct x86_cpu_id *x86_match_cpu(const struct x86_cpu_id *match); extern bool x86_cpu_has_min_microcode_rev(const struct x86_cpu_desc *table); +extern bool x86_match_min_microcode_rev(const struct x86_cpu_id *table); #endif /* _ASM_X86_CPU_DEVICE_ID */ diff --git a/arch/x86/kernel/cpu/match.c b/arch/x86/kernel/cpu/match.c index 8e7de733320a..2de2a83a7e12 100644 --- a/arch/x86/kernel/cpu/match.c +++ b/arch/x86/kernel/cpu/match.c @@ -86,3 +86,14 @@ bool x86_cpu_has_min_microcode_rev(const struct x86_cpu_desc *table) return true; } EXPORT_SYMBOL_GPL(x86_cpu_has_min_microcode_rev); + +bool x86_match_min_microcode_rev(const struct x86_cpu_id *table) +{ + const struct x86_cpu_id *res = x86_match_cpu(table); + + if (!res || res->driver_data > boot_cpu_data.microcode) + return false; + + return true; +} +EXPORT_SYMBOL_GPL(x86_match_min_microcode_rev); -- 2.51.0 From 85b08180df07b9a5984b15ae31d76b904d42a115 Mon Sep 17 00:00:00 2001 From: Dave Hansen Date: Fri, 13 Dec 2024 10:51:29 -0800 Subject: [PATCH 05/16] x86/cpu: Expose only stepping min/max interface The x86_match_cpu() infrastructure can match CPU steppings. Since there are only 16 possible steppings, the matching infrastructure goes all out and stores the stepping match as a bitmap. That means it can match any possible steppings in a single list entry. Fun. But it exposes this bitmap to each of the X86_MATCH_*() helpers when none of them really need a bitmap. It makes up for this by exporting a helper (X86_STEPPINGS()) which converts a contiguous stepping range into the bitmap which every single user leverages. Instead of a bitmap, have the main helper for this sort of thing (X86_MATCH_VFM_STEPS()) just take a stepping range. This ends up actually being even more compact than before. Leave the helper in place (renamed to __X86_STEPPINGS()) to make it more clear what is going on instead of just having a random GENMASK() in the middle of an already complicated macro. One oddity that I hit was this macro: X86_MATCH_VFM_STEPS(vfm, X86_STEPPING_MIN, max_stepping, issues) It *could* have been converted over to take a min/max stepping value for each entry. But that would have been a bit too verbose and would prevent the one oddball in the list (INTEL_COMETLAKE_L stepping 0) from sticking out. Instead, just have it take a *maximum* stepping and imply that the match is from 0=>max_stepping. This is functional for all the cases now and also retains the nice property of having INTEL_COMETLAKE_L stepping 0 stick out like a sore thumb. skx_cpuids[] is goofy. It uses the stepping match but encodes all possible steppings. Just use a normal, non-stepping match helper. Suggested-by: Ingo Molnar Signed-off-by: Dave Hansen Link: https://lore.kernel.org/all/20241213185129.65527B2A%40davehans-spike.ostc.intel.com --- arch/x86/include/asm/cpu_device_id.h | 15 +++--- arch/x86/kernel/apic/apic.c | 18 +++---- arch/x86/kernel/cpu/common.c | 78 ++++++++++++++-------------- drivers/edac/i10nm_base.c | 21 ++++---- drivers/edac/skx_base.c | 2 +- include/linux/mod_devicetable.h | 2 + 6 files changed, 70 insertions(+), 66 deletions(-) diff --git a/arch/x86/include/asm/cpu_device_id.h b/arch/x86/include/asm/cpu_device_id.h index 9c77dbee96a7..88564bb1845f 100644 --- a/arch/x86/include/asm/cpu_device_id.h +++ b/arch/x86/include/asm/cpu_device_id.h @@ -56,7 +56,6 @@ /* x86_cpu_id::flags */ #define X86_CPU_ID_FLAG_ENTRY_VALID BIT(0) -#define X86_STEPPINGS(mins, maxs) GENMASK(maxs, mins) /** * X86_MATCH_VENDOR_FAM_MODEL_STEPPINGS_FEATURE - Base macro for CPU matching * @_vendor: The vendor name, e.g. INTEL, AMD, HYGON, ..., ANY @@ -208,6 +207,7 @@ VFM_MODEL(vfm), \ X86_STEPPING_ANY, X86_FEATURE_ANY, data) +#define __X86_STEPPINGS(mins, maxs) GENMASK(maxs, mins) /** * X86_MATCH_VFM_STEPPINGS - Match encoded vendor/family/model/stepping * @vfm: Encoded 8-bits each for vendor, family, model @@ -218,12 +218,13 @@ * * feature is set to wildcard */ -#define X86_MATCH_VFM_STEPPINGS(vfm, steppings, data) \ - X86_MATCH_VENDORID_FAM_MODEL_STEPPINGS_FEATURE( \ - VFM_VENDOR(vfm), \ - VFM_FAMILY(vfm), \ - VFM_MODEL(vfm), \ - steppings, X86_FEATURE_ANY, data) +#define X86_MATCH_VFM_STEPS(vfm, min_step, max_step, data) \ + X86_MATCH_VENDORID_FAM_MODEL_STEPPINGS_FEATURE( \ + VFM_VENDOR(vfm), \ + VFM_FAMILY(vfm), \ + VFM_MODEL(vfm), \ + __X86_STEPPINGS(min_step, max_step), \ + X86_FEATURE_ANY, data) /** * X86_MATCH_VFM_FEATURE - Match encoded vendor/family/model/feature diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c index c5fb28e6451a..b16bda1ffaab 100644 --- a/arch/x86/kernel/apic/apic.c +++ b/arch/x86/kernel/apic/apic.c @@ -509,19 +509,19 @@ static struct clock_event_device lapic_clockevent = { static DEFINE_PER_CPU(struct clock_event_device, lapic_events); static const struct x86_cpu_id deadline_match[] __initconst = { - X86_MATCH_VFM_STEPPINGS(INTEL_HASWELL_X, X86_STEPPINGS(0x2, 0x2), 0x3a), /* EP */ - X86_MATCH_VFM_STEPPINGS(INTEL_HASWELL_X, X86_STEPPINGS(0x4, 0x4), 0x0f), /* EX */ + X86_MATCH_VFM_STEPS(INTEL_HASWELL_X, 0x2, 0x2, 0x3a), /* EP */ + X86_MATCH_VFM_STEPS(INTEL_HASWELL_X, 0x4, 0x4, 0x0f), /* EX */ X86_MATCH_VFM(INTEL_BROADWELL_X, 0x0b000020), - X86_MATCH_VFM_STEPPINGS(INTEL_BROADWELL_D, X86_STEPPINGS(0x2, 0x2), 0x00000011), - X86_MATCH_VFM_STEPPINGS(INTEL_BROADWELL_D, X86_STEPPINGS(0x3, 0x3), 0x0700000e), - X86_MATCH_VFM_STEPPINGS(INTEL_BROADWELL_D, X86_STEPPINGS(0x4, 0x4), 0x0f00000c), - X86_MATCH_VFM_STEPPINGS(INTEL_BROADWELL_D, X86_STEPPINGS(0x5, 0x5), 0x0e000003), + X86_MATCH_VFM_STEPS(INTEL_BROADWELL_D, 0x2, 0x2, 0x00000011), + X86_MATCH_VFM_STEPS(INTEL_BROADWELL_D, 0x3, 0x3, 0x0700000e), + X86_MATCH_VFM_STEPS(INTEL_BROADWELL_D, 0x4, 0x4, 0x0f00000c), + X86_MATCH_VFM_STEPS(INTEL_BROADWELL_D, 0x5, 0x5, 0x0e000003), - X86_MATCH_VFM_STEPPINGS(INTEL_SKYLAKE_X, X86_STEPPINGS(0x3, 0x3), 0x01000136), - X86_MATCH_VFM_STEPPINGS(INTEL_SKYLAKE_X, X86_STEPPINGS(0x4, 0x4), 0x02000014), - X86_MATCH_VFM_STEPPINGS(INTEL_SKYLAKE_X, X86_STEPPINGS(0x5, 0xf), 0), + X86_MATCH_VFM_STEPS(INTEL_SKYLAKE_X, 0x3, 0x3, 0x01000136), + X86_MATCH_VFM_STEPS(INTEL_SKYLAKE_X, 0x4, 0x4, 0x02000014), + X86_MATCH_VFM_STEPS(INTEL_SKYLAKE_X, 0x5, 0xf, 0), X86_MATCH_VFM(INTEL_HASWELL, 0x22), X86_MATCH_VFM(INTEL_HASWELL_L, 0x20), diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index a5c28975c608..d21b352bcd72 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -1201,8 +1201,8 @@ static const __initconst struct x86_cpu_id cpu_vuln_whitelist[] = { #define VULNBL(vendor, family, model, blacklist) \ X86_MATCH_VENDOR_FAM_MODEL(vendor, family, model, blacklist) -#define VULNBL_INTEL_STEPPINGS(vfm, steppings, issues) \ - X86_MATCH_VFM_STEPPINGS(vfm, steppings, issues) +#define VULNBL_INTEL_STEPS(vfm, max_stepping, issues) \ + X86_MATCH_VFM_STEPS(vfm, X86_STEP_MIN, max_stepping, issues) #define VULNBL_AMD(family, blacklist) \ VULNBL(AMD, family, X86_MODEL_ANY, blacklist) @@ -1227,43 +1227,43 @@ static const __initconst struct x86_cpu_id cpu_vuln_whitelist[] = { #define RFDS BIT(7) static const struct x86_cpu_id cpu_vuln_blacklist[] __initconst = { - VULNBL_INTEL_STEPPINGS(INTEL_IVYBRIDGE, X86_STEPPING_ANY, SRBDS), - VULNBL_INTEL_STEPPINGS(INTEL_HASWELL, X86_STEPPING_ANY, SRBDS), - VULNBL_INTEL_STEPPINGS(INTEL_HASWELL_L, X86_STEPPING_ANY, SRBDS), - VULNBL_INTEL_STEPPINGS(INTEL_HASWELL_G, X86_STEPPING_ANY, SRBDS), - VULNBL_INTEL_STEPPINGS(INTEL_HASWELL_X, X86_STEPPING_ANY, MMIO), - VULNBL_INTEL_STEPPINGS(INTEL_BROADWELL_D, X86_STEPPING_ANY, MMIO), - VULNBL_INTEL_STEPPINGS(INTEL_BROADWELL_G, X86_STEPPING_ANY, SRBDS), - VULNBL_INTEL_STEPPINGS(INTEL_BROADWELL_X, X86_STEPPING_ANY, MMIO), - VULNBL_INTEL_STEPPINGS(INTEL_BROADWELL, X86_STEPPING_ANY, SRBDS), - VULNBL_INTEL_STEPPINGS(INTEL_SKYLAKE_X, X86_STEPPING_ANY, MMIO | RETBLEED | GDS), - VULNBL_INTEL_STEPPINGS(INTEL_SKYLAKE_L, X86_STEPPING_ANY, MMIO | RETBLEED | GDS | SRBDS), - VULNBL_INTEL_STEPPINGS(INTEL_SKYLAKE, X86_STEPPING_ANY, MMIO | RETBLEED | GDS | SRBDS), - VULNBL_INTEL_STEPPINGS(INTEL_KABYLAKE_L, X86_STEPPING_ANY, MMIO | RETBLEED | GDS | SRBDS), - VULNBL_INTEL_STEPPINGS(INTEL_KABYLAKE, X86_STEPPING_ANY, MMIO | RETBLEED | GDS | SRBDS), - VULNBL_INTEL_STEPPINGS(INTEL_CANNONLAKE_L, X86_STEPPING_ANY, RETBLEED), - VULNBL_INTEL_STEPPINGS(INTEL_ICELAKE_L, X86_STEPPING_ANY, MMIO | MMIO_SBDS | RETBLEED | GDS), - VULNBL_INTEL_STEPPINGS(INTEL_ICELAKE_D, X86_STEPPING_ANY, MMIO | GDS), - VULNBL_INTEL_STEPPINGS(INTEL_ICELAKE_X, X86_STEPPING_ANY, MMIO | GDS), - VULNBL_INTEL_STEPPINGS(INTEL_COMETLAKE, X86_STEPPING_ANY, MMIO | MMIO_SBDS | RETBLEED | GDS), - VULNBL_INTEL_STEPPINGS(INTEL_COMETLAKE_L, X86_STEPPINGS(0x0, 0x0), MMIO | RETBLEED), - VULNBL_INTEL_STEPPINGS(INTEL_COMETLAKE_L, X86_STEPPING_ANY, MMIO | MMIO_SBDS | RETBLEED | GDS), - VULNBL_INTEL_STEPPINGS(INTEL_TIGERLAKE_L, X86_STEPPING_ANY, GDS), - VULNBL_INTEL_STEPPINGS(INTEL_TIGERLAKE, X86_STEPPING_ANY, GDS), - VULNBL_INTEL_STEPPINGS(INTEL_LAKEFIELD, X86_STEPPING_ANY, MMIO | MMIO_SBDS | RETBLEED), - VULNBL_INTEL_STEPPINGS(INTEL_ROCKETLAKE, X86_STEPPING_ANY, MMIO | RETBLEED | GDS), - VULNBL_INTEL_STEPPINGS(INTEL_ALDERLAKE, X86_STEPPING_ANY, RFDS), - VULNBL_INTEL_STEPPINGS(INTEL_ALDERLAKE_L, X86_STEPPING_ANY, RFDS), - VULNBL_INTEL_STEPPINGS(INTEL_RAPTORLAKE, X86_STEPPING_ANY, RFDS), - VULNBL_INTEL_STEPPINGS(INTEL_RAPTORLAKE_P, X86_STEPPING_ANY, RFDS), - VULNBL_INTEL_STEPPINGS(INTEL_RAPTORLAKE_S, X86_STEPPING_ANY, RFDS), - VULNBL_INTEL_STEPPINGS(INTEL_ATOM_GRACEMONT, X86_STEPPING_ANY, RFDS), - VULNBL_INTEL_STEPPINGS(INTEL_ATOM_TREMONT, X86_STEPPING_ANY, MMIO | MMIO_SBDS | RFDS), - VULNBL_INTEL_STEPPINGS(INTEL_ATOM_TREMONT_D, X86_STEPPING_ANY, MMIO | RFDS), - VULNBL_INTEL_STEPPINGS(INTEL_ATOM_TREMONT_L, X86_STEPPING_ANY, MMIO | MMIO_SBDS | RFDS), - VULNBL_INTEL_STEPPINGS(INTEL_ATOM_GOLDMONT, X86_STEPPING_ANY, RFDS), - VULNBL_INTEL_STEPPINGS(INTEL_ATOM_GOLDMONT_D, X86_STEPPING_ANY, RFDS), - VULNBL_INTEL_STEPPINGS(INTEL_ATOM_GOLDMONT_PLUS, X86_STEPPING_ANY, RFDS), + VULNBL_INTEL_STEPS(INTEL_IVYBRIDGE, X86_STEP_MAX, SRBDS), + VULNBL_INTEL_STEPS(INTEL_HASWELL, X86_STEP_MAX, SRBDS), + VULNBL_INTEL_STEPS(INTEL_HASWELL_L, X86_STEP_MAX, SRBDS), + VULNBL_INTEL_STEPS(INTEL_HASWELL_G, X86_STEP_MAX, SRBDS), + VULNBL_INTEL_STEPS(INTEL_HASWELL_X, X86_STEP_MAX, MMIO), + VULNBL_INTEL_STEPS(INTEL_BROADWELL_D, X86_STEP_MAX, MMIO), + VULNBL_INTEL_STEPS(INTEL_BROADWELL_G, X86_STEP_MAX, SRBDS), + VULNBL_INTEL_STEPS(INTEL_BROADWELL_X, X86_STEP_MAX, MMIO), + VULNBL_INTEL_STEPS(INTEL_BROADWELL, X86_STEP_MAX, SRBDS), + VULNBL_INTEL_STEPS(INTEL_SKYLAKE_X, X86_STEP_MAX, MMIO | RETBLEED | GDS), + VULNBL_INTEL_STEPS(INTEL_SKYLAKE_L, X86_STEP_MAX, MMIO | RETBLEED | GDS | SRBDS), + VULNBL_INTEL_STEPS(INTEL_SKYLAKE, X86_STEP_MAX, MMIO | RETBLEED | GDS | SRBDS), + VULNBL_INTEL_STEPS(INTEL_KABYLAKE_L, X86_STEP_MAX, MMIO | RETBLEED | GDS | SRBDS), + VULNBL_INTEL_STEPS(INTEL_KABYLAKE, X86_STEP_MAX, MMIO | RETBLEED | GDS | SRBDS), + VULNBL_INTEL_STEPS(INTEL_CANNONLAKE_L, X86_STEP_MAX, RETBLEED), + VULNBL_INTEL_STEPS(INTEL_ICELAKE_L, X86_STEP_MAX, MMIO | MMIO_SBDS | RETBLEED | GDS), + VULNBL_INTEL_STEPS(INTEL_ICELAKE_D, X86_STEP_MAX, MMIO | GDS), + VULNBL_INTEL_STEPS(INTEL_ICELAKE_X, X86_STEP_MAX, MMIO | GDS), + VULNBL_INTEL_STEPS(INTEL_COMETLAKE, X86_STEP_MAX, MMIO | MMIO_SBDS | RETBLEED | GDS), + VULNBL_INTEL_STEPS(INTEL_COMETLAKE_L, 0x0, MMIO | RETBLEED), + VULNBL_INTEL_STEPS(INTEL_COMETLAKE_L, X86_STEP_MAX, MMIO | MMIO_SBDS | RETBLEED | GDS), + VULNBL_INTEL_STEPS(INTEL_TIGERLAKE_L, X86_STEP_MAX, GDS), + VULNBL_INTEL_STEPS(INTEL_TIGERLAKE, X86_STEP_MAX, GDS), + VULNBL_INTEL_STEPS(INTEL_LAKEFIELD, X86_STEP_MAX, MMIO | MMIO_SBDS | RETBLEED), + VULNBL_INTEL_STEPS(INTEL_ROCKETLAKE, X86_STEP_MAX, MMIO | RETBLEED | GDS), + VULNBL_INTEL_STEPS(INTEL_ALDERLAKE, X86_STEP_MAX, RFDS), + VULNBL_INTEL_STEPS(INTEL_ALDERLAKE_L, X86_STEP_MAX, RFDS), + VULNBL_INTEL_STEPS(INTEL_RAPTORLAKE, X86_STEP_MAX, RFDS), + VULNBL_INTEL_STEPS(INTEL_RAPTORLAKE_P, X86_STEP_MAX, RFDS), + VULNBL_INTEL_STEPS(INTEL_RAPTORLAKE_S, X86_STEP_MAX, RFDS), + VULNBL_INTEL_STEPS(INTEL_ATOM_GRACEMONT, X86_STEP_MAX, RFDS), + VULNBL_INTEL_STEPS(INTEL_ATOM_TREMONT, X86_STEP_MAX, MMIO | MMIO_SBDS | RFDS), + VULNBL_INTEL_STEPS(INTEL_ATOM_TREMONT_D, X86_STEP_MAX, MMIO | RFDS), + VULNBL_INTEL_STEPS(INTEL_ATOM_TREMONT_L, X86_STEP_MAX, MMIO | MMIO_SBDS | RFDS), + VULNBL_INTEL_STEPS(INTEL_ATOM_GOLDMONT, X86_STEP_MAX, RFDS), + VULNBL_INTEL_STEPS(INTEL_ATOM_GOLDMONT_D, X86_STEP_MAX, RFDS), + VULNBL_INTEL_STEPS(INTEL_ATOM_GOLDMONT_PLUS, X86_STEP_MAX, RFDS), VULNBL_AMD(0x15, RETBLEED), VULNBL_AMD(0x16, RETBLEED), diff --git a/drivers/edac/i10nm_base.c b/drivers/edac/i10nm_base.c index 51556c72a967..09bf5a3f06bf 100644 --- a/drivers/edac/i10nm_base.c +++ b/drivers/edac/i10nm_base.c @@ -938,16 +938,17 @@ static struct res_config gnr_cfg = { }; static const struct x86_cpu_id i10nm_cpuids[] = { - X86_MATCH_VFM_STEPPINGS(INTEL_ATOM_TREMONT_D, X86_STEPPINGS(0x0, 0x3), &i10nm_cfg0), - X86_MATCH_VFM_STEPPINGS(INTEL_ATOM_TREMONT_D, X86_STEPPINGS(0x4, 0xf), &i10nm_cfg1), - X86_MATCH_VFM_STEPPINGS(INTEL_ICELAKE_X, X86_STEPPINGS(0x0, 0x3), &i10nm_cfg0), - X86_MATCH_VFM_STEPPINGS(INTEL_ICELAKE_X, X86_STEPPINGS(0x4, 0xf), &i10nm_cfg1), - X86_MATCH_VFM_STEPPINGS(INTEL_ICELAKE_D, X86_STEPPINGS(0x0, 0xf), &i10nm_cfg1), - X86_MATCH_VFM_STEPPINGS(INTEL_SAPPHIRERAPIDS_X, X86_STEPPINGS(0x0, 0xf), &spr_cfg), - X86_MATCH_VFM_STEPPINGS(INTEL_EMERALDRAPIDS_X, X86_STEPPINGS(0x0, 0xf), &spr_cfg), - X86_MATCH_VFM_STEPPINGS(INTEL_GRANITERAPIDS_X, X86_STEPPINGS(0x0, 0xf), &gnr_cfg), - X86_MATCH_VFM_STEPPINGS(INTEL_ATOM_CRESTMONT_X, X86_STEPPINGS(0x0, 0xf), &gnr_cfg), - X86_MATCH_VFM_STEPPINGS(INTEL_ATOM_CRESTMONT, X86_STEPPINGS(0x0, 0xf), &gnr_cfg), + X86_MATCH_VFM_STEPS(INTEL_ATOM_TREMONT_D, X86_STEP_MIN, 0x3, &i10nm_cfg0), + X86_MATCH_VFM_STEPS(INTEL_ATOM_TREMONT_D, 0x4, X86_STEP_MAX, &i10nm_cfg1), + X86_MATCH_VFM_STEPS(INTEL_ICELAKE_X, X86_STEP_MIN, 0x3, &i10nm_cfg0), + X86_MATCH_VFM_STEPS(INTEL_ICELAKE_X, 0x4, X86_STEP_MAX, &i10nm_cfg1), + X86_MATCH_VFM( INTEL_ICELAKE_D, &i10nm_cfg1), + + X86_MATCH_VFM(INTEL_SAPPHIRERAPIDS_X, &spr_cfg), + X86_MATCH_VFM(INTEL_EMERALDRAPIDS_X, &spr_cfg), + X86_MATCH_VFM(INTEL_GRANITERAPIDS_X, &gnr_cfg), + X86_MATCH_VFM(INTEL_ATOM_CRESTMONT_X, &gnr_cfg), + X86_MATCH_VFM(INTEL_ATOM_CRESTMONT, &gnr_cfg), {} }; MODULE_DEVICE_TABLE(x86cpu, i10nm_cpuids); diff --git a/drivers/edac/skx_base.c b/drivers/edac/skx_base.c index 14cfd394b469..fed5ecb4b0b1 100644 --- a/drivers/edac/skx_base.c +++ b/drivers/edac/skx_base.c @@ -164,7 +164,7 @@ static struct res_config skx_cfg = { }; static const struct x86_cpu_id skx_cpuids[] = { - X86_MATCH_VFM_STEPPINGS(INTEL_SKYLAKE_X, X86_STEPPINGS(0x0, 0xf), &skx_cfg), + X86_MATCH_VFM(INTEL_SKYLAKE_X, &skx_cfg), { } }; MODULE_DEVICE_TABLE(x86cpu, skx_cpuids); diff --git a/include/linux/mod_devicetable.h b/include/linux/mod_devicetable.h index 4338b1b4ac44..d67614f7b7f1 100644 --- a/include/linux/mod_devicetable.h +++ b/include/linux/mod_devicetable.h @@ -700,6 +700,8 @@ struct x86_cpu_id { #define X86_FAMILY_ANY 0 #define X86_MODEL_ANY 0 #define X86_STEPPING_ANY 0 +#define X86_STEP_MIN 0 +#define X86_STEP_MAX 0xf #define X86_FEATURE_ANY 0 /* Same as FPU, you can't test for that */ /* -- 2.51.0 From 3fa5626720c0948ce067306c4f6558d9ec86020c Mon Sep 17 00:00:00 2001 From: Dave Hansen Date: Fri, 13 Dec 2024 10:51:31 -0800 Subject: [PATCH 06/16] x86/cpu: Replace PEBS use of 'x86_cpu_desc' use with 'x86_cpu_id' The 'x86_cpu_desc' and 'x86_cpu_id' structures are very similar. Reduce duplicate infrastructure by moving the few users of 'x86_cpu_desc' to the much more common variant. The existing X86_MATCH_VFM_STEPS() helper matches ranges of steppings. Instead of introducing a single-stepping match function which could get confusing when paired with the range, just use the stepping min/max match helper and use min==max. Note that this makes the table more vertically compact because multiple entries like this: INTEL_CPU_DESC(INTEL_SKYLAKE_X, 4, 0x00000000), INTEL_CPU_DESC(INTEL_SKYLAKE_X, 5, 0x00000000), INTEL_CPU_DESC(INTEL_SKYLAKE_X, 6, 0x00000000), INTEL_CPU_DESC(INTEL_SKYLAKE_X, 7, 0x00000000), can be consolidated down to a single stepping range. Signed-off-by: Dave Hansen Link: https://lore.kernel.org/all/20241213185131.8B610039%40davehans-spike.ostc.intel.com --- arch/x86/events/intel/core.c | 62 +++++++++++++++--------------------- 1 file changed, 26 insertions(+), 36 deletions(-) diff --git a/arch/x86/events/intel/core.c b/arch/x86/events/intel/core.c index bb284aff7bfd..cd96013eea79 100644 --- a/arch/x86/events/intel/core.c +++ b/arch/x86/events/intel/core.c @@ -5371,42 +5371,32 @@ static __init void intel_clovertown_quirk(void) x86_pmu.pebs_constraints = NULL; } -static const struct x86_cpu_desc isolation_ucodes[] = { - INTEL_CPU_DESC(INTEL_HASWELL, 3, 0x0000001f), - INTEL_CPU_DESC(INTEL_HASWELL_L, 1, 0x0000001e), - INTEL_CPU_DESC(INTEL_HASWELL_G, 1, 0x00000015), - INTEL_CPU_DESC(INTEL_HASWELL_X, 2, 0x00000037), - INTEL_CPU_DESC(INTEL_HASWELL_X, 4, 0x0000000a), - INTEL_CPU_DESC(INTEL_BROADWELL, 4, 0x00000023), - INTEL_CPU_DESC(INTEL_BROADWELL_G, 1, 0x00000014), - INTEL_CPU_DESC(INTEL_BROADWELL_D, 2, 0x00000010), - INTEL_CPU_DESC(INTEL_BROADWELL_D, 3, 0x07000009), - INTEL_CPU_DESC(INTEL_BROADWELL_D, 4, 0x0f000009), - INTEL_CPU_DESC(INTEL_BROADWELL_D, 5, 0x0e000002), - INTEL_CPU_DESC(INTEL_BROADWELL_X, 1, 0x0b000014), - INTEL_CPU_DESC(INTEL_SKYLAKE_X, 3, 0x00000021), - INTEL_CPU_DESC(INTEL_SKYLAKE_X, 4, 0x00000000), - INTEL_CPU_DESC(INTEL_SKYLAKE_X, 5, 0x00000000), - INTEL_CPU_DESC(INTEL_SKYLAKE_X, 6, 0x00000000), - INTEL_CPU_DESC(INTEL_SKYLAKE_X, 7, 0x00000000), - INTEL_CPU_DESC(INTEL_SKYLAKE_X, 11, 0x00000000), - INTEL_CPU_DESC(INTEL_SKYLAKE_L, 3, 0x0000007c), - INTEL_CPU_DESC(INTEL_SKYLAKE, 3, 0x0000007c), - INTEL_CPU_DESC(INTEL_KABYLAKE, 9, 0x0000004e), - INTEL_CPU_DESC(INTEL_KABYLAKE_L, 9, 0x0000004e), - INTEL_CPU_DESC(INTEL_KABYLAKE_L, 10, 0x0000004e), - INTEL_CPU_DESC(INTEL_KABYLAKE_L, 11, 0x0000004e), - INTEL_CPU_DESC(INTEL_KABYLAKE_L, 12, 0x0000004e), - INTEL_CPU_DESC(INTEL_KABYLAKE, 10, 0x0000004e), - INTEL_CPU_DESC(INTEL_KABYLAKE, 11, 0x0000004e), - INTEL_CPU_DESC(INTEL_KABYLAKE, 12, 0x0000004e), - INTEL_CPU_DESC(INTEL_KABYLAKE, 13, 0x0000004e), +static const struct x86_cpu_id isolation_ucodes[] = { + X86_MATCH_VFM_STEPS(INTEL_HASWELL, 3, 3, 0x0000001f), + X86_MATCH_VFM_STEPS(INTEL_HASWELL_L, 1, 1, 0x0000001e), + X86_MATCH_VFM_STEPS(INTEL_HASWELL_G, 1, 1, 0x00000015), + X86_MATCH_VFM_STEPS(INTEL_HASWELL_X, 2, 2, 0x00000037), + X86_MATCH_VFM_STEPS(INTEL_HASWELL_X, 4, 4, 0x0000000a), + X86_MATCH_VFM_STEPS(INTEL_BROADWELL, 4, 4, 0x00000023), + X86_MATCH_VFM_STEPS(INTEL_BROADWELL_G, 1, 1, 0x00000014), + X86_MATCH_VFM_STEPS(INTEL_BROADWELL_D, 2, 2, 0x00000010), + X86_MATCH_VFM_STEPS(INTEL_BROADWELL_D, 3, 3, 0x07000009), + X86_MATCH_VFM_STEPS(INTEL_BROADWELL_D, 4, 4, 0x0f000009), + X86_MATCH_VFM_STEPS(INTEL_BROADWELL_D, 5, 5, 0x0e000002), + X86_MATCH_VFM_STEPS(INTEL_BROADWELL_X, 1, 1, 0x0b000014), + X86_MATCH_VFM_STEPS(INTEL_SKYLAKE_X, 3, 3, 0x00000021), + X86_MATCH_VFM_STEPS(INTEL_SKYLAKE_X, 4, 7, 0x00000000), + X86_MATCH_VFM_STEPS(INTEL_SKYLAKE_X, 11, 11, 0x00000000), + X86_MATCH_VFM_STEPS(INTEL_SKYLAKE_L, 3, 3, 0x0000007c), + X86_MATCH_VFM_STEPS(INTEL_SKYLAKE, 3, 3, 0x0000007c), + X86_MATCH_VFM_STEPS(INTEL_KABYLAKE, 9, 13, 0x0000004e), + X86_MATCH_VFM_STEPS(INTEL_KABYLAKE_L, 9, 12, 0x0000004e), {} }; static void intel_check_pebs_isolation(void) { - x86_pmu.pebs_no_isolation = !x86_cpu_has_min_microcode_rev(isolation_ucodes); + x86_pmu.pebs_no_isolation = !x86_match_min_microcode_rev(isolation_ucodes); } static __init void intel_pebs_isolation_quirk(void) @@ -5416,16 +5406,16 @@ static __init void intel_pebs_isolation_quirk(void) intel_check_pebs_isolation(); } -static const struct x86_cpu_desc pebs_ucodes[] = { - INTEL_CPU_DESC(INTEL_SANDYBRIDGE, 7, 0x00000028), - INTEL_CPU_DESC(INTEL_SANDYBRIDGE_X, 6, 0x00000618), - INTEL_CPU_DESC(INTEL_SANDYBRIDGE_X, 7, 0x0000070c), +static const struct x86_cpu_id pebs_ucodes[] = { + X86_MATCH_VFM_STEPS(INTEL_SANDYBRIDGE, 7, 7, 0x00000028), + X86_MATCH_VFM_STEPS(INTEL_SANDYBRIDGE_X, 6, 6, 0x00000618), + X86_MATCH_VFM_STEPS(INTEL_SANDYBRIDGE_X, 7, 7, 0x0000070c), {} }; static bool intel_snb_pebs_broken(void) { - return !x86_cpu_has_min_microcode_rev(pebs_ucodes); + return !x86_match_min_microcode_rev(pebs_ucodes); } static void intel_snb_check_microcode(void) -- 2.51.0 From f3f3251526739bb975b97f840c56b3054dba8638 Mon Sep 17 00:00:00 2001 From: Dave Hansen Date: Fri, 13 Dec 2024 10:51:32 -0800 Subject: [PATCH 07/16] x86/cpu: Move AMD erratum 1386 table over to 'x86_cpu_id' The AMD erratum 1386 detection code uses and old style 'x86_cpu_desc' table. Replace it with 'x86_cpu_id' so the old style can be removed. I did not create a new helper macro here. The new table is certainly more noisy than the old and it can be improved on. But I was hesitant to create a new macro just for a single site that is only two ugly lines in the end. Signed-off-by: Dave Hansen Link: https://lore.kernel.org/all/20241213185132.07555E1D%40davehans-spike.ostc.intel.com --- arch/x86/kernel/cpu/amd.c | 9 ++++----- 1 file changed, 4 insertions(+), 5 deletions(-) diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index d8408aafeed9..7bb5b1ad18c2 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -795,10 +795,9 @@ static void init_amd_bd(struct cpuinfo_x86 *c) clear_rdrand_cpuid_bit(c); } -static const struct x86_cpu_desc erratum_1386_microcode[] = { - AMD_CPU_DESC(0x17, 0x1, 0x2, 0x0800126e), - AMD_CPU_DESC(0x17, 0x31, 0x0, 0x08301052), - {}, +static const struct x86_cpu_id erratum_1386_microcode[] = { + X86_MATCH_VFM_STEPS(VFM_MAKE(X86_VENDOR_AMD, 0x17, 0x01), 0x2, 0x2, 0x0800126e), + X86_MATCH_VFM_STEPS(VFM_MAKE(X86_VENDOR_AMD, 0x17, 0x31), 0x0, 0x0, 0x08301052), }; static void fix_erratum_1386(struct cpuinfo_x86 *c) @@ -814,7 +813,7 @@ static void fix_erratum_1386(struct cpuinfo_x86 *c) * Clear the feature flag only on microcode revisions which * don't have the fix. */ - if (x86_cpu_has_min_microcode_rev(erratum_1386_microcode)) + if (x86_match_min_microcode_rev(erratum_1386_microcode)) return; clear_cpu_cap(c, X86_FEATURE_XSAVES); -- 2.51.0 From 5366d8965d35f0ea266c80e8970aa9527a9fee52 Mon Sep 17 00:00:00 2001 From: Dave Hansen Date: Fri, 13 Dec 2024 10:51:33 -0800 Subject: [PATCH 08/16] x86/cpu: Remove 'x86_cpu_desc' infrastructure All the users of 'x86_cpu_desc' are gone. Zap it from the tree. Signed-off-by: Dave Hansen Link: https://lore.kernel.org/all/20241213185133.AF0BF2BC%40davehans-spike.ostc.intel.com --- arch/x86/include/asm/cpu_device_id.h | 35 ---------------------------- arch/x86/kernel/cpu/match.c | 31 ------------------------ 2 files changed, 66 deletions(-) diff --git a/arch/x86/include/asm/cpu_device_id.h b/arch/x86/include/asm/cpu_device_id.h index 88564bb1845f..ba32e0f44cba 100644 --- a/arch/x86/include/asm/cpu_device_id.h +++ b/arch/x86/include/asm/cpu_device_id.h @@ -243,42 +243,7 @@ VFM_MODEL(vfm), \ X86_STEPPING_ANY, feature, data) -/* - * Match specific microcode revisions. - * - * vendor/family/model/stepping must be all set. - * - * Only checks against the boot CPU. When mixed-stepping configs are - * valid for a CPU model, add a quirk for every valid stepping and - * do the fine-tuning in the quirk handler. - */ - -struct x86_cpu_desc { - u8 x86_family; - u8 x86_vendor; - u8 x86_model; - u8 x86_stepping; - u32 x86_microcode_rev; -}; - -#define INTEL_CPU_DESC(vfm, stepping, revision) { \ - .x86_family = VFM_FAMILY(vfm), \ - .x86_vendor = VFM_VENDOR(vfm), \ - .x86_model = VFM_MODEL(vfm), \ - .x86_stepping = (stepping), \ - .x86_microcode_rev = (revision), \ -} - -#define AMD_CPU_DESC(fam, model, stepping, revision) { \ - .x86_family = (fam), \ - .x86_vendor = X86_VENDOR_AMD, \ - .x86_model = (model), \ - .x86_stepping = (stepping), \ - .x86_microcode_rev = (revision), \ -} - extern const struct x86_cpu_id *x86_match_cpu(const struct x86_cpu_id *match); -extern bool x86_cpu_has_min_microcode_rev(const struct x86_cpu_desc *table); extern bool x86_match_min_microcode_rev(const struct x86_cpu_id *table); #endif /* _ASM_X86_CPU_DEVICE_ID */ diff --git a/arch/x86/kernel/cpu/match.c b/arch/x86/kernel/cpu/match.c index 2de2a83a7e12..1a714f70be84 100644 --- a/arch/x86/kernel/cpu/match.c +++ b/arch/x86/kernel/cpu/match.c @@ -56,37 +56,6 @@ const struct x86_cpu_id *x86_match_cpu(const struct x86_cpu_id *match) } EXPORT_SYMBOL(x86_match_cpu); -static const struct x86_cpu_desc * -x86_match_cpu_with_stepping(const struct x86_cpu_desc *match) -{ - struct cpuinfo_x86 *c = &boot_cpu_data; - const struct x86_cpu_desc *m; - - for (m = match; m->x86_family | m->x86_model; m++) { - if (c->x86_vendor != m->x86_vendor) - continue; - if (c->x86 != m->x86_family) - continue; - if (c->x86_model != m->x86_model) - continue; - if (c->x86_stepping != m->x86_stepping) - continue; - return m; - } - return NULL; -} - -bool x86_cpu_has_min_microcode_rev(const struct x86_cpu_desc *table) -{ - const struct x86_cpu_desc *res = x86_match_cpu_with_stepping(table); - - if (!res || res->x86_microcode_rev > boot_cpu_data.microcode) - return false; - - return true; -} -EXPORT_SYMBOL_GPL(x86_cpu_has_min_microcode_rev); - bool x86_match_min_microcode_rev(const struct x86_cpu_id *table) { const struct x86_cpu_id *res = x86_match_cpu(table); -- 2.51.0 From 497f70284695bbb9b875e182554ef3f18b4a56e2 Mon Sep 17 00:00:00 2001 From: Dave Hansen Date: Fri, 13 Dec 2024 12:50:28 -0800 Subject: [PATCH 09/16] x86/cpu: Move MWAIT leaf definition to common header Begin constructing a common place to keep all CPUID leaf definitions. Move CPUID_MWAIT_LEAF to the CPUID header and include it where needed. Signed-off-by: Dave Hansen Reviewed-by: Zhao Liu Link: https://lore.kernel.org/all/20241213205028.EE94D02A%40davehans-spike.ostc.intel.com --- arch/x86/include/asm/cpuid.h | 2 ++ arch/x86/include/asm/mwait.h | 1 - arch/x86/kernel/acpi/cstate.c | 1 + arch/x86/kernel/hpet.c | 1 + arch/x86/kernel/process.c | 1 + arch/x86/kernel/smpboot.c | 1 + arch/x86/xen/enlighten_pv.c | 1 + drivers/acpi/acpi_pad.c | 1 + drivers/idle/intel_idle.c | 1 + 9 files changed, 9 insertions(+), 1 deletion(-) diff --git a/arch/x86/include/asm/cpuid.h b/arch/x86/include/asm/cpuid.h index 239b9ba5c398..13ecab94cc23 100644 --- a/arch/x86/include/asm/cpuid.h +++ b/arch/x86/include/asm/cpuid.h @@ -21,6 +21,8 @@ enum cpuid_regs_idx { CPUID_EDX, }; +#define CPUID_MWAIT_LEAF 5 + #ifdef CONFIG_X86_32 bool have_cpuid_p(void); #else diff --git a/arch/x86/include/asm/mwait.h b/arch/x86/include/asm/mwait.h index 920426d691ce..ce857ef54cf1 100644 --- a/arch/x86/include/asm/mwait.h +++ b/arch/x86/include/asm/mwait.h @@ -15,7 +15,6 @@ #define MWAIT_HINT2SUBSTATE(hint) ((hint) & MWAIT_CSTATE_MASK) #define MWAIT_C1_SUBSTATE_MASK 0xf0 -#define CPUID_MWAIT_LEAF 5 #define CPUID5_ECX_EXTENSIONS_SUPPORTED 0x1 #define CPUID5_ECX_INTERRUPT_BREAK 0x2 diff --git a/arch/x86/kernel/acpi/cstate.c b/arch/x86/kernel/acpi/cstate.c index f3ffd0a3a012..2779a93769e8 100644 --- a/arch/x86/kernel/acpi/cstate.c +++ b/arch/x86/kernel/acpi/cstate.c @@ -13,6 +13,7 @@ #include #include +#include #include #include diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c index c96ae8fee95e..25935041fee1 100644 --- a/arch/x86/kernel/hpet.c +++ b/arch/x86/kernel/hpet.c @@ -7,6 +7,7 @@ #include #include +#include #include #include #include diff --git a/arch/x86/kernel/process.c b/arch/x86/kernel/process.c index 58ead05a1c29..d40fc4965c14 100644 --- a/arch/x86/kernel/process.c +++ b/arch/x86/kernel/process.c @@ -30,6 +30,7 @@ #include #include #include +#include #include #include #include diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index b5a8f0891135..52b0d308eed9 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -64,6 +64,7 @@ #include #include +#include #include #include #include diff --git a/arch/x86/xen/enlighten_pv.c b/arch/x86/xen/enlighten_pv.c index fd2169063480..b355070e92fa 100644 --- a/arch/x86/xen/enlighten_pv.c +++ b/arch/x86/xen/enlighten_pv.c @@ -49,6 +49,7 @@ #include #include +#include #include #include #include diff --git a/drivers/acpi/acpi_pad.c b/drivers/acpi/acpi_pad.c index 4ec20fd56985..b56197467352 100644 --- a/drivers/acpi/acpi_pad.c +++ b/drivers/acpi/acpi_pad.c @@ -19,6 +19,7 @@ #include #include #include +#include #include #include diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c index ac4d8faa3886..5d8ed1a68447 100644 --- a/drivers/idle/intel_idle.c +++ b/drivers/idle/intel_idle.c @@ -51,6 +51,7 @@ #include #include #include +#include #include #include #include -- 2.51.0 From 8bd6821c9cf3b81d3c07a94fa4e3f97a3cc7b724 Mon Sep 17 00:00:00 2001 From: Dave Hansen Date: Fri, 13 Dec 2024 12:50:29 -0800 Subject: [PATCH 10/16] x86/cpu: Use MWAIT leaf definition The leaf-to-feature dependency array uses hard-coded leaf numbers. Use the new common header definition for the MWAIT leaf. Signed-off-by: Dave Hansen Reviewed-by: Zhao Liu Link: https://lore.kernel.org/all/20241213205029.5B055D6E%40davehans-spike.ostc.intel.com --- arch/x86/kernel/cpu/common.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index d21b352bcd72..853e373d2829 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -29,6 +29,7 @@ #include #include +#include #include #include #include @@ -636,7 +637,7 @@ struct cpuid_dependent_feature { static const struct cpuid_dependent_feature cpuid_dependent_features[] = { - { X86_FEATURE_MWAIT, 0x00000005 }, + { X86_FEATURE_MWAIT, CPUID_MWAIT_LEAF }, { X86_FEATURE_DCA, 0x00000009 }, { X86_FEATURE_XSAVE, 0x0000000d }, { 0, 0 } -- 2.51.0 From 262fba55708b60a063b30d103963477dc5026f8c Mon Sep 17 00:00:00 2001 From: Dave Hansen Date: Fri, 13 Dec 2024 12:50:30 -0800 Subject: [PATCH 11/16] x86/cpu: Remove unnecessary MwAIT leaf checks The CPUID leaf dependency checker will remove X86_FEATURE_MWAIT if the CPUID level is below the required level (CPUID_MWAIT_LEAF). Thus, if you check X86_FEATURE_MWAIT you do not need to also check the CPUID level. Signed-off-by: Dave Hansen Link: https://lore.kernel.org/all/20241213205030.9B42B458%40davehans-spike.ostc.intel.com --- arch/x86/kernel/hpet.c | 3 --- arch/x86/kernel/smpboot.c | 2 -- drivers/acpi/acpi_pad.c | 2 -- drivers/idle/intel_idle.c | 3 --- 4 files changed, 10 deletions(-) diff --git a/arch/x86/kernel/hpet.c b/arch/x86/kernel/hpet.c index 25935041fee1..953de5b64669 100644 --- a/arch/x86/kernel/hpet.c +++ b/arch/x86/kernel/hpet.c @@ -928,9 +928,6 @@ static bool __init mwait_pc10_supported(void) if (!cpu_feature_enabled(X86_FEATURE_MWAIT)) return false; - if (boot_cpu_data.cpuid_level < CPUID_MWAIT_LEAF) - return false; - cpuid(CPUID_MWAIT_LEAF, &eax, &ebx, &ecx, &mwait_substates); return (ecx & CPUID5_ECX_EXTENSIONS_SUPPORTED) && diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c index 52b0d308eed9..116c46f9ecbb 100644 --- a/arch/x86/kernel/smpboot.c +++ b/arch/x86/kernel/smpboot.c @@ -1292,8 +1292,6 @@ static inline void mwait_play_dead(void) return; if (!this_cpu_has(X86_FEATURE_CLFLUSH)) return; - if (__this_cpu_read(cpu_info.cpuid_level) < CPUID_MWAIT_LEAF) - return; eax = CPUID_MWAIT_LEAF; ecx = 0; diff --git a/drivers/acpi/acpi_pad.c b/drivers/acpi/acpi_pad.c index b56197467352..f3cffae0c14e 100644 --- a/drivers/acpi/acpi_pad.c +++ b/drivers/acpi/acpi_pad.c @@ -47,8 +47,6 @@ static void power_saving_mwait_init(void) if (!boot_cpu_has(X86_FEATURE_MWAIT)) return; - if (boot_cpu_data.cpuid_level < CPUID_MWAIT_LEAF) - return; cpuid(CPUID_MWAIT_LEAF, &eax, &ebx, &ecx, &edx); diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c index 5d8ed1a68447..efa32d28ed93 100644 --- a/drivers/idle/intel_idle.c +++ b/drivers/idle/intel_idle.c @@ -2317,9 +2317,6 @@ static int __init intel_idle_init(void) return -ENODEV; } - if (boot_cpu_data.cpuid_level < CPUID_MWAIT_LEAF) - return -ENODEV; - cpuid(CPUID_MWAIT_LEAF, &eax, &ebx, &ecx, &mwait_substates); if (!(ecx & CPUID5_ECX_EXTENSIONS_SUPPORTED) || -- 2.51.0 From 5d82d8e0a9ac06bfc6ac59407b96bc357eff441a Mon Sep 17 00:00:00 2001 From: Dave Hansen Date: Fri, 13 Dec 2024 12:50:32 -0800 Subject: [PATCH 12/16] x86/cpu: Refresh DCA leaf reading code The DCA leaf number is also hard-coded in the CPUID level dependency table. Move its definition to common code and use it. While at it, fix up the naming and types in the probe code. All CPUID data is provided in 32-bit registers, not 'unsigned long'. Also stop referring to "level_9". Move away from test_bit() because the type is no longer an 'unsigned long'. Signed-off-by: Dave Hansen Reviewed-by: Zhao Liu Link: https://lore.kernel.org/all/20241213205032.476A30FE%40davehans-spike.ostc.intel.com --- arch/x86/include/asm/cpuid.h | 3 ++- arch/x86/kernel/cpu/common.c | 2 +- drivers/dma/ioat/dca.c | 8 +++++--- 3 files changed, 8 insertions(+), 5 deletions(-) diff --git a/arch/x86/include/asm/cpuid.h b/arch/x86/include/asm/cpuid.h index 13ecab94cc23..8ba4d9fdc9e7 100644 --- a/arch/x86/include/asm/cpuid.h +++ b/arch/x86/include/asm/cpuid.h @@ -21,7 +21,8 @@ enum cpuid_regs_idx { CPUID_EDX, }; -#define CPUID_MWAIT_LEAF 5 +#define CPUID_MWAIT_LEAF 0x5 +#define CPUID_DCA_LEAF 0x9 #ifdef CONFIG_X86_32 bool have_cpuid_p(void); diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 853e373d2829..5ffa1f4eac38 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -638,7 +638,7 @@ struct cpuid_dependent_feature { static const struct cpuid_dependent_feature cpuid_dependent_features[] = { { X86_FEATURE_MWAIT, CPUID_MWAIT_LEAF }, - { X86_FEATURE_DCA, 0x00000009 }, + { X86_FEATURE_DCA, CPUID_DCA_LEAF }, { X86_FEATURE_XSAVE, 0x0000000d }, { 0, 0 } }; diff --git a/drivers/dma/ioat/dca.c b/drivers/dma/ioat/dca.c index 17f6b6367113..658ea2ec36f7 100644 --- a/drivers/dma/ioat/dca.c +++ b/drivers/dma/ioat/dca.c @@ -10,6 +10,8 @@ #include #include +#include + /* either a kernel change is needed, or we need something like this in kernel */ #ifndef CONFIG_SMP #include @@ -58,11 +60,11 @@ static int dca_enabled_in_bios(struct pci_dev *pdev) { /* CPUID level 9 returns DCA configuration */ /* Bit 0 indicates DCA enabled by the BIOS */ - unsigned long cpuid_level_9; + u32 eax; int res; - cpuid_level_9 = cpuid_eax(9); - res = test_bit(0, &cpuid_level_9); + eax = cpuid_eax(CPUID_DCA_LEAF); + res = eax & BIT(0); if (!res) dev_dbg(&pdev->dev, "DCA is disabled in BIOS\n"); -- 2.51.0 From a86740a77bf0942e618cc5f022336cdd99530d10 Mon Sep 17 00:00:00 2001 From: Dave Hansen Date: Fri, 13 Dec 2024 12:50:33 -0800 Subject: [PATCH 13/16] x86/cpu: Move TSC CPUID leaf definition Prepare to use the TSC CPUID leaf definition more widely by moving it to the common header. Signed-off-by: Dave Hansen Reviewed-by: Zhao Liu Link: https://lore.kernel.org/all/20241213205033.68799E53%40davehans-spike.ostc.intel.com --- arch/x86/events/intel/pt.c | 1 + arch/x86/events/intel/pt.h | 3 --- arch/x86/include/asm/cpuid.h | 1 + 3 files changed, 2 insertions(+), 3 deletions(-) diff --git a/arch/x86/events/intel/pt.c b/arch/x86/events/intel/pt.c index 4b0373bc8ab4..608145566fca 100644 --- a/arch/x86/events/intel/pt.c +++ b/arch/x86/events/intel/pt.c @@ -18,6 +18,7 @@ #include #include +#include #include #include #include diff --git a/arch/x86/events/intel/pt.h b/arch/x86/events/intel/pt.h index 7ee94fc6d7cb..2ac36250b656 100644 --- a/arch/x86/events/intel/pt.h +++ b/arch/x86/events/intel/pt.h @@ -37,9 +37,6 @@ struct topa_entry { u64 rsvd4 : 12; }; -/* TSC to Core Crystal Clock Ratio */ -#define CPUID_TSC_LEAF 0x15 - struct pt_pmu { struct pmu pmu; u32 caps[PT_CPUID_REGS_NUM * PT_CPUID_LEAVES]; diff --git a/arch/x86/include/asm/cpuid.h b/arch/x86/include/asm/cpuid.h index 8ba4d9fdc9e7..9b0d14bfd2f2 100644 --- a/arch/x86/include/asm/cpuid.h +++ b/arch/x86/include/asm/cpuid.h @@ -23,6 +23,7 @@ enum cpuid_regs_idx { #define CPUID_MWAIT_LEAF 0x5 #define CPUID_DCA_LEAF 0x9 +#define CPUID_TSC_LEAF 0x15 #ifdef CONFIG_X86_32 bool have_cpuid_p(void); -- 2.51.0 From 030c15b5610cedf7eb428dab5382f73d492a7967 Mon Sep 17 00:00:00 2001 From: Dave Hansen Date: Fri, 13 Dec 2024 12:50:34 -0800 Subject: [PATCH 14/16] x86/tsc: Move away from TSC leaf magic numbers The TSC code has a bunch of hard-coded references to leaf 0x15. Change them over to the symbolic name. Also zap the 'ART_CPUID_LEAF' definition. It was a duplicate of 'CPUID_TSC_LEAF'. Signed-off-by: Dave Hansen Link: https://lore.kernel.org/all/20241213205034.B79D6224%40davehans-spike.ostc.intel.com --- arch/x86/kernel/tsc.c | 11 +++++------ drivers/platform/x86/intel/pmc/core.c | 7 ++++--- 2 files changed, 9 insertions(+), 9 deletions(-) diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c index 67aeaba4ba9c..8091b0ea7de2 100644 --- a/arch/x86/kernel/tsc.c +++ b/arch/x86/kernel/tsc.c @@ -16,6 +16,7 @@ #include #include +#include #include #include #include @@ -665,13 +666,13 @@ unsigned long native_calibrate_tsc(void) if (boot_cpu_data.x86_vendor != X86_VENDOR_INTEL) return 0; - if (boot_cpu_data.cpuid_level < 0x15) + if (boot_cpu_data.cpuid_level < CPUID_TSC_LEAF) return 0; eax_denominator = ebx_numerator = ecx_hz = edx = 0; /* CPUID 15H TSC/Crystal ratio, plus optionally Crystal Hz */ - cpuid(0x15, &eax_denominator, &ebx_numerator, &ecx_hz, &edx); + cpuid(CPUID_TSC_LEAF, &eax_denominator, &ebx_numerator, &ecx_hz, &edx); if (ebx_numerator == 0 || eax_denominator == 0) return 0; @@ -1067,10 +1068,8 @@ core_initcall(cpufreq_register_tsc_scaling); #endif /* CONFIG_CPU_FREQ */ -#define ART_CPUID_LEAF (0x15) #define ART_MIN_DENOMINATOR (1) - /* * If ART is present detect the numerator:denominator to convert to TSC */ @@ -1078,7 +1077,7 @@ static void __init detect_art(void) { unsigned int unused; - if (boot_cpu_data.cpuid_level < ART_CPUID_LEAF) + if (boot_cpu_data.cpuid_level < CPUID_TSC_LEAF) return; /* @@ -1091,7 +1090,7 @@ static void __init detect_art(void) tsc_async_resets) return; - cpuid(ART_CPUID_LEAF, &art_base_clk.denominator, + cpuid(CPUID_TSC_LEAF, &art_base_clk.denominator, &art_base_clk.numerator, &art_base_clk.freq_khz, &unused); art_base_clk.freq_khz /= KHZ; diff --git a/drivers/platform/x86/intel/pmc/core.c b/drivers/platform/x86/intel/pmc/core.c index 3e7f99ac8c94..ac8231e2f0c6 100644 --- a/drivers/platform/x86/intel/pmc/core.c +++ b/drivers/platform/x86/intel/pmc/core.c @@ -22,6 +22,7 @@ #include #include +#include #include #include #include @@ -935,13 +936,13 @@ static unsigned int pmc_core_get_crystal_freq(void) { unsigned int eax_denominator, ebx_numerator, ecx_hz, edx; - if (boot_cpu_data.cpuid_level < 0x15) + if (boot_cpu_data.cpuid_level < CPUID_TSC_LEAF) return 0; eax_denominator = ebx_numerator = ecx_hz = edx = 0; - /* CPUID 15H TSC/Crystal ratio, plus optionally Crystal Hz */ - cpuid(0x15, &eax_denominator, &ebx_numerator, &ecx_hz, &edx); + /* TSC/Crystal ratio, plus optionally Crystal Hz */ + cpuid(CPUID_TSC_LEAF, &eax_denominator, &ebx_numerator, &ecx_hz, &edx); if (ebx_numerator == 0 || eax_denominator == 0) return 0; -- 2.51.0 From e558eadf6bd6199a3f454299de3c6338931d4e46 Mon Sep 17 00:00:00 2001 From: Dave Hansen Date: Fri, 13 Dec 2024 12:50:36 -0800 Subject: [PATCH 15/16] x86/tsc: Remove CPUID "frequency" leaf magic numbers. All the code that reads the CPUID frequency information leaf hard-codes a magic number. Give it a symbolic name and use it. Signed-off-by: Dave Hansen Reviewed-by: Zhao Liu Link: https://lore.kernel.org/all/20241213205036.4397658F%40davehans-spike.ostc.intel.com --- arch/x86/include/asm/cpuid.h | 1 + arch/x86/kernel/tsc.c | 12 ++++++------ 2 files changed, 7 insertions(+), 6 deletions(-) diff --git a/arch/x86/include/asm/cpuid.h b/arch/x86/include/asm/cpuid.h index 9b0d14bfd2f2..e7803c21a3fd 100644 --- a/arch/x86/include/asm/cpuid.h +++ b/arch/x86/include/asm/cpuid.h @@ -24,6 +24,7 @@ enum cpuid_regs_idx { #define CPUID_MWAIT_LEAF 0x5 #define CPUID_DCA_LEAF 0x9 #define CPUID_TSC_LEAF 0x15 +#define CPUID_FREQ_LEAF 0x16 #ifdef CONFIG_X86_32 bool have_cpuid_p(void); diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c index 8091b0ea7de2..678c36f5cd4a 100644 --- a/arch/x86/kernel/tsc.c +++ b/arch/x86/kernel/tsc.c @@ -681,8 +681,8 @@ unsigned long native_calibrate_tsc(void) /* * Denverton SoCs don't report crystal clock, and also don't support - * CPUID.0x16 for the calculation below, so hardcode the 25MHz crystal - * clock. + * CPUID_FREQ_LEAF for the calculation below, so hardcode the 25MHz + * crystal clock. */ if (crystal_khz == 0 && boot_cpu_data.x86_vfm == INTEL_ATOM_GOLDMONT_D) @@ -701,10 +701,10 @@ unsigned long native_calibrate_tsc(void) * clock, but we can easily calculate it to a high degree of accuracy * by considering the crystal ratio and the CPU speed. */ - if (crystal_khz == 0 && boot_cpu_data.cpuid_level >= 0x16) { + if (crystal_khz == 0 && boot_cpu_data.cpuid_level >= CPUID_FREQ_LEAF) { unsigned int eax_base_mhz, ebx, ecx, edx; - cpuid(0x16, &eax_base_mhz, &ebx, &ecx, &edx); + cpuid(CPUID_FREQ_LEAF, &eax_base_mhz, &ebx, &ecx, &edx); crystal_khz = eax_base_mhz * 1000 * eax_denominator / ebx_numerator; } @@ -739,12 +739,12 @@ static unsigned long cpu_khz_from_cpuid(void) if (boot_cpu_data.x86_vendor != X86_VENDOR_INTEL) return 0; - if (boot_cpu_data.cpuid_level < 0x16) + if (boot_cpu_data.cpuid_level < CPUID_FREQ_LEAF) return 0; eax_base_mhz = ebx_max_mhz = ecx_bus_mhz = edx = 0; - cpuid(0x16, &eax_base_mhz, &ebx_max_mhz, &ecx_bus_mhz, &edx); + cpuid(CPUID_FREQ_LEAF, &eax_base_mhz, &ebx_max_mhz, &ecx_bus_mhz, &edx); return eax_base_mhz * 1000; } -- 2.51.0 From 754aaac3bbf13bdbbed9da94b56f371e90fd9c96 Mon Sep 17 00:00:00 2001 From: Dave Hansen Date: Fri, 13 Dec 2024 12:50:37 -0800 Subject: [PATCH 16/16] x86/fpu: Move CPUID leaf definitions to common code Move the XSAVE-related CPUID leaf definitions to common code. Then, use the new definition to remove the last magic number from the CPUID level dependency table. Signed-off-by: Dave Hansen Reviewed-by: Zhao Liu Link: https://lore.kernel.org/all/20241213205037.43C57CDE%40davehans-spike.ostc.intel.com --- arch/x86/include/asm/cpuid.h | 2 ++ arch/x86/include/asm/fpu/xstate.h | 4 ---- arch/x86/kernel/cpu/common.c | 2 +- arch/x86/kernel/fpu/xstate.c | 1 + 4 files changed, 4 insertions(+), 5 deletions(-) diff --git a/arch/x86/include/asm/cpuid.h b/arch/x86/include/asm/cpuid.h index e7803c21a3fd..a86097eb26c9 100644 --- a/arch/x86/include/asm/cpuid.h +++ b/arch/x86/include/asm/cpuid.h @@ -23,8 +23,10 @@ enum cpuid_regs_idx { #define CPUID_MWAIT_LEAF 0x5 #define CPUID_DCA_LEAF 0x9 +#define XSTATE_CPUID 0x0d #define CPUID_TSC_LEAF 0x15 #define CPUID_FREQ_LEAF 0x16 +#define TILE_CPUID 0x1d #ifdef CONFIG_X86_32 bool have_cpuid_p(void); diff --git a/arch/x86/include/asm/fpu/xstate.h b/arch/x86/include/asm/fpu/xstate.h index d4427b88ee12..7f39fe7980c5 100644 --- a/arch/x86/include/asm/fpu/xstate.h +++ b/arch/x86/include/asm/fpu/xstate.h @@ -12,10 +12,6 @@ /* Bit 63 of XCR0 is reserved for future expansion */ #define XFEATURE_MASK_EXTEND (~(XFEATURE_MASK_FPSSE | (1ULL << 63))) -#define XSTATE_CPUID 0x0000000d - -#define TILE_CPUID 0x0000001d - #define FXSAVE_SIZE 512 #define XSAVE_HDR_SIZE 64 diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 5ffa1f4eac38..f5c33e155f98 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -639,7 +639,7 @@ static const struct cpuid_dependent_feature cpuid_dependent_features[] = { { X86_FEATURE_MWAIT, CPUID_MWAIT_LEAF }, { X86_FEATURE_DCA, CPUID_DCA_LEAF }, - { X86_FEATURE_XSAVE, 0x0000000d }, + { X86_FEATURE_XSAVE, XSTATE_CPUID }, { 0, 0 } }; diff --git a/arch/x86/kernel/fpu/xstate.c b/arch/x86/kernel/fpu/xstate.c index 22abb5ee0cf2..bf38b3e75425 100644 --- a/arch/x86/kernel/fpu/xstate.c +++ b/arch/x86/kernel/fpu/xstate.c @@ -20,6 +20,7 @@ #include #include +#include #include #include #include -- 2.51.0