From 5ce344beaca688f4cdea07045e0b8f03dc537e74 Mon Sep 17 00:00:00 2001 From: Adam Dunlap Date: Mon, 18 Mar 2024 16:09:27 -0700 Subject: x86/apic: Force native_apic_mem_read() to use the MOV instruction When done from a virtual machine, instructions that touch APIC memory must be emulated. By convention, MMIO accesses are typically performed via io.h helpers such as readl() or writeq() to simplify instruction emulation/decoding (ex: in KVM hosts and SEV guests) [0]. Currently, native_apic_mem_read() does not follow this convention, allowing the compiler to emit instructions other than the MOV instruction generated by readl(). In particular, when the kernel is compiled with clang and run as a SEV-ES or SEV-SNP guest, the compiler would emit a TESTL instruction which is not supported by the SEV-ES emulator, causing a boot failure in that environment. It is likely the same problem would happen in a TDX guest as that uses the same instruction emulator as SEV-ES. To make sure all emulators can emulate APIC memory reads via MOV, use the readl() function in native_apic_mem_read(). It is expected that any emulator would support MOV in any addressing mode as it is the most generic and is what is usually emitted currently. The TESTL instruction is emitted when native_apic_mem_read() is inlined into apic_mem_wait_icr_idle(). The emulator comes from insn_decode_mmio() in arch/x86/lib/insn-eval.c. It's not worth it to extend insn_decode_mmio() to support more instructions since, in theory, the compiler could choose to output nearly any instruction for such reads which would bloat the emulator beyond reason. [0] https://lore.kernel.org/all/20220405232939.73860-12-kirill.shutemov@linux.intel.com/ [ bp: Massage commit message, fix typos. ] Signed-off-by: Adam Dunlap Signed-off-by: Borislav Petkov (AMD) Reviewed-by: Thomas Gleixner Reviewed-by: Ard Biesheuvel Tested-by: Kevin Loughlin Cc: Link: https://lore.kernel.org/r/20240318230927.2191933-1-acdunlap@google.com --- arch/x86/include/asm/apic.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/include/asm/apic.h b/arch/x86/include/asm/apic.h index 94ce0f7c9d3a..e6ab0cf15ed5 100644 --- a/arch/x86/include/asm/apic.h +++ b/arch/x86/include/asm/apic.h @@ -13,6 +13,7 @@ #include #include #include +#include #define ARCH_APICTIMER_STOPS_ON_C3 1 @@ -98,7 +99,7 @@ static inline void native_apic_mem_write(u32 reg, u32 v) static inline u32 native_apic_mem_read(u32 reg) { - return *((volatile u32 *)(APIC_BASE + reg)); + return readl((void __iomem *)(APIC_BASE + reg)); } static inline void native_apic_mem_eoi(void) -- cgit v1.2.3 From 04f4230e2f86a4e961ea5466eda3db8c1762004d Mon Sep 17 00:00:00 2001 From: Daniel Sneddon Date: Tue, 9 Apr 2024 16:08:05 -0700 Subject: x86/bugs: Fix return type of spectre_bhi_state() The definition of spectre_bhi_state() incorrectly returns a const char * const. This causes the a compiler warning when building with W=1: warning: type qualifiers ignored on function return type [-Wignored-qualifiers] 2812 | static const char * const spectre_bhi_state(void) Remove the const qualifier from the pointer. Fixes: ec9404e40e8f ("x86/bhi: Add BHI mitigation knob") Reported-by: Sean Christopherson Signed-off-by: Daniel Sneddon Signed-off-by: Ingo Molnar Cc: Linus Torvalds Link: https://lore.kernel.org/r/20240409230806.1545822-1-daniel.sneddon@linux.intel.com --- arch/x86/kernel/cpu/bugs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index 295463707e68..27f5004a8e24 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -2809,7 +2809,7 @@ static char *pbrsb_eibrs_state(void) } } -static const char * const spectre_bhi_state(void) +static const char *spectre_bhi_state(void) { if (!boot_cpu_has_bug(X86_BUG_BHI)) return "; BHI: Not affected"; -- cgit v1.2.3 From a9025cd1c673a8d6eefc79d911075b8b452eba8f Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Mon, 8 Apr 2024 15:22:01 +0200 Subject: x86/topology: Don't update cpu_possible_map in topo_set_cpuids() topo_set_cpuids() updates cpu_present_map and cpu_possible map. It is invoked during enumeration and "physical hotplug" operations. In the latter case this results in a kernel crash because cpu_possible_map is marked read only after init completes. There is no reason to update cpu_possible_map in that function. During enumeration cpu_possible_map is not relevant and gets fully initialized after enumeration completed. On "physical hotplug" the bit is already set because the kernel allows only CPUs to be plugged which have been enumerated and associated to a CPU number during early boot. Remove the bogus update of cpu_possible_map. Fixes: 0e53e7b656cf ("x86/cpu/topology: Sanitize the APIC admission logic") Reported-by: Jonathan Cameron Signed-off-by: Thomas Gleixner Link: https://lore.kernel.org/r/87ttkc6kwx.ffs@tglx --- arch/x86/kernel/cpu/topology.c | 7 +++++-- 1 file changed, 5 insertions(+), 2 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/topology.c b/arch/x86/kernel/cpu/topology.c index aaca8d235dc2..d17c9b71eb4a 100644 --- a/arch/x86/kernel/cpu/topology.c +++ b/arch/x86/kernel/cpu/topology.c @@ -123,7 +123,6 @@ static void topo_set_cpuids(unsigned int cpu, u32 apic_id, u32 acpi_id) early_per_cpu(x86_cpu_to_apicid, cpu) = apic_id; early_per_cpu(x86_cpu_to_acpiid, cpu) = acpi_id; #endif - set_cpu_possible(cpu, true); set_cpu_present(cpu, true); } @@ -210,7 +209,11 @@ static __init void topo_register_apic(u32 apic_id, u32 acpi_id, bool present) topo_info.nr_disabled_cpus++; } - /* Register present and possible CPUs in the domain maps */ + /* + * Register present and possible CPUs in the domain + * maps. cpu_possible_map will be updated in + * topology_init_possible_cpus() after enumeration is done. + */ for (dom = TOPO_SMT_DOMAIN; dom < TOPO_MAX_DOMAIN; dom++) set_bit(topo_apicid(apic_id, dom), apic_maps[dom].map); } -- cgit v1.2.3 From cb2db5bb04d7f778fbc1a1ea2507aab436f1bff3 Mon Sep 17 00:00:00 2001 From: Josh Poimboeuf Date: Wed, 10 Apr 2024 22:40:46 -0700 Subject: x86/bugs: Cache the value of MSR_IA32_ARCH_CAPABILITIES There's no need to keep reading MSR_IA32_ARCH_CAPABILITIES over and over. It's even read in the BHI sysfs function which is a big no-no. Just read it once and cache it. Fixes: ec9404e40e8f ("x86/bhi: Add BHI mitigation knob") Signed-off-by: Josh Poimboeuf Signed-off-by: Ingo Molnar Reviewed-by: Nikolay Borisov Cc: Linus Torvalds Cc: Sean Christopherson Link: https://lore.kernel.org/r/9592a18a814368e75f8f4b9d74d3883aa4fd1eaf.1712813475.git.jpoimboe@kernel.org --- arch/x86/kernel/cpu/bugs.c | 22 +++++++--------------- 1 file changed, 7 insertions(+), 15 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index 27f5004a8e24..ff59fa8bb610 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -61,6 +61,8 @@ EXPORT_PER_CPU_SYMBOL_GPL(x86_spec_ctrl_current); u64 x86_pred_cmd __ro_after_init = PRED_CMD_IBPB; EXPORT_SYMBOL_GPL(x86_pred_cmd); +static u64 __ro_after_init ia32_cap; + static DEFINE_MUTEX(spec_ctrl_mutex); void (*x86_return_thunk)(void) __ro_after_init = __x86_return_thunk; @@ -144,6 +146,8 @@ void __init cpu_select_mitigations(void) x86_spec_ctrl_base &= ~SPEC_CTRL_MITIGATIONS_MASK; } + ia32_cap = x86_read_arch_cap_msr(); + /* Select the proper CPU mitigations before patching alternatives: */ spectre_v1_select_mitigation(); spectre_v2_select_mitigation(); @@ -301,8 +305,6 @@ static const char * const taa_strings[] = { static void __init taa_select_mitigation(void) { - u64 ia32_cap; - if (!boot_cpu_has_bug(X86_BUG_TAA)) { taa_mitigation = TAA_MITIGATION_OFF; return; @@ -341,7 +343,6 @@ static void __init taa_select_mitigation(void) * On MDS_NO=1 CPUs if ARCH_CAP_TSX_CTRL_MSR is not set, microcode * update is required. */ - ia32_cap = x86_read_arch_cap_msr(); if ( (ia32_cap & ARCH_CAP_MDS_NO) && !(ia32_cap & ARCH_CAP_TSX_CTRL_MSR)) taa_mitigation = TAA_MITIGATION_UCODE_NEEDED; @@ -401,8 +402,6 @@ static const char * const mmio_strings[] = { static void __init mmio_select_mitigation(void) { - u64 ia32_cap; - if (!boot_cpu_has_bug(X86_BUG_MMIO_STALE_DATA) || boot_cpu_has_bug(X86_BUG_MMIO_UNKNOWN) || cpu_mitigations_off()) { @@ -413,8 +412,6 @@ static void __init mmio_select_mitigation(void) if (mmio_mitigation == MMIO_MITIGATION_OFF) return; - ia32_cap = x86_read_arch_cap_msr(); - /* * Enable CPU buffer clear mitigation for host and VMM, if also affected * by MDS or TAA. Otherwise, enable mitigation for VMM only. @@ -508,7 +505,7 @@ static void __init rfds_select_mitigation(void) if (rfds_mitigation == RFDS_MITIGATION_OFF) return; - if (x86_read_arch_cap_msr() & ARCH_CAP_RFDS_CLEAR) + if (ia32_cap & ARCH_CAP_RFDS_CLEAR) setup_force_cpu_cap(X86_FEATURE_CLEAR_CPU_BUF); else rfds_mitigation = RFDS_MITIGATION_UCODE_NEEDED; @@ -659,8 +656,6 @@ void update_srbds_msr(void) static void __init srbds_select_mitigation(void) { - u64 ia32_cap; - if (!boot_cpu_has_bug(X86_BUG_SRBDS)) return; @@ -669,7 +664,6 @@ static void __init srbds_select_mitigation(void) * are only exposed to SRBDS when TSX is enabled or when CPU is affected * by Processor MMIO Stale Data vulnerability. */ - ia32_cap = x86_read_arch_cap_msr(); if ((ia32_cap & ARCH_CAP_MDS_NO) && !boot_cpu_has(X86_FEATURE_RTM) && !boot_cpu_has_bug(X86_BUG_MMIO_STALE_DATA)) srbds_mitigation = SRBDS_MITIGATION_TSX_OFF; @@ -813,7 +807,7 @@ static void __init gds_select_mitigation(void) /* Will verify below that mitigation _can_ be disabled */ /* No microcode */ - if (!(x86_read_arch_cap_msr() & ARCH_CAP_GDS_CTRL)) { + if (!(ia32_cap & ARCH_CAP_GDS_CTRL)) { if (gds_mitigation == GDS_MITIGATION_FORCE) { /* * This only needs to be done on the boot CPU so do it @@ -1908,8 +1902,6 @@ static void update_indir_branch_cond(void) /* Update the static key controlling the MDS CPU buffer clear in idle */ static void update_mds_branch_idle(void) { - u64 ia32_cap = x86_read_arch_cap_msr(); - /* * Enable the idle clearing if SMT is active on CPUs which are * affected only by MSBDS and not any other MDS variant. @@ -2818,7 +2810,7 @@ static const char *spectre_bhi_state(void) else if (boot_cpu_has(X86_FEATURE_CLEAR_BHB_LOOP)) return "; BHI: SW loop, KVM: SW loop"; else if (boot_cpu_has(X86_FEATURE_RETPOLINE) && - !(x86_read_arch_cap_msr() & ARCH_CAP_RRSBA)) + !(ia32_cap & ARCH_CAP_RRSBA)) return "; BHI: Retpoline"; else if (boot_cpu_has(X86_FEATURE_CLEAR_BHB_LOOP_ON_VMEXIT)) return "; BHI: Syscall hardening, KVM: SW loop"; -- cgit v1.2.3 From d0485730d2189ffe5d986d4e9e191f1e4d5ffd24 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Thu, 11 Apr 2024 09:25:36 +0200 Subject: x86/bugs: Rename various 'ia32_cap' variables to 'x86_arch_cap_msr' So we are using the 'ia32_cap' value in a number of places, which got its name from MSR_IA32_ARCH_CAPABILITIES MSR register. But there's very little 'IA32' about it - this isn't 32-bit only code, nor does it originate from there, it's just a historic quirk that many Intel MSR names are prefixed with IA32_. This is already clear from the helper method around the MSR: x86_read_arch_cap_msr(), which doesn't have the IA32 prefix. So rename 'ia32_cap' to 'x86_arch_cap_msr' to be consistent with its role and with the naming of the helper function. Signed-off-by: Ingo Molnar Cc: Josh Poimboeuf Cc: Nikolay Borisov Cc: Linus Torvalds Cc: Sean Christopherson Link: https://lore.kernel.org/r/9592a18a814368e75f8f4b9d74d3883aa4fd1eaf.1712813475.git.jpoimboe@kernel.org --- arch/x86/kernel/apic/apic.c | 6 +++--- arch/x86/kernel/cpu/bugs.c | 30 +++++++++++++-------------- arch/x86/kernel/cpu/common.c | 48 ++++++++++++++++++++++---------------------- 3 files changed, 42 insertions(+), 42 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/apic/apic.c b/arch/x86/kernel/apic/apic.c index a42d8a6f7149..c342c4aa9c68 100644 --- a/arch/x86/kernel/apic/apic.c +++ b/arch/x86/kernel/apic/apic.c @@ -1687,11 +1687,11 @@ static int x2apic_state; static bool x2apic_hw_locked(void) { - u64 ia32_cap; + u64 x86_arch_cap_msr; u64 msr; - ia32_cap = x86_read_arch_cap_msr(); - if (ia32_cap & ARCH_CAP_XAPIC_DISABLE) { + x86_arch_cap_msr = x86_read_arch_cap_msr(); + if (x86_arch_cap_msr & ARCH_CAP_XAPIC_DISABLE) { rdmsrl(MSR_IA32_XAPIC_DISABLE_STATUS, msr); return (msr & LEGACY_XAPIC_DISABLED); } diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index ff59fa8bb610..1b0cfc136432 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -61,7 +61,7 @@ EXPORT_PER_CPU_SYMBOL_GPL(x86_spec_ctrl_current); u64 x86_pred_cmd __ro_after_init = PRED_CMD_IBPB; EXPORT_SYMBOL_GPL(x86_pred_cmd); -static u64 __ro_after_init ia32_cap; +static u64 __ro_after_init x86_arch_cap_msr; static DEFINE_MUTEX(spec_ctrl_mutex); @@ -146,7 +146,7 @@ void __init cpu_select_mitigations(void) x86_spec_ctrl_base &= ~SPEC_CTRL_MITIGATIONS_MASK; } - ia32_cap = x86_read_arch_cap_msr(); + x86_arch_cap_msr = x86_read_arch_cap_msr(); /* Select the proper CPU mitigations before patching alternatives: */ spectre_v1_select_mitigation(); @@ -343,8 +343,8 @@ static void __init taa_select_mitigation(void) * On MDS_NO=1 CPUs if ARCH_CAP_TSX_CTRL_MSR is not set, microcode * update is required. */ - if ( (ia32_cap & ARCH_CAP_MDS_NO) && - !(ia32_cap & ARCH_CAP_TSX_CTRL_MSR)) + if ( (x86_arch_cap_msr & ARCH_CAP_MDS_NO) && + !(x86_arch_cap_msr & ARCH_CAP_TSX_CTRL_MSR)) taa_mitigation = TAA_MITIGATION_UCODE_NEEDED; /* @@ -434,7 +434,7 @@ static void __init mmio_select_mitigation(void) * be propagated to uncore buffers, clearing the Fill buffers on idle * is required irrespective of SMT state. */ - if (!(ia32_cap & ARCH_CAP_FBSDP_NO)) + if (!(x86_arch_cap_msr & ARCH_CAP_FBSDP_NO)) static_branch_enable(&mds_idle_clear); /* @@ -444,10 +444,10 @@ static void __init mmio_select_mitigation(void) * FB_CLEAR or by the presence of both MD_CLEAR and L1D_FLUSH on MDS * affected systems. */ - if ((ia32_cap & ARCH_CAP_FB_CLEAR) || + if ((x86_arch_cap_msr & ARCH_CAP_FB_CLEAR) || (boot_cpu_has(X86_FEATURE_MD_CLEAR) && boot_cpu_has(X86_FEATURE_FLUSH_L1D) && - !(ia32_cap & ARCH_CAP_MDS_NO))) + !(x86_arch_cap_msr & ARCH_CAP_MDS_NO))) mmio_mitigation = MMIO_MITIGATION_VERW; else mmio_mitigation = MMIO_MITIGATION_UCODE_NEEDED; @@ -505,7 +505,7 @@ static void __init rfds_select_mitigation(void) if (rfds_mitigation == RFDS_MITIGATION_OFF) return; - if (ia32_cap & ARCH_CAP_RFDS_CLEAR) + if (x86_arch_cap_msr & ARCH_CAP_RFDS_CLEAR) setup_force_cpu_cap(X86_FEATURE_CLEAR_CPU_BUF); else rfds_mitigation = RFDS_MITIGATION_UCODE_NEEDED; @@ -664,7 +664,7 @@ static void __init srbds_select_mitigation(void) * are only exposed to SRBDS when TSX is enabled or when CPU is affected * by Processor MMIO Stale Data vulnerability. */ - if ((ia32_cap & ARCH_CAP_MDS_NO) && !boot_cpu_has(X86_FEATURE_RTM) && + if ((x86_arch_cap_msr & ARCH_CAP_MDS_NO) && !boot_cpu_has(X86_FEATURE_RTM) && !boot_cpu_has_bug(X86_BUG_MMIO_STALE_DATA)) srbds_mitigation = SRBDS_MITIGATION_TSX_OFF; else if (boot_cpu_has(X86_FEATURE_HYPERVISOR)) @@ -807,7 +807,7 @@ static void __init gds_select_mitigation(void) /* Will verify below that mitigation _can_ be disabled */ /* No microcode */ - if (!(ia32_cap & ARCH_CAP_GDS_CTRL)) { + if (!(x86_arch_cap_msr & ARCH_CAP_GDS_CTRL)) { if (gds_mitigation == GDS_MITIGATION_FORCE) { /* * This only needs to be done on the boot CPU so do it @@ -1541,14 +1541,14 @@ static enum spectre_v2_mitigation __init spectre_v2_select_retpoline(void) /* Disable in-kernel use of non-RSB RET predictors */ static void __init spec_ctrl_disable_kernel_rrsba(void) { - u64 ia32_cap; + u64 x86_arch_cap_msr; if (!boot_cpu_has(X86_FEATURE_RRSBA_CTRL)) return; - ia32_cap = x86_read_arch_cap_msr(); + x86_arch_cap_msr = x86_read_arch_cap_msr(); - if (ia32_cap & ARCH_CAP_RRSBA) { + if (x86_arch_cap_msr & ARCH_CAP_RRSBA) { x86_spec_ctrl_base |= SPEC_CTRL_RRSBA_DIS_S; update_spec_ctrl(x86_spec_ctrl_base); } @@ -1916,7 +1916,7 @@ static void update_mds_branch_idle(void) if (sched_smt_active()) { static_branch_enable(&mds_idle_clear); } else if (mmio_mitigation == MMIO_MITIGATION_OFF || - (ia32_cap & ARCH_CAP_FBSDP_NO)) { + (x86_arch_cap_msr & ARCH_CAP_FBSDP_NO)) { static_branch_disable(&mds_idle_clear); } } @@ -2810,7 +2810,7 @@ static const char *spectre_bhi_state(void) else if (boot_cpu_has(X86_FEATURE_CLEAR_BHB_LOOP)) return "; BHI: SW loop, KVM: SW loop"; else if (boot_cpu_has(X86_FEATURE_RETPOLINE) && - !(ia32_cap & ARCH_CAP_RRSBA)) + !(x86_arch_cap_msr & ARCH_CAP_RRSBA)) return "; BHI: Retpoline"; else if (boot_cpu_has(X86_FEATURE_CLEAR_BHB_LOOP_ON_VMEXIT)) return "; BHI: Syscall hardening, KVM: SW loop"; diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 754d91857d63..605c26c009c8 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -1284,25 +1284,25 @@ static bool __init cpu_matches(const struct x86_cpu_id *table, unsigned long whi u64 x86_read_arch_cap_msr(void) { - u64 ia32_cap = 0; + u64 x86_arch_cap_msr = 0; if (boot_cpu_has(X86_FEATURE_ARCH_CAPABILITIES)) - rdmsrl(MSR_IA32_ARCH_CAPABILITIES, ia32_cap); + rdmsrl(MSR_IA32_ARCH_CAPABILITIES, x86_arch_cap_msr); - return ia32_cap; + return x86_arch_cap_msr; } -static bool arch_cap_mmio_immune(u64 ia32_cap) +static bool arch_cap_mmio_immune(u64 x86_arch_cap_msr) { - return (ia32_cap & ARCH_CAP_FBSDP_NO && - ia32_cap & ARCH_CAP_PSDP_NO && - ia32_cap & ARCH_CAP_SBDR_SSDP_NO); + return (x86_arch_cap_msr & ARCH_CAP_FBSDP_NO && + x86_arch_cap_msr & ARCH_CAP_PSDP_NO && + x86_arch_cap_msr & ARCH_CAP_SBDR_SSDP_NO); } -static bool __init vulnerable_to_rfds(u64 ia32_cap) +static bool __init vulnerable_to_rfds(u64 x86_arch_cap_msr) { /* The "immunity" bit trumps everything else: */ - if (ia32_cap & ARCH_CAP_RFDS_NO) + if (x86_arch_cap_msr & ARCH_CAP_RFDS_NO) return false; /* @@ -1310,7 +1310,7 @@ static bool __init vulnerable_to_rfds(u64 ia32_cap) * indicate that mitigation is needed because guest is running on a * vulnerable hardware or may migrate to such hardware: */ - if (ia32_cap & ARCH_CAP_RFDS_CLEAR) + if (x86_arch_cap_msr & ARCH_CAP_RFDS_CLEAR) return true; /* Only consult the blacklist when there is no enumeration: */ @@ -1319,11 +1319,11 @@ static bool __init vulnerable_to_rfds(u64 ia32_cap) static void __init cpu_set_bug_bits(struct cpuinfo_x86 *c) { - u64 ia32_cap = x86_read_arch_cap_msr(); + u64 x86_arch_cap_msr = x86_read_arch_cap_msr(); /* Set ITLB_MULTIHIT bug if cpu is not in the whitelist and not mitigated */ if (!cpu_matches(cpu_vuln_whitelist, NO_ITLB_MULTIHIT) && - !(ia32_cap & ARCH_CAP_PSCHANGE_MC_NO)) + !(x86_arch_cap_msr & ARCH_CAP_PSCHANGE_MC_NO)) setup_force_cpu_bug(X86_BUG_ITLB_MULTIHIT); if (cpu_matches(cpu_vuln_whitelist, NO_SPECULATION)) @@ -1335,7 +1335,7 @@ static void __init cpu_set_bug_bits(struct cpuinfo_x86 *c) setup_force_cpu_bug(X86_BUG_SPECTRE_V2); if (!cpu_matches(cpu_vuln_whitelist, NO_SSB) && - !(ia32_cap & ARCH_CAP_SSB_NO) && + !(x86_arch_cap_msr & ARCH_CAP_SSB_NO) && !cpu_has(c, X86_FEATURE_AMD_SSB_NO)) setup_force_cpu_bug(X86_BUG_SPEC_STORE_BYPASS); @@ -1346,17 +1346,17 @@ static void __init cpu_set_bug_bits(struct cpuinfo_x86 *c) * Don't use AutoIBRS when SNP is enabled because it degrades host * userspace indirect branch performance. */ - if ((ia32_cap & ARCH_CAP_IBRS_ALL) || + if ((x86_arch_cap_msr & ARCH_CAP_IBRS_ALL) || (cpu_has(c, X86_FEATURE_AUTOIBRS) && !cpu_feature_enabled(X86_FEATURE_SEV_SNP))) { setup_force_cpu_cap(X86_FEATURE_IBRS_ENHANCED); if (!cpu_matches(cpu_vuln_whitelist, NO_EIBRS_PBRSB) && - !(ia32_cap & ARCH_CAP_PBRSB_NO)) + !(x86_arch_cap_msr & ARCH_CAP_PBRSB_NO)) setup_force_cpu_bug(X86_BUG_EIBRS_PBRSB); } if (!cpu_matches(cpu_vuln_whitelist, NO_MDS) && - !(ia32_cap & ARCH_CAP_MDS_NO)) { + !(x86_arch_cap_msr & ARCH_CAP_MDS_NO)) { setup_force_cpu_bug(X86_BUG_MDS); if (cpu_matches(cpu_vuln_whitelist, MSBDS_ONLY)) setup_force_cpu_bug(X86_BUG_MSBDS_ONLY); @@ -1375,9 +1375,9 @@ static void __init cpu_set_bug_bits(struct cpuinfo_x86 *c) * TSX_CTRL check alone is not sufficient for cases when the microcode * update is not present or running as guest that don't get TSX_CTRL. */ - if (!(ia32_cap & ARCH_CAP_TAA_NO) && + if (!(x86_arch_cap_msr & ARCH_CAP_TAA_NO) && (cpu_has(c, X86_FEATURE_RTM) || - (ia32_cap & ARCH_CAP_TSX_CTRL_MSR))) + (x86_arch_cap_msr & ARCH_CAP_TSX_CTRL_MSR))) setup_force_cpu_bug(X86_BUG_TAA); /* @@ -1403,7 +1403,7 @@ static void __init cpu_set_bug_bits(struct cpuinfo_x86 *c) * Set X86_BUG_MMIO_UNKNOWN for CPUs that are neither in the blacklist, * nor in the whitelist and also don't enumerate MSR ARCH_CAP MMIO bits. */ - if (!arch_cap_mmio_immune(ia32_cap)) { + if (!arch_cap_mmio_immune(x86_arch_cap_msr)) { if (cpu_matches(cpu_vuln_blacklist, MMIO)) setup_force_cpu_bug(X86_BUG_MMIO_STALE_DATA); else if (!cpu_matches(cpu_vuln_whitelist, NO_MMIO)) @@ -1411,7 +1411,7 @@ static void __init cpu_set_bug_bits(struct cpuinfo_x86 *c) } if (!cpu_has(c, X86_FEATURE_BTC_NO)) { - if (cpu_matches(cpu_vuln_blacklist, RETBLEED) || (ia32_cap & ARCH_CAP_RSBA)) + if (cpu_matches(cpu_vuln_blacklist, RETBLEED) || (x86_arch_cap_msr & ARCH_CAP_RSBA)) setup_force_cpu_bug(X86_BUG_RETBLEED); } @@ -1429,15 +1429,15 @@ static void __init cpu_set_bug_bits(struct cpuinfo_x86 *c) * disabling AVX2. The only way to do this in HW is to clear XCR0[2], * which means that AVX will be disabled. */ - if (cpu_matches(cpu_vuln_blacklist, GDS) && !(ia32_cap & ARCH_CAP_GDS_NO) && + if (cpu_matches(cpu_vuln_blacklist, GDS) && !(x86_arch_cap_msr & ARCH_CAP_GDS_NO) && boot_cpu_has(X86_FEATURE_AVX)) setup_force_cpu_bug(X86_BUG_GDS); - if (vulnerable_to_rfds(ia32_cap)) + if (vulnerable_to_rfds(x86_arch_cap_msr)) setup_force_cpu_bug(X86_BUG_RFDS); /* When virtualized, eIBRS could be hidden, assume vulnerable */ - if (!(ia32_cap & ARCH_CAP_BHI_NO) && + if (!(x86_arch_cap_msr & ARCH_CAP_BHI_NO) && !cpu_matches(cpu_vuln_whitelist, NO_BHI) && (boot_cpu_has(X86_FEATURE_IBRS_ENHANCED) || boot_cpu_has(X86_FEATURE_HYPERVISOR))) @@ -1447,7 +1447,7 @@ static void __init cpu_set_bug_bits(struct cpuinfo_x86 *c) return; /* Rogue Data Cache Load? No! */ - if (ia32_cap & ARCH_CAP_RDCL_NO) + if (x86_arch_cap_msr & ARCH_CAP_RDCL_NO) return; setup_force_cpu_bug(X86_BUG_CPU_MELTDOWN); -- cgit v1.2.3 From 1cea8a280dfd1016148a3820676f2f03e3f5b898 Mon Sep 17 00:00:00 2001 From: Josh Poimboeuf Date: Wed, 10 Apr 2024 22:40:47 -0700 Subject: x86/bugs: Fix BHI handling of RRSBA The ARCH_CAP_RRSBA check isn't correct: RRSBA may have already been disabled by the Spectre v2 mitigation (or can otherwise be disabled by the BHI mitigation itself if needed). In that case retpolines are fine. Fixes: ec9404e40e8f ("x86/bhi: Add BHI mitigation knob") Signed-off-by: Josh Poimboeuf Signed-off-by: Ingo Molnar Cc: Linus Torvalds Cc: Sean Christopherson Link: https://lore.kernel.org/r/6f56f13da34a0834b69163467449be7f58f253dc.1712813475.git.jpoimboe@kernel.org --- arch/x86/kernel/cpu/bugs.c | 30 ++++++++++++++++++------------ 1 file changed, 18 insertions(+), 12 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index 1b0cfc136432..08dfb94fcb3a 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -1538,20 +1538,25 @@ static enum spectre_v2_mitigation __init spectre_v2_select_retpoline(void) return SPECTRE_V2_RETPOLINE; } +static bool __ro_after_init rrsba_disabled; + /* Disable in-kernel use of non-RSB RET predictors */ static void __init spec_ctrl_disable_kernel_rrsba(void) { - u64 x86_arch_cap_msr; + if (rrsba_disabled) + return; - if (!boot_cpu_has(X86_FEATURE_RRSBA_CTRL)) + if (!(x86_arch_cap_msr & ARCH_CAP_RRSBA)) { + rrsba_disabled = true; return; + } - x86_arch_cap_msr = x86_read_arch_cap_msr(); + if (!boot_cpu_has(X86_FEATURE_RRSBA_CTRL)) + return; - if (x86_arch_cap_msr & ARCH_CAP_RRSBA) { - x86_spec_ctrl_base |= SPEC_CTRL_RRSBA_DIS_S; - update_spec_ctrl(x86_spec_ctrl_base); - } + x86_spec_ctrl_base |= SPEC_CTRL_RRSBA_DIS_S; + update_spec_ctrl(x86_spec_ctrl_base); + rrsba_disabled = true; } static void __init spectre_v2_determine_rsb_fill_type_at_vmexit(enum spectre_v2_mitigation mode) @@ -1652,9 +1657,11 @@ static void __init bhi_select_mitigation(void) return; /* Retpoline mitigates against BHI unless the CPU has RRSBA behavior */ - if (cpu_feature_enabled(X86_FEATURE_RETPOLINE) && - !(x86_read_arch_cap_msr() & ARCH_CAP_RRSBA)) - return; + if (cpu_feature_enabled(X86_FEATURE_RETPOLINE)) { + spec_ctrl_disable_kernel_rrsba(); + if (rrsba_disabled) + return; + } if (spec_ctrl_bhi_dis()) return; @@ -2809,8 +2816,7 @@ static const char *spectre_bhi_state(void) return "; BHI: BHI_DIS_S"; else if (boot_cpu_has(X86_FEATURE_CLEAR_BHB_LOOP)) return "; BHI: SW loop, KVM: SW loop"; - else if (boot_cpu_has(X86_FEATURE_RETPOLINE) && - !(x86_arch_cap_msr & ARCH_CAP_RRSBA)) + else if (boot_cpu_has(X86_FEATURE_RETPOLINE) && rrsba_disabled) return "; BHI: Retpoline"; else if (boot_cpu_has(X86_FEATURE_CLEAR_BHB_LOOP_ON_VMEXIT)) return "; BHI: Syscall hardening, KVM: SW loop"; -- cgit v1.2.3 From 5f882f3b0a8bf0788d5a0ee44b1191de5319bb8a Mon Sep 17 00:00:00 2001 From: Josh Poimboeuf Date: Wed, 10 Apr 2024 22:40:48 -0700 Subject: x86/bugs: Clarify that syscall hardening isn't a BHI mitigation While syscall hardening helps prevent some BHI attacks, there's still other low-hanging fruit remaining. Don't classify it as a mitigation and make it clear that the system may still be vulnerable if it doesn't have a HW or SW mitigation enabled. Fixes: ec9404e40e8f ("x86/bhi: Add BHI mitigation knob") Signed-off-by: Josh Poimboeuf Signed-off-by: Ingo Molnar Cc: Linus Torvalds Cc: Sean Christopherson Link: https://lore.kernel.org/r/b5951dae3fdee7f1520d5136a27be3bdfe95f88b.1712813475.git.jpoimboe@kernel.org --- Documentation/admin-guide/hw-vuln/spectre.rst | 11 +++++------ Documentation/admin-guide/kernel-parameters.txt | 3 +-- arch/x86/kernel/cpu/bugs.c | 6 +++--- 3 files changed, 9 insertions(+), 11 deletions(-) (limited to 'arch/x86') diff --git a/Documentation/admin-guide/hw-vuln/spectre.rst b/Documentation/admin-guide/hw-vuln/spectre.rst index 3cf18e4a1d9a..5a39acf82483 100644 --- a/Documentation/admin-guide/hw-vuln/spectre.rst +++ b/Documentation/admin-guide/hw-vuln/spectre.rst @@ -441,10 +441,10 @@ The possible values in this file are: - System is protected by BHI_DIS_S * - BHI: SW loop, KVM SW loop - System is protected by software clearing sequence - * - BHI: Syscall hardening - - Syscalls are hardened against BHI - * - BHI: Syscall hardening, KVM: SW loop - - System is protected from userspace attacks by syscall hardening; KVM is protected by software clearing sequence + * - BHI: Vulnerable + - System is vulnerable to BHI + * - BHI: Vulnerable, KVM: SW loop + - System is vulnerable; KVM is protected by software clearing sequence Full mitigation might require a microcode update from the CPU vendor. When the necessary microcode is not available, the kernel will @@ -661,8 +661,7 @@ kernel command line. spectre_bhi= [X86] Control mitigation of Branch History Injection - (BHI) vulnerability. Syscalls are hardened against BHI - regardless of this setting. This setting affects the deployment + (BHI) vulnerability. This setting affects the deployment of the HW BHI control and the SW BHB clearing sequence. on diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt index a029ad6c4963..a3874cc97892 100644 --- a/Documentation/admin-guide/kernel-parameters.txt +++ b/Documentation/admin-guide/kernel-parameters.txt @@ -6065,8 +6065,7 @@ See Documentation/admin-guide/laptops/sonypi.rst spectre_bhi= [X86] Control mitigation of Branch History Injection - (BHI) vulnerability. Syscalls are hardened against BHI - reglardless of this setting. This setting affects the + (BHI) vulnerability. This setting affects the deployment of the HW BHI control and the SW BHB clearing sequence. diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index 08dfb94fcb3a..9eeb60f5fbb3 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -2818,10 +2818,10 @@ static const char *spectre_bhi_state(void) return "; BHI: SW loop, KVM: SW loop"; else if (boot_cpu_has(X86_FEATURE_RETPOLINE) && rrsba_disabled) return "; BHI: Retpoline"; - else if (boot_cpu_has(X86_FEATURE_CLEAR_BHB_LOOP_ON_VMEXIT)) - return "; BHI: Syscall hardening, KVM: SW loop"; + else if (boot_cpu_has(X86_FEATURE_CLEAR_BHB_LOOP_ON_VMEXIT)) + return "; BHI: Vulnerable, KVM: SW loop"; - return "; BHI: Vulnerable (Syscall hardening enabled)"; + return "; BHI: Vulnerable"; } static ssize_t spectre_v2_show_state(char *buf) -- cgit v1.2.3 From 36d4fe147c870f6d3f6602befd7ef44393a1c87a Mon Sep 17 00:00:00 2001 From: Josh Poimboeuf Date: Wed, 10 Apr 2024 22:40:50 -0700 Subject: x86/bugs: Remove CONFIG_BHI_MITIGATION_AUTO and spectre_bhi=auto Unlike most other mitigations' "auto" options, spectre_bhi=auto only mitigates newer systems, which is confusing and not particularly useful. Remove it. Signed-off-by: Josh Poimboeuf Signed-off-by: Ingo Molnar Reviewed-by: Nikolay Borisov Cc: Sean Christopherson Cc: Linus Torvalds Link: https://lore.kernel.org/r/412e9dc87971b622bbbaf64740ebc1f140bff343.1712813475.git.jpoimboe@kernel.org --- Documentation/admin-guide/hw-vuln/spectre.rst | 4 ---- Documentation/admin-guide/kernel-parameters.txt | 3 --- arch/x86/Kconfig | 5 ----- arch/x86/kernel/cpu/bugs.c | 10 +--------- 4 files changed, 1 insertion(+), 21 deletions(-) (limited to 'arch/x86') diff --git a/Documentation/admin-guide/hw-vuln/spectre.rst b/Documentation/admin-guide/hw-vuln/spectre.rst index 5a39acf82483..25a04cda4c2c 100644 --- a/Documentation/admin-guide/hw-vuln/spectre.rst +++ b/Documentation/admin-guide/hw-vuln/spectre.rst @@ -669,10 +669,6 @@ kernel command line. needed. off Disable the mitigation. - auto - Enable the HW mitigation if needed, but - *don't* enable the SW mitigation except for KVM. - The system may be vulnerable. For spectre_v2_user see Documentation/admin-guide/kernel-parameters.txt diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt index a3874cc97892..902ecd92a29f 100644 --- a/Documentation/admin-guide/kernel-parameters.txt +++ b/Documentation/admin-guide/kernel-parameters.txt @@ -6072,9 +6072,6 @@ on - (default) Enable the HW or SW mitigation as needed. off - Disable the mitigation. - auto - Enable the HW mitigation if needed, but - *don't* enable the SW mitigation except - for KVM. The system may be vulnerable. spectre_v2= [X86,EARLY] Control mitigation of Spectre variant 2 (indirect branch speculation) vulnerability. diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 10a6251f58f3..b63b6767a63d 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -2651,11 +2651,6 @@ config SPECTRE_BHI_OFF bool "off" help Equivalent to setting spectre_bhi=off command line parameter. -config SPECTRE_BHI_AUTO - bool "auto" - depends on BROKEN - help - Equivalent to setting spectre_bhi=auto command line parameter. endchoice diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index 9eeb60f5fbb3..6af4780a18ed 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -1625,13 +1625,10 @@ static bool __init spec_ctrl_bhi_dis(void) enum bhi_mitigations { BHI_MITIGATION_OFF, BHI_MITIGATION_ON, - BHI_MITIGATION_AUTO, }; static enum bhi_mitigations bhi_mitigation __ro_after_init = - IS_ENABLED(CONFIG_SPECTRE_BHI_ON) ? BHI_MITIGATION_ON : - IS_ENABLED(CONFIG_SPECTRE_BHI_OFF) ? BHI_MITIGATION_OFF : - BHI_MITIGATION_AUTO; + IS_ENABLED(CONFIG_SPECTRE_BHI_ON) ? BHI_MITIGATION_ON : BHI_MITIGATION_OFF; static int __init spectre_bhi_parse_cmdline(char *str) { @@ -1642,8 +1639,6 @@ static int __init spectre_bhi_parse_cmdline(char *str) bhi_mitigation = BHI_MITIGATION_OFF; else if (!strcmp(str, "on")) bhi_mitigation = BHI_MITIGATION_ON; - else if (!strcmp(str, "auto")) - bhi_mitigation = BHI_MITIGATION_AUTO; else pr_err("Ignoring unknown spectre_bhi option (%s)", str); @@ -1673,9 +1668,6 @@ static void __init bhi_select_mitigation(void) setup_force_cpu_cap(X86_FEATURE_CLEAR_BHB_LOOP_ON_VMEXIT); pr_info("Spectre BHI mitigation: SW BHB clearing on vm exit\n"); - if (bhi_mitigation == BHI_MITIGATION_AUTO) - return; - /* Mitigate syscalls when the mitigation is forced =on */ setup_force_cpu_cap(X86_FEATURE_CLEAR_BHB_LOOP); pr_info("Spectre BHI mitigation: SW BHB clearing on syscall\n"); -- cgit v1.2.3 From 4f511739c54b549061993b53fc0380f48dfca23b Mon Sep 17 00:00:00 2001 From: Josh Poimboeuf Date: Wed, 10 Apr 2024 22:40:51 -0700 Subject: x86/bugs: Replace CONFIG_SPECTRE_BHI_{ON,OFF} with CONFIG_MITIGATION_SPECTRE_BHI For consistency with the other CONFIG_MITIGATION_* options, replace the CONFIG_SPECTRE_BHI_{ON,OFF} options with a single CONFIG_MITIGATION_SPECTRE_BHI option. [ mingo: Fix ] Signed-off-by: Josh Poimboeuf Signed-off-by: Ingo Molnar Cc: Sean Christopherson Cc: Linus Torvalds Cc: Nikolay Borisov Link: https://lore.kernel.org/r/3833812ea63e7fdbe36bf8b932e63f70d18e2a2a.1712813475.git.jpoimboe@kernel.org --- arch/x86/Kconfig | 17 +++-------------- arch/x86/kernel/cpu/bugs.c | 2 +- 2 files changed, 4 insertions(+), 15 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index b63b6767a63d..4474bf32d0a4 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -2633,27 +2633,16 @@ config MITIGATION_RFDS stored in floating point, vector and integer registers. See also -choice - prompt "Clear branch history" +config MITIGATION_SPECTRE_BHI + bool "Mitigate Spectre-BHB (Branch History Injection)" depends on CPU_SUP_INTEL - default SPECTRE_BHI_ON + default y help Enable BHI mitigations. BHI attacks are a form of Spectre V2 attacks where the branch history buffer is poisoned to speculatively steer indirect branches. See -config SPECTRE_BHI_ON - bool "on" - help - Equivalent to setting spectre_bhi=on command line parameter. -config SPECTRE_BHI_OFF - bool "off" - help - Equivalent to setting spectre_bhi=off command line parameter. - -endchoice - endif config ARCH_HAS_ADD_PAGES diff --git a/arch/x86/kernel/cpu/bugs.c b/arch/x86/kernel/cpu/bugs.c index 6af4780a18ed..ca295b0c1eee 100644 --- a/arch/x86/kernel/cpu/bugs.c +++ b/arch/x86/kernel/cpu/bugs.c @@ -1628,7 +1628,7 @@ enum bhi_mitigations { }; static enum bhi_mitigations bhi_mitigation __ro_after_init = - IS_ENABLED(CONFIG_SPECTRE_BHI_ON) ? BHI_MITIGATION_ON : BHI_MITIGATION_OFF; + IS_ENABLED(CONFIG_MITIGATION_SPECTRE_BHI) ? BHI_MITIGATION_ON : BHI_MITIGATION_OFF; static int __init spectre_bhi_parse_cmdline(char *str) { -- cgit v1.2.3 From 1b3108f6898ef2e03973d65255182792e94e2240 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 10 Apr 2024 21:45:27 +0200 Subject: x86/cpu/amd: Make the CPUID 0x80000008 parser correct CPUID 0x80000008 ECX.cpu_nthreads describes the number of threads in the package. The parser uses this value to initialize the SMT domain level. That's wrong because cpu_nthreads does not describe the number of threads per physical core. So this needs to set the CORE domain level and let the later parsers set the SMT shift if available. Preset the SMT domain level with the assumption of one thread per core, which is correct ifrt here are no other CPUID leafs to parse, and propagate cpu_nthreads and the core level APIC bitwidth into the CORE domain. Fixes: f7fb3b2dd92c ("x86/cpu: Provide an AMD/HYGON specific topology parser") Reported-by: "kernelci.org bot" Reported-by: Laura Nao Signed-off-by: Thomas Gleixner Tested-by: Laura Nao Link: https://lore.kernel.org/r/20240410194311.535206450@linutronix.de --- arch/x86/kernel/cpu/topology_amd.c | 24 ++++++++++++++++++------ 1 file changed, 18 insertions(+), 6 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/topology_amd.c b/arch/x86/kernel/cpu/topology_amd.c index 1a8b3ad493af..79a85a4814ca 100644 --- a/arch/x86/kernel/cpu/topology_amd.c +++ b/arch/x86/kernel/cpu/topology_amd.c @@ -29,11 +29,21 @@ static bool parse_8000_0008(struct topo_scan *tscan) if (!sft) sft = get_count_order(ecx.cpu_nthreads + 1); - topology_set_dom(tscan, TOPO_SMT_DOMAIN, sft, ecx.cpu_nthreads + 1); + /* + * cpu_nthreads describes the number of threads in the package + * sft is the number of APIC ID bits per package + * + * As the number of actual threads per core is not described in + * this leaf, just set the CORE domain shift and let the later + * parsers set SMT shift. Assume one thread per core by default + * which is correct if there are no other CPUID leafs to parse. + */ + topology_update_dom(tscan, TOPO_SMT_DOMAIN, 0, 1); + topology_set_dom(tscan, TOPO_CORE_DOMAIN, sft, ecx.cpu_nthreads + 1); return true; } -static void store_node(struct topo_scan *tscan, unsigned int nr_nodes, u16 node_id) +static void store_node(struct topo_scan *tscan, u16 nr_nodes, u16 node_id) { /* * Starting with Fam 17h the DIE domain could probably be used to @@ -73,12 +83,14 @@ static bool parse_8000_001e(struct topo_scan *tscan, bool has_0xb) tscan->c->topo.initial_apicid = leaf.ext_apic_id; /* - * If leaf 0xb is available, then SMT shift is set already. If not - * take it from ecx.threads_per_core and use topo_update_dom() - - * topology_set_dom() would propagate and overwrite the already - * propagated CORE level. + * If leaf 0xb is available, then the domain shifts are set + * already and nothing to do here. */ if (!has_0xb) { + /* + * Leaf 0x80000008 set the CORE domain shift already. + * Update the SMT domain, but do not propagate it. + */ unsigned int nthreads = leaf.core_nthreads + 1; topology_update_dom(tscan, TOPO_SMT_DOMAIN, get_count_order(nthreads), nthreads); -- cgit v1.2.3 From c064b536a8f9ab7c8e204da8f5a22f7420d0b56c Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 10 Apr 2024 21:45:28 +0200 Subject: x86/cpu/amd: Make the NODEID_MSR union actually work A system with NODEID_MSR was reported to crash during early boot without any output. The reason is that the union which is used for accessing the bitfields in the MSR is written wrongly and the resulting executable code accesses the wrong part of the MSR data. As a consequence a later division by that value results in 0 and that result is used for another division as divisor, which obviously does not work well. The magic world of C, unions and bitfields: union { u64 bita : 3, bitb : 3; u64 all; } x; x.all = foo(); a = x.bita; b = x.bitb; results in the effective executable code of: a = b = x.bita; because bita and bitb are treated as union members and therefore both end up at bit offset 0. Wrapping the bitfield into an anonymous struct: union { struct { u64 bita : 3, bitb : 3; }; u64 all; } x; works like expected. Rework the NODEID_MSR union in exactly that way to cure the problem. Fixes: f7fb3b2dd92c ("x86/cpu: Provide an AMD/HYGON specific topology parser") Reported-by: "kernelci.org bot" Reported-by: Laura Nao Signed-off-by: Thomas Gleixner Tested-by: Laura Nao Link: https://lore.kernel.org/r/20240410194311.596282919@linutronix.de Closes: https://lore.kernel.org/all/20240322175210.124416-1-laura.nao@collabora.com/ --- arch/x86/kernel/cpu/topology_amd.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/topology_amd.c b/arch/x86/kernel/cpu/topology_amd.c index 79a85a4814ca..7f999aef1965 100644 --- a/arch/x86/kernel/cpu/topology_amd.c +++ b/arch/x86/kernel/cpu/topology_amd.c @@ -121,13 +121,13 @@ static bool parse_8000_001e(struct topo_scan *tscan, bool has_0xb) static bool parse_fam10h_node_id(struct topo_scan *tscan) { - struct { - union { + union { + struct { u64 node_id : 3, nodes_per_pkg : 3, unused : 58; - u64 msr; }; + u64 msr; } nid; if (!boot_cpu_has(X86_FEATURE_NODEID_MSR)) -- cgit v1.2.3 From 7211274fe0ee352332255e41ab5e628b86e83994 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Thu, 11 Apr 2024 18:55:38 +0200 Subject: x86/cpu/amd: Move TOPOEXT enablement into the topology parser The topology rework missed that early_init_amd() tries to re-enable the Topology Extensions when the BIOS disabled them. The new parser is invoked before early_init_amd() so the re-enable attempt happens too late. Move it into the AMD specific topology parser code where it belongs. Fixes: f7fb3b2dd92c ("x86/cpu: Provide an AMD/HYGON specific topology parser") Signed-off-by: Thomas Gleixner Signed-off-by: Ingo Molnar Link: https://lore.kernel.org/r/878r1j260l.ffs@tglx --- arch/x86/kernel/cpu/amd.c | 15 --------------- arch/x86/kernel/cpu/topology_amd.c | 21 +++++++++++++++++++++ 2 files changed, 21 insertions(+), 15 deletions(-) (limited to 'arch/x86') diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index 9bf17c9c29da..cb9eece55904 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -535,7 +535,6 @@ clear_sev: static void early_init_amd(struct cpuinfo_x86 *c) { - u64 value; u32 dummy; if (c->x86 >= 0xf) @@ -603,20 +602,6 @@ static void early_init_amd(struct cpuinfo_x86 *c) early_detect_mem_encrypt(c); - /* Re-enable TopologyExtensions if switched off by BIOS */ - if (c->x86 == 0x15 && - (c->x86_model >= 0x10 && c->x86_model <= 0x6f) && - !cpu_has(c, X86_FEATURE_TOPOEXT)) { - - if (msr_set_bit(0xc0011005, 54) > 0) { - rdmsrl(0xc0011005, value); - if (value & BIT_64(54)) { - set_cpu_cap(c, X86_FEATURE_TOPOEXT); - pr_info_once(FW_INFO "CPU: Re-enabling disabled Topology Extensions Support.\n"); - } - } - } - if (!cpu_has(c, X86_FEATURE_HYPERVISOR) && !cpu_has(c, X86_FEATURE_IBPB_BRTYPE)) { if (c->x86 == 0x17 && boot_cpu_has(X86_FEATURE_AMD_IBPB)) setup_force_cpu_cap(X86_FEATURE_IBPB_BRTYPE); diff --git a/arch/x86/kernel/cpu/topology_amd.c b/arch/x86/kernel/cpu/topology_amd.c index 7f999aef1965..a7aa6eff4ae5 100644 --- a/arch/x86/kernel/cpu/topology_amd.c +++ b/arch/x86/kernel/cpu/topology_amd.c @@ -147,6 +147,26 @@ static void legacy_set_llc(struct topo_scan *tscan) tscan->c->topo.llc_id = apicid >> tscan->dom_shifts[TOPO_CORE_DOMAIN]; } +static void topoext_fixup(struct topo_scan *tscan) +{ + struct cpuinfo_x86 *c = tscan->c; + u64 msrval; + + /* Try to re-enable TopologyExtensions if switched off by BIOS */ + if (cpu_has(c, X86_FEATURE_TOPOEXT) || c->x86_vendor != X86_VENDOR_AMD || + c->x86 != 0x15 || c->x86_model < 0x10 || c->x86_model > 0x6f) + return; + + if (msr_set_bit(0xc0011005, 54) <= 0) + return; + + rdmsrl(0xc0011005, msrval); + if (msrval & BIT_64(54)) { + set_cpu_cap(c, X86_FEATURE_TOPOEXT); + pr_info_once(FW_INFO "CPU: Re-enabling disabled Topology Extensions Support.\n"); + } +} + static void parse_topology_amd(struct topo_scan *tscan) { bool has_0xb = false; @@ -176,6 +196,7 @@ static void parse_topology_amd(struct topo_scan *tscan) void cpu_parse_topology_amd(struct topo_scan *tscan) { tscan->amd_nodes_per_pkg = 1; + topoext_fixup(tscan); parse_topology_amd(tscan); if (tscan->amd_nodes_per_pkg > 1) -- cgit v1.2.3