From c64b04fe6e0cb7c78e01751a44ef56cf20344e87 Mon Sep 17 00:00:00 2001 From: Jaswinder Singh Rajput Date: Sun, 14 Jun 2009 00:59:50 +0530 Subject: x86, cpu: cpu/proc.c display cache alignment and address sizes for 32 bit 32 bits can also access x86_cache_alignment, x86_phys_bits and x86_virt_bits, make them available to user space just as on 64 bits. Signed-off-by: Jaswinder Singh Rajput LKML-Reference: <1244921390.11733.30.camel@ht.satnam> Signed-off-by: H. Peter Anvin --- arch/x86/kernel/cpu/proc.c | 2 -- 1 file changed, 2 deletions(-) (limited to 'arch/x86/kernel/cpu') diff --git a/arch/x86/kernel/cpu/proc.c b/arch/x86/kernel/cpu/proc.c index d5e30397246..f82706a3901 100644 --- a/arch/x86/kernel/cpu/proc.c +++ b/arch/x86/kernel/cpu/proc.c @@ -116,11 +116,9 @@ static int show_cpuinfo(struct seq_file *m, void *v) seq_printf(m, "TLB size\t: %d 4K pages\n", c->x86_tlbsize); #endif seq_printf(m, "clflush size\t: %u\n", c->x86_clflush_size); -#ifdef CONFIG_X86_64 seq_printf(m, "cache_alignment\t: %d\n", c->x86_cache_alignment); seq_printf(m, "address sizes\t: %u bits physical, %u bits virtual\n", c->x86_phys_bits, c->x86_virt_bits); -#endif seq_printf(m, "power management:"); for (i = 0; i < 32; i++) { -- cgit v1.2.3-70-g09d2 From 6b0f43ddfa358dc71ad2a2d57bce5906c1c5dc1a Mon Sep 17 00:00:00 2001 From: Borislav Petkov Date: Mon, 31 Aug 2009 09:50:11 +0200 Subject: x86, AMD: Disable wrongly set X86_FEATURE_LAHF_LM CPUID bit fbd8b1819e80ac5a176d085fdddc3a34d1499318 turns off the bit for /proc/cpuinfo. However, a proper/full fix would be to additionally turn off the bit in the CPUID output so that future callers get correct CPU features info. Do that by basically reversing what the BIOS wrongfully does at boot. Signed-off-by: Borislav Petkov LKML-Reference: <1251705011-18636-3-git-send-email-petkovbb@gmail.com> Signed-off-by: H. Peter Anvin --- arch/x86/kernel/cpu/amd.c | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) (limited to 'arch/x86/kernel/cpu') diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index 63fddcd082c..0a717fc6aeb 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -404,9 +404,18 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c) /* * Some BIOSes incorrectly force this feature, but only K8 * revision D (model = 0x14) and later actually support it. + * (AMD Erratum #110, docId: 25759). */ - if (c->x86_model < 0x14) + if (c->x86_model < 0x14 && cpu_has(c, X86_FEATURE_LAHF_LM)) { + u64 val; + clear_cpu_cap(c, X86_FEATURE_LAHF_LM); + if (!rdmsrl_amd_safe(0xc001100d, &val)) { + val &= ~(1ULL << 32); + wrmsrl_amd_safe(0xc001100d, val); + } + } + } if (c->x86 == 0x10 || c->x86 == 0x11) set_cpu_cap(c, X86_FEATURE_REP_GOOD); -- cgit v1.2.3-70-g09d2 From 4a376ec3a2599c02207cd4cbd5dbf73783548463 Mon Sep 17 00:00:00 2001 From: Andreas Herrmann Date: Thu, 3 Sep 2009 09:40:21 +0200 Subject: x86: Fix CPU llc_shared_map information for AMD Magny-Cours Construct entire NodeID and use it as cpu_llc_id. Thus internal node siblings are stored in llc_shared_map. Signed-off-by: Andreas Herrmann Signed-off-by: H. Peter Anvin --- arch/x86/include/asm/cpufeature.h | 1 + arch/x86/kernel/cpu/amd.c | 64 ++++++++++++++++++++++++++++++++++++++- 2 files changed, 64 insertions(+), 1 deletion(-) (limited to 'arch/x86/kernel/cpu') diff --git a/arch/x86/include/asm/cpufeature.h b/arch/x86/include/asm/cpufeature.h index 4a28d22d479..847fee6493a 100644 --- a/arch/x86/include/asm/cpufeature.h +++ b/arch/x86/include/asm/cpufeature.h @@ -95,6 +95,7 @@ #define X86_FEATURE_NONSTOP_TSC (3*32+24) /* TSC does not stop in C states */ #define X86_FEATURE_CLFLUSH_MONITOR (3*32+25) /* "" clflush reqd with monitor */ #define X86_FEATURE_EXTD_APICID (3*32+26) /* has extended APICID (8 bits) */ +#define X86_FEATURE_AMD_DCM (3*32+27) /* multi-node processor */ /* Intel-defined CPU features, CPUID level 0x00000001 (ecx), word 4 */ #define X86_FEATURE_XMM3 (4*32+ 0) /* "pni" SSE-3 */ diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index 0a717fc6aeb..a76d2c18f15 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -250,6 +250,64 @@ static int __cpuinit nearby_node(int apicid) } #endif +/* + * Fixup core topology information for AMD multi-node processors. + * Assumption 1: Number of cores in each internal node is the same. + * Assumption 2: Mixed systems with both single-node and dual-node + * processors are not supported. + */ +#ifdef CONFIG_X86_HT +static void __cpuinit amd_fixup_dcm(struct cpuinfo_x86 *c) +{ +#ifdef CONFIG_PCI + u32 t, cpn; + u8 n, n_id; + int cpu = smp_processor_id(); + + /* fixup topology information only once for a core */ + if (cpu_has(c, X86_FEATURE_AMD_DCM)) + return; + + /* check for multi-node processor on boot cpu */ + t = read_pci_config(0, 24, 3, 0xe8); + if (!(t & (1 << 29))) + return; + + set_cpu_cap(c, X86_FEATURE_AMD_DCM); + + /* cores per node: each internal node has half the number of cores */ + cpn = c->x86_max_cores >> 1; + + /* even-numbered NB_id of this dual-node processor */ + n = c->phys_proc_id << 1; + + /* + * determine internal node id and assign cores fifty-fifty to + * each node of the dual-node processor + */ + t = read_pci_config(0, 24 + n, 3, 0xe8); + n = (t>>30) & 0x3; + if (n == 0) { + if (c->cpu_core_id < cpn) + n_id = 0; + else + n_id = 1; + } else { + if (c->cpu_core_id < cpn) + n_id = 1; + else + n_id = 0; + } + + /* compute entire NodeID, use llc_shared_map to store sibling info */ + per_cpu(cpu_llc_id, cpu) = (c->phys_proc_id << 1) + n_id; + + /* fixup core id to be in range from 0 to cpn */ + c->cpu_core_id = c->cpu_core_id % cpn; +#endif +} +#endif + /* * On a AMD dual core setup the lower bits of the APIC id distingush the cores. * Assumes number of cores is a power of two. @@ -267,6 +325,9 @@ static void __cpuinit amd_detect_cmp(struct cpuinfo_x86 *c) c->phys_proc_id = c->initial_apicid >> bits; /* use socket ID also for last level cache */ per_cpu(cpu_llc_id, cpu) = c->phys_proc_id; + /* fixup topology information on multi-node processors */ + if ((c->x86 == 0x10) && (c->x86_model == 9)) + amd_fixup_dcm(c); #endif } @@ -277,7 +338,8 @@ static void __cpuinit srat_detect_node(struct cpuinfo_x86 *c) int node; unsigned apicid = cpu_has_apic ? hard_smp_processor_id() : c->apicid; - node = c->phys_proc_id; + node = per_cpu(cpu_llc_id, cpu); + if (apicid_to_node[apicid] != NUMA_NO_NODE) node = apicid_to_node[apicid]; if (!node_online(node)) { -- cgit v1.2.3-70-g09d2 From a326e948c538e8ce998f30d92e146ecea8a30421 Mon Sep 17 00:00:00 2001 From: Andreas Herrmann Date: Thu, 3 Sep 2009 09:41:19 +0200 Subject: x86, cacheinfo: Fixup L3 cache information for AMD multi-node processors L3 cache size, associativity and shared_cpu information need to be adapted to show information for an internal node instead of the entire physical package. Signed-off-by: Andreas Herrmann Signed-off-by: H. Peter Anvin --- arch/x86/kernel/cpu/intel_cacheinfo.c | 32 ++++++++++++++++++++++---------- 1 file changed, 22 insertions(+), 10 deletions(-) (limited to 'arch/x86/kernel/cpu') diff --git a/arch/x86/kernel/cpu/intel_cacheinfo.c b/arch/x86/kernel/cpu/intel_cacheinfo.c index 789efe217e1..cb98a73e516 100644 --- a/arch/x86/kernel/cpu/intel_cacheinfo.c +++ b/arch/x86/kernel/cpu/intel_cacheinfo.c @@ -241,7 +241,7 @@ amd_cpuid4(int leaf, union _cpuid4_leaf_eax *eax, case 0: if (!l1->val) return; - assoc = l1->assoc; + assoc = assocs[l1->assoc]; line_size = l1->line_size; lines_per_tag = l1->lines_per_tag; size_in_kb = l1->size_in_kb; @@ -249,7 +249,7 @@ amd_cpuid4(int leaf, union _cpuid4_leaf_eax *eax, case 2: if (!l2.val) return; - assoc = l2.assoc; + assoc = assocs[l2.assoc]; line_size = l2.line_size; lines_per_tag = l2.lines_per_tag; /* cpu_data has errata corrections for K7 applied */ @@ -258,10 +258,14 @@ amd_cpuid4(int leaf, union _cpuid4_leaf_eax *eax, case 3: if (!l3.val) return; - assoc = l3.assoc; + assoc = assocs[l3.assoc]; line_size = l3.line_size; lines_per_tag = l3.lines_per_tag; size_in_kb = l3.size_encoded * 512; + if (boot_cpu_has(X86_FEATURE_AMD_DCM)) { + size_in_kb = size_in_kb >> 1; + assoc = assoc >> 1; + } break; default: return; @@ -270,18 +274,14 @@ amd_cpuid4(int leaf, union _cpuid4_leaf_eax *eax, eax->split.is_self_initializing = 1; eax->split.type = types[leaf]; eax->split.level = levels[leaf]; - if (leaf == 3) - eax->split.num_threads_sharing = - current_cpu_data.x86_max_cores - 1; - else - eax->split.num_threads_sharing = 0; + eax->split.num_threads_sharing = 0; eax->split.num_cores_on_die = current_cpu_data.x86_max_cores - 1; - if (assoc == 0xf) + if (assoc == 0xffff) eax->split.is_fully_associative = 1; ebx->split.coherency_line_size = line_size - 1; - ebx->split.ways_of_associativity = assocs[assoc] - 1; + ebx->split.ways_of_associativity = assoc - 1; ebx->split.physical_line_partition = lines_per_tag - 1; ecx->split.number_of_sets = (size_in_kb * 1024) / line_size / (ebx->split.ways_of_associativity + 1) - 1; @@ -522,6 +522,18 @@ static void __cpuinit cache_shared_cpu_map_setup(unsigned int cpu, int index) int index_msb, i; struct cpuinfo_x86 *c = &cpu_data(cpu); + if ((index == 3) && (c->x86_vendor == X86_VENDOR_AMD)) { + struct cpuinfo_x86 *d; + for_each_online_cpu(i) { + if (!per_cpu(cpuid4_info, i)) + continue; + d = &cpu_data(i); + this_leaf = CPUID4_INFO_IDX(i, index); + cpumask_copy(to_cpumask(this_leaf->shared_cpu_map), + d->llc_shared_map); + } + return; + } this_leaf = CPUID4_INFO_IDX(cpu, index); num_threads_sharing = 1 + this_leaf->eax.split.num_threads_sharing; -- cgit v1.2.3-70-g09d2 From cb9805ab5b97079f69a21b6b4e344a69d5c96157 Mon Sep 17 00:00:00 2001 From: Andreas Herrmann Date: Thu, 3 Sep 2009 09:42:20 +0200 Subject: x86, mcheck: Use correct cpumask for shared bank4 This fixes threshold_bank4 support on multi-node processors. The correct mask to use is llc_shared_map, representing an internal node on Magny-Cours. We need to create 2 sets of symlinks for sibling shared banks -- one set for each internal node, symlinks of each set should target the first core on same internal node. Currently only one set is created where all symlinks are targeting the first core of the entire socket. Signed-off-by: Andreas Herrmann Signed-off-by: H. Peter Anvin --- arch/x86/kernel/cpu/mcheck/mce_amd.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) (limited to 'arch/x86/kernel/cpu') diff --git a/arch/x86/kernel/cpu/mcheck/mce_amd.c b/arch/x86/kernel/cpu/mcheck/mce_amd.c index ddae21620bd..1fecba404fd 100644 --- a/arch/x86/kernel/cpu/mcheck/mce_amd.c +++ b/arch/x86/kernel/cpu/mcheck/mce_amd.c @@ -489,12 +489,14 @@ static __cpuinit int threshold_create_bank(unsigned int cpu, unsigned int bank) int i, err = 0; struct threshold_bank *b = NULL; char name[32]; + struct cpuinfo_x86 *c = &cpu_data(cpu); + sprintf(name, "threshold_bank%i", bank); #ifdef CONFIG_SMP if (cpu_data(cpu).cpu_core_id && shared_bank[bank]) { /* symlink */ - i = cpumask_first(cpu_core_mask(cpu)); + i = cpumask_first(c->llc_shared_map); /* first core not up yet */ if (cpu_data(i).cpu_core_id) @@ -514,7 +516,7 @@ static __cpuinit int threshold_create_bank(unsigned int cpu, unsigned int bank) if (err) goto out; - cpumask_copy(b->cpus, cpu_core_mask(cpu)); + cpumask_copy(b->cpus, c->llc_shared_map); per_cpu(threshold_banks, cpu)[bank] = b; goto out; @@ -539,7 +541,7 @@ static __cpuinit int threshold_create_bank(unsigned int cpu, unsigned int bank) #ifndef CONFIG_SMP cpumask_setall(b->cpus); #else - cpumask_copy(b->cpus, cpu_core_mask(cpu)); + cpumask_copy(b->cpus, c->llc_shared_map); #endif per_cpu(threshold_banks, cpu)[bank] = b; -- cgit v1.2.3-70-g09d2 From 0d96b9ff748b5f57d6f1d6d21209f5745245aadc Mon Sep 17 00:00:00 2001 From: Yinghai Lu Date: Sat, 29 Aug 2009 13:17:14 -0700 Subject: x86: Use hard_smp_processor_id() to get apic id for AMD K8 cpus Otherwise, system with apci id lifting will have wrong apicid in /proc/cpuinfo. and use that in srat_detect_node(). Signed-off-by: Yinghai Lu Cc: Andreas Herrmann Cc: Suresh Siddha Cc: Cyrill Gorcunov LKML-Reference: <4A998CCA.1040407@kernel.org> Signed-off-by: Ingo Molnar --- arch/x86/kernel/cpu/amd.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'arch/x86/kernel/cpu') diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index a76d2c18f15..e1600c775e1 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -336,7 +336,7 @@ static void __cpuinit srat_detect_node(struct cpuinfo_x86 *c) #if defined(CONFIG_NUMA) && defined(CONFIG_X86_64) int cpu = smp_processor_id(); int node; - unsigned apicid = cpu_has_apic ? hard_smp_processor_id() : c->apicid; + unsigned apicid = c->apicid; node = per_cpu(cpu_llc_id, cpu); @@ -481,6 +481,9 @@ static void __cpuinit init_amd(struct cpuinfo_x86 *c) } if (c->x86 == 0x10 || c->x86 == 0x11) set_cpu_cap(c, X86_FEATURE_REP_GOOD); + + /* get apicid instead of initial apic id from cpuid */ + c->apicid = hard_smp_processor_id(); #else /* -- cgit v1.2.3-70-g09d2