From 09dc68d958c67c76cf672ec78b7391af453010f8 Mon Sep 17 00:00:00 2001 From: Jan Beulich Date: Mon, 21 Oct 2013 09:35:20 +0100 Subject: x86/cpu: Track legacy CPU model data only on 32-bit kernels struct cpu_dev's c_models is only ever set inside CONFIG_X86_32 conditionals (or code that's being built for 32-bit only), so there's no use of reserving the (empty) space for the model names in a 64-bit kernel. Similarly, c_size_cache is only used in the #else of a CONFIG_X86_64 conditional, so reserving space for (and in one case even initializing) that field is pointless for 64-bit kernels too. While moving both fields to the end of the structure, I also noticed that: - the c_models array size was one too small, potentially causing table_lookup_model() to return garbage on Intel CPUs (intel.c's instance was lacking the sentinel with family being zero), so the patch bumps that by one, - c_models' vendor sub-field was unused (and anyway redundant with the base structure's c_x86_vendor field), so the patch deletes it. Also rename the legacy fields so that their legacy nature stands out and comment their declarations. Signed-off-by: Jan Beulich Link: http://lkml.kernel.org/r/5265036802000078000FC4DB@nat28.tlf.novell.com Signed-off-by: Ingo Molnar diff --git a/arch/x86/kernel/cpu/amd.c b/arch/x86/kernel/cpu/amd.c index 903a264..3daece7 100644 --- a/arch/x86/kernel/cpu/amd.c +++ b/arch/x86/kernel/cpu/amd.c @@ -823,8 +823,8 @@ static const struct cpu_dev amd_cpu_dev = { .c_vendor = "AMD", .c_ident = { "AuthenticAMD" }, #ifdef CONFIG_X86_32 - .c_models = { - { .vendor = X86_VENDOR_AMD, .family = 4, .model_names = + .legacy_models = { + { .family = 4, .model_names = { [3] = "486 DX/2", [7] = "486 DX/2-WB", @@ -835,7 +835,7 @@ static const struct cpu_dev amd_cpu_dev = { } }, }, - .c_size_cache = amd_size_cache, + .legacy_cache_size = amd_size_cache, #endif .c_early_init = early_init_amd, .c_detect_tlb = cpu_detect_tlb_amd, diff --git a/arch/x86/kernel/cpu/centaur.c b/arch/x86/kernel/cpu/centaur.c index fbf6c3b..8d5652d 100644 --- a/arch/x86/kernel/cpu/centaur.c +++ b/arch/x86/kernel/cpu/centaur.c @@ -468,10 +468,10 @@ static void init_centaur(struct cpuinfo_x86 *c) #endif } +#ifdef CONFIG_X86_32 static unsigned int centaur_size_cache(struct cpuinfo_x86 *c, unsigned int size) { -#ifdef CONFIG_X86_32 /* VIA C3 CPUs (670-68F) need further shifting. */ if ((c->x86 == 6) && ((c->x86_model == 7) || (c->x86_model == 8))) size >>= 8; @@ -484,16 +484,18 @@ centaur_size_cache(struct cpuinfo_x86 *c, unsigned int size) if ((c->x86 == 6) && (c->x86_model == 9) && (c->x86_mask == 1) && (size == 65)) size -= 1; -#endif return size; } +#endif static const struct cpu_dev centaur_cpu_dev = { .c_vendor = "Centaur", .c_ident = { "CentaurHauls" }, .c_early_init = early_init_centaur, .c_init = init_centaur, - .c_size_cache = centaur_size_cache, +#ifdef CONFIG_X86_32 + .legacy_cache_size = centaur_size_cache, +#endif .c_x86_vendor = X86_VENDOR_CENTAUR, }; diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c index 2793d1f..9ada0b3 100644 --- a/arch/x86/kernel/cpu/common.c +++ b/arch/x86/kernel/cpu/common.c @@ -346,7 +346,8 @@ static void filter_cpuid_features(struct cpuinfo_x86 *c, bool warn) /* Look up CPU names by table lookup. */ static const char *table_lookup_model(struct cpuinfo_x86 *c) { - const struct cpu_model_info *info; +#ifdef CONFIG_X86_32 + const struct legacy_cpu_model_info *info; if (c->x86_model >= 16) return NULL; /* Range check */ @@ -354,13 +355,14 @@ static const char *table_lookup_model(struct cpuinfo_x86 *c) if (!this_cpu) return NULL; - info = this_cpu->c_models; + info = this_cpu->legacy_models; - while (info && info->family) { + while (info->family) { if (info->family == c->x86) return info->model_names[c->x86_model]; info++; } +#endif return NULL; /* Not found */ } @@ -450,8 +452,8 @@ void cpu_detect_cache_sizes(struct cpuinfo_x86 *c) c->x86_tlbsize += ((ebx >> 16) & 0xfff) + (ebx & 0xfff); #else /* do processor-specific cache resizing */ - if (this_cpu->c_size_cache) - l2size = this_cpu->c_size_cache(c, l2size); + if (this_cpu->legacy_cache_size) + l2size = this_cpu->legacy_cache_size(c, l2size); /* Allow user to override all this if necessary. */ if (cachesize_override != -1) diff --git a/arch/x86/kernel/cpu/cpu.h b/arch/x86/kernel/cpu/cpu.h index 4041c24..c37dc37 100644 --- a/arch/x86/kernel/cpu/cpu.h +++ b/arch/x86/kernel/cpu/cpu.h @@ -1,12 +1,6 @@ #ifndef ARCH_X86_CPU_H #define ARCH_X86_CPU_H -struct cpu_model_info { - int vendor; - int family; - const char *model_names[16]; -}; - /* attempt to consolidate cpu attributes */ struct cpu_dev { const char *c_vendor; @@ -14,15 +8,23 @@ struct cpu_dev { /* some have two possibilities for cpuid string */ const char *c_ident[2]; - struct cpu_model_info c_models[4]; - void (*c_early_init)(struct cpuinfo_x86 *); void (*c_bsp_init)(struct cpuinfo_x86 *); void (*c_init)(struct cpuinfo_x86 *); void (*c_identify)(struct cpuinfo_x86 *); void (*c_detect_tlb)(struct cpuinfo_x86 *); - unsigned int (*c_size_cache)(struct cpuinfo_x86 *, unsigned int); int c_x86_vendor; +#ifdef CONFIG_X86_32 + /* Optional vendor specific routine to obtain the cache size. */ + unsigned int (*legacy_cache_size)(struct cpuinfo_x86 *, + unsigned int); + + /* Family/stepping-based lookup table for model names. */ + struct legacy_cpu_model_info { + int family; + const char *model_names[16]; + } legacy_models[5]; +#endif }; struct _tlb_table { diff --git a/arch/x86/kernel/cpu/intel.c b/arch/x86/kernel/cpu/intel.c index ec72995..dc1ec0d 100644 --- a/arch/x86/kernel/cpu/intel.c +++ b/arch/x86/kernel/cpu/intel.c @@ -665,8 +665,8 @@ static const struct cpu_dev intel_cpu_dev = { .c_vendor = "Intel", .c_ident = { "GenuineIntel" }, #ifdef CONFIG_X86_32 - .c_models = { - { .vendor = X86_VENDOR_INTEL, .family = 4, .model_names = + .legacy_models = { + { .family = 4, .model_names = { [0] = "486 DX-25/33", [1] = "486 DX-50", @@ -679,7 +679,7 @@ static const struct cpu_dev intel_cpu_dev = { [9] = "486 DX/4-WB" } }, - { .vendor = X86_VENDOR_INTEL, .family = 5, .model_names = + { .family = 5, .model_names = { [0] = "Pentium 60/66 A-step", [1] = "Pentium 60/66", @@ -690,7 +690,7 @@ static const struct cpu_dev intel_cpu_dev = { [8] = "Mobile Pentium MMX" } }, - { .vendor = X86_VENDOR_INTEL, .family = 6, .model_names = + { .family = 6, .model_names = { [0] = "Pentium Pro A-step", [1] = "Pentium Pro", @@ -704,7 +704,7 @@ static const struct cpu_dev intel_cpu_dev = { [11] = "Pentium III (Tualatin)", } }, - { .vendor = X86_VENDOR_INTEL, .family = 15, .model_names = + { .family = 15, .model_names = { [0] = "Pentium 4 (Unknown)", [1] = "Pentium 4 (Willamette)", @@ -714,7 +714,7 @@ static const struct cpu_dev intel_cpu_dev = { } }, }, - .c_size_cache = intel_size_cache, + .legacy_cache_size = intel_size_cache, #endif .c_detect_tlb = intel_detect_tlb, .c_early_init = early_init_intel, diff --git a/arch/x86/kernel/cpu/umc.c b/arch/x86/kernel/cpu/umc.c index 202759a..75c5ad5 100644 --- a/arch/x86/kernel/cpu/umc.c +++ b/arch/x86/kernel/cpu/umc.c @@ -11,8 +11,8 @@ static const struct cpu_dev umc_cpu_dev = { .c_vendor = "UMC", .c_ident = { "UMC UMC UMC" }, - .c_models = { - { .vendor = X86_VENDOR_UMC, .family = 4, .model_names = + .legacy_models = { + { .family = 4, .model_names = { [1] = "U5D", [2] = "U5S", -- cgit v0.10.2 From a477c8594bee3bff639739c48258a8c737ab721e Mon Sep 17 00:00:00 2001 From: HATAYAMA Daisuke Date: Tue, 5 Nov 2013 02:15:48 +0900 Subject: x86/cpu: Always print SMP information in /proc/cpuinfo Currently show_cpuinfo_core() displays cpu core information only if the number of threads per a whole cores is 2 or larger. However, this condition doesn't care about the number of sockets. For example, this condition doesn't hold on systems with two logical cpus consisting of two sockets and a single core on each socket - yet the topology information would be interesting to see in that case as well. I don't know whether or not there are processors in real world by which such configurations are possible, but at least on vitual machine environments, such configuration can occur, typically when no explicit SMP information is provided in advance. For example, on qemu/KVM, SMP information is specified via -smp command-line option, more specifically, its syntax is: -smp n[,cores=cores][,threads=threads][,sockets=sockets][,maxcpus=maxcpus] If this is not specified, qemu tells configuration with n-sockets, 1-core and 1-thread to the guest machine, on which guest, MP information is not displayed in /proc/cpuinfo. I saw this situation on VMWare guest environment, too. To fix this issue, this patch simply removes the condition because this information is useful even if there's only 1 thread. Signed-off-by: HATAYAMA Daisuke Cc: Vivek Goyal Cc: H. Peter Anvin Cc: Linus Torvalds Cc: Andrew Morton Cc: Peter Zijlstra Link: http://lkml.kernel.org/r/5277D644.4090707@jp.fujitsu.com Signed-off-by: Ingo Molnar diff --git a/arch/x86/kernel/cpu/proc.c b/arch/x86/kernel/cpu/proc.c index aee6317..06fe3ed 100644 --- a/arch/x86/kernel/cpu/proc.c +++ b/arch/x86/kernel/cpu/proc.c @@ -11,15 +11,12 @@ static void show_cpuinfo_core(struct seq_file *m, struct cpuinfo_x86 *c, unsigned int cpu) { #ifdef CONFIG_SMP - if (c->x86_max_cores * smp_num_siblings > 1) { - seq_printf(m, "physical id\t: %d\n", c->phys_proc_id); - seq_printf(m, "siblings\t: %d\n", - cpumask_weight(cpu_core_mask(cpu))); - seq_printf(m, "core id\t\t: %d\n", c->cpu_core_id); - seq_printf(m, "cpu cores\t: %d\n", c->booted_cores); - seq_printf(m, "apicid\t\t: %d\n", c->apicid); - seq_printf(m, "initial apicid\t: %d\n", c->initial_apicid); - } + seq_printf(m, "physical id\t: %d\n", c->phys_proc_id); + seq_printf(m, "siblings\t: %d\n", cpumask_weight(cpu_core_mask(cpu))); + seq_printf(m, "core id\t\t: %d\n", c->cpu_core_id); + seq_printf(m, "cpu cores\t: %d\n", c->booted_cores); + seq_printf(m, "apicid\t\t: %d\n", c->apicid); + seq_printf(m, "initial apicid\t: %d\n", c->initial_apicid); #endif } -- cgit v0.10.2 From bb61ccc7dbcbc1ed4600bef5610d99ec91fd7032 Mon Sep 17 00:00:00 2001 From: Josh Boyer Date: Tue, 5 Nov 2013 09:37:29 -0500 Subject: x86/cpu: Allow higher NR_CPUS values The current range for SMP configs is 2 - 512 CPUs, or a full 4096 in the case of MAXSMP. There are machines that have 1024 CPUs in them today and configuring a kernel for that means you are forced to set MAXSMP. This adds additional unnecessary overhead. While that overhead might be considered tiny for large machines, it isn't necessarily so if you are building a kernel that runs across a wide variety of machines. To cover the range of more common machines today, we allow NR_CPUS to be up to 4096 when CPUMASK_OFFSTACK is enabled. Signed-off-by: Josh Boyer Cc: prarit@redhat.com Cc: Russ Anderson Cc: Linus Torvalds Cc: Andrew Morton Link: http://lkml.kernel.org/r/20131105143728.GJ9944@hansolo.jdub.homelinux.org Signed-off-by: Ingo Molnar diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index f67e839..f03e428 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -825,14 +825,16 @@ config MAXSMP config NR_CPUS int "Maximum number of CPUs" if SMP && !MAXSMP range 2 8 if SMP && X86_32 && !X86_BIGSMP - range 2 512 if SMP && !MAXSMP + range 2 512 if SMP && !MAXSMP && !CPUMASK_OFFSTACK + range 2 4096 if SMP && !MAXSMP && CPUMASK_OFFSTACK && X86_64 default "1" if !SMP default "4096" if MAXSMP default "32" if SMP && (X86_NUMAQ || X86_SUMMIT || X86_BIGSMP || X86_ES7000) default "8" if SMP ---help--- This allows you to specify the maximum number of CPUs which this - kernel will support. The maximum supported value is 512 and the + kernel will support. If CPUMASK_OFFSTACK is enabled, the maximum + supported value is 4096, otherwise the maximum value is 512. The minimum value which makes sense is 2. This is purely to save memory - each supported CPU adds -- cgit v0.10.2 From b53b5eda8194214928c8243d711a75dbf51809fc Mon Sep 17 00:00:00 2001 From: Josh Boyer Date: Tue, 5 Nov 2013 09:38:16 -0500 Subject: x86/cpu: Increase max CPU count to 8192 The MAXSMP option is intended to enable silly large numbers of CPUs for testing purposes. The current value of 4096 isn't very silly any longer as there are actual SGI machines that approach 6096 CPUs when taking HT into account. Increase the value to a nice round 8192 to account for this and allow for short term future increases. Signed-off-by: Josh Boyer Cc: prarit@redhat.com Cc: Russ Anderson Cc: Linus Torvalds Cc: Andrew Morton Link: http://lkml.kernel.org/r/20131105143816.GK9944@hansolo.jdub.homelinux.org [ Tweaked it so that MAXSMP simply sets the maximum of the normal range. ] Signed-off-by: Ingo Molnar diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index f03e428..805469a 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -826,9 +826,9 @@ config NR_CPUS int "Maximum number of CPUs" if SMP && !MAXSMP range 2 8 if SMP && X86_32 && !X86_BIGSMP range 2 512 if SMP && !MAXSMP && !CPUMASK_OFFSTACK - range 2 4096 if SMP && !MAXSMP && CPUMASK_OFFSTACK && X86_64 + range 2 8192 if SMP && !MAXSMP && CPUMASK_OFFSTACK && X86_64 default "1" if !SMP - default "4096" if MAXSMP + default "8192" if MAXSMP default "32" if SMP && (X86_NUMAQ || X86_SUMMIT || X86_BIGSMP || X86_ES7000) default "8" if SMP ---help--- -- cgit v0.10.2