Re: [PATCH, resend] x86-64: don't track CPU model data that is usedby 32-bit code only

From: Ingo Molnar
Date: Fri Oct 18 2013 - 02:25:01 EST



* Jan Beulich <JBeulich@xxxxxxxx> wrote:

> 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.
>
> Signed-off-by: Jan Beulich <jbeulich@xxxxxxxx>
> ---
> arch/x86/kernel/cpu/amd.c | 2 +-
> arch/x86/kernel/cpu/centaur.c | 6 ++++--
> arch/x86/kernel/cpu/common.c | 4 +++-
> arch/x86/kernel/cpu/cpu.h | 16 +++++++---------
> arch/x86/kernel/cpu/intel.c | 8 ++++----
> arch/x86/kernel/cpu/umc.c | 2 +-
> 6 files changed, 20 insertions(+), 18 deletions(-)
>
> --- 3.12-rc5/arch/x86/kernel/cpu/amd.c
> +++ 3.12-rc5-x86-cpu-32bit-only-pieces/arch/x86/kernel/cpu/amd.c
> @@ -824,7 +824,7 @@ static const struct cpu_dev amd_cpu_dev
> .c_ident = { "AuthenticAMD" },
> #ifdef CONFIG_X86_32
> .c_models = {
> - { .vendor = X86_VENDOR_AMD, .family = 4, .model_names =
> + { .family = 4, .model_names =
> {
> [3] = "486 DX/2",
> [7] = "486 DX/2-WB",
> --- 3.12-rc5/arch/x86/kernel/cpu/centaur.c
> +++ 3.12-rc5-x86-cpu-32bit-only-pieces/arch/x86/kernel/cpu/centaur.c
> @@ -468,10 +468,10 @@ static void init_centaur(struct cpuinfo_
> #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
> 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,
> +#ifdef CONFIG_X86_32
> .c_size_cache = centaur_size_cache,
> +#endif
> .c_x86_vendor = X86_VENDOR_CENTAUR,
> };
>
> --- 3.12-rc5/arch/x86/kernel/cpu/common.c
> +++ 3.12-rc5-x86-cpu-32bit-only-pieces/arch/x86/kernel/cpu/common.c
> @@ -346,6 +346,7 @@ static void filter_cpuid_features(struct
> /* Look up CPU names by table lookup. */
> static const char *table_lookup_model(struct cpuinfo_x86 *c)
> {
> +#ifdef CONFIG_X86_32
> const struct cpu_model_info *info;
>
> if (c->x86_model >= 16)
> @@ -356,11 +357,12 @@ static const char *table_lookup_model(st
>
> info = this_cpu->c_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 */
> }
>
> --- 3.12-rc5/arch/x86/kernel/cpu/cpu.h
> +++ 3.12-rc5-x86-cpu-32bit-only-pieces/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 +8,19 @@ 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
> + unsigned int (*c_size_cache)(struct cpuinfo_x86 *, unsigned int);
> + struct cpu_model_info {
> + int family;
> + const char *model_names[16];
> + } c_models[5];
> +#endif
> };
>
> struct _tlb_table {
> --- 3.12-rc5/arch/x86/kernel/cpu/intel.c
> +++ 3.12-rc5-x86-cpu-32bit-only-pieces/arch/x86/kernel/cpu/intel.c
> @@ -666,7 +666,7 @@ static const struct cpu_dev intel_cpu_de
> .c_ident = { "GenuineIntel" },
> #ifdef CONFIG_X86_32
> .c_models = {
> - { .vendor = X86_VENDOR_INTEL, .family = 4, .model_names =
> + { .family = 4, .model_names =
> {
> [0] = "486 DX-25/33",
> [1] = "486 DX-50",
> @@ -679,7 +679,7 @@ static const struct cpu_dev intel_cpu_de
> [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_de
> [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_de
> [11] = "Pentium III (Tualatin)",
> }
> },
> - { .vendor = X86_VENDOR_INTEL, .family = 15, .model_names =
> + { .family = 15, .model_names =
> {
> [0] = "Pentium 4 (Unknown)",
> [1] = "Pentium 4 (Willamette)",
> --- 3.12-rc5/arch/x86/kernel/cpu/umc.c
> +++ 3.12-rc5-x86-cpu-32bit-only-pieces/arch/x86/kernel/cpu/umc.c
> @@ -12,7 +12,7 @@ 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 =
> + { .family = 4, .model_names =
> {
> [1] = "U5D",
> [2] = "U5S",

So I'm not totally convinced about this as it increases the #ifdef count -
that's why I didn't pick up the earlier submission.

But I guess we could do it if you do one more cleanup: please rename it
all to ->legacy_model_names, ->legacy_cache_size, etc., to make sure it's
apparent at first sight that this is an old, secondary identification
mechanism for pre-sane-CPUID CPUs.

Also maybe describe the fields in a comment line as well.

Thanks,

Ingo
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/