x86/CPU/AMD: Fix Zen5 TLB sizes reporting
PATCH 1/1 x86/CPU/AMD: Fix Zen5 TLB sizes reporting
From: "Borislav Petkov (AMD)" <[email protected]>
Starting with Zen5, TLB sizes in CPUID_Fn80000006_E[AB]X are reported as multiples of 32. There's a CPUID bit which determines that:
CPUID_Fn80000021_EAX [Extended Feature 2 EAX] (Core::X86::Cpuid::FeatureExt2Eax)
...
14: L2TlbSizeX32. Read-only. Reset: 1. Indicates that L2TLB sizes are encoded as multiples of 32.
Update the places which report that information.
With it, the numbers look correct now:
-Last level iTLB entries: 4KB 64, 2MB 64, 4MB 32 -Last level dTLB entries: 4KB 128, 2MB 128, 4MB 64, 1GB 0 +Last level iTLB entries: 4KB 2048, 2MB 2048, 4MB 1024 +Last level dTLB entries: 4KB 4096, 2MB 4096, 4MB 2048, 1GB 0
/proc/cpuinfo
-TLB size : 192 4K pages
+TLB size : 6144 4K pages
--- arch/x86/include/asm/cpufeatures.h | 2 ++ arch/x86/kernel/cpu/amd.c | 21 +++++++++++++-------- arch/x86/kernel/cpu/common.c | 4 +++- 3 files changed, 18 insertions(+), 9 deletions(-)
arch/x86/include/asm/cpufeatures.h +2 -0
@@ -473,6 +473,8 @@#define X86_FEATURE_AUTOIBRS (20*32+ 8)/* Automatic IBRS */ #define X86_FEATURE_NO_SMM_CTL_MSR (20*32+ 9)/* SMM_CTL MSR is not present */ +#define X86_FEATURE_L2_TLB_SIZE_X32 (20*32+14)/* L2 TLB sizes are encoded as multiples of 32 */ +#define X86_FEATURE_GP_ON_USER_CPUID (20*32+17)/* User CPUID faulting */ #define X86_FEATURE_PREFETCHI (20*32+20)/* Prefetch Data/Instruction to Cache Level */
arch/x86/kernel/cpu/amd.c +13 -8
@@ -1192,7 +1192,7 @@ static unsigned int amd_size_cache(struct cpuinfo_x86 *c, unsigned int size)static void cpu_detect_tlb_amd (struct cpuinfo_x86 *c) { -u32 ebx, eax, ecx, edx; +u32 ebx, eax, ecx, edx, shift, tmp;u16 mask =0xfff ;if (c->x86 <0xf )
@@ -1201,10 +1201,12 @@ static void cpu_detect_tlb_amd(struct cpuinfo_x86 *c)if (c->extended_cpuid_level <0x80000006 )return ; + shift = !!cpu_has (c,X86_FEATURE_L2_TLB_SIZE_X32 ) *5 ; +cpuid (0x80000006 , &eax, &ebx, &ecx, &edx); - tlb_lld_4k = (ebx >>16 ) & mask; - tlb_lli_4k = ebx & mask; + tlb_lld_4k = ((ebx >>16 ) & mask) << shift; + tlb_lli_4k = (ebx & mask) << shift;/* * K8 doesn't have 2M/4M entries in the L2 TLB so read out the L1 TLB
@@ -1216,16 +1218,18 @@ static void cpu_detect_tlb_amd(struct cpuinfo_x86 *c) }/* Handle DTLB 2M and 4M sizes, fall back to L1 if L2 is disabled */ -if (!((eax >>16 ) & mask)) + tmp = ((eax >>16 ) & mask) << shift; +if (!tmp) tlb_lld_2m = (cpuid_eax (0x80000005 ) >>16 ) &0xff ;else - tlb_lld_2m = (eax >>16 ) & mask; + tlb_lld_2m = tmp;/* a 4M entry uses two 2M entries */ tlb_lld_4m = tlb_lld_2m >>1 ;/* Handle ITLB 2M and 4M sizes, fall back to L1 if L2 is disabled */ -if (!(eax & mask)) { + tmp = (eax & mask) << shift; +if (!tmp) {/* Erratum 658 */ if (c->x86 ==0x15 && c->x86_model <=0x1f ) { tlb_lli_2m =1024 ;
@@ -1233,8 +1237,9 @@ static void cpu_detect_tlb_amd(struct cpuinfo_x86 *c)cpuid (0x80000005 , &eax, &ebx, &ecx, &edx); tlb_lli_2m = eax &0xff ; } - }else - tlb_lli_2m = eax & mask; + }else { + tlb_lli_2m = tmp; + } tlb_lli_4m = tlb_lli_2m >>1 ;
arch/x86/kernel/cpu/common.c +3 -1
@@ -857,7 +857,7 @@ static void get_model_name(struct cpuinfo_x86 *c)void cpu_detect_cache_sizes (struct cpuinfo_x86 *c) { -unsigned int n, dummy, ebx, ecx, edx, l2size; +unsigned int n, dummy, ebx, ecx, edx, l2size, shift; n = c->extended_cpuid_level;
@@ -875,9 +875,11 @@ void cpu_detect_cache_sizes(struct cpuinfo_x86 *c)cpuid (0x80000006 , &dummy, &ebx, &ecx, &edx); l2size = ecx >>16 ; + shift = !!cpu_has (c,X86_FEATURE_L2_TLB_SIZE_X32 ) *5 ;#ifdef CONFIG_X86_64 c->x86_tlbsize += ((ebx >>16 ) &0xfff ) + (ebx &0xfff ); + c->x86_tlbsize <<= shift;#else /* do processor-specific cache resizing */ if (this_cpu->legacy_cache_size)
Signature · 1 line
2.53.0
Borislav Petkov
maintainerQuoted 9 lines · Borislav Petkov
static void
cpu_detect_tlb_amd(struct cpuinfo_x86 *c){arch/x86/kernel/cpu/amd.c +1
-1-u32 ebx, eax, ecx, edx; +u32 ebx, eax, ecx, edx, shift, tmp;u16 mask =0xfff ;if (c->x86 <0xf )@@ -1201,10 +1201,12 @@ static void cpu_detect_tlb_amd(struct cpuinfo_x86 *c)if (c->extended_cpuid_level <0x80000006 )
arch/x86/kernel/cpu/amd.c static void cpu_detect_tlb_amd(struct cpuinfo_x86 *c) · line 1209
Quoted 10 lines
arch/x86/kernel/cpu/amd.c +4
-2return ; + shift = !!cpu_has (c,X86_FEATURE_L2_TLB_SIZE_X32 ) *5 ; +cpuid (0x80000006 , &eax, &ebx, &ecx, &edx); - tlb_lld_4k = (ebx >>16 ) & mask; - tlb_lli_4k = ebx & mask; + tlb_lld_4k = ((ebx >>16 ) & mask) << shift; + tlb_lli_4k = (ebx & mask) << shift;
Sashiko says:
| Could this shift operation overflow the target variables? | The variables tlb_lld_4k and tlb_lli_4k are defined globally as u16. | The architectural maximum raw value for the CPUID mask is 4095. When | shifted by 5 for the new X32 encoding, this yields up to 131040, which | exceeds the 65535 maximum capacity of a u16 variable. | While current Zen 5 CPUs might report smaller values that fit safely, | future CPUs returning larger raw values could be truncated.
I say, sure, let's get there first. More than 65K TLB entries - that would be cheap TLB design when it comes to power... /eyeroll.
I'll gladly touch that bunch of u16s then.
arch/x86/kernel/cpu/common.c void cpu_detect_cache_sizes(struct cpuinfo_x86 *c) · line 883
Quoted 10 lines
arch/x86/kernel/cpu/common.c +2
-0@@ -875,9 +875,11 @@ void cpu_detect_cache_sizes(struct cpuinfo_x86 *c)cpuid (0x80000006 , &dummy, &ebx, &ecx, &edx); l2size = ecx >>16 ; + shift = !!cpu_has (c,X86_FEATURE_L2_TLB_SIZE_X32 ) *5 ;#ifdef CONFIG_X86_64 c->x86_tlbsize += ((ebx >>16 ) &0xfff ) + (ebx &0xfff ); + c->x86_tlbsize <<= shift;#else
Sashiko says:
| Will this trigger a compiler warning on 32-bit builds? | The shift variable is assigned unconditionally here, but its only usage | is bounded by the CONFIG_X86_64 block just below it. On targets | where CONFIG_X86_64 is not set, shift is written to but never read.
Sure, fixed.
Thx.
Signature · 4 lines
Regards/Gruss,
Boris.
https://people.kernel.org/tglx/notes-about-netiquettekernel test robot
build botkernel test robot noticed the following build warnings:
Hi Borislav,
kernel test robot noticed the following build warnings:
[auto build test WARNING on tip/x86/core]
[also build test WARNING on tip/master linus/master v7.2 next-20260821] [cannot apply to tip/auto-latest]
[If your patch is applied to the wrong git tree, kindly drop us a note. And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch#_base_tree_information]
url: https://github.com/intel-lab-lkp/linux/commits/Borislav-Petkov/x86-CPU-AMD-Fix-Zen5-TLB-sizes-reporting/20260820-192031 base: tip/x86/core patch link: https://lore.kernel.org/r/20260821022031.946311-1-bp%40kernel.org patch subject: [PATCH] x86/CPU/AMD: Fix Zen5 TLB sizes reporting config: i386-allmodconfig (https://download.01.org/0day-ci/archive/20260824/[email protected]/config) compiler: gcc-14 (Debian 14.2.0-19) 14.2.0 reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20260824/[email protected]/reproduce)
If you fix the issue in a separate patch/commit (i.e. not just a new version of the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <[email protected]>
| Closes: https://lore.kernel.org/oe-kbuild-all/[email protected]/
All warnings (new ones prefixed by >>):
arch/x86/kernel/cpu/common.c: In function 'cpu_detect_cache_sizes':
>> arch/x86/kernel/cpu/common.c:860:55: warning: variable 'shift' set but not used [-Wunused-but-set-variable]
860 | unsigned int n, dummy, ebx, ecx, edx, l2size, shift;
| ^~~~~vim +/shift +860 arch/x86/kernel/cpu/common.c
857
858 void cpu_detect_cache_sizes(struct cpuinfo_x86 *c)
859 {
> 860 unsigned int n, dummy, ebx, ecx, edx, l2size, shift;
861
862 n = c->extended_cpuid_level;
863
864 if (n >= 0x80000005) {
865 cpuid(0x80000005, &dummy, &ebx, &ecx, &edx);
866 c->x86_cache_size = (ecx>>24) + (edx>>24);
867 #ifdef CONFIG_X86_64
868 /* On K8 L1 TLB is inclusive, so don't count it */
869 c->x86_tlbsize = 0;
870 #endif
871 }
872
873 if (n < 0x80000006) /* Some chips just has a large L1. */
874 return;
875
876 cpuid(0x80000006, &dummy, &ebx, &ecx, &edx);
877 l2size = ecx >> 16;
878 shift = !!cpu_has(c, X86_FEATURE_L2_TLB_SIZE_X32) * 5;
879
--
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki
Borislav Petkov
maintainerQuoted 24 lines · kernel test robot
Hi Borislav,
kernel test robot noticed the following build warnings:
[auto build test WARNING on tip/x86/core]
[also build test WARNING on tip/master linus/master v7.2 next-20260821] [cannot apply to tip/auto-latest]
[If your patch is applied to the wrong git tree, kindly drop us a note. And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch#_base_tree_information]url: https://github.com/intel-lab-lkp/linux/commits/Borislav-Petkov/x86-CPU-AMD-Fix-Zen5-TLB-sizes-reporting/20260820-192031 base: tip/x86/core patch link: https://lore.kernel.org/r/20260821022031.946311-1-bp%40kernel.org patch subject: [PATCH] x86/CPU/AMD: Fix Zen5 TLB sizes reporting config: i386-allmodconfig (https://download.01.org/0day-ci/archive/20260824/[email protected]/config) compiler: gcc-14 (Debian 14.2.0-19) 14.2.0 reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20260824/[email protected]/reproduce)If you fix the issue in a separate patch/commit (i.e. not just a new version of the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <[email protected]>
| Closes: https://lore.kernel.org/oe-kbuild-all/[email protected]/
Quoted 6 lines
All warnings (new ones prefixed by >>):
arch/x86/kernel/cpu/common.c: In function 'cpu_detect_cache_sizes':Quoted 1 line
Quoted 1 line
arch/x86/kernel/cpu/common.c:860:55: warning: variable 'shift' set but not used [-Wunused-but-set-variable]860 | unsigned int n, dummy, ebx, ecx, edx, l2size, shift; | ^~~~~
Yah, AI caught it already, try this:
---
From: "Borislav Petkov (AMD)" <[email protected]>
Date: Thu, 20 Aug 2026 17:02:10 -0700
Subject: [PATCH] x86/CPU/AMD: Fix Zen5 TLB sizes reporting
Starting with Zen5, TLB sizes in CPUID_Fn80000006_E[AB]X are reported as multiples of 32. There's a CPUID bit which determines that:
CPUID_Fn80000021_EAX [Extended Feature 2 EAX] (Core::X86::Cpuid::FeatureExt2Eax)
...
14: L2TlbSizeX32. Read-only. Reset: 1. Indicates that L2TLB sizes are encoded as multiples of 32.
Update the places which report that information.
With it, the numbers look correct now:
-Last level iTLB entries: 4KB 64, 2MB 64, 4MB 32 -Last level dTLB entries: 4KB 128, 2MB 128, 4MB 64, 1GB 0 +Last level iTLB entries: 4KB 2048, 2MB 2048, 4MB 1024 +Last level dTLB entries: 4KB 4096, 2MB 4096, 4MB 2048, 1GB 0
/proc/cpuinfo
-TLB size : 192 4K pages
+TLB size : 6144 4K pages
--- arch/x86/include/asm/cpufeatures.h | 2 ++ arch/x86/kernel/cpu/amd.c | 21 +++++++++++++-------- arch/x86/kernel/cpu/common.c | 4 +++- 3 files changed, 18 insertions(+), 9 deletions(-)
arch/x86/include/asm/cpufeatures.h +2 -0
@@ -473,6 +473,8 @@#define X86_FEATURE_AUTOIBRS (20*32+ 8)/* Automatic IBRS */ #define X86_FEATURE_NO_SMM_CTL_MSR (20*32+ 9)/* SMM_CTL MSR is not present */ +#define X86_FEATURE_L2_TLB_SIZE_X32 (20*32+14)/* L2 TLB sizes are encoded as multiples of 32 */ +#define X86_FEATURE_GP_ON_USER_CPUID (20*32+17)/* User CPUID faulting */ #define X86_FEATURE_PREFETCHI (20*32+20)/* Prefetch Data/Instruction to Cache Level */
arch/x86/kernel/cpu/amd.c +13 -8
@@ -1192,7 +1192,7 @@ static unsigned int amd_size_cache(struct cpuinfo_x86 *c, unsigned int size)static void cpu_detect_tlb_amd (struct cpuinfo_x86 *c) { -u32 ebx, eax, ecx, edx; +u32 ebx, eax, ecx, edx, shift, tmp;u16 mask =0xfff ;if (c->x86 <0xf )
@@ -1201,10 +1201,12 @@ static void cpu_detect_tlb_amd(struct cpuinfo_x86 *c)if (c->extended_cpuid_level <0x80000006 )return ; + shift = !!cpu_has (c,X86_FEATURE_L2_TLB_SIZE_X32 ) *5 ; +cpuid (0x80000006 , &eax, &ebx, &ecx, &edx); - tlb_lld_4k = (ebx >>16 ) & mask; - tlb_lli_4k = ebx & mask; + tlb_lld_4k = ((ebx >>16 ) & mask) << shift; + tlb_lli_4k = (ebx & mask) << shift;/* * K8 doesn't have 2M/4M entries in the L2 TLB so read out the L1 TLB
@@ -1216,16 +1218,18 @@ static void cpu_detect_tlb_amd(struct cpuinfo_x86 *c) }/* Handle DTLB 2M and 4M sizes, fall back to L1 if L2 is disabled */ -if (!((eax >>16 ) & mask)) + tmp = ((eax >>16 ) & mask) << shift; +if (!tmp) tlb_lld_2m = (cpuid_eax (0x80000005 ) >>16 ) &0xff ;else - tlb_lld_2m = (eax >>16 ) & mask; + tlb_lld_2m = tmp;/* a 4M entry uses two 2M entries */ tlb_lld_4m = tlb_lld_2m >>1 ;/* Handle ITLB 2M and 4M sizes, fall back to L1 if L2 is disabled */ -if (!(eax & mask)) { + tmp = (eax & mask) << shift; +if (!tmp) {/* Erratum 658 */ if (c->x86 ==0x15 && c->x86_model <=0x1f ) { tlb_lli_2m =1024 ;
@@ -1233,8 +1237,9 @@ static void cpu_detect_tlb_amd(struct cpuinfo_x86 *c)cpuid (0x80000005 , &eax, &ebx, &ecx, &edx); tlb_lli_2m = eax &0xff ; } - }else - tlb_lli_2m = eax & mask; + }else { + tlb_lli_2m = tmp; + } tlb_lli_4m = tlb_lli_2m >>1 ;
arch/x86/kernel/cpu/common.c +3 -1
@@ -857,7 +857,7 @@ static void get_model_name(struct cpuinfo_x86 *c)void cpu_detect_cache_sizes (struct cpuinfo_x86 *c) { -unsigned int n, dummy, ebx, ecx, edx, l2size; +unsigned int n, dummy, ebx, ecx, edx, l2size, shift__maybe_unused ; n = c->extended_cpuid_level;
@@ -877,7 +877,9 @@ void cpu_detect_cache_sizes(struct cpuinfo_x86 *c) l2size = ecx >>16 ;#ifdef CONFIG_X86_64 + shift = !!cpu_has (c,X86_FEATURE_L2_TLB_SIZE_X32 ) *5 ; c->x86_tlbsize += ((ebx >>16 ) &0xfff ) + (ebx &0xfff ); + c->x86_tlbsize <<= shift;#else /* do processor-specific cache resizing */ if (this_cpu->legacy_cache_size)
Signature · 7 lines
2.53.0
--
Regards/Gruss,
Boris.
https://people.kernel.org/tglx/notes-about-netiquette