diff options
author | H.J. Lu <hjl.tools@gmail.com> | 2020-06-29 16:36:08 -0700 |
---|---|---|
committer | H.J. Lu <hjl.tools@gmail.com> | 2020-07-13 06:05:16 -0700 |
commit | 107e6a3c2212ba7a3a4ec7cae8d82d73f7c95d0b (patch) | |
tree | b23f1c5ba166bd28f519fa36f225a31407b45270 /sysdeps/x86_64/fpu | |
parent | 10b01bd4529336bffc2c398ce43a171ed94aacc7 (diff) | |
download | glibc-107e6a3c2212ba7a3a4ec7cae8d82d73f7c95d0b.zip glibc-107e6a3c2212ba7a3a4ec7cae8d82d73f7c95d0b.tar.gz glibc-107e6a3c2212ba7a3a4ec7cae8d82d73f7c95d0b.tar.bz2 |
x86: Support usable check for all CPU features
Support usable check for all CPU features with the following changes:
1. Change struct cpu_features to
struct cpuid_features
{
struct cpuid_registers cpuid;
struct cpuid_registers usable;
};
struct cpu_features
{
struct cpu_features_basic basic;
struct cpuid_features features[COMMON_CPUID_INDEX_MAX];
unsigned int preferred[PREFERRED_FEATURE_INDEX_MAX];
...
};
so that there is a usable bit for each cpuid bit.
2. After the cpuid bits have been initialized, copy the known bits to the
usable bits. EAX/EBX from INDEX_1 and EAX from INDEX_7 aren't used for
CPU feature detection.
3. Clear the usable bits which require OS support.
4. If the feature is supported by OS, copy its cpuid bit to its usable
bit.
5. Replace HAS_CPU_FEATURE and CPU_FEATURES_CPU_P with CPU_FEATURE_USABLE
and CPU_FEATURE_USABLE_P to check if a feature is usable.
6. Add DEPR_FPU_CS_DS for INDEX_7_EBX_13.
7. Unset MPX feature since it has been deprecated.
The results are
1. If the feature is known and doesn't requre OS support, its usable bit
is copied from the cpuid bit.
2. Otherwise, its usable bit is copied from the cpuid bit only if the
feature is known to supported by OS.
3. CPU_FEATURE_USABLE/CPU_FEATURE_USABLE_P are used to check if the
feature can be used.
4. HAS_CPU_FEATURE/CPU_FEATURE_CPU_P are used to check if CPU supports
the feature.
Diffstat (limited to 'sysdeps/x86_64/fpu')
-rw-r--r-- | sysdeps/x86_64/fpu/math-tests-arch.h | 6 | ||||
-rw-r--r-- | sysdeps/x86_64/fpu/multiarch/ifunc-avx-fma4.h | 8 | ||||
-rw-r--r-- | sysdeps/x86_64/fpu/multiarch/ifunc-fma.h | 4 | ||||
-rw-r--r-- | sysdeps/x86_64/fpu/multiarch/ifunc-fma4.h | 6 | ||||
-rw-r--r-- | sysdeps/x86_64/fpu/multiarch/ifunc-mathvec-avx2.h | 4 | ||||
-rw-r--r-- | sysdeps/x86_64/fpu/multiarch/ifunc-mathvec-avx512.h | 4 | ||||
-rw-r--r-- | sysdeps/x86_64/fpu/multiarch/ifunc-mathvec-sse4_1.h | 2 | ||||
-rw-r--r-- | sysdeps/x86_64/fpu/multiarch/ifunc-sse4_1.h | 2 | ||||
-rw-r--r-- | sysdeps/x86_64/fpu/multiarch/s_fma.c | 4 | ||||
-rw-r--r-- | sysdeps/x86_64/fpu/multiarch/s_fmaf.c | 4 |
10 files changed, 22 insertions, 22 deletions
diff --git a/sysdeps/x86_64/fpu/math-tests-arch.h b/sysdeps/x86_64/fpu/math-tests-arch.h index 435ddad..33ea763 100644 --- a/sysdeps/x86_64/fpu/math-tests-arch.h +++ b/sysdeps/x86_64/fpu/math-tests-arch.h @@ -24,7 +24,7 @@ # define CHECK_ARCH_EXT \ do \ { \ - if (!HAS_ARCH_FEATURE (AVX_Usable)) return; \ + if (!CPU_FEATURE_USABLE (AVX)) return; \ } \ while (0) @@ -34,7 +34,7 @@ # define CHECK_ARCH_EXT \ do \ { \ - if (!HAS_ARCH_FEATURE (AVX2_Usable)) return; \ + if (!CPU_FEATURE_USABLE (AVX2)) return; \ } \ while (0) @@ -44,7 +44,7 @@ # define CHECK_ARCH_EXT \ do \ { \ - if (!HAS_ARCH_FEATURE (AVX512F_Usable)) return; \ + if (!CPU_FEATURE_USABLE (AVX512F)) return; \ } \ while (0) diff --git a/sysdeps/x86_64/fpu/multiarch/ifunc-avx-fma4.h b/sysdeps/x86_64/fpu/multiarch/ifunc-avx-fma4.h index 86835ee..95fe2f4 100644 --- a/sysdeps/x86_64/fpu/multiarch/ifunc-avx-fma4.h +++ b/sysdeps/x86_64/fpu/multiarch/ifunc-avx-fma4.h @@ -29,14 +29,14 @@ IFUNC_SELECTOR (void) { const struct cpu_features* cpu_features = __get_cpu_features (); - if (CPU_FEATURES_ARCH_P (cpu_features, FMA_Usable) - && CPU_FEATURES_ARCH_P (cpu_features, AVX2_Usable)) + if (CPU_FEATURE_USABLE_P (cpu_features, FMA) + && CPU_FEATURE_USABLE_P (cpu_features, AVX2)) return OPTIMIZE (fma); - if (CPU_FEATURES_ARCH_P (cpu_features, FMA4_Usable)) + if (CPU_FEATURE_USABLE_P (cpu_features, FMA4)) return OPTIMIZE (fma4); - if (CPU_FEATURES_ARCH_P (cpu_features, AVX_Usable)) + if (CPU_FEATURE_USABLE_P (cpu_features, AVX)) return OPTIMIZE (avx); return OPTIMIZE (sse2); diff --git a/sysdeps/x86_64/fpu/multiarch/ifunc-fma.h b/sysdeps/x86_64/fpu/multiarch/ifunc-fma.h index 2242d97..0a25a44 100644 --- a/sysdeps/x86_64/fpu/multiarch/ifunc-fma.h +++ b/sysdeps/x86_64/fpu/multiarch/ifunc-fma.h @@ -26,8 +26,8 @@ IFUNC_SELECTOR (void) { const struct cpu_features* cpu_features = __get_cpu_features (); - if (CPU_FEATURES_ARCH_P (cpu_features, FMA_Usable) - && CPU_FEATURES_ARCH_P (cpu_features, AVX2_Usable)) + if (CPU_FEATURE_USABLE_P (cpu_features, FMA) + && CPU_FEATURE_USABLE_P (cpu_features, AVX2)) return OPTIMIZE (fma); return OPTIMIZE (sse2); diff --git a/sysdeps/x86_64/fpu/multiarch/ifunc-fma4.h b/sysdeps/x86_64/fpu/multiarch/ifunc-fma4.h index 03adf86..7659758 100644 --- a/sysdeps/x86_64/fpu/multiarch/ifunc-fma4.h +++ b/sysdeps/x86_64/fpu/multiarch/ifunc-fma4.h @@ -28,11 +28,11 @@ IFUNC_SELECTOR (void) { const struct cpu_features* cpu_features = __get_cpu_features (); - if (CPU_FEATURES_ARCH_P (cpu_features, FMA_Usable) - && CPU_FEATURES_ARCH_P (cpu_features, AVX2_Usable)) + if (CPU_FEATURE_USABLE_P (cpu_features, FMA) + && CPU_FEATURE_USABLE_P (cpu_features, AVX2)) return OPTIMIZE (fma); - if (CPU_FEATURES_ARCH_P (cpu_features, FMA4_Usable)) + if (CPU_FEATURE_USABLE_P (cpu_features, FMA)) return OPTIMIZE (fma4); return OPTIMIZE (sse2); diff --git a/sysdeps/x86_64/fpu/multiarch/ifunc-mathvec-avx2.h b/sysdeps/x86_64/fpu/multiarch/ifunc-mathvec-avx2.h index 9c5c6f1..2655e55 100644 --- a/sysdeps/x86_64/fpu/multiarch/ifunc-mathvec-avx2.h +++ b/sysdeps/x86_64/fpu/multiarch/ifunc-mathvec-avx2.h @@ -31,8 +31,8 @@ IFUNC_SELECTOR (void) { const struct cpu_features* cpu_features = __get_cpu_features (); - if (CPU_FEATURES_ARCH_P (cpu_features, FMA_Usable) - && CPU_FEATURES_ARCH_P (cpu_features, AVX2_Usable)) + if (CPU_FEATURE_USABLE_P (cpu_features, FMA) + && CPU_FEATURE_USABLE_P (cpu_features, AVX2)) return OPTIMIZE (avx2); return OPTIMIZE (sse_wrapper); diff --git a/sysdeps/x86_64/fpu/multiarch/ifunc-mathvec-avx512.h b/sysdeps/x86_64/fpu/multiarch/ifunc-mathvec-avx512.h index 70e22c5..5f83265 100644 --- a/sysdeps/x86_64/fpu/multiarch/ifunc-mathvec-avx512.h +++ b/sysdeps/x86_64/fpu/multiarch/ifunc-mathvec-avx512.h @@ -34,10 +34,10 @@ IFUNC_SELECTOR (void) if (!CPU_FEATURES_ARCH_P (cpu_features, MathVec_Prefer_No_AVX512)) { - if (CPU_FEATURES_ARCH_P (cpu_features, AVX512DQ_Usable)) + if (CPU_FEATURE_USABLE_P (cpu_features, AVX512DQ)) return OPTIMIZE (skx); - if (CPU_FEATURES_ARCH_P (cpu_features, AVX512F_Usable)) + if (CPU_FEATURE_USABLE_P (cpu_features, AVX512F)) return OPTIMIZE (knl); } diff --git a/sysdeps/x86_64/fpu/multiarch/ifunc-mathvec-sse4_1.h b/sysdeps/x86_64/fpu/multiarch/ifunc-mathvec-sse4_1.h index 63005c0..7240e55 100644 --- a/sysdeps/x86_64/fpu/multiarch/ifunc-mathvec-sse4_1.h +++ b/sysdeps/x86_64/fpu/multiarch/ifunc-mathvec-sse4_1.h @@ -31,7 +31,7 @@ IFUNC_SELECTOR (void) { const struct cpu_features* cpu_features = __get_cpu_features (); - if (CPU_FEATURES_CPU_P (cpu_features, SSE4_1)) + if (CPU_FEATURE_USABLE_P (cpu_features, SSE4_1)) return OPTIMIZE (sse4); return OPTIMIZE (sse2); diff --git a/sysdeps/x86_64/fpu/multiarch/ifunc-sse4_1.h b/sysdeps/x86_64/fpu/multiarch/ifunc-sse4_1.h index 7f26215..e5d8a6f 100644 --- a/sysdeps/x86_64/fpu/multiarch/ifunc-sse4_1.h +++ b/sysdeps/x86_64/fpu/multiarch/ifunc-sse4_1.h @@ -26,7 +26,7 @@ IFUNC_SELECTOR (void) { const struct cpu_features* cpu_features = __get_cpu_features (); - if (CPU_FEATURES_CPU_P (cpu_features, SSE4_1)) + if (CPU_FEATURE_USABLE_P (cpu_features, SSE4_1)) return OPTIMIZE (sse41); return OPTIMIZE (c); diff --git a/sysdeps/x86_64/fpu/multiarch/s_fma.c b/sysdeps/x86_64/fpu/multiarch/s_fma.c index 9992a1e..0d8c0b7 100644 --- a/sysdeps/x86_64/fpu/multiarch/s_fma.c +++ b/sysdeps/x86_64/fpu/multiarch/s_fma.c @@ -41,8 +41,8 @@ __fma_fma4 (double x, double y, double z) } -libm_ifunc (__fma, HAS_ARCH_FEATURE (FMA_Usable) - ? __fma_fma3 : (HAS_ARCH_FEATURE (FMA4_Usable) +libm_ifunc (__fma, CPU_FEATURE_USABLE (FMA) + ? __fma_fma3 : (CPU_FEATURE_USABLE (FMA4) ? __fma_fma4 : __fma_sse2)); libm_alias_double (__fma, fma) diff --git a/sysdeps/x86_64/fpu/multiarch/s_fmaf.c b/sysdeps/x86_64/fpu/multiarch/s_fmaf.c index 4cbcf1f..c01e5a2 100644 --- a/sysdeps/x86_64/fpu/multiarch/s_fmaf.c +++ b/sysdeps/x86_64/fpu/multiarch/s_fmaf.c @@ -40,8 +40,8 @@ __fmaf_fma4 (float x, float y, float z) } -libm_ifunc (__fmaf, HAS_ARCH_FEATURE (FMA_Usable) - ? __fmaf_fma3 : (HAS_ARCH_FEATURE (FMA4_Usable) +libm_ifunc (__fmaf, CPU_FEATURE_USABLE (FMA) + ? __fmaf_fma3 : (CPU_FEATURE_USABLE (FMA4) ? __fmaf_fma4 : __fmaf_sse2)); libm_alias_float (__fma, fma) |