From 1fd8c163a83d96ace1ff78fa6bac7aee084f6f77 Mon Sep 17 00:00:00 2001 From: "H.J. Lu" Date: Fri, 5 Mar 2021 06:24:52 -0800 Subject: x86-64: Add ifunc-avx2.h functions with 256-bit EVEX Update ifunc-avx2.h, strchr.c, strcmp.c, strncmp.c and wcsnlen.c to select the function optimized with 256-bit EVEX instructions using YMM16-YMM31 registers to avoid RTM abort with usable AVX512VL, AVX512BW and BMI2 since VZEROUPPER isn't needed at function exit. For strcmp/strncmp, prefer AVX2 strcmp/strncmp if Prefer_AVX2_STRCMP is set. --- sysdeps/x86_64/multiarch/strncmp.c | 15 ++++++++++++--- 1 file changed, 12 insertions(+), 3 deletions(-) (limited to 'sysdeps/x86_64/multiarch/strncmp.c') diff --git a/sysdeps/x86_64/multiarch/strncmp.c b/sysdeps/x86_64/multiarch/strncmp.c index a565626aed..617c4e1065 100644 --- a/sysdeps/x86_64/multiarch/strncmp.c +++ b/sysdeps/x86_64/multiarch/strncmp.c @@ -30,16 +30,25 @@ extern __typeof (REDIRECT_NAME) OPTIMIZE (sse2) attribute_hidden; extern __typeof (REDIRECT_NAME) OPTIMIZE (ssse3) attribute_hidden; extern __typeof (REDIRECT_NAME) OPTIMIZE (sse42) attribute_hidden; extern __typeof (REDIRECT_NAME) OPTIMIZE (avx2) attribute_hidden; +extern __typeof (REDIRECT_NAME) OPTIMIZE (evex) attribute_hidden; static inline void * IFUNC_SELECTOR (void) { const struct cpu_features* cpu_features = __get_cpu_features (); - if (!CPU_FEATURES_ARCH_P (cpu_features, Prefer_No_VZEROUPPER) - && CPU_FEATURE_USABLE_P (cpu_features, AVX2) + if (CPU_FEATURE_USABLE_P (cpu_features, AVX2) && CPU_FEATURES_ARCH_P (cpu_features, AVX_Fast_Unaligned_Load)) - return OPTIMIZE (avx2); + { + if (CPU_FEATURE_USABLE_P (cpu_features, AVX512VL) + && CPU_FEATURE_USABLE_P (cpu_features, AVX512BW) + && CPU_FEATURE_USABLE_P (cpu_features, BMI2) + && !CPU_FEATURES_ARCH_P (cpu_features, Prefer_AVX2_STRCMP)) + return OPTIMIZE (evex); + + if (!CPU_FEATURES_ARCH_P (cpu_features, Prefer_No_VZEROUPPER)) + return OPTIMIZE (avx2); + } if (CPU_FEATURE_USABLE_P (cpu_features, SSE4_2) && !CPU_FEATURES_ARCH_P (cpu_features, Slow_SSE4_2)) -- cgit 1.4.1