From 4f0e90fae97a894247ec93336c8826cf4afb3d0d Mon Sep 17 00:00:00 2001 From: Hongtao Liu Date: Wed, 8 May 2019 10:21:40 +0000 Subject: Enable support for bfloat16 which will be in Future Cooper Lake. There are 3 instructions for AVX512BF16: VCVTNE2PS2BF16, VCVTNEPS2BF16 and DPBF16PS instructions, which are Vector Neural Network Instructions supporting: - VCVTNE2PS2BF16: Convert Two Packed Single Data to One Packed BF16 Data. - VCVTNEPS2BF16: Convert Packed Single Data to Packed BF16 Data. - VDPBF16PS: Dot Product of BF16 Pairs Accumulated into Packed Single Precision. 2019-05-07 Wei Xiao * common/config/i386/i386-common.c (OPTION_MASK_ISA_AVX512BF16_SET OPTION_MASK_ISA_AVX512BF16_UNSET, OPTION_MASK_ISA2_AVX512BW_UNSET): New. (OPTION_MASK_ISA2_AVX512F_UNSET): Add OPTION_MASK_ISA_AVX512BF16_UNSET. (ix86_handle_option): Handle -mavx512bf16. * config.gcc: Add avx512bf16vlintrin.h and avx512bf16intrin.h to extra_headers. * config/i386/avx512bf16vlintrin.h: New. * config/i386/avx512bf16intrin.h: New. * config/i386/cpuid.h (bit_AVX512BF16): New. * config/i386/driver-i386.c (host_detect_local_cpu): Detect BF16. * config/i386/i386-builtin-types.def: Add new types. * config/i386/i386-builtin.def: Add new builtins. * config/i386/i386-c.c (ix86_target_macros_internal): Define __AVX512BF16__. * config/i386/i386-option.c (ix86_target_string): Add -mavx512bf16. (ix86_option_override_internal): Handle BF16. (ix86_valid_target_attribute_inner_p): Ditto. * config/i386/i386-expand.c (ix86_expand_args_builtin): Ditto. * config/i386/i386-builtin.c (enum processor_features): Add F_AVX512BF16. (static const _isa_names_table isa_names_table): Ditto. * config/i386/i386.h (TARGET_AVX512BF16, TARGET_AVX512BF16_P): New. (PTA_AVX512BF16): Ditto. * config/i386/i386.opt: Add -mavx512bf16. * config/i386/immintrin.h: Include avx512bf16intrin.h and avx512bf16vlintrin.h. * config/i386/sse.md (avx512f_cvtne2ps2bf16_, avx512f_cvtneps2bf16_, avx512f_dpbf16ps_): New define_insn patterns. * config/i386/subst.md (mask_half): Add new subst. * doc/invoke.texi: Document -mavx512bf16. 2019-05-07 Wei Xiao * gcc.target/i386/avx512bf16-vcvtne2ps2bf16-1.c: New test. * gcc.target/i386/avx512bf16-vcvtneps2bf16-1.c: New test. * gcc.target/i386/avx512bf16-vdpbf16ps-1.c: New test. * gcc.target/i386/avx512bf16vl-vcvtne2ps2bf16-1.c: New test. * gcc.target/i386/avx512bf16vl-vcvtneps2bf16-1.c: New test. * gcc.target/i386/avx512bf16vl-vdpbf16ps-1.c: New test. * gcc.target/i386/builtin_target.c: Handle avx512bf16. * gcc.target/i386/sse-12.c: Add -mavx512bf16. * gcc.target/i386/sse-13.c: Ditto. * gcc.target/i386/sse-14.c: Ditto. * gcc.target/i386/sse-22.c: Ditto. * gcc.target/i386/sse-23.c: Ditto. * g++.dg/other/i386-2.C: Ditto. * g++.dg/other/i386-3.C: Ditto. 2019-05-07 Hongtao Liu * config/i386/cpuinfo.c (get_available_features): Detect BF16. * config/i386/cpuinfo.h (enum processor_features): Add FEATURE_AVX512BF16. From-SVN: r271006 --- libgcc/ChangeLog | 6 ++++++ libgcc/config/i386/cpuinfo.c | 6 +++++- libgcc/config/i386/cpuinfo.h | 3 ++- 3 files changed, 13 insertions(+), 2 deletions(-) (limited to 'libgcc') diff --git a/libgcc/ChangeLog b/libgcc/ChangeLog index 30e4da4..c3c7a16 100644 --- a/libgcc/ChangeLog +++ b/libgcc/ChangeLog @@ -1,3 +1,9 @@ +2019-05-07 Hongtao Liu + + * config/i386/cpuinfo.c (get_available_features): Detect BF16. + * config/i386/cpuinfo.h (enum processor_features): Add + FEATURE_AVX512BF16. + 2019-04-23 Ramana Radhakrishnan Bernd Edlinger Jakub Jelinek diff --git a/libgcc/config/i386/cpuinfo.c b/libgcc/config/i386/cpuinfo.c index d6cb2de..5659ec8 100644 --- a/libgcc/config/i386/cpuinfo.c +++ b/libgcc/config/i386/cpuinfo.c @@ -336,7 +336,7 @@ get_available_features (unsigned int ecx, unsigned int edx, set_feature (FEATURE_FMA); } - /* Get Advanced Features at level 7 (eax = 7, ecx = 0). */ + /* Get Advanced Features at level 7 (eax = 7, ecx = 0/1). */ if (max_cpuid_level >= 7) { __cpuid_count (7, 0, eax, ebx, ecx, edx); @@ -385,6 +385,10 @@ get_available_features (unsigned int ecx, unsigned int edx, set_feature (FEATURE_AVX5124VNNIW); if (edx & bit_AVX5124FMAPS) set_feature (FEATURE_AVX5124FMAPS); + + __cpuid_count (7, 1, eax, ebx, ecx, edx); + if (eax & bit_AVX512BF16) + set_feature (FEATURE_AVX512BF16); } } diff --git a/libgcc/config/i386/cpuinfo.h b/libgcc/config/i386/cpuinfo.h index b4492eb..68ca466 100644 --- a/libgcc/config/i386/cpuinfo.h +++ b/libgcc/config/i386/cpuinfo.h @@ -119,7 +119,8 @@ enum processor_features FEATURE_GFNI, FEATURE_VPCLMULQDQ, FEATURE_AVX512VNNI, - FEATURE_AVX512BITALG + FEATURE_AVX512BITALG, + FEATURE_AVX512BF16 }; extern struct __processor_model -- cgit v1.1