From: "Jiang, Haochen" <haochen.jiang@intel.com>
To: Hongtao Liu <crazylht@gmail.com>,
"Kong, Lingling" <lingling.kong@intel.com>
Cc: "gcc-patches@gcc.gnu.org" <gcc-patches@gcc.gnu.org>,
"Liu, Hongtao" <hongtao.liu@intel.com>
Subject: RE: [PATCH 2/6] Support Intel AVX-VNNI-INT8
Date: Mon, 17 Oct 2022 06:27:18 +0000 [thread overview]
Message-ID: <SA1PR11MB5946440B1D7DFF0C8F9C5416EC299@SA1PR11MB5946.namprd11.prod.outlook.com> (raw)
In-Reply-To: <CAMZc-byz7V47jJGZOU7vk=njZhcK5noaLD2T8vYpAn+7zM-8QA@mail.gmail.com>
> -----Original Message-----
> From: Hongtao Liu <crazylht@gmail.com>
> Sent: Monday, October 17, 2022 12:05 PM
> To: Jiang, Haochen <haochen.jiang@intel.com>
> Cc: gcc-patches@gcc.gnu.org; Liu, Hongtao <hongtao.liu@intel.com>
> Subject: Re: [PATCH 2/6] Support Intel AVX-VNNI-INT8
>
> On Fri, Oct 14, 2022 at 3:57 PM Haochen Jiang via Gcc-patches <gcc-
> patches@gcc.gnu.org> wrote:
> >
> > From: Kong Lingling <lingling.kong@intel.com>
> >
> > gcc/ChangeLog
> >
> > * common/config/i386/cpuinfo.h (get_available_features): Detect
> > avxvnniint8.
> > * common/config/i386/i386-common.cc
> > (OPTION_MASK_ISA2_AVXVNNIINT8_SET): New.
> > (OPTION_MASK_ISA2_AVXVNNIINT8_UNSET): Ditto.
> > (ix86_handle_option): Handle -mavxvnniint8.
> > * common/config/i386/i386-cpuinfo.h (enum processor_features):
> > Add FEATURE_AVXVNNIINT8.
> > * common/config/i386/i386-isas.h: Add ISA_NAME_TABLE_ENTRY for
> > avxvnniint8.
> > * config.gcc: Add avxvnniint8intrin.h.
> > * config/i386/avxvnniint8intrin.h: New file.
> > * config/i386/cpuid.h (bit_AVXVNNIINT8): New.
> > * config/i386/i386-builtin.def: Add new builtins.
> > * config/i386/i386-c.cc (ix86_target_macros_internal): Define
> > __AVXVNNIINT8__.
> > * config/i386/i386-options.cc (isa2_opts): Add -mavxvnniint8.
> > (ix86_valid_target_attribute_inner_p): Handle avxvnniint8.
> > * config/i386/i386-isa.def: Add DEF_PTA(AVXVNNIINT8) New..
> > * config/i386/i386.opt: Add option -mavxvnniint8.
> > * config/i386/immintrin.h: Include avxvnniint8intrin.h.
> > * config/i386/sse.md
> > (vpdp<vpdotprodtype>_<mode>): New define_insn.
> > * doc/extend.texi: Document avxvnniint8.
> > * doc/invoke.texi: Document -mavxvnniint8.
> > * doc/sourcebuild.texi: Document target avxvnniint8.
> >
> > gcc/testsuite/ChangeLog
> >
> > * g++.dg/other/i386-2.C: Add -mavxvnniint8.
> > * g++.dg/other/i386-3.C: Ditto.
> > * gcc.target/i386/avx-check.h: Add avxvnniint8 check.
> > * gcc.target/i386/sse-12.c: Add -mavxvnniint8.
> > * 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.
> > * gcc.target/i386/funcspec-56.inc: Add new target attribute.
> > * lib/target-supports.exp
> > (check_effective_target_avxvnniint8): New.
> > * gcc.target/i386/avxvnniint8-1.c: Ditto.
> > * gcc.target/i386/avxvnniint8-vpdpbssd-2.c: Ditto.
> > * gcc.target/i386/avxvnniint8-vpdpbssds-2.c: Ditto.
> > * gcc.target/i386/avxvnniint8-vpdpbsud-2.c: Ditto.
> > * gcc.target/i386/avxvnniint8-vpdpbsuds-2.c: Ditto.
> > * gcc.target/i386/avxvnniint8-vpdpbuud-2.c: Ditto.
> > * gcc.target/i386/avxvnniint8-vpdpbuuds-2.c: Ditto.
> >
> > Co-authored-by: Hongyu Wang <hongyu.wang@intel.com>
> > Co-authored-by: Haochen Jiang <haochen.jiang@intel.com>
> > ---
> > gcc/common/config/i386/cpuinfo.h | 2 +
> > gcc/common/config/i386/i386-common.cc | 22 ++-
> > gcc/common/config/i386/i386-cpuinfo.h | 1 +
> > gcc/common/config/i386/i386-isas.h | 2 +
> > gcc/config.gcc | 2 +-
> > gcc/config/i386/avxvnniint8intrin.h | 138 ++++++++++++++++++
> > gcc/config/i386/cpuid.h | 1 +
> > gcc/config/i386/i386-builtin.def | 14 ++
> > gcc/config/i386/i386-c.cc | 2 +
> > gcc/config/i386/i386-isa.def | 1 +
> > gcc/config/i386/i386-options.cc | 4 +-
> > gcc/config/i386/i386.opt | 5 +
> > gcc/config/i386/immintrin.h | 2 +
> > gcc/config/i386/sse.md | 31 ++++
> > gcc/doc/extend.texi | 5 +
> > gcc/doc/invoke.texi | 9 +-
> > gcc/doc/sourcebuild.texi | 3 +
> > gcc/testsuite/g++.dg/other/i386-2.C | 2 +-
> > gcc/testsuite/g++.dg/other/i386-3.C | 2 +-
> > gcc/testsuite/gcc.target/i386/avx-check.h | 3 +
> > gcc/testsuite/gcc.target/i386/avxvnniint8-1.c | 43 ++++++
> > .../gcc.target/i386/avxvnniint8-vpdpbssd-2.c | 72 +++++++++
> > .../gcc.target/i386/avxvnniint8-vpdpbssds-2.c | 72 +++++++++
> > .../gcc.target/i386/avxvnniint8-vpdpbsud-2.c | 72 +++++++++
> > .../gcc.target/i386/avxvnniint8-vpdpbsuds-2.c | 72 +++++++++
> > .../gcc.target/i386/avxvnniint8-vpdpbuud-2.c | 72 +++++++++
> > .../gcc.target/i386/avxvnniint8-vpdpbuuds-2.c | 72 +++++++++
> > gcc/testsuite/gcc.target/i386/funcspec-56.inc | 2 +
> > gcc/testsuite/gcc.target/i386/sse-12.c | 2 +-
> > gcc/testsuite/gcc.target/i386/sse-13.c | 2 +-
> > gcc/testsuite/gcc.target/i386/sse-14.c | 2 +-
> > gcc/testsuite/gcc.target/i386/sse-22.c | 4 +-
> > gcc/testsuite/gcc.target/i386/sse-23.c | 2 +-
> > gcc/testsuite/lib/target-supports.exp | 12 ++
> > 34 files changed, 738 insertions(+), 14 deletions(-) create mode
> > 100644 gcc/config/i386/avxvnniint8intrin.h
> > create mode 100644 gcc/testsuite/gcc.target/i386/avxvnniint8-1.c
> > create mode 100644
> > gcc/testsuite/gcc.target/i386/avxvnniint8-vpdpbssd-2.c
> > create mode 100644
> > gcc/testsuite/gcc.target/i386/avxvnniint8-vpdpbssds-2.c
> > create mode 100644
> > gcc/testsuite/gcc.target/i386/avxvnniint8-vpdpbsud-2.c
> > create mode 100644
> > gcc/testsuite/gcc.target/i386/avxvnniint8-vpdpbsuds-2.c
> > create mode 100644
> > gcc/testsuite/gcc.target/i386/avxvnniint8-vpdpbuud-2.c
> > create mode 100644
> > gcc/testsuite/gcc.target/i386/avxvnniint8-vpdpbuuds-2.c
> >
> > diff --git a/gcc/common/config/i386/cpuinfo.h
> > b/gcc/common/config/i386/cpuinfo.h
> > index 9bb21c6cacc..bed88003f8e 100644
> > --- a/gcc/common/config/i386/cpuinfo.h
> > +++ b/gcc/common/config/i386/cpuinfo.h
> > @@ -795,6 +795,8 @@ get_available_features (struct __processor_model
> *cpu_model,
> > set_feature (FEATURE_AVXVNNI);
> > if (eax & bit_AVXIFMA)
> > set_feature (FEATURE_AVXIFMA);
> > + if (edx & bit_AVXVNNIINT8)
> > + set_feature (FEATURE_AVXVNNIINT8);
> > }
> > if (avx512_usable)
> > {
> > diff --git a/gcc/common/config/i386/i386-common.cc
> > b/gcc/common/config/i386/i386-common.cc
> > index 4de7906b247..6a2a7e3d25a 100644
> > --- a/gcc/common/config/i386/i386-common.cc
> > +++ b/gcc/common/config/i386/i386-common.cc
> > @@ -108,6 +108,7 @@ along with GCC; see the file COPYING3. If not see
> > #define OPTION_MASK_ISA2_AMX_TILE_SET
> OPTION_MASK_ISA2_AMX_TILE
> > #define OPTION_MASK_ISA2_AMX_INT8_SET
> OPTION_MASK_ISA2_AMX_INT8
> > #define OPTION_MASK_ISA2_AMX_BF16_SET
> OPTION_MASK_ISA2_AMX_BF16
> > +#define OPTION_MASK_ISA2_AVXVNNIINT8_SET
> OPTION_MASK_ISA2_AVXVNNIINT8
> >
> > /* SSE4 includes both SSE4.1 and SSE4.2. -msse4 should be the same
> > as -msse4.2. */
> > @@ -214,7 +215,7 @@ along with GCC; see the file COPYING3. If not see
> > (OPTION_MASK_ISA_AVX2 | OPTION_MASK_ISA_AVX512F_UNSET)
> #define
> > OPTION_MASK_ISA2_AVX2_UNSET \
> > (OPTION_MASK_ISA2_AVXIFMA_UNSET |
> OPTION_MASK_ISA2_AVXVNNI_UNSET \
> > - | OPTION_MASK_ISA2_AVX512F_UNSET)
> > + | OPTION_MASK_ISA2_AVXVNNIINT8_UNSET |
> > + OPTION_MASK_ISA2_AVX512F_UNSET)
> > #define OPTION_MASK_ISA_AVX512F_UNSET \
> > (OPTION_MASK_ISA_AVX512F | OPTION_MASK_ISA_AVX512CD_UNSET
> \
> > | OPTION_MASK_ISA_AVX512PF_UNSET |
> OPTION_MASK_ISA_AVX512ER_UNSET
> > \ @@ -278,6 +279,7 @@ along with GCC; see the file COPYING3. If not
> > see #define OPTION_MASK_ISA2_KL_UNSET \
> > (OPTION_MASK_ISA2_KL | OPTION_MASK_ISA2_WIDEKL_UNSET)
> #define
> > OPTION_MASK_ISA2_WIDEKL_UNSET OPTION_MASK_ISA2_WIDEKL
> > +#define OPTION_MASK_ISA2_AVXVNNIINT8_UNSET
> > +OPTION_MASK_ISA2_AVXVNNIINT8
> >
> > /* SSE4 includes both SSE4.1 and SSE4.2. -mno-sse4 should the same
> > as -mno-sse4.1. */
> > @@ -1142,6 +1144,24 @@ ix86_handle_option (struct gcc_options *opts,
> > }
> > return true;
> >
> > + case OPT_mavxvnniint8:
> > + if (value)
> > + {
> > + opts->x_ix86_isa_flags2 |= OPTION_MASK_ISA2_AVXVNNIINT8_SET;
> > + opts->x_ix86_isa_flags2_explicit |=
> > + OPTION_MASK_ISA2_AVXVNNIINT8_SET;
> > + opts->x_ix86_isa_flags |= OPTION_MASK_ISA_AVX2_SET;
> > + opts->x_ix86_isa_flags_explicit |= OPTION_MASK_ISA_AVX2_SET;
> > + }
> > + else
> > + {
> > + opts->x_ix86_isa_flags2 &=
> > + ~OPTION_MASK_ISA2_AVXVNNIINT8_UNSET;
> > + opts->x_ix86_isa_flags2_explicit |=
> > + OPTION_MASK_ISA2_AVXVNNIINT8_UNSET;
> > + }
> > + return true;
> > +
> > case OPT_mfma:
> > if (value)
> > {
> > diff --git a/gcc/common/config/i386/i386-cpuinfo.h
> > b/gcc/common/config/i386/i386-cpuinfo.h
> > index 968f9a56a6c..9a6b92fab79 100644
> > --- a/gcc/common/config/i386/i386-cpuinfo.h
> > +++ b/gcc/common/config/i386/i386-cpuinfo.h
> > @@ -241,6 +241,7 @@ enum processor_features
> > FEATURE_X86_64_V3,
> > FEATURE_X86_64_V4,
> > FEATURE_AVXIFMA,
> > + FEATURE_AVXVNNIINT8,
> > CPU_FEATURE_MAX
> > };
> >
> > diff --git a/gcc/common/config/i386/i386-isas.h
> > b/gcc/common/config/i386/i386-isas.h
> > index b05b4bb8f0d..8c1f351056c 100644
> > --- a/gcc/common/config/i386/i386-isas.h
> > +++ b/gcc/common/config/i386/i386-isas.h
> > @@ -176,4 +176,6 @@ ISA_NAMES_TABLE_START
> > ISA_NAMES_TABLE_ENTRY("x86-64-v3", FEATURE_X86_64_V3,
> P_X86_64_V3, NULL)
> > ISA_NAMES_TABLE_ENTRY("x86-64-v4", FEATURE_X86_64_V4,
> P_X86_64_V4, NULL)
> > ISA_NAMES_TABLE_ENTRY("avxifma", FEATURE_AVXIFMA, P_NONE,
> > "-mavxifma")
> > + ISA_NAMES_TABLE_ENTRY("avxvnniint8", FEATURE_AVXVNNIINT8,
> > + P_NONE, "-mavxvnniint8")
> > ISA_NAMES_TABLE_END
> > diff --git a/gcc/config.gcc b/gcc/config.gcc index
> > 12365abbf86..4df78238910 100644
> > --- a/gcc/config.gcc
> > +++ b/gcc/config.gcc
> > @@ -422,7 +422,7 @@ i[34567]86-*-* | x86_64-*-*)
> > amxbf16intrin.h x86gprintrin.h uintrintrin.h
> > hresetintrin.h keylockerintrin.h avxvnniintrin.h
> > mwaitintrin.h avx512fp16intrin.h avx512fp16vlintrin.h
> > - avxifmaintrin.h"
> > + avxifmaintrin.h avxvnniint8intrin.h"
> > ;;
> > ia64-*-*)
> > extra_headers=ia64intrin.h
> > diff --git a/gcc/config/i386/avxvnniint8intrin.h
> > b/gcc/config/i386/avxvnniint8intrin.h
> > new file mode 100644
> > index 00000000000..362e6f65c2a
> > --- /dev/null
> > +++ b/gcc/config/i386/avxvnniint8intrin.h
> > @@ -0,0 +1,138 @@
> > +/* Copyright (C) 2020 Free Software Foundation, Inc.
> > +
> > + This file is part of GCC.
> > +
> > + GCC is free software; you can redistribute it and/or modify
> > + it under the terms of the GNU General Public License as published by
> > + the Free Software Foundation; either version 3, or (at your option)
> > + any later version.
> > +
> > + GCC is distributed in the hope that it will be useful,
> > + but WITHOUT ANY WARRANTY; without even the implied warranty of
> > + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> > + GNU General Public License for more details.
> > +
> > + Under Section 7 of GPL version 3, you are granted additional
> > + permissions described in the GCC Runtime Library Exception, version
> > + 3.1, as published by the Free Software Foundation.
> > +
> > + You should have received a copy of the GNU General Public License and
> > + a copy of the GCC Runtime Library Exception along with this program;
> > + see the files COPYING3 and COPYING.RUNTIME respectively. If not, see
> > + <http://www.gnu.org/licenses/>. */
> > +
> > +#if !defined _IMMINTRIN_H_INCLUDED
> > +#error "Never use <avxvnniint8vlintrin.h> directly; include <immintrin.h>
> instead."
> > +#endif
> > +
> > +#ifndef _AVXVNNIINT8INTRIN_H_INCLUDED #define
> > +_AVXVNNIINT8INTRIN_H_INCLUDED
> > +
> > +#if !defined(__AVXVNNIINT8__)
> > +#pragma GCC push_options
> > +#pragma GCC target("avxvnniint8")
> > +#define __DISABLE_AVXVNNIINT8__
> > +#endif /* __AVXVNNIINT8__ */
> > +
> > +extern __inline __m128i
> > +__attribute__((__gnu_inline__, __always_inline__, __artificial__))
> > +_mm_dpbssd_epi32 (__m128i __W, __m128i __A, __m128i __B) {
> > + return (__m128i)
> > + __builtin_ia32_vpdpbssd128 ((__v4si) __W, (__v4si) __A, (__v4si)
> > +__B); }
> > +
> > +extern __inline __m128i
> > +__attribute__((__gnu_inline__, __always_inline__, __artificial__))
> > +_mm_dpbssds_epi32 (__m128i __W, __m128i __A, __m128i __B) {
> > + return (__m128i)
> > + __builtin_ia32_vpdpbssds128 ((__v4si) __W, (__v4si) __A, (__v4si)
> > +__B); }
> > +
> > +extern __inline __m128i
> > +__attribute__((__gnu_inline__, __always_inline__, __artificial__))
> > +_mm_dpbsud_epi32 (__m128i __W, __m128i __A, __m128i __B) {
> > + return (__m128i)
> > + __builtin_ia32_vpdpbsud128 ((__v4si) __W, (__v4si) __A, (__v4si)
> > +__B); }
> > +
> > +extern __inline __m128i
> > +__attribute__((__gnu_inline__, __always_inline__, __artificial__))
> > +_mm_dpbsuds_epi32 (__m128i __W, __m128i __A, __m128i __B) {
> > + return (__m128i)
> > + __builtin_ia32_vpdpbsuds128 ((__v4si) __W, (__v4si) __A, (__v4si)
> > +__B); }
> > +
> > +extern __inline __m128i
> > +__attribute__((__gnu_inline__, __always_inline__, __artificial__))
> > +_mm_dpbuud_epi32 (__m128i __W, __m128i __A, __m128i __B) {
> > + return (__m128i)
> > + __builtin_ia32_vpdpbuud128 ((__v4si) __W, (__v4si) __A, (__v4si)
> > +__B); }
> > +
> > +extern __inline __m128i
> > +__attribute__((__gnu_inline__, __always_inline__, __artificial__))
> > +_mm_dpbuuds_epi32 (__m128i __W, __m128i __A, __m128i __B) {
> > + return (__m128i)
> > + __builtin_ia32_vpdpbuuds128 ((__v4si) __W, (__v4si) __A, (__v4si)
> > +__B); }
> > +
> > +extern __inline __m256i
> > +__attribute__((__gnu_inline__, __always_inline__, __artificial__))
> > +_mm256_dpbssd_epi32 (__m256i __W, __m256i __A, __m256i __B) {
> > + return (__m256i)
> > + __builtin_ia32_vpdpbssd256 ((__v8si) __W, (__v8si) __A, (__v8si)
> > +__B); }
> > +
> > +extern __inline __m256i
> > +__attribute__((__gnu_inline__, __always_inline__, __artificial__))
> > +_mm256_dpbssds_epi32 (__m256i __W, __m256i __A, __m256i __B) {
> > + return (__m256i)
> > + __builtin_ia32_vpdpbssds256 ((__v8si) __W, (__v8si) __A, (__v8si)
> > +__B); }
> > +
> > +extern __inline __m256i
> > +__attribute__((__gnu_inline__, __always_inline__, __artificial__))
> > +_mm256_dpbsud_epi32 (__m256i __W, __m256i __A, __m256i __B) {
> > + return (__m256i)
> > + __builtin_ia32_vpdpbsud256 ((__v8si) __W, (__v8si) __A, (__v8si)
> > +__B); }
> > +
> > +extern __inline __m256i
> > +__attribute__((__gnu_inline__, __always_inline__, __artificial__))
> > +_mm256_dpbsuds_epi32 (__m256i __W, __m256i __A, __m256i __B) {
> > + return (__m256i)
> > + __builtin_ia32_vpdpbsuds256 ((__v8si) __W, (__v8si) __A, (__v8si)
> > +__B); }
> > +
> > +extern __inline __m256i
> > +__attribute__((__gnu_inline__, __always_inline__, __artificial__))
> > +_mm256_dpbuud_epi32 (__m256i __W, __m256i __A, __m256i __B) {
> > + return (__m256i)
> > + __builtin_ia32_vpdpbuud256 ((__v8si) __W, (__v8si) __A, (__v8si)
> > +__B); }
> > +
> > +extern __inline __m256i
> > +__attribute__((__gnu_inline__, __always_inline__, __artificial__))
> > +_mm256_dpbuuds_epi32 (__m256i __W, __m256i __A, __m256i __B) {
> > + return (__m256i)
> > + __builtin_ia32_vpdpbuuds256 ((__v8si) __W, (__v8si) __A, (__v8si)
> > +__B); }
> > +
> > +#ifdef __DISABLE_AVXVNNIINT8__
> > +#undef __DISABLE_AVXVNNIINT8__
> > +#pragma GCC pop_options
> > +#endif /* __DISABLE_AVXVNNIINT8__ */
> > +
> > +#endif /* __AVXVNNIINT8INTRIN_H_INCLUDED */
> > diff --git a/gcc/config/i386/cpuid.h b/gcc/config/i386/cpuid.h index
> > 9885699efd5..f5fad22149a 100644
> > --- a/gcc/config/i386/cpuid.h
> > +++ b/gcc/config/i386/cpuid.h
> > @@ -49,6 +49,7 @@
> > #define bit_RDRND (1 << 30)
> >
> > /* %edx */
> > +#define bit_AVXVNNIINT8 (1 << 4)
> > #define bit_CMPXCHG8B (1 << 8)
> > #define bit_CMOV (1 << 15)
> > #define bit_MMX (1 << 23)
> > diff --git a/gcc/config/i386/i386-builtin.def
> > b/gcc/config/i386/i386-builtin.def
> > index 4a89099a00f..e6edae5728b 100644
> > --- a/gcc/config/i386/i386-builtin.def
> > +++ b/gcc/config/i386/i386-builtin.def
> > @@ -2696,6 +2696,20 @@ BDESC (OPTION_MASK_ISA_AVX512VNNI |
> > OPTION_MASK_ISA_AVX512VL, OPTION_MASK_ISA2_A BDESC
> > (OPTION_MASK_ISA_AVX512VNNI | OPTION_MASK_ISA_AVX512VL, 0,
> > CODE_FOR_vpdpwssds_v4si_mask,
> "__builtin_ia32_vpdpwssds_v4si_mask",
> > IX86_BUILTIN_VPDPWSSDSV4SI_MASK, UNKNOWN, (int)
> > V4SI_FTYPE_V4SI_V4SI_V4SI_UQI) BDESC
> (OPTION_MASK_ISA_AVX512VNNI |
> > OPTION_MASK_ISA_AVX512VL, 0, CODE_FOR_vpdpwssds_v4si_maskz,
> > "__builtin_ia32_vpdpwssds_v4si_maskz",
> > IX86_BUILTIN_VPDPWSSDSV4SI_MASKZ, UNKNOWN, (int)
> > V4SI_FTYPE_V4SI_V4SI_V4SI_UQI)
> >
> > +/* AVXVNNIINT8 */
> > +BDESC (0, OPTION_MASK_ISA2_AVXVNNIINT8,
> CODE_FOR_vpdpbssd_v8si,
> > +"__builtin_ia32_vpdpbssd256", IX86_BUILTIN_VPDPBSSDV8SI,
> UNKNOWN,
> > +(int) V8SI_FTYPE_V8SI_V8SI_V8SI) BDESC (0,
> > +OPTION_MASK_ISA2_AVXVNNIINT8, CODE_FOR_vpdpbssds_v8si,
> > +"__builtin_ia32_vpdpbssds256", IX86_BUILTIN_VPDPBSSDSV8SI,
> UNKNOWN,
> > +(int) V8SI_FTYPE_V8SI_V8SI_V8SI) BDESC (0,
> > +OPTION_MASK_ISA2_AVXVNNIINT8, CODE_FOR_vpdpbsud_v8si,
> > +"__builtin_ia32_vpdpbsud256", IX86_BUILTIN_VPDPBSUDV8SI,
> UNKNOWN,
> > +(int) V8SI_FTYPE_V8SI_V8SI_V8SI) BDESC (0,
> > +OPTION_MASK_ISA2_AVXVNNIINT8, CODE_FOR_vpdpbsuds_v8si,
> > +"__builtin_ia32_vpdpbsuds256", IX86_BUILTIN_VPDPBSUDSV8SI,
> UNKNOWN,
> > +(int) V8SI_FTYPE_V8SI_V8SI_V8SI) BDESC (0,
> > +OPTION_MASK_ISA2_AVXVNNIINT8, CODE_FOR_vpdpbuud_v8si,
> > +"__builtin_ia32_vpdpbuud256", IX86_BUILTIN_VPDPBUUDV8SI,
> UNKNOWN,
> > +(int) V8SI_FTYPE_V8SI_V8SI_V8SI) BDESC (0,
> > +OPTION_MASK_ISA2_AVXVNNIINT8, CODE_FOR_vpdpbuuds_v8si,
> > +"__builtin_ia32_vpdpbuuds256", IX86_BUILTIN_VPDPBUUDSV8SI,
> UNKNOWN,
> > +(int) V8SI_FTYPE_V8SI_V8SI_V8SI) BDESC (0,
> > +OPTION_MASK_ISA2_AVXVNNIINT8, CODE_FOR_vpdpbssd_v4si,
> > +"__builtin_ia32_vpdpbssd128", IX86_BUILTIN_VPDPBSSDV4SI,
> UNKNOWN,
> > +(int) V4SI_FTYPE_V4SI_V4SI_V4SI) BDESC (0,
> > +OPTION_MASK_ISA2_AVXVNNIINT8, CODE_FOR_vpdpbssds_v4si,
> > +"__builtin_ia32_vpdpbssds128", IX86_BUILTIN_VPDPBSSDSV4SI,
> UNKNOWN,
> > +(int) V4SI_FTYPE_V4SI_V4SI_V4SI) BDESC (0,
> > +OPTION_MASK_ISA2_AVXVNNIINT8, CODE_FOR_vpdpbsud_v4si,
> > +"__builtin_ia32_vpdpbsud128", IX86_BUILTIN_VPDPBSUDV4SI,
> UNKNOWN,
> > +(int) V4SI_FTYPE_V4SI_V4SI_V4SI) BDESC (0,
> > +OPTION_MASK_ISA2_AVXVNNIINT8, CODE_FOR_vpdpbsuds_v4si,
> > +"__builtin_ia32_vpdpbsuds128", IX86_BUILTIN_VPDPBSUDSV4SI,
> UNKNOWN,
> > +(int) V4SI_FTYPE_V4SI_V4SI_V4SI) BDESC (0,
> > +OPTION_MASK_ISA2_AVXVNNIINT8, CODE_FOR_vpdpbuud_v4si,
> > +"__builtin_ia32_vpdpbuud128", IX86_BUILTIN_VPDPBUUDV4SI,
> UNKNOWN,
> > +(int) V4SI_FTYPE_V4SI_V4SI_V4SI) BDESC (0,
> > +OPTION_MASK_ISA2_AVXVNNIINT8, CODE_FOR_vpdpbuuds_v4si,
> > +"__builtin_ia32_vpdpbuuds128", IX86_BUILTIN_VPDPBUUDSV4SI,
> UNKNOWN,
> > +(int) V4SI_FTYPE_V4SI_V4SI_V4SI)
> > +
> > /* VPCLMULQDQ */
> > BDESC (OPTION_MASK_ISA_VPCLMULQDQ |
> OPTION_MASK_ISA_AVX512VL, 0,
> > CODE_FOR_vpclmulqdq_v2di, "__builtin_ia32_vpclmulqdq_v2di",
> > IX86_BUILTIN_VPCLMULQDQ2, UNKNOWN, (int)
> V2DI_FTYPE_V2DI_V2DI_INT)
> > BDESC (OPTION_MASK_ISA_VPCLMULQDQ | OPTION_MASK_ISA_AVX, 0,
> > CODE_FOR_vpclmulqdq_v4di, "__builtin_ia32_vpclmulqdq_v4di",
> > IX86_BUILTIN_VPCLMULQDQ4, UNKNOWN, (int)
> V4DI_FTYPE_V4DI_V4DI_INT)
> > diff --git a/gcc/config/i386/i386-c.cc b/gcc/config/i386/i386-c.cc
> > index 3494ec035d5..a9a35c0a18a 100644
> > --- a/gcc/config/i386/i386-c.cc
> > +++ b/gcc/config/i386/i386-c.cc
> > @@ -635,6 +635,8 @@ ix86_target_macros_internal (HOST_WIDE_INT
> isa_flag,
> > def_or_undef (parse_in, "__AVXVNNI__");
> > if (isa_flag2 & OPTION_MASK_ISA2_AVXIFMA)
> > def_or_undef (parse_in, "__AVXIFMA__");
> > + if (isa_flag2 & OPTION_MASK_ISA2_AVXVNNIINT8)
> > + def_or_undef (parse_in, "__AVXVNNIINT8__");
> > if (TARGET_IAMCU)
> > {
> > def_or_undef (parse_in, "__iamcu"); diff --git
> > a/gcc/config/i386/i386-isa.def b/gcc/config/i386/i386-isa.def index
> > 6e0254ce418..c95b917c6ce 100644
> > --- a/gcc/config/i386/i386-isa.def
> > +++ b/gcc/config/i386/i386-isa.def
> > @@ -110,3 +110,4 @@ DEF_PTA(WIDEKL)
> > DEF_PTA(AVXVNNI)
> > DEF_PTA(AVX512FP16)
> > DEF_PTA(AVXIFMA)
> > +DEF_PTA(AVXVNNIINT8)
> > diff --git a/gcc/config/i386/i386-options.cc
> > b/gcc/config/i386/i386-options.cc index 5facb64c2a8..3e6d04433a6
> > 100644
> > --- a/gcc/config/i386/i386-options.cc
> > +++ b/gcc/config/i386/i386-options.cc
> > @@ -227,7 +227,8 @@ static struct ix86_target_opts isa2_opts[] =
> > { "-mwidekl", OPTION_MASK_ISA2_WIDEKL },
> > { "-mavxvnni", OPTION_MASK_ISA2_AVXVNNI },
> > { "-mavx512fp16", OPTION_MASK_ISA2_AVX512FP16 },
> > - { "-mavxifma", OPTION_MASK_ISA2_AVXIFMA }
> > + { "-mavxifma", OPTION_MASK_ISA2_AVXIFMA },
> > + { "-mavxvnniint8", OPTION_MASK_ISA2_AVXVNNIINT8 }
> > };
> > static struct ix86_target_opts isa_opts[] = { @@ -1074,6 +1075,7 @@
> > ix86_valid_target_attribute_inner_p (tree fndecl, tree args, char
> *p_strings[],
> > IX86_ATTR_ISA ("avxvnni", OPT_mavxvnni),
> > IX86_ATTR_ISA ("avx512fp16", OPT_mavx512fp16),
> > IX86_ATTR_ISA ("avxifma", OPT_mavxifma),
> > + IX86_ATTR_ISA ("avxvnniint8", OPT_mavxvnniint8),
> >
> > /* enum options */
> > IX86_ATTR_ENUM ("fpmath=", OPT_mfpmath_), diff --git
> > a/gcc/config/i386/i386.opt b/gcc/config/i386/i386.opt index
> > 36e28b7063d..53d534f6392 100644
> > --- a/gcc/config/i386/i386.opt
> > +++ b/gcc/config/i386/i386.opt
> > @@ -1219,3 +1219,8 @@ mavxifma
> > Target Mask(ISA2_AVXIFMA) Var(ix86_isa_flags2) Save Support MMX,
> > SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2, AVX, AVX2, and AVXIFMA
> > built-in functions and code generation.
> > +
> > +mavxvnniint8
> > +Target Mask(ISA2_AVXVNNIINT8) Var(ix86_isa_flags2) Save Support
> MMX,
> > +SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2, AVX, AVX2 and
> > +AVXVNNIINT8 built-in functions and code generation.
> > diff --git a/gcc/config/i386/immintrin.h b/gcc/config/i386/immintrin.h
> > index e9d4e975243..ddea249d09b 100644
> > --- a/gcc/config/i386/immintrin.h
> > +++ b/gcc/config/i386/immintrin.h
> > @@ -46,6 +46,8 @@
> >
> > #include <avxifmaintrin.h>
> >
> > +#include <avxvnniint8intrin.h>
> > +
> > #include <avx2intrin.h>
> >
> > #include <avx512fintrin.h>
> > diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md index
> > 331347569ea..49490a213ea 100644
> > --- a/gcc/config/i386/sse.md
> > +++ b/gcc/config/i386/sse.md
> > @@ -200,6 +200,13 @@
> > UNSPEC_COMPLEX_FCMUL
> > UNSPEC_COMPLEX_MASK
> >
> > + ;; For AVX-VNNI-INT8 support
> > + UNSPEC_VPDPBSSD
> > + UNSPEC_VPDPBSSDS
> > + UNSPEC_VPDPBSUD
> > + UNSPEC_VPDPBSUDS
> > + UNSPEC_VPDPBUUD
> > + UNSPEC_VPDPBUUDS
> > ])
> Can we have unified naming conversion for the UNSPECs and those in
> AVX512(AVX)_VNNI, since they have similar semantics.
> Maybe change the old UNSPEC_VPMADDUBSWACCD to the "new-style"
> UNSPEC_VPDPBUSD? Similar for other UNSPEC_VPMADD***.
I suppose we can do that in this patch to avoid confusion.
> >
> > (define_c_enum "unspecv" [
> > @@ -29241,3 +29248,27 @@
> > gcc_unreachable ();
> > DONE;
> > })
> > +
> > +(define_int_iterator VPDOTPROD
> > + [UNSPEC_VPDPBSSD
> > + UNSPEC_VPDPBSSDS
> > + UNSPEC_VPDPBSUD
> > + UNSPEC_VPDPBSUDS
> > + UNSPEC_VPDPBUUD
> > + UNSPEC_VPDPBUUDS])
> > +
> > +(define_int_attr vpdotprodtype
> > + [(UNSPEC_VPDPBSSD "bssd") (UNSPEC_VPDPBSSDS "bssds")
> > + (UNSPEC_VPDPBSUD "bsud") (UNSPEC_VPDPBSUDS "bsuds")
> > + (UNSPEC_VPDPBUUD "buud") (UNSPEC_VPDPBUUDS "buuds")])
> > +
> > +(define_insn "vpdp<vpdotprodtype>_<mode>"
> > + [(set (match_operand:VI4_AVX 0 "register_operand" "=x")
> > + (unspec:VI4_AVX
> > + [(match_operand:VI4_AVX 1 "register_operand" "0")
> > + (match_operand:VI4_AVX 2 "register_operand" "x")
> > + (match_operand:VI4_AVX 3 "nonimmediate_operand" "xm")]
> > + VPDOTPROD))]
> > + "TARGET_AVXVNNIINT8"
> > + "vpdp<vpdotprodtype>\t{%3, %2, %0|%0, %2, %3}"
> > + [(set_attr "prefix" "vex")])
> > diff --git a/gcc/doc/extend.texi b/gcc/doc/extend.texi index
> > edecf5c0070..9a8de9fc226 100644
> > --- a/gcc/doc/extend.texi
> > +++ b/gcc/doc/extend.texi
> > @@ -7065,6 +7065,11 @@ Enable/disable the generation of the AVXVNNI
> instructions.
> > @cindex @code{target("avxifma")} function attribute, x86
> > Enable/disable the generation of the AVXIFMA instructions.
> >
> > +@item avxvnniint8
> > +@itemx no-avxvnniint8
> > +@cindex @code{target("avxvnniint8")} function attribute, x86
> > +Enable/disable the generation of the AVXVNNIINT8 instructions.
> > +
> > @item cld
> > @itemx no-cld
> > @cindex @code{target("cld")} function attribute, x86 diff --git
> > a/gcc/doc/invoke.texi b/gcc/doc/invoke.texi index
> > 886fc1d0164..d4ff7549bf3 100644
> > --- a/gcc/doc/invoke.texi
> > +++ b/gcc/doc/invoke.texi
> > @@ -1436,7 +1436,7 @@ See RS/6000 and PowerPC Options.
> > -mavx5124fmaps -mavx512vnni -mavx5124vnniw -mprfchw -mrdpid
> @gol
> > -mrdseed -msgx -mavx512vp2intersect -mserialize -mtsxldtrk@gol
> > -mamx-tile -mamx-int8 -mamx-bf16 -muintr -mhreset -mavxvnni@gol
> > --mavx512fp16 -mavxifma @gol
> > +-mavx512fp16 -mavxifma -mavxvnniint8 @gol
> > -mcldemote -mms-bitfields -mno-align-stringops
> > -minline-all-stringops @gol -minline-stringops-dynamically
> > -mstringop-strategy=@var{alg} @gol -mkl -mwidekl @gol @@ -32896,6
> > +32896,9 @@ preferred alignment to @option{-mpreferred-stack-
> boundary=2}.
> > @need 200
> > @itemx -mavxifma
> > @opindex mavxifma
> > +@need 200
> > +@itemx -mavxvnniint8
> > +@opindex mavxvnniint8
> > These switches enable the use of instructions in the MMX, SSE, SSE2,
> > SSE3, SSSE3, SSE4, SSE4A, SSE4.1, SSE4.2, AVX, AVX2, AVX512F,
> > AVX512PF, AVX512ER, AVX512CD, AVX512VL, AVX512BW, AVX512DQ,
> > AVX512IFMA, AVX512VBMI, SHA, @@ -32906,8 +32909,8 @@ XSAVEOPT,
> XSAVEC,
> > XSAVES, RTM, HLE, TBM, MWAITX, CLZERO, PKU, AVX512VBMI2, GFNI,
> VAES,
> > WAITPKG, VPCLMULQDQ, AVX512BITALG, MOVDIRI, MOVDIR64B,
> AVX512BF16,
> > ENQCMD, AVX512VPOPCNTDQ, AVX5124FMAPS, AVX512VNNI,
> AVX5124VNNIW,
> > SERIALIZE, UINTR, HRESET, AMXTILE, AMXINT8, AMXBF16, KL, WIDEKL,
> > AVXVNNI, AVX512FP16, -AVXIFMA or CLDEMOTE extended instruction sets.
> Each has a corresponding -@option{-mno-} option to disable use of these
> instructions.
> > +AVXIFMA, AVXVNNIINT8 or CLDEMOTE extended instruction sets. Each
> has
> > +a corresponding @option{-mno-} option to disable use of these
> instructions.
> >
> > These extensions are also available as built-in functions: see
> > @ref{x86 Built-in Functions}, for details of the functions enabled
> > and diff --git a/gcc/doc/sourcebuild.texi b/gcc/doc/sourcebuild.texi
> > index 0173acf4a65..e21a1d381e0 100644
> > --- a/gcc/doc/sourcebuild.texi
> > +++ b/gcc/doc/sourcebuild.texi
> > @@ -2493,6 +2493,9 @@ Target supports the execution of
> @code{avx512vp2intersect} instructions.
> > @item avxifma
> > Target supports the execution of @code{avxifma} instructions.
> >
> > +@item avxvnniint8
> > +Target supports the execution of @code{avxvnniint8} instructions.
> > +
> > @item amx_tile
> > Target supports the execution of @code{amx-tile} instructions.
> >
> > diff --git a/gcc/testsuite/g++.dg/other/i386-2.C
> > b/gcc/testsuite/g++.dg/other/i386-2.C
> > index 5388606779b..ebd01fe47bc 100644
> > --- a/gcc/testsuite/g++.dg/other/i386-2.C
> > +++ b/gcc/testsuite/g++.dg/other/i386-2.C
> > @@ -1,5 +1,5 @@
> > /* { dg-do compile { target i?86-*-* x86_64-*-* } } */
> > -/* { dg-options "-O -pedantic-errors -march=k8 -msse4a -m3dnow -mavx
> > -mavx2 -mfma4 -mxop -maes -mpclmul -mpopcnt -mabm -mlzcnt -mbmi -
> mbmi2
> > -mtbm -mlwp -mfsgsbase -mrdrnd -mf16c -mfma -mrtm -mrdseed -
> mprfchw
> > -madx -mfxsr -mxsaveopt -mavx512f -mavx512er -mavx512cd -mavx512pf
> > -msha -mprefetchwt1 -mxsavec -mxsaves -mclflushopt -mavx512dq
> > -mavx512bw -mavx512vl -mavx512ifma -mavx512vbmi -mavx512vbmi2
> > -mavx5124fmaps -mavx5124vnniw -mavx512vpopcntdq -mclwb -mmwaitx
> > -mclzero -mpku -msgx -mrdpid -mgfni -mavx512bitalg -mpconfig
> > -mwbnoinvd -mavx512bf16 -menqcmd -mavx512vp2intersect -mserialize
> > -mtsxldtrk -mamx-tile -mamx-int8 -mamx-bf16 -mkl -mwidekl -mavxvnni
> > -mavx512fp16 -mavxifma" } */
> > +/* { dg-options "-O -pedantic-errors -march=k8 -msse4a -m3dnow -mavx
> > +-mavx2 -mfma4 -mxop -maes -mpclmul -mpopcnt -mabm -mlzcnt -mbmi
> > +-mbmi2 -mtbm -mlwp -mfsgsbase -mrdrnd -mf16c -mfma -mrtm -
> mrdseed
> > +-mprfchw -madx -mfxsr -mxsaveopt -mavx512f -mavx512er -mavx512cd
> > +-mavx512pf -msha -mprefetchwt1 -mxsavec -mxsaves -mclflushopt
> > +-mavx512dq -mavx512bw -mavx512vl -mavx512ifma -mavx512vbmi
> > +-mavx512vbmi2 -mavx5124fmaps -mavx5124vnniw -mavx512vpopcntdq -
> mclwb
> > +-mmwaitx -mclzero -mpku -msgx -mrdpid -mgfni -mavx512bitalg -
> mpconfig
> > +-mwbnoinvd -mavx512bf16 -menqcmd -mavx512vp2intersect -mserialize
> > +-mtsxldtrk -mamx-tile -mamx-int8 -mamx-bf16 -mkl -mwidekl -mavxvnni
> > +-mavx512fp16 -mavxifma -mavxvnniint8" } */
> >
> > /* Test that {,x,e,p,t,s,w,a,b,i}mmintrin.h, mm3dnow.h, fma4intrin.h,
> > xopintrin.h, abmintrin.h, bmiintrin.h, tbmintrin.h, lwpintrin.h,
> > diff --git a/gcc/testsuite/g++.dg/other/i386-3.C
> > b/gcc/testsuite/g++.dg/other/i386-3.C
> > index 86cedd3d32f..b66498f1d4c 100644
> > --- a/gcc/testsuite/g++.dg/other/i386-3.C
> > +++ b/gcc/testsuite/g++.dg/other/i386-3.C
> > @@ -1,5 +1,5 @@
> > /* { dg-do compile { target i?86-*-* x86_64-*-* } } */
> > -/* { dg-options "-O -fkeep-inline-functions -march=k8 -msse4a -m3dnow
> > -mavx -mavx2 -mfma4 -mxop -maes -mpclmul -mpopcnt -mabm -mlzcnt -
> mbmi
> > -mbmi2 -mtbm -mlwp -mfsgsbase -mrdrnd -mf16c -mfma -mrtm -mrdseed
> > -mprfchw -madx -mfxsr -mxsaveopt -mavx512f -mavx512er -mavx512cd
> > -mavx512pf -msha -mprefetchwt1 -mxsavec -mxsaves -mclflushopt
> > -mavx512dq -mavx512bw -mavx512vl -mavx512ifma -mavx512vbmi
> > -mavx512vbmi2 -mavx5124fmaps -mavx5124vnniw -mavx512vpopcntdq -
> mclwb
> > -mmwaitx -mclzero -mpku -msgx -mrdpid -mgfni -mavx512bitalg -mpconfig
> > -mwbnoinvd -mavx512bf16 -menqcmd -mavx512vp2intersect -mserialize
> > -mtsxldtrk -mamx-tile -mamx-int8 -mamx-bf16 -mkl -mwidekl -mavxvnni
> > -mavx512fp16 -mavxifma" } */
> > +/* { dg-options "-O -fkeep-inline-functions -march=k8 -msse4a -m3dnow
> > +-mavx -mavx2 -mfma4 -mxop -maes -mpclmul -mpopcnt -mabm -mlzcnt -
> mbmi
> > +-mbmi2 -mtbm -mlwp -mfsgsbase -mrdrnd -mf16c -mfma -mrtm -
> mrdseed
> > +-mprfchw -madx -mfxsr -mxsaveopt -mavx512f -mavx512er -mavx512cd
> > +-mavx512pf -msha -mprefetchwt1 -mxsavec -mxsaves -mclflushopt
> > +-mavx512dq -mavx512bw -mavx512vl -mavx512ifma -mavx512vbmi
> > +-mavx512vbmi2 -mavx5124fmaps -mavx5124vnniw -mavx512vpopcntdq -
> mclwb
> > +-mmwaitx -mclzero -mpku -msgx -mrdpid -mgfni -mavx512bitalg -
> mpconfig
> > +-mwbnoinvd -mavx512bf16 -menqcmd -mavx512vp2intersect -mserialize
> > +-mtsxldtrk -mamx-tile -mamx-int8 -mamx-bf16 -mkl -mwidekl -mavxvnni
> > +-mavx512fp16 -mavxifma -mavxvnniint8" } */
> >
> > /* Test that {,x,e,p,t,s,w,a,b,i}mmintrin.h, mm3dnow.h, fma4intrin.h,
> > xopintrin.h, abmintrin.h, bmiintrin.h, tbmintrin.h, lwpintrin.h,
> > diff --git a/gcc/testsuite/gcc.target/i386/avx-check.h
> > b/gcc/testsuite/gcc.target/i386/avx-check.h
> > index 24ee6ab4efd..77507ca2edc 100644
> > --- a/gcc/testsuite/gcc.target/i386/avx-check.h
> > +++ b/gcc/testsuite/gcc.target/i386/avx-check.h
> > @@ -25,6 +25,9 @@ main ()
> > && avx_os_support ()
> > #ifdef AVXIFMA
> > && __builtin_cpu_supports ("avxifma")
> > +#endif
> > +#ifdef AVXVNNIINT8
> > + && __builtin_cpu_supports ("avxvnniint8")
> > #endif
> > )
> > {
> > diff --git a/gcc/testsuite/gcc.target/i386/avxvnniint8-1.c
> > b/gcc/testsuite/gcc.target/i386/avxvnniint8-1.c
> > new file mode 100644
> > index 00000000000..d6942f34d6e
> > --- /dev/null
> > +++ b/gcc/testsuite/gcc.target/i386/avxvnniint8-1.c
> > @@ -0,0 +1,43 @@
> > +/* { dg-do compile } */
> > +/* { dg-options "-mavxvnniint8 -O2" } */
> > +/* { dg-final { scan-assembler-times "vpdpbssd\[
> > +\\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\r]*%ymm\[0-
> 9\]+\[^\n\r]*%ymm\[0-9\]
> > ++(?:\n|\[ \\t\]+#)" 1 } } */
> > +/* { dg-final { scan-assembler-times "vpdpbssd\[
> > +\\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-
> 9\]+\[^\n\r]*%xmm\[0-9\]
> > ++(?:\n|\[ \\t\]+#)" 1 } } */
> > +/* { dg-final { scan-assembler-times "vpdpbssds\[
> > +\\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\r]*%ymm\[0-
> 9\]+\[^\n\r]*%ymm\[0-9\]
> > ++(?:\n|\[ \\t\]+#)" 1 } } */
> > +/* { dg-final { scan-assembler-times "vpdpbssds\[
> > +\\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-
> 9\]+\[^\n\r]*%xmm\[0-9\]
> > ++(?:\n|\[ \\t\]+#)" 1 } } */
> > +/* { dg-final { scan-assembler-times "vpdpbsud\[
> > +\\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\r]*%ymm\[0-
> 9\]+\[^\n\r]*%ymm\[0-9\]
> > ++(?:\n|\[ \\t\]+#)" 1 } } */
> > +/* { dg-final { scan-assembler-times "vpdpbsud\[
> > +\\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-
> 9\]+\[^\n\r]*%xmm\[0-9\]
> > ++(?:\n|\[ \\t\]+#)" 1 } } */
> > +/* { dg-final { scan-assembler-times "vpdpbsuds\[
> > +\\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\r]*%ymm\[0-
> 9\]+\[^\n\r]*%ymm\[0-9\]
> > ++(?:\n|\[ \\t\]+#)" 1 } } */
> > +/* { dg-final { scan-assembler-times "vpdpbsuds\[
> > +\\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-
> 9\]+\[^\n\r]*%xmm\[0-9\]
> > ++(?:\n|\[ \\t\]+#)" 1 } } */
> > +/* { dg-final { scan-assembler-times "vpdpbuud\[
> > +\\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\r]*%ymm\[0-
> 9\]+\[^\n\r]*%ymm\[0-9\]
> > ++(?:\n|\[ \\t\]+#)" 1 } } */
> > +/* { dg-final { scan-assembler-times "vpdpbuud\[
> > +\\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-
> 9\]+\[^\n\r]*%xmm\[0-9\]
> > ++(?:\n|\[ \\t\]+#)" 1 } } */
> > +/* { dg-final { scan-assembler-times "vpdpbuuds\[
> > +\\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\r]*%ymm\[0-
> 9\]+\[^\n\r]*%ymm\[0-9\]
> > ++(?:\n|\[ \\t\]+#)" 1 } } */
> > +/* { dg-final { scan-assembler-times "vpdpbuuds\[
> > +\\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-
> 9\]+\[^\n\r]*%xmm\[0-9\]
> > ++(?:\n|\[ \\t\]+#)" 1 } } */
> > +
> > +
> > +#include <immintrin.h>
> > +
> > +volatile __m256i x,y,z;
> > +volatile __m128i x_,y_,z_;
> > +volatile __mmask8 m;
> > +
> > +void extern
> > +avxvnniint8_test (void)
> > +{
> > + x = _mm256_dpbssd_epi32 (x, y, z);
> > + x_ = _mm_dpbssd_epi32 (x_, y_, z_);
> > +
> > + x = _mm256_dpbssds_epi32 (x, y, z); x_ = _mm_dpbssds_epi32 (x_,
> > + y_, z_);
> > +
> > + x = _mm256_dpbsud_epi32 (x, y, z);
> > + x_ = _mm_dpbsud_epi32 (x_, y_, z_);
> > +
> > + x = _mm256_dpbsuds_epi32 (x, y, z); x_ = _mm_dpbsuds_epi32 (x_,
> > + y_, z_);
> > +
> > + x = _mm256_dpbuud_epi32 (x, y, z);
> > + x_ = _mm_dpbuud_epi32 (x_, y_, z_);
> > +
> > + x = _mm256_dpbuuds_epi32 (x, y, z);
> > + x_ = _mm_dpbuuds_epi32 (x_, y_, z_); }
> > diff --git a/gcc/testsuite/gcc.target/i386/avxvnniint8-vpdpbssd-2.c
> > b/gcc/testsuite/gcc.target/i386/avxvnniint8-vpdpbssd-2.c
> > new file mode 100644
> > index 00000000000..5016de39621
> > --- /dev/null
> > +++ b/gcc/testsuite/gcc.target/i386/avxvnniint8-vpdpbssd-2.c
> > @@ -0,0 +1,72 @@
> > +/* { dg-do run } */
> > +/* { dg-options "-O2 -mavxvnniint8" } */
> > +/* { dg-require-effective-target avxvnniint8 } */ #define AVXVNNIINT8
> > +#ifndef CHECK #define CHECK "avx-check.h"
> > +#endif
> > +
> > +#ifndef TEST
> > +#define TEST avx_test
> > +#endif
> > +
> > +#include CHECK
> > +
> > +static void
> > +CALC (int *r, int *dst, char *s1, char *s2, int size) {
> > + short tempres[32];
> > + for (int i = 0; i < size; i++) {
> > + tempres[i] = (short) s1[i] * (short) s2[i];
> > + }
> > + for (int i = 0; i < size / 4; i++) {
> > + long long test = (long long) dst[i] + tempres[i * 4] + tempres[i * 4 + 1]
> > + + tempres[i * 4 + 2] + tempres[i * 4 + 3];
> > + r[i] = test;
> > + }
> > +}
> > +
> > +void
> > +TEST (void)
> > +{
> > + int i;
> > + union256i_d res_256;
> > + union256i_b src2_256;
> > + union256i_b src1_256;
> > + int res_ref_256[8];
> > +
> > + for (i = 0; i < 32; i++)
> > + {
> > + int sign = i % 2 ? 1 : -1;
> > + src1_256.a[i] = 10 + 3 * i + sign;
> > + src2_256.a[i] = sign * 10 * i * i;
> > + }
> > +
> > + for (i = 0; i < 8; i++)
> > + res_256.a[i] = 0x7fffffff;
> > +
> > + CALC (res_ref_256, res_256.a, src1_256.a, src2_256.a, 32);
> > + res_256.x = _mm256_dpbssd_epi32 (res_256.x, src1_256.x, src2_256.x);
> > + if (check_union256i_d (res_256, res_ref_256))
> > + abort ();
> > +
> > + union128i_d res_128;
> > + union128i_b src2_128;
> > + union128i_b src1_128;
> > + int res_ref_128[4];
> > +
> > + for (i = 0; i < 16; i++)
> > + {
> > + int sign = i % 2 ? 1 : -1;
> > + src1_128.a[i] = 10 + 3 * i * i + sign;
> > + src2_128.a[i] = sign * 10 * i * i;
> > + }
> > +
> > + for (i = 0; i < 4; i++)
> > + res_128.a[i] = 0x7fffffff;
> > +
> > + CALC (res_ref_128, res_128.a, src1_128.a, src2_128.a, 16);
> > + res_128.x = _mm_dpbssd_epi32 (res_128.x, src1_128.x, src2_128.x);
> > + if (check_union128i_d (res_128, res_ref_128))
> > + abort ();
> > +}
> > diff --git a/gcc/testsuite/gcc.target/i386/avxvnniint8-vpdpbssds-2.c
> > b/gcc/testsuite/gcc.target/i386/avxvnniint8-vpdpbssds-2.c
> > new file mode 100644
> > index 00000000000..6de5062e917
> > --- /dev/null
> > +++ b/gcc/testsuite/gcc.target/i386/avxvnniint8-vpdpbssds-2.c
> > @@ -0,0 +1,72 @@
> > +/* { dg-do run } */
> > +/* { dg-options "-O2 -mavxvnniint8" } */
> > +/* { dg-require-effective-target avxvnniint8 } */ #define AVXVNNIINT8
> > +#ifndef CHECK #define CHECK "avx-check.h"
> > +#endif
> > +
> > +#ifndef TEST
> > +#define TEST avx_test
> > +#endif
> > +
> > +#include CHECK
> > +
> > +static void
> > +CALC (int *r, int *dst, char *s1, char *s2, int size) {
> > + short tempres[32];
> > + for (int i = 0; i < size; i++) {
> > + tempres[i] = (short) s1[i] * (short) s2[i];
> > + }
> > + for (int i = 0; i < size / 4; i++) {
> > + long long test = (long long) dst[i] + tempres[i * 4] + tempres[i * 4 + 1]
> > + + tempres[i * 4 + 2] + tempres[i * 4 + 3];
> > + r[i] = test > 0x7FFFFFFF ? 0x7FFFFFFF : test;
> > + }
> > +}
> > +
> > +void
> > +TEST (void)
> > +{
> > + int i;
> > + union256i_d res_256;
> > + union256i_b src2_256;
> > + union256i_b src1_256;
> > + int res_ref_256[8];
> > +
> > + for (i = 0; i < 32; i++)
> > + {
> > + int sign = i % 2 ? 1 : -1;
> > + src1_256.a[i] = 10 + 3 * i + sign;
> > + src2_256.a[i] = sign * 10 * i * i;
> > + }
> > +
> > + for (i = 0; i < 8; i++)
> > + res_256.a[i] = 0x7fffffff;
> > +
> > + CALC (res_ref_256, res_256.a, src1_256.a, src2_256.a, 32);
> > + res_256.x = _mm256_dpbssds_epi32 (res_256.x, src1_256.x,
> > + src2_256.x); if (check_union256i_d (res_256, res_ref_256))
> > + abort ();
> > +
> > + union128i_d res_128;
> > + union128i_b src2_128;
> > + union128i_b src1_128;
> > + int res_ref_128[4];
> > +
> > + for (i = 0; i < 16; i++)
> > + {
> > + int sign = i % 2 ? 1 : -1;
> > + src1_128.a[i] = 10 + 3 * i * i + sign;
> > + src2_128.a[i] = sign * 10 * i * i;
> > + }
> > +
> > + for (i = 0; i < 4; i++)
> > + res_128.a[i] = 0x7fffffff;
> > +
> > + CALC (res_ref_128, res_128.a, src1_128.a, src2_128.a, 16);
> > + res_128.x = _mm_dpbssds_epi32 (res_128.x, src1_128.x, src2_128.x);
> > + if (check_union128i_d (res_128, res_ref_128))
> > + abort ();
> > +}
> > diff --git a/gcc/testsuite/gcc.target/i386/avxvnniint8-vpdpbsud-2.c
> > b/gcc/testsuite/gcc.target/i386/avxvnniint8-vpdpbsud-2.c
> > new file mode 100644
> > index 00000000000..6e4ffd1c7be
> > --- /dev/null
> > +++ b/gcc/testsuite/gcc.target/i386/avxvnniint8-vpdpbsud-2.c
> > @@ -0,0 +1,72 @@
> > +/* { dg-do run } */
> > +/* { dg-options "-O2 -mavxvnniint8" } */
> > +/* { dg-require-effective-target avxvnniint8 } */ #define AVXVNNIINT8
> > +#ifndef CHECK #define CHECK "avx-check.h"
> > +#endif
> > +
> > +#ifndef TEST
> > +#define TEST avx_test
> > +#endif
> > +
> > +#include CHECK
> > +
> > +static void
> > +CALC (int *r, int *dst, char *s1, unsigned char *s2, int size) {
> > + short tempres[32];
> > + for (int i = 0; i < size; i++) {
> > + tempres[i] = (short) s1[i] * (unsigned short) s2[i];
> > + }
> > + for (int i = 0; i < size / 4; i++) {
> > + long long test = (long long) dst[i] + tempres[i * 4] + tempres[i * 4 + 1]
> > + + tempres[i * 4 + 2] + tempres[i * 4 + 3];
> > + r[i] = test;
> > + }
> > +}
> > +
> > +void
> > +TEST (void)
> > +{
> > + int i;
> > + union256i_d res_256;
> > + union256i_b src1_256;
> > + union256i_ub src2_256;
> > + int res_ref_256[8];
> > +
> > + for (i = 0; i < 32; i++)
> > + {
> > + int sign = i % 2 ? 1 : -1;
> > + src1_256.a[i] = 10 + 3 * i + sign;
> > + src2_256.a[i] = sign * 10 * i * i;
> > + }
> > +
> > + for (i = 0; i < 8; i++)
> > + res_256.a[i] = 0x7fffffff;
> > +
> > + CALC (res_ref_256, res_256.a, src1_256.a, src2_256.a, 32);
> > + res_256.x = _mm256_dpbsud_epi32 (res_256.x, src1_256.x, src2_256.x);
> > + if (check_union256i_d (res_256, res_ref_256))
> > + abort ();
> > +
> > + union128i_d res_128;
> > + union128i_b src1_128;
> > + union128i_ub src2_128;
> > + int res_ref_128[4];
> > +
> > + for (i = 0; i < 16; i++)
> > + {
> > + int sign = i % 2 ? 1 : -1;
> > + src1_128.a[i] = 10 + 3 * i * i + sign;
> > + src2_128.a[i] = sign * 10 * i * i;
> > + }
> > +
> > + for (i = 0; i < 4; i++)
> > + res_128.a[i] = 0x7fffffff;
> > +
> > + CALC (res_ref_128, res_128.a, src1_128.a, src2_128.a, 16);
> > + res_128.x = _mm_dpbsud_epi32 (res_128.x, src1_128.x, src2_128.x);
> > + if (check_union128i_d (res_128, res_ref_128))
> > + abort ();
> > +}
> > diff --git a/gcc/testsuite/gcc.target/i386/avxvnniint8-vpdpbsuds-2.c
> > b/gcc/testsuite/gcc.target/i386/avxvnniint8-vpdpbsuds-2.c
> > new file mode 100644
> > index 00000000000..ad4b6047ecd
> > --- /dev/null
> > +++ b/gcc/testsuite/gcc.target/i386/avxvnniint8-vpdpbsuds-2.c
> > @@ -0,0 +1,72 @@
> > +/* { dg-do run } */
> > +/* { dg-options "-O2 -mavxvnniint8" } */
> > +/* { dg-require-effective-target avxvnniint8 } */ #define AVXVNNIINT8
> > +#ifndef CHECK #define CHECK "avx-check.h"
> > +#endif
> > +
> > +#ifndef TEST
> > +#define TEST avx_test
> > +#endif
> > +
> > +#include CHECK
> > +
> > +static void
> > +CALC (int *r, int *dst, char *s1, unsigned char *s2, int size) {
> > + short tempres[32];
> > + for (int i = 0; i < size; i++) {
> > + tempres[i] = (short) s1[i] * (unsigned short) s2[i];
> > + }
> > + for (int i = 0; i < size / 4; i++) {
> > + long long test = (long long) dst[i] + tempres[i * 4] + tempres[i * 4 + 1]
> > + + tempres[i * 4 + 2] + tempres[i * 4 + 3];
> > + r[i] = test > 0x7FFFFFFF ? 0x7FFFFFFF : test;
> > + }
> > +}
> > +
> > +void
> > +TEST (void)
> > +{
> > + int i;
> > + union256i_d res_256;
> > + union256i_b src1_256;
> > + union256i_ub src2_256;
> > + int res_ref_256[8];
> > +
> > + for (i = 0; i < 32; i++)
> > + {
> > + int sign = i % 2 ? 1 : -1;
> > + src1_256.a[i] = 10 + 3 * i + sign;
> > + src2_256.a[i] = sign * 10 * i * i;
> > + }
> > +
> > + for (i = 0; i < 8; i++)
> > + res_256.a[i] = 0x7fffffff;
> > +
> > + CALC (res_ref_256, res_256.a, src1_256.a, src2_256.a, 32);
> > + res_256.x = _mm256_dpbsuds_epi32 (res_256.x, src1_256.x,
> > + src2_256.x); if (check_union256i_d (res_256, res_ref_256))
> > + abort ();
> > +
> > + union128i_d res_128;
> > + union128i_b src1_128;
> > + union128i_ub src2_128;
> > + int res_ref_128[4];
> > +
> > + for (i = 0; i < 16; i++)
> > + {
> > + int sign = i % 2 ? 1 : -1;
> > + src1_128.a[i] = 10 + 3 * i * i + sign;
> > + src2_128.a[i] = sign * 10 * i * i;
> > + }
> > +
> > + for (i = 0; i < 4; i++)
> > + res_128.a[i] = 0x7fffffff;
> > +
> > + CALC (res_ref_128, res_128.a, src1_128.a, src2_128.a, 16);
> > + res_128.x = _mm_dpbsuds_epi32 (res_128.x, src1_128.x, src2_128.x);
> > + if (check_union128i_d (res_128, res_ref_128))
> > + abort ();
> > +}
> > diff --git a/gcc/testsuite/gcc.target/i386/avxvnniint8-vpdpbuud-2.c
> > b/gcc/testsuite/gcc.target/i386/avxvnniint8-vpdpbuud-2.c
> > new file mode 100644
> > index 00000000000..6590915a459
> > --- /dev/null
> > +++ b/gcc/testsuite/gcc.target/i386/avxvnniint8-vpdpbuud-2.c
> > @@ -0,0 +1,72 @@
> > +/* { dg-do run } */
> > +/* { dg-options "-O2 -mavxvnniint8" } */
> > +/* { dg-require-effective-target avxvnniint8 } */ #define AVXVNNIINT8
> > +#ifndef CHECK #define CHECK "avx-check.h"
> > +#endif
> > +
> > +#ifndef TEST
> > +#define TEST avx_test
> > +#endif
> > +
> > +#include CHECK
> > +
> > +static void
> > +CALC (unsigned int *r, unsigned int *dst, unsigned char *s1, unsigned
> > +char *s2, int size) {
> > + unsigned short tempres[32];
> > + for (int i = 0; i < size; i++) {
> > + tempres[i] = (unsigned short) s1[i] * (unsigned short) s2[i];
> > + }
> > + for (int i = 0; i < size / 4; i++) {
> > + unsigned int test = (unsigned int) dst[i] + tempres[i * 4] + tempres[i * 4
> + 1]
> > + + tempres[i * 4 + 2] + tempres[i * 4 + 3];
> > + r[i] = test;
> > + }
> > +}
> > +
> > +void
> > +TEST (void)
> > +{
> > + int i;
> > + union256i_ud res_256;
> > + union256i_ub src2_256;
> > + union256i_ub src1_256;
> > + unsigned int res_ref_256[8];
> > +
> > + for (i = 0; i < 32; i++)
> > + {
> > + int sign = i % 2 ? 1 : -1;
> > + src1_256.a[i] = 10 + 3 * i + sign;
> > + src2_256.a[i] = sign * 10 * i * i;
> > + }
> > +
> > + for (i = 0; i < 8; i++)
> > + res_256.a[i] = 0x7fffffff;
> > +
> > + CALC (res_ref_256, res_256.a, src1_256.a, src2_256.a, 32);
> > + res_256.x = _mm256_dpbuud_epi32 (res_256.x, src1_256.x, src2_256.x);
> > + if (check_union256i_ud (res_256, res_ref_256))
> > + abort ();
> > +
> > + union128i_ud res_128;
> > + union128i_ub src2_128;
> > + union128i_ub src1_128;
> > + unsigned int res_ref_128[4];
> > +
> > + for (i = 0; i < 16; i++)
> > + {
> > + int sign = i % 2 ? 1 : -1;
> > + src1_128.a[i] = 10 + 3 * i * i + sign;
> > + src2_128.a[i] = sign * 10 * i * i;
> > + }
> > +
> > + for (i = 0; i < 4; i++)
> > + res_128.a[i] = 0x7fffffff;
> > +
> > + CALC (res_ref_128, res_128.a, src1_128.a, src2_128.a, 16);
> > + res_128.x = _mm_dpbuud_epi32 (res_128.x, src1_128.x, src2_128.x);
> > + if (check_union128i_ud (res_128, res_ref_128))
> > + abort ();
> > +}
> > diff --git a/gcc/testsuite/gcc.target/i386/avxvnniint8-vpdpbuuds-2.c
> > b/gcc/testsuite/gcc.target/i386/avxvnniint8-vpdpbuuds-2.c
> > new file mode 100644
> > index 00000000000..970e4a5d408
> > --- /dev/null
> > +++ b/gcc/testsuite/gcc.target/i386/avxvnniint8-vpdpbuuds-2.c
> > @@ -0,0 +1,72 @@
> > +/* { dg-do run } */
> > +/* { dg-options "-O2 -mavxvnniint8" } */
> > +/* { dg-require-effective-target avxvnniint8 } */ #define AVXVNNIINT8
> > +#ifndef CHECK #define CHECK "avx-check.h"
> > +#endif
> > +
> > +#ifndef TEST
> > +#define TEST avx_test
> > +#endif
> > +
> > +#include CHECK
> > +
> > +static void
> > +CALC (unsigned int *r, unsigned int *dst, unsigned char *s1, unsigned
> > +char *s2, int size) {
> > + unsigned short tempres[32];
> > + for (int i = 0; i < size; i++) {
> > + tempres[i] = (unsigned short) s1[i] * (unsigned short) s2[i];
> > + }
> > + for (int i = 0; i < size / 4; i++) {
> > + unsigned int test = (unsigned int) dst[i] + tempres[i * 4] + tempres[i * 4
> + 1]
> > + + tempres[i * 4 + 2] + tempres[i * 4 + 3];
> > + r[i] = test > 0xFFFFFFFF ? 0xFFFFFFFF : test;
> > + }
> > +}
> > +
> > +void
> > +TEST (void)
> > +{
> > + int i;
> > + union256i_ud res_256;
> > + union256i_ub src2_256;
> > + union256i_ub src1_256;
> > + unsigned int res_ref_256[8];
> > +
> > + for (i = 0; i < 32; i++)
> > + {
> > + int sign = i % 2 ? 1 : -1;
> > + src1_256.a[i] = 10 + 3 * i + sign;
> > + src2_256.a[i] = sign * 10 * i * i;
> > + }
> > +
> > + for (i = 0; i < 8; i++)
> > + res_256.a[i] = 0x7fffffff;
> > +
> > + CALC (res_ref_256, res_256.a, src1_256.a, src2_256.a, 32);
> > + res_256.x = _mm256_dpbuuds_epi32 (res_256.x, src1_256.x,
> > + src2_256.x); if (check_union256i_ud (res_256, res_ref_256))
> > + abort ();
> > +
> > + union128i_ud res_128;
> > + union128i_ub src2_128;
> > + union128i_ub src1_128;
> > + unsigned int res_ref_128[4];
> > +
> > + for (i = 0; i < 16; i++)
> > + {
> > + int sign = i % 2 ? 1 : -1;
> > + src1_128.a[i] = 10 + 3 * i * i + sign;
> > + src2_128.a[i] = sign * 10 * i * i;
> > + }
> > +
> > + for (i = 0; i < 4; i++)
> > + res_128.a[i] = 0x7fffffff;
> > +
> > + CALC (res_ref_128, res_128.a, src1_128.a, src2_128.a, 16);
> > + res_128.x = _mm_dpbuuds_epi32 (res_128.x, src1_128.x, src2_128.x);
> > + if (check_union128i_ud (res_128, res_ref_128))
> > + abort ();
> > +}
> > diff --git a/gcc/testsuite/gcc.target/i386/funcspec-56.inc
> > b/gcc/testsuite/gcc.target/i386/funcspec-56.inc
> > index 466555c0d06..a681bffe3e7 100644
> > --- a/gcc/testsuite/gcc.target/i386/funcspec-56.inc
> > +++ b/gcc/testsuite/gcc.target/i386/funcspec-56.inc
> > @@ -81,6 +81,7 @@ extern void test_widekl (void)
> __attribute__((__target__("widekl")));
> > extern void test_avxvnni (void)
> __attribute__((__target__("avxvnni")));
> > extern void test_avx512fp16 (void)
> __attribute__((__target__("avx512fp16")));
> > extern void test_avxifma (void)
> __attribute__((__target__("avxifma")));
> > +extern void test_avxvnniint8 (void)
> __attribute__((__target__("avxvnniint8")));
> >
> > extern void test_no_sgx (void) __attribute__((__target__("no-
> sgx")));
> > extern void test_no_avx5124fmaps(void)
> __attribute__((__target__("no-avx5124fmaps")));
> > @@ -163,6 +164,7 @@ extern void test_no_widekl (void)
> __attribute__((__target__("no-widekl")));
> > extern void test_no_avxvnni (void) __attribute__((__target__("no-
> avxvnni")));
> > extern void test_no_avx512fp16 (void)
> __attribute__((__target__("no-avx512fp16")));
> > extern void test_no_avxifma (void) __attribute__((__target__("no-
> avxifma")));
> > +extern void test_no_avxvnniint8 (void)
> __attribute__((__target__("no-avxvnniint8")));
> >
> > extern void test_arch_nocona (void)
> __attribute__((__target__("arch=nocona")));
> > extern void test_arch_core2 (void)
> __attribute__((__target__("arch=core2")));
> > diff --git a/gcc/testsuite/gcc.target/i386/sse-12.c
> > b/gcc/testsuite/gcc.target/i386/sse-12.c
> > index fde56261d8f..ddde2df6657 100644
> > --- a/gcc/testsuite/gcc.target/i386/sse-12.c
> > +++ b/gcc/testsuite/gcc.target/i386/sse-12.c
> > @@ -3,7 +3,7 @@
> > popcntintrin.h gfniintrin.h and mm_malloc.h are usable
> > with -O -std=c89 -pedantic-errors. */
> > /* { dg-do compile } */
> > -/* { dg-options "-O -std=c89 -pedantic-errors -march=k8 -msse4a
> > -m3dnow -mavx -mavx2 -mfma4 -mxop -maes -mpclmul -mpopcnt -mabm
> > -mlzcnt -mbmi -mbmi2 -mtbm -mlwp -mfsgsbase -mrdrnd -mf16c -mfma -
> mrtm
> > -mrdseed -mprfchw -madx -mfxsr -mxsaveopt -mavx512f -mavx512er
> > -mavx512cd -mavx512pf -msha -mprefetchwt1 -mxsavec -mxsaves
> > -mclflushopt -mavx512bw -mavx512dq -mavx512vl -mavx512vbmi
> > -mavx512vbmi2 -mavx512ifma -mavx5124fmaps -mavx5124vnniw
> > -mavx512vpopcntdq -mclwb -mmwaitx -mclzero -mpku -msgx -mrdpid -
> mgfni
> > -mavx512bitalg -mpconfig -mwbnoinvd -mavx512bf16 -menqcmd
> > -mavx512vp2intersect -mserialize -mtsxldtrk -mamx-tile -mamx-int8
> > -mamx-bf16 -mkl -mwidekl -mavxvnni -mavxifma" } */
> > +/* { dg-options "-O -std=c89 -pedantic-errors -march=k8 -msse4a
> > +-m3dnow -mavx -mavx2 -mfma4 -mxop -maes -mpclmul -mpopcnt -
> mabm
> > +-mlzcnt -mbmi -mbmi2 -mtbm -mlwp -mfsgsbase -mrdrnd -mf16c -mfma
> > +-mrtm -mrdseed -mprfchw -madx -mfxsr -mxsaveopt -mavx512f -
> mavx512er
> > +-mavx512cd -mavx512pf -msha -mprefetchwt1 -mxsavec -mxsaves
> > +-mclflushopt -mavx512bw -mavx512dq -mavx512vl -mavx512vbmi
> > +-mavx512vbmi2 -mavx512ifma -mavx5124fmaps -mavx5124vnniw
> > +-mavx512vpopcntdq -mclwb -mmwaitx -mclzero -mpku -msgx -mrdpid -
> mgfni
> > +-mavx512bitalg -mpconfig -mwbnoinvd -mavx512bf16 -menqcmd
> > +-mavx512vp2intersect -mserialize -mtsxldtrk -mamx-tile -mamx-int8
> > +-mamx-bf16 -mkl -mwidekl -mavxvnni -mavxifma -mavxvnniint8" } */
> >
> > #include <x86intrin.h>
> >
> > diff --git a/gcc/testsuite/gcc.target/i386/sse-13.c
> > b/gcc/testsuite/gcc.target/i386/sse-13.c
> > index bb29555babe..2b293216c6f 100644
> > --- a/gcc/testsuite/gcc.target/i386/sse-13.c
> > +++ b/gcc/testsuite/gcc.target/i386/sse-13.c
> > @@ -1,5 +1,5 @@
> > /* { dg-do compile } */
> > -/* { dg-options "-O2 -Werror-implicit-function-declaration -march=k8
> > -msse4a -m3dnow -mavx -mavx2 -mfma4 -mxop -maes -mpclmul -
> mpopcnt
> > -mabm -mlzcnt -mbmi -mbmi2 -mtbm -mlwp -mfsgsbase -mrdrnd -mf16c -
> mfma
> > -mrtm -mrdseed -mprfchw -madx -mfxsr -mxsaveopt -mavx512f -
> mavx512er
> > -mavx512cd -mavx512pf -msha -mprefetchwt1 -mxsavec -mxsaves
> > -mclflushopt -mavx512vl -mavx512dq -mavx512bw -mavx512vbmi
> > -mavx512vbmi2 -mavx512ifma -mavx5124fmaps -mavx5124vnniw
> > -mavx512vpopcntdq -mavx512vp2intersect -mclwb -mmwaitx -mclzero -
> mpku
> > -msgx -mrdpid -mgfni -mavx512bitalg -mpconfig -mwbnoinvd -
> mavx512bf16
> > -menqcmd -mserialize -mtsxldtrk -mamx-tile -mamx-int8 -mamx-bf16 -mkl
> > -mwidekl -mavxvnni -mavx512fp16 -mavxifma" } */
> > +/* { dg-options "-O2 -Werror-implicit-function-declaration -march=k8
> > +-msse4a -m3dnow -mavx -mavx2 -mfma4 -mxop -maes -mpclmul -
> mpopcnt
> > +-mabm -mlzcnt -mbmi -mbmi2 -mtbm -mlwp -mfsgsbase -mrdrnd -mf16c
> > +-mfma -mrtm -mrdseed -mprfchw -madx -mfxsr -mxsaveopt -mavx512f
> > +-mavx512er -mavx512cd -mavx512pf -msha -mprefetchwt1 -mxsavec
> > +-mxsaves -mclflushopt -mavx512vl -mavx512dq -mavx512bw -
> mavx512vbmi
> > +-mavx512vbmi2 -mavx512ifma -mavx5124fmaps -mavx5124vnniw
> > +-mavx512vpopcntdq -mavx512vp2intersect -mclwb -mmwaitx -mclzero -
> mpku
> > +-msgx -mrdpid -mgfni -mavx512bitalg -mpconfig -mwbnoinvd -
> mavx512bf16
> > +-menqcmd -mserialize -mtsxldtrk -mamx-tile -mamx-int8 -mamx-bf16 -
> mkl
> > +-mwidekl -mavxvnni -mavx512fp16 -mavxifma -mavxvnniint8" } */
> > /* { dg-add-options bind_pic_locally } */
> >
> > #include <mm_malloc.h>
> > diff --git a/gcc/testsuite/gcc.target/i386/sse-14.c
> > b/gcc/testsuite/gcc.target/i386/sse-14.c
> > index f2701ddaaf9..78b51048b90 100644
> > --- a/gcc/testsuite/gcc.target/i386/sse-14.c
> > +++ b/gcc/testsuite/gcc.target/i386/sse-14.c
> > @@ -1,5 +1,5 @@
> > /* { dg-do compile } */
> > -/* { dg-options "-O0 -Werror-implicit-function-declaration -march=k8
> > -msse4a -m3dnow -mavx -mavx2 -mfma4 -mxop -maes -mpclmul -
> mpopcnt
> > -mabm -mlzcnt -mbmi -mbmi2 -mtbm -mlwp -mfsgsbase -mrdrnd -mf16c -
> mfma
> > -mrtm -mrdseed -mprfchw -madx -mfxsr -mxsaveopt -mavx512f -
> mavx512er
> > -mavx512cd -mavx512pf -msha -mprefetchwt1 -mxsavec -mxsaves
> > -mclflushopt -mavx512dq -mavx512bw -mavx512vl -mavx512ifma
> > -mavx512vbmi -mavx512vbmi2 -mavx5124fmaps -mavx5124vnniw
> > -mavx512vpopcntdq -mclwb -mmwaitx -mclzero -mpku -msgx -mrdpid -
> mgfni
> > -mpconfig -mwbnoinvd -mavx512vl -mavx512bf16 -menqcmd
> > -mavx512vp2intersect -mserialize -mtsxldtrk -mamx-tile -mamx-int8
> > -mamx-bf16 -mkl -mwidekl -mavxvnni -mavx512fp16 -mavxifma" } */
> > +/* { dg-options "-O0 -Werror-implicit-function-declaration -march=k8
> > +-msse4a -m3dnow -mavx -mavx2 -mfma4 -mxop -maes -mpclmul -
> mpopcnt
> > +-mabm -mlzcnt -mbmi -mbmi2 -mtbm -mlwp -mfsgsbase -mrdrnd -mf16c
> > +-mfma -mrtm -mrdseed -mprfchw -madx -mfxsr -mxsaveopt -mavx512f
> > +-mavx512er -mavx512cd -mavx512pf -msha -mprefetchwt1 -mxsavec
> > +-mxsaves -mclflushopt -mavx512dq -mavx512bw -mavx512vl -
> mavx512ifma
> > +-mavx512vbmi -mavx512vbmi2 -mavx5124fmaps -mavx5124vnniw
> > +-mavx512vpopcntdq -mclwb -mmwaitx -mclzero -mpku -msgx -mrdpid -
> mgfni
> > +-mpconfig -mwbnoinvd -mavx512vl -mavx512bf16 -menqcmd
> > +-mavx512vp2intersect -mserialize -mtsxldtrk -mamx-tile -mamx-int8
> > +-mamx-bf16 -mkl -mwidekl -mavxvnni -mavx512fp16 -mavxifma
> > +-mavxvnniint8" } */
> > /* { dg-add-options bind_pic_locally } */
> >
> > #include <mm_malloc.h>
> > diff --git a/gcc/testsuite/gcc.target/i386/sse-22.c
> > b/gcc/testsuite/gcc.target/i386/sse-22.c
> > index 3d196975b1e..cc1c8cfa4be 100644
> > --- a/gcc/testsuite/gcc.target/i386/sse-22.c
> > +++ b/gcc/testsuite/gcc.target/i386/sse-22.c
> > @@ -103,7 +103,7 @@
> >
> >
> > #ifndef DIFFERENT_PRAGMAS
> > -#pragma GCC target
> > ("sse4a,3dnow,avx,avx2,fma4,xop,aes,pclmul,popcnt,abm,lzcnt,bmi,bmi2,t
> > bm,lwp,fsgsbase,rdrnd,f16c,rtm,rdseed,prfchw,adx,fxsr,xsaveopt,avx512f
> > ,avx512er,avx512cd,avx512pf,sha,prefetchwt1,avx512vl,avx512bw,avx512d
> q
> > ,avx512vbmi,avx512vbmi2,avx512ifma,avx5124fmaps,avx5124vnniw,avx512
> vpo
> > pcntdq,gfni,avx512bitalg,avx512bf16,avx512vp2intersect,serialize,tsxld
> > trk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma")
> > +#pragma GCC target
> > +("sse4a,3dnow,avx,avx2,fma4,xop,aes,pclmul,popcnt,abm,lzcnt,bmi,bmi2,
> > +tbm,lwp,fsgsbase,rdrnd,f16c,rtm,rdseed,prfchw,adx,fxsr,xsaveopt,avx51
> >
> +2f,avx512er,avx512cd,avx512pf,sha,prefetchwt1,avx512vl,avx512bw,avx51
> >
> +2dq,avx512vbmi,avx512vbmi2,avx512ifma,avx5124fmaps,avx5124vnniw,avx
> 51
> > +2vpopcntdq,gfni,avx512bitalg,avx512bf16,avx512vp2intersect,serialize,
> > +tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxi
> > +fma,avxvnniint8")
> > #endif
> >
> > /* Following intrinsics require immediate arguments. They @@ -220,7
> > +220,7 @@ test_4 (_mm_cmpestrz, int, __m128i, int, __m128i, int, 1)
> >
> > /* immintrin.h (AVX/AVX2/RDRND/FSGSBASE/F16C/RTM/AVX512F/SHA)
> */
> > #ifdef DIFFERENT_PRAGMAS -#pragma GCC target
> > ("avx,avx2,rdrnd,fsgsbase,f16c,rtm,avx512f,avx512er,avx512cd,avx512pf,
> >
> sha,avx512vl,avx512bw,avx512dq,avx512ifma,avx512vbmi,avx512vbmi2,avx5
> 1
> >
> 24fmaps,avx5124vnniw,avx512vpopcntdq,gfni,avx512bitalg,avx512bf16,avx5
> > 12vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl
> > ,avxvnni,avx512fp16,avxifma")
> > +#pragma GCC target
> > +("avx,avx2,rdrnd,fsgsbase,f16c,rtm,avx512f,avx512er,avx512cd,avx512pf
> >
> +,sha,avx512vl,avx512bw,avx512dq,avx512ifma,avx512vbmi,avx512vbmi2,av
> x
> >
> +5124fmaps,avx5124vnniw,avx512vpopcntdq,gfni,avx512bitalg,avx512bf16,a
> > +vx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,wi
> > +dekl,avxvnni,avx512fp16,avxifma,avxvnniint8")
> > #endif
> > #include <immintrin.h>
> > test_1 (_cvtss_sh, unsigned short, float, 1) diff --git
> > a/gcc/testsuite/gcc.target/i386/sse-23.c
> > b/gcc/testsuite/gcc.target/i386/sse-23.c
> > index d3a233f90fc..270f4483491 100644
> > --- a/gcc/testsuite/gcc.target/i386/sse-23.c
> > +++ b/gcc/testsuite/gcc.target/i386/sse-23.c
> > @@ -843,6 +843,6 @@
> > #define __builtin_ia32_vpclmulqdq_v2di(A, B, C)
> > __builtin_ia32_vpclmulqdq_v2di(A, B, 1) #define
> > __builtin_ia32_vpclmulqdq_v8di(A, B, C)
> > __builtin_ia32_vpclmulqdq_v8di(A, B, 1)
> >
> > -#pragma GCC target
> > ("sse4a,3dnow,avx,avx2,fma4,xop,aes,pclmul,popcnt,abm,lzcnt,bmi,bmi2,t
> > bm,lwp,fsgsbase,rdrnd,f16c,fma,rtm,rdseed,prfchw,adx,fxsr,xsaveopt,avx
> > 512f,avx512er,avx512cd,avx512pf,sha,prefetchwt1,xsavec,xsaves,clflusho
> >
> pt,avx512bw,avx512dq,avx512vl,avx512vbmi,avx512ifma,avx5124fmaps,avx5
> 1
> > 24vnniw,avx512vpopcntdq,clwb,mwaitx,clzero,pku,sgx,rdpid,gfni,avx512vb
> >
> mi2,vpclmulqdq,avx512bitalg,pconfig,wbnoinvd,avx512bf16,enqcmd,avx512v
> > p2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,av
> > xvnni,avx512fp16,avxifma")
> > +#pragma GCC target
> > +("sse4a,3dnow,avx,avx2,fma4,xop,aes,pclmul,popcnt,abm,lzcnt,bmi,bmi2,
> > +tbm,lwp,fsgsbase,rdrnd,f16c,fma,rtm,rdseed,prfchw,adx,fxsr,xsaveopt,a
> > +vx512f,avx512er,avx512cd,avx512pf,sha,prefetchwt1,xsavec,xsaves,clflu
> >
> +shopt,avx512bw,avx512dq,avx512vl,avx512vbmi,avx512ifma,avx5124fmaps,
> a
> > +vx5124vnniw,avx512vpopcntdq,clwb,mwaitx,clzero,pku,sgx,rdpid,gfni,avx
> >
> +512vbmi2,vpclmulqdq,avx512bitalg,pconfig,wbnoinvd,avx512bf16,enqcmd,
> a
> > +vx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,wi
> > +dekl,avxvnni,avx512fp16,avxifma,avxvnniint8")
> >
> > #include <x86intrin.h>
> > diff --git a/gcc/testsuite/lib/target-supports.exp
> > b/gcc/testsuite/lib/target-supports.exp
> > index 69de3b96bfc..64ccfc746bd 100644
> > --- a/gcc/testsuite/lib/target-supports.exp
> > +++ b/gcc/testsuite/lib/target-supports.exp
> > @@ -9518,6 +9518,18 @@ proc check_effective_target_avxifma { } {
> > } "-O0 -mavxifma" ]
> > }
> >
> > +# Return 1 if avxvnniint8 instructions can be compiled.
> > +proc check_effective_target_avxvnniint8 { } {
> > + return [check_no_compiler_messages avxvnniint8 object {
> > + typedef int __v8si __attribute__ ((__vector_size__ (32)));
> > + __v8si
> > + _mm256_dpbssd_epi32 (__v8si __A, __v8si __B, __v8si __C)
> > + {
> > + return __builtin_ia32_vpdpbssd256 (__A, __B, __C);
> > + }
> > + } "-O0 -mavxvnniint8" ]
> > +}
> > +
> > # Return 1 if sse instructions can be compiled.
> > proc check_effective_target_sse { } {
> > return [check_no_compiler_messages sse object {
> > --
> > 2.18.1
> >
>
>
> --
> BR,
> Hongtao
next prev parent reply other threads:[~2022-10-17 6:27 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-10-14 7:54 [PATCH 0/6] Add Intel Sierra Forest Instructions Haochen Jiang
2022-10-14 7:54 ` [PATCH 1/6] Support Intel AVX-IFMA Haochen Jiang
2022-10-19 6:03 ` [PATCH] " Hongyu Wang
2022-10-21 0:52 ` Hongtao Liu
2022-10-14 7:54 ` [PATCH 2/6] Support Intel AVX-VNNI-INT8 Haochen Jiang
2022-10-17 4:05 ` Hongtao Liu
2022-10-17 6:27 ` Jiang, Haochen [this message]
2022-10-17 6:30 ` Hongtao Liu
2022-10-18 9:10 ` [PATCH v2] " Haochen Jiang
2022-10-19 1:41 ` Hongtao Liu
2022-10-21 0:53 ` Hongtao Liu
2022-10-14 7:54 ` [PATCH 3/6] i386: Add intrinsic for vector __bf16 Haochen Jiang
2022-10-14 7:54 ` [PATCH 4/6] Support Intel AVX-NE-CONVERT Haochen Jiang
2022-10-17 5:46 ` Hongtao Liu
2022-10-24 6:20 ` Kong, Lingling
2022-10-25 5:23 ` Hongtao Liu
2022-10-28 8:57 ` Kong, Lingling
2022-10-31 1:05 ` Liu, Hongtao
2022-10-14 7:54 ` [PATCH 5/6] Support Intel CMPccXADD Haochen Jiang
2022-10-24 9:01 ` [PATCH] " Haochen Jiang
2022-10-24 9:04 ` Jiang, Haochen
2022-10-24 9:44 ` Uros Bizjak
2022-11-03 6:26 ` Haochen Jiang
2022-11-03 7:47 ` Uros Bizjak
2022-10-14 7:54 ` [PATCH 6/6] Initial Sierra Forest Support Haochen Jiang
2022-11-03 6:31 ` Hongtao Liu
2022-10-14 8:14 ` [PATCH 0/6] Add Intel Sierra Forest Instructions Iain Sandoe
2022-10-14 8:20 ` Hongtao Liu
2022-10-14 8:24 ` Iain Sandoe
2022-10-14 8:30 ` Hongtao Liu
2022-10-14 8:36 ` Iain Sandoe
2022-10-17 1:02 ` Hongtao Liu
2022-10-17 1:30 ` Bernhard Reutner-Fischer
2022-10-17 1:56 ` Hongtao Liu
2022-10-19 11:09 ` Iain Sandoe
2022-10-20 1:11 ` Hongtao Liu
2022-10-20 9:09 ` Hongtao Liu
2022-10-20 9:17 ` Iain Sandoe
2022-10-20 9:20 ` Hongtao Liu
2022-10-20 15:23 ` Iain Sandoe
2022-10-21 0:06 ` Hongtao Liu
2022-10-14 8:35 ` Hongtao Liu
2022-10-18 9:17 ` [PATCH] i386: Auto vectorize sdot_prod, udot_prod with VNNIINT8 instruction Haochen Jiang
2022-10-19 1:43 ` Hongtao Liu
2022-10-21 0:54 ` Hongtao Liu
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=SA1PR11MB5946440B1D7DFF0C8F9C5416EC299@SA1PR11MB5946.namprd11.prod.outlook.com \
--to=haochen.jiang@intel.com \
--cc=crazylht@gmail.com \
--cc=gcc-patches@gcc.gnu.org \
--cc=hongtao.liu@intel.com \
--cc=lingling.kong@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).