[PATCH 0/4] Support Intel Arrow Lake/Lunar Lake ISAs

public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed

* [PATCH 0/4] Support Intel Arrow Lake/Lunar Lake ISAs
@ 2023-07-13  6:03 Haochen Jiang
  2023-07-13  6:03 ` [PATCH 1/4] Support Intel AVX-VNNI-INT16 Haochen Jiang
                   ` (3 more replies)
  0 siblings, 4 replies; 9+ messages in thread
From: Haochen Jiang @ 2023-07-13  6:03 UTC (permalink / raw)
  To: gcc-patches; +Cc: hongtao.liu, ubizjak

Hi all,

These four patches aimed to add Intel Arrow Lake/Lunar Lake
instructions, including AVX-VNNI-INT16, SM3, SHA512 and SM4.

The information is based on newly released
Intel Architecture Instruction Set Extensions and Future Features.

The document comes following:
https://www.intel.com/content/www/us/en/develop/download/intel-architecture-instruction-set-extensions-programming-reference.html

Regtested on x86_64-pc-linux-gnu. Ok for trunk?

BRs,
Haochen

^ permalink raw reply	[flat|nested] 9+ messages in thread

* [PATCH 1/4] Support Intel AVX-VNNI-INT16
  2023-07-13  6:03 [PATCH 0/4] Support Intel Arrow Lake/Lunar Lake ISAs Haochen Jiang
@ 2023-07-13  6:03 ` Haochen Jiang
  2023-07-17  1:39   ` Hongtao Liu
  2023-07-13  6:03 ` [PATCH 2/4] Support Intel SM3 Haochen Jiang
                   ` (2 subsequent siblings)
  3 siblings, 1 reply; 9+ messages in thread
From: Haochen Jiang @ 2023-07-13  6:03 UTC (permalink / raw)
  To: gcc-patches; +Cc: hongtao.liu, ubizjak, Kong Lingling

From: Kong Lingling <lingling.kong@intel.com>

gcc/ChangeLog

	* common/config/i386/cpuinfo.h (get_available_features): Detect
	avxvnniint16.
	* common/config/i386/i386-common.cc
	(OPTION_MASK_ISA2_AVXVNNIINT16_SET): New.
	(OPTION_MASK_ISA2_AVXVNNIINT16_UNSET): Ditto.
	(ix86_handle_option): Handle -mavxvnniint16.
	* common/config/i386/i386-cpuinfo.h (enum processor_features):
	Add FEATURE_AVXVNNIINT16.
	* common/config/i386/i386-isas.h: Add ISA_NAME_TABLE_ENTRY for
	avxvnniint16.
	* config.gcc: Add avxvnniint16.h.
	* config/i386/avxvnniint16intrin.h: New file.
	* config/i386/cpuid.h (bit_AVXVNNIINT16): New.
	* config/i386/i386-builtin.def: Add new builtins.
	* config/i386/i386-c.cc (ix86_target_macros_internal): Define
	__AVXVNNIINT16__.
	* config/i386/i386-options.cc (isa2_opts): Add -mavxvnniint16.
	(ix86_valid_target_attribute_inner_p): Handle avxvnniint16intrin.h.
	* config/i386/i386-isa.def: Add DEF_PTA(AVXVNNIINT16).
	* config/i386/i386.opt: Add option -mavxvnniint16.
	* config/i386/immintrin.h: Include avxvnniint16.h.
	* config/i386/sse.md
	(vpdp<vpdpwprodtype>_<mode>): New define_insn.
	* doc/extend.texi: Document avxvnniint16.
	* doc/invoke.texi: Document -mavxvnniint16.
	* doc/sourcebuild.texi: Document target avxvnniint16.

gcc/testsuite/ChangeLog

	* g++.dg/other/i386-2.C: Add -mavxvnniint16.
	* g++.dg/other/i386-3.C: Ditto.
	* gcc.target/i386/avx-check.h: Add avxvnniint16 check.
	* gcc.target/i386/sse-12.c: Add -mavxvnniint16.
	* 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_avxvnniint16): New.
	* gcc.target/i386/avxvnniint16-1.c: Ditto.
	* gcc.target/i386/avxvnniint16-vpdpwusd-2.c: Ditto.
	* gcc.target/i386/avxvnniint16-vpdpwusds-2.c: Ditto.
	* gcc.target/i386/avxvnniint16-vpdpwsud-2.c: Ditto.
	* gcc.target/i386/avxvnniint16-vpdpwsuds-2.c: Ditto.
	* gcc.target/i386/avxvnniint16-vpdpwuud-2.c: Ditto.
	* gcc.target/i386/avxvnniint16-vpdpwuuds-2.c: Ditto.

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/avxvnniint16intrin.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                        |  32 ++++
 gcc/doc/extend.texi                           |   5 +
 gcc/doc/invoke.texi                           |  10 +-
 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.target/i386/avxvnniint16-1.c          |  43 ++++++
 .../gcc.target/i386/avxvnniint16-vpdpwsud-2.c |  71 +++++++++
 .../i386/avxvnniint16-vpdpwsuds-2.c           |  72 +++++++++
 .../gcc.target/i386/avxvnniint16-vpdpwusd-2.c |  71 +++++++++
 .../i386/avxvnniint16-vpdpwusds-2.c           |  72 +++++++++
 .../gcc.target/i386/avxvnniint16-vpdpwuud-2.c |  71 +++++++++
 .../i386/avxvnniint16-vpdpwuuds-2.c           |  71 +++++++++
 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, 735 insertions(+), 15 deletions(-)
 create mode 100644 gcc/config/i386/avxvnniint16intrin.h
 create mode 100644 gcc/testsuite/gcc.target/i386/avxvnniint16-1.c
 create mode 100644 gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwsud-2.c
 create mode 100644 gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwsuds-2.c
 create mode 100644 gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwusd-2.c
 create mode 100644 gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwusds-2.c
 create mode 100644 gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwuud-2.c
 create mode 100644 gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwuuds-2.c

diff --git a/gcc/common/config/i386/cpuinfo.h b/gcc/common/config/i386/cpuinfo.h
index 7c2565c1d93..3599f9def2c 100644
--- a/gcc/common/config/i386/cpuinfo.h
+++ b/gcc/common/config/i386/cpuinfo.h
@@ -875,6 +875,8 @@ get_available_features (struct __processor_model *cpu_model,
 	    set_feature (FEATURE_AVXVNNIINT8);
 	  if (edx & bit_AVXNECONVERT)
 	    set_feature (FEATURE_AVXNECONVERT);
+	  if (edx & bit_AVXVNNIINT16)
+	    set_feature (FEATURE_AVXVNNIINT16);
 	}
       if (avx512_usable)
 	{
diff --git a/gcc/common/config/i386/i386-common.cc b/gcc/common/config/i386/i386-common.cc
index 8cea3669239..32c6d00580d 100644
--- a/gcc/common/config/i386/i386-common.cc
+++ b/gcc/common/config/i386/i386-common.cc
@@ -119,6 +119,7 @@ along with GCC; see the file COPYING3.  If not see
 #define OPTION_MASK_ISA2_RAOINT_SET OPTION_MASK_ISA2_RAOINT
 #define OPTION_MASK_ISA2_AMX_COMPLEX_SET \
   (OPTION_MASK_ISA2_AMX_TILE | OPTION_MASK_ISA2_AMX_COMPLEX)
+#define OPTION_MASK_ISA2_AVXVNNIINT16_SET OPTION_MASK_ISA2_AVXVNNIINT16
 
 /* SSE4 includes both SSE4.1 and SSE4.2. -msse4 should be the same
    as -msse4.2.  */
@@ -228,7 +229,7 @@ along with GCC; see the file COPYING3.  If not see
 #define OPTION_MASK_ISA2_AVX2_UNSET \
   (OPTION_MASK_ISA2_AVXIFMA_UNSET | OPTION_MASK_ISA2_AVXVNNI_UNSET \
    | OPTION_MASK_ISA2_AVXVNNIINT8_UNSET | OPTION_MASK_ISA2_AVXNECONVERT_UNSET \
-   | OPTION_MASK_ISA2_AVX512F_UNSET)
+   | OPTION_MASK_ISA2_AVXVNNIINT16_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 \
@@ -301,6 +302,7 @@ along with GCC; see the file COPYING3.  If not see
 #define OPTION_MASK_ISA2_PREFETCHI_UNSET OPTION_MASK_ISA2_PREFETCHI
 #define OPTION_MASK_ISA2_RAOINT_UNSET OPTION_MASK_ISA2_RAOINT
 #define OPTION_MASK_ISA2_AMX_COMPLEX_UNSET OPTION_MASK_ISA2_AMX_COMPLEX
+#define OPTION_MASK_ISA2_AVXVNNIINT16_UNSET OPTION_MASK_ISA2_AVXVNNIINT16
 
 /* SSE4 includes both SSE4.1 and SSE4.2.  -mno-sse4 should the same
    as -mno-sse4.1. */
@@ -1268,6 +1270,24 @@ ix86_handle_option (struct gcc_options *opts,
 	}
       return true;
 
+    case OPT_mavxvnniint16:
+      if (value)
+	{
+	  opts->x_ix86_isa_flags2 |= OPTION_MASK_ISA2_AVXVNNIINT16_SET;
+	  opts->x_ix86_isa_flags2_explicit |=
+	    OPTION_MASK_ISA2_AVXVNNIINT16_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_AVXVNNIINT16_UNSET;
+	  opts->x_ix86_isa_flags2_explicit |=
+	    OPTION_MASK_ISA2_AVXVNNIINT16_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 254dfec70e5..ae4e6a02f7f 100644
--- a/gcc/common/config/i386/i386-cpuinfo.h
+++ b/gcc/common/config/i386/i386-cpuinfo.h
@@ -255,6 +255,7 @@ enum processor_features
   FEATURE_PREFETCHI,
   FEATURE_RAOINT,
   FEATURE_AMX_COMPLEX,
+  FEATURE_AVXVNNIINT16,
   CPU_FEATURE_MAX
 };
 
diff --git a/gcc/common/config/i386/i386-isas.h b/gcc/common/config/i386/i386-isas.h
index d4b0b23b417..fc6abdedf24 100644
--- a/gcc/common/config/i386/i386-isas.h
+++ b/gcc/common/config/i386/i386-isas.h
@@ -186,4 +186,6 @@ ISA_NAMES_TABLE_START
   ISA_NAMES_TABLE_ENTRY("raoint", FEATURE_RAOINT, P_NONE, "-mraoint")
   ISA_NAMES_TABLE_ENTRY("amx-complex", FEATURE_AMX_COMPLEX,
 			P_NONE, "-mamx-complex")
+  ISA_NAMES_TABLE_ENTRY("avxvnniint16", FEATURE_AVXVNNIINT16,
+			P_NONE, "-mavxvnniint16")
 ISA_NAMES_TABLE_END
diff --git a/gcc/config.gcc b/gcc/config.gcc
index 1446eb2b3ca..fc74d776048 100644
--- a/gcc/config.gcc
+++ b/gcc/config.gcc
@@ -435,7 +435,7 @@ i[34567]86-*-* | x86_64-*-*)
 		       mwaitintrin.h avx512fp16intrin.h avx512fp16vlintrin.h
 		       avxifmaintrin.h avxvnniint8intrin.h avxneconvertintrin.h
 		       cmpccxaddintrin.h amxfp16intrin.h prfchiintrin.h
-		       raointintrin.h amxcomplexintrin.h"
+		       raointintrin.h amxcomplexintrin.h avxvnniint16intrin.h"
 	;;
 ia64-*-*)
 	extra_headers=ia64intrin.h
diff --git a/gcc/config/i386/avxvnniint16intrin.h b/gcc/config/i386/avxvnniint16intrin.h
new file mode 100644
index 00000000000..f87d76cb9f3
--- /dev/null
+++ b/gcc/config/i386/avxvnniint16intrin.h
@@ -0,0 +1,138 @@
+/* Copyright (C) 2023 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 <avxvnniint16intrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef _AVXVNNIINT16INTRIN_H_INCLUDED
+#define _AVXVNNIINT16INTRIN_H_INCLUDED
+
+#if !defined(__AVXVNNIINT16__)
+#pragma GCC push_options
+#pragma GCC target("avxvnniint16")
+#define __DISABLE_AVXVNNIINT16__
+#endif /* __AVXVNNIINT16__ */
+
+extern __inline __m128i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_dpwsud_avx_epi32 (__m128i __W, __m128i __A, __m128i __B)
+{
+  return (__m128i)
+    __builtin_ia32_vpdpwsud128 ((__v4si) __W, (__v4si) __A, (__v4si) __B);
+}
+
+extern __inline __m128i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_dpwsuds_avx_epi32 (__m128i __W, __m128i __A, __m128i __B)
+{
+  return (__m128i)
+    __builtin_ia32_vpdpwsuds128 ((__v4si) __W, (__v4si) __A, (__v4si) __B);
+}
+
+extern __inline __m128i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_dpwusd_avx_epi32 (__m128i __W, __m128i __A, __m128i __B)
+{
+  return (__m128i)
+    __builtin_ia32_vpdpwusd128 ((__v4si) __W, (__v4si) __A, (__v4si) __B);
+}
+
+extern __inline __m128i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_dpwusds_avx_epi32 (__m128i __W, __m128i __A, __m128i __B)
+{
+  return (__m128i)
+    __builtin_ia32_vpdpwusds128 ((__v4si) __W, (__v4si) __A, (__v4si) __B);
+}
+
+extern __inline __m128i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_dpwuud_avx_epi32 (__m128i __W, __m128i __A, __m128i __B)
+{
+  return (__m128i)
+    __builtin_ia32_vpdpwuud128 ((__v4si) __W, (__v4si) __A, (__v4si) __B);
+}
+
+extern __inline __m128i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_dpwuuds_avx_epi32 (__m128i __W, __m128i __A, __m128i __B)
+{
+  return (__m128i)
+    __builtin_ia32_vpdpwuuds128 ((__v4si) __W, (__v4si) __A, (__v4si) __B);
+}
+
+extern __inline __m256i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_dpwsud_avx_epi32 (__m256i __W, __m256i __A, __m256i __B)
+{
+  return (__m256i)
+    __builtin_ia32_vpdpwsud256 ((__v8si) __W, (__v8si) __A, (__v8si) __B);
+}
+
+extern __inline __m256i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_dpwsuds_avx_epi32 (__m256i __W, __m256i __A, __m256i __B)
+{
+  return (__m256i)
+    __builtin_ia32_vpdpwsuds256 ((__v8si) __W, (__v8si) __A, (__v8si) __B);
+}
+
+extern __inline __m256i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_dpwusd_avx_epi32 (__m256i __W, __m256i __A, __m256i __B)
+{
+  return (__m256i)
+    __builtin_ia32_vpdpwusd256 ((__v8si) __W, (__v8si) __A, (__v8si) __B);
+}
+
+extern __inline __m256i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_dpwusds_avx_epi32 (__m256i __W, __m256i __A, __m256i __B)
+{
+  return (__m256i)
+    __builtin_ia32_vpdpwusds256 ((__v8si) __W, (__v8si) __A, (__v8si) __B);
+}
+
+extern __inline __m256i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_dpwuud_avx_epi32 (__m256i __W, __m256i __A, __m256i __B)
+{
+  return (__m256i)
+    __builtin_ia32_vpdpwuud256 ((__v8si) __W, (__v8si) __A, (__v8si) __B);
+}
+
+extern __inline __m256i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_dpwuuds_avx_epi32 (__m256i __W, __m256i __A, __m256i __B)
+{
+  return (__m256i)
+    __builtin_ia32_vpdpwuuds256 ((__v8si) __W, (__v8si) __A, (__v8si) __B);
+}
+
+#ifdef __DISABLE_AVXVNNIINT16__
+#undef __DISABLE_AVXVNNIINT16__
+#pragma GCC pop_options
+#endif /* __DISABLE_AVXVNNIINT16__ */
+
+#endif /* __AVXVNNIINT16INTRIN_H_INCLUDED */
diff --git a/gcc/config/i386/cpuid.h b/gcc/config/i386/cpuid.h
index 4cc44615cf6..98d0f193d22 100644
--- a/gcc/config/i386/cpuid.h
+++ b/gcc/config/i386/cpuid.h
@@ -144,6 +144,7 @@
 /* %edx */
 #define bit_AVXVNNIINT8 (1 << 4)
 #define bit_AVXNECONVERT	(1 << 5)
+#define bit_AVXVNNIINT16	(1 << 10)
 #define bit_PREFETCHI	(1 << 14)
 
 /* Extended State Enumeration Sub-leaf (%eax == 0xd, %ecx == 1) */
diff --git a/gcc/config/i386/i386-builtin.def b/gcc/config/i386/i386-builtin.def
index 7ba5b6a9d11..ff5b3dcbcd3 100644
--- a/gcc/config/i386/i386-builtin.def
+++ b/gcc/config/i386/i386-builtin.def
@@ -2740,6 +2740,20 @@ BDESC (0, OPTION_MASK_ISA2_AVXVNNIINT8, CODE_FOR_vpdpbsuds_v4si, "__builtin_ia32
 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)
 
+/* AVXVNNIINT16 */
+BDESC (0, OPTION_MASK_ISA2_AVXVNNIINT16, CODE_FOR_vpdpwusd_v8si, "__builtin_ia32_vpdpwusd256", IX86_BUILTIN_VPDPWUSDV8SI, UNKNOWN, (int) V8SI_FTYPE_V8SI_V8SI_V8SI)
+BDESC (0, OPTION_MASK_ISA2_AVXVNNIINT16, CODE_FOR_vpdpwusds_v8si, "__builtin_ia32_vpdpwusds256", IX86_BUILTIN_VPDPWUSDSV8SI, UNKNOWN, (int) V8SI_FTYPE_V8SI_V8SI_V8SI)
+BDESC (0, OPTION_MASK_ISA2_AVXVNNIINT16, CODE_FOR_vpdpwsud_v8si, "__builtin_ia32_vpdpwsud256", IX86_BUILTIN_VPDPWSUDV8SI, UNKNOWN, (int) V8SI_FTYPE_V8SI_V8SI_V8SI)
+BDESC (0, OPTION_MASK_ISA2_AVXVNNIINT16, CODE_FOR_vpdpwsuds_v8si, "__builtin_ia32_vpdpwsuds256", IX86_BUILTIN_VPDPWSUDSV8SI, UNKNOWN, (int) V8SI_FTYPE_V8SI_V8SI_V8SI)
+BDESC (0, OPTION_MASK_ISA2_AVXVNNIINT16, CODE_FOR_vpdpwuud_v8si, "__builtin_ia32_vpdpwuud256", IX86_BUILTIN_VPDPWUUDV8SI, UNKNOWN, (int) V8SI_FTYPE_V8SI_V8SI_V8SI)
+BDESC (0, OPTION_MASK_ISA2_AVXVNNIINT16, CODE_FOR_vpdpwuuds_v8si, "__builtin_ia32_vpdpwuuds256", IX86_BUILTIN_VPDPWUUDSV8SI, UNKNOWN, (int) V8SI_FTYPE_V8SI_V8SI_V8SI)
+BDESC (0, OPTION_MASK_ISA2_AVXVNNIINT16, CODE_FOR_vpdpwusd_v4si, "__builtin_ia32_vpdpwusd128", IX86_BUILTIN_VPDPWUSDV4SI, UNKNOWN, (int) V4SI_FTYPE_V4SI_V4SI_V4SI)
+BDESC (0, OPTION_MASK_ISA2_AVXVNNIINT16, CODE_FOR_vpdpwusds_v4si, "__builtin_ia32_vpdpwusds128", IX86_BUILTIN_VPDPWUSDSV4SI, UNKNOWN, (int) V4SI_FTYPE_V4SI_V4SI_V4SI)
+BDESC (0, OPTION_MASK_ISA2_AVXVNNIINT16, CODE_FOR_vpdpwsud_v4si, "__builtin_ia32_vpdpwsud128", IX86_BUILTIN_VPDPWSUDV4SI, UNKNOWN, (int) V4SI_FTYPE_V4SI_V4SI_V4SI)
+BDESC (0, OPTION_MASK_ISA2_AVXVNNIINT16, CODE_FOR_vpdpwsuds_v4si, "__builtin_ia32_vpdpwsuds128", IX86_BUILTIN_VPDPWSUDSV4SI, UNKNOWN, (int) V4SI_FTYPE_V4SI_V4SI_V4SI)
+BDESC (0, OPTION_MASK_ISA2_AVXVNNIINT16, CODE_FOR_vpdpwuud_v4si, "__builtin_ia32_vpdpwuud128", IX86_BUILTIN_VPDPWUUDV4SI, UNKNOWN, (int) V4SI_FTYPE_V4SI_V4SI_V4SI)
+BDESC (0, OPTION_MASK_ISA2_AVXVNNIINT16, CODE_FOR_vpdpwuuds_v4si, "__builtin_ia32_vpdpwuuds128", IX86_BUILTIN_VPDPWUUDSV4SI, 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 e7bd7cc706c..d3514dd46ac 100644
--- a/gcc/config/i386/i386-c.cc
+++ b/gcc/config/i386/i386-c.cc
@@ -677,6 +677,8 @@ ix86_target_macros_internal (HOST_WIDE_INT isa_flag,
     def_or_undef (parse_in, "__RAOINT__");
   if (isa_flag2 & OPTION_MASK_ISA2_AMX_COMPLEX)
     def_or_undef (parse_in, "__AMX_COMPLEX__");
+  if (isa_flag2 & OPTION_MASK_ISA2_AVXVNNIINT16)
+    def_or_undef (parse_in, "__AVXVNNIINT16__");
   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 0634c6f5bac..fbf22f7270a 100644
--- a/gcc/config/i386/i386-isa.def
+++ b/gcc/config/i386/i386-isa.def
@@ -117,3 +117,4 @@ DEF_PTA(AMX_FP16)
 DEF_PTA(PREFETCHI)
 DEF_PTA(RAOINT)
 DEF_PTA(AMX_COMPLEX)
+DEF_PTA(AVXVNNIINT16)
diff --git a/gcc/config/i386/i386-options.cc b/gcc/config/i386/i386-options.cc
index 37cb5a0dcc4..d981666dd87 100644
--- a/gcc/config/i386/i386-options.cc
+++ b/gcc/config/i386/i386-options.cc
@@ -239,7 +239,8 @@ static struct ix86_target_opts isa2_opts[] =
   { "-mamx-fp16",       OPTION_MASK_ISA2_AMX_FP16 },
   { "-mprefetchi",      OPTION_MASK_ISA2_PREFETCHI },
   { "-mraoint", 	OPTION_MASK_ISA2_RAOINT },
-  { "-mamx-complex",	OPTION_MASK_ISA2_AMX_COMPLEX }
+  { "-mamx-complex",	OPTION_MASK_ISA2_AMX_COMPLEX },
+  { "-mavxvnniint16",	OPTION_MASK_ISA2_AVXVNNIINT16 },
 };
 static struct ix86_target_opts isa_opts[] =
 {
@@ -1091,6 +1092,7 @@ ix86_valid_target_attribute_inner_p (tree fndecl, tree args, char *p_strings[],
     IX86_ATTR_ISA ("prefetchi",   OPT_mprefetchi),
     IX86_ATTR_ISA ("raoint", OPT_mraoint),
     IX86_ATTR_ISA ("amx-complex", OPT_mamx_complex),
+    IX86_ATTR_ISA ("avxvnniint16", OPT_mavxvnniint16),
 
     /* enum options */
     IX86_ATTR_ENUM ("fpmath=",	OPT_mfpmath_),
diff --git a/gcc/config/i386/i386.opt b/gcc/config/i386/i386.opt
index d74f6b1f8fc..618d713530f 100644
--- a/gcc/config/i386/i386.opt
+++ b/gcc/config/i386/i386.opt
@@ -1278,3 +1278,8 @@ Enum(lam_type) String(u57) Value(lam_u57)
 mamx-complex
 Target Mask(ISA2_AMX_COMPLEX) Var(ix86_isa_flags2) Save
 Support AMX-COMPLEX built-in functions and code generation.
+
+mavxvnniint16
+Target Mask(ISA2_AVXVNNIINT16) Var(ix86_isa_flags2) Save
+Support MMX, SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2, AVX, AVX2 and
+AVXVNNIINT16 built-in functions and code generation.
diff --git a/gcc/config/i386/immintrin.h b/gcc/config/i386/immintrin.h
index b220d871942..52dc35d8398 100644
--- a/gcc/config/i386/immintrin.h
+++ b/gcc/config/i386/immintrin.h
@@ -48,6 +48,8 @@
 
 #include <avxvnniint8intrin.h>
 
+#include <avxvnniint16intrin.h>
+
 #include <avx2intrin.h>
 
 #include <avx512fintrin.h>
diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md
index 6bf9c99a2c1..85a5f801e7a 100644
--- a/gcc/config/i386/sse.md
+++ b/gcc/config/i386/sse.md
@@ -204,6 +204,14 @@
   UNSPEC_VPDPBSUDS
   UNSPEC_VPDPBUUD
   UNSPEC_VPDPBUUDS
+
+  ;; For AVX-VNNI-INT16 support
+  UNSPEC_VPDPWUSD
+  UNSPEC_VPDPWUSDS
+  UNSPEC_VPDPWSUD
+  UNSPEC_VPDPWSUDS
+  UNSPEC_VPDPWUUD
+  UNSPEC_VPDPWUUDS
 ])
 
 (define_c_enum "unspecv" [
@@ -30209,3 +30217,27 @@
   "vcvtneo<bf16_ph>2ps\t{%1, %0|%0, %1}"
   [(set_attr "prefix" "vex")
    (set_attr "mode" "<sseinsnmode>")])
+
+(define_int_iterator VPDPWPROD
+  [UNSPEC_VPDPWUSD
+   UNSPEC_VPDPWUSDS
+   UNSPEC_VPDPWSUD
+   UNSPEC_VPDPWSUDS
+   UNSPEC_VPDPWUUD
+   UNSPEC_VPDPWUUDS])
+
+(define_int_attr vpdpwprodtype
+  [(UNSPEC_VPDPWUSD "wusd") (UNSPEC_VPDPWUSDS "wusds")
+   (UNSPEC_VPDPWSUD "wsud") (UNSPEC_VPDPWSUDS "wsuds")
+   (UNSPEC_VPDPWUUD "wuud") (UNSPEC_VPDPWUUDS "wuuds")])
+
+(define_insn "vpdp<vpdpwprodtype>_<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")]
+	  VPDPWPROD))]
+  "TARGET_AVXVNNIINT16"
+  "vpdp<vpdpwprodtype>\t{%3, %2, %0|%0, %2, %3}"
+   [(set_attr "prefix" "vex")])
diff --git a/gcc/doc/extend.texi b/gcc/doc/extend.texi
index 7e4c5541b11..565bf1352e2 100644
--- a/gcc/doc/extend.texi
+++ b/gcc/doc/extend.texi
@@ -7163,6 +7163,11 @@ Enable/disable the generation of the RAOINT instructions.
 @itemx no-amx-complex
 Enable/disable the generation of the AMX-COMPLEX instructions.
 
+@cindex @code{target("avxvnniint16")} function attribute, x86
+@item avxvnniint16
+@itemx no-avxvnniint16
+Enable/disable the generation of the AVXVNNIINT16 instructions.
+
 @cindex @code{target("cld")} function attribute, x86
 @item cld
 @itemx no-cld
diff --git a/gcc/doc/invoke.texi b/gcc/doc/invoke.texi
index cbc1282c274..359887db5fd 100644
--- a/gcc/doc/invoke.texi
+++ b/gcc/doc/invoke.texi
@@ -1433,7 +1433,7 @@ See RS/6000 and PowerPC Options.
 -mrdseed  -msgx -mavx512vp2intersect -mserialize -mtsxldtrk
 -mamx-tile  -mamx-int8  -mamx-bf16 -muintr -mhreset -mavxvnni
 -mavx512fp16 -mavxifma -mavxvnniint8 -mavxneconvert -mcmpccxadd -mamx-fp16
--mprefetchi -mraoint -mamx-complex
+-mprefetchi -mraoint -mamx-complex -mavxvnniint16
 -mcldemote  -mms-bitfields  -mno-align-stringops  -minline-all-stringops
 -minline-stringops-dynamically  -mstringop-strategy=@var{alg}
 -mkl -mwidekl
@@ -33552,8 +33552,10 @@ preferred alignment to @option{-mpreferred-stack-boundary=2}.
 @need 200
 @opindex mamx-complex
 @itemx -mamx-complex
+@need 200
+@opindex mavxvnniint16
+@itemx -mavxvnniint16
 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,
 AES, PCLMUL, CLFLUSHOPT, CLWB, FSGSBASE, PTWRITE, RDRND, F16C, FMA, PCONFIG,
 WBNOINVD, FMA4, PREFETCHW, RDPID, PREFETCHWT1, RDSEED, SGX, XOP, LWP,
@@ -33563,8 +33565,8 @@ GFNI, VAES, WAITPKG, VPCLMULQDQ, AVX512BITALG, MOVDIRI, MOVDIR64B, AVX512BF16,
 ENQCMD, AVX512VPOPCNTDQ, AVX5124FMAPS, AVX512VNNI, AVX5124VNNIW, SERIALIZE,
 UINTR, HRESET, AMXTILE, AMXINT8, AMXBF16, KL, WIDEKL, AVXVNNI, AVX512-FP16,
 AVXIFMA, AVXVNNIINT8, AVXNECONVERT, CMPCCXADD, AMX-FP16, PREFETCHI, RAOINT,
-AMX-COMPLEX or CLDEMOTE extended instruction sets. Each has a corresponding
-@option{-mno-} option to disable use of these instructions.
+AMX-COMPLEX, AVXVNNIINT16 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 ffb6eb1a48c..40919b30a62 100644
--- a/gcc/doc/sourcebuild.texi
+++ b/gcc/doc/sourcebuild.texi
@@ -2511,6 +2511,9 @@ Target supports the execution of @code{avxneconvert} instructions.
 @item avxvnniint8
 Target supports the execution of @code{avxvnniint8} instructions.
 
+@item avxvnniint16
+Target supports the execution of @code{avxvnniint16} 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 6fe07e18fc6..53622df2bb8 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 -mavxvnniint8 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex" } */
+/* { 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 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16" } */
 
 /* 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 55c81677300..3b76cee3af8 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 -mavxvnniint8 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex" } */
+/* { 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 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16" } */
 
 /* 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 666eff50780..3d417ea1ed5 100644
--- a/gcc/testsuite/gcc.target/i386/avx-check.h
+++ b/gcc/testsuite/gcc.target/i386/avx-check.h
@@ -31,6 +31,9 @@ main ()
 #endif
 #ifdef AVXNECONVERT
       && __builtin_cpu_supports ("avxneconvert")
+#endif
+#ifdef AVXVNNIINT16
+      && __builtin_cpu_supports ("avxvnniint16")
 #endif
       )
     {
diff --git a/gcc/testsuite/gcc.target/i386/avxvnniint16-1.c b/gcc/testsuite/gcc.target/i386/avxvnniint16-1.c
new file mode 100644
index 00000000000..6ae57b150fe
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/avxvnniint16-1.c
@@ -0,0 +1,43 @@
+/* { dg-do compile } */
+/* { dg-options "-mavxvnniint16 -O2" } */
+/* { dg-final { scan-assembler-times "vpdpwusd\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\r]*%ymm\[0-9\]+\[^\n\r]*%ymm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
+/* { dg-final { scan-assembler-times "vpdpwusd\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
+/* { dg-final { scan-assembler-times "vpdpwusds\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\r]*%ymm\[0-9\]+\[^\n\r]*%ymm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
+/* { dg-final { scan-assembler-times "vpdpwusds\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
+/* { dg-final { scan-assembler-times "vpdpwsud\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\r]*%ymm\[0-9\]+\[^\n\r]*%ymm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
+/* { dg-final { scan-assembler-times "vpdpwsud\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
+/* { dg-final { scan-assembler-times "vpdpwsuds\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\r]*%ymm\[0-9\]+\[^\n\r]*%ymm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
+/* { dg-final { scan-assembler-times "vpdpwsuds\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
+/* { dg-final { scan-assembler-times "vpdpwuud\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\r]*%ymm\[0-9\]+\[^\n\r]*%ymm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
+/* { dg-final { scan-assembler-times "vpdpwuud\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
+/* { dg-final { scan-assembler-times "vpdpwuuds\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\r]*%ymm\[0-9\]+\[^\n\r]*%ymm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
+/* { dg-final { scan-assembler-times "vpdpwuuds\[ \\t\]+\[^\{\n\]*%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
+avxvnniint16_test (void)
+{
+  x = _mm256_dpwusd_avx_epi32 (x, y, z);
+  x_ = _mm_dpwusd_avx_epi32 (x_, y_, z_);
+
+  x = _mm256_dpwusds_avx_epi32 (x, y, z);
+  x_ = _mm_dpwusds_avx_epi32 (x_, y_, z_);
+
+  x = _mm256_dpwsud_avx_epi32 (x, y, z);
+  x_ = _mm_dpwsud_avx_epi32 (x_, y_, z_);
+
+  x = _mm256_dpwsuds_avx_epi32 (x, y, z);
+  x_ = _mm_dpwsuds_avx_epi32 (x_, y_, z_);
+
+  x = _mm256_dpwuud_avx_epi32 (x, y, z);
+  x_ = _mm_dpwuud_avx_epi32 (x_, y_, z_);
+
+  x = _mm256_dpwuuds_avx_epi32 (x, y, z);
+  x_ = _mm_dpwuuds_avx_epi32 (x_, y_, z_);
+}
diff --git a/gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwsud-2.c b/gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwsud-2.c
new file mode 100644
index 00000000000..bc57a8a8078
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwsud-2.c
@@ -0,0 +1,71 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -mavxvnniint16" } */
+/* { dg-require-effective-target avxvnniint16 } */
+#define AVXVNNIINT16
+#ifndef CHECK
+#define CHECK "avx-check.h"
+#endif
+
+#ifndef TEST
+#define TEST avx_test
+#endif
+
+#include CHECK
+
+static void
+CALC (int *r, int *dst, short *s1, unsigned short *s2, int size)
+{
+  int tempres[16];
+  for (int i = 0; i < size; i++)
+    tempres[i] = (int) s1[i] * (unsigned int) s2[i];
+  for (int i = 0; i < size / 2; i++)
+  {
+    int test = (int) dst[i] + tempres[i * 2] + tempres[i * 2 + 1];
+    r[i] = test;
+  }
+}
+
+void
+TEST (void)
+{
+  int i;
+  union256i_d res_256;
+  union256i_w src1_256;
+  union256i_uw src2_256;
+  int res_ref_256[8];
+
+  for (i = 0; i < 16; 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, 16);
+  res_256.x = _mm256_dpwsud_avx_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_w src1_128;
+  union128i_uw src2_128;
+  int res_ref_128[4];
+
+  for (i = 0; i < 8; 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, 8);
+  res_128.x = _mm_dpwsud_avx_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/avxvnniint16-vpdpwsuds-2.c b/gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwsuds-2.c
new file mode 100644
index 00000000000..fbcf46a4827
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwsuds-2.c
@@ -0,0 +1,72 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -mavxvnniint16" } */
+/* { dg-require-effective-target avxvnniint16 } */
+#define AVXVNNIINT16
+#ifndef CHECK
+#define CHECK "avx-check.h"
+#endif
+
+#ifndef TEST
+#define TEST avx_test
+#endif
+
+#include CHECK
+
+static void
+CALC (int *r, int *dst, short *s1, unsigned short *s2, int size)
+{
+  int tempres[16];
+  for (int i = 0; i < size; i++)
+    tempres[i] = (int) s1[i] * (unsigned int) s2[i];
+  for (int i = 0; i < size / 2; i++)
+  {
+    long long test = (long long) dst[i] + tempres[i * 2] + tempres[i * 2 + 1];
+    r[i] = test > 0x7FFFFFFF ? 0x7FFFFFFF : 
+           (test < 0xffffffff80000000LL ? 0x80000000 : test);
+  }
+}
+
+void
+TEST (void)
+{
+  int i;
+  union256i_d res_256;
+  union256i_w src1_256;
+  union256i_uw src2_256;
+  int res_ref_256[8];
+
+  for (i = 0; i < 16; 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, 16);
+  res_256.x = _mm256_dpwsuds_avx_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_w src1_128;
+  union128i_uw src2_128;
+  int res_ref_128[4];
+
+  for (i = 0; i < 8; 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, 8);
+  res_128.x = _mm_dpwsuds_avx_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/avxvnniint16-vpdpwusd-2.c b/gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwusd-2.c
new file mode 100644
index 00000000000..54cf271acbf
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwusd-2.c
@@ -0,0 +1,71 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -mavxvnniint16" } */
+/* { dg-require-effective-target avxvnniint16 } */
+#define AVXVNNIINT16
+#ifndef CHECK
+#define CHECK "avx-check.h"
+#endif
+
+#ifndef TEST
+#define TEST avx_test
+#endif
+
+#include CHECK
+
+static void
+CALC (int *r, int *dst, unsigned short *s1, short *s2, int size)
+{
+  int tempres[16];
+  for (int i = 0; i < size; i++)
+    tempres[i] = (unsigned int) s1[i] * (int) s2[i];
+  for (int i = 0; i < size / 2; i++)
+  {
+    int test = (int) dst[i] + tempres[i * 2] + tempres[i * 2 + 1];
+    r[i] = test;
+  }
+}
+
+void
+TEST (void)
+{
+  int i;
+  union256i_d res_256;
+  union256i_uw src1_256;
+  union256i_w src2_256;
+  int res_ref_256[8];
+
+  for (i = 0; i < 16; 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, 16);
+  res_256.x = _mm256_dpwusd_avx_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_uw src1_128;
+  union128i_w src2_128;
+  int res_ref_128[4];
+
+  for (i = 0; i < 8; 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, 8);
+  res_128.x = _mm_dpwusd_avx_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/avxvnniint16-vpdpwusds-2.c b/gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwusds-2.c
new file mode 100644
index 00000000000..ed9594c708b
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwusds-2.c
@@ -0,0 +1,72 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -mavxvnniint16" } */
+/* { dg-require-effective-target avxvnniint16 } */
+#define AVXVNNIINT16
+#ifndef CHECK
+#define CHECK "avx-check.h"
+#endif
+
+#ifndef TEST
+#define TEST avx_test
+#endif
+
+#include CHECK
+
+static void
+CALC (int *r, int *dst, unsigned short *s1, short *s2, int size)
+{
+  int tempres[16];
+  for (int i = 0; i < size; i++)
+    tempres[i] = (unsigned int) s1[i] * (int) s2[i];
+  for (int i = 0; i < size / 2; i++)
+  {
+    long long test = (long long) dst[i] + tempres[i * 2] + tempres[i * 2 + 1];
+    r[i] = test > 0x7FFFFFFF ? 0x7FFFFFFF : 
+           (test < 0xffffffff80000000LL ? 0x80000000 : test);
+  }
+}
+
+void
+TEST (void)
+{
+  int i;
+  union256i_d res_256;
+  union256i_uw src1_256;
+  union256i_w src2_256;
+  int res_ref_256[8];
+
+  for (i = 0; i < 16; 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, 16);
+  res_256.x = _mm256_dpwusds_avx_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_uw src1_128;
+  union128i_w src2_128;
+  int res_ref_128[4];
+
+  for (i = 0; i < 8; 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, 8);
+  res_128.x = _mm_dpwusds_avx_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/avxvnniint16-vpdpwuud-2.c b/gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwuud-2.c
new file mode 100644
index 00000000000..8bdc4330ef0
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwuud-2.c
@@ -0,0 +1,71 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -mavxvnniint16" } */
+/* { dg-require-effective-target avxvnniint16 } */
+#define AVXVNNIINT16
+#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 short *s1, unsigned short *s2, int size)
+{
+  unsigned int tempres[16];
+  for (int i = 0; i < size; i++)
+    tempres[i] = (unsigned int) s1[i] * (unsigned int) s2[i];
+  for (int i = 0; i < size / 2; i++)
+  {
+    unsigned int test = (unsigned) dst[i] + tempres[i * 2] + tempres[i * 2 + 1];
+    r[i] = test;
+  }
+}
+
+void
+TEST (void)
+{
+  int i;
+  union256i_ud res_256;
+  union256i_uw src1_256;
+  union256i_uw src2_256;
+  unsigned int res_ref_256[8];
+
+  for (i = 0; i < 16; 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, 16);
+  res_256.x = _mm256_dpwuud_avx_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_uw src2_128;
+  union128i_uw src1_128;
+  unsigned int res_ref_128[4];
+
+  for (i = 0; i < 8; 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, 8);
+  res_128.x = _mm_dpwuud_avx_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/avxvnniint16-vpdpwuuds-2.c b/gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwuuds-2.c
new file mode 100644
index 00000000000..32204122f1c
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwuuds-2.c
@@ -0,0 +1,71 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -mavxvnniint16" } */
+/* { dg-require-effective-target avxvnniint16 } */
+#define AVXVNNIINT16
+#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 short *s1, unsigned short *s2, int size)
+{
+  unsigned int tempres[16];
+  for (int i = 0; i < size; i++)
+    tempres[i] = (unsigned int) s1[i] * (unsigned int) s2[i];
+  for (int i = 0; i < size / 2; i++)
+  {
+    unsigned int test = (unsigned) dst[i] + tempres[i * 2] + tempres[i * 2 + 1];
+    r[i] = test > 0xFFFFFFFF ? 0xFFFFFFFF : test;
+  }
+}
+
+void
+TEST (void)
+{
+  int i;
+  union256i_ud res_256;
+  union256i_uw src1_256;
+  union256i_uw src2_256;
+  unsigned int res_ref_256[8];
+
+  for (i = 0; i < 16; 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, 16);
+  res_256.x = _mm256_dpwuuds_avx_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_uw src2_128;
+  union128i_uw src1_128;
+  unsigned int res_ref_128[4];
+
+  for (i = 0; i < 8; 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, 8);
+  res_128.x = _mm_dpwuuds_avx_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 f466962c36c..bba0fa37efd 100644
--- a/gcc/testsuite/gcc.target/i386/funcspec-56.inc
+++ b/gcc/testsuite/gcc.target/i386/funcspec-56.inc
@@ -88,6 +88,7 @@ extern void test_amx_fp16 (void)		__attribute__((__target__("amx-fp16")));
 extern void test_prefetchi (void)               __attribute__((__target__("prefetchi")));
 extern void test_raoint (void)                  __attribute__((__target__("raoint")));
 extern void test_amx_complex (void)		__attribute__((__target__("amx-complex")));
+extern void test_avxvnniint16 (void)		__attribute__((__target__("avxvnniint16")));
 
 extern void test_no_sgx (void)			__attribute__((__target__("no-sgx")));
 extern void test_no_avx5124fmaps(void)		__attribute__((__target__("no-avx5124fmaps")));
@@ -177,6 +178,7 @@ extern void test_no_amx_fp16 (void)		__attribute__((__target__("no-amx-fp16")));
 extern void test_no_prefetchi (void)            __attribute__((__target__("no-prefetchi")));
 extern void test_no_raoint (void)               __attribute__((__target__("no-raoint")));
 extern void test_no_amx_complex (void)		__attribute__((__target__("no-amx-complex")));
+extern void test_no_avxvnniint16 (void)		__attribute__((__target__("no-avxvnniint16")));
 
 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 ae4ffd1975f..2b7d78c51d3 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 -mavxvnniint8 -mavxneconvert -mamx-fp16 -mraoint -mamx-complex" } */
+/* { 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 -mavxneconvert -mamx-fp16 -mraoint -mamx-complex -mavxvnniint16" } */
 
 #include <x86intrin.h>
 
diff --git a/gcc/testsuite/gcc.target/i386/sse-13.c b/gcc/testsuite/gcc.target/i386/sse-13.c
index f046a68ddbb..33693484d2a 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 -mavxvnniint8 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex" } */
+/* { 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 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16" } */
 /* { 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 05322f7e914..51c2946b25a 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 -mavxvnniint8 -mavxneconvert -mamx-fp16 -mraoint -mamx-complex" } */
+/* { 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 -mavxneconvert -mamx-fp16 -mraoint -mamx-complex -mavxvnniint16" } */
 /* { 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 53c38b70241..4982fde2a76 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,tbm,lwp,fsgsbase,rdrnd,f16c,rtm,rdseed,prfchw,adx,fxsr,xsaveopt,avx512f,avx512er,avx512cd,avx512pf,sha,prefetchwt1,avx512vl,avx512bw,avx512dq,avx512vbmi,avx512vbmi2,avx512ifma,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,gfni,avx512bitalg,avx512bf16,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,amx-fp16,raoint,amx-complex")
+#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,avx512f,avx512er,avx512cd,avx512pf,sha,prefetchwt1,avx512vl,avx512bw,avx512dq,avx512vbmi,avx512vbmi2,avx512ifma,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,gfni,avx512bitalg,avx512bf16,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,amx-fp16,raoint,amx-complex,avxvnniint16")
 #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,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,gfni,avx512bitalg,avx512bf16,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,amx-fp16,raoint,amx-complex")
+#pragma GCC target ("avx,avx2,rdrnd,fsgsbase,f16c,rtm,avx512f,avx512er,avx512cd,avx512pf,sha,avx512vl,avx512bw,avx512dq,avx512ifma,avx512vbmi,avx512vbmi2,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,gfni,avx512bitalg,avx512bf16,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,amx-fp16,raoint,amx-complex,avxvnniint16")
 #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 50bf85a3392..7e9c9f2ca2b 100644
--- a/gcc/testsuite/gcc.target/i386/sse-23.c
+++ b/gcc/testsuite/gcc.target/i386/sse-23.c
@@ -847,6 +847,6 @@
 #define __builtin_ia32_cmpccxadd(A, B, C, D) __builtin_ia32_cmpccxadd(A, B, C, 1)
 #define __builtin_ia32_cmpccxadd64(A, B, C, D) __builtin_ia32_cmpccxadd64(A, B, C, 1)
 
-#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,avx512f,avx512er,avx512cd,avx512pf,sha,prefetchwt1,xsavec,xsaves,clflushopt,avx512bw,avx512dq,avx512vl,avx512vbmi,avx512ifma,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,clwb,mwaitx,clzero,pku,sgx,rdpid,gfni,avx512vbmi2,vpclmulqdq,avx512bitalg,pconfig,wbnoinvd,avx512bf16,enqcmd,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,cmpccxadd,amx-fp16,prefetchi,raoint,amx-complex")
+#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,avx512f,avx512er,avx512cd,avx512pf,sha,prefetchwt1,xsavec,xsaves,clflushopt,avx512bw,avx512dq,avx512vl,avx512vbmi,avx512ifma,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,clwb,mwaitx,clzero,pku,sgx,rdpid,gfni,avx512vbmi2,vpclmulqdq,avx512bitalg,pconfig,wbnoinvd,avx512bf16,enqcmd,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,cmpccxadd,amx-fp16,prefetchi,raoint,amx-complex,avxvnniint16")
 
 #include <x86intrin.h>
diff --git a/gcc/testsuite/lib/target-supports.exp b/gcc/testsuite/lib/target-supports.exp
index 33482b2fa9a..60de239f1ce 100644
--- a/gcc/testsuite/lib/target-supports.exp
+++ b/gcc/testsuite/lib/target-supports.exp
@@ -9856,6 +9856,18 @@ proc check_effective_target_amx_complex { } {
     } "-mamx-complex" ]
 }
 
+# Return 1 if avxvnniint16 instructions can be compiled.
+proc check_effective_target_avxvnniint16 { } {
+    return [check_no_compiler_messages avxvnniint16 object {
+	typedef int __v8si __attribute__ ((__vector_size__ (32)));
+	__v8si
+	_mm256_dpwsud_avx_epi32  (__v8si __A, __v8si __B, __v8si __C)
+        {
+	   return __builtin_ia32_vpdpwsud256 (__A, __B, __C);
+	}
+    } "-O0 -mavxvnniint16" ]
+}
+
 # Return 1 if sse instructions can be compiled.
 proc check_effective_target_sse { } {
     return [check_no_compiler_messages sse object {
-- 
2.31.1


^ permalink raw reply	[flat|nested] 9+ messages in thread

* Re: [PATCH 1/4] Support Intel AVX-VNNI-INT16
  2023-07-13  6:03 ` [PATCH 1/4] Support Intel AVX-VNNI-INT16 Haochen Jiang
@ 2023-07-17  1:39   ` Hongtao Liu
  0 siblings, 0 replies; 9+ messages in thread
From: Hongtao Liu @ 2023-07-17  1:39 UTC (permalink / raw)
  To: Haochen Jiang; +Cc: gcc-patches, hongtao.liu, ubizjak, Kong Lingling

On Thu, Jul 13, 2023 at 2:06 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
>         avxvnniint16.
>         * common/config/i386/i386-common.cc
>         (OPTION_MASK_ISA2_AVXVNNIINT16_SET): New.
>         (OPTION_MASK_ISA2_AVXVNNIINT16_UNSET): Ditto.
>         (ix86_handle_option): Handle -mavxvnniint16.
>         * common/config/i386/i386-cpuinfo.h (enum processor_features):
>         Add FEATURE_AVXVNNIINT16.
>         * common/config/i386/i386-isas.h: Add ISA_NAME_TABLE_ENTRY for
>         avxvnniint16.
>         * config.gcc: Add avxvnniint16.h.
>         * config/i386/avxvnniint16intrin.h: New file.
>         * config/i386/cpuid.h (bit_AVXVNNIINT16): New.
>         * config/i386/i386-builtin.def: Add new builtins.
>         * config/i386/i386-c.cc (ix86_target_macros_internal): Define
>         __AVXVNNIINT16__.
>         * config/i386/i386-options.cc (isa2_opts): Add -mavxvnniint16.
>         (ix86_valid_target_attribute_inner_p): Handle avxvnniint16intrin.h.
>         * config/i386/i386-isa.def: Add DEF_PTA(AVXVNNIINT16).
>         * config/i386/i386.opt: Add option -mavxvnniint16.
>         * config/i386/immintrin.h: Include avxvnniint16.h.
>         * config/i386/sse.md
>         (vpdp<vpdpwprodtype>_<mode>): New define_insn.
>         * doc/extend.texi: Document avxvnniint16.
>         * doc/invoke.texi: Document -mavxvnniint16.
>         * doc/sourcebuild.texi: Document target avxvnniint16.
Ok.
>
> gcc/testsuite/ChangeLog
>
>         * g++.dg/other/i386-2.C: Add -mavxvnniint16.
>         * g++.dg/other/i386-3.C: Ditto.
>         * gcc.target/i386/avx-check.h: Add avxvnniint16 check.
>         * gcc.target/i386/sse-12.c: Add -mavxvnniint16.
>         * 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_avxvnniint16): New.
>         * gcc.target/i386/avxvnniint16-1.c: Ditto.
>         * gcc.target/i386/avxvnniint16-vpdpwusd-2.c: Ditto.
>         * gcc.target/i386/avxvnniint16-vpdpwusds-2.c: Ditto.
>         * gcc.target/i386/avxvnniint16-vpdpwsud-2.c: Ditto.
>         * gcc.target/i386/avxvnniint16-vpdpwsuds-2.c: Ditto.
>         * gcc.target/i386/avxvnniint16-vpdpwuud-2.c: Ditto.
>         * gcc.target/i386/avxvnniint16-vpdpwuuds-2.c: Ditto.
>
> 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/avxvnniint16intrin.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                        |  32 ++++
>  gcc/doc/extend.texi                           |   5 +
>  gcc/doc/invoke.texi                           |  10 +-
>  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.target/i386/avxvnniint16-1.c          |  43 ++++++
>  .../gcc.target/i386/avxvnniint16-vpdpwsud-2.c |  71 +++++++++
>  .../i386/avxvnniint16-vpdpwsuds-2.c           |  72 +++++++++
>  .../gcc.target/i386/avxvnniint16-vpdpwusd-2.c |  71 +++++++++
>  .../i386/avxvnniint16-vpdpwusds-2.c           |  72 +++++++++
>  .../gcc.target/i386/avxvnniint16-vpdpwuud-2.c |  71 +++++++++
>  .../i386/avxvnniint16-vpdpwuuds-2.c           |  71 +++++++++
>  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, 735 insertions(+), 15 deletions(-)
>  create mode 100644 gcc/config/i386/avxvnniint16intrin.h
>  create mode 100644 gcc/testsuite/gcc.target/i386/avxvnniint16-1.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwsud-2.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwsuds-2.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwusd-2.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwusds-2.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwuud-2.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwuuds-2.c
>
> diff --git a/gcc/common/config/i386/cpuinfo.h b/gcc/common/config/i386/cpuinfo.h
> index 7c2565c1d93..3599f9def2c 100644
> --- a/gcc/common/config/i386/cpuinfo.h
> +++ b/gcc/common/config/i386/cpuinfo.h
> @@ -875,6 +875,8 @@ get_available_features (struct __processor_model *cpu_model,
>             set_feature (FEATURE_AVXVNNIINT8);
>           if (edx & bit_AVXNECONVERT)
>             set_feature (FEATURE_AVXNECONVERT);
> +         if (edx & bit_AVXVNNIINT16)
> +           set_feature (FEATURE_AVXVNNIINT16);
>         }
>        if (avx512_usable)
>         {
> diff --git a/gcc/common/config/i386/i386-common.cc b/gcc/common/config/i386/i386-common.cc
> index 8cea3669239..32c6d00580d 100644
> --- a/gcc/common/config/i386/i386-common.cc
> +++ b/gcc/common/config/i386/i386-common.cc
> @@ -119,6 +119,7 @@ along with GCC; see the file COPYING3.  If not see
>  #define OPTION_MASK_ISA2_RAOINT_SET OPTION_MASK_ISA2_RAOINT
>  #define OPTION_MASK_ISA2_AMX_COMPLEX_SET \
>    (OPTION_MASK_ISA2_AMX_TILE | OPTION_MASK_ISA2_AMX_COMPLEX)
> +#define OPTION_MASK_ISA2_AVXVNNIINT16_SET OPTION_MASK_ISA2_AVXVNNIINT16
>
>  /* SSE4 includes both SSE4.1 and SSE4.2. -msse4 should be the same
>     as -msse4.2.  */
> @@ -228,7 +229,7 @@ along with GCC; see the file COPYING3.  If not see
>  #define OPTION_MASK_ISA2_AVX2_UNSET \
>    (OPTION_MASK_ISA2_AVXIFMA_UNSET | OPTION_MASK_ISA2_AVXVNNI_UNSET \
>     | OPTION_MASK_ISA2_AVXVNNIINT8_UNSET | OPTION_MASK_ISA2_AVXNECONVERT_UNSET \
> -   | OPTION_MASK_ISA2_AVX512F_UNSET)
> +   | OPTION_MASK_ISA2_AVXVNNIINT16_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 \
> @@ -301,6 +302,7 @@ along with GCC; see the file COPYING3.  If not see
>  #define OPTION_MASK_ISA2_PREFETCHI_UNSET OPTION_MASK_ISA2_PREFETCHI
>  #define OPTION_MASK_ISA2_RAOINT_UNSET OPTION_MASK_ISA2_RAOINT
>  #define OPTION_MASK_ISA2_AMX_COMPLEX_UNSET OPTION_MASK_ISA2_AMX_COMPLEX
> +#define OPTION_MASK_ISA2_AVXVNNIINT16_UNSET OPTION_MASK_ISA2_AVXVNNIINT16
>
>  /* SSE4 includes both SSE4.1 and SSE4.2.  -mno-sse4 should the same
>     as -mno-sse4.1. */
> @@ -1268,6 +1270,24 @@ ix86_handle_option (struct gcc_options *opts,
>         }
>        return true;
>
> +    case OPT_mavxvnniint16:
> +      if (value)
> +       {
> +         opts->x_ix86_isa_flags2 |= OPTION_MASK_ISA2_AVXVNNIINT16_SET;
> +         opts->x_ix86_isa_flags2_explicit |=
> +           OPTION_MASK_ISA2_AVXVNNIINT16_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_AVXVNNIINT16_UNSET;
> +         opts->x_ix86_isa_flags2_explicit |=
> +           OPTION_MASK_ISA2_AVXVNNIINT16_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 254dfec70e5..ae4e6a02f7f 100644
> --- a/gcc/common/config/i386/i386-cpuinfo.h
> +++ b/gcc/common/config/i386/i386-cpuinfo.h
> @@ -255,6 +255,7 @@ enum processor_features
>    FEATURE_PREFETCHI,
>    FEATURE_RAOINT,
>    FEATURE_AMX_COMPLEX,
> +  FEATURE_AVXVNNIINT16,
>    CPU_FEATURE_MAX
>  };
>
> diff --git a/gcc/common/config/i386/i386-isas.h b/gcc/common/config/i386/i386-isas.h
> index d4b0b23b417..fc6abdedf24 100644
> --- a/gcc/common/config/i386/i386-isas.h
> +++ b/gcc/common/config/i386/i386-isas.h
> @@ -186,4 +186,6 @@ ISA_NAMES_TABLE_START
>    ISA_NAMES_TABLE_ENTRY("raoint", FEATURE_RAOINT, P_NONE, "-mraoint")
>    ISA_NAMES_TABLE_ENTRY("amx-complex", FEATURE_AMX_COMPLEX,
>                         P_NONE, "-mamx-complex")
> +  ISA_NAMES_TABLE_ENTRY("avxvnniint16", FEATURE_AVXVNNIINT16,
> +                       P_NONE, "-mavxvnniint16")
>  ISA_NAMES_TABLE_END
> diff --git a/gcc/config.gcc b/gcc/config.gcc
> index 1446eb2b3ca..fc74d776048 100644
> --- a/gcc/config.gcc
> +++ b/gcc/config.gcc
> @@ -435,7 +435,7 @@ i[34567]86-*-* | x86_64-*-*)
>                        mwaitintrin.h avx512fp16intrin.h avx512fp16vlintrin.h
>                        avxifmaintrin.h avxvnniint8intrin.h avxneconvertintrin.h
>                        cmpccxaddintrin.h amxfp16intrin.h prfchiintrin.h
> -                      raointintrin.h amxcomplexintrin.h"
> +                      raointintrin.h amxcomplexintrin.h avxvnniint16intrin.h"
>         ;;
>  ia64-*-*)
>         extra_headers=ia64intrin.h
> diff --git a/gcc/config/i386/avxvnniint16intrin.h b/gcc/config/i386/avxvnniint16intrin.h
> new file mode 100644
> index 00000000000..f87d76cb9f3
> --- /dev/null
> +++ b/gcc/config/i386/avxvnniint16intrin.h
> @@ -0,0 +1,138 @@
> +/* Copyright (C) 2023 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 <avxvnniint16intrin.h> directly; include <immintrin.h> instead."
> +#endif
> +
> +#ifndef _AVXVNNIINT16INTRIN_H_INCLUDED
> +#define _AVXVNNIINT16INTRIN_H_INCLUDED
> +
> +#if !defined(__AVXVNNIINT16__)
> +#pragma GCC push_options
> +#pragma GCC target("avxvnniint16")
> +#define __DISABLE_AVXVNNIINT16__
> +#endif /* __AVXVNNIINT16__ */
> +
> +extern __inline __m128i
> +__attribute__((__gnu_inline__, __always_inline__, __artificial__))
> +_mm_dpwsud_avx_epi32 (__m128i __W, __m128i __A, __m128i __B)
> +{
> +  return (__m128i)
> +    __builtin_ia32_vpdpwsud128 ((__v4si) __W, (__v4si) __A, (__v4si) __B);
> +}
> +
> +extern __inline __m128i
> +__attribute__((__gnu_inline__, __always_inline__, __artificial__))
> +_mm_dpwsuds_avx_epi32 (__m128i __W, __m128i __A, __m128i __B)
> +{
> +  return (__m128i)
> +    __builtin_ia32_vpdpwsuds128 ((__v4si) __W, (__v4si) __A, (__v4si) __B);
> +}
> +
> +extern __inline __m128i
> +__attribute__((__gnu_inline__, __always_inline__, __artificial__))
> +_mm_dpwusd_avx_epi32 (__m128i __W, __m128i __A, __m128i __B)
> +{
> +  return (__m128i)
> +    __builtin_ia32_vpdpwusd128 ((__v4si) __W, (__v4si) __A, (__v4si) __B);
> +}
> +
> +extern __inline __m128i
> +__attribute__((__gnu_inline__, __always_inline__, __artificial__))
> +_mm_dpwusds_avx_epi32 (__m128i __W, __m128i __A, __m128i __B)
> +{
> +  return (__m128i)
> +    __builtin_ia32_vpdpwusds128 ((__v4si) __W, (__v4si) __A, (__v4si) __B);
> +}
> +
> +extern __inline __m128i
> +__attribute__((__gnu_inline__, __always_inline__, __artificial__))
> +_mm_dpwuud_avx_epi32 (__m128i __W, __m128i __A, __m128i __B)
> +{
> +  return (__m128i)
> +    __builtin_ia32_vpdpwuud128 ((__v4si) __W, (__v4si) __A, (__v4si) __B);
> +}
> +
> +extern __inline __m128i
> +__attribute__((__gnu_inline__, __always_inline__, __artificial__))
> +_mm_dpwuuds_avx_epi32 (__m128i __W, __m128i __A, __m128i __B)
> +{
> +  return (__m128i)
> +    __builtin_ia32_vpdpwuuds128 ((__v4si) __W, (__v4si) __A, (__v4si) __B);
> +}
> +
> +extern __inline __m256i
> +__attribute__((__gnu_inline__, __always_inline__, __artificial__))
> +_mm256_dpwsud_avx_epi32 (__m256i __W, __m256i __A, __m256i __B)
> +{
> +  return (__m256i)
> +    __builtin_ia32_vpdpwsud256 ((__v8si) __W, (__v8si) __A, (__v8si) __B);
> +}
> +
> +extern __inline __m256i
> +__attribute__((__gnu_inline__, __always_inline__, __artificial__))
> +_mm256_dpwsuds_avx_epi32 (__m256i __W, __m256i __A, __m256i __B)
> +{
> +  return (__m256i)
> +    __builtin_ia32_vpdpwsuds256 ((__v8si) __W, (__v8si) __A, (__v8si) __B);
> +}
> +
> +extern __inline __m256i
> +__attribute__((__gnu_inline__, __always_inline__, __artificial__))
> +_mm256_dpwusd_avx_epi32 (__m256i __W, __m256i __A, __m256i __B)
> +{
> +  return (__m256i)
> +    __builtin_ia32_vpdpwusd256 ((__v8si) __W, (__v8si) __A, (__v8si) __B);
> +}
> +
> +extern __inline __m256i
> +__attribute__((__gnu_inline__, __always_inline__, __artificial__))
> +_mm256_dpwusds_avx_epi32 (__m256i __W, __m256i __A, __m256i __B)
> +{
> +  return (__m256i)
> +    __builtin_ia32_vpdpwusds256 ((__v8si) __W, (__v8si) __A, (__v8si) __B);
> +}
> +
> +extern __inline __m256i
> +__attribute__((__gnu_inline__, __always_inline__, __artificial__))
> +_mm256_dpwuud_avx_epi32 (__m256i __W, __m256i __A, __m256i __B)
> +{
> +  return (__m256i)
> +    __builtin_ia32_vpdpwuud256 ((__v8si) __W, (__v8si) __A, (__v8si) __B);
> +}
> +
> +extern __inline __m256i
> +__attribute__((__gnu_inline__, __always_inline__, __artificial__))
> +_mm256_dpwuuds_avx_epi32 (__m256i __W, __m256i __A, __m256i __B)
> +{
> +  return (__m256i)
> +    __builtin_ia32_vpdpwuuds256 ((__v8si) __W, (__v8si) __A, (__v8si) __B);
> +}
> +
> +#ifdef __DISABLE_AVXVNNIINT16__
> +#undef __DISABLE_AVXVNNIINT16__
> +#pragma GCC pop_options
> +#endif /* __DISABLE_AVXVNNIINT16__ */
> +
> +#endif /* __AVXVNNIINT16INTRIN_H_INCLUDED */
> diff --git a/gcc/config/i386/cpuid.h b/gcc/config/i386/cpuid.h
> index 4cc44615cf6..98d0f193d22 100644
> --- a/gcc/config/i386/cpuid.h
> +++ b/gcc/config/i386/cpuid.h
> @@ -144,6 +144,7 @@
>  /* %edx */
>  #define bit_AVXVNNIINT8 (1 << 4)
>  #define bit_AVXNECONVERT       (1 << 5)
> +#define bit_AVXVNNIINT16       (1 << 10)
>  #define bit_PREFETCHI  (1 << 14)
>
>  /* Extended State Enumeration Sub-leaf (%eax == 0xd, %ecx == 1) */
> diff --git a/gcc/config/i386/i386-builtin.def b/gcc/config/i386/i386-builtin.def
> index 7ba5b6a9d11..ff5b3dcbcd3 100644
> --- a/gcc/config/i386/i386-builtin.def
> +++ b/gcc/config/i386/i386-builtin.def
> @@ -2740,6 +2740,20 @@ BDESC (0, OPTION_MASK_ISA2_AVXVNNIINT8, CODE_FOR_vpdpbsuds_v4si, "__builtin_ia32
>  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)
>
> +/* AVXVNNIINT16 */
> +BDESC (0, OPTION_MASK_ISA2_AVXVNNIINT16, CODE_FOR_vpdpwusd_v8si, "__builtin_ia32_vpdpwusd256", IX86_BUILTIN_VPDPWUSDV8SI, UNKNOWN, (int) V8SI_FTYPE_V8SI_V8SI_V8SI)
> +BDESC (0, OPTION_MASK_ISA2_AVXVNNIINT16, CODE_FOR_vpdpwusds_v8si, "__builtin_ia32_vpdpwusds256", IX86_BUILTIN_VPDPWUSDSV8SI, UNKNOWN, (int) V8SI_FTYPE_V8SI_V8SI_V8SI)
> +BDESC (0, OPTION_MASK_ISA2_AVXVNNIINT16, CODE_FOR_vpdpwsud_v8si, "__builtin_ia32_vpdpwsud256", IX86_BUILTIN_VPDPWSUDV8SI, UNKNOWN, (int) V8SI_FTYPE_V8SI_V8SI_V8SI)
> +BDESC (0, OPTION_MASK_ISA2_AVXVNNIINT16, CODE_FOR_vpdpwsuds_v8si, "__builtin_ia32_vpdpwsuds256", IX86_BUILTIN_VPDPWSUDSV8SI, UNKNOWN, (int) V8SI_FTYPE_V8SI_V8SI_V8SI)
> +BDESC (0, OPTION_MASK_ISA2_AVXVNNIINT16, CODE_FOR_vpdpwuud_v8si, "__builtin_ia32_vpdpwuud256", IX86_BUILTIN_VPDPWUUDV8SI, UNKNOWN, (int) V8SI_FTYPE_V8SI_V8SI_V8SI)
> +BDESC (0, OPTION_MASK_ISA2_AVXVNNIINT16, CODE_FOR_vpdpwuuds_v8si, "__builtin_ia32_vpdpwuuds256", IX86_BUILTIN_VPDPWUUDSV8SI, UNKNOWN, (int) V8SI_FTYPE_V8SI_V8SI_V8SI)
> +BDESC (0, OPTION_MASK_ISA2_AVXVNNIINT16, CODE_FOR_vpdpwusd_v4si, "__builtin_ia32_vpdpwusd128", IX86_BUILTIN_VPDPWUSDV4SI, UNKNOWN, (int) V4SI_FTYPE_V4SI_V4SI_V4SI)
> +BDESC (0, OPTION_MASK_ISA2_AVXVNNIINT16, CODE_FOR_vpdpwusds_v4si, "__builtin_ia32_vpdpwusds128", IX86_BUILTIN_VPDPWUSDSV4SI, UNKNOWN, (int) V4SI_FTYPE_V4SI_V4SI_V4SI)
> +BDESC (0, OPTION_MASK_ISA2_AVXVNNIINT16, CODE_FOR_vpdpwsud_v4si, "__builtin_ia32_vpdpwsud128", IX86_BUILTIN_VPDPWSUDV4SI, UNKNOWN, (int) V4SI_FTYPE_V4SI_V4SI_V4SI)
> +BDESC (0, OPTION_MASK_ISA2_AVXVNNIINT16, CODE_FOR_vpdpwsuds_v4si, "__builtin_ia32_vpdpwsuds128", IX86_BUILTIN_VPDPWSUDSV4SI, UNKNOWN, (int) V4SI_FTYPE_V4SI_V4SI_V4SI)
> +BDESC (0, OPTION_MASK_ISA2_AVXVNNIINT16, CODE_FOR_vpdpwuud_v4si, "__builtin_ia32_vpdpwuud128", IX86_BUILTIN_VPDPWUUDV4SI, UNKNOWN, (int) V4SI_FTYPE_V4SI_V4SI_V4SI)
> +BDESC (0, OPTION_MASK_ISA2_AVXVNNIINT16, CODE_FOR_vpdpwuuds_v4si, "__builtin_ia32_vpdpwuuds128", IX86_BUILTIN_VPDPWUUDSV4SI, 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 e7bd7cc706c..d3514dd46ac 100644
> --- a/gcc/config/i386/i386-c.cc
> +++ b/gcc/config/i386/i386-c.cc
> @@ -677,6 +677,8 @@ ix86_target_macros_internal (HOST_WIDE_INT isa_flag,
>      def_or_undef (parse_in, "__RAOINT__");
>    if (isa_flag2 & OPTION_MASK_ISA2_AMX_COMPLEX)
>      def_or_undef (parse_in, "__AMX_COMPLEX__");
> +  if (isa_flag2 & OPTION_MASK_ISA2_AVXVNNIINT16)
> +    def_or_undef (parse_in, "__AVXVNNIINT16__");
>    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 0634c6f5bac..fbf22f7270a 100644
> --- a/gcc/config/i386/i386-isa.def
> +++ b/gcc/config/i386/i386-isa.def
> @@ -117,3 +117,4 @@ DEF_PTA(AMX_FP16)
>  DEF_PTA(PREFETCHI)
>  DEF_PTA(RAOINT)
>  DEF_PTA(AMX_COMPLEX)
> +DEF_PTA(AVXVNNIINT16)
> diff --git a/gcc/config/i386/i386-options.cc b/gcc/config/i386/i386-options.cc
> index 37cb5a0dcc4..d981666dd87 100644
> --- a/gcc/config/i386/i386-options.cc
> +++ b/gcc/config/i386/i386-options.cc
> @@ -239,7 +239,8 @@ static struct ix86_target_opts isa2_opts[] =
>    { "-mamx-fp16",       OPTION_MASK_ISA2_AMX_FP16 },
>    { "-mprefetchi",      OPTION_MASK_ISA2_PREFETCHI },
>    { "-mraoint",        OPTION_MASK_ISA2_RAOINT },
> -  { "-mamx-complex",   OPTION_MASK_ISA2_AMX_COMPLEX }
> +  { "-mamx-complex",   OPTION_MASK_ISA2_AMX_COMPLEX },
> +  { "-mavxvnniint16",  OPTION_MASK_ISA2_AVXVNNIINT16 },
>  };
>  static struct ix86_target_opts isa_opts[] =
>  {
> @@ -1091,6 +1092,7 @@ ix86_valid_target_attribute_inner_p (tree fndecl, tree args, char *p_strings[],
>      IX86_ATTR_ISA ("prefetchi",   OPT_mprefetchi),
>      IX86_ATTR_ISA ("raoint", OPT_mraoint),
>      IX86_ATTR_ISA ("amx-complex", OPT_mamx_complex),
> +    IX86_ATTR_ISA ("avxvnniint16", OPT_mavxvnniint16),
>
>      /* enum options */
>      IX86_ATTR_ENUM ("fpmath=", OPT_mfpmath_),
> diff --git a/gcc/config/i386/i386.opt b/gcc/config/i386/i386.opt
> index d74f6b1f8fc..618d713530f 100644
> --- a/gcc/config/i386/i386.opt
> +++ b/gcc/config/i386/i386.opt
> @@ -1278,3 +1278,8 @@ Enum(lam_type) String(u57) Value(lam_u57)
>  mamx-complex
>  Target Mask(ISA2_AMX_COMPLEX) Var(ix86_isa_flags2) Save
>  Support AMX-COMPLEX built-in functions and code generation.
> +
> +mavxvnniint16
> +Target Mask(ISA2_AVXVNNIINT16) Var(ix86_isa_flags2) Save
> +Support MMX, SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2, AVX, AVX2 and
> +AVXVNNIINT16 built-in functions and code generation.
> diff --git a/gcc/config/i386/immintrin.h b/gcc/config/i386/immintrin.h
> index b220d871942..52dc35d8398 100644
> --- a/gcc/config/i386/immintrin.h
> +++ b/gcc/config/i386/immintrin.h
> @@ -48,6 +48,8 @@
>
>  #include <avxvnniint8intrin.h>
>
> +#include <avxvnniint16intrin.h>
> +
>  #include <avx2intrin.h>
>
>  #include <avx512fintrin.h>
> diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md
> index 6bf9c99a2c1..85a5f801e7a 100644
> --- a/gcc/config/i386/sse.md
> +++ b/gcc/config/i386/sse.md
> @@ -204,6 +204,14 @@
>    UNSPEC_VPDPBSUDS
>    UNSPEC_VPDPBUUD
>    UNSPEC_VPDPBUUDS
> +
> +  ;; For AVX-VNNI-INT16 support
> +  UNSPEC_VPDPWUSD
> +  UNSPEC_VPDPWUSDS
> +  UNSPEC_VPDPWSUD
> +  UNSPEC_VPDPWSUDS
> +  UNSPEC_VPDPWUUD
> +  UNSPEC_VPDPWUUDS
>  ])
>
>  (define_c_enum "unspecv" [
> @@ -30209,3 +30217,27 @@
>    "vcvtneo<bf16_ph>2ps\t{%1, %0|%0, %1}"
>    [(set_attr "prefix" "vex")
>     (set_attr "mode" "<sseinsnmode>")])
> +
> +(define_int_iterator VPDPWPROD
> +  [UNSPEC_VPDPWUSD
> +   UNSPEC_VPDPWUSDS
> +   UNSPEC_VPDPWSUD
> +   UNSPEC_VPDPWSUDS
> +   UNSPEC_VPDPWUUD
> +   UNSPEC_VPDPWUUDS])
> +
> +(define_int_attr vpdpwprodtype
> +  [(UNSPEC_VPDPWUSD "wusd") (UNSPEC_VPDPWUSDS "wusds")
> +   (UNSPEC_VPDPWSUD "wsud") (UNSPEC_VPDPWSUDS "wsuds")
> +   (UNSPEC_VPDPWUUD "wuud") (UNSPEC_VPDPWUUDS "wuuds")])
> +
> +(define_insn "vpdp<vpdpwprodtype>_<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")]
> +         VPDPWPROD))]
> +  "TARGET_AVXVNNIINT16"
> +  "vpdp<vpdpwprodtype>\t{%3, %2, %0|%0, %2, %3}"
> +   [(set_attr "prefix" "vex")])
> diff --git a/gcc/doc/extend.texi b/gcc/doc/extend.texi
> index 7e4c5541b11..565bf1352e2 100644
> --- a/gcc/doc/extend.texi
> +++ b/gcc/doc/extend.texi
> @@ -7163,6 +7163,11 @@ Enable/disable the generation of the RAOINT instructions.
>  @itemx no-amx-complex
>  Enable/disable the generation of the AMX-COMPLEX instructions.
>
> +@cindex @code{target("avxvnniint16")} function attribute, x86
> +@item avxvnniint16
> +@itemx no-avxvnniint16
> +Enable/disable the generation of the AVXVNNIINT16 instructions.
> +
>  @cindex @code{target("cld")} function attribute, x86
>  @item cld
>  @itemx no-cld
> diff --git a/gcc/doc/invoke.texi b/gcc/doc/invoke.texi
> index cbc1282c274..359887db5fd 100644
> --- a/gcc/doc/invoke.texi
> +++ b/gcc/doc/invoke.texi
> @@ -1433,7 +1433,7 @@ See RS/6000 and PowerPC Options.
>  -mrdseed  -msgx -mavx512vp2intersect -mserialize -mtsxldtrk
>  -mamx-tile  -mamx-int8  -mamx-bf16 -muintr -mhreset -mavxvnni
>  -mavx512fp16 -mavxifma -mavxvnniint8 -mavxneconvert -mcmpccxadd -mamx-fp16
> --mprefetchi -mraoint -mamx-complex
> +-mprefetchi -mraoint -mamx-complex -mavxvnniint16
>  -mcldemote  -mms-bitfields  -mno-align-stringops  -minline-all-stringops
>  -minline-stringops-dynamically  -mstringop-strategy=@var{alg}
>  -mkl -mwidekl
> @@ -33552,8 +33552,10 @@ preferred alignment to @option{-mpreferred-stack-boundary=2}.
>  @need 200
>  @opindex mamx-complex
>  @itemx -mamx-complex
> +@need 200
> +@opindex mavxvnniint16
> +@itemx -mavxvnniint16
>  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,
>  AES, PCLMUL, CLFLUSHOPT, CLWB, FSGSBASE, PTWRITE, RDRND, F16C, FMA, PCONFIG,
>  WBNOINVD, FMA4, PREFETCHW, RDPID, PREFETCHWT1, RDSEED, SGX, XOP, LWP,
> @@ -33563,8 +33565,8 @@ GFNI, VAES, WAITPKG, VPCLMULQDQ, AVX512BITALG, MOVDIRI, MOVDIR64B, AVX512BF16,
>  ENQCMD, AVX512VPOPCNTDQ, AVX5124FMAPS, AVX512VNNI, AVX5124VNNIW, SERIALIZE,
>  UINTR, HRESET, AMXTILE, AMXINT8, AMXBF16, KL, WIDEKL, AVXVNNI, AVX512-FP16,
>  AVXIFMA, AVXVNNIINT8, AVXNECONVERT, CMPCCXADD, AMX-FP16, PREFETCHI, RAOINT,
> -AMX-COMPLEX or CLDEMOTE extended instruction sets. Each has a corresponding
> -@option{-mno-} option to disable use of these instructions.
> +AMX-COMPLEX, AVXVNNIINT16 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 ffb6eb1a48c..40919b30a62 100644
> --- a/gcc/doc/sourcebuild.texi
> +++ b/gcc/doc/sourcebuild.texi
> @@ -2511,6 +2511,9 @@ Target supports the execution of @code{avxneconvert} instructions.
>  @item avxvnniint8
>  Target supports the execution of @code{avxvnniint8} instructions.
>
> +@item avxvnniint16
> +Target supports the execution of @code{avxvnniint16} 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 6fe07e18fc6..53622df2bb8 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 -mavxvnniint8 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex" } */
> +/* { 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 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16" } */
>
>  /* 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 55c81677300..3b76cee3af8 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 -mavxvnniint8 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex" } */
> +/* { 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 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16" } */
>
>  /* 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 666eff50780..3d417ea1ed5 100644
> --- a/gcc/testsuite/gcc.target/i386/avx-check.h
> +++ b/gcc/testsuite/gcc.target/i386/avx-check.h
> @@ -31,6 +31,9 @@ main ()
>  #endif
>  #ifdef AVXNECONVERT
>        && __builtin_cpu_supports ("avxneconvert")
> +#endif
> +#ifdef AVXVNNIINT16
> +      && __builtin_cpu_supports ("avxvnniint16")
>  #endif
>        )
>      {
> diff --git a/gcc/testsuite/gcc.target/i386/avxvnniint16-1.c b/gcc/testsuite/gcc.target/i386/avxvnniint16-1.c
> new file mode 100644
> index 00000000000..6ae57b150fe
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/i386/avxvnniint16-1.c
> @@ -0,0 +1,43 @@
> +/* { dg-do compile } */
> +/* { dg-options "-mavxvnniint16 -O2" } */
> +/* { dg-final { scan-assembler-times "vpdpwusd\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\r]*%ymm\[0-9\]+\[^\n\r]*%ymm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
> +/* { dg-final { scan-assembler-times "vpdpwusd\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
> +/* { dg-final { scan-assembler-times "vpdpwusds\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\r]*%ymm\[0-9\]+\[^\n\r]*%ymm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
> +/* { dg-final { scan-assembler-times "vpdpwusds\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
> +/* { dg-final { scan-assembler-times "vpdpwsud\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\r]*%ymm\[0-9\]+\[^\n\r]*%ymm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
> +/* { dg-final { scan-assembler-times "vpdpwsud\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
> +/* { dg-final { scan-assembler-times "vpdpwsuds\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\r]*%ymm\[0-9\]+\[^\n\r]*%ymm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
> +/* { dg-final { scan-assembler-times "vpdpwsuds\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
> +/* { dg-final { scan-assembler-times "vpdpwuud\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\r]*%ymm\[0-9\]+\[^\n\r]*%ymm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
> +/* { dg-final { scan-assembler-times "vpdpwuud\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
> +/* { dg-final { scan-assembler-times "vpdpwuuds\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\r]*%ymm\[0-9\]+\[^\n\r]*%ymm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
> +/* { dg-final { scan-assembler-times "vpdpwuuds\[ \\t\]+\[^\{\n\]*%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
> +avxvnniint16_test (void)
> +{
> +  x = _mm256_dpwusd_avx_epi32 (x, y, z);
> +  x_ = _mm_dpwusd_avx_epi32 (x_, y_, z_);
> +
> +  x = _mm256_dpwusds_avx_epi32 (x, y, z);
> +  x_ = _mm_dpwusds_avx_epi32 (x_, y_, z_);
> +
> +  x = _mm256_dpwsud_avx_epi32 (x, y, z);
> +  x_ = _mm_dpwsud_avx_epi32 (x_, y_, z_);
> +
> +  x = _mm256_dpwsuds_avx_epi32 (x, y, z);
> +  x_ = _mm_dpwsuds_avx_epi32 (x_, y_, z_);
> +
> +  x = _mm256_dpwuud_avx_epi32 (x, y, z);
> +  x_ = _mm_dpwuud_avx_epi32 (x_, y_, z_);
> +
> +  x = _mm256_dpwuuds_avx_epi32 (x, y, z);
> +  x_ = _mm_dpwuuds_avx_epi32 (x_, y_, z_);
> +}
> diff --git a/gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwsud-2.c b/gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwsud-2.c
> new file mode 100644
> index 00000000000..bc57a8a8078
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwsud-2.c
> @@ -0,0 +1,71 @@
> +/* { dg-do run } */
> +/* { dg-options "-O2 -mavxvnniint16" } */
> +/* { dg-require-effective-target avxvnniint16 } */
> +#define AVXVNNIINT16
> +#ifndef CHECK
> +#define CHECK "avx-check.h"
> +#endif
> +
> +#ifndef TEST
> +#define TEST avx_test
> +#endif
> +
> +#include CHECK
> +
> +static void
> +CALC (int *r, int *dst, short *s1, unsigned short *s2, int size)
> +{
> +  int tempres[16];
> +  for (int i = 0; i < size; i++)
> +    tempres[i] = (int) s1[i] * (unsigned int) s2[i];
> +  for (int i = 0; i < size / 2; i++)
> +  {
> +    int test = (int) dst[i] + tempres[i * 2] + tempres[i * 2 + 1];
> +    r[i] = test;
> +  }
> +}
> +
> +void
> +TEST (void)
> +{
> +  int i;
> +  union256i_d res_256;
> +  union256i_w src1_256;
> +  union256i_uw src2_256;
> +  int res_ref_256[8];
> +
> +  for (i = 0; i < 16; 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, 16);
> +  res_256.x = _mm256_dpwsud_avx_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_w src1_128;
> +  union128i_uw src2_128;
> +  int res_ref_128[4];
> +
> +  for (i = 0; i < 8; 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, 8);
> +  res_128.x = _mm_dpwsud_avx_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/avxvnniint16-vpdpwsuds-2.c b/gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwsuds-2.c
> new file mode 100644
> index 00000000000..fbcf46a4827
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwsuds-2.c
> @@ -0,0 +1,72 @@
> +/* { dg-do run } */
> +/* { dg-options "-O2 -mavxvnniint16" } */
> +/* { dg-require-effective-target avxvnniint16 } */
> +#define AVXVNNIINT16
> +#ifndef CHECK
> +#define CHECK "avx-check.h"
> +#endif
> +
> +#ifndef TEST
> +#define TEST avx_test
> +#endif
> +
> +#include CHECK
> +
> +static void
> +CALC (int *r, int *dst, short *s1, unsigned short *s2, int size)
> +{
> +  int tempres[16];
> +  for (int i = 0; i < size; i++)
> +    tempres[i] = (int) s1[i] * (unsigned int) s2[i];
> +  for (int i = 0; i < size / 2; i++)
> +  {
> +    long long test = (long long) dst[i] + tempres[i * 2] + tempres[i * 2 + 1];
> +    r[i] = test > 0x7FFFFFFF ? 0x7FFFFFFF :
> +           (test < 0xffffffff80000000LL ? 0x80000000 : test);
> +  }
> +}
> +
> +void
> +TEST (void)
> +{
> +  int i;
> +  union256i_d res_256;
> +  union256i_w src1_256;
> +  union256i_uw src2_256;
> +  int res_ref_256[8];
> +
> +  for (i = 0; i < 16; 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, 16);
> +  res_256.x = _mm256_dpwsuds_avx_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_w src1_128;
> +  union128i_uw src2_128;
> +  int res_ref_128[4];
> +
> +  for (i = 0; i < 8; 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, 8);
> +  res_128.x = _mm_dpwsuds_avx_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/avxvnniint16-vpdpwusd-2.c b/gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwusd-2.c
> new file mode 100644
> index 00000000000..54cf271acbf
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwusd-2.c
> @@ -0,0 +1,71 @@
> +/* { dg-do run } */
> +/* { dg-options "-O2 -mavxvnniint16" } */
> +/* { dg-require-effective-target avxvnniint16 } */
> +#define AVXVNNIINT16
> +#ifndef CHECK
> +#define CHECK "avx-check.h"
> +#endif
> +
> +#ifndef TEST
> +#define TEST avx_test
> +#endif
> +
> +#include CHECK
> +
> +static void
> +CALC (int *r, int *dst, unsigned short *s1, short *s2, int size)
> +{
> +  int tempres[16];
> +  for (int i = 0; i < size; i++)
> +    tempres[i] = (unsigned int) s1[i] * (int) s2[i];
> +  for (int i = 0; i < size / 2; i++)
> +  {
> +    int test = (int) dst[i] + tempres[i * 2] + tempres[i * 2 + 1];
> +    r[i] = test;
> +  }
> +}
> +
> +void
> +TEST (void)
> +{
> +  int i;
> +  union256i_d res_256;
> +  union256i_uw src1_256;
> +  union256i_w src2_256;
> +  int res_ref_256[8];
> +
> +  for (i = 0; i < 16; 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, 16);
> +  res_256.x = _mm256_dpwusd_avx_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_uw src1_128;
> +  union128i_w src2_128;
> +  int res_ref_128[4];
> +
> +  for (i = 0; i < 8; 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, 8);
> +  res_128.x = _mm_dpwusd_avx_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/avxvnniint16-vpdpwusds-2.c b/gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwusds-2.c
> new file mode 100644
> index 00000000000..ed9594c708b
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwusds-2.c
> @@ -0,0 +1,72 @@
> +/* { dg-do run } */
> +/* { dg-options "-O2 -mavxvnniint16" } */
> +/* { dg-require-effective-target avxvnniint16 } */
> +#define AVXVNNIINT16
> +#ifndef CHECK
> +#define CHECK "avx-check.h"
> +#endif
> +
> +#ifndef TEST
> +#define TEST avx_test
> +#endif
> +
> +#include CHECK
> +
> +static void
> +CALC (int *r, int *dst, unsigned short *s1, short *s2, int size)
> +{
> +  int tempres[16];
> +  for (int i = 0; i < size; i++)
> +    tempres[i] = (unsigned int) s1[i] * (int) s2[i];
> +  for (int i = 0; i < size / 2; i++)
> +  {
> +    long long test = (long long) dst[i] + tempres[i * 2] + tempres[i * 2 + 1];
> +    r[i] = test > 0x7FFFFFFF ? 0x7FFFFFFF :
> +           (test < 0xffffffff80000000LL ? 0x80000000 : test);
> +  }
> +}
> +
> +void
> +TEST (void)
> +{
> +  int i;
> +  union256i_d res_256;
> +  union256i_uw src1_256;
> +  union256i_w src2_256;
> +  int res_ref_256[8];
> +
> +  for (i = 0; i < 16; 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, 16);
> +  res_256.x = _mm256_dpwusds_avx_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_uw src1_128;
> +  union128i_w src2_128;
> +  int res_ref_128[4];
> +
> +  for (i = 0; i < 8; 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, 8);
> +  res_128.x = _mm_dpwusds_avx_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/avxvnniint16-vpdpwuud-2.c b/gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwuud-2.c
> new file mode 100644
> index 00000000000..8bdc4330ef0
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwuud-2.c
> @@ -0,0 +1,71 @@
> +/* { dg-do run } */
> +/* { dg-options "-O2 -mavxvnniint16" } */
> +/* { dg-require-effective-target avxvnniint16 } */
> +#define AVXVNNIINT16
> +#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 short *s1, unsigned short *s2, int size)
> +{
> +  unsigned int tempres[16];
> +  for (int i = 0; i < size; i++)
> +    tempres[i] = (unsigned int) s1[i] * (unsigned int) s2[i];
> +  for (int i = 0; i < size / 2; i++)
> +  {
> +    unsigned int test = (unsigned) dst[i] + tempres[i * 2] + tempres[i * 2 + 1];
> +    r[i] = test;
> +  }
> +}
> +
> +void
> +TEST (void)
> +{
> +  int i;
> +  union256i_ud res_256;
> +  union256i_uw src1_256;
> +  union256i_uw src2_256;
> +  unsigned int res_ref_256[8];
> +
> +  for (i = 0; i < 16; 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, 16);
> +  res_256.x = _mm256_dpwuud_avx_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_uw src2_128;
> +  union128i_uw src1_128;
> +  unsigned int res_ref_128[4];
> +
> +  for (i = 0; i < 8; 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, 8);
> +  res_128.x = _mm_dpwuud_avx_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/avxvnniint16-vpdpwuuds-2.c b/gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwuuds-2.c
> new file mode 100644
> index 00000000000..32204122f1c
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/i386/avxvnniint16-vpdpwuuds-2.c
> @@ -0,0 +1,71 @@
> +/* { dg-do run } */
> +/* { dg-options "-O2 -mavxvnniint16" } */
> +/* { dg-require-effective-target avxvnniint16 } */
> +#define AVXVNNIINT16
> +#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 short *s1, unsigned short *s2, int size)
> +{
> +  unsigned int tempres[16];
> +  for (int i = 0; i < size; i++)
> +    tempres[i] = (unsigned int) s1[i] * (unsigned int) s2[i];
> +  for (int i = 0; i < size / 2; i++)
> +  {
> +    unsigned int test = (unsigned) dst[i] + tempres[i * 2] + tempres[i * 2 + 1];
> +    r[i] = test > 0xFFFFFFFF ? 0xFFFFFFFF : test;
> +  }
> +}
> +
> +void
> +TEST (void)
> +{
> +  int i;
> +  union256i_ud res_256;
> +  union256i_uw src1_256;
> +  union256i_uw src2_256;
> +  unsigned int res_ref_256[8];
> +
> +  for (i = 0; i < 16; 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, 16);
> +  res_256.x = _mm256_dpwuuds_avx_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_uw src2_128;
> +  union128i_uw src1_128;
> +  unsigned int res_ref_128[4];
> +
> +  for (i = 0; i < 8; 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, 8);
> +  res_128.x = _mm_dpwuuds_avx_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 f466962c36c..bba0fa37efd 100644
> --- a/gcc/testsuite/gcc.target/i386/funcspec-56.inc
> +++ b/gcc/testsuite/gcc.target/i386/funcspec-56.inc
> @@ -88,6 +88,7 @@ extern void test_amx_fp16 (void)              __attribute__((__target__("amx-fp16")));
>  extern void test_prefetchi (void)               __attribute__((__target__("prefetchi")));
>  extern void test_raoint (void)                  __attribute__((__target__("raoint")));
>  extern void test_amx_complex (void)            __attribute__((__target__("amx-complex")));
> +extern void test_avxvnniint16 (void)           __attribute__((__target__("avxvnniint16")));
>
>  extern void test_no_sgx (void)                 __attribute__((__target__("no-sgx")));
>  extern void test_no_avx5124fmaps(void)         __attribute__((__target__("no-avx5124fmaps")));
> @@ -177,6 +178,7 @@ extern void test_no_amx_fp16 (void)         __attribute__((__target__("no-amx-fp16")));
>  extern void test_no_prefetchi (void)            __attribute__((__target__("no-prefetchi")));
>  extern void test_no_raoint (void)               __attribute__((__target__("no-raoint")));
>  extern void test_no_amx_complex (void)         __attribute__((__target__("no-amx-complex")));
> +extern void test_no_avxvnniint16 (void)                __attribute__((__target__("no-avxvnniint16")));
>
>  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 ae4ffd1975f..2b7d78c51d3 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 -mavxvnniint8 -mavxneconvert -mamx-fp16 -mraoint -mamx-complex" } */
> +/* { 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 -mavxneconvert -mamx-fp16 -mraoint -mamx-complex -mavxvnniint16" } */
>
>  #include <x86intrin.h>
>
> diff --git a/gcc/testsuite/gcc.target/i386/sse-13.c b/gcc/testsuite/gcc.target/i386/sse-13.c
> index f046a68ddbb..33693484d2a 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 -mavxvnniint8 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex" } */
> +/* { 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 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16" } */
>  /* { 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 05322f7e914..51c2946b25a 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 -mavxvnniint8 -mavxneconvert -mamx-fp16 -mraoint -mamx-complex" } */
> +/* { 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 -mavxneconvert -mamx-fp16 -mraoint -mamx-complex -mavxvnniint16" } */
>  /* { 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 53c38b70241..4982fde2a76 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,tbm,lwp,fsgsbase,rdrnd,f16c,rtm,rdseed,prfchw,adx,fxsr,xsaveopt,avx512f,avx512er,avx512cd,avx512pf,sha,prefetchwt1,avx512vl,avx512bw,avx512dq,avx512vbmi,avx512vbmi2,avx512ifma,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,gfni,avx512bitalg,avx512bf16,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,amx-fp16,raoint,amx-complex")
> +#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,avx512f,avx512er,avx512cd,avx512pf,sha,prefetchwt1,avx512vl,avx512bw,avx512dq,avx512vbmi,avx512vbmi2,avx512ifma,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,gfni,avx512bitalg,avx512bf16,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,amx-fp16,raoint,amx-complex,avxvnniint16")
>  #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,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,gfni,avx512bitalg,avx512bf16,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,amx-fp16,raoint,amx-complex")
> +#pragma GCC target ("avx,avx2,rdrnd,fsgsbase,f16c,rtm,avx512f,avx512er,avx512cd,avx512pf,sha,avx512vl,avx512bw,avx512dq,avx512ifma,avx512vbmi,avx512vbmi2,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,gfni,avx512bitalg,avx512bf16,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,amx-fp16,raoint,amx-complex,avxvnniint16")
>  #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 50bf85a3392..7e9c9f2ca2b 100644
> --- a/gcc/testsuite/gcc.target/i386/sse-23.c
> +++ b/gcc/testsuite/gcc.target/i386/sse-23.c
> @@ -847,6 +847,6 @@
>  #define __builtin_ia32_cmpccxadd(A, B, C, D) __builtin_ia32_cmpccxadd(A, B, C, 1)
>  #define __builtin_ia32_cmpccxadd64(A, B, C, D) __builtin_ia32_cmpccxadd64(A, B, C, 1)
>
> -#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,avx512f,avx512er,avx512cd,avx512pf,sha,prefetchwt1,xsavec,xsaves,clflushopt,avx512bw,avx512dq,avx512vl,avx512vbmi,avx512ifma,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,clwb,mwaitx,clzero,pku,sgx,rdpid,gfni,avx512vbmi2,vpclmulqdq,avx512bitalg,pconfig,wbnoinvd,avx512bf16,enqcmd,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,cmpccxadd,amx-fp16,prefetchi,raoint,amx-complex")
> +#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,avx512f,avx512er,avx512cd,avx512pf,sha,prefetchwt1,xsavec,xsaves,clflushopt,avx512bw,avx512dq,avx512vl,avx512vbmi,avx512ifma,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,clwb,mwaitx,clzero,pku,sgx,rdpid,gfni,avx512vbmi2,vpclmulqdq,avx512bitalg,pconfig,wbnoinvd,avx512bf16,enqcmd,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,cmpccxadd,amx-fp16,prefetchi,raoint,amx-complex,avxvnniint16")
>
>  #include <x86intrin.h>
> diff --git a/gcc/testsuite/lib/target-supports.exp b/gcc/testsuite/lib/target-supports.exp
> index 33482b2fa9a..60de239f1ce 100644
> --- a/gcc/testsuite/lib/target-supports.exp
> +++ b/gcc/testsuite/lib/target-supports.exp
> @@ -9856,6 +9856,18 @@ proc check_effective_target_amx_complex { } {
>      } "-mamx-complex" ]
>  }
>
> +# Return 1 if avxvnniint16 instructions can be compiled.
> +proc check_effective_target_avxvnniint16 { } {
> +    return [check_no_compiler_messages avxvnniint16 object {
> +       typedef int __v8si __attribute__ ((__vector_size__ (32)));
> +       __v8si
> +       _mm256_dpwsud_avx_epi32  (__v8si __A, __v8si __B, __v8si __C)
> +        {
> +          return __builtin_ia32_vpdpwsud256 (__A, __B, __C);
> +       }
> +    } "-O0 -mavxvnniint16" ]
> +}
> +
>  # Return 1 if sse instructions can be compiled.
>  proc check_effective_target_sse { } {
>      return [check_no_compiler_messages sse object {
> --
> 2.31.1
>


-- 
BR,
Hongtao

^ permalink raw reply	[flat|nested] 9+ messages in thread

* [PATCH 2/4] Support Intel SM3
  2023-07-13  6:03 [PATCH 0/4] Support Intel Arrow Lake/Lunar Lake ISAs Haochen Jiang
  2023-07-13  6:03 ` [PATCH 1/4] Support Intel AVX-VNNI-INT16 Haochen Jiang
@ 2023-07-13  6:03 ` Haochen Jiang
  2023-07-17  1:39   ` Hongtao Liu
  2023-07-13  6:03 ` [PATCH 3/4] Support Intel SHA512 Haochen Jiang
  2023-07-13  6:03 ` [PATCH 4/4] Support Intel SM4 Haochen Jiang
  3 siblings, 1 reply; 9+ messages in thread
From: Haochen Jiang @ 2023-07-13  6:03 UTC (permalink / raw)
  To: gcc-patches; +Cc: hongtao.liu, ubizjak

gcc/ChangeLog:

	* common/config/i386/cpuinfo.h (get_available_features):
	Detect SM3.
	* common/config/i386/i386-common.cc (OPTION_MASK_ISA2_SM3_SET,
	OPTION_MASK_ISA2_SM3_UNSET): New.
	(OPTION_MASK_ISA2_AVX_UNSET): Add SM3.
	(ix86_handle_option): Handle -msm3.
	* common/config/i386/i386-cpuinfo.h (enum processor_features):
	Add FEATURE_SM3.
	* common/config/i386/i386-isas.h: Add ISA_NAME_TABLE_ENTRY for
	SM3.
	* config.gcc: Add sm3intrin.h
	* config/i386/cpuid.h (bit_SM3): New.
	* config/i386/i386-builtin-types.def:
	Add DEF_FUNCTION_TYPE (V4SI, V4SI, V4SI, V4SI, INT).
	* config/i386/i386-builtin.def (BDESC): Add new builtins.
	* config/i386/i386-c.cc (ix86_target_macros_internal): Define
	__SM3__.
	* config/i386/i386-expand.cc (ix86_expand_args_builtin): Handle
	V4SI_FTYPE_V4SI_V4SI_V4SI_INT.
	* config/i386/i386-isa.def (SM3): Add DEF_PTA(SM3).
	* config/i386/i386-options.cc (isa2_opts): Add -msm3.
	(ix86_valid_target_attribute_inner_p): Handle sm3.
	* config/i386/i386.opt: Add option -msm3.
	* config/i386/immintrin.h: Include sm3intrin.h.
	* config/i386/sse.md (vsm3msg1): New define insn.
	(vsm3msg2): Ditto.
	(vsm3rnds2): Ditto.
	* doc/extend.texi: Document sm3.
	* doc/invoke.texi: Document -msm3.
	* doc/sourcebuild.texi: Document target sm3.
	* config/i386/sm3intrin.h: New file.

gcc/testsuite/ChangeLog:

	* g++.dg/other/i386-2.C: Add -msm3.
	* g++.dg/other/i386-3.C: Ditto.
	* gcc.target/i386/avx-1.c: Add new define for immediate.
	* gcc.target/i386/funcspec-56.inc: Add new target attribute.
	* gcc.target/i386/sse-12.c: Add -msm3.
	* gcc.target/i386/sse-13.c: Ditto.
	* gcc.target/i386/sse-14.c: Ditto.
	* gcc.target/i386/sse-22.c: Add sm3.
	* gcc.target/i386/sse-23.c: Ditto.
	* lib/target-supports.exp (check_effective_target_sm3): New.
	* gcc.target/i386/sm3-1.c: New test.
	* gcc.target/i386/sm3-check.h: Ditto.
	* gcc.target/i386/sm3msg1-2.c: Ditto.
	* gcc.target/i386/sm3msg2-2.c: Ditto.
	* gcc.target/i386/sm3rnds2-2.c: Ditto.
---
 gcc/common/config/i386/cpuinfo.h              |   2 +
 gcc/common/config/i386/i386-common.cc         |  20 +++-
 gcc/common/config/i386/i386-cpuinfo.h         |   1 +
 gcc/common/config/i386/i386-isas.h            |   1 +
 gcc/config.gcc                                |   3 +-
 gcc/config/i386/cpuid.h                       |   1 +
 gcc/config/i386/i386-builtin-types.def        |   3 +
 gcc/config/i386/i386-builtin.def              |   5 +
 gcc/config/i386/i386-c.cc                     |   2 +
 gcc/config/i386/i386-expand.cc                |   1 +
 gcc/config/i386/i386-isa.def                  |   1 +
 gcc/config/i386/i386-options.cc               |   2 +
 gcc/config/i386/i386.opt                      |   5 +
 gcc/config/i386/immintrin.h                   |   2 +
 gcc/config/i386/sm3intrin.h                   |  72 ++++++++++++
 gcc/config/i386/sse.md                        |  43 ++++++++
 gcc/doc/extend.texi                           |   5 +
 gcc/doc/invoke.texi                           |   7 +-
 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-1.c         |   3 +
 gcc/testsuite/gcc.target/i386/funcspec-56.inc |   2 +
 gcc/testsuite/gcc.target/i386/sm3-1.c         |  17 +++
 gcc/testsuite/gcc.target/i386/sm3-check.h     |  37 +++++++
 gcc/testsuite/gcc.target/i386/sm3msg1-2.c     |  54 +++++++++
 gcc/testsuite/gcc.target/i386/sm3msg2-2.c     |  57 ++++++++++
 gcc/testsuite/gcc.target/i386/sm3rnds2-2.c    | 104 ++++++++++++++++++
 gcc/testsuite/gcc.target/i386/sse-12.c        |   2 +-
 gcc/testsuite/gcc.target/i386/sse-13.c        |   5 +-
 gcc/testsuite/gcc.target/i386/sse-14.c        |   5 +-
 gcc/testsuite/gcc.target/i386/sse-22.c        |   7 +-
 gcc/testsuite/gcc.target/i386/sse-23.c        |   5 +-
 gcc/testsuite/lib/target-supports.exp         |  15 +++
 34 files changed, 484 insertions(+), 12 deletions(-)
 create mode 100644 gcc/config/i386/sm3intrin.h
 create mode 100644 gcc/testsuite/gcc.target/i386/sm3-1.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sm3-check.h
 create mode 100644 gcc/testsuite/gcc.target/i386/sm3msg1-2.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sm3msg2-2.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sm3rnds2-2.c

diff --git a/gcc/common/config/i386/cpuinfo.h b/gcc/common/config/i386/cpuinfo.h
index 3599f9def2c..e5cdffe017a 100644
--- a/gcc/common/config/i386/cpuinfo.h
+++ b/gcc/common/config/i386/cpuinfo.h
@@ -877,6 +877,8 @@ get_available_features (struct __processor_model *cpu_model,
 	    set_feature (FEATURE_AVXNECONVERT);
 	  if (edx & bit_AVXVNNIINT16)
 	    set_feature (FEATURE_AVXVNNIINT16);
+	  if (eax & bit_SM3)
+	    set_feature (FEATURE_SM3);
 	}
       if (avx512_usable)
 	{
diff --git a/gcc/common/config/i386/i386-common.cc b/gcc/common/config/i386/i386-common.cc
index 32c6d00580d..57b008ca3af 100644
--- a/gcc/common/config/i386/i386-common.cc
+++ b/gcc/common/config/i386/i386-common.cc
@@ -120,6 +120,7 @@ along with GCC; see the file COPYING3.  If not see
 #define OPTION_MASK_ISA2_AMX_COMPLEX_SET \
   (OPTION_MASK_ISA2_AMX_TILE | OPTION_MASK_ISA2_AMX_COMPLEX)
 #define OPTION_MASK_ISA2_AVXVNNIINT16_SET OPTION_MASK_ISA2_AVXVNNIINT16
+#define OPTION_MASK_ISA2_SM3_SET OPTION_MASK_ISA2_SM3
 
 /* SSE4 includes both SSE4.1 and SSE4.2. -msse4 should be the same
    as -msse4.2.  */
@@ -303,6 +304,7 @@ along with GCC; see the file COPYING3.  If not see
 #define OPTION_MASK_ISA2_RAOINT_UNSET OPTION_MASK_ISA2_RAOINT
 #define OPTION_MASK_ISA2_AMX_COMPLEX_UNSET OPTION_MASK_ISA2_AMX_COMPLEX
 #define OPTION_MASK_ISA2_AVXVNNIINT16_UNSET OPTION_MASK_ISA2_AVXVNNIINT16
+#define OPTION_MASK_ISA2_SM3_UNSET OPTION_MASK_ISA2_SM3
 
 /* SSE4 includes both SSE4.1 and SSE4.2.  -mno-sse4 should the same
    as -mno-sse4.1. */
@@ -351,7 +353,8 @@ along with GCC; see the file COPYING3.  If not see
 #define OPTION_MASK_ISA2_GENERAL_REGS_ONLY_UNSET \
   OPTION_MASK_ISA2_SSE_UNSET
 #define OPTION_MASK_ISA2_AVX_UNSET \
-  (OPTION_MASK_ISA2_AVX2_UNSET | OPTION_MASK_ISA2_VAES_UNSET)
+  (OPTION_MASK_ISA2_AVX2_UNSET | OPTION_MASK_ISA2_VAES_UNSET \
+   | OPTION_MASK_ISA2_SM3_UNSET)
 #define OPTION_MASK_ISA2_SSE4_2_UNSET OPTION_MASK_ISA2_AVX_UNSET
 #define OPTION_MASK_ISA2_SSE4_1_UNSET OPTION_MASK_ISA2_SSE4_2_UNSET
 #define OPTION_MASK_ISA2_SSE4_UNSET OPTION_MASK_ISA2_SSE4_1_UNSET
@@ -1288,6 +1291,21 @@ ix86_handle_option (struct gcc_options *opts,
 	}
       return true;
 
+    case OPT_msm3:
+      if (value)
+	{
+	  opts->x_ix86_isa_flags2 |= OPTION_MASK_ISA2_SM3_SET;
+	  opts->x_ix86_isa_flags2_explicit |= OPTION_MASK_ISA2_SM3_SET;
+	  opts->x_ix86_isa_flags |= OPTION_MASK_ISA_AVX_SET;
+	  opts->x_ix86_isa_flags_explicit |= OPTION_MASK_ISA_AVX_SET;
+	}
+      else
+	{
+	  opts->x_ix86_isa_flags2 &= ~OPTION_MASK_ISA2_SM3_UNSET;
+	  opts->x_ix86_isa_flags2_explicit |= OPTION_MASK_ISA2_SM3_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 ae4e6a02f7f..c3403090c3b 100644
--- a/gcc/common/config/i386/i386-cpuinfo.h
+++ b/gcc/common/config/i386/i386-cpuinfo.h
@@ -256,6 +256,7 @@ enum processor_features
   FEATURE_RAOINT,
   FEATURE_AMX_COMPLEX,
   FEATURE_AVXVNNIINT16,
+  FEATURE_SM3,
   CPU_FEATURE_MAX
 };
 
diff --git a/gcc/common/config/i386/i386-isas.h b/gcc/common/config/i386/i386-isas.h
index fc6abdedf24..961a7f0ccd4 100644
--- a/gcc/common/config/i386/i386-isas.h
+++ b/gcc/common/config/i386/i386-isas.h
@@ -188,4 +188,5 @@ ISA_NAMES_TABLE_START
 			P_NONE, "-mamx-complex")
   ISA_NAMES_TABLE_ENTRY("avxvnniint16", FEATURE_AVXVNNIINT16,
 			P_NONE, "-mavxvnniint16")
+  ISA_NAMES_TABLE_ENTRY("sm3", FEATURE_SM3, P_NONE, "-msm3")
 ISA_NAMES_TABLE_END
diff --git a/gcc/config.gcc b/gcc/config.gcc
index fc74d776048..fbd7360e355 100644
--- a/gcc/config.gcc
+++ b/gcc/config.gcc
@@ -435,7 +435,8 @@ i[34567]86-*-* | x86_64-*-*)
 		       mwaitintrin.h avx512fp16intrin.h avx512fp16vlintrin.h
 		       avxifmaintrin.h avxvnniint8intrin.h avxneconvertintrin.h
 		       cmpccxaddintrin.h amxfp16intrin.h prfchiintrin.h
-		       raointintrin.h amxcomplexintrin.h avxvnniint16intrin.h"
+		       raointintrin.h amxcomplexintrin.h avxvnniint16intrin.h
+		       sm3intrin.h"
 	;;
 ia64-*-*)
 	extra_headers=ia64intrin.h
diff --git a/gcc/config/i386/cpuid.h b/gcc/config/i386/cpuid.h
index 98d0f193d22..28a36ad0628 100644
--- a/gcc/config/i386/cpuid.h
+++ b/gcc/config/i386/cpuid.h
@@ -132,6 +132,7 @@
 
 /* Extended Features Sub-leaf (%eax == 7, %ecx == 1) */
 /* %eax */
+#define bit_SM3		(1 << 1)
 #define bit_RAOINT      (1 << 3)
 #define bit_AVXVNNI     (1 << 4)
 #define bit_AVX512BF16  (1 << 5)
diff --git a/gcc/config/i386/i386-builtin-types.def b/gcc/config/i386/i386-builtin-types.def
index cb2d0cd56ed..899eac1e014 100644
--- a/gcc/config/i386/i386-builtin-types.def
+++ b/gcc/config/i386/i386-builtin-types.def
@@ -1416,3 +1416,6 @@ DEF_FUNCTION_TYPE (LONGLONG, PLONGLONG, LONGLONG, LONGLONG, INT)
 # PREFETCHI builtins
 DEF_FUNCTION_TYPE (VOID, PCVOID, INT)
 DEF_FUNCTION_TYPE (VOID, PCVOID, INT, INT, INT)
+
+# SM3 builtins
+DEF_FUNCTION_TYPE (V4SI, V4SI, V4SI, V4SI, INT)
diff --git a/gcc/config/i386/i386-builtin.def b/gcc/config/i386/i386-builtin.def
index ff5b3dcbcd3..17db19c2495 100644
--- a/gcc/config/i386/i386-builtin.def
+++ b/gcc/config/i386/i386-builtin.def
@@ -1655,6 +1655,11 @@ BDESC (OPTION_MASK_ISA_SSE2, 0, CODE_FOR_sha256msg1, 0, IX86_BUILTIN_SHA256MSG1,
 BDESC (OPTION_MASK_ISA_SSE2, 0, CODE_FOR_sha256msg2, 0, IX86_BUILTIN_SHA256MSG2, UNKNOWN, (int) V4SI_FTYPE_V4SI_V4SI)
 BDESC (OPTION_MASK_ISA_SSE2, 0, CODE_FOR_sha256rnds2, 0, IX86_BUILTIN_SHA256RNDS2, UNKNOWN, (int) V4SI_FTYPE_V4SI_V4SI_V4SI)
 
+/* SM3 */
+BDESC (OPTION_MASK_ISA_AVX, OPTION_MASK_ISA2_SM3, CODE_FOR_vsm3msg1, "__builtin_ia32_vsm3msg1", IX86_BUILTIN_VSM3MSG1, UNKNOWN, (int) V4SI_FTYPE_V4SI_V4SI_V4SI)
+BDESC (OPTION_MASK_ISA_AVX, OPTION_MASK_ISA2_SM3, CODE_FOR_vsm3msg2, "__builtin_ia32_vsm3msg2", IX86_BUILTIN_VSM3MSG2, UNKNOWN, (int) V4SI_FTYPE_V4SI_V4SI_V4SI)
+BDESC (OPTION_MASK_ISA_AVX, OPTION_MASK_ISA2_SM3, CODE_FOR_vsm3rnds2, "__builtin_ia32_vsm3rnds2", IX86_BUILTIN_VSM3RNDS2, UNKNOWN, (int) V4SI_FTYPE_V4SI_V4SI_V4SI_INT)
+
 /* AVX512VL.  */
 BDESC (OPTION_MASK_ISA_AVX512BW | OPTION_MASK_ISA_AVX512VL, 0, CODE_FOR_avx2_palignrv32qi_mask, "__builtin_ia32_palignr256_mask", IX86_BUILTIN_PALIGNR256_MASK, UNKNOWN, (int) V4DI_FTYPE_V4DI_V4DI_INT_V4DI_USI_CONVERT)
 BDESC (OPTION_MASK_ISA_AVX512BW | OPTION_MASK_ISA_AVX512VL, 0, CODE_FOR_ssse3_palignrv16qi_mask, "__builtin_ia32_palignr128_mask", IX86_BUILTIN_PALIGNR128_MASK, UNKNOWN, (int) V2DI_FTYPE_V2DI_V2DI_INT_V2DI_UHI_CONVERT)
diff --git a/gcc/config/i386/i386-c.cc b/gcc/config/i386/i386-c.cc
index d3514dd46ac..0cb5a6dcce5 100644
--- a/gcc/config/i386/i386-c.cc
+++ b/gcc/config/i386/i386-c.cc
@@ -679,6 +679,8 @@ ix86_target_macros_internal (HOST_WIDE_INT isa_flag,
     def_or_undef (parse_in, "__AMX_COMPLEX__");
   if (isa_flag2 & OPTION_MASK_ISA2_AVXVNNIINT16)
     def_or_undef (parse_in, "__AVXVNNIINT16__");
+  if (isa_flag2 & OPTION_MASK_ISA2_SM3)
+    def_or_undef (parse_in, "__SM3__");
   if (TARGET_IAMCU)
     {
       def_or_undef (parse_in, "__iamcu");
diff --git a/gcc/config/i386/i386-expand.cc b/gcc/config/i386/i386-expand.cc
index 648d6098eff..f6ad54c0cfe 100644
--- a/gcc/config/i386/i386-expand.cc
+++ b/gcc/config/i386/i386-expand.cc
@@ -11202,6 +11202,7 @@ ix86_expand_args_builtin (const struct builtin_description *d,
     case V4SF_FTYPE_V4SF_V4SF_V4SI_INT:
     case V8SF_FTYPE_V8SF_V8SF_V8SI_INT:
     case V16SF_FTYPE_V16SF_V16SF_V16SI_INT:
+    case V4SI_FTYPE_V4SI_V4SI_V4SI_INT:
       nargs = 4;
       nargs_constant = 1;
       break;
diff --git a/gcc/config/i386/i386-isa.def b/gcc/config/i386/i386-isa.def
index fbf22f7270a..432c36e7f79 100644
--- a/gcc/config/i386/i386-isa.def
+++ b/gcc/config/i386/i386-isa.def
@@ -118,3 +118,4 @@ DEF_PTA(PREFETCHI)
 DEF_PTA(RAOINT)
 DEF_PTA(AMX_COMPLEX)
 DEF_PTA(AVXVNNIINT16)
+DEF_PTA(SM3)
diff --git a/gcc/config/i386/i386-options.cc b/gcc/config/i386/i386-options.cc
index d981666dd87..db2ff0c7ae1 100644
--- a/gcc/config/i386/i386-options.cc
+++ b/gcc/config/i386/i386-options.cc
@@ -241,6 +241,7 @@ static struct ix86_target_opts isa2_opts[] =
   { "-mraoint", 	OPTION_MASK_ISA2_RAOINT },
   { "-mamx-complex",	OPTION_MASK_ISA2_AMX_COMPLEX },
   { "-mavxvnniint16",	OPTION_MASK_ISA2_AVXVNNIINT16 },
+  { "-msm3",		OPTION_MASK_ISA2_SM3 }
 };
 static struct ix86_target_opts isa_opts[] =
 {
@@ -1093,6 +1094,7 @@ ix86_valid_target_attribute_inner_p (tree fndecl, tree args, char *p_strings[],
     IX86_ATTR_ISA ("raoint", OPT_mraoint),
     IX86_ATTR_ISA ("amx-complex", OPT_mamx_complex),
     IX86_ATTR_ISA ("avxvnniint16", OPT_mavxvnniint16),
+    IX86_ATTR_ISA ("sm3", OPT_msm3),
 
     /* enum options */
     IX86_ATTR_ENUM ("fpmath=",	OPT_mfpmath_),
diff --git a/gcc/config/i386/i386.opt b/gcc/config/i386/i386.opt
index 618d713530f..80a8611993c 100644
--- a/gcc/config/i386/i386.opt
+++ b/gcc/config/i386/i386.opt
@@ -1283,3 +1283,8 @@ mavxvnniint16
 Target Mask(ISA2_AVXVNNIINT16) Var(ix86_isa_flags2) Save
 Support MMX, SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2, AVX, AVX2 and
 AVXVNNIINT16 built-in functions and code generation.
+
+msm3
+Target Mask(ISA2_SM3) Var(ix86_isa_flags2) Save
+Support MMX, SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2, AVX and
+SM3 built-in functions and code generation.
diff --git a/gcc/config/i386/immintrin.h b/gcc/config/i386/immintrin.h
index 52dc35d8398..7731990131c 100644
--- a/gcc/config/i386/immintrin.h
+++ b/gcc/config/i386/immintrin.h
@@ -108,6 +108,8 @@
 
 #include <shaintrin.h>
 
+#include <sm3intrin.h>
+
 #include <fmaintrin.h>
 
 #include <f16cintrin.h>
diff --git a/gcc/config/i386/sm3intrin.h b/gcc/config/i386/sm3intrin.h
new file mode 100644
index 00000000000..378c3dd41d9
--- /dev/null
+++ b/gcc/config/i386/sm3intrin.h
@@ -0,0 +1,72 @@
+/* Copyright (C) 2023 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/>.  */
+
+#ifndef _IMMINTRIN_H_INCLUDED
+#error "Never use <sm3intrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef _SM3INTRIN_H_INCLUDED
+#define _SM3INTRIN_H_INCLUDED
+
+#ifndef __SM3__
+#pragma GCC push_options
+#pragma GCC target("sm3")
+#define __DISABLE_SM3__
+#endif /* __SM3__ */
+
+extern __inline __m128i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sm3msg1_epi32 (__m128i __A, __m128i __B, __m128i __C)
+{
+  return (__m128i) __builtin_ia32_vsm3msg1 ((__v4si) __A, (__v4si) __B,
+					    (__v4si) __C);
+}
+
+extern __inline __m128i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sm3msg2_epi32 (__m128i __A, __m128i __B, __m128i __C)
+{
+  return (__m128i) __builtin_ia32_vsm3msg2 ((__v4si) __A, (__v4si) __B,
+					    (__v4si) __C);
+}
+
+#ifdef __OPTIMIZE__
+extern __inline __m128i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sm3rnds2_epi32 (__m128i __A, __m128i __B, __m128i __C, const int __D)
+{
+  return (__m128i) __builtin_ia32_vsm3rnds2 ((__v4si) __A, (__v4si) __B,
+					     (__v4si) __C, __D);
+}
+#else
+#define _mm_sm3rnds2_epi32(A, B, C, D)					\
+  ((__m128i) __builtin_ia32_vsm3rnds2 ((__v4si) (A), (__v4si) (B),	\
+				       (__v4si) (C), (int) (D)))
+#endif
+
+#ifdef __DISABLE_SM3__
+#undef __DISABLE_SM3__
+#pragma GCC pop_options
+#endif /* __DISABLE_SM3__ */
+
+#endif /* _SM3INTRIN_H_INCLUDED */
diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md
index 85a5f801e7a..25a1e5dd780 100644
--- a/gcc/config/i386/sse.md
+++ b/gcc/config/i386/sse.md
@@ -196,6 +196,11 @@
   UNSPEC_COMPLEX_FMUL
   UNSPEC_COMPLEX_FCMUL
   UNSPEC_COMPLEX_MASK
+ 
+  ;; For SM3 support
+  UNSPEC_SM3MSG1
+  UNSPEC_SM3MSG2
+  UNSPEC_SM3RNDS2
 
   ;; For AVX-VNNI-INT8 support
   UNSPEC_VPDPBSSD
@@ -28597,6 +28602,44 @@
    (set_attr "length_immediate" "1")
    (set_attr "mode" "TI")])
 
+(define_insn "vsm3msg1"
+  [(set (match_operand:V4SI 0 "register_operand" "=x")
+	(unspec:V4SI
+	  [(match_operand:V4SI 1 "register_operand" "0")
+	   (match_operand:V4SI 2 "register_operand" "x")
+	   (match_operand:V4SI 3 "vector_operand" "xBm")]
+	  UNSPEC_SM3MSG1))]
+  "TARGET_SM3"
+  "vsm3msg1\t{%3, %2, %0|%0, %2, %3}"
+  [(set_attr "type" "other")
+   (set_attr "mode" "TI")])
+
+(define_insn "vsm3msg2"
+  [(set (match_operand:V4SI 0 "register_operand" "=x")
+	(unspec:V4SI
+	  [(match_operand:V4SI 1 "register_operand" "0")
+	   (match_operand:V4SI 2 "register_operand" "x")
+	   (match_operand:V4SI 3 "vector_operand" "xBm")]
+	  UNSPEC_SM3MSG2))]
+  "TARGET_SM3"
+  "vsm3msg2\t{%3, %2, %0|%0, %2, %3}"
+  [(set_attr "type" "other")
+   (set_attr "mode" "TI")])
+
+(define_insn "vsm3rnds2"
+  [(set (match_operand:V4SI 0 "register_operand" "=x")
+	(unspec:V4SI
+	  [(match_operand:V4SI 1 "register_operand" "0")
+	   (match_operand:V4SI 2 "register_operand" "x")
+	   (match_operand:V4SI 3 "vector_operand" "xBm")
+	   (match_operand:SI 4 "const_0_to_255_operand" "n")]
+	  UNSPEC_SM3RNDS2))]
+  "TARGET_SM3"
+  "vsm3rnds2\t{%4, %3, %2, %0|%0, %2, %3, %4}"
+  [(set_attr "type" "other")
+   (set_attr "mode" "TI")
+   (set_attr "length_immediate" "1")])
+
 (define_insn_and_split "avx512f_<castmode><avxsizesuffix>_<castmode>"
   [(set (match_operand:AVX512MODE2P 0 "nonimmediate_operand" "=x,m")
 	(vec_concat:AVX512MODE2P
diff --git a/gcc/doc/extend.texi b/gcc/doc/extend.texi
index 565bf1352e2..e76cd399a83 100644
--- a/gcc/doc/extend.texi
+++ b/gcc/doc/extend.texi
@@ -7168,6 +7168,11 @@ Enable/disable the generation of the AMX-COMPLEX instructions.
 @itemx no-avxvnniint16
 Enable/disable the generation of the AVXVNNIINT16 instructions.
 
+@cindex @code{target("sm3")} function attribute, x86
+@item sm3
+@itemx no-sm3
+Enable/disable the generation of the SM3 instructions.
+
 @cindex @code{target("cld")} function attribute, x86
 @item cld
 @itemx no-cld
diff --git a/gcc/doc/invoke.texi b/gcc/doc/invoke.texi
index 359887db5fd..2671d70736f 100644
--- a/gcc/doc/invoke.texi
+++ b/gcc/doc/invoke.texi
@@ -1433,7 +1433,7 @@ See RS/6000 and PowerPC Options.
 -mrdseed  -msgx -mavx512vp2intersect -mserialize -mtsxldtrk
 -mamx-tile  -mamx-int8  -mamx-bf16 -muintr -mhreset -mavxvnni
 -mavx512fp16 -mavxifma -mavxvnniint8 -mavxneconvert -mcmpccxadd -mamx-fp16
--mprefetchi -mraoint -mamx-complex -mavxvnniint16
+-mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3
 -mcldemote  -mms-bitfields  -mno-align-stringops  -minline-all-stringops
 -minline-stringops-dynamically  -mstringop-strategy=@var{alg}
 -mkl -mwidekl
@@ -33555,6 +33555,9 @@ preferred alignment to @option{-mpreferred-stack-boundary=2}.
 @need 200
 @opindex mavxvnniint16
 @itemx -mavxvnniint16
+@need 200
+@opindex msm3
+@itemx -msm3
 These switches enable the use of instructions in the MMX, SSE,
 AVX512ER, AVX512CD, AVX512VL, AVX512BW, AVX512DQ, AVX512IFMA, AVX512VBMI, SHA,
 AES, PCLMUL, CLFLUSHOPT, CLWB, FSGSBASE, PTWRITE, RDRND, F16C, FMA, PCONFIG,
@@ -33565,7 +33568,7 @@ GFNI, VAES, WAITPKG, VPCLMULQDQ, AVX512BITALG, MOVDIRI, MOVDIR64B, AVX512BF16,
 ENQCMD, AVX512VPOPCNTDQ, AVX5124FMAPS, AVX512VNNI, AVX5124VNNIW, SERIALIZE,
 UINTR, HRESET, AMXTILE, AMXINT8, AMXBF16, KL, WIDEKL, AVXVNNI, AVX512-FP16,
 AVXIFMA, AVXVNNIINT8, AVXNECONVERT, CMPCCXADD, AMX-FP16, PREFETCHI, RAOINT,
-AMX-COMPLEX, AVXVNNIINT16 or CLDEMOTE extended instruction sets. Each has a
+AMX-COMPLEX, AVXVNNIINT16, SM3 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
diff --git a/gcc/doc/sourcebuild.texi b/gcc/doc/sourcebuild.texi
index 40919b30a62..dae51132c42 100644
--- a/gcc/doc/sourcebuild.texi
+++ b/gcc/doc/sourcebuild.texi
@@ -2589,6 +2589,9 @@ Target supports the execution of @code{raoint} instructions.
 @item rdrand
 Target supports x86 @code{rdrand} instruction.
 
+@item sm3
+Target supports the execution of @code{sm3} instructions.
+
 @item sqrt_insn
 Target has a square root instruction that the compiler can generate.
 
diff --git a/gcc/testsuite/g++.dg/other/i386-2.C b/gcc/testsuite/g++.dg/other/i386-2.C
index 53622df2bb8..2ec93261cac 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 -mavxvnniint8 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16" } */
+/* { 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 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3" } */
 
 /* 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 3b76cee3af8..fe03143e39b 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 -mavxvnniint8 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16" } */
+/* { 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 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3" } */
 
 /* 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-1.c b/gcc/testsuite/gcc.target/i386/avx-1.c
index 0b2b68b678d..a6589deca84 100644
--- a/gcc/testsuite/gcc.target/i386/avx-1.c
+++ b/gcc/testsuite/gcc.target/i386/avx-1.c
@@ -839,6 +839,9 @@
 #define __builtin_ia32_cmpccxadd(A, B, C, D) __builtin_ia32_cmpccxadd(A, B, C, 1)
 #define __builtin_ia32_cmpccxadd64(A, B, C, D) __builtin_ia32_cmpccxadd64(A, B, C, 1)
 
+/* sm3intrin.h */
+#define __builtin_ia32_vsm3rnds2(A, B, C, D) __builtin_ia32_vsm3rnds2 (A, B, C, 1)
+
 #include <wmmintrin.h>
 #include <immintrin.h>
 #include <mm3dnow.h>
diff --git a/gcc/testsuite/gcc.target/i386/funcspec-56.inc b/gcc/testsuite/gcc.target/i386/funcspec-56.inc
index bba0fa37efd..8dd8d9bf9d8 100644
--- a/gcc/testsuite/gcc.target/i386/funcspec-56.inc
+++ b/gcc/testsuite/gcc.target/i386/funcspec-56.inc
@@ -89,6 +89,7 @@ extern void test_prefetchi (void)               __attribute__((__target__("prefe
 extern void test_raoint (void)                  __attribute__((__target__("raoint")));
 extern void test_amx_complex (void)		__attribute__((__target__("amx-complex")));
 extern void test_avxvnniint16 (void)		__attribute__((__target__("avxvnniint16")));
+extern void test_sm3 (void)			__attribute__((__target__("sm3")));
 
 extern void test_no_sgx (void)			__attribute__((__target__("no-sgx")));
 extern void test_no_avx5124fmaps(void)		__attribute__((__target__("no-avx5124fmaps")));
@@ -179,6 +180,7 @@ extern void test_no_prefetchi (void)            __attribute__((__target__("no-pr
 extern void test_no_raoint (void)               __attribute__((__target__("no-raoint")));
 extern void test_no_amx_complex (void)		__attribute__((__target__("no-amx-complex")));
 extern void test_no_avxvnniint16 (void)		__attribute__((__target__("no-avxvnniint16")));
+extern void test_no_sm3 (void)			__attribute__((__target__("no-sm3")));
 
 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/sm3-1.c b/gcc/testsuite/gcc.target/i386/sm3-1.c
new file mode 100644
index 00000000000..0a8ea658130
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sm3-1.c
@@ -0,0 +1,17 @@
+/* { dg-do compile } */
+/* { dg-options "-O2 -msm3" } */
+/* { dg-final { scan-assembler "vsm3msg1\[ \\t\]+\[^\n\]*%xmm\[0-9\]+\[^\n\]*%xmm\[0-9\]+\[^\n\]*%xmm\[0-9\]" } } */
+/* { dg-final { scan-assembler "vsm3msg2\[ \\t\]+\[^\n\]*%xmm\[0-9\]+\[^\n\]*%xmm\[0-9\]+\[^\n\]*%xmm\[0-9\]" } } */
+/* { dg-final { scan-assembler "vsm3rnds2\[ \\t\]+\[^\n\]*%xmm\[0-9\]+\[^\n\]*%xmm\[0-9\]+\[^\n\]*%xmm\[0-9\]" } } */
+
+#include <immintrin.h>
+
+volatile __m128i x, y, z;
+
+void extern
+sm3_test (void)
+{
+  x = _mm_sm3msg1_epi32 (x, y, z);
+  x = _mm_sm3msg2_epi32 (x, y, z);
+  x = _mm_sm3rnds2_epi32 (x, y, z, 1);
+}
diff --git a/gcc/testsuite/gcc.target/i386/sm3-check.h b/gcc/testsuite/gcc.target/i386/sm3-check.h
new file mode 100644
index 00000000000..ad9847402fc
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sm3-check.h
@@ -0,0 +1,37 @@
+#include <stdlib.h>
+#include "m128-check.h"
+
+static void sm3_test (void);
+
+static unsigned
+rol32 (unsigned w, int n)
+{
+  int count = n % 32;
+  return ((w << n) | (w >> (32 - n)));
+}
+
+static void
+__attribute__ ((noinline))
+do_test (void)
+{
+  sm3_test ();
+}
+
+int
+main ()
+{
+  /* Run SM3 test only if host has SM3 support.  */
+  if (__builtin_cpu_supports ("sm3"))
+    {
+      do_test ();
+#ifdef DEBUG
+      printf ("PASSED\n");
+#endif
+      return 0;
+    }
+
+#ifdef DEBUG
+  printf ("SKIPPED\n");
+#endif
+  return 0;
+}
diff --git a/gcc/testsuite/gcc.target/i386/sm3msg1-2.c b/gcc/testsuite/gcc.target/i386/sm3msg1-2.c
new file mode 100644
index 00000000000..e08abf5539d
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sm3msg1-2.c
@@ -0,0 +1,54 @@
+/* { dg-do run } */
+/* { dg-options "-O2  -msm3" } */
+/* { dg-require-effective-target sm3 } */
+
+#include "sm3-check.h"
+#include <x86intrin.h>
+#include <immintrin.h>
+
+static unsigned
+p1 (unsigned w)
+{
+  return rol32 (w, 15) ^ rol32 (w, 23) ^ w;
+}
+
+static void
+compute_sm3msg1 (int *src0, int *src1, int *src2, int *res)
+{
+  unsigned w0, w1, w2, w3, w7, w8, w9, w10, w13, w14, w15;
+
+  w0 = src2[0];
+  w1 = src2[1];
+  w2 = src2[2];
+  w3 = src2[3];
+  w7 = src0[0];
+  w8 = src0[1];
+  w9 = src0[2];
+  w10 = src0[3];
+  w13 = src1[0];
+  w14 = src1[1];
+  w15 = src1[2];
+
+  res[0] = p1 (w7 ^ w0 ^ rol32 (w13, 15));
+  res[1] = p1 (w8 ^ w1 ^ rol32 (w14, 15));
+  res[2] = p1 (w9 ^ w2 ^ rol32 (w15, 15));
+  res[3] = p1 (w10 ^ w3);
+}
+
+static void
+sm3_test (void)
+{
+  union128i_d s1, s2, s3, res;
+  int res_ref[4];
+
+  s1.x = _mm_set_epi32 (111, 222, 333, 444);
+  s2.x = _mm_set_epi32 (555, 666, 777, 888);
+  s3.x = _mm_set_epi32 (999, 123, 456, 789);
+
+  res.x = _mm_sm3msg1_epi32 (s1.x, s2.x, s3.x);
+
+  compute_sm3msg1 (s1.a, s2.a, s3.a, res_ref);
+
+  if (check_union128i_d (res, res_ref))
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sm3msg2-2.c b/gcc/testsuite/gcc.target/i386/sm3msg2-2.c
new file mode 100644
index 00000000000..f5986313c7d
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sm3msg2-2.c
@@ -0,0 +1,57 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -msm3" } */
+/* { dg-require-effective-target sm3 } */
+
+#include "sm3-check.h"
+#include <x86intrin.h>
+#include <immintrin.h>
+
+static void
+compute_sm3msg2 (int *src0, int *src1, int *src2, int *res)
+{
+  unsigned wtmp0, wtmp1, wtmp2, wtmp3, w3, w4, w5, w6, w10, w11, w12, w13,
+	   w16, w17, w18, w19;
+
+  wtmp0 = src0[0];
+  wtmp1 = src0[1];
+  wtmp2 = src0[2];
+  wtmp3 = src0[3];
+  w3 = src1[0];
+  w4 = src1[1];
+  w5 = src1[2];
+  w6 = src1[3];
+  w10 = src2[0];
+  w11 = src2[1];
+  w12 = src2[2];
+  w13 = src2[3];
+
+  w16 = rol32 (w3, 7) ^ w10 ^ wtmp0;
+  w17 = rol32 (w4, 7)  ^ w11 ^ wtmp1;
+  w18 = rol32 (w5, 7)  ^ w12 ^ wtmp2;
+  w19 = rol32 (w6, 7)  ^ w13 ^ wtmp3;
+
+  w19 = w19 ^ rol32 (w16, 6) ^ rol32 (w16, 15)  ^ rol32 (w16, 30) ;
+
+  res[0] = w16;
+  res[1] = w17;
+  res[2] = w18;
+  res[3] = w19;
+}
+
+static void
+sm3_test (void)
+{
+  union128i_d s1, s2, s3, res;
+  int res_ref[4];
+
+  s1.x = _mm_set_epi32 (111, 222, 333, 444);
+  s2.x = _mm_set_epi32 (555, 666, 777, 888);
+  s3.x = _mm_set_epi32 (999, 123, 456, 789);
+
+  res.x = _mm_sm3msg2_epi32 (s1.x, s2.x, s3.x);
+
+  compute_sm3msg2 (s1.a, s2.a, s3.a, res_ref);
+
+  if (check_union128i_d (res, res_ref))
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sm3rnds2-2.c b/gcc/testsuite/gcc.target/i386/sm3rnds2-2.c
new file mode 100644
index 00000000000..ffa3ed125f3
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sm3rnds2-2.c
@@ -0,0 +1,104 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -msm3" } */
+/* { dg-require-effective-target sm3 } */
+
+#include "sm3-check.h"
+#include <x86intrin.h>
+#include <immintrin.h>
+
+static unsigned
+p0 (unsigned w)
+{
+  return (w ^ rol32 (w, 9) ^ rol32 (w, 17));
+}
+
+static unsigned
+ff (unsigned x, unsigned y, unsigned z, int round)
+{
+  if (round < 16)
+    return (x ^ y ^ z);
+  else
+    return ((x & y) | (x & z) | (y & z));
+}
+
+static unsigned
+gg (unsigned x, unsigned y, unsigned z, int round)
+{
+  if (round < 16)
+    return (x ^ y ^ z);
+  else
+    return ((x & y) | ((~x) & z));
+}
+
+static void
+compute_sm3rnds2 (int *src0, int *src1, int *src2, int imm, int *res)
+{
+  unsigned s1, s2, t1, t2, co;
+  unsigned w[6], a[3], b[3], c[3], d[3], e[3], f[3], g[3], h[3];
+  int round, i;
+
+  a[0] = src1[3];
+  b[0] = src1[2];
+  c[0] = src0[3];
+  d[0] = src0[2];
+  e[0] = src1[1];
+  f[0] = src1[0];
+  g[0] = src0[1];
+  h[0] = src0[0];
+  w[0] = src2[0];
+  w[1] = src2[1];
+  w[4] = src2[2];
+  w[5] = src2[3];
+
+  c[0] = rol32 (c[0], 9);
+  d[0] = rol32 (d[0], 9);
+  g[0] = rol32 (g[0], 19);
+  h[0] = rol32 (h[0], 19);
+
+  round = imm & 0x3e;
+  if (round < 16)
+    co = 0x79cc4519;
+  else
+    co = 0x7a879d8a;
+  co = rol32 (co, round);
+
+  for (i = 0; i < 2; i++)
+  {
+    s1 = rol32 ((rol32 (a[i], 12) + e[i] + co), 7);
+    s2 = s1 ^ rol32 (a[i], 12);
+    t1 = ff (a[i], b[i], c[i], round) + d[i] + s2 + (w[i] ^ w[i + 4]);
+    t2 = gg (e[i], f[i], g[i], round) + h[i] + s1 + w[i];
+    d[i + 1] = c[i];
+    c[i + 1] = rol32 (b[i], 9);
+    b[i + 1] = a[i];
+    a[i + 1] = t1;
+    h[i + 1] = g[i];
+    g[i + 1] = rol32 (f[i], 19);
+    f[i + 1] = e[i];
+    e[i + 1] = p0 (t2);
+    co = rol32 (co, 1);
+  }
+
+  res[3] = a[2];
+  res[2] = b[2];
+  res[1] = e[2];
+  res[0] = f[2];
+}
+
+static void
+sm3_test (void)
+{
+  union128i_d s1, s2, s3, res;
+  int res_ref[4];
+
+  s1.x = _mm_set_epi32 (111, 222, 333, 444);
+  s2.x = _mm_set_epi32 (555, 666, 777, 888);
+  s3.x = _mm_set_epi32 (999, 123, 456, 789);
+
+  res.x = _mm_sm3rnds2_epi32 (s1.x, s2.x, s3.x, 22);
+
+  compute_sm3rnds2 (s1.a, s2.a, s3.a, 22, res_ref);
+
+  if (check_union128i_d (res, res_ref))
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse-12.c b/gcc/testsuite/gcc.target/i386/sse-12.c
index 2b7d78c51d3..5058be6f6e9 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 -mavxvnniint8 -mavxneconvert -mamx-fp16 -mraoint -mamx-complex -mavxvnniint16" } */
+/* { 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 -mavxneconvert -mamx-fp16 -mraoint -mamx-complex -mavxvnniint16 -msm3" } */
 
 #include <x86intrin.h>
 
diff --git a/gcc/testsuite/gcc.target/i386/sse-13.c b/gcc/testsuite/gcc.target/i386/sse-13.c
index 33693484d2a..d30b365564a 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 -mavxvnniint8 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16" } */
+/* { 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 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3" } */
 /* { dg-add-options bind_pic_locally } */
 
 #include <mm_malloc.h>
@@ -846,4 +846,7 @@
 #define __builtin_ia32_cmpccxadd(A, B, C, D) __builtin_ia32_cmpccxadd(A, B, C, 1)
 #define __builtin_ia32_cmpccxadd64(A, B, C, D) __builtin_ia32_cmpccxadd64(A, B, C, 1)
 
+/* sm3intrin.h */
+#define __builtin_ia32_vsm3rnds2(A, B, C, D) __builtin_ia32_vsm3rnds2 (A, B, C, 1)
+
 #include <x86intrin.h>
diff --git a/gcc/testsuite/gcc.target/i386/sse-14.c b/gcc/testsuite/gcc.target/i386/sse-14.c
index 51c2946b25a..7842005a98b 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 -mavxvnniint8 -mavxneconvert -mamx-fp16 -mraoint -mamx-complex -mavxvnniint16" } */
+/* { 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 -mavxneconvert -mamx-fp16 -mraoint -mamx-complex -mavxvnniint16 -msm3" } */
 /* { dg-add-options bind_pic_locally } */
 
 #include <mm_malloc.h>
@@ -1054,3 +1054,6 @@ test_2 (_mm512_gf2p8affineinv_epi64_epi8, __m512i, __m512i, __m512i, 1)
 test_2 (_mm_gf2p8affine_epi64_epi8, __m128i, __m128i, __m128i, 1)
 test_2 (_mm256_gf2p8affine_epi64_epi8, __m256i, __m256i, __m256i, 1)
 test_2 (_mm512_gf2p8affine_epi64_epi8, __m512i, __m512i, __m512i, 1)
+
+/* sm3intrin.h */
+test_3 (_mm_sm3rnds2_epi32, __m128i, __m128i, __m128i, __m128i, 1)
diff --git a/gcc/testsuite/gcc.target/i386/sse-22.c b/gcc/testsuite/gcc.target/i386/sse-22.c
index 4982fde2a76..7537db1ac30 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,tbm,lwp,fsgsbase,rdrnd,f16c,rtm,rdseed,prfchw,adx,fxsr,xsaveopt,avx512f,avx512er,avx512cd,avx512pf,sha,prefetchwt1,avx512vl,avx512bw,avx512dq,avx512vbmi,avx512vbmi2,avx512ifma,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,gfni,avx512bitalg,avx512bf16,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,amx-fp16,raoint,amx-complex,avxvnniint16")
+#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,avx512f,avx512er,avx512cd,avx512pf,sha,prefetchwt1,avx512vl,avx512bw,avx512dq,avx512vbmi,avx512vbmi2,avx512ifma,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,gfni,avx512bitalg,avx512bf16,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,amx-fp16,raoint,amx-complex,avxvnniint16,sm3")
 #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,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,gfni,avx512bitalg,avx512bf16,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,amx-fp16,raoint,amx-complex,avxvnniint16")
+#pragma GCC target ("avx,avx2,rdrnd,fsgsbase,f16c,rtm,avx512f,avx512er,avx512cd,avx512pf,sha,avx512vl,avx512bw,avx512dq,avx512ifma,avx512vbmi,avx512vbmi2,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,gfni,avx512bitalg,avx512bf16,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,amx-fp16,raoint,amx-complex,avxvnniint16,sm3")
 #endif
 #include <immintrin.h>
 test_1 (_cvtss_sh, unsigned short, float, 1)
@@ -1099,3 +1099,6 @@ test_1 ( __bextri_u32, unsigned int, unsigned int, 1)
 #ifdef __x86_64__
 test_1 ( __bextri_u64, unsigned long long, unsigned long long, 1)
 #endif
+
+/* sm3intrin.h */
+test_3 (_mm_sm3rnds2_epi32, __m128i, __m128i, __m128i, __m128i, 1)
diff --git a/gcc/testsuite/gcc.target/i386/sse-23.c b/gcc/testsuite/gcc.target/i386/sse-23.c
index 7e9c9f2ca2b..3fc61b50fe6 100644
--- a/gcc/testsuite/gcc.target/i386/sse-23.c
+++ b/gcc/testsuite/gcc.target/i386/sse-23.c
@@ -847,6 +847,9 @@
 #define __builtin_ia32_cmpccxadd(A, B, C, D) __builtin_ia32_cmpccxadd(A, B, C, 1)
 #define __builtin_ia32_cmpccxadd64(A, B, C, D) __builtin_ia32_cmpccxadd64(A, B, C, 1)
 
-#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,avx512f,avx512er,avx512cd,avx512pf,sha,prefetchwt1,xsavec,xsaves,clflushopt,avx512bw,avx512dq,avx512vl,avx512vbmi,avx512ifma,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,clwb,mwaitx,clzero,pku,sgx,rdpid,gfni,avx512vbmi2,vpclmulqdq,avx512bitalg,pconfig,wbnoinvd,avx512bf16,enqcmd,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,cmpccxadd,amx-fp16,prefetchi,raoint,amx-complex,avxvnniint16")
+/* sm3intrin.h */
+#define __builtin_ia32_vsm3rnds2(A, B, C, D) __builtin_ia32_vsm3rnds2 (A, B, C, 1)
+
+#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,avx512f,avx512er,avx512cd,avx512pf,sha,prefetchwt1,xsavec,xsaves,clflushopt,avx512bw,avx512dq,avx512vl,avx512vbmi,avx512ifma,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,clwb,mwaitx,clzero,pku,sgx,rdpid,gfni,avx512vbmi2,vpclmulqdq,avx512bitalg,pconfig,wbnoinvd,avx512bf16,enqcmd,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,cmpccxadd,amx-fp16,prefetchi,raoint,amx-complex,avxvnniint16,sm3")
 
 #include <x86intrin.h>
diff --git a/gcc/testsuite/lib/target-supports.exp b/gcc/testsuite/lib/target-supports.exp
index 60de239f1ce..c911a824d31 100644
--- a/gcc/testsuite/lib/target-supports.exp
+++ b/gcc/testsuite/lib/target-supports.exp
@@ -9868,6 +9868,21 @@ proc check_effective_target_avxvnniint16 { } {
     } "-O0 -mavxvnniint16" ]
 }
 
+# Return 1 if sm3 instructions can be compiled.
+proc check_effective_target_sm3 { } {
+    return [check_no_compiler_messages sm3 object {
+	typedef long long __m128i __attribute__ ((__vector_size__ (16)));
+	typedef int __v4si __attribute__ ((__vector_size__ (16)));
+	__m128i
+	_mm_sm3msg1_epi32 (__m128i __A, __m128i __B, __m128i __C)
+	{
+	  return (__m128i) __builtin_ia32_vsm3msg1 ((__v4si) __A,
+						    (__v4si) __B,
+						    (__v4si) __C);
+	}
+    } "-msm3" ]
+}
+
 # Return 1 if sse instructions can be compiled.
 proc check_effective_target_sse { } {
     return [check_no_compiler_messages sse object {
-- 
2.31.1


^ permalink raw reply	[flat|nested] 9+ messages in thread

* Re: [PATCH 2/4] Support Intel SM3
  2023-07-13  6:03 ` [PATCH 2/4] Support Intel SM3 Haochen Jiang
@ 2023-07-17  1:39   ` Hongtao Liu
  0 siblings, 0 replies; 9+ messages in thread
From: Hongtao Liu @ 2023-07-17  1:39 UTC (permalink / raw)
  To: Haochen Jiang; +Cc: gcc-patches, hongtao.liu, ubizjak

On Thu, Jul 13, 2023 at 2:04 PM Haochen Jiang via Gcc-patches
<gcc-patches@gcc.gnu.org> wrote:
>
> gcc/ChangeLog:
>
>         * common/config/i386/cpuinfo.h (get_available_features):
>         Detect SM3.
>         * common/config/i386/i386-common.cc (OPTION_MASK_ISA2_SM3_SET,
>         OPTION_MASK_ISA2_SM3_UNSET): New.
>         (OPTION_MASK_ISA2_AVX_UNSET): Add SM3.
>         (ix86_handle_option): Handle -msm3.
>         * common/config/i386/i386-cpuinfo.h (enum processor_features):
>         Add FEATURE_SM3.
>         * common/config/i386/i386-isas.h: Add ISA_NAME_TABLE_ENTRY for
>         SM3.
>         * config.gcc: Add sm3intrin.h
>         * config/i386/cpuid.h (bit_SM3): New.
>         * config/i386/i386-builtin-types.def:
>         Add DEF_FUNCTION_TYPE (V4SI, V4SI, V4SI, V4SI, INT).
>         * config/i386/i386-builtin.def (BDESC): Add new builtins.
>         * config/i386/i386-c.cc (ix86_target_macros_internal): Define
>         __SM3__.
>         * config/i386/i386-expand.cc (ix86_expand_args_builtin): Handle
>         V4SI_FTYPE_V4SI_V4SI_V4SI_INT.
>         * config/i386/i386-isa.def (SM3): Add DEF_PTA(SM3).
>         * config/i386/i386-options.cc (isa2_opts): Add -msm3.
>         (ix86_valid_target_attribute_inner_p): Handle sm3.
>         * config/i386/i386.opt: Add option -msm3.
>         * config/i386/immintrin.h: Include sm3intrin.h.
>         * config/i386/sse.md (vsm3msg1): New define insn.
>         (vsm3msg2): Ditto.
>         (vsm3rnds2): Ditto.
>         * doc/extend.texi: Document sm3.
>         * doc/invoke.texi: Document -msm3.
>         * doc/sourcebuild.texi: Document target sm3.
>         * config/i386/sm3intrin.h: New file.
>
> gcc/testsuite/ChangeLog:
>
>         * g++.dg/other/i386-2.C: Add -msm3.
>         * g++.dg/other/i386-3.C: Ditto.
>         * gcc.target/i386/avx-1.c: Add new define for immediate.
>         * gcc.target/i386/funcspec-56.inc: Add new target attribute.
>         * gcc.target/i386/sse-12.c: Add -msm3.
>         * gcc.target/i386/sse-13.c: Ditto.
>         * gcc.target/i386/sse-14.c: Ditto.
>         * gcc.target/i386/sse-22.c: Add sm3.
>         * gcc.target/i386/sse-23.c: Ditto.
>         * lib/target-supports.exp (check_effective_target_sm3): New.
>         * gcc.target/i386/sm3-1.c: New test.
>         * gcc.target/i386/sm3-check.h: Ditto.
>         * gcc.target/i386/sm3msg1-2.c: Ditto.
>         * gcc.target/i386/sm3msg2-2.c: Ditto.
>         * gcc.target/i386/sm3rnds2-2.c: Ditto.
Ok.
> ---
>  gcc/common/config/i386/cpuinfo.h              |   2 +
>  gcc/common/config/i386/i386-common.cc         |  20 +++-
>  gcc/common/config/i386/i386-cpuinfo.h         |   1 +
>  gcc/common/config/i386/i386-isas.h            |   1 +
>  gcc/config.gcc                                |   3 +-
>  gcc/config/i386/cpuid.h                       |   1 +
>  gcc/config/i386/i386-builtin-types.def        |   3 +
>  gcc/config/i386/i386-builtin.def              |   5 +
>  gcc/config/i386/i386-c.cc                     |   2 +
>  gcc/config/i386/i386-expand.cc                |   1 +
>  gcc/config/i386/i386-isa.def                  |   1 +
>  gcc/config/i386/i386-options.cc               |   2 +
>  gcc/config/i386/i386.opt                      |   5 +
>  gcc/config/i386/immintrin.h                   |   2 +
>  gcc/config/i386/sm3intrin.h                   |  72 ++++++++++++
>  gcc/config/i386/sse.md                        |  43 ++++++++
>  gcc/doc/extend.texi                           |   5 +
>  gcc/doc/invoke.texi                           |   7 +-
>  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-1.c         |   3 +
>  gcc/testsuite/gcc.target/i386/funcspec-56.inc |   2 +
>  gcc/testsuite/gcc.target/i386/sm3-1.c         |  17 +++
>  gcc/testsuite/gcc.target/i386/sm3-check.h     |  37 +++++++
>  gcc/testsuite/gcc.target/i386/sm3msg1-2.c     |  54 +++++++++
>  gcc/testsuite/gcc.target/i386/sm3msg2-2.c     |  57 ++++++++++
>  gcc/testsuite/gcc.target/i386/sm3rnds2-2.c    | 104 ++++++++++++++++++
>  gcc/testsuite/gcc.target/i386/sse-12.c        |   2 +-
>  gcc/testsuite/gcc.target/i386/sse-13.c        |   5 +-
>  gcc/testsuite/gcc.target/i386/sse-14.c        |   5 +-
>  gcc/testsuite/gcc.target/i386/sse-22.c        |   7 +-
>  gcc/testsuite/gcc.target/i386/sse-23.c        |   5 +-
>  gcc/testsuite/lib/target-supports.exp         |  15 +++
>  34 files changed, 484 insertions(+), 12 deletions(-)
>  create mode 100644 gcc/config/i386/sm3intrin.h
>  create mode 100644 gcc/testsuite/gcc.target/i386/sm3-1.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sm3-check.h
>  create mode 100644 gcc/testsuite/gcc.target/i386/sm3msg1-2.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sm3msg2-2.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sm3rnds2-2.c
>
> diff --git a/gcc/common/config/i386/cpuinfo.h b/gcc/common/config/i386/cpuinfo.h
> index 3599f9def2c..e5cdffe017a 100644
> --- a/gcc/common/config/i386/cpuinfo.h
> +++ b/gcc/common/config/i386/cpuinfo.h
> @@ -877,6 +877,8 @@ get_available_features (struct __processor_model *cpu_model,
>             set_feature (FEATURE_AVXNECONVERT);
>           if (edx & bit_AVXVNNIINT16)
>             set_feature (FEATURE_AVXVNNIINT16);
> +         if (eax & bit_SM3)
> +           set_feature (FEATURE_SM3);
>         }
>        if (avx512_usable)
>         {
> diff --git a/gcc/common/config/i386/i386-common.cc b/gcc/common/config/i386/i386-common.cc
> index 32c6d00580d..57b008ca3af 100644
> --- a/gcc/common/config/i386/i386-common.cc
> +++ b/gcc/common/config/i386/i386-common.cc
> @@ -120,6 +120,7 @@ along with GCC; see the file COPYING3.  If not see
>  #define OPTION_MASK_ISA2_AMX_COMPLEX_SET \
>    (OPTION_MASK_ISA2_AMX_TILE | OPTION_MASK_ISA2_AMX_COMPLEX)
>  #define OPTION_MASK_ISA2_AVXVNNIINT16_SET OPTION_MASK_ISA2_AVXVNNIINT16
> +#define OPTION_MASK_ISA2_SM3_SET OPTION_MASK_ISA2_SM3
>
>  /* SSE4 includes both SSE4.1 and SSE4.2. -msse4 should be the same
>     as -msse4.2.  */
> @@ -303,6 +304,7 @@ along with GCC; see the file COPYING3.  If not see
>  #define OPTION_MASK_ISA2_RAOINT_UNSET OPTION_MASK_ISA2_RAOINT
>  #define OPTION_MASK_ISA2_AMX_COMPLEX_UNSET OPTION_MASK_ISA2_AMX_COMPLEX
>  #define OPTION_MASK_ISA2_AVXVNNIINT16_UNSET OPTION_MASK_ISA2_AVXVNNIINT16
> +#define OPTION_MASK_ISA2_SM3_UNSET OPTION_MASK_ISA2_SM3
>
>  /* SSE4 includes both SSE4.1 and SSE4.2.  -mno-sse4 should the same
>     as -mno-sse4.1. */
> @@ -351,7 +353,8 @@ along with GCC; see the file COPYING3.  If not see
>  #define OPTION_MASK_ISA2_GENERAL_REGS_ONLY_UNSET \
>    OPTION_MASK_ISA2_SSE_UNSET
>  #define OPTION_MASK_ISA2_AVX_UNSET \
> -  (OPTION_MASK_ISA2_AVX2_UNSET | OPTION_MASK_ISA2_VAES_UNSET)
> +  (OPTION_MASK_ISA2_AVX2_UNSET | OPTION_MASK_ISA2_VAES_UNSET \
> +   | OPTION_MASK_ISA2_SM3_UNSET)
>  #define OPTION_MASK_ISA2_SSE4_2_UNSET OPTION_MASK_ISA2_AVX_UNSET
>  #define OPTION_MASK_ISA2_SSE4_1_UNSET OPTION_MASK_ISA2_SSE4_2_UNSET
>  #define OPTION_MASK_ISA2_SSE4_UNSET OPTION_MASK_ISA2_SSE4_1_UNSET
> @@ -1288,6 +1291,21 @@ ix86_handle_option (struct gcc_options *opts,
>         }
>        return true;
>
> +    case OPT_msm3:
> +      if (value)
> +       {
> +         opts->x_ix86_isa_flags2 |= OPTION_MASK_ISA2_SM3_SET;
> +         opts->x_ix86_isa_flags2_explicit |= OPTION_MASK_ISA2_SM3_SET;
> +         opts->x_ix86_isa_flags |= OPTION_MASK_ISA_AVX_SET;
> +         opts->x_ix86_isa_flags_explicit |= OPTION_MASK_ISA_AVX_SET;
> +       }
> +      else
> +       {
> +         opts->x_ix86_isa_flags2 &= ~OPTION_MASK_ISA2_SM3_UNSET;
> +         opts->x_ix86_isa_flags2_explicit |= OPTION_MASK_ISA2_SM3_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 ae4e6a02f7f..c3403090c3b 100644
> --- a/gcc/common/config/i386/i386-cpuinfo.h
> +++ b/gcc/common/config/i386/i386-cpuinfo.h
> @@ -256,6 +256,7 @@ enum processor_features
>    FEATURE_RAOINT,
>    FEATURE_AMX_COMPLEX,
>    FEATURE_AVXVNNIINT16,
> +  FEATURE_SM3,
>    CPU_FEATURE_MAX
>  };
>
> diff --git a/gcc/common/config/i386/i386-isas.h b/gcc/common/config/i386/i386-isas.h
> index fc6abdedf24..961a7f0ccd4 100644
> --- a/gcc/common/config/i386/i386-isas.h
> +++ b/gcc/common/config/i386/i386-isas.h
> @@ -188,4 +188,5 @@ ISA_NAMES_TABLE_START
>                         P_NONE, "-mamx-complex")
>    ISA_NAMES_TABLE_ENTRY("avxvnniint16", FEATURE_AVXVNNIINT16,
>                         P_NONE, "-mavxvnniint16")
> +  ISA_NAMES_TABLE_ENTRY("sm3", FEATURE_SM3, P_NONE, "-msm3")
>  ISA_NAMES_TABLE_END
> diff --git a/gcc/config.gcc b/gcc/config.gcc
> index fc74d776048..fbd7360e355 100644
> --- a/gcc/config.gcc
> +++ b/gcc/config.gcc
> @@ -435,7 +435,8 @@ i[34567]86-*-* | x86_64-*-*)
>                        mwaitintrin.h avx512fp16intrin.h avx512fp16vlintrin.h
>                        avxifmaintrin.h avxvnniint8intrin.h avxneconvertintrin.h
>                        cmpccxaddintrin.h amxfp16intrin.h prfchiintrin.h
> -                      raointintrin.h amxcomplexintrin.h avxvnniint16intrin.h"
> +                      raointintrin.h amxcomplexintrin.h avxvnniint16intrin.h
> +                      sm3intrin.h"
>         ;;
>  ia64-*-*)
>         extra_headers=ia64intrin.h
> diff --git a/gcc/config/i386/cpuid.h b/gcc/config/i386/cpuid.h
> index 98d0f193d22..28a36ad0628 100644
> --- a/gcc/config/i386/cpuid.h
> +++ b/gcc/config/i386/cpuid.h
> @@ -132,6 +132,7 @@
>
>  /* Extended Features Sub-leaf (%eax == 7, %ecx == 1) */
>  /* %eax */
> +#define bit_SM3                (1 << 1)
>  #define bit_RAOINT      (1 << 3)
>  #define bit_AVXVNNI     (1 << 4)
>  #define bit_AVX512BF16  (1 << 5)
> diff --git a/gcc/config/i386/i386-builtin-types.def b/gcc/config/i386/i386-builtin-types.def
> index cb2d0cd56ed..899eac1e014 100644
> --- a/gcc/config/i386/i386-builtin-types.def
> +++ b/gcc/config/i386/i386-builtin-types.def
> @@ -1416,3 +1416,6 @@ DEF_FUNCTION_TYPE (LONGLONG, PLONGLONG, LONGLONG, LONGLONG, INT)
>  # PREFETCHI builtins
>  DEF_FUNCTION_TYPE (VOID, PCVOID, INT)
>  DEF_FUNCTION_TYPE (VOID, PCVOID, INT, INT, INT)
> +
> +# SM3 builtins
> +DEF_FUNCTION_TYPE (V4SI, V4SI, V4SI, V4SI, INT)
> diff --git a/gcc/config/i386/i386-builtin.def b/gcc/config/i386/i386-builtin.def
> index ff5b3dcbcd3..17db19c2495 100644
> --- a/gcc/config/i386/i386-builtin.def
> +++ b/gcc/config/i386/i386-builtin.def
> @@ -1655,6 +1655,11 @@ BDESC (OPTION_MASK_ISA_SSE2, 0, CODE_FOR_sha256msg1, 0, IX86_BUILTIN_SHA256MSG1,
>  BDESC (OPTION_MASK_ISA_SSE2, 0, CODE_FOR_sha256msg2, 0, IX86_BUILTIN_SHA256MSG2, UNKNOWN, (int) V4SI_FTYPE_V4SI_V4SI)
>  BDESC (OPTION_MASK_ISA_SSE2, 0, CODE_FOR_sha256rnds2, 0, IX86_BUILTIN_SHA256RNDS2, UNKNOWN, (int) V4SI_FTYPE_V4SI_V4SI_V4SI)
>
> +/* SM3 */
> +BDESC (OPTION_MASK_ISA_AVX, OPTION_MASK_ISA2_SM3, CODE_FOR_vsm3msg1, "__builtin_ia32_vsm3msg1", IX86_BUILTIN_VSM3MSG1, UNKNOWN, (int) V4SI_FTYPE_V4SI_V4SI_V4SI)
> +BDESC (OPTION_MASK_ISA_AVX, OPTION_MASK_ISA2_SM3, CODE_FOR_vsm3msg2, "__builtin_ia32_vsm3msg2", IX86_BUILTIN_VSM3MSG2, UNKNOWN, (int) V4SI_FTYPE_V4SI_V4SI_V4SI)
> +BDESC (OPTION_MASK_ISA_AVX, OPTION_MASK_ISA2_SM3, CODE_FOR_vsm3rnds2, "__builtin_ia32_vsm3rnds2", IX86_BUILTIN_VSM3RNDS2, UNKNOWN, (int) V4SI_FTYPE_V4SI_V4SI_V4SI_INT)
> +
>  /* AVX512VL.  */
>  BDESC (OPTION_MASK_ISA_AVX512BW | OPTION_MASK_ISA_AVX512VL, 0, CODE_FOR_avx2_palignrv32qi_mask, "__builtin_ia32_palignr256_mask", IX86_BUILTIN_PALIGNR256_MASK, UNKNOWN, (int) V4DI_FTYPE_V4DI_V4DI_INT_V4DI_USI_CONVERT)
>  BDESC (OPTION_MASK_ISA_AVX512BW | OPTION_MASK_ISA_AVX512VL, 0, CODE_FOR_ssse3_palignrv16qi_mask, "__builtin_ia32_palignr128_mask", IX86_BUILTIN_PALIGNR128_MASK, UNKNOWN, (int) V2DI_FTYPE_V2DI_V2DI_INT_V2DI_UHI_CONVERT)
> diff --git a/gcc/config/i386/i386-c.cc b/gcc/config/i386/i386-c.cc
> index d3514dd46ac..0cb5a6dcce5 100644
> --- a/gcc/config/i386/i386-c.cc
> +++ b/gcc/config/i386/i386-c.cc
> @@ -679,6 +679,8 @@ ix86_target_macros_internal (HOST_WIDE_INT isa_flag,
>      def_or_undef (parse_in, "__AMX_COMPLEX__");
>    if (isa_flag2 & OPTION_MASK_ISA2_AVXVNNIINT16)
>      def_or_undef (parse_in, "__AVXVNNIINT16__");
> +  if (isa_flag2 & OPTION_MASK_ISA2_SM3)
> +    def_or_undef (parse_in, "__SM3__");
>    if (TARGET_IAMCU)
>      {
>        def_or_undef (parse_in, "__iamcu");
> diff --git a/gcc/config/i386/i386-expand.cc b/gcc/config/i386/i386-expand.cc
> index 648d6098eff..f6ad54c0cfe 100644
> --- a/gcc/config/i386/i386-expand.cc
> +++ b/gcc/config/i386/i386-expand.cc
> @@ -11202,6 +11202,7 @@ ix86_expand_args_builtin (const struct builtin_description *d,
>      case V4SF_FTYPE_V4SF_V4SF_V4SI_INT:
>      case V8SF_FTYPE_V8SF_V8SF_V8SI_INT:
>      case V16SF_FTYPE_V16SF_V16SF_V16SI_INT:
> +    case V4SI_FTYPE_V4SI_V4SI_V4SI_INT:
>        nargs = 4;
>        nargs_constant = 1;
>        break;
> diff --git a/gcc/config/i386/i386-isa.def b/gcc/config/i386/i386-isa.def
> index fbf22f7270a..432c36e7f79 100644
> --- a/gcc/config/i386/i386-isa.def
> +++ b/gcc/config/i386/i386-isa.def
> @@ -118,3 +118,4 @@ DEF_PTA(PREFETCHI)
>  DEF_PTA(RAOINT)
>  DEF_PTA(AMX_COMPLEX)
>  DEF_PTA(AVXVNNIINT16)
> +DEF_PTA(SM3)
> diff --git a/gcc/config/i386/i386-options.cc b/gcc/config/i386/i386-options.cc
> index d981666dd87..db2ff0c7ae1 100644
> --- a/gcc/config/i386/i386-options.cc
> +++ b/gcc/config/i386/i386-options.cc
> @@ -241,6 +241,7 @@ static struct ix86_target_opts isa2_opts[] =
>    { "-mraoint",        OPTION_MASK_ISA2_RAOINT },
>    { "-mamx-complex",   OPTION_MASK_ISA2_AMX_COMPLEX },
>    { "-mavxvnniint16",  OPTION_MASK_ISA2_AVXVNNIINT16 },
> +  { "-msm3",           OPTION_MASK_ISA2_SM3 }
>  };
>  static struct ix86_target_opts isa_opts[] =
>  {
> @@ -1093,6 +1094,7 @@ ix86_valid_target_attribute_inner_p (tree fndecl, tree args, char *p_strings[],
>      IX86_ATTR_ISA ("raoint", OPT_mraoint),
>      IX86_ATTR_ISA ("amx-complex", OPT_mamx_complex),
>      IX86_ATTR_ISA ("avxvnniint16", OPT_mavxvnniint16),
> +    IX86_ATTR_ISA ("sm3", OPT_msm3),
>
>      /* enum options */
>      IX86_ATTR_ENUM ("fpmath=", OPT_mfpmath_),
> diff --git a/gcc/config/i386/i386.opt b/gcc/config/i386/i386.opt
> index 618d713530f..80a8611993c 100644
> --- a/gcc/config/i386/i386.opt
> +++ b/gcc/config/i386/i386.opt
> @@ -1283,3 +1283,8 @@ mavxvnniint16
>  Target Mask(ISA2_AVXVNNIINT16) Var(ix86_isa_flags2) Save
>  Support MMX, SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2, AVX, AVX2 and
>  AVXVNNIINT16 built-in functions and code generation.
> +
> +msm3
> +Target Mask(ISA2_SM3) Var(ix86_isa_flags2) Save
> +Support MMX, SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2, AVX and
> +SM3 built-in functions and code generation.
> diff --git a/gcc/config/i386/immintrin.h b/gcc/config/i386/immintrin.h
> index 52dc35d8398..7731990131c 100644
> --- a/gcc/config/i386/immintrin.h
> +++ b/gcc/config/i386/immintrin.h
> @@ -108,6 +108,8 @@
>
>  #include <shaintrin.h>
>
> +#include <sm3intrin.h>
> +
>  #include <fmaintrin.h>
>
>  #include <f16cintrin.h>
> diff --git a/gcc/config/i386/sm3intrin.h b/gcc/config/i386/sm3intrin.h
> new file mode 100644
> index 00000000000..378c3dd41d9
> --- /dev/null
> +++ b/gcc/config/i386/sm3intrin.h
> @@ -0,0 +1,72 @@
> +/* Copyright (C) 2023 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/>.  */
> +
> +#ifndef _IMMINTRIN_H_INCLUDED
> +#error "Never use <sm3intrin.h> directly; include <immintrin.h> instead."
> +#endif
> +
> +#ifndef _SM3INTRIN_H_INCLUDED
> +#define _SM3INTRIN_H_INCLUDED
> +
> +#ifndef __SM3__
> +#pragma GCC push_options
> +#pragma GCC target("sm3")
> +#define __DISABLE_SM3__
> +#endif /* __SM3__ */
> +
> +extern __inline __m128i
> +__attribute__((__gnu_inline__, __always_inline__, __artificial__))
> +_mm_sm3msg1_epi32 (__m128i __A, __m128i __B, __m128i __C)
> +{
> +  return (__m128i) __builtin_ia32_vsm3msg1 ((__v4si) __A, (__v4si) __B,
> +                                           (__v4si) __C);
> +}
> +
> +extern __inline __m128i
> +__attribute__((__gnu_inline__, __always_inline__, __artificial__))
> +_mm_sm3msg2_epi32 (__m128i __A, __m128i __B, __m128i __C)
> +{
> +  return (__m128i) __builtin_ia32_vsm3msg2 ((__v4si) __A, (__v4si) __B,
> +                                           (__v4si) __C);
> +}
> +
> +#ifdef __OPTIMIZE__
> +extern __inline __m128i
> +__attribute__((__gnu_inline__, __always_inline__, __artificial__))
> +_mm_sm3rnds2_epi32 (__m128i __A, __m128i __B, __m128i __C, const int __D)
> +{
> +  return (__m128i) __builtin_ia32_vsm3rnds2 ((__v4si) __A, (__v4si) __B,
> +                                            (__v4si) __C, __D);
> +}
> +#else
> +#define _mm_sm3rnds2_epi32(A, B, C, D)                                 \
> +  ((__m128i) __builtin_ia32_vsm3rnds2 ((__v4si) (A), (__v4si) (B),     \
> +                                      (__v4si) (C), (int) (D)))
> +#endif
> +
> +#ifdef __DISABLE_SM3__
> +#undef __DISABLE_SM3__
> +#pragma GCC pop_options
> +#endif /* __DISABLE_SM3__ */
> +
> +#endif /* _SM3INTRIN_H_INCLUDED */
> diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md
> index 85a5f801e7a..25a1e5dd780 100644
> --- a/gcc/config/i386/sse.md
> +++ b/gcc/config/i386/sse.md
> @@ -196,6 +196,11 @@
>    UNSPEC_COMPLEX_FMUL
>    UNSPEC_COMPLEX_FCMUL
>    UNSPEC_COMPLEX_MASK
> +
> +  ;; For SM3 support
> +  UNSPEC_SM3MSG1
> +  UNSPEC_SM3MSG2
> +  UNSPEC_SM3RNDS2
>
>    ;; For AVX-VNNI-INT8 support
>    UNSPEC_VPDPBSSD
> @@ -28597,6 +28602,44 @@
>     (set_attr "length_immediate" "1")
>     (set_attr "mode" "TI")])
>
> +(define_insn "vsm3msg1"
> +  [(set (match_operand:V4SI 0 "register_operand" "=x")
> +       (unspec:V4SI
> +         [(match_operand:V4SI 1 "register_operand" "0")
> +          (match_operand:V4SI 2 "register_operand" "x")
> +          (match_operand:V4SI 3 "vector_operand" "xBm")]
> +         UNSPEC_SM3MSG1))]
> +  "TARGET_SM3"
> +  "vsm3msg1\t{%3, %2, %0|%0, %2, %3}"
> +  [(set_attr "type" "other")
> +   (set_attr "mode" "TI")])
> +
> +(define_insn "vsm3msg2"
> +  [(set (match_operand:V4SI 0 "register_operand" "=x")
> +       (unspec:V4SI
> +         [(match_operand:V4SI 1 "register_operand" "0")
> +          (match_operand:V4SI 2 "register_operand" "x")
> +          (match_operand:V4SI 3 "vector_operand" "xBm")]
> +         UNSPEC_SM3MSG2))]
> +  "TARGET_SM3"
> +  "vsm3msg2\t{%3, %2, %0|%0, %2, %3}"
> +  [(set_attr "type" "other")
> +   (set_attr "mode" "TI")])
> +
> +(define_insn "vsm3rnds2"
> +  [(set (match_operand:V4SI 0 "register_operand" "=x")
> +       (unspec:V4SI
> +         [(match_operand:V4SI 1 "register_operand" "0")
> +          (match_operand:V4SI 2 "register_operand" "x")
> +          (match_operand:V4SI 3 "vector_operand" "xBm")
> +          (match_operand:SI 4 "const_0_to_255_operand" "n")]
> +         UNSPEC_SM3RNDS2))]
> +  "TARGET_SM3"
> +  "vsm3rnds2\t{%4, %3, %2, %0|%0, %2, %3, %4}"
> +  [(set_attr "type" "other")
> +   (set_attr "mode" "TI")
> +   (set_attr "length_immediate" "1")])
> +
>  (define_insn_and_split "avx512f_<castmode><avxsizesuffix>_<castmode>"
>    [(set (match_operand:AVX512MODE2P 0 "nonimmediate_operand" "=x,m")
>         (vec_concat:AVX512MODE2P
> diff --git a/gcc/doc/extend.texi b/gcc/doc/extend.texi
> index 565bf1352e2..e76cd399a83 100644
> --- a/gcc/doc/extend.texi
> +++ b/gcc/doc/extend.texi
> @@ -7168,6 +7168,11 @@ Enable/disable the generation of the AMX-COMPLEX instructions.
>  @itemx no-avxvnniint16
>  Enable/disable the generation of the AVXVNNIINT16 instructions.
>
> +@cindex @code{target("sm3")} function attribute, x86
> +@item sm3
> +@itemx no-sm3
> +Enable/disable the generation of the SM3 instructions.
> +
>  @cindex @code{target("cld")} function attribute, x86
>  @item cld
>  @itemx no-cld
> diff --git a/gcc/doc/invoke.texi b/gcc/doc/invoke.texi
> index 359887db5fd..2671d70736f 100644
> --- a/gcc/doc/invoke.texi
> +++ b/gcc/doc/invoke.texi
> @@ -1433,7 +1433,7 @@ See RS/6000 and PowerPC Options.
>  -mrdseed  -msgx -mavx512vp2intersect -mserialize -mtsxldtrk
>  -mamx-tile  -mamx-int8  -mamx-bf16 -muintr -mhreset -mavxvnni
>  -mavx512fp16 -mavxifma -mavxvnniint8 -mavxneconvert -mcmpccxadd -mamx-fp16
> --mprefetchi -mraoint -mamx-complex -mavxvnniint16
> +-mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3
>  -mcldemote  -mms-bitfields  -mno-align-stringops  -minline-all-stringops
>  -minline-stringops-dynamically  -mstringop-strategy=@var{alg}
>  -mkl -mwidekl
> @@ -33555,6 +33555,9 @@ preferred alignment to @option{-mpreferred-stack-boundary=2}.
>  @need 200
>  @opindex mavxvnniint16
>  @itemx -mavxvnniint16
> +@need 200
> +@opindex msm3
> +@itemx -msm3
>  These switches enable the use of instructions in the MMX, SSE,
>  AVX512ER, AVX512CD, AVX512VL, AVX512BW, AVX512DQ, AVX512IFMA, AVX512VBMI, SHA,
>  AES, PCLMUL, CLFLUSHOPT, CLWB, FSGSBASE, PTWRITE, RDRND, F16C, FMA, PCONFIG,
> @@ -33565,7 +33568,7 @@ GFNI, VAES, WAITPKG, VPCLMULQDQ, AVX512BITALG, MOVDIRI, MOVDIR64B, AVX512BF16,
>  ENQCMD, AVX512VPOPCNTDQ, AVX5124FMAPS, AVX512VNNI, AVX5124VNNIW, SERIALIZE,
>  UINTR, HRESET, AMXTILE, AMXINT8, AMXBF16, KL, WIDEKL, AVXVNNI, AVX512-FP16,
>  AVXIFMA, AVXVNNIINT8, AVXNECONVERT, CMPCCXADD, AMX-FP16, PREFETCHI, RAOINT,
> -AMX-COMPLEX, AVXVNNIINT16 or CLDEMOTE extended instruction sets. Each has a
> +AMX-COMPLEX, AVXVNNIINT16, SM3 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
> diff --git a/gcc/doc/sourcebuild.texi b/gcc/doc/sourcebuild.texi
> index 40919b30a62..dae51132c42 100644
> --- a/gcc/doc/sourcebuild.texi
> +++ b/gcc/doc/sourcebuild.texi
> @@ -2589,6 +2589,9 @@ Target supports the execution of @code{raoint} instructions.
>  @item rdrand
>  Target supports x86 @code{rdrand} instruction.
>
> +@item sm3
> +Target supports the execution of @code{sm3} instructions.
> +
>  @item sqrt_insn
>  Target has a square root instruction that the compiler can generate.
>
> diff --git a/gcc/testsuite/g++.dg/other/i386-2.C b/gcc/testsuite/g++.dg/other/i386-2.C
> index 53622df2bb8..2ec93261cac 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 -mavxvnniint8 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16" } */
> +/* { 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 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3" } */
>
>  /* 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 3b76cee3af8..fe03143e39b 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 -mavxvnniint8 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16" } */
> +/* { 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 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3" } */
>
>  /* 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-1.c b/gcc/testsuite/gcc.target/i386/avx-1.c
> index 0b2b68b678d..a6589deca84 100644
> --- a/gcc/testsuite/gcc.target/i386/avx-1.c
> +++ b/gcc/testsuite/gcc.target/i386/avx-1.c
> @@ -839,6 +839,9 @@
>  #define __builtin_ia32_cmpccxadd(A, B, C, D) __builtin_ia32_cmpccxadd(A, B, C, 1)
>  #define __builtin_ia32_cmpccxadd64(A, B, C, D) __builtin_ia32_cmpccxadd64(A, B, C, 1)
>
> +/* sm3intrin.h */
> +#define __builtin_ia32_vsm3rnds2(A, B, C, D) __builtin_ia32_vsm3rnds2 (A, B, C, 1)
> +
>  #include <wmmintrin.h>
>  #include <immintrin.h>
>  #include <mm3dnow.h>
> diff --git a/gcc/testsuite/gcc.target/i386/funcspec-56.inc b/gcc/testsuite/gcc.target/i386/funcspec-56.inc
> index bba0fa37efd..8dd8d9bf9d8 100644
> --- a/gcc/testsuite/gcc.target/i386/funcspec-56.inc
> +++ b/gcc/testsuite/gcc.target/i386/funcspec-56.inc
> @@ -89,6 +89,7 @@ extern void test_prefetchi (void)               __attribute__((__target__("prefe
>  extern void test_raoint (void)                  __attribute__((__target__("raoint")));
>  extern void test_amx_complex (void)            __attribute__((__target__("amx-complex")));
>  extern void test_avxvnniint16 (void)           __attribute__((__target__("avxvnniint16")));
> +extern void test_sm3 (void)                    __attribute__((__target__("sm3")));
>
>  extern void test_no_sgx (void)                 __attribute__((__target__("no-sgx")));
>  extern void test_no_avx5124fmaps(void)         __attribute__((__target__("no-avx5124fmaps")));
> @@ -179,6 +180,7 @@ extern void test_no_prefetchi (void)            __attribute__((__target__("no-pr
>  extern void test_no_raoint (void)               __attribute__((__target__("no-raoint")));
>  extern void test_no_amx_complex (void)         __attribute__((__target__("no-amx-complex")));
>  extern void test_no_avxvnniint16 (void)                __attribute__((__target__("no-avxvnniint16")));
> +extern void test_no_sm3 (void)                 __attribute__((__target__("no-sm3")));
>
>  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/sm3-1.c b/gcc/testsuite/gcc.target/i386/sm3-1.c
> new file mode 100644
> index 00000000000..0a8ea658130
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/i386/sm3-1.c
> @@ -0,0 +1,17 @@
> +/* { dg-do compile } */
> +/* { dg-options "-O2 -msm3" } */
> +/* { dg-final { scan-assembler "vsm3msg1\[ \\t\]+\[^\n\]*%xmm\[0-9\]+\[^\n\]*%xmm\[0-9\]+\[^\n\]*%xmm\[0-9\]" } } */
> +/* { dg-final { scan-assembler "vsm3msg2\[ \\t\]+\[^\n\]*%xmm\[0-9\]+\[^\n\]*%xmm\[0-9\]+\[^\n\]*%xmm\[0-9\]" } } */
> +/* { dg-final { scan-assembler "vsm3rnds2\[ \\t\]+\[^\n\]*%xmm\[0-9\]+\[^\n\]*%xmm\[0-9\]+\[^\n\]*%xmm\[0-9\]" } } */
> +
> +#include <immintrin.h>
> +
> +volatile __m128i x, y, z;
> +
> +void extern
> +sm3_test (void)
> +{
> +  x = _mm_sm3msg1_epi32 (x, y, z);
> +  x = _mm_sm3msg2_epi32 (x, y, z);
> +  x = _mm_sm3rnds2_epi32 (x, y, z, 1);
> +}
> diff --git a/gcc/testsuite/gcc.target/i386/sm3-check.h b/gcc/testsuite/gcc.target/i386/sm3-check.h
> new file mode 100644
> index 00000000000..ad9847402fc
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/i386/sm3-check.h
> @@ -0,0 +1,37 @@
> +#include <stdlib.h>
> +#include "m128-check.h"
> +
> +static void sm3_test (void);
> +
> +static unsigned
> +rol32 (unsigned w, int n)
> +{
> +  int count = n % 32;
> +  return ((w << n) | (w >> (32 - n)));
> +}
> +
> +static void
> +__attribute__ ((noinline))
> +do_test (void)
> +{
> +  sm3_test ();
> +}
> +
> +int
> +main ()
> +{
> +  /* Run SM3 test only if host has SM3 support.  */
> +  if (__builtin_cpu_supports ("sm3"))
> +    {
> +      do_test ();
> +#ifdef DEBUG
> +      printf ("PASSED\n");
> +#endif
> +      return 0;
> +    }
> +
> +#ifdef DEBUG
> +  printf ("SKIPPED\n");
> +#endif
> +  return 0;
> +}
> diff --git a/gcc/testsuite/gcc.target/i386/sm3msg1-2.c b/gcc/testsuite/gcc.target/i386/sm3msg1-2.c
> new file mode 100644
> index 00000000000..e08abf5539d
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/i386/sm3msg1-2.c
> @@ -0,0 +1,54 @@
> +/* { dg-do run } */
> +/* { dg-options "-O2  -msm3" } */
> +/* { dg-require-effective-target sm3 } */
> +
> +#include "sm3-check.h"
> +#include <x86intrin.h>
> +#include <immintrin.h>
> +
> +static unsigned
> +p1 (unsigned w)
> +{
> +  return rol32 (w, 15) ^ rol32 (w, 23) ^ w;
> +}
> +
> +static void
> +compute_sm3msg1 (int *src0, int *src1, int *src2, int *res)
> +{
> +  unsigned w0, w1, w2, w3, w7, w8, w9, w10, w13, w14, w15;
> +
> +  w0 = src2[0];
> +  w1 = src2[1];
> +  w2 = src2[2];
> +  w3 = src2[3];
> +  w7 = src0[0];
> +  w8 = src0[1];
> +  w9 = src0[2];
> +  w10 = src0[3];
> +  w13 = src1[0];
> +  w14 = src1[1];
> +  w15 = src1[2];
> +
> +  res[0] = p1 (w7 ^ w0 ^ rol32 (w13, 15));
> +  res[1] = p1 (w8 ^ w1 ^ rol32 (w14, 15));
> +  res[2] = p1 (w9 ^ w2 ^ rol32 (w15, 15));
> +  res[3] = p1 (w10 ^ w3);
> +}
> +
> +static void
> +sm3_test (void)
> +{
> +  union128i_d s1, s2, s3, res;
> +  int res_ref[4];
> +
> +  s1.x = _mm_set_epi32 (111, 222, 333, 444);
> +  s2.x = _mm_set_epi32 (555, 666, 777, 888);
> +  s3.x = _mm_set_epi32 (999, 123, 456, 789);
> +
> +  res.x = _mm_sm3msg1_epi32 (s1.x, s2.x, s3.x);
> +
> +  compute_sm3msg1 (s1.a, s2.a, s3.a, res_ref);
> +
> +  if (check_union128i_d (res, res_ref))
> +    abort ();
> +}
> diff --git a/gcc/testsuite/gcc.target/i386/sm3msg2-2.c b/gcc/testsuite/gcc.target/i386/sm3msg2-2.c
> new file mode 100644
> index 00000000000..f5986313c7d
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/i386/sm3msg2-2.c
> @@ -0,0 +1,57 @@
> +/* { dg-do run } */
> +/* { dg-options "-O2 -msm3" } */
> +/* { dg-require-effective-target sm3 } */
> +
> +#include "sm3-check.h"
> +#include <x86intrin.h>
> +#include <immintrin.h>
> +
> +static void
> +compute_sm3msg2 (int *src0, int *src1, int *src2, int *res)
> +{
> +  unsigned wtmp0, wtmp1, wtmp2, wtmp3, w3, w4, w5, w6, w10, w11, w12, w13,
> +          w16, w17, w18, w19;
> +
> +  wtmp0 = src0[0];
> +  wtmp1 = src0[1];
> +  wtmp2 = src0[2];
> +  wtmp3 = src0[3];
> +  w3 = src1[0];
> +  w4 = src1[1];
> +  w5 = src1[2];
> +  w6 = src1[3];
> +  w10 = src2[0];
> +  w11 = src2[1];
> +  w12 = src2[2];
> +  w13 = src2[3];
> +
> +  w16 = rol32 (w3, 7) ^ w10 ^ wtmp0;
> +  w17 = rol32 (w4, 7)  ^ w11 ^ wtmp1;
> +  w18 = rol32 (w5, 7)  ^ w12 ^ wtmp2;
> +  w19 = rol32 (w6, 7)  ^ w13 ^ wtmp3;
> +
> +  w19 = w19 ^ rol32 (w16, 6) ^ rol32 (w16, 15)  ^ rol32 (w16, 30) ;
> +
> +  res[0] = w16;
> +  res[1] = w17;
> +  res[2] = w18;
> +  res[3] = w19;
> +}
> +
> +static void
> +sm3_test (void)
> +{
> +  union128i_d s1, s2, s3, res;
> +  int res_ref[4];
> +
> +  s1.x = _mm_set_epi32 (111, 222, 333, 444);
> +  s2.x = _mm_set_epi32 (555, 666, 777, 888);
> +  s3.x = _mm_set_epi32 (999, 123, 456, 789);
> +
> +  res.x = _mm_sm3msg2_epi32 (s1.x, s2.x, s3.x);
> +
> +  compute_sm3msg2 (s1.a, s2.a, s3.a, res_ref);
> +
> +  if (check_union128i_d (res, res_ref))
> +    abort ();
> +}
> diff --git a/gcc/testsuite/gcc.target/i386/sm3rnds2-2.c b/gcc/testsuite/gcc.target/i386/sm3rnds2-2.c
> new file mode 100644
> index 00000000000..ffa3ed125f3
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/i386/sm3rnds2-2.c
> @@ -0,0 +1,104 @@
> +/* { dg-do run } */
> +/* { dg-options "-O2 -msm3" } */
> +/* { dg-require-effective-target sm3 } */
> +
> +#include "sm3-check.h"
> +#include <x86intrin.h>
> +#include <immintrin.h>
> +
> +static unsigned
> +p0 (unsigned w)
> +{
> +  return (w ^ rol32 (w, 9) ^ rol32 (w, 17));
> +}
> +
> +static unsigned
> +ff (unsigned x, unsigned y, unsigned z, int round)
> +{
> +  if (round < 16)
> +    return (x ^ y ^ z);
> +  else
> +    return ((x & y) | (x & z) | (y & z));
> +}
> +
> +static unsigned
> +gg (unsigned x, unsigned y, unsigned z, int round)
> +{
> +  if (round < 16)
> +    return (x ^ y ^ z);
> +  else
> +    return ((x & y) | ((~x) & z));
> +}
> +
> +static void
> +compute_sm3rnds2 (int *src0, int *src1, int *src2, int imm, int *res)
> +{
> +  unsigned s1, s2, t1, t2, co;
> +  unsigned w[6], a[3], b[3], c[3], d[3], e[3], f[3], g[3], h[3];
> +  int round, i;
> +
> +  a[0] = src1[3];
> +  b[0] = src1[2];
> +  c[0] = src0[3];
> +  d[0] = src0[2];
> +  e[0] = src1[1];
> +  f[0] = src1[0];
> +  g[0] = src0[1];
> +  h[0] = src0[0];
> +  w[0] = src2[0];
> +  w[1] = src2[1];
> +  w[4] = src2[2];
> +  w[5] = src2[3];
> +
> +  c[0] = rol32 (c[0], 9);
> +  d[0] = rol32 (d[0], 9);
> +  g[0] = rol32 (g[0], 19);
> +  h[0] = rol32 (h[0], 19);
> +
> +  round = imm & 0x3e;
> +  if (round < 16)
> +    co = 0x79cc4519;
> +  else
> +    co = 0x7a879d8a;
> +  co = rol32 (co, round);
> +
> +  for (i = 0; i < 2; i++)
> +  {
> +    s1 = rol32 ((rol32 (a[i], 12) + e[i] + co), 7);
> +    s2 = s1 ^ rol32 (a[i], 12);
> +    t1 = ff (a[i], b[i], c[i], round) + d[i] + s2 + (w[i] ^ w[i + 4]);
> +    t2 = gg (e[i], f[i], g[i], round) + h[i] + s1 + w[i];
> +    d[i + 1] = c[i];
> +    c[i + 1] = rol32 (b[i], 9);
> +    b[i + 1] = a[i];
> +    a[i + 1] = t1;
> +    h[i + 1] = g[i];
> +    g[i + 1] = rol32 (f[i], 19);
> +    f[i + 1] = e[i];
> +    e[i + 1] = p0 (t2);
> +    co = rol32 (co, 1);
> +  }
> +
> +  res[3] = a[2];
> +  res[2] = b[2];
> +  res[1] = e[2];
> +  res[0] = f[2];
> +}
> +
> +static void
> +sm3_test (void)
> +{
> +  union128i_d s1, s2, s3, res;
> +  int res_ref[4];
> +
> +  s1.x = _mm_set_epi32 (111, 222, 333, 444);
> +  s2.x = _mm_set_epi32 (555, 666, 777, 888);
> +  s3.x = _mm_set_epi32 (999, 123, 456, 789);
> +
> +  res.x = _mm_sm3rnds2_epi32 (s1.x, s2.x, s3.x, 22);
> +
> +  compute_sm3rnds2 (s1.a, s2.a, s3.a, 22, res_ref);
> +
> +  if (check_union128i_d (res, res_ref))
> +    abort ();
> +}
> diff --git a/gcc/testsuite/gcc.target/i386/sse-12.c b/gcc/testsuite/gcc.target/i386/sse-12.c
> index 2b7d78c51d3..5058be6f6e9 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 -mavxvnniint8 -mavxneconvert -mamx-fp16 -mraoint -mamx-complex -mavxvnniint16" } */
> +/* { 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 -mavxneconvert -mamx-fp16 -mraoint -mamx-complex -mavxvnniint16 -msm3" } */
>
>  #include <x86intrin.h>
>
> diff --git a/gcc/testsuite/gcc.target/i386/sse-13.c b/gcc/testsuite/gcc.target/i386/sse-13.c
> index 33693484d2a..d30b365564a 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 -mavxvnniint8 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16" } */
> +/* { 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 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3" } */
>  /* { dg-add-options bind_pic_locally } */
>
>  #include <mm_malloc.h>
> @@ -846,4 +846,7 @@
>  #define __builtin_ia32_cmpccxadd(A, B, C, D) __builtin_ia32_cmpccxadd(A, B, C, 1)
>  #define __builtin_ia32_cmpccxadd64(A, B, C, D) __builtin_ia32_cmpccxadd64(A, B, C, 1)
>
> +/* sm3intrin.h */
> +#define __builtin_ia32_vsm3rnds2(A, B, C, D) __builtin_ia32_vsm3rnds2 (A, B, C, 1)
> +
>  #include <x86intrin.h>
> diff --git a/gcc/testsuite/gcc.target/i386/sse-14.c b/gcc/testsuite/gcc.target/i386/sse-14.c
> index 51c2946b25a..7842005a98b 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 -mavxvnniint8 -mavxneconvert -mamx-fp16 -mraoint -mamx-complex -mavxvnniint16" } */
> +/* { 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 -mavxneconvert -mamx-fp16 -mraoint -mamx-complex -mavxvnniint16 -msm3" } */
>  /* { dg-add-options bind_pic_locally } */
>
>  #include <mm_malloc.h>
> @@ -1054,3 +1054,6 @@ test_2 (_mm512_gf2p8affineinv_epi64_epi8, __m512i, __m512i, __m512i, 1)
>  test_2 (_mm_gf2p8affine_epi64_epi8, __m128i, __m128i, __m128i, 1)
>  test_2 (_mm256_gf2p8affine_epi64_epi8, __m256i, __m256i, __m256i, 1)
>  test_2 (_mm512_gf2p8affine_epi64_epi8, __m512i, __m512i, __m512i, 1)
> +
> +/* sm3intrin.h */
> +test_3 (_mm_sm3rnds2_epi32, __m128i, __m128i, __m128i, __m128i, 1)
> diff --git a/gcc/testsuite/gcc.target/i386/sse-22.c b/gcc/testsuite/gcc.target/i386/sse-22.c
> index 4982fde2a76..7537db1ac30 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,tbm,lwp,fsgsbase,rdrnd,f16c,rtm,rdseed,prfchw,adx,fxsr,xsaveopt,avx512f,avx512er,avx512cd,avx512pf,sha,prefetchwt1,avx512vl,avx512bw,avx512dq,avx512vbmi,avx512vbmi2,avx512ifma,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,gfni,avx512bitalg,avx512bf16,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,amx-fp16,raoint,amx-complex,avxvnniint16")
> +#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,avx512f,avx512er,avx512cd,avx512pf,sha,prefetchwt1,avx512vl,avx512bw,avx512dq,avx512vbmi,avx512vbmi2,avx512ifma,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,gfni,avx512bitalg,avx512bf16,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,amx-fp16,raoint,amx-complex,avxvnniint16,sm3")
>  #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,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,gfni,avx512bitalg,avx512bf16,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,amx-fp16,raoint,amx-complex,avxvnniint16")
> +#pragma GCC target ("avx,avx2,rdrnd,fsgsbase,f16c,rtm,avx512f,avx512er,avx512cd,avx512pf,sha,avx512vl,avx512bw,avx512dq,avx512ifma,avx512vbmi,avx512vbmi2,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,gfni,avx512bitalg,avx512bf16,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,amx-fp16,raoint,amx-complex,avxvnniint16,sm3")
>  #endif
>  #include <immintrin.h>
>  test_1 (_cvtss_sh, unsigned short, float, 1)
> @@ -1099,3 +1099,6 @@ test_1 ( __bextri_u32, unsigned int, unsigned int, 1)
>  #ifdef __x86_64__
>  test_1 ( __bextri_u64, unsigned long long, unsigned long long, 1)
>  #endif
> +
> +/* sm3intrin.h */
> +test_3 (_mm_sm3rnds2_epi32, __m128i, __m128i, __m128i, __m128i, 1)
> diff --git a/gcc/testsuite/gcc.target/i386/sse-23.c b/gcc/testsuite/gcc.target/i386/sse-23.c
> index 7e9c9f2ca2b..3fc61b50fe6 100644
> --- a/gcc/testsuite/gcc.target/i386/sse-23.c
> +++ b/gcc/testsuite/gcc.target/i386/sse-23.c
> @@ -847,6 +847,9 @@
>  #define __builtin_ia32_cmpccxadd(A, B, C, D) __builtin_ia32_cmpccxadd(A, B, C, 1)
>  #define __builtin_ia32_cmpccxadd64(A, B, C, D) __builtin_ia32_cmpccxadd64(A, B, C, 1)
>
> -#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,avx512f,avx512er,avx512cd,avx512pf,sha,prefetchwt1,xsavec,xsaves,clflushopt,avx512bw,avx512dq,avx512vl,avx512vbmi,avx512ifma,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,clwb,mwaitx,clzero,pku,sgx,rdpid,gfni,avx512vbmi2,vpclmulqdq,avx512bitalg,pconfig,wbnoinvd,avx512bf16,enqcmd,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,cmpccxadd,amx-fp16,prefetchi,raoint,amx-complex,avxvnniint16")
> +/* sm3intrin.h */
> +#define __builtin_ia32_vsm3rnds2(A, B, C, D) __builtin_ia32_vsm3rnds2 (A, B, C, 1)
> +
> +#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,avx512f,avx512er,avx512cd,avx512pf,sha,prefetchwt1,xsavec,xsaves,clflushopt,avx512bw,avx512dq,avx512vl,avx512vbmi,avx512ifma,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,clwb,mwaitx,clzero,pku,sgx,rdpid,gfni,avx512vbmi2,vpclmulqdq,avx512bitalg,pconfig,wbnoinvd,avx512bf16,enqcmd,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,cmpccxadd,amx-fp16,prefetchi,raoint,amx-complex,avxvnniint16,sm3")
>
>  #include <x86intrin.h>
> diff --git a/gcc/testsuite/lib/target-supports.exp b/gcc/testsuite/lib/target-supports.exp
> index 60de239f1ce..c911a824d31 100644
> --- a/gcc/testsuite/lib/target-supports.exp
> +++ b/gcc/testsuite/lib/target-supports.exp
> @@ -9868,6 +9868,21 @@ proc check_effective_target_avxvnniint16 { } {
>      } "-O0 -mavxvnniint16" ]
>  }
>
> +# Return 1 if sm3 instructions can be compiled.
> +proc check_effective_target_sm3 { } {
> +    return [check_no_compiler_messages sm3 object {
> +       typedef long long __m128i __attribute__ ((__vector_size__ (16)));
> +       typedef int __v4si __attribute__ ((__vector_size__ (16)));
> +       __m128i
> +       _mm_sm3msg1_epi32 (__m128i __A, __m128i __B, __m128i __C)
> +       {
> +         return (__m128i) __builtin_ia32_vsm3msg1 ((__v4si) __A,
> +                                                   (__v4si) __B,
> +                                                   (__v4si) __C);
> +       }
> +    } "-msm3" ]
> +}
> +
>  # Return 1 if sse instructions can be compiled.
>  proc check_effective_target_sse { } {
>      return [check_no_compiler_messages sse object {
> --
> 2.31.1
>


-- 
BR,
Hongtao

^ permalink raw reply	[flat|nested] 9+ messages in thread

* [PATCH 3/4] Support Intel SHA512
  2023-07-13  6:03 [PATCH 0/4] Support Intel Arrow Lake/Lunar Lake ISAs Haochen Jiang
  2023-07-13  6:03 ` [PATCH 1/4] Support Intel AVX-VNNI-INT16 Haochen Jiang
  2023-07-13  6:03 ` [PATCH 2/4] Support Intel SM3 Haochen Jiang
@ 2023-07-13  6:03 ` Haochen Jiang
  2023-07-17  1:40   ` Hongtao Liu
  2023-07-13  6:03 ` [PATCH 4/4] Support Intel SM4 Haochen Jiang
  3 siblings, 1 reply; 9+ messages in thread
From: Haochen Jiang @ 2023-07-13  6:03 UTC (permalink / raw)
  To: gcc-patches; +Cc: hongtao.liu, ubizjak

gcc/ChangeLog:

	* common/config/i386/cpuinfo.h (get_available_features):
	Detect SHA512.
	* common/config/i386/i386-common.cc (OPTION_MASK_ISA2_SHA512_SET,
	OPTION_MASK_ISA2_SHA512_UNSET): New.
	(OPTION_MASK_ISA2_AVX_UNSET): Add SHA512.
	(ix86_handle_option): Handle -msha512.
	* common/config/i386/i386-cpuinfo.h (enum processor_features):
	Add FEATURE_SHA512.
	* common/config/i386/i386-isas.h: Add ISA_NAME_TABLE_ENTRY for
	sha512.
	* config.gcc: Add sha512intrin.h.
	* config/i386/cpuid.h (bit_SHA512): New.
	* config/i386/i386-builtin-types.def:
	Add DEF_FUNCTION_TYPE (V4DI, V4DI, V4DI, V2DI).
	* config/i386/i386-builtin.def (BDESC): Add new builtins.
	* config/i386/i386-c.cc (ix86_target_macros_internal): Define
	__SHA512__.
	* config/i386/i386-expand.cc (ix86_expand_args_builtin): Handle
	V4DI_FTYPE_V4DI_V4DI_V2DI and V4DI_FTYPE_V4DI_V2DI.
	* config/i386/i386-isa.def (SHA512): Add DEF_PTA(SHA512).
	* config/i386/i386-options.cc (isa2_opts): Add -msha512.
	(ix86_valid_target_attribute_inner_p): Handle sha512.
	* config/i386/i386.opt: Add option -msha512.
	* config/i386/immintrin.h: Include sha512intrin.h.
	* config/i386/sse.md (vsha512msg1): New define insn.
        (vsha512msg2): Ditto.
        (vsha512rnds2): Ditto.
	* doc/extend.texi: Document sha512.
	* doc/invoke.texi: Document -msha512.
	* doc/sourcebuild.texi: Document target sha512.
	* config/i386/sha512intrin.h: New file.

gcc/testsuite/ChangeLog:

	* g++.dg/others/i386-2.C: Add -msha512.
	* g++.dg/others/i386-3.C: Ditto.
	* gcc.target/i386/funcspec-56.inc: Add new target attribute.
	* gcc.target/i386/sse-12.c: Add -msha512.
	* gcc.target/i386/sse-13.c: Ditto.
	* gcc.target/i386/sse-14.c: Ditto.
	* gcc.target/i386/sse-22.c: Add sha512.
 	* gcc.target/i386/sse-23.c: Ditto.
	* lib/target-supports.exp (check_effective_target_sha512): New.
	* gcc.target/i386/sha512-1.c: New test.
	* gcc.target/i386/sha512-check.h: Ditto.
	* gcc.target/i386/sha512msg1-2.c: Ditto.
	* gcc.target/i386/sha512msg2-2.c: Ditto.
	* gcc.target/i386/sha512rnds2-2.c: Ditto.
---
 gcc/common/config/i386/cpuinfo.h              |  2 +
 gcc/common/config/i386/i386-common.cc         | 19 ++++-
 gcc/common/config/i386/i386-cpuinfo.h         |  1 +
 gcc/common/config/i386/i386-isas.h            |  1 +
 gcc/config.gcc                                |  2 +-
 gcc/config/i386/cpuid.h                       |  1 +
 gcc/config/i386/i386-builtin-types.def        |  3 +
 gcc/config/i386/i386-builtin.def              |  5 ++
 gcc/config/i386/i386-c.cc                     |  2 +
 gcc/config/i386/i386-expand.cc                |  2 +
 gcc/config/i386/i386-isa.def                  |  1 +
 gcc/config/i386/i386-options.cc               |  4 +-
 gcc/config/i386/i386.opt                      | 10 +++
 gcc/config/i386/immintrin.h                   |  2 +
 gcc/config/i386/sha512intrin.h                | 64 ++++++++++++++
 gcc/config/i386/sse.md                        | 40 +++++++++
 gcc/doc/extend.texi                           |  5 ++
 gcc/doc/invoke.texi                           | 10 ++-
 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/funcspec-56.inc |  2 +
 gcc/testsuite/gcc.target/i386/sha512-1.c      | 18 ++++
 gcc/testsuite/gcc.target/i386/sha512-check.h  | 43 ++++++++++
 gcc/testsuite/gcc.target/i386/sha512msg1-2.c  | 48 +++++++++++
 gcc/testsuite/gcc.target/i386/sha512msg2-2.c  | 47 ++++++++++
 gcc/testsuite/gcc.target/i386/sha512rnds2-2.c | 85 +++++++++++++++++++
 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         | 14 +++
 33 files changed, 436 insertions(+), 14 deletions(-)
 create mode 100644 gcc/config/i386/sha512intrin.h
 create mode 100644 gcc/testsuite/gcc.target/i386/sha512-1.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sha512-check.h
 create mode 100644 gcc/testsuite/gcc.target/i386/sha512msg1-2.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sha512msg2-2.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sha512rnds2-2.c

diff --git a/gcc/common/config/i386/cpuinfo.h b/gcc/common/config/i386/cpuinfo.h
index e5cdffe017a..0cfde3ebccd 100644
--- a/gcc/common/config/i386/cpuinfo.h
+++ b/gcc/common/config/i386/cpuinfo.h
@@ -879,6 +879,8 @@ get_available_features (struct __processor_model *cpu_model,
 	    set_feature (FEATURE_AVXVNNIINT16);
 	  if (eax & bit_SM3)
 	    set_feature (FEATURE_SM3);
+	  if (eax & bit_SHA512)
+	    set_feature (FEATURE_SHA512);
 	}
       if (avx512_usable)
 	{
diff --git a/gcc/common/config/i386/i386-common.cc b/gcc/common/config/i386/i386-common.cc
index 57b008ca3af..97c3cdfe5e1 100644
--- a/gcc/common/config/i386/i386-common.cc
+++ b/gcc/common/config/i386/i386-common.cc
@@ -121,6 +121,7 @@ along with GCC; see the file COPYING3.  If not see
   (OPTION_MASK_ISA2_AMX_TILE | OPTION_MASK_ISA2_AMX_COMPLEX)
 #define OPTION_MASK_ISA2_AVXVNNIINT16_SET OPTION_MASK_ISA2_AVXVNNIINT16
 #define OPTION_MASK_ISA2_SM3_SET OPTION_MASK_ISA2_SM3
+#define OPTION_MASK_ISA2_SHA512_SET OPTION_MASK_ISA2_SHA512
 
 /* SSE4 includes both SSE4.1 and SSE4.2. -msse4 should be the same
    as -msse4.2.  */
@@ -305,6 +306,7 @@ along with GCC; see the file COPYING3.  If not see
 #define OPTION_MASK_ISA2_AMX_COMPLEX_UNSET OPTION_MASK_ISA2_AMX_COMPLEX
 #define OPTION_MASK_ISA2_AVXVNNIINT16_UNSET OPTION_MASK_ISA2_AVXVNNIINT16
 #define OPTION_MASK_ISA2_SM3_UNSET OPTION_MASK_ISA2_SM3
+#define OPTION_MASK_ISA2_SHA512_UNSET OPTION_MASK_ISA2_SHA512
 
 /* SSE4 includes both SSE4.1 and SSE4.2.  -mno-sse4 should the same
    as -mno-sse4.1. */
@@ -354,7 +356,7 @@ along with GCC; see the file COPYING3.  If not see
   OPTION_MASK_ISA2_SSE_UNSET
 #define OPTION_MASK_ISA2_AVX_UNSET \
   (OPTION_MASK_ISA2_AVX2_UNSET | OPTION_MASK_ISA2_VAES_UNSET \
-   | OPTION_MASK_ISA2_SM3_UNSET)
+   | OPTION_MASK_ISA2_SM3_UNSET | OPTION_MASK_ISA2_SHA512_UNSET)
 #define OPTION_MASK_ISA2_SSE4_2_UNSET OPTION_MASK_ISA2_AVX_UNSET
 #define OPTION_MASK_ISA2_SSE4_1_UNSET OPTION_MASK_ISA2_SSE4_2_UNSET
 #define OPTION_MASK_ISA2_SSE4_UNSET OPTION_MASK_ISA2_SSE4_1_UNSET
@@ -1306,6 +1308,21 @@ ix86_handle_option (struct gcc_options *opts,
 	}
       return true;
 
+    case OPT_msha512:
+      if (value)
+	{
+	  opts->x_ix86_isa_flags2 |= OPTION_MASK_ISA2_SHA512_SET;
+	  opts->x_ix86_isa_flags2_explicit |= OPTION_MASK_ISA2_SHA512_SET;
+	  opts->x_ix86_isa_flags |= OPTION_MASK_ISA_AVX_SET;
+	  opts->x_ix86_isa_flags_explicit |= OPTION_MASK_ISA_AVX_SET;
+	}
+      else
+	{
+	  opts->x_ix86_isa_flags2 &= ~OPTION_MASK_ISA2_SHA512_UNSET;
+	  opts->x_ix86_isa_flags2_explicit |= OPTION_MASK_ISA2_SHA512_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 c3403090c3b..a6e34d14f8e 100644
--- a/gcc/common/config/i386/i386-cpuinfo.h
+++ b/gcc/common/config/i386/i386-cpuinfo.h
@@ -257,6 +257,7 @@ enum processor_features
   FEATURE_AMX_COMPLEX,
   FEATURE_AVXVNNIINT16,
   FEATURE_SM3,
+  FEATURE_SHA512,
   CPU_FEATURE_MAX
 };
 
diff --git a/gcc/common/config/i386/i386-isas.h b/gcc/common/config/i386/i386-isas.h
index 961a7f0ccd4..250dc87764f 100644
--- a/gcc/common/config/i386/i386-isas.h
+++ b/gcc/common/config/i386/i386-isas.h
@@ -189,4 +189,5 @@ ISA_NAMES_TABLE_START
   ISA_NAMES_TABLE_ENTRY("avxvnniint16", FEATURE_AVXVNNIINT16,
 			P_NONE, "-mavxvnniint16")
   ISA_NAMES_TABLE_ENTRY("sm3", FEATURE_SM3, P_NONE, "-msm3")
+  ISA_NAMES_TABLE_ENTRY("sha512", FEATURE_SHA512, P_NONE, "-msha512")
 ISA_NAMES_TABLE_END
diff --git a/gcc/config.gcc b/gcc/config.gcc
index fbd7360e355..4e753ba7c64 100644
--- a/gcc/config.gcc
+++ b/gcc/config.gcc
@@ -436,7 +436,7 @@ i[34567]86-*-* | x86_64-*-*)
 		       avxifmaintrin.h avxvnniint8intrin.h avxneconvertintrin.h
 		       cmpccxaddintrin.h amxfp16intrin.h prfchiintrin.h
 		       raointintrin.h amxcomplexintrin.h avxvnniint16intrin.h
-		       sm3intrin.h"
+		       sm3intrin.h sha512intrin.h"
 	;;
 ia64-*-*)
 	extra_headers=ia64intrin.h
diff --git a/gcc/config/i386/cpuid.h b/gcc/config/i386/cpuid.h
index 28a36ad0628..f9103f1b1c9 100644
--- a/gcc/config/i386/cpuid.h
+++ b/gcc/config/i386/cpuid.h
@@ -132,6 +132,7 @@
 
 /* Extended Features Sub-leaf (%eax == 7, %ecx == 1) */
 /* %eax */
+#define bit_SHA512	(1 << 0)
 #define bit_SM3		(1 << 1)
 #define bit_RAOINT      (1 << 3)
 #define bit_AVXVNNI     (1 << 4)
diff --git a/gcc/config/i386/i386-builtin-types.def b/gcc/config/i386/i386-builtin-types.def
index 899eac1e014..e9463120eea 100644
--- a/gcc/config/i386/i386-builtin-types.def
+++ b/gcc/config/i386/i386-builtin-types.def
@@ -1419,3 +1419,6 @@ DEF_FUNCTION_TYPE (VOID, PCVOID, INT, INT, INT)
 
 # SM3 builtins
 DEF_FUNCTION_TYPE (V4SI, V4SI, V4SI, V4SI, INT)
+
+# SHA512 builtins
+DEF_FUNCTION_TYPE (V4DI, V4DI, V4DI, V2DI)
diff --git a/gcc/config/i386/i386-builtin.def b/gcc/config/i386/i386-builtin.def
index 17db19c2495..b9e2bad8522 100644
--- a/gcc/config/i386/i386-builtin.def
+++ b/gcc/config/i386/i386-builtin.def
@@ -1660,6 +1660,11 @@ BDESC (OPTION_MASK_ISA_AVX, OPTION_MASK_ISA2_SM3, CODE_FOR_vsm3msg1, "__builtin_
 BDESC (OPTION_MASK_ISA_AVX, OPTION_MASK_ISA2_SM3, CODE_FOR_vsm3msg2, "__builtin_ia32_vsm3msg2", IX86_BUILTIN_VSM3MSG2, UNKNOWN, (int) V4SI_FTYPE_V4SI_V4SI_V4SI)
 BDESC (OPTION_MASK_ISA_AVX, OPTION_MASK_ISA2_SM3, CODE_FOR_vsm3rnds2, "__builtin_ia32_vsm3rnds2", IX86_BUILTIN_VSM3RNDS2, UNKNOWN, (int) V4SI_FTYPE_V4SI_V4SI_V4SI_INT)
 
+/* SHA512 */
+BDESC (0, OPTION_MASK_ISA2_SHA512, CODE_FOR_vsha512msg1, "__builtin_ia32_vsha512msg1", IX86_BUILTIN_VSHA512MSG1, UNKNOWN, (int) V4DI_FTYPE_V4DI_V2DI)
+BDESC (0, OPTION_MASK_ISA2_SHA512, CODE_FOR_vsha512msg2, "__builtin_ia32_vsha512msg2", IX86_BUILTIN_VSHA512MSG2, UNKNOWN, (int) V4DI_FTYPE_V4DI_V4DI)
+BDESC (0, OPTION_MASK_ISA2_SHA512, CODE_FOR_vsha512rnds2, "__builtin_ia32_vsha512rnds2", IX86_BUILTIN_VSHA512RNDS2, UNKNOWN, (int) V4DI_FTYPE_V4DI_V4DI_V2DI)
+
 /* AVX512VL.  */
 BDESC (OPTION_MASK_ISA_AVX512BW | OPTION_MASK_ISA_AVX512VL, 0, CODE_FOR_avx2_palignrv32qi_mask, "__builtin_ia32_palignr256_mask", IX86_BUILTIN_PALIGNR256_MASK, UNKNOWN, (int) V4DI_FTYPE_V4DI_V4DI_INT_V4DI_USI_CONVERT)
 BDESC (OPTION_MASK_ISA_AVX512BW | OPTION_MASK_ISA_AVX512VL, 0, CODE_FOR_ssse3_palignrv16qi_mask, "__builtin_ia32_palignr128_mask", IX86_BUILTIN_PALIGNR128_MASK, UNKNOWN, (int) V2DI_FTYPE_V2DI_V2DI_INT_V2DI_UHI_CONVERT)
diff --git a/gcc/config/i386/i386-c.cc b/gcc/config/i386/i386-c.cc
index 0cb5a6dcce5..c6311f12cf9 100644
--- a/gcc/config/i386/i386-c.cc
+++ b/gcc/config/i386/i386-c.cc
@@ -681,6 +681,8 @@ ix86_target_macros_internal (HOST_WIDE_INT isa_flag,
     def_or_undef (parse_in, "__AVXVNNIINT16__");
   if (isa_flag2 & OPTION_MASK_ISA2_SM3)
     def_or_undef (parse_in, "__SM3__");
+  if (isa_flag2 & OPTION_MASK_ISA2_SHA512)
+    def_or_undef (parse_in, "__SHA512__");
   if (TARGET_IAMCU)
     {
       def_or_undef (parse_in, "__iamcu");
diff --git a/gcc/config/i386/i386-expand.cc b/gcc/config/i386/i386-expand.cc
index f6ad54c0cfe..e99dcf19364 100644
--- a/gcc/config/i386/i386-expand.cc
+++ b/gcc/config/i386/i386-expand.cc
@@ -10733,6 +10733,7 @@ ix86_expand_args_builtin (const struct builtin_description *d,
     case V4SF_FTYPE_V4SF_UINT:
     case V4SF_FTYPE_V4SF_DI:
     case V4SF_FTYPE_V4SF_SI:
+    case V4DI_FTYPE_V4DI_V2DI:
     case V2DI_FTYPE_V2DI_V2DI:
     case V2DI_FTYPE_V16QI_V16QI:
     case V2DI_FTYPE_V4SI_V4SI:
@@ -11030,6 +11031,7 @@ ix86_expand_args_builtin (const struct builtin_description *d,
     case V8HI_FTYPE_V8DI_V8HI_UQI:
     case V8SI_FTYPE_V8DI_V8SI_UQI:
     case V4SI_FTYPE_V4SI_V4SI_V4SI:
+    case V4DI_FTYPE_V4DI_V4DI_V2DI:
     case V16SI_FTYPE_V16SI_V16SI_V16SI:
     case V8DI_FTYPE_V8DI_V8DI_V8DI:
     case V32HI_FTYPE_V32HI_V32HI_V32HI:
diff --git a/gcc/config/i386/i386-isa.def b/gcc/config/i386/i386-isa.def
index 432c36e7f79..28f221753a9 100644
--- a/gcc/config/i386/i386-isa.def
+++ b/gcc/config/i386/i386-isa.def
@@ -119,3 +119,4 @@ DEF_PTA(RAOINT)
 DEF_PTA(AMX_COMPLEX)
 DEF_PTA(AVXVNNIINT16)
 DEF_PTA(SM3)
+DEF_PTA(SHA512)
diff --git a/gcc/config/i386/i386-options.cc b/gcc/config/i386/i386-options.cc
index db2ff0c7ae1..d79ab01bd79 100644
--- a/gcc/config/i386/i386-options.cc
+++ b/gcc/config/i386/i386-options.cc
@@ -241,7 +241,8 @@ static struct ix86_target_opts isa2_opts[] =
   { "-mraoint", 	OPTION_MASK_ISA2_RAOINT },
   { "-mamx-complex",	OPTION_MASK_ISA2_AMX_COMPLEX },
   { "-mavxvnniint16",	OPTION_MASK_ISA2_AVXVNNIINT16 },
-  { "-msm3",		OPTION_MASK_ISA2_SM3 }
+  { "-msm3",		OPTION_MASK_ISA2_SM3 },
+  { "-msha512",		OPTION_MASK_ISA2_SHA512 }
 };
 static struct ix86_target_opts isa_opts[] =
 {
@@ -1095,6 +1096,7 @@ ix86_valid_target_attribute_inner_p (tree fndecl, tree args, char *p_strings[],
     IX86_ATTR_ISA ("amx-complex", OPT_mamx_complex),
     IX86_ATTR_ISA ("avxvnniint16", OPT_mavxvnniint16),
     IX86_ATTR_ISA ("sm3", OPT_msm3),
+    IX86_ATTR_ISA ("sha512", OPT_msha512),
 
     /* enum options */
     IX86_ATTR_ENUM ("fpmath=",	OPT_mfpmath_),
diff --git a/gcc/config/i386/i386.opt b/gcc/config/i386/i386.opt
index 80a8611993c..cf9dbca58b3 100644
--- a/gcc/config/i386/i386.opt
+++ b/gcc/config/i386/i386.opt
@@ -1288,3 +1288,13 @@ msm3
 Target Mask(ISA2_SM3) Var(ix86_isa_flags2) Save
 Support MMX, SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2, AVX and
 SM3 built-in functions and code generation.
+
+mvpinsrvpextr
+Target Mask(ISA2_VPINSRVPEXTR) Var(ix86_isa_flags2) Save
+Support MMX, SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2, AVX, AVX2, AVX512F,
+AVX512VL and VPINSRVPEXTR built-in functions and code generation.
+
+msha512
+Target Mask(ISA2_SHA512) Var(ix86_isa_flags2) Save
+Support MMX, SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2, AVX and
+SHA512 built-in functions and code generation.
diff --git a/gcc/config/i386/immintrin.h b/gcc/config/i386/immintrin.h
index 7731990131c..6f2bcef6a8a 100644
--- a/gcc/config/i386/immintrin.h
+++ b/gcc/config/i386/immintrin.h
@@ -110,6 +110,8 @@
 
 #include <sm3intrin.h>
 
+#include <sha512intrin.h>
+
 #include <fmaintrin.h>
 
 #include <f16cintrin.h>
diff --git a/gcc/config/i386/sha512intrin.h b/gcc/config/i386/sha512intrin.h
new file mode 100644
index 00000000000..884c2bc6340
--- /dev/null
+++ b/gcc/config/i386/sha512intrin.h
@@ -0,0 +1,64 @@
+/* Copyright (C) 2023 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/>.  */
+
+#ifndef _IMMINTRIN_H_INCLUDED
+#error "Never use <sha512intrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef _SHA512INTRIN_H_INCLUDED
+#define _SHA512INTRIN_H_INCLUDED
+
+#ifndef __SHA512__
+#pragma GCC push_options
+#pragma GCC target("sha512")
+#define __DISABLE_SHA512__
+#endif /* __SHA512__ */
+
+extern __inline __m256i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_sha512msg1_epi64 (__m256i __A, __m128i __B)
+{
+  return (__m256i) __builtin_ia32_vsha512msg1 ((__v4di) __A, (__v2di) __B);
+}
+
+extern __inline __m256i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_sha512msg2_epi64 (__m256i __A, __m256i __B)
+{
+  return (__m256i) __builtin_ia32_vsha512msg2 ((__v4di) __A, (__v4di) __B);
+}
+
+extern __inline __m256i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_sha512rnds2_epi64 (__m256i __A, __m256i __B, __m128i __C)
+{
+  return (__m256i) __builtin_ia32_vsha512rnds2 ((__v4di) __A, (__v4di) __B,
+						(__v2di) __C);
+}
+
+#ifdef __DISABLE_SHA512__
+#undef __DISABLE_SHA512__
+#pragma GCC pop_options
+#endif /* __DISABLE_SHA512__ */
+
+#endif /* _SHA512INTRIN_H_INCLUDED */
diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md
index 25a1e5dd780..e16b2b5a6c4 100644
--- a/gcc/config/i386/sse.md
+++ b/gcc/config/i386/sse.md
@@ -217,6 +217,12 @@
   UNSPEC_VPDPWSUDS
   UNSPEC_VPDPWUUD
   UNSPEC_VPDPWUUDS
+
+  ;; For SHA512 support
+  UNSPEC_SHA512MSG1
+  UNSPEC_SHA512MSG2
+  UNSPEC_SHA512RNDS2
+
 ])
 
 (define_c_enum "unspecv" [
@@ -28640,6 +28646,40 @@
    (set_attr "mode" "TI")
    (set_attr "length_immediate" "1")])
 
+(define_insn "vsha512msg1"
+  [(set (match_operand:V4DI 0 "register_operand" "=x")
+	(unspec:V4DI
+	  [(match_operand:V4DI 1 "register_operand" "0")
+	   (match_operand:V2DI 2 "register_operand" "x")]
+	  UNSPEC_SHA512MSG1))]
+  "TARGET_SHA512"
+  "vsha512msg1\t{%2, %0|%0, %2}"
+  [(set_attr "type" "sselog1")
+   (set_attr "mode" "OI")])
+
+(define_insn "vsha512msg2"
+  [(set (match_operand:V4DI 0 "register_operand" "=x")
+	(unspec:V4DI
+	  [(match_operand:V4DI 1 "register_operand" "0")
+	   (match_operand:V4DI 2 "register_operand" "x")]
+	  UNSPEC_SHA512MSG2))]
+  "TARGET_SHA512"
+  "vsha512msg2\t{%2, %0|%0, %2}"
+  [(set_attr "type" "sselog1")
+   (set_attr "mode" "OI")])
+
+(define_insn "vsha512rnds2"
+  [(set (match_operand:V4DI 0 "register_operand" "=x")
+	(unspec:V4DI
+	  [(match_operand:V4DI 1 "register_operand" "0")
+	   (match_operand:V4DI 2 "register_operand" "x")
+	   (match_operand:V2DI 3 "register_operand" "x")]
+	  UNSPEC_SHA512RNDS2))]
+  "TARGET_SHA512"
+  "vsha512rnds2\t{%3, %2, %0|%0, %2, %3}"
+  [(set_attr "type" "sselog1")
+   (set_attr "mode" "OI")])
+
 (define_insn_and_split "avx512f_<castmode><avxsizesuffix>_<castmode>"
   [(set (match_operand:AVX512MODE2P 0 "nonimmediate_operand" "=x,m")
 	(vec_concat:AVX512MODE2P
diff --git a/gcc/doc/extend.texi b/gcc/doc/extend.texi
index e76cd399a83..5250990050b 100644
--- a/gcc/doc/extend.texi
+++ b/gcc/doc/extend.texi
@@ -7173,6 +7173,11 @@ Enable/disable the generation of the AVXVNNIINT16 instructions.
 @itemx no-sm3
 Enable/disable the generation of the SM3 instructions.
 
+@cindex @code{target("sha512")} function attribute, x86
+@item sha512
+@itemx no-sha512
+Enable/disable the generation of the SHA512 instructions.
+
 @cindex @code{target("cld")} function attribute, x86
 @item cld
 @itemx no-cld
diff --git a/gcc/doc/invoke.texi b/gcc/doc/invoke.texi
index 2671d70736f..433ccf35505 100644
--- a/gcc/doc/invoke.texi
+++ b/gcc/doc/invoke.texi
@@ -1433,7 +1433,7 @@ See RS/6000 and PowerPC Options.
 -mrdseed  -msgx -mavx512vp2intersect -mserialize -mtsxldtrk
 -mamx-tile  -mamx-int8  -mamx-bf16 -muintr -mhreset -mavxvnni
 -mavx512fp16 -mavxifma -mavxvnniint8 -mavxneconvert -mcmpccxadd -mamx-fp16
--mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3
+-mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3 -msha512
 -mcldemote  -mms-bitfields  -mno-align-stringops  -minline-all-stringops
 -minline-stringops-dynamically  -mstringop-strategy=@var{alg}
 -mkl -mwidekl
@@ -33558,6 +33558,9 @@ preferred alignment to @option{-mpreferred-stack-boundary=2}.
 @need 200
 @opindex msm3
 @itemx -msm3
+@need 200
+@opindex msha512
+@itemx -msha512
 These switches enable the use of instructions in the MMX, SSE,
 AVX512ER, AVX512CD, AVX512VL, AVX512BW, AVX512DQ, AVX512IFMA, AVX512VBMI, SHA,
 AES, PCLMUL, CLFLUSHOPT, CLWB, FSGSBASE, PTWRITE, RDRND, F16C, FMA, PCONFIG,
@@ -33568,8 +33571,9 @@ GFNI, VAES, WAITPKG, VPCLMULQDQ, AVX512BITALG, MOVDIRI, MOVDIR64B, AVX512BF16,
 ENQCMD, AVX512VPOPCNTDQ, AVX5124FMAPS, AVX512VNNI, AVX5124VNNIW, SERIALIZE,
 UINTR, HRESET, AMXTILE, AMXINT8, AMXBF16, KL, WIDEKL, AVXVNNI, AVX512-FP16,
 AVXIFMA, AVXVNNIINT8, AVXNECONVERT, CMPCCXADD, AMX-FP16, PREFETCHI, RAOINT,
-AMX-COMPLEX, AVXVNNIINT16, SM3 or CLDEMOTE extended instruction sets. Each has a
-corresponding @option{-mno-} option to disable use of these instructions.
+AMX-COMPLEX, AVXVNNIINT16, SM3, SHA512 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 dae51132c42..54a062db3fe 100644
--- a/gcc/doc/sourcebuild.texi
+++ b/gcc/doc/sourcebuild.texi
@@ -2589,6 +2589,9 @@ Target supports the execution of @code{raoint} instructions.
 @item rdrand
 Target supports x86 @code{rdrand} instruction.
 
+@item sha512
+Target supports the execution of @code{sha512} instructions.
+
 @item sm3
 Target supports the execution of @code{sm3} instructions.
 
diff --git a/gcc/testsuite/g++.dg/other/i386-2.C b/gcc/testsuite/g++.dg/other/i386-2.C
index 2ec93261cac..985f14abcbc 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 -mavxvnniint8 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3" } */
+/* { 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 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3 -msha512" } */
 
 /* 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 fe03143e39b..274b0e6256f 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 -mavxvnniint8 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3" } */
+/* { 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 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3 -msha512" } */
 
 /* 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/funcspec-56.inc b/gcc/testsuite/gcc.target/i386/funcspec-56.inc
index 8dd8d9bf9d8..eb9309819ab 100644
--- a/gcc/testsuite/gcc.target/i386/funcspec-56.inc
+++ b/gcc/testsuite/gcc.target/i386/funcspec-56.inc
@@ -90,6 +90,7 @@ extern void test_raoint (void)                  __attribute__((__target__("raoin
 extern void test_amx_complex (void)		__attribute__((__target__("amx-complex")));
 extern void test_avxvnniint16 (void)		__attribute__((__target__("avxvnniint16")));
 extern void test_sm3 (void)			__attribute__((__target__("sm3")));
+extern void test_sha512 (void)			__attribute__((__target__("sha512")));
 
 extern void test_no_sgx (void)			__attribute__((__target__("no-sgx")));
 extern void test_no_avx5124fmaps(void)		__attribute__((__target__("no-avx5124fmaps")));
@@ -181,6 +182,7 @@ extern void test_no_raoint (void)               __attribute__((__target__("no-ra
 extern void test_no_amx_complex (void)		__attribute__((__target__("no-amx-complex")));
 extern void test_no_avxvnniint16 (void)		__attribute__((__target__("no-avxvnniint16")));
 extern void test_no_sm3 (void)			__attribute__((__target__("no-sm3")));
+extern void test_no_sha512 (void)		__attribute__((__target__("no-sha512")));
 
 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/sha512-1.c b/gcc/testsuite/gcc.target/i386/sha512-1.c
new file mode 100644
index 00000000000..c66e8124d6b
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sha512-1.c
@@ -0,0 +1,18 @@
+/* { dg-do compile } */
+/* { dg-options "-O2 -msha512" } */
+/* { dg-final { scan-assembler "vsha512msg1\[ \\t\]+\[^\n\]*%xmm\[0-9\]+\[^\n\]*%ymm\[0-9\]" } } */
+/* { dg-final { scan-assembler "vsha512msg2\[ \\t\]+\[^\n\]*%ymm\[0-9\]+\[^\n\]*%ymm\[0-9\]" } } */
+/* { dg-final { scan-assembler "vsha512rnds2\[ \\t\]+\[^\n\]*%xmm\[0-9\]+\[^\n\]*%ymm\[0-9\]+\[^\n\]*%ymm\[0-9\]" } } */
+
+#include <immintrin.h>
+
+volatile __m128i x;
+volatile __m256i y;
+
+void extern
+sha512_test (void)
+{
+  y = _mm256_sha512msg1_epi64(y, x);
+  y = _mm256_sha512msg2_epi64(y, y);
+  y = _mm256_sha512rnds2_epi64(y, y, x);
+}
diff --git a/gcc/testsuite/gcc.target/i386/sha512-check.h b/gcc/testsuite/gcc.target/i386/sha512-check.h
new file mode 100644
index 00000000000..083bf3b6a1b
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sha512-check.h
@@ -0,0 +1,43 @@
+#include <stdlib.h>
+#include "m256-check.h"
+
+static void sha512_test (void);
+
+static unsigned long long
+ror64 (unsigned long long w, int n)
+{
+  int count = n % 64;
+  return ((w >> n) | (w << (64 - n)));
+}
+
+static unsigned long long
+shr64 (unsigned long long w, int n)
+{
+  return (w >> n);
+}
+
+static void
+__attribute__ ((noinline))
+do_test(void)
+{
+  sha512_test ();
+}
+
+int
+main ()
+{
+  /* Check CPU support for SHA512.  */
+  if (__builtin_cpu_supports ("sha512"))
+    {
+      do_test ();
+#ifdef DEBUG
+      printf ("PASSED\n");
+#endif
+      return 0;
+    }
+
+#ifdef DEBUG
+  printf ("SKIPPED\n");
+#endif
+  return 0;
+}
diff --git a/gcc/testsuite/gcc.target/i386/sha512msg1-2.c b/gcc/testsuite/gcc.target/i386/sha512msg1-2.c
new file mode 100644
index 00000000000..b7baff1cfb4
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sha512msg1-2.c
@@ -0,0 +1,48 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -msha512" } */
+/* { dg-require-effective-target sha512 } */
+
+#include "sha512-check.h"
+#include <emmintrin.h>
+#include <immintrin.h>
+
+static unsigned long long
+s0 (unsigned long long w)
+{
+  return ror64 (w, 1) ^ ror64 (w, 8) ^ shr64 (w, 7);
+}
+
+static void
+compute_sha512msg1(long long* src1, long long* src2, long long* res)
+{
+  unsigned long long w0, w1, w2, w3, w4;
+
+  w0 = src1[0];
+  w1 = src1[1];
+  w2 = src1[2];
+  w3 = src1[3];
+  w4 = src2[0];
+
+  res[0] = w0 + s0 (w1);
+  res[1] = w1 + s0 (w2);
+  res[2] = w2 + s0 (w3);
+  res[3] = w3 + s0 (w4);
+}
+
+static void
+sha512_test(void)
+{
+  union256i_q s1, res;
+  union128i_q s2;
+  long long res_ref[4];
+
+  s1.x = _mm256_set_epi64x (111, 222, 333, 444);
+  s2.x = _mm_set_epi64x (0, 555);
+
+  res.x = _mm256_sha512msg1_epi64 (s1.x, s2.x);
+
+  compute_sha512msg1 (s1.a, s2.a, res_ref);
+
+  if (check_union256i_q (res, res_ref))
+    abort();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sha512msg2-2.c b/gcc/testsuite/gcc.target/i386/sha512msg2-2.c
new file mode 100644
index 00000000000..e5033903f06
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sha512msg2-2.c
@@ -0,0 +1,47 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -msha512" } */
+/* { dg-require-effective-target sha512 } */
+
+#include "sha512-check.h"
+#include <immintrin.h>
+
+static unsigned long long
+s1 (unsigned long long w)
+{
+  return ror64 (w, 19) ^ ror64 (w, 61) ^ shr64 (w, 6);
+}
+
+static void
+compute_sha512msg2 (long long* src1, long long* src2, long long* res)
+{
+  unsigned long long w14, w15, w16, w17, w18, w19;
+
+  w14 = src2[2];
+  w15 = src2[3];
+  w16 = src1[0] + s1 (w14);
+  w17 = src1[1] + s1 (w15);
+  w18 = src1[2] + s1 (w16);
+  w19 = src1[3] + s1 (w17);
+
+  res[0] = w16;
+  res[1] = w17;
+  res[2] = w18;
+  res[3] = w19;
+}
+
+static void
+sha512_test (void)
+{
+  union256i_q s1, s2, res;
+  long long res_ref[4];
+
+  s1.x = _mm256_set_epi64x (111, 222, 333, 444);
+  s2.x = _mm256_set_epi64x (555, 666, 0, 0);
+
+  res.x = _mm256_sha512msg2_epi64 (s1.x, s2.x);
+
+  compute_sha512msg2 (s1.a, s2.a, res_ref);
+
+  if (check_union256i_q (res, res_ref))
+    abort();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sha512rnds2-2.c b/gcc/testsuite/gcc.target/i386/sha512rnds2-2.c
new file mode 100644
index 00000000000..3bc6a00a05b
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sha512rnds2-2.c
@@ -0,0 +1,85 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -msha512" } */
+/* { dg-require-effective-target sha512 } */
+
+#include "sha512-check.h"
+#include <emmintrin.h>
+#include <immintrin.h>
+
+static unsigned long long
+ch (unsigned long long e, unsigned long long f, unsigned long long g)
+{
+  return (e & f) ^ (~e & g);
+}
+
+static unsigned long long
+maj (unsigned long long a, unsigned long long b, unsigned long long c)
+{
+  return (a & b) ^ (a & c) ^ (b & c);
+}
+
+static unsigned long long
+s0 (unsigned long long w)
+{
+  return ror64 (w, 28) ^ ror64 (w, 34) ^ ror64 (w, 39);
+}
+
+static unsigned long long
+s1 (unsigned long long w)
+{
+  return ror64 (w, 14) ^ ror64 (w, 18) ^ ror64 (w, 41);
+}
+
+static void
+compute_sha512rnds2(long long* src0, long long* src1, long long* src2, long long* res)
+{
+  unsigned long long wk[2] = { src2[0], src2[1] };
+  unsigned long long a[3], b[3], c[3], d[3], e[3], f[3], g[3], h[3];
+
+  a[0] = src1[3];
+  b[0] = src1[2];
+  c[0] = src0[3];
+  d[0] = src0[2];
+  e[0] = src1[1];
+  f[0] = src1[0];
+  g[0] = src0[1];
+  h[0] = src0[0];
+
+  int i;
+  for (i = 0; i <= 1; i++)
+  {
+    a[i + 1] = ch (e[i], f[i], g[i]) + s1 (e[i]) + wk[i] + h[i]
+      + maj (a[i], b[i], c[i]) + s0 (a[i]);
+    b[i + 1] = a[i];
+    c[i + 1] = b[i];
+    d[i + 1] = c[i];
+    e[i + 1] = ch (e[i], f[i], g[i]) + s1 (e[i]) + wk[i] + h[i] + d[i];
+    f[i + 1] = e[i];
+    g[i + 1] = f[i];
+    h[i + 1] = g[i];
+  }
+
+  res[0] = f[2];
+  res[1] = e[2];
+  res[2] = b[2];
+  res[3] = a[2];
+}
+
+static void
+sha512_test (void)
+{
+  union256i_q s0, s1, res;
+  union128i_q s2;
+  long long res_ref[4];
+
+  s0.x = _mm256_set_epi64x (111, 222, 333, 444);
+  s1.x = _mm256_set_epi64x (555, 666, 777, 888);
+  s2.x = _mm_set_epi64x (999, 123);
+
+  res.x = _mm256_sha512rnds2_epi64 (s0.x, s1.x, s2.x);
+
+  compute_sha512rnds2 (s0.a, s1.a, s2.a, res_ref);
+
+  if (check_union256i_q (res, res_ref))
+    abort();
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse-12.c b/gcc/testsuite/gcc.target/i386/sse-12.c
index 5058be6f6e9..976541389ea 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 -mavxvnniint8 -mavxneconvert -mamx-fp16 -mraoint -mamx-complex -mavxvnniint16 -msm3" } */
+/* { 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 -mavxneconvert -mamx-fp16 -mraoint -mamx-complex -mavxvnniint16 -msm3 -msha512" } */
 
 #include <x86intrin.h>
 
diff --git a/gcc/testsuite/gcc.target/i386/sse-13.c b/gcc/testsuite/gcc.target/i386/sse-13.c
index d30b365564a..8c314e70e31 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 -mavxvnniint8 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3" } */
+/* { 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 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3 -msha512" } */
 /* { 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 7842005a98b..2b4d7bc9079 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 -mavxvnniint8 -mavxneconvert -mamx-fp16 -mraoint -mamx-complex -mavxvnniint16 -msm3" } */
+/* { 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 -mavxneconvert -mamx-fp16 -mraoint -mamx-complex -mavxvnniint16 -msm3 -msha512" } */
 /* { 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 7537db1ac30..d6f19b5e20a 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,tbm,lwp,fsgsbase,rdrnd,f16c,rtm,rdseed,prfchw,adx,fxsr,xsaveopt,avx512f,avx512er,avx512cd,avx512pf,sha,prefetchwt1,avx512vl,avx512bw,avx512dq,avx512vbmi,avx512vbmi2,avx512ifma,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,gfni,avx512bitalg,avx512bf16,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,amx-fp16,raoint,amx-complex,avxvnniint16,sm3")
+#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,avx512f,avx512er,avx512cd,avx512pf,sha,prefetchwt1,avx512vl,avx512bw,avx512dq,avx512vbmi,avx512vbmi2,avx512ifma,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,gfni,avx512bitalg,avx512bf16,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,amx-fp16,raoint,amx-complex,avxvnniint16,sm3,sha512")
 #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,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,gfni,avx512bitalg,avx512bf16,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,amx-fp16,raoint,amx-complex,avxvnniint16,sm3")
+#pragma GCC target ("avx,avx2,rdrnd,fsgsbase,f16c,rtm,avx512f,avx512er,avx512cd,avx512pf,sha,avx512vl,avx512bw,avx512dq,avx512ifma,avx512vbmi,avx512vbmi2,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,gfni,avx512bitalg,avx512bf16,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,amx-fp16,raoint,amx-complex,avxvnniint16,sm3,sha512")
 #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 3fc61b50fe6..1df66b525ca 100644
--- a/gcc/testsuite/gcc.target/i386/sse-23.c
+++ b/gcc/testsuite/gcc.target/i386/sse-23.c
@@ -850,6 +850,6 @@
 /* sm3intrin.h */
 #define __builtin_ia32_vsm3rnds2(A, B, C, D) __builtin_ia32_vsm3rnds2 (A, B, C, 1)
 
-#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,avx512f,avx512er,avx512cd,avx512pf,sha,prefetchwt1,xsavec,xsaves,clflushopt,avx512bw,avx512dq,avx512vl,avx512vbmi,avx512ifma,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,clwb,mwaitx,clzero,pku,sgx,rdpid,gfni,avx512vbmi2,vpclmulqdq,avx512bitalg,pconfig,wbnoinvd,avx512bf16,enqcmd,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,cmpccxadd,amx-fp16,prefetchi,raoint,amx-complex,avxvnniint16,sm3")
+#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,avx512f,avx512er,avx512cd,avx512pf,sha,prefetchwt1,xsavec,xsaves,clflushopt,avx512bw,avx512dq,avx512vl,avx512vbmi,avx512ifma,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,clwb,mwaitx,clzero,pku,sgx,rdpid,gfni,avx512vbmi2,vpclmulqdq,avx512bitalg,pconfig,wbnoinvd,avx512bf16,enqcmd,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,cmpccxadd,amx-fp16,prefetchi,raoint,amx-complex,avxvnniint16,sm3,sha512")
 
 #include <x86intrin.h>
diff --git a/gcc/testsuite/lib/target-supports.exp b/gcc/testsuite/lib/target-supports.exp
index c911a824d31..f376d835f8b 100644
--- a/gcc/testsuite/lib/target-supports.exp
+++ b/gcc/testsuite/lib/target-supports.exp
@@ -9883,6 +9883,20 @@ proc check_effective_target_sm3 { } {
     } "-msm3" ]
 }
 
+# Return 1 if sha512 instructions can be compiled.
+proc check_effective_target_sha512 { } {
+    return [check_no_compiler_messages sha512 object {
+	typedef long long __m256i __attribute__ ((__vector_size__ (32)));
+	typedef long long __v4di __attribute__ ((__vector_size__ (32)));
+	__m256i
+	_mm256_sha512msg2_epi64 (__m256i __A, __m256i __B)
+        {
+	   return (__m256i) __builtin_ia32_vsha512msg2 ((__v4di) __A,
+							(__v4di) __B);
+	}
+    } "-msha512" ]
+}
+
 # Return 1 if sse instructions can be compiled.
 proc check_effective_target_sse { } {
     return [check_no_compiler_messages sse object {
-- 
2.31.1


^ permalink raw reply	[flat|nested] 9+ messages in thread

* Re: [PATCH 3/4] Support Intel SHA512
  2023-07-13  6:03 ` [PATCH 3/4] Support Intel SHA512 Haochen Jiang
@ 2023-07-17  1:40   ` Hongtao Liu
  0 siblings, 0 replies; 9+ messages in thread
From: Hongtao Liu @ 2023-07-17  1:40 UTC (permalink / raw)
  To: Haochen Jiang; +Cc: gcc-patches, hongtao.liu, ubizjak

On Thu, Jul 13, 2023 at 2:06 PM Haochen Jiang via Gcc-patches
<gcc-patches@gcc.gnu.org> wrote:
>
> gcc/ChangeLog:
>
>         * common/config/i386/cpuinfo.h (get_available_features):
>         Detect SHA512.
>         * common/config/i386/i386-common.cc (OPTION_MASK_ISA2_SHA512_SET,
>         OPTION_MASK_ISA2_SHA512_UNSET): New.
>         (OPTION_MASK_ISA2_AVX_UNSET): Add SHA512.
>         (ix86_handle_option): Handle -msha512.
>         * common/config/i386/i386-cpuinfo.h (enum processor_features):
>         Add FEATURE_SHA512.
>         * common/config/i386/i386-isas.h: Add ISA_NAME_TABLE_ENTRY for
>         sha512.
>         * config.gcc: Add sha512intrin.h.
>         * config/i386/cpuid.h (bit_SHA512): New.
>         * config/i386/i386-builtin-types.def:
>         Add DEF_FUNCTION_TYPE (V4DI, V4DI, V4DI, V2DI).
>         * config/i386/i386-builtin.def (BDESC): Add new builtins.
>         * config/i386/i386-c.cc (ix86_target_macros_internal): Define
>         __SHA512__.
>         * config/i386/i386-expand.cc (ix86_expand_args_builtin): Handle
>         V4DI_FTYPE_V4DI_V4DI_V2DI and V4DI_FTYPE_V4DI_V2DI.
>         * config/i386/i386-isa.def (SHA512): Add DEF_PTA(SHA512).
>         * config/i386/i386-options.cc (isa2_opts): Add -msha512.
>         (ix86_valid_target_attribute_inner_p): Handle sha512.
>         * config/i386/i386.opt: Add option -msha512.
>         * config/i386/immintrin.h: Include sha512intrin.h.
>         * config/i386/sse.md (vsha512msg1): New define insn.
>         (vsha512msg2): Ditto.
>         (vsha512rnds2): Ditto.
>         * doc/extend.texi: Document sha512.
>         * doc/invoke.texi: Document -msha512.
>         * doc/sourcebuild.texi: Document target sha512.
>         * config/i386/sha512intrin.h: New file.
>
> gcc/testsuite/ChangeLog:
>
>         * g++.dg/others/i386-2.C: Add -msha512.
>         * g++.dg/others/i386-3.C: Ditto.
>         * gcc.target/i386/funcspec-56.inc: Add new target attribute.
>         * gcc.target/i386/sse-12.c: Add -msha512.
>         * gcc.target/i386/sse-13.c: Ditto.
>         * gcc.target/i386/sse-14.c: Ditto.
>         * gcc.target/i386/sse-22.c: Add sha512.
>         * gcc.target/i386/sse-23.c: Ditto.
>         * lib/target-supports.exp (check_effective_target_sha512): New.
>         * gcc.target/i386/sha512-1.c: New test.
>         * gcc.target/i386/sha512-check.h: Ditto.
>         * gcc.target/i386/sha512msg1-2.c: Ditto.
>         * gcc.target/i386/sha512msg2-2.c: Ditto.
>         * gcc.target/i386/sha512rnds2-2.c: Ditto.
Ok.
> ---
>  gcc/common/config/i386/cpuinfo.h              |  2 +
>  gcc/common/config/i386/i386-common.cc         | 19 ++++-
>  gcc/common/config/i386/i386-cpuinfo.h         |  1 +
>  gcc/common/config/i386/i386-isas.h            |  1 +
>  gcc/config.gcc                                |  2 +-
>  gcc/config/i386/cpuid.h                       |  1 +
>  gcc/config/i386/i386-builtin-types.def        |  3 +
>  gcc/config/i386/i386-builtin.def              |  5 ++
>  gcc/config/i386/i386-c.cc                     |  2 +
>  gcc/config/i386/i386-expand.cc                |  2 +
>  gcc/config/i386/i386-isa.def                  |  1 +
>  gcc/config/i386/i386-options.cc               |  4 +-
>  gcc/config/i386/i386.opt                      | 10 +++
>  gcc/config/i386/immintrin.h                   |  2 +
>  gcc/config/i386/sha512intrin.h                | 64 ++++++++++++++
>  gcc/config/i386/sse.md                        | 40 +++++++++
>  gcc/doc/extend.texi                           |  5 ++
>  gcc/doc/invoke.texi                           | 10 ++-
>  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/funcspec-56.inc |  2 +
>  gcc/testsuite/gcc.target/i386/sha512-1.c      | 18 ++++
>  gcc/testsuite/gcc.target/i386/sha512-check.h  | 43 ++++++++++
>  gcc/testsuite/gcc.target/i386/sha512msg1-2.c  | 48 +++++++++++
>  gcc/testsuite/gcc.target/i386/sha512msg2-2.c  | 47 ++++++++++
>  gcc/testsuite/gcc.target/i386/sha512rnds2-2.c | 85 +++++++++++++++++++
>  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         | 14 +++
>  33 files changed, 436 insertions(+), 14 deletions(-)
>  create mode 100644 gcc/config/i386/sha512intrin.h
>  create mode 100644 gcc/testsuite/gcc.target/i386/sha512-1.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sha512-check.h
>  create mode 100644 gcc/testsuite/gcc.target/i386/sha512msg1-2.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sha512msg2-2.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sha512rnds2-2.c
>
> diff --git a/gcc/common/config/i386/cpuinfo.h b/gcc/common/config/i386/cpuinfo.h
> index e5cdffe017a..0cfde3ebccd 100644
> --- a/gcc/common/config/i386/cpuinfo.h
> +++ b/gcc/common/config/i386/cpuinfo.h
> @@ -879,6 +879,8 @@ get_available_features (struct __processor_model *cpu_model,
>             set_feature (FEATURE_AVXVNNIINT16);
>           if (eax & bit_SM3)
>             set_feature (FEATURE_SM3);
> +         if (eax & bit_SHA512)
> +           set_feature (FEATURE_SHA512);
>         }
>        if (avx512_usable)
>         {
> diff --git a/gcc/common/config/i386/i386-common.cc b/gcc/common/config/i386/i386-common.cc
> index 57b008ca3af..97c3cdfe5e1 100644
> --- a/gcc/common/config/i386/i386-common.cc
> +++ b/gcc/common/config/i386/i386-common.cc
> @@ -121,6 +121,7 @@ along with GCC; see the file COPYING3.  If not see
>    (OPTION_MASK_ISA2_AMX_TILE | OPTION_MASK_ISA2_AMX_COMPLEX)
>  #define OPTION_MASK_ISA2_AVXVNNIINT16_SET OPTION_MASK_ISA2_AVXVNNIINT16
>  #define OPTION_MASK_ISA2_SM3_SET OPTION_MASK_ISA2_SM3
> +#define OPTION_MASK_ISA2_SHA512_SET OPTION_MASK_ISA2_SHA512
>
>  /* SSE4 includes both SSE4.1 and SSE4.2. -msse4 should be the same
>     as -msse4.2.  */
> @@ -305,6 +306,7 @@ along with GCC; see the file COPYING3.  If not see
>  #define OPTION_MASK_ISA2_AMX_COMPLEX_UNSET OPTION_MASK_ISA2_AMX_COMPLEX
>  #define OPTION_MASK_ISA2_AVXVNNIINT16_UNSET OPTION_MASK_ISA2_AVXVNNIINT16
>  #define OPTION_MASK_ISA2_SM3_UNSET OPTION_MASK_ISA2_SM3
> +#define OPTION_MASK_ISA2_SHA512_UNSET OPTION_MASK_ISA2_SHA512
>
>  /* SSE4 includes both SSE4.1 and SSE4.2.  -mno-sse4 should the same
>     as -mno-sse4.1. */
> @@ -354,7 +356,7 @@ along with GCC; see the file COPYING3.  If not see
>    OPTION_MASK_ISA2_SSE_UNSET
>  #define OPTION_MASK_ISA2_AVX_UNSET \
>    (OPTION_MASK_ISA2_AVX2_UNSET | OPTION_MASK_ISA2_VAES_UNSET \
> -   | OPTION_MASK_ISA2_SM3_UNSET)
> +   | OPTION_MASK_ISA2_SM3_UNSET | OPTION_MASK_ISA2_SHA512_UNSET)
>  #define OPTION_MASK_ISA2_SSE4_2_UNSET OPTION_MASK_ISA2_AVX_UNSET
>  #define OPTION_MASK_ISA2_SSE4_1_UNSET OPTION_MASK_ISA2_SSE4_2_UNSET
>  #define OPTION_MASK_ISA2_SSE4_UNSET OPTION_MASK_ISA2_SSE4_1_UNSET
> @@ -1306,6 +1308,21 @@ ix86_handle_option (struct gcc_options *opts,
>         }
>        return true;
>
> +    case OPT_msha512:
> +      if (value)
> +       {
> +         opts->x_ix86_isa_flags2 |= OPTION_MASK_ISA2_SHA512_SET;
> +         opts->x_ix86_isa_flags2_explicit |= OPTION_MASK_ISA2_SHA512_SET;
> +         opts->x_ix86_isa_flags |= OPTION_MASK_ISA_AVX_SET;
> +         opts->x_ix86_isa_flags_explicit |= OPTION_MASK_ISA_AVX_SET;
> +       }
> +      else
> +       {
> +         opts->x_ix86_isa_flags2 &= ~OPTION_MASK_ISA2_SHA512_UNSET;
> +         opts->x_ix86_isa_flags2_explicit |= OPTION_MASK_ISA2_SHA512_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 c3403090c3b..a6e34d14f8e 100644
> --- a/gcc/common/config/i386/i386-cpuinfo.h
> +++ b/gcc/common/config/i386/i386-cpuinfo.h
> @@ -257,6 +257,7 @@ enum processor_features
>    FEATURE_AMX_COMPLEX,
>    FEATURE_AVXVNNIINT16,
>    FEATURE_SM3,
> +  FEATURE_SHA512,
>    CPU_FEATURE_MAX
>  };
>
> diff --git a/gcc/common/config/i386/i386-isas.h b/gcc/common/config/i386/i386-isas.h
> index 961a7f0ccd4..250dc87764f 100644
> --- a/gcc/common/config/i386/i386-isas.h
> +++ b/gcc/common/config/i386/i386-isas.h
> @@ -189,4 +189,5 @@ ISA_NAMES_TABLE_START
>    ISA_NAMES_TABLE_ENTRY("avxvnniint16", FEATURE_AVXVNNIINT16,
>                         P_NONE, "-mavxvnniint16")
>    ISA_NAMES_TABLE_ENTRY("sm3", FEATURE_SM3, P_NONE, "-msm3")
> +  ISA_NAMES_TABLE_ENTRY("sha512", FEATURE_SHA512, P_NONE, "-msha512")
>  ISA_NAMES_TABLE_END
> diff --git a/gcc/config.gcc b/gcc/config.gcc
> index fbd7360e355..4e753ba7c64 100644
> --- a/gcc/config.gcc
> +++ b/gcc/config.gcc
> @@ -436,7 +436,7 @@ i[34567]86-*-* | x86_64-*-*)
>                        avxifmaintrin.h avxvnniint8intrin.h avxneconvertintrin.h
>                        cmpccxaddintrin.h amxfp16intrin.h prfchiintrin.h
>                        raointintrin.h amxcomplexintrin.h avxvnniint16intrin.h
> -                      sm3intrin.h"
> +                      sm3intrin.h sha512intrin.h"
>         ;;
>  ia64-*-*)
>         extra_headers=ia64intrin.h
> diff --git a/gcc/config/i386/cpuid.h b/gcc/config/i386/cpuid.h
> index 28a36ad0628..f9103f1b1c9 100644
> --- a/gcc/config/i386/cpuid.h
> +++ b/gcc/config/i386/cpuid.h
> @@ -132,6 +132,7 @@
>
>  /* Extended Features Sub-leaf (%eax == 7, %ecx == 1) */
>  /* %eax */
> +#define bit_SHA512     (1 << 0)
>  #define bit_SM3                (1 << 1)
>  #define bit_RAOINT      (1 << 3)
>  #define bit_AVXVNNI     (1 << 4)
> diff --git a/gcc/config/i386/i386-builtin-types.def b/gcc/config/i386/i386-builtin-types.def
> index 899eac1e014..e9463120eea 100644
> --- a/gcc/config/i386/i386-builtin-types.def
> +++ b/gcc/config/i386/i386-builtin-types.def
> @@ -1419,3 +1419,6 @@ DEF_FUNCTION_TYPE (VOID, PCVOID, INT, INT, INT)
>
>  # SM3 builtins
>  DEF_FUNCTION_TYPE (V4SI, V4SI, V4SI, V4SI, INT)
> +
> +# SHA512 builtins
> +DEF_FUNCTION_TYPE (V4DI, V4DI, V4DI, V2DI)
> diff --git a/gcc/config/i386/i386-builtin.def b/gcc/config/i386/i386-builtin.def
> index 17db19c2495..b9e2bad8522 100644
> --- a/gcc/config/i386/i386-builtin.def
> +++ b/gcc/config/i386/i386-builtin.def
> @@ -1660,6 +1660,11 @@ BDESC (OPTION_MASK_ISA_AVX, OPTION_MASK_ISA2_SM3, CODE_FOR_vsm3msg1, "__builtin_
>  BDESC (OPTION_MASK_ISA_AVX, OPTION_MASK_ISA2_SM3, CODE_FOR_vsm3msg2, "__builtin_ia32_vsm3msg2", IX86_BUILTIN_VSM3MSG2, UNKNOWN, (int) V4SI_FTYPE_V4SI_V4SI_V4SI)
>  BDESC (OPTION_MASK_ISA_AVX, OPTION_MASK_ISA2_SM3, CODE_FOR_vsm3rnds2, "__builtin_ia32_vsm3rnds2", IX86_BUILTIN_VSM3RNDS2, UNKNOWN, (int) V4SI_FTYPE_V4SI_V4SI_V4SI_INT)
>
> +/* SHA512 */
> +BDESC (0, OPTION_MASK_ISA2_SHA512, CODE_FOR_vsha512msg1, "__builtin_ia32_vsha512msg1", IX86_BUILTIN_VSHA512MSG1, UNKNOWN, (int) V4DI_FTYPE_V4DI_V2DI)
> +BDESC (0, OPTION_MASK_ISA2_SHA512, CODE_FOR_vsha512msg2, "__builtin_ia32_vsha512msg2", IX86_BUILTIN_VSHA512MSG2, UNKNOWN, (int) V4DI_FTYPE_V4DI_V4DI)
> +BDESC (0, OPTION_MASK_ISA2_SHA512, CODE_FOR_vsha512rnds2, "__builtin_ia32_vsha512rnds2", IX86_BUILTIN_VSHA512RNDS2, UNKNOWN, (int) V4DI_FTYPE_V4DI_V4DI_V2DI)
> +
>  /* AVX512VL.  */
>  BDESC (OPTION_MASK_ISA_AVX512BW | OPTION_MASK_ISA_AVX512VL, 0, CODE_FOR_avx2_palignrv32qi_mask, "__builtin_ia32_palignr256_mask", IX86_BUILTIN_PALIGNR256_MASK, UNKNOWN, (int) V4DI_FTYPE_V4DI_V4DI_INT_V4DI_USI_CONVERT)
>  BDESC (OPTION_MASK_ISA_AVX512BW | OPTION_MASK_ISA_AVX512VL, 0, CODE_FOR_ssse3_palignrv16qi_mask, "__builtin_ia32_palignr128_mask", IX86_BUILTIN_PALIGNR128_MASK, UNKNOWN, (int) V2DI_FTYPE_V2DI_V2DI_INT_V2DI_UHI_CONVERT)
> diff --git a/gcc/config/i386/i386-c.cc b/gcc/config/i386/i386-c.cc
> index 0cb5a6dcce5..c6311f12cf9 100644
> --- a/gcc/config/i386/i386-c.cc
> +++ b/gcc/config/i386/i386-c.cc
> @@ -681,6 +681,8 @@ ix86_target_macros_internal (HOST_WIDE_INT isa_flag,
>      def_or_undef (parse_in, "__AVXVNNIINT16__");
>    if (isa_flag2 & OPTION_MASK_ISA2_SM3)
>      def_or_undef (parse_in, "__SM3__");
> +  if (isa_flag2 & OPTION_MASK_ISA2_SHA512)
> +    def_or_undef (parse_in, "__SHA512__");
>    if (TARGET_IAMCU)
>      {
>        def_or_undef (parse_in, "__iamcu");
> diff --git a/gcc/config/i386/i386-expand.cc b/gcc/config/i386/i386-expand.cc
> index f6ad54c0cfe..e99dcf19364 100644
> --- a/gcc/config/i386/i386-expand.cc
> +++ b/gcc/config/i386/i386-expand.cc
> @@ -10733,6 +10733,7 @@ ix86_expand_args_builtin (const struct builtin_description *d,
>      case V4SF_FTYPE_V4SF_UINT:
>      case V4SF_FTYPE_V4SF_DI:
>      case V4SF_FTYPE_V4SF_SI:
> +    case V4DI_FTYPE_V4DI_V2DI:
>      case V2DI_FTYPE_V2DI_V2DI:
>      case V2DI_FTYPE_V16QI_V16QI:
>      case V2DI_FTYPE_V4SI_V4SI:
> @@ -11030,6 +11031,7 @@ ix86_expand_args_builtin (const struct builtin_description *d,
>      case V8HI_FTYPE_V8DI_V8HI_UQI:
>      case V8SI_FTYPE_V8DI_V8SI_UQI:
>      case V4SI_FTYPE_V4SI_V4SI_V4SI:
> +    case V4DI_FTYPE_V4DI_V4DI_V2DI:
>      case V16SI_FTYPE_V16SI_V16SI_V16SI:
>      case V8DI_FTYPE_V8DI_V8DI_V8DI:
>      case V32HI_FTYPE_V32HI_V32HI_V32HI:
> diff --git a/gcc/config/i386/i386-isa.def b/gcc/config/i386/i386-isa.def
> index 432c36e7f79..28f221753a9 100644
> --- a/gcc/config/i386/i386-isa.def
> +++ b/gcc/config/i386/i386-isa.def
> @@ -119,3 +119,4 @@ DEF_PTA(RAOINT)
>  DEF_PTA(AMX_COMPLEX)
>  DEF_PTA(AVXVNNIINT16)
>  DEF_PTA(SM3)
> +DEF_PTA(SHA512)
> diff --git a/gcc/config/i386/i386-options.cc b/gcc/config/i386/i386-options.cc
> index db2ff0c7ae1..d79ab01bd79 100644
> --- a/gcc/config/i386/i386-options.cc
> +++ b/gcc/config/i386/i386-options.cc
> @@ -241,7 +241,8 @@ static struct ix86_target_opts isa2_opts[] =
>    { "-mraoint",        OPTION_MASK_ISA2_RAOINT },
>    { "-mamx-complex",   OPTION_MASK_ISA2_AMX_COMPLEX },
>    { "-mavxvnniint16",  OPTION_MASK_ISA2_AVXVNNIINT16 },
> -  { "-msm3",           OPTION_MASK_ISA2_SM3 }
> +  { "-msm3",           OPTION_MASK_ISA2_SM3 },
> +  { "-msha512",                OPTION_MASK_ISA2_SHA512 }
>  };
>  static struct ix86_target_opts isa_opts[] =
>  {
> @@ -1095,6 +1096,7 @@ ix86_valid_target_attribute_inner_p (tree fndecl, tree args, char *p_strings[],
>      IX86_ATTR_ISA ("amx-complex", OPT_mamx_complex),
>      IX86_ATTR_ISA ("avxvnniint16", OPT_mavxvnniint16),
>      IX86_ATTR_ISA ("sm3", OPT_msm3),
> +    IX86_ATTR_ISA ("sha512", OPT_msha512),
>
>      /* enum options */
>      IX86_ATTR_ENUM ("fpmath=", OPT_mfpmath_),
> diff --git a/gcc/config/i386/i386.opt b/gcc/config/i386/i386.opt
> index 80a8611993c..cf9dbca58b3 100644
> --- a/gcc/config/i386/i386.opt
> +++ b/gcc/config/i386/i386.opt
> @@ -1288,3 +1288,13 @@ msm3
>  Target Mask(ISA2_SM3) Var(ix86_isa_flags2) Save
>  Support MMX, SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2, AVX and
>  SM3 built-in functions and code generation.
> +
> +mvpinsrvpextr
> +Target Mask(ISA2_VPINSRVPEXTR) Var(ix86_isa_flags2) Save
> +Support MMX, SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2, AVX, AVX2, AVX512F,
> +AVX512VL and VPINSRVPEXTR built-in functions and code generation.
> +
> +msha512
> +Target Mask(ISA2_SHA512) Var(ix86_isa_flags2) Save
> +Support MMX, SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2, AVX and
> +SHA512 built-in functions and code generation.
> diff --git a/gcc/config/i386/immintrin.h b/gcc/config/i386/immintrin.h
> index 7731990131c..6f2bcef6a8a 100644
> --- a/gcc/config/i386/immintrin.h
> +++ b/gcc/config/i386/immintrin.h
> @@ -110,6 +110,8 @@
>
>  #include <sm3intrin.h>
>
> +#include <sha512intrin.h>
> +
>  #include <fmaintrin.h>
>
>  #include <f16cintrin.h>
> diff --git a/gcc/config/i386/sha512intrin.h b/gcc/config/i386/sha512intrin.h
> new file mode 100644
> index 00000000000..884c2bc6340
> --- /dev/null
> +++ b/gcc/config/i386/sha512intrin.h
> @@ -0,0 +1,64 @@
> +/* Copyright (C) 2023 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/>.  */
> +
> +#ifndef _IMMINTRIN_H_INCLUDED
> +#error "Never use <sha512intrin.h> directly; include <immintrin.h> instead."
> +#endif
> +
> +#ifndef _SHA512INTRIN_H_INCLUDED
> +#define _SHA512INTRIN_H_INCLUDED
> +
> +#ifndef __SHA512__
> +#pragma GCC push_options
> +#pragma GCC target("sha512")
> +#define __DISABLE_SHA512__
> +#endif /* __SHA512__ */
> +
> +extern __inline __m256i
> +__attribute__((__gnu_inline__, __always_inline__, __artificial__))
> +_mm256_sha512msg1_epi64 (__m256i __A, __m128i __B)
> +{
> +  return (__m256i) __builtin_ia32_vsha512msg1 ((__v4di) __A, (__v2di) __B);
> +}
> +
> +extern __inline __m256i
> +__attribute__((__gnu_inline__, __always_inline__, __artificial__))
> +_mm256_sha512msg2_epi64 (__m256i __A, __m256i __B)
> +{
> +  return (__m256i) __builtin_ia32_vsha512msg2 ((__v4di) __A, (__v4di) __B);
> +}
> +
> +extern __inline __m256i
> +__attribute__((__gnu_inline__, __always_inline__, __artificial__))
> +_mm256_sha512rnds2_epi64 (__m256i __A, __m256i __B, __m128i __C)
> +{
> +  return (__m256i) __builtin_ia32_vsha512rnds2 ((__v4di) __A, (__v4di) __B,
> +                                               (__v2di) __C);
> +}
> +
> +#ifdef __DISABLE_SHA512__
> +#undef __DISABLE_SHA512__
> +#pragma GCC pop_options
> +#endif /* __DISABLE_SHA512__ */
> +
> +#endif /* _SHA512INTRIN_H_INCLUDED */
> diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md
> index 25a1e5dd780..e16b2b5a6c4 100644
> --- a/gcc/config/i386/sse.md
> +++ b/gcc/config/i386/sse.md
> @@ -217,6 +217,12 @@
>    UNSPEC_VPDPWSUDS
>    UNSPEC_VPDPWUUD
>    UNSPEC_VPDPWUUDS
> +
> +  ;; For SHA512 support
> +  UNSPEC_SHA512MSG1
> +  UNSPEC_SHA512MSG2
> +  UNSPEC_SHA512RNDS2
> +
>  ])
>
>  (define_c_enum "unspecv" [
> @@ -28640,6 +28646,40 @@
>     (set_attr "mode" "TI")
>     (set_attr "length_immediate" "1")])
>
> +(define_insn "vsha512msg1"
> +  [(set (match_operand:V4DI 0 "register_operand" "=x")
> +       (unspec:V4DI
> +         [(match_operand:V4DI 1 "register_operand" "0")
> +          (match_operand:V2DI 2 "register_operand" "x")]
> +         UNSPEC_SHA512MSG1))]
> +  "TARGET_SHA512"
> +  "vsha512msg1\t{%2, %0|%0, %2}"
> +  [(set_attr "type" "sselog1")
> +   (set_attr "mode" "OI")])
> +
> +(define_insn "vsha512msg2"
> +  [(set (match_operand:V4DI 0 "register_operand" "=x")
> +       (unspec:V4DI
> +         [(match_operand:V4DI 1 "register_operand" "0")
> +          (match_operand:V4DI 2 "register_operand" "x")]
> +         UNSPEC_SHA512MSG2))]
> +  "TARGET_SHA512"
> +  "vsha512msg2\t{%2, %0|%0, %2}"
> +  [(set_attr "type" "sselog1")
> +   (set_attr "mode" "OI")])
> +
> +(define_insn "vsha512rnds2"
> +  [(set (match_operand:V4DI 0 "register_operand" "=x")
> +       (unspec:V4DI
> +         [(match_operand:V4DI 1 "register_operand" "0")
> +          (match_operand:V4DI 2 "register_operand" "x")
> +          (match_operand:V2DI 3 "register_operand" "x")]
> +         UNSPEC_SHA512RNDS2))]
> +  "TARGET_SHA512"
> +  "vsha512rnds2\t{%3, %2, %0|%0, %2, %3}"
> +  [(set_attr "type" "sselog1")
> +   (set_attr "mode" "OI")])
> +
>  (define_insn_and_split "avx512f_<castmode><avxsizesuffix>_<castmode>"
>    [(set (match_operand:AVX512MODE2P 0 "nonimmediate_operand" "=x,m")
>         (vec_concat:AVX512MODE2P
> diff --git a/gcc/doc/extend.texi b/gcc/doc/extend.texi
> index e76cd399a83..5250990050b 100644
> --- a/gcc/doc/extend.texi
> +++ b/gcc/doc/extend.texi
> @@ -7173,6 +7173,11 @@ Enable/disable the generation of the AVXVNNIINT16 instructions.
>  @itemx no-sm3
>  Enable/disable the generation of the SM3 instructions.
>
> +@cindex @code{target("sha512")} function attribute, x86
> +@item sha512
> +@itemx no-sha512
> +Enable/disable the generation of the SHA512 instructions.
> +
>  @cindex @code{target("cld")} function attribute, x86
>  @item cld
>  @itemx no-cld
> diff --git a/gcc/doc/invoke.texi b/gcc/doc/invoke.texi
> index 2671d70736f..433ccf35505 100644
> --- a/gcc/doc/invoke.texi
> +++ b/gcc/doc/invoke.texi
> @@ -1433,7 +1433,7 @@ See RS/6000 and PowerPC Options.
>  -mrdseed  -msgx -mavx512vp2intersect -mserialize -mtsxldtrk
>  -mamx-tile  -mamx-int8  -mamx-bf16 -muintr -mhreset -mavxvnni
>  -mavx512fp16 -mavxifma -mavxvnniint8 -mavxneconvert -mcmpccxadd -mamx-fp16
> --mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3
> +-mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3 -msha512
>  -mcldemote  -mms-bitfields  -mno-align-stringops  -minline-all-stringops
>  -minline-stringops-dynamically  -mstringop-strategy=@var{alg}
>  -mkl -mwidekl
> @@ -33558,6 +33558,9 @@ preferred alignment to @option{-mpreferred-stack-boundary=2}.
>  @need 200
>  @opindex msm3
>  @itemx -msm3
> +@need 200
> +@opindex msha512
> +@itemx -msha512
>  These switches enable the use of instructions in the MMX, SSE,
>  AVX512ER, AVX512CD, AVX512VL, AVX512BW, AVX512DQ, AVX512IFMA, AVX512VBMI, SHA,
>  AES, PCLMUL, CLFLUSHOPT, CLWB, FSGSBASE, PTWRITE, RDRND, F16C, FMA, PCONFIG,
> @@ -33568,8 +33571,9 @@ GFNI, VAES, WAITPKG, VPCLMULQDQ, AVX512BITALG, MOVDIRI, MOVDIR64B, AVX512BF16,
>  ENQCMD, AVX512VPOPCNTDQ, AVX5124FMAPS, AVX512VNNI, AVX5124VNNIW, SERIALIZE,
>  UINTR, HRESET, AMXTILE, AMXINT8, AMXBF16, KL, WIDEKL, AVXVNNI, AVX512-FP16,
>  AVXIFMA, AVXVNNIINT8, AVXNECONVERT, CMPCCXADD, AMX-FP16, PREFETCHI, RAOINT,
> -AMX-COMPLEX, AVXVNNIINT16, SM3 or CLDEMOTE extended instruction sets. Each has a
> -corresponding @option{-mno-} option to disable use of these instructions.
> +AMX-COMPLEX, AVXVNNIINT16, SM3, SHA512 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 dae51132c42..54a062db3fe 100644
> --- a/gcc/doc/sourcebuild.texi
> +++ b/gcc/doc/sourcebuild.texi
> @@ -2589,6 +2589,9 @@ Target supports the execution of @code{raoint} instructions.
>  @item rdrand
>  Target supports x86 @code{rdrand} instruction.
>
> +@item sha512
> +Target supports the execution of @code{sha512} instructions.
> +
>  @item sm3
>  Target supports the execution of @code{sm3} instructions.
>
> diff --git a/gcc/testsuite/g++.dg/other/i386-2.C b/gcc/testsuite/g++.dg/other/i386-2.C
> index 2ec93261cac..985f14abcbc 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 -mavxvnniint8 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3" } */
> +/* { 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 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3 -msha512" } */
>
>  /* 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 fe03143e39b..274b0e6256f 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 -mavxvnniint8 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3" } */
> +/* { 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 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3 -msha512" } */
>
>  /* 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/funcspec-56.inc b/gcc/testsuite/gcc.target/i386/funcspec-56.inc
> index 8dd8d9bf9d8..eb9309819ab 100644
> --- a/gcc/testsuite/gcc.target/i386/funcspec-56.inc
> +++ b/gcc/testsuite/gcc.target/i386/funcspec-56.inc
> @@ -90,6 +90,7 @@ extern void test_raoint (void)                  __attribute__((__target__("raoin
>  extern void test_amx_complex (void)            __attribute__((__target__("amx-complex")));
>  extern void test_avxvnniint16 (void)           __attribute__((__target__("avxvnniint16")));
>  extern void test_sm3 (void)                    __attribute__((__target__("sm3")));
> +extern void test_sha512 (void)                 __attribute__((__target__("sha512")));
>
>  extern void test_no_sgx (void)                 __attribute__((__target__("no-sgx")));
>  extern void test_no_avx5124fmaps(void)         __attribute__((__target__("no-avx5124fmaps")));
> @@ -181,6 +182,7 @@ extern void test_no_raoint (void)               __attribute__((__target__("no-ra
>  extern void test_no_amx_complex (void)         __attribute__((__target__("no-amx-complex")));
>  extern void test_no_avxvnniint16 (void)                __attribute__((__target__("no-avxvnniint16")));
>  extern void test_no_sm3 (void)                 __attribute__((__target__("no-sm3")));
> +extern void test_no_sha512 (void)              __attribute__((__target__("no-sha512")));
>
>  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/sha512-1.c b/gcc/testsuite/gcc.target/i386/sha512-1.c
> new file mode 100644
> index 00000000000..c66e8124d6b
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/i386/sha512-1.c
> @@ -0,0 +1,18 @@
> +/* { dg-do compile } */
> +/* { dg-options "-O2 -msha512" } */
> +/* { dg-final { scan-assembler "vsha512msg1\[ \\t\]+\[^\n\]*%xmm\[0-9\]+\[^\n\]*%ymm\[0-9\]" } } */
> +/* { dg-final { scan-assembler "vsha512msg2\[ \\t\]+\[^\n\]*%ymm\[0-9\]+\[^\n\]*%ymm\[0-9\]" } } */
> +/* { dg-final { scan-assembler "vsha512rnds2\[ \\t\]+\[^\n\]*%xmm\[0-9\]+\[^\n\]*%ymm\[0-9\]+\[^\n\]*%ymm\[0-9\]" } } */
> +
> +#include <immintrin.h>
> +
> +volatile __m128i x;
> +volatile __m256i y;
> +
> +void extern
> +sha512_test (void)
> +{
> +  y = _mm256_sha512msg1_epi64(y, x);
> +  y = _mm256_sha512msg2_epi64(y, y);
> +  y = _mm256_sha512rnds2_epi64(y, y, x);
> +}
> diff --git a/gcc/testsuite/gcc.target/i386/sha512-check.h b/gcc/testsuite/gcc.target/i386/sha512-check.h
> new file mode 100644
> index 00000000000..083bf3b6a1b
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/i386/sha512-check.h
> @@ -0,0 +1,43 @@
> +#include <stdlib.h>
> +#include "m256-check.h"
> +
> +static void sha512_test (void);
> +
> +static unsigned long long
> +ror64 (unsigned long long w, int n)
> +{
> +  int count = n % 64;
> +  return ((w >> n) | (w << (64 - n)));
> +}
> +
> +static unsigned long long
> +shr64 (unsigned long long w, int n)
> +{
> +  return (w >> n);
> +}
> +
> +static void
> +__attribute__ ((noinline))
> +do_test(void)
> +{
> +  sha512_test ();
> +}
> +
> +int
> +main ()
> +{
> +  /* Check CPU support for SHA512.  */
> +  if (__builtin_cpu_supports ("sha512"))
> +    {
> +      do_test ();
> +#ifdef DEBUG
> +      printf ("PASSED\n");
> +#endif
> +      return 0;
> +    }
> +
> +#ifdef DEBUG
> +  printf ("SKIPPED\n");
> +#endif
> +  return 0;
> +}
> diff --git a/gcc/testsuite/gcc.target/i386/sha512msg1-2.c b/gcc/testsuite/gcc.target/i386/sha512msg1-2.c
> new file mode 100644
> index 00000000000..b7baff1cfb4
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/i386/sha512msg1-2.c
> @@ -0,0 +1,48 @@
> +/* { dg-do run } */
> +/* { dg-options "-O2 -msha512" } */
> +/* { dg-require-effective-target sha512 } */
> +
> +#include "sha512-check.h"
> +#include <emmintrin.h>
> +#include <immintrin.h>
> +
> +static unsigned long long
> +s0 (unsigned long long w)
> +{
> +  return ror64 (w, 1) ^ ror64 (w, 8) ^ shr64 (w, 7);
> +}
> +
> +static void
> +compute_sha512msg1(long long* src1, long long* src2, long long* res)
> +{
> +  unsigned long long w0, w1, w2, w3, w4;
> +
> +  w0 = src1[0];
> +  w1 = src1[1];
> +  w2 = src1[2];
> +  w3 = src1[3];
> +  w4 = src2[0];
> +
> +  res[0] = w0 + s0 (w1);
> +  res[1] = w1 + s0 (w2);
> +  res[2] = w2 + s0 (w3);
> +  res[3] = w3 + s0 (w4);
> +}
> +
> +static void
> +sha512_test(void)
> +{
> +  union256i_q s1, res;
> +  union128i_q s2;
> +  long long res_ref[4];
> +
> +  s1.x = _mm256_set_epi64x (111, 222, 333, 444);
> +  s2.x = _mm_set_epi64x (0, 555);
> +
> +  res.x = _mm256_sha512msg1_epi64 (s1.x, s2.x);
> +
> +  compute_sha512msg1 (s1.a, s2.a, res_ref);
> +
> +  if (check_union256i_q (res, res_ref))
> +    abort();
> +}
> diff --git a/gcc/testsuite/gcc.target/i386/sha512msg2-2.c b/gcc/testsuite/gcc.target/i386/sha512msg2-2.c
> new file mode 100644
> index 00000000000..e5033903f06
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/i386/sha512msg2-2.c
> @@ -0,0 +1,47 @@
> +/* { dg-do run } */
> +/* { dg-options "-O2 -msha512" } */
> +/* { dg-require-effective-target sha512 } */
> +
> +#include "sha512-check.h"
> +#include <immintrin.h>
> +
> +static unsigned long long
> +s1 (unsigned long long w)
> +{
> +  return ror64 (w, 19) ^ ror64 (w, 61) ^ shr64 (w, 6);
> +}
> +
> +static void
> +compute_sha512msg2 (long long* src1, long long* src2, long long* res)
> +{
> +  unsigned long long w14, w15, w16, w17, w18, w19;
> +
> +  w14 = src2[2];
> +  w15 = src2[3];
> +  w16 = src1[0] + s1 (w14);
> +  w17 = src1[1] + s1 (w15);
> +  w18 = src1[2] + s1 (w16);
> +  w19 = src1[3] + s1 (w17);
> +
> +  res[0] = w16;
> +  res[1] = w17;
> +  res[2] = w18;
> +  res[3] = w19;
> +}
> +
> +static void
> +sha512_test (void)
> +{
> +  union256i_q s1, s2, res;
> +  long long res_ref[4];
> +
> +  s1.x = _mm256_set_epi64x (111, 222, 333, 444);
> +  s2.x = _mm256_set_epi64x (555, 666, 0, 0);
> +
> +  res.x = _mm256_sha512msg2_epi64 (s1.x, s2.x);
> +
> +  compute_sha512msg2 (s1.a, s2.a, res_ref);
> +
> +  if (check_union256i_q (res, res_ref))
> +    abort();
> +}
> diff --git a/gcc/testsuite/gcc.target/i386/sha512rnds2-2.c b/gcc/testsuite/gcc.target/i386/sha512rnds2-2.c
> new file mode 100644
> index 00000000000..3bc6a00a05b
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/i386/sha512rnds2-2.c
> @@ -0,0 +1,85 @@
> +/* { dg-do run } */
> +/* { dg-options "-O2 -msha512" } */
> +/* { dg-require-effective-target sha512 } */
> +
> +#include "sha512-check.h"
> +#include <emmintrin.h>
> +#include <immintrin.h>
> +
> +static unsigned long long
> +ch (unsigned long long e, unsigned long long f, unsigned long long g)
> +{
> +  return (e & f) ^ (~e & g);
> +}
> +
> +static unsigned long long
> +maj (unsigned long long a, unsigned long long b, unsigned long long c)
> +{
> +  return (a & b) ^ (a & c) ^ (b & c);
> +}
> +
> +static unsigned long long
> +s0 (unsigned long long w)
> +{
> +  return ror64 (w, 28) ^ ror64 (w, 34) ^ ror64 (w, 39);
> +}
> +
> +static unsigned long long
> +s1 (unsigned long long w)
> +{
> +  return ror64 (w, 14) ^ ror64 (w, 18) ^ ror64 (w, 41);
> +}
> +
> +static void
> +compute_sha512rnds2(long long* src0, long long* src1, long long* src2, long long* res)
> +{
> +  unsigned long long wk[2] = { src2[0], src2[1] };
> +  unsigned long long a[3], b[3], c[3], d[3], e[3], f[3], g[3], h[3];
> +
> +  a[0] = src1[3];
> +  b[0] = src1[2];
> +  c[0] = src0[3];
> +  d[0] = src0[2];
> +  e[0] = src1[1];
> +  f[0] = src1[0];
> +  g[0] = src0[1];
> +  h[0] = src0[0];
> +
> +  int i;
> +  for (i = 0; i <= 1; i++)
> +  {
> +    a[i + 1] = ch (e[i], f[i], g[i]) + s1 (e[i]) + wk[i] + h[i]
> +      + maj (a[i], b[i], c[i]) + s0 (a[i]);
> +    b[i + 1] = a[i];
> +    c[i + 1] = b[i];
> +    d[i + 1] = c[i];
> +    e[i + 1] = ch (e[i], f[i], g[i]) + s1 (e[i]) + wk[i] + h[i] + d[i];
> +    f[i + 1] = e[i];
> +    g[i + 1] = f[i];
> +    h[i + 1] = g[i];
> +  }
> +
> +  res[0] = f[2];
> +  res[1] = e[2];
> +  res[2] = b[2];
> +  res[3] = a[2];
> +}
> +
> +static void
> +sha512_test (void)
> +{
> +  union256i_q s0, s1, res;
> +  union128i_q s2;
> +  long long res_ref[4];
> +
> +  s0.x = _mm256_set_epi64x (111, 222, 333, 444);
> +  s1.x = _mm256_set_epi64x (555, 666, 777, 888);
> +  s2.x = _mm_set_epi64x (999, 123);
> +
> +  res.x = _mm256_sha512rnds2_epi64 (s0.x, s1.x, s2.x);
> +
> +  compute_sha512rnds2 (s0.a, s1.a, s2.a, res_ref);
> +
> +  if (check_union256i_q (res, res_ref))
> +    abort();
> +}
> diff --git a/gcc/testsuite/gcc.target/i386/sse-12.c b/gcc/testsuite/gcc.target/i386/sse-12.c
> index 5058be6f6e9..976541389ea 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 -mavxvnniint8 -mavxneconvert -mamx-fp16 -mraoint -mamx-complex -mavxvnniint16 -msm3" } */
> +/* { 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 -mavxneconvert -mamx-fp16 -mraoint -mamx-complex -mavxvnniint16 -msm3 -msha512" } */
>
>  #include <x86intrin.h>
>
> diff --git a/gcc/testsuite/gcc.target/i386/sse-13.c b/gcc/testsuite/gcc.target/i386/sse-13.c
> index d30b365564a..8c314e70e31 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 -mavxvnniint8 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3" } */
> +/* { 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 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3 -msha512" } */
>  /* { 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 7842005a98b..2b4d7bc9079 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 -mavxvnniint8 -mavxneconvert -mamx-fp16 -mraoint -mamx-complex -mavxvnniint16 -msm3" } */
> +/* { 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 -mavxneconvert -mamx-fp16 -mraoint -mamx-complex -mavxvnniint16 -msm3 -msha512" } */
>  /* { 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 7537db1ac30..d6f19b5e20a 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,tbm,lwp,fsgsbase,rdrnd,f16c,rtm,rdseed,prfchw,adx,fxsr,xsaveopt,avx512f,avx512er,avx512cd,avx512pf,sha,prefetchwt1,avx512vl,avx512bw,avx512dq,avx512vbmi,avx512vbmi2,avx512ifma,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,gfni,avx512bitalg,avx512bf16,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,amx-fp16,raoint,amx-complex,avxvnniint16,sm3")
> +#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,avx512f,avx512er,avx512cd,avx512pf,sha,prefetchwt1,avx512vl,avx512bw,avx512dq,avx512vbmi,avx512vbmi2,avx512ifma,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,gfni,avx512bitalg,avx512bf16,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,amx-fp16,raoint,amx-complex,avxvnniint16,sm3,sha512")
>  #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,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,gfni,avx512bitalg,avx512bf16,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,amx-fp16,raoint,amx-complex,avxvnniint16,sm3")
> +#pragma GCC target ("avx,avx2,rdrnd,fsgsbase,f16c,rtm,avx512f,avx512er,avx512cd,avx512pf,sha,avx512vl,avx512bw,avx512dq,avx512ifma,avx512vbmi,avx512vbmi2,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,gfni,avx512bitalg,avx512bf16,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,amx-fp16,raoint,amx-complex,avxvnniint16,sm3,sha512")
>  #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 3fc61b50fe6..1df66b525ca 100644
> --- a/gcc/testsuite/gcc.target/i386/sse-23.c
> +++ b/gcc/testsuite/gcc.target/i386/sse-23.c
> @@ -850,6 +850,6 @@
>  /* sm3intrin.h */
>  #define __builtin_ia32_vsm3rnds2(A, B, C, D) __builtin_ia32_vsm3rnds2 (A, B, C, 1)
>
> -#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,avx512f,avx512er,avx512cd,avx512pf,sha,prefetchwt1,xsavec,xsaves,clflushopt,avx512bw,avx512dq,avx512vl,avx512vbmi,avx512ifma,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,clwb,mwaitx,clzero,pku,sgx,rdpid,gfni,avx512vbmi2,vpclmulqdq,avx512bitalg,pconfig,wbnoinvd,avx512bf16,enqcmd,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,cmpccxadd,amx-fp16,prefetchi,raoint,amx-complex,avxvnniint16,sm3")
> +#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,avx512f,avx512er,avx512cd,avx512pf,sha,prefetchwt1,xsavec,xsaves,clflushopt,avx512bw,avx512dq,avx512vl,avx512vbmi,avx512ifma,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,clwb,mwaitx,clzero,pku,sgx,rdpid,gfni,avx512vbmi2,vpclmulqdq,avx512bitalg,pconfig,wbnoinvd,avx512bf16,enqcmd,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,cmpccxadd,amx-fp16,prefetchi,raoint,amx-complex,avxvnniint16,sm3,sha512")
>
>  #include <x86intrin.h>
> diff --git a/gcc/testsuite/lib/target-supports.exp b/gcc/testsuite/lib/target-supports.exp
> index c911a824d31..f376d835f8b 100644
> --- a/gcc/testsuite/lib/target-supports.exp
> +++ b/gcc/testsuite/lib/target-supports.exp
> @@ -9883,6 +9883,20 @@ proc check_effective_target_sm3 { } {
>      } "-msm3" ]
>  }
>
> +# Return 1 if sha512 instructions can be compiled.
> +proc check_effective_target_sha512 { } {
> +    return [check_no_compiler_messages sha512 object {
> +       typedef long long __m256i __attribute__ ((__vector_size__ (32)));
> +       typedef long long __v4di __attribute__ ((__vector_size__ (32)));
> +       __m256i
> +       _mm256_sha512msg2_epi64 (__m256i __A, __m256i __B)
> +        {
> +          return (__m256i) __builtin_ia32_vsha512msg2 ((__v4di) __A,
> +                                                       (__v4di) __B);
> +       }
> +    } "-msha512" ]
> +}
> +
>  # Return 1 if sse instructions can be compiled.
>  proc check_effective_target_sse { } {
>      return [check_no_compiler_messages sse object {
> --
> 2.31.1
>


-- 
BR,
Hongtao

^ permalink raw reply	[flat|nested] 9+ messages in thread

* [PATCH 4/4] Support Intel SM4
  2023-07-13  6:03 [PATCH 0/4] Support Intel Arrow Lake/Lunar Lake ISAs Haochen Jiang
                   ` (2 preceding siblings ...)
  2023-07-13  6:03 ` [PATCH 3/4] Support Intel SHA512 Haochen Jiang
@ 2023-07-13  6:03 ` Haochen Jiang
  2023-07-17  1:40   ` Hongtao Liu
  3 siblings, 1 reply; 9+ messages in thread
From: Haochen Jiang @ 2023-07-13  6:03 UTC (permalink / raw)
  To: gcc-patches; +Cc: hongtao.liu, ubizjak

gcc/ChangeLog:

	* common/config/i386/cpuinfo.h (get_available_features):
	Detech SM4.
	* common/config/i386/i386-common.cc (OPTION_MASK_ISA2_SM4_SET,
	OPTION_MASK_ISA2_SM4_UNSET): New.
	(OPTION_MASK_ISA2_AVX_UNSET): Add SM4.
	(ix86_handle_option): Handle -msm4.
	* common/config/i386/i386-cpuinfo.h (enum processor_features):
	Add FEATURE_SM4.
	* common/config/i386/i386-isas.h: Add ISA_NAME_TABLE_ENTRY for
	sm4.
	* config.gcc: Add sm4intrin.h.
	* config/i386/cpuid.h (bit_SM4): New.
	* config/i386/i386-builtin.def (BDESC): Add new builtins.
	* config/i386/i386-c.cc (ix86_target_macros_internal): Define
	__SM4__.
	* config/i386/i386-isa.def (SM4): Add DEF_PTA(SM4).
	* config/i386/i386-options.cc (isa2_opts): Add -msm4.
	(ix86_valid_target_attribute_inner_p): Handle sm4.
	* config/i386/i386.opt: Add option -msm4.
	* config/i386/immintrin.h: Include sm4intrin.h
	* config/i386/sse.md (vsm4key4_<mode>): New define insn.
	(vsm4rnds4_<mode>): Ditto.
	* doc/extend.texi: Document sm4.
	* doc/invoke.texi: Document -msm4.
	* doc/sourcebuild.texi: Document target sm4.
	* config/i386/sm4intrin.h: New file.

gcc/testsuite/ChangeLog:

	* g++.dg/other/i386-2.C: Add -msm4.
	* g++.dg/other/i386-3.C: Ditto.
	* gcc.target/i386/funcspec-56.inc: Add new target attribute.
	* gcc.target/i386/sse-12.c: Add -msm4.
	* gcc.target/i386/sse-13.c: Ditto.
	* gcc.target/i386/sse-14.c: Ditto.
	* gcc.target/i386/sse-22.c: Add sm4.
	* gcc.target/i386/sse-23.c: Ditto.
	* lib/target-supports.exp (check_effective_target_sm4): New.
	* gcc.target/i386/sm4-1.c: New test.
	* gcc.target/i386/sm4-check.h: Ditto.
	* gcc.target/i386/sm4key4-2.c: Ditto.
        * gcc.target/i386/sm4rnds4-2.c: Ditto.
---
 gcc/common/config/i386/cpuinfo.h              |   2 +
 gcc/common/config/i386/i386-common.cc         |  20 +-
 gcc/common/config/i386/i386-cpuinfo.h         |   1 +
 gcc/common/config/i386/i386-isas.h            |   1 +
 gcc/config.gcc                                |   2 +-
 gcc/config/i386/cpuid.h                       |   1 +
 gcc/config/i386/i386-builtin.def              |   6 +
 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/sm4intrin.h                   |  70 +++++++
 gcc/config/i386/sse.md                        |  26 +++
 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/funcspec-56.inc |   2 +
 gcc/testsuite/gcc.target/i386/sm4-1.c         |  20 ++
 gcc/testsuite/gcc.target/i386/sm4-check.h     | 183 ++++++++++++++++++
 gcc/testsuite/gcc.target/i386/sm4key4-2.c     |  14 ++
 gcc/testsuite/gcc.target/i386/sm4rnds4-2.c    |  14 ++
 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         |  14 ++
 30 files changed, 409 insertions(+), 14 deletions(-)
 create mode 100644 gcc/config/i386/sm4intrin.h
 create mode 100644 gcc/testsuite/gcc.target/i386/sm4-1.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sm4-check.h
 create mode 100644 gcc/testsuite/gcc.target/i386/sm4key4-2.c
 create mode 100644 gcc/testsuite/gcc.target/i386/sm4rnds4-2.c

diff --git a/gcc/common/config/i386/cpuinfo.h b/gcc/common/config/i386/cpuinfo.h
index 0cfde3ebccd..f9434f038ea 100644
--- a/gcc/common/config/i386/cpuinfo.h
+++ b/gcc/common/config/i386/cpuinfo.h
@@ -881,6 +881,8 @@ get_available_features (struct __processor_model *cpu_model,
 	    set_feature (FEATURE_SM3);
 	  if (eax & bit_SHA512)
 	    set_feature (FEATURE_SHA512);
+	  if (eax & bit_SM4)
+	    set_feature (FEATURE_SM4);
 	}
       if (avx512_usable)
 	{
diff --git a/gcc/common/config/i386/i386-common.cc b/gcc/common/config/i386/i386-common.cc
index 97c3cdfe5e1..610cabe52c1 100644
--- a/gcc/common/config/i386/i386-common.cc
+++ b/gcc/common/config/i386/i386-common.cc
@@ -122,6 +122,7 @@ along with GCC; see the file COPYING3.  If not see
 #define OPTION_MASK_ISA2_AVXVNNIINT16_SET OPTION_MASK_ISA2_AVXVNNIINT16
 #define OPTION_MASK_ISA2_SM3_SET OPTION_MASK_ISA2_SM3
 #define OPTION_MASK_ISA2_SHA512_SET OPTION_MASK_ISA2_SHA512
+#define OPTION_MASK_ISA2_SM4_SET OPTION_MASK_ISA2_SM4
 
 /* SSE4 includes both SSE4.1 and SSE4.2. -msse4 should be the same
    as -msse4.2.  */
@@ -307,6 +308,7 @@ along with GCC; see the file COPYING3.  If not see
 #define OPTION_MASK_ISA2_AVXVNNIINT16_UNSET OPTION_MASK_ISA2_AVXVNNIINT16
 #define OPTION_MASK_ISA2_SM3_UNSET OPTION_MASK_ISA2_SM3
 #define OPTION_MASK_ISA2_SHA512_UNSET OPTION_MASK_ISA2_SHA512
+#define OPTION_MASK_ISA2_SM4_UNSET OPTION_MASK_ISA2_SM4
 
 /* SSE4 includes both SSE4.1 and SSE4.2.  -mno-sse4 should the same
    as -mno-sse4.1. */
@@ -356,7 +358,8 @@ along with GCC; see the file COPYING3.  If not see
   OPTION_MASK_ISA2_SSE_UNSET
 #define OPTION_MASK_ISA2_AVX_UNSET \
   (OPTION_MASK_ISA2_AVX2_UNSET | OPTION_MASK_ISA2_VAES_UNSET \
-   | OPTION_MASK_ISA2_SM3_UNSET | OPTION_MASK_ISA2_SHA512_UNSET)
+   | OPTION_MASK_ISA2_SM3_UNSET | OPTION_MASK_ISA2_SHA512_UNSET \
+   | OPTION_MASK_ISA2_SM4_UNSET)
 #define OPTION_MASK_ISA2_SSE4_2_UNSET OPTION_MASK_ISA2_AVX_UNSET
 #define OPTION_MASK_ISA2_SSE4_1_UNSET OPTION_MASK_ISA2_SSE4_2_UNSET
 #define OPTION_MASK_ISA2_SSE4_UNSET OPTION_MASK_ISA2_SSE4_1_UNSET
@@ -1323,6 +1326,21 @@ ix86_handle_option (struct gcc_options *opts,
 	}
       return true;
 
+    case OPT_msm4:
+      if (value)
+	{
+	  opts->x_ix86_isa_flags2 |= OPTION_MASK_ISA2_SM4_SET;
+	  opts->x_ix86_isa_flags2_explicit |= OPTION_MASK_ISA2_SM4_SET;
+	  opts->x_ix86_isa_flags |= OPTION_MASK_ISA_AVX_SET;
+	  opts->x_ix86_isa_flags_explicit |= OPTION_MASK_ISA_AVX_SET;
+	}
+      else
+	{
+	  opts->x_ix86_isa_flags2 &= ~OPTION_MASK_ISA2_SM4_UNSET;
+	  opts->x_ix86_isa_flags2_explicit |= OPTION_MASK_ISA2_SM4_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 a6e34d14f8e..be04d85c9d5 100644
--- a/gcc/common/config/i386/i386-cpuinfo.h
+++ b/gcc/common/config/i386/i386-cpuinfo.h
@@ -258,6 +258,7 @@ enum processor_features
   FEATURE_AVXVNNIINT16,
   FEATURE_SM3,
   FEATURE_SHA512,
+  FEATURE_SM4,
   CPU_FEATURE_MAX
 };
 
diff --git a/gcc/common/config/i386/i386-isas.h b/gcc/common/config/i386/i386-isas.h
index 250dc87764f..2297903a45e 100644
--- a/gcc/common/config/i386/i386-isas.h
+++ b/gcc/common/config/i386/i386-isas.h
@@ -190,4 +190,5 @@ ISA_NAMES_TABLE_START
 			P_NONE, "-mavxvnniint16")
   ISA_NAMES_TABLE_ENTRY("sm3", FEATURE_SM3, P_NONE, "-msm3")
   ISA_NAMES_TABLE_ENTRY("sha512", FEATURE_SHA512, P_NONE, "-msha512")
+  ISA_NAMES_TABLE_ENTRY("sm4", FEATURE_SM4, P_NONE, "-msm4")
 ISA_NAMES_TABLE_END
diff --git a/gcc/config.gcc b/gcc/config.gcc
index 4e753ba7c64..305e859880f 100644
--- a/gcc/config.gcc
+++ b/gcc/config.gcc
@@ -436,7 +436,7 @@ i[34567]86-*-* | x86_64-*-*)
 		       avxifmaintrin.h avxvnniint8intrin.h avxneconvertintrin.h
 		       cmpccxaddintrin.h amxfp16intrin.h prfchiintrin.h
 		       raointintrin.h amxcomplexintrin.h avxvnniint16intrin.h
-		       sm3intrin.h sha512intrin.h"
+		       sm3intrin.h sha512intrin.h sm4intrin.h"
 	;;
 ia64-*-*)
 	extra_headers=ia64intrin.h
diff --git a/gcc/config/i386/cpuid.h b/gcc/config/i386/cpuid.h
index f9103f1b1c9..03fd6fc9478 100644
--- a/gcc/config/i386/cpuid.h
+++ b/gcc/config/i386/cpuid.h
@@ -134,6 +134,7 @@
 /* %eax */
 #define bit_SHA512	(1 << 0)
 #define bit_SM3		(1 << 1)
+#define bit_SM4		(1 << 2)
 #define bit_RAOINT      (1 << 3)
 #define bit_AVXVNNI     (1 << 4)
 #define bit_AVX512BF16  (1 << 5)
diff --git a/gcc/config/i386/i386-builtin.def b/gcc/config/i386/i386-builtin.def
index b9e2bad8522..8738b3b6a8a 100644
--- a/gcc/config/i386/i386-builtin.def
+++ b/gcc/config/i386/i386-builtin.def
@@ -1660,6 +1660,12 @@ BDESC (OPTION_MASK_ISA_AVX, OPTION_MASK_ISA2_SM3, CODE_FOR_vsm3msg1, "__builtin_
 BDESC (OPTION_MASK_ISA_AVX, OPTION_MASK_ISA2_SM3, CODE_FOR_vsm3msg2, "__builtin_ia32_vsm3msg2", IX86_BUILTIN_VSM3MSG2, UNKNOWN, (int) V4SI_FTYPE_V4SI_V4SI_V4SI)
 BDESC (OPTION_MASK_ISA_AVX, OPTION_MASK_ISA2_SM3, CODE_FOR_vsm3rnds2, "__builtin_ia32_vsm3rnds2", IX86_BUILTIN_VSM3RNDS2, UNKNOWN, (int) V4SI_FTYPE_V4SI_V4SI_V4SI_INT)
 
+/* SM4 */
+BDESC (0, OPTION_MASK_ISA2_SM4, CODE_FOR_vsm4key4_v4si, "__builtin_ia32_vsm4key4128", IX86_BUILTIN_VSM4KEY4128, UNKNOWN, (int) V4SI_FTYPE_V4SI_V4SI)
+BDESC (0, OPTION_MASK_ISA2_SM4, CODE_FOR_vsm4key4_v8si, "__builtin_ia32_vsm4key4256", IX86_BUILTIN_VSM4KEY4256, UNKNOWN, (int) V8SI_FTYPE_V8SI_V8SI)
+BDESC (0, OPTION_MASK_ISA2_SM4, CODE_FOR_vsm4rnds4_v4si, "__builtin_ia32_vsm4rnds4128", IX86_BUILTIN_VSM4RNDS4128, UNKNOWN, (int) V4SI_FTYPE_V4SI_V4SI)
+BDESC (0, OPTION_MASK_ISA2_SM4, CODE_FOR_vsm4rnds4_v8si, "__builtin_ia32_vsm4rnds4256", IX86_BUILTIN_VSM4RNDS4256, UNKNOWN, (int) V8SI_FTYPE_V8SI_V8SI)
+
 /* SHA512 */
 BDESC (0, OPTION_MASK_ISA2_SHA512, CODE_FOR_vsha512msg1, "__builtin_ia32_vsha512msg1", IX86_BUILTIN_VSHA512MSG1, UNKNOWN, (int) V4DI_FTYPE_V4DI_V2DI)
 BDESC (0, OPTION_MASK_ISA2_SHA512, CODE_FOR_vsha512msg2, "__builtin_ia32_vsha512msg2", IX86_BUILTIN_VSHA512MSG2, UNKNOWN, (int) V4DI_FTYPE_V4DI_V4DI)
diff --git a/gcc/config/i386/i386-c.cc b/gcc/config/i386/i386-c.cc
index c6311f12cf9..0adec145600 100644
--- a/gcc/config/i386/i386-c.cc
+++ b/gcc/config/i386/i386-c.cc
@@ -683,6 +683,8 @@ ix86_target_macros_internal (HOST_WIDE_INT isa_flag,
     def_or_undef (parse_in, "__SM3__");
   if (isa_flag2 & OPTION_MASK_ISA2_SHA512)
     def_or_undef (parse_in, "__SHA512__");
+  if (isa_flag2 & OPTION_MASK_ISA2_SM4)
+    def_or_undef (parse_in, "__SM4__");
   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 28f221753a9..aeafcf870ac 100644
--- a/gcc/config/i386/i386-isa.def
+++ b/gcc/config/i386/i386-isa.def
@@ -120,3 +120,4 @@ DEF_PTA(AMX_COMPLEX)
 DEF_PTA(AVXVNNIINT16)
 DEF_PTA(SM3)
 DEF_PTA(SHA512)
+DEF_PTA(SM4)
diff --git a/gcc/config/i386/i386-options.cc b/gcc/config/i386/i386-options.cc
index d79ab01bd79..347ed2d210a 100644
--- a/gcc/config/i386/i386-options.cc
+++ b/gcc/config/i386/i386-options.cc
@@ -242,7 +242,8 @@ static struct ix86_target_opts isa2_opts[] =
   { "-mamx-complex",	OPTION_MASK_ISA2_AMX_COMPLEX },
   { "-mavxvnniint16",	OPTION_MASK_ISA2_AVXVNNIINT16 },
   { "-msm3",		OPTION_MASK_ISA2_SM3 },
-  { "-msha512",		OPTION_MASK_ISA2_SHA512 }
+  { "-msha512",		OPTION_MASK_ISA2_SHA512 },
+  { "-msm4",            OPTION_MASK_ISA2_SM4 }
 };
 static struct ix86_target_opts isa_opts[] =
 {
@@ -1097,6 +1098,7 @@ ix86_valid_target_attribute_inner_p (tree fndecl, tree args, char *p_strings[],
     IX86_ATTR_ISA ("avxvnniint16", OPT_mavxvnniint16),
     IX86_ATTR_ISA ("sm3", OPT_msm3),
     IX86_ATTR_ISA ("sha512", OPT_msha512),
+    IX86_ATTR_ISA ("sm4", OPT_msm4),
 
     /* enum options */
     IX86_ATTR_ENUM ("fpmath=",	OPT_mfpmath_),
diff --git a/gcc/config/i386/i386.opt b/gcc/config/i386/i386.opt
index cf9dbca58b3..db9956885e2 100644
--- a/gcc/config/i386/i386.opt
+++ b/gcc/config/i386/i386.opt
@@ -1298,3 +1298,8 @@ msha512
 Target Mask(ISA2_SHA512) Var(ix86_isa_flags2) Save
 Support MMX, SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2, AVX and
 SHA512 built-in functions and code generation.
+
+msm4
+Target Mask(ISA2_SM4) Var(ix86_isa_flags2) Save
+Support MMX, SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2, AVX and
+SM4 built-in functions and code generation.
diff --git a/gcc/config/i386/immintrin.h b/gcc/config/i386/immintrin.h
index 6f2bcef6a8a..ea14354efbc 100644
--- a/gcc/config/i386/immintrin.h
+++ b/gcc/config/i386/immintrin.h
@@ -112,6 +112,8 @@
 
 #include <sha512intrin.h>
 
+#include <sm4intrin.h>
+
 #include <fmaintrin.h>
 
 #include <f16cintrin.h>
diff --git a/gcc/config/i386/sm4intrin.h b/gcc/config/i386/sm4intrin.h
new file mode 100644
index 00000000000..f58a782bfdc
--- /dev/null
+++ b/gcc/config/i386/sm4intrin.h
@@ -0,0 +1,70 @@
+/* Copyright (C) 2023 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/>.  */
+
+#ifndef _IMMINTRIN_H_INCLUDED
+#error "Never use <sm4intrin.h> directly; include <immintrin.h> instead."
+#endif
+
+#ifndef _SM4INTRIN_H_INCLUDED
+#define _SM4INTRIN_H_INCLUDED
+
+#ifndef __SM4__
+#pragma GCC push_options
+#pragma GCC target("sm4")
+#define __DISABLE_SM4__
+#endif /* __SM4__ */
+
+extern __inline __m128i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sm4key4_epi32 (__m128i __A, __m128i __B)
+{
+  return (__m128i) __builtin_ia32_vsm4key4128 ((__v4si) __A, (__v4si) __B);
+}
+
+extern __inline __m256i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_sm4key4_epi32 (__m256i __A, __m256i __B)
+{
+  return (__m256i) __builtin_ia32_vsm4key4256 ((__v8si) __A, (__v8si) __B);
+}
+
+extern __inline __m128i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_sm4rnds4_epi32 (__m128i __A, __m128i __B)
+{
+  return (__m128i) __builtin_ia32_vsm4rnds4128 ((__v4si) __A, (__v4si) __B);
+}
+
+extern __inline __m256i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_sm4rnds4_epi32 (__m256i __A, __m256i __B)
+{
+  return (__m256i) __builtin_ia32_vsm4rnds4256 ((__v8si) __A, (__v8si) __B);
+}
+
+#ifdef __DISABLE_SM4__
+#undef __DISABLE_SM4__
+#pragma GCC pop_options
+#endif /* __DISABLE_SM4__ */
+
+#endif /* _SM4INTRIN_H_INCLUDED */
diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md
index e16b2b5a6c4..7471932b27e 100644
--- a/gcc/config/i386/sse.md
+++ b/gcc/config/i386/sse.md
@@ -223,6 +223,10 @@
   UNSPEC_SHA512MSG2
   UNSPEC_SHA512RNDS2
 
+  ;; For SM4 support
+  UNSPEC_SM4KEY4
+  UNSPEC_SM4RNDS4
+
 ])
 
 (define_c_enum "unspecv" [
@@ -28680,6 +28684,28 @@
   [(set_attr "type" "sselog1")
    (set_attr "mode" "OI")])
 
+(define_insn "vsm4key4_<mode>"
+  [(set (match_operand:VI4_AVX 0 "register_operand" "=x")
+        (unspec:VI4_AVX
+          [(match_operand:VI4_AVX 1 "register_operand" "x")
+           (match_operand:VI4_AVX 2 "vector_operand" "xBm")]
+          UNSPEC_SM4KEY4))]
+  "TARGET_SM4"
+  "vsm4key4\t{%2, %1, %0|%0, %1, %2}"
+  [(set_attr "type" "other")
+   (set_attr "mode" "<sseinsnmode>")])
+
+(define_insn "vsm4rnds4_<mode>"
+  [(set (match_operand:VI4_AVX 0 "register_operand" "=x")
+        (unspec:VI4_AVX
+          [(match_operand:VI4_AVX 1 "register_operand" "x")
+           (match_operand:VI4_AVX 2 "vector_operand" "xBm")]
+          UNSPEC_SM4RNDS4))]
+  "TARGET_SM4"
+  "vsm4rnds4\t{%2, %1, %0|%0, %1, %2}"
+  [(set_attr "type" "other")
+   (set_attr "mode" "<sseinsnmode>")])
+
 (define_insn_and_split "avx512f_<castmode><avxsizesuffix>_<castmode>"
   [(set (match_operand:AVX512MODE2P 0 "nonimmediate_operand" "=x,m")
 	(vec_concat:AVX512MODE2P
diff --git a/gcc/doc/extend.texi b/gcc/doc/extend.texi
index 5250990050b..8c83c58a30a 100644
--- a/gcc/doc/extend.texi
+++ b/gcc/doc/extend.texi
@@ -7178,6 +7178,11 @@ Enable/disable the generation of the SM3 instructions.
 @itemx no-sha512
 Enable/disable the generation of the SHA512 instructions.
 
+@cindex @code{target("sm4")} function attribute, x86
+@item sm4
+@itemx no-sm4
+Enable/disable the generation of the SM4 instructions.
+
 @cindex @code{target("cld")} function attribute, x86
 @item cld
 @itemx no-cld
diff --git a/gcc/doc/invoke.texi b/gcc/doc/invoke.texi
index 433ccf35505..dd28320185d 100644
--- a/gcc/doc/invoke.texi
+++ b/gcc/doc/invoke.texi
@@ -1433,7 +1433,7 @@ See RS/6000 and PowerPC Options.
 -mrdseed  -msgx -mavx512vp2intersect -mserialize -mtsxldtrk
 -mamx-tile  -mamx-int8  -mamx-bf16 -muintr -mhreset -mavxvnni
 -mavx512fp16 -mavxifma -mavxvnniint8 -mavxneconvert -mcmpccxadd -mamx-fp16
--mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3 -msha512
+-mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3 -msha512 -msm4
 -mcldemote  -mms-bitfields  -mno-align-stringops  -minline-all-stringops
 -minline-stringops-dynamically  -mstringop-strategy=@var{alg}
 -mkl -mwidekl
@@ -33561,6 +33561,9 @@ preferred alignment to @option{-mpreferred-stack-boundary=2}.
 @need 200
 @opindex msha512
 @itemx -msha512
+@need 200
+@opindex msm4
+@itemx -msm4
 These switches enable the use of instructions in the MMX, SSE,
 AVX512ER, AVX512CD, AVX512VL, AVX512BW, AVX512DQ, AVX512IFMA, AVX512VBMI, SHA,
 AES, PCLMUL, CLFLUSHOPT, CLWB, FSGSBASE, PTWRITE, RDRND, F16C, FMA, PCONFIG,
@@ -33571,8 +33574,8 @@ GFNI, VAES, WAITPKG, VPCLMULQDQ, AVX512BITALG, MOVDIRI, MOVDIR64B, AVX512BF16,
 ENQCMD, AVX512VPOPCNTDQ, AVX5124FMAPS, AVX512VNNI, AVX5124VNNIW, SERIALIZE,
 UINTR, HRESET, AMXTILE, AMXINT8, AMXBF16, KL, WIDEKL, AVXVNNI, AVX512-FP16,
 AVXIFMA, AVXVNNIINT8, AVXNECONVERT, CMPCCXADD, AMX-FP16, PREFETCHI, RAOINT,
-AMX-COMPLEX, AVXVNNIINT16, SM3, SHA512 or CLDEMOTE extended instruction sets.
-Each has a corresponding @option{-mno-} option to disable use of these
+AMX-COMPLEX, AVXVNNIINT16, SM3, SHA512, SM4 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
diff --git a/gcc/doc/sourcebuild.texi b/gcc/doc/sourcebuild.texi
index 54a062db3fe..e5d15d67253 100644
--- a/gcc/doc/sourcebuild.texi
+++ b/gcc/doc/sourcebuild.texi
@@ -2595,6 +2595,9 @@ Target supports the execution of @code{sha512} instructions.
 @item sm3
 Target supports the execution of @code{sm3} instructions.
 
+@item sm4
+Target supports the execution of @code{sm4} instructions.
+
 @item sqrt_insn
 Target has a square root instruction that the compiler can generate.
 
diff --git a/gcc/testsuite/g++.dg/other/i386-2.C b/gcc/testsuite/g++.dg/other/i386-2.C
index 985f14abcbc..7d68967488d 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 -mavxvnniint8 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3 -msha512" } */
+/* { 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 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3 -msha512 -msm4" } */
 
 /* 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 274b0e6256f..9b775c33ab4 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 -mavxvnniint8 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3 -msha512" } */
+/* { 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 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3 -msha512 -msm4" } */
 
 /* 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/funcspec-56.inc b/gcc/testsuite/gcc.target/i386/funcspec-56.inc
index eb9309819ab..577bfc75edf 100644
--- a/gcc/testsuite/gcc.target/i386/funcspec-56.inc
+++ b/gcc/testsuite/gcc.target/i386/funcspec-56.inc
@@ -91,6 +91,7 @@ extern void test_amx_complex (void)		__attribute__((__target__("amx-complex")));
 extern void test_avxvnniint16 (void)		__attribute__((__target__("avxvnniint16")));
 extern void test_sm3 (void)			__attribute__((__target__("sm3")));
 extern void test_sha512 (void)			__attribute__((__target__("sha512")));
+extern void test_sm4 (void)                     __attribute__((__target__("sm4")));
 
 extern void test_no_sgx (void)			__attribute__((__target__("no-sgx")));
 extern void test_no_avx5124fmaps(void)		__attribute__((__target__("no-avx5124fmaps")));
@@ -183,6 +184,7 @@ extern void test_no_amx_complex (void)		__attribute__((__target__("no-amx-comple
 extern void test_no_avxvnniint16 (void)		__attribute__((__target__("no-avxvnniint16")));
 extern void test_no_sm3 (void)			__attribute__((__target__("no-sm3")));
 extern void test_no_sha512 (void)		__attribute__((__target__("no-sha512")));
+extern void test_no_sm4 (void)                  __attribute__((__target__("no-sm4")));
 
 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/sm4-1.c b/gcc/testsuite/gcc.target/i386/sm4-1.c
new file mode 100644
index 00000000000..2d3d6c6aab8
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sm4-1.c
@@ -0,0 +1,20 @@
+/* { dg-do compile } */
+/* { dg-options "-O2 -msm4" } */
+/* { dg-final { scan-assembler "vsm4key4\[ \\t\]+\[^\n\]*%xmm\[0-9\]+\[^\n\]*%xmm\[0-9\]+\[^\n\]*%xmm\[0-9\]" } } */
+/* { dg-final { scan-assembler "vsm4key4\[ \\t\]+\[^\n\]*%ymm\[0-9\]+\[^\n\]*%ymm\[0-9\]+\[^\n\]*%ymm\[0-9\]" } } */
+/* { dg-final { scan-assembler "vsm4rnds4\[ \\t\]+\[^\n\]*%xmm\[0-9\]+\[^\n\]*%xmm\[0-9\]+\[^\n\]*%xmm\[0-9\]" } } */
+/* { dg-final { scan-assembler "vsm4rnds4\[ \\t\]+\[^\n\]*%ymm\[0-9\]+\[^\n\]*%ymm\[0-9\]+\[^\n\]*%ymm\[0-9\]" } } */
+
+#include <immintrin.h>
+
+volatile __m128i a, b, c;
+volatile __m256i d, e, f;
+
+void extern
+sm4_test (void)
+{
+  a = _mm_sm4key4_epi32 (b, c);
+  d = _mm256_sm4key4_epi32 (e, f);
+  a = _mm_sm4rnds4_epi32 (b, c);
+  d = _mm256_sm4rnds4_epi32 (e, f);
+}
diff --git a/gcc/testsuite/gcc.target/i386/sm4-check.h b/gcc/testsuite/gcc.target/i386/sm4-check.h
new file mode 100644
index 00000000000..435fcf2b17d
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sm4-check.h
@@ -0,0 +1,183 @@
+#include <stdlib.h>
+#include "m256-check.h"
+
+static void sm4_test (void);
+
+typedef union
+{
+  unsigned int x;
+  unsigned char a[4];
+} union32ui_ub;
+
+unsigned char sbox[256] = {
+0xD6, 0x90, 0xE9, 0xFE, 0xCC, 0xE1, 0x3D, 0xB7,
+0x16, 0xB6, 0x14, 0xC2, 0x28, 0xFB, 0x2C, 0x05,
+0x2B, 0x67, 0x9A, 0x76, 0x2A, 0xBE, 0x04, 0xC3,
+0xAA, 0x44, 0x13, 0x26, 0x49, 0x86, 0x06, 0x99,
+0x9C, 0x42, 0x50, 0xF4, 0x91, 0xEF, 0x98, 0x7A,
+0x33, 0x54, 0x0B, 0x43, 0xED, 0xCF, 0xAC, 0x62,
+0xE4, 0xB3, 0x1C, 0xA9, 0xC9, 0x08, 0xE8, 0x95,
+0x80, 0xDF, 0x94, 0xFA, 0x75, 0x8F, 0x3F, 0xA6,
+0x47, 0x07, 0xA7, 0xFC, 0xF3, 0x73, 0x17, 0xBA,
+0x83, 0x59, 0x3C, 0x19, 0xE6, 0x85, 0x4F, 0xA8,
+0x68, 0x6B, 0x81, 0xB2, 0x71, 0x64, 0xDA, 0x8B,
+0xF8, 0xEB, 0x0F, 0x4B, 0x70, 0x56, 0x9D, 0x35,
+0x1E, 0x24, 0x0E, 0x5E, 0x63, 0x58, 0xD1, 0xA2,
+0x25, 0x22, 0x7C, 0x3B, 0x01, 0x21, 0x78, 0x87,
+0xD4, 0x00, 0x46, 0x57, 0x9F, 0xD3, 0x27, 0x52,
+0x4C, 0x36, 0x02, 0xE7, 0xA0, 0xC4, 0xC8, 0x9E,
+0xEA, 0xBF, 0x8A, 0xD2, 0x40, 0xC7, 0x38, 0xB5,
+0xA3, 0xF7, 0xF2, 0xCE, 0xF9, 0x61, 0x15, 0xA1,
+0xE0, 0xAE, 0x5D, 0xA4, 0x9B, 0x34, 0x1A, 0x55,
+0xAD, 0x93, 0x32, 0x30, 0xF5, 0x8C, 0xB1, 0xE3,
+0x1D, 0xF6, 0xE2, 0x2E, 0x82, 0x66, 0xCA, 0x60,
+0xC0, 0x29, 0x23, 0xAB, 0x0D, 0x53, 0x4E, 0x6F,
+0xD5, 0xDB, 0x37, 0x45, 0xDE, 0xFD, 0x8E, 0x2F,
+0x03, 0xFF, 0x6A, 0x72, 0x6D, 0x6C, 0x5B, 0x51,
+0x8D, 0x1B, 0xAF, 0x92, 0xBB, 0xDD, 0xBC, 0x7F,
+0x11, 0xD9, 0x5C, 0x41, 0x1F, 0x10, 0x5A, 0xD8,
+0x0A, 0xC1, 0x31, 0x88, 0xA5, 0xCD, 0x7B, 0xBD,
+0x2D, 0x74, 0xD0, 0x12, 0xB8, 0xE5, 0xB4, 0xB0,
+0x89, 0x69, 0x97, 0x4A, 0x0C, 0x96, 0x77, 0x7E,
+0x65, 0xB9, 0xF1, 0x09, 0xC5, 0x6E, 0xC6, 0x84,
+0x18, 0xF0, 0x7D, 0xEC, 0x3A, 0xDC, 0x4D, 0x20,
+0x79, 0xEE, 0x5F, 0x3E, 0xD7, 0xCB, 0x39, 0x48
+};
+
+static unsigned
+rol32 (unsigned w, int n)
+{
+  int count = n % 32;
+  return ((w << count) | (w >> (32 - count)));
+}
+
+static unsigned char
+sbox_byte (unsigned w, int i)
+{
+  union32ui_ub tmp;
+  tmp.x = w;
+  return sbox[tmp.a[i]];
+}
+
+static unsigned
+lower_t (unsigned w)
+{
+  union32ui_ub tmp;
+  tmp.a[0] = sbox_byte (w, 0);
+  tmp.a[1] = sbox_byte (w, 1);
+  tmp.a[2] = sbox_byte (w, 2);
+  tmp.a[3] = sbox_byte (w, 3);
+  return tmp.x;
+}
+
+static unsigned
+l_key (unsigned w)
+{
+  return w ^ rol32 (w, 13) ^ rol32 (w, 23);
+}
+
+static unsigned
+l_rnds (unsigned w)
+{
+  unsigned tmp = w;
+  tmp = tmp ^ rol32 (w, 2);
+  tmp = tmp ^ rol32 (w, 10);
+  tmp = tmp ^ rol32 (w, 18);
+  tmp = tmp ^ rol32 (w, 24);
+  return tmp;
+}
+
+#define SM4_FUNC(name)					      \
+static unsigned						      \
+t_##name (unsigned w)					      \
+{							      \
+  return l_##name (lower_t (w));			      \
+}							      \
+							      \
+static unsigned						      \
+f_##name (unsigned x0, unsigned x1, unsigned x2, unsigned x3, unsigned k) \
+{							      \
+  return x0 ^ t_##name (x1 ^ x2 ^ x3 ^ k);		      \
+}							      \
+							      \
+static void						      \
+compute_sm4##name##4 (int *dst, int *src1, int *src2, int vl) \
+{							      \
+  unsigned c[4], p[4];					      \
+							      \
+  int kl = vl / 128;					      \
+  int i;						      \
+							      \
+  for (i = 0; i < kl; i++)				      \
+  {							      \
+    p[0] = src1[4 * i];					      \
+    p[1] = src1[4 * i + 1];				      \
+    p[2] = src1[4 * i + 2];				      \
+    p[3] = src1[4 * i + 3];				      \
+							      \
+    c[0] = f_##name (p[0], p[1], p[2], p[3], src2[4 * i]);    \
+    c[1] = f_##name (p[1], p[2], p[3], c[0], src2[4 * i + 1]);	\
+    c[2] = f_##name (p[2], p[3], c[0], c[1], src2[4 * i + 2]);	\
+    c[3] = f_##name (p[3], c[0], c[1], c[2], src2[4 * i + 3]);	\
+							      \
+    dst[4 * i] = c[0];					      \
+    dst[4 * i + 1] = c[1];				      \
+    dst[4 * i + 2] = c[2];				      \
+    dst[4 * i + 3] = c[3];				      \
+  }							      \
+}
+
+#define SM4_AVX_SIMULATE(name)				      \
+  union128i_d src1, src2, res1;				      \
+  int dst1[4] = {0, 0, 0, 0};				      \
+							      \
+  src1.x = _mm_set_epi32 (111, 222, 333, 444);		      \
+  src2.x = _mm_set_epi32 (555, 666, 777, 888);		      \
+  res1.x = _mm_set_epi32 (0, 0, 0, 0);			      \
+							      \
+  res1.x = _mm_sm4##name##4_epi32 (src1.x, src2.x);	      \
+							      \
+  compute_sm4##name##4 (dst1, src1.a, src2.a, 128);	      \
+							      \
+  if (check_union128i_d (res1, dst1))			      \
+    abort ();						      \
+							      \
+  union256i_d src3, src4, res2;				      \
+  int dst2[8] = {0, 0, 0, 0, 0, 0, 0, 0};		      \
+							      \
+  src3.x = _mm256_set_epi32 (111, 222, 333, 444, 555, 666, 777, 888); \
+  src4.x = _mm256_set_epi32 (999, 123, 456, 789, 135, 792, 468, 147); \
+  res2.x = _mm256_set_epi32 (0, 0, 0, 0, 0, 0, 0, 0);	      \
+							      \
+  res2.x = _mm256_sm4##name##4_epi32 (src3.x, src4.x);	      \
+							      \
+  compute_sm4##name##4 (dst2, src3.a, src4.a, 256);	      \
+							      \
+  if (check_union256i_d (res2, dst2))			      \
+    abort ();
+
+static void
+__attribute__ ((noinline))
+do_test (void)
+{
+  sm4_test ();
+}
+
+int
+main ()
+{
+  /* Check CPU support for SM4.  */
+  if (__builtin_cpu_supports ("sm4"))
+    {
+      do_test ();
+#ifdef DEBUG
+      printf ("PASSED\n");
+#endif
+      return 0;
+    }
+
+#ifdef DEBUG
+  printf ("SKIPPED\n");
+#endif
+  return 0;
+}
diff --git a/gcc/testsuite/gcc.target/i386/sm4key4-2.c b/gcc/testsuite/gcc.target/i386/sm4key4-2.c
new file mode 100644
index 00000000000..a8bec560214
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sm4key4-2.c
@@ -0,0 +1,14 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -msm4" } */
+/* { dg-require-effective-target sm4 } */
+
+#include "sm4-check.h"
+
+char key;
+SM4_FUNC (key);
+
+static void
+sm4_test (void)
+{
+  SM4_AVX_SIMULATE (key);
+}
diff --git a/gcc/testsuite/gcc.target/i386/sm4rnds4-2.c b/gcc/testsuite/gcc.target/i386/sm4rnds4-2.c
new file mode 100644
index 00000000000..0860d0dd412
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/sm4rnds4-2.c
@@ -0,0 +1,14 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -msm4" } */
+/* { dg-require-effective-target sm4 } */
+
+#include "sm4-check.h"
+
+char rnds;
+SM4_FUNC (rnds);
+
+static void
+sm4_test (void)
+{
+  SM4_AVX_SIMULATE (rnds);
+}
diff --git a/gcc/testsuite/gcc.target/i386/sse-12.c b/gcc/testsuite/gcc.target/i386/sse-12.c
index 976541389ea..a553a5202d1 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 -mavxvnniint8 -mavxneconvert -mamx-fp16 -mraoint -mamx-complex -mavxvnniint16 -msm3 -msha512" } */
+/* { 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 -mavxneconvert -mamx-fp16 -mraoint -mamx-complex -mavxvnniint16 -msm3 -msha512 -msm4" } */
 
 #include <x86intrin.h>
 
diff --git a/gcc/testsuite/gcc.target/i386/sse-13.c b/gcc/testsuite/gcc.target/i386/sse-13.c
index 8c314e70e31..946182f0e76 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 -mavxvnniint8 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3 -msha512" } */
+/* { 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 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3 -msha512 -msm4" } */
 /* { 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 2b4d7bc9079..0d07aadc7f8 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 -mavxvnniint8 -mavxneconvert -mamx-fp16 -mraoint -mamx-complex -mavxvnniint16 -msm3 -msha512" } */
+/* { 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 -mavxneconvert -mamx-fp16 -mraoint -mamx-complex -mavxvnniint16 -msm3 -msha512 -msm4" } */
 /* { 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 d6f19b5e20a..e6681f7dd12 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,tbm,lwp,fsgsbase,rdrnd,f16c,rtm,rdseed,prfchw,adx,fxsr,xsaveopt,avx512f,avx512er,avx512cd,avx512pf,sha,prefetchwt1,avx512vl,avx512bw,avx512dq,avx512vbmi,avx512vbmi2,avx512ifma,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,gfni,avx512bitalg,avx512bf16,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,amx-fp16,raoint,amx-complex,avxvnniint16,sm3,sha512")
+#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,avx512f,avx512er,avx512cd,avx512pf,sha,prefetchwt1,avx512vl,avx512bw,avx512dq,avx512vbmi,avx512vbmi2,avx512ifma,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,gfni,avx512bitalg,avx512bf16,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,amx-fp16,raoint,amx-complex,avxvnniint16,sm3,sha512,sm4")
 #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,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,gfni,avx512bitalg,avx512bf16,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,amx-fp16,raoint,amx-complex,avxvnniint16,sm3,sha512")
+#pragma GCC target ("avx,avx2,rdrnd,fsgsbase,f16c,rtm,avx512f,avx512er,avx512cd,avx512pf,sha,avx512vl,avx512bw,avx512dq,avx512ifma,avx512vbmi,avx512vbmi2,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,gfni,avx512bitalg,avx512bf16,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,amx-fp16,raoint,amx-complex,avxvnniint16,sm3,sha512,sm4")
 #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 1df66b525ca..92b1c467d95 100644
--- a/gcc/testsuite/gcc.target/i386/sse-23.c
+++ b/gcc/testsuite/gcc.target/i386/sse-23.c
@@ -850,6 +850,6 @@
 /* sm3intrin.h */
 #define __builtin_ia32_vsm3rnds2(A, B, C, D) __builtin_ia32_vsm3rnds2 (A, B, C, 1)
 
-#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,avx512f,avx512er,avx512cd,avx512pf,sha,prefetchwt1,xsavec,xsaves,clflushopt,avx512bw,avx512dq,avx512vl,avx512vbmi,avx512ifma,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,clwb,mwaitx,clzero,pku,sgx,rdpid,gfni,avx512vbmi2,vpclmulqdq,avx512bitalg,pconfig,wbnoinvd,avx512bf16,enqcmd,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,cmpccxadd,amx-fp16,prefetchi,raoint,amx-complex,avxvnniint16,sm3,sha512")
+#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,avx512f,avx512er,avx512cd,avx512pf,sha,prefetchwt1,xsavec,xsaves,clflushopt,avx512bw,avx512dq,avx512vl,avx512vbmi,avx512ifma,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,clwb,mwaitx,clzero,pku,sgx,rdpid,gfni,avx512vbmi2,vpclmulqdq,avx512bitalg,pconfig,wbnoinvd,avx512bf16,enqcmd,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,cmpccxadd,amx-fp16,prefetchi,raoint,amx-complex,avxvnniint16,sm3,sha512,sm4")
 
 #include <x86intrin.h>
diff --git a/gcc/testsuite/lib/target-supports.exp b/gcc/testsuite/lib/target-supports.exp
index f376d835f8b..8ea0d9feb1c 100644
--- a/gcc/testsuite/lib/target-supports.exp
+++ b/gcc/testsuite/lib/target-supports.exp
@@ -9897,6 +9897,20 @@ proc check_effective_target_sha512 { } {
     } "-msha512" ]
 }
 
+# Return 1 if sm4 instructions can be compiled.
+proc check_effective_target_sm4 { } {
+    return [check_no_compiler_messages sm4 object {
+        typedef long long __m128i __attribute__ ((__vector_size__ (16)));
+        typedef int __v4si __attribute__ ((__vector_size__ (16)));
+        __m128i
+        _mm_sm4key4_epi32 (__m128i __A, __m128i __B)
+        {
+          return (__m128i) __builtin_ia32_vsm4key4128 ((__v4si) __A,
+						       (__v4si) __B);
+        }
+    } "-msm4" ]
+}
+
 # Return 1 if sse instructions can be compiled.
 proc check_effective_target_sse { } {
     return [check_no_compiler_messages sse object {
-- 
2.31.1


^ permalink raw reply	[flat|nested] 9+ messages in thread

* Re: [PATCH 4/4] Support Intel SM4
  2023-07-13  6:03 ` [PATCH 4/4] Support Intel SM4 Haochen Jiang
@ 2023-07-17  1:40   ` Hongtao Liu
  0 siblings, 0 replies; 9+ messages in thread
From: Hongtao Liu @ 2023-07-17  1:40 UTC (permalink / raw)
  To: Haochen Jiang; +Cc: gcc-patches, hongtao.liu, ubizjak

On Thu, Jul 13, 2023 at 2:04 PM Haochen Jiang via Gcc-patches
<gcc-patches@gcc.gnu.org> wrote:
>
> gcc/ChangeLog:
>
>         * common/config/i386/cpuinfo.h (get_available_features):
>         Detech SM4.
>         * common/config/i386/i386-common.cc (OPTION_MASK_ISA2_SM4_SET,
>         OPTION_MASK_ISA2_SM4_UNSET): New.
>         (OPTION_MASK_ISA2_AVX_UNSET): Add SM4.
>         (ix86_handle_option): Handle -msm4.
>         * common/config/i386/i386-cpuinfo.h (enum processor_features):
>         Add FEATURE_SM4.
>         * common/config/i386/i386-isas.h: Add ISA_NAME_TABLE_ENTRY for
>         sm4.
>         * config.gcc: Add sm4intrin.h.
>         * config/i386/cpuid.h (bit_SM4): New.
>         * config/i386/i386-builtin.def (BDESC): Add new builtins.
>         * config/i386/i386-c.cc (ix86_target_macros_internal): Define
>         __SM4__.
>         * config/i386/i386-isa.def (SM4): Add DEF_PTA(SM4).
>         * config/i386/i386-options.cc (isa2_opts): Add -msm4.
>         (ix86_valid_target_attribute_inner_p): Handle sm4.
>         * config/i386/i386.opt: Add option -msm4.
>         * config/i386/immintrin.h: Include sm4intrin.h
>         * config/i386/sse.md (vsm4key4_<mode>): New define insn.
>         (vsm4rnds4_<mode>): Ditto.
>         * doc/extend.texi: Document sm4.
>         * doc/invoke.texi: Document -msm4.
>         * doc/sourcebuild.texi: Document target sm4.
>         * config/i386/sm4intrin.h: New file.
>
> gcc/testsuite/ChangeLog:
>
>         * g++.dg/other/i386-2.C: Add -msm4.
>         * g++.dg/other/i386-3.C: Ditto.
>         * gcc.target/i386/funcspec-56.inc: Add new target attribute.
>         * gcc.target/i386/sse-12.c: Add -msm4.
>         * gcc.target/i386/sse-13.c: Ditto.
>         * gcc.target/i386/sse-14.c: Ditto.
>         * gcc.target/i386/sse-22.c: Add sm4.
>         * gcc.target/i386/sse-23.c: Ditto.
>         * lib/target-supports.exp (check_effective_target_sm4): New.
>         * gcc.target/i386/sm4-1.c: New test.
>         * gcc.target/i386/sm4-check.h: Ditto.
>         * gcc.target/i386/sm4key4-2.c: Ditto.
>         * gcc.target/i386/sm4rnds4-2.c: Ditto.
Ok.
> ---
>  gcc/common/config/i386/cpuinfo.h              |   2 +
>  gcc/common/config/i386/i386-common.cc         |  20 +-
>  gcc/common/config/i386/i386-cpuinfo.h         |   1 +
>  gcc/common/config/i386/i386-isas.h            |   1 +
>  gcc/config.gcc                                |   2 +-
>  gcc/config/i386/cpuid.h                       |   1 +
>  gcc/config/i386/i386-builtin.def              |   6 +
>  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/sm4intrin.h                   |  70 +++++++
>  gcc/config/i386/sse.md                        |  26 +++
>  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/funcspec-56.inc |   2 +
>  gcc/testsuite/gcc.target/i386/sm4-1.c         |  20 ++
>  gcc/testsuite/gcc.target/i386/sm4-check.h     | 183 ++++++++++++++++++
>  gcc/testsuite/gcc.target/i386/sm4key4-2.c     |  14 ++
>  gcc/testsuite/gcc.target/i386/sm4rnds4-2.c    |  14 ++
>  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         |  14 ++
>  30 files changed, 409 insertions(+), 14 deletions(-)
>  create mode 100644 gcc/config/i386/sm4intrin.h
>  create mode 100644 gcc/testsuite/gcc.target/i386/sm4-1.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sm4-check.h
>  create mode 100644 gcc/testsuite/gcc.target/i386/sm4key4-2.c
>  create mode 100644 gcc/testsuite/gcc.target/i386/sm4rnds4-2.c
>
> diff --git a/gcc/common/config/i386/cpuinfo.h b/gcc/common/config/i386/cpuinfo.h
> index 0cfde3ebccd..f9434f038ea 100644
> --- a/gcc/common/config/i386/cpuinfo.h
> +++ b/gcc/common/config/i386/cpuinfo.h
> @@ -881,6 +881,8 @@ get_available_features (struct __processor_model *cpu_model,
>             set_feature (FEATURE_SM3);
>           if (eax & bit_SHA512)
>             set_feature (FEATURE_SHA512);
> +         if (eax & bit_SM4)
> +           set_feature (FEATURE_SM4);
>         }
>        if (avx512_usable)
>         {
> diff --git a/gcc/common/config/i386/i386-common.cc b/gcc/common/config/i386/i386-common.cc
> index 97c3cdfe5e1..610cabe52c1 100644
> --- a/gcc/common/config/i386/i386-common.cc
> +++ b/gcc/common/config/i386/i386-common.cc
> @@ -122,6 +122,7 @@ along with GCC; see the file COPYING3.  If not see
>  #define OPTION_MASK_ISA2_AVXVNNIINT16_SET OPTION_MASK_ISA2_AVXVNNIINT16
>  #define OPTION_MASK_ISA2_SM3_SET OPTION_MASK_ISA2_SM3
>  #define OPTION_MASK_ISA2_SHA512_SET OPTION_MASK_ISA2_SHA512
> +#define OPTION_MASK_ISA2_SM4_SET OPTION_MASK_ISA2_SM4
>
>  /* SSE4 includes both SSE4.1 and SSE4.2. -msse4 should be the same
>     as -msse4.2.  */
> @@ -307,6 +308,7 @@ along with GCC; see the file COPYING3.  If not see
>  #define OPTION_MASK_ISA2_AVXVNNIINT16_UNSET OPTION_MASK_ISA2_AVXVNNIINT16
>  #define OPTION_MASK_ISA2_SM3_UNSET OPTION_MASK_ISA2_SM3
>  #define OPTION_MASK_ISA2_SHA512_UNSET OPTION_MASK_ISA2_SHA512
> +#define OPTION_MASK_ISA2_SM4_UNSET OPTION_MASK_ISA2_SM4
>
>  /* SSE4 includes both SSE4.1 and SSE4.2.  -mno-sse4 should the same
>     as -mno-sse4.1. */
> @@ -356,7 +358,8 @@ along with GCC; see the file COPYING3.  If not see
>    OPTION_MASK_ISA2_SSE_UNSET
>  #define OPTION_MASK_ISA2_AVX_UNSET \
>    (OPTION_MASK_ISA2_AVX2_UNSET | OPTION_MASK_ISA2_VAES_UNSET \
> -   | OPTION_MASK_ISA2_SM3_UNSET | OPTION_MASK_ISA2_SHA512_UNSET)
> +   | OPTION_MASK_ISA2_SM3_UNSET | OPTION_MASK_ISA2_SHA512_UNSET \
> +   | OPTION_MASK_ISA2_SM4_UNSET)
>  #define OPTION_MASK_ISA2_SSE4_2_UNSET OPTION_MASK_ISA2_AVX_UNSET
>  #define OPTION_MASK_ISA2_SSE4_1_UNSET OPTION_MASK_ISA2_SSE4_2_UNSET
>  #define OPTION_MASK_ISA2_SSE4_UNSET OPTION_MASK_ISA2_SSE4_1_UNSET
> @@ -1323,6 +1326,21 @@ ix86_handle_option (struct gcc_options *opts,
>         }
>        return true;
>
> +    case OPT_msm4:
> +      if (value)
> +       {
> +         opts->x_ix86_isa_flags2 |= OPTION_MASK_ISA2_SM4_SET;
> +         opts->x_ix86_isa_flags2_explicit |= OPTION_MASK_ISA2_SM4_SET;
> +         opts->x_ix86_isa_flags |= OPTION_MASK_ISA_AVX_SET;
> +         opts->x_ix86_isa_flags_explicit |= OPTION_MASK_ISA_AVX_SET;
> +       }
> +      else
> +       {
> +         opts->x_ix86_isa_flags2 &= ~OPTION_MASK_ISA2_SM4_UNSET;
> +         opts->x_ix86_isa_flags2_explicit |= OPTION_MASK_ISA2_SM4_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 a6e34d14f8e..be04d85c9d5 100644
> --- a/gcc/common/config/i386/i386-cpuinfo.h
> +++ b/gcc/common/config/i386/i386-cpuinfo.h
> @@ -258,6 +258,7 @@ enum processor_features
>    FEATURE_AVXVNNIINT16,
>    FEATURE_SM3,
>    FEATURE_SHA512,
> +  FEATURE_SM4,
>    CPU_FEATURE_MAX
>  };
>
> diff --git a/gcc/common/config/i386/i386-isas.h b/gcc/common/config/i386/i386-isas.h
> index 250dc87764f..2297903a45e 100644
> --- a/gcc/common/config/i386/i386-isas.h
> +++ b/gcc/common/config/i386/i386-isas.h
> @@ -190,4 +190,5 @@ ISA_NAMES_TABLE_START
>                         P_NONE, "-mavxvnniint16")
>    ISA_NAMES_TABLE_ENTRY("sm3", FEATURE_SM3, P_NONE, "-msm3")
>    ISA_NAMES_TABLE_ENTRY("sha512", FEATURE_SHA512, P_NONE, "-msha512")
> +  ISA_NAMES_TABLE_ENTRY("sm4", FEATURE_SM4, P_NONE, "-msm4")
>  ISA_NAMES_TABLE_END
> diff --git a/gcc/config.gcc b/gcc/config.gcc
> index 4e753ba7c64..305e859880f 100644
> --- a/gcc/config.gcc
> +++ b/gcc/config.gcc
> @@ -436,7 +436,7 @@ i[34567]86-*-* | x86_64-*-*)
>                        avxifmaintrin.h avxvnniint8intrin.h avxneconvertintrin.h
>                        cmpccxaddintrin.h amxfp16intrin.h prfchiintrin.h
>                        raointintrin.h amxcomplexintrin.h avxvnniint16intrin.h
> -                      sm3intrin.h sha512intrin.h"
> +                      sm3intrin.h sha512intrin.h sm4intrin.h"
>         ;;
>  ia64-*-*)
>         extra_headers=ia64intrin.h
> diff --git a/gcc/config/i386/cpuid.h b/gcc/config/i386/cpuid.h
> index f9103f1b1c9..03fd6fc9478 100644
> --- a/gcc/config/i386/cpuid.h
> +++ b/gcc/config/i386/cpuid.h
> @@ -134,6 +134,7 @@
>  /* %eax */
>  #define bit_SHA512     (1 << 0)
>  #define bit_SM3                (1 << 1)
> +#define bit_SM4                (1 << 2)
>  #define bit_RAOINT      (1 << 3)
>  #define bit_AVXVNNI     (1 << 4)
>  #define bit_AVX512BF16  (1 << 5)
> diff --git a/gcc/config/i386/i386-builtin.def b/gcc/config/i386/i386-builtin.def
> index b9e2bad8522..8738b3b6a8a 100644
> --- a/gcc/config/i386/i386-builtin.def
> +++ b/gcc/config/i386/i386-builtin.def
> @@ -1660,6 +1660,12 @@ BDESC (OPTION_MASK_ISA_AVX, OPTION_MASK_ISA2_SM3, CODE_FOR_vsm3msg1, "__builtin_
>  BDESC (OPTION_MASK_ISA_AVX, OPTION_MASK_ISA2_SM3, CODE_FOR_vsm3msg2, "__builtin_ia32_vsm3msg2", IX86_BUILTIN_VSM3MSG2, UNKNOWN, (int) V4SI_FTYPE_V4SI_V4SI_V4SI)
>  BDESC (OPTION_MASK_ISA_AVX, OPTION_MASK_ISA2_SM3, CODE_FOR_vsm3rnds2, "__builtin_ia32_vsm3rnds2", IX86_BUILTIN_VSM3RNDS2, UNKNOWN, (int) V4SI_FTYPE_V4SI_V4SI_V4SI_INT)
>
> +/* SM4 */
> +BDESC (0, OPTION_MASK_ISA2_SM4, CODE_FOR_vsm4key4_v4si, "__builtin_ia32_vsm4key4128", IX86_BUILTIN_VSM4KEY4128, UNKNOWN, (int) V4SI_FTYPE_V4SI_V4SI)
> +BDESC (0, OPTION_MASK_ISA2_SM4, CODE_FOR_vsm4key4_v8si, "__builtin_ia32_vsm4key4256", IX86_BUILTIN_VSM4KEY4256, UNKNOWN, (int) V8SI_FTYPE_V8SI_V8SI)
> +BDESC (0, OPTION_MASK_ISA2_SM4, CODE_FOR_vsm4rnds4_v4si, "__builtin_ia32_vsm4rnds4128", IX86_BUILTIN_VSM4RNDS4128, UNKNOWN, (int) V4SI_FTYPE_V4SI_V4SI)
> +BDESC (0, OPTION_MASK_ISA2_SM4, CODE_FOR_vsm4rnds4_v8si, "__builtin_ia32_vsm4rnds4256", IX86_BUILTIN_VSM4RNDS4256, UNKNOWN, (int) V8SI_FTYPE_V8SI_V8SI)
> +
>  /* SHA512 */
>  BDESC (0, OPTION_MASK_ISA2_SHA512, CODE_FOR_vsha512msg1, "__builtin_ia32_vsha512msg1", IX86_BUILTIN_VSHA512MSG1, UNKNOWN, (int) V4DI_FTYPE_V4DI_V2DI)
>  BDESC (0, OPTION_MASK_ISA2_SHA512, CODE_FOR_vsha512msg2, "__builtin_ia32_vsha512msg2", IX86_BUILTIN_VSHA512MSG2, UNKNOWN, (int) V4DI_FTYPE_V4DI_V4DI)
> diff --git a/gcc/config/i386/i386-c.cc b/gcc/config/i386/i386-c.cc
> index c6311f12cf9..0adec145600 100644
> --- a/gcc/config/i386/i386-c.cc
> +++ b/gcc/config/i386/i386-c.cc
> @@ -683,6 +683,8 @@ ix86_target_macros_internal (HOST_WIDE_INT isa_flag,
>      def_or_undef (parse_in, "__SM3__");
>    if (isa_flag2 & OPTION_MASK_ISA2_SHA512)
>      def_or_undef (parse_in, "__SHA512__");
> +  if (isa_flag2 & OPTION_MASK_ISA2_SM4)
> +    def_or_undef (parse_in, "__SM4__");
>    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 28f221753a9..aeafcf870ac 100644
> --- a/gcc/config/i386/i386-isa.def
> +++ b/gcc/config/i386/i386-isa.def
> @@ -120,3 +120,4 @@ DEF_PTA(AMX_COMPLEX)
>  DEF_PTA(AVXVNNIINT16)
>  DEF_PTA(SM3)
>  DEF_PTA(SHA512)
> +DEF_PTA(SM4)
> diff --git a/gcc/config/i386/i386-options.cc b/gcc/config/i386/i386-options.cc
> index d79ab01bd79..347ed2d210a 100644
> --- a/gcc/config/i386/i386-options.cc
> +++ b/gcc/config/i386/i386-options.cc
> @@ -242,7 +242,8 @@ static struct ix86_target_opts isa2_opts[] =
>    { "-mamx-complex",   OPTION_MASK_ISA2_AMX_COMPLEX },
>    { "-mavxvnniint16",  OPTION_MASK_ISA2_AVXVNNIINT16 },
>    { "-msm3",           OPTION_MASK_ISA2_SM3 },
> -  { "-msha512",                OPTION_MASK_ISA2_SHA512 }
> +  { "-msha512",                OPTION_MASK_ISA2_SHA512 },
> +  { "-msm4",            OPTION_MASK_ISA2_SM4 }
>  };
>  static struct ix86_target_opts isa_opts[] =
>  {
> @@ -1097,6 +1098,7 @@ ix86_valid_target_attribute_inner_p (tree fndecl, tree args, char *p_strings[],
>      IX86_ATTR_ISA ("avxvnniint16", OPT_mavxvnniint16),
>      IX86_ATTR_ISA ("sm3", OPT_msm3),
>      IX86_ATTR_ISA ("sha512", OPT_msha512),
> +    IX86_ATTR_ISA ("sm4", OPT_msm4),
>
>      /* enum options */
>      IX86_ATTR_ENUM ("fpmath=", OPT_mfpmath_),
> diff --git a/gcc/config/i386/i386.opt b/gcc/config/i386/i386.opt
> index cf9dbca58b3..db9956885e2 100644
> --- a/gcc/config/i386/i386.opt
> +++ b/gcc/config/i386/i386.opt
> @@ -1298,3 +1298,8 @@ msha512
>  Target Mask(ISA2_SHA512) Var(ix86_isa_flags2) Save
>  Support MMX, SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2, AVX and
>  SHA512 built-in functions and code generation.
> +
> +msm4
> +Target Mask(ISA2_SM4) Var(ix86_isa_flags2) Save
> +Support MMX, SSE, SSE2, SSE3, SSSE3, SSE4.1, SSE4.2, AVX and
> +SM4 built-in functions and code generation.
> diff --git a/gcc/config/i386/immintrin.h b/gcc/config/i386/immintrin.h
> index 6f2bcef6a8a..ea14354efbc 100644
> --- a/gcc/config/i386/immintrin.h
> +++ b/gcc/config/i386/immintrin.h
> @@ -112,6 +112,8 @@
>
>  #include <sha512intrin.h>
>
> +#include <sm4intrin.h>
> +
>  #include <fmaintrin.h>
>
>  #include <f16cintrin.h>
> diff --git a/gcc/config/i386/sm4intrin.h b/gcc/config/i386/sm4intrin.h
> new file mode 100644
> index 00000000000..f58a782bfdc
> --- /dev/null
> +++ b/gcc/config/i386/sm4intrin.h
> @@ -0,0 +1,70 @@
> +/* Copyright (C) 2023 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/>.  */
> +
> +#ifndef _IMMINTRIN_H_INCLUDED
> +#error "Never use <sm4intrin.h> directly; include <immintrin.h> instead."
> +#endif
> +
> +#ifndef _SM4INTRIN_H_INCLUDED
> +#define _SM4INTRIN_H_INCLUDED
> +
> +#ifndef __SM4__
> +#pragma GCC push_options
> +#pragma GCC target("sm4")
> +#define __DISABLE_SM4__
> +#endif /* __SM4__ */
> +
> +extern __inline __m128i
> +__attribute__((__gnu_inline__, __always_inline__, __artificial__))
> +_mm_sm4key4_epi32 (__m128i __A, __m128i __B)
> +{
> +  return (__m128i) __builtin_ia32_vsm4key4128 ((__v4si) __A, (__v4si) __B);
> +}
> +
> +extern __inline __m256i
> +__attribute__((__gnu_inline__, __always_inline__, __artificial__))
> +_mm256_sm4key4_epi32 (__m256i __A, __m256i __B)
> +{
> +  return (__m256i) __builtin_ia32_vsm4key4256 ((__v8si) __A, (__v8si) __B);
> +}
> +
> +extern __inline __m128i
> +__attribute__((__gnu_inline__, __always_inline__, __artificial__))
> +_mm_sm4rnds4_epi32 (__m128i __A, __m128i __B)
> +{
> +  return (__m128i) __builtin_ia32_vsm4rnds4128 ((__v4si) __A, (__v4si) __B);
> +}
> +
> +extern __inline __m256i
> +__attribute__((__gnu_inline__, __always_inline__, __artificial__))
> +_mm256_sm4rnds4_epi32 (__m256i __A, __m256i __B)
> +{
> +  return (__m256i) __builtin_ia32_vsm4rnds4256 ((__v8si) __A, (__v8si) __B);
> +}
> +
> +#ifdef __DISABLE_SM4__
> +#undef __DISABLE_SM4__
> +#pragma GCC pop_options
> +#endif /* __DISABLE_SM4__ */
> +
> +#endif /* _SM4INTRIN_H_INCLUDED */
> diff --git a/gcc/config/i386/sse.md b/gcc/config/i386/sse.md
> index e16b2b5a6c4..7471932b27e 100644
> --- a/gcc/config/i386/sse.md
> +++ b/gcc/config/i386/sse.md
> @@ -223,6 +223,10 @@
>    UNSPEC_SHA512MSG2
>    UNSPEC_SHA512RNDS2
>
> +  ;; For SM4 support
> +  UNSPEC_SM4KEY4
> +  UNSPEC_SM4RNDS4
> +
>  ])
>
>  (define_c_enum "unspecv" [
> @@ -28680,6 +28684,28 @@
>    [(set_attr "type" "sselog1")
>     (set_attr "mode" "OI")])
>
> +(define_insn "vsm4key4_<mode>"
> +  [(set (match_operand:VI4_AVX 0 "register_operand" "=x")
> +        (unspec:VI4_AVX
> +          [(match_operand:VI4_AVX 1 "register_operand" "x")
> +           (match_operand:VI4_AVX 2 "vector_operand" "xBm")]
> +          UNSPEC_SM4KEY4))]
> +  "TARGET_SM4"
> +  "vsm4key4\t{%2, %1, %0|%0, %1, %2}"
> +  [(set_attr "type" "other")
> +   (set_attr "mode" "<sseinsnmode>")])
> +
> +(define_insn "vsm4rnds4_<mode>"
> +  [(set (match_operand:VI4_AVX 0 "register_operand" "=x")
> +        (unspec:VI4_AVX
> +          [(match_operand:VI4_AVX 1 "register_operand" "x")
> +           (match_operand:VI4_AVX 2 "vector_operand" "xBm")]
> +          UNSPEC_SM4RNDS4))]
> +  "TARGET_SM4"
> +  "vsm4rnds4\t{%2, %1, %0|%0, %1, %2}"
> +  [(set_attr "type" "other")
> +   (set_attr "mode" "<sseinsnmode>")])
> +
>  (define_insn_and_split "avx512f_<castmode><avxsizesuffix>_<castmode>"
>    [(set (match_operand:AVX512MODE2P 0 "nonimmediate_operand" "=x,m")
>         (vec_concat:AVX512MODE2P
> diff --git a/gcc/doc/extend.texi b/gcc/doc/extend.texi
> index 5250990050b..8c83c58a30a 100644
> --- a/gcc/doc/extend.texi
> +++ b/gcc/doc/extend.texi
> @@ -7178,6 +7178,11 @@ Enable/disable the generation of the SM3 instructions.
>  @itemx no-sha512
>  Enable/disable the generation of the SHA512 instructions.
>
> +@cindex @code{target("sm4")} function attribute, x86
> +@item sm4
> +@itemx no-sm4
> +Enable/disable the generation of the SM4 instructions.
> +
>  @cindex @code{target("cld")} function attribute, x86
>  @item cld
>  @itemx no-cld
> diff --git a/gcc/doc/invoke.texi b/gcc/doc/invoke.texi
> index 433ccf35505..dd28320185d 100644
> --- a/gcc/doc/invoke.texi
> +++ b/gcc/doc/invoke.texi
> @@ -1433,7 +1433,7 @@ See RS/6000 and PowerPC Options.
>  -mrdseed  -msgx -mavx512vp2intersect -mserialize -mtsxldtrk
>  -mamx-tile  -mamx-int8  -mamx-bf16 -muintr -mhreset -mavxvnni
>  -mavx512fp16 -mavxifma -mavxvnniint8 -mavxneconvert -mcmpccxadd -mamx-fp16
> --mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3 -msha512
> +-mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3 -msha512 -msm4
>  -mcldemote  -mms-bitfields  -mno-align-stringops  -minline-all-stringops
>  -minline-stringops-dynamically  -mstringop-strategy=@var{alg}
>  -mkl -mwidekl
> @@ -33561,6 +33561,9 @@ preferred alignment to @option{-mpreferred-stack-boundary=2}.
>  @need 200
>  @opindex msha512
>  @itemx -msha512
> +@need 200
> +@opindex msm4
> +@itemx -msm4
>  These switches enable the use of instructions in the MMX, SSE,
>  AVX512ER, AVX512CD, AVX512VL, AVX512BW, AVX512DQ, AVX512IFMA, AVX512VBMI, SHA,
>  AES, PCLMUL, CLFLUSHOPT, CLWB, FSGSBASE, PTWRITE, RDRND, F16C, FMA, PCONFIG,
> @@ -33571,8 +33574,8 @@ GFNI, VAES, WAITPKG, VPCLMULQDQ, AVX512BITALG, MOVDIRI, MOVDIR64B, AVX512BF16,
>  ENQCMD, AVX512VPOPCNTDQ, AVX5124FMAPS, AVX512VNNI, AVX5124VNNIW, SERIALIZE,
>  UINTR, HRESET, AMXTILE, AMXINT8, AMXBF16, KL, WIDEKL, AVXVNNI, AVX512-FP16,
>  AVXIFMA, AVXVNNIINT8, AVXNECONVERT, CMPCCXADD, AMX-FP16, PREFETCHI, RAOINT,
> -AMX-COMPLEX, AVXVNNIINT16, SM3, SHA512 or CLDEMOTE extended instruction sets.
> -Each has a corresponding @option{-mno-} option to disable use of these
> +AMX-COMPLEX, AVXVNNIINT16, SM3, SHA512, SM4 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
> diff --git a/gcc/doc/sourcebuild.texi b/gcc/doc/sourcebuild.texi
> index 54a062db3fe..e5d15d67253 100644
> --- a/gcc/doc/sourcebuild.texi
> +++ b/gcc/doc/sourcebuild.texi
> @@ -2595,6 +2595,9 @@ Target supports the execution of @code{sha512} instructions.
>  @item sm3
>  Target supports the execution of @code{sm3} instructions.
>
> +@item sm4
> +Target supports the execution of @code{sm4} instructions.
> +
>  @item sqrt_insn
>  Target has a square root instruction that the compiler can generate.
>
> diff --git a/gcc/testsuite/g++.dg/other/i386-2.C b/gcc/testsuite/g++.dg/other/i386-2.C
> index 985f14abcbc..7d68967488d 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 -mavxvnniint8 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3 -msha512" } */
> +/* { 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 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3 -msha512 -msm4" } */
>
>  /* 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 274b0e6256f..9b775c33ab4 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 -mavxvnniint8 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3 -msha512" } */
> +/* { 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 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3 -msha512 -msm4" } */
>
>  /* 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/funcspec-56.inc b/gcc/testsuite/gcc.target/i386/funcspec-56.inc
> index eb9309819ab..577bfc75edf 100644
> --- a/gcc/testsuite/gcc.target/i386/funcspec-56.inc
> +++ b/gcc/testsuite/gcc.target/i386/funcspec-56.inc
> @@ -91,6 +91,7 @@ extern void test_amx_complex (void)           __attribute__((__target__("amx-complex")));
>  extern void test_avxvnniint16 (void)           __attribute__((__target__("avxvnniint16")));
>  extern void test_sm3 (void)                    __attribute__((__target__("sm3")));
>  extern void test_sha512 (void)                 __attribute__((__target__("sha512")));
> +extern void test_sm4 (void)                     __attribute__((__target__("sm4")));
>
>  extern void test_no_sgx (void)                 __attribute__((__target__("no-sgx")));
>  extern void test_no_avx5124fmaps(void)         __attribute__((__target__("no-avx5124fmaps")));
> @@ -183,6 +184,7 @@ extern void test_no_amx_complex (void)              __attribute__((__target__("no-amx-comple
>  extern void test_no_avxvnniint16 (void)                __attribute__((__target__("no-avxvnniint16")));
>  extern void test_no_sm3 (void)                 __attribute__((__target__("no-sm3")));
>  extern void test_no_sha512 (void)              __attribute__((__target__("no-sha512")));
> +extern void test_no_sm4 (void)                  __attribute__((__target__("no-sm4")));
>
>  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/sm4-1.c b/gcc/testsuite/gcc.target/i386/sm4-1.c
> new file mode 100644
> index 00000000000..2d3d6c6aab8
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/i386/sm4-1.c
> @@ -0,0 +1,20 @@
> +/* { dg-do compile } */
> +/* { dg-options "-O2 -msm4" } */
> +/* { dg-final { scan-assembler "vsm4key4\[ \\t\]+\[^\n\]*%xmm\[0-9\]+\[^\n\]*%xmm\[0-9\]+\[^\n\]*%xmm\[0-9\]" } } */
> +/* { dg-final { scan-assembler "vsm4key4\[ \\t\]+\[^\n\]*%ymm\[0-9\]+\[^\n\]*%ymm\[0-9\]+\[^\n\]*%ymm\[0-9\]" } } */
> +/* { dg-final { scan-assembler "vsm4rnds4\[ \\t\]+\[^\n\]*%xmm\[0-9\]+\[^\n\]*%xmm\[0-9\]+\[^\n\]*%xmm\[0-9\]" } } */
> +/* { dg-final { scan-assembler "vsm4rnds4\[ \\t\]+\[^\n\]*%ymm\[0-9\]+\[^\n\]*%ymm\[0-9\]+\[^\n\]*%ymm\[0-9\]" } } */
> +
> +#include <immintrin.h>
> +
> +volatile __m128i a, b, c;
> +volatile __m256i d, e, f;
> +
> +void extern
> +sm4_test (void)
> +{
> +  a = _mm_sm4key4_epi32 (b, c);
> +  d = _mm256_sm4key4_epi32 (e, f);
> +  a = _mm_sm4rnds4_epi32 (b, c);
> +  d = _mm256_sm4rnds4_epi32 (e, f);
> +}
> diff --git a/gcc/testsuite/gcc.target/i386/sm4-check.h b/gcc/testsuite/gcc.target/i386/sm4-check.h
> new file mode 100644
> index 00000000000..435fcf2b17d
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/i386/sm4-check.h
> @@ -0,0 +1,183 @@
> +#include <stdlib.h>
> +#include "m256-check.h"
> +
> +static void sm4_test (void);
> +
> +typedef union
> +{
> +  unsigned int x;
> +  unsigned char a[4];
> +} union32ui_ub;
> +
> +unsigned char sbox[256] = {
> +0xD6, 0x90, 0xE9, 0xFE, 0xCC, 0xE1, 0x3D, 0xB7,
> +0x16, 0xB6, 0x14, 0xC2, 0x28, 0xFB, 0x2C, 0x05,
> +0x2B, 0x67, 0x9A, 0x76, 0x2A, 0xBE, 0x04, 0xC3,
> +0xAA, 0x44, 0x13, 0x26, 0x49, 0x86, 0x06, 0x99,
> +0x9C, 0x42, 0x50, 0xF4, 0x91, 0xEF, 0x98, 0x7A,
> +0x33, 0x54, 0x0B, 0x43, 0xED, 0xCF, 0xAC, 0x62,
> +0xE4, 0xB3, 0x1C, 0xA9, 0xC9, 0x08, 0xE8, 0x95,
> +0x80, 0xDF, 0x94, 0xFA, 0x75, 0x8F, 0x3F, 0xA6,
> +0x47, 0x07, 0xA7, 0xFC, 0xF3, 0x73, 0x17, 0xBA,
> +0x83, 0x59, 0x3C, 0x19, 0xE6, 0x85, 0x4F, 0xA8,
> +0x68, 0x6B, 0x81, 0xB2, 0x71, 0x64, 0xDA, 0x8B,
> +0xF8, 0xEB, 0x0F, 0x4B, 0x70, 0x56, 0x9D, 0x35,
> +0x1E, 0x24, 0x0E, 0x5E, 0x63, 0x58, 0xD1, 0xA2,
> +0x25, 0x22, 0x7C, 0x3B, 0x01, 0x21, 0x78, 0x87,
> +0xD4, 0x00, 0x46, 0x57, 0x9F, 0xD3, 0x27, 0x52,
> +0x4C, 0x36, 0x02, 0xE7, 0xA0, 0xC4, 0xC8, 0x9E,
> +0xEA, 0xBF, 0x8A, 0xD2, 0x40, 0xC7, 0x38, 0xB5,
> +0xA3, 0xF7, 0xF2, 0xCE, 0xF9, 0x61, 0x15, 0xA1,
> +0xE0, 0xAE, 0x5D, 0xA4, 0x9B, 0x34, 0x1A, 0x55,
> +0xAD, 0x93, 0x32, 0x30, 0xF5, 0x8C, 0xB1, 0xE3,
> +0x1D, 0xF6, 0xE2, 0x2E, 0x82, 0x66, 0xCA, 0x60,
> +0xC0, 0x29, 0x23, 0xAB, 0x0D, 0x53, 0x4E, 0x6F,
> +0xD5, 0xDB, 0x37, 0x45, 0xDE, 0xFD, 0x8E, 0x2F,
> +0x03, 0xFF, 0x6A, 0x72, 0x6D, 0x6C, 0x5B, 0x51,
> +0x8D, 0x1B, 0xAF, 0x92, 0xBB, 0xDD, 0xBC, 0x7F,
> +0x11, 0xD9, 0x5C, 0x41, 0x1F, 0x10, 0x5A, 0xD8,
> +0x0A, 0xC1, 0x31, 0x88, 0xA5, 0xCD, 0x7B, 0xBD,
> +0x2D, 0x74, 0xD0, 0x12, 0xB8, 0xE5, 0xB4, 0xB0,
> +0x89, 0x69, 0x97, 0x4A, 0x0C, 0x96, 0x77, 0x7E,
> +0x65, 0xB9, 0xF1, 0x09, 0xC5, 0x6E, 0xC6, 0x84,
> +0x18, 0xF0, 0x7D, 0xEC, 0x3A, 0xDC, 0x4D, 0x20,
> +0x79, 0xEE, 0x5F, 0x3E, 0xD7, 0xCB, 0x39, 0x48
> +};
> +
> +static unsigned
> +rol32 (unsigned w, int n)
> +{
> +  int count = n % 32;
> +  return ((w << count) | (w >> (32 - count)));
> +}
> +
> +static unsigned char
> +sbox_byte (unsigned w, int i)
> +{
> +  union32ui_ub tmp;
> +  tmp.x = w;
> +  return sbox[tmp.a[i]];
> +}
> +
> +static unsigned
> +lower_t (unsigned w)
> +{
> +  union32ui_ub tmp;
> +  tmp.a[0] = sbox_byte (w, 0);
> +  tmp.a[1] = sbox_byte (w, 1);
> +  tmp.a[2] = sbox_byte (w, 2);
> +  tmp.a[3] = sbox_byte (w, 3);
> +  return tmp.x;
> +}
> +
> +static unsigned
> +l_key (unsigned w)
> +{
> +  return w ^ rol32 (w, 13) ^ rol32 (w, 23);
> +}
> +
> +static unsigned
> +l_rnds (unsigned w)
> +{
> +  unsigned tmp = w;
> +  tmp = tmp ^ rol32 (w, 2);
> +  tmp = tmp ^ rol32 (w, 10);
> +  tmp = tmp ^ rol32 (w, 18);
> +  tmp = tmp ^ rol32 (w, 24);
> +  return tmp;
> +}
> +
> +#define SM4_FUNC(name)                                       \
> +static unsigned                                                      \
> +t_##name (unsigned w)                                        \
> +{                                                            \
> +  return l_##name (lower_t (w));                             \
> +}                                                            \
> +                                                             \
> +static unsigned                                                      \
> +f_##name (unsigned x0, unsigned x1, unsigned x2, unsigned x3, unsigned k) \
> +{                                                            \
> +  return x0 ^ t_##name (x1 ^ x2 ^ x3 ^ k);                   \
> +}                                                            \
> +                                                             \
> +static void                                                  \
> +compute_sm4##name##4 (int *dst, int *src1, int *src2, int vl) \
> +{                                                            \
> +  unsigned c[4], p[4];                                       \
> +                                                             \
> +  int kl = vl / 128;                                         \
> +  int i;                                                     \
> +                                                             \
> +  for (i = 0; i < kl; i++)                                   \
> +  {                                                          \
> +    p[0] = src1[4 * i];                                              \
> +    p[1] = src1[4 * i + 1];                                  \
> +    p[2] = src1[4 * i + 2];                                  \
> +    p[3] = src1[4 * i + 3];                                  \
> +                                                             \
> +    c[0] = f_##name (p[0], p[1], p[2], p[3], src2[4 * i]);    \
> +    c[1] = f_##name (p[1], p[2], p[3], c[0], src2[4 * i + 1]); \
> +    c[2] = f_##name (p[2], p[3], c[0], c[1], src2[4 * i + 2]); \
> +    c[3] = f_##name (p[3], c[0], c[1], c[2], src2[4 * i + 3]); \
> +                                                             \
> +    dst[4 * i] = c[0];                                       \
> +    dst[4 * i + 1] = c[1];                                   \
> +    dst[4 * i + 2] = c[2];                                   \
> +    dst[4 * i + 3] = c[3];                                   \
> +  }                                                          \
> +}
> +
> +#define SM4_AVX_SIMULATE(name)                               \
> +  union128i_d src1, src2, res1;                                      \
> +  int dst1[4] = {0, 0, 0, 0};                                \
> +                                                             \
> +  src1.x = _mm_set_epi32 (111, 222, 333, 444);               \
> +  src2.x = _mm_set_epi32 (555, 666, 777, 888);               \
> +  res1.x = _mm_set_epi32 (0, 0, 0, 0);                       \
> +                                                             \
> +  res1.x = _mm_sm4##name##4_epi32 (src1.x, src2.x);          \
> +                                                             \
> +  compute_sm4##name##4 (dst1, src1.a, src2.a, 128);          \
> +                                                             \
> +  if (check_union128i_d (res1, dst1))                        \
> +    abort ();                                                \
> +                                                             \
> +  union256i_d src3, src4, res2;                                      \
> +  int dst2[8] = {0, 0, 0, 0, 0, 0, 0, 0};                    \
> +                                                             \
> +  src3.x = _mm256_set_epi32 (111, 222, 333, 444, 555, 666, 777, 888); \
> +  src4.x = _mm256_set_epi32 (999, 123, 456, 789, 135, 792, 468, 147); \
> +  res2.x = _mm256_set_epi32 (0, 0, 0, 0, 0, 0, 0, 0);        \
> +                                                             \
> +  res2.x = _mm256_sm4##name##4_epi32 (src3.x, src4.x);       \
> +                                                             \
> +  compute_sm4##name##4 (dst2, src3.a, src4.a, 256);          \
> +                                                             \
> +  if (check_union256i_d (res2, dst2))                        \
> +    abort ();
> +
> +static void
> +__attribute__ ((noinline))
> +do_test (void)
> +{
> +  sm4_test ();
> +}
> +
> +int
> +main ()
> +{
> +  /* Check CPU support for SM4.  */
> +  if (__builtin_cpu_supports ("sm4"))
> +    {
> +      do_test ();
> +#ifdef DEBUG
> +      printf ("PASSED\n");
> +#endif
> +      return 0;
> +    }
> +
> +#ifdef DEBUG
> +  printf ("SKIPPED\n");
> +#endif
> +  return 0;
> +}
> diff --git a/gcc/testsuite/gcc.target/i386/sm4key4-2.c b/gcc/testsuite/gcc.target/i386/sm4key4-2.c
> new file mode 100644
> index 00000000000..a8bec560214
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/i386/sm4key4-2.c
> @@ -0,0 +1,14 @@
> +/* { dg-do run } */
> +/* { dg-options "-O2 -msm4" } */
> +/* { dg-require-effective-target sm4 } */
> +
> +#include "sm4-check.h"
> +
> +char key;
> +SM4_FUNC (key);
> +
> +static void
> +sm4_test (void)
> +{
> +  SM4_AVX_SIMULATE (key);
> +}
> diff --git a/gcc/testsuite/gcc.target/i386/sm4rnds4-2.c b/gcc/testsuite/gcc.target/i386/sm4rnds4-2.c
> new file mode 100644
> index 00000000000..0860d0dd412
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/i386/sm4rnds4-2.c
> @@ -0,0 +1,14 @@
> +/* { dg-do run } */
> +/* { dg-options "-O2 -msm4" } */
> +/* { dg-require-effective-target sm4 } */
> +
> +#include "sm4-check.h"
> +
> +char rnds;
> +SM4_FUNC (rnds);
> +
> +static void
> +sm4_test (void)
> +{
> +  SM4_AVX_SIMULATE (rnds);
> +}
> diff --git a/gcc/testsuite/gcc.target/i386/sse-12.c b/gcc/testsuite/gcc.target/i386/sse-12.c
> index 976541389ea..a553a5202d1 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 -mavxvnniint8 -mavxneconvert -mamx-fp16 -mraoint -mamx-complex -mavxvnniint16 -msm3 -msha512" } */
> +/* { 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 -mavxneconvert -mamx-fp16 -mraoint -mamx-complex -mavxvnniint16 -msm3 -msha512 -msm4" } */
>
>  #include <x86intrin.h>
>
> diff --git a/gcc/testsuite/gcc.target/i386/sse-13.c b/gcc/testsuite/gcc.target/i386/sse-13.c
> index 8c314e70e31..946182f0e76 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 -mavxvnniint8 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3 -msha512" } */
> +/* { 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 -mavxneconvert -mcmpccxadd -mamx-fp16 -mprefetchi -mraoint -mamx-complex -mavxvnniint16 -msm3 -msha512 -msm4" } */
>  /* { 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 2b4d7bc9079..0d07aadc7f8 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 -mavxvnniint8 -mavxneconvert -mamx-fp16 -mraoint -mamx-complex -mavxvnniint16 -msm3 -msha512" } */
> +/* { 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 -mavxneconvert -mamx-fp16 -mraoint -mamx-complex -mavxvnniint16 -msm3 -msha512 -msm4" } */
>  /* { 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 d6f19b5e20a..e6681f7dd12 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,tbm,lwp,fsgsbase,rdrnd,f16c,rtm,rdseed,prfchw,adx,fxsr,xsaveopt,avx512f,avx512er,avx512cd,avx512pf,sha,prefetchwt1,avx512vl,avx512bw,avx512dq,avx512vbmi,avx512vbmi2,avx512ifma,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,gfni,avx512bitalg,avx512bf16,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,amx-fp16,raoint,amx-complex,avxvnniint16,sm3,sha512")
> +#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,avx512f,avx512er,avx512cd,avx512pf,sha,prefetchwt1,avx512vl,avx512bw,avx512dq,avx512vbmi,avx512vbmi2,avx512ifma,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,gfni,avx512bitalg,avx512bf16,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,amx-fp16,raoint,amx-complex,avxvnniint16,sm3,sha512,sm4")
>  #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,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,gfni,avx512bitalg,avx512bf16,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,amx-fp16,raoint,amx-complex,avxvnniint16,sm3,sha512")
> +#pragma GCC target ("avx,avx2,rdrnd,fsgsbase,f16c,rtm,avx512f,avx512er,avx512cd,avx512pf,sha,avx512vl,avx512bw,avx512dq,avx512ifma,avx512vbmi,avx512vbmi2,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,gfni,avx512bitalg,avx512bf16,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,amx-fp16,raoint,amx-complex,avxvnniint16,sm3,sha512,sm4")
>  #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 1df66b525ca..92b1c467d95 100644
> --- a/gcc/testsuite/gcc.target/i386/sse-23.c
> +++ b/gcc/testsuite/gcc.target/i386/sse-23.c
> @@ -850,6 +850,6 @@
>  /* sm3intrin.h */
>  #define __builtin_ia32_vsm3rnds2(A, B, C, D) __builtin_ia32_vsm3rnds2 (A, B, C, 1)
>
> -#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,avx512f,avx512er,avx512cd,avx512pf,sha,prefetchwt1,xsavec,xsaves,clflushopt,avx512bw,avx512dq,avx512vl,avx512vbmi,avx512ifma,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,clwb,mwaitx,clzero,pku,sgx,rdpid,gfni,avx512vbmi2,vpclmulqdq,avx512bitalg,pconfig,wbnoinvd,avx512bf16,enqcmd,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,cmpccxadd,amx-fp16,prefetchi,raoint,amx-complex,avxvnniint16,sm3,sha512")
> +#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,avx512f,avx512er,avx512cd,avx512pf,sha,prefetchwt1,xsavec,xsaves,clflushopt,avx512bw,avx512dq,avx512vl,avx512vbmi,avx512ifma,avx5124fmaps,avx5124vnniw,avx512vpopcntdq,clwb,mwaitx,clzero,pku,sgx,rdpid,gfni,avx512vbmi2,vpclmulqdq,avx512bitalg,pconfig,wbnoinvd,avx512bf16,enqcmd,avx512vp2intersect,serialize,tsxldtrk,amx-tile,amx-int8,amx-bf16,kl,widekl,avxvnni,avx512fp16,avxifma,avxvnniint8,avxneconvert,cmpccxadd,amx-fp16,prefetchi,raoint,amx-complex,avxvnniint16,sm3,sha512,sm4")
>
>  #include <x86intrin.h>
> diff --git a/gcc/testsuite/lib/target-supports.exp b/gcc/testsuite/lib/target-supports.exp
> index f376d835f8b..8ea0d9feb1c 100644
> --- a/gcc/testsuite/lib/target-supports.exp
> +++ b/gcc/testsuite/lib/target-supports.exp
> @@ -9897,6 +9897,20 @@ proc check_effective_target_sha512 { } {
>      } "-msha512" ]
>  }
>
> +# Return 1 if sm4 instructions can be compiled.
> +proc check_effective_target_sm4 { } {
> +    return [check_no_compiler_messages sm4 object {
> +        typedef long long __m128i __attribute__ ((__vector_size__ (16)));
> +        typedef int __v4si __attribute__ ((__vector_size__ (16)));
> +        __m128i
> +        _mm_sm4key4_epi32 (__m128i __A, __m128i __B)
> +        {
> +          return (__m128i) __builtin_ia32_vsm4key4128 ((__v4si) __A,
> +                                                      (__v4si) __B);
> +        }
> +    } "-msm4" ]
> +}
> +
>  # Return 1 if sse instructions can be compiled.
>  proc check_effective_target_sse { } {
>      return [check_no_compiler_messages sse object {
> --
> 2.31.1
>


-- 
BR,
Hongtao

^ permalink raw reply	[flat|nested] 9+ messages in thread

end of thread, other threads:[~2023-07-17  1:40 UTC | newest]

Thread overview: 9+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-07-13  6:03 [PATCH 0/4] Support Intel Arrow Lake/Lunar Lake ISAs Haochen Jiang
2023-07-13  6:03 ` [PATCH 1/4] Support Intel AVX-VNNI-INT16 Haochen Jiang
2023-07-17  1:39   ` Hongtao Liu
2023-07-13  6:03 ` [PATCH 2/4] Support Intel SM3 Haochen Jiang
2023-07-17  1:39   ` Hongtao Liu
2023-07-13  6:03 ` [PATCH 3/4] Support Intel SHA512 Haochen Jiang
2023-07-17  1:40   ` Hongtao Liu
2023-07-13  6:03 ` [PATCH 4/4] Support Intel SM4 Haochen Jiang
2023-07-17  1:40   ` Hongtao Liu

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).