Enable VNNI support [4/5]

gcc/
	* config/i386/avx512vnniintrin.h (_mm512_dpwssd_epi32,
	_mm512_mask_dpwssd_epi32, _mm512_maskz_dpwssd_epi32): New intrinsics.
	* config/i386/avx512vnnivlintrin.h (_mm256_dpwssd_epi32,
	_mm256_mask_dpwssd_epi32, _mm256_maskz_dpwssd_epi32, _mm_dpwssd_epi32,
	_mm_mask_dpwssd_epi32, _mm_maskz_dpwssd_epi32): Ditto.

gcc/testsuite/
	* gcc.target/i386/avx512f-vnni-1.c: Add vdpwssd checks.
	* gcc.target/i386/avx512vl-vnni-1.c: Ditto.
	* gcc.target/i386/avx512f-vpdpwssd-2.c: New.
	* gcc.target/i386/avx512vl-vpdpwssd-2.c: Ditto.


git-svn-id: svn+ssh://gcc.gnu.org/svn/gcc/trunk@255496 138bc75d-0d04-0410-961f-82ee72b054a4

8 files changed, 204 insertions, 14 deletions

diff --git a/gcc/ChangeLog b/gcc/ChangeLog
index 3de4b428ae9..1d20dc5234c 100644
--- a/gcc/ChangeLog
+++ b/gcc/ChangeLog
@@ -1,14 +1,19 @@
 2017-12-08  Julia Koval  <julia.koval@intel.com>
 
-	* config.gcc: Add vaesintrin.h.
-	* config/i386/i386-builtin-types.def (V64QI_FTYPE_V64QI_V64QI): New type.
-	* config/i386/i386-builtin.def (__builtin_ia32_vaesdec_v16qi,
-	__builtin_ia32_vaesdec_v32qi, __builtin_ia32_vaesdec_v64qi): New builtins.
-	* config/i386/i386.c (ix86_expand_args_builtin): Handle new type.
-	* config/i386/immintrin.h: Include vaesintrin.h.
-	* config/i386/sse.md (vaesdec_<mode>): New pattern.
-	* config/i386/vaesintrin.h (_mm256_aesdec_epi128, _mm512_aesdec_epi128,
-	_mm_aesdec_epi128): New intrinsics.
+	* config/i386/avx512vnniintrin.h (_mm512_dpwssd_epi32,
+	_mm512_mask_dpwssd_epi32, _mm512_maskz_dpwssd_epi32): New intrinsics.
+	* config/i386/avx512vnnivlintrin.h (_mm256_dpwssd_epi32,
+	_mm256_mask_dpwssd_epi32, _mm256_maskz_dpwssd_epi32, _mm_dpwssd_epi32,
+	_mm_mask_dpwssd_epi32, _mm_maskz_dpwssd_epi32): Ditto.
+
+2017-12-08  Julia Koval  <julia.koval@intel.com>
+
+	* config/i386/avx512vnniintrin.h (_mm512_dpbusds_epi32,
+	_mm512_mask_dpbusds_epi32, _mm512_maskz_dpbusds_epi32): New.
+	* config/i386/avx512vnnivlintrin.h (_mm256_dpbusds_epi32,
+	_mm256_mask_dpbusds_epi32, _mm256_maskz_dpbusds_epi32,
+	_mm_dpbusds_epi32, _mm_mask_dpbusds_epi32,
+	_mm_maskz_dpbusds_epi32): New intrinsics.
 
 2017-12-07  Sandra Loosemore  <sandra@codesourcery.com>
 
diff --git a/gcc/config/i386/avx512vnniintrin.h b/gcc/config/i386/avx512vnniintrin.h
index c9842f3833f..c435dc14a85 100644
--- a/gcc/config/i386/avx512vnniintrin.h
+++ b/gcc/config/i386/avx512vnniintrin.h
@@ -85,6 +85,31 @@ _mm512_maskz_dpbusds_epi32 (__mmask16 __A, __m512i __B, __m512i __C,
 				(__v16si) __C, (__v16si) __D, (__mmask16)__A);
 }
 
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_dpwssd_epi32 (__m512i __A, __m512i __B, __m512i __C)
+{
+  return (__m512i) __builtin_ia32_vpdpwssd_v16si ((__v16si)__A, (__v16si) __B,
+								(__v16si) __C);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_dpwssd_epi32 (__m512i __A, __mmask16 __B, __m512i __C, __m512i __D)
+{
+  return (__m512i)__builtin_ia32_vpdpwssd_v16si_mask ((__v16si)__A,
+				(__v16si) __C, (__v16si) __D, (__mmask16)__B);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_dpwssd_epi32 (__mmask16 __A, __m512i __B, __m512i __C,
+							__m512i __D)
+{
+  return (__m512i)__builtin_ia32_vpdpwssd_v16si_maskz ((__v16si)__B,
+				(__v16si) __C, (__v16si) __D, (__mmask16)__A);
+}
+
 #ifdef __DISABLE_AVX512VNNI__
 #undef __DISABLE_AVX512VNNI__
 #pragma GCC pop_options
diff --git a/gcc/config/i386/avx512vnnivlintrin.h b/gcc/config/i386/avx512vnnivlintrin.h
index 621f0aceb71..d87feaa3815 100644
--- a/gcc/config/i386/avx512vnnivlintrin.h
+++ b/gcc/config/i386/avx512vnnivlintrin.h
@@ -131,6 +131,54 @@ _mm_maskz_dpbusds_epi32 (__mmask8 __A, __m128i __B, __m128i __C, __m128i __D)
 				(__v4si) __C, (__v4si) __D, (__mmask8)__A);
 }
 
+extern __inline __m256i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_dpwssd_epi32 (__m256i __A, __m256i __B, __m256i __C)
+{
+  return (__m256i) __builtin_ia32_vpdpwssd_v8si ((__v8si)__A, (__v8si) __B,
+								(__v8si) __C);
+}
+
+extern __inline __m256i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_dpwssd_epi32 (__m256i __A, __mmask8 __B, __m256i __C, __m256i __D)
+{
+  return (__m256i)__builtin_ia32_vpdpwssd_v8si_mask ((__v8si)__A, (__v8si) __C,
+						(__v8si) __D, (__mmask8)__B);
+}
+
+extern __inline __m256i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_dpwssd_epi32 (__mmask8 __A, __m256i __B, __m256i __C, __m256i __D)
+{
+  return (__m256i)__builtin_ia32_vpdpwssd_v8si_maskz ((__v8si)__B,
+				(__v8si) __C, (__v8si) __D, (__mmask8)__A);
+}
+
+extern __inline __m128i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_dpwssd_epi32 (__m128i __A, __m128i __B, __m128i __C)
+{
+  return (__m128i) __builtin_ia32_vpdpwssd_v4si ((__v4si)__A, (__v4si) __B,
+								(__v4si) __C);
+}
+
+extern __inline __m128i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_dpwssd_epi32 (__m128i __A, __mmask8 __B, __m128i __C, __m128i __D)
+{
+  return (__m128i)__builtin_ia32_vpdpwssd_v4si_mask ((__v4si)__A, (__v4si) __C,
+						(__v4si) __D, (__mmask8)__B);
+}
+
+extern __inline __m128i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_dpwssd_epi32 (__mmask8 __A, __m128i __B, __m128i __C, __m128i __D)
+{
+  return (__m128i)__builtin_ia32_vpdpwssd_v4si_maskz ((__v4si)__B,
+				(__v4si) __C, (__v4si) __D, (__mmask8)__A);
+}
+
 #ifdef __DISABLE_AVX512VNNIVL__
 #undef __DISABLE_AVX512VNNIVL__
 #pragma GCC pop_options
diff --git a/gcc/testsuite/ChangeLog b/gcc/testsuite/ChangeLog
index 3020f51df66..b13130d4454 100644
--- a/gcc/testsuite/ChangeLog
+++ b/gcc/testsuite/ChangeLog
@@ -1,10 +1,16 @@
 2017-12-08  Julia Koval  <julia.koval@intel.com>
 
-	* gcc.target/i386/avx512-check.h: Handle bit_VAES.
-	* gcc.target/i386/avx512f-aesdec-2.c: New test.
-	* gcc.target/i386/avx512fvl-vaes-1.c: Ditto.
-	* gcc.target/i386/avx512vl-aesdec-2.c: Ditto.
-	* gcc.target/i386/i386.exp (check_effective_target_avx512vaes): New.
+	* gcc.target/i386/avx512f-vnni-1.c: Add vdpwssd checks.
+	* gcc.target/i386/avx512vl-vnni-1.c: Ditto.
+	* gcc.target/i386/avx512f-vpdpwssd-2.c: New.
+	* gcc.target/i386/avx512vl-vpdpwssd-2.c: Ditto.
+
+2017-12-08  Julia Koval  <julia.koval@intel.com>
+
+	* gcc.target/i386/avx512f-vnni-1.c: Add vpdpbusds check.
+	* gcc.target/i386/avx512vl-vnni-1.c: Ditto.
+	* gcc.target/i386/avx512f-vpdpbusds-2.c: New.
+	* gcc.target/i386/avx512vl-vpdpbusds-2.c: Ditto.
 
 2017-12-07  Sandra Loosemore  <sandra@codesourcery.com>
 
diff --git a/gcc/testsuite/gcc.target/i386/avx512f-vnni-1.c b/gcc/testsuite/gcc.target/i386/avx512f-vnni-1.c
index b4b6d83d3a9..d6c319b8007 100644
--- a/gcc/testsuite/gcc.target/i386/avx512f-vnni-1.c
+++ b/gcc/testsuite/gcc.target/i386/avx512f-vnni-1.c
@@ -6,6 +6,9 @@
 /* { dg-final { scan-assembler-times "vpdpbusds\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\n\r]*%zmm\[0-9\]+\[^\n\r]*%zmm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
 /* { dg-final { scan-assembler-times "vpdpbusds\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\\n\\r]*%zmm\[0-9\]+\[^\\n\\r\]*%zmm\[0-9\]+\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */
 /* { dg-final { scan-assembler-times "vpdpbusds\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\\n\\r]*%zmm\[0-9\]+\[^\\n\\r\]*%zmm\[0-9\]+\{%k\[1-7\]\}\{z\}(?:\n|\[ \\t\]+#)" 1 } } */
+/* { dg-final { scan-assembler-times "vpdpwssd\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\n\r]*%zmm\[0-9\]+\[^\n\r]*%zmm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
+/* { dg-final { scan-assembler-times "vpdpwssd\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\\n\\r]*%zmm\[0-9\]+\[^\\n\\r\]*%zmm\[0-9\]+\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */
+/* { dg-final { scan-assembler-times "vpdpwssd\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\\n\\r]*%zmm\[0-9\]+\[^\\n\\r\]*%zmm\[0-9\]+\{%k\[1-7\]\}\{z\}(?:\n|\[ \\t\]+#)" 1 } } */
 
 #include <immintrin.h>
 
@@ -25,4 +28,8 @@ avx512f_test (void)
   x = _mm512_mask_dpbusds_epi32 (x, m16, y, z);
   x = _mm512_maskz_dpbusds_epi32 (m16, x, y, z);
 
+  x = _mm512_dpwssd_epi32 (x, y, z);
+  x = _mm512_mask_dpwssd_epi32 (x, m16, y, z);
+  x = _mm512_maskz_dpwssd_epi32 (m16, x, y, z);
+
 }
diff --git a/gcc/testsuite/gcc.target/i386/avx512f-vpdpwssd-2.c b/gcc/testsuite/gcc.target/i386/avx512f-vpdpwssd-2.c
new file mode 100644
index 00000000000..d8488e8251f
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/avx512f-vpdpwssd-2.c
@@ -0,0 +1,69 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -mavx512f -mavx512vnni" } */
+/* { dg-require-effective-target avx512f } */
+/* { dg-require-effective-target avx512vnni } */
+
+#define AVX512F
+
+#define AVX512VNNI
+#include "avx512f-helper.h"
+
+#define SIZE (AVX512F_LEN / 16)
+#define SIZE_RES (AVX512F_LEN / 32)
+
+#include "avx512f-mask-type.h"
+
+static void
+CALC (int *r, int *dst, short *s1, short *s2)
+{
+  short tempres[SIZE];
+  for (int i = 0; i < SIZE; i++) {
+    tempres[i] = ((int)(s1[i]) * (int)(s2[i]));
+  }
+  for (int i = 0; i < SIZE_RES; i++) {
+    long long test = (long long)dst[i] + tempres[i*2] + tempres[i*2 + 1];
+    r[i] = test;
+  }
+}
+
+void
+TEST (void)
+{
+  int i;
+  UNION_TYPE (AVX512F_LEN, i_d) res1, res2, res3;
+  UNION_TYPE (AVX512F_LEN, i_w) src1, src2;
+  MASK_TYPE mask = MASK_VALUE;
+  int res_ref[SIZE_RES];
+  int res_ref2[SIZE_RES];
+
+  for (i = 0; i < SIZE; i++)
+    {
+      src1.a[i] = 1 + i;
+      src2.a[i] = 2 + 2*i;
+    }
+
+  for (i = 0; i < SIZE_RES; i++)
+    {
+      res1.a[i] = 0x7fffffff;
+      res2.a[i] = DEFAULT_VALUE;
+      res3.a[i] = DEFAULT_VALUE;
+    }
+
+  CALC (res_ref, res1.a, src1.a, src2.a);
+  CALC (res_ref2, res2.a, src1.a, src2.a);
+
+  res1.x = INTRINSIC (_dpwssd_epi32) (res1.x, src1.x, src2.x);
+  res2.x = INTRINSIC (_mask_dpwssd_epi32) (res2.x, mask, src1.x, src2.x);
+  res3.x = INTRINSIC (_maskz_dpwssd_epi32) (mask, res3.x, src1.x, src2.x);
+
+  if (UNION_CHECK (AVX512F_LEN, i_d) (res1, res_ref))
+    abort ();
+
+  MASK_MERGE (i_d) (res_ref2, mask, SIZE_RES);
+  if (UNION_CHECK (AVX512F_LEN, i_d) (res2, res_ref2))
+    abort ();
+
+  MASK_ZERO (i_d) (res_ref2, mask, SIZE_RES);
+  if (UNION_CHECK (AVX512F_LEN, i_d) (res3, res_ref2))
+    abort ();
+}
diff --git a/gcc/testsuite/gcc.target/i386/avx512vl-vnni-1.c b/gcc/testsuite/gcc.target/i386/avx512vl-vnni-1.c
index 058c082cff1..aba98979021 100644
--- a/gcc/testsuite/gcc.target/i386/avx512vl-vnni-1.c
+++ b/gcc/testsuite/gcc.target/i386/avx512vl-vnni-1.c
@@ -12,6 +12,12 @@
 /* { dg-final { scan-assembler-times "vpdpbusds\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
 /* { dg-final { scan-assembler-times "vpdpbusds\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\\n\\r]*%xmm\[0-9\]+\[^\\n\\r\]*%xmm\[0-9\]+\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */
 /* { dg-final { scan-assembler-times "vpdpbusds\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\\n\\r]*%xmm\[0-9\]+\[^\\n\\r\]*%xmm\[0-9\]+\{%k\[1-7\]\}\{z\}(?:\n|\[ \\t\]+#)" 1 } } */
+/* { dg-final { scan-assembler-times "vpdpwssd\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\r]*%ymm\[0-9\]+\[^\n\r]*%ymm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
+/* { dg-final { scan-assembler-times "vpdpwssd\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\\n\\r]*%ymm\[0-9\]+\[^\\n\\r\]*%ymm\[0-9\]+\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */
+/* { dg-final { scan-assembler-times "vpdpwssd\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\\n\\r]*%ymm\[0-9\]+\[^\\n\\r\]*%ymm\[0-9\]+\{%k\[1-7\]\}\{z\}(?:\n|\[ \\t\]+#)" 1 } } */
+/* { dg-final { scan-assembler-times "vpdpwssd\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
+/* { dg-final { scan-assembler-times "vpdpwssd\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\\n\\r]*%xmm\[0-9\]+\[^\\n\\r\]*%xmm\[0-9\]+\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */
+/* { dg-final { scan-assembler-times "vpdpwssd\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\\n\\r]*%xmm\[0-9\]+\[^\\n\\r\]*%xmm\[0-9\]+\{%k\[1-7\]\}\{z\}(?:\n|\[ \\t\]+#)" 1 } } */
 
 
 #include <immintrin.h>
@@ -39,4 +45,12 @@ avx512f_test (void)
   x_ = _mm_mask_dpbusds_epi32 (x_, m, y_, z_);
   x_ = _mm_maskz_dpbusds_epi32 (m, x_, y_, z_);
 
+  x = _mm256_dpwssd_epi32 (x, y, z);
+  x = _mm256_mask_dpwssd_epi32 (x, m, y, z);
+  x = _mm256_maskz_dpwssd_epi32 (m, x, y, z);
+
+  x_ = _mm_dpwssd_epi32 (x_, y_, z_);
+  x_ = _mm_mask_dpwssd_epi32 (x_, m, y_, z_);
+  x_ = _mm_maskz_dpwssd_epi32 (m, x_, y_, z_);
+
 }
diff --git a/gcc/testsuite/gcc.target/i386/avx512vl-vpdpwssd-2.c b/gcc/testsuite/gcc.target/i386/avx512vl-vpdpwssd-2.c
new file mode 100644
index 00000000000..023f06f93c5
--- /dev/null
+++ b/gcc/testsuite/gcc.target/i386/avx512vl-vpdpwssd-2.c
@@ -0,0 +1,16 @@
+/* { dg-do run } */
+/* { dg-options "-O2 -mavx512vnni -mavx512vl" } */
+/* { dg-require-effective-target avx512vl } */
+/* { dg-require-effective-target avx512vnni } */
+
+#define AVX512VL
+#define AVX512F_LEN 256
+#define AVX512F_LEN_HALF 128
+#include "avx512f-vpdpwssd-2.c"
+
+#undef AVX512F_LEN
+#undef AVX512F_LEN_HALF
+
+#define AVX512F_LEN 128
+#define AVX512F_LEN_HALF 128
+#include "avx512f-vpdpwssd-2.c"