Enable VNNI support [5/5]

gcc/
	* config/i386/avx512vnniintrin.h (_mm512_dpwssds_epi32,
	_mm512_mask_dpwssds_epi32, _mm512_maskz_dpwssds_epi32): New intrinsics.
	* config/i386/avx512vnnivlintrin.h (_mm256_dpwssds_epi32,
	_mm256_mask_dpwssds_epi32, _mm256_maskz_dpwssds_epi32,
	_mm_dpwssds_epi32, _mm_mask_dpwssds_epi32,
	_mm_maskz_dpwssds_epi32): Ditto.

gcc/testsuite/
	* gcc.target/i386/avx512f-vnni-1.c: Add checks for vdpdwssds.
	* gcc.target/i386/avx512vl-vnni-1.c: Ditto.
	* gcc.target/i386/avx512f-vpdpwssds-2.c: New test.
	* gcc.target/i386/avx512vl-vpdpwssds-2.c: Ditto.

From-SVN: r255498

gcc/ChangeLog

+2017-12-08  Julia Koval  <julia.koval@intel.com>
+
+	* config/i386/avx512vnniintrin.h (_mm512_dpwssds_epi32,
+	_mm512_mask_dpwssds_epi32, _mm512_maskz_dpwssds_epi32): New intrinsics.
+	* config/i386/avx512vnnivlintrin.h (_mm256_dpwssds_epi32,
+	_mm256_mask_dpwssds_epi32, _mm256_maskz_dpwssds_epi32,
+	_mm_dpwssds_epi32, _mm_mask_dpwssds_epi32,
+	_mm_maskz_dpwssds_epi32): Ditto.
+
 2017-12-08  Richard Biener  <rguenther@suse.de>
 
 	PR tree-optimization/81303

gcc/config/i386/avx512vnniintrin.h

@@ -110,6 +110,32 @@ _mm512_maskz_dpwssd_epi32 (__mmask16 __A, __m512i __B, __m512i __C,
 				(__v16si) __C, (__v16si) __D, (__mmask16)__A);
 }
 
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_dpwssds_epi32 (__m512i __A, __m512i __B, __m512i __C)
+{
+  return (__m512i) __builtin_ia32_vpdpwssds_v16si ((__v16si)__A, (__v16si) __B,
+								(__v16si) __C);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_mask_dpwssds_epi32 (__m512i __A, __mmask16 __B, __m512i __C,
+							__m512i __D)
+{
+  return (__m512i)__builtin_ia32_vpdpwssds_v16si_mask ((__v16si)__A,
+				(__v16si) __C, (__v16si) __D, (__mmask16)__B);
+}
+
+extern __inline __m512i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm512_maskz_dpwssds_epi32 (__mmask16 __A, __m512i __B, __m512i __C,
+							__m512i __D)
+{
+  return (__m512i)__builtin_ia32_vpdpwssds_v16si_maskz ((__v16si)__B,
+				(__v16si) __C, (__v16si) __D, (__mmask16)__A);
+}
+
 #ifdef __DISABLE_AVX512VNNI__
 #undef __DISABLE_AVX512VNNI__
 #pragma GCC pop_options

gcc/config/i386/avx512vnnivlintrin.h

@@ -179,6 +179,54 @@ _mm_maskz_dpwssd_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_dpwssds_epi32 (__m256i __A, __m256i __B, __m256i __C)
+{
+  return (__m256i) __builtin_ia32_vpdpwssds_v8si ((__v8si)__A, (__v8si) __B,
+								(__v8si) __C);
+}
+
+extern __inline __m256i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_mask_dpwssds_epi32 (__m256i __A, __mmask8 __B, __m256i __C, __m256i __D)
+{
+  return (__m256i)__builtin_ia32_vpdpwssds_v8si_mask ((__v8si)__A,
+				(__v8si) __C, (__v8si) __D, (__mmask8)__B);
+}
+
+extern __inline __m256i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm256_maskz_dpwssds_epi32 (__mmask8 __A, __m256i __B, __m256i __C,
+							__m256i __D)
+{
+  return (__m256i)__builtin_ia32_vpdpwssds_v8si_maskz ((__v8si)__B,
+				(__v8si) __C, (__v8si) __D, (__mmask8)__A);
+}
+
+extern __inline __m128i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_dpwssds_epi32 (__m128i __A, __m128i __B, __m128i __C)
+{
+  return (__m128i) __builtin_ia32_vpdpwssds_v4si ((__v4si)__A, (__v4si) __B,
+								(__v4si) __C);
+}
+
+extern __inline __m128i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_mask_dpwssds_epi32 (__m128i __A, __mmask8 __B, __m128i __C, __m128i __D)
+{
+  return (__m128i)__builtin_ia32_vpdpwssds_v4si_mask ((__v4si)__A,
+				(__v4si) __C, (__v4si) __D, (__mmask8)__B);
+}
+
+extern __inline __m128i
+__attribute__((__gnu_inline__, __always_inline__, __artificial__))
+_mm_maskz_dpwssds_epi32 (__mmask8 __A, __m128i __B, __m128i __C, __m128i __D)
+{
+  return (__m128i)__builtin_ia32_vpdpwssds_v4si_maskz ((__v4si)__B,
+				(__v4si) __C, (__v4si) __D, (__mmask8)__A);
+}
 #ifdef __DISABLE_AVX512VNNIVL__
 #undef __DISABLE_AVX512VNNIVL__
 #pragma GCC pop_options

gcc/testsuite/ChangeLog

+2017-12-08  Julia Koval  <julia.koval@intel.com>
+
+	* gcc.target/i386/avx512f-vnni-1.c: Add checks for vdpdwssds.
+	* gcc.target/i386/avx512vl-vnni-1.c: Ditto.
+	* gcc.target/i386/avx512f-vpdpwssds-2.c: New test.
+	* gcc.target/i386/avx512vl-vpdpwssds-2.c: Ditto.
+
 2017-12-08  Richard Biener  <rguenther@suse.de>
 
 	PR tree-optimization/81303

gcc/testsuite/gcc.target/i386/avx512f-vnni-1.c

@@ -9,6 +9,9 @@
 /* { 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 } } */
+/* { dg-final { scan-assembler-times "vpdpwssds\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\n\r]*%zmm\[0-9\]+\[^\n\r]*%zmm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
+/* { dg-final { scan-assembler-times "vpdpwssds\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\\n\\r]*%zmm\[0-9\]+\[^\\n\\r\]*%zmm\[0-9\]+\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */
+/* { dg-final { scan-assembler-times "vpdpwssds\[ \\t\]+\[^\{\n\]*%zmm\[0-9\]+\[^\\n\\r]*%zmm\[0-9\]+\[^\\n\\r\]*%zmm\[0-9\]+\{%k\[1-7\]\}\{z\}(?:\n|\[ \\t\]+#)" 1 } } */
 
 #include <immintrin.h>
 
@@ -32,4 +35,7 @@ avx512f_test (void)
   x = _mm512_mask_dpwssd_epi32 (x, m16, y, z);
   x = _mm512_maskz_dpwssd_epi32 (m16, x, y, z);
 
+  x = _mm512_dpwssds_epi32 (x, y, z);
+  x = _mm512_mask_dpwssds_epi32 (x, m16, y, z);
+  x = _mm512_maskz_dpwssds_epi32 (m16, x, y, z);
 }

gcc/testsuite/gcc.target/i386/avx512f-vpdpwssds-2.c

+/* { 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];
+    long long max_int = 0x7FFFFFFF;
+    if (test > max_int)
+      test = 0x7FFFFFFF;
+    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 (_dpwssds_epi32) (res1.x, src1.x, src2.x);
+  res2.x = INTRINSIC (_mask_dpwssds_epi32) (res2.x, mask, src1.x, src2.x);
+  res3.x = INTRINSIC (_maskz_dpwssds_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 ();
+}

gcc/testsuite/gcc.target/i386/avx512vl-vnni-1.c

@@ -18,6 +18,12 @@
 /* { 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 } } */
+/* { dg-final { scan-assembler-times "vpdpwssds\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\n\r]*%ymm\[0-9\]+\[^\n\r]*%ymm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
+/* { dg-final { scan-assembler-times "vpdpwssds\[ \\t\]+\[^\{\n\]*%ymm\[0-9\]+\[^\\n\\r]*%ymm\[0-9\]+\[^\\n\\r\]*%ymm\[0-9\]+\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */
+/* { dg-final { scan-assembler-times "vpdpwssds\[ \\t\]+\[^\{\n\]*%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 "vpdpwssds\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+\[^\n\r]*%xmm\[0-9\]+(?:\n|\[ \\t\]+#)" 1 } } */
+/* { dg-final { scan-assembler-times "vpdpwssds\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\\n\\r]*%xmm\[0-9\]+\[^\\n\\r\]*%xmm\[0-9\]+\{%k\[1-7\]\}(?:\n|\[ \\t\]+#)" 1 } } */
+/* { dg-final { scan-assembler-times "vpdpwssds\[ \\t\]+\[^\{\n\]*%xmm\[0-9\]+\[^\\n\\r]*%xmm\[0-9\]+\[^\\n\\r\]*%xmm\[0-9\]+\{%k\[1-7\]\}\{z\}(?:\n|\[ \\t\]+#)" 1 } } */
 
 
 #include <immintrin.h>
@@ -53,4 +59,11 @@ avx512f_test (void)
   x_ = _mm_mask_dpwssd_epi32 (x_, m, y_, z_);
   x_ = _mm_maskz_dpwssd_epi32 (m, x_, y_, z_);
 
+  x = _mm256_dpwssds_epi32 (x, y, z);
+  x = _mm256_mask_dpwssds_epi32 (x, m, y, z);
+  x = _mm256_maskz_dpwssds_epi32 (m, x, y, z);
+
+  x_ = _mm_dpwssds_epi32 (x_, y_, z_);
+  x_ = _mm_mask_dpwssds_epi32 (x_, m, y_, z_);
+  x_ = _mm_maskz_dpwssds_epi32 (m, x_, y_, z_);
 }

gcc/testsuite/gcc.target/i386/avx512vl-vpdpwssds-2.c

+/* { 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-vpdpwssds-2.c"
+
+#undef AVX512F_LEN
+#undef AVX512F_LEN_HALF
+
+#define AVX512F_LEN 128
+#define AVX512F_LEN_HALF 128
+#include "avx512f-vpdpwssds-2.c"