re PR fortran/80975 (matmul for zero-length arrays)

2017-06-06  Thomas Koenig  <tkoenig@gcc.gnu.org>

	PR fortran/80975
	* m4/matmul_internal.m4:  Move zeroing before early return.
	* generated/matmul_c10.c: Regenerated.
	* generated/matmul_c16.c: Regenerated.
	* generated/matmul_c4.c: Regenerated.
	* generated/matmul_c8.c: Regenerated.
	* generated/matmul_i1.c: Regenerated.
	* generated/matmul_i16.c: Regenerated.
	* generated/matmul_i2.c: Regenerated.
	* generated/matmul_i4.c: Regenerated.
	* generated/matmul_i8.c: Regenerated.
	* generated/matmul_r10.c: Regenerated.
	* generated/matmul_r16.c: Regenerated.
	* generated/matmul_r4.c: Regenerated.
	* generated/matmul_r8.c: Regenerated.
	* generated/matmulavx128_c10.c: Regenerated.
	* generated/matmulavx128_c16.c: Regenerated.
	* generated/matmulavx128_c4.c: Regenerated.
	* generated/matmulavx128_c8.c: Regenerated.
	* generated/matmulavx128_i1.c: Regenerated.
	* generated/matmulavx128_i16.c: Regenerated.
	* generated/matmulavx128_i2.c: Regenerated.
	* generated/matmulavx128_i4.c: Regenerated.
	* generated/matmulavx128_i8.c: Regenerated.
	* generated/matmulavx128_r10.c: Regenerated.
	* generated/matmulavx128_r16.c: Regenerated.
	* generated/matmulavx128_r4.c: Regenerated.
	* generated/matmulavx128_r8.c: Regenerated.

2017-06-06  Thomas Koenig  <tkoenig@gcc.gnu.org>

	PR fortran/80975
	* gfortran.dg/matmul_16.f90: New test.
	* gfortran.dg/inline_matmul_18.f90: New test.

From-SVN: r248932

1 files changed, 25 insertions, 25 deletions

diff --git a/libgfortran/generated/matmul_r4.c b/libgfortran/generated/matmul_r4.c
index 52eea532a9b..5d13e6713d6 100644
--- a/libgfortran/generated/matmul_r4.c
+++ b/libgfortran/generated/matmul_r4.c
@@ -307,6 +307,11 @@ matmul_r4_avx (gfc_array_r4 * const restrict retarray,
       b_offset = 1 + b_dim1;
       b -= b_offset;
 
+      /* Empty c first.  */
+      for (j=1; j<=n; j++)
+	for (i=1; i<=m; i++)
+	  c[i + j * c_dim1] = (GFC_REAL_4)0;
+
       /* Early exit if possible */
       if (m == 0 || n == 0 || k == 0)
 	return;
@@ -319,11 +324,6 @@ matmul_r4_avx (gfc_array_r4 * const restrict retarray,
 
       t1 = malloc (t1_dim * sizeof(GFC_REAL_4));
 
-      /* Empty c first.  */
-      for (j=1; j<=n; j++)
-	for (i=1; i<=m; i++)
-	  c[i + j * c_dim1] = (GFC_REAL_4)0;
-
       /* Start turning the crank. */
       i1 = n;
       for (jj = 1; jj <= i1; jj += 512)
@@ -859,6 +859,11 @@ matmul_r4_avx2 (gfc_array_r4 * const restrict retarray,
       b_offset = 1 + b_dim1;
       b -= b_offset;
 
+      /* Empty c first.  */
+      for (j=1; j<=n; j++)
+	for (i=1; i<=m; i++)
+	  c[i + j * c_dim1] = (GFC_REAL_4)0;
+
       /* Early exit if possible */
       if (m == 0 || n == 0 || k == 0)
 	return;
@@ -871,11 +876,6 @@ matmul_r4_avx2 (gfc_array_r4 * const restrict retarray,
 
       t1 = malloc (t1_dim * sizeof(GFC_REAL_4));
 
-      /* Empty c first.  */
-      for (j=1; j<=n; j++)
-	for (i=1; i<=m; i++)
-	  c[i + j * c_dim1] = (GFC_REAL_4)0;
-
       /* Start turning the crank. */
       i1 = n;
       for (jj = 1; jj <= i1; jj += 512)
@@ -1411,6 +1411,11 @@ matmul_r4_avx512f (gfc_array_r4 * const restrict retarray,
       b_offset = 1 + b_dim1;
       b -= b_offset;
 
+      /* Empty c first.  */
+      for (j=1; j<=n; j++)
+	for (i=1; i<=m; i++)
+	  c[i + j * c_dim1] = (GFC_REAL_4)0;
+
       /* Early exit if possible */
       if (m == 0 || n == 0 || k == 0)
 	return;
@@ -1423,11 +1428,6 @@ matmul_r4_avx512f (gfc_array_r4 * const restrict retarray,
 
       t1 = malloc (t1_dim * sizeof(GFC_REAL_4));
 
-      /* Empty c first.  */
-      for (j=1; j<=n; j++)
-	for (i=1; i<=m; i++)
-	  c[i + j * c_dim1] = (GFC_REAL_4)0;
-
       /* Start turning the crank. */
       i1 = n;
       for (jj = 1; jj <= i1; jj += 512)
@@ -1977,6 +1977,11 @@ matmul_r4_vanilla (gfc_array_r4 * const restrict retarray,
       b_offset = 1 + b_dim1;
       b -= b_offset;
 
+      /* Empty c first.  */
+      for (j=1; j<=n; j++)
+	for (i=1; i<=m; i++)
+	  c[i + j * c_dim1] = (GFC_REAL_4)0;
+
       /* Early exit if possible */
       if (m == 0 || n == 0 || k == 0)
 	return;
@@ -1989,11 +1994,6 @@ matmul_r4_vanilla (gfc_array_r4 * const restrict retarray,
 
       t1 = malloc (t1_dim * sizeof(GFC_REAL_4));
 
-      /* Empty c first.  */
-      for (j=1; j<=n; j++)
-	for (i=1; i<=m; i++)
-	  c[i + j * c_dim1] = (GFC_REAL_4)0;
-
       /* Start turning the crank. */
       i1 = n;
       for (jj = 1; jj <= i1; jj += 512)
@@ -2603,6 +2603,11 @@ matmul_r4 (gfc_array_r4 * const restrict retarray,
       b_offset = 1 + b_dim1;
       b -= b_offset;
 
+      /* Empty c first.  */
+      for (j=1; j<=n; j++)
+	for (i=1; i<=m; i++)
+	  c[i + j * c_dim1] = (GFC_REAL_4)0;
+
       /* Early exit if possible */
       if (m == 0 || n == 0 || k == 0)
 	return;
@@ -2615,11 +2620,6 @@ matmul_r4 (gfc_array_r4 * const restrict retarray,
 
       t1 = malloc (t1_dim * sizeof(GFC_REAL_4));
 
-      /* Empty c first.  */
-      for (j=1; j<=n; j++)
-	for (i=1; i<=m; i++)
-	  c[i + j * c_dim1] = (GFC_REAL_4)0;
-
       /* Start turning the crank. */
       i1 = n;
       for (jj = 1; jj <= i1; jj += 512)