aboutsummaryrefslogtreecommitdiff
path: root/libgfortran/m4
diff options
context:
space:
mode:
authorThomas Koenig <tkoenig@gcc.gnu.org>2017-05-08 17:56:13 +0000
committerThomas Koenig <tkoenig@gcc.gnu.org>2017-05-08 17:56:13 +0000
commitfd9910392bc2854ecc5c91fd6f089fa4e66edb44 (patch)
tree644a36f58118c3fd22c9a3fc368ec218e1e7d524 /libgfortran/m4
parentd16d5eac2a0132adcd9e88ae288bedd03abfc513 (diff)
re PR libfortran/80602 (Reduce stack usage for blocked matmul)
2017-05-08 Thomas Koenig <tkoenig@gcc.gnu.org> PR fortran/80602 * m4/matmul_internal.m4: 'matmul_name`: Change t1 to a VLA of the required size. * 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. 2017-05-08 Thomas Koenig <tkoenig@gcc.gnu.org> PR fortran/80602 * gfortran.dg/matmul_15.f90: New test case. From-SVN: r247753
Diffstat (limited to 'libgfortran/m4')
-rw-r--r--libgfortran/m4/matmul_internal.m414
1 files changed, 12 insertions, 2 deletions
diff --git a/libgfortran/m4/matmul_internal.m4 b/libgfortran/m4/matmul_internal.m4
index d35968b3be5..37839567168 100644
--- a/libgfortran/m4/matmul_internal.m4
+++ b/libgfortran/m4/matmul_internal.m4
@@ -202,8 +202,7 @@ sinclude(`matmul_asm_'rtype_code`.m4')dnl
i1, i2, i3, i4, i5, i6;
/* Local variables */
- 'rtype_name` t1[65536], /* was [256][256] */
- f11, f12, f21, f22, f31, f32, f41, f42,
+ 'rtype_name` f11, f12, f21, f22, f31, f32, f41, f42,
f13, f14, f23, f24, f33, f34, f43, f44;
index_type i, j, l, ii, jj, ll;
index_type isec, jsec, lsec, uisec, ujsec, ulsec;
@@ -227,6 +226,17 @@ sinclude(`matmul_asm_'rtype_code`.m4')dnl
if (m == 0 || n == 0 || k == 0)
return;
+ /* Adjust size of t1 to what is needed. */
+ index_type t1_dim;
+ t1_dim = (a_dim1-1) * 256 + b_dim1;
+ if (t1_dim > 65536)
+ t1_dim = 65536;
+
+#pragma GCC diagnostic push
+#pragma GCC diagnostic ignored "-Wvla"
+ 'rtype_name` t1[t1_dim]; /* was [256][256] */
+#pragma GCC diagnostic pop
+
/* Empty c first. */
for (j=1; j<=n; j++)
for (i=1; i<=m; i++)