tree-optimization/104152 - add missing check for SLP constant build

This adds a missing check to verify we can actually build an
invariant vector from components when SLP vectorizing an associatable
chain.

2022-01-21  Richard Biener  <rguenther@suse.de>

	PR tree-optimization/104152
	* tree-vect-slp.cc (vect_build_slp_tree_2): Add missing
	can_duplicate_and_interleave_p check.

	* gcc.dg/vect/pr104152.c: New testcase.
This commit is contained in:
Richard Biener 2022-01-21 09:48:33 +01:00
parent 3e0b19f1d6
commit 1ad7281124
2 changed files with 51 additions and 2 deletions

View File

@ -0,0 +1,40 @@
/* { dg-do compile } */
/* { dg-additional-options "-Ofast" } */
/* { dg-additional-options "-march=armv8.2-a+sve" { target aarch64-*-* } } */
#define M_PI 3.14f
#define NK 24
#define DIM 3
typedef float rvec_ [3];
static rvec_ v0[NK] = {
{ 1, 0, 0 }, { 0, 1, 0 }, { 0, 0, 1 }, { 1, 1, 0 }, { 1, -1, 0 },
{ 1, 0, 1 }, { 1, 0, -1 }, { 0, 1, 1 }, { 0, 1, -1 }, { 1, 1, 1 },
{ 1, 1, -1 }, { 1, -1, 1 }, { -1, 1, 1 }, { 2, 0, 0 }, { 0, 2, 0 },
{ 0, 0, 2 }, { 3, 0, 0 }, { 0, 3, 0 }, { 0, 0, 3 }, { 4, 0, 0 },
{ 0, 4, 0 }, { 0, 0, 4 } };
static inline float iprod__(const rvec_ a, const rvec_ b)
{
return (a[0] * b[0] + a[1] * b[1] + a[2] * b[2]);
}
int badaboum(rvec_ cm_mol, float **tc)
{
float sx;
int k, d;
rvec_ kfac[3];
for (k = 0; k < DIM; k++)
for (d = 0; d < DIM; d++)
kfac[k][d] = M_PI * v0[k][d] / v0[d][d];
for (k = 0; k < DIM; k++)
{
sx = __builtin_sinf(iprod__(kfac[k], cm_mol));
tc[k][0] += sx * iprod__(v0[k], cm_mol);
}
return 0;
}

View File

@ -1926,8 +1926,17 @@ vect_build_slp_tree_2 (vec_info *vinfo, slp_tree node,
if (dt == vect_constant_def
|| dt == vect_external_def)
{
/* We can always build those. Might want to sort last
or defer building. */
/* Check whether we can build the invariant. If we can't
we never will be able to. */
tree type = TREE_TYPE (chains[0][n].op);
if (!GET_MODE_SIZE (vinfo->vector_mode).is_constant ()
&& (TREE_CODE (type) == BOOLEAN_TYPE
|| !can_duplicate_and_interleave_p (vinfo, group_size,
type)))
{
matches[0] = false;
goto out;
}
vec<tree> ops;
ops.create (group_size);
for (lane = 0; lane < group_size; ++lane)