On Tue, 2 Feb 2021, Jakub Jelinek wrote:
> Hi!
>
> As discussed in the PR, the reduction code isn't able to cope with type
> promotions/demotions in the reduction computation, so if we recognize an
> over-widening pattern that has vect_reduction_def type, we most likely make
> it non-vectorizable.
>
> Fixed thusly, bootstrapped/regtested on x86_64-linux and i686-linux, ok for
> turnk?
OK.
Thanks,
Richard.
> 2021-02-02 Jakub Jelinek <ja...@redhat.com>
>
> PR tree-optimization/98848
> * tree-vect-patterns.c (vect_recog_over_widening_pattern): Punt if
> STMT_VINFO_DEF_TYPE (last_stmt_info) is vect_reduction_def.
>
> * gcc.dg/vect/pr98848.c: New test.
> * gcc.dg/vect/pr92205.c: Remove xfail.
>
> --- gcc/tree-vect-patterns.c.jj 2021-01-04 10:25:38.650235896 +0100
> +++ gcc/tree-vect-patterns.c 2021-02-01 11:04:48.136814236 +0100
> @@ -1579,6 +1579,10 @@ vect_recog_over_widening_pattern (vec_in
> tree type = TREE_TYPE (lhs);
> tree_code code = gimple_assign_rhs_code (last_stmt);
>
> + /* Punt for reductions where we don't handle the type conversions. */
> + if (STMT_VINFO_DEF_TYPE (last_stmt_info) == vect_reduction_def)
> + return NULL;
> +
> /* Keep the first operand of a COND_EXPR as-is: only the other two
> operands are interesting. */
> unsigned int first_op = (code == COND_EXPR ? 2 : 1);
> --- gcc/testsuite/gcc.dg/vect/pr98848.c.jj 2021-02-01 11:15:11.965633750
> +0100
> +++ gcc/testsuite/gcc.dg/vect/pr98848.c 2021-02-01 11:15:07.348686891
> +0100
> @@ -0,0 +1,18 @@
> +/* PR tree-optimization/98848 */
> +/* { dg-do compile } */
> +/* { dg-require-effective-target vect_int } */
> +
> +short a[9000];
> +
> +int
> +foo (void)
> +{
> + int b = a[0];
> + int i;
> + for (i = 1; i < 9000; i ++)
> + if (a[i] < b)
> + b = a[i];
> + return b;
> +}
> +
> +/* { dg-final { scan-tree-dump-times "vectorized 1 loop" 1 "vect" { xfail {
> vect_no_int_add || vect_no_int_min_max } } } } */
> --- gcc/testsuite/gcc.dg/vect/pr92205.c.jj 2020-01-12 11:54:37.645394821
> +0100
> +++ gcc/testsuite/gcc.dg/vect/pr92205.c 2021-02-02 09:59:40.998104863
> +0100
> @@ -10,4 +10,4 @@ int b(int n, unsigned char *a)
> return d;
> }
>
> -/* { dg-final { scan-tree-dump "vectorized 1 loops" "vect" { xfail *-*-* } }
> } */
> +/* { dg-final { scan-tree-dump "vectorized 1 loops" "vect" { target {
> vect_unpack && { ! vect_no_bitwise } } } } } */
>
> Jakub
>
>
--
Richard Biener <rguent...@suse.de>
SUSE Software Solutions Germany GmbH, Maxfeldstrasse 5, 90409 Nuernberg,
Germany; GF: Felix Imendörffer; HRB 36809 (AG Nuernberg)
