diff options
| author | Matthew Malcomson <matthew.malcomson@arm.com> | 2023-07-26 16:53:18 +0100 |
|---|---|---|
| committer | Matthew Malcomson <matthew.malcomson@arm.com> | 2023-07-26 16:53:18 +0100 |
| commit | d74aed3d7aefbb661ae629a91ca7baa4916085f0 (patch) | |
| tree | 7cebe2f6eeae35570513316df2000de6c2982fa3 /gcc | |
| parent | b51795c832cf6e724d61919eb18a383223b76694 (diff) | |
| download | gcc-d74aed3d7aefbb661ae629a91ca7baa4916085f0.zip gcc-d74aed3d7aefbb661ae629a91ca7baa4916085f0.tar.gz gcc-d74aed3d7aefbb661ae629a91ca7baa4916085f0.tar.bz2 | |
[committed] Add check_vect in a testcase
Also reformat a comment that had too long lines.
Commit c5bd0e5870a introduced both these problems to fix.
Committed as obvious after ensuring that when running the testcase on
AArch64 it still fails before the original c5bd0e5870a commit and passes
after it.
gcc/ChangeLog:
* tree-vect-stmts.cc (get_group_load_store_type): Reformat
comment.
gcc/testsuite/ChangeLog:
* gcc.dg/vect/vect-multi-peel-gaps.c: Add `check_vect` call into
`main` of this testcase.
Diffstat (limited to 'gcc')
| -rw-r--r-- | gcc/testsuite/gcc.dg/vect/vect-multi-peel-gaps.c | 2 | ||||
| -rw-r--r-- | gcc/tree-vect-stmts.cc | 5 |
2 files changed, 5 insertions, 2 deletions
diff --git a/gcc/testsuite/gcc.dg/vect/vect-multi-peel-gaps.c b/gcc/testsuite/gcc.dg/vect/vect-multi-peel-gaps.c index 1aab4c5..7e7db3c 100644 --- a/gcc/testsuite/gcc.dg/vect/vect-multi-peel-gaps.c +++ b/gcc/testsuite/gcc.dg/vect/vect-multi-peel-gaps.c @@ -9,6 +9,7 @@ /* { dg-require-effective-target mmap } */ #include <sys/mman.h> #include <stdio.h> +#include "tree-vect.h" #define MMAP_SIZE 0x20000 #define ADDRESS 0x1122000000 @@ -24,6 +25,7 @@ void initialise_s(int *s) { } int main() { void *s_mapping; void *end_s; + check_vect (); s_mapping = mmap ((void *)ADDRESS, MMAP_SIZE, PROT_READ | PROT_WRITE, MAP_PRIVATE | MAP_ANONYMOUS, -1, 0); if (s_mapping == MAP_FAILED) diff --git a/gcc/tree-vect-stmts.cc b/gcc/tree-vect-stmts.cc index 99e61f8..5018bd2 100644 --- a/gcc/tree-vect-stmts.cc +++ b/gcc/tree-vect-stmts.cc @@ -2214,8 +2214,9 @@ get_group_load_store_type (vec_info *vinfo, stmt_vec_info stmt_info, we can end up with no gap recorded but still excess elements accessed, see PR103116. Make sure we peel for gaps if necessary and sufficient and give up if not. - If there is a combination of the access not covering the full vector and - a gap recorded then we may need to peel twice. */ + + If there is a combination of the access not covering the full + vector and a gap recorded then we may need to peel twice. */ if (loop_vinfo && *memory_access_type == VMAT_CONTIGUOUS && SLP_TREE_LOAD_PERMUTATION (slp_node).exists () |