diff options
| author | Robin Dapp <rdapp@ventanamicro.com> | 2025-10-29 16:02:51 +0100 |
|---|---|---|
| committer | Robin Dapp <rdapp@ventanamicro.com> | 2025-11-07 21:09:45 +0100 |
| commit | 573ea59a48d95fc4e9f520865eae71c5d9de614f (patch) | |
| tree | 151de219d15e7a5bfc2cf8138aefa5b613900912 /libjava/java/security/Engine.java | |
| parent | e51fffdcbb53f10acffd197379585a586720a9b3 (diff) | |
| download | gcc-573ea59a48d95fc4e9f520865eae71c5d9de614f.zip gcc-573ea59a48d95fc4e9f520865eae71c5d9de614f.tar.gz gcc-573ea59a48d95fc4e9f520865eae71c5d9de614f.tar.bz2 | |
vect: Relax gather/scatter scale handling.
Similar to the signed/unsigned patch before this one relaxes the
gather/scatter restrictions on scale factors. The basic idea is that a
natively unsupported scale factor can still be reached by emitting a
multiplication before the actual gather operation. As before, we need
to make sure that there is no overflow when multiplying.
gcc/ChangeLog:
* tree-vect-data-refs.cc (struct gather_scatter_config):
Add scale.
(vect_gather_scatter_get_configs): Try various scales.
(vect_gather_scatter_fn_p): Add scale handling.
(vect_check_gather_scatter): Add scale parameter.
* tree-vect-stmts.cc (check_load_store_for_partial_vectors):
Ditto.
(vect_truncate_gather_scatter_offset): Ditto.
(vect_use_grouped_gather): Ditto.
(get_load_store_type): Ditto.
(vectorizable_store): Scale offset if necessary.
(vectorizable_load): Ditto.
* tree-vectorizer.h (struct vect_load_store_data): Add
supported_scale.
(vect_gather_scatter_fn_p): Add argument.
Diffstat (limited to 'libjava/java/security/Engine.java')
0 files changed, 0 insertions, 0 deletions
