diff options
| author | Juzhe-Zhong <juzhe.zhong@rivai.ai> | 2023-12-14 21:45:59 +0800 |
|---|---|---|
| committer | Pan Li <pan2.li@intel.com> | 2023-12-15 08:26:45 +0800 |
| commit | c7ef2189855a8cf12427a778cd5a31d42ddf6260 (patch) | |
| tree | 6242fc448fcc3a15a09d77424d2871072594743f /libgomp/target.c | |
| parent | f998335ac0aa56261789282cba5ae1286ddf3a1f (diff) | |
| download | gcc-c7ef2189855a8cf12427a778cd5a31d42ddf6260.zip gcc-c7ef2189855a8cf12427a778cd5a31d42ddf6260.tar.gz gcc-c7ef2189855a8cf12427a778cd5a31d42ddf6260.tar.bz2 | |
Middle-end: Do not model address cost for SELECT_VL style vectorization
Follow Richard's suggestions, we should not model address cost in the loop
vectorizer for select_vl or decrement IV since other style vectorization doesn't
do that.
To make cost model comparison apple to apple.
This patch set COST from 2 to 1 which turns out have better codegen
in various codegen for RVV.
Ok for trunk ?
PR target/111153
gcc/ChangeLog:
* tree-vect-loop.cc (vect_estimate_min_profitable_iters):
Remove address cost for select_vl/decrement IV.
gcc/testsuite/ChangeLog:
* gcc.dg/vect/costmodel/riscv/rvv/pr111153.c: Moved to...
* gcc.dg/vect/costmodel/riscv/rvv/pr11153-2.c: ...here.
* gcc.dg/vect/costmodel/riscv/rvv/pr111153-1.c: New test.
Diffstat (limited to 'libgomp/target.c')
0 files changed, 0 insertions, 0 deletions