diff options
| author | chenxiaolong <chenxiaolong@loongson.cn> | 2025-01-07 21:04:51 +0800 |
|---|---|---|
| committer | Lulu Cheng <chenglulu@loongson.cn> | 2025-01-10 10:02:56 +0800 |
| commit | e8a57884ad4898fdec5c13a8933d73bcbaf06099 (patch) | |
| tree | a6a30e97a0fe08683afe9cd4660d6034d271e2e2 /gcc/c | |
| parent | 979ca3ba366da7177f427e049f67673ec3e35442 (diff) | |
| download | gcc-e8a57884ad4898fdec5c13a8933d73bcbaf06099.zip gcc-e8a57884ad4898fdec5c13a8933d73bcbaf06099.tar.gz gcc-e8a57884ad4898fdec5c13a8933d73bcbaf06099.tar.bz2 | |
LoongArch: Opitmize the cost of vec_construct.
When analyzing 525 on LoongArch architecture, it was found that the
for loop of hotspot function x264_pixel_satd_8x4 could not be quantized
256-bit due to the cost of vec_construct setting. After re-adjusting
vec_construct, the performance of 525 program was improved by 16.57%.
It was found that this function can be vectorized on the aarch64 and
x86 architectures, see [PR98138].
Co-Authored-By: Deng Jianbo <dengjianbo@loongson.cn>.
gcc/ChangeLog:
* config/loongarch/loongarch.cc
(loongarch_builtin_vectorization_cost): Modify the
construction cost of the vec_construct vector.
gcc/testsuite/ChangeLog:
* gcc.target/loongarch/vect-slp-two-operator.c: New test.
Diffstat (limited to 'gcc/c')
0 files changed, 0 insertions, 0 deletions