Re:[pushed] [PATCH v3] LoongArch: Define LOGICAL_OP_NON_SHORT_CIRCUIT

2024-01-26 Thread chenglulu
Pushed to r14-8446. 在 2024/1/16 上午10:32, Jiahao Xu 写道: Define LOGICAL_OP_NON_SHORT_CIRCUIT as 0, for a short-circuit branch, use the short-circuit operation instead of the non-short-circuit operation. SPEC2017 performance evaluation shows 1% performance improvement for fprate GEOMEAN and no obv

Re: [PATCH v3] LoongArch: Define LOGICAL_OP_NON_SHORT_CIRCUIT

2024-01-19 Thread Jiahao Xu
The test case gcc.dg/tree-ssa/copy-headers-8.c fails for a target where LOGICAL_OP_NON_SHORT_CIRCUIT is defined as 0.It is suggested to add `--param logical-op-non-short-circuit=1` to the test case to make it a target-independent testcase. see https://gcc.gnu.org/bugzilla/show_bug.cgi?id=11124

Re: [PATCH v3] LoongArch: Define LOGICAL_OP_NON_SHORT_CIRCUIT

2024-01-19 Thread chenglulu
Hi, Jiahao: This patch will introduce redundant FAIL, and the reason needs to be explained. +FAIL: gcc.dg/tree-ssa/copy-headers-8.c scan-tree-dump-times ch2 "Conditional combines static and invariant" 1 +FAIL: gcc.dg/tree-ssa/copy-headers-8.c scan-tree-dump-times ch2 "Will duplicate bb" 2 +FAI

[PATCH v3] LoongArch: Define LOGICAL_OP_NON_SHORT_CIRCUIT

2024-01-15 Thread Jiahao Xu
Define LOGICAL_OP_NON_SHORT_CIRCUIT as 0, for a short-circuit branch, use the short-circuit operation instead of the non-short-circuit operation. SPEC2017 performance evaluation shows 1% performance improvement for fprate GEOMEAN and no obvious regression for others. Especially, 526.blender_r +10.