diff options
| author | Xi Ruoyao <xry111@xry111.site> | 2023-10-17 21:55:05 +0800 |
|---|---|---|
| committer | Xi Ruoyao <xry111@xry111.site> | 2023-10-18 16:25:05 +0800 |
| commit | b588dcb77e96d77777eb5647cba9e8f454e314dc (patch) | |
| tree | 7a965d4f8df0da81174055db3c470a5d53d649c9 | |
| parent | b0372ef12f6f7f863606b0d2858d049b18a4cfda (diff) | |
| download | gcc-b588dcb77e96d77777eb5647cba9e8f454e314dc.zip gcc-b588dcb77e96d77777eb5647cba9e8f454e314dc.tar.gz gcc-b588dcb77e96d77777eb5647cba9e8f454e314dc.tar.bz2 | |
LoongArch: Use fcmp.caf.s instead of movgr2cf for zeroing a fcc
During the review of an LLVM change [1], on LA464 we found that zeroing
an fcc with fcmp.caf.s is much faster than a movgr2cf from $r0.
[1]: https://github.com/llvm/llvm-project/pull/69300
gcc/ChangeLog:
* config/loongarch/loongarch.md (movfcc): Use fcmp.caf.s for
zeroing a fcc.
| -rw-r--r-- | gcc/config/loongarch/loongarch.md | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/gcc/config/loongarch/loongarch.md b/gcc/config/loongarch/loongarch.md index 3286b0c..9f5a753 100644 --- a/gcc/config/loongarch/loongarch.md +++ b/gcc/config/loongarch/loongarch.md @@ -2150,7 +2150,7 @@ [(set (match_operand:FCC 0 "register_operand" "=z") (const_int 0))] "" - "movgr2cf\t%0,$r0") + "fcmp.caf.s\t%0,$f0,$f0") ;; Conditional move instructions. |