diff options
| author | Di Zhao OS <dizhao@os.amperecomputing.com> | 2023-06-22 15:16:57 +0100 |
|---|---|---|
| committer | Richard Sandiford <richard.sandiford@arm.com> | 2023-06-22 15:16:57 +0100 |
| commit | 4ced8363622b31910cda61796a28fe2cbf70faa7 (patch) | |
| tree | 23ed7a7220f0b712a88e942ab6a2fe2e6d6f67b4 /gcc | |
| parent | 2cd0689a79498dcaaadc8cc5c1c4d0a452a4fb09 (diff) | |
| download | gcc-4ced8363622b31910cda61796a28fe2cbf70faa7.zip gcc-4ced8363622b31910cda61796a28fe2cbf70faa7.tar.gz gcc-4ced8363622b31910cda61796a28fe2cbf70faa7.tar.bz2 | |
Change fma_reassoc_width tuning for ampere1
This patch enables reassociation of floating-point additions on ampere1.
This brings about 1% overall benefit on spec2017 fprate cases. (There
are minor regressions in 510.parest_r and 508.namd_r, analyzed here:
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110279 .)
gcc/ChangeLog:
* config/aarch64/aarch64.cc: Change fma_reassoc_width for ampere1.
Diffstat (limited to 'gcc')
| -rw-r--r-- | gcc/config/aarch64/aarch64.cc | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/gcc/config/aarch64/aarch64.cc b/gcc/config/aarch64/aarch64.cc index b99f12c..644ebde 100644 --- a/gcc/config/aarch64/aarch64.cc +++ b/gcc/config/aarch64/aarch64.cc @@ -1929,7 +1929,7 @@ static const struct tune_params ampere1_tunings = "32:16", /* loop_align. */ 2, /* int_reassoc_width. */ 4, /* fp_reassoc_width. */ - 1, /* fma_reassoc_width. */ + 4, /* fma_reassoc_width. */ 2, /* vec_reassoc_width. */ 2, /* min_div_recip_mul_sf. */ 2, /* min_div_recip_mul_df. */ |