diff options
| author | Bu Le <bule1@huawei.com> | 2020-03-12 22:39:12 +0000 |
|---|---|---|
| committer | Richard Sandiford <richard.sandiford@arm.com> | 2020-03-13 09:18:40 +0000 |
| commit | dbf3dc75888623e9d4bb7cc5e9c30caa9b24ffe7 (patch) | |
| tree | fa9b5262372b71530d61a6d4db02c46be6776f96 /gcc/tree-ssa-pre.c | |
| parent | 3e6ab5cefa81165e90fb62abf50e515f85a17e9a (diff) | |
| download | gcc-dbf3dc75888623e9d4bb7cc5e9c30caa9b24ffe7.zip gcc-dbf3dc75888623e9d4bb7cc5e9c30caa9b24ffe7.tar.gz gcc-dbf3dc75888623e9d4bb7cc5e9c30caa9b24ffe7.tar.bz2 | |
aarch64: Add --params to control the number of recip steps [PR94154]
-mlow-precision-div hard-coded the number of iterations to 2 for double
and 1 for float. This patch adds a --param to control the number.
2020-03-13 Bu Le <bule1@huawei.com>
gcc/
PR target/94154
* config/aarch64/aarch64.opt (-param=aarch64-float-recp-precision=)
(-param=aarch64-double-recp-precision=): New options.
* doc/invoke.texi: Document them.
* config/aarch64/aarch64.c (aarch64_emit_approx_div): Use them
instead of hard-coding the choice of 1 for float and 2 for double.
Diffstat (limited to 'gcc/tree-ssa-pre.c')
0 files changed, 0 insertions, 0 deletions