| commit | 1d9f99ce1b3788d1897cb53a76d57e973111b8fe | |
| author | Naohiro Tamura <naohirot@fujitsu.com> | |
| Fri, 27 Aug 2021 05:03:04 +0000 (27 05:03 +0000) | ||
| committer | Szabolcs Nagy <szabolcs.nagy@arm.com> | |
| Mon, 6 Sep 2021 09:23:24 +0000 (6 10:23 +0100) | ||
| tree | 684c81cc6e88650313797fadaa642d714fcce8a8 | treesnapshot (tar.gz zip) |
| parent | f873adf3df443f8d302677f963adcc3c22187e68 | commitdiff |
AArch64: Update A64FX memset not to degrade at 16KB
This patch updates unroll8 code so as not to degrade at the peak
performance 16KB for both FX1000 and FX700.
Inserted 2 instructions at the beginning of the unroll8 loop,
cmp and branch, are a workaround that is found heuristically.
Reviewed-by: Wilco Dijkstra <Wilco.Dijkstra@arm.com>
This patch updates unroll8 code so as not to degrade at the peak
performance 16KB for both FX1000 and FX700.
Inserted 2 instructions at the beginning of the unroll8 loop,
cmp and branch, are a workaround that is found heuristically.
Reviewed-by: Wilco Dijkstra <Wilco.Dijkstra@arm.com>
| sysdeps/aarch64/multiarch/memset_a64fx.S | diffblobblamehistory |