[all-commits] [llvm/llvm-project] abb825: tsan: optimize __tsan_read/write16
Dmitry Vyukov via All-commits
all-commits at lists.llvm.org
Tue Dec 21 02:33:40 PST 2021
Author: Dmitry Vyukov <dvyukov at google.com>
Date: 2021-12-21 (Tue, 21 Dec 2021)
tsan: optimize __tsan_read/write16
These callbacks are used for SSE vector accesses.
In some computational programs these accesses dominate.
Currently we do 2 uninlined 8-byte accesses to handle them.
Inline and optimize them similarly to unaligned accesses.
This reduces the vector access benchmark time from 8 to 3 seconds.
Depends on D112603.
Reviewed By: melver
Differential Revision: https://reviews.llvm.org/D114594
More information about the All-commits