https://gcc.gnu.org/bugzilla/show_bug.cgi?id=120943
Bug ID: 120943
Summary: [16 Regression] 5% slowdown of 527.cam4_r on Zen{4,5}
since r16-1643-gd073bb6cfc219d
Product: gcc
Version: 16.0
Status: UNCONFIRMED
Keywords: missed-optimization
Severity: normal
Priority: P3
Component: target
Assignee: unassigned at gcc dot gnu.org
Reporter: pheeck at gcc dot gnu.org
CC: hjl at gcc dot gnu.org
Blocks: 26163
Target Milestone: ---
Host: x86_64-linux
Target: x86_64-linux
As seen here
https://lnt.opensuse.org/db_default/v4/SPEC/graph?plot.0=1108.497.0
https://lnt.opensuse.org/db_default/v4/SPEC/graph?plot.0=1286.497.0
there was a 5% exec time slowdown of 527.cam4_r SPEC 2017
benchmark when run with -O2 -march=x86-64-v3 -flto -fprofile-use on an AMD Zen4
or Zen5 machine.
I bisected it to r16-1643-gd073bb6cfc219d.
d073bb6cfc219d4b6c283a0b527ee88b42e640e0 is the first bad commit
commit d073bb6cfc219d4b6c283a0b527ee88b42e640e0
Author: H.J. Lu <hjl.to...@gmail.com>
Date: Thu Mar 18 18:43:10 2021 -0700
x86: Update memcpy/memset inline strategies for -mtune=generic
Update memcpy and memset inline strategies for -mtune=generic:
1. Don't align memory.
2. For known sizes, prefer vector loop, unroll loop with 4 moves or
stores per iteration without aligning the loop, up to 256 bytes.
3. For unknown sizes, use memcpy/memset.
4. Since each loop iteration has 4 stores and 8 stores for zeroing with
unroll loop may be needed, change CLEAR_RATIO to 10 so that zeroing
up to 72 bytes are fully unrolled with 9 stores without SSE.
Referenced Bugs:
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=26163
[Bug 26163] [meta-bug] missed optimization in SPEC (2k17, 2k and 2k6 and 95)
