https://gcc.gnu.org/bugzilla/show_bug.cgi?id=126433
Bug ID: 126433
Summary: [17 Regression] 4% slowdown of imagick_r on Zen5 since
r17-2486-g60b27d915e1474
Product: gcc
Version: 17.0
Status: UNCONFIRMED
Keywords: missed-optimization
Severity: normal
Priority: P3
Component: tree-optimization
Assignee: unassigned at gcc dot gnu.org
Reporter: jmelcr at gcc dot gnu.org
CC: ubizjak at gmail dot com
Blocks: 26163
Target Milestone: ---
Host: x86_64-suse-linux
Target: x86_64-suse-linux
Build: x86_64-suse-linux
As seen here
https://lnt.opensuse.org/db_default/v4/SPEC/graph?plot.0=1252.507.0
there was a 4% exec time slowdown of the imagick_r SPEC 2017
benchmark when compiled with -Ofast -march=native -g PGO on an Zen5
machine. I bisected it to r17-2486-g60b27d915e1474.
60b27d915e14748a627deb1a4616da419b1e24b2 is the first bad commit
commit 60b27d915e14748a627deb1a4616da419b1e24b2
Author: Uros Bizjak <[email protected]>
Date: Fri Jul 17 13:28:16 2026 +0200
i386: Add ROLW %r,8 alternative to bswaphi2 patterns [PR126283]
This is a regression against GCC 16. See the comparison
here:
https://lnt.opensuse.org/db_default/v4/SPEC/graph?plot.0=1281.507.0&plot.1=1339.507.0&plot.2=1305.507.0&plot.3=1367.507.0&plot.4=1565.507.0&plot.5=1252.507.0&;
Referenced Bugs:
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=26163
[Bug 26163] [meta-bug] missed optimization in SPEC (2026, 2k17, 2k and 2k6 and
95)
