https://gcc.gnu.org/bugzilla/show_bug.cgi?id=126963
Bug ID: 126963
Summary: [17 Regression] 6% slowdown of lbm on Zen5 since
r17-2817-g5e45cc08f3bbbf
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: rguenth at gcc dot gnu.org
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=1284.240.0
there was a 6% exec time slowdown of the lbm SPEC 2006
benchmark when compiled with -Ofast -march=native -g -flto=128 on a Zen 5
machine. I bisected it to r17-2817-g5e45cc08f3bbbf.
5e45cc08f3bbbfa77b6e6524127f7acbec1d2c83 is the first bad commit
commit 5e45cc08f3bbbfa77b6e6524127f7acbec1d2c83
Author: Richard Biener <[email protected]>
Date: Wed Jul 29 14:26:42 2026 +0200
Avoid SSA copies from vect_add_slp_permutation
This is a regression against GCC 16. See the comparison
here:
https://lnt.opensuse.org/db_default/v4/SPEC/graph?plot.0=1264.240.0&plot.1=1321.240.0&plot.2=1291.240.0&plot.3=1343.240.0&plot.4=1546.240.0&plot.5=1284.240.0&;
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)
