https://gcc.gnu.org/bugzilla/show_bug.cgi?id=125858
Bug ID: 125858
Summary: [17 Regression] 19-26% slowdown of lbm_r on Zen2 with
PGO since r17-140-gf8d911e6ae3fc1
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=293.477.0
there was a 19-26% exec time slowdown of the lbm_r SPEC 2017
benchmark when compiled with -Ofast -march=native -g with PGO on a Zen2
machine. I bisected it to r17-140-gf8d911e6ae3fc1.
f8d911e6ae3fc14c33780ab950511e2b5a20484c is the first bad commit
commit f8d911e6ae3fc14c33780ab950511e2b5a20484c
Author: Richard Biener <[email protected]>
Date: Mon Mar 2 14:53:04 2026 +0100
Simplify vect_bb_slp_mark_live_stmts
This is a regression against previous GCC versions. See the comparison
here:
https://lnt.opensuse.org/db_default/v4/SPEC/graph?plot.0=333.477.0&plot.1=319.477.0&plot.2=344.477.0&plot.3=273.477.0&plot.4=396.477.0&plot.5=425.477.0&plot.6=783.477.0&plot.7=931.477.0&plot.8=1075.477.0&plot.9=1217.477.0&plot.10=1533.477.0&plot.11=293.477.0&;
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)
