https://gcc.gnu.org/bugzilla/show_bug.cgi?id=123327
Bug ID: 123327
Summary: [16 Regression] 4% slowdown of perlbench on Zen5 since
r16-6232-gb2b5ccb0a2450a
Product: gcc
Version: 16.0
Status: UNCONFIRMED
Keywords: missed-optimization
Severity: normal
Priority: P3
Component: target
Assignee: unassigned at gcc dot gnu.org
Reporter: jmelcr at gcc dot gnu.org
CC: pinskia at gcc dot gnu.org
Blocks: 26163
Target Milestone: ---
Host: x86_64-suse-linux
Target: x86_64-suse-linux
As seen here
https://lnt.opensuse.org/db_default/v4/SPEC/graph?plot.0=1241.10.0
there was a 4% exec time slowdown of the perlbench SPEC 2006
benchmark when compiled with -Ofast -march=x86-64-v3 -g -flto=128 on a Zen5
machine. I bisected it to r16-6232-gb2b5ccb0a2450a.
b2b5ccb0a2450a812526bdde72545f9c85fce473 is the first bad commit
commit b2b5ccb0a2450a812526bdde72545f9c85fce473
Author: Andrew Pinski <[email protected]>
Date: Tue Dec 16 16:04:18 2025 -0800
predict: Fix return value in unlikely_executed_stmt_p for
__builtin_unreachable/__builtin_trap
This is a regression against GCC 15. See the comparison
here:
https://lnt.opensuse.org/db_default/v4/SPEC/graph?plot.0=1260.10.0&plot.1=1322.10.0&plot.2=1290.10.0&plot.3=1350.10.0&plot.4=1241.10.0&;
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)
