https://gcc.gnu.org/bugzilla/show_bug.cgi?id=127522
Bug ID: 127522
Summary: [17 Regression] 4% slowdown of sphinx3 on znver4 since
r17-3887-ga937d8f7b557d4
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
I am seeing a 4% exec time slowdown of SPEC 2006 sphinx3 when compiled with -O2
-march=x86-64-v3 -g -flto=128 PGO on znver4:
https://lnt.opensuse.org/db_default/v4/SPEC/graph?plot.0=1109.280.0
The slowdown bisects to r17-3887-ga937d8f7b557d4:
a937d8f7b557d4d6231e71c885cfbcdd50c5de27 is the first bad commit
commit a937d8f7b557d4d6231e71c885cfbcdd50c5de27
Author: Richard Biener <[email protected]>
Date: Tue Sep 1 10:23:24 2026 +0200
Use more precise costing location for BB vectorization
This is a regression against GCC 16 and 15:
https://lnt.opensuse.org/db_default/v4/SPEC/graph?plot.0=1136.280.0&plot.1=1147.280.0&plot.2=1571.280.0&plot.3=1109.280.0&;
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)
