[Bug tree-optimization/127775] New: [17 Regression] 7% slowdown of SPEC 2017 x264_r on Zen3 since r17-4901-gf9b60a8bd7569f

[jmelcr at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=127775

            Bug ID: 127775
           Summary: [17 Regression] 7% slowdown of SPEC 2017 x264_r on
                    Zen3 since r17-4901-gf9b60a8bd7569f
           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: mjambor at suse dot cz
            Blocks: 26163
  Target Milestone: ---
              Host: x86_64-suse-linux
            Target: x86_64-suse-linux
             Build: x86_64-suse-linux

I am seeing a 7% of SPEC 2017 x264_r on Zen3 when compiling with  -Ofast
-march=native /* =znver3 */ -g -flto=128 PGO:

https://lnt.opensuse.org/db_default/v4/SPEC/graph?plot.0=477.377.0

The slowdown bisects to commit r17-4901-gf9b60a8bd7569f:
commit f9b60a8bd7569f59534a42d1383c1f66a37542cf
Author: Martin Jambor <mjam...@suse.cz>
Date:   Mon Oct 5 16:59:12 2026 +0200

    ipa-cp: Adjust PGO IPA-CP cloning heuristics

This slowdown is a regression against GCC 16:

https://lnt.opensuse.org/db_default/v4/SPEC/graph?plot.0=1053.377.0&plot.1=1165.377.0&plot.2=1504.377.0&plot.3=477.377.0&;


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)