[Bug tree-optimization/119965] New: [16 Regression] 531.deepsjeng_r binary is 50% bigger since r16-116-gcfb04e0de6aa43

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

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=119965

            Bug ID: 119965
           Summary: [16 Regression] 531.deepsjeng_r binary is 50% bigger
                    since r16-116-gcfb04e0de6aa43
           Product: gcc
           Version: 16.0
            Status: UNCONFIRMED
          Keywords: missed-optimization
          Severity: normal
          Priority: P3
         Component: tree-optimization
          Assignee: unassigned at gcc dot gnu.org
          Reporter: pheeck at gcc dot gnu.org
                CC: hubicka at gcc dot gnu.org
            Blocks: 26163
  Target Milestone: ---
              Host: x86_64-pc-linux-gnu
            Target: x86_64-pc-linux-gnu

If you compile 531.deepsjeng_r from SPEC CPU 2017 benchmark suite with -Ofast
(or -O2) -march=native PGO, you get a 50% bigger binary (70% bigger .text) than
before. Also happens with -Ofast -march=native -flto PGO but to a lesser extent
(25% bigger .text).

Happens on zen2, zen3, zen4 and Intel Ice Lake (3rd generation Xeon).

This got introduced by r16-101-g132d01d96ea9d6, but 531.deepsjeng_r couldn't be
built until r16-116-gcfb04e0de6aa43.


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)