https://gcc.gnu.org/bugzilla/show_bug.cgi?id=120957
Bug ID: 120957
Summary: [16 Regression] 6-9% slowdown of 503.bwaves_r on
Zen{2,3} since r16-1647-gc06979ff957485
Product: gcc
Version: 16.0
Status: UNCONFIRMED
Keywords: missed-optimization
Severity: normal
Priority: P3
Component: target
Assignee: unassigned at gcc dot gnu.org
Reporter: pheeck at gcc dot gnu.org
CC: liuhongt at gcc dot gnu.org
Blocks: 26163
Target Milestone: ---
Host: x86_64-pc-linux-gnu
Target: x86_64-pc-linux-gnu
As seen here
https://lnt.opensuse.org/db_default/v4/SPEC/graph?plot.0=295.427.0
https://lnt.opensuse.org/db_default/v4/SPEC/graph?plot.0=471.427.0
there was a 6% exec time slowdown of the 503.bwaves SPEC 2017
benchmark when run with -Ofast -march=native on an AMD Zen2 machine, 9%
slowdown on an AMD Zen3 machine.
I bisected this to r16-1647-gc06979ff957485 (2025-06-24).
c06979ff95748559da0c2d3aa4eda9d5999eaaf6 is the first bad commit
commit c06979ff95748559da0c2d3aa4eda9d5999eaaf6
Author: hongtao.liu <hongtao....@intel.com>
Date: Wed Mar 5 12:25:32 2025 +0100
Don't duplicate setup code cost when do group-candidate cost calucalution.
- /* Uses in a group can share setup code, so only add setup cost once.
*/
- cost -= cost.scratch;
It looks like the original code took into account avoiding double
counting, but unfortunately cost is reset inside the follow loop which
invalidates the upper code, and makes same setup code cost duplicated in
each use of the group.
The patch fix the issue. It can also improve 548.exchange_r by 6% with
-march=x86-64-v3 -O2 due to better ivopt on EMR.
No big performance impact for SPEC2017 on graviton4/SPR with -mcpu=native
-Ofast -fomit-framepointer -flto=auto.
gcc/ChangeLog:
PR target/115842
* tree-ssa-loop-ivopts.cc (determine_group_iv_cost_address):
Don't recalculate inv_expr when group-candidate cost
calucalution.
gcc/tree-ssa-loop-ivopts.cc | 9 ++++++---
1 file changed, 6 insertions(+), 3 deletions(-)
bisect found first bad commit
Connection to tiber.arch.suse.cz closed.
This is a regression against GCC 15. See the comparison (Zen2)
here:
https://lnt.opensuse.org/db_default/v4/SPEC/graph?plot.0=1070.427.0&plot.1=1219.427.0&plot.2=295.427.0&;
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)
