public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug ipa/90151] 554.roms_r regression on x86_64 at -O2 and generic march/mtune
[not found] <bug-90151-4@http.gcc.gnu.org/bugzilla/>
@ 2020-03-30 20:31 ` jamborm at gcc dot gnu.org
2022-01-20 16:50 ` jamborm at gcc dot gnu.org
2023-01-31 0:02 ` jamborm at gcc dot gnu.org
2 siblings, 0 replies; 3+ messages in thread
From: jamborm at gcc dot gnu.org @ 2020-03-30 20:31 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=90151
--- Comment #1 from Martin Jambor <jamborm at gcc dot gnu.org> ---
This year's numbers:
- on AMD Zen1, we are still 7.2% worse than GCC 7
- on AMD Zen2, the reegression is 4.6%
- in Intel Cascade Lake server CPU, it is 5.4%
This is all -O2, so perhaps not that important for a Fortran
benchmark.
^ permalink raw reply [flat|nested] 3+ messages in thread
* [Bug ipa/90151] 554.roms_r regression on x86_64 at -O2 and generic march/mtune
[not found] <bug-90151-4@http.gcc.gnu.org/bugzilla/>
2020-03-30 20:31 ` [Bug ipa/90151] 554.roms_r regression on x86_64 at -O2 and generic march/mtune jamborm at gcc dot gnu.org
@ 2022-01-20 16:50 ` jamborm at gcc dot gnu.org
2023-01-31 0:02 ` jamborm at gcc dot gnu.org
2 siblings, 0 replies; 3+ messages in thread
From: jamborm at gcc dot gnu.org @ 2022-01-20 16:50 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=90151
--- Comment #2 from Martin Jambor <jamborm at gcc dot gnu.org> ---
Reconfirmed in 2021 too, also on LNT. The best way to see current
status is probably to go to
https://lnt.opensuse.org/db_default/v4/SPEC/spec_report/branch?sorting=gcc-7%2Cgcc-8%2Cgcc-9%2Cgcc-10%2Cgcc-11%2Cgcc-trunk
and then enter "554.roms" into the "test name filter" edit field.
On Zen3 we now regress by 5.63% compared to GCC 7 at -O2.
On Zen2 we now regress by 8.35% compared to GCC 7 at -O2.
On Zen1 we now regress by 11.15% compared to GCC 7 at -O2.
On s Kabylake "client" CPU we now regress by 9.06% compared to GCC 7 at -O2.
On a Cascadelake "server" CPU we now regresss by 7.84% compared to GCC 7 at
-O2.
^ permalink raw reply [flat|nested] 3+ messages in thread
* [Bug ipa/90151] 554.roms_r regression on x86_64 at -O2 and generic march/mtune
[not found] <bug-90151-4@http.gcc.gnu.org/bugzilla/>
2020-03-30 20:31 ` [Bug ipa/90151] 554.roms_r regression on x86_64 at -O2 and generic march/mtune jamborm at gcc dot gnu.org
2022-01-20 16:50 ` jamborm at gcc dot gnu.org
@ 2023-01-31 0:02 ` jamborm at gcc dot gnu.org
2 siblings, 0 replies; 3+ messages in thread
From: jamborm at gcc dot gnu.org @ 2023-01-31 0:02 UTC (permalink / raw)
To: gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=90151
--- Comment #3 from Martin Jambor <jamborm at gcc dot gnu.org> ---
In January 2022 I see 9% regression on zen2, 8% regression on zen3 and
CascadeLake and 7% on zen4 (compared to gcc 7). I no longer track
zen1. LNT largely confirms these observations although it tracks -O2
mostly only with LTO.
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2023-01-31 0:02 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
[not found] <bug-90151-4@http.gcc.gnu.org/bugzilla/>
2020-03-30 20:31 ` [Bug ipa/90151] 554.roms_r regression on x86_64 at -O2 and generic march/mtune jamborm at gcc dot gnu.org
2022-01-20 16:50 ` jamborm at gcc dot gnu.org
2023-01-31 0:02 ` jamborm at gcc dot gnu.org
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).