public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
From: "jamborm at gcc dot gnu.org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug target/113295] New: SPEC 2006 416.gamess miscompares on Aarch64 when built with -Ofast -march=native -flto
Date: Tue, 09 Jan 2024 17:42:28 +0000 [thread overview]
Message-ID: <bug-113295-4@http.gcc.gnu.org/bugzilla/> (raw)
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=113295
Bug ID: 113295
Summary: SPEC 2006 416.gamess miscompares on Aarch64 when built
with -Ofast -march=native -flto
Product: gcc
Version: 14.0
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: target
Assignee: unassigned at gcc dot gnu.org
Reporter: jamborm at gcc dot gnu.org
Blocks: 26163
Target Milestone: ---
Host: aarch64-linux
Target: aarch64-linux
Our Aarch64 benchmarker (armv8.2-a+crypto+fp16+rcpc+dotprod+ssbs)
signals that SPEC 2006 416.gamess miscompares when built with -Ofast
-march=native -flto and master revision r14-7022-g34d339bbd0c1f5.
*** Miscompare of cytosine.2.out;
1120: 1 C 0.027630132 0.018067739 0.002234116
1 C -223.432234062 7.107716215 -9.326017293
^
1121: 2 C 0.012259576 -0.006051645 -0.000067202
2 C -205.307990130 -173.019401916 -6.442472179
^
1122: 3 C -0.012829758 0.003221329 -0.000743429
3 C -0.900001858 -263.923127366 3.131404191
^
1123: 4 N -0.041204707 0.020932737 -0.000372560
4 N 291.766837166 -257.876625173 10.788390925
^
1124: 5 C 0.057007688 0.032540385 -0.000909621
5 C -204.215830139 57.403322317 -0.929403441
^
1125: 6 N -0.015041867 -0.049945043 0.002129121
6 N 117.540483305 300.802327718 -10.562573219
^
1126: 7 O -0.076442899 -0.041673056 -0.000117411
7 O 481.672983389 238.169793443 3.121961894
^
1127: 8 N 0.034391335 -0.016048119 -0.001905357
8 N -247.780884876 91.547672097 -1.133077767
^
1128: 9 H 0.014938973 0.008953835 0.000373759
9 H -3.948185240 0.659195159 -0.007774350
^
1129: 10 H -0.002268325 0.023480419 0.000010207
10 H -0.160267091 -4.621874235 -0.178161165
Unfortunately at the moment don't have another access to another
adequate Aarch64 machine to debug further and so at this time I cannot
provide more information (and so the component "target" is likely
bogus, sorry).
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)
next reply other threads:[~2024-01-09 17:42 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-01-09 17:42 jamborm at gcc dot gnu.org [this message]
2024-01-09 17:43 ` [Bug target/113295] [14 Regression] " pinskia at gcc dot gnu.org
2024-02-19 13:13 ` tnfchris at gcc dot gnu.org
2024-02-21 10:25 ` tnfchris at gcc dot gnu.org
2024-02-21 10:32 ` tnfchris at gcc dot gnu.org
2024-02-21 11:07 ` [Bug target/113295] [14 Regression] SPEC 2006 416.gamess miscompares on Aarch64 when built with -Ofast -mcpu=native since g:2f46e3578d45ff060a0a329cb39d4f52878f9d5a tnfchris at gcc dot gnu.org
2024-02-21 11:09 ` rsandifo at gcc dot gnu.org
2024-02-21 18:37 ` rsandifo at gcc dot gnu.org
2024-02-23 14:13 ` cvs-commit at gcc dot gnu.org
2024-02-23 14:13 ` cvs-commit at gcc dot gnu.org
2024-02-23 14:17 ` rsandifo at gcc dot gnu.org
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=bug-113295-4@http.gcc.gnu.org/bugzilla/ \
--to=gcc-bugzilla@gcc.gnu.org \
--cc=gcc-bugs@gcc.gnu.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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).