add patch with workaround for Arm SVE512 test failure to GROMACS 2024+2025 easyconfigs by bedroge · Pull Request #24548 · easybuilders/easybuild-easyconfigs

bedroge · 2025-11-13T12:55:38Z

See https://gitlab.com/gromacs/gromacs/-/issues/5440 and https://gitlab.com/gromacs/gromacs/-/commit/343a96210009ae420f633c2cb2daf663f1147dee.

bedroge · 2025-11-13T14:33:42Z

@boegelbot please test @ jsc-zen3
CORE_CNT=16
EB_ARGS="--installpath /tmp/pr24548"

boegelbot · 2025-11-13T14:40:09Z

@bedroge: Request for testing this PR well received on jsczen3l1.int.jsc-zen3.fz-juelich.de

PR test command 'if [[ develop != 'develop' ]]; then EB_BRANCH=develop ./easybuild_develop.sh 2> /dev/null 1>&2; EB_PREFIX=/home/boegelbot/easybuild/develop source init_env_easybuild_develop.sh; fi; EB_PR=24548 EB_ARGS="--installpath /tmp/pr24548" EB_CONTAINER= EB_REPO=easybuild-easyconfigs EB_BRANCH=develop /opt/software/slurm/bin/sbatch --job-name test_PR_24548 --ntasks="16" ~/boegelbot/eb_from_pr_upload_jsc-zen3.sh' executed!

exit code: 0
output:

Submitted batch job 8758

Test results coming soon (I hope)...

Details

- notification for comment with ID 3528108653 processed

Message to humans: this is just bookkeeping information for me,
it is of no use to you (unless you think I have a bug, which I don't).

boegelbot · 2025-11-13T15:46:59Z

Test report by @boegelbot
FAILED
Build succeeded for 0 out of 9 (total: 1 hour 6 mins 15 secs) (9 easyconfigs in total)
jsczen3c1.int.jsc-zen3.fz-juelich.de - Linux Rocky Linux 9.6, x86_64, AMD EPYC-Milan Processor (zen3), Python 3.9.21
See https://gist.github.com/boegelbot/f037412397f8b1e6dddd1b50ef79d628 for a full test report.

boegel · 2025-11-13T16:00:19Z

@bedroge That didn't exactly go well...

/tmp/boegelbot/GROMACS/2024.1/foss-2023b/gromacs-2024.1/src/gromacs/trajectoryanalysis/tests/surfacearea.cpp:61:42: error: operator '&&' has no right operand
   61 |         && GMX_SIMD_ARM_SVE_LENGTH_VALUE == 512
      |                                          ^~

ocaisa · 2025-11-14T07:23:06Z

ChatGPT says it should be

#if defined(__GNUC__) && !defined(__clang__) && __GNUC__ >= 13 && \
    defined(GMX_SIMD_ARM_SVE) && defined(GMX_SIMD_ARM_SVE_LENGTH_VALUE) && \
    GMX_SIMD_ARM_SVE && (GMX_SIMD_ARM_SVE_LENGTH_VALUE == 512)
#define WORKAROUND_GCC13_ARM_SVE512 1
#endif

as GMX_SIMD_ARM_SVE and GMX_SIMD_ARM_SVE_LENGTH_VALUE are not defined at all on x86

…rge_requests/5593

bedroge · 2025-11-14T08:00:21Z

ChatGPT says it should be

#if defined(__GNUC__) && !defined(__clang__) && __GNUC__ >= 13 && \
    defined(GMX_SIMD_ARM_SVE) && defined(GMX_SIMD_ARM_SVE_LENGTH_VALUE) && \
    GMX_SIMD_ARM_SVE && (GMX_SIMD_ARM_SVE_LENGTH_VALUE == 512)
#define WORKAROUND_GCC13_ARM_SVE512 1
#endif

as GMX_SIMD_ARM_SVE and GMX_SIMD_ARM_SVE_LENGTH_VALUE are not defined at all on x86

I took the latest version from https://gitlab.com/gromacs/gromacs/-/merge_requests/5593, which should solve the issue (by having a nested if statement). GMX_SIMD_ARM_SVE was actually defined (set to 0), it just didn't seem to do lazy evaluation, so I think your AI friend may be hallucinating 😄

bedroge · 2025-11-14T08:00:36Z

@boegelbot please test @ jsc-zen3
CORE_CNT=16
EB_ARGS="--installpath /tmp/pr24548"

boegelbot · 2025-11-14T08:10:11Z

@bedroge: Request for testing this PR well received on jsczen3l1.int.jsc-zen3.fz-juelich.de

PR test command 'if [[ develop != 'develop' ]]; then EB_BRANCH=develop ./easybuild_develop.sh 2> /dev/null 1>&2; EB_PREFIX=/home/boegelbot/easybuild/develop source init_env_easybuild_develop.sh; fi; EB_PR=24548 EB_ARGS="--installpath /tmp/pr24548" EB_CONTAINER= EB_REPO=easybuild-easyconfigs EB_BRANCH=develop /opt/software/slurm/bin/sbatch --job-name test_PR_24548 --ntasks="16" ~/boegelbot/eb_from_pr_upload_jsc-zen3.sh' executed!

exit code: 0
output:

Submitted batch job 8763

Test results coming soon (I hope)...

Details

- notification for comment with ID 3531440932 processed

Message to humans: this is just bookkeeping information for me,
it is of no use to you (unless you think I have a bug, which I don't).

boegelbot · 2025-11-14T13:05:15Z

Test report by @boegelbot
FAILED
Build succeeded for 8 out of 9 (total: 4 hours 54 mins 43 secs) (9 easyconfigs in total)
jsczen3c1.int.jsc-zen3.fz-juelich.de - Linux Rocky Linux 9.6, x86_64, AMD EPYC-Milan Processor (zen3), Python 3.9.21
See https://gist.github.com/boegelbot/cd7303892b9433f9bed9c0adb3004219 for a full test report.

bedroge · 2025-11-14T15:05:52Z

@boegelbot please test @ jsc-zen3
CORE_CNT=16
EB_ARGS="--installpath /tmp/pr24548 GROMACS-2024.3-foss-2023b-PLUMED-2.9.2.eb"

boegelbot · 2025-11-14T15:10:12Z

@bedroge: Request for testing this PR well received on jsczen3l1.int.jsc-zen3.fz-juelich.de

PR test command 'if [[ develop != 'develop' ]]; then EB_BRANCH=develop ./easybuild_develop.sh 2> /dev/null 1>&2; EB_PREFIX=/home/boegelbot/easybuild/develop source init_env_easybuild_develop.sh; fi; EB_PR=24548 EB_ARGS="--installpath /tmp/pr24548 GROMACS-2024.3-foss-2023b-PLUMED-2.9.2.eb" EB_CONTAINER= EB_REPO=easybuild-easyconfigs EB_BRANCH=develop /opt/software/slurm/bin/sbatch --job-name test_PR_24548 --ntasks="16" ~/boegelbot/eb_from_pr_upload_jsc-zen3.sh' executed!

exit code: 0
output:

Submitted batch job 8773

Test results coming soon (I hope)...

Details

- notification for comment with ID 3533197102 processed

Message to humans: this is just bookkeeping information for me,
it is of no use to you (unless you think I have a bug, which I don't).

boegelbot · 2025-11-14T15:48:01Z

Test report by @boegelbot
SUCCESS
Build succeeded for 1 out of 1 (total: 37 mins 33 secs) (1 easyconfigs in total)
jsczen3c2.int.jsc-zen3.fz-juelich.de - Linux Rocky Linux 9.6, x86_64, AMD EPYC-Milan Processor (zen3), Python 3.9.21
See https://gist.github.com/boegelbot/b6e4a414e9170873623d91bb7bf68bb1 for a full test report.

boegel

lgtm

boegel · 2025-11-16T09:01:03Z

Going in, thanks @bedroge!

add patch with workaround for Arm SVE512 issue

864017f

bedroge added bug fix EESSI Related to EESSI project aarch64 Related to Arm 64-bit (aarch64) labels Nov 13, 2025

github-actions Bot added 2023b 2025a issues & PRs related to 2025a common toolchains 2025b issues & PRs related to 2025b common toolchains change labels Nov 13, 2025

bedroge changed the title ~~add patch with workaround for Arm SVE512 issue to GROMACS 2024+2025 easyconfigs~~ add patch with workaround for Arm SVE512 test failure to GROMACS 2024+2025 easyconfigs Nov 13, 2025

boegel mentioned this pull request Nov 13, 2025

{2023.06}[2023b, a64fx] GROMACS 2024.3 + 2024.4 EESSI/software-layer#1302

Merged

bedroge added 2 commits November 14, 2025 08:56

use latest patch version from https://gitlab.com/gromacs/gromacs/-/me…

753c6c7

…rge_requests/5593

update checksum of patch

2a45a80

boegel added this to the next release (5.2.0?) milestone Nov 16, 2025

boegel approved these changes Nov 16, 2025

View reviewed changes

boegel merged commit 091da70 into easybuilders:develop Nov 16, 2025
8 checks passed

bedroge deleted the gromacs_a64fx_fix branch November 16, 2025 09:16

boegel removed the change label Nov 19, 2025

Conversation

bedroge commented Nov 13, 2025

Uh oh!

bedroge commented Nov 13, 2025

Uh oh!

boegelbot commented Nov 13, 2025

Uh oh!

boegelbot commented Nov 13, 2025

Uh oh!

boegel commented Nov 13, 2025

Uh oh!

ocaisa commented Nov 14, 2025

Uh oh!

bedroge commented Nov 14, 2025

Uh oh!

bedroge commented Nov 14, 2025

Uh oh!

boegelbot commented Nov 14, 2025

Uh oh!

boegelbot commented Nov 14, 2025

Uh oh!

bedroge commented Nov 14, 2025

Uh oh!

boegelbot commented Nov 14, 2025

Uh oh!

boegelbot commented Nov 14, 2025

Uh oh!

boegel left a comment

Choose a reason for hiding this comment

Uh oh!

boegel commented Nov 16, 2025

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants