{2026.06}[2026.1] GROMACS 2026.3 with CUDA 13.3.0 by bedroge · Pull Request #1658 · EESSI/software-layer

bedroge · 2026-09-17T18:44:07Z

No description provided.

bedroge · 2026-09-17T18:44:34Z

bot: build repo:eessi.io-2026.06-software instance:eessi-bot-rug for:arch=x86_64/amd/zen5,accel=nvidia/cc120

eessi-bot-rug · 2026-09-17T18:44:44Z

New job on instance eessi-bot-rug for repository eessi.io-2026.06-software
Building on: amd-zen5 and accelerator nvidia/cc120
Building for: x86_64/amd/zen5 and accelerator nvidia/cc120
Job dir: /scratch/hb-eessibot/SHARED/jobs/2026.09/pr_1658/31702815

date	job status	comment
Sep 17 18:44:42 UTC 2026	submitted	job id `31702815` awaits release by job manager
Sep 17 18:44:59 UTC 2026	released	job awaits launch by Slurm scheduler
Sep 17 18:53:04 UTC 2026	running	job `31702815` is running
Sep 17 19:33:47 UTC 2026	finished	😢 FAILURE (click triangle for details) Details ✅ job output file `slurm-31702815.out` ✅ no message matching `FATAL:` ❌ found message matching `ERROR:` ❌ found message matching `FAILED:` ❌ found message matching `required modules missing:` ❌ no message matching `No missing installations` ✅ found message matching `.tar.* created!` Artefacts `eessi-2026.06-software-linux-x86_64-amd-zen5-accel-nvidia-cc120-17896735150.tar.zst` size: 0 MiB (1009435 bytes) entries: 87 modules under 2026.06/software/linux/x86_64/amd/zen5/accel/nvidia/cc120/modules/all `hwloc-CUDA/2.13.0-GCCcore-15.2.0-CUDA-13.3.0.lua` `UCX-CUDA/1.20.0-GCCcore-15.2.0-CUDA-13.3.0.lua` software under 2026.06/software/linux/x86_64/amd/zen5/accel/nvidia/cc120/software `hwloc-CUDA/2.13.0-GCCcore-15.2.0-CUDA-13.3.0` `UCX-CUDA/1.20.0-GCCcore-15.2.0-CUDA-13.3.0` reprod directories under 2026.06/software/linux/x86_64/amd/zen5/accel/nvidia/cc120/reprod `hwloc-CUDA/2.13.0-GCCcore-15.2.0-CUDA-13.3.0/20260917_185740UTC` `UCX-CUDA/1.20.0-GCCcore-15.2.0-CUDA-13.3.0/20260917_185659UTC` other under 2026.06/software/linux/x86_64/amd/zen5/accel/nvidia/cc120 no other files in tarball
Sep 17 19:33:47 UTC 2026	test result	😢 FAILURE (click triangle for details) Reason EESSI test suite was not run, test step itself failed to execute. Details ✅ job output file `slurm-31702815.out` ❌ found message matching `ERROR:` ✅ no message matching `[\sFAILED\s].Ran . test case`

bedroge · 2026-09-18T07:15:29Z

Hmm, the sanity check is failing:

== FAILED: Installation ended unsuccessfully: Sanity check failed: no file found at 'bin/gmx_d' in

That's probably due to https://github.com/easybuilders/easybuild-easyblocks/blob/develop/easybuild/easyblocks/g/gromacs.py#L751 and due to demoting CUDA to a build dependency, which is not loaded during the sanity check anymore in recent EB versions? How do we fix this? @ocaisa @boegel

boegel · 2026-09-20T20:30:27Z

Hmm, the sanity check is failing:
== FAILED: Installation ended unsuccessfully: Sanity check failed: no file found at 'bin/gmx_d' in 
That's probably due to https://github.com/easybuilders/easybuild-easyblocks/blob/develop/easybuild/easyblocks/g/gromacs.py#L751 and due to demoting CUDA to a build dependency, which is not loaded during the sanity check anymore in recent EB versions? How do we fix this? @ocaisa @boegel

Rather than checking that CUDA is loaded during the sanity check, we can check whether it's listed as a (build or runtime) dependency

bedroge · 2026-09-21T13:00:19Z

Rather than checking that CUDA is loaded during the sanity check, we can check whether it's listed as a (build or runtime) dependency

Done in easybuilders/easybuild-easyblocks#4234. Let's give it a try.

…026.3_cuda_13.3.0

bedroge · 2026-09-21T13:01:43Z

bot: build repo:eessi.io-2026.06-software instance:eessi-bot-rug for:arch=x86_64/amd/zen5,accel=nvidia/cc120

eessi-bot-rug · 2026-09-21T13:01:56Z

New job on instance eessi-bot-rug for repository eessi.io-2026.06-software
Building on: amd-zen5 and accelerator nvidia/cc120
Building for: x86_64/amd/zen5 and accelerator nvidia/cc120
Job dir: /scratch/hb-eessibot/SHARED/jobs/2026.09/pr_1658/31818232

date	job status	comment
Sep 21 13:01:54 UTC 2026	submitted	job id `31818232` awaits release by job manager
Sep 21 13:02:33 UTC 2026	released	job awaits launch by Slurm scheduler
Sep 21 13:04:38 UTC 2026	running	job `31818232` is running
Sep 21 13:41:17 UTC 2026	finished	😁 SUCCESS (click triangle for details) Details ✅ job output file `slurm-31818232.out` ✅ no message matching `FATAL:` ✅ no message matching `ERROR:` ✅ no message matching `FAILED:` ✅ no message matching `required modules missing:` ✅ found message(s) matching `No missing installations` ✅ found message matching `.tar.* created!` Artefacts `eessi-2026.06-software-linux-x86_64-amd-zen5-accel-nvidia-cc120-17899980050.tar.zst` size: 33 MiB (35160463 bytes) entries: 889 modules under 2026.06/software/linux/x86_64/amd/zen5/accel/nvidia/cc120/modules/all `GROMACS/2026.3-foss-2026.1-CUDA-13.3.0.lua` `hwloc-CUDA/2.13.0-GCCcore-15.2.0-CUDA-13.3.0.lua` `UCX-CUDA/1.20.0-GCCcore-15.2.0-CUDA-13.3.0.lua` software under 2026.06/software/linux/x86_64/amd/zen5/accel/nvidia/cc120/software `GROMACS/2026.3-foss-2026.1-CUDA-13.3.0` `hwloc-CUDA/2.13.0-GCCcore-15.2.0-CUDA-13.3.0` `UCX-CUDA/1.20.0-GCCcore-15.2.0-CUDA-13.3.0` reprod directories under 2026.06/software/linux/x86_64/amd/zen5/accel/nvidia/cc120/reprod `GROMACS/2026.3-foss-2026.1-CUDA-13.3.0/20260921_133949UTC` `hwloc-CUDA/2.13.0-GCCcore-15.2.0-CUDA-13.3.0/20260921_130840UTC` `UCX-CUDA/1.20.0-GCCcore-15.2.0-CUDA-13.3.0/20260921_130806UTC` other under 2026.06/software/linux/x86_64/amd/zen5/accel/nvidia/cc120 no other files in tarball
Sep 21 13:41:17 UTC 2026	test result	😢 FAILURE (click triangle for details) Reason EESSI test suite was not run, test step itself failed to execute. Details ✅ job output file `slurm-31818232.out` ❌ found message matching `ERROR:` ✅ no message matching `[\sFAILED\s].Ran . test case`

casparvl · 2026-09-30T12:23:21Z

Just want to check the formatting of this table for GPU builds... abusing this PR for that

bot:status last_build

eessi-bot-surf · 2026-09-30T12:23:26Z