Update jaxlib easyblock for jax/jaxlib v0.6.2 and v0.7.0 #3852

pavelToman · 2025-07-22T15:14:17Z

resolves vscentrum/vsc-software-stack#578

pavelToman · 2025-07-22T15:23:52Z

@boegelbot please test @ jsc-zen3 EB_ARGS="jax-0.4.25-gfbf-2023a.eb"

boegelbot · 2025-07-22T15:33:09Z

@pavelToman: 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=3852 EB_ARGS="jax-0.4.25-gfbf-2023a.eb" EB_REPO=easybuild-easyblocks EB_BRANCH=develop /opt/software/slurm/bin/sbatch --job-name test_PR_3852 --ntasks=8 ~/boegelbot/eb_from_pr_upload_jsc-zen3.sh' executed!

exit code: 0
output:

Submitted batch job 7310

Test results coming soon (I hope)...

Details

- notification for comment with ID 3103357751 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 · 2025-07-22T20:32:00Z

Test report by @boegelbot

Overview of tested easyconfigs (in order)

FAIL (build issue) jax-0.4.25-gfbf-2023a.eb (partial log available at https://gist.github.com/boegelbot/237037e36cbcf9097de34b4ba8cc5b93)

Build succeeded for 0 out of 1 (1 easyconfigs in total)
jsczen3c2.int.jsc-zen3.fz-juelich.de - Linux Rocky Linux 9.5, x86_64, AMD EPYC-Milan Processor (zen3), Python 3.9.21
See https://gist.github.com/boegelbot/1e2aa1906c27c4df78827318b35ae3ff for a full test report.

pavelToman · 2025-07-23T09:51:48Z

Test report by @boegelbot

Overview of tested easyconfigs (in order)
* **FAIL (build issue)** _jax-0.4.25-gfbf-2023a.eb_ (partial log available at https://gist.github.com/boegelbot/237037e36cbcf9097de34b4ba8cc5b93)
Build succeeded for 0 out of 1 (1 easyconfigs in total) jsczen3c2.int.jsc-zen3.fz-juelich.de - Linux Rocky Linux 9.5, x86_64, AMD EPYC-Milan Processor (zen3), Python 3.9.21 See https://gist.github.com/boegelbot/1e2aa1906c27c4df78827318b35ae3ff for a full test report.

Only one test fail - worker crashed

boegelbot · 2025-07-23T09:53:09Z

@pavelToman: 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=3852 EB_ARGS="jax-0.4.25-gfbf-2023a.eb" EB_REPO=easybuild-easyblocks EB_BRANCH=develop /opt/software/slurm/bin/sbatch --job-name test_PR_3852 --ntasks=8 ~/boegelbot/eb_from_pr_upload_jsc-zen3.sh' executed!

exit code: 0
output:

Submitted batch job 7317

Test results coming soon (I hope)...

Details

- notification for comment with ID 3106769335 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 · 2025-07-23T10:13:08Z

@pavelToman: 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=3852 EB_ARGS="jax-0.4.25-gfbf-2023a.eb --parallel 4" EB_REPO=easybuild-easyblocks EB_BRANCH=develop /opt/software/slurm/bin/sbatch --job-name test_PR_3852 --ntasks=8 ~/boegelbot/eb_from_pr_upload_jsc-zen3.sh' executed!

exit code: 0
output:

Submitted batch job 7318

Test results coming soon (I hope)...

Details

- notification for comment with ID 3106774717 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 · 2025-07-23T10:13:23Z

Test report by @boegelbot

Overview of tested easyconfigs (in order)

FAIL (build issue) jax-0.4.25-gfbf-2023a.eb (partial log available at https://gist.github.com/boegelbot/c0d7d3d9df39b98a14cea800f42c3033)

Build succeeded for 0 out of 1 (1 easyconfigs in total)
jsczen3c1.int.jsc-zen3.fz-juelich.de - Linux Rocky Linux 9.5, x86_64, AMD EPYC-Milan Processor (zen3), Python 3.9.21
See https://gist.github.com/boegelbot/23e55038a63ca5f5c5bfadf0a7a4c355 for a full test report.

boegelbot · 2025-07-23T14:03:08Z

Test report by @boegelbot

Overview of tested easyconfigs (in order)

FAIL (build issue) jax-0.4.25-gfbf-2023a.eb (partial log available at https://gist.github.com/boegelbot/c921b373bf41f87df0a3e3051732d245)

Build succeeded for 0 out of 1 (1 easyconfigs in total)
jsczen3c1.int.jsc-zen3.fz-juelich.de - Linux Rocky Linux 9.5, x86_64, AMD EPYC-Milan Processor (zen3), Python 3.9.21
See https://gist.github.com/boegelbot/fab82d13673222ded33c3320398e792a for a full test report.

pavelToman · 2025-07-23T14:04:32Z

@boegelbot please test @ jsc-zen3 EB_ARGS="jax-0.4.25-gfbf-2023a.eb --parallel 4"

boegelbot · 2025-07-23T14:13:08Z

@pavelToman: 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=3852 EB_ARGS="jax-0.4.25-gfbf-2023a.eb --parallel 4" EB_REPO=easybuild-easyblocks EB_BRANCH=develop /opt/software/slurm/bin/sbatch --job-name test_PR_3852 --ntasks=8 ~/boegelbot/eb_from_pr_upload_jsc-zen3.sh' executed!

exit code: 0
output:

Submitted batch job 7321

Test results coming soon (I hope)...

Details

- notification for comment with ID 3108810261 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 · 2025-07-23T16:12:19Z

Test report by @boegelbot

Overview of tested easyconfigs (in order)

SUCCESS jax-0.4.25-gfbf-2023a.eb

Build succeeded for 1 out of 1 (1 easyconfigs in total)
jsczen3c1.int.jsc-zen3.fz-juelich.de - Linux Rocky Linux 9.5, x86_64, AMD EPYC-Milan Processor (zen3), Python 3.9.21
See https://gist.github.com/boegelbot/d5ce16d3e9a97c0b8d86260e6a339f94 for a full test report.

pavelToman · 2025-08-06T14:20:06Z

@boegelbot please test @ jsc-zen3-a100 EB_ARGS="jax-0.4.25-gfbf-2023a-CUDA-12.1.1.eb --parallel 4"

boegelbot · 2025-08-06T14:33:09Z

@pavelToman: 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=3852 EB_ARGS="jax-0.4.25-gfbf-2023a-CUDA-12.1.1.eb --parallel 4" EB_REPO=easybuild-easyblocks EB_BRANCH=develop /opt/software/slurm/bin/sbatch --job-name test_PR_3852 --ntasks=8 ~/boegelbot/eb_from_pr_upload_jsc-zen3.sh' executed!

exit code: 0
output:

Submitted batch job 7505

Test results coming soon (I hope)...

Details

- notification for comment with ID 3160384833 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 · 2025-08-06T17:10:54Z

Test report by @boegelbot

Overview of tested easyconfigs (in order)

FAIL (build issue) jax-0.4.25-gfbf-2023a-CUDA-12.1.1.eb (partial log available at https://gist.github.com/boegelbot/a2bbe5d18367557eaa248a34e9194293)

Build succeeded for 0 out of 1 (1 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/13124e99a036c2867add45cc5f78733f for a full test report.

pavelToman · 2025-08-07T12:16:02Z

Test report by @boegelbot

Overview of tested easyconfigs (in order)
* **FAIL (build issue)** _jax-0.4.25-gfbf-2023a-CUDA-12.1.1.eb_ (partial log available at https://gist.github.com/boegelbot/a2bbe5d18367557eaa248a34e9194293)
Build succeeded for 0 out of 1 (1 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/13124e99a036c2867add45cc5f78733f for a full test report.

/usr/bin/bash: line 1: 197823 Segmentation fault      (core dumped) pytest -vv tests --deselect tests/host_callback_test.py::HostCallbackTapTest::test_tap_scan_custom_jvp --deselect tests/host_callback_test.py::HostCallbackTapTest::test_tap_transforms_doc --deselect tests/lax_scipy_special_functions_test.py::LaxScipySpcialFunctionsTest::testScipySpecialFun_gammainc_s_2x1x4_float32_float32

pavelToman · 2025-08-07T12:17:33Z

@boegelbot please test @ jsc-zen3-a100 EB_ARGS="jax-0.4.25-gfbf-2023a-CUDA-12.1.1.eb --parallel 4"
CORE_CNT=16

boegelbot · 2025-08-07T12:33:09Z

@pavelToman: 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=3852 EB_ARGS="jax-0.4.25-gfbf-2023a-CUDA-12.1.1.eb --parallel 4" EB_REPO=easybuild-easyblocks EB_BRANCH=develop /opt/software/slurm/bin/sbatch --job-name test_PR_3852 --ntasks="16" ~/boegelbot/eb_from_pr_upload_jsc-zen3.sh' executed!

exit code: 0
output:

Submitted batch job 7511

Test results coming soon (I hope)...

Details

- notification for comment with ID 3163947569 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 · 2025-08-07T15:02:19Z

Test report by @boegelbot

Overview of tested easyconfigs (in order)

FAIL (build issue) jax-0.4.25-gfbf-2023a-CUDA-12.1.1.eb (partial log available at https://gist.github.com/boegelbot/d8e94cb28559d85463bfe1b83416d35d)

Build succeeded for 0 out of 1 (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/991eda96ac035f3221e683112950bdf2 for a full test report.

boegel · 2025-08-12T18:29:55Z

@boegelbot please test @ jsc-zen3-a100
EB_ARGS="jax-0.4.25-gfbf-2023a-CUDA-12.1.1.eb --parallel 2"
CORE_CNT=16

boegelbot · 2025-08-12T18:33:09Z

@boegel: 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=3852 EB_ARGS="jax-0.4.25-gfbf-2023a-CUDA-12.1.1.eb --parallel 2" EB_REPO=easybuild-easyblocks EB_BRANCH=develop /opt/software/slurm/bin/sbatch --job-name test_PR_3852 --ntasks="16" ~/boegelbot/eb_from_pr_upload_jsc-zen3.sh' executed!

exit code: 0
output:

Submitted batch job 7565

Test results coming soon (I hope)...

Details

- notification for comment with ID 3180519077 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 · 2025-08-12T23:06:35Z

Test report by @boegelbot

Overview of tested easyconfigs (in order)

FAIL (build issue) jax-0.4.25-gfbf-2023a-CUDA-12.1.1.eb (partial log available at https://gist.github.com/boegelbot/b955975ffbaa753db8dc9f1024895313)

Build succeeded for 0 out of 1 (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/9a2b4058f7db8648aed53432fee6df0a for a full test report.

Thyre · 2025-09-10T13:13:25Z

@Thyre So that means that the fix in boegel/boegelbot#43 isn't working as intended?

@boegel I guess so. The EB_CONTAINER= from the same PR shows up, and trying locally (with some dummy stuff) also transfers the information. I don't see where this fails yet. Unfortunately, debugging is complicated without having the notifications and so on set up. The script itself basically matches what we have for easybuild-easyconfigs.

boegel · 2025-09-10T14:36:47Z

@Thyre So that means that the fix in boegel/boegelbot#43 isn't working as intended?

@boegel I guess so. The EB_CONTAINER= from the same PR shows up, and trying locally (with some dummy stuff) also transfers the information. I don't see where this fails yet. Unfortunately, debugging is complicated without having the notifications and so on set up. The script itself basically matches what we have for easybuild-easyconfigs.

Hopefully fixed with:

Add missing slurm_args to jsc-zen3 EasyBlock tests boegel/boegelbot#44

boegel · 2025-09-10T14:37:02Z

@boegelbot please test @ jsc-zen3-a100
EB_ARGS="jax-0.4.25-gfbf-2023a-CUDA-12.1.1.eb --parallel 4"
CORE_CNT=16

boegelbot · 2025-09-10T14:53:08Z

@boegel: 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=3852 EB_ARGS="jax-0.4.25-gfbf-2023a-CUDA-12.1.1.eb --parallel 4" EB_CONTAINER= EB_REPO=easybuild-easyblocks EB_BRANCH=develop /opt/software/slurm/bin/sbatch --job-name test_PR_3852 --ntasks="16" --partition=jsczen3g --gres=gpu:1 ~/boegelbot/eb_from_pr_upload_jsc-zen3.sh' executed!

exit code: 0
output:

Submitted batch job 7926

Test results coming soon (I hope)...

Details

- notification for comment with ID 3275278393 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).

boegel · 2025-09-10T15:01:26Z

Not it's running on the GPU node:

$ squeue
             JOBID PARTITION     NAME     USER ST       TIME  NODES NODELIST(REASON)
              7926  jsczen3g test_PR_ boegelbo  R       8:05      1 jsczen3g1

boegel · 2025-09-10T15:04:23Z

@boegelbot please test @ jsc-zen3
EB_ARGS="jax-0.3.25-foss-2022a.eb jax-0.4.25-gfbf-2023a.eb --installpath /tmp/$USER/pr3852"
CORE_CNT=16

boegelbot · 2025-09-10T15:13:09Z

@boegel: 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=3852 EB_ARGS="jax-0.3.25-foss-2022a.eb jax-0.4.25-gfbf-2023a.eb --installpath /tmp/$USER/pr3852" EB_CONTAINER= EB_REPO=easybuild-easyblocks EB_BRANCH=develop /opt/software/slurm/bin/sbatch --job-name test_PR_3852 --ntasks="16" ~/boegelbot/eb_from_pr_upload_jsc-zen3.sh' executed!

exit code: 0
output:

Submitted batch job 7927

Test results coming soon (I hope)...

Details

- notification for comment with ID 3275385035 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 · 2025-09-10T16:41:00Z

Test report by @boegelbot

Overview of tested easyconfigs (in order)

SUCCESS jax-0.3.25-foss-2022a.eb
SUCCESS jax-0.4.25-gfbf-2023a.eb

Build succeeded for 2 out of 2 (2 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/27c23310842e7dca84a8cfebe78d1d8a for a full test report.

pavelToman · 2025-09-10T17:25:11Z

test cmd:
eb jax-0.4.25-gfbf-2023a-CUDA-12.1.1.eb --rebuild --robot --include-easyblocks-from-pr=3852 --upload-test-report --cuda-compute-capabilities=9.0

boegelbot · 2025-09-10T19:17:55Z

Test report by @boegelbot

Overview of tested easyconfigs (in order)

SUCCESS jax-0.4.25-gfbf-2023a-CUDA-12.1.1.eb

Build succeeded for 1 out of 1 (1 easyconfigs in total)
jsczen3g1.int.jsc-zen3.fz-juelich.de - Linux Rocky Linux 9.6, x86_64, AMD EPYC-Milan Processor (zen3), 1 x NVIDIA NVIDIA A100 80GB PCIe, 575.57.08, Python 3.9.21
See https://gist.github.com/boegelbot/7a9f72492c6830ba18eb38bfccd397a0 for a full test report.

boegel · 2025-09-10T19:29:45Z

Test report by @boegel

Overview of tested easyconfigs (in order)

FAIL (build issue) jax-0.4.25-gfbf-2023a.eb (partial log available at https://gist.github.com/boegel/385c98851e15f5331dbfdda41994bcac)

Build succeeded for 0 out of 1 (1 easyconfigs in total)
node3590.doduo.os - Linux RHEL 9.4, x86_64, AMD EPYC 7552 48-Core Processor (zen2), Python 3.9.18
See https://gist.github.com/boegel/55c897c91bb563bfcbfb375b8ced9195 for a full test report.

pavelToman · 2025-09-10T20:41:46Z

Test report by @pavelToman

Overview of tested easyconfigs (in order)

FAIL (build issue) jax-0.4.25-gfbf-2023a-CUDA-12.1.1.eb (partial log available at https://gist.github.com/pavelToman/2f456cf44235240d11b5863eae172fa8)

Build succeeded for 0 out of 1 (1 easyconfigs in total)
node4306.litleo.os - Linux RHEL 9.4, x86_64, AMD EPYC 9454P 48-Core Processor, 1 x NVIDIA NVIDIA H100 NVL, 570.133.20, Python 3.9.18
See https://gist.github.com/pavelToman/7301e689e74d95f0d4430692dd2befda for a full test report.

pavelToman · 2025-09-11T12:34:50Z

Test report by @pavelToman

Overview of tested easyconfigs (in order)

SUCCESS jax-0.4.25-gfbf-2023a-CUDA-12.1.1.eb

Build succeeded for 1 out of 1 (1 easyconfigs in total)
node3907.accelgor.os - Linux RHEL 9.4, x86_64, AMD EPYC 7413 24-Core Processor, 1 x NVIDIA NVIDIA A100-SXM4-80GB, 570.133.20, Python 3.9.18
See https://gist.github.com/pavelToman/41c3de0c5fa1b386e5a69a892382b78c for a full test report.

pavelToman · 2025-09-11T14:52:50Z

Test report by @pavelToman

Overview of tested easyconfigs (in order)
* **FAIL (build issue)** _jax-0.4.25-gfbf-2023a-CUDA-12.1.1.eb_ (partial log available at https://gist.github.com/pavelToman/2f456cf44235240d11b5863eae172fa8)
Build succeeded for 0 out of 1 (1 easyconfigs in total) node4306.litleo.os - Linux RHEL 9.4, x86_64, AMD EPYC 9454P 48-Core Processor, 1 x NVIDIA NVIDIA H100 NVL, 570.133.20, Python 3.9.18 See https://gist.github.com/pavelToman/7301e689e74d95f0d4430692dd2befda for a full test report.

>     return backend.compile(built_c, compile_options=options)
E     jaxlib.xla_extension.XlaRuntimeError: INTERNAL: ptxas exited with non-zero error code 65280, output: ptxas /tmp/vsc47063/eb-ba1zon8g/tempfile-node4306.litleo.os-96bcc12a-3732316-63e785e47e7a2, line 49; error   : Unexpected instruction types specified for 'sub'
E     ptxas /tmp/vsc47063/eb-ba1zon8g/tempfile-node4306.litleo.os-96bcc12a-3732316-63e785e47e7a2, line 97; error   : Unexpected instruction types specified for 'sub'
E     ptxas fatal   : Ptx assembly aborted due to errors
E     : If the error message indicates that a file could not be written, please verify that sufficient filesystem space is provided.

jax/_src/compiler.py:237: XlaRuntimeError

This error seems as a problem with gpu H100 and ptxas from CUDA-12.1.1.
There is a LLVM bug?: pytorch/pytorch#118589
In original PR of jax they had the same problem with H100 gpus: easybuilders/easybuild-easyconfigs#20119 (comment)

pavelToman · 2025-09-11T16:07:35Z

Test report by @boegel

Overview of tested easyconfigs (in order)
* **FAIL (build issue)** _jax-0.4.25-gfbf-2023a.eb_ (partial log available at https://gist.github.com/boegel/385c98851e15f5331dbfdda41994bcac)
Build succeeded for 0 out of 1 (1 easyconfigs in total) node3590.doduo.os - Linux RHEL 9.4, x86_64, AMD EPYC 7552 48-Core Processor (zen2), Python 3.9.18 See https://gist.github.com/boegel/55c897c91bb563bfcbfb375b8ced9195 for a full test report.

Not enough available memory? More memory should fix the worker crashed problem

pavelToman · 2025-09-11T19:28:10Z

Test report by @pavelToman

Overview of tested easyconfigs (in order)

SUCCESS jax-0.4.25-gfbf-2023a-CUDA-12.1.1.eb

Build succeeded for 1 out of 1 (1 easyconfigs in total)
node3300.joltik.os - Linux RHEL 9.4, x86_64, Intel(R) Xeon(R) Gold 6242 CPU @ 2.80GHz, 1 x NVIDIA Tesla V100-SXM2-32GB, 570.133.20, Python 3.9.18
See https://gist.github.com/pavelToman/60b95f2b3b4096a81b69c072897b1ecc for a full test report.

pavelToman · 2025-09-12T10:39:21Z

On our gpu clusters joltik and accelgor this easyblock works for jax-0.4.25-gfbf-2023a-CUDA-12.1.1. It needs more memory for testing, I had 56gb.
On cluster litleo with Nvidia H100 jax did not pass all the tests, because of incompatibility of H100 and CUDA-12.1.1. Details: #3852 (comment)

easybuild/easyblocks/j/jaxlib.py

fix logic in jaxlib w.r.t. disabling use of MKL DNN

pavelToman · 2025-09-23T09:51:08Z

@boegelbot please test @ jsc-zen3-a100
EB_ARGS="jax-0.4.25-gfbf-2023a-CUDA-12.1.1.eb --parallel 4"
CORE_CNT=16

boegelbot · 2025-09-23T09:53:10Z

@pavelToman: 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=3852 EB_ARGS="jax-0.4.25-gfbf-2023a-CUDA-12.1.1.eb --parallel 4" EB_CONTAINER= EB_REPO=easybuild-easyblocks EB_BRANCH=develop /opt/software/slurm/bin/sbatch --job-name test_PR_3852 --ntasks="16" --partition=jsczen3g --gres=gpu:1 ~/boegelbot/eb_from_pr_upload_jsc-zen3.sh' executed!

exit code: 0
output:

Submitted batch job 8035

Test results coming soon (I hope)...

Details

- notification for comment with ID 3323206948 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).

boegel · 2025-09-23T13:12:37Z

Test report by @boegel

Overview of tested easyconfigs (in order)

SUCCESS jax-0.4.25-gfbf-2023a-CUDA-12.1.1.eb

Build succeeded for 1 out of 1 (1 easyconfigs in total)
node3300.joltik.os - Linux RHEL 9.6, x86_64, Intel(R) Xeon(R) Gold 6242 CPU @ 2.80GHz (cascadelake), 1 x NVIDIA Tesla V100-SXM2-32GB, 580.82.07, Python 3.9.21
See https://gist.github.com/boegel/a3c27101bfd44bf747eb199c7cf507b1 for a full test report.

boegelbot · 2025-09-23T20:48:49Z

Test report by @boegelbot

Overview of tested easyconfigs (in order)

SUCCESS jax-0.4.25-gfbf-2023a-CUDA-12.1.1.eb

Build succeeded for 1 out of 1 (1 easyconfigs in total)
jsczen3g1.int.jsc-zen3.fz-juelich.de - Linux Rocky Linux 9.6, x86_64, AMD EPYC-Milan Processor (zen3), 1 x NVIDIA NVIDIA A100 80GB PCIe, 580.82.07, Python 3.9.21
See https://gist.github.com/boegelbot/65a0b3e48a036159e2b6da6a5d22e45f for a full test report.

boegel

lgtm

Update jaxlib.py for jax/jaxlib v0.6.2

c45a9da

pavelToman mentioned this pull request Jul 22, 2025

updated of Python packages on top of Python 3.13 for courses (Py4Sci) vscentrum/vsc-software-stack#578

Open

25 tasks

Update jaxlib.py - fix whitespaces

76c8fb0

This was referenced Jul 29, 2025

{ai}[gfbf/2025a] jax v0.6.2 easybuilders/easybuild-easyconfigs#23526

Closed

{ai}[gfbf/2025a] jax v0.7.0 easybuilders/easybuild-easyconfigs#23530

Merged

boegel changed the title ~~Update jaxlib.py for jax/jaxlib v0.6.2~~ Update jaxlib easyblock for jax/jaxlib v0.6.2 Jul 30, 2025

boegel added the update label Jul 30, 2025

boegel added this to the release after 5.1.1 milestone Jul 30, 2025

pavelToman mentioned this pull request Jul 31, 2025

{ai}[gfbf/2024a] jax v0.6.2 easybuilders/easybuild-easyconfigs#23544

Merged

pavelToman changed the title ~~Update jaxlib easyblock for jax/jaxlib v0.6.2~~ Update jaxlib easyblock for jax/jaxlib v0.6.2 and v0.7.0 Sep 10, 2025

pavelToman mentioned this pull request Sep 12, 2025

{ai,bio}[foss/2024a] AlphaFold3 v3.0.1, jax-triton v0.2.0, jaxtyping v0.2.38, dm-tree v0.1.8 w/ CUDA 12.6.0 DRAFT! easybuilders/easybuild-easyconfigs#22067

Draft

15 tasks

fix logic in jaxlib w.r.t. disabling use of MKL DNN

15dd83b

boegel mentioned this pull request Sep 23, 2025

fix logic in jaxlib w.r.t. disabling use of MKL DNN pavelToman/easybuild-easyblocks#1

Merged

boegel requested changes Sep 23, 2025

View reviewed changes

easybuild/easyblocks/j/jaxlib.py Outdated Show resolved Hide resolved

Merge pull request #1 from boegel/jax_update

0eb0f0c

fix logic in jaxlib w.r.t. disabling use of MKL DNN

boegel approved these changes Sep 23, 2025

View reviewed changes

boegel merged commit cc4b1f8 into easybuilders:develop Sep 23, 2025
17 checks passed

pavelToman mentioned this pull request Oct 10, 2025

Update jaxlib easyblock for jax 0.6.2 with CUDA-12.6.0 #3951

Merged

Update jaxlib easyblock for jax/jaxlib v0.6.2 and v0.7.0 #3852

Update jaxlib easyblock for jax/jaxlib v0.6.2 and v0.7.0 #3852

Uh oh!

Conversation

pavelToman commented Jul 22, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

pavelToman commented Jul 22, 2025

Uh oh!

boegelbot commented Jul 22, 2025

Uh oh!

boegelbot commented Jul 22, 2025

Overview of tested easyconfigs (in order)

Uh oh!

pavelToman commented Jul 23, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Overview of tested easyconfigs (in order)

Uh oh!

boegelbot commented Jul 23, 2025

Uh oh!

boegelbot commented Jul 23, 2025

Uh oh!

boegelbot commented Jul 23, 2025

Overview of tested easyconfigs (in order)

Uh oh!

boegelbot commented Jul 23, 2025

Overview of tested easyconfigs (in order)

Uh oh!

pavelToman commented Jul 23, 2025

Uh oh!

boegelbot commented Jul 23, 2025

Uh oh!

boegelbot commented Jul 23, 2025

Overview of tested easyconfigs (in order)

Uh oh!

pavelToman commented Aug 6, 2025

Uh oh!

boegelbot commented Aug 6, 2025

Uh oh!

boegelbot commented Aug 6, 2025

Overview of tested easyconfigs (in order)

Uh oh!

pavelToman commented Aug 7, 2025

Overview of tested easyconfigs (in order)

Uh oh!

pavelToman commented Aug 7, 2025

Uh oh!

boegelbot commented Aug 7, 2025

Uh oh!

boegelbot commented Aug 7, 2025

Overview of tested easyconfigs (in order)

Uh oh!

boegel commented Aug 12, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

boegelbot commented Aug 12, 2025

Uh oh!

boegelbot commented Aug 12, 2025

Overview of tested easyconfigs (in order)

Uh oh!

Thyre commented Sep 10, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

boegel commented Sep 10, 2025

Uh oh!

boegel commented Sep 10, 2025

Uh oh!

boegelbot commented Sep 10, 2025

Uh oh!

boegel commented Sep 10, 2025

Uh oh!

boegel commented Sep 10, 2025

Uh oh!

boegelbot commented Sep 10, 2025

Uh oh!

boegelbot commented Sep 10, 2025

Overview of tested easyconfigs (in order)

Uh oh!

pavelToman commented Sep 10, 2025

Uh oh!

pavelToman commented Jul 22, 2025 •

edited

Loading

pavelToman commented Jul 23, 2025 •

edited

Loading

boegel commented Aug 12, 2025 •

edited

Loading

Thyre commented Sep 10, 2025 •

edited

Loading

pavelToman commented Sep 11, 2025 •

edited

Loading