jaxlib v0.5.3

[regro-cf-autotick-bot]

It is very likely that the current package version for this feedstock is out of date.

Checklist before merging this PR:

• Dependencies have been updated if changed: see upstream
• Tests have passed
• Updated license if changed and license_file is packaged

Information about this PR:

1. Feel free to push to the bot's branch to update this PR if needed.
2. The bot will almost always only open one PR per version.
3. The bot will stop issuing PRs if more than 3 version bump PRs generated by the bot are open. If you don't want to package a particular version please close the PR.
4. If you want these PRs to be merged automatically, make an issue with @conda-forge-admin,please add bot automerge in the title and merge the resulting PR. This command will add our bot automerge feature to your feedstock.
5. If this PR was opened in error or needs to be updated please add the bot-rerun label to this PR. The bot will close this PR and schedule another one. If you do not have permissions to add this label, you can use the phrase @conda-forge-admin, please rerun bot in a PR comment to have the conda-forge-admin add it for you.

_{This PR was created by the regro-cf-autotick-bot. The regro-cf-autotick-bot is a service to automatically track the dependency graph, migrate packages, and propose package version updates for conda-forge. Feel free to drop us a line if there are any issues! This PR was generated by https://github.com/regro/cf-scripts/actions/runs/13977737888 - please use this URL for debugging.}

[conda-forge-admin]

Hi! This is the friendly automated conda-forge-linting service.

I just wanted to let you know that I linted all conda-recipes in your PR (recipe/meta.yaml) and found it was in an excellent condition.

I do have some suggestions for making it better though...

For recipe/meta.yaml:

• ℹ️ The recipe is not parsable by parser conda-souschef (grayskull). This parser is not currently used by conda-forge, but may be in the future. We are collecting information to see which recipes are compatible with grayskull.

_{This message was generated by GitHub Actions workflow run https://github.com/conda-forge/conda-forge-webservices/actions/runs/15061437288. Examine the logs at this URL for more detail.}

[xhochy]

Not sure how to fix

FileNotFoundError: [Errno 2] No such file or directory: '/home/conda/feedstock_root/build_artifacts/jaxlib_1742524828574/_build_env/share/bazel/5dc7bdbc8139dad09bdc51c397651d85/execroot/__main__/bazel-out/k8-opt/bin/jaxlib/tools/build_wheel.runfiles/pypi_setuptools/site-packages/setuptools/_vendor/jaraco/text/Lorem ipsum.txt'

[h-vetinari]

<bazel...>/jaxlib/tools/build_wheel.runfiles/pypi_setuptools/site-packages/setuptools/_vendor/jaraco/text/Lorem ipsum.txt

Maybe something changed in newer setuptools, and we just need to pin to an older version?

[beckermr]

I am trying a pin of <75 to see what happens here.

[beckermr]

xref: Quansight/open-gpu-server#53 for the cirun failure

I thought I already had access but I think somehow my name was lost when we moved to the production system.

cc @jaimergp

[beckermr]

See here: pypa/setuptools#4605

It is an issue with Bazel interacting with setuptools vendoring and paths with spaces.

[beckermr]

@conda-forge-admin rerender

[beckermr]

So upstream uses bazel 7.4.1 which should have this issue fixed. We should get the build here to pass with that bazel version.

[h-vetinari]

So upstream uses bazel 7.4.1 which should have this issue fixed.

Which issue are you referring to? And why should the same fix not be present in 7.5 or 7.6? I don't think the minor version should matter very much here. If it does, we need to rebuild bazel 7.4 for newer protobuf.

[beckermr]

I was referring to the issue I linked in a comment. I tried with bazel 7.5 and got a really cryptic error I did not know how to debug. Maybe you can make sense of it?

[h-vetinari]

See here: pypa/setuptools#4605

It is an issue with Bazel interacting with setuptools vendoring and paths with spaces.

Based on that issue, we should use setuptools <71. You only tried <75 AFAICT

[beckermr]

Bazel bundles its own copy of setuptools. The one in the host environment is ignored. It's bundled copy is at version 70.0.0.

[h-vetinari]

OK. Bazel 7.5 seems to have worked on linux in b13564b. I can't see the errors on osx because azure deletes the logs of cancelled runs

[beckermr]

jobs are running now...

[beckermr]

INFO: ToolchainResolution: Target platform @@local_execution_config_platform//:platform: Selected execution platform @@local_execution_config_platform//:platform, type @@bazel_tools//tools/cpp:toolchain_type -> toolchain //bazel_toolchain:cc-compiler-build
ERROR: /Users/runner/miniforge3/conda-bld/jaxlib_1743547443983/work/bazel_toolchain/BUILD:42:19: in cc_toolchain_suite rule //bazel_toolchain:toolchain: cc_toolchain_suite '//bazel_toolchain:toolchain' does not contain a toolchain for cpu 'darwin_x86_64'
ERROR: /Users/runner/miniforge3/conda-bld/jaxlib_1743547443983/work/bazel_toolchain/BUILD:42:19: Analysis of target '//bazel_toolchain:toolchain' failed
ERROR: Analysis of target '//jaxlib/tools:build_wheel' failed; build aborted: Analysis failed

[h-vetinari]

Yeah, this error is probably not related to 7.5 vs. 7.4, but either a bazel 6 vs. 7 behaviour change, or that upstream jax removed some config that recognizes our cross-compiling toolchain (interestingly, osx-64 looks fine).

[beckermr]

I think we should probably solve this osx-arm64 error then since the setuptools one needs at least 7.4.

[lucascolley]

hello, is there an issue tracking what is blocking here? Or does it require more investigation?

[beckermr]

This needs more investigation and is likely an issue with the build in conda-forge, not upstream. Someone who knows more about bazel and this build will need to pick it up.

[h-vetinari]

Thanks a lot for fixing this @xhochy! 🙏

[flferretti]

@conda-forge/jaxlib this appears to be affected by #312

[ehfd]

@conda-forge/jaxlib this appears to be affected by #312

Please reconfirm. CUDNN was fixed.

[flferretti]

@conda-forge/jaxlib this appears to be affected by #312

Please reconfirm. CUDNN was fixed.

Confirmed. Fixed by conda-forge/conda-forge-repodata-patches-feedstock#1028