cutensor v2.0.2.5

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It is very likely that the current package version for this feedstock is out of date.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

Pending Dependency Version Updates

Here is a list of all the pending dependency version updates for this repo. Please double check all dependencies before merging.

Name	Upstream Version	Current Version
cudatoolkit	12.0.0

Dependency Analysis

We couldn't run dependency analysis due to an internal error in the bot, depfinder, or grayskull. :/ Help is very welcome!

_{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 - please use this URL for debugging.}

[conda-forge-webservices]

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) and found it was in an excellent condition.

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

For recipe:

• You're setting a requirement on sysroot_linux- directly; this should now be done by adding a build dependence on {{ stdlib("c") }}, and overriding c_stdlib_version in recipe/conda_build_config.yaml for the respective platform as necessary. For further details, please see META: {{ stdlib("c") }} migration conda-forge.github.io#2102.

[leofang]

With the migration to stdlib("c"), any explicit mention of sysroot in the recipe (line 58) should be removed.

xref: https://conda-forge.org/news/2024/03/24/stdlib-migration/

[mtjrider]

@conda-forge-admin, please rerender

[github-actions]

Hi! This is the friendly automated conda-forge-webservice.

I tried to rerender for you, but it looks like there was nothing to do.

This message was generated by GitHub actions workflow run https://github.com/conda-forge/cutensor-feedstock/actions/runs/9864702702.

[mtjrider]

@v0i0 and I were working through updating this pull request.
2.0.2.4 is not the correct version.
2.0.2.5 is the correct version, published this morning on devzone.

@leofang @jakirkham is best practice to manually update this PR for 2.0.2.5 or close and rerender?
Separately, should we add @v0i0 to the list of maintainers? He's a dev for cuTENSOR who will be supporting conda.

[jakirkham]

In terms of adding a new maintainer, would ask them to open a new issue using this bot request @conda-forge-admin, please add user @username (where @username is the maintainer's GH handle). Then a existing maintainer can merge the bot PR that is opened

Regarding updating, either approach is ok. Would just make a note of the issue in the prior version when updating

[conda-forge-webservices]

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

I wanted to let you know that I linted all conda-recipes in your PR (recipe/meta.yaml) and found some lint.

Here's what I've got...

For recipe/meta.yaml:

• You're setting a requirement on sysroot_linux- directly; this should now be done by adding a build dependence on {{ stdlib("c") }}, and overriding c_stdlib_version in recipe/conda_build_config.yaml for the respective platform as necessary. For further details, please see META: {{ stdlib("c") }} migration conda-forge.github.io#2102.

[mtjrider]

For more information, see below-- the archive tarball links for 2.0.2 all point to 2.0.2.4, but the redist keys reference libcutensor version 2.0.2.5 in version 2.0.2.1.

I assume the correct version to publish is 2.0.2.4... Looking for confirmation from @v0i0.

Some notes below to capture the information:
Navigating the cuTENSOR release archive and cross-referencing the cuTENSOR redist keys, we find:

• archive 2.0.2 → redist libcutensor version 2.0.2.4
• archive 2.0.2.1 → redist libcutensor version 2.0.2.5

The meta.yaml uses libcutensor versions and keys-- this leads to some confusion when making corrections.

[conda-forge-webservices]

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.

[mtjrider]

With the migration to stdlib("c"), any explicit mention of sysroot in the recipe (line 58) should be removed.

xref: https://conda-forge.org/news/2024/03/24/stdlib-migration/

Done here.

[v0i0]

The 2.0.2 / 2.0.2.4 binaries were lacking a crucial change, so we pushed 2.0.2.1 / 2.0.2.5 immediately after (in python terms, like 2.0.2post1). Thus, it doesn't matter much whether we publish 2.0.2.4 binaries to conda (they are obsolete), but we need to publisg 2.0.2.5 binaries for sure. Does that help?

…

On Tue, Jul 30, 2024, 22:31 Matthew Jones ***@***.***> wrote: For more information, see below-- the archive tarball links for 2.0.2 all point to 2.0.2.4, but the redist keys reference libcutensor version 2.0.2.5 in version 2.0.2.1. I assume the correct version to publish is 2.0.2.4... Looking for confirmation from @v0i0 <https://github.com/v0i0>. Some notes below to capture the information: Navigating the cuTENSOR release archive <https://developer.nvidia.com/cutensor-archive> and cross-referencing the cuTENSOR redist keys, we find: - archive 2.0.2 → redist libcutensor version 2.0.2.4 - archive 2.0.2.1 → redist libcutensor version 2.0.2.5 The meta.yaml uses libcutensor versions and keys-- this leads to some confusion when making corrections. — Reply to this email directly, view it on GitHub <#56 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABWRY3LBJ2JNCBNQE2RV7VLZO7ZZLAVCNFSM6AAAAABJ26F5L2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDENJZGE2TOMBXGE> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[leofang]

Then let's just publish 2.0.2.5. @mtjrider could you help update the version & checksums?

[mtjrider]

@conda-forge-admin, please rerender

[leofang]

Change this line from default to azure and rerender. This will use QEMU to emulate linux-aarch64, but since this repo just does binary repackaging, the speed is not too bad.

cutensor-feedstock/conda-forge.yml

Line 7 in 88f00db

linux_aarch64: default

[mtjrider]

Change this line from default to azure and rerender. This will use QEMU to emulate linux-aarch64, but since this repo just does binary repackaging, the speed is not too bad.

cutensor-feedstock/conda-forge.yml

Line 7 in 88f00db

linux_aarch64: default

Thanks @leofang will do.

[mtjrider]

@conda-forge-admin, please rerender

[leofang]

The cuTENSORMg compilation test for CUDA 11 + aarcht64 just failed without any error message... lol Is it something we can reproduce locally?

[mtjrider]

The cuTENSORMg compilation test for CUDA 11 + aarch64 just failed without any error message... lol Is it something we can reproduce locally?

I am guessing that it’s not parsing the line-breaks correctly in that case. It’s supposed to cat the output redirected from STDERR. I removed the line-breaks to see if that gives us more or fixes the problem.

EDIT: I say this because I see a line-break in the compile command itself for that example around -I{PREFIX}.

[mtjrider]

Added a commit that passes nvcc directly to the shell instead of redirecting output.
I’m switching to local builds for further debugging while this runs.

In previous logs, only stderr would be cat to the session. The only thing I can think is the command is too long (I’m guessing)?

[mtjrider]

The error is related to linking glibc, here.

Snapshot:

/home/conda/feedstock_root/build_artifacts/cutensor_1722533265067/_test_env_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placeho/bin/../lib/gcc/aarch64-conda-linux-gnu/11.4.0/../../../../aarch64-conda-linux-gnu/bin/ld: /home/conda/feedstock_root/build_artifacts/cutensor_1722533265067/_test_env_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placeho/lib/libcutensorMg.so: undefined reference to `logf@GLIBC_2.27'
2024-08-01T18:02:37.9332539Z /home/conda/feedstock_root/build_artifacts/cutensor_1722533265067/_test_env_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placeho/bin/../lib/gcc/aarch64-conda-linux-gnu/11.4.0/../../../../aarch64-conda-linux-gnu/bin/ld: /home/conda/feedstock_root/build_artifacts/cutensor_1722533265067/_test_env_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placehold_placeho/lib/libcutensorMg.so: undefined reference to `expf@GLIBC_2.27'

[leofang]

Ah, so it's a glibc issue, seems like we have no other way but to skip the compilation test in this particular pipeline... conda-forge has not migrated to newer glibc (>2.17) as far as I know (conda-forge/conda-forge.github.io#1941). Let's follow what we did in cuSPARSELt for simplicity.
https://github.com/conda-forge/cusparselt-feedstock/blob/0f8fda896aa5253dee51e72bef071ec1e5897432/recipe/run_test.sh#L7-L8

[mtjrider]

Good to merge @leofang @jakirkham ?

[leofang]

Thanks, @mtjrider!