search

conda-forge / suitesparse-feedstock

A conda-smithy repository for suitesparse.
BSD 3-Clause "New" or "Revised" License
1 stars 16 forks source link

Rebuild for Python 3.7, GCC 7, R 3.5.1, openBLAS 0.3.2 #40

Closed regro-cf-autotick-bot closed 6 years ago

regro-cf-autotick-bot commented 6 years ago

It is likely this feedstock needs to be rebuilt. Notes and instructions for merging this PR:

  1. Please merge the PR only after the tests have passed.
  2. Feel free to push to the bot's branch to update this PR if needed.

Please note that if you close this PR we presume that the feedstock has been rebuilt, so if you are going to perform the rebuild yourself don't close this PR until the your rebuild has been merged.

This package has the following downstream children: cvxopt julia octave scikit-sparse scikit-umfpack And potentially more.

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.

This PR was created by the cf-regro-autotick-bot. The cf-regro-autotick-bot is a service to automatically track the dependency graph, migrate packages, and propose package version updates for conda-forge. If you would like a local version of this bot, you might consider using rever. Rever is a tool for automating software releases and forms the backbone of the bot's conda-forge PRing capability. Rever is both conda (conda install -c conda-forge rever) and pip (pip install re-ver) installable. Finally, feel free to drop us a line if there are any issues!

conda-forge-linter commented 6 years ago

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.

jakirkham commented 6 years ago

@conda-forge-admin, please re-render.

conda-forge-linter commented 6 years ago

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

I tried to re-render for you, but it looks like there was nothing to do.

jakirkham commented 6 years ago

Linux build with new compilers is going to need some work. Somehow its build of libumfpack is broken as it can't find symbols from libcholmod. The rest of the builds (including macOS with both compilers) are fine though.

ref: https://circleci.com/gh/conda-forge/suitesparse-feedstock/115

minrk commented 6 years ago

@jakirkham #41 gets the builds passing by removing -Wl,--as-needed and adding -rpath-link to one flag.

jakirkham commented 6 years ago

Thanks @minrk for narrowing that down and fixing it.

Yeah -Wl,--as-needed is problematic in some cases. So it doesn't surprise me that we might need to drop that from some builds. Though am surprised how far we have gotten through the stack without needing to drop it.

minrk commented 6 years ago

The main situation I've found as-needed to fail is when one dylib exists to 'bundle' a collection of others that it doesn't itself use. umfpack apparently does this. I think it's not very common to do that, which is probably why it hasn't come up.

I really don't understand with -rpath-link is sometimes needed with the new compilers and never the old ones, though.

jakirkham commented 6 years ago

Yep, that makes perfect sense. I've also run into situations where link order mattered and was enforced through a similar strategy.

It's probably a consequence of the Anaconda compilers being built with a custom sysroot. That said, it's probably worth raising with the Anaconda team. Maybe they can offer some more insight. Also wouldn't hurt if they baked the -rpath-link option into the compiler flags for us.