Samuel Wein
Samuel Wein
fixed with https://github.com/OpenMS/OpenMS/pull/8494
Its all on several year old intel laptops. Typically i3's. They should all support AVX, but this is the only thing I could think of that would cause illegal instruction...
I can confirm that disabling AVX does make this work on at least some of the affected computers (we're still working on the rest)
> would be interesting to know what kind of hardware they were running Turns out they weren't i3's but rather celeron and pentium brand processors, which (as I just found...
@hliu56 If you have a chance, can you try installing the latest package from here: https://abibuilder.cs.uni-tuebingen.de/archive/openms/OpenMSInstaller/experimental/try_signing/ We're trying to get codesigning working on our side as well.
@jpfeuffer , can we just not count comment rows as rows for our internal count? It would shift the numbers if we referred to anything by row in output.
as of 434c024 we require > 3
> `PACKAGE_TYPE: ${{ steps.set-vars.outputs.pkg_type }}` There's a reason we removed that from the ENV for the build step, and IIRC it's because otherwise the tests don't get built.
closing in favor of https://github.com/OpenMS/OpenMS/pull/8496
@jcharkow I think the testing issues are due to the updated bokeh, rather than the python version. I'm working on adding some code to the testing script to try to...