BLD: improve M1 build support #3595
Conversation
* the changes here allow me to build MDAnalysis and its dependencies on an M1 Mac on the gcc compile farm using `python3 -m pip install --user .` * the full test suite passes, except for 1 extra `xfail` I've added in here--which is also documented by gh-1389 (it has apparently been failing on "exotic platforms" for almost 5 years)
|
Hello @tylerjereddy! Thanks for updating this PR. We checked the lines you've touched for PEP 8 issues, and found:
Comment last updated at 2022-04-03 18:27:39 UTC |
| @@ -444,6 +445,8 @@ def test_clustering_two_ensembles(self, ens1, ens2): | |||
| assert len(cluster_collection) == expected_value, "Unexpected " \ | |||
| "results: {0}".format(cluster_collection) | |||
|
|
|||
| @pytest.mark.xfail(platform.machine() == "arm64" and platform.system() == "Darwin", | |||
| reason="Fails on M1 Mac") | |||
There was a problem hiding this comment.
could also reference the GH issue proper here, I just didn't search for it until after I opened the PR
It's not been an issue on ppc64le or aarch64 (non-M1) though, it might be good to supplement that original issue with a new one, it's not immediately obvious to me why this is failing, is Line 287 in 009c467 |
Codecov Report
@@ Coverage Diff @@
## develop MDAnalysis/mdanalysis#3595 +/- ##
===========================================
+ Coverage 94.20% 94.21% +0.01%
===========================================
Files 190 190
Lines 24668 24725 +57
Branches 3313 3321 +8
===========================================
+ Hits 23238 23295 +57
+ Misses 1384 1382 -2
- Partials 46 48 +2
Continue to review full report at Codecov.
|
I tried that--it actually made the situation worse on the M1. I can post what happens if you do that in an issue later maybe. |
How very odd. Yeah if you could either open a new issue or add to the old one that'd be great. Otherwise I'm happy to merge this. |
* explicit reference the issue related to M1 Mac `xfail` for `test_clustering_three_ensembles_two_identical()`
|
Ok, I added a more detailed description of the M1 situation for this test in MDAnalysis/mdaencore#28, and revised this PR to reference that issue instead of vaguely referring to a test failure on M1. |
|
/azp run |
|
Azure Pipelines successfully started running 1 pipeline(s). |
Deals with part of #3592
the changes here allow me to build MDAnalysis and its
dependencies on an M1 Mac on the gcc compile farm using
python3 -m pip install --user .the full test suite passes, except for 1 extra
xfailI've added in here--which is also documented by FAIL: test_clustering_three_ensembles_two_identical on non-x86 64bit arches (aarch64, ppc64, ppc64le and s390x) mdaencore#38
(it has apparently been failing on "exotic platforms"
for almost 5 years)
PR Checklist