XERBLA tweaks plus a tweak to GenerateSuffixedSource.cmake - #1411
Conversation
|
@mohawk2 Do we want to add the corresponding changes in the BLAS testing here as well? |
Those changes were small enough I felt they weren't a distraction. |
a108103 to
75a7dc0
Compare
|
@ACSimon33 Force-pushed. Please "resolve" if you feel your excellent point is now addressed :-) |
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## master #1411 +/- ##
=======================================
Coverage 69.36% 69.36%
=======================================
Files 6122 6122
Lines 486711 486717 +6
Branches 23268 23268
=======================================
+ Hits 337584 337590 +6
Misses 148689 148689
Partials 438 438
Continue to review full report in Codecov by Harness.
|
0f0e3e2 to
6eee65b
Compare
6eee65b to
c674ef2
Compare
|
@ACSimon33 Thank you for all your help so far! If you have capacity, could you steer me on what else is needed here? There's not much I can do about the failing coverage check. If this were to be merged, it would greatly simplify review of #1407. |
|
@martin-frbg If you have capacity, could you look this over? Hopefully low-impact by itself. |
|
Sorry, I'm barely able to keep my own house in order (OpenBLAS) so I can only try to contribute quick and easy reviews here |
c674ef2 to
bb3f52a
Compare
|
@ACSimon33 Thank you! By the way, I wondered: you have write access, and it's saying "At least 1 approving review is required by reviewers with write access." - doesn't that include you? |
|
Also, what tool did you use for the reformatting? I'm assuming (hoping!) it wasn't by hand. |
I have write access to master, but I can't approve PRs (probably that's only maintainers?).
For stuff like that, I let my Claude Agent do it, and then I just check it afterward. |
Surprising, but it is what it is. Now I just have to wait for e.g. @langou etc to have available capacity :-) |
My understanding: The rule is that you cannot approve your own PRs. (I cannot approve my PRs.) All PRs require the approval of someone else. Then once approved, anyone with write acces can merge. I think you can approve PRs from other people. (Feel welcome to try.) |
|
Speaking of me not being able to approve my own PRs, if someone with WRITE access can review: #1328. That'd be great. This PR has been sitting for a while now and waiting either for an approval, or then, if no approval, we'll just close the PR. |
Ok ... turns out I'm so used to Gitlab, I didn't know you how to approve PRs in GitHub. Thanks for letting me know that it should work 👍 |
4 participants
Description
This superficially large PR contains a small improvement to
GenerateSuffixedSource.cmake, and a patch to the various test files that overrideXERBLAso that the override calls into the overriding replacement routine. This is intended as part of #1407, greatly reducing the "diff noise" and making it comprehensible. That is currently breaking onflang; I can fix that, but that PR will become smaller and easier to reason about after this is applied.@ACSimon33 As mentioned on that PR. I have omitted the commit on the PR that detects
ENTRYbecause that is apparently unworkable on currentflang.Checklist