Skip to content

Conversation

@DJDavies2
Copy link
Collaborator

Fix #265.

@matthewrmshin
Copy link
Collaborator

I wonder if it would be more portable using INT32 from the ISO_FORTRAN_ENV module?

@DJDavies2
Copy link
Collaborator Author

Yes, there are much more correct things to do that this.

Or look at removing the MATMUL_ESSL branch altogether? I don't think it is used (but then is this whole file even used?).

I don't particularly want to go down a rabbit hole at this point.

@DJDavies2 DJDavies2 marked this pull request as draft January 23, 2026 13:25
@DJDavies2
Copy link
Collaborator Author

Heh, I followed up with the next warning from Fortitude and it promptly complained about the "KIND=4", suggestiong use of ISO_FORTRAN_ENV. So maybe I should do that here.

@DJDavies2
Copy link
Collaborator Author

I've updated the branch to use int32. This requires updating the compiler flags to 2008 standard. There is some logic in the cmake file that tests the version; we are clearly always using at least 10.1.0 now so maybe this test can simply be removed? I have left it in for now though.

@DJDavies2 DJDavies2 marked this pull request as ready for review January 26, 2026 19:03
@matthewrmshin
Copy link
Collaborator

I've updated the branch to use int32. This requires updating the compiler flags to 2008 standard. There is some logic in the cmake file that tests the version; we are clearly always using at least 10.1.0 now so maybe this test can simply be removed? I have left it in for now though.

Yes, I believe we can remove this test, as we should be on GNU 12 or above from now on.

@DJDavies2
Copy link
Collaborator Author

I've updated the branch to use int32. This requires updating the compiler flags to 2008 standard. There is some logic in the cmake file that tests the version; we are clearly always using at least 10.1.0 now so maybe this test can simply be removed? I have left it in for now though.

Yes, I believe we can remove this test, as we should be on GNU 12 or above from now on.

I'll open another PR for this, it probably shouldn't be done here.

@matthewrmshin matthewrmshin merged commit 32c1006 into develop Jan 27, 2026
6 checks passed
@matthewrmshin matthewrmshin deleted the feature/265 branch January 27, 2026 16:22
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

INTEGER*4 declarations

4 participants