Skip to content

Conversation

@roystgnr
Copy link
Member

@roystgnr roystgnr commented Nov 6, 2025

Not sure why we don't have more recipes screaming about this, but at least a few of the devel->master recipe compilers caught it.

Not sure why we don't have more recipes screaming about this, but at least a few of the devel->master recipe compilers caught it.
Copy link
Member

@jwpeterson jwpeterson left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

If you want to merge this before all the PR testing finishes, I'm fine with it.

@roystgnr
Copy link
Member Author

roystgnr commented Nov 6, 2025

The only difference between the passing-on-devel "Min gcc" and the failing-on-devel "Min gcc debug" is opt vs dbg builds ... and there the difference is that opt doesn't set -pedantic (or -W -Wall -Wextra -Wno-long-long -Wunused -Wpointer-arith -Wformat -Wparentheses -Woverloaded-virtual ...)

Any objections to us starting to turn on all warnings on all build levels? IMHO opt should entail skipping internal checks that might slow us down at runtime, but not warnings at compile time. This obviously isn't urgent to fix, but I might just change it in ACSM and then we'll pick it up whenever we next update that for something more pressing.

@jwpeterson
Copy link
Member

Hmm... I guess I don't know why opt and dbg mode warning flags should be different, other than we have just done that historically. So I guess I'd be fine with making warning flags the same for all optimization levels.

@roystgnr roystgnr merged commit 480153a into libMesh:devel Nov 6, 2025
15 of 24 checks passed
@roystgnr roystgnr deleted the drop_semicolon branch November 6, 2025 16:44
@roystgnr
Copy link
Member Author

roystgnr commented Nov 6, 2025

This passed at least one of the previously failing recipes; not going to wait on everything.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

2 participants