DanielCChen wrote:

> > The `-finit-local=zero` (`-finit-local-zero` is an aliasing to 
> > `-finit-local=zero`) mode in PR #216164 should supersede this PR, with an 
> > exception that it doesn't yet initialize CHARACTER type array as documented 
> > as a known limitation. We can look into that as a subsequent PR.
> 
> Dear @DanielCChen,
> 
> Thanks for the work on the other PR. I am okay with merging the other PR, 
> subject to approval from other reviewers. We would still like to have support 
> for gfortran family of options, but I think that could be added as a thin 
> wiring in the driver to your option (as in a user defined `-finit-local-zero` 
> wires internally to your option). Of course, that could come as a separate 
> set of PRs (we will also contribute on that).
> 
> Hope that is okay

Thank you very much @NimishMishra for confirming!
PR #216164 should already have `-finit-local-zero` available to users with the 
underline implementation as an aliasing to `-finit-local=zero`. We can 
definitely look into it as a separate PR if it needs more tuning and adjustment.

https://github.com/llvm/llvm-project/pull/159788
_______________________________________________
cfe-commits mailing list
[email protected]
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

Reply via email to