androm3da wrote:

> > > I'm fine with this change. Not sure if there's a better fix available for 
> > > this. This change will guard the function definition but the call sites 
> > > for B0 will not work. Should consider renaming the function too?
> > 
> > 
> > Yeah - this solution is not as effective as I thought. Renaming should 
> > work, I'll do that instead.
> 
> Hmm - but renaming breaks existing callers, right? Not sure how to fix this 
> without introducing breakage.

We could - in addition to the push/pop - create aliases that don't conflict 
(e.g. lowercase like `b0`).  That way, folks can include both headers and 
nothing breaks ... until they try `B0()`, at which point they can use`b0()` 
instead.

And at some point in the future we could theoretically deprecate the upper case 
versions.  Or not.

https://github.com/llvm/llvm-project/pull/184539
_______________________________________________
cfe-commits mailing list
[email protected]
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

Reply via email to