[PATCH] D17664: Move discriminator assignment to the right place.
David Blaikie via llvm-commits
llvm-commits at lists.llvm.org
Fri Feb 26 16:41:31 PST 2016
Ah, great - thanks! (& I'd read that, but was reading the patch backwards,
as though that comment was being added, looking at it in more detail the
patch does look great/going in the right direction to be sure - will leave
it to Diego or others to sign off on, though)
On Fri, Feb 26, 2016 at 4:37 PM, Dehao Chen via llvm-commits <
llvm-commits at lists.llvm.org> wrote:
> danielcdh added a comment.
> It was mentioned in the FIXME comment I just removed:
> - // FIXME: This seems completely wrong.
> - //
> - // 1. If two modules are generated in the same context, then the second
> - // Module will get different discriminators than it would have if it
> - // generated in its own context.
> - // 2. If this function is called after round-tripping to bitcode
> instead of
> - // before, it will give a different (and potentially incorrect!)
> - //
> - // The discriminator should instead be calculated from local information
> - // where it's actually needed. This logic should be moved to
> - // AddDiscriminators::runOnFunction(), where it doesn't pollute the
> - // LLVMContext.
> llvm-commits mailing list
> llvm-commits at lists.llvm.org
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the llvm-commits