[PATCH] D78398: [DebugInfo] Factor out SalvageDebugInfoForDbgValues() from InstCombine

[Chris Jackson via Phabricator via llvm-commits]

chrisjackson added a comment.

In D78398#1992390 <https://reviews.llvm.org/D78398#1992390>, @Orlando wrote:

> Assuming D78369 <https://reviews.llvm.org/D78369> lands as it is right now; IMO it might make sense to push the OrUndef behaviour of `salvageDebugInfo` into `salvageDebugInfoForDbgValues`, and have `salvageDebugInfo` simply call `salvageDebugInfoForDbgValues(I, findDbgUsers(I))`.
>
> This is pretty similar to what happens pre-D78369, except that `salvageDebugInfo` and  `salvageDebugInfoForDbgValues` both MarkUndef if they fail to salvage. I'm fairly confident you could then simplify the instcombine sink salvaging with this setup.
>
> What do you both think?


This seems reasonable. I'll complete the requested changes to this diff then I'll submit one with @Orlando's suggested approach and we can see which is looks better.

In D78398#1992390 <https://reviews.llvm.org/D78398#1992390>, @Orlando wrote:

> Assuming D78369 <https://reviews.llvm.org/D78369> lands as it is right now; IMO it might make sense to push the OrUndef behaviour of `salvageDebugInfo` into `salvageDebugInfoForDbgValues`, and have `salvageDebugInfo` simply call `salvageDebugInfoForDbgValues(I, findDbgUsers(I))`.
>
> This is pretty similar to what happens pre-D78369, except that `salvageDebugInfo` and  `salvageDebugInfoForDbgValues` both MarkUndef if they fail to salvage. I'm fairly confident you could then simplify the instcombine sink salvaging with this setup.
>
> What do you both think?


I have uploaded a diff implementing this suggested alternative, D79863 <https://reviews.llvm.org/D79863>.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D78398/new/

https://reviews.llvm.org/D78398