[PATCH] D85456: Fix instruction counting in post-RA scheduler

[Stanislav Mekhanoshin via Phabricator via llvm-commits]

rampitec added a comment.

In D85456#2200952 <https://reviews.llvm.org/D85456#2200952>, @arsenm wrote:

> I think this is an invalid bundle.
>
> In D85456#2200548 <https://reviews.llvm.org/D85456#2200548>, @rampitec wrote:
>
>> In D85456#2200529 <https://reviews.llvm.org/D85456#2200529>, @arsenm wrote:
>>
>>> Is this really a well formed bundle? i.e. is this really a verifier bug?
>>
>> That's a good question. Technically that is possible to create such IR. It might be OK post-RA?
>> For the context, it was created by the SplitEditor::buildSingleSubRegCopy() in the D30438 <https://reviews.llvm.org/D30438>.
>
> I'm inclined to think it's invalid. It looks like using bundles at all was a hack.  I think VirtRegRewriter::expandCopyBundle just missed the case where a copy becomes a kill.
>
> Can you add a second test where this weird kill bundle thing is produced?

Added splitkit-copy-bundle.mir. It produces a bundle with KILLs.

What's interesting I have created a patch to properly bundle copies in the SplitKit, which has resulted in KILLs surviving until inst printer and crash.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D85456/new/

https://reviews.llvm.org/D85456