[PATCH] D84464: AMDGPU: Skip other terminators before inserting s_cbranch_exec[n]z

[Matt Arsenault via Phabricator via llvm-commits]

arsenm added a comment.

In D84464#2172405 <https://reviews.llvm.org/D84464#2172405>, @nhaehnle wrote:

> It's hard to say whether this is really correct given how underdefined the control flow pseudos are, but it seems like a pretty reasonable change to me. Assuming you've properly tested this change, LGTM


This isn't the only insert point trick we need to play, it was just only really bad for SI_END_CF before. It needs to be moved to insert itself before lowered phi destination copies. This is the complementary problem on the phi source block side


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D84464/new/

https://reviews.llvm.org/D84464