[PATCH] D36704: [CodeGen] Improve the consistency of instruction fusion

[Florian Hahn via Phabricator via llvm-commits]

fhahn added a comment.

I think this changes makes sense and thanks for getting rid of the `if` (that's not needed any longer) and making the debug message clearer too! I guess adding a test for the case where SecondSU has dependencies that are scheduled between FirstSU and SecondSU is difficult, as in most fused instruction pairs the second instruction only depends on the first instruction?

I am not sure about excluding control dependencies, but I am probably missing a constraint on control dependencies. Couldn't it be the case that an instruction has a control dependency on FirstSU, so after scheduling FirstSU, we could schedule this instruction instead of SecondSU?


https://reviews.llvm.org/D36704