[PATCH] D132828: Add new optimization pass of Tree Height Reduction

[KAWASHIMA Takahiro via Phabricator via llvm-commits]

kawashima-fj abandoned this revision.
kawashima-fj added a comment.

In D132828#3757973 <https://reviews.llvm.org/D132828#3757973>, @kawashima-fj wrote:

> In D132828#3757715 <https://reviews.llvm.org/D132828#3757715>, @fhahn wrote:
>
>> Have you considered implementing this as MachineFunctionPass instead of an LLVM IR pass?
>
> No. Just because the original patch D67383 <https://reviews.llvm.org/D67383> was implemented as an LLVM IR pass and I'm not familier with MachineFunctionPass.
>
>> Doing the transformation on MachineIR would allow for more precise cost estimates, including more accurate information about register usage, selected instructions and processor resource usage. `MachineCombiner.cpp` might be interesting example to look at for similar (although simpler) transformations with relatively accurate uarch-driven cost-modeling.
>
> Thanks for your advice. It seems to have an advantage but porting to MachineFunctionPass may take time for me. I'll see about it.

Thank you very much for your information. I found `MachineCombiner.cpp` does similar transformations and adding missing opcodes to an existing function can achieve similar result for AArch64. The patch is D138107 <https://reviews.llvm.org/D138107>. Please review. I abandon this patch.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D132828/new/

https://reviews.llvm.org/D132828