[PATCH] D131838: [ValueTracking] computeKnownBits - attempt to use a branch condition feeding a phi to improve known bits range (PR38280)

[Simon Pilgrim via Phabricator via llvm-commits]

RKSimon added a comment.

In D131838#3726257 <https://reviews.llvm.org/D131838#3726257>, @spatel wrote:

> The new test changes to the loop latch code are potential trouble if it harms indvars or LSR, so you might want to pipe that through to codegen to see if there's any obvious regression before pushing.

Thanks Sanjay - a quick review of the codegen changes from remove-loop-phi-fastmul.ll suggested it was mainly neutral (x86/riscv64/powerpc64), although arm/aarch64/powerpc32 could struggle on some to schedule to a code order that didn't require an extra reg move. riscv32 was the only case that saw an improvement (both with i32 and i64 indvars).


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D131838/new/

https://reviews.llvm.org/D131838