[PATCH] D36338: BlockPlacement: Consider hotness of blocks relative to a loop iteration rather than relative to the loop as a whole

[Richard Smith - zygoloid via Phabricator via llvm-commits]

rsmith added a comment.

In https://reviews.llvm.org/D36338#832543, @davidxl wrote:

> This looks fine to me. Is there any performance data with PGO?


I've done some measurements, and found that while some benchmarks improve significantly, others regress significantly. I'd like to take a look at the regressions before pushing forward more with this.


Repository:
  rL LLVM

https://reviews.llvm.org/D36338