[PATCH] D89341: [X86] Encode global symbol address in sub if possible

[Wei Wang via Phabricator via llvm-commits]

weiwang added a comment.

In D89341#2328984 <https://reviews.llvm.org/D89341#2328984>, @wenlei wrote:

> Would be good to share benchmark/performance numbers you have for this change too. Thanks.

Sure. This change is to address some codegen difference we saw between clang and gcc from internal workloads. Gcc seems to prefer encoding address immediate into `cmp` in small code model. With the change, we saw 1% perf improvement on average across multiple workloads.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D89341/new/

https://reviews.llvm.org/D89341