[PATCH] D72392: [MachineScheduler] Fix the TopDepth/BotHeightReduce latency heuristics

Thu Jan 9 04:14:13 PST 2020

fhahn added a comment.

In D72392#1811735 <https://reviews.llvm.org/D72392#1811735>, @foad wrote:

> The diff in `llvm/test/CodeGen/X86/testb-je-fusion.ll` is:
>
>           movl    %edi, %eax
>   -       addl    $-512, %eax             # imm = 0xFE00
>           movb    $1, (%rsi)
>   +       addl    $-512, %eax             # imm = 0xFE00
>           je      .LBB2_2
>   ...
>           movl    %edi, %eax
>   -       decl    %eax
>           movb    $1, (%rsi)
>   +       decl    %eax
>           je      .LBB3_2
>
>
> The scheduler prefers not to put the addl/decl immediately after the first movl because of the register dependency on eax with latency 1.

Great. My main point about the tests was that I think it would be good to have a dedicated small test that just checks on rod the improved cases, with a clear explanation, to guard against regressions by other scheduler changes with a lot of test changes

Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D72392/new/

https://reviews.llvm.org/D72392