[all-commits] [llvm/llvm-project] ac5e27: [X86] Skip AMX type lowering when AMX is not used ...

[aengelke via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: ac5e278067033636d51bbee41a6ddb9869d0455b
      https://github.com/llvm/llvm-project/commit/ac5e278067033636d51bbee41a6ddb9869d0455b
  Author: aengelke <engelke at in.tum.de>
  Date:   2024-06-06 (Thu, 06 Jun 2024)

  Changed paths:
    M llvm/lib/Target/X86/X86LowerAMXType.cpp

  Log Message:
  -----------
  [X86] Skip AMX type lowering when AMX is not used (#92910)

The pass iterates over the IR multiple times, but most code doesn't use
AMX. Therefore, do a single iteration in advance to check whether a
function uses AMX at all, and exit early if it doesn't. This makes the
function-has-AMX path slightly more expensive, but AMX users probably
care a lot less about compile time than JIT users (which tend to not use
AMX).

For us, it reduces the time spent in this pass from 0.62% to 0.12%.

Ideally, we wouldn't even need to iterate over the function to determine
that it doesn't use AMX.



To unsubscribe from these emails, change your notification settings at https://github.com/llvm/llvm-project/settings/notifications