[all-commits] [llvm/llvm-project] bb3eb0: [ARM] Test unroll behaviour on machines with low o...

[VladiKrapp-Arm via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: bb3eb0ca0cf0fe454f6845d429190cb30e6fa0f5
      https://github.com/llvm/llvm-project/commit/bb3eb0ca0cf0fe454f6845d429190cb30e6fa0f5
  Author: VladiKrapp-Arm <vladi.krapp at arm.com>
  Date:   2024-12-06 (Fri, 06 Dec 2024)

  Changed paths:
    A llvm/test/Transforms/LoopUnroll/ARM/lob-unroll.ll

  Log Message:
  -----------
  [ARM] Test unroll behaviour on machines with low overhead branching (#118692)

Add test for existing loop unroll behaviour.

Current behaviour is the single loop with fmul gets runtime unrolled by
count of 4, with the loop remainder unrolled as the 3 for.body9.us.prol
sections. This is quite a lot of compare and branch, negating the
benefits of the low overhead loop mechanism.



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