[all-commits] [llvm/llvm-project] d606ea: [LV] Stop using the legacy cost model for udiv + f...

[David Sherwood via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: d606eae2ced1884a95df9511a867e3faee09ee3c
      https://github.com/llvm/llvm-project/commit/d606eae2ced1884a95df9511a867e3faee09ee3c
  Author: David Sherwood <david.sherwood at arm.com>
  Date:   2025-08-26 (Tue, 26 Aug 2025)

  Changed paths:
    M llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
    M llvm/lib/Transforms/Vectorize/VPlanRecipes.cpp

  Log Message:
  -----------
  [LV] Stop using the legacy cost model for udiv + friends (#152707)

In VPWidenRecipe::computeCost for the instructions udiv, sdiv, urem and
srem we fall back on the legacy cost unnecessarily. At this point we
know that the vplan must be functionally correct, i.e. if the
divide/remainder is not safe to speculatively execute then we must have
either:

1. Scalarised the operation, in which case we wouldn't be using a
VPWidenRecipe, or
2. We've inserted a select for the second operand to ensure we don't
fault through divide-by-zero.

For 2) it's necessary to add the select operation to
VPInstruction::computeCost so that we mirror the cost of the legacy cost
model. The only problem with this is that we also generate selects in
vplan for predicated loops with reductions, which *aren't* accounted for
in the legacy cost model. In order to prevent asserts firing I've also
added the selects to precomputeCosts to ensure the legacy costs match
the vplan costs for reductions.



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