[all-commits] [llvm/llvm-project] a761e2: [RISCV] Allow non-loop invariant steps in RISCVGat...

[Luke Lau via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: a761e26b2364ea457b79b9a4bea6d792e4913d24
      https://github.com/llvm/llvm-project/commit/a761e26b2364ea457b79b9a4bea6d792e4913d24
  Author: Luke Lau <luke at igalia.com>
  Date:   2025-01-17 (Fri, 17 Jan 2025)

  Changed paths:
    M llvm/lib/Target/RISCV/RISCVGatherScatterLowering.cpp
    M llvm/test/CodeGen/RISCV/rvv/fixed-vectors-strided-load-store.ll
    M llvm/test/CodeGen/RISCV/rvv/strided-load-store.ll

  Log Message:
  -----------
  [RISCV] Allow non-loop invariant steps in RISCVGatherScatterLowering (#122244)

The motivation for this is to allow us to match strided accesses that
are emitted from the loop vectorizer with EVL tail folding (see #122232)

In these loops the step isn't loop invariant and is based off of
@llvm.experimental.get.vector.length.

We can relax this as long as we make sure to construct the updates after
the definition inside the loop, instead of the preheader.

I presume the restriction was previously added so that the step would
dominate the insertion point in the preheader. I can't think of why it
wouldn't be safe to calculate it in the loop otherwise.



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