[all-commits] [llvm/llvm-project] 8d16c6: [RISCV] Increase default vectorizer LMUL to 2

[Luke Lau via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 8d16c6809a080947057ae21b9f6165105b4b2ad8
      https://github.com/llvm/llvm-project/commit/8d16c6809a080947057ae21b9f6165105b4b2ad8
  Author: Luke Lau <luke at igalia.com>
  Date:   2023-03-23 (Thu, 23 Mar 2023)

  Changed paths:
    M llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
    M llvm/test/Transforms/LoopVectorize/RISCV/defaults.ll
    M llvm/test/Transforms/LoopVectorize/RISCV/divrem.ll
    M llvm/test/Transforms/LoopVectorize/RISCV/illegal-type.ll
    M llvm/test/Transforms/LoopVectorize/RISCV/inloop-reduction.ll
    M llvm/test/Transforms/LoopVectorize/RISCV/interleaved-accesses.ll
    M llvm/test/Transforms/LoopVectorize/RISCV/lmul.ll
    M llvm/test/Transforms/LoopVectorize/RISCV/low-trip-count.ll
    M llvm/test/Transforms/LoopVectorize/RISCV/mask-index-type.ll
    M llvm/test/Transforms/LoopVectorize/RISCV/masked_gather_scatter.ll
    M llvm/test/Transforms/LoopVectorize/RISCV/riscv-interleaved.ll
    M llvm/test/Transforms/LoopVectorize/RISCV/riscv-unroll.ll
    M llvm/test/Transforms/LoopVectorize/RISCV/safe-dep-distance.ll
    M llvm/test/Transforms/LoopVectorize/RISCV/scalable-basics.ll
    M llvm/test/Transforms/LoopVectorize/RISCV/scalable-reductions.ll
    M llvm/test/Transforms/LoopVectorize/RISCV/scalable-tailfold.ll
    M llvm/test/Transforms/LoopVectorize/RISCV/short-trip-count.ll
    M llvm/test/Transforms/LoopVectorize/RISCV/uniform-load-store.ll
    M llvm/test/Transforms/LoopVectorize/RISCV/zvl32b.ll
    M llvm/test/Transforms/SLPVectorizer/RISCV/rvv-min-vector-size.ll

  Log Message:
  -----------
  [RISCV] Increase default vectorizer LMUL to 2

After some discussion and experimentation, we have seen that changing the default number of vector register bits to LMUL=2 strikes a sweet spot.
Whilst we could be clever here and make the vectorizer smarter about dynamically selecting an LMUL that
a) Doesn't affect register pressure
b) Suitable for the microarchitecture
we would need to teach its heuristics about RISC-V register grouping specifics.
Instead this just does the easy, pragmatic thing by changing the default to a safe value that doesn't affect register pressure signifcantly[1], but should increase throughput and unlock more interleaving.

[1] Register spilling when compiling sqlite at various levels of `-riscv-v-register-bit-width-lmul`:

LMUL=1    2573 spills
LMUL=2    2583 spills
LMUL=4    2819 spills
LMUL=8    3256 spills

Reviewed By: craig.topper

Differential Revision: https://reviews.llvm.org/D143723