[all-commits] [llvm/llvm-project] a6ca88: [RISCV] Lower more BUILD_VECTOR sequences to RVV's...

[Fraser Cormack via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: a6ca88e908b5befcd9b0f8c8cb40f53095cc17bc
      https://github.com/llvm/llvm-project/commit/a6ca88e908b5befcd9b0f8c8cb40f53095cc17bc
  Author: Fraser Cormack <fraser at codeplay.com>
  Date:   2021-07-16 (Fri, 16 Jul 2021)

  Changed paths:
    M llvm/lib/Target/RISCV/RISCVISelLowering.cpp
    M llvm/test/CodeGen/RISCV/rvv/fixed-vectors-fp-shuffles.ll
    M llvm/test/CodeGen/RISCV/rvv/fixed-vectors-int-buildvec.ll
    M llvm/test/CodeGen/RISCV/rvv/fixed-vectors-int-shuffles.ll
    M llvm/test/CodeGen/RISCV/rvv/fixed-vectors-int.ll
    M llvm/test/CodeGen/RISCV/rvv/fixed-vectors-stepvector-rv32.ll
    M llvm/test/CodeGen/RISCV/rvv/fixed-vectors-stepvector-rv64.ll

  Log Message:
  -----------
  [RISCV] Lower more BUILD_VECTOR sequences to RVV's VID

This patch teaches the compiler to identify a wider variety of
`BUILD_VECTOR`s which form integer arithmetic sequences, and to lower
them to `vid.v` with modifications for non-unit steps and non-zero
addends.

The sequences handled by this optimization must either be monotonically
increasing or decreasing. Consecutive elements holding the same value
indicate a fractional step which, while simple mathematically,
becomes more complex to handle both in the realm of lossy integer
division and in the presence of `undef`s.

For example, a common "interleaving" shuffle index will be lowered by
LLVM to both `<0,u,1,u,2,...>` and `<u,0,u,1,u,...>` `BUILD_VECTOR`
nodes. Either of these would ideally be lowered to `vid.v` shifted right
by 1. Detection of this sequence in presence of general `undef` values
is more complicated, however: `<0,u,u,1,>` could match either
`<0,0,0,1,>` or `<0,0,1,1,>` depending on later values in the sequence.
Both are possible, so backtracking or multiple passes is inevitable.

Sticking to monotonic sequences keeps the logic simpler as it can be
done in one pass. Fractional steps will likely be a separate
optimization in a future patch.

Reviewed By: craig.topper

Differential Revision: https://reviews.llvm.org/D104921