[all-commits] [llvm/llvm-project] 7f1b9a: [RISCV] Add MachineCombiner to fold (sh3add Z, (ad...

[Craig Topper via All-commits]

  Branch: refs/heads/main
  Home:   https://github.com/llvm/llvm-project
  Commit: 7f1b9adfc8d86c77ee87a268b3d30e0eda8ed493
      https://github.com/llvm/llvm-project/commit/7f1b9adfc8d86c77ee87a268b3d30e0eda8ed493
  Author: Craig Topper <craig.topper at sifive.com>
  Date:   2024-04-10 (Wed, 10 Apr 2024)

  Changed paths:
    M llvm/include/llvm/CodeGen/MachineCombinerPattern.h
    M llvm/lib/Target/RISCV/RISCVInstrInfo.cpp
    M llvm/test/CodeGen/RISCV/rv64zba.ll

  Log Message:
  -----------
  [RISCV] Add MachineCombiner to fold (sh3add Z, (add X, (slli Y, 6))) -> (sh3add (sh3add Y, Z), X). (#87884)

This improves a pattern that occurs in 531.deepsjeng_r. Reducing the
dynamic instruction count by 0.5%.

This may be possible to improve in SelectionDAG, but given the special
cases around shXadd formation, it's not obvious it can be done in a
robust way without adding multiple special cases.

I've used a GEP with 2 indices because that mostly closely resembles the
motivating case. Most of the test cases are the simplest GEP case. One
test has a logical right shift on an index which is closer to the
deepsjeng code. This requires special handling in isel to reverse a
DAGCombiner canonicalization that turns a pair of shifts into (srl (and
X, C1), C2).



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