This patch adds incrementally-better support for SPLAT_VECTOR in a
handful of vector combines by changing a few more
isBuildVectorAllOnes/isBuildVectorAllZeros to the equivalent
isConstantSplatVectorAllOnes/Zeros calls.
Details
Details
Diff Detail
Diff Detail
- Repository
- rG LLVM Github Monorepo
Event Timeline
llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp | ||
---|---|---|
10054 | I wasn't sure if I should reword this comment. Unsure how it could be done without just making it more verbose. |
llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp | ||
---|---|---|
10054 | Perhaps just drop build_vector? |
llvm/lib/CodeGen/SelectionDAG/DAGCombiner.cpp | ||
---|---|---|
10054 | Aye that looks pretty good, thanks. |
I wasn't sure if I should reword this comment. Unsure how it could be done without just making it more verbose.