This patch promotes fmaximum/fminimum for fp16 vector to float operation.
Details
Details
Diff Detail
Diff Detail
- Repository
- rG LLVM Github Monorepo
Event Timeline
llvm/lib/Target/RISCV/RISCVISelLowering.cpp | ||
---|---|---|
5503 | nit: maybe it makes sense for this to go in lowerFMAXIMUM_FMINIMUM? |
llvm/lib/Target/RISCV/RISCVISelLowering.cpp | ||
---|---|---|
5503 | Others for promoting fp16 vector operation to float vector operation also put this checking here. |
nit: maybe it makes sense for this to go in lowerFMAXIMUM_FMINIMUM?