projects / platform / upstream / llvm.git / commit
? search:
summary | shortlog | log | commit | commitdiff | tree
(parent: 3e24135) | patch
[RISCV] Improve i64 splat vector lowering in RV32.
authorjacquesguan <Jianjian.Guan@streamcomputing.com>
Tue, 4 Jan 2022 07:41:44 +0000 (15:41 +0800)
committerjacquesguan <Jianjian.Guan@streamcomputing.com>
Fri, 14 Jan 2022 06:06:01 +0000 (14:06 +0800)
commit88c0e0806b36a32e92f97585b85db6e2ec9860ec
tree842ea718c36265cb467ec66d070ac5c6d6d8e8e1tree | snapshot
parent3e241353e1c649bcf35e14cc4498d501319403b0commit | diff
[RISCV] Improve i64 splat vector lowering in RV32.

We could use vmv.v.i/vmv.v.x whose eew is 32 to lower the i64 splat vector if the i64 constant scalar could be splitted into two same i32 scalar.

Differential Revision: https://reviews.llvm.org/D117079
llvm/lib/Target/RISCV/RISCVISelLowering.cpp diff | blob | history
llvm/test/CodeGen/RISCV/rvv/vmv.v.x-rv32.ll diff | blob | history
Domain: System / Toolchain;