Files
clang-p2996/llvm/test/CodeGen/AMDGPU/GlobalISel/inst-select-fmul.mir
Tom Stellard 46bbbc33c0 AMDGPU/GlobalISel: Implement select() for 32-bit G_FADD and G_FMUL
Reviewers: arsenm, nhaehnle

Reviewed By: arsenm

Subscribers: kzhuravl, wdng, yaxunl, rovka, kristof.beyls, dstuttard, tpr, t-tye, llvm-commits

Differential Revision: https://reviews.llvm.org/D46171

llvm-svn: 334665
2018-06-13 22:30:47 +00:00

38 lines
883 B
YAML

# RUN: llc -march=amdgcn -run-pass=instruction-select -verify-machineinstrs -global-isel %s -o - | FileCheck %s -check-prefixes=GCN
--- |
define amdgpu_kernel void @fmul(i32 addrspace(1)* %global0) {ret void}
...
---
name: fmul
legalized: true
regBankSelected: true
# GCN-LABEL: name: fmul
body: |
bb.0:
liveins: $sgpr0, $vgpr0, $vgpr1, $vgpr3_vgpr4
%0:sgpr(s32) = COPY $sgpr0
%1:vgpr(s32) = COPY $vgpr0
%2:vgpr(s32) = COPY $vgpr1
%3:vgpr(s64) = COPY $vgpr3_vgpr4
; fmul vs
; GCN: V_MUL_F32_e64
%4:vgpr(s32) = G_FMUL %1, %0
; fmul sv
; GCN: V_MUL_F32_e64
%5:vgpr(s32) = G_FMUL %0, %1
; fmul vv
; GCN: V_MUL_F32_e64
%6:vgpr(s32) = G_FMUL %1, %2
G_STORE %4, %3 :: (store 4 into %ir.global0)
G_STORE %5, %3 :: (store 4 into %ir.global0)
G_STORE %6, %3 :: (store 4 into %ir.global0)
...
---