maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity

AMDGPU: Fix v2i64<->v4f32 bitcast 

I'm not sure how to test the v2i64->v4f32 case since I can't think of
any v2i64 cases that won't legalize to v4i32.
This commit is contained in:
Matt Arsenault 2020-02-20 09:42:32 -05:00 committed by Matt Arsenault
parent 15e20dcb8f
commit 083717cf49
2 changed files with 13 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
llvm/lib/Target/AMDGPU/SIInstructions.td
View File

@ -1123,6 +1123,8 @@ def : BitConvert <v4f32, v2f64, VReg_128>;
def : BitConvert <v4i32, v2f64, VReg_128>;
def : BitConvert <v2i64, v2f64, VReg_128>;
def : BitConvert <v2f64, v2i64, VReg_128>;
def : BitConvert <v4f32, v2i64, VReg_128>;
def : BitConvert <v2i64, v4f32, VReg_128>;
// 160-bit bitcast
def : BitConvert <v5i32, v5f32, SGPR_160>;

11
llvm/test/CodeGen/AMDGPU/amdgcn.bitcast.ll
View File

@ -287,3 +287,14 @@ define amdgpu_kernel void @v2i32_to_v4f16(<4 x half> addrspace(1)* %out, <2 x i3
store <4 x half> %add.bitcast, <4 x half> addrspace(1)* %out
ret void
}
declare <4 x float> @llvm.amdgcn.s.buffer.load.v4f32(<4 x i32>, i32, i32 immarg)
; FUNC-LABEL: {{^}}bitcast_v4f32_to_v2i64:
; GCN: s_buffer_load_dwordx4
define <2 x i64> @bitcast_v4f32_to_v2i64(<2 x i64> %arg) {
%val = call <4 x float> @llvm.amdgcn.s.buffer.load.v4f32(<4 x i32> undef, i32 0, i32 0)
%cast = bitcast <4 x float> %val to <2 x i64>
%div = udiv <2 x i64> %cast, %arg
ret <2 x i64> %div
}