RPCS3/llvm
1
0
Fork 0
mirror of https://github.com/RPCS3/llvm.git synced 2024-12-11 05:35:11 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

AMDGPU/SI: Make sure to emit TargetConstant nodes when matching ds_*permute

Browse Source 
Summary:
This fixes a bug with ds_*permute instructions where if it was passed a
constant address, then the offset operand would get assigned a register
operand instead of an immediate.

Reviewers: scchan, arsenm

Subscribers: arsenm, llvm-commits

Differential Revision: http://reviews.llvm.org/D19994

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@272349 91177308-0d34-0410-b5e6-96231b3b80d8
This commit is contained in:
Tom Stellard 2016-06-10 00:01:04 +00:00
parent 6c347f62cf
commit 60f588f570
2 changed files with 11 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
lib/Target/AMDGPU/AMDGPUISelDAGToDAG.cpp
View File

@ -876,7 +876,7 @@ bool AMDGPUDAGToDAGISel::SelectDS1Addr1Offset(SDValue Addr, SDValue &Base,
Zero, Addr.getOperand(1));
Base = SDValue(MachineSub, 0);
Offset = Addr.getOperand(0);
Offset = CurDAG->getTargetConstant(ByteOffset, DL, MVT::i16);
return true;
}
}
@ -894,7 +894,7 @@ bool AMDGPUDAGToDAGISel::SelectDS1Addr1Offset(SDValue Addr, SDValue &Base,
MachineSDNode *MovZero = CurDAG->getMachineNode(AMDGPU::V_MOV_B32_e32,
DL, MVT::i32, Zero);
Base = SDValue(MovZero, 0);
Offset = Addr;
Offset = CurDAG->getTargetConstant(CAddr->getZExtValue(), DL, MVT::i16);
return true;
}
}

9
test/CodeGen/AMDGPU/llvm.amdgcn.ds.bpermute.ll
View File

@ -21,4 +21,13 @@ define void @ds_bpermute_imm_offset(i32 addrspace(1)* %out, i32 %base_index, i32
ret void
}
; CHECK-LABEL: {{^}}ds_bpermute_imm_index:
; CHECK: ds_bpermute_b32 v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}} offset:64
; CHECK: s_waitcnt lgkmcnt
define void @ds_bpermute_imm_index(i32 addrspace(1)* %out, i32 %base_index, i32 %src) nounwind {
%bpermute = call i32 @llvm.amdgcn.ds.bpermute(i32 64, i32 %src) #0
store i32 %bpermute, i32 addrspace(1)* %out, align 4
ret void
}
attributes #0 = { nounwind readnone convergent }