1
0
mirror of https://github.com/RPCS3/llvm-mirror.git synced 2024-10-21 03:53:04 +02:00
llvm-mirror/test/CodeGen/AMDGPU/llvm.amdgcn.ds.bpermute.ll
Tom Stellard 7240ff2203 AMDGPU/SI: Make sure to emit TargetConstant nodes when matching ds_*permute
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

llvm-svn: 272349
2016-06-10 00:01:04 +00:00

34 lines
1.3 KiB
LLVM

; RUN: llc -mtriple=amdgcn--amdhsa -mcpu=fiji -verify-machineinstrs < %s | FileCheck %s
declare i32 @llvm.amdgcn.ds.bpermute(i32, i32) #0
; FUNC-LABEL: {{^}}ds_bpermute:
; CHECK: ds_bpermute_b32 v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}}
; CHECK: s_waitcnt lgkmcnt
define void @ds_bpermute(i32 addrspace(1)* %out, i32 %index, i32 %src) nounwind {
%bpermute = call i32 @llvm.amdgcn.ds.bpermute(i32 %index, i32 %src) #0
store i32 %bpermute, i32 addrspace(1)* %out, align 4
ret void
}
; CHECK-LABEL: {{^}}ds_bpermute_imm_offset:
; CHECK: ds_bpermute_b32 v{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}} offset:4
; CHECK: s_waitcnt lgkmcnt
define void @ds_bpermute_imm_offset(i32 addrspace(1)* %out, i32 %base_index, i32 %src) nounwind {
%index = add i32 %base_index, 4
%bpermute = call i32 @llvm.amdgcn.ds.bpermute(i32 %index, i32 %src) #0
store i32 %bpermute, i32 addrspace(1)* %out, align 4
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 }