mirror of
https://github.com/c64scene-ar/llvm-6502.git
synced 2024-12-29 10:32:47 +00:00
1f996fa36b
This is equivalent to the AMDGPUTargetMachine now, but it is the starting point for separating R600 and GCN functionality into separate targets. It is recommened that users start using the gcn triple for GCN-based GPUs, because using the r600 triple for these GPUs will be deprecated in the future. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@225277 91177308-0d34-0410-b5e6-96231b3b80d8
45 lines
1.7 KiB
LLVM
45 lines
1.7 KiB
LLVM
; RUN: llc -march=amdgcn -mcpu=bonaire -verify-machineinstrs -mattr=+load-store-opt -enable-misched < %s | FileCheck -strict-whitespace -check-prefix=SI %s
|
|
|
|
; XFAIL: *
|
|
|
|
@lds = addrspace(3) global [512 x float] undef, align 4
|
|
|
|
; SI-LABEL: {{^}}offset_order:
|
|
|
|
; SI: ds_read_b32 v{{[0-9]+}}, v{{[0-9]+}} offset:56
|
|
; SI: ds_read2st64_b32 v[{{[0-9]+}}:{{[0-9]+}}], v{{[0-9]+}} offset0:0 offset1:4
|
|
; SI: ds_read2_b32 v[{{[0-9]+}}:{{[0-9]+}}], v{{[0-9]+}} offset0:2 offset1:3
|
|
; SI: ds_read2_b32 v[{{[0-9]+}}:{{[0-9]+}}], v{{[0-9]+}} offset0:11 offset1:1
|
|
|
|
define void @offset_order(float addrspace(1)* %out) {
|
|
entry:
|
|
%ptr0 = getelementptr inbounds [512 x float] addrspace(3)* @lds, i32 0, i32 0
|
|
%val0 = load float addrspace(3)* %ptr0
|
|
|
|
%ptr1 = getelementptr inbounds [512 x float] addrspace(3)* @lds, i32 0, i32 256
|
|
%val1 = load float addrspace(3)* %ptr1
|
|
%add1 = fadd float %val0, %val1
|
|
|
|
%ptr2 = getelementptr inbounds [512 x float] addrspace(3)* @lds, i32 0, i32 3
|
|
%val2 = load float addrspace(3)* %ptr2
|
|
%add2 = fadd float %add1, %val2
|
|
|
|
%ptr3 = getelementptr inbounds [512 x float] addrspace(3)* @lds, i32 0, i32 2
|
|
%val3 = load float addrspace(3)* %ptr3
|
|
%add3 = fadd float %add2, %val3
|
|
|
|
%ptr4 = getelementptr inbounds [512 x float] addrspace(3)* @lds, i32 0, i32 12
|
|
%val4 = load float addrspace(3)* %ptr4
|
|
%add4 = fadd float %add3, %val4
|
|
|
|
%ptr5 = getelementptr inbounds [512 x float] addrspace(3)* @lds, i32 0, i32 14
|
|
%val5 = load float addrspace(3)* %ptr5
|
|
%add5 = fadd float %add4, %val5
|
|
|
|
%ptr6 = getelementptr inbounds [512 x float] addrspace(3)* @lds, i32 0, i32 11
|
|
%val6 = load float addrspace(3)* %ptr6
|
|
%add6 = fadd float %add5, %val6
|
|
store float %add6, float addrspace(1)* %out
|
|
ret void
|
|
}
|