Changeset View
Changeset View
Standalone View
Standalone View
llvm/trunk/test/CodeGen/AMDGPU/lds-relocs.ll
1 | ; RUN: llc -mtriple=amdgcn-mesa-mesa3d -mcpu=gfx900 -verify-machineinstrs -show-mc-encoding < %s | FileCheck -check-prefixes=GCN %s | ||||
---|---|---|---|---|---|
2 | ; RUN: llc -mtriple=amdgcn-mesa-mesa3d -mcpu=gfx900 -filetype=obj < %s | llvm-readobj -r -t | FileCheck -check-prefixes=ELF %s | ||||
3 | | ||||
4 | @lds.external = external unnamed_addr addrspace(3) global [0 x i32] | ||||
5 | @lds.defined = unnamed_addr addrspace(3) global [8 x i32] undef, align 8 | ||||
6 | | ||||
7 | ; ELF: Relocations [ | ||||
8 | ; ELF-NEXT: Section (3) .rel.text { | ||||
9 | ; ELF-NEXT: 0x{{[0-9a-f]*}} R_AMDGPU_ABS32 lds.external 0x0 | ||||
10 | ; ELF-NEXT: 0x{{[0-9a-f]*}} R_AMDGPU_ABS32 lds.defined 0x0 | ||||
11 | ; ELF-NEXT: } | ||||
12 | ; ELF-NEXT: ] | ||||
13 | | ||||
14 | ; ELF: Symbol { | ||||
15 | ; ELF: Name: lds.defined | ||||
16 | ; ELF-NEXT: Value: 0x8 | ||||
17 | ; ELF-NEXT: Size: 32 | ||||
18 | ; ELF-NEXT: Binding: Global (0x1) | ||||
19 | ; ELF-NEXT: Type: Object (0x1) | ||||
20 | ; ELF-NEXT: Other: 0 | ||||
21 | ; ELF-NEXT: Section: Processor Specific (0xFF00) | ||||
22 | ; ELF-NEXT: } | ||||
23 | | ||||
24 | ; ELF: Symbol { | ||||
25 | ; ELF: Name: lds.external | ||||
26 | ; ELF-NEXT: Value: 0x4 | ||||
27 | ; ELF-NEXT: Size: 0 | ||||
28 | ; ELF-NEXT: Binding: Global (0x1) | ||||
29 | ; ELF-NEXT: Type: Object (0x1) | ||||
30 | ; ELF-NEXT: Other: 0 | ||||
31 | ; ELF-NEXT: Section: Processor Specific (0xFF00) | ||||
32 | ; ELF-NEXT: } | ||||
33 | | ||||
34 | ; GCN-LABEL: {{^}}test_basic: | ||||
35 | ; GCN: v_mov_b32_e32 v1, lds.external@abs32@lo ; encoding: [0xff,0x02,0x02,0x7e,A,A,A,A] | ||||
36 | ; GCN-NEXT: ; fixup A - offset: 4, value: lds.external@abs32@lo, kind: FK_Data_4{{$}} | ||||
37 | ; | ||||
38 | ; GCN: s_add_i32 s0, lds.defined@abs32@lo, s0 ; encoding: [0xff,0x00,0x00,0x81,A,A,A,A] | ||||
39 | ; GCN-NEXT: ; fixup A - offset: 4, value: lds.defined@abs32@lo, kind: FK_Data_4{{$}} | ||||
40 | ; | ||||
41 | ; GCN: .globl lds.external | ||||
42 | ; GCN: .amdgpu_lds lds.external, 0, 4 | ||||
43 | ; GCN: .globl lds.defined | ||||
44 | ; GCN: .amdgpu_lds lds.defined, 32, 8 | ||||
45 | define amdgpu_gs float @test_basic(i32 inreg %wave, i32 %arg1) #0 { | ||||
46 | main_body: | ||||
47 | %gep0 = getelementptr [0 x i32], [0 x i32] addrspace(3)* @lds.external, i32 0, i32 %arg1 | ||||
48 | %tmp = load i32, i32 addrspace(3)* %gep0 | ||||
49 | | ||||
50 | %mask = call i64 @llvm.amdgcn.icmp.i64.i32(i32 %tmp, i32 0, i32 0) | ||||
51 | %mask.32 = trunc i64 %mask to i32 | ||||
52 | %gep1 = getelementptr [8 x i32], [8 x i32] addrspace(3)* @lds.defined, i32 0, i32 %wave | ||||
53 | store i32 %mask.32, i32 addrspace(3)* %gep1 | ||||
54 | | ||||
55 | %r = bitcast i32 %tmp to float | ||||
56 | ret float %r | ||||
57 | } | ||||
58 | | ||||
59 | ; Function Attrs: convergent nounwind readnone | ||||
60 | declare i64 @llvm.amdgcn.icmp.i64.i32(i32, i32, i32) #4 | ||||
61 | | ||||
62 | attributes #0 = { "no-signed-zeros-fp-math"="true" } | ||||
63 | attributes #4 = { convergent nounwind readnone } |