v_cvt_pk_u8_f32.ll
2.78 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
; RUN: llc -march=amdgcn -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
; RUN: llc -march=amdgcn -mcpu=fiji -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
declare i32 @llvm.amdgcn.cvt.pk.u8.f32(float, i32, i32) #0
; GCN-LABEL: {{^}}v_cvt_pk_u8_f32_idx_0:
; GCN: v_cvt_pk_u8_f32 v{{[0-9]+}}, s{{[0-9]+}}, 0, v{{[0-9]+}}
define amdgpu_kernel void @v_cvt_pk_u8_f32_idx_0(i32 addrspace(1)* %out, float %src, i32 %reg) {
%result = call i32 @llvm.amdgcn.cvt.pk.u8.f32(float %src, i32 0, i32 %reg) #0
store i32 %result, i32 addrspace(1)* %out, align 4
ret void
}
; GCN-LABEL: {{^}}v_cvt_pk_u8_f32_idx_1:
; GCN: v_cvt_pk_u8_f32 v{{[0-9]+}}, s{{[0-9]+}}, 1, v{{[0-9]+}}
define amdgpu_kernel void @v_cvt_pk_u8_f32_idx_1(i32 addrspace(1)* %out, float %src, i32 %reg) {
%result = call i32 @llvm.amdgcn.cvt.pk.u8.f32(float %src, i32 1, i32 %reg) #0
store i32 %result, i32 addrspace(1)* %out, align 4
ret void
}
; GCN-LABEL: {{^}}v_cvt_pk_u8_f32_idx_2:
; GCN: v_cvt_pk_u8_f32 v{{[0-9]+}}, s{{[0-9]+}}, 2, v{{[0-9]+}}
define amdgpu_kernel void @v_cvt_pk_u8_f32_idx_2(i32 addrspace(1)* %out, float %src, i32 %reg) {
%result = call i32 @llvm.amdgcn.cvt.pk.u8.f32(float %src, i32 2, i32 %reg) #0
store i32 %result, i32 addrspace(1)* %out, align 4
ret void
}
; GCN-LABEL: {{^}}v_cvt_pk_u8_f32_idx_3:
; GCN: v_cvt_pk_u8_f32 v{{[0-9]+}}, s{{[0-9]+}}, 3, v{{[0-9]+}}
define amdgpu_kernel void @v_cvt_pk_u8_f32_idx_3(i32 addrspace(1)* %out, float %src, i32 %reg) {
%result = call i32 @llvm.amdgcn.cvt.pk.u8.f32(float %src, i32 3, i32 %reg) #0
store i32 %result, i32 addrspace(1)* %out, align 4
ret void
}
; GCN-LABEL: {{^}}v_cvt_pk_u8_f32_combine:
; GCN: v_cvt_pk_u8_f32 v{{[0-9]+}}, s{{[0-9]+}}, 0, v{{[0-9]+}}
; GCN: v_cvt_pk_u8_f32 v{{[0-9]+}}, s{{[0-9]+}}, 1, v{{[0-9]+}}
; GCN: v_cvt_pk_u8_f32 v{{[0-9]+}}, s{{[0-9]+}}, 2, v{{[0-9]+}}
; GCN: v_cvt_pk_u8_f32 v{{[0-9]+}}, s{{[0-9]+}}, 3, v{{[0-9]+}}
define amdgpu_kernel void @v_cvt_pk_u8_f32_combine(i32 addrspace(1)* %out, float %src, i32 %reg) {
%result0 = call i32 @llvm.amdgcn.cvt.pk.u8.f32(float %src, i32 0, i32 %reg) #0
%result1 = call i32 @llvm.amdgcn.cvt.pk.u8.f32(float %src, i32 1, i32 %result0) #0
%result2 = call i32 @llvm.amdgcn.cvt.pk.u8.f32(float %src, i32 2, i32 %result1) #0
%result3 = call i32 @llvm.amdgcn.cvt.pk.u8.f32(float %src, i32 3, i32 %result2) #0
store i32 %result3, i32 addrspace(1)* %out, align 4
ret void
}
; GCN-LABEL: {{^}}v_cvt_pk_u8_f32_idx:
; GCN: v_cvt_pk_u8_f32 v{{[0-9]+}}, s{{[0-9]+}}, v{{[0-9]+}}, v{{[0-9]+}}
define amdgpu_kernel void @v_cvt_pk_u8_f32_idx(i32 addrspace(1)* %out, float %src, i32 %idx, i32 %reg) {
%result = call i32 @llvm.amdgcn.cvt.pk.u8.f32(float %src, i32 %idx, i32 %reg) #0
store i32 %result, i32 addrspace(1)* %out, align 4
ret void
}
attributes #0 = { nounwind readnone }