11
11
; allow scheduling of other instructions which reduce RP
12
12
13
13
; CHECK-LABEL: {{^}}return_72xi32:
14
- ; GFX11-PAL: codeLenInByte = 768
15
- ; GFX11-PAL-GCNTRACKERS: codeLenInByte = 888
16
14
; GFX11-PAL: NumSgprs: 33
17
15
; GFX11-PAL-GCNTRACKERS: NumSgprs: 33
18
16
; GFX11-PAL: NumVgprs: 64
22
20
23
21
24
22
; CHECK-LABEL: {{^}}call_72xi32:
25
- ; GFX11-PAL: codeLenInByte = 1300
26
- ; GFX11-PAL-GCNTRACKERS: codeLenInByte = 1372
27
23
; GFX11-PAL: NumSgprs: 35
28
24
; GFX11-PAL-GCNTRACKERS: NumSgprs: 35
29
25
; GFX11-PAL: NumVgprs: 64
46
42
}
47
43
48
44
; CHECK-LABEL: {{^}}global_extload_v16f16_to_v16f64:
49
- ; TONGA: codeLenInByte = 420
50
- ; TONGA-GCNTRACKERS: codeLenInByte = 436
51
45
; TONGA: NumSgprs: 96
52
46
; TONGA-GCNTRACKERS: NumSgprs: 96
53
47
; TONGA: NumVgprs: 33
@@ -64,8 +58,6 @@ define amdgpu_kernel void @global_extload_v16f16_to_v16f64(ptr addrspace(1) %out
64
58
}
65
59
66
60
; CHECK-LABEL: {{^}}constant_zextload_v64i16_to_v64i32:
67
- ; GENERIC: codeLenInByte = 860
68
- ; GENERIC-GCNTRACKERS: codeLenInByte = 860
69
61
; GENERIC: NumSgprs: 71
70
62
; GENERIC-GCNTRACKERS: NumSgprs: 54
71
63
; GENERIC: NumVgprs: 16
@@ -81,8 +73,6 @@ define amdgpu_kernel void @constant_zextload_v64i16_to_v64i32(ptr addrspace(1) %
81
73
}
82
74
83
75
; CHECK-LABEL: {{^}}excess_soft_clause_reg_pressure:
84
- ; GFX908: codeLenInByte = 1436
85
- ; GFX908-GCNTRACKERS: codeLenInByte = 1436
86
76
; GFX908: NumSgprs: 56
87
77
; GFX908-GCNTRACKERS: NumSgprs: 56
88
78
; GFX908: NumVgprs: 43
@@ -644,4 +634,4 @@ declare align 4 ptr addrspace(4) @llvm.amdgcn.implicitarg.ptr() #3
644
634
attributes #0 = { nounwind "amdgpu-waves-per-eu" ="1,1" "amdgpu-flat-work-group-size" ="1,1" }
645
635
attributes #1 = { nounwind "amdgpu-num-vgpr" ="64" }
646
636
attributes #2 = { nofree nosync nounwind readnone speculatable willreturn }
647
- attributes #3 = { nounwind readnone speculatable willreturn }
637
+ attributes #3 = { nounwind readnone speculatable willreturn }
0 commit comments