1111; allow scheduling of other instructions which reduce RP
1212
1313; CHECK-LABEL: {{^}}return_72xi32:
14- ; GFX11-PAL: codeLenInByte = 768
15- ; GFX11-PAL-GCNTRACKERS: codeLenInByte = 888
1614; GFX11-PAL: NumSgprs: 33
1715; GFX11-PAL-GCNTRACKERS: NumSgprs: 33
1816; GFX11-PAL: NumVgprs: 64
2220
2321
2422; CHECK-LABEL: {{^}}call_72xi32:
25- ; GFX11-PAL: codeLenInByte = 1300
26- ; GFX11-PAL-GCNTRACKERS: codeLenInByte = 1372
2723; GFX11-PAL: NumSgprs: 35
2824; GFX11-PAL-GCNTRACKERS: NumSgprs: 35
2925; GFX11-PAL: NumVgprs: 64
4642}
4743
4844; CHECK-LABEL: {{^}}global_extload_v16f16_to_v16f64:
49- ; TONGA: codeLenInByte = 420
50- ; TONGA-GCNTRACKERS: codeLenInByte = 436
5145; TONGA: NumSgprs: 96
5246; TONGA-GCNTRACKERS: NumSgprs: 96
5347; TONGA: NumVgprs: 33
@@ -64,8 +58,6 @@ define amdgpu_kernel void @global_extload_v16f16_to_v16f64(ptr addrspace(1) %out
6458}
6559
6660; CHECK-LABEL: {{^}}constant_zextload_v64i16_to_v64i32:
67- ; GENERIC: codeLenInByte = 860
68- ; GENERIC-GCNTRACKERS: codeLenInByte = 860
6961; GENERIC: NumSgprs: 71
7062; GENERIC-GCNTRACKERS: NumSgprs: 54
7163; GENERIC: NumVgprs: 16
@@ -81,8 +73,6 @@ define amdgpu_kernel void @constant_zextload_v64i16_to_v64i32(ptr addrspace(1) %
8173}
8274
8375; CHECK-LABEL: {{^}}excess_soft_clause_reg_pressure:
84- ; GFX908: codeLenInByte = 1436
85- ; GFX908-GCNTRACKERS: codeLenInByte = 1436
8676; GFX908: NumSgprs: 56
8777; GFX908-GCNTRACKERS: NumSgprs: 56
8878; GFX908: NumVgprs: 43
@@ -644,4 +634,4 @@ declare align 4 ptr addrspace(4) @llvm.amdgcn.implicitarg.ptr() #3
644634attributes #0 = { nounwind "amdgpu-waves-per-eu" ="1,1" "amdgpu-flat-work-group-size" ="1,1" }
645635attributes #1 = { nounwind "amdgpu-num-vgpr" ="64" }
646636attributes #2 = { nofree nosync nounwind readnone speculatable willreturn }
647- attributes #3 = { nounwind readnone speculatable willreturn }
637+ attributes #3 = { nounwind readnone speculatable willreturn }
0 commit comments