reference, declarationdefinition
definition → references, declarations, derived classes, virtual overrides
reference to multiple definitions → definitions
unreferenced
    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
   61
   62
   63
   64
   65
   66
   67
   68
   69
   70
   71
   72
   73
   74
   75
   76
   77
   78
   79
   80
   81
   82
   83
   84
; RUN: llc -march=amdgcn -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s

; GCN-LABEL: {{^}}adjust_writemask_crash_0_nochain:
; GCN: image_get_lod v0, v{{[0-9]+}}, s{{\[[0-9]+:[0-9]+\]}}, s{{\[[0-9]+:[0-9]+\]}} dmask:0x2
; GCN-NOT: v1
; GCN-NOT: v0
; GCN: buffer_store_dword v0
define amdgpu_ps void @adjust_writemask_crash_0_nochain() #0 {
main_body:
  %tmp = call <2 x float> @llvm.amdgcn.image.getlod.1d.v2f32.f32(i32 3, float undef, <8 x i32> undef, <4 x i32> undef, i1 0, i32 0, i32 0)
  %tmp1 = bitcast <2 x float> %tmp to <2 x i32>
  %tmp2 = shufflevector <2 x i32> %tmp1, <2 x i32> undef, <4 x i32> <i32 1, i32 undef, i32 undef, i32 undef>
  %tmp3 = bitcast <4 x i32> %tmp2 to <4 x float>
  %tmp4 = extractelement <4 x float> %tmp3, i32 0
  store volatile float %tmp4, float addrspace(1)* undef
  ret void
}

; GCN-LABEL: {{^}}adjust_writemask_crash_1_nochain:
; GCN: image_get_lod v0, v{{[0-9]+}}, s{{\[[0-9]+:[0-9]+\]}}, s{{\[[0-9]+:[0-9]+\]}} dmask:0x1
; GCN-NOT: v1
; GCN-NOT: v0
; GCN: buffer_store_dword v0
define amdgpu_ps void @adjust_writemask_crash_1_nochain() #0 {
main_body:
  %tmp = call <2 x float> @llvm.amdgcn.image.getlod.1d.v2f32.f32(i32 3, float undef, <8 x i32> undef, <4 x i32> undef, i1 0, i32 0, i32 0)
  %tmp1 = bitcast <2 x float> %tmp to <2 x i32>
  %tmp2 = shufflevector <2 x i32> %tmp1, <2 x i32> undef, <4 x i32> <i32 1, i32 0, i32 undef, i32 undef>
  %tmp3 = bitcast <4 x i32> %tmp2 to <4 x float>
  %tmp4 = extractelement <4 x float> %tmp3, i32 1
  store volatile float %tmp4, float addrspace(1)* undef
  ret void
}

; GCN-LABEL: {{^}}adjust_writemask_crash_0_chain:
; GCN: image_sample v0, v{{[0-9]+}}, s{{\[[0-9]+:[0-9]+\]}}, s{{\[[0-9]+:[0-9]+\]}} dmask:0x2
; GCN-NOT: v1
; GCN-NOT: v0
; GCN: buffer_store_dword v0
define amdgpu_ps void @adjust_writemask_crash_0_chain() #0 {
main_body:
  %tmp = call <2 x float> @llvm.amdgcn.image.sample.1d.v2f32.f32(i32 3, float undef, <8 x i32> undef, <4 x i32> undef, i1 0, i32 0, i32 0)
  %tmp1 = bitcast <2 x float> %tmp to <2 x i32>
  %tmp2 = shufflevector <2 x i32> %tmp1, <2 x i32> undef, <4 x i32> <i32 1, i32 undef, i32 undef, i32 undef>
  %tmp3 = bitcast <4 x i32> %tmp2 to <4 x float>
  %tmp4 = extractelement <4 x float> %tmp3, i32 0
  store volatile float %tmp4, float addrspace(1)* undef
  ret void
}

; GCN-LABEL: {{^}}adjust_writemask_crash_1_chain:
; GCN: image_sample v0, v{{[0-9]+}}, s{{\[[0-9]+:[0-9]+\]}}, s{{\[[0-9]+:[0-9]+\]}} dmask:0x1
; GCN-NOT: v1
; GCN-NOT: v0
; GCN: buffer_store_dword v0
define amdgpu_ps void @adjust_writemask_crash_1_chain() #0 {
main_body:
  %tmp = call <2 x float> @llvm.amdgcn.image.sample.1d.v2f32.f32(i32 3, float undef, <8 x i32> undef, <4 x i32> undef, i1 0, i32 0, i32 0)
  %tmp1 = bitcast <2 x float> %tmp to <2 x i32>
  %tmp2 = shufflevector <2 x i32> %tmp1, <2 x i32> undef, <4 x i32> <i32 1, i32 0, i32 undef, i32 undef>
  %tmp3 = bitcast <4 x i32> %tmp2 to <4 x float>
  %tmp4 = extractelement <4 x float> %tmp3, i32 1
  store volatile float %tmp4, float addrspace(1)* undef
  ret void
}

define amdgpu_ps void @adjust_writemask_crash_0_v4() #0 {
main_body:
  %tmp = call <4 x float> @llvm.amdgcn.image.getlod.1d.v4f32.f32(i32 5, float undef, <8 x i32> undef, <4 x i32> undef, i1 0, i32 0, i32 0)
  %tmp1 = bitcast <4 x float> %tmp to <4 x i32>
  %tmp2 = shufflevector <4 x i32> %tmp1, <4 x i32> undef, <4 x i32> <i32 1, i32 undef, i32 undef, i32 undef>
  %tmp3 = bitcast <4 x i32> %tmp2 to <4 x float>
  %tmp4 = extractelement <4 x float> %tmp3, i32 0
  store volatile float %tmp4, float addrspace(1)* undef
  ret void
}


declare <2 x float> @llvm.amdgcn.image.sample.1d.v2f32.f32(i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
declare <2 x float> @llvm.amdgcn.image.getlod.1d.v2f32.f32(i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1
declare <4 x float> @llvm.amdgcn.image.getlod.1d.v4f32.f32(i32, float, <8 x i32>, <4 x i32>, i1, i32, i32) #1

attributes #0 = { nounwind }
attributes #1 = { nounwind readonly }