AMDGPU: SimplifyDemandedElts for image intrinsics

[arsenm]

Summary

Causes some VGPR usage improvements in shaderdb, but
introduces some SGPR spilling regressions due to random
scheduling changes later.

https://ghostbin.com/paste/ghrtj

[mareko]

lib/Transforms/InstCombine/InstCombineSimplifyDemanded.cpp

1734

Gather4 opcodes always return 4 VGPRs and DMASK has a different meaning. Specifically, Gather4 reads 4 texels from memory and DMASK selects which color component is returned for the texels (i.e. 4x red channel, or 4x green channel, etc.) So DMASK shouldn't be changed by the compiler for gather4 opcodes.

[arsenm]

Don't handle gather4 (is getlod OK?)

[mareko]

GETLOD should be OK.

LGTM.

[arsenm]

r300453

lib/Transforms/InstCombine/InstCombineSimplifyDemanded.cpp

case Instruction::Call: {
     // in an undefined state.
     case Intrinsic::x86_sse4a_extrq:
     case Intrinsic::x86_sse4a_extrqi:
     case Intrinsic::x86_sse4a_insertq:
     case Intrinsic::x86_sse4a_insertqi:
       UndefElts.setHighBits(VWidth / 2);
       break;
     case Intrinsic::amdgcn_buffer_load:
-    case Intrinsic::amdgcn_buffer_load_format: {
+    case Intrinsic::amdgcn_buffer_load_format:
+    case Intrinsic::amdgcn_image_sample:
+    case Intrinsic::amdgcn_image_sample_cl:
+    case Intrinsic::amdgcn_image_sample_d:
+    case Intrinsic::amdgcn_image_sample_d_cl:
+    case Intrinsic::amdgcn_image_sample_l:
+    case Intrinsic::amdgcn_image_sample_b:
+    case Intrinsic::amdgcn_image_sample_b_cl:
+    case Intrinsic::amdgcn_image_sample_lz:
+    case Intrinsic::amdgcn_image_sample_cd:
+    case Intrinsic::amdgcn_image_sample_cd_cl:
+
+    case Intrinsic::amdgcn_image_sample_c:
+    case Intrinsic::amdgcn_image_sample_c_cl:
+    case Intrinsic::amdgcn_image_sample_c_d:
+    case Intrinsic::amdgcn_image_sample_c_d_cl:
+    case Intrinsic::amdgcn_image_sample_c_l:
+    case Intrinsic::amdgcn_image_sample_c_b:
+    case Intrinsic::amdgcn_image_sample_c_b_cl:
+    case Intrinsic::amdgcn_image_sample_c_lz:
+    case Intrinsic::amdgcn_image_sample_c_cd:
+    case Intrinsic::amdgcn_image_sample_c_cd_cl:
+
+    case Intrinsic::amdgcn_image_sample_o:
+    case Intrinsic::amdgcn_image_sample_cl_o:
+    case Intrinsic::amdgcn_image_sample_d_o:
+    case Intrinsic::amdgcn_image_sample_d_cl_o:
+    case Intrinsic::amdgcn_image_sample_l_o:
+    case Intrinsic::amdgcn_image_sample_b_o:
+    case Intrinsic::amdgcn_image_sample_b_cl_o:
+    case Intrinsic::amdgcn_image_sample_lz_o:
+    case Intrinsic::amdgcn_image_sample_cd_o:
+    case Intrinsic::amdgcn_image_sample_cd_cl_o:
+
+    case Intrinsic::amdgcn_image_sample_c_o:
+    case Intrinsic::amdgcn_image_sample_c_cl_o:
+    case Intrinsic::amdgcn_image_sample_c_d_o:
+    case Intrinsic::amdgcn_image_sample_c_d_cl_o:
+    case Intrinsic::amdgcn_image_sample_c_l_o:
+    case Intrinsic::amdgcn_image_sample_c_b_o:
+    case Intrinsic::amdgcn_image_sample_c_b_cl_o:
+    case Intrinsic::amdgcn_image_sample_c_lz_o:
+    case Intrinsic::amdgcn_image_sample_c_cd_o:
+    case Intrinsic::amdgcn_image_sample_c_cd_cl_o:
+
+    case Intrinsic::amdgcn_image_getlod: {
       if (VWidth == 1 || !DemandedElts.isMask())
         return nullptr;
 
       // TODO: Handle 3 vectors when supported in code gen.
       unsigned NewNumElts = PowerOf2Ceil(DemandedElts.countTrailingOnes());
       if (NewNumElts == VWidth)
         return nullptr;
 
       Module *M = II->getParent()->getParent()->getParent();
       Type *EltTy = V->getType()->getVectorElementType();
 
       Type *NewTy = (NewNumElts == 1) ? EltTy :
         VectorType::get(EltTy, NewNumElts);
 
-      Function *NewIntrin = Intrinsic::getDeclaration(M, II->getIntrinsicID(),
-                                                      NewTy);
+      auto IID = II->getIntrinsicID();
+
+      bool IsBuffer = IID == Intrinsic::amdgcn_buffer_load ||
+                      IID == Intrinsic::amdgcn_buffer_load_format;
+
+      Function *NewIntrin = IsBuffer ?
+        Intrinsic::getDeclaration(M, IID, NewTy) :
+        // Samplers have 3 mangled types.
+        Intrinsic::getDeclaration(M, IID,
+                                  { NewTy, II->getArgOperand(0)->getType(),
+                                      II->getArgOperand(1)->getType()});
 
       SmallVector<Value *, 5> Args;
       for (unsigned I = 0, E = II->getNumArgOperands(); I != E; ++I)
         Args.push_back(II->getArgOperand(I));
 
       IRBuilderBase::InsertPointGuard Guard(*Builder);
       Builder->SetInsertPoint(II);
 
       CallInst *NewCall = Builder->CreateCall(NewIntrin, Args);
       NewCall->takeName(II);
       NewCall->copyMetadata(*II);
+
+      if (!IsBuffer) {
+        ConstantInt *DMask = dyn_cast<ConstantInt>(NewCall->getArgOperand(3));
+        if (DMask) {
+          unsigned DMaskVal = DMask->getZExtValue() & 0xf;
+
+          unsigned PopCnt = 0;
+          unsigned NewDMask = 0;
+          for (unsigned I = 0; I < 4; ++I) {
+            const unsigned Bit = 1 << I;
+            if (!!(DMaskVal & Bit)) {
+              if (++PopCnt > NewNumElts)
+                break;
+
+              NewDMask |= Bit;
+            }
+          }
+
+          NewCall->setArgOperand(3, ConstantInt::get(DMask->getType(), NewDMask));
+        }
+      }
+
+
       if (NewNumElts == 1) {
         return Builder->CreateInsertElement(UndefValue::get(V->getType()),
                                             NewCall, static_cast<uint64_t>(0));
       }
 
       SmallVector<uint32_t, 8> EltMask;
       for (unsigned I = 0; I < VWidth; ++I)
         EltMask.push_back(I);

test/Transforms/InstCombine/amdgcn-demanded-vector-elts.ll

 ; CHECK-NEXT: %data = call float @llvm.amdgcn.buffer.load.f32(<4 x i32> %rsrc, i32 %idx, i32 %ofs, i1 false, i1 false), !fpmath !0
 ; CHECK-NEXT: ret float %data
 define amdgpu_ps float @preserve_metadata_extract_elt0_buffer_load_v2f32(<4 x i32> inreg %rsrc, i32 %idx, i32 %ofs) #0 {
   %data = call <2 x float> @llvm.amdgcn.buffer.load.v2f32(<4 x i32> %rsrc, i32 %idx, i32 %ofs, i1 false, i1 false), !fpmath !0
   %elt0 = extractelement <2 x float> %data, i32 0
   ret float %elt0
 }
 
+declare float @llvm.amdgcn.buffer.load.f32(<4 x i32>, i32, i32, i1, i1) #1
+declare <1 x float> @llvm.amdgcn.buffer.load.v1f32(<4 x i32>, i32, i32, i1, i1) #1
+declare <2 x float> @llvm.amdgcn.buffer.load.v2f32(<4 x i32>, i32, i32, i1, i1) #1
+declare <3 x float> @llvm.amdgcn.buffer.load.v3f32(<4 x i32>, i32, i32, i1, i1) #1
+declare <4 x float> @llvm.amdgcn.buffer.load.v4f32(<4 x i32>, i32, i32, i1, i1) #1
+
 ; --------------------------------------------------------------------
 ; llvm.amdgcn.buffer.load.format
 ; --------------------------------------------------------------------
 
 ; CHECK-LABEL: @buffer_load_format_v1f32(
 ; CHECK-NEXT: %data = call <1 x float> @llvm.amdgcn.buffer.load.format.v1f32(<4 x i32> %rsrc, i32 %idx, i32 %ofs, i1 false, i1 true)
 ; CHECK-NEXT: ret <1 x float> %data
 define amdgpu_ps <1 x float> @buffer_load_format_v1f32(<4 x i32> inreg %rsrc, i32 %idx, i32 %ofs) #0 {

 ; CHECK-NEXT: ret i16 %tmp2
 define i16 @extract_lo16_0_bitcast_buffer_load_format_v4f32(i32 %arg) #0 {
   %tmp = call <4 x float> @llvm.amdgcn.buffer.load.format.v4f32(<4 x i32> undef, i32 %arg, i32 16, i1 false, i1 false) #3
   %tmp1 = bitcast <4 x float> %tmp to <8 x i16>
   %tmp2 = extractelement <8 x i16> %tmp1, i32 0
   ret i16 %tmp2
 }
 
-declare float @llvm.amdgcn.buffer.load.f32(<4 x i32>, i32, i32, i1, i1) #1
-declare <1 x float> @llvm.amdgcn.buffer.load.v1f32(<4 x i32>, i32, i32, i1, i1) #1
-declare <2 x float> @llvm.amdgcn.buffer.load.v2f32(<4 x i32>, i32, i32, i1, i1) #1
-declare <3 x float> @llvm.amdgcn.buffer.load.v3f32(<4 x i32>, i32, i32, i1, i1) #1
-declare <4 x float> @llvm.amdgcn.buffer.load.v4f32(<4 x i32>, i32, i32, i1, i1) #1
-
 declare float @llvm.amdgcn.buffer.load.format.f32(<4 x i32>, i32, i32, i1, i1) #1
 declare <1 x float> @llvm.amdgcn.buffer.load.format.v1f32(<4 x i32>, i32, i32, i1, i1) #1
 declare <2 x float> @llvm.amdgcn.buffer.load.format.v2f32(<4 x i32>, i32, i32, i1, i1) #1
 declare <3 x float> @llvm.amdgcn.buffer.load.format.v3f32(<4 x i32>, i32, i32, i1, i1) #1
 declare <4 x float> @llvm.amdgcn.buffer.load.format.v4f32(<4 x i32>, i32, i32, i1, i1) #1
 
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+; CHECK-LABEL: @extract_elt0_image_sample_v4f32_v4f32_v4i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.f32.v4f32.v4i32(<4 x float> %vaddr, <4 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_v4f32_v4f32_v4i32(<4 x float> %vaddr, <4 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.v4f32.v4f32.v4i32(<4 x float> %vaddr, <4 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+; CHECK-LABEL: @extract_elt0_image_sample_v4f32_v2f32_v4i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.f32.v2f32.v8i32(<2 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_v4f32_v2f32_v4i32(<2 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.v4f32.v2f32.v8i32(<2 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+; CHECK-LABEL: @extract_elt0_invalid_dmask_image_sample_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 %dmask, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_invalid_dmask_image_sample_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc, i32 %dmask) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 %dmask, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+; FIXME: Should really fold to undef
+; CHECK-LABEL: @extract_elt0_dmask_0000_image_sample_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 0, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_dmask_0000_image_sample_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 0, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+; CHECK-LABEL: @extract_elt0_dmask_0001_image_sample_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_dmask_0001_image_sample_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+; FIXME: Should really fold to undef
+; CHECK-LABEL: @extract_elt0_dmask_0010_image_sample_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 2, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_dmask_0010_image_sample_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 2, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+; FIXME: Should really fold to undef
+; CHECK-LABEL: @extract_elt0_dmask_0100_image_sample_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 4, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_dmask_0100_image_sample_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 4, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+; FIXME: Should really fold to undef
+; CHECK-LABEL: @extract_elt0_dmask_1000_image_sample_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 8, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_dmask_1000_image_sample_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 8, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+; CHECK-LABEL: @extract_elt0_dmask_1001_image_sample_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_dmask_1001_image_sample_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 9, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+; CHECK-LABEL: @extract_elt0_dmask_0011_image_sample_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_dmask_0011_image_sample_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 3, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+; CHECK-LABEL: @extract_elt0_dmask_0111_image_sample_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_dmask_0111_image_sample_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 7, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+; CHECK-LABEL: @extract_elt0_elt1_dmask_0001_image_sample_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call <2 x float> @llvm.amdgcn.image.sample.v2f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret <2 x float> %data
+define amdgpu_ps <2 x float> @extract_elt0_elt1_dmask_0001_image_sample_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %shuf = shufflevector <4 x float> %data, <4 x float> undef, <2 x i32> <i32 0, i32 1>
+  ret <2 x float> %shuf
+}
+
+; CHECK-LABEL: @extract_elt0_elt1_dmask_0011_image_sample_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call <2 x float> @llvm.amdgcn.image.sample.v2f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 3, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret <2 x float> %data
+define amdgpu_ps <2 x float> @extract_elt0_elt1_dmask_0011_image_sample_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 3, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %shuf = shufflevector <4 x float> %data, <4 x float> undef, <2 x i32> <i32 0, i32 1>
+  ret <2 x float> %shuf
+}
+
+; CHECK-LABEL: @extract_elt0_elt1_dmask_0111_image_sample_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call <2 x float> @llvm.amdgcn.image.sample.v2f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 3, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret <2 x float> %data
+define amdgpu_ps <2 x float> @extract_elt0_elt1_dmask_0111_image_sample_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 7, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %shuf = shufflevector <4 x float> %data, <4 x float> undef, <2 x i32> <i32 0, i32 1>
+  ret <2 x float> %shuf
+}
+
+; CHECK-LABEL: @extract_elt0_elt1_dmask_0101_image_sample_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call <2 x float> @llvm.amdgcn.image.sample.v2f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 5, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret <2 x float> %data
+define amdgpu_ps <2 x float> @extract_elt0_elt1_dmask_0101_image_sample_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 5, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %shuf = shufflevector <4 x float> %data, <4 x float> undef, <2 x i32> <i32 0, i32 1>
+  ret <2 x float> %shuf
+}
+
+; CHECK-LABEL: @extract_elt0_elt1_elt2_dmask_0001_image_sample_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call <4 x float> @llvm.amdgcn.image.sample.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: %shuf = shufflevector <4 x float> %data, <4 x float> undef, <3 x i32> <i32 0, i32 1, i32 2>
+; CHECK-NEXT: ret <3 x float> %shuf
+define amdgpu_ps <3 x float> @extract_elt0_elt1_elt2_dmask_0001_image_sample_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %shuf = shufflevector <4 x float> %data, <4 x float> undef, <3 x i32> <i32 0, i32 1, i32 2>
+  ret <3 x float> %shuf
+}
+
+; CHECK-LABEL: @extract_elt0_elt1_elt2_dmask_0011_image_sample_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call <4 x float> @llvm.amdgcn.image.sample.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 3, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: %shuf = shufflevector <4 x float> %data, <4 x float> undef, <3 x i32> <i32 0, i32 1, i32 2>
+; CHECK-NEXT: ret <3 x float> %shuf
+define amdgpu_ps <3 x float> @extract_elt0_elt1_elt2_dmask_0011_image_sample_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 3, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %shuf = shufflevector <4 x float> %data, <4 x float> undef, <3 x i32> <i32 0, i32 1, i32 2>
+  ret <3 x float> %shuf
+}
+
+; CHECK-LABEL: @extract_elt0_elt1_elt2_dmask_0101_image_sample_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call <4 x float> @llvm.amdgcn.image.sample.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 5, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: %shuf = shufflevector <4 x float> %data, <4 x float> undef, <3 x i32> <i32 0, i32 1, i32 2>
+; CHECK-NEXT: ret <3 x float> %shuf
+define amdgpu_ps <3 x float> @extract_elt0_elt1_elt2_dmask_0101_image_sample_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 5, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %shuf = shufflevector <4 x float> %data, <4 x float> undef, <3 x i32> <i32 0, i32 1, i32 2>
+  ret <3 x float> %shuf
+}
+
+; CHECK-LABEL: @extract_elt0_elt1_elt2_dmask_0111_image_sample_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call <4 x float> @llvm.amdgcn.image.sample.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 7, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: %shuf = shufflevector <4 x float> %data, <4 x float> undef, <3 x i32> <i32 0, i32 1, i32 2>
+; CHECK-NEXT: ret <3 x float> %shuf
+define amdgpu_ps <3 x float> @extract_elt0_elt1_elt2_dmask_0111_image_sample_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 7, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %shuf = shufflevector <4 x float> %data, <4 x float> undef, <3 x i32> <i32 0, i32 1, i32 2>
+  ret <3 x float> %shuf
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+declare <4 x float> @llvm.amdgcn.image.sample.v4f32.v4f32.v4i32(<4 x float>, <4 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+declare <4 x float> @llvm.amdgcn.image.sample.v4f32.v2f32.v8i32(<2 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.cl
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_cl_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.cl.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_cl_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.cl.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.cl.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.d
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_d_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.d.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_d_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.d.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.d.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.d.cl
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_d_cl_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.d.cl.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_d_cl_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.d.cl.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.d.cl.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.l
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_l_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.l.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_l_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.l.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.l.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.b
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_b_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.b.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_b_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.b.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.b.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.b.cl
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_b_cl_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.b.cl.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_b_cl_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.b.cl.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.b.cl.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.lz
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_lz_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.lz.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_lz_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.lz.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.lz.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.cd
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_cd_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.cd.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_cd_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.cd.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.cd.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.cd.cl
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_cd_cl_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.cd.cl.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_cd_cl_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.cd.cl.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.cd.cl.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.c
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_c_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.c.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_c_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.c.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+; CHECK-LABEL: @extract_elt0_image_sample_c_v4f32_v4f32_v4i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.c.f32.v4f32.v4i32(<4 x float> %vaddr, <4 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_c_v4f32_v4f32_v4i32(<4 x float> %vaddr, <4 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.c.v4f32.v4f32.v4i32(<4 x float> %vaddr, <4 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+; CHECK-LABEL: @extract_elt0_image_sample_c_v4f32_v2f32_v4i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.c.f32.v2f32.v8i32(<2 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_c_v4f32_v2f32_v4i32(<2 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.c.v4f32.v2f32.v8i32(<2 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.c.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.v4f32.v4f32.v4i32(<4 x float>, <4 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.v4f32.v2f32.v8i32(<2 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.c.cl
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_c_cl_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.c.cl.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_c_cl_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.c.cl.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.c.cl.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.c.d
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_c_d_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.c.d.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_c_d_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.c.d.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.c.d.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.c.d.cl
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_c_d_cl_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.c.d.cl.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_c_d_cl_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.c.d.cl.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.c.l
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_c_l_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.c.l.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_c_l_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.c.l.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.c.l.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.c.b
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_c_b_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.c.b.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_c_b_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.c.b.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.c.b.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.c.b.cl
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_c_b_cl_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.c.b.cl.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_c_b_cl_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.c.b.cl.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.c.lz
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_c_lz_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.c.lz.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_c_lz_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.c.lz.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.c.lz.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.c.cd
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_c_cd_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.c.cd.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_c_cd_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.c.cd.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.c.cd.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.c.cd.cl
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_c_cd_cl_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.c.cd.cl.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_c_cd_cl_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.o
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_o_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.o.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_o_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+; CHECK-LABEL: @extract_elt0_image_sample_o_v4f32_v4f32_v4i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.o.f32.v4f32.v4i32(<4 x float> %vaddr, <4 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_o_v4f32_v4f32_v4i32(<4 x float> %vaddr, <4 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.o.v4f32.v4f32.v4i32(<4 x float> %vaddr, <4 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+; CHECK-LABEL: @extract_elt0_image_sample_o_v4f32_v2f32_v4i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.o.f32.v2f32.v8i32(<2 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_o_v4f32_v2f32_v4i32(<2 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.o.v4f32.v2f32.v8i32(<2 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.o.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+declare <4 x float> @llvm.amdgcn.image.sample.o.v4f32.v4f32.v4i32(<4 x float>, <4 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+declare <4 x float> @llvm.amdgcn.image.sample.o.v4f32.v2f32.v8i32(<2 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.cl.o
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_cl_o_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.cl.o.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_cl_o_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.cl.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.cl.o.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.d.o
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_d_o_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.d.o.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_d_o_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.d.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.d.o.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.d.cl.o
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_d_cl_o_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.d.cl.o.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_d_cl_o_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.d.cl.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.d.cl.o.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.l.o
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_l_o_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.l.o.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_l_o_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.l.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.l.o.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.b.o
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_b_o_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.b.o.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_b_o_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.b.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.b.o.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.b.cl.o
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_b_cl_o_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.b.cl.o.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_b_cl_o_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.b.cl.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.b.cl.o.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.lz.o
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_lz_o_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.lz.o.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_lz_o_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.lz.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.lz.o.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.cd.o
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_cd_o_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.cd.o.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_cd_o_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.cd.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.cd.o.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.cd.cl.o
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_cd_cl_o_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.cd.cl.o.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_cd_cl_o_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.cd.cl.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.cd.cl.o.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.c.o
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_c_o_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.c.o.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_c_o_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.c.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+; CHECK-LABEL: @extract_elt0_image_sample_c_o_v4f32_v4f32_v4i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.c.o.f32.v4f32.v4i32(<4 x float> %vaddr, <4 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_c_o_v4f32_v4f32_v4i32(<4 x float> %vaddr, <4 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.c.o.v4f32.v4f32.v4i32(<4 x float> %vaddr, <4 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+; CHECK-LABEL: @extract_elt0_image_sample_c_o_v4f32_v2f32_v4i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.c.o.f32.v2f32.v8i32(<2 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_c_o_v4f32_v2f32_v4i32(<2 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.c.o.v4f32.v2f32.v8i32(<2 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.c.o.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.o.v4f32.v4f32.v4i32(<4 x float>, <4 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+declare <4 x float> @llvm.amdgcn.image.sample.c.o.v4f32.v2f32.v8i32(<2 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.c.cl.o
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_c_cl_o_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.c.cl.o.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_c_cl_o_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.c.cl.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.c.cl.o.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.c.d.o
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_c_d_o_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.c.d.o.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_c_d_o_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.c.d.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.c.d.o.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.c.d.cl.o
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_c_d_cl_o_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.c.d.cl.o.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_c_d_cl_o_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.c.d.cl.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.c.d.cl.o.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.c.l.o
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_c_l_o_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.c.l.o.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_c_l_o_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.c.l.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.c.l.o.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.c.b.o
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_c_b_o_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.c.b.o.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_c_b_o_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.c.b.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.c.b.o.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.c.b.cl.o
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_c_b_cl_o_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.c.b.cl.o.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_c_b_cl_o_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.c.b.cl.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.c.b.cl.o.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.c.lz.o
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_c_lz_o_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.c.lz.o.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_c_lz_o_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.c.lz.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.c.lz.o.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.c.cd.o
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_c_cd_o_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.c.cd.o.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_c_cd_o_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.c.cd.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.c.cd.o.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.sample.c.cd.cl.o
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_sample_c_cd_cl_o_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.sample.c.cd.cl.o.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_sample_c_cd_cl_o_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.sample.c.cd.cl.o.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.gather4
+; --------------------------------------------------------------------
+
+; Don't handle gather4*
+
+; CHECK-LABEL: @extract_elt0_image_gather4_v4f32_v4f32_v8i32(
+; CHECK: %data = call <4 x float> @llvm.amdgcn.image.gather4.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i3
+define amdgpu_ps float @extract_elt0_image_gather4_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.gather4.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+; CHECK-LABEL: @extract_elt0_image_gather4_v4f32_v4f32_v4i32(
+; CHECK-NEXT: %data = call <4 x float> @llvm.amdgcn.image.gather4.v4f32.v4f32.v4i32(<4 x float> %vaddr, <4 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+define amdgpu_ps float @extract_elt0_image_gather4_v4f32_v4f32_v4i32(<4 x float> %vaddr, <4 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.gather4.v4f32.v4f32.v4i32(<4 x float> %vaddr, <4 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+; CHECK-LABEL: @extract_elt0_image_gather4_v4f32_v2f32_v4i32(
+; CHECK-NEXT: %data = call <4 x float> @llvm.amdgcn.image.gather4.v4f32.v2f32.v8i32(<2 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+define amdgpu_ps float @extract_elt0_image_gather4_v4f32_v2f32_v4i32(<2 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.gather4.v4f32.v2f32.v8i32(<2 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.gather4.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.v4f32.v4f32.v4i32(<4 x float>, <4 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.v4f32.v2f32.v8i32(<2 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.gather4.cl
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_gather4_cl_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call <4 x float> @llvm.amdgcn.image.gather4.cl.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+define amdgpu_ps float @extract_elt0_image_gather4_cl_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.gather4.cl.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.gather4.cl.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.gather4.l
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_gather4_l_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call <4 x float> @llvm.amdgcn.image.gather4.l.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+define amdgpu_ps float @extract_elt0_image_gather4_l_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.gather4.l.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.gather4.l.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.gather4.b
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_gather4_b_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call <4 x float> @llvm.amdgcn.image.gather4.b.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+define amdgpu_ps float @extract_elt0_image_gather4_b_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.gather4.b.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.gather4.b.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.gather4.b.cl
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_gather4_b_cl_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call <4 x float> @llvm.amdgcn.image.gather4.b.cl.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+define amdgpu_ps float @extract_elt0_image_gather4_b_cl_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.gather4.b.cl.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.gather4.b.cl.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.gather4.lz
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_gather4_lz_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call <4 x float> @llvm.amdgcn.image.gather4.lz.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+define amdgpu_ps float @extract_elt0_image_gather4_lz_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.gather4.lz.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.gather4.lz.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.gather4.o
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_gather4_o_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call <4 x float> @llvm.amdgcn.image.gather4.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %gather4r, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+define amdgpu_ps float @extract_elt0_image_gather4_o_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.gather4.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %gather4r, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+; CHECK-LABEL: @extract_elt0_image_gather4_o_v4f32_v4f32_v4i32(
+; CHECK-NEXT: %data = call <4 x float> @llvm.amdgcn.image.gather4.o.v4f32.v4f32.v4i32(<4 x float> %vaddr, <4 x i32> %gather4r, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+define amdgpu_ps float @extract_elt0_image_gather4_o_v4f32_v4f32_v4i32(<4 x float> %vaddr, <4 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.gather4.o.v4f32.v4f32.v4i32(<4 x float> %vaddr, <4 x i32> %gather4r, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+; CHECK-LABEL: @extract_elt0_image_gather4_o_v4f32_v2f32_v4i32(
+; CHECK-NEXT: %data = call <4 x float> @llvm.amdgcn.image.gather4.o.v4f32.v2f32.v8i32(<2 x float> %vaddr, <8 x i32> %gather4r, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+define amdgpu_ps float @extract_elt0_image_gather4_o_v4f32_v2f32_v4i32(<2 x float> %vaddr, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.gather4.o.v4f32.v2f32.v8i32(<2 x float> %vaddr, <8 x i32> %gather4r, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.gather4.o.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.o.v4f32.v4f32.v4i32(<4 x float>, <4 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.o.v4f32.v2f32.v8i32(<2 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.gather4.cl.o
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_gather4_cl_o_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call <4 x float> @llvm.amdgcn.image.gather4.cl.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %gather4r, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+define amdgpu_ps float @extract_elt0_image_gather4_cl_o_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.gather4.cl.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %gather4r, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.gather4.cl.o.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.gather4.l.o
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_gather4_l_o_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call <4 x float> @llvm.amdgcn.image.gather4.l.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %gather4r, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+define amdgpu_ps float @extract_elt0_image_gather4_l_o_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.gather4.l.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %gather4r, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.gather4.l.o.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.gather4.b.o
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_gather4_b_o_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call <4 x float> @llvm.amdgcn.image.gather4.b.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %gather4r, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+define amdgpu_ps float @extract_elt0_image_gather4_b_o_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.gather4.b.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %gather4r, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.gather4.b.o.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.gather4.b.cl.o
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_gather4_b_cl_o_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call <4 x float> @llvm.amdgcn.image.gather4.b.cl.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %gather4r, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+define amdgpu_ps float @extract_elt0_image_gather4_b_cl_o_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.gather4.b.cl.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %gather4r, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.gather4.b.cl.o.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.gather4.lz.o
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_gather4_lz_o_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call <4 x float> @llvm.amdgcn.image.gather4.lz.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %gather4r, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+define amdgpu_ps float @extract_elt0_image_gather4_lz_o_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.gather4.lz.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %gather4r, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.gather4.lz.o.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.gather4.c.o
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_gather4_c_o_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call <4 x float> @llvm.amdgcn.image.gather4.c.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %gather4r, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+define amdgpu_ps float @extract_elt0_image_gather4_c_o_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.gather4.c.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %gather4r, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+; CHECK-LABEL: @extract_elt0_image_gather4_c_o_v4f32_v4f32_v4i32(
+; CHECK-NEXT: %data = call <4 x float> @llvm.amdgcn.image.gather4.c.o.v4f32.v4f32.v4i32(<4 x float> %vaddr, <4 x i32> %gather4r, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+define amdgpu_ps float @extract_elt0_image_gather4_c_o_v4f32_v4f32_v4i32(<4 x float> %vaddr, <4 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.gather4.c.o.v4f32.v4f32.v4i32(<4 x float> %vaddr, <4 x i32> %gather4r, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+; CHECK-LABEL: @extract_elt0_image_gather4_c_o_v4f32_v2f32_v4i32(
+; CHECK-NEXT: %data = call <4 x float> @llvm.amdgcn.image.gather4.c.o.v4f32.v2f32.v8i32(<2 x float> %vaddr, <8 x i32> %gather4r, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+define amdgpu_ps float @extract_elt0_image_gather4_c_o_v4f32_v2f32_v4i32(<2 x float> %vaddr, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.gather4.c.o.v4f32.v2f32.v8i32(<2 x float> %vaddr, <8 x i32> %gather4r, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.gather4.c.o.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.c.o.v4f32.v4f32.v4i32(<4 x float>, <4 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+declare <4 x float> @llvm.amdgcn.image.gather4.c.o.v4f32.v2f32.v8i32(<2 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.gather4.c.cl.o
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_gather4_c_cl_o_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call <4 x float> @llvm.amdgcn.image.gather4.c.cl.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %gather4r, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+define amdgpu_ps float @extract_elt0_image_gather4_c_cl_o_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.gather4.c.cl.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %gather4r, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.gather4.c.cl.o.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.gather4.c.l.o
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_gather4_c_l_o_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call <4 x float> @llvm.amdgcn.image.gather4.c.l.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %gather4r, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+define amdgpu_ps float @extract_elt0_image_gather4_c_l_o_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.gather4.c.l.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %gather4r, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.gather4.c.l.o.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.gather4.c.b.o
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_gather4_c_b_o_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call <4 x float> @llvm.amdgcn.image.gather4.c.b.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %gather4r, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+define amdgpu_ps float @extract_elt0_image_gather4_c_b_o_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.gather4.c.b.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %gather4r, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.gather4.c.b.o.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.gather4.c.b.cl.o
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_gather4_c_b_cl_o_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call <4 x float> @llvm.amdgcn.image.gather4.c.b.cl.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %gather4r, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+define amdgpu_ps float @extract_elt0_image_gather4_c_b_cl_o_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.gather4.c.b.cl.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %gather4r, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.gather4.c.b.cl.o.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.gather4.c.lz.o
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_gather4_c_lz_o_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call <4 x float> @llvm.amdgcn.image.gather4.c.lz.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %gather4r, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+define amdgpu_ps float @extract_elt0_image_gather4_c_lz_o_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %gather4r, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.gather4.c.lz.o.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %gather4r, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.gather4.c.lz.o.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
+; --------------------------------------------------------------------
+; llvm.amdgcn.image.getlod
+; --------------------------------------------------------------------
+
+; CHECK-LABEL: @extract_elt0_image_getlod_v4f32_v4f32_v8i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.getlod.f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_getlod_v4f32_v4f32_v8i32(<4 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.getlod.v4f32.v4f32.v8i32(<4 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+; CHECK-LABEL: @extract_elt0_image_getlod_v4f32_v4f32_v4i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.getlod.f32.v4f32.v4i32(<4 x float> %vaddr, <4 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_getlod_v4f32_v4f32_v4i32(<4 x float> %vaddr, <4 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.getlod.v4f32.v4f32.v4i32(<4 x float> %vaddr, <4 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+; CHECK-LABEL: @extract_elt0_image_getlod_v4f32_v2f32_v4i32(
+; CHECK-NEXT: %data = call float @llvm.amdgcn.image.getlod.f32.v2f32.v8i32(<2 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 1, i1 false, i1 false, i1 false, i1 false, i1 false)
+; CHECK-NEXT: ret float %data
+define amdgpu_ps float @extract_elt0_image_getlod_v4f32_v2f32_v4i32(<2 x float> %vaddr, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
+  %data = call <4 x float> @llvm.amdgcn.image.getlod.v4f32.v2f32.v8i32(<2 x float> %vaddr, <8 x i32> %sampler, <4 x i32> %rsrc, i32 15, i1 false, i1 false, i1 false, i1 false, i1 false)
+  %elt0 = extractelement <4 x float> %data, i32 0
+  ret float %elt0
+}
+
+declare <4 x float> @llvm.amdgcn.image.getlod.v4f32.v4f32.v8i32(<4 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+declare <4 x float> @llvm.amdgcn.image.getlod.v4f32.v4f32.v4i32(<4 x float>, <4 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+declare <4 x float> @llvm.amdgcn.image.getlod.v4f32.v2f32.v8i32(<2 x float>, <8 x i32>, <4 x i32>, i32, i1, i1, i1, i1, i1) #1
+
 attributes #0 = { nounwind }
 attributes #1 = { nounwind readonly }
 
 !0 = !{float 2.500000e+00}