Don't simplify image intrinsic with dmask == 0

Author: nikic

llvm/lib/Target/AMDGPU/AMDGPUInstCombineIntrinsic.cpp

@@ -1241,6 +1241,10 @@ static Value *simplifyAMDGCNMemoryIntrinsicDemanded(InstCombiner &IC,
     ConstantInt *DMask = cast<ConstantInt>(Args[DMaskIdx]);
     unsigned DMaskVal = DMask->getZExtValue() & 0xf;
 
+    // dmask 0 has special semantics, do not simplify.
+    if (DMaskVal == 0)
+      return nullptr;
+
     // Mask off values that are undefined because the dmask doesn't cover them
     DemandedElts &= (1 << llvm::popcount(DMaskVal)) - 1;
 

llvm/test/Transforms/InstCombine/AMDGPU/amdgcn-demanded-vector-elts-inseltpoison.ll

@@ -4792,7 +4792,9 @@ define amdgpu_ps float @extract_elt0_image_sample_2d_v4f32_f32(float %s, float %
 
 define amdgpu_ps float @extract_elt0_dmask_0000_image_sample_3d_v4f32_f32(float %s, float %t, float %r, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
 ; CHECK-LABEL: @extract_elt0_dmask_0000_image_sample_3d_v4f32_f32(
-; CHECK-NEXT:    ret float poison
+; CHECK-NEXT:    [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.3d.v4f32.f32(i32 0, float [[S:%.*]], float [[T:%.*]], float [[R:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT:    [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
+; CHECK-NEXT:    ret float [[ELT0]]
 ;
   %data = call <4 x float> @llvm.amdgcn.image.sample.3d.v4f32.f32(i32 0, float %s, float %t, float %r, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
   %elt0 = extractelement <4 x float> %data, i32 0

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

@@ -4791,7 +4791,9 @@ define amdgpu_ps float @extract_elt0_image_sample_2d_v4f32_f32(float %s, float %
 
 define amdgpu_ps float @extract_elt0_dmask_0000_image_sample_3d_v4f32_f32(float %s, float %t, float %r, <8 x i32> inreg %sampler, <4 x i32> inreg %rsrc) #0 {
 ; CHECK-LABEL: @extract_elt0_dmask_0000_image_sample_3d_v4f32_f32(
-; CHECK-NEXT:    ret float poison
+; CHECK-NEXT:    [[DATA:%.*]] = call <4 x float> @llvm.amdgcn.image.sample.3d.v4f32.f32(i32 0, float [[S:%.*]], float [[T:%.*]], float [[R:%.*]], <8 x i32> [[SAMPLER:%.*]], <4 x i32> [[RSRC:%.*]], i1 false, i32 0, i32 0)
+; CHECK-NEXT:    [[ELT0:%.*]] = extractelement <4 x float> [[DATA]], i64 0
+; CHECK-NEXT:    ret float [[ELT0]]
 ;
   %data = call <4 x float> @llvm.amdgcn.image.sample.3d.v4f32.f32(i32 0, float %s, float %t, float %r, <8 x i32> %sampler, <4 x i32> %rsrc, i1 false, i32 0, i32 0)
   %elt0 = extractelement <4 x float> %data, i32 0