Properly handle indirect win64 args when they're passed in memory

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@79009 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/lib/Target/X86/X86ISelLowering.cpp b/lib/Target/X86/X86ISelLowering.cpp
index ca6ff929a77d2c67b6cf119f1a0843b76bf51181..e8f4f43dc3d98cc6e0b277c2f0fc81e2c0dc6fed 100644 (file)
--- a/lib/Target/X86/X86ISelLowering.cpp
+++ b/lib/Target/X86/X86ISelLowering.cpp
@@ -1328,17 +1328,25 @@ X86TargetLowering::LowerMemArgument(SDValue Chain,
   ISD::ArgFlagsTy Flags = Ins[i].Flags;
   bool AlwaysUseMutable = (CallConv==CallingConv::Fast) && PerformTailCallOpt;
   bool isImmutable = !AlwaysUseMutable && !Flags.isByVal();
+  EVT ValVT;
+
+  // If value is passed by pointer we have address passed instead of the value
+  // itself.
+  if (VA.getLocInfo() == CCValAssign::Indirect)
+    ValVT = VA.getLocVT();
+  else
+    ValVT = VA.getValVT();
 
   // FIXME: For now, all byval parameter objects are marked mutable. This can be
   // changed with more analysis.
   // In case of tail call optimization mark all arguments mutable. Since they
   // could be overwritten by lowering of arguments in case of a tail call.
-  int FI = MFI->CreateFixedObject(VA.getValVT().getSizeInBits()/8,
+  int FI = MFI->CreateFixedObject(ValVT.getSizeInBits()/8,
                                   VA.getLocMemOffset(), isImmutable);
   SDValue FIN = DAG.getFrameIndex(FI, getPointerTy());
   if (Flags.isByVal())
     return FIN;
-  return DAG.getLoad(VA.getValVT(), dl, Chain, FIN,
+  return DAG.getLoad(ValVT, dl, Chain, FIN,
                      PseudoSourceValue::getFixedStack(FI), 0);
 }
 

diff --git a/test/CodeGen/X86/2009-08-14-Win64MemoryIndirectArg.ll b/test/CodeGen/X86/2009-08-14-Win64MemoryIndirectArg.ll
new file mode 100644 (file)
index 0000000..5431462
--- /dev/null
+++ b/test/CodeGen/X86/2009-08-14-Win64MemoryIndirectArg.ll
@@ -0,0 +1,57 @@
+; RUN: llvm-as < %s | llc
+target triple = "x86_64-mingw"
+
+; ModuleID = 'mm.bc'
+       type opaque             ; type %0
+       type opaque             ; type %1
+
+define internal fastcc float @computeMipmappingRho(%0* %shaderExecutionStatePtr, i32 %index, <4 x float> %texCoord, <4 x float> %texCoordDX, <4 x float> %texCoordDY) readonly {
+indexCheckBlock:
+       %indexCmp = icmp ugt i32 %index, 16             ; <i1> [#uses=1]
+       br i1 %indexCmp, label %zeroReturnBlock, label %primitiveTextureFetchBlock
+
+primitiveTextureFetchBlock:            ; preds = %indexCheckBlock
+       %pointerArithmeticTmp = bitcast %0* %shaderExecutionStatePtr to i8*             ; <i8*> [#uses=1]
+       %pointerArithmeticTmp1 = getelementptr i8* %pointerArithmeticTmp, i64 1808              ; <i8*> [#uses=1]
+       %pointerArithmeticTmp2 = bitcast i8* %pointerArithmeticTmp1 to %1**             ; <%1**> [#uses=1]
+       %primitivePtr = load %1** %pointerArithmeticTmp2                ; <%1*> [#uses=1]
+       %pointerArithmeticTmp3 = bitcast %1* %primitivePtr to i8*               ; <i8*> [#uses=1]
+       %pointerArithmeticTmp4 = getelementptr i8* %pointerArithmeticTmp3, i64 19408            ; <i8*> [#uses=1]
+       %pointerArithmeticTmp5 = bitcast i8* %pointerArithmeticTmp4 to %1**             ; <%1**> [#uses=1]
+       %primitiveTexturePtr = getelementptr %1** %pointerArithmeticTmp5, i32 %index            ; <%1**> [#uses=1]
+       %primitiveTexturePtr6 = load %1** %primitiveTexturePtr          ; <%1*> [#uses=2]
+       br label %textureCheckBlock
+
+textureCheckBlock:             ; preds = %primitiveTextureFetchBlock
+       %texturePtrInt = ptrtoint %1* %primitiveTexturePtr6 to i64              ; <i64> [#uses=1]
+       %testTextureNULL = icmp eq i64 %texturePtrInt, 0                ; <i1> [#uses=1]
+       br i1 %testTextureNULL, label %zeroReturnBlock, label %rhoCalculateBlock
+
+rhoCalculateBlock:             ; preds = %textureCheckBlock
+       %pointerArithmeticTmp7 = bitcast %1* %primitiveTexturePtr6 to i8*               ; <i8*> [#uses=1]
+       %pointerArithmeticTmp8 = getelementptr i8* %pointerArithmeticTmp7, i64 640              ; <i8*> [#uses=1]
+       %pointerArithmeticTmp9 = bitcast i8* %pointerArithmeticTmp8 to <4 x float>*             ; <<4 x float>*> [#uses=1]
+       %dimensionsPtr = load <4 x float>* %pointerArithmeticTmp9, align 1              ; <<4 x float>> [#uses=2]
+       %texDiffDX = fsub <4 x float> %texCoordDX, %texCoord            ; <<4 x float>> [#uses=1]
+       %texDiffDY = fsub <4 x float> %texCoordDY, %texCoord            ; <<4 x float>> [#uses=1]
+       %ddx = fmul <4 x float> %texDiffDX, %dimensionsPtr              ; <<4 x float>> [#uses=2]
+       %ddx10 = fmul <4 x float> %texDiffDY, %dimensionsPtr            ; <<4 x float>> [#uses=2]
+       %ddxSquared = fmul <4 x float> %ddx, %ddx               ; <<4 x float>> [#uses=3]
+       %0 = shufflevector <4 x float> %ddxSquared, <4 x float> %ddxSquared, <4 x i32> <i32 1, i32 0, i32 0, i32 0>             ; <<4 x float>> [#uses=1]
+       %dxSquared = fadd <4 x float> %ddxSquared, %0           ; <<4 x float>> [#uses=1]
+       %1 = call <4 x float> @llvm.x86.sse.sqrt.ss(<4 x float> %dxSquared)             ; <<4 x float>> [#uses=1]
+       %ddySquared = fmul <4 x float> %ddx10, %ddx10           ; <<4 x float>> [#uses=3]
+       %2 = shufflevector <4 x float> %ddySquared, <4 x float> %ddySquared, <4 x i32> <i32 1, i32 0, i32 0, i32 0>             ; <<4 x float>> [#uses=1]
+       %dySquared = fadd <4 x float> %ddySquared, %2           ; <<4 x float>> [#uses=1]
+       %3 = call <4 x float> @llvm.x86.sse.sqrt.ss(<4 x float> %dySquared)             ; <<4 x float>> [#uses=1]
+       %4 = call <4 x float> @llvm.x86.sse.max.ss(<4 x float> %1, <4 x float> %3)              ; <<4 x float>> [#uses=1]
+       %rho = extractelement <4 x float> %4, i32 0             ; <float> [#uses=1]
+       ret float %rho
+
+zeroReturnBlock:               ; preds = %textureCheckBlock, %indexCheckBlock
+       ret float 0.000000e+00
+}
+
+declare <4 x float> @llvm.x86.sse.sqrt.ss(<4 x float>) nounwind readnone
+
+declare <4 x float> @llvm.x86.sse.max.ss(<4 x float>, <4 x float>) nounwind readnone