[CodeGen][X86] Fix handling of __fp16 vectors.

This commit fixes a bug in IRGen where it generates completely broken
code for __fp16 vectors on X86. For example when the following code is
compiled:

half4 hv0, hv1, hv2; // these are vectors of __fp16.

void foo221() {
  hv0 = hv1 + hv2;
}

clang generates the following IR, in which two i16 vectors are added:

@hv1 = common global <4 x i16> zeroinitializer, align 8
@hv2 = common global <4 x i16> zeroinitializer, align 8
@hv0 = common global <4 x i16> zeroinitializer, align 8

define void @foo221() {
  %0 = load <4 x i16>, <4 x i16>* @hv1, align 8
  %1 = load <4 x i16>, <4 x i16>* @hv2, align 8
  %add = add <4 x i16> %0, %1
  store <4 x i16> %add, <4 x i16>* @hv0, align 8
  ret void
}

To fix the bug, this commit uses the code committed in r314056, which
modified clang to promote and truncate __fp16 vectors to and from float
vectors in the AST. It also fixes another IRGen bug where a short value
is assigned to an __fp16 variable without any integer-to-floating-point
conversion, as shown in the following example:

__fp16 a;
short b;

void foo1() {
  a = b;
}

@b = common global i16 0, align 2
@a = common global i16 0, align 2

define void @foo1() #0 {
  %0 = load i16, i16* @b, align 2
  store i16 %0, i16* @a, align 2
  ret void
}

rdar://problem/20625184

Differential Revision: https://reviews.llvm.org/D40112

llvm-svn: 320215

1 files changed, 6 insertions, 6 deletions

diff --git a/clang/lib/CodeGen/CGExprScalar.cpp b/clang/lib/CodeGen/CGExprScalar.cpp
index 7c111036177a..32e0f8853048 100644
--- a/clang/lib/CodeGen/CGExprScalar.cpp
+++ b/clang/lib/CodeGen/CGExprScalar.cpp
@@ -951,7 +951,7 @@ Value *ScalarExprEmitter::EmitScalarConversion(Value *Src, QualType SrcType,
   if (SrcType->isHalfType() && !CGF.getContext().getLangOpts().NativeHalfType) {
     // Cast to FP using the intrinsic if the half type itself isn't supported.
     if (DstTy->isFloatingPointTy()) {
-      if (!CGF.getContext().getLangOpts().HalfArgsAndReturns)
+      if (CGF.getContext().getTargetInfo().useFP16ConversionIntrinsics())
         return Builder.CreateCall(
             CGF.CGM.getIntrinsic(llvm::Intrinsic::convert_from_fp16, DstTy),
             Src);
@@ -959,7 +959,7 @@ Value *ScalarExprEmitter::EmitScalarConversion(Value *Src, QualType SrcType,
       // Cast to other types through float, using either the intrinsic or FPExt,
       // depending on whether the half type itself is supported
       // (as opposed to operations on half, available with NativeHalfType).
-      if (!CGF.getContext().getLangOpts().HalfArgsAndReturns) {
+      if (CGF.getContext().getTargetInfo().useFP16ConversionIntrinsics()) {
         Src = Builder.CreateCall(
             CGF.CGM.getIntrinsic(llvm::Intrinsic::convert_from_fp16,
                                  CGF.CGM.FloatTy),
@@ -1068,7 +1068,7 @@ Value *ScalarExprEmitter::EmitScalarConversion(Value *Src, QualType SrcType,
     if (SrcTy->isFloatingPointTy()) {
       // Use the intrinsic if the half type itself isn't supported
       // (as opposed to operations on half, available with NativeHalfType).
-      if (!CGF.getContext().getLangOpts().HalfArgsAndReturns)
+      if (CGF.getContext().getTargetInfo().useFP16ConversionIntrinsics())
         return Builder.CreateCall(
             CGF.CGM.getIntrinsic(llvm::Intrinsic::convert_to_fp16, SrcTy), Src);
       // If the half type is supported, just use an fptrunc.
@@ -1104,7 +1104,7 @@ Value *ScalarExprEmitter::EmitScalarConversion(Value *Src, QualType SrcType,
   }
 
   if (DstTy != ResTy) {
-    if (!CGF.getContext().getLangOpts().HalfArgsAndReturns) {
+    if (CGF.getContext().getTargetInfo().useFP16ConversionIntrinsics()) {
       assert(ResTy->isIntegerTy(16) && "Only half FP requires extra conversion");
       Res = Builder.CreateCall(
         CGF.CGM.getIntrinsic(llvm::Intrinsic::convert_to_fp16, CGF.CGM.FloatTy),
@@ -2028,7 +2028,7 @@ ScalarExprEmitter::EmitScalarPrePostIncDec(const UnaryOperator *E, LValue LV,
 
     if (type->isHalfType() && !CGF.getContext().getLangOpts().NativeHalfType) {
       // Another special case: half FP increment should be done via float
-      if (!CGF.getContext().getLangOpts().HalfArgsAndReturns) {
+      if (CGF.getContext().getTargetInfo().useFP16ConversionIntrinsics()) {
         value = Builder.CreateCall(
             CGF.CGM.getIntrinsic(llvm::Intrinsic::convert_from_fp16,
                                  CGF.CGM.FloatTy),
@@ -2063,7 +2063,7 @@ ScalarExprEmitter::EmitScalarPrePostIncDec(const UnaryOperator *E, LValue LV,
     value = Builder.CreateFAdd(value, amt, isInc ? "inc" : "dec");
 
     if (type->isHalfType() && !CGF.getContext().getLangOpts().NativeHalfType) {
-      if (!CGF.getContext().getLangOpts().HalfArgsAndReturns) {
+      if (CGF.getContext().getTargetInfo().useFP16ConversionIntrinsics()) {
         value = Builder.CreateCall(
             CGF.CGM.getIntrinsic(llvm::Intrinsic::convert_to_fp16,
                                  CGF.CGM.FloatTy),