Index: llvm/include/llvm/CodeGen/BasicTTIImpl.h
===================================================================
--- llvm/include/llvm/CodeGen/BasicTTIImpl.h
+++ llvm/include/llvm/CodeGen/BasicTTIImpl.h
@@ -785,13 +785,14 @@
         return 0;
 
       // If this is a zext/sext of a load, return 0 if the corresponding
-      // extending load exists on target.
+      // extending load exists on target and the result type is legal.
       if (CCH == TTI::CastContextHint::Normal) {
         EVT ExtVT = EVT::getEVT(Dst);
         EVT LoadVT = EVT::getEVT(Src);
         unsigned LType =
           ((Opcode == Instruction::ZExt) ? ISD::ZEXTLOAD : ISD::SEXTLOAD);
-        if (TLI->isLoadExtLegal(LType, ExtVT, LoadVT))
+        if (TLI->isLoadExtLegal(LType, ExtVT, LoadVT) &&
+            DstLT.first == SrcLT.first)
           return 0;
       }
       break;
Index: llvm/test/Analysis/CostModel/AArch64/cast.ll
===================================================================
--- llvm/test/Analysis/CostModel/AArch64/cast.ll
+++ llvm/test/Analysis/CostModel/AArch64/cast.ll
@@ -707,6 +707,7 @@
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %loadv2i16 = load <2 x i16>, <2 x i16>* undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %loadv4i16 = load <4 x i16>, <4 x i16>* undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %loadv2i32 = load <2 x i32>, <2 x i32>* undef
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %loadv4i32 = load <4 x i32>, <4 x i32>* undef
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %r0 = sext i8 %loadi8 to i16
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %r1 = zext i8 %loadi8 to i16
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %r2 = sext i8 %loadi8 to i32
@@ -731,6 +732,8 @@
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v9 = zext <2 x i16> %loadv2i16 to <2 x i64>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v10 = sext <2 x i32> %loadv2i32 to <2 x i64>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v11 = zext <2 x i32> %loadv2i32 to <2 x i64>
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v12 = sext <4 x i32> %loadv4i32 to <4 x i64>
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v13 = zext <4 x i32> %loadv4i32 to <4 x i64>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret i32 undef
 ;
   %loadi8 = load i8, i8* undef
@@ -742,6 +745,7 @@
   %loadv2i16 = load <2 x i16>, <2 x i16>* undef
   %loadv4i16 = load <4 x i16>, <4 x i16>* undef
   %loadv2i32 = load <2 x i32>, <2 x i32>* undef
+  %loadv4i32 = load <4 x i32>, <4 x i32>* undef
 
   %r0 = sext i8 %loadi8 to i16
   %r1 = zext i8 %loadi8 to i16
@@ -768,6 +772,8 @@
   %v9 = zext <2 x i16> %loadv2i16 to <2 x i64>
   %v10 = sext <2 x i32> %loadv2i32 to <2 x i64>
   %v11 = zext <2 x i32> %loadv2i32 to <2 x i64>
+  %v12 = sext <4 x i32> %loadv4i32 to <4 x i64>
+  %v13 = zext <4 x i32> %loadv4i32 to <4 x i64>
 
   ret i32 undef
 }
Index: llvm/test/Analysis/CostModel/ARM/cast_ldst.ll
===================================================================
--- llvm/test/Analysis/CostModel/ARM/cast_ldst.ll
+++ llvm/test/Analysis/CostModel/ARM/cast_ldst.ll
@@ -45,8 +45,8 @@
 ; CHECK-NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v4816u = zext <4 x i8> %loadv4i8 to <4 x i16>
 ; CHECK-NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4832s = sext <4 x i8> %loadv4i8 to <4 x i32>
 ; CHECK-NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4832u = zext <4 x i8> %loadv4i8 to <4 x i32>
-; CHECK-NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v4864s = sext <4 x i8> %loadv4i8 to <4 x i64>
-; CHECK-NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v4864u = zext <4 x i8> %loadv4i8 to <4 x i64>
+; CHECK-NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v4864s = sext <4 x i8> %loadv4i8 to <4 x i64>
+; CHECK-NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v4864u = zext <4 x i8> %loadv4i8 to <4 x i64>
 ; CHECK-NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v8816s = sext <8 x i8> %loadv8i8 to <8 x i16>
 ; CHECK-NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v8816u = zext <8 x i8> %loadv8i8 to <8 x i16>
 ; CHECK-NEON-RECIP-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v8832s = sext <8 x i8> %loadv8i8 to <8 x i32>
@@ -313,8 +313,8 @@
 ; CHECK-V8R-RECIP-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v4816u = zext <4 x i8> %loadv4i8 to <4 x i16>
 ; CHECK-V8R-RECIP-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4832s = sext <4 x i8> %loadv4i8 to <4 x i32>
 ; CHECK-V8R-RECIP-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %v4832u = zext <4 x i8> %loadv4i8 to <4 x i32>
-; CHECK-V8R-RECIP-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v4864s = sext <4 x i8> %loadv4i8 to <4 x i64>
-; CHECK-V8R-RECIP-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v4864u = zext <4 x i8> %loadv4i8 to <4 x i64>
+; CHECK-V8R-RECIP-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v4864s = sext <4 x i8> %loadv4i8 to <4 x i64>
+; CHECK-V8R-RECIP-NEXT:  Cost Model: Found an estimated cost of 7 for instruction: %v4864u = zext <4 x i8> %loadv4i8 to <4 x i64>
 ; CHECK-V8R-RECIP-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v8816s = sext <8 x i8> %loadv8i8 to <8 x i16>
 ; CHECK-V8R-RECIP-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v8816u = zext <8 x i8> %loadv8i8 to <8 x i16>
 ; CHECK-V8R-RECIP-NEXT:  Cost Model: Found an estimated cost of 3 for instruction: %v8832s = sext <8 x i8> %loadv8i8 to <8 x i32>
@@ -581,8 +581,8 @@
 ; CHECK-V8R-SIZE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v4816u = zext <4 x i8> %loadv4i8 to <4 x i16>
 ; CHECK-V8R-SIZE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4832s = sext <4 x i8> %loadv4i8 to <4 x i32>
 ; CHECK-V8R-SIZE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4832u = zext <4 x i8> %loadv4i8 to <4 x i32>
-; CHECK-V8R-SIZE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v4864s = sext <4 x i8> %loadv4i8 to <4 x i64>
-; CHECK-V8R-SIZE-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %v4864u = zext <4 x i8> %loadv4i8 to <4 x i64>
+; CHECK-V8R-SIZE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4864s = sext <4 x i8> %loadv4i8 to <4 x i64>
+; CHECK-V8R-SIZE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v4864u = zext <4 x i8> %loadv4i8 to <4 x i64>
 ; CHECK-V8R-SIZE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v8816s = sext <8 x i8> %loadv8i8 to <8 x i16>
 ; CHECK-V8R-SIZE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v8816u = zext <8 x i8> %loadv8i8 to <8 x i16>
 ; CHECK-V8R-SIZE-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: %v8832s = sext <8 x i8> %loadv8i8 to <8 x i32>