diff --git a/lib/Target/Mips/MipsISelLowering.cpp b/lib/Target/Mips/MipsISelLowering.cpp
index 3503c87c001..496d00fe4eb 100644
--- a/lib/Target/Mips/MipsISelLowering.cpp
+++ b/lib/Target/Mips/MipsISelLowering.cpp
@@ -975,9 +975,15 @@ static bool CC_MipsO32(unsigned ValNo, MVT ValVT,
   // argument which is not f32 or f64.
   bool AllocateFloatsInIntReg = State.isVarArg() || ValNo > 1
       || State.getFirstUnallocated(F32Regs, FloatRegsSize) != ValNo;
+  unsigned OrigAlign = ArgFlags.getOrigAlign();
+  bool isI64 = (ValVT == MVT::i32 && OrigAlign == 8);
 
   if (ValVT == MVT::i32 || (ValVT == MVT::f32 && AllocateFloatsInIntReg)) {
     Reg = State.AllocateReg(IntRegs, IntRegsSize);
+    // If this is the first part of an i64 arg,
+    // the allocated register must be either A0 or A2.
+    if (isI64 && (Reg == Mips::A1 || Reg == Mips::A3))
+      Reg = State.AllocateReg(IntRegs, IntRegsSize);
     LocVT = MVT::i32;
   } else if (ValVT == MVT::f64 && AllocateFloatsInIntReg) {
     // Allocate int register and shadow next int register. If first
@@ -1006,7 +1012,7 @@ static bool CC_MipsO32(unsigned ValNo, MVT ValVT,
 
   if (!Reg) {
     unsigned SizeInBytes = ValVT.getSizeInBits() >> 3;
-    unsigned Offset = State.AllocateStack(SizeInBytes, SizeInBytes);
+    unsigned Offset = State.AllocateStack(SizeInBytes, OrigAlign);
     State.addLoc(CCValAssign::getMem(ValNo, ValVT, Offset, LocVT, LocInfo));
   } else
     State.addLoc(CCValAssign::getReg(ValNo, ValVT, Reg, LocVT, LocInfo));
diff --git a/test/CodeGen/Mips/i64arg.ll b/test/CodeGen/Mips/i64arg.ll
new file mode 100644
index 00000000000..8e8114b27c3
--- /dev/null
+++ b/test/CodeGen/Mips/i64arg.ll
@@ -0,0 +1,34 @@
+; RUN: llc -march=mips -mcpu=4ke < %s | FileCheck %s
+
+define void @f1(i64 %ll1, float %f, i64 %ll, i32 %i, float %f2) nounwind {
+entry:
+; CHECK: addu $[[R1:[0-9]+]], $zero, $5
+; CHECK: addu $[[R0:[0-9]+]], $zero, $4
+; CHECK: ori $6, ${{[0-9]+}}, 3855
+; CHECK: ori $7, ${{[0-9]+}}, 22136
+; CHECK: lw  $25, %call16(ff1)
+; CHECK: jalr
+  tail call void @ff1(i32 %i, i64 1085102592623924856) nounwind
+; CHECK: lw $[[R2:[0-9]+]], 96($sp)
+; CHECK: lw $[[R3:[0-9]+]], 100($sp)
+; CHECK: addu $4, $zero, $[[R2]]
+; CHECK: addu $5, $zero, $[[R3]]
+; CHECK: lw $25, %call16(ff2)
+; CHECK: jalr $25
+  tail call void @ff2(i64 %ll, double 3.000000e+00) nounwind
+  %sub = add nsw i32 %i, -1
+; CHECK: sw $[[R0]], 24($sp)
+; CHECK: sw $[[R1]], 28($sp)
+; CHECK: addu $6, $zero, $[[R2]]
+; CHECK: addu $7, $zero, $[[R3]]
+; CHECK: lw $25, %call16(ff3)
+; CHECK: jalr $25
+  tail call void @ff3(i32 %i, i64 %ll, i32 %sub, i64 %ll1) nounwind
+  ret void
+}
+
+declare void @ff1(i32, i64)
+
+declare void @ff2(i64, double)
+
+declare void @ff3(i32, i64, i32, i64)