diff --git a/lib/Target/ARM/ARMInstrFormats.td b/lib/Target/ARM/ARMInstrFormats.td
index fec626acf5e..1b802f18439 100644
--- a/lib/Target/ARM/ARMInstrFormats.td
+++ b/lib/Target/ARM/ARMInstrFormats.td
@@ -911,13 +911,26 @@ class AsMul1I<bits<7> opcod, dag oops, dag iops, InstrItinClass itin,
 }
 
 // Most significant word multiply
-class AMul2I<bits<7> opcod, dag oops, dag iops, InstrItinClass itin,
-             string opc, string asm, list<dag> pattern>
+class AMul2I<bits<7> opcod, bits<4> opc7_4, dag oops, dag iops,
+             InstrItinClass itin, string opc, string asm, list<dag> pattern>
   : I<oops, iops, AddrModeNone, Size4Bytes, IndexModeNone, MulFrm, itin,
       opc, asm, "", pattern> {
-  let Inst{7-4}   = 0b1001;
+  bits<4> Rd;
+  bits<4> Rn;
+  bits<4> Rm;
+  let Inst{7-4}   = opc7_4;
   let Inst{20}    = 1;
   let Inst{27-21} = opcod;
+  let Inst{19-16} = Rd;
+  let Inst{11-8}  = Rm;
+  let Inst{3-0}   = Rn;
+}
+// MSW multiple w/ Ra operand
+class AMul2Ia<bits<7> opcod, bits<4> opc7_4, dag oops, dag iops,
+              InstrItinClass itin, string opc, string asm, list<dag> pattern>
+  : AMul2I<opcod, opc7_4, oops, iops, itin, opc, asm, pattern> {
+  bits<4> Ra;
+  let Inst{15-12} = Ra;
 }
 
 // SMUL<x><y> / SMULW<y> / SMLA<x><y> / SMLAW<x><y>
diff --git a/lib/Target/ARM/ARMInstrInfo.td b/lib/Target/ARM/ARMInstrInfo.td
index 35fa2b4507a..d28c8bc4f42 100644
--- a/lib/Target/ARM/ARMInstrInfo.td
+++ b/lib/Target/ARM/ARMInstrInfo.td
@@ -2265,8 +2265,8 @@ class AsMul1I64<bits<7> opcod, dag oops, dag iops, InstrItinClass itin,
   bits<4> RdHi;
   bits<4> Rm;
   bits<4> Rn;
-  let Inst{19-16} = RdLo;
-  let Inst{15-12} = RdHi;
+  let Inst{19-16} = RdHi;
+  let Inst{15-12} = RdLo;
   let Inst{11-8}  = Rm;
   let Inst{3-0}   = Rn;
 }
@@ -2333,49 +2333,43 @@ def UMAAL : AMul1I <0b0000010, (outs GPR:$RdLo, GPR:$RdHi),
 } // neverHasSideEffects
 
 // Most significant word multiply
-def SMMUL : AMul2I <0b0111010, (outs GPR:$dst), (ins GPR:$a, GPR:$b),
-               IIC_iMUL32, "smmul", "\t$dst, $a, $b",
-               [(set GPR:$dst, (mulhs GPR:$a, GPR:$b))]>,
+def SMMUL : AMul2I <0b0111010, 0b0001, (outs GPR:$Rd), (ins GPR:$Rn, GPR:$Rm),
+               IIC_iMUL32, "smmul", "\t$Rd, $Rn, $Rm",
+               [(set GPR:$Rd, (mulhs GPR:$Rn, GPR:$Rm))]>,
             Requires<[IsARM, HasV6]> {
-  let Inst{7-4}   = 0b0001;
   let Inst{15-12} = 0b1111;
 }
 
-def SMMULR : AMul2I <0b0111010, (outs GPR:$dst), (ins GPR:$a, GPR:$b),
-               IIC_iMUL32, "smmulr", "\t$dst, $a, $b",
+def SMMULR : AMul2I <0b0111010, 0b0011, (outs GPR:$Rd), (ins GPR:$Rn, GPR:$Rm),
+               IIC_iMUL32, "smmulr", "\t$Rd, $Rn, $Rm",
                [/* For disassembly only; pattern left blank */]>,
             Requires<[IsARM, HasV6]> {
-  let Inst{7-4}   = 0b0011; // R = 1
   let Inst{15-12} = 0b1111;
 }
 
-def SMMLA : AMul2I <0b0111010, (outs GPR:$dst), (ins GPR:$a, GPR:$b, GPR:$c),
-               IIC_iMAC32, "smmla", "\t$dst, $a, $b, $c",
-               [(set GPR:$dst, (add (mulhs GPR:$a, GPR:$b), GPR:$c))]>,
-            Requires<[IsARM, HasV6]> {
-  let Inst{7-4}   = 0b0001;
-}
+def SMMLA : AMul2Ia <0b0111010, 0b0001, (outs GPR:$Rd),
+               (ins GPR:$Rn, GPR:$Rm, GPR:$Ra),
+               IIC_iMAC32, "smmla", "\t$Rd, $Rn, $Rm, $Ra",
+               [(set GPR:$Rd, (add (mulhs GPR:$Rn, GPR:$Rm), GPR:$Ra))]>,
+            Requires<[IsARM, HasV6]>;
 
-def SMMLAR : AMul2I <0b0111010, (outs GPR:$dst), (ins GPR:$a, GPR:$b, GPR:$c),
-               IIC_iMAC32, "smmlar", "\t$dst, $a, $b, $c",
+def SMMLAR : AMul2Ia <0b0111010, 0b0011, (outs GPR:$Rd),
+               (ins GPR:$Rn, GPR:$Rm, GPR:$Ra),
+               IIC_iMAC32, "smmlar", "\t$Rd, $Rn, $Rm, $Ra",
                [/* For disassembly only; pattern left blank */]>,
-            Requires<[IsARM, HasV6]> {
-  let Inst{7-4}   = 0b0011; // R = 1
-}
+            Requires<[IsARM, HasV6]>;
 
-def SMMLS : AMul2I <0b0111010, (outs GPR:$dst), (ins GPR:$a, GPR:$b, GPR:$c),
-               IIC_iMAC32, "smmls", "\t$dst, $a, $b, $c",
-               [(set GPR:$dst, (sub GPR:$c, (mulhs GPR:$a, GPR:$b)))]>,
-            Requires<[IsARM, HasV6]> {
-  let Inst{7-4}   = 0b1101;
-}
+def SMMLS : AMul2Ia <0b0111010, 0b1101, (outs GPR:$Rd),
+               (ins GPR:$Rn, GPR:$Rm, GPR:$Ra),
+               IIC_iMAC32, "smmls", "\t$Rd, $Rn, $Rm, $Ra",
+               [(set GPR:$Rd, (sub GPR:$Ra, (mulhs GPR:$Rn, GPR:$Rm)))]>,
+            Requires<[IsARM, HasV6]>;
 
-def SMMLSR : AMul2I <0b0111010, (outs GPR:$dst), (ins GPR:$a, GPR:$b, GPR:$c),
-               IIC_iMAC32, "smmlsr", "\t$dst, $a, $b, $c",
+def SMMLSR : AMul2Ia <0b0111010, 0b1111, (outs GPR:$Rd),
+               (ins GPR:$Rn, GPR:$Rm, GPR:$Ra),
+               IIC_iMAC32, "smmlsr", "\t$Rd, $Rn, $Rm, $Ra",
                [/* For disassembly only; pattern left blank */]>,
-            Requires<[IsARM, HasV6]> {
-  let Inst{7-4}   = 0b1111; // R = 1
-}
+            Requires<[IsARM, HasV6]>;
 
 multiclass AI_smul<string opc, PatFrag opnode> {
   def BB : AMulxyI<0b0001011, (outs GPR:$dst), (ins GPR:$a, GPR:$b),
diff --git a/test/MC/ARM/simple-encoding.ll b/test/MC/ARM/simple-encoding.ll
index b8df75ab7c9..970e0e616c4 100644
--- a/test/MC/ARM/simple-encoding.ll
+++ b/test/MC/ARM/simple-encoding.ll
@@ -128,4 +128,26 @@ define i64 @f13() {
 entry:
         ret i64 9223372036854775807
 }
+
+define i32 @f14(i32 %x, i32 %y) {
+; CHECK: f14:
+; CHECK: smmul  r0, r1, r0            @ encoding: [0x11,0xf0,0x50,0xe7]
+        %tmp = sext i32 %x to i64
+        %tmp1 = sext i32 %y to i64
+        %tmp2 = mul i64 %tmp1, %tmp
+        %tmp3 = lshr i64 %tmp2, 32
+        %tmp3.upgrd.1 = trunc i64 %tmp3 to i32
+        ret i32 %tmp3.upgrd.1
+}
+
+define i32 @f15(i32 %x, i32 %y) {
+; CHECK: f15:
+; CHECK: umull  r1, r0, r1, r0        @ encoding: [0x91,0x10,0x80,0xe0]
+        %tmp = zext i32 %x to i64
+        %tmp1 = zext i32 %y to i64
+        %tmp2 = mul i64 %tmp1, %tmp
+        %tmp3 = lshr i64 %tmp2, 32
+        %tmp3.upgrd.2 = trunc i64 %tmp3 to i32
+        ret i32 %tmp3.upgrd.2
+}
 declare void @llvm.trap() nounwind