Have the X86 back-end emit the alias instead of what's being aliased. In most

cases, it's much nicer and more informative reading the alias.


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@129497 91177308-0d34-0410-b5e6-96231b3b80d8

33 files changed, 175 insertions, 163 deletions

diff --git a/lib/Target/X86/InstPrinter/X86ATTInstPrinter.cpp b/lib/Target/X86/InstPrinter/X86ATTInstPrinter.cpp
index d006eca344..69a069e438 100644
--- a/lib/Target/X86/InstPrinter/X86ATTInstPrinter.cpp
+++ b/lib/Target/X86/InstPrinter/X86ATTInstPrinter.cpp
@@ -42,7 +42,8 @@ X86ATTInstPrinter::X86ATTInstPrinter(TargetMachine &TM, const MCAsmInfo &MAI)
 }
 
 void X86ATTInstPrinter::printInst(const MCInst *MI, raw_ostream &OS) {
-  printInstruction(MI, OS);
+  if (printAliasInstr(MI, OS))
+    printInstruction(MI, OS);
   
   // If verbose assembly is enabled, we can print some informative comments.
   if (CommentStream)
diff --git a/lib/Target/X86/X86InstrInfo.td b/lib/Target/X86/X86InstrInfo.td
index 251b2fa170..177a2794a3 100644
--- a/lib/Target/X86/X86InstrInfo.td
+++ b/lib/Target/X86/X86InstrInfo.td
@@ -1437,7 +1437,7 @@ def : InstAlias<"idivq $src, %rax", (IDIV64m i64mem:$src)>;
 
 // Various unary fpstack operations default to operating on on ST1.
 // For example, "fxch" -> "fxch %st(1)"
-def : InstAlias<"faddp",        (ADD_FPrST0  ST1)>;
+def : InstAlias<"faddp",        (ADD_FPrST0  ST1), 0>;
 def : InstAlias<"fsubp",        (SUBR_FPrST0 ST1)>;
 def : InstAlias<"fsubrp",       (SUB_FPrST0  ST1)>;
 def : InstAlias<"fmulp",        (MUL_FPrST0  ST1)>;
@@ -1455,13 +1455,15 @@ def : InstAlias<"fucompi",      (UCOM_FIPr   ST1)>;
 // For example, "fadd %st(4), %st(0)" -> "fadd %st(4)".  We also disambiguate
 // instructions like "fadd %st(0), %st(0)" as "fadd %st(0)" for consistency with
 // gas.
-multiclass FpUnaryAlias<string Mnemonic, Instruction Inst> {
- def : InstAlias<!strconcat(Mnemonic, " $op, %st(0)"),    (Inst RST:$op)>;
- def : InstAlias<!strconcat(Mnemonic, " %st(0), %st(0)"), (Inst ST0)>;
+multiclass FpUnaryAlias<string Mnemonic, Instruction Inst, bit EmitAlias = 1> {
+ def : InstAlias<!strconcat(Mnemonic, " $op, %st(0)"),
+                 (Inst RST:$op), EmitAlias>;
+ def : InstAlias<!strconcat(Mnemonic, " %st(0), %st(0)"),
+                 (Inst ST0), EmitAlias>;
 }
 
 defm : FpUnaryAlias<"fadd",   ADD_FST0r>;
-defm : FpUnaryAlias<"faddp",  ADD_FPrST0>;
+defm : FpUnaryAlias<"faddp",  ADD_FPrST0, 0>;
 defm : FpUnaryAlias<"fsub",   SUB_FST0r>;
 defm : FpUnaryAlias<"fsubp",  SUBR_FPrST0>;
 defm : FpUnaryAlias<"fsubr",  SUBR_FST0r>;
@@ -1472,8 +1474,8 @@ defm : FpUnaryAlias<"fdiv",   DIV_FST0r>;
 defm : FpUnaryAlias<"fdivp",  DIVR_FPrST0>;
 defm : FpUnaryAlias<"fdivr",  DIVR_FST0r>;
 defm : FpUnaryAlias<"fdivrp", DIV_FPrST0>;
-defm : FpUnaryAlias<"fcomi",   COM_FIr>;
-defm : FpUnaryAlias<"fucomi",  UCOM_FIr>;
+defm : FpUnaryAlias<"fcomi",   COM_FIr, 0>;
+defm : FpUnaryAlias<"fucomi",  UCOM_FIr, 0>;
 defm : FpUnaryAlias<"fcompi",   COM_FIPr>;
 defm : FpUnaryAlias<"fucompi",  UCOM_FIPr>;
 
@@ -1481,7 +1483,7 @@ defm : FpUnaryAlias<"fucompi",  UCOM_FIPr>;
 // Handle "f{mulp,addp} st(0), $op" the same as "f{mulp,addp} $op", since they
 // commute.  We also allow fdiv[r]p/fsubrp even though they don't commute,
 // solely because gas supports it.
-def : InstAlias<"faddp %st(0), $op", (ADD_FPrST0 RST:$op)>;
+def : InstAlias<"faddp %st(0), $op", (ADD_FPrST0 RST:$op), 0>;
 def : InstAlias<"fmulp %st(0), $op", (MUL_FPrST0 RST:$op)>;
 def : InstAlias<"fsubrp %st(0), $op", (SUB_FPrST0 RST:$op)>;
 def : InstAlias<"fdivp %st(0), $op", (DIVR_FPrST0 RST:$op)>;
@@ -1535,9 +1537,9 @@ def : InstAlias<"movq $imm, $reg", (MOV64ri GR64:$reg, i64imm:$imm)>;
 
 // Match 'movq GR64, MMX' as an alias for movd.
 def : InstAlias<"movq $src, $dst",
-                (MMX_MOVD64to64rr VR64:$dst, GR64:$src), 0b0>;
+                (MMX_MOVD64to64rr VR64:$dst, GR64:$src), 0>;
 def : InstAlias<"movq $src, $dst",
-                (MMX_MOVD64from64rr GR64:$dst, VR64:$src), 0b0>;
+                (MMX_MOVD64from64rr GR64:$dst, VR64:$src), 0>;
 
 // movsd with no operands (as opposed to the SSE scalar move of a double) is an
 // alias for movsl. (as in rep; movsd)
diff --git a/test/CodeGen/X86/2006-11-17-IllegalMove.ll b/test/CodeGen/X86/2006-11-17-IllegalMove.ll
index affb7afb1c..091a340b8a 100644
--- a/test/CodeGen/X86/2006-11-17-IllegalMove.ll
+++ b/test/CodeGen/X86/2006-11-17-IllegalMove.ll
@@ -1,6 +1,6 @@
 ; RUN: llc < %s -march=x86-64 > %t
 ; RUN: grep movb %t | count 2
-; RUN: grep {movzb\[wl\]} %t
+; RUN: grep {movzx} %t
 
 
 define void @handle_vector_size_attribute() nounwind {
diff --git a/test/CodeGen/X86/2007-08-10-SignExtSubreg.ll b/test/CodeGen/X86/2007-08-10-SignExtSubreg.ll
index e93092f355..a492f8d37f 100644
--- a/test/CodeGen/X86/2007-08-10-SignExtSubreg.ll
+++ b/test/CodeGen/X86/2007-08-10-SignExtSubreg.ll
@@ -1,4 +1,4 @@
-; RUN: llc < %s -march=x86 | grep {movsbl}
+; RUN: llc < %s -march=x86 | grep {movsx}
 
 @X = global i32 0               ; <i32*> [#uses=1]
 
diff --git a/test/CodeGen/X86/2007-09-27-LDIntrinsics.ll b/test/CodeGen/X86/2007-09-27-LDIntrinsics.ll
index 4d6971586c..b4a986ff77 100644
--- a/test/CodeGen/X86/2007-09-27-LDIntrinsics.ll
+++ b/test/CodeGen/X86/2007-09-27-LDIntrinsics.ll
@@ -22,8 +22,8 @@ entry:
 ; CHECK: bar:
 ; CHECK: fldt 4(%esp)
 ; CHECK-NEXT: fld	%st(0)
-; CHECK-NEXT: fmul	%st(1)
-; CHECK-NEXT: fmulp	%st(1)
+; CHECK-NEXT: fmul	%st(1), %st(0)
+; CHECK-NEXT: fmulp
 ; CHECK-NEXT: ret
 }
 
diff --git a/test/CodeGen/X86/2008-08-17-UComiCodeGenBug.ll b/test/CodeGen/X86/2008-08-17-UComiCodeGenBug.ll
index 32f6ca0ce0..9cc05262e0 100644
--- a/test/CodeGen/X86/2008-08-17-UComiCodeGenBug.ll
+++ b/test/CodeGen/X86/2008-08-17-UComiCodeGenBug.ll
@@ -1,9 +1,10 @@
-; RUN: llc < %s -mtriple=x86_64-apple-darwin | grep movzbl
+; RUN: llc < %s -mtriple=x86_64-apple-darwin | FileCheck %s
 
 define i32 @foo(<4 x float> %a, <4 x float> %b) nounwind {
 entry:
-	tail call i32 @llvm.x86.sse.ucomige.ss( <4 x float> %a, <4 x float> %b ) nounwind readnone
-	ret i32 %0
+; CHECK: movzx
+  tail call i32 @llvm.x86.sse.ucomige.ss( <4 x float> %a, <4 x float> %b ) nounwind readnone
+  ret i32 %0
 }
 
 declare i32 @llvm.x86.sse.ucomige.ss(<4 x float>, <4 x float>) nounwind readnone
diff --git a/test/CodeGen/X86/2009-06-05-VZextByteShort.ll b/test/CodeGen/X86/2009-06-05-VZextByteShort.ll
index 5c514805e4..1d29fea30d 100644
--- a/test/CodeGen/X86/2009-06-05-VZextByteShort.ll
+++ b/test/CodeGen/X86/2009-06-05-VZextByteShort.ll
@@ -1,9 +1,8 @@
-; RUN: llc < %s -march=x86 -mattr=+mmx,+sse2 > %t1
-; RUN: grep movzwl %t1 | count 2
-; RUN: grep movzbl %t1 | count 2
-; RUN: grep movd %t1 | count 4
+; RUN: llc < %s -march=x86 -mattr=+mmx,+sse2 | FileCheck %s
 
 define <4 x i16> @a(i32* %x1) nounwind {
+; CHECK:      movzx
+; CHECK-NEXT: movd
   %x2 = load i32* %x1
   %x3 = lshr i32 %x2, 1
   %x = trunc i32 %x3 to i16
@@ -12,6 +11,8 @@ define <4 x i16> @a(i32* %x1) nounwind {
 }
 
 define <8 x i16> @b(i32* %x1) nounwind {
+; CHECK:      movzx
+; CHECK-NEXT: movd
   %x2 = load i32* %x1
   %x3 = lshr i32 %x2, 1
   %x = trunc i32 %x3 to i16
@@ -20,6 +21,8 @@ define <8 x i16> @b(i32* %x1) nounwind {
 }
 
 define <8 x i8> @c(i32* %x1) nounwind {
+; CHECK:      movzx
+; CHECK-NEXT: movd
   %x2 = load i32* %x1
   %x3 = lshr i32 %x2, 1
   %x = trunc i32 %x3 to i8
@@ -28,6 +31,8 @@ define <8 x i8> @c(i32* %x1) nounwind {
 }
 
 define <16 x i8> @d(i32* %x1) nounwind {
+; CHECK:      movzx
+; CHECK-NEXT: movd
   %x2 = load i32* %x1
   %x3 = lshr i32 %x2, 1
   %x = trunc i32 %x3 to i8
diff --git a/test/CodeGen/X86/anyext.ll b/test/CodeGen/X86/anyext.ll
index 106fe83661..32a48fbedc 100644
--- a/test/CodeGen/X86/anyext.ll
+++ b/test/CodeGen/X86/anyext.ll
@@ -1,8 +1,10 @@
-; RUN: llc < %s -march=x86-64 | grep movzbl | count 2
+; RUN: llc < %s -march=x86-64 | FileCheck %s
 
-; Use movzbl to avoid partial-register updates.
+; Use movzbl (aliased as movzx) to avoid partial-register updates.
 
 define i32 @foo(i32 %p, i8 zeroext %x) nounwind {
+; CHECK: movzx %dil, %eax
+; CHECK: movzx %al, %eax
   %q = trunc i32 %p to i8
   %r = udiv i8 %q, %x
   %s = zext i8 %r to i32
diff --git a/test/CodeGen/X86/avx-intrinsics-x86.ll b/test/CodeGen/X86/avx-intrinsics-x86.ll
index 5201688686..051be285e1 100644
--- a/test/CodeGen/X86/avx-intrinsics-x86.ll
+++ b/test/CodeGen/X86/avx-intrinsics-x86.ll
@@ -75,7 +75,7 @@ declare <2 x double> @llvm.x86.sse2.cmp.sd(<2 x double>, <2 x double>, i8) nounw
 define i32 @test_x86_sse2_comieq_sd(<2 x double> %a0, <2 x double> %a1) {
   ; CHECK: vcomisd
   ; CHECK: sete
-  ; CHECK: movzbl
+  ; CHECK: movzx
   %res = call i32 @llvm.x86.sse2.comieq.sd(<2 x double> %a0, <2 x double> %a1) ; <i32> [#uses=1]
   ret i32 %res
 }
@@ -85,7 +85,7 @@ declare i32 @llvm.x86.sse2.comieq.sd(<2 x double>, <2 x double>) nounwind readno
 define i32 @test_x86_sse2_comige_sd(<2 x double> %a0, <2 x double> %a1) {
   ; CHECK: vcomisd
   ; CHECK: setae
-  ; CHECK: movzbl
+  ; CHECK: movzx
   %res = call i32 @llvm.x86.sse2.comige.sd(<2 x double> %a0, <2 x double> %a1) ; <i32> [#uses=1]
   ret i32 %res
 }
@@ -95,7 +95,7 @@ declare i32 @llvm.x86.sse2.comige.sd(<2 x double>, <2 x double>) nounwind readno
 define i32 @test_x86_sse2_comigt_sd(<2 x double> %a0, <2 x double> %a1) {
   ; CHECK: vcomisd
   ; CHECK: seta
-  ; CHECK: movzbl
+  ; CHECK: movzx
   %res = call i32 @llvm.x86.sse2.comigt.sd(<2 x double> %a0, <2 x double> %a1) ; <i32> [#uses=1]
   ret i32 %res
 }
@@ -105,7 +105,7 @@ declare i32 @llvm.x86.sse2.comigt.sd(<2 x double>, <2 x double>) nounwind readno
 define i32 @test_x86_sse2_comile_sd(<2 x double> %a0, <2 x double> %a1) {
   ; CHECK: vcomisd
   ; CHECK: setbe
-  ; CHECK: movzbl
+  ; CHECK: movzx
   %res = call i32 @llvm.x86.sse2.comile.sd(<2 x double> %a0, <2 x double> %a1) ; <i32> [#uses=1]
   ret i32 %res
 }
@@ -125,7 +125,7 @@ declare i32 @llvm.x86.sse2.comilt.sd(<2 x double>, <2 x double>) nounwind readno
 define i32 @test_x86_sse2_comineq_sd(<2 x double> %a0, <2 x double> %a1) {
   ; CHECK: vcomisd
   ; CHECK: setne
-  ; CHECK: movzbl
+  ; CHECK: movzx
   %res = call i32 @llvm.x86.sse2.comineq.sd(<2 x double> %a0, <2 x double> %a1) ; <i32> [#uses=1]
   ret i32 %res
 }
@@ -786,7 +786,7 @@ declare <2 x double> @llvm.x86.sse2.sub.sd(<2 x double>, <2 x double>) nounwind
 define i32 @test_x86_sse2_ucomieq_sd(<2 x double> %a0, <2 x double> %a1) {
   ; CHECK: vucomisd
   ; CHECK: sete
-  ; CHECK: movzbl
+  ; CHECK: movzx
   %res = call i32 @llvm.x86.sse2.ucomieq.sd(<2 x double> %a0, <2 x double> %a1) ; <i32> [#uses=1]
   ret i32 %res
 }
@@ -796,7 +796,7 @@ declare i32 @llvm.x86.sse2.ucomieq.sd(<2 x double>, <2 x double>) nounwind readn
 define i32 @test_x86_sse2_ucomige_sd(<2 x double> %a0, <2 x double> %a1) {
   ; CHECK: vucomisd
   ; CHECK: setae
-  ; CHECK: movzbl
+  ; CHECK: movzx
   %res = call i32 @llvm.x86.sse2.ucomige.sd(<2 x double> %a0, <2 x double> %a1) ; <i32> [#uses=1]
   ret i32 %res
 }
@@ -806,7 +806,7 @@ declare i32 @llvm.x86.sse2.ucomige.sd(<2 x double>, <2 x double>) nounwind readn
 define i32 @test_x86_sse2_ucomigt_sd(<2 x double> %a0, <2 x double> %a1) {
   ; CHECK: vucomisd
   ; CHECK: seta
-  ; CHECK: movzbl
+  ; CHECK: movzx
   %res = call i32 @llvm.x86.sse2.ucomigt.sd(<2 x double> %a0, <2 x double> %a1) ; <i32> [#uses=1]
   ret i32 %res
 }
@@ -816,7 +816,7 @@ declare i32 @llvm.x86.sse2.ucomigt.sd(<2 x double>, <2 x double>) nounwind readn
 define i32 @test_x86_sse2_ucomile_sd(<2 x double> %a0, <2 x double> %a1) {
   ; CHECK: vucomisd
   ; CHECK: setbe
-  ; CHECK: movzbl
+  ; CHECK: movzx
   %res = call i32 @llvm.x86.sse2.ucomile.sd(<2 x double> %a0, <2 x double> %a1) ; <i32> [#uses=1]
   ret i32 %res
 }
@@ -835,7 +835,7 @@ declare i32 @llvm.x86.sse2.ucomilt.sd(<2 x double>, <2 x double>) nounwind readn
 define i32 @test_x86_sse2_ucomineq_sd(<2 x double> %a0, <2 x double> %a1) {
   ; CHECK: vucomisd
   ; CHECK: setne
-  ; CHECK: movzbl
+  ; CHECK: movzx
   %res = call i32 @llvm.x86.sse2.ucomineq.sd(<2 x double> %a0, <2 x double> %a1) ; <i32> [#uses=1]
   ret i32 %res
 }
@@ -1192,7 +1192,7 @@ declare i32 @llvm.x86.sse41.ptestc(<4 x float>, <4 x float>) nounwind readnone
 define i32 @test_x86_sse41_ptestnzc(<4 x float> %a0, <4 x float> %a1) {
   ; CHECK: vptest 
   ; CHECK: seta
-  ; CHECK: movzbl
+  ; CHECK: movzx
   %res = call i32 @llvm.x86.sse41.ptestnzc(<4 x float> %a0, <4 x float> %a1) ; <i32> [#uses=1]
   ret i32 %res
 }
@@ -1202,7 +1202,7 @@ declare i32 @llvm.x86.sse41.ptestnzc(<4 x float>, <4 x float>) nounwind readnone
 define i32 @test_x86_sse41_ptestz(<4 x float> %a0, <4 x float> %a1) {
   ; CHECK: vptest 
   ; CHECK: sete
-  ; CHECK: movzbl
+  ; CHECK: movzx
   %res = call i32 @llvm.x86.sse41.ptestz(<4 x float> %a0, <4 x float> %a1) ; <i32> [#uses=1]
   ret i32 %res
 }
@@ -1414,7 +1414,7 @@ declare <4 x float> @llvm.x86.sse.cmp.ss(<4 x float>, <4 x float>, i8) nounwind
 define i32 @test_x86_sse_comieq_ss(<4 x float> %a0, <4 x float> %a1) {
   ; CHECK: vcomiss
   ; CHECK: sete
-  ; CHECK: movzbl
+  ; CHECK: movzx
   %res = call i32 @llvm.x86.sse.comieq.ss(<4 x float> %a0, <4 x float> %a1) ; <i32> [#uses=1]
   ret i32 %res
 }
@@ -1424,7 +1424,7 @@ declare i32 @llvm.x86.sse.comieq.ss(<4 x float>, <4 x float>) nounwind readnone
 define i32 @test_x86_sse_comige_ss(<4 x float> %a0, <4 x float> %a1) {
   ; CHECK: vcomiss
   ; CHECK: setae
-  ; CHECK: movzbl
+  ; CHECK: movzx
   %res = call i32 @llvm.x86.sse.comige.ss(<4 x float> %a0, <4 x float> %a1) ; <i32> [#uses=1]
   ret i32 %res
 }
@@ -1434,7 +1434,7 @@ declare i32 @llvm.x86.sse.comige.ss(<4 x float>, <4 x float>) nounwind readnone
 define i32 @test_x86_sse_comigt_ss(<4 x float> %a0, <4 x float> %a1) {
   ; CHECK: vcomiss
   ; CHECK: seta
-  ; CHECK: movzbl
+  ; CHECK: movzx
   %res = call i32 @llvm.x86.sse.comigt.ss(<4 x float> %a0, <4 x float> %a1) ; <i32> [#uses=1]
   ret i32 %res
 }
@@ -1444,7 +1444,7 @@ declare i32 @llvm.x86.sse.comigt.ss(<4 x float>, <4 x float>) nounwind readnone
 define i32 @test_x86_sse_comile_ss(<4 x float> %a0, <4 x float> %a1) {
   ; CHECK: vcomiss
   ; CHECK: setbe
-  ; CHECK: movzbl
+  ; CHECK: movzx
   %res = call i32 @llvm.x86.sse.comile.ss(<4 x float> %a0, <4 x float> %a1) ; <i32> [#uses=1]
   ret i32 %res
 }
@@ -1463,7 +1463,7 @@ declare i32 @llvm.x86.sse.comilt.ss(<4 x float>, <4 x float>) nounwind readnone
 define i32 @test_x86_sse_comineq_ss(<4 x float> %a0, <4 x float> %a1) {
   ; CHECK: vcomiss
   ; CHECK: setne
-  ; CHECK: movzbl
+  ; CHECK: movzx
   %res = call i32 @llvm.x86.sse.comineq.ss(<4 x float> %a0, <4 x float> %a1) ; <i32> [#uses=1]
   ret i32 %res
 }
@@ -1655,7 +1655,7 @@ declare <4 x float> @llvm.x86.sse.sub.ss(<4 x float>, <4 x float>) nounwind read
 define i32 @test_x86_sse_ucomieq_ss(<4 x float> %a0, <4 x float> %a1) {
   ; CHECK: vucomiss
   ; CHECK: sete
-  ; CHECK: movzbl
+  ; CHECK: movzx
   %res = call i32 @llvm.x86.sse.ucomieq.ss(<4 x float> %a0, <4 x float> %a1) ; <i32> [#uses=1]
   ret i32 %res
 }
@@ -1665,7 +1665,7 @@ declare i32 @llvm.x86.sse.ucomieq.ss(<4 x float>, <4 x float>) nounwind readnone
 define i32 @test_x86_sse_ucomige_ss(<4 x float> %a0, <4 x float> %a1) {
   ; CHECK: vucomiss
   ; CHECK: setae
-  ; CHECK: movzbl
+  ; CHECK: movzx
   %res = call i32 @llvm.x86.sse.ucomige.ss(<4 x float> %a0, <4 x float> %a1) ; <i32> [#uses=1]
   ret i32 %res
 }
@@ -1675,7 +1675,7 @@ declare i32 @llvm.x86.sse.ucomige.ss(<4 x float>, <4 x float>) nounwind readnone
 define i32 @test_x86_sse_ucomigt_ss(<4 x float> %a0, <4 x float> %a1) {
   ; CHECK: vucomiss
   ; CHECK: seta
-  ; CHECK: movzbl
+  ; CHECK: movzx
   %res = call i32 @llvm.x86.sse.ucomigt.ss(<4 x float> %a0, <4 x float> %a1) ; <i32> [#uses=1]
   ret i32 %res
 }
@@ -1685,7 +1685,7 @@ declare i32 @llvm.x86.sse.ucomigt.ss(<4 x float>, <4 x float>) nounwind readnone
 define i32 @test_x86_sse_ucomile_ss(<4 x float> %a0, <4 x float> %a1) {
   ; CHECK: vucomiss
   ; CHECK: setbe
-  ; CHECK: movzbl
+  ; CHECK: movzx
   %res = call i32 @llvm.x86.sse.ucomile.ss(<4 x float> %a0, <4 x float> %a1) ; <i32> [#uses=1]
   ret i32 %res
 }
@@ -1704,7 +1704,7 @@ declare i32 @llvm.x86.sse.ucomilt.ss(<4 x float>, <4 x float>) nounwind readnone
 define i32 @test_x86_sse_ucomineq_ss(<4 x float> %a0, <4 x float> %a1) {
   ; CHECK: vucomiss
   ; CHECK: setne
-  ; CHECK: movzbl
+  ; CHECK: movzx
   %res = call i32 @llvm.x86.sse.ucomineq.ss(<4 x float> %a0, <4 x float> %a1) ; <i32> [#uses=1]
   ret i32 %res
 }
@@ -2179,7 +2179,7 @@ declare i32 @llvm.x86.avx.ptestc.256(<4 x i64>, <4 x i64>) nounwind readnone
 define i32 @test_x86_avx_ptestnzc_256(<4 x i64> %a0, <4 x i64> %a1) {
   ; CHECK: vptest
   ; CHECK: seta
-  ; CHECK: movzbl
+  ; CHECK: movzx
   %res = call i32 @llvm.x86.avx.ptestnzc.256(<4 x i64> %a0, <4 x i64> %a1) ; <i32> [#uses=1]
   ret i32 %res
 }
@@ -2189,7 +2189,7 @@ declare i32 @llvm.x86.avx.ptestnzc.256(<4 x i64>, <4 x i64>) nounwind readnone
 define i32 @test_x86_avx_ptestz_256(<4 x i64> %a0, <4 x i64> %a1) {
   ; CHECK: vptest
   ; CHECK: sete
-  ; CHECK: movzbl
+  ; CHECK: movzx
   %res = call i32 @llvm.x86.avx.ptestz.256(<4 x i64> %a0, <4 x i64> %a1) ; <i32> [#uses=1]
   ret i32 %res
 }
@@ -2483,7 +2483,7 @@ declare i32 @llvm.x86.avx.vtestc.ps.256(<8 x float>, <8 x float>) nounwind readn
 define i32 @test_x86_avx_vtestnzc_pd(<2 x double> %a0, <2 x double> %a1) {
   ; CHECK: vtestpd
   ; CHECK: seta
-  ; CHECK: movzbl
+  ; CHECK: movzx
   %res = call i32 @llvm.x86.avx.vtestnzc.pd(<2 x double> %a0, <2 x double> %a1) ; <i32> [#uses=1]
   ret i32 %res
 }
@@ -2493,7 +2493,7 @@ declare i32 @llvm.x86.avx.vtestnzc.pd(<2 x double>, <2 x double>) nounwind readn
 define i32 @test_x86_avx_vtestnzc_pd_256(<4 x double> %a0, <4 x double> %a1) {
   ; CHECK: vtestpd
   ; CHECK: seta
-  ; CHECK: movzbl
+  ; CHECK: movzx
   %res = call i32 @llvm.x86.avx.vtestnzc.pd.256(<4 x double> %a0, <4 x double> %a1) ; <i32> [#uses=1]
   ret i32 %res
 }
@@ -2503,7 +2503,7 @@ declare i32 @llvm.x86.avx.vtestnzc.pd.256(<4 x double>, <4 x double>) nounwind r
 define i32 @test_x86_avx_vtestnzc_ps(<4 x float> %a0, <4 x float> %a1) {
   ; CHECK: vtestps
   ; CHECK: seta
-  ; CHECK: movzbl
+  ; CHECK: movzx
   %res = call i32 @llvm.x86.avx.vtestnzc.ps(<4 x float> %a0, <4 x float> %a1) ; <i32> [#uses=1]
   ret i32 %res
 }
@@ -2513,7 +2513,7 @@ declare i32 @llvm.x86.avx.vtestnzc.ps(<4 x float>, <4 x float>) nounwind readnon
 define i32 @test_x86_avx_vtestnzc_ps_256(<8 x float> %a0, <8 x float> %a1) {
   ; CHECK: vtestps
   ; CHECK: seta
-  ; CHECK: movzbl
+  ; CHECK: movzx
   %res = call i32 @llvm.x86.avx.vtestnzc.ps.256(<8 x float> %a0, <8 x float> %a1) ; <i32> [#uses=1]
   ret i32 %res
 }
@@ -2523,7 +2523,7 @@ declare i32 @llvm.x86.avx.vtestnzc.ps.256(<8 x float>, <8 x float>) nounwind rea
 define i32 @test_x86_avx_vtestz_pd(<2 x double> %a0, <2 x double> %a1) {
   ; CHECK: vtestpd
   ; CHECK: sete
-  ; CHECK: movzbl
+  ; CHECK: movzx
   %res = call i32 @llvm.x86.avx.vtestz.pd(<2 x double> %a0, <2 x double> %a1) ; <i32> [#uses=1]
   ret i32 %res
 }
@@ -2533,7 +2533,7 @@ declare i32 @llvm.x86.avx.vtestz.pd(<2 x double>, <2 x double>) nounwind readnon
 define i32 @test_x86_avx_vtestz_pd_256(<4 x double> %a0, <4 x double> %a1) {
   ; CHECK: vtestpd
   ; CHECK: sete
-  ; CHECK: movzbl
+  ; CHECK: movzx
   %res = call i32 @llvm.x86.avx.vtestz.pd.256(<4 x double> %a0, <4 x double> %a1) ; <i32> [#uses=1]
   ret i32 %res
 }
@@ -2543,7 +2543,7 @@ declare i32 @llvm.x86.avx.vtestz.pd.256(<4 x double>, <4 x double>) nounwind rea
 define i32 @test_x86_avx_vtestz_ps(<4 x float> %a0, <4 x float> %a1) {
   ; CHECK: vtestps
   ; CHECK: sete
-  ; CHECK: movzbl
+  ; CHECK: movzx
   %res = call i32 @llvm.x86.avx.vtestz.ps(<4 x float> %a0, <4 x float> %a1) ; <i32> [#uses=1]
   ret i32 %res
 }
@@ -2553,7 +2553,7 @@ declare i32 @llvm.x86.avx.vtestz.ps(<4 x float>, <4 x float>) nounwind readnone
 define i32 @test_x86_avx_vtestz_ps_256(<8 x float> %a0, <8 x float> %a1) {
   ; CHECK: vtestps
   ; CHECK: sete
-  ; CHECK: movzbl
+  ; CHECK: movzx
   %res = call i32 @llvm.x86.avx.vtestz.ps.256(<8 x float> %a0, <8 x float> %a1) ; <i32> [#uses=1]
   ret i32 %res
 }
diff --git a/test/CodeGen/X86/bool-zext.ll b/test/CodeGen/X86/bool-zext.ll
index d2c30c64f2..91e36b0d53 100644
--- a/test/CodeGen/X86/bool-zext.ll
+++ b/test/CodeGen/X86/bool-zext.ll
@@ -1,7 +1,7 @@
 ; RUN: llc < %s -march=x86-64 | FileCheck %s
 
 ; CHECK: @bar1
-; CHECK: movzbl
+; CHECK: movzx
 ; CHECK: callq
 define void @bar1(i1 zeroext %v1) nounwind ssp {
 entry:
@@ -11,7 +11,7 @@ entry:
 }
 
 ; CHECK: @bar2
-; CHECK-NOT: movzbl
+; CHECK-NOT: movzx
 ; CHECK: callq
 define void @bar2(i8 zeroext %v1) nounwind ssp {
 entry:
@@ -22,7 +22,7 @@ entry:
 
 ; CHECK: @bar3
 ; CHECK: callq
-; CHECK-NOT: movzbl
+; CHECK-NOT: movzx
 ; CHECK-NOT: and
 ; CHECK: ret
 define zeroext i1 @bar3() nounwind ssp {
diff --git a/test/CodeGen/X86/cmov.ll b/test/CodeGen/X86/cmov.ll
index 39d9d1e9ec..2e71ef20cf 100644
--- a/test/CodeGen/X86/cmov.ll
+++ b/test/CodeGen/X86/cmov.ll
@@ -121,7 +121,7 @@ define i32 @test5(i32* nocapture %P) nounwind readonly {
 entry:
 ; CHECK: test5:
 ; CHECK: 	setg	%al
-; CHECK:	movzbl	%al, %eax
+; CHECK:	movzx	%al, %eax
 ; CHECK:	orl	$-2, %eax
 ; CHECK:	ret
 
@@ -135,7 +135,7 @@ define i32 @test6(i32* nocapture %P) nounwind readonly {
 entry:
 ; CHECK: test6:
 ; CHECK: 	setl	%al
-; CHECK:	movzbl	%al, %eax
+; CHECK:	movzx	%al, %eax
 ; CHECK:	leal	4(%rax,%rax,8), %eax
 ; CHECK:        ret
 	%0 = load i32* %P, align 4		; <i32> [#uses=1]
diff --git a/test/CodeGen/X86/cmp.ll b/test/CodeGen/X86/cmp.ll
index ef5e353e9f..6e88fc1cbe 100644
--- a/test/CodeGen/X86/cmp.ll
+++ b/test/CodeGen/X86/cmp.ll
@@ -38,7 +38,7 @@ define i64 @test3(i64 %x) nounwind {
 ; CHECK: test3:
 ; CHECK: 	testq	%rdi, %rdi
 ; CHECK: 	sete	%al
-; CHECK: 	movzbl	%al, %eax
+; CHECK: 	movzx	%al, %eax
 ; CHECK: 	ret
 }
 
@@ -49,7 +49,7 @@ define i64 @test4(i64 %x) nounwind {
 ; CHECK: test4:
 ; CHECK: 	testq	%rdi, %rdi
 ; CHECK: 	setle	%al
-; CHECK: 	movzbl	%al, %eax
+; CHECK: 	movzx	%al, %eax
 ; CHECK: 	ret
 }
 
diff --git a/test/CodeGen/X86/fast-isel-gep.ll b/test/CodeGen/X86/fast-isel-gep.ll
index 48abfd0f26..ebefd74f19 100644
--- a/test/CodeGen/X86/fast-isel-gep.ll
+++ b/test/CodeGen/X86/fast-isel-gep.ll
@@ -14,7 +14,7 @@ define i32 @test1(i32 %t3, i32* %t1) nounwind {
 ; X32:  	ret
 
 ; X64: test1:
-; X64:  	movslq	%e[[A0:di|cx]], %rax
+; X64:  	movsx	%e[[A0:di|cx]], %rax
 ; X64:  	movl	(%r[[A1:si|dx]],%rax,4), %eax
 ; X64:  	ret
 
@@ -81,7 +81,7 @@ define i64 @test5(i8* %A, i32 %I, i64 %B) nounwind {
   %v11 = add i64 %B, %v10
   ret i64 %v11
 ; X64: test5:
-; X64: movslq	%e[[A1]], %rax
+; X64: movsx	%e[[A1]], %rax
 ; X64-NEXT: movq	(%r[[A0]],%rax), %rax
 ; X64-NEXT: addq	%{{rdx|r8}}, %rax
 ; X64-NEXT: ret
diff --git a/test/CodeGen/X86/fp-stack-compare.ll b/test/CodeGen/X86/fp-stack-compare.ll
index b216914d23..f3998b67f6 100644
--- a/test/CodeGen/X86/fp-stack-compare.ll
+++ b/test/CodeGen/X86/fp-stack-compare.ll
@@ -1,11 +1,11 @@
-; RUN: llc < %s -march=x86 -mcpu=i386 | grep {fucompi.*st.\[12\]}
+; RUN: llc < %s -march=x86 -mcpu=i386 | FileCheck %s
 ; PR1012
 
 define float @foo(float* %col.2.0) {
-        %tmp = load float* %col.2.0             ; <float> [#uses=3]
-        %tmp16 = fcmp olt float %tmp, 0.000000e+00              ; <i1> [#uses=1]
-        %tmp20 = fsub float -0.000000e+00, %tmp          ; <float> [#uses=1]
-        %iftmp.2.0 = select i1 %tmp16, float %tmp20, float %tmp         ; <float> [#uses=1]
-        ret float %iftmp.2.0
+; CHECK: fucompi
+  %tmp = load float* %col.2.0
+  %tmp16 = fcmp olt float %tmp, 0.000000e+00
+  %tmp20 = fsub float -0.000000e+00, %tmp
+  %iftmp.2.0 = select i1 %tmp16, float %tmp20, float %tmp
+  ret float %iftmp.2.0
 }
-
diff --git a/test/CodeGen/X86/h-register-addressing-32.ll b/test/CodeGen/X86/h-register-addressing-32.ll
index 76ffd66524..e312bcc2e2 100644
--- a/test/CodeGen/X86/h-register-addressing-32.ll
+++ b/test/CodeGen/X86/h-register-addressing-32.ll
@@ -1,4 +1,4 @@
-; RUN: llc < %s -march=x86 | grep {movzbl	%\[abcd\]h,} | count 7
+; RUN: llc < %s -march=x86 | grep {movzx	%\[abcd\]h,} | count 7
 
 ; Use h-register extract and zero-extend.
 
diff --git a/test/CodeGen/X86/h-registers-0.ll b/test/CodeGen/X86/h-registers-0.ll
index cdc75af92e..b477d7759c 100644
--- a/test/CodeGen/X86/h-registers-0.ll
+++ b/test/CodeGen/X86/h-registers-0.ll
@@ -70,7 +70,7 @@ define i64 @qux64(i64 inreg %x) nounwind {
 ; WIN64:  movzbl %ch, %eax
 
 ; X86-32: qux64:
-; X86-32: movzbl %ah, %eax
+; X86-32: movzx %ah, %eax
   %t0 = lshr i64 %x, 8
   %t1 = and i64 %t0, 255
   ret i64 %t1
@@ -85,7 +85,7 @@ define i32 @qux32(i32 inreg %x) nounwind {
 ; WIN64:  movzbl %ch, %eax
 
 ; X86-32: qux32:
-; X86-32: movzbl %ah, %eax
+; X86-32: movzx %ah, %eax
   %t0 = lshr i32 %x, 8
   %t1 = and i32 %t0, 255
   ret i32 %t1
@@ -100,7 +100,7 @@ define i16 @qux16(i16 inreg %x) nounwind {
 ; WIN64:  movzbl %ch, %eax
 
 ; X86-32: qux16:
-; X86-32: movzbl %ah, %eax
+; X86-32: movzx %ah, %eax
   %t0 = lshr i16 %x, 8
   ret i16 %t0
 }
diff --git a/test/CodeGen/X86/h-registers-2.ll b/test/CodeGen/X86/h-registers-2.ll
index 16e13f8396..702e90bf5e 100644
--- a/test/CodeGen/X86/h-registers-2.ll
+++ b/test/CodeGen/X86/h-registers-2.ll
@@ -1,5 +1,5 @@
 ; RUN: llc < %s -march=x86 > %t
-; RUN: grep {movzbl	%\[abcd\]h,} %t | count 1
+; RUN: grep {movzx	%\[abcd\]h,} %t | count 1
 ; RUN: grep {shll	\$3,} %t | count 1
 
 ; Use an h register, but don't omit the explicit shift for
diff --git a/test/CodeGen/X86/inline-asm-mrv.ll b/test/CodeGen/X86/inline-asm-mrv.ll
index 78d7e776cf..8442d7a7c4 100644
--- a/test/CodeGen/X86/inline-asm-mrv.ll
+++ b/test/CodeGen/X86/inline-asm-mrv.ll