rip out a ton of intrinsic modernization logic from AutoUpgrade.cpp, which is

for pre-2.9 bitcode files.  We keep x86 unaligned loads, movnt, crc32, and the
target indep prefetch change.

As usual, updating the testsuite is a PITA.



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@133337 91177308-0d34-0410-b5e6-96231b3b80d8

37 files changed, 588 insertions, 795 deletions

diff --git a/test/Transforms/DeadStoreElimination/crash.ll b/test/Transforms/DeadStoreElimination/crash.ll
index bb279cdb97..148695fced 100644
--- a/test/Transforms/DeadStoreElimination/crash.ll
+++ b/test/Transforms/DeadStoreElimination/crash.ll
@@ -36,11 +36,11 @@ bb14:                                             ; preds = %bb4
   %6 = getelementptr inbounds i16* %2, i64 undef  ; <i16*> [#uses=1]
   store i16 undef, i16* %6, align 2
   %7 = getelementptr inbounds i8* %5, i64 undef   ; <i8*> [#uses=1]
-  call void @llvm.memcpy.i64(i8* %7, i8* undef, i64 undef, i32 1) nounwind
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %7, i8* undef, i64 undef, i32 1, i1 false)
   unreachable
 }
 
-declare void @llvm.memcpy.i64(i8* nocapture, i8* nocapture, i64, i32) nounwind
+declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture, i64, i32, i1) nounwind
 
 
 ; rdar://7635088
diff --git a/test/Transforms/DeadStoreElimination/lifetime.ll b/test/Transforms/DeadStoreElimination/lifetime.ll
index 2b5cc5aedb..678565315e 100644
--- a/test/Transforms/DeadStoreElimination/lifetime.ll
+++ b/test/Transforms/DeadStoreElimination/lifetime.ll
@@ -4,7 +4,7 @@ target datalayout = "E-p:64:64:64-a0:0:8-f32:32:32-f64:64:64-i1:8:8-i8:8:8-i16:1
 
 declare void @llvm.lifetime.start(i64, i8* nocapture) nounwind
 declare void @llvm.lifetime.end(i64, i8* nocapture) nounwind
-declare void @llvm.memset.i8(i8*, i8, i8, i32)
+declare void @llvm.memset.p0i8.i8(i8* nocapture, i8, i8, i32, i1) nounwind
 
 define void @test1() {
 ; CHECK: @test1
@@ -14,7 +14,7 @@ define void @test1() {
   call void @llvm.lifetime.end(i64 1, i8* %A)
 ; CHECK: lifetime.end
 
-  call void @llvm.memset.i8(i8* %A, i8 0, i8 -1, i32 0)
+  call void @llvm.memset.p0i8.i8(i8* %A, i8 0, i8 -1, i32 0, i1 false)
 ; CHECK-NOT: memset
 
   ret void
diff --git a/test/Transforms/DeadStoreElimination/memintrinsics.ll b/test/Transforms/DeadStoreElimination/memintrinsics.ll
index e31e9fa3ca..d5c5365771 100644
--- a/test/Transforms/DeadStoreElimination/memintrinsics.ll
+++ b/test/Transforms/DeadStoreElimination/memintrinsics.ll
@@ -1,8 +1,8 @@
 ; RUN: opt -S -dse < %s | FileCheck %s
 
-declare void @llvm.memcpy.i8(i8*, i8*, i8, i32)
-declare void @llvm.memmove.i8(i8*, i8*, i8, i32)
-declare void @llvm.memset.i8(i8*, i8, i8, i32)
+declare void @llvm.memcpy.p0i8.p0i8.i8(i8* nocapture, i8* nocapture, i8, i32, i1) nounwind
+declare void @llvm.memmove.p0i8.p0i8.i8(i8* nocapture, i8* nocapture, i8, i32, i1) nounwind
+declare void @llvm.memset.p0i8.i8(i8* nocapture, i8, i8, i32, i1) nounwind
 
 define void @test1() {
 ; CHECK: @test1
@@ -12,7 +12,7 @@ define void @test1() {
   store i8 0, i8* %A  ;; Written to by memcpy
 ; CHECK-NOT: store
 
-  call void @llvm.memcpy.i8(i8* %A, i8* %B, i8 -1, i32 0)
+  call void @llvm.memcpy.p0i8.p0i8.i8(i8* %A, i8* %B, i8 -1, i32 0, i1 false)
 
   ret void
 ; CHECK: ret void
@@ -26,7 +26,7 @@ define void @test2() {
   store i8 0, i8* %A  ;; Written to by memmove
 ; CHECK-NOT: store
 
-  call void @llvm.memmove.i8(i8* %A, i8* %B, i8 -1, i32 0)
+  call void @llvm.memmove.p0i8.p0i8.i8(i8* %A, i8* %B, i8 -1, i32 0, i1 false)
 
   ret void
 ; CHECK: ret void
@@ -40,7 +40,7 @@ define void @test3() {
   store i8 0, i8* %A  ;; Written to by memset
 ; CHECK-NOT: store
 
-  call void @llvm.memset.i8(i8* %A, i8 0, i8 -1, i32 0)
+  call void @llvm.memset.p0i8.i8(i8* %A, i8 0, i8 -1, i32 0, i1 false)
 
   ret void
 ; CHECK: ret void
diff --git a/test/Transforms/DeadStoreElimination/simple.ll b/test/Transforms/DeadStoreElimination/simple.ll
index 23576dadcf..5f143fcd1e 100644
--- a/test/Transforms/DeadStoreElimination/simple.ll
+++ b/test/Transforms/DeadStoreElimination/simple.ll
@@ -3,8 +3,6 @@ target datalayout = "E-p:64:64:64-a0:0:8-f32:32:32-f64:64:64-i1:8:8-i8:8:8-i16:1
 
 declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i32, i1) nounwind
 declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture, i64, i32, i1) nounwind
-declare void @llvm.memset.i64(i8*, i8, i64, i32)
-declare void @llvm.memcpy.i64(i8*, i8*, i64, i32)
 declare i8* @llvm.init.trampoline(i8*, i8*, i8*)
 
 define void @test1(i32* %Q, i32* %P) {
@@ -65,7 +63,7 @@ define void @test5(i32* %Q) {
 ; alias).
 define void @test6(i32 *%p, i8 *%q) {
   store i32 10, i32* %p, align 4       ;; dead.
-  call void @llvm.memset.i64(i8* %q, i8 42, i64 900, i32 1)
+  call void @llvm.memset.p0i8.i64(i8* %q, i8 42, i64 900, i32 1, i1 false)
   store i32 30, i32* %p, align 4
   ret void
 ; CHECK: @test6
@@ -76,7 +74,7 @@ define void @test6(i32 *%p, i8 *%q) {
 ; alias).
 define void @test7(i32 *%p, i8 *%q, i8* noalias %r) {
   store i32 10, i32* %p, align 4       ;; dead.
-  call void @llvm.memcpy.i64(i8* %q, i8* %r, i64 900, i32 1)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %q, i8* %r, i64 900, i32 1, i1 false)
   store i32 30, i32* %p, align 4
   ret void
 ; CHECK: @test7
@@ -184,8 +182,8 @@ define void @test14(i32* %Q) {
 
 ;; Fully dead overwrite of memcpy.
 define void @test15(i8* %P, i8* %Q) nounwind ssp {
-  tail call void @llvm.memcpy.i64(i8* %P, i8* %Q, i64 12, i32 1)
-  tail call void @llvm.memcpy.i64(i8* %P, i8* %Q, i64 12, i32 1)
+  tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %P, i8* %Q, i64 12, i32 1, i1 false)
+  tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %P, i8* %Q, i64 12, i32 1, i1 false)
   ret void
 ; CHECK: @test15
 ; CHECK-NEXT: call void @llvm.memcpy
@@ -194,8 +192,8 @@ define void @test15(i8* %P, i8* %Q) nounwind ssp {
 
 ;; Full overwrite of smaller memcpy.
 define void @test16(i8* %P, i8* %Q) nounwind ssp {
-  tail call void @llvm.memcpy.i64(i8* %P, i8* %Q, i64 8, i32 1)
-  tail call void @llvm.memcpy.i64(i8* %P, i8* %Q, i64 12, i32 1)
+  tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %P, i8* %Q, i64 8, i32 1, i1 false)
+  tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %P, i8* %Q, i64 12, i32 1, i1 false)
   ret void
 ; CHECK: @test16
 ; CHECK-NEXT: call void @llvm.memcpy
@@ -204,8 +202,8 @@ define void @test16(i8* %P, i8* %Q) nounwind ssp {
 
 ;; Overwrite of memset by memcpy.
 define void @test17(i8* %P, i8* noalias %Q) nounwind ssp {
-  tail call void @llvm.memset.i64(i8* %P, i8 42, i64 8, i32 1)
-  tail call void @llvm.memcpy.i64(i8* %P, i8* %Q, i64 12, i32 1)
+  tail call void @llvm.memset.p0i8.i64(i8* %P, i8 42, i64 8, i32 1, i1 false)
+  tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %P, i8* %Q, i64 12, i32 1, i1 false)
   ret void
 ; CHECK: @test17
 ; CHECK-NEXT: call void @llvm.memcpy
@@ -228,8 +226,8 @@ define void @test17v(i8* %P, i8* %Q) nounwind ssp {
 ; A = B
 ; A = A
 define void @test18(i8* %P, i8* %Q, i8* %R) nounwind ssp {
-  tail call void @llvm.memcpy.i64(i8* %P, i8* %Q, i64 12, i32 1)
-  tail call void @llvm.memcpy.i64(i8* %P, i8* %R, i64 12, i32 1)
+  tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %P, i8* %Q, i64 12, i32 1, i1 false)
+  tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %P, i8* %R, i64 12, i32 1, i1 false)
   ret void
 ; CHECK: @test18
 ; CHECK-NEXT: call void @llvm.memcpy
diff --git a/test/Transforms/FunctionAttrs/2008-10-04-LocalMemory.ll b/test/Transforms/FunctionAttrs/2008-10-04-LocalMemory.ll
deleted file mode 100644
index faac1184b6..0000000000
--- a/test/Transforms/FunctionAttrs/2008-10-04-LocalMemory.ll
+++ /dev/null
@@ -1,64 +0,0 @@
-; RUN: opt < %s -basicaa -functionattrs -S | FileCheck %s
-
-%struct.X = type { i32*, i32* }
-
-declare i32 @g(i32*) readnone
-
-define i32 @f() {
-; CHECK: @f() readnone
-	%x = alloca i32		; <i32*> [#uses=2]
-	store i32 0, i32* %x
-	%y = call i32 @g(i32* %x)		; <i32> [#uses=1]
-	ret i32 %y
-}
-
-define i32 @foo() nounwind {
-; CHECK: @foo() nounwind readonly
-entry:
-  %y = alloca %struct.X                           ; <%struct.X*> [#uses=2]
-  %x = alloca %struct.X                           ; <%struct.X*> [#uses=2]
-  %j = alloca i32                                 ; <i32*> [#uses=2]
-  %i = alloca i32                                 ; <i32*> [#uses=2]
-  %"alloca point" = bitcast i32 0 to i32          ; <i32> [#uses=0]
-  store i32 0, i32* %i, align 4
-  store i32 1, i32* %j, align 4
-  %0 = getelementptr inbounds %struct.X* %y, i32 0, i32 0 ; <i32**> [#uses=1]
-  store i32* %i, i32** %0, align 8
-  %1 = getelementptr inbounds %struct.X* %x, i32 0, i32 1 ; <i32**> [#uses=1]
-  store i32* %j, i32** %1, align 8
-  %x1 = bitcast %struct.X* %x to i8*              ; <i8*> [#uses=2]
-  %y2 = bitcast %struct.X* %y to i8*              ; <i8*> [#uses=1]
-  call void @llvm.memcpy.i64(i8* %x1, i8* %y2, i64 8, i32 1)
-  %2 = bitcast i8* %x1 to i32**                   ; <i32**> [#uses=1]
-  %3 = load i32** %2, align 8                     ; <i32*> [#uses=1]
-  %4 = load i32* %3, align 4                      ; <i32> [#uses=1]
-  br label %return
-
-return:                                           ; preds = %entry
-  ret i32 %4
-}
-
-define i32 @t(i32 %a, i32 %b, i32 %c) nounwind {
-; CHECK: @t(i32 %a, i32 %b, i32 %c) nounwind readnone
-entry:
-  %a.addr = alloca i32                            ; <i32*> [#uses=3]
-  %c.addr = alloca i32                            ; <i32*> [#uses=2]
-  store i32 %a, i32* %a.addr
-  store i32 %c, i32* %c.addr
-  %tmp = load i32* %a.addr                        ; <i32> [#uses=1]
-  %tobool = icmp ne i32 %tmp, 0                   ; <i1> [#uses=1]
-  br i1 %tobool, label %if.then, label %if.else
-
-if.then:                                          ; preds = %entry
-  br label %if.end
-
-if.else:                                          ; preds = %entry
-  br label %if.end
-
-if.end:                                           ; preds = %if.else, %if.then
-  %p.0 = phi i32* [ %a.addr, %if.then ], [ %c.addr, %if.else ] ; <i32*> [#uses=1]
-  %tmp2 = load i32* %p.0                          ; <i32> [#uses=1]
-  ret i32 %tmp2
-}
-
-declare void @llvm.memcpy.i64(i8* nocapture, i8* nocapture, i64, i32) nounwind
diff --git a/test/Transforms/GVN/2008-02-24-NonDominatedMemcpy.ll b/test/Transforms/GVN/2008-02-24-NonDominatedMemcpy.ll
deleted file mode 100644
index 9a75e1ad97..0000000000
--- a/test/Transforms/GVN/2008-02-24-NonDominatedMemcpy.ll
+++ /dev/null
@@ -1,25 +0,0 @@
-; RUN: opt < %s -gvn -dse -S | grep {call.*memcpy} | count 1
-
-target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128"
-target triple = "x86_64-apple-darwin8"
-	%struct.ggFrame3 = type { %struct.ggPoint3, %struct.ggONB3 }
-	%struct.ggHMatrix3 = type { [4 x [4 x double]] }
-	%struct.ggONB3 = type { %struct.ggPoint3, %struct.ggPoint3, %struct.ggPoint3 }
-	%struct.ggPoint3 = type { [3 x double] }
-	%struct.ggQuaternion = type { [4 x double], i32, %struct.ggHMatrix3 }
-
-declare void @llvm.memcpy.i64(i8*, i8*, i64, i32) nounwind 
-
-define void @_Z10ggCRSplineRK8ggFrame3S1_S1_S1_d(%struct.ggFrame3* noalias sret  %agg.result, %struct.ggFrame3* %f0, %struct.ggFrame3* %f1, %struct.ggFrame3* %f2, %struct.ggFrame3* %f3, double %t) nounwind  {
-entry:
-	%qresult = alloca %struct.ggQuaternion		; <%struct.ggQuaternion*> [#uses=1]
-	%tmp = alloca %struct.ggONB3		; <%struct.ggONB3*> [#uses=2]
-	call void @_ZN12ggQuaternion7getONB3Ev( %struct.ggONB3* noalias sret  %tmp, %struct.ggQuaternion* %qresult ) nounwind 
-	%tmp1.i = getelementptr %struct.ggFrame3* %agg.result, i32 0, i32 1		; <%struct.ggONB3*> [#uses=1]
-	%tmp13.i = bitcast %struct.ggONB3* %tmp1.i to i8*		; <i8*> [#uses=1]
-	%tmp24.i = bitcast %struct.ggONB3* %tmp to i8*		; <i8*> [#uses=1]
-	call void @llvm.memcpy.i64( i8* %tmp13.i, i8* %tmp24.i, i64 72, i32 8 ) nounwind 
-	ret void
-}
-
-declare void @_ZN12ggQuaternion7getONB3Ev(%struct.ggONB3* noalias sret , %struct.ggQuaternion*) nounwind 
diff --git a/test/Transforms/GVN/2008-02-26-MemCpySize.ll b/test/Transforms/GVN/2008-02-26-MemCpySize.ll
deleted file mode 100644
index 6ed8a76c0d..0000000000
--- a/test/Transforms/GVN/2008-02-26-MemCpySize.ll
+++ /dev/null
@@ -1,46 +0,0 @@
-; RUN: opt < %s -gvn -dse -S | grep {call.*memcpy.*cell} | count 2
-; PR2099
-
-target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f32:32:32-f64:32:64-v64:64:64-v128:128:128-a0:0:64-f80:128:128"
-target triple = "i686-apple-darwin9"
-	%struct.s = type { [11 x i8], i32 }
-@.str = internal constant [11 x i8] c"0123456789\00"		; <[11 x i8]*> [#uses=1]
-@cell = weak global %struct.s zeroinitializer		; <%struct.s*> [#uses=2]
-
-declare i32 @check(%struct.s* byval  %p) nounwind
-
-declare i32 @strcmp(i8*, i8*) nounwind readonly 
-
-define i32 @main() noreturn nounwind  {
-entry:
-	%p = alloca %struct.s, align 8		; <%struct.s*> [#uses=2]
-	store i32 99, i32* getelementptr (%struct.s* @cell, i32 0, i32 1), align 4
-	call void @llvm.memcpy.i32( i8* getelementptr (%struct.s* @cell, i32 0, i32 0, i32 0), i8* getelementptr ([11 x i8]* @.str, i32 0, i32 0), i32 11, i32 1 )
-	%tmp = getelementptr %struct.s* %p, i32 0, i32 0, i32 0		; <i8*> [#uses=2]
-	call void @llvm.memcpy.i64( i8* %tmp, i8* getelementptr (%struct.s* @cell, i32 0, i32 0, i32 0), i64 16, i32 8 )
-	%tmp1.i = getelementptr %struct.s* %p, i32 0, i32 1		; <i32*> [#uses=1]
-	%tmp2.i = load i32* %tmp1.i, align 4		; <i32> [#uses=1]
-	%tmp3.i = icmp eq i32 %tmp2.i, 99		; <i1> [#uses=1]
-	br i1 %tmp3.i, label %bb5.i, label %bb
-
-bb5.i:		; preds = %entry
-	%tmp91.i = call i32 @strcmp( i8* %tmp, i8* getelementptr ([11 x i8]* @.str, i32 0, i32 0) ) nounwind readonly 		; <i32> [#uses=1]
-	%tmp53 = icmp eq i32 %tmp91.i, 0		; <i1> [#uses=1]
-	br i1 %tmp53, label %bb7, label %bb
-
-bb:		; preds = %bb5.i, %entry
-	call void @abort( ) noreturn nounwind 
-	unreachable
-
-bb7:		; preds = %bb5.i
-	call void @exit( i32 0 ) noreturn nounwind 
-	unreachable
-}
-
-declare void @llvm.memcpy.i32(i8*, i8*, i32, i32) nounwind 
-
-declare void @abort() noreturn nounwind 
-
-declare void @exit(i32) noreturn nounwind 
-
-declare void @llvm.memcpy.i64(i8*, i8*, i64, i32) nounwind 
diff --git a/test/Transforms/GVN/nonescaping-malloc.ll b/test/Transforms/GVN/nonescaping-malloc.ll
index 1d50205c68..dba9d81405 100644
--- a/test/Transforms/GVN/nonescaping-malloc.ll
+++ b/test/Transforms/GVN/nonescaping-malloc.ll
@@ -20,13 +20,13 @@ target triple = "x86_64-apple-darwin10.0"
 
 define %"struct.llvm::StringMapEntry<void*>"* @_Z3fooRN4llvm9StringMapIPvNS_15MallocAllocatorEEEPKc(%"struct.llvm::StringMap<void*,llvm::MallocAllocator>"* %X, i8* %P) ssp {
 entry:
-  %tmp = alloca %"struct.llvm::StringRef", align 8 ; <%"struct.llvm::StringRef"*> [#uses=3]
-  %tmp.i = getelementptr inbounds %"struct.llvm::StringRef"* %tmp, i64 0, i32 0 ; <i8**> [#uses=1]
+  %tmp = alloca %"struct.llvm::StringRef", align 8
+  %tmp.i = getelementptr inbounds %"struct.llvm::StringRef"* %tmp, i64 0, i32 0
   store i8* %P, i8** %tmp.i, align 8
-  %tmp1.i = call i64 @strlen(i8* %P) nounwind readonly ; <i64> [#uses=1]
-  %tmp2.i = getelementptr inbounds %"struct.llvm::StringRef"* %tmp, i64 0, i32 1 ; <i64*> [#uses=1]
+  %tmp1.i = call i64 @strlen(i8* %P) nounwind readonly
+  %tmp2.i = getelementptr inbounds %"struct.llvm::StringRef"* %tmp, i64 0, i32 1
   store i64 %tmp1.i, i64* %tmp2.i, align 8
-  %tmp1 = call %"struct.llvm::StringMapEntry<void*>"* @_ZN4llvm9StringMapIPvNS_15MallocAllocatorEE16GetOrCreateValueERKNS_9StringRefE(%"struct.llvm::StringMap<void*,llvm::MallocAllocator>"* %X, %"struct.llvm::StringRef"* %tmp) ssp ; <%"struct.llvm::StringMapEntry<void*>"*> [#uses=1]
+  %tmp1 = call %"struct.llvm::StringMapEntry<void*>"* @_ZN4llvm9StringMapIPvNS_15MallocAllocatorEE16GetOrCreateValueERKNS_9StringRefE(%"struct.llvm::StringMap<void*,llvm::MallocAllocator>"* %X, %"struct.llvm::StringRef"* %tmp) ssp
   ret %"struct.llvm::StringMapEntry<void*>"* %tmp1
 }
 
@@ -34,75 +34,75 @@ declare i64 @strlen(i8* nocapture) nounwind readonly
 
 declare noalias i8* @malloc(i64) nounwind
 
-declare void @llvm.memcpy.i64(i8* nocapture, i8* nocapture, i64, i32) nounwind
-
 declare i32 @_ZN4llvm13StringMapImpl15LookupBucketForENS_9StringRefE(%"struct.llvm::StringMapImpl"*, i64, i64)
 
 define linkonce_odr %"struct.llvm::StringMapEntry<void*>"* @_ZN4llvm9StringMapIPvNS_15MallocAllocatorEE16GetOrCreateValueERKNS_9StringRefE(%"struct.llvm::StringMap<void*,llvm::MallocAllocator>"* %this, %"struct.llvm::StringRef"* nocapture %Key) ssp align 2 {
 entry:
-  %elt = bitcast %"struct.llvm::StringRef"* %Key to i64* ; <i64*> [#uses=1]
-  %val = load i64* %elt                           ; <i64> [#uses=3]
-  %tmp = getelementptr inbounds %"struct.llvm::StringRef"* %Key, i64 0, i32 1 ; <i64*> [#uses=1]
-  %val2 = load i64* %tmp                          ; <i64> [#uses=2]
-  %tmp2.i = getelementptr inbounds %"struct.llvm::StringMap<void*,llvm::MallocAllocator>"* %this, i64 0, i32 0 ; <%"struct.llvm::StringMapImpl"*> [#uses=1]
-  %tmp3.i = tail call i32 @_ZN4llvm13StringMapImpl15LookupBucketForENS_9StringRefE(%"struct.llvm::StringMapImpl"* %tmp2.i, i64 %val, i64 %val2) ; <i32> [#uses=1]
-  %tmp4.i = getelementptr inbounds %"struct.llvm::StringMap<void*,llvm::MallocAllocator>"* %this, i64 0, i32 0, i32 0 ; <%"struct.llvm::StringMapImpl::ItemBucket"**> [#uses=1]
-  %tmp5.i = load %"struct.llvm::StringMapImpl::ItemBucket"** %tmp4.i, align 8 ; <%"struct.llvm::StringMapImpl::ItemBucket"*> [#uses=1]
-  %tmp6.i = zext i32 %tmp3.i to i64               ; <i64> [#uses=1]
-  %tmp7.i = getelementptr inbounds %"struct.llvm::StringMapImpl::ItemBucket"* %tmp5.i, i64 %tmp6.i, i32 1 ; <%"struct.llvm::StringMapEntryBase"**> [#uses=2]
-  %tmp8.i = load %"struct.llvm::StringMapEntryBase"** %tmp7.i, align 8 ; <%"struct.llvm::StringMapEntryBase"*> [#uses=3]
-  %tmp9.i = icmp eq %"struct.llvm::StringMapEntryBase"* %tmp8.i, null ; <i1> [#uses=1]
-  %tmp13.i = icmp eq %"struct.llvm::StringMapEntryBase"* %tmp8.i, inttoptr (i64 -1 to %"struct.llvm::StringMapEntryBase"*) ; <i1> [#uses=1]
-  %or.cond.i = or i1 %tmp9.i, %tmp13.i            ; <i1> [#uses=1]
+  %elt = bitcast %"struct.llvm::StringRef"* %Key to i64*
+  %val = load i64* %elt
+  %tmp = getelementptr inbounds %"struct.llvm::StringRef"* %Key, i64 0, i32 1
+  %val2 = load i64* %tmp
+  %tmp2.i = getelementptr inbounds %"struct.llvm::StringMap<void*,llvm::MallocAllocator>"* %this, i64 0, i32 0
+  %tmp3.i = tail call i32 @_ZN4llvm13StringMapImpl15LookupBucketForENS_9StringRefE(%"struct.llvm::StringMapImpl"* %tmp2.i, i64 %val, i64 %val2)
+  %tmp4.i = getelementptr inbounds %"struct.llvm::StringMap<void*,llvm::MallocAllocator>"* %this, i64 0, i32 0, i32 0
+  %tmp5.i = load %"struct.llvm::StringMapImpl::ItemBucket"** %tmp4.i, align 8
+  %tmp6.i = zext i32 %tmp3.i to i64
+  %tmp7.i = getelementptr inbounds %"struct.llvm::StringMapImpl::ItemBucket"* %tmp5.i, i64 %tmp6.i, i32 1
+  %tmp8.i = load %"struct.llvm::StringMapEntryBase"** %tmp7.i, align 8
+  %tmp9.i = icmp eq %"struct.llvm::StringMapEntryBase"* %tmp8.i, null
+  %tmp13.i = icmp eq %"struct.llvm::StringMapEntryBase"* %tmp8.i, inttoptr (i64 -1 to %"struct.llvm::StringMapEntryBase"*)
+  %or.cond.i = or i1 %tmp9.i, %tmp13.i
   br i1 %or.cond.i, label %bb4.i, label %bb6.i
 
 bb4.i:                                            ; preds = %entry
-  %tmp41.i = inttoptr i64 %val to i8*             ; <i8*> [#uses=2]
-  %tmp4.i35.i = getelementptr inbounds i8* %tmp41.i, i64 %val2 ; <i8*> [#uses=1]
-  %tmp.i.i = ptrtoint i8* %tmp4.i35.i to i64      ; <i64> [#uses=1]
-  %tmp1.i.i = trunc i64 %tmp.i.i to i32           ; <i32> [#uses=1]
-  %tmp3.i.i = trunc i64 %val to i32               ; <i32> [#uses=1]
-  %tmp4.i.i = sub i32 %tmp1.i.i, %tmp3.i.i        ; <i32> [#uses=3]
-  %tmp5.i.i = add i32 %tmp4.i.i, 17               ; <i32> [#uses=1]
-  %tmp8.i.i = zext i32 %tmp5.i.i to i64           ; <i64> [#uses=1]
-  %tmp.i20.i.i = tail call noalias i8* @malloc(i64 %tmp8.i.i) nounwind ; <i8*> [#uses=7]
-  %tmp10.i.i = bitcast i8* %tmp.i20.i.i to %"struct.llvm::StringMapEntry<void*>"* ; <%"struct.llvm::StringMapEntry<void*>"*> [#uses=2]
-  %tmp12.i.i = icmp eq i8* %tmp.i20.i.i, null     ; <i1> [#uses=1]
+  %tmp41.i = inttoptr i64 %val to i8*
+  %tmp4.i35.i = getelementptr inbounds i8* %tmp41.i, i64 %val2
+  %tmp.i.i = ptrtoint i8* %tmp4.i35.i to i64
+  %tmp1.i.i = trunc i64 %tmp.i.i to i32
+  %tmp3.i.i = trunc i64 %val to i32
+  %tmp4.i.i = sub i32 %tmp1.i.i, %tmp3.i.i
+  %tmp5.i.i = add i32 %tmp4.i.i, 17
+  %tmp8.i.i = zext i32 %tmp5.i.i to i64
+  %tmp.i20.i.i = tail call noalias i8* @malloc(i64 %tmp8.i.i) nounwind
+  %tmp10.i.i = bitcast i8* %tmp.i20.i.i to %"struct.llvm::StringMapEntry<void*>"*
+  %tmp12.i.i = icmp eq i8* %tmp.i20.i.i, null
   br i1 %tmp12.i.i, label %_ZN4llvm14StringMapEntryIPvE6CreateINS_15MallocAllocatorES1_EEPS2_PKcS7_RT_T0_.exit.i, label %bb.i.i
 
 bb.i.i:                                           ; preds = %bb4.i
-  %tmp.i.i.i.i = bitcast i8* %tmp.i20.i.i to i32* ; <i32*> [#uses=1]
+  %tmp.i.i.i.i = bitcast i8* %tmp.i20.i.i to i32*
   store i32 %tmp4.i.i, i32* %tmp.i.i.i.i, align 4
-  %tmp1.i19.i.i = getelementptr inbounds i8* %tmp.i20.i.i, i64 8 ; <i8*> [#uses=1]
-  %0 = bitcast i8* %tmp1.i19.i.i to i8**          ; <i8**> [#uses=1]
+  %tmp1.i19.i.i = getelementptr inbounds i8* %tmp.i20.i.i, i64 8
+  %0 = bitcast i8* %tmp1.i19.i.i to i8**
   store i8* null, i8** %0, align 8
   br label %_ZN4llvm14StringMapEntryIPvE6CreateINS_15MallocAllocatorES1_EEPS2_PKcS7_RT_T0_.exit.i
 
-_ZN4llvm14StringMapEntryIPvE6CreateINS_15MallocAllocatorES1_EEPS2_PKcS7_RT_T0_.exit.i: ; preds = %bb4.i, %bb.i.i
-  %tmp.i18.i.i = getelementptr inbounds i8* %tmp.i20.i.i, i64 16 ; <i8*> [#uses=1]
-  %tmp15.i.i = zext i32 %tmp4.i.i to i64          ; <i64> [#uses=2]
-  tail call void @llvm.memcpy.i64(i8* %tmp.i18.i.i, i8* %tmp41.i, i64 %tmp15.i.i, i32 1) nounwind
-  %tmp.i18.sum.i.i = add i64 %tmp15.i.i, 16       ; <i64> [#uses=1]
-  %tmp17.i.i = getelementptr inbounds i8* %tmp.i20.i.i, i64 %tmp.i18.sum.i.i ; <i8*> [#uses=1]
+_ZN4llvm14StringMapEntryIPvE6CreateINS_15MallocAllocatorES1_EEPS2_PKcS7_RT_T0_.exit.i: ; preds = %bb.i.i, %bb4.i
+  %tmp.i18.i.i = getelementptr inbounds i8* %tmp.i20.i.i, i64 16
+  %tmp15.i.i = zext i32 %tmp4.i.i to i64
+  tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %tmp.i18.i.i, i8* %tmp41.i, i64 %tmp15.i.i, i32 1, i1 false)
+  %tmp.i18.sum.i.i = add i64 %tmp15.i.i, 16
+  %tmp17.i.i = getelementptr inbounds i8* %tmp.i20.i.i, i64 %tmp.i18.sum.i.i
   store i8 0, i8* %tmp17.i.i, align 1
-  %tmp.i.i.i = getelementptr inbounds i8* %tmp.i20.i.i, i64 8 ; <i8*> [#uses=1]
-  %1 = bitcast i8* %tmp.i.i.i to i8**             ; <i8**> [#uses=1]
+  %tmp.i.i.i = getelementptr inbounds i8* %tmp.i20.i.i, i64 8
+  %1 = bitcast i8* %tmp.i.i.i to i8**
   store i8* null, i8** %1, align 8
-  %tmp22.i = load %"struct.llvm::StringMapEntryBase"** %tmp7.i, align 8 ; <%"struct.llvm::StringMapEntryBase"*> [#uses=1]
-  %tmp24.i = icmp eq %"struct.llvm::StringMapEntryBase"* %tmp22.i, inttoptr (i64 -1 to %"struct.llvm::StringMapEntryBase"*) ; <i1> [#uses=1]
+  %tmp22.i = load %"struct.llvm::StringMapEntryBase"** %tmp7.i, align 8
+  %tmp24.i = icmp eq %"struct.llvm::StringMapEntryBase"* %tmp22.i, inttoptr (i64 -1 to %"struct.llvm::StringMapEntryBase"*)
   br i1 %tmp24.i, label %bb9.i, label %_ZN4llvm9StringMapIPvNS_15MallocAllocatorEE16GetOrCreateValueIS1_EERNS_14StringMapEntryIS1_EENS_9StringRefET_.exit
 
 bb6.i:                                            ; preds = %entry
-  %tmp16.i = bitcast %"struct.llvm::StringMapEntryBase"* %tmp8.i to %"struct.llvm::StringMapEntry<void*>"* ; <%"struct.llvm::StringMapEntry<void*>"*> [#uses=1]
+  %tmp16.i = bitcast %"struct.llvm::StringMapEntryBase"* %tmp8.i to %"struct.llvm::StringMapEntry<void*>"*
   ret %"struct.llvm::StringMapEntry<void*>"* %tmp16.i
 
 bb9.i:                                            ; preds = %_ZN4llvm14StringMapEntryIPvE6CreateINS_15MallocAllocatorES1_EEPS2_PKcS7_RT_T0_.exit.i
-  %tmp25.i = getelementptr inbounds %"struct.llvm::StringMap<void*,llvm::MallocAllocator>"* %this, i64 0, i32 0, i32 3 ; <i32*> [#uses=2]
-  %tmp26.i = load i32* %tmp25.i, align 8          ; <i32> [#uses=1]
-  %tmp27.i = add i32 %tmp26.i, -1                 ; <i32> [#uses=1]
+  %tmp25.i = getelementptr inbounds %"struct.llvm::StringMap<void*,llvm::MallocAllocator>"* %this, i64 0, i32 0, i32 3
+  %tmp26.i = load i32* %tmp25.i, align 8
+  %tmp27.i = add i32 %tmp26.i, -1
   store i32 %tmp27.i, i32* %tmp25.i, align 8
   ret %"struct.llvm::StringMapEntry<void*>"* %tmp10.i.i
 
 _ZN4llvm9StringMapIPvNS_15MallocAllocatorEE16GetOrCreateValueIS1_EERNS_14StringMapEntryIS1_EENS_9StringRefET_.exit: ; preds = %_ZN4llvm14StringMapEntryIPvE6CreateINS_15MallocAllocatorES1_EEPS2_PKcS7_RT_T0_.exit.i
   ret %"struct.llvm::StringMapEntry<void*>"* %tmp10.i.i
 }
+
+declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture, i64, i32, i1) nounwind
diff --git a/test/Transforms/GVN/rle.ll b/test/Transforms/GVN/rle.ll
index 28b1fc7b60..2f0d2eb367 100644
--- a/test/Transforms/GVN/rle.ll
+++ b/test/Transforms/GVN/rle.ll
@@ -135,7 +135,7 @@ define i8* @coerce_mustalias7(i64 %V, i64* %P) {
 define signext i16 @memset_to_i16_local(i16* %A) nounwind ssp {
 entry:
   %conv = bitcast i16* %A to i8* 
-  tail call void @llvm.memset.i64(i8* %conv, i8 1, i64 200, i32 1)
+  tail call void @llvm.memset.p0i8.i64(i8* %conv, i8 1, i64 200, i32 1, i1 false)
   %arrayidx = getelementptr inbounds i16* %A, i64 42
   %tmp2 = load i16* %arrayidx
   ret i16 %tmp2
@@ -148,7 +148,7 @@ entry:
 define float @memset_to_float_local(float* %A, i8 %Val) nounwind ssp {
 entry:
   %conv = bitcast float* %A to i8*                ; <i8*> [#uses=1]
-  tail call void @llvm.memset.i64(i8* %conv, i8 %Val, i64 400, i32 1)
+  tail call void @llvm.memset.p0i8.i64(i8* %conv, i8 %Val, i64 400, i32 1, i1 false)
   %arrayidx = getelementptr inbounds float* %A, i64 42 ; <float*> [#uses=1]
   %tmp2 = load float* %arrayidx                   ; <float> [#uses=1]
   ret float %tmp2
@@ -168,11 +168,11 @@ define i16 @memset_to_i16_nonlocal0(i16* %P, i1 %cond) {
   %P3 = bitcast i16* %P to i8*
   br i1 %cond, label %T, label %F
 T:
-  tail call void @llvm.memset.i64(i8* %P3, i8 1, i64 400, i32 1)
+  tail call void @llvm.memset.p0i8.i64(i8* %P3, i8 1, i64 400, i32 1, i1 false)
   br label %Cont
   
 F:
-  tail call void @llvm.memset.i64(i8* %P3, i8 2, i64 400, i32 1)
+  tail call void @llvm.memset.p0i8.i64(i8* %P3, i8 2, i64 400, i32 1, i1 false)
   br label %Cont
 
 Cont:
@@ -193,7 +193,7 @@ Cont:
 define float @memcpy_to_float_local(float* %A) nounwind ssp {
 entry:
   %conv = bitcast float* %A to i8*                ; <i8*> [#uses=1]
-  tail call void @llvm.memcpy.i64(i8* %conv, i8* bitcast ({i32, float, i32 }* @GCst to i8*), i64 12, i32 1)
+  tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %conv, i8* bitcast ({i32, float, i32 }* @GCst to i8*), i64 12, i32 1, i1 false)
   %arrayidx = getelementptr inbounds float* %A, i64 1 ; <float*> [#uses=1]
   %tmp2 = load float* %arrayidx                   ; <float> [#uses=1]
   ret float %tmp2
@@ -203,11 +203,6 @@ entry:
 }
 
 
-declare void @llvm.memset.i64(i8* nocapture, i8, i64, i32) nounwind
-declare void @llvm.memcpy.i64(i8* nocapture, i8* nocapture, i64, i32) nounwind
-
-
-
 
 ;; non-local i32/float -> i8 load forwarding.
 define i8 @coerce_mustalias_nonlocal0(i32* %P, i1 %cond) {
@@ -539,7 +534,7 @@ define i32 @memset_to_load() nounwind readnone {
 entry:
   %x = alloca [256 x i32], align 4                ; <[256 x i32]*> [#uses=2]
   %tmp = bitcast [256 x i32]* %x to i8*           ; <i8*> [#uses=1]
-  call void @llvm.memset.i64(i8* %tmp, i8 0, i64 1024, i32 4)
+  call void @llvm.memset.p0i8.i64(i8* %tmp, i8 0, i64 1024, i32 4, i1 false)
   %arraydecay = getelementptr inbounds [256 x i32]* %x, i32 0, i32 0 ; <i32*>
   %tmp1 = load i32* %arraydecay                   ; <i32> [#uses=1]
   ret i32 %tmp1
@@ -643,3 +638,7 @@ entry:
 ; CHECK-ret i32
 }
 
+declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i32, i1) nounwind
+
+declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture, i64, i32, i1) nounwind
+
diff --git a/test/Transforms/GlobalOpt/memcpy.ll b/test/Transforms/GlobalOpt/memcpy.ll
index 8f063a2fe4..94e07a0848 100644
--- a/test/Transforms/GlobalOpt/memcpy.ll
+++ b/test/Transforms/GlobalOpt/memcpy.ll
@@ -3,13 +3,11 @@
 
 @G1 = internal global [58 x i8] c"asdlfkajsdlfkajsd;lfkajds;lfkjasd;flkajsd;lkfja;sdlkfjasd\00"         ; <[58 x i8]*> [#uses=1]
 
-declare void @llvm.memcpy.i32(i8*, i8*, i32, i32)
-
 define void @foo() {
-        %Blah = alloca [58 x i8]                ; <[58 x i8]*> [#uses=1]
-        %tmp.0 = getelementptr [58 x i8]* %Blah, i32 0, i32 0           ; <i8*> [#uses=1]
-        call void @llvm.memcpy.i32( i8* %tmp.0, i8* getelementptr ([58 x i8]* @G1, i32 0, i32 0), i32 58, i32 1 )
-        ret void
+  %Blah = alloca [58 x i8]
+  %tmp.0 = getelementptr [58 x i8]* %Blah, i32 0, i32 0
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %tmp.0, i8* getelementptr inbounds ([58 x i8]* @G1, i32 0, i32 0), i32 58, i32 1, i1 false)
+  ret void
 }
 
-
+declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture, i32, i32, i1) nounwind
diff --git a/test/Transforms/GlobalOpt/memset.ll b/test/Transforms/GlobalOpt/memset.ll
index a9b9d5e6bd..3bb5ce92d0 100644
--- a/test/Transforms/GlobalOpt/memset.ll
+++ b/test/Transforms/GlobalOpt/memset.ll
@@ -1,21 +1,18 @@
 ; both globals are write only, delete them.
 
-; RUN: opt < %s -globalopt -S | \
-; RUN:   not grep internal
+; RUN: opt < %s -globalopt -S | not grep internal
 
 @G0 = internal global [58 x i8] c"asdlfkajsdlfkajsd;lfkajds;lfkjasd;flkajsd;lkfja;sdlkfjasd\00"         ; <[58 x i8]*> [#uses=1]
 @G1 = internal global [4 x i32] [ i32 1, i32 2, i32 3, i32 4 ]          ; <[4 x i32]*> [#uses=1]
 
-declare void @llvm.memcpy.i32(i8*, i8*, i32, i32)
-
-declare void @llvm.memset.i32(i8*, i8, i32, i32)
-
 define void @foo() {
-        %Blah = alloca [58 x i8]                ; <[58 x i8]*> [#uses=1]
-        %tmp3 = bitcast [58 x i8]* %Blah to i8*         ; <i8*> [#uses=1]
-        call void @llvm.memcpy.i32( i8* bitcast ([4 x i32]* @G1 to i8*), i8* %tmp3, i32 16, i32 1 )
-        call void @llvm.memset.i32( i8* getelementptr ([58 x i8]* @G0, i32 0, i32 0), i8 17, i32 58, i32 1 )
-        ret void
+  %Blah = alloca [58 x i8]
+  %tmp3 = bitcast [58 x i8]* %Blah to i8*
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* bitcast ([4 x i32]* @G1 to i8*), i8* %tmp3, i32 16, i32 1, i1 false)
+  call void @llvm.memset.p0i8.i32(i8* getelementptr inbounds ([58 x i8]* @G0, i32 0, i32 0), i8 17, i32 58, i32 1, i1 false)
+  ret void
 }
 
+declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture, i32, i32, i1) nounwind
 
+declare void @llvm.memset.p0i8.i32(i8* nocapture, i8, i32, i32, i1) nounwind
diff --git a/test/Transforms/Inline/inline-invoke-tail.ll b/test/Transforms/Inline/inline-invoke-tail.ll
index 961f6789fe..462c29a85c 100644
--- a/test/Transforms/Inline/inline-invoke-tail.ll
+++ b/test/Transforms/Inline/inline-invoke-tail.ll
@@ -4,7 +4,7 @@
 define internal void @foo(i32* %p, i32* %q) {
 	%pp = bitcast i32* %p to i8*
 	%qq = bitcast i32* %q to i8*
-	tail call void @llvm.memcpy.i32(i8* %pp, i8* %qq, i32 4, i32 1)
+	tail call void @llvm.memcpy.p0i8.p0i8.i32(i8* %pp, i8* %qq, i32 4, i32 1, i1 false)
 	ret void
 }
 
@@ -24,12 +24,14 @@ invcont:
 
 lpad:
 	%eh_ptr = call i8* @llvm.eh.exception()
-	%eh_select = call i32 (i8*, i8*, ...)* @llvm.eh.selector.i32(i8* %eh_ptr, i8* bitcast (i32 (...)* @__gxx_personality_v0 to i8*), i8* null)
+	%eh_select = call i32 (i8*, i8*, ...)* @llvm.eh.selector(i8* %eh_ptr, i8* bitcast (i32 (...)* @__gxx_personality_v0 to i8*), i8* null)
 	unreachable
 }
 
-declare i8* @llvm.eh.exception() nounwind
+declare i8* @llvm.eh.exception() nounwind readonly
 
-declare i32 @llvm.eh.selector.i32(i8*, i8*, ...) nounwind
+declare i32 @llvm.eh.selector(i8*, i8*, ...) nounwind
 
 declare i32 @__gxx_personality_v0(...)
+
+declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture, i32, i32, i1) nounwind
diff --git a/test/Transforms/InstCombine/2006-11-03-Memmove64.ll b/test/Transforms/InstCombine/2006-11-03-Memmove64.ll
deleted file mode 100644
index 35bb45e2b9..0000000000
--- a/test/Transforms/InstCombine/2006-11-03-Memmove64.ll
+++ /dev/null
@@ -1,16 +0,0 @@
-; RUN: opt < %s -instcombine -S | \
-; RUN:    not grep memmove.i32
-; Instcombine was trying to turn this into a memmove.i32
-
-target datalayout = "e-p:64:64"
-target triple = "alphaev67-unknown-linux-gnu"
-@str10 = internal constant [1 x i8] zeroinitializer             ; <[1 x i8]*> [#uses=1]
-
-define void @do_join(i8* %b) {
-entry:
-        call void @llvm.memmove.i64( i8* %b, i8* getelementptr ([1 x i8]* @str10, i32 0, i64 0), i64 1, i32 1 )
-        ret void
-}
-
-declare void @llvm.memmove.i64(i8*, i8*, i64, i32)
-
diff --git a/test/Transforms/InstCombine/2007-05-04-Crash.ll b/test/Transforms/InstCombine/2007-05-04-Crash.ll
deleted file mode 100644
index 9f50d8ac9b..0000000000
--- a/test/Transforms/InstCombine/2007-05-04-Crash.ll
+++ /dev/null
@@ -1,30 +0,0 @@
-; RUN: opt < %s -instcombine -disable-output
-; PR1384
-
-target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f32:32:32-f64:32:64-v64:64:64-v128:128:128-a0:0:64"
-target triple = "i686-apple-darwin8"
-	%struct.CFRuntimeBase = type { i32, [4 x i8] }
-	%struct.CGColor = type opaque
-	%struct.CGColorSpace = type { %struct.CFRuntimeBase, i8, i8, i8, i32, i32, i32, %struct.CGColor*, float*, %struct.CGMD5Signature, %struct.CGMD5Signature*, [0 x %struct.CGColorSpaceDescriptor] }
-	%struct.CGColorSpaceCalibratedRGBData = type { [3 x float], [3 x float], [3 x float], [9 x float] }
-	%struct.CGColorSpaceDescriptor = type { %struct.CGColorSpaceCalibratedRGBData }
-	%struct.CGColorSpaceLabData = type { [3 x float], [3 x float], [4 x float] }
-	%struct.CGMD5Signature = type { [16 x i8], i8 }
-
-declare fastcc %struct.CGColorSpace* @CGColorSpaceCreate(i32, i32)
-
-declare void @llvm.memcpy.i32(i8*, i8*, i32, i32)
-
-define %struct.CGColorSpace* @CGColorSpaceCreateLab(float* %whitePoint, float* %blackPoint, float* %range) {
-entry:
-	%tmp17 = call fastcc %struct.CGColorSpace* @CGColorSpaceCreate( i32 5, i32 3 )		; <%struct.CGColorSpace*> [#uses=2]
-	%tmp28 = getelementptr %struct.CGColorSpace* %tmp17, i32 0, i32 11		; <[0 x %struct.CGColorSpaceDescriptor]*> [#uses=1]
-	%tmp29 = getelementptr [0 x %struct.CGColorSpaceDescriptor]* %tmp28, i32 0, i32 0		; <%struct.CGColorSpaceDescriptor*> [#uses=1]
-	%tmp30 = getelementptr %struct.CGColorSpaceDescriptor* %tmp29, i32 0, i32 0		; <%struct.CGColorSpaceCalibratedRGBData*> [#uses=1]
-	%tmp3031 = bitcast %struct.CGColorSpaceCalibratedRGBData* %tmp30 to %struct.CGColorSpaceLabData*		; <%struct.CGColorSpaceLabData*> [#uses=1]
-	%tmp45 = getelementptr %struct.CGColorSpaceLabData* %tmp3031, i32 0, i32 2		; <[4 x float]*> [#uses=1]
-	%tmp46 = getelementptr [4 x float]* %tmp45, i32 0, i32 0		; <float*> [#uses=1]
-	%tmp4648 = bitcast float* %tmp46 to i8*		; <i8*> [#uses=1]
-	call void @llvm.memcpy.i32( i8* %tmp4648, i8* null, i32 16, i32 4 )
-	ret %struct.CGColorSpace* %tmp17
-}
diff --git a/test/Transforms/InstCombine/2007-10-10-EliminateMemCpy.ll b/test/Transforms/InstCombine/2007-10-10-EliminateMemCpy.ll
index 710aff274a..fe935f96e9 100644
--- a/test/Transforms/InstCombine/2007-10-10-EliminateMemCpy.ll
+++ b/test/Transforms/InstCombine/2007-10-10-EliminateMemCpy.ll
@@ -6,16 +6,15 @@ target datalayout = "E-p:64:64:64-a0:0:8-f32:32:32-f64:64:64-i1:8:8-i8:8:8-i16:1
 
 define void @foo(i8* %P) {
 entry:
-	%P_addr = alloca i8*		; <i8**> [#uses=2]
-	%"alloca point" = bitcast i32 0 to i32		; <i32> [#uses=0]
-	store i8* %P, i8** %P_addr
-	%tmp = load i8** %P_addr, align 4		; <i8*> [#uses=1]
-	%tmp1 = getelementptr [4 x i8]* @.str, i32 0, i32 0		; <i8*> [#uses=1]
-	call void @llvm.memcpy.i32( i8* %tmp, i8* %tmp1, i32 4, i32 1 )
-	br label %return
+  %P_addr = alloca i8*
+  store i8* %P, i8** %P_addr
+  %tmp = load i8** %P_addr, align 4
+  %tmp1 = getelementptr [4 x i8]* @.str, i32 0, i32 0
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %tmp, i8* %tmp1, i32 4, i32 1, i1 false)
+  br label %return
 
-return:		; preds = %entry
-	ret void
+return:                                           ; preds = %entry
+  ret void
 }
 
-declare void @llvm.memcpy.i32(i8*, i8*, i32, i32)
+declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture, i32, i32, i1) nounwind
diff --git a/test/Transforms/InstCombine/2009-02-20-InstCombine-SROA.ll b/test/Transforms/InstCombine/2009-02-20-InstCombine-SROA.ll
index b29d8d23bc..a51c47d423 100644
--- a/test/Transforms/InstCombine/2009-02-20-InstCombine-SROA.ll
+++ b/test/Transforms/InstCombine/2009-02-20-InstCombine-SROA.ll
@@ -4,275 +4,276 @@
 
 target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f32:32:32-f64:32:64-v64:64:64-v128:128:128-a0:0:64-f80:128:128"
 target triple = "i386-apple-darwin9.6"
-	%"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >" = type { i32* }
-	%"struct.std::_Vector_base<int,std::allocator<int> >" = type { %"struct.std::_Vector_base<int,std::allocator<int> >::_Vector_impl" }
-	%"struct.std::_Vector_base<int,std::allocator<int> >::_Vector_impl" = type { i32*, i32*, i32* }
-	%"struct.std::bidirectional_iterator_tag" = type <{ i8 }>
-	%"struct.std::forward_iterator_tag" = type <{ i8 }>
-	%"struct.std::input_iterator_tag" = type <{ i8 }>
-	%"struct.std::random_access_iterator_tag" = type <{ i8 }>
-	%"struct.std::vector<int,std::allocator<int> >" = type { %"struct.std::_Vector_base<int,std::allocator<int> >" }
+
+%"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >" = type { i32* }
+%"struct.std::_Vector_base<int,std::allocator<int> >" = type { %"struct.std::_Vector_base<int,std::allocator<int> >::_Vector_impl" }
+%"struct.std::_Vector_base<int,std::allocator<int> >::_Vector_impl" = type { i32*, i32*, i32* }
+%"struct.std::bidirectional_iterator_tag" = type <{ i8 }>
+%"struct.std::forward_iterator_tag" = type <{ i8 }>
+%"struct.std::input_iterator_tag" = type <{ i8 }>
+%"struct.std::random_access_iterator_tag" = type <{ i8 }>
+%"struct.std::vector<int,std::allocator<int> >" = type { %"struct.std::_Vector_base<int,std::allocator<int> >" }
 
 define i32* @_Z3fooRSt6vectorIiSaIiEE(%"struct.std::vector<int,std::allocator<int> >"* %X) {
 entry:
-	%0 = alloca %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"		; <%"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"*> [#uses=2]
-	%__first_addr.i.i = alloca %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"		; <%"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"*> [#uses=31]
-	%__last_addr.i.i = alloca %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"		; <%"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"*> [#uses=4]
-	%unnamed_arg.i = alloca %"struct.std::bidirectional_iterator_tag", align 8		; <%"struct.std::bidirectional_iterator_tag"*> [#uses=1]
-	%1 = alloca %"struct.std::bidirectional_iterator_tag"		; <%"struct.std::bidirectional_iterator_tag"*> [#uses=1]
-	%__first_addr.i = alloca %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"		; <%"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"*> [#uses=2]
-	%2 = alloca %"struct.std::bidirectional_iterator_tag"		; <%"struct.std::bidirectional_iterator_tag"*> [#uses=2]
-	%3 = alloca %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"		; <%"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"*> [#uses=2]
-	%4 = alloca i32		; <i32*> [#uses=8]
-	%"alloca point" = bitcast i32 0 to i32		; <i32> [#uses=0]
-	store i32 42, i32* %4, align 4
-	%5 = getelementptr %"struct.std::vector<int,std::allocator<int> >"* %X, i32 0, i32 0		; <%"struct.std::_Vector_base<int,std::allocator<int> >"*> [#uses=1]
-	%6 = getelementptr %"struct.std::_Vector_base<int,std::allocator<int> >"* %5, i32 0, i32 0		; <%"struct.std::_Vector_base<int,std::allocator<int> >::_Vector_impl"*> [#uses=1]
-	%7 = getelementptr %"struct.std::_Vector_base<int,std::allocator<int> >::_Vector_impl"* %6, i32 0, i32 1		; <i32**> [#uses=1]
-	%8 = load i32** %7, align 4		; <i32*> [#uses=1]
-	%9 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %3, i32 0, i32 0		; <i32**> [#uses=1]
-	store i32* %8, i32** %9, align 4
-	%10 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %3, i32 0, i32 0		; <i32**> [#uses=1]
-	%11 = load i32** %10, align 4		; <i32*> [#uses=1]
-	%tmp2.i = ptrtoint i32* %11 to i32		; <i32> [#uses=1]
-	%tmp1.i = inttoptr i32 %tmp2.i to i32*		; <i32*> [#uses=1]
-	%tmp3 = ptrtoint i32* %tmp1.i to i32		; <i32> [#uses=1]
-	%tmp2 = inttoptr i32 %tmp3 to i32*		; <i32*> [#uses=1]
-	%12 = getelementptr %"struct.std::vector<int,std::allocator<int> >"* %X, i32 0, i32 0		; <%"struct.std::_Vector_base<int,std::allocator<int> >"*> [#uses=1]
-	%13 = getelementptr %"struct.std::_Vector_base<int,std::allocator<int> >"* %12, i32 0, i32 0		; <%"struct.std::_Vector_base<int,std::allocator<int> >::_Vector_impl"*> [#uses=1]
-	%14 = getelementptr %"struct.std::_Vector_base<int,std::allocator<int> >::_Vector_impl"* %13, i32 0, i32 0		; <i32**> [#uses=1]
-	%15 = load i32** %14, align 4		; <i32*> [#uses=1]
-	%16 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %0, i32 0, i32 0		; <i32**> [#uses=1]
-	store i32* %15, i32** %16, align 4
-	%17 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %0, i32 0, i32 0		; <i32**> [#uses=1]
-	%18 = load i32** %17, align 4		; <i32*> [#uses=1]
-	%tmp2.i17 = ptrtoint i32* %18 to i32		; <i32> [#uses=1]
-	%tmp1.i18 = inttoptr i32 %tmp2.i17 to i32*		; <i32*> [#uses=1]
-	%tmp8 = ptrtoint i32* %tmp1.i18 to i32		; <i32> [#uses=1]
-	%tmp6 = inttoptr i32 %tmp8 to i32*		; <i32*> [#uses=1]
-	%19 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i, i32 0, i32 0		; <i32**> [#uses=1]
-	store i32* %tmp6, i32** %19
-	%20 = getelementptr %"struct.std::bidirectional_iterator_tag"* %1, i32 0, i32 0		; <i8*> [#uses=1]
-	%21 = load i8* %20, align 1		; <i8> [#uses=1]
-	%22 = or i8 %21, 0		; <i8> [#uses=1]
-	%23 = or i8 %22, 0		; <i8> [#uses=1]
-	%24 = or i8 %23, 0		; <i8> [#uses=0]
-	%25 = getelementptr %"struct.std::bidirectional_iterator_tag"* %2, i32 0, i32 0		; <i8*> [#uses=1]
-	store i8 0, i8* %25, align 1
-	%elt.i = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i, i32 0, i32 0		; <i32**> [#uses=1]
-	%val.i = load i32** %elt.i		; <i32*> [#uses=1]
-	%tmp.i = bitcast %"struct.std::bidirectional_iterator_tag"* %unnamed_arg.i to i8*		; <i8*> [#uses=1]
-	%tmp9.i = bitcast %"struct.std::bidirectional_iterator_tag"* %2 to i8*		; <i8*> [#uses=1]
-	call void @llvm.memcpy.i64(i8* %tmp.i, i8* %tmp9.i, i64 1, i32 1)
-	%26 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0		; <i32**> [#uses=1]
-	store i32* %val.i, i32** %26
-	%27 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__last_addr.i.i, i32 0, i32 0		; <i32**> [#uses=1]
-	store i32* %tmp2, i32** %27
-	%28 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__last_addr.i.i, i32 0, i32 0		; <i32**> [#uses=1]
-	%29 = load i32** %28, align 4		; <i32*> [#uses=1]
-	%30 = ptrtoint i32* %29 to i32		; <i32> [#uses=1]
-	%31 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0		; <i32**> [#uses=1]
-	%32 = load i32** %31, align 4		; <i32*> [#uses=1]
-	%33 = ptrtoint i32* %32 to i32		; <i32> [#uses=1]
-	%34 = sub i32 %30, %33		; <i32> [#uses=1]
-	%35 = ashr i32 %34, 2		; <i32> [#uses=1]
-	%36 = ashr i32 %35, 2		; <i32> [#uses=1]
-	br label %bb12.i.i
+  %0 = alloca %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"
+  %__first_addr.i.i = alloca %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"
+  %__last_addr.i.i = alloca %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"
+  %unnamed_arg.i = alloca %"struct.std::bidirectional_iterator_tag", align 8
+  %1 = alloca %"struct.std::bidirectional_iterator_tag"
+  %__first_addr.i = alloca %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"
+  %2 = alloca %"struct.std::bidirectional_iterator_tag"
+  %3 = alloca %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"
+  %4 = alloca i32
+  %"alloca point" = bitcast i32 0 to i32
+  store i32 42, i32* %4, align 4
+  %5 = getelementptr %"struct.std::vector<int,std::allocator<int> >"* %X, i32 0, i32 0
+  %6 = getelementptr %"struct.std::_Vector_base<int,std::allocator<int> >"* %5, i32 0, i32 0
+  %7 = getelementptr %"struct.std::_Vector_base<int,std::allocator<int> >::_Vector_impl"* %6, i32 0, i32 1
+  %8 = load i32** %7, align 4
+  %9 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %3, i32 0, i32 0
+  store i32* %8, i32** %9, align 4
+  %10 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %3, i32 0, i32 0
+  %11 = load i32** %10, align 4
+  %tmp2.i = ptrtoint i32* %11 to i32
+  %tmp1.i = inttoptr i32 %tmp2.i to i32*
+  %tmp3 = ptrtoint i32* %tmp1.i to i32
+  %tmp2 = inttoptr i32 %tmp3 to i32*
+  %12 = getelementptr %"struct.std::vector<int,std::allocator<int> >"* %X, i32 0, i32 0
+  %13 = getelementptr %"struct.std::_Vector_base<int,std::allocator<int> >"* %12, i32 0, i32 0
+  %14 = getelementptr %"struct.std::_Vector_base<int,std::allocator<int> >::_Vector_impl"* %13, i32 0, i32 0
+  %15 = load i32** %14, align 4
+  %16 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %0, i32 0, i32 0
+  store i32* %15, i32** %16, align 4
+  %17 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %0, i32 0, i32 0
+  %18 = load i32** %17, align 4
+  %tmp2.i17 = ptrtoint i32* %18 to i32
+  %tmp1.i18 = inttoptr i32 %tmp2.i17 to i32*
+  %tmp8 = ptrtoint i32* %tmp1.i18 to i32
+  %tmp6 = inttoptr i32 %tmp8 to i32*
+  %19 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i, i32 0, i32 0
+  store i32* %tmp6, i32** %19
+  %20 = getelementptr %"struct.std::bidirectional_iterator_tag"* %1, i32 0, i32 0
+  %21 = load i8* %20, align 1
+  %22 = or i8 %21, 0
+  %23 = or i8 %22, 0
+  %24 = or i8 %23, 0
+  %25 = getelementptr %"struct.std::bidirectional_iterator_tag"* %2, i32 0, i32 0
+  store i8 0, i8* %25, align 1
+  %elt.i = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i, i32 0, i32 0
+  %val.i = load i32** %elt.i
+  %tmp.i = bitcast %"struct.std::bidirectional_iterator_tag"* %unnamed_arg.i to i8*
+  %tmp9.i = bitcast %"struct.std::bidirectional_iterator_tag"* %2 to i8*
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %tmp.i, i8* %tmp9.i, i64 1, i32 1, i1 false)
+  %26 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  store i32* %val.i, i32** %26
+  %27 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__last_addr.i.i, i32 0, i32 0
+  store i32* %tmp2, i32** %27
+  %28 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__last_addr.i.i, i32 0, i32 0
+  %29 = load i32** %28, align 4
+  %30 = ptrtoint i32* %29 to i32
+  %31 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %32 = load i32** %31, align 4
+  %33 = ptrtoint i32* %32 to i32
+  %34 = sub i32 %30, %33
+  %35 = ashr i32 %34, 2
+  %36 = ashr i32 %35, 2
+  br label %bb12.i.i
 
-bb.i.i:		; preds = %bb12.i.i
-	%37 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0		; <i32**> [#uses=1]
-	%38 = load i32** %37, align 4		; <i32*> [#uses=1]
-	%39 = load i32* %38, align 4		; <i32> [#uses=1]
-	%40 = load i32* %4, align 4		; <i32> [#uses=1]
-	%41 = icmp eq i32 %39, %40		; <i1> [#uses=1]
-	%42 = zext i1 %41 to i8		; <i8> [#uses=1]
-	%toBool.i.i = icmp ne i8 %42, 0		; <i1> [#uses=1]
-	br i1 %toBool.i.i, label %bb1.i.i, label %bb2.i.i
+bb.i.i:                                           ; preds = %bb12.i.i
+  %37 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %38 = load i32** %37, align 4
+  %39 = load i32* %38, align 4
+  %40 = load i32* %4, align 4
+  %41 = icmp eq i32 %39, %40
+  %42 = zext i1 %41 to i8
+  %toBool.i.i = icmp ne i8 %42, 0
+  br i1 %toBool.i.i, label %bb1.i.i, label %bb2.i.i
 
-bb1.i.i:		; preds = %bb.i.i
-	%43 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0		; <i32**> [#uses=1]
-	%44 = load i32** %43, align 4		; <i32*> [#uses=1]
-	br label %_ZSt4findIN9__gnu_cxx17__normal_iteratorIPiSt6vectorIiSaIiEEEEiET_S7_S7_RKT0_.exit
+bb1.i.i:                                          ; preds = %bb.i.i
+  %43 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %44 = load i32** %43, align 4
+  br label %_ZSt4findIN9__gnu_cxx17__normal_iteratorIPiSt6vectorIiSaIiEEEEiET_S7_S7_RKT0_.exit
 
-bb2.i.i:		; preds = %bb.i.i
-	%45 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0		; <i32**> [#uses=1]
-	%46 = load i32** %45, align 4		; <i32*> [#uses=1]
-	%47 = getelementptr i32* %46, i64 1		; <i32*> [#uses=1]
-	%48 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0		; <i32**> [#uses=1]
-	store i32* %47, i32** %48, align 4
-	%49 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0		; <i32**> [#uses=1]
-	%50 = load i32** %49, align 4		; <i32*> [#uses=1]
-	%51 = load i32* %50, align 4		; <i32> [#uses=1]
-	%52 = load i32* %4, align 4		; <i32> [#uses=1]
-	%53 = icmp eq i32 %51, %52		; <i1> [#uses=1]
-	%54 = zext i1 %53 to i8		; <i8> [#uses=1]
-	%toBool3.i.i = icmp ne i8 %54, 0		; <i1> [#uses=1]
-	br i1 %toBool3.i.i, label %bb4.i.i, label %bb5.i.i
+bb2.i.i:                                          ; preds = %bb.i.i
+  %45 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %46 = load i32** %45, align 4
+  %47 = getelementptr i32* %46, i64 1
+  %48 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  store i32* %47, i32** %48, align 4
+  %49 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %50 = load i32** %49, align 4
+  %51 = load i32* %50, align 4
+  %52 = load i32* %4, align 4
+  %53 = icmp eq i32 %51, %52
+  %54 = zext i1 %53 to i8
+  %toBool3.i.i = icmp ne i8 %54, 0
+  br i1 %toBool3.i.i, label %bb4.i.i, label %bb5.i.i
 
-bb4.i.i:		; preds = %bb2.i.i
-	%55 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0		; <i32**> [#uses=1]
-	%56 = load i32** %55, align 4		; <i32*> [#uses=1]
-	br label %_ZSt4findIN9__gnu_cxx17__normal_iteratorIPiSt6vectorIiSaIiEEEEiET_S7_S7_RKT0_.exit
+bb4.i.i:                                          ; preds = %bb2.i.i
+  %55 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %56 = load i32** %55, align 4
+  br label %_ZSt4findIN9__gnu_cxx17__normal_iteratorIPiSt6vectorIiSaIiEEEEiET_S7_S7_RKT0_.exit
 
-bb5.i.i:		; preds = %bb2.i.i
-	%57 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0		; <i32**> [#uses=1]
-	%58 = load i32** %57, align 4		; <i32*> [#uses=1]
-	%59 = getelementptr i32* %58, i64 1		; <i32*> [#uses=1]
-	%60 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0		; <i32**> [#uses=1]
-	store i32* %59, i32** %60, align 4
-	%61 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0		; <i32**> [#uses=1]
-	%62 = load i32** %61, align 4		; <i32*> [#uses=1]
-	%63 = load i32* %62, align 4		; <i32> [#uses=1]
-	%64 = load i32* %4, align 4		; <i32> [#uses=1]
-	%65 = icmp eq i32 %63, %64		; <i1> [#uses=1]
-	%66 = zext i1 %65 to i8		; <i8> [#uses=1]
-	%toBool6.i.i = icmp ne i8 %66, 0		; <i1> [#uses=1]
-	br i1 %toBool6.i.i, label %bb7.i.i, label %bb8.i.i
+bb5.i.i:                                          ; preds = %bb2.i.i
+  %57 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %58 = load i32** %57, align 4
+  %59 = getelementptr i32* %58, i64 1
+  %60 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  store i32* %59, i32** %60, align 4
+  %61 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %62 = load i32** %61, align 4
+  %63 = load i32* %62, align 4
+  %64 = load i32* %4, align 4
+  %65 = icmp eq i32 %63, %64
+  %66 = zext i1 %65 to i8
+  %toBool6.i.i = icmp ne i8 %66, 0
+  br i1 %toBool6.i.i, label %bb7.i.i, label %bb8.i.i
 
-bb7.i.i:		; preds = %bb5.i.i
-	%67 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0		; <i32**> [#uses=1]
-	%68 = load i32** %67, align 4		; <i32*> [#uses=1]
-	br label %_ZSt4findIN9__gnu_cxx17__normal_iteratorIPiSt6vectorIiSaIiEEEEiET_S7_S7_RKT0_.exit
+bb7.i.i:                                          ; preds = %bb5.i.i
+  %67 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %68 = load i32** %67, align 4
+  br label %_ZSt4findIN9__gnu_cxx17__normal_iteratorIPiSt6vectorIiSaIiEEEEiET_S7_S7_RKT0_.exit
 
-bb8.i.i:		; preds = %bb5.i.i
-	%69 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0		; <i32**> [#uses=1]
-	%70 = load i32** %69, align 4		; <i32*> [#uses=1]
-	%71 = getelementptr i32* %70, i64 1		; <i32*> [#uses=1]
-	%72 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0		; <i32**> [#uses=1]
-	store i32* %71, i32** %72, align 4
-	%73 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0		; <i32**> [#uses=1]
-	%74 = load i32** %73, align 4		; <i32*> [#uses=1]
-	%75 = load i32* %74, align 4		; <i32> [#uses=1]
-	%76 = load i32* %4, align 4		; <i32> [#uses=1]
-	%77 = icmp eq i32 %75, %76		; <i1> [#uses=1]
-	%78 = zext i1 %77 to i8		; <i8> [#uses=1]
-	%toBool9.i.i = icmp ne i8 %78, 0		; <i1> [#uses=1]
-	br i1 %toBool9.i.i, label %bb10.i.i, label %bb11.i.i
+bb8.i.i:                                          ; preds = %bb5.i.i
+  %69 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %70 = load i32** %69, align 4
+  %71 = getelementptr i32* %70, i64 1
+  %72 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  store i32* %71, i32** %72, align 4
+  %73 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %74 = load i32** %73, align 4
+  %75 = load i32* %74, align 4
+  %76 = load i32* %4, align 4
+  %77 = icmp eq i32 %75, %76
+  %78 = zext i1 %77 to i8
+  %toBool9.i.i = icmp ne i8 %78, 0
+  br i1 %toBool9.i.i, label %bb10.i.i, label %bb11.i.i
 
-bb10.i.i:		; preds = %bb8.i.i
-	%79 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0		; <i32**> [#uses=1]
-	%80 = load i32** %79, align 4		; <i32*> [#uses=1]
-	br label %_ZSt4findIN9__gnu_cxx17__normal_iteratorIPiSt6vectorIiSaIiEEEEiET_S7_S7_RKT0_.exit
+bb10.i.i:                                         ; preds = %bb8.i.i
+  %79 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %80 = load i32** %79, align 4
+  br label %_ZSt4findIN9__gnu_cxx17__normal_iteratorIPiSt6vectorIiSaIiEEEEiET_S7_S7_RKT0_.exit
 
-bb11.i.i:		; preds = %bb8.i.i
-	%81 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0		; <i32**> [#uses=1]
-	%82 = load i32** %81, align 4		; <i32*> [#uses=1]
-	%83 = getelementptr i32* %82, i64 1		; <i32*> [#uses=1]
-	%84 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0		; <i32**> [#uses=1]
-	store i32* %83, i32** %84, align 4
-	%85 = sub i32 %__trip_count.0.i.i, 1		; <i32> [#uses=1]
-	br label %bb12.i.i
+bb11.i.i:                                         ; preds = %bb8.i.i
+  %81 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %82 = load i32** %81, align 4
+  %83 = getelementptr i32* %82, i64 1
+  %84 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  store i32* %83, i32** %84, align 4
+  %85 = sub i32 %__trip_count.0.i.i, 1
+  br label %bb12.i.i
 
-bb12.i.i:		; preds = %bb11.i.i, %entry
-	%__trip_count.0.i.i = phi i32 [ %36, %entry ], [ %85, %bb11.i.i ]		; <i32> [#uses=2]
-	%86 = icmp sgt i32 %__trip_count.0.i.i, 0		; <i1> [#uses=1]
-	br i1 %86, label %bb.i.i, label %bb13.i.i
+bb12.i.i:                                         ; preds = %bb11.i.i, %entry
+  %__trip_count.0.i.i = phi i32 [ %36, %entry ], [ %85, %bb11.i.i ]
+  %86 = icmp sgt i32 %__trip_count.0.i.i, 0
+  br i1 %86, label %bb.i.i, label %bb13.i.i
 
-bb13.i.i:		; preds = %bb12.i.i
-	%87 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__last_addr.i.i, i32 0, i32 0		; <i32**> [#uses=1]
-	%88 = load i32** %87, align 4		; <i32*> [#uses=1]
-	%89 = ptrtoint i32* %88 to i32		; <i32> [#uses=1]
-	%90 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0		; <i32**> [#uses=1]
-	%91 = load i32** %90, align 4		; <i32*> [#uses=1]
-	%92 = ptrtoint i32* %91 to i32		; <i32> [#uses=1]
-	%93 = sub i32 %89, %92		; <i32> [#uses=1]
-	%94 = ashr i32 %93, 2		; <i32> [#uses=1]
-	switch i32 %94, label %bb26.i.i [
-		i32 1, label %bb22.i.i
-		i32 2, label %bb18.i.i
-		i32 3, label %bb14.i.i
-	]
+bb13.i.i:                                         ; preds = %bb12.i.i
+  %87 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__last_addr.i.i, i32 0, i32 0
+  %88 = load i32** %87, align 4
+  %89 = ptrtoint i32* %88 to i32
+  %90 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %91 = load i32** %90, align 4
+  %92 = ptrtoint i32* %91 to i32
+  %93 = sub i32 %89, %92
+  %94 = ashr i32 %93, 2
+  switch i32 %94, label %bb26.i.i [
+    i32 1, label %bb22.i.i
+    i32 2, label %bb18.i.i
+    i32 3, label %bb14.i.i
+  ]
 
-bb14.i.i:		; preds = %bb13.i.i
-	%95 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0		; <i32**> [#uses=1]
-	%96 = load i32** %95, align 4		; <i32*> [#uses=1]
-	%97 = load i32* %96, align 4		; <i32> [#uses=1]
-	%98 = load i32* %4, align 4		; <i32> [#uses=1]
-	%99 = icmp eq i32 %97, %98		; <i1> [#uses=1]
-	%100 = zext i1 %99 to i8		; <i8> [#uses=1]
-	%toBool15.i.i = icmp ne i8 %100, 0		; <i1> [#uses=1]
-	br i1 %toBool15.i.i, label %bb16.i.i, label %bb17.i.i
+bb14.i.i:                                         ; preds = %bb13.i.i
+  %95 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %96 = load i32** %95, align 4
+  %97 = load i32* %96, align 4
+  %98 = load i32* %4, align 4
+  %99 = icmp eq i32 %97, %98
+  %100 = zext i1 %99 to i8
+  %toBool15.i.i = icmp ne i8 %100, 0
+  br i1 %toBool15.i.i, label %bb16.i.i, label %bb17.i.i
 
-bb16.i.i:		; preds = %bb14.i.i
-	%101 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0		; <i32**> [#uses=1]
-	%102 = load i32** %101, align 4		; <i32*> [#uses=1]
-	br label %_ZSt4findIN9__gnu_cxx17__normal_iteratorIPiSt6vectorIiSaIiEEEEiET_S7_S7_RKT0_.exit
+bb16.i.i:                                         ; preds = %bb14.i.i
+  %101 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %102 = load i32** %101, align 4
+  br label %_ZSt4findIN9__gnu_cxx17__normal_iteratorIPiSt6vectorIiSaIiEEEEiET_S7_S7_RKT0_.exit
 
-bb17.i.i:		; preds = %bb14.i.i
-	%103 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0		; <i32**> [#uses=1]
-	%104 = load i32** %103, align 4		; <i32*> [#uses=1]
-	%105 = getelementptr i32* %104, i64 1		; <i32*> [#uses=1]
-	%106 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0		; <i32**> [#uses=1]
-	store i32* %105, i32** %106, align 4
-	br label %bb18.i.i
+bb17.i.i:                                         ; preds = %bb14.i.i
+  %103 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %104 = load i32** %103, align 4
+  %105 = getelementptr i32* %104, i64 1
+  %106 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  store i32* %105, i32** %106, align 4
+  br label %bb18.i.i
 
-bb18.i.i:		; preds = %bb17.i.i, %bb13.i.i
-	%107 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0		; <i32**> [#uses=1]
-	%108 = load i32** %107, align 4		; <i32*> [#uses=1]
-	%109 = load i32* %108, align 4		; <i32> [#uses=1]
-	%110 = load i32* %4, align 4		; <i32> [#uses=1]
-	%111 = icmp eq i32 %109, %110		; <i1> [#uses=1]
-	%112 = zext i1 %111 to i8		; <i8> [#uses=1]
-	%toBool19.i.i = icmp ne i8 %112, 0		; <i1> [#uses=1]
-	br i1 %toBool19.i.i, label %bb20.i.i, label %bb21.i.i
+bb18.i.i:                                         ; preds = %bb17.i.i, %bb13.i.i
+  %107 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %108 = load i32** %107, align 4
+  %109 = load i32* %108, align 4
+  %110 = load i32* %4, align 4
+  %111 = icmp eq i32 %109, %110
+  %112 = zext i1 %111 to i8
+  %toBool19.i.i = icmp ne i8 %112, 0
+  br i1 %toBool19.i.i, label %bb20.i.i, label %bb21.i.i
 
-bb20.i.i:		; preds = %bb18.i.i
-	%113 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0		; <i32**> [#uses=1]
-	%114 = load i32** %113, align 4		; <i32*> [#uses=1]
-	br label %_ZSt4findIN9__gnu_cxx17__normal_iteratorIPiSt6vectorIiSaIiEEEEiET_S7_S7_RKT0_.exit
+bb20.i.i:                                         ; preds = %bb18.i.i
+  %113 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %114 = load i32** %113, align 4
+  br label %_ZSt4findIN9__gnu_cxx17__normal_iteratorIPiSt6vectorIiSaIiEEEEiET_S7_S7_RKT0_.exit
 
-bb21.i.i:		; preds = %bb18.i.i
-	%115 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0		; <i32**> [#uses=1]
-	%116 = load i32** %115, align 4		; <i32*> [#uses=1]
-	%117 = getelementptr i32* %116, i64 1		; <i32*> [#uses=1]
-	%118 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0		; <i32**> [#uses=1]
-	store i32* %117, i32** %118, align 4
-	br label %bb22.i.i
+bb21.i.i:                                         ; preds = %bb18.i.i
+  %115 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %116 = load i32** %115, align 4
+  %117 = getelementptr i32* %116, i64 1
+  %118 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  store i32* %117, i32** %118, align 4
+  br label %bb22.i.i
 
-bb22.i.i:		; preds = %bb21.i.i, %bb13.i.i
-	%119 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0		; <i32**> [#uses=1]
-	%120 = load i32** %119, align 4		; <i32*> [#uses=1]
-	%121 = load i32* %120, align 4		; <i32> [#uses=1]
-	%122 = load i32* %4, align 4		; <i32> [#uses=1]
-	%123 = icmp eq i32 %121, %122		; <i1> [#uses=1]
-	%124 = zext i1 %123 to i8		; <i8> [#uses=1]
-	%toBool23.i.i = icmp ne i8 %124, 0		; <i1> [#uses=1]
-	br i1 %toBool23.i.i, label %bb24.i.i, label %bb25.i.i
+bb22.i.i:                                         ; preds = %bb21.i.i, %bb13.i.i
+  %119 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %120 = load i32** %119, align 4
+  %121 = load i32* %120, align 4
+  %122 = load i32* %4, align 4
+  %123 = icmp eq i32 %121, %122
+  %124 = zext i1 %123 to i8
+  %toBool23.i.i = icmp ne i8 %124, 0
+  br i1 %toBool23.i.i, label %bb24.i.i, label %bb25.i.i
 
-bb24.i.i:		; preds = %bb22.i.i
-	%125 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0		; <i32**> [#uses=1]
-	%126 = load i32** %125, align 4		; <i32*> [#uses=1]
-	br label %_ZSt4findIN9__gnu_cxx17__normal_iteratorIPiSt6vectorIiSaIiEEEEiET_S7_S7_RKT0_.exit
+bb24.i.i:                                         ; preds = %bb22.i.i
+  %125 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %126 = load i32** %125, align 4
+  br label %_ZSt4findIN9__gnu_cxx17__normal_iteratorIPiSt6vectorIiSaIiEEEEiET_S7_S7_RKT0_.exit
 
-bb25.i.i:		; preds = %bb22.i.i
-	%127 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0		; <i32**> [#uses=1]
-	%128 = load i32** %127, align 4		; <i32*> [#uses=1]
-	%129 = getelementptr i32* %128, i64 1		; <i32*> [#uses=1]
-	%130 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0		; <i32**> [#uses=1]
-	store i32* %129, i32** %130, align 4
-	br label %bb26.i.i
+bb25.i.i:                                         ; preds = %bb22.i.i
+  %127 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %128 = load i32** %127, align 4
+  %129 = getelementptr i32* %128, i64 1
+  %130 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  store i32* %129, i32** %130, align 4
+  br label %bb26.i.i
 
-bb26.i.i:		; preds = %bb25.i.i, %bb13.i.i
-	%131 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__last_addr.i.i, i32 0, i32 0		; <i32**> [#uses=1]
-	%132 = load i32** %131, align 4		; <i32*> [#uses=1]
-	br label %_ZSt4findIN9__gnu_cxx17__normal_iteratorIPiSt6vectorIiSaIiEEEEiET_S7_S7_RKT0_.exit
+bb26.i.i:                                         ; preds = %bb25.i.i, %bb13.i.i
+  %131 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__last_addr.i.i, i32 0, i32 0
+  %132 = load i32** %131, align 4
+  br label %_ZSt4findIN9__gnu_cxx17__normal_iteratorIPiSt6vectorIiSaIiEEEEiET_S7_S7_RKT0_.exit
 
-_ZSt4findIN9__gnu_cxx17__normal_iteratorIPiSt6vectorIiSaIiEEEEiET_S7_S7_RKT0_.exit:		; preds = %bb26.i.i, %bb24.i.i, %bb20.i.i, %bb16.i.i, %bb10.i.i, %bb7.i.i, %bb4.i.i, %bb1.i.i
-	%.0.0.i.i = phi i32* [ %132, %bb26.i.i ], [ %126, %bb24.i.i ], [ %114, %bb20.i.i ], [ %102, %bb16.i.i ], [ %80, %bb10.i.i ], [ %68, %bb7.i.i ], [ %56, %bb4.i.i ], [ %44, %bb1.i.i ]		; <i32*> [#uses=1]
-	%tmp2.i.i = ptrtoint i32* %.0.0.i.i to i32		; <i32> [#uses=1]
-	%tmp1.i.i = inttoptr i32 %tmp2.i.i to i32*		; <i32*> [#uses=1]
-	%tmp4.i = ptrtoint i32* %tmp1.i.i to i32		; <i32> [#uses=1]
-	%tmp3.i = inttoptr i32 %tmp4.i to i32*		; <i32*> [#uses=1]
-	%tmp8.i = ptrtoint i32* %tmp3.i to i32		; <i32> [#uses=1]
-	%tmp6.i = inttoptr i32 %tmp8.i to i32*		; <i32*> [#uses=1]
-	%tmp12 = ptrtoint i32* %tmp6.i to i32		; <i32> [#uses=1]
-	%tmp10 = inttoptr i32 %tmp12 to i32*		; <i32*> [#uses=1]
-	%tmp16 = ptrtoint i32* %tmp10 to i32		; <i32> [#uses=1]
-	br label %return
+_ZSt4findIN9__gnu_cxx17__normal_iteratorIPiSt6vectorIiSaIiEEEEiET_S7_S7_RKT0_.exit: ; preds = %bb26.i.i, %bb24.i.i, %bb20.i.i, %bb16.i.i, %bb10.i.i, %bb7.i.i, %bb4.i.i, %bb1.i.i
+  %.0.0.i.i = phi i32* [ %132, %bb26.i.i ], [ %126, %bb24.i.i ], [ %114, %bb20.i.i ], [ %102, %bb16.i.i ], [ %80, %bb10.i.i ], [ %68, %bb7.i.i ], [ %56, %bb4.i.i ], [ %44, %bb1.i.i ]
+  %tmp2.i.i = ptrtoint i32* %.0.0.i.i to i32
+  %tmp1.i.i = inttoptr i32 %tmp2.i.i to i32*
+  %tmp4.i = ptrtoint i32* %tmp1.i.i to i32
+  %tmp3.i = inttoptr i32 %tmp4.i to i32*
+  %tmp8.i = ptrtoint i32* %tmp3.i to i32
+  %tmp6.i = inttoptr i32 %tmp8.i to i32*
+  %tmp12 = ptrtoint i32* %tmp6.i to i32
+  %tmp10 = inttoptr i32 %tmp12 to i32*
+  %tmp16 = ptrtoint i32* %tmp10 to i32
+  br label %return
 
-return:		; preds = %_ZSt4findIN9__gnu_cxx17__normal_iteratorIPiSt6vectorIiSaIiEEEEiET_S7_S7_RKT0_.exit
-	%tmp14 = inttoptr i32 %tmp16 to i32*		; <i32*> [#uses=1]
-	ret i32* %tmp14
+return:                                           ; preds = %_ZSt4findIN9__gnu_cxx17__normal_iteratorIPiSt6vectorIiSaIiEEEEiET_S7_S7_RKT0_.exit
+  %tmp14 = inttoptr i32 %tmp16 to i32*
+  ret i32* %tmp14
 }
 
-declare void @llvm.memcpy.i64(i8* nocapture, i8* nocapture, i64, i32) nounwind
+declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture, i64, i32, i1) nounwind
diff --git a/test/Transforms/InstCombine/memcpy-to-load.ll b/test/Transforms/InstCombine/memcpy-to-load.ll
index ebb8711af9..04aac98145 100644
--- a/test/Transforms/InstCombine/memcpy-to-load.ll
+++ b/test/Transforms/InstCombine/memcpy-to-load.ll
@@ -4,11 +4,10 @@ target triple = "i686-apple-darwin8"
 
 define void @foo(double* %X, double* %Y) {
 entry:
-	%"alloca point" = bitcast i32 0 to i32		; <i32> [#uses=0]
-	%tmp2 = bitcast double* %X to i8*		; <i8*> [#uses=1]
-	%tmp13 = bitcast double* %Y to i8*		; <i8*> [#uses=1]
-	call void @llvm.memcpy.i32( i8* %tmp2, i8* %tmp13, i32 8, i32 1 )
-	ret void
+  %tmp2 = bitcast double* %X to i8*
+  %tmp13 = bitcast double* %Y to i8*
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %tmp2, i8* %tmp13, i32 8, i32 1, i1 false)
+  ret void
 }
 
-declare void @llvm.memcpy.i32(i8*, i8*, i32, i32) nounwind 
+declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture, i32, i32, i1) nounwind
diff --git a/test/Transforms/InstCombine/memmove.ll b/test/Transforms/InstCombine/memmove.ll
index 1806cfcb54..4602c12680 100644
--- a/test/Transforms/InstCombine/memmove.ll
+++ b/test/Transforms/InstCombine/memmove.ll
@@ -1,24 +1,20 @@
 ; This test makes sure that memmove instructions are properly eliminated.
 ;
-; RUN: opt < %s -instcombine -S | \
-; RUN:    not grep {call void @llvm.memmove}
+; RUN: opt < %s -instcombine -S | not grep {call void @llvm.memmove}
 
 @S = internal constant [33 x i8] c"panic: restorelist inconsistency\00"		; <[33 x i8]*> [#uses=1]
 @h = constant [2 x i8] c"h\00"		; <[2 x i8]*> [#uses=1]
 @hel = constant [4 x i8] c"hel\00"		; <[4 x i8]*> [#uses=1]
 @hello_u = constant [8 x i8] c"hello_u\00"		; <[8 x i8]*> [#uses=1]
 
-
-declare void @llvm.memmove.i32(i8*, i8*, i32, i32)
-
 define void @test1(i8* %A, i8* %B, i32 %N) {
-	call void @llvm.memmove.i32( i8* %A, i8* %B, i32 0, i32 1 )
+	call void @llvm.memmove.p0i8.p0i8.i32(i8* %A, i8* %B, i32 0, i32 1, i1 false)
 	ret void
 }
 
 define void @test2(i8* %A, i32 %N) {
         ;; dest can't alias source since we can't write to source!
-	call void @llvm.memmove.i32( i8* %A, i8* getelementptr ([33 x i8]* @S, i32 0, i32 0), i32 %N, i32 1 )
+	call void @llvm.memmove.p0i8.p0i8.i32(i8* %A, i8* getelementptr inbounds ([33 x i8]* @S, i32 0, i32 0), i32 %N, i32 1, i1 false)
 	ret void
 }
 
@@ -28,15 +24,16 @@ define i32 @test3() {
 	%hello_u_p = getelementptr [8 x i8]* @hello_u, i32 0, i32 0		; <i8*> [#uses=1]
 	%target = alloca [1024 x i8]		; <[1024 x i8]*> [#uses=1]
 	%target_p = getelementptr [1024 x i8]* %target, i32 0, i32 0		; <i8*> [#uses=3]
-	call void @llvm.memmove.i32( i8* %target_p, i8* %h_p, i32 2, i32 2 )
-	call void @llvm.memmove.i32( i8* %target_p, i8* %hel_p, i32 4, i32 4 )
-	call void @llvm.memmove.i32( i8* %target_p, i8* %hello_u_p, i32 8, i32 8 )
+        call void @llvm.memmove.p0i8.p0i8.i32(i8* %target_p, i8* %h_p, i32 2, i32 2, i1 false)
+        call void @llvm.memmove.p0i8.p0i8.i32(i8* %target_p, i8* %hel_p, i32 4, i32 4, i1 false)
+        call void @llvm.memmove.p0i8.p0i8.i32(i8* %target_p, i8* %hello_u_p, i32 8, i32 8, i1 false)
 	ret i32 0
 }
 
 ; PR2370
 define void @test4(i8* %a) {
-        tail call void @llvm.memmove.i32( i8* %a, i8* %a, i32 100, i32 1 )
-        ret void
+  tail call void @llvm.memmove.p0i8.p0i8.i32(i8* %a, i8* %a, i32 100, i32 1, i1 false)
+  ret void
 }
 
+declare void @llvm.memmove.p0i8.p0i8.i32(i8* nocapture, i8* nocapture, i32, i32, i1) nounwind
diff --git a/test/Transforms/InstCombine/memset.ll b/test/Transforms/InstCombine/memset.ll
index 8e85694d9a..7f7bc9f86e 100644
--- a/test/Transforms/InstCombine/memset.ll
+++ b/test/Transforms/InstCombine/memset.ll
@@ -1,15 +1,14 @@
 ; RUN: opt < %s -instcombine -S | not grep {call.*llvm.memset}
 
-declare void @llvm.memset.i32(i8*, i8, i32, i32)
-
 define i32 @main() {
-	%target = alloca [1024 x i8]		; <[1024 x i8]*> [#uses=1]
-	%target_p = getelementptr [1024 x i8]* %target, i32 0, i32 0		; <i8*> [#uses=5]
-	call void @llvm.memset.i32( i8* %target_p, i8 1, i32 0, i32 1 )
-	call void @llvm.memset.i32( i8* %target_p, i8 1, i32 1, i32 1 )
-	call void @llvm.memset.i32( i8* %target_p, i8 1, i32 2, i32 2 )
-	call void @llvm.memset.i32( i8* %target_p, i8 1, i32 4, i32 4 )
-	call void @llvm.memset.i32( i8* %target_p, i8 1, i32 8, i32 8 )
-	ret i32 0
+  %target = alloca [1024 x i8]
+  %target_p = getelementptr [1024 x i8]* %target, i32 0, i32 0
+  call void @llvm.memset.p0i8.i32(i8* %target_p, i8 1, i32 0, i32 1, i1 false)
+  call void @llvm.memset.p0i8.i32(i8* %target_p, i8 1, i32 1, i32 1, i1 false)
+  call void @llvm.memset.p0i8.i32(i8* %target_p, i8 1, i32 2, i32 2, i1 false)
+  call void @llvm.memset.p0i8.i32(i8* %target_p, i8 1, i32 4, i32 4, i1 false)
+  call void @llvm.memset.p0i8.i32(i8* %target_p, i8 1, i32 8, i32 8, i1 false)
+  ret i32 0
 }
 
+declare void @llvm.memset.p0i8.i32(i8* nocapture, i8, i32, i32, i1) nounwind
diff --git a/test/Transforms/InstCombine/stack-overalign.ll b/test/Transforms/InstCombine/stack-overalign.ll
index 88b4114d58..2fc84140b3 100644
--- a/test/Transforms/InstCombine/stack-overalign.ll
+++ b/test/Transforms/InstCombine/stack-overalign.ll
@@ -17,13 +17,13 @@
 
 define void @foo() nounwind {
 entry:
-	%src = alloca [1024 x i8], align 1
-	%src1 = getelementptr [1024 x i8]* %src, i32 0, i32 0
-	call void @llvm.memcpy.i32(i8* getelementptr ([1024 x i8]* @dst, i32 0, i32 0), i8* %src1, i32 1024, i32 1)
-	call void @frob(i8* %src1) nounwind
-	ret void
+  %src = alloca [1024 x i8], align 1
+  %src1 = getelementptr [1024 x i8]* %src, i32 0, i32 0
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* getelementptr inbounds ([1024 x i8]* @dst, i32 0, i32 0), i8* %src1, i32 1024, i32 1, i1 false)
+  call void @frob(i8* %src1) nounwind
+  ret void
 }
 
-declare void @llvm.memcpy.i32(i8* nocapture, i8* nocapture, i32, i32) nounwind
-
 declare void @frob(i8*)
+
+declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture, i32, i32, i1) nounwind
diff --git a/test/Transforms/MemCpyOpt/2008-02-24-MultipleUseofSRet.ll b/test/Transforms/MemCpyOpt/2008-02-24-MultipleUseofSRet.ll
index 9f1e280467..b95ad91a36 100644
--- a/test/Transforms/MemCpyOpt/2008-02-24-MultipleUseofSRet.ll
+++ b/test/Transforms/MemCpyOpt/2008-02-24-MultipleUseofSRet.ll
@@ -4,31 +4,33 @@
 target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f32:32:32-f64:32:64-v64:64:64-v128:128:128-a0:0:64-f80:32:32"
 target triple = "i386-pc-linux-gnu"
 
-define internal fastcc void @initialize({ x86_fp80, x86_fp80 }* noalias sret  %agg.result) nounwind  {
+%0 = type { x86_fp80, x86_fp80 }
+
+define internal fastcc void @initialize(%0* noalias sret %agg.result) nounwind {
 entry:
-	%agg.result.03 = getelementptr { x86_fp80, x86_fp80 }* %agg.result, i32 0, i32 0		; <x86_fp80*> [#uses=1]
-	store x86_fp80 0xK00000000000000000000, x86_fp80* %agg.result.03
-	%agg.result.15 = getelementptr { x86_fp80, x86_fp80 }* %agg.result, i32 0, i32 1		; <x86_fp80*> [#uses=1]
-	store x86_fp80 0xK00000000000000000000, x86_fp80* %agg.result.15
-	ret void
+  %agg.result.03 = getelementptr %0* %agg.result, i32 0, i32 0
+  store x86_fp80 0xK00000000000000000000, x86_fp80* %agg.result.03
+  %agg.result.15 = getelementptr %0* %agg.result, i32 0, i32 1
+  store x86_fp80 0xK00000000000000000000, x86_fp80* %agg.result.15
+  ret void
 }
 
-declare fastcc x86_fp80 @passed_uninitialized({ x86_fp80, x86_fp80 }* %x) nounwind
+declare fastcc x86_fp80 @passed_uninitialized(%0*) nounwind
 
-define fastcc void @badly_optimized() nounwind  {
+define fastcc void @badly_optimized() nounwind {
 entry:
-	%z = alloca { x86_fp80, x86_fp80 }		; <{ x86_fp80, x86_fp80 }*> [#uses=2]
-	%tmp = alloca { x86_fp80, x86_fp80 }		; <{ x86_fp80, x86_fp80 }*> [#uses=2]
-	%memtmp = alloca { x86_fp80, x86_fp80 }, align 8		; <{ x86_fp80, x86_fp80 }*> [#uses=2]
-	call fastcc void @initialize( { x86_fp80, x86_fp80 }* noalias sret  %memtmp )
-	%tmp1 = bitcast { x86_fp80, x86_fp80 }* %tmp to i8*		; <i8*> [#uses=1]
-	%memtmp2 = bitcast { x86_fp80, x86_fp80 }* %memtmp to i8*		; <i8*> [#uses=1]
-	call void @llvm.memcpy.i32( i8* %tmp1, i8* %memtmp2, i32 24, i32 8 )
-	%z3 = bitcast { x86_fp80, x86_fp80 }* %z to i8*		; <i8*> [#uses=1]
-	%tmp4 = bitcast { x86_fp80, x86_fp80 }* %tmp to i8*		; <i8*> [#uses=1]
-	call void @llvm.memcpy.i32( i8* %z3, i8* %tmp4, i32 24, i32 8 )
-	%tmp5 = call fastcc x86_fp80 @passed_uninitialized( { x86_fp80, x86_fp80 }* %z )		; <x86_fp80> [#uses=0]
-	ret void
+  %z = alloca %0
+  %tmp = alloca %0
+  %memtmp = alloca %0, align 8
+  call fastcc void @initialize(%0* noalias sret %memtmp)
+  %tmp1 = bitcast %0* %tmp to i8*
+  %memtmp2 = bitcast %0* %memtmp to i8*
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %tmp1, i8* %memtmp2, i32 24, i32 8, i1 false)
+  %z3 = bitcast %0* %z to i8*
+  %tmp4 = bitcast %0* %tmp to i8*
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %z3, i8* %tmp4, i32 24, i32 8, i1 false)
+  %tmp5 = call fastcc x86_fp80 @passed_uninitialized(%0* %z)
+  ret void
 }
 
-declare void @llvm.memcpy.i32(i8*, i8*, i32, i32) nounwind
+declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture, i32, i32, i1) nounwind
diff --git a/test/Transforms/MemCpyOpt/2008-03-13-ReturnSlotBitcast.ll b/test/Transforms/MemCpyOpt/2008-03-13-ReturnSlotBitcast.ll
index 418761e936..24cf576a08 100644
--- a/test/Transforms/MemCpyOpt/2008-03-13-ReturnSlotBitcast.ll
+++ b/test/Transforms/MemCpyOpt/2008-03-13-ReturnSlotBitcast.ll
@@ -1,20 +1,22 @@
 ; RUN: opt < %s -basicaa -memcpyopt -S | not grep {call.*memcpy.}
 target datalayout = "E-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64"
-	%a = type { i32 }
-	%b = type { float }
 
-declare void @llvm.memcpy.i32(i8*, i8*, i32, i32) nounwind 
+%a = type { i32 }
+%b = type { float }
+
 declare void @g(%a*)
 
 define float @f() {
 entry:
-	%a_var = alloca %a
-	%b_var = alloca %b
-	call void @g(%a *%a_var)
-	%a_i8 = bitcast %a* %a_var to i8*
-	%b_i8 = bitcast %b* %b_var to i8*
-	call void @llvm.memcpy.i32(i8* %b_i8, i8* %a_i8, i32 4, i32 4)
-	%tmp1 = getelementptr %b* %b_var, i32 0, i32 0
-	%tmp2 = load float* %tmp1
-	ret float %tmp2
+  %a_var = alloca %a
+  %b_var = alloca %b
+  call void @g(%a* %a_var)
+  %a_i8 = bitcast %a* %a_var to i8*
+  %b_i8 = bitcast %b* %b_var to i8*
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %b_i8, i8* %a_i8, i32 4, i32 4, i1 false)
+  %tmp1 = getelementptr %b* %b_var, i32 0, i32 0
+  %tmp2 = load float* %tmp1
+  ret float %tmp2
 }
+
+declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture, i32, i32, i1) nounwind
diff --git a/test/Transforms/MemCpyOpt/memcpy.ll b/test/Transforms/MemCpyOpt/memcpy.ll
index fb979135e0..12519ef76c 100644
--- a/test/Transforms/MemCpyOpt/memcpy.ll
+++ b/test/Transforms/MemCpyOpt/memcpy.ll
@@ -3,17 +3,21 @@
 target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f32:32:32-f64:32:64-v64:64:64-v128:128:128-a0:0:64-f80:128:128"
 target triple = "i686-apple-darwin9"
 
+%0 = type { x86_fp80, x86_fp80 }
+%1 = type { i32, i32 }
+
 define void @test1({ x86_fp80, x86_fp80 }* sret  %agg.result, x86_fp80 %z.0, x86_fp80 %z.1) nounwind  {
 entry:
-	%tmp2 = alloca { x86_fp80, x86_fp80 }		; <{ x86_fp80, x86_fp80 }*> [#uses=1]
-	%memtmp = alloca { x86_fp80, x86_fp80 }, align 16		; <{ x86_fp80, x86_fp80 }*> [#uses=2]
-	%tmp5 = fsub x86_fp80 0xK80000000000000000000, %z.1		; <x86_fp80> [#uses=1]
-	call void @ccoshl( { x86_fp80, x86_fp80 }* sret  %memtmp, x86_fp80 %tmp5, x86_fp80 %z.0 ) nounwind 
-	%tmp219 = bitcast { x86_fp80, x86_fp80 }* %tmp2 to i8*		; <i8*> [#uses=2]
-	%memtmp20 = bitcast { x86_fp80, x86_fp80 }* %memtmp to i8*		; <i8*> [#uses=1]
-	call void @llvm.memcpy.i32( i8* %tmp219, i8* %memtmp20, i32 32, i32 16 )
-	%agg.result21 = bitcast { x86_fp80, x86_fp80 }* %agg.result to i8*		; <i8*> [#uses=1]
-	call void @llvm.memcpy.i32( i8* %agg.result21, i8* %tmp219, i32 32, i32 16 )
+  %tmp2 = alloca %0
+  %memtmp = alloca %0, align 16
+  %tmp5 = fsub x86_fp80 0xK80000000000000000000, %z.1
+  call void @ccoshl(%0* sret %memtmp, x86_fp80 %tmp5, x86_fp80 %z.0) nounwind
+  %tmp219 = bitcast %0* %tmp2 to i8*
+  %memtmp20 = bitcast %0* %memtmp to i8*
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %tmp219, i8* %memtmp20, i32 32, i32 16, i1 false)
+  %agg.result21 = bitcast %0* %agg.result to i8*
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %agg.result21, i8* %tmp219, i32 32, i32 16, i1 false)
+  ret void
 
 ; Check that one of the memcpy's are removed.
 ;; FIXME: PR 8643 We should be able to eliminate the last memcpy here.
@@ -23,22 +27,19 @@ entry:
 ; CHECK: call void @llvm.memcpy
 ; CHECK-NOT: llvm.memcpy
 ; CHECK: ret void
-	ret void
 }
 
 declare void @ccoshl({ x86_fp80, x86_fp80 }* sret , x86_fp80, x86_fp80) nounwind 
 
-declare void @llvm.memcpy.i32(i8*, i8*, i32, i32) nounwind 
-
 
 ; The intermediate alloca and one of the memcpy's should be eliminated, the
 ; other should be related with a memmove.
 define void @test2(i8* %P, i8* %Q) nounwind  {
-	%memtmp = alloca { x86_fp80, x86_fp80 }, align 16
-	%R = bitcast { x86_fp80, x86_fp80 }* %memtmp to i8*
-	call void @llvm.memcpy.i32( i8* %R, i8* %P, i32 32, i32 16 )
-	call void @llvm.memcpy.i32( i8* %Q, i8* %R, i32 32, i32 16 )
-        ret void
+  %memtmp = alloca %0, align 16
+  %R = bitcast %0* %memtmp to i8*
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %R, i8* %P, i32 32, i32 16, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %Q, i8* %R, i32 32, i32 16, i1 false)
+  ret void
         
 ; CHECK: @test2
 ; CHECK-NEXT: call void @llvm.memmove{{.*}}(i8* %Q, i8* %P
@@ -51,12 +52,12 @@ define void @test2(i8* %P, i8* %Q) nounwind  {
 @x = external global { x86_fp80, x86_fp80 }
 
 define void @test3({ x86_fp80, x86_fp80 }* noalias sret %agg.result) nounwind  {
-	%x.0 = alloca { x86_fp80, x86_fp80 }
-	%x.01 = bitcast { x86_fp80, x86_fp80 }* %x.0 to i8*
-	call void @llvm.memcpy.i32( i8* %x.01, i8* bitcast ({ x86_fp80, x86_fp80 }* @x to i8*), i32 32, i32 16 )
-	%agg.result2 = bitcast { x86_fp80, x86_fp80 }* %agg.result to i8*
-	call void @llvm.memcpy.i32( i8* %agg.result2, i8* %x.01, i32 32, i32 16 )
-	ret void
+  %x.0 = alloca %0
+  %x.01 = bitcast %0* %x.0 to i8*
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %x.01, i8* bitcast (%0* @x to i8*), i32 32, i32 16, i1 false)
+  %agg.result2 = bitcast %0* %agg.result to i8*
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %agg.result2, i8* %x.01, i32 32, i32 16, i1 false)
+  ret void
 ; CHECK: @test3
 ; CHECK-NEXT: %agg.result2 = bitcast 
 ; CHECK-NEXT: call void @llvm.memcpy
@@ -66,10 +67,10 @@ define void @test3({ x86_fp80, x86_fp80 }* noalias sret %agg.result) nounwind  {
 
 ; PR8644
 define void @test4(i8 *%P) {
-  %A = alloca {i32, i32}
-  %a = bitcast {i32, i32}* %A to i8*
+  %A = alloca %1
+  %a = bitcast %1* %A to i8*
   call void @llvm.memcpy.p0i8.p0i8.i64(i8* %a, i8* %P, i64 8, i32 4, i1 false)
-  call void @test4a(i8* byval align 1 %a) 
+  call void @test4a(i8* byval align 1 %a)
   ret void
 ; CHECK: @test4
 ; CHECK-NEXT: call void @test4a(
@@ -127,4 +128,5 @@ entry:
 
 declare i32 @g(%struct.p* byval align 8)
 
+declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture, i32, i32, i1) nounwind
 
diff --git a/test/Transforms/MemCpyOpt/memmove.ll b/test/Transforms/MemCpyOpt/memmove.ll
index 8d3fbd2b6d..7f1667a455 100644
--- a/test/Transforms/MemCpyOpt/memmove.ll
+++ b/test/Transforms/MemCpyOpt/memmove.ll
@@ -4,7 +4,7 @@
 target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128"
 target triple = "x86_64-apple-darwin9.0"
 
-declare void @llvm.memmove.i64(i8* nocapture, i8* nocapture, i64, i32) nounwind
+declare void @llvm.memmove.p0i8.p0i8.i64(i8* nocapture, i8* nocapture, i64, i32, i1) nounwind
 
 define i8* @test1(i8* nocapture %src) nounwind {
 entry:
@@ -13,8 +13,8 @@ entry:
 
   %malloccall = tail call i8* @malloc(i32 trunc (i64 mul nuw (i64 ptrtoint (i8* getelementptr (i8* null, i32 1) to i64), i64 13) to i32))
   %call3 = bitcast i8* %malloccall to [13 x i8]*
-  %call3.sub = getelementptr inbounds [13 x i8]* %call3, i64 0, i64 0 ; <i8*> [#uses=2]
-  tail call void @llvm.memmove.i64(i8* %call3.sub, i8* %src, i64 13, i32 1)
+  %call3.sub = getelementptr inbounds [13 x i8]* %call3, i64 0, i64 0
+  tail call void @llvm.memmove.p0i8.p0i8.i64(i8* %call3.sub, i8* %src, i64 13, i32 1, i1 false)
   ret i8* %call3.sub
 }
 declare noalias i8* @malloc(i32)
@@ -24,8 +24,8 @@ define void @test2(i8* %P) nounwind {
 entry:
 ; CHECK: @test2
 ; CHECK: call void @llvm.memcpy
-  %add.ptr = getelementptr i8* %P, i64 16         ; <i8*> [#uses=1]
-  tail call void @llvm.memmove.i64(i8* %P, i8* %add.ptr, i64 16, i32 1)
+  %add.ptr = getelementptr i8* %P, i64 16
+  tail call void @llvm.memmove.p0i8.p0i8.i64(i8* %P, i8* %add.ptr, i64 16, i32 1, i1 false)
   ret void
 }
 
@@ -34,7 +34,7 @@ define void @test3(i8* %P) nounwind {
 entry:
 ; CHECK: @test3
 ; CHECK: call void @llvm.memmove
-  %add.ptr = getelementptr i8* %P, i64 16         ; <i8*> [#uses=1]
-  tail call void @llvm.memmove.i64(i8* %P, i8* %add.ptr, i64 17, i32 1)
+  %add.ptr = getelementptr i8* %P, i64 16
+  tail call void @llvm.memmove.p0i8.p0i8.i64(i8* %P, i8* %add.ptr, i64 17, i32 1, i1 false)
   ret void
 }
diff --git a/test/Transforms/MemCpyOpt/sret.ll b/test/Transforms/MemCpyOpt/sret.ll
index ddfd0fd1fc..8eac7da798 100644
--- a/test/Transforms/MemCpyOpt/sret.ll
+++ b/test/Transforms/MemCpyOpt/sret.ll
@@ -3,26 +3,28 @@
 target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f32:32:32-f64:32:64-v64:64:64-v128:128:128-a0:0:64-f80:128:128"
 target triple = "i686-apple-darwin9"
 
-define void @ccosl({ x86_fp80, x86_fp80 }* noalias sret  %agg.result, { x86_fp80, x86_fp80 }* byval  align 8 %z) nounwind  {
+%0 = type { x86_fp80, x86_fp80 }
+
+define void @ccosl(%0* noalias sret %agg.result, %0* byval align 8 %z) nounwind {
 entry:
-	%iz = alloca { x86_fp80, x86_fp80 }		; <{ x86_fp80, x86_fp80 }*> [#uses=3]
-	%memtmp = alloca { x86_fp80, x86_fp80 }, align 16		; <{ x86_fp80, x86_fp80 }*> [#uses=2]
-	%tmp1 = getelementptr { x86_fp80, x86_fp80 }* %z, i32 0, i32 1		; <x86_fp80*> [#uses=1]
-	%tmp2 = load x86_fp80* %tmp1, align 16		; <x86_fp80> [#uses=1]
-	%tmp3 = fsub x86_fp80 0xK80000000000000000000, %tmp2		; <x86_fp80> [#uses=1]
-	%tmp4 = getelementptr { x86_fp80, x86_fp80 }* %iz, i32 0, i32 1		; <x86_fp80*> [#uses=1]
-	%real = getelementptr { x86_fp80, x86_fp80 }* %iz, i32 0, i32 0		; <x86_fp80*> [#uses=1]
-	%tmp7 = getelementptr { x86_fp80, x86_fp80 }* %z, i32 0, i32 0		; <x86_fp80*> [#uses=1]
-	%tmp8 = load x86_fp80* %tmp7, align 16		; <x86_fp80> [#uses=1]
-	store x86_fp80 %tmp3, x86_fp80* %real, align 16
-	store x86_fp80 %tmp8, x86_fp80* %tmp4, align 16
-	call void @ccoshl( { x86_fp80, x86_fp80 }* noalias sret  %memtmp, { x86_fp80, x86_fp80 }* byval align 8 %iz ) nounwind 
-	%memtmp14 = bitcast { x86_fp80, x86_fp80 }* %memtmp to i8*		; <i8*> [#uses=1]
-	%agg.result15 = bitcast { x86_fp80, x86_fp80 }* %agg.result to i8*		; <i8*> [#uses=1]
-	call void @llvm.memcpy.i32( i8* %agg.result15, i8* %memtmp14, i32 32, i32 16 )
-	ret void
+  %iz = alloca %0
+  %memtmp = alloca %0, align 16
+  %tmp1 = getelementptr %0* %z, i32 0, i32 1
+  %tmp2 = load x86_fp80* %tmp1, align 16
+  %tmp3 = fsub x86_fp80 0xK80000000000000000000, %tmp2
+  %tmp4 = getelementptr %0* %iz, i32 0, i32 1
+  %real = getelementptr %0* %iz, i32 0, i32 0
+  %tmp7 = getelementptr %0* %z, i32 0, i32 0
+  %tmp8 = load x86_fp80* %tmp7, align 16
+  store x86_fp80 %tmp3, x86_fp80* %real, align 16
+  store x86_fp80 %tmp8, x86_fp80* %tmp4, align 16
+  call void @ccoshl(%0* noalias sret %memtmp, %0* byval align 8 %iz) nounwind
+  %memtmp14 = bitcast %0* %memtmp to i8*
+  %agg.result15 = bitcast %0* %agg.result to i8*
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %agg.result15, i8* %memtmp14, i32 32, i32 16, i1 false)
+  ret void
 }
 
-declare void @ccoshl({ x86_fp80, x86_fp80 }* noalias sret , { x86_fp80, x86_fp80 }* byval ) nounwind 
+declare void @ccoshl(%0* noalias sret, %0* byval) nounwind
 
-declare void @llvm.memcpy.i32(i8*, i8*, i32, i32) nounwind 
+declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture, i32, i32, i1) nounwind
diff --git a/test/Transforms/ScalarRepl/2007-05-24-LargeAggregate.ll b/test/Transforms/ScalarRepl/2007-05-24-LargeAggregate.ll
deleted file mode 100644
index e67b6106cf..0000000000
--- a/test/Transforms/ScalarRepl/2007-05-24-LargeAggregate.ll
+++ /dev/null
@@ -1,27 +0,0 @@
-; RUN: opt < %s -scalarrepl -S | grep {alloca.*client_t}
-; PR1446
-target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f32:32:32-f64:32:64-v64:64:64-v128:128:128-a0:0:64"
-target triple = "i686-pc-linux-gnu"
-
-	%struct.clientSnapshot_t = type { i32, [32 x i8], %struct.playerState_t, i32, i32, i32, i32, i32 }
-	%struct.client_t = type { i32, [1024 x i8], [64 x [1024 x i8]], i32, i32, i32, i32, i32, i32, %struct.usercmd_t, i32, i32, [1024 x i8], %struct.sharedEntity_t*, [32 x i8], [64 x i8], i32, i32, i32, i32, i32, i32, [8 x i8*], [8 x i32], i32, i32, i32, i32, i32, i32, i32, i32, i32, [32 x %struct.clientSnapshot_t], i32, i32, i32, i32, i32, %struct.netchan_t, %struct.netchan_buffer_t*, %struct.netchan_buffer_t**, i32, [1025 x i32] }
-	%struct.entityShared_t = type { %struct.entityState_t, i32, i32, i32, i32, i32, [3 x float], [3 x float], i32, [3 x float], [3 x float], [3 x float], [3 x float], i32 }
-	%struct.entityState_t = type { i32, i32, i32, %struct.trajectory_t, %struct.trajectory_t, i32, i32, [3 x float], [3 x float], [3 x float], [3 x float], i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32, i32 }
-	%struct.msg_t = type { i32, i32, i32, i8*, i32, i32, i32, i32 }
-	%struct.netadr_t = type { i32, [4 x i8], [10 x i8], i16 }
-	%struct.netchan_buffer_t = type { %struct.msg_t, [16384 x i8], %struct.netchan_buffer_t* }
-	%struct.netchan_t = type { i32, i32, %struct.netadr_t, i32, i32, i32, i32, i32, [16384 x i8], i32, i32, i32, [16384 x i8] }
-	%struct.playerState_t = type { i32, i32, i32, i32, i32, [3 x float], [3 x float], i32, i32, i32, [3 x i32], i32, i32, i32, i32, i32, i32, [3 x float], i32, i32, [2 x i32], [2 x i32], i32, i32, i32, i32, i32, i32, [3 x float], i32, i32, i32, i32, i32, [16 x i32], [16 x i32], [16 x i32], [16 x i32], i32, i32, i32, i32, i32, i32, i32 }
-	%struct.sharedEntity_t = type { %struct.entityState_t, %struct.entityShared_t }
-	%struct.trajectory_t = type { i32, i32, i32, [3 x float], [3 x float] }
-	%struct.usercmd_t = type { i32, [3 x i32], i32, i8, i8, i8, i8 }
-
-declare void @llvm.memcpy.i32(i8*, i8*, i32, i32)
-
-define void @SV_DirectConnect(i64 %from.0.0, i64 %from.0.1, i32 %from.1) {
-entry:
-	%temp = alloca %struct.client_t, align 16		; <%struct.client_t*> [#uses=1]
-	%temp586 = bitcast %struct.client_t* %temp to i8*		; <i8*> [#uses=1]
-	call void @llvm.memcpy.i32( i8* null, i8* %temp586, i32 121596, i32 0 )
-	unreachable
-}
diff --git a/test/Transforms/ScalarRepl/2007-05-29-MemcpyPreserve.ll b/test/Transforms/ScalarRepl/2007-05-29-MemcpyPreserve.ll
index f1b8b80f8f..cf96c4cc20 100644
--- a/test/Transforms/ScalarRepl/2007-05-29-MemcpyPreserve.ll
+++ b/test/Transforms/ScalarRepl/2007-05-29-MemcpyPreserve.ll
@@ -3,21 +3,22 @@
 
 target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f32:32:32-f64:32:64-v64:64:64-v128:128:128-a0:0:64"
 target triple = "i686-apple-darwin8"
-	%struct.LongestMember = type { i8, i32 }
-	%struct.MyString = type { i32 }
-	%struct.UnionType = type { %struct.LongestMember }
+
+%struct.LongestMember = type { i8, i32 }
+%struct.MyString = type { i32 }
+%struct.UnionType = type { %struct.LongestMember }
 
 define void @_Z4testP9UnionTypePS0_(%struct.UnionType* %p, %struct.UnionType** %pointerToUnion) {
 entry:
-	%tmp = alloca %struct.UnionType, align 8		; <%struct.UnionType*> [#uses=2]
-	%tmp2 = getelementptr %struct.UnionType* %tmp, i32 0, i32 0, i32 0		; <i8*> [#uses=1]
-	%tmp13 = getelementptr %struct.UnionType* %p, i32 0, i32 0, i32 0		; <i8*> [#uses=1]
-	call void @llvm.memcpy.i32( i8* %tmp2, i8* %tmp13, i32 8, i32 0 )
-	%tmp5 = load %struct.UnionType** %pointerToUnion		; <%struct.UnionType*> [#uses=1]
-	%tmp56 = getelementptr %struct.UnionType* %tmp5, i32 0, i32 0, i32 0		; <i8*> [#uses=1]
-	%tmp7 = getelementptr %struct.UnionType* %tmp, i32 0, i32 0, i32 0		; <i8*> [#uses=1]
-	call void @llvm.memcpy.i32( i8* %tmp56, i8* %tmp7, i32 8, i32 0 )
-	ret void
+  %tmp = alloca %struct.UnionType, align 8
+  %tmp2 = getelementptr %struct.UnionType* %tmp, i32 0, i32 0, i32 0
+  %tmp13 = getelementptr %struct.UnionType* %p, i32 0, i32 0, i32 0
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %tmp2, i8* %tmp13, i32 8, i32 0, i1 false)
+  %tmp5 = load %struct.UnionType** %pointerToUnion
+  %tmp56 = getelementptr %struct.UnionType* %tmp5, i32 0, i32 0, i32 0
+  %tmp7 = getelementptr %struct.UnionType* %tmp, i32 0, i32 0, i32 0
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %tmp56, i8* %tmp7, i32 8, i32 0, i1 false)
+  ret void
 }
 
-declare void @llvm.memcpy.i32(i8*, i8*, i32, i32)
+declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture, i32, i32, i1) nounwind
diff --git a/test/Transforms/ScalarRepl/2008-06-22-LargeArray.ll b/test/Transforms/ScalarRepl/2008-06-22-LargeArray.ll
index b704727c11..71ba601833 100644
--- a/test/Transforms/ScalarRepl/2008-06-22-LargeArray.ll
+++ b/test/Transforms/ScalarRepl/2008-06-22-LargeArray.ll
@@ -4,14 +4,14 @@
 target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f32:32:32-f64:32:64-v64:64:64-v128:128:128-a0:0:64-f80:128:128"
 target triple = "i386-apple-darwin8"
 
-define void @memtest1(i8* %dst, i8* %src) nounwind  {
+define void @memtest1(i8* %dst, i8* %src) nounwind {
 entry:
-	%temp = alloca [200 x i8]		; <[100 x i8]*> [#uses=2]
-	%temp1 = bitcast [200 x i8]* %temp to i8*		; <i8*> [#uses=1]
-	call void @llvm.memcpy.i32( i8* %temp1, i8* %src, i32 200, i32 1 )
-	%temp3 = bitcast [200 x i8]* %temp to i8*		; <i8*> [#uses=1]
-	call void @llvm.memcpy.i32( i8* %dst, i8* %temp3, i32 200, i32 1 )
-	ret void
+  %temp = alloca [200 x i8]
+  %temp1 = bitcast [200 x i8]* %temp to i8*
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %temp1, i8* %src, i32 200, i32 1, i1 false)
+  %temp3 = bitcast [200 x i8]* %temp to i8*
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dst, i8* %temp3, i32 200, i32 1, i1 false)
+  ret void
 }
 
-declare void @llvm.memcpy.i32(i8*, i8*, i32, i32) nounwind 
+declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture, i32, i32, i1) nounwind
diff --git a/test/Transforms/ScalarRepl/2008-08-22-out-of-range-array-promote.ll b/test/Transforms/ScalarRepl/2008-08-22-out-of-range-array-promote.ll
index 1df01c1f4e..7cccb19714 100644
--- a/test/Transforms/ScalarRepl/2008-08-22-out-of-range-array-promote.ll
+++ b/test/Transforms/ScalarRepl/2008-08-22-out-of-range-array-promote.ll
@@ -2,21 +2,22 @@
 ; PR2423
 target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f32:32:32-f64:32:64-v64:64:64-v128:128:128-a0:0:64-f80:128:128"
 target triple = "i386-apple-darwin8"
-	%struct.x = type { [1 x i32], i32, i32 }
+
+%struct.x = type { [1 x i32], i32, i32 }
 
 define i32 @b() nounwind {
 entry:
-	%s = alloca %struct.x		; <%struct.x*> [#uses=2]
-	%r = alloca %struct.x		; <%struct.x*> [#uses=2]
-	call i32 @a( %struct.x* %s ) nounwind		; <i32>:0 [#uses=0]
-	%r1 = bitcast %struct.x* %r to i8*		; <i8*> [#uses=1]
-	%s2 = bitcast %struct.x* %s to i8*		; <i8*> [#uses=1]
-	call void @llvm.memcpy.i32( i8* %r1, i8* %s2, i32 12, i32 8 )
-	getelementptr %struct.x* %r, i32 0, i32 0, i32 1		; <i32*>:1 [#uses=1]
-	load i32* %1, align 4		; <i32>:2 [#uses=1]
-	ret i32 %2
+  %s = alloca %struct.x
+  %r = alloca %struct.x
+  %0 = call i32 @a(%struct.x* %s) nounwind
+  %r1 = bitcast %struct.x* %r to i8*
+  %s2 = bitcast %struct.x* %s to i8*
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %r1, i8* %s2, i32 12, i32 8, i1 false)
+  %1 = getelementptr %struct.x* %r, i32 0, i32 0, i32 1
+  %2 = load i32* %1, align 4
+  ret i32 %2
 }
 
 declare i32 @a(%struct.x*)
 
-declare void @llvm.memcpy.i32(i8*, i8*, i32, i32) nounwind
+declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture, i32, i32, i1) nounwind
diff --git a/test/Transforms/ScalarRepl/2008-09-22-vector-gep.ll b/test/Transforms/ScalarRepl/2008-09-22-vector-gep.ll
index e32e6835fc..e7a58f16e2 100644
--- a/test/Transforms/ScalarRepl/2008-09-22-vector-gep.ll
+++ b/test/Transforms/ScalarRepl/2008-09-22-vector-gep.ll
@@ -18,8 +18,8 @@ entry:
         ; because the type of the first element in %struct.two is i8.
 	%tmpS = getelementptr %struct.two* %S, i32 0, i32 0, i32 0 
 	%tmpD = bitcast %struct.two* %D to i8*
-        call void @llvm.memmove.i32(i8* %tmpD, i8* %tmpS, i32 4, i32 1)
+        call void @llvm.memmove.p0i8.p0i8.i32(i8* %tmpD, i8* %tmpS, i32 4, i32 1, i1 false)
         ret void
 }
 
-declare void @llvm.memmove.i32(i8*, i8*, i32, i32) nounwind
+declare void @llvm.memmove.p0i8.p0i8.i32(i8* nocapture, i8* nocapture, i32, i32, i1) nounwind
\ No newline at end of file
diff --git a/test/Transforms/ScalarRepl/2009-03-04-MemCpyAlign.ll b/test/Transforms/ScalarRepl/2009-03-04-MemCpyAlign.ll
index 526457be1e..3218d599d1 100644
--- a/test/Transforms/ScalarRepl/2009-03-04-MemCpyAlign.ll
+++ b/test/Transforms/ScalarRepl/2009-03-04-MemCpyAlign.ll
@@ -12,9 +12,8 @@ entry:
         %0 = getelementptr %struct.st* %s, i32 0, i32 0  ; <i16*> [#uses=1]
         store i16 1, i16* %0, align 4
         %s1 = bitcast %struct.st* %s to i8*  ; <i8*> [#uses=1]
-        call void @llvm.memcpy.i32(i8* %p, i8* %s1, i32 2, i32 1)
+        call void @llvm.memcpy.p0i8.p0i8.i32(i8* %p, i8* %s1, i32 2, i32 1, i1 false)
         ret void
 }
 
-declare void @llvm.memcpy.i32(i8* nocapture, i8* nocapture, i32, i32) nounwind
-
+declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture, i32, i32, i1) nounwind
diff --git a/test/Transforms/ScalarRepl/2009-12-11-NeonTypes.ll b/test/Transforms/ScalarRepl/2009-12-11-NeonTypes.ll
index 31d9bae6be..1993e4f526 100644
--- a/test/Transforms/ScalarRepl/2009-12-11-NeonTypes.ll
+++ b/test/Transforms/ScalarRepl/2009-12-11-NeonTypes.ll
@@ -13,54 +13,54 @@ define void @test(<8 x i16> %tmp.0, %struct.int16x8x2_t* %dst) nounwind {
 ; CHECK: @test
 ; CHECK-NOT: alloca
 ; CHECK: "alloca point"
+; CHECK: store <8 x i16>
+; CHECK: store <8 x i16>
+
 entry:
-  %tmp_addr = alloca %struct.int16x8_t            ; <%struct.int16x8_t*> [#uses=3]
-  %dst_addr = alloca %struct.int16x8x2_t*         ; <%struct.int16x8x2_t**> [#uses=2]
-  %__rv = alloca %union..0anon                    ; <%union..0anon*> [#uses=2]
-  %__bx = alloca %struct.int16x8_t                ; <%struct.int16x8_t*> [#uses=2]
-  %__ax = alloca %struct.int16x8_t                ; <%struct.int16x8_t*> [#uses=2]
-  %tmp2 = alloca %struct.int16x8x2_t              ; <%struct.int16x8x2_t*> [#uses=2]
-  %0 = alloca %struct.int16x8x2_t                 ; <%struct.int16x8x2_t*> [#uses=2]
-  %"alloca point" = bitcast i32 0 to i32          ; <i32> [#uses=0]
-  %1 = getelementptr inbounds %struct.int16x8_t* %tmp_addr, i32 0, i32 0 ; <<8 x i16>*> [#uses=1]
+  %tmp_addr = alloca %struct.int16x8_t
+  %dst_addr = alloca %struct.int16x8x2_t*
+  %__rv = alloca %union..0anon
+  %__bx = alloca %struct.int16x8_t
+  %__ax = alloca %struct.int16x8_t
+  %tmp2 = alloca %struct.int16x8x2_t
+  %0 = alloca %struct.int16x8x2_t
+  %"alloca point" = bitcast i32 0 to i32
+  %1 = getelementptr inbounds %struct.int16x8_t* %tmp_addr, i32 0, i32 0
   store <8 x i16> %tmp.0, <8 x i16>* %1
   store %struct.int16x8x2_t* %dst, %struct.int16x8x2_t** %dst_addr
-  %2 = getelementptr inbounds %struct.int16x8_t* %__ax, i32 0, i32 0 ; <<8 x i16>*> [#uses=1]
-  %3 = getelementptr inbounds %struct.int16x8_t* %tmp_addr, i32 0, i32 0 ; <<8 x i16>*> [#uses=1]
-  %4 = load <8 x i16>* %3, align 16               ; <<8 x i16>> [#uses=1]
+  %2 = getelementptr inbounds %struct.int16x8_t* %__ax, i32 0, i32 0
+  %3 = getelementptr inbounds %struct.int16x8_t* %tmp_addr, i32 0, i32 0
+  %4 = load <8 x i16>* %3, align 16
   store <8 x i16> %4, <8 x i16>* %2, align 16
-  %5 = getelementptr inbounds %struct.int16x8_t* %__bx, i32 0, i32 0 ; <<8 x i16>*> [#uses=1]
-  %6 = getelementptr inbounds %struct.int16x8_t* %tmp_addr, i32 0, i32 0 ; <<8 x i16>*> [#uses=1]
-  %7 = load <8 x i16>* %6, align 16               ; <<8 x i16>> [#uses=1]
+  %5 = getelementptr inbounds %struct.int16x8_t* %__bx, i32 0, i32 0
+  %6 = getelementptr inbounds %struct.int16x8_t* %tmp_addr, i32 0, i32 0
+  %7 = load <8 x i16>* %6, align 16
   store <8 x i16> %7, <8 x i16>* %5, align 16
-  %8 = getelementptr inbounds %struct.int16x8_t* %__ax, i32 0, i32 0 ; <<8 x i16>*> [#uses=1]
-  %9 = load <8 x i16>* %8, align 16               ; <<8 x i16>> [#uses=2]
-  %10 = getelementptr inbounds %struct.int16x8_t* %__bx, i32 0, i32 0 ; <<8 x i16>*> [#uses=1]
-  %11 = load <8 x i16>* %10, align 16             ; <<8 x i16>> [#uses=2]
-  %12 = getelementptr inbounds %union..0anon* %__rv, i32 0, i32 0 ; <%struct.int16x8x2_t*> [#uses=1]
-  %13 = bitcast %struct.int16x8x2_t* %12 to %struct.__neon_int16x8x2_t* ; <%struct.__neon_int16x8x2_t*> [#uses=2]
-  %14 = shufflevector <8 x i16> %9, <8 x i16> %11, <8 x i32> <i32 0, i32 8, i32 2, i32 10, i32 4, i32 12, i32 6, i32 14> ; <<8 x i16>> [#uses=1]
-  %15 = getelementptr inbounds %struct.__neon_int16x8x2_t* %13, i32 0, i32 0 ; <<8 x i16>*> [#uses=1]
+  %8 = getelementptr inbounds %struct.int16x8_t* %__ax, i32 0, i32 0
+  %9 = load <8 x i16>* %8, align 16
+  %10 = getelementptr inbounds %struct.int16x8_t* %__bx, i32 0, i32 0
+  %11 = load <8 x i16>* %10, align 16
+  %12 = getelementptr inbounds %union..0anon* %__rv, i32 0, i32 0
+  %13 = bitcast %struct.int16x8x2_t* %12 to %struct.__neon_int16x8x2_t*
+  %14 = shufflevector <8 x i16> %9, <8 x i16> %11, <8 x i32> <i32 0, i32 8, i32 2, i32 10, i32 4, i32 12, i32 6, i32 14>
+  %15 = getelementptr inbounds %struct.__neon_int16x8x2_t* %13, i32 0, i32 0
   store <8 x i16> %14, <8 x i16>* %15
-  %16 = shufflevector <8 x i16> %9, <8 x i16> %11, <8 x i32> <i32 1, i32 9, i32 3, i32 11, i32 5, i32 13, i32 7, i32 15> ; <<8 x i16>> [#uses=1]
-  %17 = getelementptr inbounds %struct.__neon_int16x8x2_t* %13, i32 0, i32 1 ; <<8 x i16>*> [#uses=1]
+  %16 = shufflevector <8 x i16> %9, <8 x i16> %11, <8 x i32> <i32 1, i32 9, i32 3, i32 11, i32 5, i32 13, i32 7, i32 15>
+  %17 = getelementptr inbounds %struct.__neon_int16x8x2_t* %13, i32 0, i32 1
   store <8 x i16> %16, <8 x i16>* %17
-  %18 = getelementptr inbounds %union..0anon* %__rv, i32 0, i32 0 ; <%struct.int16x8x2_t*> [#uses=1]
-  %19 = bitcast %struct.int16x8x2_t* %0 to i8*    ; <i8*> [#uses=1]
-  %20 = bitcast %struct.int16x8x2_t* %18 to i8*   ; <i8*> [#uses=1]
-  call void @llvm.memcpy.i32(i8* %19, i8* %20, i32 32, i32 16)
-  %tmp21 = bitcast %struct.int16x8x2_t* %tmp2 to i8* ; <i8*> [#uses=1]
-  %21 = bitcast %struct.int16x8x2_t* %0 to i8*    ; <i8*> [#uses=1]
-  call void @llvm.memcpy.i32(i8* %tmp21, i8* %21, i32 32, i32 16)
-  %22 = load %struct.int16x8x2_t** %dst_addr, align 4 ; <%struct.int16x8x2_t*> [#uses=1]
-  %23 = bitcast %struct.int16x8x2_t* %22 to i8*   ; <i8*> [#uses=1]
-  %tmp22 = bitcast %struct.int16x8x2_t* %tmp2 to i8* ; <i8*> [#uses=1]
-  call void @llvm.memcpy.i32(i8* %23, i8* %tmp22, i32 32, i32 16)
+  %18 = getelementptr inbounds %union..0anon* %__rv, i32 0, i32 0
+  %19 = bitcast %struct.int16x8x2_t* %0 to i8*
+  %20 = bitcast %struct.int16x8x2_t* %18 to i8*
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %19, i8* %20, i32 32, i32 16, i1 false)
+  %tmp21 = bitcast %struct.int16x8x2_t* %tmp2 to i8*
+  %21 = bitcast %struct.int16x8x2_t* %0 to i8*
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %tmp21, i8* %21, i32 32, i32 16, i1 false)
+  %22 = load %struct.int16x8x2_t** %dst_addr, align 4
+  %23 = bitcast %struct.int16x8x2_t* %22 to i8*
+  %tmp22 = bitcast %struct.int16x8x2_t* %tmp2 to i8*
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %23, i8* %tmp22, i32 32, i32 16, i1 false)
   br label %return
 
-; CHECK: store <8 x i16>
-; CHECK: store <8 x i16>
-
 return:                                           ; preds = %entry
   ret void
 }
@@ -69,21 +69,22 @@ return:                                           ; preds = %entry
 %struct._NSRange = type { i64 }
 
 define void @test_memcpy_self() nounwind {
-; CHECK: @test_memcpy_self
-; CHECK-NOT: alloca
-; CHECK: br i1
 entry:
-  %range = alloca %struct._NSRange                ; <%struct._NSRange*> [#uses=2]
+  %range = alloca %struct._NSRange
   br i1 undef, label %cond.true, label %cond.false
 
 cond.true:                                        ; preds = %entry
-  %tmp3 = bitcast %struct._NSRange* %range to i8* ; <i8*> [#uses=1]
-  %tmp4 = bitcast %struct._NSRange* %range to i8* ; <i8*> [#uses=1]
-  call void @llvm.memcpy.i32(i8* %tmp3, i8* %tmp4, i32 8, i32 8)
+  %tmp3 = bitcast %struct._NSRange* %range to i8*
+  %tmp4 = bitcast %struct._NSRange* %range to i8*
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %tmp3, i8* %tmp4, i32 8, i32 8, i1 false)
   ret void
 
 cond.false:                                       ; preds = %entry
   ret void
+
+; CHECK: @test_memcpy_self
+; CHECK-NOT: alloca
+; CHECK: br i1
 }
 
-declare void @llvm.memcpy.i32(i8* nocapture, i8* nocapture, i32, i32) nounwind
+declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture, i32, i32, i1) nounwind
diff --git a/test/Transforms/ScalarRepl/2010-01-18-SelfCopy.ll b/test/Transforms/ScalarRepl/2010-01-18-SelfCopy.ll
index 3aee399f1c..52df6d5c59 100644
--- a/test/Transforms/ScalarRepl/2010-01-18-SelfCopy.ll
+++ b/test/Transforms/ScalarRepl/2010-01-18-SelfCopy.ll
@@ -3,7 +3,7 @@
 
 target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f32:32:32-f64:32:64-v64:64:64-v128:128:128-a0:0:64-f80:128:128-n8:16:32"
 
-%struct.test = type { [3 x double ] }
+%struct.test = type { [3 x double] }
 
 define void @test_memcpy_self() nounwind {
 ; CHECK: @test_memcpy_self
@@ -11,8 +11,8 @@ define void @test_memcpy_self() nounwind {
 ; CHECK: ret void
   %1 = alloca %struct.test
   %2 = bitcast %struct.test* %1 to i8*
-  call void @llvm.memcpy.i32(i8* %2, i8* %2, i32 24, i32 4)
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %2, i8* %2, i32 24, i32 4, i1 false)
   ret void
 }
 
-declare void @llvm.memcpy.i32(i8* nocapture, i8* nocapture, i32, i32) nounwind
+declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture, i32, i32, i1) nounwind
diff --git a/test/Transforms/ScalarRepl/badarray.ll b/test/Transforms/ScalarRepl/badarray.ll
index 3ec3c01b28..768fec6306 100644
--- a/test/Transforms/ScalarRepl/badarray.ll
+++ b/test/Transforms/ScalarRepl/badarray.ll
@@ -48,10 +48,10 @@ entry:
   %callret = call %padded *@test3f() ; <i32> [#uses=2]
   %callretcast = bitcast %padded* %callret to i8*                     ; <i8*> [#uses=1]
   %var_11 = bitcast %padded* %var_1 to i8*        ; <i8*> [#uses=1]
-  call void @llvm.memcpy.i32(i8* %callretcast, i8* %var_11, i32 8, i32 4)
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %callretcast, i8* %var_11, i32 8, i32 4, i1 false)
   ret void
 }
 
-declare void @llvm.memcpy.i32(i8* nocapture, i8* nocapture, i32, i32) nounwind
+declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture, i32, i32, i1) nounwind
 
 declare %padded* @test3f()
diff --git a/test/Transforms/ScalarRepl/crash.ll b/test/Transforms/ScalarRepl/crash.ll
index 83daaaf14c..e06d0c1cd7 100644
--- a/test/Transforms/ScalarRepl/crash.ll
+++ b/test/Transforms/ScalarRepl/crash.ll
@@ -143,7 +143,6 @@ entry:
         %struct.anon = type { %struct.aal_spanarray_t }
 
 
-declare void @llvm.memcpy.i32(i8*, i8*, i32, i32)
 
 define fastcc void @test7() {
 entry:
@@ -158,7 +157,7 @@ cond_next114.i:         ; preds = %cond_true
 
 cond_next:              ; preds = %cond_true
         %SB19 = bitcast %struct.aal_spanbucket_t* %SB to i8*            ; <i8*> [#uses=1]
-        call void @llvm.memcpy.i32( i8* %SB19, i8* null, i32 12, i32 0 )
+        call void @llvm.memcpy.p0i8.p0i8.i32(i8* %SB19, i8* null, i32 12, i32 0, i1 false)
         br i1 false, label %cond_next34, label %cond_next79
 
 cond_next34:            ; preds = %cond_next
@@ -196,7 +195,7 @@ entry:
         %.compoundliteral = alloca %0           
         %tmp228 = getelementptr %0* %.compoundliteral, i32 0, i32 7
         %tmp229 = bitcast [0 x i16]* %tmp228 to i8*             
-        call void @llvm.memset.i64(i8* %tmp229, i8 0, i64 0, i32 2)
+        call void @llvm.memset.p0i8.i64(i8* %tmp229, i8 0, i64 0, i32 2, i1 false)
         unreachable
 }
 
@@ -260,3 +259,6 @@ entry:
   call void %0() nounwind
   ret void
 }
+
+declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture, i32, i32, i1) nounwind
+declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i32, i1) nounwind
diff --git a/test/Transforms/ScalarRepl/memset-aggregate.ll b/test/Transforms/ScalarRepl/memset-aggregate.ll
index 5aeefcd131..42e7a0ffdc 100644
--- a/test/Transforms/ScalarRepl/memset-aggregate.ll
+++ b/test/Transforms/ScalarRepl/memset-aggregate.ll
@@ -14,31 +14,29 @@ entry:
 	%L = alloca %struct.foo, align 8		; <%struct.foo*> [#uses=2]
 	%L2 = bitcast %struct.foo* %L to i8*		; <i8*> [#uses=1]
 	%tmp13 = bitcast %struct.foo* %P to i8*		; <i8*> [#uses=1]
-	call void @llvm.memcpy.i32( i8* %L2, i8* %tmp13, i32 8, i32 4 )
+        call void @llvm.memcpy.p0i8.p0i8.i32(i8* %L2, i8* %tmp13, i32 8, i32 4, i1 false)
 	%tmp4 = getelementptr %struct.foo* %L, i32 0, i32 0		; <i32*> [#uses=1]
 	%tmp5 = load i32* %tmp4		; <i32> [#uses=1]
 	ret i32 %tmp5
 }
 
-declare void @llvm.memcpy.i32(i8*, i8*, i32, i32)
 
 define i32 @test2() {
 entry:
 	%L = alloca [4 x %struct.foo], align 16		; <[4 x %struct.foo]*> [#uses=2]
 	%L12 = bitcast [4 x %struct.foo]* %L to i8*		; <i8*> [#uses=1]
-	call void @llvm.memset.i32( i8* %L12, i8 0, i32 32, i32 16 )
+        call void @llvm.memset.p0i8.i32(i8* %L12, i8 0, i32 32, i32 16, i1 false)
 	%tmp4 = getelementptr [4 x %struct.foo]* %L, i32 0, i32 0, i32 0		; <i32*> [#uses=1]
 	%tmp5 = load i32* %tmp4		; <i32> [#uses=1]
 	ret i32 %tmp5
 }
 
-declare void @llvm.memset.i32(i8*, i8, i32, i32)
 
 define i32 @test3() {
 entry:
 	%B = alloca %struct.bar, align 16		; <%struct.bar*> [#uses=4]
 	%B1 = bitcast %struct.bar* %B to i8*		; <i8*> [#uses=1]
-	call void @llvm.memset.i32( i8* %B1, i8 1, i32 24, i32 16 )
+	call void @llvm.memset.p0i8.i32(i8* %B1, i8 1, i32 24, i32 16, i1 false)
 	%tmp3 = getelementptr %struct.bar* %B, i32 0, i32 0, i32 0		; <i32*> [#uses=1]
 	store i32 1, i32* %tmp3
 	%tmp4 = getelementptr %struct.bar* %B, i32 0, i32 2		; <double*> [#uses=1]
@@ -58,9 +56,12 @@ entry:
 	store i32 1, i32* %0, align 8
 	%1 = getelementptr %struct.f* %A, i32 0, i32 1		; <i32*> [#uses=1]
 	%2 = bitcast i32* %1 to i8*		; <i8*> [#uses=1]
-	call void @llvm.memset.i32(i8* %2, i8 2, i32 12, i32 4)
+	call void @llvm.memset.p0i8.i32(i8* %2, i8 2, i32 12, i32 4, i1 false)
 	%3 = getelementptr %struct.f* %A, i32 0, i32 2		; <i32*> [#uses=1]
 	%4 = load i32* %3, align 8		; <i32> [#uses=1]
 	%retval12 = trunc i32 %4 to i16		; <i16> [#uses=1]
 	ret i16 %retval12
 }
+declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture, i32, i32, i1) nounwind
+
+declare void @llvm.memset.p0i8.i32(i8* nocapture, i8, i32, i32, i1) nounwind
\ No newline at end of file
diff --git a/test/Transforms/SimplifyLibCalls/MemCpy.ll b/test/Transforms/SimplifyLibCalls/MemCpy.ll
index 39662b1589..c711178fa1 100644
--- a/test/Transforms/SimplifyLibCalls/MemCpy.ll
+++ b/test/Transforms/SimplifyLibCalls/MemCpy.ll
@@ -4,17 +4,16 @@
 @hel = constant [4 x i8] c"hel\00"		; <[4 x i8]*> [#uses=1]
 @hello_u = constant [8 x i8] c"hello_u\00"		; <[8 x i8]*> [#uses=1]
 
-declare void @llvm.memcpy.i32(i8*, i8*, i32, i32)
-
 define i32 @main() {
-	%h_p = getelementptr [2 x i8]* @h, i32 0, i32 0		; <i8*> [#uses=1]
-	%hel_p = getelementptr [4 x i8]* @hel, i32 0, i32 0		; <i8*> [#uses=1]
-	%hello_u_p = getelementptr [8 x i8]* @hello_u, i32 0, i32 0		; <i8*> [#uses=1]
-	%target = alloca [1024 x i8]		; <[1024 x i8]*> [#uses=1]
-	%target_p = getelementptr [1024 x i8]* %target, i32 0, i32 0		; <i8*> [#uses=3]
-	call void @llvm.memcpy.i32( i8* %target_p, i8* %h_p, i32 2, i32 2 )
-	call void @llvm.memcpy.i32( i8* %target_p, i8* %hel_p, i32 4, i32 4 )
-	call void @llvm.memcpy.i32( i8* %target_p, i8* %hello_u_p, i32 8, i32 8 )
-	ret i32 0
+  %h_p = getelementptr [2 x i8]* @h, i32 0, i32 0
+  %hel_p = getelementptr [4 x i8]* @hel, i32 0, i32 0
+  %hello_u_p = getelementptr [8 x i8]* @hello_u, i32 0, i32 0
+  %target = alloca [1024 x i8]
+  %target_p = getelementptr [1024 x i8]* %target, i32 0, i32 0
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %target_p, i8* %h_p, i32 2, i32 2, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %target_p, i8* %hel_p, i32 4, i32 4, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %target_p, i8* %hello_u_p, i32 8, i32 8, i1 false)
+  ret i32 0
 }
 
+declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture, i32, i32, i1) nounwind