10 files changed, 109 insertions, 9 deletions
diff --git a/test/Transforms/MemCpyOpt/2008-02-24-MultipleUseofSRet.ll b/test/Transforms/MemCpyOpt/2008-02-24-MultipleUseofSRet.ll
index 30d0a6d..30c2713 100644
--- a/test/Transforms/MemCpyOpt/2008-02-24-MultipleUseofSRet.ll
+++ b/test/Transforms/MemCpyOpt/2008-02-24-MultipleUseofSRet.ll
@@ -1,4 +1,4 @@
-; RUN: llvm-as < %s | opt -memcpyopt -dse | llvm-dis | grep {call.*initialize} | not grep memtmp
+; RUN: opt < %s -memcpyopt -dse -S | grep {call.*initialize} | not grep memtmp
 ; PR2077
 
 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"
diff --git a/test/Transforms/MemCpyOpt/2008-03-13-ReturnSlotBitcast.ll b/test/Transforms/MemCpyOpt/2008-03-13-ReturnSlotBitcast.ll
index 12c9a9d..13205e6 100644
--- a/test/Transforms/MemCpyOpt/2008-03-13-ReturnSlotBitcast.ll
+++ b/test/Transforms/MemCpyOpt/2008-03-13-ReturnSlotBitcast.ll
@@ -1,4 +1,4 @@
-; RUN: llvm-as < %s | opt -memcpyopt | llvm-dis | not grep {call.*memcpy.}
+; RUN: opt < %s -memcpyopt -S | not grep {call.*memcpy.}
 	%a = type { i32 }
 	%b = type { float }
 
diff --git a/test/Transforms/MemCpyOpt/2008-04-29-SRetRemoval.ll b/test/Transforms/MemCpyOpt/2008-04-29-SRetRemoval.ll
index 1d9a35d..4fec169 100644
--- a/test/Transforms/MemCpyOpt/2008-04-29-SRetRemoval.ll
+++ b/test/Transforms/MemCpyOpt/2008-04-29-SRetRemoval.ll
@@ -1,4 +1,4 @@
-; RUN: llvm-as < %s | opt -memcpyopt | llvm-dis | grep {call.*memcpy.*agg.result}
+; RUN: opt < %s -memcpyopt -S | grep {call.*memcpy.*agg.result}
 
 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"
diff --git a/test/Transforms/MemCpyOpt/align.ll b/test/Transforms/MemCpyOpt/align.ll
new file mode 100644
index 0000000..a9d0337
--- /dev/null
+++ b/test/Transforms/MemCpyOpt/align.ll
@@ -0,0 +1,18 @@
+; RUN: opt < %s -S -memcpyopt | FileCheck %s
+
+; The resulting memset is only 4-byte aligned, despite containing
+; a 16-byte alignmed store in the middle.
+
+; CHECK: call void @llvm.memset.i64(i8* %a01, i8 0, i64 16, i32 4)
+
+define void @foo(i32* %p) {
+  %a0 = getelementptr i32* %p, i64 0
+  store i32 0, i32* %a0, align 4
+  %a1 = getelementptr i32* %p, i64 1
+  store i32 0, i32* %a1, align 16
+  %a2 = getelementptr i32* %p, i64 2
+  store i32 0, i32* %a2, align 4
+  %a3 = getelementptr i32* %p, i64 3
+  store i32 0, i32* %a3, align 4
+  ret void
+}
diff --git a/test/Transforms/MemCpyOpt/crash.ll b/test/Transforms/MemCpyOpt/crash.ll
new file mode 100644
index 0000000..bf5b234
--- /dev/null
+++ b/test/Transforms/MemCpyOpt/crash.ll
@@ -0,0 +1,45 @@
+; RUN: opt < %s -memcpyopt -disable-output
+; PR4882
+
+target datalayout = "e-p:32:32:32-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"
+target triple = "armv7-eabi"
+
+%struct.qw = type { [4 x float] }
+%struct.bar = type { %struct.qw, %struct.qw, %struct.qw, %struct.qw, %struct.qw, float, float}
+
+define arm_aapcs_vfpcc void @test1(%struct.bar* %this) {
+entry:
+  %0 = getelementptr inbounds %struct.bar* %this, i32 0, i32 0, i32 0, i32 0
+  store float 0.000000e+00, float* %0, align 4
+  %1 = getelementptr inbounds %struct.bar* %this, i32 0, i32 0, i32 0, i32 1
+  store float 0.000000e+00, float* %1, align 4
+  %2 = getelementptr inbounds %struct.bar* %this, i32 0, i32 0, i32 0, i32 2
+  store float 0.000000e+00, float* %2, align 4
+  %3 = getelementptr inbounds %struct.bar* %this, i32 0, i32 0, i32 0, i32 3
+  store float 0.000000e+00, float* %3, align 4
+  %4 = getelementptr inbounds %struct.bar* %this, i32 0, i32 1, i32 0, i32 0
+  store float 0.000000e+00, float* %4, align 4
+  %5 = getelementptr inbounds %struct.bar* %this, i32 0, i32 1, i32 0, i32 1
+  store float 0.000000e+00, float* %5, align 4
+  %6 = getelementptr inbounds %struct.bar* %this, i32 0, i32 1, i32 0, i32 2
+  store float 0.000000e+00, float* %6, align 4
+  %7 = getelementptr inbounds %struct.bar* %this, i32 0, i32 1, i32 0, i32 3
+  store float 0.000000e+00, float* %7, align 4
+  %8 = getelementptr inbounds %struct.bar* %this, i32 0, i32 3, i32 0, i32 1
+  store float 0.000000e+00, float* %8, align 4
+  %9 = getelementptr inbounds %struct.bar* %this, i32 0, i32 3, i32 0, i32 2
+  store float 0.000000e+00, float* %9, align 4
+  %10 = getelementptr inbounds %struct.bar* %this, i32 0, i32 3, i32 0, i32 3
+  store float 0.000000e+00, float* %10, align 4
+  %11 = getelementptr inbounds %struct.bar* %this, i32 0, i32 4, i32 0, i32 0
+  store float 0.000000e+00, float* %11, align 4
+  %12 = getelementptr inbounds %struct.bar* %this, i32 0, i32 4, i32 0, i32 1
+  store float 0.000000e+00, float* %12, align 4
+  %13 = getelementptr inbounds %struct.bar* %this, i32 0, i32 4, i32 0, i32 2
+  store float 0.000000e+00, float* %13, align 4
+  %14 = getelementptr inbounds %struct.bar* %this, i32 0, i32 4, i32 0, i32 3
+  store float 0.000000e+00, float* %14, align 4
+  %15 = getelementptr inbounds %struct.bar* %this, i32 0, i32 5
+  store float 0.000000e+00, float* %15, align 4
+  unreachable
+}
diff --git a/test/Transforms/MemCpyOpt/form-memset.ll b/test/Transforms/MemCpyOpt/form-memset.ll
index ffacb85..eb8dbe3 100644
--- a/test/Transforms/MemCpyOpt/form-memset.ll
+++ b/test/Transforms/MemCpyOpt/form-memset.ll
@@ -1,5 +1,5 @@
-; RUN: llvm-as < %s | opt -memcpyopt | llvm-dis | not grep store
-; RUN: llvm-as < %s | opt -memcpyopt | llvm-dis | grep {call.*llvm.memset}
+; RUN: opt < %s -memcpyopt -S | not grep store
+; RUN: opt < %s -memcpyopt -S | grep {call.*llvm.memset}
 
 ; All the stores in this example should be merged into a single memset.
 
diff --git a/test/Transforms/MemCpyOpt/form-memset2.ll b/test/Transforms/MemCpyOpt/form-memset2.ll
index 719cd47..c90af9c 100644
--- a/test/Transforms/MemCpyOpt/form-memset2.ll
+++ b/test/Transforms/MemCpyOpt/form-memset2.ll
@@ -1,5 +1,5 @@
-; RUN: llvm-as < %s | opt -memcpyopt | llvm-dis | not grep store
-; RUN: llvm-as < %s | opt -memcpyopt | llvm-dis | grep {call.*llvm.memset} | count 3
+; RUN: opt < %s -memcpyopt -S | not grep store
+; RUN: opt < %s -memcpyopt -S | grep {call.*llvm.memset} | count 3
 
 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"
diff --git a/test/Transforms/MemCpyOpt/memcpy.ll b/test/Transforms/MemCpyOpt/memcpy.ll
index 94daee0..724acfa 100644
--- a/test/Transforms/MemCpyOpt/memcpy.ll
+++ b/test/Transforms/MemCpyOpt/memcpy.ll
@@ -1,4 +1,4 @@
-; RUN: llvm-as < %s | opt -memcpyopt -dse | llvm-dis | grep {call.*memcpy} | count 1
+; RUN: opt < %s -memcpyopt -dse -S | grep {call.*memcpy} | count 1
 
 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"
diff --git a/test/Transforms/MemCpyOpt/memmove.ll b/test/Transforms/MemCpyOpt/memmove.ll
new file mode 100644
index 0000000..73bbf0b
--- /dev/null
+++ b/test/Transforms/MemCpyOpt/memmove.ll
@@ -0,0 +1,37 @@
+; RUN: opt < %s -memcpyopt -S | FileCheck %s
+; These memmoves should get optimized to memcpys.
+
+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
+
+define i8* @test1(i8* nocapture %src) nounwind {
+entry:
+; CHECK: @test1
+; CHECK: call void @llvm.memcpy
+
+  %call3 = malloc [13 x i8]                       ; <[13 x i8]*> [#uses=1]
+  %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)
+  ret i8* %call3.sub
+}
+
+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)
+  ret void
+}
+
+; This cannot be optimize because the src/dst really do overlap.
+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)
+  ret void
+}
diff --git a/test/Transforms/MemCpyOpt/sret.ll b/test/Transforms/MemCpyOpt/sret.ll
index ad9fb1b..5002875 100644
--- a/test/Transforms/MemCpyOpt/sret.ll
+++ b/test/Transforms/MemCpyOpt/sret.ll
@@ -1,4 +1,4 @@
-; RUN: llvm-as < %s | opt -memcpyopt | llvm-dis | not grep {call.*memcpy}
+; RUN: opt < %s -memcpyopt -S | not grep {call.*memcpy}
 
 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"