7 files changed, 198 insertions, 7 deletions
diff --git a/test/Transforms/DeadStoreElimination/OverwriteStoreEnd.ll b/test/Transforms/DeadStoreElimination/OverwriteStoreEnd.ll
new file mode 100644
index 0000000..ed53eb5
--- /dev/null
+++ b/test/Transforms/DeadStoreElimination/OverwriteStoreEnd.ll
@@ -0,0 +1,95 @@
+; RUN: opt < %s -basicaa -dse -S | FileCheck %s
+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-n8:16:32:64-S128"
+
+%struct.vec2 = type { <4 x i32>, <4 x i32> }
+%struct.vec2plusi = type { <4 x i32>, <4 x i32>, i32 }
+
+@glob1 = global %struct.vec2 zeroinitializer, align 16
+@glob2 = global %struct.vec2plusi zeroinitializer, align 16
+
+define void @write24to28(i32* nocapture %p) nounwind uwtable ssp {
+; CHECK: @write24to28
+entry:
+  %arrayidx0 = getelementptr inbounds i32* %p, i64 1
+  %p3 = bitcast i32* %arrayidx0 to i8*
+; CHECK: call void @llvm.memset.p0i8.i64(i8* %p3, i8 0, i64 24, i32 4, i1 false)
+  call void @llvm.memset.p0i8.i64(i8* %p3, i8 0, i64 28, i32 4, i1 false)
+  %arrayidx1 = getelementptr inbounds i32* %p, i64 7
+  store i32 1, i32* %arrayidx1, align 4
+  ret void
+}
+
+define void @write28to32(i32* nocapture %p) nounwind uwtable ssp {
+; CHECK: @write28to32
+entry:
+  %p3 = bitcast i32* %p to i8*
+; CHECK: call void @llvm.memset.p0i8.i64(i8* %p3, i8 0, i64 28, i32 4, i1 false)
+  call void @llvm.memset.p0i8.i64(i8* %p3, i8 0, i64 32, i32 4, i1 false)
+  %arrayidx1 = getelementptr inbounds i32* %p, i64 7
+  store i32 1, i32* %arrayidx1, align 4
+  ret void
+}
+
+define void @dontwrite28to32memset(i32* nocapture %p) nounwind uwtable ssp {
+; CHECK: @dontwrite28to32memset
+entry:
+  %p3 = bitcast i32* %p to i8*
+; CHECK: call void @llvm.memset.p0i8.i64(i8* %p3, i8 0, i64 32, i32 16, i1 false)
+  call void @llvm.memset.p0i8.i64(i8* %p3, i8 0, i64 32, i32 16, i1 false)
+  %arrayidx1 = getelementptr inbounds i32* %p, i64 7
+  store i32 1, i32* %arrayidx1, align 4
+  ret void
+}
+
+define void @write32to36(%struct.vec2plusi* nocapture %p) nounwind uwtable ssp {
+; CHECK: @write32to36
+entry:
+  %0 = bitcast %struct.vec2plusi* %p to i8*
+; CHECK: tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %0, i8* bitcast (%struct.vec2plusi* @glob2 to i8*), i64 32, i32 16, i1 false)
+  tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %0, i8* bitcast (%struct.vec2plusi* @glob2 to i8*), i64 36, i32 16, i1 false)
+  %c = getelementptr inbounds %struct.vec2plusi* %p, i64 0, i32 2
+  store i32 1, i32* %c, align 4
+  ret void
+}
+
+define void @write16to32(%struct.vec2* nocapture %p) nounwind uwtable ssp {
+; CHECK: @write16to32
+entry:
+  %0 = bitcast %struct.vec2* %p to i8*
+; CHECK: tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %0, i8* bitcast (%struct.vec2* @glob1 to i8*), i64 16, i32 16, i1 false)
+  tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %0, i8* bitcast (%struct.vec2* @glob1 to i8*), i64 32, i32 16, i1 false)
+  %c = getelementptr inbounds %struct.vec2* %p, i64 0, i32 1
+  store <4 x i32> <i32 1, i32 2, i32 3, i32 4>, <4 x i32>* %c, align 4
+  ret void
+}
+
+define void @dontwrite28to32memcpy(%struct.vec2* nocapture %p) nounwind uwtable ssp {
+; CHECK: @dontwrite28to32memcpy
+entry:
+  %0 = bitcast %struct.vec2* %p to i8*
+; CHECK: tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %0, i8* bitcast (%struct.vec2* @glob1 to i8*), i64 32, i32 16, i1 false)
+  tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %0, i8* bitcast (%struct.vec2* @glob1 to i8*), i64 32, i32 16, i1 false)
+  %arrayidx1 = getelementptr inbounds %struct.vec2* %p, i64 0, i32 0, i64 7
+  store i32 1, i32* %arrayidx1, align 4
+  ret void
+}
+
+declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture, i64, i32, i1) nounwind
+declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i32, i1) nounwind
+
+%struct.trapframe = type { i64, i64, i64 }
+
+; bugzilla 11455 - make sure negative GEP's don't break this optimisation
+; CHECK: @cpu_lwp_fork
+define void @cpu_lwp_fork(%struct.trapframe* %md_regs, i64 %pcb_rsp0) nounwind uwtable noinline ssp {
+entry:
+  %0 = inttoptr i64 %pcb_rsp0 to %struct.trapframe*
+  %add.ptr = getelementptr inbounds %struct.trapframe* %0, i64 -1
+  %1 = bitcast %struct.trapframe* %add.ptr to i8*
+  %2 = bitcast %struct.trapframe* %md_regs to i8*
+; CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* %1, i8* %2, i64 24, i32 1, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %1, i8* %2, i64 24, i32 1, i1 false)
+  %tf_trapno = getelementptr inbounds %struct.trapframe* %0, i64 -1, i32 1
+  store i64 3, i64* %tf_trapno, align 8
+  ret void
+}
diff --git a/test/Transforms/DeadStoreElimination/dg.exp b/test/Transforms/DeadStoreElimination/dg.exp
deleted file mode 100644
index f200589..0000000
--- a/test/Transforms/DeadStoreElimination/dg.exp
+++ /dev/null
@@ -1,3 +0,0 @@
-load_lib llvm.exp
-
-RunLLVMTests [lsort [glob -nocomplain $srcdir/$subdir/*.{ll,c,cpp}]]
diff --git a/test/Transforms/DeadStoreElimination/dominate.ll b/test/Transforms/DeadStoreElimination/dominate.ll
new file mode 100644
index 0000000..284fea4
--- /dev/null
+++ b/test/Transforms/DeadStoreElimination/dominate.ll
@@ -0,0 +1,25 @@
+; RUN: opt  %s -dse -disable-output
+; test that we don't crash
+declare void @bar()
+
+define void @foo() {
+bb1:
+  %memtmp3.i = alloca [21 x i8], align 1
+  %0 = getelementptr inbounds [21 x i8]* %memtmp3.i, i64 0, i64 0
+  br label %bb3
+
+bb2:
+  call void @llvm.lifetime.end(i64 -1, i8* %0)
+  br label %bb3
+
+bb3:
+  call void @bar()
+  call void @llvm.lifetime.end(i64 -1, i8* %0)
+  br label %bb4
+
+bb4:
+  ret void
+
+}
+
+declare void @llvm.lifetime.end(i64, i8* nocapture) nounwind
diff --git a/test/Transforms/DeadStoreElimination/free.ll b/test/Transforms/DeadStoreElimination/free.ll
index aa3f0ab..a5fbdc7 100644
--- a/test/Transforms/DeadStoreElimination/free.ll
+++ b/test/Transforms/DeadStoreElimination/free.ll
@@ -2,6 +2,9 @@
 
 target datalayout = "e-p:64:64:64"
 
+declare void @free(i8* nocapture)
+declare noalias i8* @malloc(i64)
+
 ; CHECK: @test
 ; CHECK-NEXT: bitcast
 ; CHECK-NEXT: @free
@@ -26,10 +29,10 @@ define void @test2({i32, i32}* %P) {
 	ret void
 }
 
-; CHECK: @test4
+; CHECK: @test3
 ; CHECK-NOT: store
 ; CHECK: ret void
-define void @test4() {
+define void @test3() {
   %m = call i8* @malloc(i64 24)
   store i8 0, i8* %m
   %m1 = getelementptr i8* %m, i64 1
@@ -38,5 +41,30 @@ define void @test4() {
   ret void
 }
 
-declare void @free(i8*)
-declare i8* @malloc(i64)
+; PR11240
+; CHECK: @test4
+; CHECK-NOT: store
+; CHECK: ret void
+define void @test4(i1 %x) nounwind {
+entry:
+  %alloc1 = tail call noalias i8* @malloc(i64 4) nounwind
+  br i1 %x, label %skipinit1, label %init1
+
+init1:
+  store i8 1, i8* %alloc1
+  br label %skipinit1
+
+skipinit1:
+  tail call void @free(i8* %alloc1) nounwind
+  ret void
+}
+
+; CHECK: @test5
+define void @test5() {
+  br label %bb
+
+bb:
+  tail call void @free(i8* undef) nounwind
+  br label %bb
+}
+
diff --git a/test/Transforms/DeadStoreElimination/lit.local.cfg b/test/Transforms/DeadStoreElimination/lit.local.cfg
new file mode 100644
index 0000000..19eebc0
--- /dev/null
+++ b/test/Transforms/DeadStoreElimination/lit.local.cfg
@@ -0,0 +1 @@
+config.suffixes = ['.ll', '.c', '.cpp']
diff --git a/test/Transforms/DeadStoreElimination/pr11390.ll b/test/Transforms/DeadStoreElimination/pr11390.ll
new file mode 100644
index 0000000..2ce6eea
--- /dev/null
+++ b/test/Transforms/DeadStoreElimination/pr11390.ll
@@ -0,0 +1,38 @@
+; RUN: opt -basicaa -dse -S -o - %s | FileCheck %s
+; PR11390
+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-n8:16:32:64-S128"
+target triple = "x86_64-unknown-linux-gnu"
+
+define fastcc void @cat_domain(i8* nocapture %name, i8* nocapture %domain, i8** 
+nocapture %s) nounwind uwtable {
+entry:
+  %call = tail call i64 @strlen(i8* %name) nounwind readonly
+  %call1 = tail call i64 @strlen(i8* %domain) nounwind readonly
+  %add = add i64 %call, 1
+  %add2 = add i64 %add, %call1
+  %add3 = add i64 %add2, 1
+  %call4 = tail call noalias i8* @malloc(i64 %add3) nounwind
+  store i8* %call4, i8** %s, align 8
+  %tobool = icmp eq i8* %call4, null
+  br i1 %tobool, label %return, label %if.end
+
+if.end:                                           ; preds = %entry
+  tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %call4, i8* %name, i64 %call, i32 1, i1 false)
+  %arrayidx = getelementptr inbounds i8* %call4, i64 %call
+  store i8 46, i8* %arrayidx, align 1
+; CHECK: store i8 46
+  %add.ptr5 = getelementptr inbounds i8* %call4, i64 %add
+  tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %add.ptr5, i8* %domain, i64 %call1, i32 1, i1 false)
+  %arrayidx8 = getelementptr inbounds i8* %call4, i64 %add2
+  store i8 0, i8* %arrayidx8, align 1
+  br label %return
+
+return:                                           ; preds = %if.end, %entry
+  ret void
+}
+
+declare i64 @strlen(i8* nocapture) nounwind readonly
+
+declare noalias i8* @malloc(i64) nounwind
+
+declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture, i64, i32, i1) nounwind
diff --git a/test/Transforms/DeadStoreElimination/simple.ll b/test/Transforms/DeadStoreElimination/simple.ll
index ec2f157..81eb5a8 100644
--- a/test/Transforms/DeadStoreElimination/simple.ll
+++ b/test/Transforms/DeadStoreElimination/simple.ll
@@ -251,3 +251,10 @@ bb:
 ; CHECK: call void @test19f
 }
 
+define void @test20() {
+  %m = call i8* @malloc(i32 24)
+  store i8 0, i8* %m
+  ret void
+}
+; CHECK: @test20
+; CHECK-NEXT: ret void