; On darwin10 (which is the triple in this .ll file) this loop can be turned
; into a memset_pattern call.
; rdar://9009151
-define void @test11(i32* nocapture %P) nounwind ssp {
+define void @test11_pattern(i32* nocapture %P) nounwind ssp {
entry:
br label %for.body
for.end: ; preds = %for.body
ret void
-; CHECK: @test11
+; CHECK: @test11_pattern
; CHECK-NEXT: entry:
; CHECK-NEXT: bitcast
; CHECK-NEXT: memset_pattern
; CHECK: ret void
}
+@G = global i32 5
+
+; This store-of-address loop can be turned into a memset_pattern call.
+; rdar://9009151
+define void @test13_pattern(i32** nocapture %P) nounwind ssp {
+entry:
+ br label %for.body
+
+for.body: ; preds = %entry, %for.body
+ %indvar = phi i64 [ 0, %entry ], [ %indvar.next, %for.body ]
+ %arrayidx = getelementptr i32** %P, i64 %indvar
+ store i32* @G, i32** %arrayidx, align 4
+ %indvar.next = add i64 %indvar, 1
+ %exitcond = icmp eq i64 %indvar.next, 10000
+ br i1 %exitcond, label %for.end, label %for.body
+
+for.end: ; preds = %for.body
+ ret void
+; CHECK: @test13_pattern
+; CHECK-NEXT: entry:
+; CHECK-NEXT: bitcast
+; CHECK-NEXT: memset_pattern
+; CHECK-NOT: store
+; CHECK: ret void
+}
+
+
+
+; PR9815 - This is a partial overlap case that cannot be safely transformed
+; into a memcpy.
+@g_50 = global [7 x i32] [i32 0, i32 0, i32 0, i32 0, i32 1, i32 0, i32 0], align 16
+
+define i32 @test14() nounwind {
+entry:
+ br label %for.body
+
+for.body: ; preds = %for.inc, %for.body.lr.ph
+ %tmp5 = phi i32 [ %inc, %for.body ], [ 0, %entry ]
+ %add = add nsw i32 %tmp5, 4
+ %idxprom = sext i32 %add to i64
+ %arrayidx = getelementptr inbounds [7 x i32]* @g_50, i32 0, i64 %idxprom
+ %tmp2 = load i32* %arrayidx, align 4
+ %add4 = add nsw i32 %tmp5, 5
+ %idxprom5 = sext i32 %add4 to i64
+ %arrayidx6 = getelementptr inbounds [7 x i32]* @g_50, i32 0, i64 %idxprom5
+ store i32 %tmp2, i32* %arrayidx6, align 4
+ %inc = add nsw i32 %tmp5, 1
+ %cmp = icmp slt i32 %inc, 2
+ br i1 %cmp, label %for.body, label %for.end
+
+for.end: ; preds = %for.inc
+ %tmp8 = load i32* getelementptr inbounds ([7 x i32]* @g_50, i32 0, i64 6), align 4
+ ret i32 %tmp8
+; CHECK: @test14
+; CHECK: for.body:
+; CHECK: load i32
+; CHECK: store i32
+; CHECK: br i1 %cmp
+
+}
+
+