[Hexagon] Generate correct runtime check when recognizing memmove

The check (assuming positive stride) for validity of memmove should be
(a) the destination is at a lower address than the source, or
(b) the distance between the source and destination is greater than or
    equal the number of bytes copied.

For the second part it is sufficient to assume that the destination
is at a higher address, since the opposite case is covered by (a).
The distance calculation was previously done by subtracting the
pointers in the wrong order.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@311650 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/lib/Target/Hexagon/HexagonLoopIdiomRecognition.cpp b/lib/Target/Hexagon/HexagonLoopIdiomRecognition.cpp
index 7455a02b06dc01267ab451dea9509a7047b52c2c..29e2bc32dfbbc29daf33efdf8aeb567949d66b36 100644 (file)
--- a/lib/Target/Hexagon/HexagonLoopIdiomRecognition.cpp
+++ b/lib/Target/Hexagon/HexagonLoopIdiomRecognition.cpp
@@ -2117,9 +2117,10 @@ CleanupAndExit:
     Value *CmpA = Builder.CreateICmpULT(LowA, HighA);
     Value *Cond = CmpA;
 
-    // Check for distance between pointers.
-    Value *Dist = Builder.CreateSub(HighA, LowA);
-    Value *CmpD = Builder.CreateICmpSLT(NumBytes, Dist);
+    // Check for distance between pointers. Since the case LowA < HighA
+    // is checked for above, assume LowA >= HighA.
+    Value *Dist = Builder.CreateSub(LowA, HighA);
+    Value *CmpD = Builder.CreateICmpSLE(NumBytes, Dist);
     Value *CmpEither = Builder.CreateOr(Cond, CmpD);
     Cond = CmpEither;
 

diff --git a/test/CodeGen/Hexagon/loop-idiom/memmove-rt-check.ll b/test/CodeGen/Hexagon/loop-idiom/memmove-rt-check.ll
new file mode 100644 (file)
index 0000000..ce2c17e
--- /dev/null
+++ b/test/CodeGen/Hexagon/loop-idiom/memmove-rt-check.ll
@@ -0,0 +1,45 @@
+; RUN: opt -hexagon-loop-idiom -S < %s | FileCheck %s
+
+; Make sure that we generate correct runtime checks.
+
+; CHECK: b7.old:
+; CHECK:   [[LEN:%[0-9]+]] = shl i32 %len, 3
+; CHECK:   [[SRC:%[0-9]+]] = ptrtoint i8* %src to i32
+; CHECK:   [[DST:%[0-9]+]] = ptrtoint i8* %dst to i32
+; CHECK:   [[ULT:%[0-9]+]] = icmp ult i32 [[DST]], [[SRC]]
+; CHECK:   [[SUB:%[0-9]+]] = sub i32 [[DST]], [[SRC]]
+; CHECK:   [[SLT:%[0-9]+]] = icmp sle i32 [[LEN]], [[SUB]]
+; CHECK:   [[CND:%[0-9]+]] = or i1 [[ULT]], [[SLT]]
+; CHECK:   br i1 [[CND]], label %b8.rtli, label %b8.rtli.ph
+
+target datalayout = "e-m:e-p:32:32:32-a:0-n16:32-i64:64:64-i32:32:32-i16:16:16-i1:8:8-f32:32:32-f64:64:64-v32:32:32-v64:64:64-v512:512:512-v1024:1024:1024-v2048:2048:2048"
+target triple = "hexagon"
+
+define void @fred(i8* %dst, i8* %src, i32 %len) #0 {
+b3:
+  %v4 = bitcast i8* %src to i64*
+  %v5 = bitcast i8* %dst to i64*
+  %v6 = icmp slt i32 0, %len
+  br i1 %v6, label %b7, label %b16
+
+b7:                                               ; preds = %b3
+  br label %b8
+
+b8:                                               ; preds = %b13, %b7
+  %v9 = phi i32 [ 0, %b7 ], [ %v14, %b13 ]
+  %v10 = getelementptr inbounds i64, i64* %v4, i32 %v9
+  %v11 = load i64, i64* %v10, align 8
+  %v12 = getelementptr inbounds i64, i64* %v5, i32 %v9
+  store i64 %v11, i64* %v12, align 8
+  br label %b13
+
+b13:                                              ; preds = %b8
+  %v14 = add nsw i32 %v9, 1
+  %v15 = icmp slt i32 %v14, %len
+  br i1 %v15, label %b8, label %b16
+
+b16:                                              ; preds = %b13, %b3
+  ret void
+}
+
+attributes #0 = { noinline nounwind "target-cpu"="hexagonv60" }