[CodeGen][IfCvt] Don't re-ifcvt blocks with unanalyzable terminators.

ahmedbougacha · ahmedbougacha · commit 7173b669b47f · 2015-03-21T01:23:15.000Z
If we couldn't analyze its terminator (i.e., it's an indirectbr, or some other weirdness), we can't safely re-if-convert a predicated block, because we can't tell whether the predicated terminator can fallthrough (it does). Currently, we would completely ignore the fallthrough successor. In the added testcase, this means we used to generate: ... @ %entry: cmp r5, #21 ittt ne @ %cc1f: cmpne r7, #42 @ %cc2t: strne.w r5, [r8] movne pc, r10 @ %cc1t: ... Whereas the successor of %cc1f was originally %bb1. With the fix, we get the correct: ... @ %entry: cmp r5, #21 itt eq @ %cc1t: streq.w r5, [r11] moveq pc, r0 @ %cc1f: cmp r7, #42 itt ne @ %cc2t: strne.w r5, [r8] movne pc, r10 @ %bb1: ... rdar://20192768 Differential Revision: http://reviews.llvm.org/D8509 llvm-svn: 232872
diff --git a/llvm/lib/CodeGen/IfConversion.cpp b/llvm/lib/CodeGen/IfConversion.cpp
@@ -726,6 +726,12 @@ bool IfConverter::FeasibilityAnalysis(BBInfo &BBI,
   if (BBI.IsDone || BBI.IsUnpredicable)
     return false;
 
+  // If it is already predicated but we couldn't analyze its terminator, the
+  // latter might fallthrough, but we can't determine where to.
+  // Conservatively avoid if-converting again.
+  if (BBI.Predicate.size() && !BBI.IsBrAnalyzable)
+    return false;
+
   // If it is already predicated, check if the new predicate subsumes
   // its predicate.
   if (BBI.Predicate.size() && !TII->SubsumesPredicate(Pred, BBI.Predicate))
diff --git a/llvm/test/CodeGen/ARM/ifcvt-iter-indbr.ll b/llvm/test/CodeGen/ARM/ifcvt-iter-indbr.ll
@@ -0,0 +1,56 @@
+; RUN: llc < %s -mtriple thumbv7s-apple-darwin  -asm-verbose=false | FileCheck %s
+
+declare i32 @foo(i32)
+declare i8* @bar(i32, i8*, i8*)
+
+; Verify that we don't try to iteratively re-ifconvert a block with a
+; (predicated) indirectbr terminator.
+; If we do, we would ignore its fallthrough successor.
+
+
+; CHECK-LABEL: test:
+; CHECK:       cmp {{.*}}, #21
+; CHECK-NEXT:  itt eq
+; CHECK-NEXT:  streq.w
+; CHECK-NEXT:  moveq pc
+; CHECK-NEXT: LBB{{[0-9_]+}}:
+; CHECK-NEXT:  cmp {{.*}}, #42
+; CHECK-NEXT:  itt ne
+; CHECK-NEXT:  strne.w
+; CHECK-NEXT:  movne pc
+; CHECK-NEXT: Ltmp
+; CHECK-NEXT: LBB0_2:
+; CHECK-NEXT:  movw r0, #1234
+; CHECK-NEXT:  b [[FOOCALL:LBB[0-9_]+]]
+; CHECK-NEXT: Ltmp
+; CHECK-NEXT: LBB{{[0-9_]+}}:
+; CHECK-NEXT:  movw r0, #4567
+; CHECK-NEXT: [[FOOCALL]]:
+; CHECK-NEXT:  blx _foo
+
+define i32 @test(i32 %a, i32 %a2, i32* %p, i32* %p2) {
+entry:
+  %dst1 = call i8* @bar(i32 1, i8* blockaddress(@test, %bb1), i8* blockaddress(@test, %bb2))
+  %dst2 = call i8* @bar(i32 2, i8* blockaddress(@test, %bb1), i8* blockaddress(@test, %bb2))
+  %dst3 = call i8* @bar(i32 3, i8* blockaddress(@test, %bb1), i8* blockaddress(@test, %bb2))
+  %cc1 = icmp eq i32 %a, 21
+  br i1 %cc1, label %cc1t, label %cc1f
+
+cc1t:
+  store i32 %a, i32* %p
+  indirectbr i8* %dst3, [label %bb1, label %bb2]
+
+cc1f:
+  %cc2 = icmp ne i32 %a2, 42
+  br i1 %cc2, label %cc2t, label %bb1
+cc2t:
+  store i32 %a, i32* %p2
+  indirectbr i8* %dst1, [label %bb1, label %bb2]
+
+bb1:
+  %ret_bb1 = call i32 @foo(i32 1234)
+  ret i32 %ret_bb1
+bb2:
+  %ret_bb2 = call i32 @foo(i32 4567)
+  ret i32 %ret_bb2
+}
