[VPlan] Update HCFG builder after 7293720 to fix leak.

fhahn · aaryanshukla · commit 460540dc44d3 · 2024-07-14T17:59:38.000Z
Update buildPlainCFG to re-use the vector and latch VPBBs created as part of the initial skeleton in 7293720. This should fix the leak sanitizer failure discovered by https://lab.llvm.org/buildbot/#/builders/52/builds/619.
diff --git a/llvm/lib/Transforms/Vectorize/VPlanHCFGBuilder.cpp b/llvm/lib/Transforms/Vectorize/VPlanHCFGBuilder.cpp
@@ -346,9 +346,24 @@ void PlainCFGBuilder::buildPlainCFG() {
   // latter.
   BB2VPBB[ThePreheaderBB] = VectorPreheaderVPBB;
   BasicBlock *LoopExitBB = TheLoop->getUniqueExitBlock();
+  Loop2Region[LI->getLoopFor(TheLoop->getHeader())] = TheRegion;
   assert(LoopExitBB && "Loops with multiple exits are not supported.");
   BB2VPBB[LoopExitBB] = cast<VPBasicBlock>(TheRegion->getSingleSuccessor());
 
+  // The existing vector region's entry and exiting VPBBs correspond to the loop
+  // header and latch.
+  VPBasicBlock *VectorHeaderVPBB = TheRegion->getEntryBasicBlock();
+  VPBasicBlock *VectorLatchVPBB = TheRegion->getExitingBasicBlock();
+  BB2VPBB[TheLoop->getHeader()] = VectorHeaderVPBB;
+  VectorHeaderVPBB->clearSuccessors();
+  VectorLatchVPBB->clearPredecessors();
+  if (TheLoop->getHeader() != TheLoop->getLoopLatch()) {
+    BB2VPBB[TheLoop->getLoopLatch()] = VectorLatchVPBB;
+  } else {
+    TheRegion->setExiting(VectorHeaderVPBB);
+    delete VectorLatchVPBB;
+  }
+
   // 1. Scan the body of the loop in a topological order to visit each basic
   // block after having visited its predecessor basic blocks. Create a VPBB for
   // each BB and link it to its successor and predecessor VPBBs. Note that
@@ -413,10 +428,11 @@ void PlainCFGBuilder::buildPlainCFG() {
     // of VPBB and it should be set to the exit, i.e., non-header successor,
     // except for the top region, whose successor was set when creating VPlan's
     // skeleton.
-    if (TheRegion != Region)
+    if (TheRegion != Region) {
       Region->setOneSuccessor(isHeaderVPBB(Successor0) ? Successor1
                                                        : Successor0);
-    Region->setExiting(VPBB);
+      Region->setExiting(VPBB);
+    }
   }
 
   // 2. The whole CFG has been built at this point so all the input Values must
diff --git a/llvm/test/Transforms/LoopVectorize/AArch64/outer_loop_prefer_scalable.ll b/llvm/test/Transforms/LoopVectorize/AArch64/outer_loop_prefer_scalable.ll
@@ -46,7 +46,7 @@ define void @foo() {
 ; CHECK-NEXT:    [[TMP14:%.*]] = icmp eq <vscale x 4 x i64> [[TMP13]], shufflevector (<vscale x 4 x i64> insertelement (<vscale x 4 x i64> poison, i64 512, i64 0), <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer)
 ; CHECK-NEXT:    [[TMP15:%.*]] = extractelement <vscale x 4 x i1> [[TMP14]], i32 0
 ; CHECK-NEXT:    br i1 [[TMP15]], label [[OUTER_LOOP_LATCH4]], label [[INNER_LOOP1]]
-; CHECK:       outer_loop_latch4:
+; CHECK:       vector.latch:
 ; CHECK-NEXT:    [[VEC_PHI5:%.*]] = phi <vscale x 4 x float> [ [[TMP12]], [[INNER_LOOP1]] ]
 ; CHECK-NEXT:    call void @llvm.masked.scatter.nxv4f32.nxv4p0(<vscale x 4 x float> [[VEC_PHI5]], <vscale x 4 x ptr> [[TMP10]], i32 4, <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer))
 ; CHECK-NEXT:    [[TMP16:%.*]] = add nuw nsw <vscale x 4 x i64> [[VEC_IND]], shufflevector (<vscale x 4 x i64> insertelement (<vscale x 4 x i64> poison, i64 1, i64 0), <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer)
diff --git a/llvm/test/Transforms/LoopVectorize/dbg-outer-loop-vect.ll b/llvm/test/Transforms/LoopVectorize/dbg-outer-loop-vect.ll
@@ -29,7 +29,7 @@ define void @foo(ptr %h) !dbg !4 {
 ; CHECK-NEXT:    [[TMP5:%.*]] = icmp eq <4 x i64> [[TMP4]], <i64 5, i64 5, i64 5, i64 5>, !dbg [[DBG25:![0-9]+]]
 ; CHECK-NEXT:    [[TMP6:%.*]] = extractelement <4 x i1> [[TMP5]], i32 0, !dbg [[DBG26:![0-9]+]]
 ; CHECK-NEXT:    br i1 [[TMP6]], label [[FOR_COND_CLEANUP32]], label [[FOR_COND5_PREHEADER1]], !dbg [[DBG26]]
-; CHECK:       for.cond.cleanup32:
+; CHECK:       vector.latch:
 ; CHECK-NEXT:    [[TMP7:%.*]] = add nuw nsw <4 x i64> [[VEC_IND]], <i64 1, i64 1, i64 1, i64 1>, !dbg [[DBG27:![0-9]+]]
 ; CHECK-NEXT:    [[TMP8:%.*]] = icmp eq <4 x i64> [[TMP7]], <i64 23, i64 23, i64 23, i64 23>, !dbg [[DBG28:![0-9]+]]
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
diff --git a/llvm/test/Transforms/LoopVectorize/outer-loop-vec-phi-predecessor-order.ll b/llvm/test/Transforms/LoopVectorize/outer-loop-vec-phi-predecessor-order.ll
@@ -40,7 +40,7 @@ define void @test(ptr %src, i64 %n) {
 ; CHECK-NEXT:    [[TMP6:%.*]] = icmp eq <4 x i64> [[TMP5]], [[BROADCAST_SPLAT]]
 ; CHECK-NEXT:    [[TMP7:%.*]] = extractelement <4 x i1> [[TMP6]], i32 0
 ; CHECK-NEXT:    br i1 [[TMP7]], label [[LOOP_1_LATCH5]], label [[LOOP_2_HEADER1]]
-; CHECK:       loop.1.latch5:
+; CHECK:       vector.latch:
 ; CHECK-NEXT:    [[TMP8:%.*]] = add nuw nsw <4 x i64> [[VEC_IND]], <i64 1, i64 1, i64 1, i64 1>
 ; CHECK-NEXT:    [[TMP9:%.*]] = icmp eq <4 x i64> [[TMP8]], [[BROADCAST_SPLAT]]
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
diff --git a/llvm/test/Transforms/LoopVectorize/outer_loop_hcfg_construction.ll b/llvm/test/Transforms/LoopVectorize/outer_loop_hcfg_construction.ll
@@ -51,7 +51,7 @@ define void @non_outermost_loop_hcfg_construction(i64 %n, ptr %a) {
 ; CHECK-NEXT:    [[TMP6:%.*]] = icmp eq <4 x i64> [[TMP5]], [[BROADCAST_SPLAT3]]
 ; CHECK-NEXT:    [[TMP7:%.*]] = extractelement <4 x i1> [[TMP6]], i32 0
 ; CHECK-NEXT:    br i1 [[TMP7]], label [[MIDDLE_LOOP_LATCH4]], label [[INNERMOST_LOOP1]]
-; CHECK:       middle.loop.latch4:
+; CHECK:       vector.latch:
 ; CHECK-NEXT:    [[TMP8:%.*]] = add nuw nsw <4 x i64> [[VEC_IND]], <i64 1, i64 1, i64 1, i64 1>
 ; CHECK-NEXT:    [[TMP9:%.*]] = icmp eq <4 x i64> [[TMP8]], [[BROADCAST_SPLAT3]]
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
@@ -196,7 +196,7 @@ define void @non_outermost_loop_hcfg_construction_other_loops_at_same_level(i64
 ; CHECK-NEXT:    [[TMP6:%.*]] = icmp eq <4 x i64> [[TMP5]], [[BROADCAST_SPLAT3]]
 ; CHECK-NEXT:    [[TMP7:%.*]] = extractelement <4 x i1> [[TMP6]], i32 0
 ; CHECK-NEXT:    br i1 [[TMP7]], label [[MIDDLE_LOOP_J0_CLEANUP4]], label [[INNERMOST_LOOP1]]
-; CHECK:       middle.loop.j0.cleanup4:
+; CHECK:       vector.latch:
 ; CHECK-NEXT:    [[TMP8:%.*]] = add nuw nsw <4 x i64> [[VEC_IND]], <i64 1, i64 1, i64 1, i64 1>
 ; CHECK-NEXT:    [[TMP9:%.*]] = icmp eq <4 x i64> [[TMP8]], [[BROADCAST_SPLAT3]]
 ; CHECK-NEXT:    [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
diff --git a/llvm/test/Transforms/LoopVectorize/outer_loop_scalable.ll b/llvm/test/Transforms/LoopVectorize/outer_loop_scalable.ll
@@ -52,7 +52,7 @@ define void @foo() {
 ; CHECK-NEXT:    [[TMP14:%.*]] = icmp eq <vscale x 4 x i64> [[TMP13]], shufflevector (<vscale x 4 x i64> insertelement (<vscale x 4 x i64> poison, i64 512, i64 0), <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer)
 ; CHECK-NEXT:    [[TMP15:%.*]] = extractelement <vscale x 4 x i1> [[TMP14]], i32 0
 ; CHECK-NEXT:    br i1 [[TMP15]], label [[OUTER_LOOP_LATCH4]], label [[INNER_LOOP1]]
-; CHECK:       outer_loop_latch4:
+; CHECK:       vector.latch:
 ; CHECK-NEXT:    [[VEC_PHI5:%.*]] = phi <vscale x 4 x float> [ [[TMP12]], [[INNER_LOOP1]] ]
 ; CHECK-NEXT:    call void @llvm.masked.scatter.nxv4f32.nxv4p0(<vscale x 4 x float> [[VEC_PHI5]], <vscale x 4 x ptr> [[TMP10]], i32 4, <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i64 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer))
 ; CHECK-NEXT:    [[TMP16:%.*]] = add nuw nsw <vscale x 4 x i64> [[VEC_IND]], shufflevector (<vscale x 4 x i64> insertelement (<vscale x 4 x i64> poison, i64 1, i64 0), <vscale x 4 x i64> poison, <vscale x 4 x i32> zeroinitializer)
diff --git a/llvm/test/Transforms/LoopVectorize/vplan-printing-outer-loop.ll b/llvm/test/Transforms/LoopVectorize/vplan-printing-outer-loop.ll
@@ -32,9 +32,9 @@ define void @foo(i64 %n) {
 ; CHECK-NEXT:       EMIT branch-on-cond ir<%inner.ec>
 ; CHECK-NEXT:   No successors
 ; CHECK-NEXT:  }
-; CHECK-NEXT:  Successor(s): outer.latch
+; CHECK-NEXT:  Successor(s): vector.latch
 ; CHECK-EMPTY:
-; CHECK-NEXT:   outer.latch:
+; CHECK-NEXT:   vector.latch:
 ; CHECK-NEXT:     EMIT ir<%outer.iv.next> = add ir<%outer.iv>, ir<1>
 ; CHECK-NEXT:     EMIT ir<%outer.ec> = icmp ir<%outer.iv.next>, ir<8>
 ; CHECK-NEXT:     EMIT branch-on-cond ir<%outer.ec>
