[llvm] [VPlan] Connect Entry to scalar preheader during initial construction. (PR #140132)
Florian Hahn via llvm-commits
llvm-commits at lists.llvm.org
Sat May 17 09:56:18 PDT 2025
https://github.com/fhahn updated https://github.com/llvm/llvm-project/pull/140132
>From 4acf4d41f15d34d33681b111fa9b627094a714f9 Mon Sep 17 00:00:00 2001
From: Florian Hahn <flo at fhahn.com>
Date: Thu, 15 May 2025 20:52:40 +0100
Subject: [PATCH] [VPlan] Connect Entry to scalar preheader during initial
construction.
Update initial construction to connect the Plan's entry to the scalar
preheader during initial construction. This moves a small part of the
skeleton creation out of ILV and will also enable replacing
VPInstruction::ResumePhi with regular VPPhi recipes.
Resume phis need 2 incoming values to start with, the second being the
bypass value from the scalar ph (and used to replicate the incoming
value for other bypass blocks). Adding the extra edge ensures we
incoming values for resume phis match the incoming blocks.
Still needs various VPlan printing tests to be updated.
vplan-printing.ll shows the changes.
---
.../Transforms/Vectorize/LoopVectorize.cpp | 26 ++---
llvm/lib/Transforms/Vectorize/VPlan.h | 3 +-
.../Vectorize/VPlanConstruction.cpp | 6 +
.../Transforms/Vectorize/VPlanTransforms.cpp | 2 +-
.../LoopVectorize/vplan-printing.ll | 106 +++++++++---------
5 files changed, 72 insertions(+), 71 deletions(-)
diff --git a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
index 148b187f4f25d..39b0308943685 100644
--- a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
+++ b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
@@ -490,7 +490,7 @@ class InnerLoopVectorizer {
MinProfitableTripCount(MinProfitableTripCount), UF(UnrollFactor),
Builder(PSE.getSE()->getContext()), Cost(CM), BFI(BFI), PSI(PSI),
RTChecks(RTChecks), Plan(Plan),
- VectorPHVPB(Plan.getEntry()->getSingleSuccessor()) {}
+ VectorPHVPB(Plan.getEntry()->getSuccessors()[1]) {}
virtual ~InnerLoopVectorizer() = default;
@@ -2365,14 +2365,11 @@ InnerLoopVectorizer::getOrCreateVectorTripCount(BasicBlock *InsertBlock) {
void InnerLoopVectorizer::introduceCheckBlockInVPlan(BasicBlock *CheckIRBB) {
VPBlockBase *ScalarPH = Plan.getScalarPreheader();
VPBlockBase *PreVectorPH = VectorPHVPB->getSinglePredecessor();
- if (PreVectorPH->getNumSuccessors() != 1) {
- assert(PreVectorPH->getNumSuccessors() == 2 && "Expected 2 successors");
- assert(PreVectorPH->getSuccessors()[0] == ScalarPH &&
- "Unexpected successor");
- VPIRBasicBlock *CheckVPIRBB = Plan.createVPIRBasicBlock(CheckIRBB);
- VPBlockUtils::insertOnEdge(PreVectorPH, VectorPHVPB, CheckVPIRBB);
- PreVectorPH = CheckVPIRBB;
- }
+ assert(PreVectorPH->getNumSuccessors() == 2 && "Expected 2 successors");
+ assert(PreVectorPH->getSuccessors()[0] == ScalarPH && "Unexpected successor");
+ VPIRBasicBlock *CheckVPIRBB = Plan.createVPIRBasicBlock(CheckIRBB);
+ VPBlockUtils::insertOnEdge(PreVectorPH, VectorPHVPB, CheckVPIRBB);
+ PreVectorPH = CheckVPIRBB;
VPBlockUtils::connectBlocks(PreVectorPH, ScalarPH);
PreVectorPH->swapSuccessors();
@@ -2463,9 +2460,6 @@ void InnerLoopVectorizer::emitIterationCountCheck(BasicBlock *Bypass) {
setBranchWeights(BI, MinItersBypassWeights, /*IsExpected=*/false);
ReplaceInstWithInst(TCCheckBlock->getTerminator(), &BI);
LoopBypassBlocks.push_back(TCCheckBlock);
-
- // TODO: Wrap LoopVectorPreHeader in VPIRBasicBlock here.
- introduceCheckBlockInVPlan(TCCheckBlock);
}
BasicBlock *InnerLoopVectorizer::emitSCEVChecks(BasicBlock *Bypass) {
@@ -7837,7 +7831,7 @@ DenseMap<const SCEV *, Value *> LoopVectorizationPlanner::executePlan(
// 1. Set up the skeleton for vectorization, including vector pre-header and
// middle block. The vector loop is created during VPlan execution.
- VPBasicBlock *VectorPH = cast<VPBasicBlock>(Entry->getSingleSuccessor());
+ VPBasicBlock *VectorPH = cast<VPBasicBlock>(Entry->getSuccessors()[1]);
State.CFG.PrevBB = ILV.createVectorizedLoopSkeleton();
if (VectorizingEpilogue)
VPlanTransforms::removeDeadRecipes(BestVPlan);
@@ -8070,7 +8064,8 @@ EpilogueVectorizerMainLoop::emitIterationCountCheck(BasicBlock *Bypass,
setBranchWeights(BI, MinItersBypassWeights, /*IsExpected=*/false);
ReplaceInstWithInst(TCCheckBlock->getTerminator(), &BI);
- introduceCheckBlockInVPlan(TCCheckBlock);
+ if (!ForEpilogue)
+ introduceCheckBlockInVPlan(TCCheckBlock);
return TCCheckBlock;
}
@@ -8200,7 +8195,6 @@ EpilogueVectorizerEpilogueLoop::emitMinimumVectorEpilogueIterCountCheck(
Plan.setEntry(NewEntry);
// OldEntry is now dead and will be cleaned up when the plan gets destroyed.
- introduceCheckBlockInVPlan(Insert);
return Insert;
}
@@ -9160,7 +9154,7 @@ static void addScalarResumePhis(VPRecipeBuilder &Builder, VPlan &Plan,
DenseMap<VPValue *, VPValue *> &IVEndValues) {
VPTypeAnalysis TypeInfo(Plan.getCanonicalIV()->getScalarType());
auto *ScalarPH = Plan.getScalarPreheader();
- auto *MiddleVPBB = cast<VPBasicBlock>(ScalarPH->getSinglePredecessor());
+ auto *MiddleVPBB = cast<VPBasicBlock>(ScalarPH->getPredecessors()[0]);
VPRegionBlock *VectorRegion = Plan.getVectorLoopRegion();
VPBuilder VectorPHBuilder(
cast<VPBasicBlock>(VectorRegion->getSinglePredecessor()));
diff --git a/llvm/lib/Transforms/Vectorize/VPlan.h b/llvm/lib/Transforms/Vectorize/VPlan.h
index e634de1e17c69..788e8c37ed0a3 100644
--- a/llvm/lib/Transforms/Vectorize/VPlan.h
+++ b/llvm/lib/Transforms/Vectorize/VPlan.h
@@ -4246,7 +4246,8 @@ class VPlan {
/// that this relies on unneeded branches to the scalar tail loop being
/// removed.
bool hasScalarTail() const {
- return getScalarPreheader()->getNumPredecessors() != 0;
+ return getScalarPreheader()->getNumPredecessors() != 0 &&
+ getScalarPreheader()->getSinglePredecessor() != getEntry();
}
};
diff --git a/llvm/lib/Transforms/Vectorize/VPlanConstruction.cpp b/llvm/lib/Transforms/Vectorize/VPlanConstruction.cpp
index 287bc93ce496a..78c6e2f98387f 100644
--- a/llvm/lib/Transforms/Vectorize/VPlanConstruction.cpp
+++ b/llvm/lib/Transforms/Vectorize/VPlanConstruction.cpp
@@ -546,6 +546,9 @@ void VPlanTransforms::prepareForVectorization(
if (auto *LatchExitVPB = MiddleVPBB->getSingleSuccessor())
VPBlockUtils::disconnectBlocks(MiddleVPBB, LatchExitVPB);
VPBlockUtils::connectBlocks(MiddleVPBB, ScalarPH);
+ VPBlockUtils::connectBlocks(Plan.getEntry(), ScalarPH);
+ Plan.getEntry()->swapSuccessors();
+
// The exit blocks are unreachable, remove their recipes to make sure no
// users remain that may pessimize transforms.
for (auto *EB : Plan.getExitBlocks()) {
@@ -558,6 +561,9 @@ void VPlanTransforms::prepareForVectorization(
// The connection order corresponds to the operands of the conditional branch,
// with the middle block already connected to the exit block.
VPBlockUtils::connectBlocks(MiddleVPBB, ScalarPH);
+ // Also connect the entry block to the scalar preheader.
+ VPBlockUtils::connectBlocks(Plan.getEntry(), ScalarPH);
+ Plan.getEntry()->swapSuccessors();
auto *ScalarLatchTerm = TheLoop->getLoopLatch()->getTerminator();
// Here we use the same DebugLoc as the scalar loop latch terminator instead
diff --git a/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp b/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
index f1c466e3208be..f5be8dd992198 100644
--- a/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
+++ b/llvm/lib/Transforms/Vectorize/VPlanTransforms.cpp
@@ -1732,7 +1732,7 @@ static void removeBranchOnCondTrue(VPlan &Plan) {
using namespace llvm::VPlanPatternMatch;
for (VPBasicBlock *VPBB : VPBlockUtils::blocksOnly<VPBasicBlock>(
vp_depth_first_shallow(Plan.getEntry()))) {
- if (VPBB->getNumSuccessors() != 2 ||
+ if (VPBB->getNumSuccessors() != 2 || isa<VPIRBasicBlock>(VPBB) ||
!match(&VPBB->back(), m_BranchOnCond(m_True())))
continue;
diff --git a/llvm/test/Transforms/LoopVectorize/vplan-printing.ll b/llvm/test/Transforms/LoopVectorize/vplan-printing.ll
index 727784de2cb6d..3504558d1b545 100644
--- a/llvm/test/Transforms/LoopVectorize/vplan-printing.ll
+++ b/llvm/test/Transforms/LoopVectorize/vplan-printing.ll
@@ -15,7 +15,7 @@ define void @print_call_and_memory(i64 %n, ptr noalias %y, ptr noalias %x) nounw
; CHECK-NEXT: Live-in ir<%n> = original trip-count
; CHECK-EMPTY:
; CHECK-NEXT: ir-bb<for.body.preheader>:
-; CHECK-NEXT: Successor(s): vector.ph
+; CHECK-NEXT: Successor(s): scalar.ph, vector.ph
; CHECK-EMPTY:
; CHECK-NEXT: vector.ph:
; CHECK-NEXT: Successor(s): vector loop
@@ -42,6 +42,9 @@ define void @print_call_and_memory(i64 %n, ptr noalias %y, ptr noalias %x) nounw
; CHECK-NEXT: EMIT branch-on-cond vp<[[CMP]]>
; CHECK-NEXT: Successor(s): ir-bb<for.end.loopexit>, scalar.ph
; CHECK-EMPTY:
+; CHECK-NEXT: ir-bb<for.end.loopexit>
+; CHECK-NEXT: No successors
+; CHECK-EMPTY:
; CHECK-NEXT: scalar.ph
; CHECK-NEXT: EMIT vp<[[RESUME_IV:%.+]]> = resume-phi vp<[[VTC]]>, ir<0>
; CHECK-NEXT: Successor(s): ir-bb<for.body>
@@ -50,9 +53,6 @@ define void @print_call_and_memory(i64 %n, ptr noalias %y, ptr noalias %x) nounw
; CHECK-NEXT: IR %iv = phi i64 [ %iv.next, %for.body ], [ 0, %for.body.preheader ] (extra operand: vp<[[RESUME_IV]]> from scalar.ph)
; CHECK: IR %exitcond = icmp eq i64 %iv.next, %n
; CHECK-NEXT: No successors
-; CHECK-EMPTY:
-; CHECK-NEXT: ir-bb<for.end.loopexit>
-; CHECK-NEXT: No successors
; CHECK-NEXT: }
;
entry:
@@ -83,7 +83,7 @@ define void @print_widen_gep_and_select(i64 %n, ptr noalias %y, ptr noalias %x,
; CHECK-NEXT: Live-in ir<%n> = original trip-count
; CHECK-EMPTY:
; CHECK-NEXT: ir-bb<for.body.preheader>:
-; CHECK-NEXT: Successor(s): vector.ph
+; CHECK-NEXT: Successor(s): scalar.ph, vector.ph
; CHECK-EMPTY:
; CHECK-NEXT: vector.ph:
; CHECK-NEXT: Successor(s): vector loop
@@ -113,6 +113,9 @@ define void @print_widen_gep_and_select(i64 %n, ptr noalias %y, ptr noalias %x,
; CHECK-NEXT: EMIT branch-on-cond vp<[[CMP]]>
; CHECK-NEXT: Successor(s): ir-bb<for.end.loopexit>, scalar.ph
; CHECK-EMPTY:
+; CHECK-NEXT: ir-bb<for.end.loopexit>
+; CHECK-NEXT: No successors
+; CHECK-EMPTY:
; CHECK-NEXT: scalar.ph
; CHECK-NEXT: EMIT vp<[[RESUME_IV:%.+]]> = resume-phi vp<[[VTC]]>, ir<0>
; CHECK-NEXT: Successor(s): ir-bb<for.body>
@@ -121,9 +124,6 @@ define void @print_widen_gep_and_select(i64 %n, ptr noalias %y, ptr noalias %x,
; CHECK-NEXT: IR %iv = phi i64 [ %iv.next, %for.body ], [ 0, %for.body.preheader ] (extra operand: vp<[[RESUME_IV]]> from scalar.ph)
; CHECK: IR %exitcond = icmp eq i64 %iv.next, %n
; CHECK-NEXT: No successors
-; CHECK-EMPTY:
-; CHECK-NEXT: ir-bb<for.end.loopexit>
-; CHECK-NEXT: No successors
; CHECK-NEXT: }
;
entry:
@@ -157,7 +157,7 @@ define void @print_replicate_predicated_phi(i64 %n, ptr %x) {
; CHECK-EMPTY:
; CHECK-NEXT: ir-bb<entry>:
; CHECK-NEXT: EMIT vp<[[TC]]> = EXPAND SCEV (1 smax %n)
-; CHECK-NEXT: Successor(s): vector.ph
+; CHECK-NEXT: Successor(s): scalar.ph, vector.ph
; CHECK-EMPTY:
; CHECK-NEXT: vector.ph:
; CHECK-NEXT: Successor(s): vector loop
@@ -202,6 +202,9 @@ define void @print_replicate_predicated_phi(i64 %n, ptr %x) {
; CHECK-NEXT: EMIT branch-on-cond vp<[[CMP]]>
; CHECK-NEXT: Successor(s): ir-bb<for.end>, scalar.ph
; CHECK-EMPTY:
+; CHECK-NEXT: ir-bb<for.end>
+; CHECK-NEXT: No successors
+; CHECK-EMPTY:
; CHECK-NEXT: scalar.ph
; CHECK-NEXT: EMIT vp<[[RESUME_IV:%.+]]> = resume-phi vp<[[VTC]]>, ir<0>
; CHECK-NEXT: Successor(s): ir-bb<for.body>
@@ -210,9 +213,6 @@ define void @print_replicate_predicated_phi(i64 %n, ptr %x) {
; CHECK-NEXT: IR %i = phi i64 [ 0, %entry ], [ %i.next, %for.inc ] (extra operand: vp<[[RESUME_IV]]> from scalar.ph)
; CHECK-NEXT: IR %cmp = icmp ult i64 %i, 5
; CHECK-NEXT: No successors
-; CHECK-EMPTY:
-; CHECK-NEXT: ir-bb<for.end>
-; CHECK-NEXT: No successors
; CHECK-NEXT: }
;
entry:
@@ -251,7 +251,7 @@ define void @print_interleave_groups(i32 %C, i32 %D) {
; CHECK-NEXT: Live-in ir<256> = original trip-count
; CHECK-EMPTY:
; CHECK-NEXT: ir-bb<entry>:
-; CHECK-NEXT: Successor(s): vector.ph
+; CHECK-NEXT: Successor(s): scalar.ph, vector.ph
; CHECK-EMPTY:
; CHECK-NEXT: vector.ph:
; CHECK-NEXT: vp<[[IV_END:%.+]]> = DERIVED-IV ir<0> + vp<[[VTC]]> * ir<4>
@@ -285,6 +285,9 @@ define void @print_interleave_groups(i32 %C, i32 %D) {
; CHECK-NEXT: EMIT branch-on-cond vp<[[CMP]]>
; CHECK-NEXT: Successor(s): ir-bb<for.end>, scalar.ph
; CHECK-EMPTY:
+; CHECK-NEXT: ir-bb<for.end>
+; CHECK-NEXT: No successors
+; CHECK-EMPTY:
; CHECK-NEXT: scalar.ph
; CHECK-NEXT: EMIT vp<[[RESUME_IV:%.+]]> = resume-phi vp<[[IV_END]]>, ir<0>
; CHECK-NEXT: Successor(s): ir-bb<for.body>
@@ -293,9 +296,6 @@ define void @print_interleave_groups(i32 %C, i32 %D) {
; CHECK-NEXT: IR %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ] (extra operand: vp<[[RESUME_IV]]> from scalar.ph)
; CHECK: IR %cmp = icmp slt i64 %iv.next, 1024
; CHECK-NEXT: No successors
-; CHECK-EMPTY:
-; CHECK-NEXT: ir-bb<for.end>
-; CHECK-NEXT: No successors
; CHECK-NEXT: }
;
entry:
@@ -338,7 +338,7 @@ define void @debug_loc_vpinstruction(ptr nocapture %asd, ptr nocapture %bsd) !db
; CHECK-NEXT: Live-in ir<128> = original trip-count
; CHECK-EMPTY:
; CHECK-NEXT: ir-bb<entry>:
-; CHECK-NEXT: Successor(s): vector.ph
+; CHECK-NEXT: Successor(s): scalar.ph, vector.ph
; CHECK-EMPTY:
; CHECK-NEXT: vector.ph:
; CHECK-NEXT: Successor(s): vector loop
@@ -390,6 +390,9 @@ define void @debug_loc_vpinstruction(ptr nocapture %asd, ptr nocapture %bsd) !db
; CHECK-NEXT: EMIT branch-on-cond vp<[[CMP]]>
; CHECK-NEXT: Successor(s): ir-bb<exit>, scalar.ph
; CHECK-EMPTY:
+; CHECK-NEXT: ir-bb<exit>
+; CHECK-NEXT: No successors
+; CHECK-EMPTY:
; CHECK-NEXT: scalar.ph:
; CHECK-NEXT: EMIT vp<[[RESUME_IV:%.+]]> = resume-phi vp<[[VTC]]>, ir<0>
; CHECK-NEXT: Successor(s): ir-bb<loop>
@@ -398,9 +401,6 @@ define void @debug_loc_vpinstruction(ptr nocapture %asd, ptr nocapture %bsd) !db
; CHECK-NEXT: IR %iv = phi i64 [ 0, %entry ], [ %iv.next, %if.end ] (extra operand: vp<[[RESUME_IV]]> from scalar.ph)
; CHECK: IR %cmp1 = icmp slt i32 %lsd, 100
; CHECK-NEXT: No successors
-; CHECK-EMPTY:
-; CHECK-NEXT: ir-bb<exit>
-; CHECK-NEXT: No successors
; CHECK-NEXT:}
;
entry:
@@ -449,7 +449,7 @@ define void @print_expand_scev(i64 %y, ptr %ptr) {
; CHECK-NEXT: IR %inc = add i64 %div, 1
; CHECK-NEXT: EMIT vp<[[TC]]> = EXPAND SCEV (1 + ((15 + (%y /u 492802768830814060))<nuw><nsw> /u (1 + (%y /u 492802768830814060))<nuw><nsw>))<nuw><nsw>
; CHECK-NEXT: EMIT vp<[[EXP_SCEV:%.+]]> = EXPAND SCEV (1 + (%y /u 492802768830814060))<nuw><nsw>
-; CHECK-NEXT: Successor(s): vector.ph
+; CHECK-NEXT: Successor(s): scalar.ph, vector.ph
; CHECK-EMPTY:
; CHECK-NEXT: vector.ph:
; CHECK-NEXT: vp<[[IV_END:%.+]]> = DERIVED-IV ir<0> + vp<[[VTC]]> * vp<[[EXP_SCEV]]>
@@ -475,6 +475,9 @@ define void @print_expand_scev(i64 %y, ptr %ptr) {
; CHECK-NEXT: EMIT branch-on-cond vp<[[CMP]]>
; CHECK-NEXT: Successor(s): ir-bb<loop.exit>, scalar.ph
; CHECK-EMPTY:
+; CHECK-NEXT: ir-bb<loop.exit>
+; CHECK-NEXT: No successors
+; CHECK-EMPTY:
; CHECK-NEXT: scalar.ph
; CHECK-NEXT: EMIT vp<[[RESUME_IV:%.+]]> = resume-phi vp<[[IV_END]]>, ir<0>
; CHECK-NEXT: Successor(s): ir-bb<loop>
@@ -483,9 +486,6 @@ define void @print_expand_scev(i64 %y, ptr %ptr) {
; CHECK-NEXT: IR %iv = phi i64 [ %iv.next, %loop ], [ 0, %entry ] (extra operand: vp<[[RESUME_IV]]> from scalar.ph)
; CHECK: IR %iv.next = add i64 %iv, %inc
; CHECK-NEXT: No successors
-; CHECK-EMPTY:
-; CHECK-NEXT: ir-bb<loop.exit>
-; CHECK-NEXT: No successors
; CHECK-NEXT: }
;
entry:
@@ -517,7 +517,7 @@ define i32 @print_exit_value(ptr %ptr, i32 %off) {
; CHECK-NEXT: Live-in ir<1000> = original trip-count
; CHECK-EMPTY:
; CHECK-NEXT: ir-bb<entry>:
-; CHECK-NEXT: Successor(s): vector.ph
+; CHECK-NEXT: Successor(s): scalar.ph, vector.ph
; CHECK-EMPTY:
; CHECK-NEXT: vector.ph:
; CHECK-NEXT: Successor(s): vector loop
@@ -543,6 +543,10 @@ define i32 @print_exit_value(ptr %ptr, i32 %off) {
; CHECK-NEXT: EMIT branch-on-cond vp<[[CMP]]>
; CHECK-NEXT: Successor(s): ir-bb<exit>, scalar.ph
; CHECK-EMPTY:
+; CHECK-NEXT: ir-bb<exit>
+; CHECK-NEXT: IR %lcssa = phi i32 [ %add, %loop ] (extra operand: vp<[[EXIT]]> from middle.block)
+; CHECK-NEXT: No successors
+; CHECK-EMPTY:
; CHECK-NEXT: scalar.ph
; CHECK-NEXT: EMIT vp<[[RESUME_IV:%.+]]> = resume-phi vp<[[VTC]]>, ir<0>
; CHECK-NEXT: Successor(s): ir-bb<loop>
@@ -551,10 +555,6 @@ define i32 @print_exit_value(ptr %ptr, i32 %off) {
; CHECK-NEXT: IR %iv = phi i32 [ 0, %entry ], [ %iv.next, %loop ] (extra operand: vp<[[RESUME_IV]]> from scalar.ph)
; CHECK: IR %ec = icmp eq i32 %iv.next, 1000
; CHECK-NEXT: No successors
-; CHECK-EMPTY:
-; CHECK-NEXT: ir-bb<exit>
-; CHECK-NEXT: IR %lcssa = phi i32 [ %add, %loop ] (extra operand: vp<[[EXIT]]> from middle.block)
-; CHECK-NEXT: No successors
; CHECK-NEXT: }
;
entry:
@@ -583,7 +583,7 @@ define void @print_fast_math_flags(i64 %n, ptr noalias %y, ptr noalias %x, ptr %
; CHECK-NEXT: Live-in ir<%n> = original trip-count
; CHECK-EMPTY:
; CHECK-NEXT: ir-bb<entry>:
-; CHECK-NEXT: Successor(s): vector.ph
+; CHECK-NEXT: Successor(s): scalar.ph, vector.ph
; CHECK-EMPTY:
; CHECK-NEXT: vector.ph:
; CHECK-NEXT: Successor(s): vector loop
@@ -612,6 +612,9 @@ define void @print_fast_math_flags(i64 %n, ptr noalias %y, ptr noalias %x, ptr %
; CHECK-NEXT: EMIT branch-on-cond vp<[[CMP]]>
; CHECK-NEXT: Successor(s): ir-bb<exit>, scalar.ph
; CHECK-EMPTY:
+; CHECK-NEXT: ir-bb<exit>
+; CHECK-NEXT: No successors
+; CHECK-EMPTY:
; CHECK-NEXT: scalar.ph
; CHECK-NEXT: EMIT vp<[[RESUME_IV:%.+]]> = resume-phi vp<[[VTC]]>, ir<0>
; CHECK-NEXT: Successor(s): ir-bb<loop>
@@ -620,9 +623,6 @@ define void @print_fast_math_flags(i64 %n, ptr noalias %y, ptr noalias %x, ptr %
; CHECK-NEXT: IR %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
; CHECK: IR %exitcond = icmp eq i64 %iv.next, %n
; CHECK-NEXT: No successors
-; CHECK-EMPTY:
-; CHECK-NEXT: ir-bb<exit>
-; CHECK-NEXT: No successors
; CHECK-NEXT: }
;
entry:
@@ -654,7 +654,7 @@ define void @print_exact_flags(i64 %n, ptr noalias %x) {
; CHECK-NEXT: Live-in ir<%n> = original trip-count
; CHECK-EMPTY:
; CHECK-NEXT: ir-bb<entry>:
-; CHECK-NEXT: Successor(s): vector.ph
+; CHECK-NEXT: Successor(s): scalar.ph, vector.ph
; CHECK-EMPTY:
; CHECK-NEXT: vector.ph:
; CHECK-NEXT: Successor(s): vector loop
@@ -682,6 +682,9 @@ define void @print_exact_flags(i64 %n, ptr noalias %x) {
; CHECK-NEXT: EMIT branch-on-cond vp<[[CMP]]>
; CHECK-NEXT: Successor(s): ir-bb<exit>, scalar.ph
; CHECK-EMPTY:
+; CHECK-NEXT: ir-bb<exit>
+; CHECK-NEXT: No successors
+; CHECK-EMPTY:
; CHECK-NEXT: scalar.ph
; CHECK-NEXT: EMIT vp<[[RESUME_IV:%.+]]> = resume-phi vp<[[VTC]]>, ir<0>
; CHECK-NEXT: Successor(s): ir-bb<loop>
@@ -690,10 +693,7 @@ define void @print_exact_flags(i64 %n, ptr noalias %x) {
; CHECK-NEXT: IR %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ] (extra operand: vp<[[RESUME_IV]]> from scalar.ph)
; CHECK: IR %exitcond = icmp eq i64 %iv.next, %n
; CHECK-NEXT: No successors
-; CHECK-EMPTY:
-; CHECK-NEXT: ir-bb<exit>
-; CHECK-NEXT: No successors
-; CHECK-NEXT: }
+; ; CHECK-NEXT: }
;
entry:
br label %loop
@@ -723,7 +723,7 @@ define void @print_call_flags(ptr readonly %src, ptr noalias %dest, i64 %n) {
; CHECK-NEXT: Live-in ir<%n> = original trip-count
; CHECK-EMPTY:
; CHECK-NEXT: ir-bb<entry>:
-; CHECK-NEXT: Successor(s): vector.ph
+; CHECK-NEXT: Successor(s): scalar.ph, vector.ph
; CHECK-EMPTY:
; CHECK-NEXT: vector.ph:
; CHECK-NEXT: Successor(s): vector loop
@@ -772,6 +772,9 @@ define void @print_call_flags(ptr readonly %src, ptr noalias %dest, i64 %n) {
; CHECK-NEXT: EMIT branch-on-cond vp<[[CMP]]>
; CHECK-NEXT: Successor(s): ir-bb<end>, scalar.ph
; CHECK-EMPTY:
+; CHECK-NEXT: ir-bb<end>
+; CHECK-NEXT: No successors
+; CHECK-EMPTY:
; CHECK-NEXT: scalar.ph
; CHECK-NEXT: EMIT vp<[[RESUME_IV:%.+]]> = resume-phi vp<[[VTC]]>, ir<0>
; CHECK-NEXT: Successor(s): ir-bb<for.body>
@@ -780,9 +783,6 @@ define void @print_call_flags(ptr readonly %src, ptr noalias %dest, i64 %n) {
; CHECK-NEXT: IR %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.loop ] (extra operand: vp<[[RESUME_IV]]> from scalar.ph)
; CHECK: IR %ifcond = fcmp oeq float %ld.value, 5.0
; CHECK-NEXT: No successors
-; CHECK-EMPTY:
-; CHECK-NEXT: ir-bb<end>
-; CHECK-NEXT: No successors
; CHECK-NEXT: }
;
entry:
@@ -823,7 +823,7 @@ define void @print_disjoint_flags(i64 %n, ptr noalias %x) {
; CHECK-NEXT: Live-in ir<%n> = original trip-count
; CHECK-EMPTY:
; CHECK-NEXT: ir-bb<entry>:
-; CHECK-NEXT: Successor(s): vector.ph
+; CHECK-NEXT: Successor(s): scalar.ph, vector.ph
; CHECK-EMPTY:
; CHECK-NEXT: vector.ph:
; CHECK-NEXT: Successor(s): vector loop
@@ -851,6 +851,9 @@ define void @print_disjoint_flags(i64 %n, ptr noalias %x) {
; CHECK-NEXT: EMIT branch-on-cond vp<[[CMP]]>
; CHECK-NEXT: Successor(s): ir-bb<exit>, scalar.ph
; CHECK-EMPTY:
+; CHECK-NEXT: ir-bb<exit>
+; CHECK-NEXT: No successors
+; CHECK-EMPTY:
; CHECK-NEXT: scalar.ph
; CHECK-NEXT: EMIT vp<[[RESUME_IV:%.+]]> = resume-phi vp<[[VTC]]>, ir<0>
; CHECK-NEXT: Successor(s): ir-bb<loop>
@@ -859,9 +862,6 @@ define void @print_disjoint_flags(i64 %n, ptr noalias %x) {
; CHECK-NEXT: IR %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ] (extra operand: vp<[[RESUME_IV]]> from scalar.ph)
; CHECK: IR %exitcond = icmp eq i64 %iv.next, %n
; CHECK-NEXT: No successors
-; CHECK-EMPTY:
-; CHECK-NEXT: ir-bb<exit>
-; CHECK-NEXT: No successors
; CHECK-NEXT: }
;
entry:
@@ -892,7 +892,7 @@ define void @zext_nneg(ptr noalias %p, ptr noalias %p1) {
; CHECK-NEXT: Live-in ir<1000> = original trip-count
; CHECK-EMPTY:
; CHECK-NEXT: ir-bb<entry>:
-; CHECK-NEXT: Successor(s): vector.ph
+; CHECK-NEXT: Successor(s): scalar.ph, vector.ph
; CHECK-EMPTY:
; CHECK-NEXT: vector.ph:
; CHECK-NEXT: Successor(s): vector loop
@@ -937,7 +937,7 @@ define i16 @print_first_order_recurrence_and_result(ptr %ptr) {
; CHECK-NEXT: Live-in ir<1000> = original trip-count
; CHECK-EMPTY:
; CHECK-NEXT: ir-bb<entry>:
-; CHECK-NEXT: Successor(s): vector.ph
+; CHECK-NEXT: Successor(s): scalar.ph, vector.ph
; CHECK-EMPTY:
; CHECK-NEXT: vector.ph:
; CHECK-NEXT: Successor(s): vector loop
@@ -967,6 +967,10 @@ define i16 @print_first_order_recurrence_and_result(ptr %ptr) {
; CHECK-NEXT: EMIT branch-on-cond vp<[[CMP]]>
; CHECK-NEXT: Successor(s): ir-bb<exit>, scalar.ph
; CHECK-EMPTY:
+; CHECK-NEXT: ir-bb<exit>
+; CHECK-NEXT: IR %for.1.lcssa = phi i16 [ %for.1, %loop ] (extra operand: vp<[[FOR_RESULT]]> from middle.block)
+; CHECK-NEXT: No successors
+; CHECK-EMPTY:
; CHECK-NEXT: scalar.ph
; CHECK-NEXT: EMIT vp<[[RESUME_P:%.*]]> = resume-phi vp<[[RESUME_1]]>, ir<22>
; CHECK-NEXT: EMIT vp<[[RESUME_IV:%.+]]> = resume-phi vp<[[VTC]]>, ir<0>
@@ -977,10 +981,6 @@ define i16 @print_first_order_recurrence_and_result(ptr %ptr) {
; CHECK-NEXT: IR %iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ] (extra operand: vp<[[RESUME_IV]]> from scalar.ph)
; CHECK: IR %exitcond.not = icmp eq i64 %iv.next, 1000
; CHECK-NEXT: No successors
-; CHECK-EMPTY:
-; CHECK-NEXT: ir-bb<exit>
-; CHECK-NEXT: IR %for.1.lcssa = phi i16 [ %for.1, %loop ] (extra operand: vp<[[FOR_RESULT]]> from middle.block)
-; CHECK-NEXT: No successors
; CHECK-NEXT: }
;
entry:
@@ -1010,7 +1010,7 @@ define void @print_select_with_fastmath_flags(ptr noalias %a, ptr noalias %b, pt
; CHECK-NEXT: Live-in ir<%N> = original trip-count
; CHECK-EMPTY:
; CHECK-NEXT: ir-bb<entry>:
-; CHECK-NEXT: Successor(s): vector.ph
+; CHECK-NEXT: Successor(s): scalar.ph, vector.ph
; CHECK-EMPTY:
; CHECK-NEXT: vector.ph:
; CHECK-NEXT: Successor(s): vector loop
More information about the llvm-commits
mailing list