Index: llvm/lib/Transforms/Utils/LoopVersioning.cpp =================================================================== --- llvm/lib/Transforms/Utils/LoopVersioning.cpp +++ llvm/lib/Transforms/Utils/LoopVersioning.cpp @@ -44,11 +44,11 @@ AliasChecks(Checks.begin(), Checks.end()), Preds(LAI.getPSE().getUnionPredicate()), LAI(LAI), LI(LI), DT(DT), SE(SE) { - assert(L->getUniqueExitBlock() && "No single exit block"); } void LoopVersioning::versionLoop( const SmallVectorImpl &DefsUsedOutside) { + assert(VersionedLoop->getUniqueExitBlock() && "No single exit block"); assert(VersionedLoop->isLoopSimplifyForm() && "Loop is not in loop-simplify form"); Index: llvm/lib/Transforms/Vectorize/LoopVectorize.cpp =================================================================== --- llvm/lib/Transforms/Vectorize/LoopVectorize.cpp +++ llvm/lib/Transforms/Vectorize/LoopVectorize.cpp @@ -848,7 +848,7 @@ /// Middle Block between the vector and the scalar. BasicBlock *LoopMiddleBlock; - /// The (unique) ExitBlock of the scalar loop. Note that + /// The unique ExitBlock of the scalar loop if one exists. Note that /// there can be multiple exiting edges reaching this block. BasicBlock *LoopExitBlock; @@ -3103,9 +3103,13 @@ DT->getNode(Bypass)->getIDom()) && "TC check is expected to dominate Bypass"); - // Update dominator for Bypass & LoopExit. + // Update dominator for Bypass & LoopExit (if needed). DT->changeImmediateDominator(Bypass, TCCheckBlock); - DT->changeImmediateDominator(LoopExitBlock, TCCheckBlock); + if (!Cost->requiresScalarEpilogue()) + // For loops with multiple exits, there's no edge from the middle block + // to exit blocks (as the epilogue must run) and thus no need to update + // the immediate dominator of the exit blocks. + DT->changeImmediateDominator(LoopExitBlock, TCCheckBlock); ReplaceInstWithInst( TCCheckBlock->getTerminator(), @@ -3144,7 +3148,11 @@ // Update dominator only if this is first RT check. if (LoopBypassBlocks.empty()) { DT->changeImmediateDominator(Bypass, SCEVCheckBlock); - DT->changeImmediateDominator(LoopExitBlock, SCEVCheckBlock); + if (!Cost->requiresScalarEpilogue()) + // For loops with multiple exits, there's no edge from the middle block + // to exit blocks (as the epilogue must run) and thus no need to update + // the immediate dominator of the exit blocks. + DT->changeImmediateDominator(LoopExitBlock, SCEVCheckBlock); } ReplaceInstWithInst( @@ -3200,7 +3208,11 @@ // Update dominator only if this is first RT check. if (LoopBypassBlocks.empty()) { DT->changeImmediateDominator(Bypass, MemCheckBlock); - DT->changeImmediateDominator(LoopExitBlock, MemCheckBlock); + if (!Cost->requiresScalarEpilogue()) + // For loops with multiple exits, there's no edge from the middle block + // to exit blocks (as the epilogue must run) and thus no need to update + // the immediate dominator of the exit blocks. + DT->changeImmediateDominator(LoopExitBlock, MemCheckBlock); } Instruction *FirstCheckInst; @@ -3324,9 +3336,10 @@ Loop *InnerLoopVectorizer::createVectorLoopSkeleton(StringRef Prefix) { LoopScalarBody = OrigLoop->getHeader(); LoopVectorPreHeader = OrigLoop->getLoopPreheader(); - LoopExitBlock = OrigLoop->getUniqueExitBlock(); - assert(LoopExitBlock && "Must have an exit block"); assert(LoopVectorPreHeader && "Invalid loop structure"); + LoopExitBlock = OrigLoop->getUniqueExitBlock(); // may be nullptr + assert((LoopExitBlock || Cost->requiresScalarEpilogue()) && + "multiple exit loop without required epilogue?"); LoopMiddleBlock = SplitBlock(LoopVectorPreHeader, LoopVectorPreHeader->getTerminator(), DT, @@ -3335,12 +3348,24 @@ SplitBlock(LoopMiddleBlock, LoopMiddleBlock->getTerminator(), DT, LI, nullptr, Twine(Prefix) + "scalar.ph"); - // Set up branch from middle block to the exit and scalar preheader blocks. - // completeLoopSkeleton will update the condition to use an iteration check, - // if required to decide whether to execute the remainder. - BranchInst *BrInst = - BranchInst::Create(LoopExitBlock, LoopScalarPreHeader, Builder.getTrue()); auto *ScalarLatchTerm = OrigLoop->getLoopLatch()->getTerminator(); + + // Set up the middle block terminator. Two cases: + // 1) If we know that we must execute the scalar epilogue, emit an + // unconditional branch. + // 2) Otherwise, we must have a single unique exit block (due to how we + // implement the multiple exit case). In this case, set up conditonal + // branch from middle block to loop scalar preheader, and the exit block. + // completeLoopSkeleton will update the condition to use an iteration + // check, if required to decide whether to execute the remainder. + BranchInst *BrInst = [&]() { + if (Cost->requiresScalarEpilogue()) { + return BranchInst::Create(LoopScalarPreHeader); + } + assert(LoopExitBlock && "Must have an exit block"); + return BranchInst::Create(LoopExitBlock, LoopScalarPreHeader, + Builder.getTrue()); + }(); BrInst->setDebugLoc(ScalarLatchTerm->getDebugLoc()); ReplaceInstWithInst(LoopMiddleBlock->getTerminator(), BrInst); @@ -3352,7 +3377,11 @@ nullptr, nullptr, Twine(Prefix) + "vector.body"); // Update dominator for loop exit. - DT->changeImmediateDominator(LoopExitBlock, LoopMiddleBlock); + if (!Cost->requiresScalarEpilogue()) + // For loops with multiple exits, there's no edge from the middle block + // to exit blocks (as the epilogue must run) and thus no need to update + // the immediate dominator of the exit blocks. + DT->changeImmediateDominator(LoopExitBlock, LoopMiddleBlock); // Create and register the new vector loop. Loop *Lp = LI->AllocateLoop(); @@ -3449,10 +3478,14 @@ auto *ScalarLatchTerm = OrigLoop->getLoopLatch()->getTerminator(); // Add a check in the middle block to see if we have completed - // all of the iterations in the first vector loop. - // If (N - N%VF) == N, then we *don't* need to run the remainder. - // If tail is to be folded, we know we don't need to run the remainder. - if (!Cost->foldTailByMasking()) { + // all of the iterations in the first vector loop. Three cases: + // 1) If we require scalar epilogue, there is no conditional branch as + // we unconditionally branch to scalar preheader. Do nothing. + // 2) If (N - N%VF) == N, then we *don't* need to run the remainder. + // Thus if tail is to be folded, we know we don't need to run the + // remainder and we can use the previous value for the condition (true). + // 3) Otherwise, construct a runtime check. + if (!Cost->requiresScalarEpilogue() && !Cost->foldTailByMasking()) { Instruction *CmpN = CmpInst::Create(Instruction::ICmp, CmpInst::ICMP_EQ, Count, VectorTripCount, "cmp.n", LoopMiddleBlock->getTerminator()); @@ -3516,17 +3549,17 @@ | [ ]_| <-- vector loop. | | | v - | -[ ] <--- middle-block. - | / | - | / v - -|- >[ ] <--- new preheader. + \ -[ ] <--- middle-block. + \/ | + /\ v + | ->[ ] <--- new preheader. | | - | v + (opt) v <-- edge from middle to exit iff epilogue is not required. | [ ] \ - | [ ]_| <-- old scalar loop to handle remainder. + | [ ]_| <-- old scalar loop to handle remainder (scalar epilogue). \ | \ v - >[ ] <-- exit block. + >[ ] <-- exit block(s). ... */ @@ -3590,6 +3623,7 @@ const InductionDescriptor &II, Value *CountRoundDown, Value *EndValue, BasicBlock *MiddleBlock) { + // There are two kinds of external IV usages - those that use the value // computed in the last iteration (the PHI) and those that use the penultimate // value (the value that feeds into the phi from the loop latch). @@ -3928,13 +3962,18 @@ // Forget the original basic block. PSE.getSE()->forgetLoop(OrigLoop); - // Fix-up external users of the induction variables. - for (auto &Entry : Legal->getInductionVars()) - fixupIVUsers(Entry.first, Entry.second, - getOrCreateVectorTripCount(LI->getLoopFor(LoopVectorBody)), - IVEndValues[Entry.first], LoopMiddleBlock); + // If we inserted an edge from the middle block to the unique exit block, + // update uses outside the loop (phis) to account for the newly inserted + // edge. + if (!Cost->requiresScalarEpilogue()) { + // Fix-up external users of the induction variables. + for (auto &Entry : Legal->getInductionVars()) + fixupIVUsers(Entry.first, Entry.second, + getOrCreateVectorTripCount(LI->getLoopFor(LoopVectorBody)), + IVEndValues[Entry.first], LoopMiddleBlock); - fixLCSSAPHIs(); + fixLCSSAPHIs(); + } for (Instruction *PI : PredicatedInstructions) sinkScalarOperands(&*PI); @@ -4152,12 +4191,13 @@ // recurrence in the exit block, and then add an edge for the middle block. // Note that LCSSA does not imply single entry when the original scalar loop // had multiple exiting edges (as we always run the last iteration in the - // scalar epilogue); in that case, the exiting path through middle will be - // dynamically dead and the value picked for the phi doesn't matter. - for (PHINode &LCSSAPhi : LoopExitBlock->phis()) - if (any_of(LCSSAPhi.incoming_values(), - [Phi](Value *V) { return V == Phi; })) - LCSSAPhi.addIncoming(ExtractForPhiUsedOutsideLoop, LoopMiddleBlock); + // scalar epilogue); in that case, there is no edge from middle to exit and + // and thus no phis which needed updated. + if (!Cost->requiresScalarEpilogue()) + for (PHINode &LCSSAPhi : LoopExitBlock->phis()) + if (any_of(LCSSAPhi.incoming_values(), + [Phi](Value *V) { return V == Phi; })) + LCSSAPhi.addIncoming(ExtractForPhiUsedOutsideLoop, LoopMiddleBlock); } void InnerLoopVectorizer::fixReduction(PHINode *Phi) { @@ -4319,10 +4359,11 @@ // We know that the loop is in LCSSA form. We need to update the PHI nodes // in the exit blocks. See comment on analogous loop in // fixFirstOrderRecurrence for a more complete explaination of the logic. - for (PHINode &LCSSAPhi : LoopExitBlock->phis()) - if (any_of(LCSSAPhi.incoming_values(), - [LoopExitInst](Value *V) { return V == LoopExitInst; })) - LCSSAPhi.addIncoming(ReducedPartRdx, LoopMiddleBlock); + if (!Cost->requiresScalarEpilogue()) + for (PHINode &LCSSAPhi : LoopExitBlock->phis()) + if (any_of(LCSSAPhi.incoming_values(), + [LoopExitInst](Value *V) { return V == LoopExitInst; })) + LCSSAPhi.addIncoming(ReducedPartRdx, LoopMiddleBlock); // Fix the scalar loop reduction variable with the incoming reduction sum // from the vector body and from the backedge value. @@ -4377,9 +4418,9 @@ unsigned LastLane = 0; if (isa(IncomingValue)) LastLane = Cost->isUniformAfterVectorization( - cast(IncomingValue), VF) - ? 0 - : VF.getKnownMinValue() - 1; + cast(IncomingValue), VF) + ? 0 + : VF.getKnownMinValue() - 1; assert((!VF.isScalable() || LastLane == 0) && "scalable vectors dont support non-uniform scalars yet"); // Can be a loop invariant incoming value or the last scalar value to be @@ -7810,7 +7851,11 @@ // Update dominator for Bypass & LoopExit. DT->changeImmediateDominator(Bypass, TCCheckBlock); - DT->changeImmediateDominator(LoopExitBlock, TCCheckBlock); + if (!Cost->requiresScalarEpilogue()) + // For loops with multiple exits, there's no edge from the middle block + // to exit blocks (as the epilogue must run) and thus no need to update + // the immediate dominator of the exit blocks. + DT->changeImmediateDominator(LoopExitBlock, TCCheckBlock); LoopBypassBlocks.push_back(TCCheckBlock); @@ -7874,7 +7919,12 @@ DT->changeImmediateDominator(LoopScalarPreHeader, EPI.EpilogueIterationCountCheck); - DT->changeImmediateDominator(LoopExitBlock, EPI.EpilogueIterationCountCheck); + if (!Cost->requiresScalarEpilogue()) + // For loops with multiple exits, there's no edge from the middle block + // to exit blocks (as the epilogue must run) and thus no need to update + // the immediate dominator of the exit blocks. + DT->changeImmediateDominator(LoopExitBlock, + EPI.EpilogueIterationCountCheck); // Keep track of bypass blocks, as they feed start values to the induction // phis in the scalar loop preheader. Index: llvm/test/Transforms/LoopVectorize/first-order-recurrence-complex.ll =================================================================== --- llvm/test/Transforms/LoopVectorize/first-order-recurrence-complex.ll +++ llvm/test/Transforms/LoopVectorize/first-order-recurrence-complex.ll @@ -471,10 +471,9 @@ ; CHECK-NEXT: [[TMP15:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]] ; CHECK-NEXT: br i1 [[TMP15]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], [[LOOP6:!llvm.loop !.*]] ; CHECK: middle.block: -; CHECK-NEXT: [[CMP_N:%.*]] = icmp eq i32 [[TMP2]], [[N_VEC]] ; CHECK-NEXT: [[VECTOR_RECUR_EXTRACT:%.*]] = extractelement <4 x i16> [[WIDE_LOAD]], i32 3 ; CHECK-NEXT: [[VECTOR_RECUR_EXTRACT_FOR_PHI:%.*]] = extractelement <4 x i16> [[WIDE_LOAD]], i32 2 -; CHECK-NEXT: br i1 [[CMP_N]], label [[IF_END:%.*]], label [[SCALAR_PH]] +; CHECK-NEXT: br label [[SCALAR_PH]] ; CHECK: scalar.ph: ; CHECK-NEXT: [[SCALAR_RECUR_INIT:%.*]] = phi i16 [ 0, [[ENTRY:%.*]] ], [ [[VECTOR_RECUR_EXTRACT]], [[MIDDLE_BLOCK]] ] ; CHECK-NEXT: [[BC_RESUME_VAL:%.*]] = phi i32 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY]] ] @@ -486,14 +485,14 @@ ; CHECK-NEXT: [[B:%.*]] = getelementptr inbounds i16, i16* [[P]], i64 [[IPROM]] ; CHECK-NEXT: [[REC_NEXT]] = load i16, i16* [[B]], align 2 ; CHECK-NEXT: [[CMP:%.*]] = icmp slt i32 [[I]], [[N]] -; CHECK-NEXT: br i1 [[CMP]], label [[FOR_BODY]], label [[IF_END]] +; CHECK-NEXT: br i1 [[CMP]], label [[FOR_BODY]], label [[IF_END:%.*]] ; CHECK: for.body: ; CHECK-NEXT: store i16 [[SCALAR_RECUR]], i16* [[B]], align 4 ; CHECK-NEXT: [[INC]] = add nsw i32 [[I]], 1 ; CHECK-NEXT: [[CMP2:%.*]] = icmp slt i32 [[I]], 2096 ; CHECK-NEXT: br i1 [[CMP2]], label [[FOR_COND]], label [[IF_END]], [[LOOP7:!llvm.loop !.*]] ; CHECK: if.end: -; CHECK-NEXT: [[REC_LCSSA:%.*]] = phi i16 [ [[SCALAR_RECUR]], [[FOR_BODY]] ], [ [[SCALAR_RECUR]], [[FOR_COND]] ], [ [[VECTOR_RECUR_EXTRACT_FOR_PHI]], [[MIDDLE_BLOCK]] ] +; CHECK-NEXT: [[REC_LCSSA:%.*]] = phi i16 [ [[SCALAR_RECUR]], [[FOR_BODY]] ], [ [[SCALAR_RECUR]], [[FOR_COND]] ] ; CHECK-NEXT: ret i16 [[REC_LCSSA]] ; entry: @@ -558,10 +557,9 @@ ; CHECK-NEXT: [[TMP15:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]] ; CHECK-NEXT: br i1 [[TMP15]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], [[LOOP8:!llvm.loop !.*]] ; CHECK: middle.block: -; CHECK-NEXT: [[CMP_N:%.*]] = icmp eq i32 [[TMP2]], [[N_VEC]] ; CHECK-NEXT: [[VECTOR_RECUR_EXTRACT:%.*]] = extractelement <4 x i16> [[WIDE_LOAD]], i32 3 ; CHECK-NEXT: [[VECTOR_RECUR_EXTRACT_FOR_PHI:%.*]] = extractelement <4 x i16> [[WIDE_LOAD]], i32 2 -; CHECK-NEXT: br i1 [[CMP_N]], label [[IF_END:%.*]], label [[SCALAR_PH]] +; CHECK-NEXT: br label [[SCALAR_PH]] ; CHECK: scalar.ph: ; CHECK-NEXT: [[SCALAR_RECUR_INIT:%.*]] = phi i16 [ 0, [[ENTRY:%.*]] ], [ [[VECTOR_RECUR_EXTRACT]], [[MIDDLE_BLOCK]] ] ; CHECK-NEXT: [[BC_RESUME_VAL:%.*]] = phi i32 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY]] ] @@ -573,14 +571,14 @@ ; CHECK-NEXT: [[B:%.*]] = getelementptr inbounds i16, i16* [[P]], i64 [[IPROM]] ; CHECK-NEXT: [[REC_NEXT]] = load i16, i16* [[B]], align 2 ; CHECK-NEXT: [[CMP:%.*]] = icmp slt i32 [[I]], [[N]] -; CHECK-NEXT: br i1 [[CMP]], label [[FOR_BODY]], label [[IF_END]] +; CHECK-NEXT: br i1 [[CMP]], label [[FOR_BODY]], label [[IF_END:%.*]] ; CHECK: for.body: ; CHECK-NEXT: store i16 [[SCALAR_RECUR]], i16* [[B]], align 4 ; CHECK-NEXT: [[INC]] = add nsw i32 [[I]], 1 ; CHECK-NEXT: [[CMP2:%.*]] = icmp slt i32 [[I]], 2096 ; CHECK-NEXT: br i1 [[CMP2]], label [[FOR_COND]], label [[IF_END]], [[LOOP9:!llvm.loop !.*]] ; CHECK: if.end: -; CHECK-NEXT: [[REC_LCSSA:%.*]] = phi i16 [ [[SCALAR_RECUR]], [[FOR_COND]] ], [ 10, [[FOR_BODY]] ], [ [[VECTOR_RECUR_EXTRACT_FOR_PHI]], [[MIDDLE_BLOCK]] ] +; CHECK-NEXT: [[REC_LCSSA:%.*]] = phi i16 [ [[SCALAR_RECUR]], [[FOR_COND]] ], [ 10, [[FOR_BODY]] ] ; CHECK-NEXT: ret i16 [[REC_LCSSA]] ; entry: Index: llvm/test/Transforms/LoopVectorize/interleaved-accesses.ll =================================================================== --- llvm/test/Transforms/LoopVectorize/interleaved-accesses.ll +++ llvm/test/Transforms/LoopVectorize/interleaved-accesses.ll @@ -447,7 +447,7 @@ ; CHECK-NEXT: [[TMP6:%.*]] = icmp eq i64 [[INDEX_NEXT]], 508 ; CHECK-NEXT: br i1 [[TMP6]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], [[LOOP12:!llvm.loop !.*]] ; CHECK: middle.block: -; CHECK-NEXT: br i1 false, label [[FOR_COND_CLEANUP:%.*]], label [[SCALAR_PH]] +; CHECK-NEXT: br label [[SCALAR_PH]] ; CHECK: scalar.ph: ; CHECK-NEXT: [[BC_RESUME_VAL:%.*]] = phi i64 [ 1016, [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ] ; CHECK-NEXT: br label [[FOR_BODY:%.*]] @@ -463,7 +463,7 @@ ; CHECK-NEXT: store i32 [[MUL]], i32* [[ARRAYIDX2]], align 4 ; CHECK-NEXT: [[INDVARS_IV_NEXT]] = add nuw nsw i64 [[INDVARS_IV]], 2 ; CHECK-NEXT: [[CMP:%.*]] = icmp ult i64 [[INDVARS_IV]], 1022 -; CHECK-NEXT: br i1 [[CMP]], label [[FOR_BODY]], label [[FOR_COND_CLEANUP]], [[LOOP13:!llvm.loop !.*]] +; CHECK-NEXT: br i1 [[CMP]], label [[FOR_BODY]], label [[FOR_COND_CLEANUP:%.*]], [[LOOP13:!llvm.loop !.*]] ; entry: br label %for.body @@ -528,7 +528,7 @@ ; CHECK-NEXT: [[TMP12:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]] ; CHECK-NEXT: br i1 [[TMP12]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], [[LOOP14:!llvm.loop !.*]] ; CHECK: middle.block: -; CHECK-NEXT: br i1 false, label [[FOR_COND_CLEANUP:%.*]], label [[SCALAR_PH]] +; CHECK-NEXT: br label [[SCALAR_PH]] ; CHECK: scalar.ph: ; CHECK-NEXT: [[BC_RESUME_VAL:%.*]] = phi i64 [ [[IND_END]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ] ; CHECK-NEXT: br label [[FOR_BODY:%.*]] @@ -544,7 +544,7 @@ ; CHECK-NEXT: store i32 [[MUL]], i32* [[ARRAYIDX2]], align 4 ; CHECK-NEXT: [[INDVARS_IV_NEXT]] = add nuw nsw i64 [[INDVARS_IV]], 2 ; CHECK-NEXT: [[CMP:%.*]] = icmp ult i64 [[INDVARS_IV_NEXT]], [[N]] -; CHECK-NEXT: br i1 [[CMP]], label [[FOR_BODY]], label [[FOR_COND_CLEANUP]], [[LOOP15:!llvm.loop !.*]] +; CHECK-NEXT: br i1 [[CMP]], label [[FOR_BODY]], label [[FOR_COND_CLEANUP:%.*]], [[LOOP15:!llvm.loop !.*]] ; entry: br label %for.body @@ -973,7 +973,7 @@ ; CHECK-NEXT: [[TMP19:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]] ; CHECK-NEXT: br i1 [[TMP19]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], [[LOOP24:!llvm.loop !.*]] ; CHECK: middle.block: -; CHECK-NEXT: br i1 false, label [[FOR_END:%.*]], label [[SCALAR_PH]] +; CHECK-NEXT: br label [[SCALAR_PH]] ; CHECK: scalar.ph: ; CHECK-NEXT: [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ] ; CHECK-NEXT: br label [[FOR_BODY:%.*]] @@ -985,7 +985,7 @@ ; CHECK-NEXT: store i32 [[Z]], i32* [[P_I_Y]], align 4 ; CHECK-NEXT: [[I_NEXT]] = add nuw nsw i64 [[I]], 1 ; CHECK-NEXT: [[COND:%.*]] = icmp slt i64 [[I_NEXT]], [[N]] -; CHECK-NEXT: br i1 [[COND]], label [[FOR_BODY]], label [[FOR_END]], [[LOOP25:!llvm.loop !.*]] +; CHECK-NEXT: br i1 [[COND]], label [[FOR_BODY]], label [[FOR_END:%.*]], [[LOOP25:!llvm.loop !.*]] ; CHECK: for.end: ; CHECK-NEXT: ret void ; @@ -1066,7 +1066,7 @@ ; CHECK-NEXT: [[RDX_SHUF3:%.*]] = shufflevector <4 x i32> [[BIN_RDX]], <4 x i32> poison, <4 x i32> ; CHECK-NEXT: [[BIN_RDX4:%.*]] = add <4 x i32> [[BIN_RDX]], [[RDX_SHUF3]] ; CHECK-NEXT: [[TMP19:%.*]] = extractelement <4 x i32> [[BIN_RDX4]], i32 0 -; CHECK-NEXT: br i1 false, label [[FOR_END:%.*]], label [[SCALAR_PH]] +; CHECK-NEXT: br label [[SCALAR_PH]] ; CHECK: scalar.ph: ; CHECK-NEXT: [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ] ; CHECK-NEXT: [[BC_MERGE_RDX:%.*]] = phi i32 [ [[TMP19]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY]] ] @@ -1081,10 +1081,9 @@ ; CHECK-NEXT: [[TMP21]] = add nsw i32 [[TMP20]], [[S]] ; CHECK-NEXT: [[I_NEXT]] = add nuw nsw i64 [[I]], 1 ; CHECK-NEXT: [[COND:%.*]] = icmp slt i64 [[I_NEXT]], [[N]] -; CHECK-NEXT: br i1 [[COND]], label [[FOR_BODY]], label [[FOR_END]], [[LOOP27:!llvm.loop !.*]] +; CHECK-NEXT: br i1 [[COND]], label [[FOR_BODY]], label [[FOR_END:%.*]], [[LOOP27:!llvm.loop !.*]] ; CHECK: for.end: -; CHECK-NEXT: [[TMP22:%.*]] = phi i32 [ [[TMP21]], [[FOR_BODY]] ], [ [[TMP19]], [[MIDDLE_BLOCK]] ] -; CHECK-NEXT: ret i32 [[TMP22]] +; CHECK-NEXT: ret i32 [[TMP21]] ; entry: br label %for.body @@ -1163,7 +1162,7 @@ ; CHECK-NEXT: [[TMP20:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]] ; CHECK-NEXT: br i1 [[TMP20]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], [[LOOP28:!llvm.loop !.*]] ; CHECK: middle.block: -; CHECK-NEXT: br i1 false, label [[FOR_END:%.*]], label [[SCALAR_PH]] +; CHECK-NEXT: br label [[SCALAR_PH]] ; CHECK: scalar.ph: ; CHECK-NEXT: [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ] ; CHECK-NEXT: br label [[FOR_BODY:%.*]] @@ -1177,7 +1176,7 @@ ; CHECK-NEXT: store i32 [[TMP21]], i32* [[P_I_Y]], align 4 ; CHECK-NEXT: [[I_NEXT]] = add nuw nsw i64 [[I]], 1 ; CHECK-NEXT: [[COND:%.*]] = icmp slt i64 [[I_NEXT]], [[N]] -; CHECK-NEXT: br i1 [[COND]], label [[FOR_BODY]], label [[FOR_END]], [[LOOP29:!llvm.loop !.*]] +; CHECK-NEXT: br i1 [[COND]], label [[FOR_BODY]], label [[FOR_END:%.*]], [[LOOP29:!llvm.loop !.*]] ; CHECK: for.end: ; CHECK-NEXT: ret void ; @@ -1264,7 +1263,7 @@ ; CHECK-NEXT: [[RDX_SHUF3:%.*]] = shufflevector <4 x i32> [[BIN_RDX]], <4 x i32> poison, <4 x i32> ; CHECK-NEXT: [[BIN_RDX4:%.*]] = add <4 x i32> [[BIN_RDX]], [[RDX_SHUF3]] ; CHECK-NEXT: [[TMP22:%.*]] = extractelement <4 x i32> [[BIN_RDX4]], i32 0 -; CHECK-NEXT: br i1 false, label [[FOR_END:%.*]], label [[SCALAR_PH]] +; CHECK-NEXT: br label [[SCALAR_PH]] ; CHECK: scalar.ph: ; CHECK-NEXT: [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ] ; CHECK-NEXT: [[BC_MERGE_RDX:%.*]] = phi i32 [ [[TMP22]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY]] ] @@ -1282,10 +1281,9 @@ ; CHECK-NEXT: [[TMP25]] = add nsw i32 [[TMP24]], [[S]] ; CHECK-NEXT: [[I_NEXT]] = add nuw nsw i64 [[I]], 1 ; CHECK-NEXT: [[COND:%.*]] = icmp slt i64 [[I_NEXT]], [[N]] -; CHECK-NEXT: br i1 [[COND]], label [[FOR_BODY]], label [[FOR_END]], [[LOOP31:!llvm.loop !.*]] +; CHECK-NEXT: br i1 [[COND]], label [[FOR_BODY]], label [[FOR_END:%.*]], [[LOOP31:!llvm.loop !.*]] ; CHECK: for.end: -; CHECK-NEXT: [[TMP26:%.*]] = phi i32 [ [[TMP25]], [[FOR_BODY]] ], [ [[TMP22]], [[MIDDLE_BLOCK]] ] -; CHECK-NEXT: ret i32 [[TMP26]] +; CHECK-NEXT: ret i32 [[TMP25]] ; entry: br label %for.body Index: llvm/test/Transforms/LoopVectorize/loop-form.ll =================================================================== --- llvm/test/Transforms/LoopVectorize/loop-form.ll +++ llvm/test/Transforms/LoopVectorize/loop-form.ll @@ -146,15 +146,14 @@ ; CHECK-NEXT: [[TMP10:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]] ; CHECK-NEXT: br i1 [[TMP10]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], [[LOOP4:!llvm.loop !.*]] ; CHECK: middle.block: -; CHECK-NEXT: [[CMP_N:%.*]] = icmp eq i32 [[TMP1]], [[N_VEC]] -; CHECK-NEXT: br i1 [[CMP_N]], label [[IF_END:%.*]], label [[SCALAR_PH]] +; CHECK-NEXT: br label [[SCALAR_PH]] ; CHECK: scalar.ph: ; CHECK-NEXT: [[BC_RESUME_VAL:%.*]] = phi i32 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ] ; CHECK-NEXT: br label [[FOR_COND:%.*]] ; CHECK: for.cond: ; CHECK-NEXT: [[I:%.*]] = phi i32 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[INC:%.*]], [[FOR_BODY:%.*]] ] ; CHECK-NEXT: [[CMP:%.*]] = icmp slt i32 [[I]], [[N]] -; CHECK-NEXT: br i1 [[CMP]], label [[FOR_BODY]], label [[IF_END]] +; CHECK-NEXT: br i1 [[CMP]], label [[FOR_BODY]], label [[IF_END:%.*]] ; CHECK: for.body: ; CHECK-NEXT: [[IPROM:%.*]] = sext i32 [[I]] to i64 ; CHECK-NEXT: [[B:%.*]] = getelementptr inbounds i16, i16* [[P]], i64 [[IPROM]] @@ -286,15 +285,14 @@ ; CHECK-NEXT: [[TMP11:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]] ; CHECK-NEXT: br i1 [[TMP11]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], [[LOOP6:!llvm.loop !.*]] ; CHECK: middle.block: -; CHECK-NEXT: [[CMP_N:%.*]] = icmp eq i32 [[TMP2]], [[N_VEC]] -; CHECK-NEXT: br i1 [[CMP_N]], label [[IF_END:%.*]], label [[SCALAR_PH]] +; CHECK-NEXT: br label [[SCALAR_PH]] ; CHECK: scalar.ph: ; CHECK-NEXT: [[BC_RESUME_VAL:%.*]] = phi i32 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ] ; CHECK-NEXT: br label [[FOR_COND:%.*]] ; CHECK: for.cond: ; CHECK-NEXT: [[I:%.*]] = phi i32 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[INC:%.*]], [[FOR_BODY:%.*]] ] ; CHECK-NEXT: [[CMP:%.*]] = icmp slt i32 [[I]], [[N]] -; CHECK-NEXT: br i1 [[CMP]], label [[FOR_BODY]], label [[IF_END]] +; CHECK-NEXT: br i1 [[CMP]], label [[FOR_BODY]], label [[IF_END:%.*]] ; CHECK: for.body: ; CHECK-NEXT: [[IPROM:%.*]] = sext i32 [[I]] to i64 ; CHECK-NEXT: [[B:%.*]] = getelementptr inbounds i16, i16* [[P]], i64 [[IPROM]] @@ -374,17 +372,14 @@ ; CHECK-NEXT: [[TMP11:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]] ; CHECK-NEXT: br i1 [[TMP11]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], [[LOOP8:!llvm.loop !.*]] ; CHECK: middle.block: -; CHECK-NEXT: [[CMP_N:%.*]] = icmp eq i32 [[TMP2]], [[N_VEC]] -; CHECK-NEXT: [[IND_ESCAPE:%.*]] = sub i32 [[N_VEC]], 1 -; CHECK-NEXT: [[IND_ESCAPE1:%.*]] = sub i32 [[N_VEC]], 1 -; CHECK-NEXT: br i1 [[CMP_N]], label [[IF_END:%.*]], label [[SCALAR_PH]] +; CHECK-NEXT: br label [[SCALAR_PH]] ; CHECK: scalar.ph: ; CHECK-NEXT: [[BC_RESUME_VAL:%.*]] = phi i32 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ] ; CHECK-NEXT: br label [[FOR_COND:%.*]] ; CHECK: for.cond: ; CHECK-NEXT: [[I:%.*]] = phi i32 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[INC:%.*]], [[FOR_BODY:%.*]] ] ; CHECK-NEXT: [[CMP:%.*]] = icmp slt i32 [[I]], [[N]] -; CHECK-NEXT: br i1 [[CMP]], label [[FOR_BODY]], label [[IF_END]] +; CHECK-NEXT: br i1 [[CMP]], label [[FOR_BODY]], label [[IF_END:%.*]] ; CHECK: for.body: ; CHECK-NEXT: [[IPROM:%.*]] = sext i32 [[I]] to i64 ; CHECK-NEXT: [[B:%.*]] = getelementptr inbounds i16, i16* [[P]], i64 [[IPROM]] @@ -393,7 +388,7 @@ ; CHECK-NEXT: [[CMP2:%.*]] = icmp slt i32 [[I]], 2096 ; CHECK-NEXT: br i1 [[CMP2]], label [[FOR_COND]], label [[IF_END]], [[LOOP9:!llvm.loop !.*]] ; CHECK: if.end: -; CHECK-NEXT: [[I_LCSSA:%.*]] = phi i32 [ [[I]], [[FOR_BODY]] ], [ [[I]], [[FOR_COND]] ], [ [[IND_ESCAPE1]], [[MIDDLE_BLOCK]] ] +; CHECK-NEXT: [[I_LCSSA:%.*]] = phi i32 [ [[I]], [[FOR_BODY]] ], [ [[I]], [[FOR_COND]] ] ; CHECK-NEXT: ret i32 [[I_LCSSA]] ; ; TAILFOLD-LABEL: @multiple_unique_exit2( @@ -466,15 +461,14 @@ ; CHECK-NEXT: [[TMP11:%.*]] = icmp eq i32 [[INDEX_NEXT]], [[N_VEC]] ; CHECK-NEXT: br i1 [[TMP11]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], [[LOOP10:!llvm.loop !.*]] ; CHECK: middle.block: -; CHECK-NEXT: [[CMP_N:%.*]] = icmp eq i32 [[TMP2]], [[N_VEC]] -; CHECK-NEXT: br i1 [[CMP_N]], label [[IF_END:%.*]], label [[SCALAR_PH]] +; CHECK-NEXT: br label [[SCALAR_PH]] ; CHECK: scalar.ph: ; CHECK-NEXT: [[BC_RESUME_VAL:%.*]] = phi i32 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ] ; CHECK-NEXT: br label [[FOR_COND:%.*]] ; CHECK: for.cond: ; CHECK-NEXT: [[I:%.*]] = phi i32 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[INC:%.*]], [[FOR_BODY:%.*]] ] ; CHECK-NEXT: [[CMP:%.*]] = icmp slt i32 [[I]], [[N]] -; CHECK-NEXT: br i1 [[CMP]], label [[FOR_BODY]], label [[IF_END]] +; CHECK-NEXT: br i1 [[CMP]], label [[FOR_BODY]], label [[IF_END:%.*]] ; CHECK: for.body: ; CHECK-NEXT: [[IPROM:%.*]] = sext i32 [[I]] to i64 ; CHECK-NEXT: [[B:%.*]] = getelementptr inbounds i16, i16* [[P]], i64 [[IPROM]] @@ -483,7 +477,7 @@ ; CHECK-NEXT: [[CMP2:%.*]] = icmp slt i32 [[I]], 2096 ; CHECK-NEXT: br i1 [[CMP2]], label [[FOR_COND]], label [[IF_END]], [[LOOP11:!llvm.loop !.*]] ; CHECK: if.end: -; CHECK-NEXT: [[EXIT:%.*]] = phi i32 [ 0, [[FOR_COND]] ], [ 1, [[FOR_BODY]] ], [ 0, [[MIDDLE_BLOCK]] ] +; CHECK-NEXT: [[EXIT:%.*]] = phi i32 [ 0, [[FOR_COND]] ], [ 1, [[FOR_BODY]] ] ; CHECK-NEXT: ret i32 [[EXIT]] ; ; TAILFOLD-LABEL: @multiple_unique_exit3( @@ -1000,8 +994,7 @@ ; CHECK-NEXT: [[TMP10:%.*]] = icmp eq i64 [[INDEX_NEXT]], 200 ; CHECK-NEXT: br i1 [[TMP10]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], [[LOOP12:!llvm.loop !.*]] ; CHECK: middle.block: -; CHECK-NEXT: [[CMP_N:%.*]] = icmp eq i64 201, 200 -; CHECK-NEXT: br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]] +; CHECK-NEXT: br label [[SCALAR_PH]] ; CHECK: scalar.ph: ; CHECK-NEXT: [[BC_RESUME_VAL:%.*]] = phi i64 [ 200, [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ] ; CHECK-NEXT: br label [[LOOP_HEADER:%.*]] @@ -1009,7 +1002,7 @@ ; CHECK-NEXT: [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], [[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], [[LOOP_LATCH:%.*]] ] ; CHECK-NEXT: [[GEP:%.*]] = getelementptr float, float* [[ADDR]], i64 [[IV]] ; CHECK-NEXT: [[EXITCOND_NOT:%.*]] = icmp eq i64 [[IV]], 200 -; CHECK-NEXT: br i1 [[EXITCOND_NOT]], label [[EXIT]], label [[LOOP_BODY:%.*]] +; CHECK-NEXT: br i1 [[EXITCOND_NOT]], label [[EXIT:%.*]], label [[LOOP_BODY:%.*]] ; CHECK: loop.body: ; CHECK-NEXT: [[TMP11:%.*]] = load float, float* [[GEP]], align 4 ; CHECK-NEXT: [[PRED:%.*]] = fcmp oeq float [[TMP11]], 0.000000e+00 @@ -1095,8 +1088,7 @@ ; CHECK-NEXT: [[RDX_SHUF:%.*]] = shufflevector <2 x i32> [[TMP5]], <2 x i32> poison, <2 x i32> ; CHECK-NEXT: [[BIN_RDX:%.*]] = add <2 x i32> [[TMP5]], [[RDX_SHUF]] ; CHECK-NEXT: [[TMP7:%.*]] = extractelement <2 x i32> [[BIN_RDX]], i32 0 -; CHECK-NEXT: [[CMP_N:%.*]] = icmp eq i64 201, 200 -; CHECK-NEXT: br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]] +; CHECK-NEXT: br label [[SCALAR_PH]] ; CHECK: scalar.ph: ; CHECK-NEXT: [[BC_RESUME_VAL:%.*]] = phi i64 [ 200, [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ] ; CHECK-NEXT: [[BC_MERGE_RDX:%.*]] = phi i32 [ 0, [[ENTRY]] ], [ [[TMP7]], [[MIDDLE_BLOCK]] ] @@ -1106,7 +1098,7 @@ ; CHECK-NEXT: [[ACCUM:%.*]] = phi i32 [ [[BC_MERGE_RDX]], [[SCALAR_PH]] ], [ [[ACCUM_NEXT:%.*]], [[LOOP_LATCH]] ] ; CHECK-NEXT: [[GEP:%.*]] = getelementptr i32, i32* [[ADDR]], i64 [[IV]] ; CHECK-NEXT: [[EXITCOND_NOT:%.*]] = icmp eq i64 [[IV]], 200 -; CHECK-NEXT: br i1 [[EXITCOND_NOT]], label [[EXIT]], label [[LOOP_LATCH]] +; CHECK-NEXT: br i1 [[EXITCOND_NOT]], label [[EXIT:%.*]], label [[LOOP_LATCH]] ; CHECK: loop.latch: ; CHECK-NEXT: [[TMP8:%.*]] = load i32, i32* [[GEP]], align 4 ; CHECK-NEXT: [[ACCUM_NEXT]] = add i32 [[ACCUM]], [[TMP8]] @@ -1114,7 +1106,7 @@ ; CHECK-NEXT: [[EXITCOND2_NOT:%.*]] = icmp eq i64 [[IV]], 400 ; CHECK-NEXT: br i1 [[EXITCOND2_NOT]], label [[EXIT]], label [[LOOP_HEADER]], [[LOOP15:!llvm.loop !.*]] ; CHECK: exit: -; CHECK-NEXT: [[LCSSA:%.*]] = phi i32 [ 0, [[LOOP_HEADER]] ], [ [[ACCUM_NEXT]], [[LOOP_LATCH]] ], [ [[TMP7]], [[MIDDLE_BLOCK]] ] +; CHECK-NEXT: [[LCSSA:%.*]] = phi i32 [ 0, [[LOOP_HEADER]] ], [ [[ACCUM_NEXT]], [[LOOP_LATCH]] ] ; CHECK-NEXT: ret i32 [[LCSSA]] ; ; TAILFOLD-LABEL: @me_reduction(