From 2f2df751d453566ab70fd02b9a019cd66af76bc6 Mon Sep 17 00:00:00 2001 From: Florian Hahn Date: Mon, 28 Jul 2025 12:12:27 +0100 Subject: [PATCH] [LV] Use SCEV::getElementCount in selectEpilogueVectorizationFactor. (#150018) Follow-up to https://github.com/llvm/llvm-project/pull/149789 to use getElementCount to compute the remaining iterations in selectEpilogueVectrizationFactor. PR: https://github.com/llvm/llvm-project/pull/150018 --- .../Transforms/Vectorize/LoopVectorize.cpp | 22 +- .../LoopVectorize/AArch64/sve-epilog-vect.ll | 541 +++++++++++++++--- 2 files changed, 485 insertions(+), 78 deletions(-) diff --git a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp index 6616e61f9bb8..1c2de750b0d0 100644 --- a/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp +++ b/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp @@ -4500,19 +4500,17 @@ VectorizationFactor LoopVectorizationPlanner::selectEpilogueVectorizationFactor( Type *TCType = Legal->getWidestInductionType(); const SCEV *RemainingIterations = nullptr; unsigned MaxTripCount = 0; + const SCEV *TC = + vputils::getSCEVExprForVPValue(getPlanFor(MainLoopVF).getTripCount(), SE); + assert(!isa(TC) && "Trip count SCEV must be computable"); + RemainingIterations = + SE.getURemExpr(TC, SE.getElementCount(TCType, MainLoopVF * IC)); + + // No iterations left to process in the epilogue. + if (RemainingIterations->isZero()) + return Result; + if (MainLoopVF.isFixed()) { - // TODO: extend to support scalable VFs. - const SCEV *TC = vputils::getSCEVExprForVPValue( - getPlanFor(MainLoopVF).getTripCount(), SE); - assert(!isa(TC) && - "Trip count SCEV must be computable"); - RemainingIterations = SE.getURemExpr( - TC, SE.getConstant(TCType, MainLoopVF.getFixedValue() * IC)); - - // No iterations left to process in the epilogue. - if (RemainingIterations->isZero()) - return Result; - MaxTripCount = MainLoopVF.getFixedValue() * IC - 1; if (SE.isKnownPredicate(CmpInst::ICMP_ULT, RemainingIterations, SE.getConstant(TCType, MaxTripCount))) { diff --git a/llvm/test/Transforms/LoopVectorize/AArch64/sve-epilog-vect.ll b/llvm/test/Transforms/LoopVectorize/AArch64/sve-epilog-vect.ll index 18cc3a812f92..0f407cd56545 100644 --- a/llvm/test/Transforms/LoopVectorize/AArch64/sve-epilog-vect.ll +++ b/llvm/test/Transforms/LoopVectorize/AArch64/sve-epilog-vect.ll @@ -148,30 +148,12 @@ exit: ret void } - -; DEBUG: LV: Checking a loop in 'main_vf_vscale_x_2' -; DEBUG: Create Skeleton for epilogue vectorized loop (first pass) -; DEBUG: Main Loop VF:vscale x 2, Main Loop UF:2, Epilogue Loop VF:8, Epilogue Loop UF:1 - -; DEBUG-FORCED: LV: Checking a loop in 'main_vf_vscale_x_2' -; DEBUG-FORCED: LEV: Epilogue vectorization factor is forced. -; DEBUG-FORCED: Create Skeleton for epilogue vectorized loop (first pass) -; DEBUG-FORCED: Main Loop VF:vscale x 2, Main Loop UF:2, Epilogue Loop VF:8, Epilogue Loop UF:1 - -; When the vector.body uses VF=vscale x 1 (or VF=vscale x 2 because -; that's the minimum supported VF by SVE), we could still use a wide -; fixed-width VF=8 for the epilogue if the vectors are known to be -; sufficiently wide. This information can be deduced from vscale_range or -; VScaleForTuning (set by mcpu/mtune). -define void @main_vf_vscale_x_2(ptr %A) #0 vscale_range(8, 8) { -; CHECK-LABEL: @main_vf_vscale_x_2( -; CHECK-NEXT: iter.check: -; CHECK-NEXT: br i1 false, label [[VEC_EPILOG_SCALAR_PH:%.*]], label [[VECTOR_MAIN_LOOP_ITER_CHECK:%.*]] -; CHECK: vector.main.loop.iter.check: +define void @main_vf_vscale_x_2_no_epi_iteration(ptr %A) #0 vscale_range(8, 8) { +; CHECK-LABEL: @main_vf_vscale_x_2_no_epi_iteration( +; CHECK-NEXT: entry: ; CHECK-NEXT: [[TMP0:%.*]] = call i64 @llvm.vscale.i64() ; CHECK-NEXT: [[TMP1:%.*]] = mul nuw i64 [[TMP0]], 4 -; CHECK-NEXT: [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 1024, [[TMP1]] -; CHECK-NEXT: br i1 [[MIN_ITERS_CHECK]], label [[VEC_EPILOG_PH:%.*]], label [[VECTOR_PH:%.*]] +; CHECK-NEXT: br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]] ; CHECK: vector.ph: ; CHECK-NEXT: [[TMP2:%.*]] = call i64 @llvm.vscale.i64() ; CHECK-NEXT: [[TMP3:%.*]] = mul nuw i64 [[TMP2]], 4 @@ -182,40 +164,24 @@ define void @main_vf_vscale_x_2(ptr %A) #0 vscale_range(8, 8) { ; CHECK-NEXT: br label [[VECTOR_BODY:%.*]] ; CHECK: vector.body: ; CHECK-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ] -; CHECK-NEXT: [[TMP12:%.*]] = getelementptr inbounds i64, ptr [[A:%.*]], i64 [[INDEX]] -; CHECK-NEXT: [[TMP15:%.*]] = call i64 @llvm.vscale.i64() -; CHECK-NEXT: [[TMP16:%.*]] = mul nuw i64 [[TMP15]], 2 -; CHECK-NEXT: [[TMP17:%.*]] = getelementptr inbounds i64, ptr [[TMP12]], i64 [[TMP16]] -; CHECK-NEXT: store splat (i64 1), ptr [[TMP12]], align 1 -; CHECK-NEXT: store splat (i64 1), ptr [[TMP17]], align 1 +; CHECK-NEXT: [[TMP6:%.*]] = getelementptr inbounds i64, ptr [[A:%.*]], i64 [[INDEX]] +; CHECK-NEXT: [[TMP7:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-NEXT: [[TMP8:%.*]] = mul nuw i64 [[TMP7]], 2 +; CHECK-NEXT: [[TMP9:%.*]] = getelementptr inbounds i64, ptr [[TMP6]], i64 [[TMP8]] +; CHECK-NEXT: store splat (i64 1), ptr [[TMP6]], align 1 +; CHECK-NEXT: store splat (i64 1), ptr [[TMP9]], align 1 ; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP5]] -; CHECK-NEXT: [[TMP18:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]] -; CHECK-NEXT: br i1 [[TMP18]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP5:![0-9]+]] +; CHECK-NEXT: [[TMP11:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]] +; CHECK-NEXT: br i1 [[TMP11]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP5:![0-9]+]] ; CHECK: middle.block: ; CHECK-NEXT: [[CMP_N:%.*]] = icmp eq i64 1024, [[N_VEC]] -; CHECK-NEXT: br i1 [[CMP_N]], label [[EXIT:%.*]], label [[VEC_EPILOG_ITER_CHECK:%.*]] -; CHECK: vec.epilog.iter.check: -; CHECK-NEXT: [[N_VEC_REMAINING:%.*]] = sub i64 1024, [[N_VEC]] -; CHECK-NEXT: [[MIN_EPILOG_ITERS_CHECK:%.*]] = icmp ult i64 [[N_VEC_REMAINING]], 8 -; CHECK-NEXT: br i1 [[MIN_EPILOG_ITERS_CHECK]], label [[VEC_EPILOG_SCALAR_PH]], label [[VEC_EPILOG_PH]] -; CHECK: vec.epilog.ph: -; CHECK-NEXT: [[VEC_EPILOG_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[VEC_EPILOG_ITER_CHECK]] ], [ 0, [[VECTOR_MAIN_LOOP_ITER_CHECK]] ] -; CHECK-NEXT: br label [[VEC_EPILOG_VECTOR_BODY:%.*]] -; CHECK: vec.epilog.vector.body: -; CHECK-NEXT: [[INDEX1:%.*]] = phi i64 [ [[VEC_EPILOG_RESUME_VAL]], [[VEC_EPILOG_PH]] ], [ [[INDEX_NEXT2:%.*]], [[VEC_EPILOG_VECTOR_BODY]] ] -; CHECK-NEXT: [[TMP20:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[INDEX1]] -; CHECK-NEXT: store <8 x i64> splat (i64 1), ptr [[TMP20]], align 1 -; CHECK-NEXT: [[INDEX_NEXT2]] = add nuw i64 [[INDEX1]], 8 -; CHECK-NEXT: [[TMP22:%.*]] = icmp eq i64 [[INDEX_NEXT2]], 1024 -; CHECK-NEXT: br i1 [[TMP22]], label [[VEC_EPILOG_MIDDLE_BLOCK:%.*]], label [[VEC_EPILOG_VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]] -; CHECK: vec.epilog.middle.block: -; CHECK-NEXT: br i1 true, label [[EXIT]], label [[VEC_EPILOG_SCALAR_PH]] -; CHECK: vec.epilog.scalar.ph: -; CHECK-NEXT: [[BC_RESUME_VAL:%.*]] = phi i64 [ 1024, [[VEC_EPILOG_MIDDLE_BLOCK]] ], [ [[N_VEC]], [[VEC_EPILOG_ITER_CHECK]] ], [ 0, [[ITER_CHECK:%.*]] ] +; CHECK-NEXT: br i1 [[CMP_N]], label [[EXIT:%.*]], 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:%.*]] ; CHECK: for.body: ; -; CHECK-VF8-LABEL: @main_vf_vscale_x_2( +; CHECK-VF8-LABEL: @main_vf_vscale_x_2_no_epi_iteration( ; CHECK-VF8-NEXT: iter.check: ; CHECK-VF8-NEXT: br i1 false, label [[VEC_EPILOG_SCALAR_PH:%.*]], label [[VECTOR_MAIN_LOOP_ITER_CHECK:%.*]] ; CHECK-VF8: vector.main.loop.iter.check: @@ -233,15 +199,15 @@ define void @main_vf_vscale_x_2(ptr %A) #0 vscale_range(8, 8) { ; CHECK-VF8-NEXT: br label [[VECTOR_BODY:%.*]] ; CHECK-VF8: vector.body: ; CHECK-VF8-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ] -; CHECK-VF8-NEXT: [[TMP12:%.*]] = getelementptr inbounds i64, ptr [[A:%.*]], i64 [[INDEX]] -; CHECK-VF8-NEXT: [[TMP15:%.*]] = call i64 @llvm.vscale.i64() -; CHECK-VF8-NEXT: [[TMP16:%.*]] = mul nuw i64 [[TMP15]], 2 -; CHECK-VF8-NEXT: [[TMP17:%.*]] = getelementptr inbounds i64, ptr [[TMP12]], i64 [[TMP16]] -; CHECK-VF8-NEXT: store splat (i64 1), ptr [[TMP12]], align 1 -; CHECK-VF8-NEXT: store splat (i64 1), ptr [[TMP17]], align 1 +; CHECK-VF8-NEXT: [[TMP6:%.*]] = getelementptr inbounds i64, ptr [[A:%.*]], i64 [[INDEX]] +; CHECK-VF8-NEXT: [[TMP7:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-VF8-NEXT: [[TMP8:%.*]] = mul nuw i64 [[TMP7]], 2 +; CHECK-VF8-NEXT: [[TMP9:%.*]] = getelementptr inbounds i64, ptr [[TMP6]], i64 [[TMP8]] +; CHECK-VF8-NEXT: store splat (i64 1), ptr [[TMP6]], align 1 +; CHECK-VF8-NEXT: store splat (i64 1), ptr [[TMP9]], align 1 ; CHECK-VF8-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP5]] -; CHECK-VF8-NEXT: [[TMP18:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]] -; CHECK-VF8-NEXT: br i1 [[TMP18]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP5:![0-9]+]] +; CHECK-VF8-NEXT: [[TMP10:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]] +; CHECK-VF8-NEXT: br i1 [[TMP10]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP5:![0-9]+]] ; CHECK-VF8: middle.block: ; CHECK-VF8-NEXT: [[CMP_N:%.*]] = icmp eq i64 1024, [[N_VEC]] ; CHECK-VF8-NEXT: br i1 [[CMP_N]], label [[EXIT:%.*]], label [[VEC_EPILOG_ITER_CHECK:%.*]] @@ -254,11 +220,11 @@ define void @main_vf_vscale_x_2(ptr %A) #0 vscale_range(8, 8) { ; CHECK-VF8-NEXT: br label [[VEC_EPILOG_VECTOR_BODY:%.*]] ; CHECK-VF8: vec.epilog.vector.body: ; CHECK-VF8-NEXT: [[INDEX1:%.*]] = phi i64 [ [[VEC_EPILOG_RESUME_VAL]], [[VEC_EPILOG_PH]] ], [ [[INDEX_NEXT2:%.*]], [[VEC_EPILOG_VECTOR_BODY]] ] -; CHECK-VF8-NEXT: [[TMP20:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[INDEX1]] -; CHECK-VF8-NEXT: store <8 x i64> splat (i64 1), ptr [[TMP20]], align 1 +; CHECK-VF8-NEXT: [[TMP11:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[INDEX1]] +; CHECK-VF8-NEXT: store <8 x i64> splat (i64 1), ptr [[TMP11]], align 1 ; CHECK-VF8-NEXT: [[INDEX_NEXT2]] = add nuw i64 [[INDEX1]], 8 -; CHECK-VF8-NEXT: [[TMP22:%.*]] = icmp eq i64 [[INDEX_NEXT2]], 1024 -; CHECK-VF8-NEXT: br i1 [[TMP22]], label [[VEC_EPILOG_MIDDLE_BLOCK:%.*]], label [[VEC_EPILOG_VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]] +; CHECK-VF8-NEXT: [[TMP14:%.*]] = icmp eq i64 [[INDEX_NEXT2]], 1024 +; CHECK-VF8-NEXT: br i1 [[TMP14]], label [[VEC_EPILOG_MIDDLE_BLOCK:%.*]], label [[VEC_EPILOG_VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]] ; CHECK-VF8: vec.epilog.middle.block: ; CHECK-VF8-NEXT: br i1 true, label [[EXIT]], label [[VEC_EPILOG_SCALAR_PH]] ; CHECK-VF8: vec.epilog.scalar.ph: @@ -281,6 +247,146 @@ exit: ret void } +; DEBUG: LV: Checking a loop in 'main_vf_vscale_x_2' +; DEBUG: Create Skeleton for epilogue vectorized loop (first pass) +; DEBUG: Main Loop VF:vscale x 2, Main Loop UF:2, Epilogue Loop VF:8, Epilogue Loop UF:1 + +; DEBUG-FORCED: LV: Checking a loop in 'main_vf_vscale_x_2' +; DEBUG-FORCED: LEV: Epilogue vectorization factor is forced. +; DEBUG-FORCED: Create Skeleton for epilogue vectorized loop (first pass) +; DEBUG-FORCED: Main Loop VF:vscale x 2, Main Loop UF:2, Epilogue Loop VF:8, Epilogue Loop UF:1 + +; When the vector.body uses VF=vscale x 1 (or VF=vscale x 2 because +; that's the minimum supported VF by SVE), we could still use a wide +; fixed-width VF=8 for the epilogue if the vectors are known to be +; sufficiently wide. This information can be deduced from vscale_range or +; VScaleForTuning (set by mcpu/mtune). +define void @main_vf_vscale_x_2(ptr %A, i64 %n) #0 vscale_range(8, 8) { +; CHECK-LABEL: @main_vf_vscale_x_2( +; CHECK-NEXT: iter.check: +; CHECK-NEXT: [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[N:%.*]], 8 +; CHECK-NEXT: br i1 [[MIN_ITERS_CHECK]], label [[VEC_EPILOG_SCALAR_PH:%.*]], label [[VECTOR_MAIN_LOOP_ITER_CHECK:%.*]] +; CHECK: vector.main.loop.iter.check: +; CHECK-NEXT: [[TMP0:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-NEXT: [[TMP1:%.*]] = mul nuw i64 [[TMP0]], 4 +; CHECK-NEXT: [[MIN_ITERS_CHECK1:%.*]] = icmp ult i64 [[N]], [[TMP1]] +; CHECK-NEXT: br i1 [[MIN_ITERS_CHECK1]], label [[VEC_EPILOG_PH:%.*]], label [[VECTOR_PH:%.*]] +; CHECK: vector.ph: +; CHECK-NEXT: [[TMP2:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-NEXT: [[TMP3:%.*]] = mul nuw i64 [[TMP2]], 4 +; CHECK-NEXT: [[N_MOD_VF:%.*]] = urem i64 [[N]], [[TMP3]] +; CHECK-NEXT: [[N_VEC:%.*]] = sub i64 [[N]], [[N_MOD_VF]] +; CHECK-NEXT: [[TMP4:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-NEXT: [[TMP5:%.*]] = mul nuw i64 [[TMP4]], 4 +; CHECK-NEXT: br label [[VECTOR_BODY:%.*]] +; CHECK: vector.body: +; CHECK-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ] +; CHECK-NEXT: [[TMP12:%.*]] = getelementptr inbounds i64, ptr [[A:%.*]], i64 [[INDEX]] +; CHECK-NEXT: [[TMP15:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-NEXT: [[TMP16:%.*]] = mul nuw i64 [[TMP15]], 2 +; CHECK-NEXT: [[TMP17:%.*]] = getelementptr inbounds i64, ptr [[TMP12]], i64 [[TMP16]] +; CHECK-NEXT: store splat (i64 1), ptr [[TMP12]], align 1 +; CHECK-NEXT: store splat (i64 1), ptr [[TMP17]], align 1 +; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP5]] +; CHECK-NEXT: [[TMP18:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]] +; CHECK-NEXT: br i1 [[TMP18]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP7:![0-9]+]] +; CHECK: middle.block: +; CHECK-NEXT: [[CMP_N:%.*]] = icmp eq i64 [[N]], [[N_VEC]] +; CHECK-NEXT: br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH:%.*]] +; CHECK: vec.epilog.iter.check: +; CHECK-NEXT: [[N_VEC_REMAINING:%.*]] = sub i64 [[N]], [[N_VEC]] +; CHECK-NEXT: [[MIN_EPILOG_ITERS_CHECK:%.*]] = icmp ult i64 [[N_VEC_REMAINING]], 8 +; CHECK-NEXT: br i1 [[MIN_EPILOG_ITERS_CHECK]], label [[VEC_EPILOG_SCALAR_PH]], label [[VEC_EPILOG_PH]] +; CHECK: vec.epilog.ph: +; CHECK-NEXT: [[VEC_EPILOG_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[SCALAR_PH]] ], [ 0, [[VECTOR_MAIN_LOOP_ITER_CHECK]] ] +; CHECK-NEXT: [[N_MOD_VF2:%.*]] = urem i64 [[N]], 8 +; CHECK-NEXT: [[N_VEC3:%.*]] = sub i64 [[N]], [[N_MOD_VF2]] +; CHECK-NEXT: br label [[FOR_BODY:%.*]] +; CHECK: vec.epilog.vector.body: +; CHECK-NEXT: [[INDEX4:%.*]] = phi i64 [ [[VEC_EPILOG_RESUME_VAL]], [[VEC_EPILOG_PH]] ], [ [[INDEX_NEXT5:%.*]], [[FOR_BODY]] ] +; CHECK-NEXT: [[TMP19:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[INDEX4]] +; CHECK-NEXT: store <8 x i64> splat (i64 1), ptr [[TMP19]], align 1 +; CHECK-NEXT: [[INDEX_NEXT5]] = add nuw i64 [[INDEX4]], 8 +; CHECK-NEXT: [[TMP20:%.*]] = icmp eq i64 [[INDEX_NEXT5]], [[N_VEC3]] +; CHECK-NEXT: br i1 [[TMP20]], label [[VEC_EPILOG_MIDDLE_BLOCK:%.*]], label [[FOR_BODY]], !llvm.loop [[LOOP8:![0-9]+]] +; CHECK: vec.epilog.middle.block: +; CHECK-NEXT: [[CMP_N6:%.*]] = icmp eq i64 [[N]], [[N_VEC3]] +; CHECK-NEXT: br i1 [[CMP_N6]], label [[EXIT]], label [[VEC_EPILOG_SCALAR_PH]] +; CHECK: vec.epilog.scalar.ph: +; CHECK-NEXT: [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC3]], [[VEC_EPILOG_MIDDLE_BLOCK]] ], [ [[N_VEC]], [[SCALAR_PH]] ], [ 0, [[ITER_CHECK:%.*]] ] +; CHECK-NEXT: br label [[FOR_BODY1:%.*]] +; CHECK: for.body: +; +; CHECK-VF8-LABEL: @main_vf_vscale_x_2( +; CHECK-VF8-NEXT: iter.check: +; CHECK-VF8-NEXT: [[MIN_ITERS_CHECK1:%.*]] = icmp ult i64 [[N:%.*]], 8 +; CHECK-VF8-NEXT: br i1 [[MIN_ITERS_CHECK1]], label [[VEC_EPILOG_SCALAR_PH:%.*]], label [[VECTOR_MAIN_LOOP_ITER_CHECK:%.*]] +; CHECK-VF8: vector.main.loop.iter.check: +; CHECK-VF8-NEXT: [[TMP0:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-VF8-NEXT: [[TMP1:%.*]] = mul nuw i64 [[TMP0]], 4 +; CHECK-VF8-NEXT: [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[N]], [[TMP1]] +; CHECK-VF8-NEXT: br i1 [[MIN_ITERS_CHECK]], label [[VEC_EPILOG_PH:%.*]], label [[VECTOR_PH:%.*]] +; CHECK-VF8: vector.ph: +; CHECK-VF8-NEXT: [[TMP2:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-VF8-NEXT: [[TMP3:%.*]] = mul nuw i64 [[TMP2]], 4 +; CHECK-VF8-NEXT: [[N_MOD_VF:%.*]] = urem i64 [[N]], [[TMP3]] +; CHECK-VF8-NEXT: [[N_VEC:%.*]] = sub i64 [[N]], [[N_MOD_VF]] +; CHECK-VF8-NEXT: [[TMP4:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-VF8-NEXT: [[TMP5:%.*]] = mul nuw i64 [[TMP4]], 4 +; CHECK-VF8-NEXT: br label [[VECTOR_BODY:%.*]] +; CHECK-VF8: vector.body: +; CHECK-VF8-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ] +; CHECK-VF8-NEXT: [[TMP12:%.*]] = getelementptr inbounds i64, ptr [[A:%.*]], i64 [[INDEX]] +; CHECK-VF8-NEXT: [[TMP15:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-VF8-NEXT: [[TMP16:%.*]] = mul nuw i64 [[TMP15]], 2 +; CHECK-VF8-NEXT: [[TMP17:%.*]] = getelementptr inbounds i64, ptr [[TMP12]], i64 [[TMP16]] +; CHECK-VF8-NEXT: store splat (i64 1), ptr [[TMP12]], align 1 +; CHECK-VF8-NEXT: store splat (i64 1), ptr [[TMP17]], align 1 +; CHECK-VF8-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP5]] +; CHECK-VF8-NEXT: [[TMP18:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]] +; CHECK-VF8-NEXT: br i1 [[TMP18]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP8:![0-9]+]] +; CHECK-VF8: middle.block: +; CHECK-VF8-NEXT: [[CMP_N:%.*]] = icmp eq i64 [[N]], [[N_VEC]] +; CHECK-VF8-NEXT: br i1 [[CMP_N]], label [[EXIT:%.*]], label [[VEC_EPILOG_ITER_CHECK:%.*]] +; CHECK-VF8: vec.epilog.iter.check: +; CHECK-VF8-NEXT: [[N_VEC_REMAINING:%.*]] = sub i64 [[N]], [[N_VEC]] +; CHECK-VF8-NEXT: [[MIN_EPILOG_ITERS_CHECK:%.*]] = icmp ult i64 [[N_VEC_REMAINING]], 8 +; CHECK-VF8-NEXT: br i1 [[MIN_EPILOG_ITERS_CHECK]], label [[VEC_EPILOG_SCALAR_PH]], label [[VEC_EPILOG_PH]] +; CHECK-VF8: vec.epilog.ph: +; CHECK-VF8-NEXT: [[VEC_EPILOG_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[VEC_EPILOG_ITER_CHECK]] ], [ 0, [[VECTOR_MAIN_LOOP_ITER_CHECK]] ] +; CHECK-VF8-NEXT: [[N_MOD_VF2:%.*]] = urem i64 [[N]], 8 +; CHECK-VF8-NEXT: [[N_VEC3:%.*]] = sub i64 [[N]], [[N_MOD_VF2]] +; CHECK-VF8-NEXT: br label [[VEC_EPILOG_VECTOR_BODY:%.*]] +; CHECK-VF8: vec.epilog.vector.body: +; CHECK-VF8-NEXT: [[INDEX1:%.*]] = phi i64 [ [[VEC_EPILOG_RESUME_VAL]], [[VEC_EPILOG_PH]] ], [ [[INDEX_NEXT2:%.*]], [[VEC_EPILOG_VECTOR_BODY]] ] +; CHECK-VF8-NEXT: [[TMP20:%.*]] = getelementptr inbounds i64, ptr [[A]], i64 [[INDEX1]] +; CHECK-VF8-NEXT: store <8 x i64> splat (i64 1), ptr [[TMP20]], align 1 +; CHECK-VF8-NEXT: [[INDEX_NEXT2]] = add nuw i64 [[INDEX1]], 8 +; CHECK-VF8-NEXT: [[TMP19:%.*]] = icmp eq i64 [[INDEX_NEXT2]], [[N_VEC3]] +; CHECK-VF8-NEXT: br i1 [[TMP19]], label [[VEC_EPILOG_MIDDLE_BLOCK:%.*]], label [[VEC_EPILOG_VECTOR_BODY]], !llvm.loop [[LOOP9:![0-9]+]] +; CHECK-VF8: vec.epilog.middle.block: +; CHECK-VF8-NEXT: [[CMP_N6:%.*]] = icmp eq i64 [[N]], [[N_VEC3]] +; CHECK-VF8-NEXT: br i1 [[CMP_N6]], label [[EXIT]], label [[VEC_EPILOG_SCALAR_PH]] +; CHECK-VF8: vec.epilog.scalar.ph: +; CHECK-VF8-NEXT: [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC3]], [[VEC_EPILOG_MIDDLE_BLOCK]] ], [ [[N_VEC]], [[VEC_EPILOG_ITER_CHECK]] ], [ 0, [[ITER_CHECK:%.*]] ] +; CHECK-VF8-NEXT: br label [[FOR_BODY:%.*]] +; CHECK-VF8: for.body: +; +entry: + br label %for.body + +for.body: + %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ] + %arrayidx = getelementptr inbounds i64, ptr %A, i64 %iv + store i64 1, ptr %arrayidx, align 1 + %iv.next = add nuw nsw i64 %iv, 1 + %exitcond = icmp ne i64 %iv.next, %n + br i1 %exitcond, label %for.body, label %exit + +exit: + ret void +} + ; FIXME: The epilogue loop is currently miscompiled: the pointer induction ; uses an incorrect resume value. define void @test_pr57912_pointer_induction(ptr %start) #0 { @@ -313,7 +419,7 @@ define void @test_pr57912_pointer_induction(ptr %start) #0 { ; CHECK-NEXT: store zeroinitializer, ptr [[TMP19]], align 1 ; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP7]] ; CHECK-NEXT: [[TMP20:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]] -; CHECK-NEXT: br i1 [[TMP20]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP8:![0-9]+]] +; CHECK-NEXT: br i1 [[TMP20]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP10:![0-9]+]] ; CHECK: middle.block: ; CHECK-NEXT: [[CMP_N:%.*]] = icmp eq i64 10000, [[N_VEC]] ; CHECK-NEXT: br i1 [[CMP_N]], label [[EXIT:%.*]], label [[VEC_EPILOG_ITER_CHECK:%.*]] @@ -340,7 +446,7 @@ define void @test_pr57912_pointer_induction(ptr %start) #0 { ; CHECK-NEXT: store zeroinitializer, ptr [[TMP28]], align 1 ; CHECK-NEXT: [[INDEX_NEXT8]] = add nuw i64 [[INDEX7]], [[TMP26]] ; CHECK-NEXT: [[TMP30:%.*]] = icmp eq i64 [[INDEX_NEXT8]], [[N_VEC3]] -; CHECK-NEXT: br i1 [[TMP30]], label [[VEC_EPILOG_MIDDLE_BLOCK:%.*]], label [[VEC_EPILOG_VECTOR_BODY]], !llvm.loop [[LOOP9:![0-9]+]] +; CHECK-NEXT: br i1 [[TMP30]], label [[VEC_EPILOG_MIDDLE_BLOCK:%.*]], label [[VEC_EPILOG_VECTOR_BODY]], !llvm.loop [[LOOP11:![0-9]+]] ; CHECK: vec.epilog.middle.block: ; CHECK-NEXT: [[CMP_N6:%.*]] = icmp eq i64 10000, [[N_VEC3]] ; CHECK-NEXT: br i1 [[CMP_N6]], label [[EXIT]], label [[VEC_EPILOG_SCALAR_PH]] @@ -376,7 +482,7 @@ define void @test_pr57912_pointer_induction(ptr %start) #0 { ; CHECK-VF8-NEXT: store zeroinitializer, ptr [[TMP17]], align 1 ; CHECK-VF8-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP5]] ; CHECK-VF8-NEXT: [[TMP18:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]] -; CHECK-VF8-NEXT: br i1 [[TMP18]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP8:![0-9]+]] +; CHECK-VF8-NEXT: br i1 [[TMP18]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP11:![0-9]+]] ; CHECK-VF8: middle.block: ; CHECK-VF8-NEXT: [[CMP_N:%.*]] = icmp eq i64 10000, [[N_VEC]] ; CHECK-VF8-NEXT: br i1 [[CMP_N]], label [[EXIT:%.*]], label [[VEC_EPILOG_ITER_CHECK:%.*]] @@ -395,7 +501,7 @@ define void @test_pr57912_pointer_induction(ptr %start) #0 { ; CHECK-VF8-NEXT: store <8 x i8> zeroinitializer, ptr [[TMP20]], align 1 ; CHECK-VF8-NEXT: [[INDEX_NEXT4]] = add nuw i64 [[INDEX3]], 8 ; CHECK-VF8-NEXT: [[TMP22:%.*]] = icmp eq i64 [[INDEX_NEXT4]], 10000 -; CHECK-VF8-NEXT: br i1 [[TMP22]], label [[VEC_EPILOG_MIDDLE_BLOCK:%.*]], label [[VEC_EPILOG_VECTOR_BODY]], !llvm.loop [[LOOP9:![0-9]+]] +; CHECK-VF8-NEXT: br i1 [[TMP22]], label [[VEC_EPILOG_MIDDLE_BLOCK:%.*]], label [[VEC_EPILOG_VECTOR_BODY]], !llvm.loop [[LOOP12:![0-9]+]] ; CHECK-VF8: vec.epilog.middle.block: ; CHECK-VF8-NEXT: br i1 true, label [[EXIT]], label [[VEC_EPILOG_SCALAR_PH]] ; CHECK-VF8: vec.epilog.scalar.ph: @@ -420,4 +526,307 @@ exit: ret void } +; Loop with vscale-based trip count vscale x 1033. +define void @trip_count_vscale(ptr noalias %a, ptr noalias %b) vscale_range(1, 16) #0 { +; CHECK-LABEL: @trip_count_vscale( +; CHECK-NEXT: iter.check: +; CHECK-NEXT: [[V:%.*]] = tail call i64 @llvm.vscale.i64() +; CHECK-NEXT: [[N:%.*]] = mul nuw nsw i64 [[V]], 1033 +; CHECK-NEXT: [[TMP0:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-NEXT: [[TMP1:%.*]] = mul nuw i64 [[TMP0]], 2 +; CHECK-NEXT: br i1 false, label [[VEC_EPILOG_SCALAR_PH:%.*]], label [[VECTOR_MAIN_LOOP_ITER_CHECK:%.*]] +; CHECK: vector.main.loop.iter.check: +; CHECK-NEXT: [[TMP2:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-NEXT: [[TMP3:%.*]] = mul nuw i64 [[TMP2]], 8 +; CHECK-NEXT: [[MIN_ITERS_CHECK1:%.*]] = icmp ult i64 [[N]], [[TMP3]] +; CHECK-NEXT: br i1 [[MIN_ITERS_CHECK1]], label [[VEC_EPILOG_PH:%.*]], label [[VECTOR_PH:%.*]] +; CHECK: vector.ph: +; CHECK-NEXT: [[TMP4:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-NEXT: [[TMP5:%.*]] = mul nuw i64 [[TMP4]], 8 +; CHECK-NEXT: [[N_MOD_VF:%.*]] = urem i64 [[N]], [[TMP5]] +; CHECK-NEXT: [[N_VEC:%.*]] = sub i64 [[N]], [[N_MOD_VF]] +; CHECK-NEXT: [[TMP6:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-NEXT: [[TMP7:%.*]] = mul nuw i64 [[TMP6]], 8 +; CHECK-NEXT: br label [[VECTOR_BODY:%.*]] +; CHECK: vector.body: +; CHECK-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ] +; CHECK-NEXT: [[TMP8:%.*]] = getelementptr inbounds nuw float, ptr [[A:%.*]], i64 [[INDEX]] +; CHECK-NEXT: [[TMP10:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-NEXT: [[TMP11:%.*]] = mul nuw i64 [[TMP10]], 4 +; CHECK-NEXT: [[TMP12:%.*]] = getelementptr inbounds nuw float, ptr [[TMP8]], i64 [[TMP11]] +; CHECK-NEXT: [[WIDE_LOAD:%.*]] = load , ptr [[TMP8]], align 4 +; CHECK-NEXT: [[WIDE_LOAD2:%.*]] = load , ptr [[TMP12]], align 4 +; CHECK-NEXT: [[TMP13:%.*]] = getelementptr inbounds nuw float, ptr [[B:%.*]], i64 [[INDEX]] +; CHECK-NEXT: [[TMP15:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-NEXT: [[TMP16:%.*]] = mul nuw i64 [[TMP15]], 4 +; CHECK-NEXT: [[TMP17:%.*]] = getelementptr inbounds nuw float, ptr [[TMP13]], i64 [[TMP16]] +; CHECK-NEXT: [[WIDE_LOAD3:%.*]] = load , ptr [[TMP13]], align 4 +; CHECK-NEXT: [[WIDE_LOAD4:%.*]] = load , ptr [[TMP17]], align 4 +; CHECK-NEXT: [[TMP18:%.*]] = fmul [[WIDE_LOAD]], [[WIDE_LOAD3]] +; CHECK-NEXT: [[TMP19:%.*]] = fmul [[WIDE_LOAD2]], [[WIDE_LOAD4]] +; CHECK-NEXT: [[TMP20:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-NEXT: [[TMP21:%.*]] = mul nuw i64 [[TMP20]], 4 +; CHECK-NEXT: [[TMP22:%.*]] = getelementptr inbounds nuw float, ptr [[TMP13]], i64 [[TMP21]] +; CHECK-NEXT: store [[TMP18]], ptr [[TMP13]], align 4 +; CHECK-NEXT: store [[TMP19]], ptr [[TMP22]], align 4 +; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP7]] +; CHECK-NEXT: [[TMP23:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]] +; CHECK-NEXT: br i1 [[TMP23]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP13:![0-9]+]] +; CHECK: middle.block: +; CHECK-NEXT: [[CMP_N:%.*]] = icmp eq i64 [[N]], [[N_VEC]] +; CHECK-NEXT: br i1 [[CMP_N]], label [[EXIT:%.*]], label [[VEC_EPILOG_ITER_CHECK:%.*]] +; CHECK: vec.epilog.iter.check: +; CHECK-NEXT: [[N_VEC_REMAINING:%.*]] = sub i64 [[N]], [[N_VEC]] +; CHECK-NEXT: [[TMP24:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-NEXT: [[TMP25:%.*]] = mul nuw i64 [[TMP24]], 2 +; CHECK-NEXT: [[MIN_EPILOG_ITERS_CHECK:%.*]] = icmp ult i64 [[N_VEC_REMAINING]], [[TMP25]] +; CHECK-NEXT: br i1 [[MIN_EPILOG_ITERS_CHECK]], label [[VEC_EPILOG_SCALAR_PH]], label [[VEC_EPILOG_PH]] +; CHECK: vec.epilog.ph: +; CHECK-NEXT: [[VEC_EPILOG_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[VEC_EPILOG_ITER_CHECK]] ], [ 0, [[VECTOR_MAIN_LOOP_ITER_CHECK]] ] +; CHECK-NEXT: [[TMP26:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-NEXT: [[TMP27:%.*]] = mul nuw i64 [[TMP26]], 2 +; CHECK-NEXT: [[N_MOD_VF5:%.*]] = urem i64 [[N]], [[TMP27]] +; CHECK-NEXT: [[N_VEC6:%.*]] = sub i64 [[N]], [[N_MOD_VF5]] +; CHECK-NEXT: [[TMP28:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-NEXT: [[TMP29:%.*]] = mul nuw i64 [[TMP28]], 2 +; CHECK-NEXT: br label [[VEC_EPILOG_VECTOR_BODY:%.*]] +; CHECK: vec.epilog.vector.body: +; CHECK-NEXT: [[INDEX7:%.*]] = phi i64 [ [[VEC_EPILOG_RESUME_VAL]], [[VEC_EPILOG_PH]] ], [ [[INDEX_NEXT10:%.*]], [[VEC_EPILOG_VECTOR_BODY]] ] +; CHECK-NEXT: [[TMP30:%.*]] = getelementptr inbounds nuw float, ptr [[A]], i64 [[INDEX7]] +; CHECK-NEXT: [[WIDE_LOAD8:%.*]] = load , ptr [[TMP30]], align 4 +; CHECK-NEXT: [[TMP32:%.*]] = getelementptr inbounds nuw float, ptr [[B]], i64 [[INDEX7]] +; CHECK-NEXT: [[WIDE_LOAD9:%.*]] = load , ptr [[TMP32]], align 4 +; CHECK-NEXT: [[TMP34:%.*]] = fmul [[WIDE_LOAD8]], [[WIDE_LOAD9]] +; CHECK-NEXT: store [[TMP34]], ptr [[TMP32]], align 4 +; CHECK-NEXT: [[INDEX_NEXT10]] = add nuw i64 [[INDEX7]], [[TMP29]] +; CHECK-NEXT: [[TMP35:%.*]] = icmp eq i64 [[INDEX_NEXT10]], [[N_VEC6]] +; CHECK-NEXT: br i1 [[TMP35]], label [[VEC_EPILOG_MIDDLE_BLOCK:%.*]], label [[VEC_EPILOG_VECTOR_BODY]], !llvm.loop [[LOOP14:![0-9]+]] +; CHECK: vec.epilog.middle.block: +; CHECK-NEXT: [[CMP_N11:%.*]] = icmp eq i64 [[N]], [[N_VEC6]] +; CHECK-NEXT: br i1 [[CMP_N11]], label [[EXIT]], label [[VEC_EPILOG_SCALAR_PH]] +; CHECK: vec.epilog.scalar.ph: +; CHECK-NEXT: [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC6]], [[VEC_EPILOG_MIDDLE_BLOCK]] ], [ [[N_VEC]], [[VEC_EPILOG_ITER_CHECK]] ], [ 0, [[ITER_CHECK:%.*]] ] +; CHECK-NEXT: br label [[FOR_BODY:%.*]] +; CHECK: for.body: +; +; CHECK-VF8-LABEL: @trip_count_vscale( +; CHECK-VF8-NEXT: entry: +; CHECK-VF8-NEXT: [[V:%.*]] = tail call i64 @llvm.vscale.i64() +; CHECK-VF8-NEXT: [[N:%.*]] = mul nuw nsw i64 [[V]], 1033 +; CHECK-VF8-NEXT: [[TMP0:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-VF8-NEXT: [[TMP1:%.*]] = mul nuw i64 [[TMP0]], 8 +; CHECK-VF8-NEXT: br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]] +; CHECK-VF8: vector.ph: +; CHECK-VF8-NEXT: [[TMP2:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-VF8-NEXT: [[TMP3:%.*]] = mul nuw i64 [[TMP2]], 8 +; CHECK-VF8-NEXT: [[N_MOD_VF:%.*]] = urem i64 [[N]], [[TMP3]] +; CHECK-VF8-NEXT: [[N_VEC:%.*]] = sub i64 [[N]], [[N_MOD_VF]] +; CHECK-VF8-NEXT: [[TMP4:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-VF8-NEXT: [[TMP5:%.*]] = mul nuw i64 [[TMP4]], 8 +; CHECK-VF8-NEXT: br label [[VECTOR_BODY:%.*]] +; CHECK-VF8: vector.body: +; CHECK-VF8-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ] +; CHECK-VF8-NEXT: [[TMP6:%.*]] = getelementptr inbounds nuw float, ptr [[A:%.*]], i64 [[INDEX]] +; CHECK-VF8-NEXT: [[TMP8:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-VF8-NEXT: [[TMP9:%.*]] = mul nuw i64 [[TMP8]], 4 +; CHECK-VF8-NEXT: [[TMP10:%.*]] = getelementptr inbounds nuw float, ptr [[TMP6]], i64 [[TMP9]] +; CHECK-VF8-NEXT: [[WIDE_LOAD:%.*]] = load , ptr [[TMP6]], align 4 +; CHECK-VF8-NEXT: [[WIDE_LOAD1:%.*]] = load , ptr [[TMP10]], align 4 +; CHECK-VF8-NEXT: [[TMP11:%.*]] = getelementptr inbounds nuw float, ptr [[B:%.*]], i64 [[INDEX]] +; CHECK-VF8-NEXT: [[TMP13:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-VF8-NEXT: [[TMP14:%.*]] = mul nuw i64 [[TMP13]], 4 +; CHECK-VF8-NEXT: [[TMP15:%.*]] = getelementptr inbounds nuw float, ptr [[TMP11]], i64 [[TMP14]] +; CHECK-VF8-NEXT: [[WIDE_LOAD2:%.*]] = load , ptr [[TMP11]], align 4 +; CHECK-VF8-NEXT: [[WIDE_LOAD3:%.*]] = load , ptr [[TMP15]], align 4 +; CHECK-VF8-NEXT: [[TMP16:%.*]] = fmul [[WIDE_LOAD]], [[WIDE_LOAD2]] +; CHECK-VF8-NEXT: [[TMP17:%.*]] = fmul [[WIDE_LOAD1]], [[WIDE_LOAD3]] +; CHECK-VF8-NEXT: [[TMP18:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-VF8-NEXT: [[TMP19:%.*]] = mul nuw i64 [[TMP18]], 4 +; CHECK-VF8-NEXT: [[TMP20:%.*]] = getelementptr inbounds nuw float, ptr [[TMP11]], i64 [[TMP19]] +; CHECK-VF8-NEXT: store [[TMP16]], ptr [[TMP11]], align 4 +; CHECK-VF8-NEXT: store [[TMP17]], ptr [[TMP20]], align 4 +; CHECK-VF8-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP5]] +; CHECK-VF8-NEXT: [[TMP21:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]] +; CHECK-VF8-NEXT: br i1 [[TMP21]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP14:![0-9]+]] +; CHECK-VF8: middle.block: +; CHECK-VF8-NEXT: [[CMP_N:%.*]] = icmp eq i64 [[N]], [[N_VEC]] +; CHECK-VF8-NEXT: br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]] +; CHECK-VF8: scalar.ph: +; CHECK-VF8-NEXT: [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ] +; CHECK-VF8-NEXT: br label [[FOR_BODY:%.*]] +; CHECK-VF8: for.body: +; +entry: + %v = tail call i64 @llvm.vscale.i64() + %n = mul nuw nsw i64 %v, 1033 + br label %for.body + +for.body: + %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ] + %arrayidx = getelementptr inbounds nuw float, ptr %a, i64 %iv + %l.a = load float, ptr %arrayidx, align 4 + %arrayidx3 = getelementptr inbounds nuw float, ptr %b, i64 %iv + %l.b = load float, ptr %arrayidx3, align 4 + %mul4 = fmul float %l.a, %l.b + store float %mul4, ptr %arrayidx3, align 4 + %iv.next = add nuw nsw i64 %iv, 1 + %ec = icmp eq i64 %iv.next, %n + br i1 %ec, label %exit, label %for.body + +exit: + ret void +} + +; Loop with vscale-based trip count vscale x 1024. +; TODO: No epilogue vectorizations should remain when choosing VF = vscale x 4. +define void @trip_count_vscale_no_epilogue_iterations(ptr noalias %a, ptr noalias %b) vscale_range(1, 16) #0 { +; CHECK-LABEL: @trip_count_vscale_no_epilogue_iterations( +; CHECK-NEXT: iter.check: +; CHECK-NEXT: [[V:%.*]] = tail call i64 @llvm.vscale.i64() +; CHECK-NEXT: [[N:%.*]] = mul nuw nsw i64 [[V]], 1024 +; CHECK-NEXT: [[TMP0:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-NEXT: [[TMP1:%.*]] = mul nuw i64 [[TMP0]], 2 +; CHECK-NEXT: br i1 false, label [[VEC_EPILOG_SCALAR_PH:%.*]], label [[VECTOR_MAIN_LOOP_ITER_CHECK:%.*]] +; CHECK: vector.main.loop.iter.check: +; CHECK-NEXT: [[TMP2:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-NEXT: [[TMP3:%.*]] = mul nuw i64 [[TMP2]], 8 +; CHECK-NEXT: [[MIN_ITERS_CHECK1:%.*]] = icmp ult i64 [[N]], [[TMP3]] +; CHECK-NEXT: br i1 [[MIN_ITERS_CHECK1]], label [[VEC_EPILOG_PH:%.*]], label [[VECTOR_PH:%.*]] +; CHECK: vector.ph: +; CHECK-NEXT: [[TMP4:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-NEXT: [[TMP5:%.*]] = mul nuw i64 [[TMP4]], 8 +; CHECK-NEXT: [[N_MOD_VF:%.*]] = urem i64 [[N]], [[TMP5]] +; CHECK-NEXT: [[N_VEC:%.*]] = sub i64 [[N]], [[N_MOD_VF]] +; CHECK-NEXT: [[TMP6:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-NEXT: [[TMP7:%.*]] = mul nuw i64 [[TMP6]], 8 +; CHECK-NEXT: br label [[VECTOR_BODY:%.*]] +; CHECK: vector.body: +; CHECK-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ] +; CHECK-NEXT: [[TMP8:%.*]] = getelementptr inbounds nuw float, ptr [[A:%.*]], i64 [[INDEX]] +; CHECK-NEXT: [[TMP10:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-NEXT: [[TMP11:%.*]] = mul nuw i64 [[TMP10]], 4 +; CHECK-NEXT: [[TMP12:%.*]] = getelementptr inbounds nuw float, ptr [[TMP8]], i64 [[TMP11]] +; CHECK-NEXT: [[WIDE_LOAD:%.*]] = load , ptr [[TMP8]], align 4 +; CHECK-NEXT: [[WIDE_LOAD2:%.*]] = load , ptr [[TMP12]], align 4 +; CHECK-NEXT: [[TMP13:%.*]] = getelementptr inbounds nuw float, ptr [[B:%.*]], i64 [[INDEX]] +; CHECK-NEXT: [[TMP15:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-NEXT: [[TMP16:%.*]] = mul nuw i64 [[TMP15]], 4 +; CHECK-NEXT: [[TMP17:%.*]] = getelementptr inbounds nuw float, ptr [[TMP13]], i64 [[TMP16]] +; CHECK-NEXT: [[WIDE_LOAD3:%.*]] = load , ptr [[TMP13]], align 4 +; CHECK-NEXT: [[WIDE_LOAD4:%.*]] = load , ptr [[TMP17]], align 4 +; CHECK-NEXT: [[TMP18:%.*]] = fmul [[WIDE_LOAD]], [[WIDE_LOAD3]] +; CHECK-NEXT: [[TMP19:%.*]] = fmul [[WIDE_LOAD2]], [[WIDE_LOAD4]] +; CHECK-NEXT: [[TMP20:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-NEXT: [[TMP21:%.*]] = mul nuw i64 [[TMP20]], 4 +; CHECK-NEXT: [[TMP22:%.*]] = getelementptr inbounds nuw float, ptr [[TMP13]], i64 [[TMP21]] +; CHECK-NEXT: store [[TMP18]], ptr [[TMP13]], align 4 +; CHECK-NEXT: store [[TMP19]], ptr [[TMP22]], align 4 +; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP7]] +; CHECK-NEXT: [[TMP23:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]] +; CHECK-NEXT: br i1 [[TMP23]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP16:![0-9]+]] +; CHECK: middle.block: +; CHECK-NEXT: [[CMP_N:%.*]] = icmp eq i64 [[N]], [[N_VEC]] +; CHECK-NEXT: br i1 [[CMP_N]], label [[EXIT:%.*]], label [[VEC_EPILOG_ITER_CHECK:%.*]] +; CHECK: vec.epilog.iter.check: +; CHECK-NEXT: [[N_VEC_REMAINING:%.*]] = sub i64 [[N]], [[N_VEC]] +; CHECK-NEXT: [[TMP24:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-NEXT: [[TMP25:%.*]] = mul nuw i64 [[TMP24]], 2 +; CHECK-NEXT: [[MIN_EPILOG_ITERS_CHECK:%.*]] = icmp ult i64 [[N_VEC_REMAINING]], [[TMP25]] +; CHECK-NEXT: br i1 [[MIN_EPILOG_ITERS_CHECK]], label [[VEC_EPILOG_SCALAR_PH]], label [[VEC_EPILOG_PH]] +; CHECK: vec.epilog.ph: +; CHECK-NEXT: [[VEC_EPILOG_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[VEC_EPILOG_ITER_CHECK]] ], [ 0, [[VECTOR_MAIN_LOOP_ITER_CHECK]] ] +; CHECK-NEXT: [[TMP26:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-NEXT: [[TMP27:%.*]] = mul nuw i64 [[TMP26]], 2 +; CHECK-NEXT: [[N_MOD_VF5:%.*]] = urem i64 [[N]], [[TMP27]] +; CHECK-NEXT: [[N_VEC6:%.*]] = sub i64 [[N]], [[N_MOD_VF5]] +; CHECK-NEXT: [[TMP28:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-NEXT: [[TMP29:%.*]] = mul nuw i64 [[TMP28]], 2 +; CHECK-NEXT: br label [[VEC_EPILOG_VECTOR_BODY:%.*]] +; CHECK: vec.epilog.vector.body: +; CHECK-NEXT: [[INDEX7:%.*]] = phi i64 [ [[VEC_EPILOG_RESUME_VAL]], [[VEC_EPILOG_PH]] ], [ [[INDEX_NEXT10:%.*]], [[VEC_EPILOG_VECTOR_BODY]] ] +; CHECK-NEXT: [[TMP30:%.*]] = getelementptr inbounds nuw float, ptr [[A]], i64 [[INDEX7]] +; CHECK-NEXT: [[WIDE_LOAD8:%.*]] = load , ptr [[TMP30]], align 4 +; CHECK-NEXT: [[TMP32:%.*]] = getelementptr inbounds nuw float, ptr [[B]], i64 [[INDEX7]] +; CHECK-NEXT: [[WIDE_LOAD9:%.*]] = load , ptr [[TMP32]], align 4 +; CHECK-NEXT: [[TMP34:%.*]] = fmul [[WIDE_LOAD8]], [[WIDE_LOAD9]] +; CHECK-NEXT: store [[TMP34]], ptr [[TMP32]], align 4 +; CHECK-NEXT: [[INDEX_NEXT10]] = add nuw i64 [[INDEX7]], [[TMP29]] +; CHECK-NEXT: [[TMP35:%.*]] = icmp eq i64 [[INDEX_NEXT10]], [[N_VEC6]] +; CHECK-NEXT: br i1 [[TMP35]], label [[VEC_EPILOG_MIDDLE_BLOCK:%.*]], label [[VEC_EPILOG_VECTOR_BODY]], !llvm.loop [[LOOP17:![0-9]+]] +; CHECK: vec.epilog.middle.block: +; CHECK-NEXT: [[CMP_N11:%.*]] = icmp eq i64 [[N]], [[N_VEC6]] +; CHECK-NEXT: br i1 [[CMP_N11]], label [[EXIT]], label [[VEC_EPILOG_SCALAR_PH]] +; CHECK: vec.epilog.scalar.ph: +; CHECK-NEXT: [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC6]], [[VEC_EPILOG_MIDDLE_BLOCK]] ], [ [[N_VEC]], [[VEC_EPILOG_ITER_CHECK]] ], [ 0, [[ITER_CHECK:%.*]] ] +; CHECK-NEXT: br label [[FOR_BODY:%.*]] +; CHECK: for.body: +; +; CHECK-VF8-LABEL: @trip_count_vscale_no_epilogue_iterations( +; CHECK-VF8-NEXT: entry: +; CHECK-VF8-NEXT: [[V:%.*]] = tail call i64 @llvm.vscale.i64() +; CHECK-VF8-NEXT: [[N:%.*]] = mul nuw nsw i64 [[V]], 1024 +; CHECK-VF8-NEXT: [[TMP0:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-VF8-NEXT: [[TMP1:%.*]] = mul nuw i64 [[TMP0]], 8 +; CHECK-VF8-NEXT: br i1 false, label [[SCALAR_PH:%.*]], label [[VECTOR_PH:%.*]] +; CHECK-VF8: vector.ph: +; CHECK-VF8-NEXT: [[TMP2:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-VF8-NEXT: [[TMP3:%.*]] = mul nuw i64 [[TMP2]], 8 +; CHECK-VF8-NEXT: [[N_MOD_VF:%.*]] = urem i64 [[N]], [[TMP3]] +; CHECK-VF8-NEXT: [[N_VEC:%.*]] = sub i64 [[N]], [[N_MOD_VF]] +; CHECK-VF8-NEXT: [[TMP4:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-VF8-NEXT: [[TMP5:%.*]] = mul nuw i64 [[TMP4]], 8 +; CHECK-VF8-NEXT: br label [[VECTOR_BODY:%.*]] +; CHECK-VF8: vector.body: +; CHECK-VF8-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ] +; CHECK-VF8-NEXT: [[TMP6:%.*]] = getelementptr inbounds nuw float, ptr [[A:%.*]], i64 [[INDEX]] +; CHECK-VF8-NEXT: [[TMP8:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-VF8-NEXT: [[TMP9:%.*]] = mul nuw i64 [[TMP8]], 4 +; CHECK-VF8-NEXT: [[TMP10:%.*]] = getelementptr inbounds nuw float, ptr [[TMP6]], i64 [[TMP9]] +; CHECK-VF8-NEXT: [[WIDE_LOAD:%.*]] = load , ptr [[TMP6]], align 4 +; CHECK-VF8-NEXT: [[WIDE_LOAD1:%.*]] = load , ptr [[TMP10]], align 4 +; CHECK-VF8-NEXT: [[TMP11:%.*]] = getelementptr inbounds nuw float, ptr [[B:%.*]], i64 [[INDEX]] +; CHECK-VF8-NEXT: [[TMP13:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-VF8-NEXT: [[TMP14:%.*]] = mul nuw i64 [[TMP13]], 4 +; CHECK-VF8-NEXT: [[TMP15:%.*]] = getelementptr inbounds nuw float, ptr [[TMP11]], i64 [[TMP14]] +; CHECK-VF8-NEXT: [[WIDE_LOAD2:%.*]] = load , ptr [[TMP11]], align 4 +; CHECK-VF8-NEXT: [[WIDE_LOAD3:%.*]] = load , ptr [[TMP15]], align 4 +; CHECK-VF8-NEXT: [[TMP16:%.*]] = fmul [[WIDE_LOAD]], [[WIDE_LOAD2]] +; CHECK-VF8-NEXT: [[TMP17:%.*]] = fmul [[WIDE_LOAD1]], [[WIDE_LOAD3]] +; CHECK-VF8-NEXT: [[TMP18:%.*]] = call i64 @llvm.vscale.i64() +; CHECK-VF8-NEXT: [[TMP19:%.*]] = mul nuw i64 [[TMP18]], 4 +; CHECK-VF8-NEXT: [[TMP20:%.*]] = getelementptr inbounds nuw float, ptr [[TMP11]], i64 [[TMP19]] +; CHECK-VF8-NEXT: store [[TMP16]], ptr [[TMP11]], align 4 +; CHECK-VF8-NEXT: store [[TMP17]], ptr [[TMP20]], align 4 +; CHECK-VF8-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], [[TMP5]] +; CHECK-VF8-NEXT: [[TMP21:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]] +; CHECK-VF8-NEXT: br i1 [[TMP21]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP16:![0-9]+]] +; CHECK-VF8: middle.block: +; CHECK-VF8-NEXT: [[CMP_N:%.*]] = icmp eq i64 [[N]], [[N_VEC]] +; CHECK-VF8-NEXT: br i1 [[CMP_N]], label [[EXIT:%.*]], label [[SCALAR_PH]] +; CHECK-VF8: scalar.ph: +; CHECK-VF8-NEXT: [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], [[MIDDLE_BLOCK]] ], [ 0, [[ENTRY:%.*]] ] +; CHECK-VF8-NEXT: br label [[FOR_BODY:%.*]] +; CHECK-VF8: for.body: +; +entry: + %v = tail call i64 @llvm.vscale.i64() + %n = mul nuw nsw i64 %v, 1024 + br label %for.body + +for.body: + %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ] + %arrayidx = getelementptr inbounds nuw float, ptr %a, i64 %iv + %l.a = load float, ptr %arrayidx, align 4 + %arrayidx3 = getelementptr inbounds nuw float, ptr %b, i64 %iv + %l.b = load float, ptr %arrayidx3, align 4 + %mul4 = fmul float %l.a, %l.b + store float %mul4, ptr %arrayidx3, align 4 + %iv.next = add nuw nsw i64 %iv, 1 + %ec = icmp eq i64 %iv.next, %n + br i1 %ec, label %exit, label %for.body + +exit: + ret void +} + attributes #0 = { "target-features"="+sve" }