| ; NOTE: Assertions have been autogenerated by utils/update_test_checks.py |
| ; RUN: opt -passes=loop-vectorize -S %s | FileCheck %s --check-prefix=CHECK |
| ; RUN: opt -passes=loop-vectorize -debug-only=loop-vectorize -disable-output %s 2>&1 | FileCheck %s --check-prefix=CHECK-COST |
| ; REQUIRES: asserts |
| |
| target datalayout = "e-m:e-i8:8:32-i16:16:32-i64:64-i128:128-n32:64-S128" |
| target triple = "aarch64--linux-gnu" |
| |
| ; CHECK-COST-LABEL: sadd |
| ; CHECK-COST: Found an estimated cost of 6 for VF 1 For instruction: %1 = tail call i16 @llvm.sadd.sat.i16(i16 %0, i16 %offset) |
| ; CHECK-COST: Cost of 4 for VF 2: WIDEN-INTRINSIC ir<%1> = call llvm.sadd.sat(ir<%0>, ir<%offset>) |
| ; CHECK-COST: Cost of 1 for VF 4: WIDEN-INTRINSIC ir<%1> = call llvm.sadd.sat(ir<%0>, ir<%offset>) |
| ; CHECK-COST: Cost of 1 for VF 8: WIDEN-INTRINSIC ir<%1> = call llvm.sadd.sat(ir<%0>, ir<%offset>) |
| |
| define void @saddsat(ptr nocapture readonly %pSrc, i16 signext %offset, ptr nocapture noalias %pDst, i32 %blockSize) #0 { |
| ; CHECK-LABEL: @saddsat( |
| ; CHECK-NEXT: entry: |
| ; CHECK-NEXT: [[CMP_NOT6:%.*]] = icmp eq i32 [[BLOCKSIZE:%.*]], 0 |
| ; CHECK-NEXT: br i1 [[CMP_NOT6]], label [[WHILE_END:%.*]], label [[ITER_CHECK:%.*]] |
| ; CHECK: iter.check: |
| ; CHECK-NEXT: [[TMP0:%.*]] = zext i32 [[BLOCKSIZE]] to i64 |
| ; CHECK-NEXT: [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[TMP0]], 4 |
| ; 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: [[MIN_ITERS_CHECK1:%.*]] = icmp ult i64 [[TMP0]], 16 |
| ; CHECK-NEXT: br i1 [[MIN_ITERS_CHECK1]], label [[VEC_EPILOG_PH:%.*]], label [[VECTOR_PH:%.*]] |
| ; CHECK: vector.ph: |
| ; CHECK-NEXT: [[N_MOD_VF:%.*]] = urem i64 [[TMP0]], 16 |
| ; CHECK-NEXT: [[N_VEC:%.*]] = sub i64 [[TMP0]], [[N_MOD_VF]] |
| ; CHECK-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <8 x i16> poison, i16 [[OFFSET:%.*]], i64 0 |
| ; CHECK-NEXT: [[BROADCAST_SPLAT:%.*]] = shufflevector <8 x i16> [[BROADCAST_SPLATINSERT]], <8 x i16> poison, <8 x i32> zeroinitializer |
| ; CHECK-NEXT: br label [[VECTOR_BODY:%.*]] |
| ; CHECK: vector.body: |
| ; CHECK-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ] |
| ; CHECK-NEXT: [[OFFSET_IDX:%.*]] = mul i64 [[INDEX]], 2 |
| ; CHECK-NEXT: [[NEXT_GEP:%.*]] = getelementptr i8, ptr [[PSRC:%.*]], i64 [[OFFSET_IDX]] |
| ; CHECK-NEXT: [[OFFSET_IDX2:%.*]] = mul i64 [[INDEX]], 2 |
| ; CHECK-NEXT: [[NEXT_GEP3:%.*]] = getelementptr i8, ptr [[PDST:%.*]], i64 [[OFFSET_IDX2]] |
| ; CHECK-NEXT: [[TMP1:%.*]] = getelementptr i16, ptr [[NEXT_GEP]], i32 8 |
| ; CHECK-NEXT: [[WIDE_LOAD:%.*]] = load <8 x i16>, ptr [[NEXT_GEP]], align 2 |
| ; CHECK-NEXT: [[WIDE_LOAD4:%.*]] = load <8 x i16>, ptr [[TMP1]], align 2 |
| ; CHECK-NEXT: [[TMP2:%.*]] = call <8 x i16> @llvm.sadd.sat.v8i16(<8 x i16> [[WIDE_LOAD]], <8 x i16> [[BROADCAST_SPLAT]]) |
| ; CHECK-NEXT: [[TMP3:%.*]] = call <8 x i16> @llvm.sadd.sat.v8i16(<8 x i16> [[WIDE_LOAD4]], <8 x i16> [[BROADCAST_SPLAT]]) |
| ; CHECK-NEXT: [[TMP4:%.*]] = getelementptr i16, ptr [[NEXT_GEP3]], i32 8 |
| ; CHECK-NEXT: store <8 x i16> [[TMP2]], ptr [[NEXT_GEP3]], align 2 |
| ; CHECK-NEXT: store <8 x i16> [[TMP3]], ptr [[TMP4]], align 2 |
| ; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 16 |
| ; CHECK-NEXT: [[TMP5:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]] |
| ; CHECK-NEXT: br i1 [[TMP5]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]] |
| ; CHECK: middle.block: |
| ; CHECK-NEXT: [[CMP_N:%.*]] = icmp eq i64 [[TMP0]], [[N_VEC]] |
| ; CHECK-NEXT: br i1 [[CMP_N]], label [[WHILE_END_LOOPEXIT:%.*]], label [[VEC_EPILOG_ITER_CHECK:%.*]] |
| ; CHECK: vec.epilog.iter.check: |
| ; CHECK-NEXT: [[DOTCAST1:%.*]] = trunc i64 [[N_VEC]] to i32 |
| ; CHECK-NEXT: [[IND_END8:%.*]] = sub i32 [[BLOCKSIZE]], [[DOTCAST1]] |
| ; CHECK-NEXT: [[TMP6:%.*]] = mul i64 [[N_VEC]], 2 |
| ; CHECK-NEXT: [[IND_END10:%.*]] = getelementptr i8, ptr [[PSRC]], i64 [[TMP6]] |
| ; CHECK-NEXT: [[TMP7:%.*]] = mul i64 [[N_VEC]], 2 |
| ; CHECK-NEXT: [[IND_END13:%.*]] = getelementptr i8, ptr [[PDST]], i64 [[TMP7]] |
| ; CHECK-NEXT: [[MIN_EPILOG_ITERS_CHECK:%.*]] = icmp ult i64 [[N_MOD_VF]], 4 |
| ; CHECK-NEXT: br i1 [[MIN_EPILOG_ITERS_CHECK]], label [[VEC_EPILOG_SCALAR_PH]], label [[VEC_EPILOG_PH]], !prof [[PROF3:![0-9]+]] |
| ; 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: [[N_MOD_VF4:%.*]] = urem i64 [[TMP0]], 4 |
| ; CHECK-NEXT: [[N_VEC6:%.*]] = sub i64 [[TMP0]], [[N_MOD_VF4]] |
| ; CHECK-NEXT: [[DOTCAST:%.*]] = trunc i64 [[N_VEC6]] to i32 |
| ; CHECK-NEXT: [[IND_END:%.*]] = sub i32 [[BLOCKSIZE]], [[DOTCAST]] |
| ; CHECK-NEXT: [[TMP8:%.*]] = mul i64 [[N_VEC6]], 2 |
| ; CHECK-NEXT: [[IND_END9:%.*]] = getelementptr i8, ptr [[PSRC]], i64 [[TMP8]] |
| ; CHECK-NEXT: [[TMP9:%.*]] = mul i64 [[N_VEC6]], 2 |
| ; CHECK-NEXT: [[IND_END12:%.*]] = getelementptr i8, ptr [[PDST]], i64 [[TMP9]] |
| ; CHECK-NEXT: [[BROADCAST_SPLATINSERT21:%.*]] = insertelement <4 x i16> poison, i16 [[OFFSET]], i64 0 |
| ; CHECK-NEXT: [[BROADCAST_SPLAT22:%.*]] = shufflevector <4 x i16> [[BROADCAST_SPLATINSERT21]], <4 x i16> poison, <4 x i32> zeroinitializer |
| ; CHECK-NEXT: br label [[VEC_EPILOG_VECTOR_BODY:%.*]] |
| ; CHECK: vec.epilog.vector.body: |
| ; CHECK-NEXT: [[INDEX15:%.*]] = phi i64 [ [[VEC_EPILOG_RESUME_VAL]], [[VEC_EPILOG_PH]] ], [ [[INDEX_NEXT23:%.*]], [[VEC_EPILOG_VECTOR_BODY]] ] |
| ; CHECK-NEXT: [[OFFSET_IDX16:%.*]] = mul i64 [[INDEX15]], 2 |
| ; CHECK-NEXT: [[NEXT_GEP17:%.*]] = getelementptr i8, ptr [[PSRC]], i64 [[OFFSET_IDX16]] |
| ; CHECK-NEXT: [[OFFSET_IDX18:%.*]] = mul i64 [[INDEX15]], 2 |
| ; CHECK-NEXT: [[NEXT_GEP19:%.*]] = getelementptr i8, ptr [[PDST]], i64 [[OFFSET_IDX18]] |
| ; CHECK-NEXT: [[WIDE_LOAD20:%.*]] = load <4 x i16>, ptr [[NEXT_GEP17]], align 2 |
| ; CHECK-NEXT: [[TMP10:%.*]] = call <4 x i16> @llvm.sadd.sat.v4i16(<4 x i16> [[WIDE_LOAD20]], <4 x i16> [[BROADCAST_SPLAT22]]) |
| ; CHECK-NEXT: store <4 x i16> [[TMP10]], ptr [[NEXT_GEP19]], align 2 |
| ; CHECK-NEXT: [[INDEX_NEXT23]] = add nuw i64 [[INDEX15]], 4 |
| ; CHECK-NEXT: [[TMP11:%.*]] = icmp eq i64 [[INDEX_NEXT23]], [[N_VEC6]] |
| ; CHECK-NEXT: br i1 [[TMP11]], label [[VEC_EPILOG_MIDDLE_BLOCK:%.*]], label [[VEC_EPILOG_VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]] |
| ; CHECK: vec.epilog.middle.block: |
| ; CHECK-NEXT: [[CMP_N14:%.*]] = icmp eq i64 [[TMP0]], [[N_VEC6]] |
| ; CHECK-NEXT: br i1 [[CMP_N14]], label [[WHILE_END_LOOPEXIT]], label [[VEC_EPILOG_SCALAR_PH]] |
| ; CHECK: vec.epilog.scalar.ph: |
| ; CHECK-NEXT: [[BC_RESUME_VAL:%.*]] = phi i32 [ [[IND_END]], [[VEC_EPILOG_MIDDLE_BLOCK]] ], [ [[IND_END8]], [[VEC_EPILOG_ITER_CHECK]] ], [ [[BLOCKSIZE]], [[ITER_CHECK]] ] |
| ; CHECK-NEXT: [[BC_RESUME_VAL11:%.*]] = phi ptr [ [[IND_END9]], [[VEC_EPILOG_MIDDLE_BLOCK]] ], [ [[IND_END10]], [[VEC_EPILOG_ITER_CHECK]] ], [ [[PSRC]], [[ITER_CHECK]] ] |
| ; CHECK-NEXT: [[BC_RESUME_VAL14:%.*]] = phi ptr [ [[IND_END12]], [[VEC_EPILOG_MIDDLE_BLOCK]] ], [ [[IND_END13]], [[VEC_EPILOG_ITER_CHECK]] ], [ [[PDST]], [[ITER_CHECK]] ] |
| ; CHECK-NEXT: br label [[WHILE_BODY:%.*]] |
| ; CHECK: while.body: |
| ; CHECK-NEXT: [[BLKCNT_09:%.*]] = phi i32 [ [[DEC:%.*]], [[WHILE_BODY]] ], [ [[BC_RESUME_VAL]], [[VEC_EPILOG_SCALAR_PH]] ] |
| ; CHECK-NEXT: [[PSRC_ADDR_08:%.*]] = phi ptr [ [[INCDEC_PTR:%.*]], [[WHILE_BODY]] ], [ [[BC_RESUME_VAL11]], [[VEC_EPILOG_SCALAR_PH]] ] |
| ; CHECK-NEXT: [[PDST_ADDR_07:%.*]] = phi ptr [ [[INCDEC_PTR3:%.*]], [[WHILE_BODY]] ], [ [[BC_RESUME_VAL14]], [[VEC_EPILOG_SCALAR_PH]] ] |
| ; CHECK-NEXT: [[INCDEC_PTR]] = getelementptr inbounds i16, ptr [[PSRC_ADDR_08]], i32 1 |
| ; CHECK-NEXT: [[TMP12:%.*]] = load i16, ptr [[PSRC_ADDR_08]], align 2 |
| ; CHECK-NEXT: [[TMP13:%.*]] = tail call i16 @llvm.sadd.sat.i16(i16 [[TMP12]], i16 [[OFFSET]]) |
| ; CHECK-NEXT: [[INCDEC_PTR3]] = getelementptr inbounds i16, ptr [[PDST_ADDR_07]], i32 1 |
| ; CHECK-NEXT: store i16 [[TMP13]], ptr [[PDST_ADDR_07]], align 2 |
| ; CHECK-NEXT: [[DEC]] = add i32 [[BLKCNT_09]], -1 |
| ; CHECK-NEXT: [[CMP_NOT:%.*]] = icmp eq i32 [[DEC]], 0 |
| ; CHECK-NEXT: br i1 [[CMP_NOT]], label [[WHILE_END_LOOPEXIT]], label [[WHILE_BODY]], !llvm.loop [[LOOP5:![0-9]+]] |
| ; CHECK: while.end.loopexit: |
| ; CHECK-NEXT: br label [[WHILE_END]] |
| ; CHECK: while.end: |
| ; CHECK-NEXT: ret void |
| ; |
| entry: |
| %cmp.not6 = icmp eq i32 %blockSize, 0 |
| br i1 %cmp.not6, label %while.end, label %while.body |
| |
| while.body: ; preds = %entry, %while.body |
| %blkCnt.09 = phi i32 [ %dec, %while.body ], [ %blockSize, %entry ] |
| %pSrc.addr.08 = phi ptr [ %incdec.ptr, %while.body ], [ %pSrc, %entry ] |
| %pDst.addr.07 = phi ptr [ %incdec.ptr3, %while.body ], [ %pDst, %entry ] |
| %incdec.ptr = getelementptr inbounds i16, ptr %pSrc.addr.08, i32 1 |
| %0 = load i16, ptr %pSrc.addr.08, align 2 |
| %1 = tail call i16 @llvm.sadd.sat.i16(i16 %0, i16 %offset) |
| %incdec.ptr3 = getelementptr inbounds i16, ptr %pDst.addr.07, i32 1 |
| store i16 %1, ptr %pDst.addr.07, align 2 |
| %dec = add i32 %blkCnt.09, -1 |
| %cmp.not = icmp eq i32 %dec, 0 |
| br i1 %cmp.not, label %while.end, label %while.body |
| |
| while.end: ; preds = %while.body, %entry |
| ret void |
| } |
| |
| ; CHECK-COST-LABEL: umin |
| ; CHECK-COST: Found an estimated cost of 2 for VF 1 For instruction: %1 = tail call i8 @llvm.umin.i8(i8 %0, i8 %offset) |
| ; CHECK-COST: Cost of 1 for VF 2: WIDEN-INTRINSIC ir<%1> = call llvm.umin(ir<%0>, ir<%offset>) |
| ; CHECK-COST: Cost of 1 for VF 4: WIDEN-INTRINSIC ir<%1> = call llvm.umin(ir<%0>, ir<%offset>) |
| ; CHECK-COST: Cost of 1 for VF 8: WIDEN-INTRINSIC ir<%1> = call llvm.umin(ir<%0>, ir<%offset>) |
| ; CHECK-COST: Cost of 1 for VF 16: WIDEN-INTRINSIC ir<%1> = call llvm.umin(ir<%0>, ir<%offset>) |
| |
| |
| define void @umin(ptr nocapture readonly %pSrc, i8 signext %offset, ptr nocapture noalias %pDst, i32 %blockSize) #0 { |
| ; CHECK-LABEL: @umin( |
| ; CHECK-NEXT: entry: |
| ; CHECK-NEXT: [[CMP_NOT6:%.*]] = icmp eq i32 [[BLOCKSIZE:%.*]], 0 |
| ; CHECK-NEXT: br i1 [[CMP_NOT6]], label [[WHILE_END:%.*]], label [[ITER_CHECK:%.*]] |
| ; CHECK: iter.check: |
| ; CHECK-NEXT: [[TMP0:%.*]] = zext i32 [[BLOCKSIZE]] to i64 |
| ; CHECK-NEXT: [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[TMP0]], 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: [[MIN_ITERS_CHECK1:%.*]] = icmp ult i64 [[TMP0]], 32 |
| ; CHECK-NEXT: br i1 [[MIN_ITERS_CHECK1]], label [[VEC_EPILOG_PH:%.*]], label [[VECTOR_PH:%.*]] |
| ; CHECK: vector.ph: |
| ; CHECK-NEXT: [[N_MOD_VF:%.*]] = urem i64 [[TMP0]], 32 |
| ; CHECK-NEXT: [[N_VEC:%.*]] = sub i64 [[TMP0]], [[N_MOD_VF]] |
| ; CHECK-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <16 x i8> poison, i8 [[OFFSET:%.*]], i64 0 |
| ; CHECK-NEXT: [[BROADCAST_SPLAT:%.*]] = shufflevector <16 x i8> [[BROADCAST_SPLATINSERT]], <16 x i8> poison, <16 x i32> zeroinitializer |
| ; CHECK-NEXT: br label [[VECTOR_BODY:%.*]] |
| ; CHECK: vector.body: |
| ; CHECK-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[VECTOR_BODY]] ] |
| ; CHECK-NEXT: [[NEXT_GEP:%.*]] = getelementptr i8, ptr [[PSRC:%.*]], i64 [[INDEX]] |
| ; CHECK-NEXT: [[NEXT_GEP2:%.*]] = getelementptr i8, ptr [[PDST:%.*]], i64 [[INDEX]] |
| ; CHECK-NEXT: [[TMP1:%.*]] = getelementptr i8, ptr [[NEXT_GEP]], i32 16 |
| ; CHECK-NEXT: [[WIDE_LOAD:%.*]] = load <16 x i8>, ptr [[NEXT_GEP]], align 2 |
| ; CHECK-NEXT: [[WIDE_LOAD3:%.*]] = load <16 x i8>, ptr [[TMP1]], align 2 |
| ; CHECK-NEXT: [[TMP2:%.*]] = call <16 x i8> @llvm.umin.v16i8(<16 x i8> [[WIDE_LOAD]], <16 x i8> [[BROADCAST_SPLAT]]) |
| ; CHECK-NEXT: [[TMP3:%.*]] = call <16 x i8> @llvm.umin.v16i8(<16 x i8> [[WIDE_LOAD3]], <16 x i8> [[BROADCAST_SPLAT]]) |
| ; CHECK-NEXT: [[TMP4:%.*]] = getelementptr i8, ptr [[NEXT_GEP2]], i32 16 |
| ; CHECK-NEXT: store <16 x i8> [[TMP2]], ptr [[NEXT_GEP2]], align 2 |
| ; CHECK-NEXT: store <16 x i8> [[TMP3]], ptr [[TMP4]], align 2 |
| ; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 32 |
| ; CHECK-NEXT: [[TMP5:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]] |
| ; CHECK-NEXT: br i1 [[TMP5]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP6:![0-9]+]] |
| ; CHECK: middle.block: |
| ; CHECK-NEXT: [[CMP_N:%.*]] = icmp eq i64 [[TMP0]], [[N_VEC]] |
| ; CHECK-NEXT: br i1 [[CMP_N]], label [[WHILE_END_LOOPEXIT:%.*]], label [[VEC_EPILOG_ITER_CHECK:%.*]] |
| ; CHECK: vec.epilog.iter.check: |
| ; CHECK-NEXT: [[DOTCAST6:%.*]] = trunc i64 [[N_VEC]] to i32 |
| ; CHECK-NEXT: [[IND_END7:%.*]] = sub i32 [[BLOCKSIZE]], [[DOTCAST6]] |
| ; CHECK-NEXT: [[IND_END9:%.*]] = getelementptr i8, ptr [[PSRC]], i64 [[N_VEC]] |
| ; CHECK-NEXT: [[IND_END12:%.*]] = getelementptr i8, ptr [[PDST]], i64 [[N_VEC]] |
| ; CHECK-NEXT: [[MIN_EPILOG_ITERS_CHECK:%.*]] = icmp ult i64 [[N_MOD_VF]], 8 |
| ; CHECK-NEXT: br i1 [[MIN_EPILOG_ITERS_CHECK]], label [[VEC_EPILOG_SCALAR_PH]], label [[VEC_EPILOG_PH]], !prof [[PROF7:![0-9]+]] |
| ; 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: [[N_MOD_VF4:%.*]] = urem i64 [[TMP0]], 8 |
| ; CHECK-NEXT: [[N_VEC5:%.*]] = sub i64 [[TMP0]], [[N_MOD_VF4]] |
| ; CHECK-NEXT: [[DOTCAST:%.*]] = trunc i64 [[N_VEC5]] to i32 |
| ; CHECK-NEXT: [[IND_END:%.*]] = sub i32 [[BLOCKSIZE]], [[DOTCAST]] |
| ; CHECK-NEXT: [[IND_END8:%.*]] = getelementptr i8, ptr [[PSRC]], i64 [[N_VEC5]] |
| ; CHECK-NEXT: [[IND_END11:%.*]] = getelementptr i8, ptr [[PDST]], i64 [[N_VEC5]] |
| ; CHECK-NEXT: [[BROADCAST_SPLATINSERT18:%.*]] = insertelement <8 x i8> poison, i8 [[OFFSET]], i64 0 |
| ; CHECK-NEXT: [[BROADCAST_SPLAT19:%.*]] = shufflevector <8 x i8> [[BROADCAST_SPLATINSERT18]], <8 x i8> poison, <8 x i32> zeroinitializer |
| ; CHECK-NEXT: br label [[VEC_EPILOG_VECTOR_BODY:%.*]] |
| ; CHECK: vec.epilog.vector.body: |
| ; CHECK-NEXT: [[INDEX14:%.*]] = phi i64 [ [[VEC_EPILOG_RESUME_VAL]], [[VEC_EPILOG_PH]] ], [ [[INDEX_NEXT20:%.*]], [[VEC_EPILOG_VECTOR_BODY]] ] |
| ; CHECK-NEXT: [[NEXT_GEP15:%.*]] = getelementptr i8, ptr [[PSRC]], i64 [[INDEX14]] |
| ; CHECK-NEXT: [[NEXT_GEP16:%.*]] = getelementptr i8, ptr [[PDST]], i64 [[INDEX14]] |
| ; CHECK-NEXT: [[WIDE_LOAD17:%.*]] = load <8 x i8>, ptr [[NEXT_GEP15]], align 2 |
| ; CHECK-NEXT: [[TMP6:%.*]] = call <8 x i8> @llvm.umin.v8i8(<8 x i8> [[WIDE_LOAD17]], <8 x i8> [[BROADCAST_SPLAT19]]) |
| ; CHECK-NEXT: store <8 x i8> [[TMP6]], ptr [[NEXT_GEP16]], align 2 |
| ; CHECK-NEXT: [[INDEX_NEXT20]] = add nuw i64 [[INDEX14]], 8 |
| ; CHECK-NEXT: [[TMP7:%.*]] = icmp eq i64 [[INDEX_NEXT20]], [[N_VEC5]] |
| ; CHECK-NEXT: br i1 [[TMP7]], label [[VEC_EPILOG_MIDDLE_BLOCK:%.*]], label [[VEC_EPILOG_VECTOR_BODY]], !llvm.loop [[LOOP8:![0-9]+]] |
| ; CHECK: vec.epilog.middle.block: |
| ; CHECK-NEXT: [[CMP_N13:%.*]] = icmp eq i64 [[TMP0]], [[N_VEC5]] |
| ; CHECK-NEXT: br i1 [[CMP_N13]], label [[WHILE_END_LOOPEXIT]], label [[VEC_EPILOG_SCALAR_PH]] |
| ; CHECK: vec.epilog.scalar.ph: |
| ; CHECK-NEXT: [[BC_RESUME_VAL:%.*]] = phi i32 [ [[IND_END]], [[VEC_EPILOG_MIDDLE_BLOCK]] ], [ [[IND_END7]], [[VEC_EPILOG_ITER_CHECK]] ], [ [[BLOCKSIZE]], [[ITER_CHECK]] ] |
| ; CHECK-NEXT: [[BC_RESUME_VAL10:%.*]] = phi ptr [ [[IND_END8]], [[VEC_EPILOG_MIDDLE_BLOCK]] ], [ [[IND_END9]], [[VEC_EPILOG_ITER_CHECK]] ], [ [[PSRC]], [[ITER_CHECK]] ] |
| ; CHECK-NEXT: [[BC_RESUME_VAL13:%.*]] = phi ptr [ [[IND_END11]], [[VEC_EPILOG_MIDDLE_BLOCK]] ], [ [[IND_END12]], [[VEC_EPILOG_ITER_CHECK]] ], [ [[PDST]], [[ITER_CHECK]] ] |
| ; CHECK-NEXT: br label [[WHILE_BODY:%.*]] |
| ; CHECK: while.body: |
| ; CHECK-NEXT: [[BLKCNT_09:%.*]] = phi i32 [ [[DEC:%.*]], [[WHILE_BODY]] ], [ [[BC_RESUME_VAL]], [[VEC_EPILOG_SCALAR_PH]] ] |
| ; CHECK-NEXT: [[PSRC_ADDR_08:%.*]] = phi ptr [ [[INCDEC_PTR:%.*]], [[WHILE_BODY]] ], [ [[BC_RESUME_VAL10]], [[VEC_EPILOG_SCALAR_PH]] ] |
| ; CHECK-NEXT: [[PDST_ADDR_07:%.*]] = phi ptr [ [[INCDEC_PTR3:%.*]], [[WHILE_BODY]] ], [ [[BC_RESUME_VAL13]], [[VEC_EPILOG_SCALAR_PH]] ] |
| ; CHECK-NEXT: [[INCDEC_PTR]] = getelementptr inbounds i8, ptr [[PSRC_ADDR_08]], i32 1 |
| ; CHECK-NEXT: [[TMP8:%.*]] = load i8, ptr [[PSRC_ADDR_08]], align 2 |
| ; CHECK-NEXT: [[TMP9:%.*]] = tail call i8 @llvm.umin.i8(i8 [[TMP8]], i8 [[OFFSET]]) |
| ; CHECK-NEXT: [[INCDEC_PTR3]] = getelementptr inbounds i8, ptr [[PDST_ADDR_07]], i32 1 |
| ; CHECK-NEXT: store i8 [[TMP9]], ptr [[PDST_ADDR_07]], align 2 |
| ; CHECK-NEXT: [[DEC]] = add i32 [[BLKCNT_09]], -1 |
| ; CHECK-NEXT: [[CMP_NOT:%.*]] = icmp eq i32 [[DEC]], 0 |
| ; CHECK-NEXT: br i1 [[CMP_NOT]], label [[WHILE_END_LOOPEXIT]], label [[WHILE_BODY]], !llvm.loop [[LOOP9:![0-9]+]] |
| ; CHECK: while.end.loopexit: |
| ; CHECK-NEXT: br label [[WHILE_END]] |
| ; CHECK: while.end: |
| ; CHECK-NEXT: ret void |
| ; |
| entry: |
| %cmp.not6 = icmp eq i32 %blockSize, 0 |
| br i1 %cmp.not6, label %while.end, label %while.body |
| |
| while.body: ; preds = %entry, %while.body |
| %blkCnt.09 = phi i32 [ %dec, %while.body ], [ %blockSize, %entry ] |
| %pSrc.addr.08 = phi ptr [ %incdec.ptr, %while.body ], [ %pSrc, %entry ] |
| %pDst.addr.07 = phi ptr [ %incdec.ptr3, %while.body ], [ %pDst, %entry ] |
| %incdec.ptr = getelementptr inbounds i8, ptr %pSrc.addr.08, i32 1 |
| %0 = load i8, ptr %pSrc.addr.08, align 2 |
| %1 = tail call i8 @llvm.umin.i8(i8 %0, i8 %offset) |
| %incdec.ptr3 = getelementptr inbounds i8, ptr %pDst.addr.07, i32 1 |
| store i8 %1, ptr %pDst.addr.07, align 2 |
| %dec = add i32 %blkCnt.09, -1 |
| %cmp.not = icmp eq i32 %dec, 0 |
| br i1 %cmp.not, label %while.end, label %while.body |
| |
| while.end: ; preds = %while.body, %entry |
| ret void |
| } |
| |
| declare i16 @llvm.sadd.sat.i16(i16, i16) |
| declare i8 @llvm.umin.i8(i8, i8) |
| |
| ;; NOTE: These prefixes are unused and the list is autogenerated. Do not add tests below this line: |
| ; CHECK-COST: {{.*}} |
