Doomhowl-Interactive/llvm-project
0
0
Fork 0
Code Issues Pull Requests Actions 5 Packages Projects Releases Wiki Activity
Files
81f445b7df94de485240b21034682f22432729c4
llvm-project/llvm/test/Transforms/LoopVectorize/tail-folding-vectorization-factor-1.ll
Mel Chen 8c6658aca6 [VPlan] Sink recipes from the vector loop region in licm. (#168031) 
When a recipe can be safely sunk and all of its users are outside the
vector loop region in the same dedicated exit block, the recipe does not
need to be executed on every iteration.
This patch extends the VPlan-based LICM (Loop Invariant Code Motion) to
also sink such recipes from the vector loop region into the exit block.
This reduces redundant computation and improves cost model accuracy.

TODO: Support nested loop sinking
TODO: Support sinking `VPReplicateRecipe` (requires `replicateByVF`
fixes)
TODO: Support recipes with multiple defined values (e.g., interleaved
loads)
TODO: Clone recipes without users to all exit blocks
TODO: Support PHI node users by checking incoming value blocks
TODO: Support sinking when users are in multiple blocks
TODO: Clone recipes when users are on multiple exit paths

Co-authored-by: Luke Lau <luke@igalia.com>

---------

Co-authored-by: Luke Lau <luke@igalia.com>
Co-authored-by: Luke Lau <luke_lau@icloud.com>
2026-02-03 07:57:15 +00:00

228 lines
12 KiB
LLVM
Raw Blame History

; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt < %s -passes=loop-vectorize -force-vector-interleave=4 -prefer-predicate-over-epilogue=predicate-else-scalar-epilogue -pass-remarks='loop-vectorize' -disable-output -S 2>&1 | FileCheck %s --check-prefix=CHECK-REMARKS
; RUN: opt < %s -passes=loop-vectorize -force-vector-interleave=4 -prefer-predicate-over-epilogue=predicate-else-scalar-epilogue -S | FileCheck %s
; These tests are to check that fold-tail procedure produces correct scalar code when
; loop-vectorization is only unrolling but not vectorizing.
; CHECK-REMARKS: remark: {{.*}} interleaved loop (interleaved count: 4)
; CHECK-REMARKS-NEXT: remark: {{.*}} interleaved loop (interleaved count: 4)
; CHECK-REMARKS-NOT: remark: {{.*}} vectorized loop
define void @VF1-VPlanExe(ptr %dst) {
; CHECK-LABEL: @VF1-VPlanExe(
; CHECK-NEXT: entry:
; CHECK-NEXT: br label [[VECTOR_PH:%.*]]
; CHECK: vector.ph:
; CHECK-NEXT: br label [[VECTOR_BODY:%.*]]
; CHECK: vector.body:
; CHECK-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[PRED_STORE_CONTINUE6:%.*]] ]
; CHECK-NEXT: [[TMP0:%.*]] = add i64 [[INDEX]], 0
; CHECK-NEXT: [[TMP1:%.*]] = add i64 [[INDEX]], 1
; CHECK-NEXT: [[TMP2:%.*]] = add i64 [[INDEX]], 2
; CHECK-NEXT: [[TMP3:%.*]] = add i64 [[INDEX]], 3
; CHECK-NEXT: [[TMP4:%.*]] = icmp ule i64 [[TMP0]], 14
; CHECK-NEXT: [[TMP5:%.*]] = icmp ule i64 [[TMP1]], 14
; CHECK-NEXT: [[TMP6:%.*]] = icmp ule i64 [[TMP2]], 14
; CHECK-NEXT: [[TMP7:%.*]] = icmp ule i64 [[TMP3]], 14
; CHECK-NEXT: br i1 [[TMP4]], label [[PRED_STORE_IF:%.*]], label [[PRED_STORE_CONTINUE:%.*]]
; CHECK: pred.store.if:
; CHECK-NEXT: [[TMP8:%.*]] = getelementptr inbounds i32, ptr [[DST:%.*]], i64 [[TMP0]]
; CHECK-NEXT: store i32 0, ptr [[TMP8]], align 4
; CHECK-NEXT: br label [[PRED_STORE_CONTINUE]]
; CHECK: pred.store.continue:
; CHECK-NEXT: br i1 [[TMP5]], label [[PRED_STORE_IF1:%.*]], label [[PRED_STORE_CONTINUE2:%.*]]
; CHECK: pred.store.if1:
; CHECK-NEXT: [[TMP9:%.*]] = getelementptr inbounds i32, ptr [[DST]], i64 [[TMP1]]
; CHECK-NEXT: store i32 0, ptr [[TMP9]], align 4
; CHECK-NEXT: br label [[PRED_STORE_CONTINUE2]]
; CHECK: pred.store.continue2:
; CHECK-NEXT: br i1 [[TMP6]], label [[PRED_STORE_IF3:%.*]], label [[PRED_STORE_CONTINUE4:%.*]]
; CHECK: pred.store.if3:
; CHECK-NEXT: [[TMP10:%.*]] = getelementptr inbounds i32, ptr [[DST]], i64 [[TMP2]]
; CHECK-NEXT: store i32 0, ptr [[TMP10]], align 4
; CHECK-NEXT: br label [[PRED_STORE_CONTINUE4]]
; CHECK: pred.store.continue4:
; CHECK-NEXT: br i1 [[TMP7]], label [[PRED_STORE_IF5:%.*]], label [[PRED_STORE_CONTINUE6]]
; CHECK: pred.store.if5:
; CHECK-NEXT: [[TMP11:%.*]] = getelementptr inbounds i32, ptr [[DST]], i64 [[TMP3]]
; CHECK-NEXT: store i32 0, ptr [[TMP11]], align 4
; CHECK-NEXT: br label [[PRED_STORE_CONTINUE6]]
; CHECK: pred.store.continue6:
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
; CHECK-NEXT: [[TMP12:%.*]] = icmp eq i64 [[INDEX_NEXT]], 16
; CHECK-NEXT: br i1 [[TMP12]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; CHECK: middle.block:
; CHECK-NEXT: br label [[FOR_BODY:%.*]]
; CHECK: for.cond.cleanup:
; CHECK-NEXT: ret void
;
entry:
br label %for.body
for.cond.cleanup:
ret void
for.body:
%indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
%dst.ptr = getelementptr inbounds i32, ptr %dst, i64 %indvars.iv
store i32 0, ptr %dst.ptr
%indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
%exitcond = icmp eq i64 %indvars.iv.next, 15
br i1 %exitcond, label %for.cond.cleanup, label %for.body
}
; Note: Most scalar pointer induction GEPs could be sunk into the conditional
; blocks.
define void @VF1-VPWidenCanonicalIVRecipeExe(ptr %ptr1) {
; CHECK-LABEL: @VF1-VPWidenCanonicalIVRecipeExe(
; CHECK-NEXT: entry:
; CHECK-NEXT: [[PTR2:%.*]] = getelementptr inbounds double, ptr [[PTR1:%.*]], i64 15
; CHECK-NEXT: br label [[VECTOR_PH:%.*]]
; CHECK: vector.ph:
; CHECK-NEXT: br label [[VECTOR_BODY:%.*]]
; CHECK: vector.body:
; CHECK-NEXT: [[INDEX:%.*]] = phi i64 [ 0, [[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], [[PRED_STORE_CONTINUE12:%.*]] ]
; CHECK-NEXT: [[OFFSET_IDX:%.*]] = mul i64 [[INDEX]], 8
; CHECK-NEXT: [[TMP4:%.*]] = add i64 [[OFFSET_IDX]], 8
; CHECK-NEXT: [[TMP5:%.*]] = add i64 [[OFFSET_IDX]], 16
; CHECK-NEXT: [[TMP6:%.*]] = add i64 [[OFFSET_IDX]], 24
; CHECK-NEXT: [[NEXT_GEP:%.*]] = getelementptr i8, ptr [[PTR1]], i64 [[OFFSET_IDX]]
; CHECK-NEXT: [[NEXT_GEP1:%.*]] = getelementptr i8, ptr [[PTR1]], i64 [[TMP4]]
; CHECK-NEXT: [[NEXT_GEP2:%.*]] = getelementptr i8, ptr [[PTR1]], i64 [[TMP5]]
; CHECK-NEXT: [[NEXT_GEP3:%.*]] = getelementptr i8, ptr [[PTR1]], i64 [[TMP6]]
; CHECK-NEXT: [[VEC_IV:%.*]] = add i64 [[INDEX]], 0
; CHECK-NEXT: [[VEC_IV4:%.*]] = add i64 [[INDEX]], 1
; CHECK-NEXT: [[VEC_IV5:%.*]] = add i64 [[INDEX]], 2
; CHECK-NEXT: [[VEC_IV6:%.*]] = add i64 [[INDEX]], 3
; CHECK-NEXT: [[TMP0:%.*]] = icmp ule i64 [[VEC_IV]], 14
; CHECK-NEXT: [[TMP1:%.*]] = icmp ule i64 [[VEC_IV4]], 14
; CHECK-NEXT: [[TMP2:%.*]] = icmp ule i64 [[VEC_IV5]], 14
; CHECK-NEXT: [[TMP3:%.*]] = icmp ule i64 [[VEC_IV6]], 14
; CHECK-NEXT: br i1 [[TMP0]], label [[PRED_STORE_IF:%.*]], label [[PRED_STORE_CONTINUE:%.*]]
; CHECK: pred.store.if:
; CHECK-NEXT: store double 0.000000e+00, ptr [[NEXT_GEP]], align 8
; CHECK-NEXT: br label [[PRED_STORE_CONTINUE]]
; CHECK: pred.store.continue:
; CHECK-NEXT: br i1 [[TMP1]], label [[PRED_STORE_IF7:%.*]], label [[PRED_STORE_CONTINUE8:%.*]]
; CHECK: pred.store.if7:
; CHECK-NEXT: store double 0.000000e+00, ptr [[NEXT_GEP1]], align 8
; CHECK-NEXT: br label [[PRED_STORE_CONTINUE8]]
; CHECK: pred.store.continue8:
; CHECK-NEXT: br i1 [[TMP2]], label [[PRED_STORE_IF9:%.*]], label [[PRED_STORE_CONTINUE10:%.*]]
; CHECK: pred.store.if9:
; CHECK-NEXT: store double 0.000000e+00, ptr [[NEXT_GEP2]], align 8
; CHECK-NEXT: br label [[PRED_STORE_CONTINUE10]]
; CHECK: pred.store.continue10:
; CHECK-NEXT: br i1 [[TMP3]], label [[PRED_STORE_IF11:%.*]], label [[PRED_STORE_CONTINUE12]]
; CHECK: pred.store.if11:
; CHECK-NEXT: store double 0.000000e+00, ptr [[NEXT_GEP3]], align 8
; CHECK-NEXT: br label [[PRED_STORE_CONTINUE12]]
; CHECK: pred.store.continue12:
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 4
; CHECK-NEXT: [[TMP8:%.*]] = icmp eq i64 [[INDEX_NEXT]], 16
; CHECK-NEXT: br i1 [[TMP8]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP3:![0-9]+]]
; CHECK: middle.block:
; CHECK-NEXT: br label [[FOR_BODY:%.*]]
; CHECK: for.cond.cleanup:
; CHECK-NEXT: ret void
;
entry:
%ptr2 = getelementptr inbounds double, ptr %ptr1, i64 15
br label %for.body
for.cond.cleanup:
ret void
for.body:
%addr = phi ptr [ %ptr, %for.body ], [ %ptr1, %entry ]
store double 0.0, ptr %addr
%ptr = getelementptr inbounds double, ptr %addr, i64 1
%cond = icmp eq ptr %ptr, %ptr2
br i1 %cond, label %for.cond.cleanup, label %for.body
}
define i64 @live_out_scalar_vf(i64 %n) {
; CHECK-LABEL: @live_out_scalar_vf(
; CHECK-NEXT: entry:
; CHECK-NEXT: [[TMP0:%.*]] = add i64 [[N:%.*]], 1
; CHECK-NEXT: br label [[VECTOR_PH:%.*]]
; CHECK: vector.ph:
; CHECK-NEXT: [[N_RND_UP:%.*]] = add i64 [[TMP0]], 15
; CHECK-NEXT: [[N_MOD_VF:%.*]] = urem i64 [[N_RND_UP]], 16
; CHECK-NEXT: [[N_VEC:%.*]] = sub i64 [[N_RND_UP]], [[N_MOD_VF]]
; CHECK-NEXT: [[TRIP_COUNT_MINUS_1:%.*]] = sub i64 [[TMP0]], 1
; CHECK-NEXT: [[BROADCAST_SPLATINSERT:%.*]] = insertelement <4 x i64> poison, i64 [[TRIP_COUNT_MINUS_1]], i64 0
; CHECK-NEXT: [[BROADCAST_SPLAT:%.*]] = shufflevector <4 x i64> [[BROADCAST_SPLATINSERT]], <4 x i64> poison, <4 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: [[VEC_IND:%.*]] = phi <4 x i64> [ <i64 0, i64 1, i64 2, i64 3>, [[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT: [[VECTOR_RECUR:%.*]] = phi <4 x i64> [ <i64 poison, i64 poison, i64 poison, i64 0>, [[VECTOR_PH]] ], [ [[STEP_ADD_3:%.*]], [[VECTOR_BODY]] ]
; CHECK-NEXT: [[STEP_ADD:%.*]] = add <4 x i64> [[VEC_IND]], splat (i64 4)
; CHECK-NEXT: [[STEP_ADD_2:%.*]] = add <4 x i64> [[STEP_ADD]], splat (i64 4)
; CHECK-NEXT: [[STEP_ADD_3]] = add <4 x i64> [[STEP_ADD_2]], splat (i64 4)
; CHECK-NEXT: [[INDEX_NEXT]] = add i64 [[INDEX]], 16
; CHECK-NEXT: [[VEC_IND_NEXT]] = add <4 x i64> [[STEP_ADD_3]], splat (i64 4)
; CHECK-NEXT: [[TMP5:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT: br i1 [[TMP5]], label [[MIDDLE_BLOCK:%.*]], label [[VECTOR_BODY]], !llvm.loop [[LOOP4:![0-9]+]]
; CHECK: middle.block:
; CHECK-NEXT: [[TMP1:%.*]] = icmp ugt <4 x i64> [[VEC_IND]], [[BROADCAST_SPLAT]]
; CHECK-NEXT: [[TMP2:%.*]] = icmp ugt <4 x i64> [[STEP_ADD]], [[BROADCAST_SPLAT]]
; CHECK-NEXT: [[TMP3:%.*]] = icmp ugt <4 x i64> [[STEP_ADD_2]], [[BROADCAST_SPLAT]]
; CHECK-NEXT: [[TMP4:%.*]] = icmp ugt <4 x i64> [[STEP_ADD_3]], [[BROADCAST_SPLAT]]
; CHECK-NEXT: [[FIRST_INACTIVE_LANE:%.*]] = call i64 @llvm.experimental.cttz.elts.i64.v4i1(<4 x i1> [[TMP4]], i1 false)
; CHECK-NEXT: [[TMP6:%.*]] = add i64 12, [[FIRST_INACTIVE_LANE]]
; CHECK-NEXT: [[FIRST_INACTIVE_LANE1:%.*]] = call i64 @llvm.experimental.cttz.elts.i64.v4i1(<4 x i1> [[TMP3]], i1 false)
; CHECK-NEXT: [[TMP7:%.*]] = add i64 8, [[FIRST_INACTIVE_LANE1]]
; CHECK-NEXT: [[TMP8:%.*]] = icmp ne i64 [[FIRST_INACTIVE_LANE1]], 4
; CHECK-NEXT: [[TMP9:%.*]] = select i1 [[TMP8]], i64 [[TMP7]], i64 [[TMP6]]
; CHECK-NEXT: [[FIRST_INACTIVE_LANE2:%.*]] = call i64 @llvm.experimental.cttz.elts.i64.v4i1(<4 x i1> [[TMP2]], i1 false)
; CHECK-NEXT: [[TMP10:%.*]] = add i64 4, [[FIRST_INACTIVE_LANE2]]
; CHECK-NEXT: [[TMP11:%.*]] = icmp ne i64 [[FIRST_INACTIVE_LANE2]], 4
; CHECK-NEXT: [[TMP12:%.*]] = select i1 [[TMP11]], i64 [[TMP10]], i64 [[TMP9]]
; CHECK-NEXT: [[FIRST_INACTIVE_LANE3:%.*]] = call i64 @llvm.experimental.cttz.elts.i64.v4i1(<4 x i1> [[TMP1]], i1 false)
; CHECK-NEXT: [[TMP13:%.*]] = add i64 0, [[FIRST_INACTIVE_LANE3]]
; CHECK-NEXT: [[TMP14:%.*]] = icmp ne i64 [[FIRST_INACTIVE_LANE3]], 4
; CHECK-NEXT: [[TMP15:%.*]] = select i1 [[TMP14]], i64 [[TMP13]], i64 [[TMP12]]
; CHECK-NEXT: [[LAST_ACTIVE_LANE:%.*]] = sub i64 [[TMP15]], 1
; CHECK-NEXT: [[TMP16:%.*]] = sub i64 [[LAST_ACTIVE_LANE]], 1
; CHECK-NEXT: [[TMP17:%.*]] = extractelement <4 x i64> [[VEC_IND]], i64 [[TMP16]]
; CHECK-NEXT: [[TMP18:%.*]] = sub i64 [[TMP16]], 4
; CHECK-NEXT: [[TMP32:%.*]] = extractelement <4 x i64> [[STEP_ADD]], i64 [[TMP18]]
; CHECK-NEXT: [[TMP20:%.*]] = icmp uge i64 [[TMP16]], 4
; CHECK-NEXT: [[TMP21:%.*]] = select i1 [[TMP20]], i64 [[TMP32]], i64 [[TMP17]]
; CHECK-NEXT: [[TMP22:%.*]] = sub i64 [[TMP16]], 8
; CHECK-NEXT: [[TMP23:%.*]] = extractelement <4 x i64> [[STEP_ADD_2]], i64 [[TMP22]]
; CHECK-NEXT: [[TMP24:%.*]] = icmp uge i64 [[TMP16]], 8
; CHECK-NEXT: [[TMP25:%.*]] = select i1 [[TMP24]], i64 [[TMP23]], i64 [[TMP21]]
; CHECK-NEXT: [[TMP26:%.*]] = sub i64 [[TMP16]], 12
; CHECK-NEXT: [[TMP27:%.*]] = extractelement <4 x i64> [[STEP_ADD_3]], i64 [[TMP26]]
; CHECK-NEXT: [[TMP28:%.*]] = icmp uge i64 [[TMP16]], 12
; CHECK-NEXT: [[TMP29:%.*]] = select i1 [[TMP28]], i64 [[TMP27]], i64 [[TMP25]]
; CHECK-NEXT: [[TMP30:%.*]] = extractelement <4 x i64> [[VECTOR_RECUR]], i32 3
; CHECK-NEXT: [[TMP31:%.*]] = icmp eq i64 [[LAST_ACTIVE_LANE]], 0
; CHECK-NEXT: [[TMP19:%.*]] = select i1 [[TMP31]], i64 [[TMP30]], i64 [[TMP29]]
; CHECK-NEXT: br label [[LOOP:%.*]]
; CHECK: exit:
; CHECK-NEXT: ret i64 [[TMP19]]
;
entry:
br label %loop
loop:
%iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
; Need to use a phi otherwise the header mask will use a
; VPWidenCanonicalIVRecipe instead of a VPScalarIVStepsRecipe.
%exitval = phi i64 [ 0, %entry ], [ %iv, %loop ]
%iv.next = add i64 %iv, 1
%ec = icmp eq i64 %iv, %n
br i1 %ec, label %exit, label %loop
exit:
ret i64 %exitval
}
;; NOTE: These prefixes are unused and the list is autogenerated. Do not add tests below this line:
; CHECK-REMARKS: {{.*}}
Reference in New Issue View Git Blame Copy Permalink