Doomhowl-Interactive/llvm-project
0
0
Fork 0
Code Issues Pull Requests Actions 5 Packages Projects Releases Wiki Activity
Files
main
llvm-project/llvm/test/Transforms/SLPVectorizer/X86
History
Alexey Bataev 18053e3c6a [SLP]Bail out on non-schedulable expanded binop with multi-use operand 
In tryScheduleBundle's DoesNotRequireScheduling path, an expanded binop
(shl X, 1 modeled as add X, X) doubles the dependency count of the
duplicated operand. Unlike the regular scheduling path, this branch does
not clear stale operand ScheduleData dependencies, so when that operand
has more uses than this bundle member the schedule's decrement count
exceeds calculateDependencies' increment count and UnscheduledDeps goes
negative, hitting "Expected valid number of unscheduled deps".

Fixes #194303.

Reviewers: 

Pull Request: https://github.com/llvm/llvm-project/pull/194420
2026-04-27 13:40:11 -04:00
..
aggregate.ll
…
align.ll
…
alternate-calls-inseltpoison.ll
…
alternate-calls.ll
…
alternate-cast-inseltpoison.ll
…
alternate-cast.ll
…
alternate-cmp-swapped-pred.ll
…
alternate-fp-inseltpoison.ll
…
alternate-fp.ll
…
alternate-int-inseltpoison.ll
[SLP]Exclude non-profitable subtrees.
2025-12-30 09:32:05 -08:00
alternate-int.ll
[SLP]Exclude non-profitable subtrees.
2025-12-30 09:32:05 -08:00
alternate-op-constant.ll
[SLP]Check if the value has uselist before asking for uses
2025-12-25 10:10:58 -08:00
alternate-opcode-strict-bitwidth-than-main.ll
…
arith-abs.ll
…
arith-add-load.ll
…
arith-add-saddo.ll
…
arith-add-ssat.ll
…
arith-add-uaddo.ll
…
arith-add-usat.ll
…
arith-add.ll
…
arith-and-const-load.ll
…
arith-div.ll
…
arith-fix.ll
…
arith-fp-call.ll
…
arith-fp-inseltpoison.ll
[SLP]Exclude non-profitable subtrees.
2025-12-30 09:32:05 -08:00
arith-fp.ll
[SLP]Exclude non-profitable subtrees.
2025-12-30 09:32:05 -08:00
arith-fshl-rot.ll
…
arith-fshl.ll
…
arith-fshr-rot.ll
…
arith-fshr.ll
…
arith-max-cost.ll
…
arith-mul-load.ll
…
arith-mul-smulo.ll
…
arith-mul-umulo.ll
…
arith-mul.ll
…
arith-scmp.ll
…
arith-smax.ll
…
arith-smin.ll
…
arith-sub-ssat.ll
…
arith-sub-ssubo.ll
…
arith-sub-usat.ll
…
arith-sub-usubo.ll
…
arith-sub.ll
…
arith-ucmp.ll
…
arith-umax.ll
…
arith-umin.ll
…
ashr-node-with-poison.ll
…
associate-copyable-non-copyable-op.ll
[SLP]Update deps for copyables operands, if the user is used several times in node
2026-01-09 15:18:31 -08:00
atomics.ll
…
bad_types.ll
…
bad-reduction.ll
[SLP]Remove LoadCombine workaround after handling of the copyables
2026-02-05 11:16:08 -08:00
barriercall.ll
…
BinOpSameOpcodeHelper.ll
…
bitreverse.ll
…
blending-shuffle-inseltpoison.ll
…
blending-shuffle.ll
…
bool-mask.ll
[SLP] Fix CmpInst type handling in cost model
2026-04-11 08:00:46 -04:00
bottom-to-top-reorder.ll
[SLP] Normalize copyable operand order via majority voting
2026-04-16 12:56:44 -04:00
broadcast_long.ll
…
broadcast.ll
…
bswap-i64-by-i32-chunks.ll
[SLP]Improve analysis for the shl-based reduced values with copyables (#185485)
2026-03-19 14:16:53 -04:00
bswap-reduction-aliased.ll
[SLP]Remove LoadCombine workaround after handling of the copyables
2026-02-05 11:16:08 -08:00
bswap.ll
…
buildvector_splat_extractvalue.ll
…
buildvector-float-and-extract-lane1.ll
[SLP] Initial support for ordered reductions
2026-03-20 13:45:14 -04:00
buildvector-postpone-for-dependency.ll
[SLP] Use the correct identity when combining binary opcodes with AND/MUL (#180457)
2026-02-12 09:34:44 -08:00
buildvector-reused-with-bv-subvector.ll
…
buildvector-same-lane-insert.ll
…
buildvector-schedule-for-subvector.ll
…
buildvector-shuffle-with-root.ll
…
buildvector-shuffle.ll
[SLP]Prefer copyable over alternate
2026-03-20 11:59:59 -04:00
buildvector-with-reuses.ll
…
buildvectors-parent-phi-nodes.ll
Revert "[SLP] Loop aware cost model/tree building"
2026-03-11 04:37:54 -07:00
buildvectors-with-same-parents.ll
…
bv-matched-node-reorder.ll
…
bv-root-part-of-graph.ll
[SLP]Fix operand reordering when estimating profitability of operands
2026-02-27 16:16:22 -08:00
bv-shuffle-mask.ll
[SLP]Prefer copyable over alternate
2026-03-20 11:59:59 -04:00
c-ray.ll
[SLP] Reject 2-element vectorization when vector inst count exceeds scalar
2026-04-11 08:18:36 -04:00
call.ll
…
cast-operand-extracted.ll
…
cast.ll
…
catchswitch-block-in-use.ll
…
cmp_commute-inseltpoison.ll
…
cmp_commute.ll
…
cmp_sel.ll
…
cmp-after-intrinsic-call-minbitwidth.ll
…
cmp-as-alternate-ops.ll
[SLP] Fix CmpInst type handling in cost model
2026-04-11 08:00:46 -04:00
cmp-diff-sized.ll
…
cmp-in-bv-node-type-cost.ll
…
cmp-values-non-full-registers.ll
…
combined-stores-chains.ll
…
commutable-member-in-non-commutable-node.ll
…
commutable-node-with-non-sched-parent.ll
[SLP]Prefer copyable over alternate
2026-03-20 11:59:59 -04:00
commutative-copyable-external-phi-use.ll
[SLP]Fix scheduling of copyable bundle with commutative op used outside parent PHI
2026-04-22 14:59:12 -04:00
commutativity.ll
…
compare-node-with-reuses.ll
[SLP]Do not convert inversed cmp nodes, if they reordered/reused
2026-02-20 06:04:51 -08:00
compare-reduce.ll
…
const-in-different-functions.ll
…
const-reduced-vals-resized.ll
…
constant-based-reductions.ll
[SLP] Replace TrackedToOrig DenseMap with parallel SmallVector in reduction
2026-03-31 16:22:57 -07:00
continue_vectorizing.ll
…
control-dependence.ll
…
control-dependent-schedule.ll
…
control-deps-schedule-data-recalculate.ll
…
conversion-fp16.ll
…
copyable_reorder.ll
[SLP] Normalize copyable operand order to group loads for better vectorization
2026-04-21 10:17:09 -04:00
copyable-child-node-used-outside.ll
[SLP] Use the correct identity when combining binary opcodes with AND/MUL (#180457)
2026-02-12 09:34:44 -08:00
copyable-non-inst-in-stores.ll
[SLP]Better copyable vectorization for stores with non-instructions (#174249)
2026-01-03 17:05:55 -05:00
copyable-operands-reordering.ll
[SLP]Improve analysis of copyables operands for commmutative main instruction
2026-03-26 16:03:58 -04:00
copyable-phi-used-non-copyable.ll
[SLP]Fix dominance failure when value is copyable in one PHI entry but non-copyable in another
2026-04-12 12:00:31 -04:00
copyable-used-outside-with-immediate-op.ll
[SLP]Fix scheduling of copyable bundle with commutative op used outside parent PHI
2026-04-22 14:59:12 -04:00
copyable-with-non-scheduled-parent-node.ll
…
copyable-with-used-outside-bb.ll
[SLP]Prefer instructions, ued outside the block, as the initial main copyable instructions
2025-12-08 09:46:15 -08:00
copyables-with-parent-scalars-in-phis.ll
[SLP]Fix crash with deleted non-copyable node in scheduling copyables
2026-02-12 11:42:02 -08:00
crash_7zip.ll
…
crash_binaryop.ll
…
crash_bullet3.ll
…
crash_bullet.ll
…
crash_clear_undefs.ll
…
crash_cmpop.ll
[SLP] Reject 2-element vectorization when vector inst count exceeds scalar
2026-04-11 08:18:36 -04:00
crash_dequeue.ll
…
crash_flop7.ll
…
crash_gep.ll
…
crash_getpointersdiff-nullopt.ll
[SLPVectorizer] Check std::optional coming out of getPointersDiff (#175784)
2026-01-15 09:07:13 -06:00
crash_lencod-inseltpoison.ll
…
crash_lencod.ll
…
crash_mandeltext.ll
…
crash_netbsd_decompress.ll
…
crash_reordering_undefs.ll
…
crash_scheduling-inseltpoison.ll
…
crash_scheduling.ll
…
crash_sim4b1.ll
…
crash_smallpt.ll
[SLP]Enable float point math ops as copyables elements.
2025-12-26 11:55:58 -08:00
crash_undef_mask_elem.ll
…
crash_vectorizeTree.ll
…
crash-on-out-of-bound-extract.ll
[SLP] Fix crash on extractelement with out-of-bounds index.....Fixes … (#176918)
2026-01-28 06:08:03 -05:00
cross_block_slp.ll
…
cse_extractelement.ll
…
cse.ll
[SLP] Loop aware cost model/tree building
2026-03-18 17:54:01 -04:00
ctlz.ll
…
ctpop-non-power-of-2-reduction.ll
…
ctpop.ll
…
cttz.ll
…
cycle_dup.ll
…
debug_info.ll
…
debug-counter.ll
[Support] Support debug counters in non-assertion builds (#170468)
2025-12-03 16:21:47 +01:00
debug-info-salvage.ll
…
delayed-gather-emission.ll
[SLP] Reject 2-element vectorization when vector inst count exceeds scalar
2026-04-11 08:18:36 -04:00
deleted-inst-reduction-attempt.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
deleted-instructions-clear.ll
[SLP]Initial compatibility support for shl v, 1 and add v, v
2026-04-16 10:32:57 -04:00
deleted-node-with-copyable-operands.ll
[SLP] Loop aware cost model/tree building
2026-03-18 17:54:01 -04:00
diamond_broadcast_extra_shuffle.ll
…
diamond.ll
…
different-vec-widths.ll
…
disjoint-or-reductions.ll
[SLP]Allow bitcast/bswap based reductions for types, larger than the total strided size
2026-03-08 10:37:09 -04:00
div-possibly-extended-with-poisons.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
dot-product.ll
[SLP] Initial support for ordered reductions
2026-03-20 13:45:14 -04:00
empty-vectorizable-tree.ll
…
entries-different-vf.ll
[SLP]Initial compatibility support for shl v, 1 and add v, v
2026-04-16 10:32:57 -04:00
entries-shuffled-diff-sizes.ll
…
entry-no-bundle-but-extra-use-on-vec.ll
[SLP] Reject 2-element vectorization when vector inst count exceeds scalar
2026-04-11 08:18:36 -04:00
expanded-binop-doesnotneedschedule-user.ll
[SLP]Fix stale deps for operands of non-scheduled expanded-binop parents
2026-04-21 12:40:16 -04:00
expected-prof-consecutive-access.ll
[SLP] Loop aware cost model/tree building
2026-03-18 17:54:01 -04:00
ext-used-scalar-different-bitwidth.ll
…
external_user_jumbled_load-inseltpoison.ll
…
external_user_jumbled_load.ll
…
external_user.ll
…
external-bin-op-user.ll
[SLP] Reject 2-element vectorization when vector inst count exceeds scalar
2026-04-11 08:18:36 -04:00
external-reduced-value-vectorized.ll
…
external-used-across-reductions.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
extract_in_tree_user.ll
…
extract-scalar-from-undef.ll
[SLP] Support for copyables in the reduced values (#153589)
2026-01-01 13:31:13 -08:00
extract-shuffle-inseltpoison.ll
…
extract-shuffle.ll
…
extract-subvector-long-input.ll
…
extract-used-in-many-nodes.ll
[SLP]Exclude non-profitable subtrees.
2025-12-30 09:32:05 -08:00
extract-with-undefs-shuffle.ll
…
extract.ll
…
extractcost.ll
…
extractelement-multi-register-use.ll
[SLP] Reject 2-element vectorization when vector inst count exceeds scalar
2026-04-11 08:18:36 -04:00
extractelement-multiple-uses.ll
…
extractelement-phi-in-landingpad.ll
…
extractelement-single-use-many-nodes.ll
[SLP]Enable float point math ops as copyables elements.
2025-12-26 11:55:58 -08:00
extractelement-vecop-vectorized.ll
…
extractelement.ll
…
extractelements-subnodes-same-index.ll
[SLP]Do not generate extractelement subnodes with the same indeces
2026-01-08 07:23:06 -08:00
extractelements-vector-ops-shuffle.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
extractelemets-extended-by-poison.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
extracts-non-extendable.ll
[IR] Mark reduction intrinsics as nocreateundeforpoison (#184173)
2026-03-03 09:51:12 +01:00
fabs-cost-softfp.ll
…
fabs.ll
…
fcopysign.ll
…
flag.ll
…
float-min-max.ll
…
fma.ll
…
fmaximum-fminimum.ll
…
fmaxnum.ll
[SLP] Initial support for ordered reductions
2026-03-20 13:45:14 -04:00
fminimumnum.ll
…
fminnum.ll
[SLP] Initial support for ordered reductions
2026-03-20 13:45:14 -04:00
fmuladd.ll
…
fptosi-inseltpoison.ll
…
fptosi.ll
…
fptoui.ll
…
fround.ll
…
full-match-with-poison-scalar.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
full-matched-bv-with-subvectors.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
full-non-schedulable-overlap.ll
…
funclet.ll
…
gather-extractelements-different-bbs.ll
…
gather-loads-non-power-of-2.ll
…
gather-move-out-of-loop.ll
[SLP]Remove Alternate early profitability checks in favor of throttling
2026-03-08 09:37:51 -04:00
gather-node-same-as-vect-but-order.ll
…
gather-node-same-reduced.ll
[SLP]Do not increase depth for type-changing nodes and NotProfitableForVectorization removal
2026-02-17 09:52:23 -05:00
gather-nodes-different-bb.ll
…
gather-with-cmp-user.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
gathered-delayed-nodes-with-reused-user.ll
[SLP]Remove Alternate early profitability checks in favor of throttling
2026-03-08 09:37:51 -04:00
gathered-loads-non-full-reg.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
gathered-node-with-in-order-parent.ll
…
gathered-shuffle-resized.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
gep_mismatch.ll
…
gep-nodes-with-non-gep-inst.ll
…
gep-nonconst-idx-transformed-to-const.ll
…
gep.ll
…
geps-non-pow-2.ll
[SLP] Loop aware cost model/tree building
2026-03-18 17:54:01 -04:00
hadd-inseltpoison.ll
…
hadd.ll
…
hoist.ll
…
horizontal-list.ll
[SLP] Initial support for ordered reductions
2026-03-20 13:45:14 -04:00
horizontal-minmax.ll
[SLP] Initial support for ordered reductions
2026-03-20 13:45:14 -04:00
horizontal-smax.ll
…
horizontal-store-many-uses.ll
…
horizontal.ll
…
hsub-inseltpoison.ll
…
hsub.ll
…
icmp-altopcode-after-reordering.ll
…
identity-match-splat-less-defined.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
identity-reuses-with-poisons.ll
[SLP]Fix crash in getReorderingData on all-poison reuse-mask slice
2026-04-27 10:53:48 -04:00
implicitfloat.ll
…
in-tree-user.ll
…
insert-after-bundle.ll
…
insert-subvector.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
insertelement-with-copyable-args.ll
…
insertelements-with-reused-indices.ll
…
insertvalue.ll
…
inst_size_bug.ll
…
InstructionsState-is-invalid-1.ll
…
intrinsic_with_scalar_param.ll
…
intrinsic.ll
…
inversed-icmp-to-gather.ll
[SLP] Fix CmpInst type handling in cost model
2026-04-11 08:00:46 -04:00
jumbled-load-multiuse.ll
…
jumbled-load-shuffle-placement.ll
…
jumbled-load-used-in-phi.ll
…
jumbled-load.ll
…
landing_pad.ll
…
landing-pad-for-split-node.ll
…
last-non-copyable-inst-used-outside-bb.ll
…
lhs-constant-non-cummutative.ll
[SLP] Fix misvectorization in commutative to non-commutative conversion (#185230)
2026-03-09 16:17:39 -04:00
lit.local.cfg
…
load-bitcast-vec.ll
…
load-merge-inseltpoison.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
load-merge.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
load-partial-vector-shuffle.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
long_chains.ll
…
long-full-reg-stores.ll
…
long-pointer-distance.ll
…
lookahead.ll
[SLP] Reject 2-element vectorization when vector inst count exceeds scalar
2026-04-11 08:18:36 -04:00
loopinvariant.ll
…
main-alternate-interechanged-detect.ll
…
malformed_phis.ll
[SLP] Loop aware cost model/tree building
2026-03-18 17:54:01 -04:00
many_stores.ll
…
many-uses-parent-node.ll
…
mask-size-less-common-mask.ll
[SLP]Match the mask size, when copying mask for full match
2026-03-08 05:33:30 -07:00
masked-load-compress-reordered.ll
…
matched-bv-schedulable.ll
…
matched-gather-part-of-combined.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
matched-nodes-updated.ll
…
matched-shuffled-entries.ll
[SLP]Prefer copyable over alternate
2026-03-20 11:59:59 -04:00
matching-gather-nodes-phi-users.ll
[SLP] Loop aware cost model/tree building
2026-03-18 17:54:01 -04:00
matching-insert-point-for-nodes.ll
[SLP]Prefer copyable over alternate
2026-03-20 11:59:59 -04:00
memory-runtime-checks.ll
…
metadata.ll
…
minbitwidth-cast.ll
…
minbitwidth-drop-wrapping-flags.ll
[SLP]Prefer copyable over alternate
2026-03-20 11:59:59 -04:00
minbitwidth-icmp-to-trunc.ll
…
minbitwidth-node-but-not-operands.ll
…
minbitwidth-transformed-operand.ll
…
minbw-bitcast-to-fp.ll
[SLP]Prefer copyable over alternate
2026-03-20 11:59:59 -04:00
minbw-multiused-from-gather.ll
…
minbw-node-used-twice.ll
…
minbw-user-non-sizable.ll
…
minimum-sizes.ll
…
mul64.ll
…
multi_block.ll
…
multi_user.ll
…
multi-extracts-bv-combined.ll
[SLP]Prefer copyable over alternate
2026-03-20 11:59:59 -04:00
multi-incoming-blocks-in-phi.ll
…
multi-node-for-copyable-parent.ll
[SLP]Exclude non-profitable subtrees.
2025-12-30 09:32:05 -08:00
multi-node-reuse-in-bv.ll
…
multi-node-user-with-copyable-ops.ll
[SLP]Exclude non-profitable subtrees.
2025-12-30 09:32:05 -08:00
multi-node-with-save-values.ll
…
multi-nodes-bv-vectorized.ll
…
multi-nodes-to-shuffle.ll
…
multi-parent-instr-copyable-regular.ll
[SLP] Reject 2-element vectorization when vector inst count exceeds scalar
2026-04-11 08:18:36 -04:00
multi-tracked-reduced-value.ll
[SLP] Support for copyables in the reduced values (#153589)
2026-01-01 13:31:13 -08:00
multi-use-bitcasted-reduction.ll
[SLP][NFC]Fix run line for the test, fix test name, NFC
2026-04-05 12:20:43 -04:00
no_alternate_divrem.ll
…
no-schedule-terminate-inst.ll
…
no-scheduled-instructions.ll
…
node-outside-used-only.ll
…
non-commutative-op-in-commutative-inst.ll
[SLP]Enable float point math ops as copyables elements.
2025-12-26 11:55:58 -08:00
non-load-reduced-as-part-of-bv.ll
[SLP] Fix CmpInst type handling in cost model
2026-04-11 08:00:46 -04:00
non-power-2-num-elems-reused.ll
…
non-power-of-2-bswap.ll
[SLP]Improve analysis for the shl-based reduced values with copyables (#185485)
2026-03-19 14:16:53 -04:00
non-power-of-2-num-elems.ll
…
non-power-of-2-order-detection.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
non-power-of-2-subvectors-insert.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
non-reduced-select-of-bits.ll
[SLP] Check if potential bitcast/bswap candidate is a root of reduction
2026-03-28 13:58:22 -07:00
non-sched-inst-has-copyable-before.ll
…
non-schedulable-before-main.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
non-schedulable-instructions-become-schedulable.ll
…
non-schedulable-node-with-non-schedulable-parent.ll
[SLP]Fix operand reordering when estimating profitability of operands
2026-02-27 16:16:22 -08:00
non-schedulable-parent-multi-copyables.ll
[SLP]Prefer copyable over alternate
2026-03-20 11:59:59 -04:00
non-schedulable-recalculate-deps.ll
[SLP]Recalculate dependencies for all cleared entries
2025-12-26 11:17:14 -08:00
non-schedulable-split-node.ll
…
non-schedulable-with-copyable-op.ll
[SLP]Recalculate dependencies for all cleared entries
2025-12-26 11:17:14 -08:00
non-schedulable-with-multi-used-expanded.ll
[SLP]Bail out on non-schedulable expanded binop with multi-use operand
2026-04-27 13:40:11 -04:00
non-scheduled-inst-extern-use.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
non-scheduled-inst-reused-as-last-inst.ll
[SLP]Remove Alternate early profitability checks in favor of throttling
2026-03-08 09:37:51 -04:00
non-vectorizable-inst-operand.ll
[SLP][NFC]Rename test, add other non-vectorizable inst candidates tests, NFC
2026-04-25 11:57:11 -04:00
num-uses-for-copyable-elements.ll
…
odd_store.ll
…
opaque-ptr-2.ll
…
opaque-ptr.ll
…
operand-reorder-with-copyables.ll
[SLP] Normalize copyable operand order via majority voting
2026-04-16 12:56:44 -04:00
operandorder.ll
…
opt.ll
…
ordering-bug.ll
…
ordering.ll
…
original-inst-scheduled-after-copyable.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
parent-bitcast-with-fp.ll
…
parent-node-non-schedulable.ll
[SLP]Initial support for vector register spills/reloads estimation
2026-03-26 14:27:27 -04:00
parent-node-schedulable-with-multi-copyables.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
parent-node-split-non-schedulable.ll
[SLP] Initial support for ordered reductions
2026-03-20 13:45:14 -04:00
parent-non-schedule-multi-use-in-binop.ll
…
parent-phi-node-reordered.ll
…
partail.ll
…
perfect-matched-reused-bv.ll
…
phi3.ll
…
phi_landingpad.ll
…
phi_overalignedtype.ll
…
phi-comparator-fix-vec-ops-compare.ll
…
phi-comparator.ll
…
phi-multi-same-nodes.ll
[SLP]Correctly handle vector nodes, coming from same incoming blocks in PHI nodes
2026-01-21 12:59:21 -08:00
phi-node-bitwidt-op-not.ll
…
phi-node-deletion.ll
…
phi-node-reshuffled-part.ll
…
phi-node-with-cycle.ll
…
phi-nodes-as-operand-reorder.ll
…
phi-nodes-incoming-same-blocks.ll
…
phi-operand-gathered-loads.ll
[SLP] Fix GEP cost computation for load vectorization cost estimates
2026-04-12 12:36:19 -04:00
phi-removed-on-operand-vectorization.ll
[SLP] Reject 2-element vectorization when vector inst count exceeds scalar
2026-04-11 08:18:36 -04:00
phi-reordered-reshuffled.ll
…
phi.ll
[SLP] Initial support for ordered reductions
2026-03-20 13:45:14 -04:00
phis-different-order.ll
…
poison-element-shuffle.ll
…
poison-within-divisions.ll
…
powi-regression.ll
…
powi.ll
…
powof2div.ll
…
powof2mul.ll
…
pr16571.ll
…
pr16628.ll
…
pr16899.ll
…
pr18060.ll
…
pr19657.ll
…
pr23510.ll
…
pr27163.ll
…
pr35497.ll
[SLP] Reject 2-element vectorization when vector inst count exceeds scalar
2026-04-11 08:18:36 -04:00
pr40522.ll
…
pr42022-inseltpoison.ll
…
pr42022.ll
…
pr44067-inseltpoison.ll
…
pr44067.ll
…
pr46983.ll
[SLP]Exclude non-profitable subtrees.
2025-12-30 09:32:05 -08:00
pr47623.ll
…
pr47629-inseltpoison.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
pr47629.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
pr47642.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
pr48879-sroa.ll
[SLP] Support for copyables in the reduced values (#153589)
2026-01-01 13:31:13 -08:00
pr49081.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
pr49933.ll
…
pr52275.ll
…
pr54465.ll
…
pr59867.ll
…
pr63668.ll
…
pr98978.ll
…
pr176906.ll
[CostModel][X86] reduce_add(vXi1) will lower as a scalar ctpop (#178400)
2026-01-28 11:43:30 +00:00
PR31847.ll
…
PR32086.ll
…
PR34635.ll
…
PR35628_1.ll
…
PR35628_2.ll
…
PR35777.ll
…
PR35865-inseltpoison.ll
…
PR35865.ll
…
PR36280.ll
…
PR39774.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
PR40310.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
propagate_ir_flags.ll
[SLP]Prefer copyable over alternate
2026-03-20 11:59:59 -04:00
propagate-mmra.ll
[SLP]Enable float point math ops as copyables elements.
2025-12-26 11:55:58 -08:00
reassociated-loads.ll
…
reduce-constants.ll
…
reduce-with-folded-to-consts.ll
…
reduced-bswap-with-larger-reduced-type.ll
[SLP]Fix legality checks for bswap-based transformations
2026-03-16 11:56:24 -07:00
reduced-ordered-values-update.ll
[SLP]Fix patterns for compile time blow up with ordered reductions
2026-03-22 13:42:54 -07:00
reduced-val-extracted-and-externally-used.ll
[SLP] Loop aware cost model/tree building
2026-03-18 17:54:01 -04:00
reduced-val-vectorized-in-transform.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
reduced-value-replace-extractelement.ll
…
reduced-value-stored.ll
[SLP] Fix CmpInst type handling in cost model
2026-04-11 08:00:46 -04:00
reduced-value-vectorized-later.ll
…
reduction2.ll
[SLP] Reject 2-element vectorization when vector inst count exceeds scalar
2026-04-11 08:18:36 -04:00
reduction_unrolled.ll
…
reduction-across-different-bb.ll
[SLP]Use reduction root explicitly from reduction analysis to avoid non-determinism
2026-03-23 09:46:33 -07:00
reduction-bool-logic-op-inside.ll
…
reduction-extracted-value.ll
[SLP]Improve reductions for copyables/split nodes
2026-03-19 12:03:05 -04:00
reduction-logical.ll
[IR] Mark reduction intrinsics as nocreateundeforpoison (#184173)
2026-03-03 09:51:12 +01:00
reduction-same-vals.ll
…
reduction-shl1-add-merge.ll
[SLP]Initial compatibility support for shl v, 1 and add v, v
2026-04-16 10:32:57 -04:00
reduction-transpose.ll
…
reduction-value-in-tree.ll
…
reduction-with-removed-extracts.ll
…
reduction.ll
…
redux-feed-buildvector.ll
…
redux-feed-insertelement.ll
…
remark_alternate.ll
…
remark_extract_broadcast.ll
…
remark_gather-load-redux-cost.ll
…
remark_horcost.ll
[SLP] Loop aware cost model/tree building
2026-03-18 17:54:01 -04:00
remark_listcost.ll
…
remark_not_all_parts.ll
[SLP] Loop aware cost model/tree building
2026-03-18 17:54:01 -04:00
remark_unsupported.ll
…
remark-masked-loads-consecutive-loads-same-ptr.ll
…
remark-partial-loads-vectorize.ll
…
reorder_diamond_match.ll
…
reorder_phi.ll
…
reorder_repeated_ops.ll
…
reorder_with_external_users.ll
[SLP] Reject 2-element vectorization when vector inst count exceeds scalar
2026-04-11 08:18:36 -04:00
reorder_with_reordered_users.ll
…
reorder-phi-operand.ll
…
reorder-possible-strided-node.ll
…
reorder-reused-masked-gather2.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
reorder-reused-masked-gather.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
reorder-reused-subvector.ll
…
reorder-vf-to-resize.ll
…
reordered-masked-loads.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
replaced-external-in-reduction.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
resched.ll
[SLP]Prefer copyable over alternate
2026-03-20 11:59:59 -04:00
reschedule-only-scheduled.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
resized-bv-values-non-power-of2-node.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
reuse-extracts-in-wider-vect.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
reused-extractelements.ll
…
reused-extracts.ll
…
reused-last-instruction-in-split-node.ll
[SLP] Normalize copyable operand order via majority voting
2026-04-16 12:56:44 -04:00
reused-mask-with-poison-index.ll
…
reused-reductions-with-minbitwidth.ll
…
reused-scalars-in-buildvector.ll
…
reused-undefs.ll
…
revec-estimateNodesPermuteCost.ll
…
revec-ExtractSubvector.ll
…
revec-fix-99411.ll
…
revec-getAltInstrMask.ll
…
revec-getExtractWithExtendCost.ll
…
revec-getStoreMinimumVF.ll
…
revec-load-compress.ll
…
revec-non-power-2-to-power-2-large-vect.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
revec-reduced-value-replace-extractelement.ll
…
revec-reduced-value-vectorized-later.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
revec-scalar-insertelement.ll
[SLP][REVEC] Fix InsertElement scalar type in getEntryCost
2026-04-11 09:27:10 -04:00
revec-split-entry.ll
[SLP]Fix revec in split nodes
2026-02-16 14:09:28 -08:00
revec-SplitVectorize.ll
…
revectorize-phis.ll
…
revectorized_rdx_crash.ll
…
reverse_extract_elements.ll
…
rgb_phi.ll
[SLP] Reject 2-element vectorization when vector inst count exceeds scalar
2026-04-11 08:18:36 -04:00
root-gather-reused-scalar.ll
…
root-node-reordered-reused.ll
…
same-last-instruction-different-parents.ll
[SLP]Exclude non-profitable subtrees.
2025-12-30 09:32:05 -08:00
same-operands-but-copyable.ll
[SLP]Exclude non-profitable subtrees.
2025-12-30 09:32:05 -08:00
same-values-sub-node-with-poisons.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
saxpy.ll
…
scalarize-ctlz.ll
[SLP]Exclude non-profitable subtrees.
2025-12-30 09:32:05 -08:00
scatter-vectorize-reorder-non-empty.ll
[SLP] Initial support for ordered reductions
2026-03-20 13:45:14 -04:00
scatter-vectorize-reorder.ll
…
scatter-vectorize-reused-pointer.ll
…
schedule_budget_debug_info.ll
…
schedule_budget.ll
…
schedule-bundle.ll
…
schedule-control-deps-after-copyable.ll
…
schedule-same-user-with-copyable.ll
…
scheduling.ll
…
select-copyable-cmp-poison.ll
[SLP] Reuse diamond-matched gather across asymmetric reorder/reuse
2026-04-26 12:50:49 -04:00
select-logical-or-and-i1-vector.ll
[SLP] Improve cost model for i1 select-as-or/and patterns
2026-04-20 19:22:45 -04:00
select-reduction-op.ll
…
sext-inseltpoison.ll
…
sext.ll
…
shift-ashr.ll
…
shift-lshr.ll
…
shift-shl.ll
…
shl-compatible-with-add.ll
[SLP]Initial compatibility support for shl v, 1 and add v, v
2026-04-16 10:32:57 -04:00
shl-to-add-transformation2.ll
[SLP][NFC]Add another test for shl-to-add transformation, NFC
2026-02-03 03:56:28 -08:00
shl-to-add-transformation3.ll
[SLP][NFC]Add another shl-to-add transformation test, NFC
2026-02-04 08:08:12 -08:00
shl-to-add-transformation4.ll
[SLP] Reject 2-element vectorization when vector inst count exceeds scalar
2026-04-11 08:18:36 -04:00
shl-to-add-transformation5.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
shl-to-add-transformation.ll
[SLP]Initial compatibility support for shl v, 1 and add v, v
2026-04-16 10:32:57 -04:00
shll1-add-sub-combined.ll
[SLP]Initial compatibility support for shl v, 1 and add v, v
2026-04-16 10:32:57 -04:00
shrink_after_reorder.ll
[SLP] Reuse diamond-matched gather across asymmetric reorder/reuse
2026-04-26 12:50:49 -04:00
shuffle-mask-emission.ll
…
shuffle-multiple-nodes.ll
…
shuffled-gathers-diff-size.ll
…
sign-extend-inseltpoison.ll
…
sign-extend.ll
…
simple-loop.ll
…
simplebb.ll
[SLP] Reject 2-element vectorization when vector inst count exceeds scalar
2026-04-11 08:18:36 -04:00
sin-sqrt.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
sitofp-inseltpoison.ll
…
sitofp-minbitwidth-node.ll
…
sitofp.ll
…
slp-fma-loss.ll
…
slp-schedule-use-order.ll
[SLP] Loop aware cost model/tree building
2026-03-18 17:54:01 -04:00
slp-throttle.ll
[SLP]Remove Alternate early profitability checks in favor of throttling
2026-03-08 09:37:51 -04:00
small-graph-diff-block-instructions.ll
[SLP]Remove Alternate early profitability checks in favor of throttling
2026-03-08 09:37:51 -04:00
splat-score-adjustment.ll
…
split-load8_2_unord_geps.ll
…
split-load8_2-unord.ll
…
split-node-full-match.ll
…
split-node-last-inst-vectorized.ll
…
split-node-marked-to-gather.ll
[SLP]Do not vectorize subtrees of the split node, marked as gathers.
2026-01-28 17:44:38 -08:00
split-node-no-reorder-copy.ll
…
split-node-num-operands.ll
…
split-node-parent-operands-in-spill.ll
…
split-node-reduce-root.ll
[SLP]Consider split node as potential reduction root
2026-01-02 06:42:44 -08:00
split-node-reorder-node-with-ops.ll
[SLP]Exclude non-profitable subtrees.
2025-12-30 09:32:05 -08:00
split-node-reused-in-later-vector.ll
[SLP]Do not cache sentinel position for SplitVectorize nodes
2026-04-24 15:52:33 -04:00
split-node-throttled.ll
[SLP]Do not cache sentinel position for SplitVectorize nodes
2026-04-24 15:52:33 -04:00
split-vector-operand-with-reuses.ll
…
split-vectorize-gathered-def-after-use.ll
[SLP]Initial support for non-power-of-2 vectorization
2026-04-25 18:30:32 -04:00
split-vectorize-phi-user.ll
[SLP]Do not consider split vectorize nodes as vector phi nodes
2026-03-04 14:03:20 -08:00
sqrt.ll
…
stackrestore-dependence.ll
…
stacksave-dependence.ll
…
store-abs-minbitwidth.ll
…
store-constant.ll
…
store-insertelement-minbitwidth.ll
…
store-jumbled.ll
…
stores_constant_float.ll
…
stores_mix_sizes.ll
…
stores_vectorize.ll
…
subvector-minbitwidth-unsigned-value.ll
[SLP]Support for zext i1 %x modeling as select %x, 1, 0
2026-02-10 12:54:12 -08:00
supernode.ll
[SLP] Reject 2-element vectorization when vector inst count exceeds scalar
2026-04-11 08:18:36 -04:00
tiny-tree.ll
…
trunc-node-reused.ll
…
trunc-store-value-ty-not-power-of-2.ll
…
trunced-buildvector-scalar-extended.ll
[SLP]Exclude non-profitable subtrees.
2025-12-30 09:32:05 -08:00
uitofp-with-signed-value-bitwidth.ll
…
uitofp.ll
…
undef_vect.ll
…
unreachable.ll
…
used-reduced-op.ll
[SLP] Initial support for ordered reductions
2026-03-20 13:45:14 -04:00
user-buildvector-with-minbiwidth.ll
…
user-node-no-state.ll
…
user-node-with-same-last-instr.ll
…
user-with-multi-copyable-ops.ll
[SLP]Enable float point math ops as copyables elements.
2025-12-26 11:55:58 -08:00
value-bug-inseltpoison.ll
…
value-bug.ll
…
vec3-base.ll
[SLP] Reject 2-element vectorization when vector inst count exceeds scalar
2026-04-11 08:18:36 -04:00
vec3-calls.ll
…
vec3-gather-some-loads.ll
…
vec3-reorder-reshuffle.ll
[SLP] Reject 2-element vectorization when vector inst count exceeds scalar
2026-04-11 08:18:36 -04:00
vec_list_bias_external_insert_shuffled.ll
[SLP]Add external uses estimations into tree throttling
2026-02-11 16:14:34 -05:00
vec_list_bias-inseltpoison.ll
[SLP]Add external uses estimations into tree throttling
2026-02-11 16:14:34 -05:00
vec_list_bias.ll
[SLP]Add external uses estimations into tree throttling
2026-02-11 16:14:34 -05:00
vect_copyable_in_binops.ll
[SLP]Enable float point math ops as copyables elements.
2025-12-26 11:55:58 -08:00
vect-gather-same-nodes.ll
…
vector_gep.ll
…
vector-like-instr-does-not-dominate.ll
…
vector.ll
…
vectorize-cmps.ll
…
vectorize-pair-path.ll
[SLP] Support for copyables in the reduced values (#153589)
2026-01-01 13:31:13 -08:00
vectorize-reorder-reuse.ll
…
vectorize-widest-phis.ll
[SLP] Loop aware cost model/tree building
2026-03-18 17:54:01 -04:00
whole-registers-compare.ll
…
xor-combined-opcode.ll
…
xor-with-zero-and-incompat.ll
[SLP]Mark and incompatible for 'xor %a, 0' operations
2025-12-31 08:30:50 -08:00
zext-inseltpoison.ll
…
zext.ll
…