aboutsummaryrefslogtreecommitdiff
path: root/llvm/test/Transforms/SLPVectorizer/AArch64
AgeCommit message (Expand)AuthorFilesLines
4 days[SLP]Fix types for reductions in revecAlexey Bataev1-0/+82
4 days[SLP]Do not vectorize select nodes with scalar and vector conditionsAlexey Bataev1-0/+29
2026-02-17[SLP]Fix an ArrayRef out-of-bounds access in sliceAlexey Bataev1-0/+121
2026-02-12[SLP]Fix crash with deleted non-copyable node in scheduling copyablesAlexey Bataev1-17/+21
2026-02-11[SLP]Skip operands comparing on non-matching (but compatible) instructionsAlexey Bataev1-0/+42
2026-02-11[SLP] Use the correct calling convention for vector math routines (#180759)David Sherwood1-4/+249
2026-01-09[SLP]Update deps for copyables operands, if the user is used several times in...Alexey Bataev1-0/+259
2026-01-05[SLP] Report the correct operand to getArithmeticInstrCost() when duplicated ...Ryan Buchner1-0/+44
2025-12-30[SLP]Exclude non-profitable subtrees.Alexey Bataev3-98/+48
2025-12-30[SLP][NFC]Add a test that reveals the issue with non-profitable subtrees dele...Alexey Bataev1-0/+56
2025-12-30Revert "[SLP]Exclude non-profitable subtrees."Alexey Bataev3-48/+98
2025-12-29[SLP]Exclude non-profitable subtrees.Alexey Bataev3-98/+48
2025-12-26[SLP]Enable float point math ops as copyables elements.Alexey Bataev1-3/+3
2025-12-26Revert "[SLP]Enable float point math ops as copyables elements."Alexey Bataev1-3/+3
2025-12-25[SLP]Enable float point math ops as copyables elements.Alexey Bataev1-3/+3
2025-12-22Revert "[SLP]Enable float point math ops as copyables elements."Alexey Bataev1-3/+3
2025-12-21[SLP]Enable float point math ops as copyables elements.Alexey Bataev1-3/+3
2025-12-15[AArch64] use `isTRNMask` to calculate shuffle costs (#171524)Philip Ginsbach-Chen2-6/+52
2025-11-14[SLP]Check if the copyable element is a sub instruciton with abs in isCommutableAlexey Bataev1-0/+29
2025-11-14Revert "[SLP]Check if the copyable element is a sub instruciton with abs in i...Alexey Bataev1-29/+0
2025-11-14[SLP]Check if the copyable element is a sub instruciton with abs in isCommutableAlexey Bataev1-0/+29
2025-11-04[AArch64] Improve the cost model for extending mull (#125651)David Green1-2/+2
2025-11-01[SLP][NFC]Fix UB and constant folded ops in test, NFCAlexey Bataev1-17/+17
2025-10-20[SLP]Do not pack div-like copyable valuesAlexey Bataev1-0/+53
2025-10-20[IR] Replace alignment argument with attribute on masked intrinsics (#163802)Nikita Popov1-3/+3
2025-09-12[llvm] Regenerate test checks including TBAA semantics (NFC)Antonio Frighetto3-32/+48
2025-09-02[SLP]Improved/fixed FMAD support in reductionsAlexey Bataev1-0/+47
2025-09-01[LV] Bundle sub reductions into VPExpressionRecipe (#147255)Sam Tebbs1-0/+261
2025-08-14[SLP]Support LShr as base for copyable elementsAlexey Bataev1-5/+2
2025-08-14Revert "[SLP]Support LShr as base for copyable elements"Alex Bradbury1-2/+5
2025-08-14[AArch64] Change the cost of fma and fmuladd to match fmul. (#152963)David Green2-117/+64
2025-08-14[SLP]Support LShr as base for copyable elementsAlexey Bataev1-5/+2
2025-08-11[SLP]Initial support for copyable elementsAlexey Bataev1-3/+3
2025-08-11[SLP]Initial FMAD support (#149102)Alexey Bataev3-64/+136
2025-08-10Revert "[SLP]Initial FMAD support (#149102)"David Green3-166/+70
2025-08-08[SLP]Initial FMAD support (#149102)Alexey Bataev3-70/+166
2025-08-08Revert "[SLP]Initial FMAD support (#149102)"Alexey Bataev3-166/+70
2025-08-07[SLP]Initial FMAD support (#149102)Alexey Bataev3-70/+166
2025-08-07Reland [VectorUtils] Trivially vectorize ldexp, [l]lround (#152476)Ramkumar Ramachandra2-0/+559
2025-08-01Revert "[VectorUtils] Trivially vectorize ldexp, [l]lround (#145545)"Muhammad Omair Javaid2-559/+0
2025-07-29[VectorUtils] Trivially vectorize ldexp, [l]lround (#145545)Ramkumar Ramachandra2-0/+559
2025-07-15[AArch64,TTI] Remove RealUse check for vector insert/extract costs. (#146526)Florian Hahn14-276/+820
2025-07-14[SLP,AArch64] Update build-vector test to actually build vectors.Florian Hahn1-18/+70
2025-07-11[SLP]Remove emission of vector_insert/vector_extract intrinsicsAlexey Bataev10-152/+184
2025-07-10[SLP] Emit reduction instead of 2 extracts + scalar op, when vectorizing oper...Alexey Bataev2-20/+18
2025-07-10Revert "[SLP] Emit reduction instead of 2 extracts + scalar op, when vectoriz...Alex Bradbury3-20/+28
2025-07-10[SLP] Drop unnecessary '' from around -passes=... arg lists to appease update...Simon Pilgrim5-5/+5
2025-07-09[SLP] Emit reduction instead of 2 extracts + scalar op, when vectorizing oper...Alexey Bataev3-28/+20
2025-07-02[SLP] Avoid -passes=instcombine stages in SLP tests (#146257)Hanyang (Eric) Xu7-227/+404
2025-06-23[CostModel] improveShuffleKindFromMask - recognise a SK_PermuteSingleSrc inco...Simon Pilgrim1-13/+10