aboutsummaryrefslogtreecommitdiff
path: root/llvm/lib/Analysis/VectorUtils.cpp
AgeCommit message (Expand)AuthorFilesLines
2019-08-02[LV] Avoid building interleaved group in presence of WAW dependencyHideki Saito1-0/+4
2019-06-24[Scalarizer] Add scalarizer support for smul.fix.satBjorn Pettersson1-4/+6
2019-06-11[Analysis] add isSplatValue() for vectors in IRSanjay Patel1-0/+39
2019-06-07[Analysis] simplify code for getSplatValue(); NFCSanjay Patel1-20/+11
2019-04-25Consolidate existing utilities for interpreting vector predicate maskes [NFC]Philip Reames1-0/+46
2019-02-25[Vectorizer] Add vectorization support for fixed smul/umul intrinsicsSimon Pilgrim1-0/+5
2019-02-05[NFC] fix trivial typos in commentsHiroshi Inoue1-1/+1
2019-01-23Move saturated arithmetic intrinsics to other integer intrinsics. NFCI.Simon Pilgrim1-4/+4
2019-01-19Update the file headers across all of the LLVM projects in the monorepoChandler Carruth1-4/+3
2019-01-03[SLPVectorizer] Flag ADD/SUB SSAT/USAT intrinsics trivially vectorizable (PR4...Simon Pilgrim1-0/+4
2018-12-20[NFC] Fix trailing comma after function.Clement Courbet1-1/+1
2018-12-20Introduce llvm.loop.parallel_accesses and llvm.access.group metadata.Michael Kruse1-4/+91
2018-11-13[VectorUtils] Use namespace for InterleaveGroup template specialization.Florian Hahn1-4/+6
2018-11-13[VPlan] VPlan version of InterleavedAccessInfo.Florian Hahn1-10/+24
2018-11-12[VectorUtils] add funnel-shifts to the list of vectorizable intrinsicsSanjay Patel1-0/+2
2018-11-12[VectorUtils] reorder list of vectorizable intrinsics; NFCSanjay Patel1-10/+9
2018-10-31[LV] Support vectorization of interleave-groups that require an epilog underDorit Nuzman1-2/+22
2018-10-22[IAI,LV] Avoid creating a scalar epilogue due to gaps in interleave-groups when Dorit Nuzman1-0/+24
2018-10-19[LoopVectorize] Loop vectorization for minimum and maximumThomas Lively1-0/+2
2018-10-14recommit 344472 after fixing build failure on ARM and PPC.Dorit Nuzman1-9/+20
2018-10-14revert 344472 due to failures.Dorit Nuzman1-20/+9
2018-10-14[IAI,LV] Add support for vectorizing predicated strided accesses using maskedDorit Nuzman1-9/+20
2018-10-07[IAI,LV] Avoid creating interleave-groups for predicated accesseDorit Nuzman1-1/+3
2018-09-24[Analysis] add comment to generalize finding a scalar op from vector; NFCSanjay Patel1-3/+4
2018-09-17Fix vectorization of canonicalizeMatt Arsenault1-0/+1
2018-09-12[LV] Move InterleaveGroup and InterleavedAccessInfo to VectorUtils.h (NFC)Florian Hahn1-0/+327
2018-05-01Remove \brief commands from doxygen comments.Adrian Prantl1-9/+9
2018-02-28Fixed spelling mistake in comments of LLVM Analysis passesVedant Kumar1-1/+1
2017-11-08Add an @llvm.sideeffect intrinsicDan Gohman1-1/+2
2017-07-06[Constants] If we already have a ConstantInt*, prefer to use isZero/isOne/isM...Craig Topper1-1/+1
2017-06-06Sort the remaining #include lines in include/... and lib/....Chandler Carruth1-4/+4
2017-05-09Introduce experimental generic intrinsics for horizontal vector reductions.Amara Emerson1-0/+1
2017-02-01[LV] Move interleaved access helper functions to VectorUtils (NFC)Matthew Simpson1-0/+85
2016-12-02IR: Change the gep_type_iterator API to avoid always exposing the "current" t...Peter Collingbourne1-2/+2
2016-09-11Add handling of !invariant.load to PropagateMetadata.Justin Lebar1-6/+6
2016-06-30SLPVectorizer: Move propagateMetadata to VectorUtilsMatt Arsenault1-0/+41
2016-06-04[Analysis] Enabled BITREVERSE as a vectorizable intrinsicSimon Pilgrim1-0/+1
2016-05-10Revert "[VectorUtils] Query number of sign bits to allow more truncations"James Molloy1-14/+4
2016-05-09[VectorUtils] Query number of sign bits to allow more truncationsJames Molloy1-4/+14
2016-04-19[ValueTracking, VectorUtils] Refactor getIntrinsicIDForCallDavid Majnemer1-145/+8
2016-04-13[InstCombine] We folded an fcmp to an i1 instead of a vector of i1David Majnemer1-2/+2
2016-04-06[SLPVectorizer] Vectorizing the libm sqrt to llvm's sqrt intrinsic requires nnanDavid Majnemer1-1/+3
2016-04-06[SLPVectorizer] Vectorize libcalls of sqrtDavid Majnemer1-0/+4
2016-03-30[VectorUtils] Don't try and truncate PHIs to a smaller bitwidthJames Molloy1-0/+15
2016-01-19[opaque pointer types] [NFC] GEP: replace get(Pointer)ElementType uses with g...Eduard Burtescu1-2/+1
2016-01-17[NFC] Remove one dead PointerType::getElementType() call.Manuel Jacob1-2/+0
2015-12-17[SCEV] Add and use SCEVConstant::getAPInt; NFCISanjoy Das1-2/+1
2015-12-01Fixed a failure in getSpaltValue()Elena Demikhovsky1-1/+2
2015-12-01Fixed a failure in cost calculation for vector GEPElena Demikhovsky1-3/+4
2015-11-26[LoopVectorize] Use MapVector rather than DenseMap for MinBWs.Charlie Turner1-3/+3
generated by cgit v1.1 at 2025-10-08 21:36:32 +0000