summaryrefslogtreecommitdiffstats
path: root/llvm/lib/Transforms/Vectorize/LoopVectorize.cpp
Commit message (Expand)AuthorAgeFilesLines
* [LV] Don't call recordVectorLoopValueForInductionCast for newly-created IV fr...Andrei Elovikov2018-01-151-2/+7
* [LV][VPlan] NFC patch to move LoopVectorizationPlanner class out of LoopVecto...Hal Finkel2018-01-071-203/+11
* Use phi ranges to simplify code. No functionality change intended.Benjamin Kramer2017-12-301-36/+18
* [LV] Remove unnecessary DoExtraAnalysis guard (silent bug)Florian Hahn2017-12-201-2/+2
* [LV] Extend InstWidening with CM_Widen_RecursiveHal Finkel2017-12-161-5/+16
* [LV] NFC patch for moving VP*Recipe class definitions from LoopVectorize.cpp ...Hal Finkel2017-12-161-392/+74
* [LV] Support efficient vectorization of an induction with redundant castsDorit Nuzman2017-12-141-6/+94
* [LV] Ignore the cost of values that will not appear in the vectorized loopDorit Nuzman2017-12-121-1/+2
* [LV] Interleaved access vectorization: fix computing new alias infoAdam Nemet2017-12-061-2/+16
* Add MemorySSA as loop dependency, disabled by default [NFC].Alina Sbirlea2017-11-211-1/+1
* [LV] Model masking in VPlan, introducing VPInstructionsGil Rapaport2017-11-201-117/+226
* [LV] Introduce VPBlendRecipe, VPWidenMemoryInstructionRecipeGil Rapaport2017-11-141-124/+241
* Add an @llvm.sideeffect intrinsicDan Gohman2017-11-081-1/+1
* [IR] redefine 'UnsafeAlgebra' / 'reassoc' fast-math-flags and add 'trans' fas...Sanjay Patel2017-11-061-3/+3
* [LoopVectorize] Replace manual VPlan memory management with unique_ptr.Benjamin Kramer2017-10-311-26/+10
* Do not add discriminator encoding for debug intrinsics.Dehao Chen2017-10-261-1/+2
* [Transforms] Fix some Clang-tidy modernize and Include What You Use warnings;...Eugene Zelenko2017-10-121-163/+236
* [NFC] Convert OptimizationRemarkEmitter old emit() calls to new closureVivek Pandya2017-10-111-48/+69
* [LV] Fix PR34743 - handle casts that sink after interleaved loadsAyal Zaks2017-10-051-1/+4
* [LV] Fix PR34711 - widen instruction ranges when sinking castsAyal Zaks2017-10-051-23/+22
* [LV] Use correct insertion point when type shrinking reductionsMatthew Simpson2017-09-291-1/+2
* Use a BumpPtrAllocator for Loop objectsSanjoy Das2017-09-281-1/+1
* Allow ORE.emit to take a closure to delay building the remark objectAdam Nemet2017-09-191-2/+4
* This patch fixes https://bugs.llvm.org/show_bug.cgi?id=32352 Vivek Pandya2017-09-151-9/+12
* This reverts r313381Vivek Pandya2017-09-151-12/+9
* This patch fixes https://bugs.llvm.org/show_bug.cgi?id=32352 Vivek Pandya2017-09-151-9/+12
* [LV] Fix maximum legal VF calculationAlon Kom2017-09-141-28/+18
* [LV] Avoid computing the register usage for default VF. NFCAnna Thomas2017-09-131-2/+4
* [LV] Fix PR34523 - avoid generating redundant selectsAyal Zaks2017-09-131-3/+3
* [LV] Clamp the VF to the trip countAnna Thomas2017-09-121-7/+12
* LoopVectorize: MaxVF should not be larger than the loop trip countZvi Rackover2017-09-041-4/+8
* [LoopVectorize] Turn static DenseSet into switch.Benjamin Kramer2017-09-021-16/+47
* [LoopVectorizer] Use two step casting for float to pointer types.Manoj Gupta2017-09-011-3/+40
* [LV] Fix PR34248 - recommit D32871 after revert r311304Ayal Zaks2017-08-271-523/+1056
* Revert r311077: [LV] Using VPlan ...Chandler Carruth2017-08-201-1060/+523
* [Loop Vectorize] Added a separate metadataAditya Kumar2017-08-201-9/+18
* [LV] Using VPlan to model the vectorized code and drive its transformationAyal Zaks2017-08-171-523/+1060
* [LV] Minor savings to Sink casts to unravel first order recurrenceAyal Zaks2017-08-151-1/+1
* [LoopVectorize] Fix assertion failure in Fcmp vectorizationAnna Thomas2017-08-081-1/+3
* LV: Don't insert runtime ptr checks on divergent targetsMatt Arsenault2017-08-021-0/+12
* [LV] Avoid redundant operations manipulating masksAyal Zaks2017-07-311-36/+37
* [LV] Test once if vector trip count is zero, instead of twiceAyal Zaks2017-07-191-44/+20
* [LV] Don't allow outside uses of IVs if the SCEV is predicated on loop condit...Michael Kuperstein2017-07-121-2/+7
* Revert "r306473 - re-commit r306336: Enable vectorizer-maximize-bandwidth by ...Teresa Johnson2017-07-011-1/+1
* re-commit r306336: Enable vectorizer-maximize-bandwidth by default.Teresa Johnson2017-07-011-1/+1
* revert r306336 for breaking ppc test.Teresa Johnson2017-07-011-1/+1
* Enable vectorizer-maximize-bandwidth by default.Teresa Johnson2017-07-011-1/+1
* [LV] Sink casts to unravel first order recurrenceAyal Zaks2017-06-301-1/+17
* [LV] Optimize for size when vectorizing loops with tiny trip countAyal Zaks2017-06-301-29/+30
* Revert "r306473 - re-commit r306336: Enable vectorizer-maximize-bandwidth by ...Daniel Jasper2017-06-301-1/+1
OpenPOWER on IntegriCloud