summaryrefslogtreecommitdiffstats
path: root/llvm/test/Transforms/LoopVectorize/X86
Commit message (Expand)AuthorAgeFilesLines
* Reapply r298620: [LV] Vectorize GEPsMatthew Simpson2017-04-072-107/+43
* Revert r298620: [LV] Vectorize GEPsIvan Krasin2017-03-242-43/+107
* [LV] Vectorize GEPsMatthew Simpson2017-03-232-107/+43
* [LV] These remark should have been missed remarksAdam Nemet2017-03-011-1/+1
* [AVX-512] Fix the execution domain for AVX-512 integer broadcasts.Craig Topper2017-02-261-1/+1
* Increases full-unroll threshold.Dehao Chen2017-02-181-2/+12
* [Loop Vectorizer] Cost-based decision for vectorization form of memory instru...Elena Demikhovsky2017-02-082-1/+43
* [X86] Remove PCOMMIT instruction support since Intel has deprecated this inst...Craig Topper2017-02-081-2/+2
* [LV] Also port failure remarks to new OptimizationRemarkEmitter APIAdam Nemet2017-02-021-0/+57
* [X86] enable memory interleaving for X86\SLM arch. Mohammed Agabaria2017-01-251-0/+1
* [LV] Add requires asserts to test caseMatthew Simpson2017-01-171-0/+1
* [LV] Mark non-consecutive-like pointers non-uniformMatthew Simpson2017-01-171-0/+55
* [X86] fixing failed test in commit: r291657 Mohammed Agabaria2017-01-111-0/+1
* [X86] updating TTI costs for arithmetic instructions on X86\SLM arch.Mohammed Agabaria2017-01-111-0/+144
* Currently isLikelyComplexAddressComputation tries to figure out if the given ...Mohammed Agabaria2017-01-051-0/+54
* [LV] Scalarize operands of predicated instructionsMatthew Simpson2016-12-071-0/+60
* [LoopVectorizer] When estimating reg usage, unused insts may "end" another useRobert Lougher2016-11-152-1/+135
* [X86][AVX] Fixed v16i16/v32i8 ADD/SUB costs on AVX1 subtargetsSimon Pilgrim2016-11-141-2/+4
* [LV] Stop saying "use -Rpass-analysis=loop-vectorize"Adam Nemet2016-11-113-4/+4
* Second attempt at r285517.Dorit Nuzman2016-10-311-1/+1
* Revert r285517 due to build failures.Dorit Nuzman2016-10-301-1/+1
* [LoopVectorize] Make interleaved-accesses analysis less conservative aboutDorit Nuzman2016-10-301-1/+1
* [X86] Enable interleaved memory access by defaultMichael Kuperstein2016-10-204-11/+46
* [LV] Move insertelement sequence after scalar definitionsMatthew Simpson2016-08-291-16/+16
* [LV] Unify vector and scalar mapsMatthew Simpson2016-08-241-64/+32
* [LV, X86] Be more optimistic about vectorizing shifts.Michael Kuperstein2016-08-041-0/+23
* [LoopVectorize] Change comment for isOutOfScope in collectLoopUniforms, NFCWei Mi2016-08-021-0/+27
* [LV] Generate both scalar and vector integer induction variablesMatthew Simpson2016-08-021-32/+34
* [AVX512] Don't use i128 masked gather/scatter/load/store. Do more accurately ...Igor Breger2016-08-021-0/+76
* [AVX-512] Fix a test missed in r277327.Craig Topper2016-08-011-1/+1
* Initial support for vectorization using svml (short vector math library).Matt Masten2016-07-291-0/+185
* [Loop Vectorizer] Handling loops FP induction variables.Elena Demikhovsky2016-07-241-0/+86
* [LV] Move vector int induction update to end of latchMatthew Simpson2016-07-211-2/+2
* [OptDiag,LV] Add hotness attribute to applied-optimization remarksAdam Nemet2016-07-211-4/+4
* [OptDiag,LV] Add hotness attribute to the derived analysis remarksAdam Nemet2016-07-201-0/+113
* Recommit the patch "Use uniforms set to populate VecValuesToIgnore".Wei Mi2016-07-192-5/+73
* Revert rL275912.Wei Mi2016-07-182-73/+5
* Use uniforms set to populate VecValuesToIgnore.Wei Mi2016-07-182-5/+73
* [X86] Make some cast costs more preciseMichael Kuperstein2016-07-111-2/+2
* Fixed a bug in vectorizing GEP before gather/scatter intrinsic.Elena Demikhovsky2016-07-072-3/+221
* [TTI] The cost model should not assume vector casts get completely scalarizedMichael Kuperstein2016-07-061-3/+3
* SLPVectorizer: Move propagateMetadata to VectorUtilsMatt Arsenault2016-06-301-0/+25
* Refine the set of UniformAfterVectorization instructions.Wei Mi2016-06-301-0/+50
* Support arbitrary addrspace pointers in masked load/store intrinsicsArtur Pilipenko2016-06-281-20/+95
* Revert -r273892 "Support arbitrary addrspace pointers in masked load/store in...Artur Pilipenko2016-06-271-95/+20
* Support arbitrary addrspace pointers in masked load/store intrinsicsArtur Pilipenko2016-06-271-20/+95
* [LV] For some IVs, use vector phis instead of widening in the loop bodyMichael Kuperstein2016-06-011-3/+3
* Move test to X86 directory: I think it depends on X86 TTI.Tim Northover2016-05-271-0/+75
* Look for a loop's starting location in the llvm.loop metadataHal Finkel2016-05-251-0/+74
* [x86] avoid code explosion from LoopVectorizer for gather loop (PR27826) Sanjay Patel2016-05-251-0/+41
OpenPOWER on IntegriCloud