summaryrefslogtreecommitdiffstats
path: root/llvm/test/Transforms/LoopVectorize
Commit message (Expand)AuthorAgeFilesLines
* Revert "Support arbitrary addrspace pointers in masked load/store intrinsics"Matthias Braun2016-03-221-95/+20
* [LoopVectorize] Annotate versioned loop with noalias metadataAdam Nemet2016-03-172-0/+137
* [LV] Preserve LoopInfo when store predication is usedAdam Nemet2016-03-151-4/+4
* Support arbitrary addrspace pointers in masked load/store intrinsicsArtur Pilipenko2016-03-101-20/+95
* [x86] fix cost model inaccuracy for vector memory opsSanjay Patel2016-03-091-2/+2
* add a test RUN to show unexpected behaviorSanjay Patel2016-03-091-7/+10
* [LoopUtils, LV] Fix PR26734Matthew Simpson2016-03-031-0/+49
* Fix tests that used CHECK-NEXT-NOT and CHECK-DAG-NOT.Paul Robinson2016-02-262-2/+6
* Revert r255691 "[LoopVectorizer] Refine loop vectorizer's register usage calc...Hans Wennborg2016-02-192-72/+1
* [LV] Vectorize first-order recurrencesMatthew Simpson2016-02-191-0/+209
* [LV] Fix PR26600: avoid out of bounds loads for interleaved access vectorizationSilviu Baranga2016-02-192-5/+9
* Create masked gather and scatter intrinsics in Loop Vectorizer.Elena Demikhovsky2016-02-172-1/+238
* [LV] Add support for insertelt/extractelt processing during type truncationSilviu Baranga2016-02-151-0/+47
* [SCEV][LAA] Re-commit r260085 and r260086, this time with a fix for the memorySilviu Baranga2016-02-081-6/+4
* Revert r260086 and r260085. They have broken the memorySilviu Baranga2016-02-081-4/+6
* [SCEV][LAA] Add no wrap SCEV predicates and use use them to improve strided p...Silviu Baranga2016-02-081-6/+4
* [DemandedBits] Revert r249687 due to PR26071James Molloy2016-02-031-34/+0
* AVX1 : Enable vector masked_load/store to AVX1.Igor Breger2016-01-251-30/+28
* [LoopVectorizer] Refine loop vectorizer's register usage calculator by ignori...Cong Hou2015-12-152-1/+72
* Revert r255460, which still causes test failures on some platforms.Cong Hou2015-12-132-70/+1
* [LoopVectorizer] Refine loop vectorizer's register usage calculator by ignori...Cong Hou2015-12-132-1/+70
* Revert r255454 as it leads to several test failers on buildbots.Cong Hou2015-12-132-69/+1
* [LoopVectorizer] Refine loop vectorizer's register usage calculator by ignori...Cong Hou2015-12-132-1/+69
* [LoopVectorize] Use MapVector rather than DenseMap for MinBWs.Charlie Turner2015-11-261-0/+54
* Pointers in Masked Load, Store, Gather, Scatter intrinsicsElena Demikhovsky2015-11-191-0/+142
* [LoopVectorize] Address post-commit feedback on r250032James Molloy2015-11-091-1/+1
* DI: Reverse direction of subprogram -> function edge.Peter Collingbourne2015-11-0510-28/+28
* LoopVectorizer - skip 'bitcast' between GEP and load.Elena Demikhovsky2015-11-032-86/+125
* Add a flag vectorizer-maximize-bandwidth in loop vectorizer to enable using l...Cong Hou2015-11-022-4/+50
* Revert the revision 251592 as it fails a test on some platforms.Cong Hou2015-10-292-50/+4
* Add a flag vectorizer-maximize-bandwidth in loop vectorizer to enable using l...Cong Hou2015-10-292-4/+50
* Revert r251291, "Loop Vectorizer - skipping "bitcast" before GEP"NAKAMURA Takumi2015-10-272-125/+86
* Loop Vectorizer - skipping "bitcast" before GEPElena Demikhovsky2015-10-262-86/+125
* [LoopVectorize] Shrink integer operations into the smallest type possibleJames Molloy2015-10-121-0/+243
* [LoopUtils,LV] Propagate fast-math flags on generated FCmp instructionsJames Molloy2015-09-211-52/+52
* [LV] Relax Small Size Reduction Type RequirementMatthew Simpson2015-09-101-3/+66
* [PM/AA] Rebuild LLVM's alias analysis infrastructure in a way compatibleChandler Carruth2015-09-091-1/+1
* Delay predication of stores until near the end of vector code generationJames Molloy2015-09-091-9/+34
* Simplify testcase added in r246759. NFCSilviu Baranga2015-09-041-43/+11
* [PowerPC] Enable interleaved-access vectorizationHal Finkel2015-09-041-0/+30
* [PowerPC] Always use aggressive interleaving on the A2Hal Finkel2015-09-031-0/+40
* Fix IRBuilder CreateBitOrPointerCast for vector typesSilviu Baranga2015-09-031-0/+69
* Add newline to test. NFC.Chad Rosier2015-09-021-1/+1
* [LV] Don't bail to MiddleBlock if a runtime check fails, bail to ScalarPH ins...James Molloy2015-09-024-6/+6
* [LV] Refactor all runtime check emissions into helper functions.James Molloy2015-09-021-2/+1
* [LV] Pull creation of trip counts into a helper function.James Molloy2015-09-021-1/+1
* [LV] Never widen an induction variable.James Molloy2015-09-021-0/+34
* [LV] Switch to using canonical induction variables.James Molloy2015-09-022-10/+5
* [AArch64] Turn on by default interleaved access vectorizationSilviu Baranga2015-09-012-3/+3
* [ARM] Turn on by default interleaved access vectorizationSilviu Baranga2015-09-011-1/+1
OpenPOWER on IntegriCloud