summaryrefslogtreecommitdiffstats
path: root/llvm/test/Transforms/LoopVectorize
Commit message (Expand)AuthorAgeFilesLines
* [LoopVectorizer] Let target prefer scalar addressing computations.Jonas Paulsson2017-05-241-0/+72
* [LV] Report multiple reasons for not vectorizing under allowExtraAnalysisAyal Zaks2017-05-231-16/+108
* Re-commit r302678, fixing PR33053.Amara Emerson2017-05-162-23/+59
* Revert 303174, 303176, and 303178Matthew Simpson2017-05-161-35/+0
* Make test target-specificMatthew Simpson2017-05-161-0/+4
* Fix test case to unbreak botsMatthew Simpson2017-05-161-3/+1
* [LV] Avoid potentential division by zero when selecting ICMatthew Simpson2017-05-161-0/+33
* Revert r302678 "[AArch64] Enable use of reduction intrinsics."Hans Wennborg2017-05-151-3/+23
* [LoopOptimizer][Fix]PR32859, PR24738Simon Pilgrim2017-05-131-0/+30
* [TLI] Add mapping for various '__<func>_finite' forms of the math routines to...Andrew Kaylor2017-05-121-0/+187
* [AArch64] Enable use of reduction intrinsics.Amara Emerson2017-05-101-23/+3
* [LV] Fix insertion point for shuffle vectors in first order recurrenceAnna Thomas2017-05-091-0/+45
* Support arbitrary address space pointers in masked gather/scatter intrinsics.Elad Cohen2017-05-033-11/+201
* [LV] Handle external uses of floating-point induction variablesMatthew Simpson2017-04-261-0/+63
* [LV] Make LIT test insensitive to basic block numberingGil Rapaport2017-04-251-12/+12
* [LV] Model if-converted phi node costsMatthew Simpson2017-04-211-0/+86
* [LV] Fix the vector code generation for first order recurrenceAnna Thomas2017-04-132-12/+49
* [SystemZ] Fix more target specific testsRenato Golin2017-04-121-0/+2
* [InstCombine] morph an existing instruction instead of creating a new oneSanjay Patel2017-04-121-2/+2
* [LoopVectorizer] Improve handling of branches during cost estimation.Jonas Paulsson2017-04-121-0/+38
* [LoopVectorizer, TTI] New method supportsEfficientVectorElementLoadStore()Jonas Paulsson2017-04-121-0/+33
* [SystemZ] TargetTransformInfo cost functions implemented.Jonas Paulsson2017-04-121-0/+70
* [LV] Avoid vectorizing first order recurrence when phi uses are outside loopAnna Thomas2017-04-112-3/+28
* [LV] Move first order recurrence test to common folder. NFCAnna Thomas2017-04-111-0/+0
* Add address space mangling to lifetime intrinsicsMatt Arsenault2017-04-101-12/+12
* Reapply r298620: [LV] Vectorize GEPsMatthew Simpson2017-04-074-107/+247
* [LV] Make test case more robustMatthew Simpson2017-04-051-33/+49
* [LV] Transform truncations of non-primary induction variablesMatthew Simpson2017-03-271-0/+45
* Revert r298620: [LV] Vectorize GEPsIvan Krasin2017-03-243-104/+107
* [LV] Add regression test for r297610Gil Rapaport2017-03-231-0/+36
* [LV] Vectorize GEPsMatthew Simpson2017-03-233-107/+104
* [LV] Delete unneeded scalar GEP creation codeMatthew Simpson2017-03-232-3/+2
* AMDGPU: Mark all unspecified CC functions in tests as amdgpu_kernelMatt Arsenault2017-03-211-1/+1
* [TargetTransformInfo] getIntrinsicInstrCost() scalarization estimation improvedJonas Paulsson2017-03-142-4/+4
* AMDGPU/SI: Disable unrolling in the loop vectorizer if the loop is not vector...Changpeng Fang2017-03-092-0/+30
* [LV] Select legal insert point when fixing first-order recurrencesMatthew Simpson2017-03-081-0/+22
* [LV] Make the test case for PR30183 less fragileMatthew Simpson2017-03-081-22/+31
* [LV] Add missing check labels to tests and reformatMatthew Simpson2017-03-081-115/+94
* [LV] Consider users that are memory accesses in uniforms expansion stepMatthew Simpson2017-03-071-0/+50
* [ARM/AArch64] Update costs for interleaved accesses with wide typesMatthew Simpson2017-03-022-44/+89
* [LV] Considier non-consecutive but vectorizable accesses for VF selectionMatthew Simpson2017-03-021-0/+33
* [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
* [LV] Merge floating-point and integer induction widening codeMatthew Simpson2017-02-241-63/+56
* [LV] Update floating-point induction test checks (NFC)Matthew Simpson2017-02-221-57/+129
* [LV] Add scalar floating-point induction test (NFC)Matthew Simpson2017-02-221-0/+58
* [LoopVectorize] Added address space check when analysing interleaved accessesKarl-Johan Karlsson2017-02-221-0/+37
* Increases full-unroll threshold.Dehao Chen2017-02-181-2/+12
* [LV] Remove constant restriction for vector phi creationMatthew Simpson2017-02-171-12/+89
* Reapply "[LV] Extend trunc optimization to all IVs with constant integer steps"Matthew Simpson2017-02-143-2/+64
OpenPOWER on IntegriCloud