summaryrefslogtreecommitdiff
path: root/test/Analysis
AgeCommit message (Expand)Author
2017-11-29[X86] Promote fp_to_sint v16f32->v16i16/v16i8 to avoid scalarization.Craig Topper
2017-11-22[SCEV] Strengthen variance condition in calculateLoopDispositionMax Kazantsev
2017-11-20Add heuristics for irreducible loop metadata under PGOHiroshi Yamauchi
2017-11-20[LV][X86] Support of AVX2 Gathers code generation and update the LV with thisMohammed Agabaria
2017-11-16Let llvm.invariant.group.barrier accepts pointer to any address spaceYaxun Liu
2017-11-16[TTI][X86] update costs of interleaved load\store of i64\doubleMohammed Agabaria
2017-11-15[Lint] Don't warn about passing alloca'd value to tail call if using byvalMikael Holmen
2017-11-14Simplify irreducible loop metadata test code.Hiroshi Yamauchi
2017-11-13[SCEV] Handling for ICmp occuring in the evolution chain.Jatin Bhateja
2017-11-06[LV][X86] update the cost of interleaving mem. access of floatsMohammed Agabaria
2017-11-05[REVERT][LV][X86] update the cost of interleaving mem. access of floatsMohammed Agabaria
2017-11-05[LV][X86] update the cost of interleaving mem. access of floatsMohammed Agabaria
2017-11-02Irreducible loop metadata for more accurate block frequency under PGO.Hiroshi Yamauchi
2017-11-02Allow inaccessiblememonly and inaccessiblemem_or_argmemonly to be overwriten ...Yichao Yu
2017-11-01[BranchProbabilityInfo] Handle irreducible loops.Geoff Berry
2017-10-25[SCEV] Fix an assertion failure in the max backedge taken countSanjoy Das
2017-10-24[ConstantFolding] Avoid assert when folding ptrtoint of vectorized GEPBjorn Pettersson
2017-10-18Revert "[ScalarEvolution] Handling for ICmp occuring in the evolution chain."Sanjoy Das
2017-10-18[AVX512][AVX2]Cost calculation for interleave load/store patterns {v8i8,v16i8...Michael Zuckerman
2017-10-18[ScalarEvolution] Handling for ICmp occuring in the evolution chain.Jatin Bhateja
2017-10-13[SCEV] Teach SCEV to find maxBECount when loop endbound is variantAnna Thomas
2017-10-13Revert r314923: "Recommit : Use the basic cost if a GEP is not used as addres...Daniel Jasper
2017-10-12[ValueTracking] return zero when there's conflict in known bits of a shift (P...Sanjay Patel
2017-10-04Convert an APInt to int64_t properly in TTI::getGEPCost().Justin Lebar
2017-10-04[TargetTransformInfo] Check if function pointer is valid before calling isLow...Guozhi Wei
2017-10-04Recommit : Use the basic cost if a GEP is not used as addressing modeJun Bum Lim
2017-10-03[Lint] Avoid failed assertion by fetching the proper pointer typeMikael Holmen
2017-10-03[X86] Add AVX512 check lines to the cost model truncate test.Craig Topper
2017-10-02[PassManager] Retire cl::opt that have been set for a while. NFCI.Davide Italiano
2017-10-02Move the stripping of invalid debug info from the Verifier to AutoUpgrade.Adrian Prantl
2017-09-30Revert r314435: "[JumpThreading] Preserve DT and LVI across the pass"Daniel Jasper
2017-09-29Revert "Use the basic cost if a GEP is not used as addressing mode"Alex Shlyapnikov
2017-09-29Use the basic cost if a GEP is not used as addressing modeJun Bum Lim
2017-09-28[JumpThreading] Preserve DT and LVI across the passEvandro Menezes
2017-09-27Check for overflows when calculating the offset in GetGEPCost.Justin Lebar
2017-09-22[TargetTransformInfo] Handle intrinsic call in getInstructionLatency()Guozhi Wei
2017-09-15[TargetTransformInfo] Static alloca has 0 costGuozhi Wei
2017-09-14[TargetTransformInfo] Detect 0 latency instructionsGuozhi Wei
2017-09-12[LAA] Allow more run-time alias checks by coercing pointer expressions to Add...Silviu Baranga
2017-09-08[TargetTransformInfo] Add a new public interface getInstructionCostGuozhi Wei
2017-09-07X86: Improve AVX512 fptoui loweringZvi Rackover
2017-09-01[SCEV] Add URem support to SCEVAlexandre Isoard
2017-08-31AMDGPU: Don't assert in TTI with fp32 denorms enabledMatt Arsenault
2017-08-16[CostModel][X86][XOP] Improve costs for XOP shufflesSimon Pilgrim
2017-08-15[Dominators] Include infinite loops in PostDominatorTreeJakub Kuderski
2017-08-14[ValueTracking] Don't delete assumes of side-effectful instructionsHal Finkel
2017-08-14[ValueTracking] Revert r310583 which enabled functionality that still isChandler Carruth
2017-08-10[CostModel][X86] Add SSE2 two-src shuffle costsSimon Pilgrim
2017-08-10[CostModel][X86] Add avx1 two-src shuffle costsSimon Pilgrim
2017-08-10[CostModel][X86] Add avx2 two-src shuffle costsSimon Pilgrim
generated by cgit v1.2.3 (git 2.25.1) at 2024-06-04 23:42:47 +0000