aboutsummaryrefslogtreecommitdiff
path: root/llvm/test/Transforms/PhaseOrdering
AgeCommit message (Expand)AuthorFilesLines
2 daysRevert "[VPlan] Remove unused VPExpandSCEVRecipe before expansion" (#184108)Jan Patrick Lehr1-1/+1
3 days[VPlan] Remove unused VPExpandSCEVRecipe before expansion (#181329)Mel Chen1-1/+1
6 daysRevert "[VPlan] Don't drop NUW flag on tail folded canonical IVs (#183301)" (...Luke Lau1-1/+1
6 days[VPlan] Don't drop NUW flag on tail folded canonical IVs (#183301)Luke Lau1-1/+1
6 daysAMDGPU: Stop adding uniform-work-group-size=false (#183502)Matt Arsenault1-3/+3
8 days[VPlan] Allow recursive narrowing in interleave group narrowing. (#167310)Florian Hahn1-12/+66
10 daysRevert "[InstCombine] Allow freezing multiple operands (#154336)" (#182769)Eli Friedman1-6/+6
2026-02-11[LAA] Use SCEVPtrToAddr in tryToCreateDiffChecks. (#178861)Florian Hahn2-4/+4
2026-02-10InstCombine: Use SimplifyDemandedFPClass on fmul (#177490)Matt Arsenault1-6/+6
2026-02-09[InstCombine] Drop nonnull assumes if the pointer is already known to be nonn...Nikolas Klauser1-1/+0
2026-02-09[PhaseOrdering] Regenerate test checks (NFC)Nikita Popov1-19/+25
2026-02-05[SLP]Remove LoadCombine workaround after handling of the copyablesAlexey Bataev1-204/+12
2026-02-05Revert "[SLP]Remove LoadCombine workaround after handling of the copyables"Alexey Bataev1-12/+204
2026-02-05[SLP]Remove LoadCombine workaround after handling of the copyablesAlexey Bataev1-204/+12
2026-02-03[InstCombine] fold icmp ne (and X, 1), 0 --> trunc X to i1 (#178977)Andreas Jonson1-2/+2
2026-01-26[SLP]Model disjoint or reduction of shl(zext, (0, stride, 2* stride)) as bitcastAlexey Bataev1-6/+2
2026-01-23[LV] Separate runtime check cost from total overhead in profitability check (...Mel Chen1-1/+1
2026-01-13[VPlan] Optimize BranchOnTwoConds to chain of 2 simple branches. (#174016)Florian Hahn1-24/+134
2026-01-11[ValueTracking] Support horizontal vector add in computeKnownBits (#174410)Valeriy Savchenko1-10/+10
2026-01-06Reland [VPlan] Simplify pow-of-2 (mul|udiv) -> (shl|lshr) (#174581)Ramkumar Ramachandra1-1/+1
2026-01-06Revert "[VPlan] Simplify pow-of-2 (mul|udiv) -> (shl|lshr)" (#174559)Alex Bradbury1-1/+1
2026-01-06[VPlan] Simplify pow-of-2 (mul|udiv) -> (shl|lshr) (#172477)Ramkumar Ramachandra1-1/+1
2026-01-04[IR] Reland Optimize PHINode::removeIncomingValue() and PHINode::removeIncomi...Mingjie Xu3-4/+4
2026-01-01[SLP] Support for copyables in the reduced values (#153589)Alexey Bataev4-151/+48
2026-01-01Revert "[SLP] Support for copyables in the reduced values (#153589)"Alexey Bataev4-48/+151
2026-01-01[SLP] Support for copyables in the reduced values (#153589)Alexey Bataev4-151/+48
2025-12-30[SLP]Exclude non-profitable subtrees.Alexey Bataev1-5/+4
2025-12-30Revert "[SLP]Exclude non-profitable subtrees."Alexey Bataev1-4/+5
2025-12-29Revert 159f1c048e08a8780d92858cfc80e723c90235e3 (#173893)Walter Lee3-4/+4
2025-12-29[SLP]Exclude non-profitable subtrees.Alexey Bataev1-5/+4
2025-12-18[IR] Update `PHINode::removeIncomingValueIf()` to use the swap strategy like ...Mingjie Xu1-1/+1
2025-12-17[IR] Optimize PHINode::removeIncomingValue() by swapping removed incoming val...Mingjie Xu4-5/+5
2025-12-08[VPlan] Use BlockFrequencyInfo in getPredBlockCostDivisor (#158690)Luke Lau2-1/+62
2025-12-07[Clang]: Support opt-in speculative devirtualization (#159685)Hassnaa Hamdi1-0/+60
2025-12-03Utils: Inhibit load/store folding through phis for llvm.protected.field.ptr.Peter Collingbourne1-0/+38
2025-11-24[VPlan] Set ZeroIsPoison=false for FirstActiveLane (#169298)Luke Lau1-8/+9
2025-11-18[VPlan] Hoist loads with invariant addresses using noalias metadata. (#166247)Florian Hahn1-8/+58
2025-11-17InstCombine: Stop transforming EQ/NE of SHR to 0 to ULT/UGT if >1 usePeter Collingbourne2-9/+7
2025-11-14Revert "[LICM] Sink unused l-invariant loads in preheader. #157559"Alexander Belyaev6-46/+44
2025-11-12[VectorCombine] Try to scalarize vector loads feeding bitcast instructions. (...Julian Nagele1-0/+32
2025-11-10[DropAssumes] Drop dereferenceable assumptions after vectorization. (#166947)Florian Hahn2-73/+3
2025-11-09[InstCombine] Don't sink if it would require dropping deref assumptions. (#16...Florian Hahn1-56/+46
2025-11-07[VectorCombine] support mismatching extract/insert indices for foldInsExtFNeg...hanbeom2-8/+4
2025-11-07[InstCombnine] Add test for sinking with dereferneceable assumes.Florian Hahn1-16/+125
2025-11-06Revert "[InterleavedAccess] Construct interleaved access store with shuffles"Martin Storsjö1-8/+8
2025-11-05[InterleavedAccess] Construct interleaved access store with shufflesRamkrishnan1-8/+8
2025-11-05[InstCombine] Enable FoldOpIntoSelect and foldOpIntoPhi when the Op's other p...Gábor Spaits1-52/+53
2025-11-02[LV] Add tests with hoist-able invariant loads.Florian Hahn1-0/+46
2025-10-30[LICM] Sink unused l-invariant loads in preheader. (#157559)Vigneshwar Jayakumar6-41/+42
2025-10-24[test][Transforms] Remove unsafe-fp-math uses part 3 (NFC) (#164787)paperchalice11-136/+110
generated by cgit v1.1 at 2026-03-04 23:03:22 +0000