summaryrefslogtreecommitdiff
path: root/llvm/test/Transforms
AgeCommit message (Expand)Author
2025-11-22[DFAJumpThreading] Try harder to avoid cycles in paths. (#169151)Usman Nadeem
2025-11-22[InstCombine] Generalize trunc-shift-icmp fold from (1 << Y) to (Pow2 << Y) (...Pedro Lobo
2025-11-21[unroll-and-jam] Document dependencies_multidims.ll and fix loop bounds (NFC)...Sebastian Pop
2025-11-21AMDGPU: Improve getShuffleCost accuracy for 8- and 16-bit shuffles (#168818)Nicolai Hähnle
2025-11-21[profcheck] Propagate profile metadata to Wrapper function in optimize mode o...Jin Huang
2025-11-21[VPlan] Only apply forced cost to recipes with underlying values. (#168372)Florian Hahn
2025-11-21[LoopCacheAnalysis] Replace delinearization for fixed size array (#164798)Ryotaro Kasuga
2025-11-21[VPlan] Drop poison-generating flags on induction trunc (#168922)Ramkumar Ramachandra
2025-11-20[unroll-and-jam] Document dependency patterns in dependencies.ll (NFC) (#156577)Sebastian Pop
2025-11-20[InstSimplify] Extend icmp-of-add simplification to sle/sgt/sge (#168900)Pedro Lobo
2025-11-20[LV] Add test a low-trip count test without folding the tail.Florian Hahn
2025-11-20[VPlan] Remove PtrIV::IsScalarAfterVectorization, use VPlan analysis. (#168289)Florian Hahn
2025-11-20[SLP]Check if the non-schedulable phi parent node has unique operandsAlexey Bataev
2025-11-20[LV] Add tests for loops with low trip counts requiring tail-folding.Florian Hahn
2025-11-20[profcheck] Exclude `naked`, asm-only functions from profcheck (#168447)Mircea Trofin
2025-11-20[LV] Check full partial reduction chains in order. (#168036)Florian Hahn
2025-11-20[LoopPeel] Fix BFI when peeling last iteration without guard (#168250)Joel E. Denny
2025-11-20VectorCombine/AMDGPU: Cleanup a test and add a new one (#168817)Nicolai Hähnle
2025-11-20[LV] Allow partial reductions with an extended bin op (#165536)Sam Tebbs
2025-11-20AMDGPU: Autogenerate checks in a test (#168815)Nicolai Hähnle
2025-11-19Re-land [Transform][LoadStoreVectorizer] allow redundant in Chain (#168135)Gang Chen
2025-11-19[SLP]Fix insertion point for setting for the nodesAlexey Bataev
2025-11-19[ConstantFolding] Add constant folding for scalable vector interleave intrins...Craig Topper
2025-11-19[ConstantFolding] Generalize constant folding for vector_deinterleave2 to dei...Craig Topper
2025-11-19[SLPVectorizer] Widen constant strided loads. (#162324)Mikhail Gudim
2025-11-19[LV] Simplify existing load/store sink/hoisting tests, extend coverage.Florian Hahn
2025-11-19[InstSimplify] Add whitespace to struct declarations in vector-calls.ll. NFCCraig Topper
2025-11-19[LoopInterchange] Don't consider loops with BTC=0 (#167113)Sjoerd Meijer
2025-11-19[DA] Replace delinearization for fixed size array (#161822)Ryotaro Kasuga
2025-11-19[VPlan] Print debug info for all recipes. (#168454)Florian Hahn
2025-11-19[LV]: Skip Epilogue scalable VF greater than RemainingIterations. (#156724)Hassnaa Hamdi
2025-11-18[LTT] Mark as unkown weak function tests. (#167399)Mircea Trofin
2025-11-18[VPlan] VPIRFlags kind for FCmp with predicate + fast-math flags (NFCI).Florian Hahn
2025-11-18[VPlan] Fix OpType-mismatch in getFlagsFromIndDesc (#168560)Ramkumar Ramachandra
2025-11-18[AArch64] - Improve costing for Identity shuffles for SVE targets. (#165375)Pawan Nirpal
2025-11-18[ConstantFolding] Generalize constant folding for vector_interleave2 to inter...Craig Topper
2025-11-19[InstCombine] Canonicalize signed saturated additions (#153053)AZero13
2025-11-18[VPlan] Populate and use VPIRFlags from initial VPInstruction. (#168450)Florian Hahn
2025-11-18[LLVM][InstSimplify] Add folds for SVE integer reduction intrinsics. (#167519)Paul Walker
2025-11-18[LLVM][AArch64] Mark SVE integer intrinsics as speculatable. (#167915)Paul Walker
2025-11-18Extend MemoryEffects to Support Target-Specific Memory Locations (#148650)CarolineConcatto
2025-11-18[VPlan] Hoist loads with invariant addresses using noalias metadata. (#166247)Florian Hahn
2025-11-18[SLP] Invariant loads cannot have a memory dependency on stores. (#167929)Michael Bedy
2025-11-17InstCombine: Stop transforming EQ/NE of SHR to 0 to ULT/UGT if >1 usePeter Collingbourne
2025-11-17[MemProf] Fixup edges for largest N cold contexts (#167599)Teresa Johnson
2025-11-17[LV] Add test with existing noalias metadata and runtime checks.Florian Hahn
2025-11-17Reland [VPlan] Expand WidenInt inductions with nuw/nsw (#168354)Ramkumar Ramachandra
2025-11-17[VPlan] Fix LastActiveLane assertion on scalar VF (#167897)Luke Lau
2025-11-17[AArch64] Allow forcing unrolling of small loops (#167488)Vladi Krapp
2025-11-16[SLP]Do not consider split nodes, when checking parent PHI-based nodesAlexey Bataev
generated by cgit v1.2.3 (git 2.25.1) at 2026-05-31 17:08:15 +0000