| Naie_runtime | |
| CBdPool | |
| CTxnDeviceInfo | |
| Nhrx_test | |
| CContext | |
| CBuffer | |
| CLoadedKernel | |
| CChainRun | |
| Nllvm | |
| CDenseMapInfo< xilinx::AIE::ObjectFifoAcquireOp > | |
| CDenseMapInfo< xilinx::AIE::ObjectFifoCreateOp > | |
| CDenseMapInfo< xilinx::AIE::DMAChannel > | |
| CDenseMapInfo< xilinx::AIE::Port > | |
| CDenseMapInfo< xilinx::AIE::TileID > | |
| CGraphTraits< const WaitGraph * > | |
| CToNode | |
| Nstd | |
| Cless< xilinx::AIE::Port > | |
| Chash< xilinx::AIE::Port > | |
| Chash< xilinx::AIE::TileID > | |
| Ntest_utils | |
| CParameterScratchpad | |
| Nxilinx | |
| NAIEX | |
| CTokenAnalysis | |
| CBlockwriteData | |
| CSubviewTraceResult | |
| CAIEResourceStagesOptions | Options of aie-resource-stages, named and defaulted as aiecc's flags |
| CConstStridePolicy | |
| CSsaStridePolicy | |
| CBdTemplateFields | |
| CNdDmaPattern | Innermost-first ND access pattern (d0..d3 / repeat), matching the convention used by verifyStridesWraps and NpuDmaMemcpyNdOp verification |
| CTaskBdDecomposition | How aie-decompose-large-dma-bd splits a runtime task's descriptor |
| CDmaQueueModel | Tracks, per DMA channel, the pushes not yet known to have completed along one execution path |
| CChannelState | One channel's queue: its pushes not yet known to have completed, oldest first, each true when it issues a completion token |
| CQueueEffect | What one op in a loop body does to a channel queue |
| CDmaCompletionState | Completion is distinct from queue occupancy: Task_Queue_Size excludes the running task |
| CPush | |
| NAIE | |
| CResetConfig | |
| CMemoryRun | |
| CHasValidBDs | |
| CHasValidDMAChannels | |
| CSkipAccessibilityCheckTrait | |
| CAttrBasedSymbolVisibility | |
| CIsFlowEndPoint | |
| CNamedOpTable | |
| CNamedOpTableCollection | |
| CMyOffsetSizeAndStrideOpInterfaceTrait | |
| CMyOffsetSizeAndStrideOpInterface | |
| CTrait | |
| CStreamWindow | |
| CStreamWindowStep | |
| CDmaBdField | Bits [shift, shift + width) of register word in a buffer descriptor's register block |
| CDmaBdLayout | The register layout of one tile type's DMA buffer descriptor |
| CAIETargetModel | |
| CAIE1TargetModel | |
| CAIE2TargetModel | |
| CAIE2PSTargetModel | |
| CVE3858TargetModel | |
| CVC1902TargetModel | |
| CVE2302TargetModel | |
| CVE2802TargetModel | |
| CBaseNPU1TargetModel | |
| CVirtualizedNPU1TargetModel | |
| CBaseNPU2TargetModel | |
| CNPU2TargetModel | |
| CVirtualizedNPU2TargetModel | |
| CTileMemoryLimits | A memtile is reached by DMA rather than by core vector load and store, so its bus width also covers the vector-alignment requirement |
| CBufferAllocation | |
| CCoreTurns | The order one core moves objects through several pools, read off its program, so that one DMA channel can serve those pools in turn |
| CDMAChannelAnalysis | Which DMA channels of each tile are already spoken for, so that a channel is handed out at most once across everything that programs one |
| CPacketIdSpace | The packet ids a device has already spoken for |
| CSwitchboxConnect | |
| CCell | |
| CPathEndPoint | |
| CFlow | |
| CSwitchSetting | |
| CTreeHop | |
| CPacketsApart | |
| CTreeSplit | |
| CRoutingFaults | |
| CRouting | |
| CRoutingFailure | |
| CDeadlockProneRouting | |
| CPinnedRoutingFailure | |
| CPacketConstraints | |
| CPathfinder | |
| CRouteState | |
| CIdSplit | |
| CTreeBuilder | |
| CDynamicTileAnalysis | |
| CStreamDemand | |
| CTileAvailability | |
| CPlacer | |
| CAdjacency | |
| CCollectedOps | |
| CSequentialPlacer | |
| CSAConfig | |
| CSASchedule | |
| CNetBoundingBox | |
| CNetInfo | |
| CSAPlacer | |
| CShimTransferSpans | When the runtime sequences have each shim end's transfers in flight, so that ends never in flight together can take turns on one shim channel |
| CEnd | A shim end a runtime sequence transfers through |
| CStreamEndpoint | A tile port at the edge of the stream fabric |
| CTileDMAChannel | One DMA channel of one tile |
| CDmaChannelProgram | The BD chain a DMA channel runs, as one aie.dma_start, aie.dma or runtime task programs it |
| CStreamHop | A switchbox a stream passes: the port it enters by and, where it is packet switched there, the arbiter it takes |
| CRoutedStream | One stream from a source tile port to a destination tile port, recovered from the physical switchbox and shim-mux configuration |
| CHoldCycle | A cycle of waits packet streams can deadlock in, given their routes |
| CStep | |
| CStreamVolumeAnalysis | How much data a stream carries over a run and how much a receiver takes in before it has to wait on another agent, read off BD lengths, repeat counts, lock initial values and runtime transfers |
| CStreamWaitGraph | Which agent waits on which |
| CAgent | A core, a DMA channel, or a tile's controller |
| CStreamDeadlockAnalysis | Which streams can deadlock against each other if they share an arbiter or a link |
| CStreamConflicts | The streams a device asks for or already routes, and which pairs of them must not share an arbiter or a link |
| CStructuralGraph | Ops and the labelled links between them, for ranking them by structure where a pass would otherwise fall back on op order |
| CBitFieldInfo | Bit field information for a register |
| CRegisterInfo | Register information |
| CEventInfo | Event information |
| CRegisterDatabase | Register and event database for a specific architecture |
| CTxnLocBracket | |
| CAIERTControl | |
| CAIERtImpl | |
| CTxnLocEntry | |
| CNpuConfigCache | |
| CConfig | |
| Naievec | |
| CMatMulRecipe | |
| CShuffleModeShape | |
| CAIEVecType | |
| CCanonicalizeVectorForAIEVecOptions | Options for the "canonicalize-vector-for-aievec" pipeline |
| CLowerVectorToAIEVecOptions | Options for the "lower-vector-to-aievec" pipeline |
| COptimizeAIEVecOptions | Options for the "optimize-aievec" pipeline |
| CConvertVectorToAIEVecOptions | Options for the "convert-vector-to-aievec" pipeline |
| CBufferParams | |
| CAddOpConversion | |
| CAddElemOpAIE2Conversion | |
| CDecodedAddElemOp | |
| CSubElemOpAIE2Conversion | |
| CDecodedSubElemOp | |
| CAccFloatAddSubAIE2pConversion | |
| CNegOpAIE2Conversion | |
| CNegOpAIE2pConversion | |
| CSubOpConversion | |
| CFMAOpConversion | |
| CMulOpConversion | |
| CMulElemOpConversion | |
| CDecodedMulElemOp | |
| CMulElemOpAIE2pConversion | |
| CDecodedMulElemOp | |
| CFMAElemOpAIE2pConversion | |
| CUPSOpAIE2Conversion | |
| CUPSOpAIE2pConversion | |
| CSRSOpAIE2Conversion | |
| CSRSOpAIE2pConversion | |
| CUPDOpConversion | |
| CConcatOpConversion | |
| CExtOpConversion | |
| CExtOpAIE2pConversion | |
| CSelectOpConversion | |
| CPackOpConversion | |
| CUnpackOpConversion | |
| CUnpackI4OpAIE2pConversion | |
| CBroadcastOpConversion | |
| CMaxOpConversion | |
| CMinOpConversion | |
| CMaxOpAIE2pConversion | |
| CMinOpAIE2pConversion | |
| CCmpOpConversionBase | |
| CSelOpConversionBase | |
| CBroadcastScalarOpConversion | |
| CBroadcastScalarOpAIE2pConversion | |
| CShiftOpConversion | |
| CShiftOpAIE2pConversion | |
| CExtractElemOpConversion | |
| CFMAElemOpConversion | |
| CResolvedMatMulOperand | |
| CMatMulOpConversion | |
| CMatMulOpAIE2pConversion | |
| CFoldAIECastOps | |
| CFoldAIECastOpsAIE2p | |
| CShuffleOpConversion | |
| CShuffleOpAIE2pConversion | |
| CInvOpAIE2pConversion | |
| CExpOpAIE2pConversion | |
| CAccFloatUnaryOpAIE2pConversion | |
| CRsqrtOpAIE2pConversion | |
| COneLaneBroadcastOpConversion | |
| COneLaneInsertOpConversion | |
| CFdivOpConversion | |
| CExtractElemOpAIE2pConversion | |
| CConcatOpAIE2pConversion | |
| CConvertAIEVecToLLVMPass | |
| CShuffleGather | |
| CShuffleRequest | |
| CShufflePlan | |
| CNode | |
| CAIE2PShufflePlanner | |
| CAIEVecConvAnalysis | |
| CAccessesLocalLocks | |
| CAcquireReleaseOneStateInDMABlock | |
| CAIE2PElementwiseOps | |
| Caie_libxaie_ctx_t | |
| CAIEAssignBufferDescriptorIDsPass | |
| CAIEAssignCoreLinkFilesPass | |
| CAIEAssignLockIDsPass | |
| CAIEAssignTileCtrlIDsPass | |
| CAIEBroadcastPacketPass | |
| CAIECanonicalizeDevicePass | |
| CAIECreateCoresPass | |
| CAIECreateLocksPass | |
| CAIECtrlPacketInferTilesPass | |
| CAIECtrlPacketToDmaPass | |
| CAIEDMATasksToNPUPass | |
| CAIEFindFlowsPass | |
| CAIEGenerateColumnControlOverlayPass | |
| CAIELocalizeLocksPass | |
| CAIELowerCascadeFlowsPass | |
| CAIELowerCoreResetPass | |
| CAIELowerDmaChannelResetPass | |
| CAIELowerMemcpyPass | |
| CAIELowerMulticastPass | |
| CAIELowerSetLockPass | |
| CAIEMaterializeBDChainsPass | |
| CAIEMaterializeRuntimeSequencesPass | |
| CAIENormalizeAddressSpacesPass | |
| CAIENormalizeDmaBdDimsPass | |
| CAIEOpRemoval | |
| CAieRtControl | |
| CAIETransformBfpTypesPass | |
| CAIEVecConvOpTransformationPass | |
| CAIEVecTransformationPass | |
| CAIEXOpRemoval | |
| CAIEXToStandardPass | |
| Cargs | |
| CBankAwareContext | |
| CBdIdGenerator | |
| CBF16EmulationPass | |
| CBfpToIntegerConversionPattern | |
| CBfpToIntegerConverter | |
| CBroadcastLeadingDimsToRepeatPattern | |
| CCanonicalizeTrivialReadAccessSubviewOpPattern | Pattern to canonicalize trivial vector.transfer_read operations on subviews |
| CCanonicalizeTrivialWriteAccessSubviewOpPattern | Pattern to canonicalize trivial vector.transfer_write operations on subviews |
| CCanonicalizeVectorForAIEVecPass | |
| CCircuitFanoutSeed | |
| CComputeAbsOpPattern | |
| CComputeBandAndBorOpPattern | |
| CComputeBxorAndBnegOpPattern | |
| CComputeCeilOpPattern | |
| CComputeErfOpAIE2PPattern | |
| CComputeErfOpPattern | |
| CComputeExpOpByLUTLLVMPattern | |
| CComputeExpOpByLUTPattern | |
| CComputeFloorOpPattern | |
| CComputeNegOpPattern | |
| CComputeOneLaneMathOpAIE2PPattern | |
| CComputeRsqrtOpLLVMAIE2Pattern | |
| CComputeRsqrtOpPattern | |
| CComputeScalarMathOpByVectorAIE2PPattern | |
| CComputeSigmoidOpPattern | |
| CComputeSignedIntRightShiftOpPattern | |
| CComputeSqrtOpAIE2Pattern | |
| CComputeTanhOpByLUTPattern | |
| CConnectivityAnalysis | |
| CContractionTile | |
| CConversionPattern | |
| CConvertDivFToAIEVecInvOpPattern | |
| CConvertLeadingUnitDimInsertToReshapePattern | |
| CConvertMathExpToAIEVecExpOpPattern | |
| CConvertMathToAIEVecAccFloatUnaryOpPattern | |
| CConvertMulAddFToAIEVecFMAElemOpPattern | |
| CConvertMulAddToAIEVecFMAElemOpPattern | |
| CConvertMulAddToAIEVecFMAOpPattern | |
| CConvertMulFToAIEVecMulElemOpAIE2PPattern | |
| CConvertMulFToAIEVecMulElemOpPattern | |
| CConvertMulIToAIEVecMulElemOpPattern | |
| CConvertSplatToAIEBroadcast | |
| CConvertSplatToAIEBroadcastAIE2p | |
| CConvertSplatTransferReadToBroadcastPattern | |
| CConvertVectorFMAOpToAIEVecFMAElemOpAIE2PPattern | |
| CConvertVectorFMAOpToAIEVecFMAElemOpPattern | |
| CCoreResetToMaskWrite32Pattern | |
| Cd_ary_heap_indirect | |
| CDMAAwaitTaskOpPattern | |
| CDmaChannelResetToMaskWrite32Pattern | |
| CDMAInlineBDChainPattern | |
| CDMAStartBdChainForOpPattern | |
| CDMAStartTaskOpPattern | |
| CDynamicSizeNoImplicitBroadcastPass | |
| CDynamicSizeNoImplicitBroadcastPattern | |
| CEmulateBinaryF32InBF16Pattern | Pattern to emulate f32 binary vector arithmetic ops in bf16 |
| CEmulateCmpFF32InBF16Pattern | Pattern to emulate f32 comparison ops in bf16 |
| CEmulateFMAF32InBF16Pattern | Pattern to emulate f32 vector.fma with bf16 inputs |
| CEmulateMulFInBF16Pattern | Pattern to emulate an f32 multiply with bf16 inputs |
| CEmulateSelectF32InBF16Pattern | Pattern to emulate f32 select ops in bf16 |
| CEmulateUnaryF32InBF16Pattern | Pattern to emulate f32 unary vector ops in bf16 |
| CEventMonitor | |
| CExpandUPDToUPDAndExtPattern | |
| Cext_mem_model_t | |
| CExtendUPDOpsPass | |
| CFlattenElementwiseVectorToRank1Pattern | |
| CFlattenMultDimTransferReadPattern | |
| CFlattenMultDimTransferWritePattern | |
| CFlattenVectorIterArgsPattern | |
| CFoldAIEShiftAndBroadcast | |
| CFoldBF16RoundTripPattern | Pattern to drop bf16 -> f32 -> bf16 round trips |
| CFoldMulAddChainToConvOpPattern | |
| CFoldSplatToFMAOp | |
| CFoldVectorExtractAndSplatToAIEBroadcast | |
| CFuseExtIntoUPDPattern | |
| CFuseMulAddFInBF16Pattern | Pattern to fold an f32 multiply into the add that is its only user, as a vector.fma with bf16 inputs |
| CHasSomeParent | |
| CHasSomeTerminator | |
| CInBuf | |
| CInlineRuntimeCallsPattern | |
| CInsertLoadPdiForConfigurePattern | |
| CLiftedOps | |
| CLongestConvMACChainAnalysis | This analysis builds the longest possible chain of MAC operations whose operands are a vector that may or may not be shifted, and a broadcast |
| CConvMac | |
| CConvMacChainGroup | |
| CLowerAIEMemcpy | |
| CLowerExtOpPattern | |
| CLowerExtUIOfBitcastI4ToUnpackPattern | |
| CLowerScalarMinMaxToAIEVecMinMaxOp | |
| CLowerScalarShiftClampTruncToSRS | |
| CLowerScalarShRSIToAIEVecUPSSRS | |
| CLowerTruncOpPattern | |
| CLowerVectorAddIOpToAIEVecAddOp | |
| CLowerVectorAddOrSubOpToAIEVecAddElemOrSubElemOp | |
| CLowerVectorCmpOpToAIEVecCmpOp | |
| CLowerVectorContractionOpToAIEVecMatMulPattern | |
| CLowerVectorExtractStridedSliceOpAIE2Pattern | |
| CLowerVectorExtractStridedSliceOpAIEv1Pattern | |
| CLowerVectorMinMaxOpToAIEVecMinMaxOp | |
| CLowerVectorMulIOpToAIEVecMulOp | |
| CLowerVectorReductionAddBfloat16OpAIE2 | |
| CLowerVectorReductionAddBfloat16OpAIE2P | |
| CLowerVectorReductionAddFloatOp | |
| CLowerVectorReductionAddIntOp | |
| CLowerVectorReductionMaxOp | |
| CLowerVectorReductionMinOp | |
| CLowerVectorSelectOpToAIEVecSelOp | |
| CLowerVectorSIToFPI16BF16AIE2pPattern | |
| CLowerVectorToAIEVec | Lower incoming vector operations into their corresponding AIE vector intrinsics |
| CLowerVectorTransferReadToAIEUPD | |
| CLowerVectorTransposeOpToAIEVecShuffleOpPattern | |
| CLowerVectorUMinMaxOpToCmpSelOp | |
| CMergeSingleColumnI16FMAOpPattern | |
| CNarrowingResult | |
| CNormalizeLanesAIE2PPattern | |
| COneToOneVectorOpToAIEVecOpPattern | |
| COpRewritePattern | |
| COutBuf | |
| CPassWrapper | |
| CPinnedOverlay | What the prioritized flows take routed alone (see has_ctrl_pkt_overlay) |
| CPyAieTargetModel | |
| CRedundantLoadStoreOptimizationPass | |
| CRemoveAIECalls | |
| CRemoveAIEFuncs | |
| CReorderOperationsPass | |
| CRuntimeCallGraphCyclicityAnalysis | |
| CScalarizeVectorIntFpConversionOpPattern | |
| CSetInboundsToReadStoreOpPattern | |
| CSetLockToWrite32Pattern | |
| CShiftClampTruncToSRSPattern | |
| CSimplifyUPDOpsPass | |
| CSplitUnalignedTransferReadPattern | |
| CSwapUnaryOpsPattern | |
| CToken2LockLowering | |
| CUsesOneLockInDMABlock | |
| CVectorBroadcastLoweringPass | |
| CVia | |
1.15.0