use toolchain binaries

use toolchain binaries
[2026-06-29 08:31:05] [debug] [Thread: 668] [/projects_src/kneron-piano_v2/dynasty/floating_point/floating_point/src/common/BaseInferencerImpl.cpp:63]
start to create operators....
[2026-06-29 08:31:05] [debug] [Thread: 668] [/projects_src/kneron-piano_v2/dynasty/floating_point/floating_point/src/common/BaseInferencerImpl.cpp:102]
The model allocated space: 8708.854288 Mbytes( including workspace: 0.0 Mbytes)
[array([[[ 1.9232708 , -1.0427347 ,  2.2663233 , ...,  4.9218965 ,
         -4.8801312 , -0.3491888 ],
        [ 0.04256073, -1.0873784 ,  2.9717665 , ...,  3.358795  ,
         -1.6552099 ,  3.533231  ],
        [ 2.6577165 ,  0.6868226 ,  0.84456134, ...,  3.9475377 ,
         -0.95962983,  1.2934611 ],
        ...,
        [ 2.054679  ,  0.98919183,  0.19099367, ...,  2.5516078 ,
          0.30415085,  5.344432  ],
        [ 6.1130185 ,  4.0604615 ,  3.188645  , ...,  5.3276134 ,
          1.3165636 , -0.81487703],
        [ 2.9077537 , -1.2325453 ,  1.8155068 , ...,  4.582607  ,
         -4.8986917 ,  3.3238902 ]]], dtype=float32)]
/data1/kneron_flow/input.kdp730.scaled.release.bie
ERROR:root:Analysis is required before compile.
[common][info][mem_guard.cc:122][SetMemoryCapLimit] set memory usage limit to [29953] MB
[tool][info][batch_compile.cc:375][main] Use compiler built by commit [f91f2bb2b1]
[tool][info][batch_compile.cc:376][main] Use piano built by commit [1a7b5c8]
[tool][info][batch_compile.cc:163][BatchCompile] compiling input.onnx
[common][info][compile.cc:65][LoadCfg] Loading config.
[piano][info][graph_gen.cc:26][LoadOnnxModel] Loading onnx file.
[piano][info][graph_gen.cc:55][LoadOnnxModel] Creating graph from onnx.
[piano][debug][graph.cc:394][constructGraph] Start constructing graph.
[common][info][compile_utils.cc:585][OptimizeGraph] (FE) optimize graph [1/1] (of 554 nodes)
[piano][info][graph_gen.cc:135][GraphOptBasic] Working on basic graph optimization.
[graph][info][middle_api.cc:11][IntroduceMiddleInfo] Working on hardware relevant information.
[graph][info][graph_opt_fe.cc:29][GraphOptFeHardware] Working on front-end graph optimization.
[common][info][compile_utils.cc:604][OptimizeGraph] (BE) optimize graph [1/1] (of 2400 nodes)
[common][info][tgt_graph_gen.cc:60][GraphOptBE] Working on hardware relevant optimizations.
[common][info][knerex_fp.cc:168][PatchKnerexInfo] patch FP info.
[common][debug][tgt_knerex_fp.cc:45][AddDefaultFpInfo] Cannot found fixed-point info. using default settings!
[graph][warning][cpu_judgement.cc:129][LogReason] CPU op: Reshape [/connector/Reshape_1], reason: [in/out data cannot fit into SRAM with current feature map cutting implementation]
[common][info][tgt_post_process.cc:160][WagnerPostProcess] Working on graph post process
[piano][debug][generate_onnx.cc:83][GenerateDebugOnnxFromGraph] Export debug ONNX to [debug.onnx]
[common][info][compile.cc:248][CompileSubGraph] Lowering IR (of 2957 nodes)
[cmd][debug][node_gen_utils.cc:223][CollectWeights] Totally 2133 weight nodes found in graph.
[cmd][info][cmd_node_gen.cc:591][CutAnalysis] Cutting analysis
[fmcut][info][img_cut.cc:105][ImageCut] Start image cut with mode [default]
[fmcut][debug][img_cut_tool.cc:752][operator()] cannot cut group [25] by [Row], try by [Col]
[fmcut][debug][img_cut_tool.cc:752][operator()] cannot cut group [72] by [Row], try by [Col]
[fmcut][debug][img_cut_tool.cc:752][operator()] cannot cut group [119] by [Row], try by [Col]
[fmcut][debug][img_cut_tool.cc:752][operator()] cannot cut group [166] by [Row], try by [Col]
[fmcut][debug][img_cut_tool.cc:752][operator()] cannot cut group [213] by [Row], try by [Col]
[fmcut][debug][img_cut_tool.cc:752][operator()] cannot cut group [260] by [Row], try by [Col]
[fmcut][debug][img_cut_tool.cc:752][operator()] cannot cut group [307] by [Row], try by [Col]
[fmcut][debug][img_cut_tool.cc:752][operator()] cannot cut group [354] by [Row], try by [Col]
[fmcut][debug][img_cut_tool.cc:752][operator()] cannot cut group [401] by [Row], try by [Col]
[fmcut][debug][img_cut_tool.cc:752][operator()] cannot cut group [448] by [Row], try by [Col]
[fmcut][debug][img_cut_tool.cc:752][operator()] cannot cut group [495] by [Row], try by [Col]
[fmcut][debug][img_cut_tool.cc:752][operator()] cannot cut group [542] by [Row], try by [Col]
[fmcut][debug][img_cut_tool.cc:752][operator()] cannot cut group [581] by [Col], try by [Row]
[fmcut][debug][img_cut_tool.cc:752][operator()] cannot cut group [581] by [Row], try by [Channel]
[fmcut][debug][img_cut_tool.cc:762][operator()] cannot cut group [581] by [Channel], split group and try by [Col]
[cmd][info][cmd_node_gen.cc:619][CutAnalysis] Create 15223 cmd node(s) after FM cut analysis
[mmu][info][mmu.cc:1921][MemoryManagement] Allocationg memory for 15459 nodes
[mmu][info][mmu.cc:806][LogStat] CNODE: cnt [15459], npu [15458], cpu [1]
[mmu][info][mmu.cc:809][LogStat] DRAM: in buf size [4194304]
[mmu][info][mmu.cc:810][LogStat] DRAM: out buf size [61440]
[mmu][info][mmu.cc:811][LogStat] DRAM: work buf size [26738688]
[mmu][info][mmu.cc:812][LogStat] DRAM: const_in buf size [786432]
[mmu][info][mmu.cc:814][LogStat] RDMA: cnt [4018], opt_cnt [2447], size [1398816768]
[mmu][info][mmu.cc:815][LogStat] WDMA: cnt [3003], size [881537024]
[mmu][info][mmu.cc:816][LogStat] PFUNC: cnt [1154]
[mmu][info][mmu.cc:817][LogStat] Inplace OP: cnt [5765]
[tiling][info][tiling.cc:470][TileAnalysis] Running tiling analysis
[tiling][info][tiling.cc:490][TileAnalysis] Create additional 239180 cmd node(s) after tiling analysis
[fmcut][info][multi_core.cc:147][SetMultiCoreCnt] Running multi-core analysis
[common][info][compile.cc:256][CompileSubGraph] Generating weight
[wt][debug][wt_generator.cc:639][UpdateMatMulConv1x1WtInfoAfterCut] UpdateMatMulConv1x1WtInfoAfterCut
[wt][debug][wt_generator.cc:728][UpdateWtRefNode] UpdateWtRefNode
[wt][info][wt_generator.cc:517][ProcessWtInfo] ProcessWtInfo
[wt][debug][wt_generator.cc:431][MergeContWtInfo] merge continuous weight infos
[common][info][compile.cc:259][CompileSubGraph] Generating command
[cmd][info][cmd_generator.cc:676][Generate] Generate commands based on [252012] cmd nodes
[cmd][info][cmd_generator.cc:686][Generate] 22371678 out of 25049702 CONF cmds are optimized out
[cmd][info][cmd_generator.cc:687][Generate] Total GETW: [661938624] bytes, refetch: [35603344] bytes
[common][debug][addr_info.cc:156][CheckMemRegion] Memory region [input]: [0x80000000-0x80400000]
[common][debug][addr_info.cc:156][CheckMemRegion] Memory region [output]: [0x80400000-0x8040f000]
[common][debug][addr_info.cc:156][CheckMemRegion] Memory region [cmd]: [0x8238f000-0x830b7fd8]
[common][debug][addr_info.cc:156][CheckMemRegion] Memory region [weight]: [0x830b8000-0x8c66d220]
[common][debug][addr_info.cc:156][CheckMemRegion] Memory region [dram]: [0x8040f000-0x81d8f000]
[common][debug][addr_info.cc:156][CheckMemRegion] Memory region [cpu_tmp]: [0x81d8f000-0x8238f000]
[common][info][compile.cc:275][CompileSubGraph] input [2], output [1], wt [156979744], cmd [13799384],  in_buf [4194304], out_buf [61440], data_buf [26738688]
[common][info][compile.cc:288][CompileSubGraph] Compilation completed.
[common][debug][addr_info.cc:156][CheckMemRegion] Memory region [input]: [0x80000000-0x80000000]
[common][debug][addr_info.cc:156][CheckMemRegion] Memory region [output]: [0x80400000-0x80400000]
[common][debug][addr_info.cc:156][CheckMemRegion] Memory region [cmd]: [0x8238f000-0x8238f000]
[common][debug][addr_info.cc:156][CheckMemRegion] Memory region [weight]: [0x830b8000-0x830b8000]
[common][debug][addr_info.cc:156][CheckMemRegion] Memory region [dram]: [0x8040f000-0x8040f000]
[common][debug][addr_info.cc:156][CheckMemRegion] Memory region [cpu_tmp]: [0x81d8f000-0x81d8f000]
[kne][info][kne_generator.cc:123][CountCoreNumber] Core statics: npu [1], cpu [1], total [1]
[tool][debug][batch_compile.cc:181][BatchCompile] max input size:0x00000000
[tool][debug][batch_compile.cc:182][BatchCompile] max output size:0x00000000
[tool][debug][batch_compile.cc:183][BatchCompile] max wbuf size:0x00000000
[common][debug][addr_info.cc:156][CheckMemRegion] Memory region [input]: [0x80000000-0x80400000]
[common][debug][addr_info.cc:156][CheckMemRegion] Memory region [output]: [0x80400000-0x8040f000]
[common][debug][addr_info.cc:156][CheckMemRegion] Memory region [cmd]: [0x8238f000-0x8238f000]
[common][debug][addr_info.cc:156][CheckMemRegion] Memory region [weight]: [0x8238f000-0x8238f000]
[common][debug][addr_info.cc:156][CheckMemRegion] Memory region [dram]: [0x8040f000-0x81d8f000]
[common][debug][addr_info.cc:156][CheckMemRegion] Memory region [cpu_tmp]: [0x81d8f000-0x8238f000]
models_730.kne
[tool][info][batch_compile.cc:402][main] batch_compile complete
/data1/kneron_flow/models_730.nef
NEF file with updated model info has been generated in:
/data1/kneron_flow/models_730.nef
/data1/kneron_flow/models_730.nef
The number of input node names specified is less than the number of provided preprocess results.
