search

apache / tvm

Open deep learning compiler stack for cpu, gpu and specialized accelerators
https://tvm.apache.org/
Apache License 2.0
11.67k stars 3.45k forks source link

[Bug] Can't run prequantized tflite model through MicroTVM #11371

Open zhiwei-dong opened 2 years ago

zhiwei-dong commented 2 years ago

Expected behavior

Thank for your great works on microTVM, I try to run MCUNet model, in here . I followed this tutorial. When I use the tutorial model, every thing is ok, but when I use above mcunet model, I encounter RPCError.

Actual behavior

(microtvm-3BvVWFka-py3.6) vagrant@microtvm:/Users/zhiwei.dong/test/tvm$ tvmc run \

     --device micro \
     project \
     --project-option zephyr_board=qemu_x86 \
     --fill-mode ones \
     --print-top 4

INFO:main:b'[ 2%] Built target parse_syscalls_target\n' INFO:main:b'[ 3%] Built target kobj_types_h_target\n' INFO:main:b'[ 4%] Built target syscall_list_h_target\n' INFO:main:b'[ 5%] Built target driver_validation_h_target\n' INFO:main:b'Consolidate compiler generated dependencies of target offsets\n' INFO:main:b'[ 6%] Built target offsets\n' INFO:main:b'[ 7%] Built target offsets_h\n' INFO:main:b'[ 7%] Built target zephyr_generated_headers\n' INFO:main:b'Consolidate compiler generated dependencies of target kernel\n' INFO:main:b'[ 27%] Built target kernel\n' INFO:main:b'Consolidate compiler generated dependencies of target microtvm_rpc_server\n' INFO:main:b'[ 28%] Built target microtvm_rpc_server\n' INFO:main:b'Consolidate compiler generated dependencies of target microtvm_rpc_common\n' INFO:main:b'[ 32%] Built target microtvm_rpc_common\n' INFO:main:b'Consolidate compiler generated dependencies of target common\n' INFO:main:b'[ 36%] Built target common\n' INFO:main:b'Consolidate compiler generated dependencies of target tvm_model\n' INFO:main:b'[ 39%] Built target tvm_model\n' INFO:main:b'Consolidate compiler generated dependencies of target app\n' INFO:main:b'[ 40%] Built target app\n' INFO:main:b'Consolidate compiler generated dependencies of target zephyr\n' INFO:main:b'[ 62%] Built target zephyr\n' INFO:main:b'[ 63%] Built target linker_zephyr_prebuilt_script_target\n' INFO:main:b'Consolidate compiler generated dependencies of target archx86core\n' INFO:main:b'[ 77%] Built target archx86core\n' INFO:main:b'Consolidate compiler generated dependencies of target liblibcnewlib\n' INFO:main:b'[ 78%] Built target liblibcnewlib\n' INFO:main:b'Consolidate compiler generated dependencies of target libposix\n' INFO:main:b'[ 80%] Built target libposix\n' INFO:main:b'Consolidate compiler generated dependencies of target subsysrandom\n' INFO:main:b'[ 81%] Built target subsysrandom\n' INFO:main:b'Consolidate compiler generated dependencies of target driversinterrupt_controller\n' INFO:main:b'[ 84%] Built target drivers__interrupt_controller\n' INFO:main:b'Consolidate compiler generated dependencies of target driversserial\n' INFO:main:b'[ 86%] Built target driversserial\n' INFO:main:b'Consolidate compiler generated dependencies of target driverstimer\n' INFO:main:b'[ 88%] Built target driverstimer\n' INFO:main:b'Consolidate compiler generated dependencies of target zephyr_prebuilt\n' INFO:main:b'[ 90%] Built target zephyr_prebuilt\n' INFO:main:b'[ 91%] Built target linker_zephyr_final_script_target\n' INFO:main:b'[ 91%] Built target gen_idt_output\n' INFO:main:b'[ 92%] Built target staticIdt_o\n' INFO:main:b'[ 93%] Built target irq_int_vector_map_o\n' INFO:main:b'[ 94%] Built target irq_vectors_alloc_o\n' INFO:main:b'[ 95%] Built target pagetables_bin_target\n' INFO:main:b'[ 95%] Built target pagetables_o\n' INFO:main:b'Consolidate compiler generated dependencies of target zephyr_final\n' INFO:main:b'[ 98%] Built target zephyr_final\n' INFO:main:b'[100%] [QEMU] CPU: qemu32,+nx,+pae\n' [14:54:33] /Users/zhiwei.dong/test/tvm/src/runtime/micro/micro_session.cc:367: remote: microTVM Zephyr runtime - running INFO:main:b'[100%] Built target run\n' Traceback (most recent call last): File "/usr/lib/python3.6/runpy.py", line 193, in _run_module_as_main "main", mod_spec) File "/usr/lib/python3.6/runpy.py", line 85, in _run_code exec(code, run_globals) File "/Users/zhiwei.dong/test/tvm/python/tvm/driver/tvmc/main.py", line 24, in tvmc.main.main() File "/Users/zhiwei.dong/test/tvm/python/tvm/driver/tvmc/main.py", line 115, in main sys.exit(_main(sys.argv[1:])) File "/Users/zhiwei.dong/test/tvm/python/tvm/driver/tvmc/main.py", line 103, in _main return args.func(args) File "/Users/zhiwei.dong/test/tvm/python/tvm/driver/tvmc/runner.py", line 277, in drive_run options=options, File "/Users/zhiwei.dong/test/tvm/python/tvm/driver/tvmc/runner.py", line 580, in run_module module = tvm.micro.create_local_graph_executor(tvmc_package.graph, lib, dev) File "/Users/zhiwei.dong/test/tvm/python/tvm/micro/session.py", line 214, in create_local_graph_executor fcreate(graph_json_str, mod, lookup_remote_linked_param, device_type_id) File "/Users/zhiwei.dong/test/tvm/python/tvm/_ffi/_ctypes/packed_func.py", line 237, in call raise get_last_ffi_error() tvm.error.RPCError: Traceback (most recent call last): 12: TVMFuncCall 11: tvm::runtime::PackedFuncObj::Extractor<tvm::runtime::PackedFuncSubObj<tvm::runtime::{lambda(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue)#1}> >::Call(tvm::runtime::PackedFuncObj const, tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue) 10: tvm::runtime::GraphExecutorCreate(std::cxx11::basic_string<char, std::char_traits, std::allocator > const&, tvm::runtime::Module const&, std::vector<DLDevice, std::allocator > const&, tvm::runtime::PackedFunc) 9: tvm::runtime::GraphExecutor::Init(std::__cxx11::basic_string<char, std::char_traits, std::allocator > const&, tvm::runtime::Module, std::vector<DLDevice, std::allocator > const&, tvm::runtime::PackedFunc) 8: tvm::runtime::GraphExecutor::SetupStorage() 7: tvm::runtime::NDArray::Empty(tvm::runtime::ShapeTuple, DLDataType, DLDevice, tvm::runtime::Optional) 6: tvm::runtime::RPCDeviceAPI::AllocDataSpace(DLDevice, int, long const, DLDataType, tvm::runtime::Optional) 5: tvm::runtime::RPCClientSession::AllocDataSpace(DLDevice, int, long const, DLDataType, tvm::runtime::Optional) 4: tvm::runtime::PackedFuncObj::Extractor<tvm::runtime::PackedFuncSubObj<tvm::runtime::RPCEndpoint::Init()::{lambda(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue)#2}> >::Call(tvm::runtime::PackedFuncObj const, tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) 3: tvm::runtime::RPCEndpoint::HandleUntilReturnEvent(bool, std::function<void (tvm::runtime::TVMArgs)>) 2: tvm::runtime::RPCEndpoint::EventHandler::HandleNextEvent(bool, bool, std::function<void (tvm::runtime::TVMArgs)>) 1: tvm::runtime::RPCEndpoint::EventHandler::HandleProcessPacket(std::function<void (tvm::runtime::TVMArgs)>) 0: tvm::runtime::RPCEndpoint::EventHandler::HandleReturn(tvm::runtime::RPCCode, std::function<void (tvm::runtime::TVMArgs)>) File "/Users/zhiwei.dong/test/tvm/src/runtime/rpc/rpc_endpoint.cc", line 376 RPCError: Error caught from RPC call:

Environment

Reference VM is my environment.

Steps to reproduce

Just download my model, and follow above tutorial, error will present when executing dvmc run cmd.

Theoo1997 commented 2 years ago

Hi, I face the same problem. Do we have any update about the issue?

zhiwei-dong commented 2 years ago

no, but we can contact using email for more details, dongz.cn@outlook.con