pfk-beta commented on PR #11089: URL: https://github.com/apache/tvm/pull/11089#issuecomment-1106218397
> Not sure why the tests are failing :-( I'm not expert in reading test output of TVM, but when I open full logs(https://ci.tlcpack.ai/blue/rest/organizations/jenkins/pipelines/tvm/branches/PR-11089/runs/1/nodes/348/steps/786/log/?start=0) I can see on this timestamp (2022-04-21T11:25:28.044Z) huger error, here is only part: ``` [2022-04-21T11:25:28.044Z] DEBUG:tvm.autotvm.tuner.tuner:No: 19 GFLOPS: 0.00/93.74 result: Traceback (most recent call last): [2022-04-21T11:25:28.044Z] File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 721, in __call__ [2022-04-21T11:25:28.044Z] yield remote, remote.load_module(os.path.split(build_result.filename)[1]) [2022-04-21T11:25:28.044Z] File "/workspace/python/tvm/autotvm/measure/measure_methods.py", line 685, in run_through_rpc [2022-04-21T11:25:28.044Z] costs = time_f(*args).results [2022-04-21T11:25:28.044Z] File "/workspace/python/tvm/runtime/module.py", line 297, in evaluator [2022-04-21T11:25:28.044Z] blob = feval(*args) [2022-04-21T11:25:28.044Z] File "tvm/_ffi/_cython/./packed_func.pxi", line 323, in tvm._ffi._cy3.core.PackedFuncBase.__call__ [2022-04-21T11:25:28.044Z] File "tvm/_ffi/_cython/./packed_func.pxi", line 257, in tvm._ffi._cy3.core.FuncCall [2022-04-21T11:25:28.044Z] File "tvm/_ffi/_cython/./packed_func.pxi", line 246, in tvm._ffi._cy3.core.FuncCall3 [2022-04-21T11:25:28.044Z] File "tvm/_ffi/_cython/./base.pxi", line 163, in tvm._ffi._cy3.core.CALL [2022-04-21T11:25:28.044Z] tvm._ffi.base.TVMError: Traceback (most recent call last): [2022-04-21T11:25:28.044Z] 4: TVMFuncCall [2022-04-21T11:25:28.044Z] at /workspace/src/runtime/c_runtime_api.cc:477 [2022-04-21T11:25:28.044Z] 3: tvm::runtime::PackedFuncObj::CallPacked(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const [2022-04-21T11:25:28.044Z] at /workspace/include/tvm/runtime/packed_func.h:1217 [2022-04-21T11:25:28.044Z] 2: tvm::runtime::RPCWrappedFunc::operator()(tvm::runtime::TVMArgs, tvm::runtime::TVMRetValue*) const [2022-04-21T11:25:28.044Z] at /workspace/src/runtime/rpc/rpc_module.cc:127 [2022-04-21T11:25:28.044Z] 1: tvm::runtime::RPCClientSession::CallFunc(void*, TVMValue const*, int const*, int, std::function<void (tvm::runtime::TVMArgs)> const&) [2022-04-21T11:25:28.044Z] at /workspace/src/runtime/rpc/rpc_endpoint.cc:1009 [2022-04-21T11:25:28.044Z] 0: tvm::runtime::RPCEndpoint::CallFunc(void*, TVMValue const*, int const*, int, std::function<void (tvm::runtime::TVMArgs)>) [2022-04-21T11:25:28.044Z] at /workspace/src/runtime/rpc/rpc_endpoint.cc:801 [2022-04-21T11:25:28.044Z] File "/workspace/src/runtime/rpc/rpc_endpoint.cc", line 801 [2022-04-21T11:25:28.044Z] TVMError: [2022-04-21T11:25:28.044Z] --------------------------------------------------------------- [2022-04-21T11:25:28.044Z] An error occurred during the execution of TVM. [2022-04-21T11:25:28.044Z] For more information, please see: https://tvm.apache.org/docs/errors.html [2022-04-21T11:25:28.044Z] --------------------------------------------------------------- [2022-04-21T11:25:28.044Z] Check failed: (code == RPCCode::kReturn) is false: code=kShutdown [2022-04-21T11:25:28.044Z] [2022-04-21T11:25:28.044Z] During handling of the above exception, another exception occurred: ``` > Let's leave this PR open for 1 week or 2 just in case there is any legit concern :-) I think 05.05 may be good date for merging :D > 1. ./autotvm/graph_tuner/base_graph_tuner.py: self._logger = logging.getLogger(name + "_logger") ./meta_schedule/testing BaseGraphTuner which has such construction, is inherited by 2 different GraphTuners, which are located in separate files. I suggest to use following logger name: `__name__ + "__" + name`, or remove `_logger` field from `BaseGraphTuner` and its children. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
