Open wyq-carol opened 1 year ago
I followed instructions in README.txt and compiled pytorch v1.12.0 successfully but failed after applying the patch. I got error message as followed.
Traceback (most recent call last): File "/root/anaconda3/envs/env_gnn/lib/python3.8/runpy.py", line 194, in _run_module_as_main return _run_code(code, main_globals, None, File "/root/anaconda3/envs/env_gnn/lib/python3.8/runpy.py", line 87, in _run_code exec(code, run_globals) File "/home/pytorch/torchgen/gen.py", line 2535, in <module> main() File "/home/pytorch/torchgen/gen.py", line 2393, in main parsed_yaml = parse_native_yaml(native_yaml_path, tags_yaml_path, ignore_keys) File "/home/pytorch/torchgen/gen.py", line 237, in parse_native_yaml _GLOBAL_PARSE_NATIVE_YAML_CACHE[path] = parse_native_yaml_struct( File "/home/pytorch/torchgen/gen.py", line 174, in parse_native_yaml_struct func, m = NativeFunction.from_yaml(e, loc, valid_tags, ignore_keys) File "/home/pytorch/torchgen/model.py", line 548, in from_yaml dispatch_key = DispatchKey.parse(k.strip()) File "/home/pytorch/torchgen/model.py", line 142, in parse raise AssertionError(f"unknown dispatch key {value}") AssertionError: unknown dispatch key Checkpoint in /home/pytorch/cmake/../aten/src/ATen/native/native_functions.yaml:476: add.Tensor(Tensor self, Tensor other, *, Scalar alpha=1) -> Tensor ninja: error: rebuilding 'build.ninja': subcommand failed
I tried hard to fix the bugs but the project is so complex. I finally got error message as followed.
FAILED: bin/torch_shm_manager : && /usr/bin/c++ -I/usr/local/cuda-10.2/include -I/usr/local/cuda-10.2/include -I/usr/local/cuda/include -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DUSE_KINETO -DUSE_PYTORCH_QNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -DEDGE_PROFILER_USE_KINETO -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-unused-parameter -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow -DHAVE_AVX2_CPU_DEFINITION -g -fno-omit-frame-pointer -O0 -rdynamic -rdynamic caffe2/torch/lib/libshm/CMakeFiles/torch_shm_manager.dir/manager.cpp.o -o bin/torch_shm_manager -Wl,-rpath,/home/pytorch/build/lib:/root/anaconda3/envs/env_gnn/lib:/usr/lib/x86_64-linux-gnu/lib64: lib/libshm.so -lrt lib/libtorch.so -Wl,--no-as-needed,"/home/pytorch/build/lib/libtorch_cpu.so" -Wl,--as-needed lib/libprotobufd.a -pthread /root/anaconda3/envs/env_gnn/lib/libmkl_intel_lp64.so /root/anaconda3/envs/env_gnn/lib/libmkl_gnu_thread.so /root/anaconda3/envs/env_gnn/lib/libmkl_core.so -fopenmp -lpthread -lm -ldl -Wl,--no-as-needed,"/home/pytorch/build/lib/libtorch_cuda.so" -Wl,--as-needed lib/libc10_cuda.so -lcudart -lnvToolsExt -lcufft /usr/local/cuda-10.2/lib64/libcurand.so -lcublas /usr/lib/x86_64-linux-gnu/lib64/libcudnn.so lib/libc10.so && : /home/pytorch/build/lib/libtorch_cpu.so: undefined reference to `c10::StorageImpl::pagein_manual()' /home/pytorch/build/lib/libtorch_cpu.so: undefined reference to `c10::StorageImpl::release_resources()' /home/pytorch/build/lib/libtorch_cpu.so: undefined reference to `c10::StorageImpl::need_prefetch()' /home/pytorch/build/lib/libtorch_cpu.so: undefined reference to `vtable for c10::StorageImpl' /home/pytorch/build/lib/libtorch_cpu.so: undefined reference to `c10::StorageImpl::~StorageImpl()' /home/pytorch/build/lib/libtorch_cpu.so: undefined reference to `c10::StorageImpl::pageout_manual()' /home/pytorch/build/lib/libtorch_cpu.so: undefined reference to `c10::StorageImpl::~StorageImpl()' /home/pytorch/build/lib/libtorch_cpu.so: undefined reference to `typeinfo for c10::StorageImpl' collect2: error: ld returned 1 exit status [1983/1985] Linking CXX shared library lib/libtorch_python.so ninja: build stopped: subcommand failed.
I really appreciated if there is a code version that compiles properly.
Hi, I also encountered the same problem, have you solved this problem
I followed instructions in README.txt and compiled pytorch v1.12.0 successfully but failed after applying the patch. I got error message as followed.
I tried hard to fix the bugs but the project is so complex. I finally got error message as followed.
I really appreciated if there is a code version that compiles properly.