search

pytorch / tvm

TVM integration into PyTorch
453 stars 64 forks source link

Error when installing TVM on Raspberry (PyTorch 1.1) #130

Open snippler opened 5 years ago

snippler commented 5 years ago

This is the error that I get incl. some text before. I make it to ca. 95%. Any idea anyone what is wrong?

/usr/include/c++/8/bits/range_access.h:87:5: note: 'std::begin' begin(_Tp (&arr)[_Nm]) ^~~~~ /home/pi/code/tvm/torch_tvm/compiler.cpp:201:43: error: 'end' was not declared in this scope for (const auto& elem : val.toIntList()) { ^ /home/pi/code/tvm/torch_tvm/compiler.cpp:201:43: note: suggested alternatives: In file included from /usr/include/c++/8/string:51, from /usr/include/c++/8/bits/locale_classes.h:40, from /usr/include/c++/8/bits/ios_base.h:41, from /usr/include/c++/8/ios:42, from /usr/include/c++/8/istream:38, from /usr/include/c++/8/sstream:38, from /usr/local/lib/python2.7/dist-packages/torch/include/c10/macros/Macros.h:111, from /usr/local/lib/python2.7/dist-packages/torch/include/c10/core/DeviceType.h:8, from /usr/local/lib/python2.7/dist-packages/torch/include/c10/core/Device.h:3, from /usr/local/lib/python2.7/dist-packages/torch/include/ATen/core/Tensor.h:3, from /usr/local/lib/python2.7/dist-packages/torch/include/ATen/core/TensorMethods.h:3, from /usr/local/lib/python2.7/dist-packages/torch/include/ATen/core/jit_type.h:3, from /usr/local/lib/python2.7/dist-packages/torch/include/torch/csrc/jit/argument_spec.h:3, from /home/pi/code/tvm/torch_tvm/compiler.h:3, from /home/pi/code/tvm/torch_tvm/compiler.cpp:1: /usr/include/c++/8/bits/range_access.h:97:5: note: 'std::end' end(_Tp (&__arr)[_Nm]) ^~~ In file included from /usr/local/lib/python2.7/dist-packages/torch/include/ATen/core/jit_type.h:6, from /usr/local/lib/python2.7/dist-packages/torch/include/torch/csrc/jit/argument_spec.h:3, from /home/pi/code/tvm/torch_tvm/compiler.h:3, from /home/pi/code/tvm/torch_tvm/compiler.cpp:1: /usr/local/lib/python2.7/dist-packages/torch/include/ATen/core/aten_internedstrings.h:776:9: note: 'c10::attr::end' (attr, end) \ ^~~ /usr/local/lib/python2.7/dist-packages/torch/include/ATen/core/interned_strings.h:322:35: note: in definition of macro 'DEFINE_SYMBOL' namespace ns { constexpr Symbol s(static_cast(keys::ns####s)); } ^ /usr/local/lib/python2.7/dist-packages/torch/include/ATen/core/interned_strings.h:159:3: note: in expansion of macro 'FORALL_ATTR_BASE_SYMBOLS' FORALL_ATTR_BASESYMBOLS() \ ^~~~~~~~ /usr/local/lib/python2.7/dist-packages/torch/include/ATen/core/interned_strings.h:323:1: note: in expansion of macro 'FORALL_NS_SYMBOLS' FORALL_NS_SYMBOLS(DEFINE_SYMBOL) ^~~~~ /home/pi/code/tvm/torch_tvm/compiler.cpp:209:3: error: 'TORCH_CHECK' was not declared in this scope TORCH_CHECK( ^~~ /home/pi/code/tvm/torch_tvm/compiler.cpp:209:3: note: suggested alternative: 'AT_CHECK' TORCH_CHECK( ^~~ AT_CHECK /home/pi/code/tvm/torch_tvm/compiler.cpp: In static member function 'static tvm::relay::Function TVMCompiler::convertToRelay(std::shared_ptr, TVMContext, std::unordered_map<torch::jit::Value, TVMGraphInputInfo>)': /home/pi/code/tvm/torch_tvm/compiler.cpp:221:5: error: 'TORCH_INTERNAL_ASSERT' was not declared in this scope TORCH_INTERNAL_ASSERT(input->isCompleteTensor()); ^~~~~ /home/pi/code/tvm/torch_tvm/compiler.cpp:319:5: error: 'TORCH_INTERNAL_ASSERT' was not declared in this scope TORCH_INTERNAL_ASSERT(value_map.find(sg_output) != value_map.end()); ^~~~~ /home/pi/code/tvm/torch_tvm/compiler.cpp:326:3: error: 'TORCH_CHECK' was not declared in this scope TORCH_CHECK( ^~~ /home/pi/code/tvm/torch_tvm/compiler.cpp:326:3: note: suggested alternative: 'AT_CHECK' TORCH_CHECK( ^~~ AT_CHECK /home/pi/code/tvm/torch_tvm/compiler.cpp: In constructor 'TVMCompiler::TVMCompiler(const torch::jit::Node*, int, bool, bool, bool, std::cxx11::string, std::cxx11::string, std::cxx11::string)': /home/pi/code/tvm/torch_tvm/compiler.cpp:362:3: error: 'TORCH_INTERNAL_ASSERT' was not declared in this scope TORCH_INTERNAL_ASSERT(pfb); ^~~~~ /home/pi/code/tvm/torch_tvm/compiler.cpp: In member function 'void TVMCompiler::run(torch::jit::Stack&)': /home/pi/code/tvm/torch_tvm/compiler.cpp:440:5: error: 'TORCH_INTERNAL_ASSERT' was not declared in this scope TORCH_INTERNAL_ASSERT(pfr); ^~~~~ /home/pi/code/tvm/torch_tvm/compiler.cpp:443:9: error: 'TORCH_CHECK' was not declared in this scope TORCH_CHECK(pfr, "TVM must be compiled with debug runtime. " ^~~ /home/pi/code/tvm/torch_tvm/compiler.cpp:443:9: note: suggested alternative: 'AT_CHECK' TORCH_CHECK(pfr, "TVM must be compiled with debug runtime. " ^~~ AT_CHECK /home/pi/code/tvm/torch_tvm/compiler.cpp:466:5: error: 'TORCH_CHECK' was not declared in this scope TORCH_CHECK( ^~~ /home/pi/code/tvm/torch_tvm/compiler.cpp:466:5: note: suggested alternative: 'AT_CHECK' TORCH_CHECK( ^~~ AT_CHECK make[2]: [CMakeFiles/_torch_tvm.dir/build.make:284: CMakeFiles/_torch_tvm.dir/torch_tvm/memory_utils.cpp.o] Error 1 /home/pi/code/tvm/torch_tvm/compiler.cpp: In static member function 'static tvm::relay::Expr TVMCompiler::convertToRelay(const c10::IValue&, TVMContext)': /home/pi/code/tvm/torch_tvm/compiler.cpp:211:1: warning: control reaches end of non-void function [-Wreturn-type] } ^ make[2]: [CMakeFiles/_torch_tvm.dir/build.make:63: CMakeFiles/_torch_tvm.dir/torch_tvm/compiler.cpp.o] Error 1 make[1]: [CMakeFiles/Makefile2:74: CMakeFiles/_torch_tvm.dir/all] Error 2 make: [Makefile:130: all] Error 2 Traceback (most recent call last): File "setup.py", line 273, in url='https://github.com/pytorch/tvm', File "/usr/lib/python2.7/dist-packages/setuptools/init.py", line 145, in setup return distutils.core.setup(**attrs) File "/usr/lib/python2.7/distutils/core.py", line 151, in setup dist.run_commands() File "/usr/lib/python2.7/distutils/dist.py", line 953, in run_commands self.run_command(cmd) File "/usr/lib/python2.7/distutils/dist.py", line 972, in run_command cmd_obj.run() File "setup.py", line 203, in run setuptools.command.install.install.run(self) File "/usr/lib/python2.7/dist-packages/setuptools/command/install.py", line 65, in run orig.install.run(self) File "/usr/lib/python2.7/distutils/command/install.py", line 601, in run self.run_command('build') File "/usr/lib/python2.7/distutils/cmd.py", line 326, in run_command self.distribution.run_command(command) File "/usr/lib/python2.7/distutils/dist.py", line 972, in run_command cmd_obj.run() File "/usr/lib/python2.7/distutils/command/build.py", line 128, in run self.run_command(cmd_name) File "/usr/lib/python2.7/distutils/cmd.py", line 326, in run_command self.distribution.run_command(command) File "/usr/lib/python2.7/distutils/dist.py", line 972, in run_command cmd_obj.run() File "setup.py", line 187, in run self.run_command('cmake_build') File "/usr/lib/python2.7/distutils/cmd.py", line 326, in run_command self.distribution.run_command(command) File "/usr/lib/python2.7/distutils/dist.py", line 972, in run_command cmd_obj.run() File "setup.py", line 176, in run self._run_build() File "setup.py", line 165, in _run_build subprocess.check_call(build_args) File "/usr/lib/python2.7/subprocess.py", line 190, in check_call raise CalledProcessError(retcode, cmd) subprocess.CalledProcessError: Command '[u'/usr/local/bin/cmake', u'--build', '.', u'--', u'-j', '4']' returned non-zero exit status 2

snippler commented 5 years ago

I noticed that the actual error message was displayed before:

/home/pi/tvm/torch_tvm/register.cpp:129:16: error: inconsistent types 'unsigned int' and 'long unsigned int' deduced for lambda return type return 0UL; /home/pi/tvm/torch_tvm/register.cpp:137:25: error: conversion from 'tvm::runtime::TVMArgValue' to 'size_t' {aka 'unsigned int'} is ambigous size_t id = args[0];

I thought it means that I have to change the code for 32 bit system by changing to return 0U Indeed the first error disappears but the second error is kept. What could be the reason?