More modules for nn + improvements in CUDA tests #24
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
torch/autograd/variable.py
Outdated
This comment was marked as off-topic.
This comment was marked as off-topic.
Sorry, something went wrong.
Contributor
|
this is now pushed to master |
pytorchmergebot
pushed a commit
that referenced
this pull request
Jul 28, 2023
### Description Hi! We've been fuzzing `pytorch` with [sydr-fuzz](https://github.com/ispras/oss-sydr-fuzz) and found error of out of bounds access in `torch::jit` module. pytorch version: 18bcf62 The error occurs in `import_source.cpp:560` when we get the type from the `assign.rhs()`. `assign.rhs()` has `Maybe` type, as well as `assign.type()`, so one of them can be not presented. According to [grammar](https://github.com/pytorch/pytorch/blob/22f93852a2664b3dc29544ac6a36f1ec52c6caa2/torch/csrc/jit/frontend/tree_views.h), we can have `Assign` statement, which `lhs` will be `Subscript`, `rhs` will be empty (`Maybe` type with no subtrees) and `type` will be presented. But in `import_source.cpp:560` we try to get `rhs` expression from the assignment with no check whether it is presented. This is example from the how to reproduce section from the testing input: ``` class Module(Module): __parameters__ = ["0", ] __buffers__ = [] __annotations__ = [] __annotations__["0"] : Tensor ``` When we parse the last statement of class definition, we set the type of `lhs` to `Subscript`, because the lookahead is `[` https://github.com/pytorch/pytorch/blob/76fb72e24a5a4a47ad1f50c5c94d5c0b7e703531/torch/csrc/jit/frontend/parser.cpp#L205-L207 Then in `parseAssignment` we get `maybeOp` and `type` depending on the next symbol (if it is `:`, we get only the type) https://github.com/pytorch/pytorch/blob/76fb72e24a5a4a47ad1f50c5c94d5c0b7e703531/torch/csrc/jit/frontend/parser.cpp#L437-L447 So after that, in `import_source.cpp:560`, parsing attributes, one of which is assignment with subscript type of `lhs`, we try to get type from `rhs` expression and out of bounds access occurs. To fix the error, we need to check whether the `rhs` or `type` are presented and get the type from corresponding expression. ### How to reproduce Build docker container from [here](https://github.com/ispras/oss-sydr-fuzz/tree/master/projects/pytorch): ```bash $ sudo docker build -t oss-sydr-fuzz-pytorch ``` Run docker container: ```bash $ sudo docker run --rm --privileged -v `pwd`:/fuzz -it oss-sydr-fuzz-pytorch /bin/bash ``` Run the `load_fuzz` target on the [input.txt](https://github.com/pytorch/pytorch/files/12173962/input.txt) ```bash /load_fuzz input.txt ``` You will see the following output: ``` AddressSanitizer:DEADLYSIGNAL ================================================================= ==157==ERROR: AddressSanitizer: SEGV on unknown address (pc 0x00000c163764 bp 0x7ffee71d0070 sp 0x7ffee71d0050 T0) ==157==The signal is caused by a READ memory access. ==157==Hint: this fault was caused by a dereference of a high value address (see register values below). Disassemble the provided pc to learn which register was used. #0 0xc163764 in c10::intrusive_ptr<torch::jit::Tree, c10::detail::intrusive_target_default_null_type<torch::jit::Tree> >::retain_() /pytorch/c10/util/intrusive_ptr.h:265:54 #1 0xc1697fd in c10::intrusive_ptr<torch::jit::Tree, c10::detail::intrusive_target_default_null_type<torch::jit::Tree> >::intrusive_ptr(c10::intrusive_ptr<torch::jit::Tree, c10::detail::intrusive_target_default_null_type<torch::jit::Tree> > const&) /pytorch/c10/util/intrusive_ptr.h:354:5 #2 0xc1697fd in torch::jit::Expr::Expr(c10::intrusive_ptr<torch::jit::Tree, c10::detail::intrusive_target_default_null_type<torch::jit::Tree> > const&) /pytorch/torch/csrc/jit/frontend/tree_views.h:270:49 #3 0xc1f02cb in torch::jit::Maybe<torch::jit::Expr>::get() const /pytorch/torch/csrc/jit/frontend/tree_views.h:212:12 #4 0xd194369 in torch::jit::SourceImporterImpl::importClass(c10::QualifiedName const&, torch::jit::ClassDef const&, bool) /pytorch/torch/csrc/jit/serialization/import_source.cpp:560:70 #5 0xd18c701 in torch::jit::SourceImporterImpl::importNamedType(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, torch::jit::ClassDef const&) /pytorch/torch/csrc/jit/serialization/import_source.cpp:288:5 #6 0xd18a84c in torch::jit::SourceImporterImpl::findNamedType(c10::QualifiedName const&) /pytorch/torch/csrc/jit/serialization/import_source.cpp:140:5 #7 0xd1913a8 in torch::jit::SourceImporterImpl::resolveType(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, torch::jit::SourceRange const&) /pytorch/torch/csrc/jit/serialization/import_source.cpp:261:10 #8 0xc2e422f in torch::jit::ScriptTypeParser::parseTypeFromExpr(torch::jit::Expr const&) const /pytorch/torch/csrc/jit/frontend/script_type_parser.cpp:238:24 #9 0xc2e4697 in torch::jit::ScriptTypeParser::parseType(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) /pytorch/torch/csrc/jit/frontend/script_type_parser.cpp:312:10 #10 0xd1a37d4 in torch::jit::SourceImporter::loadType(c10::QualifiedName const&) const /pytorch/torch/csrc/jit/serialization/import_source.cpp:786:27 #11 0xd121c47 in torch::jit::(anonymous namespace)::ScriptModuleDeserializer::readArchive(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)::$_0::operator()(c10::QualifiedName const&) const /pytorch/torch/csrc/jit/serialization/import.cpp:146:33 #12 0xd121c47 in c10::StrongTypePtr std::__invoke_impl<c10::StrongTypePtr, torch::jit::(anonymous namespace)::ScriptModuleDeserializer::readArchive(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)::$_0&, c10::QualifiedName const&>(std::__invoke_other, torch::jit::(anonymous namespace)::ScriptModuleDeserializer::readArchive(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)::$_0&, c10::QualifiedName const&) /usr/bin/../lib/gcc/x86_64-linux-gnu/10/../../../../include/c++/10/bits/invoke.h:60:14 #13 0xd121ad0 in std::enable_if<is_invocable_r_v<c10::StrongTypePtr, torch::jit::(anonymous namespace)::ScriptModuleDeserializer::readArchive(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)::$_0&, c10::QualifiedName const&>, c10::StrongTypePtr>::type std::__invoke_r<c10::StrongTypePtr, torch::jit::(anonymous namespace)::ScriptModuleDeserializer::readArchive(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)::$_0&, c10::QualifiedName const&>(torch::jit::(anonymous namespace)::ScriptModuleDeserializer::readArchive(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)::$_0&, c10::QualifiedName const&) /usr/bin/../lib/gcc/x86_64-linux-gnu/10/../../../../include/c++/10/bits/invoke.h:113:9 #14 0xd121926 in std::_Function_handler<c10::StrongTypePtr (c10::QualifiedName const&), torch::jit::(anonymous namespace)::ScriptModuleDeserializer::readArchive(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)::$_0>::_M_invoke(std::_Any_data const&, c10::QualifiedName const&) /usr/bin/../lib/gcc/x86_64-linux-gnu/10/../../../../include/c++/10/bits/std_function.h:291:9 #15 0xd17ec49 in std::function<c10::StrongTypePtr (c10::QualifiedName const&)>::operator()(c10::QualifiedName const&) const /usr/bin/../lib/gcc/x86_64-linux-gnu/10/../../../../include/c++/10/bits/std_function.h:622:14 #16 0xd26b802 in torch::jit::Unpickler::readGlobal(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) /pytorch/torch/csrc/jit/serialization/unpickler.cpp:844:9 #17 0xd2615fb in torch::jit::Unpickler::readInstruction() /pytorch/torch/csrc/jit/serialization/unpickler.cpp:520:7 #18 0xd25f917 in torch::jit::Unpickler::run() /pytorch/torch/csrc/jit/serialization/unpickler.cpp:253:27 #19 0xd25f5b2 in torch::jit::Unpickler::parse_ivalue() /pytorch/torch/csrc/jit/serialization/unpickler.cpp:206:3 #20 0xd186403 in torch::jit::readArchiveAndTensors(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, c10::optional<std::function<c10::StrongTypePtr (c10::QualifiedName const&)> >, c10::optional<std::function<c10::intrusive_ptr<c10::ivalue::Object, c10::detail::intrusive_target_default_null_type<c10::ivalue::Object> > (c10::StrongTypePtr, c10::IValue)> >, c10::optional<c10::Device>, caffe2::serialize::PyTorchStreamReader&, c10::Type::SingletonOrSharedTypePtr<c10::Type> (*)(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&), std::shared_ptr<torch::jit::DeserializationStorageContext>) /pytorch/torch/csrc/jit/serialization/import_read.cpp:53:20 #21 0xd12152d in torch::jit::(anonymous namespace)::ScriptModuleDeserializer::readArchive(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) /pytorch/torch/csrc/jit/serialization/import.cpp:184:10 #22 0xd117bae in torch::jit::(anonymous namespace)::ScriptModuleDeserializer::deserialize(c10::optional<c10::Device>, std::unordered_map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::hash<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::equal_to<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > > >&, bool) /pytorch/torch/csrc/jit/serialization/import.cpp:287:19 #23 0xd114074 in torch::jit::import_ir_module(std::shared_ptr<torch::jit::CompilationUnit>, std::istream&, c10::optional<c10::Device>, std::unordered_map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::hash<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::equal_to<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > > >&, bool, bool) /pytorch/torch/csrc/jit/serialization/import.cpp:389:25 #24 0xd113a27 in torch::jit::import_ir_module(std::shared_ptr<torch::jit::CompilationUnit>, std::istream&, c10::optional<c10::Device>, bool) /pytorch/torch/csrc/jit/serialization/import.cpp:325:10 #25 0xd11bb64 in torch::jit::load(std::istream&, c10::optional<c10::Device>, bool) /pytorch/torch/csrc/jit/serialization/import.cpp:485:10 #26 0x610c5c in LLVMFuzzerTestOneInput /load.cc:42:14 #27 0x537701 in fuzzer::Fuzzer::ExecuteCallback(unsigned char const*, unsigned long) /llvm-project-llvmorg-14.0.6/compiler-rt/lib/fuzzer/FuzzerLoop.cpp:611:15 #28 0x52160c in fuzzer::RunOneTest(fuzzer::Fuzzer*, char const*, unsigned long) /llvm-project-llvmorg-14.0.6/compiler-rt/lib/fuzzer/FuzzerDriver.cpp:324:6 #29 0x52735b in fuzzer::FuzzerDriver(int*, char***, int (*)(unsigned char const*, unsigned long)) /llvm-project-llvmorg-14.0.6/compiler-rt/lib/fuzzer/FuzzerDriver.cpp:860:9 #30 0x550912 in main /llvm-project-llvmorg-14.0.6/compiler-rt/lib/fuzzer/FuzzerMain.cpp:20:10 #31 0x7f06e8323082 in __libc_start_main (/lib/x86_64-linux-gnu/libc.so.6+0x24082) (BuildId: 1878e6b475720c7c51969e69ab2d276fae6d1dee) #32 0x51bf2d in _start (/load_fuzz+0x51bf2d) AddressSanitizer can not provide additional info. SUMMARY: AddressSanitizer: SEGV /pytorch/c10/util/intrusive_ptr.h:265:54 in c10::intrusive_ptr<torch::jit::Tree, c10::detail::intrusive_target_default_null_type<torch::jit::Tree> >::retain_() ==157==ABORTING ``` Pull Request resolved: #106041 Approved by: https://github.com/davidberard98
bobby-palmer
pushed a commit
to bobby-palmer/pytorch
that referenced
this pull request
Jul 29, 2023
### Description Hi! We've been fuzzing `pytorch` with [sydr-fuzz](https://github.com/ispras/oss-sydr-fuzz) and found error of out of bounds access in `torch::jit` module. pytorch version: 18bcf62 The error occurs in `import_source.cpp:560` when we get the type from the `assign.rhs()`. `assign.rhs()` has `Maybe` type, as well as `assign.type()`, so one of them can be not presented. According to [grammar](https://github.com/pytorch/pytorch/blob/22f93852a2664b3dc29544ac6a36f1ec52c6caa2/torch/csrc/jit/frontend/tree_views.h), we can have `Assign` statement, which `lhs` will be `Subscript`, `rhs` will be empty (`Maybe` type with no subtrees) and `type` will be presented. But in `import_source.cpp:560` we try to get `rhs` expression from the assignment with no check whether it is presented. This is example from the how to reproduce section from the testing input: ``` class Module(Module): __parameters__ = ["0", ] __buffers__ = [] __annotations__ = [] __annotations__["0"] : Tensor ``` When we parse the last statement of class definition, we set the type of `lhs` to `Subscript`, because the lookahead is `[` https://github.com/pytorch/pytorch/blob/76fb72e24a5a4a47ad1f50c5c94d5c0b7e703531/torch/csrc/jit/frontend/parser.cpp#L205-L207 Then in `parseAssignment` we get `maybeOp` and `type` depending on the next symbol (if it is `:`, we get only the type) https://github.com/pytorch/pytorch/blob/76fb72e24a5a4a47ad1f50c5c94d5c0b7e703531/torch/csrc/jit/frontend/parser.cpp#L437-L447 So after that, in `import_source.cpp:560`, parsing attributes, one of which is assignment with subscript type of `lhs`, we try to get type from `rhs` expression and out of bounds access occurs. To fix the error, we need to check whether the `rhs` or `type` are presented and get the type from corresponding expression. ### How to reproduce Build docker container from [here](https://github.com/ispras/oss-sydr-fuzz/tree/master/projects/pytorch): ```bash $ sudo docker build -t oss-sydr-fuzz-pytorch ``` Run docker container: ```bash $ sudo docker run --rm --privileged -v `pwd`:/fuzz -it oss-sydr-fuzz-pytorch /bin/bash ``` Run the `load_fuzz` target on the [input.txt](https://github.com/pytorch/pytorch/files/12173962/input.txt) ```bash /load_fuzz input.txt ``` You will see the following output: ``` AddressSanitizer:DEADLYSIGNAL ================================================================= ==157==ERROR: AddressSanitizer: SEGV on unknown address (pc 0x00000c163764 bp 0x7ffee71d0070 sp 0x7ffee71d0050 T0) ==157==The signal is caused by a READ memory access. ==157==Hint: this fault was caused by a dereference of a high value address (see register values below). Disassemble the provided pc to learn which register was used. #0 0xc163764 in c10::intrusive_ptr<torch::jit::Tree, c10::detail::intrusive_target_default_null_type<torch::jit::Tree> >::retain_() /pytorch/c10/util/intrusive_ptr.h:265:54 pytorch#1 0xc1697fd in c10::intrusive_ptr<torch::jit::Tree, c10::detail::intrusive_target_default_null_type<torch::jit::Tree> >::intrusive_ptr(c10::intrusive_ptr<torch::jit::Tree, c10::detail::intrusive_target_default_null_type<torch::jit::Tree> > const&) /pytorch/c10/util/intrusive_ptr.h:354:5 pytorch#2 0xc1697fd in torch::jit::Expr::Expr(c10::intrusive_ptr<torch::jit::Tree, c10::detail::intrusive_target_default_null_type<torch::jit::Tree> > const&) /pytorch/torch/csrc/jit/frontend/tree_views.h:270:49 pytorch#3 0xc1f02cb in torch::jit::Maybe<torch::jit::Expr>::get() const /pytorch/torch/csrc/jit/frontend/tree_views.h:212:12 pytorch#4 0xd194369 in torch::jit::SourceImporterImpl::importClass(c10::QualifiedName const&, torch::jit::ClassDef const&, bool) /pytorch/torch/csrc/jit/serialization/import_source.cpp:560:70 pytorch#5 0xd18c701 in torch::jit::SourceImporterImpl::importNamedType(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, torch::jit::ClassDef const&) /pytorch/torch/csrc/jit/serialization/import_source.cpp:288:5 pytorch#6 0xd18a84c in torch::jit::SourceImporterImpl::findNamedType(c10::QualifiedName const&) /pytorch/torch/csrc/jit/serialization/import_source.cpp:140:5 pytorch#7 0xd1913a8 in torch::jit::SourceImporterImpl::resolveType(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, torch::jit::SourceRange const&) /pytorch/torch/csrc/jit/serialization/import_source.cpp:261:10 pytorch#8 0xc2e422f in torch::jit::ScriptTypeParser::parseTypeFromExpr(torch::jit::Expr const&) const /pytorch/torch/csrc/jit/frontend/script_type_parser.cpp:238:24 pytorch#9 0xc2e4697 in torch::jit::ScriptTypeParser::parseType(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) /pytorch/torch/csrc/jit/frontend/script_type_parser.cpp:312:10 pytorch#10 0xd1a37d4 in torch::jit::SourceImporter::loadType(c10::QualifiedName const&) const /pytorch/torch/csrc/jit/serialization/import_source.cpp:786:27 pytorch#11 0xd121c47 in torch::jit::(anonymous namespace)::ScriptModuleDeserializer::readArchive(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)::$_0::operator()(c10::QualifiedName const&) const /pytorch/torch/csrc/jit/serialization/import.cpp:146:33 pytorch#12 0xd121c47 in c10::StrongTypePtr std::__invoke_impl<c10::StrongTypePtr, torch::jit::(anonymous namespace)::ScriptModuleDeserializer::readArchive(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)::$_0&, c10::QualifiedName const&>(std::__invoke_other, torch::jit::(anonymous namespace)::ScriptModuleDeserializer::readArchive(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)::$_0&, c10::QualifiedName const&) /usr/bin/../lib/gcc/x86_64-linux-gnu/10/../../../../include/c++/10/bits/invoke.h:60:14 pytorch#13 0xd121ad0 in std::enable_if<is_invocable_r_v<c10::StrongTypePtr, torch::jit::(anonymous namespace)::ScriptModuleDeserializer::readArchive(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)::$_0&, c10::QualifiedName const&>, c10::StrongTypePtr>::type std::__invoke_r<c10::StrongTypePtr, torch::jit::(anonymous namespace)::ScriptModuleDeserializer::readArchive(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)::$_0&, c10::QualifiedName const&>(torch::jit::(anonymous namespace)::ScriptModuleDeserializer::readArchive(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)::$_0&, c10::QualifiedName const&) /usr/bin/../lib/gcc/x86_64-linux-gnu/10/../../../../include/c++/10/bits/invoke.h:113:9 pytorch#14 0xd121926 in std::_Function_handler<c10::StrongTypePtr (c10::QualifiedName const&), torch::jit::(anonymous namespace)::ScriptModuleDeserializer::readArchive(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)::$_0>::_M_invoke(std::_Any_data const&, c10::QualifiedName const&) /usr/bin/../lib/gcc/x86_64-linux-gnu/10/../../../../include/c++/10/bits/std_function.h:291:9 pytorch#15 0xd17ec49 in std::function<c10::StrongTypePtr (c10::QualifiedName const&)>::operator()(c10::QualifiedName const&) const /usr/bin/../lib/gcc/x86_64-linux-gnu/10/../../../../include/c++/10/bits/std_function.h:622:14 pytorch#16 0xd26b802 in torch::jit::Unpickler::readGlobal(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) /pytorch/torch/csrc/jit/serialization/unpickler.cpp:844:9 pytorch#17 0xd2615fb in torch::jit::Unpickler::readInstruction() /pytorch/torch/csrc/jit/serialization/unpickler.cpp:520:7 pytorch#18 0xd25f917 in torch::jit::Unpickler::run() /pytorch/torch/csrc/jit/serialization/unpickler.cpp:253:27 pytorch#19 0xd25f5b2 in torch::jit::Unpickler::parse_ivalue() /pytorch/torch/csrc/jit/serialization/unpickler.cpp:206:3 pytorch#20 0xd186403 in torch::jit::readArchiveAndTensors(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, c10::optional<std::function<c10::StrongTypePtr (c10::QualifiedName const&)> >, c10::optional<std::function<c10::intrusive_ptr<c10::ivalue::Object, c10::detail::intrusive_target_default_null_type<c10::ivalue::Object> > (c10::StrongTypePtr, c10::IValue)> >, c10::optional<c10::Device>, caffe2::serialize::PyTorchStreamReader&, c10::Type::SingletonOrSharedTypePtr<c10::Type> (*)(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&), std::shared_ptr<torch::jit::DeserializationStorageContext>) /pytorch/torch/csrc/jit/serialization/import_read.cpp:53:20 pytorch#21 0xd12152d in torch::jit::(anonymous namespace)::ScriptModuleDeserializer::readArchive(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) /pytorch/torch/csrc/jit/serialization/import.cpp:184:10 pytorch#22 0xd117bae in torch::jit::(anonymous namespace)::ScriptModuleDeserializer::deserialize(c10::optional<c10::Device>, std::unordered_map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::hash<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::equal_to<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > > >&, bool) /pytorch/torch/csrc/jit/serialization/import.cpp:287:19 pytorch#23 0xd114074 in torch::jit::import_ir_module(std::shared_ptr<torch::jit::CompilationUnit>, std::istream&, c10::optional<c10::Device>, std::unordered_map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::hash<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::equal_to<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > > >&, bool, bool) /pytorch/torch/csrc/jit/serialization/import.cpp:389:25 pytorch#24 0xd113a27 in torch::jit::import_ir_module(std::shared_ptr<torch::jit::CompilationUnit>, std::istream&, c10::optional<c10::Device>, bool) /pytorch/torch/csrc/jit/serialization/import.cpp:325:10 pytorch#25 0xd11bb64 in torch::jit::load(std::istream&, c10::optional<c10::Device>, bool) /pytorch/torch/csrc/jit/serialization/import.cpp:485:10 pytorch#26 0x610c5c in LLVMFuzzerTestOneInput /load.cc:42:14 pytorch#27 0x537701 in fuzzer::Fuzzer::ExecuteCallback(unsigned char const*, unsigned long) /llvm-project-llvmorg-14.0.6/compiler-rt/lib/fuzzer/FuzzerLoop.cpp:611:15 pytorch#28 0x52160c in fuzzer::RunOneTest(fuzzer::Fuzzer*, char const*, unsigned long) /llvm-project-llvmorg-14.0.6/compiler-rt/lib/fuzzer/FuzzerDriver.cpp:324:6 pytorch#29 0x52735b in fuzzer::FuzzerDriver(int*, char***, int (*)(unsigned char const*, unsigned long)) /llvm-project-llvmorg-14.0.6/compiler-rt/lib/fuzzer/FuzzerDriver.cpp:860:9 pytorch#30 0x550912 in main /llvm-project-llvmorg-14.0.6/compiler-rt/lib/fuzzer/FuzzerMain.cpp:20:10 pytorch#31 0x7f06e8323082 in __libc_start_main (/lib/x86_64-linux-gnu/libc.so.6+0x24082) (BuildId: 1878e6b475720c7c51969e69ab2d276fae6d1dee) pytorch#32 0x51bf2d in _start (/load_fuzz+0x51bf2d) AddressSanitizer can not provide additional info. SUMMARY: AddressSanitizer: SEGV /pytorch/c10/util/intrusive_ptr.h:265:54 in c10::intrusive_ptr<torch::jit::Tree, c10::detail::intrusive_target_default_null_type<torch::jit::Tree> >::retain_() ==157==ABORTING ``` Pull Request resolved: pytorch#106041 Approved by: https://github.com/davidberard98
pytorchmergebot
pushed a commit
that referenced
this pull request
Sep 5, 2023
Hi! I've been fuzzing different pytorch modules with with [sydr-fuzz](https://github.com/ispras/oss-sydr-fuzz/tree/master/projects/pytorch), and found a SEGV that occurs during class deserialization in jit module. Docker to reproduce found error: [Dockerfile](https://github.com/ispras/oss-sydr-fuzz/tree/master/projects/pytorch). ### PoC: [crash-bfbab61bf86755aa712bb978e26057ae76d75fe4.txt](https://github.com/pytorch/pytorch/files/12499228/crash-bfbab61bf86755aa712bb978e26057ae76d75fe4.txt) ### ASAN report ``` ==1003115==ERROR: AddressSanitizer: SEGV on unknown address (pc 0x00000db61680 bp 0x7fffffff5e30 sp 0x7fffffff5a60 T0) ==1003115==The signal is caused by a READ memory access. ==1003115==Hint: this fault was caused by a dereference of a high value address (see register values below). Disassemble the provided pc to learn which register was used. #0 0xdb61680 in c10::intrusive_ptr<torch::jit::Tree, c10::detail::intrusive_target_default_null_type<torch::jit::Tree> >::retain_() /pytorch/c10/util/intrusive_ptr.h:265:54 #1 0xdb6721c in c10::intrusive_ptr<torch::jit::Tree, c10::detail::intrusive_target_default_null_type<torch::jit::Tree> >::intrusive_ptr(c10::intrusive_ptr<torch::jit::Tree, c10::detail::intrusive_target_default_null_type<torch::jit::Tree> > const&) /pytorch/c10/util/intrusive_ptr.h:354:5 #2 0xdb6721c in torch::jit::Expr::Expr(c10::intrusive_ptr<torch::jit::Tree, c10::detail::intrusive_target_default_null_type<torch::jit::Tree> > const&) /pytorch/torch/csrc/jit/frontend/tree_views.h:270:49 #3 0xdbf73b9 in torch::jit::Maybe<torch::jit::Expr>::get() const /pytorch/torch/csrc/jit/frontend/tree_views.h:212:12 #4 0xecac171 in torch::jit::SourceImporterImpl::importClass(c10::QualifiedName const&, torch::jit::ClassDef const&, bool) /pytorch/torch/csrc/jit/serialization/import_source.cpp:454:64 #5 0xeca0ada in torch::jit::SourceImporterImpl::importNamedType(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, torch::jit::ClassDef const&) /pytorch/torch/csrc/jit/serialization/import_source.cpp:288:5 #6 0xeca7422 in torch::jit::SourceImporterImpl::findNamedType(c10::QualifiedName const&) /pytorch/torch/csrc/jit/serialization/import_source.cpp:140:5 #7 0xeca295c in torch::jit::SourceImporterImpl::resolveType(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, torch::jit::SourceRange const&) /pytorch/torch/csrc/jit/serialization/import_source.cpp:261:10 #8 0xdd03bc8 in torch::jit::ScriptTypeParser::parseTypeFromExpr(torch::jit::Expr const&) const /pytorch/torch/csrc/jit/frontend/script_type_parser.cpp:238:24 #9 0xdcfc9b6 in torch::jit::ScriptTypeParser::parseType(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) /pytorch/torch/csrc/jit/frontend/script_type_parser.cpp:312:10 #10 0xecbac43 in torch::jit::SourceImporter::loadType(c10::QualifiedName const&) const /pytorch/torch/csrc/jit/serialization/import_source.cpp:786:27 #11 0xec2b5d3 in torch::jit::(anonymous namespace)::ScriptModuleDeserializer::readArchive(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)::$_0::operator()(c10::QualifiedName const&) const /pytorch/torch/csrc/jit/serialization/import.cpp:146:33 #12 0xec2b5d3 in c10::StrongTypePtr std::__invoke_impl<c10::StrongTypePtr, torch::jit::(anonymous namespace)::ScriptModuleDeserializer::readArchive(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)::$_0&, c10::QualifiedName const&>(std::__invoke_other, torch::jit::(anonymous namespace)::ScriptModuleDeserializer::readArchive(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)::$_0&, c10::QualifiedName const&) /usr/bin/../lib/gcc/x86_64-linux-gnu/10/../../../../include/c++/10/bits/invoke.h:60:14 #13 0xec2b4a0 in std::enable_if<is_invocable_r_v<c10::StrongTypePtr, torch::jit::(anonymous namespace)::ScriptModuleDeserializer::readArchive(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)::$_0&, c10::QualifiedName const&>, c10::StrongTypePtr>::type std::__invoke_r<c10::StrongTypePtr, torch::jit::(anonymous namespace)::ScriptModuleDeserializer::readArchive(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)::$_0&, c10::QualifiedName const&>(torch::jit::(anonymous namespace)::ScriptModuleDeserializer::readArchive(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)::$_0&, c10::QualifiedName const&) /usr/bin/../lib/gcc/x86_64-linux-gnu/10/../../../../include/c++/10/bits/invoke.h:113:9 #14 0xec2b3a0 in std::_Function_handler<c10::StrongTypePtr (c10::QualifiedName const&), torch::jit::(anonymous namespace)::ScriptModuleDeserializer::readArchive(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)::$_0>::_M_invoke(std::_Any_data const&, c10::QualifiedName const&) /usr/bin/../lib/gcc/x86_64-linux-gnu/10/../../../../include/c++/10/bits/std_function.h:291:9 #15 0xec95f7c in std::function<c10::StrongTypePtr (c10::QualifiedName const&)>::operator()(c10::QualifiedName const&) const /usr/bin/../lib/gcc/x86_64-linux-gnu/10/../../../../include/c++/10/bits/std_function.h:622:14 #16 0xed78721 in torch::jit::Unpickler::readGlobal(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) /pytorch/torch/csrc/jit/serialization/unpickler.cpp:844:9 #17 0xed87821 in torch::jit::Unpickler::readInstruction() /pytorch/torch/csrc/jit/serialization/unpickler.cpp:520:7 #18 0xed85b27 in torch::jit::Unpickler::run() /pytorch/torch/csrc/jit/serialization/unpickler.cpp:253:27 #19 0xed85781 in torch::jit::Unpickler::parse_ivalue() /pytorch/torch/csrc/jit/serialization/unpickler.cpp:206:3 #20 0xec9c7be in torch::jit::readArchiveAndTensors(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, c10::optional<std::function<c10::StrongTypePtr (c10::QualifiedName const&)> >, c10::optional<std::function<c10::intrusive_ptr<c10::ivalue::Object, c10::detail::intrusive_target_default_null_type<c10::ivalue::Object> > (c10::StrongTypePtr, c10::IValue)> >, c10::optional<c10::Device>, caffe2::serialize::PyTorchStreamReader&, c10::Type::SingletonOrSharedTypePtr<c10::Type> (*)(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&), std::shared_ptr<torch::jit::DeserializationStorageContext>) /pytorch/torch/csrc/jit/serialization/import_read.cpp:53:20 #21 0xec2b168 in torch::jit::(anonymous namespace)::ScriptModuleDeserializer::readArchive(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) /pytorch/torch/csrc/jit/serialization/import.cpp:184:10 #22 0xec27235 in torch::jit::(anonymous namespace)::ScriptModuleDeserializer::deserialize(c10::optional<c10::Device>, std::unordered_map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::hash<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::equal_to<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > > >&, bool) /pytorch/torch/csrc/jit/serialization/import.cpp:287:19 #23 0xec25644 in torch::jit::import_ir_module(std::shared_ptr<torch::jit::CompilationUnit>, std::istream&, c10::optional<c10::Device>, std::unordered_map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::hash<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::equal_to<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > > >&, bool, bool) /pytorch/torch/csrc/jit/serialization/import.cpp:389:25 #24 0xec2dcbe in torch::jit::import_ir_module(std::shared_ptr<torch::jit::CompilationUnit>, std::istream&, c10::optional<c10::Device>, bool) /pytorch/torch/csrc/jit/serialization/import.cpp:325:10 #25 0xec30659 in torch::jit::load(std::istream&, c10::optional<c10::Device>, bool) /pytorch/torch/csrc/jit/serialization/import.cpp:485:10 #26 0x8d8636 in LLVMFuzzerTestOneInput /load.cc:42:14 #27 0x8d835d in ExecuteFilesOnyByOne /AFLplusplus/utils/aflpp_driver/aflpp_driver.c:255:7 #28 0x8d8168 in LLVMFuzzerRunDriver /AFLplusplus/utils/aflpp_driver/aflpp_driver.c #29 0x8d7d28 in main /AFLplusplus/utils/aflpp_driver/aflpp_driver.c:300:10 #30 0x7ffff7a37082 in __libc_start_main (/lib/x86_64-linux-gnu/libc.so.6+0x24082) (BuildId: 1878e6b475720c7c51969e69ab2d276fae6d1dee) #31 0x817add in _start (/load_afl+0x817add) AddressSanitizer can not provide additional info. SUMMARY: AddressSanitizer: SEGV /pytorch/c10/util/intrusive_ptr.h:265:54 in c10::intrusive_ptr<torch::jit::Tree, c10::detail::intrusive_target_default_null_type<torch::jit::Tree> >::retain_() ==1003115==ABORTING ``` Pull Request resolved: #108417 Approved by: https://github.com/ezyang
pytorchmergebot
pushed a commit
that referenced
this pull request
Sep 6, 2023
…zation (#108418) Hi! I've been fuzzing different pytorch modules with with [sydr-fuzz](https://github.com/ispras/oss-sydr-fuzz/tree/master/projects/pytorch), and found a SEGV that occurs during data parsing for quantized conv deserialization. The crash occurs because of empty `optional` vector. Docker to reproduce found error: [Dockerfile](https://github.com/ispras/oss-sydr-fuzz/tree/master/projects/pytorch). ### PoC: [crash-aaa72b1c1431ac556118e34099ba163052dc0f96.txt](https://github.com/pytorch/pytorch/files/12499249/crash-aaa72b1c1431ac556118e34099ba163052dc0f96.txt) ### ASAN report ``` ==1003193==ERROR: AddressSanitizer: SEGV on unknown address 0x000000000000 (pc 0x000000cbd1b1 bp 0x7fffffff8490 sp 0x7fffffff7a30 T0) ==1003193==The signal is caused by a READ memory access. ==1003193==Hint: address points to the zero page. #0 0xcbd1b1 in c10::optional_base<at::Tensor>::optional_base(c10::optional_base<at::Tensor> const&) /pytorch/c10/util/Optional.h:222:17 #1 0x2b32336 in c10::optional<at::Tensor>::optional(c10::optional<at::Tensor> const&) /pytorch/c10/util/Optional.h:631:3 #2 0x2b32336 in std::tuple<long, std::vector<long, std::allocator<long> >, std::vector<c10::optional<at::Tensor>, std::allocator<c10::optional<at::Tensor> > > > parse_conv_serialized_state<2u>(c10::IValue) /pytorch/aten/src/ATen/native/quantized/cpu/conv_serialization.h:183:17 #3 0x2b30276 in int register_conv_params<2>()::'lambda'(c10::IValue)::operator()(c10::IValue) const /pytorch/aten/src/ATen/native/quantized/cpu/fbgemm_utils.cpp:410:49 #4 0x2b30014 in std::enable_if<!(std::is_member_pointer<std::decay<int register_conv_params<2>()::'lambda'(c10::IValue) const&>::type>::value), std::invoke_result<int register_conv_params<2>()::'lambda'(c10::IValue) const&, c10::IValue>::type>::type c10::guts::invoke<int register_conv_params<2>()::'lambda'(c10::IValue) const&, c10::IValue>(int register_conv_params<2>()::'lambda'(c10::IValue) const&, c10::IValue&&) /pytorch/c10/util/C++17.h:203:10 #5 0x2b2f7e7 in torch::class_<ConvPackedParamsBase<2> >& torch::class_<ConvPackedParamsBase<2> >::def_pickle<int register_conv_params<2>()::'lambda'(c10::intrusive_ptr<ConvPackedParamsBase<2>, c10::detail::intrusive_target_default_null_type<ConvPackedParamsBase<2> > > const&), int register_conv_params<2>()::'lambda'(c10::IValue)>(int register_conv_params<2>()::'lambda'(c10::intrusive_ptr<ConvPackedParamsBase<2>, c10::detail::intrusive_target_default_null_type<ConvPackedParamsBase<2> > > const&)&&, int register_conv_params<2>()::'lambda'(c10::IValue)&&)::'lambda'(c10::tagged_capsule<ConvPackedParamsBase<2> >, c10::IValue&&)::operator()(c10::tagged_capsule<ConvPackedParamsBase<2> >, c10::IValue&&) const /pytorch/torch/custom_class.h:328:11 #6 0x2b2f570 in c10::guts::infer_function_traits<int register_conv_params<2>()::'lambda'(c10::intrusive_ptr<ConvPackedParamsBase<2>, c10::detail::intrusive_target_default_null_type<ConvPackedParamsBase<2> > > const&)>::type::return_type torch::detail::call_torchbind_method_from_stack<torch::class_<ConvPackedParamsBase<2> >& torch::class_<ConvPackedParamsBase<2> >::def_pickle<int register_conv_params<2>()::'lambda'(c10::intrusive_ptr<ConvPackedParamsBase<2>, c10::detail::intrusive_target_default_null_type<ConvPackedParamsBase<2> > > const&), int register_conv_params<2>()::'lambda'(c10::IValue)>(int register_conv_params<2>()::'lambda'(c10::intrusive_ptr<ConvPackedParamsBase<2>, c10::detail::intrusive_target_default_null_type<ConvPackedParamsBase<2> > > const&)&&, int register_conv_params<2>()::'lambda'(c10::IValue)&&)::'lambda'(c10::tagged_capsule<ConvPackedParamsBase<2> >, c10::IValue&&), false, 0ul, 1ul>(int register_conv_params<2>()::'lambda'(c10::intrusive_ptr<ConvPackedParamsBase<2>, c10::detail::intrusive_target_default_null_type<ConvPackedParamsBase<2> > > const&)&, std::vector<c10::IValue, std::allocator<c10::IValue> >&, std::integer_sequence<unsigned long, 0ul, 1ul>) /pytorch/torch/custom_class_detail.h:139:10 #7 0x2b2f408 in c10::guts::infer_function_traits<int register_conv_params<2>()::'lambda'(c10::intrusive_ptr<ConvPackedParamsBase<2>, c10::detail::intrusive_target_default_null_type<ConvPackedParamsBase<2> > > const&)>::type::return_type torch::detail::call_torchbind_method_from_stack<torch::class_<ConvPackedParamsBase<2> >& torch::class_<ConvPackedParamsBase<2> >::def_pickle<int register_conv_params<2>()::'lambda'(c10::intrusive_ptr<ConvPackedParamsBase<2>, c10::detail::intrusive_target_default_null_type<ConvPackedParamsBase<2> > > const&), int register_conv_params<2>()::'lambda'(c10::IValue)>(int register_conv_params<2>()::'lambda'(c10::intrusive_ptr<ConvPackedParamsBase<2>, c10::detail::intrusive_target_default_null_type<ConvPackedParamsBase<2> > > const&)&&, int register_conv_params<2>()::'lambda'(c10::IValue)&&)::'lambda'(c10::tagged_capsule<ConvPackedParamsBase<2> >, c10::IValue&&), false>(int register_conv_params<2>()::'lambda'(c10::intrusive_ptr<ConvPackedParamsBase<2>, c10::detail::intrusive_target_default_null_type<ConvPackedParamsBase<2> > > const&)&, std::vector<c10::IValue, std::allocator<c10::IValue> >&) /pytorch/torch/custom_class_detail.h:153:10 #8 0x2b2f408 in torch::detail::BoxedProxy<void, torch::class_<ConvPackedParamsBase<2> >& torch::class_<ConvPackedParamsBase<2> >::def_pickle<int register_conv_params<2>()::'lambda'(c10::intrusive_ptr<ConvPackedParamsBase<2>, c10::detail::intrusive_target_default_null_type<ConvPackedParamsBase<2> > > const&), int register_conv_params<2>()::'lambda'(c10::IValue)>(int register_conv_params<2>()::'lambda'(c10::intrusive_ptr<ConvPackedParamsBase<2>, c10::detail::intrusive_target_default_null_type<ConvPackedParamsBase<2> > > const&)&&, int register_conv_params<2>()::'lambda'(c10::IValue)&&)::'lambda'(c10::tagged_capsule<ConvPackedParamsBase<2> >, c10::IValue&&)>::operator()(std::vector<c10::IValue, std::allocator<c10::IValue> >&, torch::class_<ConvPackedParamsBase<2> >& torch::class_<ConvPackedParamsBase<2> >::def_pickle<int register_conv_params<2>()::'lambda'(c10::intrusive_ptr<ConvPackedParamsBase<2>, c10::detail::intrusive_target_default_null_type<ConvPackedParamsBase<2> > > const&), int register_conv_params<2>()::'lambda'(c10::IValue)>(int register_conv_params<2>()::'lambda'(c10::intrusive_ptr<ConvPackedParamsBase<2>, c10::detail::intrusive_target_default_null_type<ConvPackedParamsBase<2> > > const&)&&, int register_conv_params<2>()::'lambda'(c10::IValue)&&)::'lambda'(c10::tagged_capsule<ConvPackedParamsBase<2> >, c10::IValue&&)&) /pytorch/torch/custom_class_detail.h:174:5 #9 0x2b2f38d in torch::jit::Function* torch::class_<ConvPackedParamsBase<2> >::defineMethod<torch::class_<ConvPackedParamsBase<2> >& torch::class_<ConvPackedParamsBase<2> >::def_pickle<int register_conv_params<2>()::'lambda'(c10::intrusive_ptr<ConvPackedParamsBase<2>, c10::detail::intrusive_target_default_null_type<ConvPackedParamsBase<2> > > const&), int register_conv_params<2>()::'lambda'(c10::IValue)>(int register_conv_params<2>()::'lambda'(c10::intrusive_ptr<ConvPackedParamsBase<2>, c10::detail::intrusive_target_default_null_type<ConvPackedParamsBase<2> > > const&)&&, int register_conv_params<2>()::'lambda'(c10::IValue)&&)::'lambda'(c10::tagged_capsule<ConvPackedParamsBase<2> >, c10::IValue&&)>(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, int register_conv_params<2>()::'lambda'(c10::intrusive_ptr<ConvPackedParamsBase<2>, c10::detail::intrusive_target_default_null_type<ConvPackedParamsBase<2> > > const&), std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::initializer_list<torch::arg>)::'lambda'(std::vector<c10::IValue, std::allocator<c10::IValue> >&)::operator()(std::vector<c10::IValue, std::allocator<c10::IValue> >&) /pytorch/torch/custom_class.h:407:7 #10 0x2b2f38d in int register_conv_params<2>()::'lambda'(c10::intrusive_ptr<ConvPackedParamsBase<2>, c10::detail::intrusive_target_default_null_type<ConvPackedParamsBase<2> > > const&) std::__invoke_impl<void, torch::jit::Function* torch::class_<ConvPackedParamsBase<2> >::defineMethod<torch::class_<ConvPackedParamsBase<2> >& torch::class_<ConvPackedParamsBase<2> >::def_pickle<int register_conv_params<2>()::'lambda'(c10::intrusive_ptr<ConvPackedParamsBase<2>, c10::detail::intrusive_target_default_null_type<ConvPackedParamsBase<2> > > const&), int register_conv_params<2>()::'lambda'(c10::IValue)>(int register_conv_params<2>()::'lambda'(c10::intrusive_ptr<ConvPackedParamsBase<2>, c10::detail::intrusive_target_default_null_type<ConvPackedParamsBase<2> > > const&)&&, int register_conv_params<2>()::'lambda'(c10::IValue)&&)::'lambda'(c10::tagged_capsule<ConvPackedParamsBase<2> >, c10::IValue&&)>(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, int register_conv_params<2>()::'lambda'(c10::intrusive_ptr<ConvPackedParamsBase<2>, c10::detail::intrusive_target_default_null_type<ConvPackedParamsBase<2> > > const&), std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::initializer_list<torch::arg>)::'lambda'(std::vector<c10::IValue, std::allocator<c10::IValue> >&)&, std::vector<c10::IValue, std::allocator<c10::IValue> >&>(std::__invoke_other, int register_conv_params<2>()::'lambda'(c10::IValue)&&, std::vector<c10::IValue, std::allocator<c10::IValue> >&) /usr/bin/../lib/gcc/x86_64-linux-gnu/10/../../../../include/c++/10/bits/invoke.h:60:14 #11 0x125654e in torch::jit::Function::operator()(std::vector<c10::IValue, std::allocator<c10::IValue> >, std::unordered_map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, c10::IValue, std::hash<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::equal_to<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const, c10::IValue> > > const&) /pytorch/aten/src/ATen/core/function.h:62:5 #12 0xec2c1c6 in torch::jit::(anonymous namespace)::ScriptModuleDeserializer::readArchive(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)::$_1::operator()(c10::StrongTypePtr const&, c10::IValue) const /pytorch/torch/csrc/jit/serialization/import.cpp:172:7 #13 0xec2c1c6 in c10::intrusive_ptr<c10::ivalue::Object, c10::detail::intrusive_target_default_null_type<c10::ivalue::Object> > std::__invoke_impl<c10::intrusive_ptr<c10::ivalue::Object, c10::detail::intrusive_target_default_null_type<c10::ivalue::Object> >, torch::jit::(anonymous namespace)::ScriptModuleDeserializer::readArchive(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)::$_1&, c10::StrongTypePtr, c10::IValue>(std::__invoke_other, torch::jit::(anonymous namespace)::ScriptModuleDeserializer::readArchive(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)::$_1&, c10::StrongTypePtr&&, c10::IValue&&) /usr/bin/../lib/gcc/x86_64-linux-gnu/10/../../../../include/c++/10/bits/invoke.h:60:14 #14 0xec2b9a0 in std::enable_if<is_invocable_r_v<c10::intrusive_ptr<c10::ivalue::Object, c10::detail::intrusive_target_default_null_type<c10::ivalue::Object> >, torch::jit::(anonymous namespace)::ScriptModuleDeserializer::readArchive(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)::$_1&, c10::StrongTypePtr, c10::IValue>, c10::intrusive_ptr<c10::ivalue::Object, c10::detail::intrusive_target_default_null_type<c10::ivalue::Object> > >::type std::__invoke_r<c10::intrusive_ptr<c10::ivalue::Object, c10::detail::intrusive_target_default_null_type<c10::ivalue::Object> >, torch::jit::(anonymous namespace)::ScriptModuleDeserializer::readArchive(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)::$_1&, c10::StrongTypePtr, c10::IValue>(torch::jit::(anonymous namespace)::ScriptModuleDeserializer::readArchive(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)::$_1&, c10::StrongTypePtr&&, c10::IValue&&) /usr/bin/../lib/gcc/x86_64-linux-gnu/10/../../../../include/c++/10/bits/invoke.h:113:9 #15 0xec2b8ae in std::_Function_handler<c10::intrusive_ptr<c10::ivalue::Object, c10::detail::intrusive_target_default_null_type<c10::ivalue::Object> > (c10::StrongTypePtr, c10::IValue), torch::jit::(anonymous namespace)::ScriptModuleDeserializer::readArchive(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)::$_1>::_M_invoke(std::_Any_data const&, c10::StrongTypePtr&&, c10::IValue&&) /usr/bin/../lib/gcc/x86_64-linux-gnu/10/../../../../include/c++/10/bits/std_function.h:291:9 #16 0xeda0c63 in std::function<c10::intrusive_ptr<c10::ivalue::Object, c10::detail::intrusive_target_default_null_type<c10::ivalue::Object> > (c10::StrongTypePtr, c10::IValue)>::operator()(c10::StrongTypePtr, c10::IValue) const /usr/bin/../lib/gcc/x86_64-linux-gnu/10/../../../../include/c++/10/bits/std_function.h:622:14 #17 0xed8062d in torch::jit::Unpickler::readGlobal(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)::$_9::operator()() const /pytorch/torch/csrc/jit/serialization/unpickler.cpp:863:20 #18 0xed8062d in void std::__invoke_impl<void, torch::jit::Unpickler::readGlobal(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)::$_9&>(std::__invoke_other, torch::jit::Unpickler::readGlobal(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)::$_9&) /usr/bin/../lib/gcc/x86_64-linux-gnu/10/../../../../include/c++/10/bits/invoke.h:60:14 #19 0xed877c6 in torch::jit::Unpickler::readInstruction() /pytorch/torch/csrc/jit/serialization/unpickler.cpp:545:7 #20 0xed85b27 in torch::jit::Unpickler::run() /pytorch/torch/csrc/jit/serialization/unpickler.cpp:253:27 #21 0xed85781 in torch::jit::Unpickler::parse_ivalue() /pytorch/torch/csrc/jit/serialization/unpickler.cpp:206:3 #22 0xec9c7be in torch::jit::readArchiveAndTensors(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, c10::optional<std::function<c10::StrongTypePtr (c10::QualifiedName const&)> >, c10::optional<std::function<c10::intrusive_ptr<c10::ivalue::Object, c10::detail::intrusive_target_default_null_type<c10::ivalue::Object> > (c10::StrongTypePtr, c10::IValue)> >, c10::optional<c10::Device>, caffe2::serialize::PyTorchStreamReader&, c10::Type::SingletonOrSharedTypePtr<c10::Type> (*)(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&), std::shared_ptr<torch::jit::DeserializationStorageContext>) /pytorch/torch/csrc/jit/serialization/import_read.cpp:53:20 #23 0xec2b168 in torch::jit::(anonymous namespace)::ScriptModuleDeserializer::readArchive(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) /pytorch/torch/csrc/jit/serialization/import.cpp:184:10 #24 0xec27235 in torch::jit::(anonymous namespace)::ScriptModuleDeserializer::deserialize(c10::optional<c10::Device>, std::unordered_map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::hash<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::equal_to<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > > >&, bool) /pytorch/torch/csrc/jit/serialization/import.cpp:287:19 #25 0xec25644 in torch::jit::import_ir_module(std::shared_ptr<torch::jit::CompilationUnit>, std::istream&, c10::optional<c10::Device>, std::unordered_map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::hash<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::equal_to<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > > >&, bool, bool) /pytorch/torch/csrc/jit/serialization/import.cpp:389:25 #26 0xec2dcbe in torch::jit::import_ir_module(std::shared_ptr<torch::jit::CompilationUnit>, std::istream&, c10::optional<c10::Device>, bool) /pytorch/torch/csrc/jit/serialization/import.cpp:325:10 #27 0xec30659 in torch::jit::load(std::istream&, c10::optional<c10::Device>, bool) /pytorch/torch/csrc/jit/serialization/import.cpp:485:10 #28 0x8d8636 in LLVMFuzzerTestOneInput /load.cc:42:14 #29 0x8d835d in ExecuteFilesOnyByOne /AFLplusplus/utils/aflpp_driver/aflpp_driver.c:255:7 #30 0x8d8168 in LLVMFuzzerRunDriver /AFLplusplus/utils/aflpp_driver/aflpp_driver.c #31 0x8d7d28 in main /AFLplusplus/utils/aflpp_driver/aflpp_driver.c:300:10 #32 0x7ffff7a37082 in __libc_start_main (/lib/x86_64-linux-gnu/libc.so.6+0x24082) (BuildId: 1878e6b475720c7c51969e69ab2d276fae6d1dee) #33 0x817add in _start (/load_afl+0x817add) AddressSanitizer can not provide additional info. SUMMARY: AddressSanitizer: SEGV /pytorch/c10/util/Optional.h:222:17 in c10::optional_base<at::Tensor>::optional_base(c10::optional_base<at::Tensor> const&) ==1003193==ABORTING ``` Pull Request resolved: #108418 Approved by: https://github.com/Skylion007
pytorchmergebot
pushed a commit
that referenced
this pull request
Dec 5, 2023
… to hang (#115124) Let's see if it helps #114913 The issues on llvm are at llvm/llvm-project#55530 and llvm/llvm-project#69369. In my CI test, I saw the following process hanged: ``` /pytorch/pytorch/.lintbin/clang-tidy -p=/pytorch/pytorch/build --extra-arg -I/usr/lib/llvm-11/include/openmp --extra-arg -I/opt/conda/envs/py_3.9/include/python3.9 --extra-arg -I/pytorch/pytorch/third_party/pybind11/include --extra-arg -I/usr/bin/../lib/gcc/x86_64-linux-gnu/11/../../../../include/c++/11 --extra-arg -I/usr/bin/../lib/gcc/x86_64-linux-gnu/11/../../../../include/x86_64-linux-gnu/c++/11 --extra-arg -I/usr/bin/../lib/gcc/x86_64-linux-gnu/11/../../../../include/c++/11/backward --extra-arg -I/usr/lib/llvm-14/lib/clang/14.0.0/include --extra-arg -I/usr/local/include --extra-arg -I/usr/include/x86_64-linux-gnu --extra-arg -I/usr/include /pytorch/pytorch/torch/csrc/autograd/python_nested_functions_manual.cpp ``` and the core dump matches the description found in llvm/llvm-project#69369 showing the stuck in `clang::tidy::bugprone::UncheckedOptionalAccessCheck::check`: ``` #0 0x00000000030c7420 in clang::dataflow::WatchedLiteralsSolverImpl::updateWatchedLiterals() () #1 0x00000000030c6c2a in clang::dataflow::WatchedLiteralsSolverImpl::solve() && () #2 0x00000000030c6572 in clang::dataflow::WatchedLiteralsSolver::solve(llvm::DenseSet<clang::dataflow::BoolValue*, llvm::DenseMapInfo<clang::dataflow::BoolValue*, void> >) () #3 0x00000000030b3bd3 in clang::dataflow::DataflowAnalysisContext::querySolver(llvm::DenseSet<clang::dataflow::BoolValue*, llvm::DenseMapInfo<clang::dataflow::BoolValue*, void> >) () #4 0x00000000030b3ca5 in clang::dataflow::DataflowAnalysisContext::flowConditionImplies(clang::dataflow::AtomicBoolValue&, clang::dataflow::BoolValue&) () #5 0x00000000030b1213 in clang::dataflow::(anonymous namespace)::diagnoseUnwrapCall(clang::Expr const*, clang::Expr const*, clang::dataflow::Environment const&) () #6 0x00000000030b1357 in std::_Function_handler<std::vector<clang::SourceLocation, std::allocator<clang::SourceLocation> > (clang::CallExpr const*, clang::ast_matchers::MatchFinder::MatchResult const&, clang::dataflow::Environment const&), clang::dataflow::(anonymous namespace)::buildDiagnoseMatchSwitch(clang::dataflow::UncheckedOptionalAccessModelOptions const&)::$_7>::_M_invoke(std::_Any_data const&, clang::CallExpr const*&&, clang::ast_matchers::MatchFinder::MatchResult const&, clang::dataflow::Environment const&) () #7 0x00000000030b1292 in std::_Function_handler<std::vector<clang::SourceLocation, std::allocator<clang::SourceLocation> > (clang::Stmt const*, clang::ast_matchers::MatchFinder::MatchResult const&, clang::dataflow::Environment const&), clang::dataflow::MatchSwitchBuilder<clang::dataflow::Environment const, std::vector<clang::SourceLocation, std::allocator<clang::SourceLocation> > >::CaseOf<clang::CallExpr>(clang::ast_matchers::internal::Matcher<clang::Stmt>, std::function<std::vector<clang::SourceLocation, std::allocator<clang::SourceLocation> > (clang::CallExpr const*, clang::ast_matchers::MatchFinder::MatchResult const&, clang::dataflow::Environment const&)>) &&::{lambda(clang::Stmt const*, clang::ast_matchers::MatchFinder::MatchResult const&, clang::dataflow::Environment const&)#1}>::_M_invoke(std::_Any_data const&, clang::Stmt const*&&, clang::ast_matchers::MatchFinder::MatchResult const&, clang::dataflow::Environment const&) () #8 0x00000000030b1995 in clang::dataflow::MatchSwitchBuilder<clang::dataflow::Environment const, std::vector<clang::SourceLocation, std::allocator<clang::SourceLocation> > >::Build() &&::{lambda(clang::Stmt const&, clang::ASTContext&, clang::dataflow::Environment const&)#1}::operator()(clang::Stmt const&, clang::ASTContext&, clang::dataflow::Environment const&) const () #9 0x00000000030b170c in std::_Function_handler<std::vector<clang::SourceLocation, std::allocator<clang::SourceLocation> > (clang::Stmt const&, clang::ASTContext&, clang::dataflow::Environment const&), clang::dataflow::MatchSwitchBuilder<clang::dataflow::Environment const, std::vector<clang::SourceLocation, std::allocator<clang::SourceLocation> > >::Build() &&::{lambda(clang::Stmt const&, clang::ASTContext&, clang::dataflow::Environment const&)#1}>::_M_invoke(std::_Any_data const&, clang::Stmt const&, clang::ASTContext&, clang::dataflow::Environment const&) () #10 0x00000000030a7c27 in clang::dataflow::UncheckedOptionalAccessDiagnoser::diagnose(clang::ASTContext&, clang::Stmt const*, clang::dataflow::Environment const&) () #11 0x0000000002931286 in std::_Function_handler<void (clang::Stmt const*, clang::dataflow::DataflowAnalysisState<clang::dataflow::NoopLattice> const&), clang::tidy::bugprone::analyzeFunction(clang::FunctionDecl const&, clang::ASTContext&)::$_0>::_M_invoke(std::_Any_data const&, clang::Stmt const*&&, clang::dataflow::DataflowAnalysisState<clang::dataflow::NoopLattice> const&) () #12 0x0000000002930b41 in clang::dataflow::runDataflowAnalysis<clang::dataflow::UncheckedOptionalAccessModel>(clang::dataflow::ControlFlowContext const&, clang::dataflow::UncheckedOptionalAccessModel&, clang::dataflow::Environment const&, std::function<void (clang::Stmt const*, clang::dataflow::DataflowAnalysisState<clang::dataflow::UncheckedOptionalAccessModel::Lattice> const&)>)::{lambda(clang::Stmt const*, clang::dataflow::TypeErasedDataflowAnalysisState const&)#1}::operator()(clang::Stmt const*, clang::dataflow::TypeErasedDataflowAnalysisState const&) const () #13 0x00000000030c18cc in std::_Function_handler<void (clang::CFGStmt const&, clang::dataflow::TypeErasedDataflowAnalysisState const&), clang::dataflow::runTypeErasedDataflowAnalysis(clang::dataflow::ControlFlowContext const&, clang::dataflow::TypeErasedDataflowAnalysis&, clang::dataflow::Environment const&, std::function<void (clang::Stmt const*, clang::dataflow::TypeErasedDataflowAnalysisState const&)>)::$_1>::_M_invoke(std::_Any_data const&, clang::CFGStmt const&, clang::dataflow::TypeErasedDataflowAnalysisState const&) () #14 0x00000000030bf069 in clang::dataflow::transferBlock(clang::dataflow::ControlFlowContext const&, std::vector<llvm::Optional<clang::dataflow::TypeErasedDataflowAnalysisState>, std::allocator<llvm::Optional<clang::dataflow::TypeErasedDataflowAnalysisState> > >&, clang::CFGBlock const&, clang::dataflow::Environment const&, clang::dataflow::TypeErasedDataflowAnalysis&, std::function<void (clang::CFGStmt const&, clang::dataflow::TypeErasedDataflowAnalysisState const&)>) () #15 0x00000000030bfaa5 in clang::dataflow::runTypeErasedDataflowAnalysis(clang::dataflow::ControlFlowContext const&, clang::dataflow::TypeErasedDataflowAnalysis&, clang::dataflow::Environment const&, std::function<void (clang::Stmt const*, clang::dataflow::TypeErasedDataflowAnalysisState const&)>) () #16 0x00000000029301b3 in llvm::Expected<std::vector<llvm::Optional<clang::dataflow::DataflowAnalysisState<clang::dataflow::UncheckedOptionalAccessModel::Lattice> >, std::allocator<llvm::Optional<clang::dataflow::DataflowAnalysisState<clang::dataflow::UncheckedOptionalAccessModel::Lattice> > > > > clang::dataflow::runDataflowAnalysis<clang::dataflow::UncheckedOptionalAccessModel>(clang::dataflow::ControlFlowContext const&, clang::dataflow::UncheckedOptionalAccessModel&, clang::dataflow::Environment const&, std::function<void (clang::Stmt const*, clang::dataflow::DataflowAnalysisState<clang::dataflow::UncheckedOptionalAccessModel::Lattice> const&)>) () #17 0x000000000292fbe8 in clang::tidy::bugprone::UncheckedOptionalAccessCheck::check(clang::ast_matchers::MatchFinder::MatchResult const&) () #18 0x00000000022e1572 in clang::ast_matchers::internal::(anonymous namespace)::MatchASTVisitor::MatchVisitor::visitMatch(clang::ast_matchers::BoundNodes const&) () #19 0x0000000002797a1c in clang::ast_matchers::internal::BoundNodesTreeBuilder::visitMatches(clang::ast_matchers::internal::BoundNodesTreeBuilder::Visitor*) () #20 0x00000000022e0dc6 in clang::ast_matchers::internal::(anonymous namespace)::MatchASTVisitor::matchWithFilter(clang::DynTypedNode const&) () #21 0x00000000022e3b57 in clang::ast_matchers::internal::(anonymous namespace)::MatchASTVisitor::TraverseDecl(clang::Decl*) () #22 0x00000000022e4c0c in clang::RecursiveASTVisitor<clang::ast_matchers::internal::(anonymous namespace)::MatchASTVisitor>::TraverseDecl(clang::Decl*) () #23 0x00000000022e3b62 in clang::ast_matchers::internal::(anonymous namespace)::MatchASTVisitor::TraverseDecl(clang::Decl*) () #24 0x00000000022e4c0c in clang::RecursiveASTVisitor<clang::ast_matchers::internal::(anonymous namespace)::MatchASTVisitor>::TraverseDecl(clang::Decl*) () #25 0x00000000022e3b62 in clang::ast_matchers::internal::(anonymous namespace)::MatchASTVisitor::TraverseDecl(clang::Decl*) () #26 0x00000000022e4c0c in clang::RecursiveASTVisitor<clang::ast_matchers::internal::(anonymous namespace)::MatchASTVisitor>::TraverseDecl(clang::Decl*) () #27 0x00000000022e3b62 in clang::ast_matchers::internal::(anonymous namespace)::MatchASTVisitor::TraverseDecl(clang::Decl*) () #28 0x00000000022e4c0c in clang::RecursiveASTVisitor<clang::ast_matchers::internal::(anonymous namespace)::MatchASTVisitor>::TraverseDecl(clang::Decl*) () #29 0x00000000022e3b62 in clang::ast_matchers::internal::(anonymous namespace)::MatchASTVisitor::TraverseDecl(clang::Decl*) () #30 0x00000000022e8791 in clang::RecursiveASTVisitor<clang::ast_matchers::internal::(anonymous namespace)::MatchASTVisitor>::TraverseDecl(clang::Decl*) () #31 0x00000000022e3b62 in clang::ast_matchers::internal::(anonymous namespace)::MatchASTVisitor::TraverseDecl(clang::Decl*) () #32 0x00000000022c017a in clang::ast_matchers::MatchFinder::matchAST(clang::ASTContext&) () #33 0x000000000370ad3c in clang::MultiplexConsumer::HandleTranslationUnit(clang::ASTContext&) () #34 0x00000000038ed4bb in clang::ParseAST(clang::Sema&, bool, bool) () #35 0x000000000369eda7 in clang::FrontendAction::Execute() () #36 0x000000000360d3f6 in clang::CompilerInstance::ExecuteAction(clang::FrontendAction&) () #37 0x00000000027c475c in clang::tooling::FrontendActionFactory::runInvocation(std::shared_ptr<clang::CompilerInvocation>, clang::FileManager*, std::shared_ptr<clang::PCHContainerOperations>, clang::DiagnosticConsumer*) () #38 0x00000000022ad486 in clang::tidy::runClangTidy(clang::tidy::ClangTidyContext&, clang::tooling::CompilationDatabase const&, llvm::ArrayRef<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, llvm::IntrusiveRefCntPtr<llvm::vfs::OverlayFileSystem>, bool, bool, llvm::StringRef)::ActionFactory::runInvocation(std::shared_ptr<clang::CompilerInvocation>, clang::FileManager*, std::shared_ptr<clang::PCHContainerOperations>, clang::DiagnosticConsumer*) () #39 0x00000000027c44c6 in clang::tooling::ToolInvocation::runInvocation(char const*, clang::driver::Compilation*, std::shared_ptr<clang::CompilerInvocation>, std::shared_ptr<clang::PCHContainerOperations>) () #40 0x00000000027c360b in clang::tooling::ToolInvocation::run() () #41 0x00000000027c5bb1 in clang::tooling::ClangTool::run(clang::tooling::ToolAction*) () #42 0x00000000022a90c7 in clang::tidy::runClangTidy(clang::tidy::ClangTidyContext&, clang::tooling::CompilationDatabase const&, llvm::ArrayRef<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, llvm::IntrusiveRefCntPtr<llvm::vfs::OverlayFileSystem>, bool, bool, llvm::StringRef) () #43 0x0000000001ebc7f2 in clang::tidy::clangTidyMain(int, char const**) () #44 0x0000000004c54ba0 in __libc_start_main () #45 0x0000000001eb76ae in _start () ``` Another note is that clang-tidy is CPU-bound. So we could consider running lintrunner job on 4xlarge if needed. Pull Request resolved: #115124 Approved by: https://github.com/kit1980, https://github.com/Skylion007, https://github.com/malfet
dmenig
pushed a commit
to dmenig/pytorch
that referenced
this pull request
Dec 21, 2023
… to hang (pytorch#115124) Let's see if it helps pytorch#114913 The issues on llvm are at llvm/llvm-project#55530 and llvm/llvm-project#69369. In my CI test, I saw the following process hanged: ``` /pytorch/pytorch/.lintbin/clang-tidy -p=/pytorch/pytorch/build --extra-arg -I/usr/lib/llvm-11/include/openmp --extra-arg -I/opt/conda/envs/py_3.9/include/python3.9 --extra-arg -I/pytorch/pytorch/third_party/pybind11/include --extra-arg -I/usr/bin/../lib/gcc/x86_64-linux-gnu/11/../../../../include/c++/11 --extra-arg -I/usr/bin/../lib/gcc/x86_64-linux-gnu/11/../../../../include/x86_64-linux-gnu/c++/11 --extra-arg -I/usr/bin/../lib/gcc/x86_64-linux-gnu/11/../../../../include/c++/11/backward --extra-arg -I/usr/lib/llvm-14/lib/clang/14.0.0/include --extra-arg -I/usr/local/include --extra-arg -I/usr/include/x86_64-linux-gnu --extra-arg -I/usr/include /pytorch/pytorch/torch/csrc/autograd/python_nested_functions_manual.cpp ``` and the core dump matches the description found in llvm/llvm-project#69369 showing the stuck in `clang::tidy::bugprone::UncheckedOptionalAccessCheck::check`: ``` #0 0x00000000030c7420 in clang::dataflow::WatchedLiteralsSolverImpl::updateWatchedLiterals() () pytorch#1 0x00000000030c6c2a in clang::dataflow::WatchedLiteralsSolverImpl::solve() && () pytorch#2 0x00000000030c6572 in clang::dataflow::WatchedLiteralsSolver::solve(llvm::DenseSet<clang::dataflow::BoolValue*, llvm::DenseMapInfo<clang::dataflow::BoolValue*, void> >) () pytorch#3 0x00000000030b3bd3 in clang::dataflow::DataflowAnalysisContext::querySolver(llvm::DenseSet<clang::dataflow::BoolValue*, llvm::DenseMapInfo<clang::dataflow::BoolValue*, void> >) () pytorch#4 0x00000000030b3ca5 in clang::dataflow::DataflowAnalysisContext::flowConditionImplies(clang::dataflow::AtomicBoolValue&, clang::dataflow::BoolValue&) () pytorch#5 0x00000000030b1213 in clang::dataflow::(anonymous namespace)::diagnoseUnwrapCall(clang::Expr const*, clang::Expr const*, clang::dataflow::Environment const&) () pytorch#6 0x00000000030b1357 in std::_Function_handler<std::vector<clang::SourceLocation, std::allocator<clang::SourceLocation> > (clang::CallExpr const*, clang::ast_matchers::MatchFinder::MatchResult const&, clang::dataflow::Environment const&), clang::dataflow::(anonymous namespace)::buildDiagnoseMatchSwitch(clang::dataflow::UncheckedOptionalAccessModelOptions const&)::$_7>::_M_invoke(std::_Any_data const&, clang::CallExpr const*&&, clang::ast_matchers::MatchFinder::MatchResult const&, clang::dataflow::Environment const&) () pytorch#7 0x00000000030b1292 in std::_Function_handler<std::vector<clang::SourceLocation, std::allocator<clang::SourceLocation> > (clang::Stmt const*, clang::ast_matchers::MatchFinder::MatchResult const&, clang::dataflow::Environment const&), clang::dataflow::MatchSwitchBuilder<clang::dataflow::Environment const, std::vector<clang::SourceLocation, std::allocator<clang::SourceLocation> > >::CaseOf<clang::CallExpr>(clang::ast_matchers::internal::Matcher<clang::Stmt>, std::function<std::vector<clang::SourceLocation, std::allocator<clang::SourceLocation> > (clang::CallExpr const*, clang::ast_matchers::MatchFinder::MatchResult const&, clang::dataflow::Environment const&)>) &&::{lambda(clang::Stmt const*, clang::ast_matchers::MatchFinder::MatchResult const&, clang::dataflow::Environment const&)pytorch#1}>::_M_invoke(std::_Any_data const&, clang::Stmt const*&&, clang::ast_matchers::MatchFinder::MatchResult const&, clang::dataflow::Environment const&) () pytorch#8 0x00000000030b1995 in clang::dataflow::MatchSwitchBuilder<clang::dataflow::Environment const, std::vector<clang::SourceLocation, std::allocator<clang::SourceLocation> > >::Build() &&::{lambda(clang::Stmt const&, clang::ASTContext&, clang::dataflow::Environment const&)pytorch#1}::operator()(clang::Stmt const&, clang::ASTContext&, clang::dataflow::Environment const&) const () pytorch#9 0x00000000030b170c in std::_Function_handler<std::vector<clang::SourceLocation, std::allocator<clang::SourceLocation> > (clang::Stmt const&, clang::ASTContext&, clang::dataflow::Environment const&), clang::dataflow::MatchSwitchBuilder<clang::dataflow::Environment const, std::vector<clang::SourceLocation, std::allocator<clang::SourceLocation> > >::Build() &&::{lambda(clang::Stmt const&, clang::ASTContext&, clang::dataflow::Environment const&)pytorch#1}>::_M_invoke(std::_Any_data const&, clang::Stmt const&, clang::ASTContext&, clang::dataflow::Environment const&) () pytorch#10 0x00000000030a7c27 in clang::dataflow::UncheckedOptionalAccessDiagnoser::diagnose(clang::ASTContext&, clang::Stmt const*, clang::dataflow::Environment const&) () pytorch#11 0x0000000002931286 in std::_Function_handler<void (clang::Stmt const*, clang::dataflow::DataflowAnalysisState<clang::dataflow::NoopLattice> const&), clang::tidy::bugprone::analyzeFunction(clang::FunctionDecl const&, clang::ASTContext&)::$_0>::_M_invoke(std::_Any_data const&, clang::Stmt const*&&, clang::dataflow::DataflowAnalysisState<clang::dataflow::NoopLattice> const&) () pytorch#12 0x0000000002930b41 in clang::dataflow::runDataflowAnalysis<clang::dataflow::UncheckedOptionalAccessModel>(clang::dataflow::ControlFlowContext const&, clang::dataflow::UncheckedOptionalAccessModel&, clang::dataflow::Environment const&, std::function<void (clang::Stmt const*, clang::dataflow::DataflowAnalysisState<clang::dataflow::UncheckedOptionalAccessModel::Lattice> const&)>)::{lambda(clang::Stmt const*, clang::dataflow::TypeErasedDataflowAnalysisState const&)pytorch#1}::operator()(clang::Stmt const*, clang::dataflow::TypeErasedDataflowAnalysisState const&) const () pytorch#13 0x00000000030c18cc in std::_Function_handler<void (clang::CFGStmt const&, clang::dataflow::TypeErasedDataflowAnalysisState const&), clang::dataflow::runTypeErasedDataflowAnalysis(clang::dataflow::ControlFlowContext const&, clang::dataflow::TypeErasedDataflowAnalysis&, clang::dataflow::Environment const&, std::function<void (clang::Stmt const*, clang::dataflow::TypeErasedDataflowAnalysisState const&)>)::$_1>::_M_invoke(std::_Any_data const&, clang::CFGStmt const&, clang::dataflow::TypeErasedDataflowAnalysisState const&) () pytorch#14 0x00000000030bf069 in clang::dataflow::transferBlock(clang::dataflow::ControlFlowContext const&, std::vector<llvm::Optional<clang::dataflow::TypeErasedDataflowAnalysisState>, std::allocator<llvm::Optional<clang::dataflow::TypeErasedDataflowAnalysisState> > >&, clang::CFGBlock const&, clang::dataflow::Environment const&, clang::dataflow::TypeErasedDataflowAnalysis&, std::function<void (clang::CFGStmt const&, clang::dataflow::TypeErasedDataflowAnalysisState const&)>) () pytorch#15 0x00000000030bfaa5 in clang::dataflow::runTypeErasedDataflowAnalysis(clang::dataflow::ControlFlowContext const&, clang::dataflow::TypeErasedDataflowAnalysis&, clang::dataflow::Environment const&, std::function<void (clang::Stmt const*, clang::dataflow::TypeErasedDataflowAnalysisState const&)>) () pytorch#16 0x00000000029301b3 in llvm::Expected<std::vector<llvm::Optional<clang::dataflow::DataflowAnalysisState<clang::dataflow::UncheckedOptionalAccessModel::Lattice> >, std::allocator<llvm::Optional<clang::dataflow::DataflowAnalysisState<clang::dataflow::UncheckedOptionalAccessModel::Lattice> > > > > clang::dataflow::runDataflowAnalysis<clang::dataflow::UncheckedOptionalAccessModel>(clang::dataflow::ControlFlowContext const&, clang::dataflow::UncheckedOptionalAccessModel&, clang::dataflow::Environment const&, std::function<void (clang::Stmt const*, clang::dataflow::DataflowAnalysisState<clang::dataflow::UncheckedOptionalAccessModel::Lattice> const&)>) () pytorch#17 0x000000000292fbe8 in clang::tidy::bugprone::UncheckedOptionalAccessCheck::check(clang::ast_matchers::MatchFinder::MatchResult const&) () pytorch#18 0x00000000022e1572 in clang::ast_matchers::internal::(anonymous namespace)::MatchASTVisitor::MatchVisitor::visitMatch(clang::ast_matchers::BoundNodes const&) () pytorch#19 0x0000000002797a1c in clang::ast_matchers::internal::BoundNodesTreeBuilder::visitMatches(clang::ast_matchers::internal::BoundNodesTreeBuilder::Visitor*) () pytorch#20 0x00000000022e0dc6 in clang::ast_matchers::internal::(anonymous namespace)::MatchASTVisitor::matchWithFilter(clang::DynTypedNode const&) () pytorch#21 0x00000000022e3b57 in clang::ast_matchers::internal::(anonymous namespace)::MatchASTVisitor::TraverseDecl(clang::Decl*) () pytorch#22 0x00000000022e4c0c in clang::RecursiveASTVisitor<clang::ast_matchers::internal::(anonymous namespace)::MatchASTVisitor>::TraverseDecl(clang::Decl*) () pytorch#23 0x00000000022e3b62 in clang::ast_matchers::internal::(anonymous namespace)::MatchASTVisitor::TraverseDecl(clang::Decl*) () pytorch#24 0x00000000022e4c0c in clang::RecursiveASTVisitor<clang::ast_matchers::internal::(anonymous namespace)::MatchASTVisitor>::TraverseDecl(clang::Decl*) () pytorch#25 0x00000000022e3b62 in clang::ast_matchers::internal::(anonymous namespace)::MatchASTVisitor::TraverseDecl(clang::Decl*) () pytorch#26 0x00000000022e4c0c in clang::RecursiveASTVisitor<clang::ast_matchers::internal::(anonymous namespace)::MatchASTVisitor>::TraverseDecl(clang::Decl*) () pytorch#27 0x00000000022e3b62 in clang::ast_matchers::internal::(anonymous namespace)::MatchASTVisitor::TraverseDecl(clang::Decl*) () pytorch#28 0x00000000022e4c0c in clang::RecursiveASTVisitor<clang::ast_matchers::internal::(anonymous namespace)::MatchASTVisitor>::TraverseDecl(clang::Decl*) () pytorch#29 0x00000000022e3b62 in clang::ast_matchers::internal::(anonymous namespace)::MatchASTVisitor::TraverseDecl(clang::Decl*) () pytorch#30 0x00000000022e8791 in clang::RecursiveASTVisitor<clang::ast_matchers::internal::(anonymous namespace)::MatchASTVisitor>::TraverseDecl(clang::Decl*) () pytorch#31 0x00000000022e3b62 in clang::ast_matchers::internal::(anonymous namespace)::MatchASTVisitor::TraverseDecl(clang::Decl*) () pytorch#32 0x00000000022c017a in clang::ast_matchers::MatchFinder::matchAST(clang::ASTContext&) () pytorch#33 0x000000000370ad3c in clang::MultiplexConsumer::HandleTranslationUnit(clang::ASTContext&) () pytorch#34 0x00000000038ed4bb in clang::ParseAST(clang::Sema&, bool, bool) () pytorch#35 0x000000000369eda7 in clang::FrontendAction::Execute() () pytorch#36 0x000000000360d3f6 in clang::CompilerInstance::ExecuteAction(clang::FrontendAction&) () pytorch#37 0x00000000027c475c in clang::tooling::FrontendActionFactory::runInvocation(std::shared_ptr<clang::CompilerInvocation>, clang::FileManager*, std::shared_ptr<clang::PCHContainerOperations>, clang::DiagnosticConsumer*) () pytorch#38 0x00000000022ad486 in clang::tidy::runClangTidy(clang::tidy::ClangTidyContext&, clang::tooling::CompilationDatabase const&, llvm::ArrayRef<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, llvm::IntrusiveRefCntPtr<llvm::vfs::OverlayFileSystem>, bool, bool, llvm::StringRef)::ActionFactory::runInvocation(std::shared_ptr<clang::CompilerInvocation>, clang::FileManager*, std::shared_ptr<clang::PCHContainerOperations>, clang::DiagnosticConsumer*) () pytorch#39 0x00000000027c44c6 in clang::tooling::ToolInvocation::runInvocation(char const*, clang::driver::Compilation*, std::shared_ptr<clang::CompilerInvocation>, std::shared_ptr<clang::PCHContainerOperations>) () pytorch#40 0x00000000027c360b in clang::tooling::ToolInvocation::run() () pytorch#41 0x00000000027c5bb1 in clang::tooling::ClangTool::run(clang::tooling::ToolAction*) () pytorch#42 0x00000000022a90c7 in clang::tidy::runClangTidy(clang::tidy::ClangTidyContext&, clang::tooling::CompilationDatabase const&, llvm::ArrayRef<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, llvm::IntrusiveRefCntPtr<llvm::vfs::OverlayFileSystem>, bool, bool, llvm::StringRef) () pytorch#43 0x0000000001ebc7f2 in clang::tidy::clangTidyMain(int, char const**) () pytorch#44 0x0000000004c54ba0 in __libc_start_main () pytorch#45 0x0000000001eb76ae in _start () ``` Another note is that clang-tidy is CPU-bound. So we could consider running lintrunner job on 4xlarge if needed. Pull Request resolved: pytorch#115124 Approved by: https://github.com/kit1980, https://github.com/Skylion007, https://github.com/malfet
pytorch-bot bot
pushed a commit
that referenced
this pull request
Jan 10, 2024
Fwd prefetch enable grad weif
pytorchmergebot
pushed a commit
that referenced
this pull request
Feb 2, 2024
user may not know which line of code called collectives in a big code base. When debugging, we can print python-cpp stacktrace in case user call ``ProcessGroup.reduce`` instead of ``torch.distributed.reduce``
```
LOG(INFO) << "ProcessGroupNCCL::_allgather_base stacktrace: "
<< get_python_cpp_trace();
```
output (using _allgather_base as an example): one example python-part trace is ``all_gather_into_tensor from /data/users/weif/pytorch/torch/distributed/distributed_c10d.py:2838``
```
ProcessGroupNCCL::_allgather_base stacktrace: #0 torch::unwind::unwind() from ??:0
#1 torch::CapturedTraceback::gather(bool, bool, bool) from ??:0
#2 c10d::get_python_cpp_trace[abi:cxx11]() from :0
#3 c10d::ProcessGroupNCCL::_allgather_base(at::Tensor&, at::Tensor&, c10d::AllgatherOptions const&) from ??:0
#4 c10d::ops::(anonymous namespace)::_allgather_base_CUDA(at::Tensor&, at::Tensor&, c10::intrusive_ptr<c10d::ProcessGroup, c10::detail::intrusive_target_default_null_type<c10d::ProcessGroup> > const&, bool, long) from Ops.cpp:0
#5 c10::impl::make_boxed_from_unboxed_functor<c10::impl::detail::WrapFunctionIntoRuntimeFunctor_<std::tuple<at::Tensor, c10::intrusive_ptr<c10d::Work, c10::detail::intrusive_target_default_null_type<c10d::Work> > > (*)(at::Tensor&, at::Tensor&, c10::intrusive_ptr<c10d::ProcessGroup, c10::detail::intrusive_target_default_null_type<c10d::ProcessGroup> > const&, bool, long), std::tuple<at::Tensor, c10::intrusive_ptr<c10d::Work, c10::detail::intrusive_target_default_null_type<c10d::Work> > >, c10::guts::typelist::typelist<at::Tensor&, at::Tensor&, c10::intrusive_ptr<c10d::ProcessGroup, c10::detail::intrusive_target_default_null_type<c10d::ProcessGroup> > const&, bool, long> >, false>::call(c10::OperatorKernel*, c10::OperatorHandle const&, c10::DispatchKeySet, std::vector<c10::IValue, std::allocator<c10::IValue> >*) from :0
#6 torch::autograd::basicAutogradNotImplementedFallbackImpl(c10::OperatorHandle const&, c10::DispatchKeySet, std::vector<c10::IValue, std::allocator<c10::IValue> >*) from autograd_not_implemented_fallback.cpp:0
#7 c10d::ProcessGroup::_allgather_base(at::Tensor&, at::Tensor&, c10d::AllgatherOptions const&) from :0
#8 pybind11::cpp_function::initialize<pybind11::cpp_function::initialize<c10::intrusive_ptr<c10d::Work, c10::detail::intrusive_target_default_null_type<c10d::Work> >, c10d::ProcessGroup, at::Tensor&, at::Tensor&, c10d::AllgatherOptions const&, pybind11::name, pybind11::is_method, pybind11::sibling, pybind11::arg, pybind11::arg, pybind11::arg_v, pybind11::call_guard<pybind11::gil_scoped_release> >(c10::intrusive_ptr<c10d::Work, c10::detail::intrusive_target_default_null_type<c10d::Work> > (c10d::ProcessGroup::*)(at::Tensor&, at::Tensor&, c10d::AllgatherOptions const&), pybind11::name const&, pybind11::is_method const&, pybind11::sibling const&, pybind11::arg const&, pybind11::arg const&, pybind11::arg_v const&, pybind11::call_guard<pybind11::gil_scoped_release> const&)::{lambda(c10d::ProcessGroup*, at::Tensor&, at::Tensor&, c10d::AllgatherOptions const&)#1}, c10::intrusive_ptr<c10d::Work, c10::detail::intrusive_target_default_null_type<c10d::Work> >, c10d::ProcessGroup*, at::Tensor&, at::Tensor&, c10d::AllgatherOptions const&, pybind11::name, pybind11::is_method, pybind11::sibling, pybind11::arg, pybind11::arg, pybind11::arg_v, pybind11::call_guard<pybind11::gil_scoped_release> >(pybind11::cpp_function::initialize<c10::intrusive_ptr<c10d::Work, c10::detail::intrusive_target_default_null_type<c10d::Work> >, c10d::ProcessGroup, at::Tensor&, at::Tensor&, c10d::AllgatherOptions const&, pybind11::name, pybind11::is_method, pybind11::sibling, pybind11::arg, pybind11::arg, pybind11::arg_v, pybind11::call_guard<pybind11::gil_scoped_release> >(c10::intrusive_ptr<c10d::Work, c10::detail::intrusive_target_default_null_type<c10d::Work> > (c10d::ProcessGroup::*)(at::Tensor&, at::Tensor&, c10d::AllgatherOptions const&), pybind11::name const&, pybind11::is_method const&, pybind11::sibling const&, pybind11::arg const&, pybind11::arg const&, pybind11::arg_v const&, pybind11::call_guard<pybind11::gil_scoped_release> const&)::{lambda(c10d::ProcessGroup*, at::Tensor&, at::Tensor&, c10d::AllgatherOptions const&)#1}&&, c10::intrusive_ptr<c10d::Work, c10::detail::intrusive_target_default_null_type<c10d::Work> > (*)(c10d::ProcessGroup*, at::Tensor&, at::Tensor&, c10d::AllgatherOptions const&), pybind11::name const&, pybind11::is_method const&, pybind11::sibling const&, pybind11::arg const&, pybind11::arg const&, pybind11::arg_v const&, pybind11::call_guard<pybind11::gil_scoped_release> const&)::{lambda(pybind11::detail::function_call&)#3}::_FUN(pybind11::detail::function_call&) from :0
#9 pybind11::cpp_function::dispatcher(_object*, _object*, _object*) from :0
#10 cfunction_call from /usr/local/src/conda/python-3.10.12/Objects/methodobject.c:543
#11 _PyObject_MakeTpCall from /usr/local/src/conda/python-3.10.12/Objects/call.c:215
#12 _PyObject_VectorcallTstate from /usr/local/src/conda/python-3.10.12/Include/cpython/abstract.h:112
#13 _PyObject_VectorcallTstate from /usr/local/src/conda/python-3.10.12/Include/cpython/abstract.h:114
#14 all_gather_into_tensor from /data/users/weif/pytorch/torch/distributed/distributed_c10d.py:2838
#15 _PyEval_EvalFrame from /usr/local/src/conda/python-3.10.12/Include/internal/pycore_ceval.h:46
#16 do_call_core from /usr/local/src/conda/python-3.10.12/Python/ceval.c:5945
#17 wrapper from /data/users/weif/pytorch/torch/distributed/c10d_logger.py:75
#18 _PyEval_EvalFrame from /usr/local/src/conda/python-3.10.12/Include/internal/pycore_ceval.h:46
#19 _PyObject_VectorcallTstate from /usr/local/src/conda/python-3.10.12/Include/cpython/abstract.h:114
#20 _all_gather_flat_param from /data/users/weif/pytorch/torch/distributed/fsdp/_flat_param.py:1399
#21 _PyEval_EvalFrame from /usr/local/src/conda/python-3.10.12/Include/internal/pycore_ceval.h:46
#22 _PyObject_VectorcallTstate from /usr/local/src/conda/python-3.10.12/Include/cpython/abstract.h:114
#23 unshard from /data/users/weif/pytorch/torch/distributed/fsdp/_flat_param.py:1308
#24 _PyEval_EvalFrame from /usr/local/src/conda/python-3.10.12/Include/internal/pycore_ceval.h:46
#25 _PyObject_VectorcallTstate from /usr/local/src/conda/python-3.10.12/Include/cpython/abstract.h:114
#26 _unshard from /data/users/weif/pytorch/torch/distributed/fsdp/_runtime_utils.py:332
#27 _PyEval_EvalFrame from /usr/local/src/conda/python-3.10.12/Include/internal/pycore_ceval.h:46
#28 _PyObject_VectorcallTstate from /usr/local/src/conda/python-3.10.12/Include/cpython/abstract.h:114
#29 _pre_forward_unshard from /data/users/weif/pytorch/torch/distributed/fsdp/_runtime_utils.py:448
#30 _PyEval_EvalFrame from /usr/local/src/conda/python-3.10.12/Include/internal/pycore_ceval.h:46
#31 _PyObject_VectorcallTstate from /usr/local/src/conda/python-3.10.12/Include/cpython/abstract.h:114
#32 _pre_forward from /data/users/weif/pytorch/torch/distributed/fsdp/_runtime_utils.py:413
#33 _PyEval_EvalFrame from /usr/local/src/conda/python-3.10.12/Include/internal/pycore_ceval.h:46
#34 _PyObject_VectorcallTstate from /usr/local/src/conda/python-3.10.12/Include/cpython/abstract.h:114
#35 forward from /data/users/weif/pytorch/torch/distributed/fsdp/fully_sharded_data_parallel.py:839
#36 _PyEval_EvalFrame from /usr/local/src/conda/python-3.10.12/Include/internal/pycore_ceval.h:46
#37 do_call_core from /usr/local/src/conda/python-3.10.12/Python/ceval.c:5945
#38 _call_impl from /data/users/weif/pytorch/torch/nn/modules/module.py:1520
#39 _PyEval_EvalFrame from /usr/local/src/conda/python-3.10.12/Include/internal/pycore_ceval.h:46
#40 do_call_core from /usr/local/src/conda/python-3.10.12/Python/ceval.c:5945
#41 _wrapped_call_impl from /data/users/weif/pytorch/torch/nn/modules/module.py:1511
#42 _PyEval_EvalFrame from /usr/local/src/conda/python-3.10.12/Include/internal/pycore_ceval.h:46
#43 _PyObject_Call_Prepend from /usr/local/src/conda/python-3.10.12/Objects/call.c:431
#44 slot_tp_call from /usr/local/src/conda/python-3.10.12/Objects/typeobject.c:7494
#45 _PyObject_MakeTpCall from /usr/local/src/conda/python-3.10.12/Objects/call.c:215
#46 _PyObject_VectorcallTstate from /usr/local/src/conda/python-3.10.12/Include/cpython/abstract.h:112
#47 inner from /data/users/weif/pytorch/run_fsdp.py:72
#48 _PyEval_EvalFrame from /usr/local/src/conda/python-3.10.12/Include/internal/pycore_ceval.h:46
#49 _PyObject_VectorcallTstate from /usr/local/src/conda/python-3.10.12/Include/cpython/abstract.h:114
#50 run from /data/users/weif/pytorch/run_fsdp.py:76
#51 _PyEval_EvalFrame from /usr/local/src/conda/python-3.10.12/Include/internal/pycore_ceval.h:46
#52 _PyObject_VectorcallTstate from /usr/local/src/conda/python-3.10.12/Include/cpython/abstract.h:114
#53 main from /data/users/weif/pytorch/run_fsdp.py:133
#54 _PyEval_EvalFrame from /usr/local/src/conda/python-3.10.12/Include/internal/pycore_ceval.h:46
#55 _PyObject_VectorcallTstate from /usr/local/src/conda/python-3.10.12/Include/cpython/abstract.h:114
#56 <module> from /data/users/weif/pytorch/run_fsdp.py:137
#57 _PyEval_EvalFrame from /usr/local/src/conda/python-3.10.12/Include/internal/pycore_ceval.h:46
#58 PyEval_EvalCode from /usr/local/src/conda/python-3.10.12/Python/ceval.c:1134
#59 run_eval_code_obj from /usr/local/src/conda/python-3.10.12/Python/pythonrun.c:1291
#60 run_mod from /usr/local/src/conda/python-3.10.12/Python/pythonrun.c:1312
#61 pyrun_file from /usr/local/src/conda/python-3.10.12/Python/pythonrun.c:1208
#62 _PyRun_SimpleFileObject from /usr/local/src/conda/python-3.10.12/Python/pythonrun.c:456
#63 _PyRun_AnyFileObject from /usr/local/src/conda/python-3.10.12/Python/pythonrun.c:90
#64 pymain_run_file_obj from /usr/local/src/conda/python-3.10.12/Modules/main.c:357
#65 Py_BytesMain from /usr/local/src/conda/python-3.10.12/Modules/main.c:1090
#66 __libc_start_call_main from ??:0
#67 <unwind unsupported> from ??:0
```
Pull Request resolved: #118924
Approved by: https://github.com/kwen2501
pytorch-bot bot
pushed a commit
that referenced
this pull request
Feb 8, 2024
user may not know which line of code called collectives in a big code base. When debugging, we can print python-cpp stacktrace in case user call ``ProcessGroup.reduce`` instead of ``torch.distributed.reduce``
```
LOG(INFO) << "ProcessGroupNCCL::_allgather_base stacktrace: "
<< get_python_cpp_trace();
```
output (using _allgather_base as an example): one example python-part trace is ``all_gather_into_tensor from /data/users/weif/pytorch/torch/distributed/distributed_c10d.py:2838``
```
ProcessGroupNCCL::_allgather_base stacktrace: #0 torch::unwind::unwind() from ??:0
#1 torch::CapturedTraceback::gather(bool, bool, bool) from ??:0
#2 c10d::get_python_cpp_trace[abi:cxx11]() from :0
#3 c10d::ProcessGroupNCCL::_allgather_base(at::Tensor&, at::Tensor&, c10d::AllgatherOptions const&) from ??:0
#4 c10d::ops::(anonymous namespace)::_allgather_base_CUDA(at::Tensor&, at::Tensor&, c10::intrusive_ptr<c10d::ProcessGroup, c10::detail::intrusive_target_default_null_type<c10d::ProcessGroup> > const&, bool, long) from Ops.cpp:0
#5 c10::impl::make_boxed_from_unboxed_functor<c10::impl::detail::WrapFunctionIntoRuntimeFunctor_<std::tuple<at::Tensor, c10::intrusive_ptr<c10d::Work, c10::detail::intrusive_target_default_null_type<c10d::Work> > > (*)(at::Tensor&, at::Tensor&, c10::intrusive_ptr<c10d::ProcessGroup, c10::detail::intrusive_target_default_null_type<c10d::ProcessGroup> > const&, bool, long), std::tuple<at::Tensor, c10::intrusive_ptr<c10d::Work, c10::detail::intrusive_target_default_null_type<c10d::Work> > >, c10::guts::typelist::typelist<at::Tensor&, at::Tensor&, c10::intrusive_ptr<c10d::ProcessGroup, c10::detail::intrusive_target_default_null_type<c10d::ProcessGroup> > const&, bool, long> >, false>::call(c10::OperatorKernel*, c10::OperatorHandle const&, c10::DispatchKeySet, std::vector<c10::IValue, std::allocator<c10::IValue> >*) from :0
#6 torch::autograd::basicAutogradNotImplementedFallbackImpl(c10::OperatorHandle const&, c10::DispatchKeySet, std::vector<c10::IValue, std::allocator<c10::IValue> >*) from autograd_not_implemented_fallback.cpp:0
#7 c10d::ProcessGroup::_allgather_base(at::Tensor&, at::Tensor&, c10d::AllgatherOptions const&) from :0
#8 pybind11::cpp_function::initialize<pybind11::cpp_function::initialize<c10::intrusive_ptr<c10d::Work, c10::detail::intrusive_target_default_null_type<c10d::Work> >, c10d::ProcessGroup, at::Tensor&, at::Tensor&, c10d::AllgatherOptions const&, pybind11::name, pybind11::is_method, pybind11::sibling, pybind11::arg, pybind11::arg, pybind11::arg_v, pybind11::call_guard<pybind11::gil_scoped_release> >(c10::intrusive_ptr<c10d::Work, c10::detail::intrusive_target_default_null_type<c10d::Work> > (c10d::ProcessGroup::*)(at::Tensor&, at::Tensor&, c10d::AllgatherOptions const&), pybind11::name const&, pybind11::is_method const&, pybind11::sibling const&, pybind11::arg const&, pybind11::arg const&, pybind11::arg_v const&, pybind11::call_guard<pybind11::gil_scoped_release> const&)::{lambda(c10d::ProcessGroup*, at::Tensor&, at::Tensor&, c10d::AllgatherOptions const&)#1}, c10::intrusive_ptr<c10d::Work, c10::detail::intrusive_target_default_null_type<c10d::Work> >, c10d::ProcessGroup*, at::Tensor&, at::Tensor&, c10d::AllgatherOptions const&, pybind11::name, pybind11::is_method, pybind11::sibling, pybind11::arg, pybind11::arg, pybind11::arg_v, pybind11::call_guard<pybind11::gil_scoped_release> >(pybind11::cpp_function::initialize<c10::intrusive_ptr<c10d::Work, c10::detail::intrusive_target_default_null_type<c10d::Work> >, c10d::ProcessGroup, at::Tensor&, at::Tensor&, c10d::AllgatherOptions const&, pybind11::name, pybind11::is_method, pybind11::sibling, pybind11::arg, pybind11::arg, pybind11::arg_v, pybind11::call_guard<pybind11::gil_scoped_release> >(c10::intrusive_ptr<c10d::Work, c10::detail::intrusive_target_default_null_type<c10d::Work> > (c10d::ProcessGroup::*)(at::Tensor&, at::Tensor&, c10d::AllgatherOptions const&), pybind11::name const&, pybind11::is_method const&, pybind11::sibling const&, pybind11::arg const&, pybind11::arg const&, pybind11::arg_v const&, pybind11::call_guard<pybind11::gil_scoped_release> const&)::{lambda(c10d::ProcessGroup*, at::Tensor&, at::Tensor&, c10d::AllgatherOptions const&)#1}&&, c10::intrusive_ptr<c10d::Work, c10::detail::intrusive_target_default_null_type<c10d::Work> > (*)(c10d::ProcessGroup*, at::Tensor&, at::Tensor&, c10d::AllgatherOptions const&), pybind11::name const&, pybind11::is_method const&, pybind11::sibling const&, pybind11::arg const&, pybind11::arg const&, pybind11::arg_v const&, pybind11::call_guard<pybind11::gil_scoped_release> const&)::{lambda(pybind11::detail::function_call&)#3}::_FUN(pybind11::detail::function_call&) from :0
#9 pybind11::cpp_function::dispatcher(_object*, _object*, _object*) from :0
#10 cfunction_call from /usr/local/src/conda/python-3.10.12/Objects/methodobject.c:543
#11 _PyObject_MakeTpCall from /usr/local/src/conda/python-3.10.12/Objects/call.c:215
#12 _PyObject_VectorcallTstate from /usr/local/src/conda/python-3.10.12/Include/cpython/abstract.h:112
#13 _PyObject_VectorcallTstate from /usr/local/src/conda/python-3.10.12/Include/cpython/abstract.h:114
#14 all_gather_into_tensor from /data/users/weif/pytorch/torch/distributed/distributed_c10d.py:2838
#15 _PyEval_EvalFrame from /usr/local/src/conda/python-3.10.12/Include/internal/pycore_ceval.h:46
#16 do_call_core from /usr/local/src/conda/python-3.10.12/Python/ceval.c:5945
#17 wrapper from /data/users/weif/pytorch/torch/distributed/c10d_logger.py:75
#18 _PyEval_EvalFrame from /usr/local/src/conda/python-3.10.12/Include/internal/pycore_ceval.h:46
#19 _PyObject_VectorcallTstate from /usr/local/src/conda/python-3.10.12/Include/cpython/abstract.h:114
#20 _all_gather_flat_param from /data/users/weif/pytorch/torch/distributed/fsdp/_flat_param.py:1399
#21 _PyEval_EvalFrame from /usr/local/src/conda/python-3.10.12/Include/internal/pycore_ceval.h:46
#22 _PyObject_VectorcallTstate from /usr/local/src/conda/python-3.10.12/Include/cpython/abstract.h:114
#23 unshard from /data/users/weif/pytorch/torch/distributed/fsdp/_flat_param.py:1308
#24 _PyEval_EvalFrame from /usr/local/src/conda/python-3.10.12/Include/internal/pycore_ceval.h:46
#25 _PyObject_VectorcallTstate from /usr/local/src/conda/python-3.10.12/Include/cpython/abstract.h:114
#26 _unshard from /data/users/weif/pytorch/torch/distributed/fsdp/_runtime_utils.py:332
#27 _PyEval_EvalFrame from /usr/local/src/conda/python-3.10.12/Include/internal/pycore_ceval.h:46
#28 _PyObject_VectorcallTstate from /usr/local/src/conda/python-3.10.12/Include/cpython/abstract.h:114
#29 _pre_forward_unshard from /data/users/weif/pytorch/torch/distributed/fsdp/_runtime_utils.py:448
#30 _PyEval_EvalFrame from /usr/local/src/conda/python-3.10.12/Include/internal/pycore_ceval.h:46
#31 _PyObject_VectorcallTstate from /usr/local/src/conda/python-3.10.12/Include/cpython/abstract.h:114
#32 _pre_forward from /data/users/weif/pytorch/torch/distributed/fsdp/_runtime_utils.py:413
#33 _PyEval_EvalFrame from /usr/local/src/conda/python-3.10.12/Include/internal/pycore_ceval.h:46
#34 _PyObject_VectorcallTstate from /usr/local/src/conda/python-3.10.12/Include/cpython/abstract.h:114
#35 forward from /data/users/weif/pytorch/torch/distributed/fsdp/fully_sharded_data_parallel.py:839
#36 _PyEval_EvalFrame from /usr/local/src/conda/python-3.10.12/Include/internal/pycore_ceval.h:46
#37 do_call_core from /usr/local/src/conda/python-3.10.12/Python/ceval.c:5945
#38 _call_impl from /data/users/weif/pytorch/torch/nn/modules/module.py:1520
#39 _PyEval_EvalFrame from /usr/local/src/conda/python-3.10.12/Include/internal/pycore_ceval.h:46
#40 do_call_core from /usr/local/src/conda/python-3.10.12/Python/ceval.c:5945
#41 _wrapped_call_impl from /data/users/weif/pytorch/torch/nn/modules/module.py:1511
#42 _PyEval_EvalFrame from /usr/local/src/conda/python-3.10.12/Include/internal/pycore_ceval.h:46
#43 _PyObject_Call_Prepend from /usr/local/src/conda/python-3.10.12/Objects/call.c:431
#44 slot_tp_call from /usr/local/src/conda/python-3.10.12/Objects/typeobject.c:7494
#45 _PyObject_MakeTpCall from /usr/local/src/conda/python-3.10.12/Objects/call.c:215
#46 _PyObject_VectorcallTstate from /usr/local/src/conda/python-3.10.12/Include/cpython/abstract.h:112
#47 inner from /data/users/weif/pytorch/run_fsdp.py:72
#48 _PyEval_EvalFrame from /usr/local/src/conda/python-3.10.12/Include/internal/pycore_ceval.h:46
#49 _PyObject_VectorcallTstate from /usr/local/src/conda/python-3.10.12/Include/cpython/abstract.h:114
#50 run from /data/users/weif/pytorch/run_fsdp.py:76
#51 _PyEval_EvalFrame from /usr/local/src/conda/python-3.10.12/Include/internal/pycore_ceval.h:46
#52 _PyObject_VectorcallTstate from /usr/local/src/conda/python-3.10.12/Include/cpython/abstract.h:114
#53 main from /data/users/weif/pytorch/run_fsdp.py:133
#54 _PyEval_EvalFrame from /usr/local/src/conda/python-3.10.12/Include/internal/pycore_ceval.h:46
#55 _PyObject_VectorcallTstate from /usr/local/src/conda/python-3.10.12/Include/cpython/abstract.h:114
#56 <module> from /data/users/weif/pytorch/run_fsdp.py:137
#57 _PyEval_EvalFrame from /usr/local/src/conda/python-3.10.12/Include/internal/pycore_ceval.h:46
#58 PyEval_EvalCode from /usr/local/src/conda/python-3.10.12/Python/ceval.c:1134
#59 run_eval_code_obj from /usr/local/src/conda/python-3.10.12/Python/pythonrun.c:1291
#60 run_mod from /usr/local/src/conda/python-3.10.12/Python/pythonrun.c:1312
#61 pyrun_file from /usr/local/src/conda/python-3.10.12/Python/pythonrun.c:1208
#62 _PyRun_SimpleFileObject from /usr/local/src/conda/python-3.10.12/Python/pythonrun.c:456
#63 _PyRun_AnyFileObject from /usr/local/src/conda/python-3.10.12/Python/pythonrun.c:90
#64 pymain_run_file_obj from /usr/local/src/conda/python-3.10.12/Modules/main.c:357
#65 Py_BytesMain from /usr/local/src/conda/python-3.10.12/Modules/main.c:1090
#66 __libc_start_call_main from ??:0
#67 <unwind unsupported> from ??:0
```
Pull Request resolved: #118924
Approved by: https://github.com/kwen2501
pytorchmergebot
pushed a commit
that referenced
this pull request
May 21, 2024
#126677) …destruction of tensors cached by autocast ## Root Cause For out-of-tree device extension it is loaded after torch (different .so), so the global variable `cached_casts` may be constructed before caching allocator and then destructed in reversed order when exit. ## Fix Lazily initialize `cached_casts` to correct the order. ## How to Reproduce && Test Modify the testcase `TestAutocastGPU.test_cast_cache_is_global` in test/test_autocast.py to run on your out-of-tree device. You will see following failure in the end of test. ```bash ---------------------------------------------------------------------- Ran 1 test in 4.812s OK free: 0x30080ff44000400 terminate called after throwing an instance of 'c10::Error' what(): invalid device pointer: 0x30080ff44000400 Exception raised from free at /projs/framework/betterman/code/pytorch_new/catch/torch_mlu/csrc/framework/core/caching_allocator.cpp:1609 (most recent call first): frame #0: <unknown function> + 0x118fe1 (0x7ffaef4d3fe1 in /projs/framework/betterman/code/pytorch_new/torch/lib/libc10.so) frame #1: <unknown function> + 0x11b1c4 (0x7ffaef4d61c4 in /projs/framework/betterman/code/pytorch_new/torch/lib/libc10.so) frame #2: <unknown function> + 0x117677 (0x7ffaef4d2677 in /projs/framework/betterman/code/pytorch_new/torch/lib/libc10.so) frame #3: <unknown function> + 0x11a2bf (0x7ffaef4d52bf in /projs/framework/betterman/code/pytorch_new/torch/lib/libc10.so) frame #4: <unknown function> + 0x11a186 (0x7ffaef4d5186 in /projs/framework/betterman/code/pytorch_new/torch/lib/libc10.so) frame #5: <unknown function> + 0x119fde (0x7ffaef4d4fde in /projs/framework/betterman/code/pytorch_new/torch/lib/libc10.so) frame #6: <unknown function> + 0x119d2e (0x7ffaef4d4d2e in /projs/framework/betterman/code/pytorch_new/torch/lib/libc10.so) frame #7: <unknown function> + 0x119be0 (0x7ffaef4d4be0 in /projs/framework/betterman/code/pytorch_new/torch/lib/libc10.so) frame #8: <unknown function> + 0x119977 (0x7ffaef4d4977 in /projs/framework/betterman/code/pytorch_new/torch/lib/libc10.so) frame #9: <unknown function> + 0x119313 (0x7ffaef4d4313 in /projs/framework/betterman/code/pytorch_new/torch/lib/libc10.so) frame #10: <unknown function> + 0x118b4c (0x7ffaef4d3b4c in /projs/framework/betterman/code/pytorch_new/torch/lib/libc10.so) frame #11: c10::Error::Error(c10::SourceLocation, std::string) + 0x34 (0x7ffaef4d27c4 in /projs/framework/betterman/code/pytorch_new/torch/lib/libc10.so) frame #12: c10::detail::torchCheckFail(char const*, char const*, unsigned int, std::string const&) + 0x7f (0x7ffaef4d04ed in /projs/framework/betterman/code/pytorch_new/torch/lib/libc10.so) frame #13: torch_mlu::MLUCachingAllocator::Native::NativeCachingAllocator::free(void*) + 0xe6 (0x7ff9a8eeb112 in /projs/framework/betterman/code/pytorch_new/catch/torch_mlu/csrc/lib/libtorch_mlu.so) frame #14: torch_mlu::MLUCachingAllocator::Native::local_raw_delete(void*) + 0x3b (0x7ff9a8ed9480 in /projs/framework/betterman/code/pytorch_new/catch/torch_mlu/csrc/lib/libtorch_mlu.so) frame #15: std::unique_ptr<void, void (*)(void*)>::~unique_ptr() + 0x50 (0x7ffb0a5ea322 in /projs/framework/betterman/code/pytorch_new/torch/lib/libtorch_python.so) frame #16: <unknown function> + 0x1269890 (0x7ffb0a5e4890 in /projs/framework/betterman/code/pytorch_new/torch/lib/libtorch_python.so) frame #17: <unknown function> + 0x1269928 (0x7ffb0a5e4928 in /projs/framework/betterman/code/pytorch_new/torch/lib/libtorch_python.so) frame #18: <unknown function> + 0x127572c (0x7ffb0a5f072c in /projs/framework/betterman/code/pytorch_new/torch/lib/libtorch_python.so) frame #19: <unknown function> + 0x1275758 (0x7ffb0a5f0758 in /projs/framework/betterman/code/pytorch_new/torch/lib/libtorch_python.so) frame #20: <unknown function> + 0xb9bc7 (0x7ffaef474bc7 in /projs/framework/betterman/code/pytorch_new/torch/lib/libc10.so) frame #21: <unknown function> + 0xb97bc (0x7ffaef4747bc in /projs/framework/betterman/code/pytorch_new/torch/lib/libc10.so) frame #22: <unknown function> + 0xdbc50 (0x7ffaef496c50 in /projs/framework/betterman/code/pytorch_new/torch/lib/libc10.so) frame #23: c10::TensorImpl::~TensorImpl() + 0x82 (0x7ffaef49157e in /projs/framework/betterman/code/pytorch_new/torch/lib/libc10.so) frame #24: c10::TensorImpl::~TensorImpl() + 0x1c (0x7ffaef4915aa in /projs/framework/betterman/code/pytorch_new/torch/lib/libc10.so) frame #25: <unknown function> + 0x2f596d9 (0x7ffaf24fc6d9 in /projs/framework/betterman/code/pytorch_new/torch/lib/libtorch_cpu.so) frame #26: <unknown function> + 0x2f589c2 (0x7ffaf24fb9c2 in /projs/framework/betterman/code/pytorch_new/torch/lib/libtorch_cpu.so) frame #27: <unknown function> + 0x2f57b92 (0x7ffaf24fab92 in /projs/framework/betterman/code/pytorch_new/torch/lib/libtorch_cpu.so) frame #28: <unknown function> + 0x2f5c228 (0x7ffaf24ff228 in /projs/framework/betterman/code/pytorch_new/torch/lib/libtorch_cpu.so) frame #29: <unknown function> + 0x30f3f70 (0x7ffaf2696f70 in /projs/framework/betterman/code/pytorch_new/torch/lib/libtorch_cpu.so) frame #30: <unknown function> + 0x30f3f90 (0x7ffaf2696f90 in /projs/framework/betterman/code/pytorch_new/torch/lib/libtorch_cpu.so) frame #31: <unknown function> + 0x30f5004 (0x7ffaf2698004 in /projs/framework/betterman/code/pytorch_new/torch/lib/libtorch_cpu.so) frame #32: <unknown function> + 0x30f5024 (0x7ffaf2698024 in /projs/framework/betterman/code/pytorch_new/torch/lib/libtorch_cpu.so) frame #33: <unknown function> + 0x31207f0 (0x7ffaf26c37f0 in /projs/framework/betterman/code/pytorch_new/torch/lib/libtorch_cpu.so) frame #34: <unknown function> + 0x3120814 (0x7ffaf26c3814 in /projs/framework/betterman/code/pytorch_new/torch/lib/libtorch_cpu.so) frame #35: <unknown function> + 0x30f51e8 (0x7ffaf26981e8 in /projs/framework/betterman/code/pytorch_new/torch/lib/libtorch_cpu.so) frame #36: <unknown function> + 0x30f5148 (0x7ffaf2698148 in /projs/framework/betterman/code/pytorch_new/torch/lib/libtorch_cpu.so) frame #37: <unknown function> + 0x316ecea (0x7ffaf2711cea in /projs/framework/betterman/code/pytorch_new/torch/lib/libtorch_cpu.so) frame #38: <unknown function> + 0x468a7 (0x7ffb0c9ed8a7 in /lib/x86_64-linux-gnu/libc.so.6) frame #39: on_exit + 0 (0x7ffb0c9eda60 in /lib/x86_64-linux-gnu/libc.so.6) <omitting python frames> frame #47: __libc_start_main + 0xf3 (0x7ffb0c9cb083 in /lib/x86_64-linux-gnu/libc.so.6) Aborted (core dumped) ``` Pull Request resolved: #126677 Approved by: https://github.com/ezyang
pytorchmergebot
pushed a commit
that referenced
this pull request
Jun 1, 2025
Which inherits from `RuntimeError` and contains `error_code`, which in case of CUDA should contain error returned by `cudaGetLastError` `torch::detail::_new_accelerator_error_object(c10::AcceleratorError&)` follows the pattern of CPython's [`PyErr_SetString`](https://github.com/python/cpython/blob/cb8a72b301f47e76d93a7fe5b259e9a5758792e1/Python/errors.c#L282), namely - Convert cstr into Python string with `PyUnicode_FromString` - Create new exception object using `PyObject_CallOneArg` just like it's done in [`_PyErr_CreateException`](https://github.com/python/cpython/blob/cb8a72b301f47e76d93a7fe5b259e9a5758792e1/Python/errors.c#L32) - Set `error_code` property using `PyObject_SetAttrString` - decref all temporary references Test that it works and captures CPP backtrace (in addition to CI) by running ```python import os os.environ['TORCH_SHOW_CPP_STACKTRACES'] = '1' import torch x = torch.rand(10, device="cuda") y = torch.arange(20, device="cuda") try: x[y] = 2 print(x) except torch.AcceleratorError as e: print("Exception was raised", e.args[0]) print("Captured error code is ", e.error_code) ``` which produces following output ``` Exception was raised CUDA error: device-side assert triggered CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. For debugging consider passing CUDA_LAUNCH_BLOCKING=1 Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions. Exception raised from c10_cuda_check_implementation at /home/ubuntu/pytorch/c10/cuda/CUDAException.cpp:41 (most recent call first): C++ CapturedTraceback: #4 std::_Function_handler<std::shared_ptr<c10::LazyValue<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > const> (), c10::SetStackTraceFetcher(std::function<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > ()>)::{lambda()#1}>::_M_invoke(std::_Any_data const&) from Logging.cpp:0 #5 c10::Error::Error(c10::SourceLocation, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >) from ??:0 #6 c10::cuda::c10_cuda_check_implementation(int, char const*, char const*, int, bool) [clone .cold] from CUDAException.cpp:0 #7 void at::native::gpu_kernel_impl<at::native::AbsFunctor<float> >(at::TensorIteratorBase&, at::native::AbsFunctor<float> const&) [clone .isra.0] from tmpxft_000191fc_00000000-6_AbsKernel.cudafe1.cpp:0 #8 at::native::abs_kernel_cuda(at::TensorIteratorBase&) from ??:0 #9 at::Tensor& at::native::unary_op_impl_with_complex_to_float_out<at::native::abs_stub_DECLARE_DISPATCH_type>(at::Tensor&, at::Tensor const&, at::native::abs_stub_DECLARE_DISPATCH_type&, bool) [clone .constprop.0] from UnaryOps.cpp:0 #10 at::(anonymous namespace)::(anonymous namespace)::wrapper_CUDA_out_abs_out(at::Tensor const&, at::Tensor&) from RegisterCUDA_0.cpp:0 #11 at::_ops::abs_out::call(at::Tensor const&, at::Tensor&) from ??:0 #12 at::native::abs(at::Tensor const&) from ??:0 #13 c10::impl::wrap_kernel_functor_unboxed_<c10::impl::detail::WrapFunctionIntoFunctor_<c10::CompileTimeFunctionPointer<at::Tensor (at::Tensor const&), &at::(anonymous namespace)::(anonymous namespace)::wrapper_CompositeExplicitAutograd__abs>, at::Tensor, c10::guts::typelist::typelist<at::Tensor const&> >, at::Tensor (at::Tensor const&)>::call(c10::OperatorKernel*, c10::DispatchKeySet, at::Tensor const&) from RegisterCompositeExplicitAutograd_0.cpp:0 #14 at::_ops::abs::redispatch(c10::DispatchKeySet, at::Tensor const&) from ??:0 #15 torch::autograd::VariableType::(anonymous namespace)::abs(c10::DispatchKeySet, at::Tensor const&) from VariableType_1.cpp:0 #16 c10::impl::wrap_kernel_functor_unboxed_<c10::impl::detail::WrapFunctionIntoFunctor_<c10::CompileTimeFunctionPointer<at::Tensor (c10::DispatchKeySet, at::Tensor const&), &torch::autograd::VariableType::(anonymous namespace)::abs>, at::Tensor, c10::guts::typelist::typelist<c10::DispatchKeySet, at::Tensor const&> >, at::Tensor (c10::DispatchKeySet, at::Tensor const&)>::call(c10::OperatorKernel*, c10::DispatchKeySet, at::Tensor const&) from VariableType_1.cpp:0 #17 at::_ops::abs::call(at::Tensor const&) from ??:0 #18 at::native::isfinite(at::Tensor const&) from ??:0 #19 c10::impl::wrap_kernel_functor_unboxed_<c10::impl::detail::WrapFunctionIntoFunctor_<c10::CompileTimeFunctionPointer<at::Tensor (at::Tensor const&), &at::(anonymous namespace)::(anonymous namespace)::wrapper_CompositeImplicitAutograd__isfinite>, at::Tensor, c10::guts::typelist::typelist<at::Tensor const&> >, at::Tensor (at::Tensor const&)>::call(c10::OperatorKernel*, c10::DispatchKeySet, at::Tensor const&) from RegisterCompositeImplicitAutograd_0.cpp:0 #20 at::_ops::isfinite::call(at::Tensor const&) from ??:0 #21 torch::autograd::THPVariable_isfinite(_object*, _object*, _object*) from python_torch_functions_2.cpp:0 #22 PyObject_CallFunctionObjArgs from ??:0 #23 _PyObject_MakeTpCall from ??:0 #24 _PyEval_EvalFrameDefault from ??:0 #25 _PyObject_FastCallDictTstate from ??:0 #26 _PyStack_AsDict from ??:0 #27 _PyObject_MakeTpCall from ??:0 #28 _PyEval_EvalFrameDefault from ??:0 #29 _PyFunction_Vectorcall from ??:0 #30 _PyEval_EvalFrameDefault from ??:0 #31 _PyFunction_Vectorcall from ??:0 #32 _PyEval_EvalFrameDefault from ??:0 #33 _PyFunction_Vectorcall from ??:0 #34 _PyEval_EvalFrameDefault from ??:0 #35 PyFrame_GetCode from ??:0 #36 PyNumber_Xor from ??:0 #37 PyObject_Str from ??:0 #38 PyFile_WriteObject from ??:0 #39 _PyWideStringList_AsList from ??:0 #40 _PyDict_NewPresized from ??:0 #41 _PyEval_EvalFrameDefault from ??:0 #42 PyEval_EvalCode from ??:0 #43 PyEval_EvalCode from ??:0 #44 PyUnicode_Tailmatch from ??:0 #45 PyInit__collections from ??:0 #46 PyUnicode_Tailmatch from ??:0 #47 _PyRun_SimpleFileObject from ??:0 #48 _PyRun_AnyFileObject from ??:0 #49 Py_RunMain from ??:0 #50 Py_BytesMain from ??:0 #51 __libc_init_first from ??:0 #52 __libc_start_main from ??:0 #53 _start from ??:0 Captured error code is 710 ``` Pull Request resolved: #152023 Approved by: https://github.com/eqy, https://github.com/mradmila, https://github.com/ngimel ghstack dependencies: #154436
ethanwee1
added a commit
to ethanwee1/pytorch
that referenced
this pull request
Jun 26, 2025
…:warp_size() (pytorch#2293) Fixes SWDEV-540240, SWDEV-540309, SWDEV-539989 ### Error ``` pytorch#24 437.7 what(): HIP error: no ROCm-capable device is detected pytorch#24 437.7 HIP kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. pytorch#24 437.7 For debugging consider passing AMD_SERIALIZE_KERNEL=3 pytorch#24 437.7 Device-side assertions were explicitly omitted for this error check; the error probably arose while initializing the DSA handlers. pytorch#24 437.7 Exception raised from c10_hip_check_implementation at /pytorch/c10/hip/HIPException.cpp:44 (most recent call first): pytorch#24 437.7 frame #0: c10::Error::Error(c10::SourceLocation, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >) + 0x88 (0x7f272de18738 in /usr/local/lib/python3.10/dist-packages/torch/lib/libc10.so) pytorch#24 437.7 frame #1: c10::detail::torchCheckFail(char const*, char const*, unsigned int, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) + 0x55 (0x7f272ddb42ed in /usr/local/lib/python3.10/dist-packages/torch/lib/libc10.so) ... pytorch#24 437.7 frame pytorch#7: at::cuda::getCurrentDeviceProperties() + 0x9 (0x7f270b5874e9 in /usr/local/lib/python3.10/dist-packages/torch/lib/libtorch_hip.so) pytorch#24 437.7 frame pytorch#8: at::cuda::warp_size() + 0x9 (0x7f270b587509 in /usr/local/lib/python3.10/dist-packages/torch/lib/libtorch_hip.so) pytorch#24 437.7 frame pytorch#9: <unknown function> + 0x81ac8b (0x7f2709c27c8b in /usr/local/lib/python3.10/dist-packages/torch/lib/libtorch_hip.so) ``` ### Explanation ROCm@80cca70 created a static global variable that used `at::cuda::warp_size()` to initialize its value, which needs GPUs to be visible to query device properties. However, GPUs are not present on CPU-only build systems. ### Solution Convert static variable into a static function, thus preventing static initialization. ### Validation http://rocm-ci.amd.com/job/pyt_whl_docker_mainline/1461/artifact/build_artifacts.txt/*view*/ Ran microbenchmark to confirm basic functionality: ``` root@ubb4-rack-22:/var/lib/jenkins/pytorch-micro-benchmarking# python3 micro_benchmarking_pytorch.py --network resnet50 INFO: running forward and backward for warmup. INFO: running the benchmark.. OK: finished running benchmark.. --------------------SUMMARY-------------------------- Microbenchmark for network : resnet50 Num devices: 1 Dtype: FP32 Mini batch size [img] : 64 Time per mini-batch : 0.10158218145370483 Throughput [img/sec] : 630.0317544289736= ```
jagadish-amd
pushed a commit
to jagadish-amd/pytorch
that referenced
this pull request
Jul 9, 2025
…:warp_size() (pytorch#2293) Fixes SWDEV-540240, SWDEV-540309, SWDEV-539989 ### Error ``` pytorch#24 437.7 what(): HIP error: no ROCm-capable device is detected pytorch#24 437.7 HIP kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. pytorch#24 437.7 For debugging consider passing AMD_SERIALIZE_KERNEL=3 pytorch#24 437.7 Device-side assertions were explicitly omitted for this error check; the error probably arose while initializing the DSA handlers. pytorch#24 437.7 Exception raised from c10_hip_check_implementation at /pytorch/c10/hip/HIPException.cpp:44 (most recent call first): pytorch#24 437.7 frame #0: c10::Error::Error(c10::SourceLocation, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >) + 0x88 (0x7f272de18738 in /usr/local/lib/python3.10/dist-packages/torch/lib/libc10.so) pytorch#24 437.7 frame pytorch#1: c10::detail::torchCheckFail(char const*, char const*, unsigned int, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) + 0x55 (0x7f272ddb42ed in /usr/local/lib/python3.10/dist-packages/torch/lib/libc10.so) ... pytorch#24 437.7 frame pytorch#7: at::cuda::getCurrentDeviceProperties() + 0x9 (0x7f270b5874e9 in /usr/local/lib/python3.10/dist-packages/torch/lib/libtorch_hip.so) pytorch#24 437.7 frame pytorch#8: at::cuda::warp_size() + 0x9 (0x7f270b587509 in /usr/local/lib/python3.10/dist-packages/torch/lib/libtorch_hip.so) pytorch#24 437.7 frame pytorch#9: <unknown function> + 0x81ac8b (0x7f2709c27c8b in /usr/local/lib/python3.10/dist-packages/torch/lib/libtorch_hip.so) ``` ### Explanation ROCm@80cca70 created a static global variable that used `at::cuda::warp_size()` to initialize its value, which needs GPUs to be visible to query device properties. However, GPUs are not present on CPU-only build systems. ### Solution Convert static variable into a static function, thus preventing static initialization. ### Validation http://rocm-ci.amd.com/job/pyt_whl_docker_mainline/1461/artifact/build_artifacts.txt/*view*/ Ran microbenchmark to confirm basic functionality: ``` root@ubb4-rack-22:/var/lib/jenkins/pytorch-micro-benchmarking# python3 micro_benchmarking_pytorch.py --network resnet50 INFO: running forward and backward for warmup. INFO: running the benchmark.. OK: finished running benchmark.. --------------------SUMMARY-------------------------- Microbenchmark for network : resnet50 Num devices: 1 Dtype: FP32 Mini batch size [img] : 64 Time per mini-batch : 0.10158218145370483 Throughput [img/sec] : 630.0317544289736= ```
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Do not merge. This PR is just for review and test purposes.