Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

build failures on wsl2-ubuntu22.04 #19875

Open
sgjzfzzf opened this issue Feb 2, 2025 · 4 comments
Open

build failures on wsl2-ubuntu22.04 #19875

sgjzfzzf opened this issue Feb 2, 2025 · 4 comments
Labels
bug 🐞 Something isn't working

Comments

@sgjzfzzf
Copy link

sgjzfzzf commented Feb 2, 2025

What happened?

Hi, I'm building IREE from the source, and it fails. The backtrace looks like

[0/2] Re-checking globbed directories...
[5/5] Generating check_cuda_ukernel_ukernel_example.mlir_module.vmfb from ukernel_example.mlir
FAILED: samples/custom_dispatch/cuda/kernels/check_cuda_ukernel_ukernel_example.mlir_module.vmfb /home/jinjieliu/code/cpp/iree-fluidml/build/samples/custom_dispatch/cuda/kernels/check_cuda_ukernel_ukernel_example.mlir_module.vmfb 
cd /home/jinjieliu/code/cpp/iree-fluidml/build/samples/custom_dispatch/cuda/kernels && /home/jinjieliu/code/cpp/iree-fluidml/build/tools/iree-compile --output-format=vm-bytecode --mlir-print-op-on-diagnostic=false --iree-hal-target-backends=cuda --iree-link-bitcode=cuda_ukernel.bc /home/jinjieliu/code/cpp/iree-fluidml/samples/custom_dispatch/cuda/kernels/ukernel_example.mlir -o check_cuda_ukernel_ukernel_example.mlir_module.vmfb --iree-hal-executable-object-search-path=\"/home/jinjieliu/code/cpp/iree-fluidml/build\"
Please report issues to https://github.com/iree-org/iree/issues and include the crash backtrace.
Stack dump:
0.      Program arguments: /home/jinjieliu/code/cpp/iree-fluidml/build/tools/iree-compile --output-format=vm-bytecode --mlir-print-op-on-diagnostic=false --iree-hal-target-backends=cuda --iree-link-bitcode=cuda_ukernel.bc /home/jinjieliu/code/cpp/iree-fluidml/samples/custom_dispatch/cuda/kernels/ukernel_example.mlir -o check_cuda_ukernel_ukernel_example.mlir_module.vmfb --iree-hal-executable-object-search-path=\"/home/jinjieliu/code/cpp/iree-fluidml/build\"
 #0 0x00007f2fdc849e6f llvm::sys::PrintStackTrace(llvm::raw_ostream&, int) /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/llvm/lib/Support/Unix/Signals.inc:727:3
 #1 0x00007f2fdc847da4 llvm::sys::RunSignalHandlers() /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/llvm/lib/Support/Signals.cpp:105:20
 #2 0x00007f2fdc8480f6 SignalHandler(int) /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/llvm/lib/Support/Unix/Signals.inc:413:1
 #3 0x00007f2fda6f2520 (/lib/x86_64-linux-gnu/libc.so.6+0x42520)
 #4 0x00007f2fdf62843f mlir::iree_compiler::PartitionableLoopsInterface::getPartitionableLoops(std::optional<unsigned int>) /home/jinjieliu/code/cpp/iree-fluidml/build/compiler/src/iree/compiler/Codegen/Interfaces/PartitionableLoopsInterface.cpp.inc:14:96
 #5 0x00007f2fde5cf53a mlir::iree_compiler::setRootDefaultConfig(mlir::iree_compiler::IREE::GPU::TargetAttr, mlir::FunctionOpInterface, mlir::Operation*) /home/jinjieliu/code/cpp/iree-fluidml/compiler/src/iree/compiler/Codegen/LLVMGPU/KernelConfig.cpp:1664:3
 #6 0x00007f2fde5d5198 operator()<mlir::Operation*> /home/jinjieliu/code/cpp/iree-fluidml/compiler/src/iree/compiler/Codegen/LLVMGPU/KernelConfig.cpp:2437:36
 #7 0x00007f2fde5d5198 Default<mlir::iree_compiler::setRootConfig(mlir::iree_compiler::IREE::GPU::TargetAttr, mlir::FunctionOpInterface, mlir::Operation*)::<lambda(auto:52)> > /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/llvm/include/llvm/ADT/TypeSwitch.h:111:21
 #8 0x00007f2fde5d5198 mlir::iree_compiler::setRootConfig(mlir::iree_compiler::IREE::GPU::TargetAttr, mlir::FunctionOpInterface, mlir::Operation*) /home/jinjieliu/code/cpp/iree-fluidml/compiler/src/iree/compiler/Codegen/LLVMGPU/KernelConfig.cpp:2428:15
 #9 0x00007f2fde5d7a2d mlir::iree_compiler::initGPULaunchConfig(mlir::FunctionOpInterface) /home/jinjieliu/code/cpp/iree-fluidml/compiler/src/iree/compiler/Codegen/LLVMGPU/KernelConfig.cpp:2555:3
#10 0x00007f2fde5b34fd mlir::iree_compiler::(anonymous namespace)::LLVMGPUSelectLoweringStrategyPass::runOnOperation() /home/jinjieliu/code/cpp/iree-fluidml/compiler/src/iree/compiler/Codegen/LLVMGPU/LLVMGPUSelectLoweringStrategy.cpp:74:5
#11 0x00007f2fdcac6bea operator() /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/lib/Pass/Pass.cpp:527:22
#12 0x00007f2fdcac6bea callback_fn<mlir::detail::OpToOpPassAdaptor::run(mlir::Pass*, mlir::Operation*, mlir::AnalysisManager, bool, unsigned int)::<lambda()> > /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/llvm/include/llvm/ADT/STLFunctionalExtras.h:46:52
#13 0x00007f2fdcac6bea llvm::function_ref<void ()>::operator()() const /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/llvm/include/llvm/ADT/STLFunctionalExtras.h:69:12
#14 0x00007f2fdcac6bea void mlir::MLIRContext::executeAction<mlir::PassExecutionAction, mlir::Pass&>(llvm::function_ref<void ()>, llvm::ArrayRef<mlir::IRUnit>, mlir::Pass&) /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/include/mlir/IR/MLIRContext.h:280:15
#15 0x00007f2fdcac6bea mlir::detail::OpToOpPassAdaptor::run(mlir::Pass*, mlir::Operation*, mlir::AnalysisManager, bool, unsigned int) /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/lib/Pass/Pass.cpp:520:55
#16 0x00007f2fdcac7629 mlir::detail::OpToOpPassAdaptor::runPipeline(mlir::OpPassManager&, mlir::Operation*, mlir::AnalysisManager, bool, unsigned int, mlir::PassInstrumentor*, mlir::PassInstrumentation::PipelineParentInfo const*) /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/lib/Pass/Pass.cpp:592:5
#17 0x00007f2fdcac7a24 mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::'lambda'(mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo&)::operator()(mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo&) const /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/lib/Pass/Pass.cpp:816:5
#18 0x00007f2fdcac642d operator++ /usr/include/c++/11/bits/stl_iterator.h:1054:2
#19 0x00007f2fdcac642d failableParallelForEach<__gnu_cxx::__normal_iterator<mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo*, std::vector<mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo> >, mlir::parallelForEach<__gnu_cxx::__normal_iterator<mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo*, std::vector<mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo> >, mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::<lambda(mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo&)> >(mlir::MLIRContext*, __gnu_cxx::__normal_iterator<mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo*, std::vector<mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo> >, __gnu_cxx::__normal_iterator<mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo*, std::vector<mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo> >, mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::<lambda(mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo&)>&&)::<lambda(auto:15&&)> > /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/include/mlir/IR/Threading.h:45:26
#20 0x00007f2fdcac642d parallelForEach<__gnu_cxx::__normal_iterator<mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo*, std::vector<mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo> >, mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::<lambda(mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo&)> > /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/include/mlir/IR/Threading.h:119:32
#21 0x00007f2fdcac642d parallelForEach<std::vector<mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo>&, mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::<lambda(mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo&)> > /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/include/mlir/IR/Threading.h:131:18
#22 0x00007f2fdcac642d mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool) /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/lib/Pass/Pass.cpp:803:18
#23 0x00007f2fdcac6f65 mlir::detail::OpToOpPassAdaptor::runOnOperation(bool) /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/lib/Pass/Pass.cpp:713:28
#24 0x00007f2fdcac6f65 operator() /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/lib/Pass/Pass.cpp:524:34
#25 0x00007f2fdcac6f65 callback_fn<mlir::detail::OpToOpPassAdaptor::run(mlir::Pass*, mlir::Operation*, mlir::AnalysisManager, bool, unsigned int)::<lambda()> > /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/llvm/include/llvm/ADT/STLFunctionalExtras.h:46:52
#26 0x00007f2fdcac6f65 llvm::function_ref<void ()>::operator()() const /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/llvm/include/llvm/ADT/STLFunctionalExtras.h:69:12
#27 0x00007f2fdcac6f65 void mlir::MLIRContext::executeAction<mlir::PassExecutionAction, mlir::Pass&>(llvm::function_ref<void ()>, llvm::ArrayRef<mlir::IRUnit>, mlir::Pass&) /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/include/mlir/IR/MLIRContext.h:280:15
#28 0x00007f2fdcac6f65 mlir::detail::OpToOpPassAdaptor::run(mlir::Pass*, mlir::Operation*, mlir::AnalysisManager, bool, unsigned int) /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/lib/Pass/Pass.cpp:520:55
#29 0x00007f2fdcac7629 mlir::detail::OpToOpPassAdaptor::runPipeline(mlir::OpPassManager&, mlir::Operation*, mlir::AnalysisManager, bool, unsigned int, mlir::PassInstrumentor*, mlir::PassInstrumentation::PipelineParentInfo const*) /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/lib/Pass/Pass.cpp:592:5
#30 0x00007f2fdcac82bb mlir::detail::OpToOpPassAdaptor::run(mlir::Pass*, mlir::Operation*, mlir::AnalysisManager, bool, unsigned int)::'lambda'(mlir::OpPassManager&, mlir::Operation*)::operator()(mlir::OpPassManager&, mlir::Operation*) const /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/lib/Pass/Pass.cpp:509:42
#31 0x00007f2fde5da7bd mlir::iree_compiler::IREE::HAL::(anonymous namespace)::ConfigureTargetExecutableVariantsPass::runOnOperation() /home/jinjieliu/code/cpp/iree-fluidml/compiler/src/iree/compiler/Dialect/HAL/Transforms/ConfigureExecutables.cpp:71:5
#32 0x00007f2fdcac6bea operator() /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/lib/Pass/Pass.cpp:527:22
#33 0x00007f2fdcac6bea callback_fn<mlir::detail::OpToOpPassAdaptor::run(mlir::Pass*, mlir::Operation*, mlir::AnalysisManager, bool, unsigned int)::<lambda()> > /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/llvm/include/llvm/ADT/STLFunctionalExtras.h:46:52
#34 0x00007f2fdcac6bea llvm::function_ref<void ()>::operator()() const /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/llvm/include/llvm/ADT/STLFunctionalExtras.h:69:12
#35 0x00007f2fdcac6bea void mlir::MLIRContext::executeAction<mlir::PassExecutionAction, mlir::Pass&>(llvm::function_ref<void ()>, llvm::ArrayRef<mlir::IRUnit>, mlir::Pass&) /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/include/mlir/IR/MLIRContext.h:280:15
#36 0x00007f2fdcac6bea mlir::detail::OpToOpPassAdaptor::run(mlir::Pass*, mlir::Operation*, mlir::AnalysisManager, bool, unsigned int) /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/lib/Pass/Pass.cpp:520:55
#37 0x00007f2fdcac7629 mlir::detail::OpToOpPassAdaptor::runPipeline(mlir::OpPassManager&, mlir::Operation*, mlir::AnalysisManager, bool, unsigned int, mlir::PassInstrumentor*, mlir::PassInstrumentation::PipelineParentInfo const*) /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/lib/Pass/Pass.cpp:592:5
#38 0x00007f2fdcac7a24 mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::'lambda'(mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo&)::operator()(mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo&) const /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/lib/Pass/Pass.cpp:816:5
#39 0x00007f2fdcac642d operator++ /usr/include/c++/11/bits/stl_iterator.h:1054:2
#40 0x00007f2fdcac642d failableParallelForEach<__gnu_cxx::__normal_iterator<mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo*, std::vector<mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo> >, mlir::parallelForEach<__gnu_cxx::__normal_iterator<mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo*, std::vector<mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo> >, mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::<lambda(mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo&)> >(mlir::MLIRContext*, __gnu_cxx::__normal_iterator<mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo*, std::vector<mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo> >, __gnu_cxx::__normal_iterator<mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo*, std::vector<mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo> >, mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::<lambda(mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo&)>&&)::<lambda(auto:15&&)> > /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/include/mlir/IR/Threading.h:45:26
#41 0x00007f2fdcac642d parallelForEach<__gnu_cxx::__normal_iterator<mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo*, std::vector<mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo> >, mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::<lambda(mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo&)> > /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/include/mlir/IR/Threading.h:119:32
#42 0x00007f2fdcac642d parallelForEach<std::vector<mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo>&, mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::<lambda(mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo&)> > /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/include/mlir/IR/Threading.h:131:18
#43 0x00007f2fdcac642d mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool) /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/lib/Pass/Pass.cpp:803:18
#44 0x00007f2fdcac6f65 mlir::detail::OpToOpPassAdaptor::runOnOperation(bool) /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/lib/Pass/Pass.cpp:713:28
#45 0x00007f2fdcac6f65 operator() /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/lib/Pass/Pass.cpp:524:34
#46 0x00007f2fdcac6f65 callback_fn<mlir::detail::OpToOpPassAdaptor::run(mlir::Pass*, mlir::Operation*, mlir::AnalysisManager, bool, unsigned int)::<lambda()> > /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/llvm/include/llvm/ADT/STLFunctionalExtras.h:46:52
#47 0x00007f2fdcac6f65 llvm::function_ref<void ()>::operator()() const /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/llvm/include/llvm/ADT/STLFunctionalExtras.h:69:12
#48 0x00007f2fdcac6f65 void mlir::MLIRContext::executeAction<mlir::PassExecutionAction, mlir::Pass&>(llvm::function_ref<void ()>, llvm::ArrayRef<mlir::IRUnit>, mlir::Pass&) /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/include/mlir/IR/MLIRContext.h:280:15
#49 0x00007f2fdcac6f65 mlir::detail::OpToOpPassAdaptor::run(mlir::Pass*, mlir::Operation*, mlir::AnalysisManager, bool, unsigned int) /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/lib/Pass/Pass.cpp:520:55
#50 0x00007f2fdcac7629 mlir::detail::OpToOpPassAdaptor::runPipeline(mlir::OpPassManager&, mlir::Operation*, mlir::AnalysisManager, bool, unsigned int, mlir::PassInstrumentor*, mlir::PassInstrumentation::PipelineParentInfo const*) /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/lib/Pass/Pass.cpp:592:5
#51 0x00007f2fdcac82bb mlir::detail::OpToOpPassAdaptor::run(mlir::Pass*, mlir::Operation*, mlir::AnalysisManager, bool, unsigned int)::'lambda'(mlir::OpPassManager&, mlir::Operation*)::operator()(mlir::OpPassManager&, mlir::Operation*) const /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/lib/Pass/Pass.cpp:509:42
#52 0x00007f2fde5dbacf mlir::iree_compiler::IREE::HAL::(anonymous namespace)::ConfigureExecutablesPass::runOnOperation() /home/jinjieliu/code/cpp/iree-fluidml/compiler/src/iree/compiler/Dialect/HAL/Transforms/ConfigureExecutables.cpp:110:5
#53 0x00007f2fdcac6bea operator() /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/lib/Pass/Pass.cpp:527:22
#54 0x00007f2fdcac6bea callback_fn<mlir::detail::OpToOpPassAdaptor::run(mlir::Pass*, mlir::Operation*, mlir::AnalysisManager, bool, unsigned int)::<lambda()> > /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/llvm/include/llvm/ADT/STLFunctionalExtras.h:46:52
#55 0x00007f2fdcac6bea llvm::function_ref<void ()>::operator()() const /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/llvm/include/llvm/ADT/STLFunctionalExtras.h:69:12
#56 0x00007f2fdcac6bea void mlir::MLIRContext::executeAction<mlir::PassExecutionAction, mlir::Pass&>(llvm::function_ref<void ()>, llvm::ArrayRef<mlir::IRUnit>, mlir::Pass&) /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/include/mlir/IR/MLIRContext.h:280:15
#57 0x00007f2fdcac6bea mlir::detail::OpToOpPassAdaptor::run(mlir::Pass*, mlir::Operation*, mlir::AnalysisManager, bool, unsigned int) /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/lib/Pass/Pass.cpp:520:55
#58 0x00007f2fdcac7629 mlir::detail::OpToOpPassAdaptor::runPipeline(mlir::OpPassManager&, mlir::Operation*, mlir::AnalysisManager, bool, unsigned int, mlir::PassInstrumentor*, mlir::PassInstrumentation::PipelineParentInfo const*) /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/lib/Pass/Pass.cpp:592:5
#59 0x00007f2fdcac7a24 mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::'lambda'(mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo&)::operator()(mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo&) const /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/lib/Pass/Pass.cpp:816:5
#60 0x00007f2fdcac642d operator++ /usr/include/c++/11/bits/stl_iterator.h:1054:2
#61 0x00007f2fdcac642d failableParallelForEach<__gnu_cxx::__normal_iterator<mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo*, std::vector<mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo> >, mlir::parallelForEach<__gnu_cxx::__normal_iterator<mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo*, std::vector<mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo> >, mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::<lambda(mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo&)> >(mlir::MLIRContext*, __gnu_cxx::__normal_iterator<mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo*, std::vector<mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo> >, __gnu_cxx::__normal_iterator<mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo*, std::vector<mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo> >, mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::<lambda(mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo&)>&&)::<lambda(auto:15&&)> > /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/include/mlir/IR/Threading.h:45:26
#62 0x00007f2fdcac642d parallelForEach<__gnu_cxx::__normal_iterator<mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo*, std::vector<mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo> >, mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::<lambda(mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo&)> > /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/include/mlir/IR/Threading.h:119:32
#63 0x00007f2fdcac642d parallelForEach<std::vector<mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo>&, mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::<lambda(mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool)::OpPMInfo&)> > /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/include/mlir/IR/Threading.h:131:18
#64 0x00007f2fdcac642d mlir::detail::OpToOpPassAdaptor::runOnOperationAsyncImpl(bool) /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/lib/Pass/Pass.cpp:803:18
#65 0x00007f2fdcac6f65 mlir::detail::OpToOpPassAdaptor::runOnOperation(bool) /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/lib/Pass/Pass.cpp:713:28
#66 0x00007f2fdcac6f65 operator() /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/lib/Pass/Pass.cpp:524:34
#67 0x00007f2fdcac6f65 callback_fn<mlir::detail::OpToOpPassAdaptor::run(mlir::Pass*, mlir::Operation*, mlir::AnalysisManager, bool, unsigned int)::<lambda()> > /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/llvm/include/llvm/ADT/STLFunctionalExtras.h:46:52
#68 0x00007f2fdcac6f65 llvm::function_ref<void ()>::operator()() const /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/llvm/include/llvm/ADT/STLFunctionalExtras.h:69:12
#69 0x00007f2fdcac6f65 void mlir::MLIRContext::executeAction<mlir::PassExecutionAction, mlir::Pass&>(llvm::function_ref<void ()>, llvm::ArrayRef<mlir::IRUnit>, mlir::Pass&) /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/include/mlir/IR/MLIRContext.h:280:15
#70 0x00007f2fdcac6f65 mlir::detail::OpToOpPassAdaptor::run(mlir::Pass*, mlir::Operation*, mlir::AnalysisManager, bool, unsigned int) /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/lib/Pass/Pass.cpp:520:55
#71 0x00007f2fdcac7629 mlir::detail::OpToOpPassAdaptor::runPipeline(mlir::OpPassManager&, mlir::Operation*, mlir::AnalysisManager, bool, unsigned int, mlir::PassInstrumentor*, mlir::PassInstrumentation::PipelineParentInfo const*) /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/lib/Pass/Pass.cpp:592:5
#72 0x00007f2fdcac89ce mlir::PassManager::runPasses(mlir::Operation*, mlir::AnalysisManager) /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/lib/Pass/Pass.cpp:905:40
#73 0x00007f2fdcac89ce mlir::PassManager::run(mlir::Operation*) /home/jinjieliu/code/cpp/iree-fluidml/third_party/llvm-project/mlir/lib/Pass/Pass.cpp:885:69
#74 0x00007f2fdc7bd8e3 mlir::iree_compiler::embed::(anonymous namespace)::Invocation::runPipeline(iree_compiler_pipeline_t) /home/jinjieliu/code/cpp/iree-fluidml/compiler/src/iree/compiler/API/Internal/CompilerDriver.cpp:1008:0
#75 0x00007f2fdc9e9cba mlir::iree_compiler::runIreecMain(int, char**)::'lambda3'(iree_compiler_source_t*)::operator()(iree_compiler_source_t*) const /home/jinjieliu/code/cpp/iree-fluidml/compiler/src/iree/compiler/Tools/iree_compile_lib.cc:254:7
#76 0x00007f2fdc9ec9db mlir::iree_compiler::runIreecMain(int, char**) /home/jinjieliu/code/cpp/iree-fluidml/compiler/src/iree/compiler/Tools/iree_compile_lib.cc:355:5
#77 0x00007f2fda6d9d90 __libc_start_call_main ./csu/../sysdeps/nptl/libc_start_call_main.h:58:16
#78 0x00007f2fda6d9e40 call_init ./csu/../csu/libc-start.c:128:20
#79 0x00007f2fda6d9e40 __libc_start_main ./csu/../csu/libc-start.c:379:5
#80 0x000055fdfc7a6095 _start (/home/jinjieliu/code/cpp/iree-fluidml/build/tools/iree-compile+0x1095)
Segmentation fault
ninja: build stopped: subcommand failed.

Steps to reproduce your issue

I'm running on wsl2-ubuntu22.04 and building it with gcc/g++. Their version follows

gcc --version
gcc (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0
Copyright (C) 2021 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

g++ --version
g++ (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0
Copyright (C) 2021 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

The tag I'm using is v3.1.0, and I use the command to start the building.

cmake -G Ninja -B build -S . -DCMAKE_BUILD_TYPE=RelWithDebInfo -DIREE_TARGET_BACKEND_CUDA=ON -DIREE_HAL_DRIVER_CUDA=ON
cmake --build build

Then, it would generate the above error when the building almost reaches its end.

What component(s) does this issue relate to?

No response

Version information

Please refer to the above block.

Additional context

I tried to use gdb to get some information and found that a member pointer conceptImpl of mlir::iree_compiler::PartitionableLoopsInterface is set as NULL, and then it calls the getImpl() method to access this pointer, invokes a method getPartitionableLoops on it, and then exits abnormally. I guess it could be a NULL pointer reference.

@sgjzfzzf sgjzfzzf added the bug 🐞 Something isn't working label Feb 2, 2025
@ziereis
Copy link
Contributor

ziereis commented Feb 2, 2025

As far as i can tell its the same error has here: #19662. the easiest fix you can apply for yourself is to remove the cmake rule for building the custom dispatch for cuda.

@sgjzfzzf
Copy link
Author

sgjzfzzf commented Feb 2, 2025

Thank you for your reply! I need to use CUDA in my IREE. Is there any other way to enable CUDA?

Also, the same failure disappeared when I built the main branch. Will this information help?

@benvanik
Copy link
Collaborator

benvanik commented Feb 2, 2025

I thought we had a bot building CUDA stuff (though nothing running it) so I'm not sure how this isn't catching - maybe we don't have the bot. It should be an easy fix and is unrelated to platform/targets/etc.

@ScottTodd
Copy link
Member

We build what we can on each platform, and #19883 will increase that coverage a bit further, but we don't currently have bots running CUDA tests: #18814.

Not sure about WSL, but that sample also has issues on standard Windows: #17933

ScottTodd added a commit that referenced this issue Feb 10, 2025
These backends default to disabled, so enable them on this CI config.
The backends are already tested in other jobs too.

I hoped this might help spot issues like
#19875, but it doesn't seem that
way.

| | Before | After |
-- | -- | --
Logs | [logs
here](https://github.com/iree-org/iree/actions/runs/13118302608/job/36597989009)
| [logs
here](https://github.com/iree-org/iree/actions/runs/13118149869/job/36597451528?pr=19883)
Number of build targets | 8442 | 8715
Number of `iree-test-deps` targets | 1067 | 1336
Number of tests | 1538 | 1552


ci-exactly: linux_x64_clang
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug 🐞 Something isn't working
Projects
None yet
Development

No branches or pull requests

4 participants