AuroraPerego
commented
8 months ago I see the same behavior when compiling our project as well, even when compiling for AMD only.
LLVM ERROR: SmallVector unable to grow. Requested capacity (94570172698048) is larger than maximum value for size type (4294967295)
PLEASE submit a bug report to https://software.intel.com/en-us/support/priority-support and include the crash backtrace.
Stack dump:
0. Program arguments: /opt/intel/oneapi/compiler/2024.0/bin/compiler/lld -flavor gnu -m elf64_amdgpu --no-undefined -shared -plugin-opt=-amdgpu-internalize-symbols -plugin-opt=mcpu=gfx900 -plugin-opt=O2 --lto-CGO2 -plugin-opt=-vector-library=SVML -plugin-opt=fintel-libirc-allowed -plugin-opt=-disable-hir-generate-mkl-call -plugin-opt=-loopopt=1 -plugin-opt=-intel-abi-compatible=true -plugin-opt=-x86-enable-unaligned-vector-move=true --whole-archive -o /tmp/BrokenLineFitOnGPU-gfx900-bbc203-de42e9.out /tmp/BrokenLineFitOnGPU-gfx900-9738be-475de6.o --no-whole-archive
#0 0x00005602d75bd443 llvm::sys::PrintStackTrace(llvm::raw_ostream&, int) (/opt/intel/oneapi/compiler/2024.0/bin/compiler/lld+0x2d62443)
#1 0x00005602d75bb930 llvm::sys::RunSignalHandlers() (/opt/intel/oneapi/compiler/2024.0/bin/compiler/lld+0x2d60930)
#2 0x00005602d75bdd74 SignalHandler(int) Signals.cpp:0:0
#3 0x00007fef72e54db0 __restore_rt (/lib64/libc.so.6+0x54db0)
#4 0x00007fef72ea154c __pthread_kill_implementation (/lib64/libc.so.6+0xa154c)
#5 0x00007fef72e54d06 gsignal (/lib64/libc.so.6+0x54d06)
#6 0x00007fef72e287f3 abort (/lib64/libc.so.6+0x287f3)
#7 0x00005602d7557062 llvm::report_fatal_error(llvm::Twine const&, bool) (/opt/intel/oneapi/compiler/2024.0/bin/compiler/lld+0x2cfc062)
#8 0x00005602d755bde0 report_at_maximum_capacity(unsigned long) SmallVector.cpp:0:0
#9 0x00005602d755ba48 llvm::SmallVectorBase<unsigned int>::grow_pod(void*, unsigned long, unsigned long) (/opt/intel/oneapi/compiler/2024.0/bin/compiler/lld+0x2d00a48)
#10 0x00005602d794bd17 llvm::SmallVector<int, 64u>::SmallVector<int, void>(llvm::ArrayRef<int>) X86ISelLowering.cpp:0:0
#11 0x00005602d7951c6f lowerShuffleAsBlend(llvm::SDLoc const&, llvm::MVT, llvm::SDValue, llvm::SDValue, llvm::ArrayRef<int>, llvm::APInt const&, llvm::X86Subtarget const&, llvm::SelectionDAG&) X86ISelLowering.cpp:0:0
#12 0x00005602d794dd48 lowerV4F32Shuffle(llvm::SDLoc const&, llvm::ArrayRef<int>, llvm::APInt const&, llvm::SDValue, llvm::SDValue, llvm::X86Subtarget const&, llvm::SelectionDAG&) X86ISelLowering.cpp:0:0
#13 0x00005602d78c0bf8 llvm::X86TargetLowering::LowerOperation(llvm::SDValue, llvm::SelectionDAG&) const X86ISelLowering.cpp:0:0
#14 0x00005602d9bd0081 llvm::DataLayout::getABITypeAlign(llvm::Type*) const (/opt/intel/oneapi/compiler/2024.0/bin/compiler/lld+0x5375081)
#15 0x00005602d877d6b3 llvm::vpo::VPlanTTICostModel::getMemInstAlignment(llvm::vpo::VPLoadStoreInst const*) const (/opt/intel/oneapi/compiler/2024.0/bin/compiler/lld+0x3f226b3)
#16 0x00005602d877f75f llvm::vpo::VPlanTTICostModel::getTTICostForVF(llvm::vpo::VPInstruction const*, unsigned int) (/opt/intel/oneapi/compiler/2024.0/bin/compiler/lld+0x3f2475f)
#17 0x00005602d877f5bc llvm::vpo::VPlanTTICostModel::getTTICost(llvm::vpo::VPInstruction const*) (/opt/intel/oneapi/compiler/2024.0/bin/compiler/lld+0x3f245bc)
#18 0x00005602d87094e0 llvm::vpo::VPlanCostModelWithHeuristics<llvm::vpo::HeuristicsList<llvm::vpo::VPInstruction const>, llvm::vpo::HeuristicsList<llvm::vpo::VPBasicBlock const>, llvm::vpo::HeuristicsList<llvm::vpo::VPlanVector const, llvm::vpo::VPlanCostModelHeuristics::HeuristicSLP, llvm::vpo::VPlanCostModelHeuristics::HeuristicGatherScatter, llvm::vpo::VPlanCostModelHeuristics::HeuristicSpillFill, llvm::vpo::VPlanCostModelHeuristics::HeuristicPsadbw, llvm::vpo::VPlanCostModelHeuristics::HeuristicUnroll>>::getCostImpl(llvm::vpo::VPBasicBlock const*, llvm::raw_ostream*) IntelLoopVectorizationPlanner.cpp:0:0
#19 0x00005602d8709716 llvm::vpo::VPInstructionCost llvm::vpo::VPlanCostModelWithHeuristics<llvm::vpo::HeuristicsList<llvm::vpo::VPInstruction const>, llvm::vpo::HeuristicsList<llvm::vpo::VPBasicBlock const>, llvm::vpo::HeuristicsList<llvm::vpo::VPlanVector const, llvm::vpo::VPlanCostModelHeuristics::HeuristicSLP, llvm::vpo::VPlanCostModelHeuristics::HeuristicGatherScatter, llvm::vpo::VPlanCostModelHeuristics::HeuristicSpillFill, llvm::vpo::VPlanCostModelHeuristics::HeuristicPsadbw, llvm::vpo::VPlanCostModelHeuristics::HeuristicUnroll>>::getRangeCost<llvm::iterator_range<llvm::vpo::sese_df_iterator<llvm::vpo::VPBasicBlock*>>>(llvm::iterator_range<llvm::vpo::sese_df_iterator<llvm::vpo::VPBasicBlock*>>, llvm::raw_ostream*) IntelLoopVectorizationPlanner.cpp:0:0
#20 0x00005602d8708ca7 llvm::vpo::VPlanCostModelWithHeuristics<llvm::vpo::HeuristicsList<llvm::vpo::VPInstruction const>, llvm::vpo::HeuristicsList<llvm::vpo::VPBasicBlock const>, llvm::vpo::HeuristicsList<llvm::vpo::VPlanVector const, llvm::vpo::VPlanCostModelHeuristics::HeuristicSLP, llvm::vpo::VPlanCostModelHeuristics::HeuristicGatherScatter, llvm::vpo::VPlanCostModelHeuristics::HeuristicSpillFill, llvm::vpo::VPlanCostModelHeuristics::HeuristicPsadbw, llvm::vpo::VPlanCostModelHeuristics::HeuristicUnroll>>::getCost(bool, llvm::vpo::VPlanPeelingVariant*, llvm::raw_ostream*) IntelLoopVectorizationPlanner.cpp:0:0
#21 0x00005602d86fc377 llvm::vpo::LoopVectorizationPlanner::selectBestPlan() (/opt/intel/oneapi/compiler/2024.0/bin/compiler/lld+0x3ea1377)
#22 0x00005602d86e9980 llvm::vpo::VPlanDriverHIRImpl::processLoop(llvm::loopopt::HLLoop*, llvm::Function&, llvm::vpo::WRNVecLoopNode*) (/opt/intel/oneapi/compiler/2024.0/bin/compiler/lld+0x3e8e980)
#23 0x00005602d86ec883 llvm::vpo::VPlanDriverHIRImpl::runImpl(llvm::Function&, llvm::loopopt::HIRFramework*, llvm::loopopt::HIRLoopStatistics*, llvm::loopopt::HIRDDAnalysis*, llvm::loopopt::HIRSafeReductionAnalysis*, llvm::OptReportVerbosity::Level, llvm::vpo::WRegionInfo*, llvm::TargetTransformInfo*, llvm::TargetLibraryInfo*, llvm::AssumptionCache*, llvm::DominatorTree*) (/opt/intel/oneapi/compiler/2024.0/bin/compiler/lld+0x3e91883)
#24 0x00005602d86ec627 llvm::vpo::VPlanDriverHIRPass::runImpl(llvm::Function&, llvm::AnalysisManager<llvm::Function>&, llvm::loopopt::HIRFramework&) (/opt/intel/oneapi/compiler/2024.0/bin/compiler/lld+0x3e91627)
#25 0x00005602d83219e5 llvm::detail::PassModel<llvm::Function, llvm::vpo::VPlanDriverHIRPass, llvm::PreservedAnalyses, llvm::AnalysisManager<llvm::Function>>::run(llvm::Function&, llvm::AnalysisManager<llvm::Function>&) PassBuilder.cpp:0:0
#26 0x00005602d9c92d63 llvm::PassManager<llvm::Function, llvm::AnalysisManager<llvm::Function>>::run(llvm::Function&, llvm::AnalysisManager<llvm::Function>&) (/opt/intel/oneapi/compiler/2024.0/bin/compiler/lld+0x5437d63)
#27 0x00005602d7ba304d llvm::detail::PassModel<llvm::Function, llvm::PassManager<llvm::Function, llvm::AnalysisManager<llvm::Function>>, llvm::PreservedAnalyses, llvm::AnalysisManager<llvm::Function>>::run(llvm::Function&, llvm::AnalysisManager<llvm::Function>&) NVPTXTargetMachine.cpp:0:0
#28 0x00005602d9c996f2 llvm::ModuleToFunctionPassAdaptor::run(llvm::Module&, llvm::AnalysisManager<llvm::Module>&) (/opt/intel/oneapi/compiler/2024.0/bin/compiler/lld+0x543e6f2)
#29 0x00005602d7ba2ddd llvm::detail::PassModel<llvm::Module, llvm::ModuleToFunctionPassAdaptor, llvm::PreservedAnalyses, llvm::AnalysisManager<llvm::Module>>::run(llvm::Module&, llvm::AnalysisManager<llvm::Module>&) NVPTXTargetMachine.cpp:0:0
#30 0x00005602d9c91d63 llvm::PassManager<llvm::Module, llvm::AnalysisManager<llvm::Module>>::run(llvm::Module&, llvm::AnalysisManager<llvm::Module>&) (/opt/intel/oneapi/compiler/2024.0/bin/compiler/lld+0x5436d63)
#31 0x00005602d8253b10 runNewPMPasses(llvm::lto::Config const&, llvm::Module&, llvm::TargetMachine*, unsigned int, bool, llvm::ModuleSummaryIndex*, llvm::ModuleSummaryIndex const*) LTOBackend.cpp:0:0
#32 0x00005602d82534bc llvm::lto::opt(llvm::lto::Config const&, llvm::TargetMachine*, unsigned int, llvm::Module&, bool, llvm::ModuleSummaryIndex*, llvm::ModuleSummaryIndex const*, std::vector<unsigned char, std::allocator<unsigned char>> const&) (/opt/intel/oneapi/compiler/2024.0/bin/compiler/lld+0x39f84bc)
#33 0x00005602d825456c llvm::lto::backend(llvm::lto::Config const&, std::function<llvm::Expected<std::unique_ptr<llvm::CachedFileStream, std::default_delete<llvm::CachedFileStream>>> (unsigned int, llvm::Twine const&)>, unsigned int, llvm::Module&, llvm::ModuleSummaryIndex&) (/opt/intel/oneapi/compiler/2024.0/bin/compiler/lld+0x39f956c)
#34 0x00005602d8245eed llvm::lto::LTO::runRegularLTO(std::function<llvm::Expected<std::unique_ptr<llvm::CachedFileStream, std::default_delete<llvm::CachedFileStream>>> (unsigned int, llvm::Twine const&)>) (/opt/intel/oneapi/compiler/2024.0/bin/compiler/lld+0x39eaeed)
#35 0x00005602d8245610 llvm::lto::LTO::run(std::function<llvm::Expected<std::unique_ptr<llvm::CachedFileStream, std::default_delete<llvm::CachedFileStream>>> (unsigned int, llvm::Twine const&)>, std::function<llvm::Expected<std::function<llvm::Expected<std::unique_ptr<llvm::CachedFileStream, std::default_delete<llvm::CachedFileStream>>> (unsigned int, llvm::Twine const&)>> (unsigned int, llvm::StringRef, llvm::Twine const&)>) (/opt/intel/oneapi/compiler/2024.0/bin/compiler/lld+0x39ea610)
#36 0x00005602d7739a45 lld::elf::BitcodeCompiler::compile() (/opt/intel/oneapi/compiler/2024.0/bin/compiler/lld+0x2edea45)
#37 0x00005602d769c7c1 lld::elf::LinkerDriver::link(llvm::opt::InputArgList&) (/opt/intel/oneapi/compiler/2024.0/bin/compiler/lld+0x2e417c1)
#38 0x00005602d768f5ec lld::elf::LinkerDriver::linkerMain(llvm::ArrayRef<char const*>) (/opt/intel/oneapi/compiler/2024.0/bin/compiler/lld+0x2e345ec)
#39 0x00005602d768d908 lld::elf::link(llvm::ArrayRef<char const*>, llvm::raw_ostream&, llvm::raw_ostream&, bool, bool) (/opt/intel/oneapi/compiler/2024.0/bin/compiler/lld+0x2e32908)
#40 0x00005602d75bf820 lld::unsafeLldMain(llvm::ArrayRef<char const*>, llvm::raw_ostream&, llvm::raw_ostream&, llvm::ArrayRef<lld::DriverDef>, bool) (/opt/intel/oneapi/compiler/2024.0/bin/compiler/lld+0x2d64820)
#41 0x00005602d7543226 lld_main(int, char**, llvm::ToolContext const&) (/opt/intel/oneapi/compiler/2024.0/bin/compiler/lld+0x2ce8226)
#42 0x00005602d754386e main (/opt/intel/oneapi/compiler/2024.0/bin/compiler/lld+0x2ce886e)
#43 0x00007fef72e3feb0 __libc_start_call_main (/lib64/libc.so.6+0x3feb0)
#44 0x00007fef72e3ff60 __libc_start_main@GLIBC_2.2.5 (/lib64/libc.so.6+0x3ff60)
#45 0x00005602d7542ee9 _start (/opt/intel/oneapi/compiler/2024.0/bin/compiler/lld+0x2ce7ee9)
llvm-foreach: Aborted (core dumped)
icpx: fatal error: amdgcn-link command failed with exit code 254 (use -v to see invocation)
Intel(R) oneAPI DPC++/C++ Compiler 2024.0.2 (2024.0.2.20231213)
Target: x86_64-unknown-linux-gnu
Thread model: posix
InstalledDir: /opt/intel/oneapi/compiler/2024.0/bin/compiler
Configuration file: /opt/intel/oneapi/compiler/2024.0/bin/compiler/../icpx.cfgthis is the file that causes the crash: cernbox and the command to reproduce it is:
/opt/intel/oneapi/compiler/2024.0/bin/compiler/lld -flavor gnu -m elf64_amdgpu --no-undefined -shared -plugin-opt=-amdgpu-internalize-symbols -plugin-opt=mcpu=gfx900 -plugin-opt=O2 --lto-CGO2 -plugin-opt=-vector-library=SVML -plugin-opt=fintel-libirc-allowed -plugin-opt=-disable-hir-generate-mkl-call -plugin-opt=-loopopt=1 -plugin-opt=-intel-abi-compatible=true -plugin-opt=-x86-enable-unaligned-vector-move=true --whole-archive -o BrokenLineFitOnGPU-gfx900-bbc203-de42e9.out BrokenLineFitOnGPU-gfx900-9738be-475de6.o --no-whole-archiveTo compile the entire project:
git clone -b sycl_bugfix https://github.com/AuroraPerego/pixeltrack-standalone.git
cd pixeltrack-standalone
make environment
. env.sh
make sycl
ashadrina
rafbiels
Describe the bug
Recently, with help from @fwyzard and Aurora (I don't know her GitHub username unfortunately... :frowning:), we did some experiments with building hybrid Intel+NVIDIA+AMD binaries using the oneAPI compiler. The winning incantation being of the following kind:
This works well for "simple enough" source code. :wink: But similar to #8065, once I try to build our "most complicated" project, things break. :frowning: However, with the current latest version of oneAPI (2024.0.1), the failure is different than what I described in #8065. During the build I get:
To reproduce
As in the other ticket, reproducing this "from scratch" is a bit of an involved process unfortunately. :frowning: But since the problem is during compilation in this case, I can "simply" give you the preprocessed sources. :smile: You can download that file (it's pretty big) from:
https://cernbox.cern.ch/s/Dc83iva0Dl2czCF
With that file downloaded, you can reproduce this issue like:
The thing is, it doesn't always crash. :thinking: Just now, I was able to execute:
While a little earlier the preprocessed file made my compilation crash as expected. :confused:
(I was in a different subdirectory on that try...)
Environment
Don't be fooled by what the compiler reports for HIP though. :thinking: The oneAPI plugin is set up to use ROCm-5.4.6.
I just happen to do a native HIP build with version 6.0.2 at the same time...
Additional context
The build for "just" an Intel+NVIDIA backend does work. Only once I try to build for all 3 in parallel, that I see this issue pop up. :thinking: