[
https://issues.apache.org/jira/browse/KUDU-3545?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17916174#comment-17916174
]
ASF subversion and git services commented on KUDU-3545:
-------------------------------------------------------
Commit d932324e802aa397324c2181dfc55aaee39f4f69 in kudu's branch
refs/heads/master from Alexey Serbin
[ https://gitbox.apache.org/repos/asf?p=kudu.git;h=d932324e8 ]
[thirdparty] fix building LLVM/CLANG 11.0.0 with GCC13
GCC13 (G++13) is the default system compiler on contemporary Linux OS
flavors such as Ubuntu 24.04 LTS. With this patch it's now possible
to compile LLVM/CLANG in Kudu's thirdparty on Ubuntu 24.04 and 24.04.1
LTS.
This patch picks up a few relevant updates from the LLVM project's
upstream repository [1] with minor tweaks to adapt to the change in
the source files layout that happened in LLVM/CLANG 12. The updates
prevent the leakage of the parts of the standard library installed
on the build machine into the components of the LLVM/CLANG
that are supposed to be completely freestanding and standalone.
I guess it should help not only with making it possible to build
LLVM/CLANG 11 with GCC13, but also address the root cause of the issues
we have hit on SLES15 and elsewhere in the codegen that's reported in
a few Jira items such as KUDU-3479 and KUDU-3545 [2][3].
However, there might be some extra patches beyond the ones already
included. Anyway, I haven't verified this hypothesis yet.
We definitely need to upgrade from LLVM 11.0.0 to LLVM 19.1.6 and I
suspect it should help us to address [2] and [3] along with other
updates and fixes. However, it's a separate task on its own,
while this patch focuses on making the current version of LLVM/CLANG
in the Kudu's thirdparty buildable on recent Linux OS releases
(such as Ubuntu 24.04 LTS).
I also snuck in an update to skip building compiler-rt tests since
they aren't run in the scope of building LLVM/CLANG as a thirdparty
component of the Apache Kudu project.
[1] https://github.com/llvm/llvm-project
[2] https://issues.apache.org/jira/browse/KUDU-3479
[3] https://issues.apache.org/jira/browse/KUDU-3545
Change-Id: I9b8282c85bbac5f25eae885e4b3ec183104bc540
Reviewed-on: http://gerrit.cloudera.org:8080/22279
Tested-by: Alexey Serbin <ale...@apache.org>
Reviewed-by: Ashwani Raina <ara...@cloudera.com>
Reviewed-by: Abhishek Chennaka <achenn...@cloudera.com>
> codegen test fails on SLES with higher libgcc version
> -----------------------------------------------------
>
> Key: KUDU-3545
> URL: https://issues.apache.org/jira/browse/KUDU-3545
> Project: Kudu
> Issue Type: Bug
> Components: codegen
> Reporter: Ashwani Raina
> Assignee: Ashwani Raina
> Priority: Minor
>
> On a SLES 15 withlibgcc_s1-13.2.1+git7813-150000.1.6.1.x86_64 version,
> codegen-test fails with following crash:
> {noformat}
> *** SIGABRT (@0x3162e) received by PID 202286 (TID 0x7f71d1bfe700) from PID
> 202286; stack trace: ***
> @ 0x7f71d41f5910 (unknown)
> @ 0x7f71d2725d2b __GI_raise
> @ 0x7f71d27273e5 __GI_abort
> @ 0x7f71d28d78d7 (unknown)
> @ 0x7f71d28f1009 __deregister_frame
> @ 0x7f71d4d6c9e0 llvm::RTDyldMemoryManager::deregisterEHFrames()
> @ 0x7f71d4976b02 llvm::MCJIT::~MCJIT()
> @ 0x7f71d4977241 llvm::MCJIT::~MCJIT()
> @ 0x7f71d481c222 std::default_delete<>::operator()()
> @ 0x7f71d481c12d std::unique_ptr<>::~unique_ptr()
> @ 0x7f71d481bfaf kudu::codegen::JITWrapper::~JITWrapper()
> @ 0x7f71d4835f34
> kudu::codegen::RowProjectorFunctions::~RowProjectorFunctions()
> @ 0x7f71d4835f50
> kudu::codegen::RowProjectorFunctions::~RowProjectorFunctions()
> @ 0x46297c kudu::RefCountedThreadSafe<>::DeleteInternal()
> @ 0x45f3d1 kudu::DefaultRefCountedThreadSafeTraits<>::Destruct()
> @ 0x45acb0 kudu::RefCountedThreadSafe<>::Release()
> @ 0x7f71d480c191
> kudu::codegen::CodeCache::EvictionCallback::EvictedEntry()
> @ 0x7f71d3c5e4bb kudu::(anonymous
> namespace)::CacheShard<>::FreeEntry()
> @ 0x7f71d3c60b31 kudu::(anonymous namespace)::CacheShard<>::Insert()
> @ 0x7f71d3c5fb73 kudu::(anonymous namespace)::ShardedCache<>::Insert()
> @ 0x7f71d480bab6 kudu::codegen::CodeCache::AddEntry()
> @ 0x7f71d4811fea kudu::codegen::(anonymous
> namespace)::CompilationTask::RunWithStatus()
> @ 0x7f71d4811a64 kudu::codegen::(anonymous
> namespace)::CompilationTask::Run()
> @ 0x7f71d481288a
> _ZZN4kudu7codegen18CompilationManager19RequestRowProjectorEPKNS_6SchemaES4_PSt10unique_ptrINS0_12RowProjectorESt14default_deleteIS6_EEENKUlvE_clEv
> @ 0x7f71d4813e72
> _ZNSt17_Function_handlerIFvvEZN4kudu7codegen18CompilationManager19RequestRowProjectorEPKNS1_6SchemaES6_PSt10unique_ptrINS2_12RowProjectorESt14default_deleteIS8_EEEUlvE_E9_M_invokeERKSt9_Any_data
> @ 0x452430 std::function<>::operator()()
> @ 0x7f71d3d98648 kudu::ThreadPool::DispatchThread()
> @ 0x7f71d3d98ee9 _ZZN4kudu10ThreadPool12CreateThreadEvENKUlvE_clEv
> @ 0x7f71d3d9a6a0
> _ZNSt17_Function_handlerIFvvEZN4kudu10ThreadPool12CreateThreadEvEUlvE_E9_M_invokeERKSt9_Any_data
> @ 0x452430 std::function<>::operator()()
> @ 0x7f71d3d89482 kudu::Thread::SuperviseThread()
> @ 0x7f71d41e96ea start_thread
> {noformat}
> From the stack frame, it seems that __deregister_frame is probably being fed
> some invalid input that is already de-initialised before calling the
> __deregister_frame.
> We seem to be hitting this assert:
> [https://github.com/gcc-mirror/gcc/blob/65e2c932019b4e36d7c1d49952dc006fa7419a3d/libgcc/unwind-dw2-fde.c#L291C11-L291C11]
> gcc_assert (in_shutdown || ob);
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
