FAQs About ABI Configuration

Symptom

Symptom 1

The error message "undefined reference to "std::__cxx11 ***" is displayed during the compilation and linking of ATB or MKI.

fffec0150000-fffec0160000 rw-p 214a70000 08:11 42336262	/datal/models/Lite-32K/SFT-V4.0-20240621-INT8/SFT-13B v20240621 longbase ep3.data
fffec0160000-fffec0170000 rw-p e8d90000 08:11 42336262	/datal/models/Lite-32K/SFT-V4.0-20240621-INT8/SFT-13B v20240621 longbase ep3.data
fffec0170000-fffec0180000 rw-p 1a03c0000 08:11 42336262	/datal/models/Lite-32K/SFT-V4.0-20240621-INT8/SFT-13B v20240621 longbase_ep3.data
fffec0180000-fffec01a0000 rw-p 1a03b0000 08:11 42336262	/datal/models/Lite-32K/SFT-V4.0-20240621-INT8/SFT-13B v20240621 longbase ep3.data
fffec01a0000-fffec01c0000 rw-p 1b8530000 08:11 42336262	/datal/models/Lite-32K/SFT-V4.0-20240621-INT8/SFT-13B v20240621 longbase ep3.data
Program received signal SIGABRT, Aborted.
[Switching to Thread 0xfff7aa3a1990 (LWP 41895)]
0x0000ffffbc3850e8 in raise () from /Lib64/libc.so.6
Missing separate debuginfos, use: debuginfo-install glibc-2.17-260.el7.aarch64 zlib-1.2.7-18.el7.aarch64
(gdb) bt
#0  0x0000ffffbc3850e8 in raise () from /lib64/libc.so.6
#1  0x0000ffffoc386760 in abort () from /lib64/libc.so.6
#2  0x0000ffffbc3c5048 in __Libc_message () from /Lib64/Libc.so.6
#3  0x0000ffffbc3cd58c in _int free () from /lib64/libc.so.6
#4  0x0000ffffadd85c08 in std::__detail::_Compiler<std::__cxx11::regex_traits<char> >::_Compiler(char const*, std::locale const&, std::tegex_con
    stants::syntax option type) () from /usr/local/Ascend/nnal/atb/latest/atb/cxx_abi_1/lib/libmki.so
#5  0x0000ffffadd7450c in Mki::LogSinkFile::DeleteOldestFile() () from /usr/local/Ascend/nnal/atb/latest/atb/cxx_abi_1/lib/libmki.so
#6  0x0000ffffadd74df4 in Mki::LogSinkFile::OpenFile() () from /usr/local/Ascend/nnal/atb/latest/atb/cxx_abi_1/lib/libmki.so
#7	0x0000ffffadd75080 in Mki::LogSinkFile::Log(char const*, unsigned long) () from /usr/local/Ascend/nnal/atb/latest/atb/cxx_abi_1/lib/libmki.so
#8  0x000Offffadd6f4e4 in Mki::LogCore::Log(char const*, unsigned long) () from /usr/local/Ascend/nnal/atb/latest/atb/cxx_abi_1/lib/libmki.so
#9  0x0000ffffadd863Oc in Mki::LogStream::~LogStream() () from /usr/local/Ascend/nnal/atb/latest/atb/cxx_abi_1/lib/libmki.so
#10 0x0000ffffb0494abO in atb::SelfAttentionOperation::PAMaskDimCheckNz(atb::SVector<atb::TensorDesc> const&) const ()
    from /usr/local/Ascend/nnal/atb/latest/atb/cxx_abi_1/lib/libatb.so
#11 0x0000ffffb0496984 in atb::SelfAttentionOperation::InferShapePADimCheck(atb::SVector<atb::TensorDesc> const&) const ()
    from /usr/local/Ascend/nnal/atb/latest/atb/cxx_abi_1/lib/libatb.so
#12 0x0000ffffb0498ae0 in atb::SelfAttentionOperation::SetupCheckImpl(atb::SVector<atb::Tensor> const&, atb::SVector<atb::Tensor> const&) const
    from /usr/local/Ascend/nnal/atb/latest/atb/cxx_abi_1/lib/libatb.so
#13 0x0000ffffb04e77ac in atb::OperationBase::SetupCheck(atb::VariantPack const&, atb::Context*) ()
    from /usr/local/Ascend/nnal/atb/latest/atb/cxx_abi_1/lib/libatb.so
#14 0x0000ffffb04eca4c in atb::OperationBase::Setup(atb::VariantPack const&, unsigned long&, atb::Context*) ()
    from /usr/local/Ascend/nnal/atb/latest/atb/cxx_abi_1/lib/libatb.so

Symptom 2

"undefined symbol: ***cxx11***" is displayed when ATB is executed.

python {source_code_path}/ascend-transformer-boost/tests/apitest/torch_atb_test/op_test/linear_parallel/matmul_allreduce_pertoken_test.py
Traceback (most recent call last):
  File "{python_install_path}/site-packages/torch_atb/__init__.py", line 25, in _load_atb_libs
    ctypes.CDLL(str(atb_lib_path / lib_file), mode=ctypes.RTLD_GLOBAL)
  File "{python_install_path}/ctypes/__init__.py", line 374, in __init__
    self._handle = _dlopen(self._name, mode)
OSError: {CANN_nnal_install_path}/nnal/atb/latest/atb/cxx_abi_1/lib/libcann_ops_adapter.so: undefined symbol: _ZN3Mki6Status10FailStatusEiRKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "{source_code_path}/ascend-transformer-boost/tests/apitest/torch_atb_test/op_test/linear_parallel/matmul_allreduce_pertoken_test.py", line 12, in <module>
    import torch_atb
  File "{python_install_path}/site-packages/torch_atb/__init__.py", line 35, in <module>
    _load_atb_libs()
  File "{python_install_path}/site-packages/torch_atb/__init__.py", line 27, in _load_atb_libs
    raise RuntimeError(f"Failed to load {lib_file}: {err}") from err
RuntimeError: Failed to load libasdops.so: {CANN_nnal_install_path}/nnal/atb/latest/atb/cxx_abi_1/lib/libcann_ops_adapter.so: undefined symbol: _ZN3Mki6Status10FailStatusEiRKNSt7__cxx1112basic_stringIcSt11char_traitsIcESaIcEEE
[ERROR] 2025-07-28-11:58:16 (PID:1974512, Device:-1, RankID:-1) ERR99999 UNKNOWN application exception

Symptom 3

An error occurs when param with members of string type is used to create an ATB operation. The log prints "atb create operation failed".

[2025-07-28 14:24:11.340266] [error] [2063924] [all_reduce_operation.cpp:86] param rsv has a non-zero value, please check the compilation version.
[2025-07-28 14:24:11.340266] [error] [2063923] [all_reduce_operation.cpp:86] param rsv has a non-zero value, please check the compilation version.
all_reduce_demo.cpp:58 [error]: all_reduce_demo.cpp:58 [error]: 11

atb create operation failedatb create operation failed

all_reduce_demo.cpp:137 [error]: 1
multithread task 0 failed
all_reduce_demo.cpp:137 [error]: 1
multithread task 1 failed

Cause Analysis

The preceding examples show errors reported when abi1 is used to compile or execute ATB-related functions in the framework of the abi0 version.
Different versions of abi (such as the old ABI and C++11 ABI) implement std::string in different ways. The old ABI (C++98/03) has unique implementation in memory management and small string optimization (SSO), while the C++11 ABI improves memory allocation and shared memory management.

These differences may cause memory conflicts, crashes, or link errors when std::string objects are transferred in programs compiled in different ABI versions. For details about the current ATB configuration logic for the ABI version, see ABI configuration method.

Solution

Torch available:

# Use torch to check the ABI version corresponding to the current framework.
echo $(python3 -c 'import torch; print(torch.compiled_with_cxx11_abi())')
# If true is printed:
source ${nnal_install_path}/nnal/atb/set_env.sh --cxx_abi=1
# Or
source ${nnal_install_path}/nnal/atb/set_env.sh
# If false is printed:
source ${nnal_install_path}/nnal/atb/set_env.sh --cxx_abi=0
# Or
source ${nnal_install_path}/nnal/atb/set_env.sh

No torch required:
- If you cannot determine which ABI version to use, check your environment setup, and decide whether to use the default version as instructed in ascendtb_0046.html#EN-US_TOPIC_0000002757055899__li13966195614383 or another version.
- You can try to use the ABI configuration opposite to the current setting.

Parent topic: Troubleshooting Cases