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.

Sign up for GitHub

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

Jump to bottom

Arithmetic exception on old AMD dual-core CPU #2716

Open
grialion opened this issue Feb 17, 2025 · 8 comments
Open

Arithmetic exception on old AMD dual-core CPU #2716

grialion opened this issue Feb 17, 2025 · 8 comments
Assignees
@kminemur
Labels
bug A confirmed library bug help wanted platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64

Comments

@grialion
Copy link

Summary

I'm trying to get Open-WebUI running on my old AMD Athlon dual-core server.

DNNL crashes with exit code 136.
GDB shows that it crashed in the dnnl library.

Version

I was not able to get any logs using the environment variable ONEDNN_VERBOSE=all /path/to/python /path/to/open-webui serve
However the project seems to have torch v2.5.1 pinned.

Environment

lscpu output:

Architecture:             x86_64
  CPU op-mode(s):         32-bit, 64-bit
  Address sizes:          40 bits physical, 48 bits virtual
  Byte Order:             Little Endian
CPU(s):                   2
  On-line CPU(s) list:    0,1
Vendor ID:                AuthenticAMD
  Model name:             AMD Athlon(tm) 64 X2 Dual Core Processor 5000+
    CPU family:           15
    Model:                107
    Thread(s) per core:   1
    Core(s) per socket:   2
    Socket(s):            1
    Stepping:             2
    CPU(s) scaling MHz:   69%
    CPU max MHz:          2600.0000
    CPU min MHz:          1000.0000
    BogoMIPS:             5223.78
    Flags:                fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 h
                          t syscall nx mmxext fxsr_opt rdtscp lm 3dnowext 3dnow rep_good nopl cpuid extd_apicid pni cx16 lahf
                          _lm cmp_legacy svm extapic cr8_legacy 3dnowprefetch vmmcall lbrv
Virtualization features:  
  Virtualization:         AMD-V
Caches (sum of all):      
  L1d:                    128 KiB (2 instances)
  L1i:                    128 KiB (2 instances)
  L2:                     1 MiB (2 instances)
NUMA:                     
  NUMA node(s):           1
  NUMA node0 CPU(s):      0,1
Vulnerabilities:          
  [...]

Steps to reproduce

Start Open-WebUI on a dual-core Athlon CPU, e.g uvx --python 3.11 open-webui@latest serve.
dnnl::impl::cpu::x64::get_effective_cpu_isa is eventually called, leading to the arithmetic exception.

Observed behavior

After starting the web application, it crashes with exit code 136.
Using bash, it shows Floating point exception (core dumped).

This is the application output:

$ ONEDNN_VERBOSE=all [...]/bin/python [...]/bin/open-webui serve
Loading WEBUI_SECRET_KEY from file, not provided as an environment variable.
Loading WEBUI_SECRET_KEY from /home/grial/ollama/.webui_secret_key
[...]/lib/python3.11/site-packages/open_webui
[...]/lib/python3.11/site-packages
[...]/lib/python3.11
Running migrations
INFO  [alembic.runtime.migration] Context impl SQLiteImpl.
INFO  [alembic.runtime.migration] Will assume non-transactional DDL.
INFO  [open_webui.env] 'DEFAULT_LOCALE' loaded from the latest database entry
INFO  [open_webui.env] 'DEFAULT_PROMPT_SUGGESTIONS' loaded from the latest database entry
WARNI [open_webui.env] 

WARNING: CORS_ALLOW_ORIGIN IS SET TO '*' - NOT RECOMMENDED FOR PRODUCTION DEPLOYMENTS.

INFO  [open_webui.env] Embedding model set: sentence-transformers/all-MiniLM-L6-v2
WARNI [langchain_community.utils.user_agent] USER_AGENT environment variable not set, consider setting it to identify your requests.

  ___                    __        __   _     _   _ ___
 / _ \ _ __   ___ _ __   \ \      / /__| |__ | | | |_ _|
| | | | '_ \ / _ \ '_ \   \ \ /\ / / _ \ '_ \| | | || |
| |_| | |_) |  __/ | | |   \ V  V /  __/ |_) | |_| || |
 \___/| .__/ \___|_| |_|    \_/\_/ \___|_.__/ \___/|___|
      |_|


v0.5.12 - building the best open-source AI user interface.

https://github.com/open-webui/open-webui

Floating point exception (core dumped)

Here's the coredumpctl log:

           PID: 8421 (python)
           UID: 1001 (user)
           GID: 1001 (user)
        Signal: 8 (FPE)
     Timestamp: Mon 2025-02-17 21:31:03 CET (52s ago)
  Command Line: [...]/bin/python [...]/bin/open-webui serve
    Executable: [...]/bin/python3.11

Using gdb to view the backtrace, I see the following error:

Program terminated with signal SIGFPE, Arithmetic exception.
#0  0x00007a57f234c383 in Xbyak::util::Cpu::Cpu() ()
   from [...]/lib/python3.11/site-packages/torch/lib/libtorch_cpu.so
Here is the truncated backtrace: 
#0  0x00007a57f234c383 in Xbyak::util::Cpu::Cpu() () from [...]/lib/python3.11/site-packages/torch/lib/libtorch_cpu.so
#1  0x00007a57f2c0e076 in dnnl::impl::cpu::x64::(anonymous namespace)::mayiuse(dnnl::impl::cpu::x64::cpu_isa_t, bool) () from [...]/lib/python3.11/site-packages/torch/lib/libtorch_cpu.so
#2  0x00007a57f2c0dddd in dnnl::impl::cpu::x64::(anonymous namespace)::mayiuse(dnnl::impl::cpu::x64::cpu_isa_t, bool) () from [...]/lib/python3.11/site-packages/torch/lib/libtorch_cpu.so
#3  0x00007a57f2c0f16e in dnnl::impl::cpu::x64::get_effective_cpu_isa() () from [...]/lib/python3.11/site-packages/torch/lib/libtorch_cpu.so
#4  0x00007a57e606b46d in ideep::has_bf16_type_support() () from [...]/lib/python3.11/site-packages/torch/lib/libtorch_cpu.so
#5  0x00007a57e606bb43 in at::native::use_mkldnn_matmul(at::Tensor const&, at::Tensor const&, at::Tensor const&) () from [...]/lib/python3.11/site-packages/torch/lib/libtorch_cpu.so
#6  0x00007a57e5836c71 in at::native::dot(at::Tensor const&, at::Tensor const&) () from [...]/lib/python3.11/site-packages/torch/lib/libtorch_cpu.so
#7  0x00007a57e68c11f0 in c10::impl::wrap_kernel_functor_unboxed_<c10::impl::detail::WrapFunctionIntoFunctor_<c10::CompileTimeFunctionPointer<at::Tensor (at::Tensor const&, at::Tensor const&), &at::(anonymous namespace)::(anonymous namespace)::wrapper_CPU__dot>, at::Tensor, c10::guts::typelist::typelist<at::Tensor const&, at::Tensor const&> >, at::Tensor (at::Tensor const&, at::Tensor const&)>::call(c10::OperatorKernel*, c10::DispatchKeySet, at::Tensor const&, at::Tensor const&) () from [...]/lib/python3.11/site-packages/torch/lib/libtorch_cpu.so
#8  0x00007a57e648a6ae in at::_ops::dot::redispatch(c10::DispatchKeySet, at::Tensor const&, at::Tensor const&) () from [...]/lib/python3.11/site-packages/torch/lib/libtorch_cpu.so
#9  0x00007a57e8abfe3f in torch::autograd::VariableType::(anonymous namespace)::dot(c10::DispatchKeySet, at::Tensor const&, at::Tensor const&) () from [...]/lib/python3.11/site-packages/torch/lib/libtorch_cpu.so
#10 0x00007a57e8ac0813 in c10::impl::wrap_kernel_functor_unboxed_<c10::impl::detail::WrapFunctionIntoFunctor_<c10::CompileTimeFunctionPointer<at::Tensor (c10::DispatchKeySet, at::Tensor const&, at::Tensor const&), &torch::autograd::VariableType::(anonymous namespace)::dot>, at::Tensor, c10::guts::typelist::typelist<c10::DispatchKeySet, at::Tensor const&, at::Tensor const&> >, at::Tensor (c10::DispatchKeySet, at::Tensor const&, at::Tensor const&)>::call(c10::OperatorKernel*, c10::DispatchKeySet, at::Tensor const&, at::Tensor const&) ()
 from [...]/lib/python3.11/site-packages/torch/lib/libtorch_cpu.so
#11 0x00007a57e64f1f80 in at::_ops::dot::call(at::Tensor const&, at::Tensor const&) () from [...]/lib/python3.11/site-packages/torch/lib/libtorch_cpu.so
#12 0x00007a57e59ab8c8 in at::native::_matmul_impl(at::Tensor&, at::Tensor const&, at::Tensor const&) () from [...]/lib/python3.11/site-packages/torch/lib/libtorch_cpu.so
#13 0x00007a57e59b4339 in at::native::matmul(at::Tensor const&, at::Tensor const&) () from [...]/lib/python3.11/site-packages/torch/lib/libtorch_cpu.so
#14 0x00007a57e6c81f90 in c10::impl::wrap_kernel_functor_unboxed_<c10::impl::detail::WrapFunctionIntoFunctor_<c10::CompileTimeFunctionPointer<at::Tensor (at::Tensor const&, at::Tensor const&), &at::(anonymous namespace)::(anonymous namespace)::wrapper_CompositeImplicitAutograd__matmul>, at::Tensor, c10::guts::typelist::typelist<at::Tensor const&, at::Tensor const&> >, at::Tensor (at::Tensor const&, at::Tensor const&)>::call(c10::OperatorKernel*, c10::DispatchKeySet, at::Tensor const&, at::Tensor const&) () from [...]/lib/python3.11/site-packages/torch/lib/libtorch_cpu.so
#15 0x00007a57e67cb750 in at::_ops::matmul::call(at::Tensor const&, at::Tensor const&) () from [...]/lib/python3.11/site-packages/torch/lib/libtorch_cpu.so
#16 0x00007a57f93f1cab in torch::autograd::THPVariable_matmul(_object*, _object*, _object*) () from [...]/lib/python3.11/site-packages/torch/lib/libtorch_python.so
#17 0x00007a580695a4db in method_vectorcall_VARARGS_KEYWORDS () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#18 0x00007a58069be949 in _PyEval_EvalFrameDefault () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#19 0x00007a5806955855 in _PyFunction_Vectorcall () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#20 0x00007a58069586be in method_vectorcall () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#21 0x00007a5806a75ff8 in _PyObject_CallFunctionVa () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#22 0x00007a5806a763dd in PyObject_CallMethod () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#23 0x00007a57f9b76d2c in torch::handle_torch_function_no_python_arg_parser(c10::ArrayRef<_object*>, _object*, _object*, char const*, _object*, char const*, torch::TorchFunctionName) () from [...]/lib/python3.11/site-packages/torch/lib/libtorch_python.so
#24 0x00007a57f9b77c37 in torch::handle_torch_function(torch::PythonArgs&, _object*, _object*, _object*, _object*, char const*, char const*) () from [...]/lib/python3.11/site-packages/torch/lib/libtorch_python.so
#25 0x00007a57f93f1afd in torch::autograd::THPVariable_matmul(_object*, _object*, _object*) () from [...]/lib/python3.11/site-packages/torch/lib/libtorch_python.so
#26 0x00007a57f93f1e17 in _object* torch::autograd::TypeError_to_NotImplemented_<&torch::autograd::THPVariable_matmul>(_object*, _object*, _object*) () from [...]/lib/python3.11/site-packages/torch/lib/libtorch_python.so
#27 0x00007a580695a4db in method_vectorcall_VARARGS_KEYWORDS () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#28 0x00007a5806aaff4f in vectorcall_maybe () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#29 0x00007a580684a8e7 in slot_nb_matrix_multiply () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#30 0x00007a5806a619ab in binary_op () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#31 0x00007a580697ac06 in cfunction_vectorcall_FASTCALL () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#32 0x00007a58069c2389 in _PyEval_EvalFrameDefault () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#33 0x00007a5806ae12fb in _PyEval_Vector () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#34 0x00007a5806ae10df in PyEval_EvalCode () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#35 0x00007a5806adf013 in builtin_exec () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#36 0x00007a580697acdd in cfunction_vectorcall_FASTCALL_KEYWORDS () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#37 0x00007a58069c758b in _PyEval_EvalFrameDefault () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#38 0x00007a5806955855 in _PyFunction_Vectorcall () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#39 0x00007a5806954c46 in _PyObject_VectorcallTstate.753 () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#40 0x00007a5806958316 in object_vacall () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#41 0x00007a5806a769a3 in PyObject_CallMethodObjArgs () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#42 0x00007a5806a0ae21 in PyImport_ImportModuleLevelObject () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#43 0x00007a58069bbca0 in _PyEval_EvalFrameDefault () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#44 0x00007a5806ae12fb in _PyEval_Vector () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#45 0x00007a5806ae10df in PyEval_EvalCode () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#46 0x00007a5806adf013 in builtin_exec () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#47 0x00007a580697acdd in cfunction_vectorcall_FASTCALL_KEYWORDS () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#48 0x00007a58069c758b in _PyEval_EvalFrameDefault () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#49 0x00007a5806955855 in _PyFunction_Vectorcall () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#50 0x00007a5806954c46 in _PyObject_VectorcallTstate.753 () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#51 0x00007a5806958316 in object_vacall () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#52 0x00007a5806a769a3 in PyObject_CallMethodObjArgs () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#53 0x00007a5806a0ae21 in PyImport_ImportModuleLevelObject () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#54 0x00007a58069bbca0 in _PyEval_EvalFrameDefault () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#55 0x00007a5806ae12fb in _PyEval_Vector () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#56 0x00007a5806ae10df in PyEval_EvalCode () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#57 0x00007a5806adf013 in builtin_exec () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#58 0x00007a580697acdd in cfunction_vectorcall_FASTCALL_KEYWORDS () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#59 0x00007a58069c758b in _PyEval_EvalFrameDefault () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#60 0x00007a5806955855 in _PyFunction_Vectorcall () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#61 0x00007a5806954c46 in _PyObject_VectorcallTstate.753 () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#62 0x00007a5806958316 in object_vacall () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#63 0x00007a5806a769a3 in PyObject_CallMethodObjArgs () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#64 0x00007a5806a0ae21 in PyImport_ImportModuleLevelObject () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#65 0x00007a58069bbca0 in _PyEval_EvalFrameDefault () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#66 0x00007a5806ae12fb in _PyEval_Vector () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#67 0x00007a5806ae10df in PyEval_EvalCode () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#68 0x00007a5806adf013 in builtin_exec () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#69 0x00007a580697acdd in cfunction_vectorcall_FASTCALL_KEYWORDS () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#70 0x00007a58069c758b in _PyEval_EvalFrameDefault () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#71 0x00007a5806955855 in _PyFunction_Vectorcall () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
#72 0x00007a5806954c46 in _PyObject_VectorcallTstate.753 () from [...]/cpython-3.11.11-linux-x86_64-gnu/bin/../lib/libpython3.11.so.1.0
[...]
@grialion grialion added the sighting Suspicious library behavior. Should be promoted to a bug when confirmed label Feb 17, 2025
@shu1chen shu1chen added the platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64 label Feb 18, 2025
@kminemur
Copy link

kminemur commented Feb 18, 2025 •
edited
Loading

@grialion could you provide more specific reproduce steps and corresponding lib versions?

e.g.,
openWeb xxx
oneDNN yyy
torch v2.5.1

@kminemur
Copy link

kminemur commented Feb 18, 2025 •
edited
Loading

Hi @grialion
can you try DNNL_VERBOSE=2 instead of ONEDNN_VERBOSE=all?

@vpirogov
Copy link
Member

Looks like Xbyak fails to detect number of cores or cache size on this processor and dies in this function. To be fair this processor is over 15 years old...

@vpirogov
Copy link
Member

If you want to make oneDNN work on this processor please work with @herumi to fix the issue in upstream Xbyak. We'll propagate the changes to oneDNN and further downstream.

And thanks for the detailed report!

@vpirogov vpirogov added help wanted bug A confirmed library bug and removed sighting Suspicious library behavior. Should be promoted to a bug when confirmed labels Feb 19, 2025
@grialion
Copy link
Author

Hi @grialion can you try DNNL_VERBOSE=2 instead of ONEDNN_VERBOSE=all?

Running it with DNNL_VERBOSE=2 yields no additional logs either @kminemur:

$ DNNL_VERBOSE=2 /path/to/open-webui serve
[...]
  ___                    __        __   _     _   _ ___
 / _ \ _ __   ___ _ __   \ \      / /__| |__ | | | |_ _|
| | | | '_ \ / _ \ '_ \   \ \ /\ / / _ \ '_ \| | | || |
| |_| | |_) |  __/ | | |   \ V  V /  __/ |_) | |_| || |
 \___/| .__/ \___|_| |_|    \_/\_/ \___|_.__/ \___/|___|
      |_|


v0.5.12 - building the best open-source AI user interface.

https://github.com/open-webui/open-webui

Floating point exception (core dumped)

If you want to make oneDNN work on this processor please work with @herumi to fix the issue in upstream Xbyak. We'll propagate the changes to oneDNN and further downstream.

And thanks for the detailed report!

Yes, I'd like to make it work, however I would like to use the computation-heavy parts on my GPU. Is it possible to off-load the work to the GPU instead of using oneDNN on the CPU?

If it's not an option, I'd help making this CPU supported.
If I understood correctly Xbyak is a header-only library and the logic is handled in oneDNN itself. Is that true?
I seem to have fixed the other issues with the machine that made me answer this late - sorry for that -, so I should be able to respond faster.
Also should we move to another platform to continue this conversation or is this issue appropriate?

@vpirogov
Copy link
Member

Yes, I'd like to make it work, however I would like to use the computation-heavy parts on my GPU. Is it possible to off-load the work to the GPU instead of using oneDNN on the CPU?

It looks like Open WebUI is using Pytorch under the hood, so you should be able to run your model on GPU as well. Discussion on how to do that probably belongs at Open WebUI specific forum.

If I understood correctly Xbyak is a header-only library and the logic is handled in oneDNN itself. Is that true?

Xbyak is a header-only library that implements certain functionality oneDNN relies on. The floating point exception you see is thrown by an Xbyak function. My guess that AMD-specific code for hardware capabilities detection does not work correctly on your particular processor.

@herumi
Copy link
Contributor

herumi commented Feb 20, 2025 •
edited
Loading

@grialion If you make an issue on Xbyak, then I'll investigate that bug.

@kminemur
Copy link

kminemur commented Feb 21, 2025 •
edited
Loading

HI @grialion
Can you make an issue ticket on Xbyak?

I couldn't reproduce this issue on my end with Intel CPU/GPUs.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@kminemur kminemur

Labels
bug A confirmed library bug help wanted platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
6 participants
@herumi @vpirogov @kminemur @grialion @shu1chen @onednnsupporttriage