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 fails on 3.14 no-GIL #613

Open
clin1234 opened this issue Jun 20, 2024 · 5 comments
Open

Build fails on 3.14 no-GIL #613

clin1234 opened this issue Jun 20, 2024 · 5 comments

Comments

@clin1234
Copy link

$ make
cython msgpack/_cmsgpack.pyx
python setup.py build_ext -i -f
running build_ext
building 'msgpack._cmsgpack' extension
creating build
creating build/temp.linux-x86_64-cpython-314
creating build/temp.linux-x86_64-cpython-314/msgpack
x86_64-linux-gnu-gcc -fno-strict-overflow -Wsign-compare -DNDEBUG -g -O2 -Wall -g -fno-omit-frame-pointer -mno-omit-leaf-frame-pointer -flto=auto -ffat-lto-objects -fstack-protector-strong -fstack-clash-protection -Wformat -Werror=format-security -fcf-protection -fPIC -I. -I/tmp/venv/include -I/usr/include/python3.14t -c msgpack/_cmsgpack.c -o build/temp.linux-x86_64-cpython-314/msgpack/_cmsgpack.o
msgpack/_cmsgpack.c:2343:80: error: unknown type name ‘__pyx_vectorcallfunc’; did you mean ‘vectorcallfunc’?
 2343 | ject *__Pyx_PyVectorcall_FastCallDict(PyObject *func, __pyx_vectorcallfunc vc, PyObject *const *args, size_t nargs, PyObject *kw);
      |                                                       ^~~~~~~~~~~~~~~~~~~~
      |                                                       vectorcallfunc
msgpack/_cmsgpack.c: In function ‘__pyx_pf_7msgpack_9_cmsgpack_6Packer_2__dealloc__’:
msgpack/_cmsgpack.c:6941:3: warning: ‘Py_OptimizeFlag’ is deprecated [-Wdeprecated-declarations]
 6941 |   if (unlikely(__pyx_assertions_enabled())) {
      |   ^~
In file included from /usr/include/python3.14t/Python.h:72,
                 from msgpack/_cmsgpack.c:16:
/usr/include/python3.14t/cpython/pydebug.h:13:37: note: declared here
   13 | Py_DEPRECATED(3.12) PyAPI_DATA(int) Py_OptimizeFlag;
      |                                     ^~~~~~~~~~~~~~~
msgpack/_cmsgpack.c: At top level:
msgpack/_cmsgpack.c:21382:69: error: unknown type name ‘__pyx_vectorcallfunc’; did you mean ‘vectorcallfunc’?
21382 | t *__Pyx_PyVectorcall_FastCallDict_kw(PyObject *func, __pyx_vectorcallfunc vc, PyObject *const *args, size_t nargs, PyObject *kw)
      |                                                       ^~~~~~~~~~~~~~~~~~~~
      |                                                       vectorcallfunc
msgpack/_cmsgpack.c:21427:80: error: unknown type name ‘__pyx_vectorcallfunc’; did you mean ‘vectorcallfunc’?
21427 | ject *__Pyx_PyVectorcall_FastCallDict(PyObject *func, __pyx_vectorcallfunc vc, PyObject *const *args, size_t nargs, PyObject *kw)
      |                                                       ^~~~~~~~~~~~~~~~~~~~
      |                                                       vectorcallfunc
msgpack/_cmsgpack.c: In function ‘__Pyx_CyFunction_CallAsMethod’:
msgpack/_cmsgpack.c:22116:6: error: unknown type name ‘__pyx_vectorcallfunc’; did you mean ‘vectorcallfunc’?
22116 |      __pyx_vectorcallfunc vc = __Pyx_CyFunction_func_vectorcall(cyfunc);
      |      ^~~~~~~~~~~~~~~~~~~~
      |      vectorcallfunc
msgpack/_cmsgpack.c:2433:45: warning: initialization of ‘int’ from ‘vectorcallfunc’ {aka ‘struct _object * (*)(struct _object *, struct _object * const*, long unsigned int,  struct _object *)’} makes integer from pointer without a cast [-Wint-conversion]
 2433 | #define __Pyx_CyFunction_func_vectorcall(f) (((PyCFunctionObject*)f)->vectorcall)
      |                                             ^
msgpack/_cmsgpack.c:22116:32: note: in expansion of macro ‘__Pyx_CyFunction_func_vectorcall’
22116 |      __pyx_vectorcallfunc vc = __Pyx_CyFunction_func_vectorcall(cyfunc);
      |                                ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
msgpack/_cmsgpack.c:22119:16: warning: implicit declaration of function ‘__Pyx_PyVectorcall_FastCallDict’; did you mean ‘__Pyx_PyObject_FastCallDict’? [-Wimplicit-function-declaration]
22119 |         return __Pyx_PyVectorcall_FastCallDict(func, vc, &PyTuple_GET_ITEM(args, 0), (size_t)PyTuple_GET_SIZE(args), kw);
      |                ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
      |                __Pyx_PyObject_FastCallDict
msgpack/_cmsgpack.c:22119:16: warning: returning ‘int’ from a function with return type ‘PyObject *’ {aka ‘struct _object *’} makes pointer from integer without a cast [-Wint-conversion]
22119 |         return __Pyx_PyVectorcall_FastCallDict(func, vc, &PyTuple_GET_ITEM(args, 0), (size_t)PyTuple_GET_SIZE(args), kw);
      |                ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
error: command '/usr/bin/x86_64-linux-gnu-gcc' failed with exit code 1
make: *** [Makefile:5: all] Error 1
@methane
Copy link
Member

methane commented Jun 20, 2024

no plan until cython support no-gil build.

@rgommers
Copy link

rgommers commented Aug 2, 2024

Note that while Cython support for free-threaded Python isn't fully complete yet, it works well enough in master now (3.1.0-dev, not in 3.0.x). A nightly wheel for Cython can be installed as documented on https://py-free-threading.github.io/ci/. It's used by projects with a large amount of Cython code like SciPy and scikit-learn for their free-threaded nightly wheels.

This issue should automatically go away, see https://py-free-threading.github.io/debugging/#cython-compilation-errors-unknown-type-name-__pyx_vectorcallfunc

@tacaswell
Copy link
Contributor

The sdist for 1.1.0rc1 on pypi contains a couple files generated by cython which makes the install fail even if you have a version of cython that will work. Is it possible to make cython a build-time dependency and regenerate the output as needed?

@methane
Copy link
Member

methane commented Aug 19, 2024

See #539.

@methane
Copy link
Member

methane commented Aug 19, 2024

no-GIL Python is experimental. Do not ask support for it.
Just wait until stable Cython release, or manually build instead of pip install.
That is what you should do it by yourself when using experimental software.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging a pull request may close this issue.

4 participants