Hi,

Victor Stinner wrote:

Hi,

2014-03-28 9:31 GMT+01:00 Andrey Ponomarenko aponomarenko@rosalab.ru:

...

The libpython library has been added to the ABI tracker: http://upstream-tracker.org/versions/python.html

The page lists library versions and changes in API/ABI.

Nice!

By the way, would it be possible to add a second page for the stable ABI of Python?

Victor

The public libpython API changes will be tracked here: http://upstream-tracker.org/versions/python_public_api.html

For now I've excluded only symbols starting with an underscore. What other symbols should be excluded?

Thanks.

On 1 Apr 2014 01:38, "Victor Stinner" victor.stinner@gmail.com wrote:

2014-03-31 13:38 GMT+02:00 Andrey Ponomarenko aponomarenko@rosalab.ru:

...

The public libpython API changes will be tracked here: http://upstream-tracker.org/versions/python_public_api.html

For now I've excluded only symbols starting with an underscore. What

other

...

symbols should be excluded?

It's not a matter of underscore. You should define Py_LIMITED_API to 0x03020000 to test the stable ABI of Python 3.2.

http://docs.python.org/dev/c-api/stable.html

Well, we have more than one ABI, with different guarantees. The "no leading underscore" one we promise not to change in maintenance releases, but we only preserve *API* compatibility in feature releases (mostly due to structs changing size).

The "stable ABI" (aka Py_LIMITED_API) is the one where we promise to hide all the memory layout details and treat it as "additive only" so that binaries built with previous releases keep working. That should never break ABI compatibility, and only get new additions if the macro definition is bumped up to match the newer release.

Cheers, Nick.

P.S. I understand it was Anatoly that put the process in motion to get this set up. Thanks for doing that Anatoly, it's a genuinely good idea.

Victor _______________________________________________ Python-Dev mailing list Python-Dev@python.org https://mail.python.org/mailman/listinfo/python-dev Unsubscribe:

https://mail.python.org/mailman/options/python-dev/ncoghlan%40gmail.com

On 1 April 2014 21:23, Andrey Ponomarenko aponomarenko@rosalab.ru wrote:

Nick Coghlan wrote:

...

On 1 Apr 2014 01:38, "Victor Stinner" victor.stinner@gmail.com wrote:

...

2014-03-31 13:38 GMT+02:00 Andrey Ponomarenko aponomarenko@rosalab.ru:

...

The public libpython API changes will be tracked here: http://upstream-tracker.org/versions/python_public_api.html

For now I've excluded only symbols starting with an underscore. What

other

...

...

symbols should be excluded?

It's not a matter of underscore. You should define Py_LIMITED_API to 0x03020000 to test the stable ABI of Python 3.2.

http://docs.python.org/dev/c-api/stable.html

Well, we have more than one ABI, with different guarantees. The "no leading underscore" one we promise not to change in maintenance releases, but we only preserve *API* compatibility in feature releases (mostly due to structs changing size).

The "stable ABI" (aka Py_LIMITED_API) is the one where we promise to hide all the memory layout details and treat it as "additive only" so that binaries built with previous releases keep working. That should never break ABI compatibility, and only get new additions if the macro definition is bumped up to match the newer release.

Cheers, Nick.

The stable libpython ABI with Py_LIMITED_API=0x03020000 will be tracked at http://upstream-tracker.org/versions/python_stable_api.html

Thanks! The "leading underscore means private" convention is also applicable here (that's a general guideline for Python related APIs).

Interesting to see the UCS2 removal there for 3.3. That's a genuine removal from the public ABI as part of PEP 393. I guess the reason nobody complained is because most 3.2 Linux builds used the UCS4 ABI instead, and the stable ABI hadn't seen broad adoption on Windows in the 3.2->3.3 time frame.

Regarding the warnings for this one - is there a way for the checker to warn if data structures are exposed directly, rather than as opaque types? It's fine if there isn't, it would just be cool if there was - one of the premises of the stable ABI is that it *doesn't* expose the type definitions directly to consuming code, just the pointers to them (hence allowing the struct size to change without actually breaking compatibility with the defined ABI).

Regardless, this service already shows we've made some mistakes with the stable ABI in previous releases - it is indicating there are new symbols in the stable ABI for 3.3 and 3.4 that aren't properly guarded with version constraints. That means it is currently possible to set Py_LIMITED_API=0x03020000 and get something that won't actually run properly on 3.2.

Georg, Larry, Benjamin - should checking these be added to PEP 101, so we don't get the same thing happening for 3.5?

I also added source-compatibility reports to the "public" API tracker: http://upstream-tracker.org/versions/python_public_api.html

Thanks again for setting these up!

Cheers, Nick.

On 2 Apr 2014 00:54, "Benjamin Peterson" benjamin@python.org wrote:

On Tue, Apr 1, 2014, at 4:45, Nick Coghlan wrote:

...

Georg, Larry, Benjamin - should checking these be added to PEP 101, so we don't get the same thing happening for 3.5?

I would like it if we could put this in the testsuite somehow.

Perhaps an independent automated daily or weekly check like Antoine's daily refleak hunter would be better?

Cheers, Nick.

Nick Coghlan wrote:

Regarding the warnings for this one - is there a way for the checker to warn if data structures are exposed directly, rather than as opaque types? It's fine if there isn't, it would just be cool if there was - one of the premises of the stable ABI is that it *doesn't* expose the type definitions directly to consuming code, just the pointers to them (hence allowing the struct size to change without actually breaking compatibility with the defined ABI).

The ABI of the library can be dumped by the abi-compliance-checker basic tool to a text file in the human readable format, so anyone can analyse it in order to find problems of any kind.

Homepage of the tool: https://github.com/lvc/abi-compliance-checker

To dump the libpython ABI type:

$> abi-compliance-checker -l libpython -dump descriptor.xml

The descriptor.xml input file ({RELPATH} - path to the python install tree, i.e. installation "prefix"):

<version> 3.4.0 </version>

<headers> {RELPATH}/include </headers>

<libs> {RELPATH}/lib/libpython3.4m.so.1.0 </libs>

<skip_including> ast.h Python-ast.h asdl.h pyexpat.h pymactoolbox.h </skip_including>

<gcc_options> -DPy_LIMITED_API=0x03020000 </gcc_options>

I've already created the sample dump of the libpython-3.4.0 stable ABI here: http://upstream-tracker.org/downloads/libpython_3.4.0.abi.tar.gz

In order to analyse data types in the ABI please read 'TypeInfo' section of the dump.

I see several structures with exposed definition in the stable ABI v3.4.0:

struct PyStructSequence_Desc struct grammar struct PyStructSequence_Field struct _node struct cchar_t struct PyType_Spec struct PyType_Slot struct dfa struct labellist struct PyMethodDef struct _Py_Identifier struct state struct PyVarObject struct arc struct label struct PyModuleDef struct PyModuleDef_Base struct PyMemberDef struct PyGetSetDef struct _object struct PyCursesWindowObject

Thanks.

Nick Coghlan wrote:

On 3 April 2014 00:18, Andrey Ponomarenko aponomarenko@rosalab.ru wrote:

...

Nick Coghlan wrote:

...

Regarding the warnings for this one - is there a way for the checker to warn if data structures are exposed directly, rather than as opaque types? It's fine if there isn't, it would just be cool if there was - one of the premises of the stable ABI is that it *doesn't* expose the type definitions directly to consuming code, just the pointers to them (hence allowing the struct size to change without actually breaking compatibility with the defined ABI).

The ABI of the library can be dumped by the abi-compliance-checker basic tool to a text file in the human readable format, so anyone can analyse it in order to find problems of any kind.

Thanks! I've now filed http://bugs.python.org/issue21142 as an RFE - I'll definitely be looking into setting up a regular run of this to help us avoid any new issues with managing the evolution of the stable ABI (and the public ABI in general).

I wrote some instructions on how to setup local ABI checker in the comment to issue 21142.

The main idea is to create ABI dump of the reference libpython version (say, 3.2) and compare it with the ABI dump of the current version. The ABI dump of the reference version can be stored somewhere in the source tree.

Thanks.