Skip to content

gh-141926: Do not unset RUNSHARED when cross-compiling#141958

Merged
FFY00 merged 1 commit intopython:mainfrom
zanieb:zb/runshared-unset
Dec 5, 2025
Merged

gh-141926: Do not unset RUNSHARED when cross-compiling#141958
FFY00 merged 1 commit intopython:mainfrom
zanieb:zb/runshared-unset

Conversation

@zanieb
Copy link
Copy Markdown
Contributor

@zanieb zanieb commented Nov 25, 2025 •
edited
Loading

An alternative to #141927 just removing what appears to be legacy code instead of providing a way to opt-in to different behavior.

Closes #141926
Closes astral-sh/python-build-standalone#864

@bedevere-app bedevere-app bot mentioned this pull request Nov 25, 2025
Closed
@zanieb zanieb force-pushed the zb/runshared-unset branch from a3272ed to 9152608 Compare November 25, 2025 19:21
@zanieb zanieb marked this pull request as ready for review November 25, 2025 19:49
@zanieb zanieb mentioned this pull request Nov 26, 2025
Closed
@zanieb zanieb requested a review from brettcannon December 1, 2025 20:20
FFY00
FFY00 approved these changes Dec 4, 2025
View reviewed changes
Copy link
Copy Markdown
Member

@FFY00 FFY00 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think RUNSHARED should only need to be cleared to run the built interpreter on edge cases. Since nowadays it's more likely we need to keep it, I think it makes sense to keep it by default, and add some logic to adjust it when needed.

@FFY00 FFY00 added the 🔨 test-with-buildbots Test PR w/ buildbots; report in status section label Dec 4, 2025
@bedevere-bot
Copy link
Copy Markdown

bedevere-bot commented Dec 4, 2025

🤖 New build scheduled with the buildbot fleet by @FFY00 for commit 9152608 🤖

Results will be shown at:

https://buildbot.python.org/all/#/grid?branch=refs%2Fpull%2F141958%2Fmerge

If you want to schedule another build, you need to add the 🔨 test-with-buildbots label again.

@bedevere-bot bedevere-bot removed the 🔨 test-with-buildbots Test PR w/ buildbots; report in status section label Dec 4, 2025
@emmatyping
Copy link
Copy Markdown
Member

emmatyping commented Dec 4, 2025

I think RUNSHARED should only need to be cleared to run the built interpreter on edge cases.

Wouldn't cross-compiling from say x86_64 to aarch64 on Linux where you haven't configured binfmt-misc to handle ARM binaries break? I know that is not uncommon to build software for SBCs like a Raspberry Pi with this setup.

@zanieb
Copy link
Copy Markdown
Contributor Author

zanieb commented Dec 4, 2025

Wouldn't cross-compiling from say x86_64 to aarch64 on Linux where you haven't configured binfmt-misc to handle ARM binaries break?

Break at what step though? It looks like, e.g., PGO still runs when cross-compiling, so clearing RUNSHARED won't save you and this shouldn't cause a regression.

@emmatyping
Copy link
Copy Markdown
Member

emmatyping commented Dec 4, 2025

Break at what step though? It looks like, e.g., PGO still runs when cross-compiling, so clearing RUNSHARED won't save you and this shouldn't cause a regression.

Yeah thinking this over a bit more, I think you're right that any use-case where this could regress is either completely or subtly broken anyway.

@zanieb
Copy link
Copy Markdown
Contributor Author

zanieb commented Dec 4, 2025 •
edited
Loading

I presume these test failures on the refleaks bots are flakes? e.g.:

======================================================================
ERROR: test_interrupt (test.test_multiprocessing_fork.test_processes.WithProcessesTestProcess.test_interrupt)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "/home/buildbot/buildarea/pull_request.cstratak-fedora-stable-s390x.refleak/build/Lib/contextlib.py", line 85, in inner
    return func(*args, **kwds)
  File "/home/buildbot/buildarea/pull_request.cstratak-fedora-stable-s390x.refleak/build/Lib/test/_test_multiprocessing.py", line 597, in test_interrupt
    exitcode = self._kill_process(multiprocessing.Process.interrupt)
  File "/home/buildbot/buildarea/pull_request.cstratak-fedora-stable-s390x.refleak/build/Lib/contextlib.py", line 85, in inner
    return func(*args, **kwds)
  File "/home/buildbot/buildarea/pull_request.cstratak-fedora-stable-s390x.refleak/build/Lib/test/_test_multiprocessing.py", line 578, in _kill_process
    self.assertEqual(join(), None)
                     ~~~~^^
  File "/home/buildbot/buildarea/pull_request.cstratak-fedora-stable-s390x.refleak/build/Lib/test/_test_multiprocessing.py", line 251, in __call__
    return self.func(*args, **kwds)
           ~~~~~~~~~^^^^^^^^^^^^^^^
  File "/home/buildbot/buildarea/pull_request.cstratak-fedora-stable-s390x.refleak/build/Lib/multiprocessing/process.py", line 156, in join
    res = self._popen.wait(timeout)
  File "/home/buildbot/buildarea/pull_request.cstratak-fedora-stable-s390x.refleak/build/Lib/multiprocessing/popen_fork.py", line 44, in wait
    return self.poll(os.WNOHANG if timeout == 0.0 else 0)
           ~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/buildbot/buildarea/pull_request.cstratak-fedora-stable-s390x.refleak/build/Lib/multiprocessing/popen_fork.py", line 28, in poll
    pid, sts = os.waitpid(self.pid, flag)
               ~~~~~~~~~~^^^^^^^^^^^^^^^^
  File "/home/buildbot/buildarea/pull_request.cstratak-fedora-stable-s390x.refleak/build/Lib/test/_test_multiprocessing.py", line 574, in handler
    raise RuntimeError('join took too long: %s' % p)
RuntimeError: join took too long: <Process name='Process-1708' pid=2226706 parent=2212089 started daemon>

----------------------------------------------------------------------
Ran 162 tests in 33.825s

FAILED (errors=1, skipped=7)
test test.test_multiprocessing_fork.test_processes failed

@FFY00
Copy link
Copy Markdown
Member

FFY00 commented Dec 5, 2025

Yeah, the buildbot failures are unrelated, we are okay.

@FFY00 FFY00 merged commit 128d316 into python:main Dec 5, 2025
103 of 105 checks passed
StanFromIreland pushed a commit to StanFromIreland/cpython that referenced this pull request Dec 6, 2025
@zanieb @StanFromIreland
pythongh-141926: Do not unset RUNSHARED when cross-compiling (pytho…
1c39079 
…n#141958)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@emmatyping emmatyping emmatyping approved these changes

@FFY00 FFY00 FFY00 approved these changes

@erlend-aasland erlend-aasland Awaiting requested review from erlend-aasland erlend-aasland is a code owner

@corona10 corona10 Awaiting requested review from corona10 corona10 is a code owner

@AA-Turner AA-Turner Awaiting requested review from AA-Turner AA-Turner is a code owner

@brettcannon brettcannon Awaiting requested review from brettcannon

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

Support PGO in cross-compiles when transparent emulation is available Upstream patch patch-dont-clear-runshared-14.patch

5 participants

@zanieb @bedevere-bot @emmatyping @FFY00 @nazeerali4325-commits