CONTRIBUTING
author Martin von Zweigbergk <martinvonz@google.com>
Fri, 14 Aug 2020 20:45:49 -0700
changeset 45409 7d24201b6447
parent 30084 a492610a2fc1
permissions -rw-r--r--
worker: don't expose readinto() on _blockingreader since pickle is picky The `pickle` module expects the input to be buffered and a whole object to be available when `pickle.load()` is called, which is not necessarily true when we send data from workers back to the parent process (i.e., it seems like a bad assumption for the `pickle` module to make). We added a workaround for that in https://phab.mercurial-scm.org/D8076, which made `read()` continue until all the requested bytes have been read. As we found out at work after a lot of investigation (I've spent the last two days on this), the native version of `pickle.load()` has started calling `readinto()` on the input since Python 3.8. That started being called in https://github.com/python/cpython/commit/91f4380cedbae32b49adbea2518014a5624c6523 (and only by the C version of `pickle.load()`)). Before that, it was only `read()` and `readline()` that were called. The problem with that was that `readinto()` on our `_blockingreader` was simply delegating to the underlying, *unbuffered* object. The symptom we saw was that `hg fix` started failing sometimes on Python 3.8 on Mac. It failed very relyable in some cases. I still haven't figured out under what circumstances it fails and I've been unable to reproduce it in test cases (I've tried writing larger amounts of data, using different numbers of workers, and making the formatters sleep). I have, however, been able to reproduce it 3-4 times on Linux, but then it stopped reproducing on the following few hundred attempts. To fix the problem, we can simply remove the implementation of `readinto()`, since the unpickler will then fall back to calling `read()`. The fallback was added a bit later, in https://github.com/python/cpython/commit/b19f7ecfa3adc6ba1544225317b9473649815b38. However, that commit also added checking that what `read()` returns is a `bytes`, so we also need to convert the `bytearray` we use into that. I was able to add a test for that failure at least. Differential Revision: https://phab.mercurial-scm.org/D8928

Our full contribution guidelines are in our wiki, please see:

https://www.mercurial-scm.org/wiki/ContributingChanges

If you just want a checklist to follow, you can go straight to

https://www.mercurial-scm.org/wiki/ContributingChanges#Submission_checklist

If you can't run the entire testsuite for some reason (it can be
difficult on Windows), please at least run `contrib/check-code.py` on
any files you've modified and run `python contrib/check-commit` on any
commits you've made (for example, `python contrib/check-commit
273ce12ad8f1` will report some style violations on a very old commit).