gh-120974: Make asyncio swap_current_task safe in free-threaded build

[colesbury]

Use a critical section around the modifications to current_tasks so that the swapping is atomic.

• Issue: asyncio thread-safety issues in the free-threaded build #120974

[colesbury]

!buildbot nogil refleaks

[bedevere-bot]

🤖 New build scheduled with the buildbot fleet by @colesbury for commit 846f6f8 🤖

The command will test the builders whose names match following regular expression: nogil refleaks

The builders matched are:

• AMD64 Fedora Rawhide NoGIL refleaks PR
• PPC64LE Fedora Rawhide NoGIL refleaks PR
• aarch64 Fedora Rawhide NoGIL refleaks PR
• AMD64 Ubuntu NoGIL Refleaks PR

[kumaraditya303]

Should this also use a lock held variant?

[colesbury]

I don't think it makes much of a difference, but we can. I'll add _PyDict_GetItemRef_KnownHash_LockHeld.

The lock held variants are important for the SetItem calls so that we don't try to re-lock the dict. But not so much for the GetItem calls, because those have fast paths that do not lock the dict.

[miss-islington-app]

Thanks @colesbury for the PR, and @kumaraditya303 for merging it 🌮🎉.. I'm working now to backport this PR to: 3.13.
🐍🍒⛏🤖

[bedevere-app]

GH-122612 is a backport of this pull request to the 3.13 branch.

[bedevere-bot]

⚠️⚠️⚠️ Buildbot failure ⚠️⚠️⚠️

Hi! The buildbot iOS ARM64 Simulator 3.13 has failed when building commit c1efeb3.

What do you need to do:

1. Don't panic.
2. Check the buildbot page in the devguide if you don't know what the buildbots are or how they work.
3. Go to the page of the buildbot that failed (https://buildbot.python.org/#builders/1386/builds/461) and take a look at the build logs.
4. Check if the failure is related to this commit (c1efeb3) or if it is a false positive.
5. If the failure is related to this commit, please, reflect that on the issue and make a new Pull Request with a fix.

You can take a look at the buildbot page here:

https://buildbot.python.org/#builders/1386/builds/461

Failed tests:

• test_httplib

Failed subtests:

• testHTTPConnectionSourceAddress - test.test_httplib.SourceAddressTest.testHTTPConnectionSourceAddress

Summary of the results of the build (if available):

==

Click to see traceback logs

Traceback (most recent call last):
  File "/Users/buildbot/Library/Developer/XCTestDevices/0CD06645-FC15-4514-B5DC-AEF3ED24C912/data/Containers/Bundle/Application/1AE2CD3C-EBB6-482B-B920-80647D6C686A/iOSTestbed.app/python/lib/python3.13/test/test_httplib.py", line 1778, in testHTTPConnectionSourceAddress
    self.conn.connect()
    ~~~~~~~~~~~~~~~~~^^
  File "/Users/buildbot/Library/Developer/XCTestDevices/0CD06645-FC15-4514-B5DC-AEF3ED24C912/data/Containers/Bundle/Application/1AE2CD3C-EBB6-482B-B920-80647D6C686A/iOSTestbed.app/python/lib/python3.13/http/client.py", line 1001, in connect
    self.sock = self._create_connection(
                ~~~~~~~~~~~~~~~~~~~~~~~^
        (self.host,self.port), self.timeout, self.source_address)
        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/buildbot/Library/Developer/XCTestDevices/0CD06645-FC15-4514-B5DC-AEF3ED24C912/data/Containers/Bundle/Application/1AE2CD3C-EBB6-482B-B920-80647D6C686A/iOSTestbed.app/python/lib/python3.13/socket.py", line 864, in create_connection
    raise exceptions[0]
  File "/Users/buildbot/Library/Developer/XCTestDevices/0CD06645-FC15-4514-B5DC-AEF3ED24C912/data/Containers/Bundle/Application/1AE2CD3C-EBB6-482B-B920-80647D6C686A/iOSTestbed.app/python/lib/python3.13/socket.py", line 848, in create_connection
    sock.bind(source_address)
    ~~~~~~~~~^^^^^^^^^^^^^^^^
OSError: [Errno 48] Address already in use