gh-107888: Fix test_mmap.test_access_parameter() on macOS 14

[vstinner]

• Issue: test_mmap.test_access_parameter failing on macOS Sonoma 14.0 beta on Apple Silicon #107888

[vstinner]

!buildbot ARM64 macOS

[bedevere-bot]

🤖 New build scheduled with the buildbot fleet by @vstinner for commit 7b401f4 🤖

The command will test the builders whose names match following regular expression: ARM64 macOS

The builders matched are:

• ARM64 macOS PR

[vstinner]

cc @ned-deily

[ned-deily]

Thanks, LGTM, with a blurb should be good to go. This probably should be backported to security releases as well to avoid CI/buildbot failures.

[gpshead]

perhaps make this a self.skipTest message?

[ned-deily]

perhaps make this a self.skipTest message?

That's probably a good idea although there are plenty of other places in test_mmap that simply do a pass for similar types of failures.

[vstinner]

test_access_parameter() is a long function: 124 lines, it contains many tests. If someone wants to use skipTest() as some places, the function should be splitted into sub-functions. Otherwise, there is a risk that newly added code below will be skipped by mistake.

Today I lost 10 min trying to understand why the flush() call that I added at the end doesn't work as expected...

def display_header(use_resources: tuple[str, ...]):
    ...  # <=== a lot more code obviously

    # This makes it easier to remember what to set in your local
    # environment when trying to reproduce a sanitizer failure.
    asan = support.check_sanitizer(address=True)
    msan = support.check_sanitizer(memory=True)
    ubsan = support.check_sanitizer(ub=True)
    sanitizers = []
    if asan:
        sanitizers.append("address")
    if msan:
        sanitizers.append("memory")
    if ubsan:
        sanitizers.append("undefined behavior")
    if not sanitizers:
        return

    print(f"== sanitizers: {', '.join(sanitizers)}")
    for sanitizer, env_var in (
        (asan, "ASAN_OPTIONS"),
        (msan, "MSAN_OPTIONS"),
        (ubsan, "UBSAN_OPTIONS"),
    ):
        options= os.environ.get(env_var)
        if sanitizer and options is not None:
            print(f"== {env_var}={options!r}")

    sys.stdout.flush()  # <====== ADDED CODE ======

[vstinner]

buildbot/ARM64 macOS PR

test_gh105829_should_not_deadlock_if_wakeup_pipe_full() of test_concurrent_futures.test_deadlock decided to hang: issue gh-109917. This failure is unrelated to test_mmap (unrelated to this change).

[miss-islington]

Thanks @vstinner for the PR 🌮🎉.. I'm working now to backport this PR to: 3.11, 3.12.
🐍🍒⛏🤖

[bedevere-app]

GH-109929 is a backport of this pull request to the 3.12 branch.

[bedevere-app]

GH-109930 is a backport of this pull request to the 3.11 branch.

[vstinner]

test_gh105829_should_not_deadlock_if_wakeup_pipe_full() of test_concurrent_futures.test_deadlock decided to hang: issue #109917. This failure is unrelated to test_mmap (unrelated to this change).

I'm not curious if this bug only occurs on macOS or not. I'm asking since I couldn't reproduce it on Linux.

[vstinner]

Thanks for quick reviews @ned-deily and @gpshead!

[ned-deily]

As I hinted at in the review, this might be worthy of a blurb as others will probably run into it. What do you think?

[vstinner]

As I hinted at in the review, this might be worthy of a blurb as others will probably run into it. What do you think?

Oh. These days I'm fixing so many test failures that I stopped documenting most of them. I'm busy fixing more test failures. If you want, please go ahead and document the fix.

[miss-islington-app]

Thanks @vstinner for the PR 🌮🎉.. I'm working now to backport this PR to: 3.8.
🐍🍒⛏🤖

[miss-islington-app]

Thanks @vstinner for the PR 🌮🎉.. I'm working now to backport this PR to: 3.9.
🐍🍒⛏🤖

[miss-islington-app]

Thanks @vstinner for the PR 🌮🎉.. I'm working now to backport this PR to: 3.10.
🐍🍒⛏🤖

[bedevere-app]

GH-114183 is a backport of this pull request to the 3.8 branch.

[bedevere-app]

GH-114184 is a backport of this pull request to the 3.9 branch.

[bedevere-app]

GH-114185 is a backport of this pull request to the 3.10 branch.