Re: [PATCH] tests/avocado/reverse_debugging: Disable the ppc64 tests by default

[John Snow <jsnow@redhat.com>]

On Thu, Nov 23, 2023 at 5:53 AM Peter Maydell <peter.maydell@linaro.org> wrote:
>
> On Mon, 20 Nov 2023 at 19:19, John Snow <jsnow@redhat.com> wrote:
> >
> > On Wed, Nov 15, 2023 at 12:23 PM Daniel P. Berrangé <berrange@redhat.com> wrote:
> > > The Python  Machine() class has passed one of a pre-created socketpair
> > > FDs for the serial port chardev. The guest is trying to write to this
> > > and blocking.  Nothing in the Machine() class is reading from the
> > > other end of the serial port console.
>
> > > The Machine class doesn't know if anything will ever use the console,
> > > so as is the change is unsafe.
> > >
> > > The original goal of John's change was to guarantee we capture early
> > > boot messages as some test need that.
> > >
> > > I think we need to be able to have a flag to say whether the caller needs
> > > an "early console" facility, and only use the pre-opened FD passing for
> > > that case. Tests we need early console will have to ask for that guarantee
> > > explicitly.
> >
> > Tch. I see. Thank you for diagnosing this.
> >
> > From the machine.py perspective, you have to *opt in* to having a
> > console, so I hadn't considered that a caller would enable the console
> > and then ... not read from it. Surely that's a bug in the caller?
>
> From an Avocado test perspective, I would expect that the test case
> should have to explicitly opt *out* of "the console messages appear
> in the avocado test log, even if the test case doesn't care about them
> for the purposes of identifying when to end the test or whatever".
> The console logs are important for after-the-fact human diagnosis
> of why a test might have failed, so we should always collect them.
>
> thanks
> -- PMM
>

Understood. In that case, fixing the test would involve engaging's the
avocado suite's draining utility to ensure that the log is being
consumed and logged.

I think there's a potential here to simplify all of the
draining-and-logging code we have split across the avocado test suite,
console_socket.py and machine.py, but I can't promise that the rewrite
I've been working on will be ready quickly, so if this is still busted
(I'm still catching back up with my mail post-holidays) then we want a
quicker fix if we haven't committed one yet.

--js