[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CABVgOSndkMOd0U+p=fMy0q-KdN29XERheY=3-3+dgVoeWf9m5g@mail.gmail.com>
Date: Thu, 7 May 2020 11:08:10 +0800
From: David Gow <davidgow@...gle.com>
To: Anders Roxell <anders.roxell@...aro.org>
Cc: Brendan Higgins <brendanhiggins@...gle.com>,
John Johansen <john.johansen@...onical.com>,
James Morris <jmorris@...ei.org>,
"Serge E. Hallyn" <serge@...lyn.com>,
"Theodore Ts'o" <tytso@....edu>,
Andreas Dilger <adilger.kernel@...ger.ca>,
Greg Kroah-Hartman <gregkh@...uxfoundation.org>,
Andrew Morton <akpm@...ux-foundation.org>,
"open list:KERNEL SELFTEST FRAMEWORK"
<linux-kselftest@...r.kernel.org>,
KUnit Development <kunit-dev@...glegroups.com>,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
linux-ext4@...r.kernel.org,
linux-security-module <linux-security-module@...r.kernel.org>,
Marco Elver <elver@...gle.com>
Subject: Re: [PATCH v2 1/6] kunit: Kconfig: enable a KUNIT_RUN_ALL fragment
On Wed, May 6, 2020 at 6:33 PM Anders Roxell <anders.roxell@...aro.org> wrote:
>
> Hi David,
>
> Thank you for the review.
>
> On Wed, 6 May 2020 at 07:08, David Gow <davidgow@...gle.com> wrote:
> >
> > On Tue, May 5, 2020 at 6:27 PM Anders Roxell <anders.roxell@...aro.org> wrote:
> > >
> > > Make it easier to enable all KUnit fragments. This is needed for kernel
> > > test-systems, so its easy to get all KUnit tests enabled and if new gets
> > > added they will be enabled as well. Fragments that has to be builtin
> > > will be missed if CONFIG_KUNIT_RUN_ALL is set as a module.
> > >
> > > Signed-off-by: Anders Roxell <anders.roxell@...aro.org>
> > > ---
> > > lib/kunit/Kconfig | 6 ++++++
> > > 1 file changed, 6 insertions(+)
> > >
> > > diff --git a/lib/kunit/Kconfig b/lib/kunit/Kconfig
> > > index 95d12e3d6d95..537f37bc8400 100644
> > > --- a/lib/kunit/Kconfig
> > > +++ b/lib/kunit/Kconfig
> > > @@ -41,4 +41,10 @@ config KUNIT_EXAMPLE_TEST
> > > is intended for curious hackers who would like to understand how to
> > > use KUnit for kernel development.
> > >
> > > +config KUNIT_RUN_ALL
> > > + tristate "KUnit run all test"
> >
> > I'm not 100% sure about this name and description. It only actually
> > "runs" the tests if they're builtin (as modules, they'll only run when
> > loaded).
> >
> > Would KUNIT_BUILD_ALL or KUNIT_ALL_TESTS
>
> I would like to go with KUNIT_ALL_TESTS if no one objects.
>
Personally, I'm fine with that.
Brendan, thoughts?
> > or similar be better?
> >
> > It also, as mentioned, only really runs all available (i.e., with
> > dependencies met in the current .config) tests (as distinct from the
> > --alltests flag to kunit.py, which builds UML with allyesconfig), it
> > might be good to add this to the description or help.
> >
> > Something like "Enable all KUnit tests" or "Enable all available KUnit
> > tests" (or even something about "all KUnit tests with satisfied
> > dependencies") might be clearer.
>
> I like "All KUnit tests with satisfied dependencies".
>
> >
> > > + help
> > > + Enables all KUnit tests, if they can be enabled.
> > > + That depends on if KUnit is enabled as a module or builtin.
> > > +
> > I like the first line here, but the second one could use a bit more
> > explanation. Maybe copy some of the boilerplate text from other tests,
> > e.g.:
> >
> > KUnit tests run during boot and output the results to the debug log
> > in TAP format (http://testanything.org/). Only useful for kernel devs
> > running the KUnit test harness, and not intended for inclusion into a
> > production build.
> >
> > For more information on KUnit and unit tests in general please refer
> > to the KUnit documentation in Documentation/dev-tools/kunit/.
> >
> > If unsure, say N.
>
> Makes much more sense, thanks.
>
> >
> > > endif # KUNIT
> > > --
> > > 2.20.1
> > >
> >
> > Otherwise, this is looking good. I've done some quick testing, and it
> > all seems to work for me. As long as it's clear what the difference
> > between this and other ways of running "all tests" (like the
> > --alltests kunit.py option),
>
> Do you think it should be made clearer in some way?
>
I think the changes above should do it.
-- David
Powered by blists - more mailing lists