[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAGS_qxp2TjOzwBTmvr98b7Tj8xxx7bU_ydZBXVvaOkqGo1qe9Q@mail.gmail.com>
Date: Mon, 7 Nov 2022 14:27:20 -0800
From: Daniel Latypov <dlatypov@...gle.com>
To: Rae Moar <rmoar@...gle.com>
Cc: brendanhiggins@...gle.com, davidgow@...gle.com,
skhan@...uxfoundation.org, mauro.chehab@...ux.intel.com,
kunit-dev@...glegroups.com, linux-kernel@...r.kernel.org,
linux-kselftest@...r.kernel.org
Subject: Re: [PATCH v1 1/2] kunit: improve KTAP compliance of KUnit test output
On Fri, Nov 4, 2022 at 12:48 PM Rae Moar <rmoar@...gle.com> wrote:
>
> Change KUnit test output to comply with KTAP version 1 specifications
> found here: https://kernel.org/doc/html/latest/dev-tools/ktap.html.
> 1) Use "KTAP version 1" instead of "TAP version 14" as test output header
> 2) Remove '-' between test number and test name on test result lines
> 2) Add KTAP version lines to each subtest header as well
>
> Original output:
>
> TAP version 14
> 1..1
> # Subtest: kunit-test-suite
> 1..3
> ok 1 - kunit_test_1
> ok 2 - kunit_test_2
> ok 3 - kunit_test_3
> # kunit-test-suite: pass:3 fail:0 skip:0 total:3
> # Totals: pass:3 fail:0 skip:0 total:3
> ok 1 - kunit-test-suite
>
> New output:
>
> KTAP version 1
> 1..1
> # Subtest: kunit-test-suite
> KTAP version 1
> 1..3
> ok 1 kunit_test_1
> ok 2 kunit_test_2
> ok 3 kunit_test_3
> # kunit-test-suite: pass:3 fail:0 skip:0 total:3
> # Totals: pass:3 fail:0 skip:0 total:3
> ok 1 kunit-test-suite
>
> Signed-off-by: Rae Moar <rmoar@...gle.com>
Reviewed-by: Daniel Latypov <dlatypov@...gle.com>
I had been worried initially that this needed kunit_parser.py changes to work.
But it doesn't, this change can go in now with minimal side effects.
More details below:
This code treats the new "KTAP version 1" subheaders as random kernel
output, so it puts it into the `test.log`
E.g. if I change to `not ok 1 kunit_test_1`, it'll print
KTAP version 1
1..3
not ok 1 kunit_test_1
After the next patch, it just prints
not ok 1 kunit_test_1
This bit of extra output on failure is something we can live with,
esp. since it gets addressed by the next patch.
Powered by blists - more mailing lists