[PATCH 0/2] drm: revert some framebuffer API tests

Maxime Ripard mripard at kernel.org
Thu Sep 26 07:05:14 UTC 2024


On Wed, Sep 25, 2024 at 09:04:39AM GMT, Guenter Roeck wrote:
> On 9/25/24 06:05, Maxime Ripard wrote:
> [ ... ]
> 
> > > We've had similar discussions in the past around unit tests that wasted
> > > too much cpu time with randomized combinatorial testing, and we've thrown
> > > those out too from drm. Test hygiene matters.
> > 
> > We had that discussion because those tests could run for up to around a
> > minute on certain platforms. It's not the same issue at all?
> > 
> 
> I added "kunit.stats_enabled=2 kunit.filter=speed>slow" to the kernel command
> line in my tests to avoid long-running tests. That works as long as slow tests
> are marked accordingly using KUNIT_CASE_SLOW(), KUNIT_SPEED_SLOW, or
> KUNIT_SPEED_VERY_SLOW.

Yeah, and I've added a warning some time if a test runs for too long but
isn't flagged as slow.

Maxime
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 273 bytes
Desc: not available
URL: <https://lists.freedesktop.org/archives/dri-devel/attachments/20240926/95611e10/attachment.sig>


More information about the dri-devel mailing list