On 01/06/2017 11:21 AM, Benjamin Gaignard wrote:
> 2017-01-06 10:58 GMT+01:00 Daniel Vetter <daniel@xxxxxxxx>:
>> On Fri, Jan 06, 2017 at 10:06:50AM +0100, Benjamin Gaignard wrote:
>>> 2017-01-06 9:22 GMT+01:00 Tomeu Vizoso <tomeu.vizoso@xxxxxxxxxxxxx>:
>>>> On 5 January 2017 at 12:12, Benjamin Gaignard
>>>> <benjamin.gaignard@xxxxxxxxxx> wrote:
>>>>> Use CRC API to retrieve the 3 crc values from hardware.
>>>>>
>>>>> Signed-off-by: Benjamin Gaignard <benjamin.gaignard@xxxxxxxxxx>
>>>>> ---
>>>>> This patch should be applied on top of drm-misc branch where Tomeu
>>>>> has change crc.lock.
>>>>>
>>>>> I think that wake_up_interruptible() could also be call at the end
>>>>> of drm_crtc_add_crc_entry() to avoid putting it in all drivers.
>>>>
>>>> Agreed, I can send such a patch if you don't have time.
>>>
>>> I just finish to test the patches I will send them asap
>>>
>>>>
>>>> Btw, do any tests from iGT that make use of CRCs pass with this? If
>>>> so, would be good to note it in the commit message.
>>>
>>> I don't run IGT just modetest and cat on crtc-0/crc/data
>>
>> Would be really good if you can give igt a spin on this ...
>
> What is the status of IGT on ARM platforms ?
> Last time (~6 months ago) I tested it, I had to include intel drm lib
> and it just allow me
> to check drm version. Does that have change ?
Yup, this and other issues have been fixed. Have been running
tests/kms_pipe_crc_basic on a RK3288 Chromebook recently without any
problems.
Make sure though to rebase onto a recent drm-misc because of
e3d19d55676b ("drm: crc: Wait for a frame before returning from open()").
Regards,
Tomeu
_______________________________________________
dri-devel mailing list
dri-devel@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/dri-devel
