Am 03.12.19 um 14:34 schrieb Axel Braun:
>
>
>> Gesendet: Montag, 02. Dezember 2019 um 17:56 Uhr
>> Von: "Richard (MQ)" <[email protected]>
>> An: "Axel Braun" <[email protected]>, [email protected]
>> Betreff: Re: [opensuse-arm] Raspi 3b / Leap 15.1 / reproducible Kernel Panic
>>
>> On 01/12/2019 18:25, Axel Braun wrote:
>>> Am Sonntag, 1. Dezember 2019, 17:39:03 CET schrieb Freek de Kruijf:
>>>> Had a similar problem with a RPi1B+, which worked fine for quite some
> time,
>>>> but started to crash and rebooted several times a day.
>>>>
>>>> After replacing the power device, it behaved like before.
>>>
>>> In this case I was using the same power device for both...
>>>
>>> Cheers
>>> Axel
>>>
>>>
>> Have you checked "journalctl -f" / dmesg / "/var/log/messages" or
>> similar for warnings about under-voltage? I had this with a 1A PSU,
>> fixed by usiang a 2A one.
>
> I guess its more this what causes the problems:
>
> Dez 03 14:11:38 gnuhealth-raspi kernel: ------------[ cut here ]------------
> Dez 03 14:11:38 gnuhealth-raspi kernel: WARNING: CPU: 3 PID: 659 at ../
> drivers/gpu/drm/vc4/vc4_plane.c:795 vc4_plane_async_set_fb+0x80/0x88 [vc4]
This is not a panic.
786 void vc4_plane_async_set_fb(struct drm_plane *plane, struct
drm_framebuffer *fb)
787 {
788 struct vc4_plane_state *vc4_state =
to_vc4_plane_state(plane->state);
789 struct drm_gem_cma_object *bo = drm_fb_cma_get_gem_obj(fb, 0);
790 uint32_t addr;
791
792 /* We're skipping the address adjustment for negative origin,
793 * because this is only called on the primary plane.
794 */
795 WARN_ON_ONCE(plane->state->crtc_x < 0 || plane->state->crtc_y < 0);
796 addr = bo->paddr + fb->offsets[0];
Whatever it is, it should not be fatal.
--
Stefan Seyfried
"For a successful technology, reality must take precedence over
public relations, for nature cannot be fooled." -- Richard Feynman
--
To unsubscribe, e-mail: [email protected]
To contact the owner, e-mail: [email protected]
