Hi Mexime,
On 29.10.2021 10:05, Maxime Ripard wrote:
> On Fri, Oct 29, 2021 at 08:23:45AM +0200, Marek Szyprowski wrote:
>> On 25.10.2021 17:15, Maxime Ripard wrote:
>>> In order to avoid any probe ordering issue, the best practice is to move
>>> the secondary MIPI-DSI device registration and attachment to the
>>> MIPI-DSI host at probe time. Let's do this.
>>>
>>> Acked-by: Sam Ravnborg <sam@xxxxxxxxxxxx>
>>> Tested-by: John Stultz <john.stultz@xxxxxxxxxx>
>>> Signed-off-by: Maxime Ripard <maxime@xxxxxxxxxx>
>> This patch landed in linux-next as commit 864c49a31d6b ("drm/bridge:
>> adv7511: Register and attach our DSI device at probe"). Sadly it causes
>> endless probe-fail-defer loop on DragonBoard 410c board
>> (arch/arm64/boot/dts/qcom/apq8016-sbc.dts):
> I'm sorry to hear that (but would have been surprised if it didn't occur)
>
> This is supposed to be fixed by 8f59ee9a570c ("drm/msm/dsi: Adjust probe
> order"). Do you have that patch applied?
Yes, I did my test directly on linux next-20211028, which also contains
it. What might be important in my case, my DragonBoard 410c doesn't have
any display attached.
I've also noticed the following error during boot:
[ 23.847651] msm_mdp 1a01000.mdp: Adding to iommu group 3
[ 23.866044] msm_mdp 1a01000.mdp: No interconnect support may cause
display underflows!
[ 23.957949] irq: no irq domain found for mdss@1a00000 !
[ 23.958014] msm_dsi 1a98000.dsi: failed to request IRQ0: -22
[ 23.962229] msm_dsi: probe of 1a98000.dsi failed with error -22
The above errors appeared in next-20211028 for the first time. I assume
that they are related.
Best regards
--
Marek Szyprowski, PhD
Samsung R&D Institute Poland
