Hi,
On 08/11/2023 14:45, Alexander Stein wrote:
> Hi Tomi,
>
> Am Mittwoch, 8. November 2023, 12:27:21 CET schrieb Tomi Valkeinen:
>> These two patches are needed to make tc358767 work in the
>> DRM_BRIDGE_ATTACH_NO_CONNECTOR case, at least when using a DP connector.
>>
>> I have tested this with TI AM654 EVM with a tc358767 add-on card
>> connected to a DP monitor.
>
> Just a question regarding the usage of this DSI-DP bridge.
> What is the state of the DSI lanes after the DSI host has been initialized,
> but before calling atomic_pre_enable? AFAIK this bridge requires LP-11 on DSI
> at any time for accessing the AUX channel.
We haven't received any test reports for the DSI-DP case... I was
looking at the datasheet, and I wonder, why do you say the bridge
requires DSI to be up for the AUX transactions?
Tomi
> Best regards,
> Alexander
>
>> Signed-off-by: Tomi Valkeinen <[email protected]>
>> ---
>> Changes in v2:
>> - Update the format negotiation patch as discussed in
>> https://lore.kernel.org/all/[email protected]/ -
>> Link to v1:
>> https://lore.kernel.org/r/20231031-tc358767-v1-0-392081ad9f4b@ideasonboard.
>> com
>>
>> ---
>> Aradhya Bhatia (1):
>> drm/bridge: tc358767: Add format negotiation hooks for DPI/DSI to
>> (e)DP
>>
>> Tomi Valkeinen (1):
>> drm/bridge: tc358767: Fix link properties discovery
>>
>> drivers/gpu/drm/bridge/tc358767.c | 32 ++++++++++++++++++++++++++++++++
>> 1 file changed, 32 insertions(+)
>> ---
>> base-commit: 9d7c8c066916f231ca0ed4e4fce6c4b58ca3e451
>> change-id: 20231031-tc358767-58e3ebdf95f0
>>
>> Best regards,
>
>
On 06/12/23 17:41, Tomi Valkeinen wrote:
> Hi,
>
> On 08/11/2023 14:45, Alexander Stein wrote:
>> Hi Tomi,
>>
>> Am Mittwoch, 8. November 2023, 12:27:21 CET schrieb Tomi Valkeinen:
>>> These two patches are needed to make tc358767 work in the
>>> DRM_BRIDGE_ATTACH_NO_CONNECTOR case, at least when using a DP connector.
>>>
>>> I have tested this with TI AM654 EVM with a tc358767 add-on card
>>> connected to a DP monitor.
>>
>> Just a question regarding the usage of this DSI-DP bridge.
>> What is the state of the DSI lanes after the DSI host has been
>> initialized,
>> but before calling atomic_pre_enable? AFAIK this bridge requires LP-11
>> on DSI
>> at any time for accessing the AUX channel.
+ Marek
Marek, Alexander,
A quick grep tells me that you have added devicetree for tc358767 in DSI
to (e)DP mode on other platforms. Could you please test these patches
and report if you find any issue?
Regards
Aradhya
>
> We haven't received any test reports for the DSI-DP case... I was
> looking at the datasheet, and I wonder, why do you say the bridge
> requires DSI to be up for the AUX transactions?
>
> Tomi
>
>> Best regards,
>> Alexander
>>
>>> Signed-off-by: Tomi Valkeinen <[email protected]>
>>> ---
>>> Changes in v2:
>>> - Update the format negotiation patch as discussed in
>>> https://lore.kernel.org/all/[email protected]/ -
>>> Link to v1:
>>> https://lore.kernel.org/r/20231031-tc358767-v1-0-392081ad9f4b@ideasonboard.
>>> com
>>>
>>> ---
>>> Aradhya Bhatia (1):
>>> drm/bridge: tc358767: Add format negotiation hooks for DPI/DSI to
>>> (e)DP
>>>
>>> Tomi Valkeinen (1):
>>> drm/bridge: tc358767: Fix link properties discovery
>>>
>>> drivers/gpu/drm/bridge/tc358767.c | 32
>>> ++++++++++++++++++++++++++++++++
>>> 1 file changed, 32 insertions(+)
>>> ---
>>> base-commit: 9d7c8c066916f231ca0ed4e4fce6c4b58ca3e451
>>> change-id: 20231031-tc358767-58e3ebdf95f0
>>>
>>> Best regards,
>>
>>
>
Hi Tomi,
Am Mittwoch, 6. Dezember 2023, 13:11:59 CET schrieb Tomi Valkeinen:
> Hi,
>
> On 08/11/2023 14:45, Alexander Stein wrote:
> > Hi Tomi,
> >
> > Am Mittwoch, 8. November 2023, 12:27:21 CET schrieb Tomi Valkeinen:
> >> These two patches are needed to make tc358767 work in the
> >> DRM_BRIDGE_ATTACH_NO_CONNECTOR case, at least when using a DP connector.
> >>
> >> I have tested this with TI AM654 EVM with a tc358767 add-on card
> >> connected to a DP monitor.
> >
> > Just a question regarding the usage of this DSI-DP bridge.
> > What is the state of the DSI lanes after the DSI host has been
> > initialized,
> > but before calling atomic_pre_enable? AFAIK this bridge requires LP-11 on
> > DSI at any time for accessing the AUX channel.
>
> We haven't received any test reports for the DSI-DP case... I was
> looking at the datasheet, and I wonder, why do you say the bridge
> requires DSI to be up for the AUX transactions?
Looking at Figure 4.20 (Power On Sequence) in the datasheet TC9595XBG (Rev 1.1
2021-06-23) you can see that RESX can be released (only) after DSI lanes went
to LP-11 state. I got information, down from the support, that LP-11 must be
up in order to use AUX channel. This also matches our observations, DSI hosts
often enable LP-11 only in atomic_prepare. That's too late so we used some
hacks to enable LP-11 right from the beginning just to get access to AUX
channel.
Best regards
Alexander
>
> Tomi
>
> > Best regards,
> > Alexander
> >
> >> Signed-off-by: Tomi Valkeinen <[email protected]>
> >> ---
> >> Changes in v2:
> >> - Update the format negotiation patch as discussed in
> >> https://lore.kernel.org/all/[email protected]/
> >> -
> >> Link to v1:
> >> https://lore.kernel.org/r/20231031-tc358767-v1-0-392081ad9f4b@ideasonboar
> >> d.
> >> com
> >>
> >> ---
> >>
> >> Aradhya Bhatia (1):
> >> drm/bridge: tc358767: Add format negotiation hooks for DPI/DSI to
> >>
> >> (e)DP
> >>
> >> Tomi Valkeinen (1):
> >> drm/bridge: tc358767: Fix link properties discovery
> >>
> >> drivers/gpu/drm/bridge/tc358767.c | 32 ++++++++++++++++++++++++++++++++
> >> 1 file changed, 32 insertions(+)
> >>
> >> ---
> >> base-commit: 9d7c8c066916f231ca0ed4e4fce6c4b58ca3e451
> >> change-id: 20231031-tc358767-58e3ebdf95f0
> >>
> >> Best regards,
--
TQ-Systems GmbH | M?hlstra?e 2, Gut Delling | 82229 Seefeld, Germany
Amtsgericht M?nchen, HRB 105018
Gesch?ftsf?hrer: Detlef Schneider, R?diger Stahl, Stefan Schneider
http://www.tq-group.com/
On 11.12.23 09:07, Aradhya Bhatia wrote:
>
>
> On 06/12/23 17:41, Tomi Valkeinen wrote:
>> Hi,
>>
>> On 08/11/2023 14:45, Alexander Stein wrote:
>>> Hi Tomi,
>>>
>>> Am Mittwoch, 8. November 2023, 12:27:21 CET schrieb Tomi Valkeinen:
>>>> These two patches are needed to make tc358767 work in the
>>>> DRM_BRIDGE_ATTACH_NO_CONNECTOR case, at least when using a DP connector.
>>>>
>>>> I have tested this with TI AM654 EVM with a tc358767 add-on card
>>>> connected to a DP monitor.
>>>
>>> Just a question regarding the usage of this DSI-DP bridge.
>>> What is the state of the DSI lanes after the DSI host has been
>>> initialized,
>>> but before calling atomic_pre_enable? AFAIK this bridge requires LP-11
>>> on DSI
>>> at any time for accessing the AUX channel.
>
> + Marek
>
> Marek, Alexander,
>
> A quick grep tells me that you have added devicetree for tc358767 in DSI
> to (e)DP mode on other platforms. Could you please test these patches
> and report if you find any issue?
Is this the last blocker to move forward with these fixes? I'd really
like to see them finally merged.
Thanks,
Jan
--
Siemens AG, Technology
Linux Expert Center
Hi everyone,
Am Donnerstag, 15. Februar 2024, 09:53:54 CET schrieb Jan Kiszka:
> On 11.12.23 09:07, Aradhya Bhatia wrote:
> > On 06/12/23 17:41, Tomi Valkeinen wrote:
> >> Hi,
> >>
> >> On 08/11/2023 14:45, Alexander Stein wrote:
> >>> Hi Tomi,
> >>>
> >>> Am Mittwoch, 8. November 2023, 12:27:21 CET schrieb Tomi Valkeinen:
> >>>> These two patches are needed to make tc358767 work in the
> >>>> DRM_BRIDGE_ATTACH_NO_CONNECTOR case, at least when using a DP
> >>>> connector.
> >>>>
> >>>> I have tested this with TI AM654 EVM with a tc358767 add-on card
> >>>> connected to a DP monitor.
> >>>
> >>> Just a question regarding the usage of this DSI-DP bridge.
> >>> What is the state of the DSI lanes after the DSI host has been
> >>> initialized,
> >>> but before calling atomic_pre_enable? AFAIK this bridge requires LP-11
> >>> on DSI
> >>> at any time for accessing the AUX channel.
> >
> > + Marek
> >
> > Marek, Alexander,
> >
> > A quick grep tells me that you have added devicetree for tc358767 in DSI
> > to (e)DP mode on other platforms. Could you please test these patches
> > and report if you find any issue?
Sorry, I can't provide any feedback here. I've yet to setup the DSI-DP
correctly.
Best regards,
Alexander
> Is this the last blocker to move forward with these fixes? I'd really
> like to see them finally merged.
>
> Thanks,
> Jan
--
TQ-Systems GmbH | M?hlstra?e 2, Gut Delling | 82229 Seefeld, Germany
Amtsgericht M?nchen, HRB 105018
Gesch?ftsf?hrer: Detlef Schneider, R?diger Stahl, Stefan Schneider
http://www.tq-group.com/
On 15/02/2024 11:03, Alexander Stein wrote:
> Hi everyone,
>
> Am Donnerstag, 15. Februar 2024, 09:53:54 CET schrieb Jan Kiszka:
>> On 11.12.23 09:07, Aradhya Bhatia wrote:
>>> On 06/12/23 17:41, Tomi Valkeinen wrote:
>>>> Hi,
>>>>
>>>> On 08/11/2023 14:45, Alexander Stein wrote:
>>>>> Hi Tomi,
>>>>>
>>>>> Am Mittwoch, 8. November 2023, 12:27:21 CET schrieb Tomi Valkeinen:
>>>>>> These two patches are needed to make tc358767 work in the
>>>>>> DRM_BRIDGE_ATTACH_NO_CONNECTOR case, at least when using a DP
>>>>>> connector.
>>>>>>
>>>>>> I have tested this with TI AM654 EVM with a tc358767 add-on card
>>>>>> connected to a DP monitor.
>>>>>
>>>>> Just a question regarding the usage of this DSI-DP bridge.
>>>>> What is the state of the DSI lanes after the DSI host has been
>>>>> initialized,
>>>>> but before calling atomic_pre_enable? AFAIK this bridge requires LP-11
>>>>> on DSI
>>>>> at any time for accessing the AUX channel.
>>>
>>> + Marek
>>>
>>> Marek, Alexander,
>>>
>>> A quick grep tells me that you have added devicetree for tc358767 in DSI
>>> to (e)DP mode on other platforms. Could you please test these patches
>>> and report if you find any issue?
>
> Sorry, I can't provide any feedback here. I've yet to setup the DSI-DP
> correctly.
Ok. Does anyone have a worry that these patches make the situation worse
for the DSI case than it was before? Afaics, if the DSI lanes are not
set up early enough by the DSI host, the driver would break with and
without these patches.
These do fix the driver for DRM_BRIDGE_ATTACH_NO_CONNECTOR and DPI, so
I'd like to merge these unless these cause a regression with the DSI case.
Tomi
On 2/16/24 10:10, Tomi Valkeinen wrote:
> On 15/02/2024 11:03, Alexander Stein wrote:
>> Hi everyone,
>>
>> Am Donnerstag, 15. Februar 2024, 09:53:54 CET schrieb Jan Kiszka:
>>> On 11.12.23 09:07, Aradhya Bhatia wrote:
>>>> On 06/12/23 17:41, Tomi Valkeinen wrote:
>>>>> Hi,
>>>>>
>>>>> On 08/11/2023 14:45, Alexander Stein wrote:
>>>>>> Hi Tomi,
>>>>>>
>>>>>> Am Mittwoch, 8. November 2023, 12:27:21 CET schrieb Tomi Valkeinen:
>>>>>>> These two patches are needed to make tc358767 work in the
>>>>>>> DRM_BRIDGE_ATTACH_NO_CONNECTOR case, at least when using a DP
>>>>>>> connector.
>>>>>>>
>>>>>>> I have tested this with TI AM654 EVM with a tc358767 add-on card
>>>>>>> connected to a DP monitor.
>>>>>>
>>>>>> Just a question regarding the usage of this DSI-DP bridge.
>>>>>> What is the state of the DSI lanes after the DSI host has been
>>>>>> initialized,
>>>>>> but before calling atomic_pre_enable? AFAIK this bridge requires
>>>>>> LP-11
>>>>>> on DSI
>>>>>> at any time for accessing the AUX channel.
>>>>
>>>> + Marek
>>>>
>>>> Marek, Alexander,
>>>>
>>>> A quick grep tells me that you have added devicetree for tc358767 in
>>>> DSI
>>>> to (e)DP mode on other platforms. Could you please test these patches
>>>> and report if you find any issue?
>>
>> Sorry, I can't provide any feedback here. I've yet to setup the DSI-DP
>> correctly.
>
> Ok. Does anyone have a worry that these patches make the situation worse
> for the DSI case than it was before? Afaics, if the DSI lanes are not
> set up early enough by the DSI host, the driver would break with and
> without these patches.
>
> These do fix the driver for DRM_BRIDGE_ATTACH_NO_CONNECTOR and DPI, so
> I'd like to merge these unless these cause a regression with the DSI case.
1/2 looks good to me, go ahead and apply .