Re: [Freedreno] [PATCH v2 0/7] drm/msm/dpu: simplify DPU encoder init

[Abhinav Kumar <quic_abhinavk@quicinc.com>]

On 5/23/2023 12:30 PM, Abhinav Kumar wrote:
> 
> 
> On 5/23/2023 12:23 PM, Dmitry Baryshkov wrote:
>> On Tue, 23 May 2023 at 22:14, Abhinav Kumar 
>> <quic_abhinavk@quicinc.com> wrote:
>>>
>>>
>>>
>>> On 5/23/2023 7:36 AM, Dmitry Baryshkov wrote:
>>>> On 23/05/2023 10:31, Neil Armstrong wrote:
>>>>> On 23/05/2023 09:20, Dmitry Baryshkov wrote:
>>>>>> On Tue, 23 May 2023 at 04:58, Abhinav Kumar
>>>>>> <quic_abhinavk@quicinc.com> wrote:
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> On 5/18/2023 7:38 PM, Dmitry Baryshkov wrote:
>>>>>>>> Rework dpu_encoder initialization code, simplifying calling 
>>>>>>>> sequences
>>>>>>>> and separating common init parts.
>>>>>>>>
>>>>>>>> Changes since v1:
>>>>>>>> - Withdrawn two pathes for a later consideration
>>>>>>>> - Changed dpu_encoder_phys_init() to return void (Abhinav)
>>>>>>>> - Added small simplifications of dpu_encoder_phys_cmd_init() and
>>>>>>>>      dpu_encoder_phys_wb_init()
>>>>>>>>
>>>>>>>
>>>>>>> I had previously given these comments on the cover letter of v1, so
>>>>>>> giving it again.
>>>>>>>
>>>>>>> Please mention that your series was made on top of
>>>>>>> https://patchwork.freedesktop.org/series/116530/.
>>>>>>>
>>>>>>> Figured it out when I tried to apply it to my branch to test.
>>>>>>>
>>>>>>> I had tested v1, and between v1 and v2 i only see very trivial 
>>>>>>> change,
>>>>>>> so i think its okay to retain:
>>>>>>>
>>>>>>> Tested-by: Abhinav Kumar <quic_abhinavk@quicinc.com> # sc7280
>>>>>>
>>>>>> Unfortunately patchwork ignores tags sent in the cover letter thread.
>>>>>
>>>>> But b4 does with -t option to b4 shazam or b4 am
>>>>
>>>> Yes. But b4 doesn't append Patchwork headers.
>>>>
>>>
>>> If thats the case, either the author can add them to the patches
>>> manually like we do sometimes for R-b tags OR I will go ahead and add it
>>> one by one for every patch now.
>>
>> I'd prefer either to have a single T-B on the latest patch on the
>> series, or a pile of replies with T-B tags. Thank you (for the testing
>> and for providing the feedback).
>> If we ever switch from git-pw to b4, this requirement will be lifted.
>>
> 
> Latest patch means, the last one in the series?
> 
> In this case, that would look a bit odd as that one just removes a temp 
> variable.
> 
> I will provide it on all the patches by tomorrow.
> 

This patch doesnt apply cleanly on msm-next-lumag now. Do you have to 
rebase this? Otherwise please list the dependency on top of msm-next-lumag.

>>>
>>> Let me know what you prefer.
>>>
>>>>>
>>>>> Neil
>>>>>
>>>>>>
>>>>>>>
>>>>>>>> Dmitry Baryshkov (7):
>>>>>>>>      drm/msm/dpu: merge dpu_encoder_init() and dpu_encoder_setup()
>>>>>>>>      drm/msm/dpu: separate common function to init physical encoder
>>>>>>>>      drm/msm/dpu: drop duplicated intf/wb indices from encoder 
>>>>>>>> structs
>>>>>>>>      drm/msm/dpu: inline dpu_encoder_get_wb()
>>>>>>>>      drm/msm/dpu: call dpu_rm_get_intf() from 
>>>>>>>> dpu_encoder_get_intf()
>>>>>>>>      drm/msm/dpu: drop temp variable from 
>>>>>>>> dpu_encoder_phys_cmd_init()
>>>>>>>>      drm/msm/dpu: simplify dpu_encoder_phys_wb_init()
>>>>>>>>
>>>>>>>>     drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c   | 178
>>>>>>>> ++++++++----------
>>>>>>>>     drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.h   |  14 +-
>>>>>>>>     .../gpu/drm/msm/disp/dpu1/dpu_encoder_phys.h  |  15 +-
>>>>>>>>     .../drm/msm/disp/dpu1/dpu_encoder_phys_cmd.c  |  35 ++--
>>>>>>>>     .../drm/msm/disp/dpu1/dpu_encoder_phys_vid.c  |  19 +-
>>>>>>>>     .../drm/msm/disp/dpu1/dpu_encoder_phys_wb.c   |  35 +---
>>>>>>>>     drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c       |  87 ++++-----
>>>>>>>>     7 files changed, 140 insertions(+), 243 deletions(-)
>>>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>
>>>>
>>
>>
>>