Re: [PATCH] drm/panel/synaptics-r63353: Use _multi variants

[Tejas Vipin <tejasvipin76@gmail.com>]

On 3/6/25 9:03 PM, Anusha Srivatsa wrote:
> On Thu, Mar 6, 2025 at 11:29 AM Dmitry Baryshkov <
> dmitry.baryshkov@linaro.org> wrote:
> 
>> On Thu, 6 Mar 2025 at 17:10, Anusha Srivatsa <asrivats@redhat.com> wrote:
>>>
>>>
>>>
>>> On Thu, Mar 6, 2025 at 4:31 AM Maxime Ripard <mripard@kernel.org> wrote:
>>>>
>>>> Hi Anusha,
>>>>
>>>> On Wed, Mar 05, 2025 at 07:01:41PM -0500, Anusha Srivatsa wrote:
>>>>> Move away from using deprecated API and use _multi
>>>>> variants if available. Use mipi_dsi_msleep()
>>>>> and mipi_dsi_usleep_range() instead of msleep()
>>>>> and usleep_range() respectively.
>>>>>
>>>>> Used Coccinelle to find the multiple occurences.
>>>>> SmPl patch:
>>>>> @rule@
>>>>> identifier dsi_var;
>>>>> identifier r;
>>>>> identifier func;
>>>>> type t;
>>>>> position p;
>>>>> expression dsi_device;
>>>>> expression list es;
>>>>> @@
>>>>> t func(...) {
>>>>> ...
>>>>> struct mipi_dsi_device *dsi_var = dsi_device;
>>>>> +struct mipi_dsi_multi_context dsi_ctx = { .dsi = dsi_var };
>>>>> <+...
>>>>> (
>>>>> -mipi_dsi_dcs_write_seq(dsi_var,es)@p;
>>>>> +mipi_dsi_dcs_write_seq_multi(&dsi_ctx,es);
>>>>> |
>>>>> -mipi_dsi_generic_write_seq(dsi_var,es)@p;
>>>>> +mipi_dsi_generic_write_seq_multi(&dsi_ctx,es);
>>>>> |
>>>>> -mipi_dsi_generic_write(dsi_var,es)@p;
>>>>> +mipi_dsi_generic_write_multi(&dsi_ctx,es);
>>>>> |
>>>>> -r = mipi_dsi_dcs_nop(dsi_var)@p;
>>>>> +mipi_dsi_dcs_nop_multi(&dsi_ctx);
>>>>> |
>>>>> ....rest of API
>>>>> ..
>>>>> )
>>>>> -if(r < 0) {
>>>>> -...
>>>>> -}
>>>>> ...+>
>>>>
>>>> The point of sending a single patch was to review the coccinelle script,
>>>> so you must put the entire script you used here.
>>>>
>>>
>>> I was actually thinking of sending patches per driver this time around
>> since Tejas also seems to be looking into similar parts....Thoughts?
>>
>> Have you discussed it with Tejas? What is his next target?
>>
>> I was hoping he will have some feedback on this patch and we could take it
> from there.....
> It *should* be okay for me to send all changes in a single series...
> 
> Anusha
>

There's 5 more panels that use dcs/generic write_seq(). Maybe I could
work on those (himax-hx8394, samsung-sofef00, samsung-s6d7aa0,
boe-bf060y8m-aj0, jdi-lpm102a188a) while you work on transitioning the
rest of the panels (excluding these) that use other functions in the 
old API? When either of us finishes before the other we could have 
another discussion about splitting work if necessary. I'm open to other
suggestions too.

>>>>> Cc: Maxime Ripard <maxime.ripard@bootlin.com>
>>>>
>>>> That hasn't been my email address for 6 years :)
>>>>
>>> My bad. Will change this.
>>
>>
>>
>> --
>> With best wishes
>> Dmitry
>>
>>
> 

-- 
Tejas Vipin