From: Abhinav Kumar <quic_abhinavk@quicinc.com>
To: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>,
Marijn Suijten <marijn.suijten@somainline.org>,
Jessica Zhang <quic_jesszhan@quicinc.com>
Cc: Sean Paul <sean@poorly.run>, <linux-kernel@vger.kernel.org>,
<dri-devel@lists.freedesktop.org>,
Rob Clark <robdclark@gmail.com>,
"Daniel Vetter" <daniel@ffwll.ch>,
<linux-arm-msm@vger.kernel.org>, David Airlie <airlied@gmail.com>,
<freedreno@lists.freedesktop.org>
Subject: Re: [Freedreno] [PATCH 2/3] drm/msm/dpu: Set DATABUS_WIDEN on command mode encoders
Date: Thu, 29 Jun 2023 10:26:31 -0700 [thread overview]
Message-ID: <53df7008-0a72-dfa7-27d9-96c07e410ac3@quicinc.com> (raw)
In-Reply-To: <0f988169-58cb-eccf-9590-f5097e085f22@quicinc.com>
On 6/22/2023 4:37 PM, Abhinav Kumar wrote:
>
>
> On 6/22/2023 4:14 PM, Dmitry Baryshkov wrote:
>> On 23/06/2023 01:37, Abhinav Kumar wrote:
>>>
>>>
>>> On 6/21/2023 4:46 PM, Dmitry Baryshkov wrote:
>>>> On 22/06/2023 02:01, Abhinav Kumar wrote:
>>>>>
>>>>>
>>>>> On 6/21/2023 9:36 AM, Dmitry Baryshkov wrote:
>>>>>> On 21/06/2023 18:17, Marijn Suijten wrote:
>>>>>>> On 2023-06-20 14:38:34, Jessica Zhang wrote:
>>>>>>> <snip>
>>>>>>>>>>>>> + if (phys_enc->hw_intf->ops.enable_widebus)
>>>>>>>>>>>>> + phys_enc->hw_intf->ops.enable_widebus(phys_enc->hw_intf);
>>>>>>>>>>>>
>>>>>>>>>>>> No. Please provide a single function which takes necessary
>>>>>>>>>>>> configuration, including compression and wide_bus_enable.
>>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> There are two ways to look at this. Your point is coming from
>>>>>>>>>>> the
>>>>>>>>>>> perspective that its programming the same register but just a
>>>>>>>>>>> different
>>>>>>>>>>> bit. But that will also make it a bit confusing.
>>>>>>>>>
>>>>>>>>> My point is to have a high-level function that configures the
>>>>>>>>> INTF for
>>>>>>>>> the CMD mode. This way it can take a structure with necessary
>>>>>>>>> configuration bits.
>>>>>>>>
>>>>>>>> Hi Dmitry,
>>>>>>>>
>>>>>>>> After discussing this approach with Abhinav, we still have a few
>>>>>>>> questions about it:
>>>>>>>>
>>>>>>>> Currently, only 3 of the 32 bits for INTF_CONFIG2 are being used
>>>>>>>> (the
>>>>>>>> rest are reserved with no plans of being programmed in the
>>>>>>>> future). Does
>>>>>>>> this still justify the use of a struct to pass in the necessary
>>>>>>>> configuration?
>>>>>>>
>>>>>>> No. The point Dmitry is making is **not** about this concidentally
>>>>>>> using the same register, but about adding a common codepath to
>>>>>>> enable
>>>>>>> compression on this hw_intf (regardless of the registers it needs to
>>>>>>> touch).
>>>>>>
>>>>>> Actually to setup INTF for CMD stream (which is equal to setting
>>>>>> up compression at this point).
>>>>>>
>>>>>
>>>>> Yes it should be setup intf for cmd and not enable compression.
>>>>>
>>>>> Widebus and compression are different features and we should be
>>>>> able to control them independently.
>>>>>
>>>>> We just enable them together for DSI. So a separation is necessary.
>>>>>
>>>>> But I am still not totally convinced we even need to go down the
>>>>> path for having an op called setup_intf_cmd() which takes in a
>>>>> struct like
>>>>>
>>>>> struct dpu_cmd_intf_cfg {
>>>>> bool data_compress;
>>>>> bool widebus_en;
>>>>> };
>>>>>
>>>>> As we have agreed that we will not touch the video mode timing
>>>>> engine path, it leaves us with only two bits.
>>>>>
>>>>> And like I said, its not that these two bits always go together. We
>>>>> want to be able to control them independently which means that its
>>>>> not necessary both bits program the same register one by one. We
>>>>> might just end up programming one of them if we just use widebus.
>>>>>
>>>>> Thats why I am still leaning on keeping this approach.
>>>>
>>>> I do not like the idea of having small functions being called
>>>> between modules. So, yes there will a config of two booleans, but it
>>>> is preferable (and more scalable) compared to separate callbacks.
>>>>
>>>
>>> I definitely agree with the scalable part and I even cross checked
>>> that the number of usable bitfields of this register is going up from
>>> one chipset to the other although once again that depends on whether
>>> we will use those features.
>>>
>>> For that reason I am not opposed to the struct idea.
>>>
>>> But there is also another pattern i am seeing which worries me.
>>> Usable bitfields sometimes even reduce. For those cases, if we go
>>> with a pre-defined struct it ends up with redundant members as those
>>> bitfields go away.
>>>
>>> With the function op based approach, we just call the op if the
>>> feature bit / core revision.
>>>
>>> So I wanted to check once more about the fact that we should consider
>>> not just expansion but also reduction.
>>
>> As we have to support all generations, there is no actual reduction.
>> Newer SoCs do not have particular feature/bit, but older ones do. By
>> having multiple optional ops we just move this knowledge from
>> ops->complex_callback() to _setup_block_ops(). But more importantly
>> the caller gets more complicated. Instead of just calling
>> ops->set_me_up(), it has to check all the optional callbacks and call
>> the one by one.
>>
>
> Alright, I am thinking that perhaps because this register is kind of
> unique that its actually controlling a specific setting in the datapath,
> downstream also has separate ops for this.
>
> But thats fine, we can go ahead with the struct based approach.
>
As data_compress has already landed, let me introduced the struct along
with the core_revision based approach in the core_revision series and
this series will expand that struct to include widebus too.
next prev parent reply other threads:[~2023-06-29 17:26 UTC|newest]
Thread overview: 55+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-06-14 1:57 [PATCH 0/3] Add support for databus widen mode Jessica Zhang
2023-06-14 1:57 ` [PATCH 1/3] drm/msm/dpu: Add DPU_INTF_DATABUS_WIDEN feature flag for DPU >= 5.0 Jessica Zhang
2023-06-14 7:50 ` Dmitry Baryshkov
2023-06-14 11:42 ` Marijn Suijten
2023-06-14 12:01 ` Dmitry Baryshkov
2023-06-14 12:23 ` Marijn Suijten
2023-06-14 19:35 ` Abhinav Kumar
2023-06-14 19:54 ` Abhinav Kumar
2023-06-14 20:39 ` [Freedreno] " Abhinav Kumar
2023-06-14 20:43 ` Dmitry Baryshkov
2023-06-16 21:10 ` Abhinav Kumar
2023-06-17 0:37 ` Dmitry Baryshkov
2023-06-20 21:37 ` Jessica Zhang
2023-06-20 22:11 ` Dmitry Baryshkov
2023-06-20 23:37 ` Jessica Zhang
2023-06-14 21:41 ` Marijn Suijten
2023-06-16 21:13 ` Abhinav Kumar
2023-06-16 21:58 ` Marijn Suijten
2023-06-14 22:49 ` Marijn Suijten
2023-06-16 21:06 ` Abhinav Kumar
2023-06-16 22:00 ` Marijn Suijten
2023-06-14 1:57 ` [PATCH 2/3] drm/msm/dpu: Set DATABUS_WIDEN on command mode encoders Jessica Zhang
2023-06-14 7:56 ` Dmitry Baryshkov
2023-06-16 21:18 ` Abhinav Kumar
2023-06-16 21:54 ` Marijn Suijten
2023-06-17 0:35 ` Dmitry Baryshkov
2023-06-20 21:38 ` Jessica Zhang
2023-06-20 22:10 ` Dmitry Baryshkov
2023-06-21 15:17 ` Marijn Suijten
2023-06-21 16:36 ` Dmitry Baryshkov
2023-06-21 19:00 ` Marijn Suijten
2023-06-21 23:01 ` Abhinav Kumar
2023-06-21 23:46 ` Dmitry Baryshkov
2023-06-22 22:37 ` Abhinav Kumar
2023-06-22 23:14 ` Dmitry Baryshkov
2023-06-22 23:37 ` [Freedreno] " Abhinav Kumar
2023-06-29 17:26 ` Abhinav Kumar [this message]
2023-06-29 17:37 ` Jessica Zhang
2023-06-14 1:57 ` [PATCH 3/3] drm/msm/dsi: Enable DATABUS_WIDEN for DSI command mode Jessica Zhang
2023-06-14 7:49 ` Dmitry Baryshkov
2023-06-14 10:03 ` Marijn Suijten
2023-06-22 23:04 ` Abhinav Kumar
2023-06-23 7:02 ` Marijn Suijten
2023-06-16 21:23 ` Abhinav Kumar
2023-06-14 9:56 ` Marijn Suijten
2023-06-23 0:01 ` [Freedreno] " Abhinav Kumar
2023-06-23 0:41 ` Dmitry Baryshkov
2023-06-23 7:19 ` Marijn Suijten
2023-06-23 17:29 ` Abhinav Kumar
2023-06-23 20:14 ` Marijn Suijten
2023-06-23 20:34 ` Abhinav Kumar
2023-06-23 21:33 ` Marijn Suijten
2023-06-23 21:47 ` Abhinav Kumar
2023-06-14 10:09 ` [PATCH 0/3] Add support for databus widen mode Marijn Suijten
2023-06-16 11:30 ` Dmitry Baryshkov
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=53df7008-0a72-dfa7-27d9-96c07e410ac3@quicinc.com \
--to=quic_abhinavk@quicinc.com \
--cc=airlied@gmail.com \
--cc=daniel@ffwll.ch \
--cc=dmitry.baryshkov@linaro.org \
--cc=dri-devel@lists.freedesktop.org \
--cc=freedreno@lists.freedesktop.org \
--cc=linux-arm-msm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=marijn.suijten@somainline.org \
--cc=quic_jesszhan@quicinc.com \
--cc=robdclark@gmail.com \
--cc=sean@poorly.run \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®