mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Vikash Garodia <quic_vgarodia@quicinc.com>
To: Krzysztof Kozlowski <krzk@kernel.org>,
	Dikshita Agarwal <quic_dikshita@quicinc.com>,
	Abhinav Kumar <quic_abhinavk@quicinc.com>,
	"Mauro Carvalho Chehab" <mchehab@kernel.org>,
	Rob Herring <robh@kernel.org>,
	Krzysztof Kozlowski <krzk+dt@kernel.org>,
	Conor Dooley <conor+dt@kernel.org>,
	Philipp Zabel <p.zabel@pengutronix.de>
Cc: <linux-media@vger.kernel.org>, <linux-arm-msm@vger.kernel.org>,
	<devicetree@vger.kernel.org>, <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH v3 04/29] media: iris: initialize power resources
Date: Fri, 6 Sep 2024 16:51:56 +0530	[thread overview]
Message-ID: <f88d8596-c6a0-356e-060e-81d68f038995@quicinc.com> (raw)
In-Reply-To: <76ffb882-10f9-4737-afa2-9bb60248835d@kernel.org>

Hi Krzysztof,

On 9/5/2024 5:27 PM, Krzysztof Kozlowski wrote:
> On 05/09/2024 13:53, Dikshita Agarwal wrote:
>>
>>
>> On 8/27/2024 4:21 PM, Krzysztof Kozlowski wrote:
>>> On 27/08/2024 12:05, Dikshita Agarwal via B4 Relay wrote:
>>>> From: Dikshita Agarwal <quic_dikshita@quicinc.com>
>>>>
>>>> Add support for initializing Iris "resources", which are clocks,
>>>> interconnects, power domains, reset clocks, and clock frequencies
>>>> used for iris hardware.
>>>>
>>>> Signed-off-by: Dikshita Agarwal <quic_dikshita@quicinc.com>
>>>> ---
>>>
>>> ...
>>>
>>>> +struct iris_platform_data sm8550_data = {
>>>> +	.icc_tbl = sm8550_icc_table,
>>>> +	.icc_tbl_size = ARRAY_SIZE(sm8550_icc_table),
>>>> +	.clk_rst_tbl = sm8550_clk_reset_table,
>>>> +	.clk_rst_tbl_size = ARRAY_SIZE(sm8550_clk_reset_table),
>>>> +	.pmdomain_tbl = sm8550_pmdomain_table,
>>>> +	.pmdomain_tbl_size = ARRAY_SIZE(sm8550_pmdomain_table),
>>>> +	.opp_pd_tbl = sm8550_opp_pd_table,
>>>> +	.opp_pd_tbl_size = ARRAY_SIZE(sm8550_opp_pd_table),
>>>> +	.clk_tbl = sm8550_clk_table,
>>>> +	.clk_tbl_size = ARRAY_SIZE(sm8550_clk_table),
>>>> +};
>>>> diff --git a/drivers/media/platform/qcom/iris/iris_probe.c b/drivers/media/platform/qcom/iris/iris_probe.c
>>>> index 0a54fdaa1ab5..2616a31224f9 100644
>>>> --- a/drivers/media/platform/qcom/iris/iris_probe.c
>>>> +++ b/drivers/media/platform/qcom/iris/iris_probe.c
>>>> @@ -69,6 +69,19 @@ static int iris_probe(struct platform_device *pdev)
>>>>  	if (core->irq < 0)
>>>>  		return core->irq;
>>>>  
>>>> +	core->iris_platform_data = of_device_get_match_data(core->dev);
>>>> +	if (!core->iris_platform_data) {
>>>> +		ret = -ENODEV;
>>>> +		dev_err_probe(core->dev, ret, "init platform failed\n");
>>>
>>> That's not even possible. I would suggest dropping entire if. But if yoi
>>> insist, then without this weird redundant code. return -EINVAL.
>>>
>> Its possible if platform data is not initialized and this is only place we
>> check it, there is no further NULL check for the same.
> 
> It is possible? Then point me to the code line. Or present some code
> flow leading to it.
Lets go with return -EINVAL in this if block.

> 
>>>> +		return ret;
>>>> +	}
>>>> +
>>>> +	ret = iris_init_resources(core);
>>>> +	if (ret) {
>>>> +		dev_err_probe(core->dev, ret, "init resource failed\n");
>>>> +		return ret;
>>>
>>> How many same errors are you printing? Not mentioning that syntax of
>>> dev_errp_rpboe is different...
>> We have these errors at multiple points to know at what point the probe
>> failed which is useful while debugging probe failures.
> 
> Duplicating is not helpful.
> 
>> But Sure we will revisit this code and fix the syntax of dev_err_probe.
> 
>>>
>>>
>>>> +	}
>>>> +
>>>>  	ret = v4l2_device_register(dev, &core->v4l2_dev);
>>>>  	if (ret)
>>>>  		return ret;
>>>> @@ -88,8 +101,14 @@ static int iris_probe(struct platform_device *pdev)
>>>>  }
>>>>  
>>>>  static const struct of_device_id iris_dt_match[] = {
>>>> -	{ .compatible = "qcom,sm8550-iris", },
>>>> -	{ .compatible = "qcom,sm8250-venus", },
>>>> +	{
>>>> +		.compatible = "qcom,sm8550-iris",
>>>> +		.data = &sm8550_data,
>>>> +	},
>>>> +	{
>>>> +		.compatible = "qcom,sm8250-venus",
>>>> +		.data = &sm8250_data,
>>>
>>> You just added this. No, please do not add code which is immediatly
>>> incorrect.
>> It's not incorrect, in earlier patch we only added the compatible strings
>> and with this patch introducing the platform data and APIs to get it.
> 
> It is incorrect to immediately remove it. You keep arguing on basic
> stuff. Sorry, but that is not how it works. If you add code and
> IMMEDIATELY remove it, then it means the code was not needed. Or was not
> correct. Choose one.
I think it is not removing it. It is adding platform data to compatibles
introduced in previous patch. Maybe it appears as if it is removing it.
> 
> ...
> 
>>>
>>> This should be just part of of main unit file, next to probe. It is
>>> unusual to see probe parts not next to probe. Sorry, that's wrong.
>>>
>> All the APIs handling(init/enable/disable) the different resources (PM
>> domains, OPP, clocks, buses) are kept in this iris_resource.c file hence
>> this API to init all those resources is kept here to not load iris_probe.c
>> file.
> 
> You introduce your own coding style and as an argument you use just "I
> do this".
> 
> The expected is to see resource initialization next to probe. Repeating
> what your code does, is not helping me to understand your design choice.
I see your point and it would be good to have the resources initialized as part
of probe.

> Best regards,
> Krzysztof

Regards,
Vikash


  reply	other threads:[~2024-09-06 11:22 UTC|newest]

Thread overview: 93+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-08-27 10:05 [PATCH v3 00/29] Qualcomm iris video decoder driver Dikshita Agarwal via B4 Relay
2024-08-27 10:05 ` [PATCH v3 01/29] dt-bindings: media: Add sm8550 dt schema Dikshita Agarwal via B4 Relay
2024-08-27 10:42   ` Krzysztof Kozlowski
2024-09-05  5:41     ` Dikshita Agarwal
2024-08-27 10:05 ` [PATCH v3 02/29] media: MAINTAINERS: Add Qualcomm Iris video accelerator driver Dikshita Agarwal via B4 Relay
2024-08-27 10:42   ` Krzysztof Kozlowski
2024-09-05  5:47     ` Dikshita Agarwal
2024-09-05 10:10       ` Dmitry Baryshkov
2024-09-05 11:02         ` Dikshita Agarwal
2024-09-05 11:02           ` Dmitry Baryshkov
2024-09-05 11:14             ` Dikshita Agarwal
2024-08-27 10:05 ` [PATCH v3 03/29] media: iris: add platform driver for iris video device Dikshita Agarwal via B4 Relay
2024-08-27 14:08   ` Bryan O'Donoghue
2024-08-29  9:13     ` Dmitry Baryshkov
2024-08-29  9:36       ` Bryan O'Donoghue
2024-09-05  6:12       ` Dikshita Agarwal
2024-09-05  6:15         ` Dikshita Agarwal
2024-09-05 10:11           ` Dmitry Baryshkov
2024-09-05 10:59             ` Dikshita Agarwal
2024-09-05 11:07               ` Dmitry Baryshkov
2024-09-05 11:13                 ` Dikshita Agarwal
2024-08-27 10:05 ` [PATCH v3 04/29] media: iris: initialize power resources Dikshita Agarwal via B4 Relay
2024-08-27 10:51   ` Krzysztof Kozlowski
2024-09-05 11:53     ` Dikshita Agarwal
2024-09-05 11:57       ` Krzysztof Kozlowski
2024-09-06 11:21         ` Vikash Garodia [this message]
2024-09-06 12:04           ` Krzysztof Kozlowski
2024-09-06 19:47             ` Vikash Garodia
2024-09-07  9:07               ` Krzysztof Kozlowski
2024-08-27 10:05 ` [PATCH v3 05/29] media: iris: implement iris v4l2 file ops Dikshita Agarwal via B4 Relay
2024-09-06 19:05   ` Markus Elfring
2024-09-07  8:52   ` Markus Elfring
2024-08-27 10:05 ` [PATCH v3 06/29] media: iris: introduce iris core state management with shared queues Dikshita Agarwal via B4 Relay
2024-08-28  2:38   ` kernel test robot
2024-08-27 10:05 ` [PATCH v3 07/29] media: iris: implement video firmware load/unload Dikshita Agarwal via B4 Relay
2024-08-27 23:13   ` kernel test robot
2024-08-31 13:18   ` Bryan O'Donoghue
2024-09-02  0:04     ` Dmitry Baryshkov
2024-09-05  6:17       ` Dikshita Agarwal
2024-08-27 10:05 ` [PATCH v3 08/29] media: iris: implement boot sequence of the firmware Dikshita Agarwal via B4 Relay
2024-09-05 12:34   ` Bryan O'Donoghue
2024-09-06 11:27     ` Vikash Garodia
2024-08-27 10:05 ` [PATCH v3 09/29] media: iris: introduce Host firmware interface with necessary hooks Dikshita Agarwal via B4 Relay
2024-09-05 12:36   ` Bryan O'Donoghue
2024-09-24  9:13     ` Dikshita Agarwal
2024-09-05 13:10   ` Bryan O'Donoghue
2024-09-06 13:31     ` Vikash Garodia
2024-08-27 10:05 ` [PATCH v3 10/29] media: iris: implement power management Dikshita Agarwal via B4 Relay
2024-09-05 13:23   ` Bryan O'Donoghue
2024-09-05 13:46     ` Krzysztof Kozlowski
2024-09-24  8:38       ` Dikshita Agarwal
2024-09-24  8:36     ` Dikshita Agarwal
2024-08-27 10:05 ` [PATCH v3 11/29] media: iris: implement reqbuf ioctl with vb2_queue_setup Dikshita Agarwal via B4 Relay
2024-09-06 12:50   ` Bryan O'Donoghue
2024-09-06 13:05     ` Bryan O'Donoghue
2024-09-26 10:47     ` Dikshita Agarwal
2024-08-27 10:05 ` [PATCH v3 12/29] media: iris: implement s_fmt, g_fmt and try_fmt ioctls Dikshita Agarwal via B4 Relay
2024-09-24 14:41   ` Bryan O'Donoghue
2024-09-26 10:49     ` Dikshita Agarwal
2024-08-27 10:05 ` [PATCH v3 13/29] media: iris: implement g_selection ioctl Dikshita Agarwal via B4 Relay
2024-08-27 10:05 ` [PATCH v3 14/29] media: iris: implement enum_fmt and enum_frameintervals ioctls Dikshita Agarwal via B4 Relay
2024-08-27 10:05 ` [PATCH v3 15/29] media: iris: implement subscribe_event and unsubscribe_event ioctls Dikshita Agarwal via B4 Relay
2024-08-27 10:05 ` [PATCH v3 16/29] media: iris: implement iris v4l2_ctrl_ops and prepare capabilities Dikshita Agarwal via B4 Relay
2024-08-29  9:33   ` Dmitry Baryshkov
2024-10-01 13:01     ` Vedang Nagar
2024-10-06 16:46       ` Dmitry Baryshkov
2024-08-27 10:05 ` [PATCH v3 17/29] media: iris: implement query_cap, query_ctrl and query_menu ioctls Dikshita Agarwal via B4 Relay
2024-09-24 14:49   ` Bryan O'Donoghue
2024-09-26 10:50     ` Dikshita Agarwal
2024-08-27 10:05 ` [PATCH v3 18/29] media: iris: implement vb2 streaming ops Dikshita Agarwal via B4 Relay
2024-08-28  0:26   ` kernel test robot
2024-08-27 10:05 ` [PATCH v3 19/29] media: iris: implement set properties to firmware during streamon Dikshita Agarwal via B4 Relay
2024-09-24 15:09   ` Bryan O'Donoghue
2024-08-27 10:05 ` [PATCH v3 20/29] media: iris: subscribe parameters and properties to firmware for hfi_gen2 Dikshita Agarwal via B4 Relay
2024-09-24 15:16   ` Bryan O'Donoghue
2024-09-26 10:55     ` Dikshita Agarwal
2024-08-27 10:05 ` [PATCH v3 21/29] media: iris: allocate, initialize and queue internal buffers Dikshita Agarwal via B4 Relay
2024-08-27 10:05 ` [PATCH v3 22/29] media: iris: implement vb2 ops for buf_queue and firmware response Dikshita Agarwal via B4 Relay
2024-08-27 10:05 ` [PATCH v3 23/29] media: iris: add support for dynamic resolution change Dikshita Agarwal via B4 Relay
2024-08-27 10:05 ` [PATCH v3 24/29] media: iris: handle streamoff/on from client in " Dikshita Agarwal via B4 Relay
2024-08-27 10:05 ` [PATCH v3 25/29] media: iris: add support for drain sequence Dikshita Agarwal via B4 Relay
2024-08-27 10:05 ` [PATCH v3 26/29] media: iris: add check whether the video session is supported or not Dikshita Agarwal via B4 Relay
2024-08-27 10:05 ` [PATCH v3 27/29] media: iris: implement power scaling for vpu2 and vpu3 Dikshita Agarwal via B4 Relay
2024-08-27 10:05 ` [PATCH v3 28/29] media: iris: add allow checks for v4l2 ioctls Dikshita Agarwal via B4 Relay
2024-08-27 10:05 ` [PATCH v3 29/29] media: iris: add check to allow sub states transitions Dikshita Agarwal via B4 Relay
2024-08-27 13:41 ` [PATCH v3 00/29] Qualcomm iris video decoder driver neil.armstrong
2024-09-24  9:13   ` Dikshita Agarwal
2024-10-01 13:28     ` Neil Armstrong
2024-08-31 15:18 ` Bryan O'Donoghue
2024-09-02  0:02   ` Dmitry Baryshkov
2024-09-06 14:19     ` Nicolas Dufresne
2024-09-06 19:26       ` Vikash Garodia
2024-09-06 16:28     ` Abhinav Kumar

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=f88d8596-c6a0-356e-060e-81d68f038995@quicinc.com \
    --to=quic_vgarodia@quicinc.com \
    --cc=conor+dt@kernel.org \
    --cc=devicetree@vger.kernel.org \
    --cc=krzk+dt@kernel.org \
    --cc=krzk@kernel.org \
    --cc=linux-arm-msm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-media@vger.kernel.org \
    --cc=mchehab@kernel.org \
    --cc=p.zabel@pengutronix.de \
    --cc=quic_abhinavk@quicinc.com \
    --cc=quic_dikshita@quicinc.com \
    --cc=robh@kernel.org \
    /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®