linux-kernel - Re: [PATCH v2 4/7] media: ov5693: move hw cfg functions into ov5693_check

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Date:   Thu, 30 Jun 2022 11:26:53 +0100
From:   Daniel Scally <djrscally@...il.com>
To:     Jacopo Mondi <jacopo@...ndi.org>,
        Tommaso Merciai <tommaso.merciai@...rulasolutions.com>
Cc:     linuxfancy@...glegroups.com, linux-amarula@...rulasolutions.com,
        quentin.schulz@...obroma-systems.com,
        Mauro Carvalho Chehab <mchehab@...nel.org>,
        linux-media@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v2 4/7] media: ov5693: move hw cfg functions into
 ov5693_check_hwcfg

Hey

On 29/06/2022 09:16, Jacopo Mondi wrote:
> Hi Tommaso,
>
> On Mon, Jun 27, 2022 at 05:04:50PM +0200, Tommaso Merciai wrote:
>> Move hw configuration functions into ov5693_check_hwcfg. This is done to
>> separe the code that handle the hw cfg from probe in a clean way
> s/separe/separate/
>
> You also seem to change the logic of the clk handling, please mention
> this in the commit message, otherwise one could be fooled into
> thinking you're only moving code around with no functional changes...
>
>> Signed-off-by: Tommaso Merciai <tommaso.merciai@...rulasolutions.com>
>> ---
>>  drivers/media/i2c/ov5693.c | 53 +++++++++++++++++++++++---------------
>>  1 file changed, 32 insertions(+), 21 deletions(-)
>>
>> diff --git a/drivers/media/i2c/ov5693.c b/drivers/media/i2c/ov5693.c
>> index d2adc5513a21..d5a934ace597 100644
>> --- a/drivers/media/i2c/ov5693.c
>> +++ b/drivers/media/i2c/ov5693.c
>> @@ -1348,6 +1348,38 @@ static int ov5693_check_hwcfg(struct ov5693_device *ov5693)
>>  	struct fwnode_handle *endpoint;
>>  	unsigned int i;
>>  	int ret;
>> +	u32 xvclk_rate;
> nit: move it up to maintain reverse-xmas-tree order (I know, it's an
> annoying comment, but since variables are already declared in this order..)
>
>> +
>> +	ov5693->xvclk = devm_clk_get(ov5693->dev, "xvclk");
> Isn't this broken ?
>
> if you use ov5693->xvclk to identify the ACPI vs OF use case shouldn't
> you use the get_optionl() version ? Otherwise in the ACPI case you will have
> -ENOENT if there's not 'xvclk' property and bail out.
>
> Unless my understanding is wrong on ACPI we have "clock-frequency" and
> on OF "xvclk" with an "assigned-clock-rates",
>
> Dan you upstreamed this driver and I assume it was tested on ACPI ?
> Can you clarify how this worked for you, as it seems the original code
> wanted a mandatory "xvclk" ? Are there ACPI tables with an actual
> 'xvclk' property ?


Sorry - late answer, but when I wrote this although it's ostensibly for
an ACPI platform, it's actually only tested with the IPU3 platforms
which work in a _weird_ way. The fix we eventually came to was to create
through the int3472-discrete driver clocks and regulators through the
normal frameworks that a dt platform would expect to consume, so even
though the devices are enumerated through ACPI, the clock/regulator
parts really work more like a dt platform.


You're right that it needs to be get_optional() here, but with that
added I think this is fine - I tested it last night and it works ok for me.

>
>> +	if (IS_ERR(ov5693->xvclk))
>> +		return dev_err_probe(ov5693->dev, PTR_ERR(ov5693->xvclk),
>> +				     "failed to get xvclk: %ld\n",
>> +				     PTR_ERR(ov5693->xvclk));
>> +
>> +	if (ov5693->xvclk) {
>> +		xvclk_rate = clk_get_rate(ov5693->xvclk);
>> +	} else {
>> +		ret = fwnode_property_read_u32(fwnode, "clock-frequency",
>> +					       &xvclk_rate);
>> +
>> +		if (ret) {
>> +			dev_err(ov5693->dev, "can't get clock frequency");
>> +			return ret;
>> +		}
>> +	}
>> +
>> +	if (xvclk_rate != OV5693_XVCLK_FREQ)
>> +		dev_warn(ov5693->dev, "Found clk freq %u, expected %u\n",
>> +			 xvclk_rate, OV5693_XVCLK_FREQ);
>> +
>> +	ret = ov5693_configure_gpios(ov5693);
>> +	if (ret)
>> +		return ret;
>> +
>> +	ret = ov5693_get_regulators(ov5693);
>> +	if (ret)
>> +		return dev_err_probe(ov5693->dev, ret,
>> +				     "Error fetching regulators\n");
>>
>>  	endpoint = fwnode_graph_get_next_endpoint(fwnode, NULL);
>>  	if (!endpoint)
>> @@ -1390,7 +1422,6 @@ static int ov5693_check_hwcfg(struct ov5693_device *ov5693)
>>  static int ov5693_probe(struct i2c_client *client)
>>  {
>>  	struct ov5693_device *ov5693;
>> -	u32 xvclk_rate;
>>  	int ret = 0;
>>
>>  	ov5693 = devm_kzalloc(&client->dev, sizeof(*ov5693), GFP_KERNEL);
>> @@ -1408,26 +1439,6 @@ static int ov5693_probe(struct i2c_client *client)
>>
>>  	v4l2_i2c_subdev_init(&ov5693->sd, client, &ov5693_ops);
>>
>> -	ov5693->xvclk = devm_clk_get(&client->dev, "xvclk");
>> -	if (IS_ERR(ov5693->xvclk)) {
>> -		dev_err(&client->dev, "Error getting clock\n");
>> -		return PTR_ERR(ov5693->xvclk);
>> -	}
>> -
>> -	xvclk_rate = clk_get_rate(ov5693->xvclk);
>> -	if (xvclk_rate != OV5693_XVCLK_FREQ)
>> -		dev_warn(&client->dev, "Found clk freq %u, expected %u\n",
>> -			 xvclk_rate, OV5693_XVCLK_FREQ);
>> -
>> -	ret = ov5693_configure_gpios(ov5693);
>> -	if (ret)
>> -		return ret;
>> -
>> -	ret = ov5693_get_regulators(ov5693);
>> -	if (ret)
>> -		return dev_err_probe(&client->dev, ret,
>> -				     "Error fetching regulators\n");
>> -
>>  	ov5693->sd.flags |= V4L2_SUBDEV_FL_HAS_DEVNODE;
>>  	ov5693->pad.flags = MEDIA_PAD_FL_SOURCE;
>>  	ov5693->sd.entity.function = MEDIA_ENT_F_CAM_SENSOR;
>> --
>> 2.25.1
>>