linux-kernel - Re: [PATCH 2/2] media: qcom: camss: Restrict endpoint bus-type to D-PHY

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] [thread-next>] [day] [month] [year] [list]

Message-ID: <12618e8f-6f7b-4578-8315-1dc3d7049688@linaro.org>
Date: Fri, 13 Dec 2024 13:49:55 +0200
From: Vladimir Zapolskiy <vladimir.zapolskiy@...aro.org>
To: Luca Weiss <luca.weiss@...rphone.com>,
 Bryan O'Donoghue <bryan.odonoghue@...aro.org>, Robert Foss
 <rfoss@...nel.org>, Todor Tomov <todor.too@...il.com>,
 Mauro Carvalho Chehab <mchehab@...nel.org>, Rob Herring <robh@...nel.org>,
 Krzysztof Kozlowski <krzk+dt@...nel.org>, Conor Dooley
 <conor+dt@...nel.org>, Barnabas Czeman <barnabas.czeman@...nlining.org>,
 AngeloGioacchino Del Regno <angelogioacchino.delregno@...ainline.org>,
 Caleb Connolly <caleb.connolly@...aro.org>, David Heidelberg <david@...t.cz>
Cc: ~postmarketos/upstreaming@...ts.sr.ht, phone-devel@...r.kernel.org,
 linux-arm-msm@...r.kernel.org, linux-media@...r.kernel.org,
 devicetree@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 2/2] media: qcom: camss: Restrict endpoint bus-type to
 D-PHY

On 12/13/24 13:22, Luca Weiss wrote:
> On Fri Dec 13, 2024 at 12:02 PM CET, Vladimir Zapolskiy wrote:
>> On 12/9/24 14:32, Bryan O'Donoghue wrote:
>>> On 09/12/2024 12:01, Luca Weiss wrote:
>>>> Currently the Qualcomm CAMSS driver only supports D-PHY while the
>>>> hardware on most SoCs also supports C-PHY. Until this support is added,
>>>> check for D-PHY to make it somewhat explicit that C-PHY won't work.
>>>>
>>>> Signed-off-by: Luca Weiss <luca.weiss@...rphone.com>
>>>> ---
>>>>     drivers/media/platform/qcom/camss/camss.c | 9 +++++++++
>>>>     1 file changed, 9 insertions(+)
>>>>
>>>> diff --git a/drivers/media/platform/qcom/camss/camss.c b/drivers/media/platform/qcom/camss/camss.c
>>>> index 9fb31f4c18adee886cd0bcf84438a8f27635e07f..b99af35074cdf6fa794a0d2f0d54ecf12ac354d9 100644
>>>> --- a/drivers/media/platform/qcom/camss/camss.c
>>>> +++ b/drivers/media/platform/qcom/camss/camss.c
>>>> @@ -1855,6 +1855,15 @@ static int camss_of_parse_endpoint_node(struct device *dev,
>>>>     	if (ret)
>>>>     		return ret;
>>>>     
>>>> +	/*
>>>> +	 * Most SoCs support both D-PHY and C-PHY standards, but currently only
>>>> +	 * D-PHY is supported in the driver.
>>>> +	 */
>>>> +	if (vep.bus_type != V4L2_MBUS_CSI2_DPHY) {
>>>> +		dev_err(dev, "Unsupported bus type %d\n", vep.bus_type);
>>>> +		return -EINVAL;
>>>> +	}
>>>> +
>>
>> Looks like it would break all old board dtbs, which is not just bad, but NAK.
>>
>> V4L2_MBUS_UNKNOWN shall be properly handled without the risk of regressions.
> 
> Please see drivers/media/v4l2-core/v4l2-fwnode.c around line 218.
> The code there sets bus_type if it's UNKNOWN
> 
>      if (bus_type == V4L2_MBUS_UNKNOWN)
>          vep->bus_type = V4L2_MBUS_CSI2_DPHY;
> 
> So setting "bus-type" in dt is not necessary, even if it makes things
> more explicit from dt side. I don't think we'll ever get UNKNOWN here in
> camss.

Thank you for pointing it out, I haven't tested the change yet, but hopefully
I will find time today to do it later on, it should exclude any doubts.

--
Best wishes,
Vladimir