[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <6067898e-eaac-4266-b4a3-388db9a918fa@xs4all.nl>
Date: Wed, 14 Feb 2024 11:54:12 +0100
From: Hans Verkuil <hverkuil-cisco@...all.nl>
To: Arnd Bergmann <arnd@...db.de>, Sakari Ailus <sakari.ailus@...ux.intel.com>
Cc: Arnd Bergmann <arnd@...nel.org>, linux-media@...r.kernel.org,
linux-kernel@...r.kernel.org, Mauro Carvalho Chehab <mchehab@...nel.org>
Subject: Re: [PATCH] media: v4l: marvell: select CONFIG_V4L2_ASYNC where
needed
On 14/02/2024 11:48, Arnd Bergmann wrote:
> On Wed, Feb 14, 2024, at 11:33, Sakari Ailus wrote:
>> Hi Hans, Arnd,
>>
>> On Wed, Feb 14, 2024 at 11:24:41AM +0100, Hans Verkuil wrote:
>>> Arnd, Sakari,
>>>
>>> Is this something that needs to go to v6.8? Or just v6.9?
>>>
>>> Do we need a Fixes tag?
>>
>> The patch seems to be related to this:
>> <URL:https://lore.kernel.org/oe-kbuild-all/202402130955.f6uxzdCA-lkp@intel.com/>.
>>
>> So most likely yes, and Cc: stable, too.
>
> Ah, so lkp bisected it to that commit, which means it was
> already broken in 6.5, but I'm fairly sure the bug is even
> older then, as your commit seems to have only uncovered
> an existing problem.
>
> It was definitely working before ff3cc65cadb5 ("media: v4l: async,
> fwnode: Improve module organisation") in linux-5.13, but it's not
> clear if that is the culprit. It's probably safe to backport
> to v5.15 and higher.
>
> Arnd
If it has been broken for so long, then should we bother with v6.8?
I'm not saying we should, I just like to get your opinion on this.
Regards,
Hans
Powered by blists - more mailing lists