Re: [RFC PATCH v2] media: Virtual camera driver

[Jarkko Sakkinen <jarkko@kernel.org>]

On Tue, Feb 03, 2026 at 10:09:25AM +0200, Jarkko Sakkinen wrote:
> On Mon, Feb 02, 2026 at 10:44:21PM +0200, Jarkko Sakkinen wrote:
> > Already a quick Google survey backs strongly that OOT drivers (e.g.,
> > v4l2loopback) are the defacto solution for streaming phone cameras in
> > video conference calls, which puts confidential discussions at risk.
> > 
> > It can be also claimed that there's enough OOT usage in the wild that
> > possible security bugs could be considered as potential zerodays for the
> > benefit of malicious actors.
> > 
> > The situation has been stagnated for however many years, which is
> > unsastainable situation, and it further factors potential security
> > risks. Therefore, a driver is needed to address the popular use case.
> > 
> > vcam is a DMA-BUF backed virtual camera driver capable of creating video
> > capture devices to which data can be streamed through /dev/vcam after
> > calling VCAM_IOC_CREATE. Frames are pushed with VCAM_IOC_QUEUE and recycled
> > with VCAM_IOC_DEQUEUE. Zero-copy semantics are supported for shared DMA-BUF
> > between capture and output.
> > 
> > This enables efficient implementation of software, which can manage network
> > video streams from phone cameras, and map those streams to video devices.
> > 
> > PipeWire or any other specific pick of userspace software cannot really
> > address the issue at scale, as e.g., the use of v4l2loopback is both wide
> > and scattered.
> > 
> > Signed-off-by: Jarkko Sakkinen <jarkko@kernel.org>
> 
> After learning a bit more about the topic and for future updates I will
> drop YUYV. NV12, MJPEG, and additionally RGBX32 and XRGB32 for testing
> and GPUs define pretty well the requirements for a software define device,
> and limit the applicability of "proprietary risk" (as that was the main
> concern raised). I honestly did not have idea what would be an
> appropriate subset of formats to constraint the driver initially.
> 
> In addition, a better name for this module would probably be swcam as it
> does not  mix up with those pre-existing devices starting with the
> letter 'v'.

In addition to phones, there's also category of devices called IP
cameras, and managing them ubiquitos to other camera device is a
convenience to say the least.

And when it comes to pure testing, not all testing where V4L2 is
involved is neither about simulating V4L2 hardware nor centered around
V4L2 as a topic. It could be e.g., embedded system testing where you
want to mock the cameras with simulated content  in software only
testing (QEMU has V4L2 passthrough)

BR, Jarkko