CVE-2026-89741

In the Linux kernel, the following vulnerability has been resolved:

Revert "media: v4l2-dev: fix error handling in video register device()"

This reverts commit 2a934fdb01db6458288fc9386d3d8ceba6dd551a.

The intentions of that patch were good, but it doesn't work.

The idea is that if device register fails, you have to do a put device to let the ref counter release resources.

However, the V4L2 API says that if video register device() fails, then you have to call video device release(), which kfree()s the video device struct.

But the put device() will already have freed the struct, so you end up in a double-free scenario.

There is not really a good way of fixing this without breaking video register device() into two parts, one that initializes everything, and one that does the actual device register, and then converting all V4L2 drivers to this new model.

That is a massive job, and it is very unlikely that device register will fail.

So rather than ending up in a double-free scenario, just revert this patch, and in that case we'll have a small memory leak. Which is a lot more robust.