I've been working on cleaning up the vboxguest drivers so that they
can be added to the mainline kernel for:
The vboxvideo driver is already upstream, atm it is in staging
because it still needs to be ported to the new atomic-kms APIs
but otherwise it is in very good state and not really of staging
quality (I've already done a lot of cleanup reducing it from
52681 in its original form to 7275 lines in staging).
Some people have expressed concerns about my plans to _temporarily_
carry patches for the vboxguest integration in the Fedora kernel pkg.
I can understand that you are reluctant to carry patches which
need to be maintained for ever and ever, but that is not the case
I've been working my ass of to get these drivers cleaned up
in time for F27, so that the _cleaned up_ version can be added
to the Fedora kernels for a kernel release or 2, see:
for all the work I've been oing on the vboxguest driver.
As mentioned on the fedora-devel list I did not contact the kernel
team before because I first wanted to have something to show and
something better then just dropping the vbox out-of-tree drivers
into the kernel as is.
The other 2 drivers needed for vboxguest integration are the vboxguest
and vboxsf drivers.
I now have the vboxguest driver ready for upstream submission,
reducing it from 100587 lines of code in /usr/src/vboxguest-5.1.24/vboxguest
to just 6324 lines for the patch I'm going to submit upstream:
I plan to also have the vboxsf driver ready coming Friday and to submit
both upstream then.
As you can see from the above link the vboxguest driver is not big and
is completely self contained, without needing changes anywhere else
in the kernel. The vboxsf driver is the same. As such carrying these
2 drivers as patches should cause very little work and I will maintain
them both while they are patched in and afterwards.
So I'm hereby asking the Fedora kernel-team for permission to add
these 2 drivers as patches to the Fedora kernel for a kernel-release
or 2 while I finish pushing them upstream.