Re: [PATCH 4/4] dt-bindings: hv: Add dt-bindings for VMBus

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Fri, Jan 20, 2023 at 12:43:40PM +0100, Krzysztof Kozlowski wrote:
> On 17/01/2023 16:52, Saurabh Singh Sengar wrote:
> > On Tue, Jan 17, 2023 at 04:41:22PM +0100, Krzysztof Kozlowski wrote:
> >> On 17/01/2023 16:13, Saurabh Singh Sengar wrote:
> >>> On Mon, Jan 16, 2023 at 07:55:13PM +0100, Krzysztof Kozlowski wrote:
> >>>> On 16/01/2023 17:48, Saurabh Sengar wrote:
> >>>>> Add dt-bindings for Hyper-V VMBus
> >>>>>
> >>>>> Signed-off-by: Saurabh Sengar <ssengar@xxxxxxxxxxxxxxxxxxx>
> >>>>> ---
> >>>>>  .../devicetree/bindings/hv/msft,vmbus.yaml         | 34 ++++++++++++++++++++++
> >>>>
> >>>> Also, there is no "hv" hardware, so that's not correct location. If your
> >>>> bindings describe firmware, this should go to firmware. Otherwise, this
> >>>> does not look like suitable for DT. We do not describe software stuff in DT.
> >>>
> >>> VMBus is a virtual device this is simmilar to virtio. I can rename this folder to vmbus.
> >>>
> >>
> >> Then virtio directory. The directories are per subsystems (hardware
> >> classes).
> > 
> > Apologies if I was not clear, I meant to say this is a device conceptually
> > similar to virtio. But this driver has nothing to do with virtio, we should
> 
> Bindings are for hardware, not drivers, so if the device serves the same
> purpose, it's driver differences do not matter.
> 
> > be creating a new folder for it OR I am fine moving it under bus if that's
> > okay.
> 
> Since you do not have children here, it's not really a bus to fit under
> bus directory...
> 
> Probably this should go together with virtio bindings to dedicated
> hypervisor interfaces directory. We do not create directories for
> specific solutions (implementations) with only one or few bindings.
> Directories are for entire classes.

I am OK to keep it anywhere, but I believe virtio is not its correct place. I am also
concern how will the virtio maintainers will perceive it. Ideally we should be renaming
virtio to virtualization OR hypervisor OR something more generic where both virtio and
VMBus can co-exist. Please let me know if renaming virtio is acceptable.

> 
> Best regards,
> Krzysztof



[Index of Archives]     [Device Tree Compilter]     [Device Tree Spec]     [Linux Driver Backports]     [Video for Linux]     [Linux USB Devel]     [Linux PCI Devel]     [Linux Audio Users]     [Linux Kernel]     [Linux SCSI]     [XFree86]     [Yosemite Backpacking]


  Powered by Linux