[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: Re: [virtio-comment] Re: [PATCH] Introducing virtio-increment-edu.
On Sun, Jun 16, 2019 at 10:37:05PM -0400, Michael S. Tsirkin wrote: > On Sun, Jun 16, 2019 at 03:01:37PM +0300, Yoni Bettan wrote: > > The main goal is to create an educational example to be used as template or > > guideline for contributors when they wish to create a new virtio > > device and to document "the right way" to do so. > > > > The device functionality is simple as possible so contributors can have > > their focus on the virtio protocol rather on the device functionality to > > give them an "easy start". > > > > It consists of several parts: > > > > 1. The device specification > > * this patch content > > > > 2. The device implementation for Qemu-KVM hypervisor > > * it will hopefully be added to the Qemu project > > * for now it can be found at https://github.com/ybettan/qemu/blob/\ > > virtio/hw/virtio/virtio-example.c > > > > 3. The device driver for linux > > * it will hopefully be added to linux > > * for now it can be found at https://github.com/ybettan/\ > > QemuDeviceDrivers/blob/master/virtio/virtio_example_driver.c > > > > 4. A blog on virtio > > * introducing the virtio concept > > * gives some motivation for virtio-devices to be used > > * bring extra documentation on "how to write": > > - device specification > > - device implementation > > - device driver for linux > > * it can be found at https://howtovms.wordpress.com > > > > Signed-off-by: Yoni Bettan <ybettan@redhat.com> > > --- > > > > > > Here are some questions I would like to here your opinion on: > > > > * do you think such device should be included in the virtio-specification ? > > Not necessarily. But I think it's a good excercise and lessons > learned can be added to a spec section explaining how to do this > and what are the common mistakes to avoid. I agree. The ./newdevice.tex file can be extended based on lessons from this device. I think a virtio-net code commentary would be valuable, too. It's a real-world device with non-trivial features. Anyone who spends some time on that will be equipped to design new devices. Stefan
Attachment:
signature.asc
Description: PGP signature
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]