mbox series

[v4,0/4] Add a vhost RPMsg API

Message ID 20200722150927.15587-1-guennadi.liakhovetski@linux.intel.com (mailing list archive)
Headers show
Series Add a vhost RPMsg API | expand

Message

Guennadi Liakhovetski July 22, 2020, 3:09 p.m. UTC
Hi,

Now that virtio-rpmsg endianness fixes have been merged we can 
proceed with the next step.

v4:
- add endianness conversions to comply with the VirtIO standard

v3:
- address several checkpatch warnings
- address comments from Mathieu Poirier

v2:
- update patch #5 with a correct vhost_dev_init() prototype
- drop patch #6 - it depends on a different patch, that is currently
  an RFC
- address comments from Pierre-Louis Bossart:
  * remove "default n" from Kconfig

Linux supports RPMsg over VirtIO for "remote processor" / AMP use
cases. It can however also be used for virtualisation scenarios,
e.g. when using KVM to run Linux on both the host and the guests.
This patch set adds a wrapper API to facilitate writing vhost
drivers for such RPMsg-based solutions. The first use case is an
audio DSP virtualisation project, currently under development, ready
for review and submission, available at
https://github.com/thesofproject/linux/pull/1501/commits

Thanks
Guennadi

Guennadi Liakhovetski (4):
  vhost: convert VHOST_VSOCK_SET_RUNNING to a generic ioctl
  rpmsg: move common structures and defines to headers
  rpmsg: update documentation
  vhost: add an RPMsg API

 Documentation/rpmsg.txt          |   6 +-
 drivers/rpmsg/virtio_rpmsg_bus.c |  78 +------
 drivers/vhost/Kconfig            |   7 +
 drivers/vhost/Makefile           |   3 +
 drivers/vhost/rpmsg.c            | 375 +++++++++++++++++++++++++++++++
 drivers/vhost/vhost_rpmsg.h      |  74 ++++++
 include/linux/virtio_rpmsg.h     |  83 +++++++
 include/uapi/linux/rpmsg.h       |   3 +
 include/uapi/linux/vhost.h       |   4 +-
 9 files changed, 553 insertions(+), 80 deletions(-)
 create mode 100644 drivers/vhost/rpmsg.c
 create mode 100644 drivers/vhost/vhost_rpmsg.h
 create mode 100644 include/linux/virtio_rpmsg.h

Comments

Michael S. Tsirkin July 30, 2020, 4:08 p.m. UTC | #1
On Wed, Jul 22, 2020 at 05:09:23PM +0200, Guennadi Liakhovetski wrote:
> Hi,
> 
> Now that virtio-rpmsg endianness fixes have been merged we can 
> proceed with the next step.

Which tree is this for?


> v4:
> - add endianness conversions to comply with the VirtIO standard
> 
> v3:
> - address several checkpatch warnings
> - address comments from Mathieu Poirier
> 
> v2:
> - update patch #5 with a correct vhost_dev_init() prototype
> - drop patch #6 - it depends on a different patch, that is currently
>   an RFC
> - address comments from Pierre-Louis Bossart:
>   * remove "default n" from Kconfig
> 
> Linux supports RPMsg over VirtIO for "remote processor" / AMP use
> cases. It can however also be used for virtualisation scenarios,
> e.g. when using KVM to run Linux on both the host and the guests.
> This patch set adds a wrapper API to facilitate writing vhost
> drivers for such RPMsg-based solutions. The first use case is an
> audio DSP virtualisation project, currently under development, ready
> for review and submission, available at
> https://github.com/thesofproject/linux/pull/1501/commits
> 
> Thanks
> Guennadi
> 
> Guennadi Liakhovetski (4):
>   vhost: convert VHOST_VSOCK_SET_RUNNING to a generic ioctl
>   rpmsg: move common structures and defines to headers
>   rpmsg: update documentation
>   vhost: add an RPMsg API
> 
>  Documentation/rpmsg.txt          |   6 +-
>  drivers/rpmsg/virtio_rpmsg_bus.c |  78 +------
>  drivers/vhost/Kconfig            |   7 +
>  drivers/vhost/Makefile           |   3 +
>  drivers/vhost/rpmsg.c            | 375 +++++++++++++++++++++++++++++++
>  drivers/vhost/vhost_rpmsg.h      |  74 ++++++
>  include/linux/virtio_rpmsg.h     |  83 +++++++
>  include/uapi/linux/rpmsg.h       |   3 +
>  include/uapi/linux/vhost.h       |   4 +-
>  9 files changed, 553 insertions(+), 80 deletions(-)
>  create mode 100644 drivers/vhost/rpmsg.c
>  create mode 100644 drivers/vhost/vhost_rpmsg.h
>  create mode 100644 include/linux/virtio_rpmsg.h
> 
> -- 
> 2.27.0
Guennadi Liakhovetski July 31, 2020, 5:47 a.m. UTC | #2
Hi Michael,

On Thu, Jul 30, 2020 at 12:08:29PM -0400, Michael S. Tsirkin wrote:
> On Wed, Jul 22, 2020 at 05:09:23PM +0200, Guennadi Liakhovetski wrote:
> > Hi,
> > 
> > Now that virtio-rpmsg endianness fixes have been merged we can 
> > proceed with the next step.
> 
> Which tree is this for?

The essential part of this series is for drivers/vhost, so, I presume 
that should be the target tree as well. There is however a small part 
for the drivers/rpmsg, should I split this series in two or shall we 
first review is as a whole to make its goals clearer?

Thanks
Guennadi

> > v4:
> > - add endianness conversions to comply with the VirtIO standard
> > 
> > v3:
> > - address several checkpatch warnings
> > - address comments from Mathieu Poirier
> > 
> > v2:
> > - update patch #5 with a correct vhost_dev_init() prototype
> > - drop patch #6 - it depends on a different patch, that is currently
> >   an RFC
> > - address comments from Pierre-Louis Bossart:
> >   * remove "default n" from Kconfig
> > 
> > Linux supports RPMsg over VirtIO for "remote processor" / AMP use
> > cases. It can however also be used for virtualisation scenarios,
> > e.g. when using KVM to run Linux on both the host and the guests.
> > This patch set adds a wrapper API to facilitate writing vhost
> > drivers for such RPMsg-based solutions. The first use case is an
> > audio DSP virtualisation project, currently under development, ready
> > for review and submission, available at
> > https://github.com/thesofproject/linux/pull/1501/commits
> > 
> > Thanks
> > Guennadi
> > 
> > Guennadi Liakhovetski (4):
> >   vhost: convert VHOST_VSOCK_SET_RUNNING to a generic ioctl
> >   rpmsg: move common structures and defines to headers
> >   rpmsg: update documentation
> >   vhost: add an RPMsg API
> > 
> >  Documentation/rpmsg.txt          |   6 +-
> >  drivers/rpmsg/virtio_rpmsg_bus.c |  78 +------
> >  drivers/vhost/Kconfig            |   7 +
> >  drivers/vhost/Makefile           |   3 +
> >  drivers/vhost/rpmsg.c            | 375 +++++++++++++++++++++++++++++++
> >  drivers/vhost/vhost_rpmsg.h      |  74 ++++++
> >  include/linux/virtio_rpmsg.h     |  83 +++++++
> >  include/uapi/linux/rpmsg.h       |   3 +
> >  include/uapi/linux/vhost.h       |   4 +-
> >  9 files changed, 553 insertions(+), 80 deletions(-)
> >  create mode 100644 drivers/vhost/rpmsg.c
> >  create mode 100644 drivers/vhost/vhost_rpmsg.h
> >  create mode 100644 include/linux/virtio_rpmsg.h
> > 
> > -- 
> > 2.27.0
>
Mathieu Poirier Aug. 3, 2020, 1:25 p.m. UTC | #3
On Thu, 30 Jul 2020 at 23:47, Guennadi Liakhovetski
<guennadi.liakhovetski@linux.intel.com> wrote:
>
> Hi Michael,
>
> On Thu, Jul 30, 2020 at 12:08:29PM -0400, Michael S. Tsirkin wrote:
> > On Wed, Jul 22, 2020 at 05:09:23PM +0200, Guennadi Liakhovetski wrote:
> > > Hi,
> > >
> > > Now that virtio-rpmsg endianness fixes have been merged we can
> > > proceed with the next step.
> >
> > Which tree is this for?
>
> The essential part of this series is for drivers/vhost, so, I presume
> that should be the target tree as well. There is however a small part
> for the drivers/rpmsg, should I split this series in two or shall we
> first review is as a whole to make its goals clearer?

I suggest to keep it whole for now.

>
> Thanks
> Guennadi
>
> > > v4:
> > > - add endianness conversions to comply with the VirtIO standard
> > >
> > > v3:
> > > - address several checkpatch warnings
> > > - address comments from Mathieu Poirier
> > >
> > > v2:
> > > - update patch #5 with a correct vhost_dev_init() prototype
> > > - drop patch #6 - it depends on a different patch, that is currently
> > >   an RFC
> > > - address comments from Pierre-Louis Bossart:
> > >   * remove "default n" from Kconfig
> > >
> > > Linux supports RPMsg over VirtIO for "remote processor" / AMP use
> > > cases. It can however also be used for virtualisation scenarios,
> > > e.g. when using KVM to run Linux on both the host and the guests.
> > > This patch set adds a wrapper API to facilitate writing vhost
> > > drivers for such RPMsg-based solutions. The first use case is an
> > > audio DSP virtualisation project, currently under development, ready
> > > for review and submission, available at
> > > https://github.com/thesofproject/linux/pull/1501/commits
> > >
> > > Thanks
> > > Guennadi
> > >
> > > Guennadi Liakhovetski (4):
> > >   vhost: convert VHOST_VSOCK_SET_RUNNING to a generic ioctl
> > >   rpmsg: move common structures and defines to headers
> > >   rpmsg: update documentation
> > >   vhost: add an RPMsg API
> > >
> > >  Documentation/rpmsg.txt          |   6 +-
> > >  drivers/rpmsg/virtio_rpmsg_bus.c |  78 +------
> > >  drivers/vhost/Kconfig            |   7 +
> > >  drivers/vhost/Makefile           |   3 +
> > >  drivers/vhost/rpmsg.c            | 375 +++++++++++++++++++++++++++++++
> > >  drivers/vhost/vhost_rpmsg.h      |  74 ++++++
> > >  include/linux/virtio_rpmsg.h     |  83 +++++++
> > >  include/uapi/linux/rpmsg.h       |   3 +
> > >  include/uapi/linux/vhost.h       |   4 +-
> > >  9 files changed, 553 insertions(+), 80 deletions(-)
> > >  create mode 100644 drivers/vhost/rpmsg.c
> > >  create mode 100644 drivers/vhost/vhost_rpmsg.h
> > >  create mode 100644 include/linux/virtio_rpmsg.h
> > >
> > > --
> > > 2.27.0
> >
Michael S. Tsirkin Aug. 3, 2020, 8:46 p.m. UTC | #4
On Mon, Aug 03, 2020 at 07:25:24AM -0600, Mathieu Poirier wrote:
> On Thu, 30 Jul 2020 at 23:47, Guennadi Liakhovetski
> <guennadi.liakhovetski@linux.intel.com> wrote:
> >
> > Hi Michael,
> >
> > On Thu, Jul 30, 2020 at 12:08:29PM -0400, Michael S. Tsirkin wrote:
> > > On Wed, Jul 22, 2020 at 05:09:23PM +0200, Guennadi Liakhovetski wrote:
> > > > Hi,
> > > >
> > > > Now that virtio-rpmsg endianness fixes have been merged we can
> > > > proceed with the next step.
> > >
> > > Which tree is this for?
> >
> > The essential part of this series is for drivers/vhost, so, I presume
> > that should be the target tree as well. There is however a small part
> > for the drivers/rpmsg, should I split this series in two or shall we
> > first review is as a whole to make its goals clearer?
> 
> I suggest to keep it whole for now.


Ok can I get some acks please?
Also, I put this in my linux-next branch on 

https://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost.git

there were some conflicts - could you pls test and report it's ok?

> >
> > Thanks
> > Guennadi
> >
> > > > v4:
> > > > - add endianness conversions to comply with the VirtIO standard
> > > >
> > > > v3:
> > > > - address several checkpatch warnings
> > > > - address comments from Mathieu Poirier
> > > >
> > > > v2:
> > > > - update patch #5 with a correct vhost_dev_init() prototype
> > > > - drop patch #6 - it depends on a different patch, that is currently
> > > >   an RFC
> > > > - address comments from Pierre-Louis Bossart:
> > > >   * remove "default n" from Kconfig
> > > >
> > > > Linux supports RPMsg over VirtIO for "remote processor" / AMP use
> > > > cases. It can however also be used for virtualisation scenarios,
> > > > e.g. when using KVM to run Linux on both the host and the guests.
> > > > This patch set adds a wrapper API to facilitate writing vhost
> > > > drivers for such RPMsg-based solutions. The first use case is an
> > > > audio DSP virtualisation project, currently under development, ready
> > > > for review and submission, available at
> > > > https://github.com/thesofproject/linux/pull/1501/commits
> > > >
> > > > Thanks
> > > > Guennadi
> > > >
> > > > Guennadi Liakhovetski (4):
> > > >   vhost: convert VHOST_VSOCK_SET_RUNNING to a generic ioctl
> > > >   rpmsg: move common structures and defines to headers
> > > >   rpmsg: update documentation
> > > >   vhost: add an RPMsg API
> > > >
> > > >  Documentation/rpmsg.txt          |   6 +-
> > > >  drivers/rpmsg/virtio_rpmsg_bus.c |  78 +------
> > > >  drivers/vhost/Kconfig            |   7 +
> > > >  drivers/vhost/Makefile           |   3 +
> > > >  drivers/vhost/rpmsg.c            | 375 +++++++++++++++++++++++++++++++
> > > >  drivers/vhost/vhost_rpmsg.h      |  74 ++++++
> > > >  include/linux/virtio_rpmsg.h     |  83 +++++++
> > > >  include/uapi/linux/rpmsg.h       |   3 +
> > > >  include/uapi/linux/vhost.h       |   4 +-
> > > >  9 files changed, 553 insertions(+), 80 deletions(-)
> > > >  create mode 100644 drivers/vhost/rpmsg.c
> > > >  create mode 100644 drivers/vhost/vhost_rpmsg.h
> > > >  create mode 100644 include/linux/virtio_rpmsg.h
> > > >
> > > > --
> > > > 2.27.0
> > >
Michael S. Tsirkin Aug. 4, 2020, 12:26 p.m. UTC | #5
On Wed, Jul 22, 2020 at 05:09:23PM +0200, Guennadi Liakhovetski wrote:
> Hi,
> 
> Now that virtio-rpmsg endianness fixes have been merged we can 
> proceed with the next step.

OK my attempts to resolve conflicts just created a mess.

I dropped these for now, could you pls rebase on top
of linux-next branch in my tree, and repost?
Thanks!


> v4:
> - add endianness conversions to comply with the VirtIO standard
> 
> v3:
> - address several checkpatch warnings
> - address comments from Mathieu Poirier
> 
> v2:
> - update patch #5 with a correct vhost_dev_init() prototype
> - drop patch #6 - it depends on a different patch, that is currently
>   an RFC
> - address comments from Pierre-Louis Bossart:
>   * remove "default n" from Kconfig
> 
> Linux supports RPMsg over VirtIO for "remote processor" / AMP use
> cases. It can however also be used for virtualisation scenarios,
> e.g. when using KVM to run Linux on both the host and the guests.
> This patch set adds a wrapper API to facilitate writing vhost
> drivers for such RPMsg-based solutions. The first use case is an
> audio DSP virtualisation project, currently under development, ready
> for review and submission, available at
> https://github.com/thesofproject/linux/pull/1501/commits
> 
> Thanks
> Guennadi
> 
> Guennadi Liakhovetski (4):
>   vhost: convert VHOST_VSOCK_SET_RUNNING to a generic ioctl
>   rpmsg: move common structures and defines to headers
>   rpmsg: update documentation
>   vhost: add an RPMsg API
> 
>  Documentation/rpmsg.txt          |   6 +-
>  drivers/rpmsg/virtio_rpmsg_bus.c |  78 +------
>  drivers/vhost/Kconfig            |   7 +
>  drivers/vhost/Makefile           |   3 +
>  drivers/vhost/rpmsg.c            | 375 +++++++++++++++++++++++++++++++
>  drivers/vhost/vhost_rpmsg.h      |  74 ++++++
>  include/linux/virtio_rpmsg.h     |  83 +++++++
>  include/uapi/linux/rpmsg.h       |   3 +
>  include/uapi/linux/vhost.h       |   4 +-
>  9 files changed, 553 insertions(+), 80 deletions(-)
>  create mode 100644 drivers/vhost/rpmsg.c
>  create mode 100644 drivers/vhost/vhost_rpmsg.h
>  create mode 100644 include/linux/virtio_rpmsg.h
> 
> -- 
> 2.27.0
Guennadi Liakhovetski Aug. 4, 2020, 1:19 p.m. UTC | #6
Hi Michael,

On Tue, Aug 04, 2020 at 08:26:53AM -0400, Michael S. Tsirkin wrote:
> On Wed, Jul 22, 2020 at 05:09:23PM +0200, Guennadi Liakhovetski wrote:
> > Hi,
> > 
> > Now that virtio-rpmsg endianness fixes have been merged we can 
> > proceed with the next step.
> 
> OK my attempts to resolve conflicts just created a mess.

You just need to apply my previous patch for virtio-rpmsg first 
https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/drivers/rpmsg/virtio_rpmsg_bus.c?id=111d1089700cdb752681ef44f54ab6137736f5c2
Then this series should apply cleanly.

Thanks
Guennadi

> I dropped these for now, could you pls rebase on top
> of linux-next branch in my tree, and repost?
> Thanks!
> 
> 
> > v4:
> > - add endianness conversions to comply with the VirtIO standard
> > 
> > v3:
> > - address several checkpatch warnings
> > - address comments from Mathieu Poirier
> > 
> > v2:
> > - update patch #5 with a correct vhost_dev_init() prototype
> > - drop patch #6 - it depends on a different patch, that is currently
> >   an RFC
> > - address comments from Pierre-Louis Bossart:
> >   * remove "default n" from Kconfig
> > 
> > Linux supports RPMsg over VirtIO for "remote processor" / AMP use
> > cases. It can however also be used for virtualisation scenarios,
> > e.g. when using KVM to run Linux on both the host and the guests.
> > This patch set adds a wrapper API to facilitate writing vhost
> > drivers for such RPMsg-based solutions. The first use case is an
> > audio DSP virtualisation project, currently under development, ready
> > for review and submission, available at
> > https://github.com/thesofproject/linux/pull/1501/commits
> > 
> > Thanks
> > Guennadi
> > 
> > Guennadi Liakhovetski (4):
> >   vhost: convert VHOST_VSOCK_SET_RUNNING to a generic ioctl
> >   rpmsg: move common structures and defines to headers
> >   rpmsg: update documentation
> >   vhost: add an RPMsg API
> > 
> >  Documentation/rpmsg.txt          |   6 +-
> >  drivers/rpmsg/virtio_rpmsg_bus.c |  78 +------
> >  drivers/vhost/Kconfig            |   7 +
> >  drivers/vhost/Makefile           |   3 +
> >  drivers/vhost/rpmsg.c            | 375 +++++++++++++++++++++++++++++++
> >  drivers/vhost/vhost_rpmsg.h      |  74 ++++++
> >  include/linux/virtio_rpmsg.h     |  83 +++++++
> >  include/uapi/linux/rpmsg.h       |   3 +
> >  include/uapi/linux/vhost.h       |   4 +-
> >  9 files changed, 553 insertions(+), 80 deletions(-)
> >  create mode 100644 drivers/vhost/rpmsg.c
> >  create mode 100644 drivers/vhost/vhost_rpmsg.h
> >  create mode 100644 include/linux/virtio_rpmsg.h
> > 
> > -- 
> > 2.27.0
>
Mathieu Poirier Aug. 4, 2020, 1:37 p.m. UTC | #7
On Mon, 3 Aug 2020 at 14:47, Michael S. Tsirkin <mst@redhat.com> wrote:
>
> On Mon, Aug 03, 2020 at 07:25:24AM -0600, Mathieu Poirier wrote:
> > On Thu, 30 Jul 2020 at 23:47, Guennadi Liakhovetski
> > <guennadi.liakhovetski@linux.intel.com> wrote:
> > >
> > > Hi Michael,
> > >
> > > On Thu, Jul 30, 2020 at 12:08:29PM -0400, Michael S. Tsirkin wrote:
> > > > On Wed, Jul 22, 2020 at 05:09:23PM +0200, Guennadi Liakhovetski wrote:
> > > > > Hi,
> > > > >
> > > > > Now that virtio-rpmsg endianness fixes have been merged we can
> > > > > proceed with the next step.
> > > >
> > > > Which tree is this for?
> > >
> > > The essential part of this series is for drivers/vhost, so, I presume
> > > that should be the target tree as well. There is however a small part
> > > for the drivers/rpmsg, should I split this series in two or shall we
> > > first review is as a whole to make its goals clearer?
> >
> > I suggest to keep it whole for now.
>
>
> Ok can I get some acks please?

Yes, as soon as I have the opportunity to review the work.  There is a
lot of volume on the linux-remoteproc mailing list lately and
patchsets are reviewed in the order they have been received.

> Also, I put this in my linux-next branch on
>
> https://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost.git
>
> there were some conflicts - could you pls test and report it's ok?
>
> > >
> > > Thanks
> > > Guennadi
> > >
> > > > > v4:
> > > > > - add endianness conversions to comply with the VirtIO standard
> > > > >
> > > > > v3:
> > > > > - address several checkpatch warnings
> > > > > - address comments from Mathieu Poirier
> > > > >
> > > > > v2:
> > > > > - update patch #5 with a correct vhost_dev_init() prototype
> > > > > - drop patch #6 - it depends on a different patch, that is currently
> > > > >   an RFC
> > > > > - address comments from Pierre-Louis Bossart:
> > > > >   * remove "default n" from Kconfig
> > > > >
> > > > > Linux supports RPMsg over VirtIO for "remote processor" / AMP use
> > > > > cases. It can however also be used for virtualisation scenarios,
> > > > > e.g. when using KVM to run Linux on both the host and the guests.
> > > > > This patch set adds a wrapper API to facilitate writing vhost
> > > > > drivers for such RPMsg-based solutions. The first use case is an
> > > > > audio DSP virtualisation project, currently under development, ready
> > > > > for review and submission, available at
> > > > > https://github.com/thesofproject/linux/pull/1501/commits
> > > > >
> > > > > Thanks
> > > > > Guennadi
> > > > >
> > > > > Guennadi Liakhovetski (4):
> > > > >   vhost: convert VHOST_VSOCK_SET_RUNNING to a generic ioctl
> > > > >   rpmsg: move common structures and defines to headers
> > > > >   rpmsg: update documentation
> > > > >   vhost: add an RPMsg API
> > > > >
> > > > >  Documentation/rpmsg.txt          |   6 +-
> > > > >  drivers/rpmsg/virtio_rpmsg_bus.c |  78 +------
> > > > >  drivers/vhost/Kconfig            |   7 +
> > > > >  drivers/vhost/Makefile           |   3 +
> > > > >  drivers/vhost/rpmsg.c            | 375 +++++++++++++++++++++++++++++++
> > > > >  drivers/vhost/vhost_rpmsg.h      |  74 ++++++
> > > > >  include/linux/virtio_rpmsg.h     |  83 +++++++
> > > > >  include/uapi/linux/rpmsg.h       |   3 +
> > > > >  include/uapi/linux/vhost.h       |   4 +-
> > > > >  9 files changed, 553 insertions(+), 80 deletions(-)
> > > > >  create mode 100644 drivers/vhost/rpmsg.c
> > > > >  create mode 100644 drivers/vhost/vhost_rpmsg.h
> > > > >  create mode 100644 include/linux/virtio_rpmsg.h
> > > > >
> > > > > --
> > > > > 2.27.0
> > > >
>
Michael S. Tsirkin Aug. 4, 2020, 2:07 p.m. UTC | #8
On Tue, Aug 04, 2020 at 07:37:49AM -0600, Mathieu Poirier wrote:
> On Mon, 3 Aug 2020 at 14:47, Michael S. Tsirkin <mst@redhat.com> wrote:
> >
> > On Mon, Aug 03, 2020 at 07:25:24AM -0600, Mathieu Poirier wrote:
> > > On Thu, 30 Jul 2020 at 23:47, Guennadi Liakhovetski
> > > <guennadi.liakhovetski@linux.intel.com> wrote:
> > > >
> > > > Hi Michael,
> > > >
> > > > On Thu, Jul 30, 2020 at 12:08:29PM -0400, Michael S. Tsirkin wrote:
> > > > > On Wed, Jul 22, 2020 at 05:09:23PM +0200, Guennadi Liakhovetski wrote:
> > > > > > Hi,
> > > > > >
> > > > > > Now that virtio-rpmsg endianness fixes have been merged we can
> > > > > > proceed with the next step.
> > > > >
> > > > > Which tree is this for?
> > > >
> > > > The essential part of this series is for drivers/vhost, so, I presume
> > > > that should be the target tree as well. There is however a small part
> > > > for the drivers/rpmsg, should I split this series in two or shall we
> > > > first review is as a whole to make its goals clearer?
> > >
> > > I suggest to keep it whole for now.
> >
> >
> > Ok can I get some acks please?
> 
> Yes, as soon as I have the opportunity to review the work.  There is a
> lot of volume on the linux-remoteproc mailing list lately and
> patchsets are reviewed in the order they have been received.

Well the merge window is open, I guess I'll merge this and
any issues can be addressed later then?

> > Also, I put this in my linux-next branch on
> >
> > https://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost.git
> >
> > there were some conflicts - could you pls test and report it's ok?
> >
> > > >
> > > > Thanks
> > > > Guennadi
> > > >
> > > > > > v4:
> > > > > > - add endianness conversions to comply with the VirtIO standard
> > > > > >
> > > > > > v3:
> > > > > > - address several checkpatch warnings
> > > > > > - address comments from Mathieu Poirier
> > > > > >
> > > > > > v2:
> > > > > > - update patch #5 with a correct vhost_dev_init() prototype
> > > > > > - drop patch #6 - it depends on a different patch, that is currently
> > > > > >   an RFC
> > > > > > - address comments from Pierre-Louis Bossart:
> > > > > >   * remove "default n" from Kconfig
> > > > > >
> > > > > > Linux supports RPMsg over VirtIO for "remote processor" / AMP use
> > > > > > cases. It can however also be used for virtualisation scenarios,
> > > > > > e.g. when using KVM to run Linux on both the host and the guests.
> > > > > > This patch set adds a wrapper API to facilitate writing vhost
> > > > > > drivers for such RPMsg-based solutions. The first use case is an
> > > > > > audio DSP virtualisation project, currently under development, ready
> > > > > > for review and submission, available at
> > > > > > https://github.com/thesofproject/linux/pull/1501/commits
> > > > > >
> > > > > > Thanks
> > > > > > Guennadi
> > > > > >
> > > > > > Guennadi Liakhovetski (4):
> > > > > >   vhost: convert VHOST_VSOCK_SET_RUNNING to a generic ioctl
> > > > > >   rpmsg: move common structures and defines to headers
> > > > > >   rpmsg: update documentation
> > > > > >   vhost: add an RPMsg API
> > > > > >
> > > > > >  Documentation/rpmsg.txt          |   6 +-
> > > > > >  drivers/rpmsg/virtio_rpmsg_bus.c |  78 +------
> > > > > >  drivers/vhost/Kconfig            |   7 +
> > > > > >  drivers/vhost/Makefile           |   3 +
> > > > > >  drivers/vhost/rpmsg.c            | 375 +++++++++++++++++++++++++++++++
> > > > > >  drivers/vhost/vhost_rpmsg.h      |  74 ++++++
> > > > > >  include/linux/virtio_rpmsg.h     |  83 +++++++
> > > > > >  include/uapi/linux/rpmsg.h       |   3 +
> > > > > >  include/uapi/linux/vhost.h       |   4 +-
> > > > > >  9 files changed, 553 insertions(+), 80 deletions(-)
> > > > > >  create mode 100644 drivers/vhost/rpmsg.c
> > > > > >  create mode 100644 drivers/vhost/vhost_rpmsg.h
> > > > > >  create mode 100644 include/linux/virtio_rpmsg.h
> > > > > >
> > > > > > --
> > > > > > 2.27.0
> > > > >
> >
Michael S. Tsirkin Aug. 4, 2020, 2:10 p.m. UTC | #9
On Tue, Aug 04, 2020 at 03:19:19PM +0200, Guennadi Liakhovetski wrote:
> Hi Michael,
> 
> On Tue, Aug 04, 2020 at 08:26:53AM -0400, Michael S. Tsirkin wrote:
> > On Wed, Jul 22, 2020 at 05:09:23PM +0200, Guennadi Liakhovetski wrote:
> > > Hi,
> > > 
> > > Now that virtio-rpmsg endianness fixes have been merged we can 
> > > proceed with the next step.
> > 
> > OK my attempts to resolve conflicts just created a mess.
> 
> You just need to apply my previous patch for virtio-rpmsg first 
> https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/drivers/rpmsg/virtio_rpmsg_bus.c?id=111d1089700cdb752681ef44f54ab6137736f5c2
> Then this series should apply cleanly.
> 
> Thanks
> Guennadi

Hmm.  Could you test the vhost branch in my tree, and tell me if that looks
good to you?

> > I dropped these for now, could you pls rebase on top
> > of linux-next branch in my tree, and repost?
> > Thanks!
> > 
> > 
> > > v4:
> > > - add endianness conversions to comply with the VirtIO standard
> > > 
> > > v3:
> > > - address several checkpatch warnings
> > > - address comments from Mathieu Poirier
> > > 
> > > v2:
> > > - update patch #5 with a correct vhost_dev_init() prototype
> > > - drop patch #6 - it depends on a different patch, that is currently
> > >   an RFC
> > > - address comments from Pierre-Louis Bossart:
> > >   * remove "default n" from Kconfig
> > > 
> > > Linux supports RPMsg over VirtIO for "remote processor" / AMP use
> > > cases. It can however also be used for virtualisation scenarios,
> > > e.g. when using KVM to run Linux on both the host and the guests.
> > > This patch set adds a wrapper API to facilitate writing vhost
> > > drivers for such RPMsg-based solutions. The first use case is an
> > > audio DSP virtualisation project, currently under development, ready
> > > for review and submission, available at
> > > https://github.com/thesofproject/linux/pull/1501/commits
> > > 
> > > Thanks
> > > Guennadi
> > > 
> > > Guennadi Liakhovetski (4):
> > >   vhost: convert VHOST_VSOCK_SET_RUNNING to a generic ioctl
> > >   rpmsg: move common structures and defines to headers
> > >   rpmsg: update documentation
> > >   vhost: add an RPMsg API
> > > 
> > >  Documentation/rpmsg.txt          |   6 +-
> > >  drivers/rpmsg/virtio_rpmsg_bus.c |  78 +------
> > >  drivers/vhost/Kconfig            |   7 +
> > >  drivers/vhost/Makefile           |   3 +
> > >  drivers/vhost/rpmsg.c            | 375 +++++++++++++++++++++++++++++++
> > >  drivers/vhost/vhost_rpmsg.h      |  74 ++++++
> > >  include/linux/virtio_rpmsg.h     |  83 +++++++
> > >  include/uapi/linux/rpmsg.h       |   3 +
> > >  include/uapi/linux/vhost.h       |   4 +-
> > >  9 files changed, 553 insertions(+), 80 deletions(-)
> > >  create mode 100644 drivers/vhost/rpmsg.c
> > >  create mode 100644 drivers/vhost/vhost_rpmsg.h
> > >  create mode 100644 include/linux/virtio_rpmsg.h
> > > 
> > > -- 
> > > 2.27.0
> >
Guennadi Liakhovetski Aug. 4, 2020, 2:46 p.m. UTC | #10
On Tue, Aug 04, 2020 at 10:10:23AM -0400, Michael S. Tsirkin wrote:
> On Tue, Aug 04, 2020 at 03:19:19PM +0200, Guennadi Liakhovetski wrote:
> > Hi Michael,
> > 
> > On Tue, Aug 04, 2020 at 08:26:53AM -0400, Michael S. Tsirkin wrote:
> > > On Wed, Jul 22, 2020 at 05:09:23PM +0200, Guennadi Liakhovetski wrote:
> > > > Hi,
> > > > 
> > > > Now that virtio-rpmsg endianness fixes have been merged we can 
> > > > proceed with the next step.
> > > 
> > > OK my attempts to resolve conflicts just created a mess.
> > 
> > You just need to apply my previous patch for virtio-rpmsg first 
> > https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/drivers/rpmsg/virtio_rpmsg_bus.c?id=111d1089700cdb752681ef44f54ab6137736f5c2
> > Then this series should apply cleanly.
> > 
> > Thanks
> > Guennadi
> 
> Hmm.  Could you test the vhost branch in my tree, and tell me if that looks
> good to you?

Sorry, I'm not sure I understand why you're trying to resolve conflicts 
manually. My previous patch is already in "next," if you don't pull from 
"next" you will have a conflict when pushing to it. What am I missing?

Thanks
Guennadi

> > > I dropped these for now, could you pls rebase on top
> > > of linux-next branch in my tree, and repost?
> > > Thanks!
> > > 
> > > 
> > > > v4:
> > > > - add endianness conversions to comply with the VirtIO standard
> > > > 
> > > > v3:
> > > > - address several checkpatch warnings
> > > > - address comments from Mathieu Poirier
> > > > 
> > > > v2:
> > > > - update patch #5 with a correct vhost_dev_init() prototype
> > > > - drop patch #6 - it depends on a different patch, that is currently
> > > >   an RFC
> > > > - address comments from Pierre-Louis Bossart:
> > > >   * remove "default n" from Kconfig
> > > > 
> > > > Linux supports RPMsg over VirtIO for "remote processor" / AMP use
> > > > cases. It can however also be used for virtualisation scenarios,
> > > > e.g. when using KVM to run Linux on both the host and the guests.
> > > > This patch set adds a wrapper API to facilitate writing vhost
> > > > drivers for such RPMsg-based solutions. The first use case is an
> > > > audio DSP virtualisation project, currently under development, ready
> > > > for review and submission, available at
> > > > https://github.com/thesofproject/linux/pull/1501/commits
> > > > 
> > > > Thanks
> > > > Guennadi
> > > > 
> > > > Guennadi Liakhovetski (4):
> > > >   vhost: convert VHOST_VSOCK_SET_RUNNING to a generic ioctl
> > > >   rpmsg: move common structures and defines to headers
> > > >   rpmsg: update documentation
> > > >   vhost: add an RPMsg API
> > > > 
> > > >  Documentation/rpmsg.txt          |   6 +-
> > > >  drivers/rpmsg/virtio_rpmsg_bus.c |  78 +------
> > > >  drivers/vhost/Kconfig            |   7 +
> > > >  drivers/vhost/Makefile           |   3 +
> > > >  drivers/vhost/rpmsg.c            | 375 +++++++++++++++++++++++++++++++
> > > >  drivers/vhost/vhost_rpmsg.h      |  74 ++++++
> > > >  include/linux/virtio_rpmsg.h     |  83 +++++++
> > > >  include/uapi/linux/rpmsg.h       |   3 +
> > > >  include/uapi/linux/vhost.h       |   4 +-
> > > >  9 files changed, 553 insertions(+), 80 deletions(-)
> > > >  create mode 100644 drivers/vhost/rpmsg.c
> > > >  create mode 100644 drivers/vhost/vhost_rpmsg.h
> > > >  create mode 100644 include/linux/virtio_rpmsg.h
> > > > 
> > > > -- 
> > > > 2.27.0
> > > 
>
Mathieu Poirier Aug. 4, 2020, 7:30 p.m. UTC | #11
On Tue, 4 Aug 2020 at 08:07, Michael S. Tsirkin <mst@redhat.com> wrote:
>
> On Tue, Aug 04, 2020 at 07:37:49AM -0600, Mathieu Poirier wrote:
> > On Mon, 3 Aug 2020 at 14:47, Michael S. Tsirkin <mst@redhat.com> wrote:
> > >
> > > On Mon, Aug 03, 2020 at 07:25:24AM -0600, Mathieu Poirier wrote:
> > > > On Thu, 30 Jul 2020 at 23:47, Guennadi Liakhovetski
> > > > <guennadi.liakhovetski@linux.intel.com> wrote:
> > > > >
> > > > > Hi Michael,
> > > > >
> > > > > On Thu, Jul 30, 2020 at 12:08:29PM -0400, Michael S. Tsirkin wrote:
> > > > > > On Wed, Jul 22, 2020 at 05:09:23PM +0200, Guennadi Liakhovetski wrote:
> > > > > > > Hi,
> > > > > > >
> > > > > > > Now that virtio-rpmsg endianness fixes have been merged we can
> > > > > > > proceed with the next step.
> > > > > >
> > > > > > Which tree is this for?
> > > > >
> > > > > The essential part of this series is for drivers/vhost, so, I presume
> > > > > that should be the target tree as well. There is however a small part
> > > > > for the drivers/rpmsg, should I split this series in two or shall we
> > > > > first review is as a whole to make its goals clearer?
> > > >
> > > > I suggest to keep it whole for now.
> > >
> > >
> > > Ok can I get some acks please?
> >
> > Yes, as soon as I have the opportunity to review the work.  There is a
> > lot of volume on the linux-remoteproc mailing list lately and
> > patchsets are reviewed in the order they have been received.
>
> Well the merge window is open, I guess I'll merge this and
> any issues can be addressed later then?

Please don't do that.  I prefer to miss a merge window than impacting
upstream consumers.  This patch will be reviewed, just not in time for
this merge window.

>
> > > Also, I put this in my linux-next branch on
> > >
> > > https://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost.git
> > >
> > > there were some conflicts - could you pls test and report it's ok?
> > >
> > > > >
> > > > > Thanks
> > > > > Guennadi
> > > > >
> > > > > > > v4:
> > > > > > > - add endianness conversions to comply with the VirtIO standard
> > > > > > >
> > > > > > > v3:
> > > > > > > - address several checkpatch warnings
> > > > > > > - address comments from Mathieu Poirier
> > > > > > >
> > > > > > > v2:
> > > > > > > - update patch #5 with a correct vhost_dev_init() prototype
> > > > > > > - drop patch #6 - it depends on a different patch, that is currently
> > > > > > >   an RFC
> > > > > > > - address comments from Pierre-Louis Bossart:
> > > > > > >   * remove "default n" from Kconfig
> > > > > > >
> > > > > > > Linux supports RPMsg over VirtIO for "remote processor" / AMP use
> > > > > > > cases. It can however also be used for virtualisation scenarios,
> > > > > > > e.g. when using KVM to run Linux on both the host and the guests.
> > > > > > > This patch set adds a wrapper API to facilitate writing vhost
> > > > > > > drivers for such RPMsg-based solutions. The first use case is an
> > > > > > > audio DSP virtualisation project, currently under development, ready
> > > > > > > for review and submission, available at
> > > > > > > https://github.com/thesofproject/linux/pull/1501/commits
> > > > > > >
> > > > > > > Thanks
> > > > > > > Guennadi
> > > > > > >
> > > > > > > Guennadi Liakhovetski (4):
> > > > > > >   vhost: convert VHOST_VSOCK_SET_RUNNING to a generic ioctl
> > > > > > >   rpmsg: move common structures and defines to headers
> > > > > > >   rpmsg: update documentation
> > > > > > >   vhost: add an RPMsg API
> > > > > > >
> > > > > > >  Documentation/rpmsg.txt          |   6 +-
> > > > > > >  drivers/rpmsg/virtio_rpmsg_bus.c |  78 +------
> > > > > > >  drivers/vhost/Kconfig            |   7 +
> > > > > > >  drivers/vhost/Makefile           |   3 +
> > > > > > >  drivers/vhost/rpmsg.c            | 375 +++++++++++++++++++++++++++++++
> > > > > > >  drivers/vhost/vhost_rpmsg.h      |  74 ++++++
> > > > > > >  include/linux/virtio_rpmsg.h     |  83 +++++++
> > > > > > >  include/uapi/linux/rpmsg.h       |   3 +
> > > > > > >  include/uapi/linux/vhost.h       |   4 +-
> > > > > > >  9 files changed, 553 insertions(+), 80 deletions(-)
> > > > > > >  create mode 100644 drivers/vhost/rpmsg.c
> > > > > > >  create mode 100644 drivers/vhost/vhost_rpmsg.h
> > > > > > >  create mode 100644 include/linux/virtio_rpmsg.h
> > > > > > >
> > > > > > > --
> > > > > > > 2.27.0
> > > > > >
> > >
>
Michael S. Tsirkin Aug. 5, 2020, 11:34 a.m. UTC | #12
On Tue, Aug 04, 2020 at 01:30:32PM -0600, Mathieu Poirier wrote:
> On Tue, 4 Aug 2020 at 08:07, Michael S. Tsirkin <mst@redhat.com> wrote:
> >
> > On Tue, Aug 04, 2020 at 07:37:49AM -0600, Mathieu Poirier wrote:
> > > On Mon, 3 Aug 2020 at 14:47, Michael S. Tsirkin <mst@redhat.com> wrote:
> > > >
> > > > On Mon, Aug 03, 2020 at 07:25:24AM -0600, Mathieu Poirier wrote:
> > > > > On Thu, 30 Jul 2020 at 23:47, Guennadi Liakhovetski
> > > > > <guennadi.liakhovetski@linux.intel.com> wrote:
> > > > > >
> > > > > > Hi Michael,
> > > > > >
> > > > > > On Thu, Jul 30, 2020 at 12:08:29PM -0400, Michael S. Tsirkin wrote:
> > > > > > > On Wed, Jul 22, 2020 at 05:09:23PM +0200, Guennadi Liakhovetski wrote:
> > > > > > > > Hi,
> > > > > > > >
> > > > > > > > Now that virtio-rpmsg endianness fixes have been merged we can
> > > > > > > > proceed with the next step.
> > > > > > >
> > > > > > > Which tree is this for?
> > > > > >
> > > > > > The essential part of this series is for drivers/vhost, so, I presume
> > > > > > that should be the target tree as well. There is however a small part
> > > > > > for the drivers/rpmsg, should I split this series in two or shall we
> > > > > > first review is as a whole to make its goals clearer?
> > > > >
> > > > > I suggest to keep it whole for now.
> > > >
> > > >
> > > > Ok can I get some acks please?
> > >
> > > Yes, as soon as I have the opportunity to review the work.  There is a
> > > lot of volume on the linux-remoteproc mailing list lately and
> > > patchsets are reviewed in the order they have been received.
> >
> > Well the merge window is open, I guess I'll merge this and
> > any issues can be addressed later then?
> 
> Please don't do that.  I prefer to miss a merge window than impacting
> upstream consumers.  This patch will be reviewed, just not in time for
> this merge window.

OK then.


> >
> > > > Also, I put this in my linux-next branch on
> > > >
> > > > https://git.kernel.org/pub/scm/linux/kernel/git/mst/vhost.git
> > > >
> > > > there were some conflicts - could you pls test and report it's ok?
> > > >
> > > > > >
> > > > > > Thanks
> > > > > > Guennadi
> > > > > >
> > > > > > > > v4:
> > > > > > > > - add endianness conversions to comply with the VirtIO standard
> > > > > > > >
> > > > > > > > v3:
> > > > > > > > - address several checkpatch warnings
> > > > > > > > - address comments from Mathieu Poirier
> > > > > > > >
> > > > > > > > v2:
> > > > > > > > - update patch #5 with a correct vhost_dev_init() prototype
> > > > > > > > - drop patch #6 - it depends on a different patch, that is currently
> > > > > > > >   an RFC
> > > > > > > > - address comments from Pierre-Louis Bossart:
> > > > > > > >   * remove "default n" from Kconfig
> > > > > > > >
> > > > > > > > Linux supports RPMsg over VirtIO for "remote processor" / AMP use
> > > > > > > > cases. It can however also be used for virtualisation scenarios,
> > > > > > > > e.g. when using KVM to run Linux on both the host and the guests.
> > > > > > > > This patch set adds a wrapper API to facilitate writing vhost
> > > > > > > > drivers for such RPMsg-based solutions. The first use case is an
> > > > > > > > audio DSP virtualisation project, currently under development, ready
> > > > > > > > for review and submission, available at
> > > > > > > > https://github.com/thesofproject/linux/pull/1501/commits
> > > > > > > >
> > > > > > > > Thanks
> > > > > > > > Guennadi
> > > > > > > >
> > > > > > > > Guennadi Liakhovetski (4):
> > > > > > > >   vhost: convert VHOST_VSOCK_SET_RUNNING to a generic ioctl
> > > > > > > >   rpmsg: move common structures and defines to headers
> > > > > > > >   rpmsg: update documentation
> > > > > > > >   vhost: add an RPMsg API
> > > > > > > >
> > > > > > > >  Documentation/rpmsg.txt          |   6 +-
> > > > > > > >  drivers/rpmsg/virtio_rpmsg_bus.c |  78 +------
> > > > > > > >  drivers/vhost/Kconfig            |   7 +
> > > > > > > >  drivers/vhost/Makefile           |   3 +
> > > > > > > >  drivers/vhost/rpmsg.c            | 375 +++++++++++++++++++++++++++++++
> > > > > > > >  drivers/vhost/vhost_rpmsg.h      |  74 ++++++
> > > > > > > >  include/linux/virtio_rpmsg.h     |  83 +++++++
> > > > > > > >  include/uapi/linux/rpmsg.h       |   3 +
> > > > > > > >  include/uapi/linux/vhost.h       |   4 +-
> > > > > > > >  9 files changed, 553 insertions(+), 80 deletions(-)
> > > > > > > >  create mode 100644 drivers/vhost/rpmsg.c
> > > > > > > >  create mode 100644 drivers/vhost/vhost_rpmsg.h
> > > > > > > >  create mode 100644 include/linux/virtio_rpmsg.h
> > > > > > > >
> > > > > > > > --
> > > > > > > > 2.27.0
> > > > > > >
> > > >
> >