Hi Alex,
Thanks for reviewing.
On Tue, 11 Jan 2022, Alex Bennée wrote:
>
> Peter Griffin <peter.grif...@linaro.org> writes:
>
> > Signed-off-by: Peter Griffin <peter.grif...@linaro.org>
> > ---
> > hw/display/Kconfig | 5 +
> > hw/display/meson.build | 3 +
> > hw/display/vhost-user-video.c | 386 +++++++++++++++++++++++++++
> > include/hw/virtio/vhost-user-video.h | 41 +++
> > 4 files changed, 435 insertions(+)
> > create mode 100644 hw/display/vhost-user-video.c
> > create mode 100644 include/hw/virtio/vhost-user-video.h
> >
> > diff --git a/hw/display/Kconfig b/hw/display/Kconfig
> > index a2306b67d8..186163b015 100644
> > --- a/hw/display/Kconfig
> > +++ b/hw/display/Kconfig
> > @@ -118,6 +118,11 @@ config VHOST_USER_VGA
> > default y
> > depends on VIRTIO_VGA && VHOST_USER_GPU
> >
> > +config VHOST_USER_VIDEO
> > + bool
> > + default y
> > + depends on VIRTIO && VHOST_USER
> > +
> > config DPCD
> > bool
> > select AUX
> > diff --git a/hw/display/meson.build b/hw/display/meson.build
> > index 861c43ff98..48284528cf 100644
> > --- a/hw/display/meson.build
> > +++ b/hw/display/meson.build
> > @@ -37,6 +37,9 @@ softmmu_ss.add(when: 'CONFIG_MACFB', if_true:
> > files('macfb.c'))
> > softmmu_ss.add(when: 'CONFIG_NEXTCUBE', if_true: files('next-fb.c'))
> >
> > specific_ss.add(when: 'CONFIG_VGA', if_true: files('vga.c'))
> > +specific_ss.add(when: 'CONFIG_VHOST_USER_VIDEO', if_true:
> > files('vhost-user-video.c'))
> > +specific_ss.add(when: ['CONFIG_VHOST_USER_VIDEO', 'CONFIG_VIRTIO_PCI' ],
> > + if_true: files('vhost-user-video-pci.c'))
> >
> > if config_all_devices.has_key('CONFIG_QXL')
> > qxl_ss = ss.source_set()
> > diff --git a/hw/display/vhost-user-video.c b/hw/display/vhost-user-video.c
> > new file mode 100644
> > index 0000000000..506e350365
> > --- /dev/null
> > +++ b/hw/display/vhost-user-video.c
> > @@ -0,0 +1,386 @@
> > +/*
> > + * Vhost-user VIDEO virtio device
> > + *
> > + * This is the boilerplate for instantiating a vhost-user device
> > + * implementing a virtio-video device.
> > + *
> > + * The virtio video decoder and encoder devices are virtual devices that
> > + * support encoding and decoding respectively.
> > + *
> > + * The actual back-end for this driver is the vhost-user-video daemon.
> > + * The code here just connects up the device in QEMU and allows it to
> > + * be instantiated.
> > + *
> > + * Copyright (c) 2021 Linaro Ltd
> > + *
> > + * SPDX-License-Identifier: GPL-2.0-or-later
> > + */
> > +
> > +#include "qemu/osdep.h"
> > +#include "qapi/error.h"
> > +#include "hw/qdev-properties.h"
> > +#include "hw/virtio/virtio-bus.h"
> > +#include "hw/virtio/vhost-user-video.h"
> > +#include "qemu/error-report.h"
> > +
> > +/* currently there is no VIDEO enc/dec defined in Linux virtio_ids.h */
> > +#define VIRTIO_ID_VIDEO_ENC 30
> > +#define VIRTIO_ID_VIDEO_DEC 31
> > +#define MAX_CAPS_LEN 4096
> > +
> > +static void vhost_user_video_get_config(VirtIODevice *vdev, uint8_t
> > *config)
> > +{
> > + VHostUserVIDEO *video = VHOST_USER_VIDEO(vdev);
> > + struct virtio_video_config *vconfig = (struct virtio_video_config
> > *)config;
> > + int ret;
> > + Error *local_err = NULL;
> > +
> > + memset(config, 0, sizeof(struct virtio_video_config));
> > +
> > + ret = vhost_dev_get_config(&video->vhost_dev, config,
> > + sizeof(struct virtio_video_config),
> > &local_err);
> > + if (ret) {
> > + error_report("vhost-user-video: get device config space failed");
> > +
> > + /*TODO vhost_dev_get_config() fails so for now lets just set
> > it here */
>
> Is this a lifetime issue?
I believe so. I spent some time trying to get it working, but then added this
workaround so I could progress on the virtio-video implementation.
I will take another look and see if I can get thie config coming from the daemon
as intended.
Looking at other examples, hw/display/vhost-user-gpu.c get_config() hook seems
to call
vhost_dev_config successfully like I'm doing here.
But hw/block/vhost-user-blk.c and hw/virtio/vhost-user-vsock.c seem to have
a slightly different mechanism of using VhostDevConfigOps
.vhost_dev_config_notifier() which I don't recall having looked into that
previously.
>
> Otherwise:
>
> Reviewed-by: Alex Bennée <alex.ben...@linaro.org>
kind regards,
Peter.
