[v3,0/5] drm/virtio: Import scanout buffers from other devices

Message

Vivek Kasireddy Nov. 19, 2024, 5:59 a.m. UTC

Having virtio-gpu import scanout buffers (via prime) from other
devices means that we'd be adding a head to headless GPUs assigned
to a Guest VM or additional heads to regular GPU devices that are
passthrough'd to the Guest. In these cases, the Guest compositor
can render into the scanout buffer using a primary GPU and has the
secondary GPU (virtio-gpu) import it for display purposes.

The main advantage with this is that the imported scanout buffer can
either be displayed locally on the Host (e.g, using Qemu + GTK UI)
or encoded and streamed to a remote client (e.g, Qemu + Spice UI).
Note that since Qemu uses udmabuf driver, there would be no copies
made of the scanout buffer as it is displayed. This should be
possible even when it might reside in device memory such has VRAM.

The specific use-case that can be supported with this series is when
running Weston or other guest compositors with "additional-devices"
feature (./weston --drm-device=card1 --additional-devices=card0).
More info about this feature can be found at:
https://gitlab.freedesktop.org/wayland/weston/-/merge_requests/736

In the above scenario, card1 could be a dGPU or an iGPU and card0
would be virtio-gpu in KMS only mode. However, the case where this
patch series could be particularly useful is when card1 is a GPU VF
that needs to share its scanout buffer (in a zero-copy way) with the
GPU PF on the Host. Or, it can also be useful when the scanout buffer
needs to be shared between any two GPU devices (assuming one of them
is assigned to a Guest VM) as long as they are P2P DMA compatible.

As part of the import, the virtio-gpu driver shares the dma
addresses and lengths with Qemu which then determines whether the
memory region they belong to is owned by a VFIO device or whether it
is part of the Guest's system ram. If it is the former, it can use
the VFIO_DEVICE_FEATURE_DMA_BUF feature flag while invoking the ioctl
against the VFIO device fd and get a dmabuf fd in return. In the
latter case, Qemu obtains the dmabuf fd using the udmabuf driver.

Note that the virtio-gpu driver registers a move_notify() callback
to track location changes associated with the scanout buffer and
sends attach/detach backing cmds to Qemu when appropriate. And,
synchronization (that is, ensuring that Guest and Host are not
using the scanout buffer at the same time) is ensured by pinning/
unpinning the dmabuf as part of prepare/cleanup fb and using a
fence in resource_flush cmd.

Changelog:

v2 -> v3:
- Rebase on 6.12

v1 -> v2:
- Use a fenced version of VIRTIO_GPU_CMD_RESOURCE_DETACH_BACKING cmd
  (Dmitry)

RFC -> v1:
- Use virtio_gpu_cleanup_object() to cleanup the imported obj
- Do pin/unpin as part of prepare and cleanup fb for the imported
  dmabuf obj instead doing it as part of plane update
- Tested with gnome-shell/mutter (wayland backend)

Patchset overview:

Patch 1:   Implement VIRTIO_GPU_CMD_RESOURCE_DETACH_BACKING cmd
Patch 2-3: Helpers to initalize, import, free imported object
Patch 4-5: Import and use buffers from other devices for scanout

This series is tested using the following method:
- Run Qemu with the following relevant options:
  qemu-system-x86_64 -m 4096m ....
  -device vfio-pci,host=0000:03:00.0
  -device virtio-vga,max_outputs=1,blob=true,xres=1920,yres=1080
  -display gtk,gl=on
  -object memory-backend-memfd,id=mem1,size=4096M
  -machine memory-backend=mem1 ...
- Run upstream Weston with the following options in the Guest VM:
  ./weston --drm-device=card1 --additional-devices=card0
- Or run Gnome-shell/Mutter (wayland backend) with this additional patch:
  https://gitlab.gnome.org/GNOME/mutter/-/merge_requests/3745
  XDG_SESSION_TYPE=wayland dbus-run-session -- /usr/bin/gnome-shell --wayland --no-x11

where card1 is a DG2 dGPU (passthrough'd and using i915 driver in Guest VM),
card0 is virtio-gpu and the Host is using a RPL iGPU.

Cc: Gerd Hoffmann <[e-mail]>
Cc: Dongwon Kim <[e-mail]>
Cc: Simona Vetter <[e-mail]>
Cc: Christian Koenig <[e-mail]>
Cc: Dmitry Osipenko <[e-mail]>
Cc: Rob Clark <[e-mail]>
Cc: Gurchetan Singh <[e-mail]>
Cc: Chia-I Wu <[e-mail]>

Vivek Kasireddy (5):
  drm/virtio: Implement VIRTIO_GPU_CMD_RESOURCE_DETACH_BACKING cmd
  drm/virtio: Add a helper to map and note the dma addrs and lengths
  drm/virtio: Add helpers to initialize and free the imported object
  drm/virtio: Import prime buffers from other devices as guest blobs
  drm/virtio: Add prepare and cleanup routines for imported dmabuf obj

 drivers/gpu/drm/virtio/virtgpu_drv.h    |  10 ++
 drivers/gpu/drm/virtio/virtgpu_object.c |  26 ++++
 drivers/gpu/drm/virtio/virtgpu_plane.c  |  65 ++++++++-
 drivers/gpu/drm/virtio/virtgpu_prime.c  | 170 +++++++++++++++++++++++-
 drivers/gpu/drm/virtio/virtgpu_vq.c     |  25 ++++
 5 files changed, 294 insertions(+), 2 deletions(-)