[0/3] dma-buf: Add an API for exporting sync files (v7)

Message

Jason Ekstrand March 16, 2021, 4:53 a.m. UTC

Modern userspace APIs like Vulkan are built on an explicit
synchronization model.  This doesn't always play nicely with the
implicit synchronization used in the kernel and assumed by X11 and
Wayland.  The client -> compositor half of the synchronization isn't too
bad, at least on intel, because we can control whether or not i915
synchronizes on the buffer and whether or not it's considered written.

The harder part is the compositor -> client synchronization when we get
the buffer back from the compositor.  We're required to be able to
provide the client with a VkSemaphore and VkFence representing the point
in time where the window system (compositor and/or display) finished
using the buffer.  With current APIs, it's very hard to do this in such
a way that we don't get confused by the Vulkan driver's access of the
buffer.  In particular, once we tell the kernel that we're rendering to
the buffer again, any CPU waits on the buffer or GPU dependencies will
wait on some of the client rendering and not just the compositor.

This new IOCTL solves this problem by allowing us to get a snapshot of
the implicit synchronization state of a given dma-buf in the form of a
sync file.  It's effectively the same as a poll() or I915_GEM_WAIT only,
instead of CPU waiting directly, it encapsulates the wait operation, at
the current moment in time, in a sync_file so we can check/wait on it
later.  As long as the Vulkan driver does the sync_file export from the
dma-buf before we re-introduce it for rendering, it will only contain
fences from the compositor or display.  This allows to accurately turn
it into a VkFence or VkSemaphore without any over- synchronization.

The only remaining TODO item as far as I know is that some kernel CI system
is sending me nastygrams about build system problems.  With this change,
dma-fence now has a dependency on sync_file and I really don't know how
best to solve it.  Should we make sync_file no longer optional?  Should I
hide the new ioctl behind a #define?  If so, what #define?  I'm a bit lost
when it comes to KConfig, I'm afraid.

Mesa MR: https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/4037
IGT tests: https://lists.freedesktop.org/archives/igt-dev/2021-March/029825.html

Cc: Christian König <[e-mail]>
Cc: Michel Dänzer <[e-mail]>
Cc: Dave Airlie <[e-mail]>
Cc: Bas Nieuwenhuizen <[e-mail]>
Cc: Daniel Stone <[e-mail]>

Christian König (2):
  dma-buf: add dma_fence_array_for_each (v2)
  dma-buf: add dma_resv_get_singleton (v2)

Jason Ekstrand (1):
  dma-buf: Add an API for exporting sync files (v6)

 drivers/dma-buf/dma-buf.c         |  55 ++++++++++++++
 drivers/dma-buf/dma-fence-array.c |  27 +++++++
 drivers/dma-buf/dma-resv.c        | 118 ++++++++++++++++++++++++++++++
 include/linux/dma-fence-array.h   |  17 +++++
 include/linux/dma-resv.h          |   3 +
 include/uapi/linux/dma-buf.h      |   6 ++
 6 files changed, 226 insertions(+)