op 23-07-14 11:47, Christian KÃnig schreef:
Am 23.07.2014 11:44, schrieb Daniel Vetter:You would still need enable_signaling, else polling on the dma-buf wouldn't work. ;-)
On Wed, Jul 23, 2014 at 11:39 AM, Daniel Vetter <daniel.vetter@xxxxxxxx> wrote:Yeah, I see. You would need some way to abort the waiting on other devices fences in case of a lockup.
The scheduler needs to keep track of a lot of fences, so I think we'llAlso see my other comment about interactions between wait_fence and
have to register callbacks, not a simple wait function. We must keep
track of all the non-i915 fences for all oustanding batches. Also, the
scheduler doesn't eliminate the hw queue, only keep it much slower so
that we can sneak in higher priority things.
Really, scheduler or not is orthogonal.
the i915 reset logic. We can't actually use it from within the
scheduler code since that would deadlock.
What about an userspace thread to offload waiting and command submission to?
Can't wait synchronously with multiple shared fences, need to poll for that.
And the dma-buf would still have fences belonging to both drivers, and it would still call from outside the driver.
~Maarten