On 7/28/20 5:08 PM, Chris Wilson wrote:
Quoting Thomas Hellström (Intel) (2020-07-27 19:19:19)
On 7/15/20 1:51 PM, Chris Wilson wrote:
It is illegal to wait on an another vma while holding the vm->mutex, as
that easily leads to ABBA deadlocks (we wait on a second vma that waits
on us to release the vm->mutex). So while the vm->mutex exists, move the
waiting outside of the lock into the async binding pipeline.
Why is it we don't just move the fence binding to a separate loop after
unlocking the vm->mutex in eb_reserve_vm()?
That is what is done. The work is called immediately when possible. Just
the loop may be deferred if the what we need to unbind are still active
OK, then
Reviewed-by: Thomas Hellström <thomas.hellstrom@xxxxxxxxx>
-Chris
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
