You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
One of the important bits of difference is that both sides of an event/barrier now know about the barrier contents. This matches DX12 split barriers, so there is an obvious benefit here.
I think what we should do is having this extra information optional in set_event. If the user provides it, then the backend can take advantage of a split barrier. Otherwise, it's the same as today. Vulkan Portability would be using the old version for a while.
The text was updated successfully, but these errors were encountered:
This appears to be more problematic than I thought. The wait_events are supposed to receive a list of the sets of barriers now, as opposed to the just a set of barriers for all the events at once.
Moreover, each individual barrier now carries its own stage mask.
This probably needs a bit more thought, doesn't need to block hal-0.8.
See https://gpuopen.com/vulkan-sync2-support/ and https://www.khronos.org/registry/vulkan/specs/1.2-extensions/man/html/VK_KHR_synchronization2.html
One of the important bits of difference is that both sides of an event/barrier now know about the barrier contents. This matches DX12 split barriers, so there is an obvious benefit here.
I think what we should do is having this extra information optional in
set_event
. If the user provides it, then the backend can take advantage of a split barrier. Otherwise, it's the same as today. Vulkan Portability would be using the old version for a while.The text was updated successfully, but these errors were encountered: