Awesome, this is something that i overlooked and definitely should be the way
to go.
The main reason that I took the current approach was because there is no clear
way to do synchronization other than the fence. Your proposed approach is
definitely better and we should just move toward this new
cc @jwfromm, @tqchen
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/dmlc/tvm/pull/3849#issuecomment-525639619
In the Vulkan API, it is recommended to launch multiple command buffers per
kernel. See
http://on-demand.gputechconf.com/gtc/2016/events/vulkanday/High_Performance_Vulkan.pdf,
https://devblogs.nvidia.com/vulkan-dos-donts/, etc. The extant TVM Vulkan
runtime uses one command buffer per kernel, w
Hybrid python programming model
transition of vision operators to hybrid mode.
what's plan for this feature?
--
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/dmlc/tvm/issues/2623#issuecomment-525611531
