The idea of a “paint thread” that’s isolated enough from the “compositor thread” so as to never jank scrolling in the latter is fundamentally at odds with the way the GPU FIFO works. (Assuming the paint thread uses the GPU.)
-
-
Context switching has been a feature of NVIDIA GPUs for quite a while, initially with command granularity but it's getting more fine-grained.
-
I don't think it was until Pascal that (GTX 1080/1070) NVIDIA implemented "fine-grain" context switching (instruction-level, aka save/restore a warp in the middle of its execution). It's also advertised as "compute preemption", so I'm not sure if graphics got the same.
- 4 more replies
New conversation -
-
-
WDDM 2.0 (guided by Windows Vista) from over 10 years ago was demanding GPU vendors implement fine-grained multi-threading. On mobile at least, this has been a reality for many years.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.