Open jansel opened 4 months ago
So this pipeline is a single scalar add operation?
I don't think any of us expected anyone to try to autoschedule a pipeline that does O(1) work. I think the appropriate schedule is gpu_single_thread(), but nobody taught the autoscheduler how to use that.
Yeah correct, it should be pretty trivial to schedule -- but it is a corner case the scheduler doesn't handle. This is coming from a unit test, but you occasionally have scalar operations (for example a learning rate update) in real models.
repro.py:
Note: you will need to update the path to
libautoschedule_anderson2021.so
for your system.Output:
This example is just adding two 1-element tensors.
Possible workarounds:
out_ptr0.set_estimates
from1
to2
(even though the real tensor is size 1). For some of the other schedulers (on CPU) I have gotten out of bounds access errors if I made the estimates larger than the actual value. Is doing this safe?