mbrookhart commented on a change in pull request #6839:
URL: https://github.com/apache/tvm/pull/6839#discussion_r538542467
##########
File path: python/tvm/topi/cuda/sort.py
##########
@@ -201,9 +201,9 @@ def sort_nms_ir(data, valid_count, output, axis, is_ascend):
nthread_tx = max_threads
nthread_bx = size // max_threads + 1
tx = te.thread_axis("threadIdx.x")
- bx = te.thread_axis("vthread")
+ bx = te.thread_axis("blockIdx.x")
ib.scope_attr(tx, "thread_extent", nthread_tx)
- ib.scope_attr(bx, "virtual_thread", nthread_bx)
+ ib.scope_attr(bx, "thread_extent", nthread_bx)
Review comment:
I have not been able to pass tests with the kernels currently in master,
so I can't compare apples to apples
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
