You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@tvm.apache.org by GitBox <gi...@apache.org> on 2019/12/23 05:40:48 UTC

[GitHub] [incubator-tvm] ghostplant edited a comment on issue #4547: CUDA source emited from hybrid script contains undefined `vthread`

ghostplant edited a comment on issue #4547: CUDA source emited from hybrid script contains undefined `vthread`
URL: https://github.com/apache/incubator-tvm/issues/4547#issuecomment-568362640
 
 
   I think there is an undetermined way to decide whether we need to bind an axis to `vthread`. Sometimes we cannot bind it because it is possible to result in `// attr [iter_var(vthread, , vthread)] thread_extent = 2` in lower code.
   Sometimes we have to bind it because lacking it might result in `invalid argument number like: float *input0, float *output, int outer_i` for a kernel in its lower code, which is not transformed into a loop.

----------------------------------------------------------------
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:
users@infra.apache.org


With regards,
Apache Git Services