marcoabreu commented on a change in pull request #15550: Infra to use tvm write
op kernels
URL: https://github.com/apache/incubator-mxnet/pull/15550#discussion_r304238016
##########
File path: Makefile
##########
@@ -589,6 +603,30 @@ $(DMLC_CORE)/libdmlc.a: DMLCCORE
DMLCCORE:
+ cd $(DMLC_CORE); $(MAKE) libdmlc.a USE_SSE=$(USE_SSE)
config=$(ROOTDIR)/$(config); cd $(ROOTDIR)
+TVM_USE_CUDA := OFF
Review comment:
Could we have that logic by checking use_cuda then instead? I see your point
and it's certainly simpler, but we have a lot of diverging logic branches
around the use_cuda flag already, so I'd like to avoid further branching by
introducing additional variables that branch off of that.
----------------------------------------------------------------
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]
With regards,
Apache Git Services
