Laurawly commented on a change in pull request #4353: [Perf] Enhance cudnn and
cublas backend and enable TensorCore
URL: https://github.com/apache/incubator-tvm/pull/4353#discussion_r347069828
##########
File path: src/runtime/contrib/cublas/cublas.cc
##########
@@ -124,35 +175,205 @@ struct CublasDgemmBatchOp {
}
};
+// Check cublas supported mix-precision computation type and return computeType
+bool CheckCublasMixPrecisionType(DLDataType in_dtype, DLDataType out_dtype,
bool int_support = true) {
Review comment:
Maybe shorten the name to just `CheckMixPrecisionType` since the function is
under cublas context
----------------------------------------------------------------
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
