You are viewing a plain text version of this content. The canonical link for it is here.

Posted to commits@mxnet.apache.org by GitBox <gi...@apache.org> on 2020/09/14 21:48:51 UTC

[GitHub] [incubator-mxnet] ptrendx commented on a change in pull request #19142: [1.x][FEATURE] CUDA graphs support

ptrendx commented on a change in pull request #19142:
URL: https://github.com/apache/incubator-mxnet/pull/19142#discussion_r488239192



##########
File path: src/operator/numpy/linalg/np_eigvals.cu
##########
@@ -29,11 +29,19 @@ namespace mxnet {
 namespace op {
 
 NNVM_REGISTER_OP(_npi_eigvals)
+.set_attr<FIsCUDAGraphsCompatible>("FIsCUDAGraphsCompatible",
+    [](const NodeAttrs&, const bool) {
+      return false;
+    })
 .set_attr<FCompute>("FCompute<gpu>", EigvalsOpForward<gpu>);
 
 #if MXNET_USE_CUSOLVER == 1
 
 NNVM_REGISTER_OP(_npi_eigvalsh)
+.set_attr<FIsCUDAGraphsCompatible>("FIsCUDAGraphsCompatible",

Review comment:
       Well, the "everywhere" here is actually way less than if we went the other way around (only a bunch of operators with `FCompute` has synchronization that is not allowed under graphs.




----------------------------------------------------------------
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