An issue popped up as part of the code review for

https://github.com/ofiwg/libfabric/pull/6185

The summary is that supporting FI_HMEM (e.g. GPU memory buffers) may have a 
negative impact on the performance of fi_inject() calls.  Either the provider 
must disable inject completely, or possibly check the buffer location on each 
inject call.

Unfortunately, the problem was not detected until now, and we're looking for 
input on how to handle this.

- Sean
_______________________________________________
ofiwg mailing list
[email protected]
https://lists.openfabrics.org/mailman/listinfo/ofiwg

Reply via email to