Thanks Rolf.  We'll try it out.

Ken

On Tue, 2011-04-19 at 13:45 -0700, Rolf vandeVaart wrote:
> WHAT: Second try to add support to send data directly from CUDA device
> memory via MPI calls.
> 
>  
> 
> TIMEOUT: 4/26/2011
> 
>  
> 
> DETAILS: Based on all the feedback (thanks to everyone who looked at
> it),  I have whittled down what I hope to accomplish with this RFC.
> There were suggestions to better modularize the CUDA registration code
> so I will take a look at that separately.  Since the registration code
> is a performance feature, it will be dropped from this RFC and
> investigated separately.  This significantly reduced the changes being
> proposed here.  With this RFC, all the changes are isolated in
> datatype and convertor code.   As mentioned before, the changes mostly
> boil down to replacing memcpy with cuMemcpy when moving the data to or
> from a CUDA device buffer.
> 
>  
> 
> Per suggestions, the choice to disable the large memory RDMA now
> happens on a per message basis.  This is done by adding a flag to the
> convertor which tells the BTLs that an intermediate buffer is needed
> when dealing with device memory.
> 
>  
> 
> As before, this code would be enabled via a configure option.  A
> mostly completed version is viewable on bitbucket although I know the
> configure code is sorely lacking.
> 
>  
> 
> This is the new list of changed files.
> 
>  
> 
> M       opal/config/opal_configure_options.m4
> 
> A       opal/datatype/opal_datatype_cuda.c
> 
> A       opal/datatype/opal_datatype_cuda.h
> 
> M       opal/datatype/opal_convertor.h
> 
> M       opal/datatype/opal_datatype_copy.c
> 
> M       opal/datatype/opal_datatype_unpack.c
> 
> M       opal/datatype/Makefile.am
> 
> M       opal/datatype/opal_datatype_pack.h
> 
> M       opal/datatype/opal_convertor.c
> 
> M       opal/datatype/opal_datatype_unpack.h
> 
> M       ompi/mca/pml/ob1/pml_ob1_sendreq.h
> 
>  
> 
> Thanks,
> 
> Rolf
> 
>  
> 
> 
> 
> ______________________________________________________________________
> This email message is for the sole use of the intended recipient(s)
> and may contain confidential information.  Any unauthorized review,
> use, disclosure or distribution is prohibited.  If you are not the
> intended recipient, please contact the sender by reply email and
> destroy all copies of the original message. 
> 
> ______________________________________________________________________
> 
> _______________________________________________
> devel mailing list
> de...@open-mpi.org
> http://www.open-mpi.org/mailman/listinfo.cgi/devel




Reply via email to