<div class="gmail_quote">On Fri, Aug 17, 2012 at 1:05 PM, John Fettig <span dir="ltr"><<a href="mailto:john.fettig@gmail.com" target="_blank">john.fettig@gmail.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Is this the role of MatGetGhosts?<br></blockquote><div><br></div><div>That's the intent, but it looks like this function was only implemented for MPIBAIJ.</div><div><br></div><div>The actual implementation of MatMult_MPI*AIJ is to start the scatter, apply the diagonal block, finish the scatter, and apply the off-diagonal part.</div>
<div><br></div><div>Why do you want raw access?</div><div><br></div><div>If you really need, you can either use MatMPIAIJGetSeqAIJ() or just include the private header.</div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<span class="HOEnZb"><font color="#888888"><br>
John<br>
</font></span><div class="HOEnZb"><div class="h5"><br>
On Fri, Aug 17, 2012 at 2:01 PM, John Fettig <<a href="mailto:john.fettig@gmail.com">john.fettig@gmail.com</a>> wrote:<br>
> Is there a scatter context available someplace that would allow me to<br>
> gather all of the elements of a vector that correspond to the ones<br>
> needed for a MatMult (MPIAIJ matrices)? When you do a MatMult, I<br>
> presume it first scatters/gathers the ghost elements of the vector<br>
> being multiplied, can I use this scatter for a vectors outside of the<br>
> MatMult?<br>
><br>
> Thanks,<br>
> John<br>
</div></div></blockquote></div><br>