<div><div><div><div><div class="gmail_quote"><div dir="auto">On Wed, 15 Nov 2017 at 05:55, Adrian Croucher <<a href="mailto:a.croucher@auckland.ac.nz" target="_blank">a.croucher@auckland.ac.nz</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">hi<br>
<br>
I'm trying to use ISGlobalToLocalMappingApplyBlock() and am a bit<br>
puzzled about the results it's giving.<br>
<br>
I've attached a small test to illustrate. It just sets up a<br>
local-to-global mapping with 10 elements. Running on two processes the<br>
first has global indices 0 - 4 and the the second has 5 - 9. I then try<br>
to find the local index corresponding to global index 8.<br>
<br>
If I set the blocksize parameter to 1, it correctly gives the results -1<br>
on rank 0 and 3 on rank 1.<br>
<br>
But if I set the blocksize to 2 (or more), the results are -253701943 on<br>
rank 0 and -1 on rank 1. Neither of these are what I expected- I thought<br>
they should be the same as in the blocksize 1 case.</blockquote><div dir="auto"><br></div></div></div></div></div></div><div><div><div><div><div class="gmail_quote"><div dir="auto">The man page says to use "block global numbering"</div></div></div></div></div></div><div><div><div><div><div class="gmail_quote"><div dir="auto"><br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><br>
<br>
I'm presuming the global indices I pass in to<br>
ISGlobalToLocalMappingApplyBlock() should be global block indices (i.e.<br>
not scaled up by blocksize). </blockquote><div dir="auto"><br></div></div></div></div></div></div><div><div><div><div><div class="gmail_quote"><div dir="auto">Yes, the indices should relate to the blocks</div></div></div></div></div></div><div><div><div><div><div class="gmail_quote"><div dir="auto"><br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">If I do scale them up it doesn't give the<br>
answers I expect either.<br>
<br>
Or am I wrong to expect this to give the same results regardless of<br>
blocksize?</blockquote><div dir="auto"><br></div></div></div></div></div></div><div><div><div><div><div class="gmail_quote"><div dir="auto">Yep.</div><div dir="auto"><br></div><div dir="auto">However the large negative number being printed looks an uninitialized variable. This seems odd as with mode = MASK nout should equal N and any requested block indices not in the IS should result in -1 being inserted in your local_indices array.</div><div dir="auto"><br></div><div dir="auto">What's the value of nout?</div><div dir="auto"><br></div><div dir="auto">Thanks,</div><div dir="auto"> Dave</div></div></div></div></div></div><div><div><div><div><div class="gmail_quote"><div dir="auto"><br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><br>
<br>
Cheers, Adrian<br>
<br>
--<br>
Dr Adrian Croucher<br>
Senior Research Fellow<br>
Department of Engineering Science<br>
University of Auckland, New Zealand<br>
email: <a href="mailto:a.croucher@auckland.ac.nz" target="_blank">a.croucher@auckland.ac.nz</a><br>
tel: +64 (0)9 923 4611<br>
<br>
</blockquote></div></div></div></div></div>