<div dir="ltr">Barry,<div><br></div><div>Thanks a lot; I should definitely learn to use the debugger. </div><div><br></div><div>I'm not sure what makes the graph "ill-formed" in ParMetis terminology. I'll keep looking into their manual to see if I can find anything. What makes this even harder is the code runs fine for certain grids and then this just happens as I refine the grid. I'll look into the function where I set up the adjacency matrix to see if I'm doing something wrong ...</div>
<div><br></div><div>I'll keep you posted.</div><div><br></div><div><br><br><div class="gmail_quote">On Mon, Apr 9, 2012 at 8:25 PM, Barry Smith <span dir="ltr"><<a href="mailto:bsmith@mcs.anl.gov">bsmith@mcs.anl.gov</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><br>
This is part of the excellent error handing in ParMetis (sarcasm intended), you would have found it rather quickly in the debugger. (ParMetis code follows)<br>
<br>
idx_t BSearch(idx_t n, idx_t *array, idx_t key)<br>
{<br>
idx_t a=0, b=n, c;<br>
<br>
while (b-a > 8) {<br>
c = (a+b)>>1;<br>
if (array[c] > key)<br>
b = c;<br>
else<br>
a = c;<br>
}<br>
<br>
for (c=a; c<b; c++) {<br>
if (array[c] == key)<br>
return c;<br>
}<br>
<br>
errexit("Key %"PRIDX" not found!\n", key);<br>
<br>
return 0;<br>
}<br>
<br>
So either it is a bug in ParMetis or ParMetis is being passed a "ill-formed" graph and not detecting the "ill-formed" graph cleanly.<br>
<br>
What to do? Well we can't spend out lives debugging other peoples libraries so ..... I am not sure what to do next? Validate that Parmetis is getting reasonable input?<br>
<span class="HOEnZb"><font color="#888888"><br>
Barry<br>
</font></span><div class="HOEnZb"><div class="h5"><br>
<br>
On Apr 9, 2012, at 10:19 PM, Mohammad Mirzadeh wrote:<br>
<br>
> -on_error_attach_debugger does not start the debugger. I traced it upto the following call: MatPartitioningApply(). It runs fine up until that function call and then gives the error.<br>
><br>
> Also, to make things even look more weird, I got the following once I was running it:<br>
><br>
> Key 10176 not found!<br>
> INTERNAL ERROR: Invalid error class (66) encountered while returning from<br>
> PMPI_Waitall. Please file a bug report.<br>
><br>
> Do you happen to know what that could mean? The funny thing is, it only happened once!<br>
><br>
> On Mon, Apr 9, 2012 at 7:57 PM, Jed Brown <<a href="mailto:jedbrown@mcs.anl.gov">jedbrown@mcs.anl.gov</a>> wrote:<br>
> On Mon, Apr 9, 2012 at 21:56, Mohammad Mirzadeh <<a href="mailto:mirzadeh@gmail.com">mirzadeh@gmail.com</a>> wrote:<br>
> :( that's all I see. Do you mean running with -on_error_attach_debugger?<br>
><br>
> Sure, or directly in a debugger. Does execution make it to main? Something is either not catching errors or is improperly calling exit() or similar.<br>
><br>
<br>
</div></div></blockquote></div><br></div></div>