Give r21836 a try and see if it still gets out of order.
Ralph
On Aug 18, 2009, at 2:18 PM, Greg Watson wrote:
Ralph,
Not sure that's it because all XML output should be via stdout.
Greg
On Aug 18, 2009, at 3:53 PM, Ralph Castain wrote:
Hmmmlet me try adding a fflush after the output
Trueyet the two outputs do come through separate code paths, so it
could be that's the issue. I honestly can't think of any other reason
as the printf for the mpirun tag comes well before any mapping occurs.
I'm not sure why the ssh would invert that order, nor how it could.
Let's try t
Ralph,
Not sure that's it because all XML output should be via stdout.
Greg
On Aug 18, 2009, at 3:53 PM, Ralph Castain wrote:
Hmmmlet me try adding a fflush after the output to
force it out. Best guess is that you are seeing a little race
condition - the map output is coming over stde
Hmmmlet me try adding a fflush after the output to force it
out. Best guess is that you are seeing a little race condition - the map
output is coming over stderr, while the tag is coming over stdout.
On Tue, Aug 18, 2009 at 12:53 PM, Greg Watson wrote:
> Hi Ralph,
>
> I'm seeing somethin
Hi Ralph,
I'm seeing something strange. When I run "mpirun -mca
orte_show_resolved_nodenames 1 -xml -display-map...", I see:
...
but when I run " ssh localhost mpirun -mca
orte_show_resolved_nodenames
Hmmmwell, actually - no. To the best of my knowledge, the only
ones using this interface are the Eclipse folks, who are the ones
requesting the changes.
Is anyone else out there using it? If so, please let us know and we'll
be more careful about procedure/docs.
On Aug 18, 2009, at 2:
On Mon, 2009-08-17 at 21:16 -0600, Ralph Castain wrote:
> Should be done on trunk with r21826 - would you please give it a try
> and let me know if that meets requirements? If so, I'll move it to
> 1.3.4.
Is there somewhere these xml changes are documented? I don't work with
OMPI xml but I do
- "Chris Samuel" wrote:
> This is most likely because it's getting an error from the
> kernel when trying to bind to a socket it's not permitted
> to access.
This is what strace reports:
18561 sched_setaffinity(18561, 8, { f0 }
18561 <... sched_setaffinity resumed> ) = -1 EINVAL (Invalid
- "Eugene Loh" wrote:
> Ah, you're missing the third secret safety switch that prevents
> hapless mortals from using this stuff accidentally! :^)
Sounds good to me. :-)
> I think you need to add
>
> --mca opal_paffinity_alone 1
Yup, looks like that's it; it fails to launch with tha
Chris Samuel wrote:
OK, grabbed that (1.4a1r21825). Configured with:
./configure --prefix=$FOO --with-openib --with-tm=/usr/
local/torque/latest --enable-static --enable-shared
It built & installed OK, but when running a trivial example
with it I don't see evidence for that code getting calle
- "Ralph Castain" wrote:
> Hi Chris
Hiya,
> The devel trunk has all of this in it - you can get that tarball from
> the OMPI web site (take the nightly snapshot).
OK, grabbed that (1.4a1r21825). Configured with:
./configure --prefix=$FOO --with-openib --with-tm=/usr/
local/torque/latest
11 matches
Mail list logo
