On Fri, Nov 9, 2012 at 1:47 PM, Kirk, Benjamin (JSC-EG311) <
[email protected]> wrote:
>
> FWIW, Roy and I were having some offline discussion, and as soon as 0.8.0
> is tagged I plan to merge automake. Like the next day…
>
The current plan is still to distribute a "prebuilt" configure sc
Good news, everyone!
Automake branch with OpenMPI, PETSc, and GCC-4.7.2 on Mountain Lion seems
to finally be working for me (as it has been working for you all for
weeks...:-/).
Thanks for all the help/suggestions!
--
John
On Fri, Nov 9, 2012 at 1:47 PM, Kirk, Benjamin (JSC-EG311) <
[email protected]> wrote:
> FWIW, Roy and I were having some offline discussion, and as soon as 0.8.0
> is tagged I plan to merge automake. Like the next day…
>
> So keep that in mind, if automake is working for you then you mig
On Fri, 9 Nov 2012, Vikram Garg wrote:
Roy, Should we shoot for some kind of time dependent adjoint support
before the release ? I think we have a first iteration of how to add that
functionality figured out.
If we can get the first-pass stuff done and debugged before John's
fixed that
Roy,
Should we shoot for some kind of time dependent adjoint support
before the release ? I think we have a first iteration of how to add that
functionality figured out.
Thanks.
On Fri, Nov 9, 2012 at 3:47 PM, Kirk, Benjamin (JSC-EG311) <
[email protected]> wrote:
>
> On Nov 9, 201
On Nov 9, 2012, at 2:33 PM, John Peterson wrote:
> At this point, I'm fairly convinced that the only way to do shared library
> linking **portably** is with libtool, but I haven't started on this patch
> yet...and it will likely take a few iterations to get correct.
FWIW, Roy and I were havin
On Fri, Nov 9, 2012 at 1:15 PM, Kirk, Benjamin (JSC-EG311) <
[email protected]> wrote:
> OK, reviving this thread. I've got one more change I'd like to roll in of
> 0.8.0 - the RB I/O optimization - but beyond that it looks like we are
> pretty close.
>
> What else is still outstanding bef
On Fri, 9 Nov 2012, Kirk, Benjamin (JSC-EG311) wrote:
> OK, reviving this thread. I've got one more change I'd like to roll
> in of 0.8.0 - the RB I/O optimization - but beyond that it looks
> like we are pretty close.
>
> What else is still outstanding before a feature freeze/release?
I've got
On Oct 25, 2012, at 5:57 PM, Roy Stogner wrote:
>
> On Thu, 25 Oct 2012, Paul T. Bauman wrote:
>
>> 1. FEMContext work - Finish off hiding public members and putting in
>> accessors. I can get this done pretty quickly so I don't think it
>> would hold up a release, but nothing is broken at the
On Nov 9, 2012, at 2:00 PM, Derek Gaston wrote:
> I just wanted to report that with libMesh HEAD our tests are currently
> running fine in both serial and parallel (both threaded and MPI and the
> combo). So, as far as I can tell, nothing has been broken on our end.
Woohoo!
Thanks for that.
So... I had a crazy week this week and I haven't given you guys feedback on
all of this.
I just wanted to report that with libMesh HEAD our tests are currently
running fine in both serial and parallel (both threaded and MPI and the
combo). So, as far as I can tell, nothing has been broken on our
On Fri, Nov 9, 2012 at 1:11 PM, Paul T. Bauman wrote:
>
> Will report back about my results with mpich2.
>
Sure enough! I get a failure on branch on adjoints_ex1 using mpich2 1.4.1p1
with a stack trace back to misaligned_stack_error_entering_dyld_stub_binder
().
I don't have these issues with O
On Fri, Nov 9, 2012 at 1:11 PM, Kirk, Benjamin (JSC-EG311) <
[email protected]> wrote:
> I'd go with OpenMPI if it is feasible.
>
I'd agree with this for the most part, except if you want to do one-sided
communication. There are known, ancient, unresolved issues in OpenMPI for
one-sided co
On Nov 9, 2012, at 1:00 PM, John Peterson wrote:
> (Correct me if I'm wrong, but everyone at Texas on a Mac is using OpenMPI,
> right?)
Definitely true for me - OpenMPI 1.6.3 here
I'd suspect the flat namespace business too - let me know if you have a
compelling reason to use mpich, I'm flue
On Fri, Nov 9, 2012 at 1:00 PM, John Peterson wrote:
> In case there's anyone that hasn't muted this conversation yet...
I'm really interested in this because the trunk failures I'd been seeing
were a real mystery to me.
> I think I've tracked the INL's issue down to mpich2. (I happen to be
In case there's anyone that hasn't muted this conversation yet...
I think I've tracked the INL's issue down to mpich2. (I happen to be using
mpich2-1.4.1p1, but the problem is probably present in mpich2-1.5 as well.)
(Correct me if I'm wrong, but everyone at Texas on a Mac is using OpenMPI,
righ
On Nov 9, 2012, at 10:59 AM, David Knezevic wrote:
> SVN head works for me now, thanks!
Great - next will be my optimization for multiple vectors, which I expect to
provide the real win.
-Ben
--
Everyone hates slow
On 11/09/2012 09:46 AM, Kirk, Benjamin (JSC-EG311) wrote:
> On Nov 9, 2012, at 1:35 AM, Roy Stogner wrote:
>
>> I'm not sure what the issue is, maybe something with recent I/O
>> changes? Stack traces (from processes interrupted in some kind of busy-loop)
>> include three processes caught here:
On Nov 9, 2012, at 8:47 AM, Roy Stogner wrote:
> This particular failure was on a --enable-everything --enable-parmesh
> build, but at least the --enable-parmesh part of that is unnecessary.
> Check out our recent BuildBot results; it seems like reduced_basis_ex6
> is hanging nearly half the tim
On Fri, 9 Nov 2012, Kirk, Benjamin (JSC-EG311) wrote:
> Could you describe the configure options that gave rise to this? I'm not
> seeing it with my default stuff. Parallel mesh or what?
>
> Let me know how you configured and I'll see if I can replicate.
This particular failure was on a --enab
On Nov 9, 2012, at 1:35 AM, Roy Stogner wrote:
> I'm not sure what the issue is, maybe something with recent I/O
> changes? Stack traces (from processes interrupted in some kind of busy-loop)
> include three processes caught here:
Got this - I missed that the reduced basis code was also manual
Could you describe the configure options that gave rise to this? I'm not
seeing it with my default stuff. Parallel mesh or what?
Let me know how you configured and I'll see if I can replicate.
-Ben
On Nov 9, 2012, at 1:35 AM, "Roy Stogner" wrote:
>
> On Thu, 8 Nov 2012, Roy Stogner wrote:
22 matches
Mail list logo
