Re: [Paraview] Paraview vs. nVidia Driver
David - I came to the same exact conclusion. When I jumped to openMPI-1.3 I was able to get a clean build and execution.Strange. David C Thompson wrote: Has anyone had any trouble building Paraview with the latest nVidia GL libraries? On both RH4 and RH5, we've not been able to compile Paraview 3.4.0 (or 3.2.1) against nVidia driver 180.22 or 180.25.We can build against nVidia driver 173.14.xx. Also, if we build on a system that has driver 173.14.xx and then try to execute 'paraview' on a system with a 180.22 driver, we also get a floating point exception error. I've had a similar problem on Fedora 10, but only when MPI support is turned on and linking to openmpi-1.2.x (where x is 4 or 8). Carefully uninstalling all vestiges of the nvidia drivers, re-installing just the 180.25 drivers, and building against openmpi-1.3.0 solved the problem. I am not sure how critical the nvidia stuff was to the process but it's something I did when I was looking into the issue -- I noticed that NVidia's *.pkg.run script had put some libraries in a different directory than the rpms from rpmfusion-nonfree and so both coexisted (and might have caused problems where we linked to one but loaded the other at runtime). In any event the floating point exception seemed to be coming from the openmpi-opal library's override of malloc. Hope this helps, David The compile failure is a floating point exception and looks like this: [89%] Built target pvpython-real Scanning dependencies of target pvbatch Scanning dependencies of target pvpython /bin/sh: line 1: 28655 Floating point exception../../bin/vtkSMExtractDocumentation /build/ParaView-3.2.1-x86_64-RHEL/Documentation /build/src/ParaView3.2.1/Qt/Components/Resources/XML/ParaViewWriters.xml make[2]: *** [Documentation/ParaViewWriters.html] Error 136 make[1]: *** [Servers/ServerManager/CMakeFiles/HTMLDocumentation.dir/all] Error 2 make[1]: *** Waiting for unfinished jobs ___ ParaView mailing list ParaView@paraview.org http://www.paraview.org/mailman/listinfo/paraview ___ ParaView mailing list ParaView@paraview.org http://www.paraview.org/mailman/listinfo/paraview
Re: [Paraview] Paraview on a tiled display
Hi Moreland, As i said when i tried running paraview on the tiled display with 6 procesors only everything worked right. I could see perfectly each model loaded on Paraview on the Tiled Display, even glgears i tested and run correctly on each screen. Now i corrected the Paraview's host file and it looks like this. 192.168.51.136 slots=4 192.168.51.134 slots=4 192.168.51.132 slots=4 192.168.51.135 slots=4 192.168.51.133 slots=4 192.168.51.131 slots=4 Each of the six computers comforming the cluster has 4 slots available meaning a total of 24 processors running in parallel. 192.168.51.136 is the first screen, 192.168.51.134 the second and so on... When i run the command with np 24 then the mapping on each screen is incorrect, as it 136 was no longer associated with process 0 and so on. So at the end you see the model but the correspondance between the tiles and process is incorrect. I don't know if i am specifying wrong the arguments tdx and tdy as tdx=3 and tdy=2 and 3*2=6 which is different to 24...maybe there is something i am going over which is causing the incorrect mapping. 2009/2/2 Utkarsh Ayachit utkarsh.ayac...@kitware.com I knew I should have looked it up before responding :), I stand corrected. On Mon, Feb 2, 2009 at 1:00 PM, Moreland, Kenneth kmo...@sandia.gov wrote: Sorry, but that's not true at all. It is in fact encouraged to have more processors than tiles when driving a tiled display. All the processors, even the non-display ones, will be involved in the processing and parallel rendering work. -Ken On 1/30/09 11:43 AM, Utkarsh Ayachit utkarsh.ayac...@kitware.com wrote: The number of processes must match the number of tiles (num of tiles = tdx * tdy). In your case you have 8 processes are only 2*1 = 3 tiles? Utkarsh On Fri, Jan 30, 2009 at 10:21 AM, Camilo Marin igetmyjunkmailh...@gmail.com wrote: Hi all, We are trying to configure and run Paraview 3.4.0 on a tiled display with the following command: mpirun -np 8 --mca btl ^openib,udapl --mca btl_tcp_if_exclude lo --hostfile /home/imagine/ParaView/hosts /bin/env DISPLAY=:0 ~/OpenFOAM/ThirdParty/ParaView3.3-cvs/platforms/linux64Gcc/bin/pvserver --server-port=1100 -tdx=1 -tdy=2 Then when we connect throught the paraview client it doesn't show in the two displays we requested. So, is there a guideline or some kind of command/configuration we are missing so it can be displayed as wished? Thnaks in advance. ___ ParaView mailing list ParaView@paraview.org http://www.paraview.org/mailman/listinfo/paraview ___ ParaView mailing list ParaView@paraview.org http://www.paraview.org/mailman/listinfo/paraview Kenneth Moreland *** Sandia National Laboratories *** *** *** *** email: kmo...@sandia.gov ** *** ** phone: (505) 844-8919 *** web: http://www.cs.unm.edu/~kmorelhttp://www.cs.unm.edu/%7Ekmorel ___ ParaView mailing list ParaView@paraview.org http://www.paraview.org/mailman/listinfo/paraview
Re: [Paraview] best way to update a custom reader
One possibility is to create a custom panel with a timer which periodically calls RequestInformation() to update the timesteps. Utkarsh On Thu, Jan 29, 2009 at 2:25 AM, Dominik Szczerba domi...@itis.ethz.ch wrote: Hi, What is the best way to update my custom reader? I have a complex pipeline and my timesteps are continuously coming in - reloading the dataset and rebuilding the settings every time is not very funny. I CAN do it by calling RequestInformation (where I set the time range etc) each time I press 'left' or 'right' on the animation tab, but slows down playback, because the function does some extra IO. Maybe a custom filter calling RequestInformation of current dataset? Any other ideas? Dominik -- Dominik Szczerba, Ph.D. Computational Physics Group IT'IS Foundation http://www.itis.ethz.ch ___ ParaView mailing list ParaView@paraview.org http://www.paraview.org/mailman/listinfo/paraview ___ ParaView mailing list ParaView@paraview.org http://www.paraview.org/mailman/listinfo/paraview
Re: [Paraview] Paraview on a tiled display
OK, if I understand correctly, you have 6 nodes with 4 processors per node.
You want to launch a job with 4 processes per node (one per processor). When
you do that, the first 4 processes are assigned to one node, and all try to
create their own tile on the same node because ParaView assumes that you have
arranged the first 6 processes on the display nodes.
Each implementation of MPI has its own way of specifying how to assign the
processes in your job. You have not said, but it looks like you are using
OpenMPI. You might want to try the -bynode flag to mpirun. This flag is
supposed to interleave the process allocation amongst nodes. Thus, 136 should
get process 0, 6, 12, and 18; 134 should get processes 1, 7, 13, and 19; and so
on.
-Ken
On 2/3/09 8:04 AM, Camilo Marin igetmyjunkmailh...@gmail.com wrote:
Hi Moreland,
As i said when i tried running paraview on the tiled display with 6 procesors
only everything worked right. I could see perfectly each model loaded on
Paraview on the Tiled Display, even glgears i tested and run correctly on each
screen. Now i corrected the Paraview's host file and it looks like this.
192.168.51.136 slots=4
192.168.51.134 slots=4
192.168.51.132 slots=4
192.168.51.135 slots=4
192.168.51.133 slots=4
192.168.51.131 slots=4
Each of the six computers comforming the cluster has 4 slots available meaning
a total of 24 processors running in parallel. 192.168.51.136 is the first
screen, 192.168.51.134 the second and so on... When i run the command with np
24 then the mapping on each screen is incorrect, as it 136 was no longer
associated with process 0 and so on. So at the end you see the model but the
correspondance between the tiles and process is incorrect. I don't know if i am
specifying wrong the arguments tdx and tdy as tdx=3 and tdy=2 and 3*2=6 which
is different to 24...maybe there is something i am going over which is causing
the incorrect mapping.
2009/2/2 Utkarsh Ayachit utkarsh.ayac...@kitware.com
I knew I should have looked it up before responding :), I stand corrected.
On Mon, Feb 2, 2009 at 1:00 PM, Moreland, Kenneth kmo...@sandia.gov wrote:
Sorry, but that's not true at all. It is in fact encouraged to have more
processors than tiles when driving a tiled display. All the processors,
even the non-display ones, will be involved in the processing and parallel
rendering work.
-Ken
On 1/30/09 11:43 AM, Utkarsh Ayachit utkarsh.ayac...@kitware.com wrote:
The number of processes must match the number of tiles (num of tiles =
tdx * tdy). In your case you have 8 processes are only 2*1 = 3 tiles?
Utkarsh
On Fri, Jan 30, 2009 at 10:21 AM, Camilo Marin
igetmyjunkmailh...@gmail.com wrote:
Hi all,
We are trying to configure and run Paraview 3.4.0 on a tiled display with
the following command:
mpirun -np 8 --mca btl ^openib,udapl --mca btl_tcp_if_exclude lo
--hostfile
/home/imagine/ParaView/hosts /bin/env DISPLAY=:0
~/OpenFOAM/ThirdParty/ParaView3.3-cvs/platforms/linux64Gcc/bin/pvserver
--server-port=1100 -tdx=1 -tdy=2
Then when we connect throught the paraview client it doesn't show in the
two
displays we requested.
So, is there a guideline or some kind of command/configuration we are
missing so it can be displayed as wished?
Thnaks in advance.
___
ParaView mailing list
ParaView@paraview.org
http://www.paraview.org/mailman/listinfo/paraview
___
ParaView mailing list
ParaView@paraview.org
http://www.paraview.org/mailman/listinfo/paraview
Kenneth Moreland
*** Sandia National Laboratories
***
*** *** *** email: kmo...@sandia.gov
** *** ** phone: (505) 844-8919
*** web: http://www.cs.unm.edu/~kmorel
http://www.cs.unm.edu/%7Ekmorel
Kenneth Moreland
*** Sandia National Laboratories
***
*** *** *** email: kmo...@sandia.gov
** *** ** phone: (505) 844-8919
*** web: http://www.cs.unm.edu/~kmorel
___
ParaView mailing list
ParaView@paraview.org
http://www.paraview.org/mailman/listinfo/paraview
Re: [Paraview] How to know the type of a selection ?
First you'll need to locate the selection source proxy i.e. the proxy that is producing the vtkSelection. Look at pqSelectionManager::getIndices to see how to locate the selection source. The selection source proxy will have a property named FieldType. That can be used to determine if it's a cell selection or point selection. Values for this property are: Entry value=0 text=CELL / Entry value=1 text=POINT / Entry value=2 text=FIELD / Entry value=3 text=VERTEX / Entry value=4 text=EDGE / Utkarsh On Tue, Feb 3, 2009 at 8:23 AM, Rafael March rafaelmar...@yahoo.com wrote: Hi, in programming of a plugin, I can get the ids of selected elements or points through pqSelectionManager. I would like to process the selection of cells and of points differently. The question is: how to get the type of the selection, if it is a point or cell one ? I've been looking at pqExtractSelectionsPanel, and it seems it only look at the comboBox displayed in the Panel. How can I know this ? Thanks all, Rafael March. ___ ParaView mailing list ParaView@paraview.org http://www.paraview.org/mailman/listinfo/paraview ___ ParaView mailing list ParaView@paraview.org http://www.paraview.org/mailman/listinfo/paraview
Re: [Paraview] Press Apply Button from Toolbar
Hi David, thanks, I haven't tried the Auto Accept yet, I'm going to try it as soon as possible... It is still a workaround, is there any other possibility? Best regards, Michael David E DeMarle schrieb: Does the Auto Accept toggle work for you? It is under Options-General On Mon, Feb 2, 2009 at 10:40 AM, Michael Geppert michaelgepp...@gmx.net wrote: Hi everyone, I wrote a toolbar plugin with a copy paste function for filter pipelines. My toolbar button just copies the selected part of the pipeline to the selected source. Now I would like my filter to auto-press the apply button of every newly-created filter, is this possible somehow? Thanks! Michael Geppert ___ ParaView mailing list ParaView@paraview.org http://www.paraview.org/mailman/listinfo/paraview ___ ParaView mailing list ParaView@paraview.org http://www.paraview.org/mailman/listinfo/paraview
Re: [Paraview] Paraview on a tiled display
This is just speculation, but it could be an issue with multiple processes
opening windows on the same graphics card. My guess is that a non display
process is drawing a window in front of the display process.
I often run multiple processes on each node in our visualization cluster for
desktop delivery. I get around the problem of X windows on top of each other
by using the --use-offscreen-rendering flag. I don't know how that flag will
behave in tile display mode. Ideally, it will make all the rendering windows
offscreen except for the display windows. If you have to make changes to
ParaView to get that to work, it seems reasonable to push those back into
ParaView proper.
-Ken
On 2/3/09 11:59 AM, Camilo Marin igetmyjunkmailh...@gmail.com wrote:
Hi all,
Thank you very much for the tip on running mpi with --bynode option. We are
nearer from a solution but it still bugging us.
The command is:
mpirun -bynode -np 24 --mca btl ^openib,udapl --mca btl_tcp_if_exclude lo
--hostfile /home/imagine/ParaView/hosts /bin/env DISPLAY=:0
~/OpenFOAM/ThirdParty/ParaView3.3-cvs/platforms/linux64Gcc/bin/pvserver
--server-port=1100 -tdx=3 -tdy=2
Now, when we launch paraview the model is correctly positioned but not all the
tiles show up. We have made several tests and it seems to be a random process
as it turns some tiles on and off without reason. By the way we are using
OpenMPI just as you said guys. This issue could be an OpenMPI problem or its a
local cluster one? Could it be a problem when connecting paraview to the
server? Even sometimes one of the cluster's machine stop responding which
blocks paraview.
I have uploaded some screenshots pointing out the problem. The first image
corresponds to a paraview run which only shows on 2 screens. The second one
runs the same configuration as above but this time shows only on one screen.
The last image corresponds when paraview blocks due to an unkonw (for the
moment) network problem.
Kenneth Moreland
*** Sandia National Laboratories
***
*** *** *** email: kmo...@sandia.gov
** *** ** phone: (505) 844-8919
*** web: http://www.cs.unm.edu/~kmorel
___
ParaView mailing list
ParaView@paraview.org
http://www.paraview.org/mailman/listinfo/paraview
Únete a Bebo
¡Ãnete a Bebo! Créeme, ¡te encantará! Descubre por qué... http://www.bebo.com/in/8585959763a386522642b135 .. Este correo electrónico te ha sido enviado a petición de Lucas Fernández Seivane queve...@gmail.com. No has sido agregado a ninguna lista de distribución. Si no deseas recibir invitaciones de NINGÃN miembro de Bebo, haz clic aquà - http://www.bebo.com/unsub/8585959763a386522642 Bebo, Inc., 795 Folsom St, 6th Floor, San Francisco, CA 94107, USA. ___ ParaView mailing list ParaView@paraview.org http://www.paraview.org/mailman/listinfo/paraview
