chromium-users — Mailing list for all users of the Chromuim rendering system

[Chromium-users] parallel rendering

[daniel h. <dan...@ya...>]

This might seem like a stupid question but does anyone know where the parallel rendering code is? The thing is I'm not really interested in rendering over multiple workstations but I am interested in rendering over multiple GPU cores in parallel. So I was wondering if the algorithms present in your code would be usefull on a single machine, with a high-spec multicore GPU?

[Chromium-users] Force certain extension on single installation?

[Doritos <mtb...@gm...>]

Hi.

Is there a way to force certain(s) extension(s) to no sync to other 
chromium installations?

For example, if I have an extension that I want it installed only on one 
machine.

What can I do?

[Chromium-users] chromium + vbox guest additions

[Aruna J (RBEI/ECF3) <Ar...@in...>]

Hi,

I have ubuntu running on vbox. Guest additions are installed successfully.
The guest addition provide libraries which export chromium APIs (e.g. VBoxOGL.so)

Now I want to use these chromium APIs as opengl renderer in my app.

A small demo linked to VBoxOGL.so, as below gives sementation fault!!
The API crWindowCreate_ptr returns valid integer value. But the app crashes after few milliseconds.

*************************************************************************************
main()
{
 const char *dpyName = "MyApp";
 int visual = CR_RGB_BIT | CR_DEPTH_BIT | CR_DOUBLE_BIT;

 #define LOAD(x) x##_ptr = (x##Proc) crGetProcAddress( #x )
 int ret = XInitThreads();

 LOAD( crWindowCreate );

 ret = crWindowCreate_ptr(dpyName, visual);
}
*************************************************************************************

Semenation fault backtrace shows below:

#0  0xsomeaddress  in XGetGeometry () from /usr/lib/libX11.so.6
#1  0xsomeaddress  in ?? () from /usr/lib/VBoxOGL.so
#2  0x0000000000 in ?? ()

Is it possible to use the chromium APIs as above?
If not what is the correct usage and configuration?
How to fix the above issue?

Please help me.

Best regards,
Cruser

[Chromium-users] help: building project with make

[Francisco O. <fra...@gm...>]

Hi,

I posted to dev, but maybe this is the more appropriate forum for it.

I'm building the project with make in Linux Mint 14 cinnamon running with
nvidia drivers.

I have been able to find many of the dependencies. However, I'm a bit
stock.

In statecopytest , I get some undefined errors :

-------------------------------------------------------------------------------
              Building spheres for Linux (RELEASE)
-------------------------------------------------------------------------------
  Done!

-------------------------------------------------------------------------------
              Building statecopytest for Linux (RELEASE)
-------------------------------------------------------------------------------
Linking statecopytest for Linux
../../lib/Linux/libcrstate.a(state_flush.o): In function `crStateDiffAPI':
state_flush.c:(.text+0x71): undefined reference to `crSPUInitDispatchTable'
state_flush.c:(.text+0x5f): undefined reference to `crSPUCopyDispatchTable'
state_flush.c:(.text+0x83): undefined reference to `crSPUCopyDispatchTable'
collect2: error: ld returned 1 exit status
gmake[3]: *** [../../bin/Linux/statecopytest] Error 1
gmake[2]: *** [dep] Error 2
gmake[1]: *** [statecopytest.subdir] Error 2
make: *** [progs.subdir] Error 2


Can you please tell me what do I need. I did install the package libcr-dev
but I'm not sure if this is the correct package.

Thanks,
Francisco

[Chromium-users] Running wall demo on Windows

[William H. <HWI...@nt...>]

Hi,

I am trying to run wall demo on Windows using Cygwin bash shell as instructed in the Chromium documentation,
After running the mothership through wall.conf script in the mothership/configs folder,
then when trying to run crserver on the same machine (localhost), the crserver quit without leaving any error message.
I tried to trace where it stops, and it leads me to this function call:

render_spu.ws.wglMakeCurrent( context->visual->device_context, context->hRC )

in render_spu.c line 597.

Seems like that function call never returns.

And also, there is something in the mothership process, which I think a bit strange, that is,
After replying to all server param requests, the socket still accepts and
processing command "server machine_name_1", that means it processes that command twice from the same machine.
And so, the mothership replies with "Never heard of server host machine_name_1. Expected one of: machine_name_2", then the crserver quit by itself.
Although in the config(or script) file, I put two different machine name (I want to use two displays).
And I have not even started the crserver on the other machine.

I tried to run the atlantis demo in a single machine and it runs fine, just that when I started using two machines for 2 tiles display,
It went wrong.

Any help would be much appreciated.

Thank you,
William

________________________________
CONFIDENTIALITY: This email is intended solely for the person(s) named and may be confidential and/or privileged. If you are not the intended recipient, please delete it, notify us and do not copy, use, or disclose its content.

Towards A Sustainable Earth: Print Only When Necessary. Thank you.

[Chromium-users] Will the Processing Langauge Binding to OpenGL work with Chromium?

[Thomas M. <tho...@gm...>]

Hi,

Will the Processing Langauge Binding [
http://processing.org/reference/libraries/opengl/index.html] to OpenGL work
with Chromium on a multi-monitor, tiled display?

Thanks,

Thomas

[Chromium-users] Jesux invited you to Dropbox

[Dropbox <no-...@dr...>]

Jesux wants you to use Dropbox to sync and share files online and across computers.

Get started here: http://www.dropbox.com/link/20.DtjAEQCoSv/NjQ4NDY4NTQ1Nw?src=referrals_bulk

- The Dropbox Team

____________________________________________________ 
To stop receiving invites from Dropbox, please go to http://www.dropbox.com/bl/0713db4f78dd/chromium-users%40lists.sourceforge.net

[Chromium-users] chromium error

[Armando A. <aro...@gm...>]

Hi to all, I am using a reassemble.conf from the examples that come with
chromium, when I try run remote by ssh the crserver in the server nodes the
next error occurs, the pcs on that run crserver there are not monitor,
keyboard and mouse

Start a crappfaker on pc3.cluster
Start a crserver on pc3.cluster
Autostart for node pc1: ['/usr/bin/ssh', '-x', '-l', 'cluster1',
'192.168.1.11', 'crserver']
Autostart for node pc2: ['/usr/bin/ssh', '-x', '-l', 'cluster2',
'192.168.1.12', 'crserver']
CR Warning(pc2:3866): Couldn't open X display named '(null)'
CR Warning(pc2:3866): Render SPU: Couldn't get a visual,
renderspu_SystemInitVisual failed
CR Warning(pc2:3866): Render SPU: Couldn't create a window,
renderspuFindVisual returned NULL
CR Error(pc2:3866): Render SPU: Couldn't get a double-buffered, RGB visual
with Z!
CR Warning(pc1:3646): Couldn't open X display named '(null)'
CR Warning(pc1:3646): Render SPU: Couldn't get a visual,
renderspu_SystemInitVisual failed
CR Warning(pc1:3646): Render SPU: Couldn't create a window,
renderspuFindVisual returned NULL
CR Error(pc1:3646): Render SPU: Couldn't get a double-buffered, RGB visual
with Z!

I configured autostart mode to run crserver automatic but the same error
occurs

If somebody had the same problem please tell me how to fix it
thanks

[Chromium-users] I need help

[Armando A. <aro...@gm...>]

Hi to all, I try install a render cluster with chromium, virtualgl and
turboVNC, I try to do a Sort-First Rendering (Image-Space Decomposition)
with Readback,
I don't want to do a display wall, I want to do is run 3d aplications very
heavy on the appnode using the render capabilities of server nodes

I have 4 pcs, the first question, Do I need to install a tool for cluster
like red hat cluster suite or Rocks??
I installed Chromium cr-1.9, I complied and run the example atlantis,
my 4 pcs are:

pc1.cluster 192.168.1.11
pc2.cluster 192.168.1.12
pc3.cluster 192.168.1.13
pc4.cluster 192.168.1.14

pc1 is appnode and pc2, pc3 and pc4 are server nodes
in the appnode use spu tilesort and in the server nodes use spu readback, is
it ok???
My configuration are the next:
I don't understand how to do autostart en the server nodes, how start
crserver of server nodes manually??
If somebody do a similar project please help me
thanks,
excuseme for my english, I speak spanish

  import sys
sys.path.append('/home/cluster1/cr-1.9/mothership/server')
from mothership import *


 app_node = CRApplicationNode('pc1.cluster')

app_spu = SPU('tilesort')
app_node.AddSPU(app_spu)


 cr = CR()
cr.AddNode(app_node)


 server_node1 = CRNetworkNode('pc2.cluster')

server_spu = SPU('readback')

server_node1.AddSPU(server_spu)

cr.AddNode(server_node1)

app_spu.AddServer(server_node1, protocol='tcpip', port=7000)



 server_node2 = CRNetworkNode('pc3.cluster')

server_spu = SPU('readback')

server_node2.AddSPU(server_spu)

cr.AddNode(server_node2)

app_spu.AddServer(server_node2, protocol='tcpip', port=7001)


 server_node3 = CRNetworkNode('pc4.cluster')

server_spu = SPU('readback')

server_node3.AddSPU(server_spu)

cr.AddNode(server_node3)

app_spu.AddServer(server_node3, protocol='tcpip', port=7002)

 cr.Go()

Re: [Chromium-users] Chromium breaks after upgrade to Ubuntu 10.04

[Angel de V. <an...@ia...>]

Hi,

On Tue, 08 Jun 2010 19:39:55 +0100, Kristopher Kalish  
<kal...@du...> wrote:

> I have a tiled display of with 16 nodes, and a master server.  This
> "display wall" previously ran Ubuntu 9.04, but I decided to move
> everything to 10.04.

[...]

> I get the following output:
>
> Real width: 13440, Real height: 4200
> This is Chromium, Version 1.9
> Autostart for node precision13: ['/usr/bin/ssh', 'precision13',
> "/bin/sh -c  
> 'LD_LIBRARY_PATH=/home/display/master4/demos/chromium/cr-1.9/lib/Linux
> /home/display/master4/demos/chromium/cr-1.9/bin/Linux/crserver
> -mothership master4'"]
> ....
> ....
> CR Warning(precision1:1640): Render SPU: Display :0 doesn't have the
> necessary visual: RGB, Doublebuffer, Z
> CR Warning(precision1:1640): Render SPU: Couldn't get a visual,
> renderspu_SystemInitVisual failed
> CR Warning(precision1:1640): Render SPU: Couldn't create a window,
> renderspuFindVisual returned NULL
> CR Error(precision1:1640): Render SPU: Couldn't get a double-buffered,
> RGB visual with Z!

[...]

> Has anyone else experienced a problem like this?

Exactly the same problem here. I decided to try it also with 10.04 and  
then I had to back down to 9.10...  
(http://angel-de-vicente.blogspot.com/2010/05/settings-for-tiled-display-wall.html)

Cheers,
Ángel de Vicente
-- 
+---------------------------------------------+
|                                             |
| http://www.iac.es/galeria/angelv/           |
|                                             |
| High Performance Computing Support PostDoc  |
| Instituto de Astrofísica de Canarias        |
|                                             |
+---------------------------------------------+

[Chromium-users] Chromium breaks after upgrade to Ubuntu 10.04

[Kristopher K. <kal...@du...>]

I have a tiled display of with 16 nodes, and a master server.  This
"display wall" previously ran Ubuntu 9.04, but I decided to move
everything to 10.04.

When I made the move to 10.04 Chromium applications no long launched.

If I do something like:

display@master4:~/master4/demos/chromium/cr-1.9/mothership/configs$
python sweetmural.conf

I get the following output:

Real width: 13440, Real height: 4200
This is Chromium, Version 1.9
Autostart for node precision13: ['/usr/bin/ssh', 'precision13',
"/bin/sh -c 'LD_LIBRARY_PATH=/home/display/master4/demos/chromium/cr-1.9/lib/Linux
/home/display/master4/demos/chromium/cr-1.9/bin/Linux/crserver
-mothership master4'"]
....
....
CR Warning(precision1:1640): Render SPU: Display :0 doesn't have the
necessary visual: RGB, Doublebuffer, Z
CR Warning(precision1:1640): Render SPU: Couldn't get a visual,
renderspu_SystemInitVisual failed
CR Warning(precision1:1640): Render SPU: Couldn't create a window,
renderspuFindVisual returned NULL
CR Error(precision1:1640): Render SPU: Couldn't get a double-buffered,
RGB visual with Z!
...
...
...


All of the nodes have the nvidia drivers. I even double checked that I
was getting hardware acceleration on the nodes with the following
commands:
display@precision1:~$ export DISPLAY=:0
display@precision1:~$ glxinfo
name of display: :0.0
display: :0  screen: 0
direct rendering: Yes
server glx vendor string: NVIDIA Corporation
server glx version string: 1.4
...
...


Has anyone else experienced a problem like this?

[Chromium-users] Chromium with InfiniBand support,

[JesuX <jj....@gm...>]

I try compile C whit IB support but the process fails

[root@master cr-1.9]# make
-------------------------------------------------------------------------------
              Building libcrutil.so for Linux (RELEASE)
-------------------------------------------------------------------------------
Compiling ib.c
ib.c:46:57: mosal.h: No such file or directory
ib.c:51:18: vapi.h: No such file or directory
ib.c:52:19: evapi.h: No such file or directory
ib.c:53:25: vapi_common.h: No such file or directory
.......more error stuff


Some idea ?




-- 
JesuX
www.linpodcast.com/jesux

[Chromium-users] error using XMLHttpRequest.open over HTTPS

[Phil O. <phi...@gm...>]

Hi chromium-users,

I'm running into an issue on the latest Chrome for Windows - 4.0.249.89
(38071). It does not affect the latest version for mac, so hopefully a fix
is already out and just hasn't made it into the Windows releases yet.

While on a page delivered over HTTPS, calls to XMLHttpRequest.send()
generate an exception: "NETWORK_ERR: XMLHttpRequest Exception 101".
readyState is 1 (Open) prior to the call. The same call succeeds when the
page is opened over plain old HTTP. Thoughts and suggestions would be much
appreciated. Here's a small test case, written in PHP (just so the test case
could be contained in a single file):

causeNetworkErr.php:
<?php
if ($_GET['jav']) {
echo '<blink>Not broke!</blink>';
exit;
}
?>

<html>
<head>
<title>Try to break chroooooome</title>

<script type="text/javascript">
function getNewContent() {
var xmlHttp=new XMLHttpRequest();

var url = "causeNetworkErr.php?jav=ya"
xmlHttp.open("GET",url,false);
xmlHttp.send(null);
var rtns=xmlHttp.responseText;

document.getElementById("survey").innerHTML=rtns;
}
</script>

</head>

<body>
<h1>Click the button to break chrome</h1>

<div id="survey">
<button onclick="surveygoinit()">Click me.</button>
</div>

</body>
</html>

[Chromium-users] Weird error when attempting to start crserver remotely.

[Ángel de V. <an...@ia...>]

Hi all,

After a few days at last I managed to run both Google Earth and Stellarium  
in Chromium, but I found something weird...

I installed Chromium, Google Earth and Stellarium from source (except  
Google Earth). I login as angelv to the tile nodes (compute-0-0 and  
compute-0-1) and to the head node. If I start the crserver program in the  
tile nodes when connected locally to them (in the console), with: crserver  
-mothership 192.168.0.1:10000 -port 7000, then I can start "crappfaker "  
in the head node and there is no problem with either Google Earth or  
Stellarium.

If instead, I try to start the cserver remotely in the head node, with:

ssh -x compute-0-0 'bash --login -c "env DISPLAY=:0.0  
/home/angelv/cr-1.9/bin/Linux/crserver -mothership 192.168.0.1:10000 -port  
7000"'

ssh -x compute-0-1 'bash --login -c "env DISPLAY=:0.0  
/home/angelv/cr-1.9/bin/Linux/crserver -mothership 192.168.0.1:10000 -port  
7000"'

then, when I start crappfaker in the head node, compute-0-0 aborts with:

[angelv@vaiven configs]$ ssh -x compute-0-0 'bash --login -c "env  
DISPLAY=:0.0 /home/angelv/cr-1.9/bin/Linux/crserver -mothership  
192.168.0.1:10000 -port 7000"'
CR Info(compute-0-0:9338): Total output dimensions = (5120, 1024)
CR Warning(compute-0-0:9338): EFAULT
CR Warning(compute-0-0:9338): Bad bad bad socket error: Bad address
CR Warning(compute-0-0:9338): Bad juju: 1048576 268760 on socket 0xc

Anyone has any idea why starting remotely like this should crash Chromium?  
Or how could I start remotely the crserver with ssh?

Cheers,
Ángel de Vicente

PS. glxgears runs fine when I start the crservers in the compute nodes via  
ssh...
-- 
+---------------------------------------------+
|                                             |
| http://www.iac.es/galeria/angelv/           |
|                                             |
| High Performance Computing Support PostDoc  |
| Instituto de Astrofísica de Canarias        |
|                                             |
+---------------------------------------------+

[Chromium-users] Graphics hardware requirements

[Simon Su <new...@gm...>]

Hello,

Do I need to have the same graphics hardware on the cluster machines for
Chromium+dmx to work? I want to create a large single desktop from the LCD
panels attached to the cluster nodes.

Chromium compiled with no error on Centos 5.4 x86_64 machines. All I did was
to turn on the DMX flag. Anything else I need to be on the look out before I
begin?


Thank you,
Simon

Re: [Chromium-users] Chromium uses wrong resolution for monitors.

[Ángel de V. <an...@ia...>]

Hi,

On Tue, 26 Jan 2010 22:46:15 -0000, Jackman <kd...@gm...> wrote:

> On Tue, Jan 26, 2010 at 3:12 AM, Angel de Vicente <an...@ia...> wrote:
>> Hi,
>>
>> I'm new to Chromium, and I'm having a (apparently) simple problem,
>> which I hope some of you guys can help me to solve.
>>
>> I'm running Chromium as a part of a Rocks Cluster installation
>> (http://www.rocksclusters.org/wordpress/). For my setting I have a
>> head node and two tile nodes, each connected
>> to two monitors. Chromium seems to be working more or less OK, but it
>> seems to believe that the resolution of my "tile" nodes is 1920x1200
>> where it actually is 1280x1024. Any pointers on where should I look so
>> that I can tell Chromium about the actual resolution of my monitors? I
>> include below the output when I run glxgears.
>
> Just out of curiosity, what video cards are you using?  Can we see
> some of your configuration details?  Thanks!

thanks for the help. In the end I found how to tell the Rocks software the  
resolution of my tile nodes, so this is not an issue anymore, although I'm  
still stuck, perhaps you can shed some light?

The graphic cards used are:
The head node: nVidia Corporation Quadro NVS 290 (rev a1)
The tile nodes: nVidia Corporation NV43GL [Quadro FX 550] (rev a2)
                 nVidia Corporation NV44 [Quadro NVS 285] (rev a1)

And now the problem is when I try to use Google Earch or Stellarium (so  
far I have only tried these two, and glxgears). glxgears is fine after  
enabling chromium, and both Google Earth and Stellarium work fine in the  
head node when Chromium is disabled, but when I enable it I get the errors  
that I attach below. Any idea? Thanks a lot,

Ángel de Vicente



GOOGLE EARTH
===============

[angelv@vaiven ~]$ googleearth
CR Warning(vaiven:9210): the OpenGL faker was loaded without crappfaker!
Defaulting to an application id of -1!
This won't work if you're debugging a parallel application!
In this case, set the CR_APPLICATION_ID_NUMBER environment
variable to the right thing (see opengl_stub/load.c)
CR Warning(vaiven:9210): Using Chromium configuration for * from
/home/angelv/.crconfigs
This is Chromium, Version 1.9
Autostart for node tile-0-0: ['/usr/bin/ssh', '-x', 'tile-0-0', "bash
--login -c 'env DISPLAY=:0.0 /opt/viz/bin/crserver -mothership
vaiven:10046'"]
Autostart for node tile-0-0: ['/usr/bin/ssh', '-x', 'tile-0-0', "bash
--login -c 'env DISPLAY=:0.1 /opt/viz/bin/crserver -mothership
vaiven:10046'"]
Autostart for node tile-0-1: ['/usr/bin/ssh', '-x', 'tile-0-1', "bash
--login -c 'env DISPLAY=:0.0 /opt/viz/bin/crserver -mothership
vaiven:10046'"]
Autostart for node tile-0-1: ['/usr/bin/ssh', '-x', 'tile-0-1', "bash
--login -c 'env DISPLAY=:0.1 /opt/viz/bin/crserver -mothership
vaiven:10046'"]
Start a crappfaker on vaiven.ll.iac.es
Mothership signalling spawning process 9210
CR Info(tile-0-0:26576): Total output dimensions = (5120, 1024)
CR Info(tile-0-1:26582): Total output dimensions = (5120, 1024)
CR Info(tile-0-0:26568): Total output dimensions = (5120, 1024)
CR Info(tile-0-1:26576): Total output dimensions = (5120, 1024)
CR Warning(vaiven:9210): glXChooseVisual: ignoring attribute 0x22
CR Warning(tile-0-0:26576): __tcpip_read_exact() error: Bad address
CR Warning(tile-0-0:26576): Bad juju: 10485760 197100 on socket 0xb
CR Warning(tile-0-0:26576): CRServer: Last client disconnected - exiting.
Google Earth has caught signal 11.



We apologize for the inconvenience, but Google Earth has crashed.
  This is a bug in the program, and should never happen under normal
  circumstances. A bug report and debugging data have been written
  to this text file:

     /home/angelv/.googleearth/crashlogs/crashlog-4b5ecb14.txt

Please include this file if you submit a bug report will to Google.
CR Warning(tile-0-0:26568): CRServer: Last client disconnected - exiting.
CR Warning(tile-0-1:26582): CRServer: Last client disconnected - exiting.
[angelv@vaiven ~]$ CR Warning(tile-0-1:26576): CRServer: Last client
disconnected - exiting.



STELLARIUM
============
[angelv@vaiven ~]$ stellarium
  -------------------------------------------------------
[ This is Stellarium 0.8.0 - http://www.stellarium.org ]
[ Copyright (C) 2000-2005 Fabien Chereau et al         ]
  -------------------------------------------------------
CR Warning(vaiven:9191): YOU ARE CALLING A NULLED FUNCTION (Viewport)
CR Warning(vaiven:9191): YOU ARE CALLING A NULLED FUNCTION (MatrixMode)
CR Warning(vaiven:9191): YOU ARE CALLING A NULLED FUNCTION (LoadMatrixd)
CR Warning(vaiven:9191): YOU ARE CALLING A NULLED FUNCTION (MatrixMode)
CR Warning(vaiven:9191): YOU ARE CALLING A NULLED FUNCTION (MatrixMode)
CR Warning(vaiven:9191): YOU ARE CALLING A NULLED FUNCTION (LoadMatrixd)
CR Warning(vaiven:9191): YOU ARE CALLING A NULLED FUNCTION (MatrixMode)
CR Warning(vaiven:9191): YOU ARE CALLING A NULLED FUNCTION (FrontFace)
CR Warning(vaiven:9191): the OpenGL faker was loaded without crappfaker!
Defaulting to an application id of -1!
This won't work if you're debugging a parallel application!
In this case, set the CR_APPLICATION_ID_NUMBER environment
variable to the right thing (see opengl_stub/load.c)
CR Warning(vaiven:9191): Using Chromium configuration for * from
/home/angelv/.crconfigs
This is Chromium, Version 1.9
Autostart for node tile-0-0: ['/usr/bin/ssh', '-x', 'tile-0-0', "bash
--login -c 'env DISPLAY=:0.0 /opt/viz/bin/crserver -mothership
vaiven:10073'"]
Autostart for node tile-0-0: ['/usr/bin/ssh', '-x', 'tile-0-0', "bash
--login -c 'env DISPLAY=:0.1 /opt/viz/bin/crserver -mothership
vaiven:10073'"]
Autostart for node tile-0-1: ['/usr/bin/ssh', '-x', 'tile-0-1', "bash
--login -c 'env DISPLAY=:0.0 /opt/viz/bin/crserver -mothership
vaiven:10073'"]
Autostart for node tile-0-1: ['/usr/bin/ssh', '-x', 'tile-0-1', "bash
--login -c 'env DISPLAY=:0.1 /opt/viz/bin/crserver -mothership
vaiven:10073'"]
Start a crappfaker on vaiven.ll.iac.es
Mothership signalling spawning process 9191
CR Info(tile-0-0:26451): Total output dimensions = (5120, 1024)
CR Info(tile-0-1:26451): Total output dimensions = (5120, 1024)
CR Info(tile-0-0:26443): Total output dimensions = (5120, 1024)
CR Info(tile-0-1:26453): Total output dimensions = (5120, 1024)
CR Warning(vaiven:9191): glXChooseVisual: ignoring attribute 0x22
Application locale is system
Loading Solar System data...(loaded)
Loading location: "Paris", on Earth (landscape is: "guereins")
CR Warning(tile-0-0:26443): __tcpip_read_exact() error: Bad address
CR Warning(tile-0-0:26443): Bad juju: 10485760 262852 on socket 0xb
CR Warning(tile-0-0:26443): CRServer: Last client disconnected - exiting.
/usr/bin/stellarium: line 3:  9191 Segmentation fault      run-stellarium
CR Warning(tile-0-0:26451): __tcpip_read_exact() error: Connection reset  
by peer
CR Warning(tile-0-0:26451): CRServer: Last client disconnected - exiting.
CR Warning(tile-0-1:26453): __tcpip_read_exact() error: Connection reset  
by peer
CR Warning(tile-0-1:26453): CRServer: Last client disconnected - exiting.
CR Warning(tile-0-1:26451): __tcpip_read_exact() error: Connection reset  
by peer
CR Warning(tile-0-1:26451): CRServer: Last client disconnected - exiting.
[angelv@vaiven ~]$


-- 
+---------------------------------------------+
|                                             |
| http://www.iac.es/galeria/angelv/           |
|                                             |
| High Performance Computing Support PostDoc  |
| Instituto de Astrofísica de Canarias        |
|                                             |
+---------------------------------------------+

[Chromium-users] Dynamic Tile Sorting for Loading Balancing

[David H. <dav...@gm...>]

Hey guys,

I'm a graduate student from Rochester Institute of Technology and I'm doing
a study on load balancing in distributed farming. I came across Chromium and
as I was reading through everything I thought it would be a good testbed for
my project.

I'm totally new to this project and I've been digging through all the
documentation, script and source. And I was wondering I could get some
directions on where I should look.

I want to do some dynamic load balancing through the 'tilesort' SPU. But I'm
not sure where the scheduler is in Chromium. I want to measure the amount of
time it takes for each tile. I have a feeling it's either in 'tilesort' or
in 'render', but i'm not sure.

Any ideas?


Thanks,

David Huynh
dx...@ri...

Re: [Chromium-users] Chromium uses wrong resolution for monitors.

[Jackman <kd...@gm...>]

On Tue, Jan 26, 2010 at 3:12 AM, Angel de Vicente <an...@ia...> wrote:
> Hi,
>
> I'm new to Chromium, and I'm having a (apparently) simple problem,
> which I hope some of you guys can help me to solve.
>
> I'm running Chromium as a part of a Rocks Cluster installation
> (http://www.rocksclusters.org/wordpress/). For my setting I have a
> head node and two tile nodes, each connected
> to two monitors. Chromium seems to be working more or less OK, but it
> seems to believe that the resolution of my "tile" nodes is 1920x1200
> where it actually is 1280x1024. Any pointers on where should I look so
> that I can tell Chromium about the actual resolution of my monitors? I
> include below the output when I run glxgears.

Just out of curiosity, what video cards are you using?  Can we see
some of your configuration details?  Thanks!

-- 
Andrew Jackman
kd...@gm...

CONFIDENTIALITY NOTICE: This e-mail message, including any
attachments, is for the sole use of the intended recipient(s) and may
contain confidential and privileged information. Any unauthorized
review, use, disclosure, or distribution is prohibited. If you are not
the intended recipient, please contact the sender by reply e-mail and
destroy all copies of the original message.  All your base are belong
to us.

[Chromium-users] Chromium uses wrong resolution for monitors.

[Angel de V. <an...@ia...>]

Hi,

I'm new to Chromium, and I'm having a (apparently) simple problem,
which I hope some of you guys can help me to solve.

I'm running Chromium as a part of a Rocks Cluster installation
(http://www.rocksclusters.org/wordpress/). For my setting I have a
head node and two tile nodes, each connected
to two monitors. Chromium seems to be working more or less OK, but it
seems to believe that the resolution of my "tile" nodes is 1920x1200
where it actually is 1280x1024. Any pointers on where should I look so
that I can tell Chromium about the actual resolution of my monitors? I
include below the output when I run glxgears.

Thanks a lot,
Ángel de Vicente

[angelv@vaiven ~]$ glxgears
CR Warning(vaiven:8016): the OpenGL faker was loaded without crappfaker!
Defaulting to an application id of -1!
This won't work if you're debugging a parallel application!
In this case, set the CR_APPLICATION_ID_NUMBER environment
variable to the right thing (see opengl_stub/load.c)
CR Warning(vaiven:8016): Using Chromium configuration for * from
/home/angelv/.crconfigs
This is Chromium, Version 1.9
Autostart for node tile-0-0: ['/usr/bin/ssh', '-x', 'tile-0-0', "bash
--login -c 'env DISPLAY=:0.0 /opt/viz/bin/crserver -mothership
vaiven:10098'"]
Autostart for node tile-0-0: ['/usr/bin/ssh', '-x', 'tile-0-0', "bash
--login -c 'env DISPLAY=:0.1 /opt/viz/bin/crserver -mothership
vaiven:10098'"]
Autostart for node tile-0-1: ['/usr/bin/ssh', '-x', 'tile-0-1', "bash
--login -c 'env DISPLAY=:0.0 /opt/viz/bin/crserver -mothership
vaiven:10098'"]
Autostart for node tile-0-1: ['/usr/bin/ssh', '-x', 'tile-0-1', "bash
--login -c 'env DISPLAY=:0.1 /opt/viz/bin/crserver -mothership
vaiven:10098'"]
Start a crappfaker on vaiven.ll.iac.es
Mothership signalling spawning process 8016
CR Info(tile-0-0:22229): Total output dimensions = (7680, 1200)
CR Warning(tile-0-0:22229): No room for 1920x1200 tile in this
server's window (1280 x 1024)!
CR Warning(tile-0-0:22229): No room for 1920x1200 tile in this
server's window (1280 x 1024)!
CR Info(tile-0-0:22231): Total output dimensions = (7680, 1200)
CR Warning(tile-0-0:22231): No room for 1920x1200 tile in this
server's window (1280 x 1024)!
CR Warning(tile-0-0:22231): No room for 1920x1200 tile in this
server's window (1280 x 1024)!
CR Info(tile-0-1:22247): Total output dimensions = (7680, 1200)
CR Warning(tile-0-1:22247): No room for 1920x1200 tile in this
server's window (1280 x 1024)!
CR Warning(tile-0-1:22247): No room for 1920x1200 tile in this
server's window (1280 x 1024)!
CR Info(tile-0-1:22251): Total output dimensions = (7680, 1200)
CR Warning(tile-0-1:22251): No room for 1920x1200 tile in this
server's window (1280 x 1024)!
CR Warning(tile-0-1:22251): No room for 1920x1200 tile in this
server's window (1280 x 1024)!
CR Warning(tile-0-0:22231): No room for 1920x1200 tile in this
server's window (1280 x 1024)!
CR Warning(tile-0-0:22231): No room for 1920x1200 tile in this
server's window (1280 x 1024)!
CR Warning(tile-0-0:22229): No room for 1920x1200 tile in this
server's window (1280 x 1024)!
CR Warning(tile-0-0:22229): No room for 1920x1200 tile in this
server's window (1280 x 1024)!
CR Warning(tile-0-1:22247): No room for 1920x1200 tile in this
server's window (1280 x 1024)!
CR Warning(tile-0-1:22247): No room for 1920x1200 tile in this
server's window (1280 x 1024)!
CR Warning(tile-0-1:22251): No room for 1920x1200 tile in this
server's window (1280 x 1024)!
CR Warning(tile-0-1:22251): No room for 1920x1200 tile in this
server's window (1280 x 1024)!

Re: [Chromium-users] Chromium on 64-bit platform?

[Jürgen L. <ga...@hn...>]

Hi,

most probably the game you are running is a 32 bit version. Thus you
need a 64 bit version of the game or a 32 bit version of chromium.

Try to set

FORCE_32BIT_ABI=1

in options.mk and run make again to build a 32 bit version of chromium.
You?ll need the 32 bit versions of libraries chromium depends on for this.

regards

Jürgen


Killian De Smedt schrieb:
> Hi
>
> I'm trying to run chromium on 2 64-bit linux (ubuntu karmic) machines.
> Compiling, running simple applications across those machines, ...
> works fine already.
>
> Now I'm trying to step up a little bit and trying to run a game (
> return to castle wolfenstein). This should work as far as I know as it
> is still using an opengl version prior to 2.0 and is based on the same
> engine as quake3 ( that has already been used in some chromium setups).
>
> Now, when running a simple chromium setup (1 appfaker and 1 server)
> the appfaker just doesn't catch the dll call, I don't know why, but it
> seems to miss the dlopen or something...
>
> Then I edited the script to load the libcrfaker.so directly into rtcw
> (using the +set glDriver libcrfaker.so). It then found the library but
> gave following errors:
> ...loading libcrfaker.so: QGL_Init: Can't load libcrfaker.so from
> /etc/ld.so.conf or current dir:
> /home/chromium/wolfenstein/libcrfaker.so: wrong ELF class: ELFCLASS64
> failed
> Afterwards it fell back at using my normal drivers:
> ...loading libGL.so.1: Initializing OpenGL display
>
> As far as I can tell this is because of a 64-bit issue? I should use a
> 32-bit library, but I can't build one on this system. Are there any
> other solutions or am I doing something else wrong?
>
> I know this list has become a little unactive, but I really need this
> to work.
>
> thank you for your time
>
> Killian De Smedt
> ------------------------------------------------------------------------
>
> ------------------------------------------------------------------------------
> Come build with us! The BlackBerry(R) Developer Conference in SF, CA
> is the only developer event you need to attend this year. Jumpstart your
> developing skills, take BlackBerry mobile applications to market and stay 
> ahead of the curve. Join us from November 9 - 12, 2009. Register now!
> http://p.sf.net/sfu/devconference
> ------------------------------------------------------------------------
>
> _______________________________________________
> Chromium-users mailing list
> Chr...@li...
> https://lists.sourceforge.net/lists/listinfo/chromium-users
>   


-- 
Hiermit widerspreche ich/wir der Nutzung oder Uebermittlung 
meiner/unserer Daten fuer Werbezwecke oder fuer die Markt- oder 
Meinungsforschung gem. Par. 28 Abs. 3 Bundesdatenschutzgesetz.

[Chromium-users] integrate Chromium with VisIt

[liu n. <ta...@gm...>]

Hi everyone,

There is a cluster doing simulation computation and I want to use
VisIt to visualize the computation results and show the image  on a
display wall. But I have no idea on how to integrate VisIt with
Chromium to accomplish what I want. Can someone to be kind enough to
gave me some instructions?

Thanks in advance.

-Ning

[Chromium-users] Chromium on 64-bit platform?

[Killian De S. <kil...@gm...>]

Hi

I'm trying to run chromium on 2 64-bit linux (ubuntu karmic) machines.
Compiling, running simple applications across those machines, ... works fine
already.

Now I'm trying to step up a little bit and trying to run a game ( return to
castle wolfenstein). This should work as far as I know as it is still using
an opengl version prior to 2.0 and is based on the same engine as quake3 (
that has already been used in some chromium setups).

Now, when running a simple chromium setup (1 appfaker and 1 server) the
appfaker just doesn't catch the dll call, I don't know why, but it seems to
miss the dlopen or something...

Then I edited the script to load the libcrfaker.so directly into rtcw (using
the +set glDriver libcrfaker.so). It then found the library but gave
following errors:
...loading libcrfaker.so: QGL_Init: Can't load libcrfaker.so from
/etc/ld.so.conf or current dir: /home/chromium/wolfenstein/libcrfaker.so:
wrong ELF class: ELFCLASS64
failed
Afterwards it fell back at using my normal drivers:
...loading libGL.so.1: Initializing OpenGL display

As far as I can tell this is because of a 64-bit issue? I should use a
32-bit library, but I can't build one on this system. Are there any other
solutions or am I doing something else wrong?

I know this list has become a little unactive, but I really need this to
work.

thank you for your time

Killian De Smedt

Re: [Chromium-users] GLEW Error

[Stephen F. <ste...@gm...>]

Hello,
In the event anyone else tries, since I couldn't find anything but one
lonely thread on the internet...
The solution is to add all of the glX*() calls to the bottom of the
function[] in getprocaddress.py, although the proper solution would probably
be to create a glXapi_parser or something.

I can provide the files for anyone interested, and running "python
crdemo.conf glewinfo > out1" gives some pretty neat information, especially
combined with a diff of "glewinfo > out2"

-Stephen

On Tue, Sep 15, 2009 at 9:33 AM, Stephen Furlani
<ste...@gm...>wrote:

> Hello,
> I know I've been saturating the list lately, but I have some more
> questions.
>
> I've downloaded and run MeVisLab fine by itself.
>
> However, when I run it with Chromium, I get the error:
> (MeVisLab) Cerr: GLEW Error: GLX 1.2 and up are not supported.
>
> And some warnings:
> CR Warning(rockstest: 24536): crPackReleaseBuffer called with no current
> buffer
>
> although it will do everything except display images...
>
> any thoughts?  can I update the GLEW with chromium?
>
> --
> -Stephen Furlani
> *I'm moving to a new email*
> ste...@gm...
> Please update your address books, thanks!
>



-- 
-Stephen Furlani
*I'm moving to a new email*
ste...@gm...
Please update your address books, thanks!

[Chromium-users] GLEW Error

[Stephen F. <ste...@gm...>]

Hello,
I know I've been saturating the list lately, but I have some more questions.

I've downloaded and run MeVisLab fine by itself.

However, when I run it with Chromium, I get the error:
(MeVisLab) Cerr: GLEW Error: GLX 1.2 and up are not supported.

And some warnings:
CR Warning(rockstest: 24536): crPackReleaseBuffer called with no current
buffer

although it will do everything except display images...

any thoughts?  can I update the GLEW with chromium?

-- 
-Stephen Furlani
*I'm moving to a new email*
ste...@gm...
Please update your address books, thanks!

Re: [Chromium-users] Spew SPU

[Stephen F. <ste...@gm...>]

Ok, so I've got it to work, marginally.
I get one of two situations depending on how I set it up:

1) app_window renders fine, but Video Wall renders at app_window resolution
in corner OR stretches pixelmap to entire wall
2) Video Wall renders fine, but app_window renders a small portion of the
video wall.

neither of these cases satisfy.

help!

-S!


On Wed, Sep 9, 2009 at 8:01 AM, Stephen Furlani
<ste...@gm...>wrote:

> Broadcast is not performing as expected...
> I've got the config:
>
> App[Tilesort - Broadcast] --> Server[Render - toAppWindow]
>    |
>   V
> Server[Tilesort - Test all Tiles] --> 2x Server[Render - toVideoWall]
>
> But the program never starts.  And if I set any render or tile size larger
> than 640x480 it complains that it can't "fit" it in the tile.  I've tried
> setting everything under the Config.py program, but I can't get this to
> work.
>
> -S!
>
> 2009/9/8 Sean Ahern <ah...@or...>
>
> Stephen Furlani wrote:
>> > Okay, so I want to make a "Spew" SPU that takes the commands and sends
>> > identical commands to all of it's nodes.
>> >
>> > ... I have to write a function for each individual OpenGL command!?
>> >  SPUs like Passthrough and Pack have 600+ commands...
>>
>> Why not just run the existing Tilesort SPU in 'Broadcast' mode?  It
>> sends all primitives to all downstream servers.
>>
>> See "bucket_mode" at
>> http://chromium.sourceforge.net/doc/configoptions.html#tilesort
>>
>> -Sean
>>
>> __
>> Sean Ahern
>> Oak Ridge National Laboratory
>> 865-241-3748
>> AIM: ornlsean
>>
>
>
>
> --
> -Stephen Furlani
> *I'm moving to a new email*
> ste...@gm...
> Please update your address books, thanks!
>



-- 
-Stephen Furlani
*I'm moving to a new email*
ste...@gm...
Please update your address books, thanks!