Re: [osg-users] OpenThreads, scheduling, etc ... policies
Hi, Checked out the latest trunk version myself and I am still having the same issues here with the OpenThread calls and the example code I sent to Nick. We are running OpenSuse 12.3 64-bit here. The calls all return good status. In fact the getSchedulePolicy and the getSchedulePriority return the same level that was set, but the linux system calls do not show any change in the threads priority, scheduling policy or affinity. Tried running the example as normal user and the root user and get the same exact results so does not seem to be any sort of issue with priviledges at this point either. Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=59262#59262 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] OpenThreads, scheduling, etc ... policies
Hi Robert, My first problem was that I had selected the option to build OpenThreads with QT in cmake for some reason. So I rebuilt with this option off.. My apologies on that issue... At least it is now using the pthreads section properly. However, as Nick stated in his post, the openthreads code to allow me to select a scheduling policy other than OTHER has been ifdef'd out for __linux__ even though these options are available on Linux. This is also affecting the ability to set a proper thread priority value. So while Nick and I can fix this by removing the __linux__ checks, I would like to see if anyone knows why this was put in there in the first place so that we do not possibly break something else as we are attempting to fix the issues that we are seeing. ... Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=59289#59289 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] OpenThreads, scheduling, etc ... policies
Nick, __linux__ is defined someplace in the gcc compiler as it finds the current arch and os, etc., installed. you can: cpp -dM /dev/null and see all the gcc defines output. I see that __linux is still there as well, at least for the time being. ALLOW_PRIORITY_SCHEDULING, however seems to be an OpenThreads specific define. ... Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=59293#59293 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
[osg-users] How to position osg Window on top or below other X11 apps on Linux
Hi everyone, I am trying to find the best method to be able to control our osg windows such that we can force them on top of other KDE/X11 apps or push them below other KDE/X11 apps as necessary while our osg based application is running. The 64bit Linux system is running a KDE desktop for the window manager. Any hints or suggestions would be greatly appreciated. ... Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=59639#59639 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] How to position osg Window on top or below other X11 apps on Linux
Hi Marco, Thank you for your reply... Actually I have already been talking with Trajce Nikolov(Nick) about this yesterday as he does some consulting work for us from time to time. He has helped us a lot and I highly recommend him for his expertise... Unfortunately, that window handle method does not seem to be available under Linux so we are searching for a method that will work under Linux. After spending some time online trying to find some info on this, seems there are some posts asking similar things around, but I did not find anyone that had a good working result. Seems that with the variety of window managers out there, they all apparently have different ways of handling this type of thing. I was able to go into my KDE desktop settings and using the application name setup the kwinrulesrc file to force one of my osg window ontop of any others. So I know there is a way to accomplish what we need, just need to find the API to accomplish this in my code now. Something we can turn on/off as necessary for various osg windows as they open and close. ... Thank you again for your reply.. Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=59643#59643 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] How to position osg Window on top or below other X11 apps on Linux
Nick, The strange thing is that KDE is a Qt based desktop, even when I use a QT based application window I cannot force it to be ontop of other applications either. If there are multiple Widgets in one application the Qt ontop stuff works, but when the widgets are from separate apps it does not. I understand that there are issues when everyone wants to be ONTOP of someone else, someone is going to loose...so the desktop manager ends up doing something for this. Not sure if in our case this is a KDE4 API issue or an X11 issue or some combination of both that we need to address. ... Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=59646#59646 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] How to position osg Window on top or below other X11 apps on Linux
Nick, I am looking at it now to see if I can figure out the API and make the calls correctly. ... Thank you! Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=59649#59649 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] How to position osg Window on top or below other X11 apps on Linux
NIck, I do not have this working yet but: what about a lowerWindow()..meothd .how do you reset this once you set it? raiseWindow takes no args so I do not see a method to return the window back to the previous state once its raised. I guess if you raise another window after this one it pushes this one back down possibly?? .. Thank you! Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=59650#59650 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
[osg-users] HowTo create an X11 GraphicsWindow without a closer decoration on Linux?
Hi, I have looked and only found an old 2009 post about the this topic. Has anything changed in this area that would allow me to easily create an X11 GraphicsWindow that does not have the closer "X" decoration on it, or at the very least a method to ignore the close event? Looking at the source to OSG 3.3.1 did not give me any clues. ... Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=60086#60086 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
[osg-users] [build] Conflict between current osg trunk 3.3.8 and QT version 5.4.1 on types GLint64 and GLuint64
Hi, I am currently getting the following compiler errors when attempting to compile against the current OSG trunk 04/28/2015, numbered as 3.3.8 and my installed QT release of 5.4.1 I am currently running on an OpenSuSE 13.2 64bit linux distro. /usr/include/qt5/QtGui/qopengl.h:242:17: error: conflicting declaration ‘typedef int64_t GLint64’ typedef int64_t GLint64; ^ In file included from /usr/local/include/osg/GLDefines:25:0, from /usr/local/include/osg/GLExtensions:18, from /usr/local/include/osg/Shader:25, from /usr/local/include/osg/StateAttribute:20, from /usr/local/include/osg/StateSet:18, from /usr/local/include/osg/Node:19, from /usr/local/include/osg/NodeVisitor:17, from /usr/local/include/osgUtil/UpdateVisitor:17, from /usr/local/include/osgViewer/CompositeViewer:18, from ../../../libs/inc/QtOsg/viewer_widget.h:5, from src/widgets/muselogowidget.h:4, from src/widgets/muselogowidget.cpp:1: /usr/local/include/osg/GL:124:31: error: ‘GLint64’ has a previous declaration as ‘typedef long long int GLint64’ typedef long long int GLint64; Same basic error for the GLuint64 type as well. This sort of looks to me like a 64bit versus 32bit operating system issue of some sort when I look at both definitions, but I am no expert in this area so my apologies if I am making this statement in error. Is there something that needs to be defined or setup properly for the code to properly compile against this new version of OSG on a 64bit operating system or am I possibly looking at a bug in the QT release I am using? Thank you everyone for you help in this matter! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=63554#63554 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] [build] Conflict between current osg trunk 3.3.8 and QT version 5.4.1 on types GLint64 and GLuint64
Hi Robert, I am not so sure I can change the include order or not. The QT project I am trying to compile pulls in the QT OpenGL stuff by itself as part of its build process, I am not explicitly including any of the QT include files in my source files directly. I will take a look and see. Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=63557#63557 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] [build] Conflict between current osg trunk 3.3.8 and QT version 5.4.1 on types GLint64 and GLuint64
Hi Robert, After much searching, I finally found the issue on my system. One of our 3rdparty products had its own copy of GL/gl.x files in it and this one was unfortunately being found first. Renaming this old include directory seems to have solved this issue. Thank you for pointing me in the correct direction... Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=63562#63562 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] [build] Conflict between current osg trunk 3.3.8 and QT version 5.4.1 on types GLint64 and GLuint64
Hi Robert, I do not see any place to close the post, or mark it as solved. Am I missing something in the forum? I do not want to leave it as is if there is something I need to do after you helped me get this solved... Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=63563#63563 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
[osg-users] Segfault occuring in -- void Text::drawImplementation(osg::State& state, const osg::Vec4& colorMultiplier) const -- after updating from osg 3.3.1 to the current trunk
Hi,
We have an application that is outputting a number of osgText::Text objects as
well as some osgLabel objects.
Everything appears to be working fine when compiled against OSG 3.3.1
On the current trunk as of yesterday, I am getting a segfault in osgText
Text.cpp drawImplementation() function.
Code:
if ( !_textureGlyphQuadMap.empty() )
{
const GlyphQuads& glyphquad = (_textureGlyphQuadMap.begin())->second;
if ( glyphquad._transformedCoords[contextID]->empty() )
{
computePositions(contextID);
}
}
In the snippet above, the glyphquad._transformedCoords has 3 items in it.
Items 0 and 1 contain what appear to be valid pointers, but item 2 contains a
NULL pointer. So when the contextID for the current incoming state is 2, we
get the segfault.
I have gone back and looked at the Text and Label objects and they are all
setup with a datavariance of Dynamic since the text can change based on user
selection during runtime, just as something I thought might be related.
I know this is a long shot with such little information, but is there anything
that anyone can tell me that might cause something like this to occur? Has
something changed in this area since OSG release 3.3.1 that we possibly need to
account for now?
Sorry about the code snippet output, not sure what is causing the ?? output in
the code area, as they are just spaces in there.
Thank you,
Cheers
Curtis
--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=63572#63572
___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] Segfault occuring in -- void Text::drawImplementation(osg::State& state, const osg::Vec4& colorMultiplier) const -- after updating from osg 3.3.1 to the current trunk
Hi Robert, The application that is having this issue is very large, so I will have to dig into it and see about finding the exact location of the segfault. The author of the app no longer works with us so this will take some time to investigate. The application is one where OSG is an object inside of a QT5 QTWidget. In the qt app, it is crashing in the paintEvent where its calling ViewerBase::frame(), so not very direct as to where the problem actually is in this case. I will see what I can find and if possible setup a small example that duplicates the issue. As for you question about the number of contexts, I can only say that at this time I am unsure if there is any graphics contexts being added as the application is running. The application emulates an aircraft Multifunction Display, so there are many different pages of text and graphics that get displayed. It is possible that a new graphics context could be being added during runtime. As I mentioned the author no longer works with us so we will have to debug this and see what we can determine. .. Thank you for your quick response.. Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=63583#63583 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] Segfault occuring in -- void Text::drawImplementation(osg::State& state, const osg::Vec4& colorMultiplier) const -- after updating from osg 3.3.1 to the current trunk
Hi Everyone,
Attached to this post is an example that causes the segv to occur that I am
referring to in my first post on the subject. This example is the combination
of portions the osgviewerQT and osgText examples as provided with this same
release of OpenSceneGraph.
Hopefully with this example failing it can be determined what is happening.
...
Thank you!
Cheers,
Curtis
--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=63591#63591
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
#include
osg::Group* createHUDText()
{
osg::Group* rootNode = new osg::Group;
osgText::Font* font = osgText::readFontFile("fonts/arial.ttf");
//osg::setNotifyLevel(osg::INFO);
osg::Geode* geode = new osg::Geode;
rootNode->addChild(geode);
bool useVBOs = false;
#if defined(OSG_GL3_AVAILABLE) && !defined(OSG_GL2_AVAILABLE) && !defined(OSG_GL1_AVAILABLE)
useVBOs = true;
osg::Program* program = new osg::Program;
program->addShader(new osg::Shader(osg::Shader::VERTEX, gl3TextVertexShader));
program->addShader(new osg::Shader(osg::Shader::FRAGMENT, gl3TextFragmentShader));
geode->getOrCreateStateSet()->setAttributeAndModes(program, osg::StateAttribute::ON);
#endif
float windowHeight = 1024.0f;
float windowWidth = 1280.0f;
float margin = 50.0f;
//
// Examples of how to set up different text layout
//
osg::Vec4 layoutColor(1.0f,1.0f,0.0f,1.0f);
float layoutCharacterSize = 20.0f;
{
osgText::Text* text = new osgText::Text;
text->setUseVertexBufferObjects(useVBOs);
text->setFont(font);
text->setColor(layoutColor);
text->setCharacterSize(layoutCharacterSize);
text->setPosition(osg::Vec3(margin,windowHeight-margin,0.0f));
// the default layout is left to right, typically used in languages
// originating from europe such as English, French, German, Spanish etc..
text->setLayout(osgText::Text::LEFT_TO_RIGHT);
text->setText("text->setLayout(osgText::Text::LEFT_TO_RIGHT);");
geode->addDrawable(text);
}
{
osgText::Text* text = new osgText::Text;
text->setUseVertexBufferObjects(useVBOs);
text->setFont(font);
text->setColor(layoutColor);
text->setCharacterSize(layoutCharacterSize);
text->setPosition(osg::Vec3(windowWidth-margin,windowHeight-margin,0.0f));
// right to left layouts would be used for hebrew or arabic fonts.
text->setLayout(osgText::Text::RIGHT_TO_LEFT);
text->setAlignment(osgText::Text::RIGHT_BASE_LINE);
text->setText("text->setLayout(osgText::Text::RIGHT_TO_LEFT);");
geode->addDrawable(text);
}
{
osgText::Text* text = new osgText::Text;
text->setUseVertexBufferObjects(useVBOs);
text->setFont(font);
text->setColor(layoutColor);
text->setPosition(osg::Vec3(margin,windowHeight-margin,0.0f));
text->setCharacterSize(layoutCharacterSize);
// vertical font layout would be used for asian fonts.
text->setLayout(osgText::Text::VERTICAL);
text->setText("text->setLayout(osgText::Text::VERTICAL);");
geode->addDrawable(text);
}
//
// Examples of how to set up different font resolution
//
osg::Vec4 fontSizeColor(0.0f,1.0f,1.0f,1.0f);
float fontSizeCharacterSize = 30;
osg::Vec3 cursor = osg::Vec3(margin*2,windowHeight-margin*2,0.0f);
{
osgText::Text* text = new osgText::Text;
text->setUseVertexBufferObjects(useVBOs);
text->setFont(font);
text->setColor(fontSizeColor);
text->setCharacterSize(fontSizeCharacterSize);
text->setPosition(cursor);
// use text that uses 10 by 10 texels as a target resolution for fonts.
text->setFontResolution(10,10); // blocky but small texture memory usage
text->setText("text->setFontResolution(10,10); // blocky but small texture memory usage");
geode->addDrawable(text);
}
cursor.y() -= fontSizeCharacterSize;
{
osgText::Text* text = new osgText::Text;
text->setUseVertexBufferObjects(useVBOs);
text->setFont(font);
text->setColor(fontSizeColor);
text->setCharacterSize(fontSizeCharacterSize);
text->setPosition(cursor);
// use text that uses 20 by 20 texels as a target resolution for fonts.
text->setFontResolution(20,20); // smoother but higher texture memory usage (but still quite low).
text->setText("text->setFontReso
Re: [osg-users] osgText::Text::GlyphQuads change in the recent versions
Hi Nick, That is an interesting note for me as this is in the exact area of the Text.cpp that the segv that I am having with the issue I posted on the forum... The segv does not occur on 3.3.1, but does now on the trunk version I am running currently. I did not test on any releases prior to 337, but would be curious to see if it runs on 3.3.3, if it does it might help with finding the cause of the problem now So I will download and test against 3.3.3, just for my issue. ... Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=63592#63592 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] Segfault occuring in -- void Text::drawImplementation(osg::State& state, const osg::Vec4& colorMultiplier) const -- after updating from osg 3.3.1 to the current trunk
Hi, Just updated the attachment, had an extra include file that was not present listed in the source that was not needed. My apologies if this caused anyone any issues before I caught the problem. ... Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=63595#63595 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] Segfault occuring in -- void Text::drawImplementation(osg::State& state, const osg::Vec4& colorMultiplier) const -- after updating from osg 3.3.1 to the current trunk
Hi Robert, Thank you for looking in to this for us... I did a quick test against the 3.3.3 release and it appears to be working OK. So seems this has something to do with changes that occurred after that release. ... Best Regards!, Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=63598#63598 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] Segfault occuring in -- void Text::drawImplementation(osg::State& state, const osg::Vec4& colorMultiplier) const -- after updating from osg 3.3.1 to the current trunk
Hi Robert,
After getting the small example to fail today, I was looking some more and came
up with a few changes here that allows me to run.
I am relatively sure that this will not be the final fix, but I am hoping that
it will at least get you closer to what is causing the problem. Here is an svn
diff listing from my modified Text.cpp
Code:
Index: src/osgText/Text.cpp
===
--- src/osgText/Text.cpp(revision 14858)
+++ src/osgText/Text.cpp(working copy)
@@ -726,6 +726,12 @@
GlyphQuads::Coords2& coords2 = glyphquad._coords;
GlyphQuads::Coords3& transformedCoords =
glyphquad._transformedCoords[contextID];
+if(transformedCoords == NULL)
+{
+ glyphquad._transformedCoords[contextID] = new osg::Vec3Array;
+ transformedCoords = glyphquad._transformedCoords[contextID];
+}
+
unsigned int numCoords = coords2->size();
if (numCoords != transformedCoords->size())
{
@@ -804,6 +810,12 @@
for( ; backdrop_index < max_backdrop_index; backdrop_index++)
{
GlyphQuads::Coords3& transformedCoords =
glyphquad._transformedBackdropCoords[backdrop_index][contextID];
+if(transformedCoords == NULL)
+{
+ glyphquad._transformedCoords[contextID] = new osg::Vec3Array;
+ transformedCoords = glyphquad._transformedCoords[contextID];
+}
+
unsigned int numCoords = coords2->size();
if (numCoords!=transformedCoords->size())
{
@@ -1288,7 +1300,7 @@
if ( !_textureGlyphQuadMap.empty() )
{
const GlyphQuads& glyphquad = (_textureGlyphQuadMap.begin())->second;
-if ( :q)
+if ( glyphquad._transformedCoords[contextID] == NULL )
{
computePositions(contextID);
}
The diff at line 1303 in my source file:
-if ( glyphquad._transformedCoords[contextID]->empty() )
+if ( glyphquad._transformedCoords[contextID] == NULL )
The glyphquad._transformedCoords[contextID] call creates an
empty buffered object entry since the contextID coming in is not present, yet
and the buffered_object API automatically resizes the array if it does not
contain the value in the [] operator. So now I check for NULL instead of
empty() and thereby now calling computePositions(contextID).
The change in the
Re: [osg-users] Segfault occuring in -- void Text::drawImplementation(osg::State& state, const osg::Vec4& colorMultiplier) const -- after updating from osg 3.3.1 to the current trunk
Robert, One other thing to mention, I have another application that is pure OSG that is to say it is NOT using an QT widgets and it is doing basically the same thing as the QT app that I have and it runs OK as far as I can tell. The example code I sent you that is failing is also based on the osgviewerQT example. I am running QT 5.4.1 right now on my OpenSuSE 64bit linux system in case you need that level of information to isolate the issue. Another co-worker is running on MacOS and has told me that he is getting the same segv on his system so does not seem to be Linux specific. ... Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=63604#63604 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] OpenThreads, scheduling, etc ... policies
Hi, I would like to revisit this topic again as I see that even in the current trunk(3.3.8) as of this date May 04, 2015, the linux platform is still crippled inside of Openthreads/PThread API, from running priority/real-time scheduling policies. I will like to open a discussion about removing the code that is currently crippling the Linux platform as I do not see any reason at this date time to leave it this way. When properly used, as on any other platform, the currently crippled threading policies work properly on the Linux platform. ... Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=63625#63625 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] OpenThreads, scheduling, etc ... policies
Robert,
When I mentioned crippled, I did not mean that as a bad or derogatory
comment. Maybe saying something like intentionally disabled would have
been a better choice of words, my apologies if my wording offended anyone in
any way.
Basic threading is working fine under Linux. If the user however wants to
take advantage of all of the available threading policies and priorities now
available to them they cannot. I think that this is just possibly just due to
this section of Openthreads not being updated when these features were fully
implemented and properly working on the linux platform.
Here is a code excerpt of the SetThreadSchedulingParams
function call within OpenThreads/phhreads/PThread.cpp
Code:
//--
// Set thread scheduling parameters. Unfortunately on Linux, there's no
// good way to set this, as pthread_setschedparam is mostly a no-op.
//
static int SetThreadSchedulingParams(Thread *thread)
{
int status = 0;
#ifdef ALLOW_PRIORITY_SCHEDULING // [
if(sysconf(_POSIX_THREAD_PRIORITY_SCHEDULING))
{
int th_policy;
int max_priority, nominal_priority, min_priority;
sched_param th_param;
pthread_getschedparam(thread->getProcessId(),
&th_policy, &th_param);
#ifndef __linux__
switch(thread->getSchedulePolicy())
{
case Thread::THREAD_SCHEDULE_FIFO:
th_policy = SCHED_FIFO;
break;
case Thread::THREAD_SCHEDULE_ROUND_ROBIN:
th_policy = SCHED_RR;
break;
case Thread::THREAD_SCHEDULE_TIME_SHARE:
th_policy = SCHED_OTHER;
break;
default:
#ifdef __sgi
th_policy = SCHED_RR;
#else
th_policy = SCHED_FIFO;
#endif
break;
};
#else
th_policy = SCHED_OTHER; // Must protect linux from realtime.
#endif
#ifdef __linux__
max_priority = 0;
min_priority = 20;
nominal_priority = (max_priority + min_priority)/2;
#else
max_priority = sched_get_priority_max(th_policy);
min_priority = sched_get_priority_min(th_policy);
nominal_priority = (max_priority + min_priority)/2;
#endif
switch(thread->getSchedulePriority())
{
case Thread::THREAD_PRIORITY_MAX:
th_param.sched_priority = max_priority;
break;
case Thread::THREAD_PRIORITY_HIGH:
th_param.sched_priority = (max_priority +
nominal_priority)/2;
break;
case Thread::THREAD_PRIORITY_NOMINAL:
th_param.sched_priority = nominal_priority;
break;
case Thread::THREAD_PRIORITY_LOW:
th_param.sched_priority = (min_priority +
nominal_priority)/2;
break;
case Thread::THREAD_PRIORITY_MIN:
th_param.sched_priority = min_priority;
break;
default:
th_param.sched_priority = max_priority;
break;
}
status = pthread_setschedparam(thread->getProcessId(),
th_policy,
&th_param);
if(getenv("OUTPUT_THREADLIB_SCHEDULING_INFO") != 0)
PrintThreadSchedulingInfo(thread);
}
#endif // ] ALLOW_PRIORITY_SCHEDULING
return status;
};
You can see in this excerpt that on Linux platforms the code to allow you to
select any of the available threading policies is allowed on all platforms but
linux. Linux is hard-coded to set the SCHED_OTHER policy.
The other sections of the code excerpt limit the range of priorities allowed to
between 0 and 20, when linux currently has a range of 0 - 99, 99 being the
highest priority allowed. The associated calls to sched_get_priority_max and
sched_get_priority_min are not allowed on the linux platform.
I will be happy to provide you with a sample application if you still would
like to have one, but I think the ifndef's in the code will let you see what I
am referring too.
Again I apologize if I offended anyone, that was not my intention.
...
Thank you!
Cheers,
Curtis
Code:
--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=63627#63627
___
osg-users mailing list
osg-users@lists.openscenegraph.org
http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] OpenThreads, scheduling, etc ... policies
Robert, There is already a define around all the priority scheduling calls to protect people from accidentally using them: #ifdef ALLOW_PRIORITY_SCHEDULING Which seems to have to be defined in the calling program as I cannot find this any place inside of OSG other than the pthreads source. Maybe this would suffice for now to protect these calls. I am going to suspect that few Linux users are defining this since the only reason to do so would be to use the calls that are currently ifdef'd out looking at the code inside of the ALLOW_PRIORITY_SCHEDULING define. I agree with you about Linux not originally supporting the full suite of the pthread extensions. Linux has come a long way in recent years. We have been using these real-time pthread scheduling parameters though for at least 5 or 6 years now on our other Linux applications that do not need OpenSceneGraph support. So the support on Linux has been there for awhile. Is it possible to contact the original author on this subject, if so who would I contact? Hopefully we will get some more comments on this topic for the OSG community... ... Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=63629#63629 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] OpenThreads, scheduling, etc ... policies
Robert,
Attached is a copy of the PThread.cpp file from OSG 3.3.7 that I modified that
allows Linux systems to take full advantage of the pthread scheduling
parameters. The changes I made are all within the
#ifdef ALLOW_PRIORITY_SCHEDULING
define and the static int SetThreadSchedulingParams(Thread *thread)
function.
Feel free to pass this around to anyone that would like to review. comment
and/or test the changes when time permits. I am not
trying to rush anything, just figured I would go ahead and send
you what I am referring too, so it will be handy when time permits
you/(or your group) to take a look at it.
I am currently running with this change here and all our old existing OSG
applications still are running OK as well as are my new test applications that
take advantage of the changes.
One thing to note is that on at least my Linux system here, unless the user is
running with root privileges or their application as been changed to run with
root privileges the Linux kernel does not allow the application to run at
anything other than the default policy of SCHEDULE_OTHER and priority 0(zero),
no matter what they attempt to set. This is another protection that seems to
be in place to protect someone from accidentally causing a system problem.
Anything else that I can do to assist in getting this change implemented at
some point, please let me know.
Thank you for your time and consideration in this matter.
Cheers,
Curtis
--
Read this topic online here:
http://forum.openscenegraph.org/viewtopic.php?p=63650#63650
/* -*-c++-*- OpenThreads library, Copyright (C) 2002 - 2007 The Open Thread Group
*
* This library is open source and may be redistributed and/or modified under
* the terms of the OpenSceneGraph Public License (OSGPL) version 0.0 or
* (at your option) any later version. The full license is in LICENSE file
* included with this distribution, and on the openscenegraph.org website.
*
* This library is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* OpenSceneGraph Public License for more details.
*/
//
// PThread.c++ - C++ Thread class built on top of posix threads.
// ~~~
#include
#include
#include
#include
#include
#include
#if defined __linux__ || defined __sun || defined __APPLE__ || ANDROID
#include
#include
#include
#if !defined ANDROID
#include
#endif
#endif
#if defined(__sgi)
#include
#endif
#if defined(__hpux)
#include
#endif
#if defined(HAVE_THREE_PARAM_SCHED_SETAFFINITY) || defined(HAVE_TWO_PARAM_SCHED_SETAFFINITY)
#include
#endif
#if defined (__FreeBSD__) || defined (__APPLE__) || defined (__MACH__)
#include
#include
#endif
#if defined(__ANDROID__)
#ifndef PAGE_SIZE
#define PAGE_SIZE 0x400
#endif
#endif
#include
#include "PThreadPrivateData.h"
#include
using namespace OpenThreads;
#ifdef DEBUG
# define DPRINTF(arg) printf arg
#else
# define DPRINTF(arg)
#endif
#define ALLOW_PRIORITY_SCHEDULING 1
//-
// Initialize the static unique ids.
//
int PThreadPrivateData::nextId = 0;
//-
// Initialize thread master priority level
//
Thread::ThreadPriority Thread::s_masterThreadPriority =
Thread::THREAD_PRIORITY_DEFAULT;
bool Thread::s_isInitialized = false;
pthread_key_t PThreadPrivateData::s_tls_key;
struct ThreadCleanupStruct
{
OpenThreads::Thread *thread;
Atomic *runflag;
};
//-
// This cleanup handler is necessary to ensure that the thread will cleanup
// and set its isRunning flag properly.
//
void thread_cleanup_handler(void *arg)
{
ThreadCleanupStruct *tcs = static_cast(arg);
tcs->thread->cancelCleanup();
(*(tcs->runflag)).exchange(0);
}
//-
// Class to support some static methods necessary for pthread's to work
// correctly.
//
namespace OpenThreads
{
class ThreadPrivateActions
{
//-
// We're friendly to Thread, so it can issue the methods.
//
friend class Thread;
private:
//-
// pthreads standard start routine.
//
static void *StartThread(void *data)
{
Thread *thread = static_cast(data);
PThreadPrivateData *pd =
static_cast(thread->_prvData);
if (pd->cpunum>=0)
{
#if defined(__sgi)
pthread_setrunon_np( pd->cpunum );
#elif defined(HAVE_PTHREAD_SETAFFINITY_NP) || defined(HAVE_THREE_PARAM_SCHED_SETAFFINITY) || defined(HAVE_TWO_PARAM_SC
Re: [osg-users] Segfault occuring in -- void Text::drawImplementation(osg::State& state, const osg::Vec4& colorMultiplier) const -- after updating from osg 3.3.1 to the current trunk
Hi, Any chance the fix for this was included in the 3.3.8 release and I just did not see any notice of it? ... Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=64098#64098 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] Segfault occuring in -- void Text::drawImplementation(osg::State& state, const osg::Vec4& colorMultiplier) const -- after updating from osg 3.3.1 to the current trunk
Hi Robert, I compiled and tested the code to make sure it failed the same way before posting it, so that is sort of strange to find this type of thing now.. I was testing on 64bit Linux here.. Well I am glad it did not waste to much of your time, I know you have been really busy lately from watching all the posts and stuff going on the last month or so... We have been busy ourselves here getting our OpenIG project ready for release to the opensource community. Hopefully that will be ready by the end of this week. Getting all 3 platforms, Linux, MacOSX and Windows all happy at the same time has been quite the chore, but I am sure you are fully aware of that already... ... Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=64101#64101 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
[osg-users] HowTo Add PostDrawCallback after making call to setUpDepthPartition?
Hi, I am relatively new to OSG so please take that into account when reading my question below We have a large visual database, extents are basically the entire country of Italy. Due to some z-buffer issues with the terrain we have been reading and studying lots of z-buffer data, etc..and have determined that we would like to try using the osgViewer DepthPartition for some testing. We can get the depthpartition created successfully and it seems to work fairly well. The problem is that we have a PostDrawCallback that we are trying to have do some stuff for us and the callback is never being called. I feel that is due to us most likely passing in the Master camera instead of the proper slave camera that was created in the DepthPartition call. My question is how do I get/find the correct camera to use in our call for: osg::Camera* camera = NULL; double zNear=1.0, zMid=50.0, zFar=10.0; osg::ref_ptr dps = new osgViewer::DepthPartitionSettings; dps->_mode = osgViewer::DepthPartitionSettings::FIXED_RANGE; dps->_zNear = zNear; ps->_zMid = zMid; dps->_zFar = zFar; viewer = new osgViewer::Viewer; viewer->setUpDepthPartition(dps.get()); camera = viewer->getCamera(); camera->setPostDrawCallback(Callback); does not seem to work as expected in this case. We get a valid camera, just not apparently the correct camera as the callBack is not called. Removal of the setUpDepthPartition call and our callBack is called as expected. Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=51862#51862 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] Happy New Year.
Hi, Just wanted to say Happy New year back to you Chris... Hope we all have a great one this year... Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=51917#51917 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] Segfault occuring in -- void Text::drawImplementation(osg::State& state, const osg::Vec4& colorMultiplier) const -- after updating from osg 3.3.1 to the current trunk
Hi Robert, I noticed some references to this reported issue in the latest Changelog for the trunk checking for the proper setup of the number of contexts in the osgText area. Do you think this entire issue is now resolved or is the possible issue with QtOSG and it not reporting the number of proper contexts possibly still present? ... Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=64637#64637 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] Segfault occuring in -- void Text::drawImplementation(osg::State& state, const osg::Vec4& colorMultiplier) const -- after updating from osg 3.3.1 to the current trunk
Robert, Just tested the trunk from a svn co from about 1 hour ago, with my test osgviewerQT.cpp source. Still getting a segv when running it. I was running in release mode for this test so cannot verify if its exactly the same fault or notbut it would seem that there is still some sort of issue with this. I will rebuild in debug mode as soon as I have some more free time and check to see if its the same bug or something different as I saw before. ... Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=64642#64642 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] Segfault occuring in -- void Text::drawImplementation(osg::State& state, const osg::Vec4& colorMultiplier) const -- after updating from osg 3.3.1 to the current trunk
Hi Robert, OK I will update my copy of the OSG trunk and let you know as soon as I have a minute. Probably something I will have to do in my off time at home as work time is pretty much taken up right now. I would not consider myself a QT expert either, but will see if I can find anything that stands out. The failing example is pretty basic stuff so hopefully something will show up when I recompile everything in debug mode so I can step into this a little at a time and see what is happening in there. Thank you very much, Curtis ... Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=64651#64651 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] Segfault occuring in -- void Text::drawImplementation(osg::State& state, const osg::Vec4& colorMultiplier) const -- after updating from osg 3.3.1 to the current trunk
Hi Robert, I updated to the latest in the trunk this morning, version reporting it as 3.5.0 So far everything appers to be working flawlessly. The example I gave you for the segv is working for me as well as all of our OSG applications seem to be working perfectly I think I even see a few percentange points of a performance increase in our applications. I will leave this version in for a few days now and will let you know if I find anythng else to report. For now though looks like the issue I reported seems to have been resolved. Is the current trunk and the current stable rc of 3.4rc11 the same right now? ... Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=64707#64707 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
[osg-users] [osgPlugins] DDS textures not loading on Mac OSX and the latest trunk(3.5.0)
Hi, I have looked a bit and did not find this specific issue listed, I think it might have been reported already so please accept my apologies if this is already in the pipeline to get taken care of On Mac OSX Yosimite 10.10.3, osg/osgviewer does not seem to be able to load DDS textures when referenced in our openflight files. This I believe was first noticed by us back on release 3.3.7 and as of updating from SVN today from the trunk(3.5.0), still appears to be not working as our terrain tiles are all showing with no textures. It would be really great if we can get this fix into the next upcoming release... At one point we had commented out line 1016 of the ReaderWriterDDS.cpp if (mipmap_offsets.size()>0) osgImage->setMipmapLevels(mipmap_offsets); and was able to get the textures to show up. So maybe this will help someone figure out the actual area that is affected I would be happy to look but this is way out of my area of knowledge. I will be happy to help in any way I can though if anyone has any suggestions they would like to try, I can put them in and retest... ... Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=64725#64725 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] [osgPlugins] DDS textures not loading on Mac OSX and the latest trunk(3.5.0)
Hi Robert, The files ended up being a bit to big to upload I guess as it kept failing. Here is a a link to my dropbox... It contains a zip of the flt file, all the dds textures and a .osgb file converted on the mac. https://www.dropbox.com/s/t5r8x1s7ee6m107/DDS_TEST.zip?dl=0 This tile will load the textures when I comment out the line I mentioned. Right now that line is back in and no textures are loading. The issue I see is that while on the MAC I cannot see the textures in osgviewer, when I unzip these files on my linux system and load the osgb file, the textures load properly. So seems its something to do only with the actual loading of the textures and not building of the .osgb file itself. We can duplicate this on all our macs, so it does not appear to be just a one bad mac issue either ... Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=64728#64728 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] [osgPlugins] DDS textures not loading on Mac OSX and the latest trunk(3.5.0)
Hi Robert, OK I will do some more digging into this and see if I can find anything myself that could help. Hopefully there are some other mac folks out there that have a better background in this area than I do, that would be nice... I can def say that OSG 3.3.1 is working OK on my mac, as I just compiled and reloaded it to make sure. So looks like maybe its something that has changed since that dev release. I will try 3.3.3 next to see if we can at least isolate when the issue first started as it might help give some ideas where to look as well. ... Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=64739#64739 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] [osgPlugins] DDS textures not loading on Mac OSX and the latest trunk(3.5.0)
Hi Robert, I know the openGL implementation on the Mac is different, we run into shader issues all the time when we setup our applications to run on WIndows, Linux and Mac. It never fails the Mac will have some sort of issue we need to account for... A real fix is going to take someone very good at openGL on the mac to find it then.. Now to find someone to help us on this...I will consult with Frank over at SunDog Software as I know with SilverLining and Triton he has someone that he uses for Mac OSX OpenGL/etc., support that might be able to help us on this too. If I get anything I will update this thread for everyone... ... Thank you for all your input on this, Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=64743#64743 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] [osgPlugins] DDS textures not loading on Mac OSX and the latest trunk(3.5.0)
Hi Robert, I am searching the OpenGL Mac forums right now to see if I can find any references to mipmap or compressed textures but so far no results. I will look a bit more before entering a new post. When I was replying earlier I missed the other updates on this thread that occurred, so I now see you did some pretty thorough testing to rule out OSG on this...thanks for spending the time, I know your busy right now with everything else going on.. I will pass on your recommendations to our other developers here as well on your suggestion on the Nodevisitor switching off mipmaps in the interim. ... Thank you very much! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=64746#64746 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] [osgPlugins] DDS textures not loading on Mac OSX and the latest trunk(3.5.0)
Hi Robert, After some checking around with our development team, we discovered that after updating to the latest OSX release the newest GL package from Apple was also installed as part of this update and this issue seems to have been resolved... The numbers from my particular mac: Version: 4.1 NVIDIA-10.4.2 310.41.35f01 Shading language version: 4.10 Then removing and reconverting all our files into .osgb again things are loading properly now. So my apologies for not thinking about updating this before getting everyone here involved. Normally all our macs are kept up to date, so was not even thinking about this as a possibility. Anyone else having similar issues should try to update their systems to the latest release of OSX to ensure they have the latest Apple versions of OpenGL installed. ... Thank you! Cheers, Curtis :-* -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=64810#64810 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] [osgPlugins] DDS textures not loading on Mac OSX and the latest trunk(3.5.0)
Hi Robert, OK thank you for taking a quick look With things the way they are now, its very strange. I have a few sets of flight files that work and also a set that does not. The dropbox link I put in my last post for the small flight test case, the dds textures do not load on the mac. So it seems that this is not failing with all cases of pre-generated mipmaps. Maybe there is something in how this was done in older versus newer versions of openflight or something. What is the difference between using the pre-gen'd mipmaps and not using them? Is there some sort of performance boost if the pre-gen'd ones are working properly, something has to be calculated on the fly as they are loading or something? Just curious as I know very little about this subject... ... Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=64828#64828 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] [osgPlugins] DDS textures not loading on Mac OSX and the latest trunk(3.5.0)
Hi Robert, We will be continuing our investigation to see if we can get a better handle on what is the culprit behind these differences... The suggestion for the NodeVistor is also something we will be looking at for the Mac systems until such a time as we can find out the exact cause of the issue or it gets fixed in a later version of the Mac OS somewhere along the line. Thank you for all the information and suggestions... Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=64836#64836 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] [osgPlugins] DDS textures not loading on Mac OSX and the latest trunk(3.5.0)
Hi all, The current version of OpenGL for Yosimite 10.10.5, seems to be working to some extent from the testing I have done here recently, depending on the data you are loading. I have a new set of flt tiles here that loading their associated dds textures OK, I can convert the flt.files to .osgb, etc. and they still load properly. However, I also have another set of flt files that will not load their associated dds textures, as well as an old .ive file that does not load the textures on Mac OSX but loads fine on Linux right now. We are not quite sure what the differences are. Could be the non-working flt files were built against an older flt spec or something as this is a relatively old set of flt files. The non-working flt files have this for the first few bytes: 00 01 01 14 54 65 72 72 65 78 00 00 00 00 06 04 |Terrex..| 0010 00 00 00 01 68 65 72 65 61 66 74 65 72 00 00 00 |hereafter...| The working flt files are different: 00 01 01 48 64 62 00 00 00 00 00 00 00 00 06 68 |...Hdb.h| 0010 00 00 00 01 31 39 2e 30 38 2e 32 30 31 35 2d 31 |19.08.2015-1| I do not know anything about the layout of a flt file, so cannot say even if this means anything in this area, or is even enough to determine anything... At any rate things are improving on OSX it seems, just backwards compatibility seems to be an issue to me. Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=64892#64892 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
[osg-users] OpenThread, each with its own viewer and performance
Hello OSG forum users, I have a new application I am working on where I am using an Openthreads, thread to setup a viewer, graphics context etc. for 3 separate screens on my computer. At the moment we are loading no data into the viewer, so when its realized its just the default blue OSG screen and the one default camera for each viewer/screen. First thread starts up, opens the screen and its osg stats shows 1200fps. Second thread starts up and both of the osg stats are showing about 6000fps, the third thread starts up and all 3 have an FPS about 400fps. So the stats are showing a drop by each screen, even though each thread has its own independent, viewer and graphics context. I have tried all the OSG threading models, including just leaving it at AUTO and did not notice any change in the FPS symptoms. Anyone have any clues as to why this is occurring?? It just seems to me that all these screens should be independent. When I do the same thing with 3 separate applications instead of threads, each screen has its own FPS of approx 1200fps... ... Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=64964#64964 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] OpenThread, each with its own viewer and performance
Hello OSG users, I am posting a simple application that uses OpenThreads. Where each thread creates a Graphics context, sets up one composite viewer with one view and turns on the osg stats. No nodes or data of any kind is added to the view. There is a define in the main() osgTest.cpp that sets up how many threads are created. Very simple example to show the problem. I am hoping that with your help we can determine if this is something we are doing incorrectly or if it is just something that is this way for a specific reason that we just do not understand. Any help, suggestions or comments would be greatly appreciated. ... Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=65185#65185 osgTest.7z Description: application/7z-compressed ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] embedded UI in OSG
Hi Nick, Also in the examples directory you can see some for WxWidgets, Qt, GTK, FLTK, Cocoa(osx), MFC, FOX, GLUT... just to name a few that I see in there. So you have a few choices nowadays... We have been successful at using Qt for all our needs, as you are aware I think.. ... Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=65550#65550 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
[osg-users] IVE, OSGB, XXX -- which one is recommended for generally best performance?
Hi, I know there might be some reasons that you would pick one format or another but in general terms, what is the recommended OSG format to use these days for best OSG overall performance? IVE, OSGB, etc.. Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=65708#65708 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] Vulkan SDKs released
Hello everyone, Speaking with the folks here at Compro, everyone seems interested in working with Vulcan's new feature set and contributing, at what level we cannot really say at this early stage in the game. But please keep us in the loop as things start to move forward so that we can jump in when/where we think that we could be of use to the project. ... Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=66312#66312 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
[osg-users] osgsimplegl3 example, is it possible to run on some combination of OSG and OSX?
Hi, I am hoping there may be some OSG & MacOSX guru's out there that could tell if me there is a combination of Mac OSX release and OpenSceneGraph release that allows the provided osgsimplegl3 example run as it is without modification. The mac is running El Capitan and we currently have OSG 3.4.0 compiled and installed on it as we had read someplace that 3.4.0 was already patched to allow for later GL support of some kind. Unfortunately, it was not able to successfully run the osgsimplegl3 example either without getting a couple of #version 140 glcompiler messages output after enabling debug output messages. We used macports to satisfy most of the required OSG dependencies. We are not GL experts, nor OSX experts here as you probably already deduced...hopefully the answer is not one that will take very much time to answer. If you need more information just let me know and I will do my best to get the information for you. ... Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=67589#67589 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] osgsimplegl3 example, is it possible to run on some combination of OSG and OSX?
Hi Robert, Thank you for your quick response, I will keep posted. ... Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=67591#67591 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] osgsimplegl3 example, is it possible to run on some combination of OSG and OSX?
Hi Robert, Sorry to bother you gain, but I have been watching and I cannot say for sure if there was any progress on this for OSX or not. Are things ready for OSX and being able to at least run the osgsimplegl3 example yet? Watching all the commits and messages I got a bit confused as to exactly how the repository is right now as far as OSX is concerned. ... Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=68220#68220 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] osgsimplegl3 example, is it possible to run on some combination of OSG and OSX?
Hi Robert, I will continue to monitor the forum then and when I see some updates to the Vertex Array Object I will update my Mac system and retest things using the OSG GL examples again to see if things are working any better. Hopefully one of the other OSX guru's will be able to take a look as well, I am not much more that a general user on OSX, just enough to test and verify whether our latest OpenIG project works on it or not. Our Forward Plus lighting plugin requires some of the GL3 API to be available. I am hopeful that if the osgsimplegl3 example runs successfully, then OpenIG's Forward Plus lighting system should be OK too on the Mac. Right now its running pretty well on Linux and Windows systems already. Thank you, Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=68232#68232 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
[osg-users] osgversion command fails with symbol lookup error on new clone of OpenSceneGraph 3.5.5 on Linux
Hi, After cloning out a copy of OpenSceneGraph 3.5.5 from GitHub, and running cmake with default options, then enabled building of examples and documentation on Linux only. After clean build with no errors, running make install;sudo ldconfig osgversion command fails with the following results: osgversion osgversion: symbol lookup error: /usr/local/lib64/libosgDB.so.145: undefined symbol: _ZN11OpenThreads6Thread20setProcessorAffinityERKNS_8AffinityE Actually seems all the default built binaries fail with the same error. Are there some other cmake flags, defines or compile options that need to be specified before compiling 3.5.5 and have it run properly on 64bit Linux? If more data is needed, just let me know what/how to get it for you and I will do my best to get it. Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=68963#68963 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] osgversion command fails with symbol lookup error on new clone of OpenSceneGraph 3.5.5 on Linux
Hi Robert, On my OpenSuSE 13.2 64bit system, it seems that at some point OSG had been installed in /usr/local/lib and /usr/local/lib64, which had conflicting versions installed and was causing this problem. For some reason OpenSuSE still keeps both of these paths alive. They are supposed to be for 32bit versus 64bit library installs, but it seems depending on your cmake options, it can go to either place during the "make install".. My apologies to you and the forumI have cleaned up the duplicated libraries and everything is working as expected now. Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=68965#68965 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] Feedback/guidance sought on move of osgQt out into it's own project/repository
Hi, Just curious how things are proceeding with this process. I saw Roberts request to test the latest 3.5.6 Master branch with the new vertex buffer changes and would like to try and do that but our entire baseline relies heavily on osgQt being present to build against. We are mainly a Linux house, but also have some Windows and MacOSX requirements from time to time. My first attempt to clone out the latest osgQt kit from GitHub does not allow me to run cmake on it successfully as a standalone project/directory on my system. I looked around a bit and I do not see how to actually get cmake to properly setup the project to build with OSG now that its been separated. Can someone point me in the direction to find a set of instructions to follow or possibly give me some hints on how to get osgQt built against the current 3.5.6 master branch of OSG. I have a little bit of time right now so, any info would be greatly appreciated so that we can start to provide back some feedback on the latest master branch of OSG as well as this new osgQt project. ... Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=68976#68976 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] Feedback/guidance sought on move of osgQt out into it's own project/repository
Hi Mathieu, Basically that is exactly what I did and am getting an error. CMake Error: File /usr/local/3rdparty/osgQt-git/packaging/ld.so.conf.d/openscenegraph.conf.in does not exist. CMake Error at CMakeLists.txt:971 (CONFIGURE_FILE): configure_file Problem configuring file The ld.so.conf.d directory does not exist within the packaging directory on my system, its actually at: /etc/ld.so.conf.d Currently ONLY the master branch 3.5.6 of osg is installed on my system in its default location of /usr/local/lib64. Here is my output of the entire process: amx@ios:/usr/local/3rdparty> cd osgQt-git/ amx@ios:/usr/local/3rdparty/osgQt-git> mkdir BUILD amx@ios:/usr/local/3rdparty/osgQt-git> cd BUILD/ amx@ios:/usr/local/3rdparty/osgQt-git/BUILD> cmake .. -- The C compiler identification is GNU 4.8.3 -- The CXX compiler identification is GNU 4.8.3 -- Check for working C compiler: /usr/bin/cc -- Check for working C compiler: /usr/bin/cc -- works -- Detecting C compiler ABI info -- Detecting C compiler ABI info - done -- Check for working CXX compiler: /usr/bin/c++ -- Check for working CXX compiler: /usr/bin/c++ -- works -- Detecting CXX compiler ABI info -- Detecting CXX compiler ABI info - done -- Found osgDB: /usr/local/lib64/libosgDB.so -- Found osgGA: /usr/local/lib64/libosgGA.so -- Found osgUtil: /usr/local/lib64/libosgUtil.so -- Found osgText: /usr/local/lib64/libosgText.so -- Found osgViewer: /usr/local/lib64/libosgViewer.so -- Found osgWidget: /usr/local/lib64/libosgWidget.so -- Found osg: /usr/local/lib64/libosg.so -- Found OpenThreads: /usr/local/lib64/libOpenThreads.so -- Found OpenSceneGraph: /usr/local/lib64/libosgDB.so;/usr/local/lib64/libosgGA.so;/usr/local/lib64/libosgUtil.so;/usr/local/lib64/libosgText.so;/usr/local/lib64/libosgViewer.so;/usr/local/lib64/libosgWidget.so;/usr/local/lib64/libosg.so;/usr/local/lib64/libOpenThreads.so (found suitable version "3.5.6", minimum required is "3.5.5") -- Looking for include file pthread.h -- Looking for include file pthread.h - found -- Looking for pthread_create -- Looking for pthread_create - not found -- Looking for pthread_create in pthreads -- Looking for pthread_create in pthreads - not found -- Looking for pthread_create in pthread -- Looking for pthread_create in pthread - found -- Found Threads: TRUE -- Looking for XOpenDisplay in /usr/lib64/libX11.so;/usr/lib64/libXext.so -- Looking for XOpenDisplay in /usr/lib64/libX11.so;/usr/lib64/libXext.so - found -- Looking for gethostbyname -- Looking for gethostbyname - found -- Looking for connect -- Looking for connect - found -- Looking for remove -- Looking for remove - found -- Looking for shmat -- Looking for shmat - found -- Looking for IceConnectionNumber in ICE -- Looking for IceConnectionNumber in ICE - found -- Found X11: /usr/lib64/libX11.so -- Found OpenGL: /usr/lib64/libGL.so The build system is configured to install libraries to /usr/local/lib64 Your applications may not be able to find your installed libraries unless you: set your LD_LIBRARY_PATH (user specific) or update your ld.so configuration (system wide) You have an ld.so.conf.d directory on your system, so if you wish to ensure that applications find the installed osg libraries, system wide, you could install an OpenSceneGraph specific ld.so configuration with: sudo make install_ld_conf CMake Error: File /usr/local/3rdparty/osgQt-git/packaging/ld.so.conf.d/openscenegraph.conf.in does not exist. CMake Error at CMakeLists.txt:971 (CONFIGURE_FILE): configure_file Problem configuring file -- Configuring incomplete, errors occurred! See also "/usr/local/3rdparty/osgQt-git/BUILD/CMakeFiles/CMakeOutput.log". See also "/usr/local/3rdparty/osgQt-git/BUILD/CMakeFiles/CMakeError.log". I was able to get past this error by running cmake-gui and manually setting the build path, then clicked configure again and then finally was able to generate the makefile. After that the osgQt project built and installed ok on my system. I currently have cmake 3.0.2 installed on my 64bit OpenSuSE 13.2 Linux system that I am using for this initial test in case that affects any of this. Thank you for your quick response.. Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=68981#68981 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] Feedback/guidance sought on move of osgQt out into it's own project/repository
Hi Mathieu, I did a pull on osg and osgQt just to make sure I had the latest of both. cmake on osgQt completed without error now with both the default settings and with the examples selected to build. Just an FYI, when I selected the documentation build in the cmake-gui... I got a cmake error...anyhow that is just so you know about it, its not anything that we need, just figured I would pass it along to you for your TODO list... CMake Error: File /usr/local/3rdparty/osgQt-git/doc/Doxyfiles/openthreads.doxyfile.cmake does not exist. CMake Error at CMakeLists.txt:888 (CONFIGURE_FILE): configure_file Problem configuring file ... Thank you for the quick fix(s)... Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=68984#68984 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] Feedback/guidance sought on move of osgQt out into it's own project/repository
Hi Mathieu, One other thing that is also not any sort of priority or any sort of bug report for you to concern yourself with, but maybe you know something about it with your work on this project. The attached PIC to this reply is what I get when starting the osgviewerQt example. You will notice the left larger window only contains one model instead of the 4 models that it should have. Just was running the example after building it and it reminded me of seeing this again. The 4 models show up properly as soon as I touch the window and resize it. Minimizing it or covering it up and then uncovering does not fix it, it needs a resize event to finally show the correct output, doesn't matter if its sized larger or smaller. This linux system has QT 5.4.2 installed on it right now, I want to say this behavior started with the first version of Qt5 we started to use after our linux distro finally updated from 4.8.x. Anyhow, just curious if you have seen this at all with your Qt work and if this is actually a Qt bug that we have to live with until they fix it or its something in the osgQt implementation. We get the same as I recall when we run this on our development Mac as well, not sure about windows we don't have the examples built on that platform right now for me to test it at the moment. ... Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=68985#68985 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] vertex_array_object branch merged with master
Hi Robert, I did some very basic testing of this for you yesterday. There is no need to respond back to me with any possible fixes or anything, I am just passing this along to you for your information right now. We have no plans on updating our OSG version in the near future. I was able to successfully compile the release on OpenSuSE 13.2 64bit Linux. Next I recompiled our Opensource OpenIG application against this release and everything compiled OK, with exception of one minor change to the source code for a missing function: osg::State* state = renderInfo.getState(); #if OSG_VERSION_LESS_OR_EQUAL(3,5,3) state->dirtyTexCoordPointer(1); #endif I used 3.5.3 above only because that is what I know for sure as I do not have a copy of 3.5.4 to check against. After this change our OpenIG application runs with no segv's, however there are a number of visual issues noted with the 3.5.6 master: 1. The one FBX model in the scene is not positioned properly anymore. Although the shadow for it seems to still be in the correct location and even shows the animation of that model, even though visually the model itself is not in that location anymore. 2.. There appears to be something going on with our terrain shaders too as when I change the TOD to 10:00, the terrain all turns bright white from its normal textures. In fact just moving around in the scene also results in the terrain textures disappearing totally. 3. Seems that our F+ lighting implementation will also need some adjustments as it appears not to be rendering properly anymore. I know very little about how the shaders were implemented in this application as most of this was done by our consultant, just passing on some of the basic results as I seem them right now for you. All of this is working properly in the 3.5.3 version we have been using so at some point between this and whats on the master branch now these visual changes have shown up. Anyhow I know its just some very basic info, but I figured I would pass it along for you in case this is not what you might have expected with this latest release. We will be curious to see what if anything anyone else reports back to you. ... Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=68993#68993 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] vertex_array_object branch merged with master
Hi Robert, That call was only used once in our source, which in this case is all opensource as its in our OpenIG opensource project available on GitHub The section of code this deals with is part of the integration of the MyGUI project with our OpenIG. At this point I would not consider it an issue as I suspect it might have just been put there as a way to clean up all things possible and not for a specific reason or just because some old example used it and was copied over this way. If I find out differently I will let you know but for now we can ignore this as far as I am concerned. ... Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=69006#69006 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] I/ITSEC 2016
Hi Chris, I'll be over there at least once while your there, not sure when yet, will know more when it gets closer, but I wouldn't mind catching up with ya and talking about some OpenIG, etc. things with ya ... Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=69170#69170 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] a linux dialog starting to blur while draging towards to an OSG QT widgets
Hi, I know the new KDE plasma desktops have effects that blur windows while dragging as well as make them semi-transparent, so I just wanted to pass on that information just in case it may be related to your problem and it has not been thought of yet. If this is not possible in your configuration, then please ignore my reply... Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=69535#69535 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] [3rd party] autodesk fbx sdk
Hi Nick, I also tried the Autodesk website and the links they provide are broken and taking you to a "page not found" default now. I did send their support people and email letting them know that it was broken. Hopefully they will fix it soon. ... Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=72213#72213 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] [3rd party] autodesk fbx sdk
NIck, I probably have a 2017 SDK around someplace for Linux and/or Windows. If that will do let me know and I will look around on Monday when I get in to work... ... Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=72219#72219 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] [3rd party] autodesk fbx sdk
Hi Nick, I got a reply from AutoDesk seems they had it down for maintenance and its back up now at this link: http://usa.autodesk.com/adsk/servlet/pc/item?siteID=123112&id=10775847 Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=72229#72229 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] [3rd party] autodesk fbx sdk
Hi, Looks like the 2018.1.1 SDK's are out now for Linux, Windows and Mac, as well as something for IOS. at least at this US link: http://usa.autodesk.com/adsk/servlet/pc/item?siteID=123112&id=26416130 ... Thank you! Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=72234#72234 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
[osg-users] OsgText compile issue with the newer versions of OSG
Hello community, I was attempting to test the latest release of OSG 3.5.8 and am now getting some compiler errors. error: ‘const struct osgText::Text::GlyphQuads’ has no member named ‘getCoords’ coords.insert(coords.end(),gq.getCoords()->begin(),gq.getCoords()->end()); Snippet of the code in question: Code: std::vector coords; osgText::Text::TextureGlyphQuadMap& tgqm = const_cast(_text->getTextureGlyphQuadMap()); const osgText::Text::GlyphQuads& gq = tgqmi->second; coords.insert(coords.end(),gq.getCoords()->begin(),gq.getCoords()->end()); We are using this code to control onscreen text command line input. What would be the proper way to fix this code under the newer versions of OSG to accomplish the same, if it is still possible in some way? Any hints would be appreciated. Thank you all as always... Cheers, Curtis[/code] -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=72393#72393 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
Re: [osg-users] The naming of VulkanSceneGraph
Hi, https://www.khronos.org/vulkan/adopters/ I think this page will alleviate anymore discussion, at least it sounds pretty straight forward to me that as long as VulcanSceneGraph passes their tests it seems OK to use the name. Of course I am no a lawyer either but this text seems to be pretty simple and in laymen's terms. Cheers, Curtis -- Read this topic online here: http://forum.openscenegraph.org/viewtopic.php?p=74159#74159 ___ osg-users mailing list osg-users@lists.openscenegraph.org http://lists.openscenegraph.org/listinfo.cgi/osg-users-openscenegraph.org
