#jogamp @ irc.freenode.net - 20131209 05:05:56 (UTC)
20131209 05:05:56 -jogamp- Previous @ http://jogamp.org/log/irc/jogamp_20131208050556.html
20131209 05:05:56 -jogamp- This channel is logged @ http://jogamp.org/log/irc/jogamp_20131209050556.html
20131209 05:37:23 * [Mike] (~Mike]@anon) Quit (Ping timeout: 250 seconds)
20131209 05:46:57 * [Mike] (~Mike]@anon) has joined #jogamp
20131209 06:26:52 * xranby-64 (~familjen@anon) Quit (Ping timeout: 264 seconds)
20131209 06:48:20 * [Mike] (~Mike]@anon) Quit ()
20131209 07:59:52 * hija (~hija@anon) Quit (Read error: Connection reset by peer)
20131209 08:04:09 * hija (~hija@anon) has joined #jogamp
20131209 08:11:03 * hharrison (~chatzilla@anon) Quit (Remote host closed the connection)
20131209 08:28:28 * monsieur_max (~maxime@anon) has joined #jogamp
20131209 08:36:02 * hija (~hija@anon) Quit (Quit: hija)
20131209 08:42:27 * monsieur_max (~maxime@anon) Quit (Quit: Leaving.)
20131209 08:44:44 * monsieur_max (~maxime@anon) has joined #jogamp
20131209 09:21:38 * xranby (~xranby@anon) has joined #jogamp
20131209 09:24:25 * hija (~hija@anon) has joined #jogamp
20131209 09:28:48 * eclesia (husky@anon) has joined #jogamp
20131209 09:29:00 <eclesia> good morning
20131209 09:42:28 * promswet (~promswet@anon) has joined #jogamp
20131209 09:42:39 * promswet (~promswet@anon) Quit (Remote host closed the connection)
20131209 09:55:00 <xranby> good morning
20131209 09:55:15 <monsieur_max> hi people
20131209 10:12:13 <eclesia> question : can we use a newt canvas inside a swing jpanel ?
20131209 10:39:35 <xranby> eclesia yes using NEWTCanvasAWT http://jogamp.org/deployment/jogamp-current/javadoc/jogl/javadoc/com/jogamp/newt/awt/NewtCanvasAWT.html
20131209 10:43:01 <xranby> http://jogamp.org/jogl/doc/NEWT-Overview.html
20131209 10:44:20 <xranby> example: http://jogamp.org/git/?p=jogl.git;a=blob;f=src/test/com/jogamp/opengl/test/junit/newt/parenting/TestParenting01cSwingAWT.java;hb=HEAD#l124
20131209 10:46:28 <eclesia> nice thanks
20131209 10:48:58 <bbbruce> has anyone encountered problems with NEWT, undecorated windows and multi screen setups before?
20131209 10:49:07 <xranby> eclesia: have you considered using the GL4ES3 interface ? to make your framework work using the shared functionality of OpenGL 4 and OpenGL ES 3 ?
20131209 10:49:08 <xranby> http://jogamp.org/deployment/jogamp-next/javadoc/jogl/javadoc/javax/media/opengl/GL4ES3.html
20131209 10:49:27 <bbbruce> whenever I hide the screen decorations and position the window on the secondary screen the window never becomes visible
20131209 10:49:37 <bbbruce> on OSX
20131209 10:51:19 <xranby> bbbruce: not really, what kind of problems?
20131209 10:52:06 <bbbruce> just the window never becoming visible
20131209 10:52:26 <eclesia> xranby: 0% of the enormous developer team of the project have a mobile device with android. so no, never considered using gl4es3
20131209 10:54:13 <xranby> eclesia: i know what you mean, i had to borrow my "moms" brand new galaxy note 3 in order to test gl4es3... mesa on linux also support it since it implement OpenGL ES 3 foss drivers for use on AMD and nvidia cards
20131209 10:56:19 <xranby> thus if you want to support the current crop of linux distributions using mesa this interface may enable more desktops to be supported
20131209 10:57:04 <xranby> and intel
20131209 11:00:39 <eclesia> xranby: It's not that I don't want to, problem is the enormous developers team == 1 developer. I can't do everything, compatibility has a very low priority. GL4 is the baseline I have set, I don't care about anything else unless someone else want to take care of it.
20131209 11:03:42 <xranby> bbbruce: hmm.. do you have access to some other system to test on?
20131209 11:03:49 <xranby> is it working on non-OSX
20131209 11:03:51 <xranby> ?
20131209 11:04:59 <bbbruce> yeah, it works on non-OSX
20131209 11:05:29 <xranby> can you update one of the jogl/src/test/com/jogamp/opengl/test/junit/newt/mm junit tests to trigger this issue on OSX?
20131209 11:07:38 <xranby> osx should behave similar to all other platforms thus please file a bug
20131209 11:07:59 <bbbruce> yeah, I was already working on the simplest program that would trigger this
20131209 11:09:23 <xranby> TestScreenMode01cNEWT.java
20131209 11:09:33 <xranby> Fullscreen on separate monitors, incl. spanning across multiple monitors.
20131209 11:09:40 <xranby> maybe this test needs improvement
20131209 11:09:51 <bbbruce> ah
20131209 11:09:52 <bbbruce> yeah
20131209 11:10:05 <bbbruce> also fullscreen on multi monitors is quite broken on osx in my experience
20131209 11:10:16 <bbbruce> but I am not sure if I am doing something wrong
20131209 11:12:07 <xranby> bbbruce: sgothel did most of the multi monitor tests using a virtual windows machine and a virtual linux machine spanning 4 monitors
20131209 11:13:14 <xranby> forum discussion: http://forum.jogamp.org/Re-Our-Road-to-2-0-2-NEWT-Multi-Monitor-Support-td4029101.html#a4029109
20131209 11:13:43 <xranby> thank you for testing it on OSX!
20131209 11:17:40 <xranby> "If you are able to compile the sources and run (at least) the 'TestScreenMode*' unit tests
20131209 11:17:40 <xranby> and give us the detailed results (7z compressed file) in case of a failure - it would help us a lot. "
20131209 11:33:33 <bbbruce> my.. that's a lot of tests :)
20131209 11:34:03 <bbbruce> (all jogl tests that is)
20131209 11:36:01 <xranby> bbbruce: you may edit jogl/make/scripts/tests.sh to only run the tests you are interested in and then run the scripts/tests-osx-x64.sh from the jogl/make directory :)
20131209 11:36:11 <xranby> ofc.. if you run all tests the better :)
20131209 11:38:45 <xranby> make all tests pass on all jogamp jenkins platforms is the release criteria :)
20131209 11:40:04 <xranby> 10000 tests are run for each jogl release! https://jogamp.org/chuck/job/jogl/
20131209 11:41:01 <xranby> https://jogamp.org/chuck/job/jogl/lastCompletedBuild/testReport/ last build got 7 failed tests of 10,146
20131209 11:42:26 <xranby> thus on osx there is only three known AWT bugs
20131209 11:44:15 <xranby> sorry i mean.. two failed awt tests and one failed swt test
20131209 11:59:32 <bbbruce> running scripts/tests-osx-x64.sh from jogl/make gives me build-dir ../build-macosx is not a directory
20131209 12:02:18 <xranby> you may change tests-osx-x64.sh it assumes the build directory to be build-macosx
20131209 12:03:31 <xranby> all this script do is helping pass the arguments to the tests.sh script
20131209 12:04:26 <xranby> if you simply run ant then the build directory will be named build
20131209 12:49:27 * xranby (~xranby@anon) Quit (Quit: Leaving.)
20131209 13:08:22 * dnkrozz (~dnkrozz@anon) has joined #jogamp
20131209 13:08:22 * dnkrozz (~dnkrozz@anon) Quit (Remote host closed the connection)
20131209 13:16:18 * xranby (~xranby@anon) has joined #jogamp
20131209 13:16:36 <xranby> https://gist.github.com/xranby/76df44c6be98e6f9d89f - Ubuntu 12.04 i386 crash in [libX11.so.6+0x28063] XEventsQueued+0x33 after re-parent undecorated fullscreen window in browser. jogamp jogl 2.1.2 GL_VENDOR NVIDIA Corporation GL_RENDERER GeForce GT 610/PCIe/SSE2 GL_VERSION 4.3.0 NVIDIA 319.32
20131209 13:17:39 <xranby> steps to reproduce: 1. _JAVA_OPTIONS="-Dnativewindow.debug=all -Djogl.debug=all -Dnewt.debug=all" firefox http://jogamp.org/deployment/jogamp-current/jogl-demos/jogl-newt-applet-runner-angelesgl2es1.html
20131209 13:17:43 <xranby> 2. press the keys in the following order: f d r r -> crash
20131209 13:38:48 <xranby> is anyone able to reproduce this crash?
20131209 13:41:19 <bbbruce> xranby I get dyld: Symbol not found: __cg_jpeg_resync_to_restart when trying to run the tests through scripts/tests-osx-x64.sh
20131209 13:41:51 <xranby> interesting
20131209 13:42:13 <xranby> there is a generated test log
20131209 13:42:52 <bbbruce> I have no problems running them from junit.run
20131209 13:42:58 <xranby> java-run.log
20131209 13:43:01 <bbbruce> ant junit.run that is
20131209 13:45:13 <bbbruce> http://pastebin.com/QXkn3jQM
20131209 13:46:06 <xranby> it hopefully contain more info
20131209 13:46:06 <xranby> bbbruce: ^
20131209 13:46:37 <xranby> you should be able to see which Test Start before the Symbol not found was triggered
20131209 13:47:06 <bbbruce> I enabled all the newt tests, every single one of them appears to do this
20131209 13:47:40 <bbbruce> (pastebin url contains my java-run.log)
20131209 13:48:08 <xranby> do you have a third party libJPEG installed on your OSX system?
20131209 13:48:10 <xranby> /usr/local/lib/libJPEG.dylib
20131209 13:50:19 <xranby> normally libJPEG.dylib is found in /System/Library/Frameworks/ImageIO.framework/Versions/A/Resources/libJPEG.dylib
20131209 13:50:48 <bbbruce> could be that brew placed that there
20131209 13:51:10 <bbbruce> ah yes.
20131209 13:52:05 <xranby> the error you saw was simply caused by /usr/local/lib/libJPEG.dylib not implementing the same methods compared to /System/Library/Frameworks/ImageIO.framework/Versions/A/Resources/libJPEG.dylib
20131209 13:52:36 <xranby> and your DYLD_LIBRARY_PATH find the lib in /usr/local/lib before /System/Library/Frameworks/ImageIO.framework/Versions/A/Resources/
20131209 13:53:25 <bbbruce> ok, but that does not really explain why I can run the tests through ant and not from the script
20131209 13:54:50 <xranby> cat scripts/tests-osx-x64.sh
20131209 13:54:55 <xranby> export DYLD_LIBRARY_PATH=/usr/local/lib:$DYLD_LIBRARY_PATH
20131209 13:54:58 <xranby> thats why
20131209 13:55:18 <xranby> the DYLD_LIBRARY_PATH added /usr/local/lib/
20131209 13:55:23 <xranby> when you used the script
20131209 13:55:43 <xranby> try comment out that line :)
20131209 13:56:05 <xranby> the export DYLD_LIBRARY_PATH=/usr/local/lib:$DYLD_LIBRARY_PATH from scripts/tests-osx-x64.sh
20131209 13:56:53 <bbbruce> more … better
20131209 16:09:23 * rallipilot (~rallipilo@anon) has joined #jogamp
20131209 16:09:25 * rallipilot (~rallipilo@anon) Quit (Remote host closed the connection)
20131209 16:56:57 * monsieur_max (~maxime@anon) Quit (Quit: Leaving.)
20131209 17:03:08 * eclesia (husky@anon) has left #jogamp
20131209 17:38:36 * xranby (~xranby@anon) Quit (Ping timeout: 260 seconds)
20131209 17:42:03 * xranby (~xranby@anon) has joined #jogamp
20131209 18:30:24 * monsieur_max (~maxime@anon) has joined #jogamp
20131209 18:32:54 * hija (~hija@anon) Quit (Quit: hija)
20131209 18:33:13 * monsieur_max (~maxime@anon) Quit (Client Quit)
20131209 18:35:10 * monsieur_max (~maxime@anon) has joined #jogamp
20131209 19:09:37 * xranby (~xranby@anon) Quit (Ping timeout: 272 seconds)
20131209 19:13:03 * xranby (~xranby@anon) has joined #jogamp
20131209 19:19:03 * hija (~hija@anon) has joined #jogamp
20131209 19:29:39 * [Mike] (~Mike]@anon) has joined #jogamp
20131209 19:46:17 * hija (~hija@anon) Quit (Quit: hija)
20131209 19:52:22 * hija (~hija@anon) has joined #jogamp
20131209 20:04:55 * hija (~hija@anon) Quit (Quit: hija)
20131209 20:17:51 * odin_ (~Odin@anon) has joined #jogamp
20131209 21:33:35 * hija (~hija@anon) has joined #jogamp
20131209 21:33:52 <hija> Can't load library: /System/Library/Frameworks/gluegen-rt.Framework/gluegen-rt
20131209 21:33:55 <hija> ahhhhhh
20131209 21:34:05 <hija> any ideas guys?
20131209 21:35:44 <monsieur_max> hija: which os, architecture, java ?
20131209 21:36:44 <hija> osx mavericks jdk 1.7.0.45
20131209 21:36:56 <hija> (64)
20131209 21:38:33 <hija> i have a seperate project in eclipse and in my build path jogl and glugen
20131209 21:38:38 <hija> not the jars, the source
20131209 21:40:00 <[Mike]> sgothel: Hi Sven, I have a question about commit 538a4184 "Fix NEWT WindowImpl reparent-recreate w/ GLEventListenerState: Bug introduced w/ commit e2506d7"
20131209 21:41:43 <[Mike]> sgothel: the wasVisible variable is added which checks if the window isVisible. The problem is seems is that in NewtCavasAWT.attachNewtChild() the newtChild visibility is being set to false before the reparent operation is called.
20131209 21:42:46 <[Mike]> sgothel: NewtCanvasAWT.java:741
20131209 21:43:29 <hija> maybe i'll just restart
20131209 21:44:13 * hija (~hija@anon) Quit (Quit: hija)
20131209 21:48:24 * hija (~hija@anon) has joined #jogamp
20131209 21:48:42 <hija> nope
20131209 21:50:54 <hija> any ideas?
20131209 21:52:54 <hija> Can't load library: /System/Library/Frameworks/nativewindow_macosx.Framework/nativewindow_macosx
20131209 21:52:56 <hija> meh
20131209 21:58:18 <[Mike]> sgothel: The end goal of mine is to make sure the GLContext does not get destroyed during reparent operation.
20131209 22:26:34 * monsieur_max (~maxime@anon) has left #jogamp
20131209 22:34:13 * xranby (~xranby@anon) Quit (Ping timeout: 248 seconds)
20131209 22:35:41 * xranby (~xranby@anon) has joined #jogamp
20131210 00:01:41 * xranby (~xranby@anon) Quit (Ping timeout: 248 seconds)
20131210 00:04:27 * xranby (~xranby@anon) has joined #jogamp
20131210 00:44:38 * rmk0_ (~rmk0@anon) has joined #jogamp
20131210 00:46:08 * rmk0 (~rmk0@anon) Quit (Ping timeout: 250 seconds)
20131210 05:05:57 -jogamp- Continue @ http://jogamp.org/log/irc/jogamp_20131210050557.html