#jogamp @ irc.freenode.net - 20151027 05:05:19 (UTC)
20151027 05:05:19 -jogamp- Previous @ http://jogamp.org/log/irc/jogamp_20151026050519.html
20151027 05:05:19 -jogamp- This channel is logged @ http://jogamp.org/log/irc/jogamp_20151027050519.html
20151027 06:00:30 * netlike (~netlike@anon) has joined #jogamp
20151027 06:35:50 * jvanek (jvanek@anon) has joined #jogamp
20151027 07:22:18 * elect (~GBarbieri@anon) has joined #jogamp
20151027 07:24:55 <elect> hi
20151027 07:25:58 * doev (~doev@anon) has joined #jogamp
20151027 07:26:59 * doev (~doev@anon) Quit (Client Quit)
20151027 08:25:44 * monsieur_max (~maxime@anon) has joined #jogamp
20151027 08:34:47 * eclesia (~husky@anon) has joined #jogamp
20151027 08:34:53 <eclesia> good morning
20151027 09:20:53 <elect> hey
20151027 09:51:25 * netlike (~netlike@anon) Quit (Quit: Leaving)
20151027 10:24:43 <xranby> rmk0: yes
20151027 10:24:58 <rmk0> nice!
20151027 10:25:17 <xranby> rmk0: i can both build and run the jogamp test suite.. however the free travis setup refuses to create logs larger than 5Mb
20151027 10:25:28 <rmk0> heh, right
20151027 10:25:34 <xranby> and if you run all tests then you do create logs larget than 5mb
20151027 10:25:48 <xranby> i have to add a smaller list of tests to run
20151027 10:26:01 <rmk0> is there a .travis.yml i can look at somewhere?
20151027 10:26:13 <xranby> https://github.com/xranby/jogl/commits/travis
20151027 10:26:20 <rmk0> thanks
20151027 10:26:52 <xranby> https://github.com/sgothel/jogl/compare/master...xranby:travis
20151027 10:26:55 <xranby> that is a good view
20151027 10:27:18 <rmk0> surprisingly short, would have assumed it would take more work than that
20151027 10:27:26 <rmk0> do they only expose a software gpu?
20151027 10:27:42 <xranby> yes, they will then test using mesa3d using software emulationn
20151027 10:27:53 <xranby> actually i have not experimented installing other mesa3d packages
20151027 10:28:08 <xranby> maybe they have a real gpu on the travis builders
20151027 10:28:28 <xranby> that you maay utilitize by installing the right mesa3d packages
20151027 10:29:39 <xranby> rmk0: if you can make this setup run a limited amount of tests that actually complete
20151027 10:29:46 <xranby> then we may push it to the main jogl branch
20151027 10:29:53 <xranby> to have travis test all jogamp commits
20151027 10:30:00 <xranby> using the mesa3d software rasterizer
20151027 10:30:20 <rmk0> right
20151027 10:31:45 <xranby> i belive travis uses ubuntu based build machines
20151027 10:32:56 <rmk0> yes, think so
20151027 11:07:41 * monsieur_max1 (~maxime@anon) has joined #jogamp
20151027 11:09:03 * monsieur_max (~maxime@anon) Quit (Ping timeout: 268 seconds)
20151027 11:55:40 <xranby> ant junit.run newt
20151027 11:55:41 <xranby> bash ./scripts/check-junit.sh ../build
20151027 12:04:58 * xranby (~xranby@anon) Quit (Quit: Leaving.)
20151027 12:16:07 * jvanek (jvanek@anon) Quit (Quit: Leaving)
20151027 12:16:22 * jvanek (jvanek@anon) has joined #jogamp
20151027 13:01:46 * rmk0 abandons any pretenses of supporting OpenGL ES
20151027 13:02:32 <eclesia> rmk0: it's hell ?
20151027 13:03:14 <rmk0> it's ... lame
20151027 13:03:59 <rmk0> and having worked with android for eight months, i now have no interest in mobile platforms whatsoever!
20151027 13:04:29 <eclesia> even those with nvidia chips and gl4 ?
20151027 13:04:48 <rmk0> if they can run java 8
20151027 13:05:35 <rmk0> i'm placing a hard requirement on opengl 3.3 until vulkan appears, then i'll proceed lemming-like off the vulkan cliff
20151027 13:07:19 <rmk0> previously i didn't use anything beyond opengl 3.1, because there weren't open drivers that supported anything higher for all hardware
20151027 13:07:32 <rmk0> but now mesa has 3.3 everywhere, so it's just a matter of people having recent drivers
20151027 13:07:38 <rmk0> if they don't have recent drivers, that's their problem
20151027 13:08:31 <rmk0> *not* requiring people to use proprietary drivers was a requirement
20151027 13:15:34 <rmk0> GL 4 has passed the unnecessarily complexity horizon, in my opinion
20151027 13:15:43 <rmk0> *unnecessary
20151027 13:16:11 <rmk0> layers and layers of irrelevant abstraction
20151027 13:16:15 <rmk0> looking forward to vulkan
20151027 13:26:55 * xranby (~xranby@anon) has joined #jogamp
20151027 13:27:48 <xranby> rmk0: i have updated the scripts to run on the new travis container based infrastructure.. however they have lowered the limbo bar.. he log length has exceeded the limit of 4 Megabytes
20151027 13:28:49 <xranby> https://github.com/sgothel/jogl/compare/master...xranby:travis-master
20151027 13:29:10 <xranby> i had to disable use of sudo to make it work on the new infrastructure
20151027 13:30:06 <xranby> here is the full log of my last build https://s3.amazonaws.com/archive.travis-ci.org/jobs/87663358/log.txt
20151027 13:31:03 * ctrueden (~ctrueden@anon) has joined #jogamp
20151027 13:34:33 <rmk0> xranby: thanks
20151027 13:34:45 <rmk0> why is it that the jogamp build logs to stdout?
20151027 13:34:59 <rmk0> is fairly standard practice in maven projects to log each test to its own file
20151027 13:35:21 <xranby> rmk0: zubzub tested to redirect the output, however then travis complained that no new output appeared within 2 min and failed for that reason :/
20151027 13:35:30 <rmk0> argh!
20151027 13:36:11 <rmk0> presumably just logging the name of each test to stdout would fix that?
20151027 13:36:13 <xranby> rmk0: i will try quiet the gluegen clone and build to check if the extra lines will make the jogl tests finish
20151027 13:36:40 <xranby> rmk0: yes resumably
20151027 13:36:42 <xranby> presumably
20151027 13:37:25 <xranby> ok we have an extra 3000 lines now
20151027 13:37:31 <xranby> https://travis-ci.org/xranby/jogl/
20151027 13:37:41 <xranby> my build is running
20151027 13:38:35 * rmk0 eyes it
20151027 13:41:41 <xranby> build sucess, now urnning the junit tests
20151027 13:41:44 <xranby> running
20151027 13:42:32 <rmk0> "This log is too long to be displayed."
20151027 13:42:39 <xranby> log longer than 10000 lines switching to raw log https://api.travis-ci.org/jobs/87668253/log.txt?deansi=true
20151027 13:42:46 <rmk0> yep
20151027 13:43:52 <xranby> [apply] GL_VENDOR VMware, Inc. [apply] GL_RENDERER Gallium 0.4 on llvmpipe (LLVM 0x300) [apply] GL_VERSION 2.1 Mesa 8.0.4
20151027 13:44:31 <xranby> The log length has exceeded the limit of 4 Megabytes (this usually means that test suite is raising the same exception over and over). The build has been terminated.
20151027 13:44:32 <rmk0> i hope that's not the highest GL_VERSION they provide /o\
20151027 13:44:32 <xranby> meh
20151027 13:45:15 <xranby> GL2 true [2.1 (Compat profile, compat[ES2], FBO, software)]
20151027 13:45:31 <xranby> supported mappings: GL2ES2 GLProfile[GL2ES2/GL2.sw]
20151027 13:45:41 <xranby> GL2ES1 GLProfile[GL2ES1/GL2.sw]
20151027 13:45:47 <xranby> GL2 GLProfile[GL2/GL2.sw]
20151027 13:45:53 <xranby> GL2GL3 GLProfile[GL2GL3/GL2.sw]
20151027 13:46:27 <xranby> looks like it... VMWare == gl 2.1
20151027 13:46:34 <rmk0> bleh!
20151027 13:46:54 <xranby> http://www.mesa3d.org/vmware-guest.html
20151027 13:47:09 <xranby> according to this document With the August 2015 Workstation 12 / Fusion 8 releases, OpenGL 3.3 is supported in the guest.
20151027 13:47:42 <xranby> however that require a newer os on the build container
20151027 13:48:01 <rmk0> what about if software rendering is forced?
20151027 13:48:38 <xranby> http://www.mesa3d.org/envvars.html
20151027 13:48:47 <xranby> LIBGL_ALWAYS_SOFTWARE
20151027 13:52:02 <rmk0> yep
20151027 13:52:18 <rmk0> would think that forcing software rendering would expose any version you want
20151027 13:55:35 * xranby now tries to use the and junit https://ant.apache.org/manual/Tasks/junit.html printsumamry
20151027 13:56:48 <xranby> ant junit printsummary
20151027 14:15:39 * monsieur_max1 is now known as monsieur_max
20151027 14:29:28 * bruce-_ is now known as bruce-
20151027 14:33:54 <xranby> this time it is still running https://api.travis-ci.org/jobs/87676179/log.txt?deansi=true
20151027 14:36:56 <xranby> .. no
20151027 14:37:38 <xranby> a new travis record.. ran for 19 min 40 sec
20151027 14:37:47 <xranby> https://travis-ci.org/xranby/jogl/builds/
20151027 14:43:15 <rmk0> seem to remember their limit being two hours
20151027 14:44:22 <xranby> running the jogl junit tests usually takes about 45min so that should be ok
20151027 14:44:47 <xranby> however need to reduce the junit chatter
20151027 14:45:15 <xranby> will try to remove the stderr during the test run.. hoping it will appear in the summary
20151027 15:02:31 * jvanek (jvanek@anon) Quit (Quit: Leaving)
20151027 15:58:07 <elect> https://medium.com/@schewick/europe-is-about-to-adopt-bad-net-neutrality-rules-here-s-how-to-fix-them-bbfa4d5df0c8#.xsqi3h701
20151027 16:21:38 * elect (~GBarbieri@anon) Quit (Ping timeout: 260 seconds)
20151027 16:27:32 * doev (~doev@anon) has joined #jogamp
20151027 16:52:29 * elect (~elect@anon) has joined #jogamp
20151027 17:02:06 * eclesia (~husky@anon) has left #jogamp
20151027 17:16:53 * doev (~doev@anon) Quit (Quit: Verlassend)
20151027 17:51:45 * doev (~doev@anon) has joined #jogamp
20151027 18:29:07 * elect (~elect@anon) Quit (Ping timeout: 252 seconds)
20151027 18:55:16 * doev (~doev@anon) Quit (Ping timeout: 268 seconds)
20151027 19:39:23 <zubzub> xranby: I stopped using travis for anything serious ever since they seem to be stuck on 12.04
20151027 19:39:26 <zubzub> I ean
20151027 19:39:27 <zubzub> *mean
20151027 19:39:30 <zubzub> it's 2015
20151027 19:39:35 <zubzub> october
20151027 19:39:40 <zubzub> 15.10
20151027 19:40:08 <zubzub> I can't run wayland based thing because the version they use is completely outdated
20151027 19:40:19 <zubzub> same for GL is seems (2.1)
20151027 19:47:41 * elect (~elect@anon) has joined #jogamp
20151027 20:42:16 * agggile (58c03e45@anon) has joined #jogamp
20151027 20:42:20 * ctrueden (~ctrueden@anon) Quit (Ping timeout: 272 seconds)
20151027 20:54:28 * agggile (58c03e45@anon) Quit (Quit: Page closed)
20151027 21:02:46 * elect (~elect@anon) Quit (Remote host closed the connection)
20151027 21:10:42 * elect (~elect@anon) has joined #jogamp
20151027 21:11:37 * monsieur_max (~maxime@anon) Quit (Quit: Leaving.)
20151027 21:13:24 * monsieur_max (~maxime@anon) has joined #jogamp
20151027 21:34:53 * ctrueden (~ctrueden@anon) has joined #jogamp
20151027 21:35:08 * ctrueden (~ctrueden@anon) Quit (Client Quit)
20151027 21:36:19 * gouessej (5279cf66@anon) has joined #jogamp
20151027 21:39:49 <gouessej> Hi
20151027 21:40:05 <gouessej> Sven isn't here :(
20151027 21:40:30 <gouessej> I plan to migrate to Git and to leave Sourceforge
20151027 21:40:59 <gouessej> I'll probably need a temporary repository until I can self-host mine
20151027 21:44:26 <gouessej> bye
20151027 21:44:35 * gouessej (5279cf66@anon) Quit (Quit: Page closed)
20151027 21:57:59 <zubzub> it's always such a shame when you find a library
20151027 21:58:07 <zubzub> that does *almost* everything you want
20151027 21:58:21 <zubzub> so you decide to fix/implement that one thing yourself
20151027 21:58:28 <zubzub> and then you look at the source code
20151027 21:58:37 <zubzub> and you throw up a little in your mouth
20151027 21:58:52 <zubzub> (no I'm not talking about jogl ;p)
20151027 22:08:01 * elect (~elect@anon) Quit (Ping timeout: 252 seconds)
20151027 22:16:30 * monsieur_max (~maxime@anon) Quit (Quit: Leaving.)
20151028 05:05:19 -jogamp- Continue @ http://jogamp.org/log/irc/jogamp_20151028050519.html