#jogamp @ irc.freenode.net - 20150903 05:05:08 (UTC)


20150903 05:05:08 -jogamp- Previous @ http://jogamp.org/log/irc/jogamp_20150901042208.html
20150903 05:05:08 -jogamp- This channel is logged @ http://jogamp.org/log/irc/jogamp_20150903050508.html
20150903 05:46:55 * monsieur_max (~maxime@anon) has joined #jogamp
20150903 06:22:50 * elect (~elect@anon) has joined #jogamp
20150903 06:31:07 <elect> hi
20150903 06:34:54 * jvanek (jvanek@anon) has joined #jogamp
20150903 06:36:58 * badshah400 (~badshah40@anon) has joined #jogamp
20150903 06:58:42 * guillaum1_mobile (~androirc@anon) has joined #jogamp
20150903 07:05:55 * guillaum1_mobile (~androirc@anon) Quit (Remote host closed the connection)
20150903 07:06:01 * guillaum1_mobile (~androirc@anon) has joined #jogamp
20150903 07:28:01 * chamo (~chamo@anon) has joined #jogamp
20150903 07:28:37 * chamo (~chamo@anon) Quit (Client Quit)
20150903 07:53:30 * guillaum1_mobile (~androirc@anon) Quit (Remote host closed the connection)
20150903 07:53:39 * guillaum1_mobile (~androirc@anon) has joined #jogamp
20150903 08:35:38 * guillaum1_mobile (~androirc@anon) Quit (Quit: Leaving...)
20150903 08:35:54 * guillaume_mobile (~androirc@anon) has joined #jogamp
20150903 08:43:15 * guillaume_mobile (~androirc@anon) Quit (Remote host closed the connection)
20150903 08:49:48 * chamo (~chamo@anon) has joined #jogamp
20150903 09:00:47 <xranby> a11mad11: after learning the base you should write a demo application demonstrating all the knowledge you have learned
20150903 09:01:51 <xranby> make this demo application something that inspire other people to knack their first line of code
20150903 09:26:46 <monsieur_max> a rainbow colored triangle fe !
20150903 09:32:15 <elect> a11mad11, what are you interested in?
20150903 11:01:51 <xranby> sgothel: i postponed bug 1187 to 2.3.3 since we do not know the root cause and it only trigger on the vc4 with the experimental driver
20150903 11:19:38 <a11mad11> i am interesting to do a game engine for a game with various graphical effect
20150903 11:36:30 * a11mad11 (~marc-anto@anon) Quit (Ping timeout: 250 seconds)
20150903 11:56:43 * a11ma11 (d8712ef2@anon) has joined #jogamp
20150903 12:09:27 * a11ma11 (d8712ef2@anon) Quit (Ping timeout: 246 seconds)
20150903 12:09:42 <xranby> sgothel: i can re-produce the 1211 quite sucessfully now https://jogamp.org/bugzilla/show_bug.cgi?id=1211#c6
20150903 12:09:54 <sgothel> great
20150903 12:11:15 <sgothel> which jvm?
20150903 12:11:23 <xranby> java -version
20150903 12:11:23 <xranby> java version "1.7.0_79"
20150903 12:11:23 <xranby> OpenJDK Runtime Environment (IcedTea 2.5.6) (7u79-2.5.6-0ubuntu1.14.04.1)
20150903 12:11:23 <xranby> OpenJDK 64-Bit Server VM (build 24.79-b02, mixed mode)
20150903 12:11:33 <sgothel> on x86_64?
20150903 12:11:41 <xranby> yes x86_64
20150903 12:11:47 <sgothel> all seems to matter .. it seems :)
20150903 12:12:10 <sgothel> I will make a setSize .. test case overriding Thread.interupt() of the test Thread
20150903 12:12:48 <sgothel> then we need to figure .. legitimacy .. for the cond-wait use-cases
20150903 12:13:26 <sgothel> legitimacy of the interruption .. if legitimate: end, otherwise: continue -> cancellation ..
20150903 12:13:48 <sgothel> 1211 opened a little Pandora box :)
20150903 12:14:57 <sgothel> I remember that earlier we always used 'persistent-wait' to fight spurious-wakeups .. dunno why I ended that
20150903 12:15:19 <sgothel> also .. swallowing interrupts was quite common .. even though questionable design
20150903 12:22:08 <xranby> i have happy gnomes delivering fibre connection into my garden... https://www.facebook.com/xranby/posts/10153251174755805
20150903 12:23:20 <xranby> sgothel: yes i am quite surprised 1211 even existed
20150903 12:23:21 <elect> which speed will you reach?
20150903 12:23:46 <xranby> elect: depend how much I want i can get up to 1000Mbit
20150903 12:24:10 <elect> what are you gonna choose?
20150903 12:24:46 <xranby> i will likely settle for 100/100
20150903 12:24:51 <xranby> https://www.bahnhof.se/telia-oppen-fiber/telia-oppen-fiber/villa#orderNow
20150903 12:24:52 <elect> wow, symmetric
20150903 12:24:58 <elect> awesome
20150903 12:25:02 <elect> how much is gonna cost?
20150903 12:25:13 <xranby> about 40€ a month
20150903 12:25:33 <elect> here I pay 19,50 for a flat tel + 20/1 Mb
20150903 12:26:17 <xranby> i will likely use bahnhof as my isp because they have a good track record on privacy
20150903 12:26:53 <elect> it's not bad 40
20150903 12:28:01 <elect> kabeldeutschland offers 100/6 for 40€
20150903 12:28:48 <xranby> i am making room to have servers running at home in a sound isolated place
20150903 12:29:27 <elect> you have to take a look to this http://www.phoronix.com/scan.php?page=article&item=basement-server-room&num=1 :D
20150903 12:30:07 <xranby> what? no servers in the toilet?
20150903 12:30:15 <elect> :s
20150903 12:30:50 <elect> cool, with 100mb you can run all the servers you want
20150903 12:32:13 <xranby> sgothel: kudos,i will run the test to the groud!
20150903 12:32:19 <xranby> ground
20150903 12:36:37 <sgothel> fiber .. sweet and healthy :)
20150903 12:38:53 <xranby> yes.. it i one of those things you get when they feel you are ready for it... i sent out querys to install fibre 1.5 year ago ... and now.. boom.. they push fibre to the whole neighbourhood
20150903 12:39:00 <xranby> it is
20150903 12:40:08 <sgothel> have to wait a bit here, new neighborhood still uses copper .. but fiber is on its way .. just a matter of month/years :)
20150903 12:40:32 <sgothel> still .. synchronous rates are rarely provided
20150903 12:42:26 <rmk0> a lot of consumer isps are hostile to running servers
20150903 12:42:59 <xranby> sgothel: do you have any debug flags i should try turn on that may provide additional information to solve 1211?
20150903 12:43:01 <rmk0> is stupid. "we'll give you 80mbps up! but DON'T YOU DARE TRY USING IT FOR ANYTHING."
20150903 12:43:23 <xranby> rmk0: i suspect they will cap everything in 5 years
20150903 12:43:26 <xranby> for example
20150903 12:43:30 <sgothel> yeah .. the upload rate is ridiculous
20150903 12:44:04 <sgothel> @xerxes: nope .. we need a prepared test here to learn about the interrupt issuer
20150903 12:44:19 <xranby> when they pushed wireless broadband they gave everyone unlimited contracts. two years later you get a letter that they will replace your contract with a limited one and if you are unhappy they will terminate it free of charge
20150903 12:44:58 <xranby> thus we had 3 years of cheap unlimited wireless broadband in sweden
20150903 12:45:04 <xranby> was great to hack jogl on the bus
20150903 12:45:08 <xranby> but this is history now
20150903 12:45:35 <sgothel> some powers trying to unregulate the market, where the EU regulation was the source of broad availability
20150903 12:45:49 <sgothel> there goes your freedom again
20150903 12:46:12 <sgothel> ahead of ttip and all the crappy BS
20150903 12:46:37 <rmk0> i'm going to switch to http://idnet.net/data_products/fttc-self-install.php soon
20150903 12:46:50 <rmk0> they actually told me i could run my own freebsd server instead of using their provided router, if i wanted
20150903 12:46:53 <rmk0> ♥ ♥ ♥
20150903 12:47:40 <rmk0> free static ipv4 address...
20150903 12:47:50 <sgothel> yeah .. thats another thing, using your own networking stuff, EU is discussing it .. can only takes years .. until, wait, oh .. hurting a 'foreign' company *ttip*
20150903 12:47:55 <sgothel> lol
20150903 12:48:05 <sgothel> whats ipv4 good for anymore?
20150903 12:48:25 <sgothel> my stupid provider here .. one contact: no ipv6 avail .. ridiculous
20150903 12:48:26 <rmk0> well it's more than i'd get for free with anyone else. free ipv6 subnet too
20150903 12:48:52 <sgothel> just saying .. there are millions of people w/o ipv4
20150903 12:49:07 <rmk0> yep
20150903 12:49:27 <rmk0> i really hope everyone gets it together to deploy ipv6 before everything gets fucked up with carrier grade nat
20150903 12:49:54 <xranby> rmk0: +1!
20150903 13:05:27 <monsieur_max> https://blog.scaleway.com/2015/09/02/we-are-slashing-the-c1-price-by-70-percent/
20150903 13:05:46 <monsieur_max> very interesting offer
20150903 13:07:08 <xranby> monsieur_max: interesting provider, so they hav a ARm cluster?
20150903 13:07:20 <xranby> "We spawn BareMetal arm SSD cloud servers"
20150903 13:07:38 <monsieur_max> xranby: yep they built all by themselves, made in france
20150903 13:11:54 <xranby> https://twitter.com/MkHeck/status/639261417061810176 - Mark Heckler : Today was my last day at Oracle. Looking for a batteries-included #cloud/#IoT/#Java evangelist who knows it & shares it? Put me to work!
20150903 13:18:20 * a11mad11 (d8712ef2@anon) has joined #jogamp
20150903 13:19:20 <xranby> http://www.enterprisetech.com/2015/09/01/researchers-take-wraps-off-open-source-gpu/ - Open Source GPU capable of running OpenCL
20150903 13:30:20 <xranby> Oracle will implement java for mips, but outside OpenJDK "the MIPS port will not be done in OpenJDK. We do our work for general purpose desktop/server hardware in OpenJDK, but not the work for the Java embedded platforms." https://blogs.oracle.com/henrik/entry/oracle_announces_plans_for_java
20150903 13:45:58 <xranby> sgothel: if a user create an enourmous GLWindow lets say size 8000,6000 and use setResizable(false) before making a GLWindow visible, should this window still be auto-resized to fit within the screen limits?
20150903 13:46:29 <xranby> i have creates a junit test that expose this issue ( if we expect the window to stay 8000x6000)
20150903 13:47:19 <xranby> the user surely expects the window to stay 8000x6000
20150903 13:56:19 <xranby> sgothel: junit test https://github.com/xranby/jogl/commit/401a01ba31cf6c98d95a67dfa1021af854438044
20150903 14:01:38 <xranby> the above testcase fail with AssertionError: expected:<8000> but was:<1855>
20150903 14:01:52 <xranby> since newt has auto resized the window
20150903 15:00:53 <sgothel> clipping .. is expected
20150903 15:02:09 <sgothel> I assume they use it this way due to some hi-res screenshot or printin?
20150903 15:02:14 <sgothel> printing
20150903 15:02:36 <sgothel> then they shall use offscreen and maybe even tilted-renderer .. as we use for printing
20150903 15:03:03 <sgothel> b/c no platform can guarantee such window sized >> virtual desktop size
20150903 15:04:27 <sgothel> Oracle's Henrik: "but not the work for the Java embedded platforms". *No shit Oracle*
20150903 15:04:52 <sgothel> Oracle's Henrik: "We do our work for general purpose.." "We do" .. "We"
20150903 15:05:01 <sgothel> *No shit*
20150903 15:05:23 <sgothel> exactly the fucking reason I hate the OpenJDK chair
20150903 15:05:27 <sgothel> forgive me my language
20150903 15:06:05 <sgothel> one would say: 'half ass free software'
20150903 15:06:37 <elect> lol
20150903 15:11:20 * jvanek (jvanek@anon) Quit (Quit: Leaving)
20150903 15:12:59 <sgothel> @Xerxes: That 'open source' GPU .. is not / it uses AMD IP AFAIK
20150903 15:13:51 <sgothel> it is merely tolerated and allowed to be used for research .. AFAIK
20150903 15:16:06 <sgothel> reminder Oracle + embedded: https://jogamp.org/bugzilla/show_bug.cgi?id=698
20150903 15:22:22 <sgothel> https://jogamp.org/bugzilla/show_bug.cgi?id=698#c7 -> "Or use OpenJDK" -> the reason why they won't merge IcedTea
20150903 15:22:45 <sgothel> (and other JVMs)
20150903 15:23:46 <sgothel> aka 'OpenJDK has the implied blessing to be free of patent claims .. etc', hence we ensure it won't ruin our business strategy of 60 cents per unit :)
20150903 15:24:15 <sgothel> (instead of simply trusting ones performance .. asking for cents per unit due to own's great work)
20150903 15:24:58 <sgothel> looks like the blue ray TVs are still generating some income? :)
20150903 15:29:45 * elect (~elect@anon) Quit (Ping timeout: 250 seconds)
20150903 15:34:54 * a11mad11 (d8712ef2@anon) Quit (Ping timeout: 246 seconds)
20150903 15:37:43 * elect (~elect@anon) has joined #jogamp
20150903 15:44:19 <sgothel> @Xerxes: test this .. http://jogamp.org/git/?p=jogl.git;a=commit;h=5f5553f1c0b6731970db6df24d79654661238247
20150903 15:44:36 <sgothel> my desktop does not produce interruption .. trying other machines :-/
20150903 15:47:50 <sgothel> testing on one test node linux-nvidia, where it happened sometimes ..
20150903 15:52:44 * monsieur_max (~maxime@anon) Quit (Quit: Leaving.)
20150903 15:52:57 <sgothel> "TestBug1211IRQ00NEWT -time00 10000 -time 01 10000" on test node .. no interrupt :-/
20150903 15:55:35 <sgothel> each *trigger* is ~100ms .. so 10000 -> 100 times
20150903 15:55:41 <sgothel> testing w/ 1000 times .. dunno
20150903 15:57:13 * elect (~elect@anon) Quit (Ping timeout: 252 seconds)
20150903 16:26:27 * elect (~elect@anon) has joined #jogamp
20150903 16:31:19 * elect (~elect@anon) Quit (Ping timeout: 252 seconds)
20150903 16:50:38 * chamo (~chamo@anon) Quit (Remote host closed the connection)
20150903 16:51:50 * monsieur_max (~maxime@anon) has joined #jogamp
20150903 18:55:56 * a11mad11 (d8712ef2@anon) has joined #jogamp
20150903 19:52:30 * a11mad11 (d8712ef2@anon) Quit (Ping timeout: 246 seconds)
20150903 20:04:39 * elect (~elect@anon) has joined #jogamp
20150903 20:35:44 * monsieur_max (~maxime@anon) Quit (Quit: Leaving.)
20150903 20:53:35 * elect (~elect@anon) Quit (Ping timeout: 256 seconds)
20150903 21:19:47 * xranby_f22 (~familjen@anon) has joined #jogamp
20150903 21:20:02 <sgothel> last try .. 32bit java7 on that particular linux node ..
20150903 21:20:24 <xranby_f22> sgothel: hi, i have a usage pattern that makes your test case crash gnome shell
20150903 21:20:24 <sgothel> if I cannot reproduce it .. hmm, I guess I have to 'blindly' patch it
20150903 21:20:30 <xranby_f22> in fedora f22
20150903 21:20:34 <sgothel> uh
20150903 21:20:53 <sgothel> better than nothing .. almost got the idea that we need a broken WM for it :)
20150903 21:21:02 <xranby_f22> sh scripts/tests.sh /usr/lib/jvm/java-1.8.0-openjdk/bin/java " " ../build
20150903 21:21:02 <xranby_f22> then grab the firefox window with your left mouse button and hold the ouse dragged to the right side
20150903 21:21:05 <sgothel> since I use Kwin here .. which works flawless
20150903 21:21:49 <xranby_f22> to make firefox expand to cover half the screen, when the test enters the second test phase then it will instantly crash gnome shell
20150903 21:22:05 <xranby_f22> test00.resize[9]: 900x700 -> 800x600
20150903 21:22:05 <xranby_f22> ++++ TestCase.tearDown: com.jogamp.opengl.test.junit.newt.TestBug1211IRQ00NEWT - test00
20150903 21:22:05 <xranby_f22> .++++ TestCase.setUp: com.jogamp.opengl.test.junit.newt.TestBug1211IRQ00NEWT - test01
20150903 21:22:05 <xranby_f22> test01.create[0]: [visible, focused, ], 800x600
20150903 21:22:05 <xranby_f22> *gnomeshell crashed here* and our test is locked up
20150903 21:22:08 <sgothel> I did this w/ KWin ..
20150903 21:22:13 <sgothel> working well
20150903 21:22:25 <sgothel> well .. that is another result .. which is unusable
20150903 21:22:33 <sgothel> i.e. a gnome crash .. duh
20150903 21:22:53 <xranby_f22> if i do not keep the mousebutton pressed holding ta firefox window then it work as normal and i can run it 1000's of times
20150903 21:23:15 <sgothel> so .. hard to reproduce it .. to say the least
20150903 21:23:19 <xranby_f22> yes yes
20150903 21:23:59 <sgothel> just curious who is the interruptor :)
20150903 21:24:03 <xranby_f22> i will try the testcase using ubuntu 14.04 tomorrow with their unity window manager
20150903 21:24:27 <sgothel> yeah .. I will try w/ a virtualmachine one more time ..
20150903 21:24:30 <sgothel> thx a lot
20150903 21:24:36 <xranby_f22> the bugreport was using the unity window manager ( that is a compiz plugin)
20150903 21:24:45 <sgothel> in the end .. I guess we have to make it Noncancelable ..
20150903 21:24:48 <sgothel> *anyways*
20150903 21:24:55 <sgothel> ha
20150903 21:25:07 <sgothel> the perfect broken combination :)
20150903 21:25:14 <sgothel> but our linux-nvidia node ..
20150903 21:25:17 <xranby_f22> what machine and setup do you manage to reproduce it with?
20150903 21:25:26 <sgothel> uses kwin
20150903 21:25:32 <sgothel> ubuntu .. old ..
20150903 21:25:38 <xranby_f22> ah, ok well good for me that it can be reproduced using kwin
20150903 21:25:40 <xranby_f22> hehe
20150903 21:25:58 <sgothel> Ubuntu 14.04.2 LTS
20150903 21:26:07 <xranby_f22> using one of your machines as well
20150903 21:26:13 <sgothel> right .. I updated it lately
20150903 21:26:47 <sgothel> ok .. ubuntu 14.04 unity+compiz .. earmarked
20150903 21:28:45 * xranby_f22 (~familjen@anon) Quit (Quit: Leaving.)
20150903 21:29:19 * xranby_f22 (~familjen@anon) has joined #jogamp
20150903 21:29:43 <xranby_f22> fun, if you crash gnome shell three times in a row, it logs you out explaining it is sorry crashing on you
20150903 21:30:42 <xranby_f22> at least i have a very reliable way of crashing gnome
20150903 21:36:57 <sgothel> hillarious .. virtualbox + ubuntu 15.04 - unity .. -> XIO error on glxgears -info
20150903 21:37:45 <sgothel> (pressing window-close button in window decoration)
20150903 21:39:21 <xranby_f22> i have sent a bugrport to robovm, their jvm teardown may cause the nvidia driver to crash https://github.com/robovm/robovm/issues/85
20150903 21:39:42 <xranby_f22> (also triggred by running our GLWindow)
20150903 21:39:50 <sgothel> then .. this is also (or more) an NV driver issue
20150903 21:39:55 <sgothel> dinner - laters
20150903 21:41:49 <xranby_f22> later
20150903 21:42:18 <xranby_f22> see you tomorrow, getting some sleep now before more unity testing at work
20150903 22:04:45 * a11mad11 (~marc-anto@anon) has joined #jogamp
20150904 03:39:46 * badshah400 (~badshah40@anon) Quit (Quit: badshah400)
20150904 05:05:08 -jogamp- Continue @ http://jogamp.org/log/irc/jogamp_20150904050508.html