| MCR1 | Hi :) Anyone here ? | 14:50 |
|---|---|---|
| smspillaz | MCR1: what up ? | 14:52 |
| MCR1 | Oh - Hi Sam :) Great to see you working on the missing functionality - but do not expect that I can help much there yet ;) | 14:53 |
| MCR1 | smspillaz, seems Expo is broken in trunk - Can you confirm ? | 14:53 |
| smspillaz | not right now | 14:54 |
| smspillaz | MCR1: what is "broken" about it ? | 14:54 |
| smspillaz | MCR1: I wrote a porting guide, check my blog for the link, I haven't got access to it at the moment (long story) | 14:54 |
| MCR1 | must have been a very recent commit that broke it... it crashes Compiz - restart via virtual terminal is needed | 14:55 |
| MCR1 | smspillaz: I have read it ;) | 14:55 |
| smspillaz | MCR1: nothing has been merged in the past month | 14:55 |
| smspillaz | MCR1: can you get a stacktrace ? | 14:55 |
| MCR1 | maybe it was a nux commit that broke it... | 14:55 |
| MCR1 | smspillaz: I can try, but I can try to find out which one broke it also - one min | 14:56 |
| smspillaz | MCR1: a stacktrace is more useful | 14:56 |
| smspillaz | use valgrind or something | 14:56 |
| MCR1 | you already told me once and I've saved it somewhere, but could you please repeat the steps necessary as I failed with trying to get a gdb stacktrace and have no experience using valgrind | 14:58 |
| smspillaz | MCR1: sudo apt-get install valgrind; valgrind compiz --replace ccp | 14:58 |
| smspillaz | it'll run slower, but you'll get a stack when it crashes, plus some other info as to why it crashed | 14:59 |
| MCR1 | ok | 14:59 |
| smspillaz | in other news, fuck me I'm an idiot. Just spent the last three hours trying to figure out why a pass-through shader that renders the contents of the backbuffer obtained by glCopyTexSubImage2D as "transparent" - when that's exactly what its meant to do | 15:00 |
| smspillaz | at least now I know the problem is in the texel fetch logic and not in the backbuffer copy | 15:01 |
| smspillaz | debugging opengl is a giant pain in the ass | 15:01 |
| smspillaz | ARGH TYPOES | 15:03 |
| smspillaz | s/2/3/ | 15:03 |
| MCR1 | valgrind setsid unity from VT seems to work... | 15:05 |
| smspillaz | MCR1: if you want copy-and-paste-ability just use a normal terminal | 15:05 |
| MCR1 | ok | 15:06 |
| MCR1 | Segmentation fault is all I get :( | 15:12 |
| smspillaz | MCR1: valgrind compiz --replace ccp in a normal term ? | 15:17 |
| smspillaz | surely not .... | 15:17 |
| MCR1 | smspillaz: Sorry - had to reboot - valgrind compiz --replace ccp fails: --9012:0:aspacem Valgrind: FATAL: VG_N_SEGNAMES is too low. --9012:0:aspacem Increase it and rebuild. Exiting now. | 15:27 |
| MCR1 | smspillaz: Something completely different: branch fixing firepaint is done :) | 15:28 |
| smspillaz | wtf | 15:31 |
| smspillaz | MCR1: you have weird hardware / screwed installed | 15:31 |
| smspillaz | MCR1: in any case, to get gdb output into a file, its something like | 15:32 |
| smspillaz | gdb --args compiz --replace ccp | 15:32 |
| smspillaz | set logging-file /home/username/gdb-log.txt | 15:32 |
| smspillaz | bt | 15:32 |
| smspillaz | bt full | 15:32 |
| smspillaz | q | 15:32 |
| smspillaz | thanks for the firepaint fixes, I'll take a look | 15:33 |
| smspillaz | though we need to port firepaint to not use quads, which I suspect will be a pain | 15:33 |
| smspillaz | since we actually have to do it, and not just emulate it with gl_triangles and primitive restarts since I'm pretty sure gles doesn't support that | 15:33 |
| MCR1 | well, I just created the branches, all the credit goes to Michail Bitzes | 15:33 |
| smspillaz | MCR1: you shouldn't underestimate your own ability | 15:34 |
| smspillaz | try your hand at porting some of the simpler plugins | 15:34 |
| smspillaz | its pretty easy | 15:34 |
| smspillaz | you just convert all the glVertex3f () crap to streamingBuffer->addVertices (1, vertex); etc | 15:35 |
| MCR1 | smspillaz: For now, firepaint and showmouse are disabled for GLES anyway, so I guess it should be okay... | 15:35 |
| smspillaz | well, one day phones and tablets will take over and all we'll have is GLES :P | 15:36 |
| MCR1 | see plugins/CMakeLists.txt - but ofc you are right - but recently I had not much time and failed to fix some of the grid issues I took on... | 15:37 |
| MCR1 | namely grid-keyboard-resized windows jumping around workspaces... :( | 15:37 |
| smspillaz | keep trying | 15:38 |
| MCR1 | no idea why those behave so much different, than the mouse-resized ones | 15:38 |
| smspillaz | :) | 15:38 |
| MCR1 | it is some strangeness going on there :P | 15:39 |
| smspillaz | MCR1: in case you were wondering what all my swearing about was earlier, I am trying to por^W^W^Wrewrite the blur plugin | 15:40 |
| smspillaz | it really is that bad | 15:40 |
| MCR1 | hehe | 15:41 |
| MCR1 | if you ask me 4 preferences regarding missing plugins... I would say -> bring back animations... | 15:42 |
| MCR1 | but ofc I am very happy about each of them coming back, although I did not use blur personally... | 15:42 |
| smspillaz | animations is easy | 15:42 |
| smspillaz | you could probably do it | 15:42 |
| smspillaz | GPU assembly is such indecipherable garbage ..... | 15:44 |
| MCR1 | lets make a deal -> you take one of the animations plugins and convert it, I'll study your diff and take on the others ? - how 'bout that one ? | 15:44 |
| smspillaz | done | 15:44 |
| MCR1 | we have simple, extra, experimental :) | 15:44 |
| smspillaz | not right now though, need to focus on getting blur done and then I have to fix some bugs in my experimental ppa | 15:45 |
| MCR1 | and animationsaddon IIRC | 15:45 |
| MCR1 | sure, whenever you find some time 4 it :) | 15:45 |
| MCR1 | & thanks a lot 4 your efforts ;) | 15:46 |
| smspillaz | MCR1: there we go WIP http://i.imgur.com/qjESr.png | 15:47 |
| smspillaz | MCR1: eh, I'm learning a fair bit | 15:47 |
| smspillaz | having to rip some of the more complicated plugins apart to get them to work forces you to learn how they work | 15:47 |
| smspillaz | the compiz blur algorithm is actually really neat | 15:48 |
| smspillaz | doesn't work the way you'd expect it to, has lots of tricks | 15:48 |
| MCR1 | I guess 'lots of tricks' is the correct way to describe the whole Compiz ;) | 15:49 |
| MCR1 | and it is 'compiz or bust' imho ;) - do not underestimate what we have ;) | 15:50 |
| smspillaz | just the parts that dennis wrote | 15:50 |
| smspillaz | MCR1: well, hopefully people will take the good bits and make it work on other compositors | 15:50 |
| smspillaz | fragmentation is stupid, I'm not going to contribute to that problem | 15:51 |
| MCR1 | yes, that is true - I never quite fully understood that open-source problem | 15:51 |
| MCR1 | all devs should gather around the best/most advanced project, but this is not the way it happens | 15:52 |
| MCR1 | smspillaz: Good news. Got the wizard to work with post-r3320 as well 8-) - was the same problem as with the other particle plugins ;) | 16:27 |
| MCR1 | smspillaz: Here it is: https://code.launchpad.net/~mc-return/compiz/compiz0.9.9.merge-plugin-wizard/+merge/141459 | 16:37 |
| kucksdorfs | Hello, can anyone help me with the login screen? I want to be able to only type my username and password, not choose my username from a list. | 17:22 |
| === rsalveti_ is now known as rsalveti | ||
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!
