IRC log for #maemo-ssu on 20121002

00:14.05merlin1991hm new kernel in the debian repos
00:17.33*** join/#maemo-ssu merlin1991_wrb (5472b26f@gateway/web/freenode/ip.84.114.178.111)
00:18.08*** join/#maemo-ssu merlin1991 (~merlin@Maemo/community/cssu/merlin1991)
00:19.27merlin1991server rebooted :D
00:24.51*** part/#maemo-ssu merlin1991_wrb (5472b26f@gateway/web/freenode/ip.84.114.178.111)
00:28.10merlin1991server back up with new kernel
01:31.23*** join/#maemo-ssu LaoLang_cool (~LaoLang_c@221.226.175.141)
01:45.05*** join/#maemo-ssu _rd (~rd@p57B494BD.dip0.t-ipconnect.de)
02:13.09*** join/#maemo-ssu amiconn_ (amiconn@rockbox/developer/amiconn)
02:27.15*** join/#maemo-ssu LaoLang_cool (~LaoLang_c@221.226.175.141)
03:03.42*** join/#maemo-ssu DocScrutinizer06 (~HaleBopp@openmoko/engineers/joerg)
05:29.32*** join/#maemo-ssu luf (luf@nat/ibm/x-ulsxrqhciababfvd)
06:35.13*** join/#maemo-ssu int_ua (~int_ua@ip-602f.proline.net.ua)
06:38.12freemangordonluf: you did find the bug in obexd?
06:39.30luffreemangordon: yes. As I disabled mas plugin also syncevolution plugin needs to be disabled (just config) due to bluetooth synchronization.
06:40.12luffreemangordon: the bug I mentioned is in MAP and I finally have stable initial map (this isn't in released obexd).
06:41.20freemangordongreat
06:58.42*** join/#maemo-ssu int_ua (~int_ua@ip-602f.proline.net.ua)
08:11.30freemangordonkerio: wanna try thumb cstreamer?
08:11.34freemangordon*gstreamer
08:25.26*** join/#maemo-ssu dhbiker (~dhbiker@95.87.145.172)
08:28.25*** join/#maemo-ssu M4rtinK (~M4rtinK@mail.melf.eu)
08:37.29freemangordonkerio: if you want to, .debs are here http://merlin1991.at/~freemangordon/cssu-thumb/gstreamer/
08:37.41freemangordonnot tested, so be careful
08:43.47*** join/#maemo-ssu rantom (rantom@kapsi.fi)
08:44.48rantomHi. Can I ask a quick question about tuning the transitions with/for CSSU?
08:47.29tadzikgo ahead
08:48.46rantomThanks. :) So I just moved back for, now, to my N900 and tried to customize the transitions with CSSU Features/etc. That didn't work that well though (I'm on CSSU-Stable) and I thought what could I use in its place
08:49.24rantomNow I found cssutransitionstuner in packages, that's stuck in -testing, would that be good alternative to use since I already have Theme Customizer isntalled?
08:50.15tadzikhuh. I just switched this .ini file somewhere )
08:50.20tadziknever used any tool for that
08:51.14rantomHeh, ok
08:51.33tadzikI don't change the transitions every few hours, so that seems sensible
08:51.56tadzikfreemangordon: actually, I can try this gstreamer
08:52.01rantomYeah, it'll be once the settings are complete
08:52.15tadzikfreemangordon: just download it all and dpkg -i it?
08:52.23rantomJust curious overall is there a reason not to use cssutransitionstuner
08:52.29rantome.g. incompatibility
08:52.44rantom(which would be a bit ironic considering the name :)
09:03.43*** join/#maemo-ssu CodenameStrike (~assassin@175.138.115.29)
09:15.37*** join/#maemo-ssu NIN101 (~NIN@p5DD281D1.dip0.t-ipconnect.de)
09:30.07*** join/#maemo-ssu int_ua_ (~int_ua@ip-602f.proline.net.ua)
09:31.38*** join/#maemo-ssu int_ua_ (~int_ua@ip-602f.proline.net.ua)
09:37.10lufIs here someone with tracker knowledge? I need to understand what part of query means:
09:37.12luf" FILTER (!BOUND(?msg)) "
09:38.31lufDoes MAP message listing in /, /telecom, /telecom/msg returns some/all/none messages?
09:56.49*** join/#maemo-ssu arcean (~Arcean@aadb8.neoplus.adsl.tpnet.pl)
10:25.39*** join/#maemo-ssu lizardo (lizardo@nat/indt/x-pratzasmwigefsxm)
10:28.37DocScrutinizer05luf: ?? I fthat's what I think it is, you should read about sparQL
10:44.55DocScrutinizer05I'd guess it means there's a filter on a table, to return all records where the field 'msg' is NOT bound to any data
10:46.35DocScrutinizer05luf: http://www.w3.org/TR/rdf-sparql-query/#func-bound
10:48.14DocScrutinizer05luf: http://www.w3.org/TR/rdf-sparql-query/#tests
11:00.36*** join/#maemo-ssu dhbiker (dhbiker@95.87.145.172)
11:06.13rantomBut thanks
11:06.14*** part/#maemo-ssu rantom (rantom@kapsi.fi)
11:10.39*** join/#maemo-ssu CodenameStrike (~assassin@175.144.139.85)
11:15.18lufDocScrutinizer05: yeah that's it. Thanks.
12:07.39*** join/#maemo-ssu andre__ (~andre@Maemo/community/bugmaster/andre)
12:13.39keriofreemangordon: not now, but i'll try it later
12:14.14kerioi'm at the uni, no easy way to recover
12:14.22freemangordonaah, I see
12:14.41lufkerio: just ask the blond girl she'll help you ;)
12:14.47freemangordonhehe
12:14.55keriohaha
12:15.08freemangordonkerio: better ask wise, not blone one :P
12:15.16freemangordon*blond
12:32.42lufUgh obexd patch againist orig grow from 130kB to 400kB :D
12:48.35DocScrutinizer05luf: honestly, SPARQL is such a braindamaged crap
12:50.18lufDocScrutinizer05: true :)
12:51.52DocScrutinizer05where in a decent language I'd define a condition like NOT ISEMPTY(data-element), in SPARQL I need a mindfsckng OPTIONAL { ?x dc:date ?date } . FILTER (!bound(?date))
12:52.41lufDocScrutinizer05: :D that's the reason I don't want study such crapy language.
12:54.51*** join/#maemo-ssu _rd (~rd@p57B494BD.dip0.t-ipconnect.de)
12:57.26lufWow. I finished initial MAP support for obexd. I'll upload it to the git and announce on the TMO.
12:57.41lufIs someone here interested to test it? :)
12:58.38*** join/#maemo-ssu Pali (~pali@unaffiliated/pali)
12:59.24DocScrutinizer05I'd not know how to test it
13:00.23DocScrutinizer05for fun, figure how to literally transcode OPTIONAL { ?x dc:date ?date } . FILTER (!bound(?date)) to SQL ;-D
13:00.55lufI'm still a little bit normal :D
13:03.20DocScrutinizer05select a.x a.y from mytab as a, b.x  b.z mytab as b where notempty(b.z) and b.x = a.x
13:03.25DocScrutinizer05somthing like that
13:04.05DocScrutinizer05s/not//
13:05.59DocScrutinizer05OPTIONAL  --- pfffffbwahaha
13:06.02lufDocScrutinizer05: do you want to kill as? Or just kill out brains? :D
13:06.12lufs/as/us/
13:06.26lufs/out/our/
13:06.56*** join/#maemo-ssu M13 (~Miha@170.133-224-87.telenet.ru)
13:07.08DocScrutinizer05I know whom to kill: the one who invented that SPARQL shit
13:10.59*** join/#maemo-ssu dafox (~dafox@ip51cc571d.speed.planet.nl)
13:13.37DocScrutinizer05I could bet it's been a math graduate with second study CS who never wrote a decent working program before he finished his math study
13:13.58DocScrutinizer05during his first semester of CS
13:42.13*** join/#maemo-ssu dafox (~dafox@ip51cc571d.speed.planet.nl)
13:47.52*** join/#maemo-ssu _rd (~rd@p57B494BD.dip0.t-ipconnect.de)
13:54.01m0shbearDocScrutinizer05: here's a question
13:54.16m0shbearwtf happened to just using ye olde sql for that kind of domain?
13:56.57*** join/#maemo-ssu arcean_ (~Arcean@aacs84.neoplus.adsl.tpnet.pl)
14:08.28lufok, new obexd is in git and also merlin1991.at :) I'll announce it on TMO in the evening or tomorrow. I need to fix dependency for librtcom-eventlogger1-async and rtcom-eventlogger-async.
14:09.06luffreemangordon: I think I'll take a look on bluez upgrade tomorrow or the day after tomorrow.
14:09.12freemangordongood, now you're back to BT :P
14:09.34luffreemangordon: :D as you wish :D
14:09.35freemangordonluf: don't forget my re-packaging ;)
14:09.44freemangordon:D:D:D
14:10.08lufSure it's the first point. I want to learn because it should be interesting also for obexd package ;)
14:10.12lufI have to leave now. Bye.
14:11.29merlin1991what did I think when I set up that domain?
14:11.46merlin1991I kinda missed the point of unintentional highlights on irc :D
14:29.56m0shbearmerlin1991: it's to troll those with op status in order to get /kicked :P
14:31.21merlin1991hehe
14:36.44*** join/#maemo-ssu int_ua (~int_ua@ip-602f.proline.net.ua)
14:48.03*** join/#maemo-ssu Sc0rpius (~naikel@190.201.107.179)
14:52.40keriofreemangordon: ok, i'm at home
14:52.43keriowhat should i test and how
14:52.53kerioactually first i gotta poop
14:53.05keriobut i have a laptop, so don't worry
14:57.03keriofreemangordon: that's a lot of debs D:
14:57.13kerioalso the versions aren't +thumb
15:01.03m0shbeari wonder how well-optimized mednafen's and desmume's virtualizations are, in order to avoid as much arm->arm translation as possible
15:01.44m0shbearsince a ds is pretty much an arm965e and gba an arm7tdmi with a few peripheral fancinesses
15:09.55freemangordonkerio: yes, no -thumb in version, but that should not be a problem
15:10.18kerioit IS >:C
15:10.28freemangordonI thoink you have all, but gstreamer0.10-tools installed
15:10.59freemangordonkerio: not, it is not, if everything is OK, +thumb version will come with next -pr thing
15:11.15keriolink again?
15:11.32freemangordon<PROTECTED>
15:12.12freemangordonis on his way home
15:12.57keriofreemangordon: how do i test if everything is working fine?
15:15.43*** join/#maemo-ssu Sc0rpius (~naikel@190.201.107.179)
15:24.34*** join/#maemo-ssu Milhouse (~irc_milho@Maemo/community/contributor/Milhouse)
15:34.51*** join/#maemo-ssu andre__ (~andre@ip-89-176-24-140.net.upcbroadband.cz)
15:34.51*** join/#maemo-ssu andre__ (~andre@Maemo/community/bugmaster/andre)
15:44.06keriofreemangordon: *poke*
15:44.50*** join/#maemo-ssu dhbiker (~dhbiker@95.87.145.172)
15:50.00keriofreemangordon: this is srsly weird, half of those packages don't even exist
15:55.47freemangordonkerio: ?
15:55.55freemangordonfor dure they do
15:56.00freemangordon*sure
15:56.04keriofreemangordon: gstreamer0.10-plugins-good-extra doesn't exist
15:56.17freemangordonwhere?
15:56.55kerioactually no
15:57.01keriogstreamer0.10-plugins-base-extra doesn't exist
15:57.11kerioanywhere
15:57.14kerioit's not in the nokia repos
15:57.17kerioit's not in extras-devel
15:58.37freemangordonhmm
16:00.22kerioalso, should i install stuff like gstreamer0.10-plugins-good-extra?
16:00.24kerioit's not installed now
16:00.46freemangordonhttps://bugs.maemo.org/show_bug.cgi?id=5463
16:00.49povbotBug 5463: Provide gstreamer0.10-plugins-base-extra (for gdppay or gdpdepay)
16:00.52freemangordonWTF?!?
16:01.19kerioWTF indeed
16:02.19freemangordonwell, you may want to not install it
16:02.44freemangordonthough it contains libgstgdp.so and libgstpango.so
16:02.54freemangordonwhatever those are used for
16:05.37freemangordonkerio: it is to construct gstreamer pipeline over a network :D
16:05.50kerioneat
16:06.43keriofreemangordon: what are you going to add to the mp?
16:07.27freemangordononly what is in stock, but will leave that extra package on the repo too (no option actually, because of the .changes file)
16:07.44kerioi see
16:07.53kerioanyway, i'm actually not going to test it >:D
16:08.04freemangordonwhich one?
16:08.08kerioneither
16:08.17kerios/neither/none/
16:08.21freemangordonup to you
16:08.30keriodo you actually need me to test it?
16:08.36freemangordonno
16:08.38kerioi don't want to copypaste a bunch of URLs :(
16:10.08freemangordoni'll install it on my primary device for couple of days
16:10.20freemangordons/it/them/
16:10.23*** join/#maemo-ssu Pali (~pali@unaffiliated/pali)
16:14.00*** join/#maemo-ssu CodenameStrike (~assassin@175.144.136.127)
16:21.42kerioPali: sup
16:21.50keriothere could be a kernel patch for you
16:21.58keriofreemangordon: what's the unaligned access thing again?
16:22.00Paliwhere?
16:22.07Paliwhat?
16:22.37kerioPali: unaligned memory access is trapped in old kernels, but more recent ones let that slide on arm 6+
16:25.58kerioPali: http://git.kernel.org/?p=linux/kernel/git/torvalds/linux.git;a=commit;h=8428e84d42179c2a00f5f6450866e70d802d1d05
16:31.45*** join/#maemo-ssu Milhouse (~irc_milho@Maemo/community/contributor/Milhouse)
16:32.34*** join/#maemo-ssu M13 (~Miha@170.133-224-87.telenet.ru)
16:34.54*** join/#maemo-ssu M13 (~Miha@170.133-224-87.telenet.ru)
16:42.06*** join/#maemo-ssu CodenameStrike (~assassin@175.144.136.127)
16:42.54freemangordonwell, all of the videos i have here still play, camera works, the music plays, so I assume gstreamer-thumb is ok for the next update
16:44.54*** join/#maemo-ssu amiconn_ (amiconn@rockbox/developer/amiconn)
16:49.11kerioyay
16:49.13keriois it faster?
16:49.24kerioalso yay gcc4.7 for mp3
16:50.59freemangordonfaster? how to check that?
16:51.10freemangordonunfortunately mp3 is nokia :(
16:51.23kerio...wtf
16:51.23keriowhy?
16:51.27kerioit's fukken gstreamer
16:51.30freemangordonno idea
16:51.37keriowell fix it
16:51.49freemangordonlibgstnokiamp3dec.so
16:52.43freemangordonpackage gstreamer0.10-nokia-speech
16:52.52keriothat's bullshit and we both know it
16:52.56keriofix it
16:53.00kerioplease ;_;
16:54.34freemangordonunfortunately the parts that will benefit the most from using gcc4.7.2 are closed source - pulseaudio nokia plugin and gst mp3 decoder :(
16:55.15freemangordoncurse Nokia for that
16:58.38freemangordonthere are FOSS mp3 decoders, but those are out of the scope of CSSU-thumb :P
16:59.52*** join/#maemo-ssu m0shbear (~moshbear@servbox.moshbear.net)
17:00.25freemangordonbtw did it come clear why tracker update screwed flac and ogg?
17:00.30tadzikfreemangordon: why out of scope?
17:00.56freemangordontadzik: as I am not going to reinvent maemo5
17:01.28tadzikI see
17:01.55freemangordonfeel free to pester merlin1991 and chem|st to add foss mp3 codec replacement in CSSU-T/S, -thumb will follow shortly :)
17:02.51ShadowJKiirc kulve already tested ffmpeg's mp3 and vorbis decoder back in 09 or 10, and performance was good
17:03.39freemangordonShadowJK: hmm, I remember something about that.
17:04.07freemangordonAnd I suspect it will be much better when optimized with gcc 4.7.2
17:04.24freemangordonbut please, understand that I am the only one working on that
17:04.51ShadowJKAlso ffmpeg/libav itself will have improved since then
17:06.49ShadowJKI'd also claim that current perf is "good enough"
17:07.26ShadowJKand that the difference in battery life measured back then came  probably mostly from different wakeup behaviour rather than cpuseconds used
17:13.48kerioShadowJK: idk, there's a *big* difference between gcc 4.2 and 4.7-linaro
17:13.59keriooptimizers are black magic
17:16.16freemangordonShadowJK: did you see the comparison in openssl benchmarks?
17:17.07freemangordonthe one compiled with gcc4.7.2 is about 2 times faster for some algos
17:18.17freemangordonbtw does anyone know if sshd uses openssl?
17:18.34kerioi'd guess it does
17:18.47kerioactually no, maybe not
17:19.06kerioyes it does, it uses libcrypto
17:19.45freemangordonas i've never seen winscp copying a file with > 1MB/s speed. until an hour ago :D
17:20.16m0shbearkerio: less black magic than tons of program analysis math
17:20.35m0shbearthey're becoming as advanced as decompilers, if not at that level, imo
17:21.59freemangordonShadowJK: http://pastebin.com/Bpdt8nr7
17:24.18ShadowJKis that in cssu thumb?
17:24.24freemangordonyes
17:24.47freemangordondevice not overclocked
17:24.51keriofreemangordon: i wonder how much of that is thumb and how much is gcc 4.7
17:25.05freemangordonit is deffinitely gcc
17:25.45freemangordonthough AFAIK thumb2 allows compiler to schedule more effectively
17:28.35freemangordonwonders is it a good idea to recompile the kernel with gcc4.7
17:29.17*** join/#maemo-ssu NIN101 (~NIN@p5DD281D1.dip0.t-ipconnect.de)
17:31.29keriodo it
17:31.30kerioi dare you
17:32.30freemangordonhehe. I already used 4.6.2 with meego kernel back in the days I was playing with ubuntu 12.04. it was ok
17:36.12ShadowJKis thumb kernel possible?
17:38.00*** join/#maemo-ssu sixwheeledbeast (~sixwheele@host-78-148-129-185.as13285.net)
17:39.16freemangordonShadowJK: I didn't mean thumb-compiled, but gcc4.7 compiled
17:39.37freemangordonAFAIK our kernel is too old to be compiled with thumb2
17:40.23freemangordontoo many patches are missing
17:40.29ShadowJKYeah I know you meant normal
17:40.49freemangordonotherwise even meego kernel can be thumb2-compiled
17:40.57freemangordon(iirc)
17:43.02freemangordonthough I cannot imagine what would benefit from gcc4.7 besides filesystems(maybe).
17:44.19kerio...everything?
17:44.19kerioidk
17:51.34ShadowJKThat's why I asked about thumb, smaller codesize could benefit cpu cache etc :)
18:04.08freemangordonShadowJK: yeah, but it is too much of a backporting :(
18:21.22keriofreemangordon: hm, why won't it compile with -mthumb?
18:26.37ShadowJKnoticed sys cpu use was 20% playing video, but i guess that's mostly shuffling data back and forth
18:27.29freemangordonkerio: it will compile, but it won't work
18:27.33keriowhy? :o
18:28.38freemangordonbecause of things like signal/exception handling and generation. MPU/FPU context, etc,etc
18:30.07keriohm
18:30.21keriowell, it's time we got an up-to-date kernel anyway ;)
18:30.45freemangordonwell, I agree, when are you going to release it?
18:31.06keriotomorrow
18:31.09kerioif nothing else comes up
18:31.18freemangordongood, i'll wait for it
18:32.13freemangordonShadowJK: it is your DSP that decodes it then :)
18:34.26freemangordonthough most of those those 20% are most probably used by PA, there is only one large memcpy happening - from GST_BUFFER to framebuffer
18:46.55ShadowJKi didnt think pa used any measurable amounts of sys
18:47.40ShadowJKwonders if a2dp encoder is closed sauce
18:48.14kerioShadowJK: it is
18:48.19kerioin the sense that you can't recompile a physical chip
18:48.57keriountil 3d printers become much better, at least
19:23.11ShadowJKuh
19:23.35ShadowJKI thought the a2dp encoder was software
19:23.35ShadowJKsbc or whatever the codec is called
19:25.32keriothere's not much change in cpu usage between audio through cordful headphones and bt headphones
19:27.17ShadowJKoh? I've got like 40% use with bt headphones
19:32.56*** join/#maemo-ssu dhbiker (~dhbiker@95.87.145.172)
20:03.38*** join/#maemo-ssu Woody14619 (~Woody@66-162-186-66.static.twtelecom.net)
20:03.38*** join/#maemo-ssu Woody14619 (~Woody@Maemo/Community/council/Woody14619)
20:16.08*** join/#maemo-ssu wmarone (~wmarone@c-67-174-151-253.hsd1.ca.comcast.net)
20:38.41*** join/#maemo-ssu MrPingu (~MrPingu@86.92.226.97)
20:49.43*** join/#maemo-ssu MrPingu (~MrPingu@86.92.226.97)
20:58.00*** join/#maemo-ssu nox- (noident@freebsd/developer/nox)
21:00.22*** join/#maemo-ssu MrPingu (~MrPingu@86.92.226.97)
21:05.39*** join/#maemo-ssu _rd (~rd@p57B494BD.dip0.t-ipconnect.de)
21:42.54*** join/#maemo-ssu int_ua (~int_ua@ip-602f.proline.net.ua)
21:56.28*** join/#maemo-ssu dhbiker (1000@95.87.145.172)
23:12.26*** part/#maemo-ssu sixwheeledbeast (~sixwheele@host-78-148-129-185.as13285.net)

Generated by irclog2html.pl Modified by Tim Riker to work with infobot.