Previous day | Jump to hour: 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 | Next day

Seconds: Show Hide | Joins: Show Hide | View raw
Font: Serif Sans-Serif Monospace | Size: Small Medium Large

Click in the nick column to highlight everything a person has said.
The Logo icon identifies that the person is a core developer (has commit access).

#rockbox log for 2009-04-22

00:01:19 Join tree [0] (
00:01:45 Join robin0800 [0] (
00:02:17 Nick tree is now known as trisiak (
00:02:23 Nick fxb is now known as fxb__ (
00:02:27 Join stripwax [0] (
00:05:16 Join dfkt_ [0] (
00:05:51 Quit BXCracer (Remote closed the connection)
00:05:57 Quit dfkt (Nick collision from services.)
00:06:02 Quit evilnick_7 (" hometime")
00:06:26 Nick dfkt_ is now known as dfkt (
00:07:15 Quit stripwax (Client Quit)
00:07:49 Part mcuelenaere ("Leaving.")
00:09:08 Quit bertrik ("Sleeo")
00:12:00 Join dfkt_ [0] (
00:12:27 Quit dfkt (Nick collision from services.)
00:13:35 Nick dfkt_ is now known as dfkt (
00:14:23 Quit MT (Remote closed the connection)
00:15:16 Join robin0800_ [0] (
00:16:42 Join stripwax [0] (
00:18:24 Quit robin0800_ (Remote closed the connection)
00:20:44 Join robin0800_ [0] (
00:24:28 Quit stripwax ("")
00:26:16 Quit flydutch ("/* empty */")
00:26:54 Quit blithe ("Lost terminal")
00:27:05 Join blithe [0] (
00:30:00 Quit robin0800 (Read error: 113 (No route to host))
00:30:01 Quit robin0800_ ("No Ping reply in 90 seconds.")
00:30:35 Join robin0800 [0] (
00:33:50***Saving seen data "./dancer.seen"
00:34:44 Quit perrikwp (" ajax IRC Client")
00:35:04 Join perrikwp [0] (i=18ac0c41@gateway/web/ajax/
00:37:23 Quit robin0800 ("No Ping reply in 90 seconds.")
00:37:27 Quit PaulJam (".")
00:37:55 Join robin0800 [0] (
00:40:03 Quit EternalRains (Read error: 104 (Connection reset by peer))
00:45:14kugelFlynDice: did you upload?
00:59:48 Quit gregzx ("ChatZilla 0.9.84 [Firefox 3.0.8/2009032609]")
01:02:47 Join kako [0] (
01:03:55tmztkugel: do you know the fix for the sysfont.h problem on fuze? (current svn)
01:04:20kugelwhat problem?
01:04:24 Join barrywardell [0] (n=barrywar@
01:04:41tmzttmzt@shini:~/src/rockbox/build$ PATH=/usr/local/arm-elf/bin:$PATH make
01:04:41tmztCC bootloader/common.c
01:04:41tmztIn file included from /home/tmzt/src/rockbox/bootloader/common.c:23:
01:04:42DBUGEnqueued KICK tmzt
01:04:42tmzt/home/tmzt/src/rockbox/firmware/export/font.h:33:21: error: sysfont.h: No such file or directory
01:04:47kugel there doesn't seem to be a problem
01:07:35tmztwhat do you recommend to diagnose this then? I can give you the full output in paste if you need it
01:07:49linuxstbtmzt: You should change your PATH before running configure, not before running make.
01:08:03tmztI'll try that
01:08:44tmztoutput is shorter but the error is still there
01:10:24tmztis make building something that wouldn't be tested with the automated tools?
01:10:29 Quit kako ("using sirc version 2.211+KSIRC/1.3.12")
01:10:53kugeltmzt: can't you just mess with the PATH before messing with confiure and make?
01:11:09tmztI'm sorry?
01:11:12linuxstbLook at the generated Makefile - is it including the correct paths to arm-elf-gcc etc?
01:11:16tmztI ran configure with the correct patch
01:11:37tmztexport CC=/usr/local/arm-elf/bin/arm-elf-gcc
01:11:46 Join CaptainKewl [0] (
01:11:47tmztfrom Makefile
01:12:17linuxstbAnd all the others look OK? Include the HOST versions?
01:12:42tmztyes, those are 64bit though
01:12:46tmztor multilib
01:13:35linuxstbDo a "make clean", then try "make V=1 &> logfile" to get more verbose errors.
01:14:50 Quit faemir ("Leaving")
01:16:27 Quit barrywardell ()
01:18:48 Quit dfkt ("-= SysReset 2.53=- Ph'nglui mglw'nafh Cthulhu R'lyeh wgah'nagl fhtagn.")
01:22:14 Join barrywardell [0] (n=barrywar@
01:23:30tmztthe build is taking a long time but hasn't crashed
01:23:47tmztmake reconf doesn't rebuild Makefile? you have to do make clean?
01:24:18 Quit barrywardell (Client Quit)
01:33:31linuxstbIt should rebuild the Makefile - testing if the Makefile is older than configure is what prompts you to run "make reconf".
01:38:14 Quit daurnimator (Remote closed the connection)
01:39:07 Quit robin0800 (Read error: 104 (Connection reset by peer))
01:39:22tmztah, it's been done. I was trying to figure out why it was stuck on sudoko
01:39:55tmztand I have a rockbox.bin
01:40:03tmztand rockbox.sansa
01:40:28 Quit Thundercloud (Remote closed the connection)
01:41:32 Join robin0800 [0] (
01:44:59 Quit DataGhost (Nick collision from services.)
01:45:07 Join DataGhost [0] (i=dataghos@unaffiliated/dataghost)
01:47:35 Join _lifeless [0] (n=lifeless@
01:48:53 Quit apo (Read error: 60 (Operation timed out))
01:51:09FlynDicekugel: mmu patch uploaded, lots of work left though....
01:52:52 Join apo [0] (
01:53:03 Quit amiconn (Nick collision from services.)
01:53:07 Join amiconn_ [50] (n=jens@rockbox/developer/amiconn)
01:53:14 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
01:53:22 Join pixelma_ [50] (n=pixelma@rockbox/staff/pixelma)
01:53:22 Quit pixelma (Nick collision from services.)
01:53:44 Nick pixelma_ is now known as pixelma (n=pixelma@rockbox/staff/pixelma)
01:57:40 Quit dreamlayers ("back later")
01:59:57 Nick bubsy is now known as [8bit]bubsy (i=Bubsy@unaffiliated/bubsy)
02:01:06 Quit rwcr (Read error: 113 (No route to host))
02:01:40 Quit tchan ("WeeChat 0.2.7-dev")
02:02:42 Join tchan [0] (n=tchan@lunar-linux/developer/tchan)
02:05:15 Quit KBH (Read error: 104 (Connection reset by peer))
02:05:21 Join HBK [0] (
02:08:41 Part toffe82
02:09:05 Join rwcr [0] (
02:18:36 Join kkurbjun [0] (n=kkurbjun@rockbox/developer/kkurbjun)
02:18:50 Join Blue_Dude [0] (
02:20:08Blue_DudeHi. I'm considering implementing a new feature and wanted to just jump in and test the water here. I am new to collaborative development.
02:20:50 Quit robin0800 ("No Ping reply in 90 seconds.")
02:21:21 Join robin0800 [0] (
02:22:00kkurbjunBlue_Dude: not too many devs are active at this time of the day
02:22:25Blue_DudeThat's OK. When's a good time to jump in?
02:22:36kkurbjungenerally if you have questions on the development flow and ask general questions people will try and pitch in
02:23:07kkurbjunthe mornings in the America's are usually pretty active
02:23:51kkurbjunI guess it would be late in the evening in eastern europe
02:24:04kkurbjunworry, western that is
02:25:29tmztis only charging support by usb on ams?
02:26:02Blue_DudeI'm on EDT, so that works for me. I'll check back in the AM. Thanks!
02:26:17kkurbjunbut usually unless you find someone else that is interested in working on the same feature people work on their own code and if they run into questions they will just throw them out, and then ask for reviews of the code before committing it depending on what parts of rockbox you are touching.
02:26:30kkurbjunno problem
02:27:34Blue_DudeThat's pretty much what I had in mind. I was considering a dynamic compressor/limiter feature. It's not especially invasive but I would need to find my way around the code.
02:27:58Unhelpfulhrm... another oddity in our jpeg decoder, it seems to key subsampling parameters on the first component... even though it's a per-component option?
02:28:26UnhelpfulBlue_Dude: we have a DSP framework and several other options (replaygain, EQ) that use it.
02:29:36Blue_DudeI didn't see any combination of features that would pre-amp audio without clipping.
02:31:40Unhelpfulin the replaygain case, the tag includes the peak sample value, so that the gain level can be limited appropriately
02:33:52***Saving seen data "./dancer.seen"
02:34:12 Join xenonsniper [0] (
02:34:16Blue_DudeWhat I had in mind was to pre-amp the signal by an user-selectable amount (without regard to metadata), then hard limit the output to prevent clipping. The idea is to permit playback in noisy environments.
02:34:36xenonsniperhey all, i just heard about Rockbox and i was wondering if there's any Zune support
02:34:56UnhelpfulBlue_Dude: by "hard limit", what exactly do you mean?
02:37:36 Quit kugel ("ChatZilla 0.9.84 [Firefox 3.0.8/2009032609]")
02:38:03Unhelpfuli think the best bet for dynamic volume control would be to take either the plain RMS average of the signal, or evaluate it via an equal-loudness function, feed that to a moving average, and use that to set an amp value (using the RG data to stop short of clipping, if needed)
02:38:07Blue_DudeA hard limiter ramps down the audio to prevent clipping. An intelligent limiter should only lower volume when approaching a peak and restore the volume afterward. If the signal is pre-amped, you can get an on-the-fly compressor effect.
02:39:28 Quit itcheg (" ajax IRC Client")
02:39:57Unhelpfulsince we do DSP at greater-than-output depth, actually, you shouldn't have a problem with *actual* clipping during DSP, and then you can apply whatever conversion you want to get large values low enough to prevent them clipping.
02:41:43Blue_DudeExactly. I'm not worried about a data overrun.
02:41:45Unhelpfuli think the moving-average idea is best to have in front of a peak limiter, though, because limiting just the peaks == distortion
02:44:53Blue_DudeI had in mind a look ahead buffer to identify peaks, then ramp down the volume to meet the peak, followed by a slower release back to normal volume. That would keep distortion to a minimum (there will always be some). I'm not sure what an algorithm based on RMS would sound like, or how to tune it.
02:45:26 Join jmillikin [0] (
02:47:52Unhelpfuli would favor equal loudness over a plain RMS, since it takes into account the varying sensitivity of the ear with regard to frequency. and it would sound like the volume going up and down... either obviously, or subtly, depending on how long your average is.
02:48:34 Join dmb [0] (n=dmb@unaffiliated/dmb)
02:50:49 Join claydoh [0] (n=quassel@ubuntu/member/claydoh)
02:51:26Blue_DudeI've been playing around with wav editor software. The attack time (for ramping down the volume) is only about 7ms. So the clamp down happens pretty fast. The release time is a bit longer, about 100ms. So the adjustments are pretty fast.
02:53:08Unhelpfulyou're also basically using peak amplitude as your measure of loudness. peak is not perceived loudness.
02:55:15Blue_DudeThat's true. But I'm not trying to limit loudness as such, just prevent digital clipping at the DAC, and in a way that prevents square waves from appearing there. I'm thinking more of protecting the equipment here.
02:58:10Unhelpfuli thought this was about range compression for noisy environments. we're talking about two different things here, and i think that a well-designed clipping limiter is a good idea, but i would *not* rely on that+preamp to take care of noisy environments.
02:59:53Blue_DudeYep, two different things, but my ignorance of dynamic range compression is boundless. But a quick acting intelligent volume manager might be in reach.
03:00:57Unhelpfuli would still say an equal-loudness- or rms-driven amp, and a limiter to make sure *that* doesn't clip
03:03:33Blue_DudeWhat are the processing implications of that? Since it would have to run a process every time a sample is played, is there enough headroom to allow it?
03:06:35Unhelpfuli would actually suggest processing a chunk of samples at some regular interval - maybe with a small fuzz factor on the interval so it doesn't screw up on periodic signals, but that's probably overdesign. you don't need to evaluate loudness per-sample, it's certainly not perceived that way.
03:07:21Blue_DudeNot loudness per sample, but ... OK never mind I see where you're going with it.
03:08:20Blue_DudeCheck every, say, 50 samples and see what's coming down the pike for the next 300 samples or so, do an analysis and start limiting as necessary.
03:10:08Blue_DudeThat's not a bad way to go for limiting. But if you aim to increase loudness, if not pre-amping, then what?
03:11:22 Quit DataGhost (Nick collision from services.)
03:11:24Unhelpfulno, i mean, don't even bother evaluating loudness most of the time. take a FFT of, say, every fifth block of 2048 samples. my point is that a *fixed* pre-amp is not the way to manage dynamic range
03:11:30 Join DataGhost [0] (i=dataghos@unaffiliated/dataghost)
03:11:37 Join Xerion_ [0] (
03:11:48Blue_Dude(I was thinking of the cheap, easy route, not a full-blown DSP plugin. But I'm all ears.)
03:13:02 Join midijunkie [0] (
03:13:14kkurbjunxenonsniper: there is no zune support for rockbox right now.
03:13:30 Quit Xerion (Read error: 104 (Connection reset by peer))
03:13:30 Nick Xerion_ is now known as Xerion (
03:14:21xenonsniperkkurbjun: thanks, i read a couple of webpages, the wiki for rockbox, and some other stuff but i didn't find anything. good to know though
03:14:54kkurbjunno problem
03:15:28kkurbjunthe gigabeat S is close to the Zune in terms of hardware, but the zune uses alot of the hardware features to lock down what code is run on it.
03:22:46Blue_DudeThanks very much, Unhelpful. You were not unhelpful! I'll come slinking back in the morning.
03:30:19 Quit midijunkie ("?(???~~)?")
03:33:51 Join Traveler9 [0] (
03:34:03Traveler9rockbox is able to use fgets() method, correct?
03:34:55 Join EternalRains [0] (
03:38:36 Quit tvelocity (Remote closed the connection)
03:41:51 Quit Blue_Dude ("ChatZilla 0.9.84 [Firefox 3.0.8/2009032609]")
03:41:52 Nick Traveler9 is now known as traverser (
03:42:55 Quit CaptainKewl ("( :: NoNameScript 4.02 :: )")
03:47:28traverseror is fgetc a method able to be used in rockbox?
03:51:14 Part claydoh (" - Chat comfortably. Anywhere.")
03:53:08 Quit xenonsniper ("ChatZilla 0.9.84 [Firefox 3.0.8/2009032609]")
03:54:01 Quit timc (Remote closed the connection)
03:57:39 Quit EternalRains (Read error: 104 (Connection reset by peer))
03:59:13 Quit traverser ("Java user signed off")
04:06:54 Quit froggyman ("CGI:IRC (Ping timeout)")
04:18:17FlynDicecan someone tell me how to get enable test_codec
04:21:33 Join mc2739 [0] (
04:22:24mc2739FlynDice: I think you add test_codec.c to apps/plugins/SOURCES
04:22:45 Join JdGordonn [0] (n=irchon@
04:24:22 Quit JdGordonn (Client Quit)
04:31:08FlynDicemc2739: well I tried that, does it come up under applications like test_disk?
04:32:29mc2739FlynDice: did you look for it in the plugins folder?
04:32:57mc2739FlynDice: oops, not plugins, but rocks
04:33:56***Saving seen data "./dancer.seen"
04:34:24FlynDiceyes I found the file but when I added it to SOURCES it didn't seem to come up, I'll try again thanks
04:36:32mc2739FlynDice: no, I meant in the zip or on the device in the .rockbox/rocks folder or subfolders
04:40:08 Quit mc2739 ("ChatZilla 0.9.84 [Firefox 3.0.8/2009032609]")
04:47:11 Quit robin0800 ("No Ping reply in 90 seconds.")
04:47:46 Join robin0800 [0] (
05:01:54 Join hittudiv [0] (n=hittudiv@
05:02:20 Quit hittudiv (Client Quit)
05:03:22 Join fenugrec [0] (
05:07:06 Quit fenugrec (Client Quit)
05:18:02 Quit perrikwp (" ajax IRC Client")
05:18:14 Join perrikwp [0] (i=18ac0c41@gateway/web/ajax/
05:30:05 Quit Horschti ("Verlassend")
05:42:48 Join Trista266 [0] (
05:43:26 Quit evilnick (Read error: 104 (Connection reset by peer))
05:43:37 Join evilnick [0] (
05:43:59 Join kushalone [0] (n=kushal@
05:47:32 Quit _Auron_ (Read error: 104 (Connection reset by peer))
05:56:29 Quit Tristan (Read error: 110 (Connection timed out))
05:59:37 Quit Trista266 (Connection timed out)
06:02:09 Join krazykit [0] (n=kkit@
06:02:12 Quit krazykit (Client Quit)
06:02:31 Join _Tristan [0] (
06:04:15 Quit perrikwp (" ajax IRC Client")
06:07:09 Join dreamlayers [0] (
06:08:20 Quit jmillikin (Read error: 110 (Connection timed out))
06:08:22 Join perrikwp [0] (i=18ac0c41@gateway/web/ajax/
06:12:05 Join timc [0] (n=aoeu@
06:16:47 Quit parafin (Remote closed the connection)
06:17:26 Join robin0800_ [0] (
06:18:35 Join kachna|lappy [0] (
06:18:55 Join parafin [0] (
06:18:57 Quit evilnick (Read error: 113 (No route to host))
06:19:07 Join evilnick [0] (
06:19:16 Join Rob2222 [0] (
06:19:30 Quit pixelma ("No Ping reply in 90 seconds.")
06:19:33 Join pixelma [50] (n=pixelma@rockbox/staff/pixelma)
06:19:46 Quit Rob2223 (Read error: 104 (Connection reset by peer))
06:20:07 Join Xerion_ [0] (
06:21:01 Quit Xerion (Read error: 113 (No route to host))
06:21:11 Nick Xerion_ is now known as Xerion (
06:22:20 Quit robin0800 (Read error: 113 (No route to host))
06:27:18 Quit kachna (Connection timed out)
06:33:58***Saving seen data "./dancer.seen"
06:35:04 Quit Tuplanolla (Read error: 110 (Connection timed out))
06:35:17 Join Traveler0 [0] (
06:40:03Traveler0what would be the simplest way to read in a line from a text file? I'm hoping to put the line into a 100 cell char array.
06:40:15 Quit AndyI ()
06:49:13 Join _Auron_ [0] (
06:52:53 Join saratoga [0] (n=41becb3b@gateway/web/cgi-irc/
06:53:49saratogaTraveler0: take a look at apps/plugin.h
06:56:09Traveler0alright I'll take a look at that, thanks
06:56:13 Quit Traveler0 ("Java user signed off")
06:56:36 Join AndyI [0] (i=AndyI@
06:56:50 Quit gevaerts (Read error: 60 (Operation timed out))
06:59:13 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
07:00:36 Quit Unhelpful (Read error: 60 (Operation timed out))
07:00:41 Join strikerz911 [0] (
07:01:19 Quit timc (Remote closed the connection)
07:01:35strikerz911Hello there.
07:01:51strikerz911I could use some help here.
07:02:00strikerz911I am running rockbox on my ipod 5th generation
07:02:20strikerz911I have read the manual but I'm still not sure how to create a playlist.
07:02:41strikerz911Could anyone please help?
07:03:19strikerz911By the way, I am going to backup and delete all my songs off iTunes and just use the file browser to play mp3s
07:04:24lucentstrikerz911: I don't know how to create a playlist, but I think that you can use "select" in the file browser mode as a kind of "right click" to get an option menu
07:04:33lucentand that might have an option to add the track to a playlist?
07:04:35lucentjust a guess
07:05:11strikerz911Well I haven't actually used the file browser method, but I plan on doing so, but while reading the manual, I still don't get how I would create a playlist
07:05:33strikerz911I have so many songs, but I would like to divide them up into playlist to avoid the hassle of listening to them all as one big playlist
07:05:34 Join FOAD_ [0] (
07:06:14lucentstrikerz911: there is also the database option, built from metadata tags
07:06:26lucentstrikerz911: you'll have to read the manual -and- try it out to learn
07:06:37robin0800_strikerz911: Yes in the context menu you can choose insert or evan insert shuffeld
07:06:38 Quit saratoga ("CGI:IRC (EOF)")
07:08:07robin0800_strikerz911: You could also use the Database to brouse by Artist or Album etc
07:08:29strikerz911I would really prefer using the regular File Browser method to avoid using iTunes entirely.
07:08:50 Join BXCracer [0] (
07:08:52strikerz911I'll try creating a playlist on my computer using winamp and see if I can use it like that.
07:09:09robin0800_strikerz911: Both methods can be done by Rockbox
07:11:07 Quit AndyI (Read error: 60 (Operation timed out))
07:11:08 Quit BXCracer (Read error: 104 (Connection reset by peer))
07:12:58 Join einhirn [0] (
07:15:15 Quit dreamlayers ()
07:15:28 Join AndyI [0] (i=AndyI@
07:17:11 Quit tarbo (Success)
07:18:29 Join Unhelpful [0] (n=Militant@rockbox/developer/Unhelpful)
07:21:11 Quit FOAD (Read error: 110 (Connection timed out))
07:21:11 Nick FOAD_ is now known as FOAD (
07:22:16 Quit robin0800_ (Remote closed the connection)
07:33:45 Quit jeffdameth (Read error: 110 (Connection timed out))
07:35:07 Join evilnick1 [0] (
07:35:07 Quit evilnick (Read error: 104 (Connection reset by peer))
07:35:08 Quit evilnick1 (Read error: 104 (Connection reset by peer))
07:35:50 Join jeffdameth [0] (
07:36:56 Quit _Tristan (Read error: 104 (Connection reset by peer))
07:37:39 Join _Tristan [0] (
07:39:11 Join tarbo [0] (n=me@unaffiliated/tarbo)
07:51:34 Join BHSPitMonkey [0] (n=stephen@unaffiliated/bhspitmonkey)
07:54:47 Quit parafin (Read error: 110 (Connection timed out))
07:55:35 Quit intrados_ (Read error: 104 (Connection reset by peer))
08:01:03amiconnUnhelpful: If you have questions regarding our jpeg decoder, you could try to mail [IDC]Dragon. He wrote that decoder
08:03:45 Join robin0800 [0] (
08:04:20amiconnThe original decoder was Y only; colour decoding was added much later
08:07:43 Quit robin0800 (Remote closed the connection)
08:07:50 Quit Zarggg ()
08:09:02 Join Zarggg [0] (
08:12:31 Join ender` [0] (
08:21:45 Quit strikerz911 ()
08:25:12 Join parafin [0] (
08:26:15 Quit Unhelpful (Read error: 60 (Operation timed out))
08:27:10 Join Unhelpful [0] (n=Militant@rockbox/developer/Unhelpful)
08:34:02***Saving seen data "./dancer.seen"
08:37:21 Join evilnick [0] (
08:38:11 Quit BHSPitMonkey ("Ex-Chat")
08:42:12 Join Rob2223 [0] (
08:43:29 Join B4gder [241] (n=daniel@rockbox/developer/bagder)
08:44:00 Join petur [50] (n=petur@rockbox/developer/petur)
08:57:01 Join Zarggg_ [0] (
08:59:11 Quit Rob2222 (Connection timed out)
09:01:18 Quit Zarggg (Read error: 60 (Operation timed out))
09:12:53 Join Thundercloud [0] (
09:18:53 Quit Seed ("cu, Andre")
09:22:15 Join petur2 [50] (n=petur@rockbox/developer/petur)
09:22:36 Quit petur (Nick collision from services.)
09:22:42 Nick petur2 is now known as petur (n=petur@rockbox/developer/petur)
09:26:47 Join Dieterbe_ [0] (
09:34:37 Join l403 [0] (n=l@
09:38:38 Quit Dieterbe (Read error: 110 (Connection timed out))
09:39:11GodEaterhas the topic of customisable menus come up again recently much ?
09:41:20 Quit l403 (Read error: 113 (No route to host))
09:42:28CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
09:42:28*GodEater sees that the rockbox channel has become "Full on tumbleweed racing. No holds barred! Place your bets!"
09:43:21martian67pick the next join/part
09:43:23martian6720:1 odds
09:43:43GodEaterI'll put a tenner on it.
09:43:51GodEateroh wait
09:43:53GodEaterhe's here already
09:43:55*GodEater rubs his eyes
09:44:17 Quit Dieterbe_ (Client Quit)
09:44:22martian67you lose
09:44:31martian67no rigging it either ;)
09:44:53GodEaterI wouldn't stoop that low :)
09:45:17GodEateralthough technically a "quit" isn't a "part"
09:51:40 Join l403 [0] (n=l@
09:52:04 Join Icingdeath [0] (n=d584ab1b@gateway/web/cgi-irc/
09:52:29 Quit Icingdeath (Client Quit)
09:55:13 Quit Thundercloud (Remote closed the connection)
10:13:32 Join pyro_maniac [0] (
10:16:14 Join barrywardell [0] (n=barrywar@
10:16:15 Join Sedgewick [0] (n=Sedgewic@
10:29:49 Quit barrywardell ()
10:34:04***Saving seen data "./dancer.seen"
10:46:18 Join petur2 [50] (n=petur@rockbox/developer/petur)
10:48:58 Join Zagor [242] (n=bjorn@rockbox/developer/Zagor)
10:52:44 Quit petur (Read error: 60 (Operation timed out))
10:54:17 Nick petur2 is now known as petur (n=petur@rockbox/developer/petur)
11:06:46 Join barrywardell [0] (n=barry@
11:09:36ajbDoes anyone claim to undertand the mp4 metadata code?
11:09:55ajbOr the structure of MP4 metadata?
11:16:53fuzzie<-, but i have a lecture now :|
11:17:01 Join daurnimator [0] (n=daurnima@unaffiliated/daurnimator)
11:22:26ajbfuzzie: OK, I shall ping you later if thats OK
11:23:07ajbDoes anyone actually have a copy of documentation describing the format of MP4 files?
11:24:38*ajb finds the draft on the ISO website
11:30:55 Join timc [0] (n=aoeu@
11:32:10 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
11:39:46 Quit l403 (Remote closed the connection)
11:42:24 Join EternalRains [0] (
11:50:18 Join codesquid [0] (n=quassel@kong.kawo1.RWTH-Aachen.DE)
11:52:48 Join l403 [0] (n=l@
12:04:33 Quit kushalone ("Leaving. I cannot promise to be back but most likely will.")
12:05:26 Join kushal_12_27_200 [0] (n=kushal@
12:12:54 Join daurn [0] (n=daurnima@unaffiliated/daurnimator)
12:13:20 Join Lynx_ [0] (n=Lynx@
12:18:49 Quit daurnimator (Read error: 60 (Operation timed out))
12:20:33 Nick fxb__ is now known as fxb (
12:26:03 Join tvelocity [0] (
12:34:06***Saving seen data "./dancer.seen"
12:42:46 Quit EternalRains (Read error: 104 (Connection reset by peer))
12:49:55*ajb gets confused by the seeking and skipping between read_mp4_atom and read_mp4_container
13:02:39 Quit kushal_12_27_200 ("Leaving. I cannot promise to be back but most likely will.")
13:11:22ajbI assume id3->filesize represents the sum of the size of the audio bitsream and not the size of the file on disk
13:19:12 Quit Sedgewick ("off")
13:26:49linuxstbajb: I think that's used in a different way by different codecs. Some may reduce it (to remove trailing tags). You can see how it's set for mp4 in the metadata parsing code - apps/metadata/
13:27:20ajblinuxstb: Yeah, I'm currently digging through that.
13:27:51ajblinuxstb: It seems my dodgy m4a breaks the "mdat is the last tag RB can handle" case
13:28:42linuxstbYes, I think there are known problems if m4a files are not following the expected structure - i.e. that all metadata is at the start of the file.
13:28:57linuxstb"known problems" = "we don't support such files"...
13:29:59ajblinuxstb: I don't think the spec says it has to be..., I now have the metadata displaying properly after a few fixes. However it won't play (I assume because the codec has issues as well?)
13:31:13linuxstbajb: I don't really know the details. Lear is the person who has worked most on our mp4 parser. I just did the very first version, which I copied from the open source ALAC decoder, without really understanding it too much.
13:35:10ajblinuxstb: I shall sumbit a WIP patch to Flyspray later on for people to comment on.
13:37:10linuxstbajb: OK. Hopefully Lear will spot that.
13:50:17 Join Tuplanolla [0] (n=jani@unaffiliated/tuplanolla)
13:55:05 Join petur2 [50] (n=petur@rockbox/developer/petur)
13:55:26 Quit petur (Nick collision from services.)
13:55:30 Nick petur2 is now known as petur (n=petur@rockbox/developer/petur)
14:01:20 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
14:06:46 Join funman [0] (i=56421242@gateway/web/ajax/
14:07:15funmanFlynDice: to use test_codec, you must go through the context menu of a music file (long press select, open with, test_codec)
14:08:05funmanFlynDice: also take care that the file length must be less or equal than the audiobuffer size (~256kb on Clip, but with 8MB of RAM 5MB might fit)
14:08:41 Quit funman (Client Quit)
14:23:29*ajb sumbits #FS10160 - Clean up MP4 metadata parsing to handle files with tags after 'mdat' atom
14:24:10ajbLear: Could you take a look at FS10160 when you get a chance?
14:28:13 Join PyroBor [0] (
14:32:34 Join LambdaCalculus37 [0] (i=44a0430d@rockbox/staff/LambdaCalculus37)
14:34:09***Saving seen data "./dancer.seen"
14:38:12 Quit l403 (Remote closed the connection)
14:43:01 Join kugel [0] (n=kugel@rockbox/developer/kugel)
14:53:48 Quit FlynDice (Remote closed the connection)
14:56:36ajbNow onto the second problem
14:56:53ajbWhen I select the tune it starts paused but there isn't any indication of what failed
14:57:13ajbThe last log output says: "Looking for album art for /03 - This House Is Not for Sale.m4a" and that is it
14:57:29Lloreanajb: What device is this on?
14:57:46ajbipod 5g simulator
14:59:57 Quit B4gder ("It is time to say moo")
15:01:45kugelZagor: did you already send the fuze?
15:04:29Zagorkugel: sorry, still not. I forgot yesterday.
15:10:36 Join evilnick_7 [0] (i=0c140464@gateway/web/ajax/
15:13:12 Join archivator [0] (n=archivat@
15:14:04 Quit archivator (Client Quit)
15:15:34LambdaCalculus37Speaking of sending... gotta ask saratoga if he's sent out the Gigabeat T yet.
15:16:25 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
15:19:01 Join RoC_MasterMind [0] (
15:21:41mcuelenaereis there already some kind of script in Rockbox?
15:22:12kugelmcuelenaere: yes, the release script. I didn't figure how it works though
15:22:12Lloreanmcuelenaere: Isn't there at least the release build script?
15:22:24*mcuelenaere looks
15:23:49mcuelenaerehmm that also seems to build the manuals
15:25:30 Quit kugel ("ChatZilla 0.9.84-rdmsoft [XULRunner]")
15:26:34mcuelenaerekugel: 'mcuelenaere@debian:~/rockbox$ tools/release/ -v ipodcolor' seems to work
15:26:43mcuelenaereit really slowed down my system though
15:28:07 Join BeholdMyGlory [0] (
15:29:05 Quit tchan ("WeeChat 0.2.7-dev")
15:30:27 Join tchan [0] (n=tchan@lunar-linux/developer/tchan)
15:35:06BeholdMyGloryI know you probably get this a lot, but is an ipod classic port on the way/started/planned? Or should I not get my hopes up at all? (I _have_ done a bit of googling, but I can't seem to find anything from 2009...)
15:35:34GodEaterBeholdMyGlory: no-one is working on one to our knowledge
15:38:37LloreanBeholdMyGlory: Generally speaking, if someone's working on something they'll post updates. So a lack of recent news on something should be taken to mean "there's no progress to report at this time" in the future as well.
15:38:42 Quit gevaerts (Nick collision from services.)
15:39:09 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
15:40:46linuxstbBeholdMyGlory: The only activity on hacking the new ipods that I know about is the "linux4nano" project - that sort of covers the Classic as well (similar hardware). If they make progress, then I'm sure activity will quickly start on a Rockbox port.
15:49:40 Join Blue_Dude [0] (
15:51:47Blue_DudeI'm considering adding a new feature. Any devs on?
15:52:25LloreanBlue_Dude: Did you need specific hep?
15:52:27Lloreanhelp even
15:53:23Blue_DudeSpecific? No. Just trying to get a feel for whether it's a good idea.
15:55:06Blue_DudeLooking at a kind of compressor/limiter feature to help with listening in noisy environments. I'm new to collaborative development so I came here.
15:55:29*mcuelenaere wonders whether it is possible to build a specific plugin instead of a whole build
15:55:54mcuelenaereI already found 'make -j4 /home/mcuelenaere/rockbox/build4/apps/plugins/lua.rock', but as this is an overlay it doesn't build all the required *.c files
16:00:00LloreanBlue_Dude: Weren't you in here asking basically the same thing yesterday though?
16:00:10LloreanIf you think it's a good idea, you should write a patch for consideration.
16:00:34Blue_DudeYes indeed. But was told the devs weren't in.
16:01:03GodEaterwonder who by ?
16:01:06GodEatermost of them are here all the time
16:02:02evilnick_7GodEater: This was late at night US-time so there were very few active users
16:04:03GodEateryes, but the important ones read the logs ;)
16:04:13Blue_DudeI do think it's a good idea, but if I'm the only one, I probably won't write the patch.
16:05:21Blue_DudeOr am I putting the cart before the horse here?
16:05:55gevaertsBlue_Dude: I think a well-working compressor would be nice to have
16:09:53LloreanBlue_Dude: Before an idea can really be judged wholly, it needs to be working.
16:10:09Blue_DudeFair enough.
16:10:09LloreanYou can't judge the usefulness relative to cost on theory alone. :-P
16:13:21LloreanBlue_Dude: Most common DSP functions are likely to be considered (I think). Especially those with a practical use, like a compressor.
16:15:35Blue_DudeThis kind of gets back to a conversation last night. I don't know all that much at all about DSP design, but I could build an intelligent volume control. It's not the same as a real compressor, but it'll help.
16:16:20LloreanI think you'd want a real compressor if you really wanted it to be considered.
16:26:58 Join kugel [0] (n=kugel@rockbox/developer/kugel)
16:30:50 Join l403 [0] (n=l@
16:31:27CIA-43New commit by mcuelenaere (r20775): Add exit() implementation for plugins
16:31:28 Join evilnick_230 [0] (i=0c140464@gateway/web/ajax/
16:32:55 Quit l403 (Read error: 104 (Connection reset by peer))
16:33:41 Join toffe82 [0] (n=chatzill@
16:34:13***Saving seen data "./dancer.seen"
16:35:09 Join l403 [0] (n=l@
16:40:58 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
16:43:00 Quit Blue_Dude ("ChatZilla 0.9.84 [Firefox 3.0.8/2009032609]")
16:48:10CIA-43New commit by mcuelenaere (r20776): Fix yellow
16:52:34at0mcompressor, or even limiter would be nice (cheaper on DSP normally)
16:54:31at0mlove to go to bed with some podcast, and the jingles just are jumping out too much sometimes. for those annoyances it would be fantastic
16:57:13 Quit lucent (Read error: 60 (Operation timed out))
16:58:43 Join wpyh [0] (n=william@
17:00:46 Quit Zagor ("Don't panic")
17:02:42 Quit codesquid (Remote closed the connection)
17:02:49 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
17:04:29CIA-43New commit by mcuelenaere (r20777): Fix red
17:05:41 Join archivator [0] (n=archivat@
17:09:36 Join Ubuntuxer [0] (
17:10:44 Join petur2 [50] (n=petur@rockbox/developer/petur)
17:11:03 Quit petur (Nick collision from services.)
17:11:05 Nick petur2 is now known as petur (n=petur@rockbox/developer/petur)
17:18:40 Join Lynx [0] (
17:19:03 Nick Lynx is now known as Guest5885 (
17:25:08 Quit Ubuntuxer ("Leaving.")
17:26:14 Join cornwall [0] (
17:27:14cornwallHello, could someone please help me apply a patch? I'm unsure of how to do so
17:27:22cornwallif someone is here, of course :P
17:27:36Lloreancornwall: Do you have a development environment set up?
17:27:48cornwalloh, hello again
17:27:52cornwallwhat do you mean?
17:28:06Lloreanhave you installed the tools and compilers necessary to compile Rockbox?
17:28:11 Join CaptainKwel [0] (i=2669ecc2@gateway/web/ajax/
17:28:27cornwalloi, nope, I haven't
17:28:37Lloreancornwall: You should see the SimpleGuideToCompiling wiki page then
17:28:43cornwallcool, thank you
17:28:44LloreanPatches are applied to the source code, which you then must compile
17:29:01cornwallthanks a lot!
17:30:01 Quit cornwall (Client Quit)
17:31:02 Join saratoga [0] (i=9803c6dd@gateway/web/ajax/
17:31:43saratogaBagder: my network's admin is unhappy with me due to a large number of SSH session connect attempts from, any idea why that might be?
17:32:17saratogaobviously for the build server, but I don't see why it should connect repeatidly
17:35:24 Quit Lynx_ (Connection timed out)
17:35:24 Nick Guest5885 is now known as Lynx_ (
17:41:16linuxstbsaratoga: Are these successful or unsuccessful connection attempts?
17:42:12 Quit petur ("work->...")
17:42:35saratogalinuxstb: thats what the admin sent me when I complained about being black listed
17:43:27wpyhI'm wondering if it's feasible to have one single build for the 5g and 5.5g ipods
17:43:28 Join funman [0] (i=53c2a0b9@rockbox/developer/funman)
17:43:41saratogaI think we already do
17:43:53GodEaterwe absolutely do
17:43:53wpyhsaratoga: huh?
17:44:02GodEaterthere's no difference between them at all
17:44:11linuxstbsaratoga: So no indication of what time period those 233 hits were in?
17:44:16wpyhI still get asked to select 32MB and 64MB builds when running configure
17:44:27GodEaterwpyh: that's nothing to do with 5G vs 5.5G
17:44:36saratogalinuxstb: no, but I regularly connect from home repeatidly without being blacklisted, so it couldn't have been months
17:44:41GodEaterwpyh: that's to do with whether or not you have a 30GB or 60/80GB ipod
17:44:49GodEaterwhich have different amounts of memory
17:44:51wpyhum... ok, I didn't ask the proper question...
17:44:59saratogayou did not
17:45:04GodEaterand there's no way to have a single build for those
17:45:11wpyhGodEater: you mean the 60GB ipod 5g has 64MB of RAM?
17:45:16GodEaterwpyh: correct
17:45:18oboAccording to logrotate, I get anywhere between 25 and 175 SSH connections per day from Rockbox
17:45:37oboand mine isn't the quickest build server out there...
17:45:38saratogawhy is it so many?
17:45:46wpyhI was under the impression that only the 80GB ipods have 64MB of RAM
17:45:48linuxstbsaratoga: But I guess it's easy enough to provide them with the whitelisting information - they even know it already...
17:45:50oboone to svn up, one to build, one to sftp?
17:45:55GodEaterwpyh: you would be under the wrong impression then
17:46:00wpyhah, OK
17:46:30saratogalinuxstb: well they want to know the IP of all the machines I access from
17:46:44saratogawhich is problematic since I have many and often use DHCP via wireless
17:46:53linuxstbsaratoga: I thought it was the other way around - the incoming traffic?
17:47:01 Quit toffe82 (Read error: 104 (Connection reset by peer))
17:47:04GodEaterI got 188 connections from rbclient on the 21st of April
17:47:11saratogawhy is that so high?
17:47:26saratogait makes little sense to me, shouldn't each commit be one for the SSH and one for SFTP?
17:47:38linuxstbYes, but also multiple builds per commit.
17:47:43GodEaterah hang on - not all of those are connections
17:47:49*GodEater needs to divide by 4
17:49:09 Join gregzx [0] (
17:49:19 Join toffe82 [0] (n=chatzill@
17:49:25saratogai'd also prefer them to not ask why some server in the EU needs SSH access to our network, so less connections is better in that respect too
17:49:51*obo pings Bagder
17:51:33 Quit funman (" ajax IRC Client")
17:54:42obofrom, it seems to create connections to svn up, build, scp, and then cleanup
17:55:52oboso one svn up, then multiple rounds of build, scp and cleanup depending how many builds the server does that round?
18:03:26 Part wpyh ("Kopete 0.12.7 :")
18:04:35Bagderobo: correct
18:10:08obohmm, the only way I could see to reduce the number of connections would be to have a single-shot clean-up at the end?
18:10:42Bagderor possibly some kind of mode that keeps the ssh connection up and the master sends new commands to it
18:10:53Bagderbut that would require some pretty large changes
18:15:24BigBambisaratoga: As all servers in the EU are nasty? :)
18:15:39 Join faemir [0] (
18:16:10 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
18:19:57 Join flydutch [0] (
18:20:10 Join dan_ [0] (
18:21:32 Quit dan_ (Client Quit)
18:22:34 Nick l403 is now known as randomnickgen (n=l@
18:23:19 Join Tichodroma [0] (
18:27:22TichodromaIs it in principle straightforward to simulate the shortcuts plugin using rockboxui? I tried placing some symlinks to the relevant .rock files in the appropriate places under simdisk/.rockbox but didn't quite get there. Just wanted to check I'm not attempting to do something impossible.
18:29:05 Join stoffel [0] (
18:29:07dionoeaI thought that shortcut stuff was for .lnk files
18:31:51 Quit CaptainKwel (" ajax IRC Client")
18:33:31TichodromaYes it is. I initially tried it with simdisk/.rockbox a symlink to my ipod's .rockbox. However when you try to access a shortcut like that you this error dlopen(simdisk/.rockbox/rocks/viewers/shortcuts_view.rock): simdisk/.rockbox/rocks/viewers/shortcuts_view.rock: invalid ELF header, presumably because that file was compiled for my ipod and not for the simulator. But the simulator build dir contains this file: ./apps/plugins/shortcuts/shortcuts_
18:33:31Tichodromaview.rock so I thought it might work if I symlinked from simdisk/.rockbox to that.
18:34:14***Saving seen data "./dancer.seen"
18:36:10 Nick randomnickgen is now known as l401 (n=l@
18:37:50linuxstbTichodroma: What exactly are you trying to do. As you've discovered, you can't link from a simulator .rockbox dir to a target .rockbox dir. Other links should work fine though I would have though.
18:37:52LloreanThe simulator can't use plugins compiled for the target.
18:38:16LloreanSimulators (and its plugins) are compiled for the host architecture. There's no emulation, so it has no capability of running plugins compiled for the target.
18:41:36TichodromaYes that was what I realised when I got the 'invalid ELF header error' So I thought that the following might do the trick:
18:41:40Tichodromamkdir -p simdisk/.rockbox/rocks/viewers
18:41:51Tichodromaln -s apps/plugins/shortcuts/shortcuts_view.rock simdisk/.rockbox/rocks/viewers/shortcuts_view.rock
18:43:04saratogawhat are you trying to do
18:43:17*kugel has not the slightest idea
18:44:05*domonoky suggest "make install" to Tichodroma
18:45:09TichodromaI'm playing around with altering the code in tree.c / filetree.c to see if I can (a) understand it and (b) arrange for <prev> to return to the previous directory (rather than the parent) after visiting a link.
18:45:34saratogawhy does this involve making symlinks
18:47:43domonoky"make install" will copy the freshly build rockbox files in the simdisk directory, and then shortcuts.rock will work like normal. no need for symlinks
18:48:03TichodromaBecause if you run rockboxui, and hit select on a .link file, the shortcuts plugin does not work out-of-the-box. I'm getting a blank screen when I select a *.link file.
18:48:25Tichodromadomonoky: ok thanks I'll try that
18:50:07 Join JdGordon| [0] (i=836b0070@gateway/web/ajax/
18:50:15 Join bertrik [0] (
18:51:10saratogaTichodroma: you realize that rockbox as an OS doesn't use symlinks right?
18:52:34TichodromaYes I realise. I understand the .link files to be a sort of rockbox version of symlinks. make install was what I needed.
18:53:09LloreanTichodroma: Were you planning on keeping track of all links traversed, or just the most recent one?
18:53:19 Quit pyro_maniac ("Leaving.")
18:55:14kugelTichodroma: hard links, rather
18:56:05Lloreankugel: Uhm, how are they like hard links?
18:56:16TichodromaI was thinking just the most recent one, at least to start with.
18:56:52kugelaren't they just text files which contain the target?
18:57:45Lloreankugel: They point to it yes. But you can't manipulate the data by way of them, afaik
18:57:55LloreanThey're much more like a windows "shortcut" than a hard link
18:58:11 Quit JdGordon| (" ajax IRC Client")
18:59:22kugelthe link file is a text file
18:59:28kugeleach line contains the path to a target
19:00:11saratogai think hard links are part of the file system, not files in it
19:00:19Lloreansaratoga: Exactly.
19:00:49TichodromaAIUI the behaviour's differs depending on whether the link file contains one link or several: if just one then select goes straight to it; otherwise it starts up the shortcut viewer.
19:00:52 Join midijunkie [0] (
19:01:13Lloreankugel: I'm not sure what you mean by "Hard Links" but have you looked at perhaps the wikipedia article on it?
19:01:17LloreanMaybe this is a definitions difference.
19:01:18saratogai don't think hard links would work on fat
19:01:37kugellooks my definition is different from the real one :p
19:01:43kugellooks like*
19:01:48saratogathough i can never keep track of all the differn link, junction point crap
19:02:37TichodromaSo *ideally* after visiting one of several possible links, I would like <prev> to return to the viewer plugin. But that's probably harder than returning to the directory containing the .link file.
19:03:31kugelhard? in my imagination it's just a matter of keeping track of the visited dirs in a visted[][] array
19:04:04LloreanTichodroma: I don't really see a practical benefit to this.
19:04:26LambdaCalculus37saratoga: Have you sent out the Gigabeat T yet?
19:06:55TichodromaFair question. I've written a program that creates a system of .link files which allows you to navigate your music library in additional ways to those offered by the file tree and the id3db. E.g. by musical similarity (defined e.g. by I find that if you start using .link files lot for moving through your library, you start to wish that <prev> sent you back where you came from rather than to the parent dir.
19:07:29LloreanAnd it's impossible to update tagnavi to add these new paths of navigation?
19:09:42TichodromaI admit I need to look into / read up on that.
19:15:11 Join einhirn [0] (
19:17:10 Join moos [0] (i=mustapha@rockbox/staff/moos)
19:19:46 Join jeffdameth1 [0] (
19:19:59 Quit jeffdameth (Read error: 60 (Operation timed out))
19:24:07Tichodromakugel: the thing I was imagining would be harder was returning to the shortcuts viewer plugin, as opposed to the directory.
19:25:02LloreanTichodroma: How would that work if it was a single-link shortcut?
19:25:16kugelhm, not too hard I think. Look at plugin_load(). Some places in the core load plugins as well, like properties or credits
19:26:46 Quit einhirn (Read error: 104 (Connection reset by peer))
19:27:55TichodromaOK. Llorean: yes I was thinking that in principle the code would have to be different for the two cases (single and multiple links)
19:30:56 Join bluebrother [0] (n=dom@rockbox/developer/bluebrother)
19:31:50*bluebrother is annoyed by incomplete bug reports
19:36:43saratogaLambdaCalculus37: no i've been away, I'll go mail it now
19:38:12saratogaactually could you PM me your address again, I seem to have misplaced it
19:40:44 Quit Tichodroma ("Leaving")
19:41:48*kugel wonders what happened about the beast
19:42:09kugeldidn't we plan (or rather consider) to release it or putting it into supported?
19:42:50BigBambiThe installers aren't ready
19:46:02 Join Hillshum [0] (n=chatzill@unaffiliated/hillshum)
19:48:42 Join A_M [0] (
19:49:36linuxstbkugel: The to-do list is still the same...
19:54:43kugelso no progress, that's sad
19:54:43 Join JdGordon| [0] (i=836b0071@gateway/web/ajax/
19:57:51 Join jgarvey [0] (
19:58:33 Nick fxb is now known as fxb__ (
19:59:40 Quit stoffel ("leaving")
20:02:34 Join pyro_maniac [0] (
20:10:42 Quit Horscht ("Verlassend")
20:11:33 Quit Hillshum ("ChatZilla 0.9.83 [Firefox 3.0.3/2008092417]")
20:13:16kugelmcuelenaere: heh, nice lua build. Is it possible to show line numbers when the interpreter finds errors?
20:13:39mcuelenaereyou mean on target?
20:14:01Lloreanmcuelenaere: Are those really builds of Rockbox 3.0?
20:14:16mcuelenaereno, that's just the how the release script names then
20:14:24*mcuelenaere needs to place a note about that in the post
20:14:38 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
20:14:54LloreanYou might want to tell what SVN revision they're from, so people can have better odds of extracting it over a compatible Rockbox version
20:16:12 Join petur [50] (n=petur@rockbox/developer/petur)
20:19:00kugelmcuelenaere: line numbers..
20:19:13kugelah that was in response to Llorean, sorry
20:19:24kugelyes on target (where else?)
20:19:32mcuelenaere(as in DEBUGF)
20:19:53kugelwell, it should work the same on the sim
20:20:15kugelthe interpreter already prints error messages, but not the corresponding line number
20:20:46Lloreankugel: Line numbers?
20:20:47*mcuelenaere looks in the lua source
20:21:03mcuelenaerekugel: what kind of message did you get?
20:21:31kugeland it doesn't yield :(
20:22:20kugelmcuelenaere: something like "expected ')' before 'do'"
20:22:47mcuelenaerekugel: was that message a splashf?
20:23:01kugelsplash or splashf, yes
20:23:37kugellua.g: or so whas the very first part of the string
20:24:09*mcuelenaere thinks he found the line
20:24:15mcuelenaeredid it appear for 5 seconds?
20:24:21*Llorean wonders why there are builds for some unsupported targets there, but not others.
20:24:48mcuelenaereLlorean: I didn't modify tools/release/ much
20:25:02LloreanIt builds for the M:Robe 500 and D2?
20:25:44archivatorQuick question - what's the difference between xlcd's bitmap blipping and the one exposed via rb->*bitmap ?
20:25:47mcuelenaereLlorean: M:Robe 500 not by default, D2 it does
20:25:58mcuelenaeres/it does/yes/
20:25:59Lloreanmcuelenaere: Why'd you add it in?
20:26:07mcuelenaereI just uncommented the line
20:26:38mcuelenaereare unsupported targets not allowed in the test build forum?
20:26:39LloreanYes, but why are there "official test builds" for unsupported targets?
20:26:47LloreanOr rather, why leave out other targets that could use equivalent testing?
20:27:10LloreanWell, it seems to me you should either do as many as it can reasonably be tested on, or just the supported, but not really pick and choose.
20:27:36kugel"test.lua.g: attempt to call global 'string' (a table value)" ??
20:27:45mcuelenaereI didn't really pick and choose, just uncommented that particular line and ran the script
20:27:50mcuelenaerebut I'll remove the unsupported builds
20:27:57 Quit ajb (Remote closed the connection)
20:29:12kugelhaha, my press-a-button-to-exit function even works :p
20:31:07kugelnice work. with rb.sleep it also yields (because sleep does it itself, I guess)
20:31:30kugelbut a while 1 do end loop keeps my dap busy (and locked)
20:31:47mcuelenaereyes, there's no escape function atm..
20:32:16pyro_maniacmcuelenaere: have you tested the test script?
20:32:37mcuelenaerean older version of it, yes
20:32:49pyro_maniacmcuelenaere: there seems to be an issue on the current patch
20:33:09kugelthe test script works for me
20:33:11pyro_maniacthe time was not correct
20:33:20kugeloh, I didn't messure that
20:33:23mcuelenaerepyro_maniac: which time?
20:33:39mcuelenaere(the sleep or the splashf?)
20:33:42pyro_maniacthe test script should run 5 seconds
20:33:50mcuelenaereno, 10sec
20:33:59pyro_maniacon v4 of the patch it worked correct but not on v5
20:34:15***Saving seen data "./dancer.seen"
20:34:16mcuelenaereyes, but I changed the behaviour of rb.splash() to mimic the C function more
20:34:35mcuelenaere(there wasn't a need to multiplicate with rb.HZ at first, but I added that)
20:34:35kugelmcuelenaere: is a thread created for the interpreter? I think an instance should regulary post something, and let the interpreter exit if it doesn't get a message
20:35:12kugelis there a standard for lua?
20:35:19*mcuelenaere doesn't know very much about the internals of the patch, safetydan did the most
20:35:27kugelstandard functions like sleep I mean
20:35:58kugelif yes, I think it should rather follow the lua standard, not the rockbox internals
20:36:00mcuelenaereI don't think so, it looks like everything works through libraries (although Lua tries to push a standard library)
20:36:29mcuelenaeresee the os.* functions
20:38:45bluebrotherok, there are two targets that are affected by the naming stuff for theme installation via rbutil: h100 and ipodmini2g
20:39:27pyro_maniacmcuelenaere: i dont understand the change of rb.splash
20:39:27bluebrotherany idea how we should deal with them? Add a workaround so we can release a new rbutil and then fix it properly?
20:39:38LloreanWhat's the issue?
20:40:21bluebrotherthemes list is always empty −− the theme site doesn't know about ipodmini2g while rbutil transmits the configure name
20:40:34mcuelenaerepyro_maniac: before I changed it, the splash wrapper would multiplicate the value with HZ (so you could pass seconds as an argument to rb.splash), now it needs ticks as an argument (meaning that if you want to wait for 5 sec, you need to wait for 5*rb.HZ)
20:41:24Lloreanbluebrother: Why can't we fix the theme site?
20:41:51pyro_maniacmcuelenaere: yes and in the testscript 5 seconds are calculated of 5*rb.HZ
20:41:53kugelwe have a malloc now? :)
20:42:00mcuelenaerekugel: lua has :)
20:42:03bluebrotheras far as I understood the theme site code there is no easy way to have an alternate name for a target
20:42:10mcuelenaere(see the Flyspray entry for more info)
20:42:27mcuelenaerepyro_maniac: did you use the new testscript with the old code?
20:42:33bluebrotherrasher? domonoky? Any thoughts on that?
20:42:33mcuelenaere(or vice versa)
20:42:33Lloreanbluebrother: Why can't we just have more ipodmini2g as another target with the same properties as the ipod mini then?
20:42:41pyro_maniaci will test
20:42:44BigBambigevaerts: Does a buffer i/o error in dmesg followed by my S disconnecting and Rockbox locking on the USB screen indicate a hardware or a software problem? Or either? :)
20:43:02kugelmcuelenaere: I see that __unlikely and __likely are empty, even though rockbox actually makes use of them
20:43:03mcuelenaerepyro_maniac: also, I fixed an error where rb.HZ actually containt LCD_WIDTH (or vice versa, can't remember what exactly)
20:43:06Lloreanbluebrother: I don't think there's significant harm in having a couple redundant targets on the page.
20:43:22bluebrotherLlorean: I don't see why we couldn't. No idea if that's a good thing. rasher should be able telling more I guess
20:43:34bluebrotherthe same issue exists for h100 vs h120 btw.
20:43:51*Llorean thinks the same solution is applicable there. =P
20:44:03bluebrotherhehe :)
20:44:10LloreanA better solution can come, but I'd rather the solution we choose now be hacks on the website side so we don't need yet another RBUtil after
20:44:13mcuelenaerepyro_maniac: I'll fix it
20:44:33*domonoky also think extending the theme site would be good.
20:44:34LloreanAnd the most "consistent" hack would be to have a couple extra targets displayed for a while.
20:44:41Lloreandomonoky: Extending?
20:45:05domonokyextenting so it knows all targets from configure.
20:45:38bluebrothermake it bigger! Enlarge the font / images / screenshots! ;-)
20:45:50LloreanWell, it's going to have to be updated to the new website theme soon enough
20:46:37bluebrotherurgh. I'm not too fond of that new site, especially all those bling-bling stuff
20:46:44LloreanNor am I.
20:46:54LloreanBut it looks like there's a lot of weight behind it now.
20:46:57BigBambibluebrother: The proposal is to have a seperate "dev" site
20:47:05LloreanAnd maybe it'll lead to the separate 'dev' site soon enough.
20:47:06BigBambiLlorean: Indeed there is, I like it :P
20:47:22LloreanBigBambi: I don't mind it too much except, as I said, I'm really not fond of the white bar in the middle.
20:47:25bluebrothercouldn't we scrap all those animated drop-shadow whatever stuff? The design itself isn't too bad
20:47:33LloreanIt feels like there's a lot of "noise".
20:47:40bluebrotherBigBambi: I'm aware of that. I still don't like the site proposal :)
20:47:41pyro_maniacmcuelenaere: no i think it was just me false
20:48:08pyro_maniaci tested the new test script and it seems to work right
20:48:10BigBambibluebrother: Fair enough :)
20:48:40*Llorean thinks we also need to run it by the blind and see how navigable it is.
20:49:10mcuelenaerewhen an error is thrown, the error message seems to get put on stack and you just need to get it off and display it; but I can't seem to find anything relating to the line number..
20:52:41 Quit timc (Success)
20:52:47 Join Tomers [0] (
20:56:04kugelmcuelenaere: maybe put the line number into the message?
20:56:22mcuelenaerethat would require messing with the Lua internals (and the documentation says it already does that :/)
20:57:02kugelthis implementation apparently doesn't thoug :(
20:58:49gevaertsBigBambi: I'm not sure...
20:59:11 Join Thundercloud [0] (
20:59:23 Join MT [0] (n=chatzill@
20:59:47BigBambigevaerts: It has just done it again - both times it was when opening an image in gimp direct from the beast
20:59:56gevaertsBigBambi: same image?
21:00:04BigBambigevaerts: No, different one
21:00:16BigBambiAnd the image opened fine, it was when I went to save as in fact
21:00:23 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
21:00:27gevaertstry fscking it
21:00:46BigBambiAnd in order to get it to connect again I restarted X, as otherwise it sits there saying waiting to settle in dmesg
21:00:51BigBambiWill do
21:01:22BigBambiI guess Rockbox freezes as it doesn't disconnect properly? (whatever causes it)
21:01:48 Join arohtar [0] (
21:04:09 Join robin0800 [0] (
21:05:40 Quit faemir (Read error: 104 (Connection reset by peer))
21:06:13BigBambigevaerts: Free cluster summary was off by 5 clusters, that's it. I've corrected that and I'll try again :)
21:06:43gevaertshm, that shouldn't have horrible effects...
21:07:44mcuelenaerekugel: I seem to have line numbers on errors?
21:08:18mcuelenaerehow did you produce that error?
21:08:37BigBambigevaerts: Same again, with yet another different image - everything works fine, browsing, displaying, deleting etc, except for when I press save as in gimp
21:08:55kugelmcuelenaere: I've head different errors, neither had the line number
21:09:09mcuelenaerehave you tried adding some garbage?
21:09:13kugelwhere's the line number supposed written?
21:09:19mcuelenaerein the front of the message :)
21:09:20kugel+to be
21:09:29kugelas in?
21:09:39kugelbefore the filename or after?
21:09:45gevaertsBigBambi: I'll spend some time on it during the weekend
21:09:53mcuelenaereI get something like './test.lua:18: malformed number near '1dsfsdfs''
21:09:59kugelit's probably there where I get this weird 'g'
21:10:04BigBambigevaerts: cheers very much - I just hope it isn't hardware
21:10:09BigBambigevaerts: dmesg
21:10:14mcuelenaerehmm, I'll try on target
21:10:34kugelI have this: "/test.lua:g: 'do' expected near 'end'"
21:11:36 Join schrottplatz [0] (
21:11:37kugelmcuelenaere: I tried your error, still no line number
21:17:44Lloreankugel: Why don't you just dig up the old ML thread and show where the consensus was what you said it is if there was a consensus.
21:18:00LloreanRather than persisting the whole "Well, my opinion is that everyone agreed this" without any more to back it up than he has.
21:18:20 Join Zagor [242] (
21:20:01kugel"I just stated what I recall."
21:20:19linuxstbkugel: That's why we have these discussions in publically logged places...
21:20:33Lloreankugel: Yes, and we're beyond the point where memory is at all useful.
21:22:25kugelLlorean: I think if it was possible to quote someone who posted the exact consensus, fml wouldn't have posted to the list again.
21:22:45BigBambiDoesn't that suggest that there wasn't a consensus then?
21:22:58BigBambiIf there was one, it would be possible to quote it surely?
21:23:24 Quit robin0800 (Remote closed the connection)
21:23:39kugeland since no-one did it, the only thing we can do is to state how/what we felt the consensus was
21:23:55LloreanIf you can't show there was a consensus, your opinion on whether there was one doesn't really matter.
21:24:03LloreanYou either open discussion again and tally opinions, or just have a vote.
21:24:04BigBambiA consenus isn't a matter of what you feel
21:24:13BigBambiEither there was or wasn't one
21:24:26kugelok, then I'll just shut up
21:25:05 Quit Tomers (Remote closed the connection)
21:25:48 Quit barrywardell (Remote closed the connection)
21:26:40 Quit martian67 (Remote closed the connection)
21:35:53BigBambigevaerts: OK, super wierd - it just did the exact same thing when I clicked save as in gimp when the picture I opened and was manipulating was on a completely different disk
21:36:02BigBambigevaerts: I wasn't doing anything with the beast
21:36:41BigBambibut as soon as I clicked save as in gimp, the file browser window I had open on the beast went empty and same thing in dmesg
21:37:53mcuelenaerekugel: I also get that 'g' instead of the line number...
21:41:25gevaertsBigBambi: you didn't put your gimp swap files on the beast by accident?
21:41:56*gevaerts has no idea about what could be going on
21:42:08BigBambigevaerts: I've never touched them, and I don't normally have the S connected when using it, but I can check
21:42:39gevaertsI'll have to try to reproduce this somehow
21:43:21BigBambiIt is incrdibly weird
21:43:38BigBambiI'll try with another players OF and see what happens :)
21:45:33BigBambiWell, the H140 doesn't mind :)
21:45:50 Quit Zarggg_ ()
21:49:54mcuelenaerehmm somehow the lua printf() implementation converts 12 to g? (%d)
21:50:14mcuelenaeres/12/any number/
21:51:48 Join __lifeless [0] (n=lifeless@
21:52:12 Quit moos ("Rockbox rules the DAP world")
21:56:57Lloreanlinuxstb: Good point re: number of targets. It needs to basically be infinitely expandable.
21:57:01domonokymcuelenaere: look for LUA_NUMBER_FMT
21:57:19domonokythe printf in rockbox doesnt support float
21:57:55 Quit LambdaCalculus37 (" bye!")
21:58:03mcuelenaereLUA_NUMBER_FMT is set to %.14g
21:58:43 Join Zarggg [0] (
21:58:54 Quit _Auron_ (Read error: 104 (Connection reset by peer))
21:59:04 Join bs66_1 [0] (
21:59:27 Join _Auron_ [0] (
21:59:38domonokywhich rockbox's snprintf probably doesnt understand.
22:01:34*mcuelenaere wonders whether he should adjust luaconf.h or rockconf.h
22:02:04mcuelenaereif you have #define X a\n#define Y X\n#undef X\define X b what will Y give you?
22:04:10mcuelenaeres/define X b/#define X b/
22:04:14amiconnarchivator: The functions in xlcd complement the ones exported from the core.
22:04:18peturand a slap on the wrist for writing code that's hard to read ;)
22:04:31*mcuelenaere doesn't want to multipaste here :)
22:05:32 Quit _lifeless (Read error: 110 (Connection timed out))
22:06:05archivatoramiconn: I gathered as much. GraphicsAPI lists the *scroll_<dir> functions so I thought those were exported in rb-> as well. Turns out, they're not :)
22:06:28 Join Tomers [0] (
22:06:32mcuelenaeredomonoky: thanks, works now!
22:07:14amiconnarchivator: See footnote 3) ...
22:10:08archivatorRight, I saw it *after* I asked, sorry about that. On a related note, why would grey_clear_display() fail with a segmentation fault when memset()-ing the buffer? I've called grey_init(), grey_show(true), anything I'm missing?
22:11:00TomersGood evening guys. I have some questions, that I'll be glad if you can help me first. Where can I found documentation for the USB module of PP5024? Wiki mentions that it is similar to FreeScale's i.MX31 and that this system has lots of documentation available. do we have any documents which can be helpful in developing the USB driver (I need to use Interrupt transfers)?
22:12:01BigBambiThere are no docs AFAIK
22:12:17Bagderi.MX31 has docs for those things I believe
22:12:20BigBambifor the PP that is
22:12:54Bagderright, but its the same as the mx31 so the guys use the mx31 docs
22:13:28BigBambiyep, I was replying to the " Where can I found documentation for the USB module of PP5024?" bit
22:13:42TomersDo we have the mx31 documents? Do you think I can find them on FreeScale's site?
22:13:59amiconnarchivator: Depends on a lot of details. First one: Did you select buffered or unbuffered mode?
22:14:02TomersAre there any sub-versions of mx31?
22:14:04BagderTomers: they are public, or at least they used to be
22:14:19 Quit bs66_ (Read error: 110 (Connection timed out))
22:14:31saratogatheres a link to th mx31 docs on the beast wikipage
22:14:38saratogait worked about a week ago
22:15:25TomersBTW why does it called 'beast'? Is it the manufacturer name?
22:15:54bluebrotherIt's Gigabeat S with a typo
22:15:55BigBambiit is a gigaBEAT S and it has a massive processor
22:16:01archivatoramiconn: Buffered mode (I need text handling).
22:16:02saratogaGigaBEAT S with the S and T flipped
22:16:20BigBambiIf something is very powerful you can desribe it as a beast, and stick the S into the beat...
22:16:41 Join nibbler [0] (
22:18:25amiconnarchivator: Hmm, ok. The buffer you're passing to grey_init() is sufficiently large? Are you checking the result of it?
22:18:46 Join Grahack1 [0] (
22:18:47amiconnWhether the greyscale overlay is visible doesn't matter at all for the drawing functions
22:20:21archivatoramiconn: I followed the other plugins' approach - using plugin_get_buffer(). grey_init() is definitely returning true. I'm building on M5 simulator, if that is of any help.
22:21:27archivatorAh, I saw what I was doing wrong.
22:21:58bluebrotherhmm. What do people think about not showing a preview in the theme install window if more than one theme is selected?
22:22:16*amiconn is curious
22:24:32archivatoramiconn: I was doing it wrong. I had GREY_ON_COP|GREY_BUFFERED #define'd (I know, bad idea) and #ifdef'd at that (don't ask, it's fixed now). The #ifdef was wrong and so it defaulted to GREY_ON_COP. Silly me.
22:24:53 Quit Grahack1 ("Leaving.")
22:25:15pyro_maniacmcuelenaere: will you post a patch today?
22:25:28mcuelenaerepyro_maniac: I will, I'm currently a bit busy though
22:27:17pyro_maniacmcuelenaere: so it will take some time
22:29:07 Join lymeca [0] (
22:29:27 Join moredhel [0] (
22:29:43 Quit Tomers ("ChatZilla 0.9.84 [Firefox 3.0.9/2009040821]")
22:32:26 Quit arohtar (Read error: 54 (Connection reset by peer))
22:32:33 Join Toored [0] (n=Toored@
22:34:05 Join robin0800 [0] (
22:34:17***Saving seen data "./dancer.seen"
22:36:26 Quit _Auron_ ("Infinity repeatedly denies rumours of plotting with zero to bring down the Universe.")
22:37:51 Join EternalRains [0] (
22:38:30 Join robin0800_ [0] (
22:43:53 Quit nibbler ("Ex-Chat")
22:48:58 Quit moredhel (Client Quit)
22:49:10 Join faemir [0] (
22:53:59 Quit robin0800 (Read error: 113 (No route to host))
22:56:46 Part pyro_maniac ("Leaving.")
22:57:10 Quit schrottplatz (Remote closed the connection)
22:58:27 Quit archivator ()
22:59:04 Quit bertrik (Remote closed the connection)
22:59:50 Join bertrik [0] (
22:59:56 Part saratoga
23:00:29 Join barrywardell [0] (n=barrywar@
23:00:39 Join saratoga [0] (i=9803c6dd@gateway/web/ajax/
23:07:58 Join Conic [0] (
23:08:52 Join stripwax [0] (
23:10:08 Quit Toored (Connection timed out)
23:10:15 Join Toored [0] (n=Toored@
23:10:57 Join robin0800 [0] (
23:11:05 Quit A_M (" HydraIRC -> <- In tests, 0x09 out of 0x0A l33t h4x0rz prefer it :)")
23:17:47 Quit petur ("Zzzzz")
23:28:23 Quit robin0800_ (Read error: 113 (No route to host))
23:30:46 Join Seed [0] (
23:32:58 Join Tomers [0] (
23:32:59 Quit stripwax ("")
23:34:08TomersThis may sound like a (very) stupid question: The code defines only 3 endpoints for PP5024. Does this is a limitation of the HW, or does it set because we only need 3?
23:35:51gevaertsI think there are three endpoint pairs on PP502x
23:36:23 Join stripwax [0] (
23:36:25gevaertsif you want to make sure, there's a register somewhere
23:36:28 Quit bmbl (Read error: 104 (Connection reset by peer))
23:36:49TomersNormally, we use ep#0, and one pair of ep's for Mass (ep#1) so we left with ep#2 IN/OUT and ep#3 in/out. right?
23:38:37TomersI'll look for it. The code refers to manual sections. Do we have it? I've asked previously and I got the answer that we got nothing. I went to the links people pointed me to, but got no useful specs - definitely not one that have the sections mentioned in the code comments
23:39:16gevaertsthe sections it refers to are in the imx31 manual
23:39:35TomersI couldn't find it. Any help? hint?
23:40:17 Quit stripwax (Client Quit)
23:40:24saratogaits linked from the gigabeatS wiki page
23:40:51saratogai looked at it a few days ago [can't bring up the wiki from this computer]
23:41:00TomersI didn't find it there. I found link to general PDF. Not the one the code refers to.
23:41:14saratogaits about 2000 page long right?
23:41:14TomersI'll try again. Maybe I missed it
23:41:35gevaertsyou need MCIMX31RM
23:42:31 Join CIA-63 [0] (
23:43:03 Quit mcuelenaere (Nick collision from services.)
23:43:04 Join mcuelenaere1 [0] (
23:43:10 Nick mcuelenaere1 is now known as mcuelenaere (
23:43:30 Quit CIA-43 ()
23:45:57 Join stripwax [0] (
23:46:35TomersTerrific!!! Thanks! That solves the problem... I'll continue tomorrow. Good night!
23:46:37 Quit gniourf_gniourf (Remote closed the connection)
23:46:45 Quit Llorean (Read error: 104 (Connection reset by peer))
23:46:51 Nick fxb__ is now known as fxb (
23:47:07 Join Llorean [0] (
23:47:22gevaertsHave fun with it :)
23:48:13 Quit ender` (" I don't see what C++ has to do with keeping people from shooting themselves in the foot. C++ will happily load the gun, off")
23:48:38 Quit domonoky (Read error: 104 (Connection reset by peer))
23:49:24 Quit Tomers ("ChatZilla 0.9.84 [Firefox 3.0.9/2009040821]")
23:55:20 Quit jgarvey ("Leaving")
23:58:48 Join timc [0] (n=aoeu@

Previous day | Next day