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).

Notice: Only Gecko based browsers prior to FF4 support the multipart/mixed "server push" method used by this log reader to auto-update. Since you do not appear to use such a browser, this page will simply show the current log, and not automatically update.

#rockbox log for 2010-04-14

00:00:24Tornelinuxstb: it uses the same code as the main build, so 1g/2g it works for firewire which is all they have, and 3g it only works for firewire even though it has usb.
00:00:52Torneenough of the usb driver is compiled into the bootloader already that this is just a tiny patch to bootloader/ipod.c
00:00:55Watermark@linuxstb, is there anything else you can think of that might fix my S9? if not then i am going to look into may be just buying the COWON J3
00:01:08 Join rhodan [0] (~quassel@2001:1608:12:2::38)
00:01:15rhodanI added two albums.
00:01:27rhodanNow Rockbox is updating the collection.
00:01:31CIA-5New commit by 03funman (r25642): s3c2440: fix a typo, 4|1 == 5, not 0x41
00:01:39rhodanFive minutes now.
00:03:20Tornelinuxstb: i'm not actually going to release new bootloaders just yet, i have at least one more pending change to go into it (sector size workaround)
00:03:32linuxstbWatermark: Nope, sorry. I'm pretty sure it's just a matter of someone figuring out the SDCFG value though.
00:03:34Tornelinuxstb: but i will at some point soon, to pick up the various bootloader stuff i have changed :)
00:03:58linuxstbYes, it's been years since an ipod bootloader release hasn't it?
00:04:03Watermarkalright, i'll see about getting that. i will report back
00:04:22 Quit toffe82 (Read error: Connection reset by peer)
00:05:48Tornelinuxstb: yah, but not a lot has changed
00:06:08Tornelinuxstb: my OSOS boot stuff, the addition of large drive support (which i've only half done), and this, are the three things I've changed.. :)
00:06:45Tornethere was a discussion earlier about the fact that the default contrast for the 3G has been changed, which affects the bootloader
00:06:57Torneand may make bootloader errors completely invisible on some people's screens ;(
00:07:21 Quit Luca_S (Quit: CGI:IRC (EOF))
00:08:12Torneis there a good reason why we avoid including button.h in the bootloader and instead define arbitrary values for BUTTON_LEFT et al?
00:08:18Tornebecause that causes a conflict when you #include usb.h
00:08:21linuxstbDoesn't sound desirable - our 3G users will both be upset...
00:08:45Tornei just nuked the definitions in the bootloader and everything appears to still work, they get defined as different arbitary values but it's still *consistent* :)
00:09:06Tornethe alternative would be to fix usb.h i guess
00:09:28Tornelinuxstb: wel it was a 3g user that asked to ahve the contrast changed int he first place because the previous default made everything almost unreadably *dark*
00:09:50Tornebut jhMikeS found that the new default makes his screen so light it's literally blank ;)
00:09:56gevaertslinuxstb: not *both* of them. One of them wanted this change!
00:09:59 Quit halmi (Quit: halmi)
00:10:02Tornewhich confused him a bit when the bootloader error was invisible
00:10:29Tornethe default in firmware is used in the bootloader but is also the default for the main build, so.. yeah. i dunno. ;)
00:10:41Tornei'm hoping someone with a b/w ipod will come up with a better idea
00:11:38 Join M3DLG [0] (
00:15:46pixelmagevaerts: the other two! I know of at least three 3G Ipod owners
00:16:13gevaertsWow, it's more popular than the Ondio! ;)
00:17:45CIA-5New commit by 03torne (r25643): FS #11199: ipod bootloader: reboot to disk mode on cable insert, following a fatal error ...
00:18:39pixelmagevaerts: I know of more than three Ondio owners
00:24:33 Quit ender` (Quit: If you have a few hundred followers, and you let some of them molest children, they call you a cult leader. If you have a billion, they call you Pope.— Bill Maher)
00:27:37 Quit jgarvey (Quit: Leaving)
00:31:37wodzHD200 OF reads adc in ISR :-)
00:36:40 Quit phanboy4 (Read error: Connection reset by peer)
00:41:07 Quit liar (Ping timeout: 248 seconds)
00:43:59 Quit wodz (Quit: Leaving)
00:48:02 Quit arbingordon (Quit: `)
00:48:59 Join mt [0] (~chatzilla@
00:55:03ssorgatemfunman: left button OF boot doesn't work
01:00:23ssorgatemoh well, it's pretty late now
01:00:32ssorgatemi'm going to sleep
01:01:12 Quit MethoS- (Read error: Connection reset by peer)
01:01:13ssorgatem'night !
01:01:22 Join Llorean [0] (~DarkkOne@rockbox/user/Llorean)
01:01:22 Quit ssorgatem (Read error: Connection reset by peer)
01:03:44 Quit evilnick (Quit: Page closed)
01:07:07***Saving seen data "./dancer.seen"
01:09:32 Quit jordan` (Read error: Operation timed out)
01:17:43 Join flyback [0] (
01:18:11 Join MethoS- [0] (~clemens@
01:18:22*flyback is sorry for past drama in channel he didn't even think about when he vented his frustration with life and used a global message
01:19:44 Quit DerPapst (Quit: Leaving.)
01:24:33 Nick fxb__ is now known as fxb (
01:27:19 Quit xiainx (Ping timeout: 246 seconds)
01:29:59 Nick fxb is now known as fxb__ (
01:32:20 Quit MethoS- (Remote host closed the connection)
01:33:16 Join Adubb [0] (~Aldubuc@
01:33:28 Quit M3DLG (Ping timeout: 264 seconds)
01:33:38 Part Adubb
01:34:50 Nick fxb__ is now known as fxb (
01:43:33 Join CaptainKewl [0] (
01:43:46 Join xiainx [0] (
01:44:52 Join S_a_i_n_t [0] (S_a_i_n_t@
01:46:40 Quit Stephen__ (Quit: Leaving)
01:47:20 Join jordan` [0] (
01:51:28 Quit Schmogel (Quit: Miranda IM! Smaller, Faster, Easier.
01:54:58 Join Strife89 [0] (~michael@
01:55:48 Join kisak [0] (
01:57:03 Join S_a_i_n_t_ [0] (S_a_i_n_t@
01:57:20 Quit S_a_i_n_t (Ping timeout: 276 seconds)
01:57:29 Quit Strife89 (Client Quit)
01:58:40kisakhello, I noticed a wiki formatting discrepancy
01:58:57 Nick fxb is now known as fxb__ (
01:59:28saratogastripwax: (for the logs) any movement on the Tremor patch?
01:59:47kisakoh oops ... sorry worng project
01:59:51 Part kisak
02:07:26 Quit antil33t (Read error: Connection reset by peer)
02:08:49 Quit CaptainKewl (Remote host closed the connection)
02:10:13 Quit komputes (Ping timeout: 265 seconds)
02:11:38 Quit Watermark (Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100401080539])
02:13:14 Join antil33t [0] (
02:14:26 Quit S_a_i_n_t_ (Ping timeout: 248 seconds)
02:15:11 Join S_a_i_n_t [0] (S_a_i_n_t@
02:17:59 Quit Koreth_ (Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100401080539])
02:34:35 Quit adnyxo (Ping timeout: 240 seconds)
02:34:53 Join arbingordon [0] (~w@unaffiliated/arbingordon)
02:39:31kugelI almost finished porting pth
02:39:49saratogai missed the earlier talk, what does pth bring?
02:40:41kugelit's a cooperative threading library
02:40:56*flyback bbl
02:40:59kugeland it supports priority
02:41:08saratogadon't we already have that?
02:41:25kugelthat's two advantages over the current sdl thread hack. it's also supposedly more portable than sdl
02:41:59kugelsdl uses the underlying system for threads (i.e. preemptive on modern OSes), however it can be configured to link against pth
02:42:12saratogaah ok
02:42:59kugelpth implements a user space library (the kernel sees only the main process with a single thread), which is actually a disadvantage
02:43:12kugelwe will see if it brings us anything, I only know it's my assessment task :p
02:45:24kugelthe disadvantage is also the reason it's so portable (plus that it's written in ansi c), that might be handy when it comes to porting to android where you need java for OS calls
02:46:18 Part flyback ("Leaving")
02:46:27kugelbut android implements a subset of pthread so it might not be that handy, depending on how make-rockbox-run-under-preemptive-thread turns out
02:47:21saratogais there an advantage to using preemptive threads on a device like android?
02:49:11kugelbeing independent on the underlying thread would be an advantage on it's own, I can't say yet if preemptive threads would really be a win in general
02:49:23 Join JdGordon [0] (~7bf38c1f@gateway/web/freenode/x-ecamqjqyvuyhrzkt)
02:50:17kugelour concept if one big main thread, and a few small background ones fits very well to cooperative threads. but I can also imagine that we only have this concept because we only ever had cooperative threads
02:52:54funmanif the threads don't need to be real time then they don't need to be prempted
02:54:39 Quit kugel (Remote host closed the connection)
02:54:51funmanbtw there is a patch on flyspray (from jhmikes?) which shows how much time each thread has been running
03:01:35*jhMikeS comes by and sees thread talk
03:01:46jhMikeSfunman: no, the patch isn't mine
03:02:38funmanFS #11176 (from mcuelenaere)
03:03:44jhMikeSkugel: I've thought about preemtive threads, being explicitely created to be such. They would have some real advantage for realtime tasks in thread context, low-latency audio, and perhaps certain system tasks that should run. It need not be global.
03:05:17 Join anewuser [0] (anewuser@unaffiliated/anewuser)
03:06:15funmandoesn't audio/pcm 'thread' run from isr already ?
03:07:08jhMikeSfunman: I wouldn't call it a thread. It's not a scheduler entity, thus cannot really use the functions.
03:07:10***Saving seen data "./dancer.seen"
03:07:58funmanright, i mean there's no latency since it interrupts the other threads to send data down to the headphones
03:08:50jhMikeSfunman: but, doing low-latency mixing in a thread would allow the ISR to interrupt it and use of kernal object. doing alot of heavy work in the ISR is not possible, due to limited IIS queue and blocking out everything else.
03:09:18 Join Strife89 [0] (
03:10:10jhMikeSall in all, it would be no worse than coprocessor threads, which is as normal as it gets, and would have certain limited access
03:10:39 Quit GeekShadow (Quit: The cake is a lie !)
03:14:41 Quit funman (Quit: free(random());)
03:38:49 Join stasis [0] (
03:39:20 Join hebz0rl_ [0] (
03:40:38stasishello all
03:41:35 Quit stasis (Client Quit)
03:43:03 Quit hebz0rl (Ping timeout: 258 seconds)
04:23:14 Join Barahir_ [0] (
04:23:34 Join Rob2223 [0] (
04:24:36 Quit Barahir (Read error: Operation timed out)
04:27:08 Quit Rob2222 (Ping timeout: 265 seconds)
04:31:17 Join CGL [0] (~CGL@
04:37:07 Join mikroflops_ [0] (
04:40:33 Quit mikroflops (Ping timeout: 252 seconds)
04:50:16 Join Rondom_ [0] (
04:54:02 Quit Rondom (Ping timeout: 248 seconds)
05:01:20 Join n1s [0] (~n1s@rockbox/developer/n1s)
05:07:16***Saving seen data "./dancer.seen"
05:21:34 Quit Boldfilter (Quit: Boldfilter)
05:21:39 Join RadicalR2 [0] (
05:24:40 Quit RadicalR (Ping timeout: 245 seconds)
05:27:36 Join Boldfilter [0] (
05:30:34 Join RadicalR [0] (
05:31:20 Quit RadicalR2 (Ping timeout: 245 seconds)
05:38:21 Quit Strife89 (Quit: Sleep.)
05:41:06 Quit Horscht (Quit: Verlassend)
05:55:26 Join kramer3d [0] (~kramer@unaffiliated/kramer3d)
05:57:04 Quit arbingordon (Quit: `)
06:08:25 Quit RadicalR (Ping timeout: 245 seconds)
06:09:03 Join RadicalR [0] (
06:09:28 Quit Boldfilter (Quit: Boldfilter)
06:17:09 Join Adubb [0] (~Aldubuc@
06:18:55 Join merbanan [0] (
06:21:49 Quit MagusG (Read error: Connection reset by peer)
06:21:53 Quit xiainx (Quit: Good Bye)
06:22:18 Join MagusG [0] (
06:23:11 Join Tetracomm [0] (~nicholas@unaffiliated/tetracomm)
06:23:35TetracommAny progress with understanding Creative MP3 players?
06:23:57saratogano one has worked on them in a couple years
06:27:18 Quit panni_ (Read error: Connection reset by peer)
06:30:01 Quit Tetracomm (Quit: Ex-Chat)
06:39:31 Join Adubbb [0] (~Aldubuc@
06:39:48 Part Adubbb
06:42:33kramer3dwhere can i see recent rockbox commits?
06:47:44saratogakramer3d: theres a list on the front page
06:55:41 Join Boldfilter [0] (
06:56:59 Quit n1s (Ping timeout: 264 seconds)
07:00:53 Join BHSPitMonkey [0] (~stephen@unaffiliated/bhspitmonkey)
07:03:28 Quit anewuser (Quit: What do you know...THE WORLD'S first NTRQ (that's for NES/FAMICOM) tracking compo. Have powerpak? Try it out! Otherwise ROM IMAGE.)
07:05:06FingerSSo is the port to Fuse V2 moving along pretty well now?
07:07:09 Join Watermark [0] (
07:07:18***Saving seen data "./dancer.seen"
07:07:35Watermark@linuxstb you still on?
07:08:09 Join einhirn [0] (
07:09:18 Quit Watermark (Client Quit)
07:39:18 Quit mt (Remote host closed the connection)
07:45:32 Quit saratoga (Quit: Page closed)
07:51:15 Join phanboy4 [0] (
07:56:18 Part Boldfilter
07:58:12 Quit kramer3d (Ping timeout: 240 seconds)
08:00:42 Join mt [0] (~chatzilla@
08:05:07 Join Buschel [0] (
08:16:03 Join wnmnkh [0] (
08:16:34 Quit JdGordon (Ping timeout: 248 seconds)
08:22:46 Join kifo [0] (
08:23:15kifoI was hunting around for builds of the clip+ but they seem to have vanished off of the face of the earth
08:23:24kifostarted earlier today, about 12 hours ago
08:23:35kifoany insight on that?
08:24:04kifosansa clip+ by the way
08:24:46wnmnkhyou mean the files on site?
08:24:57wnmnkhor on the actual player?
08:25:00 Join ender` [0] (
08:25:01kifoon the site
08:25:10kifoits usually on the build page
08:25:19kifowhere i am
08:25:25kifoits not there or in the daily builds
08:25:28wnmnkhit's gone
08:25:33kifowhats up with that
08:25:43wnmnkhbecause it was demoted to "Unusuable" from "Unstable"
08:25:58kifooh wow
08:26:04wnmnkhwith too much errors and lacking features
08:26:13kifocant say i disagree with the move
08:26:15wnmnkhI mean, come on, it needs more work
08:26:32kifothe last 20 or so revs have all crashed on playback for me
08:27:16wnmnkhmy Clip+ also crashes like every 15 min, so it is not a strange decision.
08:27:36kifoits not a total loss, theres not too much thats horrible about the interface thats on there already
08:27:57kifoespecially compared to how much work woulda had to be done just to get basic things working
08:28:00kifoor atleast how much i would assume
08:28:07wnmnkhAt least
08:28:16wnmnkhmy Clip v1 works quite well
08:28:23kifoim not really a uhm
08:28:34kifodeveloper or programmer or anything
08:28:38kifoso couldnt say for sure
08:28:43kifoi just like tinkerin with stuff
08:28:50kifosansa players in general are really nice
08:29:16kifoi was excited enough about it actually being on the clip+ i was willing to risk bricking it to test it out
08:29:22kifoi still would be
08:29:29wnmnkhwell, I got Clip+ as gift...
08:29:37wnmnkhand I already have Clip v1....
08:29:51kifoi give them as gifts to my nephews and such
08:29:55kifoonly 2gb ones though
08:30:11wnmnkhthing is, as far as I understand, we really do not fully understand hardware enough to make stable firmware yet
08:30:30wnmnkhI usually buy Cowon players for gifts...
08:30:35kifoim surprised sansa isnt working with it
08:30:40kifoor at least giving some info or something
08:30:57kifoprobably not in their best interest
08:31:05wnmnkhThere are some legal concerns
08:31:24wnmnkhfor 3rd party firmware on players.
08:31:36kifowell i mean with them uhm
08:31:46kifothey seem to be pretty good about the no lock-out thing
08:31:53kifoi would think that they would take it from the start
08:31:56kifoas in not develop their own
08:32:04kifoitd be a huge step but
08:32:21kifowith even just basic info it seems making a functional rockbox firmware for it
08:32:26kifoinsstead of them developing their own
08:32:30kifobe a lot more cost effective
08:33:00kifothat probably has legal implications also
08:33:17wnmnkhThey have to pay royalities for certain codecs
08:33:19wnmnkhsuch as
08:33:33wnmnkhAAC, WMA, etc
08:34:03wnmnkhthat may makes the player more expensive
08:34:24kifoif people learned how to change the most basic settings in their media player WMA would fall off of the face of the earth
08:34:46wnmnkhit already has fall off
08:35:01wnmnkhhow many stores actually use WMA as main format?
08:35:03kifoheh itd be nice if you could reject the license like you can with a laptop OEM install
08:35:11kifoyea but
08:35:14wnmnkheven Zune store has changed into mp3 a while ago.
08:35:15kifopeople rip to wma all the time
08:35:27kifotheyre somewhat on public trackers
08:35:32kifotorrent trackers
08:35:41kifojust because people dont know how or cant be bothered to change it
08:36:43wnmnkhActually WMP does not give you much options in the first place....
08:37:20kifoterrible player
08:37:25 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
08:37:25kifobut what can you do
08:37:28wnmnkhthing is MP3 encoder on WMP is total trash compared to LAME, so basically all you have is WMA and Wave file.
08:37:29kifoends up being like IE
08:37:37kifohas a userbase because its already :there:
08:38:00kifoyea the clip+ can play my huge library of protected wma files
08:38:08kifothank god i was worried
08:38:36kifoactually i dont think ive ever seen a protected wma
08:39:04wnmnkhwhere did you get protected wma files?
08:39:11kifobeing sarcastic lol
08:39:28wnmnkhI mean I believe Microsoft stop selling music as in WMA
08:39:41wnmnkhexcept 'rental' ones.
08:39:49kifoi have the large majority of my music in flac for archiving and v0 for listening
08:40:15wnmnkhspeaking of FLAC
08:40:20kifojust because it makes it easier to uhm
08:40:39kifoive bought from them before
08:40:43wnmnkhvery nice flac selections with rather reasonable price
08:40:47kifoonly to fill a request on a private though
08:40:56kifobut yea very high quality release
08:41:02kifonot only are the tracks perfect
08:41:06kifohuge resolution album art
08:41:28wnmnkhand there is
08:41:41wnmnkhvery nice classicial catalogs
08:41:49wnmnkhin flac
08:41:56kifoi mainly use torrents now, i buy some stuff if i like it but never in flac
08:41:59kifojust sorta uhm
08:42:16 Join mitk [0] (~mitk@
08:42:22kifokinda messed up to get music you like for free and just be like i dunno
08:42:26kifohit and run on it
08:42:30kifoif it sucks, then yea
08:43:11kifoive always thought that artists/labels/whoever should have donation programs set up
08:43:14kifoat least the artists
08:43:19 Join arbingordon [0] (~w@unaffiliated/arbingordon)
08:43:26*linuxstb points out this is #rockbox...
08:43:42kifolol oh yea! sorry.
08:43:49 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
08:43:50 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
08:43:53 Quit pixelma (Disconnected by services)
08:43:53 Quit amiconn (Disconnected by services)
08:43:54kifoI have about 40 thousand irssi windows open right now
08:44:00kifoim gonna go to sleep class in 5 hours
08:44:03 Quit kifo (Quit: sanhlrkcouii)
08:44:08 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
08:44:12 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
08:46:55 Quit Buschel (Ping timeout: 252 seconds)
08:48:50 Join kramer3d [0] (~kramer@
08:48:52 Quit kramer3d (Changing host)
08:48:53 Join kramer3d [0] (~kramer@unaffiliated/kramer3d)
08:55:37 Quit mt (Remote host closed the connection)
08:55:55 Join petur [0] (~petur@rockbox/developer/petur)
09:00:25 Join flydutch [0] (
09:07:12 Join DerPapst [0] (
09:07:20***Saving seen data "./dancer.seen"
09:11:30 Join Zagor [0] (~bjst@rockbox/developer/Zagor)
09:15:33 Quit RadicalR (Read error: Connection reset by peer)
09:17:09 Join [1]Lynx_ [0] (~Lynx@
09:19:21 Quit evilnick_ (Read error: Connection reset by peer)
09:20:19 Join LinusN [0] (~linus@rockbox/developer/LinusN)
09:20:56 Quit Lynx_ (Ping timeout: 276 seconds)
09:20:57 Nick [1]Lynx_ is now known as Lynx_ (~Lynx@
09:24:11JdGordongevaerts: hey, did you try out that patch?
09:33:39 Quit kramer3d (Quit: Leaving)
09:37:11 Quit FingerS (Read error: Connection reset by peer)
09:37:51 Join mischasworld [0] (
09:44:17 Quit mischasworld (Remote host closed the connection)
09:45:23 Quit DerPapst (Quit: Leaving.)
09:47:48 Join efyx [0] (
09:51:21 Quit shaggy-w ()
09:54:53 Join shaggy-h [0] (
10:01:25 Join Casainho [0] (~chatzilla@
10:06:49 Quit wnmnkh (Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100401080539])
10:12:38gevaertsJdGordon: oops, sorry. I forgot all about it
10:14:34 Join Buschel [0] (
10:15:53 Join DerPapst [0] (
10:16:03 Join wodz [0] (
10:18:28 Quit sfury ()
10:20:00 Quit arbingordon (Quit: `)
10:25:21 Join esperegu [0] (~quassel@
10:28:46 Quit Casainho (Ping timeout: 264 seconds)
10:30:55 Join DerPapst1 [0] (
10:30:59 Quit BHSPitMonkey (Remote host closed the connection)
10:32:50 Quit DerPapst (Ping timeout: 260 seconds)
10:45:55 Quit Buschel (Ping timeout: 264 seconds)
10:52:07 Quit hebz0rl_ (Quit: Ex-Chat)
10:57:52 Join kugel [0] (~kugel@rockbox/developer/kugel)
11:04:58 Join S_a_i_n_t_ [0] (S_a_i_n_t@
11:06:04 Quit S_a_i_n_t (Ping timeout: 252 seconds)
11:07:24***Saving seen data "./dancer.seen"
11:08:16jnsspardon for asking, is there an easy method of making the device work again when in recovery mode
11:18:16gevaerts"the device"?
11:19:01wodzmoving ADC readings to ISR works but only when unboosted :-/
11:23:16wodzhmm: values are pushed to the global variable so this must be the problem with debouncing algorithm
11:25:18kugelI think I finished my assessment task
11:37:23 Join dfkt [0] (dfkt@unaffiliated/dfkt)
11:42:39JdGordonkugel: you still have the stock linux install on the mini2440 yeah?
11:44:01JdGordonoh? which?
11:46:28 Quit wodz (Quit: Leaving)
11:46:48kugelJdGordon: I can upload a dd'd image tomorrow if you want
11:47:11JdGordonmore intereted in which instructions you followed and what you have installed?
11:50:07kugel mostly
11:50:47kugelexcept I used a single ext3 partition and jtag to flash uboot instead of tftp
11:51:10JdGordonok, so the same i followed (except I have mine booting from sd)
11:52:57 Quit phanboy4 (Ping timeout: 245 seconds)
11:53:42kugelI used the uboot of, since the one from their git doesn't build with gcc-4.4. I also boot off the sd, only uboot is on the flash
11:55:45Tornekugel: i've found int he past that parts of u-boot are broken on various gcc versions, as a warning; it's generally best to go with whatever the consensus is of a compiler to use :)
11:55:59Tornekugel: where by broken i mean "builds and then functions wrong"
11:56:20Tornee.g. u-boot built with gcc 4.x for Lubbock can't write to the flash, at all, but built with 3.x it can
11:56:24kugelthe prebuild one there works fine so far :)
11:56:49kugellooking at the uboot source is a nightmare also
11:57:55 Quit Lynx_ (Ping timeout: 264 seconds)
11:58:43Torneyes, u-boot is terrifying
11:58:50Torneusing someone else's binary is basically the winning path
12:03:03Torneor relying on openembedded to build it for you, if that doesn't work.
12:03:20TorneOE's build recipies are pretty reliable
12:05:41 Join Lynx_ [0] (~Lynx@
12:12:37kugelgevaerts and gsoc experts: how do I proceed with my assessment task? I think I'm finished (although the build system integration is still a bit of a hack)
12:13:56 Join anewuser [0] (anewuser@unaffiliated/anewuser)
12:14:14gevaertskugel: as soon as you think it's ok, post a comment on the webapp with information on where to find it. Since you seem to have some time left, maybe build system integration would be a good idea to work on though :)
12:14:30gevaertskugel: also don't forget to post a comment about available times
12:16:02CIA-5New commit by 03uchida (r25644): new text viewer
12:16:53jnsscan i get rockbox onto the device when it is in recovery mode?
12:17:31kugelgreat commit message....
12:17:52gevaertsjnss: that question is missing some quite essential information
12:18:02 Join M3DLG [0] (~M3DLG@
12:18:17jnssgevaerts, i have messed up partition tables on e260,
12:18:18kugelgevaerts: sounds good, thanks
12:18:29kugelI might also try to keep the sdl threads working :)
12:18:49 Join xavieran [0] (
12:19:12 Quit Ipsilon (Ping timeout: 252 seconds)
12:19:54gevaertskugel: that's probably pretty important if you want to commit this (which I'd assume is the goal), since we have to evaluate if it's actually a win first, and since you don't have priorities yet
12:20:52kugeluntil now I mostly wanted to finish what I have been asked for :)
12:21:19gevaertsjnss: you'll first have to use the standard recovery procedures. See . Don't be creative though, and if you don't understand some things ask first.
12:21:26kugelan assessment task is usually not something that aims for being commited, is it?
12:22:02gevaertskugel: well, yes and no. If it's something useful, I see no point in throwing it away and doing the same work again later on
12:22:32gevaertsSo I think it would be a pity to let this rot in some git tree and come back to it in a year to find that it's hard to merge
12:23:49gevaertsAnd the more you do, the more you impress the people who have to decide ;)
12:27:19kugelgevaerts: are you saying my assessment task isn't considered completed until I implement priorty?
12:27:41jnssit worked! restoring the bootloader fixed it, thanks gevaerts ;)
12:27:58 Join Luca_S [0] (
12:29:25gevaertskugel: no. I'd say that priority is probably needed before we throw out the SDL threads, which means that it should be possible to pick which to use at compile time before it goes into svn. I think neither of those is essential for the purposes of GSoC qualification though
12:29:29jnssi tried all methods on that list *except* restoring the bootloader.
12:30:31kugelgevaerts: ok, thanks
12:30:48gevaertskugel: I do think that it's probably a good idea to add the needed #ifdefs and things soon, while you still know what everything does
12:30:58gevaertsthat's up to you though
12:31:44gevaertskugel: oh, and the #define SDL_THREAD_* should probably be renamed in the pth thread source :)
12:32:53kugelgevaerts: of course. right now I'm at the point where I have a working thread implementation using gnu pth which is why I said I mostly finished the task (which asked me to port gnu pth without details). a bit of working needs to be done still to be actually ready to commit
12:35:40 Join piotrekm [0] (~piotrek@unaffiliated/piotrekm)
12:39:40 Join tomers [0] (
12:39:47 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
12:41:07tomersThanks god for uchida for splitting the horrific viewer.c file into several files, but the log message is very confuzing. Actually it is not true - this is not a *new* text viewer, rather it is a re-arrangement of existing one. Uchida should mention if he added any new functionality (I guess not)
12:41:51*tomers wish uchida would be on IRC
12:48:07kugeltomers: I recomment you reading the -dev ml :)
13:01:24pixelmaI agree with tomers wish regardless
13:02:12kugelso I do, see above
13:03:38 Quit JdGordon (Ping timeout: 276 seconds)
13:07:09 Quit kugel (Ping timeout: 264 seconds)
13:07:26***Saving seen data "./dancer.seen"
13:13:54 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
13:17:04pixelmabtw. tomers: I played brickmania a bit yesterday on my Ondio (so, very small screen) and it somehow felt less precise. When the ball was hitting the paddle it sometimes vanished halfway, sometimes not (actually more visible when it had to stick to it) - the ball is only 2x2 on that screen. In the beginning I also had an invisible paddle once (until ball was lost) and one time a only half-visible paddle (diagonal). I also wondered if it was always
13:17:04pixelma the case that a new ball power-up let the new ball appear from the bottom center, it seemed odd to me and so I thought new balls would appear from where you caught the power-up (not sure though)
13:18:15pixelmahmm... "vanished halfway"... I mean it looked like you could only see the upper half with the lower "inside" the paddle
13:19:32CIA-5New commit by 03uchida (r25645): rvert r25644
13:20:00tomerspixelma: I will check on Ondio simulator, but I'm not sure I will be able to reproduce this.
13:20:55 Quit anewuser (Quit: What do you know...THE WORLD'S first NTRQ (that's for NES/FAMICOM) tracking compo. Have powerpak? Try it out! Otherwise ROM IMAGE.)
13:22:13 Join ucchan [0] (
13:25:01ucchanI delete all files committed by mistake. very very sorry.
13:25:56gevaertsucchan: this happens to everyone once in a while. Don't worry too much about it
13:26:00tomersicchan: Relax, that's not a big deal :-)
13:26:23 Quit mirsal (Ping timeout: 276 seconds)
13:26:32tomerspixelma: Regarding the ball powerup, I didn't touch that logic. I wondered about this myself
13:26:57tomersucchan: That a good idea to split the viewer.c file
13:27:29pixelmatomers: ok... as I said, I wasn't sure
13:27:51ucchanOther files were not broken and it was really good.
13:28:31ucchancommited textviewer is old. then may not move.
13:28:33tomersucchan: i think that tv_ prefix to filename is not needed, since all these files are under the text_viewer directory.
13:28:57tomersucchan: Also I would s/textviewer/text_viewer/
13:29:40tomersucchan: yes, you must sync with viewer.c before splitting...
13:29:41ucchantomers comments thanks.
13:30:55ucchanI try delete "tv_" and change textviewer -> text_viewer.
13:33:46 Quit JdGordon (Quit: Leaving.)
13:34:12 Quit perfectdrug (Ping timeout: 265 seconds)
13:34:36 Join perfectdrug [0] (
13:35:08ucchanBecause this has been almost completed, I think that I can send to the tracker perhaps on the weekend.
13:36:37tomersuchida: I suggest the when committing the split, try to avoind having changes to logic (do it separately). If you must change some logic, please write it explicitly in the FS tracker or in the commit message
13:36:55 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
13:38:53tomersucchan: Have a look at "FS #11201 - Viewer: Bookmarks data gets deleted"
13:40:21 Join kugel [0] (~kugel@rockbox/developer/kugel)
13:40:44 Join mirsal [0] (~mirsal@videolan/developer/Mirsal)
13:46:08 Join n1s [0] (~n1s@rockbox/developer/n1s)
13:46:41 Quit TillW (Quit: this concludes our broadcast day)
13:47:18 Quit Xerion (Ping timeout: 246 seconds)
13:47:50ucchanI try confirm about FS #11201 on my environment.
13:47:53 Nick fxb__ is now known as fxb (
13:49:16 Join Xerion [0] (~xerion@
13:52:23 Join mt [0] (~chatzilla@
14:04:15 Join wodz [0] (
14:11:20 Quit ucchan (Ping timeout: 264 seconds)
14:15:37 Join TillW [0] (
14:19:00pixelmatomers: the half vanishing ball has probably to do with r25372 - with a 2x2 ball, the result will be visibly different
14:19:12pixelmaif I understand correctly...
14:23:35 Join Strife89|PalmTX [0] (~cstrife89@
14:28:17 Join ucchan [0] (
14:28:39 Join Schmogel [0] (
14:29:59ucchantomers, I confirmed FS #11201. this is not a bug. this is a correct.(reason: I wrote in the tracker)
14:41:19kugelintegration into the build system is done, now I have a shot at priority
14:42:45 Quit jordan` (Ping timeout: 252 seconds)
14:46:52 Quit ucchan (Quit: Leaving...)
14:49:41 Join Watermark [0] (
14:50:26Watermarkhello Linuxstb
14:50:42Watermarki was going to ask you a quick question if i may
14:51:19linuxstbWatermark: Go ahead
14:51:41Watermarkhow dfo you check the SDCFG Value? sorry if this has been asked before. me tired here
14:52:50linuxstbWhat do you mean by "check" ? You use it in tcctool (you need to edit the source code and recompile tcctool), and if it works, it works.
14:53:51Watermarktoo bad i have no friends around who would know how to do it
14:54:21Watermarkyou think you could show me how. since i have the S9 anyhow
14:54:46Watermarkjust like write up a Guide and send it to me whenever?
14:54:55 Join jgarvey [0] (
14:55:01Watermarkne more the happy to learn
14:55:27wodzWatermark: start from reading docs from wikipage
14:56:23 Quit n1s (Read error: Connection reset by peer)
14:57:06Watermarkwhat exactly would i search for?
14:58:56linuxstbWatermark: Firstly, you need to either install linux (it can be in a virtual machine), or install cygwin. But I don't know for sure if tcctool will compile with cygwin - I haven't tested that... So safer option is to install Linux.
14:59:03wodz and look for paragraph 'For developers'
14:59:24 Part LinusN
14:59:57linuxstbwodz: Watermark wants to compile/change tcctool, which is a PC applicatoin. So no cross-compilers etc needed.
15:00:43Watermarkwhich Linux do i need?
15:01:07Watermarki have ubuntu on Disc
15:01:52Strife89|PalmTXUbuntu is fine.
15:02:14Strife89|PalmTX9.10 is the most recent stable release.
15:02:52wodzlinuxstb: ok but working with patches, general compile concept etc. is still valid
15:02:59Strife89|PalmTXOur Linux guide was actually written with Ubuntu in mind.
15:05:05Watermarki am installing ubuntu within a Virtual System
15:05:27Watermarkbe aware i am new to Linux and i don't know many commands
15:06:05wodzehh, I moved adc scans to ISR and there is no performance gain at all :-/
15:06:19Strife89|PalmTXWatermark: Our Linux guide was written with that in mind. ;)
15:06:54linuxstbWatermark: After you get Ubuntu working, you'll need to install the "mingw32" package. e.g. type "sudo apt-get install mingw32" in a terminal (or use the GUI package manager)
15:07:04 Quit tomers (Quit: CGI:IRC (Ping timeout))
15:07:26linuxstbWatermark: Then you will need to download a copy of the Rockbox source code from SVN - the wiki should have instructions for that.
15:07:27***Saving seen data "./dancer.seen"
15:08:09 Join Buschel [0] (
15:08:12Watermarki will new open a Tabb for just me and you to make things easier
15:08:50linuxstbNo, please don't.... I don't have time to be your personal tutor.
15:09:07 Join DataGhost [0] (
15:09:08 Quit DataGhost (Changing host)
15:09:08 Join DataGhost [0] (~dataghost@unaffiliated/dataghost)
15:10:24Watermarki'm just wanting to do that becausde it'll be easier to obtain all the Info. without my going up to look for it
15:13:00 Quit kugel (Ping timeout: 276 seconds)
15:13:27 Join n1s [0] (~n1s@rockbox/developer/n1s)
15:14:07Watermarkstill installing...
15:18:03 Join dys [0] (
15:21:14 Quit wodz (Quit: Leaving)
15:22:12Watermarkso when do i need to have my S9 connected? just asking
15:23:40 Quit CGL (Quit: Saliendo)
15:30:36 Quit M3DLG (Quit: RAGE QUIT)
15:36:14Watermarkcurrently downloading package then i'll get SVN
15:41:02 Quit Buschel (Ping timeout: 252 seconds)
15:41:14 Quit Rob2223 (Quit: Rob2223)
15:42:53Watermarkhaving trouble installing SVN
15:44:02perfectdrugWatermark: sudo apt-get install subversion
15:44:53Watermarkthanks installing
15:45:05Watermarkok now what?
15:45:25Watermarko hold on
15:45:31Watermarkforgot a step
15:45:36 Quit Lynx_ (Quit: HydraIRC -> <- Now with extra fish!)
15:47:53 Quit piotrekm (Quit: piotrekm)
15:49:50Watermarkhaving trouble getting the source. says URL doesn't exist
15:51:01pixelmaare you following our guide?
15:51:17 Part Strife89|PalmTX
15:51:34Watermarkyes, currently reading "How to check out the source code"
15:52:32 Join evilnick_B [0] (~0c140464@rockbox/staff/evilnick)
15:53:15 Quit mitk (Quit: Leaving)
15:54:03Watermarkok so what do i do?
15:55:23pixelmawell, I would try if checking out SVN and building from a clean tree works (provided you have done all the other steps)
15:55:44TorneWatermark: you are typing something wrong, then
15:55:57Tornepaste the command you are using
15:56:38Watermark svn co svn:// rockbox
15:56:53TorneWell, our svn repository is working. :)
15:56:53Watermarksvn checkout svn:// -r 12345 rockbox
15:57:10Tornethe first one is right.
15:57:20Tornethe other is for a particular revision, and I doubt you want revision 12345 :)
15:57:44Tornewhat exactly does subversoin say?
15:58:13 Join Strife89 [0] (~Strife89@
15:59:08Watermarksays: svn: URL svn:// doesn't exist
15:59:10 Join MethoS- [0] (~clemens@
15:59:28 Quit SirFunk (Quit: Ex-Chat)
15:59:56gevaertsnetwork issues?
16:00:07Torneare you behind some kind of proxy/firewall, perhaps at work?
16:00:30Watermarki'm not at work. me at home and no, not behind any Friewall
16:01:18 Join kugel [0] (~kugel@rockbox/developer/kugel)
16:01:40pixelmasorry for not being helpful, I missed the "trouble" in your statement before
16:02:09Watermarko it's just the URL isn't working
16:02:18*GodEater would guess it's a networking issue with the VM that's just had ubuntu installed into it?
16:02:43Watermarkmaybe so
16:04:16Watermarki may just go ahead and buy a new S9. this is taking too much time
16:05:02B4gderthis? you just started!
16:06:10Watermarkyes but i keep running into issues and it's pissing me off a bit
16:06:21Watermarki'm not a Linux user here
16:06:34B4gderin a couple of weeks you've learned!
16:07:44Watermarktrue but how do wee know if this will even work? how do we know this will fix my Cowon S9?
16:08:05B4gderwe don't
16:08:56Watermarki just wish i never did what i did but it was for the noob users out there just in case they try to do it. now i can tell them not to
16:09:32Watermarktrying to figure out the sdcfg value is just too much hassle for me
16:12:01 Quit Watermark (Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100401080539])
16:12:30 Quit Strife89 (Quit: Restart.)
16:20:32 Join Strife89 [0] (~Strife89@
16:28:03kugelcan't we just offer source zips again?
16:29:34 Join liar [0] (
16:31:17B4gderhow would that help for this?
16:31:30B4gderhe would need network access for that as well
16:35:44kugelhe could download it on the host and transfer it manually at least
16:35:50 Quit Xerion (Quit: )
16:36:05gevaertswe do have source zips, just not the archived daily ones
16:36:16B4gderhe could svn checkout on the host too
16:38:22kugelgevaerts: none for the current build either
16:39:19gevaertshm, right. I was confused apparently
16:39:45kugelgevaerts: implemention priority seems a bit tricky, because kernel.c is involved there too
16:40:07CIA-5New commit by 03torne (r25646): Make builds with MAX_PHYS_SECTOR_SIZE defined (ipod video) work with larger sizes if the drive is sane. ...
16:40:18 Join Rob2222 [0] (
16:40:25 Join archivator [0] (~archivato@
16:42:01 Join toffe82 [0] (~chatzilla@
16:45:48kugelgevaerts: also I think we would need to use pth's mutex instead of ours, needing to introduce a typedef for struct mutex
16:48:17 Quit TillW (Quit: This now concludes our broatcast day.)
16:48:29 Join Llorean1 [0] (~DarkkOne@
16:48:56 Quit Llorean (Ping timeout: 268 seconds)
16:49:02 Nick Llorean1 is now known as Llorean (~DarkkOne@
16:49:05 Quit Llorean (Changing host)
16:49:05 Join Llorean [0] (~DarkkOne@rockbox/user/Llorean)
16:49:29 Join Blue_Dude [0] (~chatzilla@
16:50:33 Join mikroflops [0] (
16:51:32Blue_DudeNoticed that there is a leftover directory from Uchida's commit earlier. If I svn del it and then commit it, will that take care of it?
16:51:43 Quit Strife89 (Quit: Going to class.)
16:51:49Torneshould do
16:51:58kugelmaybe jhMikeS can give me info on this
16:52:24Blue_DudeHm. I'll try it.
16:52:53CIA-5New commit by 03Blue_Dude (r25647): Empty folder cleanup
16:53:04Blue_DudeSeems ok.
16:53:56 Quit mikroflops_ (Ping timeout: 258 seconds)
16:55:54 Join JohannesSM64 [0] (
16:56:52Torneamiconn: did you think of anything sensible to do re contrast on the mono ipods?
16:57:04Tornelinuxstb: how do we go about doing an ipod bootloader release?
17:01:13*kugel cries at prio_add_entry()
17:02:11 Quit JohannesSM64 (Ping timeout: 246 seconds)
17:03:00 Quit Zagor (Quit: Leaving)
17:06:35linuxstbTorne: I guess the first thing is to build the bootloaders - you do something like "make VERSION=v3.0" to set the version string - and make them available somewhere for people to test. I guess because it's been so long since a release, we should give them a proper test.
17:07:02linuxstbIn the past, the releases were quite close together, so testing just involved finding one person with that target to check it worked.
17:07:20Torneright. that's probably a good plan
17:07:28***Saving seen data "./dancer.seen"
17:07:35Tornedo you think it's worth trying to do something about contrast, btw?
17:07:57Tornei am indifferent since i am assuming that if thos eplayers were widely used and the new defaults were useless people would've complained
17:08:13linuxstbMaybe a test checklist would be useful - e.g. to make sure dual-booting the OF still works, and that the OF itself is still functional. I remember some strange bugs in the past where the OF would boot, but music wouldn't play for example...
17:08:40linuxstbSo there seems to be different LCD types on the 3G, requiring a different default contrast?
17:08:50linuxstbAlso, do your changes affect the nano2g?
17:09:13kugelTorne: what do you think of the idea to provide a emergency mode in the rockbox bootloader?
17:09:27Tornekugel: you mean include the usb stack?
17:09:31Tornekugel: i had considered that too
17:09:37Tornewe could do.
17:09:37kugelespecially on PP since the patcher allow complete removal of the of
17:09:51Tornerom disk mode still works..
17:09:56Torneit's not a complete removal :)
17:09:57linuxstbIt doesn't seem that useful on ipods, as that has the Apple disk mode in NOR...
17:10:02 Quit Blue_Dude (Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100401080539])
17:10:03Tornelinuxstb: i've not done anything to nano2g
17:10:17Tornelinuxstb: the reboot-on-cable-insert from fatal error thing could be done on nano2g though
17:10:22kugelI have removed it long ago, but I needed to resort to e200tool once
17:10:24Tornelinuxstb: using more or less identical code.
17:10:43linuxstbkugel: Don't confuse Sansas and ipods...
17:10:53 Join TillW [0] (
17:11:00kugellinuxstb: which is why I said PP
17:11:05Tornethe ipod bootloaders are not the PP bootloaders
17:11:10Tornethey are totally unrelated
17:11:15Torneipod doesn't use the normal pp bootloader code at all
17:11:34kugelI know
17:11:37linuxstbkugel: So what targets did you mean by "PP" ?
17:11:39Tornethe only one of my hcanges that affects anything other than PP ipods is turning on LBA48
17:12:05linuxstbkugel: And "the patcher" refers to which patcher?
17:12:15kugellinuxstb: PP ipods and sansas mainly
17:12:37kugelif recovering from a broken rockbox.mi4 is still dead simple on the ipods then it's probably not needed
17:12:46Tornekugel: lots of targets already include the usb mass storage mode in the bootloader
17:12:46linuxstbThere isn't a rockbox.mi4 on ipods...
17:13:03Tornekugel: so if the non-ipod PP targets would benefit from that it's easily done
17:13:09Tornebut for ipod it's probably irrelevant.
17:13:16Tornebecause of hte always-available rom disk mode
17:13:23Torne(which is now easier to boot into) :)
17:14:24Tornei pondered doing it but i decided that just making it easier to boot to disk mode on error was sufficient
17:14:42kugela rockbox'ish solution would be a bit cooler though :)
17:15:40 Join JohannesSM64 [0] (
17:15:48linuxstbI don't think it's straightforward, at least on ipods (I've no idea about the other PP targets). The ipod bootloader is running completely from IRAM (although using some SDRAM for bss) and doesn't enable interrupts.
17:16:04kugelsince our goal is to be a complete replacement we should also provide some recovery mechanism if possible instead of relying on the OF
17:16:21Tornekugel: if we were going ot be a complete replacement we would have a loooot further to go on ipod :)
17:16:35Tornewe rely on *loads* of hardware setup by the NOR bootloader
17:16:53kugelI wasn't aware of that
17:18:24 Join Boldfilter [0] (
17:21:13 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
17:22:17 Quit antil33t ()
17:25:50Tornekugel: well, I say that
17:25:58Torneit might not be loads, really
17:26:01Tornebut some, certainly
17:26:26Torneit would take some reasonable effort and probably further reverse-engineering of apple bootloaders for our bootloader to be installable in rom
17:26:45Torneand, unless someone thinks of a clever trick we don't currently know, this would cause faulty bootloader installations to brick the player
17:27:06Torneso i suspect most people would say that this is *not* desirable to pursue
17:27:26Tornethe flash bootloader on ipod is fairly quick; it doesn't take much time to get to the point where it hands us control.
17:28:28kugelnobody talked about removing the disk mode ;)
17:28:59Tornedisk mode is loade dby the NOR bootloader, though
17:29:08 Quit mirsal (Ping timeout: 246 seconds)
17:29:11Torneso if we replaced that with our own code, then a bug in our code would potentially render both rockbox and the rom disk mode unbootable
17:29:15Torneleaving your player bricked.
17:29:23kugelwell, I mean whatever in the NOR is
17:29:41 Quit Minataku (Ping timeout: 276 seconds)
17:30:06Torneso my point is, we don't want to tamper with the flash unless there's a really big advantage to be had thereby, because it's not a *bad* bootloader and it has a nice recovery optoin..
17:30:18Torneso why bother to provide our own recovery option as well?
17:30:36kugelI just corrupted my rockbox.mi4 on my e200 once and was bitter than the bootloader couldn't help me recover
17:31:00Tornei don't know how e200 works, so i can't really comment
17:31:13 Join FOAD_ [0] (
17:31:41kugelTorne: I just asked because IIRC the ipods have a similar ability to replace the OF as the e200 (where the e200 loses the OF provided recovery mode)
17:32:01kugelI didn't know the ipods have another easy way to recover
17:32:15Torneon ipod we install by either replacing or modifying the contents of a file on the firmware parittion with either the bootloader or a rockbox build
17:32:22Tornenumerous combinations work :)
17:32:34Tornebut in all cases the flash is untouched and you can still get the rom disk mode
17:32:46kugelthat's also how it's done on the e200 basically
17:33:02Torneright. but presumably the e200's flash loader doesn't have a usb recovery option?
17:33:08Torneor at least not a convenient one that's mass storage
17:33:53kugelthe e200 OF bootloader (I don't exactly know where it's stored, but I assumed in flash so far) has a recovery mode
17:34:51kugelthere's another mode which does not expose a UMS drive but you can recover from that with the e200tool
17:36:39kugelyou can replace the former with a rockbox binary (or patch it to load one from the firmware partiton, that's the default procedure) in which case you lose the recovery mode
17:44:06linuxstbTorne: The Sansas are similar to the ipods (they have a firmware paritiion), but the OF bootloader is stored in that partition as well. So easier to "brick" than ipods.
17:44:55kugelbut still almost impossible
17:45:19linuxstbThe boot ROM on the Sansas just provide a kind of "usb boot" facility, allowing you to upload code to run. So yes, very hard to destroy that...
17:45:39Torneok, so yeah. it's still unbrickable, but when you "brick" it it's more hassle to fix
17:45:53Tornewhereas the ipod's recovery mode is functionally equivalent to our own usb stack :)
17:46:23Lloreankugel: Basically, through the tools we provide it should be nearly impossible / impossible to remove UMS access from an iPod, and recovering should always be as simple as hooking it up in "disk mode", recovering an MBR, formatting, and having iPodpatcher install a binary of your choice, in the worst case.
17:46:44 Join stoffel [0] (
17:47:16gevaertsTorne: well, e200tool also provides the means to "brick" the things more solidly
17:48:08LloreanThough you could -wf an image that contained a replacement flash image if you really wanted to.
17:48:22LloreanBut that'd be similar to any other device - if you try to brick it all bets are off.
17:48:39TorneYeah, you'd have to overwrit ehte whole partition with one that had somehting invalid in AUPD
17:48:45Torneand the updat-eflash flag set to true
17:48:58Tornethis is very very unlikely to happen by anything other than a really determined intentional attempt
18:00:37 Quit JohannesSM64 (Ping timeout: 248 seconds)
18:02:11 Join einhirn [0] (
18:03:54 Join Rondom [0] (
18:04:48 Quit petur (Quit: *plop*)
18:05:20 Join domonoky [0] (~Domonoky@rockbox/developer/domonoky)
18:07:10 Quit TillW (Quit: This now concludes our broatcast day.)
18:08:27 Quit Rondom_ (Ping timeout: 276 seconds)
18:11:14 Quit DerPapst1 (Quit: Leaving.)
18:12:16 Join robin0800 [0] (
18:16:01 Join JohannesSM64 [0] (
18:16:02 Join halmi [0] (~netbook@
18:17:53 Join sbpaul [0] (~sbpaul@unaffiliated/sbpaul)
18:17:59sbpaulcan somebody remind me how to refresh my library on an ipod 4g
18:19:15Tornethe manual can
18:20:01sbpaulno time man
18:20:04sbpaulclass in 20
18:20:24Tornei don't have time to read the manual for you either
18:20:32sbpaulthen we're stuck i guess
18:21:50sbpaulthe manual is wrong
18:22:41sbpauloic n/m
18:22:42 Part sbpaul
18:22:51*Torne eyerolls
18:23:32Tornei was about to say, i bet it's not.
18:24:14kugelTorne: I wouldn't bet on that
18:25:47 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
18:26:44 Quit JohannesSM64 (Quit: WeeChat 0.3.2-dev)
18:27:32 Join JohannesSM64 [0] (
18:28:53pixelmasays the one who didn't manage to get the playlist viewer shortcut into the manual as long as it existed in its pure form ;)
18:29:17 Join mirsal [0] (~mirsal@videolan/developer/Mirsal)
18:30:21 Join dfkt_ [0] (dfkt@unaffiliated/dfkt)
18:31:14 Quit Luca_S (Quit: CGI:IRC)
18:34:10 Quit dfkt (Ping timeout: 245 seconds)
18:34:50 Join dfkt [0] (~dfkt@unaffiliated/dfkt)
18:36:16 Quit mirsal (Ping timeout: 264 seconds)
18:38:20 Quit dfkt_ (Ping timeout: 245 seconds)
18:46:15 Quit dfkt (Ping timeout: 245 seconds)
18:46:49 Join DerPapst [0] (
18:51:27 Join kramer3d [0] (~kramer@unaffiliated/kramer3d)
18:54:14 Join jordan` [0] (
19:01:26 Join dfkt [0] (dfkt@unaffiliated/dfkt)
19:02:25n1shmm, gcc4.5 is just about released, something to play with over the summer :) (and a testbuild for h300 i did last week went through almost without problems)
19:02:46n1sonly thing was that the nsf codec no longer fit iram
19:03:24n1sand a few more warnings about apparenlty possibly undefined behaviour
19:04:51n1sof course i haven't tested to see if it actually works either :)
19:05:31 Join xiainx [0] (
19:05:52bertrikn1s, nice, maybe the undefined behaviour warning are dormant bugs
19:06:11bertrikthey may point to bugs in rockbox I mean
19:06:28n1safter a brief glance they seem to be pretty hard to read code at least
19:06:41 Join mischasworld [0] (~quassel@
19:07:29***Saving seen data "./dancer.seen"
19:08:12 Nick S_a_i_n_t_ is now known as S_a_i_n_t (S_a_i_n_t@
19:09:00 Join komputes [0] (~komputes@ubuntu/member/komputes)
19:12:11 Join funman [0] (~fun@rockbox/developer/funman)
19:12:57 Quit FOAD (Quit: Lost terminal)
19:12:57 Nick FOAD_ is now known as FOAD (
19:16:53 Quit S_a_i_n_t (Quit: Restart)
19:19:15amiconnTorne: The idea is to implement hw version handling like the OF.
19:19:42amiconnThe problem lies in finding the respective code... :\
19:19:55Torneamiconn: is that likely to happen in the near-term future? suitable for a bootloader release in the next, say, month?
19:20:21amiconnDepends on my luck in RE. Maybe today, maybe never...
19:21:15amiconnWell, of course you could also try to find the responsible code :)
19:21:19Torneok. well, i'm gonna do some bootloader test builds soonish, so i guess we'll see what b&w ipod users think of it
19:21:25Torneand yeah, er, about that. ;)
19:22:09 Join S_a_i_n_t [0] (S_a_i_n_t@
19:23:01amiconnThis is the table:
19:23:27amiconnAs you can see, the table evolved not just in number of entries, but also changed format with newer versions of the ROM
19:24:12amiconnThe Mini G2 version exists twice in the ROM, and at some point I had a better disassembly which had a reference to one of these
19:24:24amiconnBut now I just can't find it again :(
19:24:52 Join Horscht [0] (~Horscht2@xbmc/user/horscht)
19:25:01amiconnI've commented all fields I was able to figure out
19:25:37amiconnWe can't use those values as-is, but we need to find out how the OF picks the line to apply
19:27:43 Join stripwax [0] (
19:28:26 Quit stripwax (Client Quit)
19:30:53pixelmaTorne: what are b&w Ipod users?
19:30:54Torneblack and white
19:30:58Tornewell, greyscale.
19:31:46 Quit bertrik (Remote host closed the connection)
19:31:47pixelmawhich Ipods are you referring to exactly? E.g. there are also greyscale 4G Ipods etc...
19:32:40 Join TheSeven [0] (~theseven@rockbox/developer/TheSeven)
19:32:44Tornepixelma: yes. all of them. 1g-4g, and mini 1g/2g
19:33:05Tornethe others are colour
19:33:16Tornethe greyscale ones all use the same driver
19:33:22Tornejust with slightly different settings
19:33:27 Quit DerPapst (Quit: Leaving.)
19:34:01pixelmadoes the difference also exist between greyscale and colour 4G Ipods?
19:34:26Tornether eis no problem with the colour LCDs, afaik
19:35:07 Quit GeekShadow (Read error: Connection reset by peer)
19:35:27pixelmaumm... silly me. For some reason I didn't remember it was about the LCDs
19:40:06 Join Xerion [0] (~xerion@
19:40:57n1shmm,yeah a[u] = ++a[u] | c; seems undefined to me
19:41:11 Nick funman is now known as funbot (~fun@rockbox/developer/funman)
19:42:30n1sthe question is just, what the intended behaviour is :)
19:43:05n1sa[u] = a[u+1] | c; a++; ?
19:43:10 Join DerPapst [0] (
19:43:52 Nick funbot is now known as funman (~fun@rockbox/developer/funman)
19:44:39 Join hebz0rl [0] (
19:44:45funmanperhaps the intended behaviour is undefined behaviour ? :)
19:45:06pixelmaLlorean: regarding your question in the unsupport build section about the included theme - I wondered about license issues there too. Isn't he responsible for this alone, or does the fact that he uses our code make us responsible too, or is the problem that it is well not hosted but at least given some publicity?
19:45:30pixelmain our forums, I mean
19:45:34n1sthere's also a enum comparison warning in mpegplayer, a weird "warning: case value ‘4294967295’ not in enumerated type ‘gamestate_t’" warning in doom
19:46:04n1sand another undefined warning in png
19:46:23kugelwhat are you toying with?
19:46:24 Join mirsal [0] (~mirsal@videolan/developer/Mirsal)
19:47:00funmangcc 4.5 ?
19:47:01Lloreanpixelma: It's more just if he's distributing it through our boards, it should be something he's legally allowed to distribute.
19:47:13n1sfunman: yep
19:47:15 Join Luca_S [0] (
19:47:18kugelI knew it :p
19:47:53pixelmahe already said so a while ago
19:47:55LloreanBasically, I think if someone wouldn't be allowed to post a theme to our site for copyright reasons, they shouldn't be allowed to bypass that by simply posting it somewhere on the forums instead.
19:48:03 Part Llorean
19:48:46funmanLlorean: makes perfect sense
19:50:56S_a_i_n_tRegarding the themesite...does anyone actually pay any attention if themes are reported as infringing a copyright?
19:51:08S_a_i_n_tI dunno how many times I've reported now...
19:51:12S_a_i_n_tbut, it's a lot.
19:51:22*n1s wonders what line 197 of doom/d_main.c is supposed to mean and why that is valid code
19:51:49n1sgamestate_t is an enum typedef from doomdef.h...
19:51:58 Join phanboy4 [0] (
19:56:22kugeln1s: doom has a lot of problematic code, enums are the reason it doesn't run under eabi
19:56:23n1suum, right it's a cast but still wth
19:56:28 Nick fxb is now known as fxb__ (
19:57:28 Quit phanboy4 (Read error: Connection reset by peer)
19:57:48n1syeah, this line should be harmless though, it will never be reached unless there is some major hackery in tere
19:57:50 Join watto [0] (~watto@
19:57:54amiconnWell, regarding the C standard it's not valid code
19:58:38amiconnTechnically it is though, since an enum is an int type of some sort
19:59:06pixelmaS_a_i_n_t: there are only a few theme site admins and I don't know how active they are (and what exactly they need to do with a reported theme)
19:59:07amiconnBut the fact that doom freely mixes enums with int is why it crashes in eabi
19:59:32amiconnThat's in fact not a problem of eabi itself, but rather just because eabi defaults to short enums
19:59:53amiconnIf you compile doom with old abi and short enums, it will crash too
20:00:07n1si wonder why gcc doesn't even war when an enum type is assigned an int value (that doesn't match any of the possible enums)
20:00:12kugellikewise it runs with eabi and long enums
20:00:34kugeln1s: because that's not a problem
20:00:46n1shmm, this one particular bug seems easy to fix
20:00:53kugeland it indeed does, in case it can know (it does in a few places in doom, but there's more to fix)
20:01:39kugelthe problem is when you write enum value at a location pointed to by a pointer of a different size
20:02:09n1skugel: an enum shouldn't have any values other than are defined for that enum type, so i think it's aproblem, also why not just use an int then?
20:02:17kugel*int_ptr = some_enum_val; <−− that's the bug
20:02:27kugeland doom hides many of these by casting
20:03:00n1syeah, that may be why it crashes but i think assigning "impossible" values to enum types is a bug too
20:03:03perfectdrugpixelma: do you know why we name the prev/next buttons in the manual sometimes Left/Right?
20:03:44kugelas you can see, doom makes excessive use of enums so nobody bothered fixing it yet
20:03:45perfectdrugin the SVG's i mean
20:04:20pixelmathat usually depends on the target and how the buttons are referred to in the text
20:04:28pixelmasome use prev and next
20:05:02 Quit stoffel (Remote host closed the connection)
20:05:36perfectdrugah ok because b0hoon asked me to change that so that he can more easily write the manual, so fair enough
20:06:15perfectdrugand it would be more consistent with the sourcecode he says so alright, thanks
20:06:54amiconnLeft/Right are called that because these are the only two buttons which consistently exist on all targets
20:07:33amiconnHmm, although the manual sometimes calls them different
20:07:55pixelmaperfectdrug: the image with the labels is exactly there to show how the buttons will be named throughout the manual
20:08:39perfectdrughm ok, and should I call it Prev or Previous
20:09:01 Quit kugel (Remote host closed the connection)
20:09:17pixelmaamiconn: the platform (keymap) files let you specify some text that will be placed if e.g. the \ButtonLeft macro is used in the source
20:10:33pixelmaperfectdrug: I think Prev is more common then (actually that's the only form I remember reading)
20:10:46perfectdrugok thank you
20:10:55 Join Buschel [0] (
20:14:21 Quit n1s (Ping timeout: 252 seconds)
20:15:48*gevaerts thinks someone should finish the meizu M3 port
20:15:54 Part Adubb
20:16:03*funman thanks gevaerts for volunteering
20:16:11perfectdrugcan someone reopen FS #10947 for this changes or should I email him the new files? (he gave me his email)
20:16:24gevaertsfunman: I can't! Someone else has it!
20:16:36funmangevaerts: hopefully the work of TheSeven will work on them
20:16:54gevaertsBut it's needed to disprove this idea that all players have left and right buttons :)
20:18:01 Quit xiainx (Ping timeout: 276 seconds)
20:18:40 Quit kramer3d (Quit: Leaving)
20:22:10 Join xiainx [0] (
20:27:50*TheSeven wonders what funman is referring to
20:28:06 Join n1s [0] (~n1s@rockbox/developer/n1s)
20:28:17TheSevenwhat's blocking the M3? Whimory?
20:29:25 Join pamaury [0] (~c2c7a50a@rockbox/developer/pamaury)
20:31:19 Nick YPSY is now known as Ypsy (
20:31:33funmanTheSeven: meizus use the same SoC than ipodnano2g?
20:31:45TheSevennot exactly, only something similar
20:32:35Buschelpixelma: could take a look at fs #11198? I would like to commit if there are no offenses
20:34:25*TheSeven wonders what to do about the nano4g
20:34:58TheSevenida just isn't good enough to decipher that bootloader mess
20:35:00 Quit jd (Read error: Connection reset by peer)
20:35:08 Quit JohannesSM64 (Quit: WeeChat 0.3.2-dev)
20:35:15n1sTheSeven: looking into the n4g+
20:35:16 Join jd [0] (
20:35:17 Quit jd (Changing host)
20:35:17 Join jd [0] (~jd@Wikipedia/HellDragon)
20:35:23n1seh s/+/?
20:36:05TheSevenyes, but it doesn't look good at all
20:36:17n1show come?
20:36:46TheSevenapple went totally crazy when designing the bootloader for that one
20:37:01n1slooking for a better way to load custom code?
20:37:30TheSevenno, i'm looking for a way to understand the disassembly, to figure out how to use the hardware
20:38:08Luca_Syou already have lcd from what it seems from the screenshot on the linux4nano wiki?
20:38:24TheSevenusb, lcd, backlight, but not much mure
20:39:37TheSeventheir bootloader consists of 43 little EXE files (yes, relocatable PE executables), which use guid-based lookup tables that are created dynamically during runtime (the modules are registering themselves) to access driver functions
20:39:56 Join bmbl [0] (~Miranda@unaffiliated/bmbl)
20:40:02TheSevenfollowing xrefs is practically impossible that way
20:40:36 Quit shai (Quit: Leaving)
20:47:41 Quit flydutch (Quit: /* empty */)
20:49:22n1scould it be run in an emulator or something?
20:49:50pixelmaBuschel: consistency and correct use is nice, so I like the kHz etc. changes. The added space looks a bit odd to me but I don't know how it is correctly (or most commonly used) in British English - maybe a native speaker could comment?
20:50:06n1salso no sensible disasm i could look at then...
20:50:11TheSevencoldboot attacks have also revealed a bunch of things
20:50:43 Quit einhirn (Read error: Connection reset by peer)
20:51:38n1sis what we know in the wik (rb or l4n) or someplace else?
20:51:42Buschelpixelma: you are talking about the spaces between numbers and the unit?
20:51:57pixelmaBuschel: yes
20:53:03Buschelpixelma: I have browsed wikipedia a bit (e.g. for SI). the spacing between number and unit is commonly used.
20:53:24funmanTheSeven: what is the SoC inside nano3g and nano4g ?
20:53:54TheSevennano3g and all ipod classic generations: s5l8702, nano4g: s5l8720
20:57:50 Quit funman (Quit: free(random());)
21:04:12*gevaerts points more people at FS #11175
21:07:31***Saving seen data "./dancer.seen"
21:09:36n1sgevaerts: that bug is just too weird, also is it happening in the sim?
21:09:52gevaertsn1s: I've not managed to reproduce it in the sim, no
21:09:59gevaertsit's real though
21:10:40gevaertsCenterArt v2.0 has this fancy WPS playlist viewer, which might be related
21:10:52 Part watto
21:12:10n1smight be
21:12:53n1smaybe one approach could be to eliminate as much of the wps as possible to see which elements trigger this?
21:13:07 Quit esperegu (Ping timeout: 246 seconds)
21:16:31 Quit xiainx (Ping timeout: 276 seconds)
21:18:07*gevaerts gets confused by that wps
21:18:08 Join grndslm [0] (
21:18:28 Join CGL [0] (~CGL@
21:18:39gevaertsAre you allowed to have multiple %Vl tags with the same identifier?
21:19:14gevaertsHow does that work then?
21:19:27gevaertsit shows all of them?
21:19:47pixelmathey'll all show up at the same time (as defined by THE RESPECTIVE %Vd)
21:19:59pixelmasorry dor caps
21:20:20gevaertshm, that's useful to know
21:21:13 Join xiainx [0] (
21:22:46 Join esperegu [0] (~quassel@
21:23:28gevaertsIt seems to be the %Vp tag indeed
21:23:40gevaertsremoving it makes things work
21:25:43 Quit toffe82 (Read error: Connection reset by peer)
21:28:46 Nick Ypsy is now known as YPSY (
21:32:48pixelmabesides the center viewports (I mean the two "a"s and the two "b"s are also overlapping
21:33:35pixelmato clarify: the second "a" overlaps the first "a"
21:33:57gevaertsoverlapping viewports shouldn't cause buffering/playback issues though
21:36:30n1sTheSeven: could there be anything useful that might not be so crazy in the iphone OS for ipod tough 2g to disassemble to figure out the s5l8720?
21:37:36TheSevenbut to work out things like the flash memory controller setup, we should really use code belonging to that device
21:38:03TheSeventhe best way would probably be to write some script that tracks down all those calls between the efi modules automatically
21:41:08n1si still hope to be able to help but this stuff is still out of my league :/
21:44:09CIA-5New commit by 03Buschel (r25648): Consistent use of Hz, kHz, dB, dbFS, ms, s and min. Add spaces between number and unit −− e.g. '5 dB'.
21:44:36 Quit xiainx (Ping timeout: 248 seconds)
21:44:41Buschelpixelma: if the spaces should be a problem (what I do not think), we can easily exchange them again.
21:45:46gevaertsBuschel: shouldn't those be non-breaking spaces?
21:46:24 Join toffe82 [0] (~chatzilla@
21:47:22pixelmathat would be important I guess (didn't think of it so far though)
21:47:55Buschelhmm, checking...
21:50:08pixelmawhat do you check? Just asking because I'm a bit irritated by this answer :)
21:50:11 Join xiainx [0] (
21:50:24Buschelchecking how to add those and test it
21:50:52pixelmaah, ok
21:50:55BuschelI did not think of adding those either
21:52:29 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
21:53:15 Join TillW [0] (
21:53:34 Join Adubb [0] (~Aldubuc@
21:57:01 Join cst [0] (
21:57:55 Quit cst (Client Quit)
22:00:19 Quit n1s (Quit: Lämnar)
22:02:57pixelmagevaerts: I'm a bit confused about this playlist viewer viewport and its handling, not sure I can explain it though. Do you know if %Vp works the same as viewports or the same as elements you can put inside viewports? If the former. it is always there, just covered by album art - could this be a problem?
22:03:12 Join petur [0] (~peter@rockbox/developer/petur)
22:03:40pixelmaI lost overview about the most recent additions to WPS/SBS tags
22:04:27 Quit xiainx (Ping timeout: 264 seconds)
22:04:30 Part Adubb
22:04:46gevaertspixelma: I don't know actually, but I don't have album art on that player
22:04:57pixelmagevaerts: what happens if you make the them non-conditional?
22:05:29pixelmajust keeping the playlist viewer part
22:08:12pixelmaah hmm, looking at its parameters though it's probably more an element you can put inside the viewport
22:08:41pixelmathe "%V" it it confused me though
22:09:44 Join xiainx [0] (
22:11:46 Join lpereira [0] (~lucien@
22:13:29pixelmagevaerts: maybe it's still an interesting test - gettung rid of the "a" viewports and keeping one of the "b"s and the conditional with that handles displaying them
22:14:37 Quit mischasworld (Ping timeout: 240 seconds)
22:14:43pixelmamaybe not, I'm currently confusing myself :\
22:15:34gevaertsI'm currently testing with the most minimal %Vp-based theme I can think of
22:16:55 Join fml [0] (
22:17:50fmlgevaerts, Buschel: the spaces between the number and the unit *should* be non-breakable, i.e. a '~' should be used instead of just space.
22:18:16pixelmathat would be my approach too which reminds me about the thread in the forums about %C in two different viewports to get the art into different positions under certain circumstances. I tested myself and it really didn't work
22:18:35pixelma^ gevaerts
22:18:41 Quit CGL (Ping timeout: 276 seconds)
22:19:24Buschelfml: just preparing this
22:19:52gevaertsWell, my entire wps is now "%Vp|0|%pp - %it|%pp - %fn|", and it still happens
22:20:11gevaertsSo I'm prepared to state that conditional viewports are not involved
22:20:27fmlBuschel: there is also the latex package SIunits, but that would make typesetting units even more complicated IMO
22:22:58 Quit fml (Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100401080539])
22:23:14*gevaerts thinks that this is a really impressive bug
22:23:17pixelmaI wouldn't like adding more dependencies unless there is a good reason to
22:26:56pixelmagevaerts: did you try whether there are differences depending on which info you let it display?
22:28:03pixelmaI'm thinking of the playlist position especially as there was (maybe still is) an issue with it advancing earlier than the rest on track changes etc.
22:28:30pixelmasomething like this anyways
22:30:17gevaertspixelma: I first tried with the line from the centerart theme, and then with the line from the CustomWPS page. I tried the latter with both 0 and one as the starting position. No difference
22:31:01pixelmaI meant taking out the %pp
22:31:15pixelmaor other tags
22:32:03 Quit n17ikh (Ping timeout: 264 seconds)
22:32:28pixelmaCenterArt doesn't have it... there goes that idea :\
22:34:39*pixelma always wanted the next tracks' playlist position available (instrad of some "Next" etc.) and sees that this would be a means for it... even for hwcodec
22:35:08 Join M3DLG [0] (~M3DLG@
22:35:53 Join n17ikh [0] (
22:37:15 Quit archivator (Quit: Leaving)
22:38:08 Quit perfectdrug (Quit: Leaving.)
22:38:23CIA-5New commit by 03Buschel (r25649): More work on consistent usage of units in the manual. Use non-breaking spaces if applicable.
22:43:23pixelmaBuschel: I don't understand the change to N/A in appendix/config_file_options.tex although I see there are different units (ms and s)
22:44:17*pixelma fetches the Ondio
22:44:54Buschelpixelma: other setting with "off" and different units (ms, s) also use N/A.
22:45:53gevaertseven more fun
22:46:08Buschele.g. skip length, peak meter, ...
22:46:37 Quit xiainx (Ping timeout: 276 seconds)
22:47:13pixelmait's a bit weird, the settings list is "off", "20ms" and then all the seconds values don't show the unit
22:47:51Buschelwe shall define a minor set of rules for such settings
22:48:13pixelmaand the settings don't have a space between number and unit
22:48:56Buschelyes, that's intentional. the config file uses the setting without spaces -> e.g. "5min"
22:49:26 Join kugel [0] (~kugel@rockbox/developer/kugel)
22:51:02 Join xiainx [0] (
22:55:18 Join Kitr88 [0] (~Kitr88@
22:55:49 Quit M3DLG (Ping timeout: 268 seconds)
22:57:34 Quit Kitar|st (Ping timeout: 240 seconds)
22:57:39 Join Kitar|st [0] (Kitr88@
22:59:54 Quit Kitr88 (Ping timeout: 260 seconds)
23:00:43gevaertspixelma: can I stop simplifying the test case when I get data aborts instead of strange playlist shuffling?
23:01:23 Quit xavieran (Ping timeout: 248 seconds)
23:03:03pixelmaprobably. I guess I should try on hwcodec but I'm too tired.
23:03:33gevaertsyes, same here...
23:03:49pixelmadid you try on different arcitectures, maybe coldfire?
23:04:11gevaertsnot yet. Maybe tomorrow...
23:04:12pixelmait sometimes gives different results
23:04:15 Quit efyx (Quit: Quitte)
23:05:19gevaertsThe data abort is in wps_parse(), so I'll leave it to those who know that code for now
23:06:14*gevaerts dislikes the reset settings button on ipod video
23:07:35***Saving seen data "./dancer.seen"
23:11:08pixelmasomething completely different - could file system errors result in completely trashing the file system if files are written? I'm still trying to find out what happened to my Ondio recently and remember that not long ago I was affected by the bug with file system errors when deleting dirs and I ran disktidy quite a bit recently. I don't *know* if there were any errors before, just asking if there was a possibility
23:12:08Buschelis it intentional that the iPods manuals are using the naming Ipod instead the correct iPod? The same error is contained for Itunes and iTunes.
23:12:29pixelmagevaerts: thanks, that could explain it
23:14:50 Join xavieran [0] (
23:15:44pixelmaBuschel: the discussion comes up every now and then. There was a convention that was agreed upon to just capitalise names. The reason was that Iriver changed their own spelling and just not caring about manufacturer's conventions makes the manual independent from that
23:17:12pixelmait's probably still controversial about Ipods
23:17:28Buschelpixelma: 'iPod' looks much more pleasing to my eyes.
23:17:54Buschelpixelma: is it agreed to keep it the way like it is? I have a patch at hand ;o)
23:18:45gevaertsBuschel: well, since it was a conscious decision, I guess it needs a consensus to change :)
23:19:24BuschelI wondered if the reason for the wrong spelling is some legal issue...
23:19:26pixelmaI personally prefer Ipod - for independency (and personal dislike of iThings I admit
23:21:26pixelmaat least I wouldn't call it "wrong" spelling
23:21:57Buschelwell, for me the 'Ipod' spelling is irritating. But that is personal taste...
23:23:31Buschelpixelma: Apple would have another opinion ;)
23:24:05 Quit alexbobp (Ping timeout: 252 seconds)
23:24:36 Join alexbobp [0] (~alex@
23:24:50 Quit bmbl (Quit: Bye!)
23:25:53gevaertsBuschel: If we agreed with Apple's taste, we wouldn't be having this discussion at all ;)
23:29:56Buschelat least the decision will be to either a) capitalize all products and companies names (and −− imho −− loose some product identification) or b) use the appropriate spelling, if someone keeps it up-to-date. I would prefer b) and letting mr. someone do the work.
23:31:23 Quit liar (Ping timeout: 260 seconds)
23:31:31Buschelmr. someone = me (for iPods :)
23:34:10 Join hd [0] (
23:34:10 Quit hd (Changing host)
23:34:10 Join hd [0] (~jd@Wikipedia/HellDragon)
23:36:33 Part lpereira
23:37:26 Quit jd (Ping timeout: 264 seconds)
23:39:13CIA-5New commit by 03b0hoon (r25650): Packard Bell Vibe 500: change a not so lucky keymap for the bookmark delete.
23:44:50 Quit Luca_S (Quit: CGI:IRC)
23:44:52 Join liar [0] (
23:45:07 Quit mirsal (Ping timeout: 276 seconds)
23:46:15 Quit liar (Read error: Connection reset by peer)
23:46:28 Quit domonoky (Read error: Connection reset by peer)
23:50:20 Quit pamaury (Ping timeout: 248 seconds)
23:50:50 Quit Buschel ()
23:52:03 Quit jgarvey (Quit: Leaving)

Previous day | Next day