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 2009-02-19

00:00:19gartralumm, the page at says no plugins are loaded too the Ipod 3g >.> is this correct, or a mistake?
00:00:32 Join itcheg [0] (i=62db4767@gateway/web/ajax/
00:01:25rasherSpeaking of the plugin pages, is the intention to remove them and instead refer to the manual?
00:01:28 Part akur
00:06:32 Quit bluebrother ("leaving")
00:07:29 Join Thundercloud [0] (
00:10:18 Join amiconn_ [50] (n=jens@rockbox/developer/amiconn)
00:10:18 Quit amiconn (Nick collision from services.)
00:10:18 Quit pixelma (Nick collision from services.)
00:10:19 Join pixelma_ [50] (n=pixelma@rockbox/staff/pixelma)
00:10:31 Nick pixelma_ is now known as pixelma (n=pixelma@rockbox/staff/pixelma)
00:10:36 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
00:20:04JdGordonrasher: yeh, i tihnk that was the plan
00:20:34rasherI guess keeping the PluginIndex matrix is still valuable
00:21:46LloreanI don't think we want to remove *all* plugin pages
00:21:56LloreanThe manual currently refers to some of them for more complex stuff (video encoding)
00:22:03 Quit dfkt (Read error: 110 (Connection timed out))
00:22:16LloreanBut we could drop them with say "VideoEncodingForRockbox" and "DictionarySetup" type pages instead of "PluginDict" etc
00:22:38 Quit ender` (Read error: 60 (Operation timed out))
00:23:03rasherThat seems reasonable
00:23:12 Join midijunkie [0] (
00:25:02pixelmaI agree about the index page
00:30:26rasherThe index page should be kept (more or less as-is), and pages that are referenced in the manual should be split out
00:30:32***Saving seen data "./dancer.seen"
00:31:06*rasher discovers there are no screenshots of his ancient Logo plugin
00:32:32 Nick fxb__ is now known as fxb (
00:32:45pixelmathe index page is generated from the PluginBlah subpages and their checkmark list in the end, so you probably can't just delete PluginBlah and keep the index as-is. Or do I overlook something?
00:33:33rasherNo, you don't
00:33:51rasherIt's a matter of re-creating the table staticly
00:35:17 Join Beta2K___ [0] (
00:38:57 Quit mcuelenaere ()
00:39:24 Join cool_walking_ [0] (i=cb3b81c3@gateway/web/ajax/
00:40:18 Join moos [0] (i=Mustapha@rockbox/staff/moos)
00:41:07*rasher detects an americanism in the jpegviewer
00:41:46gevaertsThrow it out!
00:42:22rasherI will!
00:44:48rasherThe settings seem to use Grayscale as well :\
00:45:21rasherNo sense in changing that though, I guess.
00:45:45gartraltoo bad there isn't a "mass replace X with Y" function too TWiki...
00:45:58LloreanIt'd be too dangerous anyway.
00:46:09LloreanYou get it replacing examples in guidelines of "what not to do" and such like.
00:46:20gartralyou can limit it too Admins Only
00:46:40LloreanDangerous in the sense of "even someone who's thought it through can very easily make stupid mistakes and not realize it for years"
00:49:03 Quit Beta2K__ (Read error: 113 (No route to host))
00:54:09 Part LinusN
01:04:38 Join amiconn_ [50] (n=jens@rockbox/developer/amiconn)
01:06:31 Quit itcheg (" ajax IRC Client")
01:08:39 Join Laemhax [0] (
01:10:04 Quit amiconn_ (".")
01:15:55pixelmawt... ACTION_LISTTREE_PGUP/PGDOWN is mapped in 3 different ways for the gigabeat F/X in context_list. I only looked at this because BigBambisaid when reworking the bookmark chapter that the delete button combo doesn't work there and I see that the list context is chained to the boobmark context and both want to use the "A" button in different ways...
01:16:18*pixelma blames time for typos
01:17:07 Join sharperguy [0] (
01:17:24sharperguydid anyone ever port python to rockbox?
01:17:35rasherNo. No one did.
01:17:39pixelmathe question is how to best fix this -
01:18:13rashersharperguy: Doesn't seem like a very good fit if you ask me. There's an initial port of LUA in the tracker though
01:22:21 Join CaptainKewl [0] (
01:23:07 Join isthatall [0] (n=guy2@unaffiliated/isthatall)
01:23:14*rasher needed to install texlive-latex-extra in addition to the packages listed on ManualHowto on Debian
01:24:02rasherAlso, evince doesn't take kindly to filenames with a colon in them, it seems. Odd.
01:25:14Unhelpfulsharperguy: if there's something you want a plugin to do, it will probably be easier to write a plugin to do it than a plugin to interpret a python script to do it. this is likely true even if you want ten or one hundred new plugins. :)
01:25:21gartrallua in rockbox?
01:27:11 Join yhuang [0] (n=yhuang@unaffiliated/yhuang)
01:27:50Unhelpfulgartral: yes, search in flyspray for it. no idea how out-of-sync it might be, at this point.
01:29:06*gartral has only seen LUA in two places; 1) cell phones, and 2) Garry's Mod >.>
01:29:51 Join MxxCon [0] (
01:30:35MxxConhey folks, do you know how/if hackable Slacker player is?
01:30:40 Quit Thundercloud (Read error: 104 (Connection reset by peer))
01:32:29UnhelpfulMxxCon: this channel is for rockbox in particular. the front page at lists supported players, and to my knowledge, there's no work done at all toward running rockbox on Slacker...
01:34:41MxxConyes, i'm aware this channel is for rockbox. just sincey ou guys are emerged in this enviroment, just thought somebody might know or directly me where to
01:35:21 Join akur1 [0] (
01:36:40 Part akur1
01:38:44gartralit would be nice if disk tidy could include the amount of space freed by the clean up >.>
01:42:22 Join Jetoni [0] (n=4d1cbad6@gateway/web/cgi-irc/
01:43:03JetoniI'm getting continous data abort at 000090A4 (0) on my sansa e200 with the latest build with some mp3 podcasts
01:43:08 Quit toffe82 ("ChatZilla 0.9.84 [Firefox 3.0.6/2009011913]")
01:43:21pixelmakadoban: I just found a few tabs in a diff of my tree which is still containing your goban overlay patch - in the goban make file. Just a minor minor nitpick in case you have a minute...
01:43:38Jetonii've re-installed Rockbox, no help.
01:44:09pixelmaJetoni: sounds like to me
01:44:09Jetonihere's the mp3 file that causes the data abort
01:45:09Jetonii believe so , not sure thogh.
01:45:37JetoniBut this happens mostly with large files(podcasts), maybe the buffer fills in.
01:45:37 Part MxxCon
01:45:45gartralwhat it was for me, was the media was encoded badly, rarly, VBR fix fixed it
01:46:17rasherpixelma: Makefiles are supposed to have tabs afaik
01:46:28gartralhow far back does the calender plugin go for the E200s?
01:46:49 Quit midijunkie (Read error: 104 (Connection reset by peer))
01:47:08pixelmarasher: thanks for the info
01:47:20 Quit JdGordon (Read error: 110 (Connection timed out))
01:47:34Unhelpfulpixelma: yeah, that's a make thing, it *insists* on one-tab indents for rules, i believe.
01:47:57pixelmaok, nevermind then kadoban :)
01:48:37 Quit Laemhax (Remote closed the connection)
01:51:39 Quit isthatall (Remote closed the connection)
01:51:59JetoniAlso, when i start a podcast or a song, the playback "stutters" after 2-3 seconds.
01:53:20gartralyea, i have the same beavior, ive noticed that the buffer tends too underflow sometimes, and doesnt load enough data from the disk to keep going while it buffers other songs >.>
01:53:56 Join BHSPitMonkey [0] (n=stephen@unaffiliated/bhspitmonkey)
01:54:17FlynDiceWhere do bugs for unsupported targets(ie sansa v2's) get tracked/listed. Is this simply through forum posts or are wiki and flyspray involved?
01:57:02ZagorFlynDice: we don't track bugs for work in progress.
01:57:14gartralhuh, found a strange one, the game plugin spacerocks saves its highscores as Astrorocks.hs in the /.rockbox//rocks/games dir >.>
01:57:42Zagorthe very definition of "work in progress" is "does not work yet", i.e. "has bugs and/or missing functionality"
01:58:43FlynDiceZagor: I understand that so is it just forum posts and changes to the wiki then?
02:00:57ZagorFlynDice: I hate to be blunt, but nobody is interested in bug reports on code we already know is not working.
02:02:30 Quit moos (Read error: 110 (Connection timed out))
02:04:33gartralsoduko prevents you from leaving if you hve all spaces filled, but the puzzle isnt solved >.>
02:05:13gartralactually, it prevents you from doing anything but record the numbers >.>
02:05:58 Quit Jetoni ("CGI:IRC")
02:07:32rasherpixelma: Any thoughts on where to put albumart info? An appendix? My first thought was to put it under WPS, but that wouldn't make sense for targets that only have AA in PF
02:09:44Unhelpfulthoughts on adding resize to buffer_alloc: on resize down, obviously, just chop the end f the block, and add a proper head to the new freed block. on resize up, check if a free block follows, and expand into it, else compact and try to move the block to end of allocation, freeing the old block, else (ugh) memmove everything after the block, move alloc_end, and walk the chain of following blocks to update their handle pointers
02:22:43 Join isthatall [0] (n=guy2@unaffiliated/isthatall)
02:28:04 Quit Zagor ("Clint excited")
02:28:06isthatallhmm, ok, I'm not a linux noob, but I just want to delete, the .rockbox folder and it won't let because it's read-only. I have tried to chmod it but it doesn't allow that either, there are weird filenames in there from corruption I'm guessing, filenames with crazy characters, etc... I've tried to chown the folder to have absolute control over it, nope. the rbutil won't remove rockbox either, any ideas?
02:28:42krazykitisthatall, fsck the filesystem
02:28:59isthatallhm, didn't even think about that
02:30:33***Saving seen data "./dancer.seen"
02:33:38isthatallno go, bad superblock
02:33:49isthatalltried alternate
02:41:37isthatallif I gparted it can I nuke the 20MB 'unknown' partition without bricking it?
02:43:20isthatallthought not
02:44:11isthatallI'm wondering if the permissions got freaked out by having it plugged into 2 separate ubuntu boxes, one at work, one at home, both with the same username.
02:45:13Unhelpfulisthatall: no, that actually doesn't even make sense
02:46:14isthatallwell, I'll think about it, I gotta go make some stuffed crab but I'll be back later, troublesome stuff! thanks.
02:50:51 Join help [0] (n=4cf02476@gateway/web/cgi-irc/
02:51:12helphey ne one here to help?
02:51:22 Nick help is now known as Guest34723 (n=4cf02476@gateway/web/cgi-irc/
02:52:06 Quit Guest34723 (Client Quit)
02:52:26 Quit n1s ("Lämnar")
02:53:50gartraldont ask to ask, just ask, if someone knows the answer, we'll share, ussually
03:02:54 Nick fxb is now known as fxb__ (
03:04:23 Join lymeca [0] (
03:08:29 Quit faemir ("Lost terminal")
03:26:28 Quit parafin (Read error: 60 (Operation timed out))
03:28:00 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
03:43:08 Quit Barahir (Read error: 60 (Operation timed out))
03:43:43 Quit Beta2K___ (Read error: 113 (No route to host))
03:48:47 Part gartral
03:52:44 Join tvelocity[away] [0] (
03:53:31 Quit tvelocity[away] (Client Quit)
03:58:26 Join Beta2K_ [0] (
03:59:36 Join JdGordon [0] (
03:59:44 Quit tvelocity[a] (Read error: 60 (Operation timed out))
04:05:28 Join TNT [0] (
04:05:37TNThello can i have help?
04:06:08krazykityes, but we don't have crystal balls, so you'll need to be more specific in your questions
04:06:34cool_walking_don't ask to ask, just ask. the guidelines (linked in the channel topic) say this.
04:06:35TNTokay thats fine. i'm trying to install rockbox on my ipod but the bootloader will not install.
04:06:46krazykitwhich ipod?
04:06:48TNTi have a 5g ipod video
04:07:01krazykitare you sure it's a video and not a classic?
04:07:16TNTyes. i've ran rockbox and linux before but uninstalled it.
04:07:24TNTand i want it back lol.
04:07:47cool_walking_Are you using Rockbox Utility? Because versions 1.0.9 and 1.0.8 have a bug where they can't install the bootloader on Sansas and iPods.
04:08:21TNTi tried the rockbox utility and it didnt work then i tried ipodpatcher and it didnt work.
04:08:59cool_walking_"didn't work"? error message? computer exploded?
04:09:38TNTlol it says....
04:10:33TNTerror reading from disk: data error (cyclic redundancy check).
04:11:03cool_walking_Do a scandisk.
04:11:22TNThow do i do that?
04:11:48cool_walking_What OS are you on?
04:11:56TNTWindows XP
04:12:15 Join apaininth [0] (
04:12:42cool_walking_Though judging from the error, your disk is physically broken.
04:13:04TNTthe hell?
04:13:16TNTi've never had problems before and my ipod works fine.
04:13:42cool_walking_The only thing I've ever seen that message for was scratched CDs.
04:14:51TNTdo you think maybe having videos on my ipod that have stopped working is part of the problem?
04:15:06TNToh and im on scan disk, what boxes should i check before i start
04:15:16cool_walking_Check both of them
04:15:36TNTokay scanning
04:15:41cool_walking_It'll take quite a while.
04:16:26TNTsays i need to restart computer
04:16:29 Quit apaininth (Remote closed the connection)
04:16:33cool_walking_it shouldn't
04:16:35 Join AndyI [0] (i=AndyI@
04:16:50 Join apaininth [0] (
04:16:52cool_walking_unless something (like iTunes) is still accessing your iPod
04:17:02cool_walking_or you've got the wrong drive
04:17:31TNTdo ctrl+alt+delete and end processes?
04:17:52 Quit apaininth (Client Quit)
04:18:23 Join apaininth [0] (
04:18:25cool_walking_Yeah, check for anything apple-ish. Also try a program called "unlocker", which should tell you if anything's accessing the disk. By the way, restarting won't actually work, so don't do that.
04:19:06TNTits sayin "needs exclusive access to some of the files"
04:19:17cool_walking_What's saying?
04:19:25TNTthe disk check
04:19:27 Quit apaininth (Client Quit)
04:19:45 Join apaininth [0] (
04:20:25saratogathis probably isn't the correct place to learn how to use windows
04:20:35cool_walking_Are you running as an administrator?
04:20:44saratogayou could try a tech support website or even rockbox-community
04:20:52TNTsaratoga we're fine.
04:21:02TNTand the disc check went to phase 1
04:21:24TNTi closed winrar which had my ipod directory open
04:21:40saratogaTNT: I mean to take this elsewhere
04:21:59TNTcool_walking is helping a lot.
04:22:10saratogahe can help you in rockbox-community
04:22:14cool_walking_I'm in #rockbox-community, too :)
04:22:29TNTlol how do i get there? ive never used IRC, sorry
04:22:42cool_walking_type /join #rockbox-community
04:23:28 Quit TNT ()
04:26:43 Join austin__ [0] (n=46bcb4eb@gateway/web/cgi-irc/
04:27:02 Quit austin__ (Client Quit)
04:27:21 Quit AndyIL (Read error: 110 (Connection timed out))
04:30:36***Saving seen data "./dancer.seen"
04:31:47 Quit apaininth (Remote closed the connection)
04:32:58 Join apaininth [0] (
04:34:34 Quit apaininth (Client Quit)
04:37:26 Join perrikwp [0] (i=18ac0c41@gateway/web/ajax/
04:53:47 Join blkhawk- [0] (
05:00:03 Join fdinel [0] (
05:03:18isthatallso I'm seeing differences in how this sansa C240 behaves in linux vs. XP, I can umark it as read only in XP but linux does *not* cooperate at all in that regard.
05:03:27isthatallvery strange, but it's FAT so go figger
05:09:54 Quit blkhawk (Read error: 110 (Connection timed out))
05:10:45 Nick blkhawk- is now known as blkhawk (
05:14:29 Quit saratoga ("CGI:IRC (EOF)")
05:17:28Topic"meh" by ChanServ (ChanServ@services.)
05:18:39 Join Darksair [0] (n=user@
05:18:43Topic"Please read before speaking: | Please direct offtopic/social chat to #rockbox-community" by ChanServ (ChanServ@services.)
05:20:45 Join sarixe [0] (
05:25:21 Quit cool_walking_ (" ajax IRC Client")
05:30:48 Quit Horscht ("Verlassend")
05:32:26 Join cool_walking_ [0] (i=cb3b81c3@gateway/web/ajax/
05:33:53 Quit isthatall (Read error: 110 (Connection timed out))
05:37:04 Join isthatall [0] (n=guy2@unaffiliated/isthatall)
05:55:27 Join Barahir [0] (
06:01:14 Quit AndyI ()
06:05:40 Join AndyI [0] (n=pasha_in@
06:06:17 Quit shadearg ("goodbye.")
06:11:23 Join shadearg [0] (
06:15:13 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
06:30:40***Saving seen data "./dancer.seen"
06:35:10 Quit fdinel ("Miranda IM! Smaller, Faster, Easier.")
06:50:58 Join parafin [0] (
07:08:15 Quit lymeca (Read error: 110 (Connection timed out))
07:12:30 Join MrDuck [0] (
07:13:17 Quit fyrestorm (Read error: 54 (Connection reset by peer))
07:16:46 Quit kachna|lappy (Read error: 113 (No route to host))
07:19:28 Join Bagderr [241] (n=daniel@rockbox/developer/bagder)
07:23:24 Nick Bagderr is now known as B4gder (n=daniel@rockbox/developer/bagder)
07:34:54 Quit CaptainKewl (Read error: 110 (Connection timed out))
07:35:17 Join einhirn [0] (
07:38:30kadobanah, thanks pixelma (about the tabs). i need to work on that patch some more, figure out the weirdness
07:38:41 Quit z35 ("Leaving")
07:38:50Unhelpfulhrm. well, pictureflow is working beautifully on e200 sim :/
07:39:10Unhelpfulit locks hard on the device, without any message telling me what's happening :/
07:51:16 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
07:59:27 Join crwl [0] (
08:01:23 Quit _lifeless (Remote closed the connection)
08:01:34 Join _lifeless [0] (n=lifeless@
08:06:38 Join Zagor [242] (
08:19:15 Join n1s [0] (n=n1s@rockbox/developer/n1s)
08:23:00 Join Rob2223 [0] (
08:23:09 Join ender` [0] (
08:30:03 Quit cool_walking_ (" ajax IRC Client")
08:30:43***Saving seen data "./dancer.seen"
08:41:13 Quit Rob2222 (Read error: 110 (Connection timed out))
08:50:10 Join lymeca [0] (
08:52:30 Quit BHSPitMonkey ("Ex-Chat")
08:55:33pixelmakadoban: I've been told that the tabs need to be there though, so took it back :)
09:23:38 Join petur [50] (n=petur@rockbox/developer/petur)
09:26:02 Quit MrDuck (Read error: 54 (Connection reset by peer))
09:58:28 Join Thundercloud [0] (
10:01:46GodEater_nice read B4gder
10:02:15scorche"too bad it doesnt ever plan to support classic" <−− too bad we dont ever plan to support *anything*...
10:03:33*linuxstb also enjoys B4gder's bedtime story
10:08:33linuxstbAlthough it misses out the very beginnings - everyone thinking it would be next to impossible to implement software USB on PP, then someone (MrH?) figuring out that the USB hardware was register-compatible with something that had docs.
10:08:47B4gderhm indeed
10:08:52B4gderI should add something about that
10:14:29n1sspeaking usb, does anyone know if the dma-setting for some bridge chips done by that tool someone posted in the wiki could be done device side?
10:16:37linuxstbB4gder: Are you sure about the AS3514 bit? I thought that was just the audio codec (and power management) ?
10:17:02B4gderit was imx31 similar or something
10:17:27n1syeah very similar to the one in imx31
10:17:35B4gderthanks, me fix more
10:19:35GodEater_which is how come it works on the beast
10:20:44GodEater_do you supply your own stock photos B4gder ? Or do you get them from somewhere else ?
10:21:42UnhelpfulFS #9919 updated with my latest work on pictureflow buffer_alloc. i'd still like to see about stopping playback, and moving the end of the audio buffer to steal a largish piece of RAM to do allocations from, without having to rely entirely on rb->bufalloc or the (rather small) plugin buffer
10:21:43B4gderI try to do my own, but I also borrow them from elsewhere when in need
10:22:58Unhelpfulthat's a nice usb plug, that is ;)
10:23:43B4gderthat's an example of the latter ;-)
10:24:16Unhelpfuli'm guessing i'll end up having to ping Nico_P about the buffer question?
10:25:50ZagorB4gder: actually I wrote "my" stack from scratch. our approaches were so wildly different there was little point in reusing his code.
10:26:16ZagorI did read his code, but as far as I remember I never copied any of it
10:26:51B4gderok, I'll update, thanks
10:27:40 Join bandan [0] (
10:29:43B4gderthere, better?
10:30:47***Saving seen data "./dancer.seen"
10:34:29LinusNi am a little curious about the pluginlib buffer_alloc patch in the tracker. i thought we wanted a true malloc() implementation in the plugin lib
10:34:34 Quit linuxstb (Read error: 113 (No route to host))
10:35:15ZagorB4gder: looks good
10:35:33 Quit Thundercloud (Remote closed the connection)
10:38:15 Quit einhirn (Read error: 110 (Connection timed out))
10:38:28UnhelpfulLinusN: i see them as being useful for different purposes... malloc is useful where we can't reasonably remove the use of the malloc API, or where moving the data is unsafe (multiple cores using the same buffer, for instance). the buffer_alloc allocator can do things we can't do with malloc, such as completely reclaim space lost to fragmentation by compacting the allocation buffer. i see reasons for using both in different places, and i
10:38:28Unhelpful thought that compaction would be more important in pictureflow than the other concerns.
10:39:31LinusNUnhelpful: i think both may be useful. i think the malloc() pool could be allocated using buffer_alloc
10:39:42 Join rocko [0] (
10:41:51UnhelpfulLinusN: i don't think so, unless you want to lose most of the features malloc offers, besides compatibility with an API used widely in other codebases. any time an allocation is performed, any allocated buffers may be moved. pointers will be left pointing at the wrong data if not updated, which is why code using buffer_alloc needs to call buf_get_data if "something else" may have touched the allocator.
10:43:28LinusNyou mean that a buffer allocated with buffer_alloc() may move at any time??????
10:43:34LinusNhow so?
10:45:11Unhelpfulnot at *any* time, but only when there are allocations following a free. buf_free creates holes in the pool, and if no individual hole is large enough to provide the needed memory for a later allocation, buf_alloc will compact the pool by shifting everything down in address as far as possible, and then try to allocate again from the remaining space.
10:48:39Unhelpfulthis is why buf_alloc, like rb->bufalloc, returns handles rather than pointers. the handle can be used to fetch the current address of an allocation, and the design is optimized to do this quickly, in some cases at the expense of a bit of space efficiency.
11:01:00 Join kachna [0] (
11:06:21 Join kachna|lappy [0] (
11:07:50 Quit bmbl ("Woah!")
11:09:10 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
11:10:59rockowhat firmware should I download ? ipod 4g 20gb
11:11:56rocko4g (10/3.1.1.) or 4g (4/3.1.1.) ?
11:14:43UnhelpfulNico_P: i had a question (around that i thought you might know about, if you're not too busy. :)
11:16:52 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
11:17:00Nico_PUnhelpful: the question is "how to get RAM from the playback engine?"?
11:18:35UnhelpfulNico_P: yes, basically, but a large chunk with a fixed address, rather than a movable piece with a handle as the bufalloc api provides. stopping playback, if needed, to get it would be ok.
11:19:12Nico_PI don't remember the details, but corssfade does basically that
11:19:24 Part bandan
11:19:49 Quit kachna (Read error: 110 (Connection timed out))
11:20:41Unhelpfulhrm. i didn't think to look outside of plugins... i couldn't find any examples there, things that need more than the plugin buffer either use bufalloc, and deal with it failing, or stop playback and grab the whole audio buffer.
11:23:36B4gderin two
11:24:34Nico_PUnhelpful: I think it's done in audio_set_crossfade() (playback.c:797)
11:33:22 Join Zoxc [0] (
11:35:11Unhelpfulthat function seems a bit odd... it looks like the "real" magic is happening in pcmbuf.c, and i'm pretty sure that plugins shouldn't do what that does ;)
11:36:16Unhelpfuli think the 1MB plugin buffer would pretty much remove any need for stolen audio buffer on color targets, and greyscale targets will likely manage OK with .5MB plugin buffer, as their slide bitmaps are 1/2 the size (per pixel)
11:36:49 Join {phoenix} [0] (
11:38:00 Join Zoxc- [0] (
11:38:01Nico_Pthere's the call to audio_get_buffer()
11:39:21Nico_Pwhich is the same as what is used by plugins
11:39:25 Quit rvvs89 (Read error: 113 (No route to host))
11:39:54UnhelpfulNico_P: doesn't that just stop playback and provide a pointer to, and length of, the entire audio buffer?
11:40:22Nico_Pcrossfade uses it to resize the audio buffer and restart audio
11:41:28Nico_Pah I see what you mean, the actual size is defined by pcmbuf
11:41:57 Join rvvs89 [0] (n=ivo@pdpc/supporter/base/rvvs89)
11:42:28 Quit jfc (Read error: 104 (Connection reset by peer))
11:42:43Unhelpfulright, the size parameter is only used to pass back the buffer size.
11:45:11 Part B4gder
11:45:16Nico_PI think the same apporach could still work. we just need to be careful as to how we set the buffers up
11:45:18Unhelpfulheh, i can duplicate the cache fighting seen on e200 in sim, but not the hang... i might need to get serial logf working to figure out what's going on.
11:45:46 Quit rocko ("Leaving")
11:49:41 Quit Zoxc- ()
11:50:00UnhelpfulNico_P: how do you tell core, when it starts playback again, that you've only used *some* of the buffer you took, though? it seems like there might need to be some new magic for that, maybe a parameter specifying the amount wanted?
11:50:06 Join fyrestorm [0] (
11:50:33 Join rocko [0] (
11:51:04 Quit lymeca (Read error: 110 (Connection timed out))
11:51:47Nico_PUnhelpful: I think pcmbuf takes care of doing the buffer layout, but I haven't looked for where yet
11:54:06Nico_PI'm sorry I can't help you more than that, but I have to go
11:55:36 Quit kachna|lappy (Remote closed the connection)
11:55:45 Quit Zoxc (Read error: 110 (Connection timed out))
11:56:26Unhelpfulno big deal, i'm about out of work time, anyway. i'll look into it further tonight, but you've been a big help getting started :)
11:56:54Nico_PI'll try to look into it a bit more later this afternoon
11:58:30 Join kachna [0] (
12:06:44 Quit gevaerts (Nick collision from services.)
12:06:53 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
12:17:55 Join einhirn [0] (
12:22:06 Join Xenovalent [0] (n=xenovale@
12:22:26linuxstbDoes anyone know if the random freezing when connecting usb (without the Rockbox stack) also happens with the Rockbox stack?
12:23:27gevaertsI've never seen it
12:25:51advcomp2019on my e200 with the usb test build, my wheel did not work after unplugging it but the buttons worked.. i had that happen only once so far
12:26:42UnhelpfulNico_P: thanks. an alternative (which might be useful for other things) would be making bufalloc, possibly only when passed a special flag, force just enough audio data out from the end of the buffer for the alloc to succed. then it could make allocations that way, possibly mixed with ones made from the plugin buffer, without having to stop playback at all.
12:27:46linuxstbgevaerts: (and others) What do you think about enabling the usb stack now, to give it as long as possible for testing (and fixing of any bugs that are found) before the 3.2 release? If your tracer tests give worrying results, it could always be reverted prior to release.
12:29:19*Llorean is for that
12:30:36*Zagor too
12:30:50***Saving seen data "./dancer.seen"
12:34:00*GodEater_ is too
12:34:34 Join TheSphinX^ [0] (
12:35:05linuxstbadvcomp2019: You mean that most of the time the wheel works after usb, but one time it didn't?
12:36:52advcomp2019linuxstb, yea
12:37:44JdGordonlinuxstb: i said the same thing yesterday ... although if the tacer test is probably going to be soon then maybe we hould wait till then.. jut in case...
12:38:06*Llorean has seen several cases of people reporting a data abort or similar the first time they use our USB stack, and then not after.
12:38:20Lloreanadvcomp2019: Was the scroll wheel thing after your first use of it?
12:38:42linuxstbJdGordon: gevaerts said in #rockbox-community that he should receive the tracer towards the end of next week.
12:39:22JdGordonyeah, i saw that... i guess it depends how soon before 3.2 we want it in?
12:39:39advcomp2019yea and i tried it again and it did not happened.. i can try it again
12:40:50linuxstbJdGordon: That's what I'm saying - the sooner the better, to give more time for testing and bug fixing.
12:40:50 Join TheSphinX^_ [0] (
12:41:32LloreanWe should put a "expected to work, but if it doesn't may result in corrupted data, please back up your music" in the commit activating it.
12:41:37JdGordonend of next wek still gives 2 weeks before the freeze...
12:41:55JdGordoni dont think most uses read the commit messages
12:42:31 Quit rocko ("Leaving")
12:42:38LloreanI don't think we're obligated to avoid slight risks to help better test the code.
12:42:42Unhelpfulmaybe in the news area? it's a pretty big change, USB being on.
12:43:33JdGordonI woudn't actually object to postponing 3.2 by one or two weeks to make sure usb gets in....
12:43:38LloreanWe have 3.1 for people who want to avoid risk, and bleeding edge for people who want to try new stuff.
12:43:54JdGordonif we asume peple only upgrade on releases its a huge improvemnt which people shouldnt wait another 3 months for
12:43:55*Llorean doesn't think we should ever postpone a release again.
12:44:06LloreanIf we aren't confident in USB when the time comes, we drop it and it can wait for 3.3
12:44:07*Bagder agrees with Llorean on both usb and postponing
12:44:57LloreanWe've seen what happens when we decide to delay "just a little bit more"
12:45:23 Quit linuxstb (Read error: 60 (Operation timed out))
12:45:48JdGordonI mean, if we are just waiting to make sure it works then wating an extra week for testing is OK... if it comes out that its still broken then it should be pulled
12:47:06LloreanThree months isn't a long time considering people have been putting up with reboot-USB for much, much longer. I don't think "trying to squeeze it in one week earlier" is a good enough reason to break our release schedule.
12:47:46LloreanWe could even release a 3.2.1 point release on top of the 3.2 branch with USB enabled if we do decide it's good enough, if it's that big of a deal.
12:48:05JdGordonyeah, thats a very possible optino
12:48:09 Quit fyrestorm (Read error: 104 (Connection reset by peer))
12:51:38*freqmod_gq experienced that the radio on sansa clip have much better quality when tuned into a frequency that ends with xxx.x0 than with xxx.x5 and wonders if he should report it as a bug
12:52:03 Quit TheSphinX^_ (Remote closed the connection)
12:52:14advcomp2019linuxstb, just tested it like five times, the wheel still works after disconnecting it from usb
12:52:42Unhelpfulfreqmod_gq: not a bug, normal stations (in the US, at least) should not have frequencies xxx.x5
12:52:49Unhelpful(i think)
12:53:09rasherThere were a few reports of data aborts after unplugging in the forum
12:54:29*rasher wouldn't hold usb out for this, unless it causes corruption
12:54:36freqmod_gqi made the preset file with sansa e200v1 and there the x5 worked best but it is probably because of noise, crosstalk, multipath etc.
12:54:54JdGordonrasher: that sounds pretty bad...
12:55:27LloreanJdGordon: Every single one it was a "it happened once, then never again" I think
12:55:55JdGordonah, ok
12:55:58LloreanThere may be something up with the first time you use the stack.
12:56:03JdGordonfun to debug :p
12:56:12rasherHard to tell what's going on. Could be completely unrelated of course
12:56:48rasherBut I guess the "shut down usb" code hasn't had a good shaking down yet
12:57:45rasherStill, a data abort *after* disconnecting is not quite as critical I'd think
12:58:26advcomp2019Llorean, yea that is just like my wheel not working after unplugging.. it happened once and it has not happened yet
12:58:29JdGordonthey are annoying though because of the slow reboot
12:58:41 Quit TheSphinX^ (Read error: 110 (Connection timed out))
13:02:51 Quit Rob2223 ()
13:02:51 Nick Xenovalent is now known as XenovalentAFK (n=xenovale@
13:04:16 Join faemir [0] (
13:08:08 Join jaykay [0] (
13:08:52jaykaycould someone look into and verify this is not happening?
13:09:06jaykayits about freezing of e200 when connecting it to a charger
13:12:14pixelmajaykay: for comparison reason - do you test with a downloaded build or one you built yourself?
13:12:32jaykayboth i think
13:14:01advcomp2019jaykay, with the usb test build, it does not do that with my charger
13:14:50jaykayadvcomp2019: which usb test build? and you cant reproduce it, right?
13:15:34advcomp2019jaykay, i am using this: and yea, i can not reproduce it
13:16:52 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
13:17:19pixelmasounds like the "freeze when it actually should reboot to OF" bug which happens in some builds. The reason I asked is that when I first encountered this bug with a borrowed Ipod Mini it happened with a downloaded build but not with one I built myself - with the same revision
13:19:04 Join Rob2222 [0] (
13:19:12pixelmaas far as I know many portalplayer devices were (are still?) affected and I saw that quite a few times on my c200 as well. I'm using a Rockbox USB enabled build for a while which of course doesn't reboot to the OF and so I can't comment on if it still exists
13:19:55pixelmait doesn't reboot on USB connection while Rockbox is already running, I mean
13:20:49jaykaywhen connected to a charger, it doesnt, right - but when connecting to a computer it does
13:20:57jaykayimo everythings fine
13:22:53LloreanWell, if his H100 really isn't rebooting on USB insert, something isn't fine. H100s don't have USB chargers.
13:22:58pixelmajust saying that I don't doubt the bug report this much
13:22:59LloreanBut that'd be a separate bug.
13:24:28Llorean"My iriver H100, when plugging in a charger or usb cable, will continue playing in rockbox and charge the battery."
13:24:40*Llorean doubts this very much.
13:25:45*pixelma was more focuesed to the original bug report
13:26:20LloreanThe original bug report sounds like he's simply complaining that it reboots on charger insertion.
13:26:24pixelmaprobably "on"
13:26:43LloreanThe hang is bad, but he complains about the existence of the reboot in general in the comments.
13:27:24Llorean"the USB cable caused a reboot into original firmware, and if that is what still happens why has the website changed from unsupported to supported."
13:27:34pixelmaI thought it was about the hang which definitely existed for me on c200 as well in *some* build, it's just that I haven't tested for a while
13:27:50LloreanThe impression I get is that he's not getting the hang now, just the reboot.
13:28:00LloreanAs of his last comment.
13:29:11jaykaywhy does it reboot?
13:29:21jaykayas i said my e200 doesnt do that
13:29:28LloreanYou probably have a different charger than him
13:29:29jaykay(when connecting it to a charger)
13:29:35LloreanSome chargers can't be distinguished from a USB connection.
13:29:51jaykaythen its a problem of the charger, not of rockbox
13:30:01LloreanIf that's what's happening, yes.
13:30:04LloreanBut we need to wait for him to respond.
13:30:45jaykayis usb support enabled by default since 20032?
13:31:09LloreanIt says in the commit message that it's not...
13:33:05jaykayoh sorry...
13:40:07 Join fyrestorm [0] (
13:40:07 Quit Aurix_Lexico (Remote closed the connection)
13:42:46 Quit _lifeless (Remote closed the connection)
13:43:02 Join _lifeless [0] (n=lifeless@
13:44:11 Nick fxb__ is now known as fxb (
13:45:06jaykayare anywhere on the rockbox-website old manuals available?
13:45:27 Quit JdGordon (Remote closed the connection)
13:46:12LloreanThe manual's built from SVN. You should be able to compile basically any version of the 'modern' manual.
13:47:02jaykayno, thats the problem
13:47:31LloreanWhat's the problem?
13:51:05jaykaythere are some packages missing.... i dont know which ones. scorche said a new vmware-image is in work, im waiting for this.
13:51:34LloreanSo the problem isn't that it can't be built, but that you haven't got things set up right to build it?
13:51:50LloreanIt's possible to download and/or install any of the packages yourself.
13:52:04LloreanWhy, exactly, do you need a *built* old manual?
13:52:08jaykayi guess so. as i said i dont know which ones
13:52:24jaykayi do not really need a old manual, it was just for comparison with the "new" one (r20034)
13:53:03LloreanYou could just look at the diff.
13:56:27 Join rocko [0] (
13:58:00rockowhat should I do "dd if=mbr-4g-20gb.bin of=/dev/disk
13:58:00rockodd: opening `/dev/disk': Is a directory"
13:58:04 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
13:58:15rockodisk is the name of my ipod when mounted
13:59:21 Quit kachna (Read error: 60 (Operation timed out))
14:00:51rockoit is /dev/sdc1
14:05:16 Quit _lifeless (Remote closed the connection)
14:05:33 Join _lifeless [0] (n=lifeless@
14:09:22 Join robin0800 [0] (
14:10:03 Join pyro_maniac [0] (
14:13:04linuxstbrocko: No, you want to "dd" the MBR to the "whole-disk" device, which in your case is "/dev/sdc". "/dev/sdc1" refers to the first partition on that disk.
14:13:15 Join midijunkie [0] (
14:13:54rockoI see ok thanks LinusN
14:15:23 Quit fyrestorm (Read error: 54 (Connection reset by peer))
14:16:26rockolinuxstb I get his "/dev/sdc:
14:16:26rocko re-reading partition table
14:16:26rocko BLKRRPART failed: Inappropriate ioctl for device
14:16:32rockowhat shooould I do
14:20:00 Quit DataGhost (Nick collision from services.)
14:20:08 Join DataGhost [0] (i=dataghos@unaffiliated/dataghost)
14:21:09 Join CaptainKewl [0] (
14:24:10linuxstbrocko: What is saying that?
14:24:40rockohdparm -z /dev/sdc
14:25:10linuxstbJust unplug your ipod, and then reattach it. You will probably need to force the ipod into disk mode (hold SELECT+PLAY as it's booting)
14:26:23gevaertsrasher: do you still happen to have .elf and .map files for the usb test builds?
14:26:28rockonow what should I do?
14:26:32 Join jfc [0] (
14:28:01linuxstbrocko: I assume you're following these instructions?
14:28:48linuxstbThen step 4 ;)
14:29:07 Join LambdaCalculus37 [0] (i=44a04303@rockbox/staff/LambdaCalculus37)
14:29:55rockohow do I know which one to use for 4g ipod to use? linuxstb
14:30:06rockothere are two of them
14:30:16rockofor firmware
14:30:31LambdaCalculus37That's easy. Do you have a greyscale screen or a color one?
14:30:55***Saving seen data "./dancer.seen"
14:30:57preglowanyone know if the timestretcher patch fixed the binsize issue?
14:30:59rockoso is it 10 or 4
14:31:05LambdaCalculus37Then I suppose you'll want the 4th gen Greyscale build.
14:31:23rockono that is not what I am talking about LambdaCalculus37
14:31:29rockoI already know that
14:31:32*LambdaCalculus37 goes to the logs
14:31:39rockoI am talking about the apple firmware
14:31:51rockofor restoring ipod
14:33:03LambdaCalculus37Ah, okay.
14:33:07rockothere is two of them which one should I use
14:33:08 Join grdxyxy [0] (n=eric@
14:33:13LambdaCalculus37Try 10.
14:33:28jaykayBigBambi: why did you use /index in installation.tex, line 73?and why does it produce a note?
14:36:46*LambdaCalculus37 tries out rasher's USB-enabled build for his c250
14:40:26rockolinuxstb what if it tells me that it already contains a fat32 filesystem should I just leave it
14:40:45rockoI partition it in fat32 before I did the restore
14:41:06rockoevery thing else worked
14:43:34linuxstbrocko: I would still run the mkfs.vfat commant again - just to be sure it's using the values from the new partition table.
14:44:07 Quit _lifeless (Remote closed the connection)
14:44:09linuxstbAnd I don't think it matters which of the firmwares you download - IIRC, they're identical.
14:44:23 Join _lifeless [0] (n=lifeless@
14:45:22jaykaythe manual for the ipod3g is form r20014....
14:45:26jaykayis that normal?
14:46:54LambdaCalculus37jaykay: I just looked at it and it's listed as r20045.
14:48:13jaykaylambdacalculus37: weird
14:48:32jaykayi checked it many times, also deleted the cache of firefox
14:49:41pixelmaLambdaCalculus37: jaykay is right, it's broken in pictureflow (pdf is from the 16th)
14:51:57 Join __lifeless [0] (n=lifeless@
14:52:11pixelmathe html is incomplete and stops after the "keys" headline in the pictureflow chapter
14:52:18 Quit _lifeless (Remote closed the connection)
14:53:06rockolinuxstb I did what you said but I updated the firmware first and then umounted the ipod and did the format and it seems to work is that ok
14:53:34rockothe directory was empty when I formatted it then mounted it
14:53:47rockobut after I unplugged it and plugged it back iin
14:53:59rockonew directories came up
14:54:12rockoso I guess I am okay
14:54:17LambdaCalculus37pixelma: Strange. I looked at the online version.
14:54:36 Quit FlynDice ("Konversation terminated!")
14:55:12pixelmayes, which is incomplete - like #8819 describes
14:56:32pixelma1st and 2nd gen Ipod is also broken
14:57:43*LambdaCalculus37 fetches the manual for the 1st/2nd gen iPod
14:58:23LambdaCalculus37pixelma: Indeed; it's dated 16 February.
14:58:59 Join moos [0] (i=Mustapha@rockbox/staff/moos)
15:02:50 Join tyfoo [0] (
15:04:39*pixelma wonders why the 1st through 3rd gen Ipods don't HAVE_SCROLLWHEEL defined in their config-* files
15:05:14*LambdaCalculus37 shrugs
15:06:02LambdaCalculus37Don't they need HAVE_SCROLLWHEEL anyway?
15:07:53 Quit blkhawk (Remote closed the connection)
15:08:23 Join blkhawk [0] (
15:09:53pixelma1st gen Mini doesn't have it either but 2nd gen Mini. I know they have different wheels (one is position aware the other not) but both should be "scrollwheels". Confusing
15:12:54pixelmaand btw. I don't like that the c200's pictureflow button table in the manual describes the current buggy behaviour, if it described how it should be maybe people would start bugging the developers to fix it ;)
15:13:54LambdaCalculus37pixelma: We'll have to bug whoever wrote it to fix it up, unless one of us beats him to it. ;)
15:15:18 Join fdinel [0] (
15:15:36pixelmameh, PLA and mixing contexts :\
15:17:40jaykayi guess its wanted that 1g2g has {\UseOption,IPOD_3G_PAD} right?
15:19:50 Join miepchen^schlaf [0] (
15:20:17 Quit {phoenix} (Read error: 104 (Connection reset by peer))
15:24:47pixelmanot sure anymore but if I remember correctly it doesn't hurt at the moment because they use the same mappings everywhere (? - at least almost) currently
15:31:52 Part LinusN
15:32:23jaykaypixelma: could you check whether you can compile the manual?
15:34:18pixelmafix for the broken Ipod manuals is - either add IPOD_3G_PAD to the \opts and \nopts in the button table or to define HAVE_SCROLLWHEEL in their config-*.h files which may have influence on something else so I don't know
15:34:25pixelmano, I can't
15:37:01pixelmathe former in pictureflow.tex
15:37:32jaykayyou mean to all the opts with scrollwheel?
15:40:08 Quit parafin ("reboot")
15:40:15GodEater_seems to be good news on the MTP code
15:41:44amiconnjaykay: No, 1g2g should have IPOD_1G1G_PAD
15:41:55amiconnErr, IPOD_1G2G_PAD
15:42:05 Quit grdxyxy ("Leaving.")
15:42:42 Quit faemir ("Lost terminal")
15:43:14jaykayamiconn: that would brake MANY things..... because IPOD_1G2G_PAD is never used
15:43:19jaykayexcept in 4 plugins
15:43:43rasherpixelma: seen my aa manual patch?
15:44:21rockolinuxstb thanks for your help
15:44:33rockomy ipod is rocking rockbox again :D
15:44:54rockoI just wanted to restore my ipod because it was acting slow and messed up
15:45:05rockonow it is normal and fast again
15:45:29LambdaCalculus37rasher: What's the FS#?
15:45:55rasherErr.. Can't check now
15:46:37amiconnjaykay: Then that needs to be fixed. The 1G/2G pad is quite different.
15:46:47gevaertsGodEater_: where?
15:47:25jaykaylambdacalculus37: 9928
15:47:55LambdaCalculus37jaykay: Thanks.
15:48:29LambdaCalculus37rasher: By the way, I tested your USB enabled build on my Sansa c250. Managed to transfer 800MB so far to the microSDHC card without a hitch.
15:48:42*LambdaCalculus37 will transfer more when he gets a chance
15:51:01jaykayamiconn: i hope you will have fun when checking the buttonmaps of ~60 .tex where IPOD_3G_PAD is used....
15:52:44 Join kubiix [0] (n=511bc91c@gateway/web/cgi-irc/
15:53:13 Quit miepchen^schlaf ()
15:57:33 Join parafin [0] (
15:57:36 Quit CaptainKewl (Read error: 110 (Connection timed out))
16:02:09 Quit __lifeless (Remote closed the connection)
16:02:25 Join __lifeless [0] (n=lifeless@
16:06:59jaykaypixelma: can you post a pstebin with your opt-changes? i would include that in a patch
16:09:05pixelmaI could commit it but am not sure if that's the right fix
16:10:00jaykaydo what you want to do :) if nothing is broken....
16:24:48 Join Multiplex [0] (n=5773d21c@gateway/web/cgi-irc/
16:26:16 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
16:26:42mcuelenaerelinuxstb: have you read the mailing list mails about the licensing issues?
16:29:35 Join MethoS- [0] (
16:29:35 Quit kubiix ("CGI:IRC (EOF)")
16:30:56***Saving seen data "./dancer.seen"
16:31:32 Quit Zagor ("Clint excited")
16:32:26 Join toffe82 [0] (
16:35:17GodEater_gevaerts: dev list
16:36:46rashergevaerts: unfortunately not
16:39:10rashergevaerts: I could try recreating them, but I don't know how reliable that is
16:39:18 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
16:40:30 Quit __lifeless (Read error: 104 (Connection reset by peer))
16:43:14 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
16:43:22saratogaI've had the sansa wheel stop working before
16:43:31saratogausually just hitting any button fixes it though
16:43:34saratogahappened this morning to me
16:44:32rasherYeah, I've had that once in a while, seems to happen more or less randomly
16:45:12 Quit Multiplex ("CGI:IRC (EOF)")
16:45:52pixelmalooks like the HAVE_SCROLLWHEEL define was introduced with the scrolling in lists accelaration (first e200, then the "clickwheel" Ipods). It's not enabled for the Ipods with the older wheel and the changes were in firmware/target/arm/ipod/button-clickwheel.c so it's probably correct to not define it
16:47:33 Join _lifeless [0] (n=lifeless@
16:47:46 Quit fdinel ("Miranda IM! Smaller, Faster, Easier.")
16:48:44 Join jgarvey [0] (
16:48:45saratogaonce USB is enabled, how hard is it to enable bootloader USB mode on PP? I know we'd need interrupts for that
16:49:37rasherGood news on the mtp code
16:49:57GodEater_rasher: I thought so too
16:50:04GodEater_that reads to me as "license it however you like"
16:50:17rasher"for any use" means just that, afaik
16:52:07LambdaCalculus37I guess we're in the clear with the MTP code. :)
16:52:41GodEater_assuming no-one here comes up with a reason not to use it
16:53:00*LambdaCalculus37 can't think of one
16:53:22LambdaCalculus37We should buy that guy from Creative a crate of beer for giving us good news, though. :)
16:53:41 Join __lifeless [0] (n=lifeless@
16:54:10*jaykay agreed
16:55:04jaykaymay i ask how much money comes in with the ads on the hompage?
16:55:11gevaertsGodEater_: ah indeed, *that* MTP... :)
16:56:18GodEater_I didn't think we were still running those ads
16:56:25saratogawe don't
16:56:41rasherRight, so I'm not going crazy
16:56:44rasherjaykay: which ads?
16:57:21jaykayi/me doesnt know
16:57:28*jaykay doesnt know
16:57:40jaykaywait a moment
16:58:54GodEater_we don't have them any more
16:59:16jaykayhow about deleting that page?
16:59:47saratogayou could just mark it as old
17:00:53jaykayand everyone who sees the "this page is no longer valid" will wonder why its still existing
17:02:08LambdaCalculus37I think "This page is no longer valid" is enough of an explanation.
17:02:31LambdaCalculus37Or say "This is page is old and only exists for historical purposes".
17:02:45jaykayok thats better
17:02:54 Quit Nico_P (Remote closed the connection)
17:03:56jaykayrasher did that already, this cant be missed :)
17:07:11 Join kachna [0] (
17:07:50*n1s thinks FS #9929 is a bug in the user's expectations...
17:09:38jaykayn1s: this was really confusing when i tried rockbox and was happy about this delete-function....
17:10:02jaykaythe first few tims i always checked twice whether the file was really deleted
17:11:25n1sI don't understand how this can be confusing, you tell it to delete the file, and it does
17:11:36 Quit _lifeless (Read error: 110 (Connection timed out))
17:12:04jaykaybut you think its still existing because rockbox is still playing this file
17:12:29saratogait does still exist, just not on the storage
17:12:52saratogawhat happens if the file isn't entirely buffered when you delete?
17:13:06saratogaor if the file is bigger then the audio buffer?
17:14:49n1ssaratoga: i think it just skips to the next song if the file is missing when it tries to rebuffer
17:18:26jaykaythe quickscreen is missing from the manual...
17:25:20pixelmalast time I looked it wasn't
17:25:59pixelmawasn't missing, I mean - so should be there
17:28:51 Quit __lifeless (Remote closed the connection)
17:29:07 Join __lifeless [0] (n=lifeless@
17:30:59jaykaypixelma: i didnt find it... i searched with adobe reader for "quickscreen", only one result (in the changelog)
17:35:26rashergevaerts: Recreated the builds now and saved .map and .elf.. *might* not be accurate, but I guess it's worth a shot
17:36:09jaykayis there any guideline for the maximum of characters in one line in the manual files?
17:36:36rasherAnd yes, that's in the wiki
17:39:18jaykayah i found it, sorry...
17:39:19 Join nibbler [0] (
17:40:05gevaertsrasher: great. Let's see if those reported errors make sense :)
17:40:14 Quit robin0800 (Read error: 104 (Connection reset by peer))
17:42:05GodEater_jaykay: the quick screen is referred to as just that "quick screen" - not as one word in the manual
17:42:15 Join japc [0] (n=japc@
17:42:16n1srasher: if you compare md5 of the final binary with the one posted in the forum you can tell for sure if they are the same
17:42:41gevaertsn1s: different date
17:42:54rasherThere were a few bytes difference in the filesizes
17:43:01rasher(of the .zips)
17:43:52 Join flydutch [0] (
17:44:21 Join CaptainKwel [0] (i=2669ecc2@gateway/web/ajax/
17:45:38rashergevaerts: all the builds have rockbox-*.map/elf uploaded now
17:46:39n1sgevaerts: ah
17:47:23 Join caladan_ [0] (
17:47:34gevaerts00008228 would be in draw_eq_slider() if (a) that number is correct and (b) the builds are equivalent
17:48:24rasherI re-uploaded the builds, so any new downloads should match the .map/elf files
17:48:55gevaertsactually... Do you still have the old ones?
17:49:17rasherOf course not - do you expect me to be organised?
17:49:59*gevaerts wanted to compare them around this address
17:51:55BigBambijaykay: I can't look until I get home
17:52:09jaykayBigBambi: ok....
17:52:53BigBambipixelma: Sorry about the pictureflow early ipod issue - I specifically added the minis, but missed that the early ipods didn't have scrollwheel defined
17:53:32BigBambipixelma: Also, I stand by describing how the c200 actually works - I think we should tell the users what actually happens, not what is supposed to happen until it actually does :)
17:56:20 Join SndBstrd [0] (
17:57:36 Quit petur ("work->real life")
18:00:11 Nick fxb is now known as fxb__ (
18:00:49n1suh, my c200 hangs at the OF "Refresh database" afer i ran fsck on it, how do i recover from this?
18:01:51 Join bertrik [0] (
18:02:37LambdaCalculus37n1s: Can you force it to power off?
18:02:50n1sLambdaCalculus37: yeah
18:02:54rashergevaerts: that seems like an odd place to crash..
18:03:49 Part SndBstrd
18:03:55 Join kushal_12_27_200 [0] (n=kushal@
18:04:23LambdaCalculus37n1s: Does Rockbox work normally on it?
18:04:34LambdaCalculus37Or at least seem to be working normally?
18:05:11 Join evilnick [0] (i=0c140464@gateway/web/ajax/
18:05:31n1sLambdaCalculus37: no, i think the partition had gotten messed up pretty badly, the booloader is just flashing the logo over and over again
18:05:42n1sthe sansa.fmt thing doesn't work on c200 does it?
18:05:49 Join SndBstrd [0] (
18:06:19gevaertsn1s: it doesn't
18:06:40n1sso, does this leave me with running your eraser tool again? :)
18:06:51gevaertsIf nothing else works, yes
18:11:52 Quit kushal_12_27_200 ("Leaving")
18:11:55n1sok, installing a fresh of in recovery mode doesn't help so i guess that's what i have to do, is the source and build instructions for c240 erase available somewhere btw?
18:12:54 Quit perrikwp (" ajax IRC Client")
18:13:19gevaertsSure. It's all in svn
18:13:59gevaertsConfigure for c200, then choose E and follow the plot
18:14:13n1sgevaerts: aha, thanks
18:15:32*n1s spots breakage
18:16:23n1s"/home/nils/rockbox/bootloader/main-c240wipe.c:195:53: error: macro "storage_write_sectors" requires 4 arguments, but only 3 given" this have anything to do with storage rework stuff?
18:17:42gevaertsseems so
18:18:01*gevaerts fixes
18:19:52gevaertsn1s: done
18:20:23n1sgreat, thanks :)
18:21:51 Join faemir [0] (
18:21:52 Part SndBstrd
18:25:12 Quit caladan_ ("ChatZilla 0.9.84 [Firefox 3.0.6/2009021207]")
18:25:42n1sis it the bootloader.bin file i want to use?
18:27:39 Part pyro_maniac ("Leaving.")
18:30:58***Saving seen data "./dancer.seen"
18:51:05 Join cafaro [0] (
18:52:47 Quit rvvs89 (
18:52:47 Quit Barahir (
18:52:47 Quit blithe_ (
18:52:47 Quit Ridayah (
18:52:47 Quit slyyf (
18:53:35 Quit moos ("Rockbox rules the DAP world")
18:54:40NJoinrvvs89 [0] (n=ivo@pdpc/supporter/base/rvvs89)
18:54:40NJoinBarahir [0] (
18:54:40NJoinblithe_ [0] (
18:54:40NJoinRidayah [0] (
18:54:40NJoinslyyf [0] (
18:54:58 Join Barahir_ [0] (
18:55:28 Quit Darksair ("To Arch or Gentoo? That is the question...")
18:56:07 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
18:58:26 Quit yhuang ("Leaving")
18:58:36linuxstbmcuelenaere: Yes, I've seen your emails. It's frustrating they don't use a real license, but I think it's good enough for us. Although I would like to know what others think.
18:58:58rasherI'm pretty sure "for any use" is basically PD
18:59:03 Quit Barahir (Connection reset by peer)
19:00:12linuxstbI'm guessing mcuelenaere didn't say "hacking your own players" as the purpose of sendfirm though ;)
19:00:19linuxstb(sendfirm is used for the Zen IIUC)
19:00:56 Join Zoxc [0] (
19:01:01mcuelenaereI specifically avoided mentioning Rockbox, perhaps they wouldn't answer then :)
19:01:37 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
19:02:33linuxstbmcuelenaere: But I think that's good news. IMO the next step would be for you (as the (C) holder) to change the license of sendfirm_win.c to something more permissive (BSD?) so we can link to MS's lib. You could also commit those emails (e.g. in a README, or a comment in sendfirm_win.cpp
19:03:13linuxstb(assuming you're happy to use BSD of course...)
19:03:20mcuelenaereyeah, sure no problem :)
19:03:42linuxstbSo sendfirm is also being used for the Zen?
19:04:18mcuelenaereit's a generic MTP firmware upload program
19:04:34mcuelenaereit could be used as a replacement for sending the OF
19:06:35linuxstbOK, so if "someone" writes a beast-specific install tool, we should keep sendfirm around as a generic tool.
19:08:25 Quit sarixe ("Leaving")
19:09:19 Part LinusN
19:11:59BigBambijaykay: The short answer is I didn't
19:12:09BigBambijaykay: It was already there, I just moved it
19:16:34 Quit gevaerts ("rebooting")
19:18:25 Quit domonoky (Read error: 104 (Connection reset by peer))
19:20:21 Join Thundercloud [0] (
19:21:29 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
19:21:38 Quit scorche (Nick collision from services.)
19:21:56 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
19:22:07 Join {phoenix} [0] (
19:22:24 Join scorche [0] (i=Blah@rockbox/administrator/scorche)
19:24:02 Join anirban [0] (n=anirban@
19:24:30anirbanI have video iPod and installed rockbox
19:24:41anirbanbut it is not at all stable
19:24:45bertrikjhMikeS, anything holding back the PP502x audio DMA patch?
19:24:53anirbanit is hanging and rebooting all the time
19:25:14anirbanmodel no is A1136
19:25:43anirbanbut from site i got that 5th gen video ipod build is very stable
19:25:58BigBambiIt is normally
19:26:01n1sanirban: does it display any messages when it's rebooting?
19:26:07n1sor hanging
19:26:19BigBambiAnd what are you doing when it does?
19:26:24BigBambiAnd what version of Rockbox?
19:26:34anirbanno my screen just getting blank or frezes up
19:27:14n1sdid you install the version for the 60/80 models?
19:27:42jhMikeSbertrik: not really
19:27:54 Join miepchen^schlaf [0] (
19:27:57anirbanwait let me check the version of rockbox
19:28:27 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
19:28:34anirbanIthink it is ver-19036
19:28:39 Quit nibbler (Remote closed the connection)
19:29:04BigBambianirban: What version did you install? 30GB or 60/80 GB?
19:29:17BigBambin1s: oops, sorry - didn't see you asked that already :)
19:29:17anirban30 GB
19:29:59anirbanmost of the things works but main prob is hang and reboot
19:31:58gevaertsDoes this happen randomly, or e.g. when you play certain files?
19:32:01anirbanplease tell me which version of rockbox to install
19:32:21anirbannot even playing
19:32:59 Join bluebrother [0] (n=dom@rockbox/developer/bluebrother)
19:33:01anirbanit happens when I am browsing my ipod or accessing the database or even while controlling the equalizer
19:33:09BigBambianirban: Try the latest one, you are 1000 revisions out of date
19:33:18gevaertsAlso check the filesystem for errors
19:33:30anirbanok let me install it now
19:33:35anirbanhow to do that
19:38:07gevaertsYou mean checking the filesystem? What OS are you on?
19:40:07 Quit linuxstb (Read error: 110 (Connection timed out))
19:42:06 Quit dfkt (Read error: 104 (Connection reset by peer))
19:45:07lostlogicugh, anyone want to help a relative mac newb with why his is failing on osx?
19:45:43 Join Lss [0] (
19:46:01n1slostlogic: iirc preglow had lots of trouble building crosscompilers on osx
19:46:17n1smaybe he knows something
19:46:51 Join perrikwp [0] (i=98214eea@gateway/web/ajax/
19:47:07 Quit perrikwp (Client Quit)
19:49:08n1slostlogic: also how is it failing?
19:49:53lostlogic"make -r" fixed −− was getting the error mentioned in this post:
19:50:07lostlogicso now I need to teach to figure out when to use make -r
19:51:09 Join fdinel [0] (
19:51:17Bagderor use gnu make perhaps?
19:52:03rasherSounds smarter
19:52:47lostlogiccould install gnu make
19:53:15 Quit parafin (Read error: 113 (No route to host))
19:53:24BagderI bet you'll get problems to build rockbox without it anyway
19:53:34lostlogicdidn't realize what make -r did, just found a posting that said it fixed this error :-P
19:53:50 Join parafin [0] (
19:53:58 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
19:54:17n1slostlogic: so, you planning to get back into rockbox hacking? ;)
19:55:24lostlogicdepends, definitely want to get it building on here regardless −− there'd need to be some feature that I want that isn't there for me to get back into it is the long standing problem...
19:58:00kadobanhas anyone ever done much towards getting gdb serial working over usb? i'm doing a little research into it and i was wondering if there's anything rockbox specific i should read/look at.
19:58:35Bagderkadoban: LinusN had gdb serial working on the archos models
19:58:42 Quit MethoS- (Remote closed the connection)
19:58:57Bagderbut I'm not sure that helps anything
19:59:14 Quit bs66_ (Connection timed out)
19:59:24kadobanBagder: ah okay, that's mostly it though, right? (i saw that, but haven't looked at it in depth yet)
19:59:43Bagderyeah, I don't think we've had any other gdb experiments
19:59:50kadobanokay, thanks
20:00:40 Join bs66_ [0] (
20:03:55anirbani am trying to download the rockbox installer for 64 bit linux
20:04:08anirbanand it seems the archive is broken
20:04:13anirbanunable to unzip it
20:04:23 Quit domonoky (Read error: 104 (Connection reset by peer))
20:04:40anirbanI am using this link
20:05:21rasheranirban: Works for me
20:05:35anirbanwhat command you use to unzip it
20:05:46anirbanI am using tar -jxvf file_name
20:05:56kadobananirban: bzip2 -d <file>
20:06:05gevaertsthis isn't a tar file
20:06:08kadobanit's not a tarball, just
20:06:10kadobanzipped up
20:06:14rasherbunzip2 rbutilqt-v1.0.7-64bit.bz2
20:06:26 Join lymeca [0] (
20:06:26*gevaerts is confused
20:06:36lostlogic*scowl* macports gmake is stalling on activate
20:06:59anirbanohh ok sorry it is not a tar
20:07:35gevaertsWhy doesn't this mr100 give lots of problems? It's behind three or five hubs depending on how you count
20:08:10*gevaerts is starting to suspect that he did something wrong last night
20:08:16 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
20:10:03 Quit domonoky (Read error: 104 (Connection reset by peer))
20:12:19rashergevaerts: So it might actually fix things on 5020 as well?
20:12:57gevaertsI need to retest a bit, but IIRC the build on my mr100 does not have the PP5022 fix
20:13:25 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
20:14:05gevaertsOr maybe there's a really weird thing that makes the resets go away if you keep adding hubs
20:14:22rasherThat does indeed sound *really* weird
20:17:38 Nick jaked is now known as jaked_ (
20:18:40 Quit parafin (Nick collision from services.)
20:18:45 Join parafin|away [0] (
20:20:10 Nick parafin|away is now known as parafin (
20:21:45 Join pyro_maniac [0] (
20:25:41 Quit japc (Read error: 145 (Connection timed out))
20:30:59***Saving seen data "./dancer.seen"
20:33:23pixelmaBigBambi: you then could scrap the "move up/down through the album list" for the c200 manual completely as it also doesn't wrap currently and you'll only ever be able to highlight the firsdt track... ;\
20:33:36pixelmafirst too
20:37:31 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
20:38:48anirbanI just tried the latest rockbox installer on AMD 64
20:39:36anirbanall things load properly but when I reboot ipod i got error mesg that can not load rockbox.ipod read failed
20:39:56rasherIs rockbox.ipod in the correct location?
20:40:01anirbanshould I try to install the bootloader manually
20:40:12rasherThe bootloader is installed correctly
20:40:21rasherI'd rather try installing rockbox manually
20:40:38 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
20:41:02rasherWhich consists of downloading and unzipping rockbox to the root of your ipod
20:41:49Lssdont forget to flash too
20:42:11rasherLss: What are you talking about?
20:42:24 Join kugel [0] (n=kugel@rockbox/developer/kugel)
20:42:35Lssoh you have the bootloader done ok
20:43:00rasherNo flashing involved.
20:46:15anirbanis there any prob if I previously loaded ipod linux in my ipod
20:46:21anirbanand try to install rockbox now
20:46:56*rasher has no idea
20:47:27rasheranirban: do you have a .rockbox dir on your ipod?
20:48:01anirbanyeah it is there
20:48:27rasherIs it partitioned in any interesting way?
20:50:20anirbanyeah I have one fat32
20:50:30anirbanand ext2 about 200 MB
20:51:07pixelmaspeaking of the "album list" in pf on the c200. Currently it neither scrolls horizontally if a track name is too long not vertically if it's higher than the display, as I remember Unhelpful saying it should...?
20:51:13rasheranirban: That sounds like it could cause trouble
20:51:29anirbanso what to do?
20:51:35anirbando I have to reformat
20:51:57 Join z35 [0] (
20:52:41rasherWell, restoring to the original partitioning should help. How that's best done, or if there's another solution, I don't know
20:52:43n1sanirban: iirc rockbox only mounts and looks for its files in the first partition so if that is not the fat32 partition it will not find them
20:53:32gevaertshm, I seem to remember differently
20:53:48rasheranirban: you could try simply deleting the ext2 partition
20:54:12rasherThere's a chance that might do it
20:54:14anirbanfat32 is the first partition
20:54:38*rasher doesn't really know how the bootloader works in great detail
20:55:17anirbanactually fdisk -l /onmyipod
20:55:21anirbanshows 3 entry
20:55:49rasherThere's a small firmware partition as well
20:55:52anirbanfrist 80293 blocks are with empty partion
20:56:09anirbanand it shows as /dev/sdc1
20:56:14 Join Windlord [0] (
20:56:21anirbanthen /dev/sdc2 as W95 FAT32
20:56:37anirbanthen /dev/sdc3 as Linux
20:57:05 Quit pyro_maniac (Read error: 110 (Connection timed out))
20:57:08gevaertsanirban: not sure if this has been asked before, but does the Apple firmware work properly?
20:57:25anirbanyeah it works wihtout any problem
20:58:52gevaertsdid you check the filesystem?
20:59:27anirbanno actually I do not know how to do it
20:59:40gevaertsfsck.vfat /dev/sdb2
20:59:42anirbando you mean fsck the file system
21:00:02anirbanI shall try first let me uninstall the whole stuff
21:00:08anirbanit is in the process
21:01:36 Quit Windlord ("")
21:04:34 Quit lymeca ("KVIrc 3.4.0 Virgo")
21:05:11 Join pyro_maniac [0] (
21:05:49 Quit miepchen^schlaf ()
21:13:09*gevaerts can get his mini to not connect properly by not boosting
21:13:43 Join perrikwp [0] (i=98214eea@gateway/web/ajax/
21:15:04 Join roman_ [0] (n=roman@
21:20:16 Join moos [0] (i=Mustapha@rockbox/staff/moos)
21:21:50 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
21:22:52 Quit jaykay (Read error: 110 (Connection timed out))
21:24:27*bluebrother wonders if he should nag domonoky and LambdaCalculus37 again about the OS X patch or simply commit it and push a new rbutil release
21:25:46*LambdaCalculus37 wonders why the last rbutilqt version domonoky gave him just quits immediately when launched
21:26:27saratogamaybe someone could setup bluebrother with VNC access to an OSX box
21:26:41domonokymaybe because i didnt built it correctly ? ( the mac was remade, so i had to reinstall Qt and libusb)
21:27:08LambdaCalculus37domonoky: Maybe. You also have to make sure it builds as a universal build as well.
21:27:36domonokyi know, but maybe i missed someting ?
21:27:43pyro_maniachey there. is there is sleep function supported in rockbox?
21:28:06rasherpyro_maniac: check the manual
21:28:07 Join lymeca [0] (
21:28:49pyro_maniacoh no I mean in the code, sorry for that
21:28:52 Nick fxb__ is now known as fxb (
21:29:02pyro_maniacI need to wait after a function
21:29:24pyro_maniacis it simply sleep(10); ?
21:29:55domonokyhm, the rbutil binary is universal. maybe i forgot some lib ?
21:31:09domonokyah, libusb isnt included as static lib. lets hunt for a static and universal libusb lib for mac :-)
21:31:54*kugel wonders if domonoky tried to play xobox on his e200v2
21:32:12rasherpyro_maniac: I'd check some of the other code
21:32:24*domonoky didnt try much with his e200v2, why ?
21:32:47kugelit's pretty much broken on my fuze
21:32:51linuxstbpyro_maniac: If you want to sleep for 10 seconds, do "sleep(10*HZ);"
21:32:51kugelconstantly pausing
21:33:03*bluebrother thinks he should try to eventually get a mac :)
21:33:24pyro_maniaclinuxstb: ok thanks!!
21:36:50 Join tessarakt [0] (
21:37:39anirbanthank you all for the help
21:37:40*rasher thinks the Flyspray user settings db should be cleaned of people has their date and time-format string set to their username
21:37:56anirbanI need some sleep now will be back tomorrow
21:38:52 Quit anirban (Remote closed the connection)
21:40:34kugeldomonoky: I committed the e200v2 lcd driver patch, a test would be appreciated
21:41:07 Quit cafaro ("cafaro has left the building")
21:45:02 Quit pyro_maniac (Read error: 113 (No route to host))
21:48:15 Join low_light [0] (i=c730190a@gateway/web/ajax/
21:48:53 Quit perrikwp (" ajax IRC Client")
21:50:44 Join perrikwp [0] (i=98214eea@gateway/web/ajax/
21:50:47 Quit low_light (Client Quit)
21:54:21kugeldomonoky: ok, xobox works ok now
21:55:48 Join `Assassin [0] (
21:56:36BigBambipixelma: Oh, OK :/
21:57:29 Quit LambdaCalculus37 (" ajax IRC Client")
21:57:44*gevaerts thinks understands the data aborts on usb disconnect
21:57:57kugelgevaerts: cool, I have them too sometimes
21:57:58 Join SirFunk_ [0] (
21:58:21*gevaerts inserts a "he" in there
21:59:07rashergevaerts: awesome
21:59:44gevaertsThe problem is that some code calls usb_wait_for_disconnect(), and then assumes that its data is still there afterwards. Since the software USB stack uses the playback buffer, this is not a valid assumption
21:59:54*domonoky has now a new, and hopefully working rbutil mac test binary here">
22:00:17gevaertsSee e.g. buffering.c line 1385. After that it calls update_data_counters(), which is where it crashes for
22:00:24rashergevaerts: uh oh
22:00:39amiconngevaerts: Why does the usb stack use the playback buffer?
22:00:57 Join RoC_MasterMind [0] (
22:01:25`AssassinWhy do certain songs in my library always play for a few seconds with quality worse than normal and produce a data abort crash? I'm using an iPod Video (30 gb) with the latest rockbox build.
22:02:25gevaertsamiconn: because it needs comparatively big amounts of memory (32+K), and it only needs them while connected.
22:04:57gevaertsIt grabs the buffer using audio_get_buffer()
22:05:23 Quit Llorean (Connection reset by peer)
22:06:17 Quit RoC_MasterMind ("Leaving")
22:06:20amiconngevaerts: Hmm, if it does that, and yet playback/buffering still tries to access data afterwards, that's a major bug in playback/buffering
22:07:10 Join Llorean [0] (
22:08:24gevaertsamiconn: maybe the problem is that it only calls audio_get_buffer() *after* sending SYS_USB_CONNECTED?
22:09:04gevaertsAt least it used to do that
22:09:24 Join miepchen^schlaf [0] (
22:09:27amiconnThat may be the problem, but it shouldn't
22:09:30jhMikeSafter audio_get_buffer happens, the only thing could cause playback to access anything is more messages.
22:10:24amiconnaudio_get_buffer() means playback has to back off immediately, and not try to grab it again unless started explicitly. And then it has to rebuffer everything
22:10:31jhMikeSanother stop message shouldn't cause further access though or that's how it was when I last left it
22:12:43 Quit SirFunk (Read error: 110 (Connection timed out))
22:13:12jhMikeSonce audio is waiting for USB, it's message should be discarded and any sync messages auto-replied with 0
22:15:19 Join Lear [0] (n=chatzill@rockbox/developer/lear)
22:24:37 Quit blithe_ (Remote closed the connection)
22:25:23 Join blithe [0] (
22:25:55 Quit evilnick (" ajax IRC Client")
22:25:55 Join matsl [0] (
22:26:20 Quit Horscht ("Verlassend")
22:30:23 Join petur [50] (n=petur@rockbox/developer/petur)
22:30:53n1sjhMikeS: keyclick isn't supposed to work when playback is paused, right?
22:31:00***Saving seen data "./dancer.seen"
22:32:14 Join pyro_maniac [0] (
22:32:41 Join Massa [0] (
22:34:36kugeln1s: well, "supposed" isn't quite right, but yea
22:35:17MassaHi everybody!
22:36:05pyro_maniacgevaerts: I have to bother you again. I wanted to test with with an earlyer usb init as you told me but I always got a system freeze on connecting
22:36:24MassaI lately have some problems with the database at my iriver H340
22:36:46 Join MethoS- [0] (
22:37:03 Quit roman_ (Read error: 110 (Connection timed out))
22:37:13MassaIt does not initialize - it just starts and then stops immediately again without doing anytihing - does somebody else notice that?
22:38:36gevaertspyro_maniac: I haven't done this. You'll really have to experiment a bit yourself...
22:38:41jhMikeSn1s: It's still in the same situation since "paused" is still considered a state above "playing".
22:38:45 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
22:39:33n1sok, so the same as voice not playing while paused
22:40:02pyro_maniacgevaerts: and you have no idea, where this freeze come from? is there a more safty method to connect?
22:40:30gevaertsHow should I know?
22:41:32gevaertsThere probably is something that's not fully initialised
22:43:18mcuelenaeregevaerts: would it be 'normal' that when doing 3 logf() without any delays between them (with USB serial enabled), the device crashes?
22:43:27mcuelenaerewhen I add sleep(5); between the calls, it doesn't
22:43:38MassaAnybody in the mood to talk with me?
22:43:52gevaertsmcuelenaere: that sounds like a bug
22:44:08mcuelenaeregevaerts: a bug in the USB driver or in the USB core?
22:44:22gevaertsthe usb serial driver most likely
22:45:07advcomp2019Massa, just hold on
22:45:58 Join Aurix_Lexico [0] (
22:46:12mcuelenaeregevaerts: oh, couldn't it be one in the USB driver (as in the USB SoC driver)?
22:46:34Massaadvcomp2019: ok - I'll try to be more patient ;)
22:46:54gevaertsIt could be there as well of course.
22:47:57mcuelenaereperhaps you could check whether it's also on PP targets?
22:49:33gevaertsWhat exactly are you doing? (i.e. where are the logf() calls?)
22:52:11mcuelenaerethey're in a function a (new) .C file (in firmware/drivers/), which gets called from the bootloader
22:52:20 Nick XenovalentAFK is now known as Xenovalent (n=xenovale@
22:52:43mcuelenaerethis is without sleep(5); calls between the logf()'s
22:52:49mcuelenaere(which crashes)
22:52:57gevaertsOK, so not from within the USB thread
22:53:02mcuelenaereno AFAIK
22:53:39gevaertsThat should work too of course (and AFAIK it does), but it might be slightly different
22:54:44mcuelenaereactually, all logf()'s from within usb_core.c worked
22:54:52mcuelenaere(I'm not sure whether that's from inside the USB thread or not)
22:55:09 Join avis [0] (n=ident@pdpc/supporter/student/avis)
22:55:23gevaertsA lot of them are
22:56:20`AssassinDoes anyone else get frequent data abort errors with your iPod?
22:58:43bertrikI just got some data aborts on my e200 today, but it wasn't an entirely clean svn version
22:59:26Llorean`Assassin: Under what circumstances, what SVN revision, and what iPod?
23:02:27`AssassinParticular songs will always produce a data abort error (They play for a few seconds with terrible quality compared to normal. Then they crash.) I'm using a 5th generation iPod Video (30 gb) with the latest stable build of rockbox.
23:02:47 Join fml [0] (n=4fd3cd9f@gateway/web/cgi-irc/
23:02:50LloreanBy "latest stable build" you mean the 3.1 release/
23:03:00saratogaprobably broken files then
23:03:19`AssassinYes Llorean.
23:03:20saratogayou could try copying one such file to a different folder with nothing else in it [to rule out buffering bugs] and then playing it
23:03:30saratogaif you get teh same crash, then its a broken file
23:03:34saratogaif not, then its a rockbox bug
23:03:39jhMikeSbertrik: I take it it's not from the DMA change since you were querying on what was holding it back and said "no problems"
23:03:40n1syeah, but we shouldn't crash on them even if they are broken
23:03:44 Quit Thundercloud (Remote closed the connection)
23:03:54n1sso either way it's a rockbox bug
23:04:05fmlMassa: when using the DB, it's possible to skip some directories by placing a special file into it. Are you sure you don't have such a file in the root dir? See wiki/manual for details.
23:04:15jhMikeSbroken rockbox files caused Mouser_X's beast not to boot properly
23:04:17bertrikjhMikeS, no saw it without the DMA patch too earlier this week / last weekend
23:04:40bertrikseems to happen at the end of buffering on specific albums
23:06:07jhMikeSI had a few mysterious ones myself after not having seen them for a looooong time but not enough to associate it with particular thing
23:06:09`AssassinI'm not sure why any of these files would be broken. They work fine on my computer unless something happened during the transfer to my iPod. :|
23:06:23fmlkugel: hello. This is a tiny matter compared to the great work being done but I sent you a message re. inaccurate comment about the y field in the progress bar struct. Did you notice it?
23:06:40gevaerts`Assassin: what codec do they use?
23:06:40Massafml: I don't know - will have a look; although the behaviour is really strange: when I try to open the database it says "Building database... 12981 found (STOP for return)" but then also does nothing...
23:06:55kugelfml: I noticed your "complain"
23:07:04Llorean`Assassin: I'd recommend also always using the most up to date build when reporting issues here, anyway. You're always going to be asked to try the same test on the current build if it's an issue that needs further investigation.
23:07:07kugelbut in C, 0 is positive, so it's all fine
23:07:22jhMikeSn1s: my personal philosophy is to treat anything put into the metadata parsers and database as random gibberish with a correct file extension/name.
23:07:30fmlMassa: try to delete all the index files (don't know exactly where the are, in .rockbox?)
23:07:56fmlMassa: I don't use the DB myself, just telling you what I have read/heard here
23:08:09bluebrotherthe database files are .rockbox/*.tcd
23:08:52n1sjhMikeS: yeah, we can't really assume much about metadata :) I think that it's ok to reject files that are obviously broken and not jump through hoops to try to play them though
23:08:56fmlMassa: if that doesn't help try to look for weird files (very long name or bad tags)
23:09:13jhMikeS.cfg and .nvram should probably have the same treament. any setting that is invalid should be clamped or defaulted.
23:09:18Massafml: I already deleted the *.tcd files
23:09:25`AssassinOne of the files in question is an MP3 and iTunes says "Encoded with: unknown"
23:09:52bluebrotherMassa: have you chkdsk /f the drive? I had issues with the database recently and they were related to filesystem issues
23:10:04aviswhats the name of the rockbox general channel ?
23:10:24fmlMassa: can you build a sim? I'd then try to build the DB on the sim. There you'll hopefully get some debug messages.
23:10:27Lloreanavis: This is the Rockbox channel.
23:10:32Massabluebrother: good idea, I'll give it a try - although it'll take a long time ;)
23:10:33n1sjhMikeS: i think cfg is pretty good at that
23:10:36 Join akur [0] (
23:10:39 Part akur
23:10:41avisi thought there might be an offtopic channel
23:10:42bluebrotheravis: general channel? That's #rockbox, but if you're referring to general offtopic-chatter it's #rockbox-community
23:10:49n1ss/cfg/our cfg parser/
23:10:49avisooh thank you
23:10:58avisi just put rockbox on me e270 :)
23:11:01bluebrotherofftopic isn't general ;-)
23:11:12fmlavis: the other channel is rockbox-community
23:11:15BigBambiavis: You should try reading the topic
23:11:22 Join TNT [0] (
23:11:41 Quit fml ("CGI:IRC 0.5.9 (2006/06/06)")
23:12:19TNTdoes any1 know about cyclic redundancy error?
23:12:24Massafml: of course I can - it's also a good idea (I assume you mean that I should put the sim on my device to use the same files/directory structure and to make the test reliable)
23:12:45Massafml: but first I'll try to chkdsk the device...
23:12:46jhMikeSn1s: was it changed? I had an invalid setting for poweroff timeout cause the player to shut down almost immediately after entering the FW.
23:12:54BigBambiTNT: Generally it means the disk/CD/DVD whatever is buggered
23:13:22TNTi get it when trying to install the rockbox bootloader on my ipod.
23:13:22bluebrotherMassa: beware that you can't use the same .rockbox folder as on the player, and the "player" is expected in a subfolder of the sim ...
23:13:33bluebrothernothing that should be problematic on linux at least ;-)
23:13:34TNTcould it possibly me a bad cord?
23:13:36n1sjhMikeS: not recently no, well i _thought_ it was good at that sort of thing but i am obviously wrong then :/
23:13:47BigBambiTNT: possibly I guess
23:13:53 Join Llorean1 [0] (
23:13:59BigBambiTNT: It is certainly worth trying another
23:15:11Massabluebrother: I'll use Linux - much easier to act with symbolic links than to copy the whole player content to my computer's harddisk :)
23:16:06 Quit Llorean (Nick collision from services.)
23:16:08 Nick Llorean1 is now known as Llorean (
23:16:28TNTthanks man\
23:16:31TNTthanks man
23:16:48`Assassinlol I'm sure these files of mine are not broken. Rockbox simply can't handle them for whatever reason. I'll see what the latest of the latest build can do for my problem.
23:17:20bluebrotherMassa: indeed. If you have access to a windows machine you might want to try windows chkdsk /f too −− at least from my experience it's less forgiving (thus finding more issues) than fsck.vfat
23:19:31 Quit buk_ (Read error: 104 (Connection reset by peer))
23:19:35Massabluebrother: actually I ususally (have to) run windows most of the time - but I also have a lot of VMs with Linux which I often use :)
23:21:01Massabluebrother: I just ran chkdsk - and it did find some lost chains :( - I'll retry the database building shortly
23:21:50 Quit perrikwp (" ajax IRC Client")
23:22:00`AssassinLol I'm still getting a data abort. It says "Data abort at 0000994C (0)" if that means anything to anyone.
23:22:52n1sit means that the cpu tried to do an unaligned memory acces at that address
23:23:42n1sit would be helpful if you could provide a sample file that causes this in some way
23:23:45bertrikor maybe accessed an address with nothing "behind it"
23:23:54jhMikeSn1s: the instruction's at that address-8 I think
23:24:20Massaflm bluebrother: the database building seems to work now - thanks a lot for your help :) - it seems that the found/lost files are all database files...
23:24:24Lear`Assassin: And the address suggests you've encountered a bug in the buffering code.
23:24:26n1sjhMikeS: ah, right
23:24:33`AssassinIt's no problem for me to give someone the file. n1s would you like a DCC send?
23:25:14`Assassinor I could put it on for anyone to look at.
23:25:29n1sdcc, isn't working with my router so that won't work, is it a file that you can distribute freely?
23:26:17 Nick fxb is now known as fxb__ (
23:26:18`Assassinuh.. You mean legally?
23:26:42 Quit pyro_maniac (Read error: 104 (Connection reset by peer))
23:26:43`Assassinlol Nope.
23:26:57n1sok, but could you mail it to me?
23:27:45`AssassinSure. Your email is?
23:27:52pixelmareminds me once again of FS #9827
23:28:10n1syes, it seems to be the same thing...
23:28:23 Quit Zoxc ()
23:28:54 Join LambdaCalculus37 [0] (n=rmenes@rockbox/staff/LambdaCalculus37)
23:28:58pixelma`Assassin: sorry if I missed it, but what player do you have?
23:28:59n1sAssassin, could you try to copy the file into a dir by itself and play it by itself?
23:29:44n1sor just stop playback and long click on it to insert it in a playlist so it's the only song in the playlist?
23:30:30`AssassinYes I can. pixelma, it's an iPod video 5th generation (30 gb).
23:30:55rashergevaerts: did you see the other report indicating buffering trouble?
23:30:57 Join film42 [0] (
23:31:01LearSounds like I should re-implement the buffering debugging code I wrote a while back... Nothing fancy, just validate memory_handle pointers now and then, and if invalid, dump the logf buffer.
23:31:51n1sLear: there is a patch in FS #9827 that supposedly fixes it
23:31:54LearMakes it easier to reproduce the bug and get a handle on what's going on.
23:31:58 Quit {phoenix} (Remote closed the connection)
23:33:23LearHm, maybe I should wait and see if that one really does fix all the problems. :)
23:34:25n1smcuelenaere: more red
23:34:35mcuelenaeren1s: I know
23:34:52n1sgood :)
23:35:31kugelLear: it would be nice if you could say anything to the attached patch. I'm having this data aborts too on a certain album
23:35:37kugelplaylist, rather
23:35:45gevaertsrasher: I have now
23:37:11`AssassinOk I just did a clean install of Rockbox and then I put one of the problem files in a new folder called Test and it's working without a problem. I'm going back to the original folder to see if the clean install fixed the problem.
23:37:14gevaertsmcuelenaere: I seem to be able to reproduce it here
23:37:36mcuelenaeregevaerts: ok, so it's not SoC specific :)
23:37:41Learkugel: I'm not that familiar with buffering.c, so I can't really say if that code is the right solution...
23:38:03gevaertsmcuelenaere: indeed not. You're still allowed to help debug it though :)
23:38:04 Quit ender` (" Sometimes I lie awake at night, and I ask, "Where have I gone wrong?" Then a voice says to me, "This is going to take more")
23:38:24 Quit petur (Remote closed the connection)
23:38:30mcuelenaeregevaerts: I would, however the exception handler can't get as far as displaying the error on the LCD :/
23:38:58`AssassinIt seems that the clean install with the latest rockbox build is what fixed the problem. I hope the problem doesn't appear again. :(
23:39:28pixelmaBagder: just noticed a tiny tiny quirk in the build table - the used image height is a tad bit too small for the Philipps HDsomething "- Boot" and "- Sim", it's cut off a bit
23:39:35`AssassinMaybe my database problems will be fixed too. :D
23:40:00mcuelenaeregevaerts: do you think this is an USB serial bug or USB core?
23:40:19gevaertsmcuelenaere: usb serial I'd say.
23:40:22n1sAssassin: It seems this is caused by a buffering bug and not actually related to specific files only their sizes
23:40:37n1sand order in the buffer, etc
23:40:38Bagderpixelma: yeah I noticed, I just didn't care enough yet to do anything about it
23:40:57 Quit Lear ("ChatZilla 0.9.84 [Firefox 3.1b3pre/20090213033101]")
23:41:09`AssassinSo will I see this bug again? :(
23:41:28pixelmaBagder: yeah, it's not really important
23:42:16gevaertsmcuelenaere: I suspect that there is a locking bug somewhere in usb_serial.c
23:42:19 Quit Lss (Read error: 110 (Connection timed out))
23:42:30n1sAssassin: possibly, until it's fixed
23:42:54`AssassinI guess that means a reinstall everytime it breaks. :(
23:43:29LloreanA reinstall won't fix it.
23:43:38LloreanThe fact that it's gone after the reinstall is just coincidence.
23:43:41n1sa reinstall _shouldn't_ fix it... so it's probably something else you did differently
23:43:47`AssassinBut it JUST did fix it. (temporarily anyways)
23:44:45n1sa changed setting for example could make it go away, a different build could trigger the bug for a different size etc.
23:44:57 Quit n1s ("Lämnar")
23:45:17 Quit bmbl ("Woah!")
23:46:42 Quit bertrik ("Leaving")
23:47:31gevaertsmcuelenaere: at one point vitja said that there was a bug in usb_serial, but I didn't really look into it at the time. I can't find it now :(
23:47:53mcuelenaeredid he mentioned it on #rockbox?
23:47:57 Quit rocko ("Leaving")
23:48:04`AssassinI hope it's fixed by the next build. :O Now to tackle my database bug again. I think the database just refuses to read the ID3 info on certain songs correctly and puts these songs in the database as <untagged> or just in the wrong place. The ID3 info that shows gets cut off. Anyone know why?
23:48:15gevaertsI think so. Looking through logs now
23:48:34*mcuelenaere too
23:48:40 Join buk_ [0] (
23:49:30linuxstb`Assassin: Do you have large comments or similar in your tags? That can cause Rockbox to not have enough space to store the important tags, if they come after the large field.
23:50:01 Quit tyfoo (Read error: 104 (Connection reset by peer))
23:51:06Massabye everybody - I'll go to see my bed ;)
23:51:14 Quit Massa ("ChatZilla 0.9.84 [Firefox 3.0.6/2009011913]")
23:51:43 Quit domonoky (Read error: 104 (Connection reset by peer))
23:51:48mcuelenaerehmm /me thinks we need to contact vitja :)
23:51:53gevaertsFS #9437
23:52:19`AssassinThat's interesting. You're the first person to say that. I think quite a few of these songs come with long comments. How could I make the important data come before the comments? lol Also, is anyone working on fixing this issue?
23:53:16 Part LinusN
23:54:14 Join yapper [0] (n=18ec7c6f@gateway/web/cgi-irc/
23:55:13gevaertsI guess that bit needs to be reviewed and merged. If you have time for it, feel free to have a go. I can't look into it before saturday
23:55:14 Quit yapper (Client Quit)
23:55:33 Quit jgarvey ("Leaving")
23:56:30linuxstb`Assassin: I think the general solution is to not store essays in the comments tag...
23:56:54 Quit sharperguy (Remote closed the connection)
23:57:15rasherlinuxstb: I thought MoB was supposed to make room for arbitrarily large tags?
23:57:30`AssassinBut sometimes I want essays. :( Oh well... can't hurt to delete them considering I didn't put them there anyways.
23:57:34mcuelenaeregevaerts: I don't I'm going to touch much of that without the USB tracer
23:57:42Lloreanrasher: On the buffer. it's the database that seems to havei ssues still
23:57:44mcuelenaeres/I don't/I don't think/
23:58:09Lloreanrasher: The problem often is that the WPS will show the title/artist/album fine, but the database will have them truncated because something giant came before them in the metadata.
23:58:18linuxstbrasher: AFAIK, the metadata struct is still a fixed size. It needs to be, as Rockbox copies the structs for the current and next tracks to static buffers.
23:58:22 Quit CaptainKwel (" ajax IRC Client")
23:58:48linuxstbLlorean: Really? I'm not sure how they could be different.
23:58:56Lloreanlinuxstb: This is usually what I see reported.

Previous day | Next day