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 2007-08-12

00:01:23 Quit dan_a (Remote closed the connection)
00:05:22 Join salty_horse [0] (n=ori@pdpc/supporter/active/salty-horse)
00:13:26*rasher enters the menu "Sound settings > Crossfeed > Crossfeed"
00:13:30rasherWho thought up this?
00:13:48*Bagder is innocent!
00:17:19rasherBagder: md5sum change committed. Of course, this shouldn't matter to you, except your next voicebuild will be slower.
00:17:37rasher(fingers crossed)
00:18:03Bagderwe'll find out tomorrow morning ;-)
00:19:07 Quit haemmy ()
00:20:06rasherYeah, we don't need to worry about missing any voice bugs.
00:20:49 Quit matsl (Remote closed the connection)
00:21:03rasherApparantly festival can speak various other languages. Such as russian and italian
00:21:12rasherI have NO idea how though.
00:21:51 Quit salty-horse (Read error: 110 (Connection timed out))
00:22:32 Join pixelma [0] (i=57b9f5b2@rockbox/staff/pixelma)
00:24:24Bagder"I had to install and uninstall about 15 times before it ran at full speed"
00:26:23 Join amiconn [0] (
00:29:28*rasher listens to Festival speaking Italian
00:30:09Bagderbtw, we should allow the language selector in configure use the language name and not just the number
00:30:25Bagderotherwise automated non-english ones will be a pain
00:30:56rasherNow to see if I can somehow configure festival through.. configure
00:31:01amiconnI would like to be able to select multiple languages for building voices
00:31:10BagderI noticed this when I brushed up my script this time since english had moved number
00:31:23amiconnNow that the md5sum is used for clip names, this md5sum could include the language
00:31:52rasheramiconn: Probably a good idea. Same strings in different languages are not pronounced the same.
00:31:56amiconnAlso, the .wav clips are intermediates which can be deleted after encoding to .mp3, even when using the pool
00:32:17amiconnThis would save space
00:32:19rasherBut building multiple voices at once (without reconfigure) is not going to be as easy
00:32:32rasherOr.. maybe
00:32:50rasherI guess VOICE_LANGUAGE could be a delimited list of some sort
00:33:46amiconnI also find it irritating that selecting a voice build still creates a standard Makefile, and 'make' tries to build that target instead of voice
00:33:56rasheramiconn: that's a feature
00:34:08rasherso you can make && make voice and have a build and a voicefile
00:34:14amiconn...but 'make voices' in a standard build directory doesn't work, unlike 'make manual'
00:34:32rasherThat's because the manual needs no configuring
00:34:35Bagdermake manual has no choices like the voice build has
00:34:39rasheryou're asking the impossible
00:34:54amiconnI know *why* that is, but I still find it somewhat irritating
00:35:15rasherWell, I guess the questions could appear once you type make voices, but that's not terribly clean either
00:35:35rasherHaving it all happen inside configure feels nicest
00:35:39amiconnA standard build dir shouldn't allow 'make voices'
00:36:06*Bagder goes to bed
00:37:57rasheramiconn: same could be said for make install. I don't recall that being an issue. Why would you call make voice if you hadn't configured voice building?
00:38:46rasheroh, hrm, ignore me
00:41:15rasheramiconn: fixed
00:42:50rasherNow about multiple languages. We're going to need seperate TTS settings for each language. I'm not quite sure how to do that. For now I'll just make it accept a comma-delimited list of languages to build. They'll all be built with the same options (for now). Babysteps.
00:42:57 Join barrywardell [0] (
00:48:19rasherHrm. Genlang isn't very good about telling what needs to be done for voice-entries
00:51:18 Quit ender` (" Trying to establish voice contact ... please yell into keyboard.")
00:51:46 Quit petur ("Zzzzz")
01:05:01 Quit thegeek ()
01:13:32 Part juxtap
01:23:39 Join hannesd_ [0] (
01:27:37 Quit salty_horse ("Leaving")
01:28:48 Join perrikwp [0] (n=chatzill@
01:30:20 Quit barrywardell (Read error: 110 (Connection timed out))
01:31:04 Quit miepchen^schlaf ("Verlassend")
01:34:38 Quit My_Sic ("Miranda IM! Smaller, Faster, Easier.")
01:39:06 Quit hannesd (Read error: 110 (Connection timed out))
01:39:07 Nick hannesd_ is now known as hannesd (
01:44:32 Join dg10050 [0] (
01:49:06 Join chrisjs169__ [0] (
01:51:38dg10050anyone know about an issue with Rockbox on 5.5g iPods? When I run it, I get the message "No .rockbox directory", even though I have the .rockbox directory in the root of the fat32 partition
01:51:43dg10050I'm in linux btw
01:57:59***Saving seen data "./dancer.seen"
01:58:57 Join thegeek [0] (
02:00:58pixelmaI guess your using loader2 ?
02:02:27dg10050is that bad?
02:04:06 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
02:04:31pixelmathen you should ask in the ipodlinux channel - for exampel I know nothing else about it other than it exists, there's also a thread about it in the "unsupported builds" section of the rockbox forum
02:05:08 Quit chrisjs169_ (Read error: 110 (Connection timed out))
02:06:11dg10050rockbox runs, it just gives me that error message, then it goes to the menu and I can't really do anything
02:07:14rasherCan you reproduce it with the Rockbox bootloader?
02:08:17dg10050well, I'd rather not, but I can try
02:08:35dg10050keep in mind loader 2 is simply loading the rockbox bootloader
02:08:47scorcheno it isnt
02:09:09dg10050yeah, just realized that
02:09:55scorchereally though...please go into #ipodlinux for help with loader2
02:10:28dg10050k, I just wasn't sure it was loader2's fault
02:10:36dg10050so I came here
02:10:46*scorche nods
02:25:36 Quit dg10050 (Remote closed the connection)
02:29:05 Part pixelma
02:32:06iambenanyone else find it a bit counterintuitive that at the 'balance' screen, on the sansa's, rolling the wheel left (CCW) shifts sound to the right?
02:32:57 Quit ToHellWithGA (Read error: 110 (Connection timed out))
02:36:59 Join saratoga [0] (i=98039a34@gateway/web/cgi-irc/
02:37:07DataGhostwhy in the world is loader2 related to a problem with rockbox AFTER rockbox is successfully loaded and executed!?
02:40:52scorcheDataGhost: there have been issues of that in the past due to it not initializing things correctly, but i seemed to have missed that part for some odd reason...
02:41:21DataGhostI don't see how... looks like a rb bug to me then
02:41:40DataGhost.rockbox folder not found errors occur AFTER detecting partitions, filesystems, etc.
02:42:08scorchethat is why we asked if he could replicate it using rockbox's bootloader i suppose then
02:42:31DataGhostyeah well I didn't read it all but I just wondered why someone would assume that
02:44:52 Quit BigMac (Remote closed the connection)
02:51:15 Quit ompaul ("later")
02:51:33 Join BigMac [0] (
02:54:53 Join midgey [0] (
02:58:35*Domonoky has put the first, early version of his wavsplit plugin on the tracker (FS #7585 ) :-)
03:03:51 Part Domonoky
03:08:02 Quit BigMac (Remote closed the connection)
03:09:24 Join BigMac [0] (
03:16:46 Join erchino [0] (n=erchino@
03:19:14 Join aliask [0] (
03:21:45 Join chrisjs169_ [0] (
03:25:12 Quit jhMikeS (Nick collision from services.)
03:25:18 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
03:25:37Lloreaniamben: You can't roll a wheel to the "left" or "right", at any given time 50% of it is going both ways.
03:25:43LloreanYou can only roll "clockwise" and "counterclockwise"
03:27:35iambencounterclockwise is widely accepted as "left"
03:27:47LloreanDataGhost: There have been a few times in the past (at least three different that I recall) where differences in how Loader2 inited things caused unpredictable (and seemingly unrelated) behaviour in Rockbox.
03:27:49iambenever driven a car?
03:27:59Lloreaniamben: Yes, but a car isn't an MP3 player.
03:28:18iambenjust feel like arguing, or you really think im WAY off on this one?
03:28:27LloreanI don't think you're "WAY off"
03:28:35LloreanI just think that it's a rather arbitrary point.
03:28:37iambenthe current setup works, its just counterintuitive to me
03:28:50LloreanI would rather that the wheel work in opposite directions as it does now in lists, for example, because I'm left handed.
03:29:18iambenmaybe the wheel setup should be reversable then
03:29:54LloreanOr maybe the list in balance should just be reversed.
03:30:12JdGordonthat would be confusing
03:30:24iambeni think i should take this opportunity to finally dive into the source and make a patch myself
03:31:33JdGordonwe cant easily make the wheel work backwards... but we can very easily flip the list, so it would act the same...
03:31:54LloreanJdGordon: That's why I suggested the list be reversed...
03:32:18JdGordonhmm... missread, I thought i read the wheel whould be reversed
03:32:53LloreanNah, the list
03:34:04JdGordonit should only be flipped on wheel targets though
03:34:40JdGordonwhich makes it a bit more annoying
03:35:53 Quit erchino ()
03:36:41 Quit chrisjs169__ (Read error: 110 (Connection timed out))
03:37:18JdGordonmy settings_list.c is currently beinig worked in... but if you remind me later ill do this
03:38:09LloreanDon't we have an actual scrollwheel define now?
03:38:23JdGordononly set for sansa i think
03:38:28LloreanAh, true for now
03:40:27 Quit spiorf (Remote closed the connection)
03:42:25*JdGordon trying to figure out why mp3's are getting the vbrfix icon instead of the uilt in one :(
03:43:38*JdGordon loves finding the bug literally 1 sec after telling the whole world he is stuck :p
03:48:54 Join EbErT [0] (
03:58:03***Saving seen data "./dancer.seen"
04:00:07 Quit sneakums ("reblort")
04:05:06 Join sneakums [0] (
04:13:02JdGordonLlorean: do all the ipods have a scrollwheel?
04:13:26JdGordonor just the 3g and 4g keypad ones?
04:13:50saratogaJdGordon: they all have some sort of wheel
04:14:11saratogathough the actual implementation varies
04:14:29 Quit Febs (Read error: 110 (Connection timed out))
04:15:17 Join midkay [0] (n=midkay@rockbox/developer/midkay)
04:18:52*JdGordon blames iamben and Llorean for any badness coming out of most recent commit
04:20:09iambenill take the fall
04:20:24JdGordonhmm... that delta is a bit surprising
04:28:04 Join murilobsd [0] (
04:28:56murilobsdircop here ?
04:29:21scorchemurilobsd: what are you looking for?
04:30:41 Quit Thundercloud (Remote closed the connection)
04:30:44 Quit Soap (Read error: 104 (Connection reset by peer))
04:31:08 Join Soap [0] (n=Soap@rockbox/staff/soap)
04:32:00murilobsdgive me a cloak
04:32:18scorchemurilobsd: the channel you are looking for is #freenode
04:32:31scorchei dont know how you came to this channel instead...
04:32:32murilobsdscorche, okzz thankz
04:32:54iambenscorche: happened to me too
04:33:02iambenso i bought a music player and installed rockbox
04:33:08LloreanJdGordon: Last boot + 8 seems overkill.
04:33:19LloreanCan't you open the file and count how many you need?
04:46:49 Quit murilobsd (Remote closed the connection)
04:53:56JdGordonLlorean: +8 means 64 bytes wasted or something... no big deal
04:54:41LloreanIt just seems tracking it across boots and assigning +8 (how often is anyone likely to add more than one, and why can't we just add a please reboot to the "array full" message and match the number)?
04:56:30JdGordonwell.. the reason I did it this way is because we cant know how many more we need if we run out of room, and the wated ram for the extra 8 is worth it if a few types get added in one hit
04:57:40JdGordon13*8 bytes wasted MAX
04:57:41LloreanWhy can't they be counted on boot, exactly?
04:57:54JdGordonbecause they are coming from text files
04:58:10LloreanI don't see the connection, though
04:58:12JdGordonand they all need to be alloc'ed in one hit...
04:58:25LloreanSo can't the file be loaded and counted before the alloc?
04:58:37JdGordonno, that would mean having to read it twice
04:58:49JdGordon+ not every item in the file gets added
04:59:31LloreanBut wouldn't reading it twice absolutely assure there will never be a file type array full?
05:00:01JdGordonyes, but I would rather it load faster and possibly have that error instead of slow every time
05:00:54JdGordonthe only time it should ever come up is if more than 8 new types are added in one update, or someones global_status structy was corrupted and the filtetype count was left very small
05:01:17 Join Tino|Home [0] (
05:01:27LloreanWhat if there's no global status struct at all?
05:01:46JdGordonthen 128 are alloced
05:01:59JdGordonand then next time the amount used + 8 are alloced
05:04:21saratogaJdGordon: can you help me with a stupid c problem i was bugging Llorean about a couple days ago?
05:04:33JdGordonI can try
05:04:39LloreanIsn't 128 an awful lot?
05:04:43LloreanHow much do we allocate right now?
05:04:57JdGordonyes, and thats the point... its nearly double the old max
05:05:05LloreanThat's a lot of overkill
05:05:08JdGordonthat amount is only alloced the first time...
05:05:15LloreanWouldn't +10 from the old max be a good starting point?
05:05:26JdGordonthe second time only the amount actually used is alloced
05:05:46JdGordonwe could do that, but then may have to up it soon, 128 means no upping for quite some time
05:06:02JdGordonand its swcodec only.... so its not like the ram is really needed
05:06:07LloreanWell, anyone who adds an "official" filetype should up it themselves anyway
05:06:20LloreanSo it should be only "user" filetypes we care about
05:06:37LloreanAs for swcoded, I'd rather have absolutely as much RAM free as possible anyway.
05:07:09saratogaJdGordon: I want to change this struct entry to be a pointer to a buffer so that I can put it in IRAM
05:07:15saratoga fixed32 frame_out[MAX_CHANNELS][BLOCK_MAX_SIZE * 2];
05:07:21 Quit perrikwp (Read error: 104 (Connection reset by peer))
05:07:46saratogaso i changed it to " fixed32 (*frame_out)[MAX_CHANNELS][BLOCK_MAX_SIZE * 2];"
05:08:04saratogahowever I can't seem to use it properly after i do that
05:08:16JdGordonjust change it to *frame_out; and then set it to the global variable of the correct size when you have it in iram?
05:09:36JdGordonLlorean: exactly... whats to say a user wont try adding 15 types for themselves? having such a high max means they dont need an extra patch for that one little thing
05:09:58saratogathats what i attempted to do, however, when I access the varibale it doesn't seem to actually store anything in it
05:10:00LloreanJdGordon: Yeah, but with yours, it just means they need to reboot once if it's +10 by default
05:10:15saratogafor instance
05:11:03LloreanJdGordon: You could always just make it a menu entry, like the Dir Buffer
05:11:47 Quit midgey ()
05:11:48JdGordonLlorean: I tihnk your mis-understanding... very first boot, max: 128, used:15, 2nd boot max:23, used 15.... <adds 10 items> boot max:23, used: 23 <array full, reboot> max:128, used 23...
05:12:02JdGordonadding a menu item would waste probably double the ram
05:13:01LloreanIt just doesn't feel like the right way to me.
05:14:08JdGordonsaratoga: ok, i see your problem
05:14:26JdGordonyou cant use a pointer like a 2d array...
05:14:49JdGordonyou have to do blaa[row_length*row + col]
05:14:59saratogai don't see why that should be the case
05:15:03JdGordonreason is the compiler doesnt know how wide each row is
05:15:19saratogait should: fixed32 (*frame_out)[MAX_CHANNELS][BLOCK_MAX_SIZE * 2];
05:15:34saratogathe struct and buffer declarations both include that info
05:15:46JdGordonok, well imho that looks horrible....
05:15:57saratogahaha, well its 2D arrays in c
05:16:21JdGordonthe (*name) bit is confusing me
05:16:28saratogawhat baffles me is i do the same thing for another 2D array and it works fine
05:16:44saratoga"fixed32 (*coefs)[MAX_CHANNELS][BLOCK_MAX_SIZE*2];"
05:16:51saratogaworks great using array notation
05:17:41 Join perrikwp [0] (n=chatzill@
05:18:02JdGordonok, well... the //was lines look much nicer than the //is lines, and blaa[row_length*row + col] would look much much nicer than //is.... but if it works, it might be an iram thing?
05:18:05JdGordonI dont know
05:18:25 Quit TinoM| (Read error: 110 (Connection timed out))
05:19:56saratogai'm actually trying it on my linux box now and its doing the same thing
05:20:02saratogadefinately some c thing i'm not understanding
05:20:06saratogai'll continue to play with it
05:20:59*JdGordon hates pointers
05:21:46JdGordonwma_window(s, output, &s->frame_out[ch*MAX_CHANNELS + index]); ftw
05:21:48 Nick BHSPitMonkey_ is now known as BHSPitMonkey (
05:22:24*JdGordon bbl
05:22:43 Join lincoln [0] (n=lincoln@
05:23:12lincolnhi, i'm having problems with translations...
05:26:27 Join ctaylorr [0] (
05:27:30Lloreanlincoln: What sort of problem?
05:28:07 Quit XavierGr (Read error: 110 (Connection timed out))
05:28:36lincolnwell, i switch to a new theme and the menu start to show wrong titles...
05:29:17lincolni try to reload the lang file, but it doesn't works...
05:31:29LloreanWhat SVN revision are you using?
05:32:25lincolni'musing the current build 14288
05:32:56lincolnand i'm using Sansa e270
05:33:22LloreanAre you sure you extracted the whole build? There was a problem with translations until recently and if your language files weren't overwritten on update, you would still experience the problem
05:36:26saratogaJdGordon: are you still around?
05:36:40JdGordonsaratoga: just back
05:38:03lincolnLlorean, I do not update, I remove the .rockbox and extract it again... I tring this under linux and windows...
05:38:07saratogacan you look at this?
05:38:22saratogai'm trying doing it with just pointers
05:40:05JdGordonsaratoga: get rid of the [...] in the struct
05:40:13JdGordonand the & when setting the struct pointer
05:40:45JdGordonyou should comment it though after that change...
05:41:08 Quit BigMac (Remote closed the connection)
05:41:15JdGordonunless you want an array of [...] pointers?
05:41:17Lloreanlincoln: Well, themes don't / should not change language
05:41:31Lloreanlincoln: Unless you mean you're getting boxes instead of characters, which simply means the font doesn't have those characters
05:42:16 Join BigMac [0] (
05:42:21 Join Adam88 [0] (
05:43:06lincolnLlorean, this happens when I swtch to a new lang too, the "language" menu becomes " Ask - Recent only" and the "Theme" menu becomes "Cleared"...
05:43:26Lloreanlincoln: Are you absolutely sure you're using a current SVN, this was fixed recently and works for me.
05:43:30LloreanTry updating again.
05:44:05lincolni'm downloading from
05:44:21saratogaJdGordon: no its just supposed to a be 2 arrays of samples, one per channel
05:44:50JdGordonthen yeah, the paste i pasted should work
05:44:52saratogai'm still getting that warning when i init the pointer to the buffer though
05:45:19JdGordondid you get rid fo the &?
05:45:33saratogadon't i need it?
05:45:44Lloreanlincoln: If you simply change from English to English does this problem happen? And what player have you?
05:45:55JdGordonyes, you dont need it
05:46:07 Part Adam88
05:46:21saratogathen why the warning?
05:46:26JdGordonsaratoga: paste it how it is now...
05:46:47saratogathe struct member is this: fixed32 *frame_out;
05:46:49lincolnLlorean, yes... i'm using a e270 6GB
05:47:11saratogaand the initialization is: s->coefs = coefsarray;
05:47:34JdGordon... 2 different vairables there...
05:47:48 Part Llorean
05:48:35saratogathat got it
05:49:53saratogathough the music is all clicky, so i'm guessing the pointer i pass into wma_window is a bit off
05:53:46saratogaJdGordon: frame_out[ch][index] becomes s->frame_out + ch*BLOCK_MAX_SIZE + index right?
05:54:12saratogaassuming BLOCK_MAX_SIZE is the width of the dim addressed by "index"
05:54:29JdGordonyeah, or you can index it also s->fram_out[c*BLOXK_SIX + index];
05:56:32saratogaoh its "BLOCK_MAX_SIZE * 2" in the original code!
05:56:49saratogaamazing i could even understand it since I wouldn't be applying windowing to the output of the MDCT in that case
05:57:05 Quit inmotion ()
05:58:07***Saving seen data "./dancer.seen"
06:10:39 Quit lincoln ("bye")
06:16:18 Join MonkeyTamer [0] (
06:24:47 Join advcomp2019- [0] (n=advcomp2@
06:29:11 Quit advcomp2019 (Nick collision from services.)
06:29:14 Nick advcomp2019- is now known as advcomp2019 (n=advcomp2@
06:41:37 Join jhulst [0] (
06:42:53 Join midgey [0] (
06:51:02 Quit Soul-Slayer (Read error: 104 (Connection reset by peer))
06:52:56 Join ddalton [0] (
07:01:22 Quit gtkspert (Remote closed the connection)
07:06:48daurnimatorJdGordon: you around?
07:08:13TiMiD[FD]the today's build doesn't work at all on the iriver
07:08:19TiMiD[FD]it crashes when accessing the disk
07:08:29TiMiD[FD]the 2 days ago's one works fine
07:10:04 Join gtkspert [0] (
07:14:32aliaskdaurnimator: What did you want yesterday?
07:14:43 Join Llorean [0] (
07:15:52JdGordondaurnimator: hey
07:16:20JdGordonTiMiD[FD]: ?
07:19:44 Join webguest30 [0] (i=46101058@gateway/web/cgi-irc/
07:20:28 Quit webguest30 (Client Quit)
07:20:36 Join webguest86 [0] (i=7aa71f57@gateway/web/cgi-irc/
07:21:33webguest86anyone here who can answer a few questions about GIGABEAT
07:21:43LloreanWhat are the questions?
07:21:53webguest86hi llorean
07:22:26webguest86wud rockbox enable video formats on the F series
07:24:25TiMiD[FD]JdGordon: I just made svn update, compiled, installed
07:24:49TiMiD[FD]didn't reset the settings though ...
07:24:55TiMiD[FD]but the disk didn't stopped spinning
07:25:10JdGordonsounds like you have the fun task of finding out which commit broke it :D
07:25:29TiMiD[FD]I'm quite busy today
07:25:46TiMiD[FD]I go on a trip tomorrow so I don't have a lot of time
07:25:52TiMiD[FD]I'm already late on my planning
07:26:07TiMiD[FD]I'll try yesterday's build
07:26:45 Quit webguest86 ("CGI:IRC (EOF)")
07:27:10*Llorean is probably not suitable for answering questions at the moment, it seems. :(
07:27:15 Join webguest86 [0] (i=7aa71f57@gateway/web/cgi-irc/
07:27:27 Quit hcs (Read error: 110 (Connection timed out))
07:33:35amiconnTiMiD[FD]: Current builds are working fine on all my targets, which include an iriver H1x0
07:41:29amiconnTiMiD[FD]: Did you reconfigure and make clean?
07:45:58saratogamarkun: do you know if we're using the stock tremor code or the tremor lowmem code for vorbis?
07:58:09***Saving seen data "./dancer.seen"
07:58:42*amiconn didn't check the filetype thing yet
08:00:07 Quit saratoga ("CGI:IRC (Ping timeout)")
08:07:17ddaltonI can't build a voice file. on cygwin. What do I need.? If there is something wrong with english.lang will this affect the building. I am using the latest revision
08:10:09amiconnJdGordon: Your filetype handling is the culprit.
08:10:23amiconnWith it in place, H1x0 hangs at the rockbox logo, disk spinning
08:11:34*amiconn wonders whether JdGordon *ever* tests somewhat thoroughly before commit
08:12:25markunscorche: stock
08:12:36amiconnAnd I think the problem is that the setting was just added to the nvram data, without bumping the version. The setting can have *any* value after load
08:12:47amiconnJust relying on that is badness
08:12:51scorchemarkun: huh?
08:12:54markunhm, that was for saratoga
08:13:01 Join DarthShrine [0] (n=Angus@pdpc/supporter/student/DarthShrine)
08:13:05amiconnYou can't buffer_alloc 410238972 bytes
08:13:58 Join DarthShr1ne [0] (n=Angus@pdpc/supporter/student/DarthShrine)
08:14:16 Quit DarthShrine (Nick collision from services.)
08:14:18 Nick DarthShr1ne is now known as DarthShrine (n=Angus@pdpc/supporter/student/DarthShrine)
08:15:18 Quit ctaylorr (Read error: 110 (Connection timed out))
08:17:51 Part toffe82
08:23:42 Quit ddalton ("I was using BOFHNet IRC version 1.2 by fmillion - get your copy today from !")
08:23:51 Join ddalton [0] (
08:26:29 Join hcs [0] (n=agashlin@rockbox/contributor/hcs)
08:26:34 Quit DarthShrine ("Leaving")
08:27:28 Join Rob2222 [0] (
08:30:59ddaltonI think I fixed my voice building problem
08:35:27amiconnJdGordon: I can confirm that your filetypes handling commit is the problem. With it in place, H1x0 hangs on boot, always, and even after fixing some quirks myself
08:35:53amiconnReverted to svn immediately before it, and it works like a charm again
08:40:58 Quit webguest86 ("CGI:IRC (EOF)")
08:44:43JdGordonhmm... odd
08:44:55 Quit Rob222241 (Read error: 110 (Connection timed out))
08:46:02JdGordonIm sure I remember there was a size value for the nvram settings so you could add settings on the end and it wouldnt corrupt it
08:46:50*JdGordon was right
08:46:58JdGordonthere is... must be a bug with it though
08:47:36amiconnEven with the nvram version bumped, a sane maximum in place, fixing an alignment problem (that only hits colour targets), and setting the default earlier, it still hangs
08:48:53JdGordonim not arguing about the max again, 128 is perfectly sane
08:49:36amiconnI meant a sane maximum if the loaded value is going wild
08:49:47amiconnI used 1000
08:50:21amiconnCalling buffer_alloc with an nvram value without range checking doesn't sound like a good idea
08:51:12JdGordonoh, ok
08:51:49JdGordonwell, found the bug in settings so vnram settings can be added without bumping the version
08:52:27 Join TiMiD[FD1 [0] (n=TiMiD[
08:53:10*Llorean wonders why 96 isn't sane, it's still a "lot"
08:53:15JdGordonwhats the alignemnt issue?
08:54:06amiconncustom_colors = (int*)buffer_alloc(sizeof(int)*max_types+1);
08:54:17amiconnYou certainly want custom_colors = (int*)buffer_alloc(sizeof(int)*(max_types+1));
08:54:22JdGordonah yes
08:56:42JdGordonok, well its _still_ working fine here, so i'll commit, and bump the version to stop bug reports
08:58:12amiconnHmm, but even with those measures, and some more, it doesn't work on H1x0
08:58:44amiconn(although I don't know what you found in nvram settings - maybe that fixes it)
08:58:55*amiconn hates needing a paper clip or equivalent
08:59:04JdGordonjust checked on my sansa and it works, but yeah, I'm a bit confused
08:59:20JdGordonill put up a patch...
09:00:56 Join Siku [0] (
09:01:54JdGordondoh, yeah im a dill... bumped the nvram version so didnt even test the setting fix... 1 min
09:01:54JdGordonfortunatly i purposly backup up my nvram.bin before updating
09:02:00 Join perrikwp_ [0] (n=chatzill@
09:02:19 Join RaRe` [0] (n=Laffin_B@
09:02:24amiconnWhat target(s) do you have, again (apart from the sansa)?
09:02:25 Quit ddalton (Read error: 104 (Connection reset by peer))
09:02:42JdGordonyep, definatly working now
09:02:47JdGordonh300 and nano, but neither are accessable atm"> is the fix
09:04:08JdGordonif you modify the nvram.bin so the 4th byte is 1 less than the current that should show its fixed
09:04:45 Quit TiMiD[FD] (Read error: 110 (Connection timed out))
09:07:46 Nick TiMiD[FD1 is now known as TiMiD[FD] (n=TiMiD[
09:08:47 Join haemmy [0] (i=stefan@
09:13:20amiconnJdGordon: Still the same (hangs)
09:13:52amiconnIt did the dircache scan in the froreground, so the nvram invalidation worked
09:14:18*amiconn sighs
09:14:25 Quit perrikwp_ (Read error: 104 (Connection reset by peer))
09:14:46amiconnI need to work on another aspect, but current rockbox is definitely borked and hence unusable for that :(
09:15:00JdGordongood old version control....
09:15:15JdGordonunless the another aspect is viewports? then ill revert the whole commit to get you working :D
09:15:35amiconnNo, shutdown
09:16:15amiconnI need to make it work the same way as USB handling. The shutdown message is broadcast, and each thread does what it needs and then acknowledges
09:16:43amiconnI need that for making suspend work.
09:17:37amiconnI don't even know yet when I'll find the time to do viewports :(
09:17:57JdGordonis suspend going to happen instead of shutdown? I think it should only happen when it turns of because of no acitivity
09:18:52amiconnOn 1st/2nd gen suspend is the only possible way
09:19:00amiconnThe hardware has *no* shutdown
09:19:14JdGordonrest of PP?
09:20:31 Quit perrikwp (Read error: 110 (Connection timed out))
09:20:34JdGordonok, im going to commit this fix and then find my h300 to hunt this down
09:20:43 Quit RaRe (Read error: 110 (Connection timed out))
09:23:29amiconnye, H300 also hangs
09:27:02JdGordonok wierd.... i see what you mean with the hanging
09:27:13JdGordonmaybe its dircache related which would explain why I didnt see it on the sansa
09:27:17 Join ddalton [0] (
09:27:44amiconnMaybe... should try on a target where I have dircache disabled
09:28:21*JdGordon checks
09:28:26amiconnMeanwhile I have dircache enabled on targets with slow hdd (that excludes all flash targets and mini G2), and at least 32MB ram (that excludes X5)
09:30:23JdGordoninteresting... freeze on shutdown
09:31:01JdGordongrr... not enough batt to test properly
09:34:45 Quit EbErT ()
09:43:07JdGordondamn, this is going to be painful.. its a dircache issue almost certainly... and it doesnt effect the sim
09:45:02JdGordonholding rec on the h300 so settings get reset makes it go into the rec screen, and then a few sec after leaving that (in the main menu) it freezes!
09:46:03*amiconn doesn't remember whether he ever used settings reset on iriver
09:51:14JdGordonhmm... I think maybe that commit was useless... settings_reset() is called before the nvram settings are loaded, so it should have got a normal value anyway
09:58:13***Saving seen data "./dancer.seen"
10:02:23 Join davina [0] (
10:03:33 Join pixelma [0] (i=57b9d368@rockbox/staff/pixelma)
10:09:17 Join desowin [0] (n=desowin@
10:12:05*JdGordon annoyed
10:13:39ddaltonsorry i was in the wrong window
10:20:34JdGordonoh goody, the bug isnt in the filetypes commit... its definatly a problem with the nvram settings... just need to find out why
10:25:07 Quit MonkeyTamer ("ChatZilla [Firefox]")
10:26:01TiMiD[FD]JdGordon: may that be related to the dircache bug ?
10:26:17JdGordonit could be
10:26:55TiMiD[FD]it seems taht at some point at least, the settings struct gets corrupted
10:28:36 Join fm2 [0] (i=543ab6f2@gateway/web/cgi-irc/
10:29:58 Quit tchan (Read error: 110 (Connection timed out))
10:30:56fm2JdGordon: ping
10:31:40fm2JdGordon: since GodEater isn't here, would you like to discuss some things about the shortcuts plugin?
10:32:00JdGordoni only commited it...
10:32:43fm2JdGordon: ah, ok. Then I'll wait for him to come in. Or will try later.
10:33:34fm2But in the hope that he reads log, I'll ask one more question (some were already asked on Fri, look for fm2)
10:34:30fm2In the plugin, the extension ".link" is hard coded. OTOH we want the association of file exts to viewers be configurable in the config file.
10:36:19fm2This is a contradiction IMHO, and it can be resolved by splitting the plugin into two. One would be only responsible for adding files to the default shortcuts file and would be called via the context menu (add). The other would load the files and show the links there. It would be configured as the viewer for .link files.
10:36:51fm2Then we could just remove the code that checks whether the file is a .link file from the plugin.
10:38:31 Join spiorf [0] (
10:38:37 Quit midgey (Read error: 113 (No route to host))
10:38:39JdGordonamiconn: Im 99% sure the bug isnt in the filetypes part of the commit, but rather in the nvram part, so to get a stable build again should I change it to write the setting to config.cfg instead of nvram.bin untill the fix is figured out?
10:39:06JdGordonreverting the whole commit is the other option, which imho isnt needed
10:39:16amiconnI'd rather revert the commit
10:40:10JdGordonthats not where the bug is though
10:41:41JdGordonamiconn: does this crash/freeze the recorder also?
10:41:55amiconnDidn't try...
10:42:52 Part fm2 ("@GodEater: please read the logs.")
10:48:17 Join My_Sic [0] (
10:51:02amiconnJdGordon: At least 2 reasons speak against saving status in config files: (1) Loading a .cfg would restore the status as well, which would be very wrong and unwanted
10:51:45amiconn(2) Saving a .cfg often is a lot more overhead than saving nvram.bin. Even more overhead is caused on targets with real nvram
10:52:37JdGordonhmm... ok
10:54:37amiconnI also wonder why you think the problem lies in the nvram thing
10:55:33amiconnBoth nvram and dircache are there for quite some time, and the freeze started with the filetypes commit...
10:55:53JdGordoncrap.... i think i found the problem
10:56:00JdGordonnvram is only 44bytes..
10:56:12amiconnBut the irivers have the advantage that you can use logf() with the remote lcd
10:56:24JdGordonall the current vnram settings are 4 bytes... and there were 9, now 10
10:56:30JdGordonbloody buffer overrun
10:56:54amiconnTake care about the size in general. nvram *must* not grow beyond 44 bytes
10:57:41amiconn(well, at least on recorders)
10:57:55JdGordonit alwaysread/ writes the full 44 bytes, so thats safe.. the rest of the code is causing the problem
10:59:13JdGordon is our problem
10:59:21JdGordoncan any of them use less than 4 bytes?
10:59:31JdGordonfiletype_count can, but thats still over the 44 bytes
11:01:10*JdGordon is open to suggestions here
11:05:13 Join ender` [0] (
11:06:01 Join kretender [0] (
11:06:39 Join robin0800 [0] (
11:08:55 Join reeyal [0] (i=d2317d7d@gateway/web/cgi-irc/
11:09:12JdGordondinner time, but the bug is found, so itll be fixed shortly
11:09:27reeyalhi there, i think i just killed my iriver H340. can anyone help me pls? ;(
11:10:24 Quit robin0800 (Client Quit)
11:10:38 Join robin0800 [0] (
11:11:18 Quit robin0800 (Client Quit)
11:11:34 Join robin0800 [0] (
11:12:46GodEater_reeyal: killed it how? I'm not an H300 expert, but I'm willing to listen...
11:13:14 Join leftright [0] (i=d9e1c015@gateway/web/cgi-irc/
11:15:49leftrightwith latest build, unable to get version number, when player reboots, I get the rockbox logo and file type array full splash screen then the system locks, reset via pin, it wont reboot after that
11:16:31leftrightH140, flashed but set to load firmware froom HD
11:16:57GodEater_leftright: it's a known issue as of this morning - it's being worked on
11:17:05GodEater_in the meantime, revert to an older build
11:17:26leftrightstable build flag ;)
11:18:32 Part leftright
11:20:21 Quit haemmy (Read error: 104 (Connection reset by peer))
11:21:01reeyali don't know what happened. i put rockbox onto my iriver and it worked for a second but then all the buttons froze up and i had to reset. then when i rebooted no buttons worked. i tried booted into the iriver o/s but that didn't work. i plugged the usb cable in and it said usb boot. then that disappeared and now nothing works. can't switch on or connect to the computer
11:21:02amiconnJdGordon: For the filetypes: how many bytes is the filetype array in total?
11:21:41amiconnMaybe the nvram setting is unneeded if it's small enough (say, around one KB).
11:22:02reeyalgodeater: it was saying that filetype array full msg
11:22:18amiconnLet the function read it into a temporary array (on stack) and count the types as you go. Then buffer_alloc() and memcpy()
11:23:15amiconnBut if it's small enough, I don't see the necessity of dynamic handling at all. The old static array worked well enough
11:25:49pixelmaGodEater_: re. the forum "charge" post: isn't it in the wrong section? And I'm not sure if I understood correctly but I thought he meant charging with a wall charger - isn't that supposed to just work with the Ipods (thought Rockbox can detect if it's connected to a USB data line or not)
11:27:55pixelmaI meant just plugging the charger in without holding a button
11:28:32amiconnJdGordon: I'd say just revert the commit, and set the filetype count somewhat more generous. 64 types for hwcodec and 256 types for swcodec should be sufficient for quite some time to go
11:29:07 Quit amiconn (" HydraIRC -> <- Like it? Visit #hydrairc on EFNet")
11:31:47TiMiD[FD]reeyal: uh ?
11:31:54TiMiD[FD]you mean it doesn't power up ?
11:32:14reeyalyes, nothing happens :(
11:32:27TiMiD[FD]even after a reset ?
11:32:49reeyalreset doesn't do anything now. i did it a few times and it was resetting but now seems to have died
11:33:01reeyalit was saying 3.7V battery. is that low?
11:33:18TiMiD[FD]I don't know I've got an h140
11:33:38Llorean3.7v isn't a full charge by any means.
11:33:40TiMiD[FD]once it behaved like yours while I was in the train, but a charging did the trick
11:34:18TiMiD[FD]I guess there is some kind of hw protection that prevent powering up when battery it too low
11:34:18reeyali've plugged it into the charger but it's not saying anything. do you think it's charging?
11:34:43TiMiD[FD]usually does it says something ?
11:34:48TiMiD[FD]when you hcarge ?
11:35:05TiMiD[FD]I'm not an h300 expert, I only saw one in my life ...
11:35:43reeyalyes it normally says charging but that's with the iriver o/s. this is the first time i've used rockbox, but at the moment, i'm getting no response from it at all
11:36:06TiMiD[FD]on my h140 there is a led that indicates wether it's charging or not
11:36:25TiMiD[FD]I don't know if rb has that kind of screen ...
11:36:33TiMiD[FD]try to powerit up
11:36:48reeyalnothing happens
11:36:50TiMiD[FD]with the plug it should work (again that's how my h140 behaves)
11:37:20TiMiD[FD]it's highly impossible that rb did anything to your hardware anyway
11:37:55TiMiD[FD]btw the today's build is completely broken, you should try yesterday's one if you power it on one day
11:38:21TiMiD[FD]someone with a h300 could help you better than I
11:38:30TiMiD[FD]maybe the battery is just too low
11:39:24reeyalbut i've got it plugged in and it doesn't look like it's charging
11:39:53TiMiD[FD]maybe it needs software to boot first to charge
11:39:59TiMiD[FD]I don't know how it works
11:41:57TiMiD[FD]anyone with a H300 here ?
11:44:33reeyalthanks for listening guys. i will leave it on charge overnight and see whether anything happens.
11:44:36pixelmaJdGordon has one (but having dinner, he said)
11:44:36 Quit reeyal ("CGI:IRC (EOF)")
11:44:58*JdGordon back
11:45:47 Join ompaul [0] (n=ompaul@freenode/staff/gnewsense.ompaul)
11:45:55pixelmajust when he left...
11:46:07JdGordonoh well
11:54:00JdGordonbuilds are fixed again... /me apologises for the inconivience :p
11:58:16***Saving seen data "./dancer.seen"
12:07:11ddaltonTiMiD: I have one why?
12:07:29ddaltonJdGordon: what was wrong with them?
12:07:43JdGordonbuffer overrun
12:08:21ddaltonwhich caused what?
12:10:22 Nick Tino|Home is now known as TinoM (
12:11:15 Join amiconn [0] (
12:12:34TiMiD[FD] ddalton nah I was trying to help that guy who seems to have bricked his h300
12:12:51TiMiD[FD](read the log above)
12:17:11 Join Guile`` [0] (n=Guile@
12:18:41ddaltonIs p7561 ok?
12:19:58 Quit ddalton ("I was using BOFHNet IRC version 1.2 by fmillion - get your copy today from !")
12:20:04 Join ddalton [0] (
12:28:46 Part ldarby
12:30:21 Quit pixelma ("bbl")
12:42:21 Join wippeout [0] (n=wippeout@unaffiliated/wippeout)
12:42:50wippeoutdoes the devices you manage support the mtp ?
12:43:39 Join petur [0] (n=petur@rockbox/developer/petur)
12:47:31 Join sp00n_ [0] (i=3ef892ec@gateway/web/cgi-irc/
12:48:41sp00n_can someone help me with rockbox?
12:48:54peturjust ask the question
12:49:58sp00n_well, where I can find the right version for ipod nano? or does it matter ?
12:50:16peturlook at the current build page
12:50:21ddaltonMake sure it is a first gen
12:50:31sp00n_my ipod?
12:50:38sp00n_it's not.
12:50:47peturipod nano 2nd gen?
12:50:53sp00n_that's why i wonder should i pick it
12:50:57peturnot supported
12:51:01ddaltonyour ipod must be a first gen otherwise it won't work
12:51:35peturcomplete new hardware and eccrypted firmware, Jobs doesn't want you to use rockbox
12:51:49sp00n_that sucks
12:52:13sp00n_itunes sucks hard and i thought this would have solved my problems
12:52:21 Join Rondom [0] (
12:53:09sp00n_is there coming a supported build for 2nd gen at all?
12:53:48rasherNever say never, but as it looks now, it's quite unlikely.
12:54:14ddaltonapparently it is completely different from the first gen but I don't think any time soon. Anyway you could make a port to it maybe. I don't know. Just search for info.
12:54:47ddaltoncan you code?
12:55:18sp00n_not rly
12:55:28ddaltondo you know c at all?
12:55:55sp00n_you mean c or c++
12:56:24ddaltonrockbox isn't in c++
12:57:01sp00n_well, I used to code c more than 3-4kk ago
12:57:14sp00n_I don't remember much
12:57:32sp00n_cuz I started to code c++
12:57:46sp00n_which i was told to be better language :)
12:58:02ddaltonwell you will need to code c. But it is a lot of work from what I can tell.
12:58:37sp00n_ok, I'm not so good at c so I think I won't even try. I have no idea what to even do.
12:59:47ddaltonWell you could just get information on the hardware if there is a port and if developers have one and time they my try to port it.
12:59:50advcomp2019i am in the same boat sp00n_ if i knew another about c myself i would help with the r series
13:00:30ddaltonwell if I could code better I would be doing a lot more things. mainly to the voice. (Still learning)
13:01:17sp00n_what do you mean with "port" exactly, (I'm not a native english speaker :)
13:01:36 Join PaulJam [0] (
13:02:06ddaltonwell make rockbox run on a certain player. In your case the ipod nano second gen.
13:02:56 Quit sp00n_ ("CGI:IRC (EOF)")
13:03:18ddaltonand look for the heading "Ports "
13:03:34 Join sp00n [0] (i=3ef892ec@gateway/web/cgi-irc/
13:05:26ddaltonI don't think there even is a port to the second gen nano. Is there?
13:06:41sp00ni'm not sure
13:07:13aliaskNot at the moment
13:07:22aliaskIt will be a very difficult task
13:08:33sp00nhave some1 tried ipodlinux?
13:09:00scorchesp00n: that is offtopic
13:09:13 Join Domonoky [0] (
13:09:26sp00noh ok
13:09:55wippeoutno port for zen vision M ? :)
13:10:31scorchewippeout: you can see the thread about it in the forums
13:11:55PaulJamhey, what happened to the clock plugin? why was the digital mode (where you had those numbers consisting of segments) removed?
13:12:26scorchePaulJam: talk to TiMiD[FD]
13:12:41 Quit davina (Remote closed the connection)
13:13:31midkayPaulJam: some deliciously confusing new navigation system by TiMiD[FD]. try up/down in the digital "plain" mode.
13:13:54 Quit spiorf (Remote closed the connection)
13:14:05 Join davina [0] (
13:14:08PaulJamah, thank you. that works
13:14:32 Join spiorf [0] (
13:14:47midkaynp :)
13:14:59 Quit davina (Remote closed the connection)
13:16:15 Quit spiorf (Client Quit)
13:17:32ddaltonok forgot about that.
13:17:32 Join pixelma [0] (i=57b9d368@rockbox/staff/pixelma)
13:17:36 Join Nico_P [0] (n=nicolas@rockbox/developer/NicoP)
13:18:04 Join advcomp2019- [0] (n=advcomp2@
13:18:18 Quit advcomp2019 (Nick collision from services.)
13:18:22 Nick advcomp2019- is now known as advcomp2019 (n=advcomp2@
13:19:38 Quit sp00n ("CGI:IRC (EOF)")
13:26:02petur"there is no spoon"
13:27:18*petur excuses himself for the spam marker in his dev-ml post
13:28:14*scorche looks for said spam marker
13:38:18Nico_Ppetur: why "[Spam][100.0%]" ? JdGordon's mails are spam to your filter ?
13:38:40Nico_P(specifically, why the 100% ?)
13:38:48peturblacklist :)
13:39:08peturI'll explain
13:40:09peturit's a way I use to not have new e-mail alerts for mailinglists. They get marked as spam by the filter but are afterwards moved to the correct folder
13:40:47amiconnStrange mechanism...
13:41:00*petur uses outlook - explains it all
13:41:35*amiconn uses outlook at work (exchange server), but would never use it without that single reason
13:43:08 Part wippeout
13:45:20petursyncing contacts with PDA is my only reason left atm.... no time to look into changing the whole mail/pda setup
13:46:07*amiconn has no pda and doesn't want one either
13:46:21scorcheterribly handy
13:46:30*amiconn waits for sim build to complete...
13:47:14peturplus I have all passwords on it, In case I have to go to somebody and he forgot his password...
13:47:33scorchenone your own, i hope =)
13:48:04 Quit ddalton ("I was using BOFHNet IRC version 1.2 by fmillion - get your copy today from !")
13:48:49*petur has lots of passwords of relatives and business, it's that "hey, you know something of computers, help me out" thing
13:49:04*petur stops OT talk
13:49:25JdGordonpetur: I tihnk commiting them all isnt a very good idea
13:49:40amiconnThe passwords? ;)
13:49:53peturI was just stating what was said on devcon
13:50:00peturyou should come next time ;)
13:50:00*scorche wishes more people join #rockbox-community so we can have more of these moments
13:50:09 Join Febs [0] (
13:50:37peturfollowing this channel is more than enough for me, thanks
13:53:13 Quit aliask ("ChatZilla [Firefox]")
13:53:16 Join p3tur [0] (n=petur@rockbox/developer/petur)
13:53:32 Join aliask [0] (
13:53:44*petur will try wifi-reception outside :)
13:53:54 Quit petur ("I'm still here :)")
13:54:05 Nick p3tur is now known as petur (n=petur@rockbox/developer/petur)
13:57:05JdGordonNico_P: so hows mob going?
13:58:17***Saving seen data "./dancer.seen"
14:07:27thegeekspeaking of pda's, the unofficial axim x50v windows mobile 6 update is insanely fast, all the "windows mobile" lag is _completely_ gone, it is imho a must have for any x50v owner
14:08:02scorchethegeek: offtopic...
14:08:21thegeekI know, which is why I made it a single line, with no followup
14:08:34scorchea single line is still offtopic
14:08:42thegeekindeed, I'm sorry
14:09:04amiconnrrrrrrr, hanging build
14:09:43amiconnBagder: ping...
14:12:43*Domonoky is trying to optimize his wavsplit plugin.. does anybody know how big a good blocksize for reading/writing on fat (in rockbox) is ?
14:12:50Nico_PJdGordon: slowly atm
14:13:04Nico_PI was doing more work when on vacation :s
14:13:28*petur wonders if it was really needed to set the filetypes array to 128...
14:13:57 Join DerPapst [0] (
14:14:50*amiconn even suggested 256
14:15:00amiconn(for swcodec)
14:15:08peturwhat, you? :p
14:15:27Nico_Pwhat size is a filetype ?
14:15:56amiconn16 bytes in the filetype array, + 4 bytes icon, + 4 bytes colour (on colour targets only
14:17:45Nico_Pok, so not big enough to require strict economy, right ?
14:18:12amiconn256 types would be 6KB on colour targets, and 5KB on monochrome
14:18:34amiconnOn hwcodec this would be a bit much, hence I suggested 64types there
14:19:05 Join leftright [0] (i=d9e1c015@gateway/web/cgi-irc/
14:19:17leftrightbuild table needs a "Kick"
14:19:49 Join MySic [0] (
14:20:38 Quit My_Sic (Read error: 113 (No route to host))
14:20:40 Part leftright
14:24:34 Join BigBambi [0] (n=Alex@
14:29:39 Join ctaylorr [0] (
14:38:36 Join Nick_Brackley [0] (i=cb57494b@gateway/web/cgi-irc/
14:39:28 Quit kretender (Read error: 110 (Connection timed out))
14:41:21 Join austriancoder [0] (n=austrian@rockbox/developer/austriancoder)
14:42:31 Join Thundercloud [0] (
14:43:10*petur waves at austriancoder
14:43:20austriancoderpetur: hi
14:45:50petur austriancoder: where was that register dump again?
14:46:19peturand could you bring some order into your directory of patches?
14:46:58 Join XavierGr [0] (
14:47:29peturthe last one (looking at date) is Aug 9? doesn't apply and build...
14:47:51pixelmaBigBambi: who could forget this one of the GoldenQuotes... ;)
14:48:27BigBambiHaha, yeah!
14:48:49austriancoderpetur: I know but i dont generate every day a patch..
14:49:10BigBambiWhy hasn't anyone done radio for H10 yet? I paid a lot of money for rockbox, and expect all the features to work...
14:49:31peturaustriancoder: you don't have to, but that list is a bit of a mess right now
14:51:15austriancoderpetur: have done a clean up and generated a new diff
14:52:13peturnow where was that usb regsiter dump again?
14:52:18TiMiD[FD]BigBambi: it's a contribution based project, not a company ...
14:52:39BigBambiTimid I know, it was a joke taken from a forum topic
14:52:41austriancoderpetur: pastebin...
14:52:56TiMiD[FD]ah ok
14:52:57BigBambinote "I paid a lot of money"
14:53:27peturaustriancoder: still have it around somewhere?
14:53:37austriancoderpetur: but I have the used register here.. wait
14:53:48pixelmaTiMiD[FD]: see the link above
14:54:01TiMiD[FD]this post is indeed quite funny ^^
14:54:35DerPapsti expect more from you guys too
14:54:38BigBambiThe joy of people not used to open source, contribution, free projects!
14:55:05DerPapsti want dualboot on my iPod as well
14:55:23BigBambiI want my gigabeat to make me lunch
14:55:35BigBambiBut I may have to resort to doing it myself
14:56:32DerPapsti tried to install rockbox on my toaster.. but i couldn't find the cable
14:56:36FebsI was reading through that thread thinking, "why didn't I just ban this guy?!" Then I got to post 63. :)
14:56:47BigBambiI shouted at it for a bit, but not even a cup of tea was forthcoming :(
14:56:51BigBambiFebs: :)
14:58:34pixelmaTiMiD[FD]: by the way - could you make the clock dial (the circular one) deal with the rectangular pixels of the Archos displays - currently it's oval because on those displays the width of the pixels is 80% of their height (if I saw correctly that was the same before your changes but since my target - the Ondio - doesn't have an RTC I didn't discover it yet)
14:58:56TiMiD[FD]pixelma: ah ?
14:59:06TiMiD[FD]I'll take a look
14:59:11pixelmathe clock in the background of the "bounce" plugin does
15:00:18pixelmawill then look a bit weird in the sim - just so that you can imagine it better: the chessboard is 80x64 pixels but looks perfectly square on these displays
15:00:54pixelmathanks :)
15:02:33TiMiD[FD]II didn't told you it was for free ^^
15:03:28TiMiD[FD]so basically
15:04:00TiMiD[FD]which dimension has the longest pixels ?
15:04:13TiMiD[FD]and in which proportion ?
15:04:34TiMiD[FD]I've better look at the chessboard
15:05:35 Quit PaulJam (".")
15:05:56pixelmawell all of the 112x64 pixels are the same - with an aspect ratio of 8:10 (width:height)
15:06:09pixelmaor 4:5
15:07:17TiMiD[FD]should'nt be difficult I guess
15:09:11 Join Lear [0] (i=chatzill@rockbox/developer/lear)
15:12:20 Quit rasher (
15:12:20 Quit Hadaka (
15:13:58 Join rasher [0] (
15:14:30 Join davina [0] (
15:14:41 Quit davina (Remote closed the connection)
15:17:09pixelmaTiMiD[FD]: as I said, maybe it's helpful to look at the bounce demo plugin too - it has a correction for it (which is why it looked so weird for a while on all the other targets)
15:18:11 Quit iamben (Read error: 104 (Connection reset by peer))
15:18:52 Quit MySic ("Miranda IM! Smaller, Faster, Easier.")
15:20:06 Join davina [0] (
15:24:15peturaustriancoder: how about dumping the contens of the usb memory range both in e200tool and rockbox for a compare?
15:24:41 Join merbanan [0] (n=banan@
15:24:56austriancoderpetur: I dont want to run e200tool... as i dont want to break my player
15:25:07TiMiD[FD]pixelma: I implemented it
15:25:18TiMiD[FD]it's only one line ...
15:25:37TiMiD[FD]now it looks elongated on the width
15:25:46*petur is bidding on a sansa and an H10...
15:26:11TiMiD[FD]I don't have an archos so I cannot test
15:28:06pixelmaTiMiD[FD]: I can fetch amiconn's recorder to test
15:28:20LearWouldn't a simulator do?
15:28:37TiMiD[FD]I'm too lazy to put a pathc on the tracker
15:28:45TiMiD[FD]I'llcommit directly
15:28:58TiMiD[FD]besides I've got my luggage to prepare
15:29:26pixelmawell you should see a difference but you can't decide if it'll look completely right in the sim
15:29:40TiMiD[FD]wanna test ?
15:29:43pixelmait should look wider
15:29:57TiMiD[FD]I applied the same correction as in the chess plugin
15:30:33pixelmathe build system is hanging at the moment anyways...
15:30:57pixelmaso I could take the time to test
15:31:49pixelmayou'll probably have to mail me the patch because I doubt dcc send works in the webclient
15:32:51 Join My_Sic [0] (
15:33:23 Quit Nick_Brackley ("CGI:IRC (EOF)")
15:34:58peturunless you want to have the binary from the server, committing should make it available - the svn server is not down
15:36:27peturaustriancoder: can you dump the usb memory range in rockbox and put it online?
15:37:03austriancoderpetur: shure.. but do expect from it?
15:37:17austriancoderbut what do you expect from it?
15:38:37peturno idea, compare with the datasheet, maybe something's different? Better than doing nothing.
15:39:32 Quit aliask ("ChatZilla [Firefox]")
15:42:36austriancoderpetur: I found out what registers are used via disassembling. I will dumb registers before modifing and after. Also I will track memory address of data to send. With this we should get a clear picture if all register contents are okay
15:42:51 Join Naked [0] (i=naked@
15:42:56 Nick Naked is now known as Hadaka (i=naked@
15:42:58***Server message 505: 'logbot :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( )'
15:44:08 Join hannesd_ [0] (
15:44:14 Quit hannesd_ (Remote closed the connection)
15:51:34dionoeaamiconn: do you know how to open the apple_os.bin file with IDA Pro? (It can't find any valid ARM instructions with IDA's default settings)
15:56:10 Quit JdGordon ("Konversation terminated!")
15:56:11 Quit DerPapst (Read error: 104 (Connection reset by peer))
15:56:19TiMiD[FD]pixelma: how does it look ?
15:57:05 Join roolku [0] (
15:58:21***Saving seen data "./dancer.seen"
15:58:44 Quit jhulst (Remote closed the connection)
15:59:19 Quit hannesd (Read error: 110 (Connection timed out))
16:09:12 Join thehungus [0] (
16:09:26 Join DerPapst [0] (
16:09:49 Join darkraven [0] (
16:10:21 Part thehungus
16:15:19 Join x1jmp [0] (
16:15:52 Nick chrisjs169_ is now known as chrisjs169 (
16:24:20 Join miepchen^schlaf [0] (
16:28:47 Join brazzzil [0] (
16:30:43*petur moves back inside as it starts to rain :(
16:32:06 Quit idnar (Nick collision from services.)
16:32:08 Join idnar_ [0] (i=mithrand@unaffiliated/idnar)
16:36:55 Join robin_0800 [0] (
16:42:39 Quit brazzzil ("Leaving")
16:47:57 Join webguest07 [0] (i=d2317d7d@gateway/web/cgi-irc/
16:49:01 Quit robin0800 (Read error: 110 (Connection timed out))
16:55:42 Quit Hadaka (
16:59:19webguest07hi, does anyone here have an iriver h340?
16:59:59*petur raises hand
17:00:20peturan h380 actually :)
17:00:51webguest07i think i just killed my h340
17:01:01peturoh... how?
17:01:27webguest07i put rockbox on for the first time. it worked for a minute, then all the keys froze up and i had to press the reset button
17:01:47webguest07then it kept loading but no buttons would work
17:02:18peturdoes OF still load if you hold REC while powering on?
17:02:37webguest07i tried that, and it wouldn't load into iriver.
17:02:57peturhold both rec an play long enough
17:03:01Nico_Pwebguest07: you have to keep pressing play
17:03:02TiMiD[FD]webguest07: aren't you "reeyal" from some hours ago ?
17:03:02webguest07but now the whole thing is dead and i can't turn it or or anything
17:03:15webguest07yeah i'm reeyal from a few hours ago
17:03:30peturinsert charger?
17:03:31webguest07i can't believe that i managed to kill my iriver
17:03:43webguest07i left it charging for the last 5 hours
17:03:46webguest07still nothing
17:03:52TiMiD[FD]I don't think it's dead
17:03:56webguest07any ideas?
17:04:17TiMiD[FD]at least putting rb on it couldn't have kill it that easily
17:04:33TiMiD[FD]maybe you should try to open it and charge the battery manually ?
17:04:39webguest07no, but what can i try doing?
17:04:48peturTiMiD[FD]: ?
17:04:49webguest07how do you do that?
17:04:57TiMiD[FD]I don't know
17:05:01 Join hannesd [0] (
17:05:29TiMiD[FD]if it doesn't power up, it could be a battery related issue
17:05:37peturwebguest07: remove from charger, reset, then hold rec and press play for at least 2 seconds
17:05:42TiMiD[FD]maybe your battery died ...
17:05:56webguest07yes maybe battery died
17:06:21TiMiD[FD]try to open it and test the voltage
17:06:57 Quit chrisjs169 (Remote closed the connection)
17:07:13webguest07the only other symptom i have, is that recently the up button has been playing up and acting as though i've pressed the back button. but i don't see how that could be related? or maybe?
17:07:41 Join advcomp2019- [0] (n=advcomp2@
17:07:47 Quit advcomp2019 (Nick collision from services.)
17:07:55 Nick advcomp2019- is now known as advcomp2019 (n=advcomp2@
17:08:06 Nick webguest07 is now known as reeyal (i=d2317d7d@gateway/web/cgi-irc/
17:08:31 Join chrisjs169 [0] (
17:08:34TiMiD[FD]who knows ...
17:11:19 Quit desowin ("use linux")
17:11:30peturif it said 3,7V then battery was prettly low. I hope it was on charger when you flashed it?
17:11:41*petur reading the log
17:13:13peturreeyal: after pressing reset, if you connect the charger, does it give the charging screen of iriver?
17:15:20 Quit chrisjs169 (Remote closed the connection)
17:15:41*petur taps Bagder with a very long stick
17:16:04peturbuild servers need a kick
17:16:16*Bagder brings out his extra long stick
17:16:22*petur hides
17:16:49 Join chrisjs169 [0] (
17:19:09 Join leftright [0] (i=d9e1c015@gateway/web/cgi-irc/
17:19:20 Join sarixe [0] (
17:19:50leftrightwebguest07: there are one or two recent builds that will make irivers freeze, try another build
17:21:00peturcurrent build is ok
17:21:06leftrightonce Bagder has finished jiggling the servers with his long stick try the most recent build
17:21:08TiMiD[FD]he cannot even power it up ...
17:21:25leftrightthat the problem I had with the dodgy builds
17:23:15leftrightah, its stone dead, the dodgy builds didn't do that to me, just wouldn't get past boot screen
17:23:50TiMiD[FD]had the freeze problem with this morning's build too, but it didn't brick
17:24:24Bagderbuilt fine now
17:24:34leftrightbrick, implies rendering your player non-useable due to firmware issue
17:24:34*Bagder goes to make dinner for family
17:24:50amiconnTiMiD[FD]: Very strange: I now compiled a clock with aspect correct for archos on my laptop. It works...
17:24:52leftrightleave your long stick behind
17:25:10TiMiD[FD]amiconn: ah with my change ?
17:25:22 Quit chrisjs169 (Remote closed the connection)
17:25:22 Part leftright
17:25:26amiconnA similar one with higher precision
17:25:36TiMiD[FD]then feel free to commit it
17:25:41amiconn(doing the * 5 / 4 before the right shift)
17:26:21TiMiD[FD]by the way make the hour hand longer too in your commit
17:26:27amiconnBut the second hand is broken, due to the use of xlcd_filltriangle
17:26:30TiMiD[FD]2*ANALOG_MINUTE_RADIUS(screen, round)/3
17:26:36 Quit My_Sic (Read error: 104 (Connection reset by peer))
17:26:43peturreeyal are you still here?
17:26:46TiMiD[FD]it's too short on archos
17:26:52TiMiD[FD]hmm yes
17:27:04TiMiD[FD]maybe adding some lines draw ?
17:27:07 Join chrisjs169 [0] (
17:27:09TiMiD[FD]as you suggested
17:28:20TiMiD[FD]now that the code is concentrated in the same place, it's really easier to change :)
17:28:50amiconnBtw, all functions (at least in clock_draw_analog.c) violate the rockbox coding style
17:29:01TiMiD[FD]ah !?
17:29:09TiMiD[FD]you mean braces ? I pointed out on the dev list
17:29:20amiconnThe rules say that the opening brace of a function *must always* be placed on its own line
17:29:27amiconn...not behind the header
17:29:42TiMiD[FD]ah ? I read somewhere that braces placement was free
17:30:27TiMiD[FD]if it's that then all the plugin has to be reindented
17:30:56amiconnYes, general brace placement is how the inital contributor selects it, but should be consistent throughout the file
17:31:16amiconnHowever, top-level braces of functions are special
17:31:30TiMiD[FD]didn't knew that one ...
17:33:49 Quit darkraven (Read error: 113 (No route to host))
17:34:03 Join dan_a [0] (n=dan_a@
17:34:45 Join darkraven90 [0] (
17:35:41 Join toffe82 [0] (
17:36:43*amiconn tries a sim build; hopefully the check will make sure the clock is perfectly circular on target *and* sim
17:37:07amiconn(checking for the LCD type instead of resolution)
17:39:03TiMiD[FD]yes way better to check the lcd type :)
17:42:51 Nick datasleep is now known as datachild (
17:43:39 Join My_Sic [0] (
17:45:35TiMiD[FD]amiconn: commited ?
17:45:58amiconnI'm just testing the added lines on a larger display
17:46:01amiconn(H300 sim)
17:46:58TiMiD[FD]shouldn't matter I guess
17:47:13TiMiD[FD]if you test with the lcd type ...
17:47:32 Quit My_Sic (Client Quit)
17:50:25TiMiD[FD]yuck :]
17:51:34reeyalsorry guys, i'm still here
17:52:04reeyaldo you know whether theres a firmware reset combination or anything that i could try pressing?
17:53:00TiMiD[FD]I don't think there is any one
17:53:10TiMiD[FD]besides your problem seems more hardware related
17:53:47TiMiD[FD]there is probably no point in trying to reset the fw if the cpu doesn't even gets juice ...
17:54:11TiMiD[FD]try to open it and test the battery with a multimeter, that's the simplest thing you could do
17:55:23peturwhat happens if you press reset while the charger is connected?
17:56:02 Quit thegeek (Read error: 104 (Connection reset by peer))
17:56:20reeyalnothing happens
17:56:33peturblack display?
17:57:34reeyalno, nothing
17:57:42reeyalsame as when it's off
17:58:01peturthat's what I meant...
17:58:22***Saving seen data "./dancer.seen"
17:58:48peturthe problem is (I think) that it will not boot if battery is too low, regardless of charger.
17:59:05reeyalso how can i charge it?
17:59:35TiMiD[FD]I guess the charging is enabled by the software ?
18:00:06peturI'm not sure
18:00:30peturI think it's a pcf thing
18:00:35TiMiD[FD]I somehow remeber some threads relating same problems at the time rb was young on the h300
18:00:46TiMiD[FD]2 years ago
18:01:00TiMiD[FD]I'mnot sure though
18:01:08peturLinusN is the expert we need...
18:02:07reeyalhe's not around?
18:02:23FebsJust plug it in and leave it for several hours.
18:02:28peturnot today, no
18:02:38peturFebs: he did
18:03:08TiMiD[FD]reeyal: try to search on the misticriver and rb's forums
18:03:17TiMiD[FD]I'm sure there were same issues before
18:03:35peturI don't know what happens if the battery falls too deep, maybe the pcf no longer works or something like that
18:04:01petur(insert 'voltage' between battery and falls)
18:04:39FebsIt will still charge.
18:05:08FebsBut it will not display the "charging" screen.
18:05:21peturah ok
18:05:39peturso his battery is dead then...
18:07:30reeyali'm not sure the battery is dead
18:07:50reeyali left it on the charger for 5 hours, although it didn't appear to be charging
18:08:03 Quit darkraven90 (Read error: 104 (Connection reset by peer))
18:08:08FebsUSB charger or mains?
18:08:24TiMiD[FD]In the SAVE state, "Only a minimum of modules are activated. This explains why the player will not turn on even when connected to a power source when the battery is very low
18:08:34 Join My_Sic [0] (
18:09:38FebsAre you certain that the charger was plugged in and the outlet it is plugged into turn on? (believe it or not, I've seen several instances before where this was the problem).
18:10:58reeyali tried a couple of different power points
18:11:32 Join bdgraue [0] (
18:13:24 Join midgey [0] (
18:18:44 Quit BigMac (Read error: 110 (Connection timed out))
18:20:19TiMiD[FD]I'm off ot bed, have a plane to take tomorrow morning
18:21:28 Join desowin [0] (n=desowin@
18:23:54 Quit TiMiD[FD] ("leaving")
18:25:58 Join mcphail [0] (n=mcphail@unaffiliated/mcphail)
18:27:19 Quit desowin ("use linux")
18:27:49mcphailCan anyone tell me what tools has been used to make the pdf images in the manual from the inkscape svgs?
18:29:46 Join thegeek [0] (
18:32:31 Quit mcphail ("leaving")
18:32:32 Quit My_Sic (Read error: 104 (Connection reset by peer))
18:34:49 Join desowin [0] (n=desowin@
18:35:19 Join My_Sic [0] (
18:37:53*Domonoky updated his wavsplit plugin (FS #7585) :-)
18:38:49 Join robin0800 [0] (
18:42:33 Join [1]robin0800 [0] (
18:45:32 Quit Seed ("cu, Andre")
18:45:34 Join bluebrother [0] (i=iad15mm2@rockbox/staff/bluebrother)
18:45:48 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn)
18:45:53bluebrotherDomonoky: you forgot to commit the uninstallation files ...
18:46:43bluebrotherbut fyi, I'm progressing with themes installation :)
18:46:46sarixeis there a way to restrict the tag database to one folder?
18:47:13bluebrotherand another idea: why save the number of installs in the log file? Why not use the version string here?
18:47:16peturhmmm something screws up in the settings title when selecting another language....
18:47:17bluebrothersarixe: no
18:47:26sarixek, thanks
18:47:41bluebrotherthat way we could also add a small info screen that simply reads the log file and can then thell the user which file is installed in what version
18:48:00bluebrotherand find outdated files by revision number
18:48:16Domonokybluebrother: version would be good.. it just the number because i didnt know something better :-)
18:48:20 Join stripwax [0] (
18:48:24 Quit pixelma ("CGI:IRC (Ping timeout)")
18:48:34bluebrotherhehe ... I never understood what's the purpose of that number ;-)
18:48:56 Join Seed [0] (
18:50:50bluebrotherah, and I figured out that Qt has it's own md5 implementation.
18:50:52Domonokyi commited the uninstall files.. i hope i didnt forgot more :-)
18:51:08bluebrotherI thought about using md5sums of the urls for caching data
18:51:23sarixewhere's the source for the database, in the source tree?
18:51:24bluebrotherto remove the need of duplicating some folder structure
18:51:36bluebrothersarixe: the database is called tagcache in the source
18:51:46 Quit DataGhost ("NTOSKRNL.EXE caused a buffer overflow in System Idle Proce")
18:52:12sarixeO.o it has a text picture of the API... how useful
18:52:17 Join pixelma [0] (i=57b9f434@rockbox/staff/pixelma)
18:52:29bluebrotherDomonoky: uninstallfrm.ui is still missing :(
18:52:54bluebrotherQString("missing: %1").args(files) ;-)
18:53:21 Quit roolku ()
18:53:39*Domonoky commited again..
18:54:00pixelmapetur: the title in the settings screen being wrong when changing the language is not new - seen that before the langV2 rework already
18:55:18Nico_PDomonoky, bluebrother: isn't there a way to make the building a bit less verbose ? maybe by displaying only the name of the file being compiled instead of the whole command ?
18:55:31 Join DataGhost [0] (
18:55:45bluebrotherNico_P: I thought about that too, but I haven't found a good way
18:56:05 Join spiorf [0] (
18:56:12bluebrotheras the Makefile is created by qmake −− need to read up on the qmake manual a bit further
18:56:18Nico_Pit's a bit hard to kniw what's going on
18:56:21bluebrotherbut maybe obo knows
18:56:30sarixeNico_P : i think that's got to do with make... but then i've seen builds that do what you're talking about
18:56:43Slasherisarixe: basically just add the folder to the build_tagcache() function call, then it will be restricted on that
18:56:44bluebrother(or someone else who is more familiar with Qt ;-)
18:57:09 Quit robin_0800 (Read error: 110 (Connection timed out))
18:57:09 Nick [1]robin0800 is now known as robin_0800 (
18:57:15sarixeSlasheri : uh... where would i put that? (i have not experience with the api here)
18:57:40Slasherisarixe: directly in tagcache.c, just search every call of tagcache_build and modify those
18:57:59sarixebuild_tagcache, you mean?
18:58:06Slasherinope, just as parameter
18:58:20Slasheriyes, not the function itself but the function parameter
18:58:28sarixeyeah, but you said tagcache_build
18:58:38Slasheriups.. :)
18:58:41sarixei see what you mean
18:59:18sarixethere's only four calls in the whole file
18:59:23Slasheriand feel free to implement a better way to do that :)
18:59:38Slasherithat sounds enough
18:59:46Nico_Pbluebrother: is there a kdevelop project file for rbutilqt ?
19:00:29sarixeyep... i'm thinking of adding a menu option to let the user choose what folders are scanned, and probably have a text file to load/save the list
19:00:48 Quit robin0800 (Read error: 110 (Connection timed out))
19:01:22 Join hannesd_ [0] (
19:01:44sarixesay i have all my music in /MUSIC... would i put build_tagcache("/MUSIC") or build_tagcache ("/MUSIC/")?
19:02:40sarixemeh, i'll try both
19:03:46 Quit amiconn (Read error: 110 (Connection timed out))
19:03:46 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
19:06:30bluebrotherNico_P: I have one locally.
19:07:02Nico_Pbluebrother: I remember it was in the svn tree... I found it handy but maybe it makes sense not to include it ?
19:07:26bluebrotherit's in svn? That's a mistake.
19:07:43Nico_Pit used to be
19:08:06sarixeSlasheri : seems to be working... i'll see if it only took stuff from /MUSIC
19:08:09bluebrotherah. I remember my initial commit added a few files that weren't intended to be in svn
19:08:24Nico_Pbluebrother: why shouldn't it be in ? it can be a good idea to use the same for all devs
19:08:55Nico_Pbut I don't really know what's in it so I may be completely wrong...
19:09:09bluebrotherwell, I never bothered much about it −− in fact I rebuilt the project file after a while by simply importing the files again
19:09:48bluebrotherIMO we should avoid IDE-specific files in svn unless it really makes sense (like the Code::Blocks project file for rbutil)
19:10:38 Join fm2 [0] (i=543abbc0@gateway/web/cgi-irc/
19:11:06fm2GodEater_: ping
19:12:33sarixeSlasheri : looks like it worked
19:13:28 Quit Febs (Read error: 110 (Connection timed out))
19:16:16fm2To irc experts: how can I tell a real person from a ghost on IRC? I.e. a person whose nich is listed but who is actually not here?
19:17:04 Quit hannesd (Read error: 110 (Connection timed out))
19:17:05 Nick hannesd_ is now known as hannesd (
19:22:14sarixefm2: /whois, and check the idle time
19:22:33sarixeeh, nevermind that
19:22:51sarixei was wrong
19:27:06bluebrotherfm2: if the nick is registered you can get some infos by nickserv −− /msg nickserv info
19:29:47 Join jhulst [0] (
19:30:30 Join Naked [0] (
19:30:45 Nick Naked is now known as Hadaka (
19:31:06fm2bluebrother: thanks. I'm particularly interested to know if GodEater_ is really here or is just listed. I assume the latter since he doesn't respond
19:34:27Nico_Pwho's Dave Bryant ?
19:35:09 Join JonStamps [0] (i=4b001dcd@gateway/web/cgi-irc/
19:35:12SikuWavPack developer
19:35:30*amiconn finally had success in wrestling with linux wlan
19:35:45amiconnHad to compile the driver myself from latest cvs, grrr
19:35:50Nico_PSiku: does he hang arounf here ?
19:36:39SikuI've no idea
19:36:59JonStampsNeed wiki access to contribute, please
19:37:30peturJonStamps: wiki name please
19:37:31 Quit amiconn (" HydraIRC -> <- s0 d4Mn l33t |t'z 5c4rY!")
19:37:48 Quit x1jmp (Remote closed the connection)
19:39:07JonStampspetur: thx!
19:40:36 Join n1s [0] (
19:41:29pixelmabluebrother: do you remember the steps to turn an svg into pdf/png for the manuals? I made one for the 1/2gen Ipods (but without the text yet...
19:42:17pixelmabut don't have time now, will be back later
19:44:37 Part JonStamps
19:45:01 Quit stripwax ("Miranda IM! Smaller, Faster, Easier.")
19:45:33 Quit pixelma ("CGI:IRC 0.5.7 (2005/06/19)")
19:47:30 Join advcomp2019- [0] (n=advcomp2@
19:57:13 Join stripwax [0] (
19:57:45LloreanNico_P: afaik, Dave Bryant doesn't hang around here, but he does drop by from time to time
19:58:23***Saving seen data "./dancer.seen"
20:00:36peturIf any display/lang expert has time to look at FS #7564, I've written down my debugging results so far... back in a few hours...
20:00:39 Quit petur ("real life calling....")
20:02:02Nico_PLlorean: ok. actually I thought he might be able to give me an opinion on a change I made in prevision of MoB integration
20:02:26Nico_P(it's in roughly the same area as his commit)
20:03:08 Join lazka [0] (
20:03:29LloreanUnfortunately, I don't know how to get in contact with him other than if he's posted an email or something at the wavpack site
20:03:34 Quit advcomp2019 (Read error: 110 (Connection timed out))
20:04:58 Quit midgey ()
20:13:57 Quit stripwax ("Miranda IM! Smaller, Faster, Easier.")
20:18:10 Join BigMac [0] (
20:19:49Nico_PLlorean: if he only works on wavpack he's probably not the best person to review my change
20:24:08 Quit jhulst (Read error: 104 (Connection reset by peer))
20:36:19 Quit Jon-Kha (Read error: 131 (Connection reset by peer))
20:36:30 Join Jon-Kha [0] (
20:39:36 Quit Jon-Kha (Read error: 131 (Connection reset by peer))
20:41:35 Join Jon-Kha [0] (
20:41:59 Join Wofl [0] (
20:44:59 Join x1jmp [0] (n=x1jmp@
20:45:24*n1s dances the build-every-target-and-sim-dance....
20:48:16 Quit ctaylorr (Read error: 110 (Connection timed out))
20:54:46*Nico_P likes FS #7588
20:57:13 Join astronouth7303 [0] (
20:57:44astronouth7303on a 4th gen B&W ipod, the boot sector and its backup should match, right?
21:01:53DerPapsthow did youmake the backup?
21:02:09 Join stripwax [0] (
21:02:15astronouth7303the boot sector backup is part of FAT
21:02:45astronouth7303the rockbox bootloader was installed according to the manual
21:02:47DerPapstdo you mean by bootsector the MBR?
21:03:04DerPapstand how did you make it?
21:03:17astronouth7303you don't explicitly make the backup
21:03:30astronouth7303its part of the filesystem control structure
21:04:36DerPapsti don't understand what you're talking about.
21:04:38astronouth7303if something changes the MBR, it should also change the backup
21:04:57DerPapstwhat backup? where is the backup stored?
21:05:24bluebrotherthe mbr is _not_ part of the FAT filesystem
21:05:32astronouth7303the backup is
21:05:56DerPapstdo you mix up the MBR with the fat superblock or something?
21:06:09astronouth7303on FAT32, the MBR is backed up at sector 6
21:06:24bluebrotherno. Definitely not.
21:06:34astronouth7303at least according to WP
21:06:42bluebrotherthere is a backup of the first FAT sector. That is not the MBR.
21:06:58DerPapstthat is the fat superblock
21:07:25astronouth7303"For FAT32 file systems, the reserved sectors include a Backup Boot Sector at Sector 6." <>
21:07:34bluebrotherthe FAT itself has some vital data in its first sector. This does _not_ match sector 0 of the disc, which is the MBR.
21:07:53*bluebrother pulls out the fat specs
21:08:02astronouth7303fsck says "There are differences between boot sector and its backup."
21:08:27bluebrotherwell, just read 1. at the link you provided:
21:09:07bluebrotherthe first reserved sector contains the BPB which is not part of the MBR but the FAT.
21:09:23bluebrotherlater that paragraph says that there is a backup boot sector.
21:09:35astronouth7303so it's not the MBR
21:09:40bluebrotheryou need to distinguish between the boot sector of the disc and the boot sector of a partition.
21:09:48astronouth7303is that the "FAT superblock" you keep refering to?
21:10:25bluebrotherFAT specs don't call it superblock, but quite some people call it that way −− there is no similar name for FAT
21:10:30astronouth7303does the installer back up Apples FAT boot sector?
21:10:46bluebrotherwhich installer? ipodpatcher?
21:11:05bluebrotherno, as the fat isn't altered.
21:11:22bluebrotheri.e. the FAT of the data partition. It gets only data written to the partition
21:11:35bluebrotheripodpatcher changed the hidden firmware partition.
21:11:45 Nick idnar_ is now known as idnar (i=mithrand@unaffiliated/idnar)
21:11:47astronouth7303that's right, the iPod has a firmware partition
21:11:53astronouth7303or psuedo-partition
21:12:10bluebrotherwell, it's a partition. It's just marked as empty and doesn't contain a file system
21:13:03DerPapsthowever it can convert your ipod from a hfs+ formated one to a fat32 one and it can format the fat32 partition if you ask it to.
21:13:21astronouth7303it's currently VFAT
21:13:24bluebrotheroh, has the conversion added to ipodpatcher?
21:13:27*DerPapst is talking about ipodpatcher
21:13:33astronouth7303and rockbox already been installed
21:14:15astronouth7303and I'm having issues with it, eg it locking up if I try to transfer files on linux, rockbox failing to boot
21:14:37bluebrotherdisc mode is handled by apples firmware.
21:14:53bluebrotherif you have problems with that you might experience hardware problems
21:15:26DerPapsthow did you format the fat32 partition?
21:15:30bluebrotheror the filesystem could be corrupt −− running chkdsk /f might help
21:15:47bluebrother(and according to my experience fsck on windows finds more issues than fsck.vfat)
21:15:55astronouth7303already ran fsck, fs is partially corrupted
21:15:58astronouth7303hm, no linux here
21:16:02astronouth7303no windows
21:16:09DerPapsthow did you format the fat32 partition?
21:16:28 Join tedr0ck [0] (
21:16:29 Join EbErT [0] (
21:21:06 Quit Xerion (Read error: 104 (Connection reset by peer))
21:21:13 Join Xerion [0] (
21:24:20 Quit desowin (Remote closed the connection)
21:24:25 Quit Hadaka (
21:26:14 Join astronouth7304 [0] (
21:26:23astronouth7304sry, internet died on me
21:27:11astronouth7304the FAT was formatted using Apple's firmware updater
21:28:59 Join advcomp2019 [0] (n=advcomp2@
21:29:40astronouth7304basically, the FAT file system should be clean, right?
21:29:50bluebrotherif you just installed it, yes.
21:31:02DerPapstotherwise try to format it with ipodpatcher. but you have to compile it yourself. afaik the version that can format is not yet available for download
21:31:28 Quit tedrock (Read error: 110 (Connection timed out))
21:32:41Aveis there something that could be or should be done, in order to iron out the current ipod nano breakage?
21:32:53 Quit astronouth7303 (Read error: 110 (Connection timed out))
21:33:30LloreanAve: Yes, someone with a failing iPod Nano needs to figure out exactly what is the cause of the problem
21:34:37Aveits sporadic
21:34:53Avesometimes it works, and then suddendly starts to skip and garble
21:35:28Aveof course I could poke in the code randomly in the clock setup area
21:35:35Wofldid you try increasing the anti-skip buffer?
21:35:37 Quit fm2 ("CGI:IRC")
21:35:50AveI'm not using it, or it is at minimum
21:36:01Avecurrently I'm running r14004
21:36:23LloreanWofl: This is a specific hardware problem.
21:36:29LloreanAve: It's not the clock setup that's the problem.
21:36:33Wofli see
21:36:34LloreanThe clock setup has brought to light another problem.
21:36:36Woflmy bad
21:36:38Aveyou can see the diff here:
21:36:42Avestart figuring out..
21:37:49Avebtw how does one tell which version of "apple software" one has?
21:37:53LloreanAve: Start figuring *what* out?
21:37:56Aveor bootloader
21:38:22Aveso is ht the different style of setting up cpu speed that breaks IO?
21:38:29Aveir the higher than before cpu speed
21:38:39LloreanSeems to be a bit of both
21:38:49LloreanBut the different style is actually more correct as far as we can tell
21:38:54Aveso they say
21:39:13 Join Jeton [0] (i=551e5adf@gateway/web/cgi-irc/
21:41:07LloreanAve: If you don't believe us, feel free to reverse engineer the apple code and verify the findings.
21:41:25LloreanBut in all honesty, if you're not going to accept what we say the problem is, you should do some research yourself.
21:41:46Aveno all I'm saying is that if they say so it must be so
21:42:02Avebut it is unfortunate that this triggers some bug elsewhere that cannot be pinpointed
21:42:13 Quit stripwax ("Miranda IM! Smaller, Faster, Easier.")
21:46:46 Quit advcomp2019- (Read error: 110 (Connection timed out))
21:47:03LloreanWell, it seems to be some hardware difference on certain Nanos
21:47:09LloreanMy Nano, for example, works 100% perfectly
21:47:27 Quit davina (Remote closed the connection)
21:47:40 Quit Lear ("ChatZilla [Firefox]")
21:49:04astronouth7304ok, on windows, `chkdsk /R` is failing to complete (device timeouts). Would this be an indicator of replacing the harddisk?
21:49:28bluebrotherit might be, it could also be trouble with your usb and / or cable
21:49:46astronouth7304I'm reasonably sure that it's not the port
21:50:02bluebrotherdo you have a usb hub between pc and Ipod?
21:50:29astronouth7304and it only happens when doing a full sector validation ("/R")
21:50:36 Join tucoz [0] (
21:50:47bluebrotherhmm. Then it sounds pretty much like a hardware issue :(
21:50:56bluebrotherDomonoky: theme uninstallation might get problematic with themes installing fonts
21:51:33tucozpixelma: if you happen to read the logs. to convert a svg to pdf and png, just follow the instructions in the wiki:
21:52:12 Quit Jeton ("CGI:IRC")
21:52:47tucozbluebrother, the qt util is shapeing up. nice initiative i must say.
21:53:05bluebrotherthanks. I'm starting to really like Qt ;-)
21:53:22tucozlooks like domonoky does as well ;)
21:54:08bluebrotheryeah. And I'm learning c++ ...
21:54:25bluebrotherneed to work on the class interfaces thingy more.
21:54:40tucozare there big differences code wise between the different platforms (win,linux,osx)?
21:54:51bluebrothernot at all.
21:54:52AveLlorean: oh hm which version do you have? you havent posted your own specs in the bug page
21:54:55tucozi mean, not with qt, but otherwise
21:55:21bluebrotherah −− well, as soon as you start doing GUIs everything is different :(
21:55:25tucozaha. how cool. so it compiles without ifdef-ing on all platforms?
21:56:09 Quit Wofl (Remote closed the connection)
21:56:28bluebrothertucoz: $ grep '#if' *.cpp | wc -l
21:56:46 Join Buschel [0] (
21:57:44LloreanAve: True, but one of the non-working iPods matches my specs exactly as far as the numbers we've looked at so far
21:58:00LloreanI have the old flash firmware version, but the 1.3.1 apple firmware and flash bootloader
21:58:10astronouth7304thanks for your help
21:58:12Averight, hold for me aswell
21:58:12 Part astronouth7304
21:58:27***Saving seen data "./dancer.seen"
21:58:41AveI dont really know what version of I have..
21:58:58Aveof any mentioned
21:59:33LloreanYou can boot the apple firmware and check in the settings I believe
22:00:24Avelets see
22:00:44n1sweee, rev 14300
22:00:59 Quit Bagder ("*plopp*")
22:01:24Avehm it says "version 1.0" MA099FB
22:01:25*bluebrother wonders if he should attach a screenshot of rbutilqt to the wiki
22:01:37Aveperhaps I should get itunes and upgrade it?
22:01:51Aveor NOT, because this model could help solve the bug
22:01:56Domonokybluebrother would be good.. not long and we can replace the wx version..
22:02:16Buschelhi folks, is it known that switching to other languages than english (tested with dansk und deutsch) does not seem to work anymore? or am i doing some kind of mistake? svn #14299
22:02:26bluebrotherok ... I'm currently searching about the mingwm.dll dependency my static binary has
22:03:08n1sBuschel: are you using an official build or have you built yourself?
22:03:25tucozDomonoky, how would you compare wx and qt in terms of consistency and working with in general?
22:03:34 Join Bagder [0] (n=daniel@rockbox/developer/bagder)
22:03:40Buscheln1s: built by myself
22:03:59n1sBuschel: have you recobfigured in the last week or so?
22:04:48Bagdern1s: lovely sim-commit!
22:05:18n1sBagder: thanks, I was a little surprised by the lack of red and yellow :-)
22:05:21Domonokytucoz: qt seems to be better..there are more ready made widgets and classes in qt.. the nice signal/slot mechanisms, and better dokumentantion... but wx also isnt bad..
22:05:22Buscheln1s: no, I didn't. do I have to?
22:05:41 Join Shishire [0] (
22:05:48LloreanBuschel: If you aren't using downloaded builds, you should *always* confirm with a completely new build folder or at least a clean and reconfigure.
22:06:02n1sBuschel: yes, the build system changed a bit with the lang cleanup so a rerun of configure is necessary
22:06:14tucozDomonoky, ok. interesting. my impression is also that qt seems to be more complete so-to-speak.
22:06:29BuschelLlorean: performed "make clean" n1s: okay, gotta try
22:06:44LloreanBuschel: Notice the "clean and reconfigure" line.
22:08:11ShishireI'm having a problem while trying to "make zip" yesterday's build on cygwin. I'm compiling a normal for ipod video, and it gives me two errors, one that there is no directory to copy wps from, and that it doesn't know what xargs:zip: is
22:09:34BagderShishire: you have zip installed?
22:09:43Shishirezip what?
22:09:49Bagderzip the tool
22:09:57Shishirelet me check
22:10:08Bagdertype 'zip' and press return
22:10:15Buschelso, just deleted and re-created/re-configured my build-directory. coming back with (hopefully good:) news in some minutes
22:10:17Shishirenot found
22:11:41BagderShishire: regarding the wps files, they are in the wps/ subdir and I don't know why you'd get a warning about that
22:11:54 Join jhulst [0] (
22:13:18Shishirethat should do it
22:13:32Shishireas for the wps, I can add those manually if need be.
22:19:58 Join Wofl [0] (
22:20:43 Part tucoz ("Leaving")
22:21:42 Quit My_Sic ("Miranda IM! Smaller, Faster, Easier.")
22:21:52 Join My_Sic [0] (
22:22:25 Quit My_Sic (Client Quit)
22:29:22bluebrotherDomonoky: added some screenshots to the wiki page. No luna as I did them on w2k ;-)
22:31:11 Join Gursikh [0] (n=Gursikh@tremulous/officialdevannoyer/khalsa)
22:34:26Domonokybluebrother: it much nicer with luna :-)
22:34:52bluebrotherwell, I usually turn of luna. But feel free to create luna-based ones ;-)
22:35:08 Quit Shishire ()
22:35:59 Join Naked [0] (
22:36:09 Nick Naked is now known as Hadaka (
22:36:28*Domonoky is thinking about the console support for rbutil.. the problem is windows: if i add console support to rbutil a console window pops up when you start rbutil.. :-/
22:37:14Domonokyon windows, as a normal gui app, i cant access the console .. :-/
22:38:14bluebrothercan't you read argv[]?
22:38:38Domonokythats not the problem, the problem is output.. :-)
22:38:50 Join barrywardell [0] (
22:38:56bluebrotherah, so completely console-controlled?
22:39:46Domonokyi thought if you give parameters to install something, you want feedback in the console, and not in a window..
22:40:05 Join inversion [0] (n=none@
22:40:33 Nick DiDjCodt is now known as h (
22:40:41 Nick h is now known as DiDjCodt (
22:41:19Domonokyso you write something like: rbutilqt: -m "e:/" -t h120 -install fonts :-)
22:42:31 Part Gursikh ("Leaving")
22:42:44Domonokybut on windows i can only output to the console if i switch on console support from qt, and the a console window popsup if you start it from the gui..
22:43:03 Join Tempis [0] (
22:43:23bluebrothermaybe make it a different build?
22:43:25Tempisanyone here able to answer an Archos question?
22:43:36Domonokyso we would need different binarys for windows.. yes. thats a way.
22:43:45bluebrotherTempis: just ask. If someone can you you'll get an answer
22:43:53Tempisok, thanks
22:44:05bluebrotherin that case it could be completely gui-less.
22:44:25bluebrotherperhaps some curses-like interface ;)
22:44:44TempisI have an Archos Recorder that I put a 60GB drive in... The unit cut out and the batteries started heating up
22:44:59TempisI don't know what the problem might be
22:45:00 Quit Wofl (Remote closed the connection)
22:46:02Tempisand when I say heating up, I don't mean they just got a bit warm... they were getting very hot
22:46:07Tempisalmost too hot to touch
22:46:24Tempisand I'm sure if I had waited they would have overloded and perhaps exploded
22:48:23DomonokyTempis: sounds like you made a shortcut somewhere when swapping the harddrive
22:49:10Tempisit's been woring but just all of a sudden it did that
22:49:14Tempisand now it won't turn on
22:49:14Domonokysomewhere is + and - connected
22:49:33Domonokycheck the connectors..
22:49:40TempisI'll do that
22:50:17 Part Domonoky
22:52:53 Join amiconn [0] (
22:57:07 Quit BigMac (Read error: 110 (Connection timed out))
22:58:24 Join BigMac [0] (
23:00:47 Join Nibbier [0] (
23:04:01 Join petur [0] (n=petur@rockbox/developer/petur)
23:07:04n1samiconn: Could firmware/drivers/power.c be moved to firmware/target/sh/archos/power-archos.c ? it is just used by the sh-archos targets...
23:09:19amiconnHmm. It should be split per target properly
23:09:26amiconnLots of ifdefs in there :(
23:09:46Buscheln1s: after reconfiguration and rebuild the languages work again :o) thanks for your help
23:09:54n1samiconn: ah, I'll leave it for someone with an archos target then...
23:10:00n1sBuschel: ah, good :-)
23:10:35amiconnOh, it also seems to contain the simulator stubs
23:12:25*rasher wrestles with building multiple voicefiles in one go
23:12:45rasherI'd need (at least) to have different TTS-config per language
23:13:32rasherAnd right now I don't even have any tts-config at all.
23:13:35n1samiconn: I don't know if these are needed anymore though, will check
23:13:40rasherThis should be interesting.
23:15:33 Join Wofl [0] (
23:15:45 Quit Wofl (Remote closed the connection)
23:16:44*amiconn is still thinking about the best solution for distributed shutdown handling (similar to how usb handling works; each thread processes the broadcast message and signals if it is ready for shutdown)
23:17:19peturthat would certainly get my vote
23:17:22 Quit Buschel ()
23:17:38 Join davina [0] (
23:17:39amiconnThe problem is that if the power management thread waits for acknowledgement from every thread, lots of threads would need to be sprinkled with processing of additional messages
23:17:54amiconn(all that have a public message queue)
23:18:43 Join My_Sic [0] (
23:18:49amiconnAnd another problem is that the ata thread needs to know when the other threads are done with saving status information
23:18:57amiconn(btw also a problem with usb)
23:19:27amiconnpetur: I *need* it for proper suspend...
23:20:08peturI'm surprised we don't have it yet....
23:21:45amiconnThe drawback is that many threads have a public message queue (16 in total, of course not all of them present on all targets. Some are mutually exclusive)
23:21:48 Join Wofl [0] (
23:22:00 Quit Wofl (Remote closed the connection)
23:23:19amiconnOf course I could make the "shutdown master thread" only expect a few acknowledges, but that's error prone when threads get added
23:24:13amiconnAnd it still doesn't solve the problem that the ata/ata_mmc/ata-e200 thread needs to know when the other threads are ready
23:24:43peturmaybe the threads should subscribe to those events and only the subscribed ones must signal they are ready?
23:25:09amiconnYeah, but that only moves the complexity from the threads to the registration
23:25:37amiconnRight now we have queue_broadcast(), and e.g. the USB thread counts the acknowledges it gets
23:25:51 Join Wofl [0] (
23:25:57amiconnIt expects (number_of_public_queues -1) acknowledges
23:26:08amiconn(the -1 is for its own queue)
23:26:51rasherAh, festival configuration might be the first LISP code added to Rockbox svn.
23:27:05amiconnThe current system is simple, every thread gets every system event, and the only ones that needs acknowledge are the USB events
23:27:15peturmaybe when it is satisfied it could broadcast another one saying it is done? Or work with a central thread to monitor the responses and use callbacks for notifications
23:27:31rasherOr not, there's an emacs config somewhere
23:28:25amiconnHmm, perhaps a 'light' registration system would be helpful?
23:29:58amiconnSimple system messages would still be broadcast to all threads, but for messages that need acknowledgement, the thread would register on queue creation if it needs to process them
23:30:32amiconnqueue_init would just maintain a count of how many acknowledges to expect for that event
23:30:57amiconnHmm, but that may be difficult to maintain since queues can be deleted....
23:31:25amiconn...unless we *require* each thread to properly unregister all events it registered
23:35:11 Join ddalton [0] (
23:35:48ddaltondoes anyone know how the current radio station is displayed in the fm radio? like 101.9 mhz
23:36:35 Quit ompaul ("morning when you join IRC night when you leave where did the day go?")
23:38:52rasherddalton: It uses LANG_FM_STATION which is "Station: %d.%02d MHz" for English.
23:39:14rasherddalton: so, exactly like you said, except with capital M and H
23:39:22rasherddalton: and Station: in front of it
23:40:21rasheramiconn: would you want/expect separate encoder/TTS selection per language, or just TTS options?
23:41:03 Join ctaylorr [0] (
23:41:07n1shmm, build system seems to be stuck
23:42:16Bagderit hates us today
23:42:52amiconnrasher: Hmm, separate encoder settings might be useful for fine-tuning voice file size vs. quality
23:43:17rasherRight now I don't even query for encoder settings..
23:44:12rasherSomething's broken in italiano.lang
23:44:41 Quit hcs ("Leaving.")
23:45:04rasherThere's a BOM and some strings are mangled (such as LANG_FM_NO_FREE_PRESETS)
23:46:21amiconnHmm, now how could a thread register for events when initialising the queue?
23:46:32amiconnIt could be interested in more than one event...
23:46:45amiconn...and they're not single bits in a mask
23:47:42amiconnA variadic function might be an option, using an end-of-list marker...
23:47:52 Quit Wofl (Remote closed the connection)
23:48:02 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
23:48:25 Join Wofl [0] (
23:48:45amiconnNow how to keep it enough KISS :(
23:50:43 Join perrikwp [0] (n=chatzill@
23:52:12 Quit Wofl (Remote closed the connection)
23:52:22ddaltonrasher: What file are you talking about? radio.c in the record dir?
23:52:31ddaltonrecorder i mean
23:54:44rasherddalton: I'm not sure I understand your question. The string that's used when displaying the radio station is LANG_FM_STATION. I imagine it happens in radio.c, yes.
23:54:58bluebrotherddalton: just grep for that string ...
23:55:54 Join Wofl [0] (
23:55:54 Join hcs [0] (n=agashlin@rockbox/contributor/hcs)
23:58:28***Saving seen data "./dancer.seen"

Previous day | Next day