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

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

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

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

#rockbox log for 2009-02-21

00:00:21rasherjhMikeS: didn't want to accuse you of sloppy coding, just seemed like the most likely culprit. Sorry!
00:00:49jhMikeSI don't know. Maybe I botched something. I'm only human (mostly)
00:01:20 Join avis [0] (n=ident@pdpc/supporter/student/avis)
00:01:37*BigBambi silly - thought jhMikeS meant voice whilst paused :)
00:02:02jhMikeSOf course mistakes with memory map changes tend to just crash outright
00:02:53*jhMikeS will be very specific about things if he tackles that one
00:07:34 Quit gevaerts (Nick collision from services.)
00:07:44 Join gevaerts [0] (
00:11:24 Part CaptainKwel
00:11:41 Quit amiconn (Nick collision from services.)
00:11:42 Join amiconn_ [50] (n=jens@rockbox/developer/amiconn)
00:11:46 Join pixelma_ [50] (n=pixelma@rockbox/staff/pixelma)
00:11:46 Quit pixelma (Nick collision from services.)
00:11:59 Nick pixelma_ is now known as pixelma (n=pixelma@rockbox/staff/pixelma)
00:12:00 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
00:21:45 Quit Seed (Nick collision from services.)
00:21:49 Join Seedy [0] (
00:23:42rasherOkay, I can't even place opt{lcd_non-mono}{} *around* an example block, it seems
00:24:41moosred builds fix is comming...
00:24:44pixelmadid you try optv (see my paste above). Pure guesswork though...
00:25:43 Quit {phoenix} (Remote closed the connection)
00:25:56pixelmarasher: ^ and it should also have a questionmark there
00:26:11rasherpixelma: That works.. but it seems there's no \noptv .. so it gets problematic
00:26:38 Join faemir [0] (
00:27:56*jhMikeS kicks the PP memory controller
00:28:15pixelmarasher: IIRC this \nopt is a self-made command for the Rockbox manuals misusing the \opt by inverting the match or something, someone said it's a bit nasty so maybe not a big surprise
00:29:24rasherYeah.. looking at preamble.tex now.
00:29:40*rasher doesn't understand what's so unnatural about having a way to invert a match
00:31:32***Saving seen data "./dancer.seen"
00:32:37 Quit intrados (Success)
00:32:45 Quit isthatall (Read error: 104 (Connection reset by peer))
00:33:45 Join intrados [0] (
00:34:53 Quit jhMikeS (Nick collision from services.)
00:34:59 Join jhMikeS [50] (n=jethead7@rockbox/developer/jhMikeS)
00:38:04 Quit HBK ()
00:50:35 Quit flydutch ("/* empty */")
00:54:54 Join BHSPitMonkey [0] (n=stephen@unaffiliated/bhspitmonkey)
01:01:37 Join tessarakt [0] (
01:22:29 Join goffa [0] (n=goffa@
01:22:33 Quit goffa_ (Read error: 60 (Operation timed out))
01:25:32Unhelpfulhrm, anybody have any other ideas for debugging something that only happens on sim? i've tried serial logf, but my e200 hangs on usb plug with that enabled, i've tried adding splashfs, but the first one never returns, so it's hard to get much data back, and i can not manage to get the sim to fail the same as target. :/
01:27:51 Quit midijunkie (Read error: 110 (Connection timed out))
01:29:59gevaertsUnhelpful: it shouldn't hang. Do you have mass storage enabled?
01:30:23 Join Thundercloud [0] (
01:31:39Unhelpfulgevaerts: i know it shouldn't. i followed the directions from, and commented out #define USB_STORAGE in firmware/export/usb-core.h
01:32:18gevaertsUnhelpful: which revision?
01:34:11Unhelpful20061 is the last i pulled from SVN, but i had the same problem when i tried logf for something back in december, as well.
01:37:43 Part toffe82
01:38:45gevaertsUnhelpful: I can reproduce the hang here
01:38:48 Join `Assassin [0] (
01:40:03 Quit tessarakt ("Client exiting")
01:40:14gevaertsUnhelpful: a (possible) workaround is to edit usb.c and everywhere it calls a bunch of usb_core_enable_driver(), remove them and put in a single usb_core_enable_driver(USB_DRIVER_SERIAL,true);
01:41:58*gevaerts doesn't understand why that's different, but it seems to be
01:42:42gevaertsUnhelpful: if you do that, don't use the "enable logf over USB" debug menu item
01:43:02Unhelpfuldoes it work? :)
01:43:28gevaerts"it" being?
01:44:16Unhelpfulah, you didn't try that workaround, yet, either. i'll test it shortly.
01:44:33gevaertsI did actually
01:44:48gevaertsI just was a bit confused...
01:45:14gevaertshm, maybe I should tell you. Yes, it works here
01:46:20gevaertsThe problem is that usb serial as a whole seems to be a bit unstable these days. Sometimes (or often, depending on how much you send) it just stops transmitting data
01:47:53Unhelpfulshould the debug menu option perhaps be disabling the mass storage driver (if it's been built) when enabling serial?
01:48:14gevaertsNot really. On PP you can do both at the same time
01:48:40 Join gartral [0] (
01:49:38Unhelpfulthe mprotect hack apparently doing nothing on sim would seem to suggest that the allocator itself works fine... perhaps the mess is somewhere else.
01:51:23rasherI wonder how much WPS stuff we want to put in the manual. And if we want to copy or move it there.
01:51:30gartralok, back, as i was saying earlier, my e250 will only play a little bit of an ogg file, skip, freeze up BAD (but doesn't actually crash) play again, repeats this 2-4 times before it gives up on the track and emediatly jumps too the next, and starts the whole proccess all over
01:51:58 Join HBK [0] (
01:52:00rashergartral: And this doesn't happen with r20044?
01:52:41gartralive been testing older builds, all the way back too 20040, all with the same effects, ide go back further, but the server i was building from went down
01:53:34gevaertsgartral: does this happen with all ogg files, or only with some? Is your filesystem clean?
01:54:28gartralthe audio files are all either q6-q8 VBR oggs, made with oggenc 2 on EAC, and ive tested my FS, and havent found any problems, only one album from jamendo didnt skip, everything else did
01:55:40 Quit tyfoo (Read error: 104 (Connection reset by peer))
01:55:52gartraland i dont have any speex files too test, or i would have >.>
01:57:11gartraleverything else works, and i noticed the skipping is worse post 20044
01:57:39rashergartral: does it happen with all your vorbis files?
01:57:56rasherIgnore me, as I can't read.
01:58:31gartrallol, don't worry, happens too me all the time
01:58:42rashergartral: did you reset your settings to the default?
01:58:49Lloreangartral: Did you attempt playing them back in a strict PC-side player? Just because your filesystem isn't corrupt doesn't mean the files aren't.
01:59:31gartraland yes, both on the internal drive, and my MSD, and yes, they all play fine in ffmpeg/winamp/vlc/foobar\
02:00:03gartralsongbird had issues, but thats songbird...
02:00:35Unhelpfuland ffmpeg doesn't display any errors when decoding them?
02:01:08gartralnope, all reported ok, cept for one i deliberatly broke to test
02:01:55Lloreangartral: Please remember our real-words rule.
02:02:33gartralohh, sorry, i use 'cept in the real world, and too me, it is a real word
02:02:57LloreanWe mean "word non-English speakers can look up"
02:03:15LloreanDid you watch the terminal window when playing in ffmpeg for warnings?
02:04:43gartralyes, and as i said, it was clean, except for one i deliberately broke, to make sure ffmpeg wasn't screwed up in one way or another
02:04:49Unhelpfulgevaerts: hrm, it still hangs for me, with the same "unable to enumerate" message on the host :/
02:04:49Unhelpfulgartral: except that it's still very much not an actual word.
02:05:22 Quit bertrik ("Leaving")
02:05:25gevaertsUnhelpful: I have a suspicion. let me check
02:06:55rasherIf someone with LaTeX smarts could have a look at FS #9934, that'd be great
02:07:42Unhelpfulgevaerts: thanks. if i'm being a bother, let me know, i'll go back to other ways of trying to troubleshoot this. also, when are you writing the gdb usb serial support? ;)
02:07:51Lloreangartral: I'd still take a clue from the fact that other players have a problem with the files that it's just a case of some recovering from errors in the files better than others.
02:08:16gevaertsUnhelpful: I've just committed a fix. gdb will have to wait a bit longer :)
02:09:32 Quit awake_ ()
02:11:52 Join tessarakt [0] (
02:12:00gartralbut none of the others have a problem
02:13:45 Quit trisiak (Read error: 110 (Connection timed out))
02:14:12 Quit moos ("Rockbox rules the DAP world")
02:14:49Lloreangartral: I thought Songbird did...
02:15:14LloreanIf two audio playback softwares have a problem with the files, and nobody else is currently reporting vorbis playback issues except you, odds are pretty good it's still a case of something on your end.
02:15:54gartralim sorry, i should have elaborated; songbird doesnt load on this computer, meaning nothing plays in it, and the only play that DID have problems was my DAp
02:16:41gevaertsThen why did you even mention it?
02:17:31 Join BossColo [0] (n=4350effe@gateway/web/cgi-irc/
02:18:10gartralrandom line of thought, im mildly autistic... >.>
02:18:24BossColoHi, I'm hoping that someone may be able to help me.
02:18:37BossColoI'm having some rockbox (i think) related woes
02:18:46gartralBossColo: dont ask too ask, just ask
02:19:05BossColoSo, I decided to salvage my old h340. The hard drive was busted, so I figured I'd just buy a new one. I got the MK4006GAH, so I don't think the trouble's there. However, after charging it up, I turn it on, and I'm greeted with "I00: at 400C2304". The hard drive isn't even spinning up. The background is oscillating from grey to a lighter grey. I have no idea where to go from here. Thank you for any help you may
02:19:59LloreanBossColo: Did you put Rockbox files on the new HD before putting it in?
02:20:29LloreanAnd it does this every time you reset it?
02:20:31gartral[this may be somewhat stupid] is the cable connecting it upside-down?
02:20:44BossColoi have everything in correctly
02:20:55BossColoand i had flashed rockbox before the old hard drive failed
02:21:04 Quit tessarakt ("Client exiting")
02:21:06LloreanYou flashed the Rockbox bootloader.
02:21:13LloreanYou still need Rockbox files on the disk for Rockbox to boot.
02:21:15BossColoso i figured i'd be able to start the usb bootloader and format the hard drive
02:21:38LloreanYou haven't answered my question on whether it does this every time you reset it.
02:21:57LloreanI wasn't sure if you were saying "I'm stuck on this screen" or "I keep getting this screen"
02:22:05LloreanWhat version of the bootloader do you have installed?
02:22:17BossColothat much comes up
02:22:24BossColoit says rockbox bootloader v5
02:22:26LloreanAnd you boot it with the USB cable plugged in?
02:22:30BossColothen the battery voltage
02:22:36LloreanHave you tried booting to the Original Firmware?
02:22:43BossColothat doesn't work at all
02:22:55LloreanIn what way? That phrase isn't very descriptive.
02:23:04BossColoit doesn't make a difference whether or not the usb cable is in, i get the same message
02:23:05gartralit might be either a bad drive, or an unformatted drive :|
02:23:15BossColothe drive doesn't even spin up
02:23:42BossColowhen i try to boot the original firmware, it says "booting original firmware", then just shuts off
02:23:53BossColoi just got something new
02:24:01LloreanOn the H300 you have to hold the button down for a little bit to try booting the OF, even after it says that, I think
02:24:05BossColoATA error -11 insert usb cable and press a button
02:24:26BossColoand now it's connected to my computer
02:24:32BossColobut the drive still isn't spinning up
02:25:00LloreanThen it sounds like hardware issues.
02:25:29BossColoin the device manager, its listed as USB Mass storage devic
02:25:39gartralmight be a bad drive, did you get a warrenty?
02:25:45BossColoi can still rma
02:27:36BossColoand i'm getting a sound from the hard drive, a rythmic buzz.
02:30:04BossColoalright guys, thank you
02:30:07gartralthats a bad drive. or a bad battery, the drive wont spin up properly if it isnt getting enough power, try with a power cable and battery removed...
02:31:09 Quit rakslice_work (Read error: 104 (Connection reset by peer))
02:31:14 Join rakslice_work [0] (
02:31:35***Saving seen data "./dancer.seen"
02:34:25 Quit domonoky (Read error: 104 (Connection reset by peer))
02:34:27 Quit timc (Remote closed the connection)
02:34:54BossColowithout the battery, i'm getting nothing
02:35:16BossColowait, now the screen is white
02:35:25 Join timc [0] (n=aoeu@
02:35:36BossColobut only when the usb cable is in
02:36:08BossColosorry, the white screen is only when both are plugged in
02:37:01gartralhmm, id'e say bad drive and RMA it
02:37:21BossColoand the motherboard right under the controls is getting hot
02:37:32BossColowhere the drive connector is
02:38:42gevaertsBossColo: are you sure your old hard drive was broken, and that it's not a logic board issue?
02:38:51gartralthats a hardware issue, and shouldnt come up in #rockbox, technically, im getting off topic by helping you, but ide rather see your DAP work than avoid a slap on the wrist >.>
02:39:19BossColothe old hard drive had a click
02:39:28Lloreangartral: Intentionally ignoring channel guidelines will get you a ban, not just a slap on the wrist.
02:39:55BossColoi appreciate the help, but you're probably right, this old iriver has seen better days
02:40:14LloreanBut hardware issues are often a gray area.
02:40:20BossColoi don't want to get anyone banned
02:40:24 Quit BossColo ("CGI:IRC")
02:40:33Unhelpfulif you think a question belongs in the other channel, you can say so. that's what generally ought to be done when it's clear a conversation is getting off-topic... hardware issues are often not very clear :/
02:43:17flyinghippoYou could also take it into private messages.
02:46:41 Quit n1s (Read error: 104 (Connection reset by peer))
03:05:19 Join nino [0] (
03:05:43 Nick nino is now known as nino_ (
03:05:53 Nick nino_ is now known as nino (
03:06:19 Nick nino is now known as nino_ (
03:08:52 Quit nino_ (Client Quit)
03:18:27 Quit BHSPitMonkey (Remote closed the connection)
03:27:58 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
03:32:18 Quit Thundercloud (Remote closed the connection)
03:32:39 Join Darksair [0] (n=user@
03:38:24 Quit flyinghippo ("Nettalk6 -")
03:48:06 Quit dfkt ("-= SysReset 2.53=- Ph'nglui mglw'nafh Cthulhu R'lyeh wgah'nagl fhtagn.")
03:52:07 Join Lss [0] (
03:55:16 Quit Zoxc- ()
03:57:23 Quit perrikwp (" ajax IRC Client")
04:10:23 Join Barahir_ [0] (
04:13:45 Part Aurix_Lexico
04:15:53 Quit AndyIL (Read error: 60 (Operation timed out))
04:16:53 Join AndyI [0] (i=AndyI@
04:22:18 Quit Barahir (Read error: 113 (No route to host))
04:31:38***Saving seen data "./dancer.seen"
04:32:41 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
04:40:30 Quit Chesteta (Read error: 110 (Connection timed out))
04:49:30 Quit intrados (Read error: 104 (Connection reset by peer))
04:50:41 Join intrados [0] (
04:51:27 Join blkhawk- [0] (
04:59:00 Quit Lss (Read error: 110 (Connection timed out))
05:00:38 Join Xenovalent [0] (n=xenovale@
05:04:16 Part Xenovalent
05:07:31 Quit blkhawk (Read error: 110 (Connection timed out))
05:08:25 Nick blkhawk- is now known as blkhawk (
05:10:02 Join ucchan [0] (
05:10:49ucchanI made a lot of patch files of the Text viewer plugin.
05:11:37ucchansee FS #8445, FS #9546, FS #9853, FS #9855, FS #9892, FS #9893, FS #9898, FS #9902
05:12:21ucchanBecause I do not correct this plugin any further, could you confirm the operation of it after the patch is applied ?
05:13:39 Quit miepchen^schlaf (Read error: 101 (Network is unreachable))
05:13:47ucchanPlease apply the patch in order of FS #9855, FS #9892, FS #9893, FS #9898, FS #9902, FS #9853, FS #9546, FS #8445, and FS #9387.
05:16:10ucchanShould I prepare the patch file where all the corrections are included ?
05:17:42 Join webguest50 [0] (n=62c6b8b8@gateway/web/cgi-irc/
05:18:00webguest50Hello =o
05:20:13krazykitdo you need help? if so, just ask
05:20:24webguest50Yes actually xD
05:20:31 Join awake_ [0] (
05:20:37webguest50Apparently my Sandisk wants to freeze on me =o
05:20:55webguest50E200R =o
05:21:51webguest50Now it won't even connect to the computer
05:22:57 Join FlynDice [0] (n=jack@
05:23:15webguest50can anyone help?
05:23:58gartralare you turning the DAP off before you plug it in?
05:24:24webguest50Well the main issue is it just freezes when i power it on
05:25:06webguest50it just says SanDisk and does nothing
05:25:09webguest50completely nothing
05:25:17gartralare you sure your useing the E200R bootloader?
05:25:35webguest50I can't even get the bootloader onto it, it won't connect.
05:27:01saratogadid it ever work?
05:27:08gartrali tke it you just gooot this e200?
05:27:23webguest50lol it worked before
05:27:42webguest50yes i have the e200
05:28:00saratogawell saying how it broke is a good start then
05:28:26gartralok, try too turn it on while holding the "left" or "rewind" key
05:28:30webguest50Lol, um then only thing i can say is i press power and it stays on SanDisk
05:29:10gartralthat should load the sandisk firmware, if not, your gonna have too use the recovery mode
05:29:13webguest50nothing so far
05:29:31gartralstaying on "sandisk"?
05:29:34webguest50recovery mode? trying to learn sorry.
05:29:36webguest50Yes it is.
05:30:17webguest50This is worse than the ipod with a theme hack .-.
05:31:22saratogaif you hold it for a long time it'll power off, but if you can't say how it stopped working its hard to suggest a fix
05:32:01webguest50I think i know what i did wrong in a odd way.
05:32:19webguest50I was reading when it said extract .rockbox and add the two files to system and root.
05:32:24webguest50then i disconnected
05:32:47saratogathat won't cause problems, so you probably did something else
05:33:00saratogai guess this happened while you were trying to install? if so, thats a good thing to mention
05:34:27webguest50So i did the two files things then went to manufacture mode and ran the patcher
05:34:38webguest50im a mess up but it takes time to learn i guess
05:36:40webguest50Okay so now im in recovery mode
05:36:58webguest50Now what should i do?
05:37:03saratogadid e200rpatcher say anything interesting when you ran it?
05:37:09 Join sarixe [0] (
05:37:21webguest50No it just said patch complete
05:37:30webguest50then my Mp3 disconeccted
05:37:34gartralok, your in recovery mode?
05:37:37webguest50Yes i am.
05:37:51webguest501 file in there called VERSION
05:38:04gartraldon't touch it
05:38:08webguest50i know
05:38:15webguest50i don't wanna touch anything until im told to D:
05:38:24saratogai guess you can just follow the unbrick guide
05:38:29saratogathen start over when you get it fixed
05:38:40saratogai don't think the R is any different then the normal in that respect
05:38:43gartralthe guide doesn't have the e200R firmwares
05:38:55webguest50is that bad or good
05:39:10 Join Hillshum [0] (n=chatzill@unaffiliated/hillshum)
05:39:11gartralneither, as of this second
05:39:27webguest50I see...
05:39:52webguest50what are those?
05:40:01 Join isthatall [0] (n=guy2@unaffiliated/isthatall)
05:40:18FlynDiceanyone available to get me wiki write access?
05:40:18gartralyou want the sandisk sansa e250r one from that page
05:40:27webguest50so download it =o
05:40:32saratogaFlynDice: sure whats your name
05:40:40FlynDiceJack Halpin
05:40:44gartralyes, but make absolutly sure you have the right one
05:40:54saratogagartral: theres normal and R firmwares on that page
05:41:08webguest50i don't know if i have R or 250 .-.
05:41:47gartralwebguest50: it may say on the back plate in the lower right hand corner
05:42:41webguest50iys 360
05:42:48webguest50no R
05:42:57krazykitgartral, that's hardly accurate, especially with refurbs
05:43:03saratogadon't use enter when you want a comma . . .
05:43:15saratogaFlynDice: added
05:43:45webguest50so which should i get? =o
05:45:19gartralhmm.. first you need too know which one you have, im not sure how too get that information without having access too the firmwares...
05:45:35webguest50it says 250
05:45:37webguest50without a R
05:45:40webguest50on the plate.
05:46:04saratogathat doesn't mean a whole lot though, since you could have either firmware on either device
05:46:05gartralyes, but as krazykit said, thats not very accurate
05:46:28saratogastop doing that
05:46:36webguest50Atm im down for taking 250R as a test.
05:46:40webguest50And sorry D:
05:47:02slyyfANybody here have experience with the Codwin D2 code?
05:47:09 Quit slyyf ("leaving")
05:47:40webguest50Saratoga, should i take my chances with the 250R?
05:48:39saratogai'm actually not sure what will happen if you guess wrong
05:48:51saratogai know its possible to convert one to the other but I don't remember the process
05:49:07webguest50I can always try taking my chances =o
05:49:39gartralanother way would be to open the DAP and look at the logic board
05:50:27gartralDigital Audio Player
05:50:43webguest50and look at the Logic Board?
05:51:24gartralyea, but i dont know what strings too look for...
05:51:42webguest50lets take our chances with 250.
05:53:48saratogagartral: the R and no-R have identical hardware
05:53:57saratogathats why they both run the same software
05:54:30webguest50What do i do with the .mi4?
05:54:45saratogaanyway, it might be a good idea to ask when someone who is familar with the R players is online
05:55:39 Quit saratoga ("CGI:IRC (EOF)")
05:56:10tmztwebguest50: check the usb settings menu
05:56:29webguest50Yeah its frozen on the SanDisk screen
05:56:41tmztah, so you are trying to recover it?
05:57:43tmztsaratoga is right of course, you want someone who knows how
05:58:30webguest50i was saying lets just go ahead and test the R mi4.
05:58:35advcomp2019webguest50, boot into recovery mode and use pastebin and post your version.txt
05:58:40webguest50Ill deal with conseqences.
05:58:50advcomp2019i have an e200R
06:00:36advcomp2019you have a corrupted version.txt file it looks like
06:00:54webguest50D: is that what im gonna sounds like a idiot and say bad?
06:01:26gartralwebguest50: where did you get that dap?
06:01:26advcomp2019looks like you need to use the e200tool in linux
06:01:40webguest50A friend.
06:01:47gartraladvcomp2019: he's a 14 year old kid...
06:02:01webguest50.-. lol
06:02:10webguest50I do alot of computer wise stuff but not this type.
06:02:29webguest50Could i get a link to the e200tool ? ^-^
06:02:35gartralwebguest50: you need too figure out where HE got it... if it came from bestbut, or the rhpsody store, then it is an R, else, its just an e200
06:03:06webguest50then its probably a e200
06:03:43advcomp2019gartral, yea but that is the only way to fix it from my experience
06:04:17webguest50Where can i get the tool?
06:04:18tmztwebguest50: do you remember what the usb settings options were?
06:04:23advcomp2019plus i have seen this more with the e200 series then with the e200r series
06:04:33webguest50the MSC thing?
06:04:45gartralthats it, its an E200
06:04:57webguest50Yay? D:
06:05:26webguest50Im just happy i can get help.
06:05:26gartralif it were playsforsure and something else, its an R its MSC/MTP its a "vanilla"
06:05:50webguest50Wait what?
06:05:54tmztPlaysForSure and Raphsody?
06:06:19webguest50Im lost.
06:06:31gartralit doesnt matter right now, what matters is that we know which one you have, as long as your sure it was MSC/MTP
06:06:32tmztMSC/MTp is non R
06:07:02webguest50Yes it has MSC/MTP
06:07:12webguest50like under the USB mode under settings right?
06:07:25gartralwhere do you live?
06:07:33webguest50Texas =o
06:07:58gartralok, get the american e200 MI4
06:08:26webguest50wait which one
06:08:36webguest503 american ones.
06:09:29gartral i believe is the proper
06:09:42webguest50one xD
06:10:21webguest50Now what should i do?
06:10:33gartralnow, extract the zip to your desktop, then move just the.mi4 to your 16 meg drive (the recovery partition)
06:11:16webguest50it wasn't a zip it was just the file itself, is that fine?
06:11:57webguest50okay i put it in the 16MB-FORMAT.
06:12:12gartraland that is the ONLY thing you put in there
06:12:37advcomp2019it needs to be PP5022.mi4 unless you had a beta firmware it needs to be firmware.mi4
06:12:56webguest50when i downloaded it it just said 1.02.15a.mi4
06:13:05advcomp2019you can rename it
06:13:13webguest50To which though?
06:13:17webguest50Firmware or PP?
06:14:06gartral(side note) the manual says nothing of renaming the file
06:14:23webguest50okay i renamed it
06:14:27webguest50Now what should i do?
06:14:30advcomp2019gartral, to recovery it you need too
06:15:04gartralcopy it over and disconnect, i believe, but i think that needs varification
06:15:20webguest50okay so disconnect now?
06:15:51gartraladvcomp2019: regardless, the manual doesnt say in any point to rename the file, nor does the e200unbrick page
06:16:50advcomp2019gartral, if it is named anything else, it will not work
06:17:02webguest50o.o...Should i disconnect now?
06:17:11advcomp2019yea you can
06:17:28gartraladvcomp2019: put it in that wiki pages then, that is crucial information and should not be left out!
06:17:40webguest50turning it on
06:17:56gartralcross your fingers, kiddo
06:18:02webguest50It worked!
06:18:05webguest50I love you guys xD
06:18:28advcomp2019gartral, it says that in the unbricking wiki
06:18:30webguest50Now maybe i can do this the right way.
06:19:00gartralso you tryed to patch the bootloaderr of a non R with the R method, is what it looks like to me
06:19:19webguest50>.> and i feel stupid but that happens alot for me.
06:19:46webguest50mind helping me through rockbox too?
06:19:48gartraladvcomp2019: where does it say to rename the file, i cant find it
06:19:51advcomp2019webguest50, if it is working, you can use the rockbox utility
06:20:19webguest50D: I would appreciate being walked through. Once i learn i can probably help someone like yall one time.
06:20:26gartralwebguest50: go get the rockbox utility preferablt version 1.0.7 or 1.0.7b
06:21:08webguest50windows, linux?
06:21:23advcomp2019gartral, the last i looked it did, but i just looked it does not tho
06:21:32gartralwebguest50: if your using the RButil.. it will automagically install everything
06:21:57webguest50no im just saying which should i download?
06:22:02gartraland you'll need the one for your os
06:22:19webguest50So windows?
06:22:21gartralif your running windows xp/vista/7 get windows
06:22:24kadobanthat depends...are you running windows?
06:22:53gartralif your running ubuntu, slackware, or another linux distro, get linux
06:23:52webguest50I did the complete install and it said rockbox install detected backup first?
06:23:55advcomp2019gartral, i have tried other names on my sansas, and PP5022.mi4 was the only one that worked for me
06:24:44webguest50Click yes?
06:26:29webguest50So click no?
06:26:41 Quit Hillshum ("ChatZilla 0.9.83 [Firefox 3.0.3/2008092417]")
06:27:14gartrali wouldnt backup, but its your choice
06:27:35webguest50okay its installing
06:27:37isthatallbackups are for wussies! (tm)
06:28:16gartralthis should be done momentarily
06:28:27isthatallthat's what she said
06:28:44webguest50So how do i install games themes etc? =o
06:28:59gartralisthatall: your getting very offtopic
06:29:09gartralgames are already installed
06:29:09isthatallyes, my apologies
06:29:30webguest50Ah and themes?
06:29:59gartralthemes you can get from and just extract them into your .rockbox directory
06:30:12webguest50it came up with a error saying i might need to install themes manually
06:30:37gartraland you'll want the fonts package, that you should be able too install from rbutil
06:30:40isthatallyes, webguest50 they are redoing that part
06:30:59gartralit's been down for a few months now
06:31:30 Join lightbulbjim [0] (
06:31:40***Saving seen data "./dancer.seen"
06:32:35webguest50You guys are amazing
06:33:15gartralactually, you could have done all this yourself just by reading the manual and wiki...
06:33:51webguest50Either way its awesome!
06:34:10gartralin either case, your welcome
06:34:18webguest50I appreciate it Highly, do you have msn or something?
06:34:20advcomp2019gartral, i just tried it on my sansas and only PP5022.mi4 worked for me unless my sansas are picky heh
06:34:59gartraladvcomp2019: why are you reporting this too me?!? go fix the smegging wiki!
06:35:34advcomp2019i will.. i thought i say it in the logs ;)
06:40:12webguest50is there extra games downloadable?
06:40:32tmztgames are included, what extras?
06:40:52gartralnope, all pluging come compiled into the files
06:40:54webguest50i mean like games other than the ones provided
06:41:01webguest50okay. ^^
06:41:06gartralplugins* damn autocorrection
06:48:21webguest50Night and thank you all.
06:48:32 Quit webguest50 ("CGI:IRC (EOF)")
06:51:09gartralis there still a WPS string that allows a static string of words to be displayed
06:51:44Unhelpfulum, as in, some text? yeah, you put that text in your WPS as-is.
06:52:39gartraland the S tag goes in front or behind for scrolling?
06:53:32Unhelpfuli believe the S tag will scroll all of the text on the line, regardless of where it appears?
06:56:50 Quit intrados (Connection reset by peer)
06:58:01 Join intrados [0] (
06:59:39 Quit intrados (Connection reset by peer)
07:01:41 Join intrados [0] (
07:02:29 Quit intrados (Connection reset by peer)
07:02:55 Join Beta2K__ [0] (
07:03:37 Join intrados [0] (
07:11:27 Quit rocko ("Leaving")
07:14:05gartralwhy are there two "day of the week" tags?
07:14:40 Join Beta2K___ [0] (
07:15:07 Quit CaptainKewl ("( :: NoNameScript 4.02 :: )")
07:15:40 Quit Beta2K_ (Read error: 110 (Connection timed out))
07:20:46 Join ender` [0] (
07:26:33 Quit Beta2K__ (Read error: 110 (Connection timed out))
07:44:49gartralwhy is the sim for the e200 so old?
07:45:26LloreanWe don't provide sim downloads.
07:46:27Unhelpfuli was about to say that they're only of use to developers, but i guess that's not *strictly* true, if you consider WPS creation as non-dev work.
07:57:00 Part gartral
08:02:39 Quit ender` (" Light a fire for a man, and he's warm for a day. Light a man on fire, and he's warm for the rest of his life.")
08:11:13 Nick Barahir_ is now known as Barahir (
08:23:34 Join Rob2223 [0] (
08:26:00 Join flydutch [0] (
08:29:54 Quit amiconn (Remote closed the connection)
08:29:55 Quit pixelma (Read error: 104 (Connection reset by peer))
08:31:40 Join amiconn [50] (n=jens@rockbox/developer/amiconn)
08:31:41 Join pixelma [50] (n=pixelma@rockbox/staff/pixelma)
08:31:42***Saving seen data "./dancer.seen"
08:31:58 Quit perrikwp (" ajax IRC Client")
08:33:22 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
08:39:01 Quit z35 (Read error: 104 (Connection reset by peer))
08:39:29 Join z35 [0] (
08:40:52 Join parkerhiggins [0] (
08:41:53 Part parkerhiggins
08:49:44 Quit Rob2222 (Connection timed out)
08:51:31 Quit ucchan ("Leaving...")
08:52:16 Quit amiconn (Remote closed the connection)
08:52:16 Quit pixelma (Remote closed the connection)
08:53:57 Join amiconn [50] (n=jens@rockbox/developer/amiconn)
08:53:58 Join pixelma [50] (n=pixelma@rockbox/staff/pixelma)
09:00:02 Quit advcomp2019 (Read error: 104 (Connection reset by peer))
09:09:00 Join homielowe [0] (n=ce748651@gateway/web/cgi-irc/
09:23:32 Join fyrestorm [0] (
09:24:24 Quit isthatall (Read error: 110 (Connection timed out))
09:30:19 Join advcomp2019 [0] (n=advcomp2@unaffiliated/advcomp2019)
09:31:37 Join gartral [0] (
09:32:09gartralis there a way to generate a mass-runtime on a folder or m3u/8?
09:34:20gartralactually, while im on the subject, rockbox reports all files to have ridiculously long runtimes
09:35:12Unhelpfuli'm very much inclined to blame your files, or your build. it reports correct runtimes on both of my devices.
09:36:00gartralin the WPS es, its fine, but long select>properties reports times 3 times + longer than there real times
09:37:10gartraland im using the officil build 20074
09:38:00 Join n1s [0] (n=n1s@rockbox/developer/n1s)
09:42:06gartraland it does that fr any file type.. like goban.rock is reported as being 4:49.. which is impossible
09:42:15homieloweWould any dev like a brand new ipod nano 4G 16GB to borrow indefinetly?(its blue :P) my contact email is linked through my wiki name on the IrcNicks wiki page.
09:42:55gartralhomielowe: thats something for #rockbox-community
09:43:54homielowegatral: The reason I put that here so it can be recored in the logs, I think donation of hardware is rockbox related plus not all devs are in -community.
09:54:02gartralUnhelpful: no ideas?
09:55:53Unhelpfulgartral: as i said, either the build or the files are likely to blame. are any of the files ones that you could upload freely?
09:56:23Lloreangartral: Are you sure you're not confusing "Time" (as in Date / Time) with "Run time"?
09:56:47gartralyessir, but try properties on a plugin file as im uploading, youll see what i mean
09:57:02LloreanI see "Time"
09:58:03pixelmathe time reported by "properties" is off for me too, and that it tries to display a time for non-music files is a bit weird
09:58:14Lloreanpixelma: It's the filesystem time, I think
09:58:16LloreanNot the length of the file
09:58:19LloreanNote the line above is "date"
09:58:27gartralif the time entry under date corrisponds to time created, i think documenting that would be a good idea, if not, then its grossly miscalculated
09:58:28LloreanIt's just a HH:MM clock time.
09:58:46tmztfat date?
09:59:28pixelmathen it could have said "date"
09:59:36Lloreanpixelma: There's a "Date" line too
09:59:40LloreanThere's "date" and "time"
09:59:51LloreanMeanwhile in track info we refer to the track runtime as "length"
09:59:52UnhelpfulAH! well, that explains a lot, i looked at the track properties from WPS, which gives track metadata. you're looking from the file browser, which gives file metadata.
10:00:15LloreanIt's the difference between "Show Track Info" and "Properties"
10:00:32pixelmatrue but still somehow misleading, especially when you're not fully awake yet
10:00:39gartrali clearified that, bilieve
10:01:06LloreanI dunno, it seems pretty obvious if you don't immediately assume it's a bug.
10:01:10LloreanDate, Time
10:01:15gartralbut there's no "show trck info"
10:01:28LloreanThere's "Show Track Info" from the WPS when the track is playing.
10:01:47gartralyes, but not from the long select context menu
10:02:04pixelmaok, shown "time" is correct... /me goes to the corner
10:02:18LloreanI don't know what we could call it other than "Time" except maybe "Write Time" or something
10:02:27gartraland i think the ability to check info like that from the browser would be very helpful
10:02:43Lloreangartral: I don't understand why that means you assume an option for *every* file has a value in it that's only valid for audio files, and is a bug for everything else...
10:03:12gartralLlorean: read bac, i assumed it was a gross miscalculation
10:03:18LloreanYes, you assumed it was a bug.
10:03:38pixelmagartral: Rockbox would have to spin up the disk for reading the metadata each and every time unless maybe it could rely on the database *and* when this is loaded to RAM
10:03:44 Quit Llorean (Read error: 104 (Connection reset by peer))
10:03:49gartralit is very misleading, cause windows and linux both show "time" as the playtime...
10:04:06 Join Llorean [0] (
10:04:09pixelmaon HD based players, of course but still
10:04:16Lloreangartral: Do we show *any* audio data related information there?
10:04:20gartralpixelma: doesnt it have too spin up regardless too check the properties?
10:04:51gartralartist title and album
10:05:05pixelmamaybe /me should have stayed in the corner
10:05:08Unhelpfulgartral: by "linux" and "windows", you mean some music player? some specific file browser? you really must be specific.
10:05:12Lloreangartral: Only on audio files.
10:05:24LloreanThere aren't blanks for them on non-audio files.. this should be a clue that "Time" isn't an audio property
10:06:10LloreanAnd the majority of file browsers I've used show time/date modified, but only a specific few and in very certain modes show audio data.
10:06:10gartralpixelma: wouldnt the drive have too spin up to read properties regardless?
10:06:20Lloreangartral: Probably not with dircache enabled.
10:06:27LloreanThough I don't know if it takes advantage of that
10:07:01LloreanAs it stands, we call it "Length" elsewhere, and if it's added to "Properties" it will almost certainly be called "Length" there as well
10:07:58gartralwell, i believe it would be nice for DJing to know how long each song is *before* you play it, as well as timing trips and such
10:08:26LloreanBy default, Windows Explorer does not show audio track lengths.
10:08:47gartraland flash targets wouldnt loose much as far as battery goes
10:08:51LloreanIf you do, it's named "Length" not "Time"
10:09:03LloreanSo please, when you're making claims like that, at least double check first.
10:09:35LloreanThere's nobody stopping you from attempting to add such a feature.
10:10:31 Join bertrik [0] (
10:11:07gartrali did
10:11:27Lloreangartral: That says "duration"
10:11:32LloreanWhich also isn't "time"
10:11:36gartralahh, but i dont have a serer too build from, nor a local copy of the trunk
10:11:59gartralok, well, the ppoint is, windows DOES show the duration
10:12:11gartralof MP3s at least
10:12:13LloreanI never said it didn't. I said it's not part of the default display.
10:12:57gartralbut it is, you just hover the file
10:14:34LloreanI was thinking of the default display as "what it displays to you without action".
10:14:48gartralohh, well, here
10:15:00LloreanAnyway, as I've said several times now, you're welcome to work on it
10:15:10LloreanI don't know what you're trying to convince anyone of at this point.
10:15:20LloreanWe don't accept feature requests any more.
10:15:32amiconnLlorean: Disk has to spin up anyway when showing properties, even with dircache enabled, as properties display is handled by a plugin
10:15:49Lloreangartral: Duration does not show there in windows' default settings
10:16:45gartralyes it is, its displayed with the "details" view option
10:16:57Llorean"default settings"
10:16:59LloreanIs this so confusing?
10:17:18gartralXP Pro's default IS details ;)
10:17:28LloreanVista Pro, it's not, and that's newer.
10:17:42gartralnewer does not always equal better
10:18:01LloreanIn that case, I declare your suggestion "not better" since it's apparently arbitrary. We're done here.
10:18:49LloreanBut this really doesn't matter.
10:18:57LloreanWe don't accept feature requests.
10:21:16gartralsomeone's sore... and besides, i think displaying durations of tracks is a bit more important than say, a game, and i would happily try my hand at implimenting this, if i had the enviroment to do so at the moment, but i dont....
10:21:17 Join isthatall [0] (n=guy2@unaffiliated/isthatall)
10:21:30Llorean"Important" doesn't matter, at all.
10:21:43LloreanIt's whatever interests the person doing the work.
10:22:09LloreanBesides, it's not important in the slightest to people who don't need the feature, exactly like games. It's all very subjective.
10:25:29gartrali was more throwing it out there as a possibility, passively and as something to maybe give though too, there's no need for it, true, but in your argument's light, there's really no need for rockbox itself, but were all here, and using it, so whats really stopping a dev in the future from picking this idea up and implementing it?
10:25:56LloreanI never said anything was.
10:26:10tmztrockbox can only get duration from the codec right?
10:26:22Lloreantmzt: I believe so.
10:26:25tmztit's not stored in the database
10:26:35LloreanSuch a feature shouldn't be dependent on the database.
10:26:43pixelmagartral: the database can show you the length - even in the browser already. You can even sort by length (the default tagnavi uses it when viewing album > albumname > tracks (length)
10:28:31pixelmaif I remember correctly, the database has som virtual minutes and seconds "tag" (Lm and Ls according to the DataBase wiki)
10:30:22gartralLlorean: yes you did, by your rather quick, and militant "we don't accept feature requests" by that statement, you effectively bound all devs working on rockbox from accepting others ideas... but in any case, its true, i would like too see this functionality, and when jd gets his server up, ill putsome time into seeing what it would take, and possible, bring the idea too reality, with some luck and brain-grease
10:30:53Lloreangartral: What part of "we don't accept feature requests" says "nobody is allowed to work on an idea if someone else suggested it"?
10:31:09 Quit homielowe ("CGI:IRC 0.5.9 (2006/06/06)")
10:31:12LloreanIt means we don't accept requests, yes.
10:31:36LloreanThat means "please don't spend time or energy trying to play up your idea here"
10:31:43LloreanPost it in the feature ideas forum, or work on it.
10:31:46***Saving seen data "./dancer.seen"
10:31:53*gartral moves this too community
10:38:46 Join Buschel [0] (
10:39:09*amiconn doesn't want to play the devil's advocate, but since displaying properties is a plugin, that plugin could as well call the core metadata functions to display some more details about audio files
10:40:12Lloreanamiconn: It already does for audio files.
10:40:15LloreanIt just needs expanding.
10:41:49amiconnIndeed it does....
10:50:09bertrikmaybe this is old news, but I just read on the sansa v2 forum thread that AMS now publishes the as3525 datasheet on their web site
10:51:44gartralpixelma: thats great, but i dont use the database as it has a nasty habit of show triplicates of my songs
10:52:15freqmod_gqbertrik: all the wiki pages are updated with references to the datasheet
10:54:50 Join blkhawk- [0] (
10:54:50 Quit blkhawk (Read error: 104 (Connection reset by peer))
10:55:48 Nick blkhawk- is now known as blkhawk (
11:03:43pixelmaUnhelpful: for what it's worth - yesterday I updated some of the files on my c200 (worked on the tags a little bit) and at the moment can't reproduce the crash when entering the <untagged> "album"
11:04:30pixelmathere are still enough tracks in that group
11:07:25Unhelpfulpixelma: hrm. well, i've changed some things about the album name cache in my local tree, although right now, *other* things related to the pf bufalloc removal are keeping me a bit stuck on committing my local work. :/
11:09:02gartralwait, is this crash in the latest builds?
11:09:34Unhelpfulgartral: in pictureflow, and seems to have gone away for her.
11:10:25*gartral has 124 <untagged> files....
11:10:52 Join JdGordon [0] (
11:11:05gartrali... cant scroll in pictureflow...
11:11:51pixelmagartral: which player?
11:11:57gartralwoaw... was your crash a bunch of flickering "dots" making little perferation lines down your screen?
11:12:00pixelmaand scrolling - where?
11:12:11gartralonce "looking" at an album
11:12:34gartrali can move though covers, but going into an album just crashed my player
11:12:40pixelmano, the crash was a "data abort" or "prefetch abort"
11:13:12gartralmy crash happend on the first "named" album
11:13:29Unhelpfulgartral: the wheel should scroll the covers left or right, as well as the tracks up or down.
11:13:45gartralright, it wont scroll tracks
11:14:08gartraland i have a reprocable graphic crash
11:14:25 Quit z35 ("Leaving")
11:14:25 Quit bertrik ("Leaving")
11:14:41gartralbut it happens in the "blank" album, next too the <untagged> one
11:15:06gartralalso, it only shows "gobbledygook" for track names
11:15:27gartralfor like a half a second before what looks like a memory leak kills the player
11:17:25Lloreangartral: Just because you've had a history of corrupt files on your device, have you checked that your Rockbox files are good? (md5sums?)
11:18:08gartralyes, these are MP3s that dont have info in there id3 tag fields
11:18:24LloreanThat's not at all what i asked about...
11:19:06LloreanYour Rockbox files. As in, the files that make up Rockbox. Specifically the plugin, in this case.
11:19:15amiconnLlorean, gartral: It looks like adding duration display to properties.rock would be very easy...
11:19:40 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
11:19:48gartralwell, i have three separate bugs in pictureflow, HIGHLY doubtful that a bad file causes some albums to be able to scroll through track, and some not
11:20:20Buschelanything speaking against submitting FS #9890 (optional LCD and BCM shutdown for iPod 5G)?
11:20:44Buscheldreamlayers found a nice solution imho without changing platform code
11:20:50amiconnSo if this is wanted....
11:21:06LloreanBuschel: What options does it add?
11:21:25BuschelLlorean: the patch uses HAVE_LCD_SLEEP_SETTING
11:21:45Lloreangartral: It'd take you just a few seconds to verify. Better safe than spending time trying other things.
11:22:13Lloreangartral: I'd also reinit your database and see if that changes anything
11:22:28LloreanBuschel: So it uses entirely existing options?
11:22:33BuschelLlorean: you can let the LCD/BCM shut down x seconds after backlight was siwtched off. or you can still let the LCD/BCM active as the 5G has a transflective display
11:22:40*amiconn spots JdGordon and slaps him around a bit with r19894
11:22:48gartralok, heres what i have, <Untagged> wont allow me too scroll through the songs, AT ALL, the "blank" album next to it totally crashes the play in weird manner, and some AA is totally messed up looking
11:23:01BuschelLlorean: yes, the patch does not add new options in the code, but enables an option for the 5G
11:23:17LloreanBuschel: If the BCM shuts down, the screen is cleared?
11:24:08BuschelLlorean: yes, it simply looks like the ipod was switched off. the BCM/LCD will be woken up again as soon as the user uses scrollwheel again
11:24:18LloreanSounds pretty good to me, then.
11:24:47BuschelLlorean: ok, I will check a few things then and may submit today
11:25:19JdGordonamiconn: *wink* some people like being slapped :D
11:25:26JdGordonwtf is 19894 thouhg?
11:26:19JdGordonwhat did i break with that?
11:26:47amiconnIt was breaking all icon display on the Player. I fixed it in r20062 yesterday
11:27:13amiconnLlorean: So what's the conclusing on duration display?
11:27:36Lloreanamiconn: I don't see any reason why it shouldn't be there.
11:28:35 Quit lightbulbjim ("leaving")
11:29:36 Quit jhMikeS (Nick collision from services.)
11:29:42 Join jhMikeS [50] (n=jethead7@rockbox/developer/jhMikeS)
11:29:44Lloreangartral: "totally messed up" isn't very descriptive. Also, have you tried re-initializing the database. You've already mentioned having trouble with it, and pictureflow depends on it.
11:30:39gartralohh, that may be the source of my woes, yes, as i dont even have databaes on >.>
11:32:06Unhelpfuli'm pretty sure PF shouldn't even start if the DB is disabled.
11:32:19gartralLlorean: totally messed up = FUBAR
11:33:08Unhelpfulgartral: that is *still* not descriptive in any useful sense. what do you see, how does it respond to input, etc.
11:33:30LloreanUnhelpful: What if you'd enabled it, then disabled it, so the database files are still around?
11:33:36pixelmaUnhelpful: that gives me an idea. What would happen if you have the database enabled but it's not up-to-date (if you can say from the knowledge of the code)?
11:34:23Unhelpfulthat pretty much depends on what tagcache_search and friends do under those circumstances
11:34:39pixelmaI don't use auto-update so that could have been the case yesterday, now I intitialized again
11:34:55gartralit doesnt respond after the graphics crash, and describe? liiks like someone painted starts on my screen, little 1x1 pixel ones, all over, at random, and, oh, sometimes they flicker, kinda pretty, but i would assume damngous to leave as such
11:35:43LloreanMaybe pictureflow should refuse to run unless auto-update is enabled.
11:36:34gartralor at least check to make sure an update was run withing recent memory? (since last rockbox update)
11:36:53Unhelpfulwell, as long as the tagcache functions return sane data, i'm not really sure how it would crash
11:37:05Lloreangartral: That wouldn't ensure any more valid a database as not caring when the last update was.
11:38:05pixelmaLlorean: I'd be against that (the auto-update dependency), the database also doesn't crash without it. And there are some targets without it
11:38:16gartralwell, the update killed the crash, but i still cant scroll through the tracks of the first 2 albums
11:38:21 Quit faemir ("Lost terminal")
11:38:51gartralpixelma: it PF that's crashing, the database was fine
11:38:52Lloreanpixelma: Maybe trigger a manual update on running Pictureflow?
11:39:05gartralLlorean: bad idea
11:39:20gartralLlorean: that would send it into a restart loop
11:39:31pixelmathey don't have pictureflow though (yet). But I don't want to use it, somehow I don't trust it on targets with removeable storage, seeing what a manual "update now" did or did not on my Ondo
11:39:54 Join faemir [0] (
11:39:55Lloreangartral: There's no reason why it'd have to send it into a restart loop.
11:40:26gartralLlorean: are we talking PFs internal databse of AA, or the main database
11:40:39Unhelpfulgartral: PF does not maintain any database of anything.
11:41:03gartralcause on my e250, a database update requires the DAP to restart to commit the databaeses changes
11:41:37Unhelpfulit queries the core's database to get a list of album names on startup, searches for album art, and stores pre-scaled copies of album art on disk.
11:41:39gartralthe cache of AA isnt a database?
11:42:25Unhelpfulin the sense that a directory with files in it is a database, sure.
11:42:27 Join tyfoo [0] (
11:42:29*gartral beats himself with a stupid-mallot
11:42:40LloreanUpdates shouldn't require a player restart, I thought. Only initialization should.
11:43:01*Unhelpful has never needed a player restart except when first enabling DB.
11:43:51gartralit says in the database info page of debug after an update "databse ready: no" commit postponed: yes"
11:45:29gartralwow, just got a data abort from PF at 0006A06C
11:45:46gartraland it has a half purple screen, half black
11:46:31gartralagain, on the "blank" album with gobbledygook for contents
11:47:59gartralwhy does the database info screen report a progress of -1% when idle?
11:48:06*Unhelpful really wishes sometimes that we had handy debug objects for official builds
11:49:51Lloreangartral: Honestly, I'd try to figure out what's wrong with your media normally before trying to investigate other bugs.
11:49:54gartraland i just hit an odd little minor bug, if you wait until the very-absolute last possible second to hit a key before the display starts its "fade" routine, it will except the key press and keep fading
11:50:02LloreanYou seem to have enough problems with your player that at this point I'm wondering if maybe *it* has problems.
11:50:14tmztyeah meant OF database? not really relevant here
11:50:26gartralLlorean: i cleared all the bad files, i know everything is fine now
11:51:05tmztxxd /dev/sda > /dev/null might be a good one to run
11:51:26gartraltmzt: who are you talking too?
11:51:38tmztgartral: you
11:51:42gartralif its me, im on a winblows box
11:51:57Lloreangartral: You've been in here a few times that I've seen now, where you've been absolutely sure Rockbox had a bug, no doubt whatsoever, up to and including making claims that turned out to be false later about having up to date bootloaders or whatever, that I have a hard time being confident in your player and files.
11:52:02tmztwhat ever the equivalent is there then
11:52:25LloreanJust by the number of times your player has demonstrated corruption, I have a hard time thinking that the player itself is in physically good condition internally.
11:52:39gartrali dont even know what that command is supposed too do, tmzt
11:52:56 Join lightbulbjim [0] (
11:53:07gartralLlorean: i reencoded my entire liabrary as far as i had, im pretty sure everything is fine
11:53:45gartralim running your official build, bootload 5, and have followed your instrutctions too the best of my abilities
11:54:02Unhelpfulgartral: your system hangs if you try to compile anything, i don't trust files encoded on it.
11:54:21gartralwho ever said i used this comp, Unhelpful
11:54:26LloreanI'm suggesting that the problem may not be Rockbox at all, but your player, considering how many times now you've had problems and issues nobody else had.
11:54:47LloreanAnd how many times those issues turned out not to be Rockbox problems, but something that just "happened" locally with no known source.
11:54:51gartralwell, its a firm possibility, as it was a refurb
11:54:57BigBambiGiven the amount of information usually left out, nobody can be sure of anything
11:54:59Unhelpfulgartral: you never said you used another. i assumed you use the same computer you use for other things, which is fairly reasonable.
11:55:39gartralwell, just goes too show, go too assume something, and you make an ass outta u and me
11:56:23Lloreangartral: I wouldn't use that line considering how recently you "assumed" the Time entry was wrong and a bug.
11:56:40BigBambiadded to all the other assumng you have been doing
11:56:52gartralwell, its true, i made an ass outta myself, didnt i?
11:57:16BigBambiBut Unhelpful did not
12:04:14 Quit faemir ("Lost terminal")
12:18:08 Join MethoS- [0] (
12:19:27 Join miepchen^schlaf [0] (
12:26:52 Join Thundercloud [0] (
12:28:23 Quit blkhawk (Broken pipe)
12:28:45 Join blkhawk [0] (
12:30:29 Join blkhawk- [0] (
12:31:50***Saving seen data "./dancer.seen"
12:33:43 Join Chesteta [0] (
12:34:44 Join jaykay [0] (
12:34:59 Quit `Assassin (Read error: 110 (Connection timed out))
12:35:36gartralLlorean: ping
12:36:58 Join {phoenix} [0] (
12:47:20gartralLlorean: if its still of any consolation, the hashes were identical...
12:47:37 Quit blkhawk (Read error: 110 (Connection timed out))
12:48:28 Nick blkhawk- is now known as blkhawk (
12:51:16 Quit Chesteta ()
12:57:19 Join zimba42 [0] (
12:57:38 Join moos [0] (
12:59:22 Quit killan_ ("( :: NoNameScript 4.22 :: )")
13:09:15 Part gartral
13:11:40 Join killan [0] (
13:12:06 Quit {phoenix} (Remote closed the connection)
13:25:36 Nick Horschti is now known as Horscht (n=Horscht@xbmc/user/horscht)
13:25:59 Join GodEater [0] (n=ge@rockbox/staff/GodEater)
13:26:53 Quit miepchen^schlaf ()
13:28:53 Quit jaykay (Read error: 110 (Connection timed out))
13:42:01 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
13:51:59 Join kugel [0] (n=kugel@rockbox/developer/kugel)
13:54:43 Join J [0] (
13:55:13 Quit J (Client Quit)
13:57:07 Join morrijr [0] (
13:57:12 Quit morrijr (Client Quit)
13:58:40 Join J [0] (
13:59:09 Nick J is now known as Guest51519 (
13:59:31 Quit Rob2223 (Read error: 104 (Connection reset by peer))
13:59:46 Quit Guest51519 (Client Quit)
14:00:00 Join Rob2222 [0] (
14:10:13 Quit Thundercloud (Remote closed the connection)
14:12:56*freqmod_gq made a simple patch that displays an error when amssansa models tries to access untransferable blocks
14:13:25 Join casainho [0] (
14:18:09casainhohello :-)
14:23:44 Quit Rob2222 ()
14:24:10 Join Rob2222 [0] (
14:24:42 Quit tyfoo (Read error: 104 (Connection reset by peer))
14:24:52casainhoyesterday I made a presentation to Rockbox and ROckbox Player
14:25:21casainhohere the pictures:
14:25:55casainhoIf I want to make audio drivers,. I just need to work this main fucntion? void pcm_play_dma_start(const void *addr, size_t size) ??
14:26:04casainhocan I do it without using DMA?
14:26:36casainhoso, I just need to send the data from that addr and with length size?
14:26:56casainhohow is data organizated? 16bits per left/right channel?
14:31:53***Saving seen data "./dancer.seen"
14:34:35 Join miepchen^schlaf [0] (
14:35:24linuxstbcasainho: Yes, you can do it without DMA. The ipod audio drivers are probably a good example - they use the FIQ (fast interrupt) to feed the audio FIFO. But the FIQ handler is ARM assembler, so you'll need to adapt that.
14:37:44 Join faemir [0] (
14:37:55 Quit isthatall (Remote closed the connection)
14:41:09casainhobut, audio buffer is sent to where?
14:41:13casainhoto DAC IC driverS?
14:41:18casainhodo dac ic?
14:41:24casainhoI think is like that...
14:41:26linuxstbThe ipods use I2S to send the data to the DAC
14:41:44casainhothanks you!! great - I iwll meditate on this info
14:41:53casainhoIm ust go now - take a train and read the code :-)
14:41:56casainhobye bye :-)
14:42:53rasherI guess a new bootloader is in order when rockbox usb is enabled...
14:43:27linuxstbWhat for? To not reboot to the OF when USB is detected?
14:43:34*kugel would like to have a USB mode in the bootloader, for recovery purposes
14:43:58rasherlinuxstb: Yes indeed
14:44:04linuxstbkugel: Then work on enabling interrupts and making sure you don't break dual-booting
14:45:46linuxstbrasher: I think that only affects the Sansas then - I don't think the ipod bootloaders check for USB.
14:46:10 Part zimba42
14:46:11*linuxstb always thinks "ipod" when people talk PP, but senses others always think "sansa"...
14:46:53*BigBambi always thinks "urgh"
14:47:04*kugel never thinks
14:48:48kugellinuxstb: is there a known problem with interrupts, bootloader and dual booting?
14:49:53linuxstbNo, but at least the ipod OF seems quite sensitive to the state that Rockbox leaves the hardware in. So it wouldn't surprise me if that caused problems (hopefully it won't...)
14:57:31 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
15:01:12 Quit casainho (Read error: 110 (Connection timed out))
15:03:01 Quit kadoban (Read error: 104 (Connection reset by peer))
15:03:51 Join kadoban [0] (
15:06:17 Join midijunkie [0] (
15:08:48*linuxstb wonders what itunes will think of an ipod in Rockbox USB mode
15:08:59 Quit miepchen^schlaf ()
15:10:38 Quit J-23 (Remote closed the connection)
15:20:59 Quit Buschel ()
15:24:09linuxstbgevaerts: Regarding the discusssions of supporting various USB classes in -community, would it be relatively easy to do this as regular plugins? i.e. the user would start a specific plugin, and that would make Rockbox act as a usb device of that class?
15:27:06gevaertslinuxstb: I think so, yes. The actual interface between the stack and the driver is reasonably small and well defined (defined, not documented :)), so this is definitely possible. The main thing you need to be careful about is that you can't really change the USB configuration while connected, so these plugins need to be started before connecting, and can't be stopped before disconnecting
15:27:09 Quit JdGordon (Read error: 60 (Operation timed out))
15:27:18 Join einhirn [0] (
15:27:49 Join SirFunk_ [0] (
15:28:14gevaertsOne other thing we need to think about is whether we want to support multiple simultaneous device classes on hardware that can do this (enough endpoints)
15:28:35linuxstbHmm, that complicates things...
15:29:36linuxstbBut I guess it's also dependent on what the classes do - i.e. I'm guessing you can't have MSC at the same time as any other class needing disk access.
15:29:44 Join tessarakt [0] (
15:30:26gevaertsI think most combinations don't make very much sense, only very few of them really do.
15:31:01gevaertshm, maybe they do...
15:31:07*gevaerts thinks a bit
15:31:17linuxstbMaybe the plugins for classes that make sense to do together could be combined.
15:31:54linuxstbSo what is the list of potential classes?
15:31:57gevaertsYou can combine MTP with anything except UMS, HID can go with anything, audio can go with anything, ...
15:32:45linuxstbIt depends what audio is implemented - i.e. audio from computer -> Rockbox, or Rockbox -> Computer
15:33:20gevaertsI think we should think about MSC (possibly with MMC), MTP, HID, Audio, (possibly) Ethernet, serial
15:34:39 Quit MethoS- (Read error: 104 (Connection reset by peer))
15:34:51linuxstbWhy would Ethernet be useful? Is there even a Class for that?
15:35:08 Join MethoS- [0] (
15:35:08gevaertsAudio is a very flexible thing. You can do PC->DAP, DAP->PC, or DAP->DAP (i.e. expose the volume and EQ settings to the PC)
15:36:00Nico_PLlorean: are we going to aim towards including the gigabeat S into the next release?
15:36:15gevaertsThere is a class for that, yes. I would possibly start with PPP though (because we have a lot of serial already). It would need lwip or a similar IP stack, and then we can do direct podcast download, streaming audio, and other weirdness
15:36:21 Quit FlynDice (Remote closed the connection)
15:36:47Nico_Pgevaerts: you mean the DAP could access the internet?
15:36:56linuxstbgevaerts: I don't understand the point though - wouldn't it need to be attached to a PC anyway?
15:37:07gevaertsNico_P: yes, if your PC acts as a router
15:37:14*linuxstb is trying to keep this discussion sensible ;)
15:37:16Nico_Pthat's pretty cool
15:37:49gevaertslinuxstb: you don't want an AJAX web WPS?
15:38:02linuxstbErrr.... No.
15:38:33*freqmod_gq sees a rockbox unit connected to an openwrt firewall with usb downloading automatic podcasts from the internet
15:38:35 Join fdinel [0] (
15:39:50gevaertsNetwork probably makes a lot more sense if you have USB host
15:40:11 Quit gromit` (Read error: 60 (Operation timed out))
15:40:21linuxstbgevaerts: Which target have enough endpoints?
15:40:54gevaertslinuxstb: for now, the ARC controller, so portalplayer and imx31
15:40:56 Quit SirFunk (Read error: 110 (Connection timed out))
15:41:53linuxstbHow many simultaneous classes could they have?
15:44:13gevaertsDepends on the class. IIRC portalplayer has 3 endpoint pairs (so 3 IN and 3 OUT), while imx31 has 8. This includes EP0, so practically it's 2 and 7
15:44:35gevaertsThat means that PP can usually combine two classes, while imx31 can do whatever you like
15:45:27gevaertshm, the ZVM may have enough endpoints for multiple classes as well
15:45:52linuxstbSo EP0 is always used for the "handshaking" between host and device, and the others are used for the classes themselves?
15:46:14gevaertsAudio is a bit special as it uses isochronous, which isn't necessarily possible everywhere (again, ARC seems to do it, the others not)
15:46:57 Join gromit` [0] (
15:47:27gevaertsIt's a bit more complicated than that, as e.g. audio also uses EP0 for control-style things like setting the volume. IIRC you can have a pure-control audio device with only EP0 (so you can combine it with anything else on all targets)
15:47:50linuxstbI think I get the idea.
15:47:59 Quit midijunkie (Read error: 104 (Connection reset by peer))
15:48:10gevaertsIt's complicated :)
15:50:37Nico_Pgevaerts: how far are you from trying out the tracer on the USB stack?
15:51:26linuxstbNico_P: Regarding putting the Beast in 3.2, it could be useful to make a "todo" list of what needs to be done.
15:51:34gevaertsNico_P: about a week. mcuelenaere has the tracer now, and he's going to send it to me on tuesday or wednesday, so I should have it on thursday or friday at the latest
15:51:57Nico_Plinuxstb: yes, I'll do it
15:52:20linuxstbNico_P: Maybe somewhere on the GigabeatSPort page/
15:52:56Nico_PI don't really know what needs to be done actually
15:53:00Nico_Pgevaerts: good news :)
15:53:07linuxstbI guess firstly we need an official bootloader release - on the download server.
15:53:37BigBambiI'll have a go at putting the install instructions in the manual
15:53:39linuxstbWe then need to decide if we're happy with the current sendfirm install method, or if we want to write something that's easier to use.
15:53:49BigBambiI guess a beastpatcher type thing would be nice
15:54:10linuxstbThen decide if we want to officially support dual-boot, and if so, write instructions on how to do that.
15:54:33BigBambiand also rbutil integration - at least to the extent of it calling an external .exe for Windows as licence etc
15:54:51linuxstbIf we want dual-boot, we need to either make binary releases of mknkboot, or incorporate that functionality in sendfirm (or beastpatcher if it exists)
15:55:10BigBambilinuxstb: I don't think there is any reason not to support dual-boot now there is a download of the OF available
15:55:24linuxstbThe reason is that it makes the install process more complex
15:55:42linuxstb(what I just said...)
15:56:06linuxstbBut now Rockbox supports usb and charging on the S, dual-booting isn't a requirement for release (IMO).
15:56:12Nico_Psingle boot is also faster
15:56:18BigBambiI was going on that we would have a beastpatcher type thing eventually
15:56:20gevaertsNico_P: did you see the data abort issue in the buffering code after USB disconnect?
15:56:37 Join tyfoo [0] (
15:56:37BigBambiBut no, go for single at first IMO - faster, usb, charging etc
15:56:38Nico_Pgevaerts: yes, I wanted to discuss it with pondlife
15:56:42linuxstbBigBambi: So we could ignore dual-boot for now, but add it as a feature of beastpatcher.
15:56:53BigBambilinuxstb: yes, sounds sensible
15:57:07linuxstbOr someone could continue with saratoga's idea to "undelete" the OF nk.bin from the firmware partition.
15:57:08gevaertsNico_P: I'll open an FS entry to keep track of it
15:57:19Nico_Pgevaerts: you mean the one in FS #8894, correct?
15:57:22BigBambiAlso, dual boot doesn't really bring anything - nothing is compatible (music, videos etc)
15:57:41BigBambiAnd we have radio, charging, USB, ....
15:57:48gevaertsNico_P: no.
15:57:57linuxstbDoes the S encrypt media files like the F?
15:58:18linuxstbOr is there something else stopping Rockbox playing OF audio files?
15:58:31Nico_Pgevaerts: oh, I didn't see that. I though you were referring to pondlife's comment here:
15:59:26BigBambilinuxstb: Not sure actually now you mention it - I thought it wasn't possible but I'll have to check
15:59:38BigBambiExcept I don't have a dual-boot bootloader
16:00:06 Quit SirFunk_ (Read error: 110 (Connection timed out))
16:00:08linuxstbNico_P: Shall I start a "todo" list for the Beast?
16:00:21Nico_Plinuxstb: I've started editing
16:00:27linuxstbYes, I just noticed ;)
16:01:07 Join SirFunk_ [0] (
16:04:32Nico_Plinuxstb: done:
16:04:36 Join Aurix_Lexico [0] (
16:06:09n1slinuxstb: rockbox can play files transferred in the OF fine but they are renamed sor of like itunes style
16:06:26gevaertsFS #9935
16:06:29BigBambiOK, my error - cheers :)
16:07:01 Join TheSphinX^ [0] (
16:07:20linuxstbNico_P: Are you editing it again?
16:07:29Nico_Plinuxstb: no, I'm finished
16:07:41Nico_Pwhy, it says I am?
16:07:42linuxstbYou still have the lock on the page.
16:07:53Nico_Pand now?
16:08:04linuxstbOK now.
16:08:14Nico_PI had forgotten to cancel the editing
16:08:17*kugel thinks dual boot on the S should be officially supported, but not that it should delay the release
16:08:22*gevaerts sends all playback and buffering specialists to FS #9935
16:08:43Nico_Pgevaerts: so it seems to be the same thing pondlife reported at
16:08:52kugelgevaerts Nico_P needs to comment on first
16:09:21Nico_PI guess audio_get_buffer() simply needs to warn the buffering thread somehow
16:09:23gevaertsNico_P: hm, indeed. I didn't read that properly the first time...
16:10:09*gevaerts was mislead by that task's global topic
16:10:44 Join SirFunk__ [0] (
16:12:02gevaertsIf needed 3.2 could statically allocate the needed buffers for USB (about 32K), but I'd like to avoid that
16:12:31Nico_PI think it should be pretty fixable
16:12:43kugelwould that make audio playling while usb at least?
16:13:11gevaertskugel: no. That's not possible in general because you can't access the disk
16:13:23gevaertsThis is to stop rockbox crashing
16:13:36kugelhow about the plugin buffer?
16:13:54kugelor codec buffer?
16:14:24kugelcodec buffer is unused anyway
16:15:14 Quit MethoS- (Remote closed the connection)
16:15:17gevaertsPlugin buffer would be tricky, as we don't force plugins to close. Codec buffer should be exactly as unused as the audio buffer
16:16:10 Join jaykay [0] (
16:17:16gevaertsThe difference is that there is actually a public exposed function to grab the audio buffer, which there isn't for the codec buffer
16:18:48kugeldoes there need to be a function? the codec buffer is a section, so you should be able to do something like *p = CODEC_START
16:19:08kugelanyway, just saying because grabbing audio buffer seems to cause problems
16:22:12gevaertsI see two problems with that. (a) are you sure there are no hidden assumptions in the code about the codec buffer remaining untouched?, and (b) from what I can see you can hit exactly the same bug with other code that grabs the audio buffer (i.e. various plugins) (although it may be less likely)
16:22:13linuxstbNico_P: I've expanded the TODO list...
16:22:32Nico_Pgevaerts: there are a few things to take care of in order to do it properly, but it should be OK
16:23:26Nico_Plinuxstb: do you think it is possible to use another name than nk.bin?
16:23:43linuxstbNico_P: Why not?
16:24:01 Quit SirFunk_ (Read error: 110 (Connection timed out))
16:24:03linuxstbOr does sendfirm send the actual filename as well?
16:24:12Nico_PI thought maybe it is hard coded into the first stage BL
16:24:36Nico_Poh right, it will just get renamed by MTP
16:25:46linuxstbLooking at sendfirm.c, it looks like it's hard-coded to send "nk.bin" as the filename anyway.
16:26:32 Quit Rob2222 ()
16:27:27*gevaerts should test what happens if you connect while playing doom
16:27:52 Quit TheSphinX^ (Read error: 110 (Connection timed out))
16:28:02 Join miepchen^schlaf [0] (
16:30:13*n1s guesses on "doom" :)
16:31:57***Saving seen data "./dancer.seen"
16:32:28 Join rocko [0] (
16:32:44*BigBambi is attempting to manualify the beast install instructions
16:33:01rockoROCKBOX RULES
16:33:13BigBambiyes indeed
16:33:50Nico_Pgevaerts: how frequent is the crash supposed to be?
16:34:51gevaertsNico_P: I think I've seen it in about one in two or one in three cases with USB disconnects when I connected with audio playing
16:35:34*linuxstb bites the bullet and starts beastpatcher
16:36:26linuxstbI guess moving utils/MTP to rbutil/.MTP makes sense? Even though it's not going to be directly linked to rbutil?
16:36:33rockolinuxstb how are you buddy?
16:36:54BigBambirocko: Please read the guidlines, this is an on-topic channel
16:37:07Nico_Pgevaerts: what is required to to for audio_get_buffer to be called?
16:37:55gevaertsNico_P: if you build with USE_ROCKBOX_USB, it will all happen automatically. What target are you testing on?
16:38:11Nico_Pgigabeat S, which has USE_ROCKBOX_USB
16:38:20Nico_PI thought maybe I needed to copy a file or something
16:39:25Nico_Pso far it refuses to crash
16:39:36gevaertshm, I need to try on my beast
16:40:10gevaertsI've reproduced it on my mini G2, and people on the forums have done it on sansa
16:40:30Nico_Phas anybody tried it in the sim?
16:40:52gevaertsnot that I know of
16:41:03 Join J-23 [0] (
16:43:16BigBambilinuxstb: Is the plan to get beastpatcher to also fix the partition table?
16:43:53 Join TheSphinX^ [0] (
16:44:41linuxstbBigBambi: No. Beastpatcher will just use MTP. (at least initially).
16:45:06BigBambilinuxstb: OK, that's fine - I'll put the fixing the partition table bit in the manual then
16:46:37 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
16:46:44 Quit kugel (Nick collision from services.)
16:46:49 Join kugel [0] (n=kugel@rockbox/developer/kugel)
16:47:49 Join grdxyxy [0] (n=eric@
16:48:27 Join Zoxc [0] (
16:50:58linuxstbNico_P: Do you know why sendfirm.c bothers to define _LARGEFILE_SOURCE ? We're only sending the small firmware file.
16:51:40Nico_Plinuxstb: that's from the original mtp-sendfile, that was supposed to be able to send anything. maybe it's not needed anymore
16:51:49Nico_Ptbh I don't even know what it means
16:52:08linuxstbIt means the app will use 64-bit integers for things like file sizes and offsets.
16:52:40linuxstbI think it's only needed on Linux - BSD/OS X use 64-bit by default I think.
16:53:15linuxstbsansapatcher needed it because the firmware partition is > 4GB into the disk (and it opens the whole-disk device and seeks in it).
16:58:41gevaertsNico_P: copying some stuff seems to help reproducing this on my mini
16:58:45*gevaerts thinks
16:59:02gevaertsDoes the beast actually crash on unaligned accesses at all?
16:59:39kugelgevaerts: I think it doesn't
16:59:54*kugel remembers jhMikeS talking about that as a feature of armv6
16:59:59gevaertskugel: thanks. That would explain why it doesn't (in this case) :)
17:00:39Nico_Pthen I guess I need to try it on the ipod
17:07:05 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
17:09:12 Join saratoga [0] (n=41becb3b@gateway/web/cgi-irc/
17:13:34Nico_Pbuffering seems to have gone back to being rather slow on the 5.5G
17:14:41*freqmod_gq is unhappy with ARM for removing it's PL180 manual because it sees the PL180 as obsolete
17:15:20kugelfreqmod_gq: I have a copy, you want it?
17:15:41kugel(if I'm allowed to give it away, that is)
17:15:55freqmod_gqkugel: well, if you know that it doesn't contain any information on bank switching i don't need it
17:16:03BigBambiWhen the beast bootloader fails to find the firmware, does it just say "File Not Found" or does it also give a -1 error?
17:16:04kugelit didn't help me much, anyway
17:16:42*kugel hands BigBambi a copy of the source
17:16:49freqmod_gqi have coded the framework code that detect which bank an adress is in, and gives a panic when the bank is not bank 0
17:17:02kugelthe patch you mentioned?
17:17:03freqmod_gqbut i don't have any clue on dissassembling the firmware so i am kind of stuck
17:17:07kugelthat was hardly based on banks
17:17:14freqmod_gqyes, however it is not updated
17:17:20BigBambikugel: yes, obviously. However if anyone knew off the top of their heads it sould be much quicker
17:17:25freqmod_gqi know it is no hardware stuff there
17:17:27freqmod_gqi know
17:17:36linuxstbBigBambi: No bootloaders should say "-1" anymore.
17:17:47freqmod_gqthe only interesting thing i found out is that the banks is at least bigger that 1GB
17:17:51*linuxstb wonders if all that untar stuff is still needed in the beast bootloader
17:17:52freqmod_gq(most problably i GIB
17:18:00BigBambilinuxstb: Ah, OK - that's a general issue with the manual then
17:18:10BigBambilinuxstb: Do they all say "File Not Found" now then I guess?
17:18:24linuxstbBigBambi: I don't think the iriver (h1x0/h3x0) bootloader have been updated for a long time though - so they may still say it.
17:18:53linuxstbi.e. it was changed in SVN a long time ago - but not all targets have had new bootloaders released.
17:20:29freqmod_gqkugel: well, is a bit more elaborate, but doesn't contain any reverse engineering actually fixing the bug
17:22:01freqmod_gq(and it is not tested)
17:22:14Nico_Pgevaerts: not much luck on the ipod either, I'll try the sansa
17:22:48kugelfreqmod_gq: Actually, I think it's worth looking into the PP sd driver again
17:23:06kugelit has bank switching, and the recent macro renaming made it a bit more readable
17:23:10freqmod_gqi looked at it and it sends a custom command and then writes to a register
17:25:57freqmod_gqehh... looking at the copy_write_sectors it looks like it writes to the device (MMC_DATA_FIFO)
17:25:57saratogamaybe check if that register is present in the AMS code?
17:27:00freqmod_gqwell, i would suggest looking for the custom ATA command, however i am not good at understanding assembly
17:27:14 Join Lear [0] (n=chatzill@rockbox/developer/lear)
17:29:20kugelsaratoga: ata-sd-pp sends a custom command (35)
17:29:31kugelvender specific, apparently
17:29:45kugeland it's not part of the command definitions in firmware/export/sd.h
17:30:15gevaertsUnhelpful: have you had more issues with usb serial? Which device are you using?
17:30:33 Quit saratoga ("CGI:IRC (EOF)")
17:30:56 Join saratoga [0] (n=41becb3b@gateway/web/cgi-irc/
17:31:24saratogaso checking if that same command is sent in the AMS binaries is probably not trivial
17:31:32 Quit __lifeless (Read error: 104 (Connection reset by peer))
17:31:54kugelfreqmod_gq: do you have a disassembly handy?
17:32:04LearNeat. Seems like the newlib setjmp (ARM version) just worked. Now the big question is, should I add file headers? :)
17:32:44freqmod_gqwell, i dissasembled the sd_recoved part with dissarm, but i don't think i did it properly, i don't understand it, and i can't find the instructions in the ARM ISA manuals
17:32:54freqmod_gqso for all practial purposes, no
17:33:22kugelfreqmod_gq: have you found a sd_cmd call?
17:33:32*freqmod_gq needs to prospone reverse engineering until he has taken compilator tecnic
17:33:44freqmod_gqi haven't found anything in the firmware
17:33:58kugelyou could search for the MCICommand address
17:34:13 Join _lifeless [0] (n=lifeless@
17:34:29kugeland watch what's written into the registers just before the corresponding b(l) instrunction
17:34:57 Quit TheSphinX^ (Remote closed the connection)
17:36:38*freqmod_gq sees lots of 0xc8000000 (the base address from the now publicized AS3525 datasheet)
17:37:33 Join mirak [0] (
17:38:13freqmod_gqwhere lots is 3-4 occurences
17:38:22gevaertsmcuelenaere: I think the current usb_serial.c is fine. There seems to be an issue with PP502x that makes it stop sending data now and then, but on imx31 it works perfectly. I haven't had it lock up the player
17:38:34kugelfreqmod_gq: well, 1 is probably enough
17:39:01mcuelenaeregevaerts: hmmm ok, so SoC specific bug after all
17:39:04freqmod_gqis bicccs an arm instruction?
17:39:08kugelMCICommand is MCIBASE+0x0C
17:39:13 Join TheSphinX^ [0] (
17:39:24kugelfreqmod_gq: yes, but that's not thumb code
17:39:35kugelfrom my experience, most of the OF is thumb code
17:39:42freqmod_gqmaybe disarm is not the best dissasembler out there,
17:39:51freqmod_gqdoes anybody have any recomendations
17:39:59freqmod_gqobjdump would not recognize the files
17:40:18freqmod_gqmaybe i sendt wrong flags
17:40:18kugelI could do it with objdump
17:40:26kugelyou need arm-elf-objdump of course
17:40:31gevaertsmcuelenaere: probably. There were bugs in it until last night though, somaybe your issues are gone anyway
17:41:10*freqmod_gq tries objdump some more
17:41:31linuxstbfreqmod_gq: You want something like "arm-elf-objdump -D -b binary -m arm file.bin > file.txt"
17:41:37 Quit faemir ("Lost terminal")
17:41:52kugelfreqmod_gq: I may send you my disasm file
17:42:02linuxstbAnd add -M force-thumb (I think) for a thumb mode disassembly.
17:42:49freqmod_gqi think i haven't managed to strip of the segment header properly
17:43:26freqmod_gqi cut the file right after the last DEADBEEF before sd_reload
17:45:13freqmod_gq(offset C604 in m300a.bin from the firmware archive)
17:45:42 Quit kugel (Nick collision from services.)
17:45:47 Join kugel [0] (n=kugel@rockbox/developer/kugel)
17:46:57kugelwell, I can't manage to find c800 in my disasm file
17:50:07freqmod_gqmaybe i didn't manage to cut it properly
17:50:32 Quit Horscht ("Verlassend")
17:50:45freqmod_gqthere is a sd_reload in the middle of the file (at F399)
17:51:03freqmod_gqand then the point was to search backwards to the last DEADBEEF before that
17:52:56*freqmod_gq begins to wonder if the offset display in okteta works properly
17:53:16rasherHm, if copying a .mi4 and bootloader .rom and sansa.fmt to my sansa while in recovery mode results in the e200 still starting in manufacturer mode, is there any hope left?
17:54:14 Join ender` [0] (
17:54:39 Join faemir [0] (
17:54:40freqmod_gqahh 24f39D, that makes more sense (that's the sd_recover)
17:54:44kugelfreqmod_gq: I have only disassembled the firmware part. is the SD stuff in a "plugin"?
17:54:52freqmod_gqkugel: yes
17:54:58freqmod_gqthat's the last library
17:55:39freqmod_gqhalfway down the page
17:55:51freqmod_gqthere is a list of library blocks
17:56:05 Quit grdxyxy ("Leaving.")
17:56:36 Quit faemir (Client Quit)
17:56:56gevaertsrasher: check the forums in the hardware section. IIRC this can mean that you just have to click the memory module back in place, and if it's not that I think some people have managed to recover as well
17:59:43rashergevaerts: And the fact that recovery mode appears to function correctly isn't an issue?
18:00:25gevaertsrasher: I don't really know
18:00:33rasherAlright, I'll read a bit
18:02:38 Join faemir [0] (
18:02:49 Quit kugel (Nick collision from services.)
18:02:52rasherThis looks problematic... crikey
18:02:53 Join kugel [0] (n=kugel@rockbox/developer/kugel)
18:02:59freqmod_gqkugel: do you know what magic i have to have at the start to make arm-elf-objdump accept the dump (currently i have cut out 23c600 - 24f39d)
18:03:01 Join stripwax [0] (
18:05:22kugelfreqmod_gq: no, sorry. the commands linuxstb gave you worked for me
18:05:24 Quit bs66_ (Read error: 110 (Connection timed out))
18:05:40freqmod_gqkugel: which offset do you use?
18:06:03linuxstbfreqmod_gq: What is the problem? arm-elf-objdump is giving you an error, or there is problem with the output?
18:06:12freqmod_gqarm-elf-objdump -M force-thumb -d fwct2 arm-elf-objdump -M force-thumb -d fwct2
18:06:17freqmod_gqehh (one time)
18:06:24freqmod_gqgives arm-elf-objdump: fwct2: File format not recognized
18:06:41freqmod_gq(the arm-elf-objdump is created by the rockbox cross compile script)
18:06:51linuxstbfreqmod_gq: I suggested the right command-line earlier.
18:07:38*freqmod_gq does sloppy reading, and tries that command
18:08:00kugel"arm-elf-objdump -D -b binary -m arm fwct2 -M force-dumb > file.txt"
18:08:41freqmod_gqthat worked (with force-thumb)
18:09:36kugeloh, typo
18:10:02kugelfreqmod_gq: can you send me the file?
18:10:17 Join bs66_ [0] (
18:13:30freqmod_gqi don't think the output is that useful, i can't find the adresses i found with disarm
18:14:03*freqmod_gq tries to find out what he sent to disarm and send the same to objdump
18:16:55linuxstbmcuelenaere: Sorry to always be the license poilice, but what about that touchscreen code? Code shouldn't be committed to SVN (and hence current builds) if the license doesn't allow it....
18:17:15mcuelenaerelinuxstb: yes I know, I'm currently a bit busy with USB though
18:17:21mcuelenaereI'll handle it later this evening, is that ok?
18:18:47 Quit faemir ("Lost terminal")
18:18:54linuxstbDo you know what you can do about it?
18:19:13 Join blkhawk- [0] (
18:19:16mcuelenaereI'll probably either try contacting the author or rewrite it
18:19:51mcuelenaereMy first idea was to base myself upon tslib, but that uses floats and my rewritten version of that didn't seem to work quite good
18:22:21linuxstbThe discussion in IRC the other night suggested reverting it, then people said that it was OK as builds weren't being distributed. But we do have cowond2 and mrobe500 builds for download...
18:23:31*linuxstb doesn't ultimately care though, as it's not on his servers, and it's unlikely the author is going to complain.
18:24:31kugelwell, the author cannot complain, can he? we don't break his license as far as I see
18:24:38mcuelenaereoh ok, then I should probably revert it for now?
18:25:49linuxstbmcuelenaere: Maybe just wait and see what others think, but if it was my code, I would revert it.
18:25:58kugelthe only issue I see is, that it's incomatible with our license, so we basically break our license.
18:26:22gevaertskugel: and that of all projects that we use code from
18:26:50 Quit Tristan (Remote closed the connection)
18:26:54*freqmod_gq found out that disarm puts out totally different addresses than objdump and that it is problably totaly confused
18:27:09 Quit blkhawk (Read error: 110 (Connection timed out))
18:27:30kugelgevaerts: uhm, yes. But not the license of the touchscreen code author :)
18:27:48kugel(which was my point, it's nevertheless incompatible)
18:27:51Nico_Pgevaerts: I believe I have a fix for FS #9935
18:28:16Nico_PI just added a call to buffering_reset() in audio_get_buffer()
18:29:59gevaertsNico_P: do you have a patch to test?
18:30:08 Quit einhirn (Read error: 104 (Connection reset by peer))
18:30:46Nico_Pgevaerts: sure,
18:31:24 Join midijunkie [0] (
18:31:33Nico_PI managed to repro the crash by transferring files (It didn't happen without transfers), and now it seems to be gone
18:32:02***Saving seen data "./dancer.seen"
18:32:57*gevaerts builds
18:34:00Nico_Pwhat I'm not quite sure of is whether the audio_stop_playback call that happens on USB connect actually manages to close all the handles
18:34:34 Join faemir [0] (
18:34:35Nico_Pif the buffering thread has acknowledged the USB connect before the audio thread, I guess the answer is no and I would even have expected a deadlock
18:35:26 Quit midijunkie (Client Quit)
18:35:30*mcuelenaere sent a mail to Carlos E. Vidales
18:37:56gevaertsNico_P: it seems to be gone here as well
18:39:24Nico_Pgevaerts: what happens when the caller of audio_get_buffer is done with the buffer?
18:40:23gevaertsNico_P: as far as I know, nothing, until playback is restarted. There is no audio_release_buffer or anything like that
18:40:50 Join LambdaCalculus37 [0] (n=rmenes@rockbox/staff/LambdaCalculus37)
18:42:39 Quit Lear ("ChatZilla 0.9.84 [Firefox 3.1b3pre/20090217034531]")
18:45:05 Quit Darksair ("People who are zhuangbility want to show their niubility but only reflect their shability.")
18:45:26 Quit tessarakt ("Client exiting")
18:45:42 Quit faemir ("Leaving")
18:45:51Nico_PI think reference to release 3.2 needs to be added to the tracker
18:46:01 Join faemir [0] (
18:48:25*mcuelenaere thinks he needs to dump the registers on exception panicf's
18:49:17mcuelenaeregevaerts: I got a Load Address Error exception in writeFIFO() when doing rapid logf()
18:49:26mcuelenaerecould usb_serial somehow pass invalid pointers?
18:51:04gevaertsit shouldn't
18:51:24gevaertsDo you need aligned data in some way?
18:51:36*mcuelenaere doesn't know
18:52:35 Join lymeca [0] (
18:52:53 Quit jaykay (Read error: 110 (Connection timed out))
18:53:06gevaertsusb_serial assumes that it can pass data to usb_drv_send*() that's not aligned in any specific way
18:53:51mcuelenaereaccording to the manual, an address error exception occurs when: Fetch an instruction, load a word, or store a word that is not aligned on a word boundary.
18:54:00 Join {phoenix} [0] (
18:54:36gevaertsIs that a 16-bit FIFO?
18:54:37 Join Thundercloud [0] (
18:55:10mcuelenaerenope, it's 32-bit; but it can be accessed in 32-, 16- and 8-bit
18:55:27Nico_Pgevaerts: I'll wait for some feedback on the FS entry before committing
18:55:48 Quit saratoga ("CGI:IRC (EOF)")
18:55:57gevaertsNico_P: good idea
18:56:57Nico_PI'd especially like to hear from jhMikeS about what happens to queue messages when USB gets connected (i.e. the issue I mention in my FS comment)
18:57:38 Nick blkhawk- is now known as blkhawk (
18:57:48gevaertsyes. The SYS_USB_CONNECTED is somewhat special as it tends to stop all processing
18:59:26 Quit _lifeless (Remote closed the connection)
18:59:42 Join _lifeless [0] (n=lifeless@
19:00:38mcuelenaerehmm the bad address is 0x80417569 (which gets loaded as word, so probably it needs aligning)
19:01:13*Nico_P has to go
19:01:15 Quit Nico_P (Remote closed the connection)
19:01:57gevaertsI'm not sure about the best way to make USB transfers use aligned data. For UMS it's easy, but for others it will probably involve lots of copying data around
19:02:29mcuelenaereit's probably something I can handle in writeFIFO() (write data using sb instead of sw instructions)
19:07:28 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
19:07:54 Join MethoS- [0] (
19:09:42*freqmod_gq sees a lot of bx'es in the thumb code and wonders if not all the firmware code is in thumb after all
19:10:24mcuelenaeregevaerts: ok, it looks to be fixed
19:10:51gevaertsI hope it doesn't cost too much in terms of speed
19:11:13mcuelenaerewell, the driver should get merged to using DMA anyway
19:11:15 Quit jhMikeS (Nick collision from services.)
19:11:21 Join jhMikeS [50] (n=jethead7@rockbox/developer/jhMikeS)
19:11:23mcuelenaereperhaps I should look into this now, as I have the tracer..
19:12:33 Quit kugel ("ChatZilla 0.9.84-rdmsoft [XULRunner]")
19:13:19 Quit lymeca ("KVIrc 3.4.0 Virgo")
19:13:33 Join lymeca [0] (
19:15:38 Quit TheSphinX^ ("XChat@Linux")
19:18:06 Join Horschti [0] (n=Horscht@xbmc/user/horscht)
19:19:53 Quit flydutch ("/* empty */")
19:22:11 Quit Thundercloud (Remote closed the connection)
19:29:59 Join bluebrother [0] (n=dom@rockbox/developer/bluebrother)
19:32:16 Join Thundercloud [0] (
19:34:13 Join Dogg006 [0] (
19:36:03 Quit Horscht (Read error: 110 (Connection timed out))
19:37:32 Quit Thundercloud (Remote closed the connection)
19:38:14BigBambiFirst attempt at the Beast installation instructions in the manual: and
19:38:34BigBambiIt will be easy to change as/when beastpatcher becomes available
19:40:03LambdaCalculus37BigBambi: 404'd. Both links.
19:40:54mcuelenaerenot here
19:40:57 Quit bmbl ("Woah!")
19:41:00BigBambier, second is
19:41:04BigBambiFirst works fine here
19:41:35 Join evilnick [0] (i=0c140464@gateway/web/ajax/
19:42:17BigBambiIt has been a while since I did any install stuff on the beast, so beast owners please check for inaccuracies as well as other comments
19:42:33 Join MethoS-- [0] (
19:45:51 Join Thundercloud [0] (
19:46:32linuxstbBigBambi: I'm just reading now...
19:46:42BigBambilinuxstb: thanks
19:47:29LambdaCalculus37bluebrother: I'm downloading the new rbutilqt image right now.
19:47:40linuxstbIt sounds like Rockbox could automatically fix the partition table - it reads the MBR at boot anyway, so could simply check if the boot flag(s) are valid, and if not, write a fixed version back.
19:47:43LambdaCalculus37Progress is slow, though... I blame JdGordon's connection. ;)
19:48:25BigBambilinuxstb: That would be ideal
19:49:42linuxstbBigBambi: You say that the OF is required, but then don't do anything with it...
19:49:54BigBambilinuxstb: I do in uninstall
19:50:36BigBambilinuxstb: I wanted to make sure that they always have a copy, otherwise we might get people not being able to uninstall if Toshiba move the update
19:51:01linuxstbAh yes, I missed the "un" in that sentence. I read it as being required for the install.
19:52:17BigBambiBetter wording is always welcome :)
19:53:29 Join countrymonkey [0] (n=4b05639a@gateway/web/cgi-irc/
19:53:57countrymonkeyWhat's the holdup on fs#s9754, 9774, 9816, 9829, 9832, 9837, 9862, ?
19:54:36LambdaCalculus37bluebrother: The new one works.
19:54:59bluebrotherLambdaCalculus37: nice. Works as in the wait-for-remount works too?
19:55:26 Join Tristan [0] (
19:55:39LambdaCalculus37bluebrother: Can't test that right now; I have neither my iPods nor my c250 right now. Just my beast and my H340.
19:55:49LambdaCalculus37I'll try it tonight and I'll mention it for the logs.
19:55:50 Quit MethoS- (Read error: 110 (Connection timed out))
19:56:30domonokybut it least i know, that my mac build enviroment is working again :-)
19:56:36LambdaCalculus37Actually, let me try something.
19:56:41bluebrotherok. I won't be around too much the next couple of days, but if it works I'd really like to commit and push a new release so it finally has that issue resolved
19:56:43linuxstbcountrymonkey: That no-one is being paid to work on Rockbox?
19:57:06bluebrotherseems like it was at least a good build setup verification ;)
19:57:10LambdaCalculus37bluebrother: My girlfriend just lent me her c240.
19:57:39LambdaCalculus37Haha... seems I did have my iPod video, as well. :)
19:57:49*LambdaCalculus37 found his video stuck in a pocket in his bag
19:58:07 Quit Dogg006 (Read error: 104 (Connection reset by peer))
20:00:27LambdaCalculus37bluebrother: What would you like me to test? Bootloader and build install?
20:01:18bluebrotherLambdaCalculus37: bootloader install. It should install as before, and then add a "waiting for OS to remount player" in the log window until it gets remounted
20:01:35 Join FlynDice [0] (
20:02:01LambdaCalculus37iPod video's up first.
20:02:21LambdaCalculus37"Could not open Ipod in R/W mode"
20:02:27bluebrotheryou can try the automated install too −− it should wait after the bootloader install until the player gets remounted.
20:02:37 Quit FlynDice (Remote closed the connection)
20:02:58linuxstbBigBambi: OS X instructions are missing. I guess for now you could share the Linux instructions - maybe call that something like "Bootloader installation from Unix (Linux, Mac OS X)".
20:03:06bluebrotherhmm, that sounds like a permission issue now ... depending on how this is handled on OS X
20:03:26BigBambilinuxstb: Yes, that was going to be a question - is OS X the same as linux?
20:03:52linuxstbI expect so - if someone can build sendfirm/beastpatcher for it...
20:04:14LambdaCalculus37bluebrother: Yep. Let me try sudo'ing rbutilqt this time.
20:04:23BigBambilinuxstb: OK
20:04:54*LambdaCalculus37 sudos rbutilqt and tries again
20:06:11*bluebrother should give the startup detection stuff a go
20:06:33LambdaCalculus37bluebrother: Still waiting for the system to remount the player...
20:06:38*linuxstb is annoyed that there doesn't seem to be an "upload file from memory buffer" function in libMTP, so he will need to write the embedded bootloader to a temporary file
20:06:48LambdaCalculus37It's stuck!
20:06:59LambdaCalculus37The player's already remounted, and rbutil just got stuck!
20:08:41bluebrotherstuck as in it still waits for the remount?
20:08:56bluebrotherthat should timeout after 5 minutes
20:08:57 Quit moos ("Rockbox rules the DAP world")
20:09:25bluebrotherthough it shouldn't get stuck.
20:11:14LambdaCalculus37It timed out.
20:11:44bluebrothercan you check if the player was mounted differently?
20:12:01LambdaCalculus37bluebrother: I think one thing that should be done on OS X to allow raw disk access is to have rbutilqt prompt the user for their password.
20:12:19LambdaCalculus37bluebrother: No, it's still in the same location as before (on mine it's /dev/disk1s2).
20:12:27linuxstbLambdaCalculus37: So it didn't work without being root?
20:12:38LambdaCalculus37linuxstb: No.
20:12:40countrymonkeyCould I please have SVN access for apps/lang?
20:12:48countrymonkeyI'd maintain Chinese.
20:12:51BigBambior with root seemingly
20:12:52linuxstbLambdaCalculus37: Can you test ipodpatcher? IIRC, that never needed root access.
20:13:04LambdaCalculus37linuxstb: Sure.
20:13:12 Quit {phoenix} (Remote closed the connection)
20:13:12LambdaCalculus37Let me finish updating my build first.
20:13:35LambdaCalculus37countrymonkey: You can send us patches through Flyspray.
20:13:52LambdaCalculus37The Tagalog maintainer does that; she has no SVN access.
20:14:05BigBambimost translators do
20:14:37LambdaCalculus37linuxstb: Doesn't work.
20:14:41BigBambiif not all (except for those that have it for other reasons)
20:14:53LambdaCalculus37linuxstb: Scratch that!
20:15:09LambdaCalculus37linuxstb: It works.
20:15:52Unhelpfulgevaerts: it's been working, although logf seems to insert two newlines per call (there are none in the format strings), and it seems to runs of characters here and there. it's not been much help debugging, but if the stack is being smashed, that might mess up logf just as it did splashf.
20:16:33mcuelenaerelinuxstb: can't you write your memory buffer to a temporary file and upload that?
20:16:51linuxstbmcuelenaere: Yes, that's what I'm annoyed about having to do...
20:16:56gevaertsUnhelpful: logf inserts \r\n for the serial line, to make it come out properly in most terminal emulators
20:17:20gevaertsUnhelpful: if you log a lot very fast, you may want to increase the buffer size in usb_serial.c
20:17:45mcuelenaeregevaerts: USB serial is acting weird here.. it seems it only displays some messages while others not
20:17:52Unhelpfuli'll try that, thanks :)
20:17:54mcuelenaereprobably the same issue as Unhelpful
20:18:28gevaertsUnhelpful: also, it seems to work better on the beast than on other players. Not sure if that helps you though
20:18:50*gevaerts welcomes all helpful eyes for usb_serial.c :)
20:19:46BigBambilinuxstb: Other than the OS X addition, anything else?
20:21:02linuxstbBigBambi: Not at the moment. Are you planning on committing it?
20:21:13mcuelenaerehmm ok, cool. now it seems to be 'stuck' transferring logf() data while the data transfer function does logf(), which results in a loop :)
20:21:36LambdaCalculus37bluebrother: iPodpatcher seems to work without needing sudo on OS X; ditto for Sansapatcher.
20:21:36gevaertsah, that's not supported I think :)
20:21:39Unhelpfulgevaerts: oh *really*? i didn't even try there, since i got data aborts trying to use logf (without serial) ages ago. also, the PF slide size is larger on beast than e200, while the plugin buffer is not... pictureflow desperately needs to grab some other buffer space to allocate from.
20:22:01BigBambilinuxstb: Probably, unless there are objections - it is missing e.g. download locations as they don't exist yet, but a) it isn't supported and b) it is better than what is there
20:22:06gevaertsUnhelpful: no promises about yesterday, but *today* it works fine on the beast :)
20:22:17BigBambilinuxstb: And it should be easier to adapt this than what is there now
20:22:31bluebrotherLambdaCalculus37: interesting. Maybe OS X treats them differently because those are non-gui?
20:22:31BigBambiWhen the final install procedure is live
20:23:28LambdaCalculus37bluebrother: That's rather unusual, though.
20:24:10LambdaCalculus37bluebrother: One easy solution is to have rbutil prompt the user for their password before attempting to do a bootloader install on the iPod or Sansa.
20:25:21bluebrotherLambdaCalculus37: I'm not sure one can elevate the permission level during runtime. Vista has something for that, but I don't know of anything like this on *nix
20:26:09 Quit stripwax (Read error: 104 (Connection reset by peer))
20:27:34 Quit Thundercloud (Remote closed the connection)
20:27:54 Join Dogg006 [0] (
20:28:09 Quit Dogg006 (Remote closed the connection)
20:28:50linuxstbLambdaCalculus37: What does "ls -l /dev/diskS*" show as the permissions for the ipod's disk device(s) ?
20:29:01 Quit faemir ("Leaving")
20:29:34*linuxstb can't believe OS X distinguishes between gui apps and console apps when accessing device nodes...
20:29:48 Quit countrymonkey ("CGI:IRC (EOF)")
20:30:09*bluebrother finds that strange too but didn't have any other idea
20:30:44linuxstbIt would be different if we were going through Apple APIs, but the ipodpatcher code just opens the disk device nodes...
20:31:54LambdaCalculus37linuxstb: On mine, the permissions are rw-r−−−−-.
20:32:04***Saving seen data "./dancer.seen"
20:32:47linuxstbWhat owner and group?
20:34:09linuxstbAnd do they change if you unmount it - i.e. with "diskutil unmount /dev/diskWHATEVER" ?
20:35:12LambdaCalculus37linuxstb: My username is the owner, and the group says "operator" (i.e. rmenes operator).
20:36:05 Quit lymeca (Read error: 110 (Connection timed out))
20:36:19LambdaCalculus37Let me unmount it and remount it.
20:36:37 Join faemir [0] (
20:37:00linuxstbLambdaCalculus37: The useful times to look at the permissions are after you've first plugged it in, and then after you unmount it (diskutil umount ...)
20:38:37LambdaCalculus37linuxstb: Nothing changes when I unmount it.
20:40:43linuxstbCan you try rbutil again? Maybe it was just co-incidence that it failed then...
20:42:45LambdaCalculus37Trying it now.
20:44:18LambdaCalculus37It seems to have successfully installed the bootloader this time without needing to be sudo'ed first.
20:44:33LambdaCalculus37But it's still timing out after waiting to remount the player.
20:46:09 Join arohtar [0] (
20:46:22 Quit arohtar (Client Quit)
20:51:31 Join petur [50] (n=petur@rockbox/developer/petur)
20:57:26 Quit MethoS-- (Read error: 60 (Operation timed out))
20:57:47 Join lymeca [0] (
20:57:52 Join MethoS-- [0] (
20:58:21LambdaCalculus37bluebrother: It's doing the same with the c240.
21:01:40 Join toffe82 [0] (
21:03:01bluebrotherLambdaCalculus37: strange. I'm using almost the same code that autodetection uses. Guess I need a mac to try myself :(
21:04:18 Join MethoS- [0] (
21:05:02LambdaCalculus37bluebrother: Ask JdGordon to hook you up with an account on his Mac. ;)
21:07:00 Quit miepchen^schlaf ()
21:09:28 Join gromit`` [0] (
21:09:45 Quit avis (Remote closed the connection)
21:10:36 Join avis [0] (n=ident@pdpc/supporter/student/avis)
21:10:42 Join miepchen^schlaf [0] (
21:16:49 Quit gromit` (Read error: 110 (Connection timed out))
21:17:58 Quit MethoS-- (Read error: 113 (No route to host))
21:26:28 Quit MethoS- (Read error: 110 (Connection timed out))
21:35:08 Quit J-23 ("ZNC -")
21:38:20 Join `Assassin [0] (
21:38:47 Quit bluebrother ("leaving")
21:39:24toffe82low_light : I am trying the last build on the hdd1630 and it doesn't seems to work...
21:40:20Unhelpfulgevaerts: um, hey, how exactly does usb logf work? do be specific, if the logf is inside a non-yielding busy loop, will my output get eaten?
21:40:40 Quit toffe82 ("ChatZilla 0.9.84 [Firefox 3.0.6/2009011913]")
21:47:10 Join keby [0] (
21:53:39 Join Annie [0] (n=43bbe4a0@gateway/web/cgi-irc/
21:54:33 Join z35 [0] (
21:54:34 Quit Annie (Client Quit)
21:56:12 Join Annie [0] (n=43bbe4a0@gateway/web/cgi-irc/
21:58:09AnnieI am trying to resolve issues with my e250 sansa because it will not recognize sdhc card and allow access through menu. will the rockbox firmware fix this? and additionally will it render my rhapsody (.rax) no longer accessible?
21:59:05krazykitrockbox will read your microSDHC card, but will not play rhapsody files. the original firmware should still play those, however
22:00:56Annieif I load the rockbox firmware, it doesn't affect the original firmware?
22:01:12LloreanIt isn't supposed to.
22:02:22LloreanThere have been one or two reports of losing Rhapsody support entirely, but they've been cases where the person wasn't careful and did not follow the installation instructions exactly, so it *may* be possible to lose Rhapsody support temporarily or permanently if you do things wrong, but we really don't have much information on this (only two people have reported it, and they never came back for us to see if we could learn more / fix it)
22:02:35LloreanIt's also possible it came back, which is why they didn't ask about it further
22:03:33 Join low_light [0] (i=ad58bb86@gateway/web/ajax/
22:04:07linuxstbLlorean: One of the reasons to write "beastpatcher" is to solve all the release hurdles - i.e. streamline and simplify the install process.
22:04:16Annieso. to clarify, the .rax files will not be impacted if I add this firmware? Original firmware will still be intact (likely) and the rockbox firmware would only augment my player capabilities by allowing it to read the slot?
22:05:01UnhelpfulLlorean: some users got it working after using dual-boot firmware. i mean to use the official updater at some point, to see if it updates the flash loader in a way that lets me duplicate the problem.
22:05:26LambdaCalculus37Annie: Yes, and then some. It'll allow you to do so much more.
22:06:04LloreanAnnie: It's a dual boot. You can only use one, or the other, at once but you can reboot to swap between them. So, for example, .rax files on the MicroSDHC would be unusable entirely without copying them to the player itself.
22:06:17 Join funman [0] (n=fun@rockbox/developer/funman)
22:06:29LambdaCalculus37funman: Welcome back!
22:07:01linuxstbWelcome back?
22:07:16linuxstbOr are you just saying hello? ;)
22:07:28funman2nd choice :/
22:07:41Anniegreat. i am not highly technically saavy so i'm trying to make sure I understand the implications. I appreciate the help. What I have gathered indicates that adding this firmware is all-in-all a good idea and will only enhance my player abilities?. correct?
22:07:41LloreanUnhelpful: Well I remember two issues, right. One was that, on occasion it'd just demand recovery mode for no apparent reason. The other is that we couldn't install after an official 1.2 update (rather than an indirect update with Rockbox install). But I never heard about recent status of either of these.
22:07:57LloreanAnnie: Yes.
22:08:00funmani'm looking for a job and an habitation, so I can't afford an internet connection nor a Sansa Clip to replace the one I broke
22:09:31AnnieLlorean, will I need to re-boot/swap while connected to computer or will i be able to do so from the player while in my car?
22:09:51LloreanFrom the player.
22:10:21Annieawesome... thank you for all your help. i will likely give this a try. have a great day!
22:11:41 Quit Annie ("CGI:IRC (EOF)")
22:11:52 Join maddler [0] (
22:16:01 Quit funman ("leaving")
22:18:06low_lighttoffe82: current svn seems to work fine on my hdd1630. What does yours do?
22:19:43 Join Rob2222 [0] (
22:22:11low_lightgevaerts: what's the status on the mr100 usb?
22:22:55UnhelpfulLlorean: for some users, at least, using dual-boot breaks the recovery-loop-on-rockbox-install cycle. random recoveries probably has to do with it not liking something about the partition table or filesystem, and won't be easy to fix unless we can get at the flash, i'd think.
22:23:11Unhelpfuland i'm not sure we want to replace the flash, since it can fix an otherwise-bricked player
22:24:36 Join LambdaCalculus3_ [0] (
22:26:12linuxstbUnhelpful: If we put Rockbox code in flash, that would have a recovery mode as well (i.e. UMS Mode)
22:32:05***Saving seen data "./dancer.seen"
22:32:41*freqmod_gq has made more unusful code experimenting with bank shifting as in the pp-driver
22:32:59 Quit LambdaCalculus37 (Read error: 110 (Connection timed out))
22:33:04LloreanDo we know how to flash the Beast?
22:33:07 Nick LambdaCalculus3_ is now known as LambdaCalculus37 (
22:33:26freqmod_gqi.e. it compiles and runs, but the actual bank shifting does not have any effect (except if you enable the #ifdef 0, then it gets a bit slower and more unsable)
22:54:20 Join toffe82 [0] (
22:54:44UnhelpfulLlorean: i don't know that we do. it would certainly solve some other issues, to have our own recovery mode, and i guess it doesn't *need* to do its own partitioning and formatting, when a host can do that :)
22:57:00toffe82low_light: ping
22:57:32low_lighttoffe82: here
22:57:42toffe82low_light : it is a 1830 and it come to a black screen but the lights of the toucpad are on
22:58:50low_lightand I assume the OF works?
22:59:56low_lightdid the 1830 work with previously with Rockbox?
22:59:57toffe82low_light : I am thining that I perhaps have a problem with svn as I made the mod for the hdd6330 before and did an update , Did the update erase previous change ?
23:01:15toffe82I have to rebuild again
23:01:38low_lightthe lcds are different between the 6330 and the 1630/1830
23:02:18low_lightjust use an svn build of the 1630 on the 1830
23:03:08amiconngevaerts: Regarding USB host - did you ever start something for the H300's ISP1362?
23:03:24toffe82rebuilding ....
23:04:37gevaertsUnhelpful: if the usb thread doesn't get to run, your logfs won't get to the other side
23:04:54 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
23:05:49Unhelpfulso, if i have a suspected busy-loop hang, i'd best put some logf *and* yield in it?
23:06:06low_lighttoffe82: got to go...hopefully it works
23:06:11 Quit low_light (" ajax IRC Client")
23:07:11gevaertslow_light: my tests on mr100 were a bit inconclusive. As long as I looked carefully, everything went fine. I tried plain svn, the PP5022 sequence and the PP5020 sequence(s) you found, with zero, two and three hubs, and I got no errors at all. Once or twice however I got serious problems after several gigabytes of data that caused linux to decide not to talk to the device any more
23:07:30gevaertsThis may or may not be related to usb. For all I know it could also be a disk issue
23:08:03gevaertsamiconn: no. I know it's OHCI, but I didn't do anything except looking at datasheets yet
23:08:13gevaertsUnhelpful: probably, yes
23:08:22amiconngevaerts: :\
23:13:32 Quit lightbulbjim (Read error: 101 (Network is unreachable))
23:14:12Unhelpfulwould something like this be generally useful? have i written something that we actually have already?
23:15:12Unhelpfulbasically, a printf-style DEBUG_OUT that prefixes function name and line, and appends a "\n" to the format if it's using DEBUGF, but not for LOGF. also an ASSERT that uses DEBUG_OUT
23:15:36amiconnLlorean: I remember that I once got that unmotivated recovery mode on my beast, but unlike for others, it didn't reformat my data partition. Maybe this is related to the fact that I don't connect the beast to linux, and hence didn't apply the partition table fix (bootable flag)?
23:16:27amiconnThis would be one reason to implement on-the-fly partition table modification. Then the on-disk version wouldn't need to be modified, and we could hide the firmware partition at the same time.
23:17:41*Unhelpful still worries a bit about handling writes to it, in that case... perhaps we return a magic one from memory if the one on disk matches the stock table?
23:18:43amiconnWriting should probably just go through
23:20:07amiconnReading would fix the bootable flag(s) on the fly if it finds the "broken" one, and hide the first partition by changing the partition type to hidden-fat32 if it finds plain fat32
23:26:33toffe82low_light : it is working , I still had the mod for the hdd6330 :)
23:31:57amiconngevaerts: Regarding USB_HANDLED_BY_OF there's now a new problem on the PP502x ipods (except G5 and nano) when we enable the usb stack.
23:32:17amiconnIt means that USB is handled within rockbox, while firewire is still handled by the OF
23:33:18 Quit maddler (Connection timed out)
23:33:23gevaertsamiconn: for now the only practical effect of having USB_HANDLED_BY_OF defined is that the rockbox upgrade detection is there
23:33:50amiconnYou probably mean "is not there"?
23:34:02gevaertsah, yes
23:34:05amiconnI wonder why this commit caused a red delta, btw
23:34:50amiconnActually a red delta for the ipods, and a green delta for clip & fuze. That can't be correct
23:35:01 Join rakslice_work_ [0] (
23:35:02gevaertsipods didn't have upgrade detection. Since the plan is to enable it soon, I thought it best to change that
23:36:08amiconnHmm. Still not really clean
23:36:26amiconnYou could have made it depend on !USE_ROCKBOX_USB
23:36:35LloreanFor 3.2, we probably want new bootloaders for any USB-getting devices that default to OF in the bootloader when a charger / USB is detected
23:37:08gevaertstrue. This needs further work
23:37:10amiconnLlorean: Ideally I'd want bootloader usb for the PP targets, but this might be problematic
23:37:33 Join Administrateur_ [0] (
23:37:57amiconnAfaik USB will require enabling interrupts, and we need to make sure the OF will still load properly, for dual boot
23:38:07 Quit Administrateur_ (Client Quit)
23:38:13gevaertsLlorean: at least for sansa that's a simple recompile with -DUSE_ROCKBOX_USB. Not sure about the others
23:38:13LloreanI seemed to recall that bootloader USB might be problematic.
23:38:22LloreanFor 3.2 we can probably settle for "don't reboot into the OF"
23:38:50amiconnLlorean: There is a problem with that though, which I observed on various targets.
23:39:24amiconnIf you connect usb during boot, and it's not caught by the bootloader so that rockbox goes straight into usb mode, the next dircache scan will be a foreground scan
23:39:41 Quit rakslice_work (Read error: 60 (Operation timed out))
23:39:46*amiconn always waits for dircache to settle before plugging usb in order to avoid this
23:40:07LloreanI don't see this as a too terribly significant problem, honestly.
23:40:16LloreanIt'd be nice to avoid, but it's still better than booting into the OF.
23:41:09 Join webguest34 [0] (n=549c7df0@gateway/web/cgi-irc/
23:42:40 Quit webguest34 (Client Quit)
23:47:02 Quit LambdaCalculus37 ("Fwump")
23:47:07 Join {phoenix} [0] (
23:51:36Unhelpfulgevaerts: erm, you said usb logf should be outputting \r\n?
23:52:33gevaertsUnhelpful: that's what logf.c has, yes
23:53:24Unhelpfulhuh. that's what i see there, as well. i was trying to use recode to conver the CRLF to LF, and it kept choking. so i hexdumped the output: 0000010: 3239 3920 6c69 7374 2064 756d 7073 0a0a 299 list dumps..
23:53:39*Unhelpful wonders what, exactly, is turning it to \n\n
23:53:52gevaertsYou could try playing with stty
23:54:42n1sspeaking of bootloaders, there's a small bug in the c200 bootloader. If it can't find the main binary it gets into a loop flashing the logo and the "file not found" message and a black screen alternating
23:54:59Unhelpfulthe large buffer fixed the missing data, by the way :)
23:55:48gevaertsgood :)
23:57:37 Join bertrik [0] (
23:57:53*domonoky tries test_disk on his e200v2, and gets strange results. The generated test file always starts with ff (and a few more wrong entrys) before it looks sensible. But a test with writing increasing numbers instead of random ones into the file, showed correct values in the first bytes. :-/
23:58:43 Quit rocko ("Leaving")
23:59:00 Part toffe82

Previous day | Next day