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

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

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

#rockbox log for 2008-04-28

00:00:08preglowi think perhaps also the struct members might be uppercase
00:00:22 Quit hannesd (Client Quit)
00:00:26 Quit FOAD ("I'll be back")
00:00:37jhMikeSI tried that but wasn't fond of it.
00:01:08preglowbut no, i agree, consistent naming of the base pointer is good
00:01:17 Join FOAD [0] (
00:02:06 Quit dabujo (Read error: 113 (No route to host))
00:02:35jhMikeS"base" when it might be one of a number of identical modules since the serial and i2c driver will need to use more than one.
00:03:29 Quit pandrew (Remote closed the connection)
00:03:48preglowwhat is debug screen code doing in the target tree?
00:06:53jhMikeSit's been there for the gigabeats for some time. I don't know what more is _not_ doing in the target tree. :)
00:08:00amiconnThe debug code shouldn't be in the target tree
00:08:38amiconn*debug menu code
00:08:51jhMikeSany suggestions on how to fix #ifdef hell there?
00:09:57Nico_Psome sort of target tree in apps/ ?
00:10:18preglowwell, the target tree is definitely not the right place for it
00:10:26preglowbut i agree that ifdef hells aren't very nice
00:10:42 Join bertrik [0] (
00:11:05 Quit ompaul (Client Quit)
00:11:16 Nick fxb is now known as fxb__ (
00:11:17 Quit jhulst ("Konversation terminated!")
00:12:19amiconnjhMikeS: The way to fix that ifdef hell is to make the hardware specific menu items superfluous...
00:12:55 Join knittl [0] (n=knittl@
00:13:13jhMikeSwhich seems like a contradiction given that it shows hardware specific information
00:13:45amiconnThe goal is that showing most of this info is no longer necessary
00:14:00amiconn...because everything works properly
00:14:03 Quit cbr|w (Read error: 110 (Connection timed out))
00:14:09*jhMikeS chuckles
00:15:21 Quit Mathiasdm ("Yuuw!")
00:28:04 Quit pixelma (" reboot")
00:30:30 Quit desowin ("KVIrc 4.0.0 Insomnia")
00:33:22 Quit gevaerts ("sleeping time")
00:34:36 Join stripwax [0] (
00:35:49 Quit Thundercloud (Remote closed the connection)
00:36:15 Quit atrus ()
00:37:51 Join stripwax_ [0] (
00:45:41 Quit herrwaldo ("Konversation terminated!")
00:53:33 Quit ender` (" The reward for work well-done is more work.")
00:54:39 Quit DerDome (Nick collision from services.)
00:54:40 Join DerDom1 [0] (
00:54:52 Nick DerDom1 is now known as DerDome (
00:55:39 Quit DerDome (Client Quit)
00:59:49 Quit einhirn_ (Read error: 104 (Connection reset by peer))
01:01:02 Quit lee-qid ("aufwiederbyebientotsayonara")
01:01:31Nico_PjhMikeS: are you still far away from audio on the S?
01:01:53 Quit NotSoCreative (Read error: 110 (Connection timed out))
01:06:41 Quit Nico_P (Remote closed the connection)
01:10:38 Join Shaid [0] (
01:10:55bertrikis there a styleguide or coding guide for rockbox?
01:11:43bertrik(maybe this is the wrong time to ask ... :) )
01:16:36bertrikI was wondering about this specifically with regard to header files
01:16:50bertrik* when to use #include <> and when to use #include ""
01:17:00bertrik* including header files within header files
01:17:25bertrik* foo.c including foo.h to make sure that foo.h and foo.c are consistent
01:18:00bertrik* which #defines go in the .h file and which ones go into the .c files
01:18:46bertrikI think I'll write a message to rockbox-dev
01:19:13*DerPapst goes too sleep
01:19:16DerPapstnight all
01:21:12 Quit gregzx ("ChatZilla 0.9.81 [Firefox]")
01:25:24***Saving seen data "./dancer.seen"
01:33:53 Join EspeonEefi [0] (i=espeonee@STRATTON-THREE-O-EIGHT.MIT.EDU)
01:38:58 Quit Shaid ("*toink*")
01:41:38 Quit DerPapst (Read error: 113 (No route to host))
01:43:53 Quit stripwax_ (Read error: 110 (Connection timed out))
01:54:49 Join midgey [0] (
01:58:41 Quit stripwax (Read error: 104 (Connection reset by peer))
02:03:35 Quit knittl (Remote closed the connection)
02:20:33jhMikeSNico_P: (for the logs) probably not too far. three things need doing 1) audio mux 2) pcm driver (we'll skip SDMA for now since that needs a program for the DMA RISC core and afaik I'd have to patch the assembler to have an assembler for it) 3) Doing the codec driver (the WM8758 is rather compatible but that needs redoing anyhow)
02:22:48 Quit bertrik ("bye")
02:25:06 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
02:28:57 Join uncfan [0] (n=473a3a91@gateway/web/cgi-irc/
02:29:24 Quit uncfan (Client Quit)
02:31:32 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
02:34:29 Join pixelma [50] (i=pixelma@rockbox/staff/pixelma)
02:35:07 Join tedrock [0] (
02:35:09 Quit tessarakt ("Client exiting")
02:35:22saratogajhMikeS: how close is sound on the gigabeast?
02:38:31pixelmaI think that question was answered about 15 minutes ago (IIUC, reading logs too)
02:38:37jhMikeSyes, logs :)
02:42:43 Quit perrikwp (" ajax IRC Client")
03:00:17 Quit jhulst ("Konversation terminated!")
03:00:38 Quit csc` ("If you can't do anything smart, do somethin right ~Serenity")
03:15:50 Part pixelma
03:15:57 Join axlgreasetires [0] (n=473ce4bd@gateway/web/cgi-irc/
03:25:26***Saving seen data "./dancer.seen"
03:27:43 Join alienbiker99 [0] (
03:29:25 Quit alienbiker99 (Read error: 104 (Connection reset by peer))
03:36:26axlgreasetiresnothing happenin
03:37:13 Quit axlgreasetires ("CGI:IRC")
03:51:44 Quit saratoga ("CGI:IRC (EOF)")
04:02:56 Join sarixe [0] (
04:03:09 Join Lynx- [0] (
04:05:25 Join alienbiker99 [0] (
04:08:35 Join cool_walking_ [0] (
04:09:17*JdGordon has no idea why that last warning is still there for the mr500
04:09:30JdGordoneither the sim and the main build should have it, or neither!
04:11:29alienbiker99hmm. i seem to keep getting this error abut FPA instructions in cygwin when im building
04:20:39 Quit Lynx_ (Read error: 110 (Connection timed out))
04:20:39 Nick Lynx- is now known as Lynx_ (
04:21:50 Quit BHSPitLappy (Remote closed the connection)
04:28:49 Quit Llorean ("Leaving.")
04:29:13 Join Llorean [0] (
04:31:26 Quit EspeonEefi ("さよなら")
04:37:20 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
04:40:14 Quit alienbiker99 ("( :: NoNameScript 4.02 :: )")
04:41:15 Join miepchen^schlaf [0] (
04:44:30 Quit sarixe ("(EE) Failed to load "quit" module")
04:47:36 Join EspeonEefi [0] (i=espeonee@STRATTON-THREE-O-EIGHT.MIT.EDU)
04:54:30 Quit miepchen^schlaf_ (Connection timed out)
05:07:54 Quit SUSaiyan (Read error: 104 (Connection reset by peer))
05:13:43 Join SUSaiyan [0] (
05:22:16 Quit leox ("La fe mueve montañas... de pelotudos")
05:22:36 Join leox [0] (
05:25:30***Saving seen data "./dancer.seen"
05:27:00 Join csc` [0] (n=csc@archlinux/user/csc)
05:31:48purpleposeidonRight, I got a sansa e250
05:38:10 Quit Horscht ("I am root. If you see me laughing, you better have a backup")
05:43:59 Quit csc` ("If you can't do anything smart, do somethin right ~Serenity")
05:45:26 Join Shaid [0] (
05:49:34 Join toffe82 [0] (n=chatzill@
05:54:34jhMikeStoffe82: got the board a couple days ago, thanks.
05:55:37*toffe82 is stuck in mexico :(
05:55:57Shaidhe forgot his torch
05:56:06Shaidand now there are grue circling.
05:56:43toffe82you don't say stuck when you are somwhere and can go away ? I have to learn english :)
05:57:37jhMikeSif you _can't_ go away for some reason. hopefully it's not some passport/INS trouble. :O
05:58:26soaptransporting fruit across the California border is a serious crime!
05:58:34toffe82passport could be a problem , I will see on tuesday
05:58:36jhMikeSeither way?
05:58:55toffe82I don't have any visa on my passport :)
05:59:22jhMikeSso they think you're an illegal mexican trying into the US and take 'er jerbs?
05:59:50Shaidyew curnt haf my jerb!
05:59:55Shaidactually, you can take it
06:00:07*scorche coughs
06:00:29*jhMikeS hands scorche a Ricola
06:00:35toffe82jhMikeS: exactly , sometimes it is easier to be illegal tnah tryimg to be legal
06:00:46ShaidWe're off-topic, we should shush.
06:01:21toffe82jhMikeS: did you work on the board or not yet ?
06:02:30jhMikeStoffe82: yes. verified the known connection. I wasn't also to get a trace on the audio outputs though so had to g
06:02:35jhMikeSgo "pro"
06:03:57jhMikeSonly got so far (up to the LDO itself) and couldn't find it even with an ohmmeter :\
06:04:47toffe82I had a lot of problem tracing on this board, I don't know why the contact are so bad...
06:06:15 Quit leox (Connection timed out)
06:06:20 Join toffe82_ [0] (n=chatzill@
06:06:30 Quit toffe82 (Read error: 104 (Connection reset by peer))
06:06:30 Nick toffe82_ is now known as toffe82 (n=chatzill@
06:06:54jhMikeSA few traces around some components in the output area were broken along with a few holes bored in some pads :) so that could have broken something.
06:08:08toffe82When I was looking for the mute, I broke one trace I think, but after I think it is just missinge some resistor or cap perhpas
06:10:51 Join goffa [0] (n=goffa@
06:11:43jhMikeSI know a GPIO line to carry a DC signal can't have a series cap and if the resistor is missing I just test both sides of its pads. I was still able to double check things against the block diagrams and see that having the analogue enabled wasn't important to using the i2c.
06:12:47 Quit Zom (Remote closed the connection)
06:12:58 Join Zom [0] (
06:14:05jhMikeSin short it let me know I'd better double-check my code
06:15:18 Quit perrikwp (" ajax IRC Client")
06:15:40 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
06:16:24 Quit miepchen^schlaf ()
06:16:29jhMikeSand it all came down to reversing the r/w address bit in the i2c driver :P
06:19:50 Quit toffe82 (Read error: 104 (Connection reset by peer))
06:21:14 Join sarixe [0] (
06:27:55 Quit sarixe ("(EE) Failed to load "quit" module")
06:35:53 Quit Llorean ("Leaving.")
06:36:08 Join Llorean [0] (
06:36:20 Quit perrikwp (" ajax IRC Client")
06:42:05 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
06:49:24 Quit Shaid ("*toink*")
06:51:38 Join DerPapst [0] (
06:51:41 Join leox [0] (
06:56:35purpleposeidonRockbox is HELLA AWESOME. The only bug is that if you plug it into USB without powering it down, it freezes and you have to hold the power button down for like 5 seconds
06:58:03purpleposeidonAnd something funny happened while I was trying to record, it froze, but I could still hear sound
06:58:48JdGordonif they are the only bugs you've found your doing well :)
07:01:53purpleposeidonOh, no. But those were the more spectacular ones
07:02:37krazykitpurpleposeidon, well, those are target-specific, it sounds like. i'd guess you're using a sansa?
07:02:52purpleposeidonyez. Ai juzt gaut it 2dai! :D
07:03:21*DerPapst has a small bug-farm on his daps. Too bad the most don't show up often....
07:03:28scorchewe like english in here...
07:03:41krazykitpurpleposeidon, please review the IrcGuidelines in the topic. writing like that is not only illegible, it burns the eyes.
07:03:53scorcheand our goggles dont do anything
07:07:30 Join advcomp2019 [0] (n=advcomp2@unaffiliated/advcomp2019)
07:18:27goffawhee.. updated my h140 and x5l... havent done that in like a year
07:18:31goffathink i have the h140 sold
07:25:32***Saving seen data "./dancer.seen"
07:30:51 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
07:35:34 Join pabs_ [0] (
07:35:40markunlinuxstb: I didn't fix the UIE problem yet and defined a dummy function in my bootloader file to get on with the rest. Did you fix it?
07:37:24 Join DerDome [0] (
07:40:01 Quit pabs (Read error: 110 (Connection timed out))
07:40:04 Join neptunepink [0] (
07:40:50 Quit purpleposeidon (Nick collision from services.)
07:40:52 Nick neptunepink is now known as purpleposeidon (
07:53:20 Quit DerPapst (Read error: 113 (No route to host))
07:57:00Dhraakellianheh... another new Sansa user... Must be the Woot.
07:57:15 Quit J (Read error: 110 (Connection timed out))
07:57:34 Quit midgey ()
08:11:36 Quit JdGordon (Remote closed the connection)
08:13:27 Join JdGordon [0] (
08:28:14 Join Thundercloud [0] (
08:30:27 Quit BigBambi (Remote closed the connection)
08:30:48linuxstbmarkun: Not really, but I think the problem is that system.o isn't being linked with the bootloader.
08:30:56 Join Bagderr- [0] (n=daniel@gateway/web/cgi-irc/
08:31:15linuxstbMy "fix" for now is to add a call to UIE() in my bootloader (at the very end, which isn't executed).
08:31:22 Nick Bagderr- is now known as B4gder (n=daniel@gateway/web/cgi-irc/
08:32:28*linuxstb now has a tick interrupt working (very quickly...) on the M260 and DAX
08:33:43amiconnI don't get that. It shouldn't matter whether it's called from C or from asm
08:33:57 Join GodEater [0] (i=c2cbc962@rockbox/staff/GodEater)
08:34:54amiconnPerhaps it needs to be declared in crt0.S
08:35:49amiconn.extern UIE should do that
08:35:50linuxstbHere's my (work in progress) crt0.S -
08:35:55linuxstbI'll try that.
08:37:02amiconnYou have .global UIE in that file
08:37:38amiconnSame for irq and fiq
08:37:48linuxstbIs that wrong? Other crt0.S files have it.
08:37:50amiconnThat can't work exports a symbol from that asm file
08:38:49linuxstbThat's what I thought, but removing it made no difference.
08:39:08amiconnSo the linker thinks that crt0.S defines UIE, and doesn't even try to link system.o if that's not needed for other reasons
08:39:50linuxstbDeleting those .global lines makes no difference though.
08:40:40linuxstb(and adding .extern UIE)
08:40:51 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
08:41:47purpleposeidonKen Perlin has some interesting java simulations applets for input with few buttons.
08:42:05amiconnThat's odd. Also, why does it work for main() then?
08:44:26 Quit DerDome (Read error: 110 (Connection timed out))
08:47:11amiconnThe gas manual says that .extern isn't necessary (but I'd still use it to indicate what external symbols the asm code needs
08:47:23 Quit JdGordon ("Konversation terminated!")
08:47:52 Join JdGordon [0] ( is definitely wrong there though
08:48:45*amiconn would like to see that effect himself
08:49:08 Join Rob2222 [0] (
08:49:28 Join ender` [0] (
08:53:17linuxstbamiconn: If you add the line "bl UIE" to crt0-pp-bl.S and then try and build an ipod bootloader, you get the same issue. UIE is compiled in system.o
08:54:50 Join bertrik [0] (
08:56:37 Quit Slasheri (Read error: 145 (Connection timed out))
08:56:58 Join davina [0] (
08:59:16jhMikeSlinuxstb: if it's what I think it is, see the bootloader/gigabeat-s.c for a solution (only way I could make it work)
09:01:29amiconnThat looks like an utter hack, and is iiuc a similar solution to what linuxstb did. I'd like to find out how to do this properly
09:01:34linuxstbHmm, that's more or less the same as my "fix"...
09:02:08amiconnFor the linker it shouldn't matter at all whether a function is called from C, asm, or both
09:03:13jhMikeSIt seems if c never references the file, then it's never seen as referenced unless there's some way to force a symbol reference
09:03:39amiconnThe asm references the symbol
09:04:02jhMikeSdoesn't seem to count since the compiler doesn't see that
09:04:11amiconnSince you can link all sorts of objects (C, asm, pascal, fortran, whatever)
09:04:13linuxstbBut it's a linker issue, not a compiler issue. shouldn't matter
09:04:29jhMikeSwhat generates the dependencies?
09:06:42 Quit Rob2223 (Read error: 110 (Connection timed out))
09:10:12 Join petur [50] (n=petur@rockbox/developer/petur)
09:10:32amiconnThe linker
09:18:04 Join nplus [0] (
09:19:22*amiconn has a suspicion
09:25:14markunamiconn: what is it, or do you want to verify it first?
09:25:34***Saving seen data "./dancer.seen"
09:26:43 Quit fxb__ ("changing servers")
09:27:06 Join fxb__ [0] (
09:27:52jhMikeSamiconn: cough it up :) I don't really like that hack either.
09:30:21 Join Squirm [0] (
09:38:00 Quit Squirm ("Ex-Chat")
09:51:33 Quit jhulst ("Konversation terminated!")
09:58:48 Join J [0] (
10:00:38 Quit GodEater (" ajax IRC Client")
10:04:11 Join CaptainSquid [0] (
10:21:08 Join Slasheri [0] (i=miipekk@rockbox/developer/Slasheri)
10:25:35JdGordonBagder: think its worth fixing the regex on the web site to ignore mr500 o it doesnt tihnk its a revision link?
10:25:36 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
10:26:14 Join austriancoder [0] (n=austrian@rockbox/developer/austriancoder)
10:31:44B4gderyeah, I'll look into that later on
10:36:17markunB4gder: do you think it's too early to add support for big-endian arm in our
10:37:08 Join pixelma [50] (i=pixelma@rockbox/staff/pixelma)
10:37:51*linuxstb threatens to add another new target to the build system....
10:38:08B4gder38 isn't enough!
10:38:15B4gderor whatever the current amount is
10:38:25markunB4gder: this patch does the trick:
10:38:29linuxstbI'm more concerned about the 3 unfinished ones I have there...
10:39:30B4gdermarkun: what are you gonna name the toolchain? armbe-elf-gcc?
10:40:49markunB4gder: no, it's still arm-elf-cc but with -mbig-endian
10:40:50linuxstbIt needs a separate compiler?
10:41:22B4gderso we'll need to rebuild all build servers arm chains before it gets to the build table later on
10:41:47markunI added a arm940tbecc to tools/configure
10:42:13markunyes, and they shouldn't reuse the old patch, so maybe we should give it another name
10:42:19linuxstbI think some _ characters would help there...
10:42:39linuxstbi.e. arm940t_be_cc
10:42:46markunyes, looks better
10:43:06linuxstbOr even s/be/bigendian/
10:43:07markunbtw, should it be called arm9tdmi_be_cc?
10:43:08B4gdermarkun: yes, use a different name to prevent accidents
10:43:56markunB4gder: I can't upload the new patch, or can I?
10:44:21LinusNgah, does this mean i have to build new cygwin compilers? :-(
10:44:34markunLinusN: only for meizu users :)
10:44:39 Quit spiorf (Remote closed the connection)
10:45:00markunand maybe we can find a way to switch to little-endian mode if that makes things easier
10:45:49B4gdermarkun: you can, in the gcc dir in the www part of svn
10:46:30B4gdermarkun: if the OF runs be, I bet all peripherals etc are made for it so it'll be awkward to switch
10:47:17markunB4gder: the coprosessor (calmrisc) is also be
10:47:48*gevaerts looks around and seens that a lot of ata.c people are awake...
10:49:15gevaertsCould someone with ata.c knowledge have a look at FS #8943 ?
10:52:09gevaertsThe issue is that ata.c does a blocking wait for usb disconnect, which makes sleep timeout handling not work any more, so the disk keeps spinning
10:52:54 Join GodEater [0] (i=c2cbc962@rockbox/staff/GodEater)
10:53:56 Quit nplus (Read error: 101 (Network is unreachable))
10:54:59amiconngevaerts: That blocking is necessary on devices with hardware usb-ata bridge, but of course not for software usb
10:55:23preglowmarkun: all other arm targets are little endian, so that would probably give the least amount of surprises
10:55:34gevaertsamiconn: the thing is that I don't understand ata.c enough to know if it's safe to just disable it for software usb
10:56:57markunpreglow: what would?
10:57:08preglowmarkun: switching it to little endian
10:57:47markunwell, maybe, but for now I'm going to work on it as a big-endian target because that's easier
10:58:24 Quit cool_walking_ (Remote closed the connection)
10:58:29amiconnThe thread is quite easy to understand. Just take care to not call ata_idle_notifys if usb is active (those might write to the disk)
10:59:10amiconnBut I don't think that not spinning down is a real issue
10:59:56peturit is if you want to charge...
10:59:59gevaertsIt is as long as we don't get charging right. With older batteries boost+spinning disk is enough to make it not charge
11:00:38preglowmarkun: switching to little endian shouldn't be hard, what kind of an arm core is it?
11:00:52gevaertsI'l try to do a patch tonight.
11:00:59amiconnMany usb-ata bridges don't spin down the disk. It's a problem of our charging handling, not of the ata thread
11:01:19*gevaerts doesn't disagree
11:01:27preglowhmm, then it doesn't have the setend instruction, no
11:02:42markunand how do we compile the first few instructions big-endian to switch to little-endian mode?
11:03:13amiconnmarkun: If it's really only a few, I'd just use .word <hex code>
11:04:06markunamiconn: I guess it's just the first jump from the vector table and the instruction after the jump to switch
11:05:41 Part linuxstb ("Leaving")
11:07:10 Join Manoj` [0] (
11:07:17markunaustriancoder: wouldn't it be simpler to just define AUDIOHW_HAVE_TREBLE etc in the header files of the codecs?
11:08:06markunor what do we gain by this extra AUDIOHW_CAPS step?
11:08:13Manoj`hi giuys. I installed from the 23rd on my ipod video 60GB, and now rockbox loads −− and does nothing
11:08:34Manoj`I can't even shut it down, I have to wait until the battery runs out
11:09:05markunI think the reset combination should always work
11:09:24Manoj`I can type in the information it presents when booting
11:10:27markunmaybe it's better to reset, enter USB mode and try today's build. But I'll wait for someone with an ipod to help you.
11:10:33Manoj`Rockbox boot loader; Version 20060126-1839
11:10:49markun2006, wow
11:10:55markunah, bootloader :)
11:11:04pixelmalooks like you need to update the bootloader then
11:12:29Manoj`I am pressin menu and select, which should be the reset
11:13:05Manoj`The thing is, tis is with the USB cable connected (I had to connect the USB cable for the ipod to charge)
11:13:25 Join Zom_ [0] (
11:13:42Manoj`So, it initially shows just the battery icon, and then rockbox boots −− but never goes into USB mode
11:13:43JdGordonBagder: the ifp label is still in the build table..
11:14:03B4gderyes, as always when builds are removed! ;-)
11:14:10austriancodermarkun: its just an one-liner to define CAPS.. looks nicer then defining 2-n defines.. I have "stolen" the idead of caps from /firmware/export/pcm_sampr.h
11:14:12B4gdersince they're still around, not just shown in the table
11:14:12*JdGordon congratulates everyone on again making the table almost too wide for my lcd :p
11:14:16B4gderit'll die in a few days
11:14:19JdGordonah ok
11:14:48B4gderthe table is built "in memory" to be all builds "high", but just the 20 top-post rows are shown
11:15:20B4gderall builds being the last week or so
11:15:53gevaertsManoj`: when you manage to reset it again, hold select+play to get into disk mode and let it charge for a while first
11:16:14pixelmaManoj`: for entering disk mode you need to press select+play...
11:16:25Manoj`gevaerts: I am hoping I can get it to reset
11:16:57markunaustriancoder: yes, maybe if you add a lot more capabilities it becomes nicer
11:17:36Manoj`aha!. I did manage to get it to reset −− and hitting select plus play before it rebooted worked!
11:18:29 Quit Zom (Read error: 110 (Connection timed out))
11:19:31Manoj`now that I have it in USB mode −− should I upgrade the bootloader?
11:21:05pixelmathe installation instructions in the manual also explain how to update the bootloader
11:25:38***Saving seen data "./dancer.seen"
11:31:46Manoj`gah. rbutilqt can't find
11:32:09Manoj` has address
11:32:17Manoj`so dns is indeed working
11:35:13peturmaybe one of the servers is down?
11:35:21 Part Miles
11:36:33B4gderManoj`: does it say specifically that isn't found?
11:37:15gevaertsall three download.rockbox.orgs seem to respond
11:37:26Manoj`it does say that
11:38:17Manoj`"Network error: Host not found. Please check your network and proxy settings."
11:38:28Manoj`I have the proxy set to "no proxy"
11:38:58Manoj`Hmm. maybe it does not like 64bit machines?
11:39:56gevaertsYou could try the "manual" method (i.e. using ipodpatcher).
11:45:11Manoj`[INFO] Using internal bootloader - 50136 bytes
11:45:11Manoj`[1] 4542 segmentation fault sudo ./ipodpatcher
11:45:18Manoj`This does not bode well.
11:45:54B4gderthat the 64bit version?
11:47:32 Join eightfold_ [0] (
11:47:41 Part eightfold_
11:47:43Manoj`actually, even the 32bit ipodpatcher seems to fail
11:48:12Manoj`They both scanned cprrectly, found the ipod −−- and failed to install
11:49:46 Join eightfold [0] (
11:50:13soapamiconn, I know the Gigabeat F doesn't spin down the disk while in USB mode either, but mine gets quite hot when left for hours plugged into USB, Ditto (heat) with the way the iPods currently behave with the Rockbox USB stack. Is that safe (for the hardware?)
11:50:20eightfoldi'm about to install rockbox an ipod mini. the color of the arrows on the click wheel is gray
11:50:28eightfoldthat means it's first gen, right+
11:50:41soapeightfold, what color is the mini?
11:50:50eightfoldsoap: blue
11:51:06gevaertsManoj`: does it give an error message ?
11:51:10soapThen I'd say that's a pretty good indication it is a first gen, or a Frankenstein.
11:51:19eightfoldisn't there a way to check with serial number?
11:51:29eightfoldsoap: what's a frankenstein?
11:51:51soapeightfold, A mismatched motherboard and case/buttons.
11:52:16eightfoldsoap: apple themselves did that?
11:52:25Manoj`[1] 4679 segmentation fault ./ipodpatcher
11:52:29soapNo, but eBay sellers do all the time.
11:52:36Manoj`even strace does not help much
11:52:44eightfoldsoap: ah, but this is new, out of the box
11:52:54soapeightfold, Instead of looking at the back with your reading glasses on trying to copy down the serial number, take a quick glance at the back and see if the capacity is etched on the back of the player - if so it is a 2nd gen.
11:53:26Manoj`open("/dev/sde", O_RDWR) = 7
11:53:29soapeightfold, New? Is this a mini or a Nano?
11:53:49Manoj`then it writes it is going to use the internal boot loader −− and then boom, Segmentation fault
11:53:50eightfoldsoap: what did you just say? :) what should i check for? if the serial is on the back?
11:53:58eightfoldeightfold: new as in unused
11:54:04eightfoldeightfold: a mini
11:54:21pixelmaeightfold: I believe another indicator is if the it has the storage engraved on the back or not
11:54:36eightfoldmodel #
11:55:04soaptake a quick glance at the back and see if the capacity is etched on the back of the player - if so it is a 2nd gen
11:55:52soapI was (trying to) say that instead of looking up a serial number you can simply look at the back, check for the capacity etching, and know if it is a 1st or 2nd gen. (2nd gen has capacity etched on back)
11:58:31preglowamiconn: i just confirmed the template struct approach shaves off bytes like mad in places with heavy access
11:58:47preglowat least on arm
11:59:36amiconnsoap: I would think so, because the disk obviously cannot spin down when it's accessed continuously.
11:59:54eightfoldsoap: no size etched
11:59:58amiconnpreglow: I know, that's the part I didn't uqestion. My concern was about readability
12:00:18preglowamiconn: i think the readability matter can be seen to
12:00:38eightfoldthen so, it's first gen
12:00:52preglowi wonder how the compiler would handle template structs in places with just a single access
12:00:58preglowit might end up bigger if it's stupid
12:01:34soapamiconn, but in a typical "mounted as storage" situation a drive is not accessed continuously. I totally grok your comments regarding the charging justification for spindown, but isn't heat production and drive wear justification enough to spin down the drive?
12:01:46*gevaerts looks around to see if rbutil or ipodpatcher specialists are around
12:02:16eightfoldthe manual here says 2g
12:02:31eightfoldin the file name: rockbox-ipodmini2g.pdf
12:02:31*Llorean saw an ipodpatcher expert make a statement about an hour and 20 ago.
12:02:42eightfoldbut it should apply to 1st gen too?
12:03:18*gevaerts saw that ipodpatcher expert leave 20 minutes after that...
12:03:56LloreanWhich is unfortunate
12:04:06LloreanBecause I'm pretty sure he's the only one really "expert" on the matter.
12:04:35*Llorean assumes the rbutil expert is expected to be one who's familiar with the ipodpatcher bits of the whole.
12:05:01 Quit GodEater (" ajax IRC Client")
12:05:18preglowamiconn: i kind of do think that having members be uppercase would fix the readability, after all, that's the only visual clue we currently have to spot these io registers
12:05:28gevaertsDepends. rbutil couldn't find in this case...
12:05:32 Nick Zom_ is now known as Zom (
12:07:44LloreanWell, at least I can confirm it finds fine for me.
12:08:53gevaertseightfold: mini 1st and 2nd gen are very similar.
12:09:35eightfoldgevaerts: ok, thanks
12:16:15 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
12:16:26eightfoldshould i do this installation thingy in windows or os x? the rbutil verisons don't seem as fresh for os x.
12:18:53gevaertseightfold: I think 1.0.4 is good enough, but I'm not entirely sure
12:20:49gevaertsManoj`: maybe an older rbutilqt could work. Proxy and url handling changed a bit between 1.0.4 and 1.0.5. Otherwise I really have no idea
12:21:48gevaertseightfold: on second thoughts, maybe 1.0.5 is the way to go for a full install...
12:22:25Manoj`gevaerts: I'll try 1.0.4
12:22:44Manoj`the thing is, the commandlin stuff also failed
12:22:52Manoj`which I don't like
12:23:08Manoj`perhaps I should borrow a windws machine from somewhare and try that
12:23:31Manoj`At this point, I am just happy that my ipod is not a brick ...
12:25:23 Join DerDome [0] (
12:30:03eightfoldok, i just ran the "complete installation" in windows (rockbox utility 1.0.5)
12:30:27eightfoldthat should be it? must say you've done an excellent job with the installer
12:30:55gevaertsThat should be it, yes
12:31:21Manoj`OK. I ran it in a 32bit chroot, and it can find the network
12:31:26Manoj`so it is a 64bit issue
12:31:41Manoj`[INFO] Scanning disk devices...
12:31:43Manoj`[INFO] Ipod found - Video (aka 5th Generation) ("winpod") - /dev/sdf
12:31:43eightfoldso, now i should jump to 2.4 in the manual (running rockbox) disconnect the device and press menu + select?
12:31:55Manoj`[INFO] Checksum OK in /tmp//qt_temp.X12588
12:31:56Manoj`[ERR] Input file too big for buffer
12:32:05eightfoldwith select you mean the button without an icon in the middle (i've never used an ipod extensively before)
12:33:32gevaertseightfold: yes
12:34:03pixelmaeightfold: there's a picture which explain the buttons at the top of section 3
12:35:45gevaertsManoj`: there's unfortunately not much we can do until the right people show up
12:36:14eightfoldseems i hadn't turned off the ipod
12:36:25eightfoldand held menu + select
12:36:28Manoj`gevaerts: I understand
12:36:55eightfoldholding play means shutting off the player, right?
12:36:56Manoj`gevaerts: and believe me, I am grateful for the work the rockbox community has put together
12:37:25Manoj`but it is so frustrating ... so close, and yet −− :-)
12:38:16Manoj`It downloads the bootloader-ipodvideo,ipod file
12:38:29Manoj`the checksum is the same every time it downloads it
12:38:41Manoj`it is prbably trying to dd it to some place
12:38:55eightfoldok, holding the menu + select made the screen show the apple and flash. then it started the regular ipod firmware
12:39:01eightfoldshould i try again?
12:39:27Manoj`eightfold: that is the point when it should have started rockbox ...
12:39:34eightfoldManoj`: mmm
12:39:39eightfoldobviously it didn't
12:39:51eightfoldwould it hurt trying again before doing anything special?
12:40:11bertrikhmmm, the build page is not updating it seems
12:40:52eightfoldit seems i can't shut the player down
12:41:05eightfoldand i didn't do that before holding menu + select the first time either
12:41:09gevaertsManoj`: (starting on real guesswork now...) One thing you could try is first run ipodpatcher -d (delete the old bootloader). Maybe that will work, and clear everything up enough to make installation work
12:41:17eightfoldholding play lights down the display
12:41:31eightfoldbut pressing menu makes the menu show up _instantly_
12:41:43eightfoldit shouldn't do that if is completely off, right?
12:42:47Manoj`gevaerts: I'll try. But this is 05:42am, I better get to bed
12:42:55Manoj`gevaerts: thanks
12:43:45eightfoldshould the startup of the mini be instant?
12:43:49eightfoldif it's completely off?
12:44:08eightfoldi press the menu button from what i believe is the "off state" and the menu shows up directly
12:45:33 Join ctaylorr [0] (
12:46:03eightfoldpressing menu + select just lights up the screen and activates the menu
12:46:59gevaertseightfold: the apple firmware never really shuts down.
12:47:11eightfoldoh, i'm running rockbox now :)
12:47:18gevaertsah, ok :)
12:47:34eightfoldso happy i am right now :)
12:47:58eightfoldbig thank you to to the rockbox community!
12:48:03gevaertsto turn it off, you have to press play for a onger time (5 seconds or so)
12:48:11gevaertsIt will tell you it's shutting down
12:48:12eightfoldoki doke
12:49:41eightfoldand now i should be able to to use it like a regular usb mass storage device? connecting to os x or windows and just move to it?
12:51:07gevaertsWell, if you plug it in, it will reboot to Emergency Disk Mode for usb mass storage (still working on native rockbox usb...), but yes, you just copy your files to anywhere you want on it
12:51:36preglowamiconn: hmm, assembler files can use defines, but not structs. i guess that's a negative side of the template structs as well
12:53:07 Join austriancoder_ [0] (n=austrian@
12:59:32 Join GodEater [0] (i=c2cbc962@rockbox/staff/GodEater)
13:07:30 Quit austriancoder (Read error: 110 (Connection timed out))
13:07:57eightfoldhey, i love this! enlightenment! rockbox is just wonderful!
13:17:06 Join nplus [0] (n=npl@141.25.Globcom.Net)
13:17:18 Nick fxb__ is now known as fxb (
13:25:39***Saving seen data "./dancer.seen"
13:29:28markuneightfold: great to hear that you like it
13:35:23 Quit ctaylorr (Read error: 110 (Connection timed out))
13:35:33eightfoldi've heard mixed success reports of people changing their ipod mini hdds to larger flash based storage
13:35:47eightfoldthis mini seems to have only 2 gb
13:35:51eightfoldwhich leaves me asking for more
13:36:51 Join J-23 [0] (
13:39:08 Quit austriancoder_ (Remote closed the connection)
13:45:40 Quit eightfold ("thanks for everything")
13:54:32B4gderdid any gigabeat user try that austriancoder wm8751 commit?
13:54:50B4gderit seems so strange to me that it can just divide with 10
13:55:06*petur wondered about that too
13:55:32B4gderand the if below the calculation just _have_ to be wrong now
13:56:13LloreanWhat is that the value of? I have a gigabeat on-hand and can test if I know what I should be looking for.
13:56:22*Llorean rarely, if ever, uses any sort of tone control
13:57:00B4gderused in bass and treble settings
13:58:36markunbass and trebble as in steps of 1.5dB
13:58:50LloreanSo I should notice a different range in what I hear?
14:00:30LloreanYeah, +9 bass has almost no effect with current SVN rockbox
14:01:34B4gderyeah, as I read it the values should only be a 10th of what they used to be
14:02:52LloreanI'd assumed they would be from the diff, but I hadn't read through the whole file so had no clue where it's actually used, or if it relates to other changed files
14:03:33preglowthose 1.5 db steps don't work right with rockbox right now
14:03:46preglowi think i did a patch that supported them. god knows what happened to it
14:04:24Lloreanpreglow: The 1.5 steps of bass/treble?
14:04:34preglowrockbox pretends the steps are 1 db
14:04:36 Quit nplus ("Leaving")
14:04:44LloreanIt calls them 1.5 in the menu
14:04:46 Join nplus [0] (n=npl@141.25.Globcom.Net)
14:04:48preglowreally, now
14:04:51preglowthen something has changed
14:04:57Llorean0, 1.5, 3.0, etc
14:05:04*amiconn would just stick with 1dB steps, and then round, like we do for volume on Player
14:05:18preglowi wouldn't, it feels inaccurate for the user
14:05:36LloreanBagder: Removing the /10 doesn't seem to have fixed the problem for me.
14:06:02preglowthere will be no difference between settings for 1 and 2 db, for example
14:06:33 Quit Seed ("cu, Andre")
14:06:58 Join Seed [0] (
14:09:05 Quit perrikwp (" ajax IRC Client")
14:09:37amiconnpreglow: I know, but it's simple, and if you want to only expose the actual physical steps, it will become really nasty in the mentioned example
14:10:16 Join miepchen^schlaf [0] (
14:10:21amiconn..because the DAC3550 goes in 1.5dB steps from +18 to -54 dB, and then in 3 dB steps down to -75 dB (-78 is mute)
14:11:04amiconnNow try to represent that in a generic way, including prescaling correction for treble/bass...
14:11:23preglowbut when it's more consistent, like for wm codecs, i think perhaps we could use the native step
14:11:30amiconnTreble and bass are in 1 dB steps, so there's an inherent impreciseness anyway
14:23:17pixelmabertrik: late reply but it happens from time to time that the build script doesn't pick up a commit
14:23:30pixelma(for a next build round)
14:24:01bertrikOk. Does it fix itself on the commit after that?
14:24:27B4gderyes, it is in svn so the next build will use that
14:27:54 Nick JdGordon is now known as JdGordon|zzz (i=jonno@rockbox/developer/JdGordon)
14:38:10*Llorean thinks austriancoder's latest commit should've been tested *first* rather than saying "looking for testers" in the commit message. :(
14:40:14B4gderat least when combined to a general absence and no other questions put in big letters somewhere where people can provide feedback to him
14:40:46LloreanYeah, at the very least a note to the dev mailing list saying "guys, i've been working on this, but don't have the hardware to test this bit..." would've been nice.
14:40:51*Llorean thinks the -dev mailing list is underused.
14:40:58B4gderI agree
14:41:20pixelmahe posted it to the tracker and asked for opinion on it yesterday
14:41:32pixelmanot that it makes it much better
14:45:00 Join kugel [0] (n=kugel@unaffiliated/kugel)
14:45:10kugelLlorean: ping
14:46:33kugelLlorean: I really liked to tell you the precice status, but I can't since jott doesn't answer me. The only thing that I can tell is, that he gave me permission to include it into my build (and he was aware that I distribute it)
14:48:20LloreanBut was he aware that doing so means it MUST be GPL licensed?
14:48:29LloreanAs it stands, you have no explicit license for the code, just a "go ahead and use it"?
14:49:08kugelLlorean: He was aware of that too. Also, his font converter is GPL
14:50:09LloreanHis other program being GPL doesn't say anything about this code.
14:50:23LloreanWhy was chrisjs concerned it might not be GPL'ed?
14:51:46kugelDon't know, but probably since he never specifically said "my AA font patch is GPL"
14:53:12LloreanThe problem is you can't really implicitly license code.
14:53:43kugelIs "no license" GPL compatible?
14:53:50LloreanNot in the slightest.
14:53:56Llorean"no license" means "all rights reserved"
14:54:14LloreanRight now you've got "some rights reserved." He gave you permission to use the code, but not explicitly to GPL it, which is pretty iffy.
14:54:49markunif it a patch against existing GPL code?
14:54:56markunif -> is
14:55:09Lloreanmarkun: You can patch against GPL and still not GPL your patch, though.
14:55:14LloreanYou just can't legally distribute it.
14:55:22kugelI must GPL his patch, if I include it. And he was aware of that
14:55:23markunhow? by not distributing it?
14:56:02LloreanOr by distributing it illegally. GPL violations do happen quite a lot.
14:57:45Lloreankugel: Again though, you're assuming it's implicit. He may have meant it, but you don't have a license on the code, and really need to try to get in contact with him.
14:58:05kugelI try that for about 6 month now
14:58:07LloreanThe act of including it does *not* GPL it.
14:58:37markunI've seen him here less than 6 months ago. Next time I'll ask him about it.
14:58:44 Join LambdaCalculus37 [0] (n=44a0430f@gateway/web/cgi-irc/
14:59:14 Join desowin [0] (n=desowin@atheme/member/desowin)
14:59:33Lloreankugel: Did he code it from scratch?
15:01:26kugelLlorean: Not sure
15:01:57markunkugel: do you remember his nick on IRC?
15:02:13kugeljott or j0tt
15:02:21kugelBut jott most of the time
15:02:24Lloreankugel: These things are important. Where the code came from determines whether he even *can* license it GPL
15:02:42 Part J-23
15:03:12kugelLlorean: I didn't ask him about if he did it from scratch. The comments doesn't tell anything like "based on xx algorithm"
15:03:51 Join Mathiasdm [0] (
15:04:08LloreanWell it's pretty important to have your licensing clear.
15:04:21LloreanOne might argue essential.
15:06:49 Quit basti (Remote closed the connection)
15:08:05 Join basti [0] (i=bnc@
15:09:11kugelAnyone know how to append an image to another? I'm making a theme and like to use bitmap strips
15:09:46 Join DerPapst [0] (
15:10:43B4gderI'd assume "convert" from imagemagick can do it
15:11:09markunyes, it has various options to combine imgages, don't know which one he needs.
15:11:39LambdaCalculus37I think GIMP can do that, too.
15:12:22B4gdercommand line is king ;-)
15:13:23kugelI know gimp can do it, but I searched for an simple and fast solution :P
15:13:40LambdaCalculus37kugel: Then B4gder hit it bang on the mark. :P
15:13:44 Join perrikwp [0] (i=9821707c@gateway/web/ajax/
15:14:11kugelB4gder: Thanks, that's it
15:14:35kugelconvert -append vol* did almost work
15:15:15kugelit just appended vol0-vol10-vol1x-vol1-vol2
15:16:37gevaertskugel: composite -tile might also be helpful
15:17:09kugelgevaerts: thanks, but B4gder's way allready worked :P
15:17:22 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
15:18:11kugelmeh, it didn't
15:19:28pixelmaif you're using imagemagick - montage is quite nice...
15:20:25*gevaerts agrees. montage was the one he was actually looking for
15:21:25kugelIs that a GUI for imagemagick?
15:22:24gevaertslinuxstb: there are some ipodpatcher problems in the logs
15:25:43***Saving seen data "./dancer.seen"
15:27:20linuxstbgevaerts: Thank you, kind sir...
15:28:27kugelI can't find montage..., google gives me a lot of results, but mainly those for the German word "Montage"
15:28:57linuxstbkugel: It comes installed as one of the imagemagick utilities.
15:30:34kugelI see
15:31:16linuxstbamiconn: I assume you've had no luck solving the "UIE problem" ?
15:32:53 Join barrywardell [0] (
15:44:51 Join feisar_ [0] (i=jljhook@
15:44:51 Quit feisar (Read error: 104 (Connection reset by peer))
15:48:13 Join Pichet_Goulu [0] (
15:48:15 Nick Pichet_Goulu is now known as Pichet (
15:55:50 Quit Pichet ("ChatZilla 0.9.81 [Firefox]")
15:58:35 Join mf0102 [0] (n=michi@
16:01:48 Quit mf0102 (Client Quit)
16:02:45 Join mf0102 [0] (n=michi@
16:09:40 Join [CBR]Unspoken|w [0] (n=cbr@
16:09:48 Join miepchen^schlaf_ [0] (
16:14:40 Join psycost [0] (
16:14:43 Quit desowin (Read error: 113 (No route to host))
16:15:17 Join perrikwp|class [0] (i=98214a02@gateway/web/ajax/
16:16:40 Quit miepchen^schlaf (Connection timed out)
16:18:39 Join dabujo [0] (
16:25:09 Quit psycost ()
16:26:22 Join desowin [0] (n=desowin@atheme/member/desowin)
16:37:22 Quit perrikwp|class (" ajax IRC Client")
16:39:57 Quit DerDome (Read error: 113 (No route to host))
16:40:54markunB4gder: shall I call the new patch rockbox-multilibs-arm-elf-gcc-4.0.3_2.diff ?
16:42:01B4gderworks for me, the name isn't that important
16:43:03 Nick pabs_ is now known as pabs (
16:47:05 Quit corevette (Remote closed the connection)
16:49:05markunB4gder: the meizu bootloader loads the flashed image to SDRAM and maps it to 0x0. Should make our first tests easy.
16:50:32markunI wrote some code which should toggle the backlight on and off when pressing PLAY. It should work on all M6 and M3 targets.
16:50:42markunBut first I need to get it to work here of course :)
16:51:06markunlet's try to have rockbox running before the devcon :)
16:54:00 Quit DerPapst (Nick collision from services.)
16:54:10 Join DerPapst_ [0] (
17:04:31 Join mrkiko [0] (
17:05:27 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
17:06:29 Join DerDome [0] (
17:07:21bertrikIs there any specific reason we have an atoi.h? (it's normally part of stdlib AFAIK so could go perfectly well into stdlib.h)
17:09:17 Part LinusN
17:18:38linuxstbI thought we did?
17:19:28linuxstbAh, you mean why we _do_ have it...
17:19:54*linuxstb should probably go home
17:21:38linuxstbYes, moving it into stdlib.h sounds sensible to me.
17:22:15 Quit CaptainSquid ("Miranda IM!")
17:22:30 Quit nplus (Remote closed the connection)
17:23:36*B4gder agrees
17:25:44***Saving seen data "./dancer.seen"
17:26:25linuxstbbertrik: BTW, I've been thinking for a while that a #include cleanup is needed, so am happy you're doing something...
17:27:58B4gderyeps, nice spring cleaning!
17:29:10bertriklinuxstb, B4gder : thank you for your support :P
17:30:56*linuxstb goes home
17:30:57 Part linuxstb ("Leaving")
17:32:16markunB4gder: I've added the patch to www/gcc but can't access it over http
17:32:53markundoes it take some time?
17:33:02bertrikone particular file that nags me is plugin.h. It #includes lots of other includes but doesn't actually need them (plugin.c does). This often causes *all* plugins to be rebuilt when changing something trivial.
17:33:08B4gdermarkun: right, it doesn't update automatically - I fixed it now
17:35:46 Quit XavierGr (Nick collision from services.)
17:35:57 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
17:39:00markunB4gder: ok, the arm toolchain should now be built with big-endian support
17:39:13B4gderI noticed
17:39:28B4gderconfigure changes coming up I presume?
17:39:46markunI'll leave that for the initial meizu commit
17:40:17markunafter I've verified that the testcode runs
17:42:16B4gdergevaerts: had a chance to log any usb traffic on a meizu upgrade?
17:42:48 Join herrwaldo [0] (
17:43:06gevaertsB4gder: actually, I can do that later today
17:43:29markunB4gder: for the meizu m6sl we will also need a linux tool to combine our bootloader image with the nand loader image from samsung
17:44:08B4gdermarkun: hm, where do we get that name image from?
17:44:13B4gderfrom the unit itself?
17:44:18B4gdernand even
17:44:38markunno, it's supplied with the DFUManager from Samsung
17:46:30LambdaCalculus37markun: With your recent commit, does that mean we have to update our build environment or no?
17:46:47B4gderLambdaCalculus37: if you wanna build meizu rockbox, yes
17:46:52markunLambdaCalculus37: only if you want to compile a meizu build (in future)
17:46:54B4gderuntil then no
17:47:42LambdaCalculus37Ah, okay.
18:04:37markunB4gder: do you have any idea why system.o doesn't get linked when UIE is only used in crt0.S?
18:04:53gevaertsmarkun: which dfu tool do you want me to trace ?
18:08:36 Join Chronon [0] (
18:11:29B4gdermarkun: I've only seen it discussed here, I haven't had time to check it out myself
18:13:18 Quit petur ("work->home")
18:15:11 Part B4gder
18:15:21 Quit DerPapst_ (Read error: 110 (Connection timed out))
18:20:00markungevaerts: will that work?
18:20:45 Join sitwon [0] (
18:21:28gevaertsmarkun: I'm not sure if it actually writes new firmware, but at least it sent data and the player reset at the end
18:22:49*gevaerts waits for the PC with tracer software to come back. It decided to reboot
18:23:13markunwhich firmware did you upload?
18:23:59 Join DerPapst [0] (
18:34:17Manoj`hi folks. Any ipodpatcher/rbutilqt folks around? are seg-faults on amd64 machines a known issue?
18:34:40 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
18:35:24*gevaerts points to domonoky
18:35:53*domonoky looks around... turns to the logs.. :-)
18:36:11 Join bluebrother [0] (n=dom@rockbox/staff/bluebrother)
18:36:13gevaertsdomonoky: last line before you joined :)
18:36:22Manoj`__> ./ipodpatcher −−list :: Image 1: Main firmware - 11720704 bytes; Third-party bootloader - 52796 bytes
18:36:28Manoj`[INFO] Ipod found - Video (aka 5th Generation) ("winpod") - /dev/sdf
18:36:54Manoj`[INFO] Using internal bootloader - 50136 bytes
18:36:54Manoj`Segmentation fault
18:36:57*desowin suggests alwasys attach backtrace to any segfault report
18:37:01desowinas segfault != other segfault
18:37:15*domonoky thinks, yes there are known problems with ipodpatcher on amd64..
18:37:17Manoj`I get this using either the 64bit ipodpatcher; or the 32bit one running in a 32bit chroot
18:37:39domonokyi think you have to run it with "linux32 ./ipodpatcher"
18:38:24Manoj`the 32bit version?
18:38:44domonokybut i am not sure, as i use windows.. :-)
18:40:41bluebrotherdon't we have a 64bit binary?
18:41:09Manoj`bluebrother: we do
18:41:17Manoj`bluebrother: same segfault
18:42:10Manoj`The strace is identical
18:42:29bluebrotherlooks like strace?
18:42:34Manoj`I can put up the full strace on a web server if that would be useful
18:43:05Manoj`The tail end can be seen at
18:43:18bluebrothercan you backtrace it using gdb? Or are the binaries stripped?
18:43:42Manoj`They are stripped
18:43:50Manoj`ipodpatcher: ELF 32-bit LSB executable, Intel 80386, version 1 (SYSV), for GNU/Linux 2.6.1, dynamically linked (uses shared libs), stripped
18:43:56 Join BigBambi [0] (n=Alex@rockbox/staff/BigBambi)
18:43:58 Join gregzx [0] (
18:44:22bluebrotherhmm. I guess a backtrace wouldn't tell much in that case. Can you build yourself?
18:45:15Manoj`I suppose I can, unless it needs cross compilation for bits
18:45:18markunhow can I put some .word values in crt0.S which will end up at the very end of the image?
18:45:34bluebrotherno, you don't need a cross compiler.
18:46:18bluebrotherget rbutil/ipodpatcher from svn. Building should be as simple as calling make
18:46:48Manoj`bluebrother: ok.
18:52:38 Quit perrikwp (" ajax IRC Client")
18:52:47 Join ackbahr [0] (
18:53:00amiconnDoes someone intend to fix the austriancoder's audio mess soon(ish)? Should it perhaps be reverted?
18:54:09 Join nplus [0] (
18:54:18amiconnmarkun: If you need to make sure these words are always last, you need to introduce a separate section, and use the linker script to put it last
18:54:28 Join n1s [0] (n=nils@rockbox/developer/n1s)
18:54:50*amiconn wonders what that could be necessary for though
18:56:00Manoj`Hmm. perhaps I should not use git-svn
18:58:18bluebrotherfor getting just ipodpatcher it might be easier to pull it directly ;-)
18:58:34 Quit gevaerts ("work->home")
18:58:45Manoj`I'm just pulling trunk/rbutil
18:58:58Manoj`I would like to help resolve the rbutil segfault too
18:59:11*bluebrother goes checking the log
18:59:50domonokyManoj`: rbutil just uses ipodpatcher, so if the segfault in ipodpatcher is fixed, rbutil should also work (if its the same problem) ..
19:00:03n1sfestival is the speech engine used for the voice files we provide, right?
19:00:10Manoj`ok. I'll see what I can do
19:01:18Nico_Pbertrik: nice commits :)
19:01:31Nico_PI'm all for the plugin.h cleanup you mentioned too
19:02:25bertrikNico_P: yeah, I hate it to see all plugins recompile (especially DOOM)
19:02:46Nico_Psame here (although I usually use make bin because of that)
19:03:24*bertrik etches mental note about 'make bin' in brain
19:03:46Manoj`fat32format.c: In function 'format_partition': fat32format.c:465: warning: format `%llu' expects type 'long long unsigned int', but argument 3 has type 'long unsigned int'
19:03:55Manoj`not an error, but ...
19:04:18 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
19:04:22Nico_PManoj`: %lu
19:04:28 Join petur [50] (n=petur@rockbox/developer/petur)
19:05:21*amiconn checks the log to find out what bertrik is plaiing to rework
19:05:41*linuxstb spots lots of unsigned longs in ipodpatcher...
19:06:49Nico_Pamiconn: removing most of the includes in plugin.h, as they're useless
19:07:01*n1s is happy as long as noone assumes sizeof(long)==sizeof(int)==sizeof(*int)
19:07:15amiconnlinuxstb: No, I wasn't able to solve the UIE() problem yet. It's really strange, all symbols are defined as they should be, yet the linker doesn't pick system.o
19:07:23linuxstbManoj`: The obvious thing to fix, is to change "unsigned long" to "uint32_t" and "long" to "int32_t"
19:08:07amiconnNico_P: Why are they useless? Many of them are rather useful...
19:08:08*linuxstb should have known better than to use longs, but it was about 2.5 years ago...
19:09:00Nico_Pamiconn: plugin.h doesn't need to have access to the prototypes of the functions you want to add to the plugin API, does it?
19:09:04amiconnPlugins aren't allowed to include anything from the core except plugin.h, so plugin.h should provide everything that's generally defined
19:09:34Manoj`linuxstb: I can do that. Also, the pre-build ipodpatcher seems to already have an internal bootloader, the svn version does not
19:09:45amiconnAnd there is a whole bunch of inline functions and macros which are needed in plugins
19:10:03Nico_Pdata structures too
19:10:12bertrikamiconn: yes, I see not that it's not that bad, but there are some header files in plugin.h that do not provide anything useful at all, except for an annoying dependency
19:10:21 Join bughunter2 [0] (
19:10:30bertrik(remove first not)
19:10:50amiconnbertrik: In fact I'm not annoyed by a few extra dependencies, but I'm *very* annoyed about missing dependencies
19:10:59 Join bluebroth3r [0] (n=dom@rockbox/staff/bluebrother)
19:11:15domonokyManoj`: yes, you can build ipodpatcher with and without internal bootloader ( the bootloaders are also on
19:11:22linuxstbManoj`: If you read the instructions in the Makefile, that tells you how to build ipodpatcher with the embedded bootloaders (there is one for each ipod model)
19:11:43amiconnLike the codecs not relinking if a codec lib changed... and then wondering why nothing changes when testing....
19:12:53linuxstbshotofadds: Ping...
19:13:12n1swhere am i supposed to add speech engine specific tweaks?
19:13:30bertrikamiconn: ok, can I do something about that?
19:13:47 Quit bluebrother (Read error: 104 (Connection reset by peer))
19:13:50 Join houbysoft [0] (
19:14:01n1sbertrik: you get beer if you fix that one! :)
19:14:01domonokyn1s: speech engine specific tweaks for what ?
19:14:18amiconnbertrik: If I would know the solution, I would already have fixed it....
19:14:38n1sdomonoky: pronunciation of words that the engine gets wrong...
19:14:48bertrikamiconn: probably something in the Makefiles?
19:15:18n1sah, thanks
19:16:17domonokythat reminds me, that i should add pronunciation correction into rbutil.. :-)
19:16:27amiconnIt's supposed to be broken down by language->TTS->voice vendor
19:17:50n1samiconn: yes, the problem is that festival pronounces 'a' as the word a when trying to spell (is sounds like a very short ö
19:18:07amiconnYes, try to quote it
19:18:50amiconnThis helped for several sapi voices (but for an unknown reason isn't necessary anymore)
19:19:40n1sdoesn't seem to make a difference when trying directly in festival from the command line, "Ay" works pretty well though
19:20:21amiconnYou probably spotted the ploogin protection in that function ;)
19:20:49 Nick bluebroth3r is now known as bluebrother (n=dom@rockbox/staff/bluebrother)
19:21:19n1syes but now i wonder how to match "A" but nothing else, my sed-fu isn't that great
19:21:37amiconnThis is plain regex
19:21:50 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
19:22:22amiconnYou want an a that has no letter before or after, so it should be s/\ba\b/ay/ig
19:22:43amiconn(matches small and capital letter a)
19:22:51n1sthanks, I'm basically regex illiterate :)
19:23:50Manoj`#0 0x00002b8f38c06e6b in memcpy () from /lib/
19:23:50Manoj`#1 0x000000000040344f in add_bootloader ()
19:23:50Manoj`#2 0x0000000000402077 in main ()
19:23:56 Join Buschel [0] (
19:24:11 Join ompaul [0] (n=ompaul@gnewsense/friend/ompaul)
19:25:48***Saving seen data "./dancer.seen"
19:27:12domonokyManoj`: is it also segfaulting if you try an external bootloader ?
19:28:31Manoj` ./ipodpatcher -a bootloader-ipodvideo.ipod
19:28:40Manoj`[INFO] Ipod model: Video (aka 5th Generation) ("winpod")
19:28:40Manoj`[INFO] Checksum OK in bootloader-ipodvideo.ipod
19:28:40Manoj`[ERR] Input file too big for buffer
19:28:40DBUGEnqueued KICK Manoj`
19:28:40Manoj`[ERR] −−add-bootloader failed.
19:29:14CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
19:29:14*Buschel asked a collegue to build up a iPod dock connector debug-board
19:29:14 Quit gregzx ("ChatZilla 0.9.81 [Firefox]")
19:29:34Buschelhopefully this will help finding a proper solution for the accessory detect
19:30:04Buschelwe've got lots of connector cables in our company, so mis-using one of them should be ok ;o)
19:30:57Buschelalso dip switches for changing the resistor coding will be part of it
19:30:59Manoj`when I run gdb, it says it is looking for a file called init.c
19:31:29Manoj`I have no idea where it is getting init.c from
19:31:39Manoj`but I can't list the source lines
19:32:24domonokyhm Inputfile to big for buffer is strange, your bootloader is surely smaller than 8MB, i think..
19:32:50Manoj`__> ls -als bootloader-ipodvideo.ipod
19:32:50Manoj`56 -rw-r−−r−− 1 root srivasta 50528 2007-05-23 13:19 bootloader-ipodvideo.ipod
19:36:38domonokyare "int"s also different on amd64 ?
19:36:54Manoj`an int is 64bit
19:38:22linuxstbI thought ints were 32-bit, and longs/pointers were 64-bit?
19:39:18n1slinuxstb: correct
19:39:49 Join weeandykidd [0] (n=519c26a9@gateway/web/cgi-irc/
19:40:15 Join perrikwp [0] (i=9821707c@gateway/web/ajax/
19:40:37BigBambiweeandykidd: If you have a question, just ask it
19:40:50weeandykiddi need help on how to add optoins to the open with menu
19:41:00n1samiconn: that didn't seem to work, I assume the /i means it's case insensitive?
19:41:31linuxstbweeandykidd: What exactly do you want to do?
19:41:53weeandykiddcan you get idarcnes in the rockbox OS
19:42:23BigBambiIs that a NES emulator?
19:42:35BigBambiIf so, the answer is theoretically yes, if you port it
19:42:51BigBambiYou cannot just 'install' it though
19:43:40weeandykiddiyes but when i set it up the rockbox doesnt know what to open the roms with, so i hold the center button selct open with but the emulator isnt there
19:43:50 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
19:43:58BigBambiI didn't really understand that question
19:44:02linuxstbweeandykidd: You're right, it's not there...
19:44:18BigBambiBut as Rockbox does not contain a NES emulator, of course you can't open a ROM with it
19:44:33weeandykiddi have placed one on it
19:44:56BigBambiYou changed the code to work with Rockbox (porting it), then you compiled rockbox?
19:45:01*BigBambi thinks not
19:45:07weeandykiddis there anyway to add entries to the open with menu; thats all i really want to know
19:45:22 Join lee-qid [0] (
19:45:29BigBambiWhy? You have not added the NES emulator, I am sure
19:45:32DerPapstweeandykidd: yes. if you know how to program with c
19:45:55linuxstbweeandykidd: You edit the .rockbox/viewers.config file to point to the ".rock" file (Rockbox plugin) that will open the file.
19:46:07Manoj`I stand corrected. The size of int is: 4
19:46:18weeandykiddmild experiennce. ill work at it now then. just wanted to know if i was wasting my time
19:46:31 Quit weeandykidd ("CGI:IRC")
19:46:46 Quit DerPapst (Read error: 104 (Connection reset by peer))
19:46:49 Join DerPapst_ [0] (
19:46:55 Quit DerPapst_ (Remote closed the connection)
19:47:08 Join DerPapst [0] (
19:47:14*linuxstb scratches head
19:47:55*LambdaCalculus37 wonders why kids keep coming in asking about emulators for Rockbox
19:48:16 Quit barrywardell (Remote closed the connection)
19:49:29domonokyManoj`: you could add some printfs to add_bootloader, to findout what goes wrong..(like checking entryOffset and padded lenght, to see if they are sane) but i dont really know where it could go wrong, or if even more is broken..
19:50:59Manoj` memcpy(sectorbuf+entryOffset,ipod->bootloader,ipod->bootloader_len);
19:51:36Manoj`I think that is where thigs die.
19:51:51domonokyyes, there is the segfault for sure..
19:52:40Manoj` fprintf(stderr,"[INFO] Using internal bootloader - %d bytes\n",ipod->bootloader_len);
19:52:40Manoj` memcpy(sectorbuf+entryOffset,ipod->bootloader,ipod->bootloader_len);
19:52:40Manoj` fprintf(stderr,"[INFO] Done\n");
19:52:41domonokyyou could printout entryOffset and ipod->bootloader_len before to see if the look right..
19:52:48Manoj`[INFO] Using internal bootloader - 50520 bytes
19:52:48Manoj`Segmentation fault
19:53:13Manoj`so ipod->bootloader_len == 50520 bytes
19:53:19domonokyso it looks like entryOffset is wrong.. then len is sane..
19:54:32Manoj`[INFO] Offset - 11720704 bytes
19:54:44Manoj`umm, what?
19:54:57 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
19:55:06amiconnn1s: yes
19:55:15Manoj`That looks a bit high
19:55:39domonokyjup, thats surely wrong..
19:55:49n1samiconn: in pearl does matching one case make it ignore the rest?
19:56:23Manoj`So, perhaps my main firmware got hosed?
19:56:27Manoj`using −−list:
19:56:29Manoj`Image 1:
19:56:29Manoj` Main firmware - 11720704 bytes
19:56:29Manoj` Third-party bootloader - 52796 bytes
19:57:05 Join jgarvey [0] (
19:57:15 Join shotofadds [0] (n=rob@rockbox/developer/shotofadds)
19:57:20domonokylinuxstb: does this sound right ?, i dont have too much knowledge of ipods..
19:57:53shotofaddslinuxstb: if it's about UIE, I don't know the answer. I don't recall having to do any magic for the D2 build.
19:58:36linuxstbdomonoky: Yes, it sounds about right, although 11MB seems a bit large...
19:58:42domonokybut the entryOffset can only be wrong, as sectorBuf only has 8MB space, and you go to offset 11MB...
19:59:28linuxstbshotofadds: No, it was actually about your interrupt handler, but I found the answer (I thought you were doing something wrong, but it turns out the tcc780x's interrupt controller can do a lot more than the tcc77x's).
20:00:18shotofaddsyeah, the 78x interrupt controller is quite fully featured...
20:00:55*n1s found the answer and it's "yes", no wonder my change didn't work :)
20:02:23amiconnn1s: Well, you're applying the regexps in sequence, so all should match their respective patterns
20:03:31n1samiconn: the regexp wasn't the problem, it was me assuming a switch worked like in c (for some reason...)
20:03:51Manoj`I seem to have created backups a long time ago: apple_os.bin apple_sw_5g_rcsc.bin bootloader-video.bin bootpartition.bin mbr.bin ...
20:05:46n1sworks nicely now :)
20:06:35Manoj`I have, unfortunately, forgotten how to restore these backup versions of the apple OS software
20:06:35domonokyManoj`: do you have access to a 32bit machine to test if it is your ipod and not the 64bit OS ?
20:06:53Manoj`Unfortunately, no
20:07:08linuxstbManoj`: You just do "dd if=bootpartition.bin of=/dev/sdX1"
20:07:12Manoj`I know there is something wrong with the ipod
20:07:31 Quit ackbahr (Read error: 104 (Connection reset by peer))
20:07:41Manoj`I can no longer boot into orckbox, so I am willing to trust that something got messed up
20:07:47Manoj`on the ipod, I mean
20:08:18linuxstbamiconn: Do you know if ipodpatcher ever worked on amd64? i.e. did you use it?
20:08:21Manoj`I can do the dd
20:08:25domonokylinuxstb: if it is his ipod, we surely should add a check for (sectorbuf+entryOffset < BUFFSIZE) before the first memcpy in add_bootloader
20:08:33Manoj`What are the other bits I saved?
20:08:58Manoj`the apple_os.bin, etc?
20:09:16domonokythe extracted apple_os (two files) ..
20:12:24amiconnlinuxstb: I think I tried it once, but I'm not sure
20:15:18*amiconn renews his question
20:15:51 Join faemir [0] (
20:22:41 Join gregzx [0] (
20:24:56Manoj`Success. I dd'd the bootpartition.bin over, and then ipodpatcher did work.
20:26:05gevaertsgreat :)
20:26:27bluebrotherok, so the question remains: will the prebuilt binary work?
20:27:31Manoj`Oh. I dunno. This version I am using has all the long int's and unsigned long's converted to u?int32_t
20:27:54gevaertsmarkun: the tracer software works with wine (at least for viewing files)
20:28:53Manoj`The proper offset was 6477824; btw
20:30:26domonoky6MB + 50k surely fit into 8Mb buffer... linuxstb could you please put a safety check before the memcopy in add_bootloader :-)
20:34:03 Quit houbysoft ("")
20:44:05 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
20:44:10 Join grndslm [0] (
20:47:20 Quit Buschel ()
20:49:28 Join midgey [0] (
20:53:39 Quit linuxstb (Nick collision from services.)
20:53:40 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
20:59:25 Join spiorf [0] (
21:03:47Manoj`gevaerts, domonoky: Thanks for your help. With the restoration of the bootpartition, everything seems to be working, and now the major decision is which theme to use
21:04:20domonokyManoj`: good to hear :-)
21:04:33gevaertsThat's a hard decision to make :)
21:04:59Manoj`arboxWidgets vs. phkOSC
21:05:44 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
21:05:53 Quit linuxstb (Nick collision from services.)
21:05:56Manoj`I am surprised that no one has packaged the tbutil for the linux distributions
21:05:57 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
21:07:13bluebrotherthere is a contributed ebuild file in the wiki
21:07:31*gevaerts guesses that Manoj` doesn't care about ebuilds ;)
21:07:57domonokyManoj`: feel free to help out.. :-)
21:07:59*bluebrother could create a rpm file but hasn't gotten around actually doing it.
21:09:03bluebrothermaybe we should create tarballs of release versions
21:09:13bluebrotherwould at least make that easier.
21:10:31*amiconn renews his question regarding austriancoder's audio stuff once more
21:12:53markunamiconn: thanks for your reply about the linker section
21:13:12markunand no, I don't plan to fix any audio code
21:14:22*gevaerts also doesn't plan to work on the audio code. No sense in making it worse...
21:15:11bertrikviewport.h contains some #defines like VP_SELECTIONCOLORS which are not used anywhere
21:16:49 Join einhirn_ [0] (
21:17:06linuxstbbertrik: I know JDGordon was working on various viewport-related things, and only some have been committed, so maybe they are left from that...
21:17:08 Quit grndslm (Read error: 113 (No route to host))
21:18:37bertrikok, I'll leave those alone until I speak to him again
21:19:43 Join grndslm [0] (
21:23:24Manoj`domonoky: I'll put rbutil/ipodpatcher on my list of TODO things for Debian (I need to get SELinux in shape first, though)
21:23:34Manoj`Thanks for a useful tool
21:25:21 Quit TMM (Remote closed the connection)
21:25:53***Saving seen data "./dancer.seen"
21:26:22 Quit ApplianceBlue ()
21:26:22 Part Manoj` ("ERC Version 5.3 (IRC client for Emacs)")
21:35:49 Quit kugel (Read error: 110 (Connection timed out))
21:39:16 Quit desowin (Excess Flood)
21:39:37 Join desowin [0] (n=desowin@atheme/member/desowin)
21:53:02 Quit desowin ("KVIrc 4.0.0 Insomnia")
21:56:08 Quit LambdaCalculus37 ("CGI:IRC 0.5.9 (2006/06/06)")
21:56:57 Join w1ll14m [0] (
21:57:13 Join Rincewind [0] (
21:57:37 Quit nplus ("Leaving")
21:58:02 Join barrywardell [0] (n=barrywar@
22:09:11 Quit XavierGr (Read error: 104 (Connection reset by peer))
22:09:24 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
22:09:34 Quit linuxstb (Read error: 110 (Connection timed out))
22:09:46 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
22:21:07 Quit leox (Read error: 104 (Connection reset by peer))
22:24:49 Join TMM [0] (
22:26:12 Quit Mathiasdm ("Yuuw!")
22:26:41 Join dabujo_ [0] (
22:26:59 Join Jckf [0] (i=bounce@
22:27:05 Part Jckf ("Leaving")
22:40:31 Join leox [0] (
22:44:30 Quit dabujo (Connection timed out)
22:47:32 Join PaulJam [0] (
22:48:14gevaertsmarkun: I tried your code, but nothing much happened
22:49:12 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
22:49:28 Quit linuxstb (Nick collision from services.)
22:49:32 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
22:50:58 Quit dabujo_ (Connection timed out)
22:51:33 Join lee-qid_ [0] (
22:58:08 Join csc` [0] (n=csc@archlinux/user/csc)
22:58:42 Quit bughunter2 ("Leaving.")
22:59:55 Quit mf0102 ("Ex-Chat")
23:02:31bertrikamiconn: still here?
23:07:25DerPapstamiconn doesn't go to bed before 1am :-P
23:08:47 Quit bluebrother ("leaving")
23:09:17 Quit lee-qid (Read error: 110 (Connection timed out))
23:14:10 Quit n1s ()
23:15:21DerPapst...normally ;-)
23:16:07 Join bertrik_ [0] (
23:16:39 Quit jhulst ("Konversation terminated!")
23:18:05 Quit petur ("Zzzz")
23:20:32 Part domonoky
23:24:37 Quit mrkiko (Remote closed the connection)
23:25:55***Saving seen data "./dancer.seen"
23:27:24 Quit OlivierBorowski (Remote closed the connection)
23:31:13 Quit davina (Remote closed the connection)
23:32:38 Quit bertrik (Read error: 110 (Connection timed out))
23:34:54 Join austriancoder [0] (n=austrian@rockbox/developer/austriancoder)
23:36:34Bagderaustriancoder: your commit seems to not work on the gigabeat
23:37:59austriancoderBagder: define... "seems not to work" please
23:38:06Bagdersee log
23:38:10BagderI don't have a gigabeat
23:39:21BagderI just read the patch, thought it looked weird and people confirmed it seems to not work
23:39:25 Quit DerPapst (Read error: 104 (Connection reset by peer))
23:39:47 Join DerPapst [0] (
23:39:53austriancoderBagder: okay found it.. will try to fix it
23:41:16amiconnaustriancoder: I think that things changing behaviour on many targets like your audio stuff needs testing on the affected targets before commit.
23:41:47amiconnThis is not that much an issue with new stuff, but the audio settings were working properly before
23:42:33amiconnbertrik_: The atoi.h was most probably from before we even had stdlib.h
23:42:38austriancoderamiconn: next time i will testers test my patch, if i do changes like the wm8751 change
23:43:38bertrik_amiconn: I decided not to change plugin.h by the way, it didn't really contain any unused header files
23:46:57 Quit jgarvey ("Leaving")
23:48:05 Quit amiconn (Nick collision from services.)
23:48:12 Join amiconn [50] (n=jens@rockbox/developer/amiconn)
23:48:15 Nick bertrik_ is now known as bertrik (
23:51:19 Quit DerDome ("Leaving.")
23:51:48austriancoderany gigabeat user online?
23:52:28DerPapstiirc gevaerts and markun have one
23:52:34*gevaerts waves
23:53:23austriancodergevaerts: fine... can i send you in some minutes a patch to test?
23:54:13DerPapstthe devcon dap list should be taken and put on a separate wiki page so people know whom to annoy if they need a tester :-P
23:54:31gevaertsaustriancoder: sure, but be aware that I don't have very good ears :)
23:55:21*bertrik was going to mention that
23:55:27DerPapstwow, gevaerts... good work
23:55:35DerPapstin such a short time :-P
23:56:30 Join ^Parrot^ [0] (
23:56:35*DerPapst fills a gap
23:57:20bertrikhow about using a stub for function ata_disk_is_active in stubs.c instead of using #ifdef SIMULATOR?
23:57:20austriancodergevaerts: please test volume and bass settings
23:57:38^Parrot^evening. does the latest build have AA support, bmp resize built into it?
23:57:54gevaertsaustriancoder: What should I test for ? Identical behaviour as older builds ?
23:58:12PaulJamAA yes, resizing no.
23:58:16austriancodergevaerts: yes..
23:58:29^Parrot^how about scrolling?
23:58:57^Parrot^I keep seeing themes that require these patches

Previous day | Next day