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 2011-10-19

00:00:03CIA-14r30791 build result: 4 errors, 0 warnings (pamaury committed)
00:00:51CIA-14New commit by pamaury: r30793): imx233/fuze+: move to a more generic xtal derived clock enabling function, fix timrot accordingly
00:00:54CIA-14New commit by pamaury: r30794): imx233/fuze+: add volume handling to sound.c
00:01:12***Saving seen data "./dancer.seen"
00:01:58CIA-14r30792 build result: 4 errors, 0 warnings (pamaury committed)
00:02:28 Join ffowcs [0] (
00:02:52pamaurygit svn dcommit is excruciatingly slow...
00:03:27CIA-14New commit by pamaury: r30795): imx233/fuze+: allow dma info retrieval; wait for end of channel reset before returning; fix typo
00:03:30CIA-14New commit by pamaury: r30796): imx233/fuze+: fix codec prototype header
00:03:35CIA-14r30794 build result: 4 errors, 2 warnings (pamaury committed)
00:03:44ffowcsIs it possible to customize the button configuration (keymap) for rockbox?
00:05:06CIA-14r30796 build result: 4 errors, 0 warnings (pamaury committed)
00:06:03CIA-14New commit by pamaury: r30797): imx233/fuze+: implement hw info debug screen to show dma channels state
00:06:06CIA-14New commit by pamaury: r30798): imx233/fuze+: implement audioout functions (init, volume, frequency); add stubs for audioin
00:06:50 Nick [Sanyi] is now known as [sanyi] (~sanyi@unaffiliated/alexanderc)
00:07:56CIA-14r30797 build result: 4 errors, 0 warnings (pamaury committed)
00:08:39CIA-14New commit by pamaury: r30799): imx233/fuze+: add prototype for rtc block (only for register defines for now)
00:08:43CIA-14New commit by pamaury: r30800): imx233/fuze+: implement pcm out and audio codec, update SOURCES
00:09:02pamaurythat's weird, are the svn hooks somehow blocking the commit until the build is complete ? Because my git svn dcommit keeps sending two commits then wait for the build results, then send two commits, ...
00:09:26 Quit domonoky (Read error: Connection reset by peer)
00:09:49CIA-14r30798 build result: 4 errors, 0 warnings (pamaury committed)
00:09:53Zagorpamaury: no, there is no waiting. it's just git->svn being very slow.
00:10:01pamauryusually it's quite fast
00:11:29CIA-14r30800 build result: All green
00:11:54pamauryfinally :)
00:14:04pamaurythere is something wrong with usb, svn head not longer has working usb on my fuze+, acts like it's never detected
00:14:50pamauryI'll investigate that tomorrow
00:15:17gevaertspamaury: r30701 ?
00:17:27pamauryDon't know, I'm too tired to have a look but it's quite possible
00:18:39pamauryI must admit I never had a close look at this commit but I will because it look suspicious
00:19:29CIA-14New commit by zagor: r30801): Replace puts* with printf.
00:19:35 Quit pamaury (Remote host closed the connection)
00:30:20 Quit Jerom (Quit: Leaving.)
00:30:59 Quit liar (Remote host closed the connection)
00:32:12 Part Zagor
00:32:32 Quit ffowcs (Quit: Leaving)
00:32:48 Join Strife1989 [0] (
00:35:47 Quit Strife89 (Ping timeout: 240 seconds)
00:36:30 Quit bertrik (Ping timeout: 248 seconds)
00:37:02 Join Scromple [0] (
00:41:45 Join D-Strife89 [0] (~Strife89@
00:42:07 Join helloworld [0] (
00:42:12 Quit helloworld (Client Quit)
00:42:12 Nick D-Strife89 is now known as L-Strife89 (~Strife89@
00:42:23 Join kev2011 [0] (
00:42:43 Quit Strife1989 (Read error: Operation timed out)
00:43:39kev2011can someone tell how did rockbox team did to hack ipod and other devices, how can i have more info about the way to do a similar thing?
00:48:18saratogakev2011: on most targets its though a combination of disassembly of the retail firmware's bootloader and referencing any datasheets that are available
00:49:18kev2011do you have any usefull resources to learn doing that?
00:49:36saratogalearning which part specifically?
00:49:56kev2011well disassembly of a firmware
00:51:25saratogado you know any assembly programming?
00:51:34kev2011yep x86
00:51:45saratogathen it should be straighfoward for you
00:52:05saratogamost of the time people use objdump or IDA
00:52:10saratogathen go from there
00:53:08saratogayou'd have to get up to speed on ARM first though
00:53:20kev2011arm assembly ?
00:53:50saratogaits fairly simple to learn
00:54:37saratogawhich player do you own
00:54:44kev2011ipod nano 2g
00:55:31kev2011i do i have access to the memory of it? LIke to "see" the firmware?
00:55:38kev2011*how di I
00:55:44kev2011** how do I
00:56:18saratogaits complicated on that player and I can't remember the details
00:56:27saratogacheck on the wiki or freemyipod, the info should be there
00:57:30kev2011ok thank you for your help, I will check around these things
01:00:13 Quit kev2011 (Quit: CGI:IRC (EOF))
01:02:49 Join kev2011 [0] (
01:03:06 Quit Rob2222 (Read error: No route to host)
01:03:12 Join Rob2222 [0] (
01:05:27kev2011You said that firmware dissasembly of ipod nano 2g would be complicated, is it easier for a mobile phone or something like that?
01:07:21wodzkev2011: ipod firmware is encrypted - you need to decrypt it first - that's what complicated
01:07:35kev2011oh i see..
01:08:18wodzkev2011: If you simply want to get familiar with arm asm - go dissasm rockbox binary for some arm target - you will get reference in source for free :-)
01:09:14kev2011well I want to learn how to modify a firmware
01:10:01wodzthats very wide topic
01:10:15kev2011yeah i don't doubt it
01:10:19 Join Resistance128 [0] (
01:10:25Resistance128anyone there?
01:10:54 Quit Resistance128 (Client Quit)
01:11:14kev2011i know C C++ X86 asm on windows
01:12:01kev2011so I need to learn arm asm
01:13:35kev2011how do i get started with dumping a firmware and maybe understand it ?
01:15:53wodzThat strongly depend on platform.
01:16:35 Join hugiec1 [0] (
01:17:09wodzBasically you get firmware update file and mangle it unless it gives meaningful disassembly. Then you compare with datasheet addresses to get some feeling what it does
01:18:36 Quit hugiec1 (Client Quit)
01:18:42kev2011i though to try on the ipod nano 2g because it seem "simple" to me but maybe not..
01:19:24wodzOther strategy is to exploit some bug in OF and run custom code from there. Then dump content of the rom and disassm it
01:19:35wodznano2g is difficult target
01:21:25kev2011do you know easier target?
01:21:38wodzThere are various reverse engineer techniques - go google
01:26:23 Quit wodz (Quit: Leaving)
01:26:26saratogamost of the sandisk players are very easy, sandisk distributes the firmware on their website, and our tools can dump it into sections for each of its functions
01:32:08kev2011like sansa clip + ?
01:49:19 Quit L-Strife89 (Ping timeout: 244 seconds)
01:49:44 Join L-Strife89 [0] (~Strife89@
01:52:36 Quit L-Strife89 (Client Quit)
01:52:55 Join L-Strife89 [0] (~Strife89@
02:01:13***Saving seen data "./dancer.seen"
02:02:35 Quit froggyman (Ping timeout: 240 seconds)
02:05:35 Quit ReimuHak_ (Quit: Leaving...)
02:24:35 Quit L-Strife89 (Quit: Going home)
02:33:53 Quit hilbert (Quit: Textual IRC Client:
02:39:28kev2011saratoga: Thank you for your help. I got to go.
02:39:46 Quit kev2011 (Quit: CGI:IRC)
02:40:59 Quit mudd1 (Ping timeout: 244 seconds)
02:45:23 Join ReimuHak_ [0] (
03:08:32 Join robin0800 [0] (
03:12:37 Quit saratoga (Ping timeout: 265 seconds)
03:34:22 Quit robin0800 (Quit: Leaving)
03:48:06 Join robin0800 [0] (
03:52:11 Join sinthetek [0] (
03:52:12 Quit sinthetek (Changing host)
03:52:12 Join sinthetek [0] (~sinthetek@unaffiliated/sinthetek)
04:01:17***Saving seen data "./dancer.seen"
04:12:46 Quit pixelma (Disconnected by services)
04:12:48 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
04:12:50 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
04:13:57 Quit amiconn (Disconnected by services)
04:13:57 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
04:14:19 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
04:28:10 Quit [7] (Disconnected by services)
04:28:10 Join froggyman [0] (~seth@unaffiliated/froggyman)
04:28:21 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
04:29:52 Join hilbert [0] (
04:38:22 Join Keripo [0] (
04:43:13 Quit Keripo (Ping timeout: 276 seconds)
04:43:22 Quit hilbert (Quit: Textual IRC Client:
04:53:07 Quit robin0800 (Quit: Leaving)
04:56:12 Quit chkktri_ (Ping timeout: 256 seconds)
05:10:14 Quit MethoS- (Remote host closed the connection)
05:39:23 Quit Horscht (Quit: Verlassend)
05:48:47 Quit freddyb (Quit: Leaving)
05:55:50 Join Rob2223 [0] (
05:59:35 Quit Rob2222 (Ping timeout: 240 seconds)
06:01:19***Saving seen data "./dancer.seen"
06:35:01 Nick kugelp is now known as kugel (~kugel@rockbox/developer/kugel)
07:20:35n1sr30801 was never built it seems
07:27:21 Quit n1s (Remote host closed the connection)
07:55:45 Join Buschel [0] (
08:00:34 Join Zagor [242] (~bjst@rockbox/developer/Zagor)
08:01:22***Saving seen data "./dancer.seen"
08:02:36CIA-14New commit by buschel (r30802): Reviewed and updated recording samples rates in the manual. Fixes FS #12318.
08:04:32CIA-14r30802 build result: All green
08:05:34Zagorn1s: r30801 was a branch commit. only trunk is built.
08:07:19 Quit swilde (Remote host closed the connection)
08:08:59 Join einhirn [0] (
08:11:50*JdGordon isnt having much luck getting the skin tree into buflib :/
08:16:53 Join GodEater_ [0] (93722cc8@rockbox/staff/GodEater)
08:21:08 Quit Buschel (Quit: ChatZilla 0.9.87 [Firefox 7.0.1/20110928134238])
08:26:00 Quit GodEater_ (Quit: Page closed)
08:26:26 Join mystica555_ [0] (
08:26:53amiconnJdGordon: Imo that shouldn't be done just now. We already have more than enough instability...
08:34:26JdGordonI keep hearing about this instabilit... is there any actual lists of what is unstable at tht emoment?
08:36:31 Quit n17ikh (Ping timeout: 248 seconds)
08:37:05amiconnDon't you use rockbox on a device?
08:37:05 Join Keripo1 [0] (
08:37:38JdGordonworks fine on my clip once the fonts were fixed
08:38:02amiconnDo you use voice? Does it save resume position?
08:38:15amiconnDoes it even shutdown cleanly?
08:38:40JdGordonno, seems to work, seems to work
08:38:55*amiconn needs to cross check this on other targets, but with voice enabled, he has to *always* use the paperclip at shutdown
08:39:10amiconn(after playback, that is)
08:39:37amiconnOtherwise it hangs at the "shutting down" splash indefinitely
08:40:16amiconnAnd when that happens (100% reproducable with my in-car settings) it also doesn't save resume position
08:42:34amiconnIf you try to get to the menu from the wps, it also hangs
08:43:44 Quit Keripo1 (Ping timeout: 255 seconds)
08:44:04amiconnIow, it's almost unusable atm if you need/ want voice
08:47:12amiconnIt even freezes when you want to get to the menu after stopping playback again (maybe that one's not 100% - got it in one of two tries)
08:49:16 Join ender` [0] (
08:52:22 Join GodEater_ [0] (93722cd0@rockbox/staff/GodEater)
08:53:28 Join bertrik [0] (
08:53:28 Quit bertrik (Changing host)
08:53:28 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
09:07:56 Quit Scromple (Quit: Leaving)
09:08:38 Quit bertrik (Quit: And That, My Liege, Is How We Know the Earth to Be Banana Shaped)
09:10:17 Quit froggyman (Ping timeout: 255 seconds)
09:10:35 Join markun [0] (~markun@rockbox/developer/markun)
09:10:46 Join froggyman [0] (~seth@unaffiliated/froggyman)
09:21:26 Join hilbert [0] (
09:28:48 Quit hilbert (Quit: Computer has gone to sleep.)
09:30:32 Join hilbert [0] (
09:32:39 Join petur [0] (~petur@rockbox/developer/petur)
09:44:15 Quit Guinness (Ping timeout: 248 seconds)
09:44:45 Join pamaury [0] (
09:44:45 Quit pamaury (Changing host)
09:44:45 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
09:45:23 Join kadoban [0] (
09:45:46 Join Keripo [0] (
09:49:50 Quit Keripo (Ping timeout: 244 seconds)
10:01:23***Saving seen data "./dancer.seen"
10:17:54 Join wodz|work [0] (
10:19:53 Join Guinness [0] (
10:22:04 Part Zagor
10:22:13 Join mortalis [0] (
10:26:26 Join Zagor [242] (~bjst@rockbox/developer/Zagor)
10:29:34 Join n17ikh [0] (
10:41:38 Quit wodz|work (Quit: CGI:IRC (EOF))
10:56:25kugelamiconn: what would help is actual bug reports on FS
10:57:46kugelZagor: calling it printf() in struct screen is calling for touble with system headers, isn't it? is printf necessary?
10:58:42Zagorwe have tons of name collisions already. just grep for "index".
10:59:04Zagorprintf is the normal name for formatting output functions
10:59:19kugelbut this printf has two params more
10:59:47kugelputsf() was just fine
10:59:55kugeland even shorter
11:01:49Zagorwell, it's not a puts. it's a printf.
11:01:51 Join dfkt [0] (dfkt@unaffiliated/dfkt)
11:03:04kugelwhat's the difference?
11:03:34Zagorputs outputs a string. printf formats a string.
11:03:41Zagor# man puts
11:04:18Zagor"printf outputs a formatted string", I mean
11:04:21kugeli suggested putsf to denote the formatting character
11:04:52kugeland I didn't know puts is a standard function (in linux)
11:05:38Zagorit's C
11:05:50kugeloh right, sorry
11:06:52kugelwell, I'm not really about correctness or not, I just would like to avoid name clashes with standard names if possible
11:07:29kugelthose were a big pita in my raaa gsoc project
11:07:59Zagoryes, if it was a standalone function it would cause problems. but now it's just a pointer in a struct, so it doesn't really collide with anything.
11:08:24Zagorat least not any more than before
11:08:44kugelthat's not my experience. see at the top of plugin.h
11:09:15kugelit's still a problem if the system header do things like #define printf XX
11:11:21Zagoras I said, we already have precisely the same collition today with screen.puts. it has worked for ~20000 revisions.
11:11:21kugelthe top of plugin.h has a workaround for platforms that do #define strcmp
11:11:53Zagorso unless you identify an actual problem, I don't see the point of this argument
11:12:04kugelokay, I see that and I wasn't aware of that yet. just saying it would perhaps be better for new code to avoid ti
11:12:35kugel(besides, our puts has a matching signature, which helps)
11:12:41kugelah, ignore that
11:23:11kugelhm, now there's a lot of calls in the (0, line++ * height, ...) style
11:24:02Zagorthat's temporary. we need some sort of text area widget to get away from pixel positioning.
11:24:13Zagorbut I haven't started with that yet
11:24:32kugelah interesting
11:44:51 Quit bluebrother^ (Read error: Connection reset by peer)
11:44:51 Quit fs-bluebot (Read error: Connection reset by peer)
11:45:58amiconnkugel: Optimisation possibility: Make 'line' refer to pixels and use line += height
11:46:42kugelamiconn: I was more concerned about the duplication
11:47:54 Quit hilbert (Quit: Textual IRC Client:
11:49:17Zagoramiconn: that requires separate statements, since the value of "line += height" is the result post-add.
11:49:28Zagorthat's why I did it like this
11:49:38 Join fs-bluebot [0] (
11:49:49amiconnStill better than multiplying every time
11:50:00 Join bluebrother [0] (
11:50:00 Quit bluebrother (Changing host)
11:50:00 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
11:50:46Zagornot for this
11:51:17kugelcould adjust the initial. otoh gcc is probably able to figure that out anyway
11:52:10Zagorthat will never be committed to trunk, so I wouldn't spend time worrying about how efficient it is
12:01:25***Saving seen data "./dancer.seen"
12:05:51kugelZagor: if you're at it :) what would be nice is if icons were a proper part of the line.
12:06:34kugele.g. drawing the selector over (or under rather) the icons or line indentation
12:07:22kugelright now a list line is assembled of icons (apps code) and text+selector (firmware code)
12:11:33kugelexcept the selector part is apps also if it's the cursor selector
12:24:03 Quit Rob2223 (Quit: Rob2223)
12:24:37 Join Rob2222 [0] (
12:28:50 Join ben22 [0] (
12:30:59ben22I have been following a guide to 'triple boot' my iPod 3g with linux and rockbox. Having followed the guide, I now get the apple logo, followed by an image of folder with an exclamation mark superimposed on it, and then if the iPod is plugged in to my PC it will enter disk mode.
12:31:10ben22Does anybody know what causes this problem?
12:33:57 Join hilbert [0] (
12:50:06Zagorkugel: yeah, I ... have a few other things first :-)
12:52:26 Quit [Saint] (Read error: Connection reset by peer)
13:02:18 Quit ben22 (Quit: Leaving)
13:38:19 Quit hilbert (Quit: Textual IRC Client:
13:44:08 Join hilbert [0] (
14:01:11Siliciumkugel: btw, i found the problem
14:01:27***Saving seen data "./dancer.seen"
14:01:50Siliciumkugel: it works fine with 2.11 firmware (cowon firmware which will be patched) but from 2.13 it wont detect any partitions
14:01:57Siliciummaybe they altered the partition on 2.13
14:02:02Siliciumor filesystem
14:02:29 Join robin0800 [0] (
14:03:15 Join [Saint] [0] (~Saint]@unaffiliated/saint/x-8516940)
14:06:07 Quit hilbert (Quit: Textual IRC Client:
14:15:08 Join petur2 [0] (~petur@
14:15:13 Quit petur2 (Changing host)
14:15:13 Join petur2 [0] (~petur@rockbox/developer/petur)
14:15:31 Nick petur is now known as Guest87424 (~petur@rockbox/developer/petur)
14:16:00 Nick petur2 is now known as petur (~petur@rockbox/developer/petur)
14:18:28 Join ben22 [0] (
14:18:36 Quit Guest87424 (Ping timeout: 252 seconds)
14:20:24ben22I have managed to install and run Rockbox on an iPod 3g. I have a dual firewire / usb connector for this iPod, and when connecting it only to firewire, the iPod reboots into disk mode, which prevents me from charging whilst listening to music. Is this behaviour caused by the cable, or the software? If it is software, is there a way to disable it?
14:21:42Torneben22: we don't suppoort firewire
14:21:50Torneso, it has to reboot to disk mode to do it.
14:21:57Tornesorry :)
14:22:13TorneOh, if you just want to charge taht should be possible
14:22:19Tornesee the manual
14:22:28ben22Torne, Ok, thanks.
14:22:37Tornehold menu while inserting the cable
14:22:39Tornei think
14:22:42Torneworks for usb on other ipods :)
14:25:38ben22Unfortunately there is nothing relevant I can find for 'charge' or 'charging' in the manual, and holding 'menu' has no effect
14:26:01[Saint]Any button.
14:26:20[Saint]Its not been "just menu" for a while now.
14:26:58[Saint]Holding any key except hold should enter "charge only" mode on the ipods.
14:27:12Torne[Saint]: ipod3g is all weird, though; it's possible it doesn't work
14:27:25Torneipod3g is the one with both usb and firewire, but we support neither and so reboot to diskmode
14:27:51Torne(we don't support firewire at all, and we don't know how to drive usb on ipod3g specifically, it's not the same usb controller as later ipods)
14:27:55[Saint]Ah, right. Yeah I forgot about the need to reboot.
14:28:05Torneit should in *theory* still work the same, though :)
14:28:07ben22Would a stripped down firewire cable work? with just voltage applied, nothing else?
14:28:11Torneben22: yes
14:28:24Tornei'm reading the code atm, gimma s ec
14:28:42 Join hilbert [0] (
14:29:01ben22The battery on this one is cooked, looking into an external battery pack. Can't listen without some form of power other than internal battery.
14:29:20Torneyeah this looks like it may have been broken when someone refactored usb stuff
14:30:03Torneit's not obvious how this is supposed to work and it's not obvious that it does work :)
14:31:22Torneactualyl this looks *very* weird.
14:31:33Tornewhen firewire is detected it posts USB_REQUEST_REBOOT
14:31:44Tornewhich does if(usb_reboot_button()) try_reboot();
14:32:09Torneusb_reboot_button always returns false on devices that don't ahve USB power
14:32:13Tornewhich seems wrong :)
14:32:26Tornethat seems like it should *never* reboot to disk mode
14:32:28Tornenot always
14:33:05Torneben22: raise a bug
14:33:10Torneunless one exists alreaday
14:33:16Tornethis seems like it was probably broken bya ccident at some point
14:33:44Tornethe fake-usb firewire handling is weird and complicated and only used by very old ipods that few people have, so it's not a surprise you are the first one to notice :)
14:35:54ben22I will try a power only firewire connection before posting a bug. Thank you for your quick replies.
14:36:06Torneno, please post a bug anyway
14:36:15Torneholding a button on insert is supposed to work
14:36:15[Saint]Please still post a bug, though.
14:36:28TorneI doubt a power only cable will make any difference
14:36:41Tornei don't think we have enough of a firewire stack to detect the host not being there
14:36:50Tornewe *probablky* just trigger it by getting 12V on the firewire power pin
14:37:03Tornebut even if it does, file a big anyway :)
14:43:23ben22OK, #13240 filed
14:44:05ben22er, #12340 I mean
14:44:27[Saint]Nice number :)
14:48:27ben22OK, thanks for your help. I look forward to being able to use Rockbox :)
14:48:32 Quit ben22 (Quit: Leaving)
14:51:38 Join MethoS- [0] (~clemens@
15:01:09 Quit antil33t (Read error: Connection reset by peer)
15:01:32 Join antil33t [0] (
15:08:06 Quit antil33t (Read error: Connection reset by peer)
15:08:32 Join antil33t [0] (
15:20:00 Join y4n [0] (y4n@unaffiliated/y4ndexx)
15:22:37 Quit robin0800 (Remote host closed the connection)
15:26:22 Join Sundiver [0] (~angel@
16:01:28***Saving seen data "./dancer.seen"
16:07:47 Quit mikroflops (Ping timeout: 240 seconds)
16:12:43 Quit mortalis (Quit: CGI:IRC)
16:22:43 Join mikroflops [0] (
16:28:01 Join ChickeNES [0] (~ChickeNES@
16:29:33 Quit ChickeNES (Client Quit)
16:30:28 Join ChickeNES [0] (~ChickeNES@
16:30:39 Nick petur is now known as geveaerts (~petur@rockbox/developer/petur)
16:30:49 Nick geveaerts is now known as petur (~petur@rockbox/developer/petur)
16:31:17 Nick petur is now known as gevearts (~petur@rockbox/developer/petur)
16:33:45 Nick gevearts is now known as not_gevearts (~petur@rockbox/developer/petur)
16:34:01 Nick not_gevearts is now known as gevearts (~petur@rockbox/developer/petur)
16:34:17 Nick gevearts is now known as petur (~petur@rockbox/developer/petur)
16:35:13 Nick petur is now known as gevearts (~petur@rockbox/developer/petur)
16:35:37 Nick gevearts is now known as petur (~petur@rockbox/developer/petur)
16:44:31CIA-14New commit by zagor (r30803): Charcell adaptation.
16:45:16 Nick Zagor is now known as would (~bjst@rockbox/developer/Zagor)
16:45:48 Nick would is now known as Zagor (~bjst@rockbox/developer/Zagor)
16:46:11 Nick [Saint] is now known as would (~Saint]@unaffiliated/saint/x-8516940)
16:46:37 Nick would is now known as [Saint] (~Saint]@unaffiliated/saint/x-8516940)
16:48:01 Nick [Saint] is now known as would (~Saint]@unaffiliated/saint/x-8516940)
16:48:26 Nick would is now known as [Saint] (~Saint]@unaffiliated/saint/x-8516940)
17:01:51 Part Zagor
17:03:36 Join T44 [0] (
17:04:28 Quit Sundiver (Quit: leaving)
17:05:59 Quit fyre^OS (Read error: Connection reset by peer)
17:07:45 Quit Topy (Ping timeout: 256 seconds)
17:08:43 Quit T44 (Ping timeout: 252 seconds)
17:09:58 Join ffowcs [0] (
17:10:04 Join Topy44 [0] (
17:19:46 Quit ChickeNES (Quit: Computer has gone to sleep.)
17:20:43 Join domonoky [0] (~Domonoky@rockbox/developer/domonoky)
17:34:03 Join benedikt93 [0] (~benedikt9@unaffiliated/benedikt93)
17:35:10 Quit Topy44 (Ping timeout: 240 seconds)
17:36:20 Join Topy44 [0] (
17:50:53 Join mortalis [0] (
17:54:34 Quit GodEater_ (Quit: Page closed)
18:01:32***Saving seen data "./dancer.seen"
18:04:42 Quit petur (Quit: *plop*)
18:05:55 Join T44 [0] (
18:07:58 Quit Topy44 (Ping timeout: 240 seconds)
18:08:56ffowcsGood morning!
18:09:22ffowcsCan anyone tell me if it is possible to map WPS features to different buttons or disable them entirely?
18:15:10 Join ChickeNES [0] (~ChickeNES@
18:18:45 Join Buschel [0] (
18:20:35 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
18:21:00 Join freddyb [0] (
18:28:06 Join thomasjfox [0] (~thomasjfo@rockbox/developer/thomasjfox)
18:31:06BuscheljhMikeS: hi, it seems r30546 / r30549 strikes back... my fix for the iPod Video USB results in just another USB issue on e200 :(
18:31:20Buschel-> FS #12324
18:31:21fs-bluebot e200v1 USB connection problems since r30701 (bugs, new)
18:33:35CIA-14New commit by fredwbauer (r30804): Revert skin font allocation size to the behavior prior to r30589. i.e. Default to 256 glyphs rather than trying to load MAX_FONT_SIZE. This matches ...
18:35:16CIA-14r30804 build result: All green
18:59:27 Quit ffowcs (Quit: Leaving)
19:01:49 Quit ChickeNES (Read error: Connection reset by peer)
19:01:56 Join ChickeNES [0] (~ChickeNES@
19:02:11 Quit bieber (Ping timeout: 240 seconds)
19:03:44 Quit ChickeNES (Read error: Connection reset by peer)
19:03:51 Join ChickeNES [0] (~ChickeNES@
19:07:22 Join fatline [0] (
19:07:37 Join bieber [0] (
19:08:58gevaertsBuschel: could you try calling usb_core_init() instead of usb_drv_init() from usb_drv_attach()? I don't think that's a good final fix, but I suspect it might well make things work again for everyone
19:09:21 Quit amiconn (Remote host closed the connection)
19:09:21 Quit pixelma (Remote host closed the connection)
19:09:28gevaertsusb_core_init() does call usb_drv_init(), but it checks first if it has done that already
19:09:31 Join robin0800 [0] (
19:10:37gevaertsA good final fix would involve simplifying the code
19:13:45 Quit Buschel (Ping timeout: 252 seconds)
19:14:02gevaertsmc2739: could you also try that?
19:14:37*gevaerts doesn't really understand yet why that would be needed, but most other drivers do this (sometimes via usb_enable())
19:15:07 Join Buschel [0] (
19:16:17 Join amiconn [0] (quassel@rockbox/developer/amiconn)
19:16:19 Join pixelma [0] (quassel@rockbox/staff/pixelma)
19:16:21 Quit ChickeNES (Read error: Connection reset by peer)
19:16:40 Join ChickeNES [0] (~ChickeNES@
19:19:18 Quit fatline (Quit: Leaving)
19:19:55 Quit ChickeNES (Read error: Connection reset by peer)
19:20:14 Join ChickeNES [0] (~ChickeNES@
19:22:12 Quit ChickeNES (Client Quit)
19:24:36 Join Horscht [0] (
19:24:36 Quit Horscht (Changing host)
19:24:36 Join Horscht [0] (~Horscht@xbmc/user/horscht)
19:25:48Buschelgevaerts: using usb_core_init() doesn't work for me
19:31:21 Join TheLemonMan [0] (
19:36:24 Quit dfkt (Quit: -= SysReset 2.55=- Sic gorgiamus allos subjectatos nunc.)
19:40:56 Quit hilbert (Quit: Textual IRC Client:
19:41:22 Join hilbert [0] (
19:41:37Buschelgevaerts: some timing issue?
19:43:14 Quit hilbert (Client Quit)
19:43:37 Join hilbert [0] (
19:45:33 Quit hilbert (Client Quit)
19:45:50 Join hilbert [0] (
19:47:01 Quit Staphylo (Ping timeout: 252 seconds)
19:47:14 Join Staphylo [0] (
19:47:34 Quit froggyman (Ping timeout: 252 seconds)
19:47:47 Quit hilbert (Client Quit)
19:48:08 Join hilbert [0] (
19:48:16 Quit hilbert (Remote host closed the connection)
19:48:38 Join hilbert [0] (
19:49:14 Quit markun (Ping timeout: 260 seconds)
19:49:44 Quit hilbert (Remote host closed the connection)
19:50:07 Join hilbert [0] (
19:52:47 Join froggyman [0] (~seth@unaffiliated/froggyman)
19:57:58gevaertsBuschel: maybe, but I still don't really see it.
19:59:27gevaertsBefore usb_drv_init() nothing should happen apart from going to POWERED (because the driver can't do much yet), and after that you shouldn't need another usb_drv_init()
20:01:24 Quit mortalis (Quit: CGI:IRC (EOF))
20:01:33***Saving seen data "./dancer.seen"
20:02:51Buschelthe logic seems to tell that usb_enable(true) −− this uses usb_core_init() −− is called before usb_attach() (= uses usb_drv_attach()).
20:03:08Buschelusing usb_core_init() in usb_drv_attach() -> no USB connection
20:03:22Buschelusing usb_drv_init() -> USB connection
20:03:36gevaertsYes, so why do you need usb_drv_init() twice? I don't understand...
20:04:02Buschelso, usb_enable(true) seems to be called before the HW is ready? something like that?
20:04:40Buschelusb_core_init() will avoid a 2nd initialization, but the first one wasn't successful
20:05:03Buschelat least on the iPod Video
20:06:11gevaertshm, usb_enable() does more than usb_init_drv()
20:07:41Buschelyes, and it always did
20:07:47*gevaerts nods
20:07:53gevaertsThat can't be the problem
20:07:54 Quit robin0800 (Quit: Leaving)
20:08:51 Quit hilbert (Quit: Textual IRC Client:
20:08:59gevaertsAh, a difference!
20:09:04Buschelusb_core_init / _exit are not thread safe, correct?
20:09:10 Join hilbert [0] (
20:09:22gevaertscorrect, I think
20:10:10gevaertsBut they're not supposed to be called from all over the place
20:10:51Buschelforget about that, was just a brain fart...
20:11:36Buschelyou've found a difference?
20:12:19gevaertsWell, ipods have HAVE_USB_CHARGING_ENABLE while sansas do not (that code hasn't been written yet...), which might make usb_core_init() behave a bit differently.
20:12:29gevaertsThe timeout there is 10 seconds though, so it shouldn't matter
20:13:00Buschelso, for a test I should disable this for my iPod Video?
20:13:25gevaertsmaybe. I doubt if it will make any difference though
20:14:06Buschelworth a try, compiling now
20:15:19gevaertsAnother thing that might be worth trying is increasing the timeout in usb_insert_int() in usb-fw-pp502x.c. That will (via a long way) delay the initial usb_drv_init() a bit more
20:15:28*gevaerts checks if his ipod has some battery
20:17:04gevaertsok, it will need to charge a bit before I can do any testing
20:17:14Buschelundefining HAVE_USB_CHARGING_ENABLE does not make a difference
20:18:10Buschelnow raising the timeout to HZ
20:21:06Buschelno change
20:22:51 Quit TheLemonMan (Quit: WeeChat 0.3.5)
20:25:38Buschelbtw, any idea what causes the deadlock when inserting USB after/during playback on the nano2G?
20:25:55*Buschel crosses finger gevaerts does not say "buflib"
20:26:07gevaertsI could say buflib if you like :)
20:26:18*Buschel knew it
20:26:57gevaertsOk, the interesting news is that I can't reproduce the issue with my 30GB video on linux
20:28:02Buschel? you commented out usb_drv_init() in usb_drv_attach() (usb-drv-arc.c) ?
20:29:27 Join dfkt [0] (dfkt@unaffiliated/dfkt)
20:30:48Buschelwin7 here
20:32:03gevaertshm, I should be able to find a win7 machine
20:32:53Buschelthat's good, my Linux is inside a VM. not the test environment we're searching for
20:34:03gevaertsNot really, no
20:39:30gevaertsThat seems to work
20:39:43gevaertsI'm not working from latest svn though... Let me update
20:39:43 Join bertrik [0] (
20:39:44 Quit bertrik (Changing host)
20:39:44 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
20:44:32gevaertsBuschel: still works. Just to make sure, is what I'm testing with
20:44:57Buschelthat's correct
20:46:14Buscheldo you inster USB after startup or during playback?
20:46:47gevaertsafter startup
20:47:10Buschelsame here
20:48:28 Quit factor (Read error: Connection reset by peer)
20:49:04*gevaerts can't really think of other possible factors
20:49:28*Buschel just changed the WPS to cabbiev2
20:49:37 Quit jhMikeS ()
20:49:40Buschelstill does not work :)
20:50:15gevaertsDo you have HID enabled?
20:50:23Buschelno, it's off
20:51:25gevaertsNot that it should make any difference...
20:53:44Buschelif I enable HID win7 gives me a warning popup that the required drivers haven't been installed
20:53:53Buschelat least win7 gives a reaction
20:54:29 Join mudd1 [0] (
20:55:40gevaertsYou do get the USB screen?
20:56:35*gevaerts was looking at usb_drv_usb_detect_event
20:57:10Buschelah, wait. you're talking of the rockbox USB screen? -> yes. that is always visible
20:57:46 Join jhMikeS [0] (
20:57:47 Quit jhMikeS (Changing host)
20:57:47 Join jhMikeS [0] (~jethead71@rockbox/developer/jhMikeS)
20:58:25gevaertsI'd prefer if you didn't. The rockbox USB screen means you did get at least one valid USB descriptor request
20:58:32gevaertsWhich means the controller works...
21:00:06gevaertsWell, obviously, since changing usb_drv_attach() changes behaviour
21:03:42gevaertsthe relevant usb_attach() should be the one in usb_slave_mode(). usb_configure_drivers() only calls that if nothing wants exclusive access, which we know isn't the case (both by what we expect and by the fact that you get the usb screen, which you only get if usb_configure_drivers() returns true)
21:03:53 Quit Xerion (Read error: Connection reset by peer)
21:07:03 Join Xerion [0] (
21:08:11gevaertsBuschel: could you enable logf for usb_core.c?
21:11:06Buschelrecompiling now
21:11:29 Join icarusfactor [0] (~factor@
21:12:18 Join robin0800 [0] (
21:12:27 Nick icarusfactor is now known as factor (~factor@
21:13:13gevaertsAfter that, maybe make a copy of usb_drv_init() and only enable some bits in there to find out what we're missing? I suspect the hardware stuff is fine (but let's wait for the core logf output...), which would leave init_control_queue_heads() and/or memset(td_array)
21:15:45Buschelwhere would this log file be placed?
21:16:37gevaertsIn the root I think, after you select the right option (dump logf output?) in the debug menu
21:16:52gevaertsi.e. it collects first, then you save to disk
21:16:57*Buschel slaps forehead
21:17:05Buschelwait a minute
21:27:39gevaertsOK, that looks like a full enumeration
21:27:52Buschelthis shows startup -> usb insert -> usb remove -> dump
21:29:54gevaertsBuschel: ok. I'd try just init_control_queue_heads();memset(td_array, 0, sizeof td_array); instead of the full init in usb_drv_attach()
21:32:17Buschellet's see
21:33:50BuschelUSB does not connect ->
21:35:51*gevaerts doesn't see it at all then :\
21:40:08Buschelso, I'll just #ifdef it for IPOD_VIDEO for now and submit to have a working svn
21:42:22gevaertsThe thing is, usb_drv_init() really shouldn't be called twice...
21:45:02gevaertsBuschel: can you also enable logf in usb_storage.c?
21:51:36CIA-14New commit by learman (r30805): Update Swedish translation.
21:51:47 Join Lear [0] (
21:53:15CIA-14r30805 build result: All green
21:57:49Buschelgevaerts: any idea? any further requests?
22:01:34***Saving seen data "./dancer.seen"
22:02:41gevaertsNot really. I don't see anything
22:05:32 Quit y4n (Quit: 6,000,000 ways to die — choose one.)
22:07:06Buschelhowever, I will need to get some sleep now. thanks for your support, gevaerts!
22:07:20 Quit Buschel (Quit: ChatZilla 0.9.87 [Firefox 7.0.1/20110928134238])
22:17:17 Quit thomasjfox (Remote host closed the connection)
22:19:09 Quit mc2739 (Quit: leaving)
22:20:14 Join Strife89 [0] (~Strife89@
22:22:40amiconnTorne: On ipod 1st and 2nd Gen power-only firewire (i.e. the original firewire charger) works as intended (i.e. just charges and doesn't reboot)
22:22:56amiconnIt's possible that it's broken on 3rd Gen - can't test that
22:31:22 Quit Unhelpful (Quit: - Chat comfortably. Anywhere.)
22:31:43 Join Unhelpful [0] (~quassel@rockbox/developer/Unhelpful)
22:42:38 Quit Lear (Quit: ChatZilla 0.9.87 [Firefox 9.0a2/20111019042022])
22:45:26 Quit benedikt93 (Quit: Bye ;))
22:45:29bluebrotherhmm, let's see if it's possible to install Rockbox on the beast via MTP ... the SA9200 did accept the file I was sending. The only problem is that it didn't like flashing it.
22:46:00Lalufubertrik: I've got rid of the serbuf_lock in IAP
22:48:49gevaertsjhMikeS: the weird thing is that Buschel's video gets through enumeration without r30701, it just doesn't go further
22:49:42 Join stephen_ [0] (~stephen@
22:51:18CIA-14New commit by bluebrother (r30806): Remove svn keyword lines from sources.
22:51:40jhMikeSgevaerts: I recall he said the USB screen shows up, so yeah, it gets pretty far the then poops out
22:52:54CIA-14r30806 build result: All green
22:53:51CIA-14New commit by bluebrother (r30807): Rockbox Utility: add GoGear SA9200 as disabled target.
22:55:08gevaertsjhMikeS: it would be useful if we found an actual clue :)
22:55:34CIA-14r30807 build result: All green
23:00:00jhMikeSgevaerts: Don't look at me. :-) Well, I could try to requisition my sister's 30 gig video and see.
23:00:30*gevaerts looks at jhMikeS anyway!
23:01:00*jhMikeS grabs some garlic
23:03:03gevaertsThe thing seems to go through enumeration without problems, but when the host is supposed to do this control request to get the number of LUNS, we see nothing. That's still EP0, so it uses exactly the same flow we used during enumeration
23:03:51*jhMikeS wonders why he gets double fs mailings for any task he's posted on...which is a recent phenomenon
23:04:23jhMikeSone from rockbox-sf and another to his own email
23:04:40jhMikeSs/one from/one to/
23:07:01gevaertsI've always had separate mails for watched tasks
23:07:17jhMikeSif one's already on the list, they probably don't need the personal notification
23:07:51jhMikeSIt started a few months ago, but I wasn't sure of the pattern
23:08:58gevaertsI like getting the duplicates. The "personal" ones go to my inbox, and sometimes I clear out the rockbox-sf mailbox without really looking
23:16:57 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
23:18:50jhMikeSIs there a setting to suppress that?
23:20:05 Part domonoky
23:45:52CIA-14New commit by mc2739 (r30808): Update english-us translation.
23:46:54 Quit factor (Read error: Connection reset by peer)
23:47:45CIA-14r30808 build result: All green
23:50:03CIA-14New commit by bluebrother (r30809): Rockbox Utility now requires voice-corrections.txt. ...
23:51:41CIA-14r30809 build result: All green
23:56:20 Quit mc2739 (Quit: leaving)
23:56:28 Join factor [0] (~factor@
23:57:36 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
23:57:48 Quit bertrik (Quit: And That, My Liege, Is How We Know the Earth to Be Banana Shaped)
23:58:45 Quit pamaury (Remote host closed the connection)

Previous day | Next day