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 2010-09-20

00:00:16puetzkfunman: I had another thought that seems to be right: since the AGC and the "Limiter Gain Overdrive" are the only things in the spec with resolution of 0.375dB, I bet they are the same thing
00:00:28puetzkWhich says the AGC never turns it up, just turns it down to avoid clipping
00:00:31puetzkwhich also seems sensible
00:01:27puetzkThat would imply that the mic ouptput is way too hot to start with, even though the current code keeps the preamp as low as it can
00:01:53 Quit mikroflops (Read error: Operation timed out)
00:01:54saratogaoh wow this wma code is a mess
00:03:07puetzkI see in the code that the Mic supply is turned only only for AS3543; do you know what the value for VMICS in AUDIOSET2 was chosen based on?
00:03:26puetzk(it's currently 0, or so maybe it's based on "never got set")
00:06:43saratogaapps/codecs/lib/ffmpeg_get_bits.h:435:44: warning: variable ‘re_cache’ set but not used [-Wunused-but-set-variable]
00:06:49 Join mikroflops [0] (
00:06:50saratogathat one sounds like it probably impacts performance
00:07:45 Quit bertrik (Quit: :tiuQ)
00:07:53 Join fdinel [0] (
00:09:15CIA-81New commit by saratoga (r28118): Remove some unused variables.
00:11:03CIA-81r28118 build result: All green
00:20:35 Quit Jerom (Read error: Connection reset by peer)
00:21:28 Join Jerom [0] (~jerome@
00:25:06n1ssaratoga: that one is still in ffmpeg svn so i didn't look much closer at it since i guess they'll fix it soonish but why should it affect performance, if gcc can figure out it's not used it should just delete it
00:25:35saratogan1s: well gcc 4.6 can figure it out, does 4.4?
00:25:36 Quit TheSeven (Ping timeout: 245 seconds)
00:26:29 Quit Jerom (Quit: Leaving.)
00:26:57 Join Jerom [0] (~jerome@
00:27:25n1scan't know that without looking at the generated asm of course, but we could just comment it out
00:28:06 Quit robin0800 (Remote host closed the connection)
00:28:36 Join robin0800 [0] (~robin0800@
00:30:40n1sit's only in the skip_bits function, and that shouldnt' be called that often
00:37:01 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
00:37:10n1swell, it's because of their macros, the OPEN_READER thing creates the unused var and then the UPDATE_CACHE assigns it some data from memory but they even say in the comment that they rely on compiler optimizations to remove all this stuff as the function could be just one simple line as long as we use the "ALT_BITSREAM_READER"
00:37:39n1sso i think it can be left alone untill ffmpeg fixes it or we want to switch to 4.6
00:41:20 Quit Jerom (Quit: Leaving.)
00:41:30 Quit n1s (Quit: Lmnar)
00:44:50 Quit ender` (Quit: I think I remember an episode of MacGyver where he overthrew a violent dictator with a rubber band, 2 bottle caps, and some navel lint.)
00:53:01 Quit funman (Quit: Lost terminal)
01:04:47***Saving seen data "./dancer.seen"
01:27:38 Quit robin0800 (Remote host closed the connection)
01:28:06 Join robin0800 [0] (~robin0800@
01:38:34 Quit robin0800 (Remote host closed the connection)
01:53:06 Quit mc2739 (Ping timeout: 272 seconds)
01:54:53 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
01:59:13 Quit _____s1gma (Quit: Leaving)
02:01:46 Join JdGordon| [0] (
02:01:46 Quit JdGordon| (Changing host)
02:01:46 Join JdGordon| [0] (~jonno@rockbox/developer/JdGordon)
02:08:40 Quit kugel (Remote host closed the connection)
02:28:53 Quit Judas_PhD (Quit: This is a quitting message)
02:32:21 Join krabador [0] (
02:44:38 Join webguest10 [0] (
02:45:12 Join Rondom [0] (
02:45:52 Quit Rondom_ (Read error: Operation timed out)
02:51:03 Quit webguest10 (Quit: CGI:IRC)
02:55:18 Join Judas_PhD [0] (
03:01:13 Join Evilnick_ [0] (
03:02:34 Quit evilnick (Ping timeout: 252 seconds)
03:04:50***Saving seen data "./dancer.seen"
03:07:17 Join hebz0rl_ [0] (
03:11:28 Quit hebz0rl (Ping timeout: 272 seconds)
03:17:38 Quit Zambezi (Read error: Connection reset by peer)
03:18:38 Join Zambezi [0] (Zulu@
03:56:35 Join anewuser [0] (anewuser@unaffiliated/anewuser)
04:12:38 Quit panni_ (Read error: Connection reset by peer)
04:29:35 Join moparx [0] (~moparx@unaffiliated/moparx)
04:32:32 Quit puetzk (Quit: puetzk)
04:33:04 Quit amiconn (Disconnected by services)
04:33:04 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
04:33:24 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
04:33:49 Quit pixelma (Disconnected by services)
04:33:51 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
04:33:53 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
04:38:28 Quit krabador (Quit: Sto andando via)
04:39:48docgnomedoes rockbox handle multiple artists in id3 tags?
04:40:24JdGordon|depends how they are listed I tihnk
04:41:23docgnomeany idea which way works? i think i'm supposed to use null characters in 2.4... no idea how to add those to a tag though
04:43:11JdGordon|you'd have to try it and see... null seperator is more likely to not work
04:56:24 Quit Barahir (Ping timeout: 240 seconds)
04:56:32 Part moparx
04:58:13 Join Barahir [0] (
05:04:52***Saving seen data "./dancer.seen"
05:09:31 Quit balintx (Ping timeout: 240 seconds)
05:12:41 Quit krazykit (Ping timeout: 276 seconds)
05:13:44docgnomeyeah... / didn't work
05:17:37JdGordon|It only shows the first author?
05:19:04docgnomeit shows arist a/artist b
05:19:28JdGordon|and you were expecting?
05:19:33JdGordon|that looks like it worked to me?
05:19:34docgnomei want a separate entry for each
05:19:53docgnomeno cause if i have stuff later that just has artist a on it i now have two entries to find stuff with artist a
05:20:12JdGordon|you're talking about the database?
05:24:40docgnomealso using rockbox 3.6
05:33:56docgnomealso the file browser seems to think that a bunch of files i just loaded are not there
05:34:02docgnomebut they show up and play in the db
05:34:04 Quit ps-auxw (Ping timeout: 240 seconds)
05:37:19docgnomesame problem in the current build, with the multi-artist tag i mean
05:41:33 Quit Horscht (Quit: Verlassend)
05:45:34 Join ps-auxw [0] (
05:48:02saratogai was pretty sure we didn't support multiple artists
05:48:45JdGordon|if they are showing up nicely like the above then we could easily support then in the skins :)
05:48:52JdGordon|the db is a different beast though
05:56:30 Quit Judas_PhD (Quit: This is a quitting message)
06:02:55 Quit clone4crw (Ping timeout: 252 seconds)
06:10:51 Join bluebrother [0] (
06:10:51 Quit bluebrother (Changing host)
06:10:51 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
06:14:35 Quit bluebroth3r (Ping timeout: 276 seconds)
06:17:46 Quit Dreamxtreme (Ping timeout: 252 seconds)
06:22:28 Quit crwl (Quit: reboot)
06:22:29 Join Dreamxtreme [0] (~Dreamxtre@
06:27:46 Join crwl [0] (
06:37:39 Join balintx [0] (
06:42:38 Quit balintx (Ping timeout: 264 seconds)
06:42:56 Quit Barahir (Ping timeout: 272 seconds)
06:55:42 Quit fdinel (Quit: Miranda IM! Smaller, Faster, Easier.
07:04:54***Saving seen data "./dancer.seen"
07:08:47 Quit sasquatch (Ping timeout: 255 seconds)
07:17:53 Quit anewuser ()
07:23:26 Quit BHSPitMonkey (Read error: Connection reset by peer)
07:30:48 Join Judas_PhD [0] (
07:36:15 Quit fxb__ (Read error: Operation timed out)
07:55:16 Join fxb__ [0] (
07:58:14 Join Barahir [0] (
08:06:23 Quit avacore (Ping timeout: 255 seconds)
08:07:21 Join avacore [0] (
08:11:46 Join solexx [0] (
08:15:13 Join sasquatch [0] (~username@
08:15:18 Quit solexx_ (Ping timeout: 276 seconds)
08:21:06 Join ender` [0] (
08:22:14 Quit sasquatch (Ping timeout: 264 seconds)
08:23:13 Join bmbl [0] (~Miranda@unaffiliated/bmbl)
08:24:00 Quit Judas_PhD (Quit: This is a quitting message)
08:26:33 Join Judas_PhD [0] (
08:28:08 Quit Judas_PhD (Client Quit)
08:29:50 Join Adubb [0] (~Aldubuc@
08:34:32 Join sasquatch [0] (~username@
08:34:57 Join Judas_PhD [0] (
08:41:19 Join wodz [0] (
08:41:24wodzJdGordon: ping
08:42:46wodzI came back to work on wm8750/51 bass setting based on your patch (UI part) and I have problems with setting saving
08:44:28wodzin config.cfg there is line 'bass:' which is never set to anything
08:45:08JdGordon|it gets written as empty?
08:46:12wodzthis is the part which is I think related
08:46:59wodzI don't understand this - is the setting saved as string?
08:47:06JdGordon|ah hehe, printf doesnt do what I thought it does obviously
08:47:20JdGordon|line 14 is probably writing the \0
08:48:08wodza yes you may be right
08:48:28wodzwhy not store just number?
08:49:01wodzit can be easly filtered elsewhere
08:49:05JdGordon|change it to anything other than *
08:49:10JdGordon|the code above it will handle it
08:49:54 Join LinusN [0] (
08:49:54 Quit LinusN (Changing host)
08:49:54 Join LinusN [0] (~linus@rockbox/developer/LinusN)
08:50:18wodzbut why you want to introduce special char in setting - thats the part I don't get
08:51:00JdGordon|the point was to know which mode it is in IIRC
08:51:01 Quit Judas_PhD (Quit: This is a quitting message)
08:51:28wodzyes but we have MSB bit set for this
08:51:44wodzso why introduce something else for saving config
08:51:47JdGordon|yeah, but that makes the number very user-unfriendly
08:52:23wodz* makes it better?
08:53:02JdGordon|*4 instead of 134217732
08:53:31wodzyou mean decimal representation yes
08:53:53wodzwhy there is += atoi(value) and not = atoi(value)
08:54:39JdGordon|because the MSB is set before that
08:54:45JdGordon|you'd need to change the code order for it to work
08:56:28wodzI don't get - val variable seem to be set and not used at all so how it is supposed to influence the setting?
08:56:55JdGordon|val is a pointer into global_settings
08:57:42JdGordon|now I see what you're saying... yeah var needs to get the MSB set, val doesnt need to be used at all
08:57:53*JdGordon| obviously didnt test this code...
08:57:59JdGordon|now that I could of course
08:58:07wodzhehe you said it worked in sim :P
08:58:18JdGordon|no i didnt! :D
09:03:34wodzJdGordon|: does it look right? (a for adaptive, l for linear)
09:04:46wodzI have also some other idea how to handle switching between modes -> maybe the best solution is to disable the setting on mode switch. This will force user to interact with setting after switch
09:04:55***Saving seen data "./dancer.seen"
09:04:59 Quit Guest72658 (Ping timeout: 252 seconds)
09:04:59 Quit T44 (Ping timeout: 265 seconds)
09:08:27 Join petur [0] (d408b802@rockbox/developer/petur)
09:09:23 Join Zagor [0] (~bjst@rockbox/developer/Zagor)
09:10:38wodzJdGordon|: it doesn't work now setting is not written to config at all
09:11:10 Join Rob2223 [0] (
09:11:36JdGordon|odd... it should :p
09:14:27wodzwhy wm875xbass_write_to_cfg is triggered on system startup?
09:14:56JdGordon|probably because it does the load then a save when it maybe doesnt need it
09:15:13 Quit Rob2222 (Ping timeout: 265 seconds)
09:15:53wodzis it a general pattern? If so such behaviour whould kill flash based palyers
09:16:26JdGordon|I highly doubt that
09:20:05wodzthis is tha backtrace
09:21:27JdGordon|nothing out of the ordinary ther
09:21:57JdGordon|it isnt like it is saving every single tick (which would be about how often it would need to trash a flash chip)
09:22:36wodzbut rewriting config on every startup is a waste
09:23:31 Join timc [0] (~tim@
09:23:36 Nick timc is now known as Guest28570 (~tim@
09:24:21wodzFS it?
09:24:45JdGordon|if you want, but its unlikely to ever be looked at...
09:25:42JdGordon|it probably isnt very hard to fix, its just not very high priority
09:26:30 Join Judas_PhD [0] (
09:28:10 Quit JdGordon| (Quit: leaving)
09:28:18wodzwhy the hell settings_load_config() call explicitly config_save() at the end????
09:31:07wodzIt looks dumb for me - I'll FS it
09:33:00 Join funman [0] (~fun@rockbox/developer/funman)
09:35:12 Join JdGord [0] (~jdg@
09:38:07wodzFS #11634
09:44:19 Quit wodz (Quit: Leaving)
09:49:02LinusNdoes anybody know what has happened with the CAPTCHA on the forum registration page?
09:49:43pixelmawhat's wrong with it?
09:50:44funmansometimes the letters are missing, just hitting reload fix that for me
09:56:26LinusNi had to press reload several times, and sometimes I get only one letter
09:56:42 Join haim_ [0] (~HaimN@
09:56:47 Quit haim_ (Client Quit)
09:57:05LinusNand for some reason, it very often gives the same sequence of letters
09:57:08 Join HaimN [0] (~HaimN@
10:00:29 Join kugel [0] (~AndChat@
10:00:34HaimNHi, I try to compile Rockbox and wen I type ../tools/configure and then the number of my player and N (for 'normal') I get the massege "../tools/configure: 2800: arm-elf-gcc: not found ../tools/configure: 2808: arm-elf-ld: not found [WARNING] The compiler you must use (arm-elf-gcc) is not in your path!"
10:01:01HaimNwhat I have to do?
10:01:21 Join swilde [0] (
10:01:33kugelhow old is your checkout?
10:01:48 Quit Adubb (Read error: Connection reset by peer)
10:02:26kugelwe use arm-elf-eabi-gcc since a while
10:03:37HaimNkugel: What do you mean "your checkout?"
10:04:23kugeleither way, you need the cross compiler in your $PATH. shows what I mean after building the compiler
10:05:13kugelyour svn checkout (copy) of the source code
10:05:29 Join einhirn [0] (
10:06:36HaimNI have Rockbox 3.6
10:07:35kugelget the latest code and run tools/
10:08:46HaimNRockbox 3.6 isn't the latest?
10:13:32 Join esperegu [0] (~quassel@
10:14:48StummiHaimN, how did you get the code?
10:25:11HaimNStummi: from here:
10:25:30HaimNand from here:
10:25:32HaimNsvn co svn:// rockbox
10:25:58HaimNI tried twice
10:26:21funmanif you absolutely want to build 3.6, you can use /tools/configure −−eabi
10:26:21Stummierm, that are two differend things
10:26:32funmanbut it's advised that you use svn
10:26:32Stummifor the first you don't need to compile anything
10:27:07Stummidid you made this "svn co" today?
10:28:05HaimNno, yesterday
10:28:21HaimNto do it again?
10:29:01Stummiok, that should do it again. It would only be a problem, if this would be a few weeks or months old
10:29:38gevaertsHaimN: did you actually install the crosscompilers?
10:29:40StummiHaimN, which OS do you use?
10:29:59HaimNi use ubuntu
10:30:39Stummiok, should work
10:30:47HaimNI work with this guide
10:31:44 Join JdGordon1 [0] (~jonno@
10:31:52gevaertsHaimN: do you have arm-elf-eabi-gcc in your path?
10:32:11StummiHaimN, you did also the part with
10:32:12HaimNhow i check it?
10:32:33HaimNStummi: yes
10:33:23StummiHaimN, which compiler did you choosed there?
10:33:55 Quit JdGordon (Ping timeout: 265 seconds)
10:34:42funmanhow do you check it? type "arm-elf-eabi-gcc"
10:35:00Stummido you have an folder "/usr/local/arm-elf-eabi/"?
10:38:06HaimNStummi: yes
10:39:04HaimNfunman: I get "arm-elf-eabi-gcc: command not found"
10:39:48StummiHaimN, what does "echo $PATH" says?
10:40:03Stummido you have a "/usr/local/bin/" there?
10:41:00gevaertsHaimN: did you run tools/ from the 3.6 source or from trunk?
10:41:03 Join DerPapst [0] (
10:41:10gevaertsThe installation path changed in between...
10:42:05HaimNI run both
10:42:38gevaertsthat's probably not a good idea...
10:43:22HaimNI will try to re-download and start again
10:43:38gevaertsre-download isn't needed. Cleaning up *is*
10:43:59gevaertsRemove /usr/local/arm-elf-eabi, and then run tools/ from *trunk* (not 3.6)
10:44:38gevaertsMaybe also remove everything that starts with arm-elf-eabi from /usr/local//bin
10:45:15 Quit JdGordon1 (Quit: Leaving.)
10:46:57HaimNi need to rmove also /usr/local/arm-elf ?
10:47:34StummiHaimN, better, delete your downloaded 3.6-source, so you can't accidently use this ;)
10:47:46Stummiand then work only with this source, you got with the "svn co"-command
10:48:05Stummiand yes, also remove /usr/local/arm-elf
10:49:01scorcheLinusN: i have been meaning to do an update on the forum software for a while now, but havent had the time...
10:50:13 Join wodz [0] (
10:51:39wodzfunman: Why do you think config_save() call in settings_load_config() was introduced in r2479? look at
10:52:42gevaertsHaimN: if you ever want to compile 3.6 (or older), also have build the old arm toolchain ('a')
10:53:50Stummiin most cases, you don't want this.
10:54:14funmanwodz: this commit just moves code around
10:54:26gevaertsStummi: maybe, but that's not for us to decide :)
10:54:33HaimNThank you all! Help me, I'll try to do what you said
10:55:14 Join DerPapst1 [0] (
10:55:48CIA-81New commit by funman (r28119): format() (and its alias vuprintf) return values are uncheck -> void
10:55:51 Quit DerPapst (Ping timeout: 264 seconds)
10:56:02wodzwould it be safe to just drop config_save() call?
10:56:11funmangevaerts: you can build eabi binaries with 3.6 (it's just not the default so you have to run configure −−eabi)
10:56:48funmanwodz: i imagined it would save the current config so modifications are not lost but i don't know this code at all
10:56:54gevaertsfunman: I know, but with the changed paths I think it's better to build the toolchains from trunk, to avoid problems later
10:57:17gevaertsfunman: oh, right. You meant something else...
10:57:20HaimNgevaerts: how i "Cleaning up *is*"
10:57:36 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
10:57:37CIA-81r28119 build result: All green
10:57:48gevaertsHaimN: I meant, cleaning up is needed. See the two lines after that
10:57:52wodzfunman: but what is the point in calling save_ form something which just loads config from file
10:58:08HaimNgevaerts :)
10:58:34gevaertsfunman: were all eabi-related bugs fixed before 3.6?
10:59:11funmanwodz: ah perhaps it should be moved before the call to this functions (only if/where needed of course)
10:59:14funmangevaerts: dunno
10:59:44gevaertsfunman: neither do I, which is why I'd recommend to not build 3.6 with eabi unless you really know what you're doing
11:00:00funmanif you're building 3.6 you don't know what you're doing
11:01:46 Quit JdGordon (Client Quit)
11:05:00***Saving seen data "./dancer.seen"
11:05:32 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
11:09:05 Quit JdGordon (Remote host closed the connection)
11:09:19 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
11:31:58 Quit wodz (Quit: CGI:IRC (Ping timeout))
11:33:58 Quit JdGordon (Quit: Leaving.)
11:43:08 Join JdGordon [0] (6fdcf25b@gateway/web/freenode/ip.
12:06:49 Quit funman (Read error: Operation timed out)
12:07:00 Part LinusN
12:11:05 Join funman [0] (~fun@rockbox/developer/funman)
12:39:43 Join LinusN [0] (~linus@rockbox/developer/LinusN)
12:45:28 Join giovanni_ [0] (~giovanni@
12:48:42 Quit Kohlrabi (Remote host closed the connection)
12:49:44 Join Kohlrabi [0] (
12:52:20 Quit kugel (Ping timeout: 240 seconds)
12:55:15 Join Topy44 [0] (
12:59:42 Join Jerom [0] (~jerome@
13:05:02***Saving seen data "./dancer.seen"
13:05:05funmananyone has a arm-elf-eabi gcc 4.5 around?
13:10:20 Join T44 [0] (
13:10:40 Quit Topy44 (Read error: Connection reset by peer)
13:24:29 Quit funman (Quit: leaving)
13:25:24Stummihm, something like swapping would be a funny thing in rockbox :)
13:26:02 Quit Jerom (Quit: Leaving.)
13:27:33 Join wodz [0] (
13:28:42wodzI think I found out why HD300 hangs and do not repond to BDM commands. There seem to be missing Vcore (1.8V). The question is why.
13:45:13 Quit DerPapst1 (Read error: Connection reset by peer)
13:45:30 Join DerPapst [0] (
13:46:05wodzAs of my understanding it should be safe to drop save_settings() line from settings_load_config()
13:52:39 Part LinusN
13:53:40 Quit AlexP (Remote host closed the connection)
13:53:40 Part Zagor
13:54:04 Join Zagor [0] (~bjst@rockbox/developer/Zagor)
14:01:20 Join AlexP [0] (~alex@rockbox/staff/AlexP)
14:04:33 Quit Bagder (Ping timeout: 276 seconds)
14:05:21 Join LinusN [0] (~linus@rockbox/developer/LinusN)
14:09:33 Quit JdGord (Ping timeout: 255 seconds)
14:09:57 Quit drizztbsd (Ping timeout: 240 seconds)
14:10:15 Join JdGord [0] (~jdg@
14:10:30 Quit JdGord (Client Quit)
14:12:20 Join JdGordon1 [0] (~jonno@
14:13:38 Quit JdGordon (Ping timeout: 252 seconds)
14:13:40 Join kugel [0] (~AndChat@
14:20:20JdGordon1wodz: just do it :)
14:20:49JdGordon1if it causes problems its not like its a big deal to revert
14:22:30wodzI am testing this in sim now
14:28:25wodzhow to apply settings in Graphical EQ screen?
14:28:33 Join anewuser [0] (anewuser@unaffiliated/anewuser)
14:30:07JdGordon1presumably they just apply when you accept the changes
14:30:14JdGordon1which could just mean leaving the screen
14:32:42wodzI don't see writes to config file from this screen (svn head) - I'll test this later
14:32:47 Quit wodz (Quit: Leaving)
14:39:19 Join krazykit [0] (
14:42:59giovanni_JdGordon1: what do you think about those issues?
14:46:40 Quit JdGordon1 (Ping timeout: 276 seconds)
15:05:05***Saving seen data "./dancer.seen"
15:17:24 Quit Guest28570 (Read error: Connection reset by peer)
15:20:44 Join DerPapst1 [0] (
15:21:39 Join jgarvey [0] (
15:23:05 Quit DerPapst (Ping timeout: 245 seconds)
15:28:55 Quit giovanni_ (Ping timeout: 252 seconds)
15:30:43 Join DerPapst [0] (
15:32:32 Join utanapischti [0] (~username@
15:32:58 Quit sasquatch (Ping timeout: 276 seconds)
15:33:07 Quit DerPapst1 (Ping timeout: 265 seconds)
15:33:55 Join DerPapst1 [0] (
15:34:29CIA-81New commit by wodz (r28120): Remove the call to settings_save() from settings_load_config() as it seems unneeded any more and causes unecessary writes on every startup.
15:36:50 Quit DerPapst (Ping timeout: 245 seconds)
15:38:13 Quit solexx (Quit: leaving)
15:38:21CIA-81r28120 build result: All green
15:42:26 Join [sko] [0] (~sko]
15:43:31amiconnMeh, wodz breaking rules :\
15:44:51 Part LinusN
15:48:07 Join wodz [0] (
15:48:24wodzamiconn: ping
16:00:53 Part Zagor
16:01:15 Join Zagor [0] (~bjst@rockbox/developer/Zagor)
16:04:20 Join domonoky [0] (~Domonoky@rockbox/developer/domonoky)
16:10:41HaimNHi, somone now were can I find in the scours code the file of the main munu of the player becaus I want to remove the "fm fadio" line (I don't want fadio in the player), thanks!
16:11:24 Quit AlexP (Remote host closed the connection)
16:13:15saratogafuze 8GB on woot
16:15:29gevaertsHaimN: the CONFIG_TUNER lines
16:15:47gevaertsin firmware/export/config/<your_player>.h
16:15:49HaimNwhen I did the compile I get the error:
16:15:49HaimN/home/haim/rockbox/apps/menus/recording_menu.c: In function recsource_func:
16:15:49DBUGEnqueued KICK HaimN
16:15:49HaimN/home/haim/rockbox/apps/menus/recording_menu.c:87: error: LANG_FM_RADIO undeclared (first use in this function)
16:15:49HaimN/home/haim/rockbox/apps/menus/recording_menu.c:87: error: (Each undeclared identifier is reported only once
16:15:53HaimN/home/haim/rockbox/apps/menus/recording_menu.c:87: error: for each function it appears in.)
16:15:55HaimN/home/haim/rockbox/apps/menus/recording_menu.c:87: warning: missing initializer
16:15:57HaimN/home/haim/rockbox/apps/menus/recording_menu.c:87: warning: (near initialization for names[1].string)
16:16:00HaimN/home/haim/rockbox/apps/menus/recording_menu.c:93: warning: implicit declaration of function radio_hardware_present
16:16:03HaimNmake: *** [/home/haim/rockbox/build/apps/menus/recording_menu.o] 1 הלקת
16:16:04pixelmastop it
16:16:12wodzHaimN: do you want to completely remove FM support? If so edit config file for the player
16:16:29wodzpixelma: You wanted me to look at something
16:16:38HaimNwodz: yes
16:16:57pixelmaHaimN: if you need to poste more lines, please use something like and just post the link here
16:17:00HaimNwaht i need to do?
16:17:27HaimNpixelma: ok
16:17:50 Join AlexP [0] (~alex@rockbox/staff/AlexP)
16:17:56wodzHaimN: look at firmware/export/config/
16:18:03 Join pamaury [0] (
16:18:03 Quit pamaury (Changing host)
16:18:03 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
16:18:21wodzfind .h for your player and edit it to not include FM support
16:18:32CIA-81New commit by pamaury (r28121): usb: remove unused variable and correctly stall on unhandled control requests
16:18:56Stummihm, an feature where a user can add shortcuts for a file to the main-menu would be cool. Has somebody already tried something like this?
16:19:00HaimNi did that but i got errore in compile
16:19:14pixelmawodz: there was an update to the Polish translation, funman already committed it though. I told him that I'd prefer asking a committer who is a native speaker to look at it first and also commit it - if we there is an active committer
16:19:34Stummiso, that i can choose this file directly on starting and it gots executed (if it is a binary) or opened with the viewer
16:19:50CIA-81r28121 build result: All green
16:22:55amiconnwodz: The settings_save() you removed looks like a leftover from when settings where saved in a reserved disk block
16:23:05amiconnThat's indeed no longer necessary
16:23:44pamauryis there any good reason why a long press in the simulator doesn't "power off"=exit
16:24:27Stummia long press on what?
16:24:34Stummiyou mean the power-button?
16:24:39pamauryon the button mapped to the power button
16:24:49pamauryit has been ifdef for month in the code
16:25:10leavittxpixelma: sorry, you askem me yesterday haven't i "accidentally enabled "absolute" mode / mouse control"... no. the settings says 3x3 grid, as usual... :(
16:26:45pixelmapamaury: my guess is that there is no implementation for the simulator to turn itself off
16:26:49 Join LinusN [0] (~linus@rockbox/developer/LinusN)
16:27:08pamauryhum, let's see
16:27:52wodzpamaury: this bit in simulator is missing - I hacked this some time ago but than massive code changes from kugel caused that I dropped the work
16:28:22pamauryyeah, kugel made all patches to the simulator obsolete recently :)
16:28:39AlexPgood, shut them all :)
16:30:34kugelso that should teach you to get patches into svn ASAP :-)
16:31:47HaimNwodz: if I remove "/* FM Tuner */ #define CONFIG_TUNER SI4700" Compilation does not work
16:31:47leavittxpixelma: i'm pretty sure that this behavior appeared only in recently revisions.
16:32:08CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
16:32:08*pamaury decides that he doesn't have the courage to touch the simulator code
16:32:34 Quit CGL (Ping timeout: 276 seconds)
16:32:55shuffle2hm, i just svn up'd, which caused nearly a full rebuild of rockbox code doesn't work :p a blocking usb send doesn't return, and all of rockbox freezes.
16:33:57shuffle2i don't think any of the recent commits did it, it's just a side-effect of recompiling...any idea what's up? :s
16:33:58evilnick_B(for the logs) There's a very minor error (I think) in the Fuze v2 manual. In 2.1 (Installation) it says "The Fuze v2 is available in multiple versions" where it ought to be "The Fuze is available in multiple versions" (i.e. no "v2")
16:34:34pamauryshuffle2: I just commit a change to usb_core but I would be surprised if it were responsible for the problem
16:34:50shuffle2it happened before that
16:35:28wodzHaimN: try editing appextra in tools/configure also
16:36:01 Join francesco_ [0] (
16:36:20kugelevilnick_B: it might be correct if it means that the released mkamsboot can't patch newer fuze v2s
16:36:22francesco_does anyone know how to change microSD1 from path into something else?
16:37:27pamauryfrancesco_: what do you mean ? changing the "<microSD1>" in the rockbox file browser ?
16:37:45evilnick_Bkugel: I agree, but it then goes on to describe how to tell v1 and v2 versions apart (the same text is used in the Fuze v1 manual, so I'm assuming that the playername macro is responsible)
16:37:54pamauryshuffle2: the blocking usb send means the host did not ack the transfer probably, which is rather strange
16:38:10francesco_pamaury: yes, just give it a different name
16:39:10AlexPevilnick_B: It should just say Fuze, but it says fuze v2 as it uses the \playertype{} macro, which for the fuze v2 is "Fuze v2"
16:39:31pamauryfrancesco_: no, it's not possible currently
16:40:00 Join domonoky1 [0] (
16:40:08francesco_pamaury: not even by editing code & recompiling (i've managed to disable some folders..)
16:40:11AlexPall of the ones that are xxx v2 will be a bit odd like this
16:40:30evilnick_BAlexP: As I suspected. In that case one fix would be to use regular text rather than the macro (in that one instance at least)
16:40:46pamauryfrancesco_: you could eventually change it to "something%d" by recompiling
16:40:54 Quit domonoky (Ping timeout: 252 seconds)
16:41:01AlexPyeah, but regular text requires an arse load of \opt and \nopt
16:41:01saratogaHaimN: try removing "HAVE_RECORDING" from the config file
16:41:12francesco_pamaury: that's ok for me. Could you tell me where to look in the many code files?
16:41:13saratogai think removing the tuner and disabling recording together should work
16:41:48AlexPevilnick_B: See the start of
16:41:54pamauryfrancesco_: just a minute, I'l trying to remember in which header it is and the conditions on it
16:42:22HaimNsaratoga: after I will finish the cimpile i'll cheack it
16:42:30pamauryfrancesco_: in firmware/include/dir.h
16:42:43francesco_pamaury: many thanks . I'll try that
16:42:50pamauryyou can change VOL_NAMES and need to tweak VOL_ENUM_POS accordingly
16:43:01saratogaHaimN: just tried now and it works (or at least compiles)
16:43:02evilnick_BAlexP: I see. Well, it's relatively minor, it just looks a little bit odd for a new user
16:43:24AlexPhmmm, the nano 2G just says nano, the e200 and c200 say e.g. e200v2 series and the clip anf fuze say e.g. "Fuze v2"
16:43:29AlexPNicely consistent!
16:43:49AlexPWe could have a new macro called \playerseries or something
16:44:43AlexPah, not needed as in there is already "\newcommand{\playerlongtype}{\playertype{} 2nd generation}"
16:44:52HaimNsaratoga: I can't do it now because I at the middle of the compilw
16:44:54AlexPSo the playertype is just wrong for the majority of them
16:45:44AlexPand playerlongtype is wrong for a few of them too
16:45:50AlexPI'll sort it out when I get home
16:46:01AlexPevilnick_B: You might need to remind me :)
16:46:24*evilnick_B will try to remember
16:51:04 Part LinusN
16:59:34 Join toffe82 [0] (~chatzilla@
17:00:21 Join kugel_ [0] (
17:00:33 Quit kugel (Disconnected by services)
17:00:37 Nick kugel_ is now known as kugel (
17:00:41 Quit kugel (Changing host)
17:00:42 Join kugel [0] (~kugel@rockbox/developer/kugel)
17:01:57amiconnAlexP: e200 & friends are "series" because there are several actual modelsa
17:02:07amiconne250, e250, e270, ...
17:05:06***Saving seen data "./dancer.seen"
17:05:16 Part Zagor
17:06:57AlexPamiconn: Yeah, good point
17:07:06AlexPThe use or not of v2 is still iffy though
17:07:28AlexPEspecially given playerlongtype exists apparently for that
17:07:42CIA-81New commit by pamaury (r28122): rbutil: update french translation
17:09:21CIA-81r28122 build result: All green
17:13:01saratogathe sansa players are actually called "Clip V2" by sandisk, which is probably why
17:13:07saratogaat least mine has that printed on the player
17:14:28kugelit says fuze on my fuze v2
17:14:36kugel"Fuze" to be precise
17:15:19 Join _s1gma [0] (~d.d.derp@
17:18:19francesco_pamaury: after changing the appropriate line in dir.h , can't understand what to set VOL_ENUM_POS to...
17:18:29 Join Bagder [0] (~daniel@rockbox/developer/bagder)
17:21:33pamauryVOL_ENUM_POS is the position of the "%d" is the string. If you have VOL_ENUM defined to "<%d>" for example, then it's 1, if VOL_ENUM is set of "<bla%dbli>", then it's 4, ...
17:26:00 Join AndroUser [0] (~androirc@
17:28:06 Quit Zarggg (Ping timeout: 240 seconds)
17:28:16pixelmaAlexP: be a bit careful, one of those macros is also used for the page footer telling what manual it exactly is. I semi-recently stumbled across this too when introducing the seriesimg macro to make the Clip+ use the correct screenshots but keeping a difference between the other Clips
17:29:45 Join Zarggg [0] (
17:29:57 Quit AndroUser (Remote host closed the connection)
17:35:07 Join giovanni [0] (~androirc@
17:35:35AlexPI'll have a look, I might just invent a new playerseries one or something if it is too intertwined
17:35:41AlexP@ pixelma
17:35:45pixelmamy point is just that you can find use of the different macros in lots of places, some maybe not so obvious places like preamble.tex
17:35:57AlexPAlthough I think the manual name one should be longplayertype
17:36:03AlexPTo give the full name
17:36:26 Join funman [0] (~fun@rockbox/developer/funman)
17:41:51 Quit markun (Read error: Connection reset by peer)
17:52:43 Quit einhirn (Ping timeout: 276 seconds)
17:52:58 Join komputes [0] (~komputes@ubuntu/member/komputes)
17:59:27 Quit petur (Quit: Page closed)
18:03:50 Join t0rc [0] (~t0rc@unaffiliated/t0rc/x-5233201)
18:08:45 Quit komputes (Remote host closed the connection)
18:09:11 Quit giovanni (Quit: Bye)
18:09:27 Join giovanni [0] (~androirc@
18:09:31 Quit giovanni (Client Quit)
18:13:26shuffle2blehhh i still can't find what went wrong :/
18:13:54 Join CGL [0] (~CGL@
18:14:02 Join markun [0] (
18:14:09 Quit markun (Changing host)
18:14:09 Join markun [0] (~markun@rockbox/developer/markun)
18:17:26 Join einhirn [0] (
18:19:51 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
18:20:57 Join komputes [0] (~komputes@ubuntu/member/komputes)
18:22:49shuffle2i'm not sure it's actually stalling on a blocking send
18:23:08shuffle2the logf()s are not printed to the screen instantly, so idk where it's actually locking up
18:23:21shuffle2which is a problem in itself :)
18:25:33 Quit t0rc (Quit: Give someone code, help them with one project. Teach someone to code, help them rule the world.)
18:26:58 Quit einhirn (Ping timeout: 276 seconds)
18:31:35shuffle2also if i put usb in highspeed instead of full, it will not lock up. however, it doesn't work then, either :p
18:31:38 Quit swilde (Quit: ERC Version 5.3 (IRC client for Emacs))
18:33:51pamauryhard to tell without an analyzer. You can try to tweaks logf to log to disk so that you can see everything after reboot
18:35:05shuffle2i have a wireshark log (i know that's not what you mean, tho)
18:35:17shuffle2fwiw, it interacts with a pc "fine": no lockups
18:35:28shuffle2silly thing!
18:39:24CIA-81New commit by funman (r28123): file_exists(): fix DEBUGF() ...
18:41:04CIA-81r28123 build result: All green
18:43:16 Join n1s [0] (
18:43:16 Quit n1s (Changing host)
18:43:16 Join n1s [0] (~n1s@rockbox/developer/n1s)
18:43:37shuffle2pamaury not sure if you were around at the time, but this was completely working last night :p
18:44:38pamauryreally ?
18:48:56pamauryand now it doesn't work anymore :(
18:49:21shuffle2indeed. very exasperating
18:49:47 Join einhirn [0] (
18:50:26 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
18:50:32 Quit DerPapst1 (Quit: Leaving.)
18:50:55pamauryand what changed ?
18:52:12shuffle2i'm pretty certain all i did was update svn, which pulled changes unrelated to usb
18:52:35shuffle2but it caused the entire project to rebuild, and that's when it died
18:52:46shuffle2maybe i should try with no gcc optimizations?
18:53:12 Join Strife89TX [0] (~cstrife89@
18:53:43shuffle2the root cause is probably just usb interrupts conflicting with (my) crappy thread :>
18:53:47shuffle2but it worked before! heh
18:54:30funmanwhat is STUBOFFSET in DEBUG builds?
18:56:02 Join Jerom [0] (~jerome@
18:58:12funmansomething sh-specific
18:59:58wodzfunman: this is related to serial stub for gdb
19:01:43funmanwodz: only for archos or other targets as well?
19:02:21 Quit einhirn (Ping timeout: 240 seconds)
19:03:44 Quit Strife89TX (Quit: Switching apps.)
19:03:58funmanafaik gdb only works on (some?) archos and ifp
19:05:07***Saving seen data "./dancer.seen"
19:06:01HaimNsaratoga: You wrote to me "try removing "HAVE_RECORDING" from the config file" were is the config file?
19:09:01wodzfunman: I think it is for archos only but better ask amiconn
19:09:31funmanthere is some ifp code
19:09:40funmanin fact i wanted to see how possible it is to use gdb on clipv1
19:10:02CIA-81New commit by funman (r28124): */ remove STUBOFFSET ...
19:10:29wodzfunman: you have to have some way to interact with device - JTAG, serial, usb probably
19:10:59funmani want to look at usb
19:11:12funmanno idea how this works though
19:11:37CIA-81r28124 build result: All green
19:11:45wodzthere is a patch on FS which implements some debug output channel for arms
19:13:02funmanFS #8454?
19:14:23funmanthanks :)
19:17:41 Quit antil33t (Read error: Connection reset by peer)
19:17:54 Join antil33t [0] (
19:21:24 Join panni_ [0] (
19:21:35funmanranma: have you seen FS #8454 ?
19:25:52funmandocs/COMMITTERS says tomal is Tomasz but says Tomek ?¿
19:25:54 Join Kitr88 [0] (
19:26:28kugeltomal != tomek ?
19:27:06funmanthey have the same surname and both worked on ifp7xx, i think it's too big for a coincidence
19:27:43funmanperhaps it is a diminutive, anyway he last committed 3 years ago
19:28:19kugelperhaps silblings?
19:28:28bertrikI think we also have someone in CREDITS under two names
19:28:36 Quit Kitar|st (Ping timeout: 265 seconds)
19:28:37ranmafunman: No, but looks neat, I should try that one...
19:30:35 Quit Kitr88 (Ping timeout: 272 seconds)
19:31:22bertrikPhinitnun Chanasabaeng and Pinitnun Shanasabang, the same person?
19:32:43n1sI'd say probably, Thai translation IIRC so if both those come from updates to that i'd say "Yes"
19:33:37kugelfunman: btw, you recently made a lang commit. IIUC new translators shall be added to the top of the lang file
19:34:05funman has a different spelling than in thai.lang file
19:34:47funman confirms what bertrik says
19:35:33 Join Kitar|st [0] (
19:36:07funmanrasher: ^ where did you read 'Shanasabang' ?
19:36:45funmankugel: ok, pixelma also told me i should not commit lang updates if there's a committer who is a native speaker
19:37:32rasherfunman: I assume he/she changed the name in Flyspray
19:38:07n1sI'd say yust use the name from FS
19:38:20n1ss/ yust/, just/
19:38:42funmanupdate thai.lang too
19:38:49CIA-81New commit by kugel (r28125): Redirect (L)DEBUGFs to adb logcat in DEBUG enabled builds. ...
19:40:44kugelI can't git pull −−ff-only from anymore
19:40:51CIA-81r28125 build result: 0 errors, 23988 warnings (kugel committed)
19:41:02kugeldue to that commit message post-editing thing
19:41:20funmankugel: the git repo shouldn't have changed ?¿
19:41:42kugeldunno, maybe it conflicts with svn rebase
19:41:57 Quit pamaury (Remote host closed the connection)
19:42:13funmanyou could try to reset −−hard to before this commit and rebase again?
19:42:49CIA-81New commit by kugel (r28126): Fix typo
19:43:31kugelfunman: I can do that once, but every subsequent ff-only pull fails
19:44:28CIA-81r28126 build result: All green
19:48:21kugelthat seriously sucks
19:48:23funmando you have a commit b5c5325e98273 ? (28078 for me)
19:48:30funmanthe log is the original one ("grumble")
19:48:45kugelseems so
19:51:27kugelfunman: I have also 08bd3157173b742b846588fa4b6964838b85e01c which is the "fixed" one
19:51:30bertrikso, is anyone going to update thai.lang and/or docs/CREDITS?
19:52:43 Quit Zambezi (Changing host)
19:52:43 Join Zambezi [0] (Zulu@unaffiliated/zambezi)
19:53:00funmanbertrik: you?
19:53:14kugelmy entire work flow is broken now
19:53:18funmankugel: ah, it's not there (but i'm using git svn rebase)
19:53:26kugelI'm using both
19:53:33HaimNsaratoga: it's works! (the removig of the radio) thanks!! but the "record" line gone too, what to do?
19:53:39funmanperhaps it's fixable by hacking the git repo
19:54:18kugelwell, I guess it needs to re-clone from svn
19:55:11 Join Lear [0] (chatzilla@rockbox/developer/lear)
19:55:29funmanif you clone from my repo would it be faster?
19:56:08kugelfunman: I want to be able to "git pull", your repo won't help me
19:56:17kugelI need a real git mirror
19:56:25funmanand cp -r?
19:56:50kugelmy tree will be broken once I pull from the git mirror I'm afraid
19:58:29kugelgit pull works, as long as I don't svn rebase
19:58:55kugelthat gets me the "dirty" post-edited commit
19:59:22kugelbut I do need svn rebase to commit :\
20:00:26 Join drizztbsd [0] (
20:00:54funmanlet's switch to git completely then
20:01:17kugeleasier said than done
20:02:16funmansetup ssh, ask a key from each commiter, done
20:02:27kugeland you can't sensibly post-edit commit messages in git (not without breaking everyone's checkouts), so we could also just stop using this svn feature again
20:02:35saratogaHaimN: if you want one but not the other you're going to have to find another way to do it
20:02:36funmanthat's true
20:03:11 Quit alexbobp (Read error: Connection reset by peer)
20:03:33HaimNsaratoga: ok, thanks a lot!
20:03:54 Join alexbobP [0] (
20:04:37funmanifp gdb stub doesn't build anymore
20:07:06 Join Horscht [0] (~Horscht@xbmc/user/horscht)
20:13:47 Join robin0800 [0] (~robin0800@
20:16:40 Join haim_ [0] (~HaimN@
20:16:46 Quit HaimN (Ping timeout: 255 seconds)
20:16:53 Nick haim_ is now known as HaimN (~HaimN@
20:25:04kugelwe should go back to DEBUG builds for the sim
20:25:06 Join notlistening [0] (
20:25:57amiconnfunman: The stub offset is necessary so that the main binary doesn't overwrite the gdb stub
20:26:39 Quit Jerom (Quit: Leaving.)
20:26:49notlisteningkugel is there a repository for of the apk's you working on now.. like a daily build?
20:26:55 Quit robin0800 (Remote host closed the connection)
20:27:12kugelwell, the repository is svn
20:27:17kugelyou can build apks from it
20:27:29kugelbut there's no daily built (yet)
20:27:41 Join Jerom [0] (~heidi@
20:30:41 Join Buschel [0] (
20:32:35kugelfunman: the whole debug thing is a complete mess. firmware/ contains code for sh&cf which would be a better fit in the target tree
20:33:08funmani only looked at apps/debug_menu.c and that was horrific enough
20:34:07funmanamiconn: is the gdb stub built as a standalone binary? i was thinking of running it from rockbox
20:34:08kugelI also kinda doubt that anyone has used gdbstub recently
20:34:14funmanthat would prevent 'load' command of course
20:34:28amiconnYes, the stub is a separate binary
20:35:38amiconnIt is loaded like an ordinary firmware, then starts listening on the serial port, to which you then connect from the pc and transfer the debug build, which is loaded at stuboffset
20:36:12amiconnOne reason for this is that a debug build is *way* larger than what the archos rom firmware loader would accept
20:36:35funmanwhy is that?
20:36:36kugeldebug built with -g?
20:36:59kugelyou can debug without -g too, as long as you have a -g enabled .elf on the host IIUC
20:37:13amiconnThey're also built with -O0
20:37:16funman-g gives the same .elf than without it for AMS but i think because debug sections are discarded by the linker
20:37:22kugel(I debugged on my mini2440 this way)
20:37:43kugel-O0 isn't necessary for -g
20:37:58funmani think if i work on this i would look at ifp code
20:38:14kugel-O0 doesn't affect -g (except inlining and <value optimized out>)
20:41:22Buschelhmm, the nano2g has defined LCD_SHUTDOWN, but not LCD_SLEEP? is it not possible to wake it up via lcd_init_device()?
20:41:29kugeldebug.c is a funny file
20:42:44kugelI especially like the #ifdef __GNUC__ at the end :)
20:48:41 Join captainkewllll [0] (2669ecc2@gateway/web/freenode/ip.
20:51:17kugelbtw, why do we have no .gitignore in svn?
20:51:40pixelmathat already reads wrong
20:53:00CIA-81New commit by kugel (r28127): Remove from version control, it's auto-generated.
20:56:03CIA-81r28127 build result: All green
20:58:33funmankugel: because you can use svn:ignore
20:58:54 Join Zagor [0] (~bjst@rockbox/developer/Zagor)
21:00:13kugelthat works with git?
21:00:59kugelmaybe that explains why I have a pretty complete .gitignore I don't know where it comes from
21:02:54 Join DerPapst [0] (
21:03:05funman git svn show-ignore >> .git/info/exclude
21:03:11funman(from man git-svn)
21:03:43kugelthat makes sense now :)
21:03:58funmanhowever i don't share files to be ignored with other, it only works for local copy
21:05:10***Saving seen data "./dancer.seen"
21:05:39 Part Zagor
21:07:09kugel should be ignored, would be nice if someone using svn could add them
21:07:13 Join Zagor [0] (~bjst@rockbox/developer/Zagor)
21:10:30 Join drizztbsd_ [0] (
21:11:00 Part toffe82
21:12:22 Quit drizztbsd (Ping timeout: 240 seconds)
21:14:10funmanlet's wait 3 days, if it's not applied it means nobody uses svn so we can switch to git losslessly
21:15:24 Quit pixelma (Read error: Operation timed out)
21:15:25 Join robin0800 [0] (~robin0800@
21:15:46 Quit esperegu (Read error: Connection reset by peer)
21:15:50funmankugel: ah i understand now: you ran git svn rebase after the commit was changed. it works for me because i already had the old commit
21:16:15kugelsvn rebase works for me
21:16:18 Quit amiconn (Disconnected by services)
21:16:19kugelgit pull doesn't
21:16:19 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
21:16:19 Join pixelma [0] (quassel@rockbox/staff/pixelma)
21:16:28kugelI get the new commit message after svn rebase
21:16:41kugelpull expects the old one, hence fast-forward fails
21:17:39 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
21:18:00wodzAs I remember the main problem in switching to git is the fact that it doesn't work well on windows + doesn't have such tools as tostoisesvn (or how does it spell)
21:18:59funmani know people who use it on windows so i'm not sure what the problem is
21:19:00Zagorkugel: apkbuilder is deprecated? are they herding everone towards eclipse?
21:19:03gevaertsThat's *one* of the reasons, yes
21:19:11kugelZagor: no idea
21:19:51funmanother reasons are: developers not wanting to change the tool they know, and no need to change a working system
21:20:24Zagorand: git being very complex and hard to learn
21:20:39gevaertsThe main reason for me is *serious* annoyance about how lots of git proponents present their case
21:20:46kugelwodz: tortoisegit exists IIRC
21:20:56simonrvni only pull from svn, so ;p
21:21:26kugelI have no problem with sticking with svn, but i would like to continue how I work currently
21:21:41wodzconsidering problems with cygwin I think we should ship development env as virtual machine image officially
21:22:07funmanZagor: can you reset the git repo to have the new r28078 log?
21:22:15Zagorfunman: sure. if you tell me how :-)
21:22:16 Quit Lear (Ping timeout: 245 seconds)
21:22:32gevaertswodz: we ship Do we really have to provide more?
21:22:49funmangit reset −−hard e33918c0e98157a6b75336b289be43fd86ba8d5a; git svn rebase # maybe
21:23:02funmanlet me verify
21:23:22pixelmaI also see no "problem" with cygwin, except that you need a bit of patience
21:23:24funmanno it doesn't fetch from svn
21:23:28 Quit Buschel (Ping timeout: 240 seconds)
21:23:42wodzok - I shut up
21:24:10gevaertsSure, cygwwin packages and prepared vms are nice, but I don't think we as a project are responsible for providing those
21:24:24funmanZagor: git svn reset -r 28077 ; git svn rebase
21:24:26 Quit lestatar (Read error: Connection reset by peer)
21:24:34Zagorwodz: if anyone builds one, I'll be happy to host it on
21:25:46*gevaerts thinks that toolchain binaries should remain as best-effort volunteer-provided things. They can be hosted on, but not having them is *not* a disaster
21:25:55pixelmaZagor (while you are around): is it possible for you to check somehow why I don't get notification mails from flyspray anymore? It once worked, don't know when it broke exactly as I just recently started using the "watch task" feature a bit more again
21:26:14funmanpixelma: it works for me now
21:26:50Zagorpixelma: that's odd
21:27:19Zagorfunman: rebase running...
21:27:24pixelmafunman: since when? I think I should have gotten a task closed mail yesterday or on Saturday which I didn't
21:28:35funmanDate: Mon, 20 Sep 2010 15:42:22 +0200
21:28:39 Quit Rondom (Disconnected by services)
21:28:45kugelgevaerts: what do you mean with "git proponents present their case"?
21:28:46funmanDate: Sun, 19 Sep 2010 21:03:51 +0200
21:28:54 Join Rondom_ [0] (
21:29:17Zagorfunman: "Unable to determine upstream SVN information from history". how does it look now?
21:29:20funmangevaerts: don't use "proponent", it is not a war
21:29:31kugelZagor: bad :)
21:29:34funmanZagor: hmm i don't know
21:29:39gevaertskugel: they way many (not all) git users tend to imply that everyone who doesn't use git must either be uninformed or stupid
21:29:55kugelbut it's true!
21:30:01kugelj/k :)
21:31:22kugelZagor: is this a simple checkout which runs svn rebase in a cronjob?
21:31:38kugelif yes then I had expected funman's command to work
21:32:06funmanZagor: ah perhaps you need the 'git reset' too
21:33:59 Join lestatar [0] (
21:34:01Zagorit does rebase and push each svn commit
21:34:16Zagorbut now it won't push
21:34:25Zagor ! [rejected] master -> master (non-fast-forward)
21:34:32funmanwhere is HEAD at ?
21:34:45Zagorcommands, please
21:35:03funmangit show|grep ^commit
21:35:12Zagorcommit b1945427afdbe3bb82a0a7132aa4196f86d3fc77
21:35:36funmani don't have it here but my history has the old commit log...
21:36:10kugelb1945427afdbe3bb82a0a7132aa4196f86d3fc77 the commit I just did
21:36:31funmanwhy does it need to push?
21:36:35Zagoroups. I ran the command in the naked repo instead of the place where it's normally ran
21:36:49funmanso the repo where it pushes was not updated?
21:37:10Zagorit is now
21:37:20CIA-81New commit by bertrik (r28128): iap: make variable serbuf non-global
21:37:22funmankugel: no, this one is dbf803a97d4bf5df9c7f7e67bb6b72643750be12 !
21:37:59kugelfatal: bad object dbf803a97d4bf5df9c7f7e67bb6b72643750be12
21:38:36kugelfunman: I did my commands from master which are up to the git mirror (I didn't do svn rebase yet)
21:38:52kugelso I expect our trees to be different
21:38:58CIA-81r28128 build result: All green
21:39:10funmanit is different only starting at r28078
21:39:50kugelis the git mirror fixed now?
21:40:01 Join Coty [0] (
21:41:07kugelno :(
21:41:07Zagoryou're not looking at me, are you? I have no idea what I just did, let alone fix it.
21:42:40funmanZagor: you have reset the git repo state just 1 rev before r28078 (the revision which log was edited by me)
21:42:49funmanand then fetched the new history again
21:42:55funmanwith the new commit log
21:44:14kugelfunman: it's still not working as before for me
21:44:28ZagorBagder: help?
21:45:11BagderI don't know anything about those things either
21:45:16kugelfunman: but I received the changed commit via git pull so it probably should work
21:45:25Zagoryay, blind leading blind ;-)
21:46:48kugelfunman: I still get 08bd3157173b742b846588fa4b6964838b85e01c to be the first different from upstream (origin/master)
21:46:52 Quit robin0800 (Remote host closed the connection)
21:47:21 Join robin0800 [0] (~robin0800@
21:48:04funmankugel: even if you reset?
21:48:18 Join COTY1 [0] (
21:48:27 Quit COTY1 (Client Quit)
21:48:38kugelI did git svn reset -r 28078 && git reset −−hard e33918c0e98157a6b75336b289be43fd86ba8d5a
21:48:52kugelthen git pull origin master && git svn rebase
21:48:56kugeldoing it again fails
21:49:17funmanZagor: what is the sha1 for HEAD?
21:49:32funmaned4cef3082405 here
21:50:25Bagdergit log -1 −−oneline perhaps?
21:50:55 Join mshathlonxp [0] (
21:51:21 Quit Coty (Ping timeout: 272 seconds)
21:51:43funmangit show|grep ^commit
21:51:44kugelbtw, bertrik's commit isn't in the mirror
21:51:52Zagorwell, they're obviously different in the naked and the clothed repo. 7f80fac in the svn repo, dbf803a in the naked
21:52:23kugeldbf803a for me too, but that's r28127 not r28128
21:52:27funmani have dbf803a97 and you shouldn't have it
21:52:28 Join Lear [0] (chatzilla@rockbox/developer/lear)
21:52:35funmanwhat's "naked" and "clothed" ?
21:53:17Zagornaked is what you get. "clothed" is the checked out repo which is rebased and pushed to the naked for each commit
21:53:37Zagorexcept now the pushing is broken
21:54:26Zagorisn't "naked" the word git uses?
21:54:43funmannever read that before
21:55:01funmanso for some reason the naked repo isn't updated
21:55:29Zagornot "some" reason. this reason: ! [rejected] master -> master (non-fast-forward)
21:55:42kugelZagor: yes, because the clothed has a different history
21:55:46Zagorwhich most likely is because I ran your first command in the naked repo
21:55:56kugelyou can −−force to overwrite the naked's history with the one of clothed
21:56:19Zagorok, force done. better now?
21:57:15kugelwin! \o/
21:57:34Zagornow, why isn't the android build working? :-)
21:57:47kugelit isn't?
21:57:51*gevaerts wonders...
21:57:57Zagormake works for me, but not "make apk".
21:58:11gevaertsIs all this resetting a sensible thing to do automatically?
21:58:21kugelput a make zip between them
21:58:29mshathlonxpI'm looking for some info about rockbox languages but I can't find anything about that in webpage in forums, can anyone help?
21:58:30gevaertsAnd if so, won't it break people's repositories anyway?
21:58:50Zagorkugel: ok, now I get make: *** No rule to make target `/home/bjst/bin/rockbox/android/bin/org/rockbox/RockboxActivity.class', needed by `/home/bjst/src/rockbox/android/bin/classes.dex
21:58:52gevaertsmshathlonxp: something like ?
21:59:01kugelgevaerts: yes. even if the mirror did it automatically, it still breaks all clones from the mirror
21:59:27kugelZagor: that is strange, it has a rule
21:59:40Zagorhistory seems to show the git mirror only breaks if I'm touching it
21:59:40kugel(and WFM too)
21:59:49mshathlonxpgevaerts thx :D
22:00:47kugelrebasing or resetting (and applying a different history) in a branch always breaks all checkouts that pull from the branch (because it rewrites the history)
22:02:28gevaertsok, makes sense. In that case, I think we should disable history changing again
22:03:49AlexPand just require people to do it properly in the first place :)
22:03:49gevaertsYes, we just have to practise our shouting a bit
22:04:24kugelI guess svn doesn't have this problem because the main repo is on the server. at git, the complete history is on the local machine
22:05:25gevaertsYes. With svn only the current checkout is local, everything else (such as the messages) only exists in one place
22:06:36 Quit mshathlonxp (Quit: let's rock!)
22:07:23Zagorkugel: where is the target for RockboxActivity?
22:07:45 Quit markun (Read error: Connection reset by peer)
22:09:42Zagoroh, looks like BUILDDIR is wrong somehow
22:10:33kugelZagor: oh, I guess "JAVA_OBJ:= $(call java2class,$(subst /src/,/bin/,$(JAVA_SRC)))" doesn't work well on your system
22:11:01Bagdermake isn't really good for java
22:11:03Zagorheh no since JAVA_SRC doesn't exist
22:11:17ZagorBagder: it's better than ant :-)
22:11:24Bagderactually it isn't
22:11:27Zagor/bin/sh is better than ant
22:11:30 Join Adublaptop [0] (~Aldubuc@
22:11:46Bagderant knows java magic while sh and make don't
22:12:00Bagderof course ant is awful still
22:12:09kugelZagor: according to your error, /src/ has been replaced by /bin/, but twice (since you have the rockbox dir in src/)
22:13:10Zagorwhat is ANDROID_DIR supposed to point at?
22:13:21Zagoroh, ANDROID_DIR=$(ROOTDIR)/android
22:13:55Zagorandroid.make looks a bit more complex than necessary
22:14:45kugelpossibly, my make-fu isn't very good
22:15:12 Quit pixelma (Disconnected by services)
22:15:12 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
22:15:12 Quit amiconn (Disconnected by services)
22:15:14 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
22:15:17 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
22:16:04 Quit Adublaptop (Read error: Connection reset by peer)
22:16:33kugelBagder: it works good enough for this small set of java files (for me, anyway)
22:16:49 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
22:17:00Bagderyeah, I guess you can make it work fine for a small enough project
22:17:27kugelone thing I thought might confuse make is that javac automatically compiles other java files needed by the one passed to it, but it appared it can deal with it
22:17:32Bagderbut for a small enough project, you probably are better off just running javac on all java files every time
22:17:38*Zagor decides against launching his rant against ant
22:17:44 Quit xnyhps (Read error: Operation timed out)
22:18:17Bagderant is a great example why XML is crappy for config files ;-)
22:19:01 Join xnyhps [0] (
22:21:24Zagorkugel: ah, you're right. it fails because my JAVA_SRC path contains two "src"
22:21:42 Quit robin0800 (Read error: Connection reset by peer)
22:22:07 Join robin0800 [0] (~robin0800@
22:25:04Zagorkugel: is it "keytool -genseckey" I need to run?
22:26:06kugelZagor: yes, but I think need to figure out how to deal with it
22:26:38kugelI guess we want a single signature for all official binaries
22:26:54kugelin that case we probably don't want to give the private key to the build clients
22:27:22Zagorgood point
22:27:22kugelsomeone had the idea to let the build clients only build the unsigned apk, and then let the build master sign the apks
22:27:38Zagorthat's not really any safer :)
22:27:58 Quit drizztbsd_ (Ping timeout: 240 seconds)
22:28:12n1sthen the build master just trusts the clients anyway
22:28:18Zagorwell the signing is to avoid malicious code (which is a real problem in android). and if we sign anything we get sent, we bypass that.
22:29:26kugelhm, right
22:29:26kugelbut we at least keep the key private
22:29:26n1show many android builds are there? maybe the master can build them itself?
22:29:26kugelwhich means we can switch to a more safe solution at any time
22:29:26ZagorI think we need to be a bit more paranoid about the android port than our regular mp3 player builds
22:29:26 Join giovanni_ [0] (~giovanni@
22:29:59Zagorn1s: yes I was thinking that too. it's just three different, isn't it?
22:30:55kugelwell, there are very few popular resolutions, but you could build for any resultion you can imagine (ignoring that not having a cabbie port sucks)
22:31:23kugeland only two resolutions work well currently because they have a cabbiev2
22:32:58Bagderthey'er not only "popular", they are also the only google "supported" resolutions
22:33:05kugelwe could also build a resolution independant binary, but then we need still need to figure out how to handle the themes we ship, and plugins if we ever build them
22:33:51 Quit [sko] (Read error: Connection reset by peer)
22:34:20 Quit robin0800 (Remote host closed the connection)
22:34:50 Join robin0800 [0] (~robin0800@
22:37:59 Join t0rc [0] (~t0rc@unaffiliated/t0rc/x-5233201)
22:38:18kugelZagor: all the malicous code I heard about for android was properly signed :)
22:40:47kugelwell, I thought the signing thing was to avoid from untrusted sources (ie. the only trusted source is the home of the project), if we don't trust our clients we shouldn't sign their apks, but I guess it's more about avoid apks from outside of our project (custom builds)
22:40:55 Quit robin0800 (Remote host closed the connection)
22:41:23 Join robin0800 [0] (~robin0800@
22:42:02Zagorkugel: we don't filter our clients. anyone can trivially modify a build client, connect and upload poisoned binaries
22:42:56 Quit funman (Quit: leaving)
22:43:14Zagorwe haven't bother securing it yet, because there's not much harm you can do to stand-alone players. phones, on the other hand, are a much more coveted target.
22:43:48bertrikWhat does a signed apk really mean anyway? It's not like there is a chain of trust, right?
22:44:33 Join markun [0] (~markun@rockbox/developer/markun)
22:44:53kugelit means person X and not Y built it
22:45:35Zagorexactly. that's why the signing isn't as important as knowing what the apk contains
22:45:51Bagderso the first time, how does it know X from Y?
22:45:55ZagorI'd say for now we build the android targets on our server
22:46:19kugelBagder: the one that comes first wins :)
22:46:25 Quit robin0800 (Remote host closed the connection)
22:46:25ZagorBagder: I guess you get a key from google to upload to their Market
22:46:34ZagorI don't know though
22:46:40*kugel neither
22:46:41Bagderah yes, that might be so
22:46:53 Join robin0800 [0] (~robin0800@
22:48:09scorcheZagor: you might want to go ahead and request a PO:−−-2010
22:48:33Zagorwoo, paperwork! :-)
22:48:52scorcheZagor: as you likely are going to be using the same info as last year, you just need to fill out the payment request form at [1] if you remember the details...
22:49:58ZagorI just need to be let in first... :)
22:51:42notlisteningkugel, thanks for the reponse ealier
22:51:50scorcheZagor: to what? arent on the mentor list, are you...
22:52:05Zagornope. I just applied.
22:53:55 Quit n1s (Quit: Lmnar)
22:55:53notlisteningkugel based on the conversation now how far is the rockbox app awake from the android marketplace?
22:56:05kugelI can't tell
22:56:31notlisteninglol thought that might have been the answer :D
22:56:39notlisteningwhat do you hope for...
22:56:42notlisteningit mabe better
22:56:48notlisteningis maybe...
22:57:22ZagorI don't think we should be in a hurry to get it to Market. let's polished it a bit first.
22:57:45notlisteningyeah i agree I just was not sure how polished it was yet
22:58:00kugelyou can always try
22:58:15notlisteningI imagine it will rank as one of the Top apps very quickly
22:59:00notlisteningreleasing it too early can be fatal
22:59:11 Quit robin0800 (Remote host closed the connection)
22:59:14 Join timc [0] (~tim@
22:59:19notlisteningif it does not work it gets uninstall and forgotten
22:59:27 Nick timc is now known as Guest59303 (~tim@
22:59:40 Join robin0800 [0] (~robin0800@
23:01:39notlisteningwill it support all the codec that rockbox is boasting now...or is that limited by the OS?
23:03:29 Quit bertrik (Quit: :tiuQ)
23:04:02Bagderit should feature them all
23:04:49Zagorthe gigabeast is finally slain
23:05:12***Saving seen data "./dancer.seen"
23:05:21 Quit t0rc (Quit: work)
23:05:38kugelZagor: I'm not actually sure :)
23:05:56 Quit domonoky1 (Read error: Connection reset by peer)
23:05:57Zagoroh I'm pretty sure the 1GHz+ monsters are beating it
23:06:16kugelin theory sure, but we of course run a good deal slower as app than as os
23:07:00Zagortrue, I don't know how android's on-demand clocking works
23:07:13kugelsyscalls are more expensive, PIC code is probably a bit slower and bigger, no access to fast iram, sharing cpu time with lots of other apps
23:07:49Bagderbut better caches probably
23:07:53notlisteningis it avaiable on a non root'd os?
23:08:15Zagornotlistening: yes, just install it manually
23:08:24Zagorit's not an iphone :-)
23:08:56notlisteningZagor :D
23:09:11notlisteningu know how these phone operators like to think it is
23:10:38 Quit Lear (Quit: ChatZilla 0.9.86 [Firefox 4.0b7pre/20100920031745])
23:11:31notlisteningbut i am not sure but if rockbox wanted to unclock over clock the processor at least on my device it would not be allowed as i have not rooted it
23:12:19Zagorno we don't do any special tricks
23:12:54 Join MxxCon [0] (
23:14:09MxxConi'm sorry if this is a repeated question, but now that vlc have been approved for iphone/ipad appstore, do you guys consider porting rockbox to ios?
23:14:46krazykitonly if someone wanted to write the port
23:15:11MxxConnobody expressed that interest?
23:15:49gevaertsIt's not at all clear if the appstore can be used for GPL software...
23:16:10 Quit evilnick_B (Quit: Page closed)
23:17:18Zagorgevaerts: vlc is gpl
23:17:32Zagorso it *can* be. whether it's legal is another issue...
23:17:50gevaertsZagor: yes, this means the vlc people have decided to allow it. That doesn't mean we have to :)
23:18:21notlisteningAndroid is going to be much more popular in the end
23:18:36gevaertsIt's not a question of which is more popular here...
23:18:43kugelI thought it was impossible to port gpl to ios?
23:19:05gevaertskugel: not exactly
23:19:29Zagorgevaerts: well, vlc contains a boatload of external libs which may not be so happy to forfeit the gpl
23:19:30gevaertsThe itunes app store seems to have some terms that are seriously incompatible with the GPL.
23:19:44gevaertsZagor: yes, indeed. We'll have to wait and see there
23:20:04MxxConwhy wouldn't you want to allow rockbox on ios?
23:20:47MxxCon(i'd rather use rockbox than ipod-player)
23:20:48gevaertsIt's not a question of rockbox on ios. The problem is rockbox on the itunes app store
23:21:16ZagorMxxCon: apple adds limitations that are incompatible with the gpl
23:21:33gevaertsAnd the problem there is that Apple imposes terms that are (a) incompatible with the GPL, and (b) not necessarily acceptable to the people who own the copyright
23:21:46 Join clone4crw [0] (
23:22:58 Quit Dreamxtreme (Quit: Never look down on someone unless you're helping them up.)
23:23:46kugelgevaerts: sorry, I meant on the app store
23:24:11gevaertsActually, aren't these app store requirements secret, so you can't technically even ask the copyright holders if it's ok?
23:24:14 Join Dreamxtreme [0] (~Dreamxtre@
23:25:22kugelbut vlc is on the appstore now
23:25:59MxxConwordpress's offical iphone publishing app is opensource
23:26:28Zagorkugel: yes, but is it legal?
23:26:42kugelZagor: I thought it isn't but maybe it works unless someone seriously questions the whole thing
23:26:56Zagori.e. do they have the consent of the copyright holders of every 3rd party lib they include?
23:27:17Zagorsure, but that's not how I want Rockbox to act
23:27:36*gevaerts agrees with Zagor
23:28:40 Quit robin0800 (Remote host closed the connection)
23:28:58 Join user890104 [0] (
23:29:05 Join robin0800 [0] (~robin0800@
23:29:13gevaertsMxxCon: it's perfectly possible to have open source software on the app store. It's *not* possible to have GPL software on the app store without adding exceptions to the license
23:29:25AlexPMxxCon: You are better of getting a phone that let's you be in charge and doesn't have brain dead people running it :)
23:29:38gevaertsThe BSD license e.g. doesn't conflict with anything much
23:30:06notlisteningAlex which phone have you gpt in mind...?
23:30:10Zagorkugel: can you post the keytool line you use to create the keystore?
23:30:18AlexPnotlistening: basically not apple :)
23:30:27kugelZagor: I'm afraid I can't :)
23:30:37notlisteningAlexP ok
23:30:59kugelI'm using the key generated by eclipse (the debug key which we must not use for the market)
23:31:07MxxConhopefully android version will be released soon-ish
23:32:06Zagorgrr, eclipse :-/
23:32:26kugelI'm indifferent about this license&appstore thing. I'd just find it said if license issue disallow porting great software to many platforms
23:32:59gevaertskugel: it's sad, sure, but only Apple can change this...
23:33:49notlisteningit happened with eSpeak and rockbox
23:34:25 Quit DerPapst (Quit: Leaving.)
23:36:42MxxConohhh..intersting. never heard of espeak. is that something that'll go into rockbox?
23:37:19Zagorkeytool -genkey -alias androiddebugkey -keystore $HOME/.android/debug.keystore -storepass android -keypass android -validity 365 -dname "CN=Android Debug,O=Android,C=US"
23:37:37Zagoradding it to the README
23:39:29kugelgood idea
23:39:33*amiconn wonders how the ape filters could be modified to *optionally* allow runtime selection of code paths depending on cpu
23:40:31amiconnFor the standalone tools as well as for raaa
23:42:11 Quit shai (Quit: Leaving)
23:42:19kugelon android there's a little C program for this in the ndk
23:42:19notlisteningMxxCon in theory yes, but the licenses where in compatible version (lisenses are a big issue)
23:42:37kugelit basically just reads out /proc/cpuinfo
23:44:27 Quit TheSeven (Remote host closed the connection)
23:44:44 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
23:45:10 Part TheSeven
23:45:13 Join toffe82 [0] (~chatzilla@
23:45:40 Join Rondom [0] (
23:46:21 Quit Rondom_ (Read error: Operation timed out)
23:48:13Zagorkugel: do you know how to install my app in a virtual device?
23:48:24kugelthe emulator?
23:48:39kugeljust launch it, adb will select it if there's no other device
23:48:57Zagorlaunch what?
23:49:01kugelthe emulator
23:49:16kugeladb install -r will select it automatically
23:49:21Zagorah, ok
23:49:45kugelif you have more than one devices you can look up the serial number with adb devices, and pass that to adb with -s <number>
23:50:05AlexPbluebrother: Is there anything holding back a RBUtil release? There seem to be quite a lot of people these days with the newer fuzev2s that the latest release won't deal with
23:51:34 Join DemonX [0] (
23:52:26DemonXI'm trying to install Rockbox on an iPod Nano 1g, and it is producing the error "Partition not found. Connect usb cable and fix it"
23:54:09Zagoroops. Sktov tagcache
23:55:09amiconnkugel: It's not about finding the variant to use, but how to include all the variants
23:56:15kugelcompiling small libraries?
23:56:47amiconnRight now the code is compiled depending on what ARM_ARCH is set to
23:57:27 Quit wodz (Quit: Leaving)
23:58:03amiconnThere's even a difference in data types between armv4 and all higher ones (armv4 uses 32 bit filter coefficients, v5 and higher uses 16 bit like all other archs and the default C code)

Previous day | Next day