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

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

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

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

#rockbox log for 2011-04-24

00:09:07 Quit leavittx (Ping timeout: 276 seconds)
00:19:20TheSevenhm, will interrupts be disabled when an abort happens?
00:19:28 Quit mc2739 (Ping timeout: 260 seconds)
00:19:50 Quit Keripo (Quit: Leaving.)
00:19:51 Quit linuxstb (Ping timeout: 246 seconds)
00:20:42TheSevenapparently IRQs are disabled, but not FIQs
00:20:59 Join linuxstb [0] (~linuxstb@rockbox/developer/linuxstb)
00:21:27 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
00:28:50CIA-87New commit by 03Buschel (r29771): get_sample_info() has been removed.
00:31:54CIA-87r29771 build result: All green
00:32:33 Quit Buschel (Quit: ChatZilla [Firefox 3.6.16/20110319135224])
00:36:35 Join Keripo [0] (
00:38:50 Join clmoe [0] (442fd73e@gateway/web/freenode/ip.
00:39:05clmoeCan someone help me with my clipp?
00:43:31clmoeThink it's somewhat bricked, not sure how it got there
00:43:47clmoeall I know is that it had sugar inside of it
00:45:36 Quit [Saint] (Ping timeout: 260 seconds)
00:46:18 Quit lovasoa (Quit: Quitte)
00:49:26 Quit wavedigit (Read error: Connection reset by peer)
00:49:51 Join wavedigit [0] (
00:54:11krazykitclmoe, sugar? was there water, too?
00:54:23clmoeNo clue
00:54:34clmoenot any when I looked at least
00:54:50clmoeRight now I've got access to the onboard flash, and the screen is black
00:54:59clmoeI'm not sure if recovery mode is what I want to do
00:55:01krazykithow on earth did you get sugar inside your mp3 player?
00:55:10clmoeI'm also on wincoolze so not sure about that sudo stuff...
00:55:16clmoeI bought a broken clipp
00:55:22clmoeit came like this
00:55:33clmoeI suspect it could have also been sand or salt, but it tasted sweet
00:55:39clmoemaybe cocaine? who knows
00:55:44clmoeI got it all out of there
00:56:40 Join [Saint] [0] (~st.lasciv@
01:02:33pamauryThe fuze+ lcd controller is responding \o/ I have kind 7783 whatever it means ;)
01:08:51clmoeSo what exactly would putting it in recovery mode do?
01:10:04gevaertsclmoe: I assume you saw ?
01:10:15clmoethat's how I ended uo here
01:10:25clmoenot sure exactly how I should go about doing this
01:10:56gevaertsThis doesn't seem to be a time when people who've done this are online apparently
01:11:41 Join GeekShadow [0] (~Antoine@
01:11:41 Quit GeekShadow (Changing host)
01:11:42 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
01:13:25clmoeGuess I'll idle out for a bit and try later
01:14:25AlexPIn about 8 or 9 hours when people wake up
01:14:51gevaertsWell, Sundays aren't our busiest days either
01:19:49 Quit maraz (Ping timeout: 250 seconds)
01:19:50 Quit bluebrother (Disconnected by services)
01:19:52 Join bluebroth3r [0] (~dom@rockbox/developer/bluebrother)
01:28:05 Join bertrik [0] (
01:28:05 Quit bertrik (Changing host)
01:28:05 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
01:33:35*bertrik finds an apple document describing the ipod accessory protocol
01:37:05pamaurypublic one ?
01:37:19bertrikkind of
01:38:18bertrikit's available to read on some chinese website, but haven't figured out yet how to download it:
01:38:52pamauryomfg no, I've just loose one hour because (1 << 16) = 0x10000 and not 0x1000 :'(
01:44:52***Saving seen data "./dancer.seen"
01:45:11 Quit ChickeNES (Quit: Computer has gone to sleep.)
01:52:53pamaurywtf ? this is a strange init sequence: reset, wait 50 ticks, clear reset, wait 20 ticks, reset, wait 200 ticks, send lots of commands
01:57:45saratogaalso related
01:58:08bertrikyes, I have the pdf for the R30 now
01:58:56bertrikNot having to reverse engineer stuff, should make iap a lot easier now. Also we can probably fix a couple of bugs with info like this
01:59:59 Quit n1s (Ping timeout: 264 seconds)
02:00:34saratogaah was just going to say google has a direct link to that pdf
02:02:07 Quit MethoS- (Remote host closed the connection)
02:03:45saratogadoes the extended interface document have anything uesful in it?
02:04:57saratogaah I see, the extended interface is for sending the UI to other devices
02:05:14 Quit robin0800 (Quit: Leaving)
02:06:15saratoga"IAP over USB"
02:06:23saratogaoh and it explains how they detected chargers
02:06:46pamaurywow, the lcd init sequence is scary :-s
02:06:54bertriksoap, are you familiar with these documents?
02:08:24saratogawow theres a LOT of protocols
02:08:34saratogaand it looks like several can be over UART or USB HID
02:08:54saratogano wonder devices randomly do'nt work
02:09:38saratogagah is there some way to remove this stupid overlay from the PDF so I can actually read it?
02:14:37 Join maraz [0] (
02:19:46TheSevennice leaks :)
02:20:12 Quit Strife89 (Quit: Rearranging network hardware.)
02:20:25pamaurywhy a lcd init sequence needs 40 commands ??
02:22:10 Join Strife89 [0] (~Strife89@
02:22:34bertrikI don't know, maybe they want to make sure that they configure each and every one register of the lcd I guess
02:23:24 Quit wavedigit (Ping timeout: 258 seconds)
02:24:12 Join wavedigit [0] (
02:24:16pamauryand I'm still not done, the lcd doesn't show anything yet, I need to setup the lcd block and to have backlight
02:24:40 Join Strife89-iBook [0] (~Strife19@
02:30:09 Quit pamaury (Remote host closed the connection)
02:55:50 Quit bertrik (Ping timeout: 248 seconds)
03:16:50 Quit clmoe (Quit: Page closed)
03:24:03 Quit pepsi_ (Ping timeout: 240 seconds)
03:29:33JdGordonsaratoga: nice, how do you download the .pdf though??
03:31:49 Join pepsi [0] (~pepsi@unaffiliated/jbutera)
03:44:55***Saving seen data "./dancer.seen"
03:56:36 Quit Judas_PhD (Quit: This is a quitting message)
03:58:00 Join Judas_PhD [0] (
03:59:07 Quit Judas_PhD (Client Quit)
03:59:40JdGordonanyone got a wheel target and a non-wheel target avilable?
04:04:11 Quit saratoga (Changing host)
04:04:11 Join saratoga [0] (9803c6dd@rockbox/developer/saratoga)
04:06:22 Join webguest685 [0] (
04:06:53 Quit webguest685 (Client Quit)
04:09:16 Quit sasquatch (Ping timeout: 276 seconds)
04:11:10Strife89JdGordon: Wheel target: Fuze and iPod Video. Non-wheel: Clip+ and c250
04:12:01JdGordonyeah, dont worry, built some sims which is good enough :)
04:13:39 Join pepsi_ [0] (~pepsi@unaffiliated/jbutera)
04:16:28 Quit pepsi (Ping timeout: 252 seconds)
04:20:33 Quit Keripo (Quit: Leaving.)
04:28:10 Part pepsi_
04:29:30 Join Beta2K [0] (
04:30:53 Join Keripo [0] (
04:34:18 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
04:34:20 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
04:34:29 Quit amiconn (Disconnected by services)
04:34:29 Quit pixelma (Disconnected by services)
04:34:36 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
04:41:21 Join kugel_ [0] (~kugel@rockbox/developer/kugel)
04:42:41 Quit kugel (Read error: Operation timed out)
04:43:21 Quit TheSeven (Disconnected by services)
04:43:30 Join [7] [0] (~TheSeven@rockbox/developer/TheSeven)
04:44:00 Quit wavedigit (Quit: Leaving)
04:48:59 Quit GeekShadow (Quit: The cake is a lie !)
05:18:06 Quit Keripo (Read error: Connection reset by peer)
05:21:33 Quit mystica555_ (Ping timeout: 248 seconds)
05:23:35 Join Keripo [0] (
05:26:23 Join Rob2222 [0] (
05:30:06 Quit Rob2223 (Ping timeout: 264 seconds)
05:35:03 Quit Keripo (Remote host closed the connection)
05:36:46 Join Keripo [0] (~keripo@
05:41:13 Join mystica555 [0] (
05:44:59***Saving seen data "./dancer.seen"
05:46:22 Join mystica555_ [0] (
06:08:54 Join Horschti [0] (~Horscht@xbmc/user/horscht)
06:12:35 Join sasquatch [0] (~username@
06:12:45 Quit Horscht (Ping timeout: 248 seconds)
06:15:55 Quit mt (Read error: Connection reset by peer)
06:16:30 Join mt [0] (~mtee@rockbox/developer/mt)
06:32:57 Join BHSPitMonkey [0] (~stephen@unaffiliated/bhspitmonkey)
06:44:09 Quit [Saint] (Disconnected by services)
06:44:11 Join S_a_i_n_t [0] (~st.lasciv@
07:03:03 Join ChickeNES [0] (~ChickeNES@
07:05:12 Join cscho0415 [0] (
07:06:19 Quit cscho0415 (Client Quit)
07:13:31 Join cscho0415 [0] (
07:14:25cscho0415do u guys support the ipod classic 6th gen
07:23:02 Quit S_a_i_n_t (Disconnected by services)
07:23:04 Join [Saint] [0] (~st.lasciv@
07:23:25cscho0415JdGordon: what do u mean by almost?
07:23:39saratogawe don't support it yet
07:23:52saratogai.e. its on the unusable list here:
07:23:54JdGordoniiuc its working but not technically supported
07:24:05saratogabut hopefully it'll be supported soonish
07:24:16cscho0415whats wrong with it that it isnt supported?
07:24:33saratogacheck the wiki
07:24:42saratogaIIRC battery life and general stability
07:24:50cscho0415i saw it...
07:24:59cscho0415is it hard to isntall?
07:25:05cscho0415is it worth it?
07:25:18cscho0415ok thank you
07:25:23saratogahaha just kidding
07:25:32saratogaseriously read the instructions and see if you think its worth trying
07:25:41saratogai have no idea if its worth it for you
07:25:44cscho0415im just worried about bricking my ipod
07:25:56saratogai don't think you can brick it, although fixing it can be fairly complicated
07:26:15cscho0415i mean to fix it is it a simple restore or is it a pain
07:26:16saratogaIIRC, this is also something you should probably read about if you're concerned
07:26:31cscho0415ok thank you
07:44:12 Quit cscho0415 (Quit: cscho0415)
07:45:01***Saving seen data "./dancer.seen"
07:47:53 Join stoffel [0] (
07:59:00 Quit liar (Read error: Connection reset by peer)
07:59:39 Join liar [0] (
08:09:45 Quit jae (Ping timeout: 276 seconds)
08:11:09 Quit sinthetek (Ping timeout: 250 seconds)
08:21:27 Join jae [0] (
08:28:51 Quit jae (Ping timeout: 260 seconds)
08:30:51 Quit saratoga (Ping timeout: 252 seconds)
08:34:08 Join sinthetek [0] (~sinthetek@unaffiliated/sinthetek)
08:34:50 Join smk [0] (~smk@
08:45:45 Join jae [0] (
08:57:43JdGordonour troll is in the forums
09:02:00 Quit jae (Read error: Operation timed out)
09:02:06 Join jae [0] (
09:04:55CIA-87New commit by 03jdgordon (r29772): invert the quickscreen setting direction on wheel targets for settings which are number lists so pressing up makes it bigger (i.e contrast). fixes ...
09:06:09JdGordonscorche|sh: keep an eye on;u=29210 or just premptive ban him?
09:06:13 Quit BHSPitMonkey (Remote host closed the connection)
09:07:06JdGordonand as much as he is a dick, he is right about the manual being too long (which I've been saying for ages)
09:07:12 Quit jae (Read error: Operation timed out)
09:07:17 Join jae [0] (
09:07:27scorchebecause ctrl+f is so hard...
09:07:37[Saint]Um...why am I banned from the forums?
09:07:50[Saint]"I have spammed for the last time"...WTF!?!
09:08:37CIA-87r29772 build result: All green
09:09:03*[Saint] doesn't see how the manual could be much more concise.
09:09:06scorcheJdGordon: why are we removing this person's posts?
09:09:27scorchesure, i see he seems a bit...misguided, but what has he done wrong
09:09:51JdGordonapart from us knowing he is a troll, he replied to a completly unrelated thread
09:09:56 Quit Keripo (Remote host closed the connection)
09:09:59JdGordon[Saint]: unbanned... dunno who banned you though :D
09:10:11*[Saint] has an idea :P
09:10:55JdGordonanyway, just saying keep an eye on him as i'll be afk
09:11:12 Quit balintx (Remote host closed the connection)
09:11:27 Join balintx [0] (
09:11:32 Join Keripo [0] (
09:15:06[Saint]I've looked for places to cut the fat out of the manuals, but for the life of me I can never seem to find much of an opportunity to cut out more than a few words here and there from a poorly (yet perfectly acceptable still) worded sentence.
09:15:22[Saint]there really isn't much room to trim it down.
09:16:34[Saint]There's a "Quick Start” section...what more do people want?
09:16:44scorchethis is why there are chapters...and why you should be able to search through it...
09:17:58[Saint]Searching would be nice, yeah.
09:18:19[Saint]Bottom line is the manual is huge because there’s a foolish amount of settings/variables to explain.
09:18:37[Saint]it *needs* to be the size it is, it’s impressive it’s not bigger really.
09:37:22 Join n1s [0] (
09:37:22 Quit n1s (Changing host)
09:37:22 Join n1s [0] (~n1s@rockbox/developer/n1s)
09:37:50 Join sideral [0] (~sideral@rockbox/developer/sideral)
09:45:03***Saving seen data "./dancer.seen"
09:49:41 Join mudd1 [0] (
09:53:09 Quit Keripo (Read error: Connection reset by peer)
09:57:07 Join thomasjfox [0] (~thomasjfo@rockbox/developer/thomasjfox)
09:57:27 Join ender` [0] (
10:15:31 Join leavittx [0] (~leavittx@
10:26:46 Quit leavittx (Ping timeout: 250 seconds)
10:30:17 Join cig [0] (
10:32:33 Quit cig (Client Quit)
10:33:10*bluebroth3r suggests JdGordon to write a new manual that doesn't have the size "bug"
10:33:40bluebroth3rthough I realize the size can scare of first time users I don't see a way to make it shorter.
10:33:46bluebroth3rmaybe we have too much features?
10:34:00smkhi. is it possible to just get the source code of the android port of rockbox, without having to download the entire tree by svn?
10:34:33bluebroth3rThere is one thing I could think about: split the manual. Either split out the plugins or split it up into a Quickstart, Plugins and everything else part
10:34:47bluebroth3rsmk: you can check out files / folders from svn selectively.
10:34:59bluebroth3ri.e. you can svn co svn://
10:35:27bluebroth3rbut that won't build because the android port uses the Rockbox sources (and the build system and tools etc)
10:36:03bluebroth3rso except for manual rbutil and utils there isn't much point in not getting the other folders
10:36:16bluebroth3rhmm, you might also be able to leave out wps backdrops and fonts
10:37:36bluebroth3rwow, what this guy describes as "colossal cheek" is his post by itself. Too bad it was removed, I could have written a german response :)
10:38:02 Quit linuxstb (Ping timeout: 250 seconds)
10:38:38bluebroth3rsmk: and once you have checked out the sources updating doesn't require much data to be transmitted so I wouldn't go for getting parts of the tree for working on int
10:39:16bluebroth3r[Saint]: re the logos: maybe put the logo original sources into docs/?
10:44:08 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
10:50:06scorchebluebroth3r: i could always un-remove it and move it to its own thread (which probably should have been done anyway...)
10:51:30 Join factor [0] (
10:52:54bluebroth3rscorche: well, not responding to it is less work for me (this post is really completely off the mark)
10:53:23scorchebluebroth3r: i would be surprised it if is his last post
10:58:44bluebroth3rtrue, so I'm likely to get another chance some time later :)
11:00:01bluebroth3rthough maybe it might be a good idea to let it pass :)
11:02:08bertrikbluebroth3r, I think this guy is "Florian" on irc
11:02:36bertrikhe was making trouble yesterday on irc, so he got muted and finally banned
11:02:40bluebroth3rah, the one that wanted to get kicked yesterday?
11:04:27 Nick [7] is now known as TheSeven (~TheSeven@rockbox/developer/TheSeven)
11:06:05bertrikbluebroth3r, it very much seems so, see
11:07:15 Part thomasjfox ("Konversation terminated!")
11:21:05 Join robin0800 [0] (
11:22:52bluebroth3rbertrik: I agree that Rockbox is a hoax! :)
11:26:30 Join dv [0] (
11:27:16dvtrying to build libtremor separately... did anybody try?
11:27:42 Part smk ("Leaving")
11:31:09 Join smk [0] (~smk@
11:31:41 Quit sideral (Quit: Leaving.)
11:32:54smkbluebroth3r: since the android port is basically an app, it would be independent of the firmware code right?
11:33:46smkby firmware, i mean the kernel.
11:34:16 Quit robin0800 (Quit: Leaving)
11:40:37 Part smk ("Leaving")
11:43:33 Join robin0800 [0] (
11:45:04***Saving seen data "./dancer.seen"
11:47:10 Quit n1s (Ping timeout: 240 seconds)
11:56:00 Quit plux (Quit: qux)
11:56:20 Join TheSphinX^ [0] (
11:57:08 Quit TheSphinX^ (Read error: Connection reset by peer)
11:58:35 Join lovasoa [0] (~lovasoa@2a01:e35:8a2e:8080:226:bbff:fe15:33b0)
12:00:57bertrikhi lovasoa
12:00:59lovasoaI'm currently hacking on iap.c (iPod accessory protocol implementation), and there are two commands that I don't know how to implement: get total number of playlists, and get all playlists names
12:01:37lovasoaCan someone tell me which functions I should use (and in which header file they are) ?
12:03:16 Join plux [0] (
12:10:12 Quit robin0800 (Quit: Leaving)
12:18:11 Join n1s [0] (~n1s@rockbox/developer/n1s)
12:21:13lovasoaI'm currently hacking on iap.c (iPod accessory protocol implementation), and there are two commands that I don't know how to implement: get total number of playlists, and get all playlists names. Can someone tell me which functions I should use (and in which header file they are) ?
12:25:15 Join Stummi [0] (~Stummi@rockbox/developer/Stummi)
12:25:18 Nick kugel_ is now known as kugelp (~kugel@rockbox/developer/kugel)
12:25:49 Join robin0800 [0] (
12:26:57 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
12:30:33bertriklovasoa, I don't know, but it's nice to see people hacking on rockbox :)
12:32:03[Saint]bluebroth3r: regarding building for Android, you pretty much need to grab everything in the svn tree.
12:32:47[Saint]If you’re building for a res that happens to have a non-RaaA target you’ll need themes and backdrops too
12:32:49AlexP[Saint]: That's what he said
12:33:04AlexPPretty much :)
12:33:17[Saint]AlexP: the second part was the more critical part there.
12:33:31AlexPAnyway, I'm sure he knows
12:33:38AlexPLike term dev and all that :)
12:33:46bluebroth3r[Saint]: personally I don't see much point in getting only parts of the repository anyway
12:34:14[Saint]AlexP: it was regarding “[20:34] <bluebroth3r> hmm, you might also be able to leave out wps backdrops and fonts”
12:34:27bluebroth3rexcept maybe for automated building (as for Rockbox Utility where I have only get the required parts via svn export to avoid transferring more data and to make sure my list is complete :)
12:34:51bluebroth3rI said "might" for a reason ;-)
12:35:04AlexP[Saint]: You probably don't need that, I'd expect you'ld get built in fallback
12:35:24AlexPWhich depending on if you are just building for testing might be acceptable
12:35:37AlexPAnyway, we all agree just get everything :)
12:36:20 Quit stoffel (Ping timeout: 240 seconds)
12:36:45bluebroth3rAlexP: any thoughs on putting the Rockbox Logo original files (png, svg) in docs/?
12:36:57AlexPWhere are they now?
12:37:09bluebroth3rscattered around :)
12:37:22AlexPYes then :)
12:37:49AlexPIt'd be nice to have them collated, and I guess docs/ is as good as any and better than most :)
12:37:53bluebroth3rrockbox-clef.svg is in the rbutil tree, but also used by android (though a copy converted to png). etc. pp. :)
12:37:59bluebroth3ryes, that was my thought.
12:39:00bluebroth3rI also have the "Rb" logo as svg around that I made from the Rockbox svg some time. It's not used anywhere yet but I'd like to commit that −− the "Rb" logo is part of the ico file for Rockbox Utility on Windows (and Mac IIRC) but that was made before the svg existed.
12:39:10AlexPI remember looking for them a while ago, and not finding them
12:39:11bluebroth3rnevertheless I think it would be good to have it around
12:39:36AlexPA subdir of docs seems good
12:39:39bluebroth3rthe "unused" "Rb" one that is
12:39:57*bluebroth3r likes the latter
12:40:20 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
12:40:34AlexPyeah, sounds sensible
12:43:47 Nick kugelp is now known as kugel (~kugel@rockbox/developer/kugel)
12:44:31kugelputting them in docs/ makes them even more scattered around, no?
12:45:09kugelalso, docs/ should have documents...not icons IMO
12:45:14[Saint]scattered around compred to?
12:45:58 Join stoffel [0] (
12:48:37AlexPkugel: How is all being in one place more scattered than all over the tree?
12:49:56bluebroth3rhaving the _original_ files (mostly svg) in one place is an improvement IMO.
12:50:21bluebroth3rif f.e. Android needs a png version of the icon then it's fine to have that separately to avoid converting it all the time.
12:50:37bluebroth3rbut having like the Android icon original in the rbutil tree is a bit weird at least
12:51:19 Quit sinthetek (Ping timeout: 260 seconds)
12:51:20bluebroth3rI'm not completely sure if it makes sense to put e.g. the Windows ico file over there as well or only keep the original sources
12:51:54 Join DerPapst [0] (
12:52:03*bluebroth3r considers the logo as document as well
12:57:51 Join sinthetek [0] (~sinthetek@unaffiliated/sinthetek)
13:06:07 Quit n1s (Ping timeout: 240 seconds)
13:08:57 Quit AlexP (Remote host closed the connection)
13:09:44 Quit [Saint] (Quit: I'm only going to Heaven if it feels like Hell, I'm only going to Heaven if it tastes like caramel...)
13:10:40 Quit lovasoa (Read error: Operation timed out)
13:12:09 Join [Saint] [0] (~st.lasciv@
13:15:13 Join AlexP [0] (~alex@rockbox/staff/AlexP)
13:17:23 Join lovasoa [0] (~lovasoa@2a01:e35:8a2e:8080:226:bbff:fe15:33b0)
13:45:06***Saving seen data "./dancer.seen"
13:49:33 Quit stoffel (Ping timeout: 240 seconds)
13:57:56 Join smk [0] (~smk@
14:04:15 Join leavittx [0] (~leavittx@
14:06:38 Quit liar (Read error: Connection reset by peer)
14:07:33 Join liar [0] (
14:08:21 Quit user890104 ()
14:11:28pamauryI've found the backlight on the fuze+ \o/
14:15:32lovasoaSomeone knows how to get the total number of playlists?
14:17:02 Join user890104 [0] (
14:18:03smkhi. m unable to configure rockbox for android. m getting the ANDROID_SDK_PATH variable error, inspite of having set it to the correct value
14:18:08lovasoaSomeone knows how to get the list of all of playlists?
14:19:18smki am on linux. and the error goes like this:
14:19:18smkERROR: You need the Android SDK installed and have the ANDROID_SDK_PATH
14:19:19smkenvironment variable point to the root directory of the Android SDK.
14:21:06 Join JackWinter [0] (
14:23:27smki have the sdk installed and pointed the variable to the top directory of the sdk. still getting the error :(
14:24:23[Saint]so, silly question...but if you do "cd $ANDROID_SDK_PATH && pwd" it returns as expected?
14:25:07 Join clmoe [0] (442fd73e@gateway/web/freenode/ip.
14:25:54clmoeHey looks like some people might be awake today
14:26:36clmoeAnyone good with AMS hardware?
14:28:13 Join stoffel [0] (
14:28:15smkand the configure script doesn't use the SDK variable anywhere.. it's using the NDK variable.. but i see only one reference to the SDK path
14:29:12[Saint]smk: did you remember to do "./android" in the tools dir?
14:29:55jhMikeSlovasoa: there's no such thing as far as dynamic playlists
14:30:42lovasoaHmm? There is no way to know how much playlists are saved ?
14:31:21bluebroth3rshow should that work?
14:31:22jhMikeSscan for files then
14:31:26lovasoaIn "Playlist > View Catalog" where do the playlists come from ?
14:31:37bluebroth3rfrom /playlists
14:31:46smk[Saint]: no i hadn't start the AVD. but now i did, and it didn't solve the problem
14:31:47bluebroth3rbut a playlist can be anywhere on disk
14:32:03bluebroth3ryou don't need to start the AVD
14:32:06 Join clmoe_ [0] (442fd73e@gateway/web/freenode/ip.
14:32:20clmoe_>BSOD out of nowhere
14:32:21[Saint]smk: did you also update, or simply start it?
14:32:29bluebroth3ryou need both ANDROID_SDK_PATH and ANDROID_NDK_PATH set before running configure
14:32:33lovasoaSo, to talk to an accessory, I just have to list the content of /playlists ?
14:32:45jhMikeSthere's a playlists directory /Playlists
14:32:47smkbluebroth3r : both are configured
14:32:55bluebroth3raccessory? How is an accessory related to playlists?
14:33:05 Quit clmoe (Ping timeout: 252 seconds)
14:33:28lovasoaI am hacking iAP.c, to get my iPod dock to work with rockbox
14:33:31 Nick clmoe_ is now known as clmoe (442fd73e@gateway/web/freenode/ip.
14:33:49smk[Saint]: update what?
14:34:06lovasoaAnd the dock asks the iPod how many playlists there are, and their names
14:34:24[Saint]smk: Make sure it has the correct APIs to use.
14:34:40bluebroth3rah. Well, you can either limit yourself to the playlists from the playlist catalog (which are in /playlists) or you need to scan the disk
14:34:45jhMikeSlovasoa: you'll have to scan if you want to be sure you got 'em all if they're saved in alternative places
14:35:37*jhMikeS is colliding with bluebroth3r
14:35:50[Saint]smk: this is why InstallToolchain does "$SDK_PATH/tools/android update sdk −−no-ui −−filter platform,tool"
14:35:57lovasoaThis risks to be long... The browsing from the accessory risks not to be fluent
14:36:26bluebroth3rshouldn't take long if dircache is enabled
14:36:29smk[Saint] : i have API 8 and API 11.
14:37:07bluebroth3rthis sounds to me like the ANDROID_SDK_PATH variable is wrong
14:37:19*jhMikeS alwasy balks @ "this feature works nicely _iff_ dircache is enabled"
14:37:51smkdoes having 2 APIs cause some conflict or something?
14:38:04bluebroth3rsmk: no
14:38:28bluebroth3rlovasoa: you could make it scan /playlists only if dircache is disabled, and query dircache if enabled
14:38:29smkbluebroth3r: cd $ANDROID_SDK_PATH goes to the correct directory..
14:38:52lovasoaAnd is there a function which do that in playlist.c or I need to write my own function ?
14:39:07bluebroth3rsmk: how exactly does "correct" look?
14:39:21bluebroth3ralso, is that variable set or exported?
14:40:16smki have set it in my bashrc
14:41:01bluebroth3rset or exported?
14:41:22[Saint]pastebin the relevant section, so we can rule that out
14:41:32smknow i exported :)
14:41:40smkworked. :P
14:41:52smksad mistake :(
14:41:56bluebroth3rthat's why I asked about _set_ or _exported_
14:42:07*bluebroth3r fell into that trap himself some time ago
14:42:14*[Saint] too
14:42:41clmoeso, what exactly does recovery mode for the ams do?
14:43:00[Saint]In this case though, I looked at what installToolchain was trying to do, and did it manually...WFM
14:43:34smkbluebroth3r , [Saint] : hmm i thought putting it in bashrc eliminates the export part as it makes the variable available over the system..
14:45:58[Saint]smk: if it's being used by processes other than the shell itself, you need to export it.
14:48:10[Saint]An expert is more than happy to correct me, but as far as I know, that's how it works
14:48:33[Saint]s/is/should be/
14:49:07smktrue maybe.. but i never exported a var after putting it in bashrc.. but then i also never used it in processes other than shell :P
14:50:08 Join wavedigit [0] (
14:51:07wavedigithi, if I get the sad ipod logo at startup, and disk mode doesn't work, is it the only possibility that the HDD is dying?
14:51:20wavedigit(iPod 80gb 5g)
14:51:44[Saint]that's not the *only* possibility
14:51:51[Saint]...but, it's definitely a possibility
14:51:54bluebroth3rhow does "disk mode doesn't work" look like?
14:52:47bluebroth3rdisk mode _always_ works. If not you have a hardware failure, and in that case I wouldn't bet it's only the hard drive
14:53:42kugelsmk: you need api 4
14:53:43 Quit user890104 (Read error: Connection reset by peer)
14:53:47 Join user890104_ [0] (
14:54:02smkkugel : it's building far.
14:54:28 Nick user890104_ is now known as user890104 (
14:54:48kugelapi 4 for the ndk, we use api 11 (in a backward compatible way) for the java part. IIRC
14:55:10kugelbut then the ndk .zip ships all APIs so you're fine
14:56:17 Quit wavedigit (Ping timeout: 240 seconds)
14:56:52 Join wavedigit [0] (
14:56:55wavedigitdamn wifi
14:57:37wavedigitso as I was saying, disk mode won't work, so it's probably a hardware issue
14:58:02 Quit ps-auxw (Ping timeout: 248 seconds)
14:58:23wavedigitand seeing as if HDD is not the only probable cause, it became much harder to troubleshoot
14:58:37gevaertswavedigit: the 5g (maybe others too) also has a diagnostic mode
14:58:41[Saint]please define "won't work"
14:59:26 Join ps-auxw [0] (~arneb@2001:470:c807:0:1532:4e5f:2ad3:4123)
14:59:39wavedigit[Saint], the commands I give the ipod to go in disk mode (boot, press menu+center, afterwards press center+play/pause), only give me a sad ipod logo with a link to apple support
15:00:10wavedigitgavaerts, I'm checking google/rockbox as we speak, thanks
15:01:36[Saint]diagnostic mode == menu+select, then previous+select
15:02:32wavedigitoh great I'll try that
15:03:11 Join ChickeNE_ [0] (~ChickeNES@
15:03:59wavedigitit has the same behavior, first I see a dim apple logo, then the screen turns white with a sad ipod logo
15:04:28[Saint]your iPod, sir, is borked.
15:05:00gevaertsWhat happened to it? That's not just a bad disk...
15:05:29wavedigitI opened it up to change the front cover with an ivue clear panel
15:05:30 Quit ChickeNES (Ping timeout: 248 seconds)
15:05:32[Saint]No, it should be able to access disk mode with a bad disk.
15:05:48wavedigitobviously I did something wrong
15:05:50gevaertsMaybe a loose cable somewhere else in there?
15:06:30Strife89[Saint]: That sounds like my iPod Color's pre-death symptoms ....
15:06:39wavedigitpossibly, or I messed the mainboard up with some kind of static electricity
15:07:04 Join linuxstb [0] (~linuxstb@rockbox/developer/linuxstb)
15:07:05wavedigiteither way the HDD is most important, I hope it still has a life somewhere in it
15:07:07Strife89Or maybe the ribbon cable itself is broken
15:07:25[Saint]ESD isn't exactly a myth, but generally speaking it's pretty hard to kill something with it.
15:07:37wavedigitstrife89, could that mess up the HDD as well?
15:08:23Strife89It's possible
15:09:14wavedigitI'll re-open and try to re-connect everything, but if that fails I have to go on a hardware hunt
15:09:21Strife89I'd take my chances and try reseating the HDD/cable.
15:09:56wavedigitluckily there's someone who's selling loads of ipod 5gen hardware parts 2nd hand
15:10:00gevaertswavedigit: don't pay *too* much attention to what Strife89 can break. he's better at this than most people...
15:10:31wavedigitlol as of now I pay most attention to the fact that I don't have any music when I'm traveling :p
15:11:01wavedigit(unless you consider surrounding noise as music, but that's a bit of a different discussion here)
15:11:22 Quit krazykit (Remote host closed the connection)
15:12:08wavedigitin any case, thanks for all the advice here :)
15:12:55 Join n1s [0] (
15:12:55 Quit n1s (Changing host)
15:12:55 Join n1s [0] (~n1s@rockbox/developer/n1s)
15:16:00Strife89wavedigit: Don't pay *too* much attention to what gevaerts says about me. I usually fix >=half the stuff I break. ;)
15:18:36wavedigitstrife89, I usually break all the stuff I try to fix :(
15:19:51 Join Buschel [0] (
15:32:38*smk finally runs rockbox.apk on avd :)
15:33:05clmoeCan anyone tell me what exactly recovery mode on a clipp would be doing?
15:39:36[Saint]clmoe: I think you perhaps need to be more (maybe even less?) specific...
15:39:53clmoeI bought a broken clipp for cheap
15:40:01clmoeit had a bunch of what seemed to be sugar inside it
15:40:06clmoeno clue what's going on with that
15:40:28clmoeright now I've got no responce from the screen, but if I plug it in then I get the onboard flash
15:40:39 Part smk ("Leaving")
15:40:53clmoeI copied all the data from my current working clip w/ rockbox but it still doesn't do anything
15:41:08clmoeI'm wondering if I would need recovery mode to make any progress
15:41:16n1sso it's working but the screen is dead?
15:41:26n1ssounds like a hw issue to me
15:41:28clmoeno, I did get some responces from the screen at one point
15:41:40clmoedidn't get past the sansa logo though
15:41:42clmoenow it's just nothing
15:41:56clmoeI thought the screen was dead at one point myself as well
15:42:05bertrikclmoe, do you see a drive with the full original capacity?
15:42:12 Quit wavedigit (Ping timeout: 250 seconds)
15:42:23clmoe3.68 for a 4gb
15:42:37clmoeno clue if that's the correct rounding or not
15:43:57bertrikthat does not appear to be the recovery mode
15:45:00clmoeso I'll have to get in to recovery mode to replace the firmware?
15:45:09***Saving seen data "./dancer.seen"
15:45:16 Quit mt (Ping timeout: 260 seconds)
15:45:41bertrikno, I think there's just something wrong with the hardware
15:46:33clmoeWell is there anything I can do about that?
15:46:44[Saint]attempting the unbrick procedures couldn't hurt at this point.
15:47:19gevaertsYou could work on the hypothesis that it's a broken screen, try to install rockbox blindly, and put a voice file on it
15:47:21clmoeThat's what I figured, doesn't really matter if I don't get this fixed but I'd like to try
15:47:39clmoebut the screan was working at one point
15:47:40bertrikI disagree, that would only break it further IMO
15:47:52clmoeit doesn't matter if it gets broken further
15:48:02clmoeit's not working right now and I've got two other clips to work with
15:48:13[Saint]gevaerts: If it were a broken screen, wouldn't the original FW files be visible on the disk?
15:48:20[Saint]assuming the OF has booted at some stage
15:48:58clmoeI think I stopped getting the OF screen after I resoldered the battery connections back on, one came off at one point
15:49:16clmoewhen I started getting disk access it had to be formatted
15:49:59clmoeI figured if there was anything terribly important on there I could image my other clipp and went ahead and did that
15:51:08 Quit sinthetek (Read error: Operation timed out)
15:53:13 Join clmoe_ [0] (442fd73e@gateway/web/freenode/ip.
15:54:05CIA-87New commit by 03Buschel (r29773): Clean up array sizes in the SBR + PS part of libfaad. Add MAX_NTSRPS (=38), use MAX_NTSR and MAX_NTSRHFG at the right places. RAM (IRAM) usage is ...
15:55:38 Quit Rondom (Quit: Bye)
15:56:09 Join Rondom [0] (
15:56:40 Quit clmoe (Ping timeout: 252 seconds)
15:56:47 Nick clmoe_ is now known as clmoe (442fd73e@gateway/web/freenode/ip.
15:57:11 Quit Rondom (Client Quit)
15:57:17CIA-87r29773 build result: All green
15:57:40 Nick FoolOnHill is now known as EasterBasketCase (
16:00:20clmoeam I gonna have to desolder the battery to do recovery or can I just hold the power button?
16:00:22 Join sinthetek [0] (~sinthetek@unaffiliated/sinthetek)
16:08:38 Join mshathlonxp [0] (
16:08:49bluebroth3ris it just me or what happened with the Rockbox png logo? I only see the "box" part:
16:09:22[Saint]Not just you...
16:09:33clmoeIt's changing to ibox
16:09:38clmoegotta keep up wit the times
16:10:09bluebroth3rthat file hasn't been changed since 5 years :o
16:11:17 Join Rondom [0] (~rondom@2a01:488:66:1000:b24d:4f2f:0:1)
16:11:27bluebroth3rrockbox-web.png looks ok, and I created that from the big one IIRC
16:13:10bluebroth3rat least it does show the greyscale parts that have been in the original logo but aren't present in the svg version
16:14:23bluebroth3rah, possibly I've taken that from the web. That has the same dimensions ...
16:14:34 Join LambdaCalculus37 [0] (
16:14:34 Quit LambdaCalculus37 (Changing host)
16:14:34 Join LambdaCalculus37 [0] (~rmenes@rockbox/staff/LambdaCalculus37)
16:14:45bluebroth3rthough I think to remember the original sized png to look correctly at some point in the past
16:24:04 Join captainkwel [0] (
16:24:30 Join wodz [0] (
16:24:52 Quit Buschel (Ping timeout: 258 seconds)
16:26:33wodzI am facing such problem when trying to compile for arm7ej-s (armv5tej core):
16:26:45wodzany expert can help?
16:27:46wodzI selected ARM_V5 (or something like this)
16:31:04pamaurythat's strange
16:32:46wodzI am quite new to arm so almost everthing is strange for me :-)
16:33:22pamaurywodz: shouldn't there be a -mcpu=.. switch on the command line ?
16:33:44wodzthere is no mcpu for arm7ej-s
16:33:46pamauryalso -march=armv5t seems strange for armv5tej
16:34:24 Quit captainkwel ()
16:34:45wodzhmm maybe armv5te than
16:35:44pamauryI think the "e" is for enhanced DSP instruction, perhaps this is one of those, I don't know this instruction
16:36:03 Join Artursk [0] (
16:38:07ArturskI'm triying to unbrick sansa clip,but dd command doesnt work
16:40:13wodzpamaury: yeh, -march=armv5te did the trick
16:40:39pamauryactually perhaps there is no -mcpu for it because it's the only cpu for this arch ;)
16:41:52pamauryhum, no, strange there is no mcpu for it then
16:42:01ArturskIt says that there is no such a file ,probably im getting wrong divice its only 32 mb large - or that is normal for a Clip +
16:46:16 Quit mshathlonxp (Quit: Leaving)
16:46:38 Join CaptainKewl [0] (
16:46:44[Saint]Artursk: If you're reading the SansaAMSUnbrick wiki page, you can see that's incorrect.
16:47:02[Saint]it should mount as a 979.75MB volume
16:47:44jhMikeSisn't that a recovery partition of some sort (the 32MB)?
16:48:12TheSevenjhMikeS: yes, but I'm still investigating how to use it :)
16:51:29[Saint]Artursk: Also, be *very* careful with the dd command.
16:51:43[Saint]incorrect use could see you lose all the data on your system
16:54:20Arturskyes i'm reading that page- not shure what im I doing wrong- maybe two points has to be already unbridged while tiping sudo fdisk -l ?
16:56:36*TheSeven is struggling with a weird makefile issue
16:56:44TheSeveni have a makefile that works roughly like this:
16:57:48TheSevenassuming everything is up to date, and i touch some source file in "subproject", and then run make on that makefile, it rebuilds the subproject, but doesn't rebuild anything else. why?
16:58:26 Quit CaptainKewl (Quit: ( :: NoNameScript 4.22 :: ))
16:59:23pamauryTheSeven: what would you expect ?
16:59:49 Join Leif [0] (
17:00:21 Join CaptainKewl [0] (
17:00:41TheSevensubproject/targetfile, patchedsubprojectfile, patchedsubprojectfile, intermediatefile and targetfile to be rebuilt as well
17:01:34pamauryTheSeven: the subproject/targetfile didn't change when first checking
17:01:45pamauryit only changes after the subproject rebuild
17:02:03TheSevenyeah, but it depends on the phony, and will have changed when the phony rule finishes building
17:02:43pamaurytargetfile depends on phony
17:02:53pamaurybut subproject/targetfile does not
17:02:58TheSeven"subproject/targetfile: subprojectphony"
17:03:20pamauryhuh, 1) not sure you can do this 2) this is not in your pastebin
17:03:35pamauryah, you updated it
17:03:54 Quit n1s (Ping timeout: 248 seconds)
17:04:34pamauryI think I would do "subproject/targetfile" -> subproject_targetfile; I'm not sure if Makefile works on such targets
17:06:04 Quit Artursk (Quit: CGI:IRC (EOF))
17:06:04TheSevenwhat do you mean?
17:06:22pamaurydon't write targets with a "/" in the middle
17:06:40TheSevenwell, that's a path, I can't do anything about that
17:07:25pamaurywhen you write this: "subproject/targetfile: subprojectphony", you are using a file as target, that's strange
17:07:36pamaurybecause it has a "/"
17:07:47pamauryperhaps Makefile doesn't like it
17:09:00TheSevenbut how else would you tell make that it needs to execute some sub-make to potentially regenerate that file if some of its deps have changed?
17:09:24pamauryuse a sub Makefile
17:09:35TheSevenyeah, that's what i'm doing!
17:09:36pamauryI already run into a similar problem and never sorted it out
17:09:50gevaertspamaury: proper paths aren't a problem for make
17:10:11TheSevengevaerts: the fun begins if those paths contain /./ and simiar things :)
17:10:32pamaurythen perhaps make doesn't propagate phony targets correctly
17:10:36TheSevenor rather *some* of them, as it seems to do string comparisons on them, and not compare the realpaths
17:11:27*gevaerts thinks that using phony targets internally is a bad idea
17:12:07*gevaerts also thinks that recursive make invocations are a bad idea
17:13:44TheSevenhow else would you do it?
17:13:54 Join u42p [0] (
17:16:35 Join Artursk [0] (
17:18:07gevaertsTheSeven: the way the rockbox build system does it, i.e. include the proper sub-makefile (if needed), and use regular dependencies
17:19:49TheSeventhe inner makefile is completely independent, it has to work standalone as well... does that work out?
17:20:05gevaertsThat would need some thinking...
17:20:56gevaertsOne way could be to have a subproject.make that gets included by both Makefile as subproject/Makefile
17:21:46 Join n1s [0] (
17:21:46 Quit n1s (Changing host)
17:21:46 Join n1s [0] (~n1s@rockbox/developer/n1s)
17:22:20gevaertsWhy don't you just call make for subproject/targetfile?
17:23:17TheSevenisn't that what I'm doing?
17:23:31 Quit z35 (Read error: Connection reset by peer)
17:24:07TheSevenyou mean making subproject/targetfile phony, moving the make command there, and removing the subproject rule?
17:24:47gevaertsI think I'd have to see the actual situation and experiment a bit before I could give really sensible answers...
17:24:53TheSevenactually i'd like the sub-makefile approach more, but I have no idea how to make this work and what side-effects it has
17:25:26[Saint]I think gevaerts just asked for freemyipod svn access ;P
17:25:37 Quit Artursk (Quit: CGI:IRC (Ping timeout))
17:25:53gevaerts[Saint]: I didn't :)
17:26:09gevaertsAnd I definitely don't need commit access to look at code
17:27:00TheSevenhm, how are relative paths in a submakefile inside a subdirectory interpreted?
17:27:09TheSevenwill they be relative to the sub or root makefile?
17:27:40gevaertsWhich is a bit annoying
17:27:49TheSevendammit :/
17:27:53 Quit Leif (Ping timeout: 240 seconds)
17:28:16gevaertsBut you could define $(ROOT) which indicates whichever root you want, and base stuff off that
17:28:18TheSevenwell, one could slap a prefix variable in front of every file name, which would be passed by the parent makefile
17:33:29 Quit DerPapst (Quit: Leaving.)
17:33:40TheSevenhm, moving the make -C invocation to the subproject/targetfile: subprojectphony rule seems to have fixed it
17:40:10 Join Leif [0] (
17:45:10***Saving seen data "./dancer.seen"
17:55:10[Saint]TheSeven: I realize you're a busy man, but, if you get you think you could look at the svn N2G bootloader? AFAIK you're one of the few that knows the code, and it'd be really good to get a consistent N2G bootloader released.
17:55:35[Saint]the current behaviour is a bit dated in the release N2G bootloader
17:55:35TheSevenwhat's wrong with it?
17:56:08TheSevenI don't think there were any changes in that area... does the SVN HEAD one work?
17:56:18[Saint]a: doesn't boot rb, b: missing features compared to svn (detection of a missing/corrupt binary)
17:57:28[Saint]I know that people with sense don't use the N2G bootloader...but, it's one of those :loose ends"
17:58:32TheSeventhe last time i've seen that thing, it was working perfectly fine... and it certainly should detect missing/wrong binaries
17:58:57TheSevenso you say that the svn head bootloader is broken (fails to boot rb)?
17:59:15[Saint]Hmmm, simply detect it, or display the (fairly recent-ish) "do something about it to fix it" warning?
17:59:28 Join z35 [0] (
17:59:33[Saint]And yes, it does indeed fail to boot rb.
17:59:40TheSevenhm, interesting
17:59:48TheSeveni haven't looked at that one for like a year
17:59:50[Saint]the release bootloader it literally years old iirc
18:00:11[Saint]it was just retageed for the last time it was released, nothing changed ;)
18:02:49 Quit LambdaCalculus37 (Quit: This computer has gone to sleep)
18:09:26 Join Highlander [0] (
18:11:00 Quit n1s (Ping timeout: 240 seconds)
18:12:16 Join utanapischti [0] (~username@
18:12:38 Quit sasquatch (Ping timeout: 264 seconds)
18:13:01 Quit Highlander (Client Quit)
18:14:34 Join n1s [0] (
18:14:35 Quit n1s (Changing host)
18:14:35 Join n1s [0] (~n1s@rockbox/developer/n1s)
18:15:20 Join Buschel [0] (
18:16:26 Quit robin0800 (Quit: Leaving)
18:17:34 Join Buschel_ [0] (
18:19:05 Quit liar (Ping timeout: 258 seconds)
18:20:09 Quit Buschel (Ping timeout: 248 seconds)
18:20:17 Nick Buschel_ is now known as Buschel (
18:21:20 Join liar [0] (
18:23:52 Quit stoffel (Ping timeout: 240 seconds)
18:40:10 Quit liar (Ping timeout: 258 seconds)
18:42:53 Quit CaptainKewl (Ping timeout: 250 seconds)
18:49:35 Join liar [0] (
18:50:25 Quit JackWinter (Remote host closed the connection)
18:53:06 Join JackWinter [0] (~jack@
18:56:44 Quit GeekShadow (Read error: Connection reset by peer)
18:56:58pamauryI'm a bit puzzled, what protocol would you use to transfer a framebuffer in 24BPP format on a 18-bit wide databus :-/
18:57:30 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
18:58:26wodzpamaury: this is how lcd is wired on rk27xx.
18:58:58wodzpamaury: some data lines are omitted
18:59:04bertrikI guess you just throw away some bits
18:59:31pamauryon mine, when doing transfers using gpio, it uses a weird packing but yeah some bits are ignoreds
19:00:28wodz <- this takes 16bit value and converts to 18bit interface
19:01:13pamauryIn my case the lcdif block is responsible for the conversion it seems
19:01:31pamauryor it actually sends them in two times
19:01:37wodzpamaury: here too, but you can set it to bypass mode
19:02:08pamaurythe OF doesn't use bypass mode and since I know nothing about the lcd controller....
19:02:34wodzit is lcdif block property not lcd controller
19:03:02pamauryyes but I don't know which format expect the lcd, so I will let the magic happen :)
19:03:37pamaurythe lcdif seems to handle all kind of setups but the manual is quite vague
19:03:48wodzif you know how - go ahead
19:03:56 Join CaptainKewl [0] (
19:04:18wodzI can't figure out how to use lcdif in buffered mode on rk
19:04:46pamaurywhat do you mean by buffered ?
19:06:25TheSevenpamaury: do i understand correctly that the clip+ has the same OTG as the nano2g, and *not* the one described in the as3525 datasheet?
19:06:39pamaurynearly the same yes
19:06:40wodzrk have 2048x32bit sram which can be mapped to lcd adress space
19:06:50 Part u42p ("Leaving")
19:06:57pamauryah, my lcdif is a dma master so it handles everything
19:07:13wodzthis allows to do hw scalling and alpha blitting
19:08:11pamauryit also has a pixel pipeline which does various stuff but I'm not using it for now
19:09:58pamauryTheSeven: why such a question ?
19:10:22TheSeventrying to hunt down the code that handles USB mass storage in that bootrom
19:13:15bertriklovasoa, I'm not sure if I can help you with the playlist stuff, but if you have other code that fixes iap for your accessory I can try to help get it incorporated
19:35:21 Quit lovasoa (Ping timeout: 248 seconds)
19:36:11 Join lovasoa [0] (~lovasoa@2a01:e35:8a2e:8080:226:bbff:fe15:33b0)
19:38:59lovasoa@bertrik Yes, I have a fix already
19:40:29lovasoaI posted a patch here :
19:45:12***Saving seen data "./dancer.seen"
19:50:23 Quit advcomp2019 (Read error: Connection reset by peer)
19:50:49 Join advcomp2019 [0] (~advcomp20@unaffiliated/advcomp2019)
19:54:27[Saint]in 2009!?!
19:55:14*[Saint] was expecting something more, current.
19:58:47 Join kevku [0] (~kevku@2001:470:28:773:babe:feed:dead:bee)
20:10:38pamaurywow, lcd working \o/
20:10:58pamauryI don't understand why but I got the logic wrong somewhere between enable and disable
20:11:14pamauryand apparently I'm using the wrong RGB ordering
20:16:13bertrikI think I'll update the comments for the commands in iap.c now that we have more documentation
20:16:29bertrikand possibly look at supporting large frames too
20:20:44bertrikor maybe we should just make defines for the commands
20:26:05 Join sideral [0] (~sideral@
20:26:05 Quit sideral (Changing host)
20:26:05 Join sideral [0] (~sideral@rockbox/developer/sideral)
20:33:22pamauryI finally got it \o/
20:33:40CIA-87New commit by 03Buschel (r29774): Ensure proper termination of m4a lookup table.
20:34:19sideralcongrats pamaury, nice picture :)
20:34:32 Quit lovasoa (Ping timeout: 248 seconds)
20:36:31pamaurythanks :)
20:36:35CIA-87r29774 build result: All green
20:37:49 Join lovasoa [0] (~lovasoa@2a01:e35:8a2e:8080:226:bbff:fe15:33b0)
20:37:54sideralBuschel: I have a few minutes now to test FS #12077. I should use SVN current + latest patch, right?
20:38:12Buschelyes. very appreciated!
20:39:02 Join benedikt93 [0] (~benedikt9@unaffiliated/benedikt93)
20:47:22 Join saratoga [0] (9803c6dd@gateway/web/freenode/ip.
20:47:32 Quit saratoga (Changing host)
20:47:32 Join saratoga [0] (9803c6dd@rockbox/developer/saratoga)
20:54:35 Quit mystica555 (Ping timeout: 240 seconds)
20:56:26CIA-87New commit by 03Buschel (r29775): Use dedicated function to internally reset aac decoder synthesis after seek.
20:57:52 Join stoffel [0] (
20:59:07CIA-87r29775 build result: 0 errors, 48 warnings (Buschel committed)
21:03:59sideralBuschel: r29774 + faad_no_malloc_v15.patch works fine on my ClipV2
21:04:55Buschelsideral: good news :) what happens if you set "#define FAAD_STATIC_ALLOC" in libfaad/common.h ?
21:05:04Buscheldo those long files still play then?
21:05:18siderallet me check :)
21:05:25CIA-87New commit by 03bertrik (r29776): iap: order lingo 0x00 commands by id and add comments
21:06:36 Quit Stummi (Quit: Bye!)
21:08:02sideralBuschel: Yep, still appears to work fine
21:08:15CIA-87r29776 build result: 0 errors, 48 warnings (bertrik committed)
21:08:30sideralis there a specific failure mode you'd like me to test?
21:09:33bluebroth3ranyone around that did experience the voice file creation in Rockbox Utility not working?
21:09:36BuschelI was not sure if static allocation of AAC-HE buffers would affect the playability of long AAC-LC files −− when FAAD_STATIC_ALLOC is set. you just tested it
21:10:13BuschelI am pretty sure it will affect playability, maybe the memory size reserved for codecs is still large enough
21:11:19*TheSeven just discovered that the clip+ bootrom uses malloc extensively :/
21:12:10saratogahave you gotten anywhere looking at that?
21:12:51bertrikTheSeven, maybe it's written in C++
21:13:10CIA-87New commit by 03Buschel (r29777): Fix yellow.
21:13:12TheSevendoesn't seem like that
21:13:26sideralBuschel: Yep, that's what I tested. Thanks for looking after the M4A codec! :)
21:13:33TheSevensaratoga: not yet, still trying to spot the interesting code
21:15:09bertrikTheSeven, what I find interesting, is that some bits of GPIO C are used as a boot mode and some of these are also connected to buttons on the clip+
21:16:16TheSevenBuschel: you just reddened by release builds!
21:16:24CIA-87r29777 build result: All green
21:16:46 Join LambdaCalculus37 [0] (~rmenes@rockbox/staff/LambdaCalculus37)
21:17:02TheSevenbertrik: which level of boot mode? to select ROM vs. external storage as initial bootloader, or where the ROM will load its code from?
21:17:06TheSevenand which buttons are those?
21:17:12TheSeven(and which gpios)
21:17:39 Part sideral
21:17:45 Join sideral [0] (~sideral@rockbox/developer/sideral)
21:18:36BuschelTheSeven: ? talking of the voltage issue ?
21:19:02TheSevenwhich voltage issue?
21:19:32BuschelTheSeven: what do you want to tell me with your last comment?
21:19:47TheSevenoops, s/by/my/
21:20:14Buschelahh, ok :) sorry for that
21:20:42TheSeveni'm still a bit confused as to why i got reds while the build grid apparently only got yellows
21:21:14 Quit LambdaCalculus37 (Ping timeout: 260 seconds)
21:21:38bertrikTheSeven, according to the as3525 datasheet, GPIO C0-C3 influence the boot mode, C1 = menu, C2 = play, C3 = left (I think I also saw processing of these bits in the boot rom somewhere)
21:22:18 Quit liar (Ping timeout: 258 seconds)
21:22:27 Join LambdaCalculus37 [0] (~rmenes@rockbox/staff/LambdaCalculus37)
21:22:42bertrikmaybe that was just dead code, I don't know
21:23:03pamauryTheSeven: on which bootrom are you working ?
21:23:21TheSevenclip+, so as3525v2
21:23:35pamaurythat's the bootrom, not the OF ?
21:24:31pamaurythen perhaps you'll have an opportunity to solve our usb issue because the clip+ OF usb code is too complicated
21:24:56TheSevenBuschel: I'm still getting reds for iPod Classic, with SVN HEAD and make veryclean
21:25:26TheSevenpamaury: I doubt this one is much better
21:26:19TheSevenalready identified 8 distinct structures/objects it uses while looking at the first couple of functions, and I have absolutely no clue what they are doing
21:26:38TheSeven(found some entry points into the USB code by hunting down the IRQ handler)
21:27:39 Join liar [0] (
21:27:56pamaurydoes it look like c++ ?
21:28:55BuschelTheSeven: this looks very strange. 1) it seems you are still using an old m4a.h 2) I do not understand why "idx" is not declared right at the top of the function.
21:28:55TheSevenfrom the part i've looked at so far it's hard to tell
21:32:04 Quit CaptainKewl (Ping timeout: 240 seconds)
21:33:08 Join Keripo [0] (
21:36:00 Join Keripo1 [0] (
21:37:22 Quit Keripo (Ping timeout: 246 seconds)
21:45:15***Saving seen data "./dancer.seen"
21:47:46 Quit LambdaCalculus37 (Quit: BOUNCE)
21:50:17clmoeyeeee is back
21:51:19clmoeso, do I need to unsolder the battery in a clipp for recovery mode or can I just hold the power button?
21:54:40 Quit stoffel (Remote host closed the connection)
21:55:04TheSevenclmoe: unsoldering the battery should only be neccessary if you can't manage to power it down by other means
21:55:38clmoewell it's powering down by itself so
21:55:40clmoeshould be good
21:58:52clmoeright now I've got no clipp screen response
21:59:06clmoebut when I plug it in I have access to 3.68gb of the flash
21:59:11clmoewhat should be different in recovery mode?
21:59:36TheSeventhat sounds good
21:59:45TheSevennow dd the firmware file to it
22:00:09saratogadon't you have to DD an actual firmware partition image to it?
22:00:14wodzhmm, on rockchip kernel_init() hangs :/
22:00:49clmoenot sure, I've got the OF and I've got another clip that works
22:01:18clmoeI tried just copying everything from the working clipp to it
22:01:21clmoedidn't do anything
22:01:28TheSevenno, that can't work
22:01:34bertrikI wonder if it's really in recovery mode now, perhaps just showing a working FAT filesystem
22:01:43TheSeventhe size shouldn't match either, as the firmware space is exposed on the broken one, but not the working one
22:02:11bertrikIf you can copy stuff, that means there's a filesystem present. The recovery mode just shows a raw block device, IIRC
22:02:53clmoeI think the working is a 2gb and broken is 4
22:03:06bertrikthat's odd
22:03:18clmoeI had to format to FAT when it started showing me the flash. At one point I was getting the sansa logo but it would hang there
22:03:39bertrikyou *had* to format it?
22:03:47clmoeit was showing as raw I think
22:04:17bertrikPerhaps we can tell what mode it is in by looking at the USB vendor/product ids
22:04:19clmoeso I just decided to format it to fat32
22:04:46clmoeI'll check that in a sec, gotta wait for the working clip to finish REFRESHING MY MEDIA
22:04:56clmoeany way to disable that? It makes me MAF
22:05:26clmoe>working clip is 3.65gb
22:05:30clmoeso it's already in recovery mode then
22:06:04clmoeso I just need an image then
22:07:13bertrikit could be that your sugar-filled clip+ is a 2GB model and your working clip+ is a 4 GB. I think you can't determine recovery mode from that.
22:07:26clmoethey are both 4gb
22:07:38clmoehow could sugar ray be 2gb if it shows 3.68gb?
22:08:02wodzI don't get it - init_threads() work, tick_start() work and the whole kernel_init() hangs
22:08:27bertrikI don't know. I just think this is a very poor way to determine whether you're in recovery mode or not.
22:08:42 Join petur [0] (~petur@rockbox/developer/petur)
22:08:51gevaertsbertrik: the idea here is that one is *slightly* bigger than the other
22:09:07clmoe30mb or so
22:09:14gevaertsi.e. 3.65gb vs. 3.88gb
22:09:34gevaertsand *that* fits the recovery mode idea
22:10:09bertrikah, ok, so go ahead if you feel comfortable doing that ......
22:10:56clmoeagain, I bought this broken and have two other clipps. I would be fine attempting to replace circutry on the board even with my terrible sodlering skills
22:11:08clmoeso how do I write the firmware to it in windoze?
22:11:50bertrikI think there are dd-clones for windows, can't name any though
22:11:53gevaertsThere are windows versions of dd around
22:12:17bertrikthis one perhaps:
22:14:08clmoeclippa.bin right?
22:14:12bluebroth3ror use a linux live CD ;-)
22:14:17bertrikclmoe, yes
22:14:31clmoehow do I do that in windd?
22:14:46clmoebluebroth3r, then I'd have to reset my uptime counter
22:14:51bluebroth3rhmm, broken. scorche, rasher, who else is in charge?
22:15:01clmoekeeping uptime in windows is like lunatic mode, but harder
22:15:22clmoethink I had almost 2 months one time on this win7 64 before I had to install something
22:15:31bluebroth3rwho cares about uptime counters? ;-)
22:15:39clmoelinux users and me
22:15:54saratogawell wait until you're going to reboot anyway
22:16:06clmoethat might be a while
22:16:18bluebroth3rhmm, opening a bug report because the theme site has a problem sounds like a somewhat ... interesting idea to me
22:17:10clmoeoh, so I'm not using the gui then
22:18:05bluebroth3rI don't care about using a GUI or not. That's at least a dd version for Windows I know about.
22:18:21clmoeoh, bertrik linked winDD and I was trying to figure that out
22:18:37bluebroth3rmight even have used that once or so. Though it's more likely I've booted into Linux those few times I actually needed something like that
22:19:10CIA-87New commit by 03Buschel (r29778): Rework of libfaad in several areas. Allow removal of malloc with a new define FAAD_STATIC_ALLOC (in common.h). For now malloc is not fully removed but ...
22:20:31 Quit leavittx (Ping timeout: 276 seconds)
22:22:01lovasoaWhich function can I call to open a specific file with the default application ?
22:22:05CIA-87r29778 build result: All green
22:22:25saratogafrom a plugin?
22:22:32clmoe>trying that other dd program
22:22:40CIA-87New commit by 03bertrik (r29779): iap: add more comments to iap command and responses
22:22:59clmoe>typed in if=h:\clippa.bin of=\\.\h:
22:23:04clmoeand now it's just sitting there
22:24:13*clmoe doesn't know how to use these silly linix based programs
22:25:05saratogaits probably in the manual for the command
22:25:20saratogait probably works somewhat differently then on linux since windows mount points work differently
22:25:57CIA-87r29779 build result: All green
22:26:08bertrikyou possibly just killed the contents of your h: drive
22:26:44BuschelTheSeven: could figure out your build problem?
22:26:57TheSevenit was actually svn acting up
22:27:02clmoeit's empty anyway
22:27:08TheSeveni suddenly had hundreds of local changes that i had never seen before :)
22:27:23TheSevenit seems to just have skipped a couple of things while updating, for whatever reason
22:27:36Buschelresync did it?
22:27:48TheSevensvnversion / svn info / svn up all thought that the working copy would be up to date
22:28:18TheSevenfixing that was a bit funny as I had local changes that I wanted to keep :)
22:29:33*Buschel can imagine
22:30:22lovasoaso, You don't know ? It should be possible, as the file browser does exactly this
22:31:38wodzwhat the hell it hangs on memset
22:31:47wodzany ideas?
22:31:58wodzcaches disabled
22:32:12TheSevenoverwriting something important?
22:32:29TheSevenhow are you locating the hang? hang/reset feedback?
22:32:40*bluebroth3r wonders why command line programs should be silly
22:32:42TheSevenor do you already have a debugging console?
22:33:11wodzI turn on backlight
22:33:43TheSevenare the memset arguments constant?
22:34:10wodzmemset(tick_funcs, 0, sizeof(tick_funcs));
22:34:53wodzin kernel.c
22:35:24wodzbut I believe this is rock solid piece of code
22:37:49wodzunsigned int test[10]; memset(test, 0, sizeof(test)); hangs as well
22:38:45clmoe>only copies entire disks
22:38:53clmoe>that other dd program
22:38:55bertrikIsn't tick_funcs run in interrupt context? so clearing it with interrupts enabled perhaps causes some race and a null pointer access
22:38:58clmoe>doesn't seem to want to do anything
22:39:24wodzbertrik: not the case - irqs are disabled
22:39:31clmoeis there any way to get the OF on there without dd?
22:40:01wodzbertrik: anyway simple memset on array hangs
22:40:17saratogaclmoe: theres probably other programs that can write in raw mode to the disk, check google
22:40:41wodzwhat is ldrpc cond=eq ?
22:41:31wodzit's in memset-arm.S
22:41:35saratogafunman's macro
22:41:40saratogaits for thumb mode IIRC
22:42:11saratogain 32 bit mode it just evaluates to popping the registers off the stack with one of the ldm commands
22:42:28saratogaand updating the PC
22:43:57clmoeparagon will let me write to the partition in raw
22:44:09clmoehow do I get the bin file in that way though?
22:44:28*bluebroth3r suggest rebooting. Or use a VM
22:45:35clmoeI'll try a VM, but I'll try writing the bin from sector zero first, for gigs
22:46:09bluebroth3rhmm, how about making beastpatcher checking for nk.bin in its folder and offering an option to install dual-boot if present?
22:46:18bluebroth3rwould make installation instructions simpler :P
22:48:24 Quit [Saint] (Ping timeout: 248 seconds)
22:48:50 Join [Saint] [0] (~st.lasciv@
22:52:07 Quit benedikt93 (Quit: HIP-HOP sounds best when you listen to METAL instead.)
22:53:53CIA-87New commit by 03bluebrother (r29780): Add a warning about WMP11 being necessary on Windows for beastpatcher.
22:56:46CIA-87r29780 build result: All green
22:57:10TheSeveni just found what i think is a shitload of c++ vtables in that bootrom
23:03:31 Join ChickeNES [0] (~ChickeNES@
23:06:46 Quit ChickeNE_ (Ping timeout: 250 seconds)
23:09:36 Quit kevku (Ping timeout: 260 seconds)
23:25:41wodz\o/ show_logo() succeeded
23:26:28wodzcolors are a bit wrong but generally I have 400x240 debuging output :-)
23:33:09TheSevenbertrik, pamaury: lots of weird stuff that smells like c++, but I can't really identify what it is...
23:34:00pamauryTheSeven: then I think it's C++, the USB code I found in the clip+ OF was C++ so perhaps it's the same
23:35:02TheSevensome pointer (this?) being passed around as the first argument to each function, and each function writes a relative address of itself to the that pointer
23:35:20TheSeventhe fun part is that this address is loaded from some memory location, and is actually relative to that location
23:35:51TheSevenso copying it there makes it kinda worthless as the base for the relative address isn't known
23:37:05TheSevenand then there's several hundreds of those relative addresses in a row at some memory address, with some veneers in between (purpose still unclear)
23:37:38TheSevenmost of them pointing to function heads, some to instructions in the middle of functions, and a few even into the middle of a 4-byte thumb instruction
23:37:50TheSevenever seen that kind of thing before?
23:38:15TorneTheSeven: that sounds *vaguely* like pointer-to-member-function logic
23:38:45Tornepointer-to-member for virtual methods works more or less like a vtable offset
23:39:24TheSevenit's not a pointer relative to the vtable's head, but to the very entry it is being loaded from
23:40:10TheSeveni've seen those things being dereferenced in a few places as well, but most of them are copied into this "this" object or whatever it is
23:40:34pamauryI think I saw those things
23:40:38pamauryin the clip+ OF !
23:41:04 Join silbo [0] (~quassel@
23:41:05 Join DerPapst [0] (
23:41:08TheSevenhm, or maybe it's lots of very small vtables, so that it looks like the entrys are relative to the entries, and the weird ones are actually members of bigger vtables
23:42:41 Quit sideral (Quit: Leaving.)
23:42:47 Quit Leif (Quit: Leaving)
23:43:02TheSevenpamaury: ever figured out what those things are actually good for?
23:43:13pamauryI don't remember
23:43:21pamauryI could have a look if you *really* want
23:43:45pamauryanyway I gave up because the code was too complicated
23:44:03TheSevenhave you ever identified some USB code in there?
23:44:12TheSevenmaybe I can find it's bootrom counterpart that way
23:44:37TheSeveni really need to get through to that mass storage code
23:44:57TheSevenjust throwing random binaries at it didn't work out :P
23:45:11pamauryI identified the entry point (the place where everything seems to start) iirc but I'm not sure I ever managed to say "this is writing to a usb register" because the C++ class is given the usb base as parameter and then everything is indirect
23:45:16***Saving seen data "./dancer.seen"
23:51:59 Quit Buschel (Quit: ChatZilla [Firefox 3.6.16/20110319135224])
23:52:55 Join krazykit [0] (
23:54:52 Quit ender` (Quit: facebook is like a jail: you sit around, waste time, have a profile picture, write on walls and get poked by guys you don't really know.)

Previous day | Next day