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 2012-03-16

00:00:22 Part kadoban_ ("Konversation terminated!")
00:00:37 Quit dfkt (Quit: -= SysReset 2.55=- Sic gorgiamus allos subjectatos nunc.)
00:08:31 Quit kadoban (Disconnected by services)
00:08:55 Join kadoban [0] (
00:09:05 Quit kadoban (Disconnected by services)
00:09:32 Join kadoban_ [0] (
00:13:24 Quit lebellium (Quit: ChatZilla [Firefox 11.0/20120310173008])
00:34:42 Join [Saint] [0] (~Saint]@unaffiliated/saint/x-8516940)
00:39:14 Join TexasRockbox [0] (46740228@gateway/web/freenode/ip.
00:39:36***Saving seen data "./dancer.seen"
00:41:22 Quit TexasRockbox (Client Quit)
00:44:44 Join [Saint_] [0] (~Saint]@
00:44:44 Quit [Saint_] (Changing host)
00:44:44 Join [Saint_] [0] (~Saint]@unaffiliated/saint/x-8516940)
00:45:01 Quit [Saint] (Remote host closed the connection)
00:45:17 Nick [Saint_] is now known as [Saint] (~Saint]@unaffiliated/saint/x-8516940)
01:05:01 Quit bertrik (Ping timeout: 265 seconds)
01:06:38 Quit ender` (Quit: #include <unistd.h> | #define Bork for | #define bork fork() | int main() { Bork(bork ;; bork) bork; })
01:09:23 Join passstab [0] (
01:15:29 Quit bitcraft_ (Remote host closed the connection)
01:15:43 Join enthdegree [0] (~BitchX@unaffiliated/enthdegree)
01:16:38 Quit Mir (Ping timeout: 276 seconds)
01:18:37 Join Mir [0] (
01:31:19ScrompleOn the iRiver 3xx.. I you have the hold switch turned on on the side of the device and attempt to turn on, Rockbox displays a warning message. It's only there for about a half second. Is there a setting to lengthen that amount of time??
01:32:40ScrompleDid some checking and didn't see one, but as I suffer selective blindness thought I check in case I missed it.
01:40:04ScrompleOkedoke... Thanks.
01:43:07[Saint]what JdGordon means is that "no, not without editing the sources manually and recompiling", rather.
01:43:12[Saint]its certainly posible.
01:45:51 Join Keripo [0] (
01:54:18 Join Strife89 [0] (~Strife89@
02:12:50 Quit Strife89 (Quit: Heading home)
02:14:59 Join [Saint_] [0] (~Saint]@unaffiliated/saint/x-8516940)
02:15:25 Quit [Saint_] (Client Quit)
02:17:47 Join mathieui [0] (
02:20:59mathieui*PANIC* ata: -2
02:21:58mathieuiDo the ATA errors mean that the drive is basically dead/completely corrupted?
02:39:38***Saving seen data "./dancer.seen"
02:55:25 Quit Scromple (Read error: Connection reset by peer)
02:56:35[Saint]mathieui: which device is this?
02:56:52mathieuisansa clip+
02:57:30mathieui(now it refuses even to boot or connect to the computer, even after the "reset" thing)
02:58:18[Saint]a brief look at the code sugegsts that its a failuer to reset the storage, or identify it.
02:58:56 Join Scromple [0] (~Simon@
02:59:22[Saint]which basically means that the bootloader can't find the internal storage, as far as I can see.
02:59:35mathieuiAlso, I was getting random freezes before that (everything just stops then I have to reset it several times to have it work again)
02:59:38[Saint]Or, it identified incorrectly, and it backed out.
02:59:51[Saint]Can you get to the original firmware?
03:00:08[Saint](hold left during boot)
03:01:10mathieuiseems to be blocked on the sansa logo
03:02:29[Saint]It seems rather likely that your device is toast, then. If you can't read from/write to the disk, there's not a lot that can be done here.
03:02:55mathieuiI can sometimes
03:04:06[Saint]the best bet would be to see if you can get it to boot (and then mount), and check for filesystem errors, or get it to boot into the OF and use the OF's format option.
03:05:20[Saint]If the OF itself won't boot though, it seems rather suggestive that the area of the disk not exposed as mass storage is corrupt.
03:06:46[Saint]If you can get it to mount, an OF "upgrade" (you don't actually have to upgrade, just use the same version firmware) should re-write this and maybe you'll get lucky.
03:07:54Scromple[Saint]: It's not that big of a problem. Turning the device on a couple of times while the penny drops as to why it won't boot does not do any damage as far as I can see.
03:10:36mathieui[Saint]: also worth noting, before the *PANIC* thingie, I had a nice after-ata-error screen, with something mentionning GPIO and voltages
03:12:40[Saint]Off the top of my head I can't even think what that would be.
03:21:41 Quit thegeek_ (Read error: Connection reset by peer)
03:31:34mathieuiWell, I cannot get it to boot right now, will try later. (this device has been dead for me for long, anyway)
03:35:36 Quit MethoS- (Quit: Konversation terminated!)
03:39:38 Quit wtachi (Ping timeout: 276 seconds)
03:40:56 Join [Saint_] [0] (~Saint]@unaffiliated/saint/x-8516940)
03:41:37 Quit passstab (Remote host closed the connection)
03:48:46 Join [Saint__] [0] (~Saint]@unaffiliated/saint/x-8516940)
03:49:11 Quit factor (Ping timeout: 244 seconds)
03:52:44 Quit [Saint_] (Ping timeout: 265 seconds)
03:54:07 Join factor [0] (
03:58:20 Join [Saint_] [0] (~Saint]@unaffiliated/saint/x-8516940)
04:02:25 Quit [Saint__] (Ping timeout: 265 seconds)
04:05:20 Quit anewuser_ (Read error: Connection reset by peer)
04:10:41 Quit amiconn (Disconnected by services)
04:10:41 Join amiconn_ [0] (amiconn@rockbox/developer/amiconn)
04:11:03 Nick amiconn_ is now known as amiconn (amiconn@rockbox/developer/amiconn)
04:11:40 Quit pixelma (Disconnected by services)
04:11:43 Join pixelma_ [0] (pixelma@rockbox/staff/pixelma)
04:11:45 Nick pixelma_ is now known as pixelma (pixelma@rockbox/staff/pixelma)
04:27:18 Quit TheSeven (Disconnected by services)
04:27:52 Join [7] [0] (~TheSeven@rockbox/developer/TheSeven)
04:32:36 Quit enthdegree (Ping timeout: 252 seconds)
04:35:24 Join Osix1 [0] (
04:36:15 Part Osix1
04:38:16 Quit Zarggg (Quit: Rebooting client...)
04:38:28 Join Osix1 [0] (
04:38:42 Part Osix1
04:38:51 Join Osix1 [0] (
04:39:14 Part Osix1
04:39:39***Saving seen data "./dancer.seen"
04:39:56 Join enthdegree [0] (~BitchX@unaffiliated/enthdegree)
04:43:22 Quit XavierGr (Ping timeout: 245 seconds)
04:50:51 Join Rower85 [0] (
04:57:32 Quit curtism (Quit: Live Long and Prosper)
05:03:30 Join Rob2222 [0] (
05:07:12 Quit Rob2223 (Ping timeout: 246 seconds)
05:08:19 Join Osix1 [0] (
05:09:53Osix1I want to use Pauls new bootloader for the fuze+ that charges. Can I update through RockboxUtility? Or do I need the version 9 metaphys built?
05:14:05[Saint]Yes, you may.
05:14:34[Saint]Download the latest version of RBUtil, and in the configure screen, select "Show Disabled Targets".
05:16:29Osix1OK I was just wonder if that would get that build with the charging fix. I have used the Utility alot so its all setup
05:17:39[Saint]Be mindful though the it only charges in the bootloader, and for this case to happen you basically need to trash your Rockbox installation.
05:18:01[Saint]Bootloader USB is used when the device cannot boot the Rockbox image for <whatever_reason>.
05:18:14[Saint]This change isn't related to "normal" charging.
05:20:05 Quit remlap (Ping timeout: 248 seconds)
05:21:42Osix1OK what has been on my mind as of late is that when you just plug in a fuze+ it does not startup full Rockbox. Just a USB bootloader. Ive been concerned that if you let it get fully drained, you basically would be bricking the fuze+ because you cant charge it again. Is this a valid concern or not?
05:22:25 Join remlap [0] (
05:25:36[Saint]It is a possible scenario, yes.
05:28:48Osix1I think that is why Paul added charging code to the boot loader and metaphys created the V9 Under alternative methods. I want to use that boot loader but it would have been easier to use the Utility other than shoving that custom bootloder in. I was just wondering if the Utility would use a version with Paul's new code
05:38:16 Quit [Saint_] (Read error: Connection reset by peer)
05:40:14 Join wtachi [0] (
05:58:35 Join Transformer [0] (
06:00:08 Quit Transformer (Excess Flood)
06:06:32[Saint]Osix1: If you select a current build, then, yes. It will.
06:06:51[Saint](where "will" == "_should_" ;) )
06:07:43 Quit Youbi (Ping timeout: 252 seconds)
06:11:18Osix1Ah OK ^^
06:16:21 Quit [Saint] (Quit: brb)
06:19:40 Quit Keripo (Quit: Leaving.)
06:20:00 Quit perrikwp_ (Read error: Connection reset by peer)
06:21:27 Join perrikwp [0] (
06:23:39 Quit Osix1 (Quit: Leaving.)
06:27:05 Join [Saint] [0] (~Saint]@
06:27:06 Quit [Saint] (Changing host)
06:27:06 Join [Saint] [0] (~Saint]@unaffiliated/saint/x-8516940)
06:38:15 Quit enthdegree (Ping timeout: 250 seconds)
06:39:43***Saving seen data "./dancer.seen"
06:48:30 Quit XMen ()
07:39:15 Quit mshathlonxp (Ping timeout: 252 seconds)
07:49:44 Quit remlap (Quit: Leaving.)
07:50:46 Join stoffel [0] (
08:00:26 Join Youbi [0] (
08:17:05 Join Zagor [242] (~bjst@rockbox/developer/Zagor)
08:26:06 Join XavierGr [0] (~xavier@rockbox/staff/XavierGr)
08:28:53 Quit Galois (Remote host closed the connection)
08:29:07 Join Galois [0] (
08:30:48 Quit XavierGr (Client Quit)
08:32:10 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
08:38:28 Join nick_p [0] (
08:39:47***Saving seen data "./dancer.seen"
08:45:57 Quit Scromple (Quit: Leaving)
08:46:01 Join ender` [0] (
08:46:19amiconnZagor, Bagder: The order of commits 'Code changes' (whether normal, extended, "last 4 weeks" or "since last release") is rather messed up...
08:46:43Zagoramiconn: wow, indeed
08:47:06[Saint]i noticed this too.
08:47:17amiconnThe build table has the correct order
08:49:07 Quit pamaury (Ping timeout: 240 seconds)
08:49:07Zagorit's not the front page as much as it is "git log". try in your own repo.
08:49:33Zagoraah, it's timezone confusion
08:49:52Zagorno, not even that
08:57:35Zagorthe order on the front and build pages is the same. only the displayed timestamps differ.
08:58:05Zagorthe build page shows the time of building. the front page shows the time of patch *submission* (not commit)
08:58:13Zagoras git log does
08:59:13Zagorso it is correct, but slightly confusing
08:59:26 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
09:09:07 Join einhirn [0] (
09:16:24amiconnSounds not very useful
09:17:06*amiconn would very much prefer to see the timestamps of when something got pushed to the central repo
09:18:58Zagoryes. there doesn't appear to be a simple option to show commit date instead of author date. looks like you have to make a fully custom message with % codes
09:23:16 Quit factor (Read error: Operation timed out)
09:23:22 Quit kugel (Ping timeout: 244 seconds)
09:26:35 Join factor [0] (
09:40:37 Quit bertrik (Ping timeout: 240 seconds)
09:47:40*[Saint] wonders if he's perhaps set a recorc for most operating systems on an Archos A101IT Gen8:
09:48:28[Saint]Froyo, Gingerbread, Angstrom, Debian, Ubuntu and Windows for Workgroups 3.11
09:49:20 Join evilnick [0] (~evilnick@rockbox/staff/evilnick)
09:49:28[Saint]Don't ask why...because I really don't have a valid explanation. I don't even have a reasonably viable excuse.
09:50:01[Saint]But, six operating systems on one device seemed like a good idea at the time.
09:50:17[Saint](and it was piss-easy to setup)
09:54:27[Saint]Angstrom for the Archos A101 is a remarkable effort in how to take a reasonably nice system and make a half assed complete piece of shit out of it.
09:59:07 Quit ThomasAH (Ping timeout: 265 seconds)
09:59:43 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
10:16:07 Join wodz [0] (
10:16:26*[Saint] skulks away...
10:17:28pamauryI think it could be useful to implement a recovery procedure in RbUtil for the fuze+ (and the imx233 in general). Some users may need it and the procedure it not trivial for end users
10:17:56pamaurybluebrother: does that sound sensible ? Does RbUtil have such a thing for other targets ?
10:21:33wodzpamaury: what about your idea of unified reg mangling plugin?
10:22:19 Quit adnap (Ping timeout: 265 seconds)
10:22:22pamaurywodz: I didn't have time to implement it, I just built the map for the imx233 but recently I've had only little free time
10:23:38pamauryotoh, I've implement a symbol table for backtrace, to have a more meaningful output
10:24:58 Join TomColler [0] (
10:25:33wodzpamaury: you mean the patch which is on FS?
10:26:10pamauryimproved version, and synced with head
10:26:40 Join ThomasAH [0] (
10:26:47wodzcould you push it to gerrit?
10:28:51pamauryI'll do, I just need to tweak something and possible have a way to opt out this
10:30:01wodzyeah some nice define which can be enabled in configure will be nice
10:30:40pamaurythis requires an external program and makefile changes so unfortunately it's not as simple as a define
10:34:58 Join adnap [0] (
10:38:18*gevaerts still thinks (also) supporting the inline symbol names gcc can produce would be good too
10:38:53pamaurygevaerts: what are "inline symbol names" ?
10:39:16gevaertsI'd say "man gcc tells all", but I can't remember the name of the feature!
10:39:18gevaertsLooking now
10:39:51***Saving seen data "./dancer.seen"
10:47:07pamaurygevaerts: perhaps you mean the debug symbols ?
10:47:23gevaertsNo, IIRC it was something else
10:47:58gevaertsUnless my memory is *really* broken, there's a way to have gcc function names just before every function surrounded by some magic numbers
10:48:21gevaertsSo if you have an address that's supposed to be inside a function, you just go back until you encounter those magic numbers
10:48:34pamauryah yes
10:49:13pamaurybut there is a fundamental problem with this approach: our backtrace code finds the exit points of the function, not the entry one
10:50:09gevaertsWhy is that a problem? If those magic numbers are indeed there and not just in my imagination, can't you just search back for them?
10:50:09pamauryI don't remember the name of this feature either
10:51:09pamaurythat leaves a number of problem: when do you stop to search ? what if this function does not have a magic value ? (like handwritten assembly ones)
10:51:49gevaertsWhen you looped throught the entire address space :)
10:52:00gevaertsYes, that obviously needs a bit of thinking
10:52:14pamaurysure :) I admit I haven't investigated it to much
10:52:46gevaertsThe first step would be to find the feature again so we can actually read what it does :)
10:55:53pamaurythe problem is that the "magic" value is not quite magic
10:55:56gevaertsAh, yes
11:01:58gevaertsUnless there's a way to add yet another marker there somehow
11:11:28wodzthe magic is probably valid for instructions but I guess it can be easily hit in data.
11:16:04 Quit adnap (Ping timeout: 252 seconds)
11:16:56pamauryor we could search for the pattern and do some checks: the pattern gives the string size so we know where it begins and can check whether it's a string or not
11:19:04wodzthats an option
11:24:48wodzall in all it seems simpler than crafting symbol table and interatively relink
11:29:19 Join fragilematter1 [0] (
11:29:19 Part fragilematter1
11:34:12pamaurythe manual is unclear about thumb functions, I guess it aligned
11:34:22pamauryon a word boundary
11:34:40pamauryI'll try to implement this
11:58:34 Quit [Saint] (Quit: Quit)
12:05:36 Quit tchan (Read error: Connection reset by peer)
12:05:56 Join adnap [0] (
12:06:10 Join Guinness` [0] (
12:06:21 Join tchan [0] (~tchan@lunar-linux/developer/tchan)
12:06:36 Join [Saint] [0] (~Saint]@unaffiliated/saint/x-8516940)
12:07:02 Quit Guinness (Read error: Connection reset by peer)
12:15:23 Quit [Saint] (Ping timeout: 255 seconds)
12:17:58 Join [Saint] [0] (~Saint]@unaffiliated/saint/x-8516940)
12:29:30 Join z180 [0] (
12:31:36 Join anewuser [0] (~anewuser@
12:31:37 Quit anewuser (Changing host)
12:31:37 Join anewuser [0] (~anewuser@unaffiliated/anewuser)
12:35:21 Join [Saint_] [0] (~Saint]@unaffiliated/saint/x-8516940)
12:37:27 Quit [Saint] (Ping timeout: 255 seconds)
12:39:53***Saving seen data "./dancer.seen"
12:42:27pamaurydo we have a function to poke at some address and check if it readable ?
12:42:47pamaury(on arm)
12:44:42Tornei was going to implement one for the benefit fo the unwinder
12:44:48Torne(by trapping the abort)
12:44:52Tornebut haven't gotten around to it yet
12:47:42pamaurythat would be nice, since I would need one to implement symbol retrieval with -mpoke-function-name
12:48:01Torneit's pretty easy, if you want to do it :)
12:48:22Tornewrite asm function that takes an address and an out-parameter and returns bool
12:48:43pamauryI don't know much about arm, so you will be faster than me
12:48:56Tornewell i may not have time to do it any time soon :)
12:49:00Torneso that wont' be faster
12:49:08Tornebug me at the weekend
12:49:21pamauryok, I'll try then. How do I do that ?
12:49:28z180wasnt it sothat rockbo doesnt support MMUs?
12:49:35Tornez180: you dont' need an MMU for this
12:49:42Tornealso, we do support MMUs, we just don't require one
12:50:03Tornepamaury: basically what you need to do is have an asm function so you can set a label on the ldr instruction that might abort
12:50:10Tornethen, in the abort handler, check if the return address was that instruction
12:50:19Torneif so, set some global "you aborted" flag to true
12:50:24Tornethen return to the instruction after the load
12:50:43Tornethen the function can just check taht global, and return true/false for whether the load was valid (as well as, if it's valid, setting the value in an out-parameter)
12:50:52pamauryoh right, dead simple
12:50:53Torneyou don't need an mmu for this since you only check r14_abt
12:50:59Tornenot any MMU control registers.
12:51:56Tornethe only fiddly part is getting the offsets right for the return address. i forget what r14 is offset by on entry to abt
12:52:00 Quit nick_p (Quit: Leaving)
12:52:20Torneand if you return from abort "normally" it will return to the *same* instruction, so you need to increment it
12:53:11Tornei'm not sure if you wer ebeing sarcastic there or not :)
12:53:43pamauryno no, I was thinking about using the mmu but this approach is much simpler
12:53:48TorneAh, right
12:54:03TorneYeah, this has the advantage that it will work on hardware without an MMU as long as it generates aborts for totally invalid addresses
12:54:16Tornewith no MMU you will get some false-negatives where the data returned is non-aborting but garbage
12:54:20Tornebut that's generally ok
12:54:26Torneyou just don't want to crash
12:55:09Tornethis is another trick from the symbian kernel :)
12:56:50Tornethe ldrt/strt instructions to access user-provided addresses were trapped this way using "magic" addresses, in order to convert the abort into a regular error return
12:57:50z180but symbian is bloatware ,opposed to EPOC
12:57:53 Join devilXnuX [0] (~CZzz@
12:57:56Torneer, what?
12:58:58*Torne afks, lunch
13:00:03 Join devilXnuX_ [0] (
13:00:58 Quit devilXnuX (Read error: Connection reset by peer)
13:00:59 Nick devilXnuX_ is now known as devilXnuX (
13:07:54devilXnuXHello. I want to ask you all about a bug in chip8 emulator. Why the player hangs when I want to "play" chip8 program? Is there any solution to this problem? My rockbox version is 3.10 build 680c6fc-120222. Thank's in advance
13:09:26evilnickdevilXnuX: What player are you using?
13:09:48devilXnuXAh yes, forgot to mention, I use Sansa Clip+
13:10:09evilnickAnd does this happen on all the chip8 programs?
13:11:50devilXnuXYes, I "play" every single chip8 program that I found in rockbox Wiki Page. The same thing happen to all of program
13:12:15devilXnuXEven my player hangs on "TEST" program
13:15:05 Join devilXnuX_ [0] (~CZzz@
13:16:18devilXnuX_Sorry, my connection isn't stable. It disconnected just now
13:16:42 Join dfkt [0] (
13:16:43 Quit dfkt (Changing host)
13:16:43 Join dfkt [0] (dfkt@unaffiliated/dfkt)
13:17:19 Quit devilXnuX (Ping timeout: 252 seconds)
13:17:31 Nick devilXnuX_ is now known as devilXnuX (~CZzz@
13:19:07devilXnuXevilnick: So, how to overcome this problem? Thank you so much for your help.
13:19:50gevaertsDoes anyone really want to play chip8 games?
13:20:08evilnickdevilXnuX: Sorry, I'm not familiar with that player. I was just making sure there were no simple explanations
13:23:17 Quit ved (Ping timeout: 260 seconds)
13:23:17 Quit tuxx- (Ping timeout: 260 seconds)
13:23:41devilXnuXgevaerts: At least, the games can be added. So when we get bored to standard rockbox game, We can add some new games to it. At least.
13:24:00gevaertsdevilXnuX: have you ever seen chip8 games?
13:24:15gevaertsI'm not sure if I often get *that* bored
13:24:16devilXnuXevilnick: OK. Thank you very much for your assistance.
13:24:54devilXnuXYes, I've played some of it in PC version of chip8 emulator.
13:24:56 Join tuxx- [0] (
13:25:11devilXnuXgevaerts: Yes, I've played some of it in PC version of chip8 emulator.
13:26:36devilXnuXI found it match with the nature of Sansa Clip+. No color, low resolution. Unlike the rockboy things.
13:27:07gevaertsAnd no 4x4 keypad
13:27:57devilXnuXAs mentioned in rockbox Wiki, the keypad can be remapped. At least I can play some games in it.
13:28:28gevaertsWell, I guess that means someone is interested in maintaining this :)
13:30:32 Quit devilXnuX (Read error: Connection reset by peer)
13:31:45 Join devilXnuX [0] (
13:31:52devilXnuXIf only I knew C & C++ :(
13:33:00gevaertsNo C++ needed, so that's half the problem solved!
13:37:23devilXnuXHmm, but problem still not solved
13:46:34 Join ved [0] (~ved@2001:41d0:1:5914::2)
13:47:11Tornei had a fiddle with chip8 on ipodvideo aeons ago and somewhere between 80% and 100% of the files i tried crashed or did nothing
13:47:19Torneso it's likely that it's just busted and has probably been that way a long time
13:47:38Torneit's unlikely to get fixed unless someone who actually cares about playing chip8 games fixes it
13:47:55Torneand that's a very small set of people to begin with, since their limited scope makes them almsot guaranteed to suck
13:48:10Tornepretty sure the number of committers who care is zero
13:48:22Torneit's more one of those "ported because we could" things
13:48:31Torne(on the other hand, if you find a bug in frotz i will fix it for you) :)
13:48:49 Quit devilXnuX (Read error: Connection reset by peer)
13:50:40 Join MethoS- [0] (~clemens@
13:51:08 Join TheLemonMan [0] (
13:57:53 Join Keripo [0] (
14:01:30 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
14:07:30 Join mortalis [0] (~mortalis@
14:33:20 Quit Keripo (Quit: Leaving.)
14:39:56***Saving seen data "./dancer.seen"
14:47:12 Join remlap [0] (~Patrick@
14:51:31 Join bertrik [0] (
14:51:32 Quit bertrik (Changing host)
14:51:32 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
15:05:10 Join petur [0] (~petur@rockbox/developer/petur)
15:13:59 Quit z180 (Ping timeout: 260 seconds)
15:21:02 Quit wodz (Quit: Leaving)
15:24:20 Join enthdegree [0] (~BitchX@unaffiliated/enthdegree)
15:28:20 Quit dfkt (Quit: -= SysReset 2.55=- Sic gorgiamus allos subjectatos nunc.)
15:34:29 Join Zarggg [0] (
15:42:49 Quit randumbum (Quit: ChatZilla [Firefox 10.0.2/20120216101208])
15:43:23 Join passstab [0] (
15:45:04 Join jlbiasini [0] (
15:45:47 Join y4n [0] (~y4n@unaffiliated/y4ndexx)
15:58:15 Quit petur (Quit: here today, gone tomorrow)
16:06:09 Join bitcraft [0] (~bitcraft@
16:06:12 Quit bitcraft (Remote host closed the connection)
16:06:15 Join bitcraft [0] (~bitcraft@
16:13:44 Join randumbum [0] (
16:14:21 Join WalkGood [0] (~4@unaffiliated/walkgood)
16:21:00jlbiasiniI'm having a look to pdbox plugin. If I activate it on the fuze+, the uisimilator hang in loop giving an infinite list of "Thread creation failed. Retryingmake_context() : Cannot allocate memory. OTOH it seems to work on the device.
16:21:33jlbiasiniarf this seems to be the same on other target, simulator doesn't handle pdbox
16:27:14pamauryTorne: here is my try, seems to work but needs more testing
16:27:39pamaury(there are some unrelated stuff, the patch is not clean, only work the imx233 because of the local crt0.S changes)
16:28:13Tornei'd probably have done it in the actual abort handler in asm instead of UIE
16:28:54jlbiasinihum there seems to be mixing issue with pd box on the fuze+ it starts ok but after a few second the sound just remains the same untill ones exit the plugin (no hang)
16:29:07 Join mshathlonxp [0] (msh@
16:29:07Tornealso, you probably don't really want to do is_memory-readable
16:29:28Torneyou want to do read_memory_safely(uint32_t addr, uint32_t* value) and just actually set the value if it didn't fail
16:29:36Tornebecause the caller is going to have to do the read anyway
16:29:48Torneyou might as well do it for them, then you have atomically gotten the value :)
16:30:13Torne(in theory you could read it to check if it aborts, find it doesn't, tell the caller, then it aborts when they read it - this is possible with peripheral memory, though rare)
16:30:52Torneoh, hm, you are supporting a range, not just one location
16:31:08Torneis that actually needed?
16:31:15pamauryI don't know :)
16:31:18Torne..generall i would do basically all of this in asm
16:31:39Tornein fact probably literally all of it
16:31:44Tornei was imagining something that was maybe 20 instructions ;)
16:32:08pamaurybut I think I will drop them, just keep a read_memory_safely like you suggest
16:32:30Tornebool saferead8(uint8_t* addr, uint8_t* out)
16:32:32Tornerepeat for 16/32
16:32:54Tornethen you don't need a C wrapper to decide what size it is
16:33:22pamauryright, one question though: how do I get rid of the warning because the arguments of the naked functions are unused ?
16:33:28Tornedon't name them
16:33:31Torneor comment out the names
16:33:43pamauryah thanks, didn't think about it
16:33:50Tornebool saferead8(uint8_t* /*addr*/, uint8_t* /*out*/)
16:34:30pamauryactual the major changes is in data_abort_handler which must be replicated in all crt0.S
16:34:31Torneso, yeah. i think a nicer structure might be for the platform specific data abort vector to just branch (regular b, no bl) to a non-platform-specific asm function
16:34:40Tornewhich can do this logic and if it fails just b to UIE]
16:34:44Tornerather than doing it inside UIE
16:34:56Torneyou still need to modify every abort vector, but only changing one line
16:35:00Tornenot adding any more instructions.
16:35:02pamauryyes, that would be nicer
16:35:26pamauryI'll change that and submit my changes again, thanks for the feedback
16:35:29Torneif you do it in asm you also should only need to save two registers :)
16:35:31Tornepossibly one
16:36:20pamaurywhere should I put this ?
16:36:29Torneer, i dunno.
16:36:30pamaurycreate a file in target/arm/ ?
16:36:32Tornenew file?
16:36:36Torneor in lib/unwarminder
16:36:52Tornesince that needs to use it (there are function stubs for doing this in there but they are implemented as just straight loads right now)
16:37:45pamaurythat's not specific to the unwinder although it's not used anywhere else
16:38:00Tornei don't think there is any use for this on rockbox other than debugging
16:38:03Torneso it might as well live in there
16:38:20Tornethe only other use for this kind of thing is for, say, kernels to use on user-provideda ddresses, which isn't a thing we have
16:40:00***Saving seen data "./dancer.seen"
17:00:49 Part TomColler
17:02:20 Part Zagor
17:03:05 Quit randumbum (Remote host closed the connection)
17:03:44 Quit [Saint_] (Remote host closed the connection)
17:21:44 Quit jlbiasini (Ping timeout: 260 seconds)
17:21:49 Join jlbiasini [0] (
17:23:11 Quit pamaury (Ping timeout: 276 seconds)
17:27:47 Quit jlbiasini (Ping timeout: 260 seconds)
17:32:53 Quit mshathlonxp (Ping timeout: 252 seconds)
17:35:49 Quit enthdegree (Ping timeout: 252 seconds)
17:38:35funmansaratoga: still no nano :(
17:45:59gevaertsSo presumably we can't release on Sunday
17:47:55 Join LeoNerd [0] (
18:00:14 Quit remlap (Ping timeout: 260 seconds)
18:01:04 Join remlap [0] (
18:03:31 Join pamaury [0] (
18:03:31 Quit pamaury (Changing host)
18:03:31 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
18:04:01 Quit evilnick (Quit: Bye)
18:07:46 Quit remlap (Ping timeout: 244 seconds)
18:09:41 Join remlap [0] (
18:14:32 Join T44 [0] (
18:15:15 Quit passstab (Read error: Connection reset by peer)
18:17:39 Quit bluebrother (Disconnected by services)
18:17:44 Join bluebrother^ [0] (~dom@rockbox/developer/bluebrother)
18:17:47 Quit Topy44 (Ping timeout: 276 seconds)
18:20:07 Quit fs-bluebot (Ping timeout: 246 seconds)
18:20:53 Join jlbiasini [0] (
18:21:14 Join fs-bluebot [0] (
18:22:48 Join lebellium [0] (
18:22:59jlbiasinibluebrother^: where should be the RockboxUtility's changelog file? would rbutil/rbutilqt/CHANGELOG be ok?
18:23:24jlbiasiniI meant RockboxUtility's changelog text file
18:26:54lebelliumhello jlbiasini
18:28:07lebelliumdid you try the version 1.50 beta 3 of my theme? It doesn't use the hide_background anymore. Would be curious to know if it still crashes on USB connection with your fuze+ :)
18:31:52jlbiasinilebellium: cool let me try it!!
18:33:41lebelliumnow it displays the UI viewport full screen when the mini-player is off thanks to the latest skin engine improvements
18:34:14LeoNerdHeh. I had come in here to ask a question about WPS themes, but I've since solved it. :) So thanks all anyway for Rockbox...
18:34:16 Part LeoNerd
18:34:51lebelliumbut I wait for JdGordon to fix the skin engine bug before uploading the final 1.50 version on the theme website :)
18:40:03***Saving seen data "./dancer.seen"
18:44:52jlbiasinilebellium: this is VERY strange: it does work now but is still displaying corrupted graphics on LCD while connecting USB. Anyway after unplug it revert to normal mode with apparently no problem
18:45:42jlbiasiniThere are some glitches but I suppose that this is what JdGordon is suppose to solve?
18:46:50jlbiasinialso: don't forget to rename it to its proper name for the moment your themes name is "test"
18:47:26jlbiasinias long as it works i'm happy!
18:47:45jlbiasinithank for your work, this theme is really great
18:47:56 Quit stoffel (Read error: Operation timed out)
18:49:13lebelliumwhat are the glitches? The bug JdGordon tries to fix is this one:
18:51:14 Quit WalkGood (Quit: ♪ ♫ ♪ ♫ ♪ ♫ ♪)
18:51:16 Quit remlap (Ping timeout: 246 seconds)
18:52:09lebelliumhe committed a working fix some days ago but unfortunately this fix always broke the file browser with cabbie theme so he reverted back to the previous build.
18:54:13 Join biopyte [0] (~kiwi@
18:56:47biopytehi. i want to bye my mother an audiobook. however, she says, her mp3 (sony) player does not support bookmarking, therefor she doesnt listen audiobooks though she would like to. Question: Does Rockbox support audiobook playing well? so i might get her a different mp3 player and install rockbox. its all about convenient bookmarking, so the last position resumes next time listenig.
18:57:13biopyte bye -> buy
18:59:38gevaertsbiopyte: rockbox does have bookmarking, yes. I like it, but of course everyone has different opinions about how such things should work
18:59:43jlbiasiniI've been listening to audio books into rockbox for a while and I would say that bookmarking is not only well supported but also really sophisticate
19:00:04gevaertsYou don't need bookmarking for resuming at the right place though, rockbox does resume properly
19:00:22jlbiasinialso right
19:00:54gevaertsI'd still use it though, so you can play something else for a while and come back
19:02:06gevaertsIf you want to try it out, you can download a simulator from">
19:02:13jlbiasiniyou can even configure stuff like restarting a few second before your last stop so that you don"t lost the context of what you were listening
19:02:16biopytemyself, i have absolutely no experience with audibooks. so, using rockbox, when stopping listening, you can tag the position and it is memorized, completely independent from the file format?
19:02:25biopytejlbiasini, thx
19:03:11biopyte"you can even configure stuff like restarting a few second before"
19:03:53jlbiasiniI think that this is right for most of the common format but you might want to go to the audio format page of the wiki to check which format actually support it
19:08:18pamauryTorne: new version
19:09:32Tornethat looks nice
19:09:52Tornethe weak symbol override is lovely.
19:10:19Torneand, moving data_abort_handler out of crt0 entirely is a good plan too, since it's likely that anything we want to do with it in future is also not really target dependant
19:10:31Torneyou might even consider doing that for all the vectors :p
19:10:46Torneat least, all the ones that UIE
19:11:01 Join XavierGr [0] (~xavier@rockbox/staff/XavierGr)
19:11:21Torneanyway, yeah, taht's definitely the righ approach
19:11:26Tornethanks a lot for saving me doing it :p
19:12:21Torne(one nit: you only want \n in teh asm() block. \r is redundant)
19:12:41Torne(\n will already expand to the "right" newline sequence if you are on a mad platform taht really wants crnl
19:20:27 Quit anewuser ()
19:20:45pamauryok, I will do the change for all handlers that call UIE and for all crt0, testing will be needed. I really like the weak overriding, that's an elegant solution
19:21:31Torneyeah. basically any vectors that currently just handle the r14 offset, set a code and b UIE can stop being in crt0
19:21:39Torneand then in future we can implement, i dunno, paging :p
19:22:36 Join anewuser [0] (~anewuser@
19:22:38 Quit anewuser (Changing host)
19:22:38 Join anewuser [0] (~anewuser@unaffiliated/anewuser)
19:22:51pamauryI thought about implementing paging and virtual memory for some creative that only has 512Kb memory
19:22:59pamaurybut that sounds a little bit crazy :)
19:31:47 Join stoffel [0] (
19:32:54jlbiasinibluebrother^: see c/rockbox/+/187">g#187 :)
19:32:56fs-bluebotGerrit review #187 at,187 : Rockbox Utility: add a Changelog tab in "about" menu by Jean-Louis Biasini (changes/87/187/1)
19:38:46jlbiasinibluebrother^: I'm also wondering if all those text browser tab shouldn't adapt the text to the size of the border. (licence, speex license and changelog)
19:40:30 Join enthdegree [0] (~BitchX@unaffiliated/enthdegree)
19:40:34jlbiasinifor the next step I suppose we should change the "about" entry in the help menu to "about Rockbox Utility" and then create a second "about Rockbox" one. You do you think?
19:47:31 Quit lembas (Ping timeout: 245 seconds)
19:50:23 Quit kadoban_ (Read error: Connection reset by peer)
19:50:46 Join kadoban_ [0] (
19:52:18 Join passstab [0] (
19:54:53 Join curtism [0] (
19:54:53 Quit curtism (Changing host)
19:54:53 Join curtism [0] (~curtis@unaffiliated/programble)
19:55:38 Nick Guinness` is now known as Guinness (
20:03:50 Join remlap [0] (
20:13:03 Quit y4n (Quit: Today is the perfect day for a perfect day.)
20:23:48 Join Keripo [0] (
20:24:13 Join shanttu [0] (
20:30:30 Part jlbiasini
20:30:50 Join lembas [0] (
20:40:06***Saving seen data "./dancer.seen"
20:45:46 Join Misanthropos [0] (
20:49:05 Quit Keripo (Quit: Leaving.)
20:53:42 Join jlbiasini [0] (
20:56:15 Quit mortalis (Quit: KVIrc 4.1.1 Equilibrium
20:57:55 Join wodz [0] (
20:58:33 Join dfkt [0] (dfkt@unaffiliated/dfkt)
20:59:16wodzpamaury: I think you have an error in safe_read8. Shouldn't it be strneqb instead of streqb in line of 131
20:59:31 Quit wodz (Client Quit)
21:02:28 Quit biopyte (Quit: Leaving)
21:03:29 Quit TheLemonMan (Quit: WeeChat 0.3.7)
21:10:23 Join benedikt93 [0] (~benedikt9@unaffiliated/benedikt93)
21:13:24 Join c [0] (~charitwo@wikia/Charitwo)
21:13:43 Quit stoffel (Remote host closed the connection)
21:21:05 Quit XavierGr (Ping timeout: 264 seconds)
21:38:30 Join fml [0] (
21:46:37 Join stoffel [0] (
21:47:06 Join Zagor_ [242] (~bjst@rockbox/developer/Zagor)
21:53:23 Quit bitcraft (Remote host closed the connection)
22:02:12 Join Keripo [0] (
22:05:56 Quit enthdegree (Ping timeout: 250 seconds)
22:16:33 Quit stoffel (Remote host closed the connection)
22:19:27 Quit Keripo (Quit: Leaving.)
22:21:12 Join vagh [0] (~vagh@
22:27:18 Join topsyandpip56[PC [0] (
22:27:22 Nick topsyandpip56[PC is now known as topsyandpip56 (
22:29:16topsyandpip56Hello fellow Rockboxers, I'm having an issue relating to the iPod Mini 2G version of the software. The dualboot function of the RockBox bootloader seems broken, upon attempt to boot into the normal iPod OS causes a hang on the Apple screen until the iPod eventually reboots back into Rockbox. Any help would be appreciated, thanks.
22:34:37passstabwhat version?
22:35:28topsyandpip56The newest, installed yesterday
22:35:41topsyandpip56Wait, you mean Rockbox or the iPod OS?
22:36:28scorche|shdefine "the newest" - a current build? - a release? - what number does it say under rockbox info?
22:37:29topsyandpip56Bootscreen says 3.10
22:38:05CIA-44Commit 7676dae in rockbox by Alexander Levin: More meaningful name and type for 'root menu' setting
22:40:10***Saving seen data "./dancer.seen"
22:41:21topsyandpip56So I'm guessing 3.10 is a current build
22:41:32topsyandpip56Although the topic says something about 3.11 being branched
22:41:34CIA-447676dae build result: All green
22:42:36topsyandpip56*release not current build
22:49:50 Quit benedikt93 (Quit: Bye ;))
22:55:08 Quit fml (Remote host closed the connection)
22:56:30 Join Keripo [0] (
22:59:56 Quit vagh (Quit: Leaving)
23:03:15topsyandpip56Is anybody able to assist with this problem?
23:08:14 Join n1s [0] (
23:08:14 Quit n1s (Changing host)
23:08:14 Join n1s [0] (~n1s@rockbox/developer/n1s)
23:09:25 Join [Saint] [0] (~Saint]@
23:09:25 Quit [Saint] (Changing host)
23:09:25 Join [Saint] [0] (~Saint]@unaffiliated/saint/x-8516940)
23:12:23 Quit shanttu (Quit: Leaving)
23:23:42 Join stripwax [0] (
23:28:55Misanthroposi have a sansa clip+ with a very recent rockbox svn version but an older patched bootloader - i want to boot rockbox on default - even when connecting to usb if the device was turned off
23:29:24Misanthroposthe current tells me it wil boot OF if applied
23:29:48MisanthroposI mean if I reinstall the OF with the bootloader
23:30:09MisanthroposI read something about that on the mailinglist
23:30:33Misanthroposhave you reverted the bootloader to old behavior or not?
23:31:27Misanthroposfrom that given link: You've successfully installed the bootloader and Rockbox. It boots by default. For booting the OF press |<< very quickly or plug the cable in while the device off (plugging the cable doesn't work on Clipv2 and Clip+).
23:31:42Misanthroposis the cable section still true?
23:36:49dfktmy clip+ has the 4.0 bootloader and a recent-ish firmare build - it boots rockbox when plugged in, and the of when plugged in while holding the left button
23:37:04dfkts/of/OF :)
23:38:25Misanthroposthank you dfkt, how do i know the version of the bootloader?
23:38:42dfktit shows it in the first second while booting
23:38:47dfktmaybe less than a second
23:39:01Misanthroposoh i - see . is it the one on: ?
23:39:16dfkti don't know, i've compiled my own
23:39:23Misanthroposi c
23:39:49 Quit pamaury (Remote host closed the connection)
23:40:01dfktbut i guess 4.0 is still the latest one
23:40:04Misanthropossorry i forgot.. you dont like abbreviations :)
23:40:58dfktyou can use mine if you want -
23:41:13dfkti mean, ready-made OF with 4.0 bootloader
23:42:54Misanthroposi see that - do i have to boot OF to reinstall it?
23:43:53dfktyou can drop it over in rockbox, but you have to initiate an OF boot for the update
23:44:38Misanthroposok - thank you - have you used the lates OF update?
23:45:03Misanthroposon my way then copying the file
23:45:31 Quit Keripo (Quit: Leaving.)
23:47:20Misanthroposupgrade in progress... hopefully i will never see the #PORT# or #bla# directories again :-D
23:47:33dfktheh, indeed
23:48:40Misanthroposgreat!! thank you! it just works as expected
23:48:53dfktyou're welcome :)
23:51:29 Quit Zagor_ (Quit: Leaving)
23:57:08 Quit remlap (Read error: Operation timed out)

Previous day | Next day