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

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

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

#rockbox log for 2010-04-08

00:00:27 Join soap_ [0] (~soap@rockbox/staff/soap)
00:00:50 Quit DataGhost (Ping timeout: 264 seconds)
00:00:59rd1381by copying files
00:01:24gevaertswhich operating system?
00:01:50rd1381doesnt matter
00:01:55rd1381both linux and windows
00:02:20rd1381had this issue in opensuse and xp and vista
00:02:27rd1381and now ubuntu and windows 7
00:02:36gevaertsIn windows, you can probably get it faster by enabling write caching. In linux, try checking if it's not mounted with -o sync
00:02:42rd1381so i think its safe to say its not system related
00:03:20rd1381as a side note i remember that in windows every time it was recognized as something diferent
00:03:40rd1381like it would say toshiba somthing attached
00:03:49rd1381and next time something else
00:04:11rd1381but booting with ipod firmware always recognized it correctly
00:04:21rd1381so maybe that the issue
00:04:23Tornetoshiba MKthingy is correct
00:04:28Tornethat's the model of hard drive in the player
00:05:40rd1381oh
00:05:45 Quit domonoky (Read error: Connection reset by peer)
00:05:49rd1381nice
00:05:49 Quit Topy (Ping timeout: 276 seconds)
00:06:13 Join Topy44 [0] (~Topy44@f048209063.adsl.alicedsl.de)
00:06:16rd1381cant you tell it to ,you know recognize it as rockbox?
00:06:18Tornebut no, rockbox does identify itself with different strings to the original firmware
00:06:24rd1381that for windows to decide?
00:06:28 Quit TopyMobile (Ping timeout: 276 seconds)
00:06:36Tornewindows doesn't use the strings for anything except displaying them to you
00:06:45Torneso it doesn't make any difference
00:06:55rd1381so the speed issue is not from that?
00:07:07Torneno
00:07:36TorneI'm not sure which devices we have a known issue on
00:07:45Tornebut in general there are models of ipod where our disk access over usb is slower
00:08:09Tornewe don't know what the problem is, really, or we'd've done something about ti
00:08:30rd1381oh i see
00:09:20 Quit jgarvey (Quit: Leaving)
00:09:23rd1381do you use a diferent protocol than the one ipod use? or is it hardware access that you both use?
00:10:00Tornewe have our own implementation of USB and disk access and so oon.. we don't use any of the code from the original firmware
00:10:16Tornethe ipod's USB stack is software, so the OF's implementation is completely separate
00:10:38rd1381but u use some hardware command to access harddrive right? maybe its not the same as iod
00:11:09gevaertsWell, our code is available, so you can check
00:11:16 Nick Ypsy is now known as YPSY (~ypsy@geekpadawan.de)
00:13:36rd1381sorry i am not a probramer (not yet) i just read some stuuf on internet and wanted to put my 2 cent
00:13:51rd1381thats all
00:14:12 Part Boldfilter
00:14:13gevaertswell, we can only assure you that we don't make things slower on purpose
00:15:00rd1381what i mean tis that specification for that toshiba hardrive and its bus command and other stuff is not available to you>?
00:15:39 Join mirak [0] (~mirak@85-171-108-41.rev.numericable.fr)
00:16:04gevaertsThat one is easy, it's standard ATA. We don't have a specification for most of the rest of the ipod though
00:17:26 Join Darkknight512 [0] (~Darkknigh@CPE00212968356c-CM00186845dd46.cpe.net.cable.rogers.com)
00:17:54 Quit Darkknight512 (Client Quit)
00:18:07rd1381so you mean some where in those specific ipod somethings slow down speed that apple know about and you dont? so its not code related(not hardware access code but rockbox base code i mean)
00:19:33rd1381thanks for all your answers. gotta go .bye
00:19:39 Quit rd1381 (Quit: ChatZilla 0.9.86 [Firefox 3.6.4pre/20100405044455])
00:19:54 Join Boldfilter [0] (~Boldfilte@adsl-178-246-170.jax.bellsouth.net)
00:22:50 Quit mirak (Quit: Ex-Chat)
00:25:21 Join Topy [0] (~topy@my.fastsh.it)
00:28:08 Quit Topy44 (Quit: Leaving)
00:31:43 Join mikroflops_ [0] (~yogurt@90-227-45-110-no112.tbcn.telia.com)
00:33:46 Join selectohh [0] (~irc@ip216-239-86-80.vif.net)
00:34:40selectohhwhen downloading a build from svn co, when i make the build and try and play an mp3 it says "codec failure".. any ideas what i'm missing here?
00:35:02Tornedid you install the entire build? the zipfile?
00:35:38selectohhi've been using svn co svn://svn.rockbox.org/rockbox/trunk
00:35:44selectohhi guess that's the wrong way?
00:36:07 Quit mikroflops (Ping timeout: 265 seconds)
00:36:23Torneno, i mean after you comiled it, did you install all the files, or just the main rockbox file?
00:36:49Torneyou need to update the codecs/plugins/etc if you replace the main build
00:36:52selectohhya i just dragged the whole thing into the rockbox folder on the mp3 player.. after deleting the rockbox folder and all it's subfolders on the mp3 player
00:37:11Tornethe build directory?
00:37:16Tornethat's not laid out correctly
00:37:28Torneyou need to run make zip or make fullzip to get a zip like the normal downloaded builds
00:37:51selectohhtorne: ya, like i took everything from the rockbox folder on the vmware virtual drive and dragged it into the rockbox folder of the mp3 player
00:38:02Tornedon 't do that.
00:38:04selectohhtorne: does make zip make a different chunk of stuff? i thought it just made a zip of the same files
00:38:09Torneno
00:38:28Torneit copies the files into the correct structure for the .rockbox directory
00:38:29Tornethen zips that
00:38:52selectohhtorne: ah the missing link i suppose (between me and happy codecs). thanks i'll try that. so make zip and then unzip instead of make + drag/drop
00:39:01Torneyes.
00:39:23selectohhtorne: i appreciate it thanks. about 4 hours of fussing solved hopefully.
00:40:13 Quit anewuser (Ping timeout: 268 seconds)
00:44:06 Join Rotting [0] (~Adubb@67.201.160.144)
00:44:06 Nick Rotting is now known as Adubb (~Adubb@67.201.160.144)
00:55:53 Quit Adubbb ()
00:56:56 Quit Adubb (Quit: <Tobin> japanese men make hot trannies)
00:58:16 Join Adubb [0] (~Aldubuc@67.201.160.144)
01:00
01:01:17 Quit ender` (Quit: Please help Conserve Gravity - Play Chess, not Basketball.)
01:03:29 Join Adubbb [0] (~Aldubuc@67.201.160.144)
01:04:04***Saving seen data "./dancer.seen"
01:10:21 Quit Adubbb ()
01:10:22 Join JohannesSM64 [0] (~johannes@cm-84.215.75.42.getinternet.no)
01:12:29 Quit linuxstb (Ping timeout: 246 seconds)
01:13:33 Join linuxstb [0] (~linuxstb@rockbox/developer/linuxstb)
01:21:17 Quit antil33t (Ping timeout: 258 seconds)
01:28:36 Join antil33t [0] (~Mudkips@203-184-54-232.callplus.net.nz)
01:31:38 Quit kugel (Remote host closed the connection)
01:38:53 Nick fxb__ is now known as fxb (~felixbrun@h1252615.stratoserver.net)
01:41:02 Quit Stephen__ (Quit: Leaving)
01:41:37 Part toffe82
01:46:39 Join CaptainKewl [0] (jds@207-237-106-60.c3-0.nyr-ubr1.nyr.ny.cable.rcn.com)
01:50:38 Quit selectohh ()
02:00
02:00:02 Quit Schmogel (Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org)
02:06:22 Join Adubbb [0] (~Aldubuc@67.201.160.144)
02:06:47 Join RadicalR [0] (~radicalr@c-69-255-49-110.hsd1.va.comcast.net)
02:08:23 Quit Adubb (Read error: Connection reset by peer)
02:09:38 Quit yegord (Ping timeout: 268 seconds)
02:16:04 Quit GeekShadow (Quit: The cake is a lie !)
02:22:51 Quit evilnick_B (Quit: Page closed)
02:26:30 Quit drostie (Quit: even marathon runners need to nap / i ran all the way there and then ran back / but back was gone...)
02:29:23 Join JdGordon [0] (~7bf38c1f@gateway/web/freenode/x-uzkgpooaeaqzowfm)
02:33:10 Quit komputes (Read error: Connection reset by peer)
02:39:31 Nick fxb is now known as fxb__ (~felixbrun@h1252615.stratoserver.net)
03:00
03:04:07***Saving seen data "./dancer.seen"
03:12:03 Quit CaptainKewl (Quit: ( www.nnscript.com :: NoNameScript 4.22 :: www.esnation.com ))
03:14:13 Quit dfkt (Quit: -= SysReset 2.53=- Ph'nglui mglw'nafh Cthulhu R'lyeh wgah'nagl fhtagn.)
03:35:27 Quit adnyxo (Remote host closed the connection)
03:41:33CIA-5New commit by Blue_Dude (r25528): Manual update for hotkey reference
03:43:52CIA-5New commit by Blue_Dude (r25529): Hotkey: better settings handling, fewer saved variables, localizable hotkey info list
03:47:59 Join Blue_Dude [0] (~chatzilla@adsl-235-206-131.mco.bellsouth.net)
03:49:01Blue_DudeWill be AFK, but please check if FS #11181 is fixed with r25529. Thanks.
03:50:37 Join phanboy4 [0] (~benji@c-174-49-112-244.hsd1.ga.comcast.net)
03:51:43 Join lennyk [0] (~lenny@adsl-75-5-198-147.dsl.snlo01.sbcglobal.net)
03:51:53 Join anewuser [0] (anewuser@unaffiliated/anewuser)
03:54:34 Quit S_a_i_n_t (Ping timeout: 240 seconds)
04:00
04:14:06 Quit TheSeven (Disconnected by services)
04:14:18 Join The_Seven [0] (~theseven@rockbox/developer/TheSeven)
04:14:28 Nick The_Seven is now known as TheSeven (~theseven@rockbox/developer/TheSeven)
04:24:18JdGordonBlue_Dude: one thiing which would be nice to consider... add the lang strings for the settings so the hotkeys can be added to the quickscreen
04:25:11 Join Barahir [0] (~jonathan@gssn-5f756869.pool.mediaWays.net)
04:26:12 Quit Barahir_ (Read error: Operation timed out)
04:30:58 Join Rob2223 [0] (~Miranda@p4FDCBA4C.dip.t-dialin.net)
04:32:55 Part Adubbb
04:34:31 Quit Rob2222 (Ping timeout: 265 seconds)
04:56:34CIA-5New commit by Blue_Dude (r25530): Add lang strings to hotkey setting macros
05:00
05:00:38 Quit maraz (Remote host closed the connection)
05:01:25Blue_DudeJdGordon: That should probably do it. I don't have a LANG yet for the setting itself. So far I haven't needed one.
05:02:01Blue_DudeI wonder if the data abort on Nano's is fixed.
05:04:11***Saving seen data "./dancer.seen"
05:04:28 Join S_a_i_n_t [0] (S_a_i_n_t@203.184.3.121)
05:05:19JdGordonI think putting the hotkey setting on the quickscreen makes the most sense
05:06:21Blue_DudeYou mean as opposed to a special keystroke on the context menu? That's not a bad idea.
05:06:37JdGordonnot instead of
05:06:42JdGordonhave both
05:06:47Blue_DudeIn addition to?
05:07:08S_a_i_n_tI think just adding "select as hotkey" (or similar) in the context menu kinda makes the most sense...but that's just me.
05:07:47JdGordoncontxt menus on context menus are NOT a good idea
05:08:03Blue_DudeYou'd still need either a menu within a menu, or another context menu item with a list of acceptable functions.
05:08:04JdGordonbut pressing it on the contetx menu sort of works
05:08:25Blue_DudeSort of.
05:08:53Blue_DudeI originally wanted to use the context menu keystroke to set the hotkey but it was complicated.
05:09:37Blue_DudeAnd probably a bad idea.
05:09:57S_a_i_n_twhat about using "long select" as "set hotkey" on the context items that can be set? long select doesn't seem to do anything in the context menu.
05:10:20 Join BHSPitMonkey [0] (~stephen@unaffiliated/bhspitmonkey)
05:10:39Blue_DudeLong select *is* the context menu key on some targets. And it means defining yet another special keystroke. There are only so many to go around.
05:11:46S_a_i_n_tI mean, once one is in the context menu.
05:11:52S_a_i_n_tnot to get to it.
05:12:08JdGordonlong-select is the context menu key on all targets
05:12:31S_a_i_n_tyes, but once your in the conetxt menu, long select does nothing.
05:12:33Blue_DudeYeah, but that's the problem I had. You'd be using the context menu key within the context menu. It was complicated.
05:12:41S_a_i_n_tOnly short select seems to trigger the selections.
05:13:24Blue_DudeActually.... Hm. I think I already solved the complications by setting a flag within menu.c. That might work now.
05:14:08Blue_DudeIf that's the way you want to go. It might be better than a combo-key on some targets.
05:15:18Blue_DudeYou *could* set without leaving the select key and the scrollwheel. Easier to navigate...
05:15:22Blue_DudeHm.
05:17:57S_a_i_n_tWhat i'm thinking is that, long select (whilst already in the context menu) defines the currently selected item (if it can be set), maybe you get a splash for confirmation...but not a yes/no screen. I'm pretty sure you could get away with it and still leave the function of "short select" intact.
05:18:02Blue_DudeYeah, I see no reason now why that wouldn't work. I solved the main problems while solving a different one and never went back to the context key idea.
05:18:13JdGordoni disagree... there is a disconnect between using context to set the hotkey, and then using the hotkey combo later
05:18:36S_a_i_n_tHmmm...this is true.
05:19:14S_a_i_n_tI'm just trying to think to take the fuck-assery out of acually setting a hotkey.
05:19:19Blue_DudeI would still rather have the confirmation screen. It's just too easy to change the setting by accident otherwise.
05:20:29 Nick krazykit` is now known as krazykit (~kkit@ppp-70-236-32-206.dsl.ipltin.ameritech.net)
05:20:31Blue_DudeAnother context menu submenu maybe. There's precedent for that. Playlist for example.
05:21:22Blue_DudeWhich leaves a disconnect of another kind: it separates the context menu from the hotkey setting. It's a bit artificial.
05:21:37Blue_Dude*shrug*
05:27:37Blue_DudeJdGordon: I was just browsing around the menu.c code. You were thinking about adding a hotkey setting option in there along with reset setting, etc? I didn't think menu items were quickscreen-able.
05:33:58Blue_DudeI changed menu.c to set the hotkey with the context key. It was *really* straightforward. I can commit this if that's where we want to go.
05:37:28Blue_DudeActually, I can hijack just about any keypress in this one case. Not many of them make much sense, though.
05:45:09 Join krazykit` [0] (~kkit@ppp-70-236-46-205.dsl.ipltin.ameritech.net)
05:47:31 Quit shai (Read error: Connection reset by peer)
05:47:43 Join shai [0] (~Shai@l192-117-110-233.cable.actcom.net.il)
05:48:08 Quit Blue_Dude (Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100401080539])
05:48:28 Quit krazykit (Ping timeout: 265 seconds)
05:49:49 Quit stavrob (Ping timeout: 246 seconds)
05:50:05 Join stavrob [0] (~sam@78-105-125-218.zone3.bethere.co.uk)
05:54:10 Join Blue_Dude [0] (~chatzilla@adsl-235-206-131.mco.bellsouth.net)
05:55:50Blue_DudeOne more quick idea for consideration: what about making the default behavior for both hotkeys be off. So reset settings would disable both of them. Also make the default "off" behavior do nothing at all, not even the splash. You could turn them on by setting them, but otherwise the key does nothing.
05:56:04Blue_DudeAnyway, off to bed.
05:56:15 Quit Blue_Dude (Client Quit)
06:00
06:02:35 Quit xiainx (Ping timeout: 245 seconds)
06:20:08 Quit anewuser (Quit: http://xrl.us/Renoise Like renoise + like music? 3 days to submit your entry!)
06:22:25 Join xiainx [0] (~iain@modemcable091.119-201-24.mc.videotron.ca)
06:31:31*S_a_i_n_t grumbles about http://themes.rockbox.org/index.php?themeid=419&target=ipodnano1g still being in the 1/2G Nano themes.
06:31:52S_a_i_n_tIt's managed to clock a lot of downloads for someting that shouldn't be there.
06:34:22S_a_i_n_tAnd, if anyone is so inclined...can http://themes.rockbox.org/index.php?themeid=406&target=ipodnano1g be taken down also? It was always my intention to remove that theme when I'd finished "Symmetry".
06:34:51 Quit Horscht (Quit: Verlassend)
06:43:42 Quit moos (Ping timeout: 246 seconds)
06:47:22 Join TillW [0] (~Till@h49-net09.simres.netcampus.ca)
06:48:32S_a_i_n_tHmmmm, amI right in thinking that "idle timout" only starts counting down after the backlight timeout+whatever fade duration you have set? Should it not be calculaed at "idle timeout = idle timeout set by user -backlight timeout period -backlight fade duration"?
06:50:18 Quit xiainx (Quit: Good Bye!)
07:00
07:02:35 Join JdGordon_ [0] (~7bf38c1f@gateway/web/freenode/x-thqarlrhgiuimeyq)
07:04:14***Saving seen data "./dancer.seen"
07:05:07 Quit JdGordon (Ping timeout: 248 seconds)
07:11:44 Quit amr (Ping timeout: 245 seconds)
07:13:32 Quit alexbobp (Read error: Operation timed out)
07:13:41 Join alexbobp [0] (~alex@66.112.249.238)
07:17:38 Quit panni_ (Quit: ( www.nnscript.de :: NoNameScript 3.81 :: www.XLhost.de ))
07:22:48S_a_i_n_tTorne: Just had something a little crappy happen, need to turn one of my nanos on using hard reset.
07:23:09S_a_i_n_toh, *with* the alternative (iram) shutdown fix :/
07:24:36S_a_i_n_tFrom what I remember, I didn't shut down the ipod manually, I just pulled out the earbuds while it was playing and it paused, and I guess timed out in the usual fashion.
07:24:45S_a_i_n_tNeeded to hard resatrt it just now.
07:28:37 Join amr [0] (~quassel@41.234.26.97)
07:29:53S_a_i_n_tTorne: it is even reproducable....fuck it.
07:31:15S_a_i_n_tHave track playing in wps, pull out earbuds (set to pause/resume on headphone insertion), player pauses, eventually times out and shuts down, needs a hard restart.
07:32:41S_a_i_n_tHmmm...just started up fine trying to get it to do it a fourth time. Odd.
07:33:04 Quit amr (Remote host closed the connection)
07:35:48 Join lostlogic [0] (~lostlogic@rockbox/developer/lostlogic)
07:36:07 Quit alexbobp (Ping timeout: 248 seconds)
07:36:07 Quit logiclost (Ping timeout: 248 seconds)
07:37:09 Quit blithe (Ping timeout: 276 seconds)
07:37:54 Join alexbobp [0] (~alex@66.112.249.238)
07:38:36 Join blithe [0] (~blithe@72.14.176.144)
07:38:37S_a_i_n_tGah! Bloody typical...apparently my method to reproduce it is wrong/inconsistent, or not valid at all. I managed to get it to need hard restart to power on 3 times in a row and now I can't get it to do it anymore. Worked fine for the last ~10 shutdowns straight trying to reproduce it consistently.
07:40:57pixelmaS_a_i_n_t: is it reproducible with a "usual" pause too? Also... I think there was something in the guidelines about this being a family-friendly channel
07:42:38pixelmaon another note and a bit late but - not all targets even have a Select button (just saying be careful with assumptions)
07:45:55S_a_i_n_tpixelma: As for "reporducible" I jumped the gun there slightly, it needing hard restart three times in a row seemed rather conclusive...untill the fourth/fifth/sixth time etc.
07:47:55S_a_i_n_tIt seemed like I had a formula to do it fine, then it just started working perfectly again...its very odd. My targets never even experienced the "shutdown issue" often enough to call myself "affected" by it, then it does it three times in a row withthe new fix that seems to be working fine for the one's that *are* affected by the shutdown issue.
07:48:05S_a_i_n_tIts nuts.
07:51:44 Join yegord [0] (~yegor@2001:67c:7c:40d4:224:7eff:fedf:124d)
07:52:11pixelmaI wouldn't be surprised if idle power-off wasn't a very clean shutdown. That's only my impression as a user because I saw weird things happen from time to time if it kicks in especially when in plugins (on all targets I have, not even an Ipod).
07:54:39pixelmaweird things ...e.g. backlight comes on like in a usual shutdown but then the player just sits in some plugin screen, sometimes a keypress helps continuing the shutdown process
07:55:50 Join stoffel [0] (~quassel@p57B4B975.dip.t-dialin.net)
08:00
08:00:35S_a_i_n_tpixelma: I'm just baffled by the fact that I managed to get it to do it 3 times in a row (or, it was indeed random...which is statistically nuts), and now its jsut working fine again. I've power cycled it in all the ways I can imagine to try reproduce the flaw, but now I just can't.
08:01:53S_a_i_n_tMaking me thing that it happening three times in a row was indeed random, but the odds of that happening (especially considering the fact that I wasn't plagued by this problem to begin with (before the patch)) are simply remarkable.
08:02:06S_a_i_n_ts/thing/think/
08:10:49 Join JdGordon [0] (~jd@110.22.112.19)
08:10:49 Quit JdGordon (Changing host)
08:10:49 Join JdGordon [0] (~jd@rockbox/developer/JdGordon)
08:11:27 Quit xavieran (Remote host closed the connection)
08:12:30 Join xavieran [0] (~xavieran@dsl-220-235-119-86.sa.westnet.com.au)
08:12:47CIA-5New commit by FlynDice (r25531): sd-as3525v2: Start out with all MCI Interrupts masked. ...
08:12:53CIA-5New commit by FlynDice (r25532): Sansa AMSv2: Do not disable DMA synchronization logic for as3525v2. The DMA controller and the SD controller run at different clock speeds.
08:13:45*FlynDice got 6 hrs without freezeups with ^^
08:20:12 Quit Boldfilter (Quit: Boldfilter)
08:23:38 Join _jhMikeS_ [0] (~jethead71@adsl-75-45-228-135.dsl.sfldmi.sbcglobal.net)
08:23:38 Quit _jhMikeS_ (Changing host)
08:23:38 Join _jhMikeS_ [0] (~jethead71@rockbox/developer/jhMikeS)
08:23:38 Quit jhMikeS (Disconnected by services)
08:26:07 Join Zagor [0] (~bjst@rockbox/developer/Zagor)
08:28:28 Join kramer3d [0] (~kramer@unaffiliated/kramer3d)
08:35:10 Join ender` [0] (krneki@foo.eternallybored.org)
08:39:36 Quit pixelma (Disconnected by services)
08:39:38 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
08:39:57 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
08:40:25 Quit amiconn (Disconnected by services)
08:40:27 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
08:40:48 Join mitk [0] (~mitk@195.117.162.130)
08:40:49 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
08:49:17 Join r2k000 [0] (~r2000@c-67-164-21-71.hsd1.ca.comcast.net)
08:52:35 Join flydutch [0] (~flydutch@host225-165-dynamic.15-87-r.retail.telecomitalia.it)
08:54:35 Join petur [0] (~petur@rockbox/developer/petur)
08:57:05 Quit yegord (Quit: ะฃั…ะพะถัƒ ั ะพั‚ ะฒะฐั (xchat 2.4.5 ะธะปะธ ัั‚ะฐั€ัˆะต))
09:00
09:04:17***Saving seen data "./dancer.seen"
09:09:59topikgetting any closer to the writability holy grail, FlynDice ?
09:19:23GodEaterseems I need to take a look at FS #11149 too. My iPod is powering itself on all by itself at odd intervals
09:21:28 Join lpereira [0] (~lucien@did75-8-82-226-27-213.fbx.proxad.net)
09:23:09pixelmaGodEater: I remember reports of this earlier and relating it to alarm settings (maybe in the OF, it's been a while)... could be other reasons now and I don't have an Ipod so this is just what I remember reading here)
09:23:19 Join wodz [0] (~wodz@skatol.ch.pw.edu.pl)
09:24:40 Join LinusN [0] (~linus@rockbox/developer/LinusN)
09:26:41pixelmafrom the theme site (Fuze themes): " I just added some Vista Backgrounds, Vista copyright from Microsoft." errmm... Vista copyright?
09:28:01wodzpixelma: have You tried FS #11189?
09:30:24pixelmano, not yet but I compiled it. Still need to put it on the Ondio and test a bit which I can't do while at work
09:31:21wodzok. Please leave a comment in tracker than.
09:33:50 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
09:39:29 Join kramer3d_ [0] (~kramer@unaffiliated/kramer3d)
09:40:22 Quit kramer3d (Disconnected by services)
09:40:26 Nick kramer3d_ is now known as kramer3d (~kramer@unaffiliated/kramer3d)
10:00
10:00:15 Quit jordan` (Remote host closed the connection)
10:05:41 Quit lpereira (Ping timeout: 240 seconds)
10:09:49 Quit BHSPitMonkey (Remote host closed the connection)
10:10:59JdGordon_pixelma: yeah, looks like it should be removed
10:11:12JdGordon_or did we decide to not activly look for (c) infringments?
10:11:16*JdGordon_ cant remember
10:15:05 Quit JdGordon (Quit: Bye)
10:20:27 Join Adubb [0] (~Aldubuc@67.201.160.144)
10:23:15Llorean We don't actively look for them, but we remove them if they're randomly spotted or reported to us.
10:27:50 Join lpereira [0] (~lucien@did75-8-82-226-27-213.fbx.proxad.net)
10:33:51S_a_i_n_tIn that case...http://themes.rockbox.org/index.php?themeid=419&target=ipodnano1g (apple_ipod) needs to go. I've reported it on IRC and through the theme site a few times now...
10:35:45wodzThe engineer form MPIO who decided to byteswap ATA lines should fry in hell
10:36:02Zagorreminds me of archos
10:36:16S_a_i_n_twodz: that's errr...."nice" :-S
10:37:09Zagor"we'll fix that in software"
10:37:31LinusNa classic example of lack of communication between hw and sw engineers
10:37:59wodzok. I hacked ata-as-coldfire.S to byteswap (it is working) but looks ugly
10:40:50wodzI am doing something wrong apparently - debug screen shows ~108MHz when playing mp3
10:41:07wodzIs there a way to find out what hogs cpu?
10:44:27wodzplaying ogg shows 124MHz (100% boost) but is realtime
10:45:21wodzok it is a bit lower ~121MHz (97% boost)
10:48:34 Join rvvs89_ [0] (robotnik@bright-snat.ucc.asn.au)
10:48:45 Quit rvvs89_ (Changing host)
10:48:45 Join rvvs89_ [0] (robotnik@pdpc/supporter/base/rvvs89)
10:49:32wodzdelta between ogg and mp3 is comparable to results posted in http://www.rockbox.org/wiki/CodecPerformanceComparison so I suppose it is not problem in codec code.
10:51:22 Quit rvvs89 (Quit: Changing hosts.)
10:52:03 Nick rvvs89_ is now known as rvvs89 (robotnik@pdpc/supporter/base/rvvs89)
10:52:08*wodz looks for advice how to narrow performance problem
10:59:21 Quit stoffel (Remote host closed the connection)
11:00
11:01:52 Quit bluebrother (Disconnected by services)
11:01:53 Join bluebroth3r [0] (~dom@rockbox/developer/bluebrother)
11:02:15linuxstbwodz: I think there's a #define somewhere which will enable logging of calls to cpu_boost(). That may help you if I'm understanding your problem correctly...
11:03:54wodzisn't there some 'top' equivalent? I mean to find out what cpu time each thread gets
11:04:18***Saving seen data "./dancer.seen"
11:10:40 Join Kitr88 [0] (~Kitr88@BSN-142-84-96.dial-up.dsl.siol.net)
11:10:55 Part LinusN
11:11:16 Join LinusN [0] (~linus@rockbox/developer/LinusN)
11:11:35 Part LinusN
11:11:39 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
11:11:50 Quit lpereira (Quit: Leaving.)
11:11:52 Join lpereira [0] (~lucien@did75-8-82-226-27-213.fbx.proxad.net)
11:12:56 Quit Kitar|st (Ping timeout: 276 seconds)
11:14:26 Join LinusN [0] (~linus@giant.haxx.se)
11:14:27 Quit LinusN (Changing host)
11:14:27 Join LinusN [0] (~linus@rockbox/developer/LinusN)
11:15:34 Quit Kitr88 (Ping timeout: 264 seconds)
11:15:38wodzHmm it is not a problem of running inside gdb session. Normal run gives exactly the same result.
11:15:59 Quit lpereira (Ping timeout: 246 seconds)
11:20:11 Join Kitar|st [0] (Kitr88@BSN-182-7-204.dial-up.dsl.siol.net)
11:23:03 Join TheMusicGuy [0] (~TheMusicG@24-217-49-9.dhcp.stls.mo.charter.com)
11:26:14TheMusicGuyHi. I was wondering if anybody knows what the hangup is with the lack of file write support for the Sansa Clip+ (in somewhat un-technical terms if possible)
11:26:58TheMusicGuynot asking when it will be fixed, just wondering what has to happen before it can be fixed.
11:32:19 Join lpereira [0] (~lucien@did75-8-82-226-27-213.fbx.proxad.net)
11:33:58CIA-5New commit by pamaury (r25533): rbutil: update french translation.
11:39:27 Part TheMusicGuy
11:42:38 Quit ender` (Read error: Connection reset by peer)
11:43:01 Join ender` [0] (krneki@foo.eternallybored.org)
11:44:20 Quit kramer3d (Ping timeout: 246 seconds)
11:46:10 Quit arbingordon (Ping timeout: 264 seconds)
11:46:58 Join arbingordon [0] (~w@unaffiliated/arbingordon)
11:48:54wodzstrange I replaced blocking loop in adc_scan with sleep(HZ/10) and now getting *PANIC* stkov 40000001
11:49:37 Join halmi [0] (~Miranda@80-123-36-32.adsl.highway.telekom.at)
11:49:46TorneS_a_i_n_t: hm. drat? :)
11:50:12Tornepixelma: idle poweroff should shut down cleanly, but *all* shutdowns, if impeded by the thing running in the foreground, force a hard shutdown after a timeout
11:50:29S_a_i_n_tTorne: Yeah...I *really* don;t know what was going on there...
11:50:35Tornepixelma: when a shutdown event gets queued, either from a manual or idle shutdown, it broadcasts the event to all threads and starts a timer
11:51:03S_a_i_n_tI must have power cycled the player about ~40 times or so trying to reproduce it after those three times...but its running fine.
11:51:05TorneIf the timer actually makes it to expiry we power off anyway
11:51:07 Join evilnick__ [0] (~evilnick@ool-457bccf5.dyn.optonline.net)
11:51:14S_a_i_n_tWhich is good, but a bit weird.
11:51:17Tornethe main thread is supposed to do the clean shutdown stuff and power off before hten
11:51:40Torneif it gets blocked by, say, a blocking UI call (e.g. the bookmarking bug I fixed recently, or a plugin) then it will indeed hard poweroff
11:51:54Tornehowever my clearing-IRAM code is in the hard poweroff handler :)
11:52:08Torneso for that purpose it shouldn't matter whether it was a clean shutdown or not.
11:52:26S_a_i_n_theh...well, no.
11:53:25Tornei still haven't had my player needing a hard reset.
11:53:32 Nick halmi is now known as halmi_ (~Miranda@80-123-36-32.adsl.highway.telekom.at)
11:54:38 Nick fxb__ is now known as fxb (~felixbrun@h1252615.stratoserver.net)
11:54:45 Nick fxb is now known as fxb__ (~felixbrun@h1252615.stratoserver.net)
11:55:03 Quit evilnick_ (Ping timeout: 276 seconds)
11:55:27S_a_i_n_tI tried to work out the odds of having to hard restart the player 3 times in a row with *seemingly* the same recipie for reproducing it and that recipe apparently having absolutely nothing to do with it. I'm just baffled about those three startups.
11:56:09S_a_i_n_tIt really did seem like I found a recipie to reproduce it, then it just works fine again. :-S
11:57:02CIA-5New commit by pamaury (r25534): rbutil: update french translation.
11:57:50 Nick halmi_ is now known as halmi (~Miranda@80-123-36-32.adsl.highway.telekom.at)
11:59:42Tornecan you test if your settings allow idle poweroff to cleanly shutdown?
12:00
12:00:02S_a_i_n_thow?
12:00:30 Nick halmi is now known as halmi_ (~Miranda@80-123-36-32.adsl.highway.telekom.at)
12:01:06Tornewatch it? :)
12:03:27S_a_i_n_tOh right...yes, I have seen it power off a few times in the last few hors. It seems to shutdown cleanly, except for maybe every 1/10 shutdowns leaves something from the screen it was displaying at shutdown slowly fading out on the screen after it shuts diwn.
12:03:32S_a_i_n_t*down
12:05:29TorneThat's not a clean shutdown then
12:05:35S_a_i_n_tIt usually leaves an icon from my .sbs slowly fading out until the screen runs out of juice in that case. It seems to be the battery icon most times.
12:06:12Torneyeah, that doens't sound like a clean shutdown.
12:06:18Torneit should turn the LCD off properly
12:07:03S_a_i_n_tThat's maybe every 1 out of 10 times though, possibly less. But something I've definitely noticed in te past few hours trying to reproduce that elusive startup issue.
12:07:57 Nick halmi_ is now known as halmi (~Miranda@80-123-36-32.adsl.highway.telekom.at)
12:09:18Torneyeah, if you see clean_shutdown() in apps/misc.c it clears the screen near the start
12:09:37Tornewait, but then it does splashes after that.
12:09:40Tornehmm
12:09:57S_a_i_n_tI've also been noticing more and more how weird it is that the backlight flicks on just before shutdown. Could this be the culprit?
12:10:13S_a_i_n_ts/the/a/
12:10:35Tornethat's weird?
12:10:42Tornethe shutdown event is an event.
12:11:00Torneit causes the things which are detecting idleness to think idleness is over, because there was an event :)
12:11:03Torneso yes, the backlight comes on
12:11:18S_a_i_n_tI know...it does seem weirs to me that it bothers to switch the backlight on just to switch it off immediately.
12:11:48Tornebut it might be about to display stuff on the screen
12:11:55S_a_i_n_tI know *why* it does it..I still think its weird though ;P
12:12:31 Join einhirn [0] (~Miranda@bsod.rz.tu-clausthal.de)
12:12:33Tornethe various things called in clean_shutdown() might indeed do splashes or other output
12:13:10Tornewhich player are you using?
12:13:39S_a_i_n_tNano 1 g at the moent.
12:13:47Torneright. that's one of the players we don' thave lcd shutdown support
12:14:04S_a_i_n_tBut, I use a 2g often as well...same case I guess?
12:14:07Torneno
12:14:52Tornenano 2g has lcd shutdown.
12:14:57Torneso, we call that on poweroff_hw()
12:15:02Torneso it should entirely disable the lcd
12:15:06 Join watto [0] (~watto@193.203.81.165)
12:15:43Torneon the nano 1g we clear the screen in power_off() though
12:15:58Torneso in fact on nano 1g it shouldn't leave anything on the screen even if it *doesn't* shut down cleanly
12:16:06Tornesince that gets run anyway
12:16:10Torne:)
12:16:40S_a_i_n_tWell, theres definitly some "residue" left on the screen to fade slowly on some shutdowns...
12:16:51TorneYeah, that shouldn't happen
12:17:03S_a_i_n_tHmmmmm :/
12:17:08Tornethere is explicitly code there to clear the screen, update the lcd, and even sleep for a bit to give it time ot do it
12:17:21Torneand it's in power_off() in target/arm/ipod/power-ipod.c
12:17:40Tornewhich is the only way to turn the power off :)
12:20:44wodzany idea what maybe the reason for battery graph not updating (in debug menu) while if You scroll screen down battery reading is changing?
12:23:30 Quit xavieran (Read error: Connection reset by peer)
12:24:32 Join xavieran [0] (~xavieran@dsl-220-235-119-86.sa.westnet.com.au)
12:44:29 Join mikroflops [0] (~yogurt@90-227-45-110-no112.tbcn.telia.com)
12:45:15 Quit shai (Quit: Leaving)
12:45:26 Quit lennyk (Quit: lennyk)
12:47:57 Quit mikroflops_ (Ping timeout: 264 seconds)
12:50:10 Join kramer3d [0] (~kramer@unaffiliated/kramer3d)
12:50:56 Quit halmi (Quit: halmi)
12:54:21 Nick evilnick__ is now known as evilnick (~evilnick@ool-457bccf5.dyn.optonline.net)
12:54:32 Quit evilnick (Changing host)
12:54:32 Join evilnick [0] (~evilnick@rockbox/staff/evilnick)
12:57:27 Join UnhelpfulWeb [0] (~a49c99e6@giant.haxx.se)
12:57:32 Join shai [0] (~Shai@l192-117-110-233.cable.actcom.net.il)
12:57:56UnhelpfulWebpetur: ugh. stupid winmo phone wasn't charging. :/
12:58:24UnhelpfulWebanyway my plan for now is probably a 4-bay NAS and 2x2TB drives, and raid1
12:58:44UnhelpfulWeblater it can move to 4 drives and raid5, or btrfs if that's mature by that point
13:00
13:00:18 Quit UnhelpfulWeb (Client Quit)
13:01:38 Quit JohannesSM64 (Quit: WeeChat 0.3.2-dev)
13:04:22***Saving seen data "./dancer.seen"
13:05:59 Join mikroflops_ [0] (~yogurt@90-227-45-110-no112.tbcn.telia.com)
13:06:42 Join jordan` [0] (~jordan@78.235.252.137)
13:07:48 Join mikroflo1s_ [0] (~yogurt@90-227-45-110-no112.tbcn.telia.com)
13:08:51 Join mikroflo2s_ [0] (~yogurt@90-227-45-110-no112.tbcn.telia.com)
13:08:51***Alert Mode level 1
13:08:51DBUGEnqueued KICK mikroflops
13:08:51DBUGEnqueued KICK mikroflops_
13:08:51***Alert Mode level 2
13:08:51DBUGEnqueued KICK mikroflo1s_
13:08:51DBUGEnqueued KICK mikroflo2s_
13:08:51***Alert Mode level 3
13:10:09 Quit mikroflops (Ping timeout: 258 seconds)
13:11:55 Quit mikroflops_ (Ping timeout: 260 seconds)
13:12:10 Quit mikroflo1s_ (Ping timeout: 246 seconds)
13:17:51 Part LinusN
13:18:52***Alert Mode OFF
13:21:27 Join LinusN [0] (~linus@rockbox/developer/LinusN)
13:21:58CIA-5New commit by Blue_Dude (r25535): Rename lang strings to something more appropriate
13:37:14peturUnhelpful: you can try to blame your winmo phone for writing this in the wrong channel ;)
13:40:44CIA-5New commit by pamaury (r25536): rbutil: update french translation.
13:45:49 Join kugel [0] (~kugel@rockbox/developer/kugel)
13:51:29 Join dfkt [0] (dfkt@unaffiliated/dfkt)
14:00
14:13:51 Quit mitk (Quit: Leaving)
14:14:04 Join mikroflops [0] (~yogurt@90-227-45-110-no112.tbcn.telia.com)
14:16:34 Part jacekowski
14:18:11 Quit mikroflo2s_ (Ping timeout: 260 seconds)
14:24:07 Join Schmogel [0] (~Miranda@p3EE21B65.dip0.t-ipconnect.de)
14:26:01 Join emrecelikten [0] (~c18cf902@gateway/web/freenode/x-oxzgkdisdelhzxjt)
14:35:51 Join anewuser [0] (anewuser@unaffiliated/anewuser)
14:39:53 Join bmbl [0] (~Miranda@unaffiliated/bmbl)
14:47:19 Join froggymana [0] (~187b533e@giant.haxx.se)
14:49:28 Part LinusN
14:52:47 Quit froggymana (Quit: CGI:IRC (Ping timeout))
14:53:20 Quit kramer3d (Ping timeout: 264 seconds)
14:56:35 Join froggymana [0] (~187b533e@giant.haxx.se)
14:57:13 Join panni_ [0] (hannes@ip-95-222-52-93.unitymediagroup.de)
14:59:22 Join maraz [0] (maraz@kapsi.fi)
15:00
15:04:23***Saving seen data "./dancer.seen"
15:07:00 Quit wodz (Read error: Connection reset by peer)
15:09:45 Quit antil33t (Read error: Connection reset by peer)
15:09:50 Join antil33t [0] (~Mudkips@203-184-54-232.callplus.net.nz)
15:10:56 Join wodz [0] (~wodz@skatol.ch.pw.edu.pl)
15:11:25wodzwhat ata_enable() is supposed to do?
15:12:50 Quit anewuser (Quit: http://xrl.us/Renoise Like renoise + like music? 3 days to submit your entry!)
15:14:39Tornewodz: enable the ATA controller
15:15:15wodzata controller of the soc or of the usb bridge?
15:15:26Tornethe processor
15:15:35wodzI see
15:15:46Torneit doesn't do anything for quite a few targets, though
15:15:55Torne:q
15:15:57Torneoops
15:16:22 Join Blue_Dude [0] (~chatzilla@adsl-235-206-131.mco.bellsouth.net)
15:17:00Torneit's probably not necessary
15:17:11Blue_DudeYay! I got my very first data abort on the e200 target. Progress!
15:18:51Blue_DudeTrying to reproduce. If I can, then hopefully I can track down the problem.
15:19:05 Join jgarvey [0] (~jgarvey@cpe-065-190-066-089.nc.res.rr.com)
15:21:43Blue_DudeReproduceable. Working on it.
15:24:32wodzHa! rockbox enters USB bridge mode on my MPIO HD200 :-)))
15:25:27wodzand leaves it without a crash :-))
15:39:23 Quit Zagor (Quit: Leaving)
15:40:41 Join robin0800 [0] (~robin0800@cpc2-brig8-0-0-cust964.brig.cable.ntl.com)
15:42:40 Quit rvvs89 (Remote host closed the connection)
15:43:09 Join adnyxo [0] (~aaron@adsl-065-013-002-216.sip.asm.bellsouth.net)
15:46:30 Join rvvs89 [0] (ivo@robotnik.ucc.gu.uwa.edu.au)
15:49:34 Quit froggymana (Quit: time for classes)
15:53:54 Quit rvvs89 (Changing host)
15:53:54 Join rvvs89 [0] (ivo@pdpc/supporter/base/rvvs89)
15:58:23 Quit emrecelikten (Ping timeout: 248 seconds)
16:00
16:01:28 Join xiainx [0] (xiainx@wpa062090.Wireless.McGill.CA)
16:02:26Blue_DudeOK, I have a line that says: switch (do_menu(menu, NULL, NULL, false))
16:02:58Blue_DudeAnd the very first line after the switch clears a flag. But if you exit the menu without doing anything, the flag doesn't hit. It's not even in a case. Why is that?
16:03:27Blue_DudeShouldn't the first line always run, before the case hits?
16:04:17Blue_DudeOr does the entire procedure after the switch get skipped under certain conditions?
16:06:41gevaertsthe first line within the switch block?
16:06:42 Quit kugel (Ping timeout: 276 seconds)
16:06:56Blue_DudeYes. It doesn't always get run it seems.
16:07:05 Join emrecelikten [0] (~c18cf902@gateway/web/freenode/x-tmlinwanxhgoxxwy)
16:07:46JdGordon_pastebin it
16:08:17JdGordon_wait, you have code between the switch and the first case? is that even legal?
16:08:38gevaertsit's legal, yes. It won't get run though :)
16:09:13gevaertsunless you do silly tricks like Duff's Device
16:09:13 Join Zagor [0] (~bjst@rockbox/developer/Zagor)
16:09:26Blue_Dudehttp://pastie.org/909339
16:09:48gevaertsright. That won't ever work
16:09:49Blue_DudeC can be weird. I didn't know that.
16:10:01Blue_DudeBut, sometimes it's run...
16:10:21TorneThen the compiler is broken
16:10:25Torneor you are mistaken :)
16:10:35TorneThis is specifically mentioned in the spec
16:10:54Blue_DudeSo just put the line in each case and it should work?
16:11:22Tornethat's very silly
16:11:32JdGordon_store the ret val from do_menu
16:11:36Tornedo int selection = do_menu()
16:11:37JdGordon_then do that, then switch
16:11:40Torneer yeah
16:11:53Blue_DudeOh, yeah OK.
16:12:22TorneThe only "valid" use of having code before the first case is variable declarations for the scope of the switch
16:12:31Tornebut even then their initialisers won't get run if they have any ;)
16:12:35CtcpIgnored 4 channel CTCP requests in 0 seconds at the last flood
16:12:35*Blue_Dude takes out the Obvious Hammer and hits self with it.
16:13:01rvvs89Torne: Not even the initialisers? Huh, didn't know that.
16:13:19Tornervvs89: there's a specific example in the C spec, amusingly
16:13:39Tornewhich, paraphrased, says "this is legal but doesn't ever do anything you would want or expect"
16:13:43Torne"suck it programmers"
16:13:50Torne:)
16:14:00 Quit adnyxo (Remote host closed the connection)
16:14:07rvvs89Welp, it says that all over the place in the C spec.
16:14:27rvvs89Undefined behaviour is a pretty awesome thing.
16:14:33TorneNo, that's different
16:14:42Tornethis is specifically defined as "there cannot ever be a reason to put code here"
16:14:46Tornebut you are still allowed to put code there
16:14:53Torneeven though it is 100% guaranteed to never run ever :)
16:14:55JdGordon_thats retarted!
16:15:18Torneyup
16:15:19JdGordon_what the heck could the rational be for allowing it but saying it would never execute?
16:15:33gevaertsWell, you're also allowed to do if(0){some_code();}...
16:15:36Tornei presume because specifically allowing only variable declarations is too difficult
16:15:48Tornea sufficiently smart compiler would warn you that the code is unreachable
16:15:51JdGordon_gevaerts: thats obviously different though
16:15:52gevaertsThat's also perfectly legal. Should they make that illegal too?
16:16:02Tornegevaerts: a sufficiently smart compiler warns you about that :)
16:16:04gevaertsJdGordon_: well, in my view it's obviously the same!
16:16:11Tornei'm somewhat disappointed that gcc *doesn't* warn for Blue_Dude's code
16:16:14JdGordon_course you'd see it that way :p
16:16:20gevaertsTorne: well yes, but that's another issue :)
16:16:26Tornetrue.
16:16:27rvvs89Pretty much, if(0) sounds like something you should be using the preprocessor for.
16:16:29JdGordon_one is the user saying its not getting executed, the other is the language spec!
16:16:59gevaertsHow about "goto mylabel;some_code();mylabel:"? Is that legal? :)
16:17:21gevaertsJdGordon_: you mean the language spec doesn't sway what if() should do?
16:17:48rvvs89Torne: If a user places code before the first switch label then clearly they don't want that code to be executed either.
16:17:54rvvs89Err, JdGordon_
16:18:11Tornervvs89: generally compilers assume that any code you wrote, you want to be executed at least *sometime*
16:18:18Tornewhich is why they tend to warn on unreachable code
16:18:31Tornethis should just be another such case, really
16:18:35Tornemaybe gcc doesn't notice, though.
16:18:54rvvs89Not enough warnings enabled :P
16:20:06Blue_DudeHa! That did it. Thanks!
16:20:12Blue_DudeNo more data abort.
16:21:17Blue_DudeS_a_i_n_t will be thrilled. ;)
16:21:35JdGordon_rvvs89: well yes, but most people would assume that the code there would be executed, whereas a blind man would know that code inside if(0) wont!
16:22:13gevaertsWell, code between the start of the switch block and the first case label *can* be executed, if you do the right things
16:22:15*JdGordon_ looks for sdoyan to prove his point
16:22:18rvvs89I figured initialisers would be run, but not any other code.
16:22:35*Blue_Dude needs to go back and review C For Dummies. Or maybe C Traps and Pitfalls.
16:23:25Blue_Dudegevaerts: Under what conditions would that code be executed. It had to have been sometimes or the abort would have shown up sooner.
16:23:55TorneBlue_Dude: no, really, never. unless there's a bug in the compiler.
16:24:05gevaertsBlue_Dude: http://pastie.org/909373
16:24:07TorneBlue_Dude: the C spec *specifically states* that code before the first case: is unreachable
16:24:32 Quit emrecelikten (Quit: Page closed)
16:24:35TorneOh yeah, that ;)
16:24:54Torneok, so actually it's "when you use duff's device"
16:25:00rvvs89You could have a regular label before your case label(s) too.
16:25:03Blue_DudeTorne: I believe you. It's just that sometimes it really did run. Which is why I didn't get the problem sooner.
16:25:07rvvs89Which I suppose could also be a duff's device.
16:25:11wodzPlease take a look at #11137. It is more or less in a shape I think is sufficient to commit.
16:25:14Tornervvs89: yeah, that's just duff's device again
16:25:41TorneBlue_Dude: well if you can reproduce that, raise a gcc bug, seriously
16:26:01 Join adnyxo [0] (~aaron@adsl-065-013-002-216.sip.asm.bellsouth.net)
16:26:26 Quit wodz (Quit: Leaving)
16:27:47JdGordon_gevaerts: does that run?
16:28:08Blue_DudeWhile we're on the subject (sort of): should the hotkey set key be the context menu key or the hotkey?
16:28:15TorneJdGordon_: yeah, it will do.
16:28:23TorneJdGordon_: duff's device is fun
16:28:41Tornethe cases of a switch can be *anywhere*, including inside a sub-block. even if that block is a loop.
16:28:49Torneas long as the sub-blocks are contained entirely within the switch
16:29:14Torneonce the switch jumps to the right case the code will execute as if the switch/case wasn't there, i.e. the loop will proceed as normal
16:29:21Torneisn't C fun
16:30:24rvvs89Sure is.
16:33:39*JdGordon_ head explodes
16:34:16CIA-5New commit by Blue_Dude (r25537): Fix data abort due to malformed switch code. My bad.
16:35:48Blue_DudeJdGordon: turns out there's no conflict between using the same key to set the hotkey vs the quickscreen set menu. They should never hit on the same menu item.
16:37:55 Nick YPSY is now known as Ypsy (~ypsy@geekpadawan.de)
16:40:48 Join mitk [0] (~mitk@chello089078013146.chello.pl)
16:43:48Blue_DudeTorne: now, as per usual Murphy procedure, I can't reproduce the bug. Ah well.
16:44:13S_a_i_n_t;) Woooo! Hotkeys!
16:44:17*S_a_i_n_t compiles
16:44:33Blue_DudeMaybe it never did execute and I just never hit the right condition before.
16:45:31 Join Boldfilter [0] (~Boldfilte@adsl-71-217-14.jax.bellsouth.net)
16:45:32 Quit lpereira (Ping timeout: 264 seconds)
16:46:05 Join halmi [0] (~Miranda@93-82-34-72.adsl.highway.telekom.at)
16:46:06 Quit r2k000 (Quit: r2k000)
16:47:06 Quit halmi (Client Quit)
16:48:36 Quit antil33t (Read error: Connection reset by peer)
16:48:43 Join antil33t [0] (~Mudkips@203-184-54-232.callplus.net.nz)
16:49:49 Join halmi [0] (~Miranda@93-82-34-72.adsl.highway.telekom.at)
16:50:18S_a_i_n_tC can be a cruel mistress...
16:50:38Blue_DudeCheap lay sometimes.
16:51:21S_a_i_n_t...but, she said she loved me! ;)
16:52:08Blue_DudeDidn't get tested. Doh!
16:57:22 Quit halmi (Quit: halmi)
17:00
17:04:25***Saving seen data "./dancer.seen"
17:05:27 Join lpereira [0] (~lucien@did75-8-82-226-27-213.fbx.proxad.net)
17:06:47 Join liar [0] (~liar@clnet-p09-185.ikbnet.co.at)
17:09:43TorneDoes anyone have an ipod video who can test this build for me: http://forums.rockbox.org/index.php?topic=24439.0
17:09:57TorneUpgraded/replaced drives, or original drives, both are useful
17:10:21gevaertsTorne: I can test it on a 30G 5G later today
17:10:30Tornecool, that'd be handy
17:11:07 Join halmi [0] (~Miranda@93-82-34-72.adsl.highway.telekom.at)
17:13:29 Join Xerion [0] (~xerion@cp209907-b.landg1.lb.home.nl)
17:15:24Blue_DudeOK, ready to commit the hotkey set key to the context menu key. Any objections?
17:16:10S_a_i_n_tYes, I'll need to compile again!
17:16:11S_a_i_n_t;P
17:16:12JdGordon_yes
17:16:18JdGordon_i dont tihnk it should work that way
17:16:20 Join evilnick_B [0] (~0c140464@gateway/web/freenode/x-itffmoyxwigdjsmu)
17:16:32JdGordon_unless it brings up a menu
17:16:34Blue_DudeIt would work the same way as the quickscreen set menu...
17:17:09JdGordon_so a menu?
17:17:23 Join efyx [0] (~efyx@lap34-1-82-225-185-146.fbx.proxad.net)
17:17:25Blue_DudeA confirmation dialog.
17:17:45JdGordon_screenshots?
17:17:49Blue_DudeNo need for a menu.
17:18:03Blue_DudeSame confirmation dialog as now. A yesno screen.
17:18:22JdGordon_the qs settings dont show a confirmation
17:18:36evilnick_Bhalmi: Would you like to ask about GSoC?
17:19:20halmiYes, I would like to get some more information about your requirements...
17:19:40linuxstbhalmi: What, specifically?
17:19:45Blue_DudeNo, a menu pops up for that. No menu needed here, just a yesno. And there's no conflict with the quickscreen, since the quickscreen only works on value setting items.
17:20:17halmiIf I could participate, I would like to get Rockbox running under Android.
17:20:17 Join stoffel [0] (~quassel@p57B4A6C3.dip.t-dialin.net)
17:20:21JdGordon_and how do you clear/reset it?
17:21:00Blue_DudeThere's a settings menu for that. Under general settings->hotkey. No change there.
17:21:13JdGordon_this all sounds rather messy
17:21:24 Join TheSphinX^ [0] (~cold@p54A5F47B.dip.t-dialin.net)
17:21:25JdGordon_cant it just have a proper listing of the options?
17:21:40linuxstbhalmi: The general view (I think) on the Rockbox-as-an-App (RaaA) project is that a very large part of the work is refactoring the existing Rockbox code to enable Rockbox to be cleanly compiled as an application for various different systems. The actual port to something like Android will be a relatively minor part of the work needed.
17:21:49Blue_DudeI could have another dropdown menu in there and drop the whole keyset idea.
17:22:07Blue_DudeBut the keyset already works, so...
17:22:09linuxstbhalmi: So you need to be prepared to (very quickly) learn about how Rockbox currently works in quite a lot of detail.
17:22:13 Quit TheSphinX^ (Client Quit)
17:22:27JdGordon_I think that is better, and then it could be on the QS which would give it the speed people might want
17:22:33 Quit lpereira (Quit: Leaving.)
17:22:46halmilinuxstb: Okay, thats an important fact.
17:23:38Blue_DudeThat would mean dedicating a QS button to a hotkey set menu.
17:23:46JdGordon_a case could be made to just get the current SDL sim working with no mods on android
17:24:07Tornegevaerts: so far my evidence is "it works on my 1024-byte sector 80gb 5.5G"
17:24:10JdGordon_Blue_Dude: the QS options are customisable, if the user wants it they can put it there
17:24:15S_a_i_n_tfrom defalut...there is a QS item free, no?
17:24:23Tornegevaerts: logically it should work on 512-byte sector drives in a trivial way :)
17:25:04Blue_DudeI didn't mean set a QS button there by default. I just meant that a QS button would have to be used up to get there quickly.
17:25:14halmilinuxstb: So the primary goal is to get rid of simulating a Rockbox device using SDL and let it run as a standalone Application.
17:25:19JdGordon_only to change the hotkey
17:25:20 Join TheSphinX^ [0] (~cold@p54A5F47B.dip.t-dialin.net)
17:25:35*JdGordon_ wouldnt mind changing the default from "show files" to "set hotkey action"
17:25:39Tornehalmi: not get rid of, we still need the simulator
17:25:39 Quit Zagor (Quit: Leaving)
17:25:57linuxstbhalmi: Yes. Although there are of course a lot of similarities between the two.
17:26:11linuxstb(and as Torne said, the simulator has different uses, so needs to stay)
17:26:21S_a_i_n_tbeither would I actually, but then...I liked the context menu set idea also :S
17:26:28S_a_i_n_t*neither
17:27:13linuxstbhalmi: Have you used Rockbox in the past?
17:27:48halmilinuxstb: I have used Rockbox since my iriver H320... I think it was summer 2005
17:27:52Blue_DudeThe hotkey setting itself is context sensitive. You would need different set menus for the File Browser and the WPS. So making a QS doesn't make as much sense.
17:28:55Blue_DudeUnless you want to make the QS context sensitive too. :p
17:29:17halmilinuxstb: Then I followed the Development for the Sansa AMS devices and since that port is usable, I have it on my Fuze.
17:29:43JdGordon_Blue_Dude: no, but I would guess that the file browser hotkey action would be more prone to changing than the wps one
17:30:44 Quit JdGordon_ (Quit: Page closed)
17:32:04Blue_DudeOK, I think maybe we ought to table the idea for a few days, until there's more experience with the interface. I don't want to add menus (and lang settings, etc.) unless they're going to be permanent.
17:32:29linuxstbhalmi: Do you have any more questions? Are you interested in any of the other ideas? (I'm trying to sell the video playback project...)
17:32:50Blue_DudeSo you'd rather leave the set key on the hotkey vs. the context key? That seems to be the main issue at the moment.
17:33:52S_a_i_n_tTBH... S_a_i_n_t doesn;t care how its set...as long as it *can* be set, and he knows how to set it.
17:34:12Blue_DudeSo, did it work on your DAP?
17:34:22*Blue_Dude is dying to know.
17:34:48halmilinuxstb: I have some more questions. Sorry, that I'm so slow, but I try to write proper English and that takes some time.
17:34:52S_a_i_n_tHaven't tried it yet, my dock conenctor is it home.
17:34:57S_a_i_n_t...and I'm not.
17:35:05gevaertshalmi: there has been some discussion here about RaaA, I'll try to dig up some useful links to the logs when I get home
17:35:25Blue_Dudeok. Let me know.
17:35:26 Join emrecelikten [0] (~c18cf902@gateway/web/freenode/x-nwglidcjrdibwwuo)
17:35:26halmigevaerts: Okay, that would be great. Thank you.
17:35:29 Join b0hoon [0] (~quassel@xdsl-7253.bielsko.dialog.net.pl)
17:35:47GodEatermostly that discussion has been contentious posturing about the best way to implement Raaa :)
17:36:29b0hoonBlue_Dude: after your last commits it works ok on the vibe 500
17:36:49halmiGodEater: Has it come to an conclusion about the best way?
17:36:50Blue_DudeWhew. Got that going at least. Thanks.
17:36:58GodEaterhalmi: not yet :)
17:37:30 Quit Boldfilter (Quit: Boldfilter)
17:37:40halmilinuxstb: Video playback sounds also very interesting, but there it comes to my problem...
17:37:55b0hoonBlue_Dude: i can test it on the ipod 5.5g video too.
17:37:58Blue_DudeAnd in other news: Anybody want pitch and speed info stored in bookmarks?
17:38:11Blue_Dudeb0hoon: Thanks, that would be great.
17:39:43halmilinuxstb: My programming-experience is not the best. I am quite good at php, and have some knowledge in java, python and perl. But I kind of lack knowledge in c.
17:39:53 Quit xiainx (Ping timeout: 264 seconds)
17:40:42Blue_Dudehalmi: Come on in, the water's fine! You know what they say about how to learn to swim...
17:41:11halmiI have read quite much c-code (especially when I followed the Sansa AMS development) and I usually understand what the code does and so. But I have never really written c.
17:41:57linuxstbhalmi: That in itself isn't a problem. But you may be competing against people with a lot of C experience...
17:43:45 Quit b0hoon (Remote host closed the connection)
17:43:53linuxstbhalmi: In your application, I think you would need to convince us that you can learn C fairly quickly.
17:44:39 Join b0hoon [0] (~quassel@xdsl-7253.bielsko.dialog.net.pl)
17:46:01 Part b0hoon
17:46:47halmilinuxstb: That is the thing I've been thinking about. In the past, I usually learned programming languages quite fast, and I think that it would not be a big problem to learn c.
17:47:34halmiBut it is clear to me, that someone who is confident with c will be able to work more efficiently (especially in the beginning)
17:51:36halmilinuxstb: The video playback also sounds interesting.
17:52:14gevaertshalmi: it's supposed to *look* interesting! Things that merely sound interesting already work fine in rockbox ;)
17:53:08emreceliktenExcept AAC decoding, I guess? :)
17:53:13halmiOkay, thats a point... Even if ac3 and aac would rather sound...
18:00
18:02:17emreceliktenAre we able to edit our proposal after the deadline?
18:03:13 Quit pamaury (Quit: Quitte)
18:03:25gevaertsIn previous years students could edit proposals after mentors had added public comments (questions, usually). To make sure, ask in #gsoc
18:04:17halmiI just found a piece of discussion about RaaA in the mail archive. Starting at: http://www.rockbox.org/mail/archive/rockbox-dev-archive-2009-03/0363.shtml
18:04:49linuxstbhalmi: It's been discussed within the last week or so in this channel. See the logs http://www.rockbox.org/irc/
18:05:09halmiOkay, I will look at that too.
18:05:23 Quit petur (Quit: *plop*)
18:07:32 Join xiainx [0] (~xiainx@modemcable195.238-202-24.mc.videotron.ca)
18:10:06 Join komputes [0] (~komputes@ubuntu/member/komputes)
18:17:47 Quit Blue_Dude (Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100401080539])
18:18:05 Quit robin0800 (Remote host closed the connection)
18:18:06 Join fyre^OS [0] (~nnscript@static-71-249-251-152.nycmny.east.verizon.net)
18:35:34 Quit xiainx (Quit: Good Bye)
18:38:28emreceliktenWhat's the best way to profile the processor usage of functions in a library? I used gprof, but I seem to fail at getting good results
18:39:41 Join lennyk [0] (~lenny@adsl-75-5-198-147.dsl.snlo01.sbcglobal.net)
18:40:10 Quit adnyxo (Quit: Leaving)
18:40:33 Join xiainx [0] (~iain@modemcable195.238-202-24.mc.videotron.ca)
18:49:13 Join domonoky [0] (~Domonoky@rockbox/developer/domonoky)
18:54:12 Join CGL [0] (~CGL@190.79.151.211)
18:56:29 Join amr [0] (~quassel@41.234.26.97)
18:57:11 Join adnyxo [0] (~aaron@adsl-065-013-002-216.sip.asm.bellsouth.net)
19:00
19:02:49 Join kugel [0] (~kugel@rockbox/developer/kugel)
19:03:03 Join r2k000 [0] (~r2000@sjs-130-65-109-108.sjsu.edu)
19:04:29***Saving seen data "./dancer.seen"
19:05:44kugelweeh, there's a step-by-step how to to install android on the mini2440
19:07:20emreceliktenDone it, used callgrind (for anyone interested)
19:08:32 Quit stoffel (Remote host closed the connection)
19:09:44kugeltoo bad we killed supervivi!
19:18:06 Join Horscht [0] (~Horscht2@xbmc/user/horscht)
19:22:00gevaertshalmi: there's some discussion of RaaA at http://www.rockbox.org/irc/log-20100326#22:42:33 and http://www.rockbox.org/irc/log-20100406#10:41:47 (and around 17:00 on the same day again). By reading that you should be able to get some ideas about what's needed, what might be needed, and what people disagree on
19:25:50gevaertshalmi: also, if you're interested in both RaaA and video, you can always submit two proposals, although of course you only have (slightly less than) 24 hours to write them...
19:26:18halmiGreat, thank you.
19:27:51halmiI did search for "application" and "raaa" on rockbox.org/irc with google, but this two logs are probably not yet in the index, as I did not find them.
19:30:03gevaertshm, maybe
19:30:33 Quit mitk (Quit: Leaving)
19:31:46halmiAnd I know that I'm a bit late and time is running short. :)
19:32:29 Join Strife89 [0] (~michael@168.16.237.214)
19:38:27ranmakugel: FYI: Added two battery bench runs to http://www.rockbox.org/wiki/SansaRuntime
19:39:21ranma9h 16min and 9h 39min on my C200v2, 320kbps mp3 looped with display off
19:39:27 Join pamaury [0] (~c2c7a50a@gateway/web/freenode/x-xwubtqtspyygujbj)
19:39:40 Quit pamaury (Changing host)
19:39:40 Join pamaury [0] (~c2c7a50a@rockbox/developer/pamaury)
19:39:48kugelranma: not bad
19:40:26saratogaemrecelikten: have you tried an AAC-HE file? that will probably be more interesting to look at then AAC-LC
19:40:43domonokykugel: you could restore supervivi if you want. its on cd, and i think i have a backup of it.
19:40:58kugelyes I know :)
19:41:00emreceliktenWill do, currently trying to profile FFmpeg as well
19:41:09emreceliktenDownloaded and compiled them both
19:41:17emreceliktenUsed Nero encoder to create a sample file
19:41:24emreceliktenI will try both of them
19:41:38emreceliktenBut I am still not sure how to obtain a good profiling from callgrind
19:41:41kugeldomonoky: I think we should try to make rockbox boot via uboot so that people are able to boot rockbox/linux/whatever depending what's on the sd card
19:42:17domonokykugel: would be nice, can supervivi or uboot from nand do that ?
19:42:19emreceliktenI think I have got the number of instructions though
19:42:59ranmaInteresting: The first bench I ever made (before the FM fix) it did not turn off at 3.3V and proceeded until 1.85V battery voltage until it failed due to low voltage.
19:43:25ranmaBased on that I'd expect it to run about 1.6 times longer when the shutdown at 3.3V is disabled.
19:43:36kugeldomonoky: but who will do the work? :(
19:43:54domonokykugel: Mr.Someone :-)
19:44:09kugelranma: 1.85 sounds very dangerous for the battery
19:45:56ranmakugel: For comparison: The Neo Freerunner battery has it's hardware discharge cutoff at 2.5V
19:46:20kugelranma: how long was the runtime for 3.3V to 1.85V?
19:46:32emreceliktensaratoga: Do you have any recommendations for profiling?
19:46:39kugelstill, I'd expect the battery to take a hit with that low voltage
19:47:05saratogaemrecelikten: i've only proflied codecs once or twice, but IIRC I just added the compile flag and then ran the binary through gprof
19:47:30saratogathat seemed to work, provided the compiler didn't try to inline everything
19:47:39 Quit TheSphinX^ (Quit: XChat)
19:47:42emreceliktensaratoga: gprof does not seem to work because most of the calls are to libraries. So I resorted to using callgrind
19:48:07 Join dfkt_ [0] (dfkt@unaffiliated/dfkt)
19:48:16kugelmost profiling tools only work static libraries IIUC
19:48:35saratogaemrecelikten: but isn't the goal to profile libfaad itself?
19:48:45kugeldomonoky: I begin to dislike that lazy dude! :p
19:49:12ranmakugel: Hmm, wait. The voltage values in that very first run were quite wacky anyway, I just noticed it stated at 5V there for some reason...
19:49:42ranma(i.e. this one: http://forums.rockbox.org/index.php?topic=14064.msg162092#msg162092)
19:49:59emreceliktensaratoga: Well I have compiled everything with -pg flag and tried to profile 'lt-faad' executable, maybe I am doing something wrong
19:50:11emreceliktensaratoga: I have gotten good results with callgrind though
19:50:42saratogamaybe faad does something weird in its make system
19:51:02saratogaif callgrind works, just use that (never used it though so I don't know what the difference is)
19:51:11kugelranma: yea that one can go directly to the trash bin :p
19:51:40 Quit dfkt (Ping timeout: 245 seconds)
19:52:06kugelwhat happened to the guy that wanted (or did?) change the codecs to also build as stand-alone?
19:53:05ranmaAccording to http://batteryuniversity.com/partone-12.htm: "The equipment protects the battery by cutting off when the cell reaches 2.7 to 3.0V/cell"
19:53:09saratogahe posted a patch for it to the tracker somewhere, but it basically only worked on codecs that didn't use the codec library
19:53:16saratogasince it didn't include the codecs/lib folder
19:53:40saratogaprobably not too hard to fix if you're good with makefiles
19:55:03kugelfs#?
19:55:49saratogahttp://www.rockbox.org/tracker/task/10243?string=codec&project=1&search_name=&type[0]=4&sev[0]=&pri[0]=&due[0]=&reported[0]=&cat[0]=&status[0]=open&percent[0]=&opened=&dev=&closed=&duedatefrom=&duedateto=&changedfrom=&changedto=&openedfrom=&openedto=&closedfrom=&closedto=
19:56:00saratogaoh looks like it was updated, that version might work better
19:56:57saratogait'd be nice to just add an option to our configure script to build all the codecs as a separate binary
20:00
20:01:33emreceliktenDuh, so much to do, so little time. Wish I could work on this earlier
20:10:20gevaertsTorne: are you expecting possible USB speed differences?
20:12:08ranmakugel: ranma/c240v2_battery_bench_percent.png">http://uguu.de/~ranma/c240v2_battery_bench_percent.png
20:12:26ranmaCutoff looks alright after all :)
20:12:51 Join fml [0] (~chatzilla@port-83-236-234-85.static.qsc.de)
20:13:19fmlWhat is the remote hotkey for H100 (and other players)?
20:13:20kugellooks like it needs to be re-calibrated
20:13:51ranmaWell it does say 'TODO: simple uncalibrated curve, linear except for first 10%' in the source code after all :)
20:13:58kugeldoes the strong drop-off at the end indicate that the 3.3V shutoff is alright?
20:14:02 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org)
20:14:16ranmakugel: Yeah, it does indicate exactly that.
20:14:51kugelcan you upload the script you used for the plotting somewhere?
20:15:09 Quit flydutch (Quit: /* empty */)
20:15:40fmlIt seems remotes were left out in the hotkey patch
20:15:41ranmaIt's not even a script, I just typed the command into gnuplot after googling for 'gnuplot howto' since I always forget the gnuplot commands :)
20:15:58kugelso I do :P
20:16:18ranmaAnd I edited the battery bench data with vim by hand to make it palatable to gnuplot :)
20:16:26kugelwould be nice to have the command line written down somewhere on the wiki for reuse still
20:16:41ranmaYeah
20:17:01kugelso we need someone else to contribute the appropriate sed command line
20:17:32ranmaWell, let's see, should be easy enough :)
20:17:34kugelOR we fix battery bench do spit out something more useful, like a csv
20:17:37CIA-5New commit by alle (r25538): Add placeholder for the remote hotkey (aka repair WPS keymap table)
20:19:06 Nick dfkt_ is now known as dfkt (dfkt@unaffiliated/dfkt)
20:19:35CIA-5New commit by alle (r25539): Remove the double full stop
20:20:32ranmased -e 's@^[0-9]\+:[0-9]\+:[0-9]\+, *\([0-9]\+\), *[0-9]\+%, *[0-9]\+:[0-9]\+, *\([0-9]\+\),.*@\1 \2@'
20:20:51kugellooks insane enough to work :)
20:24:13 Part watto
20:26:35 Quit r2k000 (Quit: r2k000)
20:27:42 Join notlistening [0] (~tom@94-195-105-95.zone9.bethere.co.uk)
20:30:09notlisteningFlynDice, where are things at with the AMSv2 work?
20:31:05notlisteningI have seen recent commits but I am intrested specifically about the clip+
20:32:17 Quit bmbl (Quit: Bye!)
20:37:35ranmakugel: I've put it at the bottom of http://www.rockbox.org/wiki/SansaRuntime
20:43:09kugelranma: cool
20:43:23FlynDicenotlistening: I've pretty much thrown what I know out on the forums in several posts. Writing to SD is probably the biggest thing right now but a seemingly unrelated commit dealing with bookmarks has somehow impacted the uSD functionality.. go figure.
20:44:06emreceliktensaratoga: Hmm libfaad is pretty inefficient really. 11.500.000.000~ instructions against 7.500.000.000~, most of it is in DCT
20:44:19emrecelikten(7b is for FFmpeg)
20:44:25fmlblue_dude: the field lang_id in the struct hotkey_assignment is not needed since it can be extracted from the menu item
20:46:07ranmakugel: Anything against committing a change to powermgmt-as3525.c with this battery profile?
20:46:20kugelI don't think so
20:50:14 Nick fxb__ is now known as fxb (~felixbrun@h1252615.stratoserver.net)
20:50:32CIA-5New commit by ranma (r25540): Use a real battery discharge curve for calibration data
20:52:05 Join merbanan [0] (~banan@c-62-220-165-110.cust.bredband2.com)
20:59:49kugelranma: I'm assuming playback works ok on the c200v2 now?
21:00
21:00:33 Quit kugel (Remote host closed the connection)
21:01:29 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
21:03:37Tornegevaerts: i'm expecting general disk speed difference, maybe
21:03:43Tornegevaerts: usb is the easiest one for people to looka t
21:04:13 Join petur [0] (~peter@rockbox/developer/petur)
21:04:31***Saving seen data "./dancer.seen"
21:06:00gevaertsprobably, yes
21:06:51fmlPlease, anyone interested (and Blue_dude in particular) review FS #11191. I haven't try to compile it since I have no development env now.
21:07:38 Join Luca_S [0] (~5712526b@giant.haxx.se)
21:08:13 Quit emrecelikten (Quit: bye everybody)
21:08:28bertrikranma, you modified firmware/target/arm/as3525/powermgmt-as3525.c but it's no longer used, we might as well have deleted it
21:09:40 Nick fxb is now known as fxb__ (~felixbrun@h1252615.stratoserver.net)
21:12:24Luca_Sdamn... it seems like i've lost my sansa charging cable :(
21:15:43 Quit notlistening (Ping timeout: 260 seconds)
21:15:54 Quit killan (Quit: ( www.nnscript.com :: NoNameScript 4.22 :: www.esnation.com ))
21:16:09ranmabertrik: Oh, I see. I'll add the change to sansa-c200v2/powermgmt-c200v2.c.
21:16:16 Join killan [0] (~nnscript@c-38fd70d5.06-397-67626721.cust.bredbandsbolaget.se)
21:18:00bertrikyes please, thanks.
21:18:56bertrikas far as I'm concerned you can delete firmware/target/arm/as3525/powermgmt-as3525.c in the same commit, if you want
21:19:39 Join robin0800 [0] (~robin0800@cpc2-brig8-0-0-cust964.brig.cable.ntl.com)
21:22:15 Quit halmi (Ping timeout: 260 seconds)
21:24:15 Quit phanboy4 (Read error: Connection reset by peer)
21:28:30Luca_Sgroan.. I wish sandisk made its players with normal miniusb cables :(
21:30:04robin0800Luca_S: its like phone makers too taking ages to use proper usb connectors
21:30:34saratogathe clip series has normal USB
21:30:56 Part fml
21:31:04saratogaemrecelikten: I suspected replacing the libfaad DCT and the rest of the SBR filterbanks will make up much of the difference between the two
21:31:05CIA-5New commit by ranma (r25541): Added battery profile change to correct file, removed unused powermgmt-as3525.c
21:31:45saratogain theory I think it should be possible to form a DCT out of the various fourier transforms we already have in the codec lib, but I haven't looked into it yet
21:32:00 Quit ender` (Ping timeout: 240 seconds)
21:32:20saratogaand of course theres already DCT-like code in mp3 and mpc since they use very similar filterbanks as libfaad
21:35:31gevaertsTorne: would you be very upset if I commit something that improves USB speed today, thereby possibly messing up possible speed change reports?
21:36:14Tornegevaerts: naw, it's fine ;)
21:36:17pamaurygevaerts: what does you want to change ?
21:36:25Tornegevaerts: it's probably not going to make any difference to people :)
21:36:30Tornewhich was mostly what i wanted to confirm
21:37:03ranmaHmm, the "Player - Sim" build errors look unrelated?
21:37:24gevaertspamaury: I've done some thinking about fixing r24333, and it doesn't look too hard
21:37:33 Quit Luca_S (Quit: CGI:IRC)
21:37:44 Join Luca_S [0] (~5712526b@giant.haxx.se)
21:37:45gevaertsranma: yes, known build system glitch
21:37:54ranmaOk
21:38:24pamaurygevaerts: what was the exact problem ?
21:38:34 Quit merbanan (Ping timeout: 246 seconds)
21:39:24gevaertspamaury: the problem is that the way things were done (and are done again now in my tree) the CBW and new command transfers can complete in any order, while the code assumed that the CBW would complete first
21:39:54pamauryok
21:40:19gevaertsa few extra states fixes that
21:42:34gevaertspamaury: http://pastie.org/909945.txt is what I plan to commit in a few minutes
21:42:52gevaertsIt adds one to two megabytes per second (back) on my ipod video
21:43:12 Join merbanan [0] (~banan@c-62-220-165-110.cust.bredband2.com)
21:43:24gevaertsoh, and it reduces ram usage for some targets as well. I cleaned up the CBW allocation a bit
21:43:53 Join halmi8249 [0] (~Miranda@93-82-44-213.adsl.highway.telekom.at)
21:44:15gevaertss/CBW/CSW in my earlier explanation...
21:44:18pamaurythe change makes sense, I didn't look at precisely but at least it looks sensible
21:45:08gevaertsWell, it Works for Me (tm)!
21:45:53pamaurygevaerts: I currently have a problem with usb audio. If I put use usb strings, for some reason the linux kernel doesn't send me volume/mute update but if I don't use them, it does ! This could to be a problem with control transfers, any idea ?
21:46:26gevaertsthat's weird...
21:46:33gevaertsIt's not a memory overflow somewhere?
21:47:32pamauryI can't see how, everything looks fine, but who knows... And I can't make it works under Windows for now also
21:48:03CIA-5New commit by gevaerts (r25542): Make usb storage wait for new commands in parallel with sending out the CSW again, for a nice speed improvement. ...
21:48:15pamauryI currently suspect that it could be an iso transfer completion during a control transfer completion (control completion uses queue whereas iso completion calls from interrupt handler)
21:48:43gevaertsthat could be nasty :\
21:49:24pamauryYeah but it's on different endpoints and iso transfers don't prime anything (there is a 16 TD queue !) so i can't see how, I'm a bit puzzled
21:50:16pamauryAs always, it will end up being a super stupid bug that I left somewhere...
21:50:50 Quit Luca_S (Quit: CGI:IRC (Ping timeout))
21:51:05pamauryAnd I can't debug it easily because wireshark output craps for control transfers ! Perhaps they broke something recently
21:52:35 Quit GeekShadow (Quit: The cake is a lie !)
21:58:32 Join Boldfilter [0] (~Boldfilte@adsl-82-106-139.jax.bellsouth.net)
21:58:43 Quit Strife89 (Quit: Clocking out.)
22:00
22:00:18*ThomasAH gets "Undefined instruction at 30063360" when opening a .txt file on Clip+ with current SVN
22:00:30ThomasAH(it works in the simulator though)
22:01:03gevaertsThomasAH: the simulator isn't running on a RISC cpu though, so it has more defined instructions available
22:01:06 Join Luca_s [0] (~5711feea@giant.haxx.se)
22:01:42 Quit robin0800 (Remote host closed the connection)
22:02:38ThomasAHeven viewing rockbox-info.txt does not work, I'm pretty sure this worked with an earlier svn revision
22:03:01Luca_sdoes it work when forcing boosting?
22:03:36ThomasAHLuca_s: How do I do that? (pressing a key on boot?)
22:04:14Luca_ssystem > debug > cpu freq > change the boost counter to 1
22:05:29ThomasAHLuca_s: no, still undef. instruction
22:05:57ThomasAH(on internal memory and u(how do I compose micro?)SD)
22:06:28 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
22:06:42ThomasAHยต <- compose-m-u :)
22:08:22Luca_sjust tested on fuzev2, it works fine.it has to be something specific to the clip
22:09:11ThomasAHFlynDice: can you verify if viewing .txt files with clip+ on current svn is broken for you, too?
22:13:06ThomasAHviewing .jpg no longer works, too ... this definitely worked on April 1st and some days after that (smurfs :))
22:13:27ThomasAHbut no undefined instruction here, just "Plugin returned error"
22:15:15ThomasAHanyway, I have to go to bed now (and will check listening to .ogg there ... .ogg and .spx seemed to work during a quick test)
22:16:39 Quit Schmogel (Ping timeout: 245 seconds)
22:17:03 Quit halmi8249 (Ping timeout: 264 seconds)
22:19:00 Quit CGL (Remote host closed the connection)
22:20:06 Quit petur (Quit: *plop*)
22:21:34 Join CGL [0] (~CGL@190.79.151.211)
22:22:00 Quit GeekShadow (Quit: The cake is a lie !)
22:22:19 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
22:22:29 Join Rondom [0] (~quassel@dslb-084-057-207-091.pools.arcor-ip.net)
22:27:09 Quit GeekShadow (Client Quit)
22:27:14 Join halmi [0] (~Miranda@93-82-44-213.adsl.highway.telekom.at)
22:27:51 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
22:29:10 Quit adnyxo (Remote host closed the connection)
22:29:27 Quit linuxstb (Ping timeout: 260 seconds)
22:31:36 Join linuxstb [0] (~linuxstb@rockbox/developer/linuxstb)
22:36:30 Nick jfc^2 is now known as jfc (~john@dpc6682208002.direcpc.com)
22:36:58 Join kugel [0] (~kugel@rockbox/developer/kugel)
22:37:32 Quit linuxstb (Ping timeout: 265 seconds)
22:38:59 Join robin0800 [0] (~robin0800@cpc2-brig8-0-0-cust964.brig.cable.ntl.com)
22:45:03 Join Darkknight512 [0] (~Darkknigh@CPE00212968356c-CM00186845dd46.cpe.net.cable.rogers.com)
22:45:22 Quit Darkknight512 (Client Quit)
22:46:53 Join ender` [0] (krneki@foo.eternallybored.org)
22:48:18 Quit pamaury (Quit: Page closed)
22:48:59 Join linuxstb [0] (~linuxstb@rockbox/developer/linuxstb)
22:49:10 Quit halmi (Quit: halmi)
22:53:37 Quit linuxstb (Ping timeout: 240 seconds)
22:55:05 Quit merbanan (Ping timeout: 264 seconds)
22:56:58 Part domonoky
22:57:12 Join killan_ [0] (~nnscript@c-38fd70d5.06-397-67626721.cust.bredbandsbolaget.se)
22:58:17 Join panni__ [0] (hannes@ip-95-222-52-93.unitymediagroup.de)
22:59:55 Join sevard_ [0] (sev@216.164.6.24)
23:00
23:00:37 Quit panni_ (Read error: Connection reset by peer)
23:00:37 Quit killan (Read error: Connection reset by peer)
23:00:38 Quit sevard (Ping timeout: 264 seconds)
23:01:46 Join froggyman [0] (~me@unaffiliated/froggyman)
23:02:19 Part froggyman
23:04:32***Saving seen data "./dancer.seen"
23:06:58 Join linuxstb [0] (~linuxstb@rockbox/developer/linuxstb)
23:07:02 Quit fyre^OS (Read error: Connection reset by peer)
23:08:03CIA-5New commit by tomers (r25543): rbutil: Add RockboxUtility to svn:ignore
23:12:49 Quit linuxstb (Ping timeout: 265 seconds)
23:15:41 Nick Ypsy is now known as YPSY (~ypsy@geekpadawan.de)
23:16:30Luca_sthe vmware dev platform is broken? i get a bunch of linker errors when compiling (Normal mode, fuzev2). libgcc.a uses FPA instructions whereas rockbox.elf does not
23:18:39 Quit evilnick_B (Quit: Page closed)
23:20:20 Nick _jhMikeS_ is now known as jhMikeS (~jethead71@rockbox/developer/jhMikeS)
23:23:21S_a_i_n_tBlue_Dude: Well, hotkeys work now with the Nano1/2g!!!
23:23:40S_a_i_n_tThey're still *really* hard to set, but they work... ;)
23:23:53 Quit niekie (Quit: http://quassel-irc.org - Chat comfortably. Anywhere.)
23:24:53kugeltomers ??
23:27:04kugelI'm not sure I understand that commit
23:27:50 Quit GeekShadow (Quit: The cake is a lie !)
23:30:01 Quit efyx (Remote host closed the connection)
23:35:41 Quit liar (Ping timeout: 258 seconds)
23:49:10 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
23:52:33 Quit CGL (Ping timeout: 248 seconds)
23:55:18 Quit Adubb (Read error: Connection reset by peer)
23:56:05CIA-5New commit by tomers (r25544): Text viewer: Handle font loading failures ...
23:56:18 Quit mc2739 (Ping timeout: 265 seconds)
23:58:10 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)

Previous day | Next day