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 2009-09-20

00:06:29 Join kugel [0] (n=kugel@rockbox/developer/kugel)
00:07:39 Part Maltin
00:31:46***Saving seen data "./dancer.seen"
00:44:35 Quit bluebrother ("leaving")
00:46:53 Join GeekShadow [0] (n=Antoine@reactos/tester/GeekShadow)
00:53:29 Join wincent_balin [0] (
00:57:23 Nick fxb__ is now known as fxb (
01:01:09 Quit esperegu (Read error: 60 (Operation timed out))
01:04:08 Quit GeekShado_ (Read error: 110 (Connection timed out))
01:05:24saratogaplaying with the sim some more, it seems like the codec will occasionally be reloaded during normal track changes (end of track, etc), and that these cause a lot of the crashes on the clip
01:07:32kugelmost crashes are during tracks?
01:08:46saratogaare they?
01:08:55saratogai think i've noticed both
01:12:21saratogai think the ones i see on the sim are nearly all during track load
01:22:08 Join inversions [0] (
01:23:10 Part toffe82
01:26:43 Quit robin0800 (Remote closed the connection)
01:28:36kugelsaratoga: from just listening to the bug, I never connected it to track change, I get them during tracks mostly yes
01:30:46 Quit bughunter2 ("Leaving.")
01:31:14CIA-43New commit by kugel (r22741): Fix the statusbar to redraw after changing colors in the color picker screen. ...
01:32:17kugelwe should branch for 3.4 in the next few hours, so that we get at least a few days for testing the RC builds (if we get some). I'll do it after I get up unless someone beats me to it
01:32:23 Join robin0800 [0] (
01:35:07saratogakugel: its possible the crash i'm pursuing is artificial due to the small buffer i'm using
01:35:34saratogaor at least that it occurs more regularly then on the clip
01:37:56CIA-43New commit by kugel (r22742): And also redraw the statusbar after resetting the colors.
01:38:10 Quit ender` (" Where python would say you can skin a cat with a knife, so adding other methods would just make the language hard to learn,")
01:39:57saratogahow do I save a playlist on the clip?
01:40:09saratogawhen i get to the keyboard screen i can't find a way to exit
01:42:03saratogaok i believe i've tried every key combination
01:42:46 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
01:42:50kugelsaratoga: very possible that there is none
01:43:11saratogawhat is BUTTON_REPEAT?
01:43:25saratogakeymaps says BUTTON_SELECT|BUTTON_REPEAT
01:43:47pixelmaa long button press
01:44:02pixelma(of select)
01:44:48CIA-43New commit by Domonoky (r22743): rbutil: mark e200v2 and fuze as unstable.
01:44:55pixelmawhat's bah about that?
01:44:57saratogakugel: i think you're right about this track change bug, while it seems to be a problem on the sim, i just flipped back and forth between wma, mp3 and aac about a hundred times without encountering it
01:45:07saratogaeach one of those should have forced a codec rebuffer
01:46:08kugelso you were on the wrong route?
01:47:03saratogawell its definately a bug i'm just not sure how much it really matters
01:47:39saratogai suppose the next thing to try is logf on the real clip and see what its doing
01:47:56saratogado you know if logf slows the system down a lot?
01:48:23saratogai'd imagine its pretty cheap on flash targets, but what i'd really like to do is print out essentially the entire call history of the playback thread
01:48:34saratogasince it takes so long to encounter a deadlock
01:49:33kugelit doesn't really slow things down
01:50:45kugelit doesn't hit the disk at all until you actually choose to dump it
01:51:51saratogait doesn't dump on spinup?
01:53:00kugelbut it decreases the audio buffer a bit more, I think logf takes some 10-20k binsize/ram
01:54:22saratogawell if i want to generate say a 100KB of log output a minute
01:54:40saratogawill it just overwrite stuff?
01:55:06Torneyes, it's jsut a ring buffer
01:55:58domonokyyou would need logf-over usb for that. but unfortunatly we still dont have usb on AMS :-/
01:58:27 Quit moos (Read error: 131 (Connection reset by peer))
01:58:50 Join moos [0] (i=mostafa@rockbox/staff/moos)
02:04:41CIA-43New commit by kugel (r22744): Redraw the list after exiting the context menu to get rid of various non-blocking splashes in it if custom ui vp is used.
02:04:57kugelpixelma: I fixed your color picker bug
02:05:30saratogatheres no way to have it dump to disk?
02:05:35kugeldomonoky: no problem since the problem triggers easily on all targets
02:05:59saratogaFS #9253 - logf file output patch.
02:06:05kugeljust decrease to audio buffer to a "clip'ish" level
02:07:04domonokykugel: then trying it on a target with usb-serial and logf-over-usb would probably help :-)
02:07:42domonokyand putting many logf calls into playback/buffering of course.
02:08:05 Quit DarkDefender ("Leaving")
02:09:20saratogafor a flash target i don't see why logf over usb is any better then logf to disk
02:09:46saratogathe clip rebuffers every other second anyway, can't hurt to also dump a couple KB of text to the disk at teh same time
02:09:50kugelthat's a useful idea actually, with logf over usb it should be possible to catch the exact moment it crashes with a call trace
02:10:17saratogawhat do you mean call trace? doesn't it just send text?
02:10:29kugelyes, but you have 2 screens :9
02:10:54kugelI imagine you would miss the moment until you get into the logf screen or being able to view the dump
02:11:39kugelsince it doesn't freeze, I'd think there are many more calls which pollute the logf after the actual playback crash there
02:12:02saratogawell my plan was to use that patch to dump it to disk continuously
02:12:27saratogatheres actually very little output when it deadlocks, just 2x a second or so a queue message
02:12:59domonokythere are so many patches in our tracker, we should make a clean-up week after the release.. :-)
02:13:19kugeldomonoky: I remember we planned that for 2-3 weeks after the devcon actually
02:13:25saratogathis is another Uchida patch
02:13:39saratogahes got ~55 open ones on the tracker
02:13:49saratogabut its impossible to communicate with him
02:14:21domonokythere are 2 of these "japanese sounding" people with many patches in the tracker. we should try to care a bit about those patches :-)
02:14:52kugelwho's the other one?
02:15:09saratogai think the other one got svn
02:15:10domonokyTeruaki Kawashima
02:15:23kugelsaratoga: has anyone looked at his WAV patches? They seem to add many (probably unpopular) codecs
02:15:26saratogayeah teru has svn
02:15:51saratogakugel: i skimmed them and they looked useful, just i couldn't find the time or interest to do anything with them
02:16:05saratogaand it didn't help that Uchida didn't return my emails
02:16:15domonokyif he has svn access, i wonder why there are still so many patches from him in the tracker.. :-)
02:16:38kugelI have a lot open tasks too
02:16:49saratogahes been slowly commiting them
02:16:56saratogateru has actually done a lot of clean up in the last couple months
02:17:02saratogamostly minor things that no one noticed though
02:17:14domonokyah, didnt notice :-)
02:17:37saratogai'm half tempted to have an email translated in japanese and send it to him
02:17:54saratogabut since he didn't even reply i'm not sure the email he gave is even any good
02:18:22domonokyabout the pcm patches: some codec guy should look overthem, and if their are sound, just commit them... codecs are rockbox core business :-)
02:18:45saratogabug linuxstb once he has the nano2g working
02:19:00saratogai honestly don't understand the codecs that well, just the insides of them
02:19:08saratogaor maybe mt if hes bored
02:19:50 Join TopyMobile [0] (
02:20:05kugelsaratoga: forget mt, they're not in a rm container :D
02:20:24saratogaspeaking of which we should really find him a clipv2 or something and send it to him
02:20:43saratogai think he was thinking of trying to work on his old M200
02:21:09kugelI haven't seen him lately, where's he gone?
02:21:23saratogahes been around
02:21:45saratogahe said he wanted to work on hardware next in spite of my gentle suggestion that wma pro still didn't work with rockbox
02:22:03*domonoky thinks if mt want a new player to play. the rockbox fund should get him one..
02:22:59saratogayeah we're getting some google money for him right?
02:23:15kugelon the other hand, he'll be given $2500 buy a whole bunch of targets :)
02:23:40saratogaapparenlty sandisk doesn't sell AMS devices in his country though
02:23:50domonokybut he deserves it for his super gsoc expirience. :-)
02:24:03kugel(that was for saratoga)
02:24:17domonokyone more arugemt to just send him one :-)
02:24:29saratogai think its pretty common, last I heard the new Zune wasn't even going to be sold in the EU, let alone Egypt
02:24:50kugelthe zune is special, that's USA only
02:25:20kugelbut I can imagine they don't sell the latest high tech toys in africa
02:26:31mooslots of fake ones unfortunatly...
02:26:50GodEaterZagor: (for the logs) I wanted qwebirc installed on a rockbox server because anything with "webchat" in the title is blocked by the RBS proxy server. So freenode's is no use to me.
02:27:05 Nick fxb is now known as fxb__ (
02:27:53domonokyGodEater: create a tunnel ? :-)
02:28:55saratogahow much is the clip+ in the EU? might make a nice present has it for ~€35 (2gig) or ~60€ for 8gig
02:31:09saratogathe 8GB is ripoff given the price of uSD
02:31:50***Saving seen data "./dancer.seen"
02:32:44 Quit gregzx ("ChatZilla 0.9.85 [Firefox 3.5.2/20090729225027]")
02:35:38pixelmait's more like 55€, not a great difference... - and by the way - all the offers state that the device will only available from October 10th. But I think you can find different prices in other countries of the EU and maybe other dates of availability
02:35:38 Quit moos (Read error: 131 (Connection reset by peer))
02:35:57 Join moos [0] (i=mostafa@rockbox/staff/moos)
02:36:21pixelmafor the Clip+ and
02:37:29 Quit gevaerts (Nick collision from services.)
02:37:41 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
02:38:55GodEaterdomonoky: it's more complicated to do that than you think
02:39:09GodEaterthe proxy there is very clever at spotting the traffic it's passing
02:54:52 Quit Lss (Read error: 104 (Connection reset by peer))
03:05:05 Quit dfkt ("-= SysReset 2.53=- Ph'nglui mglw'nafh Cthulhu R'lyeh wgah'nagl fhtagn.")
03:06:30 Quit kugel (Read error: 104 (Connection reset by peer))
03:12:59 Part wincent_balin ("Kopete 0.12.7 :")
03:13:29 Join BHSPitLappy [0] (n=BHSPitLa@unaffiliated/bhspitmonkey)
03:19:02 Part domonoky
03:34:26 Quit GeekShadow (Read error: 54 (Connection reset by peer))
03:37:13 Join moonscapex [0] (
03:38:30 Quit moonscapex (Client Quit)
03:44:00 Quit amiconn (Nick collision from services.)
03:44:03 Join amiconn_ [0] (i=quassel@rockbox/developer/amiconn)
03:44:10 Quit pixelma (Nick collision from services.)
03:44:11 Join moonscapex [0] (
03:44:13 Join pixelma_ [0] (i=quassel@rockbox/staff/pixelma)
03:44:23 Nick amiconn_ is now known as amiconn (i=quassel@rockbox/developer/amiconn)
03:44:32 Nick pixelma_ is now known as pixelma (i=quassel@rockbox/staff/pixelma)
03:45:37moonscapexHas anyone got fft.v5 to compile? It looks like a cool plugin, but I can't add it to my build.
03:50:44 Quit moonscapex (Read error: 104 (Connection reset by peer))
04:03:38 Quit BHSPitLappy (Read error: 110 (Connection timed out))
04:07:58 Quit TheSeven (Nick collision from services.)
04:08:14 Join The_Seven [0] (
04:08:23 Nick The_Seven is now known as TheSeven (
04:24:24 Quit bubsy ("I'll be back somewhere in time")
04:24:33 Join bubsy [0] (n=bubsy@
04:26:59 Quit bubsy (Client Quit)
04:27:10 Join bubsy [0] (n=bubsy@
04:28:19 Quit MethoS- (Remote closed the connection)
04:31:52***Saving seen data "./dancer.seen"
04:32:50 Join lennyk [0] (
04:32:53 Quit timc (Remote closed the connection)
04:37:57 Join BHSPitLappy [0] (n=BHSPitLa@unaffiliated/bhspitmonkey)
04:43:57 Join GreatBeaver [0] (
04:44:05GreatBeaverwhat is the preferred defragmentator for mp3 players? im using diskkeeper
04:44:07 Quit Rondom (Nick collision from services.)
04:44:18 Join Rondom [0] (
04:44:51LloreanGreatBeaver: Generally defragmenting shouldn't be necessary.
04:45:46GreatBeaveri was just wondering because diskeeper takes way too long
04:50:26GreatBeaverdiskeeper says i'll save 7% on three files if i defrag, and defrag seems like it will take at least 20 minutes
04:50:32GreatBeaver3 files out of 6000 hehe
04:55:08 Quit chandoo_ ("Leaving")
04:56:12 Quit GreatBeaver ("ChatZilla 0.9.85 [Firefox 3.5.3/20090824101458]")
04:56:22 Join FOAD_ [0] (
05:13:14 Quit FOAD (Read error: 110 (Connection timed out))
05:13:14 Nick FOAD_ is now known as FOAD (
05:33:34 Quit HBK (Read error: 113 (No route to host))
05:49:04 Join donnald [0] (
05:49:23 Nick donnald is now known as Matthew (
05:49:34Matthewis any one here
05:49:53 Nick Matthew is now known as Guest56791 (
05:50:16 Quit Guest56791 (Client Quit)
06:31:55***Saving seen data "./dancer.seen"
06:33:31 Join wincent_balin [0] (
06:33:31 Quit gartral ("Lost terminal")
06:33:43 Part wincent_balin ("Kopete 0.12.7 :")
06:55:40 Quit CaptainKwel (Remote closed the connection)
07:10:22 Join Horschti [0] (n=Horscht2@xbmc/user/horscht)
07:14:29 Quit robin0800 (Remote closed the connection)
07:14:40 Quit saratoga ("Page closed")
07:28:50 Quit Horscht (Read error: 110 (Connection timed out))
08:14:36 Join Rob2223 [0] (
08:23:30 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
08:31:56***Saving seen data "./dancer.seen"
08:32:49 Quit Rob2222 (Read error: 110 (Connection timed out))
09:19:20 Join stoffel [0] (
09:32:55 Join vitja [0] (n=vitja@
09:35:13 Join genia4 [0] (
09:44:24 Join flydutch [0] (
09:57:39 Join bluebrother [0] (n=dom@rockbox/developer/bluebrother)
10:10:58 Join jamie_ [0] (
10:18:52 Quit panni_ (Read error: 104 (Connection reset by peer))
10:28:20CIA-43New commit by bluebrother (r22745): Integrate mknkboot into beastpatcher. ...
10:31:58***Saving seen data "./dancer.seen"
10:34:58 Nick Horschti is now known as Horscht (n=Horscht2@xbmc/user/horscht)
10:35:55 Join GeekShadow [0] (n=Antoine@reactos/tester/GeekShadow)
10:37:31 Join DarkDefender [0] (
10:42:06 Quit BHSPitLappy (Remote closed the connection)
10:43:49 Join esperegu [0] (n=quassel@
10:49:01 Quit stoffel (Remote closed the connection)
10:54:56 Join stoffel [0] (n=quassel@
10:58:41 Join ender` [0] (
10:58:55 Quit stoffel (Remote closed the connection)
11:00:36 Quit fyrestorm (Read error: 104 (Connection reset by peer))
11:02:49 Join fyrestorm [0] (
11:04:00 Quit daurn (Remote closed the connection)
11:05:37*n1s wonders if anyone with an ipod video could reproduce fs#10610, not a lot of detail in that report. Well any 3G or newer ipod should do i guess, they all have the same rtc
11:12:46bluebrothern1s: an Ipod mini too?
11:13:25 Join funman [0] (n=fun@rockbox/developer/funman)
11:15:34n1sbluebrother: yes, it has the same rtc as the video
11:15:53bluebrotherwell, I can't see any problem on my mini with current svn
11:16:04n1si see no reason for this bug to hit only one player using that driver
11:16:10n1sbluebrother: thanks fro testing
11:16:37n1salso such an obvious thing should have been reported ages ago
11:16:44*bluebrother thinks we should consider closing such tasks immediately. If reporters don't take the time to follow our guidelines and put basic information in, why should we?
11:17:28 Quit shodanX_ ("leaving")
11:18:41rasherbluebrother: because they might have reported an actual bug that we do want fixed?
11:19:56bluebrotherrasher: I don't fear that we will loose bugs ... software always contains enough of them
11:22:09rasherThat doesn't mean they'll get reported
11:23:52*bluebrother doesn't want bugs to get fixed, he wants bugs to not be present :)
11:26:48*bluebrother wonders how hard it would be to have the task description field prefilled with some headlines
11:26:57 Join Grahack [0] (
11:28:24 Quit funman (Read error: 60 (Operation timed out))
11:39:22 Join stoffel [0] (
11:46:00 Join funman [0] (n=fun@rockbox/developer/funman)
11:51:25 Quit stoffel (Remote closed the connection)
11:51:37 Join Ubuntuxer [0] (
11:57:59 Quit Ubuntuxer (Remote closed the connection)
11:58:31 Quit TopyMobile (Remote closed the connection)
12:00:06 Join stoffel [0] (
12:01:07 Join gregzx [0] (
12:07:21 Join kugel [0] (i=kugel@rockbox/developer/kugel)
12:08:00kugelBagder: I have images now, for fuze and yh925
12:08:41 Quit funman ("leaving")
12:10:57kugelI assume the current build page is in SVN too?
12:11:29 Quit stoffel (Remote closed the connection)
12:14:37 Join stoffel [0] (
12:16:11kugelhow do I test my changes?
12:18:52 Quit kugel ("ChatZilla 0.9.85 [Firefox 3.5.3/20090824101458]")
12:19:03 Join kugel|win [0] (
12:19:25 Quit kugel|win (Client Quit)
12:19:35 Join kugel [0] (i=kugel|wi@rockbox/developer/kugel)
12:23:02 Join gapan [0] (
12:26:05 Quit inversions (Read error: 104 (Connection reset by peer))
12:26:26 Join inversions [0] (
12:30:28 Quit stoffel (Remote closed the connection)
12:31:06 Join stoffel [0] (
12:32:00***Saving seen data "./dancer.seen"
12:38:05 Join kyle6513 [0] (n=kyle6513@
12:40:22kugelBagder: heh, fcpp almost compiles on windows (using mingw)
12:42:56*kugel can't manage to make the page on windows
12:43:05 Join gtkspert [0] (
12:45:35 Quit esperegu (Read error: 104 (Connection reset by peer))
12:46:31 Nick liar|away is now known as liar (n=liar@
12:50:56 Quit stoffel (Remote closed the connection)
12:53:10 Join wincent [0] (
12:56:10 Quit gtkspert_ (Read error: 101 (Network is unreachable))
12:59:29 Join daurnimator [0] (i=daurnima@freenode/staff/daurnimator)
13:03:43 Quit moos ("Rockbox rules the DAP world")
13:10:49 Part kyle6513
13:15:54kugelwww/ doesn't seem to be up-to-date
13:23:34 Quit bluebrother (Read error: 113 (No route to host))
13:24:00 Join FOAD_ [0] (
13:33:10 Quit wincent (Read error: 110 (Connection timed out))
13:34:03 Join stoffel [0] (n=quassel@
13:35:31gapanhi! I have read a report that rockbox runs on a 160GB ipod classic. I thought they couldn't be supported. Was any work done on those?
13:36:54kugelyou read lies
13:36:58 Quit FOAD (Read error: 110 (Connection timed out))
13:36:59 Nick FOAD_ is now known as FOAD (
13:37:42gapanwell, I know that guy that reported that (online) and he's not inclined to lie
13:37:49AlexPHe is
13:38:10AlexPPerhaps the people that make Rockbox might know? Oh wait, that's us
13:38:21gapanI know :D
13:38:35gapanthe report is here:
13:38:39gapanunfortunately in greek :D
13:38:42AlexPlinux4nano have found an exploit that could potentially be used, but it is a long way from anything useable
13:38:48AlexPgapan: That report is wrong
13:38:55AlexPNo matter what language it is in
13:39:07gapanhe says he tried the ipod60/80 installation files
13:39:14gapan1st time he tried it failed
13:39:15AlexPWhich is rubbish
13:39:37AlexPHe is talking utter utter bollocks
13:40:45AlexPAs I say, linux4nano have found an exploit, but nobody has even started to do anything about it. Even if they eventually do, it will not use the files for any other ipod as the hardware is different.
13:41:53gapanhe says that the 2nd time he tried the installation started but failed with errors
13:42:02gapanand then when he rebooted it just worked!
13:42:06AlexPHe is lying
13:42:13gapanI'll convey that to him :D
13:42:19AlexPFor the last time, IT DOESN'T WORK, IT CANNOT WORK
13:42:30gapanyep, got it, thanks :)
13:42:34AlexPYes, perhaps you could post so that others don't get confused
13:43:44AlexPMaybe he is just confused and has a supported ipod with a bigger disk? Unlikely I guess
13:44:08gapanis there a supported ipod with a 160GB HD?
13:44:17gapanso no
13:44:28AlexPBut you *could* change the disk for a bigger one in one of the supported ipods
13:44:42AlexPWhich is what I meant
13:45:14gapanI don't think he did that
13:45:35AlexPSo we get back to him just not having done what he said he did :)
13:46:17 Quit Rondom (Nick collision from services.)
13:46:28 Join Rondom [0] (
13:49:46AlexPgapan: Sorry to have "shouted" at you, I was just being very clear :)
13:53:16gapanAlexP: no problem :)
13:55:55 Quit Rondom (Read error: 104 (Connection reset by peer))
13:56:41kugelare we going to have 3.4 for unstable devices?
13:57:26 Join Rondom [0] (
13:59:27AlexPinterestiung question
13:59:42 Nick fxb__ is now known as fxb (
13:59:46AlexPI *think* we said that we would just release stable, but support unstable
14:00:19AlexPTo my mind a release should be more "complete" - rbutil, manual, more of the features etc.
14:01:40 Join timc [0] (n=aoeu@
14:05:50kugelin my mind a release is just a stable'fied snapshot of the source, and there's no real reason to not have it running on other ports
14:06:22kugeleven users of unstable ports may which to run the most stable rockbox :)
14:06:57AlexPprobably less though as it is likely to be changing at a faster rate
14:07:41kugelwho's going to make the release zips?
14:07:45AlexPmaybe we could do something like have 3.4 - preliminary support for (insert unstable targets here)
14:13:14 Join robin0800 [0] (
14:15:49 Quit DarkDefender (Read error: 54 (Connection reset by peer))
14:21:29 Quit gregzx (Read error: 110 (Connection timed out))
14:22:28CIA-43New commit by kugel (r22746): Make Samsung YH* bootloaders non-verbose (i.e. show the bootlogo)
14:23:55 Quit Farthen (Remote closed the connection)
14:25:04CIA-43New commit by kugel (r22747): Make Samsung YH* bootloaders non-verbose (i.e. show the bootlogo) (backported from the ams/pp bootloader branch)
14:28:41CIA-43New commit by kugel (r22748): Branch for Rockbox 3.4
14:30:27 Join bipak [0] (
14:30:31 Quit lennyk ()
14:32:04***Saving seen data "./dancer.seen"
14:32:15bipaki have an ihp120 with a very old bootloader (2006). you could boot into the original firmware by pressing play on the remote control. unfortunately i can't find my remote at the moment. is there any other way to boot into it?
14:35:08pixelmathere is a description on how to do this in our manual, I don't know though if this also applies to the old bootloader (but you can always try), I don't even have an H120
14:35:56AlexPI think that was there on most bootloaders
14:35:59bipakoh yeah!
14:36:26AlexPand it should be in the manual (I'll just check)
14:36:37bipakthanks, pressing play while holding record down works :)
14:36:58AlexPyep -
14:39:39bipaknew rockbox is running :D
14:39:47bipakthanks alot for this great firmware
14:39:57AlexPyou are welcome :)
14:44:09 Quit sbhsu_ (Read error: 110 (Connection timed out))
14:45:50 Quit n17ikh (Read error: 131 (Connection reset by peer))
14:48:35 Join Omlet [0] (
14:50:23bipakis there any lowcost player you can suggest?
14:53:55 Join n17ikh [0] (
15:03:10 Join DarkDefender [0] (
15:08:49kugelAlexP: I was wondering if FS #10553 can go in before 3.4?
15:09:05kugelit doesn't look like, does it?
15:09:29AlexPkugel: I'll have a look this afternoon
15:09:46AlexPkugel: But I don't think it is hugely important - we don't have a release manual
15:10:01AlexPanyway, we have the week of the branch
15:10:19kugelno week, 3 days
15:10:24kugeland we have a release branch
15:10:32kugelrelease manual *
15:11:00AlexPWhere do you get that from?
15:11:29AlexPah, found it
15:11:35AlexPI never knew about that :)
15:11:54AlexPOK, well I have a look this afternoon anyway
15:13:19kugelwhich reminds me, I wanted to document the playlist viewer feature :S
15:13:25*kugel totally forgot about that
15:15:26 Quit FOAD (Read error: 110 (Connection timed out))
15:17:40 Nick fxb is now known as fxb__ (
15:19:44pixelmakugel: if you want more up-to-date translations maybe send a call out on the ml, this seems to have helped in previous releases
15:19:57kugelit helped yes
15:20:20kugelbut I just branched. the branch phase is really not indented to be the translation phase
15:21:21pixelmasays who? I know it's a bit harder because you have to update both branches, but I remember having done so before
15:22:13 Join bluebrother [0] (n=dom@rockbox/developer/bluebrother)
15:22:31 Join sbhsu [0] (n=a6530466@
15:24:49 Quit DarkDefender (Read error: 104 (Connection reset by peer))
15:26:31kugelwell, I'm not going to encourage this. Release branches should be left untouched ideally after the initial branch happened
15:28:32pixelmaI think it's ok for translations as it's very unlikely to break something there just making the language support better
15:29:05bluebrotherwhat's the point of a branch if you don't change it?
15:30:04 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
15:30:49kugelit's not a "cool-new-feature" branch, it's a release branch. any commits should be avoided to not risk its stability
15:31:28kugelI thought that would be clear after 3 releases of this kind
15:31:34bluebrother"release branch" doesn't mean "no commits branch". It's just a "fixes only" branch.
15:31:36kugel4 even
15:31:57kugelwe have the freeze phase for "fixes only"
15:32:02bluebrotherkugel: what's the point of this stupid remark?
15:32:42bluebrotherif you don't want to change a branch at all you can just write down the revision you'd have branched otherwise and tag that instead of doing a branch at all.
15:33:02bluebrothermissing strings in a translation is something that is part of fixing.
15:33:13*bluebrother shakes head
15:33:18*kugel too
15:34:12bluebrotherI've done more than 4 releases in the last couple of years. It just happened to not be Rockbox releases.
15:34:20bluebrotherso stop the crap.
15:34:31kugelI'm not going to discuss it further. not my problem you missed how we did the previous releases
15:34:32pixelmathe branch also allows you to e.g. disable broken features that you don't want to disable in the main code in the hope that it'll serve as a reminder to fix it, e.g. metronome
15:35:21bluebrotherkugel: would you *please* stop this nonsense? Why on earth do you think I missed how the last releases were done?
15:36:10bluebrotherpixelma: disabling broken features in a branch is changing the precious branch! How dare we did that in the past!
15:36:14domonokykugel: if you look into the history, you will see that we have done changes on the release branches..
15:37:09Bagderbug fixes done from now until the release are typically done in both branches
15:37:19Bagderand the release branch usually gets "cleanups" for the release
15:37:30pixelmaI don't like that style of discussion
15:37:46bluebrotherpixelma: me neither.
15:37:55LloreanThe whole point of the branch is for it to receive minimal changes that are very safe. Language changes are pretty darn safe.
15:38:30Bagderand if we want to disable stuff in the release we don't do in trunk, that's also perfectly fine
15:38:58LloreanYes, disabling things are usually safer than trying to fix them at the last moment, so that fits into the whole safe changes thing. :)
15:39:08LloreanBut "not touching it at all" defeats the purpose of even having a branch.
15:39:19bluebrother<sarcasm>That's a change in the precious branch!</sarcasm>
15:39:43kugelyes, disabling features that aren't working sufficiently and release stuff like the version number, that's what the branch is for. urgent last minute fixes are OK too
15:39:46LloreanCode should be as hands off as possible, which just means "don't change it unless there's a very good reason" but we had this discussion around 3.0 or 3.1 I think where we decided languages are pretty much fine.
15:40:10LloreanWell, not such a discussion then.
15:40:21LloreanJust a sort of "yeah, sure, sounds good" couple of statements.
15:40:45kugeltranslations are pretty safe, but I'm not going to encourage having the branch phase as translation phase (i.e. I won't send any "call for translator" mail)
15:41:15aidywhy not
15:41:41aidytranslation fixes aren't new translations anyway
15:41:42bluebrotherso I missed how we did releases in the past simply because I don't have a problem with calling translators during the release branch phase? *great!*
15:42:01*bluebrother is *really* impressed by this logic
15:42:01LloreanThat doesn't encourage it as being the "translation phase" that encourages getting the translation done before the release even though we got it started late.
15:42:33domonokythe call for translators still should get out.. rasher thats your job ! :-)
15:43:13*rasher doesn't see the point of this
15:43:27rasherThat is, differentiating between the freeze and branch phase for translations
15:44:00domonokytrue, we just want fine translations before we release, doesnt matter when they get in...
15:44:26rasherI do think it's a bit late now though, possibly
15:44:30CIA-43New commit by Domonoky (r22749): rbutil: bump version in preparation for release.
15:44:44domonokyrasher: better late, then never :-)
15:44:48 Quit genia4 ("leaving")
15:44:57pixelmayes, seems we are all a bit release tired... ;)
15:45:09rasherThink it was just bad timing
15:45:38pixelmathe whole release schedule suffers from being "a bit late"
15:47:23CIA-43New commit by Domonoky (r22750): tag rbutil 1.2.3 release.
15:48:37domonokybluebrother: rbutil release tag is ready. time to build some binarys :-)
15:48:54bluebrotherdomonoky: did I miss something? Did we decide to tag *now*?
15:49:14kugelI haven't followed it closely, what's the status of the unstable ports in rbutil?
15:49:53domonokyi asked yesterday if there is anything against releaseing this weekend. the only thing was the marking of unstable (which is done)
15:50:02kugeland, can you use it to install it on fuze/e200v2 and samsungs (they're supposed to have rbutil support)?
15:50:33domonokykugel: only fuze/e200v2 is enabled in rbutil.
15:50:45kugelI guess you can't install on the samsungs, no bootloaders on the download server
15:51:04kugelI was about to send Bagder some these days
15:53:36 Join guest101 [0] (
15:56:06 Join pamaury [0] (
15:56:29domonokykugel: the samsungs are not even in the unstable category... so way to early for rbutil :-)
15:56:44*bluebrother would have tried to update the german rbutil translation, but as we needed to tag immediately ... :/
15:56:55kugeldomonoky: actually I looked at putting those into the category today
15:57:12kugelbut www/ doesn't contain the latest website (as I mentioned earlier)
15:57:32domonokybluebrother: your chance to say this was yesterday :-) but if you want, we can move the tag.
15:58:23bluebrotheryou didn't tell that you are going to tag during the mid of the day, and that you are going to do it. It was only talk about considering a release ...
15:58:32guest101hi. i just installed rockbox, but i can't seem to find where to you enable time stretch. i can't find it in the sound settings. any help?
15:58:36 Join wincent [0] (
15:59:08kugelBagder: but you'll still be doing th zips?
15:59:12domonokyguest101: take a look at the manual ?
15:59:17BagderI don't know
15:59:25LloreanCould we maybe schedule an automated email for the second of the appropriate months announcing the freezes on the 9th (1 week later) to both serve as a reminder for us, and a general announcement, that it's release time?
15:59:29kugel(re: "personally I'm just in the bg for this release")
15:59:42BagderI think I'll manage to make them
15:59:54LloreanDidn't someone else make the .zips last time?
16:00:02kugelBagder: do you have an idea why www/ is outdated?
16:00:17Bagderbecause zagor edited the files locally I assume
16:00:21pixelmaguest101: which player do you use?
16:00:26guest101domonoky: already read the manual.. it's under sound settings, but i can't find it there.
16:00:36guest101pixelma: ipod 5.5
16:00:41pixelmaguest101: and which revision (build)?
16:00:45kugelI wanted to put the samsungs into unstable, but there's little point as of now
16:00:51Bagderyeah, anyone can build the zips, that's why we have the zip building scripts in svn
16:03:30guest101pixelma: i just run the rockbox installer tool.. how do you know which build you have?
16:03:56kugelguest101: the latest release (3.3) probably which doesn't have timestretch
16:04:51guest101kugel: oh i guess that's it. what release has time stretch? i want to speed up playback without affecting pitch for audiobooks.
16:04:55domonokyguest101: use the "Install Rockbox" button in rbutil, and select latest build. then you should have timestretch
16:05:13kugelor you wait a few days for 3.4 :)
16:05:30bluebrotherguest101: Rockbox Utility does tell you which build its going to install. Afterwards, check System / Rockbox Info (surprise surprise! ;-)
16:06:04bluebrotherhmm, the beast battery calculation is great. 91%, 127h :)
16:06:09 Join gregzx [0] (
16:06:47guest101bluebrother: oh ok. i get it now. thanks for all the help you guys!
16:09:01 Quit guest101 ("CGI:IRC (EOF)")
16:12:39*pixelma wonders if this lack of enthusiasm about the release is a sign that a 3-months release cycle is a bit short
16:14:04kugelpossibly, I don't like that 3-month cycle
16:16:19 Quit Grahack ("Leaving.")
16:21:03LloreanI think a longer cycle isn't really going to make remembering to do it easier, though.
16:22:27n1si think a longer cycle would be better
16:22:39LloreanIt would better accomplish what?
16:22:46*domonoky thinks we just need to automate the release thing more..
16:23:28LloreanI mean, a longer cycle's fine if it meets the goals determined by releasing, but you should figure out what those are first, then fit your release cycle around it.
16:23:38 Quit Omlet ()
16:23:45n1sbut more importantly we need someone to step up and suggest a schedule, do branching, send mails etc to get the ball rolling, something Bagder has done previously
16:24:04domonokyespecially automated reminder mails might be a good thing.
16:24:13BagderI think so too, I don't think prolonging the periods will make anything better
16:24:18n1sLlorean: i think it would help build up more enthusiasm for a release
16:24:37*n1s see no point in automated mails
16:24:40Bagderthe problem is that us devs don't really have any benefits from releases
16:24:56Bagderother than getting users more up to speed with what's on trunk
16:25:09n1sthe tracker reminder mails go directly in the spam dir...
16:26:24 Join DarkDefender [0] (
16:26:52domonokyn1s: its your fault to let them go into the spam dir. automated mails might help so we dont forget we are near release time.
16:27:37LloreanI think the problem was more "lack of memory" than "lack of enthusiasm".
16:28:00n1sdomonoky: yes, it's my fault because i set up a filter do do that
16:28:04LloreanWell, for the most part. Apparently the one person who did remember of those of you who're active here wasn't enthusiastic, so chose not to remind you guys.
16:28:54LloreanBut I mean, any of you can choose not to participate whether it's a person or an email who reminds you, so I don't think email reminders are any worse than an active person.
16:29:13LloreanAs long as you don't intentionally choose to filter out release emails (which you could do just by trashing any personal emails about the release too, anyway).
16:29:35LloreanBut again - what's the goal with the release?
16:30:12LloreanRight now at 3 months it's short enough that we can say to people "just roll back to the release for now" when there's a bug around, knowing they usually won't lose but one or two significant features.
16:30:21LloreanWhat's the proposed longer time frame?
16:31:11n1si think something like twice a year would be better
16:32:08***Saving seen data "./dancer.seen"
16:32:17*domonoky thinks that at devcon, we sayed to stay with the 3month cycle.. ( or at least we try to:-) )
16:34:19kugelyea, I brought that up
16:35:18*domonoky thinks that with automated reminders, and selecting a "release manager" for every release, we should be able to orgnaise this in a good way.
16:36:20Lloreandomonoky: See if you can get a volunteer when 3.4 comes out for a 3.5 RM, that way you don't have to remember to pick one later. :)
16:37:49domonokymaybe we should let "rsb" select the RM, then rsb has something todo :-)
16:38:32LloreanWell, you need more than one volunteer before a decision needs to be made anyway
16:38:37LloreanYou can't really just assign someone the job.
16:38:49domonokyyes, just joking...
16:38:57 Quit robin0800 (Remote closed the connection)
16:39:52LloreanWell, just wanted to say that since people keep saying "pick an RM" but you don't even have a single volunteer yet. :)
16:39:53domonokywe should just ask shortly before the release, who wants to volunteer to organise the next release... maybe give some bonus for this person ? (a rockbox release manager shirt for example)
16:40:58 Join panni_ [0] (
16:41:22LloreanT-shirts are always a fun incentive.
16:41:38bluebrotherthe release dates might be a bit of a problem ... next release would be almost on christmas.
16:42:12bluebrothershifting it two weeks earlier (or a month later) might be a good thing, at least for 3.5
16:42:15LloreanWe could do three releases a year. Six months, three months, three months. So we skip over Christmas, and give everyone a break?
16:42:31domonokypeople even mentor a student for the whole summer for a shirt, so a "Rockbox 3.4 Release Manager" Shirt should be a good incentive :-)
16:43:25domonokyBagder: can you please move this: to the download server ?
16:44:07*bluebrother could consider trying to do some organizational work for the next release, but no idea about internet connectivity during christmas
16:44:16bluebrotherplus, there's 26c3 after christmas.
16:44:43domonokylets have a release party at 26c3 :-)
16:44:57bluebrotherdomonoky: you coming? It was much fun last year
16:46:45 Join matsl [0] (
16:47:06 Join fml [0] (
16:47:59 Quit stoffel (Read error: 148 (No route to host))
16:48:10fmlHello. I'm trying to apply a patch that I created myself, but I get the following error: "patch: **** malformed patch at line xx" How can this be? And what can I do to avoid this?
16:48:29*bluebrother wonders if there's something similar to ldd for windows binaries, i.e. some command line version of depends.exe
16:48:34fmlIt's the latest patch from FS #8768
16:49:12domonokybluebrother: try depends.exe /c :-)
16:51:50 Join robin0800 [0] (
16:51:59 Quit TheSeven ("#linux4nano-core")
16:52:37 Join TheSeven [0] (n=theseven@
16:55:55AlexPWith viewports, if you define x and y and leave the width and height blank it'll stretch from the x,y to fill the rest of the screen right?
16:56:09*AlexP is still bashing FS #10553 into shape
16:56:16bluebrotherdomonoky: nice, but not exactly what I'm looking for. Might be useable enough though.
16:59:31CIA-43New commit by bluebrother (r22751): Fix short version of qmake command option. Add display of building time.
17:00:03bluebrotherBagder: rbutil 1.2.3 w32 and linux binaries at Can you move them?
17:06:07pamaurybluebrother: what are you looking for exactly ?
17:06:33fmlCan anybody please try to apply this patch to the current SVN: ? And tell me whether there are any problems?
17:07:32bluebrotherpamaury: I'd like to have some checking in the release script that gives me a warning if the binary is dynamically linked against some dlls. Especially mingwm.dll on Windows.
17:07:34rasherAlexP: did you look at the manual bits I wrote?
17:07:51rasherFS #9934
17:08:01AlexPI didn't know about that
17:08:05AlexPI'll have a look
17:08:13rasherDoesn't look like anyone did
17:08:18rasher(me included)
17:08:21bluebrotherfml: malformed patch at line 44
17:08:26AlexPThis patch is a pain as well - full of errors, tabs etc.
17:08:34AlexPI've nearly got it done though
17:08:42rasherWhen I saw the opened task, my first reaction was "what, I added that ages ago!"
17:08:46AlexPrasher: I'll finish this then try and merge yours
17:08:47rasherTurns out I didn't.
17:09:01AlexPOr is yours ready to go straight in?
17:09:16rasherI don't remember.
17:10:11CIA-43New commit by bluebrother (r22752): Substract the correct number of minutes when calculating the durations seconds part.
17:10:52bluebrothernote to self: when testing something that calculates a duration in min:sec, don't try with a tree that builds in less
17:11:29pamaurybluebrother: and what is the problem with depends.exe ? Isn't it interactive AFAIR ?
17:11:53bluebrotherpamaury: yes, but I like to have that in the deploy script. I.e. non-interactive
17:13:22fmlbluebrother: yes, I get the same error. But I don't understand why. I also get errors with other patches from that task :-/
17:13:29pamaurybluebrother: it's quite easy to create a program that list the dll a exe/dll depends on. Perhaps I even have one on my computer
17:19:36 Quit domonoky (Read error: 104 (Connection reset by peer))
17:20:08 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
17:26:15 Quit jamie_ (Remote closed the connection)
17:35:18 Quit shaggy-h ()
17:38:26 Join Lear [0] (i=chatzill@rockbox/developer/lear)
17:40:21 Join chandoo [0] (
17:44:23*fml has no clue why he keeps getting error messages when patching. The patch wasn't edited manually!
17:45:33pixelmalinefeed issues perhaps?
17:46:23bluebrotherno. I've tried dos2unix the patch, didn't change anything
17:46:54bluebrotherthe only thing I can think of is line numbers being off. Though that shouldn't happend if the patch was generated directly from svn and not changed manually
17:47:55AlexPrasher: It looks like the patch I've been playing with contains what yours does (in fact, looking at the text some of it seems to have come from there...) except it puts the description in advanced topics and just the tags in the wps tags section
17:48:07fmlI even tried delete everything but the first hunk −− no success.
17:48:09AlexPrasher: Would you like to have a look at my cleaned up version of this one?
17:48:49kugelAlexP: viewports isn't a WPS-only thing anymore, which is probably why it's under andvanced topics
17:49:02AlexPyes, I know
17:49:15AlexPIt is as two seperate topics
17:49:22AlexPwps viewports and ui viewports
17:49:38AlexPboth of which should be under advanced topics (for the description) anyway
17:50:23kugelI think the description and UI vp could be under "Customising the User Interface", the wps syntax where the other tags are described
17:50:41kugelwith the addition of conditionals VPsfor the wps
17:52:25AlexPyes, that is what it does
17:53:17bluebrotheris there something like "mkdir -p" on windows? I.e. a way to make mkdir *not* fail if the folder already exists?
17:53:22 Part bipak
17:57:38pamaurybluebrother: are you still interested in a ldd like for windows ? I have a self contained program that can list dependencies (dll only, not imported functions), that works at least on linux (ie not tested on windows) and at least for 32-bit programs :)
17:58:15bluebrotherpamaury: definitely
17:59:02 Join stoffel [0] (
17:59:44pamauryhope it can helps, it's not very robust but it should work
18:00:39bluebrotherpamaury: I'll check that, it's better than having nothing to start with :) Thanks.
18:01:17AlexPkugel, rasher:
18:01:24pamaurybluebrother: it's nothing, if you want to get rid of all the preliminary definition, just include windows.h and it should works
18:03:16 Quit pamaury ("exit(*(int *)0 / 0);")
18:03:51kugelAlexP: the patch doesn't tell me much :( can you build me a pdf?
18:04:11AlexPkugel: It tells you everything!
18:04:17AlexPkugel: What target?
18:04:39kugelcolor and/or grey
18:05:18*AlexP doesn't know this "color" thing :)
18:05:29kugeldo you have an idea why {\textbar} is needed for the \config thing, but not for \begin{example} ?
18:05:34kugelcolour* :)
18:05:54kugelfirst hunk of the patch
18:05:58kugellines 8 and 22
18:06:26AlexPno, I hadn't noticed that
18:06:30AlexPI'll have a play
18:07:13AlexPI presume because \example{} takes it verbatim
18:07:18kugelbluebrother: does python not know the % operator?
18:08:06AlexPkugel: Note the % doesn't need escaping either
18:09:28AlexPkugel: and
18:12:04 Quit gregzx ("ChatZilla 0.9.85 [Firefox 3.5.2/20090729225027]")
18:14:16kugelAlexP: Looks largely fine to me, although the major part of the discription of viewports is in the WPS section, which comes after teh UI viewport section
18:14:49kugelnot a big deal, since there's a nice reference, but I can imagine that viewports are described very generally before UI vp and WPS vps
18:15:32AlexPkugel: Yes, a good point - I'll see if I can shuffle it around
18:16:35CIA-43New commit by alle (r22753): Pitch Detector: add the possibility to set the reference frequency of A (last part of FS #8768)
18:18:35 Join robertoss [0] (n=rob@
18:20:14robertosshey everyone, quick question about turning on the ipod video, i noticed that i can turn it on by quickly pressing menu, and then also boot to the apple OS with a Long Menu, but the manual says to use menu/select? Is this something that is well known or is it new? BTW i have the large disk build from bobbozo
18:22:31 Join saratoga [0] (
18:22:52saratogacan someone add the Fuze to the theme site?
18:23:08 Quit wincent (Read error: 110 (Connection timed out))
18:24:08kugelsaratoga: done, it needs a picture though
18:25:37kugelI have one, but I have no idea how make it show up there
18:26:20 Quit robin0800 (Remote closed the connection)
18:26:30kugelrasher: ^ ?
18:26:55saratogakugel: shouldn't the fuze use ipod color themes
18:27:20kugelyea, for some reason it has no theme
18:29:33rasherkugel: needs to be on the main site
18:30:01rasherkugel: in
18:30:03CIA-43New commit by kugel (r22754): Add small pictures for Sansa Fuze and Samsung YH925.
18:30:27kugelrasher: like this?
18:30:57rasherkugel: don't know if they get pushed or if you need a swede to update www
18:32:06rasherkugel: but yes
18:32:12***Saving seen data "./dancer.seen"
18:32:37saratogano theory as to why it didn't group the ipod color themes with the fuze?
18:33:26CIA-43New commit by alle (r22755): Pitch Detector: use special functions for choosing bool values
18:35:31domonokyeither the wrong resolution was entered or it has todo with the checkwps checking.
18:36:16 Join Farthen [0] (
18:36:21kugel220x176 is what I entered
18:37:31kugelhrm, it seems I should've used "Sandisk Sansa Fuze" :/
18:37:34kugelcan that be changed?
18:38:04kugelit doesn't matter for the themes, but it would be more consistent with the other sansas
18:38:53*rasher will fix
18:39:27domonokyalso the e200v2 should be added, or rbutil wont find themes for it.
18:39:30kugelalthough the ipods don't have apple in front
18:42:10rasherscorche: we need libsdl1.2-dev on the theme server (to build checkwps)
18:42:49 Join Grahack [0] (
18:44:29saratogathe svg images on the download site look so much nicer then the sim pictures
18:47:12rasherYeah, why aren't we using these for the sim?
18:47:34saratogado we have higher res versions?
18:47:39saratogaor is svg scalable
18:47:51rashersvg is scalable
18:47:58rasherThey're the same ones used in the manual
18:48:48CIA-43New commit by alex (r22756): Add viewports to the manual. Adapted from FS #9934 by Jonas Häggqvist and FS #10553 by David Kauffmann
18:49:11AlexPDid we branch for 3.4? I guess I should put that in there too
18:49:15rasherJust needs an overlay for the buttons
18:49:22rasherAlexP: not sure it matters for the manual, but yes we did
18:49:34rashernot much work, I guess
18:49:40AlexPProbably best too - 3.4 contains the viewports :)
18:49:51AlexPYeah, I'll do it shortly
18:53:41Topic"3.4 has now been branched | Please read before speaking: | Please direct offtopic/social chat to #rockbox-community" by ChanServ (ChanServ@services.)
18:54:39 Join stevetuc [0] (i=www-data@
18:57:56 Quit stevetuc (Client Quit)
18:59:30*domonoky just added e200v2 to the theme page. but it also didnt get any themes. so it probably needs new checkwps binarys for the new targets.
19:03:13CIA-43New commit by alex (r22757): Port r22756 to the 3.4 branch - Add viewports to the manual. Adapted from FS #9934 by Jonas Häggqvist and FS #10553 by David Kauffmann.
19:03:32CIA-43New commit by alle (r22758): Pitch Detector: group note related data together
19:04:00CIA-43New commit by bluebrother (r22759): Always format error message using ANSI functions to get rid of type-punning when building with rbutil.
19:10:18CIA-43New commit by alle (r22760): Clarify the unit of note frequency
19:11:55 Quit LinuxMafia (Remote closed the connection)
19:15:00AlexPA-B repeat is on all players right?
19:15:53dionoeaBagder: Hi. it looks like the since-4weeks.html page isn't updated anymore.
19:16:10*Bagder blames a random kid brother
19:17:01pixelmaAlexP: I'm not sure, even if it is there are two ways to control it (depending on the amount of usable buttons)
19:17:26pixelmaI'm not sure the Ondio (and the Player maybe) have it
19:19:45pixelmathe Ondio seems to have it (at least it's in the menu)
19:20:47AlexPpixelma: OK, I'll think about doing something about it at some point :)
19:23:22pixelmaurgh, if you enable A-B repeat on the Ondio you can't go from the WPS to the filebrowser anymore...
19:23:55 Join webguest26 [0] (
19:23:55 Quit webguest26 (Client Quit)
19:25:05amiconnYou can... just not directly
19:26:05pixelmawell, context menu > menu > file browser? Ok...
19:26:43amiconnOf course there is a tradeoff
19:27:24 Quit fml ("CGI:IRC 0.5.9 (2006/06/06)")
19:27:31amiconnI think that A-B repeat needs quick access if enabled, otherwise it could be put into the wps context menu
19:31:03 Quit langzeitstudent_ ("Verlassend")
19:31:48 Quit saratoga ("CGI:IRC (EOF)")
19:31:50Learkugel: regarding the gcc 4.4.1 test, was the binutils snapshot needed for some reason?
19:33:22pixelmaentering the filebrowser through the menu is a bit different than entering directly from the WPS if you have "follow playlist" enabled because you get to folder you last browsed not to the curently playing track, not a big deal though
19:34:57*amiconn thinks that this is a bug
19:35:56pixelmaI think you can see it both ways
19:37:00pixelmaconsistency when entering the file browser through the menu or not
19:37:40pixelmaor for "folliow playlist"
19:38:49kugelLear: 2.19 didn't work without modifications on the previous test
19:41:03 Join robin0800 [0] (
19:48:07kugelLear: if you're running x86_84 I can hand you the binaries
19:48:15kugelfor arm
19:51:43kugelx86_64 of course :)
19:55:04Learkugel: Thanks, but I use Cygwin...
19:56:07kugelThat hurts me bad
19:58:57 Quit inversions (Read error: 110 (Connection timed out))
20:00:52 Nick fxb__ is now known as fxb (
20:09:46 Join Lear_ [0] (i=chatzill@rockbox/developer/lear)
20:10:27 Join merbanan [0] (n=banan@
20:10:40 Join wincent [0] (
20:12:02 Quit flydutch ("/* empty */")
20:12:41 Join Lear__ [0] (
20:18:54 Quit Lear_ (Read error: 60 (Operation timed out))
20:22:18 Join funman [0] (n=fun@rockbox/developer/funman)
20:22:45 Quit stoffel (Remote closed the connection)
20:23:20 Join bughunter2 [0] (n=bughunte@unaffiliated/bughunter2)
20:25:00 Nick Lear__ is now known as Lear_ (
20:27:03Lear_kugel: Hrm, the binutils snapshot from the 20th didn't build for me...
20:28:10 Quit Lear (Read error: 110 (Connection timed out))
20:30:51Lear_kugel: Ah, just a missing lib, easy to fix.
20:32:14***Saving seen data "./dancer.seen"
20:33:44Lear_Now what, test_codec made an empty log file (for a folder run).
20:35:46kugelI didn't do folder runs
20:41:05funmanbughunter2: what action should cause an USB interrupt on Sansa AMS, simply plugging the device on USB ?
20:41:24bughunter2I'd guess so?
20:41:45funmanshouldn't the first action come from the device towards the usb controller ?
20:42:06bughunter2I'm not that experienced with USB, so I'd have to read up on that
20:42:10bughunter2but you could definitely be right
20:42:32funmanjust suggesting, i'm not experienced with usb either
20:42:46Lear_File was written, (chdsk found it). If there was a crash, there was no panic screen...
20:45:14 Quit tarbo (Remote closed the connection)
20:45:29bughunter2funman: maybe you could check the OF init routine once more, to be sure?
20:46:15funmanthere is not "one init routine" iirc, what are you looking for exactly ?
20:46:29bughunter2the usb driver initialization?
20:46:50 Join bertrik [0] (
20:46:50funmanwell i believe the initialization is split in several places
20:47:37funmansome functions can be hard to follow due to several reasons (function pointers comes to my mind)
20:49:24bughunter2I'd guess that a lot of the USB packets going to the host are the same on most players... if that's the case then couldn't we just compare other rockbox code and check whether rockbox sends the first packet?
20:50:26funmansounds good
20:51:41 Quit Galois ("Leaving")
20:52:03 Quit GeekShadow (Read error: 54 (Connection reset by peer))
20:52:09bertrikbughunter2, as far as I've seen so far USB controllers give interrupts for the following two broad cases: device interrupts (plugin/out, speed-change) and endpoint interrupts (packet sent or received for example)
20:53:01bertrikthe usual case is that the device activates a pull-up on the D+ line (or inserted with the D+ pullup already active), this is detected by the host which does a USB device reset
20:53:02bughunter2so then we should get an interrupt, if interrupts are working..
20:53:23bughunter2sorry I don't know what the D+ line is
20:53:54bertrikit's one of the four signal lines used in USB (ground, power, D+, D-)
20:54:14bertrikone of the middle two in a flat USB connector
20:55:34 Quit matsl ("Leaving")
20:55:44 Join matsl [0] (
20:56:18bughunter2bertrik: so once I'd connect my e200v2, I should immediately get an interrupt, after rockbox shows the USB logo?
20:56:23 Part Grahack
20:56:26bughunter2or even before that/
20:56:51bughunter2to test whether interrupts work, I simply put a panicf() in INT_USB, but it never paniced :/
20:58:23bertrikI'd expect so, yet
20:58:40bertrik*yes. Does the host see anything happen when you plug it in?
20:58:54bughunter2nope, nothing in the device manager either
21:00:56bughunter2I just inserted some extra panics, let's see if something happens
21:01:02bertrikdo you see anything happen in the dmesg log of a linux machine?
21:01:16bughunter2I'll try that, too
21:04:27 Join HBK [0] (
21:04:38bughunter2bertrik: nothing in dmesg
21:06:39bertrikin that case I think, USB is not active at all yet. It could be that the pull-up on D+ is not activated from the device side.
21:07:01bughunter2so, the device isn't initialized correctly? what else could it be?
21:07:28 Quit Horscht ("Verlassend")
21:08:14bughunter2aw, even with panics all over the place, not even one panic is displayed :P
21:08:23bughunter2except the one in usb_attach that I tried out of paranoia
21:11:18bertrikI'll check the datasheet
21:14:12 Quit funman ("leaving")
21:16:20 Quit Lear_ ("ChatZilla 0.9.85 [Firefox 3.5.3/20090824101458]")
21:17:23bertrikwhich part (chapter) of the datasheet applies to the USB device controller in the as3525?
21:17:34bughunter2of the amd sheet?
21:18:14bughunter2just the ones that match the register names, I'd guess
21:18:22bughunter2haven't read the full thing yet
21:19:27bughunter2the names can slightly differ though
21:19:40bughunter2for e.g. in AS sheet: USB_DEV_CTL, and in AMD's sheet: DEVCTL
21:19:53bughunter2err, USB_DEV_CFG / DEVCFG
21:20:30bughunter2I'm trying to make a little trace, to see what the OF sends to the host
21:22:17bertrikah chapter 6.4.5
21:23:03bughunter2just found out as well
21:24:41 Join wincent_balin [0] (
21:24:53bertrikwe may need to have to configure the USB PHY, the part that does the low-level signalling and I think this can be found out easiest by studying the OF
21:25:14bughunter2afaik funman's code did configure that
21:25:24bughunter2or maybe just partly
21:25:26bughunter2not sure
21:25:39bertrikalso, in registers RXFIFOMEM there is a bit 'APU' that configures how the pull-up works (automatically or under software control)
21:25:40bughunter2 CGU_USB = 1<<5 /* enable */ | (CLK_DIV(AS3525_PLLA_FREQ, 48000000) / 2) << 2 | 1; /* source = PLLA */
21:25:43bughunter2that's from funman
21:26:03bughunter2should I try to set that to automatic, right now?
21:26:18bughunter2also, I just captured the packets the OF sends to the host
21:26:29 Join Galois [0] (
21:26:30bertrikthe problem is that the as3525 does not have this particular register ...
21:27:01bertrikCGU_USB just enables a clock towards the USB device controller and is only a really small part of initialisation I guess
21:27:34bughunter2it looks like the host indeed sends the first few packets, before getting packets from the device
21:27:47bughunter2so, we should indeed get interrupts immediately after plugging the device in
21:28:16bertrikthe device cannot really take the initiative to send packets to the host, it's done only on request of the host
21:28:24 Join AndyIL [0] (n=pasha_in@
21:28:47bughunter2so that in itself is enough to be sure we should get interrupts
21:29:57bughunter2bertrik: how'd you figured the RXFIFOMEM thing out btw?
21:30:15bertrikrasher, I am thinking of committing FS#
21:30:57bertrikbughunter2, I was searching for "pull" in the datasheet (from pull-up)
21:31:16bughunter2how'd you know it's related to that register though?
21:31:33rasherbertrik: sounds good. I did in fact test it on my e200 as well
21:31:38bughunter2I only stumble upon APU by searching for pull
21:33:12bughunter2bertrik: could " GPIO Pull-Up Enable (GPIO[x]_PU_EN)" be related?
21:33:41bertrikrasher, IIRC I did test it on sansa clip too (and it worked). I'll wait for the freeze to end.
21:34:42bertrikbughunter2, yes I think so
21:34:55 Quit bluebrother (Nick collision from services.)
21:35:00 Join bluebrother [0] (n=dom@rockbox/developer/bluebrother)
21:35:00bughunter2I'm not sure how to enable it, if it's possible that way..
21:37:22 Quit domonoky (Read error: 104 (Connection reset by peer))
21:38:30 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
21:40:24bughunter2bertrik: it says 'GPIO I/O Offset 18h', but how can I write to this? should I add 18h to some base register? If so, which?
21:44:52bertrikhm on closer inspection, those pull-ups probably have nothing to do with USB and are part of a different device module
21:45:10 Quit robertoss ()
21:45:21 Quit wincent (Read error: 110 (Connection timed out))
21:45:23 Join froggyman [0] (
21:45:30bughunter2in the comments of the linux amd5536udc driver, it says: " * Make sure that UDC is assigned to port 4 by BIOS settings (port can also be used as host port) and UOC bits PAD_EN and APU are set (should be done by BIOS init)."
21:46:34bughunter2uoc being usb option controller
21:48:17 Quit AndyI (Read error: 113 (No route to host))
21:51:43bughunter2bertrik: I just added some code which I thought might enable APU, and now rockbox says (when connecting) "data abort at 30058e00"
21:52:33bughunter2probably not good
21:53:13bertrikit could be that the data abort fires when you try to access a peripheral and it doesn't respond (no peripheral registers at this address), but I'm not really an ARM expert
21:53:17CIA-43New commit by vitja (r22761): i7: pcm-telechips fix sampling rate that was about 40000 instead of 44100 ...
21:53:42bughunter2I added: *(unsigned long *) (0x5120000b /* UOC base */ + 0x00 /* UOCCAP */) |= (1<<15);
21:54:26 Quit matsl (Read error: 110 (Connection timed out))
21:55:14 Join calman_ [0] (
21:55:28bughunter2probably should add volatile.. not that it would magically make it work
21:55:50 Part calman_
21:56:15AlexPer, what the hell is the first one on here:
21:56:16 Join calman_ [0] (
21:56:24*AlexP investigates with a few to deleting
21:58:52AlexPrasher: I just went to delete a theme - I think it has gone but I got Warning: rmdir(/home/themes/www//themes/220x176/.rockbox) [function.rmdir]: Directory not empty in /home/themes/private/themesite.class.php on line 295
21:59:19AlexPkugel: Someone submitted cabbie v2 with the screenshot a badly cropped photograph of his screen
21:59:36kugelyea, I saw that. I just didn't understand what you said
21:59:48AlexPWhich line?
21:59:59kugelthe two directly above
22:00:41AlexPinvestigates with a view to deleting = looking at it pretty sure that I'm going to delete it
22:01:52kugelaye, I wasn't familiar with that phrase
22:02:37 Join petur [50] (n=petur@rockbox/developer/petur)
22:02:52rasherAlexP: that looks... wrong
22:03:04AlexPrasher: indeed it does :)
22:04:25rasherAlexP: It's gone.. not sure what happened :|
22:05:31AlexPrasher: Me neither, just wanted to let you know
22:07:44 Quit billenium (Connection timed out)
22:08:10bughunter2bertrik: some news :).. I at least found some opcodes in the OF disassembly that match funman's code :D
22:08:49rasher> Some person on gave this theme as part of the firmware (rockbox) as freeware.
22:08:55rasherThat seems more disturbing
22:11:32 Quit n17ikh (Connection timed out)
22:11:53 Join wincent [0] (
22:14:44AlexPrasher: also, the e200v2 is showing zero themes - it should be showing the e200 ones, no?
22:16:12 Quit wincent_balin (Read error: 110 (Connection timed out))
22:17:25 Part Farthen
22:17:32rasherAlexP: It should, but I have no checkwps for e200v2
22:17:55rasherAlexP: and only themes which have passed checkwps get showed
22:17:57AlexPah right
22:18:09AlexPthat makes sense
22:18:20rasherJust need scorche to install sdl so I can compile checkwps
22:19:43 Join wincent_balin [0] (
22:22:15 Join PaulJam [0] (
22:25:19kugelbertrik: the freeze ended
22:26:11kugelrasher: you need sdl for checkwps?
22:27:21rasherkugel: seems so
22:27:54kugeldoes it give an error?
22:28:37bughunter2bertrik: will continue another day, bedtime :)
22:28:41bughunter2thanks for the help
22:28:47kugelI wasn't aware of this, an I don't think it's needed
22:31:43kugelrasher: the build system builds checkwpsm wouldn't it be useful to use that=
22:32:00 Quit vitja ("Ex-Chat")
22:32:09kugelit isn't uploaded yet, but I think that could be changed for the theme site
22:32:15***Saving seen data "./dancer.seen"
22:33:44rasherkugel: seems needlessly complicated
22:33:57rasherIt takes what.. at most a minute to build them all?
22:34:24 Quit wincent (Read error: 110 (Connection timed out))
22:34:27kugeldepends on the machine
22:34:36kugelmy c2d needs more
22:35:14kugelhow old is the checkwps that's currently in use?
22:35:51kugelif checkwps isn't always up-to-date, how do you garuantee that it's actually correct when it says "does not work with the current build"?
22:38:44rasherit is always up to date
22:38:49rasherexcept when scorche breaks everything :)
22:41:02 Join moonscapex [0] (
22:41:45 Quit moonscapex (Client Quit)
22:42:31 Quit wincent_balin (Read error: 110 (Connection timed out))
22:47:50 Join n17ikh [0] (
22:56:00 Quit petur ("Zzzzz")
22:57:40 Join Horscht [0] (n=Horscht2@xbmc/user/horscht)
23:03:39 Join webguest90 [0] (
23:04:39 Quit webguest90 (Client Quit)
23:04:48 Join webguest42 [0] (
23:05:40webguest42Hey, does rockbox work on samsung yps3?
23:07:39bertrikno, there's just a simple demo program, based on a rockbox bootloader
23:07:55webguest42crap in a biscuit.
23:08:07webguest42Thanks anyway.
23:09:06webguest42Is there any way I could contact the rockbox crew and request that?
23:10:09TheSevenwebguest42: I don't think you need to, and you've more-or-less already done it hereby
23:10:29 Join casainho [0] (
23:10:31TheSevenbertrik: are you just working on the meizus or is the yps3 also yours?
23:10:52casainhohello Rockbox hackers :-)
23:10:58webguest42I guess I'll wait.
23:11:03webguest42Bye now.
23:11:12 Quit bubsy ("I'll be back somewhere in time")
23:11:22 Quit webguest42 ("CGI:IRC (EOF)")
23:11:28 Quit DarkDefender (Remote closed the connection)
23:12:03casainhoI would like to know if Rockbox already have USB support?
23:12:18TheSevencasainho: for which target?
23:12:18 Quit Galois (Remote closed the connection)
23:12:29 Join Galois [0] (
23:12:43 Join webguest42 [0] (
23:12:51casainhowell, I am looking for adapt on some costum hardware like this one:
23:12:56 Quit webguest42 (Client Quit)
23:13:04AlexPcasainho: Yes, in general
23:13:26AlexPcasainho: There is a USB stack, but of course drivers for individual hardware is needed
23:13:30casainhobut would Rockbox work as a USB Host?
23:14:01casainhoso, the Rockbox USB Stack is working now on Sansa E200, for example?
23:14:56casainhoand the same question goes for Ethernet connection. I would like to know If Rockbox would play files over Ethernet, or something like that
23:15:28AlexPUSB host, no
23:15:53AlexPThe e200 uses the Rockbox USB stack for UMS, yes
23:16:01AlexPAnd Rockbox has no networking stack
23:16:26TheSevencasainho: wow, this is a really cool device for that price!
23:16:57casainhoAlexP: ok, thanks :-)
23:17:45casainhoTheSeven: yes, looks like it is :-) there are all the schematics and datasheets available on product donwload page:
23:18:53casainhoTheSeven: one developer already bought that hardware and I am looking if I would buy also an join, for porting Rockbox for it:
23:25:03kugelcasainho: yet again a hardware change?
23:26:11kugelIIRC rockbox is mostly complete for s3c2440. The Gigabeat F and X use that SoC
23:26:20casainhokugel: yes... we are still waiting for the ARMopendous-DAP boards, but is difficult, the guy designing the boards have problems with it. Now another person got near the project and bought that hardware, to port Rockbox for it.
23:26:34casainhokugel: oh, nice to know :-)
23:27:21domonokythat hardware really looks nice. and it already contains everything needed.
23:27:22 Join tarbo [0] (n=me@unaffiliated/tarbo)
23:27:33casainhoand costs $85 :-)
23:27:36 Join bubsy [0] (n=bubsy@
23:28:17casainhowould be nice to have a kind of Rockbox hardware for having at home, car? −− because that hardware is not portable...
23:29:05 Part malouin
23:29:26 Join wincent_balin [0] (
23:29:31 Quit wincent_balin (Remote closed the connection)
23:30:07 Join wincent [0] (
23:30:16 Join MG_Man [0] (
23:31:35kugelit has vga and ethernet connectors. that already exceedes the limites of Rockbox. but one could surely make funny experiments with it
23:32:15domonokykugel: yes, rockbox would surely not use all features. but it has everything needed for a audio player (and more)
23:32:38BagderI figure there's a reason the screen on that device shows it running android...
23:33:07domonokyethernet shouldnt be too complicated, there is this nice lwip stack.. you only need a hardware driver.
23:33:20 Quit bluebrother ("leaving")
23:33:40Bagderwell, and probably some apps that use the network ;-)
23:33:55domonokytrue :-)
23:34:02kugeland viruses, rootkits and the like :p
23:34:56kugel"We have lan! We cannot prove it since we have no apps, but we're quite surely it's working" :)
23:35:37casainhobut, would people be interested in using such hardware running Rockbox? (or should I wait for a TRUE open hardware like the ARMopendous DAP?? )
23:35:39kugelwell, with a small stack we should at least be able to respond to pings
23:36:09kugelwe're interesting in anything that potentially runs our os :)
23:36:49kugelbut getting rockbox run on that seems like a no brainer actually
23:36:57 Join wincent_balin [0] (
23:38:10domonokythe only thing missing on this board for perfect home/car usage is a ata-port to connect a big-big storage.
23:38:44TheSevendomonoky: bitbanging GPIOs? :-P
23:38:59Bagderdomonoky: attach a 2TB usb-disk
23:39:09TheSevenand you can always use USB-to-SATA adapters of course
23:39:13kugelor a NAS via lan :)
23:39:22domonokyBagder: if you write the usb-host stack, shure :-)
23:39:30TheSeven"wow! we found an application for the LAN port!"
23:39:51TheSevendomonoky: USB host for just one mass storage device isn't too hard
23:39:55Bagdermass-storage shouldn't be too hard
23:40:01domonokyBagder: did you see the new rbutil binarys to move to the download server ?
23:40:11TheSevenas soon as there are multiple devices things start to get tricky
23:40:23casainhook - thanks for all ideas ;-)
23:40:26kugel"(5) Bitmap to C Language transfer tool " < BMP2RB ?
23:40:36Bagderdomonoky: oh, I noticed but I forgot. Can I have the url again?
23:41:02 Quit merbanan (Read error: 145 (Connection timed out))
23:41:12casainhoI just don't know if the audio DAC IC used on that board have a good quality... (at least so good as a "normal" DAP)
23:41:26domonoky and
23:41:28kugelit looks like a nice toy I admit
23:42:59domonokywith this toy, you could also just run the sim on linux, then the drivers (usb, ethernet etc) are free :-)
23:43:37kugelthey ship some nice cloned software with it
23:43:46Bagderbluebrother's site is fast and fine, 4309k took a full second to transfer ;-)
23:44:41Bagderour new server's 100mbit connection is nice
23:44:52 Join GeekShadow [0] (n=Antoine@reactos/tester/GeekShadow)
23:45:27kugelBagder: can I send you some bootloaders while you're at it?
23:49:38 Join safetydan [0] (n=deverton@rockbox/developer/safetydan)
23:51:21 Quit casainho ("ChatZilla 0.9.85 [Firefox 3.0.14/2009090216]")
23:51:47 Join lennyk [0] (
23:52:00 Quit MG_Man ("...GOT AWAY SAFELY!")
23:52:17 Quit wincent (Read error: 110 (Connection timed out))
23:56:49 Quit robin0800 (Remote closed the connection)
23:58:18 Join BHSPitLappy [0] (n=BHSPitLa@unaffiliated/bhspitmonkey)

Previous day | Next day