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 2011-06-20

00:03:50***Saving seen data "./dancer.seen"
00:06:11 Join Thra11_ [0] (~thrall@
00:12:44 Join mudd1 [0] (
00:13:47 Quit domonoky1 (Read error: Connection reset by peer)
00:20:05 Quit balintx (Remote host closed the connection)
00:20:05 Join balintx [0] (
00:28:49 Quit liar (Quit: hallowed are the ori!)
00:30:56 Quit ReimuHakurei (Read error: Connection reset by peer)
00:30:59 Join ReimuHakurei_ [0] (~reimu@
00:34:42CIA-27New commit by amiconn (r30019): Fix non-working .talk clips on hwcodec (follow-up to r28672 and r29432): .talk clips ...
00:36:50 Nick neferty is now known as nef_away (~andor@
00:39:51CIA-27r30019 build result: All green
00:40:10 Join liar [0] (
00:40:20 Quit pamaury (Remote host closed the connection)
00:45:55 Join Judas_PhD [0] (
00:47:24 Quit ender` (Quit: When I was a kid I used to pray every night for a new bicycle. Then I realised that the Lord doesn't work that way so I stole one and asked Him to forgive me.)
01:04:58 Quit keyb_gr (Ping timeout: 240 seconds)
01:27:02 Quit sideral (Quit: Leaving.)
01:30:33 Quit Rondom (Ping timeout: 260 seconds)
01:33:48 Join Rondom [0] (~rondom@2a01:488:66:1000:b24d:4f2f:0:1)
01:51:57JdGordon[Saint]: chaining? or changing?
01:56:11 Quit ReimuHakurei_ (Read error: Connection reset by peer)
02:03:53***Saving seen data "./dancer.seen"
02:09:57 Quit MethoS- (Read error: Connection reset by peer)
02:12:37 Quit Thra11_ (Quit: kthxbai)
02:23:53 Join bieber [0] (
02:58:32 Quit Judas_PhD (Quit: This is a quitting message)
03:01:47 Quit kevku (Remote host closed the connection)
03:11:37 Quit liar (Quit: hallowed are the ori!)
03:22:19 Join wtachi [0] (
03:35:16 Join fdinel [0] (
03:40:03 Quit rasher (Ping timeout: 260 seconds)
03:42:36 Join rasher [0] (~rasher@rockbox/developer/rasher)
03:59:48 Part timccc
04:03:57***Saving seen data "./dancer.seen"
04:18:19 Quit TheSeven (Disconnected by services)
04:18:28 Join [7] [0] (~TheSeven@rockbox/developer/TheSeven)
04:32:53 Quit amiconn (Disconnected by services)
04:32:54 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
04:33:12 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
04:33:33 Quit pixelma (Disconnected by services)
04:33:35 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
04:33:37 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
04:38:50 Join JoshuaChang [0] (~JoshuaCha@
04:40:40JoshuaChangi found a strange thing: when disable asm optimize in spc codecs and using -O3 complie switch, the cowon d2 target get 10% speed gain?
04:41:32JdGordondissasemble and figure out why gcc's arm is better than ours
04:44:53 Join kugel_ [0] (~kugel@rockbox/developer/kugel)
04:48:08 Quit kugel (Ping timeout: 258 seconds)
04:49:47 Join philipballew [0] (~philip@ubuntu/member/philipballew)
04:50:19philipballewcan someone help me install rockbox on my ipod?
04:55:22JdGordonhave you tried the manual?
04:56:12philipballewi need to find the best one
04:56:29philipballewim running x8664 linux
04:58:07JoshuaChangthere is no best way, all roads lead to rome
05:02:24philipballewabout how long does it take
05:04:05JdGordontoo long
05:04:33 Join ChickeNES [0] (
05:10:00 Join d3remark [0] (
05:13:56 Quit ChickeNES (Read error: Connection reset by peer)
05:14:28 Join ChickeNES [0] (
05:18:46 Quit ChickeNES (Read error: Connection reset by peer)
05:19:22 Join ChickeNES [0] (
05:20:07philipballewdo i only need to download the RockboxUtility to install?
05:20:59JdGordonread the manual
05:22:45philipballewi did, it says housekeeping tool as well. what is that though?
05:32:14JoshuaChang, this page is gone, where can i found it?
05:34:57JdGordongood question.... google cache is failing me
05:35:08 Quit ps-auxw (Ping timeout: 255 seconds)
05:35:09JdGordonmight have to see if Zagor or Bagder has a backup
05:37:48 Quit Horscht (Quit: Verlassend)
05:38:35 Join Horscht [0] (
05:38:35 Quit Horscht (Changing host)
05:38:35 Join Horscht [0] (~Horscht@xbmc/user/horscht)
05:38:57 Quit Horscht (Client Quit)
05:39:49 Quit fdinel (Quit: Miranda IM! Smaller, Faster, Easier.
05:41:51 Quit philipballew (Read error: Connection reset by peer)
05:43:43 Quit ChickeNES (Read error: Connection reset by peer)
05:44:20 Join ChickeNES [0] (
05:45:17 Join ps-auxw [0] (
05:49:49 Join Rob2223 [0] (
05:53:38 Quit Rob2222 (Ping timeout: 250 seconds)
05:56:19 Quit Keripo (Read error: Connection reset by peer)
06:02:27 Join Keripo [0] (
06:04:00***Saving seen data "./dancer.seen"
06:04:38 Quit [Saint] (Ping timeout: 246 seconds)
06:05:46 Quit ChickeNES (Read error: Connection reset by peer)
06:06:21 Join ChickeNES [0] (
06:09:26 Join [Saint] [0] (
06:13:56 Quit krazykit (Ping timeout: 264 seconds)
06:14:59 Quit ChickeNES (Read error: Connection reset by peer)
06:15:34 Join ChickeNES [0] (
06:18:41 Join stoffel [0] (
06:19:48 Join ReimuHakurei [0] (~reimu@
06:21:50 Quit ReimuHakurei (Read error: Connection reset by peer)
06:21:52 Join ReimuHakurei_ [0] (~reimu@
06:21:59 Quit ChickeNES (Read error: Connection reset by peer)
06:22:25 Join ChickeNES [0] (
06:29:10 Quit ChickeNES (Read error: Connection reset by peer)
06:29:42 Join ChickeNES [0] (
06:35:08 Quit ChickeNES (Read error: Connection reset by peer)
06:36:00 Join ChickeNES [0] (
06:36:16 Quit stoffel (Remote host closed the connection)
06:45:12 Join sideral [0] (~sideral@
06:45:12 Quit sideral (Changing host)
06:45:12 Join sideral [0] (~sideral@rockbox/developer/sideral)
06:45:37 Quit ChickeNES (Read error: Connection reset by peer)
06:46:44 Join ChickeNES [0] (
06:47:21 Quit factor (Ping timeout: 258 seconds)
06:48:16 Quit ChickeNES (Read error: Connection reset by peer)
06:48:59 Join ChickeNES [0] (
06:50:38 Quit ChickeNES (Read error: Connection reset by peer)
06:51:13 Join ChickeNES [0] (
06:55:38 Quit ChickeNES (Client Quit)
07:10:21 Quit d3remark (Ping timeout: 258 seconds)
07:16:09 Quit JoshuaChang (Quit: ChatZilla 0.9.87 [Firefox 4.0.2pre/20110429182132])
07:31:50 Quit [Saint] (Quit: Imagination is for turbo-nerds who can't handle how kick-butt reality is. I'm a kick-butt reality master! I would rather die, than be imaginative. I mean that.)
07:33:12 Quit Strife89 (Quit: Heading off to bed.)
07:33:35 Join d3remark [0] (
07:33:42 Join [Saint] [0] (
07:39:35 Nick kugel_ is now known as kugelp (~kugel@rockbox/developer/kugel)
07:48:08ruskiepixelma, thanks
07:55:21 Join factor [0] (~factor@
07:58:13 Quit balintx (Remote host closed the connection)
07:58:31 Join balintx [0] (
08:01:18pixelmabluebrother: did you try a recent android build? I wonder if I'm the only one where playback doesn't work and it just skips through the songs (that's with r30012)
08:04:04***Saving seen data "./dancer.seen"
08:16:47 Quit FoolOnHill (Ping timeout: 276 seconds)
08:25:46 Join Zagor [242] (~bjst@rockbox/developer/Zagor)
08:28:17 Join FoH [0] (
08:31:53 Join GodEater [0] (~bibble@rockbox/staff/GodEater)
08:37:47 Quit wtachi (Quit: &)
08:50:29 Join balintx_ [0] (
08:50:55 Quit balintx (Read error: Connection reset by peer)
08:54:50 Join ender` [0] (
09:13:05 Join kevku [0] (x@2001:470:28:773:babe:feed:dead:bee)
09:14:24 Quit kevku (Client Quit)
09:14:35 Join kevku [0] (x@2001:470:28:773:babe:feed:dead:bee)
09:26:39 Quit mc2739 (Ping timeout: 244 seconds)
09:26:42 Join petur [0] (~petur@rockbox/developer/petur)
09:27:00 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
09:29:27 Join utanapischti [0] (
09:32:32 Quit sasquatch (Ping timeout: 246 seconds)
09:48:49 Join antil33t [0] (
10:04:07***Saving seen data "./dancer.seen"
10:16:47*JdGordon throws a big squishy "Told you so"! at [Saint]
10:21:33 Join efyx [0] (
10:22:27 Join pamaury [0] (
10:22:27 Quit pamaury (Changing host)
10:22:27 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
10:29:29 Nick nef_away is now known as neferty (~andor@
10:32:22 Join markun [0] (~markun@rockbox/developer/markun)
10:33:00 Quit kevku (Quit: KVIrc 4.0.4 Insomnia
10:33:09 Join einhirn [0] (
11:09:41 Quit d3remark (Ping timeout: 276 seconds)
11:22:58 Quit sideral (Quit: Leaving.)
11:38:55 Quit kugelp (Ping timeout: 244 seconds)
11:41:09 Join kugel [0] (
11:41:09 Quit kugel (Changing host)
11:41:09 Join kugel [0] (~kugel@rockbox/developer/kugel)
11:55:49 Join sideral [0] (~sideral@rockbox/developer/sideral)
11:56:41 Join kugel_ [0] (~kugel@rockbox/developer/kugel)
11:59:32 Quit kugel (Ping timeout: 264 seconds)
12:04:10***Saving seen data "./dancer.seen"
12:08:45 Quit antil33t ()
13:32:07 Join kugel [0] (~kugel@rockbox/developer/kugel)
13:35:00 Quit kugel_ (Ping timeout: 240 seconds)
13:57:04 Quit Keripo (Quit: Leaving.)
14:02:06 Quit factor (Quit: Leaving)
14:02:43 Quit [Saint] (Quit: Imagination is for turbo-nerds who can't handle how kick-butt reality is. I'm a kick-butt reality master! I would rather die, than be imaginative. I mean that.)
14:03:26 Join krazykit [0] (~krazykit@
14:04:12***Saving seen data "./dancer.seen"
14:06:44 Join liar [0] (
14:08:21kugelZagor: irc log viewer issues back with FF5
14:08:43Zagorwell, it's not like I have changed anything
14:10:17Zagorif they choose to retire multipart/mixed push long before they add a replacement, there is not a lot I can do about it
14:10:17kugelright, so FF5 isn't in the blacklist yet
14:11:31Zagorah right, I added a blacklist
14:11:45Zagortry now
14:13:06*kugel can now uninstall the user agent switcher addon
14:13:08Zagorif ($ENV{'HTTP_USER_AGENT'} =~ m|Firefox/[4-9]|) { $push = 0; }
14:19:11 Join B4gder [0] (
14:19:11 Quit B4gder (Changing host)
14:19:11 Join B4gder [241] (~daniel@rockbox/developer/bagder)
14:37:58 Join benedikt93 [0] (~benedikt9@unaffiliated/benedikt93)
14:50:29 Join kugel_ [0] (~kugel@rockbox/developer/kugel)
14:50:44 Quit kugel (Read error: Operation timed out)
15:00:14 Join [Saint] [0] (
15:02:04 Quit ReimuHakurei_ (Read error: Connection reset by peer)
15:02:09 Join ReimuHakurei [0] (~reimu@2001:0:4137:9e76:38eb:5cbb:b58f:2bf0)
15:03:20 Join guest [0] (~4fba9942@
15:03:28 Quit guest (Client Quit)
15:03:51 Join guest_clive [0] (
15:05:21 Quit guest_clive (Client Quit)
15:06:43 Join evilnick_B [0] (0c140464@rockbox/staff/evilnick)
15:26:22 Quit ReimuHakurei (Read error: Connection reset by peer)
15:26:39 Join ReimuHakurei [0] (~reimu@
15:44:37 Quit sideral (Remote host closed the connection)
15:49:00 Join Guest5258 [0] (~sideral@
15:49:49 Quit shai (Ping timeout: 255 seconds)
15:53:12 Quit Guest5258 (Quit: Leaving.)
15:53:15 Nick kugel_ is now known as kugel (~kugel@rockbox/developer/kugel)
15:59:31 Quit mc2739 (Ping timeout: 264 seconds)
16:00:56 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
16:02:59 Join shai [0] (
16:04:13***Saving seen data "./dancer.seen"
16:05:14 Quit user890104 (Read error: Connection reset by peer)
16:06:39 Join user890104 [0] (
16:06:53 Join sideral [0] (~sideral@rockbox/developer/sideral)
16:15:01gevaertsZagor: is the builds file in www/buildserver in svn up to date?
16:15:27Zagorno. fixing.
16:16:07CIA-27New commit by zagor (r30020): Build system updates
16:18:59 Join Thra11 [0] (~thrall@
16:21:34 Join GeekShadow [0] (
16:21:34 Quit GeekShadow (Changing host)
16:21:34 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
16:21:35 Join petur2 [0] (~petur@rockbox/developer/petur)
16:22:10 Quit petur (Read error: Connection reset by peer)
16:26:04 Join kugel_ [0] (~kugel@rockbox/developer/kugel)
16:27:05 Nick petur2 is now known as petur (~petur@rockbox/developer/petur)
16:28:40 Quit kugel (Ping timeout: 246 seconds)
16:28:45 Quit B4gder (Quit: Konversation terminated!)
16:37:06 Join keyb_gr [0] (
16:42:27 Join captainkewler [0] (cf57a9a2@gateway/web/freenode/ip.
16:43:22 Join Thra11_ [0] (~thrall@
16:46:13 Quit Thra11 (Ping timeout: 250 seconds)
16:49:09 Join hobby16 [0] (~ubuntu@
16:53:37 Quit Thra11_ (Remote host closed the connection)
16:54:11hobby16hi all, I've written a patch to solve my usb problem. Now when USB is on, I can switch my Sansa between storage and power-alone modes with the On/Off slider.
16:55:36hobby16So I can upload a file (in storage mode), switch to power-alone mode, test my file on Sansa, reswitch to storage mode, reupload, etc... without connect/disconnect the cable
16:56:50hobby16if somebody is interested by this patch, let me know (don't know how to use rockbox's svn)
16:57:47 Join Thra11 [0] (
16:59:37hobby16Now my question is (again), how to hide a directory in storage mode (so the PC doesn't see a specific directory on the rockbox player)
16:59:50gevaertsYou can't
17:00:04gevaertsUnless you do weird FAT stuff
17:00:46Torneand if you do weird FAT stuff you are virtually guaranteeing data loss eventually ;)
17:01:22sideralwell, the Sansa OF does it on a regular basis, without being particularly prone to data loss
17:01:42Tornedoes it?
17:02:03sideralAFAIK it marks two directories with the volume-label bit in addition to the dir bit, which cause both Windows and Linux to ignore the directory
17:02:03*Torne suspects if you chkdsk it enough it will trash your files.
17:02:13pamaurythat's a weird fat stuff
17:02:20sideralyes it is :)
17:02:25Torneyes. that's something I would expect filesystem checking tools to "fix"
17:02:31pamauryif you have several labels, it's ok to trash them
17:02:31Tornequite plausibly by destroying data
17:02:53sideralinterestingly, neither chkdsk nor fsck.vfat seem to bother
17:03:24sideralbut I agree it's beyond spec and highly implementation specific
17:03:50 Part Zagor
17:04:05Slasherihmm, just increased my logf buffer size to 6 MB.. and commented out displayremote(), it would make logf basically freeze
17:04:26Torneanyway, we don't have any of the FAT structural awareness in the mass storage driver that would be required to actually do this
17:04:32gevaertsSlasheri: there's a patch around somewhere that makes logf log straight to a file
17:04:35Torneso it would be a big project whatever teh details of how you actualyl do it
17:04:51pamauryif it's just marking them with a bit, it's ok
17:04:58hobby16sideral : "AFAIK it marks two directories with the volume-label bit in addition to the dir bit". Nice tip sideral, where do you think I can find more infos about this ?
17:04:59pamaury(I think)
17:05:27Tornepamaury: Oh, I guess you could do it before unmounting
17:05:30Torneand change it back after
17:05:33sideralTorne: the way to implement this is to toggle the bits in the mount/umount code, not in the mass-storage driver
17:05:38Tornesideral: that's no fun ;)
17:05:48Tornerewriting on the fly is more entertaining ;)
17:05:50 Join ReimuHakurei_ [0] (~reimu@2001:0:4137:9e76:38eb:5cbb:b58f:2bf0)
17:05:59Slasherigevaerts: maybe that would be better, to log straight into file.. now the 6MB buffer is barely enough for me ;)
17:06:16 Quit ReimuHakurei (Read error: Connection reset by peer)
17:06:57sideralhobby16: with this information, you should be able to google something up. additional keywords are "##MUSIC" and "#PORT", the names of these Sansa directories. I don't have the specifics available at present
17:07:51sideralhobby16: I suggest to first read up on the FAT filesystem disk format, if you haven't done that already
17:07:53gevaertsSlasheri: there's FS #9253 but I believe saratoga once made a quick and dirty much simpler patch that does this too
17:07:54fs-bluebot logf file output patch. (patches, unconfirmed)
17:08:59gevaertsSlasheri: parts of
17:09:25hobby16@sideral, yessss, I'll try it. So if I didn't miss anything, no need to patch the firmware, just modify the directory control bits before the upload and mere mortals won't see the directory, right ?
17:10:43sideralSlasheri: I now seem to be able to trigger the DB-shows-trash bug I mentioned yesterday again. The runtime commit seems to be a precondition. (After regenerating the DB from scratch, I can now once more commit at runtime; probably my DB had accumulated too many deleted files, which previously prevented commit.)
17:10:50Tornehobby16: rockbox won't see it either
17:11:05Torneand you can't modify it *before* copying it onto the player..
17:11:12sideralTorne: Rockbox does see it.
17:11:18Tornesideral: Then we're broken
17:11:21 Quit bluebrother (Disconnected by services)
17:11:23 Join bluebroth3r [0] (~dom@rockbox/developer/bluebrother)
17:11:50sideralTorne: Hence the numerous "what are the ##music / ##port dirs for" questions :)
17:12:09sideralAh wait
17:12:23Torneis it actually still set at that point?
17:12:28sideralThat's because the bit has already been reset after the OF USB disconnect
17:12:35TorneThat's what I assumed.
17:12:48sideralOK, so I don't know whether we'd see it
17:13:09Tornehobby16: what is this actualyl for?
17:13:22hobby16sideral : "Rockbox does see it". That's perfect.
17:13:22Torneare you specifically wanting to hide something from a host pc but not from rockbox?
17:13:29Torneor just hidden entirely
17:13:31Torneor what
17:13:38sideralhobby16: You will need to patch Rockbox, but not the OF, for this to work
17:13:40Tornehobby16: we don't actually know :)
17:14:25 Quit fs-bluebot (Ping timeout: 255 seconds)
17:14:32*Torne has a look
17:15:35hobby16@Torne : I want to use rockbox to launch Lua script and only that. so want directories hidden to host pc (final users don't need to see), but visible to rockbox (of course)
17:15:48 Join fs-bluebot [0] (
17:16:19Torneok, so currently the code does only skip things if they are volume labels but not directories
17:17:54sideralhobby16: You can do it like the OF and toggle the bit in the mount/umount code. Note that we currently don't have a full-blown umount, as Rockbox doesn't need it right now, but FS #11883 may be of help
17:17:55fs-bluebot Close all files and safely unmount disks before entering USB mode, shutting down, or rebooting (patches, unconfirmed)
17:18:04hobby16@sideral : I don't want to patch the OF. But I thought hiding a directory but just changing its control bits ON THE HOST pc only was enough ? If patching RB is needed, does the patch already exist ?
17:18:30 Join Thra11_ [0] (~thrall@
17:19:11hobby16@sideral, oh, concurrent posts. Nice, i'll look at FS #11883, thank you !
17:19:12fs-bluebot Close all files and safely unmount disks before entering USB mode, shutting down, or rebooting (patches, unconfirmed)
17:19:18sideralhobby16: Carefully read what Torne said :)
17:19:27Torneit will work
17:19:53Torneyes, my sentence may have been ahrd to parse :)
17:20:07TorneRockbox will ignore the volume label bit being set, *if* it's a directory
17:20:17kugel_freeze ends today?
17:20:29Tornekugel_: freeze ends when AlexP creates the release branch :)
17:20:43 Join T44 [0] (
17:20:56Torneas long as you only intend to hide directories this way you can indeed just do it by copying the files on there and then opening the drive in a FAT editor and hacking the directory entry for the directory.
17:21:10 Quit Thra11 (Ping timeout: 240 seconds)
17:21:16Torneor a hex editor if you are studly enough
17:21:35 Quit fs-bluebot (Quit: timeout?)
17:21:35hobby16sideral: "Carefully read what Torne said :)" What specifically (better be sure before I grill my dear Sansa ;) )
17:21:38 Join fs-bluebot [0] (
17:22:00 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
17:22:26sideralthe last few sentences. He said you like won't need to patch Rockbox.
17:22:51sideralGood luck with your project!
17:24:19 Quit Topy (Ping timeout: 255 seconds)
17:25:05hobby16@sideral : do you confirm Torne's "hacking the directory entry for the directory" (no offense to you Torne, of course). If so, is the hidden dir likely to be corrupted if I paste a lot of files to the "hacked" filesystem on my player
17:25:27TorneYou can't access the directory after you do that
17:25:34TorneIt won't be visible any more
17:25:56TorneYou will have to clear the bit again before you can actually see it :)
17:26:04sideralI confirm everything Torne says. I learned most of my FAT32 from him. The hard way. ;)
17:27:16hobby16@Torne, No problem, I want it to be invisible. If needed, I reformat my player and upload again everything. Just wanted to know if changing files on the storage (with hacked directory) won't corrupt anything
17:27:25TorneChanging files won't
17:27:31TorneRunning chkdsk or similar might, potentially
17:27:57Torneyou dont' need to reformat it, you can just change the bit back again the same way you set it the first time :)
17:28:04hobby16Torne: "Changing files won't". Great, and how about dding new files or directory ?
17:28:20hobby16 Great, and how about Adding new files or directory ?
17:30:35sideralInside Rockbox, that should work as well. On the host, you won't be able to unless you remove the volume-label bit first
17:34:22hobby16Can you confirm (sorry to be so annoying instead of jumping the shark but I know nothing yet of FAT): I have / , /dir1, /dir2. I hack the storage to hide /dir1, now only / and /dir2 are seen on the host pc, and /, /dir1, /dir2 are seen by rockbox. And I can do all that just by hacking the storage thank to the host PC, no need to change the firmware ?
17:35:13Tornehobby16: you don't have to keep asking the same question phrased ten different ways
17:35:33Torne*none* of this is guaranteed to work at all, the host pc is perfectly entitled to crap its pants and destroy your filesystem
17:35:39Tornebut in practise everything will just work as you want
17:40:00 Quit GodEater (Quit: WeeChat 0.3.6-dev)
17:42:05 Quit Thra11_ (Remote host closed the connection)
17:42:20hobby16@Torne, ok ! Now, I'll stop asking and do the real thing, promise. THANK YOU for everything. Cheers
17:42:30 Join domonoky [0] (~Domonoky@rockbox/developer/domonoky)
17:43:13 Join Thra11 [0] (~thrall@
17:46:48 Quit Thra11 (Remote host closed the connection)
17:47:24 Join Judas_PhD [0] (
17:53:34CIA-27New commit by miipekk (r30021): Fixed a changelog export/import problem with tagcache where <CR> or <LF> ...
17:55:17Slasherisideral: now my worst annoyance with db fixed :)
17:57:50CIA-27r30021 build result: All green
17:59:12 Join Thra11 [0] (~thrall@
17:59:17 Quit petur (Quit: *plop*)
18:00:30 Quit Thra11 (Remote host closed the connection)
18:04:15***Saving seen data "./dancer.seen"
18:13:13sideralSlasheri: Whops, CRLF in tags... Nice catch! Was that in a comment tag?
18:13:48sideralSlasheri: seen my note about DB junk reproducibility?
18:17:50 Join Thra11 [0] (~thrall@
18:19:51Slasherisideral: yep, that was a comment tag :)
18:20:04Slasherisideral: i saw that, i will look into it soon
18:26:04sideralSlasheri: Thanks!
18:33:24 Quit Thra11 (Remote host closed the connection)
18:38:14 Quit GeekShadow (Read error: Connection reset by peer)
18:38:34 Join GeekShad0w [0] (
18:40:51 Join wtachi [0] (
18:48:20 Join Thra11 [0] (
18:53:05 Join kugel [0] (
18:53:05 Quit kugel (Changing host)
18:53:05 Join kugel [0] (~kugel@rockbox/developer/kugel)
18:56:18 Quit kugel_ (Ping timeout: 260 seconds)
18:57:47 Quit sideral (Quit: Leaving.)
18:58:09 Quit Thra11 (Remote host closed the connection)
18:58:29 Join sideral [0] (~sideral@rockbox/developer/sideral)
19:00:16 Quit sideral (Client Quit)
19:00:26 Join sideral [0] (~sideral@
19:00:26 Quit sideral (Changing host)
19:00:26 Join sideral [0] (~sideral@rockbox/developer/sideral)
19:02:42Slasherisideral: so you re-init db (commit without rebooting), and then you do something to get trashed query results?
19:03:46sideralAlmost. The steps seem to be:
19:04:23 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
19:04:43 Join Thra11 [0] (
19:04:57sideralHave autoupdate enabled. Add some tracks to the player via USB (using the OF's USB mode in my case). Boot Rockbox. Wait for autoupdate / commit to complete. Enter DB browser, verify the new tracks are there. Then... some track from the DB and play it. After some time, go back to the DB browser. Now it shows funny results
19:06:19Slasheriok, i will try that :)
19:08:47sideralThe queries I'm using all have some condition set (e.g., require a specific directory) and the tracks that are shown don't belong in there; or there are no tracks shown at all
19:09:54Slasherihmm.. i should really get a bigger disk for my H140..
19:10:43Slasherisideral: does that issue occur even after you reboot the player?
19:11:33sideralNo. When I reboot after the issue has occurred, everything is normal. So it seems to have to do something with the DB update
19:11:34AlexPkugel: Yep, shortly
19:11:36 Join ChickeNES [0] (~ChickeNES@
19:12:14Slasherisideral: ok, interesting
19:12:37 Join casainho [0] (
19:14:14Slasherisideral: just added two new albums, and going to database - Same as current... and all items work as expected (playing those newly added tracks)
19:15:02sideralI somehow expected that. :)
19:15:40sideralDoes recently played / recently added look sane to you?
19:16:02hobby16hi sideral, I've tried to hide a directory like described above by setting its volume-bit to 1 (file attribute offset is changed from 0x10 to 0x18). I use a hex editor to do it but it replies "access denied". I've tried it on rockbox and a simple USB key, same (non) result. Do you have any idea ?
19:16:27Slasheriyes, at least so far everything is working.. but i will leave the iriver playing for a while
19:17:28 Join JessicaRN [0] (~JessicaRN@
19:17:43Tornehobby16: your hex editor probably doesn't know how, or doesn't ahve permission to, write to the raw disk
19:18:21hobby16The hex editor is HxD (windows freeware), do you know anything else which can write to raw disk Torne ?
19:18:34sideralSlasheri: thanks. I'll try reproducing this as well to verify the steps. Then I can try a simulator build
19:18:41JessicaRNhi all. I've got an old archos recorder jukebox w/ a 20gb drive in it. I'm about to up it to a 120gb disk. Will the automated installed prep the new disk, or do i need to partition it myself?
19:19:02Slasherisideral: it would be great if you could manage to reproduce it using the simulator
19:19:08TorneJessicaRN: installing rockbox pretty much always requires that the original firmware be functional first
19:19:35JessicaRNso, i cant start w/ a scrubbed disk?
19:19:53gevaertsIt will need to be partitioned and have a filesystem
19:20:06JessicaRNfat32? or can i use ntfs?
19:20:15CIA-27New commit by alex (r30022): Branch for 3.9
19:20:28AlexPThere we go :)
19:20:51JessicaRNugh. I have an xp laptop i'm trying to build this on. it will only let me creat an ntfs disk. any ideas?
19:20:52sideralAlexP: yay!
19:21:06gevaertsI suspect you can do the partitioning over usb after installing the disk
19:21:40Topic"Now branched for 3.9 | Rockbox accepted for GSoC 2011 | Please read before speaking: | Please direct offtopic/social chat to #rockbox-community | This channel is logged at | An upgrade caused issues with the wiki. We are on the case" by ChanServ (ChanServ@services.)
19:21:40TorneJessicaRN: you need something other than windows's built in format tool to format something >32GB as FAT
19:21:50sideralSlasheri: can you try again and delete some tracks at the same time as well as adding some? I think my sync process typically removes and adds some
19:22:20 Quit soap (Ping timeout: 240 seconds)
19:22:20JessicaRNtorne: what freeware util do you suggest? or should I just boot ubuntu and do it from there?
19:22:45gevaertsJessicaRN: I believe people have successfully used something called swissknife
19:23:41AlexPor fat32format
19:23:44AlexPboth work
19:23:50JessicaRNi'm always a bit spooked about using freeware that hasn't been vetted. just asking for viruses that way.
19:24:08AlexPA linux liveCD then :)
19:24:21Slasherisideral: ok, lets try
19:24:24JessicaRNHey! I'm carefull. I only surf the web with condoms on my fingers
19:26:11AlexPHmmm, some RC builds would probably be good
19:26:53JessicaRNthanks. i'll be back
19:26:56 Quit JessicaRN (Quit:
19:27:01 Join Horscht [0] (
19:27:01 Quit Horscht (Changing host)
19:27:01 Join Horscht [0] (~Horscht@xbmc/user/horscht)
19:29:49Slasherisideral: now playing & browsing while tagcache is working to mark deleted files and commit new ones
19:30:04 Join kugel_ [0] (~kugel@rockbox/developer/kugel)
19:32:08 Quit casainho (Remote host closed the connection)
19:32:26CIA-27New commit by alex (r30023): Bump release scripts to 3.9RC
19:32:36 Quit kugel (Ping timeout: 240 seconds)
19:33:11Slasherisideral: ok, now "same as current directory" displays results from wrong directory, every other query works ok
19:33:38Slasherithere maybe dircache pointer invalidation problem or something like that
19:33:42sideralSlasheri: ah, glad you found something
19:34:40sideralif that's the case, that could perhaps explain the dreaded duplicate entries in DB problem as well
19:35:42Slasherimaybe.. now lets find out what could cause that
19:38:04Slasherisideral: and "recently played tracks" now halted the whole query
19:39:06Slasherioh, no.. it did work, it just took a long time (ram cache was not used for that query)
19:40:49 Join kugel [0] (~kugel@rockbox/developer/kugel)
19:41:45 Join robin0800 [0] (~quassel@
19:43:53 Quit kugel_ (Ping timeout: 250 seconds)
19:47:50sideralSlasheri: I can't seem to trigger it right now, but just noticed that I left out the play a track part of my recipe
19:49:29Slasherisideral: i have the problem occurring only with the filename tag (that tag has special handling because it will never be loaded to ram, but it could point to the dircache)
19:49:36 Quit benedikt93 (Quit: Read the fucking binary.)
19:50:11kugelpamaury: I'll commit the dircache patches soon, want to have look before?
19:50:39kugelSlasheri: ^
19:50:45sideralSlasheri: That could explain why it mostly affects queries with a filename condition
19:51:08Slasherikugel: sure :)
19:51:48 Quit sideral (Quit: Leaving.)
19:51:55 Join sideral [0] (~sideral@
19:51:55 Quit sideral (Changing host)
19:51:55 Join sideral [0] (~sideral@rockbox/developer/sideral)
19:52:01 Quit hobby16 (Ping timeout: 258 seconds)
19:52:44 Quit ChickeNES (Quit: Computer has gone to sleep.)
19:53:33 Join soap [0] (
19:53:33 Quit soap (Changing host)
19:53:33 Join soap [0] (~soap@rockbox/staff/soap)
19:54:48 Join Buschel [0] (
19:54:49 Join JessicaRN [0] (
19:55:44 Quit sideral (Client Quit)
19:56:02JessicaRNback again! I just formatted the new disk as 120gb fat32. do I just need to unpack the install files to the root of the new volume?
19:56:08 Join sideral [0] (~sideral@rockbox/developer/sideral)
19:56:08 Quit captainkewler (Ping timeout: 252 seconds)
19:56:21 Quit sideral (Client Quit)
19:56:30 Join sideral [0] (~sideral@
19:56:30 Quit sideral (Changing host)
19:56:30 Join sideral [0] (~sideral@rockbox/developer/sideral)
19:58:06kugelSlasheri: sure have a look or sure commit? :)
19:59:12Slasherikugel: those patches alters the memory layout, removing linked lists?
19:59:30kugellinked lists are not removed
20:00:02Slasherioh.. what patches are you going to commit?
20:02:12kugelSlasheri: all FS#of 12133
20:02:18kugelFS #12133
20:02:19fs-bluebot Dircache patches (patches, new)
20:03:04 Quit JessicaRN (Quit:
20:04:16***Saving seen data "./dancer.seen"
20:08:50 Join fyrestorm [0] (
20:09:04 Join kugel_ [0] (~kugel@rockbox/developer/kugel)
20:11:40 Quit kugel (Ping timeout: 240 seconds)
20:12:47 Quit robin0800 (Read error: Connection reset by peer)
20:13:10pamaurykugel_: I'll have a look today !
20:14:15kugel_pamaury: don't promise, that encourages me to wait :)
20:16:19 Nick kugel_ is now known as kugel (~kugel@rockbox/developer/kugel)
20:16:33pamauryI'm making coffee and have a look just after so you won't wait too long hopefully :)
20:18:48 Quit Thra11 (Remote host closed the connection)
20:21:45kugeldircache uses still pointers internally, but it could also use array indices there, however I'm not sure if it's worth it
20:22:09BuschelAlexP: freeze on Trunk is over?
20:22:23AlexPBuschel: yep
20:22:30pamaurykugel: you want comments on FS ?
20:22:56Buschelok :)
20:23:04kugelpamaury: just drop them here, the task is (hopefully) closed soon :)
20:24:10 Quit user890104 ()
20:24:35pamauryOk, I think I already told you patch 1, 2 & 3 were ok and suggested an optimization
20:24:46pamaurycan I have a look at 8 without 4-7 ?
20:25:46pamaury8 looks ok but I would just add a small comment to say the '1' comes from the '/'
20:27:03pamaury4 is ok I guess
20:28:03kugelpamaury: is dircache_copy_path() after 8 if it helps
20:28:43pamauryah ok, there already is a comment
20:29:29pamaurypersonal comment: I don't like the indentation of 5 with respect to global variables
20:32:43kugelpamaury: it looks better in the source than in the diff :)
20:37:08Slasherisideral: it could be that tagcache itself is working correctly but wrong path is given by playlist or some other subsystem. But that is just speculation and needs to be verified
20:37:38Slasheriat least playlist uses dircache to some extend
20:37:55CIA-27New commit by buschel (r30024): Submit first part of FS #12154. Whitespace removal and correction of several defines in WM8758 driver according to data sheet, no functional change. ...
20:39:20pamaurykugel: should offset_d_names be of type ptrdiff_t ?
20:39:34sideralSlasheri: It could be that I had a deleted file in my current dynamic playlist when the problem occurred. I have no other playlists on the system. But did you have any playlist related to the deleted file?
20:40:41pamaurykugel: In your commit, "since" should be "Since" ;)
20:40:52Slasherisideral: i just checked and most likely playlist doesn't handle dircache pointer invalidation.. it keeps running using the old pointers
20:41:13Slasherithat could explain the problem, unless you stopped playback and created a new dynamic playlist
20:41:47CIA-27r30024 build result: All green
20:41:51*bluebroth3r would really like to see people to wrap commit messages at 72 chars or so
20:42:13Slasheripamaury: because during commit() dircache gets rebuilt, and old pointers can be incorrect
20:42:33bluebroth3ris it possible to rewrap all "long" commit messages when converting to git in a sensible way? Some existing ones are kinda unreadable on the terminal right now.
20:43:43 Quit ps-auxw (Quit: leaving)
20:43:52sideralSlasheri: yep. that whole concept of breaking some other subsystem's pointers seems rather brittle to me. That's one reason why I'm scared by kugel's plans for moving allocations around
20:44:20 Quit kugel (Ping timeout: 240 seconds)
20:44:21 Join boghog [0] (~aphax@2001:980:34c7:0:1e6f:65ff:fe86:1e03)
20:44:48pamauryarg, kugel disappeared :(
20:46:54 Join kugel [0] (
20:46:54 Quit kugel (Changing host)
20:46:54 Join kugel [0] (~kugel@rockbox/developer/kugel)
20:47:21pamaurykugel: did you see my comments ?
20:48:12pamaurykugel: in the /** compact the dircache buffer **/ part, I don't really like the way you structure you if statement. Since the function just return if res<0, if you just if(ret < 0) return; to make the code clearer
20:48:22pamaurysame thing for if (offset > 0)
20:48:33kugelone moment :)
20:50:38 Join ps-auxw [0] (~arneb@2001:470:c807:0:1532:4e5f:2ad3:4123)
20:51:00pamauryis there a significant difference in performance between memmove and memcpy which justify such a distinction ?
20:51:11pamaurymemove will always work no ?
20:51:53kugelbut it's nearly 0% probability that they overlap
20:52:06pamaurybut memmove will see that
20:52:32CIA-27New commit by buschel (r30025): Submit final part of FS #12154. Optimization for WM8758 used in iPod Video: Significantly reduce popping noise during startup. Additionally improve ...
20:53:39pamaurykugel: what exactly compaction does ?
20:55:13kugelmove strings down
20:55:29kugelthey are allocated from the end of the audiobuf
20:55:31pamauryso that the reserve size is exactly DIRCACHE_RESERVE
20:55:51CIA-27r30025 build result: All green
20:56:19pamauryYou should make this clearer
20:56:30pamaury(that there are allocated from the end of audiobuf)
20:57:15pamauryAdd a line to the comment just before audiobuf = (char*)(((intptr_t)audiobuf & ~0x03) + 0x04); for example
20:57:18kugelpamaury: you mean like in the comment above?
20:57:36kugel /* struct dircache_entrys are allocated from the beginning, * their corresponding d_name from the end
20:57:40 Join soap_ [0] (~soap@rockbox/staff/soap)
20:57:51pamauryyeah, be a bit more explicit
20:58:00pamauryfrom the end *of the audiobuf*
20:58:07kugelthe comment above the allocation is quite clear if you ask me
20:59:08pamauryIf you ask me, it is not :) In your comment you don't mention audiobuf, so you have to look at the code to understand
20:59:30kugelcomments shouldn't duplicate code
20:59:53pamauryBut mentioning audiobuf add a few characters and makes it easy and explicit
21:00:10 Join ChickeNES [0] (~ChickeNES@
21:00:21 Quit soap_ (Client Quit)
21:00:38kugelaudiobuf is mentioned 2 lines below in the code...
21:01:44kugelalso, I changed the code according to your suggestion (with the if()s). I can't say it looks nicer
21:02:14pamauryI believe it does, you gain two levels of indentation
21:02:28pamauryotherwise you have you far away else statement which just returns
21:02:36pamaurywhich is clearly an exception case
21:04:14pamauryAlso don't forget that the code looks clear to you because you modified it a lot; but after a few months, you'll have forgotten the details and having redundant comments is a time win.
21:04:34 Quit sideral (Quit: Leaving.)
21:06:05kugelpamaury: how about this?
21:06:41kugelthe function (no diff):
21:06:41pamauryperfect :)
21:07:21GodEater_if only we had a patch review system ;)
21:07:31pamauryAppart from those comments on the style, it look perfectly ok to me
21:08:29pamauryI must admit that the entries vs names distinction makes the code nicer
21:08:42kugelpamaury: there's one problem with ptrdiff_t, it's (theoretical) only defined for pointers within the same array
21:08:54kugelintptr_t covers the entire memory
21:10:00pamauryhum, true; but since the result is a difference between pointers, it's weird :) Do it the way you like anyway, we know it's a 32-bit system anyway :D
21:10:03kugelIIUC anyway
21:10:38pamaury6 introduces no functional changes so I guess it's no problem
21:12:30kugelintptr_t is an integral type as large as pointers, ptrdiff_t is an integral type large enough to hold differences of pointers to the same array. I don't think they make a difference in practise
21:12:47kugelbut intptr_t is always correct :)
21:13:52pamauryok, you win :)
21:14:17amiconnIsn't it branching day today?
21:14:29Torneamiconn: done already :)
21:14:43kugelpamaury: I don't really care actually
21:14:48pamauryneither me
21:15:05amiconnAh, missed that one
21:15:24*amiconn thinks branch commits should also appear on the frontpage
21:16:27pamaurypatch 7 looks ok. I assume you tested it. I'm happy that it finally don't expose dircache pointers.
21:17:22pamauryHowever, it doesn't check for full bounds on indexes (0<=index<entry_count); don't know if it's usuful but it might catches some difficult bugs
21:17:40kugelI tested somewhat yes, not extensively as I don't use targets with dircache anymore :)
21:18:02 Join MethoS- [0] (~clemens@
21:18:50pamauryok, so overall, that's all ok as long as it's correctly tested
21:19:09kugelpamaury: well, as you're supposed to pass in IDs that you previously got from the dircache (and this is how it's used) bound checking is a bit redundant
21:20:29kugelyou don't pass IDs that you just made up
21:21:37amiconnkugel: Did you have a look at r30019?
21:21:53 Join sideral [0] (~sideral@rockbox/developer/sideral)
21:22:32kugelamiconn: yes
21:22:49amiconnI hope my way to fix .talk is ok?
21:23:29kugelyes, should be. it appears the file extension was previously ignored
21:23:52kugelso whitelisting .talk for mp3 should be alright
21:24:17amiconnYes, because pre-r29432 mp3info() was specifically mpeg audio only
21:24:53amiconnNow it's actually a generic metadata function which checks extension before probing further
21:25:01kugelwell it still is but now calls common code instead
21:25:08amiconnNo it's not
21:25:24amiconnYou changed it from calling get_mp3_metadata() to calling get_metadata()
21:25:55kugelwhat I said
21:25:59amiconnIt's ok even for the old hwcodec engine, but that's why .talk must be whitelisted
21:26:58amiconnOn hwcodec talk clip queueing needs to check metadata in order to skip tags, vbr headers & stuff. On swcodec .talk clips are raw speex with no tags or headers at all
21:27:05*kugel prepares for a big pile of commits
21:27:27amiconnThat's why this wasn't a problem on swcodec
21:27:37 Quit sideral (Quit: Leaving.)
21:27:47 Join sideral [0] (~sideral@
21:27:47 Quit sideral (Changing host)
21:27:47 Join sideral [0] (~sideral@rockbox/developer/sideral)
21:27:52kugelgreat it works again
21:30:37 Quit sideral (Client Quit)
21:30:54amiconnThe metadata functions could really profit from a thorough cleanup
21:30:58 Join ReimuHakurei [0] (~reimu@
21:32:34*amiconn wonders when we'll see the score table working again :\\
21:32:44CIA-27New commit by kugel (r30026): Move buffer.h to firmware/include.h to replace a useless malloc header.
21:32:49CIA-27New commit by kugel (r30027): Use enum themable_icons in struct file_type and struct filetype (who made those names?). ...
21:32:53 Quit ReimuHakurei_ (Read error: Connection reset by peer)
21:32:53CIA-27New commit by kugel (r30028): Move struct search_instruction into struct menu root, as they're not allocated or used independently.
21:32:58CIA-27New commit by kugel (r30029): tagtree: Refactor memory allocation to local functions.
21:33:05CIA-27New commit by kugel (r30030): Remove unused code path from playlist_create_ex().
21:33:10CIA-27New commit by kugel (r30031): Fix yellow. Shouldn't have reordered struct members here.
21:33:25 Quit rudi_s (Quit: leaving)
21:35:15CIA-27r30026 build result: All green
21:35:33*gevaerts takes bets on the accuracy of that "fix yellow" :)
21:35:59kugelthe build system might jump over :)
21:38:14 Quit GeekShad0w (Read error: Connection reset by peer)
21:38:28CIA-27r30031 build result: All green
21:38:44 Join GeekShad0w [0] (
21:38:55Slasherikugel: tagtree.c, } si;
21:39:03Slasheriwhy did you add that to the struct definition?
21:39:52kugelSlasheri: so that only one allocation is needed
21:39:52bertrikkugel, can't you squash/fixup those fixes before committing?
21:40:27kugelbertrik: I can, but it's a lot more effort
21:41:04 Join Thra11 [0] (~thrall@
21:41:47Slasherikugel: hmm, i don't see the point how that would make any difference
21:42:12 Quit Thra11 (Remote host closed the connection)
21:42:58kugelwhy should they be seperate if they're tied together in reality? it just adds overhead
21:43:06kugeland less allocations are good for my gsoc project
21:43:12 Join rudi_s [0] (
21:43:37Slasherikugel: i didn't mean that, but your extra "si"
21:44:35kugelI don't understand
21:45:13kugelanonymous nested structs are gnu(/ms) extensions, not sure if we're fine with them
21:45:16 Quit Xerrao (Read error: Connection reset by peer)
21:45:53Slasheriit should work perfectly fine.. those are used everywhere
21:46:02 Join Thra11 [0] (~thrall@
21:46:17kugelthat needs more changes in the code too, I wanted to keep the diff smaller
21:46:43kugelbertrik: and by a lot, I mean significantly more effor
21:47:24Slasherikugel: hmm, sorry, i think i misread the diff :) need to check again
21:47:58bertrikoh I thought git with git-svn should make that very simple
21:48:02 Join fml [0] (
21:48:21kugelbertrik: it's simple if you need to fix the top-most commit, not the ones below
21:48:28Slasherikugel: ah, yes.. i didn't notice you nested the struct by reading the diff
21:48:39fmlWhy isn't the new search box in the online manual displayed on the title page?
21:48:53bertrikit annoys me a bit to have basically dummy commits for that in SVN
21:49:59kugelbertrik: I don't think it matters
21:50:03Slasherii just don't know if that kind of nesting is a good idea.. i would declare the struct outside and allocate it inside the other struct
21:50:06kugelnot to me anyway
21:50:23bertrikgreat, that annoys me even more. I stopped my build client, I'm not compiling crap like that.
21:51:25pixelmafml: I don't think that this was a design decision (either an oversight or harder to do than expected). I remember bluebroth3r cursing htlatex a few times in the process
21:51:29kugelisn't that a bit of an overreaction?
21:52:44*pixelma still wonders if she's the only one who can't listen to music with an r30012 android build
21:53:46 Join Xerrao [0] (
21:54:27kugelpixelma: I have problems as well, but I suspected a patch I have applied
21:56:36 Quit fml (Quit: ChatZilla 0.9.87 [Firefox 4.0.1/20110413222027])
21:59:01Buschelis anyone working on the fPIC and visibility warnings that come up when compiling a win32 simu under ubuntu? I see hundreds of that when building.
22:00:19Buschelseems to be connected to r29983
22:01:22EspreonAlexP: Well, it's Monday now.
22:01:31TorneBuschel: I don't think so, but I keep telling people what needs doing :)
22:01:38AlexPEspreon: Indeed it is
22:01:43Tornethose options should only be used when the target is linux
22:01:45AlexPAnd I still have a real life
22:01:46Torneor possibly mac
22:01:48Tornei'm not sure about mac
22:01:55Tornebut it shouldn't be used on windows (or on targets)
22:01:59AlexPBut don't worry, it'll get into the release
22:03:09EspreonAll right then...
22:03:29AlexPBelieve me or not, I don't really care
22:04:20***Saving seen data "./dancer.seen"
22:04:46EspreonVery well.
22:04:47 Part Espreon
22:07:06 Quit Thra11 (Remote host closed the connection)
22:07:32 Join kugel_ [0] (~kugel@rockbox/developer/kugel)
22:07:37CIA-27New commit by kugel (r30032): Dircache: Remove dircache_entry::name_len. ...
22:08:07kugel_something went wrong
22:10:01 Join Thra11 [0] (~thrall@
22:10:04 Quit kugel (Ping timeout: 255 seconds)
22:10:51CIA-27r30032 build result: All green
22:11:06BuschelTorne: commenting line 229 in tools/configure does it for me
22:11:21 Join JessicaRN [0] (
22:11:57TorneBuschel: yes but that breaks it for linux
22:12:09TorneI just said: those options need to be passed when building for linux, but not windows
22:12:12CIA-27New commit by kugel (r30033): Dircache: Rework and simplify dircache_copy_path(). ...
22:12:16CIA-27New commit by kugel (r30034): Dircache: Return the size of the result string in dircache_copy_path() so that callers don't need to call strlen on it.
22:12:23Tornei don't have time to actually do this atm :)
22:12:25Tornebut feel free
22:12:27CIA-27New commit by kugel (r30035): Fix dircache_size calculation bug introcuced 3 revisions ago.
22:12:33CIA-27New commit by kugel (r30036): Dircache: Change internal cache layout. ...
22:12:37CIA-27New commit by kugel (r30037): Dircache: Move struct maindata declaration to dircache.c and actually check for DIRCACHE_MAGIC when loading from disk.
22:12:44CIA-27New commit by kugel (r30038): Dircache: Don't expose struct dircache_entry and pointers into the cache, use IDs instead. ...
22:12:50CIA-27New commit by kugel (r30039): Optimize new dircache_copy_path so that the helper (strlcat) doesn't need to walk through the entire string repeatedly. ...
22:12:54CIA-27New commit by kugel (r30040): Dircache: A bit of follow-up code cleanup suggested by Amaury Pouly.
22:12:59CIA-27New commit by kugel (r30041): Use ALIGN_UP() macro for alignment, in a more correct way also.
22:13:09JessicaRNto all of you that have helped develop RB, THANK YOU! It's a great program, and you've kept my spiff Archos Jukebox Recorder working fantastic for me!!!!
22:13:24BuschelTorne: I am not really an expert to build processes... at least I now have a working workaround. sorry to pass this back to you
22:14:03JessicaRNAlso, TYVM to the folks who support RB in this room. You all are great as well!
22:14:26bluebroth3rfml: because the crosslinks block where it's attached to isn't shown on the title page. Feel free to move it around :)
22:14:48CIA-27r30033 build result: All green
22:15:07 Quit Sundiver (Ping timeout: 264 seconds)
22:15:26bluebroth3rthough I'm not sure if the search box is that useful after all −− when I've tried it today Google seemed to only give me pdf results, and I haven't figured how to make Google not include pdf results
22:16:07*Buschel really liked to read the change logs, but this is getting noisy now
22:16:11*amiconn wonders whether this 'all green' is real
22:16:42amiconnOne stillf can't verify that at all :\
22:17:56 Join kugel [0] (
22:17:56 Quit kugel (Changing host)
22:17:56 Join kugel [0] (~kugel@rockbox/developer/kugel)
22:18:06CIA-27r30041 build result: All green
22:18:58amiconnAnd now the delta table became rather sparseƟ!
22:19:54bluebroth3rthe build result table is still stuck at r30006
22:20:02 Quit JessicaRN (Quit:
22:20:37Buscheln1s, Torne: I re-opened FS #12144
22:20:38fs-bluebot Use -fPIC -fvisibility=hidden for 32 bit x86 builds too to fix link errors. (patches, new)
22:20:39kugelbluebroth3r: but IIRC CIA-27 reports correctly
22:20:41bluebroth3rBagder? Or is that Zagor's thing?
22:21:07 Quit kugel_ (Ping timeout: 264 seconds)
22:21:14kugelBuschel: can you post the build log?
22:21:18Bagderthat sounds like work so it must be Zagor's thing =)
22:21:42Buschelkugel: yep, wait a minute
22:21:57CIA-27New commit by alex (r30042): Add US English by Steven Panek.
22:22:21kugelBuschel: IIRC visibility isn't even supported for mingw targets
22:23:02Buschelwhat Torne said a few minutes before, I just don't know how to fix /tools/configure
22:23:45CIA-27New commit by alex (r30043): Add US English by Steven Panek.
22:24:43kugelI wonder if it made more sense to only translate strings that are actually different in the US lang file
22:25:06kugelbut I suspect doesn't work nicely with translate.r.o?
22:25:11bluebroth3rpixelma: current svn is broken for me too :(
22:25:12AlexPBut then it'd mess up the stats :)
22:25:57kugelalso, one more language...Torne, how's your genlang work coming along? :)
22:26:06Tornenot had time yet ;)
22:26:13CIA-27r30043 build result: All green
22:26:20Tornei think i know how to do it though
22:26:43 Quit domonoky (Ping timeout: 255 seconds)
22:26:51kugeljob half done!
22:27:01bluebroth3rRockbox does play the file, but no audio output :(
22:27:34 Join domonoky [0] (~Domonoky@rockbox/developer/domonoky)
22:27:44bluebroth3rsometimes I need to skip around in the file before it starts "playing"
22:27:55*bluebroth3r wonders when playback broke again
22:28:09AlexPSometime in the last half an hour probably :)
22:28:56kugelAlexP: are you saying we were bug-free before that? ;)
22:29:06AlexPyep, 100% :)
22:29:55Buschel"make > log.txt" does not pipe compiler warnings to the file. what is the missing parameter?
22:30:10kugelBuschel: "make &> log.txt"
22:30:16kugel> does only stdout
22:30:33bluebroth3rwow. It _does_ playback the file but it's extremely soft.
22:30:45bluebroth3rdoesn't seem to happen for mp3, only ogg
22:31:17Buschelthanks, works now
22:31:46bluebroth3rpixelma: what filetype are you playing?
22:32:28Buschelwow, the log file is 1.8 MB
22:34:53Buschelkugel: zipped and attached to FS #12144
22:34:54fs-bluebot Use -fPIC -fvisibility=hidden for 32 bit x86 builds too to fix link errors. (patches, new)
22:34:54CIA-27r30043 build result: 325 errors, 0 warnings (alex committed)
22:36:16AlexPoi, that was green a minute ago!
22:36:52 Join hobby16 [0] (
22:37:21*AlexP kicks kugel instead
22:37:28AlexPThat clearly isn't my fault :)
22:37:37 Join Zagor [242] (~bjst@rockbox/developer/Zagor)
22:38:05bluebroth3rnot building each individual revision might not be that good it seems
22:38:23Zagorbluebroth3r: it wasn't intentional
22:38:40bluebroth3rZagor: ah, that explains things. I was surprised by that change ;-)
22:39:01bluebroth3ranyone else having problem with playing non-mp3 formats on Android with current svn?
22:39:13Zagorwhen I fixed the delta table I broke the compile results. only I didn't notice until recently.
22:39:16hobby16can someone tell me more about usb-serial mode (for debugging) please ? does it work on Windows (I can't get USB work on my Ubuntu Virtualbox, grrrr)
22:39:49hobby16maybe gevaerts ?
22:40:21gevaertshobby16: I don't think anyone ever got it to work in windows
22:40:30gevaertsFeel free to fix that though :)
22:40:49kugelbluebroth3r: on my phone it crashes in the dynamic linker, but again I'm not sure if my other patch causes this
22:41:00kugelI believe my problematic files are also non-mp3
22:41:20*bluebroth3r wonders if other targets are affected as well
22:41:52Buschelgoot go now, see you tomorrow
22:41:58 Quit Buschel (Quit: ChatZilla 0.9.87 [Firefox 3.6.17/20110420140830])
22:43:16hobby16gevaerts @, oh no, I'm electronician, fixing USB is way beyond my capabilities. So I'll have to find a linux with USB.
22:43:56kugelbluebroth3r: let's hope not. broken playback wouldn't be a compelling feature for 3.9
22:44:23bluebroth3rwell, let's see. I'm building for e200 now ...
22:44:41AlexPIs it broken in the 3.9 branch?
22:44:55bluebroth3rand it's not broken, it's just like ... playing at -80dB or something like that :)
22:45:08AlexPIs it doing that in the 3.9 branch? :)
22:45:27bluebroth3rwe'll see.
22:45:40CIA-27New commit by kugel (r30044): Fix red in fft plugin.
22:45:50AlexPI'm still doing RC builds, it is taking forever
22:45:58bluebroth3rbut we can't upgrade Android to unstable with that issue IMO
22:46:20bluebroth3rhow good the fft plugin isn't built on Android
22:46:34*bluebroth3r rebases
22:49:31CIA-27r30044 build result: All green
22:50:14bluebroth3rseems to be fine on e200
22:50:28bluebroth3rat least ogg files play at normal volume so fwar
22:51:01bluebroth3rwhy would something happen on Android only?
22:51:36*bluebroth3r grumbles
22:59:13 Quit ChickeNES (Quit: Computer has gone to sleep.)
23:00:24 Quit Xerion (Read error: No route to host)
23:00:39bluebroth3rwe really need a patch cleanup month. There are quite a couple of nice ideas around that would only need someone to finish them
23:01:14kugelthat assumes people participate
23:01:23AlexPI thought we tried that and nobody did anything
23:02:07kugelwe can have that if you don't force people to stop developing other stuff in the meantime
23:02:27kugelotherwise it'll just result in lots of sad developer faces :(
23:02:35bluebroth3rthere are a couple of patches I'd like to give a look. I just need to get around doing so
23:02:55 Join Espreon [0] (~espreon@wesnoth/developer/espreon)
23:02:56kugelbluebroth3r: start your very personal patch cleanup month :)
23:02:58bluebroth3rwe need to keep our house clean, even if that means not working on cool new features for a couple of days ...
23:03:04 Join kevku [0] (x@2001:470:28:773:babe:feed:dead:bee)
23:03:41kugelI don't think I'm going to work on stuff I don't want to work on
23:03:54bluebroth3rkugel: I'm already at it. I just need less headache tonight and more time in general
23:04:22kugelbut see, that's what we discussed on the devcon. the open patches are *not* our responsibility
23:05:02bluebroth3rthat doesn't mean it's not a good idea to go through them and try to finish some
23:05:22kugelit's not a good idea to force us to do it
23:05:29 Quit GeekShad0w (Ping timeout: 240 seconds)
23:05:53EspreonAlexP: You forgot to set svn:keywords on the file. Also, I made an update. Shall I pastebin the patch?
23:06:20bluebroth3rplus, we haven't communicated that policy in the past. And a lot of patches haven't been closed simply because there are good ideas.
23:06:28kugela patch cleanup weak is quite contrary to what we decided on the devcon
23:06:34bluebroth3rEspreon: svn:keywords will go away pretty soon
23:06:46bluebroth3rkugel: I disagree.
23:06:47kugelor month even
23:06:48AlexPI didn't forget them :)
23:07:00AlexPEspreon: And yes, pastebin is fine
23:07:11kugelbluebroth3r: it's making us responsible for the patches
23:07:17kugelwhich is not something we want
23:07:27bluebroth3rkugel: no. Picking a patch up doesn't make us responsible.
23:07:39kugela patch cleanup month does
23:07:54bluebroth3rnot for the patch. We're responsible for the result, and we're responsible for the code anyway.
23:08:09bluebroth3rand a cleanup month doesn't make us responsible for patches either.
23:08:20bluebroth3rit's just asking people to give it a look.
23:08:31kugelan official one anyway, I'm not stopping anyone from doing it voluntarily
23:08:34bluebroth3ras with all open source: no promises :)
23:09:06 Quit ReimuHakurei (Quit: If I use this, I will disappear, and Shana-tan will remain...)
23:09:28bluebroth3rofficially asking people to consider picking up patches doesn't make us any more responsible than not asking.
23:09:39bluebroth3rit's _asking_ people after all.
23:10:06kugelI don't want to be asked to pick up patches
23:11:22kugelreviewing patches, OK, but not finishing them which should've been done by the author
23:11:39 Part Zagor
23:11:46Tornebluebroth3r: i don't think anyone who isn't interested in doing it anyway is going to do it
23:12:41bluebroth3rkugel: people can still ask you to do it. It's up to you to follow it or not.
23:13:13*bluebroth3r has been asked to join working on a Nano 3G port recently and hasn't been asked if he liked to be asked or not
23:13:33kugelpeople can ask me 20 times, I'll say no 20 times. nothing is gained
23:14:04bluebroth3rso? People have asked you, regardless if you like it or not.
23:14:25kugelyes, but a patch cleanup month implies I'm not allowed to do other stuff in the meantime
23:14:31CIA-27New commit by alex (r30045): Update US English by Steven Panek.
23:14:42kugelif it doesn't, it's completely useless from the start
23:14:45bluebroth3rseems you don't want to understand what I'm saying
23:14:46CIA-27New commit by alex (r30046): Update US English by Steven Panek.
23:14:50*bluebroth3r gives up
23:15:16pamaurykugel: you are supposed to not make bugs and bugs happen :)
23:15:40 Join kramer3d_ [0] (~kramer@unaffiliated/kramer3d)
23:16:33EspreonAlexP: Thanks.
23:18:22 Join ReimuHakurei [0] (~reimu@
23:18:49CIA-27r30045 build result: 2 errors, 0 warnings (alex committed)
23:19:53AlexPI'm not claiming responsibility for that either!
23:20:01AlexPI've been got twice tonight
23:20:08 Quit pamaury (Remote host closed the connection)
23:21:17gevaertsAlexP: looks like Zagor's fault :)
23:21:40*AlexP readies the pointy stick
23:31:51 Join ChickeNES [0] (~ChickeNES@
23:34:50 Join robin0800 [0] (~quassel@
23:39:31 Quit robin0800 (Read error: Connection reset by peer)
23:41:22 Quit amiconn (Disconnected by services)
23:41:23 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
23:41:26 Join shai_ [0] (
23:41:40 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
23:42:36 Join domonoky1 [0] (
23:42:45 Quit pixelma (Disconnected by services)
23:42:46 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
23:43:05 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
23:44:57 Quit evilnick_B (Quit: Page closed)
23:45:05 Join Strife89 [0] (
23:46:12 Join aevin_ [0] (
23:46:15 Quit ChickeNES (Quit: Computer has gone to sleep.)
23:49:37 Join JesusChrysler_ [0] (
23:50:21 Quit domonoky (*.net *.split)
23:50:23 Quit fs-bluebot (*.net *.split)
23:50:23 Quit shai (*.net *.split)
23:50:25 Quit JesusChrysler (*.net *.split)
23:50:25 Quit evilnick (*.net *.split)
23:50:28 Quit aevin (*.net *.split)
23:50:28 Quit otih (*.net *.split)
23:50:29 Quit linuxstb (*.net *.split)
23:50:29 Quit jordan` (*.net *.split)
23:50:30 Nick JesusChrysler_ is now known as JesusChrysler (
23:54:37 Join fs-bluebot [0] (
23:55:29 Join kugel_ [0] (~kugel@rockbox/developer/kugel)
23:56:48 Join linuxstb [0] (~linuxstb@rockbox/developer/linuxstb)
23:57:30 Join evilnick [0] (
23:57:30 Quit evilnick (Changing host)
23:57:30 Join evilnick [0] (~evilnick@rockbox/staff/evilnick)
23:57:50 Nick neferty is now known as nef_away (~andor@
23:57:52 Join Xerion [0] (
23:58:14 Quit kugel (Ping timeout: 250 seconds)
23:58:24 Nick aevin_ is now known as aevin (
23:58:25 Quit aevin (Changing host)
23:58:25 Join aevin [0] (eivindsy@unaffiliated/aevin)

Previous day | Next day