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 2011-06-05

00:01:12 Quit Buschel (Ping timeout: 248 seconds)
00:08:58 Quit jhMikeS ()
00:12:27 Join jhMikeS [0] (
00:12:27 Quit jhMikeS (Changing host)
00:12:27 Join jhMikeS [0] (~jethead71@rockbox/developer/jhMikeS)
00:16:54 Quit mudd1 (Ping timeout: 276 seconds)
00:18:00 Quit GeekShadow (Ping timeout: 244 seconds)
00:23:35 Quit jhMikeS ()
00:25:35kugelZagor: honeycomb is 11, isn't it?
00:27:20 Quit Cassandra2 (Quit: CGI:IRC (Ping timeout))
00:27:48 Quit B4gder (Quit: It is time to say moo)
00:28:06 Quit petur (Quit: here today, gone tomorrow)
00:28:43 Quit wodz (Quit: Leaving)
00:28:44Zagorkugel: no, 11 is 3.0. honeycomb is some preview thing that for some reason got its' own sdk id
00:29:15 Join ChickeNES [0] (~ChickeNES@
00:29:48kugelZagor: ah right, I rememeber there was a preview
00:30:00kugelbut actually all, the preview, 3.0 and 3.1 are honeycomb :)
00:31:25 Quit b1uebrother (Quit: leaving)
00:31:35pamaurygps working \o/
00:32:39 Quit ChickeNE_ (Ping timeout: 240 seconds)
00:32:50bertrikYou found the data sheet?
00:33:13pamauryI have the sirf protocol manual but it's actually sending nmea data
00:33:25 Part LinusN
00:33:26pamauryI "just" needed to figure out the init step and the uart settings
00:34:18 Quit pamaury (Remote host closed the connection)
00:34:24 Quit robin0800 (Quit: Leaving)
00:35:36bertrikhm, enabling wide bus mode for the internal sd memory on the clip v1 results in 70% faster read speeds for large reads
00:36:47bertrikbut apparently this fails on some models, according to the source code comments. I wonder if it failed on the internal memory, or on the external memory.
00:38:14bertrikmaybe the original fail has been fixed in the meantime
00:38:57 Quit Zagor (Ping timeout: 260 seconds)
01:21:46 Quit bertrik (Read error: Operation timed out)
01:29:20 Join romki [0] (510dba3e@gateway/web/freenode/ip.
01:41:14 Quit ChickeNES (Quit: Computer has gone to sleep.)
01:47:33romkiHello. I just installed rockbox (stable) on a sansa clip plus. The playlists are saved in the root directory. So they don't appear in the catalog. As reported here: looks like.
01:48:58 Join ChickeNES [0] (~ChickeNES@
01:53:59JdGordonbluebroth3r: err, not really
01:55:36JdGordonit isnt commited because it isnt really needed yet
01:55:36 Quit ChickeNES (Quit: Computer has gone to sleep.)
01:56:11***Saving seen data "./dancer.seen"
02:03:12 Quit factor (Read error: Connection reset by peer)
02:04:02 Quit ender` (Quit: Religion is regarded by the common people as true, by the wise as false, and by the rulers as useful. -- Seneca)
02:09:07 Join factor [0] (~factor@
02:17:08 Quit keyb_gr (Ping timeout: 240 seconds)
02:24:42pixelmaany other RaaA on Android user having the impression to be able to hear buffering or so? At low volume in a calm surrounding I can hear noise which gets worse on track start and stop with actual silence at the actual track change. I first thought it was my hardware but the bit of silence makes me wonder if it's a software problem too
02:30:31pixelmahmm, the built-in player seems to have the same problem though
02:31:03JdGordonnoisy flash like the sansas?
02:31:30mystica555_i've never heard that sort of noise on my G1.. what kind of phone, and, have you tried the stock mediaplayer for comparison?
02:33:11pixelmaJdGordon, reminds me of it yes, I notice sometimes on my c200
02:34:58pixelmamystica555_: SE X10 mini pro, and that's what I meant with 'built-in'. I'm comparing right now
02:37:40 Part romki
02:45:27 Join johnney5 [0] (
02:48:49 Join kramer3d_ [0] (~kramer@unaffiliated/kramer3d)
02:49:30 Join Peddy [0] (~apc@unaffiliated/peddy)
02:51:38 Join mshathlonxp [0] (
02:52:14 Quit kramer3d (Ping timeout: 276 seconds)
02:59:39 Nick kramer3d_ is now known as burnt (~kramer@unaffiliated/kramer3d)
02:59:44 Nick burnt is now known as burrnt (~kramer@unaffiliated/kramer3d)
03:00:38 Nick burrnt is now known as kramer3d (~kramer@unaffiliated/kramer3d)
03:24:33 Quit HaimN (Remote host closed the connection)
03:34:35 Join jhMikeS [0] (
03:34:35 Quit jhMikeS (Changing host)
03:34:35 Join jhMikeS [0] (~jethead71@rockbox/developer/jhMikeS)
03:37:51 Quit mshathlonxp (Quit: reb00t)
03:41:16 Join mshathlonxp [0] (
03:45:20 Join krazykit [0] (~krazykit@
03:47:08 Join liar [0] (
03:50:46 Join webguest1 [0] (
03:53:18 Quit webguest1 (Client Quit)
03:56:14***Saving seen data "./dancer.seen"
03:59:47 Quit mshathlonxp (Quit: HDD installation & vcore decrease)
04:05:53 Quit timccc (Ping timeout: 252 seconds)
04:08:27 Join timccc [0] (~aoeu@
04:09:10 Quit Peddy (Ping timeout: 240 seconds)
04:10:21 Join mshathlonxp [0] (
04:15:57 Join ChickeNES [0] (~ChickeNES@
04:30:19 Quit TheSeven (Disconnected by services)
04:30:28 Join [7] [0] (~TheSeven@rockbox/developer/TheSeven)
04:43:31 Quit liar (Ping timeout: 240 seconds)
04:49:24 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
04:49:31 Quit amiconn (Disconnected by services)
04:50:26 Quit pixelma (Disconnected by services)
04:50:29 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
04:50:31 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
04:53:12 Join kugel_ [0] (~kugel@rockbox/developer/kugel)
04:56:04 Quit kugel (Ping timeout: 260 seconds)
05:05:17 Join GeekShadow [0] (
05:05:17 Quit GeekShadow (Changing host)
05:05:17 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
05:10:11 Join Rob2223 [0] (
05:10:48 Quit GeekShadow (Ping timeout: 260 seconds)
05:13:33 Quit Rob2222 (Ping timeout: 244 seconds)
05:44:44 Quit mystica555 (Read error: Connection reset by peer)
05:50:58 Join mystica555 [0] (
05:53:23 Quit krazykit (Ping timeout: 260 seconds)
05:56:17***Saving seen data "./dancer.seen"
06:06:18 Quit sinthetek (Ping timeout: 258 seconds)
06:11:22 Join sinthetek [0] (
06:11:22 Quit sinthetek (Changing host)
06:11:22 Join sinthetek [0] (~sinthetek@unaffiliated/sinthetek)
06:25:45 Join Horschti [0] (~Horscht@xbmc/user/horscht)
06:29:05 Quit Horscht (Ping timeout: 240 seconds)
07:20:09 Join antil33t [0] (
07:21:53 Quit antil33t (Client Quit)
07:53:16 Join antil33t [0] (
07:56:18***Saving seen data "./dancer.seen"
08:00:35[Saint]kugel_: I realise it's *way* late....but, you rang?
08:00:55[Saint](get back to me when you can if it's still important and/or valid/required)
08:05:39[Saint]bluebroth3r: Congrats, btw...the only major patch of yours left in my tree now is the resources on uSD one. I wonder if that could be a compile-time option that defaults off? Would that please the masses I wonder?
08:07:58 Join robin0800 [0] (~robin0800@
08:32:42 Quit simonlnu (Read error: Operation timed out)
08:33:47 Join simonlnu [0] (simon@unaffiliated/simonrvn)
08:48:14 Quit factor (Ping timeout: 260 seconds)
08:51:00 Quit simonlnu (Read error: Connection reset by peer)
08:51:30 Join factor [0] (~factor@
08:52:02 Join simonlnu [0] (
08:52:02 Quit simonlnu (Changing host)
08:52:02 Join simonlnu [0] (simon@unaffiliated/simonrvn)
08:52:36 Quit kramer3d (Quit: Leaving)
08:53:36 Join simonlnu_ [0] (simon@unaffiliated/simonrvn)
09:06:59 Join robin0800_ [0] (~robin0800@
09:07:59 Quit robin0800 (Ping timeout: 255 seconds)
09:11:53 Quit simonlnu (Quit: .)
09:11:55 Nick simonlnu_ is now known as simonlnu (simon@unaffiliated/simonrvn)
09:20:00 Quit robin0800_ (Ping timeout: 244 seconds)
09:27:12 Join robin0800 [0] (
09:52:06 Join Lyrical [0] (
09:56:20***Saving seen data "./dancer.seen"
09:57:48 Quit Lyrical (Quit: CGI:IRC (EOF))
10:00:37 Join n1s [0] (~quassel@rockbox/developer/n1s)
10:01:51 Quit robin0800 (Ping timeout: 260 seconds)
10:26:24 Join Keripo [0] (
10:31:14 Quit jhMikeS (Ping timeout: 255 seconds)
10:32:11 Quit timccc (Ping timeout: 259 seconds)
10:35:22 Join bertrik [0] (
10:35:22 Quit bertrik (Changing host)
10:35:22 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
10:38:09bluebroth3r[Saint]: what I could think of is asking the user on first startup where to extractthe data to.
10:38:31*[Saint] nods.
10:39:00[Saint]I probably don't understand all involved, but I just think it's a good idea to cram as much as possible onto the uSD
10:39:43bluebroth3ryou can't do thatasasetting because atthat time Rockbox is starting - and you need to know where to extractto cirst.
10:41:38[Saint]assuming widgets end up as seperate .apks eventually and it's safe to move RaaA to the uSD on 2.2+ then the size of the isntall isn't too much an issue.
10:41:55[Saint]But 2.1 and under users will appreciate as much as possible on the uSD
10:46:30 Join timccc [0] (~aoeu@
10:51:30 Join stoffel [0] (
10:54:05bluebroth3r[Saint]: see mycomment in the FS task about the size thing.
10:57:08 Quit mystica555 (Remote host closed the connection)
10:59:51 Quit ChickeNES (Quit: Computer has gone to sleep.)
11:06:15 Join bzed_ [0] (
11:06:39 Join robin0800 [0] (
11:07:27 Quit bzed (Ping timeout: 244 seconds)
11:07:31 Nick bzed_ is now known as bzed (
11:08:53 Join mystica555 [0] (
11:10:12 Quit [Saint] (Quit: have to reboot too)
11:12:23 Join [Saint] [0] (
11:16:40 Quit robin0800 (Ping timeout: 255 seconds)
11:19:45n1ssaratoga: about that tremor iram patch giving a speedup even if the codec is in riam already is probably because tremor does some extra copying to try to use iram efficiently on targets with small iram this is also the default configuration so "no choice" == small iram, see FS #11268
11:19:45bluebrother-bot Optimization for tremor on targets that don't use iram. (patches, new)
11:20:26n1si should probably try to see if i can fix and commit that patch
11:21:05n1smaybe it has been magically fixed by other things :)
11:28:00CIA-58New commit by nls (r29962): Delete old unused macro.
11:29:00 Join Zagor [0] (~bjst@
11:29:00 Quit Zagor (Changing host)
11:29:00 Join Zagor [0] (~bjst@rockbox/developer/Zagor)
11:30:09 Join pamaury [0] (~quassel@
11:30:10 Quit pamaury (Changing host)
11:30:10 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
11:30:27 Join petur [0] (~petur@rockbox/developer/petur)
11:30:28 Join b1uebrother [0] (~dom@
11:31:16CIA-58r29962 build result: All green
11:32:27amiconn_Stream ist online again:
11:32:35 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
11:33:39 Join B4gder [0] (~daniel@
11:33:39 Quit B4gder (Changing host)
11:33:39 Join B4gder [0] (~daniel@rockbox/developer/bagder)
11:37:14 Nick kugel_ is now known as kugel (~kugel@rockbox/developer/kugel)
11:38:08bertrikbluebroth3r, there's still a long-standing bug with the lcd/backlight on both the c200 and e200
11:38:22bertrikI get a white screen when switching from "Off" to "On"
11:39:04bertrikSo I think something is still not right between powering up/down the lcd and the backlight
11:39:26 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.)
11:39:54bertrik(in relation to FS #9874)
11:39:54bluebrother-bot sansa - Make lcd_sleep work (patches, closed)
11:40:33 Join robin0800 [0] (
11:40:36b1uebrotherbertrik: feel free to reopen the task −− I'm not aware of such an issue and are trying to get the number of tasks down a bit so I might be a bit too trigger happy right now :)
11:40:56n1sbertrik: that's FS #10966
11:40:57bluebrother-bot c200 v1 LCD doesn't turn on when changing Backlight setting from Off to On (bugs, new)
11:41:17bertrikah yes, thanks n1s
11:41:38 Join [Saint] [0] (
11:44:50Zagorwhat is the minimum android sdk level we require?
11:44:59CIA-58New commit by kugel (r29963): Android: Greatly simplify the pcm callback mechanism on both, the Java and the C side. Should be more reliable now (if the old wasn't already).
11:45:12 Quit robin0800 (Ping timeout: 248 seconds)
11:45:15kugelZagor: 8 or 9
11:45:26Zagorhow do we find out?
11:45:54kugelby trying to compile with different versions :)
11:46:03kugelI'll try, I have a couple installed
11:46:18 Quit [Saint] (Ping timeout: 258 seconds)
11:46:40Zagorlevel 8 is android 2.2. do we really require such a recent version?
11:47:05pixelmait works on my 2.1 phone at least
11:47:35kugelZagor: no, we have compatibility code for downto 1.5
11:48:00gevaertsBut we probably require it to *build*
11:48:07CIA-58r29963 build result: All green
11:48:07Zagorgevaerts: ?
11:48:38kugelthe build time requirement is higher than the runtime requirement
11:48:47Zagoroh, ok
11:49:37kugel8 works
11:49:56ZagorI'll use that in the check then. thanks.
11:50:58kugelbut clients will need to upgrade if we use new APIs, even if they're already available now, so it might be better to require the latest one
11:51:42 Join robin0800 [0] (
11:52:52Zagoryeah, I guess it can't hurt.
11:54:35amiconnNew stream url:
11:54:48amiconnMr. Someone pulled the usb cable...
11:56:21***Saving seen data "./dancer.seen"
11:58:10 Join [Saint] [0] (
11:58:55kugelGodEater: perhaps, but rather unlikely, your random stops are gone
12:00:14[Saint]kugel: You pinged?
12:00:34kugel[Saint]: yea, can you try if latest svn fixes your audio problems?
12:01:19[Saint]yeah, will do. I need to sync up my tree after lots of recent commits anyway. I'll do it now.
12:01:28 Quit [Saint] (Client Quit)
12:05:35 Join [Saint] [0] (~Saint]
12:09:50pixelmaMr. Someone is here? Go catch him, he has stuff to do ;)
12:10:20 Join TheLemonMan [0] (~lem0n@
12:11:30pixelmakugel: the audio problems seem to be fixed for me, it worked fine the 2 times half an hour stress test this night and morning
12:11:46kugel great
12:16:18 Join JdGord [0] (~AndChat@
12:16:47 Join LinusN [0] (
12:23:10amiconnHmm. X5 dualboot code is basically working, but not much luck with either firmware on my X5 at least
12:23:57amiconnThe OF doesn't seem to like my SSD, and the rockbox part of the bootloader shuts down because of the on_button/ hold switch check
12:24:37amiconnThat check conflicts with the way the dualboot code decides which firmware to boot
12:25:26*amiconn isn't sure how to solve that yet
12:25:46 Join ender` [0] (
12:25:52 Join HaimN [0] (~quassel@
12:26:08CIA-58New commit by wodz (r29964): Move dbg_hw_info() into target tree. FS #11735 by me
12:26:09bluebrother-bot move dbg_hw_info() to target tree (patches, new)
12:26:28 Join wodz [0] (~wodz@
12:26:42bertrikDid anyone bring an ipod accessory?
12:28:53CIA-58New commit by bluebrother (r29965): Manual: describe key clicks feature. ...
12:29:39CIA-58r29964 build result: 3 errors, 1 warnings (wodz committed)
12:30:06bertrikZagor, I'm getting some perl warning on the rbtest client
12:30:26Zagoryeah, I'm fixing those now. going "use strict" even.
12:31:15b1uebrotherisn't FS #10766 pretty similar to what skip length does?
12:31:15bluebrother-bot Enable Binary Skip / Search as an option for Skip Length (patches, unconfirmed)
12:32:51CIA-58r29965 build result: 3 errors, 1 warnings (bluebrother committed)
12:33:44CIA-58New commit by wodz (r29966): fix red
12:37:10CIA-58r29966 build result: All green
12:39:05pixelmabluebroth3r: it's a bit different, I believe there was also a forum thread about it
12:41:31amiconnSo how to solve the dualboot conflict on the X5?
12:41:44amiconn(also on the other iAudios)
12:42:17amiconnWe cannot use another button than Power (or Remote power) without using the PCF code really early in the bootloader
12:43:23amiconnPatch is based on the last one in fs#5289
12:43:24bluebrother-bot Dual Boot for X5 (patches, unconfirmed)
12:44:23amiconnBut the rockbox part of the bootloader nowadays compares on button state with hold state in order to decide whether it has to shut down again
12:45:13amiconnSince you have to let go the On button in order to boot rockbox with that dualboot solution, it won't be pressed when the later code checks it
12:45:37amiconn-> Rockbox just says "Shutting down.." (and does it)
12:48:32pixelmarasher: seen ?
12:48:57pixelmaI first was confused because he used english as example but it also happens for other languages
12:49:56[Saint]kugel: after some very heavy stress testing with playback, I can confirm svn works for me.
12:50:40[Saint]lots of stops, starts, heavy seeking, forced rebuffer...everything that made it fail previously without a patched build now "just works".
12:50:45 Quit Horschti (Read error: Connection reset by peer)
12:51:01 Join Horschti [0] (~Horscht@xbmc/user/horscht)
12:52:50Zagorbertrik: I made a mistake, you need to grab a new client manually: bertrik: I made a mistake. curl -O
12:53:04bertrikok, will do
12:54:11[Saint]so, is the official git repo up yet?
12:54:20AlexPyes, but it is a secret
12:54:31amiconnHmm, the only solution I can think of is somehow save the state of the power buttons detected by the dualboot code, and use that in the rockbox boot code path later on
12:54:48amiconnThat leaves the question how to handle this in a single boot bootloader
12:55:32bertrik[Saint], you know we already have a git mirror, right?
12:56:00[Saint]yes, but I was wondering if that one was going to be kept, if a new one started, <other>?
12:57:54CIA-58New commit by rasher (r29967): Fix problems.php not knowing that files are now in the rockbox subdir
12:58:01b1uebrother[Saint]: what's the state of FS #11717? I'd like to get that in
12:58:02bluebrother-bot RaaA (Or Similarly Large Target) Iconset (patches, unconfirmed)
12:58:07Zagor[Saint]: we'll start a new one
12:58:18pixelmarasher: nice, thanks
12:58:53rasherpixelma: Was an easy fix!
12:59:32[Saint]bluebroth3r: the consensus was that I needed to fix the rest of the iconsets to match before that went in.
12:59:55[Saint]...I should probably get around to that.
13:00:04amiconnHmph. New stream url again:
13:00:12b1uebrotherI'd really like to (a) get a 24x24 icon set in and (b) close that task.
13:00:15*b1uebrother is on a closing spree
13:00:47*Zagor cheers for b1uebrother
13:01:21b1uebrotherFS itself isn't the problem, it's the amound of outdated stuff that's in it (it still might not be the best tool, but well ...)
13:01:31b1uebrotherstill over 600 open issues :(
13:01:54*b1uebrother has an idea
13:03:02*b1uebrother goes closing all patches that do not have a real name attached
13:03:56kugelpamaury: I uploaded a dircache patch for dircache_copy_path()
13:04:13b1uebrothermay I ask all Rockbox developers to put their real name in their FS profile?
13:04:58bertrikHow about patches that add plugins? I've seen a few that I personally just don't care about that much, but could be useful to other people. And, being a plugin, it doesn't have much impact on the rest of the code.
13:05:15AlexPLoads of them are only one target etc. though I think
13:06:00kugelwe once decided that we should have a low bar for plugins
13:06:04pixelmabertrik: any opinion on following his suggestion of changing "signal strength" to just "signal"? Sounds like a sensible solution to me and it currently takes up some space on small displays
13:06:19pixelmaerr... is what I'm talking about
13:07:05bertrikpixelma, fine with me
13:07:07pixelmaI'd like to close the task for the rest of it
13:07:54pixelmait's really the problem with scrolling lines and dynamic content
13:08:06bertrikor maybe use "RSSI", but I guess a lot of people don't know what that means, so "signal" is fine.
13:08:21amiconnAnd again a new one:
13:09:24pixelmathe µ is something we can't do for the built-in one though
13:09:29pixelmaI guess
13:09:49wodzgevaerts: what about FS #10239?
13:09:49bluebrother-bot USB storage write caching (patches, new)
13:11:19 Quit HaimN (Quit: No Ping reply in 180 seconds.)
13:12:05 Join HaimN [0] (~quassel@
13:13:37pamaurykugel: ok, I'll have a look
13:14:18b1uebrotherany time set for building the tower bridge?
13:14:57b1uebrotherwe _need_ it to be a tower bridge this year!
13:14:57 Quit TheLemonMan (Ping timeout: 276 seconds)
13:15:00*b1uebrother repeating himself
13:15:47 Join TheLemonMan [0] (
13:17:32ZagorAlexP, bertrik: I'm afraid the test client has soiled your directories a bit. feel free to run "rm -rf build-*"
13:19:09b1uebrotheranyone with H300 aware of the state of FS #9501?
13:19:10bluebrother-bot severe screen corruption on H300 (bugs, unconfirmed)
13:19:40 Join Xerion_ [0] (
13:20:08bertrikZagor, it wasn't that bad for me. BTW, I'm still seeing "Use of uninitialized value $data in length at line 167, <GEN0> line 6."
13:20:33AlexPbluebrother: I haven't seen any other reports, and it is old
13:20:37n1sbluebrother: Well we haven't heard any more from the reporter so it's impossible to develop a fix
13:20:45 Quit Xerion (Read error: Connection reset by peer)
13:20:45 Nick Xerion_ is now known as Xerion (
13:21:50n1si remember a few people using a patch to slow down the lcd somewhat a very long time ago but i think that was committed
13:22:42kugelI'm going to commit FS #5153
13:22:44bluebrother-bot Doom Scrollwheel Patch (patches, assigned)
13:22:54kugelit's a bit of a hack, but it's better than svn and better than closing it
13:23:55CIA-58New commit by kugel (r29968): Accept FS #5153 - Doom Scrollwheel Patch ...
13:23:56bluebrother-bot Doom Scrollwheel Patch (patches, assigned)
13:24:28n1syay for 5 year old patches :)
13:24:30amiconnIt might be a problem to implement this for M3, but M5 should be easy
13:25:04 Quit ender` (Read error: Connection reset by peer)
13:25:37 Join ender` [0] (
13:25:59*pixelma scratches head a bit about what to do with the language string exactly if done correctly
13:26:42n1spixelma: what are you doing?
13:26:51pixelmaI don't want to change all languages because some translators may prefer a different word for that then
13:27:11pixelman1s: changing "Signal strength:" to just "Signal:" for the radio screen
13:27:16CIA-58r29968 build result: All green
13:27:17pixelmato shorten it
13:27:58n1ssince the meaning changes the Right Thing would be to deprecate the old string and make a new one
13:28:38n1sOtherwise just chnge english and the translators will get a "source changed" notice on the next update
13:29:11kugeln1s: 6239 was actually accepted right (in r14517)? I'm going to close it
13:29:15pixelmayes, is there a difference to getting "this string doesn't exist" message?
13:30:00n1skugel: no idea
13:32:14n1spixelma: yeah, the source changed message pops up by the already present translation which might save the translators some work, perhaps
13:32:20b1uebrotherrasher: is FS #11783 still an issue?
13:32:21bluebrother-bot Themesite not accepting %T with setting_inc or setting_dec (bugs, unconfirmed)
13:32:58rasherAbsolutely no idea. I haven't been involved with that code at all
13:33:32 Join MonGoliat [0] (
13:33:53MonGoliatHi, anyone here good with Terminal?
13:34:15*kugel added a comment to FS #8668, it should be closed IMO
13:34:15[Saint]I'd say a few of us are, yes ;)
13:34:16bluebrother-bot battery runtime: experimental gui boost (patches, assigned)
13:34:25MonGoliatCool! =D
13:34:56MonGoliatIm having trouble with the terminal part in the manual install of RB on my Sansa Clip+
13:35:19[Saint]what are you having trouble with specifically?
13:35:36MonGoliatWhen i run the mkamsboot-mac it sais "Could not open bootloader-clipplus.sansa for reading"
13:36:19MonGoliatany idea why that acurrs?
13:37:10 Quit Xerion (Read error: Connection reset by peer)
13:37:14wodzpermissions problem probably
13:37:25MonGoliathow do i set the permition?
13:37:32 Join Xerion [0] (
13:37:33GodEateror the bootloader isn't in the same directory?
13:37:51MonGoliatIt is in the same direcory as the rest of the files.
13:38:09n1skugel: it seems it was committed in r18046 by sdoyon himself actually so yeah, closing it was fine
13:38:39[Saint]what do you mean by "mkamsboot-mac", should it not be either mkamsboot,exe or ./mkamsboot?
13:38:47*n1s does not like tasks that have multiple patches doing different things
13:39:15GodEaterdoesn't the mac version have a different name?
13:39:29[Saint]if it does, the manual has no idea of it.
13:39:31MonGoliatthe file I got out of the mkamsboot.dmg was a called "mkamsboot-mac"
13:40:02 Quit HaimN (Quit: No Ping reply in 180 seconds.)
13:40:03GodEaterseems reasonable to me then
13:40:29GodEaterso what permissions are on the .sansa file currently? (hint: 'ls -l *.sansa')
13:40:37MonGoliatThe manual actualy said ./mkamsboot but that didnt make any sense..
13:40:48 Join HaimN [0] (~quassel@
13:41:24AlexPIt does for linux
13:41:31AlexPAnd the installation section is the same
13:41:45AlexPWe could either separate the linux/mac bits, or rename the mac binary
13:41:46GodEaterwho built the .dmg - do we know?
13:42:21 Join jordan` [0] (gromit@2a01:e34:eebf:c890:21a:4dff:fe63:6966)
13:42:35b1uebrotherZagor: do we have wiki markup in FS again?
13:42:39MonGoliatthe permission? Im a total terminal dim-wad so I dont know how to check.
13:42:48Zagorb1uebrother: no
13:42:55AlexPMonGoliat: He gave you the command :)
13:43:00AlexPls -l *.sansa
13:43:14MonGoliat*slaps his face*
13:43:18b1uebrotherZagor: interesting, I just saw something that looked like that in a task :)
13:44:32*[Saint] is driven insane by why RaaA causes haptic feedback events on shutdown
13:44:45[Saint](either via the main menu, killing the task, or idle shutdown)
13:44:51 Quit Xerion (Read error: Connection reset by peer)
13:44:59[Saint]it's the only app on my handset that does this.
13:45:12GodEater[Saint]: doesn't do it on my phone :P
13:45:20 Join Xerion [0] (
13:45:20AlexPNor mine
13:45:29[Saint]I've seen a Glaxy Tab user report it too.
13:46:00[Saint]also the same "RaaA doesn't show up as a task in the task manager unless there's playback, and sometimes not even then" thing.
13:46:02kugelNor mine :)
13:46:18b1uebrotherGodEater: from reading the description this sounds like the issue you're experiencing? FS #12006
13:46:19bluebrother-bot Android: OGG, SPX, MP3 playback problem (bugs, unconfirmed)
13:46:22[Saint]perhaps more crazy Samsung voodoo
13:46:30pixelman1s: well, so go for "source changed" or "doesn't exist", I'm a bit confused by your answer
13:47:01GodEaterbluebrother: nope
13:47:09GodEaterit's not about playback starting on my phone
13:47:14n1spixelma: i think source changed is fine for this
13:47:14GodEaterit's playback stopping randomly mid-track
13:47:24GodEateralthough I've not had that problem since kugel's commit this morning
13:47:30GodEaterso fingers crossed :)
13:47:38MonGoliatEhmm. I think i managed to check the permission. is "staff" a valid answer?
13:47:48GodEaterpaste the entire line you got back
13:48:03MonGoliat-rw-r−−rw-@ 1 myname staff 44948 Jun 4 19:54 rbinstall/bootloader-clipplus.sansa
13:48:10kugel[Saint]: we just decided to ignore all weirdness on your "madness ROM"/phone
13:48:27GodEaterMonGoliat: can you type "pwd" now please?
13:48:38AlexPkugel: That's an official devcon decision :)
13:49:02b1uebrotheryay, we're down to less than 600 open tasks!
13:49:19AlexPTower time!
13:49:47b1uebrotherAlexP: Tower Bridge!
13:49:52[Saint]it's the buffer/audio stop patch causing it.
13:50:07AlexPAlways test SVN before reporting!
13:50:17[Saint]svn from prior to that commit breaks audio...but no haptic feedback.
13:50:20GodEater"Present working directory"
13:50:29GodEaterjust type it in the terminal and tell us what it says
13:50:32 Quit Xerion (Ping timeout: 240 seconds)
13:50:36[Saint]AlexP: it's a patch just committed
13:50:49AlexP[Saint]: I missed a :) :)
13:51:24MonGoliatehm.. could you please us terms i understand? dos pwd mean /rbinstall/?
13:51:43AlexPIt means type "pwd"
13:51:48[Saint]I had an older version of it in my tree for ages to make audio work...never assumed that could cause bloody haptic feedback events on shutdown.
13:51:53[Saint]looking at it, nor should it.
13:52:01[Saint]yet, it apparently does.
13:52:15MonGoliat"/Users/*username* "
13:52:15[Saint]Yep....madness. Samsung voodo. :/
13:54:30GodEaterthat doesn't look very literal output
13:54:34 Join Buschel [0] (
13:54:39 Join Xerion [0] (
13:54:55MonGoliatehm... cuz i dont wanna write my full name here
13:55:00[Saint]29960 (clean) == broken audio but no haptic feedback event on shutdown, 29961 (clean) == fixed audio and haptic feedback events approximately ~2 seconds after shutdown.
13:55:00Zagorhow do you group commands in bash?
13:55:25 Quit mshathlonxp (Read error: Connection reset by peer)
13:55:50GodEaterMonGoliat: paranoid much? :)
13:56:07Zagor"echo foo && echo bar >> logfile" only outputs "bar" to the logfile
13:56:12MonGoliatnot much, just enough. =p
13:56:22***Saving seen data "./dancer.seen"
13:56:29GodEaterok - type "cd rbinstall", then "ls", and then cut and paste the full output to - and then tell me what link pastebin gives you
13:57:26 Quit JdGord (Quit: Bye)
13:58:30 Quit robin0800 (Quit: Leaving)
13:59:51GodEaterthe Tower Bridge of Rockbox has started construction :)
14:00:00GodEaterunfortunately it's out of sight of the webcam
14:00:28 Quit Xerion (Ping timeout: 255 seconds)
14:01:52GodEaterMonGoliat: srsly dude - paste it in the channel
14:04:18Zagorahh, finally successful android build
14:04:49GodEaterMonGoliat: what command did you type earlier to run this?
14:05:00GodEater(the mkamsboot-mac bit)
14:05:12MonGoliatnon. U just typed what you told me to.
14:05:30MonGoliatI started a new terminal window
14:05:36GodEaterso how did you get the message about not being able to open the file?
14:06:17GodEaterdid you run mkamsboot or not?
14:06:20MonGoliatIve tried again and now the message says permission error
14:06:22GodEaterI'm serisouly confused now
14:06:31GodEaterwhat message?
14:08:17GodEaterso you're running the command from the wrong directory
14:08:39GodEateryou'll note there is NO mention in that message about the wrong permissions
14:08:40 Join Xerion [0] (
14:08:55GodEaterso please try to report *exactly* what you read, not invent messages ;)
14:09:00MonGoliatI was the last time i trired. Had do try again.
14:09:16GodEaterdo you still have the terminal open from when I told what to type?
14:09:29MonGoliatno. sorry
14:09:44GodEaterwhy on earth did you close it?
14:09:47 Join robin0800 [0] (
14:09:48MonGoliatI though I did the same thing both times. Apparently didnt
14:10:06GodEaterok so in whatever one you have open currently
14:10:12GodEaterplease type "cd rbinstall"
14:10:20GodEaterand then "./mkamsboot-mac clppa.bin bootloader-clipplus.sansa patched.bin"
14:11:41MonGoliatI got the pached.bin now
14:12:19MonGoliatGodEater! Thanks for your patience
14:13:41GodEaterbtw, is that *you* burying the dead naked chick?
14:14:55Zagorbertrik: are you still getting warnings in the latest build round?
14:15:31 Quit factor (Read error: Connection reset by peer)
14:15:42bertrikZagor, no, revision 47 seems to give no warnings anymore
14:16:20GodEaterI said burying, not corrowing
14:16:22ZagorI think we're good to go live with this version
14:16:23GodEater*borrowing even
14:16:53MonGoliatARe you reffering to the photographer with the same name as me?
14:17:13GodEatermaybe ;)
14:17:23GodEaterjust having a little fun with you
14:17:52MonGoliatI got that. :p
14:17:57kugelZagor: awesome
14:18:52bertrikWhat's actually different between the new and the current build client?
14:19:19Zagorthe client isn't so different, chiefly added android tool check. the server has changed more.
14:19:35Zagorto make it more generic and expandable
14:19:49Zagor"fit for public consumption" :-)
14:20:06MonGoliatI now have RB running. Finaly, som volume!
14:21:12MonGoliatAny must have plugins or themes?
14:21:29GodEateryou have all the plugins already
14:21:43GodEaterthemes are entirely a personal preference, so bit of a pointless question imo
14:22:02GodEaterI never bother with them anyway. I use Rockbox to listen to music, not to admire the pretty pictures ;)
14:22:12*GodEater runs away from [Saint]
14:22:37 Quit johnney5 (Ping timeout: 260 seconds)
14:23:10[Saint]it's a Clip isn't it?
14:23:24[Saint]No pretty pictures to be had there :P
14:23:32 Join krazykit [0] (~krazykit@
14:25:14GodEaterI thought pictureflow worked on it?
14:25:23 Quit robin0800 (Ping timeout: 260 seconds)
14:26:41[Saint]Ah, does. that's right, plugins can use greylib but not the core.
14:26:53 Join robin0800 [0] (
14:27:06*[Saint] learnt that a few days ago so it's still embedded.
14:27:45Zagorwe have the best tower ever this year
14:27:55Zagorwith a london theme
14:27:56GodEaterTower Bridge of Rockbox is complete!!!
14:29:16GodEateron their way
14:31:39 Quit TheLemonMan (Ping timeout: 276 seconds)
14:32:27 Join TheLemonMan [0] (
14:32:43 Join factor [0] (~factor@
14:33:27Buscheln1s: what do you think of FS #12146 ? if someone could verify the patch on a Coldfire target it is fine to commit imho. nice finding
14:33:28bluebrother-bot Fix libcook bugs introduced in r22055 (patches, unconfirmed)
14:34:20 Quit Keripo (Quit: Leaving.)
14:38:24n1sBuschel: yeah, it looks good, i can test on cf in a while
14:38:43n1sin fact, that cf asm could be faster :)
14:39:33Buschelthen you should make it faster and submit the whole thing :)
14:39:47n1syeah, i'll do that
14:40:30bertrikbluebroth3r, saratoga : how about creating a new clip+ bootloader ( ) ?
14:40:52bertrikWhat's the work involved? Just someone building it and uploading it to
14:42:18n1sbertrik: you should tag it in svn too
14:45:49 Join johnney5 [0] (
14:48:13*[Saint] has a WIp that drops both "popup tabs" but still keeps the "popup windoes" available.
14:48:23[Saint]windows too
14:48:55[Saint](I also moved the volume popup window permanently and solely to the volume icon.
14:49:44wodzbertrik: what is the story of FS #11767?
14:49:44bluebrother-bot Meier crossfeed (patches, new)
14:50:05[Saint]If I address the font...that *should*, in theory, be the immediate concerns of everyone (that I'm aware of) taken care of.
14:50:20[Saint]GodEater: Anything I'm forgetting?
14:50:33[Saint]the tabs and the font were your two major piss-offs?
14:51:35[Saint]I'm not sure if I'll have it clean enough to make a "release" just yet...the theme is spattered with test code, but I'll continue it through the night.
14:52:31Buscheln1s: do you see anything that might speak against accepting FS #12143? the patch seems fine for me and saves some binsize
14:52:31bluebrother-bot Clean up the NSF codec (patches, unconfirmed)
14:52:51[Saint] minor thing remains is blocking the alternate popup window from opening if/when the other is open...but this is possible iiuc, I just have to find the right magic to get things firing at the correct time.
14:53:22n1sBuschel: i haven't looked at it but it sounds fine to me
14:53:55[Saint]AlexP: I should point that out to you also, of the above listed things I mention...anything else you remember that I'm forgetting?
14:54:39[Saint](Oh...yeah, I fixed the spacing inthe lists up for the UI viewport in the 480x800 theme too...again, I've just got these changes locally though, not public, yet)
14:55:38[Saint]Even if it has a short shelf life...I think there's a chance this theme *can* be shaped into something quite usable 'til there's a full native alternative.
14:56:22[Saint]*s/quite usable/quite usable for all/
14:58:20bertrikwodz, I'm still undecided about the meier crossfeed, I found a bug in the coefficient calculation, but people seem to like to old version better with the bug still in it, so I have to make a choice between the old or new one.
15:00:19 Quit HaimN (Ping timeout: 255 seconds)
15:01:18*Zagor crosses his fingers and starts the new buildserver
15:06:46bertrikI get "Unterminated <> operator at line 1."
15:07:22bertrikthat's a piece of html
15:08:18 Quit user890104 (Remote host closed the connection)
15:08:28 Join user890104 [0] (
15:09:02Zagoryes, I saw that too. looks like every client got that for some reason.
15:09:23Zagornice of viewvc to croak right now...
15:09:35Zagoryou need to grab it manually: curl -o ""
15:10:51bertrikI get a warning "Use of uninitialized value $buildmaster in concatenation (.) or string at line 41."
15:12:54CIA-58New commit by nls (r29973): FS #12146: Fix libcook bugs introduced in r22055 by Sean Bartell.
15:12:54bluebrother-bot Fix libcook bugs introduced in r22055 (patches, unconfirmed)
15:16:37 Join HaimN [0] (~quassel@
15:17:03Zagorbertrik: yeah, I saw that too. so even if viewvc hadn't broken the farm, I would have :-)
15:19:05 Quit krazykit (Ping timeout: 252 seconds)
15:22:11 Join Sundiver_ [0] (~angel@
15:24:18 Quit Sundiver (Ping timeout: 276 seconds)
15:24:39Zagorok, update manually again: curl -o ""
15:28:09 Join MethoS- [0] (~clemens@
15:29:03 Quit antil33t (Read error: Connection reset by peer)
15:29:12 Join antil33t [0] (
15:29:14Zagorthis applies to *ALL* build client owners. you need to manually update rbcient: curl -o ""
15:29:23Zagorthis is not a drill :-)
15:29:55bertrikok, I have two clients upgraded
15:32:21bertrikI guess we lost roolku's farm for the time being :|
15:33:59Zagoryes. I'm writing a mail to the rbclient list now
15:38:20GodEaterupdated ;)
15:46:42 Quit wodz (Quit: Leaving)
15:47:10JdGordonsomeone mentioned wanting to enforce a commit style with git... how are we going to do that if we want people to commit locally for us to pull instead of us apply patches?
15:51:49Zagorwe can't ask to pull. most people don't have servers we can pull from.
15:52:39JdGordonI figure alot would clone to github
15:52:42bertrikZagor, I got another "Use of uninitialized value $data in length at line 168, <GEN0> line 18." with version 47 of the client
15:52:45JdGordonor similar
15:52:46B4gderwell, most people could post to github etc
15:52:57ZagorB4gder: true, that would work
15:52:57B4gderbut I don't think we need to do it that way
15:53:23Zagoryeah, that's only for patches. committers will push themselves.
15:53:27B4gderI prefer us all pushing, as that's more similar to how we've worked in the past
15:54:09JdGordonwas a timeline discussed?
15:56:25***Saving seen data "./dancer.seen"
15:59:24 Quit B4gder (Remote host closed the connection)
16:04:10 Quit robin0800 (Quit: Leaving)
16:04:14JdGordonpetur: you didnt add anything skin related for the hystogram changes right?
16:08:16JdGordondoesnt look like it. do i assume you dont plan to?
16:08:40JdGordon(which would be the only way it is useable for playback... unless i've misread the commit)
16:11:45 Quit HaimN (Remote host closed the connection)
16:23:38 Quit stoffel (Ping timeout: 240 seconds)
16:32:30 Join kevku [0] (~kevku@2001:470:28:773:babe:feed:dead:bee)
16:36:18 Join Topy44 [0] (
16:52:25 Part LinusN
16:53:20 Quit Zagor (Quit: Leaving)
16:54:37peturJdGordon: no, one thing at a time....
16:54:52peturand I'm not into skinning anyway ;)
16:55:25peturmaybe wait until I also added another new peakmeter feature which allows them to work vertically :)
17:04:09 Quit ReimuHakurei_ (Ping timeout: 240 seconds)
17:05:47 Quit bluebrother-bot (Ping timeout: 250 seconds)
17:06:22 Quit bluebroth3r (Read error: Operation timed out)
17:06:32 Quit MonGoliat (Quit: CGI:IRC)
17:08:35 Join antil33t| [0] (
17:08:38 Quit antil33t (Disconnected by services)
17:09:17 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
17:14:04 Join robin0800 [0] (
17:16:46 Join bluebrother-bot [0] (
17:17:20 Quit antil33t| ()
17:22:34Buscheldoes anybody have an idea why FS #12147 results in a codec speedup? I thought as3525 targets have the whole codec in IRAM
17:22:35bluebrother-bot Assigns bigger IRAM for libtremor (patches, unconfirmed)
17:22:54 Join stoffel [0] (
17:24:54bertrikdoes it indeed result in a speedup?
17:25:05bertrikthe clipv1 is very memory-limited IIRC
17:26:03Buschelwell, the author *states* the patch speeds up the codec on his clip+
17:32:25 Quit FOAD (Remote host closed the connection)
17:36:03 Join FOAD [0] (~dok@
17:36:11 Quit robin0800 (Quit: Leaving)
17:36:40 Join robin0800 [0] (
17:41:47 Quit petur (Quit: here today, gone tomorrow)
17:44:16kugelBuschel: I suspect he did something wrong
17:44:43Buschelkugel: maybe...
17:44:59Buscheldoes anybody have a clip+ to verifiy?
17:45:18kugelcodec is entirely in IRAM already
17:45:22kugelwe have some clip
17:45:40kugelclip+ here, but there's no need if you ask me. could also compare the map files
17:47:39Buschelgood point. is it possible that his patch changes the memory layout (e.g. moving an data array to another position in IRAM)?
17:49:43Buschelthis could maybe affect chaching
17:50:02Buschelcaching, of course :/
17:50:59n1sBuschel: i think i know why
17:51:24n1si wrote for saratoga this morning
17:51:31n1s"n1ssaratoga: about that tremor iram patch giving a speedup even if the codec is in riam already is probably because tremor does some extra copying to try to use iram efficiently on targets with small iram this is also the default configuration so "no choice" == small iram, see FS #11268"
17:51:55 Join keyb_gr [0] (
17:52:02n1sthe same happens on targets that dont have or use iram at all
17:52:14n1swastes about 2MHz on the beast
17:53:18bertrikIs there a way to tell if the LCD is on? I'd like to stop the polling of the radio tuner chip if the lcd is off to avoid interference on the clip+
17:54:38n1sextern bool lcd_active(void); sounds probable
17:56:27***Saving seen data "./dancer.seen"
17:56:50bertrikhm, the radio stuff is polled from the skin_engine, so it gets a little trickier to make it stop polling I guess
17:59:44pixelmalcd updates are also audible on stations with bad reception on the ondio and the FM Recorder - which has a system to stop lcd updates (enabled by a button press - the "freeze" button, but that's not possible on the Ondio due to lack of buttons
18:00:25Buscheln1s: where is this copying hidden? I cannot believe 2 MHz are consumed by copying within the init-routines.
18:01:06pixelmaturning off the lcd completely won't be too nice because it's nicely readable
18:01:50pixelmaI'm not sure if it's still working though, the FM Recorder is a rare target and of the ones I know, only LinusN has one IIRC
18:02:04bertrikI suppose the LCD on the ondio and FM recorder takes almost no power (no backlight), right?
18:02:39amiconnActually on the newer Ondio FM (Philips tuner) it isn't that bad, but both the older Ondio FM and the FM recorder have a Samsung tuner, and there interference is really bad
18:02:44bertrikI'm not completely sure if the interference is caused by the display update, or by access over i2c to the tuner chip
18:02:50pixelmathe FM Recorder has backlight, only stock Ondios don't. Mine does thoguh
18:03:15amiconnWe even go as far as slightly shifting the MAS PLL when the FM frequency would otherwise be a harmonic of it
18:04:08amiconnThis does have a noticeable effect
18:04:22bertrikfunny, the tuner in the clip+ has a similar problem, it thinks there's a strong station at 96 MHz, probably a harmonic of the 24 MHz main crystal, I guess we can't do anything about this.
18:05:18bertrikthe OF seems to have a work-around for this, effectively ignoring detected stations around 96.0 MHz
18:05:40n1sBuschel: see FS #11268, iirc it copies blocks to iram before running the mdct on them or somesuch so it's not just a start up cost
18:05:41bluebrother-bot Optimization for tremor on targets that don't use iram. (patches, new)
18:05:43 Quit robin0800 (Ping timeout: 255 seconds)
18:07:40n1slook for iram_pcm_doublebuffer in block.c
18:08:05 Join robin0800 [0] (~robin0800@
18:08:10n1sthe patch fixing this used to cause a problem on sims i couldn't figure out but i haven't looked at it for a year
18:08:43Buscheln1s: now I got it! thanks :)
18:10:07 Quit stoffel (Ping timeout: 260 seconds)
18:10:18Buscheln1s: so, the patch itself is somehow correct. because now the codec *assumes* its iram is large enough
18:11:23Buschelare there any other targets / CPU's which completely have the codecs in IRAM?
18:24:09 Quit robin0800 (Quit: Leaving)
18:26:01keyb_grbertrik: Have you ever tried obtaining a current RDA5802(E) datasheet @RDA Micro? (Think they're at 3.0.) Otherwise I'll fill out their little form now.
18:27:14bertrikI found a more recent datasheet on a chinese web site (I think it was or something similar), I couldn't download the pdf, but the contents looked the same as the pdf we do have.
18:27:23Buscheln1s: do you want to search for a better solution or is FS #12147 good enough to submit?
18:27:23bluebrother-bot Assigns bigger IRAM for libtremor (patches, unconfirmed)
18:27:24bertrikI haven't tried asking RDA
18:28:06bertrik(or whoever made that chip)
18:31:08keyb_grOK, gave it a shot. Haven't seen a 2KB/s download in a while. Aww, great, that's the first page of rev 1.7. ;-/ Oh well...
18:32:56 Join stoffel [0] (
18:34:22keyb_grSo no progress on these undocumented registers yet.
18:35:09 Quit sinthetek (Ping timeout: 240 seconds)
18:42:11 Join sinthetek [0] (
18:42:11 Quit sinthetek (Changing host)
18:42:11 Join sinthetek [0] (~sinthetek@unaffiliated/sinthetek)
18:42:56 Quit b1uebrother (Quit: leaving)
18:50:14Buschelkugel: I just recognized you closed FS #8668. I disagree. The gui boost works very well for several people, and especially the 24 MHz clock gains lots of battery runtime. imho it is good to keep this open and visible
18:50:15bluebrother-bot battery runtime: experimental gui boost (patches, assigned)
18:53:47 Quit z35 (Quit: Leaving)
18:54:36 Nick Horschti is now known as Horscht (~Horscht@xbmc/user/horscht)
18:57:04 Join froggyman [0] (~seth@unaffiliated/froggyman)
19:09:43 Join Lear [0] (
19:10:43 Quit pamaury (Remote host closed the connection)
19:19:38 Quit stoffel (Ping timeout: 240 seconds)
19:28:44 Join Llorean [0] (
19:28:47 Quit Llorean (Changing host)
19:28:47 Join Llorean [0] (~DarkkOne@rockbox/user/Llorean)
19:34:12 Quit Lear (Quit: ChatZilla 0.9.87 [Firefox 4.0.1/20110413222027])
19:45:37Buschelkugel: any comment?
19:56:29***Saving seen data "./dancer.seen"
20:02:20 Join kramer3d [0] (~kramer@unaffiliated/kramer3d)
20:03:17 Join GeekShadow [0] (
20:03:17 Quit GeekShadow (Changing host)
20:03:17 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
20:05:26 Quit Buschel (Ping timeout: 240 seconds)
20:11:28 Quit bertrik (Read error: Connection timed out)
20:11:58 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
20:14:42 Join domonoky [0] (~Domonoky@rockbox/developer/domonoky)
20:17:31 Join ReimuHakurei [0] (~reimu@
20:19:04 Join webguest97 [0] (
20:20:18 Quit webguest97 (Client Quit)
20:34:42 Join Keripo [0] (
20:39:03 Quit Keripo (Ping timeout: 240 seconds)
20:41:24 Quit factor (Read error: Connection reset by peer)
20:59:11 Join factor [0] (~factor@
21:01:22 Quit simonlnu (Read error: Operation timed out)
21:15:36 Quit bertrik (Read error: Connection timed out)
21:16:05 Join bertrik [0] (
21:16:06 Quit bertrik (Changing host)
21:16:06 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
21:29:54n1sBuschel: I don't think FS #12147 is the right aproach, i'd like to take another look at the thing and see if i can fix my patch
21:29:55bluebrother-bot Assigns bigger IRAM for libtremor (patches, unconfirmed)
21:30:13*n1s wonders if stripwax is around much anymore
21:39:12 Join jhMikeS [0] (
21:39:12 Quit jhMikeS (Changing host)
21:39:12 Join jhMikeS [0] (~jethead71@rockbox/developer/jhMikeS)
21:39:39 Join sirrozha [0] (~sirrozha@
21:39:51 Quit sirrozha (Client Quit)
21:45:28 Quit mc2739 (Ping timeout: 240 seconds)
21:45:33 Quit GeekShadow (Quit: The cake is a lie !)
21:47:31 Join petur [0] (~petur@rockbox/developer/petur)
21:47:41 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
21:56:33***Saving seen data "./dancer.seen"
22:11:14 Join stoffel [0] (
22:13:20 Quit stoffel (Remote host closed the connection)
22:17:51 Join krazykit [0] (~krazykit@
22:17:59 Join Buschel [0] (
22:18:32CIA-58New commit by nls (r29975): Safe part of FS #11268, saving some codec buffer space for targets that don't use separate iram.
22:18:33bluebrother-bot Optimization for tremor on targets that don't use iram. (patches, new)
22:19:17n1shmm, that does in fact not save any space yet, just the copy...
22:19:17 Join Buschel_ [0] (
22:19:24*n1s facepalms
22:21:23 Join mshathlonxp [0] (
22:22:35 Quit Buschel (Ping timeout: 252 seconds)
22:22:44 Nick Buschel_ is now known as Buschel (
22:29:18 Quit mshathlonxp (Ping timeout: 240 seconds)
22:29:40 Join mshathlonxp [0] (
22:31:14Buscheln1s: goal of FS #11268 is to be able to set TREMOR_USE_IRAM only in case the codec uses both IRAM and DRAM, right? for as3525 this define will not be set, to avoid the extra copy?
22:31:15bluebrother-bot Optimization for tremor on targets that don't use iram. (patches, new)
22:32:21n1sBuschel: yes
22:33:27Buschelthe naming of the #define is a bit misleading imho, but not worth changing when commented reasonably ;)
22:33:43 Quit mshathlonxp (Read error: No route to host)
22:34:37n1sthe special case with codec buffer in iram is handled like it's not in iram everywhere, so USE_IRAM is unset and the various ATTR's expand to nothing
22:36:02 Quit petur (Ping timeout: 258 seconds)
22:36:18n1sbut sure i can add a comment
22:38:02n1si'm looking for the bug in the patch atm, but haven't gotten any furter
22:38:03Buschelwould be a good idea as this dynamic special handling (using the result of iram_malloc()) not common.
22:38:33n1syeah, that is nbot my creation
22:39:29Buschelit is not bad, it is just special. libfaad might get more efficient with such handling as well. but not much.
22:40:02 Join liar [0] (
22:41:16Buscheln1s: the residual (sporadic) bug is the noise when skipping back to the beginning
22:41:44n1ssomething needs to be reinited somhow is my guess
22:41:49Buschelwith TREMOR_USE_IRAM defined
22:42:10n1sno, in the no iram case, the iram case still works fine
22:42:24Buschellet me see
22:42:47 Join mshathlonxp [0] (
22:45:13n1si have a synced patch if you want
22:45:40Buschelno, the v3 patch in flyspray should be fine
22:45:48Buschelthanks :)
22:49:07n1sforward seeking does it too
22:53:28Buscheln1s: this pcm_copy[] usage looks weird. but it is not impacted by TREMOR_USE_IRAM
22:54:51n1sthose pointers are only so we can free the allocated buffers easily
22:55:04n1sbut no, it's not terribly pretty
22:58:27Buschelhmm, wait. pcm_copy[] is only accessed if TREMOR_USE_IRAM is not defined (or if the IRAM is too small)
22:59:05n1syes, they are copies of the pointers from ogg_malloc
23:06:12Buschelwhat happens if you comment both free(pcm_copy)?
23:06:50n1sif i make sure _vds_init is called when seeking it works...
23:06:59 Join evilnick [0] (~evilnick@rockbox/staff/evilnick)
23:07:47n1snot freeing the pointers makes no difference
23:09:20n1sor well, making sure vorbis_synthesis_init is called on seek
23:09:27 Join Self-Perfection [0] (
23:11:29 Quit domonoky (Read error: Connection reset by peer)
23:12:00n1scalling _vds_init does fix it so something from the setup needs to be redone it seems
23:12:33 Quit TheLemonMan (Quit: Ex-Chat)
23:13:18Buschelwhat happens in case you have USE_IRAM set and v->iram_double_pcm is NULL?
23:17:20 Quit evilnick (Quit: Leaving)
23:17:32n1sv->iram_double_pcm is null if the iram_malloc fails and then the second buffer is calloced
23:18:48Buschelsorry, my question was misleading. does the error occur in this case?
23:22:31 Quit bertrik (Read error: Connection timed out)
23:23:19 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
23:32:02n1sah, hmmm
23:36:55Buschelsorry, I was not of many help tonight :/
23:37:01Buschelgood night!
23:37:07 Quit Buschel (Quit: ChatZilla 0.9.87 [Firefox 3.6.17/20110420140830])
23:37:09n1sno problem, thanks for looking
23:40:53 Quit kevku (Quit: KVIrc 4.0.4 Insomnia
23:44:26n1sthe bridge is awesome :)
23:49:20 Quit bertrik (Read error: Connection timed out)
23:49:47 Join bertrik [0] (
23:49:47 Quit bertrik (Changing host)
23:49:47 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
23:51:57 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
23:54:13 Quit Self-Perfection (Remote host closed the connection)
23:54:22 Join Keripo [0] (
23:56:37***Saving seen data "./dancer.seen"
23:58:01 Quit balintx (Remote host closed the connection)
23:58:18 Join balintx [0] (

Previous day | Next day