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).

Notice: Only Gecko based browsers prior to FF4 support the multipart/mixed "server push" method used by this log reader to auto-update. Since you do not appear to use such a browser, this page will simply show the current log, and not automatically update.

#rockbox log for 2009-07-21

00:00:12linuxstbbertrik: (if you read the logs): ping
00:00:17stripwaxAnything inlined (as we talked about last time for vorbis and especially wma) can give rise to slight crazy counts.
00:04:49 Quit at0m (Read error: 110 (Connection timed out))
00:05:28 Quit barrywardell ()
00:06:33*pixelma has a suspicion which revision broke it and the one before works so now for the proof
00:07:02pixelmait = music playback on my Ondio
00:07:42 Join wincent [0] (n=wincent@host-091-097-067-213.ewe-ip-backbone.de)
00:08:36JdGordon|which one?
00:09:01*JdGordon| hopes its one from 2 months ago
00:11:09pixelmayep... it was r21863 (the strlcpy one). So not your fault but I wonder how (or when) you tested or if it's an Ondio specific problem or
00:11:30stripwaxsaratoga - woah, cook uses a RNG that needs a mult?
00:11:47pixelma2 months ago can't be since my old revision wasn't that old
00:12:19Unhelpfulamiconn: actually i think he was referring to my arm mdct changes anyway... and i used test_codec for that. also, markun thought you'd had objections to merging some lcd driver things together... would y mind taking a look at FS #4817
00:12:26pixelmaaround DevCon-ish
00:12:43gevaertshm, the strlcpy one is annoying. That's a big one to go through
00:13:56pixelmathe crash happens in gui_wps_update which could explain why it happens in cuesheet in recent revisions (address is still "close")
00:14:27pixelmaif I read the map file correctly
00:16:52gevaertsif you have the address and rockbox.elf, you can get more accurate results
00:18:32gevaerts"sh-elf-objdump −−disassemble rockbox.elf" will give you a full disassembly, with the addresses in the first column
00:18:48 Join funman [0] (n=fun@rockbox/developer/funman)
00:19:26funmansimpler: ../utils/analysis/find_addr.pl 0xabcdef # provided rockbox.elf & rockbox.map are in the current directory
00:19:50funmanI had to install http://search.cpan.org/~jhi/String-Scanf-2.1/lib/String/Scanf.pm on ubuntu 9.04
00:20:52 Join at0m [0] (n=at0m@94-225-90-23.access.telenet.be)
00:23:32pixelmaJdGordon|: is your Recorder flashed or just "running" Rockbox?
00:23:58JdGordon|just running
00:24:42wincentfunman: Your proposition (in the e-mail) of logf() would mean that I have to enter it into every function. Do you have any other ideas?
00:24:43pixelmaI'm still trying to understand why the problem only showed up completely after I flashed the new build
00:25:11 Join barrywardell [0] (n=barrywar@86-45-4-10-dynamic.b-ras2.prp.dublin.eircom.net)
00:25:37funmanwincent: you could have a look at profiling (gcc -pg)
00:26:04funmanI think it's more complex to setup, but gcc will add a call to a certain function in each function entry
00:27:57pixelmafunman: that utility doesn't seem to work in my installation (cygwin) "Can't locate String/Scanf.pm in @INC"
00:28:37pixelmaanother difference is that my build dir is not inside the source dir
00:28:55funmanpixelma: you need to install the package as well (download the .tar.gz; untar; perl Makefile.PM; make; make install)
00:28:56pixelmabut I have the path right as tab completion shows
00:28:58wincentfunman: Does profiling information include the calling log or does it produce calling statistics only?
00:29:20funmanpixelma: you just need to call the find_addr.pl from the build dir
00:29:33wincentfunman: I do not need a calling graph, but the calling log.
00:30:02pixelmafunman: then that's not easier for now... yes, path was no problem, I just told you in case that makes a difference
00:30:31funmanwincent: you can modify the added function (__cyg_profile_* in firmware/profile.c) but i think using logf() is simpler: how much functions do you need to track ?
00:30:46funmanpixelma: will be easier when you have other problems in the future ;)
00:31:09funmanor perhaps when someone removes the dependancy of this script on extra perl modules
00:31:28wincentfunman: The whole pdbox :-) And for the reason of comparison the whole pd-anywhere (original application).
00:31:42funmanwincent: i don't know pdbox at all
00:32:18wincentfunman: pdbox = plug-in I am working at.
00:32:59 Join BHSPitMonkey [0] (n=stephen@unaffiliated/bhspitmonkey)
00:33:56funmando you need to check if functions are called in every .c file of pdbox/ ?
00:34:13wincentyes.
00:34:41 Quit BradC (Read error: 110 (Connection timed out))
00:35:21 Quit PaulJam_ (".")
00:35:54funmani think using gcc -pg with a custom firmware/profile.c is shorter then
00:36:20wincentI will give it a shot.
00:37:03funmani can't tell how you can get the function name however
00:38:13funmanperhaps store the pc in an array and compare it to pdbox.map
00:38:19funmanthe pcs*
00:38:26wincentHmm. I envision a little perl script that puts "puts("functionname");" after the entry of every function in a file...
00:38:53wincentAnd another one that removes the "puts" line afterwards :-)
00:39:11funmanI can help with the second one: "svn revert . -R" ;)
00:39:33wincentThanks. That will do :-)
00:44:19 Quit n00b81 ("Leaving")
00:45:35stripwaxwincent - what is it you need to do exactly? do you need to profile on-target on just in simulator?
00:57:33 Join Havoc65 [0] (n=chris@24.2.23.216)
00:58:14 Join BradC [0] (i=heh16911@202-89-178-44.static.dsl.amnet.net.au)
00:58:27 Quit ender` (" Man created gods. The opposite remains to be proved. -- Serge Gainsbourg")
00:58:59 Part Havoc65 ("Konversation terminated!")
01:00
01:00:51***Saving seen data "./dancer.seen"
01:01:11wincentstripwax: I need to compare function calls in the original application, which runs smoothly, and in the Rockbox plug-in, which crashes while loading.
01:01:26stripwaxbut in sim, or on-target?
01:02:33wincentIn the simulator.
01:03:14 Quit B4gder ("It is time to say moo")
01:03:57pixelmagevaerts: what do I do with the completel line from objdump?
01:04:53stripwaxwincent - ok. compiling sim with (rockbox) profiling enabled is pretty broken, so logf and/or puts might be the easy way to go
01:06:12wincentstripwax: Thank you. I think that I might try a libduma memchecker (pdbox crashes because of segmentation fault, caused by a corrupted memory).
01:06:35gevaertspixelma: send the output to a text file (e.g. "sh-elf-objdump −−disassemble rockbox.elf > output.txt"), and then search for the address you want in the first column. As soon as you find it, move up in the file until you see a line with he function name in it. That's the function you're looking for
01:13:05 Join goffa___ [0] (n=goffa@216.220.23.105)
01:13:59pixelmaI did that and all I see is e.g. "_gui_wps_update+0x8c"
01:14:27pixelmathe map already told me the first part
01:15:14gevaertsexcept you're now sure it's not a nearby static function
01:17:02 Quit DarkDefender ("Leaving")
01:17:20 Quit Thundercloud (Remote closed the connection)
01:17:28pixelmaaha. Can I do something that helps tracking dowm the bug some more? Bug report should follow tomorrow as I really should go to sleep now
01:18:32 Quit Zagor ("Leaving")
01:18:46stripwaxsaratoga - yeah, so this should be more meaningful. I had to move a bunch of inline code to noninlined functions to see what their contribution was; and put a bunch of __attribute__ ((no_instrument_function)) around everything else that I wouldn't care about. http://pastebin.com/m76d6cb69
01:19:32gevaertspixelma: not sure. I think someone needs to look at it first
01:19:46stripwaxIn doing so I've probably affected the performance a bit but them's the breaks.
01:20:26 Quit goffa (Read error: 110 (Connection timed out))
01:21:02pixelmagevaerts: I see, thanks for the help with that. Still puzzling though why it doesn't seem to hit if you haven't flashed Rockbox (my first try and probably what JdGordon| saw)
01:21:51 Quit AndyI (Read error: 110 (Connection timed out))
01:23:06funmanwincent: did you try running the sim into valgrind?
01:23:14 Quit roolku ()
01:25:20 Quit BHSPitMonkey (Read error: 104 (Connection reset by peer))
01:25:23funmanpixelma: gevaerts: IIRC the bug mentioned http://forums.rockbox.org/index.php?topic=22137.msg152934#msg152934 happens in gui_wps_update() in cuesheet related code
01:27:39Unhelpfulstripwax: here's my benchmark plugin for jpeg idct... it wouldn't be hard to add other files to it, change the functions it benchmarks, etc. it also has some support functions for logging the results of transformations, so that you can compare implementations.
01:27:42Unhelpfulhttp://pastie.org/552713
01:28:12pixelmafunman: that would be before the r21863 revision which breaks playback (maybe "just" switching ti the WPS) for me, in my case it makes no difference if I have cuesheet support enabled or disabled
01:28:24stripwaxUnhelpful - cheers
01:29:24 Quit shotofadds ("Leaving")
01:31:26 Join BHSPitMonkey [0] (n=stephen@unaffiliated/bhspitmonkey)
01:34:42stripwaxsaratoga - updated again. sprinkled a few __attribute__ ((noinline)) around static functions that gcc might otherwise inline, to get a bit more detail into cook.c http://pastebin.com/m7abb46e7
01:36:12stripwaxunpack_SVQH stands out a little, as does categorize.
01:36:26derekjaI installed rockbox on a 60 gb ipod. At first I installed the 60gb version, however when I started having crashes, I uninstalled that one and installed the 30gb version. It now crashes less frequently, but still does. Why would this be?
01:37:08stripwaxderekja - any info on what sort of crashes you mean?
01:37:36stripwaxNote that rockbox doesn't really have a 60gb version versus a 30gb version: it has a "64MB of RAM" version versus a "32MB of RAM" version.
01:37:40derekjastripwax: It goes black and needs to be restarted.
01:37:53stripwaxDoes the screen say anything?
01:38:05derekjastripwax: how can I tell which the ipod is?
01:38:11derekjastripwax:
01:38:12derekjano
01:38:21 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
01:38:39stripwaxderekja - generally the 60gb ipod has 64MB but I think some refurbished ipods have only 32MB.
01:38:46 Quit linuxstb (Nick collision from services.)
01:38:50 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
01:39:10derekjastripwax: it crashed with the 64mb version too
01:39:37stripwaxWhat are you doing at the exact moment the rockbox screen goes black? Any sound coming out? Is this scrollwheel unresponsive? What format audio files are you playing, and is the crash reproducible consistently when you play the same track?
01:39:40 Quit barrywardell ()
01:39:56stripwaxAlso very important, which version of rockbox (including build number)
01:42:16 Join CaptainKewl [0] (n=jason@207-237-172-77.c3-0.nyr-ubr4.nyr.ny.cable.rcn.com)
01:42:38 Join barrywardell [0] (n=barrywar@86-45-4-10-dynamic.b-ras2.prp.dublin.eircom.net)
01:42:45derekjastripwax: it happens when I try to change the song or open a directory. Everything is unresponsive, it does not happen if a track is playing but you are not touching it. version 3.3
01:44:43stripwaxderekja - does it happen when you skip (left/right) or only when you try and select a new track (select)? when the screen is blank is it still playing music?
01:46:59derekjastripwax: only when a new track is selected, and it is not still playing music?
01:47:08derekjaignore the question mark
01:54:02stripwaxSo you can use the file browser ok, but when you select the track to play, rockbox freezes, no audio, and no screen? I don't think I've heard of that issue before. Are you using a custom build of any sort? Which format audio files are these?
01:54:45stripwaxIs it possible your disk has a filesystem error on it (e.g. try running chkdsk on windows or fsck.vfat on *nix)
01:56:06 Quit JdGordon| (Read error: 104 (Connection reset by peer))
01:59:36stripwaxsaratoga - last time for tonight, before I sleep. So there's a bunch of inline stuff in bitstream.h that is certainly skewing some of those results. This version (hopefully) has all the inlines excluded from profiling. http://pastebin.com/m2b98ab6b
02:00
02:00:20 Quit at0m (Read error: 60 (Operation timed out))
02:01:31 Nick Ypsy is now known as YpsyZNC (n=ypsy@geekpadawan.de)
02:04:20 Quit robin0800 ("Leaving")
02:15:14 Join JdGordon| [0] (i=ae91727f@gateway/web/freenode/x-f9b1c4c77b8860f5)
02:22:03 Join at0m [0] (n=at0m@94-225-90-23.access.telenet.be)
02:28:59 Quit dys (Connection timed out)
02:30:49 Nick fxb is now known as fxb__ (n=felixbru@h1252615.stratoserver.net)
02:42:52 Join archstech [0] (n=archstec@206.251.250.215)
02:43:21 Join sindfil [0] (n=sindfil@206.251.250.217)
02:52:08 Quit JdGordon| ("Page closed")
02:59:15wincentfunman: This is what I intend to do.
03:00
03:00:55***Saving seen data "./dancer.seen"
03:16:00 Quit sindfil (K-lined)
03:16:00 Quit archstech (K-lined)
03:17:09 Quit funman ("doh")
03:47:13 Quit tomcat_ha ("If visible identification is not possible, the pathologist may be able to take fingerprints from the body until decay sets in)
03:50:20 Join LambdaCalculus37 [0] (n=rmenes@rockbox/staff/LambdaCalculus37)
03:57:16 Quit fuzzie (Read error: 60 (Operation timed out))
04:00
04:00:25 Part wincent ("Kopete 0.12.7 : http://kopete.kde.org")
04:01:01 Part derekja
04:04:39 Quit dmb (No route to host)
04:06:08 Quit TheSeven (Nick collision from services.)
04:06:22 Join The_Seven [0] (n=theseven@dslb-084-056-133-030.pools.arcor-ip.net)
04:06:26 Nick The_Seven is now known as TheSeven (n=theseven@dslb-084-056-133-030.pools.arcor-ip.net)
04:12:50 Join dmb [0] (n=Dmb@unaffiliated/dmb)
04:25:27 Quit Sajber^ (Read error: 104 (Connection reset by peer))
04:53:59 Quit Rob2222 ()
04:58:17CIA-69New commit by 03kkurbjun (r21987): Pegbox: Add support for 640x480 screens
05:00
05:00:57***Saving seen data "./dancer.seen"
05:11:42 Quit jhulst (Remote closed the connection)
05:13:57 Join supacool [0] (n=1824bf02@gateway/web/cgi-irc/labb.contactor.se/x-fd731fd04c2e4865)
05:14:00 Join jhulst [0] (n=jhulst@jhulst.com)
05:18:13 Quit safetydan ("Leaving.")
05:18:23 Join safetydan [0] (n=deverton@rockbox/developer/safetydan)
05:18:32 Part safetydan ("Leaving.")
05:18:40 Join safetydan [0] (n=deverton@rockbox/developer/safetydan)
05:19:04supacoolhi folks... i've got a bit of a wrangled problem if anyone has any suggestions: a file has two different composer tags (vorbis comments); but I can't figure out how to get my wps screen to display both composers... %ic only ever reads one of the tags
05:23:46evilnickHow are they stored? i.e. are they Composer=composer1;composer2 or are they Composer1=composer1 and Composer2=composer2
05:25:35supacoolCOMPOSER tags for both; as far as I know Rockbox's wps identifiers would have no way of parsing custom tags like COMPOSER1 and COMPOSER2
05:26:07supacooltwo separate tags for two separate composers
05:30:14supacoolI'd prefer not to resort to listing all composers in one tag comment because it defeats the advantages you gain by keeping them separate (i.e., browsing a database by composer
05:35:45supacoolUpdate: I just checked and it seems the database is affected also... that is, only one composer tag per file is ever parsed
05:39:11 Quit LambdaCalculus37 ("Fwump")
05:39:21 Join FlynDice [0] (n=FlynDice@m150436d0.tmodns.net)
05:40:55CIA-69New commit by 03saratoga (r21988): Rearrange loop to avoid one branch per iteration.
05:43:06 Join Rob2222 [0] (n=Miranda@p4FDCD6F4.dip.t-dialin.net)
05:47:43CIA-69New commit by 03saratoga (r21989): Commited the wrong thing, fix zeroing in cook.
05:51:12JdGordonsupacool: you cant do it
05:52:11supacoolyeah, i realized that... i'm gonna starting looking through some source because it should be something that can be done
05:53:01CIA-69New commit by 03kkurbjun (r21990): Star, Flipit, Rockblox, Sokoban, Wormlet: Add support for 640x480 screens
05:53:20 Quit Horscht ("Verlassend")
05:53:23supacoolmaybe tomorrow though, today I'm tired; thanks folks
05:53:27evilnickHow about setting one composer to the Comment tag? Wouldn't be ideal, but it's a quick fix
05:53:52supacoolnot if the file already has a comment tag
05:54:42supacoolplus I hate hacky solutions
05:56:44supacooleverything seems to stem from the fact that vorbis tags are arbitrarily forced to fit the ID3 scheme because that's what was used in Rockbox first
05:58:52safetydanit would be nice if there was some sort of map structure for tags
05:59:18safetydana wps could be written like %tag{COMPOSER}
05:59:56safetydanyou could then add indexes for duplicates like %tag{COMPOSER}[1]
06:00
06:01:15*daurnimator says something about his ultra cool lua audio metadata library
06:02:34 Quit evilnick ("Page closed")
06:06:35 Quit fdinel ("Miranda IM! Smaller, Faster, Easier. http://miranda-im.org")
06:07:25Dhraakellianoooh
06:22:21CIA-69New commit by 03kkurbjun (r21991): Snake2: Add support for 640x480 screens
06:24:41 Join iwonder [0] (i=chris243@c-24-118-138-250.hsd1.mn.comcast.net)
06:25:04iwonderwhy is there so much hard drive activity just switching between folders?
06:25:19iwonderon the iPod video
06:25:47iwonderthere's about a good 5 to 10 second latency on switching folders
06:26:15kkurbjuniwonder: it is much faster if you enable diskcache
06:26:21iwonderah
06:26:28Llorean5-10 seconds suggests something's wrong though
06:26:29iwonderwhere
06:26:31kkurbjunor you can turn up the spindown time
06:26:47LloreanIt shouldn't be more than a couple.
06:26:53LloreanAnd it's "dircache"
06:26:59kkurbjunLlorean: sounds like spinup
06:27:27Lloreankkurbjun: 5-10 seconds is a lot longer than spinup *should* be though.
06:28:07supacoolfragmentation of the data on the drive might be an actual issue, no?
06:28:32LloreanIsn't this just FAT stuff, fragmentation shouldn't matter?
06:33:10iwonderok, so WHY wasn't Directory Cache *enabled* at default?
06:33:17iwonderit works almost instantly now
06:33:20iwonderas it should
06:33:55LloreanSome people would rather not waste RAM on making browsing a small amount faster when most of the time they're just listening to music.
06:34:10Lloreanas well, dircache has some limitations (size limits) still that can cause problems for some users.
06:34:15iwonderit's not small at all
06:34:19iwonderit's night & day
06:34:30LloreanFor you, because there's something else wrong.
06:34:49iwonder<kkurbjun> iwonder: it is much faster if you enable diskcache
06:35:00iwonderhmm.. someone else seems to have the same issue
06:35:01LloreanYes, I was here, I saw that.
06:35:17LloreanDircache speeds things up, but it shouldn't be 5-10 seconds without it. It should be 2-3 tops.
06:35:33iwonderwell, it is at least 5 seconds
06:35:41LloreanWhich is why I said there's probably something else wrong.
06:35:50LloreanYou're seeing unusually long times.
06:35:54iwonderprobably because i transferred huge amounts of mp3's & data files to it
06:35:56kkurbjuniwonder: there was some talk about changing the defaults, but it has not been done yet
06:35:58iwonder20+ gigs
06:36:02iwondernot sure
06:36:14supacooliwonder: i'm curious how many files on average you have per directory... a giant flat structure for all your music would cause delays
06:36:19kkurbjunyou might also want to run a scan on your player just to make sure there are no problems on the HD
06:36:40iwondersupacool, i minimized most of my folders to 100 or less
06:36:52iwonderi had over 1,000 in one and it said "dir buffer is full"
06:37:17iwonderbut now, i'm not sure if it would be an issue with directory cache enabled
06:37:23kkurbjuniwonder: you can increase the maximum size of that as well in the menus
06:37:32kkurbjunit will still say that with dircache
06:37:42kkurbjunat least I believe it will
06:41:18 Quit supacool ("CGI:IRC (EOF)")
06:41:31iwonderoh & thanks to the person who created this http://www.rockbox.org/twiki/bin/view/Main/RockboxFontConvertor
06:41:55iwonderi now use Century Gothic size 15 as the Perfect font for iPod video!
06:43:34iwondersome other cool style fonts: Jokerman, Chiller, Lucida Calligraphy
06:44:33iwonderthough, Century Gothic, Trebuchet, Verdana, & Bookman old style are probably the most legible
06:46:01 Join __lifeless [0] (n=lifeless@188.16.82.134)
06:46:05iwonderinstead of stating "dir buffer is full" it could have said something like "more than 1,000 files in folder. try minimizing files in folder or you can specify the max size here"
06:46:43scorcheiwonder: keep in mind that rockbox runs on a *lot* of devices with very different LCDs...that massive message wouldnt hope to fit on anything except the largets screens
06:47:00scorchelargest
06:47:15 Quit _lifeless (Read error: 60 (Operation timed out))
06:47:57 Join _Shaid [0] (i=adam@lurking.shaid.net)
06:48:08iwonderor Rockbox could time the latency of switching between folders, seeking, forward/next tracks, and have a customizable threshold (default of say 2 seconds) to notify user to automatically enable Directory Cache
06:48:25iwonderor just have it as a default (and allow it to be disabled) if one really needs the memory
06:50:04 Quit Shaid (Read error: 104 (Connection reset by peer))
06:50:11 Nick _Shaid is now known as Shaid (i=adam@lurking.shaid.net)
06:50:37safetydaniwonder: the last option is the one most likely to happen. everything else just sounds complicated
06:51:22iwonderyep
06:51:32iwonder:)
06:51:39iwondersorry for complicating things unnecessarily
06:51:47iwonderwe need to have less burdens, not more
06:52:01 Join BHSPitMonkey_ [0] (n=stephen@pool-173-74-208-124.dllstx.fios.verizon.net)
06:52:02iwonderesp. if the pay-off is not too great for the amount of time/head-aches involved ;)
06:52:54 Quit BHSPitMonkey (Read error: 104 (Connection reset by peer))
06:54:05iwonderi also think a font similar to Century Gothic size 15 would be a Great default for iPod video
06:54:23iwonderand certainly not the default one chosen, which is far too small
06:54:38iwonderif you have seen that font on the iPod video, you'd probably agree
06:56:29safetydaniwonder: if you can find a suitably licensed font that fits, it can always be added
07:00
07:01:02***Saving seen data "./dancer.seen"
07:06:00iwonderis there a way to quick search the folder?
07:06:17iwonderi don't see it in the manual
07:07:33iwondera filter as you type
07:11:15iwondercould be opened by holding down the play/pause button in the file list, and then you can scroll through each letter of the alphabet
07:11:38iwonderselect button confirms letter & jumps to the first hit in the file list
07:12:00iwonderscrolling & selecting more letters filters the list more
07:12:09iwonderthat's how it was on my Nomad JB3
07:12:27 Nick fxb__ is now known as fxb (n=felixbru@h1252615.stratoserver.net)
07:12:27iwondermakes things a lot easier, when you don't have playlists setup (yet)
07:12:52iwonderbut even playlists it can be useful if there are many songs
07:15:25iwonderalso makes it less important to have such a tiny font size and have to squint to read each song
07:15:27iwonder;)
07:15:37iwonderbecause what you want to see, is seen clearly
07:15:47iwonderyou just filter & jump to it
07:16:21iwonderis there a way to make the sensitivity of the ipod wheel be less 'jumpy'?
07:22:51iwonderfiltering the song list could be sped up further by Rockbox detecting the letters in the song list that are in that sequence (thus saving you the time & movement) of scrolling thru letters that are not in that same sequence
07:27:55 Quit CaptainKewl (Remote closed the connection)
07:28:21iwondereg: already having the filter letters of 'per' will only only show the letters 'k' & 'c' if the only songs in the list with "per" in them are perk* or perc*
07:30:42 Quit advcomp2019 (Read error: 104 (Connection reset by peer))
07:31:12 Join advcomp2019 [0] (n=advcomp2@unaffiliated/advcomp2019)
07:32:44iwonder by holding down the Track Forward button one can have Rockbox look for those words, but unlike normally, do not have to be the first letters of the song/filename
07:33:40iwonderTrack Forward button ie when you are already in the filter mode
07:36:21iwondera good default would be to have Rockbox have its filter mode search for Song titles in the format of "Artist - Song name.mp3" or be able to switch to filter by Artists
07:36:36iwonderwithout having to have id3 tags
07:39:03iwonderit could also support wildcards, ??? * or even further, boolean :)
07:39:50evilnick_homeWhy *wouldn't* you want to tag your music files?
07:40:01iwondernext
07:40:03iwonder!
07:40:06iwonder;)
07:40:50iwonderyou obviously do not understand
07:41:04iwonderabout designing good software adaptive to the user
07:41:24iwonderyou do not value it
07:41:36iwonderat least not as much as i do
07:42:47scorche|shyou really
07:42:49scorche|shshould stop
07:42:53scorche|shtalking like this...
07:43:17iwonderwhen you can make things more accessible, without slowing down or causing other things more important to break, and the time/energy is worth the added usefulness to all the users who use it, I say - go for it!
07:43:24iwonderhehe
07:44:13evilnick_homeiwonder: Features like that might be useful for you personally but for people with properly tagged music wouldn't be much use and they'd cut down on available memory for all users
07:44:22safetydaniwonder: with Rockbox it's always a trade off between RAM usage and usability for all
07:44:40iwonderevilnick - don't speak as if i'm the only one who could use this
07:44:59Lloreaniwonder: So put up a patch showing how little RAM-use increase it can be done with.
07:45:06scorche|shthat said, i think you are misunderstanding out "target audience" if you will...also, are you very familiar with embedded development?...everything we add has an impact...each new 'feature' needs to be carefully looked over because every aded code adds to the core size of rockbox...the bigger rockbox gets, the more memory it takes up...the more memory it takes up, the less there is for the audio buffer...the less there is for the audio
07:45:28evilnick_homeI assume that the majority of people do tag their music. Is that an invalid assumption?
07:45:53Mikachuscorche|sh: use enter more :P was it supposed to end at "the less there is for the audio"?
07:46:06scorche|shsome of this stuff might be nice for some people, but if it negatively impacts the rest, it isnt worth implementing without careful consideration
07:46:11scorche|shthe less there is for the audio buffer...the less there is for the audio buffer, the less battery life we get
07:46:59iwonderyou have to know how to code well i guess to do it
07:47:07iwonderif you don't, than the other stuff may suffer
07:47:17iwonderyou have to know the code intimately obviously
07:47:24Mikachuyes, or be able to argue your point without insulting everyone who responds ;)
07:47:32iwonderi don't see it as a impossibility
07:47:35scorche|sheven if you do know how to code well, some things that people (including you) request would still be a massive addition for not much gain
07:47:40iwonderif the nomad jb3 with very limited cpu/memory could do this
07:47:53Lloreaniwonder: "can do" isn't the same as "should do"
07:48:08LloreanIt may be that they could've offered better battery life if they'd been more considerate of what is or isn't truly necessary
07:48:16iwonderwhat is the goal of Rockbox?
07:48:46iwonderto make things easier for people to access, play, & manipulate their music?
07:48:51LloreanPossibly "To be good at playing audio"
07:48:59LloreanPossibly "to be what the people actually writing it want it to be"
07:49:46LloreanYou seem to be missing the point that adding more features slowly makes it a *worse* audio player, which is basically the core function, so it needs to be decided which features are worth degrading this core function for.
07:49:48evilnick_homeiwonder: Most of what you've suggested is already possible using the database. I wonder why you don't want to tag your files.
07:49:51iwonderLlorean, are you very familiar with the code Rockbox uses?
07:50:02iwonderand understand how it works quite well?
07:50:27LloreanWhat's the point of being able to more quickly get to your files for playback, if playback is bad?
07:50:29iwonderif you don't, your statements are pretty baseless
07:50:50iwonderin fact, they are just arrogant ignorance masquerading as somehow true
07:51:21evilnick_homeiwonder: Do you reside in a greenhouse?
07:51:28Lloreaniwonder: You really should be careful throwing random insults around.
07:52:46iwonderi think most would users & programmers that know Rockbox's code well would agree: filter as you type searches is doable & very handy
07:52:52iwonderfor many purposes
07:52:58iwonder-would :)
07:53:14Mikachuand you base this on everyone telling you the opposite?
07:53:39scorche|shno one said it wasnt doable either...we are just saying that it really isnt worth it - especially considering what one can do with database
07:53:50Lloreaniwonder: Nobody said it wasn't doable. They just said it *might* not be worth the cost.
07:54:04Lloreaniwonder: As I said, feel free to prove us wrong by showing how little code it can be done with
07:54:22scorche|shbut if you can come up with a patch that implements it in a very minimal way codebloat-wise, then i would be happy to see it..
07:57:41iwonder-.-
07:58:34Lloreaniwonder: You do realize this is entirely volunteer work - people do what they're interested in. If you want it badly enough, do it. If you don't, don't complain that other programmers really aren't interested in it.
07:59:18iwonderso you speak with the 'other' programmers voice? do you know what they think about this?
07:59:20 Join advcomp2019_ [0] (n=advcomp2@unaffiliated/advcomp2019)
07:59:26iwonderin the way I described filter as you type above?
07:59:34scorche|shiwonder: yes he does...because we are in here talking to you too
07:59:43iwonderif not, you really should not put their voice to make it fit with your own
07:59:53scorche|shhe speaks with his voice, but that voice is mostly in line with the rest of our thoughts
07:59:56Lloreaniwonder: I'm only stating that several other programmers have already told you their opinions.
08:00
08:01:04scorche|shand in the same vein, you really should stop acting as "the users'" voice when you dont knwo for sure what the majority of them think about this
08:01:12evilnick_homeiwonder: I don't understand why you would rather Rockbox implement something that would benefit only a very small number of users, rather than you using a saner directory structure or by tagging your files.
08:01:15*iwonder sighs :Oo
08:01:19*iwonder knocks on wood
08:01:30iwonderyou don't know what i'm talking about then
08:01:45pixelmaiwonder: did you know that you can just let the WPS display the song name (with or without the file name) if there are is e.g. no artist tag? And maybe the database filters could also help you... you should really look into this
08:01:51iwonderjust a basic filter as you type - doesn't have to have to extra features i stated (at least for now)
08:01:57evilnick_homePlease explain how your suggestion is better.
08:02:08pixelmaerrr... I meant "display the filename"
08:02:09Lloreaniwonder: As we've said - feel free to implement it.
08:02:31scorche|shiwonder: things can appear "basic" to users, but programming those features can often turn out to be quite the opposite of basic
08:02:54iwonder:O\
08:02:56iwonderi know this
08:02:59pixelmaand "with or without the extension"
08:03:02iwonder(somewhat :)
08:03:30iwonderone thing leads to another, especially with programming
08:03:48iwonderok, than don't
08:04:18iwondersimple. problem solved. i need to put my attention on stuff that will actually be of help
08:04:59LloreanYou need to put your attention on *doing* stuff, rather than saying "hey, you guys really should do what I say"
08:05:18iwonderprobably because i don't have much energy left
08:05:27iwonderin many ways
08:05:31iwonderanyway, i need to go
08:05:56iwonderhopefully my ideas will eventually be implemented
08:06:00evilnick_homeBy all means come up with suggestions, but the way you phrased them was a little antagonistic
08:06:09iwondernot my ideas, but many others
08:06:38scorche|shand if you want something done, insulting the people you are trying to get to implement them is not the recommended way of accomplishing your goals..
08:06:53iwonderinsulting?
08:06:56iwonderhow so?
08:07:01evilnick_homeAnd your proof is where? You're claiming "many others"
08:07:35iwonderyou don't need proof to know something
08:07:44Lloreaniwonder: Throwing phrases like "arrogant ignorance" around makes you sound insulting.
08:07:56LloreanAnd to know what the user base wants, you do need at least some significant polling.
08:08:08iwonderor you can use your intuition
08:08:24iwonderyou can use your experience of what works for you
08:08:24LloreanThat's not knowing. That's guessing, blindly.
08:08:31evilnick_homeYOU know that YOU could use this new feature. That's one user. Then you claim "many users" so you can easily prove that one user would use it, you cannot prove that many would.
08:08:39LloreanI would bet you a significant amount of money I have more experience with the *Rockbox* userbase than anyone else here.
08:08:48iwonderknowing what works for one, often works for others, especially if it's already been done before and shown to work well
08:09:08LloreanIn fact, I probably have more experience interacting directly with our users than any three people in this room combined.
08:09:11evilnick_homeiwonder: If it works so well, how many other DAPs employ a similar feature?
08:09:25iwonderLlorean, maybe so, but you do not hear the words of those who usually do not speak or voice their opinion
08:09:34scorche|shiwonder: and you do?
08:09:42iwonderi am that voice
08:09:45iwonder;o)
08:09:47scorche|sh...
08:09:49Mikachu(this is hilarious)
08:10:00iwonderat least it makes someone laugh ;)
08:10:09Lloreaniwonder: So you magically know what everyone's been keeping silent about? And yet, our intuitions are all wrong, and your lone one is right?
08:10:29iwonderthis is true, Llorean
08:10:30safetydaniwonder: stick your ideas on the Feature Ideas section of the forum. Then you can get some other users to comment.
08:12:21Lloreaniwonder: If you want people to take you at all seriously, you really need to stop claiming to "know" what anyone other than you wants.
08:12:28r0b-why would rockbox show 441 songs when theres only 435
08:12:47Lloreanr0b-: Recycle binned a few?
08:13:42r0b-even running disktidy didnt do anything
08:13:42evilnick_homer0b-: Run the Disktidy plugin to clear out the Recycle Bin (or equivalent) and then Initialise Database again - that ought to fix it.
08:14:22Mikachudidn't someone say that the count includes non-music files too?
08:14:29Mikachu(although nobody said which count we are talking about yet)
08:14:54r0b-i guess now i check sub folders
08:15:47 Quit advcomp2019 (Read error: 110 (Connection timed out))
08:15:54evilnick_homer0b-: Where are you getting the count from? The stats plugin? The database?
08:16:05r0b-stats plugin
08:16:36evilnick_homeWhat player?
08:16:52r0b-e200
08:20:28*evilnick_home wonders how r0b- knows there are 7 less songs on the DAP
08:20:48r0b-i used the properties of the folders
08:21:07r0b-My Music contains 436 files
08:21:15r0b-theres no other music on the thing
08:21:35r0b-hmmm
08:21:38r0b-just found them :P
08:21:41r0b-Recordings
08:21:41evilnick_homeAha!
08:22:50r0b-for the record "I AM NOT INSANE"
08:23:18r0b-now im going to bed :)
08:30:17 Quit safetydan ("Leaving.")
08:35:49 Quit TheSeven (Read error: 113 (No route to host))
08:38:23 Join ender` [0] (i=krneki@foo.eternallybored.org)
08:40:54 Join Rob2223 [0] (n=Miranda@p4FDCD758.dip.t-dialin.net)
08:42:16 Join pixelma_ [0] (i=quassel@rockbox/staff/pixelma)
08:42:16 Quit pixelma (Nick collision from services.)
08:42:36 Nick pixelma_ is now known as pixelma (i=quassel@rockbox/staff/pixelma)
08:43:28 Quit amiconn (Nick collision from services.)
08:43:32 Join amiconn_ [0] (i=quassel@rockbox/developer/amiconn)
08:43:41 Nick amiconn_ is now known as amiconn (i=quassel@rockbox/developer/amiconn)
08:46:49 Join flydutch [0] (n=flydutch@host87-202-dynamic.15-87-r.retail.telecomitalia.it)
08:49:58 Join bertrik [0] (n=bertrik@ip117-49-211-87.adsl2.static.versatel.nl)
08:55:46 Join daurn [0] (n=daurnima@unaffiliated/daurnimator)
08:58:32 Quit Rob2222 (Read error: 110 (Connection timed out))
09:00
09:01:03***Saving seen data "./dancer.seen"
09:03:06 Join robin0800 [0] (n=robin080@cpc3-brig8-0-0-cust436.brig.cable.ntl.com)
09:11:13 Join Ubuntuxer [0] (n=johannes@dslb-094-220-227-168.pools.arcor-ip.net)
09:12:05 Quit daurnimator (Read error: 101 (Network is unreachable))
09:13:11 Quit daurn| (Read error: 110 (Connection timed out))
09:18:37 Join daurnimator [0] (n=daurnima@unaffiliated/daurnimator)
09:20:05 Quit Stephen__ ("Leaving")
09:26:22 Quit BHSPitMonkey_ ("Ex-Chat")
09:31:04 Join Thundercloud [0] (i=thunderc@persistence.flat.devzero.co.uk)
09:36:44 Quit stripwax ("http://miranda-im.org")
09:39:28 Quit bertrik (Read error: 113 (No route to host))
09:42:11 Join petur [50] (n=petur@rockbox/developer/petur)
09:44:37iwonderanyone see the movie Joe versus The Volcano?
09:45:45markuniwonder: wrong channel
09:45:56iwonderi intended it here
09:45:58iwonder;)
09:47:13 Quit iwonder ()
09:53:02 Quit Thundercloud (Remote closed the connection)
09:53:15 Part Ubuntuxer
09:56:54 Quit MrDuck (Read error: 113 (No route to host))
10:00
10:04:55 Join mt [0] (n=mt@196.221.199.202)
10:24:50mtlinuxstb: Is AAC audio always non-interleaved stereo ? (The only channel configuration I can find in codecs/aac.c)
10:29:26 Join Grahack [0] (n=chri@ip-159.net-82-216-142.rev.numericable.fr)
10:30:21 Part Grahack
10:47:20 Join PaulJam [0] (i=Paule@vpn-3112.gwdg.de)
10:52:21 Join DarkDefender [0] (n=rob@78-69-30-229-no36.tbcn.telia.com)
10:52:35 Join Sajber^ [0] (n=Sajber@h-142-120.A213.priv.bahnhof.se)
10:53:36 Join _zic [0] (n=user@91-165-225-24.rev.libertysurf.net)
11:00
11:01:07***Saving seen data "./dancer.seen"
11:03:22 Join MrDuck [0] (n=kachna@r3g248.net.upc.cz)
11:08:20 Join einhirn [0] (n=Miranda@bsod.rz.tu-clausthal.de)
11:19:39 Quit aditya (Read error: 60 (Operation timed out))
11:32:13 Join aditya [0] (n=aditya@59.95.24.42)
11:52:34 Quit HBK (Read error: 104 (Connection reset by peer))
11:52:50 Join HBK [0] (n=hbk@pool-71-96-74-73.dfw.dsl-w.verizon.net)
11:52:57 Join mankind_ [0] (n=mankind@bl13-2-219.dsl.telepac.pt)
12:00
12:18:27 Quit dash32 ("Verlassend")
12:23:07 Join KBH [0] (n=hbk@pool-71-96-74-73.dfw.dsl-w.verizon.net)
12:26:55 Quit Rob2223 ()
12:29:06 Quit mankind_ ("Leaving")
12:36:49 Quit MrDuck (Read error: 104 (Connection reset by peer))
12:39:15 Quit HBK (Read error: 110 (Connection timed out))
12:40:44 Quit scorche (Read error: 104 (Connection reset by peer))
12:41:44 Join scorche [50] (n=scorche@rockbox/administrator/scorche)
12:53:50 Join vreg [0] (i=vreg@70-101-64-183.dsl1-field.roc.ny.frontiernet.net)
13:00
13:01:11***Saving seen data "./dancer.seen"
13:06:10 Quit Lss__ (Read error: 110 (Connection timed out))
13:16:57 Join Grahack [0] (n=chri@ip-159.net-82-216-142.rev.numericable.fr)
13:23:36 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
13:26:27 Join mankind_ [0] (n=mankind@bl13-2-219.dsl.telepac.pt)
13:27:29 Quit barrywardell ()
13:28:32 Part vreg
13:31:08 Nick fxb is now known as fxb__ (n=felixbru@h1252615.stratoserver.net)
13:31:23 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
13:47:43 Quit CIA-69 (Read error: 60 (Operation timed out))
13:48:54 Join fyrestorm [0] (n=nnscript@cpe-24-90-84-156.nyc.res.rr.com)
14:00
14:00:47 Join CIA-8 [0] (n=CIA@208.69.182.149)
14:02:41 Join barrywardell [0] (n=barry@barry-workstation.ucd.ie)
14:17:30 Join roolku [0] (n=roolku@77-99-113-75.cable.ubr16.sgyl.blueyonder.co.uk)
14:28:24 Quit FlynDice (Remote closed the connection)
14:37:00 Join kachna [0] (n=kachna@r3g248.net.upc.cz)
14:37:22 Join barrywardell_ [0] (n=barrywar@dhcp-892b3c76.ucd.ie)
14:39:35 Quit dmb (Read error: 113 (No route to host))
14:41:18 Quit roolku ()
14:44:01 Nick YpsyZNC is now known as Ypsy (n=ypsy@geekpadawan.de)
14:45:31 Nick fxb__ is now known as fxb (n=felixbru@h1252615.stratoserver.net)
14:46:47 Join mc2739 [0] (n=mc2739@cpe-67-10-234-29.satx.res.rr.com)
14:50:10 Quit __lifeless (Read error: 110 (Connection timed out))
14:52:02 Part mc2739
14:57:04 Join _lifeless [0] (n=lifeless@188.16.84.212)
14:57:39 Join LambdaCalculus37 [0] (n=44a0430d@rockbox/staff/LambdaCalculus37)
14:57:45 Join wincent [0] (n=wincent@host-091-097-067-213.ewe-ip-backbone.de)
14:59:41 Quit barrywardell_ ()
15:00
15:01:16***Saving seen data "./dancer.seen"
15:04:49 Join teru [0] (n=teru@KD059133112132.ppp.dion.ne.jp)
15:08:47 Quit mankind_ ("Leaving")
15:08:47 Quit LambdaCalculus37 ("CGI:IRC (EOF)")
15:15:31 Join LambdaCalculus37 [0] (n=44a0430d@rockbox/staff/LambdaCalculus37)
15:18:39 Join AndyI [0] (i=AndyI@212.14.205.32)
15:33:44 Quit faemir (Read error: 104 (Connection reset by peer))
15:38:28 Join Lss [0] (n=Lss@cm40.delta91.maxonline.com.sg)
15:47:24 Join faemir [0] (n=daniel@78.33.109.163)
15:53:13 Join efyx_ [0] (n=efyx@lap34-1-82-224-140-171.fbx.proxad.net)
15:59:02 Quit kachna (Read error: 110 (Connection timed out))
16:00
16:01:54CIA-8New commit by 03peter (r21993): By Wincent Balin: add a new (better) memory allocation library for pdbox. Next commit will remove dbestfit and make pdbox use this one.
16:03:38linuxstbpetur: In what way is it better?
16:04:06*petur relays the question to wincent
16:04:23peturI think it was better errorchecking
16:06:08peturand he suspected dbestfiit contained a bug
16:06:17*petur shakes wincent awake
16:06:51peturseems he's away for the moment
16:11:15*linuxstb waits...
16:11:37 Join TheSeven [0] (n=theseven@dslb-084-056-133-030.pools.arcor-ip.net)
16:13:05peturI suspect he ran off to do something, I was chatting with him just before the commit
16:15:07 Join n00b81 [0] (n=n00b81@unaffiliated/n00b81)
16:18:03 Quit AlexP (Read error: 104 (Connection reset by peer))
16:19:52 Join AlexP [0] (n=alex@rockbox/staff/AlexP)
16:21:33 Quit Grahack ("Leaving.")
16:23:02 Join jgarvey [0] (n=jgarvey@cpe-098-026-065-013.nc.res.rr.com)
16:26:56linuxstbpetur: Ah, I've just read wincent's mail to the dev list last night. IIUC, he's been trying various "malloc" libraries, and tlsf was the only one he got working correctly.
16:27:14daurntrying to use some lua stuff in sim: "Incompatible version"
16:27:29linuxstbDid you do a "make install" after compiling?
16:27:53 Join BryanJacobs [0] (n=bryanjac@e33.cs.rochester.edu)
16:28:12daurnuh, no :P
16:31:19 Quit _zic ("Ухожу")
16:33:04 Join Ubuntuxer [0] (n=johannes@dslb-094-221-080-249.pools.arcor-ip.net)
16:34:36wincentlinuxstb, peter: Sorry, as Peter suspected correctly, I was away from cumputer.
16:35:16wincentlinuxstb: You are right, the lack of crashing was the cause of choosing TLSF.
16:35:32CIA-8New commit by 03teru (r21994): text_editor: don't set filename "/" before input new filename so that user can easily create copy in same directory. ...
16:35:53linuxstbwincent: OK. I was just curious what "better" meant in the commit message.
16:36:09wincentlinuxstb: Also, it is quite fast.
16:37:57linuxstbwincent: I'm not quite sure what you are asking on the mailing list - are you asking for some kind of profiling, which will tell you which functions are (and are not) called/
16:37:58linuxstb?
16:38:13wincentlinuxstb: Yes.
16:38:53 Part LinusN
16:39:03linuxstbThere are already ways to profile Rockbox, but I've never done it, so don't know the details.
16:39:11BryanJacobslike you'd get from a debugger stack trace?
16:39:29daurnlua plugin looks in wrong places when requiring
16:40:05BryanJacobsfor behavioral analysis, strace is very helpful
16:40:10 Join n1s [0] (n=n1s@rockbox/developer/n1s)
16:40:19BryanJacobsif you strace the sim you'll see every system call made - that includes open()
16:41:45wincentBryanJacobs: I would like to see every function, not every system call.
16:42:05BryanJacobswell then, you can either insert logging statements at the function entry or use breakpoints
16:42:10BryanJacobsor integer counters
16:42:27BryanJacobsor you can use oprof or some such but I'm not sure how that'd work
16:43:15wincentBryanJacobs: Last night I spoke with funman, and what we got to agree on was a scripts that inserts puts("functionname"); into every function.
16:43:30BryanJacobsheh.
16:43:59BryanJacobsthat'll be fun... are you changing the compiler, or just using a macro you add by hand to every function?
16:45:16linuxstbwincent: There's some documentation on profiling in docs/TECH (in the source code).
16:48:03wincentlinuxstb: Profinling is not what I need, because I need to compare function running logs from the original PDa application and from pdbox, so the aforementioned script which enters puts("functionname") is the sole way to go, or so it seems.
16:48:58Unhelpfulyou want not just counts for function calls and time spent in functions, but an in-order list of individual function calls?
16:49:37wincentUnhelpful: Yes, this list is what I intend to receive.
16:50:41Unhelpfuladd a global for the current stack depth, and make it a tree ;)
16:53:20wincentUnhelpful: That will be unhelpful :-)
16:54:38wincentwincent: Because I intend to compare both lists, and the stack depth will not be of much use in this comparison.
16:55:01Unhelpfulyou need to tell yourself that? ;)
16:55:07wincentUnhelpful: −−−−−−−−−−−−−−^
16:55:32wincentUnhelpful: This is what I thought too :->
16:57:42 Join saratoga [0] (i=9803c6dd@gateway/web/freenode/x-af3196804037f01e)
17:00
17:00:15 Part aditya ("WeeChat 0.3.0-rc1")
17:01:18***Saving seen data "./dancer.seen"
17:02:56 Join evilnick [0] (i=0c140464@gateway/web/freenode/x-1c6820546f21cb54)
17:07:14 Join AsaelReiter [0] (n=d44c6247@gateway/web/cgi-irc/labb.contactor.se/x-4373a99547137431)
17:08:33AsaelReiterHi
17:08:45mtlinuxstb: (question ~7 hours ago)
17:10:16AsaelReiterI installed rockbox on my Sansa Clip, and it works mostly fine.
17:11:09Tornei hear a but coming
17:12:10AsaelReiterI saw that you want testers, and I think I can help
17:13:08mtalso, I figured the extradata bytes for aac are for object_type, sampling_rate and channel_configuration, the init function now works fine, as does the frames parsing. Something is still wrong with decoding though (always getting a 'Maximum number of scalefactor bands exceeded' error).
17:13:09Unhelpfulwincent: ltrace and/or gdb can't do this for you? :/
17:13:55AsaelReiterSo, can you add me to the list (http://www.rockbox.org/twiki/bin/view/Main/RockboxTesting)? (or give me write access?)
17:14:17wincentUnhelpful: gdb might do only with breakpoint at the entry of every function, which would be too time-consuming.
17:15:05 Quit teru ("Quit")
17:15:05saratogaif you just want to see what functions are called, can't you use DEBUGF?
17:16:16 Join mankind_ [0] (n=mankind@bl13-2-219.dsl.telepac.pt)
17:16:18wincentUnhelpful: ltrace ... I am looking into it. I know he tool already.
17:16:39wincentsaratoga: Too much DEBUGFs. (Every functionin PDBox... )
17:16:57BryanJacobswhy not be a little more methodical?
17:17:01BryanJacobsuse the mprotect call
17:17:11BryanJacobsprotect every function from reading, install your own segfault handler
17:17:27BryanJacobsand then at the end, the pages which never triggered a SEGV were not read and thus the function was not called
17:17:43linuxstbmt: Without reading through the code, I don't know. But I would be surprised if AAC didn't support mono...
17:17:57BryanJacobsyou'd need to disable ASLR probably in order to get the function-to-address mapping right
17:18:36linuxstbmt: Although thinking about it, maybe Rockbox doesn't understand mono/stereo, and codecs are always required to generate stereo streams...
17:18:40saratogahow many functions are called when you get your crash?
17:19:04saratogayou're using a test file right? presumably it doesn't use every function in the code
17:19:06BryanJacobslinuxstb: isn't there a ci->dsp option for that?
17:20:06wincentsaratoga: The real trick is the question which functions does the original application use, the one that, using the same mem-allocator, does not crash?
17:20:51BryanJacobswincent: all you're trying to do is track down a crash? Can't you just gdb the sim and run a bt when the crash happens?
17:21:38Unhelpfulltrace crashes for me on various functions depending on which library i ask it to trace
17:21:51mtlinuxstb: I'm just asking because of this line : ci->configure(DSP_SET_STEREO_MODE, STEREO_NONINTERLEAVED); - This is not the case for cook, or a number of other codecs. But why are you saying that maybe rockbox doesn't understand stereo/mono ?
17:22:22linuxstbmt: I meant that it's not stored in the metadata struct. But yes, I understand your question now...
17:23:17linuxstbmt: This line seems important - "conf->outputFormat = FAAD_FMT_24BIT; /* irrelevant, we don't convert */" - maybe the AAC decoder has been hacked to always return stereo samples. Or maybe we don't support mono AAC streams, and no-one noticed...
17:24:02wincentBryanJacobs: Did so. The cognition was that an internal variable in mem-allocator becomes a value it never should. After that I made my own allocator, and then I got SEGFAULTs. But the original application worked with my allocator perfectly. Hence the whole "methodicity" :-> .
17:24:03saratogaisn't that line just saying that we don't truncate/clip samples to 24 bit?
17:24:36linuxstbsaratoga: Probably, yes.
17:24:49wincentUnhelpful: ltrace works with PDa (original application) very well. But it does not want to work with simulator.
17:25:19 Join funman [0] (n=fun@rockbox/developer/funman)
17:25:24BryanJacobswell anyhow, the method I described above will work so long as each function is on its own page
17:25:41Tornehm, is the wps supposed to say "Battery: 294h 0m"? :)
17:25:46Torneit's plugged in and fully charged
17:26:05BryanJacobsTorne: maybe it's drawing just a little more than the wall can provide :-)
17:26:15funmanTorne: it means the power consumption of your target has never been measured
17:26:26Tornefunman: it has, honest
17:26:31BryanJacobsor maybe it's just like http://xkcd.com/
17:26:45Torneit displays the right value under other circumstances
17:26:58Torneit's just ipodvideo
17:27:08funmanwhat does system > rockbox info say ?
17:27:22Torne98% 294h 0m
17:28:11Tornebattery status says charged, with ibat=0mA
17:28:43Torneso all indication is that it's fully charged and just sitting there running off the usb power
17:28:47funmanTorne: did you check r21565 ?
17:29:23TorneHrm
17:29:45Torneif i pull the usb cable it immediately becomes the number i expect
17:30:18saratogawhich player?
17:30:23Torneipodvideo
17:31:12funmanTorne: i believe when charging, the number displayed is the time needed until 100% charged ?
17:31:23BryanJacobsthat's a long time to cap things off
17:31:28Tornefunman: that's possible. i've not really paid attention to it
17:31:33AsaelReitercan somebody add me to http://www.rockbox.org/twiki/bin/view/Main/RockboxTesting ?
17:31:40Tornebut it says "charged", not charging, for the status
17:31:52Torneand it's with the charging patch so it is able to pull 500mA from the USB port
17:32:01Tornethe PCF says there is no current flowing in or out of the battery
17:32:21saratogai didn't even know we tried to estimate charging time
17:32:30funmanAsaelReiter: what is your wiki name ? i can give you writing rights so you could do it yourself ;)
17:32:36wincentUnhelpful: ltrace gets stuck on SDL_Init . What do you think of it?
17:33:01Torneaha
17:33:12Tornethere's no CURRENT_USB definition for ipods
17:33:17Tornemaybe that's the culprit?
17:33:33Torneit only has CURRENT_{NORMAL,BACKLIGHT,RECORD}
17:33:43wincentDoes anybody have an experience how to run the simulator under ltrace?
17:33:49AsaelReiterfunman: my wiki name is AsaelReiter
17:34:02saratogathats not defined for anything but archos, are you sure its still used?
17:34:18Torneit looks uses in powermgmt.c, yes
17:34:36Torneunder a complicated set of conditions that you can look at for yourself :)
17:34:46Tornebut which i believe does apply to the ipod
17:35:00linuxstbAdding a #warning would confirm...
17:35:20Torne..hm, this makes it rather unclear what CURRENT_USB is supposed ot be
17:35:50TorneI've not really used it with usb power connected but not in usb mode before
17:35:54 Quit petur ("and this concludes our afternoon session")
17:35:56Torneso had not previously noticed
17:36:10saratogai don't understand how you would estimate battery charging time using a single current number . . .
17:36:15funmanAsaelReiter: now you should be able to modify the wiki
17:36:30AsaelReiterthanks!
17:36:34Tornesaratoga: it's a really rough guess :)
17:36:44Torneoh wait
17:36:46Torneno, i see what yo umean
17:37:17Torneit's not using a single number
17:37:23Torneit's CURRENT_MAX_CHG - runcurrent()
17:38:03Torne...and CURRENT_MAX_CHG is not defined for most targets either
17:38:06saratogadefined for a total of two targets
17:38:11Torneit defaults to 350
17:38:11saratoga1 unsupported
17:38:14Tornewhich is entirely arbitrary
17:38:29AsaelReiterAnd thanks again for the work on the AMS models.
17:38:48Torneso it is indeed working out how long it takes for the last 2%
17:38:53Tornewhich the power controller is ignoring anyway
17:38:57saratogai don't think the logic in there even works anyway
17:39:00 Join barrywardell_ [0] (n=barrywar@dhcp-892b3c76.ucd.ie)
17:39:01Torneand it's doing it based on entirely bogus values
17:39:03saratogaat least for a lipoly battery
17:39:06Torneokay then :)
17:39:08 Quit barrywardell_ (Remote closed the connection)
17:39:27Tornethe pcf has happily stopped charging at what rockbox thinks is 98%
17:39:43saratogai think for that to work you'd want to ask the charge controller how much current its feeding the battery like I htink the gigabeatS does
17:39:52Torneanyway, that answers the question
17:40:11saratogaalthough just disabling that on targets without CURRENT_MAX_CHG would probably be the best solution
17:40:23TorneYah, possibly..
17:41:49 Quit AsaelReiter ("CGI:IRC")
17:41:56*Torne puts it on his list :)
17:42:07funmanAsaelReiter: if you want to help with the Clip port, something that doesn't need programming skills is looking at keymap (I remember radio screen keymap is pretty bad, but perhaps i'm confused with the Fuze keymap..)
17:42:13 Join dash32 [0] (n=dash32@p54AB5406.dip.t-dialin.net)
17:42:59 Quit Ubuntuxer ("Leaving.")
17:44:16 Quit dash32 (Client Quit)
17:44:54Torneyeah, it's not the charging time..
17:45:16Torneif i let it discharge a bit then plug it in again it comes up with a charging time, which is a couple minutes and goes down too fast (because the numbers are wrong)
17:45:48Tornebut once it fills up and the charging icon vanishes it immediately goes to being huge again. :)
17:46:40*Torne will investigate properly later.
17:48:15 Quit BryanJacobs ("Java user signed off")
17:48:19 Join BryanJacobs [0] (n=bryanjac@e33.cs.rochester.edu)
17:48:48LambdaCalculus37linuxstb: Question... I tried iBugger again to load the Rockbox bootloader onto my nano2g. When I invoke control.py execute 0x222000000, am I supposed to see something immediately?
17:49:37linuxstbLambdaCalculus37: Things are a bit unreliable, so it depends on what versions of things you're using. Are you using iBugger-0.1e and latest SVN? Also, the address 0x22000000
17:50:36LambdaCalculus37linuxstb: I just noticed I entered it wrong. :P
17:50:56LambdaCalculus37Also I didn't see iBugger-0.1e was released.
17:51:16*LambdaCalculus37 shall update iBugger and try again
17:51:29 Quit funman ("free(random());")
17:52:43linuxstbLambdaCalculus37: Yes, for some reason Rockbox stopped working with 0.1d... I never bothered to investigate though (things are too early in the port).
17:54:21 Quit mt (Read error: 113 (No route to host))
17:56:43LambdaCalculus37linuxstb: I was wondering why iBugger-0.1c was working for me at first and 0.1d wasn't. But at least I know that I can try again.
17:56:46 Join fdinel [0] (n=Miranda@modemcable204.232-203-24.mc.videotron.ca)
17:57:52LambdaCalculus37linuxstb: Also (slightly OT) how do you sleep your nano2g when the main menu on the OF is frozen and the nano isn't responding to button presses except MENU+SELECT?
17:58:13linuxstbLambdaCalculus37: The syntax for iControl.py has changed slightly - you need to give it both an execute address and stack address. Rockbox sets its own stack, so I just use 0 for that.
17:58:28linuxstbLambdaCalculus37: I go into diagnostics mode, then Sleep -> Deep Sleep Test
17:58:54linuxstb(or remove the notes file)
17:59:26TheSevenlinuxstb: you won't need you "return to ibugger" code any more :-P
17:59:57TheSevenibugger is now able of re-uploading the code while it's running
18:00
18:01:57 Join JdGordon| [0] (i=441b0728@gateway/web/freenode/x-e03088d6abd28ec2)
18:10:53 Quit n1s (Read error: 110 (Connection timed out))
18:21:26 Quit faemir (Read error: 60 (Operation timed out))
18:30:37 Quit BryanJacobs ("Java user signed off")
18:31:17 Join tmzt_ [0] (n=tmzt@adsl-76-244-147-106.dsl.akrnoh.sbcglobal.net)
18:33:12JdGordon|prize for anyone that can come up with a reasonable explanation why viewports break cuesheets in the WPS... FS #10460
18:36:30 Join pixthor [0] (n=48c1aacc@gateway/web/cgi-irc/labb.contactor.se/x-cbb06ddabf75e651)
18:36:31PaulJamdoesn't cabbie use viewports too? he writes the inbuilt ones are all ok.
18:37:00gevaertsit doesn't
18:39:18 Quit JdGordon| (Ping timeout: 180 seconds)
18:40:49PaulJamok. (the h300 version uses viewports, so i thought this is true for the other targets too.)
18:41:07*linuxstb would first challenge the assumption that it's related to viewports...
18:41:19 Join n1s [0] (n=n1s@rockbox/developer/n1s)
18:41:59 Quit pixthor ("CGI:IRC (Ping timeout)")
18:42:28 Quit tmzt (Connection timed out)
18:42:29 Join bertrik [0] (n=bertrik@ip117-49-211-87.adsl2.static.versatel.nl)
18:44:49 Nick fxb is now known as fxb__ (n=felixbru@h1252615.stratoserver.net)
18:45:22 Join pixthor [0] (n=48c1aacc@gateway/web/cgi-irc/labb.contactor.se/x-2b1cf7df1f358d42)
18:49:30 Join JdGordon| [0] (n=Miranda@nat/microsoft/x-0ba61d5e77ae6594)
18:50:50 Quit pixthor ("CGI:IRC (Ping timeout)")
18:52:58 Join kkurbjunW [0] (n=karlk@12.41.166.8)
18:53:19kkurbjunWgevaerts: I didn't realize that you had an M:Robe 500 too
18:53:33gevaertskkurbjunW: I haven't had it for long :)
18:53:38 Join pixthor [0] (n=48c1aacc@gateway/web/cgi-irc/labb.contactor.se/x-f166b3c66a8b5183)
18:54:04kkurbjunW:), have you gotten a chance to run rockbox on it yet?
18:54:38gevaertsit's installed, yes. I've even tried a 640x480 build
18:54:54*gevaerts wanted to see exactly how slow mpegplayer would be :)
18:54:54CtcpPing from gevaerts!n=fg@rockbox/developer/gevaerts
18:55:05kkurbjunWoh, nice :)
18:56:00kkurbjunWif you are interested in messing around with the hardware on it I can point you to some documentation and information we have on the player
18:56:00 Quit pixthor (Client Quit)
18:56:55kkurbjunWwhat's your take on the supportable status for the player? Do you think it's in a usable state for most people?
18:57:20gevaertsIt seems to work fine for general playback, so yes
19:00
19:00:24 Nick Ypsy is now known as YpsyZNC (n=ypsy@geekpadawan.de)
19:01:15 Nick YpsyZNC is now known as Ypsy (n=ypsy@geekpadawan.de)
19:01:22***Saving seen data "./dancer.seen"
19:03:25 Join _zic [0] (n=user@91-165-225-24.rev.libertysurf.net)
19:03:46 Quit robin0800 ("Leaving")
19:04:28 Join HellDragon_ [0] (i=jd@modemcable178.248-201-24.mc.videotron.ca)
19:04:32*ej0rge hasn't installed rockbox on either of his mr500
19:04:35 Quit HellDragon_ (Read error: 104 (Connection reset by peer))
19:04:42 Quit HellDragon (Read error: 104 (Connection reset by peer))
19:06:38kkurbjunWej0rge: you should try it :), installation is pretty easy - the wiki has general instructions and the downloads
19:09:14ej0rgekkurbjunW: It'll rise to the top priority soon enough. i hope. work/life balance isn't good lately.
19:09:41 Join fyre^OS [0] (n=nnscript@cpe-24-90-84-156.nyc.res.rr.com)
19:12:04 Quit LambdaCalculus37 (simmons.freenode.net irc.freenode.net)
19:12:04NSplitsimmons.freenode.net irc.freenode.net
19:12:04 Quit fyrestorm (simmons.freenode.net irc.freenode.net)
19:12:04 Quit jhulst (simmons.freenode.net irc.freenode.net)
19:12:04 Quit goffa___ (simmons.freenode.net irc.freenode.net)
19:12:04 Quit fred_2 (simmons.freenode.net irc.freenode.net)
19:12:04 Quit r00s (simmons.freenode.net irc.freenode.net)
19:12:14NHealsimmons.freenode.net irc.freenode.net
19:12:14NJoinfred_2 [0] (i=fred@hpc-cluster.hamburgnet.de)
19:12:19 Join robin0800 [0] (n=robin080@cpc3-brig8-0-0-cust436.brig.cable.ntl.com)
19:12:45 Join HellDragon [0] (i=christop@modemcable178.248-201-24.mc.videotron.ca)
19:13:18 Quit kkurbjunW (Remote closed the connection)
19:13:20NJoinjhulst [0] (n=jhulst@jhulst.com)
19:14:03 Quit HellDragon (Success)
19:14:04 Join kkurbjunW [0] (n=karlk@12.41.166.9)
19:14:06 Join HellDragon_ [0] (i=jd@modemcable178.248-201-24.mc.videotron.ca)
19:14:12 Join goffa [0] (n=goffa@216.220.23.105)
19:14:48 Quit kkurbjunW (Remote closed the connection)
19:15:41 Join kkurbjunW [0] (n=karlk@12.41.166.8)
19:18:15 Nick tmzt_ is now known as tmzt (n=tmzt@adsl-76-244-147-106.dsl.akrnoh.sbcglobal.net)
19:20:50ej0rgegetting home from work 13 hours after leaving for work in the morning stinks.
19:21:11kkurbjunWouch, I would agree
19:21:58ej0rgeI got home at 8:45pm last night.
19:22:52ej0rgeI've got 4 more weeks of this to look forward to.
19:26:57 Join Horscht [0] (n=Horscht2@xbmc/user/horscht)
19:29:51CIA-8New commit by 03Ubuntuxer (r21995): Add a standard menu, add highscore table and clean up the code in spacerocks (part of FS #10193)
19:30:29PaulJamJdGordon|: well, i tried a cuesheet on my h300 and i can at least confirm that there is something wrong when the WPS has viewports (at least that was the common factor for the themes i tried). What happens for me is that the tilie always shows that from the first song when the WPS has viewports. a wps with no viewports shows the correct title. one thing that may give a clue is that upon entering a non-VP WPS the wrong title (from the first song) is sh
19:30:29PaulJamown for a split second and then changes to the right title.
19:30:32 Join Ubuntuxer [0] (n=johannes@dslb-094-221-080-143.pools.arcor-ip.net)
19:31:05PaulJams/tilie/title
19:33:08JdGordon|yeah, except when the wps is redrawn the id3 should already have been spoofed, and even if it hadnt, it will be spoofed next draw which will force a full redraw
19:33:32linuxstbUbuntuxer: What do you mean with "standard menu" ? The standard "in-game" plugin menu, or the standard "before the game starts" menu?
19:35:39 Quit amiconn (Nick collision from services.)
19:35:40 Join AlexP_ [0] (n=alex@rockbox/staff/AlexP)
19:35:43 Join amiconn_ [0] (i=quassel@rockbox/developer/amiconn)
19:35:44UbuntuxerI added a menu, which looks similiar to other game plugin menus.
19:35:45 Join HellDragon [0] (i=jd@modemcable178.248-201-24.mc.videotron.ca)
19:35:50 Nick amiconn_ is now known as amiconn (i=quassel@rockbox/developer/amiconn)
19:36:21 Join obo_ [0] (n=obo@77-99-230-49.cable.ubr04.trow.blueyonder.co.uk)
19:36:25 Join JdGordon|| [0] (n=Miranda@nat/microsoft/x-99533a8e7b5a2828)
19:37:25 Quit HellDragon_ (simmons.freenode.net irc.freenode.net)
19:37:25 Quit robin0800 (simmons.freenode.net irc.freenode.net)
19:37:25 Quit JdGordon| (simmons.freenode.net irc.freenode.net)
19:37:25 Quit bertrik (simmons.freenode.net irc.freenode.net)
19:37:25 Quit fdinel (simmons.freenode.net irc.freenode.net)
19:37:25 Quit mankind_ (simmons.freenode.net irc.freenode.net)
19:37:25 Quit AlexP (simmons.freenode.net irc.freenode.net)
19:37:25 Quit n00b81 (simmons.freenode.net irc.freenode.net)
19:37:25 Quit scorche (simmons.freenode.net irc.freenode.net)
19:37:25 Quit pixelma (simmons.freenode.net irc.freenode.net)
19:37:25 Quit r0b- (simmons.freenode.net irc.freenode.net)
19:37:26 Quit MuscleNerd (simmons.freenode.net irc.freenode.net)
19:37:26 Quit Kohlrabi (simmons.freenode.net irc.freenode.net)
19:37:26 Quit sbhsu (simmons.freenode.net irc.freenode.net)
19:37:26 Quit Ypsy (simmons.freenode.net irc.freenode.net)
19:37:26 Quit obo (simmons.freenode.net irc.freenode.net)
19:37:26 Quit JdGordon (simmons.freenode.net irc.freenode.net)
19:37:26 Quit sitwon (simmons.freenode.net irc.freenode.net)
19:37:26 Quit J-23 (simmons.freenode.net irc.freenode.net)
19:37:26UbuntuxerThere is displayed a entry resume game, if you ingame
19:37:31NJoinpixelma [0] (i=quassel@rockbox/staff/pixelma)
19:38:08 Join HellDragon_ [0] (i=jd@Wikipedia/HellDragon)
19:38:08NJoinrobin0800 [0] (n=robin080@cpc3-brig8-0-0-cust436.brig.cable.ntl.com)
19:38:08NJoinJdGordon| [0] (n=Miranda@nat/microsoft/x-0ba61d5e77ae6594)
19:38:08NJoinbertrik [0] (n=bertrik@ip117-49-211-87.adsl2.static.versatel.nl)
19:38:08NJoinfdinel [0] (n=Miranda@modemcable204.232-203-24.mc.videotron.ca)
19:38:08NJoinmankind_ [0] (n=mankind@bl13-2-219.dsl.telepac.pt)
19:38:08NJoinn00b81 [0] (n=n00b81@unaffiliated/n00b81)
19:38:08NJoinscorche [50] (n=scorche@rockbox/administrator/scorche)
19:38:08NJoinr0b- [0] (n=rob@adsl-76-235-183-168.dsl.klmzmi.sbcglobal.net)
19:38:08NJoinJdGordon [0] (n=jonno@rockbox/developer/JdGordon)
19:38:08NJoinJ-23 [0] (n=zelazko@unix.net.pl)
19:38:08NJoinYpsy [0] (n=ypsy@geekpadawan.de)
19:38:08NJoinsitwon [0] (n=adam@pool-173-79-65-210.washdc.fios.verizon.net)
19:38:08NJoinsbhsu [0] (n=a6530466@Zion.dorm.au.edu.tw)
19:38:08NJoinKohlrabi [0] (n=Kohlrabi@frustrum.nosebud.de)
19:38:08NJoinMuscleNerd [0] (i=eric@adsl-75-31-138-204.dsl.irvnca.sbcglobal.net)
19:38:36 Quit JdGordon| (Connection reset by peer)
19:38:54 Quit TheSeven (Nick collision from services.)
19:39:09 Join The_Seven [0] (n=theseven@dslb-084-056-133-030.pools.arcor-ip.net)
19:39:13 Nick The_Seven is now known as TheSeven (n=theseven@dslb-084-056-133-030.pools.arcor-ip.net)
19:39:57 Join Stephen__ [0] (n=S@86-40-168-77-dynamic.b-ras2.srl.dublin.eircom.net)
19:40:26 Quit barrywardell (Remote closed the connection)
19:41:26 Quit HellDragon_ (SendQ exceeded)
19:41:55 Quit Ubuntuxer ("Leaving.")
19:42:20 Join stripwax [0] (n=Miranda@87-194-34-169.bethere.co.uk)
19:44:22 Nick AlexP_ is now known as AlexP (n=alex@rockbox/staff/AlexP)
19:47:16 Quit goffa (simmons.freenode.net irc.freenode.net)
19:47:16NSplitsimmons.freenode.net irc.freenode.net
19:47:16 Quit jhulst (simmons.freenode.net irc.freenode.net)
19:47:16 Quit AndyI (simmons.freenode.net irc.freenode.net)
19:47:16 Quit daurnimator (simmons.freenode.net irc.freenode.net)
19:48:13bertrikI think I'll have another go at s5l8700 usb tonight
19:49:13NHealsimmons.freenode.net irc.freenode.net
19:49:13NJoingoffa [0] (n=goffa@216.220.23.105)
19:49:20NJoinjhulst [0] (n=jhulst@jhulst.com)
19:49:20NJoinAndyI [0] (i=AndyI@212.14.205.32)
19:49:20NJoindaurnimator [0] (n=daurnima@unaffiliated/daurnimator)
19:53:12 Quit n1s (Read error: 60 (Operation timed out))
19:53:25 Quit n00b81 (Read error: 110 (Connection timed out))
19:54:33 Quit fdinel (Connection timed out)
19:56:37 Quit crwl (Read error: 104 (Connection reset by peer))
19:57:06 Join crwl [0] (n=crawlie@a91-156-100-168.elisa-laajakaista.fi)
19:58:49 Join LambdaCalculus37 [0] (n=44a0430d@rockbox/staff/LambdaCalculus37)
19:59:26 Join stripwax5443 [0] (n=Miranda@87-194-34-169.bethere.co.uk)
19:59:30saratogaadding a couple lines of code was enough to overfill text_editor on archos?
19:59:33saratogahow short on space was it?
20:00
20:00:06 Quit stripwax (simmons.freenode.net irc.freenode.net)
20:00:06 Quit fyre^OS (simmons.freenode.net irc.freenode.net)
20:00:06 Quit AfterDeath (simmons.freenode.net irc.freenode.net)
20:00:06 Join funman [0] (n=fun@rockbox/developer/funman)
20:00:58linuxstbsaratoga: Obviously very short...
20:01:22saratogashould i just reduce the number of characters allowed in the buffer by a hundred or so?
20:02:02gevaertssaratoga: I think the correct way would be to use the plugin buffer properly. A lot more work though
20:02:49saratogaproperly how?
20:03:17NJoinAfterDeath [0] (n=icxcnika@freenode/weird-exception/network-troll/afterdeath)
20:03:35NJoinstripwax [0] (n=Miranda@87-194-34-169.bethere.co.uk)
20:03:35NJoinfyre^OS [0] (n=nnscript@cpe-24-90-84-156.nyc.res.rr.com)
20:03:42gevaertsusing rb->plugin_get_buffer()
20:04:17gevaertsthat just gets all available space. No need to guess
20:04:27saratogathat stops playback?
20:04:40 Quit fyre^OS (Read error: 54 (Connection reset by peer))
20:04:52JdGordon||no
20:04:53 Join fyrestorm [0] (n=nnscript@cpe-24-90-84-156.nyc.res.rr.com)
20:05:03 Nick JdGordon|| is now known as JdGordon| (n=Miranda@nat/microsoft/x-99533a8e7b5a2828)
20:05:20gevaertsno. This one gets all memory from the plugin buffer that's not already used by the plugin
20:07:29 Quit stripwax (Read error: 60 (Operation timed out))
20:17:25 Join Rob2222 [0] (n=Miranda@p4FDCD758.dip.t-dialin.net)
20:17:37 Join fyre^OS [0] (n=nnscript@cpe-24-90-84-156.nyc.res.rr.com)
20:18:06 Quit scorche (Nick collision from services.)
20:18:53 Join scorche [50] (n=scorche@rockbox/administrator/scorche)
20:21:14 Quit fyrestorm (simmons.freenode.net irc.freenode.net)
20:21:14NSplitsimmons.freenode.net irc.freenode.net
20:21:14 Quit MuscleNerd (simmons.freenode.net irc.freenode.net)
20:21:14 Quit Kohlrabi (simmons.freenode.net irc.freenode.net)
20:21:14 Quit sbhsu (simmons.freenode.net irc.freenode.net)
20:21:14 Quit Ypsy (simmons.freenode.net irc.freenode.net)
20:21:14 Quit JdGordon (simmons.freenode.net irc.freenode.net)
20:21:14 Quit r0b- (simmons.freenode.net irc.freenode.net)
20:21:14 Quit mankind_ (simmons.freenode.net irc.freenode.net)
20:21:14 Quit bertrik (simmons.freenode.net irc.freenode.net)
20:21:14 Quit robin0800 (simmons.freenode.net irc.freenode.net)
20:21:14 Quit sitwon (simmons.freenode.net irc.freenode.net)
20:21:15 Quit J-23 (simmons.freenode.net irc.freenode.net)
20:21:16 Join FlynDice [0] (n=FlynDice@12.173.189.3)
20:21:50NHealsimmons.freenode.net irc.freenode.net
20:21:50NJoinsbhsu [0] (n=a6530466@Zion.dorm.au.edu.tw)
20:22:10NJoinsitwon [0] (n=adam@pool-173-79-65-210.washdc.fios.verizon.net)
20:22:21NJoinKohlrabi [0] (n=Kohlrabi@frustrum.nosebud.de)
20:22:25 Quit DarkDefender (Read error: 110 (Connection timed out))
20:23:44NJoinMuscleNerd [0] (i=eric@adsl-75-31-138-204.dsl.irvnca.sbcglobal.net)
20:26:22NJoinJ-23 [0] (n=zelazko@unix.net.pl)
20:27:39 Join Zagor [242] (n=bjst@rockbox/developer/Zagor)
20:33:53NJoinrobin0800 [0] (n=robin080@cpc3-brig8-0-0-cust436.brig.cable.ntl.com)
20:34:14NJoinbertrik [0] (n=bertrik@ip117-49-211-87.adsl2.static.versatel.nl)
20:34:22NJoinYpsy [0] (n=ypsy@geekpadawan.de)
20:35:32 Nick Ypsy is now known as YpsyZNC (n=ypsy@geekpadawan.de)
20:36:00NJoinJdGordon [0] (n=jonno@rockbox/developer/JdGordon)
20:36:19 Nick YpsyZNC is now known as Ypsy (n=ypsy@geekpadawan.de)
20:39:30ZagorFYI: I'll be testing the new best-fit server code, which means a lot of server restarts. do not be alarmed.
20:45:52 Quit einhirn ("Miranda IM! Smaller, Faster, Easier. http://miranda-im.org")
20:47:23pixelmaJdGordon|: widecabbie has viewports but still draws in the main screen
20:49:51pixelmaI should update and try on one of my targets and my theme which has quite a lot of viewports but nothing in the "main screen"
20:51:28LambdaCalculus37The GoGear SA9200 needs a Cabbiev2 port still.
20:51:38LambdaCalculus37The LCD is 128x160 on that one.
20:54:22JdGordon|pixelma: umm, does it actually draw anything in the main viewport? I'm pretty sure we made it so if any viewports are defined nothing will draw in the default one
20:55:54pixelmaprogress bar and some icons (hold etc.) are defined before the first %V
20:56:34JdGordon|defined is ok, where do they actually get drawn?
20:56:40pixelmaI still have the file open in my text editor, if you want I can paste it somewhere
20:56:40JdGordon|the tags to draw them i mean
20:56:58JdGordon|na better not.. need to work :p
20:57:57pixelmawell, progress bar is only one tag now so I assume it's "define and draw" and I meant the %xd tags for the battery icons
20:58:22pixelmaand hold icons etc.
21:00
21:01:25***Saving seen data "./dancer.seen"
21:11:33 Quit FlynDice (Remote closed the connection)
21:12:34UnhelpfulZagor: "best-fit"... as in trying to find the best fit of compute resources to builds?
21:12:40ZagorUnhelpful: yes
21:13:25Zagor"Total work: 1177729 points", "Total speed: 6235 points/sec (49 clients)", "Ideal time: 189 seconds", "Realistic time: 192.78 seconds"
21:13:33Unhelpfulwith the goal of keeping all clients busy and minimizing lost work?
21:14:01Zagoryes, with the overall goal to provide the fastest possible build rounds
21:14:30Unhelpfulenjoy solving knapsack ;)
21:14:39Zagor:)
21:15:37 Quit funman ("free(random());")
21:15:54JdGordon|the server killed again? I've got a "starting client" line but no hello?
21:26:41ZagorJdGordon|: yes, I'm futzing with it
21:27:09JdGordon|righto, I assumed we'd be seeing more stop starting :)
21:27:19*JdGordon| hides putty and gets back to work :p
21:27:40 Join Hillshum [0] (n=hillshum@75-165-246-39.slkc.qwest.net)
21:27:49 Quit Hillshum (Remote closed the connection)
21:29:30 Join r0b- [0] (n=rob@adsl-76-235-183-168.dsl.klmzmi.sbcglobal.net)
21:31:59 Join Hillshum [0] (n=hillshum@75-165-246-39.slkc.qwest.net)
21:35:03 Quit stripwax5443 ("http://miranda-im.org")
21:37:50 Join mcuelenaere [0] (n=mcuelena@78-21-191-122.access.telenet.be)
21:38:23HillshumI'm compiling qt for rbutil and it's giving me errors over the lang files like the wiki said it would. But is it supposed to keep going or stop after those errors?
21:47:56HillshumIs the qmake part supposed to take quite a while?
21:48:13 Join Strife89 [0] (n=Michael@168.16.239.215)
21:48:31JdGordon|doesnt it fix up qt to statically link? so yeah maybe it does take ages
21:49:05HillshumMine was real fast
21:51:02saratogadoes a build server need the rbutil and manual source code? i'm trying to slim it down as much as possible
21:52:08saratogaactually, if I don't have sdl anyway, can I remove the uisimulator?
21:52:55JdGordon|you're really that strapped for space?
21:53:06gevaertssaratoga: you're going to get M builds then
21:53:12JdGordon|and no.. it does an svn up so it will grab them again anyway
21:53:15saratogaits not my server so I'd really not waste much disk if I can avoid it
21:56:05saratogai can run whatever i want but if I'm using a lot of disk space someone might ask if I really need to have a complete source code checkout in my home folder
21:56:58Hillshumdefine "a lot of disk space"
21:57:09saratogano
21:59:45 Quit LambdaCalculus37 ("CGI:IRC 0.5.9 (2006/06/06)")
22:00
22:07:53HillshumNow I'm getting this trying to compile http://rockbox.pastebin.com/f10e22788
22:08:47saratogamaybe install libusb?
22:11:43 Join Grahack [0] (n=chri@ip-159.net-82-216-142.rev.numericable.fr)
22:12:03 Join LambdaCalculus37 [0] (n=rmenes@rockbox/staff/LambdaCalculus37)
22:12:17Hillshumsaratoga, will this one from the ubnuntu repos work libusb++-0.1-4c2 ?
22:15:24bertrikI'm not so sure anymore if I should try to re-use the tcc-usb.c driver for the s5l8700 based players
22:17:52 Nick Ypsy is now known as YpsyZNC (n=ypsy@geekpadawan.de)
22:27:13 Join kachna [0] (n=kachna@r4ax178.net.upc.cz)
22:31:07 Join Thundercloud [0] (i=thunderc@persistence.flat.devzero.co.uk)
22:39:56 Join petur [0] (n=peter@rockbox/developer/petur)
22:41:53 Quit _zic (Read error: 101 (Network is unreachable))
22:45:11 Join fwaokda [0] (n=fwaokda@adsl-157-16-182.jan.bellsouth.net)
22:45:32fwaokdaIs the Sansa Fuze supported with RockBox?
22:45:50JdGordon|we are currently discussing that question
22:45:52saratoganope
22:45:56LambdaCalculus37No.
22:45:59JdGordon|currently being on the mailing list
22:46:05LambdaCalculus37It works, but it's not what we call "supported".
22:46:22saratogaits definately not going to be supported while we're probably not clocking the SD card correctly
22:46:58pixelmanice, with cuesheet support enabled I get a data abort on my c200 (with an older build which I wanted to try for comparison reasons first)
22:47:16Hillshumwe "could" release without SD card support right?
22:47:58fwaokdaSo I shouldn't use it if I'm using a SDHC card I gues...
22:48:28 Join stripwax [0] (n=Miranda@87-194-34-169.bethere.co.uk)
22:48:38*gevaerts isn't entirely sure if this clocking issue only applies to the SD slot
22:49:28Hillshumfwaokda, If you're willing to reformat the card more often then go ahead
22:49:41fwaokdaHillshum, nah im good
22:49:44fwaokdaty though ;)
22:51:05 Quit saratoga ("Page closed")
22:54:30 Join FlynDice [0] (n=FlynDice@c-24-19-225-90.hsd1.wa.comcast.net)
22:55:02pixelmahmm... teru left some red in the build table... :\
22:55:18 Quit mcuelenaere ()
22:57:08gevaertsZagor: could we have the mail-on-red back?
22:57:12JdGordon|yes and no....
22:57:26JdGordon|the red is the plugin being too big isnt it?
22:57:36Zagorgevaerts: certainly. but not right now.
22:57:44gevaertsJdGordon|: yes
22:57:57 Join n1s [0] (n=n1s@rockbox/developer/n1s)
22:58:07gevaertsEasy to fix by just reducing the thing of course
22:58:22gevaertsI'm just not sure if that's the right fix
22:58:34 Join Zarggg_ [0] (n=zarggg@65-78-69-194.c3-0.eas-ubr6.atw-eas.pa.cable.rcn.com)
22:58:41 Quit Zarggg_ (Client Quit)
22:58:46pixelmawhich "thing" to reduce?
22:58:55gevaertsZagor: did you put the remaining tasks through the best-fit algorithm as well? ;)
22:59:03Zagor:)
22:59:20 Quit fyre^OS (Read error: 104 (Connection reset by peer))
22:59:24gevaertspixelma: MAX_CHARS
22:59:31 Join fyrestorm [0] (n=nnscript@cpe-24-90-84-156.nyc.res.rr.com)
23:00
23:00:17pixelmawhat's MAX_CHARS used for?
23:01:29***Saving seen data "./dancer.seen"
23:02:08gevaertsif I read things correctly, it's the maximum file size it can handle
23:02:31 Join dmb [0] (n=Dmb@unaffiliated/dmb)
23:04:14 Join stripwax5443 [0] (n=Miranda@87-194-34-169.bethere.co.uk)
23:04:50gevaertsI think that buffer array should be replaced by a pointer, and it should use plugin_get_buffer(). That won't break on code changes and it will always use every possible byte. If that's done, getting the audio buffer for larger files is also easier to implement
23:05:09*gevaerts will try to have a look at this tomorrow
23:07:30 Quit r0b- (Read error: 110 (Connection timed out))
23:08:20 Quit Zarggg (Read error: 104 (Connection reset by peer))
23:08:32*JdGordon| was going to have a look at it, but is happy to leave it for gevaerts
23:09:05 Part Grahack
23:09:21 Quit stripwax (Read error: 60 (Operation timed out))
23:09:52 Join B4gder [241] (n=daniel@rockbox/developer/bagder)
23:10:24gevaertsI guess in the short term we should fix this red though...
23:10:53*stripwax5443 hmms
23:10:56 Nick stripwax5443 is now known as stripwax (n=Miranda@87-194-34-169.bethere.co.uk)
23:11:27*gevaerts gets busy
23:14:39 Join Zarggg [0] (n=zarggg@65-78-69-194.c3-0.eas-ubr6.atw-eas.pa.cable.rcn.com)
23:15:49 Quit fwaokda (Remote closed the connection)
23:19:51CIA-8New commit by 03gevaerts (r21996): Decrease the text buffer a tiny bit to make it fit again on archos. This really should use plugin_get_buffer() (and possibly plugin_get_audio_buffer() ...
23:21:18 Quit LambdaCalculus37 ("Fwump")
23:21:36 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
23:26:53 Join webguest38 [0] (n=186dc797@gateway/web/cgi-irc/labb.contactor.se/x-56f52861725d7535)
23:27:53 Quit webguest38 (Client Quit)
23:29:21 Join webguest21 [0] (n=186dc797@gateway/web/cgi-irc/labb.contactor.se/x-150f46656fd77f13)
23:30:46webguest21i had changed some files then rebooted and went to playlist>insert and now my ipod nano 1g is locked up and won't do anything when i plug it into my computer
23:31:28Hillshumthere's a certain button combo that should reset. google should know it
23:31:46AlexPmenu + select
23:32:06AlexPyou might have to toggle old on then off beforehand
23:32:16webguest21cool worked
23:32:38webguest21is there an easier way to update songs without having to reinstall rockbox everytime it seems
23:33:04AlexPThat is not normal, of course
23:33:18AlexPAnd when did you have to reinstall?
23:34:15stripwaxwebguest21 - you can just copy the song onto your ipod .. why do you think you need to reinstall rockbox just to update your songs?
23:34:37webguest21i just had installed rockbox decided i didn't like the songs so i pluged it in and when i unplugged it was back to the ipod firmware
23:34:53AlexPRockbox can dual boot
23:34:55stripwaxwebguest21 - you can reboot (hold menu+select)
23:35:00Hillshumreboot and you will be back to Rockbox
23:35:13stripwaxperhaps your Hold switch was on: that will cause it to reboot to Apple operating system instead of Rockbox
23:35:30stripwaxor perhaps your rockbox installation is corrupted (read the screen carefully when it boots - it should say if this is the case)
23:35:45 Part wincent ("Kopete 0.12.7 : http://kopete.kde.org")
23:37:04webguest21ahh ok the hoold was on
23:37:21pixelmaJdGordon|: on my c200 I get the following with cuesheet support: iCatcher which has no viewports works flawlessly, cabbiev2 which has viewports and draws progressbar and some images into the main screen has the update problem (disabling/commenting out the progress bar helps, stuffing things neatly away into own vps doesn't), my WPS crashes Rockbox (I guess it's because it uses 2 progressbars, conditionally)
23:38:01pixelmait crashed with an older build too though
23:38:01 Quit Hillshum ("Leaving")
23:39:08JdGordon|ok, ill look into it toinght... can you put all that in the FS entry?
23:39:53pixelmayes, I'll paste from here I guess
23:40:07 Quit dfkt ("-= SysReset 2.53=- Ph'nglui mglw'nafh Cthulhu R'lyeh wgah'nagl fhtagn.")
23:42:00pixelmaat one time I saw bogus info in the "next track" info with cabbiev2 after skipping to the next or previous subsong of the cuesheet but couldn't reproduce
23:44:28 Join safetydan [0] (n=deverton@rockbox/developer/safetydan)
23:47:45 Join r0b- [0] (n=rob@adsl-76-235-183-168.dsl.klmzmi.sbcglobal.net)
23:48:45stripwaxwebguest21 - may I take this opportunity to suggest you read the *EXCELLENT* manual where this and many other questions you may have will be answered :-)
23:50:20 Quit jgarvey ("Leaving")
23:52:32*webguest21 hangs head in shame
23:52:59*linuxstb has some noise coming out of his nano2g
23:53:16JdGordon|good noise or bad noise?
23:53:38linuxstbbad noise, but intentionally bad... i.e. audio playback, but just playing back some noise
23:53:40webguest21i just figured i'ed try doing things on my own first since i've moded so many consoles and such over the years and when i was working on this it went so smoothly it totally caught me off guard when it didn't cooperate with me
23:54:20 Quit Strife89 ("Gone to dinner.")
23:55:09JdGordon|linuxstb: woo!
23:56:53stripwax'we have sound'? or doesn't that qualify?
23:56:57linuxstbThe problem is that the i2c interrupt is working fine, but I can't the timer (and now DMA) interrupts to work...
23:57:14linuxstbstripwax: "sound" is stretching it... I got the piezo buzzing a few days ago as well...
23:57:39stripwaxwebguest21 - ok .. but be warned, if you ask a question that is covered in detail the manual we might just point you at the manual again ...
23:57:42linuxstbIt's basically bertrik's patch for audio on the Meizu, attached to the existing WM8975 driver
23:58:48 Quit fyrestorm (Read error: 104 (Connection reset by peer))
23:58:57Zagorgah. best-fitting the builds to the clients is easy. handling the anomalities of ever-changing client speeds is a lot more tricky.

Previous day | Next day