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 2008-11-23

00:03:04 Join robin0800 [0] (
00:08:46gevaertsZagor: if I make a symlink to the source dir in /tmp (e.g. /tmp/rb), and build in there, it works
00:10:15gevaertsAnd if I put the source deep enough on my etch system, it fails
00:10:25Zagorwhat's your path?
00:10:30Zagorthat breaks
00:10:42gevaertsNo -s or .s involved
00:11:01gevaertsBuild dir was /home/fg/rockbox/testtesttesttesttest/trunk/build
00:12:21Zagor /home/bjst/src/a-very-long-dirname-to-provoke-strange-errors/rockbox/build is not enough to trigger it here :-(
00:13:05 Quit robin0800 (Read error: 104 (Connection reset by peer))
00:14:01gevaertsr19184 seems to have fixed this as well
00:14:25Zagorahh, of course.
00:14:35Zagorsilly me
00:15:45gevaertsAlways nice to find that everything is caused by this single bug :)
00:16:08Zagorworld peace will now ensue
00:17:56*gevaerts feels silly for not having spotted the lack of absolute paths
00:19:48 Quit ender` (" Do not believe any statistic you didn't falsify yourself.")
00:21:23 Quit HBK- ()
00:23:46*amiconn wonders what Zagor did to the build master...
00:24:54Zagormy bad breath, probably
00:27:14Zagorodd, I still get new jobs sent to my home machine
00:33:22 Quit {phoenix} (Remote closed the connection)
00:45:00 Quit _lifeless (Remote closed the connection)
00:46:07 Quit Zagor ("Client exiting")
00:47:53 Quit jhulst (Read error: 113 (No route to host))
00:48:56 Quit HellDragon (Client Quit)
00:50:08 Join HellDragon [0] (
00:59:40kugeljhMikeS: still not here? I was playing around with reducing the messages the last few hours. The problem is, that backlight_on() will cause a fade up message always when the backlight_timer is reset. IMHO not a real problem, since that's how it always worked. But if I try to reduce the message anyway, I'm getting problems with the always-off and always-on backlight settings (especially in...
00:59:42kugel...the moment they're set)
01:00:14 Quit bluebrother ("leaving")
01:00:14 Join blkhawk- [0] (
01:00:58 Quit blkhawk (Read error: 113 (No route to host))
01:01:06 Nick blkhawk- is now known as blkhawk (
01:03:38kugeljhMikeS: but with your queue_wait_w_tmo I also get problems with that two options. I'd just go with the many messages (it's not really much, once per button pressed) since that just works
01:08:49 Quit Lynx_ (" HydraIRC -> <- Leading Edge IRC")
01:09:38 Quit kugel ("ChatZilla 0.9.84 [Firefox 3.0.4/2008111318]")
01:23:53 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
01:26:45 Quit Thundercloud (Remote closed the connection)
01:30:08 Join Thundercloud [0] (
01:34:17 Quit culture (Success)
01:35:14 Quit JdGordon (Remote closed the connection)
01:35:39 Quit bertrik ("Leaving")
01:41:18 Join Schmogel [0] (
01:42:59 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
01:50:04***Saving seen data "./dancer.seen"
01:54:38 Quit lasser (Read error: 110 (Connection timed out))
01:59:31 Join Am0nXz [0] (n=normanz@
02:07:59 Quit Strife89 (Read error: 104 (Connection reset by peer))
02:07:59 Join Strife1989 [0] (n=michael@
02:09:28 Nick Strife1989 is now known as Strife89 (n=michael@
02:13:50 Part Am0nXz
02:38:42 Quit Schmogel (Read error: 104 (Connection reset by peer))
02:46:59 Quit nuonguy ("This computer has gone to sleep")
02:49:03 Quit fdinel ("Miranda IM! Smaller, Faster, Easier.")
03:02:25 Quit Nico_P (Remote closed the connection)
03:02:43 Quit herrwaldo ("Konversation terminated!")
03:03:11 Quit DerDome (Remote closed the connection)
03:04:04 Quit faemir ("Leaving")
03:06:50 Quit Thundercloud (Remote closed the connection)
03:10:13 Quit Bensawsome ("The awsome is gone :(")
03:10:51 Join Bensawsome [0] (n=Bensawso@unaffiliated/bensawsome)
03:11:32Unhelpfuli can't seem to get logf to work... i just need to #define LOGF_ENABLE before including logf.h, and make an advanced->logf build, right? but nothing ever seems to show up in system->debug->logf...
03:15:52Unhelpfulhrm. it must not have been making it to my logf statements, for whatever reason. now it just causes a data abort instead of failing silently :/
03:16:41betaWell failing with x message is better then failing with no message :)
03:17:56Unhelpfulbeta: not really, since i now have a data abort and *not* the log message my debug code is trying to give me
03:26:57 Quit arohtar (Read error: 110 (Connection timed out))
03:28:42 Join nuonguy [0] (
03:29:07 Join Darksair [0] (n=user@
03:37:15 Join whj [0] (
03:48:24 Quit mc2739 ()
03:50:05***Saving seen data "./dancer.seen"
03:55:39 Quit Strife89 ("Goodnight, all!")
03:59:23 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
04:02:02 Join miepchen^schlaf_ [0] (
04:03:31 Quit Rob2222 ()
04:16:40 Join blkhawk- [0] (
04:17:38 Quit miepchen^schlaf (Connection timed out)
04:18:34 Join Rob2222 [0] (
04:23:42 Quit tyfoo ("Carpe diem")
04:26:14 Join Rob2223 [0] (
04:30:42 Join HBK [0] (
04:32:15 Quit MethoS-- (Remote closed the connection)
04:32:58 Quit blkhawk (Read error: 110 (Connection timed out))
04:33:18 Nick blkhawk- is now known as blkhawk (
04:37:22 Quit Rob2222 (Read error: 110 (Connection timed out))
04:50:10 Join DMJC [0] (n=DMJC@
04:50:43DMJCI used to own an iriver ihp-140 before I dropped it and it stopped working
04:50:49DMJCnow I own a 160gb ipod classic
04:51:08DMJCwhat I want to know is, how much is known about the iriver ihp-140 design?
04:51:32DMJCwould it be possible to create from scratch an iriver mp3 player?
04:55:56LloreanIt wouldn't exactly be an iRiver MP3 player if you were the one making it.
04:56:26LloreanIt'd be possible to create from scratch a relatively similar player, I'm sure. It'd be rather expensive to try to make one identical, I imagine.
05:03:31 Join BHSPitLappy [0] (n=BHSPitLa@unaffiliated/bhspitmonkey)
05:09:04 Join miepchen^schlaf [0] (
05:16:31 Quit miepchen^schlaf_ (Read error: 110 (Connection timed out))
05:17:06 Nick Darksair is now known as Awaysair (n=user@
05:17:54 Quit DMJC (Connection timed out)
05:19:42 Part toffe82
05:29:11UnhelpfulLlorean: what exactly should i be measuring in terms of binsize? i'm actually getting to the point where it might be worth checking.
05:29:25LloreanI think rockboxinfo.txt reports binsize and ram usage
05:29:28*Llorean isn't sure
05:32:59Unhelpfulok. there needs to be *at least* a 3*LCD_WIDTH-byte static buffer for reading in unscaled bitmaps. there are also 36*LCD_WIDTH bytes of static buffer used by the scaler.
05:33:54Unhelpfulthe first pretty much has to stay static, so that backdrops, icons etc can be read in contexts where there's not a convenient plugin/audio bufer to use
05:34:32Unhelpful3 rows of uint32
05:34:44LloreanAnd LCD_WIDTH is used for what reason?
05:36:17Unhelpfulthe scaler can take partial lines as input, but can't output partial lines because more than one input line contributes to each output line. the buffers have to be the width of a scaler output line, and i can't imagine any good reason to scale an image to a size that is larger than the display ;)
05:40:21 Quit nuonguy ("This computer has gone to sleep")
05:42:56Unhelpfuli've noticed one gotcha so far, which is that album art on the buffer stays whatever size it was loaded at if the WPS changes. i'm not really sure that this is any different from the vanilla svn case with different cover files for different sizes?
05:43:21LloreanI think that's what the vanilla SVN does.
05:43:44LloreanI'm not sure it's a considerable problem as long as it's cropped to the display size (or centered if too small) and we document the fact that it won't be re-resized.
05:43:56Unhelpfulin any case, if the new WPS uses AA at a different size, the album art is displayed cropped or centered.
05:45:16Unhelpfulthe fixes i can think of are all not too wonderful - the most obvious is to always take enough room for the maximum reasonable AA (i'd assume a square at MIN(LCD_WIDTH,LCD_HEIGHT)), store the filename, and reload if wps changes
05:46:54LloreanI think we're okay just saying "Album Art is resized when the music it goes with is buffered. This means unless you stop and resume playback, if you change WPSes Album Art may be the wrong size for as much as thirty minutes or more, until it needs to read new songs from the disk."
05:49:47Unhelpfulwe could also store bitmaps at the end of the current audio buffer, or somewhere else not between tracks, and move or reload them as needed, but i'm not sure how big a hit the extra memcpy work might be.
05:50:07***Saving seen data "./dancer.seen"
05:51:10JdGordonit wouldnt be very commong to change wps whiel AA is loaded so it shouldnt be a big deal
05:51:33JdGordonand if the new size is smaller couldnt it be resized again anyway?
05:51:37UnhelpfulJdGordon: my only concern about that would be WPS modes, if that ever goes in.
05:52:12JdGordona 2nd img buffer would need to be used if each mode allowed different sized AA
05:52:19JdGordonwhich I doubt would ever be allowed
05:52:52Unhelpfuli had also thought of scaling to max size on load, and scaling to display size on draw, but that seems unlikely to fly
05:55:39 Nick Awaysair is now known as Darksair (n=user@
06:03:52Unhelpfulhrm, right now, memory and binsize are a bit up vs the previous patch. probably can be made a little more sane by converting all the other users for read_one_line to use read_part_line
06:05:18 Quit agaffney (Read error: 104 (Connection reset by peer))
06:05:28 Join agaffney [0] (n=agaffney@gentoo/developer/agaffney)
06:05:45Unhelpfulmy numbers are (bin/ram): 522124/1632116 for vanilla, 527224/1647252 for the "classic" resize-on-load, 528380/1650868 for mine
06:07:07Unhelpfulthose are on gigabeat S
06:07:21 Quit JdGordon (Remote closed the connection)
06:08:19Unhelpfulhrm, how to make GCC not warn on a value that *i* know will be initialized on the first pass through the loop?
06:12:38 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
06:29:25 Join massiveH [0] (
06:30:32 Quit aarcane ("Leaving")
06:33:49 Quit saratoga ("CGI:IRC (EOF)")
07:12:50 Quit Llorean (Read error: 104 (Connection reset by peer))
07:13:23 Join Llorean [0] (
07:14:46 Join jchiller1p [0] (
07:14:54 Quit jchillerup (Remote closed the connection)
07:24:27 Quit Bensawsome ("The awsome is gone :(")
07:27:24 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
07:28:14saratogagevaerts: [for the logs] you mentioned a while ago that putting buffers in IRAM prevented the SD card corruption bug, is that still believed to be the case?
07:29:44 Quit jchiller1p (Remote closed the connection)
07:29:53 Join jchillerup [0] (
07:30:44 Quit Darksair ("(define zero (lambda (f) (lambda (x) x)))")
07:31:58Lloreansaratoga: I *think* further testing proved that not to be true
07:43:32 Quit saratoga ("CGI:IRC (EOF)")
07:49:22 Quit massiveH ("Leaving")
07:50:08***Saving seen data "./dancer.seen"
07:54:45 Join Darksair [0] (n=user@
07:56:18amiconnUnhelpful: Afaik you can't. Simply initialize it to 0 in the definition to stop gcc whining.
07:57:17amiconnUnhelpful: Regarding re-resizing (on the fly): This would be good for a real headache. Did you think about 2bpp targets?
08:10:00 Quit Darksair ("Emacs = ESC-Meta-Alt-Ctrl-Shift")
08:15:05 Join Darksair [0] (n=user@
08:16:55 Part XavierGr
08:21:20 Join Rob2222 [0] (
08:25:38J-23RButil doesn't compile for me, make returns "Project ERROR: Qt >= 4.3 required!"
08:27:46LloreanAnd what version of Qt do you have?
08:28:46 Quit Darksair ("(define zero (lambda (f) (lambda (x) x)))")
08:29:37J-23more exactly 4.4.3
08:29:45 Join Acksaw [0] (
08:32:16 Join aarcane [0] (
08:33:15 Join Darksair [0] (n=user@
08:39:25 Quit Rob2223 (Read error: 110 (Connection timed out))
08:41:19 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
08:42:26 Join nuonguy [0] (
08:57:30 Quit Darksair ("Everything that has a beginning has an end.")
09:00:01 Quit BHSPitLappy (Remote closed the connection)
09:01:51 Quit aarcane (Remote closed the connection)
09:03:44 Join Darksair [0] (n=user@
09:12:02 Quit Darksair ("(define zero (lambda (f) (lambda (x) x)))")
09:28:30 Quit JdGordon (Remote closed the connection)
09:29:52 Join skipper [0] (
09:29:56 Join JdGordon [0] (
09:35:07 Quit bmbl (Read error: 104 (Connection reset by peer))
09:36:36 Join mib_8omgh4 [0] (i=18b02d06@gateway/web/ajax/
09:37:04 Quit mib_8omgh4 (Client Quit)
09:39:06 Join DMJC [0] (n=DMJC@
09:47:21 Quit jhulst (Read error: 60 (Operation timed out))
09:50:12***Saving seen data "./dancer.seen"
09:51:37 Join n1s [0] (n=nils@rockbox/developer/n1s)
09:57:13 Join lasser [0] (
10:00:43 Join Darksair [0] (n=user@
10:02:57 Join nplus [0] (n=nplus@141.25.Globcom.Net)
10:05:46 Join bertrik [0] (
10:12:04 Join jeffdameth [0] (
10:15:54 Quit jeffdameth1 (Read error: 60 (Operation timed out))
10:24:23 Quit DMJC (Connection timed out)
10:25:47 Quit Kopfgeldjaeger (Read error: 104 (Connection reset by peer))
10:26:24 Join Kopfgeldjaeger [0] (
10:33:57 Quit Kopfgeldjaeger (Read error: 60 (Operation timed out))
10:37:09 Quit JdGordon (Read error: 104 (Connection reset by peer))
10:39:01 Join JdGordon [0] (
10:39:28 Quit Bagder (Read error: 60 (Operation timed out))
10:40:07 Join {phoenix} [0] (
10:43:32 Join Lynx_ [0] (
10:57:22bertrikI'm confused by the adc-target.h file for the various ams sansa targets
10:57:40 Quit n1s ()
10:57:57bertrikthere's already an adc-target.h in firmware/target/arm/as3525
10:58:50bertrikthe bootloader and the main rockbox seem to use different adc-target.h files
11:04:46 Join [rmd] [0] (
11:05:07 Join cg [0] (
11:06:24 Part [rmd]
11:07:31 Quit nuonguy ("This computer has gone to sleep")
11:08:26 Join nuonguy [0] (
11:10:10 Quit nuonguy (Client Quit)
11:11:43 Nick Darksair is now known as Awaysair (n=user@
11:17:28bertrikthe clip seems to suffer from the same problem with battery readout with charger connected as other sandisk targets with an ams codec: battery voltage is way too high with the charger connected
11:19:24 Quit pixelma2 ("-")
11:19:46 Join Horschti [0] (
11:20:32 Quit Horscht (Nick collision from services.)
11:21:02 Join pixelma [0] (n=Marianne@rockbox/staff/pixelma)
11:21:06 Join Lear [0] (i=chatzill@rockbox/developer/lear)
11:22:05bertrikall ams sansas use a lithium battery, except for the m200, right?
11:23:14 Join herrwaldo [0] (
11:27:40*amiconn still suspects that we're using the wrong adc channel or sth like this
11:28:03 Join Bagder [241] (n=daniel@rockbox/developer/bagder)
11:28:59 Join stoffel_ [0] (
11:29:24 Quit amiconn (" bbl")
11:29:50 Join ender` [0] (
11:30:35 Join Trista109 [0] (
11:32:13 Join kugel [0] (n=chatzill@unaffiliated/kugel)
11:47:56kugelUnhelpful: so your solution uses more binsize and ram?
11:48:08 Quit JdGordon (Read error: 54 (Connection reset by peer))
11:49:02 Nick Awaysair is now known as Darksair (n=user@
11:49:54 Quit {phoenix} (Read error: 110 (Connection timed out))
11:50:13***Saving seen data "./dancer.seen"
11:52:22 Quit pixelma (Read error: 110 (Connection timed out))
11:55:04 Join n1s [0] (n=nils@rockbox/developer/n1s)
12:13:15bertrikthe voltage I'm getting now seems reasonable: 3.8V and slowly dropping, OF indicates about 50% charge
12:18:16 Join culture [0] (
12:24:13 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
12:24:53kugelbertrik: cool
12:24:59 Join {phoenix} [0] (
12:29:39 Join Kopfgeldjaeger [0] (
12:33:14bertrikrasher, Clip builds if you disable plugins in the Makefile after running ../tools/configure
12:33:30 Join MethoS- [0] (
12:34:41 Quit Kopfgeldjaeger ("Serverwechsel")
12:35:00 Join Kopfgeldjaeger [0] (
12:35:04bertrikwow, it seems a mutex_init is missing for as3525 codec, yet it seems to work
12:36:00 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
12:38:50 Join Thundercloud [0] (
12:40:54rasherbertrik: Okay. Well I'm not going to do divert from SVN for my simulator builds. When SVN builds, it'll be built again.
12:45:00 Join amiconn [50] (n=jens@rockbox/developer/amiconn)
12:46:11 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
12:49:29 Join petur [50] (n=petur@rockbox/developer/petur)
12:52:24 Quit stoffel_ (Read error: 113 (No route to host))
12:52:58J-23bertrik: why not make bin?
13:01:00bertrikI don't know, I'd rather just type make && make gzip
13:03:24 Quit Lynx_ (" HydraIRC -> <- The dawn of a new IRC era")
13:04:10 Join stoffel_ [0] (
13:05:40linuxstbbertrik: Why not just set plugins="no" in tools/configure for the Clip, until they are all building?
13:05:51 Join Lynx_ [0] (
13:06:46JdGordonit would be nice if the bin could be linked or whatever when its dependancies are built like it used to... it shouldnt need to wait for the codec/plugins to build first
13:07:06Unhelpfulkugel: i'm kind of surprised at the more ram, given that it gets rid of a multi-line input buffer. it's also not finished yet, it's a work in progress.
13:07:44bertriklinuxstb, sounds like a good idea
13:08:40bertrikso at least the default will build, anyone experimenting with plugins can turn them back on locally
13:09:06kugelUnhelpful: I'm surprised as well. Given that you start your work with the target to be simpler and more restricted o
13:09:15kugelin order to save binsize and ram usage
13:11:06linuxstbbertrik: Seems funman changed plugins to "yes" in this commit, but he didn't say why -;revision=19045
13:11:24rasherLooks like a mistake to me
13:11:52Unhelpfulpart of the WIP portion will involve stripping old code once it's no longer used - there's a new read_part_line that's almost entirely the same as read_one_line, and read_one_line can go away when things are converted to use it.
13:12:42Unhelpfulthe mono-for-remote scaling should probably just go away entirely. scaling mono bitmaps is pretty much guaranteed to give poor results.
13:14:15bertrikshall I reverse that change in tool/configure?
13:19:41kugelbertrik: seems reasonable. other ams sansas don't build plugins
13:20:59kugelUnhelpful: well, that could the bmp resize on load patch on the tracker do as well I suppose
13:21:43Unhelpfulyes, the patch on the tracker could do that part as well.
13:22:34kugelwhich would mean it'd be also less binsize&ram using
13:23:14 Quit Rob2222 ()
13:23:56kugeljhMikeS: ping
13:25:40 Quit stoffel_ ("Reconnecting")
13:25:50 Join stoffel_ [0] (
13:25:56 Join Rob2222 [0] (
13:34:35 Quit Thundercloud (Remote closed the connection)
13:35:53 Join faemir [0] (
13:44:47 Quit Horschti ("electromagnetic radiation from satellite debris")
13:47:27 Join pixelma [0] (n=Marianne@rockbox/staff/pixelma)
13:50:16***Saving seen data "./dancer.seen"
13:50:21 Quit JdGordon (Remote closed the connection)
13:50:42 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
14:00:27 Join Photoguy [0] (n=Adams@
14:02:07Unhelpfulpixelma, kugel: the scaler and bitmap buffers both limit the width of the bitmap, not the height at all. the scaler buffer could be reduced to the min dimension if we assume square output, or increased to the max dimension if we want rotated output.
14:02:15 Nick fxb__ is now known as fxb (
14:03:50Unhelpfulif we support loading a column-first image format, we'd need to rotate it for display.
14:05:24gevaertsAre there such formats? (except jpeg with the exif rotate bit)
14:07:45pixelmaso basing the buffer size on LCD_WIDTH is also due to how the scaler works?
14:13:53 Join fdinel [0] (
14:14:43pixelmanot sure, I can follow completely at the moment (so maybe the question doesn't make much sense). Just imagined when reading logs that lcd width was not the only important factor but also the height, depending on screen layout (portrait vs. landscape vs. square)
14:27:02 Join Thundercloud [0] (
14:29:29 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
14:29:31Photoguy*Thinks it looks like rain.
14:30:42Unhelpfulpixelma: i suppose it is based on how this scaler works, in a way. it scales one row of input image to one row of output image - so the buffer size is based on the width of the output.
14:31:06Unhelpfulgevaerts: are JPEG macroblocks stored row-first? i've looked a bit without finding anything that will tell me :/
14:31:20gevaertsUnhelpful: no idea
14:38:03 Nick J-23 is now known as J-23_ (
14:38:08 Quit skipper (Read error: 110 (Connection timed out))
14:38:10 Nick J-23_ is now known as J-23 (
14:41:10 Join skipper [0] (
14:42:11 Join tyfoo [0] (
14:45:55 Join hangin [0] (n=543b3462@gateway/web/cgi-irc/
14:46:49hangini made my own theme (for the nano) but i took some pics of the cabbie 2 theme (the icons and i modified the cabbie 2 background). my question is am i allowed to use them? and to upload my theme?
14:48:20 Quit hangin (Client Quit)
14:49:10 Join hangin [0] (n=543b3462@gateway/web/cgi-irc/
14:50:32JdGordoncabbiev2 should be fine to copy and distrobute (as long as you play by the rules)
14:51:13 Quit hangin (Client Quit)
14:51:19JdGordonn1s: what ever happened with the bootloaders for 3.0?
14:51:56 Join josh__ [0] (
14:52:58josh__sry i just was disconnected....
14:53:14josh__still the same question
14:53:16josh__i made my own theme (for the nano) but i took some pics of the cabbie 2 theme (the icons and i modified the cabbie 2 background). my question is am i allowed to use them? and to
14:54:26kugeljosh__: sure, as long as you credit the original authors
14:54:38josh__how do i credit?
14:54:46josh__the original authors
14:55:14kugela comment in the theme files and/or in the description saying "based on xx by yy" or something like that
14:55:51josh__k thx
14:56:11 Join uurguu [0] (
14:56:19josh__and what is about the licensing i read about it in the forum but didn't understood it...
14:56:38uurguuhi everyone
14:56:58kugelyou can only chose a license which the licenses of the themes you taken parts from allow you to
14:57:24uurguucould anyone help me with an install problem?
14:57:46josh__ok.... so i just credit the auot in the wps file and then i can upload my them ?
14:58:02josh__whats your porblem with installin?
14:58:05kugelbut in any event, you should use CC-BY-SA, since that the preferred license if you want to upload it on a site affiliated to rockbox
14:58:41josh__i hope i understood everything correct :) thx for help :)
14:58:44uurguuanyone? :-(
14:59:00josh__hi whats your problem?
14:59:26uurguui am trying to install rockbox 3.0.1 on an Archos Recorder 20
14:59:51uurguubut the stupid archos firmware is qstill there
15:00:07uurguuI have tried installing through the utility
15:00:08josh__to kugel thx (do you speak german?)
15:00:20uurguui have also tried installing manually
15:00:39uurguuwiping out the FAT table with dirsnooper
15:00:53uurguunothing works
15:01:18uurguuand the hard disk shows 3GB free out of 20GB
15:01:33bertrikthere is now a single powermgmt-xxx.c file for all as3525 targets containing the battery discharge/charge curves, I think there should probably be one for each target
15:01:35josh__i think your player isnt supported...
15:01:39uurguuwith only 200MBof MP3 on the disk
15:01:59josh__to kugel darf man hier deutsch sprechen oder ist das net erlaubt
15:02:08kugelnot allowed
15:02:13kugelplease read the guidelines
15:02:14uurguumy player shows up fine in the rockbox utility ?
15:02:33 Join mc2739 [0] (
15:02:48josh__to kugel do you have icq skype or msn or another forum etc where we can speak german?
15:03:07uurguuand rockbox loaded fine at least once
15:03:10kugelbertrik: why? Shouldn't powermanagement the same for all targets (maybe with #ifs depending on the actual battery capacity?
15:03:15 Quit Lear ("ChatZilla 0.9.84 [Firefox 3.1b2pre/20081121034512]")
15:03:19josh__to uurguu i cant find your player here
15:03:25kugeljosh__: you can /query me
15:03:57josh__to kugel what the heck is /query?
15:04:05kugeljosh__: is a german forum
15:04:19uurguui think the rockbox util is trying to install a recorder 8MB firmware
15:04:25kugeltype "/query kugel" tp provat chat with me
15:04:30uurguuat least, that's how it shows once downloaded
15:04:31kugelto privat*
15:04:35gevaertsuurguu: you could try a manual install
15:04:56JdGordoncan someone test mpegplayer quickly? if audio is paused before mpegpalyer is started does sound work?
15:05:03gevaertsOr maybe not having rbutil autodetect your player but select manually
15:05:06 Join MethoS-- [0] (
15:05:46uurguugevaerts : tried manual install
15:06:02uurguugevaerts : selected the player manually in rbutil
15:06:03 Quit stoffel_ (Read error: 113 (No route to host))
15:06:20n1sJdGordon: ne bootloaders for ipods and gigabeat f were released, there were inconclusive reports about the sansa bootloader (something about usb detection not working) and i couldn't test myself so not much to do about them, i cant test any of the other players either so it basically needs someone to test them but i tire of trying to get people to test...
15:06:22 Quit {phoenix} (Connection timed out)
15:07:17JdGordonok, so the mini should be able to boot from CF with the latest bootloader then?
15:07:51n1sit fixed _one_ of the cf issues at least, but there seems to be other problems
15:08:00uurguu <−− shows archos recorder 20 as supported
15:08:13n1suurguu: it is
15:08:35gevaertsuurguu: do you have a file called ajbrec.ajz in the root of your player, and a directory named /.rockbox ?
15:08:58bertrikkugel, likely not all ams targets use the same battery, so cannot use the same battery discharge/charge curves
15:09:01JdGordonn1s: I'm looking at 8901.. can it be closed?
15:09:02 Join stoffel_ [0] (
15:09:28uurguugevaerts: yes and yes (both file & dir are present)
15:09:28n1sJdGordon: so FS #8901 is fixed for ipods and gigabeat, not the others
15:09:51n1sbut ipod mini is by far the most common target to cf mod so i dunno
15:10:06pixelmauurguu: yours is not a "Jukebox Recorder" with a colour display though? (Just making sure)
15:10:32uurguupixelma: nope, LCD B&W display
15:10:42pixelmaeh... Multimedia, I meant
15:10:52uurguuand please note rockbox did boot up once
15:11:22uurguuand then failed with an "incompatible device" message
15:11:36uurguucould not start MP3 playing
15:11:56bertrikkugel, the source file is basically just those curves, a dangerous and shutoff battery level and a battery_adc_voltage function
15:12:08gevaertsuurguu: the only thing I can think of is that you have the wrong build. There are v1 and v2 recorders, did you pick the right one?
15:12:37uurguuno idea, i think it was recorder 8mb
15:12:47 Quit Horscht ("User was distributing pornography on server; system seized by FBI")
15:12:55uurguuagain, i let the rockbox util chose the version
15:13:22pixelmayou probably don't have an 8MB Recorder, this is only possible with a hardware mod
15:13:42uurguuthe v2 recorder does not lokk like mine
15:14:13gevaertsok. then you have a v1, but as pixelma said most probably not an 8MB
15:14:38uurguushould i take recorder v1?
15:15:02uurguuOK, i'll try a manual install then
15:15:12 Quit stoffel_ ("leaving")
15:16:20n1sJdGordon: the other cf problem is FS #9342 which looks pp specific judging by the "fix"
15:16:20 Quit skipper (Read error: 54 (Connection reset by peer))
15:17:03uurguucopied on device
15:17:46 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
15:18:22pixelmaI'm not sure if the Rockbox Utility can tell all of the Archos targets apart but if it detected a Recorder v1, it *should* install the "normal" build - not the 8MB one
15:18:33JdGordonn1s: hmm.. not a great patch thouhg
15:20:06JdGordongevaerts: close 9441 as out of date/fixed?
15:20:55JdGordonhmm reread and ignore me
15:20:59*JdGordon goes to bed
15:21:03gevaertspixelma: it may install the 8MB one if the previous (incorrect) install was an 8MB install
15:21:10 Nick JdGordon is now known as JdGordon|zzz (n=jonno@rockbox/developer/JdGordon)
15:21:27bertrikkugel, what I'll probably do right now is to use the same simple uncalibrated charging/discharging curve for all ams targets so they can all get a basic battery reading, and add a comment to the source file on how to add calibrated curves as they become available
15:21:33uurguugreat! works perfectly!
15:22:16 Quit MethoS- (Read error: 113 (No route to host))
15:22:23uurguustrange thing the utility installed an 8mb version instead of the correct one
15:22:42 Join Yggdrazil [0] (n=vegan@
15:23:22kugelbertrik: ok. I know the fuze has a li-polymer battery. I'm not sure about the capacity though
15:23:27uurguuthanks a lot everyone, this was driving me crazy
15:23:28Yggdrazili can use rockbox for my ipod ?
15:23:34n1suurguu: it can not tell a 8MB modded player from a regular so it goes by what build you had already installed
15:23:48uurguuYggdrazil: yes, check the models available
15:23:54n1sYggdrazil: is it listed as supported on the frontpage?
15:23:59 Quit josh__ ("ChatZilla 0.9.83 [Firefox 3.0.4/2008102920]")
15:24:28Yggdraziluhm, ive downloaded it, and run the quickstart complete installation
15:24:36Yggdrazilit was a very primitive installation, i didnt understand much
15:24:51Yggdrazillike, where is the program itself ?
15:24:53pixelmadepends on the exact Ipod model (no "newer" ones, 2nd gen Nano or later, new Ipod "Classic" etc.)
15:25:04Yggdrazili have a 30 gb video ipod
15:25:21uurguuprobably won't work?
15:25:23n1sYggdrazil: rockbox.ipod is the main binary
15:25:35n1sipod video is supported
15:25:50Yggdrazilyikes it has taken over the ipod
15:26:00uurguun1s: i stand corrected :-)
15:26:06n1sYggdrazil: i suggest oyu read the manual
15:26:14Yggdrazilwindows cant open this file
15:26:54 Quit Photoguy (Read error: 104 (Connection reset by peer))
15:27:23n1sYggdrazil: why do you want to open it?
15:28:30peturYggdrazil: rockbox is a replacement operating system for your player, not a simple program
15:28:47Yggdrazilokay, it doesnt go away from the ipod when i press uninstall either
15:29:09Yggdrazilhow do i get back to itunes ?
15:29:58n1sYggdrazil: check the manual, and please try to use correct terms, itunes is an app on your computer
15:30:38Yggdrazili used itunes, then i tried this rockbox thing, i didnt understand anything of the rockbox, how do i remove the program from my ipod and go back to whatever the ipod comes with ?
15:31:18n1sYggdrazil: maybe by looking at the uninstallation instructions in the manual?
15:31:28uurguuall right chaps time to transfer GB of MP3s onto.
15:31:33uurguu my archos
15:31:51uurguuthanks again for the help!
15:31:55 Quit uurguu ()
15:37:55Yggdrazilit says uninstallation finished in the rockbox utility program
15:38:17Yggdrazilbut the rockbox is still on my ipod
15:40:03 Quit MethoS-- (Remote closed the connection)
15:40:54Yggdrazili cant find where the uninstallation is in the manual either
15:41:17gevaertsYggdrazil: did you also click on "Remove the bootloader" ?
15:42:36Yggdrazilyes but that doesnt find my ipod
15:44:10gevaertsOK. Boot the ipod to the emergency disk mode then, and try again
15:44:29Yggdrazilhow do i boot it into emergency disk mode ?
15:45:53 Join MethoS- [0] (
15:46:12gevaerts explains all about it
15:50:19***Saving seen data "./dancer.seen"
15:50:54Yggdrazildo you use an ipod yourself ?
15:51:13gevaertsAmongst others, yes
15:51:43Yggdrazildo you use rockbox on your ipod ?
15:51:58Yggdrazili use windows xp and a 30 gb video ipod
15:52:09Yggdrazili would like to use rockbox to get to play .flac files
15:53:30gevaertsI use rockbox on all my players, yes
15:53:46Yggdrazilallright uhm
15:54:02Yggdrazildo you have time to guide me through the downloading - installing the rockbox - getting flacs over to my ipod
15:54:08Yggdrazilits that and only that i need help with
15:55:14gevaertsIt seems to me that you did have it installed...
15:55:36Yggdrazilit didnt seem to be properly installed
15:55:59Yggdrazilill download it from scratch
15:56:44Yggdrazil this will give me all i need to download for installing the rockbox - getting flacs over to my ipod 30gb Video?
15:57:09 Join Arinohyoshi [0] (
15:57:35gevaertsYou do know that rockbox is a replacement firmware and not just an add-on to the apple firmware, right ?
15:57:58Yggdrazili dont even know what that means
15:58:08Yggdrazilit means it takes over the entire ipod, right ?
15:58:09 Nick Darksair is now known as Awaysair (n=user@
15:58:19bertrikkugel, do you know the marking on the fuze battery?
15:58:31 Join Prom_cat|ZzZzz [0] (n=bretthil@Wikipedia/Prom3th3an)
15:58:34Yggdrazili dont have a problem with that, now that youve shown me how i reset the ipod
15:59:02 Join MethoS-- [0] (
15:59:03gevaertsAnyway, just use rockbox utility to install, it's the easiest way
15:59:35gevaertsAnd read the manual, so you know about how to boot the Apple firmware when you want to and things like that
16:00:10 Join stoffel_ [0] (
16:00:17kugelbertrik: no, sorry. I actually can't remember anything written on the battery when I had the fuze open
16:00:31kugeli.e. iirc there was nothing written
16:01:49kugelthe battery was packed in a silver foil, connected with 3 cables to the main board
16:01:54ArinohyoshiIs Rockbox ever going to support iPod Classic? As I own one I would like to see the support for it. And if I can in any way help to speed up its support, I'm willing to try get some info out of it.
16:01:58Yggdrazilstarting bootloader installation - searching for ipods - no ipod found is all it says when i try the installation with the rockbox utility ?
16:01:59bertrikthis image shows there is some text, but it's too small to read :
16:02:01 Part Prom_cat|ZzZzz
16:02:15 Quit MethoS- (Read error: 60 (Operation timed out))
16:02:38kugelbertrik: you may ask saratoga, afair he has his fuze opened too (or still open)
16:03:09kugelbertrik: or ask enzoten @abi, he made the disassembly
16:04:21bertrikthe manufacturer is probably Amperex Technology Limited, you may be able to find out the capacity from the type number on their web site
16:05:18 Quit faemir (Read error: 60 (Operation timed out))
16:05:33bertrikTBH, I'm not so motivated to go chase after information for a target that I don't personally own
16:06:19 Join faemir [0] (
16:06:22 Join {phoenix} [0] (
16:10:10 Quit stoffel_ ("leaving")
16:10:30 Nick Awaysair is now known as Darksair (n=user@
16:11:42kugelbertrik: I understand. And I'm not really motivated to open my fuze again (I already broke one with that) if others may have the needed information handy
16:12:06Yggdrazili actually need to have the ipod plugged in to install rockbox ?
16:12:11kugelof course, if it's really needed and others can't help either, I'll open it carefully
16:12:33bertrikOk. Sorry to hear about the broken fuze.
16:13:14kugelno problem, it got replaced
16:13:33bertrikActual capacity is not that important, mostly just nice-to-know
16:17:44bertrikI suppose the fuze is like the clip in that it cannot be opened without permanently damaging the casing
16:18:00kugelthats true
16:18:39kugelyou can close it again, but if one has a detailed look, he'll notice that it was cracked open
16:20:11 Quit Zarggg ()
16:20:57 Join Zarggg [0] (
16:23:27ArinohyoshiIf I may ask again, as I didn't find any answer in FAQ or such: Is Rockbox ever going to support iPod Classic? As I own one I would like to see the support for it. And if I can in any way help to speed up its support, I'm willing to try get some info out of it.
16:24:23gevaertsArinohyoshi: Nobody is working seriously on it, and in my opinion the firmware encryption will be a too big obstacle to overcome
16:24:57Yggdrazili actually need to have the ipod plugged in to install rockbox ?
16:25:22gevaertsYggdrazil: obviously, yes. How else can you install?
16:25:37Yggdrazillike i install all other programs on a puter
16:25:45Yggdrazilthe program is only for the ipod ?
16:25:53Arinohyoshigevaerts: thats a shame. 80GB iPod would've made a good external player to have inside a car as it could hold a good amount of .flac's
16:26:08kugelYggdrazil: it's a *firmware replacement*
16:26:16gevaertsArinohyoshi: the 80GB ipod video (5.5gen) is supported
16:26:22Yggdrazilso how do i get the flacs over to the ipod from my computer then ?
16:26:40kugelyou boot into the original firmware to transfer files
16:27:50Arinohyoshigevaerts: you mean these ?
16:28:28gevaertsArinohyoshi: no. That's the 6th generation, which isn't supported
16:29:24kugel7th even
16:29:31ArinohyoshiApple doesn't have the 80GB version listed anymore as that 120GB version has taken over it, but 80GB version of it
16:30:00Arinohyoshiyes, thats the one
16:30:14Arinohyoshinice, then its supported
16:30:16gevaertsThat's _not_ the one apple calls Classic
16:30:51gevaertsThis is the one you could get in 30GB or 80GB
16:32:08Arinohyoshiah. looked the same but its not. i've got the classic one then, as it has album arts on the right side of the screen
16:33:28 Join tessarakt [0] (
16:33:50Arinohyoshiyes, definitely that one, checked the last three digits. YMV
16:34:18 Join stoffel_ [0] (
16:34:44Arinohyoshiso we can conclude that its not supported by Rockbox, and likely is not going to be?
16:41:35gevaertsThat's about it, yes
16:42:19 Join kharo [0] (
16:57:27 Quit Thundercloud (Remote closed the connection)
17:05:23 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
17:13:56 Quit stoffel_ (Read error: 113 (No route to host))
17:23:36 Quit fdinel (Read error: 110 (Connection timed out))
17:25:35 Join FOAD_ [0] (
17:25:47bertrikcan anyone with an ams sansa make sense of the RTC time?
17:26:14bertrikfor me, it seems to get reset to 1 jan 2080 0:00 in some cases, but not always
17:27:46kugelbertrik: don't forget that e200v2 and fuze still have some difficulties to boot into rockbox
17:28:46kugelonly <=1GB devices can reliably boot
17:29:34kugelI can't boot into rockbox with svn code
17:30:48bertrikbecause of sd driver problems?
17:31:07kugelyes, it panics upon sd init
17:31:40kugelnot in the bootloader though, so it's possibly related to multivolume and/or hotswap
17:34:07 Quit Yggdrazil (Read error: 148 (No route to host))
17:35:48 Quit FOAD (Read error: 110 (Connection timed out))
17:35:49 Nick FOAD_ is now known as FOAD (
17:40:42bertrikI might experiment a bit with i2c tonight. On the fuze it's pretty clear which pins are i2c, I also have a suspicion for some i2c pins on the sansa clip
17:42:10kugelisn't i2c working already? I can change the backlight brightness which works over i2c
17:44:07 Quit jhMikeS (Nick collision from services.)
17:44:13 Join jhMikeS [50] (n=jethead7@rockbox/developer/jhMikeS)
17:44:24bertrikthere's another i2c bus which probably controls the radio chip
17:48:53 Join bluebrother [0] (n=dom@rockbox/staff/bluebrother)
17:50:24***Saving seen data "./dancer.seen"
17:53:40kugeljhMikeS: ping (I'll ping you until you answer ;) )
17:55:18 Join stoffel_ [0] (
17:58:04 Quit {phoenix} (Remote closed the connection)
18:01:04 Quit tyfoo ("Carpe diem")
18:04:13 Join Thundercloud [0] (
18:05:56 Join tyfoo [0] (
18:11:50 Join Yggdrazil [0] (n=vegan@
18:23:55 Quit Darksair ("Do you hear that? This is the sound of inevitability. This is the sound of your death, Mr. Anderson.")
18:29:31 Join mathias_ [0] (n=psde@
18:31:17mathias_Can I still load my rockbox player from usb while playing music? Holding 'record' does not work anymore
18:31:50n1smathias_: which player?
18:32:53n1saren't you supposed to hold "AB" on that player?
18:33:00 Join EspeonEefi [0] (i=eefi@SYDNEYPACIFIC-THREE-NINETY.MIT.EDU)
18:33:12mathias_Okay - never done that before.
18:33:51mathias_Eh.. yes. it does.
18:33:58mathias_Weird, I always pressed record
18:35:38mathias_Anyways, its working now. thanks!
18:47:50 Join MedicaMental [0] (n=vegan@
18:49:56 Quit Lynx_ ("Konversation terminated!")
18:51:29 Part mathias_
18:58:40 Quit kharo (Read error: 110 (Connection timed out))
18:59:52 Join kharo [0] (
19:01:14 Join toffe82 [0] (
19:04:22 Quit Yggdrazil (Read error: 113 (No route to host))
19:05:40 Join Yggdrazil [0] (n=vegan@
19:06:46 Quit MedicaMental (Read error: 113 (No route to host))
19:09:11 Join BHSPitLappy [0] (n=BHSPitLa@unaffiliated/bhspitmonkey)
19:15:17 Quit tessarakt (Read error: 110 (Connection timed out))
19:15:54 Join tessarakt [0] (
19:20:05 Join ap0 [0] (
19:29:56 Quit amiconn (" bbl")
19:33:47pixelmabluebrother: I'm still thinking about an M3 manual (and integrating other targets with remotes) and there are some problems I don't have a nice solution for.
19:33:57 Quit tessarakt (Connection timed out)
19:34:17pixelmaE.g., I started off mentioning both buttons in the \ActionBlah definitions in the platform file - this was the most simple way to only use it for the M3 but it will look weird when used in button tables (and it is). The other way would be to invent \ActionRemoteBlah but that would mean \opting for "remote" or m3 as a start everywhere it is used, the former means touching all platform files of affe
19:34:17pixelmacted targets
19:34:24 Join tessarakt [0] (
19:35:01 Quit Yggdrazil (Read error: 113 (No route to host))
19:35:46pixelmathe other problem is, if we want to use an additional column in button tables, it would mean opting an entire part of each line in all the tables
19:36:15pixelmasounds a bit complicated to me...
19:37:29pixelmaI somehow hope there is some feature of LaTeX that I don't know of which could help there
19:43:43 Join Schmogel [0] (
19:45:35 Join nuonguy [0] (
19:46:04pixelmaor maybe it's easier to add a second table, don't know
19:47:01nanokpixelma: is the latex code generated by some sort of script? or is it "pure latex" :) ?
19:50:06pixelmacurrently the code for button tables is not autogenerated. As for as I'm aware there is only one generated thing - parsing the features.txt into usable options
19:50:26***Saving seen data "./dancer.seen"
19:50:29nanoki think aome sort of script, if i understand correctly. either way, the more i think of it, it seems more reasonable (even from a reader point of view) to have a separate section for the remote
19:50:41nanokaham, i see
19:51:51 Quit stoffel_ ("leaving")
19:52:23nanokso it would be nice to first find a way to automate that a bit more too, and than make sure we can destroy (or handle somehow in an ellegant way, like state just "$target does not support a remote with rockbox" where needed, without too much fuss
19:53:06*nanok 's eyes are shinning thinking maybe he can finaly help with something
19:54:36nanokpixelma: autogenerating the buttons table should be simple enough, unless i am missing something: we need standard "names" of actions, specific to rockbox, which apply to rockboz across targets (and i think we already have that?)
19:54:55 Quit faemir (Remote closed the connection)
19:55:50nanoki think there are oddities speciffic to each target, so some stuff needs to be written by hand, though, but the functionality of each button can be standardized across targets, am i right?
19:58:37pixelmacan't see how that should work - for the tables you need to mention long/short presses of buttons or combos - you have action definitions in the code (see keymap files) or button definitions (with or without pre conditions or whatnot) (see some plugins). I don't know what you mean with standardized across targets
19:59:23 Join dany_21a_ [0] (
20:03:13nanokpixelma: i mean the rockbox functionality itself is rather standard across targets, as far as what the buttons do goes, how we use each button to do stuff on various targets, depending on the physical layout, number and type of buttons and so on is what varies, right?
20:07:54 Quit miepchen^schlaf (Client Quit)
20:15:33 Quit perrikwp (" ajax IRC Client")
20:15:45 Join miepchen^schlaf [0] (
20:17:30 Join faemir [0] (
20:17:31 Join Strife89 [0] (n=michael@
20:17:45 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
20:18:59 Quit bmbl ("Woah!")
20:24:18bertrikhmm, can't use sprintf in the bootloader?
20:27:45 Quit ap0 (Read error: 104 (Connection reset by peer))
20:28:39kugelbertrik: use snprintf
20:28:55kugelthat definitely works
20:32:16 Join ZincAlloy [0] (n=d9eee842@gateway/web/cgi-irc/
20:33:21 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
20:36:57 Part pixelma
20:37:12 Join ap0 [0] (
20:37:17 Join pixelma2 [0] (n=marianne@rockbox/staff/pixelma)
20:37:42 Quit ZincAlloy ("CGI:IRC (Ping timeout)")
20:47:04kugelbertrik: while you're at the bootloaders, can you take a look at ?
20:49:06 Quit ap0 ("Baï")
20:50:08 Join AndyIL [0] (i=AndyI@
20:50:11 Quit Horscht (Remote closed the connection)
20:54:40 Quit Schmogel (Read error: 110 (Connection timed out))
20:59:43 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
21:02:43 Quit AndyI (Read error: 110 (Connection timed out))
21:08:11 Quit nuonguy ("This computer has gone to sleep")
21:10:58 Join aarcane [0] (
21:11:57 Quit miepchen^schlaf (Client Quit)
21:13:16 Join DerDome [0] (
21:14:29 Part dany_21a_
21:19:30 Quit bluebrother ("Lost terminal")
21:20:21 Quit Strife89 ("Leaving for now.")
21:21:19 Quit Zarggg ()
21:27:17 Join bluebrother [0] (n=dom@rockbox/staff/bluebrother)
21:29:10 Join Zarggg [0] (
21:29:12 Join funman [0] (
21:30:04 Join krazykit [0] (
21:31:38funmanbertrik: I have an opened clip, do you need some details on the battery ?
21:33:05kugelfunman: hi. did you read my forum post?
21:33:56funmankugel: hi, yes i read it. I use svn, and i have no idea why you can't apply the patch
21:34:12kugelwell I thoug
21:34:26kugelt that it's due to git and different branches
21:35:21kugelfirmware/target/arm/as3525/dma-pl081.c doesn't exist in svn
21:36:36 Join amiconn [50] (n=jens@rockbox/developer/amiconn)
21:38:24funmankugel: ok i pasted the full patch again :
21:38:59kugelthat looks complete :)
21:39:20funmanis the plugins compilation meant to keep going on errors in plugins ?
21:40:02kugelfunman: it's meant to be disabled at such an early stage of a port :S
21:40:24 Quit bmbl ("Woah!")
21:40:26 Join Zagor [242] (
21:40:31kugelsee r19188
21:40:34funmankugel: that's not related to what I said
21:40:51kugelbut still
21:40:56funmanI expect the build to stop on the first error
21:41:07kugelask Zagor about that
21:41:15amiconnZagor: Still dependency problems....
21:41:24 Join miepchen^schlaf [0] (
21:41:30kugelI think before the make remake (pun intended!) it did stop
21:41:57funmanZagor: shouldn't the build stop on the first error ? because it keeps going, even if some plugins fail to build
21:42:07amiconnTry building an (arbitrary) swcodec build. Afterwards, 'touch apps/codecs/demac/libdemac/filter.c', then 'make' again -> it will do nothing...
21:42:56Zagoramiconn: wow, didn't we test that yesterday?
21:43:28amiconnI tried it again today, using r19185
21:44:18amiconnI still need to apply my workaround in order to be sure to test my latest code (rm apps/codecs/ape.* apps/codecs/libdemac.a apps/codecs/demac/libdemac/* && make codecs)
21:44:36 Join MethoS_ [0] (
21:44:37Zagorit rebuilds here...
21:44:39 Join {phoenix} [0] (
21:45:28amiconnI'm doing that on cygwin, in case it matters
21:45:34Unhelpfulodd, it rebuilds for me (gb f sim, in this case)
21:45:35 Quit toffe82 (Read error: 110 (Connection timed out))
21:45:49Zagoramiconn: it might. I'll see if I can reproduce it.
21:46:08Zagorkugel, funman: what's this about not stopping on first error?
21:46:23amiconnDo I need to rebuild the .dep file? I thought the script fix should work with the existing one...
21:46:44 Quit MethoS-- (Read error: 60 (Operation timed out))
21:47:03funmanZagor: make clean, add #error in a built plugin; make; notice the error doesn't stop the build
21:47:04Zagoramiconn: you need to rebuild it after r19184 but not after touch
21:47:17 Join MethoS- [0] (
21:48:09amiconnZagor: Does alter the .dep file itself?
21:48:37Zagoramiconn: yes and no. it alters the data before it is put into make.dep
21:48:47amiconnAh, that may explain it...
21:48:50*amiconn tries
21:50:28***Saving seen data "./dancer.seen"
21:51:21Zagorfunman: it stops for me
21:52:37amiconnZagor: Do you know why 'make clean' waits for a while before the actual cleaning (which is fast even on cygwin)?
21:52:48Zagoramiconn: it reads make.dep
21:53:03amiconnDoes it need that for cleaning??
21:53:16Zagorno, we could make that conditional
21:53:38*amiconn thinks this would be a good thing
21:54:42Zagorfunman: what environment are you getting that in?
21:55:10funmanZagor: In fact I think these non-stopping errors I see are the dependencies generation - i'm running linux
21:56:12Zagorfunman: ahh, yes. do you get it in pegbox?
21:57:10funmanZagor: in various, numerous, plugins which don't build on the Clip; nothing wrong.
22:01:03Zagoramiconn: actually the reading of .make files takes longer than the reading of make.dep (at least here). and we "need" to read .make since CLEANOBJS is populated in them
22:01:29Zagorthe reading of .make files also includes parsing of SOURCES etc.
22:02:10 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
22:02:14amiconnPerhaps that could be made faster?
22:02:38 Quit MethoS_ (Read error: 110 (Connection timed out))
22:02:50 Quit tyfoo ("Carpe diem")
22:03:32 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
22:03:32ZagorIf we want a very fast clean I'd say we rather skip reading the submakefiles and keep the whole clean file list in root.make
22:03:42Zagorhow much of a delay is it for you?
22:04:54 Quit nplus (Remote closed the connection)
22:05:23 Nick kachna|lappy is now known as kachna (
22:06:06Zagorand why do you keep using cygwin instead of vmware?
22:06:59 Join tyfoo [0] (
22:07:47amiconnAround 20 seconds
22:07:59amiconnCygwin is simply more conveninet, even if it is slower
22:08:01kugeluh 20s?
22:08:14kugelit's ~2s on my slow linux machine
22:08:35Zagorcygwin is so extremely slow I can't imagine how can ever feel it convenient. but it's your time...
22:08:36amiconnZagor: After rebuilding the .dep, the dependencies work, btw. Sorry for that...
22:08:39*kugel almost forgot how slow cygwin is
22:08:54Zagoramiconn: ok good. I can understand you don't wanting to rebuild more than you have to in cygwin... :-)
22:09:21amiconnkugel: Well, a full swcodec build takes around 12 minutes, so...
22:10:31kugeli think I used cygwin once or twice before getting sick from that slowness and switchted to virtualbox
22:10:42amiconnCygwin speed depends quite a bit on fragmentation; directly after defragmenting it's significantly faster
22:10:51funmanamiconn: did you think about using cygwin's terminal to ssh into a faster vmware? that way you could keep the convenience of cygwin
22:11:22 Join jeffdameth1 [0] (
22:11:45bluebrotherwhat's the convenience of cygwin if you only ssh to another box? You can use putty for that too.
22:11:47amiconnfunman: The terminal is the easy part, I know. The nastiness is accessing the source files from windows with the editor
22:12:15Zagoramiconn: full ipodvideo build including make.dep takes 36 seconds here...
22:12:33 Join jbesclapez [0] (n=chatzill@
22:12:37kugelamiconn: there's a bunch of text editors that can edit files over network
22:12:42amiconnCertianly on quite a bit more powerful hardware...
22:13:09jbesclapezhi there!! it is my first time in this channel!! Rockbox is soo good i love it!!!
22:13:14 Quit jeffdameth (Read error: 60 (Operation timed out))
22:13:25amiconnkugel: That isn't the problem. Sure I could use samba. But then I would have to have the VM *always* running when editing the sources
22:13:26jbesclapezI saw today that there was a V3!!!
22:13:31Zagoramiconn: perhaps, but still not very fast. this box is 3-4 years old.
22:13:39jbesclapezdo some of you have tried it?
22:13:55amiconnAnd a VM uses significantly more RAM than cygwin
22:13:58bluebrotherZagor: you cheating with ccache? ;-)
22:14:04kugelI don't see a problem in that
22:14:10jbesclapezCan it now handle Videos???
22:14:22Zagorbluebrother: oops, I did. testing with cleared cache now...
22:14:27amiconnBuilding in a linux VM takes around 2:30 here (at least it did with the old make system)
22:14:53advcomp2019jbesclapez, yea.. you can read the manual too
22:15:01Zagorbluebrother: otoh ccache is a vital part of the build system, so I wouldn't call it cheating
22:15:30gevaertsZagor: not on cygwin :)
22:15:45kugelhow many ram do you have? giving the vm only 256MB should be very much enough to be still significantly faster than cygwin
22:15:53bluebrotherZagor: I agree that ccache can be quite vital, but when comparing with cygwin it might have interesting effects :)
22:15:59amiconn1GB. But that's often almost full even without a VM running...
22:16:24Unhelpfuli'm imagining that moving a 24-byte bmp read buffer to the stack was probably OK to do... moving the huge scaler buffers would not be?
22:16:53gevaertsUnhelpful: how huge?
22:16:54kugeland what's bad about the ram being full? you watch many videos on youtube while coding?
22:17:00jbesclapezI know i can read the manual... but it is faster to ask! i just hvae 2 questions i am using ROckbox for 3 years!!!
22:17:16jbesclapezwith ipodlinux and Ilaunch
22:17:16kugelalso, ram is extremely cheap these days, an upgrade might be worth it
22:17:44gevaertsjbesclapez: it may be faster for you, but if everyone just asks because it's faster, we'd be spending all our time answering the same questions again and again
22:17:46amiconnI can't upgrade further
22:18:11amiconnThis is a laptop, and the chipset doesn't support more than 1GB (512MB onboard + 1x 512MB SO-DIMM)
22:18:11jbesclapezOk leave it!!!!
22:18:20jbesclapezit was short YES or NO
22:18:26Unhelpfulgevaerts: down to 24xLCD_WIDTH bytes
22:18:36jbesclapezsorry to hvae taken so much of your precious time
22:18:40Zagorjbesclapez: you did get a yes.
22:18:46 Quit {phoenix} (Remote closed the connection)
22:18:51jbesclapezok cool
22:18:53Unhelpfulcan't go smaller without sacrificing either downscaler max scalefactor or image quality
22:19:00gevaertsUnhelpful: nearly 8k. Too much for the stack
22:19:00amiconnUnhelpful: Definitely too much for the stack
22:19:14jbesclapezi will go to bed now...
22:19:15amiconngevaerts: Even more e.g. on the Ondas
22:19:21jbesclapezwill talk to you tomorow guys
22:19:34 Quit jbesclapez ("ChatZilla 0.9.84 [Firefox]")
22:19:38gevaertsindeed. How wide are those? 480 pixels?
22:20:10Unhelpfulguh, 9KB
22:20:14amiconnBut the m:robe 500 display is even 640 pixels wide
22:20:30*amiconn wonders what happened to that port, btw
22:20:45bertriko hi funman, don't bother to look at the battery of the clip, I can see the markings on the abi disassembly pictures
22:20:51gevaertsAnyway, I don't think anything above 1k or so can go to the stack
22:26:24 Join Bensawsome [0] (n=Bensawso@unaffiliated/bensawsome)
22:26:52funmanbertrik: would battery_bench be useful to figure the battery levels ?
22:26:56 Quit kharo (Read error: 110 (Connection timed out))
22:27:57 Join markun [0] (i=5418e50a@gateway/web/ajax/
22:28:03 Join kharo [0] (
22:28:59bertrikfunman, yeah battery_bench would definitely help, but it requires sd write support of course...
22:29:30bertrikI think it's not that important to have calibrated discharge curves at this point though ... nice but far from critical
22:29:35funmanbertrik: sd write support is already done .. didn't you read the forum ? :)
22:31:07bertrikfunman, oh didn't read that part apparently
22:31:17kugelZagor: what did you think about ?
22:34:00J-23funman: normal build for e200v2 with your patch returns that TIMER_FREQ is undefined for many plugins
22:34:06Zagorkugel: I like it, just as I like FS #9567
22:34:27ZagorI haven't reviewed either in detail though
22:34:50kugelZagor: would you commit it then? I'm searching for anyone who commits it for the past few days :/
22:35:14kugelmy patch is very quickly reviewed. The other one might not, no idea
22:35:29funmanJ-23: you can use 'make bin' to skip plugins compilation
22:36:11J-23I'll rather wait for this code to be commited to SVN repo
22:36:13J-23good nigth!
22:36:48Zagorkugel: yeah I'll commit your patch
22:36:58 Join fastolfe [0] (n=446c60f3@gateway/web/cgi-irc/
22:38:13 Quit pixelma2 ("-")
22:38:25 Join pixelma [50] (i=pixelma@rockbox/staff/pixelma)
22:38:25 Quit fastolfe (Client Quit)
22:40:28 Quit funman ("leaving")
22:41:15kugeldamn, why's funman already gone :/
22:47:17pixelmabluebrother: did you read my longish questions earlier?
22:47:26 Quit kharo (Read error: 110 (Connection timed out))
22:49:28 Join kharo [0] (
22:53:45 Join toffe82 [0] (
22:53:53*kugel sees the main menu on his fuze!!!!!!!!!!!!!!!!1
22:54:44kugelI'd so love to scroll through everything
22:54:50kugelbut I can't :(
22:55:26 Quit kharo (Remote closed the connection)
22:55:41kugelthe display needs some calibration I think
22:55:54kugelit looks a bit green'ish
23:05:03Unhelpfulkugel: i should point out that i've done nothing to the grayscale resizer, nor the mono one. also that they're both broken until converted to use the store_part callback, and that they probably won't see any great improvement, anyway.
23:05:30Unhelpfuli doubt i can make any of the nice scalers smaller than the nearest-neigbor scaler provided for those, and there's really no point, anyway.
23:05:40kugelUnhelpful: well, I think for grey scale the simple resize (which is in svn) will be sufficient
23:06:13Unhelpfulthe nearest-neighbor resize is in vanilla svn, for grey and bw? i had not known that.
23:06:42kugelUnhelpful: svn has 2 resize algroithms (for plugins only): smooth and simple
23:07:02Unhelpfuloh, the ones from the pluginlib, yes.
23:07:04kugelthe smooth one is as of now only used for color, and simple for greyscale
23:07:04bluebrotherpixelma: no ... will check the logs
23:07:27kugeliirc the simple one works by removing some lines
23:07:41kugeland doubling some at scaling up respectifly
23:08:21Unhelpfulright, they're nearest-neigbor.
23:08:54amiconnb&w does not need resize in the core
23:08:55kugelsmooth is area sampling afaik
23:09:17*amiconn thinks we need a way to share source files between core and plugins
23:09:39 Quit EspeonEefi ("さよなら")
23:10:16Zagoramiconn: we use two ways already. neither is terribly elegant though.
23:10:18amiconnIt would be useful to offer the smooth resize to greylib plugins on b&w and greyscale targets
23:10:46 Quit petur ("Zzzz")
23:10:51Unhelpfulamiconn: can't anything in core theoretically get an export in plugin API?
23:11:02amiconnBut the greylib has (of course) a different pixel format that any core, so the code needs to be recompiled with some macros defined differently
23:11:23amiconnUnhelpful: The code won't be in the core on b&w target. So nothing to export
23:11:45Unhelpfuli meant re: a way to share source files
23:12:02kugelI propose moving the plugin dir to the root, and creating a share/ dir for shared c files
23:12:34Unhelpfulamiconn: or do you mean, you want something to be possibly in the core and plugin API on one target, and in the pluginlib on another?
23:12:38bluebrotherwhat's wrong with having both alongside?
23:12:45amiconnUnhelpful: Yes, or even both
23:13:20Unhelpfulamiconn: i was thinking about the same kind of thing the other day, though i can't recall why.
23:13:40bluebrotherwould it work to simply have libs around that get linked to either the core or the pluginlib?
23:13:47amiconnPictureflow on greyscale targets should use the greylib, and then we'd need that
23:14:15amiconnThe core loads the .bmp as 2-bit, pixel-packed bitmap. The greylib wants 8-bit
23:14:25Unhelpfulbluebrother: it gets complicated more than a little if you want the same plugin to build with the same source and use either the API export or the pluginlib function, based on target
23:14:27bluebrotherpixelma: sorry, no real idea on this. At least as of now −− will try to shuffle it in my head the next days
23:14:48Unhelpfulsomething with macros, or some trick of "registering" pluginlib stuff into the API, or vice versa.
23:15:01amiconnUnhelpful: Not really. Several plugins already do something like this...
23:15:40amiconnCheck e.g. mandelbrot.c, or (extreme case) cube.c (which even works on the archos Player, i.e. charcell)
23:16:32pixelmabluebrother: thanks, I am thinking about it for a while but I know far less about LaTeX (only what I used so far in my edits of the manual)
23:17:17Unhelpfulamiconn: if we add an output line buffer for the scaler, then it could be feasible to have its output formats be pluggable per invocation, as well as its inputs
23:17:36Unhelpfulwhich would let you ask it for 8-bit-or-2-bit greys
23:18:06amiconnIt would increase core binsize unnecessarily though
23:18:36Unhelpfuldepends on what a particular user sees as "necessary", but yes, it definitely would.
23:18:47amiconnSome core functions (e.g. screendump() ) offer hooks for plugin functions though
23:19:02Unhelpfulit *could* be done without much bloat with a per-pixel output callback, but that would be just horrid. ;)
23:19:03amiconnThe greylib uses that hook to provide proper screendumps
23:19:19 Quit miepchen^schlaf (Client Quit)
23:19:55amiconnThis hook is an all-replacement hook, just to have it called the same way as normal screendumps
23:20:07*kugel adds +1 to his patches in svn count
23:20:19kugelZagor: thanks :)
23:21:02BagderZagor: a minor nit: it would be nice with the author's name included when a patch is committed
23:21:09Zagordoom and rockboy import sscanf.c from core. libtremor imports ctype.c.
23:21:36Unhelpfulto get at non-api-exported functions, i assume?
23:21:55ZagorUnhelpful: yes
23:22:28ZagorBagder: right, sloppy of me
23:22:35Unhelpfulok, my latest is on FS #9458, broken parts and all. don't anybody say i didn't warn them.
23:22:52Unhelpfulbig block letters, only useful on color targets.
23:23:00 Join miepchen^schlaf [0] (
23:23:39kugelUnhelpful: the smooth scaling in svn isn't nearest neighbor, is it?
23:24:18bluebrothersmooth scaling in svn?
23:24:43kugelyes, in pluginlib
23:24:51Unhelpfulkugel: the pluginlib one is the area/bilinear algos originally ripped for the resize-on-load patch.
23:25:05Unhelpfuli don't know if its limitations are the same
23:25:39kugelUnhelpful: he said he uses area sampling for downscaling
23:26:00Unhelpfulif the bitmap scaler expects to have the whole bitmap in memory, it doesn't have to limit itself at a cache of X lines, like the resize-on-load one does for area sampling.
23:26:27kugelare you sure about that?
23:27:35 Quit domonoky (Read error: 104 (Connection reset by peer))
23:28:02kugelI mean that he expects the bitmap already loaded? afaik he changed l the loading mechanism to load line-wise, and scaling them on the load when the resize bit is set
23:28:30kugelotherwise it wouldn't be resize on load
23:28:45Unhelpfulkugel: in that i was referring to the pluginlib one.
23:28:57kugel"like the resize-on-load one does for area sampling."
23:30:12kugelerr, re-reading your sentence reveals the real meaning
23:30:23 Join EspeonEefi [0] (i=eefi@STRATTON-TWO-EIGHTY.MIT.EDU)
23:30:39kugelyes, the plugin one needs the whole bmp loaded, that's why it's not in the core
23:30:50 Join EspeonEfi [0] (i=eefi@STRATTON-TWO-EIGHTY.MIT.EDU)
23:30:59 Part EspeonEfi ("さよなら")
23:31:33 Quit Zagor ("Client exiting")
23:34:08 Quit BHSPitLappy (Remote closed the connection)
23:43:23 Join bluefoxx [0] (
23:44:11bluefoxxi had to reinstall the rockbox bootloader a while ago, and i recall the previus one did something like 'disable database rebuild' when connecting to usb, what happened ot that?
23:44:16bluefoxxas it no longer does this
23:46:52 Quit reacocard (Remote closed the connection)
23:47:44 Join reacocard [0] (n=reacocar@WL-135.CINE.HMC.Edu)
23:48:48BigBambiI assume you are talking about the e200?
23:49:53bluefoxxof course
23:50:29***Saving seen data "./dancer.seen"
23:50:36bluefoxxif a link could be PM'd to me, that would be nice as i am about to head out the door
23:50:43BigBambiWhy of course?
23:50:48bluefoxxlinux or windows would both work, i use either
23:50:53BigBambiThere are other players that do database refresh
23:51:00bluefoxxbecause i havent got the cash to own anything else
23:51:04Bagderbluefoxx: it depends on the OF version
23:51:13BigBambibluebrother: How the hell should I know that?
23:51:35*Bagder offers BigBambi an improved tab completion ;-)
23:51:43BigBambiCheers :)
23:51:58BigBambiMore like improved proofreading skills :)
23:55:07bluefoxxwell, doing something anyways...thanks all
23:57:39 Part Arinohyoshi
23:58:00bluefoxxit's still long should it take?
23:58:15BigBambiWriting what?
23:58:22bluefoxxsansapatcher.exe −−install

Previous day | Next day