Previous day | Jump to hour: 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 | Next day

Seconds: Show Hide | Joins: Show Hide | View raw
Font: Serif Sans-Serif Monospace | Size: Small Medium Large

Click in the nick column to highlight everything a person has said.
The Logo icon identifies that the person is a core developer (has commit access).

#rockbox log for 2009-11-16

00:00:19kugelso is using literal numbers with enums IMO :)
00:00:29gevaertsthat's different
00:01:22gevaertsit should be avoided of course, but you can't always do that
00:01:36kugelwhere can't you?
00:02:12kugelUnhelpful: you also can't really do ++ and −− on enums anymore, right?
00:02:32liargevaerts: it does not lock up if i do tick_add_task(usb_tick) after the disk active led shuts off the first time... TheSeven is going to look at the driver tomorrow
00:04:00gevaertskugel: hm, it depends on what you mean by "literal numbers" :). You don't need the numbers in the code of course, but assigning an enum from another integer can be needed when e.g. talking to other things (hardware, files,...)
00:05:17 Quit FOAD (Read error: 110 (Connection timed out))
00:05:17 Nick FOAD_ is now known as FOAD (
00:05:32kugelIMO enums shouldn't be used if it forces one to use litelal numbers (I mean literal numbers as in: 1, 10, -23, 0x2f3ull)
00:05:56kugelalthough I think we do that right now in a few places
00:07:10 Join AaronM [0] (
00:07:11gevaertsyou mean in the enum declaration?
00:08:39*JdGordon is somewaht surprised works at all for sbs files
00:09:07gevaertsso something like "enum enum_type var = 33"?
00:09:09kugelJdGordon: I put some energy into it, but not too much it seems
00:09:15kugelgevaerts: yea
00:09:28gevaertsHow can you ever be forced to do that?
00:09:34JdGordonthere is still lots of forced ".wps" additions for finding the files
00:09:39kugelI asked you to give me an example :)
00:09:56gevaertsand I said that it depends on what literal means :)
00:10:14gevaertsbut yes, doing that is broken
00:10:16 Quit ender` (Read error: 110 (Connection timed out))
00:11:31 Quit domonoky1 (Read error: 131 (Connection reset by peer))
00:14:29Unhelpfulkugel: math on enums appears to be valid, still. there's a loop on enum in doom, and all i had to do was fix the end condition
00:14:59kugelwhat if a value is skipped?
00:15:16kugelenum { ONE = 1, THREE = 3 }?
00:15:40Unhelpfulkugel: the compiler probably won't mind.
00:15:54Unhelpfulit's really only compares where you run into trouble.
00:16:16 Quit thegeek (Read error: 104 (Connection reset by peer))
00:16:27 Join thegeek [0] (
00:16:31kugelI don't know what happens if you do enum enum_type var = ONE; var++;
00:16:55JdGordonor more fun when you copy the value of one enum to a different one :)
00:17:12JdGordonarnt enums just ints in c? so var++ will set the value of var to 2
00:17:31 Join Lss__ [0] (
00:20:16kugeleabi changes enum behavior a bit
00:20:53 Join Trista281 [0] (
00:22:05 Quit tmzt (Remote closed the connection)
00:22:16 Join tmzt [0] (
00:24:06 Quit bertrik ("De groeten")
00:24:32kugelpixelma: did you try the splash?
00:27:23 Quit Lss (Read error: 104 (Connection reset by peer))
00:28:41 Join saratoga [0] (i=9803c6dd@gateway/web/freenode/x-oobdxotaljhmugtn)
00:30:14saratogaJdGordon: were you still planning to commit the ipod accessory stuff?
00:30:25JdGordonunless you want to?
00:30:30saratoganope :)
00:31:13JdGordonill do them soonish
00:31:53 Join BHSPitLappy [0] (n=BHSPitLa@unaffiliated/bhspitmonkey)
00:32:17UnhelpfulJdGordon: the size of a C enum is not defined. a compiler *may* use the smallest type which can represent all values. EABI does by default.
00:32:58kugelpixelma: adding a splash after "send_event(GUI_EVENT_REFRESH, gwps_enter_wps);" in wps.c to see when the initial display of the wps finished
00:33:48 Quit Tristan (Read error: 110 (Connection timed out))
00:33:56JdGordonArnt 1bit bmp's supposed to use the colours from the fore/background? so black is the foreground and white is background?
00:33:57Unhelpfulit really only matters when casting from a small, unsigned enum to a larger signed type, or when taking a pointer.
00:34:01 Quit fdinel (Read error: 110 (Connection timed out))
00:34:03JdGordonor did I mess something up?
00:35:42kugelUnhelpful: I still don't quite understand the problem of casting an unsigned enum to a larger signed type
00:35:56pixelmaJdGordon: works for me this way in my viewport colours WPS
00:36:52JdGordondo I need to save the bmp's in a special way for it to work?
00:36:54 Join salty-horse [0] (
00:37:46salty-horsehi. when trying to play an m4a file (extracted from an flv) I get this error with the simulator: Unknown MP4 file type: 'isom' −− is it possible to support it? I have not seen an open bug or any mention of this elsewhere
00:38:09 Join salty_horse [0] (
00:38:09pixelmaor I guess white is transparent (so a backdrop would show through if there is one, hmm... I think I saw some weirdness in an sbs though but that was drawing on a different bmp so I thought that was the reason). Well they would be saved as 1-bit
00:38:19 Quit salty-horse (Nick collision from services.)
00:38:27 Quit n17ikh ()
00:38:29salty_horseRE: hi. when trying to play an m4a file (extracted from an flv) I get this error with the simulator: Unknown MP4 file type: 'isom' −− is it possible to support it? I have not seen an open bug or any mention of this elsewhere
00:38:34 Nick salty_horse is now known as salty-horse (
00:38:50saratogasalty-horse: try repacking the file in foobar2000 using its "optimize mp4 layout" tool
00:39:09salty-horsesaratoga, got a unix tool?
00:39:19saratogamp4box or something like that basically
00:39:26saratogamaybe ffmpeg can do it
00:39:49kugelwhat can if ffmpeg can't? :p
00:39:54pixelmakugel: no I didn't
00:40:01saratogaalternatively, you could improve our mp4 parser so it supports all those oddball mp4 layouts
00:40:12kugelcan you do that? I would find that very interesting
00:43:09saratogawe need a way to search Flyspray for people with a lot of patches
00:43:20saratogaso I can see who commits a lot of code but doesn't yet have svn access
00:43:23kugelpixelma: would do that, and show the time it took
00:43:44JdGordonsaratoga: well you can sort patche by submitter
00:43:51JdGordonwell... opener anyway
00:44:08salty-horsesaratoga, is this all I'll generally need for documentation? :D
00:44:56kugelsaratoga: luckily we established a commit template for that now
00:45:05saratogai don't even know what isom is
00:45:12saratogakugel: ?
00:45:47kugelwe usually mention the author of patches and flyspray number when committing someone else's work
00:46:10saratogawell i'd still have to search the commit logs though
00:46:25saratogai just meant a way to figure out which users on flyspray have opened a lot of tasks
00:46:52kugelisn't the number of patches they got in more important?
00:47:05pixelmakugel: that diff doesn't patch with current svn (one hunk fails)
00:47:06kugelsvn log| grep "Author: " works to get an overview
00:47:20kugelpixelma: that was supposed to be applied on top of my other patch
00:47:22saratogasure but writing a lot of patches is a good start to getting a lot in, and I can check from there
00:47:25kugel(sorry for not saying this)
00:47:43pixelmaI thought I did that first
00:47:53pixelmanot sure now though
00:48:19 Quit Omlet05 ("( :: NoNameScript 4.22 :: )")
00:49:25*kugel thinks the number of open taks doesn't correlate to the chance of getting a lot in
00:49:36kugelsee Uchida :)
00:50:02kugelwell, he's got a lot patches in too
00:50:34kugelpixelma: it's possible that it doesn't apply to my earlier patch
00:51:10pixelmanow I'm confused
00:51:10*JdGordon 's remove statusbar patch has maggically stopped working!
00:52:40 Quit AaronM ("Emo Time In My Corner... //_-")
00:52:44kugelpixelma: I've made some more changes in between. this one is the complete one, apply against svn
00:53:47 Quit ender (" Remember: A secretary isn't permanent until she's been screwed on the desk...")
00:54:35saratogaUchida's patches are of very good quality and nearly all should go in
00:54:55saratogai just can't figure out what hes saying and he doesn't respond to my emails so I can't nominate him for SVN
00:56:34kugelsaratoga: yyea, someone should look at the lot of wav extra codecs (or whatever that is exactly)
00:57:11salty-horseaccording to codecs/libfaad/README.rockbox it added a clause to the GPL license to make it incompatible with rockbox. any idea what that clause is? the website says it's GPL
00:57:23saratogathats been resolved
00:57:39saratogathe clause was removed 3 or 4 years ago and GPL compatibility restored
00:58:06saratogathey wanted you to put a copyright notice in a gui or something like that which was arguably not gpl depending on how you read it
00:59:49salty-horseaccording to logs, the code doesn't seem to have synched up with the new releases
01:00:09saratogayeah its never been resynced
01:00:15saratogabut libfaad is a piece of shit anyway
01:00:30saratogai'm slowly replacing it all with better code
01:00:41saratogaemphasis on slowly
01:01:03saratogamaybe we'll get a GSOC student to work on codecs again this year
01:01:03pixelmakugel: around 7 seconds (results vary slightly). As I said it only seems the tag data that takes so long
01:01:05salty-horseis the mp4 code in libfaad or libm4a?
01:01:08JdGordonpixelma: how do I make sure these bmp's are saved as 1bit images? I'm limited to mspaint :p
01:01:15 Join n17ikh [0] (
01:01:18saratogalibfaad is just pure AAC stuff
01:01:28kugelpixelma: something goes really wrong there
01:01:29saratogalilbm4a is our homemade mp4 parser
01:01:36saratogaits not much better then libfaad
01:01:44*kugel goes searching for fishy HWCODEC #ifs
01:02:05salty-horsesaratoga, ok. I'll see what I can figure out. thanks
01:02:13 Quit thegeek (Read error: 104 (Connection reset by peer))
01:02:25saratogaif you feel like digging into m4a parsing, we could use soemone who knows the standard
01:02:34 Join thegeek [0] (n=nnscript@
01:02:41saratogathe current parser also has issues with large files due to how it parses the TOC atom (whatever its called)
01:02:45pixelmaJdGordon: uh, don't know how that works in ms paint
01:02:58JdGordonor what should I be using?
01:03:21kugelJdGordon: any idea why the initial update of the wps takes 7(!) seconds?
01:03:25kugelon the ondio
01:03:30saratogaok i've reviewed all the patches submitted in the last 12 months and we have all the major contributors in SVN or excluded for good reason, the exception being Uchida
01:03:33JdGordonno, thats fubar!
01:03:43saratogaand a few actively being considered
01:03:54JdGordonactually yes... it might actually take that long for the id3 info to be ready
01:04:01JdGordonondio is crazy slow disk speed isnt it?
01:04:14 Quit freqmod_qu (Read error: 104 (Connection reset by peer))
01:04:16 Join freqmod_qu [0] (
01:04:28saratogaso i guess we're not doing that awful about leaving people out
01:05:15pixelmayeah, but (1) it didn't always take so long and (2) everything in the WPS shows quite quickly, just not the tag info and (3) at least there is no spin up time on the Ondio
01:05:19saratogado we have any developers who speak japanese? teru i guess?
01:05:28kugelJdGordon: I think I removed the code that waits for it?
01:05:29saratogamaybe one of them could email uchida and ask him what his story is
01:06:09kugelpixelma: that splashes how much time it spends in the skin engine
01:06:13pixelmaTiMiD too I guess but he's seldom around too (Kévin Ferrare)
01:06:15 Quit liar (No route to host)
01:07:04kugelthere's no #f HWCODEC/SWCODEC in skin_display.c :/
01:07:12JdGordondamn right there isnt!
01:07:27pixelmaJdGordon: there is a seperate item for saving as 1-bit bmp in the list of Paint here
01:07:48JdGordonI wonder if it was removed in win7..
01:08:04kugelgimp can do it I think
01:08:08 Nick Zarggg_ is now known as Zarggg (
01:08:10JdGordonfuck off gimp :D
01:08:14pixelmatried imagemagick?
01:08:49pixelmakugel: this instead of the other patch?
01:08:53kugelno, on top
01:12:37 Quit amiconn (Nick collision from services.)
01:12:39 Join amiconn_ [0] (i=quassel@rockbox/developer/amiconn)
01:12:51JdGordonhow do I use imagemagic to convert to 1bit?
01:12:59 Nick amiconn_ is now known as amiconn (i=quassel@rockbox/developer/amiconn)
01:13:08JdGordonidentify seems to think bmps saved in mspaint as mono are still 8bit
01:13:16 Quit pixelma (Nick collision from services.)
01:13:17 Join pixelma_ [0] (i=quassel@rockbox/staff/pixelma)
01:13:35kugel(just a guess) convert X.bmp -depth 1 Y.bmp ?
01:13:37 Nick pixelma_ is now known as pixelma (i=quassel@rockbox/staff/pixelma)
01:14:41pixelmakugel: a few trys all give me 0:something
01:15:05pixelmatries too
01:15:06kugelalright, very interesting
01:15:38***Saving seen data "./dancer.seen"
01:15:50kugelso the drawing is almost as fast as on other targets
01:16:15pixelmain a range of 0:3 to 0:7
01:16:19JdGordonpixelma: everything static is being shown instantly right?
01:16:27JdGordonits just the id3 stuff being slow?
01:17:03pixelmayes, and all placeholders (my fallbacks and 0:00 for playtime etc)
01:17:23JdGordonthey get shown or they dont?
01:17:53 Quit BHSPitLappy (Read error: 110 (Connection timed out))
01:18:04pixelmathose show immediately and the ID3 is slow (and playback only starts, well listenable when that shows up)
01:18:19kugelhm, thinking about it, the wps shouldn't show anything for those 6-7 seconds
01:18:40kugelthe function call I put the splash around in the latest patch is the very first drawing
01:18:41pixelmait almost looks like it waits until first buffering is finished
01:19:07pixelmaI think it didn't always do that
01:19:58kugelthe first splash patch showed 7 seconds, and the second on <1 second?
01:20:42kugeland you get to see some parts of the wps during the first 6 seconds?
01:21:32pixelmayes, everything except filled out ID3 info and actual playback time/progressbar
01:21:47kugelthat shouldn't be possible :/
01:22:09kugelhow long to parts of the list show?
01:23:01pixelmaall the time I'd say
01:23:16pixelmaI mean the 7 seconds
01:24:12pixelmasometimes it gets cleared but not regularly (with your patches)
01:25:00 Quit stripwax ("")
01:25:20 Join BHSPitLappy [0] (n=BHSPitLa@unaffiliated/bhspitmonkey)
01:25:24kugelcan you test if that one still shows 7s please?
01:26:01 Quit efyx_ (Remote closed the connection)
01:26:01kugelI got it
01:26:14kugelforget the latest one
01:26:50kugeltry if this one fixes the list instead
01:28:01pixelmaagain on top of the others?
01:29:32salty-horsesaratoga, I'm really confused. all I did was add the isom magic number to the places that give an error when they see it, and it plays fine. I really don't know anything about mpeg, so I must be missing something −−
01:29:55kugelyea, but on top of one from 1.05 (skip the 1:25 one)
01:32:39 Quit mikroflops (Read error: 110 (Connection timed out))
01:33:36saratogasalty-horse: neat
01:34:40saratogagoogle suggests its an aternative to the apple tag
01:35:17salty-horse ?
01:35:59saratogaif you can find some explination of the difference that would be good, but if not we could just ask Lear if he thinks its ok and then commit it
01:36:20saratogaeither way you should open a flyspray task for it
01:38:05salty-horseI tried looking for info about it and didn't find much besides ffmpeg's source code, and all its isom.c file has is lists of tags and a conversion function between text encodings (or something similar)
01:39:42kugelpixelma: one of the patches that went in before sbs was to modify the skin engine to not actually wait for ready id3 tags anymore. I left over a check for ready taginfo which delays the very first full update
01:40:33salty-horsesaratoga, wait, it works in the simulator but not on the sansa
01:40:41pixelmakugel: seems a lot quicker already until the WPS shows fully and playback starts - around 3 second, the splash still the time spent in the skin engine and that is down a bit too (0:2 is often seen)
01:41:07kugelah that's nice :)
01:41:14pixelmawhen music starts, ID3 info shows at the same time so it already seems to be ready
01:41:25kugelI'm surprised that playback starts earler too, but it can't hurt :)
01:42:08pixelmayes, feels a lot better already, not that sluggish anymore
01:42:35kugelalright, I'm going to commit the changes later then
01:43:02kugelexcept the splashes that is :)
01:43:49saratogasalty-horse: which sansa?
01:44:10salty-horse280 v1. can I view stderr somehow?
01:44:14 Quit Thundercloud (Remote closed the connection)
01:44:16pixelmakugel: nice, good to see some result from nagging and testing :)
01:45:14saratogasalty-horse: on device you can configure a build with logf and then enable logf in the codec
01:45:23saratogabut i'm not 100% how it works for codecs
01:45:46saratogathough i think double checking this on the sim would be a good idea
01:45:50saratogathere should not be any difference
01:46:15saratogaif anything the sim is usually more strict since it has an MMU and will crash on illegal accesses more readily
01:47:21 Join AaronM [0] (
01:49:48kugelsaratoga: yea, it's more strict for those, but less strict (not strict at all) for alignment problems
01:49:51salty-horsebuilding with DEBUG and Logf
01:51:06salty-horseI connected the usb and the usb icon came up with the current selection in the file browser scrolling side to side on top of it
01:52:36salty-horse/home/ori/devel/rockbox/apps/gui/skin_engine/wps_debug.c: In function ‘debug_skin_usage’:
01:52:36salty-horse/home/ori/devel/rockbox/apps/gui/skin_engine/wps_debug.c:638: error: ‘wps_verbose_level’ undeclared (first use in this function)
01:52:36salty-horse/home/ori/devel/rockbox/apps/gui/skin_engine/wps_debug.c:638: error: (Each undeclared identifier is reported only once
01:52:36DBUGEnqueued KICK salty-horse
01:52:36salty-horse/home/ori/devel/rockbox/apps/gui/skin_engine/wps_debug.c:638: error: for each function it appears in.)
01:53:14saratogathis is just a logf build?
01:53:21salty-horseand DEBUG
01:53:25salty-horseshould I disable DEBUG?
01:54:27salty-horsemissing ifdef around debug_skin_usage
01:54:35saratogai don't think you can compile with DEBUG on target
01:54:37saratogaits a sim thing
01:54:44saratogaDEBUGF that is
01:55:10saratogalogf should be enough, but you'll have to define LOGF_ENABLE and include logf.h in any file that you want to see logf output from
01:55:23saratogait'll show up in the debug menu, you'll see a logf option there
01:55:48saratogaalternatively if you think its an alignment problem, you could get someone with a coldfire target to try
01:55:53saratogathose can do unaligned accesses
01:56:40salty-horsethe configure prompts should have a "> "
02:00:14 Quit hillshum (Read error: 145 (Connection timed out))
02:03:02salty-horsesaratoga, logf says nothing after trying to play the file
02:03:29saratogasalty-horse: at very least you should get the same debugf prints you saw on the sim
02:03:58salty-horsewhat's the logfdump option?
02:04:08saratogaare you sure you've enabled LOGF in the correct files (probably the m4a metadata parser in apps/metadata)
02:04:14 Quit fyrestorm ("Ur skills' fireproof like a wooden panel -- U got feds talking leet on your IRC channel!")
02:04:23salty-horsebtw, isom standard:
02:04:46salty-horseall I did was run tools/configure and chose advanced->logf
02:04:49 Join fyrestorm [0] (
02:04:54saratogathe configure switch enables the logf option in the menus, but you still have to include logf.h and #define it in each file you want to see output from
02:05:03salty-horseah :/
02:05:17saratogaotherwise the logf and debugf calls all filter to nulls
02:05:46salty-horse#define what? LOGF?
02:06:02saratogacheck logf.h
02:06:06saratogai forget what it is
02:06:57salty-horseROCKBOX_HAS_LOGF ?
02:08:07saratogayeah looks like it
02:08:27salty-horseautoconf.h already defines it
02:08:46salty-horseso I just included it in demux.c
02:10:08saratogaapps\metadata\mp4.c might be neat too
02:10:21salty-horseit already has logf included
02:10:28salty-horseok still nothing in the logf
02:10:30saratogathough if the device is going to crash you won't be able to see anything
02:10:44salty-horseit doesn't crash
02:10:53saratogathen its not an alignment issue
02:11:04salty-horsejust tries all the files in the dir (while reading their metadata!) and returns to file list
02:11:15saratogatrying to do an unaligned access on hardware that can't do that will kill the whole system
02:11:34saratogaROCKBOX_HAS_LOGF has to be defined in each file you want to see output from
02:11:36salty-horsewant one of the files? it's from youtube
02:11:36saratogadid you do that?
02:11:47saratogapost it in an FS task with your patch
02:11:51salty-horseyes, and I got warnings from the compiler
02:11:55salty-horseI'm not sure if it's legal
02:12:09saratogatruncate it to be less then 30 seconds and then its legal
02:12:52salty-horseI'm not sure how to do that with the tools at hand. otherwise I would have already removed the isom compatibility header
02:13:12kugeldid you define LOGF_ENABLE before #include "logf.h"?
02:13:30salty-horsenope :/
02:13:31saratogai would just open it in an editor, copy a couple hundred KB to the clipboard, and then paste it to a new file
02:13:56saratogaas long as you get the mp4 header, which should be at the front of the file, it won't really matter that you've horribly corrupted the file for our purposes
02:15:23saratogai'm sort of at a loss how you could manage to fail out of the parser on target without doing so on the sim though
02:15:41Unhelpfulkugel: some code sets these small unsigned enums to -1 as a flag, although -1 is not in the enum type. since the enum type is unsigned char, it's set to 255. therefore (signed)enum_var == -1 can never be true, since gcc will zero-extend when casting to a larger unsigned type
02:15:56salty-horsekugel, your suggestion didn't help.
02:17:21Unhelpfulyou'd have the same thing if you used uint8_t in place of the enum. you can set it to -1 without an error, but if you cast it to signed after that and compare to -1 it will *never* evaluate as equal
02:18:08Unhelpfulthere are flag conditions in mpegplayer and doom which can never be detected due to this, and i suspect there is also a loop which can't terminate :)
02:18:08kugelah yea
02:18:24 Quit saratoga (Ping timeout: 180 seconds)
02:21:55Unhelpfulif it's only an equality comparison you can fix it by casting -1 to the enum type
02:22:30Unhelpfulthere's one in mpegplayer that's worse, since it's if ((signed)enum_var > -1)
02:23:23Unhelpfulobviously that is always true, even if the flag value has been set. worse, if you cast the -1 instead, it becomes always false, since casting -1 to an unsigned type yields the maximum value of that type
02:33:18 Quit killan ("( :: NoNameScript 4.22 :: )")
02:34:10salty-horsesaratoga (if you read the logs),
02:35:54 Quit panni_ ("( :: NoNameScript 3.81 :: )")
02:38:32CIA-80New commit by kugel (r23644): Fix some drawing problems when displaying the WPS initially with sbs enabled. Move initial display of the wps to wps.c too. Should fix bug 3 of ...
02:38:35CIA-80New commit by kugel (r23645): Avoid two subsequent full update.
02:38:39CIA-80New commit by kugel (r23646): Remove a left-over check for metadata info being ready that delayed the initial draw of the wps quite a bit (most notably on slowish hwcodec). The ...
02:38:49kugelJdGordon: there's a bug with updating next track info
02:41:04kugelit doesn't work half of the time
02:41:14kugelwhen skipping manually
02:45:37kugelhm, maybe it's only a problem on the fuze
02:47:18 Join calman_ [0] (
02:47:41 Join Strife89 [0] (
02:49:38 Join Llorean [0] (
02:51:10 Quit salty-horse ("Leaving")
02:59:06 Quit kugel (Remote closed the connection)
03:00:32 Quit Llorean ("Leaving.")
03:00:38mc2739kugel: I was seeing weird next track info also, playing track 1 - it was showing track 3 as next for about 3 seconds - then it showed track two. I cleared settings and that seems to have corrected the problem
03:09:38 Join Jobo [0] (
03:10:16JoboHello, I had some trouble with my sansa e280 v1. I reformatted the player to erase anything that was on it previously.
03:10:37JoboI am now trying to remember how to go about reinstalling rockbox.
03:11:44 Join JdGordon| [0] (
03:12:17mc2739Jobo: the easiest way is to use the installer - just click the releases linh on the nav bar of the Rockbox home page
03:12:49 Nick fxb is now known as fxb__ (
03:12:58mc2739Jobo: you can then choose to install the latest release or the current development build
03:13:12JoboI skimmed by something that said something about a "sansa patcher" I was looking through the manual to find out what that was for.
03:13:29 Join FOAD_ [0] (
03:13:51JoboI know I am supposed to read the manual and the faq before I ask questions here... I could use some extra help though if possible.
03:14:20 Quit fyrestorm (Read error: 104 (Connection reset by peer))
03:14:41JoboI think I know how to install with the installer... I wanted to find out more about the current builds.
03:15:40***Saving seen data "./dancer.seen"
03:15:43JoboIs there any information you can give me to help me??
03:16:18Unhelpfulread the manual, and use rbutil, it's the easiest way to install it.
03:16:58JoboShould I install an unstable version? a current build? what is a sansa patcher for?
03:17:26mc2739Jobo: As I said, the installer is the easiet way. Just click the installation tab and you will be able to install the bootloader (the installer uses sansapatcher to do that).
03:17:39 Quit MethoS- (Remote closed the connection)
03:17:52Joboalright thank you.
03:18:00JoboI really appreciate it.
03:18:18mc2739The current builds have bug fixes and may have additonal features that are not in the latest release.
03:18:25 Quit JdGordon| ("Miranda IM! Smaller, Faster, Easier.")
03:18:36 Join fyrestorm [0] (
03:18:38 Join saratoga [0] (i=9803c6dd@rockbox/developer/saratoga)
03:20:14Joboyou are speaking of "the current build" meaning the "stable" or "unstable" release?
03:20:19 Join froggyman [0] (
03:22:39mc2739the current builds are not necessarily unstable. There is the possibility that minor problems may be introduced as new features are added, but they are usually fixed quickly
03:23:51mc2739Once Rockbox is installed on your e280, an update is as simple as unzipping the new version over your existing version
03:24:11saratogastable and unstable aren't releases, they're classifications of ports, not actual builds you can install
03:24:49Unhelpfulthere is no unstable "release". a "release" has a version number, the most recent one is 3.4.
03:24:54 Join webslut2009 [0] (
03:25:07JoboWhat would you recomend I should do?
03:25:21saratogatheres no reason to ever use anything but the newest version
03:25:22Strife89Get the installed and read the manual.
03:25:39JoboI am confused
03:25:56Strife89Okay, let's take it from the top;.
03:26:03Strife89Jobo, what is your device?
03:26:16saratogalets not take it from the top
03:26:21saratogago read the damn manual and do what it says
03:26:24JoboI am
03:26:25Strife89Jobo: Okay.
03:26:29saratogathis isn't that hard and you don't need someone to read it to you
03:26:32JoboI know
03:26:38Jobothere are just a few things I am confused about
03:26:41Strife89Jobo: Wait, I'll make this as simple as I possibly can.
03:26:45saratogathen ask questions or go away
03:26:56JoboI dont mean to bother everyone
03:27:01JoboI will just do it my damn self
03:27:11JoboI wanted to know
03:27:20saratogawant to know what you haven't even asked a question
03:27:21Strife89Jobo: If you look at Chapter 2 of the manual:
03:27:40JoboI want to know
03:27:43Jobowhat is the difference
03:28:00Jobobetween the newest releases
03:28:01Unhelpfulbetween what and what?
03:28:14saratogareleases are numbered
03:28:19Jobothe current builds or the stable builds and stuff
03:28:20saratogalarger numbers indicate newer releases
03:28:25saratogathat is the difference
03:28:26JoboI wanted to know what is better for me to install
03:28:33saratogadid you see my reply?
03:28:33Strife89Jobo: Release version 3.4, while fairly new, does not contain the VERY LATEST improvements and features.
03:28:50 Quit FOAD (Read error: 110 (Connection timed out))
03:28:51 Nick FOAD_ is now known as FOAD (
03:29:00Strife89Jobo: The "current build" is made several times a day, and contains all the latest stuff.
03:29:29Joboso is it "stable" to use them??
03:29:31Strife89Jobo: You can use the Automated Installer to install 3.4, then manually update it with the current build.
03:29:32Joboare they buggy?
03:29:54Jobook so if I download the current build
03:29:55Strife89Jobo: Usually, the current build is fine, but ON OCCASION someone might goof up something.
03:29:56 Quit webslut2009 ("CGI:IRC (Ping timeout)")
03:30:05Joboand then drag it over to the .rockbox file
03:30:15JoboI understand
03:30:34JoboI was also wondering what kind of features are on the newest build?
03:30:35saratogafrom the install directions: "The current build is built at each source code change to the Rockbox SVN repository and represents the current state of Rockbox development. This means that the build could contain bugs but most of the time is safe to use."
03:30:40Jobothat differ from 3.4
03:30:48saratogatheres a link on the front page listing them
03:30:54JoboI know
03:31:02Jobobut I wanted to hear what you guys think...
03:31:16saratogai think you should check out #rockbox-community
03:31:17Strife89Jobo: Most of us just use the Current Build.
03:31:18Unhelpful"we guys" wrote the stuff you read already. :/
03:31:19Jobosee I would have just installed the "stable build" because I dont want it to act up since I am not so savy
03:31:26 Join killan [0] (
03:31:46Jobobut hearing from you folks I think it would be best to install the latest realease.
03:31:52JoboI am sorry to pester you all
03:31:55Strife89Jobo: If you are more concerned about stability, then just install 3.4 and be done with it until 3.5.
03:32:16Jobothats why I got on this channel to ask you
03:32:17mc2739Jobo: then that is what you should do. The new features will be in the next release and you can update again when it comes out
03:32:38Jobowell you were saying the latest release will be fine
03:32:56Strife89Exactly. The devs focus on stability before each release, so if you only install releases, you're quite safe.
03:33:08 Join webslut2009 [0] (
03:33:28webslut2009I are the bestest people where it coming two used grammer and I as well no how too utility me punctualization to,
03:33:39Strife89Whereas if you go for the *very latest* version, you run a risk, however minor, of hitting unforseen bugs.
03:33:55Jobowould the benefits outway the risks/
03:34:12Strife89Jobo: From my experience, yes.
03:34:18Jobothat is what I wanted to hear
03:34:31JoboI am sorry for any complications I may have caused.
03:34:41JoboI only wanted to find clarity to my situation.
03:34:56saratoganext time just read the manual, everything you asked was in the first 2 paragraphs of the install directions
03:35:00Strife89Jobo: Just be sure to read up next time, k? :)
03:35:02saratogawould have taken you 1/10th as long
03:35:07froggymanJobo: i would also agree with Strife89 that it is usually best to use the latest version
03:35:28saratogaand would save a couple dozen people reading the logs int he morning the trouble of learning about you learning about rockbox
03:35:29JoboI know I have read it... I just wanted to talk to someone about the latest daily builds
03:35:45JoboI am sorry
03:35:49saratogaif you just want to chat, check out rockbox-community
03:35:55JoboI am not trying to chat
03:36:01JoboI am trying to find answers!!!
03:36:11Strife89Jobo: He means if you want to hang out after this.
03:36:11saratogaok talk about things that don't belong here
03:36:14JoboI just dont like always getting the cold shoulder
03:36:17Jobothank you
03:36:37Strife89saratoga: At least, I think you do. :/
03:37:28webslut2009Saltwater Niggers!
03:38:30Mode"#rockbox +o Unhelpful " by ChanServ (ChanServ@services.)
03:39:10Mode"#rockbox +o scorche " by ChanServ (ChanServ@services.)
03:39:14Mode"#rockbox +b %webslut2009!*@* " by scorche (n=scorche@rockbox/administrator/scorche)
03:39:35scorchewebslut2009: you ahve been muted for 10 minutes...please use this time to go over the guidelines of this channel as linked in the topic
03:40:00 Quit webslut2009 (Client Quit)
03:47:59Mode"#rockbox -b %webslut2009!*@* " by scorche (n=scorche@rockbox/administrator/scorche)
03:48:10Mode"#rockbox -o Unhelpful " by ChanServ (ChanServ@services.)
03:48:17Mode"#rockbox -o scorche " by ChanServ (ChanServ@services.)
03:48:30Mode"#rockbox +o logbot " by ChanServ (ChanServ@services.)
03:56:28 Join T44 [0] (
03:57:08 Quit BHSPitLappy (Read error: 110 (Connection timed out))
03:58:16 Join fdinel [0] (
04:04:22 Quit togetic ("WeeChat 0.3.0")
04:04:37 Join togetic [0] (n=togetic@unaffiliated/ibuffy)
04:08:25 Quit Rondom (Nick collision from services.)
04:08:35 Join Rondom [0] (
04:14:01 Quit Topy (Read error: 113 (No route to host))
04:15:00 Part Jobo
04:20:03 Quit TheSeven (Nick collision from services.)
04:20:22 Join The_Seven [0] (n=theseven@rockbox/developer/TheSeven)
04:20:34 Nick The_Seven is now known as TheSeven (n=theseven@rockbox/developer/TheSeven)
04:25:25 Quit froggyman ("CGI:IRC")
04:29:21 Quit CaptainKewl ("( :: NoNameScript 4.22 :: )")
04:51:44CIA-80New commit by jdgordon (r23647): add some bmp strips needed for the rec screen which wernt already added...
04:53:51CIA-80New commit by jdgordon (r23648): get ready for classic_statusbar... get it added to the .zip. needs fixing to work with .sbs files better
04:59:23 Quit togetic ("WeeChat 0.3.0")
05:02:08 Join togetic [0] (n=togetic@unaffiliated/ibuffy)
05:15:05 Quit Strife89 ("Bed.")
05:15:42***Saving seen data "./dancer.seen"
05:17:02JdGordonalrighty.... who knows why these damn bitmaps arnt showing properly?
05:18:31 Join Blue_Dude [0] (n=chatzill@rockbox/developer/Blue-Dude)
05:19:34Blue_DudeJdGordon: ping!
05:19:39JdGordongood evening
05:21:04Blue_DudeHey there. I was reading the log and saw a concern from pixelma that may concern my patches. He was complaining about playback not buffering mp3 data soon enough on the wps. Did I read that right?
05:21:23JdGordonpixelma is a she :)
05:21:39Blue_DudeOops. Sorry pixelma, when you read this...
05:21:43JdGordonthats what it sounds like
05:24:02Blue_DudeHm. I noticed something in the audio thread in passing that I fixed a few weeks ago. The audio thread would keep hammering away at buffering to try to load the next mp3 data even when it is idle, i.e. no playlist. So a couple of times a second it do this mindless routing even though nothing was playing. I fixed that, but I wonder if it may have caused some of the non-responsiveness she...
05:24:04Blue_Dude...talked about.
05:24:15Blue_DudeOn the other hand, I haven't noticed a problem...
05:25:01 Quit Lss__ (Read error: 104 (Connection reset by peer))
05:26:04JdGordonthe ondio is hwcodec, so in all rights shouldnt be affected by your patches at all no?
05:28:09Blue_DudeNever mind.
05:28:33Blue_DudeI don't know which is which most of the time.
05:29:52JdGordonall the archos are hwcodec, the rest are usable :)
05:30:02saratogai didn't realize HWCODEC bypasses pcm buffering
05:30:12saratogai suppose the codec just dumps the pcm directly to the dac?
05:30:19Blue_DudeI found the patch anyway. It's r23487. The audio thread would try to calculate the watermark even when not playing. Lame.
05:31:12JdGordonsaratoga: yeah, hwcodec just pumps mpeg data to the codec and *magic* :)
05:31:17JdGordon(may be slightly simplified)
05:31:21Blue_DudeThe codec is hardware, so I'm sure it uses an internal mechanism to pass data.
05:31:34saratogaanyone think making test builds of the clip buffering patch for other swcodec targets makes sense?
05:31:46saratogait seems like a general buffering bug thats uncommon on most targets
05:32:21Blue_DudeI looked at the flyspray thread and I couldn't even follow it. I know next to nothing about buffering.c.
05:32:34JdGordonsaratoga: I really cant imagine it causing probalems on other targets
05:32:42Blue_DudeAs far as I'm concerned, it's all magic function calls.
05:33:21saratogaJdGordon: because you're sure the fix is correct or another reason?
05:33:59 Join AndyIL [0] (n=pasha_in@
05:34:37JdGordonisnt it fixing paths which are almost never hit on high mem targets?
05:35:32saratogaJdGordon: its fixing how handle locations are calculated
05:35:45saratogathis will definately change how all devices buffer
05:36:02saratogaits just that the corner case it fixes becomes nearly impossible if the buffer is large enough
05:37:23Blue_DudeI'm glad you guys are on that. Tracking buffer logic gives me a headache.
05:37:30saratogai'm barely on it
05:37:37saratogai just want to be able to use my clip
05:38:16JdGordonI'm just here giving not very useful responses
05:38:47Blue_DudeAnd I gave you a few thousand bytes back for the clip. Use 'em!
05:38:51saratogai guess i'll just start using it on my players and see if anything breaks
05:39:40JdGordonwhats the fs#?
05:40:40saratogafunman put a nice comment in buffering.c explaining the bug
05:41:02JdGordonwhich patch?
05:41:38Blue_DudeBarely tested update inbound...
05:41:44saratogathe last one in the thread
05:42:38CIA-80New commit by blue_dude (r23649): pcmbuf cleanup
05:43:24Blue_DudeAnd that should about do it for pcmbuf. Whatever's left I've got other plans for...
05:44:04 Quit Horscht ("Verlassend")
05:44:30saratogahas ui responsiveness been improved?
05:45:07 Quit AndyI (Read error: 110 (Connection timed out))
05:46:53Blue_DudeDoubtful. It was mostly saving memory. Some stuff might run faster but it would be marginal.
05:47:08Blue_DudeThat and it's 100% easier to read now.
05:47:11saratogawhat were you saying about that then?
05:48:21Blue_DudeI changed playback.c a couple of weeks ago to stop hunting for metadata when the player is idle. I was thinking that it might have some bearing on her problem. But with a hwcodec target it's very doubtful. :)
05:48:44saratogai mean a week or two ago you mentioned UI responsiveness
05:49:21Blue_DudeOh, that's playback.c. Other than split out codec_thread and clean it up, I haven't addressed it yet. It may be unfixable.
05:49:28JdGordonBlue_Dude: you shold get it to preemtivly buffer the playlist the user is going to load!
05:50:09Blue_DudeI should have it wake the player up five seconds before it's turned on. That would be a good patch.
05:51:12JdGordonof course!
05:51:59Blue_DudeWow, that last commit didn't save squat. I think we've definitely hit the wall of diminished returns. Next file.
05:54:01Blue_DudeI guess I ought to get started on playback.c. *sigh*
05:54:26JdGordonwell... its still getting to cleaner code isnt it?
05:54:34Blue_DudeI've still got to come back to pcmbuf and do the software mixer.
05:54:56JdGordonany chance you could do a proper writeup of how it works and how it should be used and stuff?
05:54:57Blue_DudeIt is. I'll settle for bare readability.
05:55:22CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
05:55:22*Blue_Dude shrieks.
05:55:36 Quit fdinel ("Miranda IM! Smaller, Faster, Easier.")
05:55:46Blue_DudeGo figure it out! Puts hair on your chest.
05:56:26JdGordondoesnt the stiff drink you need before going into that code do that?
05:56:57Blue_DudeAnyway, wussy crying is over. I think I have wiki access. I'll probably have to write up something. What needs explanation?
05:57:33Blue_DudeThat's... broad.
05:59:20 Join orly_owl [0] (
05:59:39orly_owlInstalling Rockbox will wipe the music off my ipod won't it?
05:59:58saratogai guess a high level explanation of how it fits in with the audio drivers and codec files would be useful
06:00:29orly_owlno to me?
06:01:31 Join Blue_Dude_ [0] (n=chatzill@rockbox/developer/Blue-Dude)
06:01:52JdGordonyes, you wont lose your music
06:02:03orly_owloh really? cool
06:02:23orly_owlim guessing it doesnt reformat the parition then
06:02:26Blue_Dude_I'm back. Hotel internet sucks.
06:02:37saratogaonly if it isn't already fat32
06:02:52orly_owlyeah its fat32
06:02:52*Blue_Dude_ checks logs.
06:05:31Blue_Dude_I suppose I could write up something about program flow and how data gets passed back and forth. I've never tried to write for a wiki before, so it might be a while.
06:07:12JdGordonanything that helps other people undestand how it all works
06:08:12Blue_Dude_It's simpler than I thought it was. Correct me if I'm wrong but wasn't pcmbuf split off from playback?
06:09:24orly_owlWhat license is Rockbox Utility released under?
06:09:27Blue_Dude_I think it never quite cut ties. There's still a lot of traffic back and forth. I'm wondering if we might be better off if pcmbuf ran in its own thread.
06:09:34 Join kisuke [0] (
06:10:25orly_owlOK. I didn't see thst mentioned anywhere though.
06:10:26Blue_Dude_And data could pass back and forth via queues rather than globals and function calls.
06:11:26saratogatheres a license in the header of its source files
06:12:22kisukehow does one not get a short read error when trying to install to an iPod video?
06:13:54saratogawhat operating system, and what is the message?
06:14:08Blue_Dude_Anyway. I'm off. Have a good night.
06:14:36 Quit Blue_Dude_ ("ChatZilla 0.9.85 [Firefox 3.5.5/20091102152451]")
06:15:49orly_owloh ok. I was just looking for a LICENSE or COPYING file
06:16:49kisukesaratoga: linux x86 and terminal or log out put?
06:18:03 Quit Blue_Dude (Read error: 110 (Connection timed out))
06:18:29saratogakisuke: what software is this?
06:19:18kisukesaratoga: the automated installer, if i understand what you are asking
06:19:56saratogawhat error did it give you?
06:20:42 Join BHSPitMonkey [0] (n=stephen@unaffiliated/bhspitmonkey)
06:21:13kisukesaratoga: failed to add bootloader
06:21:39kisukesaratoga: it is running as root
06:23:26saratogai don't see that in the source code, so I donno
06:28:44kisukenothing huh
06:31:01saratogaits possible you just don't have the right partitions and such on your ipod if you've messed with paritioning before
06:36:03saratogawell i wrote a faster version of the clip buffering patch that uses a memmove when possible, and it seems to be working fine
06:36:15saratogain that it doesn't instantly corrupt the buffer
06:43:02orly_owlwow this is pretty cool
06:43:02 Quit Sajber^ ("Leaving.")
06:43:07orly_owlthanks everyone
06:59:42 Join webguest74 [0] (
07:02:23webguest74what exactly does the picture flow need to operate? i have folder.jpgs for all albums and they show during playback, then most of my songs i am positive have the art in the tags but only three albums show in the flow mode. I see something about cover.bmp files but no real explanation
07:02:36webguest74i'm on e200
07:08:24 Quit webguest74 ("CGI:IRC (EOF)")
07:15:45***Saving seen data "./dancer.seen"
07:27:51 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
07:29:02 Quit shai (Read error: 104 (Connection reset by peer))
07:29:18 Quit kadoban_ (Remote closed the connection)
07:30:12 Join Guest78013 [0] (
07:30:34 Quit Guest78013 (Connection reset by peer)
07:33:05 Join einhirn [0] (
07:36:41 Join kadoban [0] (
07:43:17 Join Tomis2 [0] (n=Tomis@
07:46:33 Join stoffel [0] (
07:59:57 Join Grahack [0] (
08:00:15 Quit Tomis (Connection timed out)
08:00:15 Nick Tomis2 is now known as Tomis (n=Tomis@
08:01:22 Join Rob2223 [0] (
08:07:36 Part kisuke ("Leaving.")
08:10:07togeticcan you play mp4 files in rockbox?
08:10:36 Quit Tomis (Read error: 104 (Connection reset by peer))
08:17:54 Join Tomis [0] (n=Tomis@
08:19:37 Quit Rob2222 (Read error: 110 (Connection timed out))
08:22:46 Quit FlynDice (Read error: 60 (Operation timed out))
08:22:59 Join ender` [0] (
08:28:56 Join FlynDice [0] (n=FlynDice@
08:29:58 Join Tomis2 [0] (n=Tomis@
08:35:10 Quit TheSeven (Read error: 113 (No route to host))
08:35:48 Quit Tomis (Read error: 60 (Operation timed out))
08:35:48 Nick Tomis2 is now known as Tomis (n=Tomis@
08:37:53 Join lyngaas1 [0] (
08:40:08 Quit lyngaas (Read error: 110 (Connection timed out))
08:44:43 Quit Grahack ("Leaving.")
08:51:47 Quit Rob2223 ()
08:54:00 Join maruk [0] (
08:57:54 Quit calman_ (Read error: 110 (Connection timed out))
09:00:20 Join flydutch [0] (
09:10:20 Join petur [50] (n=petur@rockbox/developer/petur)
09:14:30linuxstbtogetic: Rockbox supports AAC audio and Apple Lossless audio in mp4 files. It doesn't support any kind of video i mp4.
09:15:47***Saving seen data "./dancer.seen"
09:22:35 Join Thundercloud [0] (
09:28:53 Join stripwax [0] (
09:29:27 Quit stripwax (Client Quit)
09:31:07 Join Tomis2 [0] (n=Tomis@
09:45:46 Quit Tomis (Read error: 110 (Connection timed out))
09:45:47 Nick Tomis2 is now known as Tomis (n=Tomis@
09:46:34 Quit Thundercloud (Remote closed the connection)
09:49:38 Join pamaury [0] (n=pamaury@
09:50:19 Join Bagder [0] (
09:54:15 Join DerPapst [0] (
10:15:56 Join leafwiz [0] (n=quassel@
10:16:18leafwizHi, I'm running Amarok 1.4.10, on the latest Arch. I'm trying to connect to my rockbox ipod. I can mount the ipod, and see the files on it in Amarok. But I would like to mount the database on the ipod
10:16:52leafwizIt's connected at a generic device.
10:17:07leafwizAmarok didn't autofind it. But I have hal and dbus running
10:19:10 Join fyre^OS [0] (
10:19:27linuxstbleafwiz: Are you using an official release of Rockbox (e.g. 3.4) or a current build?
10:19:49 Quit freqmod_qu (Read error: 60 (Operation timed out))
10:19:59 Join freqmod_qu [0] (n=quassel@
10:20:34 Quit phanboy4 (Read error: 60 (Operation timed out))
10:21:42leafwizlinuxstb: I'm using 3.0
10:22:28leafwizlinuxstb: I must upgrade?
10:27:56 Quit pamaury (Read error: 148 (No route to host))
10:30:54 Join phanboy4 [0] (
10:33:36 Join pamaury [0] (
10:36:34 Join liar [0] (n=liar@
10:37:12 Quit fyrestorm (Read error: 110 (Connection timed out))
10:39:59 Join Rob2222 [0] (
10:46:57leafwizlinuxstb: I got it to work. It was just something with the filepermissions
10:48:29leafwizBut I also have a problem with updating the amarok musik db. I have selected a dir on a ntfs mount. Amarok tells me that it is searching through my files. But there isn't any updated in the status field
10:50:42leafwizbut that's a questiong for the amarok guys.. :)
10:51:04 Part orly_owl
10:54:27 Join salty-horse [0] (
10:56:09salty-horsesaratoga, seen my tracker item?
10:58:08linuxstbleafwiz: 3.0 is _very_ old in Rockbox terms. But you don't have to upgrade if you're happy - see for what's changed since 3.0
10:59:04leafwizI have upgraded. It's so easy to do.
10:59:14linuxstbsalty-horse: saratoga normally isn't around at this time - wait another 12 hours or so...
10:59:20leafwizSo now I'm on 3.4 and everything works.. :)
11:00:04salty-horselinuxstb, thanks. can I manually add people to cc? yesterday he said Lear might want to look at it
11:15:48***Saving seen data "./dancer.seen"
11:22:29 Join toemrs [0] (
11:22:30 Quit toemrs (Client Quit)
11:22:35 Join tomers [0] (
11:23:04tomersJdGordon: I saw your comment for the rtl tag patch.
11:24:55 Quit tomers (Client Quit)
11:36:17 Join kugel [0] (i=kugel@rockbox/developer/kugel)
11:42:29salty-horseshalom tomers
11:46:30 Quit BHSPitMonkey (Read error: 113 (No route to host))
11:51:58 Join MethoS- [0] (n=clemens@
12:00:31 Quit parafin (Read error: 60 (Operation timed out))
12:02:21 Join parafin [0] (
12:03:25 Join Sajber^ [0] (
12:17:30 Join tomers [0] (n=86bfe845@
12:17:33tomerssalty-horse: Ping
12:18:12salty-horsehi :)
12:18:28salty-horsenice work on the wps :)
12:18:36tomersHi, We should start a group :-)
12:19:37 Quit parafin (Read error: 60 (Operation timed out))
12:19:39tomersThanks. Let me know if there are other Hebrew related bugs
12:22:41 Join parafin [0] (
12:22:43salty-horseI'm not using it (like english), but it will be useful for non-computer-literates to use it
12:23:03salty-horseI notice there's no indentation between the first item (title) and the rest of the items
12:23:16salty-horse(the english layout has a few pixels of offset)
12:25:01tomerssalty-horse: This is not a bug. It relates to the position of the scrollbar. If you set it to the left - you'll see this offset. If you switch to English and set the scrollbar to the right, you'll see the offset is gone.
12:25:18salty-horseok :)
12:26:17tomersInitially, I thought the scrollbar setting should be normal/opposite rather than left/right, so that for RTL languages normal=right, but other developers prefered to keep it left/right
12:30:22salty-horsewhat setting controls it?
12:32:04tomerssalty-horse: Settings -> Theme Settings -> Status-/Scrollbar -> Scroll Bar
12:32:23salty-horsethere's also display > scrolling. confusing?
12:32:33salty-horsegeneral > display > scrolling
12:32:55tomersScrolling means scrolling text lines which are longer than screen width...
12:33:23tomersI dislike the lack of convention in using both Scrollbar and Scroll Bar
12:34:03salty-horseI agree. the "-/" is also odd
12:35:16tomersWikipedia uses the term 'Scrollbar' (
12:35:27salty-horseare you interested in adding LRM and RLM unicode mark support for the bidi algorithm? I have an old patch lying around (it also changes heb/eng variable names to ltr/rtl to be more neutral towards other RTL languages)
12:37:08tomersAlthough I read a book about Unicode once, I need to refresh my knowledge about these marks. What benefit would we get from it?
12:38:13salty-horsehebrew song titles ending with exclamation marks could be displayed correctly :)
12:38:34tomerssalty-horse: We need to provide more supoprt for Hebrew characters in RB fonts. See
12:38:44salty-horseoh I have that
12:38:49salty-horseI'm using a dejavu font
12:38:55salty-horseI have a pseudopatch for it
12:39:13tomerssalty-horse: That would be great. Do you have a FlySpray item for that?
12:39:26salty-horsein the tracker, but apparently I gave a post-processed font instead of a bdf. I need to rework it. sec
12:41:00salty-horselet me see if I remember how to generate them. I used two ttf2bdf tools and one of them had nicer output
12:44:14tomerssalty-horse: I can't take responsibility for this right now, due to my lack of available time and also since I'm not that familiar with fonts and the considerations needed to be taken specifically for RB. It looks like you did a good job there, and this is already a work in progress. Maybe you should push it forward more, and it will be committed eventually.
12:44:48salty-horseI'll try to do it now. I'm not asking for you to take any responsibility, but feel fry to try those fonts :)
12:48:13tomerssalty-horse: is some other important patch related to this
12:52:42 Join martian67 [0] (i=lol3@about/linux/regular/martian67)
12:53:29 Quit pamaury (Read error: 113 (No route to host))
12:54:51 Quit parafin (Read error: 60 (Operation timed out))
12:55:04 Join Llorean [0] (
12:57:28 Join parafin [0] (
12:57:57 Quit tomers ("CGI:IRC")
13:01:40 Quit DerPapst ("Leaving.")
13:05:48 Join ifonefox [0] (
13:05:53 Nick ifonefox is now known as CATS (
13:05:55 Quit CATS (Remote closed the connection)
13:06:09 Join ifonefox [0] (
13:06:19 Nick ifonefox is now known as CATS (
13:08:07 Quit CATS (Remote closed the connection)
13:09:18 Quit seani ("Leaving")
13:14:59 Quit MethoS- (Remote closed the connection)
13:15:50***Saving seen data "./dancer.seen"
13:20:35 Join Zagor [242] (n=bjorn@rockbox/developer/Zagor)
13:26:36 Join Grahack [0] (
13:27:42 Part Grahack
13:27:43 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
13:38:23 Quit kugel (Read error: 110 (Connection timed out))
13:50:04 Join teru [0] (n=teru@
13:54:26 Quit evilnick (Read error: 104 (Connection reset by peer))
13:54:34 Join evilnick [0] (
13:56:23 Part LinusN
13:57:40 Join hebz0rl [0] (
14:02:30 Quit ChanServ (
14:04:36NJoinChanServ [0] (ChanServ@services.)
14:04:36Mode"#rockbox +o ChanServ " by
14:10:00 Quit ChanServ (
14:13:01 Join funman [0] (
14:14:28 Nick funman is now known as Guest1753 (
14:14:41NJoinChanServ [0] (ChanServ@services.)
14:14:41Mode"#rockbox +o ChanServ " by
14:15:12 Quit ChanServ (
14:17:12 Nick Guest1753 is now known as funman_ (
14:18:22NJoinChanServ [0] (ChanServ@services.)
14:18:22Mode"#rockbox +o ChanServ " by
14:18:25 Nick funman_ is now known as funm4n (
14:22:38 Join blablabla [0] (
14:22:41 Quit blablabla (Read error: 104 (Connection reset by peer))
14:22:50 Join blablabla [0] (
14:22:53 Quit blablabla (Read error: 104 (Connection reset by peer))
14:23:01 Join blablabla [0] (
14:23:14 Quit blablabla (Read error: 54 (Connection reset by peer))
14:23:22 Join blablabla [0] (
14:24:03 Quit blablabla (Read error: 104 (Connection reset by peer))
14:26:13 Join blablabla [0] (
14:26:51 Quit blablabla (Read error: 104 (Connection reset by peer))
14:33:55 Join kush|adium [0] (n=Adium@
14:43:03 Join seani [0] (n=seani@
14:59:15 Join Lss [0] (
15:04:44 Join Horscht [0] (n=Horscht2@xbmc/user/horscht)
15:09:56 Quit FlynDice (Remote closed the connection)
15:15:54***Saving seen data "./dancer.seen"
15:30:23 Join evilnick_B [0] (i=0c140464@rockbox/staff/evilnick)
15:32:54 Quit Res1 (Read error: 60 (Operation timed out))
15:33:55 Join Res1 [0] (
15:44:25 Quit ChanServ (
15:44:25 Join Strife89 [0] (n=michael@
15:45:12Strife89I haven't seen a battery bench for the Fuze on the Wiki yet, so I'm running one. JSYK.
15:45:40NJoinChanServ [0] (ChanServ@services.)
15:45:40Mode"#rockbox +o ChanServ " by
15:46:59funm4nStrife89: nice, did you measure the OF battery life ?
15:47:06Strife89funm4n: Not yet.
15:48:32Strife89Using VBR ogg vorbis files, ~<192 kbps, -30dB.
15:48:50 Join DerPapst [0] (
15:50:05Strife89It's a used Fuze, but if I'm lucky, I suspect it won't be dead until I turn in tonight.
15:55:19 Quit kush|adium ("Leaving.")
15:55:31 Quit ChanServ (
15:57:14 Quit jordan`` ("Coyote finally caught me")
15:57:33 Join jordan` [0] (n=jordan@
16:01:08NJoinChanServ [0] (ChanServ@services.)
16:01:08Mode"#rockbox +o ChanServ " by
16:01:51 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
16:04:22 Join AEnima1577 [0] (
16:06:04 Quit tchan (Read error: 60 (Operation timed out))
16:10:43 Join tchan [0] (n=tchan@lunar-linux/developer/tchan)
16:10:59 Join Jaykay [0] (
16:12:09 Quit teru ("Quit")
16:16:59 Part LinusN
16:17:25 Quit linuxguy3 (Remote closed the connection)
16:17:36 Join linuxguy3 [0] (
16:31:22 Join kush|adium [0] (n=Adium@
16:33:08 Part kush|adium
16:35:43 Quit Strife89 (Read error: 60 (Operation timed out))
16:37:25 Quit antil33t (Read error: 104 (Connection reset by peer))
16:37:31 Join antil33t [0] (
16:37:43 Join Tomis2 [0] (n=Tomis@
16:39:05 Join mcuelenaere [0] (
16:40:51 Join toffe82 [0] (
16:45:05 Quit Tomis (Success)
16:45:05 Nick Tomis2 is now known as Tomis (n=Tomis@
16:45:33 Join Strife89 [0] (n=michael@
16:49:19 Quit funm4n ("free(random());")
16:50:43 Join kugel [0] (
16:53:43 Join kugel__ [0] (
16:53:51 Quit kugel (Nick collision from services.)
16:53:57 Nick kugel__ is now known as kugel (
16:57:48Strife89Could someone add a Fuze section to the chart in ? I tried doing it myself, but I chickened out each time because I'm still not used to the new editor, and my input consistently appeared to be wrong.
16:58:56ZagorStrife89: you can switch back to the "old" editor
16:58:57Strife89And a row for results, please. I'm running a benchmark now; should be done by 11 PM EST.
16:59:13Strife89Zagor: Ah, never saw that option.
16:59:22Zagorit's one of the buttons
17:00:06kugelZagor: I seem to be unable to close tasks
17:01:10kugeloh wait, it works again...or maybe it was a google chrome issue?
17:01:31 Quit AEnima1577 ("Leaving.")
17:07:46 Join AEnima1577 [0] (
17:08:10 Quit evilnick (Read error: 113 (No route to host))
17:08:13 Join evilnick [0] (
17:11:43kugelcan anyone explain the difference between bufread and bufgetdata?
17:15:56***Saving seen data "./dancer.seen"
17:16:26 Join kugel__ [0] (
17:16:38 Quit kugel (Nick collision from services.)
17:16:46 Nick kugel__ is now known as kugel (
17:22:19 Join Omlet [0] (i=omlet05@
17:26:34 Part Bagder
17:32:04 Join TheSeven [0] (n=theseven@rockbox/developer/TheSeven)
17:34:20 Join mikroflops [0] (n=yogurt@
17:35:01 Join brert [0] (
17:35:29 Join rbert [0] (
17:35:38 Quit rbert (Client Quit)
17:38:29 Join rbert [0] (
17:38:32 Join funman [0] (n=fun@rockbox/developer/funman)
17:39:13*Unhelpful needs to hack mono support into convttf :/
17:39:26 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
17:39:44Unhelpfulwould convttf be considered committable with mono support added and AA-font support disabled until a decision is reached on the feature?
17:40:02 Quit brert ("CGI:IRC (Ping timeout)")
17:40:19 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
17:40:41 Quit Zagor ("Don't panic")
17:44:59kugelUnhelpful: why do you need that?
17:48:08CIA-80New commit by moos (r23650): Updated french translation.
17:54:27 Join |DaMaGeD| [0] (n=|DaMaGeD@
17:58:14 Quit Strife89 ("Leaving class.")
18:01:15 Quit bluebrother (Nick collision from services.)
18:01:17 Join bluebroth3r [0] (n=dom@rockbox/developer/bluebrother)
18:03:11 Quit Jaykay (Read error: 110 (Connection timed out))
18:04:32 Quit petur ("work->home")
18:04:48 Join freddyb [0] (
18:10:43 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
18:11:31JdGordonkugel: have you seen the latest remove statusbar patch? any ideas why when enabling/disabling the bar all menus except the root menu seems to change possitions correctly?
18:15:20 Join lyngaas [0] (
18:16:32 Quit rbert ("CGI:IRC (EOF)")
18:19:19 Quit lyngaas1 (Read error: 110 (Connection timed out))
18:20:41 Quit Tomis (Read error: 104 (Connection reset by peer))
18:27:54 Join pamaury [0] (
18:29:15 Join Tomis [0] (n=Tomis@
18:31:12 Join JdGordon| [0] (
18:32:01 Quit phanboy4 (Read error: 104 (Connection reset by peer))
18:32:02 Quit maruk ("Leaving.")
18:32:06JdGordon|kugel: nope to which? or both :)
18:33:09 Quit Tomis (Read error: 104 (Connection reset by peer))
18:33:11JdGordon|its damn wierd... if you load a sbs then the themes menu will reposition correctly (which mean list_init_something is called.. but then when you keep going back up to the root it uses the lists from before
18:33:31JdGordon|but that doesnt make sense before do_menu() is renetered up there so it should get a new list
18:33:39JdGordon|so... im stumped
18:38:51 Join n1s [0] (n=n1s@rockbox/developer/n1s)
18:41:52 Join Tomis [0] (n=Tomis@
18:44:21 Join Tomis2 [0] (n=Tomis@
18:49:54JdGordon|is anyone actually going to change statusbars without rebooting? I mean really? :p
18:50:21JdGordon|... aslso, how do i fix the bitmaps so the same ones are drawn on 16bit lcd's transparent like on mono displays?
18:50:45JdGordon|the images in svn are 8 bit apparently even though I saved them as mono
18:51:01 Join Grahack [0] (
18:56:21 Join bertrik [0] (
18:58:32pixelmaJdGordon|: those seem to be in a weird format, some of my programs couldn't open them (including Windows' preview). Got it open in one though and the program thinks those are RGB - and I can see some pixels that aren't fully black. How did you generate those?
18:59:14JdGordon|I might have to remake them :(
18:59:20pixelmaAlso the battery one at least wastes a row of pixels per subimage
19:00:04pixelmanah, if you give me some time I take care of them - got to do some shopping first though (shops close in one hour)
19:00:40 Quit flydutch ("/* empty */")
19:00:57 Quit Tomis (Read error: 110 (Connection timed out))
19:00:57 Nick Tomis2 is now known as Tomis (n=Tomis@
19:01:50JdGordon|I'm at work for the next 7 hours... so yeah if you could fix them that would be great :)
19:02:03pixelmaabout the battery one - there doesn't seem to be a fully empty battery and no "unknown" state which I actually see currently during boot
19:02:35JdGordon|I got lazy... feel free to make them better :)
19:06:11 Quit stoffel (Read error: 104 (Connection reset by peer))
19:07:38 Quit JdGordon| ("Miranda IM! Smaller, Faster, Easier.")
19:08:15pixelmathe rec bitmaps also need the quality ones for hwcodec... ;)
19:09:26 Quit hebz0rl ()
19:10:41 Quit pixelma (Nick collision from services.)
19:10:41 Join pixelma_ [0] (i=quassel@rockbox/staff/pixelma)
19:10:50 Quit kugel (Read error: 104 (Connection reset by peer))
19:11:01 Nick pixelma_ is now known as pixelma (i=quassel@rockbox/staff/pixelma)
19:11:18 Quit amiconn (Nick collision from services.)
19:11:21 Join amiconn_ [0] (i=quassel@rockbox/developer/amiconn)
19:11:41 Nick amiconn_ is now known as amiconn (i=quassel@rockbox/developer/amiconn)
19:12:37 Quit freddyb (Read error: 113 (No route to host))
19:14:48 Join JdGordon| [0] (n=Miranda@nat/microsoft/x-qgtehzmwtytvletw)
19:14:58 Join panni_ [0] (
19:15:07 Join kugel [0] (n=kugel@rockbox/developer/kugel)
19:15:33 Join fredb [0] (
19:16:00***Saving seen data "./dancer.seen"
19:21:43 Join kugel__ [0] (
19:21:50 Quit kugel (Nick collision from services.)
19:21:57 Nick kugel__ is now known as kugel (
19:27:14 Join fir3 [0] (
19:29:51fir3i'm looking for a cheap player with ~4gb and long battery lifetime
19:30:05salty-horsesansa e240?
19:30:58fir3what do you think about this one? or would you recommend other players?
19:31:19Unhelpfulwell, it won't run rockbox
19:31:59Unhelpfulkugel: i mean, would it be considered committable *before* a consensus is reached on AAF if it could also generate "normal" fonts.
19:32:23Unhelpfulit would do much to streamline the process of using ttf fonts for rockbox :)
19:32:26fir3i'd be ok with that, only very fewe players i like run rockbox :/
19:32:38salty-horsefir3, then why ask here? :)
19:32:44salty-horseeveryone here is biased
19:32:53kugelUnhelpful: but you don't *need* to hack in mono support ?
19:33:08fir3where would be a good place to ask?
19:33:39Unhelpfulkugel: don't really need to, no. but it would be useful to more people sooner with mono font support
19:34:01 Quit DerPapst ("Leaving.")
19:35:11salty-horsefir3, google, blogs, reviews
19:35:20LloreanUnhelpful: I think a one-step ttf->font tool in SVN would be handy if there's not one.
19:35:44salty-horseLlorean, there is:
19:36:10Lloreansalty-horse: Is that in SVN?
19:36:17LloreanAnd, bdf is not fnt.
19:36:23LloreanSo that's hardly one step.
19:36:36salty-horseI thought that's what you meant. the tool could just glue the two
19:36:49kugelUnhelpful: how's your AAF progress?
19:37:15salty-horseLlorean, this + tools/convbdf will do what you want
19:37:18 Quit Tomis (Read error: 104 (Connection reset by peer))
19:37:46Unhelpfulkugel: i've not done more than a math test. i'm confident that 2-bit subpixel AA fonts can be blended with only 3 multiplies, and simulated rendering of such look good
19:37:50salty-horsealthuogh the otf2bdf tool isn't very accurate with font sizes. if you ask a size 12pt it may give you something with 16pt
19:38:07 Quit liar (Read error: 113 (No route to host))
19:38:11fredbfir3: did a recording test on my Fuze last night and it made it to 10 hours before the battery gave up. I bought it for $40 (8GB). You need one with the 1.0x firmware, tho.
19:38:35*kugel whishes he would understand what subpixel means exactly :/
19:38:47Unhelpfulsalty-horse: convttf also does rather well in that area. it can be off by one, since some pixel sizes aren't really acheivable due to hinting.
19:39:05 Join efyx_ [0] (
19:39:43fir3fredb: where did you get it for 40$? the cheapest one i can find is 70€ (germany) :/
19:39:45Unhelpfulalso because fonts are measured up and down from a baseline, and sometimes the same increment in outline scaling will make the bitmaps one pixel larger on the top *and* the bottom
19:39:53 Join Tomis [0] (n=Tomis@
19:40:50fredbfir3: used on craigs list. I also have a on old iAudio X5 that was great but I don't know if you could find one anymore...
19:40:56Unhelpfulkugel: if you want to get a good idea what subpixel does, look at on your e200, then open it in an image editor and zoom in close.
19:41:34salty-horseUnhelpful, where's convttf?
19:41:55kugelUnhelpful: is that 2bit?
19:42:41salty-horsekugel, stating the obvious:
19:42:55Unhelpfulkugel: yes.
19:44:15kugelit looks ok
19:44:37kugelthe colorful pixels are quite noticable to me though
19:45:17LloreanOn the actual player, or on your monitor?
19:45:23 Join Strife89 [0] (n=michael@
19:45:44LloreanThen don't use AA fonts, I guess.
19:46:37salty-horseUnhelpful, got a link to convttf please? :)
19:46:46kugelit looks better than non-AAF still
19:51:08 Join DerPapst [0] (
19:51:31 Join liar [0] (n=liar@
19:52:26 Join stripwax [0] (
19:52:45 Quit stripwax (Client Quit)
19:58:28 Quit fredb ("Konversation terminated!")
20:01:36 Join einhirn [0] (
20:11:05 Join Utchybann [0] (
20:16:48 Quit bluebroth3r (Read error: 131 (Connection reset by peer))
20:17:01 Join bluebrother [0] (
20:19:49 Join Jaykay [0] (
20:22:15 Join domonoky1 [0] (
20:24:05 Join phanboy4 [0] (
20:27:51 Quit phanboy4 (Read error: 104 (Connection reset by peer))
20:28:16 Join phanboy4 [0] (
20:32:24 Quit bluebrother (Read error: 54 (Connection reset by peer))
20:32:32Unhelpfulkugel: looks great on mine. maybe yours has a BGR screen? or vertical pixels?
20:32:44 Join bluebrother [0] (
20:33:02kugelmy e200?
20:34:04Torneare the error codes returned by the ATA driver supposed to mean anything in particular, or ar they jus supposed to be identifiable to someone reading the source? :)
20:35:07kugelthe latter, unless you got the dreadful ATA ERROR -666
20:35:35JdGordon|ESATAN ?
20:36:43Tornei'm looking at unifying read/write in there, and there's a few differences in how they work atm
20:36:58Torneand tey return crazy different error codes (and the write code tends to panic rather than returning errors)
20:37:37JdGordon|arnt the ata codes supposed to be passed around so you can actually work out here -3 comes from and where -9 is?
20:37:47 Quit leafwiz (Read error: 113 (No route to host))
20:37:51Tornewell, i expect so, yes :0
20:38:00Tornebut ata_{read,write}_sectors are pretty much the originator of those
20:39:00Torneis anyone likely to be unhappy if i make writing more forgiving? :)
20:39:27Torneatm it doesn't retry failures, and if you try to write past the end of the disk it panics, and various other fun.
20:40:25 Quit domonoky (Read error: 110 (Connection timed out))
20:41:16*domonoky1 would like it some safety features stay with panics, then we know when we get problem..
20:41:38domonoky1retrying is fine.
20:41:54Tornethe thing is that the fat code is going to panic anyway when ATA returns an error, pretty much ;)
20:42:16Torneand if you want to panic for doign stupid things (like accessing off the end of the disk) then why doesn't read do it?
20:42:37*Torne would be happy either way tbh but doesn't see why they *should* be different :)
20:43:01 Join einhirn_ [0] (
20:43:35*domonoky1 thinks writing over end of disk should be fine with error code. but for example writing to the OF part, should be protected by panic
20:44:08gevaertsdomonoky1: in the ATA driver?
20:44:18gevaertsWriting to the OF part is needed there
20:44:39Tornealso also, why don't we use WRITE MULTIPLE?
20:44:59*domonoky1 is ab it confused, which player do we tak about ? :-)
20:45:07Tornedomonoky1: all players with ATA
20:45:09gevaertsdomonoky1: players with ATA :)
20:45:13Tornei'm talking about the generic ata driver.
20:45:29domonoky1ah.. then ignore me :-)
20:45:48 Quit |DaMaGeD| ("When two people dream the same dream, it ceases to be an illusion. KVIrc 3.4.3 Shiny(svn-3438)")
20:45:52*domonoky1 though about things like ata_sd_as3525.c
20:45:58Torneno, not those
20:46:02 Join fredb [0] (
20:47:11TorneI'm lkooking at fixing FS #9721 - we don't check for write errors properly. But not checking leads to panics, because writes dont' retry (i have experienced this)
20:47:21Torneer, but checking leads to panics, even
20:47:48Torneso unifying the two so that reads/writes are handled the same and do retries and so on seems like a good idea, as it will be safer and also probably save binsize
20:52:19JdGordon|just as long as it doesnt go into an endless retry loop...
20:52:46Torneit only tries, er, five times or something in the read :0
20:52:47Torneso it would be the same
20:53:20Torneit does a soft reset between, and assumes that the error was probably due to drive being shaken ;)
20:53:31Tornewhich is probably true. I'm prettysure that's how i've managed to make it happen :)
20:55:15 Quit einhirn (Read error: 110 (Connection timed out))
21:00:14 Nick fxb__ is now known as fxb (
21:05:57 Join FlynDice [0] (
21:07:22 Join Blue_Dude [0] (n=chatzill@rockbox/developer/Blue-Dude)
21:07:58Blue_DudeWhy exactly is mpeg.c called that? Shouldn't it be hwcodec.c or something?
21:09:49CIA-80New commit by blue_dude (r23651): Cleanup audio.h, related functions
21:09:55pixelmabecause you can only play mp3/mp2 with it?
21:10:18pixelma(wav is possible but currently only in a plugin)
21:10:33Blue_DudeYeah, but it's only used for hwcodec targets. It's the hwcodec version of playback and codecs.
21:11:15JdGordon|filenames arnt something you should assume to make sense...
21:11:30Blue_DudeI keep confusing it with the mpegplayer plugin. I have to keep reminding myself what its for...
21:11:50pixelmaand the current supported codec only does mp3 and mp2 (and maybe that's kept for historic reasons)
21:12:22Blue_DudeI"m considering renaming the thing.
21:14:04 Quit fir3 (Remote closed the connection)
21:16:04***Saving seen data "./dancer.seen"
21:28:10Unhelpfulkugel: well, the test doesn't really suffer from color fringing on mine..
21:33:35kugelUnhelpful: do you think they used different lcd types?
21:33:44 Join petur [0] (
21:35:27kugelmaybe it's just my eyes
21:39:15 Join gb_master [0] (i=82c03212@gateway/web/freenode/x-qemiervfoqhatgov)
21:47:36 Quit AEnima1577 ("Leaving.")
21:52:59 Join AEnima1577 [0] (
21:54:13 Quit phanboy4 ("Leaving")
21:55:13 Quit Grahack ("Leaving.")
21:57:22 Quit gb_master ("Page closed")
21:58:35 Quit kugel (Read error: 104 (Connection reset by peer))
22:01:51 Quit fredb (Remote closed the connection)
22:02:01 Join kugel [0] (n=kugel@rockbox/developer/kugel)
22:07:19 Quit Jaykay ("ChatZilla 0.9.85 [Firefox 3.5.5/20091102152451]")
22:11:32 Quit einhirn_ (Read error: 101 (Network is unreachable))
22:12:25pixelmaJdGordon|: any reason why your battery and volume icons in the classic statusbar sbs has less steps than the real classic one?
22:13:06JdGordon|because I'm lazy... I thought 10 steps would be fine...
22:13:19JdGordon|I tihnk also because I thought 10 steps were the max in a strip
22:13:22JdGordon|but thats wrong
22:13:33gevaertsit definitely is. I've used 52
22:14:37*gevaerts wanted a smooth reverse progress bar!
22:14:59kugeldid that work?
22:15:59 Join phanboy4 [0] (
22:16:19gevaertsIt's half a year old by now, so maybe there are now better ways to do some of the things, but yes, it works
22:16:45gevaertsit's for 240x320 if you want to have a look at it
22:18:06gevaertsoh, I used only 50 per strip I think, not 52
22:18:46*kugel shouldn't accapt themes for targets he doesn't own without, except it comes with a written promise to receive that target too in the near future
22:19:34gevaertsdon't expect this one to actually look nice :)
22:19:46gevaertsit's purely a technology demo
22:21:27pamaurygevaerts: hello, did you have a look at those M$ descriptors ?
22:21:45gevaertsum, no, not yet
22:22:12Unhelpfulkugel: it's *possible*.
22:22:43 Join hillshum [0] (
22:26:00 Quit Omlet ("( :: NoNameScript 4.22 :: )")
22:26:38Utchybannperhaps a stupid question but why nano2g does not use WHEEL_ACCELERATION ?
22:28:27linuxstbUtchybann: I tried enabling it once, but things went horribly slow. I didn't investigate much further.
22:30:27kugelgevaerts: it does work, it just wasn't what I thought a reversed progressbar would be :)
22:30:55gevaertskugel: what did you think? RTL?
22:31:07gevaertsYou can use the same technique for that
22:31:18Utchybannlinuxstb: ok. still related to the wheel. I found that on the nano2g, just touching the wheel to turn backlight on generally change the volume by 1 (in the wps).
22:31:45Tomisyea it's overly sensitive Utchybann
22:31:49Tomisi noticed that too
22:32:10*gevaerts recommends setting the appropriate setting for this
22:32:51Utchybannlinuxstb: the scrolling speed is good, just too sensitive when just 'touching' the wheel.
22:34:24 Join mirak [0] (
22:34:34Utchybannlinuxstb: I have made some experiment with WHEEL_SENSITIVITY with no luck (it just give me slower scrolling speed).
22:34:59kugelI hope to get my hands on a nano2g again to look at the wheel
22:35:44kugelI think I identified the problem that makes the generic wheel code work so poorly
22:38:44Utchybannkugel: if you have code to test... my nano is ready.
22:39:04Blue_DudeWhat is change_dir in wps.c used for? Since it's only useful for swcodec, can we make it so that it behaves correctly for hwcodec?
22:39:35 Join shotofadds [0] (n=rob@rockbox/developer/shotofadds)
22:40:20 Join LambdaCalculus37 [0] (n=LambdaCa@rockbox/staff/LambdaCalculus37)
22:40:23 Quit kugel (Read error: 60 (Operation timed out))
22:41:17Blue_DudeI've got this funky define in playback.h that nullifies two functions in that routine, but the define doesn't really belong in playback.h. It ought to be in wps.c.
22:43:43gevaertspamaury: if I understand these things correctly, we'd mainly need the CompatID things?
22:44:07 Quit funman ("free(random());")
22:44:22gevaertsand maybe an icon to be fancy :)
22:45:08*shotofadds has been sent an SD card that exhibits a weird problem on his D2
22:45:28pamaurygevaerts: I'm not sure about everything but yes, we NEED the CompatID thing. The thing I'm less sure about is the extended property things. I think there are needed but libmtp at least can work without (but prints a warning) [if I read the code correctly]
22:45:37shotofaddsThe bootloader is able to power up the card and load rockbox.d2, but the card then fails to power up when it's re-initialised by the main binary. It powers up fine when booted via tcctool, and also when re-initialised via ROLO.
22:45:51shotofaddsHas anyone seen something like this before? I'm stumped :/
22:47:10gevaertspamaury: an extended property descriptor with wCount=0 shouldn't be too hard to do I guess
22:47:27gevaertsshotofadds: that reminds me of some of the things I've heard about MMC cards
22:47:45gevaertsnot exactly the same though
22:48:29shotofaddsI wonder why loading the binary via ROLO works, but not the bootloader? maybe the longer delay makes a difference
22:48:51gevaertssounds likely
22:49:08pamaurygevaerts: I'm not sure about the extended property thing. Let me look at the libmtp code once more. Theorically, it should not be necessary but if it's implemented then we should implement one property (at least) I think
22:49:36gevaertspamaury: well, we can do the label without too much overhead
22:50:09 Join stripwax [0] (
22:50:26Blue_DudeMoved the funky define to audio.h...
22:50:38*shotofadds doesn't think adding a 10sec delay to the start of main.c will go down well with users ;-)
22:51:03JdGordon|just put it in a nice loading splash and they wont care
22:51:23shotofaddswe could animate it and everything!
22:51:26Unhelpfulplay a little tune and a video
22:51:29gevaertsshotofadds: if it fails, does retrying after a delay work?
22:52:01shotofaddsgevaerts: currently it re-tries in a loop, but with a very very short delay in between
22:52:32pamaurygevaerts: the thing I curently don't understand is why libmtp look at the extended properties. It look at CompatIDs because it's written and well defined but I've not find anything about MTP and extended properties
22:53:06gevaertspamaury: my guess is "because windows does it"
22:53:24gevaertsor to get this label? Is that used somewhere?
22:53:37shotofaddsalso I guess it should send it back to the idle state before re-trying
22:54:08 Quit petur ("Zzzz")
22:54:29pamaurygevaerts: no it just checks 3 bytes values 'MTP' but it doesn't matter what the rest is, what format is uses. It's a pure hack
22:54:41 Quit AEnima1577 ("Leaving.")
22:54:47gevaertsfun :(
22:54:56gevaertsStill, there's not much we can do about it
22:55:09pamauryThere is a comment in libmtp code the the link is dead:
22:55:16pamaury * This checks if a device has an MTP descriptor. The descriptor was
22:55:21pamaury * elaborated about in gPhoto bug 1482084, and some official documentation
22:55:25pamaury * with no strings attached was published by Microsoft at
22:55:29pamaury *
22:55:41Blue_DudeDoes it matter that change dir doesn't function correctly for hw codec?
22:57:09 Quit Sajber^ ("Leaving.")
22:57:35*pamaury is reading gPhoto 1482084 bug report
22:59:42 Quit AndyIL ()
23:00:17hillshumpamaury: Use a pastebin next time?
23:00:19 Join AEnima1577 [0] (
23:00:25pamauryyes :)
23:00:27 Join salty_horse [0] (
23:01:53pamaurygevaerts: the gPhoto bug reports is not very instructive, the only useful thing is the link but it's dead :(
23:02:09CIA-80New commit by blue_dude (r23652): mpeg.h/c cleanup
23:02:19gevaertspamaury: you could try asking in #gphoto if they know more
23:06:51pamaurygevaerts: yes perhaps... I tried on my sansa e200 and now that I understand fully the messages, I can say that the OF does not implement extended properties. Perhaps they just dumped several devices, saw that some implemented this label thing and implemented it in libmtp
23:07:20gevaertsah, ok
23:08:11pamauryI think the best way to know is implement compat IDs, check with windows and then conclude :)
23:08:36gevaertspamaury: one thing, the reason for the usb_core_ack_control() changes was that some controllers need the usb_drv_recv() before the corresponding usb_drv_send() to avoid races (they don't NAK properly in that case apparently)
23:08:59gevaertsWe tested on all targets, and doing the usb_drv_recv() first is safe
23:10:26 Join AndyI [0] (n=pasha_in@
23:10:45pamaurySo the way it's now implemented is correct ? (I don't know if you checked the code on github).
23:11:33gevaertsI checked, yes. The way you did it will break on 2nd gen nano
23:11:45 Join bluefoxx___ [0] (
23:12:07pamauryFor now it's implemented by "usb_drv_recv(EP_CONTROL, NULL, 0)". Isn't it the correct way you describe ?
23:12:17bluefoxx___Can someone Please tell me if there's a bootloader for the sansa E200v1 that completely bypasses the stock firmware, altogether
23:12:36 Quit n1s ("Lmnar")
23:12:49gevaertsyes, but it needs to be done *before* the usb_drv_send() (if there is one)
23:12:50bluefoxx___I'm sick of plugging my sansa in while it's off only to find it's gone into the stock f/w and crashed because it can't handle the idea of a SD card larger than 2 Gb
23:12:59pamauryAh !!!
23:13:19gevaertsbluefoxx___: yes, the latest released bootloader does that
23:13:21bluefoxx___Or having my battery run dead and plugging it in to find the same happening
23:13:24pamauryThen you're right, it will break on some tarhets :)
23:13:31bluefoxx___Link please?
23:13:36pamaury* targets
23:14:08 Join Thundercloud [0] (
23:14:11pamaurySo I should ack just after receiving the packet and before sending anything.
23:14:39gevaertsyes. Well, tell it to ack, it's still host-initiated :)
23:14:59gevaertsbluefoxx___: the manual should have links, or just download a recent rbutil
23:15:14pamauryyes :)
23:15:21bluefoxx___I prefer to manually flash
23:15:46bluefoxx___Fiddled with the rbutil once, decided I don't want another doeverythingforme app
23:15:46gevaertsbluefoxx___: yes, see the manual for the download url
23:16:05***Saving seen data "./dancer.seen"
23:16:25*bluefoxx___ searches his firefox history for the link
23:16:55Blue_DudeWhat's going on with the build?
23:17:36 Quit salty-horse (Read error: 110 (Connection timed out))
23:17:42pamaurymc2739: it's great that you find it but actually I already read this faq (without noticing it's was the faq I searched...) and I learn nothing those extended propertis. But thanks you for your research
23:17:46 Quit domonoky1 (Read error: 104 (Connection reset by peer))
23:17:51gevaertsBlue_Dude: "Server connection stalled. Exiting in 5s" doesn't sound good
23:18:06 Quit bmbl ("Bye!")
23:20:02 Join flydutch [0] (
23:20:41bluefoxx___Oh cool. The stock f/w crashed my USB bus. Bye-bye cardreader and external drives I was backing up to.
23:21:04*bluefoxx___ makes a noise indicating his disgust at the stock F/W for the sansa and reboots
23:21:11 Quit bluefoxx___ ()
23:22:25Strife89Usually it's Windows' fault for that. :/
23:27:16 Join Stephen_ [0] (
23:30:07pixelmaBlue_Dude: how "safe" are your changes to hwcodec playback and recording? I'm just curious because I know it works quite a bit different (especially buffering and getting tag data) but I don't understand what you actually did there except some code shuffeling
23:30:21 Quit Strife89 ("See ya.")
23:32:07Blue_DudeI cleaned out mpeg.h because most all the info there was either only referenced by mpeg.c (the defines) or not referenced at all (the functions).
23:32:46Blue_DudeThe functions were only prototypes that weren't called and didn't have matching code anywhere.
23:32:55 Quit toffe82 (Remote closed the connection)
23:34:22Blue_DudeMost of the rest of the changes was removing mpeg.h includes.
23:34:48 Join fdinel [0] (
23:40:14 Quit AaronM ("Emo Time In My Corner... //_-")
23:42:47 Quit phanboy4 (Read error: 104 (Connection reset by peer))
23:43:10 Join AaronM [0] (
23:46:16saratogaif I have a data abort address in a static function, is there anyway to figure out which function it is?
23:46:25saratogawithout recompiling and trying to reproduce the bug
23:46:41 Join Sajber^ [0] (
23:47:12gevaertssaratoga: do you have the .elf file?
23:47:26saratogai do
23:48:04gevaertsin that case arm-elf-objdump −−syms rockbox.elf|sort
23:49:56saratogaok that gives me good agreement with dumping buffering.o
23:50:05Blue_DudeI have to go. I hope the build is green... Back later tonight.
23:50:09 Quit Blue_Dude ("ChatZilla 0.9.85 [Firefox 3.5.5/20091102152451]")
23:50:37saratogaa data abort is probably an attempt to dereference a null pointer?
23:50:56saratogaor at least something thats not a valid pointer
23:51:08 Quit evilnick_B ("Page closed")
23:51:21gevaertson arm it's often an unaligned pointer
23:53:20 Quit pamaury ("exit(*(int *)0 / 0);")
23:57:52 Quit stripwax ("")

Previous day | Next day