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 2009-05-24

00:03:42 Join {phoenix} [0] (
00:04:26 Quit andrewbeveridge (" all hail rockbox development!")
00:06:57 Quit amiconn (Nick collision from services.)
00:06:59 Join amiconn_ [50] (n=jens@rockbox/developer/amiconn)
00:07:19 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
00:08:09 Quit pixelma (Nick collision from services.)
00:08:11 Join pixelma_ [50] (n=pixelma@rockbox/staff/pixelma)
00:08:29 Nick pixelma_ is now known as pixelma (n=pixelma@rockbox/staff/pixelma)
00:09:07 Quit trisiak (Read error: 104 (Connection reset by peer))
00:09:12 Join trisiak [0] (
00:14:32 Quit faemir (Read error: 110 (Connection timed out))
00:14:42 Quit alexbobp (Read error: 104 (Connection reset by peer))
00:24:12 Join alexbobp [0] (
00:31:27 Quit {phoenix} (Remote closed the connection)
00:41:31 Join matsl [0] (
00:43:34 Quit ender` (" Information travels more surely to those with a lesser need to know.")
00:48:35 Quit matsl (Remote closed the connection)
00:49:04 Quit tessarakt ("Client exiting")
00:49:23 Quit n1s ("Lmnar")
00:50:21 Join fdinel [0] (
00:51:36 Join matsl [0] (
00:54:50 Join matsl_ [0] (
01:01:48 Join HellDragon [0] (
01:11:10 Quit matsl (Read error: 110 (Connection timed out))
01:20:43 Quit HellDragon (Client Quit)
01:33:47 Join HellDragon [0] (
01:34:48***Saving seen data "./dancer.seen"
01:39:14 Quit evilnick_home (Read error: 104 (Connection reset by peer))
01:46:33 Join bagawk_ [0] (
01:49:02 Join renke [0] (
01:54:19 Quit BXCracer (Remote closed the connection)
01:59:39 Quit bagawk (Read error: 110 (Connection timed out))
02:05:10 Quit xnyhps ("Leaving.")
02:05:24 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
02:06:28 Quit matsl_ (Read error: 110 (Connection timed out))
02:08:45 Quit domonoky (Read error: 104 (Connection reset by peer))
02:12:17 Join Xerion_ [0] (
02:27:00 Join lee321987 [0] (
02:29:17lee321987My DAP keeps disconnecting from my computer randomly. Is it possible to copy the contents of a new RB zip file to the external SD card, and then overwrite the .rockbox folder while RB is running? (Sansa c200v1)
02:29:17 Quit Xerion (Connection timed out)
02:29:17 Nick Xerion_ is now known as Xerion (
02:30:32lee321987Also - how do you guys clean your DAPs' USB jacks?
02:36:39 Quit dfkt ("-= SysReset 2.53=- Ph'nglui mglw'nafh Cthulhu R'lyeh wgah'nagl fhtagn.")
02:37:41 Quit renke (Read error: 110 (Connection timed out))
02:39:51 Quit moos ("Rockbox rules the DAP world")
02:45:11lee321987I tried to run the lua script on −−−− on the line "rb.clear()" I get this error "attempt to call field 'clear' (a nil value)"
02:52:07Unhelpfulboo... counting hardware multiplies (2 each for 16x32, 4 for the final 32x32->40, 3 for 32x32->32) says the new scaler is a loser on sh-1. the best i can do for sh-1 is use the old math, try to fold the round into the 32x32->40 asm, and make as many of the multiplies as possible 16x16->32 hw multiplies.
02:52:44Unhelpfulthe newscaler wins for upscaling, but really, that'll never happen on an archos.
03:12:20 Quit Horscht ("Verlassend")
03:16:24 Join shodanX_ [0] (
03:23:38 Join mirak [0] (
03:32:33 Quit shodanX__ (Read error: 110 (Connection timed out))
03:34:50***Saving seen data "./dancer.seen"
03:38:26 Join Munkie [0] (n=29b118a3@gateway/web/cgi-irc/
03:39:21MunkieHow do i control my PC music player using my ipod
03:51:04mcuelenaerelee321987: that lua script is heavily outdated
03:52:28mcuelenaerethis is a newer version:
03:52:36 Quit efyx_ (Remote closed the connection)
03:53:46lee321987mcuelenaere: Thanks. A little investigation (with little C knowledge) lead me to try these:
03:54:01lee321987rb.lcd_drawline(0, 0, rb.LCD_WIDTH, rb.LCD_HEIGHT)
03:54:09lee321987which worked
03:54:19CIA-38New commit by mcuelenaere (r21063): Lua: add image handling + some other wrappers
03:54:53mcuelenaerelee321987: all wrappers are located in apps/plugins/lua/rocklib.c
03:55:02mcuelenaeredocumentation for them will (hopefully) come later
03:55:51lee321987mcuelenaere: thank you. But for now, just learning a little lua will help - yes?
03:56:50mcuelenaereprobably, yes
03:57:22lee321987just went to your pastie −−- awesome. Thanks.
04:00:04mcuelenaereUnhelpful: is there an uniform way of determing the maxsize for read_image_file() based on the width & height?
04:01:22mcuelenaerehmm and is there some helper function to get the width & height of a bmp/jpeg without reading the whole file into memory?
04:02:33 Join jmillikin [0] (
04:03:33mcuelenaerelee321987: if you want to play with the new lcd_bitmap_* wrappers: (do svn up first)
04:04:29lee321987thank you.
04:07:51MunkieHow do I assign the keys of my ipod to winamp using the HID controller
04:10:22 Quit antil33t (Read error: 110 (Connection timed out))
04:16:09 Quit Munkie ("CGI:IRC")
04:18:06CIA-38New commit by mcuelenaere (r21064): Lua: fix rocklua_image issue + add LCD_RGBPACK & LCD_RGBUNPACK wrappers
04:21:24 Join itcheg [0] (i=62db4c46@gateway/web/ajax/
04:23:26lee321987mcuelenaere: Should I remove the line: dofile("actions.lua")
04:24:30mcuelenaerelee321987: no you shouldn't, are you testing this on sim or on target?
04:24:39mcuelenaere(actions.lua gets included in a fresh rockbox build)
04:24:44mcuelenaeretry doing make install
04:25:42lee321987probably should have done a new configure....
04:27:00FlynDiceI'm trying to test funman's bank-boundary patch and I need to write a 3gb file system so as not to cross the 4GB boundary. Can I do this with mkfs and is there a howto somewhere on the site someone could point me to?
04:28:07CIA-38New commit by mcuelenaere (r21065): Lua: /me should learn about C arrays.. :/
04:29:54 Quit mcuelenaere ("Zzzzzz")
04:40:58 Join evilnick_home [0] (
04:42:45lee321987What determines whether I need to run "configure" again? I'm sure that I once altered a file, and ran "make" without re-running "configure" and it worked right...?
04:51:35 Quit Llorean (Read error: 54 (Connection reset by peer))
04:56:44lee321987mcuelenaere: WARNING, bad file name lacks slash: actions.lua
04:59:44 Quit miepchen^schla (Read error: 101 (Network is unreachable))
05:23:20 Quit mirak ("Ex-Chat")
05:29:54 Join Llorean [0] (
05:33:24 Join lee321987_ [0] (
05:33:43 Quit lee321987_ (Client Quit)
05:34:52***Saving seen data "./dancer.seen"
05:46:59 Quit lee321987 (Read error: 110 (Connection timed out))
05:50:27 Join lee321987 [0] (
05:53:55lee321987lua script: Is there a function to copy a file, without using either the "open" or "read" functions?
06:00:43 Join daurnimator [0] (n=daurnima@unaffiliated/daurnimator)
06:41:09 Quit KBH (Read error: 110 (Connection timed out))
06:52:05 Join n1s [0] (n=n1s@rockbox/developer/n1s)
06:54:28 Quit itcheg (" ajax IRC Client")
06:58:04 Nick JdGordon|afk is now known as JdGordon (n=jonno@rockbox/developer/JdGordon)
06:59:01JdGordonnoone had any comments on the touchscreen email? :(
07:18:05 Quit fdinel ("Miranda IM! Smaller, Faster, Easier.")
07:28:34 Join Zarggg_ [0] (
07:34:55***Saving seen data "./dancer.seen"
07:43:06 Quit martian67 (Remote closed the connection)
07:44:58 Join martian67 [0] (n=martian6@about/linux/regular/martian67)
07:45:52 Quit Zarggg (Read error: 110 (Connection timed out))
08:21:46 Quit rvvs89 (Read error: 104 (Connection reset by peer))
08:27:46 Join rvvs89 [0] (
08:27:58rvvs89ugh, whups, sorry
08:31:38 Nick bagawk_ is now known as bagawk (
08:35:51 Join dmb [0] (n=dmb@unaffiliated/dmb)
08:37:10 Quit martian67 ("out")
08:38:02 Join martian67 [0] (n=martian6@about/linux/regular/martian67)
08:38:38 Join Rob2223 [0] (
08:43:07 Quit martian67 ("out")
08:43:34 Join martian67 [0] (n=martian6@about/linux/regular/martian67)
08:53:04 Quit martian67 ("out")
08:53:20 Join martian67 [0] (n=martian6@about/linux/regular/martian67)
08:54:49 Quit martian67 (Client Quit)
08:55:27 Join martian67 [0] (n=martian6@about/linux/regular/martian67)
08:56:23 Quit Rob2222 (Read error: 110 (Connection timed out))
08:57:55 Quit martian67 (Client Quit)
08:59:48 Join martian67 [0] (n=martian6@about/linux/regular/martian67)
09:01:03 Quit jmillikin (Read error: 60 (Operation timed out))
09:01:05 Join itcheg [0] (i=62db4c46@gateway/web/ajax/
09:04:47 Join jmillikin [0] (
09:13:09 Quit itcheg (" ajax IRC Client")
09:21:21 Join stoffel [0] (
09:27:57 Join bertrik [0] (
09:34:56***Saving seen data "./dancer.seen"
09:44:55 Quit perrikwp (K-lined)
09:44:55 Quit saratoga (K-lined)
09:49:17 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
09:52:18 Quit lymeca ("KVIrc 3.4.0 Virgo")
09:52:33 Quit martian67 (Remote closed the connection)
09:52:55 Join martian67 [0] (n=martian6@about/linux/regular/martian67)
10:03:13 Quit lee321987 ("ChatZilla 0.9.84 [Firefox 3.0.10/2009042316]")
10:05:06 Join itcheg [0] (i=62db4c46@gateway/web/ajax/
10:06:21 Quit martian67 ("out")
10:06:53 Join martian67 [0] (n=martian6@about/linux/regular/martian67)
10:07:07 Quit martian67 (SendQ exceeded)
10:07:16 Join flydutch [0] (
10:07:36 Join martian67 [0] (n=martian6@about/linux/regular/martian67)
10:08:36 Quit martian67 (SendQ exceeded)
10:10:06 Join martian67 [0] (n=martian6@about/linux/regular/martian67)
10:13:15 Join antil33t [0] (n=Mudkips@
10:13:55 Quit martian67 (Client Quit)
10:14:48 Join martian67 [0] (n=martian6@about/linux/regular/martian67)
10:18:45 Quit martian67 (Remote closed the connection)
10:25:10 Join wincent [0] (
10:29:13 Join mt_ [0] (n=MTee@
10:33:56 Join martian67 [0] (n=martian6@about/linux/regular/martian67)
10:36:05 Quit martian67 (Client Quit)
10:37:00 Join martian67 [0] (n=martian6@about/linux/regular/martian67)
10:39:22 Join ender` [0] (
10:46:57 Quit mt (Read error: 110 (Connection timed out))
10:54:23 Join lymeca [0] (
10:55:40 Join george [0] (
10:58:17 Nick george is now known as jago (
11:00:40 Quit itcheg (" ajax IRC Client")
11:06:26 Nick jago is now known as jago_ (
11:10:23 Nick jago_ is now known as jago (
11:25:41 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
11:30:39 Quit rvvs89 (Read error: 110 (Connection timed out))
11:34:59***Saving seen data "./dancer.seen"
11:35:12 Quit Horscht ("Verlassend")
11:35:25 Join rvvs89 [0] (
11:44:01 Join mehrab [0] (n=quassel@
11:50:20 Quit jago ("ChatZilla 0.9.84 [Firefox 3.0.10/2009042316]")
11:51:05 Join itcheg [0] (i=62db4c46@gateway/web/ajax/
11:55:34 Quit ender` (" Do not believe any statistic you didn't falsify yourself.")
12:00:30 Quit flydutch ("/* empty */")
12:03:12 Quit mehrab (Remote closed the connection)
12:11:43 Join Ubuntuxer [0] (
12:11:59 Quit itcheg (" ajax IRC Client")
12:18:56 Join ufoman [0] (
12:21:39 Quit ufoman (Client Quit)
12:22:05 Join ufoman [0] (
12:23:06 Quit ufoman (Client Quit)
12:25:29 Join Rondom [0] (
12:27:28 Nick mt_ is now known as mt (n=MTee@
12:34:55 Join raphi [0] (
12:40:12 Quit stoffel ("leaving")
12:53:07 Quit linuxstb (Remote closed the connection)
12:57:08Unhelpfulmcuelenaere: the scaler needs a quantity of buffer determined by the output width. the jpeg loader needs an additional amount that is not easy to determine. i had considered a flag for read_* that would cause them to return early, after determining the space needed, but there has not yet been anything that actually requires it.
12:58:43 Join robin0800 [0] (
12:59:21 Join mcuelenaere [0] (
13:03:30 Quit bmbl ("Woah!")
13:05:07mcuelenaereUnhelpful: thanks. I'm trying to read in images in Lua and I'm able to alloc any kind of buffer; do you have any idea how I could read in arbitrary images (and if possible, scale them)?
13:06:53 Join renke [0] (
13:07:17Unhelpfulmcuelenaere: if you can use such a solution, the way the album art loader does it is probably easiest (see buffering.c:load_image). it basically hands the image loader the entire available buffer, then completes the allocation when the loader returns the size used by the image.
13:08:02mcuelenaerehmm I can't hand over the complete available memory though
13:08:10*mcuelenaere looks at that file
13:09:14Unhelpfulin that case, implementing the early-return-with-needed-buffer might be ideal
13:09:58Unhelpfulbuffering.c can only do it by blocking further allocation until the loader returns :/
13:10:17 Join juane414 [0] (n=chatzill@
13:10:19 Join barrywardell [0] (
13:10:42 Quit Ubuntuxer (Read error: 110 (Connection timed out))
13:10:43juane414hey guys whats the procedure for patching?
13:12:35 Join itcheg [0] (i=62db4c46@gateway/web/ajax/
13:12:55 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
13:13:09mcuelenaereUnhelpful: yes, early-return-with-needed-buffer is ideal for this situation I think :) basically what I need here is the buffer size, malloc() it and load the image into that buffer
13:13:58mcuelenaereI'm not sure whether scaling should come in the picture here though
13:16:14juane414how do i put a patch on my player? anyone know?
13:19:04juane414I already read that and didn't really understand that
13:19:16 Join miepchen^schla [0] (
13:20:08Mikachujuane414: you can start with just compiling from sourc
13:20:50juane414it seems like the whole process is too complicated for me
13:24:22 Join Jaykay [0] (
13:24:45Jaykayisn't FS #10211 fixed now?
13:25:28 Nick robin0800 is now known as robin0800_ (
13:28:26 Join moos [0] (i=mustapha@rockbox/staff/moos)
13:29:17 Join ender` [0] (
13:35:00***Saving seen data "./dancer.seen"
13:35:13 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
13:42:14 Quit ender` (" On the other hand, you have different fingers.")
13:42:21 Nick Zarggg_ is now known as Zarggg (
13:42:27 Join ender` [0] (
13:42:31 Join efyx_ [0] (
13:42:49Unhelpfulmcuelenaere: if you are using scaling, *or* JPEG, there will be a substantial excess buffer needed to load the image.
13:44:01Unhelpfulthe scaler, used if a BMP is not an exact fit, or if a JPEG is not exactly the requested size, or 1/2, 1/4, or 1/8 of it, requires 12*output_width bytes (triple that on color targets)
13:44:40 Quit robin0800_ ("No Ping reply in 90 seconds.")
13:44:58 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
13:45:00mcuelenaereok, then I'll implement that as a read_scaled_image function (which will be discouraged to use)
13:45:11 Join robin0800 [0] (
13:46:33Unhelpfulthe JPEG loader requires sizeof(struct jpeg), which is a bit over 8KB on most targets, plus up to ((input_width + 15) & ~15) * 16 bytes of decode buffer.
13:47:12 Nick fxb__ is now known as fxb (
13:47:55mcuelenaerecan this struct jpeg be declared static or do I need to alloc it for every jpeg image?
13:48:54Unhelpfulthe loader allocates it from the buffer at present. does your malloc have a realloc that could return the extra space? that might be the best way around this...
13:51:20 Join ender [0] (
13:52:25mcuelenaerehmm for that I could perhaps do a real malloc(), copy only the image to another buffer which gets 'lua_malloc()'ed (<- this one gets garbage collected) and free() the first buffer
13:54:52Unhelpfulthe buflib doesn't have a buflib_resize_buf yet, but it could. it's not very helpful for things that already use malloc, though - the interface is much closer to core buf* than to malloc, you get a handle.
13:56:03 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
13:57:13 Quit ender` (Read error: 60 (Operation timed out))
13:57:55Unhelpfulyou could avoid the copy if your malloc lib could return the last N bytes of an allocation to the buffer. it needn't be capable of a "real" realloc - i'd even considered having a buflib_shrink_buf.
13:59:08 Quit robin0800 ("No Ping reply in 90 seconds.")
13:59:34 Join robin0800 [0] (
14:05:37 Join BdN3504 [0] (n=55b2232e@gateway/web/cgi-irc/
14:07:31 Quit itcheg (" ajax IRC Client")
14:09:05 Join itcheg [0] (i=62db4c46@gateway/web/ajax/
14:46:36 Quit Jaykay ("ChatZilla 0.9.84 [Firefox 3.0.10/2009042316]")
14:49:09 Join mirak [0] (
15:02:48 Quit moos (Read error: 104 (Connection reset by peer))
15:09:18 Join teru [0] (
15:23:44 Join {phoenix} [0] (
15:27:14 Quit {phoenix} (Remote closed the connection)
15:27:19 Join BXCracer [0] (
15:27:58 Join {phoenix} [0] (
15:30:21 Quit SirFunk (Read error: 110 (Connection timed out))
15:35:02***Saving seen data "./dancer.seen"
15:36:00 Join stoffel [0] (
15:38:50trisiakhi, are there any plugins/apps in rockbox that needs optimization
15:41:12 Join Riku [0] (n=Lss@
15:42:58 Quit juane414 ("ChatZilla 0.9.84 [Firefox 3.0.10/2009042316]")
15:44:31 Join Thundercloud [0] (
15:47:00domonokytrisiak: sure, for example rockboy (gameboy emulator) could need more optimisations .. :-)
15:48:29trisiakdomonoky, are there some known problems that could be cleared without diving in the assembly code?
15:49:07trisiaki need to optimize some opensource project for my studies ;)
15:49:28domonokyi dont know... but feel free to dive into the code :-)
15:50:55 Join robin0800_ [0] (
15:51:09*gevaerts wonders if he should go for a very fancy solution with different USB write queues for different disks
15:51:58mcuelenaereUnhelpful: I got read_bmp_file working on Lua with an early-return-with-needed-buffer approach; should I just add FORMAT_DETERMINE_SIZE or .. ?
15:52:20bertrikwhat kind of cool stuff can we do with the new lua plugin?
15:52:29mcuelenaerebertrik: malloc! :)
15:52:45mcuelenaereyou can very easily load and play with images
15:52:57mcuelenaereno FP though
15:58:31mtIn metadata parsing, there's no problem with data copying, right ?
15:59:56mtlinuxstb & saratoga : ^
15:59:58 Join Lss__ [0] (
16:00:10 Quit Lss (Read error: 110 (Connection timed out))
16:01:04mtIn get_rm_metada, I copied the code from the parser in the test program and modified to take in struct mp3entry instead of RMContext.
16:01:15mtI'm thinking of changing that.
16:01:37 Quit prona ()
16:02:37mtI want to leave the current code in the test program intact, send it a RMContext from the metadata parser, then the metadata parser would have a function that would copy the required data from RMContext to mp3entry.
16:03:03mtI think this is will be better for debugging and readability.
16:05:33 Quit Lss__ ()
16:07:11gevaertsCan I stop a thread just by letting its function return?
16:08:11mcuelenaeregevaerts: I think you need to do thread_exit()
16:09:25 Quit renke (Read error: 110 (Connection timed out))
16:09:30 Quit mcuelenaere (Remote closed the connection)
16:09:47*gevaerts decide that leaving the thread running at all times is simpler for now
16:10:54CIA-38New commit by mt (r21066): Move rm.[c/h] from libcook to librm. No functional changes.
16:13:14 Quit robin0800 (Read error: 110 (Connection timed out))
16:13:37 Join flydutch [0] (
16:14:58 Join Lss [0] (
16:16:43 Join Ubuntuxer [0] (
16:17:20Unhelpfulmcuelenaere: something like that, yes. the same for jpeg will be a *little* more tricky. also, if you want to include the scaler buffers in the "needed size" value, you should look at using the BM_SCALED_SIZE macro - just use zero height and the desired output width. that + BM_SIZE or the return of get_size would work for BMP. for JPEG, also add the decode_buf_size that is calculated in read_jpeg_file and sizeof(struct jpeg)
16:18:47 Join Xerion_ [0] (
16:19:13 Quit Xerion (Read error: 60 (Operation timed out))
16:19:16 Quit Riku (Read error: 110 (Connection timed out))
16:19:45 Nick Xerion_ is now known as Xerion (
16:21:03 Quit robin0800_ (Read error: 110 (Connection timed out))
16:21:24 Join robin0800 [0] (
16:22:26Unhelpfulugh, misaligned on keyboard. BM_SCALED_SIZE is also preferable to use in the event that the allocation changes later, perhaps due to a less memory-hungry scaler, or a decision to cache-align the scaler buffers.
16:28:00 Quit stoffel ("leaving")
16:41:07 Join faemir [0] (
16:45:45trisiakcan i use static variables inside profstop() function?
16:46:13trisiaki want to avoid overwriting existing /profile.out file so that i can profile apps more than one time
16:49:53 Join |ahmina| [0] (
16:56:31 Quit BdN3504 ("CGI:IRC")
16:57:48 Join einhirn [0] (
16:58:48 Quit robin0800 ("No Ping reply in 90 seconds.")
17:00:26 Join robin0800 [0] (
17:01:43 Quit shodanX_ ("leaving")
17:03:58domonokytrisiak: using a static var in profstop() to change the filename is surely possible.. but i dont know how good this profiling thing works. you have to experiment with it a litte :-)
17:07:29 Quit _Auron_ ("Infinity repeatedly denies rumours of plotting with zero to bring down the Universe.")
17:07:39 Quit lymeca (Read error: 113 (No route to host))
17:09:27 Quit mirak ("Ex-Chat")
17:13:34domonoky misses iaudo m3. its commented in the source (, does anybody know why ?
17:15:16 Quit robin0800 (Remote closed the connection)
17:16:17domonokyalso misses iaudiom3 und releases, was iaudiom3 not included in the last release ?
17:25:12 Quit Ubuntuxer ("Leaving.")
17:25:14 Quit itcheg (" ajax IRC Client")
17:26:41 Part teru ("Leaving...")
17:29:41trisiakhmm, tools/configure allows to enable simulator and profile build at the same time, but its not possible to have profile.c included in firmware/SOURCES when building for simulator
17:30:30 Quit einhirn (Read error: 104 (Connection reset by peer))
17:30:50*AlexP points people that might know about tag reading at
17:32:49 Join SirFunk [0] (
17:35:04***Saving seen data "./dancer.seen"
17:45:47 Quit tvelocity[away] ("Αποχώρησε")
17:47:10n1strisiak: profiling the sim will probably not give results that are very representative for targets, which is probably why noone has med profiling work on the sim
17:47:21 Join robin0800 [0] (
17:53:01 Quit bittin```` (Remote closed the connection)
17:57:35 Join kugel [0] (n=kugel@rockbox/developer/kugel)
17:58:17 Join bittin```` [0] (
17:59:20 Quit robin0800 (Remote closed the connection)
18:02:31 Join robin0800 [0] (
18:06:34 Join tvelocity [0] (
18:07:43 Quit SirFunk (Read error: 110 (Connection timed out))
18:08:45 Join SirFunk [0] (
18:09:03CIA-38New commit by kugel (r21067): Fix FS #10231. The gwps_leave_wps() call at powering down is completely unnecessary anyway as it's meant to prepare the displaying of the next screen, ...
18:12:15kugelJdGordon: not splash related :P
18:12:39JdGordonlies! lies and slander!
18:13:09JdGordonkugel: ok, whats the state of the custom list patch?
18:13:38kugelit's good so far, however I'm still stuck with the zero-timeout splashes
18:14:45JdGordonah righto
18:18:02kugelI surely welcome any ideas regarding this
18:18:33JdGordonmy view on this is well known.. isnt it?
18:19:15kugelI can't recall it, no :(
18:19:49JdGordonbassically ignore the issue...
18:20:15 Quit Rondom (Read error: 104 (Connection reset by peer))
18:20:32JdGordonhow much of the ui works with the custom vp without extra fiddling?
18:20:36kugeloh, well..
18:20:58kugelit replaces the NULL viewport
18:21:26JdGordonok, nice
18:22:26JdGordonarg, the patch isnt exactly small :p
18:25:41JdGordonbah! the splash doesnt use the cvp right? so all thats needed is to set some flag to dirty if a screen uses the real full screen (or part of it) and then next time viewportmanager_redraw() is called blank out the are around the cvp
18:26:12JdGordonor.... do the extra work to hook up the wps to be used instead of the statusbar for the rest of the screen :)
18:26:32JdGordonactually.. that problem might still happen with the background wps
18:27:38 Join renke [0] (
18:28:54kugelbtw, is there a way to do FS #10233 without duplicating the menu?
18:30:21JdGordonyes, but needs to use a MAKE_MENU() macro instead os its a bit more work....
18:30:25JdGordonthat patch is fine
18:31:05 Quit perrikwp (" ajax IRC Client")
18:31:09 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
18:32:26 Quit robin0800 (Remote closed the connection)
18:37:16 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
18:37:33*JdGordon adds some comments to 8799
18:40:45 Join robin0800 [0] (
18:44:49 Quit Thundercloud (Remote closed the connection)
18:51:28 Quit robin0800 (Remote closed the connection)
18:52:52 Quit SirFunk (Connection timed out)
18:53:47 Join SirFunk [0] (
18:58:01*kugel answered
19:04:56 Join merbanan [0] (
19:06:53 Quit faemir ("Leaving")
19:07:19 Join Jaykay [0] (
19:07:45Jaykayis FS #4817 still valid?
19:08:34kugelJaykay: why shouldn't it?
19:09:15kugellcd-generic.c seems like a good thing to do
19:09:25Jaykaybecause rockbox is a open source-project and recieves new features, fixes and improvements every day.
19:10:26Jaykaydammit, tweaks, not improvements.
19:10:37kugelimprovements too, imo
19:10:54kugelbut that doesn't explain why it shouldn't be valid anymore
19:11:01 Join dany_21a_ [0] (
19:11:01Jaykaythats a quote from the manual btw.
19:11:24kugelthe patch seems to unify the scrolling code, and that's still seperate in SVN
19:11:48Jaykayi just thought maybe it is out of date because someone did this in svn without knowing the patch
19:11:51Jaykaywell, ok.
19:14:11kugelit's obviously heavily out of sync
19:14:34 Join p3tur [0] (
19:16:32 Nick p3tur is now known as petur (n=peter@rockbox/developer/petur)
19:18:22Jaykaykugel: i guess i would need some more knowledge of rockbox (and c in general) to get this working, right?
19:18:37 Quit {phoenix} (Remote closed the connection)
19:19:53kugelThat seems like a good guess. The whole drivers are viewport'ified now, so that patch needs adaption. Also, in hits all targets (mono, greyscale, color) so it's a tough job
19:20:55Jaykaynot a job for me then, thanks :)
19:28:24 Quit kugel (Nick collision from services.)
19:29:59 Join kugel [0] (n=kugel@rockbox/developer/kugel)
19:30:43 Join stoffel [0] (
19:35:05***Saving seen data "./dancer.seen"
19:40:59 Quit raphi ("leaving...")
19:46:41Jaykayanother old patch: FS #2648. jdgordon wrote he was about to commit this but needed the real name of rubberglove. we'll never get his name so i suggest committing it without the name or closing it
19:49:00 Quit SirFunk (Read error: 60 (Operation timed out))
19:50:03 Join lee321987 [0] (
19:51:00 Join fdinel [0] (
19:51:31 Join SirFunk [0] (
19:51:37 Join EspeonEefi [0] (i=eefi@SAFFRONCITY.MIT.EDU)
19:53:59lee321987(in RB) System -> Rockbox Info -> Buffer −−−− Is that how much RAM I have to work with for any plugins/lua scripts?
19:55:11Unhelpfullee321987: no. plugins load into a buffer that, on most targets, is 512MB. that space must suffice for their code and data, for the most part.
19:55:32kugel512MB? Not even the beast has that :)
19:55:36Jaykaynot kb?
19:56:31kugelkb is correct
19:56:53trisiakoptimizing rockboy is beyond my skills :P aren't there any other plugins to optimize?
19:57:01lee321987kugel: all targets?
19:57:06UnhelpfulKB. eesh.
19:57:08kugelNo, not all
19:57:18kugelbut for your c200 it should be 512K
19:58:14Unhelpfula plugin *may* call bufalloc to obtain allocations from the space used by the audio buffer. this will fail if the audio buffer is too full.
19:58:42Jaykaytrisiak: only because they work it doesn't mean they are fully optimized.... you could start with any of them :)
19:59:11Unhelpfulalso, this mechanism is not available on the archos targets, which have their own buffering code.
19:59:22 Quit EspeonEefi ("さよなら")
19:59:27lee321987trisiak: Bubbles gets slow/fast on my Sansa c200 sometimes
19:59:44trisiaklee321987, will check that
19:59:47kugeltrisiak: xobox maybe? :) It acts quite funny if you have it at max speed, i.e. it'll slow down every now and then depending on the boost state
20:00:35kugelsame goes for bubbles
20:00:57kugelit can get really slow at higher levels on my e200
20:02:07kugelpick any, before Unhelpful gets his hands on it :)
20:02:33 Quit petur ("rebooting linux *again*")
20:04:46 Join petur [0] (
20:07:44 Join AndyI [0] (i=AndyI@
20:09:55Unhelpfuli would expect that bubbles can render quite fast enough on most targets... it should be using the tick to make sure animations don't change speed. on targets where it really can't keep up without boost on, it should just boost itself.
20:11:16Unhelpfuland i've only done any real work on *one* plugin, and it's mostly been optimizing for memory.
20:11:50 Quit SirFunk (Read error: 110 (Connection timed out))
20:12:31 Join SirFunk [0] (
20:21:27 Quit AndyIL (Read error: 113 (No route to host))
20:21:52 Join matsl_ [0] (
20:22:56 Join Riku [0] (n=Lss@
20:23:35 Quit perrikwp (" ajax IRC Client")
20:31:13Horschtso, I am trying to build the manual using rev 21067 and this is the error i get:
20:31:34 Quit flydutch ("/* empty */")
20:35:51 Quit miepchen^schla ()
20:35:52 Quit Lss (Read error: 113 (No route to host))
20:42:34 Join miepchen^schla [0] (
20:46:40 Join moos [0] (i=mustapha@rockbox/staff/moos)
20:47:03 Quit JdGordon (Read error: 110 (Connection timed out))
20:52:29 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
20:54:18AlexPHorscht: I think I found it - one mo
20:54:56 Join KBH [0] (
20:55:33 Quit lee321987 ("ChatZilla 0.9.84 [Firefox 3.0.10/2009042316]")
20:56:29AlexPHorscht: Yep, just doubling checking then I'll commit it
20:58:02 Quit Riku (Read error: 60 (Operation timed out))
20:59:50CIA-38New commit by alex (r21068): Fix manual building.
21:00:59AlexPHorscht: Try again now
21:01:22FlynDiceSo when I run a battery bench do I really need to let it run all the way down or can I get some useful info after a shorter period?
21:02:53HorschtAlexP, works fine now+
21:02:54kugelI'd recommend doing a full bench
21:02:56n1sFlynDice: what info is it that you want?
21:02:59AlexPIt would be very useful to have a build table for the manual to be able to more easily see when it is broken
21:03:05AlexPHorscht: cool
21:03:10n1sAlexP: very
21:03:23kugelparticularly on in-developement targets, the numbers reported by the plugin might not be accurate enough
21:03:49FlynDicejust looking to see how different frequencies/configurations use battery power
21:05:10FlynDicekugel: I figured that was going to be the answer but thought I'd ask in case there was a better way...
21:05:25kugelyou managed to get mp3 @90MHz? that sounds nice
21:06:22kugel160 or whatever it is currently is just way to much, no matter of the mmu/caching stuff
21:06:26FlynDicewell, upping PCLK seems to be the key, it lets it run unboosted for 80% of the time and lowers the average
21:07:34FlynDiceI'm trying to check on the power cost now thogh
21:07:41kugelI mean, the e200v1 does mp3 below 30MHz. That's with dual core, but even 60MHz would be better than what we do (and the v1 doesn't have any caches)
21:07:43n1sFlynDice: the best method for measuring power draw is using a current meter in series with the battery (or a power source)
21:07:58n1sthere's always uncertainty with battery benches
21:09:00n1skugel: the pp's have cache
21:09:06kugelthey have?
21:09:24kugelhow much?
21:09:46n1syes, i don't remember the details but they cache both code and data (unlike coldfires)
21:09:59FlynDicen1s: OK, thanks, I don't know that we really need to be that precise right now I'm looking for more of a ballpark figure so to speak
21:10:44n1sFlynDice: the current meter has the added benefit of giving a result much quicker than a battery bench
21:10:54kugelah, I see:
21:10:55 Join mirak [0] (
21:11:00n1sbut sure a battery bench is a good indication
21:13:24 Part dany_21a_
21:17:32 Quit kugel (Nick collision from services.)
21:17:43 Join kugel [0] (n=kugel@rockbox/developer/kugel)
21:17:53 Join Mathiasdm [0] (
21:19:13 Join wincent_balin [0] (
21:22:56mtIs it possible for CIA to display multiline messages ?
21:25:10mthm .. some orgs have multiline messages with CIA. (FreeBSD for example - see #commits)
21:26:23 Quit wincent (Read error: 110 (Connection timed out))
21:28:41 Nick J-23 is now known as Moarc (
21:30:11 Join saratoga [0] (i=9803c6dd@rockbox/developer/saratoga)
21:30:29 Join itcheg [0] (i=62db4c46@gateway/web/ajax/
21:31:33saratogaFlynDice: I just won an e200v2 on ebay, so I can make whatever power measurements you like via a DMM once i get it
21:32:52 Quit JdGordon ("Leaving.")
21:35:06***Saving seen data "./dancer.seen"
21:36:44 Part wincent_balin ("Kopete 0.12.7 :")
21:37:12 Join archivator [0] (n=archivat@
21:39:08 Nick Moarc is now known as J-23 (
21:39:17 Quit Jaykay ("ChatZilla 0.9.84 [Firefox 3.0.10/2009042316]")
21:40:14 Join {phoenix} [0] (
21:42:52kugelsaratoga: "won"? nice
21:44:33 Join JdGordon1 [0] (
21:45:10kugelgood that e200v2 messures apply to the fuze too
21:54:53 Quit archivator (Read error: 113 (No route to host))
21:55:05 Join archivator [0] (n=archivat@
21:58:54 Quit Horscht ("Verlassend")
22:00:08 Quit Tristan (Client Quit)
22:00:27gevaertsdoes ata_write_sectors() yield for dma writes?
22:02:17 Quit |ahmina| ("KVIrc 3.4.0 Virgo")
22:02:26 Quit matsl_ (Read error: 104 (Connection reset by peer))
22:05:07 Join Tristan [0] (
22:07:52 Join matsl_ [0] (
22:09:46 Quit mt (Read error: 104 (Connection reset by peer))
22:10:15BXCracerhello, how can i make .lng file from .lang ? I am trying to translate rockbox to other language and cant figure this out
22:11:53 Nick J-23 is now known as Moarc (
22:12:04 Nick Moarc is now known as J-23 (
22:12:19n1sBXCracer: the easies way is using the regular build system and just dropping your .lang file in the dir with the other language sources
22:13:04n1sit *can* be done manually with the genlang script but requires a quite complicated set of target dependant command line options
22:20:09 Quit linuxstb (Remote closed the connection)
22:20:55 Quit BXCracer (Connection timed out)
22:23:20 Quit kugel (Read error: 60 (Operation timed out))
22:29:08 Join kugel [0] (n=kugel@rockbox/developer/kugel)
22:34:52 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
22:36:28 Quit n1s ("Lmnar")
22:45:20 Join BXCracer [0] (
22:46:59CIA-38New commit by gevaerts (r21069): always call the class driver init function. This is needed because they are called before the usb_core_enable_driver() calls, so depending on enabled ...
22:48:07 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
22:51:20archivatorCan we get a "Commits since 3.2 (or last stable release)" page? The 3.0 page has gotten quite large ..
22:54:53kugelpictureflow doesn't do anything on a remote, does it?
22:56:27 Quit tmzt (Read error: 110 (Connection timed out))
22:57:35gevaertsIf people want faster USB *and* don't care too much about their data, there's FS #10239
22:58:57 Quit HellDragon (Client Quit)
23:03:50 Join HellDragon [0] (
23:04:40 Quit miepchen^schla (Read error: 60 (Operation timed out))
23:04:58archivatorFeature request (kinda): Isn't there *any* way we can get metadata in "View current playlist"? My iPod is full of these 4-letter filenames (courtesy of iTunes, of course) and thus that part of rockbox is quite unusable as is..
23:08:59Lloreangevaerts: Re FS #10239 (write caching), which targets might it get better than the OF on?
23:10:07 Quit {phoenix} (Remote closed the connection)
23:10:22Lloreanarchivator: Many tagging tools can rewrite all the tags rather quickly.
23:10:25gevaertsLlorean: on e200 it's now slightly faster than the OF, and on my ipod video, combined with FS #9708 (set to udma4) I get writes on windows at 12.1MB/s while the OF gets 9.8
23:10:33Lloreanarchivator: Er, all the filenames
23:10:59Lloreanarchivator: It could take quite a large amount of time for Rockbox to retrieve the Title tag for ever file in your playlist, especially if you don't have the database initialized.
23:11:16Lloreangevaerts: Impressive.
23:11:51gevaertsLlorean: part of the windows difference may be due to us not telling the OS about the caching, so it never tries to flush the cache
23:11:51HorschtI might give it a whirl tomorrow on my Ipod Video 80GB. I get ~12Mb with the OF
23:12:34Horschtbut now it's time to check for any bugs in my matress
23:12:38 Quit domonoky (Read error: 104 (Connection reset by peer))
23:14:01gevaertsLlorean: on linear read and write tests (dd to /dev/sdb), we're still slower than the apple OF
23:14:02archivatorLlorean: I actually still use iTunes occasionally to manage the tracks it knows about.. In the case of initialized and loaded into RAM database, can't we query it to retrieve the metadata? Metadata retrieval doesn't have to work in all use cases but I'd imagine that one is fairly common ..
23:15:03gevaerts13.8 MB/s read and 12.5 MB/s write with FS #9708 and FS #10239 vs 14.3 MB/s read and 14.0 MB/s write with the OF
23:15:53Lloreanarchivator: It's still a lot of special-case code primarily for iTunes users.
23:16:28 Quit stoffel ("leaving")
23:16:28gevaertsThe main things that makes FS #10239 risky right now are (a) not much testing yet, and (b) it hides write errors, so if you have a bad disk writes will fail silently
23:17:05archivatorLlorean: While we are particularly hit by the lack of this feature, I figure is actually a reasonable user experience improvement..
23:17:39Lloreanarchivator: A human readable collection will already have human readable filenames.
23:17:53LloreanI'm not sure how it significantly benefits most people, just those with mangled filenames.
23:17:55 Quit merbanan (Read error: 104 (Connection reset by peer))
23:17:57 Join miepchen^schla [0] (
23:18:51Bagderarchivator: rockbox is also doing playlist highly effectively a lot by _not_ checking for meta data of the files
23:19:11Bagderhm, oddly put but you might understand anyway
23:20:37LloreanBagder: Isn't the playlist viewer using the plugin buffer?
23:20:55Bagderah it might, yes
23:20:57*gevaerts looks at the speeds reported in and concludes that we probably can't go much faster anyway
23:21:01LloreanBagder: Wouuld changes to it impact binsize? Or, more interestingly, would it be possible to have alternative "advanced" viewers somehow?
23:21:35Bagderiirc, there is a patch somewhere (or was at least) that did (parts of) what archivator suggests
23:22:34LloreanYeah, it went through a few iterations, I seem to recall hearing it was incredibly slow but I think someone fixed it to just use database.
23:24:31LloreanI don't really like the idea of coded solutions to manager-specific problems, but if there's a real reason it'd be particularly useful to users that might be different.
23:24:52LloreanI just don't really see a reason why one pass with a filename fixing tool isn't *generally* a solution with the exception of people who want to keep using iTunes.
23:24:54 Join Chrono- [0] (i=Chaterof@
23:27:07linuxstbLlorean: I see it as being a useful feature, considering the number of ipod video users we probably have, and that we are missing at least one major feature of the OF on that device, so interoperability with the OF is probably desirable for many users.
23:28:19Lloreanlinuxstb: I don't understand why interoperability is so desirable?
23:28:20CIA-38New commit by peter (r21070): Cut the files in half and it might work better (note to self: check your tree is really clean before patching)
23:28:30LloreanEven if you go with "they need the OF for videos" it's audio files we're talking aobut.
23:29:30 Part Chrono-
23:32:42 Join lee321987 [0] (
23:35:08saratogaif you try to put a V1 Fuze/Clip firmware on a V2 device, it will fail the checksum and reject the update right?
23:35:10***Saving seen data "./dancer.seen"
23:36:29 Quit lee321987 (Client Quit)
23:37:01 Quit matsl_ (Read error: 110 (Connection timed out))
23:37:13 Quit renke ("leaving")
23:42:28gevaertsok, the scsi protocol bits of this caching scheme shouldn't be too hard to implement
23:42:49gevaertsWould it make sense to also do read-ahead caching for USB?
23:47:25 Quit Mathiasdm ("Yuuw!")
23:48:08 Quit BXCracer (Remote closed the connection)
23:48:34linuxstbgevaerts: Could it hurt?
23:49:15archivatorgevaerts: would readahead as implemented in FS #9621 work (could save you some work)?
23:49:15gevaertsit shouldn't, except in terms of a bit of binsize
23:49:17 Join kushalone [0] (n=kushal@
23:50:08gevaertsarchivator: I'm not working at the fat level here, so probably not
23:51:39gevaertsall I can really do on usb is read the next bit
23:52:57 Quit moos (Read error: 104 (Connection reset by peer))
23:52:59 Join DrMoos [0] (
23:55:16 Quit fyrestorm (Read error: 104 (Connection reset by peer))
23:55:23 Quit bmbl ("Woah!")

Previous day | Next day