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

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

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

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

#rockbox log for 2010-02-22

00:01:00 Quit CGL (Quit: Saliendo)
00:01:08mc2739pamaury: for the logs - r24835 did not help
00:04:15 Join CGL [0] (~CGL@
00:10:40Unhelpfulkugel: it ought to be...
00:11:29 Quit Buschel ()
00:12:36 Quit ansuz (Remote host closed the connection)
00:13:16 Quit JdGordon (Quit: Leaving.)
00:15:29 Quit robin0800 (Ping timeout: 256 seconds)
00:15:44 Join JdGordon [0] (
00:15:44 Quit JdGordon (Changing host)
00:15:44 Join JdGordon [0] (~Miranda@rockbox/developer/JdGordon)
00:18:25S_a_i_n_twhere can I find info on the "fft" plugin? does any exist yet?
00:20:59stripwaxS_a_i_n_t : - isn't it in the manual ?
00:22:11 Quit flydutch (Quit: /* empty */)
00:22:44stripwaxoh. hm. nope.
00:23:12S_a_i_n_tyeah...nope, just looked :P
00:23:38stripwaxS_a_i_n_t : - in that case probably the only documentation is the FS# from which it came ..
00:23:48stripwaxat least until the author (or somebody else) adds to the manual
00:24:09stripwaxApart from Division, why is APE slow?
00:24:25saratogareally complicated filters used to predict each sample
00:24:36S_a_i_n_tbecause it hasn't evolved yet?
00:24:38*S_a_i_n_t sniggers
00:24:47*stripwax wow
00:24:57stripwaxsaratoga - has it been profiled?
00:26:20amiconnstripwax: ape is a symmetric codec, meaning that encoding and decoding have about the same cpu requirements
00:27:00 Quit bertrik (Quit: De groeten)
00:27:49amiconnIt's already quite optimiised. The higher levels use high order filters (even multiple "layers")
00:28:28amiconnIt's designed for tight compression, not for fast decoding
00:30:45 Quit anewuser (Quit: =ooo ϢINTER ϾHIP 5iVE is OOON!! omg 7 days left =oo)
00:31:24stripwaxamiconn - yeah .. but reading around makes it sound like it's fast and highly efficient - I guess that's for encoding though rather than decoding? It also seems to state it's quicker [to encode, presumably] than e.g. FLAC. is FLAC slower encoder/faster decoder?
00:31:35 Quit ender` (Quit: On-line, adj. The idea that a human being should always be accessible to a computer.)
00:33:26amiconnI don't know how the encoding speeds actually compare.
00:34:03 Part domonoky
00:34:55TheSevenit's quite fast for encoding
00:35:18TheSevenat least at the lower compression levels, it's a lot faster than flac/mp3
00:35:29TheSevenbut for decoding, it's obviously the other way round
00:35:49 Quit JdGordon (Read error: Connection reset by peer)
00:36:07 Quit efyx_ (Remote host closed the connection)
00:36:57 Join anewuser [0] (anewuser@unaffiliated/anewuser)
00:37:18linuxstbTheSeven: Where do you get that from? (faster to encode than flac) ?
00:38:53TheSeveni tried it once
00:39:22 Join JdGordon [0] (
00:39:23 Quit JdGordon (Changing host)
00:39:23 Join JdGordon [0] (~Miranda@rockbox/developer/JdGordon)
00:40:29linuxstbTheSeven: The latest FLAC comparison claims FLAC is faster than even "APE fast" -
00:40:55stripwaxTheSeven hm, not very symmetric then
00:41:05linuxstbI'm pretty sure recent FLAC versions are much faster than older versions at encoding.
00:41:34 Quit n17ikh ()
00:41:35stripwaxlinuxstb - on the Comparison link - - but it does state FLAC 0.1
00:42:04linuxstbstripwax: Yes, I saw that. I don't think a comparison with 0.1 is fair ;)
00:42:35 Join n17ikh [0] (
00:43:14TheSevenlinuxstb: indeed. i just retried it using monkeys audio 4.06 and flac via goldwave
00:43:30TheSevenc1000 is about the same speed (7 seconds for a 4:40 file) as flac high
00:43:43 Join veeloc [0] (
00:44:13linuxstbIs there much difference in filesize?
00:45:16TheSeven32628k (ape), 33950k (flac)
00:45:24TheSevenso c1000 is better than flac high
00:45:36TheSeven(at least for that file)
00:47:38 Quit bmbl (Quit: Bye!)
00:54:59 Quit JdGordon (Read error: Connection reset by peer)
00:56:32 Quit iq (Remote host closed the connection)
00:57:09 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
00:58:04***Saving seen data "./dancer.seen"
00:58:56 Join Rob2223 [0] (
01:02:43 Quit Rob2222 (Ping timeout: 272 seconds)
01:17:44 Quit stripwax (Quit:
01:24:40 Join Casainho [0] (
01:24:45Casainhoj reprap
01:27:16 Join iq [0] (~iq@unaffiliated/iq)
01:33:55 Quit mc2739 (Ping timeout: 245 seconds)
01:35:55 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
01:38:22 Quit JdGordon_ (Read error: Connection reset by peer)
01:38:27Unhelpfulnewer flac versions use several heuristics and iterative-solution methods for the various levels where old flac uses various degrees of brute-force searching.
01:38:38 Join JdGordon_ [0] (
01:44:17 Join nn [0] (
01:47:15amiconnWith the optimised divison, ape entropy decoding is more efficient on PP5002 (31.8MHz) and PP5022 (30.0MHz) than on coldfire (33.4MHz) (!)
01:47:32amiconnErr, PP5022 is 30.9MHz
01:48:55amiconni.MX31 needs 27.6MHz, PP5020 needs 35.5MHz due to the IRAM problem
01:49:18 Part nn
01:52:04 Quit Frakenstein (Read error: Connection reset by peer)
01:56:29*kugel has resizable skin buffer working
01:58:26 Quit JdGordon_ (Read error: Connection reset by peer)
01:58:37S_a_i_n_tkugel: Sounds interesting...I keep finding new ways to overflow the skin buffer with eye-candies WPS'
01:58:54 Join JdGordon_ [0] (
01:59:27 Quit Tomis (Quit: Tomis)
02:00:11S_a_i_n_tIf I didn't seem to feel I needed to animate *everything*, I doubt I'd have this problem.
02:00:11 Join Barahir_ [0] (
02:00:31kugelit's probably not fail-safe right now though
02:02:12 Quit Casainho (Ping timeout: 260 seconds)
02:02:57 Quit Barahir (Ping timeout: 248 seconds)
02:03:14S_a_i_n_tkugel: what percentage of RB actually is? :P
02:03:20saratogafor encoding speed, flac w/ CUDA is pretty hard to beat
02:03:37S_a_i_n_tpeople seem to find new and interesting ways to make "fail-safe" things fail :D
02:03:42 Join AndChat- [0] (
02:05:33kugelUnhelpful: could I change the return value of read_bmp_file to -totalsize if the given buffer is too small?
02:06:02kugelso one could check for that error and still retrieve the buffer size needed
02:06:32kugeloh, I see that FORMAT_RETURN_SIZE nopw
02:08:01kugellovely \o/ that solves half of my problems
02:08:05 Quit JdGordon_ (Ping timeout: 265 seconds)
02:09:31 Quit AndChat- (Read error: Connection reset by peer)
02:12:30kugelS_a_i_n_t: it just has gotten a lot more fail-safe :)
02:12:37 Join Tomis [0] (~Tomis@
02:20:43 Quit iq (Remote host closed the connection)
02:33:53 Quit kugel (Ping timeout: 248 seconds)
02:42:22 Join iq [0] (~chatzilla@unaffiliated/iq)
02:48:17 Join kugel [0] (~kugel@rockbox/developer/kugel)
02:49:54 Join taksato [0] (
02:50:16taksatoAnyone here have familiarity with the ipod serial code/registers/logic?
02:50:45taksatoI have been investigating it and had just posted on the forums. Thought an interactive dialog would be better heere.
02:51:18taksatoI enabled the code for sansa e200 v1 and had one ambiguity that will double research efforts.
02:51:53taksatothe ambiguity is the serial_setup routine.
02:52:29taksatoI have to try all tests using both combo of the clearing of bit C or bit 0xc000
02:52:38taksatomake that 0xc00
02:54:22taksatoNeither initialisation routines seemed to work well and show any different data in my edited debug i/o screen registers and mem locations
02:58:05***Saving seen data "./dancer.seen"
03:00:35linuxstbtaksato: I can't recall much discussion of the ipod serial port code happening here in IRC - I'm not sure who actually worked on the code, apart from LinusN doing the very early work. The original flyspray task should tell you though.
03:01:52taksatogood call i will check it out. do you happen to know where the 0x7000008C and etc. comes from?
03:02:12 Quit jfc (Read error: Connection reset by peer)
03:02:19taksatobasically it is setting a bit in it to zero to route the pins
03:02:36 Join jfc [0] (
03:02:42 Quit advcomp2019 (Read error: Connection reset by peer)
03:03:03saratogataksato: have you been able to look at what the signals on your accessory look like electrically?
03:03:07 Join advcomp2019 [0] (~advcomp20@unaffiliated/advcomp2019)
03:03:09saratogais it rs232?
03:03:43taksatoi guess so, saratoga. i took your advice on the thread in the forum for enabling the ipod serial and observing
03:04:07saratogabut you haven't looked at the analog signal to be sure its rs232 or similar?
03:04:52taksatono. considering ordering a set of dd1 connectors to sit between the dock and the sansa which io can try to tap into if i can get lab access at work
03:05:31saratogatheres actually a few places that sell sandisk break out boards
03:05:35saratogathey're not too expensive
03:05:52taksatothese styleL :
03:06:03taksatoah, that sounds interesting.
03:06:09saratogaalthough if you're not in N. America i don't know how much it'll cost to ship
03:06:33taksatoi am in usa
03:07:21taksatodo you have a reference to the breakout board? found some stuff on digikey and not sure if it is right
03:08:39taksatoi did open the altec lansig im413, but those pins did not look easy to tap and then untap later and keep my functionality
03:09:28taksatoalso was not sure if they map to the correspongind out pin, because the continuity checker in my mm was ringing in place where i didnt expect rings
03:10:11saratogatheres a couple places i've seen these for sale
03:10:13saratogalet me look
03:10:19taksatothank you
03:12:19taksatogoogle seems to be censoring the internet. search for sanbreakout and it is not the first item. bing has it first
03:13:21saratogathese are recommended elsewhere, but are rather expensive
03:16:07taksatoexcellent, thank you. bookmarked til i can find out if i can get eng lab access at work
03:16:37saratogaa simple scope should tell you if its rs232 or similar
03:16:37taksatothen need some combo of this and a female dd1 spreader
03:16:43saratogaverses just something like a GPIO
03:17:13taksatoill have to figure it out. we have scopes in that lab. when i was working with an fm radio part on a mobile phone i went in there with our ee guy
03:17:32saratogathat should work fine
03:18:25 Quit Sajber^ (Ping timeout: 246 seconds)
03:19:36FlynDicesaratoga: re SansaAMS PCLK & mpmc_clk, I have gotten a white screen any time I tried to use PCLK_DIV1_SEL = 1 to make PCLK = mpmc_clock/2. I started reading the pl172 docs to investigate further but never got too deep into it.
03:20:05 Join JdGordon_ [0] (
03:20:11saratogaFlynDice: were you in the bootloader or the main binary?
03:20:39FlynDicenever tried it in the bootloader
03:20:50saratogafunman's patch implies that changing it is likely to corrupt the DRAM
03:20:54 Quit MethoS- (Read error: Connection reset by peer)
03:20:59saratogaso it probably has to be set in the bootloader before the DRAM is enabled
03:21:27saratogatheres a comment to the effect of "don't change this while the system is running"
03:21:40 Part froggyman
03:21:46FlynDiceI think that would be consistent with what I observed
03:22:06saratogado you have a copy of the linux patches?
03:22:54taksatosooo... say I find out that it is using i2c or uart when i do the breakout and scope... on the RB Side, is it goign to be trial and error on choosing the right address and bit to set if it is uart?
03:22:57FlynDiceyes, are you talking about a recent patch from funman or the linux patches he just got?
03:23:07saratogathe ones he just got
03:23:15FlynDiceI have the linux patches...
03:23:20saratogathey have a lot of interesting comments about how various things work that are not mentioned in the specs
03:23:23saratogaok good
03:23:54saratogataksato: well if its GPIO then I think it would be fairly easy
03:24:14saratogaif its serial its probably also not too hard but it will take some trial and error
03:24:21FlynDiceI saw he found the patches but haven't had a chance to look yet, not much free time today so far
03:24:30saratogayeah same here, i should get back to work
03:24:30taksatoif UART, then basically just hammer on a lot of mem addr i guess
03:24:56saratogaif its UART its probably the same as the ipod, though maybe something else has to be inited that isn't on the sansa
03:25:32taksatothe ipod has two different MODEL_NUMBER init paths
03:25:42taksatodifferent addr and different bit
03:26:28taksatoprocessors on them seem different than the apple's ones
03:28:52taksatoi met an ex portal player founder, all the things i wanted to ask him, but didnt!
03:29:18saratogathe sansa v1s use similar processors to the ipod video/nano
03:29:29saratogathe other path is probably for the older PP chips
03:30:02taksatoyea, looks liek ipods use older and successively newer, but sansa uses newer than the last ipod supported under serial for RB
03:33:00saratogathe sansa is PP5024 (e200v1) or PP5022(c200v1)
03:33:12saratogathey're both basically the same thing except one has an onboard DAC
03:33:26saratogalater ipods used PP5020 or PP5022
03:34:53saratogai guess in theory they could probably use something like i2c for the accessories too
03:35:46taksatoLooks like he supported ipodvideo, ipodcolor nano1g and ipod4g
03:36:28taksatoactually that's an if preproc... no it's an if else if preproc. i think i did not try the empty case
03:38:53taksatowas encouraging, but no luck
03:41:38 Quit mt (Ping timeout: 240 seconds)
03:42:08 Quit iq (Quit: ChatZilla 0.9.86 [Firefox 3.5.5/20091102152451])
03:42:50CIA-8New commit by 03funman (r24838): sd-as3525v2: working transfers, still awfully slow ...
03:42:52CIA-8New commit by 03funman (r24839): as3525v2: core_sleep()
03:43:00CIA-8New commit by 03funman (r24840): as3525v2: build normal firmware properly (Clipv2/+)
03:43:58taksatoi am looking fwd to when i can woot for docks <G>
03:45:48taksatoThe topic or board you are looking for appears to be either missing or off limits to you
03:45:55taksatohow ot access?
03:46:20kugelhaha funman, nice
03:46:30kugelI guess he couldn't resist working on it :)
03:47:52kugeldamn, I need to work on my fuzev2 finally
03:48:09 Join funman [0] (~fun@rockbox/developer/funman)
03:48:15 Join elinenbe [0] (
03:48:16funmankugel: ;)
03:48:24elinenbeIs Rafal Carr here?
03:48:41funmanunfortunately i can't load rockbox yet, a 1kB file loads but not the 400 or so kB rockbox.sansa
03:48:47funmanelinenbe: hi, it's me
03:49:15elinenbeAh, yes, I was just looking up your nick on the site!
03:49:21elinenbeQuick question...
03:49:39taksatoi fgound a thread on google search and it says missing or i am not allowed to see it
03:49:41elinenbeThe recent commits you made −− they are for Clip V2 and + ?
03:50:01funmanyes, but only tested on Clip+
03:50:13elinenbeso, transfers now work on the Clip+?
03:50:26elinenbeThat's what I'm currently using right now.
03:50:27funmansomehow, i can only read small files
03:51:08funmanso the code is alive but still very buggy
03:51:30elinenbeokay. So I shouldn't expect to use this as my primary f/w for anytime soon ;-)
03:51:46elinenbeOh, one other thing... is this just for internal memory?
03:52:08funmanright now i'm only testing internal memory but sd slot should work one day
03:52:55elinenbeWell, yes, that would be the goal, but thank you for the answers!
03:53:09elinenbeSounds good, and looking forward to that day!
03:53:25taksatodo you think the thread got deleted or i am having an access problem?
03:53:53saratogaelinenbe: if you want to use rockbox on your player you might want to wait for the commit saying audio works to start asking questions
03:54:54taksatothis thread looks very relevant to giving background info that would also help my effort
03:55:07saratogawhat thread
03:55:10funmansaratoga: hopefully there will be no such commit if the as3525 code can be reused as is
03:55:30taksatocan i privmsg you?
03:56:33 Quit JdGordon_ (Ping timeout: 248 seconds)
03:57:31 Quit taksato (Remote host closed the connection)
03:57:58 Join taksato [0] (
03:58:16funmanperhaps there is an extra step needed for wide bus mode on clip+
03:58:16Unhelpfulfunman: still need me to try your clipv2 iram suggestion?
03:58:40 Join JdGordon_ [0] (
03:58:43funmanUnhelpful: yep, if the clipv2 has no iram it will need a special map
04:00:12Unhelpfulso add a CGU_PERI |= (1<<25) before the test?
04:00:22funmanyes, it should enable the "1TRAM"
04:01:02 Quit JdGordon_ (Read error: Connection reset by peer)
04:01:34elinenbesaratoga: Sounds good! How far along is the audio progress?
04:01:54saratogaelinenbe: see above
04:02:35saratogathe linux patches mention SRAM rather then TRAM so it sounds like IRAM is different on teh 353x at least
04:02:50 Join JdGordon_ [0] (
04:03:13Unhelpfulfunman: still hangs after attempting to write to 0x81000018.
04:03:53funmansaratoga: keep in mind we have no as353x targets
04:03:56funmanUnhelpful: :/
04:04:19saratogayeah i know, and looking closer the limit is still 66MHz so its probably the same hardware
04:04:22Unhelpfulno iram? or at least not where we think?
04:04:34saratogawhat address are you trying?
04:05:07 Quit YPSY (Ping timeout: 272 seconds)
04:05:16 Quit JdGordon_ (Read error: Connection reset by peer)
04:05:27taksatosaratoga, thank you for the thread. i am done reading it in case you wanted to move it again
04:06:50 Join JdGordon_ [0] (
04:06:52Unhelpfulsaratoga: the test code tries to write starting at 0x81000000. not sure where funman got that address though, he's kind of doing the development on my hardware.
04:07:43 Quit TheSeven (Disconnected by services)
04:07:55 Join The_Seven [0] (~theseven@rockbox/developer/TheSeven)
04:07:58saratogathe 35 defines 0x00100000 as the SRAM base, so that might be worth looking at too
04:08:06 Nick The_Seven is now known as TheSeven (~theseven@rockbox/developer/TheSeven)
04:08:42funmanUnhelpful: the address is in the memory map of as3525 described in the datasheet: there's an alias at 0xc1000000
04:09:25 Quit Barahir_ (Ping timeout: 265 seconds)
04:09:33 Quit anewuser (Quit: =ooo ϢINTER ϾHIP 5iVE is OOON!! omg 7 days left =oo)
04:09:56kugelJdGordon: ping
04:11:47kugelJdGordon: can you have a quick look at this patch? it forms the fond_ids array into a linked list, so that the actual loading can happen later at parsing (I'm needing that for resizing on boot)
04:13:10S_a_i_n_tiPod Nano1g's main stack size is defined in "rockbox\firmware\target\arm\ipod\"...correct?
04:14:37kugelif you're on windows, then yes :)
04:15:42kugelS_a_i_n_t: err, no
04:16:08 Quit mc2739 (Ping timeout: 260 seconds)
04:17:17kugelS_a_i_n_t: excuse me, I'm too tired to read properly. your first path was correct (I read nano2g..)
04:18:05 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
04:19:01 Quit kugel (Remote host closed the connection)
04:19:31 Quit Strife89|PalmTX (Quit: Bed.)
04:20:20 Quit flyback (Excess Flood)
04:20:45S_a_i_n_tNo worries, I figured it out for Nano2g (well, I asked TheSeven...) earlier :P
04:20:47 Join flyback [0] (
04:21:21S_a_i_n_tAny idea *why* pictureflow is overflowing now?
04:26:22 Join iq [0] (~iq@unaffiliated/iq)
04:26:27 Join iq_ [0] (
04:26:44 Quit iq_ (Remote host closed the connection)
04:27:21 Quit JdGordon_ (Read error: Connection reset by peer)
04:28:07 Join DerPapst1 [0] (
04:28:25taksatodid anyone write mi4code.c for e200 series?
04:29:12 Quit DerPapst (Ping timeout: 260 seconds)
04:31:24saratogacheck the SVN logs
04:31:38 Join teru [0] (
04:31:59 Quit iq (Remote host closed the connection)
04:33:16 Join fyrestorm [0] (
04:34:22 Join BHSPitMonkey [0] (~stephen@unaffiliated/bhspitmonkey)
04:35:45 Quit taksato (Ping timeout: 256 seconds)
04:36:56S_a_i_n_tIf anyone cares...the keymap for Pegbox on Nano2g is totally wrong. menu+select changes level, not brings up the in-game menu as the manual suggests. menu+play is the *actual* ingame menu.
04:41:47S_a_i_n_t*errr....whoops, I totally read it wrong.
04:46:53*S_a_i_n_t would *really* like a manual entry for SuperDomination...but guesses he'd need to figure out how to play it before he could even *attempt* to write one.
04:58:08***Saving seen data "./dancer.seen"
05:00:09 Join Barahir [0] (
05:02:10 Join anewuser [0] (anewuser@
05:02:10 Quit anewuser (Changing host)
05:02:10 Join anewuser [0] (anewuser@unaffiliated/anewuser)
05:02:29 Quit veeloc (Quit: is idle)
05:12:24 Join Razimel [0] (
05:12:46RazimelDamn im angry
05:14:38 Quit panni_ (Quit: ( :: NoNameScript 3.81 :: ))
05:15:38RazimelI was working on five different wps and themes for the ipod nano 2g, and what do you know. after two week s of modifying and testing.
05:16:15RazimelI wake up to find that the ipod would not enter iloader
05:16:32Razimeland it kept telling me I had to recover it using itunes
05:16:34 Join webguest83 [0] (
05:16:35 Quit webguest83 (Client Quit)
05:16:49Razimelfuck that I thought
05:17:03Razimel so I just reformatted and reinstalled rockbox
05:17:21Razimelbut I hadn't backed up my most recent themes
05:17:31Razimelso they're gone now
05:17:53Razimel, but im glad I at least saved two of em
05:18:00RazimelLesson learned
05:18:10RazimelBackup constantly
05:18:47RazimelThe first theme is Cowboy bebop
05:19:00Razimelwith ed and ein as the main backdrop
05:19:09Razimeland Faye in the Wps
05:19:52 Quit avacore (Ping timeout: 246 seconds)
05:20:06Razimelthe other was
05:20:09Razimela doom theme
05:21:19Razimelwhich had an awesome wps with a commando zombie
05:21:35Razimeland the status bar would fill up as a belt of bullets
05:22:04 Join avacore [0] (
05:22:59Razimelthe boxes wre black with red outlines giving it a badass neo-goth apppearance
05:23:09Razimelbut it's gone now
05:23:27RazimelI even had an experimental francesca dani theme
05:23:36Razimeloh well
05:23:47Razimel"back to the drawing board"
05:24:09saratogaRazimel: you should probably check out rockbox-community if you want to chat about things like that
05:24:47 Join veeloc [0] (
05:24:57 Quit veeloc (Client Quit)
05:26:58 Quit CGL (Quit: Saliendo)
05:27:06 Quit Razimel (Quit: CGI:IRC (EOF))
05:27:22 Join Razimel [0] (
05:28:12 Join veeloc [0] (
05:32:05 Quit Razimel (Client Quit)
05:32:22 Join Razimel [0] (
05:32:33 Join toffe82_ [0] (
05:32:36 Quit Horschti (Quit: Verlassend)
05:33:04 Quit moos (Quit: ChatZilla 0.9.86 [Firefox 3.6/20100115144158])
05:33:07 Quit Razimel (Client Quit)
05:33:37 Quit toffe82 (Ping timeout: 248 seconds)
05:34:42 Join CGL [0] (~CGL@
05:37:27 Quit veeloc (Quit: is gone)
05:43:08 Quit DerPapst1 (Quit: Leaving.)
05:46:00 Join mt [0] (~mtee@rockbox/developer/mt)
06:05:09 Quit CGL (Remote host closed the connection)
06:13:32funmanclip+ bootloader is stuck in disk_mount_all()
06:16:13CIA-8New commit by 03teru (r24841): fix typo of wps and sbs.
06:22:24 Join Ypsy [0] (
06:22:33 Nick Ypsy is now known as YPSY (
06:26:22funmanstill this mysterious "HW LOCKED WHILE ERR"
06:41:07CIA-8New commit by 03funman (r24842): Clip+ boots with iram mapped at 0x0, only the clipv2 is different
06:41:10CIA-8New commit by 03funman (r24843): mkamsboot: use properly
06:42:02funmanUnhelpful: remember what was the error on your Clipv2: ATA error -4 ?
06:50:28 Quit S_a_i_n_t (Quit: *someone* needs to use the rude.)
06:58:12***Saving seen data "./dancer.seen"
07:00:51CIA-8New commit by 03teru (r24844): correct checking size in function get_image_filename.
07:03:03 Join CGL [0] (~CGL@
07:04:52 Part toffe82_
07:12:51 Quit liar (Quit: Verlassend)
07:17:52ThomasAHfunman: the commits look promising ... anything new to test? (but now away for breakfast -> shower -> work)
07:19:09 Join S_a_i_n_t [0] (S_a_i_n_t@
07:21:17funmanThomasAH: nope, still looking why it doesn't work anymore
07:21:24S_a_i_n_tRB compatible file extensions...have I missed any? −−>
07:22:02funmansaratoga: .aac isn't handled ?
07:22:23saratogafunamn: no, just m4a/mp4/etc
07:22:34 Quit anewuser (Quit: =ooo ϢINTER ϾHIP 5iVE is OOON!! omg 6 days left =oo)
07:22:36saratoga.aac is a raw aac stream, I don't think we can parse those
07:23:12saratogai'm not sure they're entirely standard, sort of like making a raw .vorbis stream without putting it in ogg
07:23:29funmanadts & latm muxing are specified in the standard
07:23:42funmannot sure if there are a lot of raw files in the wild though
07:23:56S_a_i_n_taac wasn't in rockbox\apps\filetypes
07:24:08S_a_i_n_t...but neither are a few of those extensions...
07:24:09saratogadon't you need the container to tell you sample rate or something
07:24:30saratogai thought there was a good reason not to use, but i can't remember clearly why i thought that
07:25:41S_a_i_n_tmy ultimate goal is to give each filetype an as to not see gaps or black squares when viewing "All Files" etc.
07:26:54S_a_i_n_tif I just use " *: 0 " in an .icons file...will it catch anything I havent specified?
07:27:34*S_a_i_n_t assumes the * wildcard will work, but isn't positive.
07:27:52 Join stoffel [0] (
07:28:00funmansaratoga: no it should be in here
07:28:09funmanadts & loas are designed for transport though, not storage
07:29:02saratogathough maybe encoders stick that info in somewhere unofficially
07:29:03funmanah there is only channels / bitrate
07:29:12funmanit would be transmitted by the transporter
07:31:12saratogayeah that makes sense
07:31:47 Quit CaptainKewl (Remote host closed the connection)
07:32:47 Quit saratoga (Quit: Page closed)
07:34:56 Join taksato [0] (
07:34:59 Join shai [0] (
07:36:19taksatoMore update on the serial issue
07:36:45taksatoBoth the addresses used by ipod are shown as always zero on the sansa e200 series v1
07:37:51taksatoSo I dont think it is properly unsetting the bit it needs to unset in serial initialisation and hence not working. no word on why the thing reads an 0xFF as first char when one of the init routines is used. Doing a memory dump in debug screen before chasing that down, because i think that would be wasting time to chase down. the 0xFF thing
07:40:03taksatoweird thing is that up and down are both going to next page on my memory output screen
07:47:29S_a_i_n_tis there nowhere I can find a *FULL* list of RB supported .ext's? I've got *most* of them...but target specific ones (like .iaudio, .iriver etc.) remain elusive...any suggestions? Feel free to add to my pastebin from earlier.
07:47:35S_a_i_n_tIn fact, please do :)
07:49:56JdGordonapps/filetypes.c will have every one
07:50:03JdGordonexcept the ones in viewers.config
07:52:18*S_a_i_n_t doesn't see quite a few in apps\ list seems a LOT longer actually, and I know I'm missing quite a few.
08:00:21JdGordonteru: your last change is good, but annoying at the same time :)
08:01:26S_a_i_n_tthe ones I'm stuck on ( 'ones' being .ext's) are target specific, like .ajz, .iaudio, .iriver etc. and without going through each and every manual...I can't see a way to find them all.
08:03:42JdGordongrep BOOTFILE firmware/export/configs/*.h
08:05:18CIA-8New commit by 03jdgordon (r24845): fix remote WPS loading
08:07:00taksatohm, seems the wiki is wrong
08:09:19taksatohere says e200 series is PP5024, but my v1 e250 at mem location 0x70000004 says 0x50503530 (PP50) and 0x70000000 says 0x32324320 (22C ) combining is PP5022C
08:10:24funmaniirc PP5024 is written on the chip
08:11:40taksatoversion C of 22 must be 24 then? else, someone forgot to update their version.h before shipping ;)
08:12:56amiconnThe Sansa c200v1 also says PP5022C, and has PP5022 written on the chip
08:13:28amiconnThe PP5024 is a PP5022 with an integrated AS3514
08:13:57taksato0x70000100 and 0x70000104 says same thing PP5022C
08:14:25funmanso is PP5024 a rockbox invention?
08:14:48taksatodont know, just reading out my mem dump for discussion
08:14:57amiconn[08:13:29] <amiconn> The PP5024 is a PP5022 with an integrated AS3514
08:15:25taksatoi saw the data sheet, so it is a product like amiconn mentioned
08:17:18CIA-8New commit by 03teru (r24846): some changes to use of display_text. ...
08:18:34S_a_i_n_tJdGordon: thanks SOOO much, my naive Windows brain made me forget entirely about grep
08:18:39S_a_i_n_t*slaps forhead*
08:18:42taksatoevery 0x100 they seem to repeat/alias themselves
08:19:09funmanhmm rockbox.sansa loads but checksum is off by 2
08:19:44CIA-8New commit by 03jdgordon (r24847): do r24817 betterer to make it more generic and stuff. Skins will always be in the .rockbox/wps/ dir so there is no real need to add lots of duplicated ...
08:24:06 Join bmbl [0] (~Miranda@unaffiliated/bmbl)
08:24:22 Join ender` [0] (
08:25:14CIA-8New commit by 03jdgordon (r24848): fix charcell
08:27:43funmanscramble.c / config/sansaclip*.h are not accorded together
08:29:27 Join Zagor [0] (
08:29:27 Quit Zagor (Changing host)
08:29:28 Join Zagor [0] (~bjst@rockbox/developer/Zagor)
08:30:26*JdGordon sends a virtual nose punch kugels way
08:30:49JdGordonwtf did you do r24796 for?
08:31:38CIA-8New commit by 03funman (r24849): as3525v2: init sdram
08:31:42CIA-8New commit by 03funman (r24850): Fix model number for Clipv2 & Clip+ (same than in scramble.c)
08:31:46CIA-8New commit by 03funman (r24851): sd-as3525v2: working bootloader ...
08:34:24funmanhow do i add clipv2 & clip+ to the build table ?
08:34:42funmanbootloader & normal build compile for both without warnings
08:39:12taksatohmm, nada for memory changes in suspected places when placed on dock and then pressing remote buttons
08:40:35funmanso clip+ boots but i got nothing but a black screen, perhaps incompatibilities with i2c registers layout
08:44:28JdGordondoes anyone have a problem with;r2=24847;pathrev=24847 ? does it take pointer hacks a bit too far for what it does?
08:45:05 Join Speedy2 [0] (
08:46:13Mode"#rockbox +o rasher" by ChanServ (ChanServ@services.)
08:46:16Mode"#rockbox +b *!*mike@*" by rasher (~rasher@rockbox/developer/rasher)
08:46:16Kick(#rockbox Speedy2 :Speedy2) by rasher!~rasher@rockbox/developer/rasher
08:46:33Mode"#rockbox -b *!*mike@*" by rasher (~rasher@rockbox/developer/rasher)
08:46:47Mode"#rockbox +b *!*" by rasher (~rasher@rockbox/developer/rasher)
08:46:57 Join flydutch [0] (
08:47:17Mode"#rockbox +b speedy2!*@*" by rasher (~rasher@rockbox/developer/rasher)
08:47:35rasher(seen with the same nick, on another channel on another network)
08:48:22JdGordonhe's been here before also
08:58:08taksatono gpio changes on docking
08:58:16***Saving seen data "./dancer.seen"
08:59:08taksatono gpio changes when docked and remote buttons pressed
09:02:09 Quit bmbl (Ping timeout: 272 seconds)
09:06:17 Quit shai (Quit: Leaving)
09:06:38 Join shai [0] (
09:12:50 Quit S_a_i_n_t (Quit: There are 10 types of people, those who understand binary, and those who don't.)
09:13:34 Join S_a_i_n_t [0] (S_a_i_n_t@
09:14:00 Join petur [0] (~petur@rockbox/developer/petur)
09:15:46 Quit taksato (Read error: Connection reset by peer)
09:20:28 Nick S_a_i_n_t is now known as S_a_i_n_t_ (S_a_i_n_t@
09:21:02 Nick S_a_i_n_t_ is now known as S_a_i_n_t (S_a_i_n_t@
09:27:04 Join Grahack [0] (
09:28:08 Quit TopyMobile (Ping timeout: 252 seconds)
09:28:56CIA-8New commit by 03funman (r24852): Print correctly the rockbox logo for Clipv2/Clip+, not only Clipv1
09:29:01CIA-8New commit by 03funman (r24853): crt0: init correctly as3525v2
09:29:04CIA-8New commit by 03funman (r24854): as3525v2: do not modify CGU_PERI in rockbox.sansa ...
09:29:45funmansd doesn't work in rockbox.sansa :o
09:33:44 Join liar_ [0] (~53af7f67@gateway/web/freenode/x-cyunhlgttdyvjqip)
09:37:27 Join kaniini [0] (
09:37:43S_a_i_n_tfunman: only 6 more to go 'til can do it! :p
09:41:12*Bagder teaches S_a_i_n_t about how 60 comes after 59, I fucked that one up massively.
09:47:32JdGordonits not liked this channel is logged and anyone will know...
09:48:31S_a_i_n_tsomehow I saw 248*9*4
09:49:11 Quit petur (Quit: reboot)
09:49:35 Quit perfectdrug (Remote host closed the connection)
09:52:10 Join petur [0] (~petur@rockbox/developer/petur)
09:54:58funmanprobably timing related, it works with dcache/icache disabled
10:07:37 Quit shai (Ping timeout: 252 seconds)
10:13:52 Join pondlife [0] (
10:13:54 Quit pondlife (Changing host)
10:13:54 Join pondlife [0] (~Steve@rockbox/developer/pondlife)
10:20:01 Quit phanboy4 (Ping timeout: 248 seconds)
10:21:56funmanok i can make it to the menu but i see weird screen corruption and buttons do not work reliabily
10:35:17*funman plays chopper on clip+
10:39:04dionoeafunman: does that mean that you've got a working sd driver ?
10:39:41funman'working' is a bit overrated
10:40:05funman'somehow alive' is more descriptive :)
10:40:15dionoeahehe ;)
10:40:59 Join pamaury [0] (
10:58:20***Saving seen data "./dancer.seen"
11:02:07CIA-8New commit by 03funman (r24855): Clip+: fix buttons driver ...
11:02:12CIA-8New commit by 03funman (r24856): as3525v2: show I/O ports in debug menu
11:02:17CIA-8New commit by 03funman (r24857): as3525v2: use iram (might need to be disabled for clipv2)
11:02:22CIA-8New commit by 03funman (r24858): sd-as3525v2: use an uncached buffer, disable write for the moment
11:04:04 Join einhirn [0] (
11:12:42 Part pondlife
11:20:39 Join shai [0] (
11:22:03ThomasAHfunman: builds, logo shows, but then a PANIC occurs: wait for state faile... (IDENT)
11:22:25funmanThomasAH: the bootlaoder?
11:22:48funmani need to disable dcache & icache in mmu-arm.S to get rockbox.sansa working
11:23:22ThomasAHfunman: I built and installed the bootloader and built and unpacked
11:23:39ThomasAHfunman: after pressing this or that key, "ATA error: -2" appears
11:23:50funmanATA error: -2 is an error in rockbox.sansa
11:24:14funmantry with
11:25:04ThomasAHfunman: only rebuild
11:25:04CIA-8New commit by 03funman (r24859): Clip+: show the same real time in OF and rockbox
11:25:28funmanor even rockbox.sansa (with make bin and copy rockbox.sansa to .rockbox
11:26:13funmansound doesn't work, write doesn't work, codecs crash randomly, sd read doesn't work with caches
11:26:18funmanbattery looks ok
11:26:26funmangreylib asm for clip+ to be done
11:28:09ThomasAHstill at "Refreshing your media" ...
11:28:20 Quit BHSPitMonkey (Remote host closed the connection)
11:30:15ThomasAHumount ...
11:30:56ThomasAHargl ... again at "Refreshing your media" ... about 14GB of music and audio books
11:31:42funmanfaster to remove the µSD
11:35:52CIA-8New commit by 03funman (r24860): Clip+: greylib
11:35:59funmanS_a_i_n_t: r24900 ;)
11:36:58ThomasAHfunman: hmm, bad checksum
11:37:18ThomasAHand varying model name
11:37:35funmanvarying how ?
11:38:14ThomasAHsometimes cli+ (no p!), sometimes special char, z <
11:38:23funman'cli+' is alright
11:38:25ThomasAHah, now it booted!
11:38:31ThomasAHmenu! :)
11:41:27ThomasAHbut more often bad checksum than not
11:41:46funmanbad things (tm)
11:42:13ThomasAHbut chopper worked :)
11:42:58funmanyeah that's all we need ;)
11:44:06ThomasAHbut now I have to work a bit again :)
11:44:08gevaertschopper is actually pretty easy on the clip :)
11:47:41 Quit stoffel (Remote host closed the connection)
11:50:44 Join MethoS- [0] (~clemens@
11:50:47 Join efyx_ [0] (
11:51:36 Join TheSphinX^ [0] (
11:56:26 Join kugel [0] (~kugel@rockbox/developer/kugel)
11:56:33kugelgevaerts: it's not!
11:56:45kugelJdGordon: did you look at my patch?
11:56:58kugelfunman: awesome job! \o/
12:02:16funmankugel: it's much faster with docs :)
12:02:35funmannext steps: fix SD / get sound
12:02:41 Join topik [0] (
12:03:05topikgreat work so far, funman
12:04:25kugelfunman: hopefully we can use the as3214 code?
12:04:51 Nick evilnick_ is now known as evilnick (
12:04:51funmani suppose so, i made a quick&dirty try to use LSP (speaker) instead of HPH (headphones)
12:05:02 Quit evilnick (Changing host)
12:05:02 Join evilnick [0] (~evilnick@rockbox/staff/evilnick)
12:05:15funmandoes bertrik have a clipv2 or clip+ ?
12:05:51kugeldamn, I have pretty much disassembled button and lcd for the fuzev2 but I don't get to doing something
12:07:17funmanperhaps the priority should be understanding clocking code
12:08:01funmanunfortunately it isn't documented in the linux patches
12:11:17 Join dfkt [0] (dfkt@unaffiliated/dfkt)
12:14:45 Quit funman (Quit: free(random());)
12:15:07topiki'm ordering my clip+ today. i know it's not finished but it's great to have rockbox on a dap that's still actively sold
12:19:13gevaertstopik: if funman continues like this, it won't be actively sold for much longer!
12:20:29topiktrue, it's doomed this way
12:20:59*S_a_i_n_t waits for a day when DAP's come with a Rockbox install disk in the packaging also...when they finally realise that Open Source projects are just....better.
12:21:24topikbetter is relative
12:21:25 Join watto [0] (~watto@
12:21:39 Join Farthen [0] (
12:21:43topikmore free (as in speech) for sure
12:23:17S_a_i_n_timagine how good RB would be if the devs got the same pay-packet that the Sansa, or iPod devs do? or if it was fulltime instead of free-time
12:23:26S_a_i_n_tit'd be awesome(er)
12:23:27gevaertsit would be horrible
12:23:59 Quit liar_ (Ping timeout: 252 seconds)
12:24:18 Quit kugel (Read error: Operation timed out)
12:24:49gevaertsYou'd have a two-class system for starters, with some people being paid, some not. Those who are paid (some of them at least) would assume that they are more important
12:25:06 Quit evilnick (Quit: Leaving)
12:25:26gevaertsLots of what makes rockbox fun is that *everyone* does it for fun
12:25:34AlexPAnd things would happen because the person paying wanted them, not because it was best overall
12:25:37 Quit lyngaas (Ping timeout: 240 seconds)
12:25:48S_a_i_n_tHmmm...I see your point. I guess I was imagining more of a "eutopian" environment there
12:25:52 Join kugel [0] (~kugel@rockbox/developer/kugel)
12:27:14topiki am almost sure that more people are working on rockbox for sansa players than sansa has developers themselves
12:27:45kugeldepends on how you count the rockbox guys working on them
12:27:58topikif they were willing to supply docs for the hardware it would make a huge difference already
12:28:06AlexPYeah, I don't count kugel at all :)
12:28:22S_a_i_n_tthey gave you a JTagged unit didn;t they?
12:28:25topiki know kugel, but from the rate of development and response to bugs it can't be more than 1-3 at sandisk.
12:28:25BagderI don't think sandisk had/has a lot of docs, I think they use APIs provided by the chip guys
12:28:27S_a_i_n_tthat was nice...I guess.
12:28:32AlexPnot really
12:33:21 Join evilnick [0] (~evilnick@rockbox/staff/evilnick)
12:38:59 Join DerPapst [0] (
12:42:12 Join ecio1 [0] (
12:42:12 Quit ecio (Read error: Connection reset by peer)
12:44:35topiki think you scared funman away now, gevaerts. sandisk will be pleased they can keep selling their little dap.
12:45:00 Join perfectdrug [0] (
12:57:30perfectdrugfunman is gone? any other clip+ owner around? I would really like to finish this:
12:58:00Bagderoh nice
12:58:24***Saving seen data "./dancer.seen"
12:58:59perfectdrugbuttons are missing and I don't exactly know where to put them
12:59:46topikno pics on the web perfectdrug as inspiration?
13:00:28perfectdrughm maybe...
13:00:29 Quit parafin (Ping timeout: 256 seconds)
13:00:50topikabi's review has a lot
13:01:35 Join parafin [0] (
13:02:05perfectdruguh that's nice thanks
13:02:45 Join TopyMobile [0] (
13:03:04 Quit evilnick (Remote host closed the connection)
13:04:04perfectdrugthis thing is reaaaly small :-O
13:04:15ThomasAHperfectdrug: has some more photos, too ... and I have the real device here if you have more questions
13:04:54 Join evilnick [0] (~evilnick@rockbox/staff/evilnick)
13:06:43perfectdrugThomasAH: so as I see it there is in the right top round powerbutton and on the left is volume up and down (as I guess) with volume up at the top? and no hold switch is this right?
13:07:08ThomasAHperfectdrug: right (Home is used as hold)
13:07:16topikthe disassembly is nice if you want to model the underside of the buttons too
13:07:20ThomasAH"right" as in "correct" :)
13:07:55topikunrelated question, but ThomasAH, does the clip+ keep its display on while charging (OF) ?
13:07:57ThomasAHperfectdrug: and the volume up/down rocker is centered at the sansa logo
13:08:38perfectdrugtopik: :D ThomasAH: how does the hold with the home button work? should I label it Home/Hold?
13:08:47ThomasAHtopik: Not while charging (but it stays on), only if hooked to a computer (mass storage)
13:09:28topikprobably the biggest non-feature of rockbox on my fuze i appreciate most is that the display ALWAYS goes off when you switch it to 'hold
13:09:40ThomasAHperfectdrug: pressing it for some seconds triggers the button lock
13:09:59Unhelpfulfunman: i believe it was ATA error -4
13:10:00ThomasAHperfectdrug: and no, the house it quite fine as this is on the original device, too
13:10:11ThomasAHUnhelpful: funman left
13:11:06ThomasAHperfectdrug: and the power button is right-aligned with the right display edge
13:11:28ThomasAHi.e. right side of power button with right edge of display
13:13:17perfectdrugok nice, will do
13:13:19perfectdrugbtw how does the clip v2 look? needs it any different image for the manual?
13:13:26UnhelpfulThomasAH: channel's logged. ;)
13:13:51ThomasAHperfectdrug: IIRC it looks much differently
13:14:42 Join perfectdrug_ [0] (
13:15:13 Quit parafin (Ping timeout: 256 seconds)
13:15:44ThomasAHperfectdrug: circle d-pad instead of just rounded corners, recessed upper and lower side
13:15:55ThomasAH(quite ugly if you ask me :))
13:18:39 Quit perfectdrug (Ping timeout: 272 seconds)
13:18:57 Quit kugel (Ping timeout: 265 seconds)
13:19:42perfectdrug_ThomasAH: yes I mean wrt the existing clipv1 image:)
13:20:37ThomasAHperfectdrug_: ah, not that I know of (and people wanting to tell the difference always had to look at the firmware revision)
13:23:07 Join lyngaas [0] (
13:29:13 Join ecio [0] (
13:30:33 Quit TopyMobile (Read error: Operation timed out)
13:30:45 Quit ecio1 (Ping timeout: 252 seconds)
13:31:50TheSevenS_a_i_n_t: pictureflow is probably using a little more stack space than other things, and the fat driver needs a little more space on nano2g than on other devices, and the combination of that is just a little too much...
13:33:56S_a_i_n_tawesome, I love belated answers...well, I love trying to figure out what the question I asked was in the first place.
13:34:13S_a_i_n_tI've upped the stack size, and all is well :D
13:36:40 Join flide [0] (~flide@
13:38:08 Quit flide (Remote host closed the connection)
13:38:37 Join flide [0] (~flide@
13:40:59 Nick flide is now known as fLiDe (~flide@
13:42:47 Quit mt (Ping timeout: 252 seconds)
13:43:09perfectdrug_ThomasAH: can you please check if this is correct now ? I put it on flyspray then.
13:44:41 Join _flide_ [0] (~flide@
13:45:02_flide_ hello people, I bought a Transcend MP840 mp3 player about 6 months ago but its firmware sucks and I heard about this rockbox
13:45:09 Join mt [0] (~mtee@rockbox/developer/mt)
13:45:09_flide_ but I don't see my player being listed on the rockbox homepage
13:45:21_flide_but I am good at programming
13:45:24S_a_i_n_tit isn't....thats why
13:45:53 Quit CGL (Ping timeout: 272 seconds)
13:45:53S_a_i_n_t*isn't supported tht is
13:45:55_flide_so is there any way I can modify rockbox to suit my pmp
13:45:58 Join b0hoon [0] (
13:46:09 Quit fLiDe (Ping timeout: 252 seconds)
13:46:12 Join b0hoon2 [0] (
13:46:37 Nick _flide_ is now known as fLiDe (~flide@
13:47:59fLiDethanks saint, I felt lost in the room (this is my first time on an irc channel :))
13:48:09 Join TopyMobile [0] (
13:48:44b0hoon2hi, i have a request, the small picture on the web site has an ugly white corners (pb vibe), can someone replace it, please? The fixed one is on the flyspray: FS #10947.
13:49:51perfectdrug_^along with please:)
13:50:22S_a_i_n_tfLiDe: ah, well...the link i posted is the best wy to start, you'll findout *pretty* quickly that a ne port is a massive job
13:50:48fLiDeyes that i see
13:51:00 Quit b0hoon (Quit: CGI:IRC (Ping timeout))
13:51:15fLiDeit is huge....
13:51:27perfectdrug_Zagor: maybe
13:52:18b0hoon2also last time i forgot about to mention about adding automatic sim build and the checkWPS (i have never used it before but anyway if it's necessary it can be added, it builds clean).
13:53:00S_a_i_n_tfLiDe: if you don't have a good grounding in C programming language, start there.
13:53:19S_a_i_n_tas you'll need it.
13:53:40b0hoon2I have spotted that something is wrong with the deltas table after pb vibe was added.
13:53:47 Quit mt (Ping timeout: 252 seconds)
13:54:53fLiDesaint, I already said I am good at programming
13:55:00fLiDeso that won't be a trouble
13:55:27 Join froggyman [0] (
13:55:53S_a_i_n_tthat doesn't necessarily mean C though... ;D
13:56:07 Join mt [0] (~mtee@rockbox/developer/mt)
13:56:52 Join old-landen [0] (
13:57:25fLiDeit is c in this case...
13:57:37fLiDe... in my case...
13:57:46TheSevenb0hoon2: wait until it makes its way down, then it'll be fine again :-)
13:58:37TheSevenfLiDe: ever done programming on embedded systems before?
13:58:55fLiDeTheSeven, no.... :(
13:59:10TheSevenand first of all: are there firmware updates for this player, what kind of CPU is it using, and are those updates unencrypted?
13:59:19TornefLiDe: so you'll discover how much fun it cna be to have no documentation, no debug facilities, etc :)
13:59:51TheSevenfLiDe: ever done assembly programming / reading disassemblies / reverse engineering?
13:59:58S_a_i_n_t...don't scare the boy off :D
14:00:06fLiDeupdates? ok I must search it...
14:00:32fLiDeoohhh man, you people are reaaaaalllly scaring me
14:00:47fLiDeI get it, it's not easy
14:00:49b0hoon2TheSeven: If you mean the last unnamed column?
14:00:49fLiDebut I will try
14:00:52S_a_i_n_tits hard work.....realy.
14:01:38TheSevenb0hoon2: yes
14:01:56TheSeventhe title for the missing one will turn up when the column for it has made it all the way down
14:02:22fLiDeok, I will start step-by-step, but I don't want to brick my pmp, so how long can I avoid breaking it?
14:02:35TheSevenfLiDe: I can't find any pictures of the inside of this thing on the web :-/
14:02:37S_a_i_n_tbe careful
14:02:54 Join anewuser [0] (anewuser@unaffiliated/anewuser)
14:02:56TheSevenfLiDe: that totally depends on their firmware update and recovery mechanisms
14:03:09TheSeven(apart from mechanical damage, of course :-D )
14:03:12TornefLiDe: there's a firmware on their website, which if you unpack it enough has a bunch of .bin files in it
14:03:19Tornelike.. 130+ of them
14:03:25Tornemostly 30kb or so each :)
14:03:29Torneif that's the OS then, er, hmm.
14:03:35TheSevenlet me check
14:03:37Torneit looks like it is though
14:04:04fLiDeyou people are fast... really fast...
14:04:04Torneheavy use of overlays?
14:04:17S_a_i_n_t130+ .bins for fw? ....yuck
14:04:29fLiDewhat are overlays?
14:04:31Torneit's in an exe, which updates the firmware i guess
14:04:36Tornebut 7zip can open it as an archive
14:04:38b0hoon2TheSeven: ok, thanks for answer.
14:05:11Tornecan't really tell you anything much else form looking except "it's not unencrypted ARM code"
14:05:24mc2739b0hoon2: the last unamed column is the average for all builds
14:06:05fLiDeok, "not unencrypted arm code" ... check
14:06:48gevaertsDoes it compress well? If so, it's probably not encrypted code
14:07:24perfectdrug_task added:)
14:07:49Torneyup, they all compress by ~50%
14:07:54Torneso yah, it's probably not encrypted
14:08:07Torneso it's probably just not ARM
14:08:21Tornei can't recognise any other instruction set from just a hexdump
14:08:56fLiDewhat i get from here, suggests I am a total noob....
14:08:58b0hoon2mc2739: heh, i could swear that it wasn't there before ... :)
14:09:10S_a_i_n_tdo we know the processor in this target? i can't find much documentation for it.
14:09:13TornefLiDe: the easiest way to figure out what it is is to take it apart
14:09:19Torneand take hi res scans of the board
14:09:35fLiDeyou serious?
14:09:41TorneOf course
14:09:48TorneGoogle doesn't seem to know anything about it
14:09:55Torneand it's not immediately obvious to *me* what the code is
14:10:34Tornehere's a hexdump of 000INIT.BIN
14:10:37fLiDeactually, it has no screws or any thing , so I guess i would take some forcing and .... time
14:10:47Torneif anyone else can perhaps spot what arch that might be :)
14:11:36S_a_i_n_tfLiDe, you'll need something like "iTool", or a "spudger" to open it.
14:11:38Tornetry disassembling that for thumb/mips/sh/etc
14:11:46Tornemight be plausible code for one of them ;)
14:11:49*gevaerts makes an uneducated guess without looking at the hexdump and says: Z80
14:12:35TheSevenwhatever this is, it has 40-bit instructions...
14:12:52Tornewait what?
14:13:19TheSevenbytes like to repeat at modulo-5=0 offsets in that file
14:13:54TheSevenTorne: it's neither arm nor thumb for sure
14:13:59TorneTheSeven: yah
14:14:06Torneit might not be literal code at all
14:14:12Tornebut it's a good guess
14:15:03TheSeventake a look at the occurrences of 0x90 bytes
14:15:48pamaury130+ likes file makes me think about EFI and there pseudo assembly language
14:16:13pamauryBut I'm a complete noob :)
14:16:43fLiDeok, before I "disassemble" my pmp, is there any way I can extract the firmware from within, that will for sure remove the doubt of os
14:16:49TheSevenpamaury: then it would have PE headers
14:17:06TornefLiDe: not in any standard way
14:17:07pamauryTheSeven: if you say so... I believe you !
14:17:31Tornewhat type of usb device is it?
14:17:32linuxstbTorne: Where did you get that firmware from? The only file I could find was a .zip which contains a Windows .exe, which I can't run...
14:17:37Tornelinuxstb: the windows exe
14:17:44Torneopen it in a suitably powerful archiving tool
14:17:54Torneit has a zip or something embedded
14:17:58linuxstbAh, unzip wasn't suitably powerful...
14:18:02Torneuse 7za
14:19:09TheSevenfLiDe: which os are you using?
14:19:27fLiDelinux ubuntu
14:19:49TheSevenfollow the firmware update procedure, just don't run the tool, then do lsusb
14:20:08TheSeveni.e. once while holding the buttons, and then after letting them go
14:20:20TheSeven(what does happen if you let them go? does it continue to boot up normally?)
14:20:32fLiDeI lost you man
14:20:52*linuxstb sees the text "ALiOS1.0" which seems familiar
14:21:05fLiDewhat do you want to do me exactly?
14:21:11TheSeventhe pdf says something about powering off the player by holding a button for 2 seconds, and then holding some buttons while connecting to the tool
14:21:43fLiDeno I don't need to press any buttons while connecting it
14:21:55TheSevenlinuxstb: this?
14:23:20linuxstbTheSeven: I just recognise the name from other mp3 player firmwares.
14:23:50TheSevencan you remember which ones?
14:24:02TornefLiDe: we are looking at the instructions for the firmware update utility
14:24:15Tornethere are instructions for putting the player into firmware upgrade mode
14:24:20Tornethis quite probably changes how it appears over usb
14:24:27Torneperhaps to a device type that we recognise from other players.
14:24:45*linuxstb has a feeling of deja-vu -
14:25:42fLiDepeople, I am really sorry , but I am new to all this, and more importantly "SLOW" .... :(
14:26:24TornefLiDe: lots of mp3 player chips have a built in mode to update/recover their firmware, and this is normally standard for every player with a given cpu
14:26:33gevaertsfLiDe: don't worry. We do have some really bright people here, but most of us aren't
14:26:50Torneso, if you put your player into the firmware update mode, and then run "lsusb" on your linux machine while it's connected
14:26:57Torneit might be obvious what type of cpu it has.
14:27:03Torne(to us, at least) :)
14:27:16Tornedownload the firmware update if yo uhaven't already, and read the instructions in it
14:27:20fLiDecool, understood
14:27:32Torneyou don't need to actually run the update, just press whatever buttons on the player it is
14:27:47fLiDeprocessing lot of info...
14:29:01fLiDewhile my download is on , can you tell me what button does it say's to press
14:29:46Torneer, i don't know. i deleted the instructions already :)
14:30:20fLiDeme slow, my net slow, slow flow...
14:30:33 Join Schmogel [0] (
14:30:57linuxstb"Press and hold the “Reset” button on MP840 for 2 seconds." then Press and hold the “Volume Down“ button (and keep that button pressed until you've run lsusb)
14:32:01TheSevenis there anything on the display while you hold the volume down button?
14:32:02TheSevenwhen you let go of the volume down button, does it just continue to boot up normally?
14:32:49fLiDeooh man It is not booting anymore
14:32:55pamaurymc2739: ping
14:33:01TheSevenfLiDe: now run lsusb
14:33:16linuxstbfLiDe: Yes, that's the idea. it's now in some kind of special firmware-upgrade mode.
14:33:42fLiDejust a sec
14:34:02Tornethe ALiOS1.0 string is in the fat16/fat32 drivers
14:34:16Tornethe strings it is near make it seem likely that that's the system string used when formatting a partition
14:34:17TheSevenTorne: drivers? i saw it in boot sectors
14:34:30fLiDeBus 005 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
14:34:33Tornemaybe not drivers?
14:34:33fLiDeBus 006 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
14:34:33DBUGEnqueued KICK fLiDe
14:34:33fLiDeBus 004 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
14:34:33fLiDeBus 007 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
14:34:33***Alert Mode level 1
14:34:33fLiDeBus 002 Device 010: ID 0402:5668 ALi Corp.
14:34:33***Alert Mode level 2
14:34:33fLiDeBus 002 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
14:34:35fLiDeBus 001 Device 002: ID 064e:a101 Suyin Corp. Acer CrystalEye Webcam
14:34:38fLiDeBus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
14:34:38fLiDeBus 003 Device 001: ID 1d6b:0001 Linux Foundation 1.1 root hub
14:34:44TheSevenand yes, it was used as a system identifier of the formatting system in there
14:34:44fLiDethis is lsusb output
14:34:44Torne1) don't paste in here
14:34:59fLiDeand i guess the player is ALi corp
14:34:59Torne2) you need to use lsusb -v and we're only interested in the device which is the mp3 player :)
14:35:18TheSevenfLiDe: unplug the player and see if that ALi thing vanishes
14:35:30gevaertsso "lsusb -v -d 0402:5668", and use
14:35:44TorneTheSeven: yeah I think the file is actually a template for formatting
14:35:54fLiDeyes it does vanish
14:36:09 Join knine [0] (
14:36:37 Quit kaniini (Read error: Connection reset by peer)
14:37:36Tornegoogle suggests this is an ALi M5661R
14:37:38CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
14:37:38*linuxstb guesses the CPU might be an M5661X
14:37:39fLiDethe above command didn't produce any output?
14:37:39TheSevensame usb id
14:37:46Torneor similar
14:37:52 Join kaniini [0] (
14:37:56linuxstbThe first google hit for M5661X is interesting...
14:38:01fLiDeno outputs friends....
14:38:24fLiDewhat do i do, does that command saves the output in some file?
14:38:32linuxstbSeems some Russian site has the SDK for that chip...
14:38:36Torne <- here we go
14:38:40Tornefirmware format matches
14:38:43Torneso yes, it'll be an m5661
14:38:48gevaertsfLiDe: if it's not plugged in anymore, lsusb won't find it
14:39:21*linuxstb reads
14:39:36 Quit ecio (Ping timeout: 260 seconds)
14:41:35 Quit knine (Ping timeout: 276 seconds)
14:41:48TorneYeah there is basically no chance of getting rockbox on this
14:42:00Torneit's a miniscule intel 8051-ish MCU with a DSP
14:42:04fLiDenot good
14:42:39fLiDey ?
14:42:48Torneit's not powerful enough
14:43:21fLiDecan I scale down the rockbox code to make it enough?
14:44:27Torneno, it probably only has 40kb of ram :)
14:44:34***Alert Mode OFF
14:44:51TheSeven8051 disassembly makes sense
14:45:17TornefLiDe: the chip is only a 16 bit microcontroller
14:45:21Tornelike a PIC or AVR
14:45:25fLiDejust 40kb, and how did you find it out?
14:45:26CIA-8New commit by 03teru (r24861): plugin: implement highscore_show for player and use it in rockblox.
14:45:29Torneit relies on the DSP chip to do all video/audio decoding
14:45:47 Join bmbl [0] (~Miranda@unaffiliated/bmbl)
14:46:11TornefLiDe: because it's a 16-bit microcontroller: its address space is 64kb
14:46:15fLiDeDSP : as in line "digital signal processor"?
14:46:16Torneand it needs some of it to address peripherals
14:46:53fLiDenot possible, cause it has got 4gb of memory
14:47:06Torneno it doens't
14:47:14Torneit has 4gb of flash
14:47:21fLiDeooh yes
14:47:26fLiDewhat's the difference?
14:47:47Tornei thought you said you were good at C programming :)
14:47:56Torneand you don't know the difference between memory and storage?
14:47:58fLiDeeven if it uses 20 bit addressing it won't be able to use 4gb
14:48:13fLiDeit only will get to 1gb
14:48:14TorneIt's not a word addressed flash
14:48:18Torneit's a block device
14:48:26Torneall NAND flash is.
14:48:46Torneyou only need a few bytes for its registers, and you read one word at a time from a block into ram
14:48:53Tornejust like a hard disk or any other block device
14:49:08fLiDe64B at a time?
14:49:16fLiDeisn't it too low?
14:49:16Torneno, probably two bytes at a time
14:49:41Tornewell, it may have DMA, but maybe not
14:49:53Torneeven if it does that'll still just be single pages (2048 bytes or so)
14:50:17Torneanyway, rockbox is ~20 times larger than this device's entire ram, and rockbox requires a 32-bit processor
14:50:29Torneand even if it ran the processor would be too slow to decode mp3
14:50:45Tornewe don't ahve any DSP code to do decoding, so you'd have to rewrite all the codecs in suitable DSP code :)
14:50:46fLiDebasically, I am done...
14:51:00Tornethe s1mp3 project is more like it
14:51:09Tornebut they don't run on this particular device either
14:51:17Tornethey run on a *different* system on chip which is a MCU and DSP
14:51:28Torneso their code also won't work on this player, but at least it's in the right ballpark :)
14:51:54*linuxstb suggests fLiDe buys a different mp3 player...
14:51:55fLiDeso, thank you friends
14:52:01Tornes1mp3 runs on players with Z80 MCUs and a DSp
14:52:05Tornethis is a 8051 MCU
14:52:12Torneso, totally different, but comparable memory size/etc
14:52:20fLiDesorry linuxstb, haven't got enough money
14:55:08Tornegevaerts: you were nearly right, then :)
14:55:24 Quit fLiDe (Quit: Thanks for all the help, you guys rock)
14:55:40gevaertsTorne: at least there's a "80" in both :)
14:55:44Tornethe structure of the firmware was kinda a giveaway; the files seperated like that were pretty much guaranteed to be overlays for a device with miniscule amounts of ram
14:56:06gevaertsand therefore (probably) also address space
14:56:47Tornethe fact that people generally are happy to assume these devices are equivalent to players with 32-bit microprocessors is a tribute to how powerful these DSP chips are in the right hands, i guess :)
14:58:26***Saving seen data "./dancer.seen"
15:00:13 Join bug2000 [0] (~bug@unaffiliated/bug2000)
15:00:39gevaertssometimes they actually are equivalent out of the box. If you look at what e.g. a c200 can do out of the box, it's not much more than one of those s1mp3 things
15:01:04gevaertsThat's a software issue of course, but you can't expect users to realise that
15:06:45ThomasAHperfectdrug_: looks perfect :)
15:07:21ThomasAHperfectdrug_: (I can easily view .svg, too)
15:11:10 Join ecio [0] (
15:12:01 Quit shai (Ping timeout: 272 seconds)
15:12:58ranmachanBTW: Has anyone ever looked at what powers a Sansa Shaker? I googled it a bit and didn't find anything useful
15:13:05S_a_i_n_twhy are all the plugins/viewers etc. named without capitol letters...but the main menu does have capitol letters? it seemn weird, and inconsistent.
15:13:33ranmachanA friend of mine has a borked Shaker, maybe I'll habe a look at it when I'm back in germany.
15:13:49ranmachanThough I guess it's not a good target since it doesn't have a display...
15:16:50TorneS_a_i_n_t: because they're filenames :)
15:16:52 Join parafin [0] (
15:17:34rasherTorne: not really an excuse, is it?
15:18:12S_a_i_n_tis there some convention that dictates filnames must not have capital letters? :P
15:18:56linuxstbIsn't the reason that the real solution is to localise the names with the lang files, but no-one has done that?
15:19:46rasherThat could also be it
15:19:59*S_a_i_n_t would be prepared for this arduous English that is, if it means a bit of consistency is gained from it.
15:20:54linuxstbS_a_i_n_t: You can't really produce a patch for that. It needs someone with svn access to do it directly. But I don't think it's worthwhile for the reason I just said - we want proper localised names (in some clever and efficient way...)
15:21:53rasherDoesn't it just need a sort of lookup table from filename -> langid?
15:22:05rasherIt's not like it needs to be fast or anything
15:24:49*S_a_i_n_t doesn't quite understand why a capital letter borks efficiency...but is prepared to run with it. It just...shit, I dunno, it just looks weird to me.
15:25:24 Quit anewuser (Quit: =ooo ϢINTER ϾHIP 5iVE is OOON!! omg 5 days left =oo)
15:25:31linuxstbS_a_i_n_t: We're talking about translating plugin names into other languages.
15:25:35linuxstbi.e. not displaying the filename at all.
15:25:59TorneS_a_i_n_t: and yes, there *is* a convention that says filenames don't normally have capital letters; it's called "unix" :)
15:26:09linuxstbAnd no spaces...
15:33:54 Join Sajber^ [0] (~Sajber^
15:34:15pamauryThere is a convention that say that all filenames must be in capital and in the 8.3 format :)
15:38:14linuxstbS_a_i_n_t: I think people agree that it would be better to have the plugin names capitalised. But I at least would prefer the filenames to stay as they are, and to change the displayed plugin names via a localisation system.
15:38:22linuxstbSo that a) they can be translated; and b) they are spoken (afaik they're not at the moment, unless the user creates .talk files)
15:41:24S_a_i_n_tif you look, I only mentioned plugins/viewers...not filenames.
15:41:27S_a_i_n_tbut I agree.
15:41:50Torneplugins/viewers *are* filenames
15:41:56mc2739pamaury: pong
15:41:57Tornethe plugin menu is just a file browser
15:42:03Tornewhich happens to browse /.rockbox/rocks
15:42:18linuxstbS_a_i_n_t: "15.18.12 # <S_a_i_n_t> is there some convention that dictates filnames must not have capital letters? :P"
15:43:33S_a_i_n_tyeah...bad way of explaining it...I mean, like Dict.rock, Pictureflow.rock etc. but of course there no need for things like rockbox.ipod etc to be capitalised
15:43:58Tornewe don't want *any* filenames to be capitalised
15:46:51rasherJust hardcode the filebrowser to do filename[0]-=32 when browsing plugins!
15:47:02rasherThe perfect crime
15:47:07linuxstbrasher: commit!
15:47:45rasherPretty sure I couldn't live with the shame.
15:48:28TheSevenrasher: if anything: *filename&=~32
15:48:45rasherYou're evil.
15:48:58TheSevenwell, that's uppercase-safe at least :-)
15:49:11TheSeven( won't corrupt chars that are already uppercase)
15:49:20rasherYeah, but we already decided those don't exist!
15:51:43 Quit teru (Quit: Quit)
15:52:16pamaurymc2739: do you have time to test things on your device to investigate the stkov ?
15:53:30mc2739pamaury: yes, at least a little but of time before I leave for work
15:57:41pamaurymc2739: ok, I'm coding a little debug thing to see at which point the overflow arises
15:58:30 Quit mc2739 (Ping timeout: 268 seconds)
15:59:25 Quit b0hoon2 (Quit: CGI:IRC 0.5.9 (2006/06/06))
15:59:37 Join b0hoon [0] (~quassel@
16:00:01 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
16:04:02pamauryoops, my e200 can't find the of anymore...
16:04:28pamauryHow do I recover from this ?
16:05:06pamauryI was about to open that page :)
16:06:56pamaurymc2739: I think you will have to wait a bit before testing :)
16:08:20mc2739pamaury: no problem, just post a pastebin link or in the flyspray task and I will test when I get a chance
16:08:49pamaurymc2739: I have one but as I didn't test it myself, I will wait a bit
16:11:52*FlynDice sees funman had a bit of "insomnia" last night!! Decides to look at SD code today.
16:16:36 Quit Kitar|st (Ping timeout: 264 seconds)
16:17:31 Quit gr1b (Ping timeout: 246 seconds)
16:21:27 Quit linuxstb (Read error: Operation timed out)
16:21:58 Join linuxstb [0] (~linuxstb@rockbox/developer/linuxstb)
16:22:05 Join gr1b [0] (
16:30:37 Join jgarvey [0] (
16:36:42 Join toffe82 [0] (~chatzilla@
16:42:22 Quit linuxstb (Ping timeout: 246 seconds)
16:53:47 Quit S_a_i_n_t (Quit: Obligatory "attempt to sleep for an hour or so" time...)
16:56:02 Join linuxstb [0] (~linuxstb@rockbox/developer/linuxstb)
16:57:32 Quit b0hoon (Quit: - Chat comfortably. Anywhere.)
16:58:03Bagderwe should have a linux/scripts/ for rockbox
16:58:28***Saving seen data "./dancer.seen"
17:00:31Zagorthat would indeed be nice
17:02:45pamauryBagder: what is that script supposed to do ?
17:03:08Bagdercheck that a patch or a file fulfills the linux coding guidelines
17:03:35Bagderindentation, line lengths, spaces between "if ()", brace positions etc
17:03:57pamauryAh, I don't like spaces between "if" and "("
17:04:11pamauryAnd braces are better on a new line !
17:04:13Bagderthe linux guidlines are not rockbox guidelines
17:04:22pamauryhopefully ;)
17:04:24Torneyah, our guidelines are different :)
17:04:34Tornebut ultimately whether people like them is irrelevant ;)
17:04:40BagderI meant that the idea with a script that checks them is nice
17:04:42pamauryRockbox is not very consistent regarding certain points
17:04:54Tornewell, our guidelines are not very detailed. we mention only a few things
17:05:01Zagorpamaury: that's by design
17:05:14Torneand we have loads of imported code from other projects
17:05:23TheSevenit still should be at least *consistent within a file*, no matter what that means.
17:05:24Bagderright, that's what'll break the script idea
17:06:04pamaurySomeone should write a very intelligent script that adapt to each file
17:06:08 Quit linuxstb (Ping timeout: 260 seconds)
17:08:13 Join linuxstb [0] (~linuxstb@rockbox/developer/linuxstb)
17:19:40 Quit bmbl (Ping timeout: 260 seconds)
17:22:12 Quit Zagor (Quit: Clint excited)
17:26:55 Quit fyrestorm (Quit: lamers envy me like they envy bill g -- main boot xp, just the way it should be!)
17:34:51 Quit Grahack (Quit: Tu m'as vu ?)
17:35:00 Join iq [0] (~iq@unaffiliated/iq)
17:38:27 Join anewuser [0] (anewuser@unaffiliated/anewuser)
17:38:29 Quit iq (Read error: Connection reset by peer)
17:39:01 Join iq [0] (~iq@unaffiliated/iq)
17:40:15perfectdrug_is this desired? just a little cosmetic, if yes I add the numbers
17:41:45 Join liar [0] (
17:43:33 Join fyrestorm [0] (
17:45:12 Join watto1 [0] (~watto@
17:45:28 Join phanboy4 [0] (
17:45:53 Quit watto (Ping timeout: 248 seconds)
18:01:03 Quit froggyman (Read error: Connection reset by peer)
18:08:38 Join komputes [0] (~komputes@ubuntu/member/komputes)
18:10:55perfectdrug_ and the same file for UI-sansae200.bmp
18:11:35 Quit xavieran (Ping timeout: 256 seconds)
18:14:38 Join bertrik [0] (
18:15:54linuxstbperfectdrug_: What is showing us?
18:16:43 Join piotrekm [0] (~piotrek@unaffiliated/piotrekm)
18:17:12perfectdrug_linuxstb: IMO the right original e200 simulator picture looked something wrong so I changed it to the left
18:20:46linuxstbDid you change the SVG, or just the UI sim bitmap?
18:22:10perfectdrug_just the sim bmp it can be replaced and should work just like the other one, everything is at the exact right position as it was
18:22:23 Join Strife89|PalmTX [0] (
18:22:55perfectdrug_just a cosmetic thing
18:23:49 Join xavieran [0] (
18:26:50 Quit petur (Quit: *plop*)
18:28:05 Quit liar (Ping timeout: 245 seconds)
18:28:31 Join saul [0] (
18:30:30 Join liar [0] (
18:35:55 Quit Strife89|PalmTX (Quit: Lunch break over.)
18:38:37 Join bmbl [0] (~Miranda@unaffiliated/bmbl)
18:41:24 Quit saul (Quit: CGI:IRC (EOF))
18:41:52 Join domonoky [0] (~Domonoky@rockbox/developer/domonoky)
18:45:54 Join karashata [0] (
18:58:32***Saving seen data "./dancer.seen"
19:00:27 Join panni_ [0] (
19:04:09 Join m3dlg [0] (~m3dlg@
19:05:01 Quit linuxstb (Ping timeout: 256 seconds)
19:10:29 Join linuxstb [0] (~linuxstb@rockbox/developer/linuxstb)
19:10:41 Quit m3dlg (Ping timeout: 256 seconds)
19:27:53 Quit linuxstb (Ping timeout: 272 seconds)
19:32:21 Quit phanboy4 (Read error: Connection reset by peer)
19:33:48 Join linuxstb [0] (~linuxstb@rockbox/developer/linuxstb)
19:33:56 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
19:37:11 Join JdGordon_ [0] (
19:47:03 Join Horscht [0] (~Horscht2@xbmc/user/horscht)
19:55:01 Quit grndslm (Remote host closed the connection)
20:01:10 Quit JdGordon_ (Quit: Bye)
20:02:30 Quit linuxstb (Read error: Operation timed out)
20:05:18 Join linuxstb [0] (~linuxstb@rockbox/developer/linuxstb)
20:17:10 Join JdGordon_ [0] (~Miranda@rockbox/developer/JdGordon)
20:22:34 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
20:22:34 Quit pixelma (Disconnected by services)
20:22:43 Quit amiconn (Disconnected by services)
20:22:45 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
20:22:51 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
20:23:09 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
20:29:26 Quit linuxstb (Ping timeout: 256 seconds)
20:32:32 Join linuxstb [0] (~linuxstb@rockbox/developer/linuxstb)
20:36:45 Join Kitr88 [0] (
20:40:29 Join saratoga [0] (~9803c6dd@gateway/web/freenode/x-bmoxsjprukfjvmji)
20:41:21 Quit CIA-8 ()
20:42:10pamauryHey, CIA just left !
20:42:21saratogahow much memory does the clip+ have?
20:46:46FlynDicesaratoga: I don't know the answer for sure but right now it's being treated the same as fuze, e200v2 with /* 16 bits external bus, high performance SDRAM, 64 Mbits = 8 Mbytes */
20:53:40 Join stripwax [0] (
20:53:58 Quit flydutch (Quit: /* empty */)
20:54:56saratogaBuschel: (for the logs) removing tabs made parts of libmad look weird
20:55:10saratogai don't think you substituted the correct number of spaces per tab
20:55:18gevaertsWhat does "moveal %a5@(4),%a0" mean? (coldfire)
20:55:34saratogalayer3.c looks odd in particular
20:55:47 Part martian67 ("Leaving")
20:56:10*amiconn wonders what kind of answer gevaerts expects
20:58:27gevaertsamiconn: I know it moves something to a0, and from the context I know the something is addressed relative to whatever is in a5. Is it from a5+4?
20:58:33***Saving seen data "./dancer.seen"
20:59:15amiconnI guess this is from a disassembly made by objdump?
20:59:47amiconnobjdump just uses another combination of the various alternatives for writing m68k asm
21:00:11amiconnI'd write it move.l (4,%a5), %a0
21:01:26amiconnThe 'a' in 'moveal' indicates that this moves an address register. It's redundant, because you can figure it out from the parameters
21:02:43amiconnThe difference is that 'movea' doesn't touch flags while 'move' (the real one, i.e. involving data registers) does
21:03:56gevaertsok, I think I'm beginning to understand why pebbles don't work on m68k
21:04:54gevaertsFor some reason it uses the wrong location for the GOP, which obviously causes it to jump to wrong addresses
21:06:08 Join CIA-5 [0] (cia@
21:07:21 Join Grahack [0] (
21:08:06 Join Buschel [0] (
21:09:26pamauryCIA is back !
21:10:03Buschelsaratoga: the tabs are removed by an automatism in my editor −− which in general works very well. maybe it had a hickup in libmad... nevertheless the code is still readable better than before (imho)
21:12:22gevaertsapparently arm uses the relocated .got section and m68k expects the relocated pointers to be in .got.plt, based on .got
21:12:41 Quit piotrekm (Ping timeout: 252 seconds)
21:13:36linuxstbBuschel: Looking at your patch, your editor seems to have got more things wrong than right.... Maybe your editor was displaying the tabs oddly before, which is why you think it's better now.
21:14:00Buschelhmm, can you give an example? which tab?
21:14:24linuxstbI'm currently looking at liba52/downmix.c
21:16:49 Join m3dlg [0] (~m3dlg@
21:18:25 Join Omlet [0] (
21:20:55gevaertsJdGordon_: if I cache the pointer to the viewport I got in skin_parser.c, I seem to get changing viewports (it jumps a bit every now and then). Is that expected?
21:21:56JdGordon_show me what you're doing?
21:21:58 Part watto1
21:22:06gevaertshm, it could be something else...
21:22:34gevaertsI probably should call lcd_set_viewport()
21:22:37JdGordon_unless you are caching the wrong pointer?
21:22:49gevaertsno, it's the right viewport every now and then
21:22:52JdGordon_not in the parser
21:23:01gevaertsno, but before actually drawing
21:23:11*gevaerts tries
21:23:22JdGordon_except that might cause wierdness elsewhere
21:24:02gevaertsis there an lcd_get_viewport? I could restore the previous one then
21:24:44JdGordon_there isnt
21:25:10Buschellinuxstb: ok, my editor is set to tabs =4 spaces... :/
21:25:17pixelmajumping viewports? Is that something visible?
21:25:42gevaertspixelma: you need FS #11027 to see it
21:26:51JdGordon_jumping viewports isnt really special
21:26:59JdGordon_2 lines in the lcd driver can do that :)
21:30:07pixelmagevaerts: I know what you are working on, just got alerted because you report problems on a coldfire target which didn't seem to be there on your arm targets. Thinking of the weird problems I have with the FMS patch on my M5 and Ondio while it seemed to work flawlessly for AlexP on the Gigabeat S
21:30:49AlexPTo the extent that I tested - I haven't yet tried loading a different fms for instance
21:31:11gevaertspixelma: the jumping also happens on ARM, it's just that now that the things actually don't freeze or crash on any of my players that I'm starting to concentrate on details like actally working properly
21:31:30pixelmaah, ok
21:32:28pixelmamaybe I should try the c200 as well
21:33:06saratogaBuschel: depending on the library, i think some need tabs=8 to look right
21:33:21 Quit m3dlg (Ping timeout: 268 seconds)
21:33:26saratogasince IME a lot of them mix 8 spaces in some areas with tabs in others
21:33:31Buschelsaratoga: i will undo the change and check for each lib :(
21:33:38saratogai don't think thats needed
21:33:49saratogamaybe just go through and correct libmad and liba52?
21:33:58saratogaIMO what you have now is no worse then what it was before
21:34:18saratogais there some tool that can auto space files to rockbox specification?
21:34:51Buschelmaybe... but not my editor...
21:35:50saratogaI think visual studio can but i don't have it installed anymore
21:36:02TheSevenlet me have a look
21:36:07linuxstbsaratoga: I think it will be easier to do it properly if that patch is reverted. It _is_ worse than now.
21:36:20saratogalinuxstb: ok never mind then
21:37:29 Quit jgarvey (Read error: Connection reset by peer)
21:37:52amiconnlinuxstb: I would like to split the filter calls in demac (separate calls per channel) for dualcore operation. Do you think that's ok?
21:37:54 Join froggyman [0] (
21:38:55TheSevenwhich commit are you talking about? 24862?
21:39:04 Join S_a_i_n_t [0] (S_a_i_n_t@
21:42:58linuxstbamiconn: Seems fine to me.
21:43:25linuxstbTheSeven: I am, yes...
21:49:12 Quit evilnick (Read error: Connection reset by peer)
21:49:39 Join evilnick [0] (
21:49:39 Quit evilnick (Client Quit)
21:50:10 Join evilnick [0] (~evilnick@rockbox/staff/evilnick)
21:51:15TheSevendamn, it the number of spaces per tab isn't even consistent within single files!
21:51:54*gevaerts is chasing bad pointers again :\
21:52:55JdGordon_gevaerts: can you show me the diff for skin_displa.c ? the correct viewport shold be set before your pebble is called
21:54:20JdGordon_hmm not much there :p
21:57:06JdGordon_you are storing the pebble index in the token value.i, and then using that as the viewport?
21:57:44JdGordon_or am I too tired to see what's actually happening?
21:58:39JdGordon_the latter
21:58:54gevaertsJdGordon_: is what happens in skin_parser.c
22:00:11JdGordon_return false in parse_pebble shuold be return <0
22:00:24 Quit Grahack (Quit: Tu m'as vu ?)
22:00:28JdGordon_negative return values are errors, 0 is fine
22:00:44gevaertsah, ok
22:00:57JdGordon_&(curr_vp->vp) *should* be perfectly safe
22:01:08 Join jgarvey [0] (
22:01:21 Quit ender` (Quit: It always takes longer than you expect, even when you take Hofstadter's Law into account. -- Hofstadter's Law)
22:02:59CIA-5New commit by 03Buschel (r24863): Change liba52 and libmad to the original 8-spaced tabbing.
22:04:28 Join sudoman [0] (~c05041eb@gateway/web/freenode/x-bobwxzoymahfimqt)
22:07:20JdGordon_kugel: saving the fonts like that seems pretty wasteful, and it doesnt look like you fix the font numbers on curr_screen==SCREEN_REMOTE, but otherwsie it looks fine
22:10:05 Join veeloc [0] (
22:12:10 Join piotrekm [0] (~piotrek@
22:12:10 Quit piotrekm (Changing host)
22:12:10 Join piotrekm [0] (~piotrek@unaffiliated/piotrekm)
22:13:28 Quit iq (Remote host closed the connection)
22:15:36Buschelsaratoga/linuxstb: you are right. it looks better now :o)
22:15:52pamauryWhy does metadata contains de mp3info function and a get_metadata one ? First of all, mp3info is a special case of the second and has reversed returned value
22:16:56linuxstbpamaury: Be careful of hwcodec if you change that.
22:17:07domonokypamaury: probably because it evolved from pure mp3 metadata, to the full system we have :-)
22:19:13*linuxstb wonders if hwcodec needs that anynmore
22:22:13pamaurylinuxstb: it just metadata so swcodec or hwcodec is the same no ?
22:22:37linuxstbpamaury: No - hwcodec metadata is just mp3 (and mp2). swcodec metadata is everything.
22:23:22linuxstbBut yes, perhaps that function can be removed. Trouble is that there are almost no hwcodec devs left...
22:23:25pamaurylinuxstb: yes but in this case, get_metadata will fail for all others. It's implemented in software in both cases, on ?
22:24:11CIA-5New commit by 03Buschel (r24864): Remove tabs in firmware path (taking into account the original spacing).
22:24:13linuxstbmp3info looks like it's only used by the hwcodec playback engine (apps/mpeg.c). get_metadata() is used by the database and the swcodec playback engine.
22:24:26pamauryOk, I have to leave to few moments, if you could have a look at it and tell me if you think it's safe, that would be great.
22:24:36linuxstbIf what is safe?
22:30:00*gevaerts doesn't understand the .got.plt section
22:31:43 Join Riffer_ [0] (
22:34:50Riffer_good evening
22:34:58 Nick Riffer_ is now known as Riffer (
22:35:22 Quit toffe82 (Quit: ChatZilla 0.9.86 [Firefox 3.6/20100115144158])
22:36:04Rifferwho should I talk to about applying a patch to the repository?
22:36:13JdGordon_which patch?
22:36:31JdGordon_AlexP: dunno if oyu saw the fm update. it wasnt a straight forward sync :/
22:36:38Rifferthis one: FS #11042
22:40:17JdGordon_ignoring weather we want it or not... I see lots of tabs instead of spaces
22:40:38RifferI wonder who decides about having the things like they are... for example who says or decides about the position of some menus? I think the system menu is a little bit of a setting but it has a extra position?
22:40:43domonokyRiffer: i cant really comment on the technical side, but this patch looks like it should be cleaned up a bit more
22:40:44 Join S_a_i_n_t_ [0] (S_a_i_n_t@
22:41:18Rifferok, please tell me what exactly? first remove all tabs - ok, will be done.
22:41:46Rifferextra debugging output? remove?
22:41:54domonokythere are sections commented out. if they are not needed, remove it.
22:42:23saratogaand use /**/ comments rather then c++ comments
22:42:41 Quit Omlet (Quit: ( :: NoNameScript 4.22 :: ))
22:42:50stripwaxfft plugin - seems to think the frequency range is half what it actually is, so frequencies above N/4 get reflected and plotted 'downwards' on the scale. also, the amount that the screen scrolls seems to little - a frequency sweep has an obvious 'knee' at the point the screen begins scrolling
22:43:04domonoky+1 for also including a manual patch :-)
22:43:13 Quit S_a_i_n_t (Ping timeout: 276 seconds)
22:43:36AlexPRiffer: Menus etc. are decided by an argument, then doing nothing, then another argument, etc. After a few arguments we might reach a consensus
22:43:57JdGordon_close but no
22:43:59stripwaxalso log vs linear seems to be *amplitude* rather than frequency!
22:44:01domonokyAlexP: nice description :-)
22:44:05JdGordon_they are decided arbirtarily, then argued about
22:44:06Rifferoh - the manual patch is not from me ... as you may see a great part of the patch is made by Nick Peskett
22:44:22JdGordon_the root_menu.c change makes no sense
22:44:54JdGordon_if sleep timer on startup is true then why shuold it matter if the start screen is WPS?
22:46:46Rifferthe sleep timer is only startet on play - so if someone decided to have the wps as its start screen (like me) the sleep timer has to be started
22:47:52Rifferits a sleep timer not the normal timeout shutdown while standing still (not playing anything)
22:48:21JdGordon_we already have a sleep timer right?
22:49:04JdGordon_that timer shouldnt be linked to any screen or playback state
22:49:23JdGordon_what if I want it to turn off after an hour or recording or radio?
22:50:33JdGordon_also why did you create a new task for it instead of just continuing on 10849?
22:50:35Riffersee it as an escalation: timeout is always active if the user sets it but it only shut down while the player is stopped - when it plays the sleep timer is responsable to pause playing after the set time.
22:51:45AlexPSurely it should shutdown?
22:51:50JdGordon_that doesn't change anything. I still want to be able to shut down after an hour of radio which I can do now and that patch wont allow
22:51:50Rifferok - thats a point - i do not have radio in my device so I did not overcome to this point... hmm... another task because I wanted to merge two of them.
22:52:06AlexPIf I set the sleep timer for an hour I want to to e.g. play music for an hour then shut down
22:52:26pamaurylinuxstb: *safe* to remove it and replace it by get_metadata. The word is not well chosen
22:52:31 Quit anewuser (Quit: ϢINTER ϾHIP 5iVE is OOON!! omg ~4 days left =oo)
22:53:20JdGordon_AlexP: I can see the argument for changing it to pause, which will shutdown anyway if the idle timeout is set (which might be a stupid assumption)
22:53:38AlexPI think that might be highly confusing
22:53:49Rifferreally? I think after having the time of the sleep timer as a playtime (if radio or files) it should stop playing but not shutdown. The shutdown is a task for the timeout, isnt it?
22:54:26AlexPI might have a really long idle shutdown for whatever reason - I listen to it at work and often have to pause and go and do things for instance
22:54:44AlexPSo now if I want to use the sleep timer I have to set two things - the sleep timer and the idle shutdown
22:54:52linuxstbpamaury: No, I don't think so. Firstly, mp3info() takes a filename, get_metadata() takes a file descriptor. But maybe mp3info could be made internal to the hwcodec playback engine, as a kind of wrapper for get_metadata().
22:55:03linuxstb(hwcodec playback engine = apps/mpeg.c)
22:55:11pamauryyes that's what I propose.
22:55:15AlexPI don't think I should have to set a different setting to make another setting work properly
22:55:41pamaurylinuxstb: because it's rather strange to have a general function and another one which is specialized
22:56:07pamaurylinuxstb: and also, why get_metadata takes a filename and a fd ? Couldn't it take just a filename ?
22:56:19pamauryIs it an optimization ?
22:56:25*pamaury leave again for a few moments
22:56:54JdGordon_doesnt it only need te filename to get the extension? and seen as its opeend already passing the fd means one less open
22:57:06RifferI see we have different meanings of what a sleep timer does. I use it really for sleeping. Going to bed starting the player and listen. After a while the sleep timer runs out and pauses playing. If I am still awake and want to listen further I just start playing. If not the shutdown timer shuts off the device...
22:57:27AlexPMost sleep timers that I know turn off the device
22:57:39AlexPAnd that is how it functions now isn't it?
22:58:37***Saving seen data "./dancer.seen"
22:58:59Riffernot really. at the moment the setting for sleep timer is non-permanent. :-)
22:59:01linuxstbAlexP: Sleep timers can often give you a prompt to confirm you still want to turn it off. I guess what Riffer is proposing is the equivalent of that.
22:59:22Riffer@linuxstb: yes - thats what I want from it
22:59:23AlexPIt just seems to me that one setting should rely on another
22:59:30AlexP*Seems wrong to me
23:00:21AlexPlinuxstb: And if you don't cancel they turn off anyway. So maybe the sleep timer should pause and then shut down e.g. two minutes later if you don't do anything
23:00:32AlexPIndependent of the idle shutdown
23:00:41Rifferi agree it relys on the other - buts exactly what I want from it.
23:00:55AlexPI wouldn't object to that, I just don't think that the sleep timer should rely on the idle shutdown
23:01:01JdGordon_more settings bloat! make the option pause or shutdown
23:02:04Riffer@JdGordon: good idea
23:02:33AlexPOr just make it pause for two mins, then shutdown if you don't do anything - have its own non-configurable idle timeout
23:03:13Rifferanother good idea - but for all this the sleep timer should be a permanent option I think?
23:03:39AlexPYou mean it remembers the time, so you just have to enable it to use it?
23:04:32RifferIt remembers the time and rembers if active or not on startup.
23:04:47AlexPI'm not sure about remembering it is active
23:05:03AlexPIf I use it to go to sleep to, I don't then want it active the next morning
23:05:17JdGordon_Riffer: no its not a good idea! we have too many settings already
23:05:25AlexPAnd just turning it on when wanted is half the work of turning it on when wanted then off when not wanted
23:05:33RifferI am - its the way I use my device - its annoying to start the sleep timer anytime it starts to have it sleep when I do...
23:05:52AlexPI would be against it staying on
23:06:05AlexPRemember the time sure, but not the state
23:06:37 Quit jgarvey (Ping timeout: 276 seconds)
23:06:38Rifferbut you can change this at any time? the system menu is already at a pole position for some reasons I do not know?
23:07:07AlexPI imagine that most people use their daps during the day too
23:07:27AlexPSo if it remembers the state you have to turn it on in the evening, turn it off in the morning, on in eveneing etc
23:07:32 Join JdGordon1 [0] (~Miranda@nat/microsoft/x-rmkgnbnwjqulzzyk)
23:07:33RifferAnd many do it at evening and nights - if they use it for audio books.
23:07:39AlexPIf it doesn't remember state then you just turn it on in the evening
23:07:52AlexPIt shouldn't stay enabled
23:08:16domonokyhave a setting for it :-)
23:08:28JdGordon1doing fancy things like waiting 2min between pause and shutdown is just asking for trouble
23:09:21AlexPJdGordon1: You mean codewise?
23:09:34Rifferthis why at the moment I let the setting of shutdown timer do this for me - its my favorite solution - I see you are agains it - maybe you use your dap only for music?
23:09:46JdGordon1yeah, adding extra states to the powermanagement thread
23:09:54 Quit Xerion (Ping timeout: 240 seconds)
23:10:10 Quit bluebrother (Disconnected by services)
23:10:11 Join bluebroth3r [0] (~dom@rockbox/developer/bluebrother)
23:10:11 Quit JdGordon_ (Ping timeout: 272 seconds)
23:10:16AlexPRiffer: music and audiobooks - but that isn't the point - the point is that most people use their daps during the day than at night I suspect
23:10:24AlexPJdGordon1: Fair enough
23:10:32AlexP*more during the day
23:11:02Rifferis rockbox only for daywalkers? ;-)
23:11:32RifferI think have this as an option does not hurt anyone.
23:11:34domonokyno, but we should find a way to please both day- and night-walkers :-)
23:11:56domonokywithout introducing too much new settings.
23:12:58Rifferhmm... ok, I see - I will have to maintain the patch a long as I live... not making it to the repository
23:12:59AlexPGiven that at night you probably want the sleep timer and during the day you don't, having the sleep timer remember its state means changing the setting everytime you use your dap
23:13:41JdGordon1you drongos :)
23:13:48domonokyif the target has a rtc, we could even enable it only in the evening *hehe*
23:14:01JdGordon1just have the options "on and pause", "on and poweroff", "off" for the enable and boot optins
23:14:44JdGordon1Riffer: oh come on :) have a look at the wikipedia patch before you go all melodromatic :)
23:15:12Rifferok will do it an be back soonish (a week or a day that depends on my time)
23:15:27AlexPJdGordon1: That'd be fine too
23:15:38 Quit bmbl (Quit: Bye!)
23:16:00Rifferhmm... now I ask myself wether this client saves the log or not...
23:16:13gevaertsIs there a reason why lcd_clear_viewport() isn't in the plugin api (apart from the usual one)?
23:16:15Rifferdo i remember right that it is already on the server tomorrow?
23:16:29gevaertsRiffer: the channel is logged in real time
23:16:36Rifferoops :-)
23:16:39Rifferthank you
23:16:59JdGordon1gevaerts: rb->screens[i]->clear_viewport()
23:17:24RifferJdGordon: what exactly do you mean about the wiki? the point in the rockbox wiki about patching?
23:17:26gevaertsah, ok
23:17:55AlexPRiffer: No, there is a wikipedia viewer patch on flyspray
23:18:04JdGordon1Riffer: no, the wikipedia patch has been maintained for YEARS and wont be commited
23:18:52Rifferok .. seems to be a "running gag" to me :-)
23:19:04S_a_i_n_t_how so?
23:19:08gevaertsthere are unfortunately good reasons for not committing it
23:19:34Rifferwhat exactly does the patch do? maybe a link for me?
23:19:58linuxstbJdGordon1: Why will it never be committed?
23:20:03 Join sudoman_ [0] (~d8ecfceb@gateway/web/freenode/x-kwvxluytplpdyfzc)
23:20:39JdGordon1I dunno, noone feels the need to? I've never looked at the patch so my never might be wrong :p
23:21:02gevaertsIf the existing issues are resolved, it will go in
23:21:24 Quit sudoman (Ping timeout: 252 seconds)
23:21:43Rifferanother question: which editors are you guys (and girls?) using at linux for the rockbox development?
23:21:47 Join jgarvey [0] (
23:21:51S_a_i_n_t_who manages the .dict plugin? Is the author still around?
23:22:12AlexPRiffer: Whatever your favourite text editor is
23:22:21linuxstbgevaerts: The main issue (when I last reviewed it) seemed to be some anonymous contributions. mcuelenaere seems to be working on cleaning it up though.
23:22:36gevaertslinuxstb: yes, that's the main one. The rest is easy
23:22:54Rifferat the current I use geany - its nice, I am a little fool and love namecompletion and hovering parameter lists ...
23:23:15AlexPRiffer: This is off-topic however - see #rockbox-community for that sort of thing :)
23:23:28Rifferoh, sorry - thanks
23:23:52RifferI wish a good night to you all - see you in a while....
23:25:00JdGordon1gevaerts: are you sure the viewport needs to be set in the pebble?
23:25:04 Quit Riffer (Quit: Ex-Chat)
23:25:36gevaertsJdGordon1: if I don't, the pebble sometimes draws at an offset
23:25:54gevaertsfeel free to fix that :)
23:26:19JdGordon1are there other viewports in the wps? is the offset actually one of those?
23:27:20gevaertsno. In my tests I use a smallish viewport that's at 20 pixels from the top and left sides. Sometimes it draws in the corner (i.e. at offset 0,0 instead of 20,20)
23:28:07sudoman_S_a_i_n_t_: have you seen FS #6697 ?, but I'm about to ;)
23:29:36 Quit domonoky (Read error: Connection reset by peer)
23:30:11JdGordon1gevaerts: is there anything special in your wps? conditional viewports or images?
23:30:41gevaertsJdGordon1: the wps I quote in the task is the entire wps
23:30:42JdGordon1I see the playlist viweer also sets the viewport so you could just leave it and just make it known that you cant have other things in the viewport *(which probably makes sense anyway)
23:31:42JdGordon1in which case, you should probably set the viewport before calling pebble_draw() so you dont need to do it in each pebbble
23:32:01JdGordon1which you did :)
23:33:05S_a_i_n_t_sudoman: it doesn't use dict.desc/dict.index though does it....?
23:33:19*S_a_i_n_t_ prefers the Wordnet dict files
23:34:02sudoman_Saint: not sure.
23:34:46sudoman_also there seems to be two wiki pages for the dict plugin: and
23:37:16 Join dfkt_ [0] (dfkt@unaffiliated/dfkt)
23:37:17S_a_i_n_t_Hmmmm...I'm willing to bet that FS #6697 doesn't compile anymore (not cleanly at least) was last sync'd in Oct '09
23:37:28 Quit Farthen (Ping timeout: 246 seconds)
23:37:31 Quit dfkt_ (Client Quit)
23:39:48*Buschel removed all tabs from plugins path locally... quite some work :/
23:40:41 Join matsl [0] (
23:41:13 Quit dfkt (Ping timeout: 272 seconds)
23:42:34linuxstbBuschel: Is there really a need to do this cleanup? It breaks "svn blame"...
23:45:09*linuxstb misses that Buschel did the same in firmware earlier tonight...
23:45:55 Quit jgarvey (Quit: Leaving)
23:50:05Buschelthat were few changes compared to the plugins path.
23:50:23Buschelsomehow I am fed up with this now...
23:51:14linuxstbIs it even necessary? i.e. is the badness (misleading svn history, maybe breaking patches) worth it?
23:51:30JdGordon1I thought we wanted to keep tabs in imported code?
23:52:07Buscheldo we? if so, I could have saved some hours this night...
23:52:09*linuxstb recalls having a conversation before about mass-fixing tabs, and it was decided to leave them
23:52:45Buschelwell, at least I could save a patch for my local changes before undoing them
23:53:39 Quit karashata (Quit: The fluffy dragon has left completely!)
23:54:04 Quit sudoman_ (Quit: Page closed)
23:55:10saratogai think the codec changes are worthwhile
23:55:23JdGordon1is a unionFS type thing big enough for GSoC?
23:55:26saratogasince libmad hasn't been updated in 6 years and has been half rewriten now
23:55:38saratogaJdGordon1: whats the idea?
23:55:54JdGordon1merging the 2 filesystems together somehow
23:55:59JdGordon1internal and sd storage
23:56:16saratogayou need a new file system for that?
23:56:48JdGordon1no, but the idea obviously rips off unionfs so its easier to refer to it like than
23:57:21gevaertsJdGordon1: that's a *hard* project to do right
23:57:29pamaurywhy ?
23:57:52gevaertsthe code might be easy, but you have to get the semantics just right
23:58:11saratogaI thought the original idea of just merging the the two volume's FAT tables was enough
23:58:13pamauryyes and nobody will agree about it probably ;)
23:58:41pamaurysaratoga: I think you have to do this at the file level and not the fs level
23:58:54JdGordon1isnt merging the FAT's going to be very error prone?

Previous day | Next day