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

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

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

#rockbox log for 2005-05-08

00:06:56[IDC]Dragongotta go soon
00:07:55[IDC]Dragonjust 1 thing: my vector table is evenly spaced, entries pointing to values 4 bytes apart
00:08:29amiconnHmm. I don't have an idea what may cause this...
00:08:30[IDC]Dragonthe firt one to 0x09006600, second to 0x09006604, and so on
00:08:40[IDC]Dragon first
00:09:04amiconniiuc it must be that way for most vectors
00:09:26amiconnLook into firmware/system.c why
00:09:27[IDC]Dragonno, they point to code, some even to the same
00:09:51amiconnYes, they point into that UIE "code array"
00:10:33amiconnIt's implemented that way to get both the source address and the exception type
00:10:49amiconnOnly some vectors point to "productive" code
00:11:41[IDC]DragonI know, but mine is really a strict counter
00:12:06[IDC]Dragonat 0x09006600 I really find a table, but with wild entries
00:13:57amiconnMaybe your code doesn't include anything that requires interrupts?
00:14:11[IDC]Dragonthe drivers do
00:14:13amiconnI just looked at the vector table of my current ondiosp binary
00:14:34amiconn_most_ entries also look as you describe, only the base address is different
00:14:56amiconnThat is without sectioned compilation
00:15:07[IDC]Dragon0x09006600 is within UIE for my build
00:15:41[IDC]Dragonstrange, I will continue later
00:15:50 Quit [IDC]Dragon ()
00:17:33preglowall the float functions in dumb will have to be replaced altogether
00:18:20preglowhcl will be doing this :PPP
00:18:35BagderI just get bored by all the warnings
00:18:40Bagderso I nuke a few
00:18:48preglowyes, sure
00:20:36amiconnBagder: I can't see a reduction of warnings by your dict.c commit...
00:21:21Bagderno, I don't think the cvs build table shows that
00:21:27BagderI got it locally though
00:21:50amiconnDo you compile with even more warnings enabled?
00:22:08Bagderno, but I have a different set of gcc versions
00:24:43amiconnTry with gcc 4 :P
00:25:01Bagdersuicide mission ;-)
00:25:22amiconnyou'll get a heapload of new warnings
00:25:32preglowpointer signedness
00:25:35amiconn...which are correct, btw
00:25:40amiconnpreglow: yups
00:25:46preglowyes indeed
00:25:47 Join LinusN [0] (
00:25:56BagderI guess many of those can be enabled already with 3.4
00:26:51*Bagder nominates midi2wav.c as the ugliest piece of code in Rockbox
00:27:23preglowi'm sure stevenm will be honoured
00:27:28 Quit Harpy (Read error: 145 (Connection timed out))
00:27:30Bagder#including 5 .c files
00:27:56preglowyes, it's a temp thing as of yet, i believe
00:28:01preglownot sure if it'll end up as a codec
00:28:10preglowbut still
00:28:16preglowpretty it is not
00:29:49***Saving seen data "./dancer.seen"
00:30:02preglowi wonder what became of the wavpack guy
00:31:45amiconnBagder: The changelogs in the daily build table are sometimes off, plus the changes sometimes show up too early if something is committed in the middle of a build
00:32:17Bagderthe changelogs are from the last 20 minutes
00:32:45Bagderif a change in CVS is detected
00:32:52amiconnI mean, the changelog sometimes show the same change twice
00:32:53Bagderwhich it does by actually cvs update
00:33:06Bagderthat's because it always uses 20 minutes backlog
00:33:16Bagderand a build is typically 13-15 minutes
00:33:33amiconnImho that's somewhat irritating
00:33:47amiconnI'd expect it to show the changes which triggered the build
00:33:59Bagderwell, it doesn't know that
00:34:09Bagderit updates and see a diff => rebuild
00:34:23preglowLinusN: are you responsive at the moment?
00:34:49Bagderbut I agree it is slightly annoying
00:35:01LinusNpreglow: sure
00:35:05amiconnLinusN: Btw, I got your box to load rockbox from disk (by reformatting and putting back archos.mod as the first file
00:35:15LinusNamiconn: good
00:35:38preglowLinusN: i tried clearing both CACR and ACR0 and ACR1, shouldn't that disable the code cache?
00:35:40amiconnNow I'll wait for that plastic cover to continue...
00:36:03amiconnPlease don't forget to include the button caps and the screws...
00:36:43Bagder16 warnings less
00:36:55preglowwell, i can detect no differences in execution time
00:37:27preglowbut i also can't see anything wrong in the cache setup
00:37:40Bagderbtw, I have the endian stuff finally fixed in my local build now
00:37:53amiconnNice :)
00:38:00LinusNpreglow: clearing CACR should do it
00:38:08amiconnLinusN: Did you read about my dead code digging?
00:38:21LinusNyes, some of it
00:38:25preglowLinusN: then ACR0 must be set incorrectly, or something, because i still get exactly the same codec run times
00:38:44HClsounds promising :)
00:39:09preglowbut it looks correctly set up to me
00:39:14preglowthen again, i don't really know much about this
00:39:57*Bagder decides to go for it now
00:40:06Bagderstand by for biggish commit
00:40:14*preglow gets coffee
00:40:19*LinusN crawls under the table
00:40:24amiconnLinusN: I'm not sure what to do about all that dead code.
00:40:34LinusN"all that"
00:40:49LinusNhow much is it, and what do you want to do with it?
00:41:04amiconnThe functions which are completely unneeded for some platforms should be disabled imho
00:41:38Bagdernow update and tell me what breaks! ;-)
00:41:57amiconnHowever, there are some functions which are currently unused on *all* platforms, but which shouldn't go away imho
00:42:41amiconnWe could switch the build to use sectioned compiling and link with garbage collection like I did as an experiment...
00:43:16amiconn...but (1) I couldn't test on iriver, maybe there are problems and (2) this encourages laziness...
00:43:31preglowBagder: works just dandy here
00:43:52Bagderthe build dir is less cluttered now
00:44:05preglowso i see
00:44:19amiconnLinusN: look at my comprehensive table, the gain is 1046...2090 bytes atm (SH platforms only)
00:44:32amiconnThe completely unused functions are marked in dark red
00:45:40HCloh. oops. no wonder my invertrect didn't work >.>;
00:45:55preglowHCl: i take it dumb is not a priority?
00:46:07HClpreglow: not very.. at the moment.
00:46:18HClsearch engine/database needs a tiny bit of work.
00:46:30HClafter that i should be free to look at that, or something else for that matter
00:46:48preglowsomeone else working on the codecs would be nice...
00:46:59LinusNamiconn: where is this comprehensive table?"> I think you're reading backlogs ;)
00:48:33HClcan m$ office open that?
00:48:47amiconnHCl: nope
00:48:51HClpreglow: i'll take a look at it after i'm done with rundb
00:48:59amiconnBut oo can save as msoffice :P
00:49:18preglow.txt for ever!
00:49:38amiconnSpreadsheets as .txt ???
00:49:42preglowok, pdf!
00:50:07Bagderthe adobe reader is soon the same size as open office ;-)
00:50:23preglowyes, it's grown quite a bit
00:50:33preglowunfortunately it is also the only pdf reader that i can use without screaming in agony
00:50:44amiconnOpenoffice is still smaller than ms office ;)
00:50:51Bagderit uses *96* megabytes after install
00:51:20preglowhaven't had use for any office suites for a while
00:51:53HClmy code works :3
00:53:07amiconnBagder: reconfiguration required?
00:53:40 Join Moof [0] (
00:54:32Moofthis port of rockbox to the iRiver sounds cool. Is there anything as a non-C-friendly guy I can do to help?
00:55:05Bagderyeps, learn C and join in ;-O
00:56:19BagderI don't know, I guess code is what we lack the most atm
00:56:51Bagderor someone more than Linus with a bdm and h3x0
00:57:04preglowyup, can't think of anything non-code-related that needs doing
00:57:13Moofoh well
00:57:21Moofmind if I pitch in an observation though?
00:57:28Bagderfire away
00:57:33Moofreading the audio API bit, it doesn't seem to have much about seeking in it. Are you goign to be able to start playing a file half way through?
00:58:05Bagderthat's a MUST
00:58:09Moofright now my biggest gripe with my iRiver is that if I lose my place in an audiobook, I have to seek through 4 hours of stuff to get to it
00:58:23Bagderwe already do mid-song resume on Archos
00:58:25 Join Strath [0] ( bookmarking
00:58:42Moofbrilliant ;)
00:59:20Moofand can I put in a cheeky request for Speex on the list of codecs? ;)
00:59:27BagderI don't think Linus followed that proposal very closely when he wrote the initial code
00:59:41Bagderat least it didn't look like that too me ;-)
00:59:45Bagderto me
00:59:45MoofBagder: aaah
01:00:13BagderMoof: the problem is not to add codecs, we all want all codecs, the problem is that we need someone to work on it
01:00:26Mooffair enough
01:00:28 Quit Sucka ("a bird in the bush is worth two in your house")
01:00:30MoofI did say it was cheeky
01:00:42 Join Sucka [0] (
01:00:48Bagderamiconn: build working fine?
01:00:55amiconnYes it does
01:01:10amiconnHowever, my way of updating my boxes got a bit more difficult :/
01:01:24Bagderwhat way is that?
01:01:56amiconnNow I need to go 2 dirs down to grab the plugins
01:02:14Bagderwrite a script for it!
01:02:34amiconnHmm, probably I should do that
01:02:41Bagderfind . -name "*.rock"
01:02:57Moofanyway, I'll try to keep up with the wiki. when you need spanish translators or stress testers or UI testers or other things a non-C guy can do, stick something up there :)
01:03:33BagderMoof: we'll do that for sure
01:03:47amiconnBagder: It's not that simple. Probably the easiest way would be to start from the makezip script
01:03:47BagderI think the spanish translation is fairly up-to-date
01:04:03Bagderamiconn: yes, that has a somewhat extended check
01:04:08Moofthanks guys. Keep up the great work :)
01:04:11amiconnBasically I want almost what this script does, only that I don't want all languages
01:04:36amiconn...but only english and german for me, and additionally swedish for my sis
01:04:53Bagdershould be an easy edit
01:05:13amiconnAnd of course it shouldn't zip the result, but put it on the box
01:05:29amiconnIt would need a way to know the drive letter....
01:06:52amiconnDid you check whether simulator builds still work (especially the 'make install' part)?
01:07:13BagderI tried make zip
01:07:19Bagderso I think it works
01:07:56amiconnYellow and red builds...
01:10:18preglowLinusN: and yeah, stuffing the main imdct_l loop in iram speeded stuff up from 180% realtime to 220%, i really don't think there should be such a dramatic speedup if the cache was working properly
01:11:23*LinusN is running some cache tests
01:15:55 Quit Sucka ("a bird in the bush is worth two in your house")
01:17:20LinusNinitial tests show that the cache does have impact
01:17:50LinusNfor(i = 0;i < 10000000;i++){} takes 84 ticks with cache and 2163 without :-)
01:18:03preglowyes, i too am pretty certain the cache works :/
01:18:14preglowbut i'm still puzzled why i can't seem to turn it off
01:18:50preglowprobably did something wrong
01:19:09BagderI'm off to bed. If there are any remaining build problems and no one else fixes them, I'll continue working on the tomorrow evening
01:21:00 Quit einhirn_ (Read error: 131 (Connection reset by peer))
01:25:51 Join einhirn [0] (
01:26:18LinusNpreglow: how do i include your imdct_mcf5249.S in my build?
01:26:40 Join stripwax_ [0] (
01:27:13preglowyou have to replace the one that's already there, obviously
01:27:18preglowand declare ASO_IMDCT someplace
01:28:22preglowi just #define ASO_IMDCT if CONFIG_CPU==MCF5249 etc etc in layer3.c right now
01:33:52 Join asdsd____ [0] (
01:33:56 Part asdsd____
01:40:14*HCl goes to sleep and bites his editor.
01:40:18HClthat code is an utter mess.
01:40:23HClbut i have no idea how to cleanify it.
01:40:24preglowprettify it
01:40:36Adityaany of you guys have suggestions for good books for low level programming?
01:40:42Adityalike the kind of stuff you guys are doing
01:41:10preglownot all rockbox code is very low level
01:41:20LinusNhehe, i can't remember ever reading a book on that...
01:41:25preglowme neither
01:41:26*HCl considers designing his code in OO first to get it organized, then code it with structs and functions...
01:41:46Adityawell I want to help with the h3xx stuff.. but eh.. maybe I am getting too ambitious
01:41:55preglowHCl: if you've got it planned out how to work with oo, just using structs in the first place shouldn't be that hard
01:42:22preglowAditya: if you know nothing of low-level programming as it is, then yes, i think you are ;)
01:42:40LinusNAditya: i'm afraid nobody will be able to work on the H300 until i have ported the bootloader to it
01:43:05preglowand most of our code should work well
01:43:08stripwax_well, it's possible that someone else could port the bootloader ;-)
01:43:23LinusNport it, yes
01:43:57LinusNstripwax_: is the encryption the same btw?
01:44:19stripwax_yep. different key but the ihpfirm sources handle it anyway
01:45:23Adityawell I know how to program in C.. but I am not familiar with firmware and such
01:45:57AdityaI guess I will just wait a couple of weeks, get one of those BDM wigglers and fool around
01:46:09Adityayou can always flash a dead unit if you have a BDM wiggler right?
01:46:14preglowAditya: indeed
01:46:24Adityaor is there any way I can completely destroy my iRiver?
01:46:25LinusNAditya: do you know how to connect it?
01:46:25preglowAditya: but you really don't need one once the bootloader is done
01:46:42HClah, much better
01:46:48amiconnI know what causes the warnings in bmp.c. It's a design error... readshort() / readlong() are functions taking a pointer and returning a short/long for big endian, but they're replaced by identity macros for little endian....
01:46:54Adityawhen will this boot loader be done? anytime soon?
01:47:01preglowprobably not
01:47:12LinusNi'm a busy man
01:47:12AdityaI wanted to do something different over the summer anyways so I will prob mess around with it either way
01:47:32Adityaand hopefully not blow up my H320 :P
01:47:44preglowi haven't managed to have my h120 explode yet, at least
01:47:50stripwax_Aditya - it's always possible to kill your iRiver by electrostatic discharge ;-)
01:48:02Adityastop scaring me!
01:48:05LinusNbut i'll probably try to connect the BDM to my H320 pretty soon
01:48:13Adityaheh.. I did ruin my mobo once though..
01:48:17preglowwith a bit of luck, little work will be required
01:48:23Adityaalthough that was my bro's fault..
01:48:43Adityaidiot comes in with his fur coat..
01:48:45preglowsomeone's found the lcd display used as well, i see
01:48:50Adityawhoop de doo. mobo goes to hell
01:48:57stripwax_Linus - oh cool, didn't realise you'd already sourced an H3xx. is it US or non-US model (believe the hardware's different)
01:48:59LinusNpreglow: hopefully, yes
01:49:22LinusNhe didn't mention why he was so sure that was the one
01:49:33LinusNand we still don't have the data sheets
01:49:36Adityaso I am guessing the main thing you guys are missing for H3xx are the drivers for the actual hardware?
01:49:47preglowAditya: i think we've got most of them
01:49:49LinusNstripwax_: it's a US model
01:49:51preglowAditya: apart from the lcd driver
01:50:06stripwax_LinusN - so no USB-on-the-go (is that right?)
01:50:18LinusNstripwax_: well, i can mod it
01:50:18preglowstripwax_: can be enabled with a simple latch, i think
01:50:25stripwax_heh ;-)
01:50:33AdityaI have a US model too.. and I can open it up over the summer if needed.. but I think linusn will prob have better luck doing that and not killing the player
01:53:31preglowi just had to open mine
01:53:39preglowopenable things tend to be opened around me
01:55:08*HCl writes pseudocode and goes to sleep
01:56:07amiconnThe bmp.c warnings are tricky :( I would like to replace these functions with the neat SWAB32/SWAB16 macros, but can't do so because the bmp file header isn't properly aligned
02:00:04MoosCamarogood night (and other) all
02:00:53 Part MoosCamaro
02:01:17 Join ashridah [0] (
02:02:29stripwax_gotta go.
02:02:32 Part stripwax_
02:03:49preglowi like the fact that the coldfire init utility code generator doesn't invalidate the code cache
02:04:38preglowoh, it does
02:09:48LinusNinit utility code generator?
02:10:05preglowyeah, there's a utility to generate coldfire init code
02:11:15preglowjust wanted to see if its idea of cache disabling code was the same as mine, heh
02:11:37LinusNdoes it enable it in the same way as i do?
02:12:36preglowinvalidate with 0x01000000, then 0x80000000 to enable cache
02:12:39preglowthen set up acr
02:12:57LinusNand the acr is set up in the same way?
02:13:52preglowdoesn't let me use base address 0x31000000
02:14:08 Quit cYmen_ ("zZz")
02:14:35preglownope, complains the base address has to be a multiple of the memory region size
02:15:14LinusNthat is generally true, the 31000000 is a hack to work around the A24 bug in the chip
02:17:06preglowbut yeah, i can't see anything wrong
02:18:45preglowbut unless the cache mode is reset some other place, i don't understand why i can't measure a difference
02:19:46LinusNasm(" move.l #0,%d0\n movec %d0,%cacr");
02:20:05preglowthat's what i'm doing
02:23:00preglowit seems to boot slower
02:27:26preglowit almost certainly does boot slower when i disable the cache
02:29:07ashridahwhoa. you people have been busy in the last four and a half hours
02:29:24preglowso i get a difference, just not when using codecs
02:29:31preglowashridah: bagder has been busy ;)
02:29:50***Saving seen data "./dancer.seen"
02:30:48LinusNpreglow: i did a cache test in helloworld.rock
02:31:06LinusNthe cache gave great performance there too
02:31:31LinusNso there must be something about your code that makes the cache impact small
02:31:55preglowi just don't see why i can measure _no_ difference
02:32:02preglowthere should be tons
02:33:14LinusNmaybe not tons, but there should be a difference...
02:33:30preglowthere is a noticable difference when booting
02:33:40preglowthat should translate to a measurable difference, and that easily
02:34:59preglowalmost seems like the cache is in freeze mode
02:35:05preglowbut it's not
02:39:36preglowinvalidate_icache() enables the cache
02:39:40preglowi didn't think of that
02:40:01preglowsmall wonder there's no difference, then
02:40:23preglowthe cache is enabled before a plugin is loaded
02:40:26preglowvia that function
02:43:30preglowwrong window, yes
02:48:41preglowwell, well
02:48:48preglowi have no idea whatsoever makes the code slow
02:49:02preglowconsidering everything i've learned so far, it should be faster
02:51:55LinusNwe'll find out soon enough
02:52:05LinusNi have to go to bed now
02:52:11LinusNcu around
02:52:13preglowyes, me to
02:52:17 Part LinusN
02:52:17 Quit preglow ("leaving")
03:11:11 Join t0mas [0] (
03:11:24t0masnight :)
03:14:05 Join DMJC [0] (
03:22:37 Nick ashridah is now known as Lost-ash (
03:32:32 Join ]RowaN[ [0] (
03:32:49]RowaN[can someone tell me where i can get ? the link i found for it seems to be dead
03:33:07t0masyou can compile it...
03:33:18t0masget the patch from the patch tracker...
03:33:30t0masand the source from cvs, or a daily tar.gz
03:33:53]RowaN[that sounds too difficult for 2:33 am
03:34:03t0mas3:34 here ;)
03:34:22]RowaN[i just installed rockbxo on my ihp-120 you see
03:34:34]RowaN[i thought it would have jpg & rockboy plugins included but it doesnt seem to
03:34:54t0maswell... if you installed... you have a build environment?
03:35:19]RowaN[nah i downloaded premade bootloader firmware
03:35:28t0masfrom where?
03:35:37]RowaN[err.. google.. some forum
03:35:54t0mashm... shouldn't be there..
03:36:17 Nick Lost-ash is now known as ashridah (
03:36:19t0masrockbox isn't ready... and "we" will never release a complete .hex file
03:36:22t0mascopyright issues
03:36:26t0maswb ashridah
03:36:38]RowaN[yeah i know its not made by you guys
03:36:51]RowaN[and i know the risks flashing my player
03:37:53 Join XShocK [0] (
03:38:25]RowaN[erm, so anyway, those plugins, jpeg & rockboy.. not included in the normal daily builds right? or am i missing something
03:41:07 Quit ]RowaN[ ("i love sleep")
03:41:29t0masthey are included... but I don't know what he downloaded :)
03:57:09AdityaI take this moment to say threads suck
04:06:04 Join QT_ [0] (as@area51.users.madwifi)
04:06:10 Quit Seed (Nick collision from services.)
04:06:18 Join Seed [0] (
04:06:44*ashridah huggles threads, just to be contrary
04:09:45Adityathey are cool when they work
04:09:49Adityabut to program them.. its a bitch
04:09:54Adityasynchronization and such
04:10:17ashridahwell, granted, you need to be able to keep track of resources properly
04:10:24ashridahwhich can get convoluted
04:11:25Adityaespecially when you are dealing with 4 datastructures and n number of threads.. where you don't know what n is
04:11:51ashridahooh. do they all need access to the four structures in random orders as well?
04:12:26Adityathats what I am making
04:13:50ashridahlooks more fun than the crap i'm doing atm
04:14:45ashridahsoftware engineering lecturer has us trying to solve an intractible problem in a tractible way
04:17:19 Quit QT (Read error: 113 (No route to host))
04:26:51 Join kaltek [0] (
04:27:37kaltekhey i am having trouble with me studio jukebox can any1 help me?
04:28:24Adityawhats the problem?
04:29:27kalteki think its being reconized as a unknown device but i cant get drivers installed
04:29:53***Saving seen data "./dancer.seen"
04:30:08kaltekits says usb active
04:31:11kaltekbut when i try to install the drivers from the archos website it doesnt like them
04:40:53 Quit kaltek (Read error: 131 (Connection reset by peer))
04:45:20 Quit ashridah ("Leaving")
04:53:16 Join kaltek1 [0] (
05:28:43 Quit XShocK (" HydraIRC -> <- The future of IRC")
05:37:15 Quit kaltek1 (Read error: 60 (Operation timed out))
05:39:35 Part Aditya
05:41:32 Quit dwihno (Read error: 110 (Connection timed out))
06:29:57***Saving seen data "./dancer.seen"
06:41:51 Quit ehntoo ("Leaving")
07:37:29 Join kaltek [0] (
07:56:54 Join ashridah [0] (
08:25:31 Join JackDaniels85 [0] (
08:29:59***Saving seen data "./dancer.seen"
08:34:04 Quit kaltek (Read error: 60 (Operation timed out))
08:41:47 Quit courtc (Read error: 110 (Connection timed out))
08:45:21 Part JackDaniels85
08:47:21 Join JackDaniels85 [0] (
09:10:52t0masmorning again
09:11:10t0masgetting to sleep at 5 is ok... if you can just keep on sleeping for at least 5 to 6 hours
09:12:18JackDaniels85I have too. I have sleep 4 hours. :-)
09:12:44t0masyes, and I need coffee
09:13:06t0masnote to self.... don't drink to much when you have to get up early
09:15:06JackDaniels85hi hci
09:17:48HCli have a bit of a problem..
09:17:58*HCl will just have to adjust databox.
09:18:37HClhow do i call the virtual keyboard to give me an inputted string back?
09:18:55Rickrockbox has a standard vkeyboard now?
09:19:09HCldidn't it have one for a long time? o.o
09:19:16Rickno idea....
09:19:29 Part JackDaniels85
09:21:01HClamiconn: ?
09:22:07HClrb->kbd_input ?
09:22:21*HCl will try that :)
09:22:46Rickis that a fullscreen vkb?
09:37:35t0masit has Hadaka
09:37:37t0maseh HCl
09:38:37t0masrb->kbd_input(inbuffer, sizeof(inbuffer));
09:51:34HClyea, thanks.
10:02:14 Join courtc_ [0] (
10:02:57 Nick courtc_ is now known as courtc (
10:17:09 Join Harpy [0] (
10:21:03 Nick QT_ is now known as QT (as@area51.users.madwifi)
10:30:00***Saving seen data "./dancer.seen"
11:01:11 Join amiconn_ [0] (
11:01:25 Quit amiconn (Nick collision from services.)
11:01:26 Nick amiconn_ is now known as amiconn (
11:10:44 Join preglow [0] (
11:20:53 Join Sucka [0] (
11:31:46 Join kergoth [0] (
11:31:48 Part amiconn
12:02:56t0maswhat was the cvs commando to update new dirs also?
12:04:13t0masoh nevermind
12:04:14t0masgot it :)
12:30:03***Saving seen data "./dancer.seen"
12:43:15 Join MoosCamaro [0] (
12:43:37MoosCamarohi all
13:04:21 Join amiconn [0] (
13:15:12 Join cYmen [0] (
13:47:48 Nick Sucka is now known as Sucka`away (
14:01:39 Join Aditya [0] (
14:02:38Adityawhat's up.. is this when people are awake or when everyone's sleeping? :P
14:03:01MoosCamarohi Aditya
14:03:39MoosCamaroF1 race
14:05:02rasherAditya: some people awake I'd say
14:06:14t0masit's day in europe
14:06:48t0masand afaik the biggest part of the people here are from europe...
14:07:38AdityaI am from the US.. but eh.. I am up at the most odd times.. so its ok
14:07:50Adityabonjour mooscamaro.. cava?
14:07:52t0masbut some people here are too ;)
14:07:58Aditya(I know some french btw)
14:08:03t0masJe ne parle francais pas ;)
14:08:04Adityaand of course like to show off
14:08:11Adityasure you dont tomas
14:08:17MoosCamaroou quel français
14:08:22t0maswe have to learn it :P
14:08:43MoosCamarosalut Aditya, ca va?
14:08:53t0masalmost all dutch people speak some french
14:08:56Adityaje suis bien, et toi?
14:09:00t0masor at least should ;)
14:09:07MoosCamarotrés bi1 merci
14:09:16rasheramiconn: that's crazy :)
14:09:22Adityayou're using all weird characters..
14:09:29rasher(credits scroller)
14:09:31t0mas C'est unicode ;)
14:09:45t0masou utf-8 ?
14:09:51amiconnrasher: really?
14:10:02rasherNo, it's great.. how was it previously?
14:10:11MoosCamaroI watch F1 GP of Spain
14:11:18amiconnrasher: Previously it just displayed 2 names (one per line) for a second, then advanced to the next 2
14:11:36rasherAh, boring :)
14:11:40amiconnIt was non-scrolling, so longer names were displayed incomplete
14:11:51rashereven worse!
14:11:53AdityaI am on unicode
14:12:47HClshouldn't that block till a button is received?
14:13:03amiconnyes, almost
14:13:26HClits immediately passing through on my iriver..
14:13:39amiconnIt returns when either a button press, release, or repeat is received, or a system event like usb connected etc
14:14:01HCli'll just make my own buttonpress thing..
14:14:07Adityaits that high level?
14:14:11Adityalemme at it!
14:14:12amiconnHCl: No need to do so
14:14:31amiconnYou should just take care of releases/ repeats
14:14:52amiconnIn a plugin, most often the very first even you receive is a release
14:15:18amiconn...of course. It's the release of the button press with which you started the plugin
14:19:23HCltoo late
14:19:30HClalready wrote my own buttonpress thing.
14:19:42Adityaheh..gotta love OSS
14:20:09Adityahow exactly did you guys manage to write a gameboy emulator lol
14:20:21t0masHCl did
14:20:34HClpeople were joking about it :p
14:20:41HClso then i ported gnuboy and stripped it a bit
14:20:54*t0mas starts jokinh about wifi suppoer
14:20:55Adityaman.. you are sooo my hero lol
14:20:56HCland wrote an lcd and button driver..
14:21:03HClt0mas: :P
14:21:19Adityacause I always wanted to port an emulator once I got caught with the programming details
14:21:51Adityaoh btw.. can I already start tweaking the H1xx code if I have a BDM wiggler?
14:21:56Adityafor H3xx that is
14:22:03t0masyou can try
14:22:19t0maswith a bdm you can start writing a bootloader
14:22:39t0maswhen that works you can start porting drivers :)
14:22:58AdityaI will most likely just mod the current stuff
14:23:06t0masyes, that's whay I ment ;)
14:23:10AdityaI am not experienced enough to write it from scratch :P
14:23:15amiconnHCl: Doing your own button handling isn't a good idea imho. If you do that, you have to take care about numerous things yourself...
14:23:25t0masmake the bootloader work, and port the H1xx driver suff to H3xx
14:23:37HClamiconn: like yield?
14:23:39preglowAditya: if you're not experience with low-level stuff, steer clear of getting a bdm the responsiveness is usually worse
14:24:05HCli use the same button driver as the one i wrote for rockboy
14:24:07HClit should work fine.
14:24:10rasherAditya: But feel free to buy one and donate it to someone who is :)
14:24:12amiconnHCl: Like queueing, receiving buttons while the gui thread is busy...
14:24:24HClyea, well, blah :/
14:24:40amiconnHCl: The rockboy button handling is another story, as it needs to be realtime.
14:24:55amiconnIt has its drawbacks though, like e.g. not handling usb
14:25:16t0masHCl: just get w/o blocking the first time
14:25:22t0masand start blocking after that
14:25:31HClt0mas: hrm.
14:25:38t0mashrm = ?
14:25:42HClwell, just gonna try my current thing.
14:25:46amiconnThere are tons of button event loop examples in the current rockbox code
14:25:51HCli don't like the whole queuing anyways.
14:26:31preglowyou don't like queueing? :P
14:26:33amiconnI couldn't live without queueing. The archos firmware does this. It sometimes feels really sluggish because of that
14:26:49Adityapreglow.. I canot really kill my iriver can I? I mean.. I am not dumb.. just dont know much about firmware.. but I am going to read up first :P
14:27:03preglowAditya: what, you mean using a bdm?
14:27:03HClpreglow: sometimes when software lags, and it queues the keys
14:27:08HCland you've pressed like lots when it doesn't respond.
14:27:10amiconnLike, not reacting on a keypress just because it started reading the disk at the same moment...
14:27:11preglowHCl: yes, and that'a good thing
14:27:14HClit'll go process the entire queue.
14:27:17t0masAditya: how hardware knowledgable are you? :)
14:27:24HClinstead of dropping the keys while it was busy.
14:27:35HCli wouldn't want a queue bigger than 2
14:27:35amiconnHCl: Why should it rop them?
14:27:37preglowwell, you never should need to press a button several times
14:27:39preglowthat's the whole deal
14:27:48HClamiconn: because i've had times where the system was busy, cached all my keypresses.
14:27:48preglowyou know it'll be handled anyway
14:27:51HCland i spend the next 5 minutes
14:27:57HClhaving to watch it going up and down my playlist
14:28:01HCltill it finally responds again.
14:28:01rasherwell you shouldn't have pressed all them keys :)
14:28:13Adityayes bmd, using a bdm wiggler.
14:28:14HClwhen something doesn't respond i press lots of keys.
14:28:25Adityatomas : I am pretty good with hardware in general
14:28:27preglowAditya: well, you need to solder the thing to your player, of course you can destroy it
14:28:32amiconnI like being able to press buttons beforehand, like answering the resume request before it's even displayed, when I know I want to resume
14:28:47t0masAditya: well... see preglow's comment... you have to solder it...
14:28:50Adityawell yeah.. I COULD.. but I have soldered too many things in engineering :P
14:29:06t0masok, then it isn't a proble
14:29:16HClis the lcd_clear_display broken?
14:29:20preglowAditya: but my advice is, if you're not very knowledgable about very low-level debugging, don't get a bdm
14:29:25amiconnHCl: Nope
14:29:29HCli do the onscreenkeyboard, but it doesn't clear my iriver screen..
14:29:43HCl if(rb->kbd_input(filename, 100)) {
14:29:45HCl ....
14:29:48t0masHCl: true
14:29:51HCl rb->lcd_clear_display();
14:29:52Adityapreglow: I know.. I prob wont do anything now.. just going to read up first
14:29:53HCldoesn't work
14:29:55preglowAditya: if you aren't able to write a bootloader, you won't get to use it anyhow
14:29:58t0massee my dict plugin... it has the input stuff you want...
14:30:04***No seen item changed, no save performed.
14:30:11Adityathen do shit on random stuff like my clock first :P
14:30:11amiconnHCl: You probably forgot rb->lcd_update() afterwards
14:30:15HCli got input settled...
14:30:31HClamiconn: no, i lcd_update, and it just writes my text on top of the onscreen keyboard
14:30:43HCldo i have to explicitly lcd_update immediately after a clean?
14:30:53amiconnOf course
14:31:05HCli figured it would just set the internal buffer to 0
14:31:12amiconnWell, not if you print something after the clear
14:31:13HCland then i can write to it
14:31:20amiconnYes it does
14:31:21HCland lcd-update
14:31:27HClthen whats going wrong? :/
14:32:10amiconnI don't know
14:32:11*HCl will just add an update right after the clean, though it shouldn't really matter...
14:35:14HClit helped..
14:35:18HCli don't understand why it helped though.
14:40:30preglowpft, not important
14:49:10amiconnBagder: r u there?
15:12:12HClsearchquery editor is processing steadily
15:12:22*HCl is hoping to finish it today
15:16:48HClkind of odd that the editor for the searchengine is actually more complicated than the searchengine itself.
15:16:55preglownot really
15:17:04preglowuser interfaces are pretty complex
15:17:07preglowto code, that is
15:17:32preglowi've seen plenty of user interfaces that actually consist of more code than what it's supposed to wrap
15:18:26rashersomeone port glade to rockbox
15:18:35HCli'm just hoping that sometime someone will prettify the interface.
15:18:52HCleditor is *almost* done.
15:18:56preglowsomeone ought to make widgets
15:18:56HClworking great so far
15:19:01preglowthat would be pretty cool
15:19:43 Join ]RowaN[ [0] (
15:20:21preglowam i the only one that thinks sideways text scrolling is too fast for the iriver lcd?
15:20:36HClno idea
15:20:37]RowaN[slow it down then?
15:20:49HClyea, widgets would be nice :/
15:20:54]RowaN[can someone point me in the direction of grayscale.tgz ? (rockboy greyscale hack)
15:21:17amiconnpreglow: You can adjust the scrolling to your likings
15:21:32preglowyeah, i know, but the default is too fast!! :>
15:21:49amiconnReally? Then change the default
15:21:51preglowdon't think i'll ever wait for the scrolling to start during normal use anyway
15:21:53amiconn...for iriver
15:22:20amiconnOn archos, I actually bump it up a little from the default
15:23:50]RowaN[i renamed a mp3 file from the rockbox gui, and now i cant seem to delete that file.. any ideas ppl?
15:24:27Adityaits a feature in the shareware version
15:24:41Adityayou pay them 20 dollars and your file will be deletable magically
15:25:08]RowaN[save it for #comedy aditya
15:25:25]RowaN[or #lameness
15:25:37Adityaseriously, I think something is not updating datastructures they are using right
15:25:45rasher]RowaN[: is it reproducible?
15:25:54]RowaN[err lets see
15:26:00t0mas]RowaN[: Again: Where did you download this copy?
15:26:05SlasheriHmm, interesting. I just made some code and now testing realtime mp3 playback on iriver :) Trying to play /sample.mp3
15:26:12t0masAs you don't have rockboy... it sounds like a weird build to me
15:26:21]RowaN[rasher: mysticriver forums
15:26:33AdityaI would get the latest one
15:26:37]RowaN[for the bootloader that is, then for the build
15:26:43t0masSlasheri: 15 seconds? :)
15:26:43Adityathat way bug reports would actually be useful
15:26:54t0mas]RowaN[: wich build from
15:27:01]RowaN[latest one from yesterday
15:27:02Slasherit0mas: yes :) sorry, i have been little busy..
15:27:06rasher]RowaN[: which build do you have? menu>info>version
15:27:28t0masSlasheri: no problem, I've been quit busy meself...
15:28:04Slasherit0mas: okay, great :) I will make soon the dict converter changes
15:28:13t0masok, I still have to look at scrolling
15:28:33t0mas]RowaN[: then you have rockboy
15:29:01]RowaN[yeah i figured out how to use rockboy, just had to enable viewing of non music files too
15:29:09]RowaN[which was off by default
15:29:22*]RowaN[ lives & learns
15:29:31t0masand for grayscale... use the patch... and if you don't know how to do that... learn it.
15:30:15]RowaN[lerning? on a sunday?
15:31:40]RowaN[ah, error doesnt reproduce.. i just renamed another file within rockbox, and im able to still rename/delete it
15:32:00]RowaN[so im just stuck with one undeletable file heh.. lets see what scandisk has to say
15:32:19*t0mas prods amiconn
15:32:30t0masmaybe an intresting bug in the fat stuff?
15:32:45rasherthere seems to be a few quirks with the fat, yes.. but it also seems hard to narrow down
15:33:23]RowaN[scandisk didnt see any probs
15:33:33]RowaN[is there any reason why i shouldnt defrag my iriver hd?
15:33:50HClmy ftp is temporarily down.
15:33:53HClcause i moved it to port 2121
15:34:03t0masand forgot to update your firewall?
15:34:10t0masor can we just use it on port 2121?
15:34:10HClcause my parents' isp blocks port 21 to my server :/
15:34:15HClyou can use it on port 2121
15:34:46rasherThat's one horrible isp :(
15:36:41]RowaN[in dos the undeletable file doesnt show up at all.. shows in windwos explorer tho
15:37:19rashersure it's not listed as some weird filename?
15:37:37amiconnSounds like the dir entry/entries are incorrect somehow
15:37:47]RowaN[ah yes it is listed in dos
15:39:20SlasheriHmm, how can I allocate large amount of static memory? I think that static unsigned char codec_memory[CODEC_MEM_SIZE]; is not the right way..
15:39:42t0masSlasheri: no that's not the right way
15:39:47HCluse the mp3buffer.
15:39:52t0masbut from where?
15:39:54t0masand what do you want?
15:40:04t0masfrom a small plugin you can use the plugin buffer... for some data...
15:40:30t0masand for a lot of data... use mp3 buffer... but that will kill playback...
15:40:32HClplugin buffer is quite big on iriver
15:40:33Slasherit0mas: Hmm, okay. Now I am trying to write some dynamic memory allocation stuff for codecs
15:40:46t0masHCl: I know, I used it
15:40:57t0masSlasheri: that's something discussed a lot...
15:41:01HClit was mostly made that big so rockboy would fit ;x
15:41:30t0masa long time ago there was some malloc() function
15:41:33t0masBagder wrote it...
15:41:39t0masbut it was removed... as it wasn't needed..
15:42:33preglowdoes codecs really need dynamic memory allocation?
15:42:51SlasheriHmm, yep. At least some of the codecs needs malloc etc, unless the codec is modified
15:43:12Slasheripreglow: current libmad implementation at least needs
15:43:13preglowi think we'll try to remove as many mallocs as we can
15:43:37t0masSlasheri: Bagder wrote a good algorithm for allocating the memory in blocks... and being able to free them again... and if you just want to allocate something.. I have a _very_ simple malloc() using the pluginbuffer in dict.c
15:44:07Slasherit0mas: great, i will check that :)
15:44:21t0masfor useful things you should check the one Bagder wrote
15:44:35preglowSlasheri: i think we've removed all the mallocs in libmad
15:44:41t0masit's somewhere on
15:44:46preglowSlasheri: not actually removed, but the actuall mallocs aren't called anymore
15:45:09amiconnt0mas: It's also in rockbox cvs, in one of the 'attic' directories
15:45:19t0masah ok, I missed that :)
15:46:00t0masamiconn? playing graveyard administrator today? ;) "More dead code..."
15:46:02Slasheripreglow: Hmm.. I tried to compile the libmad but linker gave some errors about undefined codec_alloc, codec_free, memmove etc.
15:46:10amiconnt0mas: Haha, yes
15:46:15amiconngtg, cu
15:46:19 Part amiconn
15:46:41preglowSlasheri: what, you mean seperately? the libmad in the rockbox cvs compiles just fine
15:47:43SlasheriYes, it compiles fine. But I tried to link it directly (for testing purposes) with rockbox firmware and then I saw the errors..
15:48:00preglowwell, of course
15:48:06preglowit needs to be linked with the plugin subsystem
15:48:09 Join arfo [0] (
15:48:09preglowwhich has a simple malloc
15:48:21preglowthat is, the xxx2wav boys have
15:48:38Slasheriah, okay :)
15:48:44preglowbut anyway
15:48:50preglowi think we'll remove all the mallocs we can
15:48:58preglownot having to bother about dynamic memory handling is nice
15:56:51HClgood news
15:56:57HClthe editor is *almost* finished
16:03:12HClcan the player display inverted text?
16:03:37 Join |it|ChasKi` [0] (
16:04:29preglowmost charcell displays can
16:08:21HCldoes anyone know why kbd_input loads a lot of crap chars when i boot it at first
16:08:57HCli'm guessing its my button handling?
16:09:08|it|ChasKi`Anyone got rockbox running on a ihp140?
16:09:14HClahyea. my button handling doesn't clear the queue, ofcourse..
16:09:19HClyes O.o.
16:09:28*HCl is working on the searchengine on his h140
16:19:48 Join lodesi [0] (
16:22:02*HCl gnaws on his bug.
16:22:14*HCl captures it in a jar and releases it into the wild
16:28:41 Part arfo
16:30:05***Saving seen data "./dancer.seen"
16:41:21 Quit ashridah ("Leaving")
16:41:22 Quit MoosCamaro (Read error: 104 (Connection reset by peer))
16:41:46 Join MoosCamaro [0] (
16:41:50HClweeee. commit.
16:41:59*HCl waits for the daily builds to catch fire.
16:42:16*HCl prods t0mas
16:42:32HCli need people to test databox
16:42:37rasherI'll get right to it
16:42:57rasheranything I need to prepare to test it?
16:43:00 Join asdsd____ [0] (
16:43:46*t0mas wakes up?
16:43:53HCl :3
16:44:01*t0mas cvs-updates
16:44:06HCldon't forget -d
16:44:46 Quit thegeek (Read error: 131 (Connection reset by peer))
16:45:19rashersomething exploded
16:45:29HCl*goes to look at daily builds*
16:47:04rasherIn file included from /home/rasher/rockbox/build-env/src/rockbox/apps/plugin.h:36,
16:47:08rasher from databox.h:23,
16:47:10rasher from databox.c:19:
16:47:13rasher/home/rasher/rockbox/build-env/src/rockbox/firmware/export/config.h:23:22: autoconf.h: No such file or directory
16:47:27rashercould be my fault.. not sure why though
16:47:32*t0mas points at Bagder
16:47:35t0masoh wait
16:47:38t0masdid you reconfigure?
16:47:45t0massince.. eh... last week?
16:48:18rasherI did just now
16:48:39t0masand then it works?
16:48:56rasherno, I mean I did before that
16:49:17rasherlet me try again to be sure
16:50:12HClsounds like autoconf broke
16:50:15*HCl will try
16:51:11rashersame error
16:51:57t0masfirst line...
16:51:58t0masIn file included from <command line>:2:
16:51:58t0masexport/config.h:23:22: autoconf.h: No such file or directory
16:52:00preglowwhat're you building?
16:52:09t0masHCl's plugin
16:52:09rasherrockbox for iriver
16:52:12t0maswhere is that?
16:52:27t0massome place autoconf.h isn't included?
16:52:28rasherCC databox.c
16:52:38rasheris the line above
16:52:50*HCl is building...
16:56:40Adityabum bum bum
16:56:51Adityaanyone know good places to sell a laptop?
16:56:55AdityaI need to get rid of mine
16:57:14rasherdepends on which country you're in, I'd say
16:57:20HCl ebay
16:57:38AdityaI was gonna do ebay but they changed their fees.. bastards take like 4% of the closing cost
16:57:40rasherI'd guess ebay as well
16:58:00*HCl will fix errors.
16:58:03HClits due to the new makefiles..
17:00:18HClrasher, t0mas: try again?
17:03:49t0mascompiling ;)
17:04:05t0masmy laptop is pretty slow...
17:04:20t0masand I had to do a total recompile :)
17:04:26rasher"enter filename" .. of what?
17:04:57rasheralso, please lose the emoticon, it nearly overflowed with my font
17:05:03HClsorry :P
17:06:04AdityaI disable emoticons
17:06:09Adityaemoticons = evil!
17:07:31t0mashe ment in code ;)
17:07:33t0masor errors
17:08:20rashercan't I enter "end"?
17:09:02t0mashm... someone started using basename in rockbox? or I killed my basename command... and it was used before :)
17:09:53DMJCnothing in the GPL says you can't sell it
17:10:03t0masno DMJC
17:10:14t0masyou can sell GPL
17:12:58HClrasher: only when the search query is valid.
17:13:28rasherwell then I'm hosed
17:13:31rashercan't I go back?
17:13:32 Join bagawk [0] (~Lee@bagawk.user)
17:14:16HClyou can't cancel it when you selected a token to edit.
17:14:23HClbut you can just select the old token o.o
17:14:30HCland you can go back as long as you're not editing one
17:14:46HCljust left/right
17:14:56rasherthat just changes
17:14:59rasherwhat I'm on
17:15:11HClthen press select when you're satisfied and then you can go left and right.
17:15:16HClwhat do you mean with go back?
17:15:16t0masrec button = backspacew
17:15:51rasherHCl: then I just get a new thing to edit o.O
17:16:32rasherlooks like I'm stuck
17:16:40HClwhen you're at the end o.o..
17:16:43HClit might do that XD
17:16:56rasherit totally does
17:16:58t0masHCl? how does it work?
17:17:03t0masenter what filename?
17:17:04rasherI'll have to reset now
17:17:06HCllet me add the stop button stuff..
17:17:11HClyou don't.
17:17:15HCljust make it form a valid query.
17:17:19HCland then you can select end.
17:17:23rasherI can't
17:17:26HClwhy not?
17:17:35rasherwell it's invalid at the beginning, it seems
17:17:42HClthen yea:X
17:17:46HCli'll go fix that right now...
17:18:03t0masrasher: what did you type?
17:18:15rasherhow could I even possibly end up with an invalid query?
17:18:17t0masfilename of what? :)
17:18:24rasherdon't you check for valid inputs?
17:18:32rashert0mas: the searchfile generated, it seems
17:18:36t0masoooh ok
17:18:54t0masand then?
17:19:00t0masI can select <END>
17:19:03t0masbut what else?
17:19:54rasherit behaves weirdly
17:20:20rasherif I press right a bunch of times, it suddenly shows not
17:20:33rasherthen and
17:20:35t0masthat selecting a token
17:20:39rasherthen <end> a while
17:20:45rasherthen tokentyperror
17:21:14t0maswell I have a long query
17:21:19t0masbut I can't exit it :P
17:21:41 Join thegeek [0] (
17:22:20HClrasher: it has different editing modes.
17:22:24HClone enforces it to always be correct
17:22:30HClthis one allows easier editing.
17:22:40HCli added support for cancelling with the off button
17:23:07t0masHCl: but how should I use it??
17:23:22HClpretty much
17:23:32HClyou compose a search query by selecting the various possibilities
17:23:42HCland when you're done you input an <END>
17:23:44t0maswell... I'm to stupid...
17:23:45HCland it'll write the file to disk
17:23:49HCland then you can just click the file
17:23:52t0masstep 1: enter filename (what? nonexistant?)
17:23:55HCland it'll get opened with the searchengine
17:24:01HClyou can feed it test
17:24:02t0masok, done
17:24:03t0masand then?
17:24:05HClit can exist or non exist
17:24:15HClthen select something like "year"
17:24:22HClno extension needed
17:24:28HClit'll add .rsp by default
17:24:42t0maswell... it comes up with <end>
17:24:54t0maswhat button should I press to get to something usefull?
17:24:55HClyou should be able to move the joypad
17:25:00HClto change it into something else
17:25:01t0masleft/right or up/down?
17:25:06HClall will work ;p
17:25:14t0masyes, but I did that the first time
17:25:20HClleft/down do the same and up/right do the same.
17:25:23t0mascame up with and, or, not
17:25:30HClthats not correct o.o;
17:25:40rasherwell that's what happens
17:25:42*HCl checks
17:25:44t0masleft/right does nothing
17:25:50t0masup does something... sometimes
17:25:57t0masdown does nothing... (untill now)
17:26:23HClmy bad.
17:26:24t0masand a lot of tokentype errors
17:26:35HCli made a tiny screwup when trying to make the interface nicer
17:26:38HClthat happens with new files
17:26:38HCl ;x
17:26:51*lodesi is away: Occupé
17:27:16HClhold on
17:29:13rasherstop doesn't seem to quit
17:29:38HClno, it doesn't.
17:29:45HCli'm getting to that x.x;
17:29:53HCllike i said, its a crude editor at the moment, heh.
17:30:43*Moof idly ponders
17:30:55rasherother than that, it seems to work
17:31:04rasherbut what happens when you "run" such a file?
17:31:13HClthe search engine executes the search query
17:31:13rasherit said "hits: 281" for my query
17:31:21HCland writes it to /search.m3u
17:31:21rasherare those written anywhere?
17:31:27Moofis there any particular reason I couldn't just put in another hard drive into my iRiver?
17:31:35HClMoof: they're hard to get.
17:31:41rasherMoof: not if you can find one that fits
17:32:39rasherHCl: ah, wasn't there when it exited
17:32:46rasherdidn't update dir contents
17:32:47preglowMoof: another? like two disks in one player?
17:32:52HClit should
17:33:02rashergoing into another dir and back, it is now there
17:33:28HCladd that to the tofix list
17:33:34HClin the thingstodo wiki page
17:33:45HClupdate dir contents when returning from plugin
17:34:05rasheryes.. there was another place where that was needed.. can't remember what it was
17:34:52Moofpreglow: no, just a larger drive
17:35:11preglowMoof: no problem, you can fit a 30 gig in a h120 and a 60 gig in the h140
17:35:27preglowthey're even pretty easy to get
17:35:52Moofpreglow: got any suppliers to hand?
17:36:38Adityaman I want to work with u people.. damn my low level programming n00bness
17:36:57rasherYou don't need low level experience really
17:37:02preglowyou'll have plenty of chance when the h3x0 bootloader is done
17:37:07rashermost of rockbox is straight forward c
17:37:16Adityawell then
17:37:25*Aditya sits and waits till bootloader is done
17:37:34rashercould be a while :)
17:37:39preglowyou can also use the simulators :PP
17:37:46Adityaeh.. got hw and an exam to keep be busy
17:40:11Adityacan you plase explain what these simulators are?
17:40:20Adityacan I actually do stuff for the h300 using the simulators?
17:40:40Adityaor even help with the h1xx stuff.. e
17:45:23t0masyou can help the H1xx with sim
17:45:39t0massetup a build environment... and download the rockbox-devel source from CVS
17:45:53t0masand when running configure, chose the Simulator option
17:48:26HClalmost works
17:48:28HClexcept tha
17:48:35HClnow it also accepts empty files as valid
17:48:41*HCl goes to fix
17:49:03rasheror you could just assume that an empty file is the same as a file containging "<END>"
17:49:27HClyes, that is the case
17:49:33AdityaI will try simulator then
17:50:37 Join amiconn [0] (
17:50:47amiconnhi again
17:52:14Adityado you guys just have the devel source zipped up somewherE?
17:52:30Adityaeh.. eclipse is a pain in the ass
17:52:30HClbut you can download it in zipped form, i think.
17:52:58HClstand by for an databox commit
17:53:06t0masAditya: are you using windows or linux?
17:53:07HClfixed the bug with new files
17:53:10AdityaI can get the latest source.. is that the devel source?
17:53:19amiconnHCl: Pls fix the red builds...
17:53:28HCli'm working on it >.<
17:53:32t0masAditya: yes it's devel
17:53:34Adityatomas: I have both on dual boot.. I will prob use my laptop for this though.. it has slack
17:53:44t0masok, then just install cvs...
17:53:50t0masen read the w
17:53:52HClah, button select undeclared.
17:54:17t0massome way my Xchat copies when clicking on them...
17:54:23Adityalots of naked people??
17:54:33t0mas*and read the wiki*
17:54:36t0maswas my line
17:54:43t0masbut I clicked a link at the same time :)
17:54:58Adityaok be back
17:55:03 Nick Aditya is now known as Aditya|Shower (
17:55:05t0massome weird artist :X
17:57:43HClhm o.o..
17:58:02rasher"<end>" disappears mysteriously
17:58:32rasherI have "<genre> contains rock <end>"
17:58:51rasherwhen I move the cursor to one of the three first, <end> disappears
17:58:53HClyea, thats normal.
17:59:40rasherany plans to merge databox with the viewer?
18:00:09rasherso the viewer opens up the editor, and then there's some way to generate the playlist?
18:00:37HCli need a plugin_exec(filename, parameter)
18:00:40HClbefore that can happen
18:00:56HClto run the searchengine on the generated file
18:01:08HClthey're never actually gonna get merged..
18:01:11rasherwhy not just have it all in one plugin?
18:01:31HCli guess it wouldn't be hard..
18:01:45HClbut i don't want to! at least not right now XD
18:01:50rasherI'm just not seeing the reason not to, but if there is one, do tell
18:02:01HClwell, you can store intermediate search files
18:02:05HClthat you can click to update the playlist for them
18:02:32HCli know my search will often be something like "all songs with rating >= <something> and rating <= <something>"
18:02:57HCland its nice if i can quickly update the playlist without having to re-input the search query.
18:03:07HClafter rating more songs
18:03:21rasherI don't see how that would be any different using "my" approach
18:03:45rasheropen the searchfile, press a button and it runs the query
18:04:39HClyea, i know you're right
18:04:47HCli thought you meant something like without a searchfile.
18:04:52rasheroh, no
18:04:55HClit should be easy to merge them.
18:05:00rasheryeah, that
18:05:01HCljust stick searchengine right after databox
18:05:05rasher's what I thought
18:05:13HCli guess you have a good point.
18:05:17HCli'll get on that tonight.
18:05:30HCl :)
18:05:40rasherwell, I'd stick searchengine in a function, called if some button is pressed
18:05:41HClbut i've been like, fixing computers for 2 days straight.
18:05:44HCland i need a break XD
18:05:52HClrasher: sounds good to me, bbiab, food
18:06:00HClhopefully red is gone now
18:06:09rasherHeh, sure enough, just coming up with ideas
18:06:31HCl :)
18:06:36HClyea, and ideas are good.
18:06:40HClotherwise we can't improve :p
18:08:48 Quit bagawk ("Leaving")
18:09:12rasherI'd also love if it asked for filename of the playlist - defaulting (if v-keyb allows) to search I guess
18:09:58rasherseems it does, at least rename defaults to the current filename
18:10:38t0masyes, it's possible to set the buffer before calling the keyboard function
18:10:45t0masthen it has a default value :)
18:10:51rasherah, that's clever
18:12:15rasherHCl: merging them would also remove the need to run a plugin (which we all know is lots of work :)
18:12:40rasheroh.. except for new searches
18:13:32rasherthat'd not be possible
18:13:40rasherguess my approach needs some more work
18:15:53amiconnThe plugin could be used both as a viewer and a standalone plugin
18:16:15amiconnIf started as a viewer, it would offer to run the query or edit it
18:16:29amiconnIf run standalone, it would allow to create a new query
18:17:16rasherdidn't know that worked
18:17:31rasherthen yes, indeed it would
18:17:40amiconnIt's actually simple
18:17:53amiconnA plugin always gets a parameter
18:17:59rasherit would also cut the two-step creation of a new search+playlist down to one stop
18:18:04amiconnIf run standalone, this parameter is NULL
18:18:06HClyes, thats easy to do.
18:18:11HCli'm not gonna merge them
18:18:26preglowwhy not? pretty logical to have them both in one plug, if you ask me
18:18:37rasheramiconn: but viewers don't show up in the pluginlist do they?
18:18:37HCli wasn't finished
18:18:45rasherHeh, do explain
18:18:49HCli'm not gonna merge them while databox doesn't run on player yet.
18:19:08amiconnrasher: They don't (yet), since until now we don't have plugins for which this makes sense
18:19:26rasherAh, add a new parameter in viewers.conf?
18:20:10rasherHCl: alright, that makes sense
18:20:32*HCl was sort of hoping amiconn could look at it for the player :X
18:20:39rasherwhy doesn't it, by the way?
18:20:44HCli have no idea what the player even looks like
18:20:49amiconnHCl: Please don't change the button #defines!
18:20:51HClbecause the interface isn't designed for it o.o
18:20:57HClamiconn: i was just adding :X it made sense to me.
18:21:05 Quit ]RowaN[ ()
18:21:06amiconnIt doesn't
18:21:10HClwhy not x.x
18:21:26HClnow you can do BUTTON_PLAY and actually get the play button on iriver
18:21:27amiconnThere should in no case be two defines with the same value
18:21:29t0masdamn... I want gapless </groupie-mode>
18:21:39HClfine :/
18:21:47amiconnThis will most certain screw the button handling in the sims at least
18:21:56t0mashm... that's something I wonderd about... why isn't the PLAY button the iriver play button?
18:22:13rasherthere's no dedicated iriver play button
18:22:16rasherI guess
18:22:58preglowBUTTON_PLAY is joystick pushed?
18:23:05t0mashuh? I ment why is BUTTON_PLAY linked to the "select" button (joystick)
18:23:09t0masyes preglow
18:23:11preglowthe on button is what has the actualy play symbol
18:23:36HClon and play are the same button.
18:23:44HClit makes no sense to me to not do #define BUTTON_PLAY BUTTON_ON
18:23:50HClbut ok
18:23:53HCli took my define out
18:23:54t0masand I know it now... but when it asks to press PLAY to reboot.. I can imagine some people pressing the actual play butoon...
18:23:55HCland moved it to databox
18:24:09HClah, so thats why rolo wasn't working
18:24:11preglowyeah, i actually consider that quite confusing
18:24:15rashert0mas: that's because it's using an old string
18:24:26t0masamiconn? what's your problem with the buttons?
18:24:29rasherthere's only one language-string for that dialog
18:24:54t0masI did it as viewer does... #define LP_STOP = some_key and on player some_other_key
18:25:17t0masrasher: yes, but switching the buttons would solve that too?
18:25:22rashersame thing with delete
18:25:35rasherbut that's not the plan
18:25:42rasherit's the right button, text is wrong
18:26:22t0mashm... maybe the keypad thing should be done different...
18:27:15t0masI don't know to what atm... but I guess someone can come up with something smart ;)
18:27:33rasherseems sane enough to me
18:28:10amiconnt0mas: Yes, defining local macros depending on CONFIG_KEYPAD is the way that almost all places use to deal with the varying button layout
18:28:23t0masok, what did HCl do then?
18:28:24rashereach function just has to keep in mind that the devices don't use the same buttons
18:28:29t0masredefining the original?
18:28:39t0masah, ok
18:28:39rasherhe changed it now to do basically what you did
18:28:53rasherexcept in less obvious way
18:28:59amiconnThis method (along with the standard way to distinguish short/ long press) was invented after [IDC]Dragon and me started the Ondio port
18:29:19amiconnBefore we had that, there were serious compilation problems due to double #defines
18:29:30|it|ChasKi`Guys, what build do i download for the iHP-140?
18:29:59amiconnSo that's why I said there should never be double #defines again...
18:30:08***Saving seen data "./dancer.seen"
18:30:09t0mas|it|ChasKi`: what do you want with it?
18:30:40|it|ChasKi`t0mas: Ehhmm, i'm a n00b to all this, just want to put Rockbox on meh iHP-140 hehe
18:30:46amiconnbecause, if you use e.g #define BUTTON_PLAY BUTTON_SELECT , and then use both in a switch, you'll get a compilation error...
18:30:57t0mas|it|ChasKi`: have you read the iriver for n00bs section in wiki?
18:31:47amiconnImho the button assignments for the iriver are indeed a bit misleading, because they're based on the recorder keypad... but only half-way
18:31:59 Join TCK- [0] (
18:32:27 Quit MoosCamaro (Read error: 104 (Connection reset by peer))
18:32:33t0masamiconn: it's not a problem now... for us... but when a lot of people start using rockbox... and don't know the recorder, or any other rockbox background, it can cause some trouble
18:32:35 Join MoosCamaro [0] (
18:33:09rasherThe only problem is changing the text for these few places where the button is mentioned by name
18:34:38t0masah we will find those... or other people will... and start asking questions
18:34:54rasherRolo, Delete
18:35:35 Nick Aditya|Shower is now known as Aditya (
18:35:44amiconnHCl: Your button handling is... strange....
18:37:07 Join ]RowaN[ [0] (
18:37:16Adityahey rowan
18:38:53amiconnHCl: I could try to adapt the plugin for the player as well, but I don't want to interfere with your changes
18:39:38amiconn...Ondio as well
18:40:14rasherI don't understand why it shouldn't work
18:40:48amiconnSome changes required in the button handling a differnt print function for player (charcells)
18:41:27amiconnBtw, the plugin also doesn't handle USB
18:44:18HClgo ahead..
18:44:32HCli'm done with making changes, for now.
18:44:40HCljust don't change the main loop too much o.o.
18:44:49HClit still has some operations that need to be added
18:44:53HCllike insert, and delete
18:45:03HCland proper implementation of the 3 editing modes
18:45:21*HCl in the meanwhile fixes his dads pc which is having an illegal (!) windows xp pro key
18:45:22amiconnI'll change the button handling to use a switch(), and proper #defines per model for the various button functions
18:45:30HCland its refusing to upgrade to win xp sp2 cause of that
18:45:43HClsounds okay
18:45:43amiconnChange the key...
18:45:49HCli am.
18:45:58HClgot one.
18:45:59t0masinstall linux? :P
18:46:03amiconnThen I'll try to add a player print function
18:46:16amiconnOf course this can't use invert...
18:46:31amiconnThinking about using brackets or similar
18:50:10 Quit TCK (Read error: 110 (Connection timed out))
18:50:17preglowthe wavpack guy has submitted a patch
18:50:34preglowi wonder i might not just as well upgrade to a newer wavpack
18:52:16preglowahh, there is no new tiny decoder
18:53:55t0masso nothing to fix for rockbox?
18:54:51preglowthis is a wv2wav fix, it seems
18:55:59preglowi'll just commit this patch
18:57:09*lodesi is back (gone 01:30:18)
18:57:52rasherthis is the guy who doesn't use irc or im, right?
18:58:43Adityahey guys.. just a question.. is the simulator a prog that will actually show what the UI will look like on the h1xx?
18:58:57rasherit shows the screen
18:59:00 Join austriancoder [0] (~austrianc@
18:59:10rasherexactly like on the device
18:59:20 Quit lodesi ("Leaving")
18:59:26rasherI think the windows sims show an image of the device as well?
18:59:32rasherevening austriancoder
19:00:25 Join Chamois [0] (
19:01:00preglowrasher: correct
19:01:12preglowso, how do i go about closing the patch or whatever?
19:01:37rasherI think you need to be added as a patch-admin or something like that
19:01:50 Join Stryke` [0] (~Chairman8@resnet-241-86.resnet.UMBC.EDU)
19:02:07*preglow wipes the dust off of his sf login
19:02:43*rasher prods bagder
19:03:30 Join StrathAFK [0] (
19:10:45BagderI'm not here ;-)
19:10:51BagderI'll be back around midnight
19:11:17*preglow gets a surreal feeling
19:11:29rashernothing important, just that the wavpack patch needed to be closed, or preglow made a patchadmin or whatever
19:11:35rasherpreglow: ?
19:12:07preglowdid anything ever happen with the proportional font support?
19:12:13rasherin the viewer?
19:12:15preglowi see pillos attempt is in the patch tracker
19:12:26rasherI've had it applied to my viewer for a while
19:12:30rasherseems to work just fine
19:12:42 Join ardy [0] (
19:12:48preglowthen what's keeping it away from the main tree? player support?
19:12:55rasherno idea
19:13:02rasherI don't think amiconn has tested it
19:13:09rasheror anyone else, for that matter
19:13:14preglowi'll test it now
19:13:49preglowdo you need a special font, or?
19:14:02rasherI tried, but realised I didn't have anything in my archos dir for the sim, got frustrated as I couldn't remember the keys for closing it, then got distracted
19:14:08rasherno, it Just Works
19:17:49Adityayou mean It Just Works (tm) :P
19:18:07 Quit Aditya ("Download Gaim:")
19:18:07rasherYes, sorry
19:18:25preglowcan't say i notice much of a difference?
19:18:38 Part ardy ("Leaving")
19:18:45 Join Aditya [0] (
19:18:59Aditya*sigh* .. closed all of gaim instead of jus tthe buddy list..
19:19:03 Quit Strath (Read error: 110 (Connection timed out))
19:19:37rasherpreglow: no, but on bitmap displays..
19:20:17rasherwith a proportional font, that is
19:20:24preglowahh, yes
19:20:37preglowmy h120 doesn't have a bitmap display? :P
19:20:46rasherI thought you were trying on the playersim
19:20:58preglowi never use the sims
19:21:01rasher(which looks fine, btw)
19:21:20preglowthe viewer with proportional font is Sweet now
19:21:30rashertry the reflow mode
19:21:40rasherit's absolutely beautiful
19:21:48rasherI have some horribly formatted textfiles
19:21:59rasherwith linebreaks all over the place
19:22:02amiconnI know what causes the strange chars in the filename input in databox...
19:22:05rasherwhich end up being nice
19:22:11preglowrasher: how do i reflow?
19:22:18preglowfound it :P
19:22:23preglowthis is nice
19:22:41rasherthe indenting is a bit much if you ask me
19:22:52preglowit's way too much
19:22:58preglowbut that's very easily fixable
19:23:10rasherI changed a 9 to a 5 or a 5 to a 9, somewhere
19:23:20preglow9 to a 5, i'd guess
19:23:30preglowthink i'd just use 4
19:23:36rasherno, it was some mad calculation
19:23:46rasherI'm pretty sure it was 5 to 9, actually
19:25:08rasherokay, maybe not
19:26:22Adityayou guys suck
19:26:26Adityanow I realllly want to see :P
19:26:30Adityadamn hw
19:26:38rasherset up a sim
19:27:17preglowthis patch is permanent in my hacked rockbox tree from now on
19:27:27rasherit should be in the main tree
19:27:41rasherneed someone to test it on a real player I guess
19:28:05rasheror maybe not
19:28:17preglowwell, yes
19:28:20rasherI'm guessing it doesn't touch much non-bitmap related
19:28:28preglowthis thing suddenly become a viable choice for reading shit for me now
19:28:31rasherrecorder is probably more needing for fix
19:29:07preglowthere's just one proportional font?
19:29:13rasherthere are a bunch
19:29:39rasherI actually think most are
19:30:04rasherexcept the YxZ ones'
19:30:16preglowthe standard one was pretty decent
19:30:55rasherokay, not a lot of proportional fonts it seems :)
19:32:56rashera bit on the small size imho
19:32:58preglowdoes someone have to add me to the project?
19:33:00preglowor can i do that myself?
19:33:09preglowim talking about sf, yes
19:33:14rasherI don't think you can
19:33:25preglowi like it small
19:33:31preglowfits a lot onto the screen
19:33:51rasherthat it does
19:36:27Adityahey would it hurt if I install all the packages for cygwin?
19:37:27rasherprobably not, just take a while
19:42:35Adityahmm k
19:42:38t0mas[19:22:11] <amiconn> I know what causes the strange chars in the filename input in databox... <−− the buffer wasn't set to '\0' or something?
19:42:48amiconnt0mas: exactly
19:42:55t0masyes, I had that too
19:43:06amiconnI added a memset(filename, 0, sizeof filename) locally
19:43:15amiconnCurrently fiddling with the button handling
19:43:19*t0mas just did buffer[0] = '\0'
19:43:33amiconnIt _should_ work now for all models
19:43:40austriancoderdebugf calls.. where can i see the output?
19:43:45amiconn...only I didn't add printing for charcells yet
19:43:46t0masin the simulator
19:43:53t0masaustriancoder: not on device..
19:44:18amiconnaustriancoder: Either in the simulator, or when you build a debug version (and have a gdb stub and hookup to use it)
19:44:25austriancodernot good
19:44:25t0masfor on device debugging you must write to a .txt file somewhere... and afterwards read it with the viewer :)
19:44:38t0masamiconn: there is no gdb stub yet afaik?
19:44:41austriancoder.txt file is what i need
19:44:47amiconnnot for iriver
19:44:59amiconnOnly for archos recorders and player
19:45:04amiconn(not possible on Ondio)
19:45:07t0masaustriancoder: take a look at HCl's dynarec core
19:45:23preglowamiconn: did you try the new viewer thing?
19:45:25t0masyou can just copy/paste his debug to file code :)
19:45:26amiconnaustriancoder: I usually use a lot of splash()es for debugging
19:45:48t0masthat's a way too...
19:46:15*t0mas just redefined the DEBUGF() to write to a file...
19:46:29austriancodersplashes seems to be the way
19:46:38austriancoderi need it for dirver debugging
19:46:50t0masah, then it isn't a problem to fuck up your scree
19:47:03amiconnaustriancoder: That's what I did for MMC driver debugging
19:48:36austriancoderwhere is splash defined?
19:49:05Adityaaustriancoder? ... I keeel you!
19:50:13Adityayou don't get it? :P
19:50:18AdityaI dont know.. I had to say that..
19:50:43austriancoderfound it ;)
19:50:48austriancoderi am a lazy body
19:50:50Aditya(arnold schwarzenegger)
19:51:00 Join rob- [0] (
19:51:21Adityais there a todo list that actually has somehting thats not being worked on?
19:51:35Adityaall the ones on .. are already taken
19:51:59amiconnHCl: Is there a reason why left/down increase the selections, while right/up decrease them? Imho the other way round would make more sense...
19:52:43Adityahe was holding it upside down?
19:52:54amiconnhaha ;)
19:53:06Adityahmm.. still dling..
19:53:11amiconnThe flipped display mode already cares of flipping the direction buttons
19:53:25*t0mas thinks he has a partner in crime for late night bad jokes :P
19:54:10 Quit thegeek (Read error: 131 (Connection reset by peer))
19:54:28t0masyou're from the US right? :P
19:54:31Adityacygwin stopped instaling and went back to start of install
19:54:37Adityayeah I am from the US
19:54:48t0masok, that's the right wakeup time :)
19:54:49Adityaand yes.. I am def ur partner if it is about bad jokes :P
19:55:45Adityaoh screw this.. I will just use linux
19:57:16preglowdo so
19:57:20preglowit's ten times nicer
19:57:31*t0mas still uses a mix
19:57:38rasherbuilding takes ages on cygwin, I hear
19:57:45t0masyes it does
19:58:04t0masfull iriver build takes about 6 to 8 minutes... on Athlon XP 3200
19:58:33rashertakes about 2 minutes on this xp2400
19:58:47preglowfull iriver build is done in about two minutes on duron 1.3ghz
19:59:00t0masyes, my laptop does it in 2 to 3 minutes...
19:59:04t0masceleron 800
19:59:26t0masnever tried the athlon in linux... don't even know if I have a crosscompiler there :)
20:00:37preglow1 minute and 2 seconds on 1.3 ghz duron
20:01:10*t0mas gets tempted to reboot and try in linux :P
20:01:12preglowso if cygwin uses SIX minutes for that, it's slow indeed
20:01:25t0masyes it takes 6 minutes
20:01:50*t0mas can get a drink... search for some chips... read 3 emails and be bored for a while when it's compiling in windows
20:02:05rasherwhat the..
20:02:08rasherreal 1m33.804s
20:02:12*rasher punches things
20:02:32rashermandb running at 99,5% cpu might have something to do with that
20:02:34*preglow strokes gentoo
20:03:19rashercome to think of it, don't think I compiled the crosscompiler for this processor
20:03:43*t0mas times a make in cygwin
20:03:46preglowme neither
20:03:50t0masand starts writing his article again
20:05:01rasherreal 0m54.212s
20:05:04rasherthat's more like it
20:05:13*rasher curses at mandb
20:05:27*t0mas doesn't have mandb...
20:05:46t0masand the updatedb of locate is only done at night
20:06:27rasherI'm not even sure what mandb is doing
20:06:34rasherbut it seems to have been at it for a while
20:08:05 Nick Sucka`away is now known as Sucka (
20:09:08austriancoderi need someone with a rockbox player with an fm tuner - no iriver people
20:09:22t0masmy compile is still running :X
20:09:35t0masits at CC gray_fillrect.c now
20:10:02t0mas(or doesn't his recorder have fm?)
20:12:54t0maspreglow, rasher: real 9m5.774s
20:13:04t0masit's even longer than I thought it was...
20:13:09rasheron a computer faster than mine
20:13:25t0masjup... not really a slow pc... XP 3200+ 1024 mb ram...
20:14:11austriancodertomas: is this the compile time for the river rockbox?
20:14:13t0masimagine my celeron 800 laptop running windows+cygwin :X
20:14:20t0masaustriancoder: yes, full compile in cygwin
20:14:50austriancoderah.. my linux is much faster ;)
20:14:58t0masyes, mine too :)
20:15:04rashercygwin is a pig
20:15:22austriancoderreal 0m38.207s
20:15:33preglowt0mas: dAMN, man
20:15:35t0masit's about 2 minutes on my celeron...
20:15:50t0masjust 800 mhz... and iirc 256 mb mem... maybe 384
20:16:14*austriancoder has a pentium M 1.70GHz
20:16:21t0masrebooting to linux, compiling and rebooting to windows might even be faster than compiling in cygwin here
20:16:44rasheraustriancoder: now that's interesting
20:16:52t0mas"real 0m11.468s"
20:16:57t0mascygwin can be fast...
20:17:07t0maswhen there's nothing to do...
20:17:20austriancodernobody here with a fm tuner working with rockbox?!
20:17:25rasherreal 0m0.855s
20:17:47rasherI think it's mostly iriver people these days
20:17:54t0masrasher: that's make time when nothing has to be done???
20:18:07t0mascygwin really is slow...
20:18:17austriancoderrasher: ok..
20:18:45t0masaustriancoder: you've got it working for iriver? but not sure if it still works for the archos chip?
20:20:04austriancodertomas: not yet.. i am working on the audio output of the fm part..
20:21:19austriancodershould be that hard ;)
20:24:37 Join amiconn_ [0] (
20:25:37 Quit amiconn (Read error: 148 (No route to host))
20:25:37 Nick amiconn_ is now known as amiconn (
20:27:17amiconnaustriancoder: What's the problem with fm?
20:27:42amiconn(and no, I don't have an fm recorder, but my sis has an Ondio fm)
20:28:28rasherthere should be a list of which developers/peoplewillingtotest owns which devices
20:28:44amiconnStart 6ne ;
20:29:11rasherI guess :)
20:29:19*t0mas smiles gently at rasher "You can stat a wiki page."
20:29:27amiconnThis list should also contain which sim platforms this dev could test on
20:29:28t0masbut it might be a lot of work ;)
20:29:47*rasher does indeed start a wikipage
20:30:04rasheramiconn: can't all people test all platforms' sims?
20:30:09***Saving seen data "./dancer.seen"
20:30:20amiconnI mean windows vs. linux
20:30:31amiconnSome can test both though
20:30:45rasherI'll add that
20:31:01amiconnHCl: r u there?
20:31:01rasherany ideas for names?
20:31:05t0masjust let everybody add him (don't have to add her?) self?
20:31:27t0masDeviceTestingList or just TestingList
20:31:28rasheryes, I'm certainly not going to add everyone :)
20:31:31t0masor PlatformTesting
20:31:33AdityaI already have my wiki page all setup
20:31:45rasherwith WikiWikiCaps, that is
20:31:46t0masthat snouds ok
20:31:57*t0mas doesn't like the caps...
20:32:39Aditya −− wheeeeee
20:33:53*t0mas doesn't say anthing about his page... nor disclose his lastname :P
20:35:29t0mas"I like Windows in general because programs just work. Macs are ok, but OSX annoys me to no end. Linux is secure and stable, but I will probably never use it as a desktop." <−− You know that remark you always get with the "Linux doesn't work" comment? :P
20:35:55t0mas"It's just better to use things that work than jump around excited because a program actually compiles."
20:36:39t0mashosted on: Initiating server query ...
20:36:39t0masLooking up IP address for domain:
20:36:39t0masThe IP address for the domain is:
20:36:39DBUGEnqueued KICK t0mas
20:36:39t0masConnecting to the server on standard HTTP port: 80
20:36:39t0mas[Connected] Requesting the server's default page.
20:36:40***Alert Mode level 1
20:36:40t0masThe server returned the following response headers:
20:36:42t0masHTTP/1.1 200 OK
20:36:44t0masDate: Sun, 08 May 2005 18:36:34 GMT
20:36:46t0masServer: Apache/1.3.33 (Unix) mod_auth_passthrough/1.8 mod_log_bytes/1.2 mod_bwlimited/1.4 PHP/4.3.9 FrontPage/ mod_ssl/2.8.22 OpenSSL/0.9.7a
20:36:49t0masX-Powered-By: PHP/4.3.9
20:36:51t0masConnection: close
20:36:53t0masTransfer-Encoding: chunked
20:36:55t0masContent-Type: text/html
20:36:57t0masQuery complete.
20:37:01t0maswoooops, copied a little to much
20:37:13t0mas"Apache/1.3.33 (Unix)" was what I ment to copy :)
20:39:57Adityaur point?
20:40:02Adityayou mean its on unix?
20:40:10Adityaof course
20:40:15Adityaa server != desktop
20:40:19 Join TCK [0] (
20:40:37 Join amiconn_ [0] (
20:40:41t0masand I don't like the "Windows is ok because it's simple" point
20:40:49 Quit amiconn (Nick collision from services.)
20:40:49 Nick amiconn_ is now known as amiconn (
20:40:55Adityawhen did I say that?
20:41:15t0masoh wait, you said it was more usable
20:41:23AdityaI said stuff installs and it works..
20:41:39rasherIt does too here
20:41:41Adityaas apposed me having to search for libraries and dependencies
20:41:46t0masyes, and it does that on linux too ;)
20:41:48rasherin fact, much easier than on windows
20:41:54t0masapt-get install someprogram
20:41:59t0masor emerge on gentoo
20:42:05t0masor urpmi on mandrake
20:42:06rasheror urpmi on mandrake
20:42:08rasherdamn you
20:42:18Adityayeah.. that part is getting better.. especially on slack..
20:42:36Adityabut then you want to use photoshop.. or worse.. flash..
20:42:39rasherwell I think your experience with slack is seriously affecting that idea of yours
20:43:10AdityaI dont know.. for most things, windows works for me.. I dont really see much of a point in switching right now
20:43:10t0masAditya: flash installed with a few clicks here..
20:43:20Adityat0mas.. I mean flash editor
20:43:22Adityanot player
20:43:25t0masah ok
20:43:26AdityaI do web design stuff
20:43:27rasheryou haven't had to hunt down libraries or dependencies for many years on most distros
20:43:29Adityaso I need that
20:43:31t0maswell... I'm running windows now :)
20:43:38t0masbut that's not because I like it..
20:43:50AdityaI am not saying I dont like linux
20:43:56t0massome custom tools for my work are only availible on windows :(
20:44:00AdityaI think its awesome.. just wont use it for a desktop right now
20:44:10Adityaand I think I made that point clear
20:44:25t0masjup, maybe but that on the page...
20:44:28AdityaI just made that page because I am sick of people putting up similar pages (mainly for macs)
20:44:45Adityaand I have turned into some sort of a mac disliking person latelhy
20:44:51t0masmaybe the other way around is better?
20:44:59rasherthe other way round?
20:45:06t0masmake a list of devs... and a collumn for the player.. filled with V or empty
20:45:22Adityaactually that makes more sense
20:45:23t0masas this lists are going to become very very long I guess
20:45:40rasherI'd say you'd usually want to take a player, see who has one of those
20:45:49rasherso it should be a group of people for each device
20:45:56amiconnHCl: YOur token_buf is waaaay too large for the archos plugin space...
20:46:22t0masyeah... but looking up one collumn for people having it isn't much more work...
20:46:29t0masand it's way better readable...
20:46:41***Alert Mode OFF
20:47:12rasherhang on
20:47:13t0masbut... this is usable too :)
20:47:39t0masoh... and maybe add a note: "Please don't forget to release the lock when adding yourself."
20:48:01rasherwell, you can still edit the page
20:49:09rasherwell, release
20:50:09amiconnIt crashes :(
20:50:12t0masI didn't mean you had to release it...
20:57:30amiconnmuch better...
20:58:42rashert0mas: well unreleased pages are annoying
20:58:50austriancoderi have a question
20:59:04t0masshoot it austriancoder
20:59:41 Join thegeek [0] (
20:59:52austriancoderin radio.c functions like audio_init_playback are used.. should i port them to iriver or should i use #ifdef IRIVER_H100
21:00:36 Quit TCK- (Read error: 110 (Connection timed out))
21:01:13amiconnaustriancoder: I'd say #ifdef them out for iriver for now; they're for controlling the audio thread
21:01:23amiconn...which is not yet functional on iriver
21:02:13austriancoderok.. will go this way
21:06:26t0masAditya? that free mp3 player... what have you bought from them?
21:06:37t0masor am I misunderstanding something on their site?
21:11:53Adityayou dont buy anything
21:11:54 Quit TCK (Read error: 104 (Connection reset by peer))
21:12:03Adityayou sign up (free)
21:12:09Adityayou do one of the offers on the offers page
21:12:35Adityathen you get 5 people to do the same (they will use ur referral link so the free iriver people know ur friends signed up under you)
21:13:00Adityaits pretty clear in their FAQs
21:13:00rasheroh, one of theses pyramid scheme things
21:13:08Adityanot a pyramid scheme
21:13:10t0masI read that
21:13:14rashersure it is
21:13:15t0masbut what is that offer thing?
21:13:27Adityaoffers are basically trial offers
21:13:31Adityalike blockbuster or w/e
21:13:38t0masah, and what did you do?
21:13:49rasher"then you get 5 people to do the same (they will use ur referral link so the free iriver people know ur friends signed up under you)" == pyramid
21:14:06Adityaa pyramid scheme works differently
21:14:11thegeekno ;)
21:14:20thegeekthere might be twists
21:14:24thegeekbut it's still a pyramid
21:14:35Adityawell .. I don't need to prove anything.. I got my free iriver..
21:14:43Adityaif you dont want to try.. ur loss
21:14:55Adityabut as I was saying.. this is like a referral program
21:15:09thegeekit's a pyramid
21:15:10rasherwhich is a fancy word for pyramid scheme
21:15:10Adityayou always have to do exactly 1 offer and have to refer exactly 5 people
21:15:20CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
21:15:20*t0mas has seen this happening before
21:15:27t0masargument leading nowhere...
21:15:33AdityaI dont care if you guys dont want it.. I already have mine
21:15:34thegeek99% of pyramids are scams
21:15:42thegeekthere is always that 1%
21:15:44thegeekbut still
21:15:44t0masexcept for effnet... there it's normally leading to a flood war
21:15:47preglowtake this privately, please
21:15:58Adityacheck their forums
21:16:04Adityaloads of people getting their free stuff
21:16:14Adityathis has also been featured on BBC and such
21:16:15Adityaso yeah
21:16:24t0masbut my question was... w/o starting this again... "What offer did you do Aditya?"
21:16:38AdityaI did the AOL offer.. but I don't think thats there anymore
21:17:01Adityamainly because my cable was down and I needed a replacement for a while.. fitted perfectly :P
21:17:13Adityathen just cancelled after I got credit (after like 2-3 days)
21:17:25AdityaI would read up before signing up for offers
21:17:28Adityalike on forums and such
21:17:39Adityasome offers are better in the sense that they give credit faster.. and are easier to cancel
21:17:57t0masok, but I don't live in the US
21:18:08Adityawhere do you live?
21:18:11t0masand I don't trust these things normally... but that's my personal opinion
21:18:13rasherFor the record, I never said it was a scam, or that people don't get their stuff.
21:18:19t0masthe Netherlands
21:18:23 Join TCK [0] (
21:18:28Adityayeah its canada + usa only
21:18:31t0masyes, the weed country
21:18:34t0masto have said that :P
21:18:42Aditya −− for those who don't think its legit
21:18:52Adityarasher.. I was mainly arguing thegeek.. but ok :P
21:19:04thegeekI know a lot of 4free stuff is for real
21:19:09thegeekbut there are even more scams
21:19:26preglow75% of the internet is a scam
21:20:18AdityaI was skeptical
21:20:20t0maspreglow: "free porn"
21:20:24Adityabut a bunch of friends got it
21:20:26Adityaso I tried it
21:20:28Adityagot mine =P
21:20:51Adityanow I need to get one last referral for cameras4free and I will have 300 bucks
21:21:26Adityathx tomas..
21:21:33Adityau just took up 20 minutes
21:21:41t0massorry :)
21:21:49Adityanow I have half hour to finish this stupid online quiz
21:21:57 Nick Aditya is now known as Aditya|BeBackQui (
21:22:30t0mas21:35 - 06:34 = 15:01
21:22:39t0masit's not that bad :P
21:26:38HClohno, its me.
21:26:44HClwhat'd i miss
21:28:06HClamiconn: the reason for the selection down/up is because the array goes from the top down eg, item1 is stored in 0, item2 in 1 etc o.o
21:33:09amiconnI already have databox sort of working on player
21:33:13*HCl goes to shower and clean and other stuff
21:33:14HClnice :3
21:33:18amiconnOnly one thing is needed
21:33:18HCli'll look at it when i get back
21:33:40amiconnI have to make the printing keep the marked item on-screen
21:33:55amiconn..and it seems I found a bug in the player keypad
21:34:11HClwhat bug
21:34:16amiconnIt seems to not disable the blinking cursor
21:34:33amiconnLooking somewhat funny...
21:36:07amiconnBtw, this 'keep marked item visible' handling is also needed for the other devices, if you really want to use up the allowed token count...
21:36:48amiconnI'll leave that out for now...
21:37:49 Join gromit` [0] (~gromit`
21:40:15 Join amiconn_ [0] (
21:40:27 Quit amiconn (Nick collision from services.)
21:40:27 Nick amiconn_ is now known as amiconn (
21:40:33amiconnmrf, wlan...
21:42:52*t0mas wlan + vnc's only... then IRC isn't dropped when I walk out of reach
21:43:13*rasher runs most things inside screen
21:44:53t0masscreen rocks on servers yes... compiling a new kernel (or firefox) in a screen session is useful..
21:45:00t0masas you can reconnect later to see if it has finished
21:49:07 Join amiconn_ [0] (
21:49:13 Quit amiconn (Nick collision from services.)
21:49:14 Nick amiconn_ is now known as amiconn (
21:52:49*HCl wlans with putty
21:52:58HClwith screen :)
21:54:01 Nick austriancoder is now known as ac_away (~austrianc@
21:57:38 Join amiconn_ [0] (
21:58:07 Quit amiconn (Nick collision from services.)
21:58:08 Nick amiconn_ is now known as amiconn (
21:58:12amiconnhmpf :((
22:00:30 Part amiconn
22:02:31 Quit lostlogic (
22:02:31 Quit Ka (
22:02:42NJoinlostlogic [0] (
22:02:42NJoinKa [0] (
22:03:15*HCl ish back
22:07:34*HCl doesn't usually like bash quotes, but this one is okish :p
22:17:13t0masweird.. is _very_ slow here
22:20:04t0masthat one is funney yeah
22:21:17HClhmm you're right
22:21:22HClbash is suddenly slow for me too
22:22:25t0mashm.. some lamo kiddie flooding it or something
22:24:30t0mas20 packets transmitted, 6 packets received, 70% packet loss
22:28:22Aditya|BeBackQui −− lol
22:28:35Aditya|BeBackQuiand damn HCl.. got me reading bash lol
22:28:38 Nick Aditya|BeBackQui is now known as Aditya (
22:29:02t0masthat one is good yes...
22:30:05CoCoLUSi hate such sites
22:30:10***Saving seen data "./dancer.seen"
22:30:10CoCoLUSeverytime you start reading, you can't stop
22:30:16CoCoLUSanother one, another one.. :)
22:30:23HCllol o.o
22:30:57Adityammm.. I love this one quote..
22:31:09Adityathats my all time favorite..
22:33:04HClmmm, what have i started, heh.
22:33:34Aditya97/100.. good enough
22:33:56 Join Tang [0] (
22:37:12Adityaoff to the gym
22:37:18 Nick Aditya is now known as Aditya|GYM (
22:38:19t0masthe only funny quote on activenet :P
22:38:38t0masand the only international one
22:45:29 Join amiconn [0] (
22:46:12amiconnBack on wired lan :)
22:48:22amiconnDid you get my remark concerning 'keep active item visible'?
22:48:50HClyou mean when someone inserts an insane search query
22:49:27amiconnDepends on the display size... and later on the selection of fonts, perhaps
22:50:08amiconnYour original code allocated space for 200 tokens. Though I had to reduce that to 70 to make it fit on archos, this is still a lot...
22:50:21HClshould be plenty.
22:50:44amiconnOn player, the 'keep visible problem already strikes at the 3rd..4th token
22:51:01HClhow did you solve it?
22:51:31amiconnNot yet
22:51:38amiconnAnd for some reason the lcd_clear_display() doesn't seem to work right here too... strange...
22:52:07HClit does when you put an lcd_update straight after it..
22:52:20amiconnThere is not lcd_update() on the player...
22:52:38HClneeds some looking at
22:52:42HClmaybe something for the tofix list.
22:53:53amiconnSuch things should be fixed asap, imho. Putting them on an arbitrary list doesn't help anything in most cases
22:54:13HClwell okay
22:54:17HClbut i don't have time at the moment.
22:54:25amiconnThose lists have a tendency to continuallly grow...
22:58:00 Join matsl [0] (
23:00:54HCljust saying that you should either try to fix it
23:00:57HClor add it to the list
23:01:39 Quit Chamois (" Want to be different? HydraIRC -> <-")
23:02:06preglownot many other choices, are there? :P
23:02:10preglowapart from ignoring it, of course
23:02:23HClyes, i'm just saying
23:02:26HClwe mustn't forget
23:02:29HCland i'm not gonna do it.
23:02:39 Quit Tang ("Chatzilla 0.9.68a [Firefox 1.0.3/20050414]")
23:09:10 Join arfo [0] (
23:14:47amiconnah... it's a different problem. The lcd_clear_display() is working correctly, only the printing should take care not to print to invisible lines
23:14:59HClwhat? o.o;
23:15:00amiconnThis is causing trouble...
23:15:32amiconnI had that on the player...
23:19:09HClit doesn't clear on iriver either though, and it can't print to invisible lines on iriver o.o
23:23:50amiconnIt should clear with current cvs; it does work on player, recorder, and Ondio now.
23:24:12*HCl will check when he has some time
23:24:15HClbusy busy days :/
23:27:57 Quit Harpy (Read error: 145 (Connection timed out))
23:28:06 Quit matsl (Remote closed the connection)
23:30:19preglowi wish all codec coders were as considerate as the wavpack guy, heh
23:30:29preglowjust static data, very small code, no malloc
23:30:39preglowthe entire wavpack codec fits in iram
23:32:07amiconnHCl: I fixed your error message for the filename input.
23:32:18preglow265% realtime
23:32:21preglowwith no opts
23:32:32amiconnThe only reason why kbd_input() returns nonzero is when the input is cancelled
23:32:36HCltook my emoticon out, didn't you ;/
23:34:04amiconnBtw, there is no way to cancel an edit (yet)
23:34:15HClthere was o.o.
23:34:18HCldid you break it?
23:34:40HClyes, pushing off while in currentindex browsing mode
23:34:55HClso pushing off repeatedly will abort it
23:35:45HClit will still save it if its correct though.
23:35:48HClmight want to change that.
23:36:16amiconnHmz :/ For me, off saves the current editing.
23:36:32amiconnYes, I mean _not_ saving on exit
23:36:36HClthe abort saves it if the query is correct XD
23:36:39HClyou might want to like
23:36:48amiconnThen it's no abort...
23:36:51HClah meh hold on i'll fix it
23:37:38amiconnIt's not that important for now, but imho you should consider it when completing the editor with delete token etc
23:38:35HClthe code is already there, pretty much.....
23:38:51HClthe code used in the old tokentool just needs to be implemented into the new interface
23:38:54HClit should be fairly trivial
23:39:03t0masok, next good bash quote... "Saying that Java is nice because it works on all OS's is like saying that anal sex is nice because it works on all genders"
23:39:43amiconn"... because it works on all OS's" is an invalid statement...
23:39:47 Join ashridah [0] (
23:43:03*Bagder pops in
23:43:30amiconnhi Bagder
23:43:52amiconnI found some quirks with the new build system :(
23:44:22amiconn(1) 'make clean' tries to delete some strange dirs:
23:44:40amiconnrm -f obj/unix/release/*.o
23:44:40amiconnrm -f obj/unix/debug/*.o
23:44:40amiconnrm -f examples/*.o
23:45:02HClthats from rockboy.
23:45:11HCli don't know why that got back in though?
23:45:15 Quit lostlogic ("Going to the moon")
23:45:17Bagderthat hcl's fault
23:45:20HClnuhuh O.o.
23:45:24HCl :p
23:46:08amiconn(2) Dependency problem: rockbox.elf and descendants don't seem to depend on librockbox.a, so when I e.g. only change a driver, the binary doesn't get rebuilt
23:46:14Bagderthat should learn Hcl to cvs update, and not copy files over existing ones...
23:46:26HCleh? O.o.
23:46:37HCli haven't touched the rockboy makefile afaik? :X
23:46:39Bagderyou didn't cvs update did you?
23:46:47Bagderyour commit showed it
23:46:51amiconn(3) The strange dependency problem with the simulators are still there...
23:47:08HClwell i had a commit at first, but it failed because i indeed hadn't updated
23:47:14BagderI didn't modify the (3)
23:47:14HClso then i did and recommitted
23:47:39HClBagder: which commit? when? o.o
23:48:02Bagderno, I'm wrong
23:48:43amiconnBagder: Didn't you say rockboy?
23:48:55BagderI said I'm wrong
23:49:06HClor is it from libdumb?
23:49:07amiconnumm, sorry
23:49:15preglowBagder: i need to close a patch
23:49:41HCli think that might be from libdumb o.o just remembered that though rockboy had an obj/unix, it did not have an examples or release or debug
23:49:51Bagderpreglow: your sf user name?
23:50:15*amiconn tries a grep...
23:50:23preglowBagder: preglow :)
23:50:36Bagderok, hang on
23:50:41preglowcan do
23:51:38 Join lostlogic [0] (
23:51:59Bagderpreglow: it should work for you now
23:52:08*preglow goes for a test drive
23:52:39 Quit lostlogic (Client Quit)
23:54:50 Join lostlogic [0] (
23:54:50 Quit asdsd____ (Read error: 104 (Connection reset by peer))
23:55:10 Join asdsd____ [0] (
23:55:33preglow'closed' is the proper status for applied patches, i take it?
23:56:45amiconnpreglow: fyi (had to learn that back then as well): the proper status for fixed bugs is _not_ closed though
23:56:54amiconn...unlike with the patches
23:57:16Bagdernowadays I close them all
23:57:19Bagderwhen fixed
23:57:25amiconnAh, that's new
23:57:44amiconnI remember bugs did get marked fixed, but left open until the next release
23:57:59Bagderit is just too big a job to close them all ;-)
23:58:10preglowBagder: but the patches table doesn't get updated until tomorrow, no?
23:58:18Bagderpreglow: correct
23:58:27preglowseems i successfully closed the patch without destroying the world
23:58:35amiconnHmm, my grep didn't work :(

Previous day | Next day