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 2007-07-06

00:00:26chuklzsorry to interupt but i'd like to know if there's any way i can get involved
00:01:13stripwaxchuklz - the more the merrier! are you technical, programming/hardware/debugging experience, etc?
00:02:41chuklzwell im pretty young...15....but i just got a gigabeat s and im sick of the MCE firmware. I know some Python, and am interested in hardware, etc. Anything you think I would be able to do I'd be glad to get some experience
00:02:48 Quit Siltaar (Remote closed the connection)
00:03:07dionoeaoh ... Python :( that's not really what's used in rockbox :)
00:03:23dionoea(but it's an awesome language)
00:03:27chuklzWhat is Rockbox written in? C?
00:03:38dionoeaC and assembler
00:04:06chuklzwow, I'm new to the whole linux, IRC, etc, so don't mind me if I ask dumb questions.
00:04:09amiconnpetur: Your return code changes break the logical order of return codes from this function
00:04:24 Quit davina (Remote closed the connection)
00:04:25 Join saratoga [0] (i=9803c6dd@gateway/web/cgi-irc/
00:05:02peturamiconn: oops, I couln't quite figure out how they were laid out...
00:05:20amiconnThe system is really simple
00:05:34amiconnFirst possible point of failure returns -1, second one -2 etc
00:05:36 Quit chuklz ("Leaving")
00:05:57amiconnIf there's a downlevel return code involved, we use downlevel_return_code * 10 - n
00:06:11amiconn-1 and -2 already exist in the function
00:06:30peturthat's why it started at an offset
00:06:41amiconnIt didn't
00:06:44*petur slaps forehead and hopes it doesn't log in
00:06:45amiconnIt started at -1
00:07:09*amiconn would like to change the *10, btw
00:07:15peturI mean the downlevel return codes
00:07:29peturyes, I'd like that too
00:07:51amiconnIt's nasty for 2 reasons: (1) it limits the number of possible error returns to 9. If a function has more, the return code gets ambiguous
00:08:08dionoeaanyone have power consumption measurements (in mW) of rockbox playing a 160 kb/s mp3 ?
00:08:12amiconn(2) *10 is a non-trivial cpu instruction
00:08:33dionoeaon pp5022
00:08:37bluebrotherso using * 0x10 instead?
00:08:39amiconn(on many targets)
00:09:03amiconnbluebrother: Maybe, but then the problem is how to display them
00:09:16amiconn%x and %lx are unsigned, and return codes are negative
00:09:30 Quit Lear ("ChatZilla [Firefox]")
00:09:40amiconnerror return codes of course
00:11:32amiconnAnd a return code combined by using *0x10 isn't easily recognisable when displayed using %d
00:12:01 Quit perplexity (Read error: 113 (No route to host))
00:12:06stripwaxamiconn - 0x80000000 + 0x10 * downstream code + error code
00:12:15bluebrotherwell, how about extending %x to support signed?
00:12:22stripwaxor even just 0x1000 * downstream code
00:12:37*stripwax meant 0x10000 but whatever
00:12:45bluebrotherotoh, -1 would yield 0xffffffff. Wouldn't that make it clear somehow?
00:12:46amiconnThere can be more than one downstream error code
00:13:07stripwaxamiconn - yep, I'm suggesting multiply by 65536 rather than 10
00:13:20stripwaxand top bit set to make it negative
00:13:27amiconnyep, and that's what my remark was related to
00:13:57amiconnIf you multiply by 0x10000 and have 2 downlevel codes, you'll get an overflow
00:14:31stripwaxamiconn - oh −− I see what you mean. No, I meant 0x10000 * (downstream error codes * 0x10).
00:14:40stripwaxHow many downlevel codes can be there at the same time
00:15:32amiconnIt just depends on the nesting count. Theoretically - as many as the stack allows
00:15:49 Join Xerion [0] (
00:16:02amiconnBut in practice I didn't see more than 3 digits
00:16:29stripwaxoh. well, if you can limit it to 6 then you're probably ok .. 0x8123456c
00:17:06 Quit ender` (" The propeller is just a big fan in front of the plane used to keep the pilot cool. When it stops, you can actually watch the)
00:17:10amiconnAnd you can't use 0x8000000 + x because each level would try that
00:17:42amiconnHmm, you can
00:17:57stripwaxas long as you set top bit every time
00:18:01amiconnMultiplying the downlevel code by 16 will shift out the msb
00:18:22amiconnBut using | 0x80000000 makes it clearer
00:19:00stripwaxbut then last code in stack is 0-7 not 0-15
00:19:21amiconnThat idea I don't like
00:19:48amiconnToo many special measures depending on whether the return value is int or long, and what architecture we're running on
00:20:04amiconn(64 bit sims ....)
00:20:14stripwaxcould just make it int_32 ?
00:20:32stripwaxactually, which fn are we talking about (sorry..)
00:20:34amiconnNot for every function
00:20:45amiconnAny function that returns errors
00:20:53 Quit entheh ("^~")
00:20:56amiconnSome have a return data type defined by posix
00:21:19stripwaxdoesn't posix define the error codes in those cases?
00:21:27amiconn(e.g. open() or read()
00:22:10stripwaxoh, didn't know that.
00:22:40pondlifeMaybe it would be best to only keep the "original" error code and not attempt to combine multiple levels?
00:22:56stripwaxseems easiest
00:23:03pondlifeWould that allow for diagnosis without losing too much info?
00:23:28stripwaxmaybe you could have a small global stack of error codes that get pushed as you come back up the call stack
00:23:36stripwax8 words or so
00:23:44amiconnThe current mechanism shows the call chain
00:24:02stripwaxrather than trying to fit all the possible error codes and call chain into one (potentially) 31 bit value
00:24:03amiconnBut in fact we could use rc * 0x10 - new_rc
00:24:31stripwaxso long as a fn can fail for only one of 16 reasons
00:24:41pondlifeAnd display it negated in hex, i.e. as a positive value?
00:24:46amiconn...and then display it with snprintf(buf, sizeof bu, "-%x", -rc)
00:25:44amiconnAnd the good thing with *0x10 is that it equals <<4, a single instruction on most targets
00:25:48amiconn(2 on SH1)
00:26:11pondlifeWill gcc spot that optimisation?
00:26:26amiconnIt should
00:27:06amiconn*10 is also just 2 insns on arm, but more on the other archs
00:27:06stripwaxwho/what displays the error code? a short stack could be used, and then all error codes on the stack displayed? or are we talking about time sensitive (codec?) errors?
00:27:26stripwax1 on coldfire surely
00:27:41 Join midkay [0] (n=midkay@rockbox/developer/midkay)
00:27:46amiconnHmm, true
00:27:58amiconnAnd the cycle count doesn't really matter for an eror return
00:28:06dionoeaamiconn: how do you use arm-elf-objdump to disassemble the apple os ? (gives me File format not recognized here ... but that must be because i don't know how that works)
00:28:14amiconnThe error code is e.g. displayed in the panic screen
00:28:27amiconnstripwax: Ah, no, it's also at least 2
00:28:30pondlifeBinary size?
00:28:40 Quit Wiwie (Read error: 110 (Connection timed out))
00:28:43amiconnThe multiply instruction can't use an immediate constant
00:28:52stripwaxsurely panic can display a stack of errors
00:29:08oboamiconn: undefined instructions @20204004 with that patch
00:29:31dionoeaah, found it on google. arm-elf-objdump -b binary -marm -D apple_os.bin
00:30:41amiconnstripwax: My idea was about saving binary size, not increasing it
00:31:26stripwaxstacking is only going to be a couple instructions too though
00:31:36stripwax(although not displaying the errors.. true..)
00:31:43amiconn(and also allowing a few more causes for error returns to be distinguished)
00:32:24amiconn9->15 isn't bad. There aren't that many monster functions with lots of different error returns
00:32:57amiconnobo: Does that happen after running for a while, or quite early?
00:33:33stripwaxrc<<4-c sounds good enough then
00:33:42obo40ish minutes this time
00:34:28amiconnReally odd
00:34:51pixelmaon a next buffer fill maybe?
00:34:52amiconnThe G5.5/80 is already playing for ~80 minutes
00:35:26 Join krazykit_ [0] (
00:35:28stripwaxIs there a way to profile rockbox to see how much (real) time is spent by each thread? (and would that even be useful?)
00:35:34amiconnAnd I used all sorts of plugins in the beginning
00:35:44obopixelma: it's crashed both before and after buffer refills
00:35:58stripwaxafaict the existing on-device profiling is per thread only rather than 'global'
00:36:56stripwaxamiconn - quite right (2 instruction mult) - sorry about that
00:37:02 Part pixelma
00:37:54 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
00:38:37 Part TrueJournals
00:39:58 Quit pondlife ("disconnected has pondlife")
00:43:37 Quit krazykit (Nick collision from services.)
00:43:39 Nick krazykit_ is now known as krazykit (
00:44:43DerPapstis the PP chip in 5G iPods the same as in 5.5G (PP5022C)
00:47:27saratogaDerPapst: yes I believe so
00:49:46DerPapstthen it's even more strange that the 5G crashes ways more often than the 5.5G one
00:51:01 Quit Rob2222 (Read error: 104 (Connection reset by peer))
00:54:08 Quit My_Sic (Read error: 110 (Connection timed out))
00:54:41 Join Rob2222 [0] (
00:55:00XavierGrlear: yes, right now the filetype array is full (at least on H100) so if you use a build with test_codec you will get the error
00:56:07amiconnXavierGr: Already checked petur's latest commit?
01:00:44 Quit SirFunk (Success)
01:01:47 Quit Thundercloud (Remote closed the connection)
01:02:38 Nick Mouser_X is now known as Bac (
01:02:45BHSPitLappyanyone have DAP recommendations in the <$100 range?
01:03:07 Nick Bac is now known as Mouser_X (
01:03:38 Join SirFunk [0] (
01:04:13 Nick Mouser_X is now known as Gone_again (
01:04:22 Nick Gone_again is now known as Mouser_X (
01:04:26 Join andrew__2 [0] (
01:04:38peturebay for an iHP120/140 ?
01:05:07DerPapstif you're lucky you might even get a Jiggerbeat F20
01:05:14***Saving seen data "./dancer.seen"
01:05:20 Join crashmatrix [0] (
01:05:36DerPapstor some cool <insert_dap> chinese rip off
01:06:28peturw/o rockbox then...
01:06:38 Quit saratoga ("CGI:IRC")
01:06:53BHSPitLappyyeah, it's not so much a requirement :/
01:06:57BHSPitLappyas it isn't for me
01:10:11 Quit Arathis ("Bye, bye")
01:10:37 Quit andrew__ (Read error: 110 (Connection timed out))
01:11:56DerPapstwho need that bos anyways....
01:15:22 Quit gromit` ("Coyote finally caught me")
01:15:24amiconnobo:"> <= back to 75MHz. Remove the other patch first
01:16:31DerPapstamiconn: do you really think that causes the problem?
01:16:39amiconnI want to rule it out
01:17:35DerPapsthmh... :)
01:19:46XavierGramiconn: gah! I should read the commit log before typing...
01:19:49 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
01:22:09 Quit stripwax ("Miranda IM! Smaller, Faster, Easier.")
01:23:41 Quit spiorf_ (Remote closed the connection)
01:24:39 Quit Entasis_ ("Leaving")
01:27:10obothanks amiconn - I don't have time to test it now, but I've put the build on and will see how it goes in the morning
01:29:27*petur wonders if there are complaints about his last commits before going to bed
01:30:03*DerPapst prepares to start complaining...
01:30:45DerPapstno :)
01:30:54*petur prepares to go to bed faster
01:31:10DerPapstgood night petur ;)
01:31:29 Quit petur ("Zzzzz")
01:32:37 Quit obo ("bye")
01:32:45 Quit jhMikeS (Read error: 110 (Connection timed out))
01:37:54 Join hcs [0] (n=agashlin@rockbox/contributor/hcs)
01:41:28 Quit bluebrother ("leaving")
01:48:12 Quit _pill ("changing servers")
01:49:25 Join pill [0] (
01:59:14 Quit Xerion (" ")
02:03:30 Quit DerPapst ("*plop*")
02:09:09 Quit jgarvey ("Leaving")
02:12:17 Join piroko [0] (
02:13:00pirokoI've been playing a lot of nsf files lately, and I think I've encountered a bug in the plugin
02:13:05 Quit kismet ()
02:13:58pirokoI can't remember the exact error message, but I will tell you in the next thirty minutes when I get it
02:14:49hcsif I drop out say it anyway, I'll check the log
02:15:03pirokoShould I put anything special to search for in it?
02:15:03*amiconn has a suspicion regarding the instabiliy of 5g
02:15:16amiconnWill test tomorrow
02:15:20pirokoLike... 2a03?
02:15:35hcspiroko: nsf would probably be fine
02:16:03pirokohcs: So I assume you haven't encountered problems with it?
02:16:35hcsnothing serious
02:17:10pirokoHuh. The "bug" I found completely locks up my player :-O
02:17:47hcshmm, well, if it turns up regularly I'll try to track it dow
02:18:00pirokohcs: Alright. It happens quite often to me
02:18:14hcswhat player do you have?
02:18:21pirokoiPod 4g grey
02:18:31pirokoI also get occasional pauses in sound output
02:18:37pirokoLike it can't handle it or something
02:18:56pirokoBut that's no big deal
02:19:45 Join secleinteer [0] (n=scl@
02:28:16 Quit piroko ("Brb")
02:28:21 Join valis [0] (
02:28:48 Quit PaulJam (".")
02:29:01 Join piroko [0] (
02:30:37 Join Thundercloud [0] (
02:30:49bdgrauehello, my wife tried to play .wma with rockbox, but the rockbox gives an error and freeze everytime, so i have to reset the rockbox
02:31:16bdgraueis there a way to get back to normal working rockbox without reset?
02:34:39valisanyone have experience with the iaudioX5? I'm curous about rockbox's stability on it.
02:39:10 Join the_conley [0] (n=conley@
02:47:31 Quit piroko ("POOF")
02:49:28 Quit valis ("WeeChat 0.2.5")
02:55:01 Join seclein2 [0] (n=scl@
03:01:00 Quit Thundercloud (Remote closed the connection)
03:01:15 Join Nico_P [0] (
03:05:15***Saving seen data "./dancer.seen"
03:12:41 Quit secleinteer (Success)
03:18:57 Join aliask [0] (
03:20:04 Join saratoga [0] (i=9803c6dd@gateway/web/cgi-irc/
03:20:26saratogaanyone know about that annoying "differences in signedness" warning in GCC?
03:21:25 Quit bdgraue (Remote closed the connection) are comparing an unsigned int with an int
03:22:46the_conleyor something like that...if you are talking about specifics in rockbox, dunno
03:23:20saratogaits some random gcc problem that cropped up
03:23:28saratogai get thousands of those messages now when I build rockbox
03:23:31saratogaand they're all nonsense
03:25:48the_conleyah, dunno...I know I get more signedness warnings with different versions of gcc
03:26:01the_conleyI think 4+ is gives tons of warnings over it
03:26:08saratogathats the odd thing, i didn't update gcc
03:26:20saratogastill using the stock 4.0.3 build i've had for months
03:27:33the_conleyhm...don't know what to tell you...I've never actually built rockbox
03:27:48the_conleyI'm waiting for some more gigabeat s support first
03:28:21saratogayou may be in for a bit of a wait
03:30:40the_conleyyea, I've been in that wait for a while
03:30:56the_conleythe S wiki gets updated every now and then, so I guess they are doing something
03:31:08saratogathey have an IRC channel
03:31:13saratogatheres actually work being done
03:31:20 Part maffe
03:31:50the_conleyyea, I know, I'm hanging out there too
03:32:02saratogaah cool
03:32:15saratogawe need another developer to help ptw419 out
03:32:18the_conleyI just saw your question and answered it...I actually thought I was on the fedora forum
03:32:43the_conleyyea...I'm a programmer...but I'm not sure I have the time to really commit to this
03:32:51saratogawhich question?
03:33:03the_conleyI considered it a while ago, when I won the gigabeat
03:33:07saratogamake time :)
03:33:10the_conleyo, signedness
03:33:22saratogathats how i helped make the wma codec
03:33:29saratogaoh you saw a solution to that?
03:33:37krazykitsaratoga, yeah, but aren't you getting paid by google too?
03:33:53the_conleyI start getting paid soon...I'm so happy
03:33:57saratogakrazykit: yeah that helped too
03:34:07the_conley(I'm a student)
03:34:20saratogathough to be fair, I would have got paid the same if I'd gotten funded from my lab anyway
03:34:25saratogaso its not that big of an incentive
03:34:33krazykiti wish my job would hurry up and pay me. direct deposit may not deposit until the second pay period
03:34:46saratogaok got the new wma decoder improvements working
03:34:49saratogaseems a lot faster
03:35:05saratogagetting 170% realtime on my sansa @ 192k :)
03:35:13the_conleydo you use wma?
03:35:39krazykitwhat does the sansa have? a 100mhz ARM9?
03:35:47 Quit Beta-guy (Read error: 104 (Connection reset by peer))
03:36:07krazykiti need to make an updated build, bah.
03:36:40saratogasame CPU as the ipods
03:36:45saratoga75 MHz ARM7
03:38:05saratogawell 2x ARM7 but we don't use the second one much yet
03:40:28Battousaisansa has 2 arms?
03:40:36saratogayeah all the PP systems do
03:40:37scorche2 cores
03:40:44saratogathey really suck
03:40:49Battousaii assume thats one for playback and one for everything else?
03:40:55saratogathats the idea
03:41:03saratogawe don't really use the second one much though
03:45:08saratogakind of funny how slow the CPU in the ipods really was
03:45:18saratogacompared to what was used in everything else in that price range
03:47:37 Join z35 [0] (
03:49:28 Part the_conley
03:50:39saratogaok wma improvements are committed
03:50:45krazykitaw damn
03:50:52krazykiti JUST updated 5 minutes ago
03:50:57 Part chrisjs169|afk
03:51:16 Join chrisjs169|afk [0] (
03:51:52saratogai said I'd gotten it a lot faster ;)
03:51:59saratogashould have guessed that meant a new build was incoming
03:53:06krazykiteh, i don't use wma anyway :P
03:56:52krazykitnice to know it's there, though
03:57:21saratogaoh then no sense upgrading
03:57:31saratogawe'll be adding a lot of new changes to the wma decoder soon anyway
03:57:45krazykitwell, i was making a new build today anyway.
03:57:49saratogahah, got the wma decoder faster then the aac decoder on my sansa
03:58:40 Quit Nico_P (Remote closed the connection)
04:02:12 Quit JdGordon ("Konversation terminated!")
04:03:32 Join safetydan [0] (i=cbca159f@rockbox/developer/safetydan)
04:09:35 Join TrueJournals [0] (
04:12:38 Join TiMiD[FD] [0] (n=TiMiD[
04:23:31 Join n17ikh|Lappy [0] (i=n17ikh@dial-51.r3.ncbrvr.InfoAve.Net)
04:30:52 Join piroko [0] (
04:31:04 Part TrueJournals
04:32:08pirokohcs: I found an nsf file that rockbox skips really really badly on
04:33:08pirokoWhich is sad because it's a really really good one :)
04:33:39 Quit n17ikh|Lappy ()
04:40:09 Join Segadude [0] (
04:40:46SegadudeYou guys have to help me! My Ipod won't turn on!!:(
04:41:46Segadudehello anyone here?
04:41:53pirokoI am here
04:43:02Segadudethen help me please!!!!!!!!!!
04:43:19SegadudeMy ipod won't turn on!
04:43:34pirokoHave you plugged it in?
04:43:53pirokoHeld down menu and select for at least 8 seconds?
04:44:06piroko(With the hold switch off?) ;)
04:44:33 Join bspray [0] (
04:44:54Segadudei'l try that
04:45:21saratogamore wma improvements
04:45:28SegadudeThank you so much!!!!!!
04:45:36pirokoSegadude: Lol. No problem
04:45:37Segadudethat worked!
04:45:57Segadudesee ya!
04:46:10 Quit Segadude ("Quitting!")
04:49:02 Join kismet [0] (
04:52:39 Part bspray
04:59:50pirokosaratoga: Hmm... wmav2 encoded with ffmpeg isn't working for me in rockbox. It still just sits there doing nothing. What encoder would you recommend I use?
05:00:34TiMiD[FD]oggenc ?
05:01:13pirokoHaha. For testing out wma ;)
05:01:23pirokoI use ogg for my own collection
05:01:43TiMiD[FD]hmm ok
05:01:55TiMiD[FD]I was wondering why one would want to use wma actually
05:02:19pirokoIt supposedly has a better quality than mp3 from what I've read
05:02:45pirokoAnd some people use Windows Media Player to import their cd's, and thus have collections of wma files
05:03:13pirokoAnd it looks better for rockbox to have the support I'm sure :-D
05:03:54 Join n17ikh|Lappy [0] (i=n17ikh@dial-51.r3.ncbrvr.InfoAve.Net)
05:04:05 Quit n17ikh|Lappy (Remote closed the connection)
05:05:16***Saving seen data "./dancer.seen"
05:13:34 Join sarixe [0] (
05:15:33 Quit kismet ()
05:15:49sarixealright, i have a question regarding building rockbox from source. i wasn't able to find m68k-elf-gcc anywhere but from a package called "coldfire", and when i use that one, i get the following error:
05:15:51sarixeAssembler messages:
05:15:51sarixeError: already selected `5206e' processor
05:15:51sarixeError: unrecognized option -m5249
05:15:51DBUGEnqueued KICK sarixe
05:15:51sarixemake[1]: *** [/home/sarixe/Programs/rockbox/source/rockbox-20070705/build/firmware/ata_idle_notify.o] Error 1
05:15:52sarixemake: *** [build] Error 2
05:15:57 Join kismet [0] (
05:19:01sarixeany help?
05:20:18saratogapiroko: thanks for the heads up
05:20:26saratogai will need to test the ffmpeg encoder
05:20:48saratogain the meantime, you should probably use the MS encoder
05:20:58saratogaits quite well behaved compared to the ffmpeg encoder
05:21:03scorchesarixe: use
05:21:16sarixethat'll get me m68k-elf-gcc?
05:21:31sarixesweet, thanks
05:21:50saratogapiroko: what settings did you use to test the ffmpeg encoder anyway?
05:24:35pirokosaratoga: This was the command: ffmpeg -acodec wmav2 -i blah.ogg ~/test.wma
05:25:01 Quit saratoga ("CGI:IRC (EOF)")
05:25:08 Join saratoga [0] (i=9803c6dd@gateway/web/cgi-irc/
05:25:12saratogapiroko: ok so just normal settings
05:25:17saratogawill take a look tomorrow
05:25:20pirokoAnd mplayer recognizes and plays it with this to say: Selected audio codec: [ffwmav2] afm: ffmpeg (DivX audio v2 (FFmpeg))
05:25:30pirokoAlright :)
05:25:34saratogacould be either the codec or linuxstb's parser
05:25:57 Quit sarixe ("Peace")
05:26:20 Quit saratoga (Client Quit)
05:36:15pirokohcs: I got the error again. "Undefined instruction at 00057C4C"
05:38:02hcswith a specific file?
05:38:44pirokoI'm not sure. :(
05:42:40 Quit piroko ("Brb")
05:44:09 Join jhulst [0] (
05:50:37CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
05:50:37*safetydan notes fsincos from the eq has made it in to the wma decoder
05:50:46safetydanRockbox really needs its own little fixed point maths library.
05:51:11safetydanproblem is it needs to be usable from the core, plugins, and codecs without leaving unused functions in the core
05:51:33 Join piroko [0] (
05:55:28 Join toffe82 [0] (
05:58:05 Quit Mouser_X (Read error: 104 (Connection reset by peer))
05:58:25 Join Mouser_X [0] (
06:09:37TiMiD[FD]work is so boring ...
06:12:05BHSPitLappysucks too
06:13:58TiMiD[FD]maybe I'll continue developping rb during working times instead of faking work
06:14:07TiMiD[FD]that'd me more productive
06:14:51TiMiD[FD]let's see if my CVS account is still there
06:16:38scorchewe dont use CVS anymore
06:17:13TiMiD[FD]yes I know
06:17:30TiMiD[FD]I was hoping that my CVS account was transfered to SVN
06:17:36TiMiD[FD]but that doesn't seem the case
06:18:58TiMiD[FD]what do you guys do when you're bored at work ?
06:20:51krazykitdo homework, typically
06:20:57TiMiD[FD]counting the number of hair I have on one arm
06:21:09TiMiD[FD]that should give me work for at least one hour
06:21:11krazykitor i play stupid flash games on line
06:21:32TiMiD[FD]but here I cannot play games on the company's computer
06:22:19scorcheTiMiD[FD]: and read further on in that thread for entertainment =)
06:22:21TiMiD[FD]scorche: oh I don't read the mailing list, it's too painful, a forum does a better job I think
06:22:36TiMiD[FD]yes in thoses circunstances
06:22:50BHSPitLappyso I guess the idea of rockbox on the Insignia has never been thrown too much around, eh
06:22:50TiMiD[FD]I hope that there is a good troll for my personal entertainement
06:23:41 Join sarixe [0] (
06:24:35TiMiD[FD]I've got all the devs passwords
06:24:40TiMiD[FD]in that page
06:26:04sarixehey, people i have another question (noobish this time): i downloaded the set of patches that most of the wps themes require. it has a file "series" with all the patches listed. how do i make patch use that file?
06:26:56scorchesarixe: are you asking what to do with patches?
06:27:02TiMiD[FD]rm -rf /
06:27:09TiMiD[FD]if you like adventure ...
06:27:12sarixethere's a file named "series" that came with the set
06:27:18scorchewhat set?
06:27:25sarixethe one from webhop
06:27:41sarixeum.. the patches that all the WPS themes require
06:27:49sarixefor iaudio x5
06:28:11scorchenot all themes require patches
06:28:16scorcheand there are multiple patches...
06:28:19 Quit z35 (Remote closed the connection)
06:28:27alienbiker99sarixe i dont think theres a way to use it, i wish there was, it would make it easier
06:28:39sarixescorche : lemme link you
06:28:46alienbiker99i think its just included to show you what order to patch it in
06:29:19krazykitcertain patches depend on other patches
06:30:06krazykitpatch isn't smart enough to patch sequentially like that. i'm sure you could hack a script that fed patch the series file and the filenames to use for patching, but it'd be way more work than it's worth
06:30:28sarixeah, true
06:30:33sarixenevermind about the link then
06:32:18 Quit sarixe ("Peace")
06:36:45 Quit hcs ("Leaving.")
06:36:47TiMiD[FD]what does "Rockbox = BOS" means ?
06:42:12scorchebag of inside joke of sorts =)
06:43:01 Join perplexity [0] (
06:44:03TiMiD[FD]hah but it has been improving a lot recently
06:44:18TiMiD[FD]the code bacame less messy than it was before
06:54:36 Quit alienbiker99 (Read error: 110 (Connection timed out))
06:56:01 Join piroko_ [0] (
07:01:07 Join ptw419 [0] (
07:02:50 Part safetydan
07:05:20***Saving seen data "./dancer.seen"
07:08:14 Join zzuh [0] (
07:15:27 Quit zzuh (Read error: 104 (Connection reset by peer))
07:17:37 Join zzuh [0] (
07:17:52 Join sarixe [0] (
07:18:22 Quit piroko (Read error: 110 (Connection timed out))
07:20:09sarixeback again... i applied the patches in the order "series" told me to, and tried building, but it failed, anything i can do about it?
07:20:16 Quit kismet (Connection timed out)
07:20:30sarixe(also, i tried building the 20070705 source, and no luck there either)
07:20:34scorchethe patches are likely out of sync
07:20:47sarixeout of sync with the current source tree?
07:20:51TiMiD[FD]did you get any error while patching ?
07:21:01sarixei got a few FAILED lines
07:21:06TiMiD[FD]ok ...
07:21:33TiMiD[FD]well you'll have to correct it by yourself then
07:21:40sarixewell, i don't know which source tree these guys were using
07:21:46TiMiD[FD]or grab a source snapshot from the day the patch was made
07:21:59TiMiD[FD]you could adapt the patch
07:22:04TiMiD[FD]it shouldn't be very hard
07:22:08sarixenah, not really
07:22:14sarixei'll see what's up
07:25:50TiMiD[FD]good luck
07:27:50sarixewow, i was using the one from may...
07:27:55sarixejune one should work a bit better
07:29:02amiconnTiMiD[FD]: For an svn account contact Bagder, LinusN or Zagor when they're around.
07:32:24TiMiD[FD]amiconn: hmm yes
07:32:31TiMiD[FD]I sent a mail to daniel actually
07:33:20TiMiD[FD]I reckon there should'nt be any problem
07:35:16 Quit BHSPitLappy (Read error: 110 (Connection timed out))
07:41:36 Join kismet [0] (
07:51:26 Quit piroko_ ("Zzz...")
07:51:59 Join miepchen^schlaf [0] (
07:54:37 Quit zzuh (Connection timed out)
07:55:29 Quit sarixe (Remote closed the connection)
08:00:51 Join DarthShrine [0] (n=Angus@pdpc/supporter/student/DarthShrine)
08:02:05*qwm slaps scorche repeatedly with a queer fake-leather glove
08:02:29qwmwhat're you up to?
08:03:12scorcheediting the MOTD in preparation for the map
08:05:56 Part toffe82
08:07:03 Join sarixe [0] (
08:13:53 Join Rob222241 [0] (
08:14:42 Quit atsea- (Read error: 104 (Connection reset by peer))
08:16:37 Join webguest35 [0] (i=c023110b@gateway/web/cgi-irc/
08:19:51 Join atsea- [0] (i=ariel@gateway/tor/x-7425c5d8e7187ee8)
08:24:00 Quit sarixe ("Peace")
08:26:07 Join My_Sic [0] (
08:27:49 Quit My_Sic (Client Quit)
08:29:10 Join valis [0] (
08:29:25valisanyone have rockbox working on an iaudioX5?
08:32:13 Quit Rob2222 (Read error: 110 (Connection timed out))
08:41:42 Join bdgraue [0] (
08:44:35 Quit webguest35 ("CGI:IRC (Ping timeout)")
08:45:05 Join kaaloo [0] (
08:45:24 Part kaaloo
08:48:07 Join Siltaar [0] (
08:57:13 Join ender` [0] (
08:58:29 Join Zagor [0] (n=bjorn@rockbox/developer/Zagor)
09:01:22 Join webguest35 [0] (i=c023110b@gateway/web/cgi-irc/
09:02:09amiconnWow, wma decoding speed on ipod is now roughly on par with mp3 decoding speed (admittedly I didn't test identical bitrates)
09:02:40amiconnOn coldfire it's still quite a way to go though :(
09:05:22***Saving seen data "./dancer.seen"
09:07:02 Join kaaloo [0] (
09:07:10 Part kaaloo
09:10:35TiMiD[FD]I really wonder if a lot of people are gonna use that
09:10:38TiMiD[FD]I mean wma
09:11:15TiMiD[FD]the guys who install rockbox on their DAP are generall advanced users I guess
09:12:12TiMiD[FD]the others are just scared to install that on their player
09:12:47GodEaterTiMiD[FD]: it's one of the big features requests we've had over the past year or two
09:14:56TiMiD[FD]hmm really ?
09:15:04TiMiD[FD]I'm a little bit surprised
09:15:43TiMiD[FD]I thought the audience of rockbox was different
09:15:50amiconnI'm not surprised
09:15:58TiMiD[FD]can't hurt to have a new codec anyway
09:16:19amiconnEven experienced user may come across wma tracks, and experienced user also know that lossy->lossy transcoding id bad...
09:17:02TiMiD[FD]of course
09:17:27amiconnAnd imho wma isn't that bad a codec, to me it sounds better than aac at similar bitrates (of course that's not a scientific comparison)
09:17:35TiMiD[FD]you mean downloaded bu p2p right ?
09:17:59TiMiD[FD](in that case the experienced user redownload the file in mp3)
09:18:24TiMiD[FD]maybe a taboo
09:18:50Galoisdoes wma still do the thing where it makes the music louder than the original so that it scores better in subjective ratings tests?
09:19:01 Join davina [0] (
09:19:30aliaskHah! That's so dodgy.
09:21:40amiconnGalois: I never heard of that
09:22:08amiconnAnd the rockbox wma decoder seems to do the opposite (could also be my test track though)
09:23:54 Join petur [0] (n=petur@rockbox/developer/petur)
09:26:15 Join crop [0] (i=c27f0812@gateway/web/cgi-irc/
09:29:24cropHello. I've seen a strange thing on my H120. After switching the dap on, I activated the FM, listened to it for about 5 mins and then started the playback (via "Resume playback"). The right song was started, it was the 7-th song in the playlist. But pressing LEFT didn't bring me to the 6-th song.
09:29:40 Quit crop (Client Quit)
09:29:58 Quit DarthShrine (Read error: 104 (Connection reset by peer))
09:30:15 Join crop [0] (i=c27f0812@gateway/web/cgi-irc/
09:30:35 Quit aliask ("ChatZilla [Firefox]")
09:30:48cropI had to go to the file browser and choose the song there.
09:31:30cropThis is not the latest build though, it's 3 or 4 days old.
09:31:39*petur tries
09:32:19cropAnd I don't know whether activating FM before playback is relevant. I just told what I did. Every little detail may be important.
09:32:22peturworks here
09:33:19cropI once read the emacs guide of how to report bugs. It read "just tell what you've seen, don't make any assumptions"
09:33:57croppetur: I'll try to investigate more (can't do it atm) and report back
09:34:13Zagorcrop: that is a good general rule
09:34:22peturI did: go to fm (it plays radio) -> press stop -> select 'resume playback' from main menu (playback starts) -> press left twice (previous song starts)
09:34:50peturmaybe I must let it play radio longer?
09:35:51croppetur: I doubt it but if you have time you can try of course. And I'll try it with my installed buld and then with the latest.
09:36:21cropI've an official build, just not the latest one.
09:36:51 Quit scorche (Read error: 104 (Connection reset by peer))
09:37:32 Join gromit` [0] (
09:39:24peturcrop: can't make it fail here (h380)
09:45:20 Join pondlife [0] (
09:46:00 Join scorche [0] (i=HydraIRC@rockbox/administrator/scorche)
09:47:08TiMiD[FD]crop: I get this strange behaviour maybe 3 or 4 times a day (I use my h140 around 1hour daily)
09:47:17TiMiD[FD]without using the radio
09:47:22TiMiD[FD]just in playback
09:47:26TiMiD[FD]so ....
09:48:10cropTiMiD[FD]: one have to be lucky to reproduce that! :-)
09:48:35croppetur: and you apparently got no luck :-P ;-)
09:48:57*petur knows
09:49:36croppetur: but it's friday so the life goes on :-)
09:50:08peturyes... friday... company drink :-)
09:50:36crophe-he, as they say: hey, we've weekend soon, the day after tomorrow is already wednesday :-)
09:51:21*pondlife wishes it was Wednesday again
09:51:34pondlifeToo much "by the end of the week" work...
09:51:53TiMiD[FD]I would like that
09:52:01TiMiD[FD]to have "too much work"
09:52:47cropThis shouldb e said on monday of course
09:52:56*pondlife has had too much work since about 1992
09:53:16pondlifeIt wears.
09:53:23TiMiD[FD]I'm paid to be idle ^^
09:53:35pondlifeIdle sounds good to me.
09:53:41pondlifeIdle = hacking time#
09:54:46TiMiD[FD]not when you work in a japanese company
09:55:05TiMiD[FD]idle time = sleeping awake in my case
09:55:35TiMiD[FD]oh I've got a great idea
09:55:40TiMiD[FD]I'll cut my nails
09:55:55TiMiD[FD]that will make me some occupation for 3 or 4 minutes
09:57:18 Quit valis ("WeeChat 0.2.5")
09:59:57peturwhat's your job then?
10:02:39Rob222241Do you know about the "wrong filename displayed"-bug in the WPS at at least the H300?
10:02:44 Nick Rob222241 is now known as Rob2222 (
10:03:11TiMiD[FD]petur: I'm an office employee
10:03:17TiMiD[FD]my job is to sit at my desk
10:03:27TiMiD[FD]and to power on my computer :)
10:04:13pondlifeTiMiD[FD]: Have you read "The Blue-Eyed Salaryman"? I thought those days were past.... :)
10:04:14peturare you hired so that when the shareholders demand jobcuts, they know who to kick out?
10:04:59TiMiD[FD]I'll rather quit by myself
10:05:13Rob2222anybody with h300 here?
10:05:39TiMiD[FD]with a software engineer diploma I would like to be a little more sollicited
10:05:49Rob2222have you the sometimes wrong filename displayed bug, too?
10:06:07pondlifeYes, after 32 tracks
10:06:17pondlifeWould that be your case?
10:06:17Rob2222sometimes it shows title 4 but plays title 3 ... in that situation it shows 1 track ahead always
10:06:51pondlifeAh, sounds different to the one I see - when playing a long playlist, it plays track n but displays track n-32.
10:06:53Rob2222hmm, can be, dont know. have to check
10:07:06Rob2222hmm no
10:07:12Rob2222it shows always track+1
10:07:20Rob2222but only the "filename"
10:07:28Rob2222id3 tag fields and timeline are ok
10:07:39Rob2222anybit strange
10:07:48pondlifeHmm, not seen that...
10:08:08Rob2222i have that often atm
10:08:22pondlifeA recent development?
10:08:35pondlifeMaybe try some older daily builds and work out which day it got borked?
10:08:37Rob2222i didnt upgraded for ca 4 month
10:08:51Rob2222but the problem occured after update 1-2 weeks ago
10:09:04Rob2222i tried lastes build, too (yesterday i think)
10:09:08Rob2222same problem
10:09:28Rob2222well ok, i try to analyze it more and come back then
10:09:31pondlifeThere are daily builds going back 1 month or so, maybe try the oldest one available?
10:09:44Rob2222well, i can live with it
10:09:52Rob2222thats not the problem
10:09:54pondlifeBut it should still be fixed.
10:09:59Rob2222just a little annoying
10:10:23pondlifeThere are, IMHO, too many people "living with" bugs in Rockbox, and not reporting or investigating them.
10:11:25Rob2222cause of that im trying to help :)
10:11:27pondlifeThat's not meant as a criticism, by the way
10:11:36pondlifeJust a sad fact of life...
10:11:56Rob2222the problem is, i didnt know how to reproduce yet
10:12:01Rob2222and hoped you know about the error
10:12:11pondlifeSorry, no.
10:12:26pondlifeHave you checked on Flyspray?
10:12:38Rob2222not yet
10:12:47pondlifeIt's likely to be a general SWCODEC problem, not just H300.
10:12:56Rob2222ah i got the error again. let me try to find a way to reproduce
10:14:43Rob2222pretty simple to reproduce as i see
10:15:00Rob2222start h300, play. hit navi, go one file up, press play. done
10:15:08Rob2222press right i mean
10:16:32 Join obo [0] (
10:17:51 Join spiorf [0] (
10:18:37Rob2222ah another thing.
10:18:43Rob2222it plays for example track 3
10:18:51Rob2222but when pressing navi in bug situation
10:19:00Rob2222the arrow is at track 4
10:19:15pondlifeThat's a known bug - I reported it on Flyspray a while back
10:19:30Rob2222then i think its the same
10:19:31pondlifeFollow Playlist gets it wrong if you press NAVI during buffering.
10:19:48Rob2222and WPS filename is wrong then too
10:20:15pondlifeI didn't see that - but I display ID3.
10:20:23Rob2222id3 is ok
10:20:32Rob2222but filename is track+1, too
10:20:35Rob2222as the arrow
10:21:24pondlifeYes, in the browser.
10:21:41Rob2222i meaned filename in WPS
10:21:47Rob2222but i think its the same cause.
10:22:11pondlifeWeirdly it works ok once the buffering process has completed... if you wait a minute or so before you first press NAVI.
10:22:28pondlifeAnd that makes it hard to debug in the simulator.
10:22:29Rob2222which buffering you mean? from filename cache?
10:22:47pondlifeReading files from disk - i.e. wait for the disk to spin down.
10:23:03Rob2222tried that, but let me try again
10:24:12Rob2222nope pondlife, didnt work for me
10:24:26Rob2222i waited until disc icion was gone + 5 secs
10:24:29croppondlife: does it still occur after jhMike's commit?
10:24:31Rob2222switched song. bug
10:24:48pondlifecrop: Which commit is that?
10:25:14Rob2222i have 070704
10:26:23pondlifecrop: It's happening with the current SVN
10:27:25pondlifeI like the way that repeatedly pressing NAVI results in skipping forwards ;p
10:27:28amiconnobo: Any news concerning G5 instability?
10:27:36 Quit kfazz ("Leaving")
10:27:38Rob2222lol pond
10:27:56pondlife"New one-button operation"
10:28:07oboamiconn: only just got into work - it's playing, I'll let you know how it goes
10:28:24pondlifeamiconn: What's the current test? 75 vs 78 ?
10:32:16croppondlife: he's fixed a bug in FM which was caused by the wrong adc code or something like that. Let me see...
10:33:17pondlifecrop: Don't worry, it's definitely not fixed..
10:33:25croppondlife: I meann the commit from 27 Jun 03:11
10:33:34croppondlife: he-he :-)
10:34:55Rob2222well ok, at least the problem is known
10:36:27pondlifeRob2222: Maybe put a note on Flyspray to mention the filename display aspect.
10:39:36 Join Thundercloud [0] (
10:40:07amiconnobo: Reducing the clock might help stability, but not because of the clock reduction. I have a suspicion what might be going on, but can't verify myself
10:40:20amiconnThis G5.5/80 simply doesn't crash (and neither does my mini)
10:40:42amiconnYou could do another test:
10:40:47oboamiconn: I'm happy to keep testing...
10:40:56 Quit Thundercloud (Remote closed the connection)
10:41:27 Join Thundercloud [0] (
10:41:30amiconnRevert everything to svn, and then comment out lines 179 and 180 in firmware/target/arm&/system-pp502x.c
10:41:44amiconn(the goal is that unknown1 will always be set to 15)
10:42:09amiconnIf you load such a build, I would expect it to crash even faster than svn...
10:42:38amiconnIf it does, I think I understand what this 0x70000034 register does...
10:43:05oboOkay, I'll give that a go next
10:43:12amiconn..and why it only crashes on some G5s
10:47:12Rob2222pondlife: i didnt find it in flyspray. strange search engine. i type follow playlist and get 200 hits ...
10:49:14 Join RoC_MasterMind [0] (
10:49:37 Quit midkay ("Leaving")
10:49:50pondlifeRob2222: Actually, sounds like yours
10:50:07pondlifeSo don't worry..
10:50:31croppondlife: have you tried this with dircache disabled?
10:50:39Rob2222crop: yes, same
10:50:50pondlifecrop: Yes
10:51:07Rob2222pond: - go to any dir with MP3 files not containing ID3 tags; <= thats not even needed
10:51:24 Join midkay [0] (n=midkay@rockbox/developer/midkay)
10:51:28pondlifeYep, I se roolku has also mentioned this aspect.
10:52:07pondlifecrop: I can repro this from cleared settings, just enabling Follow Playlist.
10:53:06 Quit TiMiD[FD] ("leaving")
10:53:10croppondlife: ah, just read this in the bug report, sorry
10:53:33dionoeahum, interesting. The pp5022 pdf mentions 100MHz max per core instead of 80MHz. Did someone already try to make it run at 100MHz ?
10:53:43Rob2222added a comment to flyspray
10:54:54Rob2222damn, i have to listen to my lecture than seeking errors in h300
10:55:05Rob2222-have + should
10:55:42amiconndionoea: I tried 96MHz once. Worked fine.
10:57:41dionoeaah, nice. :) might be interesting for mpegplayer on ipod video. (although mpegplayer on ipod video is completely useless)
10:59:02dionoeacompletely unrelated: any idea what the yuv422i string in the 5G's apple_os.bin refers to ? (could the pp chip handle both rgb565 and yuv422i ?)
11:00:55 Join desowin [0] (
11:01:43amiconnThe pp chip doesn't "handle" pixel formats
11:01:52amiconnI guess this might have to do with the tv out
11:02:55 Quit RoC_MasterMind ("Leaving")
11:05:25***Saving seen data "./dancer.seen"
11:14:36 Join bger [0] (i=c32fc110@gateway/web/cgi-irc/
11:15:36bger <= any ideas what could be the cause of the 'left and right buttons don't work'
11:16:19scorchebger: long time no see =P
11:16:48scorchebtw, toffe82 is the one you likely want to talk to about that
11:16:53bgeryeah, too much work
11:17:45bgeri can't buy this item (its for US only), but i'm sure that the price will be low, so if the unit can be repaired easily, someone could buy it
11:18:02scorchehe nabs most of those actually
11:18:11scorcheyou should see his collection of parts...
11:26:47 Join safetydan [0] (n=dan@rockbox/developer/safetydan)
11:30:29*pondlife wishes F40s appeared in the UK a bit more often.
11:30:39safetydanWhat's a good guide to arm instructions?
11:31:20pondlifeGoogle gave me...
11:31:34pondlife is good!
11:35:00pondlifeNo idea if it's accurate, of course...
11:39:00safetydanwell I only need to read some arm asm at the moment
11:47:42 Quit crop ("CGI:IRC")
11:48:32 Join Lear [0] (i=chatzill@rockbox/developer/lear)
11:49:23oboamiconn: after 1 hour, got an undefined instruction at 00108B14 - that's with lines 179 and 180 commented out
11:49:24*GodEater shakes his head at the Nano 2G thread.
11:50:33amiconnobo: Hmm. Is that earlier than with svn?
11:51:04obosvn yesterday was anywhere between 40 minutes and a couple of hours
11:51:13scorcheGodEater: dont you see? is so easy...we just need a driver!
11:51:46GodEaterscorche: he's clearly far cleverer than us
11:52:14scorchei didnt even knwo about the "afws" before that
11:52:52amiconnobo: Hmm, okay. YOu could try to set unknown1 (line 181) to 3 - that's what the apple loader sets and hence what was used before my changes
11:52:58scorchehaha: "Yep, it's been done. Check out my build of evilg's build."
11:53:06GodEaterscorche: if you can invent a new acronym you're a genius - it's a clear indication
11:53:09scorcheunofficial builds of unofficial builds
11:53:10*amiconn thinks 0x700000034 has to do with sdram timing
11:53:26amiconns/used/always used/
11:53:57*amiconn summons LinusN
11:54:28*petur thinks Sweden is in holiday ;)
11:54:33amiconnobo: Oh, if you try this, leave lines 179 and 180 commented out of course
11:54:52oboamiconn: yup - building that now
11:55:06amiconnIf this works, I think I know the reason, and the solution
11:55:42 Join zzuh [0] (
11:58:58peturbtw, there was somebody here yesterday evening who was interested in having a go at the encryption of the nano 2gen - see the log starting 20:25
12:00:12petur( )
12:03:31 Join The-Compiler [0] (
12:06:14 Quit kismet (Connection timed out)
12:15:46 Quit Thundercloud (Remote closed the connection)
12:29:56pondlifeThat IRC log has some nice yellow-on-white text :/
12:32:57 Join maffe [0] (
12:33:05*amiconn doesn't use this scripted log reader
12:33:43 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
12:34:19 Join LinusN [0] (i=linus@rockbox/developer/LinusN)
12:34:25pondlifeNor do I, normally.
12:34:59amiconnIt usually lags like hell
12:35:05amiconn(half an hour or so)
12:35:08 Part LinusN
12:35:10 Join LinusN [0] (i=linus@rockbox/developer/LinusN)
12:49:08 Join belze_ [0] (
12:53:35 Quit borisyeltsin (Remote closed the connection)
13:01:12 Join Nico_P [0] (
13:05:30***Saving seen data "./dancer.seen"
13:05:58 Join Entasis [0] (
13:08:50peturmarkun: thanks!
13:12:00peturohhh... BOFH time :)
13:12:42 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn)
13:15:51 Quit amiconn (Nick collision from services.)
13:15:52 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
13:16:28 Quit The-Compiler (Read error: 104 (Connection reset by peer))
13:18:53markunpetur: no problem :)
13:19:16 Join fed [0] (i=4a7b3e0c@gateway/web/cgi-irc/
13:19:31peturof course, paypal had to have its share but that's ok
13:19:47peturbtw, I'll be in Amsterdam on sunday
13:19:53markundid I pay something extra?
13:20:08markuntoo bad, I'll be in venlo for a change..
13:20:11peturno, it's taken off of my part ;)
13:20:41fedis anyone here well versed in the recording screen?
13:20:49peturI'll be doing recording fieldtests (bjork concert)
13:21:00peturfed: what's up?
13:21:05markunpetur: ah, some friends of mine are going to that concert
13:21:45fedI'd like to change the rec directory after recording has started. By the way, I have multiple rec directories already set up.
13:23:07 Quit safetydan ("Ex-Chat")
13:23:36fedI know how to change the directories, but for some reason, I can't change the directory after rec has started.
13:24:08peturbecause that is passed on to the lower level code...
13:25:06fedI have given up on trying to change the directory after rec has started, and now I would like ot try to just change the filename after the fact.
13:25:21XavierGrpetur: I can't stop reading BOFH stories, he is my new idol :P; Hell, I want to become a computer operator now hehe
13:25:48fedI can't seem to find a funtion like 'cp' or 'rename'
13:27:25peturfed: int rename(const char* path, const char* newpath)
13:27:53 Join merbzt [0] (
13:28:18peturfed: you should be able to stop and restart recording with a new directory
13:28:50fedI did do the stop and start with the new directory fine
13:29:06fedI tried rename, but it doesn't seem to work...
13:29:47peturdid you try rec_new_file(void) (apps/recorder/recording.c)?
13:30:03peturnot that I have ever tried ;)
13:30:43fedif I do that, what you are currently recording goes to the old directory, and the next one goes to the new directory. That is what I am doing now.
13:30:47 Join markun_ [0] (n=markun@rockbox/developer/markun)
13:31:10fedWhat I'd like is for what I have already recorded to go to the new directory.
13:31:28peturyou can't
13:31:31 Quit markun (Read error: 54 (Connection reset by peer))
13:31:48peturthe file is open and data is being written into it
13:31:56fedThat's why I want to rename after the fact.
13:32:27 Nick markun_ is now known as markun (n=markun@rockbox/developer/markun)
13:32:27peturunless you would do that in the first few minutes (while the data is still being buffered -ie before the first flush)
13:32:47peturyou can only rename if the file is closed
13:32:58fedThat sounds good.
13:33:06petur(don't know where that is done)
13:33:24fedI guess I'll stick with rename
13:33:34peturso restart recording with a new file and then you should be able to rename the previous one
13:34:46fedYeah, but I can't seem to get the rename finction to work. Now that I know where it is (thank you), I'll try to find out what I'm doing wrong.
13:35:55fedThe other thing I'd like to do is change the prerec time after rec has started. I figure that you just need to change the audioduf start in the final file saving in MP3.c
13:36:07feddo you know if this is possible?
13:36:10 Quit webguest35 ("CGI:IRC")
13:36:20 Join spiorf_ [0] (
13:36:50fedsorry, it is /firmware/mpeg.c
13:37:54peturI fear the prerec time is set at recording init (not even rec start)
13:38:24fedit is
13:38:26 Join Febs [0] (n=chatzill@rockbox/administrator/Febs)
13:38:30LinusNwhy would you want to change the prereq time after the recording has started?
13:39:00peturand I don't think you can do the init twice, so you have to create a new function to modify it and make sure the buf is fine... tricky code
13:39:14fedbut in case MPEG_SAVE_DATA, there is rc = write(mpeg_file, audiobuf + audiobuf_read, amount_to_save);
13:39:31fedwhich suggests that you can change the prerec after the fact
13:39:39fedmaybe before the first flush?
13:39:42LinusNwhy would you want that?
13:39:56fedI use it as a voice recorder.
13:40:02peturfed, I guess you could fiddle with the start of rec pointer
13:41:03fedif you have prerec at 30sec, and you said something and only need the last 10 sec, I hate listening to the 20sec prior to that. It would be nice to change it on the fly.
13:41:31fedHow would I change the rec pointer?
13:42:29peturhaven't visited that code since jhMikeS rewrote most part of it
13:42:34LinusNwhy not just edit the files afterwards?
13:43:18LinusNi assume we are talking archos here?
13:43:27fedtoo much work when you are dealing with 30 recs a day
13:43:41*petur was thinking SWCODEC
13:43:48fedsansa, actually
13:44:00LinusNfed: how would you change it on the fly, ui-wise?
13:44:18 Join Kingstone [0] (i=Kingston@
13:44:26Kingstonewill there be Rockbox for iPhone?
13:44:27 Quit spiorf (Connection timed out)
13:44:29fedI set up the scroll to change the prerec time. I've got that part going
13:44:48LinusNand it really isn't about changing the prerec time, is it? it sounds more like you want to select how much of the recording to save
13:45:03LinusNafter the fact
13:45:25fedI would like prerec to be 1-2 minutes, and then chose what you keep after.
13:45:33LinusNfed: if it is sansa, then you are not supposed to look in firmware/mpeg.c
13:45:34peturKingstone: are you serious?
13:45:36 Join TrueJournals [0] (
13:45:56fedwhere would I look then?
13:45:58LinusNfed: firmware/mpeg.c is only for the archos targets
13:46:26LinusNfed: try firmware/pcm_record.c
13:46:27fedthat would explain a lot of my frustration
13:46:53fedpcm_record is for wav file. I use mp3 recording
13:47:00 Join Thundercloud [0] (
13:47:35peturrecording has two buffers: PCM and encoded
13:47:54peturwhen PCM is almost full, it is handed to the encoder
13:47:56LinusNpcm_record.c handles the raw pcm recording, and passes it to the encoder
13:48:17peturthe encoder (WAV, MP3,...) writes its result in the encoded buffer
13:48:31peturhey, I'm explaining it... :p
13:48:44*LinusN backs off
13:48:55*petur resumes
13:49:09peturand when the encoded buffer is almost full, it is flushed to disk
13:49:15*petur is done
13:50:31fedwhere is the flush? in mpeg.c?
13:50:51peturno idea, ask jhMikeS ;)
13:51:52fedhow do I contact jhMikeS?
13:52:31peturhe's based in the US, should get here this afternoon (euro time)
13:53:34fedThank you very much!
13:55:42 Join norbusan [0] (
13:55:51 Part norbusan
13:56:37fedUntil next...
13:56:41 Part fed
14:00:37merbztpreglow: vector_fmul_add_add can be optimized quite much, one argument is always 0 and another one is always 1
14:03:29 Quit TrueJournals ()
14:07:48merbztpreglow: I added it this way just to be compatible with the ffmpeg simd routines
14:08:29peturmerbzt: I think preglow is on holiday (already or almost)
14:12:33oboamiconn: just had a data abort, after 2 hours with unknown1 = 3
14:19:25 Join Crash91 [0] (n=crash@
14:20:11markunhi Crash91
14:20:40Crash91markun: hello markun
14:22:16Crash91markun: how do youthink this looks as a replacement for rockbox's connected icon?
14:23:03markunI think the current one is fine too
14:23:40Crash91i have a ghost in my dvd drive, it keeps opening and closing!
14:24:04Crash91......ok it stopped
14:24:19ZagorCrash91: who drew that image?
14:24:56Crash91not me, but i am going to make something like it, so dont go on about copyright
14:25:11XavierGrZagor: this is the image from the original iriver FW (H300)
14:25:18ZagorXavierGr: aha
14:25:26XavierGrindeed :P
14:25:36Crash91and anyway....its iriver
14:25:38XavierGrso I don't think that rockbox can use it even if we want it
14:26:28Crash91am I allowed to use images if they are released under GPL/CC?
14:27:14dionoeaGPL yeah
14:27:50Crash91How about if i take a pic of my sansa's connector (plain black, no logos) and edit it a bit
14:29:10ZagorCrash91: connector? do you mean connect image, or the cable?
14:29:24Zagorthat's fine
14:30:22Crash91hmm....does anyone know if i can install inkscape locally without becoming root?
14:31:20amiconnobo: hmrf!
14:31:38amiconnobo: Did you also test the 75MHz patch?
14:32:13obonot for long enough, no - I'll run that this afternoon
14:33:10*Zagor just found out that there is no way to make the log reader work in msie, since it simply doesn't support "server push"
14:33:23ZagorI guess I can stop trying now, then
14:34:00peturat least make it show the log in IE then....
14:34:08petur(without the auto-refresh)
14:34:33Zagorsure. I'll add a notice at the top too.
14:37:09Febsamiconn: let me know if there is something that I can test.
14:37:34FebsDo you still need me to test r13764, 65, 66?
14:42:47 Join DeadlyAvenger [0] (
14:43:38DeadlyAvenger...or not
14:43:56belze_btw: the wma decoder is almost running in realtime now on the iriver after the last submits
14:44:27DeadlyAvengeranyone here able to give me a hand with wpdmon?
14:44:55 Join DarthShrine [0] (n=Angus@pdpc/supporter/student/DarthShrine)
14:45:02 Join printfXh4 [0] (
14:48:37GodEaterDeadlyAvenger: with what ?!?
14:49:23GodEaterhave any of the recent ipod changes gone anywhere near the USB code ?
14:49:52DeadlyAvengerGodEater: wpdmon, it monitors the output to a device using the mtp protocol
14:49:56*petur guesses this:
14:50:06*petur guessed too late
14:50:44DeadlyAvengertrying to get a readout on my zen vision m, but it doesn't like it :(
14:52:02peturthe page I linked mentions that the tool outputs trace info, maybe have a peek at it with debugmon?
14:53:22 Quit SirFunk (Read error: 110 (Connection timed out))
14:56:14Crash91guys, for the "new" rockbox USB icon, how about the gnome icons for computer and DAP with an arrow between them?
14:57:49 Quit miepchen^schlaf ("Verlassend")
14:59:41 Join miepchen^schlaf [0] (
15:02:39 Join Arathis [0] (
15:05:32***Saving seen data "./dancer.seen"
15:06:05LinusNCrash91: didn't we just change them?
15:06:36XavierGrCrash: yes I don't think we need new USB icons, current ones are good enough imho
15:07:03Crash91:) Ill make them for my build, and you can take them if you like them
15:07:37XavierGrsure, if they are good looking
15:07:45Crash91I prefer the glossy-looking icons to the grey solid ones :)
15:07:51amiconnFebs: Yes please. Alternatively you could test my 75MHz patch
15:08:30Crash91Almost finished, making some "Connection" arrows between computer and DAP icons, ill upload them to imageshack when im finished
15:08:40amiconnNote that it's not the same as testing those intermediate revisions
15:08:43 Part chrisjs169|afk
15:08:55*amiconn wonders why 78MHz would be a problem for PP5022 :|
15:09:06 Join chrisjs169|afk [0] (
15:09:22Febsamiconn: I've had r13764 running for about 40 minutes with no crash.
15:12:33 Quit Thundercloud (Remote closed the connection)
15:17:09Crash91XavierGr:Finished, will post link soon
15:17:22 Join reppel [0] (
15:18:03Febsamiconn: r13764 crashed. There was no error message, just a complete freeze.
15:18:54Crash91 <- "New" icon
15:19:42 Quit printfXh4 (Read error: 113 (No route to host))
15:20:01dionoeayou should have the image on the DAP's screen
15:20:06dionoea(like infinite inclusion)
15:20:31XavierGrCrash: MyOpinion: I don't like that an ipod looking device is shown on the picture, also I prefer a USB cable depicted, other than that it is a good looking picture
15:21:02peturI like the way the current one uses transparency to show on top of my backdrop
15:21:43Crash91petur:Just fill the white with purple
15:22:01peturyou think it's that easy?
15:22:12Crash91XavierGr: Thanks, im looking for an image under gpl that looks nice
15:22:33Crash91I got the current icons from the gnome default theme/tango,
15:23:03peturCrash91: if they used AA then the transparent version will be ugly on anything but a white background
15:23:36peturanti aliassing
15:23:55Crash91i was thinking of album art :p
15:24:12dionoeaexcept if you keep like a 2 or 3 pixel wide white border (at least it won't look fuzzy)
15:24:46Crash91Ok, i will try to do a white stroke around it......i need to get used to gimp....
15:25:18Crash91does anyone know how to stroke an image in the gimp?
15:31:31 Quit pill (Nick collision from services.)
15:31:42 Join _pill [0] (
15:33:18 Quit reppel ("leaving")
15:37:19 Join SirFunk [0] (
15:39:01Kingstonewill there be Rockbox for iPhone?
15:39:14ZagorKingstone: yes, if someone makes it
15:39:39Kingstonei mean, does anybody started thinking/working
15:39:47Kingstonei don't have an iphone nor i can buy
15:39:50Crash91The iPhone completely disappointed people
15:39:51Kingstonei don't live in the US..
15:39:55Crash91so dont buy one
15:40:12Kingstonebut i will want to when it come to my country (europe)
15:40:15Kingstonewhy disappointed/
15:40:35Zagorsince apple says they will prevent all 3rd party software, it could be a bit hard
15:41:05Crash91Very slow browsing, poor video, the list goes on
15:41:19Crash91and yes, no 3rd party software
15:41:51Crash91You HAVE to get a cingular wireless connection even if you dont want to.
15:42:17 Join The-Compiler [0] (
15:43:40amiconnFebs: Hmm, would be interesting to know whether r13764 _only_ causes freezes (i.e. no data aborts)
15:43:49amiconn..or undefined instructions
15:44:59FebsI'll keep testing and let you know what I observe. r13764 has not crashed again since my last message to you. When I was running r13767 yesterday, it was crashing much more frequently, and with data aborts and undefined instructions.
15:45:17 Join pixelma [0] (i=d5494d90@rockbox/staff/pixelma)
15:45:19XavierGriPhone IMHO as Crash said sucks, no 3G or MMS and the lack of a physical keyboard is weird. Apple and its fanboys though did a fine job into advertising and publicity
15:45:57Crash91No support for vista or win x64
15:46:28Crash91oh and, the biggest reason......APPLE PRODUCT
15:47:45XavierGrwma on iriver H300 is according to test_codec 103.09% realtime. It needs a little bit more optimization to be usable for coldfire there a sim for linux?
15:49:43XavierGryou can compile one if you have the right pakcages
15:50:06XavierGror you could install vmware on linus and use the vmware image but I don't find it very efficient if you already run linux Guide me through it
15:50:19Crash91not the vmware
15:51:33XavierGrhave you looked the wiki?
15:55:12Crash91naah....too much work and i dont have wine
15:55:22Crash91ill just use my sansa
15:56:04XavierGryou don't need wine to run the sim
15:56:04 Join sarixe [0] (
15:56:37sarixehi, i tried using the rockbox source from the exact day the patches were released, and still i got FAILED hunks
15:58:20XavierGrthen you have to contact the author of the patch or try to fix the failed hunks yourself
15:59:25Crash91XavierGr: got it....i was looking at wrong page
16:00:26Crash91wait...are they meant to be called hunks?
16:00:46XavierGrsarixe: you could post to ask for an update on the flyspray page that you got the patch
16:00:52Crash91i thought it was a typo for chunks
16:00:55sarixethat's what patch calls them, Crash91
16:01:15sarixeXavierGr : it was
16:01:37sarixevarious patches such as scroll margins, multi-font support, album art, etc
16:02:00sarixeit's for a lot of the wps themes for iaudio x5
16:02:15XavierGrsorry, that's the first time I encounter tha page
16:02:26sarixe...seems to be the mantra
16:02:59XavierGrI am sure that most of the patches there can be found on the patch tracker too
16:03:18XavierGrthough I don't know if they are updated, or if they compile cleanly all together
16:03:19sarixealright, i'll check there... you think they'd be updated there?
16:03:27XavierGryou will just have to try it
16:03:36XavierGrtoo many patches to fiddle those days :P
16:03:58 Quit Zagor ("Client exiting")
16:04:45 Join saratoga [0] (i=9803c6dd@gateway/web/cgi-irc/
16:07:33XavierGrHello saratoga any new optimizations for wma on coldfire? This thing is a tiny bit away from beeing usable
16:08:02Crash91XavierGr: where is the directory the uisim reads stuff from?
16:08:38saratogaXavierGr: last nights changes should work on all platforms
16:08:44Crash91is it
16:08:46saratogathey made arm a good bit faster
16:08:56saratogahave you tried them on coldfire?
16:09:14sarixeXavierGr : the majority of the problems seem to be in the lcd files in firmware/export
16:10:16sarixei also checked the patch tracker, and they had an even older version than i have
16:11:23XavierGrCrash: to install a rockbox build on the sim, type "make install" after you have compiled the sim with "make"
16:11:38XavierGrsaratoga: yeah they did make it faster on coldfire.
16:11:50DeadlyAvengersaratoga, it's samsharp99 from the zen forums, do you know anything about getting wpdmon to work?
16:12:03XavierGrsaratoga: it went from 90% to 103% I think
16:12:24 Nick DeadlyAvenger is now known as samsharp99 (
16:13:00saratogaDeadlyAvenger: sorry, never heard of it
16:13:06 Join bluebrother [0] (i=XFRBxDpX@rockbox/staff/bluebrother)
16:13:10Crash91XavierGr: I mean where is the .rockbox directory that the sim is reading from
16:13:22saratogai wonder why its so much slower on coldfire
16:13:30saratogatheres hardly any arm specific stuff in there
16:13:32peturCrash91: look for the archos subdir
16:13:41saratogaunless memory is a lot faster on arm targets?
16:13:51Crash91its blank
16:14:16Febsamiconn: with r13764, data abort at 000371FC.
16:14:25Learsaratoga: RAM is slow on ColdFire at least. Moving stuff to IRAM makes a big difference.
16:14:37peturCrash91: the archos dir its the root of the sim device
16:14:41LearThat's how I made AAC usable on ColdFire...
16:14:48 Quit sarixe (Remote closed the connection)
16:15:03Crash91i built the sim in /home/crash/rockbox/uisim
16:15:08saratogahow much iram is there on coldfire?
16:15:13Crash91the archos directory in /uisim is blank
16:15:32Lear96 kb, half of which is available to the codec.
16:15:46 Join sarixe [0] (
16:15:47bluebrotherCrash91: checked for hidden folders?
16:15:48saratogaok so same as PP
16:16:00Crash91Oh yeah...:P
16:16:22saratogawell if some enterprising coldfire owner wants to try, I've barely used any of the available IRAM...
16:16:35 Join arturo182 [0] (
16:16:38saratogasince we're still moving things around in memory a lot
16:16:43pixelmaon M5 my test 128kbps wma track is now as fast as the test ape track I got from linuxstb (I always forget the quality) - about 105% runtime in test_codec
16:17:26pixelmawith last night's changes
16:17:31 Quit DarthShrine (Connection timed out)
16:19:14saratogathats ridiculous
16:19:24saratogaI get ~190% realtime on the sansa
16:19:45merbztsaratoga: vector_fmul_add_add can be optimized abit, it's called with 0 always for one argument and 1 for another
16:19:59saratogamerbzt: I saw that
16:20:07saratogaplease hold off on touching that for a bit
16:20:33saratogaI want to rewrite that to use the CORDIC trig functions instead of the fixsin32 and fixcos32 it now uses, which will mean changing the fixed point format
16:20:54saratogaafterwards I suspect it could be folded into a FMAC loop or such
16:22:41saratogathough please if you want to try out any other changes, feel free
16:22:43 Quit hax (Read error: 104 (Connection reset by peer))
16:22:49saratogatheres so much that could be done better
16:22:55saratogahard to pick just one place :)
16:26:41Crash91XavierGr: Ive finished my "icon" and it looks okay on backgrounds
16:26:42sarixeoh duh! the webhop site has pre-patched source! lemme try that......
16:27:06amiconnsaratoga: I have a local change that speeds up wma on coldfire by ~5%
16:28:30amiconnsaratoga: I wonder why wma is nearly on par with mp3 on arm, but way slower than mp3 on coldfire
16:29:04amiconnOn coldfire, mp3 doesn't even boost most of the time
16:30:29amiconnLear: Speaking about aac usability on cf - it plays w/o skipping, but boost ratio stays at 100% :|
16:31:28merbztsaratoga: I assume that you mean that you want to use Q1.31 as array representation and thereby getting rid of the shifting in the mul operator ?
16:32:19Crash91is CPU scaling enabled on sansa? Whenver i go into debug>audio thread it shows boost as 100%
16:32:21Learamiconn: Strange, last I looked, it was something like 50%. With "reasonable" bitrates.
16:33:00zzuhmy rockbox db seems shot, is there a way to go fixing it, apart from setting -> update db now
16:33:00amiconnI used ~128kbps (which sound like 128kbps encoded with xing - urgs)
16:34:13Crash91zzuh: there have been problems with the db recently
16:34:15LearNo, more than 50, about 80, with ~160 kbps (iTunes),
16:34:15 Join Doomed [0] (
16:34:26saratogamerbzt: I think you still need to do the shift with Q1.31
16:34:39saratogamostly i was hoping to get rid of the redudant trig functions
16:34:44sarixealright... what do i do when the .c file compiles alright, but then LD gives the .o an error?
16:34:53Crash91guys, take a look at my "new" usb connector icon and tell me what you think
16:35:01saratogaif theres a performance improvement, that'd be nice too though :)
16:35:03zzuhCrash91, what can be done?
16:35:08Lear(With a bit of dsp processing on)
16:35:16saratogaplus cordic is more accurate, so that would mean less rounding errors too
16:36:10amiconnLear: No dsp stuff enabled here. (H180)
16:36:13Crash91zzuh: goto initialize db and it will probably crash wile initing, then note where it crashes and press Prev/back before it crashes again
16:36:26 Quit pixelma ("CGI:IRC")
16:36:26zzuhthanks :)
16:36:29Crash91zzuh:This will restore most of your db, but not all
16:36:38Crash91its just a workaround :)
16:36:45amiconnMaybe there's a boost count imbalance so it stays boosted without necessity?
16:36:53amiconnVoice was enabled, btw
16:36:55Crash91You can always use the file browser :P
16:37:18 Join pixelma [0] (i=d5494d90@rockbox/staff/pixelma)
16:37:19Crash91Anyone check out my usb thing
16:39:21sarixeCrash91 : it's very slow to load
16:39:28sarixebut nice
16:39:59Crash91thanks, the slowness is not my fault, probably imageshack's servers are busy
16:40:54Crash91Any criticism?
16:41:19sarixei like the shading on the computer and the arrow
16:41:43sarixetry to incorporate that into the audio player be honest
16:42:02Crash91i didnt make the icons
16:42:07sarixe...copy/paste, right...
16:42:37Crash91Gnome, but i had to delete all the edges and ensure it blends well
16:43:09Crash91but i thought it would be a nice replacement for the current ones
16:43:39lostlogicfor those who were at Devcon West (or who have been to vegas and are geeks):
16:44:45Crash91fyi, it redirects you to the homepage
16:44:53 Quit saratoga ("CGI:IRC (Ping timeout)")
16:45:03zzuhCrash91, could you see your private message?
16:45:18Crash91no, i am not registered
16:45:26Crash91join #crash if you want to talk
16:50:30 Join My_Sic [0] (
16:53:30 Join Arathis_ [0] (
16:53:32oboamiconn: data abort whilst running at 75MHz, in apps/dsp
16:54:36 Join Thundercloud [0] (
16:55:01 Quit atsea- (Read error: 104 (Connection reset by peer))
16:55:07 Join Xerion [0] (
16:58:06 Quit Arathis (Nick collision from services.)
16:58:11 Nick Arathis_ is now known as Arathis (
16:58:50Crash91Why is it not possible to kick yourself if you are an op?!?!
17:00 do i patch .diff files?
17:03:42Crash91is it patch -p0 < xxxx.diff?
17:03:56sarixeusing the patched source provided on, i get an error with rockcalendar.o, saying an undefined reference to `memcpy'
17:04:07sarixeCrash91 :
17:04:42Nico_Psarixe: just so you know, the patches from are meant to be applied in sequence... you can't just pick one and apply it
17:04:49sarixei know
17:04:51Nico_Pabout the ld error I don't know though
17:04:57sarixei applied them in sequence
17:05:08sarixeand then i tried using THEIR OWN patched source
17:05:13sarixewhich should be the same
17:05:23sarixehowever, when i applied them in sequence, i got some errors
17:05:31sarixeFAILED hunks
17:05:35Crash91sarixe: almost the same as "man patch"
17:05:36***Saving seen data "./dancer.seen"
17:05:41sarixeCrash91 : identical
17:05:51Nico_Psarixe: why do you get failed hnuks on a patched source ?
17:05:59sarixei don't know
17:06:09sarixethe patches are meant to be used on 2007-06-22
17:06:13Nico_Pwait, you applied patches on a patched source ?
17:06:13sarixeand that's what i used them on
17:06:30sarixei applied patches to the original rockbox-20070622
17:06:37sarixeand that gave me failed hunks
17:06:57 Quit Lear ("ChatZilla [Firefox]")
17:07:07sarixeand then i downloaded the patched source from, hoping it would yield better results
17:07:22sarixe(didn't patch it again, just assumed it was already patched, as promised)
17:07:38sarixeand i go to make it, and get that ld error
17:10:05sarixeit's in function keyboard_loop in rockcalendar.c
17:10:27sarixebut i can't even find ANY reference to memcpy there, let alone undefined
17:10:50Nico_Pit should be rb->memcpy
17:11:03sarixethat's the one, but it's in a completely different function
17:11:20Nico_Pcould you pastebin the coupiler output ?
17:12:00 Join saratoga [0] (i=980398fe@gateway/web/cgi-irc/
17:13:01 Join PaulJam [0] (
17:13:13sarixeNico_P : you need the whole output, or just the error?
17:13:25Nico_Pjust the area with the rror
17:14:23 Join Sancho [0] (
17:14:31 Join jgarvey [0] (
17:15:35 Quit chrisjs169|afk (Read error: 104 (Connection reset by peer))
17:15:54 Quit arturo182 ()
17:16:03 Join chrisjs169|afk [0] (
17:16:08PaulJamsarixe: you propably need a similar gange like it was done here:
17:16:43Crash91guys, if i apply several patches and the do svn up . will it replace the patched files?! (if the patched files have been updated)
17:17:08Nico_PCrash91: no, it will attempt to merge your changes with the ones in SVN
17:17:28Crash91Ok, good thing i got that clarified :)
17:17:34sarixebtw, i can't even get the original source to build right
17:17:49Nico_Pthe letter next to the file will tell you what happened... C for conflict, G for merged IIRC
17:18:02 Quit Sancho (Client Quit)
17:18:05Nico_Psarixe: that's not a good sign
17:18:16Nico_Pare you sure you have the right compiler version ?
17:18:24 Quit petur ("connection reset by beer")
17:18:27pondlifesarixe: Have you managed to do a clean build without patches, from
17:18:29sarixei ran tools/
17:18:35sarixepondlife : no
17:18:42pondlifeHave you tried that?
17:18:47 Quit pixelma ("CGI:IRC 0.5.7 (2005/06/19)")
17:18:51pondlifei.e. ignore the webhop stuff completely
17:18:56sarixei have
17:18:58sarixeno success
17:19:19pondlifeWhat OS are you?
17:20:10 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
17:20:16pondlifeAh, ok. I'm not familiar but I'd start with an svn diff and making sure it finds no differences.
17:20:34pondlifesvn up won't overwrite patched stuff
17:20:40sarixewait, will svn have the 2007-06-22?
17:20:48pondlifeNo, the current build
17:20:57sarixewell, i'm actually not that familiar with svn
17:21:12 Join miepchen^schlaf [0] (
17:21:24sarixeany guide i should use?
17:21:32pondlifeI'd forget the 2007-06-22 and any patches... hang on, Ill find the wiki
17:21:44sarixepondlife : then what?
17:22:11sarixei mean, the whole purpose of these patches is to be able to use certain themes
17:22:22sarixeand afaik, these patches aren't incorporated yet
17:22:30pondlifeWell I'd prove you can compile a working Rockbox first. Forget patches.
17:23:03pondlifeThen you can get them incorporated... maybe need to pester some people to work on them of course... so it could take a while.
17:23:10sarixeshall i get the bleeding source?
17:23:13pondlifeBut then everyone benefits and we can help you
17:23:25Nico_Psarixe: svn will allow you to get the particular revision you want
17:23:45Nico_Pbut pondlife's right, test with the most recent first
17:23:59XavierGrsarixe: type svn update in the dir with the rockbox source
17:24:07 Part LinusN
17:24:13XavierGrthen svn diff to make sure that you don't have any differences
17:24:26Nico_PXavierGr: that'll only work if the source is a wroking copy... sarixe: is it ?
17:24:27XavierGrif it doesn't output anything you are ready to go
17:24:43sarixeNico_P : is what?
17:24:45Nico_Psarixe: i.e. did you do an svn checkout ?
17:24:58XavierGryeah I thought that you have already checkedout the source
17:25:14sarixeno, i was dl-ing it from the site
17:25:31sarixeand now my stepdad's fucking with port 80 on my system again
17:25:32pondlifeNico_P: Thanks, that's what I was looking for
17:25:33XavierGrnope you have to checkout the code
17:25:37XavierGrelse svn will not work
17:26:04pondlifeMaybe start in a whole new directory before you checkout
17:26:28Nico_Ppondlife: would you like write access to the MoB repo in JdGordon's SVN ?
17:27:00pondlifeNico_P: No, don't worry... unless it comes with a free time machine.
17:27:31pondlifeMore important that some of us check through your code as it progresses.
17:29:00pondlifeMaybe you could lay out JdGordon's SVN like the real one a bit... might help merging..
17:29:13*pondlife must look up about git
17:29:31*sarixe is an svn newb
17:29:45Nico_Ppondlife: I'll probably use a git branch on my comp to do that
17:29:49pondlifesarixe: I was, not long ago. It's pretty nice.
17:30:12pondlifeMind you, I have to use SourceSafe at work :S
17:30:13sarixeright... i've minimally used cvs, and yes, it's nice
17:30:13Nico_Ppondlife: ideally, Bagder would set me up with git hosting on
17:30:58pondlifeAre you ready to start using buffering.c in the playback code now?
17:31:06sarixealright, i just downloaded rockbox-bleeding
17:31:12sarixeattempting to make
17:31:19pondlifeNico_P: Or is there some multithreading testing to be done in file copy first?
17:31:46Nico_Ppondlife: I don't really know
17:32:08 Quit Siltaar (Remote closed the connection)
17:32:49 Join fed [0] (i=4a7b3e0c@gateway/web/cgi-irc/
17:32:49dionoeaNico_P: md5s match on target now ? (was it the alignement issue?)
17:32:55pondlifeI'd think the next step is to carry on copying files in a plugin, but with two threads.... one internal running the buffering and another calling the API and writing the output.
17:33:03Nico_Pdionoea: yes, it was an alignment issue
17:33:08dionoeanice :)
17:33:15Nico_Pworks fine on target now, except it fails the LL test
17:33:22dionoeaLL ?
17:33:39dionoealinked list ?
17:33:42Nico_Plinked list... I wrote a small test suite to make sure everything's OK, and a part of it fails on target
17:33:48*pondlife was wondering what that failure was about
17:33:50dionoeaah ok.
17:34:01fedI am using rename () , and the return value was -72. Does anyone have an idea of what this means?
17:34:07pondlifeAh, you said it was a move_handle issue, right?
17:34:20Nico_Pyes. I thought it would be the alignment but no
17:34:37Nico_PI think I'll investigate that
17:34:41pondlifeIf it's on target only, then maybe there's another alignment issue?
17:34:54Nico_Pmaybe yes
17:35:06Nico_Por some other weird issue :)
17:35:42pondlifeDo you need to have appropriate struct member alignment? I've heard rumours about that (but assume gcc would pack)..
17:35:52amiconnNico_P: For target tests, an ipod (or even an archos) would be better. Both will throw an exception (data abort on arm, cpuadrerr on sh) on unaligned accesses
17:36:15Nico_Pamiconn: I only have a gigabeat and an H300
17:36:28Nico_Pmy brother will be back with his H10 on the 25th
17:36:28pondlifeamiconn: That can be enabled on H300, right?
17:36:42amiconnpondlife: If the struct itself is aligned, there should be no problem. gcc aligns struct members, as long as you don't declare a packed struct
17:36:56Crash91how do i remover a patch?
17:37:09dionoeapatch -R < thepatch
17:37:09pondlifepatch -p0 -R < xx.patch
17:37:10Nico_PCrash91: patch -R -p0 < diff
17:37:10amiconnpondlife: Nope. Coldfire doesn't care about alignment and happily accesses unaligned values (jsut slower)
17:37:25Crash91ok thanks
17:37:38amiconnBut perhaps it can be enabled on gigabeat
17:37:42Nico_PCrash91: man patch too ;)
17:37:57Crash91lol...keep forgetting
17:40:00amiconnDoes the gigabeat's arm core include the system control coprocessor?
17:40:59 Join zylche [0] (
17:41:22zylchewhats up with rocklife? just upgraded rockbox there, 06/06/07 −−> 06/07/07
17:41:36fedCan anyone let me know how I access the DEBUGF outputs?
17:41:43sarixebadabing, i actually got it to build this time
17:42:03sarixebleeding source builds fine here
17:42:27sarixepondlife : now what?
17:43:20amiconnAh, looks like it does :)
17:43:34 Quit Caliban (Read error: 110 (Connection timed out))
17:43:45sarixeamiconn : who were you talking to?
17:43:56dionoeazylche: what's up what ?
17:44:05*amiconn answered his own question
17:44:59pondlifesarixe: OK, now work out which patches you need (as few as possible) and attempt to apply them one at a time. Flyspray is the place where you'll find the latest versions.
17:45:16Nico_Pamiconn: is that good or bad ?
17:45:36Crash91i cant get pegbox to build properly :(
17:45:37pondlifeSome may fail to patch as they are out of date. You'll need to pester the author to update or have a hack yourself...
17:45:46amiconnNico_P: It is good. You can enable abort on unaligned accesses
17:45:55Crash91 < i have done everything that says but no joy
17:46:27pondlifesarixe: Also, some patches may fail when combined with others. I suggest you do a make after each successful patch.
17:47:02 Part fed
17:47:32sarixepondlife : just for the record... what do i do after make?
17:47:40 Join Entasis_ [0] (
17:47:40sarixemake zip?
17:48:00amiconnNico_P: In firmware/target/arm/s3c2440/gigabeat-fx/mmu-meg-fx.c, try the following:
17:48:04pondlifeYes, if you need a zip to install on the device.
17:48:28pondlifesarixe: But I meant to rebuild after each patch just to save time if one is broken.
17:48:29amiconnInsert the following line between line 91 and 92:
17:48:41sarixeit'll make for me? or the files to put in it?
17:48:52amiconn"orr r0, r0, #1<<1\n" /* enable alignment fault check */
17:48:59pondlifesarixe: If it doesn't compile, you can patch -R and try again.
17:49:03sarixepondlife : ok, i'll try that
17:49:20saratogafor unix apps that isntall with a Makefile.PL, is there some command to change the install path?
17:49:30pondlifeYes, make zip makes - but you don't need to do that until you have constructed your final build
17:50:02pondlifeAlthough you may want to check you can make zip and install your initial unpatched build first.
17:50:20pondlife(And that it plays music!)
17:50:36Nico_Pamiconn: ok
17:51:41sarixepondlife : the latest versions of these patches all appear to be 20070622
17:52:43sarixeunder "Builds based on SVN repository from 20070622"
17:53:33amiconnNico_P: This should crash with a data abort if you run yesterday's problematic code (with the broken alignment)
17:53:57pondlifesarixe: You'll need to search Flyspray for the patches you want... the webhop one is frozen in time
17:54:04Nico_Pamiconn: I've just build a logf build with this change
17:54:09amiconnThe displayed address should match the function in question
17:54:12Nico_PI expect it to crash in the LL test
17:54:46pondlifesarixe: IMHO, with Rockbox, progress is so fast there's little point in worrying about "unofficial" sources...
17:55:03Nico_Phmm, no crash
17:55:13sarixeyou think these patches will be incorporated soon, then?
17:55:19pondlifesarixe: Unlikely
17:55:29pondlifeBut Flyspray has the latest versions
17:55:56sarixewould there just be similar ones put in? (stable ones that are permanent)
17:56:30pondlifeDepends... The WPS ones probably won't be as there is a plan for viewports that will make them redundant
17:56:45sarixewhat's that?
17:57:15pondlifeSearch the wiki... :)
17:57:41pondlifeIt's a can of worms :)
17:58:15pondlifeAnyway, best you pick which patches you want and attempt to grab them from Flyspray, I'd say.
17:58:51pondlifee.g. for multifont
17:59:02Nico_Pamiconn: still no crash, even with the alignment check comment out
17:59:20pondlifeNico_P: Even if you remove the +3)&~3 ?
17:59:45Nico_Ppondlife: that's what I commented out
18:00:03pondlifeSo bad files came out...?
18:00:35Crash91how do i change the coordinates of the USB logo?
18:00:41Nico_Ppondlife: yes
18:01:29pondlifeMaybe something else is resetting the alignment fault bit?
18:03:05sarixepondlife : rockbox boots and plays music
18:03:26pondlifeSo you can definitely build from scratch!
18:03:43sarixei don't know why it failed before
18:03:46pondlifeBut the challenge is in getting a set of synced patches and source.
18:03:47 Join ^BeN^ [0] (n=Paprica@rockbox/developer/paprica)
18:04:01Nico_Psarixe you can svn up to any rev you want
18:04:05Nico_Psvn up -rN
18:04:17sarixereplace N with the date?
18:04:24Crash91Guys, does anyone know where to set the coordinates at which the usb logo/rockbox logo is displayed??
18:04:26Nico_Pthe rev or mayb the date
18:04:32Nico_Ptype svn help up
18:04:33pondlifeDate isn't accurate enough really.
18:04:33 Nick ^BeN^ is now known as Paprica (n=Paprica@rockbox/developer/paprica)
18:05:19 Join Rori [0] (
18:05:27pondlifeHi Paprica
18:05:47Papricasomeone remmber me? =\
18:05:54RoriHi guys. Does anyone know if I can use this on my 1st gen Nano with Rockbox? Or would I have to boot to the Apple firmware?
18:06:01pondlifeYes, now you've decloaked :)
18:06:03 Part zzuh ("brb")
18:06:31Papricalol what is decloaked
18:07:25dionoeahi Paprica . Did you see my comment on the Spider solitaire entry in FS ?
18:07:50zylchehmm.. is rockipedia still getting worked on?
18:08:04zylcheAnd is there a plugin to do it for an ipod video 5g?
18:08:39Papricadionoea no =\
18:08:52dionoeaok :)
18:09:32pondlifePaprica: I didn't recognise ^BeN^...
18:09:56Papricahihi :)
18:10:21pondlifeOK, I know I've asked before, but would a 4GB iPod Nano in black or white be a 1st gen for certain...?
18:10:40pondlifeFor example, this one:
18:11:01sarixeis there a 2nd gen ipod nano?
18:11:08Crash91the 1st gen has metal back
18:11:19pondlifesarixe: There is, and Rockbox won't work on it.
18:11:20Nico_Pamiconn: my gigabeat refueses to crash
18:11:22Crash91the 2nd gen has curved edges and is all plactic
18:11:26pondlifeNico_P: :)
18:11:55Crash91pondlife: Get_a_Sansa
18:12:16Crash91pondlife:the link you set has 1st gens
18:12:23pondlifeCrash91: Is Sansa smaller than a Nano?
18:12:37Crash91pondlife:no but 500% more features
18:12:42pondlifeThis is for mrs pondlife, she wants no HD and tiny
18:12:53Crash91sansa = flash
18:12:58pondlifeFeatures not important, size is.
18:13:22Crash91hoold on
18:13:23Rori"2006-12-17: Recording on iPod 4G/5G/Nano and H10" How do you record on a Nano??? Also still need an answer on Nano accessories.
18:13:30pondlifeI don't think I can justify the expense of a nano really - maybe just get a tiny unbranded one.
18:13:50Crash91 < the nano in the pic is 2nd gen, 1st get is somewhat thicker (2mm?)
18:14:25Crash91sansa = 1cm thick, nano = .5cm thick
18:15:03Crash91IPod Nano — Dimensions: 90x40x6.9 (first generation) 90x40x6.5 (second generation)
18:15:03pondlife is cheap... but no Rockbox.
18:15:10sarixepondlife : i think you're right about the patches... none of them seem to work
18:15:17Crash91sansa = 4.4 cm wide x 8.9 cm long x 1.3 cm high
18:15:42Crash91pondlife: fyi sansas are cheaper than nanos while having more features
18:15:50sarixeeither way, is there a way i can have svn automatically update the source tree, and have an automatic build process, like the daily build server?
18:16:23pondlifesarixe: "svn up" will update your source
18:16:28 Quit Entasis (Connection timed out)
18:17:37Crash91is there something like task schedulaer in linux? then you could make it automatically execute a shell script
18:17:48pondlifeat ?
18:18:04 Quit Thundercloud (Remote closed the connection)
18:18:15sarixeaye, cron
18:18:19sarixei should have though of that
18:19:19 Part Crash91
18:21:12 Quit obo ("KVIrc 3.2.6 Anomalies")
18:21:32 Join Crash91 [0] (n=crash@
18:21:44zylcheIs Rockipedia a deserted idea then?
18:21:53Crash91pondlife: i recommend you get a creative MuVo
18:21:59 Part Crash91
18:23:45Nico_Pmemmove seems to e causing me problems
18:24:49sarixealright... anyone wanna give me a quick howto on getting source from the svn?
18:25:03sarixelike, how to login and then what to do from there...?
18:25:27dionoeayou don't need to login for svn
18:25:31dionoeato get the source
18:25:45sarixeso... what DO i do?
18:25:56sarixethanks ANY dock accessories for iPod work in Rockbox?
18:32:42PaulJamRori: i think there is a wikipage. IpodAccesories or something like that.
18:33:15 Part maffe
18:33:38Rorik thx
18:33:54 Join maffe [0] (
18:35:06 Join webguest17 [0] (i=1895b9de@gateway/web/cgi-irc/
18:37:44Rorithis is more useful
18:38:15 Quit desowin ("use linux")
18:43:59webguest17hi, i have a GBC rom saved to my ipod but when i go to open it in rockbox files it wont open it goes into a white screen then flashes unknown ROM size 61d54 and then pluggin error , help please?
18:45:06 Join Domonoky_ [0] (
18:46:53 Join Rincewind [0] (
18:52:41RoriI don't understand why finding out the iPod universal dock serial protocol etc is so hard. There are MANY dock accessories and those companies don't seem to have a problem obtaining the documentation. Unless you have to pay for it I can't see why someone could not just ask for it from either Apple or one of those companies that make the addons
18:52:47 Quit webguest17 ("CGI:IRC (EOF)")
18:55:28 Quit foleyjd ("( :: NoNameScript 4.02 :: )")
18:55:35PaulJami guess these companies have to sign some kind of NDA to get them.
18:56:14pondlife any use?
18:57:01saratogaRori: the dock protocol i swell understood
18:58:38 Join petur [0] (n=petur@rockbox/developer/petur)
19:01:25PaulJamhey petur, thanks for fixing FS7388 and 7359
19:01:33 Quit Domonoky_ (Read error: 104 (Connection reset by peer))
19:01:49sarixesomeone have quick help for this line? :
19:01:51sarixecp rockbox-${date +%Y%m%d}.zip
19:01:52peturthanks for spotting
19:02:21sarixei'm trying to copy the output of make zip to a file with the date in the name
19:02:49peturPaulJam: your bugreports are nice, you almost give the solution. Sometimes I wonder why you didn't attach the patch ;)
19:03:14PaulJampetur: no realname :p
19:03:25peturah right.. pitty
19:03:33peturyou deserve credit ;)
19:03:59 Part maffe
19:04:11 Join maffe [0] (
19:05:14sarixenvm, i got it
19:05:17 Quit sarixe ("Peace")
19:05:38***Saving seen data "./dancer.seen"
19:06:28 Join Arathis_ [0] (
19:06:45 Join pixelma [0] (i=pixelma@rockbox/staff/pixelma)
19:07:45 Quit Arathis (Nick collision from services.)
19:07:49 Nick Arathis_ is now known as Arathis (
19:08:41 Quit My_Sic (Read error: 110 (Connection timed out))
19:15:24Rori"<saratoga> Rori: the dock protocol i swell understood" I meant in getting accessories to work over the serial interface
19:17:17RoriThere are so many addons for the iPod I am amazed that this has not been added yet. Guess there are not that many Rockbox iPod users who need them as I thought. Maybe it's just the Non-Rockbox crowd who usually buy that stuff.
19:17:36FebsGodEater: are you around?
19:17:52 Part maffe
19:18:13 Join maffe [0] (
19:18:16 Quit maffe (Remote closed the connection)
19:18:33 Join maffe [0] (
19:19:41saratogaRori: the interface was documented ages ago
19:19:53saratogaits not supported because no one has figured out how the PP serial interface works
19:20:25Rorias I stated
19:20:31Roriwould need docs
19:20:38Rorimaybe NDA would screw that up
19:20:57 Join Arathis_ [0] (
19:20:58 Quit Nico_P (Remote closed the connection)
19:21:04saratogaRori: what companies produce is something else
19:21:06 Quit Arathis_ (Read error: 104 (Connection reset by peer))
19:21:13saratogathey don't need this information to make an ipod accessory
19:21:28saratogasince apple provides the firmware
19:21:42Roriso they don't even need to know
19:21:45Roriwhat goes on inside
19:21:54Rorijust instructions to interface with the apple fw
19:21:57Roriyeah makes sense
19:22:29saratogatheres some simple protocol with just a few commands
19:22:34saratogalist files, play file, etc
19:22:34Roriso they can still keep the serial PP to themselves...hmmm how annoying
19:22:40 Join obo [0] (
19:23:03Rorikinda like Nano 2G has an encryption?
19:23:18saratogawhat do you mean?
19:23:38RoriI heard they could not crack 2G Nano or something which is why it's not got a Rockbox
19:23:42 Join Caliban [0] (
19:23:56Doomedi dont think its encryption though
19:24:01RoriI don't understand it technically
19:24:57RoriPerhaps it's interfacing with the hardware that is the problem
19:25:00AveI thought that it was bc nano 2g has different internals hw-wise, like not portalplayer based anymore
19:25:12Roriyeah diff chipset
19:25:17RoriI read that too
19:25:41Rorioh well. I want to know how good 2G's gapless playback is
19:25:56Avewow, it has gapless playback... but does it play vorbis?
19:25:58RoriI may buy an 8GB one if it's good enough
19:26:00Ave.. thought so
19:26:02saratogathe 2G does encrypt the firmware and it is based on different hardware
19:26:05Rorinot bothered about Ogg etc
19:26:11saratogai just didn't understand how that was like the serial port
19:26:16RoriJust want a player that works fully and has gapless
19:26:32RoriMP3 and AAC are just fine for me
19:26:48Rorialthough working in iTunes sucks a bit
19:26:54Aveone thing that annoys me about vanilla apple is the need of itunes
19:26:56RoriI prefer easy folder and filenames
19:27:04Aveor specialized obfuscated uploader or whatnot
19:27:23Roriwell I could put up with that. I don't change my music on it that often anyhow
19:27:51Roriso has anyone any experience with 2G Nano's gapless playback?
19:27:51AveI think there are some linux tools so you can upload to plain ipod, cant buy songs tho
19:28:05saratogathis isn't the right place to discuss the apple firmware
19:28:17Roriwill check ilounge
19:29:00Aveoh yeah, keeping in mind that gapless is kind of impossible with eg mp3 encoded files unless youre lucky
19:29:21saratogaAve: mp3 works fine with gapless
19:29:37saratogarockbox, itunes, etc will decode it gaplessly
19:29:45Aveyou need more oompah in the gapless engine detecting when there's silence or end
19:29:59RoriiTunes perhaps but it's the player I am interested in
19:30:02Avewith vorbis you KNOW when the song ends precisely
19:30:06RoriRockbox does a good job of gapless
19:30:25AveI also like that rb has dithering
19:30:47Doomedmp3 gapless fine
19:31:12saratogaAve: gapless mp3 as implemented in rockbox, itunes, etc does not detect where silence is
19:31:29saratogaits actual gapless decoding, same as ogg or mpc
19:31:34Aveoh ok
19:31:46Avethere has been some advancements
19:32:08Doomedoops sorry, didnt write a complete sentence, =\
19:33:38RoriDoomed are you saying 2G Nano has decent gapless decoding with iTunes 7?
19:34:10Doomedno, i was wanted to say that mp3s play fine gaplessly on rockbox
19:34:35Doomedmy friend has an 8gb nano 2g, but i dont remeber offhand how the gapless was
19:35:44Rorimight be worth finding out if he will let you play with it :)
19:36:21GodEater_iTunes's gapless implementation is "ok". I don't think it's as good as Rockbox's - but it's a vast improvement over "no gapless at all" in previous iTunes.
19:37:49 Join Wiwie [0] (
19:38:11RoriHmm you can buy refurbed ipods from apple's online store. Wonder what the warranty is though
19:38:20 Quit Arathis (Read error: 110 (Connection timed out))
19:38:26Rorisame 1 year hmm
19:39:08Avehows the price?
19:39:08Rori135 for a 8gb nano as opposed to 169 (UK ripoff prices again)
19:39:34Rorihave to see if I can find cheaper on google product search lets check
19:39:59AveI wouldnt mind a 8G 1st gen nano if such exists
19:41:18 Join thegeek [0] (
19:42:14 Join ompaul [0] (n=ompaul@freenode/staff/gnewsense.ompaul)
19:42:43GodEater_Ave: it doesn't
19:42:43Roriyou would need to mod the nand flash memory
19:42:47Rorichange it out
19:43:03Rori128.79 is cheapest I can find 8GB 2G Nano
19:43:05Averats, so the 4gig is max
19:43:27AveI currently have 2G one which is fine and all, but at times it gets a little tight on space..
19:43:33Roriex postage and no idea if that includes vat it wants me to register to checkout (Hates those online shops)
19:43:42 Quit chrisjs169|afk (Read error: 110 (Connection timed out))
19:44:26Roriah vat included
19:44:38Roridepending on shipping that would not be a bad deal
19:44:54Rorinew rather than a refurb
19:45:28Avehave you considered anything other than nano? or apple
19:45:33Roriheh. nah I don't need to right now. Maybe later.
19:45:40GodEater_Febs: got your message - will make sure we leave a message for you if we actually work out where to meet up
19:45:43RoriReason I would stick with Apple is for the accessories
19:45:47Avemy experience with other brands has not been all that good
19:45:57Galoisthe accessories aren't usually supported in rockbox anyway
19:46:01Rorinobody has as many addons
19:46:06RoriI know
19:46:13Roriwhich is why I was thinking of a 2G for gapless
19:46:21pondlifeRob2222: Fix for your WPS filename available here -
19:46:24GaloisSansa is 8GB and works pretty well even without accessories
19:46:31GodEater_Rori: Why not a 5.5G ?
19:46:39GodEater_then you could run Rockbox on it too
19:46:53Avethe sans build quality was little flimsy, compared to apple
19:47:12FebsGodEater: thanks!
19:47:16Rori1. I want gapless playback (Rockbox delivers). 2. I want accessories to work as per hardware spec (Not on Rockbox for many addons). 3. more memory (2G has 8GB)
19:47:21Galoisat least the problems are solvable, unlike the nano 2g
19:47:29GodEater_Febs: otherwise I think we'll turn up at the hotel front desk and go from there :)
19:47:30Rori4. I want small.
19:48:07 Join docgnome [0] (
19:48:25Rori5. I don't want to wait years for Rockbox to support accessories fully. If I could do it myself I would heh. I ain't complaining though since Rockbox is free ;)
19:48:27docgnomedoes anyone know anything about mounting an Archos Jukebox under OS X 10.4?
19:48:58GodEater_docgnome: I'm not entirely sure it's possible
19:49:03GodEater_but if it is - amiconn will know
19:49:04docgnomeit's a Studio 20 running rockbox.
19:49:07RoriPlayers like Sansa etc are nice and all but I do like my addons :)
19:49:11amiconnmarkun: around?
19:49:18docgnomeGodEater_: what makes you say that?
19:49:30*pondlife wanders off, wondering if anyone would like to look at the one-line mod attached to FS
19:49:39GodEater_as I understand it - the old Archoses don't implement USB mass storage
19:49:48GodEater_though I'm ready to be corrected if I'm wrong
19:49:51amiconndocgnome: The archos Player/Studio has an USB chip that does not use the usb msd standard
19:50:21amiconnYou need a specific ISD200 driver; don't know whether that's available for OS X
19:50:39docgnomeamiconn: figures. Strange though that I can mount it just fine under Linux.
19:51:02GodEater_docgnome: this is because there's a driver under linux :)
19:51:06amiconnLinux has an ISD200 driver, courtesy of Zagor
19:51:42amiconnGodEater: This is correct, or specifically, it applies to the archos Player/Studio (all capacity versions), and the archos recorder v1 (usb1.1 versions, which are the 6, 10 and 15GB models)
19:52:07GodEater_is the list of archoses for which it isn't true shorter ? :)
19:52:12amiconnThe recorder v1 20GB and the fm/v2 recorders are usb2.0 and use the ISD300 which is msd compatible
19:52:20GodEater_hehe - thought so
19:52:28GodEater_I should really try to remember that
19:52:33amiconnAnd the Ondios are also msd compatible (though usb1.1 only)
19:52:41GodEater_ouch - slow!
19:52:47amiconnNah, it's flash
19:53:07amiconnAnd the bridge is much slower than what usb1.1 at max would be
19:53:39amiconnUpdating rockbox on my archos Studio 20 is several times (read: more than 10x) faster than on Ondio
19:53:42docgnomeIt seems that there is a driver for OS X 10.1 that came on the CD that came with the Jukebox but it's for 10.1 and it doesn't seem to do... anything after I installed it on 10.4
19:54:03GodEater_I thought USB1.1 was something like 11Mbps, while USB2.0 was up around 400 ?
19:54:37amiconnActually this doesn't have to do with USB versions; it's a common misbelief
19:55:06GodEater_maybe my old Rio600 is the culprit for me believing this
19:55:16GodEater_it was SO slow to load anything on to
19:55:20amiconnUSB1.1 defines a protocol, and 2 transfer speeds: low speed (1.5Mbit/s) and full speed (12 Mbit/s)
19:55:48amiconnUSB2.0 extends the protocol, *and* adds a 3rd transfer speed: high speed (480Mbit/s)
19:56:09RoriHmm from what I am reading 2G Nano's gapless playback only works if the album is in the iTunes database. If not then you are screwed.
19:56:20amiconnBut: an USB2.0 compatible device does not *require* high speed to be supported
19:56:43GodEater_amiconn: I see what you're saying - I was aware USB 2.0 *could* be slower
19:56:52GodEater_I've just not come across many devices which are
19:57:06amiconnBut even if the bridge implements high speed, the 480Mbit/s is a theoretical limit
19:57:08GodEater_I *didn't* know there were two speeds for 1.1 though
19:57:27GodEater_amiconn: Much like my 22Mbit/s broadband :)
19:57:43*GodEater_ checks his router. Yep. Only 8Mb/s
19:57:43RoriI should go into an Apple store with a USB drive with some CD mix tracks on it and get them to transfer them onto a player and test it.
19:57:47amiconnIt would mean that fast hdd connected via USB2.0 could yield 60MByte/s, but in practice you'll see something like 30
19:58:25GodEater_Rori: Do you mean iTunes Store Database ?
19:58:38amiconnGodEater: Low speed is often used for HIDs (mice, keyboards). More than enough for them, and allows cheaper designs
19:58:40GodEater_Rori: because it it's not in the iTunes player database, then it won't play at all ;)
19:58:55GodEater_amiconn: very good point - I'd just not thought about it I guess
19:59:02GaloisI'm running my main hard drive here off USB1.1 (full speed, but still)
19:59:07GodEater_it's not often you worry about the transfer speed of your mouse
19:59:31GodEater_Rori: you said you can only get gapless if it's in the iTunes database
19:59:35amiconnGamers might ;)
19:59:38 Quit Rincewind (Read error: 113 (No route to host))
19:59:42Roriif you add MP3's of your own to iTunes it won't play them because it's not in their database?
19:59:53GodEater_amiconn: I met Thresh (of Quake / Doom) fame once - he was obsessive about his mouse
20:00:18GodEater_Rori: when you add anything to your local iTunes - it goes into the database
20:00:36GodEater_Rori: so I'm not sure what you're saying won't play back gaplessly
20:00:56GodEater_actually, I met him twice, once in Stockholm, and once in London
20:00:59PaulJamRori: afaik itunes reads the lame header when you impot mp3s (and plays them gaplessly)
20:01:20RoriBut what I read on iLounge was that if the album does not contain the info (IE you have a bunch of MP3's from an album not listed on iTunes) then the gapless feature would not work?
20:01:25 Join TTSbox [0] (
20:01:35amiconnGodEater: Perhaps he'd buy an USB2.0 high-speed capable mouse ;)
20:01:37Rorimaybe I should install iTunes 7 and test
20:01:43GodEater_Rori: again - what do you mean by "Not listed in iTunes" ?
20:01:46GodEater_do you mean the online store ?
20:01:52GodEater_or do you mean the program on your computer
20:02:15RoriI mean when sending the MP3's to the player for gapless playback
20:02:21GodEater_Rori: well that's a complete fabrication - NONE of my music is on the iTunes store - and my iPod managed to play it back gaplessly just fine
20:02:31RoriNot sure how iTunes and the player handles it
20:02:43GodEater_iTunes DOES have to preprocess it all before it goes onto the player though
20:03:04GodEater_the preprocessing can take a LONG time if you have a lot of music
20:03:16GodEater_I have about 6000 tunes, and it took about 4 hours
20:03:32GodEater_Rockbox does it on the fly :)
20:03:54Roriwhat the hell is it doing? converting to aac?
20:04:02GodEater_No - it scans for gaps
20:04:13GodEater_and records information about it in the iTunesDB on the player
20:04:30Rorihow does it know if an album is meant to be gapless or nor?
20:04:31GodEater_so the local playing mechanism on the iPod doesn't have to do the same trick
20:04:37GodEater_Rori: it doesn't
20:04:57GodEater_although you can tag them as such in iTunes
20:05:09GodEater_not that I ever found that made the slightest bit of difference
20:05:10RoriI'd rather tag
20:05:46GodEater_anyway - I think that's more than enough talk about iTunes
20:05:55RoriFraunhoffer we love you for the MP3 codec but ffs why didn't you include a gapless tagging feature? :P
20:05:58amiconnWhat the hell does itunes need to scan?
20:06:10GodEater_every track in your database
20:06:19GodEater_from beginning to end as far as I could make out
20:06:35amiconnGapless playback just means that the playback engine doesn't add null samples between 2 adjacent tracks
20:07:05RoriI should 'upgrade' all my older MP3's anyhow
20:07:24Galoisupgrade to ogg YEAH
20:07:32amiconnTrimming the tracks is not gapless playback by definition, it's gap removal (or track mangling, as I would call it - brr)
20:07:36Roritoo lazy to go ripping again (Yeah right) ;)
20:08:02GodEater_amiconn: I don't believe it actually alters the tracks (although I could be wrong)
20:08:11GodEater_I think it writes meta data to the iPod database
20:08:18amiconnSo why does it need to scan them?
20:08:21GodEater_to indicate where to stop playing one track, and when to play the next
20:08:33amiconnHmm, and that takes 4 hours for just 6000 tracks?
20:08:43GodEater_yes - it's not fantastic
20:08:49GodEater_which is why I prefer rockbox
20:08:52amiconnRockbox can do that within minutes - on an 11MHz CPU (!!)
20:08:58GodEater_indeed ;)
20:09:05docgnomeHrm... I guess the only real solution would be to have a Linux virtual machine and mount the jukebox there and drag and drop the files into it
20:09:14amiconnadd *even* where you see fit
20:09:33GodEater_Steve Job's monkeys have a lot to learn from Rockbox
20:09:54docgnomewhich is better than having to throw everything on a pen drive and go downstairs to load it... but still annoying.
20:10:02GodEater_I'm still dumbfounded by how long the apple firmware takes to boot
20:10:11GodEater_it seems about 10x rockbox's loading speed
20:10:21Rorithat's to load all the drm
20:10:25GodEater_10x as long I mean
20:11:00amiconnI don't know whether the boot speed depends on whether the itunesdb is populated, but to mee the apple boot speed seems quite fast compared to the iriver OF (with its db enabled)
20:11:04 Quit docgnome ()
20:11:26Rorido you think apple will ever add bluetooth to their players (Forgetting iPhone for now)?
20:11:27Doomedoh the irvier db load takes soooo long
20:11:39GodEater_amiconn: ah yes - I remember that too
20:12:03amiconnI never used it for real myself (got my 2 irivers mainly for rockbox hacking)
20:12:15GodEater_I can't actually recall the last time I ever used the OF on the iriver though
20:12:25amiconnWith rockbox, the H140 (now H180) has become one of my main daps
20:12:41Doomedi used it for about a week and i couldnt read the titles cause they had a stupid graphic on the left side so, i remembered rockbox and put it on
20:12:42GodEater_amiconn: it's what I use in the car
20:12:48RoriI have an H140 sitting on the side
20:12:54Roriyou put an 80GB drive in?
20:12:56Doomedthat was about when db was getting commited
20:13:15amiconnIn car and at home. For when I need a small unit, I use the Ondio FM most of the time
20:13:19GodEater_heh - I had my H140 before RB was ever ported to it
20:13:21Doomedor should i say tagcache
20:13:29The-CompilerStrange problem here: I've reformatted my ipod 5.5G with the mtools (because of the bug in mkfs.vfat), now I get these outputs: But I can only copy 8GB of datas, then I get "no space left on device"...
20:13:30GodEater_I followed LinusN's progress with it closely :)
20:13:46RoriI can't think of a use for the H140 these days. Too much of a brick
20:13:53Doomedi had it i ithink december 04?
20:14:06GodEater_Rori: that's tragic - it's one of the best RB targets there is
20:14:18*amiconn agrees with GodEater
20:14:32Doomedi broke on of my h120 boards so i bought another h120 to just have cause i love it
20:14:43amiconnOne big advantage of the H1x0 (and the archos recorder) is that they have s/pdif out
20:14:53Doomedi might install rockbox on the flash, amiconn whats that?
20:14:58amiconnI don't want to miss that
20:15:16GodEater_Doomed: removes the iriver firmware forever ;)
20:15:23GodEater_and makes booting super fast
20:15:38Doomedyeah i know i think i want it. i meant whats that as in s/pdif
20:15:40Roriperhaps I should sell the H140 to a person who may find it more usefull
20:15:42GodEater_think you'll need the unreleased V7.0 bootloader though
20:15:54GodEater_it's digital line out
20:16:25Doomedah ok. oh Rori ill take the back of the H140 case off your hands haha
20:16:29RoriI even had iRiver fix the click
20:16:43GodEater_Rori: if you truly don't want it - then how about donating it to a RB dev ?
20:17:03RoriI'll think about it :)
20:17:26GodEater_I suspect some of them would offer to pay for it too if you're not feeling that generous
20:17:32RoriI have loads of hardware I never sell. too lazy
20:18:27Rorigot a 7600GT AGP card in a box. Hardly ever used. Should have ebayed it ages ago lol
20:19:09Roriand an XDA Mini palm mobile
20:19:15Roriand a SE P900
20:19:34Roriunused and boxed
20:20:57 Join chrisjs169|afk [0] (
20:22:10 Join Thundercloud [0] (
20:27:39 Quit ptw419 ()
20:29:46Rorianyhow later
20:29:50 Quit Rori ()
20:31:00 Join toyoyo [0] (
20:31:53toyoyoI need some help
20:32:14toyoyoinstaiiling rockbox on my ipod video
20:32:21toyoyoanybody there?
20:33:50 Join docgnome [0] (
20:34:29saratogatoyoyo: just do what the manual says
20:34:54 Join petur2 [0] (n=petur@rockbox/developer/petur)
20:35:10 Quit petur (Nick collision from services.)
20:35:15 Nick petur2 is now known as petur (n=petur@rockbox/developer/petur)
20:36:16 Part docgnome
20:39:32toyoyoI tried doing that but I get an error
20:40:17toyoyowhen I press"i" at the prompt for the ipodpatcher,
20:40:35toyoyoIt says "Error reading from disk: The request could not be performed because of an I/O device error"
20:40:40toyoyoand exits
20:41:28toyoyoAny idea what to do?
20:42:42*amiconn summons Nico_P
20:43:08*amiconn pings markun again
20:43:42 Join piroko [0] (
20:43:52 Join My_Sic [0] (
20:44:14amiconntoyoyo: What os are you running ipodpatcher on, and do you use the latest ipodpatcher?
20:44:15toyoyoCan someone help me? :(
20:44:31toyoyoI'm running Windows XP Home
20:44:44toyoyoand yeah i'm running the latest ipodpatcher
20:45:08amiconnNote that you need administrative right in order to use ipodpatcher
20:45:18toyoyoYup.. I have those
20:46:06toyoyoWould it be easier if I booted into linux and tried?
20:46:16 Quit piroko (Client Quit)
20:46:30amiconnIn theory it shouldn't matter
20:46:51amiconnlinuxstb is the ipodpatcher expert (but isn't around atm)
20:46:54toyoyoI know.. but no matter what I try, it still shows the same error
20:49:06 Join piroko [0] (
20:49:15toyoyoI even tried closing the iTunesHelper and ipodservice processes, in case they're preventing ipodpatcher from doing I/O
20:49:50*amiconn wonders whether XP home might be the problem, but doubts it
20:50:53toyoyoI knew XP Home wasn't the right choice but I had no choice :P
20:51:22piroko"XP" looks like someone sticking his tongue out in disgust
20:51:23amiconnIirc there is an option in itunes which enables or disables direct disk access
20:51:32amiconnI can't tell for sure; never used itunes
20:51:35pirokoJust "enable disk use"
20:52:00toyoyoOh.. let me try
20:53:43toyoyo"Enable disk use" is checked all right
20:54:44toyoyoBut still it says "Write failed" when I use ipodpatcher
21:00:45saratogacan you add a text file or whatever toyour ipod's disk
21:01:14toyoyoyeah, sure
21:02:06toyoyoHey, one thing...
21:02:38 Quit Thundercloud (Remote closed the connection)
21:02:43toyoyowhen I double click on the iPod icon in my computer... it shows some error that "copy.exe" was not found
21:03:00toyoyoand I have to right click and click on open to see the files
21:03:05 Join atsea- [0] (i=ariel@gateway/tor/x-f63c14b0e22d4ace)
21:03:12toyoyoDoes this have anything to do with my problem?
21:03:35saratogaprobably not
21:03:56saratogaassuming you can write arbitrary files to the ipod, i don't know why this wouldn't work
21:03:59 Join Robin0800 [0] (
21:04:18saratogahave you done anything funny to the ipod like restored it with a 3rd party tool or installed some other hack?
21:04:35toyoyoNothing of thst sort
21:04:55toyoyoThis is the first extra I ever tried to install
21:05:40***Saving seen data "./dancer.seen"
21:05:42Robin0800Soap,Why is the Ipod running at 78 mhz?
21:06:06SoapI'm an odd person to ask
21:07:05SoapSVN r 13767
21:07:19toyoyoJust so you know.. I was able to copy the .rockbox folder to the iPod root folder without any problems
21:07:27Soap3rd of July @ 00:42
21:07:29Robin0800Soap,according to this forum Topic: playback crashes on 5g (60GB) ipod it makes it crash
21:07:47 Quit Kingstone ()
21:07:54SoapI just read that forum thread 30 seconds ago.
21:08:24SoapNowhere in that thread is proof that the commit which changed clock scaling also is the cause of 5th gen crashes.
21:08:44toyoyoGuess I have to try it out on Linux then
21:08:56 Join desowin [0] (
21:09:30Robin0800Soap,I can confirm it crashes Ive been trying to find out why for two days!
21:09:38desowinobo: congrats on getting commit access!
21:09:48Soapwell, test with 13767 and 13766
21:10:05Soapif 13767 crashes and 13766 it would be nice to know.
21:10:06obodesowin: thanks :)
21:10:20Soap*and 13766 doesn't
21:11:12Robin0800Soap,my ipod has never crashed in abut the last six months I think it must be this
21:11:20FebsSoap, it crashes with r13764.
21:11:33SoapRobin0800: see
21:11:37FebsThough crashes with r13767 seem to be more frequent.
21:12:36 Join AceNik_ [0] (n=AceNik@
21:13:33AceNik_hey guys i broke my headphones pin , i opened it to solder it also , i have lesser volume in one side now , can you please help
21:13:37Soapok, Robin0800, then test 13763 vs 13762
21:14:03Soap(though I am behind in my log reading - has 13763 already been confirmed as the source of the crashes?)
21:14:39Robin0800Soap,Can you get older builds on svn
21:16:26bluebrotherRobin0800: svn co -r <revisionnumber>
21:16:47FebsSoap, I have not been able to replicate the crashes on r13661, but I have on r13764. It seems unlikely that the changes to reversi cause the problem, so r13763 seems to be the likely culprit.
21:17:49Soapipod video Robin0800 ?
21:17:58 Quit My_Sic (Read error: 110 (Connection timed out))
21:18:20Robin0800Soap,yes 60gb
21:18:39Soapyou be around in a bit?
21:19:04amiconnI wonder why it *only* crashes on the video, and not on any other PP5022 target (nano or mini G2)
21:19:22amiconnAre crashes reported for G5 32MB models?
21:19:44Soapvideo is 5021 treated as 5020
21:19:59Robin0800amiconn,perhaps its the extra 3mhz
21:19:59Soapso is nano IIRC
21:20:06 Quit AceNik_ ("bye guys have fun , enjoy !!!!!")
21:20:12pirokoIt's ironic that a fix to stabilize the older generation ipods breaks other ones :P
21:20:16SoapRobin0800: crashes start _before_ that commit.
21:20:24amiconnNo, video and nano are labeled 5021 but report as 5022 to software and are handled as such
21:21:22amiconnThe OF does exactly the same, I disassembled video, mini and color/photo firmware and compared the clock switching routines
21:21:49amiconnAnd I'm quite sure it's not the extra 3MHz. The 5022 is specced up to 100MHz
21:22:21Robin0800Soap,It was either tuesday or wednesday night watever builds those were?
21:22:45Soap<Febs> Soap, it crashes with r13764.
21:22:53 Quit petur ("switch")
21:23:11Soapwhich is prior to the max frequency change, but post amiconn's first PP rework.
21:23:28amiconnThe video firmware checks for 3 PP types in that routine though: PP5020, 5022 and 5026 (!)
21:23:38 Join petur [0] (n=petur@rockbox/developer/petur)
21:24:17amiconnFebs: What PP version is reported for your G5 in the debug menu?
21:27:54 Join crop [0] (i=53f23e69@gateway/web/cgi-irc/
21:28:58cropIf I copy/move/etc files in USB mode, how are the ops made? Via the RB's fat driver or through the host PC's one?
21:29:10Soapis there any point in me changing config-ipodvideo.h back to "5020" from "5022" and seeing if the crashing is still there?
21:30:09amiconnThe 5020 and 5022 *do* differ in how clock setup is done
21:30:17amiconnANd the 5020 handling was also changed
21:30:18bluebrothercrop: usb mode is handled in hardware make it not crash anymore
21:30:36bluebrotherexcept for the PP targets, which use apples disk mode
21:30:46Robin0800Soap,the debug h/w sayes 76 Mhz clock cpu and audio thead say 78Mhz
21:30:57bluebrotherso, for usb there is no part of Rockbox involved. Rockbox only enables usb mode.
21:31:09amiconnThe HW info one is an estimation (hence "est.")
21:31:31amiconnThis was added by me to verify whether the setup is correct
21:31:37cropbluebrother: could you elaborate? Who's moving bytes and bits on the HD then? I.e. was the recent fix by petur relevant if I only managed files in USB mode?
21:32:05bluebrothercrop: well, most players have a dedicated usb-ata brigde chip
21:32:11cropbluebrother: ah, you type rather fast! :-)
21:32:42bluebrotherif Rockbox detects a usb connection it enables this chip. As only one device can access the disk Rockbox can't access the disk that time
21:32:49peturwe do not have a software usb stack yet, so our fat code is never involved during usb
21:33:17bluebrotherand for PP targets it's quite similar, the major difference is that it's apples code that does usb
21:33:36bluebrotherbesides, if the devices comes up as external hard disk on the pc, the pc will do the filesystem handling
21:33:48bluebrotheras an external hard disk is ... just a disk ;-)
21:33:49amiconnAnd it won't even be involved if we have an usb stack one day
21:33:59croppetur: my HD is still sane then :-)
21:34:19amiconnusb msd works at the block level, it's always the PC filesystem which is responsible for messing up ;)
21:34:32Soapamiconn: I wasn't trying to second-guess your decision to treat the 5021 as a 5022.
21:35:00amiconnSoap: It's not my decision... the 5021 is actually a 5022
21:35:18amiconnEither it's only a marketing thing, or it's somewhat reduced in specs
21:35:19peturindeed out fat code will never be involved... silly me
21:35:29 Quit Febs ("switching computers ...")
21:35:35amiconnBut certainly not below the 5020 specs
21:35:43 Join Febs [0] (
21:36:01 Part maffe
21:36:09 Join maffe [0] (
21:36:24Soapthe crashes appear to start after rockbox started treating the 5021 as a 5022.
21:36:58Soap(and everything else which got changed @ that time)
21:37:38amiconn5020 and 5022 are treated exactly the same everywhere in rockbox, except for clock setup
21:44:39 Quit crop ("CGI:IRC (EOF)")
21:46:11 Join Bagder [0] (n=daniel@rockbox/developer/bagder)
21:48:34 Quit bdgraue (Remote closed the connection)
21:49:13 Join bdgraue [0] (
21:49:20Robin0800Soap,looking at the change logs it looks like the 2nd and 3rd of july are where the problems started
21:50:39Soapaye, day is known, and specific commit is strongly suspected.
21:50:41 Quit bdgraue (Remote closed the connection)
21:53:33amiconnSeems I found something...
21:56:27 Join bdgraue [0] (
21:58:43 Quit bdgraue (Client Quit)
21:59:18petur... very long drumroll ...
21:59:39XavierGramiconn: we are waiting for your discovery :P
21:59:55Bagderyes, and we're turning blue now from holding our breath
22:00:53amiconnThe G5 OF does handle the PP5022 a little different than the mini firmweare
22:01:20amiconnThere's a small udelay() *before* checking the 'pll locked' bit
22:01:33*amiconn will make a new patch to test
22:01:47 Join bdgraue [0] (
22:01:47 Quit bdgraue (Remote closed the connection)
22:01:47 Quit piroko (Connection timed out)
22:02:49amiconnHmm, in fact no need to make a full blown patch
22:02:58amiconnFebs, obo: around?
22:03:38amiconnI have a new thing to test: First revert system-pp502x.c to current svn
22:04:15amiconnThen add a new line 192 (before the while(...) line), just:
22:07:44FebsWhere is that file located?
22:08:03 Join bdgraue [0] (
22:08:03 Quit bdgraue (Read error: 104 (Connection reset by peer))
22:09:43FebsCompiling now ...
22:11:12 Join bdgraue [0] (
22:11:14 Quit bdgraue (Remote closed the connection)
22:16:59FebsOK, it's running now. I'll let you know of any issues.
22:18:28Robin0800Febs,what target are you running it on?
22:18:45 Part maffe
22:18:56SoapRobin0800: what behavior leads to crashes?
22:19:04 Join maffe [0] (
22:20:48 Quit The-Compiler (Connection timed out)
22:20:55 Join The-Compiler [0] (
22:21:20 Join Thundercloud [0] (
22:21:52FebsRobin0800: iPod Video 60 GB.
22:22:06Robin0800Soap,during normal playback with shuffle on half hour commute any time 5 mins to 25 mins
22:30:37 Join bdgraue [0] (
22:31:51 Join datsit40ther1 [0] (n=peter@
22:36:34 Join merbanan [0] (n=banan@
22:41:17 Quit petur ("*plop*")
22:41:46datsit40ther1No creative Zen support? Seems not from what I have read. My goal as a 52 year old is to have a larger font on my zen v plus. I also use an iriver h10 which I see IS supported. I will have to check out the h10 stuff and come back here later.
22:42:47 Join acb5649 [0] (i=480ec220@gateway/web/cgi-irc/
22:43:27pondlifeDon't suppose anyone knows any more than the IRC log about Nico_P's memmove() problem?
22:43:59acb5649Can anybody please tell me what kind of harddrive i could use with my H10?
22:44:36pondlifeacb5649: might help...
22:44:47pondlifeor it might not... :)
22:45:28*ender` yawns
22:45:33*acb5649 waits for the H10 radio to work with RockboX.
22:45:51pondlifeWhy wait when you could be doing it..?
22:46:49 Quit The-Compiler (Connection timed out)
22:49:20acb5649I would not know how to code. I am waiting for the head of the H10 port to give me more info.
22:51:44 Quit acb5649 ("CGI:IRC")
22:54:22 Join webguest79 [0] (i=480ec220@gateway/web/cgi-irc/
22:54:50 Quit webguest79 (Client Quit)
22:56:35*amiconn wonders whether there are different revisions of the PP5022C
22:56:56pondlifeHave you found any way to make it spew out an ID string?
22:57:49 Quit SirFunk (Read error: 113 (No route to host))
22:58:25pondlifeHmm, a build should have been triggered by now...
22:59:14PaulJamheh, i just wanted to ask if something is wrong with the buildserver.
22:59:28Bagderpondlife: ah, that reminds me...
22:59:33pondlifeIt's gone to the pub along with most of us...
22:59:47pondlifeBagder: ..?
22:59:55Bagderthere's a little prob
23:00:06Bagderwith multiple scripts doing 'svn up'
23:00:18Bagderso the build mater sometimes doesn't "catch" the change
23:01:28pondlifeShall I commit something silly (a comment change?) to give it another chance?
23:01:37Bagdernah, I poke it manually now
23:01:46Bagderand I think I fixed the actual problem too
23:02:25pondlifeI noticed the diff script on the builds page quite often gets the revision numbers wrong: e.g.
23:02:29pondlifeIs that related?
23:02:43saratogaanyone know why i'm suddenly getting hundreds of "signedness" warnings in arm-elf-gcc?
23:02:48pondlifeOr is that just due to the manual poking?
23:02:55 Quit Thundercloud (Remote closed the connection)
23:02:58saratogaso far i just ignore them, but its a little odd that it started happening so suddenly
23:03:01Bagderin this case it is due to the manual way to provoke a rebuild
23:03:13pondlifesaratoga: In the WMA codec only, or everywhere?
23:03:52Bagdersaratoga: no new -W options?
23:04:13amiconnsaratoga: sim build?
23:04:15 Quit saratoga ("CGI:IRC")
23:04:21 Join saratoga [0] (i=980398fe@gateway/web/cgi-irc/
23:04:30amiconnAh no, you said arm-elf-gcc
23:05:13amiconnTry reconfiguring anyway, especially if you might have changed the target Makefile manually
23:05:28amiconnPerhaps some -W options were lost by accident
23:05:30saratogai haven't changed anything
23:05:37saratogai get it even with SVN builds
23:05:41***Saving seen data "./dancer.seen"
23:05:48saratogaand for all targets I've tried to build
23:05:57amiconnDid you update your arm-elf-gcc?
23:05:58pondlifearm-elf-gcc version?
23:06:00saratogagcc 4.0.3 if it matters
23:06:11amiconnThat's the recommended one... odd...
23:06:32saratogawould updating be an option?
23:07:05amiconn4.1.x is know to produce unstable rockbox binaries for arm
23:07:11saratogaah ok
23:07:16saratogawell its not a big deal
23:07:21saratogai can just ignore it
23:07:27pondlifesaratoga: Not that it'd cause signedness warnings, but do you have "rockbox patch #1" on 4.0.3?
23:07:45saratogapondlife: i used the install script, so i have whatever it has
23:07:52amiconnrockbox patch #1 only matters for SH1
23:07:59saratogaCentos linux (redhat clone)
23:14:15Soapamiconn: 30 minutes with the addition of the udelay on my Nano and no crashes yet.
23:14:35amiconnSoap: Did the nano also crash with current svn?
23:14:56amiconnThis would be the first report for nano....
23:15:22SoapI hadn't seen any reports, but A:it's the same processor as the Video is it not? B:Febs is testing on the video so I thought best to try different.
23:15:51SoapC:perhaps the test isn't that valuable. :()
23:15:55amiconnI didn't manage to crash the G5.5/80 with svn, for hours
23:16:19amiconnSo it's certainly not all of them, that's why I think there might be several revisions of the cpu
23:17:08 Quit datsit40ther1 ()
23:17:46Soaprevisions we'd be able to detect?
23:18:15amiconnnot afaik
23:18:38pondlifeThe OF must have to cope with any differences.
23:18:39 Join My_Sic [0] (
23:18:55pondlifeOr could there be multiple OFs too? Sounds unlikely.
23:19:09amiconnIt looks like the OF just plays safe here
23:19:13SoapRobin0800: are you still here?
23:19:14saratogawould it be poor form to have the guts of a codec call pcmbuf_insert() directly?
23:19:22pondlifesaratoga: Yes
23:19:26saratogai'm kind of tempted to kill the output buffer
23:19:47amiconnSome revisions of the PP5022 might not reset the PLL_LOCKED bit fast enough, so the OF waits a bit, and *then* starts checking the bit
23:20:16amiconnThe PP5022B doesn't seem to suffer from this problem, because the ipod mini (g1/g2) firmware doesn't do this
23:20:42 Join S1gn [0] (
23:22:02Robin0800Soap,still here just!
23:22:06saratogaso i should build up an output buffer and then submit it to pcmbuf_insert() all at once?
23:22:40pondlifeWooh, small memory corruption problem spotted...?
23:23:57SoapRobin0800: do you want a video build with amicon's "proposed" fix? Seeing as you were having problems, perhaps you should try the fix.
23:24:48 Quit desowin ("use linux")
23:24:54amiconnsaratoga: There are some float tables in wmadata.h. Are these used at all?
23:25:18Febsamiconn: 1 hour plus and still running.
23:25:26amiconnAlso, is this style (actual data defined in .h file) a legacy from ffmpeg?
23:25:35pondlifestruct mp3entry defines path of size MAX_PATH. We then memcpy MAX_PATH+1 bytes from it... not quite right.
23:25:41FebsThough my battery is down to 30%.
23:25:51saratogaamiconn: which tables?
23:25:56amiconnFebs: Just connect the charger...
23:26:12S1gncan I view non english texts in built-in text viewer in rockbox? no success so far
23:26:14amiconnsaratoga: lsp_codebook
23:26:14*Febs tries to remember if he started from a full battery.
23:26:28saratogaamiconn: the file is from ffmpeg
23:26:51saratogalsp_codebook isn't used yet
23:26:57saratogait will be once I get LSP working
23:26:57SoapFebs: my battery isn't dropping fast, as best I can tell
23:27:07amiconnWhat's lsp?
23:27:14saratogaline spectral pairs
23:27:22saratogaits a low bitrate thing used by Ogg and WMA
23:27:27saratogai don't understand how it works yet
23:27:38saratogabut the encoder switches it on at 24kbps
23:28:09*amiconn is reading
23:28:20FebsSoap: it may be my battery. I installed a replacement last week and I'm not yet sure that it is any good. I also can't say for certain that I started this test with a full battery.
23:28:28saratogai need to sit down with a DSP text and learn what the heck they do
23:28:48pondlifeamiconn: Shouldn't pathname variables be stored in char[MAX_PATH+1]? I think so, but I fear for the binary size a little.
23:28:49PaulJamS1gn: the text viewer has its own codepage setting, mybe that is your problem?
23:29:24pondlifeAh no - actually it should be MAX_PATH everywhere by the look of it.
23:30:09S1gnPaulJam: probably.. but how do I set codepage in viewer ?
23:30:23 Quit merbanan (Remote closed the connection)
23:30:54PaulJamS1gn: look in the rockbox manual for your device how to access the menu of the textviewer.
23:36:48S1gnPaulJam: no word in manual about codepage in viewer :-/
23:37:15S1gnonly word wrap, scrollbar and stuff like that
23:38:19 Quit saratoga ("CGI:IRC (EOF)")
23:38:41PaulJamah, looks like the manual isn't quite up to date. what player do you have?
23:39:52S1gnSanDisk Sansa e200 (but now use uisimulator on linux box)
23:41:14S1gnoh wait... i pressed "select" and there is something interesting there
23:41:39S1gnoh yes! it worked )
23:44:19 Part maffe
23:47:39 Join maffe [0] (
23:47:58 Join Davide-NYC [0] (
23:51:17Robin0800Soap,no not realy unless you can post it as a patch or diff file
23:51:19 Part maffe
23:51:34Soapit's one line
23:51:35 Join maffe [0] (
23:51:42SoapI was going to offer you the build
23:54:14Robin0800Soap,very kind of you but I would also like to add patches
23:55:29PaulJamif you test stability you schouldnt use other patches
23:55:40Soap1 hour and 50 minutes ago ami con said in this channel the line which needs added and where
23:55:53*bluebrother thinks when hunting bugs no additional patches make sense at all
23:56:48 Quit ender` (" On-line, adj. The idea that a human being should always be accessible to a computer.")
23:56:58Soapso were your complaints in the forums about a stock build or a custom one?
23:57:42 Join kubiix [0] (

Previous day | Next day