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 2004-09-15

00:00:07DigitalDivawell i find the difference between an analogue and digital recording almost impossible to tell
00:00:28Zagorthe funny thing is that mp3 itself is a SCMS system, so adding a limiter is completely pointless
00:00:39DigitalDivareally?
00:00:42amiconnZagor: The archos recording screen does have one advantage compared to the rockbox version: It shows the average bitrate next to the file size
00:00:53DigitalDivaIm gonna respray my archos one day, the silver paint is wearing odd
00:00:54DigitalDivaoff
00:01:02ZagorDigitalDiva: yes, if you mp3 an mp3 file, it will sound worse than the original
00:01:39Zagorit's not a deliberate SCMS system, it's just gets that way due to the mp3 data reduction process
00:01:41DigitalDivaokay
00:02:17DigitalDivaim not particulary keen on propiretoory li-ion batteries in newer devices
00:02:25amiconnZagor: Yes, but once you have an mp3, you can copy it digitally without further degradation
00:02:44Zagoramiconn: should be easy enough to add. in fact we've been discussing a while-recording-screen and a while-radio-screen
00:02:59amiconnSounds interesting!
00:03:05Zagoramiconn: correct, so the spdif block does not help
00:03:24Zagorall it does is annoy the user
00:04:02Zagor...and *force* him to make bit-perfect copies instead of degraded rerecorded copies!
00:04:03amiconnZagor: For while-recording-screen and while-radio-screen, the extension would both be .wrs ?? ;)
00:04:35Zagorthe extension naming committee is still in session about that :)
00:05:00zehow about while-fm-screen .wfs
00:05:26Zagorsounds good. i'll tell the committee ;)
00:10:31[IDC]DragonUSB mode works!
00:10:35Zagoryay
00:10:46[IDC]Dragon(for external MMC so far)
00:10:49amiconnyay!
00:10:50amiconn[IDC]Dragon: What was the problem?
00:10:54Bagderyou rock Jörg
00:11:08[IDC]Dragonmissing chip select, in addition to the collision
00:11:43amiconnAnd I just read that you found the pin that signals bridge activity to the cpu :)
00:12:15amiconnI though that this signal must be somewhere, because the archos fw shows those dots..
00:12:34amiconn*thought
00:13:16[IDC]Dragonyes, so did I, but I haven't checked thorougly because of low prio
00:13:39[IDC]Dragonit's just an optical gimmick, and doesn't work very well
00:14:04[IDC]Dragon(after unmount, it still animates)
00:14:32amiconnI think it's nice to have, and of course we will implement it better ;)
00:14:49 Quit Bagder ("Leaving")
00:14:53[IDC]Dragonif the signal is better, yes
00:19:58[IDC]Dragonthe bridge drives the MMC with 12 MHz!?
00:21:04amiconnThat's within specs.
00:21:19amiconnDepending on the card, MMC works up to 20 MHz
00:23:27 Join bagawk [0] (lee@IC61.library.oregonstate.edu)
00:23:53[IDC]Dragonthis idle power off during USB is annoying
00:24:20[IDC]DragonI wonder why that happens, the USB detectionis obviously correct
00:24:51amiconnAlready committed working usb mode?
00:25:03[IDC]DragonI'll do that
00:25:15[IDC]Dragonbut it's external MMC only
00:25:36amiconnWhy's that?
00:25:47[IDC]Dragondon't know
00:26:06[IDC]Dragonmaybe the chip select for the internal is wrong
00:28:06amiconnDo you have to assert the MMC chip select (PA9) for usb mode to work??
00:28:31[IDC]Dragonyes
00:28:38amiconnStrange...
00:28:53amiconnThe data sheet tells us that MMC mode does not use chip select
00:29:01[IDC]Dragonah, yes
00:29:13[IDC]Dragonwhich level should it have then?
00:29:28*amiconn is looking it up
00:30:17amiconnhigh
00:30:27amiconnlow is active
00:31:00amiconnThe card is switched to SPI mode by sending CMD0 while CS is low
00:31:36[IDC]DragonI switched it high, this was bullshit
00:32:14amiconnIt *has* to be high for MMC mode (=bridge) to work
00:32:19[IDC]Dragonthen probably the internal card didn't change from SPI to MMC mode
00:32:54[IDC]Dragonhigh is good then, but my understanding was wrong and the implementation inverted
00:40:26 Quit AciD ("ou double !")
00:40:28bagawk[IDC]Dragon, Would keeping the backlight off in the charging screen save a little power, and charge better?
00:40:39amiconn[IDC]Dragon: Internal flash chip select *is* PA10
00:41:15[IDC]Dragonbagawk: not worth it
00:41:30[IDC]Dragonamiconn: I think so, too
00:41:38[IDC]Dragonwhat leads you there?
00:41:39amiconnI'm pretty sure
00:41:59[IDC]Dragonthe internal flash has a reset
00:42:10amiconnAhem, disassembler listing :)
00:42:29amiconnFound the internal/external card selector routine
00:42:56[IDC]Dragonit that reset is controllable, we could bring the card back to MMC mode
00:43:22amiconnThere must be a way to do that
00:43:51[IDC]Dragonthe external has no reset
00:44:17 Join PaulS [0] (~0d0274bd@labb.contactor.se)
00:44:51amiconn[IDC]Dragon: We need to know where this is connected
00:45:11[IDC]Dragontricky, this is BGA
00:45:37[IDC]Dragonbut not many ports left ;-)
00:45:54amiconnIf I find the routine...
00:46:06PaulSI added a little to IriverPort to explain how to get SDRAM and SRAM images out of the firmware.
00:46:13bagawki am out of disk space for my jukebox now :(
00:46:59ZagorPaulS: nice
00:48:06***Saving seen data "./dancer.seen"
00:48:15 Quit mecraw_ ("Trillian (http://www.ceruleanstudios.com)")
00:48:22PaulSI didn't want to take up more space with the actual dd commands, but someone "skilled in the art" should be able to deduce that. :-)
00:48:24[IDC]Dragoninternal MMC works!
00:50:07[IDC]Dragonbut this broke the external
00:50:36amiconn???
00:51:06ZagorPaulS: indeed
00:52:51[IDC]DragonPA7 seems to be the reset
00:57:53PaulSHm.. I threw in a dd example anyway..
00:58:06 Quit DigitalDiva ("CGI:IRC (Ping timeout)")
01:00
01:00:42amiconn[IDC]Dragon: It seems so, I found something that looks like a reset routine. It uses PA7
01:01:33[IDC]Dragonsee above, yes
01:02:50amiconnThe SanDisk datasheet is rather valuable
01:05:26[IDC]Dragonok, now both are working
01:06:22[IDC]Dragoncommitted
01:06:23amiconnIt seems I should start working on the driver itself :)
01:06:39[IDC]Dragonyes, the time is ripe
01:06:56[IDC]Dragonbut we don't mind a debugged recording
01:07:34amiconnI'll have to wait for Linus doing some logic analyzing. Until he did that, I could work on the driver...
01:08:24[IDC]DragonI work on my sleep now
01:08:32amiconn;)
01:09:09amiconnNite.
01:09:22[IDC]Dragonthe cvs code is not what I'm running, I made a hack in init() to loop the main menu.
01:09:31[IDC]DragonI haven't commited that.
01:10:34[IDC]Dragon while(1) main_menu();
01:10:50[IDC]Dragonafter the usb_start_monitoring() call.
01:15:36amiconnyEAH, ROCKBOX USB LOGO!!
01:15:44amiconnOoops, caps lock
01:16:35midk:)
01:20:12Zagorbed time, bye
01:20:12[IDC]Dragonyou're building Ondio?
01:20:14 Quit Zagor ("Client exiting")
01:20:29amiconn[IDC]Dragon: I did...
01:21:08amiconnAfter extracting USB, it fails with *PANIC* disk: NULL
01:21:20amiconnThe only way to switch it off then is to cut power
01:21:44[IDC]Dragonyes, because it tries to mount a partition
01:21:59amiconnOf course, I only wanted to tell
01:22:09[IDC]Dragon;-)
01:22:28 Quit _aLF ("good night")
01:22:35[IDC]Dragonyou can hold OFF for 10 sec to power off
01:22:50amiconnI wonder how we will handle the case that someone inserts or extracts a card while rockbox is running
01:23:04amiconnThe file system will have to be remounted
01:23:20amiconnThis renders all open file descriptors invalid
01:23:24[IDC]Dragonright now I only check at init
01:23:33[IDC]Dragonlet's worry later
01:23:39 Quit PaulS ("CGI:IRC")
01:23:44amiconnYup, only a thought
01:23:47 Join PaulS [0] (~0d0274bd@labb.contactor.se)
01:23:52[IDC]Dragongoodnight, happy Ondio hacking!
01:23:58 Quit [IDC]Dragon ()
01:54:39 Quit bagawk ("Leaving")
02:00
02:03:49 Part amiconn
02:19:57 Quit midk (Remote closed the connection)
02:21:41 Join midk [0] (~midk@c66-235-14-120.sea2.cablespeed.com)
02:23:15 Join BC_ [0] (~BlueChip@cpc3-colc1-3-0-cust61.colc.cable.ntl.com)
02:27:33BC_just popped in to say "hi"
02:27:37 Nick BC_ is now known as BC (~BlueChip@cpc3-colc1-3-0-cust61.colc.cable.ntl.com)
02:27:45midkhey bc
02:27:53BChey mk
02:28:12midkthere was something i wanted to commend you on, but i forget now :)
02:28:21BClol - thanks
02:28:22midkhm, perhaps there wasn't...
02:28:27BClol
02:29:13BCjust got my head really stuck in to my new book and got fed a new bunch of MAS docs
02:29:51midkoh, right, that reminds me, didn't you locate a set of karaoke registers?
02:29:55midkreal karaoke, that is
02:29:56BCyeah
02:30:09midkthat sounded fascinating
02:30:34BCjust plug a mic into the unit and use the in-built mixer
02:30:50midkvery cool
02:30:58BCcouple that with the centre removal system and you have a pretty good karaoke machine
02:31:18BCalso sussed mixing of stereo mp3s
02:31:35midkyou should consider tossing in a true karaoke mode.. i suppose there isn't room in audio-3587 though
02:31:43BCin fact you could do both at the same time
02:31:52BCwhat is "true" karaoke?
02:32:10BCdo you mean a cdg displayer?
02:32:19midki suppose 'true' karaoke wouldn't be possible
02:32:50BCyou can convert true karaoke disks to a mp3/cdg pair and play those
02:32:52midkalthough if there's a seperate set of registers it'd have to be a little better than the algorithm rocbox uses... isn't it stereo-mono?
02:33:27BCcan you explain what you mean by "true" karaoke?
02:33:48midkjust that - actual vocal removal instead of just taking mono away from stereo
02:34:19BCrighty - no afiak nobody has sussed it yet
02:34:29BCfor anything anywehere
02:35:06midkyeah, now it sounds a bit more impossible when i have to actually think about it :)
02:36:16BCkaraoke mode is a two minute job
02:36:51midkwhat kind of karaoke mode?
02:36:57midkthe existing, or a new one using these registers?
02:37:18BCexisting + mic mixer
02:38:43midka karaoke plugin would be neat
02:39:30BCyou just have to write a number to a register - it's all documented in the pdf on the rockbx site
02:40:08midkbut with a few options i mean, like how much of a mix between mic and song, or toggling 'imitation karaoke' on or off
02:41:17BCyep, iirc, there is also a mic-volume level and the "st-mono" is just another couple of registers - well documented again, probably easiest to nick the code from Audio-3587
02:41:44BCyou could even have a dial for how much to remove the original voices
02:41:54midkyep, that'd be neat
02:48:10***Saving seen data "./dancer.seen"
02:50:01BCit's on page 42 of the docs
02:51:33BCnite all
02:51:46 Part BC
02:57:35 Join bagawk [0] (Lee@ACC3220B.ipt.aol.com)
03:00
03:01:00 Quit bagawk (Client Quit)
03:13:12 Join pike|| [0] (amiga@h234n1fls22o1064.bredband.comhem.se)
03:13:34 Quit pike (Nick collision from services.)
03:13:34 Nick pike|| is now known as pike (amiga@h234n1fls22o1064.bredband.comhem.se)
03:16:13 Join bagawk [0] (Lee@ACC0DB51.ipt.aol.com)
03:31:21 Quit bagawk ("umount /dev/brain")
03:35:40 Join plok [0] (s336156@student.uq.edu.au)
03:41:14*plok is away - Automatically set away. - messages will be saved.
04:00
04:02:07 Quit PaulS ("CGI:IRC (EOF)")
04:19:36 Quit midk (Read error: 104 (Connection reset by peer))
04:19:48 Join midk_ [0] (~midk@c66-235-14-120.sea2.cablespeed.com)
04:25:07 Nick midk_ is now known as midk (~midk@c66-235-14-120.sea2.cablespeed.com)
04:48:14***Saving seen data "./dancer.seen"
05:00
05:28:55 Join grogro [0] (~gromit@ALagny-151-1-22-167.w83-114.abo.wanadoo.fr)
05:37:49 Join dreed [0] (~45912877@labb.contactor.se)
05:38:50dreedhey i am having a problem compiling i think....
05:39:13dreedI downloaded the 2.2 source and didn't change anything, just compiled it and this is what I got....
05:39:17dreed/opt/sh1/lib/gcc-lib/sh-elf/3.3.1/../../../../sh-elf/bin/ld: section .iram [0000
05:39:17dreed00000903dd68 -> 000000000903ea9b] overlaps section .stack [000000000903dd68 -> 0
05:39:17dreed00000000903fd67]
05:39:17DBUGEnqueued KICK dreed
05:39:17dreedcollect2: ld returned 1 exit status
05:39:17dreedmake[1]: *** [/home/David/rockbox/build/rockbox.elf] Error 1
05:39:18***Alert Mode level 1
05:39:18dreedmake[1]: Leaving directory `/home/David/rockbox/apps'
05:39:20dreedmake: *** [apps] Error 2
05:40:35midkget a daily, first off.
05:40:52midkit also looks like something isn't installed correctly, tjat
05:41:00midkthat's not a rockbox error
05:41:43dreedi didnt think it was a rockbox error but i just installed cygwin and i probably didnt do it right
05:41:56dreedrockbox rox!!
05:41:58midkhow did you install? every package?
05:42:06midk(implied by the name, yes. :))
05:43:08dreedi follwed the instructions step by step here:http://rockbox.haxx.se/twiki/bin/view/Main/CygwinDevelopment
05:43:44midkyou sure you did all of them?
05:43:55dreedthe only thing i saw is that it said to type in that one URL to get those other two packages and i did that but the version that was available was older than the one listed on your page
05:44:32dreedsh-gcc latest version is 3.4.1 but the one on the site was like 3.3
05:45:28midki'd still follow it exactly: if it doesn't work exactly as he says then you've got reason to complain
05:45:37midki suggest an uninstall, and use bc's devkit instead.
05:45:48midkhttp://homepage.ntlworld.com/cyborgsystems/CS_Main/RockBox/RockBox.htm
05:45:50midkit's down the page a bit.
05:46:06dreedwell i am not complaiing I am just a newbie and was lost :)
05:46:20dreedokay thanks, I'll try that
05:46:38midkit's np :)
05:46:52midki didn't mean you were complaining, just that you should if it doesn't work like he says. :)
05:49:20***Alert Mode OFF
05:50:21 Quit maikeul (Read error: 110 (Connection timed out))
05:51:14 Quit dreed ("CGI:IRC (EOF)")
06:00
06:42:24 Join LinusN [0] (~linus@labb.contactor.se)
06:43:43dwihnoWelcome back to *drumroll* Rockbox, your 24/7 support \o/ ;)
06:43:50dwihnoNah
06:43:50dwihno:)
06:44:36dwihnoWhat should I do? My left shift button on my keyboard is not always de-clicking correctly :/
06:44:47midki know!
06:44:49midk(lightbulb)
06:44:50midkfix it.
06:44:54midk:)
06:45:01dwihnoHow? :)
06:45:29midkby.. fixing it.
06:45:31midk:)
06:45:36midkpop it off, perhaps it's gunky.
06:45:51midkpry it off, clean it, and the place it rests
06:47:02dwihnoYou think that's possible? I mean, this is a laptop keyboard
06:47:43midkoh.
06:47:44midkhm.
06:48:15***Saving seen data "./dancer.seen"
06:48:16LinusNdwihno: dell?
06:49:53LinusNyes, you can remove the keys and clean them, i've done it myself on my dell laptop
06:50:19dwihnoAh
06:50:24LinusNthey are a bit tricky to put back, but it works
06:50:26dwihnoNice, I have a dell aswell
06:51:02LinusNi spilled coffee on it, and some keys got stuck
06:51:09dwihnoDo you have to remove the keyboard from the laptop, or do you just pry the keys off?
06:51:16dwihnoCoffe and computers just don't mix :)
06:51:32dwihnoI think I've managed to do that three times
06:51:34midkjust pop off the keys.
06:51:35LinusNit's easier if you remove the keyboard
06:51:39midkhmm
06:51:41midki stand corrected
06:51:48CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
06:51:48*midk quiets down and shuts off lightbulb
06:51:49dwihnoIs it hard to remove it?
06:52:14LinusNprying it off is simple with the keyboard in place, but putting it back is easier when you have the keyboard in your hand
06:52:32LinusNyou unscrew the top plate and then you can just lift it off
06:54:23dwihnotop plate?
06:54:40LinusNthe frame that surrounds the keyboard
06:55:16dwihnookay, so one of the screws on the bottom of the box then
06:55:20LinusNthere are screws under the laptop marked "K" for keyboard
06:55:26dwihno:-)
06:57:04dwihnoAh
06:57:07dwihno4 screws
06:57:11dwihnoOne shared with "m"
06:57:18dwihnoand one attached to some kind of compartment
06:57:32dwihnoNow THAT'S where the small elves' gotta be!
06:57:34LinusNLatitude?
06:58:12dwihnoinspiron
06:58:16LinusNok
06:58:27dwihnoit's taller than that korean basket playing guy
06:59:04dwihnoabout 3 cm thick, and then the screen is flipped up
06:59:17dwihnoadd 8mm's or such
06:59:24dwihnobut I love it!
06:59:32dwihnoThe only bad thing is the lack of onboard usb2
06:59:52LinusNi have a latitude C840, really nice
07:00
07:00:11dwihnodoes it have those replaceable hand plates?
07:00:42LinusNhand plates?
07:00:47dwihnoI'd like to remove these and clean it as well; I can see a lot of dust
07:01:12dwihnowhere you keep your arms, there are replaceable pads
07:01:14dwihnoplastic
07:01:20dwihnoyou can buy different colors etc
07:01:29dwihnoWould be pretty cool with yellow ones :)
07:02:16LinusNno, the latitude doesn't have those
07:02:24dwihno+−−−−−−-+
07:02:24dwihno|:::::::|
07:02:24dwihno| |
07:02:24DBUGEnqueued KICK dwihno
07:02:24dwihno| [ ] |
07:02:24dwihno| |
07:02:26dwihno+( )_( )+
07:02:29dwihnosomething like that
07:02:44dwihnothe square being that pad thingy
07:02:51dwihnomouse pad, whateveryoucallit
07:04:01LinusNtouch pad, i believe it's called
07:04:12dwihnoprobably
07:04:24dwihnomouse pad thingy \o/ :)
07:04:47dwihnoTime for some hot choco!
07:10:00dwihnoLinusN: I am grateful for all your guidance and help (not just this once). I just want you to know, that it's greatly apperciated.
07:10:43LinusNyou're welcome
07:23:59 Quit scott666_ ("i'll be back...eventually...")
07:25:21plokWhat sort of instruction set does the iRiver CPU have?
07:27:02LinusNplok: very similar to the 68000
07:27:28LinusNkind of a RISC version of the 68k
07:30:15plokAah. Is there a compatible C++ compiler for it?
07:31:57LinusNgcc
07:32:22LinusNrockbox is plain C though
07:34:20plokta. Yeah I noticed that. It should be possible to write C++ code and build it with the rest of rockbox though shouldn't it (eg if someone's just messing around with the code)
07:36:20LinusNthe linker may not be happy
07:37:00LinusNwhy is it so important with C++?
07:38:12plokIt's not important, I just prefer C++. It's not a big deal if I can't get it going :)
07:39:59plokProbably the easiest thing for me to do would be to see how close rockbox goes to compiling with a C++ compiler
07:46:17LinusNit will compile with a c++ compiler
07:52:15plok:)
07:54:57 Join Zagor [242] (~bjst@labb.contactor.se)
07:56:04LinusNmorning zagor
07:56:44Zagorhi
07:57:57plokbinutils-2.15 is not on the australian gnu mirror. Can I use 2.14 or should I get 2.15 from another server?
07:58:26Zagor2.14 might work but I'd use 2.15 just to be sure
07:58:40plokok, thanks.
08:00
08:22:38plokis the syntax simply −−target=sh-elf −−enable-languages=c,c++ if I want to include a c++ compiler when building the cross compiler?
08:24:34Zagorplok: yes, but you won't succeed :)
08:24:52Zagorthe standard c++ library is not as portable as the compiler itself
08:26:23 Join amiconn [0] (~jens@pD9E7DFB8.dip.t-dialin.net)
08:26:39plok!! I suppose that explains why I got an error when building with c++ enabled.. oh well
08:27:01amiconngood morning rockbox world
08:28:44plokgood morning. I just noticed binutils is an anagram of builtins
08:29:29amiconnLinusN: r u there?
08:29:50Zagorplok: :)
08:32:46LinusNamiconn: here now
08:32:49plokoops, I also hadn't added my newly built binutils to the path
08:33:42LinusNplok: i suggest you stick with the plain c compiler
08:33:43amiconnLinusN: Did you read last night's backlogs? I have somer really interesting finding concerning the recording transfer, but need you r help with logic analyzing.
08:33:56amiconn*some
08:33:59LinusNplok: we won't accept any c++ conde contributions anyway
08:34:04LinusNs/conde/code/
08:34:21LinusNamiconn: hang on, will read up
08:34:24plokhmm.. Linus, just switched back to the plain c compiler, trying to get gcc to build. got this weird error though:
08:34:25plokinsn-emit.c:8549: Internal compiler error in simplify_subreg, at simplify-rtx.c:2452
08:36:05plokdouble oops, added the root sh1 dir to the path, not sh1/bin
08:37:26amiconnLinusN: I would really appreciate if you could do the following: Hook up your LA and capture the recording transfer, but this time *including* PB14 (MAS demand IRQ). Please do this both for rockbox and archos fw, and if at all possible also capture the start of recording.
08:39:19LinusNamiconn: i'm pretty sure that EOD is the PB14 pin in my older analysis, that's why it's inverted
08:40:12LinusNthe "Note:" sentence on top of the web page describes that
08:40:39amiconnBut EOD is PA15 ???
08:40:48LinusNhmmm
08:42:05LinusNPB15 = RTW in the port pin table
08:43:01amiconnYes. P_B_15 = RTW, P_A_15 is EOD
08:43:37LinusNPB14 is the inverted PA15
08:44:07amiconnreally?
08:44:13LinusNso there will be a few nanosecs delay on one of the pins, have to check the schematics to see which one is delayed
08:44:56LinusNthere is an inverter, since the irq can only be edge triggered on the falling edge
08:47:40LinusNi don't remember which one i soldered my probe to, but the polarity in the analyzer trace should give a clue
08:48:17***Saving seen data "./dancer.seen"
08:48:23LinusNamiconn: i can't find your findings in the irc log, maybe i'm blind
08:48:29amiconnI can't find that inverter in the schematics....
08:49:45amiconnLinusN: Look at http://rockbox.haxx.se/irc/rockbox-20040914.txt , juster after 23:26
08:50:19amiconnBut if PB14 is just inverted PA15, it may be that it is not that important
09:00
09:00:18LinusNthe mas doesn't send any data until there is a S/PDIF signal
09:00:19amiconnAh, the inverter is on the far left in the schematics
09:00:34LinusNyup
09:02:02 Join amiconn_ [0] (~jens@pD9E7E9EB.dip.t-dialin.net)
09:02:15 Quit amiconn (Nick collision from services.)
09:02:15 Nick amiconn_ is now known as amiconn (~jens@pD9E7E9EB.dip.t-dialin.net)
09:03:38amiconnI'll try to write an all-new recording transfer routine in the evening
09:05:09LinusNamiconn: what's wrong with the current one?
09:06:40amiconnI'll try removing the 30 byte limit, and adding a timeout to the PRTW polling.
09:07:37amiconnPerhaps we do without those additional delays and drain_dma_buffer() calls etc.
09:07:43amiconn*we can do
09:13:33LinusNgood
09:13:37amiconnAnd using PB14 instead of PA15 should be better, because of less delay
09:14:10LinusNsure
09:20:33amiconnHmm. Did you measure the recording transfer with archos fw some time?
09:20:49 Join Bagder [0] (~daniel@1-1-5-26a.hud.sth.bostream.se)
09:22:35LinusNamiconn: i don't think so
09:30:22Zagoroh bugger, making display size run-time changeable is really messy...
09:30:32Zagori think i'll shelf it for now
09:31:32 Join Paco [0] (~527fc887@labb.contactor.se)
09:33:05 Quit Paco (Client Quit)
09:33:58 Join [av]bani [0] (~goemon@washuu.anime.net)
09:34:02[av]bani"The uClinux-dist also contains
09:34:02[av]banian integer only mp3 player (mp3play) which runs on the 5249.
09:34:02[av]baniAlthough it needs some optimizations to play higher bit rate
09:34:02DBUGEnqueued KICK [av]bani
09:34:02[av]banimp3s in real time (I have played with this and have it working,
09:34:04[av]banijust not in uClinux dist currently)."
09:34:10[av]bani:D
09:34:36Bagdernice
09:36:10[av]bani"For toolchains I use the uClinux m68k-elf toolchain (this is
09:36:10[av]banijust gcc-2.95.3 and binutils-2.14 with a couple of patches).
09:36:10[av]baniFollow the links for tool chains from:
09:36:10[av]bani http://www.uclinux.org/pub/uClinux/dist/
09:36:19[av]baniThis toolchain is in extensive use on ColdFire platforms, it
09:36:19[av]baniis well tested and generates quite reasonable code.
09:36:26LinusNthe problem is the encoder
09:36:38[av]baniIn terms of CPU and platform support get the latest uClinux-dist
09:36:38[av]banisource code package (from the above link) and start there.
09:36:38[av]bani From the top level if you run "make xconfig" you can select the
09:36:38***Alert Mode level 1
09:36:38[av]baniMotorola/M5249C3 target and compile an image (combined kernel
09:36:38***Alert Mode level 2
09:36:38[av]baniand root filesystem) that is ready to run on the M5249C3 board.
09:36:38***Alert Mode level 3
09:36:38[av]baniIt is very easy.
09:37:13[av]banilinus, integerized asm mp3 encoders abound for 68k
09:37:35[av]baniatari and amiga having been quite heavily abused by demo coders :)
09:37:43Zagormy first option is the MAD decoder. it's pretty fast but above all Really Realy Good.
09:38:10[av]banianyone confirmed if the array of pads on the left is the bdm?
09:38:36LinusNkonrad and friends say so
09:38:54LinusNbut it's easier to solder on the marked pads on the pcb
09:38:57[av]baniit must be so! :)
09:39:12[av]baniwell, if there's a socket that can match the pads...
09:39:27LinusN[av]bani: you have seen fast mp3 encoders for 68k?
09:39:32[av]baniyes
09:39:34Zagor[av]bani: if you can find a nice integer encoder with free (libre) source, please toss it this way
09:40:12Zagorwe're out of the amiga scene since, oh, about a decade ;)
09:40:18[av]banii assume nobody's actually got a working bdm to the ihp yet though
09:40:31LinusNi've only made one demo for the amiga, and that didn't include an mp3 encoder :-)
09:42:04[av]baniand correct me if i'm wrong, the ISSI 4C01-3G can't possibly be a config rom for the cypress usb bridge
09:42:14LinusNwhy not?
09:42:28[av]banithe datasheet requires minimum 256bytes config space
09:42:43[av]bani1kbit = 128 bytes
09:43:33[av]baniand on top of that, where would the ihp's persistent data be stored then?
09:44:02[av]banii'm guessing the 1kbit is for the ihp, and the actual cypress config rom is on the underside of the board
09:44:17LinusNperhaps
09:44:20ZagorLinusN: it's time to remove the usb daughterboard
09:44:27Zagor:)
09:44:31LinusNi'll do a LA trace of the i2c data
09:44:52[av]baniyes, someone needs to get scans of the underside
09:44:54[av]bani(s)
09:45:02[av]banii'm guessing there are more chips hiding
09:45:15LinusNcertainly
09:45:39[av]banifwiw the LCDs with 6800 interface are good candidates, since 68k works natively with 6800 peripherals
09:46:06Zagormy bet is on the solomon
09:46:06[av]banithey can easily be mapped into the cpu's address space
09:46:15[av]banitrivial
09:46:39***Alert Mode OFF
09:46:44Zagorif they use parallell access, yes. archos didn't.
09:46:55Zagor(not that serial is hard either, of course)
09:47:03[av]baniandi denied it, though they didnt answer my followup questions
09:47:14[av]banii just told them iwanted a replacement lcd for a busted ihp :)
09:47:36[av]baniand if they had an equivalent part
09:47:41Zagor[av]bani: ah you are Dan Hollis?
09:47:48[av]baniyeh
09:47:57Zagorah, good
09:48:14Zagorgood digging
09:50:41[av]banihmm, serial would preclude any readback access to the lcd
09:50:49[av]baniblind writes...
09:51:40[av]baniwould be kind of silly to pick an lcd controller with 6800 interface and not use it
09:51:43LinusNwhy would you want to read from the lcd?
09:51:48[av]banibut who knows what those koreans think :)
09:51:59[av]baniread from the controller...
09:52:15LinusNwhy would you want to read from the controller
09:52:20[av]banistatus bits, etc
09:52:31[av]banishrug
09:52:43Zagorwe've managed fine with just writing so far :)
09:53:57[av]baniyeah, though higher performance would likely be possible on parallel
09:54:11[av]baninot to mention lower cpu usage to drive it
09:54:16[av]banivs bit banging it out
09:54:26Zagorabsolutely
09:54:33LinusNyes, of course, i just don't see why reading the controller is a benefit
09:54:53 Join [IDC]Dragon [0] (~d90a3255@labb.contactor.se)
09:54:53[av]banisynchronization, if the controller equires it
09:55:07[av]banidepends on the lcd, imho its quite likely for color LCDs
09:55:15[av]baniihp-3xx
09:55:33LinusNyou probably know more about lcd's than i do
09:55:36[av]banipretty unlikely its serial interface on that :)
09:55:41LinusN:-)
09:56:14[IDC]Dragon'morning guys!
09:56:21LinusNmorning
09:56:22[av]baninobody got working bdm yet though?
09:57:01Zagornot yet. our pod is in the mail.
09:57:31LinusNit will take a few days before i get it
09:57:39[av]banihm, the ssd1845 has a status register
09:57:47[av]banione of the bits indicates if the controller is busy or not
09:57:53[av]banicould be important, or not :)
09:59:33[av]banisomoene should trace the lcd connections so its easier to make a match to a datasheet
10:00
10:00:09Zagorit's difficult to trace the cpu, being a bga
10:00:36Zagorthat's why we are looking for a broken device, where we can remove the cpu
10:01:08Zagordoesn't look like we are going to find any broken ones though, iriver appears to be very nice about replacing broken units
10:01:29[IDC]Dragonhave you checked if the LCD is connected to the bus?
10:01:37[IDC]Dragone.g. at the SDRAM?
10:01:47[av]baniwell yeah, tracing where the lcd goes to...
10:01:47LinusNno
10:02:18LinusN[IDC]Dragon: good idea, didn't think of that
10:03:12[av]baniwell itll go into some kind of glue chip
10:03:56 Join LePoulpe303 [0] (~lpos@AMontpellier-251-1-35-39.w83-113.abo.wanadoo.fr)
10:03:58[av]baniand if its off the shelf part (likely) then you can know how its interfaced
10:04:03[av]baniand possibly, work out the lcd from that
10:04:24[av]banijust need to take off that daughterboard :)
10:05:05LinusNi doubt that the daughterboard has anything to do with the lcd
10:05:33LinusNit doesn't make any sense to have the lcd signal path go all that way
10:08:12CtcpIgnored 2 channel CTCP requests in 2 hours and 31 minutes at the last flood
10:08:12*[IDC]Dragon is master of tracing PCBs ;-)
10:15:01[IDC]DragonBagder: do you read?
10:15:08BagderI do
10:15:52[IDC]Dragonin my overly excitement I was thinking if we should already include the Ondio in the daily build
10:16:04[IDC]Dragonthe get those colors :-)
10:16:10[IDC]Dragonto get ...
10:16:20Bagderyes, we could start with including it in the build table
10:17:06BagderI'll set it up
10:17:08[IDC]Dragonthe Neo is in there too
10:17:16[IDC]Dragonbut an orphan
10:17:25LinusNwe should remove the neo
10:17:46LinusNuntil someone that cares joins the team
10:18:09[IDC]Dragonthe Ondio has two flavours, as you probably know, SP and FM
10:18:21Bagderyes, I've noticed
10:18:41Bagderthe plugins/lib doesn't build
10:18:57[IDC]Dragonah, yes
10:19:05[IDC]DragonI forgot
10:19:08Bagderbut that'll show up ;-)
10:19:12[av]banianyone poked at the xclef yet?
10:19:16[IDC]Dragonbetter not include it then
10:19:23LinusN[av]bani: nope
10:19:58Bagder[IDC]Dragon: we'll fix that quirk sooner if it shows up red!
10:20:31[IDC]DragonBagder: amiconn has an uncommitted fix for the plugin make, but then lots of plugins don't build because of the limited keypad
10:20:50Bagderok
10:21:12[IDC]Dragonbut currently, plugins make no sense. We first need disk I/O
10:21:18BagderI agree
10:21:59Bagderthey build ajbrec.ajz files, right?
10:22:05[IDC]Dragonyes
10:22:28Bagderadded
10:22:44Bagderafter next commit there should be two new columns in the table
10:22:46[IDC]Dragonno useable .ucl yet ;-)
10:22:54[IDC]Dragonwhow, that was quick!
10:23:25Bagderthe build system is actually quite nicely written to allow easily adding new ones
10:23:34[av]banithe bdm you purchased, works with gdb?
10:23:39LinusNyes
10:23:44[av]baniyay
10:23:46[av]baniwhich one?
10:23:49LinusNthat's why we bought it
10:24:11LinusNwe have a wiggler, but we need to write drivers for it
10:24:18[av]bani:o
10:24:24LinusNwe bought a p&e wiggler
10:24:26[IDC]Dragonis there some "real" hardware in it, or just a printer port adapter?
10:24:43[av]baniif its cheap, its likely just parallel port :)
10:24:44LinusNtheres a 16V8 PAL
10:24:47LinusNon it
10:24:54[av]banibuffers likely
10:24:54[IDC]Dragon:-(
10:25:21LinusNthe wiggler we have is just an LPT adapter with buffers for voltage diffs
10:26:08LinusNthe p&e wiggler contains some extra logic for BKPT timing and singlestepping
10:26:26LinusNwe have schematics and PAL descriptions for it
10:26:40LinusNbut we don't feel like wasting time building one
10:27:09LinusNwhen we can use the donated money buying one instead
10:27:55[av]banihttp://www.ucdot.org/article.pl?sid=04/01/20/014230&mode=thread
10:28:02LinusN[IDC]Dragon: how about ondio schematics?
10:29:01LinusN[av]bani: yes. too bad their project doesn't have a driver for the macraigor wiggler
10:29:22LinusN(the wiggler we have is a macraigor)
10:29:37[av]banilooks like it 'almos works' with the p&e
10:29:39[IDC]DragonLinusN: too much work, imho not worth the effort
10:29:57[IDC]Dragonwhat we really needed is the port assignment
10:30:15LinusNi agree
10:30:27[IDC]DragonI care only about the programmable part, not how it regulates power, etc.
10:30:53[av]baniooh
10:30:59[av]banihttp://cmp.felk.cvut.cz/~pisa/m683xx/bdm_driver-old.html
10:31:54[av]baniif its implemented like the gal16v8 on that page, then i guess gdb already works with it
10:32:42[IDC]DragonI worked a lot with GALs a dozen years ago
10:32:48[av]baniThe new experimental version for gdb-4.18 is based on latest Magin's code and contains all my changes for the Linux 2.2.x support and better timing ( at least I hope so ). This version can be foud in the archive gdb-4.18-bdm-patches-pi1.tar.gz. There is an initial version of the ispGAL22V8 based EFICD, as well.
10:33:56LinusNgdb 4.18, that's old
10:34:19[av]banithen presumably it's been finished by now :)
10:37:51 Join AciD [0] (~gni@longchamp44-1-82-67-133-87.fbx.proxad.net)
10:45:38[av]baniwoo
10:46:45 Quit midk (Read error: 113 (No route to host))
10:47:05[av]banihttp://cmp.felk.cvut.cz/~pisa/#m683xx
10:47:08[av]baniany of those suitable?
10:48:18***Saving seen data "./dancer.seen"
10:49:07LinusNthe ones on http://sourceforge.net/projects/bdm will probably do just fine
10:49:18LinusNif you mean bdm drivers
10:50:27*LinusN just wrote his first PIC assembly code
10:50:36[av]bani:o
10:52:59[IDC]Dragonit's a horrible architecture, I prefer Atmel
10:53:32Bagderok, now hopefully ondio builds will come around in... 5 minutes or so
10:53:34LinusNi have had a look at atmel, and so far i agree
10:53:48[IDC]Dragon(bank-switching already wit 128 bytes, brr)
10:54:27[IDC]DragonBagder: you committed something, to force a build?
10:54:34Bagderyes
10:54:39Bagderminor docs fix
10:54:58[IDC]Dragonhehe, our docs now benefit fromthe Ondio
10:55:05Bagder:-)
10:55:06 Part [av]bani
11:00
11:06:31[IDC]DragonBagder: I see the Ondio :-)
11:06:34Bagderme too
11:06:42Bagderbut they aren't red as they should be
11:06:57[IDC]Dragonohhh
11:07:20[IDC]Dragonthe general order of the columns is a bit weird
11:07:40Bagderits sorted alphabetically
11:07:48[IDC]Dragonah
11:07:54Bagderbut we should change that
11:08:11Bagderand I'll look into why they don't turn red
11:08:42[IDC]Dragonthere's plenty of error in the log
11:08:48Bagderyes
11:09:17[IDC]Dragonand you need a pretty picture
11:09:40[IDC]Dragonfor the zip part above
11:09:53Bagderyeps
11:10:00 Join pillo [0] (~trillian@navlab03.dei.unipd.it)
11:10:00 Quit pillo (Remote closed the connection)
11:10:56 Join pillo [0] (~trillian@navlab03.dei.unipd.it)
11:17:08Bagderbetter
11:19:36[IDC]Dragonworse :-/
11:20:03Bagderhehe
11:20:34[IDC]Dragonit even re-colored some old build
11:20:52Bagderyes, I found a parser flaw
11:21:17Bagderit missed that warning before
11:21:23[IDC]Dragonoops
11:22:38[IDC]Dragonhow about access to the compile result? (zip)
11:23:37Bagderthat's separate
11:24:12BagderI'll fix that later, I need to go now
11:25:35[IDC]Dragonsure
11:25:38[IDC]Dragonthanks
11:26:51Bagderthat number "30" should keep you busy :-)
11:26:56Bagdersee ya around
11:26:59 Quit Bagder ("Leaving")
11:29:12 Join midk [0] (~midk@c66-235-14-120.sea2.cablespeed.com)
11:39:42amiconn[IDC]Dragon: We'll actually need two images for the zip builds. In addition to the different marking, the top part of the SP case is violet, while the FM seems to have a blue top part of the case
11:51:40 Join Lynx_ [0] (lynx@134.95.189.59)
12:00
12:25:18 Join lImbus [0] (~manuel@kernel.cycos.net)
12:33:42[IDC]Dragonhi Jens
12:33:53[IDC]Dragonjust got back from lunch
12:34:55[IDC]DragonDaniel will probably ask us to photograph our boxes, while showing Rockbox logo, on gray background
12:41:38[IDC]Dragonaniconn: he, you've traded the 30 errors with 84 errors 2 warnings
12:42:01Zagorooh, nice trade ;)
12:46:19[IDC]Dragonclearcase installation forces me to reboot
12:46:27 Part LinusN
12:46:46 Part [IDC]Dragon
12:47:41 Join LinusN [0] (~linus@labb.contactor.se)
12:48:22***Saving seen data "./dancer.seen"
12:53:37 Join webguest05 [0] (~d9e8e020@labb.contactor.se)
12:55:06 Join [IDC]Dragon [0] (~d90a3255@labb.contactor.se)
13:00
13:06:47 Join ripnetUK [0] (~mirc@82-70-100-230.dsl.in-addr.zen.co.uk)
13:08:38 Quit webguest05 ("CGI:IRC (Ping timeout)")
13:38:08 Quit [IDC]Dragon (leguin.freenode.net irc.freenode.net)
13:38:08NSplitleguin.freenode.net irc.freenode.net
13:38:08dwihnoare there any ways to make gcc remove all unused code when compiling?
13:38:08Zagorput the code in libs
13:38:58dwihnoit only includes used stuff then?
13:39:01NHealleguin.freenode.net irc.freenode.net
13:39:01NJoin[IDC]Dragon [0] (~d90a3255@labb.contactor.se)
13:39:05Zagoryes
13:39:05dwihnoneato :D
13:39:11dwihnoranlib something something all .o, something something?
13:40:03Zagorar, not ranlib
13:40:32dwihnohm
13:40:53Zagorsee firmware/Makefile
13:59:33webmind_my archos has become obsolete.. now what to do with the remains...
13:59:46 Nick webmind_ is now known as webmind (~random@217-195-236-172.dsl.esined.net)
14:00
14:00:07Zagorwhich model is it?
14:04:48Zagoryikes, 160x128 is a lot bigger than 112x64...
14:05:27Zagor15 lines + statusbar with default font...
14:15:35[IDC]Dragonwebmind: send it to me, I have a little junkyard
14:16:44webmindZagor, player
14:16:47webmindstudio 20
14:21:20dwihnoYum! Huge screen! Me likey!
14:22:10Zagori think we'll be getting more demands for pretty icons
14:22:50dwihnoText interface forever! :)
14:23:05[IDC]Dragonvoice interface forever!
14:23:09Zagor:)
14:24:38dwihno:-)
14:25:06[IDC]Dragondwihno has a longer nose than Zagor
14:25:22dwihno:-¨¨)
14:25:27*dwihno is having a cold
14:25:30dwihnoDamn it sucks :(
14:31:05*ripnetUK dreams of a Rockbox user defined WPS on my iRiver
14:32:34ripnetUKin theory, the iRiver /SHOULD/ be able to support a gameboy emu :) how cool would that be...
14:33:08ZagorI haven't even begun thinking about all the fun stuff we can do with this hardware :)
14:33:39Zagorlots of cpu, grayscale graphics, pcm sound. there possibilities are massive.
14:35:28LinusNfirst, we should try to execute our first CPU instruction :-)
14:35:34ripnetUKhehe
14:36:16dwihnoDreamwrecker :)
14:37:06ripnetUKso i guess the citical thing holding us up is that we need to know how to recover from a bad flash? using the debug BDM interface? i know nothing about that - is that a way of taking control of the bus externally (like a xbox mod chip) to inject our own code?
14:37:15 Join ashridah [0] (~ashridah@icculus.org)
14:37:23Zagorhttp://bjorn.haxx.se/h100-boot.png
14:37:43ripnetUKzagor - tease :)
14:37:50ashridahheh. emulated progress? :)
14:38:37Zagoryup
14:38:50[IDC]Dragonmy Ondio picture is for real :)
14:38:56Zagorhehe
14:40:12LinusNripnetUK: i don't know of any other way than the bdm
14:41:33Zagorwe're basically treading water in the code execution department until the bdm pod we've ordered arrives
14:41:38ripnetUKi dont know anything about bdm - is it a standard debug interface
14:41:47Zagorso some of us are playing with the sim instead :)
14:42:13ripnetUK(daft question - i know how to google!)
14:42:44Zagorhttp://bjorn.haxx.se/h100-browser.png
14:43:16LinusNripnetUK: Background Debug Mode, a builtin debug interface in the CPU
14:43:47ripnetUKaha - looks like they start at $500ish ...
14:44:01LinusNrather $150ish
14:44:14dwihnoZagor: wee! big screen :)
14:44:15ripnetUKcool
14:44:19dwihnoZagor: me likey
14:44:25dwihnoImagine prop font on that one :)
14:44:37ripnetUKlooks good
14:44:47*ashridah would be decidedly happy if a clock actually works (and maintains time, but somehow i doubt that) with the iriver h1xx using rockbox
14:45:32ripnetUKso basically if u get a bad flash, you would use bdm pod to upload a flash routine and a boot loader? even if the flash was totally broke?
14:45:35Zagoryeah, that'll be rather difficult...
14:46:01Zagor(clock)
14:46:14ashridahmm. i imagine without a cmos-alike and a timer to keep it ticking it's a pipedream
14:46:18ashridahpity
14:46:19LinusNripnetUK: yes, we can recover from a bad flash with the bdm
14:46:52LinusNsome people at miticriver seem to think that a real time clock is the key to good shuffling :-)
14:46:54ripnetUKand I guess verify that your code is acutally running before you know the LED registers etc
14:47:45ashridahLinusN: some of the people at misticriver think that the delays in firmware releases are a conspiracy, and that it's grounds for a class action suit
14:47:57LinusNlol
14:48:01 Join Bagder [0] (~daniel@1-1-5-26a.hud.sth.bostream.se)
14:48:14Zagorhaha, grounds how? irreparable damage?
14:48:23***Saving seen data "./dancer.seen"
14:48:29LinusNcurrent bdm wiggler location: Newark
14:48:41ZagorLinusN: :)
14:48:54ashridahnevermind that they can't prove damages, and at worst, it's an advertising misdemeanor, and that's ONLY if they never had any intension of actually delivering
14:49:19ashridahif they did, but were unavoidably delayed by circumstances outside their control, bye bye suit.
14:49:34ripnetUKlawsuit indeed :) rotfl
14:51:36ZagorBagder: have you given any thought to how you'd like to see the simulator config system (or lack thereof) expanded? I now need LCD_WIDTH and HAVE_RTC and a bunch of other stuff. we should think about a good solution.
14:52:15BagderI think we need to make it use the config-*.h files
14:52:49 Quit lImbus (Remote closed the connection)
14:52:52Zagorthus having #ifdef SIMULATOR in the config files or in the code?
14:53:20Bagderno, but having LCD_WIDTH in the config-*.h files
14:53:41 Join lImbus [0] (~manuel@kernel.cycos.net)
14:54:09Zagoryeah, that I've done already. i'm just thinking setting all those HAVE_ means a lot of hardware-specific code will be compiled. but maybe it works good enough.
14:55:46Bagderlet's start from there
14:55:50Zagoryup
14:58:38Bagderthe reason we didn't do it before was my desire to allow "weird" mixing of features
14:58:45Bagderwhen doing simulation builds
14:59:06Bagderbut I now no longer think it is sensible
15:00
15:02:07ripnetUKsurely we need to have a variable, not a #define for lcd_width, especially as the iRiver has two lcd_widths (for main and remote LCDs)
15:03:01Bagderwe still need a buffer etc
15:03:08Bagderit can't be based on a variable
15:03:13ripnetUKof course! no malloc
15:03:38Bagdermalloc wouldn't make a difference for this, as I see it
15:04:35Zagorthe double display problem is hairy. it won't be solved immediately...
15:04:58Bagdermy gut reaction is that they should be treated as two screens
15:05:13Zagorwe do not only need two displays, with two buffers. we also need to run all rendering code twice!
15:05:34Zagor...and that's just no fun :)
15:07:15Zagorso this will happen one step at a time. first setting up the simulators for more varying configs. then allowing dynamic display size changes (which will break plugins). then, later, we'll start thinking of dual displays.
15:07:46Bagderyes, I think the first step could be made on the primary single display
15:07:53Zagoryup
15:08:51 Quit midk ("just STOP it arspy")
15:09:54 Join midk [0] (~midk@c66-235-14-120.sea2.cablespeed.com)
15:10:35ZagorBagder: did you see this http://bjorn.haxx.se/h100-browser.png
15:10:42ripnetUKit IS fun :) looking (way) forward, I will want 2 seperate WPS configs for remote and main
15:10:54BagderZagor: mucho neato
15:11:05BagderripnetUK: I agree
15:11:07LinusNand different fonts for the two
15:11:11midkhaha, very cool
15:11:11Bagderthat too
15:11:25BagderLinusN: but we'd solve that with font-support in the wps config
15:11:34Bagderimho
15:11:38LinusNgood idea
15:12:00Zagornot quite. we'd also need support for >1 loaded font
15:12:14Bagderright, but I meant the choosing of which font where
15:12:23Zagorok
15:13:11ripnetUKregarding http://bjorn.haxx.se/h100-browser.png - the iRiver has a BLUE backlight :) too cool to not have in the sim
15:13:26Zagorhehe, me fix
15:13:35Bagderhahaha
15:14:04Bagderany pretty-picture(s) of the Ondio(s) around?
15:14:12Bagder(for the daily-build)
15:14:21Zagorthere the big one in the wiki
15:15:39BagderI'll annoy Jörg and Jens when they get around
15:16:22Zagorbut i don't think we should have it in the top daily list until it runs reasonably well
15:16:34Zagorthe bleeding list though
15:17:59Bagdermakes sense
15:19:03 Quit ashridah ("leaving")
15:22:25[IDC]DragonBagder: I'm around
15:22:39[IDC]Dragonbut not my Ondio
15:22:42Bagderhehe
15:23:11Bagderwell, whenever you get time, make a clear shot of the Ondio for the daily stuff
15:23:37[IDC]Dragonsure
15:24:04Bagderdaily Ondio builds start tomorrow, but there is no links provided for them just yet
15:24:05[IDC]Dragonthe first image on the wiki pake has one
15:24:23[IDC]Dragons/pake/page
15:24:33BagderI know, but it is just so close to the recorder
15:24:39[IDC]Dragonbut with my JBR next to it
15:24:41Bagderit would be better with a more stand-alone
15:25:01[IDC]Dragonhow did you do the others? (background)
15:25:10BagderLinusN made them I believe
15:25:19LinusNi used my flatbed scanner
15:25:59LinusNthe background was a white sheet of paper
15:26:01midkvery cool daily builds page now
15:26:03[IDC]Dragonfor this icon-ish picture we need no 1200 dpi, I think
15:26:21LinusNlook at the device comparison page
15:26:30LinusNthe pics are clickable
15:26:51Bagdermidk: yes, I like the new look too
15:27:15[IDC]Dragonoh, I need to run Rockbox 2.2 for the photo
15:27:24LinusN:-)
15:27:49midki think it would be nice to have an archive of daily build tarballs and/or released packages, but that would probably take up a lot of space
15:28:05Bagdernow check the bleeding edge box
15:28:18Bagderondio ajbrec links
15:28:45midkyou should span them over the two that were previously empty
15:28:50[IDC]Dragonah, nice
15:28:54midkinstead of making the recorder column longer
15:28:57midkimo.
15:29:10Bagderok then ;-)
15:29:26[IDC]DragonI can remove the .ajz from the twiki page then
15:29:33Bagderyes
15:31:50Bagderwant me to disable the plugin build for the Ondio?
15:36:39 Join hiolo [0] (Lupettino@host178-121.pool80182.interbusiness.it)
15:37:45[IDC]Dragonby doing what?
15:37:56Bagderif'ing in the makefile
15:38:25[IDC]Dragonyes, I think it's better for now
15:38:57Bagderok, done
15:39:04[IDC]Dragonthanks
15:39:12[IDC]Dragon(no cvs here)
15:39:29[IDC]Dragon(firewalled)
15:39:37 Quit hiolo (Excess Flood)
15:40:04[IDC]Dragonbtw, I like the new, brief make output
15:40:25Bagderit makes it easier to detect actual problems
15:40:29[IDC]Dragonwith the old, flooding one I regulary overlooked warnings
15:40:34Bagderyes
15:40:38BagderI agree
15:40:47[IDC]Dragonunless I did make > /dev/null
15:40:59Bagderthat's my primary reason for doing this new output
15:45:32[IDC]Dragonbuilds green now :-)
15:45:48Bagdergreen is nice
15:46:02[IDC]Dragonlooks friendly
15:47:03BagderI'll cut off the seconds part from the batch time
15:47:53[IDC]Dragon?
15:48:10Bagderto narrow the batch column
15:48:52[IDC]Dragonnot showing the build time any more?
15:49:06Bagdernot the seconds field, hour:minute is good enough
15:49:08Zagor09:40 instead of 09:40:00
15:49:14[IDC]Dragonah, yes
15:55:24[IDC]DragonI don't know how to "attract" Ondio users. So far, onlyJens and I have tried it.
15:55:45[IDC]Dragon(it=running ROckbox, checking hardware mask)
15:56:03 Quit LePoulpe303 (Read error: 232 (Connection reset by peer))
15:56:18Bagderperhaps posting on the two user forums could help
15:57:01[IDC]Dragonwhich forums?
15:57:05*Bagder hears a little girl waking up the next room
15:57:13LinusNyahoo and newmp3tech
15:57:25Bagdersee ya
15:57:26[IDC]Dragonoh, I typed too loud
15:57:27 Quit Bagder ("Leaving")
15:57:38LinusNand on the rockbox front page
15:57:46[IDC]Dragonmaybe i use too much upper case
15:58:07LinusNgotta go, cu around
15:58:07Zagorwhich is the lesser evil: adding #ifdef SIMULATOR to the config-xxx.h files or adding lots of new #ifdef SIMULATOR in the drivers etc?
15:58:35LinusNi think you will need both
15:59:01Zagorgee thanks :)
15:59:13LinusNHAVE_RTC, for example, may have a simulated RTC component
15:59:59LinusNso you will end up defining HAVE_SIM_RTC or something in the config file
16:00
16:00:11LinusNand still have ifdefs in the drivers
16:00:17LinusNor did i miss something?
16:00:32Zagorno if we have a simulated driver, the config file has nothing special. there will just be an #ifdef in the driver
16:01:27Zagorthis is more of a transition problem, since up to now the code has never been compiled with for example HAVE_MAS3587F and SIMULATOR at the same time
16:01:30LinusNthe config file tells which hardware/features exist on the target, SIMULATOR tells if it should be simulated or not
16:02:04LinusNgotta run
16:02:07 Part LinusN
16:02:07Zagor...which means you advocate leaving the config files untouched and solving it in the code?
16:02:08Zagorok byte
16:02:11Zagorbye even
16:02:49Zagori don't think I want to solve this right now, so I'll #ifdef the config files
16:03:29[IDC]Dragonconfig hell. I wonder how other projects do this, with a lot of variants
16:04:04Zagorpretty much like this. "config.h" is usually autogenerated by a configure script.
16:04:29[IDC]Dragonimho, we have way too many #ifdef SIMULATOR in the code
16:04:51[IDC]Dragonwhich indicates a bad (incomplete) simulation layer
16:05:12[IDC]Dragonor, not enough hardware abstraction
16:05:13Zagorwell we need simulated code. the other option is to have that code moved into separate files, but that means lots of code duplication instead. again: which is the lesser evil...
16:06:12Zagorwe have never aimed for hardware abstraction. it rarely works well.
16:06:49[IDC]Dragonwell, we have the driver "layer".
16:07:16[IDC]Dragonexchanging the driver with a sim that provides the same interface will work then,.
16:07:28Zagoryes, that is what we do
16:07:53[IDC]Dragoninteresting days for Rockbox
16:20:45 Join GhUl [0] (~tim@pD9FF1928.dip.t-dialin.net)
16:25:54 Join methangas [0] (methangas@0x50a476b2.virnxx10.adsl-dhcp.tele.dk)
16:34:18 Nick grogro is now known as gromit` (~gromit@ALagny-151-1-22-167.w83-114.abo.wanadoo.fr)
16:46:12 Join mecraw_ [0] (~lmarlow@69.2.235.2)
16:48:26***Saving seen data "./dancer.seen"
16:49:52 Join LePoulpe303 [0] (~lpos@AMontpellier-251-1-35-39.w83-113.abo.wanadoo.fr)
16:50:19 Quit LePoulpe303 (Read error: 104 (Connection reset by peer))
16:51:01 Join LePoulpe303 [0] (~lpos@AMontpellier-251-1-35-39.w83-113.abo.wanadoo.fr)
17:00
17:19:21 Part Zagor
17:20:02 Join PaulS [0] (~437e19f6@labb.contactor.se)
17:20:48PaulSIt's kinda sad that it apears that LinusN and I are never awake at the same time.. :-)
17:24:46 Quit PaulS (Client Quit)
18:00
18:04:41 Quit lImbus (Remote closed the connection)
18:05:47 Quit GhUl (Read error: 60 (Operation timed out))
18:06:34 Part [IDC]Dragon
18:30:16 Join lImbus [0] (~manuel@kernel.cycos.net)
18:30:31 Quit mecraw_ (Read error: 104 (Connection reset by peer))
18:31:30 Quit lImbus (Remote closed the connection)
18:35:21 Join lImbus [0] (~manuel@kernel.cycos.net)
18:42:32 Part pillo
18:47:50 Join mecraw [0] (~lmarlow@69.2.235.2)
18:48:27***Saving seen data "./dancer.seen"
18:51:05 Quit AciD (Remote closed the connection)
18:52:41 Join _aLF [0] (~alex@mutualite-3-82-67-66-128.fbx.proxad.net)
18:52:43_aLFhi
18:55:18 Part lImbus
18:55:41 Join AciD [0] (~acid@longchamp44-1-82-67-133-87.fbx.proxad.net)
19:00
19:25:50 Join dreed [0] (~45912877@labb.contactor.se)
19:25:56 Quit Lynx_ (" HydraIRC -> http://www.hydrairc.com <- irc client ownage!")
19:29:59 Join R3nTiL [0] (~zorroz@164-250-30-217.kgts.ru)
19:30:35 Quit dreed (Client Quit)
19:45:07 Join webguest26 [0] (~4128f37f@labb.contactor.se)
19:46:03 Quit webguest26 (Client Quit)
19:57:47 Quit R3nTiL ()
19:59:33 Quit ripnetUK ()
20:00
20:33:34 Quit _aLF (Read error: 54 (Connection reset by peer))
20:35:48 Join _aLF [0] (~alex@mutualite-3-82-67-66-128.fbx.proxad.net)
20:43:26 Quit AciD (Read error: 60 (Operation timed out))
20:44:58amiconnI am silly...
20:47:21amiconnNote to self: check both schematics and current source before getting excited about some finding
20:48:29***Saving seen data "./dancer.seen"
20:54:05 Join zeekoe [0] (zeekoe@zeekoe.kabel.utwente.nl)
20:54:31*zeekoe is away: UT
21:00
21:19:17*zeekoe is back (gone 00:24:53)
21:50:25 Join [IDC]Dragon [0] (~idc-drago@pD9E34188.dip.t-dialin.net)
21:50:41[IDC]Dragongood evening
21:50:51amiconnhello Jörg
21:57:35 Join SmoothOperator [0] (~acb978b4@labb.contactor.se)
21:59:26 Quit SmoothOperator (Client Quit)
22:00
22:05:43 Join elinenbe [0] (elinenbe_@207-237-224-49.c3-0.nyr-ubr1.nyr.ny.cable.rcn.com)
22:11:00 Join scott666_ [0] (~scott666@c-24-245-58-48.mn.client2.attbi.com)
22:20:01*[IDC]Dragon needs to sleep early tonight
22:20:05[IDC]Dragonbye
22:20:14 Quit [IDC]Dragon ()
22:35:08 Join AciD [0] (~acid@longchamp44-1-82-67-133-87.fbx.proxad.net)
22:48:32***Saving seen data "./dancer.seen"
23:00
23:05:59 Quit pike (Read error: 110 (Connection timed out))
23:08:43 Quit webmind (Remote closed the connection)
23:08:45 Join webmind [0] (~random@217-195-236-172.dsl.esined.net)
23:15:24 Join bagawk [0] (lee@IC62.library.oregonstate.edu)
23:28:44 Quit methangas (" HydraIRC -> http://www.hydrairc.com <- \o/")
23:35:17*zeekoe is away: busy
23:53:24 Quit AciD ("http://swpat.ffii.org/")
23:53:24 Quit scott666_ (Read error: 110 (Connection timed out))
23:53:25 Join scott666_ [0] (~scott666@c-24-245-58-48.mn.client2.attbi.com)
23:58:38*zeekoe is away: I'm busy
23:58:52 Join LinusN [0] (~linus@labb.contactor.se)

Previous day | Next day