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

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

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

#rockbox log for 2004-10-05

00:07:13amiconnZagor: The 256 sector ata limit has nothing to do with the fat code. That's why I consider this ugly. Imho ata_(read|write)_sectors should handle this
00:08:27Zagorit could, but that would only add extra code with no benefit
00:10:40 Join Servo888 [0] (
00:10:45Servo888afternoon all
00:11:39Servo888I've just ran into this project 1 minute ago. Does your firmware run on the iFP-190?
00:12:48Servo888ok - thanks a bunch
00:26:43 Quit _aLF ("Leaving")
00:26:49 Join RemakingEden [0] (
00:27:10 Nick RemakingEden is now known as Professor (
00:28:11 Quit AciD (Read error: 232 (Connection reset by peer))
00:31:13amiconnBagder: How is that SOURCES file mechanism supposed to work. I.e. what would I have to do if I want to exclude some source files for some targets?
00:32:25Zagorit's preprocessed, so add #ifdefs to it
00:32:52amiconnAh ok.
00:36:41amiconnThat means the compilation for new source files is no longer automatic, they have to be added to the SOURCES file, right?
00:38:20amiconnThat makes life a bit harder for first-time plugin developers. On the other hand, that allows to eg. exclude plugins for the wrong target completely. No more zero-sized .rocks :)
00:40:06Zagoryes :)
00:43:55amiconnDoes that mechanism work correctly for the simulator builds?
00:44:59Zagorno, the sims are not using this system yet
00:51:53Zagorhehe, a new mp3 player called "Rocbox"!
00:52:01***Saving seen data "./dancer.seen"
00:53:49amiconnDoes it use a SH1 CPU? ;-)
00:54:36Zagori doubt it. the company is called "Roc Digital" and is a branch off a hiphop record company called Roc-A-Fella.
00:59:16Zagorhmm, the dependencies have gotten worse with the new build system i think
01:03:30amiconnIn what respect? The source files itself are still the same now, and some #ifdefs that are in there today can now be removed.
01:04:08amiconnThe system does also save compile time, e.g. by not compiling the grayscale lib files at all for non-bitmap lcd.
01:05:06Zagoryes, but we don't calculate dependencies now. such as which .c files should be recompiled when a .h file has been modified.
01:05:17amiconnA propos grayscale: If someome with X11 programming experience could change the X11 sim to use a (grayscale) bitmap instead of pixel banging, I'd be happy to add grayscale support to the sim. The Win32 sim does already use a bitmap
01:06:41amiconn(dependencies) Hmm, that's indeed worse than before
01:20:26amiconnThere is nothing concerning the Ondio in the News section...
01:26:00Zagordo you feel it's time to announce it?
01:27:33amiconnI only thought about it because the iRiver is already mentioned 2 times, though that port is in a much earlier stage
01:28:24ZagorI'll add a blurb if you like
01:28:33Zagorany suggetion?
01:32:55amiconnHmm. Perhaps something like that rockbox is currently ported to the flash-based Ondio, and there is some success already. Something like the early announcements for rockbox (before 1.0 was out)
01:34:23amiconnMaybe is draws more attention from potential users if it is in the News section. We could need some hardware survey results (different hw masks etc.)
01:34:37amiconn*Maybe it
01:34:46Zagorwell, why not that then? a call for help.
01:35:04Zagorjust like we had for the iriver
01:36:33amiconnOkay. This could be linked to the wiki topic
01:39:01Zagorok, will you update the wiki page with what kind of help you need?
01:39:23 Join iRiverMan [0] (
01:40:44iRiverManhi zag
01:40:48 Quit mecraw ("Trillian (")
01:41:02Zagoramiconn: front page updated
01:41:18amiconnZagor: Ok, thanks! I'll modify the wiki page a bit
01:42:37 Join LinusN [0] (
01:43:41LinusNhi there
01:43:55Zagorwohoo, handle_shift() removed
01:43:56iRiverMando u have an iriver/
01:44:45LinusNiRiverMan: yes i do
01:45:06iRiverMangetting the knack of it/
01:47:31LinusNi'm running gdb on it as we speak
01:47:42iRiverManwhats gdb?
01:47:47LinusNi have just dumped the flash contents to a file
01:47:57LinusNgdb == gnu debugger
01:48:01iRiverManon-the-fly playlists!!!!!
01:48:11iRiverManshould be first perogative
01:48:13LinusNpatience grasshopper
01:50:10iRiverMani think the f/w size for the iriver is 1.8mb
01:51:14Zagoramiconn: can you test if my button fixes are good for ondio?
01:51:35amiconnZagor: Just a moment, currently editing the wiki
01:51:51Zagorsure, no rush
01:52:46 Join PaulS [0] (
01:53:11Zagorit's rather unfortunate that we're calling the browser context menu "onplay". many devices don't even have an on button.
01:54:54Zagori'm cruelly removing pgup and pgdn for players. i can't imagine anyone ever used them anyway.
01:55:14LinusNPaulS: are you interested in a dump of the ihp-120 flash?
01:56:58PaulSI'm not sure that I don't already have one, from the hex files...
01:57:18PaulSI suppose if there are differences between the hex dump and the flash image, I'd be interested.
01:58:31PaulSBrief Q, Linus −− I guess it's two in one: 1) Did you need to configure anything special for the BDM? I.e, apart from the signal name translations (TRST->DSTCLK) is the BDM interface more or less straightforward? I'm having trouble getting any kind of a response from the iRiver in either JTAG or BDM mode. I would have guessed based on the description that for a barebones BDM I should be able to ignore the PSTn pins, do a reset using the reset pin, th
02:00:46PaulSI also tried to use nBKPT which also didn't have the intended effect (halt).
02:01:12LinusNfirst of all, are you sure you have connected to the correct pads?
02:02:10PaulSYou tell me. :-) I used my "staring" approach, and I thought I knew which was which, but at one point you said that approach wouldn't work and that at some point you might make a picture.
02:02:14LinusNsecondly, you must hold BKPT low right from when you release RESET
02:02:26LinusNthe staring approach failed for me
02:04:02PaulSYeah, I was trying that. I think I've got RESET right, at least, since that pin will restart the bootup if held for ~20ms. Longer and it will stay off.
02:04:47LinusNyou want the correct pad for the TRST pin?
02:04:52LinusNhang on
02:05:37 Quit iRiverMan ("CGI:IRC (Ping timeout)")
02:06:10amiconnZagor: wiki topic updated, now testing button code
02:06:31PaulSI don't have a bare board to be absolutely certain of the other pins. I tried to use a combination of your pics and probing at the backsides of the vias, but that technique (esp with a populated board) is problematic. TRST, DSO(TDO), DSI(TDI), DSTCLK(TRST) and TMS(BKPT) are the ones I think I need.
02:06:45PaulS(Whoops, said TRST twice).
02:07:00Zagoramiconn: wait a minute, more updates coming
02:07:37PaulSOf course I can also wait patiently for you to do the documentation. I don't need to open my iRiver again soon.
02:08:06LinusNi can do a nice picture, but not today
02:09:00amiconnZagor: Do your latest changes fix the odd behaviour for the recorder/player/whatever too?
02:09:39Zagorhopefully :)
02:10:01PaulSLinusN: Not necessary. My question was about anything special you might have done. If it's just that I have the pins wrong, I can either solve the problem myself (go back to the vias) or just wait.
02:11:30LinusNno, i didn't have to do anything special
02:11:34LinusNone thing, though
02:12:02LinusNhow do you supply power?
02:12:16PaulSI'm using the battery.
02:12:19amiconnZagor: Now the main menu is callable from the browser, but only if you are fast enough: If the menu button is held down, the Ondio switches browser<->menu continuously...
02:12:56LinusNif you pull RESET, the power will disappear
02:13:03Zagorhmm, ok
02:13:16LinusNsince the port pin that controls the power gets tristated
02:13:20PaulSYeah, I noticed that if I held it > 20ms.
02:13:58amiconnZagor: There is still no way to stop playback from the wps. Only pause is possible
02:14:04PaulSThat's really the problem that foiled my JTAG efforts.
02:14:30Zagoramiconn: yes I know. but do you really need to stop? if so, on what button?
02:14:43LinusNsee the 6-legged IC next to the reset button?
02:14:45PaulS(Though it turns out that BDM will be a _lot_ faster.)
02:15:52Zagoramiconn: also it is strange that it leaves the menu immediately. it's only supposed to leave on menu press, not menu repeat
02:16:24amiconnHmm. It does...
02:18:50amiconnZagor: The original firmware pauses with the first press of Menu, and stops on the second. Un-pause is done with Right (=Play)
02:19:30amiconnA long press of Menu is the context menu. On/Off is unused in the "wps" well as in the browser
02:20:06LinusNPaulS: pin 4 of U28 (near the S in SW) controls the power, pull it up to VCC (i used 10k) to maintain power between resets
02:20:23Zagorthe key question is when do you use stop?
02:20:50amiconnAhem, when I want to stop playback?
02:21:10PaulSLinusN: Thanks!
02:21:18Zagoramiconn: why not pause?
02:22:30amiconnI have to admit that I use pause very rarely, most of the time I use stop. Rockbox resume function is great
02:23:23Zagorok so we remove pause and use stop instead. the point is do we really need both? we don't have buttons for both, without making a complex multilevel gui
02:23:58LinusNok, gotta sleep
02:24:01LinusNcu folks
02:24:07amiconnnite LinusN
02:24:22PaulSNite LinusN.
02:24:29 Part LinusN
02:25:28amiconnZagor: Hmm. Perhaps someone will complain if there is no pause. I think the Archos approach is not that bad for this single problem
02:26:04Zagori really don't like having pause and unpause on different buttons
02:28:26amiconnHmm, I wouldn't have a problem with that. Imagine that in the not-so-far there will be a device that features separate play and pause buttons.
02:29:02Zagoryes, but then it will be obvious because the buttons are marked. anyway, i've got to go to bed now. let's discuss this more tomorrow.
02:29:02amiconn*not-so-fat future
02:29:22amiconnPerhaps I should do the same.
02:29:32 Part Zagor ("Client exiting")
02:34:25 Part amiconn
02:49:30 Quit Chronic007 (Read error: 104 (Connection reset by peer))
02:52:04***Saving seen data "./dancer.seen"
03:15:18 Join telliott [0] (
03:18:38telliottWhat does PLAY+REPEAT mean?
03:31:21 Quit telliott (Read error: 104 (Connection reset by peer))
03:59:07 Join ashridah [0] (ashridah@
04:03:47 Quit PaulS ("CGI:IRC (EOF)")
04:16:13 Part scott666
04:40:47 Quit Servo888 ("using sirc version 2.211+KSIRC/1.3.11")
04:52:08***Saving seen data "./dancer.seen"
05:15:00 Join ashridah_ [0] (
05:23:36 Quit ashridah (Connection timed out)
05:35:30 Nick ashridah_ is now known as ashridah (
06:52:11***Saving seen data "./dancer.seen"
07:09:26 Join AciD [0] (
07:12:55 Join wake [0] (
07:15:24 Quit wake (Client Quit)
07:18:20 Join LinusN [0] (
07:23:09 Join wake [0] (
07:26:16 Quit dwihno (Read error: 113 (No route to host))
07:33:54 Quit wake ("leaving")
08:07:03 Join dwihno [0] (~dw@
08:14:32 Join Zagor [242] (
08:20:30ZagorBagder: are you here?
08:22:04 Join [IDC]Dragon [0] (
08:31:17Zagor[IDC]Dragon: can you test the new button code on ondio? amiconn reported a strange issue with the menu yesterday
08:31:36[IDC]Dragonyes, saw that
08:31:55[IDC]Dragonshould it be different here/now?
08:32:12Zagorno, but we need to examine it more
08:32:42Zagoroh, found it...
08:33:55[IDC]DragonI'm about to build anyway, just decorating the panic
08:35:57 Join MisticJeff [0] (
08:39:07 Join amiconn [0] (
08:40:04amiconnGood morning
08:41:01[IDC]DragonJens: should I remove the main menu hack?
08:42:51amiconnYes, if the menu now works correctly (Zagor?)
08:43:01amiconn[IDC]Dragon: Just a quick one: I traced the transfer() write accesses as well. Results: sec 264, then 264...279 as single sectors, then 680 (16 sectors), then 168 (112 sectors) -> panic
08:43:05Zagorit should. try it.
08:43:48amiconnZagor: There is now an issue with the On+Play menu, at least on recorder. It disappears as soon as you release Play
08:44:17[IDC]Dragonbuilding now...
08:44:22Zagornot for me
08:44:42LinusNZagor: have you tried both with play-repeat and on+play?
08:45:13LinusNreleasing the buttons in different order?
08:45:42Zagortrying all kinds of combos, can't get it to happen
08:47:23[IDC]Dragon(ondio) I can enter the menu
08:47:42[IDC]Dragonholding it is stable
08:47:51amiconn[IDC]Dragon: Sectors 264...279 is root dir (first pseudo-cluster), not fat
08:47:56amiconnGotta run
08:52:14***Saving seen data "./dancer.seen"
08:54:08Zagor[IDC]Dragon: did you read our pause/stop discussion yesterday? any opionions?
08:54:35Zagorondio does not have enough buttons for both in wps
08:55:05Zagorunless you make it a key sequence, which is what archos did
08:55:38Zagori personally think stop is overrated
08:55:40LinusNstop shouldn't really be necessary
08:56:04Zagoramiconn didn't agree with that though
08:56:30LinusNi can't make the onplay menu disappear on play-release
08:56:41LinusNworks perfect for me
08:56:58LinusNi like the play-repeat invocation
08:58:23LinusNi'm normally not fond of using held keys like that, but i guess it's a necessary evil with that few keys
08:58:55[IDC]Dragon(pause/stop) this is really a drag
08:59:17[IDC]DragonI'm fine with pause for now
08:59:46[IDC]Dragonah, no
08:59:56[IDC]Dragonthe voice UI won't work then
09:04:19[IDC]Dragonhow about on/off+repeat for stop?
09:04:35Zagorhow long will actually turn if off?
09:04:54[IDC]Dragonhardware-wise 10 sec
09:05:07[IDC]DragonRockbox after 4 sec or so, iirc
09:06:21Zagor40 key repeats, which got a lot shorter lately :)
09:06:32Zagorbut let's try it. i'll add it.
09:07:42[IDC]Dragonwith my file system bug, I haven't really arrived in the land of UI discussions :-/
09:09:06*LinusN should continue what he started and implement the SYS_POWEROFF event...
09:09:34[IDC]Dragonyes, that would be a nice thing.
09:10:25[IDC]Dragonsince the hardware timeout is so far away for the ondio, itcould write the resume pos (congig in general) then
09:10:56[IDC]Dragonsame perhaps for RTC models
09:11:25Zagorcrapola. scsi0: Unexpected busfree in Message-out phase
09:11:37Zagorwe need to change disks in the server
09:20:05[IDC]Dragondid I mention that a flashed Ondio starts Real Quick (tm)
09:20:24Zagorheh, I can imagine
09:20:50[IDC]Dragonthe browser is there in ~1 sec
09:23:09Zagorhow fast is the original firmware?
09:23:57[IDC]Dragonpretty slow, like on JB, before they start doind something "useful"
09:24:35[IDC]Dragoncan't really tell, because my boot loader cuts it already
09:24:55[IDC]Dragonwe'll have to ask Jens
09:25:11[IDC]Dragonor I need to restore the original
09:26:31Zagornah, I was just wondering if it too was faster than the jb firmware
09:30:01LinusN[IDC]Dragon: you don't really use BYTE_FLASH and BYTE_FLASH16 in minimon, do you?
09:33:29[IDC]Dragonfor flashing, I do
09:35:11Bagderwe don't have any config-* define for "flashable", right?
09:35:56[IDC]DragonBagder: no. Everything with a proper chip is flashable
09:35:58LinusNyour uart_boot code performs the flash sequence itself, it doesn't send the FLASH_xxx commands to minimon
09:36:22*Bagder edits the SOURCES files to only build relevant files
09:36:37LinusNBagder: great
09:37:14[IDC]DragonLinusN: mine does...
09:37:38[IDC]Dragondo you have an ancient version?
09:38:42[IDC]Dragonin the very beginning, I did it with all byte writes, but that was slow as hell, because of all the address phases
09:39:36ZagorBagder: dependencies don't work anymore
09:40:22Bagderhow silly
09:40:32Bagder"/home/daniel/src/rockbox/build-recorder/bookmark.o.o: bookmark.c"
09:40:36LinusN[IDC]Dragon: i'm reading the CVS version
09:41:44[IDC]DragonLinusN: client.c has a function FlashByteMultiple()
09:41:59LinusNah, see it now
09:42:04LinusNwas looking in flash.c
09:43:00[IDC]Dragonthe ID reading, and erase, yes, thats done "by hand" (not worth optimizing)
09:43:27LinusNwas looking at Programbytes()
09:43:30[IDC]Dragonyou're looking at how to wiggle the flash of the iriver?
09:43:39LinusNno, i'm porting the minimon
09:44:00BagderZagor: fixing it now
09:44:31[IDC]Dragonhehe, my little minimon gets famous
09:44:32LinusNand i need to implement FLASH_WORD
09:44:49[IDC]Dragon16 bit wide flash?
09:44:57[IDC]Dragonfeel free
09:45:05LinusNi feel very free
09:45:25dwihnoFree like a flying daemon?
09:45:28Bagderyou even look free from here! ;-)
09:45:36LinusNeuro 4ever
09:45:48dwihno\o/ Linus \o/
09:46:14Bagdertime for a coffee
09:47:07[IDC]DragonLinusN: are you porting uart_boot, too?
09:47:36[IDC]Dragonthat never made it to linux, although prepared
09:49:24 Join kurzhaarrocker [0] (
09:51:21*kurzhaarrocker had no trouble with a 6h recording using an acient rockbox back from jan 04.
09:52:43 Part kurzhaarrocker
09:52:59Zagorwow, that was a quick visit
09:56:04LinusN[IDC]Dragon: uart_boot doesn't work the same way with the iriver, but we need something similar, yes
09:56:49 Join amiconn_ [0] (
09:59:22Bagderno more gray* builds on the player
10:01:13Bagderthe #ifndef HAVE_MMC in ata.c seems wrong
10:01:39Bagderwouldn't it rather depend on the presense of something?
10:02:00Zagorit's not necessary at all with the new SOURCES system
10:02:39Zagorme fix
10:02:56BagderI would rather have it ifdef [feature]
10:03:03Bagderrather than ifndef [another feature]
10:03:14Zagor#ifdef HAVE_MMC driver/ata_mmc.c #else driver/ata.c #endif
10:03:28Bagderok then ;-)
10:10:28 Quit [IDC]Dragon (Read error: 110 (Connection timed out))
10:13:45 Quit MisticJeff ()
10:14:12 Quit amiconn (Read error: 110 (Connection timed out))
10:14:12 Nick amiconn_ is now known as amiconn (
10:30:37 Join [IDC]Dragon [0] (
10:31:13[IDC]Dragonhi again, did I miss something (log doesn't work)
10:31:46LinusNlog works fine
10:33:28[IDC]DragonIE shows it, firefox doesn't
10:33:43*Bagder tries
10:33:51[IDC]Dragonjust here, I'd say
10:34:04Bagderprobably, yes
10:34:36LinusNmozilla/firefox has a silly approach to .txt files in URL:s
10:34:37[IDC]Dragonsome local quirk, I get a very fast error message, it didn't realyy go out and try hard
10:34:51BagderLinusN: on windows I bet?
10:35:18amiconn[IDC]Dragon: Log works fine here
10:35:20LinusNit examines the data and tries to be smart how to interpret it, sometimes it decides that the data is binary
10:35:34LinusNon linux as well
10:35:36Bagderah, right, that one
10:36:06[IDC]DragonBagder: (ifndef HAVE_MMC) that was to mutually exclude ata.c vs. ata_mmc.c
10:36:27BagderI just think it looked odd
10:36:57[IDC]Dragonwhith your new system, we can make an ondio subdir and put it there, excluding the normal ata.c in SOURCES
10:37:24Bagderzagor already made an ifdef - else in SOURCES
10:37:35[IDC]Dragonit sure looks odd, but we didn't have your source composer until yesterday ;-)
10:38:19[IDC]Dragonso one file got the ifdef and the other the ifndef
10:44:09amiconnBagder: 2 comments regarding docs/HISTORY:
10:45:12amiconn(1) It wasn't me who invented rombox, I did only debug it. It was Jörg's invention, and was prepared since long in his bootloader
10:45:50Bagderfeel free to update/correc/fill in
10:46:15amiconn(2) The timescale is wrong, we finally got it running this year, and not 2003
10:46:32Bagderoh right
10:46:44Bagderwrong july
10:46:54amiconnAdditionally, the Ondio is not yet mentioned
10:47:12Bagderjump in, fill in, commit! ;-)
10:48:06*[IDC]Dragon didn't dare to credit himself
10:48:46[IDC]DragonJens did a good debugging job on rombox, and all that const'ing
10:52:15***Saving seen data "./dancer.seen"
10:53:15[IDC]Dragonaniconn: I tried rom dump on the real ondio this morning, it just doesn't do anything. Like it won't open the file, perhaps.
10:54:24amiconnStrange. It works for me on FAT32, on FAT16 I get 2 files with zero length (and sometimes a corrupt file system)
10:54:53[IDC]Dragonah, perhaps I got 0 length, too
10:55:03[IDC]Dragonbut no panic
10:55:53[IDC]Dragonof course I only tried FAT16, for that bug, and with a new panic condition for write behind the end
11:00:21amiconnNo panic, yes.
11:02:13amiconnDo you have a working MMC available now?
11:02:58[IDC]DragonI can probably find one here, if I really need
11:03:43amiconnI hope to fill in the MMC debug menu item today, then I would be interested in the parameters from various MMCs
11:05:43[IDC]DragonI can gather what's here
11:59:24 Join iRiverMan [0] (
12:00:27 Quit iRiverMan (Client Quit)
12:17:10amiconnBagder: (HISTORY) Is there a reason why sometimes the full name is mentioned, and sometimes the first name only?
12:20:29Bagderno, just sloppiness
12:34:10 Join MooMaunder [0] (~me@
12:44:34Zagoramiconn: are you still being exited from onplay on play|release ?
12:44:53amiconnZagor: Can't test atm, will do so asap
12:52:18***Saving seen data "./dancer.seen"
12:52:50[IDC]Dragonchanging history, hehe
13:40:54[IDC]Dragonthe FAT16 problem is a mystery to me
13:41:13Zagorwhat is the problem?
13:41:16[IDC]DragonI can't "unrail" it with the test code
13:41:38[IDC]Dragoncode-staring also didn't help
14:00:30 Join R3nTiL [0] (
14:01:26 Quit R3nTiL (Read error: 104 (Connection reset by peer))
14:23:37amiconn[IDC]Dragon: I suspect some variable initialization problem. You did add some members to the fat_file struct, didn't you?
14:27:50amiconnPerhaps it's just a coincidende, but 680 (first sector) - 168 (wrong sectornum) == 512
14:30:17 Join elinenbe [0] (~elinenbe_@
14:33:08[IDC]Dragonamiconn: I added members, but the whole struct is memsetted to 0
14:33:29[IDC]Dragonso it should be consistent across platforms
14:34:25 Join _Scrat_ [0] (
14:34:32_Scrat_Hi guys
14:35:00_Scrat_I just wanted to ask a simple question...
14:35:14_Scrat_about your awesome archos project
14:35:38_Scrat_Do you plan to change anything to the just released gmini 400 ?
14:36:25Bagderchange anything?
14:36:27LinusNno plans at all
14:36:28[IDC]Dragon_Scrat_: it's not working on the gmini400, and very likely this won't change
14:37:01Zagor_Scrat_: we don't change firmware, we write it
14:37:23_Scrat_oups you are right
14:37:48_Scrat_sorry, did you plan to release a firmware for it then ?
14:38:23Zagorno. it's not one of the platforms we are working on
14:38:40Bagder_Scrat_: feel free to start working!
14:39:08_Scrat_ok. No other known projects on that ?
14:39:15 Quit MooMaunder ()
14:39:32Zagorthere are some people looking at the gmini 120, but not the 400 afaik
14:39:48[IDC]Dragonthere's work on the 3x0 models, which may have similar hardware
14:40:03[IDC]Dragon(the linav and avos projects)
14:40:12_Scrat_Do you thing archos will add OGG support ?
14:40:27Bagder_Scrat_: no
14:40:28Zagor_Scrat_: no
14:40:48_Scrat_ok guys, thanks for your time
14:40:54[IDC]Dragonthe 400 uses a TMS320DM270
14:41:26Zagorsounds like the linav/avos guys could be interested then
14:41:31[IDC]Dragonwhich one was it on the 3x0 model again?
14:43:47[IDC]Dragon_Scrat_: so go there, tell them ;-)
14:45:44 Quit _Scrat_ ()
14:49:12amiconn[IDC]Dragon: I could try to display more variables on the problematic call to fat_readwrite() and transfer().
14:49:58[IDC]Dragonthat could help, yes
14:50:16amiconnIt's a pity that the player display is so small. Is it hard to do the serial mod?
14:50:41[IDC]DragonI'd like to get a test setup, but it'll take a while
14:50:44amiconnAlthough this would require that I get an usb->serial converter. Then I could try to run gdb
14:50:57[IDC]Dragondo you have a level converter?
14:51:13amiconnNope, but this shouldn't be hard, right?
14:51:45[IDC]Dragonminimally, 2 transistors, some diodes resistors
14:52:02[IDC]Dragonor a MAX chip
14:52:20***Saving seen data "./dancer.seen"
14:52:40[IDC]DragonI did the transistor kind, crammed it all in the SUB-D shell
14:53:47amiconnWith transistors, it's easier than the spdif converter (usually I don't have that TORX chip laying around, but a bunch of transistors)
14:54:01[IDC]Dragonthe nice think about that solution is that it need no external power
14:55:41[IDC]Dragonthing, needs
14:58:05elinenbeLinusN: you getting excited working on the iriver?
14:59:10LinusNit's fun
14:59:28LinusNZagor: you have broken a few things in tree.c with your button changes
14:59:38BagderI like the physical size of the iriver
15:00:02ashridah"fits right in my pocket"</grampa simpson>
15:00:08LinusNthe "PLAY to resume" and the "PLAY to rolo" dialogs both fail
15:00:14[IDC]DragonBagder: get an Ondio ;-)
15:00:27Bagderyeah, about the same storage capacity ;-O
15:00:40*ashridah isn't sure he's seen an ondio in stores here in .au
15:01:10BagderI think its too bad the 2.5" disks don't come in big enough sizes
15:01:13[IDC]DragonBagder: the get a matchbox worth of MMCs with it ;-))
15:01:30ZagorLinusN: fail how?
15:01:39[IDC]Dragon60 GB on 2.5" is rumoured
15:01:46Zagoryou mean 1.8"?
15:01:53[IDC]Dragonah, yes
15:01:59BagderI meant 1.8 yes
15:02:06[IDC]DragonI just copied
15:02:08LinusNZagor: they wait for one key event, and quits on all other
15:02:28LinusNand since the first event doesn't contain BUTTON_REL, they "fail"
15:03:55[IDC]Dragonthe Ondio doesn't even have a Play button...
15:05:18Zagorit uses TREE_RUN, not play. feel free to change the text :)
15:06:17LinusNZagor: TREE_RUN is a release event, and those two dialogs quit before the release event arrives
15:06:28Zagori'm fixing now
15:06:56LinusNbtw, the filetypes system doesn't seem to allow two plugins for the same file type :-(
15:06:58[IDC]DragonZagor probably regrets opening the pandora box of keys
15:07:36Zagorno, these minor issues were expected. the button code has gotten significantly simpler.
15:07:40BagderLinusN: no, how could it run who plugins for the same one?
15:07:58Bagderone after the other?
15:08:03LinusNin the "open with..." dialo, i can select which one to run
15:08:14Zagorthat worked before
15:08:33Bagderbut that menu shouldn't care about what extensions they are registered for
15:08:39Zagoror maybe i'm confused :)
15:08:44LinusNnow i can't have "favorites.rock" and "vbrfix.rock" in the "open with" menu
15:09:08LinusNor is there a maximum number of menu entries?
15:09:20Zagoropen with... shows *all* viewers, doesn't it?
15:09:28Bagderbecause it only adds viewers that are regged for an extension?
15:09:29LinusNi see it
15:09:43[IDC]Dragonsince we have no malloc, there has to be a limit
15:09:55LinusNit had nothing to do with the extension, there was a maximum of 8 entries
15:09:55Bagdereven with malloc, there would be a limit
15:10:10Bagderseems a bit low
15:10:42Zagorbtw, do you feel we should add F3 as a third way to get to the onplay menu? or should we remove one or two ways?
15:11:20[IDC]Dragon3 ways sound like a luxury
15:11:21Zagorthe wps context menu will only be on F3, so it would add consistency. however removing it from on+play would make a lot of people confused
15:11:23midkZagor, what two ways exist?
15:11:32Zagoron+play and play+repeat
15:11:34midki must have missed something
15:11:41LinusNf3 doesn't exist on all platforms
15:11:43midkplay+repeat sounds best
15:11:45Bagderplay+repeat is brand new
15:11:47[IDC]Dragonwhat happened to the configurable menus?
15:11:58ZagorLinusN: right, i'm only talking recorder here
15:13:04LinusNconfigurable menus haven't exactly been top priority
15:13:27Zagor[IDC]Dragon: they're still in their can, with the other worms :)
15:15:34ZagorBagder: yes it is. the question is should we move it from there to f3? i'm not so sure since play+repeat is much more likely to be a cross-platform key while f3 is strictly recorder. so it feels like we're going to have three ways...
15:15:52Bagderyeah, three ways it is
15:15:57Zagorif we want f3 to be context menu
15:16:52Zagoror should we use play+repeat in wps too...
15:17:01Zagorvastly more cross-platform
15:17:02LinusNgets my vote
15:17:19ZagorLinusN: then you get to name the key to use for ondio :)
15:20:44[IDC]Dragonno play, no onplay, no F3 :-(
15:21:37Zagorwe could use menu+right for onplay
15:21:51midkor menu+play? :)
15:21:53[IDC]Dragonhow about right repeat?
15:21:55LinusNwhat is TREE_RUN on Ondio?
15:22:15LinusNright repeat is FF in the WPS
15:22:33[IDC]Dragondoes the WPS have onplay?
15:22:40Zagormidk: there is no play key on the ondio
15:22:41LinusNit soon will
15:23:00midkmenu+right then i suppose
15:23:33[IDC]Dragonit's problematic to use menu as shift, if it already is used for long press
15:24:19[IDC]Dragonyou'd have to press them in a very tight sequence
15:24:20Zagoractually context menu (onplay) is right+repeat right in browser now on ondio, but should be changed if we want to be consistent in tree and wps
15:24:21Zagorwow, that came out strange... :)
15:24:51Zagoralthough not terribly tight. repeat is almost a second.
15:24:56Zagorrepeat delay
15:25:29Zagor...and we thought we had UI problems with the players ;)
15:26:10[IDC]Dragonsame # of buttons
15:26:43Zagoryes, but the off button is not practical to use for combos
15:26:54 Join Sp|rit-B0y [0] (~someone@
15:27:16[IDC]DragonI find combos impractical in general
15:27:22LinusNme too
15:28:26Zagoryeah but you can't have the cake and eat it you know
15:29:37[IDC]Dragonlong presses give us a 2nd layer as well, and don't "eat" a shift key
15:30:08Zagoryes but they are all used up in the wps (up,down,left,right,menu)
15:30:28Zagoreven off
15:31:36[IDC]Dragonhmm, you're right
15:32:12Sp|rit-B0yhello, i was just wondering when abouts the rockbox project for iriver will be ready, December? January? I'm really excited that someone finally decided to do s'th about the crappy original firmware from iriver which they lag to update.
15:33:10ZagorSp|rit-B0y: we don't make promises or deadlines.
15:33:39Zagoralso, software is never ready. it's just released.
15:33:56LinusNwe can't make any promises, since we dont know how long it takes
15:34:01[IDC]Dragontalking about "ready", it will never be ;-)
15:35:09Sp|rit-B0yyeah i'm talking more about the first release to the general public, do you expect it to take more than 3-4 months until that though?
15:35:11LinusNwe all do this on our spare time, and spare time is a scarce resource
15:35:23LinusNdon't expect anything this year
15:35:24Bagderwe don't expect
15:35:30Bagderwe'll just see
15:35:56Sp|rit-B0ythanks for putting your time into it though :P, iriver has been delaying their stuff for ages
15:37:17 Quit Sp|rit-B0y ()
15:38:14*LinusN feels the pressure to finish the boot loader
15:39:06 Join methangas [0] (
15:39:58[IDC]DragonLinusN: where do you want to place it, how should it work?
15:40:19LinusNi'll do it similar to the archos loader
15:40:27dwihnoLinus: Less firmware, more kebab :)
15:40:34LinusNcheck for a character on the serial port
15:40:51LinusNthen load the firmware from rom or disk
15:41:17[IDC]Dragondisk? sounds tough
15:41:39[IDC]Dragonlots of s/w has to work for that
15:42:28LinusNthe same amount of software has to work to reflash it the normal way
15:42:35[IDC]Dragonbut I mean, where's your entry point into the iriver? Do you want to nil the flash by bdm?
15:42:58LinusNi don't get it
15:43:23[IDC]Dragontoss away the whole content, build up your loader?
15:43:42LinusNi guess i'll do a dual-boot first, like we do on the archos
15:43:53LinusNwith a UCL packed original
15:44:10[IDC]Dragondoes iriver run the code from RAM?
15:44:25[IDC]Dragonsounds familiar then ;-)
15:45:07LinusNanyway, i'm not going great lengths to keep the original firmware
15:45:11[IDC]DragonI'm just missing the first steps to break in. With Archos, they already have a bootloader, which I could exploit.
15:45:52[IDC]Dragonso you wipe the flash with it, place your stuff?
15:46:04LinusNnot necessarily
15:46:25LinusNi can use gdb to load minimon, and use minimon for flashing
15:46:39LinusNi have several options
15:46:56[IDC]Dragonthat's always nice
15:47:24LinusNsince i can run gdb with the bdm, i don't have to flash until i feel secure about it
15:47:45[IDC]Dragonyes, neither did I
15:48:18[IDC]DragonI used minimon to load and start rockbox, the cold way
15:48:32amiconn[IDC]Dragon: Talking about original firmware and dual boot, and as you are the "port init" guy: Although RoLo already works on the Ondio, it does only work for rockbox, the archos firmware does 2 strange things.
15:49:35amiconn(1) It loads, but the lcd contrast is wrong (too weak). (2) When you try to play music, it freezes instead
15:50:22[IDC]Dragonlcd contrast sounds like mask values, but that's not influenced by ports
15:51:01[IDC]Dragonmusic is probably some MAS pin being wrong
15:51:41[IDC]DragonI'll deal with that once the file write is ok :-/
15:52:13[IDC]Dragonhow about an open issues section on the wiki page?
15:52:18amiconnI wonder how the original fw could read the mask wrong, as it is in rom...
15:52:32amiconn(open issue) Nice idea!
15:52:45[IDC]Dragonso we don't forget
15:53:11[IDC]Dragonusing the bug tracker is probably a bit overpowered in the early stage
15:56:27LinusNtime to go
15:56:30LinusNcu lads
15:56:32 Part LinusN
16:11:00amiconn[IDC]Dragon: There is a new MMC 4.0 standard (well, not that new. It was completed Jan 2004), supporting faster alternative transfer modes. I suppose it leverages the 4 GB barrier as well
16:11:23amiconnHowever, I couldn't find a copy describing that standard yet.
16:18:14[IDC]Dragonnor a card, perhaps
16:18:47[IDC]DragonMMC is supposed to be an open standard, as opposed to SD
16:19:19Zagor"open standards" often doesn't mean the specifications are publicly available
16:20:26amiconn[IDC]Dragon: The mmc standard is only "sort of" open, if you want the complete specs, you have to pay for it
16:21:21[IDC]Dragonfor new features, I only see multi bit serial, higher clock, and 1.8V
16:21:21amiconnThere are already mmc 4.0 cards available (1 GB and 2 GB, by Pretec)
16:21:38[IDC]Dragonno new addressing commands
16:22:08amiconnYes, but maybe this enhancement is hidden in the details
16:24:17amiconnThe barrier was not yet hit by the available cards (but it is foreseeable)
16:25:24 Join cRuNcHiE [0] (
16:25:54cRuNcHiEhey guys, just fixed my mates archos recorder 20gig (was giving hdd error)
16:26:26cRuNcHiEjust put rockbox on it though cos i used to use it when i had an archos, its giving 'error accessing playlist control file' whenver i choose to play an mp3 thoug
16:26:36cRuNcHiEit plays the mp3 fine though :)
16:27:56cRuNcHiEah just spotted sommit ;)
16:28:14cRuNcHiEnever had to extract whole zip in my archos days :)
16:30:11cRuNcHiEi used to have an old recorder which i modded to have 40 gig drive.. but then the lcd got cracked and at that time noone sold replacements... :(
16:30:30cRuNcHiEso not long ago i chucked thing fecker.. then the gits brought out the replacement! grrrr
16:32:56cRuNcHiE<not impressed :(
16:33:36cRuNcHiEhad been in my drawer broken for 2yrs, then finally thought ah well bin here it comes... then literally month later they brouhgt out replacements
16:52:21***Saving seen data "./dancer.seen"
16:54:27 Join mecraw [0] (~lmarlow@
17:00:18 Quit [IDC]Dragon ("CGI:IRC")
17:11:18 Join Rand [0] (
17:16:19 Quit Rand ("CGI:IRC (EOF)")
17:18:30 Part Zagor
17:27:52 Quit ashridah ("sleep")
17:40:51 Quit cRuNcHiE ()
17:45:49 Join gromit`` [0] (
17:46:33 Quit gromit``` (Read error: 110 (Connection timed out))
17:48:39 Quit elinenbe (" HydraIRC rocks! -> <-")
18:18:32 Join marc [0] (
18:20:45 Quit marc (Client Quit)
18:21:01 Join marc [0] (
18:21:30 Quit marc (Client Quit)
18:21:56 Join marcl [0] (
18:23:34 Join webguest53 [0] (
18:41:36 Quit webguest53 ("CGI:IRC")
18:52:25***Saving seen data "./dancer.seen"
18:59:19 Join nelliep [0] (
19:00:31 Quit marcl (Read error: 104 (Connection reset by peer))
19:00:50 Join marcl [0] (
19:03:31 Join [1]marcl [0] (
19:03:31 Quit marcl (Read error: 54 (Connection reset by peer))
19:03:35 Nick [1]marcl is now known as marcl (
19:10:22 Quit nelliep ("CGI:IRC")
19:20:32 Join [1]marcl [0] (
19:20:33 Quit marcl (Read error: 54 (Connection reset by peer))
19:20:36 Nick [1]marcl is now known as marcl (
19:33:21 Part marcl
19:53:22 Quit Sebulba03 (Success)
20:16:03 Join webguest21 [0] (
20:19:08 Quit webguest21 (Client Quit)
20:19:34 Join webguest91 [0] (
20:20:15 Quit webguest91 (Client Quit)
20:52:28***Saving seen data "./dancer.seen"
21:06:53 Join amiconn_ [0] (
21:09:03 Quit amiconn (
21:09:15 Nick amiconn_ is now known as amiconn (
21:52:11 Join scott666 [0] (
22:19:18 Quit midk (Read error: 110 (Connection timed out))
22:27:00 Join Sebulba03 [0] (
22:29:56 Join LinusN [0] (
22:30:11amiconnhi LinusN
22:33:17amiconnMore key handling woes :(
22:34:44LinusNwhat now?
22:35:30amiconnWhile the On+Play menu itself does work now, some menu items do not: On+Play (or Play+Repeat) on a file -> menu. Select "Delete". The "play to confirm" question disappears as soon as the Play button is released again (from pressing it to get there)
22:36:31amiconnTo be precise, it's the only item that does not work
22:38:01LinusNme fix
22:39:08amiconnAnother oddity: If I use OnPlay on a playlist, then select playlist->view, sometimes the playback starts on exit from the viewer
22:39:25LinusNstray release event, as usual
22:40:10amiconnMaybe it does start already when entering the viever (currently no headphone attached to the recorder)
22:40:29amiconnStill likely a stray release event
22:41:52LinusNwe should have a standard dialog code for "are you sure" type questions
22:52:32***Saving seen data "./dancer.seen"
22:53:03 Join webguest98 [0] (
22:54:17webguest98can anyone tell me if my recorder 20 with grey bumpers will accept the recorder 20 with blue bumpers firmware or is there no difference between the two other then the color of the bumpers
22:57:29amiconnwebguest98: If both are v1 recorders (identical shape) there is no other difference than the bumper color
22:58:02 Quit webguest98 (Client Quit)
22:58:21 Join webguest98 [0] (
22:59:07 Quit webguest98 (Client Quit)
23:15:15 Quit methangas (" HydraIRC -> <- irc client ownage!")
23:15:48LinusNwee, i just enabled the backlight on my iriver, via gdb
23:20:15scott666how are the new ports coming along?
23:20:41amiconnLinusN: Why do you set mas InterfaceControl to 0x01 for the better spdif signal strength? Iiuc this does also enable SDO, which is not used I think setting it to 0x09 would do. Furthermore, the comment above it is now incorrect.
23:28:48 Join Sebulba02 [0] (
23:30:41 Part Sebulba03
23:48:13LinusNamiconn: silly me, i meant to disable SDO
23:48:20LinusNthanis for the heads up
23:49:12amiconnYou see: I had to deal a lot with the mas datasheet in the last time
23:54:32amiconnWe could disable spdif output completely on platforms that don't have it (Ondio FM/SP). This mights save a bit of power, but would require an additional config-xx.h define
23:58:53 Join iRiverMan [0] (

Previous day | Next day