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 2004-10-01

00:01:50 Join LinusN [0] (
00:02:17amiconnLinusN: ??
00:02:17*LinusN is wiggling!
00:02:48amiconnLinusN: You own a player, right?
00:03:14amiconnIs it normal that it makes buzzing noises while charging?
00:03:26LinusNsome of them do, some don't
00:03:33LinusNmine doesn't
00:03:44LinusNare the noises loud?
00:04:44amiconnThey are not loud, but quite noticeable when sitting at the quite desk, with the box in ~1 m distance.
00:05:14amiconn(the second occurrence)
00:05:59LinusNwhich charger?
00:07:03amiconnThe one that came with the box. Dunno if it is the original one, since I got my Studio10 from eBay. It's a bit different from my Recorder's charger
00:07:55 Join bagawk [0] (Lee@bagawk.user)
00:08:25LinusNKings Electronics
00:08:36bagawkthey made the charger i think on the old players
00:08:42LinusNthe original charger is marked with that
00:08:46LinusNis it regulated?
00:09:27 Quit AciD (Read error: 104 (Connection reset by peer))
00:09:31amiconn(1) My recorder's charger is made by Kings Electric (2) Both chargers are unregulated (transformer type, not switchers)
00:14:40***Saving seen data "./dancer.seen"
00:17:15 Join webguest67 [0] (
00:21:22webguest67anybody here?
00:21:53webguest67is there any way to tell what build I'm running?
00:22:15webguest67but what about daily builds?
00:22:25bagawkwebguest67: then the build you are using is 2.2
00:22:25webguest67is there a build code?
00:22:36 Join ripnetUK [0] (
00:22:40webguest67for each build in sequence?
00:22:55bagawkthen the daily build have a versoin of 20040903-1234 or so
00:23:14webguest67where would i check that?
00:23:41bagawkwebguest67: at boot time, when the logo shows up, and it is at the bottom
00:24:10amiconnwebguest67: Or from the menu: Info->Version
00:24:55webguest67amiconn: all i see there is "2.2"
00:25:02webguest67then credits roll
00:25:10LinusNwebguest67: then you are running 2.2
00:25:52 Join Val-NM [0] (
00:25:54webguest67The reason I ask is because I *may* have seen a bug. :(
00:26:21 Quit zeekoe (Read error: 110 (Connection timed out))
00:26:24bagawkwebguest67: explain
00:26:33webguest67I was recording, and when I stop and start a new recording, the bytecount doesn't zero.
00:26:53webguest67Surely that isn't intended. (?)
00:27:17webguest67The time started over at zero, just not bytes.
00:27:33bagawkwebguest67: not sure
00:27:42LinusNdid you really stop it in between?
00:27:43ripnetUKdid i hear linusN say wiggleing has commenced :-)
00:27:47webguest67No big deal, just wanted to mention it.
00:27:49bagawkwebguest67: you could always use a daily build
00:27:53LinusNripnetUK: YES!
00:27:57bagawk*lots* of things have changed since then
00:28:12ripnetUKthats great news well done
00:28:31webguest67OK, maybe I'll give it a try. Is that relatively low-risk?
00:28:54LinusNah, the binutils project has applied my patch in their CVS
00:29:01bagawkwebguest67: yes, no risk at all
00:29:19bagawkwebguest67: just download the zip file for the daily build, and extract to your archos
00:29:48bagawkLinusN: what was your patch? :)
00:29:49webguest67By the way, many thanks for the nice flashing procedure. Plenty of feedback.
00:30:29LinusNbagawk: the coldfire assembler choke on move instructions with the EMAC registers
00:31:02bagawkwhat are EMAC registers?
00:31:08LinusNdiscovered it when i was porting the rockbox scheduler to coldfire
00:31:08 Quit webguest67 ("CGI:IRC (EOF)")
00:31:30LinusNExtended Multiply and ACcumulate
00:31:39LinusNDSP stuff
00:35:18LinusNgotta get some sleep now
00:35:21LinusNcu all
00:35:26amiconnnite LinusN
00:35:40 Part LinusN
00:35:50 Quit ripnetUK ()
00:40:56 Quit Val-NM ("CGI:IRC (Ping timeout)")
00:52:34 Join bagawk_ [0] (Lee@bagawk.user)
00:53:06 Quit bagawk (Nick collision from services.)
00:53:18 Nick bagawk_ is now known as bagawk (Lee@bagawk.user)
01:01:50 Quit _aLF ("bye")
01:12:06 Quit mecraw ("Trillian (")
01:18:59 Quit plok ("I'm outta here!")
01:26:13 Quit bagawk ("umount /dev/brain")
01:29:56 Nick dwihno_ is now known as dwihno (~dw@
01:31:04 Join webguest67 [0] (
02:14:43***Saving seen data "./dancer.seen"
02:27:38 Part amiconn
03:54:29 Join ashridah [0] (ashridah@
04:02:31 Quit webguest67 ("CGI:IRC (EOF)")
04:14:44***Saving seen data "./dancer.seen"
05:39:53 Quit scott666_ ("i'll be back...eventually...")
05:49:30 Join LinusN [0] (
06:14:45***Saving seen data "./dancer.seen"
06:24:07 Join traecer_ [0] (
06:28:39 Part traecer_
07:30:34 Join [IDC]Dragon [0] (
07:39:30 Join amiconn [0] (
07:41:05amiconnGood morning
07:41:51[IDC]Dragonamiconn: can you send/post your rom dump?
07:43:24 Join AciD [0] (
07:44:29amiconn[IDC]Dragon: Found my wiki addition?
07:45:03*[IDC]Dragon looks
07:46:00[IDC]Dragonyea, fame on you, blame on me :-/
07:46:18amiconn[IDC]Dragon: I mean the table...
07:47:01amiconnBtw, LinusN: If you have a look at the ArchosOndio topic too - the 3rd meaning of mask byte 1 bit 0 is also used on recorder, but not yet accounted for in rockbox
07:47:05[IDC]Dragonare the mask bits "compatible" with the classic recorder ones?
07:48:02amiconn[IDC]Dragon: Bit 0 and 1 are (rec v1). Dunno if the FM uses bit 3. Bit 2 is obviously new
07:48:55[IDC]Dragonok, then we don't need special code for the polarities
07:50:11[IDC]DragonI could flash them differently, and see what happens
07:51:10amiconnI expect that we won't find an Ondio with bit 0 or 1 of the mask unset.
07:51:44[IDC]Dragondidn't we spot an Ondio tuner bit yet?
07:51:54[IDC]Dragonthe one different between ours?
07:52:19amiconnAhem, look what I say about bit 3
07:52:39[IDC]Dragonit says FMR
07:53:10LinusNSIBC clock invert...interesting
07:53:12[IDC]Dragonsorry, that confused me, I thought you meant the classic FM Recorder
07:54:39amiconnLinusN: Yes, bit 2 of MAS register MainIOControl (0x7f1 on mas3587f) is set according to that
07:55:28[IDC]Dragonto what feature is that related? recording perhaps?
07:55:50LinusNor, no
07:56:25LinusNamiconn: is it set in encoding or decoding mode?
07:57:31amiconnLooks like decoding mode.
07:59:01LinusNhmmm lemme see
07:59:19[IDC]Dragonamiconn: the SP firmware is a lot smaller
08:00:35LinusNamiconn: can you confirm that the SIBC setting is done in the recorder firmware as well?
08:01:07amiconnLinusN: Yup.
08:01:22amiconngotta run, so more explanations later
08:09:22 Quit dwihno ("Mek0r!")
08:11:28 Join Zagor [242] (
08:14:47***Saving seen data "./dancer.seen"
08:32:11 Join Bagder_ [0] (
08:32:18LinusNhi Bagder
08:32:24LinusNhi Bagder_
08:32:27LinusNor whatever
08:32:41Bagder_I am two! ;-)
08:33:26Bagder_which one is schizo? :-P
08:34:24LinusNBagder_: heard about the wiggler news?
08:34:39Bagder_nope, you getting closer?
08:34:48LinusNread the front page news
08:35:00*Bagder_ faints
08:35:54LinusNyeah, i managed to get it running, I only fried one LPT port
08:36:16LinusNi still have two left ;-)
08:36:31LinusNbought a pci card with two extra ports yesterday
08:37:09LinusNi'm glad i didn't fry the wiggler
08:37:18LinusNor the target
08:37:43LinusNactually, i'm not even sure i really fried the first port either :-)
08:37:54LinusNi haven't tried it with the printer yet
08:39:11LinusNBagder_: my patch is in the binutils cvs now :-)
08:39:36Bagder_you are the king LinusN, any comments about that on the binutils list?
08:39:56LinusN2004-09-30 Linus Nielsen Feltzing <>
08:40:04LinusN * config/tc-m68k.c (select_control_regs): Add mcf5249.
08:40:25Bagder_good enough!
08:40:42LinusNthere's a mailing list post too:
08:40:44 Quit Bagder (Read error: 60 (Operation timed out))
08:40:50LinusNThanks very much for reporting this bug and supplying a patch to fix it.
08:41:00LinusNI have applied your patch along with this ChangeLog entry:
08:41:04LinusNbla bla
08:41:20LinusNfeels good
08:41:34LinusNi suspect that it won't be the last patch though...
08:51:10 Join plok [0] (
08:53:47 Quit ashridah ("gone")
08:54:34 Join methangas [0] (
08:57:07*plok is away - Automatically set away. - messages will be saved.
08:59:48[IDC]DragonLinusN: your email just got into the log (08:39)
09:00:34[IDC]Dragondunno if you care
09:05:17 Join amiconn_ [0] (
09:09:12LinusN[IDC]Dragon: ah
09:12:33 Join PaulS [0] (
09:13:07 Quit amiconn (Nick collision from services.)
09:13:07 Nick amiconn_ is now known as amiconn (
09:13:18PaulSCan we all join in chorus (4 part harmony with feeling) in saying how much LinusN rocks right about now?
09:14:06PaulSThis being #rockbox, this means a lot...
09:14:53LinusNhi PaulS
09:15:39*PaulS throws up a hi-5 in LinusN's direction.
09:15:43LinusNamiconn: i wonder if the sibc polarity has any bearing on the s/pdif playback problem some people are experiencing
09:15:54LinusNPaulS: *smack*
09:16:18LinusNPaulS: now tell me about your jtag experiences...
09:17:45 Quit webmind (Read error: 60 (Operation timed out))
09:18:33PaulSJTAG is coming along. I spent the evening very deliberately scoping out and writing a new cable driver for my Macraigor Wiggler-NP, which appears to be significantly different in function from the wiggler that the OpenWinCE folks had on hand. Mine has two line drivers in series −− one powered by the parallel port, and one powered by the JTAG target. Once I saw that, I knew I wanted go really slow and get it all right.
09:18:38amiconnLinusN: It shouldn't. Iiuc this has to do with playback only, as serial port B is used for that. I wonder if there is any recorder in the wild with mask bit 0 == 0 (for which rockbox would use the wrong SIBC invert setting)
09:19:19LinusNamiconn: i wrote s/pdif *playback* problems...
09:20:06LinusNelinenbe and kurzhaarrocker has mask bit 0 == 0
09:20:32amiconnOops. But anyway, the wrong polarity should lead to failing playback with any output (iiuc, again)
09:20:35 Join dwihno [0] (~dw@
09:21:01PaulSThis means I didn't take my iRiver apart tonight at all. At this point, I now have changes to the OpenWinCE code in the form of a new device definition, a new bus driver (toot toot!), and a new cable driver. Maybe I can get away without writing a new flash chip driver, but I'm not too confident of it. At that point I could have started from scratch! :-)
09:21:56 Quit [IDC]Dragon ()
09:22:15LinusNPaulS: let's hope you did it right
09:24:33PaulSHeheh.. Thanks for the vote of confidence. I've sat around with the wiggler torn apart and toggling individual pins on the parallel port enough that I'm lulled into thinking that I have that part right enough that I won't be blowing things up. Past that point, the amount of damage I can do is manageable.
09:28:16PaulS_Maybe_ tomorrow I can get as far as shifting out an IDCODE out of the BSR chain. I'm tempted to do that, but first I should pay some careful attention to the datasheet of the flash to see if it matches any of the definitions precompiled (CFI, JEDEC) so I can get away without writing that as well. Next time I open the iRiver (and have to solder on a TCK again) I want to be writing bits.
09:29:51 Join webmind [0] (
09:31:25LinusNit says "JEDEC standard"
09:32:21PaulSI saw that too. But there are different bit widths, and I really want to make sure that the read-write sequences match exactly.
09:32:31LinusNbut it is CFI compliant as well
09:35:25PaulSYep. And there's a "cfi.c" and a "jedec.c" in the jtag-0.5.1 distribution.
09:42:27LinusNamiconn: SIBI control...interesting indeed
09:44:05LinusNPaulS: well, we can agree that you will burn your flash, in one way or another :-)
09:44:54PaulSLinusN: I'm working on improving my odds in one of those directions...
09:50:44amiconnLinusN: I think the assembler snippet"> (13 instructions) is pretty clear
09:53:49LinusNamiconn: no, i mean SIBI control
09:54:02LinusNin the ondio
09:58:24LinusNsibi control is necessary for pass-through wav playback
09:59:20amiconnLinusN: Yes. But this is set to low and then not touched anymore by the original fw
10:00:57amiconnBtw: the assembler snippet is from recorder fw version 1.27d, but is present in the same form in the ondio fw (although at a different address)
10:01:18LinusNi'll add the same thing in rockbox
10:09:55 Join [IDC]Dragon [0] (
10:14:31[IDC]DragonZagor: will you fix the button, or is that our issue?
10:14:50***Saving seen data "./dancer.seen"
10:15:23amiconnLinusN: (latest adc.c change): sleep(1) does already sleep 2 ticks...
10:16:24Zagor[IDC]Dragon: i'll fix it. i've just been busy, sorry.
10:17:15[IDC]Dragonno problem, I was just asking
10:17:55[IDC]Dragonwho's job it should be
10:22:51amiconnI wonder if the shift function is still needed at all. Since LinusN wrote the new adc driver, scrolling is pretty damn fast even without shift
10:23:55Zagori still want paging. scrolling through hundreds of entries is no fun even with the faster repeat
10:27:45[IDC]Dragonon an MMC card device?
10:28:05Zagorah, no I meant in general.
10:28:21Zagoron the ondio I agree (as I wrote yesterday) that shift is unnecessary
10:28:21[IDC]DragonI thought with future 4GB MMC
10:28:49[IDC]Dragonon the HD model, I also want paging
10:28:55LinusNamiconn: yeah, i remember that now
10:29:12LinusNamiconn: then i wonder why my fix cured the charging screen bug...
10:30:24[IDC]Dragonamiconn: with the FAT test code, I can't reproduce the problem
10:30:58[IDC]DragonI used a disk image now, instead of a partition image, but worked well, too
10:31:22[IDC]DragonZagor: what kind of image was it in your tests?
10:31:42Zagorpartition image
10:32:37[IDC]Dragonwith such, the partition offset is zero, I was suspecting undetected bugs, but found none
10:33:58amiconn[IDC]Dragon: I checked the number (with splash() ) on the player. Don't remember them exactly, but the code tried to write quite a number of sectors before the allowed start sector.
10:34:17amiconnUnfortunately I can't test on the target(s) atm, since they are at home.
10:35:12[IDC]DragonI'd like to get such a partition image, or even a disk image
10:35:59[IDC]Dragonif freshly formatted, they should zip very well
10:36:05amiconnIf you want a disk image, I'd better reformat again without quick format, to ensure good compressability
10:36:06Zagori'll make you one
10:36:27[IDC]DragonI want one that fails
10:36:44[IDC]DragonI have one which I grabbed from the Ondio
10:36:45Zagorok I'll let you two work it out
10:37:20[IDC]Dragonquick format is not enough for compressability
10:37:55Zagordd if=/dev/zero of=disk.img
10:38:21[IDC]Dragonon linux, yes
10:38:43ZagorI thought amiconn was using linux?
10:38:48[IDC]DragonI should venture into vmware
10:40:05amiconnZagor: No, WinXP
10:40:31Zagorok. well good luck, i'm off for lunch
10:40:54amiconn[IDC]Dragon: How did you produce the Ondio image? Did you do it on windows?
10:41:49dwihnoerhm... with apache, how does the virtualhost stuff work? first match is displayed?
10:41:52amiconnLinusN: (charging screen issue) Weird indeed. That reminds me I should look at the bug reports more often...
10:42:49LinusNsomehow the d/c voltage wasn't correctly sensed, so it skipped the charging screen
10:43:06LinusNlunch time, cu l8r
10:44:58[IDC]Dragonamiconn: I used a program called hexedit or so
10:45:11[IDC]Dragonit can edit files, but also disks
10:47:31[IDC]Dragonsorry, it's called WinHex
10:50:08[IDC]Dragonthere may be more simple and free programs, but with a quick search I found none
10:51:25[IDC]Dragonbut I found a free tool to mount such partition images
10:52:11[IDC]Dragoncalled VDK,
10:56:06 Join ashridah [0] (
10:56:58amiconn[IDC]Dragon: I googled a bit, and using the posix /dev/sdxN devices in cygwin is possible too:
10:59:32[IDC]Dragonbut you need specific tools to create and check images, right?
11:01:00[IDC]Dragonlike /sbin/mkdosfs and /sbin/dosfsck used in the test script
11:03:29amiconnIf you want an image of my player test, a simple dd if=/dev/sdg of=blabla.img should do
11:04:45[IDC]Dragonif there is a dd
11:05:27amiconnit is there
11:05:43amiconnI tested it with my local hd, it works
11:06:17[IDC]Dragonwell, then I hope we can nail that bug
11:06:21amiconnBeware that if you unpack my image, you will be facing a ~3.3 GB file
11:06:45ashridahheh. what's it compress down to? :)
11:06:45[IDC]Dragonno problem, as long as it compresses well
11:07:11[IDC]Dragonthey call such a compression bomb
11:07:41[IDC]Dragonspecial files to tease virus checkers
11:09:16amiconnashridah: tdb.
11:10:22ashridah[IDC]Dragon: heh. a friend gave me an msdos floppy disk with a loop inside the fat entry for a particular file. damned thing grew out to fill a 1GB disk that only had 200MB on it
11:10:30ashridahi wasn't impressed ;0
11:10:32ashridah:) even
11:11:35 Join R3nTiL [0] (
11:12:06[IDC]Dragonlunch time
11:15:21 Join MisticJeff [0] (
11:15:48MisticJeffmornin' gents
11:29:11 Quit ashridah ("Client exiting")
11:29:30 Join ashridah [0] (ashridah@
11:36:06 Quit R3nTiL ()
11:43:34 Join domasofan [0] (
11:47:01 Quit domasofan (Client Quit)
11:55:20 Join DomasoFan [0] (~some@
11:56:26DomasoFani have a archos jukebox studio 20. i want to install rockbox with speaking menus. can someone help me doing that?
12:07:07[IDC]DragonDomasoFan: yes, what's your status with it?
12:11:34DomasoFanmbr status?
12:12:08[IDC]Dragondo you hve Rockbox running, do you know it already?
12:12:45[IDC]Dragonbasically, you only need to copy a .voice file into /.rockbox/langs/
12:13:12 Quit DomasoFan (Read error: 54 (Connection reset by peer))
12:13:20 Join R3nTiL [0] (
12:14:52***Saving seen data "./dancer.seen"
12:16:24 Join DomasoFan [0] (~some@
12:16:58DomasoFanafter i connected my archos the pc crashed. it restarted. don't know what the error is
12:21:04DomasoFanit was the second time the error occurred. os: win xp home
12:22:09 Quit MisticJeff ()
12:24:58amiconn[IDC]Dragon: (sibi changes) (1) Do you have working playback now? (2) Perhaps this would be better placed into a (yet to be created) mas_init() function? In addition, this could do the poweron reset for MAS3587/3539, and the WSEN thingy for the 3507
12:27:51DomasoFanis here someone which can help me? pc crashes when i connect the archos jukebox studio 20 when i turn it on.
12:28:03ZagorDomasoFan: have you installed the isd200 driver?
12:29:26[IDC]Dragonamiconn: no playback, sorry
12:29:41[IDC]Dragonbut I haven't tried today
12:30:05[IDC]Dragonsince I formatted the Ondio, there are no mp3s on it any more
12:30:25[IDC]DragonI was rather working on the port setup
12:31:33[IDC]Dragonthe SIBI state was already OK for warm boot, so the code I added doesn't change the behaviour
12:32:00DomasoFan@zagor: no. i haven't. where can i find it?
12:32:01[IDC]Dragonbut I agree on mas_init()
12:32:14[IDC]Dragonwe also have no usb_init()
12:38:44amiconn[IDC]Dragon: The usb code needs some changes for the Ondio anyway: (1) it should monitor the mmc_inserted switch, and deassert/reassert the card detect to the bridge on a state change to make sure the PC recognizes the card change
12:39:03DomasoFan@zagor: downloaded and uncompressed.
12:39:26amiconn[IDC]Dragon: (2) It should display that "Please remove..." info on usb detection if a card is inserted
12:40:58amiconn@all developers: With supporting more and more platforms, the localization will contain more and more strings that are only relevant on some platforms. Any ideas how to handle that?
12:41:16amiconnThis becomes more important for the voice file
12:43:51 Join hesim_512 [0] (
12:44:11Zagoramiconn: this is only the case if each new model adds unique features, which is far from certain
12:44:24 Quit hesim_512 (Client Quit)
12:44:27DomasoFancan someone tell me how to install the usb driver? the windows has made an error by installing the driver while it recognices the disk allone
12:44:53Zagoramiconn: so far I don't think it's a problem. if it becomes one, we'll deal with it then.
12:45:23ZagorDomasoFan: it recognizes the disk alone?
12:45:27[IDC]Dragonamiconn: (usb monitoring) yes, I was having such in mind, too
12:46:01DomasoFannot complete. but: how to install the usb driver?
12:46:17amiconnThe MMC thread should perhaps monitor the mmc_inserted status too, and force a file system remount in case of a change
12:48:36 Join matula_ [0] (
12:49:11ZagorDomasoFan: is there no install program in the archive?
12:49:29DomasoFanthere are some exe-files there.
12:50:46ZagorDomasoFan: anyone called install.exe or setup.exe?
12:53:51amiconnDomasoFan: Did you download
12:57:43amiconnYou have to install this with the device manager. Alternatively you could download TPPINST_5_04.exe
13:01:26 Quit DomasoFan ()
13:05:06 Quit R3nTiL ()
13:06:32Zagorwhat's the cheapest way to ship a small package from us to europe?
13:08:39[IDC]Dragondunno exactly, but postage is generally very affordable in the US
13:08:40ashridahZagor: good news about the working wiggler
13:09:07[IDC]Dragonsomebody who shipped me his FM paid $5 or so, can tell when I'm home
13:09:08Zagorindeed. we celebrated with a lunch pizza :)
13:09:25ashridahso what was the issue?
13:09:46Zagor[IDC]Dragon: i'm looking at the USPS website and can't find anything below $15, unless you classify it as "mail" instead of "parcel".
13:10:23Zagorashridah: I'd better let linus explain, it is his feat.
13:10:51[IDC]Dragon(wiggler) yeah, sounds rather lame, that it takes days to hook up a simple development tool ;-)
13:10:55LinusNashridah: to begin with, i had connected it wrong
13:11:16 Join elinenbe [0] (~elinenbe_@
13:11:31LinusNthe parallel port freaked out on me, but it worked with a PCI card with two ports on it
13:11:39LinusN[IDC]Dragon: :-)
13:11:50elinenbejust read the mailing list emails. Good luck with the wiggler! −− bdm is the coolest!
13:11:52ashridahLinusN: haha. reminds me of the time i freaked out about my 68hc11 evbu board being broken. turns out the serial cable i was given was wired badly. :)
13:14:52LinusNashridah: i had a similar experience with a 68HC11 many years ago, wrote a simple test loop that blinked a LED
13:15:02LinusNLEd was blinking, but nothing else worked
13:15:45LinusNafter hours of head scratching, it turned out that the CPU was dead, and i had used a blinking LED by mistake :-)
13:16:00ashridahthat's classic :)
13:21:30 Join DomasoFan [0] (~not@
13:22:19DomasoFanhi guys. im back and my archos is now running correcltly. now, how to install rockbox? copying only the .rockbox and archos.mod file into the root of the player?
13:22:33Bagder_unzip a whole zip
13:22:37 Nick Bagder_ is now known as Bagder (
13:22:42 Quit matula_ (Read error: 104 (Connection reset by peer))
13:24:19amiconn[IDC]Dragon: How do you beep traces where both ends are on different sides of the board and are connected to QFP chips?
13:26:51LinusNhave sharp probes, so you can keep them in place with the hand that holds the pcb
13:28:48amiconnYes, but you would need 3 hands for that. One for each probe, and one to hold the board...
13:29:33LinusNi use my two hands for that
13:29:38DomasoFandoes rockbox recognize the .voice file when i copy it into the lang dir?
13:30:02LinusNit should
13:31:50DomasoFan@LinusN: have i only to copy the .rockbox folder and the archos.mod into the root of the player? and: when i have copied all files on the player how i can start the player with the new firmware?. im totally blind so i have here a little bit probs :-)
13:32:09Bagder599 members of the mailing list today
13:32:16Bagder−−-><−−- this close to 600
13:32:32LinusNDomasoFan: you are supposed to unzip the entire zip file to the jukebox
13:32:53LinusNtoday, all it contains is the archos.mod and the .rockbox folder, so yes
13:33:12Bagderthen restart the player
13:33:18LinusNafter you have installed rockbox, you turn it off and on again
13:33:49LinusNamiconn: or you solder a wire to one of the pins
13:33:55DomasoFanok. uncompressed and copyed the .voice file into the lang dir.
13:34:14LinusNDomasoFan: did you install a daily build?
13:34:32DomasoFanyes. daily build of 01.10.2004
13:35:56DomasoFangood. does now the voice work? or have i to restart the player?
13:36:25LinusNyou might want to restart it, but i don't think you need to
13:36:59LinusNcan you confirm that rockbox is running on it, and not the original firmware?
13:37:24 Join ripnetUK [0] (
13:37:29DomasoFanhmm. where is the headphone plug? archos jukebox studio 20
13:37:40LinusNsometimes the studio firmware has problems loading the firmware
13:37:44LinusNtop left
13:38:14DomasoFanhm. left side near the usb plug.?
13:38:25LinusNno, top left
13:38:47LinusNopposite side of the usb plug side
13:40:03elinenbeLinusN: how is the rev-eng coming?
13:41:17ripnetUKshould be 600 on mailing list now as I just (re) joined (left when i sold my archos and bought a iRiver)
13:41:42NibbIeririver is better then archos?
13:42:04[IDC]Dragonamicon: (beeping) I place the board on the side, sometimes holding it with tha same hand as the probe. Often I held the probe between pins, to get a more stable rest
13:42:47[IDC]Dragonaminconn: gone beeping now?
13:42:52Bagdernah, still "only" 599
13:44:38LinusNsomebody PLEASE join! :-)
13:44:39 Quit DomasoFan (Read error: 104 (Connection reset by peer))
13:45:11LinusNfirst version of iriver ATA schematics uploaded
13:45:39LinusNit's so much easier with a naked PCB
13:45:42 Join DomasoFan [0] (~not@
13:46:43DomasoFanhmmm. my rockbox made some kreaking sounds when it turns on and wheni have connected it to the pc the pc crashes. do you know this?
13:47:10LinusNDomasoFan: sounds like battery or hard drive problems
13:47:28DomasoFangood idea battery :-).
13:48:05[IDC]DragonLinusN: iriver byteswaps "in hardware", good idea
13:48:33DomasoFanwhere is the plug for the power? next to the headphone plug?
13:48:39LinusN[IDC]Dragon: nice eh?
13:48:49 Join MooMaunder [0] (~me@
13:49:35[IDC]Dragondo we now need to byteswap the command traffic?
13:50:14LinusNthe commands are bytes
13:52:17amiconn[IDC]Dragon: Nah, not beeping now. Last night I wanted to double-check the mas por and i2c connections. I *want* that damn sound working!
13:52:59DomasoFan@linusN: do you know where the power plug is? don't find it
13:53:04[IDC]Dragonwhat a pity that we can't bitswap by re-wireing a serial line
13:53:42[IDC]DragonDomasoFan: bottom right, next to the USB jack
13:53:59DomasoFanleft or right?
13:54:40LinusNDomasoFan: it's located next to the usb plug
13:55:04[IDC]DragonDomasoFan: do you know the blind FAQ?
13:55:40[IDC]Dragonit tells about jack and button positions, etc.
13:58:48DomasoFanthe player is quiet after i started. no error message but it don't speak. have here a seeing person now.
14:01:12LinusNgood, see if he/she can see if rockbox is running, or the original firmware
14:01:42ZagorDomasoFan: try pressing MENU. it's the lower of the two circular buttons below the display
14:02:18LinusNto the left
14:04:17Bagderrerepiing your entire CD-collection gives you an interesting problem
14:04:41Bagderfinding the old version of the new rip and replace it
14:05:15Bagderbig magic perl script does the job pretty good now
14:13:17LinusNnow all you need is a cd changer to run it all night
14:13:31DomasoFanhow i can turn off the player?
14:13:51LinusNhold STOP for a few seconds
14:14:00LinusNbut remove the charger first
14:14:54***Saving seen data "./dancer.seen"
14:16:51Zagoranyone want a pink iriver?
14:19:06ashridah15 bucks? seems a tad steep for such a... uh. pink fashion item
14:25:37DomasoFandoes anyone if the deutsch.voice file is to big for the archos jukebox studio 20? i tried and tried but it don't speaks to me
14:25:58Zagortry the english.voice then and see if it works better
14:26:15Zagorhave you been able to verify that you are indeed running rockbox?
14:27:16DomasoFanhm. the voices menu in the player is available but i couldn't access nothing with the voice. voices are enabled.
14:29:09DomasoFani wait until my brother is at home because he and i are the only english speaking people here. my mother and my father only understand german :-).
14:31:36DomasoFanbut thanks for the good help you give me.
14:31:47[IDC]DragonDomasoFan: the selected language has to match the .voice file
14:32:15[IDC]Dragoni.e. if you run in german, you need to have deutsch.voice
14:32:48LinusNand vice versa, you need to select german as your language to use the german voice file
14:34:00Zagorwell, i'm going away over the weekend. see you all monday.
14:34:06 Part Zagor
14:38:41DomasoFan@[IDC]Dragon: OK i know whats wrong now :-). good. does anyone know how to change the language? I think that this wasn't a topic in the blind faq.
14:39:49[IDC]Dragonis it talking now?
14:40:19DomasoFanno because the language is english at the monment.
14:40:44[IDC]Dragonthen put a english.voice on it ;-)
14:41:25DomasoFanbut i need german language because my titles sound like shit when i listen the words in english :-)
14:41:42[IDC]Dragonthat's a different thing
14:42:03[IDC]Dragonwe were talking about the menu voice
14:42:43[IDC]Dragonfor your directories, you generate little mp3 clips with the spoken name, on the disk
14:43:08DomasoFanbut i need the menu voice in german to .
14:43:51[IDC]Dragonanyway, language selection is in the menu, item General Settings, subitem Language
14:44:09DomasoFanah ok. thx
14:47:20[IDC]Dragonjust above the Voice entry
14:47:54[IDC]DragonI can chat in german, too, if that helps
14:48:18 Join matula_ [0] (
14:53:34DomasoFancool. my player talks to me! :-)
14:54:02 Join Zagor [0] (
14:54:02 Join _Zagor_ [0] (
14:54:20Bagderzagor and his clone!
14:54:30 Quit Zagor (Client Quit)
14:54:31 Quit _Zagor_ (Client Quit)
14:56:12[IDC]DragonDomasoFan: now you need to generate all those directory clips, if you want that feature.
14:56:34[IDC]Dragonand if you have directories, that is.
14:58:12DomasoFancan i reset my language to english? i made an error by chosing the wrong language because the voice was wrong in talking back the language name.
15:00:08[IDC]Dragonsure, do the selection again
15:01:38DomasoFanin a foreign lang? :-)
15:01:43[IDC]Dragonor, delete all .language files except those you want to use
15:02:08[IDC]Dragonif it can't find it on boot, it will use english again
15:05:11amiconnDomasoFan: I recommend that you keep only englisch.lng, deutsch.lng, english.voice and deutsch.voice in /.rockbox/langs to avoid unnecessary confusion.
15:08:31DomasoFanah ok.
15:10:58 Part LinusN
15:34:17 Join Simon86 [0] (~not@
15:34:17 Quit DomasoFan (Read error: 232 (Connection reset by peer))
15:35:16Simon86hi here is DomasoFan. My nick has at the moment errors. thanks for helping me. now my player runs correctly with german language. thanks.
15:36:59amiconn[IDC]Dragon: Did you already prepare a flash image for Ondio SP?
15:37:34Simon86can i create .voice files by myself? i am a programmer to but at the moment only in lpc, vb basic, qbasic and some other programming languages.
15:38:57[IDC]Dragonamiconn: yes
15:38:57 Join webguest20 [0] (
15:39:29[IDC]DragonSimon86: yes
15:39:40*[IDC]Dragon is very positive right now
15:40:30amiconn[IDC]Dragon: Is it safe to flash (read: not more unsafe than flashing e.g. a JBR)? Or are some unclear things left?
15:41:51 Quit webguest20 (Client Quit)
15:41:53[IDC]Dragonamiconn: should be safe
15:42:08[IDC]Dragonthe Archos image runs fine
15:42:17[IDC]Dragonrockbox to be done
15:42:47[IDC]Dragonand plugin support, of course
15:42:47Simon86hm. how to create .voice files?
15:43:17amiconn[IDC]Dragon: Of course I would have to do this with firmware_flash.rock. No UART boot mod...
15:43:35[IDC]DragonSimon86: best wit amiconn's scripts
15:43:52[IDC]Dragondo we have a wiki page? me looks
15:43:53 Join webguest92 [0] (
15:43:56Simon86nice. where can i find tehm?
15:44:02amiconnSimon86: I wanted to write a howto for that...
15:44:52[IDC]Dragonif you are a semi-programmer, you'll have no problem with it
15:45:19[IDC]Dragonhow do you want to make the clips? TTS or record them?
15:45:20Simon86i program since abozut 1999. so i hope i have no problems :-)
15:45:42Simon86i can try it when it don't works i can use the other voice files :)
15:48:27[IDC]Dragongrab the stuff from:
15:49:11 Quit webguest92 ("CGI:IRC (EOF)")
15:51:13[IDC]Dragonyou also need a .lang file of your choice
15:52:09[IDC]Dragonfresh from cvs:
16:03:06 Quit NibbIer (Read error: 113 (No route to host))
16:07:56 Part PaulS
16:14:56***Saving seen data "./dancer.seen"
16:15:40 Join SmoothOperator [0] (
16:15:50SmoothOperatorGot my brand new mp3 player today! :-)
16:17:31SmoothOperatorand its a million times better than the ipod
16:17:35Simon86i too :-)
16:17:59SmoothOperatorwhat make is yours simon?
16:18:16Simon86and had a few problems :-). but now it runs quite fine. but in the amazon review where problems listet but not these which i had :-)
16:18:31SmoothOperatorwhat make is ur mp3 player?
16:18:34Simon86archos jukebox studio 20
16:18:50SmoothOperatorive got an iriver ihp120
16:18:55 Quit ashridah ("insomniac zombie: sleep. sleeeeeeeeeeeeeeeeep. sleeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeep")
16:19:42Simon86how much it was?
16:19:50SmoothOperatorits a 20gigger
16:20:00SmoothOperatorplays WAV, MP3, WMA, OGG, and ASF
16:20:12Simon86i taken mine about 200 from amazon. 20 gb.
16:20:36SmoothOperatorit came with a backlit LCD remote, leather holster and external microphone
16:21:02Simon86hehe. it comes with a earphone :-)
16:21:05SmoothOperatorit feels sublime compared to the plasticky ipod
16:21:11SmoothOperatoryea comes with earphones as well :-)
16:21:40Simon86mine felels like it is made of steel or something.
16:22:00Simon86ah no. it is plastic
16:22:24SmoothOperatorthe iriver is solid magnesium with chrome ends
16:22:39Simon86but the normal price is over 300 but i am not shure.
16:22:41SmoothOperatoreven its control joystick is metal
16:23:17SmoothOperatorits a beauty, you won't get a better MP3 player for the money
16:23:24Simon86i take it only because it runs with rockbox :-)
16:23:27SmoothOperatorand it supports simple drag and drop just like hte archos
16:23:37SmoothOperatorno need for stupid transfer software
16:23:51SmoothOperatorI think rockbox for the iriver is being worked on :-)
16:24:02Simon86yes. for blind people it is cool. and the rockbox supports speaking menus and other things
16:24:34SmoothOperatorthe iriver also has a great sounding FM radio
16:25:01 Join oxygen77 [0] (
16:25:05Simon86hmmmm... thank you :-)
16:25:20SmoothOperatorand it accepts line in, SPDIF optical in and external mic in, and it can record to WAV or MP3
16:25:48Simon86can we exchange the players? ;-)
16:26:09SmoothOperatorno :-)
16:26:19Simon86ah shit. :-)
16:26:51Simon86but i have line in too. and other plugs which i don't know at the monent.
16:27:14Simon86but it has a about 2 mb buffer
16:27:48SmoothOperatori dunno what the iriver has, but i imagine it has a large buffer in order to play WAV files
16:28:18Simon86yes, i see we like our players :-)
16:29:44SmoothOperatorthe remote is awsome
16:40:29 Join R3nTiL [0] (
16:42:42 Quit R3nTiL (Client Quit)
16:48:35 Join R3nTiL [0] (
16:50:20SmoothOperatorSimon the iriver has a 32Mb buffer
16:52:38 Quit R3nTiL (Client Quit)
17:00:30 Join webguest17 [0] (
17:01:11 Quit webguest17 (Client Quit)
17:08:13 Quit [IDC]Dragon ("CGI:IRC")
17:15:48 Join alxdotnet [0] (
17:17:36Simon86good afternoon/evening/night/morning/noon :-)
17:17:47 Quit kaouete (Remote closed the connection)
17:18:01 Join kaouete [0] (
17:18:12alxdotnetsame to you simon
17:19:09alxdotnetso...congratulations to the iRiver port team on getting BDM to work
17:19:47alxdotnetwhat kind of a time frame are we looking at before RockBox runs on the iHP?
17:21:46Simon86i heard that it is going to bee a few months
17:23:27elinenbei heard it will be up and running later today!
17:23:34elinenbeoh −− wait... wrong project!
17:23:38Simon86really? donno
17:28:45alxdotneti can't wait to load it...the iRiver firmware really..well...complete this sentence ;)
17:29:28Simon86i have a archos jukebox:-)
17:32:10alxdotnetwell, software decoding of wav, mp3, wma, and ogg isn't bad...but the company doesn't really listen to anything we ask gapless playback or OTF playlists
17:32:49alxdotneton the other hand, the device itself rocks
17:33:01Simon86like archos?;-).
17:33:23Simon86i have heard (features) that archos don't fixes bugs as iriver
17:34:02alxdotnet140 mhz coldfire processor
17:34:04alxdotnetfor one
17:34:18alxdotnetallowing on-the-fly software mp3 recording and possibly transcoding
17:34:34alxdotnetmaybe even software crossfading...that would be cool
17:34:58Simon86nice's a great architecture with really bad software
17:35:44alxdotnetfor instance, the default firmware is known to drop frames while recording from the line in
17:36:01alxdotnetwhich is blasphemy to any serious audiophile
17:36:21Simon86away. cu
17:36:24alxdotnetoh ok
17:39:03 Join [IDC]Dragon [0] (
17:42:10amiconn[IDC]Dragon: hi again!
17:43:24[IDC]Dragonyes, hi
17:44:44 Join Nibbler [0] (
17:50:42 Quit SmoothOperator ("CGI:IRC (EOF)")
17:52:41 Quit matula_ (
17:52:41 Quit MooMaunder (
17:52:41 Quit ripnetUK (
17:53:28NJoinmatula_ [0] (
17:53:28NJoinMooMaunder [0] (~me@
17:53:28NJoinripnetUK [0] (
17:54:56 Quit alxdotnet ()
18:00:14amiconn[IDC]Dragon: Long format of the test partitions done, still panics when trying to write to fat16 partition. Currently dumping the disk
18:00:45[IDC]DragonUSB1, I forgot :-(
18:01:08[IDC]Dragonlet's hope the power lasts
18:02:14amiconnIt's freshly charged, and estimated time will be ~1 hour. After that my processor will get some work to do with zipping that file
18:02:37[IDC]Dragonmaybe dumping the FAT16 partition is enough...
18:03:07[IDC]Dragonhave you reformatted the small partitionafter the panic?
18:03:23[IDC]Dragonelse it might be in strange shape
18:04:22 Join _aLF [0] (
18:05:00amiconn[IDC]Dragon: Should I interrupt the copy, and reformat the FAT16 part?
18:07:34[IDC]Dragonhow far is it?
18:07:47amiconnI could also try the reverse order of partitions, i.e. placing the FAT16 part before the FAT32 one. But I doubt that with that layout the panic will be triggered
18:08:02amiconn500 MB done (of 3.5 GB total)
18:08:11[IDC]Dragondid it happen with a fresh partition?
18:08:40[IDC]Dragonjust archos.mod on it, I guess
18:09:01[IDC]Dragonno, that was the other
18:09:03amiconnYes, on both the FAT32 and the FAT16. FAT16 also has .rockbox dir
18:09:04 Quit MooMaunder (Read error: 110 (Connection timed out))
18:09:29[IDC]Dragonit would be good to prepare like before it happened, imho
18:10:19 Join mecraw [0] (~lmarlow@
18:11:38 Quit ripnetUK (Read error: 110 (Connection timed out))
18:12:07 Quit matula_ (Read error: 110 (Connection timed out))
18:14:57***Saving seen data "./dancer.seen"
18:20:23amiconn[IDC]Dragon: Image prepared, dumping...
18:25:42amiconnWhile the dump is running... Do you have an idea what might cause the non-working playback? Did you measure whether the mas "DVS" pin (pin 15) is low or high?
18:26:14[IDC]DragonI haven't really investingated into playback, no
18:26:42[IDC]Dragoncurrently, I'm still checking port pins
18:28:14amiconnFor cold-start? You could do what I already did, putting a debug function that displays all P(A|B)(D|IO)R register values at the start of rockbox execution
18:28:34amiconn...or I provide you with the values
18:28:35Simon86@dragon: private message :-).
18:28:49[IDC]Dragonthat's what I did
18:29:00[IDC]Dragonnow thinning down the bits
18:29:20[IDC]DragonSimon86: sorry
18:30:13amiconn[IDC]Dragon: You can set PA13 to GPIO in. The IRQ is unusable
18:30:38[IDC]DragonArchos seems to set it to IRQ
18:31:20amiconnYes, but they set it to GPIO within all read and write functions. One more example of a half-thought-of feature:
18:32:59amiconnI think the intention was to be able to get the info that the card is no longer busy via an interrupt. However, the busy pin goes low->high in that case. Interrupts can only be triggered by high->low (or low level) on the SH1...
18:33:15[IDC]Dragonah, that again
18:33:56amiconn...unless there is an inverter, but I doubt that. Furthermore, it would only work for the internal flash
18:34:30[IDC]Dragonthat's clear, yes
18:35:18amiconnThere is no IRQ1 interrupt vector in the archos fw
18:40:26[IDC]Dragonfound the evil bit: PADR, bit 12
18:41:04[IDC]Dragonthis was missing in our init? checking..
18:41:10amiconnWhat was wrong with it? That's the clock gate, and it is set in ata_init()
18:42:27amiconnOr rather, it doesn't need to be initialized, because select_card sets the value
18:42:38amiconnIt only has to be set to GP out
18:44:57[IDC]Dragonno, PA12 is not set
18:45:17[IDC]Dragon(at least, not in the ata_init I'm facing)
18:45:35amiconnNo, as it doesn't need to be set there.
18:46:11amiconnOr do you mean setting PACR1?
18:46:52[IDC]Dragonno, the bit in PADR, I'm sure
18:47:11amiconn[IDC]Dragon: [19:33:49] <amiconn> Or rather, it doesn't need to be initialized, because select_card sets the value
18:48:07 Join zeekoe [0] (
18:48:55[IDC]Dragonmaybe that's too late?
18:49:23amiconn[IDC]Dragon: Now I know what's going wrong... my card init needs some rework. If the clock gate is set wrong, the card doesn't receive the initial sync pulses...
18:50:42amiconnOr rather, the select_card function itself needs the rework. Should be easy!
18:50:59[IDC]Dragonbut you haven't touched the card yet?
18:51:58amiconnlook @ata_mmc.c, lines 138..156. The PA12 setting comes after the sync clocks, which is obviously wrong.
18:52:13[IDC]Dragonah, now I see
18:52:41amiconnThe CS assertion has to take place after the sync clocks, so I have to split this in two
18:53:12amiconnThis doesn't hurt anyway, since the first one later has to be extended to take mask bit 2 into account
18:53:48[IDC]Dragonok, please fix ;-)
18:54:28[IDC]DragonI've added some PACR inits meanwhile
18:54:39 Quit AciD (Read error: 54 (Connection reset by peer))
18:54:44[IDC]Dragonshould I commit that first, or merge later?
18:55:24 Join AciD [0] (
18:55:52[IDC]Dragon(depends if you already have local changes)
18:55:59amiconn[IDC]Dragon: commit first, my change is most likely simpler
18:56:42 Join MIKE- [0] (
18:56:55 Quit MIKE- (Client Quit)
18:57:08[IDC]Dragonmine is 3 lines
18:57:29amiconnI think it will merge cleanly anyway
18:57:34[IDC]Dragonbut I'll commit it then
18:58:04[IDC]DragonOK, done
18:58:56 Quit Bagder ("Off to search for that connect-resetting peer guy!")
18:59:31amiconn[IDC]Dragon: Clean merge :)
19:00:20zeekoehi! /me back
19:02:11amiconn[IDC]Dragon: Committed card_select() fix. Now try to cold-start!
19:03:56[IDC]Dragonflashing takes a while, with 14kBaud
19:04:16amiconnNo working firmware_flash.rock yet?
19:04:36[IDC]Dragonhaha, no
19:06:04amiconnThere are a lot of mails concerning the Ondio on the ml :)
19:06:44zeekoei have a question... there's all this talk on wiggling. i googled, dictionaried, etc, found all kinds of interesting stuff, but still dont know what wiggling is
19:06:51zeekoecan someone please explain to me?
19:07:21[IDC]Dragonthe wiggler is a low level debug tool
19:07:48[IDC]Dragonso the folks here just created a verb from it, I guess
19:08:56zeekoeit all becomes clear now, at last...
19:11:23[IDC]Dragon(had to remove my test code first)
19:11:39[IDC]Dragonwhich I remembered half-way
19:14:28[IDC]Dragonbetter now, yes
19:16:02amiconnThere's some interesting thing with the timing: If I implement early return from writing the last block, we could write to both internal & external card with almost no speed loss!
19:18:02[IDC]Dragonwhat do you mean, both? At the same time?
19:18:31amiconnYes. Writing one block to internal, one to external, one to internal...
19:18:40[IDC]DragonRAID, hehe
19:19:04[IDC]Dragonteach that to the USB bridge
19:19:06amiconnThis way, we always write to one card while the other one is busy
19:21:25amiconnI already have an idea how to implement early return. In order to not decrease performance in other cases, the driver has to remember whether the card might be busy (from a previous write call) and only check for "still busy" in that case.
19:22:09[IDC]DragonI'm more worried about reading...
19:22:53[IDC]Dragonstill no sound, sorry.
19:23:22amiconnFor reading, we have to wait for the data block to arrive
19:24:09amiconnI'll measure the latencies, and check whether yield() ing in between would hurt performance
19:26:59amiconn[IDC]Dragon: Disk dump done, zipping...
19:27:13[IDC]Dragonlet's see
19:28:40amiconnArgh! ZIP can't handle files >= 2 GB
19:29:38[IDC]Dragonmaybe rar, or others
19:29:49amiconntrying gzip...
19:30:50amiconnThe .gz grows way too large.. :(
19:36:55[IDC]Dragonhow about .ucl? ;-)
19:37:09amiconnCurrently trying bzip2
19:37:28[IDC]Dragonit has to be something whichis good about long boring streaks
19:38:59amiconnBah, I wonder what is wrong with the dump. The .bz2 is already >300 MB :-(
19:44:53 Join ripnetUK [0] (
19:44:55[IDC]Dragonlook inside
19:45:20[IDC]DragonI hope the formatting didn't init the sectors with wild values
19:45:45amiconnWith what tool? I think no tool can load such a huge file at once
19:47:27amiconnGot an idea...
19:49:01amiconnArgh! It seems that a windows long format is in fact no long format...
19:49:09[IDC]DragonTotalCommander can split
19:50:08amiconnHave to do all that again... choosing a bit smaller partitions as well (but checking first if the panic still happens)
19:53:04[IDC]Dragonoh, sorry
19:53:19[IDC]Dragonwhat will you use for formatting now?
19:53:38 Quit Simon86 ()
19:54:19amiconn...using format /u
19:54:26[IDC]Dragoncan you restrict dd?
19:54:46[IDC]Dragonthen youcould do a small FAT32 and a small FAT16
19:54:50amiconnyes, and I already did this
19:54:56[IDC]Dragonand dump only up to there
19:55:16amiconnThe disk is 10 GB in total, and I had a 3 GB FAT32 and 256 MB FAT16
19:55:45[IDC]DragonI'm still finding many port differences
19:56:11[IDC]DragonI have moved my check now after all inits, before the menu invocation
19:56:48[IDC]Dragoncomparing cold and warm boot
19:56:58amiconnNow I have a 512 MB FAT32 in front of the 256 MB FAT16. Bug still happens
19:57:37[IDC]Dragonbut still plenty
19:58:17amiconnThis time the dump should be zippable...
19:59:19 Quit mecraw (Read error: 232 (Connection reset by peer))
20:00:26amiconnI just remember Win9x can't handle 2 primary partitions with FAT... have an idea how to solve this
20:00:55_aLFwth extended partition it works
20:01:17amiconn_aLF: Yes, but rockbox in turn doesn't handle extended partitons
20:07:52[IDC]DragonI found one port difference which may be important:
20:08:03amiconnWin98 also does no real long format...
20:08:25[IDC]Dragoncold-started, PA5 is high, warm booted it is low
20:08:50[IDC]DragonPA5 is USB enable, high = enabled
20:09:31amiconnDidn't you say we need an usb_init() ?
20:10:49[IDC]Dragoncurrently, it's all done in usb_enable()
20:11:04[IDC]Dragoneven the "static" part
20:13:36[IDC]DragonPA5 is set there, it is part of the init, how can this be?
20:14:14[IDC]Dragonwe have a usb_init(), sorry
20:14:33[IDC]Dragonbut all the hardware dependent code is in usb_enable()
20:14:36amiconnNow I know how to zero out the disk, doing it (first GB)
20:14:44[IDC]Dragonbetter leave it like that
20:14:59***Saving seen data "./dancer.seen"
20:15:19amiconndd if=/dev/zero of=/dev/sdg count=2097152
20:15:35amiconnAfterwards creating the partitions
20:19:49amiconn(pa5) Hmm, I though this could have to do with the usb init being done before the adc init needed for usb detection, but that is not the case.
20:19:50 Join mecraw [0] (~lmarlow@
20:21:08amiconn[IDC]Dragon: This could still cause the bug: Interrupts are disabled early in the init. Which function does enable them? Without interrupts, the adc driver wouldn't work...
20:22:31[IDC]Dragoninterrupts get enabled in init()
20:22:56[IDC]Dragonquite early, after the logo
20:24:03amiconnAh, oops, yes. I am silly, level 0 means enabled...
20:24:18[IDC]DragonOT: the disk info is funny, 0.1GB
20:24:44amiconnI know. I think we should go for dynamic scaling here
20:27:09[IDC]Dragonthe MAS registers look OK now
20:28:25amiconnWhat did you change?
20:28:45[IDC]Dragonnothing, I just haven't looked for a long time
20:29:33[IDC]Dragonmaybe the playback fails because the disk is just too slow?
20:29:52[IDC]Dragonlike some race condition at play start, which we now loose?
20:32:10amiconnMy mas still looks wrong :(
20:35:04[IDC]Dragonall zero, or what?
20:35:33amiconnRegisters are all zero, codec registers are all 0x0000FFFF
20:36:12[IDC]Dragonthat indeed looks no good
20:36:23[IDC]Dragonexcuse me for a dinner break
20:37:03amiconnI'll have to trace the call flow. As I don't even get the faint "click" at start, which comes from the poweron reset, I suspect init_playback() doesn't even get called
20:40:10amiconnAh, found it. There is a typo within my MAS3539F conditionals...
20:41:38 Join iSheep [0] (
20:43:20[IDC]Dragonsomething just popped into my mind: is the I2C address the same?
20:43:33[IDC]Dragon...continuing dinner
20:43:37amiconnIt is, all checked
20:44:15amiconnLooks gooood now :)
20:50:07amiconnStill no sound :-/
20:50:44 Join scott666_ [0] (
20:50:52 Join benschi [0] (
20:51:01benschiwhatz up?!
20:56:12[IDC]Dragonamiconn: what did you change?
20:56:37[IDC]Dragonah, your typo
20:57:15[IDC]Dragonthen we're on the same level now, I'd say?
20:57:32amiconnYes I think so
20:57:46benschiany nwes about the iriverport?
20:57:52[IDC]Dragonhow do you like my "too slow" theory?
20:58:14[IDC]Dragonbenschi: wrong crowd, this is Ondio
20:58:37amiconn[IDC]Dragon: I think that it doesn't cause the non-working playback. If it would, the voice ui should work. It doesnt...
20:59:01[IDC]Dragonhaven't tested that
20:59:17[IDC]Dragonhow long to load the voice file?
20:59:35amiconnI just did. The voice file gets loaded (menu doesn't react to keypresses for several seconds), but no voice
21:00:14amiconnLoading deutsch.voice takes ~10 s
21:00:15[IDC]Dragonbut it's not stuck?
21:00:28[IDC]Dragonwoo, this is *bad*
21:00:58[IDC]DragonI should do a load on demand then
21:01:01amiconnI think we won't get below ~6s
21:01:07benschioh, kk,
21:01:10benschicya good night
21:01:27amiconnbenschi: night
21:02:24amiconn[IDC]Dragon: New image dumped, compressing...
21:02:25 Quit benschi ("CGI:IRC (EOF)")
21:02:34[IDC]Dragonamiconn: how's your image coming along?
21:03:03amiconnI *love* crossing messages ;)
21:03:19[IDC]Dragonyes, it reads so strange in the log
21:05:56amiconnImage compressed down to 981 KB
21:06:15[IDC]Dragonthat's a lot better
21:07:27iSheephey guys −− how do I make the daily build boot up on default? Right now 2.2 boots, and the only way to get to the latest build is selecting abjrec.ajz
21:09:10 Nick scott666_ is now known as scott666 (
21:09:12amiconn[IDC]Dragon:"> The bug still happens with that (checked directly after creation)
21:09:53scott666iSheep: assuming youve flashed 2.2, youll need to load a daily build ajz
21:11:02iSheepyeah, I think I flashed 2.2. basically, my problem is with 2.2 now I get that "Dir buffer full" error, and the daily build doesn't give me that problem
21:11:33scott666there should be a file called rockbox.ucl in the .rockbox dir
21:11:40scott666just 'play' that
21:12:25[IDC]Dragonamiconn: got it
21:12:47iSheepgreat, I'll give that a shot
21:13:36amiconn[IDC]Dragon: I'm comparing mas register values: ondio vs. recorder
21:28:29 Join Xiang-Lee [0] (
21:29:00Xiang-Leeno one here...
21:30:02iSheepHi there
21:30:03 Quit Xiang-Lee (Client Quit)
21:30:09iSheepBye there :)
21:30:28 Join Xiang-Lee [0] (
21:31:20Xiang-Leeso quiet..
21:31:33[IDC]Dragonwe're working.
21:31:37iSheepHey Xiang
21:31:41Xiang-Leeah i see
21:31:58Xiang-Leewanna donate something too
21:32:19Xiang-Leeif it helps..
21:32:59[IDC]DragonXiang-Lee: you're most welcome
21:33:05 Quit scott666 ("i'll be back...eventually...")
21:33:08Xiang-Leehaha, thanks
21:33:31Xiang-Leecause i wanted already to sell my iriver, because of 2 reasons, the paint and the bad bad firmware...
21:33:40[IDC]Dragonamiconn: ok, test suite compiled and ready (I hope)
21:33:40 Join scott666 [0] (
21:34:11[IDC]DragonXiang-Lee: what kind of?
21:34:38Xiang-Leehow do u mean? my iriver? my english is a little bad...
21:34:44Xiang-Leeihp 120
21:35:04Xiang-Leehave it for 4 weeks, and the paint is already peeling off...
21:35:09[IDC]Dragonthat's the one they're trying to port Rockbox on, yes
21:35:19[IDC]Dragonpaint is no matter
21:35:19Xiang-Leei know, i know..
21:35:45[IDC]Dragonfrom the core developer, only Linus has one, currently
21:35:51Xiang-Leei will send it back to get a new ihp 120...
21:35:55Xiang-Leei know..
21:36:07[IDC]Dragonso I suggest Zagor is the man to equip next
21:36:22Xiang-Leebut my thing is not broken.., otherwise i would send it to u guys..
21:36:43[IDC]Dragonthey got a broken one now, from ebay
21:36:48Xiang-Leei know
21:37:09[IDC]Dragonnext step would be to have working units for the core developers
21:37:20Xiang-Leei'm surfing everyday, to the mistic river board, and to the rockbox page, checking out the wiki..
21:37:44[IDC]Dragonamiconn: what should I try first? Create a file? 64KB?
21:37:56Xiang-Leeis it already so much progressed, that u have all the access to all hardware components ?
21:38:40amiconn[IDC]Dragon: Perhaps, yes. I tried to dump the ROMs. The dir entry for the boot rom gets created, then the box panics
21:39:56 Quit iSheep ()
21:40:23[IDC]Dragonpartition type is 0x06 for FAT16, 0x0B for FAT32?
21:41:04amiconn0x06 is the fat16 one
21:43:13amiconnFAT32 seems to have different types depending on the size. In the image you got it is 0x0B, while on my recorder it is 0x0C
21:43:40 Quit oxygen77 ("Cho")
21:48:28Xiang-Leetoo complicated for me, hehe..
21:49:18[IDC]Dragonstuck withthe basics: the test code can't find the file
21:53:12 Quit Xiang-Lee ()
21:56:11[IDC]Dragonfound it, it was write-protected
21:59:10[IDC]Dragonamiconn: there is only a "archos.mod" and an empte ".rockbox" on it?
21:59:51amiconnNo, Windows also put it's dreaded "System volume information" there
22:00:08[IDC]DragonI don't see that
22:00:16amiconnIt's a hidden dir
22:00:50[IDC]Dragonmaybe rockbox does not show those
22:01:35amiconnIt does, if "Show files" is set to "all"
22:02:30 Quit kaouete (Remote closed the connection)
22:02:54[IDC]Dragonhmm, the low level test code didn't show it
22:03:14[IDC]Dragonand attributes are not filtered, afaict
22:03:21amiconn[IDC]Dragon: Is there an inverter between PB12 (SCK0) and mas SIBC ?
22:04:41amiconnOn the recorder schematics there is one...
22:05:08[IDC]Dragondo all have that?
22:05:17[IDC]Dragonor is is a mask bit?
22:06:13amiconnIt seems clear that this is the 3rd meaning of mask bit 0 (mas SIBC clock invert). However, just tried the other variant, with no luck
22:06:57[IDC]Dragonour mask indicates inverted, or not?
22:07:19[IDC]Dragongrr, the test code went OK
22:07:24[IDC]Dragonno panic
22:08:05amiconnHmm, strange
22:08:27[IDC]Dragonmaybe a porting issue? liitle/big endian?
22:08:46amiconnyes, maybe
22:10:11[IDC]Dragonthe 64KB file is the first one, and didn't succeed?
22:11:22 Quit _aLF (Read error: 54 (Connection reset by peer))
22:11:23[IDC]Dragoncan you somehow dump the contents of the fat_bpb struct?
22:11:34 Join _aLF [0] (
22:11:38[IDC]Dragonthen we can compare the parameter extraction
22:12:05[IDC]Dragontedious, I know
22:12:13 Join SmoothiRiverOwne [0] (
22:15:00***Saving seen data "./dancer.seen"
22:15:03amiconn[IDC]Dragon: How should I dump if writing to disk panics?
22:15:18[IDC]Dragonto screen...
22:15:49amiconnUrgs. With 2 lines 11 chars... this will take some time
22:16:16[IDC]Dragonok, I try staring at the code first
22:16:16 Quit _aLF (Read error: 54 (Connection reset by peer))
22:16:37 Quit zeekoe (Read error: 60 (Operation timed out))
22:16:41 Join _aLF [0] (
22:16:41 Quit _aLF (Read error: 54 (Connection reset by peer))
22:19:56amiconn[IDC]Dragon: What would you need? All members of the bpb struct?
22:20:50[IDC]Dragonnot the strings
22:21:40[IDC]DragonI don't know "what" I'd need, I'm looking for differences
22:21:56[IDC]Dragonall the info of the fat module is in there
22:22:28[IDC]Dragonif we find a different offset or so, we've won
22:23:06amiconnThere is a struct within the struct. What about that one?
22:24:20[IDC]Dragonthat has only 2 members
22:24:38 Join webguest81 [0] (
22:24:47[IDC]Dragonwe can "unfold" it
22:29:16[IDC]Dragonsystem volume info is there, sorry, I overlooked that
22:30:09[IDC]Dragonfor the dump, probably write the name into line1 , value in line2, then click through member by member
22:30:27 Quit webguest81 ("CGI:IRC (EOF)")
22:30:37[IDC]Dragonit's copy+paste code
22:31:04amiconnYes, already preparing that. Do you want decimal or hex numbers?
22:31:18[IDC]Dragondecimal, I'd say
22:31:33[IDC]Dragonbut it doesn't really matter
22:31:43[IDC]Dragondebugger is switchable
22:40:36 Join _aLF [0] (
22:40:38 Quit methangas (" HydraIRC -> <- The dawn of a new age")
22:47:02amiconn[IDC]Dragon: Test build compiled, let's see if it works
22:48:32amiconnArgh! fat_mount gets called while the display is still in double-height mode...
22:49:34[IDC]Dragondo it later, at the end of init()
22:49:45[IDC]Dragonor what's disabling this?
22:50:15amiconnI try disabling show_logo() sinced this might enable it
22:51:29amiconnWorks now. Should I type the values in here directly?
22:53:20amiconnbpb_bytespersec: 512
22:53:27amiconnbpb_secperclus: 16
22:53:33[IDC]Dragonlet me get mine ready, side-by-side
22:54:11[IDC]Dragonok so far
22:54:37amiconnbpb_rsvdseccnt: 4
22:54:51amiconnbpb_numfats: 2
22:55:14amiconnbpb_rootentcnt: 512
22:55:26amiconnbpb_totsec16: 0
22:55:36[IDC]Dragonjust the number is enough...
22:55:39amiconnbpb_media: 248
22:55:48amiconnbpb_fatsz16: 130
22:55:56amiconnbpb_secpertrk: 63
22:56:04amiconnbpb_numhead: 255
22:56:19amiconnbpb_hiddsec: 1060290
22:56:36amiconnbpb_totsec32: 530145
22:56:46amiconnlast_word: 43605
22:56:53amiconnbs_drvnum: 128
22:57:00amiconnbs_bootsig: 41
22:57:15amiconnbs_volid: 1823131562
22:57:48amiconnbpb_fatsz32: -1440153472 (okay, should have used unsigned display)
22:57:58amiconnbpb_extflags: 0
22:58:04amiconnbpb_fsver: 0
22:58:08 Quit gromit`` ("Client exiting")
22:58:12amiconnbpb_rootclus: -2
22:58:19amiconnbpb_fsinfo: 0
22:58:30amiconnbpb_bkbootsec: 0
22:58:35amiconnfatsize: 130
22:58:45amiconntotalsectors: 530145
22:58:54amiconnrootdirsector: 264
22:59:05amiconnfirstdatasector: 296
22:59:14amiconnstartsector: 1060290
22:59:23amiconndataclusters: 33115
22:59:40amiconn.freecount: 33092 (may differ from yours)
22:59:50[IDC]Dragonone less
22:59:53amiconn.nextfree: 25 (dito)
22:59:59[IDC]Dragonone more
22:59:59 Quit SmoothiRiverOwne ("CGI:IRC (Ping timeout)")
23:00:04amiconnrootdirsectors: 32
23:00:35[IDC]Dragonlooks OK, besides the free stuff
23:00:57amiconnThe free difference is because of the new archos.mod
23:01:22[IDC]Dragonyou have that on the FAT16 partition?
23:01:25 Join gromit` [0] (
23:01:44amiconn[IDC]Dragon: On both. FAT32 for boot, and FAT16 for RoLo
23:02:12[IDC]Dragondepressing, no valuable finding :(
23:03:18[IDC]Dragonyou could do this and the stack vars in the panic
23:03:42[IDC]Dragonbut this is wild probing, no systematic debugging
23:04:59[IDC]Dragonenough for today
23:05:55 Quit [IDC]Dragon ()
23:19:14 Join mecraw_ [0] (~lmarlow@
23:19:14 Quit mecraw (Read error: 104 (Connection reset by peer))
23:22:25 Quit ripnetUK ()
23:22:33 Join SmoothiRiverOwne [0] (
23:23:07SmoothiRiverOwneThe iriver can shuffle the entire drive without a playlist. I wondered if it would save battery power if i made a huge playlist of all the files and shuffled that instead
23:35:57 Join gromit`` [0] (
23:36:47 Quit AciD (""Linux ? Ya moins bien, mais c'est plus cher"")
23:38:41 Quit gromit` (Read error: 110 (Connection timed out))
23:41:22 Join scott666_ [0] (

Previous day | Next day