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

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

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

Notice: Only Gecko based browsers prior to FF4 support the multipart/mixed "server push" method used by this log reader to auto-update. Since you do not appear to use such a browser, this page will simply show the current log, and not automatically update.

#rockbox log for 2008-04-16

00:00:54 Quit Thundercloud (Read error: 110 (Connection timed out))
00:01:37 Quit Mathiasdm ("Yuuw!")
00:02:05 Join CyBergRind|w [0] (n=cbr@
00:08:45*Nico_P finds that buffering after a skip seems much faster with dircache than without
00:09:23Nico_Peven on a flash target, surprisingly
00:11:47 Quit Zarggg_ (Read error: 104 (Connection reset by peer))
00:18:34 Quit cbr|w (Read error: 110 (Connection timed out))
00:20:59 Quit amiconn (Nick collision from services.)
00:21:06 Join amiconn [50] (n=jens@rockbox/developer/amiconn)
00:21:07 Join Zarggg [0] (
00:27:52 Quit ender` (" I'm a complex person. I have a real and an imaginary part.")
00:28:21 Quit BlakeJohnson86 (Read error: 110 (Connection timed out))
00:29:16 Quit bertrik ("bye")
00:30:39 Join knitt1 [0] (n=knittl@
00:36:13amiconnOh, 3 more yllow columns in the table
00:36:18amiconn*yellow even
00:36:59 Quit ompaul (Client Quit)
00:38:27 Join |AhIoRoS| [0] (n=ahioros@
00:39:50 Quit gevaerts ("sleeping time")
00:43:35 Quit herrwaldo ("Konversation terminated!")
00:44:58 Quit knittl (Read error: 110 (Connection timed out))
00:46:31 Nick knitt1 is now known as knittl (n=knittl@
00:53:35Nico_PjhMikeS: ping
00:53:59 Join csc` [0] (n=csc@archlinux/user/csc)
00:54:47Nico_PI've made the changes to the beast's BL to allow updating only the main bin
00:54:57Nico_PI'm testing and will probably commit soon
00:55:21jhMikeSsounds like what's needed about now
00:55:55Nico_Ponly problem is the svn version doesn't seem to be updating itself... I don't know whether that means my change isn't working
00:56:18Nico_P(it happens quite often, even with other builds)
00:56:32Nico_Pso I need a significant change I could check with... any idea?
00:56:33jhMikeSupdating itself?
00:56:59Nico_Pthe svn rev number in the loading screen doesn't change
00:57:50 Quit ]3six5[ ("fish? fish. fish? fish!")
00:57:59jhMikeSMaybe you ran into the same trouble I did? I had to format the storage.
00:58:02 Quit jgarvey ("Leaving")
00:58:30jhMikeSIt actually is extracting?
00:58:42Nico_Pthe file is detected correctly and seems to be renamed correctly too
00:58:55Nico_PI just need a change that will prove it was actually updated
00:59:09amiconnDoesn't the current bootloader just untar a file?
00:59:12jhMikeStoss in a splash screen after the logo
00:59:34*amiconn doesn't understand why this method wouldn't allow to update just the main binary
00:59:56csc`heya rockbox
00:59:57jhMikeSwe want to avoid having to even make the tars
01:00:13amiconnI'd think the bootloader just untars what's in the file, be it a complete file or just the rockbox binary
01:00:20Nico_Puntarring takes time
01:00:22amiconn*complete build
01:00:29 Quit tessarakt ("Client exiting")
01:01:03amiconnI also thought this would be a temporary method anyway
01:01:12Nico_PjhMikeS: OK it works :)
01:01:22jhMikeSbut I'm sick of "make tar", copy that, untar it. better to be able to just drop the binary alone and move it.
01:01:56Nico_PjhMikeS: do you want a sleep call too so that the result is visible?
01:01:58*amiconn wonders which part takes longer - the search for the new file, or the untar action
01:02:11Nico_Pamiconn: the untar action takes time, believe me
01:02:20jhMikeSof course I could shell script it on the make side
01:02:29*amiconn would think it's the former, given the overpowered beast hardware
01:03:07jhMikeSuntarring a full build is pretty lenghty because of the disk reading and writing
01:04:12jhMikeSNico_P: 1/2 to 3/4 second would be ok
01:04:58Nico_Phow many ticks would that be? ~200?
01:05:49jhMikeSNico_P: don't you just load the whole tar into RAM? it's only 5.5megs or so.
01:06:32Nico_Pno, it's copied 512 bytes at a time
01:07:04Nico_PI guess it could be read to memory first
01:07:10amiconnThat's bound to be slooow on a real hdd
01:07:44*amiconn would expect the untar on the beast to be as fast as unzipping a build to another target with the same hdd type
01:07:53Nico_Pok, I'll improve it after the first commit
01:08:21jhMikeSwatch the interrupt vectors, mem is free after that up to the bl load address
01:09:28jhMikeSwell, currently the OFBL sets the core clock divider to 2 (so it's 264MHz) which might slow it down. ;)
01:11:38*amiconn would even expect untar on an archos to be disk bound, not cpu bound
01:15:56 Quit Llorean ("Leaving.")
01:16:08 Quit knittl (Read error: 110 (Connection timed out))
01:17:18 Join Llorean [0] (
01:17:22Nico_PjhMikeS: committed. I'll try to improve untarring now, but I might not commit it tonight
01:18:57***Saving seen data "./dancer.seen"
01:21:34jhMikeSok. no rush with that. cutting out the other step is the most important thing.
01:22:13Nico_Pdo you know where the interrupts vectors end? ie where I could write the tar file?
01:27:47jhMikeSat 0x00000064
01:27:58jhMikeSduh, 0x00000040 :P
01:29:40 Part toffe82
01:30:07 Nick JdGordon|zzz is now known as JdGordon (i=jonno@rockbox/developer/JdGordon)
01:31:07 Join roolku [0] (
01:31:12 Nick fxb is now known as fxb__ (
01:31:13*Nico_P should really write some doc for file.c funcs
01:32:32jhMikeShow many ticks (looking back up). HZ/2 to 3*HZ/4 ? :)
01:32:41 Quit returntoreality ()
01:33:08Nico_Pheh, I went for 100, which is quite a bit shorter but (IMHO) enough :)
01:34:05 Join returntoreality [0] (
01:34:52amiconnNico_P: 100 is longer....
01:35:05amiconnHZ is 100 in rockbox
01:35:28Nico_Poh... I thought HZ was 1000
01:36:49 Join elinenbe [0] (
01:37:31amiconnIt never was... and letting gcc figure out the constants is better than hardcoding anyway, in case HZ will ever change
01:37:43amiconn(it never did in rockbox throughout the years)
01:37:52Nico_PI'll change it them
01:38:11Nico_P(with my next commit of course)
01:40:40 Quit returntoreality ()
01:41:10Nico_PjhMikeS: have you gotten results with USB?
01:43:39 Join Thundercloud [0] (
01:58:14 Quit Thundercloud_ (Read error: 110 (Connection timed out))
01:59:58jhMikeSNico_P: not too much beyond a response and then a hang. can't figure why yet.
02:00:33jhMikeSctrl received 373860 <= ??
02:01:22jhMikeSit all ends there
02:02:58 Part pixelma
02:05:19 Join TaylorKillia1 [0] (
02:05:19 Quit TaylorKillian (Read error: 104 (Connection reset by peer))
02:09:00 Join toffe82 [0] (
02:09:27 Join elinenbe_ [0] (
02:09:35 Quit roolku ()
02:09:53 Quit elinenbe (Read error: 104 (Connection reset by peer))
02:09:53 Nick elinenbe_ is now known as elinenbe (
02:32:13*Nico_P doesn't notice any speed difference
02:41:51 Quit |AhIoRoS| (Remote closed the connection)
02:41:57 Quit Nico_P (Remote closed the connection)
02:49:06 Quit DerPapst (Read error: 113 (No route to host))
03:11:21 Quit Thundercloud (Remote closed the connection)
03:11:57 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
03:12:37 Quit troy__ ()
03:12:54 Join tedrock [0] (
03:19:00***Saving seen data "./dancer.seen"
03:19:33 Join ctaylorr [0] (
03:19:53 Join gagnon [0] (
03:20:31gagnonwhat and where do i need to download to play games over my archos 404 series ?
03:22:20advcomp2019gagnon, since the 404 is not supported you are offtopic
03:23:16gagnonwill it be supported a day ?
03:25:13 Join DavidS1 [0] (
03:25:14advcomp2019that is unknown.. you need to look in the forums
03:26:26gagnonkk thanks
03:27:04 Join daven_ [0] (
03:28:44 Quit daven (Read error: 110 (Connection timed out))
03:29:27 Quit gagnon ()
03:30:01 Nick daven_ is now known as daven (
03:37:16 Join DavidS2 [0] (
03:38:38 Join homielowe [0] (
03:41:01 Join jburd [0] (n=jburd@
03:42:45 Join homielowe_ [0] (
03:42:50 Quit DavidSG (Read error: 110 (Connection timed out))
03:52:16 Quit homielowe_ ()
03:58:04 Quit DavidS1 (Read error: 110 (Connection timed out))
03:59:51 Quit homielowe (Read error: 110 (Connection timed out))
04:00:34 Join homielowe [0] (
04:05:24 Quit TaylorKillia1 (Read error: 110 (Connection timed out))
04:09:47 Join BlakeJohnson86 [0] (
04:17:44 Join midgey_ [0] (
04:22:48 Join midgey__ [0] (
04:23:17 Join boxerorange [0] (
04:24:33boxerorangeI'm putting ubuntu on my old computer. :D
04:31:05 Join miepchen^schlaf_ [0] (
04:35:03 Quit midgey (Read error: 110 (Connection timed out))
04:35:14 Join cool_walking_ [0] (
04:40:07 Quit midgey_ (Read error: 110 (Connection timed out))
04:46:37 Quit miepchen^schlaf (Connection timed out)
04:47:49 Quit jburd (Read error: 110 (Connection timed out))
04:52:34 Join javama1 [0] (
04:58:06 Join javama2 [0] (
04:58:12 Quit javama2 (Client Quit)
05:01:37 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
05:06:00 Quit javaman (Read error: 110 (Connection timed out))
05:08:16 Quit m0f0x ()
05:09:19 Join Shaid [0] (
05:09:33 Quit boxerorange ("ChatZilla 0.9.81 [Firefox]")
05:10:35 Quit ctaylorr (Read error: 110 (Connection timed out))
05:14:08 Quit javama1 (Read error: 110 (Connection timed out))
05:14:40 Join countrymonkey [0] (n=4b05639a@gateway/web/cgi-irc/
05:15:29countrymonkeyWhat's better; sysfont strings with roman letters spell out the language or just plain english?
05:19:03***Saving seen data "./dancer.seen"
05:22:04homieloweHas anyone had problems with using sendfirm to send a rockbox.tar to there gigabeat S? I can send the patched nk.bin and have the gigabeat boot, but after sending rockbox.tar it won't boot and I need to restore
05:24:47 Quit Horscht ("User was distributing pornography on server; system seized by FBI")
05:25:52 Quit flynux (Remote closed the connection)
05:26:00 Join flynux [0] (n=flynux@2a01:38:0:0:0:0:0:1)
05:30:39 Quit tedrock ()
05:31:22 Quit homielowe (Remote closed the connection)
05:41:42 Join homielowe [0] (
05:52:31 Join DavidSG [0] (
05:53:24 Quit homielowe (Remote closed the connection)
05:58:53 Join MobileShaid [0] (
06:00:51 Quit countrymonkey ("CGI:IRC")
06:02:47 Join miepchen^schlaf [0] (
06:10:31 Quit DavidS2 (Read error: 110 (Connection timed out))
06:10:47 Join Shaid3 [0] (
06:16:11 Quit miepchen^schlaf_ (Read error: 110 (Connection timed out))
06:18:04 Quit Shaid (Read error: 110 (Connection timed out))
06:18:06 Nick Shaid3 is now known as Shaid (
06:19:32 Quit jhulst (Remote closed the connection)
06:19:52 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
06:20:39 Join midkay_ [0] (
06:21:27 Quit midkay (Nick collision from services.)
06:21:28 Nick midkay_ is now known as midkay (
06:24:42 Quit csc` ("If you can't do anything smart, do somethin right ~Serenity")
06:29:02 Quit MobileShaid (Read error: 110 (Connection timed out))
06:38:42 Quit miepchen^schlaf ()
06:39:20 Join homielowe [0] (
06:43:27 Join midkay_ [0] (
06:43:43 Quit midkay (Nick collision from services.)
06:43:45 Nick midkay_ is now known as midkay (
06:56:01 Quit midgey__ ()
06:56:43 Join midkay_ [0] (
06:56:53 Quit midkay (Nick collision from services.)
06:56:55 Nick midkay_ is now known as midkay (
07:09:15 Quit jhulst (Remote closed the connection)
07:12:18 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
07:14:17 Join linuxstb__ [0] (
07:19:06***Saving seen data "./dancer.seen"
07:22:09 Quit Battousai (Remote closed the connection)
07:26:17 Quit linuxstb_ (Read error: 113 (No route to host))
07:42:11 Join miepchen^schlaf [0] (
07:52:47 Join davina [0] (
08:03:41 Join Battousai [0] (
08:04:54 Part toffe82
08:05:27 Join mrkiko [0] (
08:06:42 Join jhulst_ [0] (n=jhulst@unaffiliated/jhulst)
08:08:02 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
08:11:51 Quit jhulst (Read error: 113 (No route to host))
08:12:10 Join pondlife [50] (n=Steve@rockbox/developer/pondlife)
08:15:23mrkikoHi all! Rockbox runs so much better in terms of performance
08:15:32mrkikoI apreciated it very much!
08:15:45pondlifeAuto-dir change stopped working though :/
08:18:21pondlifehehe, that build table is getting obese
08:19:32mrkikopondlife: I know
08:19:41mrkikopondlife: aniway you can't have all :)
08:20:57pondlifeHmm, it seems ok now..
08:21:07pondlifeWonder what went wrong?
08:21:20Mouser_XI like big build tables. It means more targets!
08:21:37pondlifeand more sims, and more bootloaders :)
08:21:57Mouser_XWhich are part of the targets.
08:23:32pondlifeAh, I got it... manual dir skip makes playback unhappy..
08:23:54pondlifeSubsequent dir skips are ignored, or lock up
08:23:55Mouser_XThough, it's getting big enough that I wonder if it should be adapted a bit... Make it so that only the names (Sansa e200; Gigabeat F; iAudio M3; etc) show up, and if you mouse over that column, it would show the additional information...
08:24:15pondlife3 tables - one for main, one for bootloader, one for sim.
08:24:29pondlifeAll aligned, of course
08:24:42 Nick linuxstb__ is now known as linuxstb (
08:24:49Mouser_XI guess there's that method as well.
08:24:50pondlifeAnd, finally the main delta
08:25:33 Quit jhulst_ (Remote closed the connection)
08:25:53 Join jhulst_ [0] (n=jhulst@unaffiliated/jhulst)
08:26:55 Join DerPapst [0] (
08:37:36 Quit pabs (Read error: 110 (Connection timed out))
08:42:26 Join Bagderr [0] (n=daniel@gateway/web/cgi-irc/
08:42:45 Join aliask [0] (n=aliask@rockbox/developer/aliask)
08:42:51 Nick Bagderr is now known as B4gder (n=daniel@gateway/web/cgi-irc/
08:44:34 Quit simonrvn ("bbiab, trying to fix something")
08:45:04aliaskhomielowe: I don't think sendfirm will work on non-firmware files - you'll probably want the libmtp application "connect" and use it's sendfile function
08:47:40GodEaterit definitely won't work with the rockbox.tar - even if it did, the .tar file would end up on the wrong partition
08:47:58GodEateryou need mtp-sendfile to transfer the rockbox.tar
08:50:23 Join Rob2222 [0] (
08:51:44 Quit jhulst_ (Read error: 110 (Connection timed out))
08:51:47 Join EspeonEefi [0] (i=espeonee@CERULEANCITY.MIT.EDU)
08:53:51 Join petur [50] (n=petur@rockbox/developer/petur)
08:56:11aliaskAnyone know how rolo_restart actually /restarts/? To me it looks like only PP502x would work
08:56:17 Join icemang [0] (
08:56:38*B4gder likes green tables
08:56:59icemangto update my rockbox build, do I have to wipe out .rockbox first, or can I unzip over the existing .rockbox?
08:57:42Shaidyou can just (usually) unzip it over the top
08:58:06icemangthanks shaid =)
09:00:30homielowealiask: Thanks for the reply, If I have time today, I'll document some stuff in the wiki.
09:01:11aliaskThat'd be really helpful :)
09:01:37Shaiddoes anyone else see a box of 'hidden' advert links on the frontpage? on the left, next to the subversion bit.
09:01:38linuxstbaliask: Which targets are you talking about? IIUC, it just branches to the entry point of the new firmware.
09:01:43*homielowe reboots into ubuntu
09:01:44 Quit mrkiko (Remote closed the connection)
09:01:46 Part homielowe
09:01:57 Join ender` [0] (
09:02:11LloreanCan anyone confirm/test for me? NSF audio doesn't seem to be working on my Gigabeat with a current build.
09:03:01aliasklinuxstb: in firmware/rolo.c - rolo_restart. There's an ifdef'd ASM block which jumps to (what I assume to be) the load address, but other targets (ie, non PP502x) just hit the end of the function.
09:03:32aliaskOh, looks like coldfire also gets a jump
09:03:54aliaskBut that leaves all other ARM targets (and potentially others) left hanging
09:03:57linuxstbaliask: And rolo isn't implemented for all targets either... e.g. not the F.
09:04:27aliaskThat might be why then :P
09:04:51B4gder79 builds in 4:33 makes 3.45 seconds per build...
09:04:57B4gderit might be a new record
09:04:59linuxstbSo yes, I think SH, Coldfire and PP502x are the only CPUs with rolo.
09:05:10linuxstb(SH has an external rolo_restart)
09:05:42*amiconn was about to say that
09:05:54LloreanB4gder: The distributed build system is pretty impressive, overall
09:05:58amiconnIt's in firmware/target/sh/archos/descramble.S
09:06:00aliaskCool. I'll see about implementing it for the S based on the PP502x then
09:06:51icemangHave there been any reports of trouble with today's Sansa e200 builds?
09:08:20B4gderLlorean: it scales really well too, the last ~15 builds added haven't added much to the total build time
09:08:31amiconnClock skipping udelay() has no measurable effect on runtime :\
09:08:45*amiconn tried inlined as well as real function variants
09:08:49 Quit Rob2223 (Read error: 110 (Connection timed out))
09:10:00ShaidRe: the ads on frontpage of the website.
09:10:13ShaidDoes anyone else see those?
09:10:24LloreanB4gder: Do you know of any projects that do as many builds as we do each commit? I'm thinking we probably have the record.
09:10:31 Join homielowe [0] (
09:10:33Mouser_XShaid: They've been for awhile.
09:10:40ShaidAre they meant to be there?
09:10:47Mouser_XThough, by "awhile" I mean for a few weeks.
09:11:10B4gderLlorean: nope, I'm not aware of any
09:11:19ShaidFair enough
09:11:44B4gderShaid: consider them as project sponsors! ;-)
09:11:51Mouser_XPretty much.
09:11:57Shaidadwords, hey
09:12:25B4gderadwords are crap in comparsion...
09:13:13 Join simonrvn [0] (i=simon@unaffiliated/simonrvn)
09:13:18*scorche thinks that a wikipage would fix this and immediately hides so no one tells him to make one
09:13:38*B4gder points in a direction where nobody sits. *he* can do it!
09:13:59B4gder(yes excluding the shes in the audience! :-)
09:14:23scorcheamiconn: have you seem my previous messages about perhaps posting something to the list re: devcon?
09:16:06*homielowe boots rockbox on his gigabeaSt
09:16:25aliaskNice feeling?
09:16:36*homielowe feels all tingly
09:19:07***Saving seen data "./dancer.seen"
09:21:19B4gder <= feel free to extend and improve
09:22:28scorchecan we put a link to it on the front page?...i will try and improve it later
09:22:46 Join MethoS- [0] (
09:22:47scorche(front page right above the ads if possible
09:23:25Mouser_XLlorean: With the newest build, I get strange behaviour...
09:24:00Mouser_XI inserted a NSF into my already existing playlist (insert next), and it played not the NSF, not the file after that, but the one after that...
09:24:20Mouser_X(Though, if I "view current playlist" it says it's playing the NSF.)
09:25:43Mouser_XI skipped back to the NSF, and I get a codec failure.
09:27:01LloreanMine's just skipping over it.
09:27:10LloreanShows (ERR) in front of it in the playlist viewer.
09:27:29 Join PaulJam [0] (
09:27:39Mouser_XHere's what happens: The current song (the one I restarted from poweroff) plays. The NSF doesn't play, but the file after it says (ERR) infront of it.
09:27:52Mouser_XAnd the file after that ERR plays.
09:27:52 Quit PaulJam (Read error: 104 (Connection reset by peer))
09:28:02Mouser_X(So I'm getting 2 files that don't play.)
09:28:11 Join PaulJam [0] (
09:28:26LloreanWell, at least confirmation that something's broken with that format.
09:28:27 Quit OlivierBorowski_ (Read error: 104 (Connection reset by peer))
09:28:45LloreanThough I guess with the strange way it handles tracks-within-tracks it might not be so surprising that it'd break from time to time.
09:28:57Mouser_XThough, if I *don't* play the NSF, and play the (ERR) track first (clarify: before I've attemtped to load the NSF, and therefore there's no (ERR) infront of it), it plays.
09:29:26 Quit TMM (Read error: 110 (Connection timed out))
09:30:04Mouser_XYou have confirmation that NSF is broken (and for me, the track after it), so use it well. :P
09:30:49Mouser_XAh, *now* it's saying (ERR) in front of the NSF...
09:31:07Mouser_X(And the track after it is playing.)
09:33:55 Join OlivierBorowski [0] (
09:35:11 Quit OlivierBorowski (Client Quit)
09:36:38pondlifeMouser_X: Could you try older builds and work out when it broke?
09:36:54 Join Gartral [0] (
09:36:56Lloreanpondlife: Actually, I'm the one who should be doing that.
09:37:06Gartralwhere do i find pacman roms?
09:37:07Mouser_XLlorean beat me to it.
09:37:07pondlifeAh, ok
09:37:14Lloreanhe has no build environment, and I was just asking him to verify something I'd noticed.
09:37:32Mouser_XGartral: It is *very* obviously stated that you are *NOT* to ask that here.
09:38:03Mouser_XRead the manual again.
09:38:11Lloreanpondlife: I have a suspicion as to which revision might've caused it, and will test it later and then file a proper bug report once I know exactly which one does it.
09:38:13Gartralfine... ill go to community
09:38:52Mouser_XRead the manual... It says not to ask *at all*
09:38:58LloreanGartral: It's not that. They're copyright, you need to own an arcade machine and dump them.
09:39:19pondlifeWe respect copyright here.
09:39:31Gartrali technically do..... its just, the power supply is deads
09:40:28scorcheGartral: we respect copyright in -community too...
09:40:32Mouser_XGartral: Then use a search engine, and find out how to fix it.
09:41:13Gartrali did... it requires solder... wich i A) dont have; and B) even if i did, i really suck at soldering
09:41:35aliaskLooks like you're out of luck then.
09:41:37 Quit perrikwp (Remote closed the connection)
09:41:37 Quit GodEater (Remote closed the connection)
09:42:01*Gartral merfs
09:44:42*daurnimator uses dictionarary on Gartral
09:45:04*scorche wonders why daurnimator went and brought that in here
09:45:20daurnimatordifferent universe
09:45:27daurnimatoralso.... merf isn't a word
09:45:55Gartralyour right, its a sound...
09:46:10Gartralmade by me, when disgruntled
09:46:24pondlife"a sound made by a furry character expressing discomfort, uneasiness, surprise, boredom, etc."
09:46:45pondlifeThe kind of noise I make when we go way OT
09:47:40Gartrale.g.- ME the resident furry fox, gotta problem with that?
09:48:13*pondlife only has a problem with the sim constantly segfaulting...
09:49:00scorcheGartral: i do when this conversation isnt happening where it belongs
09:49:15pondlifegdb reckons it's in gwps-common.c:1441... and an attempt to print value gives "$1 = 0x2a8ed249 <Address 0x2a8ed249 out of bounds>" :/
09:51:04 Join DerDome [0] (
09:52:28Lloreanpondlife: It looks like it's revision 17109 that broke it, as I suspected.
09:52:35 Join GodEater [0] (i=c2cbc962@rockbox/staff/GodEater)
09:52:38*Llorean now just needs to flag down Nico_P
09:53:02 Quit Shaid (Read error: 110 (Connection timed out))
09:53:14*B4gder polishes up the "I broke the build" badge in preparation
09:53:16pondlifeMight be that 17109 just revealed a problem with the codec
09:53:37 Quit DataGhost (Nick collision from services.)
09:53:45 Join DataGhost [0] (i=dataghos@unaffiliated/dataghost)
09:53:59Mouser_XSort of how
09:54:04Mouser_XADX doesn't work?
09:54:13Mouser_X(Well, doesn't work how it's supposed to.)
09:54:34Mouser_X(ADXs glitch out when attempting to loop rather often.)
09:55:10Lloreanpondlife: Well, the codec uses some strangeness in supporting single files that contain several tracks, so I imagine the playback rework may not have taken that into account.
09:55:41pondlifeThe problem may be that the codec/playback interface isn't clearly documented...
09:56:06*pondlife wonders if cuesheets are working at the moment.
09:56:43Gartralohh... a bug i found... rockbox will randomly skip songs in the middle of playing them, but only if there are multiple file types too be played from playlist
09:58:56JdGordonI think ive had that, but all tracks being the same type
09:59:16pondlifeI've had that in the past, but not recently
09:59:42LloreanI'm pretty sure that's supposed to have been resolved.
09:59:44 Quit homielowe (Read error: 104 (Connection reset by peer))
10:00:37pondlifeBackward directory skip (either way) is definitely unhappy at the moment.
10:00:48*pondlife packs a pin.
10:00:58 Quit linuxstb (Read error: 101 (Network is unreachable))
10:01:08 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
10:01:15 Join advcomp2019_ [0] (n=advcomp2@unaffiliated/advcomp2019)
10:01:36Lloreanpondlife: I'm having no problem with double-press back on my Gigabeat.
10:01:44 Quit advcomp2019 (Nick collision from services.)
10:01:57 Nick advcomp2019_ is now known as advcomp2019 (n=advcomp2@unaffiliated/advcomp2019)
10:02:14pondlifeHmm, I can segfault the sim reproducibly, and also managed to hard lock my H300
10:02:25LloreanCompletely up to date?
10:03:15LloreanI'm having no problems. I've just done it six times, and hit MP3, FLAC, SPX and AAC audio without any problems.
10:03:24pondlifeMight depend on WPS.
10:03:31pondlifeIt's in gwps-common.c
10:03:39LloreanI'm just using CabbieV2
10:03:40pondlifeI'm using a home brew WPS too
10:03:47pondlifeI'll try resetting settings
10:04:15*Mouser_X already reverted his build to April 7th.
10:04:23Mouser_X(I'm using a custom build.)
10:05:52 Join MethoS-- [0] (
10:06:26Lloreanpondlife: Album art in your WPS?
10:06:55LloreanAnd it happens in "evaluate conditional"?
10:08:11LloreanAnd can you reproduce it with the standard WPS?
10:08:43pondlifeA second issue is that repeatedly back skipping (a track at a time, with auto-change-dir enabled) will sometimes get into a state where I can navigate but playback is stalled. In this state it locks hard when I press STOP.
10:08:49pondlifeHang on, not tried yet
10:09:06pondlifeBut the WPS shouldn't be able to cause this sort of problem - it passes checkwps etc...
10:09:59pondlifeHmm, it works as long as I let the dircache scan complete, I think.
10:10:10*pondlife disables dircache
10:11:11 Join knittl [0] (n=knittl@
10:12:14pondlifeLlorean: Do you have dircache enabled?
10:12:25LloreanDoesn't work without dircache?
10:12:43pondlifeCould you perhaps try disabling dircache, and enable auto-dir-change
10:12:49pondlife(and disable repeat)
10:13:07LloreanIt has to be the auto-change, not manual changes?
10:13:34pondlifeGo to track 1 in a directory somewhere half-way through your collection
10:13:46 Join Mathiasdm [0] (
10:13:50pondlifeThen repeatedly press LEFT / RIGHT
10:14:01pondlifeHence skipping back a dir, then forward again
10:14:10pondlifeWhat happens?
10:14:34pondlifeSkip quickly several times, then wait for it to catch up, perhaps
10:15:02pondlifeOnly 2 tracks should be involved - track 1 in dir B and track n in dir A
10:16:00pondlifeI get a skip back to the last track in the last directory of my entire collection, and it's stuck in non-playing mode
10:16:29Lloreanpondlife: No problems.
10:16:32LloreanWhat format is your audio?
10:16:38pondlifeAll MP3
10:16:44LloreanOkay, that's what I just tested with.
10:16:56pondlifeI'm using the sim, maybe relevant
10:17:31LloreanIt will sometimes say "10 of 0" as playlist position.
10:17:38pondlifeOK, I've just reset settings... trying again
10:18:45 Join gevaerts [0] (
10:18:45pondlifeNope, same problem.. I also saw "1 of 0"
10:18:55LloreanBut I can't cause my gigabeat to crash when that happens.
10:19:21pondlifeIt's not a hard crash this one, but 100% reproducible
10:19:40 Join TMM [0] (
10:19:53pondlifeSo - I have reset settings, then disabled repeat and enabled auto-change-dir
10:21:12pondlifeI go into Files > (artist beginning with M) > (first album) > (track 1)
10:21:49pondlifeDo a back track skip, then go RIGHT-LEFT-RIGHT-LEFT-....
10:22:06pondlifeSo it should be skipping between two tracks.
10:22:22pondlifeIt ends up playing (artist beginning with Z)
10:22:38pondlifeSometimes playing, sometimes stuck
10:23:18Lloreanpondlife: I just followed your recipe exactly, no problems.
10:23:29pondlifeMaybe it's a sim-only thing?
10:23:54pondlifeI've got to work now, will test more later
10:23:58LloreanI thought you said it was also showing problems on your H300.
10:24:19pondlifeYes, a hard lock. But I've been diverted onto this more reproducible problem.. ;)
10:24:47pondlifeI think I've seen 3 different symptoms
10:24:57LloreanAnd I can't reproduce any. :(
10:25:06pondlifeLucky you!!
10:25:14 Quit MethoS- (Read error: 113 (No route to host))
10:25:28LloreanThough I did reproduce a problem with rapidly backskipping through a folder of FLACs
10:25:29 Join pabs [0] (
10:25:37LloreanWhich is why I asked about the format.
10:26:03pondlifeThe common factor in my problems would be a backwards track skip across directories, that's all.
10:26:18 Join ompaul [0] (n=ompaul@gnewsense/friend/ompaul)
10:27:12LloreanAnd see my FLAC problem showed up without a directory barrier being hit.
10:27:37LloreanBut when I tested with MP3s I went back and forth across the directory a few dozen times, both slowly and as quickly as I could press the buttons, without any problem
10:28:25 Join MethoS- [0] (
10:36:11 Quit cool_walking_ (Remote closed the connection)
10:36:26 Join cool_walking_ [0] (
10:46:02 Quit MethoS-- (Read error: 110 (Connection timed out))
10:46:53 Join ddalton [0] (
10:47:00ddaltonpondlife: around...?
10:47:48pondlifeWell, kind of... working mainly..
10:48:07ddaltonpondlife: Did you see my last comment: - with the grep stuff and - with the request for testers? FS #8895... Sorry, I didn't look at the feature request you mentioned... Ran out of time.
10:48:19pondlifeddalton: Not yet.
10:48:39pondlifeThe REC key is pretty much out of bounds until the USB stuff is fixed though, IMHO.
10:48:44ddaltonpondlife: When you get a chance let me know. I should be around for a couple of hours
10:48:51pondlifeWon't be today
10:49:19pondlifeYou have to have a very good reason to change keymaps too, you know..
10:49:34pondlifeI don't think that REC -> FM screen will get into SVN
10:49:41ddaltonI'll do some testing... When you get a chance for a comment just add it and I'll see if I can find the time to look.
10:49:43pondlifeBut, that's just my opinion
10:49:47 Join MethoS-- [0] (
10:50:29Mouser_XCall me crazy, but if the button says REC on it, I'd want it to go to the recording screen.
10:50:50Mouser_X(Unless recording isn't supported on that target, in which case, why does it have a REC button?)
10:51:07ddaltonwell... yes, but short_rec was used for nothing in the menus and file tree. if you guys don't like it after a few months, I'll just keep it in sync for myself... And not sure if I will keep uploading patches if its this hard to get stuff in...
10:52:15ddaltonMouser_X: Well we have rel and repeat... repeat (holding it down) goes to rec screen, and short rec didn't do a whole lot before so now go_to_fm...
10:52:23ddaltoncheck FS #8895 for details
10:52:51ddaltonI'll build a build with this patch and do some tests...
10:52:54pondlifeWhat's so hard about pressing MENU and selecting FM radio?
10:53:04pondlifeSeems more logical
10:53:16pondlifeAnd works from anywhere, pretty much
10:53:23ddaltonpondlife: how exactly should this usb thing act? Where is the keymap for the usb mode you mentioned?
10:53:41ddaltongrep -n 'BUTTON_REC' keymapfile gave no clues...
10:54:20pondlifeIf you plug in a USB cable it will normally connect (and stop playback). If you want it to charge and continue to allow use of Rockbox, then you press REC as you plug the cable in.
10:54:26ddaltonpondlife: Its easy to say the same thing about going to the rec screen...
10:54:37pondlifeUnfortunately, this currently tends to go into the recording screen
10:55:02pondlifeSo, the plan (as much as one exists) is to use the hold button for the USB control.,
10:55:22pondlifeIt makes a bit of sense for a REC button to go to the recording screen.
10:56:00pondlifeLong REC, because you don't want to do it accidentally - think microphone, amp and feedback
10:56:11ddaltonpondlife: sure, but check out FS #7979 or FS #7980 before you do that.
10:56:38ddaltonoh and well... why not change the keymap to long play? From all I can tell long play only does something in the radio?
10:56:58ddaltonOh and grep the keymap file using -n and show me the line with the usb keymap...?
10:57:04pondlifeIt's not a keymap
10:57:09pondlifeIt's hard-coded somewhere
10:57:36ddaltonpondlife: where? in the bootloader?
10:57:49pondlifeNo, in the main code somewhere... I don't know.
10:58:53ddaltonpondlife: for actions to be triggered don't you need a keymap in a keymap file? if not where should i start grepping?
10:59:15pondlifeI don't know, as I just said
10:59:34 Quit cool_walking_ (Remote closed the connection)
11:00:09ddaltonok, but you have seen this actually work, and it is in the manual somewhere?
11:02:02pondlifeYes, and don't know
11:02:08pondlifeI use it all the time
11:04:30ddaltonpondlife: before I waste my time opening another task for another keymap thing I have... do you know if play and rec does anything?
11:05:00pondlifeI don't know, sorry. I'm not an expert on this stuff..
11:05:09 Part Gartral
11:05:15pondlifeBut keymaps should have logical reasons to exist
11:05:30pondlifeDon't go using up combinations just because they're not used, please.
11:05:43pondlifeI mean, is MENU so hard to use?
11:05:52pondlife(for the radio, I mean)
11:06:54pondlifeI suspect it's because you're changing keymaps that your tasks aren't being used.
11:07:18 Quit MethoS- (Read error: 110 (Connection timed out))
11:07:33 Join MethoS- [0] (
11:17:26 Join typhoon07 [0] (
11:18:17typhoon07Just tried to install rockbox with rockbox util and it's killed my ipod, any ideas whats gone wrong?
11:19:08***Saving seen data "./dancer.seen"
11:19:41ddaltonpondlife: ok... well then why does long rec be used for going to the rec screen? why not just press menu and arrow down?
11:20:11pondlifeBecause it's a REC button?
11:20:46gevaertstyphoon07: please provide a bit more information (what model ipod, what exactly you did, what went wrong,...)
11:20:49JdGordonis the miezue music card the miezu possible target? or something different?
11:20:50pondlifeIf you can find a RADIO button, feel free to use it ;p
11:21:06PaulJamtyphoon07: you should be a little bit more specific if you expect someone to help you (what did you do, which ipod, what happens when you try to turn it on)
11:21:31PaulJamops, sorry for the echo.
11:21:34LloreanJdGordon: different.
11:22:10LloreanJdGordon: AFAIK, this is the current Meizu work is being done on:
11:22:26typhoon07ok, 30 gb video ipod, downloaded rockbox util, autodetected the mount point etc, ran complete install, Now when I try to turn in on i get the apple logo in the middle of a dark screen and can hear the HD spinning up and stopping and thats it
11:22:46ddaltonpondlife: so why waste it when it is quite easy to get to the rec screen anyway? I know we don't like looking at the of, but I believe that uses rec to bring up a special menu...?
11:23:45gevaertstyphoon07: what operating system are you using, and what version of rbutil is this ?
11:24:08pondlifeIt's not wasted - although personally I prefer to use the menu and wouldn't mind if it was an unused button (aside from turn on backlight)...
11:24:18pondlifeYou should probably ask whoever added it
11:24:28pondlifeBut you never answered my question
11:24:51JdGordon$200 for 8gb... /me needs a bloody job :p
11:24:54*PaulJam wonders why the A-B button doesn't always set ab markers when the label of the button is so important :P
11:25:12ddaltonpondlife: what was your question?
11:25:13pondlifeIf we had a MENU button, I expect the would
11:25:25pondlife"Is it so hard to use MENU and select FM Radio?"
11:26:11typhoon07im using ubuntu linux, util version m1.0.4
11:26:15ddaltonwell its a lot quicker with a short press especially with voice... and most likely the same reason the long rec keymap was add and the play to resume playback and all the other little shortcuts.
11:27:11 Quit MethoS-- (Read error: 113 (No route to host))
11:27:15pondlifeI don't object massively, but the USB issue needs to be fixed first, that's all.
11:27:50pondlifeWe already have 2 functions on the short REC key (reset settings, USB charge)
11:28:05ddaltonwell if you can show me the page in the manual if you have time so I check the usb stuff and see how it works then great... I don't understand how it is using the keymap though...
11:28:22pondlifeIt's not using the keymap
11:28:43ddaltonyes... so its in the usb code then?
11:28:55pondlifeYes, somewhere
11:28:58typhoon07gevearts −− any ideas?
11:29:54 Part typhoon07
11:29:55PaulJamFrom the RockboxUtility wikipage for rbutil m1.0.4: "Complete Installation still seems buggy. Please use the install options from the second tab, until it's fixed." does taht mean this was fixed or is it still an issue in this in this version
11:30:20 Join typhoon07 [0] (
11:30:42ddaltonpondlife: I'll do some testing/looking when I can be bothered later... :-)
11:30:44pondlifeddalton: I don't see it in the manual, it's missing
11:30:50LloreanPaulJam: If it says "Until it's fixed", and the version provided is still the unfixed version number...
11:31:13PaulJamLlorean: ok
11:31:36ddaltonoh ok... well I'll do some testing here... if it doesn't work for me and is not in the manual I'll assume it doesn't exist and ask about it...
11:31:39ddaltonon the dev list.
11:31:52*gevaerts suggests that someone who has actually used rbutil tries helping typhoon07
11:32:11typhoon07that would be nice
11:32:29 Quit MethoS- (Read error: 110 (Connection timed out))
11:32:30ddaltonpondlife: one other thing, please explain to me what you meant with your comment to sdoyon's patch in 7979?
11:32:52pondlifeIgnore that - I missed the bit about headphones being unplugged
11:32:55Lloreantyphoon07: Try holding Right while booting and see if it causes some debug output.
11:33:31pondlifeI just though that the "Shutting down" voice was added so you'd know if you'd managed to shut it down.
11:33:43PaulJamtyphoon07: it seems as if the complete installation in your version of rbutil is buggy, the wiki says to use the install optons from the second tab. but i'm not sure if that'll work with your ipod in its current state. maybe you'll have to resore it first.
11:33:54pondlifeHence 7979 would be unneedd.
11:33:56typhoon07llorean holding right does nothing
11:34:22ddaltonyes and your saying you should know if you have flicked a switch?
11:35:01typhoon07Pauljam i doubt i can do anything with it in its current state. How do I go about restoring it - plug it into itunes?
11:35:51pondlifeddalton: I just thought that if you heard "shutting down" you'd know you'd got the right button. But, of course, not if you don't have headphones
11:36:00pondlifeSo, ignore that
11:36:14gevaertstyphoon07: reset it (hold menu-select for a while), and then (quickly) hold menu-down (someone check?)
11:36:22PaulJamtyphoon07: i'm not sure. i think you have to force it into diskmode and then resstore with itunes.
11:37:11*gevaerts doubts if itunes restore is really needed. EDM + rbutil (with individual steps) might be sufficient
11:38:08typhoon07think ill throw it in the bin, it wont even reset
11:38:12ddaltonyes, its just a quick way to shutdown and put the player on hold and to no its actually off... doesn't matter about the shutdown message... weather you hear that or not. no point removing it.
11:38:35Lloreantyphoon07: If it won't reset, how did you test my suggestion?'
11:38:59gevaertstyphoon07: try again a few times. It's extremely unlikely to be really broken
11:40:42PaulJamtyphoon07: sometimes it is suggested to turn hold on and off before holding menu+select to reset.
11:42:19LloreanPaulJam: By "sometimes" you mean "it's part of the official Apple instructions, and anyone telling you to should always include this step"? :-P
11:42:34typhoon07that worked to reset
11:42:35PaulJamLlorean: yes
11:43:24 Quit ze (Read error: 110 (Connection timed out))
11:43:45 Join ze [0] (
11:45:31typhoon07no matter what I try after I reset it the HD just spins up then stops repeateddly
11:45:56 Join tvelocity [0] (n=tony@
11:46:18 Join dionoea_ [0] (
11:46:27gevaertstyphoon07: then you should be able to get into emergency disk mode has instructions
11:47:30ddaltonany sansa v2 progress...?
11:47:33BigBambigevaerts: jhMikeS Um, I have an admission and an apology
11:48:09BigBambigevaerts, jhMikeS: It seems when I was doing USB stuff for you yesterday, the beast was suffering from the not updating the build issue
11:48:36typhoon07im in disk mode, still what next?
11:49:22BigBambigevaerts, jhMikeS: I've just formatted it, and transferred a build with USB patch v2 (full speed only) installed, and I have different results than yesterday. Pastebin to follow
11:49:26 Quit dionoea (Nick collision from services.)
11:49:34gevaertstyphoon07: what I would try now is plug it in, re-run rbutil, but instead of doing a complete install, install the bootloader and rockbox separately (I think that is in the second tab)
11:49:40 Nick dionoea_ is now known as dionoea (
11:50:02typhoon07will give that i try cheers
11:55:44typhoon07PaulJam can you link the wiki page you were reading re linux rbutil
11:59:55BigBambigevaerts, jhMikeS: Interesting results:
12:00:31BigBambiI'm just retrying the highspeed to be sre
12:02:03 Join returntoreality [0] (
12:03:31 Quit DerDome (Remote closed the connection)
12:04:09 Join DerDome [0] (
12:05:49BigBambigevaerts, jhMikeS: Confirmed. I disabled tagcache for this build so I'm sure it is the correct one, and the same as last time for the high speed version
12:06:12gevaertsBigBambi: could you try "mount none -t debugfs /sys/kernel/debug/" and "modprobe usbmon" ? If that works, we can get more interesting info
12:06:26BigBambigevaerts: With both versions?
12:06:42gevaertsBigBambi: first try if they work. If they do, I'll provide further steps
12:07:43BigBambigevaerts: Yes, no return from either
12:08:00gevaertsOK. Then "cd /sys/kernel/debug/usbmon"
12:08:14 Quit DerDome (Client Quit)
12:08:45typhoon07it lives
12:08:48BigBambiOK as in done :)
12:09:09typhoon07looks like manual install is the way to go
12:09:30typhoon07cheers for your help guys
12:09:33gevaertsAnd then, "cat 1t > /tmp/1.out" (or 2, depends on the usb 1-1 or usb 2-1) while plugging in. That will capture all usb traffic
12:09:38gevaertstyphoon07: you're welcome :)
12:09:58BigBambigevaerts: So that command, then plug in?
12:10:39gevaertsBigBambi: yes. While the cat is running, all usb traffic will be captured. After plugging in (maybe wait a few seconds), ctrl-c to stop it
12:11:24gevaertsIf the output file is empty, it means you captured the wrong bus number. Then look in dmesg for the correct one
12:11:56 Join DerDome [0] (
12:12:02BigBambigevaerts: OK, will do
12:20:19BigBambigevaerts: <−− full speed
12:24:08 Join FOAD_ [0] (
12:24:54gevaertsBigBambi: is the dmesg for that similar to the one you posted earlier ?
12:25:39BigBambiI'm just doing both dmesg and the other for high speed
12:26:09typhoon07what software would you guys recommend for managing a rockbox enabled player?
12:27:24 Part pondlife
12:27:36BigBambityphoon07: I copy and paste
12:27:38gevaertstyphoon07: I use cp, rm and mv :)
12:27:39PaulJamwhat do you mean by "manage". in order to put music on it you can just use your favourite filebrowser
12:28:07typhoon07oh right no need for any itunes replacement then
12:28:10 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
12:28:12typhoon07thats cool
12:30:42 Join OlivierBorowski [0] (
12:32:11BigBambigevaerts: OK, it appears that sometimes with high speed when I connect it the USB screen etc doesn't happen and dmesg output is different
12:33:31BigBambigevaerts, jhMikeS Both full speed + high speed, on occasions where it connected 'properly':
12:33:37BigBambiSorry for the mucking about
12:35:10gevaertsBigBambi: once you realise that you're just mucking about, you're on your way to enlightenment :)
12:35:22 Part typhoon07
12:35:23BigBambihehe, cheers :)
12:38:32gevaertsBigBambi: the high speed one looks reasonably good : the PC gets correct-looking data back, and it manages to set the usb address correctly. The device then responds to this new address...
12:38:43BigBambiwow, cool
12:39:17Mouser_XIndeed. Good to hear.
12:39:24gevaertsThat means that the ULPI/UTMI/whatever is set up correctly for high speed
12:39:27Mouser_X(Is this being done on the S?)
12:39:33BigBambiMouser_X: yes
12:39:38Mouser_XAwesome news...
12:39:45 Join lastebil_ [0] (
12:39:53*Mouser_X looks forward to continued S progress...
12:40:28*Nico_P cheers for aliask's ROLO work :)
12:40:46Nico_Ponce usb works, testing will be a breeze
12:41:03BigBambiIndeed :)
12:41:37*Nico_P is off to lunch
12:41:45 Quit FOAD (Read error: 110 (Connection timed out))
12:41:45 Nick FOAD_ is now known as FOAD (
12:41:57*gevaerts suspects that the remaining high-speed problem is not usb-hardware related
12:42:12 Quit lastebil (Read error: 113 (No route to host))
12:42:12BigBambiWhat are the other options?
12:42:16 Quit DerPapst (Read error: 110 (Connection timed out))
12:43:14gevaertscaching stuff in usb_core.c
12:43:37BigBambiOne of your specialties? ;)
12:43:50gevaertsno, jhMikeS's speciality :)
12:43:55BigBambihaha :)
12:44:09*GodEater thought anything cacheing related was Cacheri ;)
12:45:22Mouser_XQuestion: How do you get the bootloader onto the S in the first place (when using Windows)? The hacked updater is unreliable.
12:45:41Mouser_X(This is for future referrance.)
12:46:10BigBambiMouser_X: jhMikeS uses windows I think - he would probably know the best way at the moment
12:46:31Mouser_XI'll ask if/when I see him then.
12:46:49Mouser_X(That might not be for awhile... I don't get home until rather late these days.)
12:46:55gevaertsBigBambi: can you add to your stack of patches ?
12:47:08BigBambigevaerts: Sure. On top of the present one?
12:47:17GodEaterafaik the hacked updater is the only way to do it on windows
12:47:29BigBambiWill do
12:47:41 Quit DerDome (Read error: 110 (Connection timed out))
12:47:54gevaertsJust concentrate on high speed for now. full speed seems to need more work
12:49:58gevaertsI wouldn't actually be surprised if the force full speed trick we use doesn't work on the beast. It's explicitely documented as a debug feature, and it might depend on colaboration from the PHY chip
12:50:02*jhMikeS drops in and sees some messing around with USB
12:50:18BigBambijhMikeS: Yeah, sorry
12:50:29BigBambijhMikeS: I cocked up a couple of tests yesterday
12:50:38BigBambiBut that has lead to good news today :)
12:50:50jhMikeSbtw, the S doesn't use IRAM
12:51:10gevaertsOK. then this latest patch will do no good
12:51:43jhMikeSIt can do uncached aliases but I tries a different memory config for the various buffers and used cache coherency operations for transfers with no real change
12:52:01Mouser_XjhMikeS: Is there a reliable method to put the bootloader on the beast? The updater crashes for me. (using WinXP)
12:52:13gevaertsjhMikeS: what we're seeing is that the data that comes back after going through response_data in usb_core.c is bogus
12:52:14jhMikeSI used the windows version of sendfirm
12:52:33Mouser_XAh... Where could I find that? Google?
12:52:59jhMikeSgevaerts: I was going to ask if it actually verified the packet
12:53:00BigBambigevaerts, jhMikeS With the additional patch, no lock up on USB screen (get the screen though) and new dmesg:
12:53:35jhMikeSI have a patch I can give that has the mem changes
12:54:08gevaertsjhMikeS: the device descriptor is OK (comes from a static struct), it responds correctly to set_address, but the config descriptor is not ok
12:54:33Mouser_XI searched Google for "sendfirm windows" and every result I found comes from an IRC log (either #rockbox or #gigabeat)
12:54:50jhMikeScould the lockup the from an invalid config?
12:55:27BigBambigevaerts, jhMikeS And to go with that
12:56:46BigBambijhMikeS: With gevaerts' additional patch it doesn't lock up anymore, and gives output as per my last two pastebins
12:57:06jhMikeSMouser_X: you have to create an .inf, install the LibUSB drivers, use sendfirm, remove the LibUSB drivers. (the driver conflicts with normal MTP)
12:58:18jhMikeSMouser_X: make sure the beast is plugged before running inf-wizard.exe and save the files in the directory with the .dlls
12:59:20Mouser_XjhMikeS: Thanks. It's plugged in already. I was going to leave it all night, and attempt to fix it tomorrow from a different machine.
12:59:31Mouser_XBut I'll try that tool instead.
13:00:24gevaertsBigBambi: the same data now seems to be a different kind of bogus
13:01:29gevaertsI think it's actually more bogus than the first one, so I would suggest to revert that latest patch and wait for jhMikeS to suggest something
13:02:09*jhMikeS begins to suspect he did something stupid with the latest diff
13:02:42BigBambipatch -p0 -R < usb_v2_additional.diff
13:02:45BigBambiheh, oops
13:02:50jhMikeSlike misplace or delete it :O
13:03:42Mouser_XjhMikeS: Create a INF? Manually, or is there a tool to do this (is that what inf-wizard.exe is for)?
13:05:00Mouser_XCan I use sendfirm for other stuff as well? I've been using Winamp to do MTP transfers, but it works poorly.
13:09:16Mouser_XSo I want to send nk.bin?
13:09:56jhMikeSI'm pretty sure
13:10:21*gevaerts suspects that BigBambi still might have the diff that jhMikeS lost
13:10:37jhMikeSthat diff never made it out :(
13:10:46jhMikeSeasy to redo though
13:11:38Mouser_XIt tells me there's no devices attatched...
13:11:57Mouser_X(I made sure it was attached before I started.)
13:13:29Mouser_XAm I supposed to specify the device somehow?
13:14:02jhMikeSwhat do you mean?
13:14:20jhMikeSat what point are you?
13:15:25Mouser_XI made the INF, I right-clicked and selected install (twice now), and I ran sendfirm suing this command: sendfirm nk.bin
13:15:34Mouser_XI get this:
13:15:35Mouser_XLIBMTP_Get_First_Device: No Devices Attached
13:15:35Mouser_XNo devices.
13:15:37jhMikeSno, don't do that
13:16:06Mouser_XSo... I don't know what to do then...
13:16:07jhMikeSupdate from the device manager. select the S under portable devices
13:18:32 Join Thundercloud [0] (
13:18:49*GodEater hugs linux :)
13:19:12 Join tvelocity_ [0] (
13:19:13***Saving seen data "./dancer.seen"
13:19:20 Join miepchen^schlaf_ [0] (
13:19:26*Mouser_X is envious of GodEater right now...
13:19:46*Mouser_X gives up for now.
13:21:26*GodEater can't wait to see what the rbutil boys are going to come up with for the S
13:21:55*Mouser_X was thinking the exact same thing...
13:21:56jhMikeSI'm sure any practical sendfirm for windows will have to not be LibUSB/MTP based
13:22:28*gevaerts spots "practical" and "windows" in the same sentence
13:22:42*GodEater was also pondering this apparent dichotomy
13:23:43aliaskjhMikeS: What speed is the S' CPU running at by default?
13:29:14 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
13:29:46jhMikeSIf you want 528, change the post-divider to 1
13:30:02jhMikeS div by 1 that is
13:30:35aliaskWhat register is that in?
13:33:13aliaskCool. Lets see how high the FPS goes.
13:33:24aliaskIt's already the fastest by a mile.
13:33:31jhMikeSFPS for what?
13:33:51*jhMikeS has branch prediction turned off atm too
13:33:54aliask(I assume)
13:34:24 Quit tvelocity (Read error: 110 (Connection timed out))
13:34:32jhMikeSThings seem somewhat similar to the F at that core speed
13:34:36amiconnaliask: Is it? m:robe 100 gets >1000fps when boosted
13:35:05Mouser_XWoah. Not bad at all...
13:35:06aliaskOh actually, no it's not... you're right amiconn
13:35:35aliaskWe only get 1329.5 fps (not boosted), but sansa has a higher FPS
13:35:35amiconnThe wonders of monochrome lcds :P
13:36:13aliask68,275,200 bytes/s
13:36:40aliaskIt has the highest datarate at least. (of the ones on LcdFrameRate)
13:38:39 Quit tvelocity_ (Read error: 110 (Connection timed out))
13:42:36 Join ctaylorr [0] (
13:43:58 Join Thundercloud_ [0] (
13:44:25 Quit Thundercloud_ (Remote closed the connection)
13:49:39 Join Batelje [0] (
13:50:48BateljeHeey everyone! I'm scrobbling my songs from my nano to using .scrobbler.log .But can i deselect a map from scrobbling?
13:51:13*gevaerts translates : map = directory
13:51:28 Quit J (Read error: 110 (Connection timed out))
13:52:02Bateljei have a directory 'music' with my favourite music in it, and 'Chinese' , where i keep my chinese podcast lessons in it. But i don't want that lessons to be scrobbled. How can i prevent that ?
13:52:35GodEaterI don't think you can at the moment
13:52:40GodEatersounds like a good feature though
13:52:57B4gdersounds like a filter feature for the PC-side tool to me
13:54:56BateljeOk thanks, but what if i only give one tag to the lessons ( for example : Les.1 as title-name for Lesson number 1) so will not accept them?
13:57:07PaulJamBatelje: you could just edit the scrobbler.log file and remove the lines with your language lessons.">
14:00:21Bateljeok thnx pauljam
14:01:55BigBambijhMikeS: On it!
14:02:52 Quit Mathiasdm ("Yuuw!")
14:05:05 Part Batelje
14:10:40 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
14:10:59 Join DefineByte [0] (
14:11:06 Quit returntoreality ()
14:12:53BigBambijhMikeS, gevaerts:
14:13:13DefineByteThis thread got me to thinking: - would it be a terrible idea (probably) to be able to configure Rockbox with RBUtil? I realise the situations where this would be useful are limited.
14:13:28aliaskDoubling the core frequency didn't double the FPS, but it did give a nice bump.
14:14:06jhMikeSaliask: what is the FPS?
14:14:53aliask519 FPS full screen
14:15:10DefineByteI meant up above.
14:15:40gevaertsjhMikeS, BigBambi: here the device descriptor data is bogus as well
14:16:41*jhMikeS checked the pad and pin configuration for the UTMI and it looked good
14:18:02gevaertsjhMikeS: it looks like the structs in usb_core.c need the v2 UNCACHED_ADDR
14:18:26jhMikeSusing UTMI is the only thing that gets a response but it locks...the others don't lock but no response
14:18:51jhMikeSwhich structs...all?
14:18:53 Join TaylorKillian [0] (
14:19:22jhMikeSaren't they const data actually?
14:20:01gevaertsAt least not all of them. The device descriptor is (and is marked as such)
14:21:01jhMikeSare they accessed by the DMA directly or copied?
14:21:13*gevaerts checks
14:21:50 Join DerDome [0] (
14:22:09jhMikeSit appears to all go through usb_drv_send which cleans the lines now
14:22:53 Quit linuxstb (Read error: 110 (Connection timed out))
14:23:03*jhMikeS wonders if it needs the physical addresses for the DMA though
14:23:23gevaertsDoes this thing have an MMU ?
14:23:53GodEatermalloc() ftw \o/
14:24:16jhMikeSand memory isn't mapped as flat...the section with the framebuffer is skipped since I don't know where to change it atm
14:25:10gevaertsok. Then it probably need physical addresses I guess. I think the remapping should be done in prepare_td()
14:27:30jhMikeSit doesn't say anything. this MMU even lets anything at all be remapped so I'm not certain. Before the addresses were physical.
14:29:47jhMikeSI'll put the translation in send/receive
14:30:57gevaertsThe qh->dtd.next_td_ptr in prime_transfer() also needs translation, although it seems to work
14:33:40jhMikeStd_array is already phyical in location
14:33:58gevaertsok. that explains it :)
14:34:13jhMikeSI put the usb buffers in the 1MB block with the ttb and framebuffer
14:40:07 Quit DefineByte ("Bye all")
14:41:08 Quit elinenbe (Read error: 110 (Connection timed out))
14:43:42 Part B4gder
14:45:09 Join mf0102 [0] (n=michi@
14:52:07 Quit aliask (Read error: 104 (Connection reset by peer))
14:53:18 Join aliask [0] (n=aliask@rockbox/developer/aliask)
14:54:16 Nick fxb__ is now known as fxb (
14:56:33 Quit ddalton ("Lost terminal")
14:56:59 Join DarkSaboteur [0] (
14:57:27jhMikeShmmm...who turned my gigabeat into a rockbox media player?
14:58:01 Quit ashes (Read error: 104 (Connection reset by peer))
14:58:15 Join ashes [0] (
14:59:09jhMikeSIt working now but no drive letter
14:59:13aliaskjhMikeS: Sign of working?
14:59:27aliaskHooray :D
14:59:34 Join LambdaCalculus37 [0] (n=44a0430f@gateway/web/cgi-irc/
14:59:35jhMikeSstill finding hardware so maybe it'll show up
14:59:39BigBambijhMikeS: woah
14:59:42jhMikeS"disk drive"
14:59:45BigBambicongrats :)
14:59:58LambdaCalculus37What did jhMikeS do now?
15:00:10Nico_Pwoohoo \o/
15:00:12aliaskThe beginnings of USB :D
15:00:14BigBambiLambdaCalculus37: see the last few lines
15:00:21*LambdaCalculus37 goes to check the logs
15:00:45BigBambiLambdaCalculus37: Sorry, I didn't notice you just joined
15:00:49*aliask hopes it works with his crappy motherboard
15:00:59BigBambijhMikeS: Perhaps a linux test would be useful.... :)
15:01:32LambdaCalculus37Woot! \o/
15:01:45LambdaCalculus37jhMikeS: Congrats! :)
15:03:12 Join perrikwp [0] (i=9821698e@gateway/web/ajax/
15:03:44*GodEater wants to see a commit asap ;)
15:03:47jhMikeS:) drive mounting is getting confused though...may I have to reboot here cause things are running badly anyway. will post a patch first though.
15:04:12GodEater\o/ /o\ \o/
15:04:24*BigBambi is poised... :)
15:05:01 Quit perrikwp (Client Quit)
15:05:10*LambdaCalculus37 readies his Gigabeast... :)
15:05:13jhMikeSI had to plug it twice though. perhaps the tranceiver needs a reset at or a wakeup when first connected.
15:05:41 Quit ctaylorr (Read error: 110 (Connection timed out))
15:05:49jhMikeSyeah, drive mounting isn't going so well on a second try either. :\
15:06:00jhMikeSooh, wait it did it.
15:06:10BigBambijhMikeS: I got that a bit with previous patches - sometimes it worked, other times it didn't
15:06:24BigBambi'worked' as in went to the usb screen for instance
15:06:42GodEaterBigBambi: thought that was fullspeed only though ?
15:06:55BigBambiNo, high speed is/was better than full speed
15:07:13GodEaterthat's what I meant
15:07:24gevaertsBigBambi: once you're testing, you could try rmmodding ehci_hcd, to force full speed from the other side. Maybe that works
15:07:26GodEaterI must have misunderstood you earlier
15:07:29jhMikeSI didn't get any messages saying "This device can perform faster"
15:07:37BigBambiGodEater: Or I wasn't clear
15:07:52BigBambigevaerts: Er, with instructions I could, yes :)
15:08:13GodEaterBigBambi: that *is* the instruction, there's not much else to do
15:08:19BigBambiah, ok
15:08:25BigBambiShows what I know
15:08:26jhMikeSwindows doesn't seem to actually be able to read the disk though (I just hear the head clunking occasionally)
15:08:28gevaertss/ding// :)
15:08:31GodEaterrmmod, then plug it in
15:08:42BigBambigot it! :)
15:10:04 Join midgey [0] (">
15:13:08jhMikeSHow much of the audiobuffer is used?
15:14:23gevaertsOr a few bytes more if the start of the audio buffer isn't 32-byte aligned
15:17:14BigBambijhMikeS, gevaerts:
15:17:46jhMikeSgevaerts: is there any place besides the send/recv functions where I'd have to translate to physical?
15:18:15GodEaterBigBambi: is that all you got?
15:18:19GodEaternot device assigned to it ?
15:18:28BigBambijhMikeS, gevaerts, GodEater No, I cut ot off by mistake:
15:18:57GodEaterstill not quite there then :(
15:19:14***Saving seen data "./dancer.seen"
15:19:38BigBambiHeh, there is still more
15:19:48BigBambiOne mo, I'll post a complete plug unplug cycle
15:21:20 Join perrikwp [0] (i=9821698e@gateway/web/ajax/
15:21:38gevaertsBigBambi: could you enable USB_SERIAL in export/usb_core.h, and enable LOGF_ENABLE in usb_storage.c, usb_core.c and usb-drv-pp502x.c ?
15:22:30BigBambigevaerts, jhMikeS That is from plug, wait a bit, unplug
15:22:34gevaertsBigBambi: and then do a logf build. After that, you'll need to have "cu" installed
15:22:35BigBambigevaerts: Sure, one mo
15:25:32BigBambiOK, just building
15:26:21 Join tvelocity [0] (
15:27:13*GodEater can't get the patch to apply to the tree he just downloaded
15:27:29jhMikeSgevaerts: those descriptors should use NOCACHEDATA_ATTR btw
15:27:47jhMikeSor else cache line flushing is needed after modifiying them
15:28:25gevaertsjhMikeS: some of them are memcpy-ed in place
15:28:36BigBambigevaerts: OK
15:28:58gevaertsBigBambi: once everything seems ready : on the linux side "modprobe usbserial vendor=0x0930 product=0x0010"
15:29:17BigBambithen just plug in?
15:29:19gevaertsOn the DAP side : debug menu -> enable logf over usb
15:29:22jhMikeSthey won't be coherent after the memcpy
15:29:24BigBambiah, ok :)
15:29:43GodEateris the rockbox.7z available for download up to date ?
15:29:52gevaertsThen prepare "cu -l /dev/ttyUSB0" but no enter yet
15:29:56GodEaterI keep getting release r16841
15:30:00BigBambioops, on mo
15:30:11*jhMikeS just recommends NOCACHEDATA_ATTR to make things easy :)
15:30:23gevaertsYou need to start the cu within ten seconds after pluging in
15:31:04 Quit DarkSaboteur ("ChatZilla 0.9.81 [Firefox 3.0b5/2008032620]")
15:31:19BigBambiGot to go and get the mains power
15:32:06BigBambiAnd transferring the logf build not the normal one would help too
15:32:29BigBambiHmmm, the OF thinks the battery is empty, but Rb shows one notch off full
15:32:58jhMikeSI don't think the scale is calibrated yet :)
15:33:00gevaertsjhMikeS: feel free to fix any bug you see :)
15:34:01jhMikeSI don't think we need to use message queues to signal endpoints either. There's the simpler wakeup object now.
15:34:14BigBambigevaerts: OK, ready. So plug, then immediately run cu?
15:34:52gevaertsDon't forget the enable logf over usb debug item
15:35:40BigBambigevaerts: I get permission denied and line in use from cu
15:36:09gevaertsBigBambi: fun. Are you in group dialout ?
15:36:22BigBambinot sure, i'll check
15:36:52gevaertsI usually have to do this as a normal user who's a member of dialout. root gives permission issues
15:37:05BigBambigevaerts: yes
15:37:29BigBambisorry, that was with sudo :) I'll try without :)
15:37:54BigBambigevaerts: OK, I get stuff now
15:38:11BigBambimy terminal is full of "pipe 12 err 8"
15:38:24BigBambiThere was briefly something before that but I can't see it now
15:38:27jhMikeSIn VMWare: Bus 001 Device 019: ID 0930:0010 Toshiba Corp
15:38:36 Join elinenbe [0] (n=elinenbe@
15:39:52gevaertsBigBambi: try setting NUM_ENDPOINTS to 3 in usb_core.h.
15:41:59BigBambihmmm, can't connect to the OF now
15:42:21BigBambiIs that something to do with having modprobed usbserial?
15:42:47gevaertsyou never know... rmmod usbserial
15:43:18BigBambiyes, connects now :)
15:45:33BigBambigevaerts: with endpoints = 3
15:46:03 Join mrkiko [0] (
15:48:13 Nick shenson_not_here is now known as shenson (n=shenson@nat/redhat/x-8d2d9b67ba79f9d0)
15:50:56 Join low_light [0] (i=c730180b@gateway/web/ajax/
15:50:56 Quit alienbiker99 (Read error: 104 (Connection reset by peer))
15:51:20 Quit elinenbe (" Try HydraIRC -> <-")
15:52:56 Quit midgey ()
15:52:57aliaskjhMikeS: What's the story with IRAM on the S? It /is/ there, but we aren't using it? Anything that uses IRAM seems to data abort (at some IRAM address)
15:53:48jhMikeSIt is there, it works but it's only 16K. The bootloader uses it for the remap stub.
15:54:36aliaskIt's used by codecs (primarily) - should we do like the gigabeat f does and fake the IRAM in DRAM?
15:54:47jhMikeSit does that already
15:55:45aliaskAre you sure? Trying to play a file (and thus loading and executing the codec) gives a data abort at 0x1FFFC840, and using mpegplayer 0x1FFFC760
15:58:24LambdaCalculus37aliask: Same here.
15:58:35gevaertsBigBambi: I don't immediately see what's going on. Could you do this again with a corresponding usbmon dump and dmesg output ? I don't have too much time now, but I'll look in depth later tonight
15:58:48BigBambiSure, no probs
15:59:16aliaskAnyway, I'll read the logs. I have to be at uni in 8 hours, so I'm heading to bed :(
15:59:36 Quit aliask ("ChatZilla 0.9.81 [Firefox]")
16:00:54jhMikeSaliask: I'm sure since it copies the stub there and then jumps to 0x1fffc000
16:02:12low_lightjhMikeS: I want to move adc-c200_e200.c and ata-c200_e200.c up on directory to make way for the Philips SA9200 port. I propose to rename them to adc-as3514.c and ata-sd-pp5022.c (or ata-nand-pp5022.c). Any objections?
16:02:58gevaertsSpeaking of renaming... I guess usb-drv-pp502x.c will soon need to be renamed as wll
16:03:05jhMikeSlow_light: not at all. btw, you never committed that change for the radio_power function :)
16:04:05jhMikeSit needs to have a bit more memory buffer abstraction than it does as well
16:04:13gevaertsmaybe. Although ARC is now something else...
16:04:34gevaertsI'll leave those to you :)
16:05:41jhMikeSgevaerts: what do you get from the logs about the continued difficulty?
16:06:58gevaertsjhMikeS: I need more time to look at it than I have right now. I'll look at them later (also at more data that BigBambi should be able to provide)
16:08:40*jhMikeS is wondering if it's just something simple like this last thing. will double-check the pointer fiddling.
16:08:52low_lightjhMikeS: bit sure which radio changes you mean...I commited the fix for the c200 tuning problem
16:10:02BigBambigevaerts, jhMikeS: I hope this is what you were after:
16:10:57jhMikeSlow_light: ah ok, the disable for bit 5
16:11:12 Quit EspeonEefi ("さよなら")
16:11:36 Join jgarvey [0] (
16:12:00gevaertsBigBambi: Those should be enough to start having fun with
16:12:12BigBambiShout if there is anything else I can do :)
16:14:04 Join perrikwp|class [0] (i=982135d2@gateway/web/ajax/
16:18:56 Join toffe82 [0] (
16:31:14 Join gregzx [0] (
16:34:22jhMikeSaren't usb_string_iManufacturer and usb_string_iProduct const too?
16:35:18gevaertsThose could be, yes
16:35:54jhMikeSwhat about usb_strings?
16:36:12jhMikeSdoesn't look modified
16:36:53 Quit perrikwp|class (" ajax IRC Client")
16:37:02 Join FOAD_ [0] (
16:37:10gevaertsthat too
16:38:20SlasheriNico_P: nice that you have done great improvements to the playback.c. Now i like again the way the code going :)
16:38:53Nico_PSlasheri: thanks, that's encouraging :)
16:39:14Nico_PI welcome any suggestions/review of course
16:40:26Nico_PI want to wait a bit for now, until I can fix the bugs I introduced lately, and then finish rewriting audio_check_new_track
16:40:40 Quit gevaerts ("back later")
16:46:55 Join tedrock [0] (
16:53:02 Quit FOAD (Read error: 110 (Connection timed out))
16:53:02 Nick FOAD_ is now known as FOAD (
16:58:29 Join cbr|w [0] (n=cbr@
16:59:05 Quit CyBergRind|w (Read error: 104 (Connection reset by peer))
16:59:52 Quit Nico_P (
16:59:52 Quit HEx (
16:59:52 Quit inakicojk (
17:00:54NJoinNico_P [50] (n=nicolas@rockbox/developer/NicoP)
17:01:09NJoinHEx [0] (
17:01:09NJoininakicojk [0] (i=0@
17:02:37 Join tessarakt [0] (
17:02:37 Join clocKwize [0] (n=mbaldry@
17:02:40 Join linuxstb__ [0] (
17:04:29 Part clocKwize
17:04:47 Join DerPapst [0] (
17:06:30markunjhMikeS: congrats on the USB progress!
17:08:25jhMikeSmarkun: thanks
17:08:51markunjhMikeS: do you use the original vendor id and model id?
17:09:28markunwould be cool if rockbox had it's own vendor id :) but I believe it's rather expensive
17:09:47jhMikeSmarkun: yes, it's pulled from it
17:11:33 Quit linuxstb_ (Read error: 104 (Connection reset by peer))
17:11:40markunnice to see that the port is progressing so quickly now after a period of relative silence
17:12:09 Join returntoreality [0] (
17:12:17jhMikeSwhere'd everyone go that originated it?
17:12:28LambdaCalculus37Who knows?
17:13:03markun'Zunepet' was never really interested in the gigabeat
17:13:10Nico_Pzunepet went back the the zune and ptw419 completely disappeared
17:13:37Nico_Phe never answered any of my emails
17:13:46jhMikeShope nothing bad happened
17:13:52Nico_Pme too
17:14:00LambdaCalculus37Me too.
17:14:45LambdaCalculus37jhMikeS: Once the Gigabeat S is up to a working state, how hard do you think it would be to adapt the port for the V and T series? Just curious.
17:15:19markunLambdaCalculus37: also depends if Toshiba fixed but which makes the exploit possible
17:15:21toffe82T is a flash base player
17:15:21jhMikeSLambdaCalculus37: I'm trying to write things such that's it's not difficult at all.
17:15:24markun(like they did for the Zune)
17:15:42jhMikeSwell, that's another story
17:16:03*GodEater notes it's apparently stripwax's birthday today
17:16:21*LambdaCalculus37 runs off for a moment
17:16:28jhMikeSwhat's keeping the rb stack from being the default on some players? I'm browsing around the H10 just fine.
17:16:38*LambdaCalculus37 wheels back in a cake for stripwax and some birthday beers
17:16:43markunGodEater: would be nice to have a rockbox birthday reminder :)
17:16:46Nico_Pyeah, I'd very much like it to be default on the ipod too
17:16:56GodEaterLambdaCalculus37: you're a bit early - he's not here ;)
17:17:08 Join gevaerts [0] (
17:17:08markunjhMikeS: the speed?
17:17:19markunor does it work at 2.0 speeds?
17:17:28BigBambimarkun: Yeah, high speed works
17:17:33LambdaCalculus37GodEater: We'll light the candles when he gets here. :)
17:17:34jhMikeSthat would a bad reason to completely not include it :)
17:18:05markungevaerts: do you think it's ok to enable it by default now?
17:18:10BigBambiI *believe* it is the flash driver bug on the Sansa's and sometimes signal quality issues (depending on port/hub etc)
17:18:25markunah yes, the hub problems
17:18:39jhMikeShub problems?
17:19:01jhMikeSdrive config isn't right?
17:19:06BigBambiIt gives lots of resets with some hubs and some ports
17:19:18***Saving seen data "./dancer.seen"
17:19:37jhMikeSsounds like a tranceiver issue
17:19:43BigBambiBut I think I'd better leave it to gevaerts to explain, or I'll just get it wrong :)
17:19:57low_lightmarkun: I read in the logs that you were working on "tap" and "slide" type button presses for the gigabeat (or meizu). I've been pondering a similar setup for the touchpads on the m:robe 100 & philips sa9200.
17:20:19low_lightI wonder if the framework could be common
17:20:27markunlow_light: I wrote some code but it works very poorly right now
17:20:35jhMikeSthere's not really much to setup in the ARC controller so it would most likely be hardware if it's not a protocol issue
17:21:07markunlow_light: I think part of the code should be called by the timer tick or interrupt driven
17:21:19jhMikeSuse a vector sum for each line/direction
17:21:51*jhMikeS has the goal to make the S the first tick-task free port
17:22:00markuncool :)
17:22:21gevaertsmarkun: sorry, bit busy right now
17:22:23markunjhMikeS: but the gigabeat F touchpad unfortunately need to be polled
17:22:23low_lightmarkun: what are the issues?
17:22:27 Join n1s [0] (n=nils@rockbox/developer/n1s)
17:23:06markunlow_light: a lot of the touchpad changes were missed by the driver, unless I did something else wrong
17:23:15 Join ZincAlloy [0] (n=d9eedab1@gateway/web/cgi-irc/
17:23:25markunsliding down should have moved the cursor something like 13 positions down in the list
17:23:25jhMikeSmarkun: on the F, the ADC reading _is_ best done the in a tick since it's on SoC. so there's always hardware that it makes sense on.
17:23:41markunbut it usually it was 0 to 2 lines
17:24:33jhMikeSdoes F have edge vs. level sensitivity?
17:24:39markunI don't know
17:25:03toffe82low_light: any progress on the sa9200 ? when do you put it on svn ?
17:26:46jhMikeStoffe82: hey, want to trace-out the rest of the lines on that block diagram (radio lines, headphone mute, others)? :)
17:27:30 Quit mf0102 (Read error: 110 (Connection timed out))
17:27:49 Part LinusN
17:27:51low_lighttoffe82: Hopefully this week. I have the bootloader build done. I need to get the normal build to compile & boot.
17:27:52 Join mf0102 [0] (n=michi@
17:27:57toffe82I need time, kids are really keeping me busy :)
17:28:54*jhMikeS suggests teaching hw hacking to the kids
17:29:14toffe82I try to but they don't listen ;)
17:29:15scorche|shtoffe82: you are supposed to be making them do all the work
17:29:34markunlow_light: does the philips sa9200 only have a vertical pad?
17:29:36LambdaCalculus37You could teach them how to play with a Rockbox sim. :)
17:29:41jhMikeSyes, leverage the child labor cost savings
17:30:12low_lighttoffe82: I've never used an MTP only device. Can you unzip a rockbox build via MTP?
17:30:43LambdaCalculus37low_light: Which MTP only device?
17:30:47BigBambilow_light: For the gigabeat S, the bootloader untars a tar we send it
17:30:50toffe82I don't think so
17:31:05low_lightLambdaCalculus37: the philips
17:31:50LambdaCalculus37low_light: Ah. Like BigBambi said, the bootloader untars a tar that's sent to it. I used Gnomad2 to copy the rockbox.tar to my Gigabeast.
17:32:37LambdaCalculus37I figure that any DAP management app that can handle MTP should allow you to copy a or rockbox.tar to it to allow it to load.
17:33:18jhMikeSe200 works nicely here and the main and card mount nicely
17:33:22Nico_Pyou can now send the plain binary for faster testing
17:33:49BigBambijhMikeS: I think the problem with the e200 is that a bug in the flash driver can occasionally corrupt data transferred over usb
17:34:42jhMikeSyeah, that's a bugger and a mystery
17:34:50LambdaCalculus37Nico_P: I also saw that the bootloader for the S was revised twice.
17:35:00LambdaCalculus37Since Monday, anyway.
17:35:03toffe82low_light: you can copy the file to the system partition and have the bootloader read from there if you stay in mtp mode
17:35:19Nico_PLambdaCalculus37: I added the ability to send a plain binary, then improved the untarring
17:36:10amiconnjhMikeS: Why would you want to do that? Tick tasks are often very useful...
17:36:21LambdaCalculus37Nico_P: I'll see if I can update tonight.
17:36:38LambdaCalculus37But I'm going to be quite busy; I'm looking at a couple of apartments tonight.
17:36:39jhMikeSamiconn: not remove the facility, but the dependence
17:36:40toffe82Nico_P: do you have an idea why somwetimes the album art doesn't display ?
17:37:04jhMikeSit really doesn't depend on them now
17:37:08Nico_Ptoffe82: could you ellaborate?
17:37:18amiconnjhMikeS: Yeah, but why? Sometimes using tick tasks is the easiest and simplest way to do something
17:38:11jhMikeSnot nescessarily the best though. I'd rather the CPU have nearly nothing to attend to.
17:38:53 Quit tessarakt ("Client exiting")
17:39:31 Quit TMM (Read error: 110 (Connection timed out))
17:40:14jhMikeSas it is now it really doesn't. PMIC events set variables that end up getting polled so no work is done there but I wouldn't event need a button tick except to generate repeat events and such.
17:40:31toffe82Nico_P: I have a bmp file on each of my folder and sometimes, it doesn't show
17:41:02Nico_Ptoffe82: is it consistently for the same tracks that it doesn't appear?
17:41:08 Join elinenbe [0] (n=elinenbe@
17:41:29low_lighttoffe82: I'll have to try an MTP transfer sometime. I had success flashing a usb enabled e200 build. But with the potential bug, I'm a bit worried about FS corruption especially since I can't find figure out how to get into manufacturer mode.
17:45:35toffe82no , I think it is random, I listen to my music in shuffle mode so sometimes I see the cover and sometimes not for the some album
17:46:51 Join MacSob [0] (n=c0ea95e0@gateway/web/cgi-irc/
17:47:16Nico_Ptoffe82: I had never heard of such an issue...
17:48:06toffe82I will have a look , but how I listen most of the time in my car, I don't look the display so much
17:51:11 Quit MacSob (Client Quit)
17:54:58toffe82low_light: you have a copy of the original bootloader o?
17:56:09 Quit PaulJam (Read error: 110 (Connection timed out))
17:56:23jhMikeSThe S will probably have a variable CPUFREQ_NORMAL too since it has DVFS and can scale frequency and voltage using hardware based on load
17:59:34markunjhMikeS: is it the only target where we will scale the voltage?
17:59:42 Join webguest22 [0] (n=4df813ab@gateway/web/cgi-irc/
18:00:23jhMikeSmarkun: I don't know. It's specifically designed to work with the MC13783 to accomplish it without much software intervention.
18:01:44 Quit tedrock ()
18:02:31 Quit webguest22 (Client Quit)
18:02:51markunless /tmp/pkg_version
18:02:54 Join giles [0] (n=4df813ab@gateway/web/cgi-irc/
18:03:37markunwhy does file tab completion work in irc ;)
18:04:18*jhMikeS gets yellow spots
18:04:27 Join tedrock [0] (
18:05:12 Quit petur ("work->home")
18:05:17jhMikeS...and I wonder why I didn't see that when building
18:05:34 Join DerPapst_ [0] (
18:05:55 Join roolku [0] (
18:06:27gileshello I'd like write access to the Wiki, to add a new FM transmitter device to TWiki > Main > IpodPort > IpodAccessories, anyone?
18:07:10roolkuNico_P: I can confirm toffe82's experience. Didn't want to report it until I had tried an unpatched build though. When I stop and resume, the albumart is displayed
18:07:16LambdaCalculus37giles: Did you already register?
18:07:28gilesyes, user: GilesNorthRow
18:07:40LambdaCalculus37Okay, one moment.
18:08:51LambdaCalculus37Done. Welcome to Rockbox!
18:08:57LambdaCalculus37No spamming, now! :)
18:09:39 Quit tedrock ()
18:10:59Nico_Proolku: no wps changes involved?
18:11:59roolkuNico_P: no, I usually have the gigabeat in the cradle at my desk and every so often I look up and there is no album art
18:12:24toffe82Nico_P: it is with the default theme cabbie,
18:12:35Nico_Pwell the best would be to find the pattern and describe the circumstances
18:13:33 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
18:14:00roolkuNico_P: yes, I'll try, but other than staring at the display at every track change it is difficult to investigate issue
18:14:55Nico_Proolku: what type of AA do you have? cover.bmp?
18:15:39roolkuNico_P: it is a mixture of all size in the filename
18:16:14roolkuNico_P: I'll try to narrow it down somemore
18:16:28Nico_Pthat would be helpful
18:16:34 Join DavidS1 [0] (
18:17:01UncleRemusA little bit related to rockbox, I hope: Does anyone have a tip for a headphone that is very well suited for listening on trains, buses, etc. and not the "in-ear" type.
18:17:06 Join krz|afk [0] (
18:18:27LambdaCalculus37UncleRemus: That's a matter of personal preference, really.
18:18:44LambdaCalculus37But you could try and buy noise-cancelling headphones.
18:20:53UncleRemusLambdaCalculus37: Yes, thanks. I have tried them, but I think the sound was strange (it was Bose). It wasn't natural sound.
18:21:06n1sor regular closed ones could work pretty well
18:21:11*jhMikeS is amused that rolo works on S before it even plays sound
18:21:28markunand it still doesn't work on the F :)
18:21:56markunjhMikeS: did you make it work just now?
18:22:06Nico_Paliask did
18:22:22UncleRemusn1s: Yes, I thought of closed ones, but closed ones that really isolates from surrounding noise.
18:22:23toffe82jhMikeS: will you have the mpeg working before sound, like on the F :)
18:22:25LambdaCalculus37UncleRemus: Shop around, then. To be honest, not everyone's ears are the same, so just try some headphones until you find your personal preference.
18:22:35 Quit mcuelenaere ()
18:23:09*Nico_P is curious to see the new mpegplayer FPS records :)
18:23:35UncleRemusLambdaCalculus37: I kinda like the Koss PortaPro, but the don't keep the noise out so I have to have volume so high.
18:23:35*LambdaCalculus37 is curious to know what kind of record we can hit on the Gigabeat S :)
18:23:41Nico_PjhMikeS: will mpegplayer be able to make use of the beast's FPU btw?
18:24:42 Join Xerion [0] (
18:26:54 Quit DerPapst (Read error: 110 (Connection timed out))
18:28:02ZincAlloyuncleremus: akg make a closed version of their portable headphones..
18:29:30UncleRemusZincAlloy: Thanks!
18:31:54 Quit DavidSG (Read error: 110 (Connection timed out))
18:32:10ZincAlloyuncleremus: they're called k 414 p and k 416 p. but don't expect too much from them. the open version sounds far better
18:32:29 Join GodEater_ [50] (n=bryan@rockbox/staff/GodEater)
18:34:01 Quit TaylorKillian (Read error: 110 (Connection timed out))
18:34:10 Join Rincewind [0] (
18:35:37 Join EspeonEefi [0] (i=espeonee@STRATTON-ONE-FIFTEEN.MIT.EDU)
18:36:01 Join TaylorKillian [0] (
18:36:31*scorche|sh waves at TaylorKillian
18:37:25 Quit knittl (Read error: 104 (Connection reset by peer))
18:37:33TaylorKillianwhats up scorche?
18:37:44scorche|shnothing...just saying hi :)
18:38:12 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
18:38:13 Join knittl [0] (n=knittl@
18:38:17*scorche|sh had a question, but then found that it was answered
18:38:36GodEater_those are the best sorts of questions
18:38:40preglowbah, no resampler for gsoc...
18:41:08 Join MajorC [0] (
18:41:42 Join tedrock [0] (
18:42:56 Join nplus [0] (
18:44:48 Quit nplus (Remote closed the connection)
18:47:33 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
18:55:05 Join TMM [0] (
18:56:55jhMikeSNico_P: why the FPU? I'd rather it use the IPU.
18:58:54n1spreglow: :(
18:59:07jhMikeSpreglow: no? why not?
19:01:23 Part GodEater_ ("There are two major products that come out of Berkeley: LSD and UNIX. We don't believe this to be a coincidence.")
19:01:32 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
19:06:25 Join PaulJam [0] (
19:08:29 Quit returntoreality ()
19:09:59Nico_PjhMikeS: don't know, I figured maybe mpeg2 decoding would be faster with an fpu, but I guess I assumed wrong :)
19:10:05Nico_Pwhat can the IPU do?
19:11:22toffe82hardware decoding
19:12:49preglowjhMikeS: student doesn't have time for gsoc
19:12:50Nico_Pincluding mpeg2?
19:13:53toffe82sorry, misread, video processing, not hardware decoding
19:14:42 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
19:15:28toffe82Nico_P: The role of the Image Processing Unit (IPU) is to provide hardware acceleration and control for processing and displaying visual data (video and graphics)
19:16:06Nico_Presizing and such, I guess?
19:16:28Nico_Pwhere can I find concise info on the soc?
19:16:38 Quit LambdaCalculus37 ("CGI:IRC (EOF)")
19:18:03jhMikeSNico_P: 2300 pages isn't concise? :)
19:18:26Nico_Phehe, not as much as I'd like ;)
19:18:31jhMikeSthere's a smaller one
19:18:42toffe82page 79 of the 2364
19:18:46 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
19:19:06mcuelenaereis NOCACHEBSS_ATTR defined on non-PP targets?
19:19:10Nico_PI have a 122 page one too
19:19:22***Saving seen data "./dancer.seen"
19:19:32Nico_Ptoffe82: thanks
19:19:45jhMikeSmcuelenaere: no
19:19:55jhMikeSwell, it's defined as nothing
19:19:56toffe82Nico_P: you want the big one
19:20:24mcuelenaerejhMikeS: are you sure it is defined as nothing? because I get a gcc syntax error on it
19:20:42mcuelenaereand it seems in config.h it is only defined ifdef CPU_PP
19:21:27Nico_Pthat IPU looks lovely
19:22:16jhMikeSmcuelenaere: I guess I erred and forgot a default #define
19:23:18jhMikeSwtf. how can I compile gigabeat S with the USB stack then?
19:23:46mcuelenaeredefine NOCACHEBSS_ATTR as nothing?
19:23:59jhMikeSyeah, if not otherwise defined
19:26:02*jhMikeS is confusing himself about his own work
19:28:59 Join LambdaCalculus37 [0] (n=44a0430f@gateway/web/cgi-irc/
19:29:22 Join phrakture [0] (n=griff@archlinux/developer/phrakture)
19:29:22 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
19:30:05phraktureHello all. I just installed rockbox on my iPod 4G, and it's really amazing. I was wondering if it worked on the newer "iPod Classic" - is that technically the same as "iPod Video"?
19:30:21Nico_Punfortunately, no
19:30:53phraktureaha, someone just sent me the link to the IpodPort wiki page, and I see it now
19:31:03phrakture6G has different hardware
19:33:00phrakturethe 5.5G works though, correct? It looks like those are still for sale
19:33:08Nico_Pyes, it works
19:33:15LambdaCalculus37Apple doesn't sell the 5.5G anymore, though.
19:33:24LambdaCalculus37Your best bet is to go onto eBay or Amazon.
19:33:26phraktureamazon does, heh
19:33:36LambdaCalculus37Well, depends.
19:33:43LambdaCalculus37Some sellers on Amazon may have a few in stock.
19:34:52phraktureoddly enough, rockbox fixed my touch-wheel too.... it stopped working at some point, but I didn't think it was hardware failure. it works now though /me shrugs
19:35:42LambdaCalculus37Little do people know, but Rockbox was touched by His Noodly Appendage. :)
19:36:27phraktureIt has Healed my hardware!
19:37:08 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
19:37:13 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
19:37:23 Quit DavidS1 (Read error: 110 (Connection timed out))
19:39:03 Join Thundercloud_ [0] (
19:45:40ZincAlloyomg, the flying spaghetti monster is a rockboxer? I had no idea...
19:45:59LambdaCalculus37ZincAlloy: Of course He is.
19:46:20LambdaCalculus37So are Goddess Eris and J.R. "Bob" Dobbs. :)
19:47:49 Join feindbild [0] (
19:48:53feindbildhi ^^
19:49:27ZincAlloycool, how about the invisible pink unicorn and eric clapton?
19:50:02 Quit linuxstb__ (Read error: 110 (Connection timed out))
19:50:30LambdaCalculus37I can't see if the invisible pink unicorn is using it or not. :P
19:53:12 Quit XavierGr ()
19:54:06 Quit Thundercloud (Read error: 113 (No route to host))
19:54:40 Join pixelma [50] (i=pixelma@rockbox/staff/pixelma)
19:56:31 Join desowin [0] (n=desowin@atheme/developer/desowin)
19:59:27phraktureso here's a side question. because the apple firmware is still on here, can I let itunes update that? or will that blow rockbox away?
19:59:59domonokyphrakture: that will remove the rockbox bootloader...
20:00:16LambdaCalculus37Which means having to reinstall it.
20:00:19scorche|shhowever, it is simple to put it back on
20:00:44scorche|shalthough i dont knwo why one would want to update the apple firmware...
20:01:04phrakturei dunno, I just plugged it into my work box, and itunes started yelling at me
20:01:14phraktureI didn't even know I had itunes installed, this is weird
20:03:33 Join Mathiasdm [0] (
20:04:37 Join bluebrother [0] (n=dom@rockbox/staff/bluebrother)
20:05:01bluebrothergevaerts: just read the SoC mail. What project is the mentioned student?
20:05:16gevaertsbluebrother: usb
20:05:39bluebrotherah, ok. And who is the other guy who withdrew his application?
20:05:45bluebrothers/who/what project/
20:06:08bluebrothertoo bad :(
20:06:20bluebrotherespecially for the resampling project
20:07:06phrakturethanks for the help guys
20:07:07*phrakture waves
20:07:10 Part phrakture ("WeeChat 0.2.6")
20:08:17mcuelenaerewho made ifp_usb_serial.c ?
20:08:27mcuelenaereor does anyone know if it ever worked?
20:09:02*bluebrother guesses tomal
20:09:29 Quit desowin (Read error: 113 (No route to host))
20:10:10bluebrotherheh, seems I was right ;-)
20:11:06Nico_PI have an iFP799 (IIRC), maybe I should test the port
20:13:28bluebrotherhmm. Any news on a DevCon date?
20:13:51LambdaCalculus37bluebrother: I was talking to scorche about one for DCW.
20:14:33bluebrotherLambdaCalculus37: hmm. Nice, but I'm in europe ... ;-)
20:15:02scorche|shbluebrother: well, we need a place first...i am about ready to just say lets go to Stockholm as that seems the most ready at the moment and we need decisions more than a bit more of a central location at this point in the game
20:15:08 Join simonrvn_ [0] (i=simon@unaffiliated/simonrvn)
20:16:24 Join simonrvn1 [0] (
20:16:47bluebrotherI agree that we need decisions. It's kinda hard planning for an event if you don't have the date.
20:17:07 Quit simonrvn (Nick collision from services.)
20:17:07low_lightjhMikeS: seems like the e200 doesn't like to be compiled without HAVE_HOTSWAP
20:17:12 Nick simonrvn1 is now known as simonrvn (
20:17:13bluebrotherand I personally won't mind stockholm, as long as there is enough time to get affordable flights
20:18:06low_lightjhMikeS: the sa9200 doesn't have a sd card, and I'm running into problems if I don't define it
20:18:26low_light^ it = have_hotswap
20:18:34scorche|shbluebrother: which is why i say we need a decision more than a central location at this point
20:19:01bluebrotherI completely agree.
20:19:24bluebrotherit's quite a bit surprising how (not) much discussion happenes about devcon ... or am I only missing it?
20:19:50domonokyyes, its very silent about DevCon..
20:20:55bluebrotheramiconn said that the location in berlin can't get used for sleeping. Does anyone knows about the available alternatives for that?
20:21:08 Quit andrew_ (Nick collision from services.)
20:21:20 Join andrew__ [0] (
20:22:21scorche|shbluebrother: well, i have been trying to nag him about writing up a short mail about some info on it to the ML, but we cant afford (literally) to wait much longer
20:23:17 Join petur [50] (n=petur@rockbox/developer/petur)
20:23:19pixelmaI know he did some research about it (how to organise sleeping) but I don't know the outcome...
20:23:33pixelmaonly vague parts of it
20:24:09scorche|shwell, we can cram into large hotel rooms nearby or hostels, etc
20:25:06bluebrotherwe could postpone devcon after the summer holidays, but that won't solve the decision issue (and I'm not sure if it would be good postponing it that much)
20:26:09scorche|shit also affects the DCW date
20:26:33bluebrotherI wouldn't mind DCW coming first ;-)
20:26:48 Join Bagder_ [0] (
20:26:49scorche|shwell, it is a bit too late for that...
20:27:27 Quit feindbild (Remote closed the connection)
20:30:08 Join desowin [0] (
20:31:58HorschtCharging an ipod 5.5G from within rockbox (holding menu when plugging USB cable) takes a really long time, is this known?
20:32:21 Quit simonrvn_ (Read error: 113 (No route to host))
20:32:24LambdaCalculus37Horscht: Mine takes a long time to charge as well, so you're not alone.
20:32:50LambdaCalculus37Is yours a 30GB or an 80GB?
20:33:01scorche|shHorscht: i think it is still operating at 100 there, so yes
20:33:16HorschtI charged it 4 hours, and it only went up to 40%
20:34:07Horschtso there is no way to connect the ipod to the PC to charge without it being "installed" as a MSD?
20:34:27LambdaCalculus37Horscht: Have you tried with a wall charger?
20:34:33HorschtI don't have one
20:34:42HorschtI ordered one, but it isn't here yet
20:34:54LambdaCalculus37A worthy investment, it was.
20:35:09Horschtit's a Wall -> USB plug to be exact
20:36:25domonokyHorscht: i think there is a button you can press while inserting usb, so that rockbox only charges, and isnt seen as a disk on PC..
20:36:46Horschtyes, I know that
20:36:59Horschtthat's what I did, but it took 4h to charge 40%
20:37:20Nico_Phow does one charge a c200? USB?
20:37:50DerPapst_Horscht: usb doesn't give much power.
20:38:00scorche|shas i said, it only requests 100 (at that stage), so slow charging is expected...especially while the device is still on...if you just want it to charge quickly, boot into the original firmware for now
20:38:13Horschtyes, but chargig 100% with the OF takes only 4h
20:38:25DerPapst_Horscht: i charge my ipod 5.5G 80GBwith firewire and it get 50% after less than 1h
20:38:41domonokyso rockbox doesnt request the full 500ma till now ? wasnt that fixed ?
20:38:41scorche|shHorscht: yes...i just gave you the reason why...
20:39:06DerPapst_Horscht: yes, but it sleeps (idle clock at 32KHz not Mhz)
20:39:12scorche|shdomonoky: i dont think it was for just regular charging, but i am not 100% positive
20:39:29domonokyor did the preparations for the new USB stack remove it again.. :-)
20:39:38HorschtDerPapst_, I think it's actualy taking a lot less than 4h, but I never realy checked :D
20:40:58pixelmaNico_P: yes, via USB (and I trust the OF more for charging)
20:41:46 Quit ompaul (Client Quit)
20:42:32 Quit Bagder (Read error: 110 (Connection timed out))
20:43:23 Join Bagder [241] (n=daniel@rockbox/developer/bagder)
20:43:48Nico_Ppixelma: thanks
20:45:17gevaertsdomonoky: it does _request_ them. It just doesn't use them
20:45:47domonokygevaerts: good to know.. :-)
20:47:33 Join bluebroth3r [0] (
20:47:47 Quit Mathiasdm ("Yuuw!")
20:48:49 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
20:48:52 Join Mathiasdm [0] (
20:52:46gevaertsdomonoky: I actually want to add a secondary 100mA configuration at some point, so that if we connect to a system that can't give 500mA we still can get a data connection
20:52:55 Join GodEater_ [50] (n=bryan@rockbox/staff/GodEater)
20:53:17*Nico_P summons pondlife
20:55:02MajorCgevaerts, what seems to be the main obsticle in the pp usb-implementation?
20:56:23scorche|shMajorC: have you read the wiki page?
20:56:50scorche|shgevaerts: could you clarify a bit about "it requests them, but doenst use them"?
20:57:15 Quit bluebrother (Read error: 111 (Connection refused))
20:57:25gevaertsscorche|sh: it requests 500 mA over USB, but AFAIK there is no code that tells the ipod to actually use more than 100mA
20:57:51scorche|shthat seems a bit...backwards :)
20:57:53 Join Manic [0] (
20:58:08 Nick shenson is now known as shenson_not_here (n=shenson@nat/redhat/x-8d2d9b67ba79f9d0)
20:58:55 Nick shenson_not_here is now known as shenson (n=shenson@nat/redhat/x-8d2d9b67ba79f9d0)
20:59:55 Quit MajorC ()
21:01:33 Quit low_light (" ajax IRC Client")
21:02:00 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
21:02:07amiconnscorche: It means that we have to find out how to tell the charger: "500mA are available, use them if necessary"
21:02:37 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
21:02:39 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
21:04:16scorche|shyes...i realise that...i just wasnt aware of the exact issue and made an off-hand comment..
21:04:59*amiconn scraps his experimental udelay() patches
21:10:34Nico_PMouser_X: you've had problems with recent builds?
21:11:41 Nick bluebroth3r is now known as bluebrother (
21:13:10bluebrotherhmm. Do we need a separate manual for the m3? I guess so
21:14:17 Join lee-qid [0] (
21:15:51 Join ompaul [0] (n=ompaul@gnewsense/friend/ompaul)
21:16:45pixelmabluebrother: and we probably need a solution for "how to include remotes" for it
21:17:18bluebrotheraaah, right. There was this thing about the m3 ...
21:17:37bluebrotherfor now, we could treat the m3 remote simply as main control
21:17:57scorche|shamiconn: thought anymore about getting a mail up about berlin as a location?
21:18:17pixelmabluebrother: the main device also has buttons and you can perfectly control the unit without the remote (except the plugins)
21:18:31 Quit OlivierBorowski (Remote closed the connection)
21:19:27***Saving seen data "./dancer.seen"
21:19:47bluebrotherhmpf. Makes it somewhat hard ... :(
21:19:57bluebrotherhow does the OF handle lcd-less mode on the m3?
21:20:27pixelmafine too (afaik)
21:20:51scorche|shamiconn: what was that? was a bit muffled ;)
21:21:15bluebrotherwhat's the resolution of the remote? Same as the m5 remote?
21:21:55pixelmait's basically the same remote for all three Iaudio Rockbox targets
21:21:55bluebrotherhmm, DeviceChart doesn't hold the m3.
21:22:47 Quit Manic ("")
21:23:11pixelmaI mean the ones with a working port (I believe there was another port started Iaudio7 or so)
21:23:31bluebrotherisn't the D2 an iaudio as well?
21:23:43pixelmaah yes
21:23:48scorche|shat least it is cowon
21:24:02LambdaCalculus37Cowon's site doesn't list it as an iAUDIO device.
21:24:08 Quit Mouser_X (Read error: 104 (Connection reset by peer))
21:24:14pixelmaok :)
21:24:24 Join Mouser_X [0] (
21:24:41LambdaCalculus37I don't think Cowon designates any of their PMP devices as iAUDIO.
21:26:00Bagderbut they do use the iaudio brand in there at times still
21:26:31BagderI recall my D2 having it at some places, can't remember exactly where right now
21:26:51 Join midgey [0] (
21:27:06LambdaCalculus37Bagder: Ah.
21:27:31LambdaCalculus37I've only gotten a brief climpse of the D2 up close, so I don't remember if it did or not.
21:28:22 Join returntoreality [0] (
21:28:51 Quit Seed (Nick collision from services.)
21:28:59 Join Seed [0] (
21:30:20*LambdaCalculus37 is (sort of) considering a Meizu M6SL as well
21:30:26 Quit ch4os_ ("leaving")
21:30:42 Join ch4os_ [0] (n=ch4os@unaffiliated/ch4os/x-059673)
21:31:27bluebrotherpixelma: any objections against committing a preliminary m3 platform file? At least it makes the manual build ;-)
21:31:38bluebrother(and I won't be able filling the buttons −− no device)
21:32:58 Join FOAD_ [0] (
21:33:55*amiconn thinks it doesn't make much sense to fill in button tables for the M3 without having a solution for remote buttons first
21:34:34*scorche|sh wonders if amiconn is ignoring him
21:34:34 Quit LambdaCalculus37 ("CGI:IRC (EOF)")
21:35:26 Part GodEater_ ("Hi, I'm a quit message virus. Please replace your old line with this line and help me take over the world of IRC.")
21:35:35 Join Llorea1 [0] (
21:37:44 Join shisken [0] (
21:38:55 Part shisken ("Ik ga weg")
21:39:40 Join shisken [0] (
21:40:11 Quit desowin (Read error: 113 (No route to host))
21:40:32 Join desowin [0] (n=desowin@atheme/developer/desowin)
21:40:45 Quit midgey ()
21:44:44 Quit shisken ("Ik ga weg")
21:45:29 Join LambdaCalculus37 [0] (n=44a0430f@gateway/web/cgi-irc/
21:45:41 Join shisken [0] (
21:48:04 Join robin0800 [0] (
21:48:17 Quit krazykit (Remote closed the connection)
21:49:17 Quit FOAD (Read error: 110 (Connection timed out))
21:49:18 Nick FOAD_ is now known as FOAD (
21:50:40 Quit mf0102 ("Verlassend")
21:55:30 Quit Bagder (
21:55:30 Quit Llorean (
21:55:30 Quit ali_as (
21:55:30 Quit lostlogic (
21:55:40 Quit shisken ("Ik ga weg")
21:56:27NJoinBagder [241] (n=daniel@rockbox/developer/bagder)
21:56:27 Join Llorean [0] (n=DarkkOne@rockbox/administrator/Llorean)
21:56:27NJoinali_as [0] (
21:56:27NJoinlostlogic [50] (n=lostlogi@rockbox/developer/lostlogic)
21:56:28***Server message 505: 'logbot :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( )'
21:56:44 Quit LambdaCalculus37 ("CGI:IRC 0.5.9 (2006/06/06)")
21:58:00 Join shisken [0] (
22:01:09 Quit shisken (Client Quit)
22:01:24 Join fml [0] (n=4fd3fa12@gateway/web/cgi-irc/
22:02:55fmlmarkun: good morning! Have you seen FS #8899?
22:04:30 Quit davina (Remote closed the connection)
22:05:57 Quit Llorean (Connection timed out)
22:10:09 Quit Rob2222 ()
22:10:33 Join Rob2222 [0] (
22:14:11 Nick Llorea1 is now known as Llorean (
22:15:16 Quit Bagder ("It is time to say moo")
22:15:35 Quit mcuelenaere ()
22:16:56 Join Hilton [0] (
22:17:57 Nick Bagder_ is now known as Bagder (
22:18:11 Join miepchen^schlaf [0] (
22:18:29Nico_PLlorean: hi. things wrong with my latest commits?
22:21:09LloreanNico_P: Haven't checked it yet.
22:22:08Nico_PI saw some talk in the logs
22:25:20LloreanYou already closed the task on flyspray?
22:25:30LloreanFor 17109 I think it was?
22:27:45Nico_PI fixed NSF
22:28:14 Nick Hilton is now known as Hillshum (
22:28:42 Join BuXY95 [0] (n=5b78afc5@gateway/web/cgi-irc/
22:28:46BuXY95hi all
22:28:56 Quit miepchen^schlaf_ (Connection timed out)
22:29:14BuXY95i tried to check all the faq's but i didn't find an important thing (or idea)
22:29:24LloreanNico_P: Okay, that was the one problem I've tracked down and reproduced.
22:29:28BuXY95how about using the iriver ihp-120 as a usb sound card?
22:29:45LloreanNico_P: Pondlife has some problems with the folder skip, when changing folders backward, but I couldn't reproduce on my Gigabeat at all.
22:29:48dionoeaBuXY95: gevearts was something about something like that i believe
22:29:50n1sBuXY95: not possible because of the hardware unfortunately
22:30:03LloreanNico_P: I experienced a "Codec Error" when rapidly skipping FLACs backward, but haven't reproduced it yet or tracked down a revision.
22:30:04dionoeaah ok :) take n1s's answer then :p
22:30:06gevaertsdionoea: not on H100
22:30:37n1sthe usb on iriver h120 (ihp-120) is hardware controlled and only UMS
22:30:40BuXY95too bad, i wanted to buy a usb sound card too, and it looks like i really have to :)
22:31:59BuXY95so i have my "brand new" used ihp-120 here, with original firmware 1.33 (E?). what do you say, should i upgrade it to the latest 1.66 and then install rockbox?
22:32:24Nico_PLlorean: I think I fixed the codec error too
22:32:35Nico_PI tried reproducing pondlife's problem but couldn't
22:33:00n1sBuXY95: no need to upgrade it before, you will need to install a new original firmware together with the rockbox bootloader anyway
22:33:05*Nico_P has to leave
22:33:11 Quit Nico_P (Remote closed the connection)
22:33:27BuXY95oh, stupid me, didn't read the install manual yet :( :)
22:33:49BuXY95you're so kind people you know! :)
22:36:14BuXY95last question: can i download the existing firmware from my ihp120?
22:36:46BagderBuXY95: no, it's only available in flash
22:36:57Bagderso not that easily anyway
22:38:08BuXY95okay, so goodbye mr. 1.33 then!
22:38:28 Quit gregzx ("ChatZilla 0.9.81 [Firefox]")
22:39:30 Join csc` [0] (n=csc@archlinux/user/csc)
22:39:36 Join krz2 [0] (
22:40:43 Quit krz|afk (Read error: 104 (Connection reset by peer))
22:41:53 Join FOAD_ [0] (
22:43:45 Quit fml ("CGI:IRC (EOF)")
22:43:50 Join krazykit [0] (n=kkit@
22:44:29 Quit giles ("CGI:IRC")
22:45:23 Quit lee-qid ("aufwiederbyebientotsayonara")
22:47:52 Join bertrik [0] (
22:48:07preglowjhMikeS: mmm.... armv6...
22:48:16preglowjhMikeS: can you tell me what's the point of the cpy instruction?
22:51:05jhMikeSI'm just doing quicky commits as I discover things and haven't seen that yet.
22:51:38preglowit seems cpy is a version of mov that doesn't touch flags, i just don't get the meaning of that
22:51:46 Join delta009 [0] (
22:52:52delta009May I have write permissions for the wiki, please ? My WikiName is MathieuGuimondMorganti.
22:53:10jhMikeSpreglow: doesn't mov without the s accomplish that?
22:53:36 Quit TaylorKillian (Read error: 110 (Connection timed out))
22:53:44jhMikeSmaybe it's shorthand for some ldr thing
22:53:51preglowjhMikeS: that's what i'm thinking...
22:54:07preglowjhMikeS: only place i've ever seen 'cpy' is the arm quick ref thing
22:54:45jhMikeSI'll just try it
22:56:43jhMikeSgcc doesn't understand it
22:58:37delta009I'm really sorry to interrupt your conversation, but ios there someone here who could grant me the write permission for the Wiki? I would appreciate it!
22:58:55domonokydelta009: your wiki name ?
22:59:13 Quit FOAD (Read error: 110 (Connection timed out))
22:59:14 Nick FOAD_ is now known as FOAD (
22:59:46 Join Alex_SC2e [0] (n=Admin@
23:00:02Alex_SC2eHey guys! I have a question about Rockbox.
23:00:03domonokydone, promise not to spam !
23:00:05 Join phinze [0] (
23:00:26*scorche|sh waves at phinze
23:00:43scorche|shphinze: so what is your gallery project if i may ask
23:00:50 Quit krz2 (Read error: 110 (Connection timed out))
23:00:53jhMikeSpreglow: it looks like a pseudoinstuction that ARM's assemblers use
23:00:56Alex_SC2eI've been using it for about half a year now, but the recent builds are REALLY slow on my Sansa e250r. Also, I have to click before I can use the wheel in every single menu. Does this happen to anyone else, and is there a way to fix it?
23:01:13delta009I promise ;) altough English is not my native language : I'll will correct my mistakes as much as possible using a dictionary before committing any changes to the Wiki.
23:01:24delta009Thank you!
23:01:59 Quit delta009 ("ChatZilla [Iceape 1.1.5/2007100801]")
23:02:18*gevaerts also waves at phinze
23:03:12Alex_SC2eDoes anyone know what I am talking about?
23:04:06Alex_SC2eI know you guys are active, so I know you can hear me. Please answer.
23:04:17LloreanAlex_SC2e: Please don't do that.
23:04:17Bagderwe can read you fine
23:04:36n1sAlex_SC2e: _what_ is it that's slow
23:04:36LloreanAlex_SC2e: Not everyone here is a sansa user, for example.
23:04:36preglowjhMikeS: for what?
23:05:06Alex_SC2eWell, it starts skipping mp3's @ 320 Kbps.
23:05:17LloreanAlex_SC2e: Using what revision?
23:05:29Alex_SC2eAlso, when playing a song, navigating menus is horribly slow. I'm currently using the most updated build.
23:05:37n1sAlex_SC2e: are you using eq/crossfeed/replaygain?
23:05:39LloreanWhat _revision_?
23:06:00Alex_SC2eit's been doing this for about a month. I am not using crossfeed/replaygain. What do you mean by revision?
23:06:07LloreanThe specific version number.
23:06:11 Quit elinenbe (Read error: 110 (Connection timed out))
23:06:23DerPapst_r17234 for example
23:07:12Alex_SC2eOh, ok. Whatever the most recent one is.. I believe r17148
23:07:28DerPapst_you can get it in Debug > Rockbox Info
23:07:40jhMikeSpreglow: "The CPY instruction is a synonym for MOV without the S suffix, and with Operand2 an unshifted register. It copies the unshifted value of Rm into Rd." So IOW I don't get the point over mov Rd, Rm
23:08:24DerPapst_s/debug/system./.. whatever :-P
23:09:26Alex_SC2eMy version is r17148-080416.
23:09:27LloreanAlex_SC2e: And as n1s asked, EQ, Crossfeed, or Replaygain?
23:09:38Alex_SC2eEQ, yes. Everything else, no
23:09:56LloreanHow many bands are set?
23:10:37Alex_SC2eAll 5 are set to at least .5 more
23:10:52LloreanThat's most likely your problem then for the skipping
23:10:58Alex_SC2eshould I disable it?
23:10:59Lloreanthe EQ is very CPU intensive, and the Sansa is not particularly fast.
23:11:22LloreanYou may not need to disable every band, the less of them that are active the less load it creates.
23:11:36*jhMikeS always sets the EQ to have at least one band be 0 gain
23:11:38Alex_SC2eOh. O.o Yeah, the sansa blows. I want to get a new mp3 player, preferably iPod Classic, but I need to wait for Rockbox to come out for it.
23:11:53bluebrotherdon't hold your breath
23:12:18Alex_SC2eYeah, Development's been awfully slow, hasn't it?
23:12:29n1sAlex_SC2e: playing mp3 at such a high bitrate will eat about half the available cpu time
23:12:29bluebrotherthe classic is a completely different beast.
23:12:43bluebrotherthere is no development for the Ipod Classic
23:12:47Alex_SC2en1s: Music sucks at lower qualities xD
23:13:07Alex_SC2eYeah, that's why I'm saying it's slow. It's supposed to be funny.
23:13:12 Quit ompaul (Client Quit)
23:13:39Alex_SC2eBecause if they ever come out with one for classic, it'll be a long time from now, when I'll want a new one anyhow. Maybe I'll get a 16GB Touch.
23:14:00Alex_SC2eWLan = Epic Win, In my opinion. The sansa is slower than a dead horse, I hate it.,
23:14:02BagderRockbox is _we_ not _they_
23:14:06 Nick shenson is now known as shenson_not_here (n=shenson@nat/redhat/x-8d2d9b67ba79f9d0)
23:14:18Bagder... and I like the sansa ;-)
23:14:19LloreanAlex_SC2e: The Sansa has low enough sound quality anyway that 320kbps is definitely a waste.
23:14:21jhMikeSSansas is not slower than any other PP-based device
23:14:56 Quit phinze ()
23:15:00LloreanAlex_SC2e: If you're going to be encoding at high bitrates, you might consider trying musepack. I believe our decoder for that is faster.
23:15:26jhMikeSactually it's a bit on the fast side over things like H10
23:16:26 Quit DerDome ("Leaving.")
23:17:11 Quit PaulJam (".")
23:19:28***Saving seen data "./dancer.seen"
23:21:44 Quit domonoky (Read error: 104 (Connection reset by peer))
23:22:54Alex_SC2eSorry. How do I use musepack on rockbox?
23:23:10BigBambiSame as anything else?
23:23:13Alex_SC2eOh. my apologies, I'm dumb
23:23:16Alex_SC2eIt's a file format. xD
23:23:18BigBambiCopy it over, play it
23:24:09Alex_SC2eThat still doesn't solve the problem of having to click twice for menus
23:24:17 Quit knittl (Remote closed the connection)
23:24:22LloreanAlex_SC2e: I'm not experiencing that problem on my e200.
23:24:26LloreanCould you describe how to reproduce it?
23:25:00 Join phinze [0] (
23:25:03*jhMikeS wanted to recommend APE
23:25:49BigBambijhMikeS is evil :P
23:26:09*n1s want's to commit that japanese translation update but doesn't understand a word of it, do we have a policy or something for unmaintained translations?
23:26:27scorche|shphinze: so what is your gallery project if i may ask
23:27:02 Quit Mouser_X (Nick collision from services.)
23:27:11 Join Mouser_X [0] (
23:27:26Alex_SC2eLlorean: I play an mp3 from my database. From now on, I have to hit a key before the wheel works in menus.
23:28:39*jhMikeS has had that happen but only after exiting plugins
23:28:41LloreanAlex_SC2e: How do you get to the menu without pressing a key? Please, be more descriptive.
23:28:45phinzescorche|sh: "Facebook / Flickr Style Image Region Based Tagging"">
23:29:44scorche|shphinze: well, i wish you luck, and hope you stay a bit with us anyway...even to chat :)
23:29:57 Join waldo [0] (
23:30:41phinzescorche|sh: yeah i'll definitely be around... hoping i can ease into some rockbox community contributions
23:33:50Alex_SC2eLlorean: It's hard. Um. Ok. I hit the select key when I'm highlighting database. After 3 or 4 seconds, it is in the database menu. Now the wheel doesn't work, until I press a key. Now the wheel is enabled (Still in the database menu). It will now do that anywhere and everywhere, all menus.
23:34:43LloreanAlex_SC2e: It doesn't take me 3 or 4 seconds to get into the database menu (about 0.5 to 1) and my wheel is instantly usable...
23:35:19phinzescorche|sh: you definitely have a great base of developers... it's been really enjoyable getting to know the community a bit... seems like a worthwhile place and project to spend some of my spare time on :)
23:35:26 Join Zone-Blaze [0] (
23:36:20Alex_SC2eLlorean: Well, my brother and I both have this issue, it's been happening for a long time
23:37:11Rincewindphinze: I just read your application. I wish I could write half as good as you.
23:37:13LloreanAlex_SC2e: Well, please test to figure out when it began then, and file a proper bug report. It needs someone who can reproduce it to investigate.
23:37:38n1sAlex_SC2e: does it only happen while playing music or always?
23:38:04phinzeRincewind: thanks. :) i almost double majored in computer science and english lit.
23:38:27Zone-BlazeI need help
23:38:27 Quit scorche|sh (Remote closed the connection)
23:38:34gevaertsphinze: are you saying that you're only good at writing applications ? ;)
23:39:07Alex_SC2en1s: Always
23:39:09phinzegevaerts: i'd like to think i'm decent at writing code as well, but i guess i will be finding that out this summer. :D
23:39:55n1sAlex_SC2e: could you test with reset settings? (just a wild guess)
23:40:30 Quit Zone-Blaze (".<UPP>.")
23:40:34 Join scorche|sh [0] (
23:40:40LloreanAlex_SC2e: Do you have "first keypress enables backlight" on?
23:40:55 Quit blithe (Remote closed the connection)
23:41:03*scorche frowns
23:41:05 Join blithe [0] (
23:41:48 Join ze0 [0] (
23:42:13 Quit ze (Nick collision from services.)
23:42:18 Nick ze0 is now known as ze (
23:42:49 Quit tvelocity ("Αποχώρησε")
23:44:35 Quit Rincewind ("bye")
23:45:02phinzesomething got you down, scorche?
23:45:06 Nick fxb is now known as fxb__ (
23:45:32scorchephinze: eh...temporary packet loss
23:45:59phinzethose darn packets, always losing their way
23:46:45 Quit markun ("leaving")
23:49:33 Quit bertrik ("bye")
23:50:43 Quit waldo ("Konversation terminated!")
23:57:19 Join markun [50] (n=markun@rockbox/developer/markun)
23:57:39 Quit bluebrother ("leaving")

Previous day | Next day