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

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

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

#rockbox log for 2009-06-01

00:06:15 Quit merbanan (Read error: 110 (Connection timed out))
00:07:35 Join SirFunk_ [0] (
00:08:03Res1A really nice feature for rockbox would be the ability to use it as a USB soundcard
00:08:15Res1Probably not feasible
00:08:22 Quit SirFunk (Read error: 60 (Operation timed out))
00:08:47krazykitit is on software-usb devices, but what's the point? your computer's DAC is likely better than the DAP
00:09:38 Join matsl__ [0] (
00:09:41Res1Yes, but still useful in a pinch
00:11:02 Join jouke [0] (n=jouke@
00:11:29 Quit wincent (Read error: 110 (Connection timed out))
00:11:39 Quit matsl_ (Read error: 60 (Operation timed out))
00:12:17joukeHi all, Rockbox thinks my new iPod mini battery is empty when the default firmware thinks it's 50%
00:12:36joukecan I force Rockbox to recalibrate?
00:12:36Mikachukrazykit: if you don't have any sound card at all it could be useful, or if you get crap in your laptop line-out or something maybe? :)
00:12:57joukeit's a replacement battery with 700 (real) mAh
00:13:12Mikachui think there's a setting for how big your battery is somewhere
00:13:28Mikachubut i also think it's only used for time remaining estamates, not % left calculations
00:13:57 Quit pixelma (Nick collision from services.)
00:13:59 Join pixelma_ [50] (n=pixelma@rockbox/staff/pixelma)
00:14:12 Quit amiconn (Nick collision from services.)
00:14:14 Join amiconn_ [50] (n=jens@rockbox/developer/amiconn)
00:14:17 Nick pixelma_ is now known as pixelma (n=pixelma@rockbox/staff/pixelma)
00:14:34 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
00:15:17 Quit matsl__ ("Leaving")
00:15:28 Join matsl__ [0] (
00:16:57 Quit ender` (" I have always wished that my computer would be as easy to use as my telephone. My wish has come true. I no longer know how")
00:18:20 Quit matsl__ (Client Quit)
00:19:16 Join lyngaas [0] (
00:21:14soapcorrect, Mikachu
00:22:44soapjouke, I can tell you rockbox is displaying a battery percentage based upon the voltage of the battery, and the known maximum and minimum voltage levels for your battery chemistry. What Apple is doing I can not say - but the /percentage/ displayed by Rockbox should be pretty accurate.
00:23:15joukeokay, since the battery is quite new I will try some more charges
00:23:42joukethe ipod firmware could play for a couple of hours more
00:23:57joukeso perhaps the discharge characteristics are different from standard
00:25:44soapif you battery's discharge curve had, for example, plateaus in it - I would expect Rockbox to keep on ticking and the percentage displayed would simply plateau as well. Is Rockbox shutting down on you "prematurely"?
00:30:12Mikachuif he can play music for hours after i'd say yes
00:30:49joukeyes, Rockbox shuts down with a battery symbol
00:31:19joukebut, again, this is only my first cycle
00:31:20soapand if you reboot into Apple firmware immediately after you can continue to play?
00:31:27joukeyes, that is correct
00:31:29 Join evilnick_home [0] (
00:32:10Res1hm, yes, disk mode is seemingly unstable on my ipod
00:32:51soapwhich version of Rockbox are you seeing this unstableness on, Res1?
00:32:54 Join matsl [0] (
00:33:16Mikachuthe feature i would want is a "safe to unplug" screen
00:33:47 Quit flydutch ("/* empty */")
00:34:26Res1I installed the latest current, version r21151-090531
00:34:50soapand what instability in the disk mode are you seeing?
00:35:07bertrikMikachu, that should be doable technically
00:35:49Mikachui have a hotkey for unmounting the ipod, but it doesn't have any feedback, so i'm always paranoid and look in the logs that it's unmounted, so it doesn't really save any time ;)
00:36:08bertrikbut it's usually the host that tells you it's safe, right?
00:36:46Mikachuyeah, but if i run it from a hotkey i don't get anything
00:38:02Unhelpfulthere could be safety issues even after unmount, if data has not been flushed. unmount *and sync* is safe.
00:40:04Mikachui usually use eject
00:40:22bertrikIIRC, there is a start/stop SCSI command, if the host sent this I think we can assume it's safe
00:40:41bertrikI'm not sure whether all major OSs send this command
00:40:43Mikachuthat seems to be what the apple firmware does, if i just umount, it still shows the stop sign, but after eject it says safe to unplug
00:41:01Mikachuif i umount, i can mount again, but after eject i have to reseat the usb cable
00:41:16Mikachu(still with apple firmware)
00:41:39Mikachuthe same is true for rockbox actually, except it doesn't show anything on screen
00:41:47Mikachuif i eject, the partition devices disappear
00:42:14 Join blaxbb [0] (
00:44:40bertrikI guess an umount does not send the start/stop unit SCSI command yet, but eject does
00:44:57bertrikI think it's best to ask gevaerts when he's back online
00:45:03Mikachuyes, it would be very annoying if umount did
00:47:44 Quit {phoenix} (Remote closed the connection)
00:48:01Mikachuah, gevaerts did the usb stuff?
00:48:02 Quit gregzx ("ChatZilla 0.9.84 [Firefox 3.0.10/2009042316]")
00:48:41 Quit evilnick_home1 (Read error: 110 (Connection timed out))
00:58:28 Quit matsl (Read error: 110 (Connection timed out))
01:01:56 Quit jouke ()
01:04:54 Quit froggyman ("CGI:IRC")
01:04:58 Quit kachna (Read error: 110 (Connection timed out))
01:19:51 Quit bertrik ("Leaving")
01:29:20 Quit PaulJam_ (".")
01:29:41 Join shadoxx [0] (
01:38:50***Saving seen data "./dancer.seen"
01:39:02 Quit Thundercloud (Remote closed the connection)
01:43:40 Join Administrator [0] (
01:44:41 Nick Administrator is now known as Freeman (
01:45:04Freemananyone alive here :)
01:49:19Freemanok, anyway i have the following question
01:49:34Freemanits probably a lame one ...
01:49:50tmztjust ask your question
01:50:17Freemanhow can i compile Sansa Fuze boot file
01:50:46tmztread the instructions, install the cross compiler, build from svn
01:50:55tmztask specfic questions if needed
01:50:59Freemani have done that
01:51:06Freemanthe specific question is
01:51:22Freemanhow to install the bootloader
01:51:41tmztyou have already made one with mkamsboot?
01:51:53Freemanthats just the thing
01:51:56tmztwhat filename did it create
01:52:02Freemani have the original firmware
01:52:14Freemanbut how shall i get the second file
01:52:35tmztselect fuze and bootloader
01:52:58tmztI haven't done this in a while though
01:53:40saratogaFreeman: if the second file is the rockbox bootloader, you compile it from the rockbox source code
01:54:48Freemanyes, thats exactly the second file
01:55:23Freemanthanks a lot for the help, I will read some more and try to make it
02:06:24Freemanin the mkamsboot README
02:06:36Freemansays that I shall scramble
02:06:48Freemanthe rockbox bootfile
02:07:17Freemanany idea which options
02:07:19tmztit compresses the OF bootloader and add the rockbox bootloader to it
02:07:27Freemanshall i use
02:07:31tmztso it can boot both
02:08:00tmztyou need the OF bootloader which it will check to make sure is valid
02:08:05tmztand the rockbox bootloader
02:08:13tmztrun mkamsboot and follow the directions
02:12:12 Quit fyrestorm (Read error: 104 (Connection reset by peer))
02:19:56 Join fyrestorm [0] (
02:28:05 Quit jordan` (Read error: 104 (Connection reset by peer))
02:28:38 Join jordan` [0] (i=gromit@
02:31:15Freemanyeah :D
02:31:35Freemanfist sound from my Fuze running Rockbox
02:33:42Freemangreat job guys !
02:34:21FreemanI will try it out and report ;-)
02:34:35 Quit Freeman ("ChatZilla 0.9.84 [Firefox 3.0.10/2009042316]")
02:53:03 Quit lyngaas (Remote closed the connection)
02:57:00 Join n17ikh [0] (
03:01:27 Quit killan (Read error: 104 (Connection reset by peer))
03:01:28 Join killan_ [0] (
03:06:54 Join froggyman [0] (n=47ba0b80@gateway/web/cgi-irc/
03:06:56 Quit killan_ (Read error: 104 (Connection reset by peer))
03:07:06 Join killan [0] (
03:19:20 Join Emobathtub [0] (
03:19:32EmobathtubHello! is anyone here?
03:20:28EmobathtubI have a question and google cannot answer it
03:22:10saratogaask it
03:22:42EmobathtubI can't seem to get the "Recently Added" option in the database to work. I have a sansa e250R
03:24:29 Quit perrikwp (" ajax IRC Client")
03:26:35 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
03:27:27EmobathtubAh, but silly me to think you'd answer. Good night.
03:27:29 Part Emobathtub
03:32:26 Quit froggyman ("CGI:IRC")
03:32:51 Join froggyman [0] (n=47ba0b80@gateway/web/cgi-irc/
03:38:54***Saving seen data "./dancer.seen"
03:40:00 Join AndrewRB [0] (
03:40:26 Quit dmb (Read error: 104 (Connection reset by peer))
03:40:35AndrewRBhey, could anybody help me figure out what I am doing wrong with patch?
03:42:03AndrewRBfresh svn checkout, i'm in the rockbox directory... "patch < patchfile" always says can't find file to patch
03:42:17Mikachutry -p0 or -p1 (before the <)
03:42:33Mikachulook in the patch file what path it has for the file
03:42:33AndrewRBi'm in the root dir. that doesnt work
03:42:57AndrewRBwhenever i copy/paste the path which it prints out (from the patch file), it works.
03:43:14 Join dmb [0] (n=dmb@unaffiliated/dmb)
03:43:30AndrewRBhold on, let me explain better.
03:45:22AndrewRBthat shows me trying to apply a patch which I just created.
03:45:41Mikachui notice you're not using -p0 there
03:46:32AndrewRBah. ok now i feel really stupid. i could have sword i tried that after reading the man page.
03:46:40Mikachuor after i asked you to?
03:46:53AndrewRBi disregarded that as i was dead-certain i had tried it
03:47:10Mikachuokay, i'll remember to tell you to do things twice in the future :)
03:47:44AndrewRBtwice makes me doubt myself ;-)
04:00:28 Quit pat_mulchrone (Read error: 110 (Connection timed out))
04:03:02 Quit froggyman ("CGI:IRC")
04:09:14AndrewRBcould anybody suggest a reason for changing this line in the D2's button-target.h file:
04:09:19AndrewRB#define BUTTON_MAIN 0x3FFF
04:09:22AndrewRB#define BUTTON_MAIN 0xFFFF
04:15:58 Join pat_mulchrone [0] (
04:23:09 Quit dfkt ("-= SysReset 2.53=- Ph'nglui mglw'nafh Cthulhu R'lyeh wgah'nagl fhtagn.")
04:42:03AndrewRBno idea? or did I just ask a stupid question (again ><)
04:50:20 Join onlysoaa [0] (i=42248456@gateway/web/ajax/
04:51:12onlysoaaHello all, anyone got a clue if/how I can build Rockbox as a sim for Windows Mobile?
04:52:49evilnick_homeI don't *think* you can, but I'm not positive
04:53:27Unhelpfulif you have SDL for windows mobile, it might be possible. i doubt it will happen without some work, though. i don't think anybody's tried, really.
04:53:54onlysoaaI read something obscure about cegcc; apparently someone tried before but didn't follow up.
04:54:34onlysoaaSee, I have two devices, and I'd like Rockbox on one of them. One is my HTC Touch Diamond that runs WinMo, and the other is a Samsung P2.
04:55:13onlysoaaSome might remember me fail-attempting to port it to the P2 before. I just lack experience with porting apps, but I'm willing to learn.
04:56:04onlysoaaI figure it's probably easier to port it to the P2 since the Cowon D2 uses the same SoC and firmware format... I have no clue where to start though, as I've never coded drivers before.
04:57:45onlysoaaWould anyone be willing to explain to a casual hacker how the an LCD driver for Rockbox would work?
04:58:26krazykitit seems like reading the code would be the most obvious place to start
05:00:20 Join kugel [0] (n=kugel@rockbox/developer/kugel)
05:00:49kugelonlysoaa: IIRC, the D2 Rockbox binary runs on the P2 without modifications
05:01:02onlysoaaI've tried before. ;) But with little success. It was like a totally foerign language, although I guess another go can't hurt.
05:01:34kugelhowever, the lcd driver needs some adjusting, as it's pretty much hardcoded for the D2's resolution
05:01:39onlysoaakugel: Whoa, what? Does it? o_o I guess I'll give it a go and see what works and what doesn't.
05:02:05onlysoaaI'd also need to reimplement a touch driver of course. The P2 uses a capacitance screen.
05:02:26onlysoaaWell thanks for the notice. *takes the leap*
05:02:43kugelonlysoaa: please read this forum thread:
05:03:08kugelat the end, I posted a link to the IRC logs where I was talking with the guy that just tried to run the D2's rockbox version
05:03:47kugelHowever, in either case, even the D2 Rockbox is unsupport and very premature. That of course applies to the P2 too
05:04:58onlysoaaCool, thanks. I see there's no current build for the D2 either.
05:05:19kugelNo, you'll need to compile it yourself
05:05:53onlysoaaAlrighty. Thankfully I already know how. :P
05:06:15kugelPlease post your findings to that forum thread, I think with some effort the P2 port can step up to the D2 port quite quickly
05:07:07onlysoaaWhoa, isn't that asking a bit too much? :P I guess I'll contribute in whatever way I can, but I don't think I'll be able to go much farther than where the D2 port is at.
05:15:01kugelonlysoaa: no, I just asked to post your findings to the forum thread
05:15:23kugelthe 2nd sentence was just a general statement from my point of view
05:15:42onlysoaaOhh. For sure I'll post my findings.
05:18:04 Quit kugel ("exit(0);")
05:21:10Unhelpfulare our htole macros pretty fast? i assume they translate to machine swap instructions where possible?
05:24:26 Quit AndrewRB ("Goodnight")
05:29:55 Quit chandoo (Read error: 104 (Connection reset by peer))
05:33:48 Part blaxbb
05:38:12 Quit fdinel ("Miranda IM! Smaller, Faster, Easier.")
05:38:55***Saving seen data "./dancer.seen"
05:48:15 Join grndslm [0] (n=grndslm@
06:01:00Unhelpfulany thoughts on this? i intend to use it to create invariant key values for the pictureflow cache, instead of using the index in the cover list. this will allow adding/removing albums without the cached covers becoming associated with the wrong albums.
06:02:17Unhelpfuli'm planning to word-align and zero-pad the strings so that i can just use hashw, which is smaller/faster, but i figured the byte-array variants might be useful as well.
06:20:34 Join lucent [0] (i=lucent@unaffiliated/shadows)
06:35:40 Join AndyI [0] (i=AndyI@
06:36:01 Quit yosafbridge ("Coyote finally caught me")
06:36:21 Join yosafbridge [0] (
06:46:57 Quit dmb ("Leaving")
06:49:38 Quit AndyIL (Read error: 110 (Connection timed out))
06:53:04 Join dmb [0] (n=dmb@unaffiliated/dmb)
07:04:38 Quit jmillikin (Read error: 110 (Connection timed out))
07:09:14 Join killan_ [0] (
07:12:25 Quit martian67 (
07:12:25 Quit grndslm (
07:12:25 Quit killan (
07:12:25 Quit safetydan (
07:12:25 Quit MarcGuay (
07:12:25 Quit vedlith (
07:12:25 Quit Horschti (
07:12:25 Quit advcomp2019 (
07:12:25 Quit goffa (
07:12:34NJoingrndslm [0] (n=grndslm@
07:12:34NJoinkillan [0] (
07:12:34NJoinsafetydan [0] (n=deverton@rockbox/developer/safetydan)
07:12:34NJoinmartian67 [0] (n=martian6@about/linux/regular/martian67)
07:12:34NJoinMarcGuay [0] (
07:12:34NJoinvedlith [0] (
07:12:34NJoinHorschti [0] (n=Horscht@xbmc/user/horscht)
07:12:34NJoinadvcomp2019 [0] (n=advcomp2@unaffiliated/advcomp2019)
07:12:34NJoingoffa [0] (n=goffa@
07:12:47 Quit killan (Read error: 104 (Connection reset by peer))
07:13:24 Quit goffa_ (Remote closed the connection)
07:13:27 Join goffa_ [0] (n=goffa@
07:15:10 Quit shadoxx (Read error: 113 (No route to host))
07:21:44 Quit pat_mulchrone (Read error: 110 (Connection timed out))
07:26:29 Join n1s [0] (
07:29:08 Part toffe82
07:35:08 Join kachna [0] (
07:37:21n1shmm, when i try to use viewvc iceweasel says it can't connect to but a regular svn up works...
07:38:25 Join __lifeless [0] (n=lifeless@
07:38:57***Saving seen data "./dancer.seen"
07:40:55 Quit _lifeless (Read error: 60 (Operation timed out))
07:55:53 Quit Res1 (":\")
07:57:13 Join Horscht86 [0] (
07:59:04 Quit kachna (Read error: 113 (No route to host))
08:01:34 Join pat_mulchrone [0] (
08:13:20 Quit Horschti (Read error: 110 (Connection timed out))
08:16:59 Join bertrik [0] (
08:18:02 Quit Xerion (Read error: 54 (Connection reset by peer))
08:18:34 Join Xerion [0] (
08:19:46 Join ender` [0] (
08:20:47 Quit safetydan ("Leaving.")
08:23:24 Join Bagder [241] (n=daniel@rockbox/developer/bagder)
08:28:54 Join Rob2223 [0] (
08:35:46 Join matsl [0] (
08:36:33bertrikthe as3525 has two timers which already seem to be used (TIMER1 as a general timer and TIMER2 to call the tick tasks)
08:37:02bertrikit would be convenient to have a timer to generate delays on the order of microseconds that are independent of the CPU clock
08:37:48bertrikthe as3525 also has a programmable watchdog timer, how about using that for a delay generation timer?
08:39:12 Join Res1 [0] (
08:41:10 Join Bagder_ [0] (
08:41:36 Quit Bagder (Nick collision from services.)
08:41:37 Nick Bagder_ is now known as Bagder (
08:42:02 Join B4gder [241] (n=daniel@rockbox/developer/bagder)
08:45:36 Quit Rob2222 (Read error: 110 (Connection timed out))
08:45:48 Quit B4gder (Client Quit)
09:01:08 Join ender1 [0] (
09:01:48 Quit ender` (Read error: 60 (Operation timed out))
09:12:07 Join flydutch [0] (
09:12:17lucentbertrik: I figured out that adding a mci_delay() makes my 8gb uSD card work on boot for fuze 8gb v1
09:13:59bertriklucent, cool, I'm not very familiar with that part of the code (better discuss with funman when he's back on irc)
09:15:26lucentbertrik: okay!
09:21:32 Join petur [50] (n=petur@rockbox/developer/petur)
09:23:10 Join Thundercloud [0] (
09:38:23 Join stoffel [0] (
09:38:59***Saving seen data "./dancer.seen"
09:52:53 Quit Thundercloud (Remote closed the connection)
10:00:52 Join kachna [0] (
10:09:11 Join PaulJam [0] (
10:15:51 Join gregzx [0] (
10:18:48 Nick Horscht86 is now known as Horscht (
10:24:35 Join Sedgewick [0] (n=Sedgewic@
10:32:11 Join lyngaas [0] (
10:33:53 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
10:46:41 Quit kachna (Read error: 110 (Connection timed out))
10:49:22 Join trmss [0] (n=loker@
10:54:15trmssi am using rockbox with 2nd gen. ipod mini and recently I am troubled with a extreme dim screen. When I boot the original firmware, the screen seems to work perfectly but after I boot rockbox, the screen fades away. Any ideas?
10:56:14linuxstbThis has changed recently?
10:56:57trmssyep, and strangely without any updates or other possible trigger... and also I've tried the recent release just in case but no chance
10:57:14bertriktrmss, if you are able to compile your own build from svn, you could try various revisions and bisect it to see exactly which revision caused it
10:58:09bertrikoh in that case, bisecting won't help unfortunately ...
10:59:14linuxstbtrmss: I can't recall anyone else reporting that... Are you saying the backlight works initially when you boot Rockbox, then fades to off?
10:59:15trmssit seems so :( I really couldn't guess where to look with this kind of problem and thought that maybe someone here experienced some similar case...
11:00:01trmssyes but from the first second even, there are some dim lines vertically... and it gets dimer and dimer when I go into some folders...
11:00:46trmssbut in some random moments it works properly... that's why I was suspicious about the hardware and tried out with original firmware...
11:02:15 Join PaulJam_ [0] (
11:04:08linuxstbSo there's no sign at all of the problem in the original firmware?
11:05:59 Join p3tur [50] (n=petur@rockbox/developer/petur)
11:07:20 Quit petur (Read error: 110 (Connection timed out))
11:08:17trmssyes, original firmware works perfectly...
11:08:29 Quit PaulJam (Read error: 104 (Connection reset by peer))
11:08:42trmssI was at least expecting the vertical dim lines, but... no problems at all
11:09:23 Join Lynx_ [0] (
11:15:02LloreanYou said it can take a little bit to show up. how long did you spend in the original firmware, and did you do many things that trigger accesses to the disk?
11:19:10pixelmadoes the original firmware have backlight fade (settings)? I think it doesn't but believe Rockbox does
11:19:25pixelma(on the Ipod Mini I mean, of course)
11:22:28trmssLlorean: The dim just appears, not after a while... and specially if I enter into any folder... and I tried to the exact same actions in ipodos as well
11:23:03trmsspixelma: I can check if the original firmware has a setting like that... I didn't change any setting in that side actually but...
11:23:34pixelmaI believe backlight fade is enabled by default
11:24:11Lloreantrmss: How can you perform the "Exact same actions"?
11:24:22LloreanThe firmwares work entirely differently, load different things into RAM, access the disk at different times.
11:24:32LloreanThat's why I asked how long you spent in the OF, and about the general types of actions.
11:25:48trmsswell go into third level of folder and choose a file to play... in rockbox... not the "same" in that manner in OF but browse many folders and play songs in ipod as well...
11:26:14 Join Chris_Black [0] (n=Sedgewic@
11:26:15LloreanAnd how long did you spend on the OF?
11:26:19LloreanI've asked this three times now...
11:26:31 Quit Chris_Black (Read error: 104 (Connection reset by peer))
11:26:35trmssspending same time yes, and it was working without changing the BL settings before... I'm using RB for about two years... with several updates...
11:26:46 Quit Sedgewick (Read error: 104 (Connection reset by peer))
11:26:50trmssjust phone 1 sec
11:26:52LloreanPlease, I didn't ask if you spent the same amount of time.
11:26:56LloreanI asked how much time in the OF.
11:27:20LloreanI would recommend using the OF for several days regular use, just see if the problem takes a little longer to show up there, before ruling it out as a hardware issue.
11:28:02 Join Lynx [0] (
11:28:29 Nick Lynx is now known as Guest69464 (
11:32:39 Quit stoffel (Read error: 113 (No route to host))
11:32:49trmssLlorean, OK, sorry for that... I got it now...
11:33:34trmssYep I didn't assume that a harware problem needs time as well to appear... I'll try that and see what happens...
11:33:53trmssthanks... take care...
11:33:54 Part trmss
11:38:22 Quit gregzx (Read error: 110 (Connection timed out))
11:39:00***Saving seen data "./dancer.seen"
11:42:24 Join petur [50] (n=petur@rockbox/developer/petur)
11:44:59 Quit Lynx_ (Read error: 110 (Connection timed out))
11:44:59 Nick Guest69464 is now known as Lynx_ (
11:56:37 Quit p3tur (Read error: 110 (Connection timed out))
12:12:58 Quit petur ("later")
12:33:33 Quit Rob2223 (Read error: 104 (Connection reset by peer))
12:43:17 Join Rob2222 [0] (
12:45:23 Join Rondom [0] (
12:50:02 Join gregzx [0] (
12:51:09CIA-38New commit by funman (r21152): Sansa AMS : higher a bit the delay between acmd41 loops, needed for some microsd cards. FS #10268 by Eric Shattow
12:54:32 Quit Rondom ("Ex-Chat")
13:04:03 Quit efyx_ (Remote closed the connection)
13:06:34 Join efyx_ [0] (
13:15:24 Join robin0800 [0] (
13:34:50 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
13:39:04***Saving seen data "./dancer.seen"
13:43:38 Join PaulJam [0] (
13:59:49gevaertsLlorean: the Rockbox Utility forum seems to be sorted in a suboptimal way
14:01:07gevaertsI think it's sorted by first post
14:02:14 Quit PaulJam_ (Read error: 113 (No route to host))
14:15:39LloreanFor some reason the new boards I created seem to have ended up like that, rather than being created set to use the forum default
14:18:48 Quit amiconn (Read error: 110 (Connection timed out))
14:18:53 Quit pixelma (Read error: 110 (Connection timed out))
14:28:47MarcGuayIs it possible to ignore a file when SVN commit-ing or to check in a particular .diff file?
14:29:41Tornesvn commit file1 file2 file3
14:29:47Tornealas not the other way around :)
14:29:59MarcGuayTorne: Crummy. Thanks.
14:30:10n1sMarcGuay: you can select specific files (like Torne said) or specific subdirs of the source
14:30:13gevaertsif you only want to commit part of a file, you have to be creative
14:30:15Tornegrep/cut/etc on the output of status to generate the list? :)
14:30:39Torneor, yaknow. check the branch out with bzr instead and use shelve or interactive :)
14:37:28CIA-38New commit by MarcGuay (r21153): c100: Get a regular build compiling. Very little functionality.
14:38:47MarcGuayCaveman style copy & paste from the diff worked fine.
14:40:12 Join LambdaCalculus37 [0] (i=44a0430d@rockbox/staff/LambdaCalculus37)
14:41:33CIA-38New commit by amiconn (r21154): typo−−
14:41:37 Join Sedgewick [0] (n=Sedgewic@
14:42:39LambdaCalculus37MarcGuay: Bonjour to you; congrats on getting the c100 to build. ;)
14:44:47 Join amiconn [0] (
14:54:59*LambdaCalculus37 managed to make a little more progress on getting the GoGear HDD6330 to build yesterday, but it still errors out at spots
14:56:22 Join PaulJam_ [0] (
14:57:19 Quit Lynx_ (" HydraIRC -> <- Go on, try it!")
14:58:22 Join Chris_Black [0] (n=Sedgewic@
14:58:59MarcGuayLambdaCalculus37: Thanks.
15:02:18 Quit PaulJam (Read error: 113 (No route to host))
15:02:31 Quit SirFunk_ (Read error: 110 (Connection timed out))
15:03:37LambdaCalculus37Although for some odd reason, I can't svn add the new adc-target.h file and two other *.h files. They're just a copy/paste from the HDD1630 files.
15:04:06LambdaCalculus37Should I perhaps svn remove those files (I have them in firmware/target/arm/philips/hdd6330 in my local trunk), then try to svn add them?
15:04:20linuxstbIf you've copied the hdd1630 files, you should use "svn cp", rather than adding them as brand new files.
15:05:08 Join SirFunk_ [0] (
15:05:48 Join pixelma2 [0] (n=Marianne@rockbox/staff/pixelma)
15:05:49 Join FlynDice [0] (
15:07:54LambdaCalculus37linuxstb: So just "svn cp <path/to/file> </path/to/destination>?
15:08:35linuxstbLambdaCalculus37: Yes, I think so.
15:10:51LambdaCalculus37linuxstb: Okay, I'll give it a shot.
15:11:58LambdaCalculus37After that, I have to figure out why it errors out when building the debug menu (and I need to get a pastebin of the error).
15:14:49 Quit Sedgewick (Read error: 110 (Connection timed out))
15:24:08 Quit robin0800 (Remote closed the connection)
15:24:28 Join robin0800 [0] (
15:32:51 Join SirFunk__ [0] (
15:39:06***Saving seen data "./dancer.seen"
15:40:07 Quit gregzx (Read error: 110 (Connection timed out))
15:40:33CIA-38New commit by unhelpful (r21155): Add Bob Jenkins' lookup3 32-bit and 64-bit hashes to pluginlib.
15:44:48 Quit onlysoaa (" ajax IRC Client")
15:47:33 Quit SirFunk_ (Read error: 110 (Connection timed out))
15:58:11 Join mt [0] (n=MTee@rockbox/developer/mt)
15:58:27 Quit advcomp2019 (Read error: 110 (Connection timed out))
16:05:28 Join PaulJam [0] (
16:06:21 Join advcomp2019 [0] (
16:06:31 Join evilnick [0] (i=0c140464@gateway/web/ajax/
16:07:06 Join PaulJam__ [0] (
16:10:14 Quit matsl (Read error: 110 (Connection timed out))
16:22:12 Quit PaulJam_ (Read error: 113 (No route to host))
16:24:59 Quit PaulJam (Read error: 110 (Connection timed out))
16:31:30 Quit bmbl ("Woah!")
16:33:40 Join DONO [0] (
16:39:05 Join kugel [0] (n=kugel@rockbox/developer/kugel)
16:39:32DONOG'day im havin a bit of trouble loadin Rockbox to my 4th Gen (colour) . Anyone around able to help?
16:40:13kugelDONO: Not if you don't describe your problem
16:40:25*GodEater 's psychic powers are on the blink today
16:41:01DONOk , in Rockbox it states my colour 4thgen is on G:\ after auto detect . i click Complete Installation but the window brings up No Ipod Detected
16:41:31DONOi read thru FAQ n stuff of course . but couldnt really.. see anything of this.
16:41:58Torne1) is it actually on G: ?
16:42:12DONOyup , under My Computer its G:\
16:42:24 Quit robin0800 (Read error: 110 (Connection timed out))
16:42:58GodEaterhave you ever had Rockbox installed on it before ?
16:43:26gevaertsdo you have administrator rights?
16:43:29DONOsure do
16:44:00GodEaterCan you try going to the "Installation" tab, and choosing the second option "Install Rockbox" first.
16:44:03kugelso, is anyone opposed to adding the ability for plugins to go to the wps after exiting?
16:44:20GodEaterkugel: assuming you have music playing, sounds like a fine idea
16:44:37DONOGodEater - Doin that now
16:44:57DONOk installed.
16:45:05kugelGodEater: it would also work without music playing. it would just start playback in this case
16:45:12GodEaterok, now try the first option, "Install the bootloader"
16:45:25DONONo Ipod Detected.
16:45:26GodEaterkugel: I'm not sure that's such a great plan
16:45:42GodEaterDONO: which OS do you use ? Windows ? Linux ? OSX ?
16:45:47gevaertskugel: what would it start exactly?
16:45:51DONOWindows XP SP2
16:46:02GodEaterDONO: have you ever used the command prompt before ?
16:46:05kugelwhere-ever you left off the last time
16:46:07DONOof course heh.
16:46:25kugelit's the same thing as hitting the wps button in the filetree or mainmenu
16:46:27GodEaterDONO: you'd be amazed how many people haven't
16:46:44DONOi was a part time network admin. so , thats that. heh
16:46:46GodEaterDONO: well, in the manual (which I assume you've downloaded) there should be a link to ipodpatcher
16:47:37kugelin fact, it's absolutely the same code that starts the wps (it works by adding a new return value for plugins, which then sets the same variable to true as if you selected music or hit the wps button)
16:47:40linuxstbipodpatcher doesn't even need the command prompt - you can just double-click ipodpatcher.exe and follow the prompts.
16:48:05DONOahk beaudy , ill get back when thats done
16:48:20GodEaterlinuxstb: true, but some people get scared by that awful text only window :)
16:48:20kugelthe modifications are rather minimal, and it's just meant as a ability, no current plugin would actually do it until it's adapted
16:48:58gevaertskugel: I think I see what you want to achieve, and I think it's a valid usecase, but I'd hope that only one or two plugins would actually do that...
16:49:53kugelthe aim should be to either add a menu entry for the wps or react to the main wps button, and not to replace the default and current exit behavior
16:50:01GodEaterI'd only want the WPS after a plugin exited if I'd had music playing already when I started the plugin
16:50:14 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
16:50:20*gevaerts thinks that kugel should explain what he's actually trying to do
16:50:47kugelmaking PF able to go to the wps directly on top of the normal exiting
16:50:53gevaertsGodEater: I guess this is about plugins actually starting playback
16:50:53GodEaterDONO: how you getting on ?
16:50:57DONOipodpatcher.exe even says no Ipods found
16:51:17*GodEater throws the support badge to linuxstb
16:51:19DONOjust one thing , Nano-Chromatics are 4th Gen correct?
16:51:25GodEaterer no
16:51:33 Join daurn| [0] (
16:51:36GodEaterthat's a 4th gen Nano
16:51:39n1sDONO: not supported, as the frontpage says
16:51:39GodEaternot a 4th gen ipod
16:51:43saratogaheh I was going to ask if you had a 4G
16:51:55saratogabut it always seems so insulting
16:52:36DONOah shit i read read wrong , i saw 1 through 5.5 .. whereas its 1 & 5.5
16:52:48DONOwithout the 2 reads lol
16:52:48Torneno, the generations are just seperate for each model
16:52:59kugelGodEater: Why would you not want that?
16:52:59Torne1st-5.5th gen original ipods work
16:53:10DONOah righti , got it
16:53:46gevaertskugel: would that be unconditional? I think going to the WPS if you start playback from PF is perfectly reasonable, but going to the WPS if you just looked at PF from the plugin menu to see what it was about sounds wrong
16:53:56kugelalso, it's meant as an additional way of exiting. The return to plugin browse should always be the default behavior when pressing the dedicated quit button
16:54:34DONOive been mislead by the whole wording differently between 'ipod' & 'nano' .. considering its iPod Nano. lol
16:54:39kugelgevaerts: Why? That's how it works in the core too
16:54:42DONOanyway.. ah hell i feel stupid lol.
16:54:43GodEaterkugel: sorry, I see where you going now - that makes perfect sense
16:55:05gevaertskugel: I'm a bit confused now. That's how what works?
16:55:21GodEaterDONO: how would you suggest we word it ?
16:55:59kugelyou can go to the wps in the core too without music playing, like you browse the unrelated .rockbox folder or themes, then you hit the wps button
16:56:15gevaertsah ok. I see what you mean now. Yes, that's perfectly reasonable
16:56:16kugelthis will start playback where you left off and get you to the wps
16:56:41Tornecan rbutil not tell people they're trying to install on an unsupported model?
16:57:05DONOi guess ' Nano-Chromatic ' would have made it more blaint . but .. now i understand theres a difference between iPod and Nano heh.
16:57:10 Quit FOAD (Read error: 104 (Connection reset by peer))
16:57:16GodEaterI've never heard them called that =/
16:57:26 Quit Zagor ("Don't panic")
16:57:40kugelI could also imagine this works nicely for plugins utilized in the core (credits, properties) so that they're even more integrated
16:57:43 Join FOAD [0] (n=dok@
16:57:44DONOGodEater - −−>
16:58:05*GodEater shoots Steve Jobs
16:58:13GodEaterI wish they'd settle on just one f***ing name
16:58:19DONOi wasnt sure if there was a 4th Gen and somehow Nano-Chromatic was different
16:58:24DONOso , thats why I was puzzled heh.
16:58:29GodEatersadly not
16:58:35GodEaterdaft name for it if you ask me though
16:58:43DONOyeah lol
16:58:46GodEatereven the 2nd gen Nano was available in multiple colours
16:59:05linuxstbThat doesn't seem the name of it, just an advertising slogan.
16:59:21GodEaterperhaps, but you could be forgiven for thinking it was the name
16:59:23gevaertskugel: I agree. Maybe it would also (in time) allow us to move more core stuff out to plugins if we ever want to
16:59:25GodEaterit's pretty big on the page
16:59:36linuxstbGodEater: Not really - everywhere else on the site simply refers to it as "ipod nano"
16:59:42Tornesrsly though, shouldn't the autodetection tell yuo that it's not supported? i thought rbutil knew all the USB ids
16:59:42DONOoh in Australia its spewed everywhere as CHROMATIC ..
16:59:55DONOso somehow i figured it was different to 4th Gen ._.
17:00:24gevaertsTorne: it should, but I'm not sure if all ids are in there
17:00:35Tornehm. theyr'e listed on the rbutil development wiki page
17:00:46gevaertsDONO: which version of rbutil did you use?
17:00:54Torneand if it wasn't listed surely it would've failed autodetection altogether?
17:01:16GodEaterDONO: we go off this page with our naming anyway :
17:01:59kugelthis link should probably be pasted on a dozens of places on and rbutil
17:02:14Torneoh, hm. does it just invoke ipodpatcher?
17:02:19DONOyeah thats the one i viewed and i saw 4th Gen and figured oh rockbox is supported , didnt realise theres a difference between 4th Gen Nano and 4th Gen iPod
17:02:41GodEateryes, they're inconsistent about what they call the ipod generations
17:02:48GodEaterthey call the 4th gen the "Click wheel" I think
17:02:48TorneDONO: there's not a particularly useful name to call the original ipod
17:02:54kugelyou saw the 4th gen but didn't see that it doesn't actually match your player?
17:02:57Torneyou can't call it the ipod classic because, er, apple have made a new thing called that ;)
17:03:07Torneso it's tricky to be specific about that
17:03:22GodEaterkugel: point to the "4th gen" on that page
17:03:25GodEaterthey don't list one
17:03:34GodEaterit's either the photo, or the click wheel
17:03:50DONOi just didnt see the difference in Nano & iPod considering its an iPod .. but now the message at the top of Rockbox's homepage makes sense hah
17:03:54n1sI agree that it would be nice if the installer told the user about an unsupported ipod since they are the most common to confuse
17:04:03kugeloh lol, sorry
17:04:16gevaertsapparently it warns about some of them, but not all
17:04:20linuxstbn1s: It should do, but it looks like rbutil doesn't have the USB IDs of all the newer Nanos, just one.
17:04:25 Quit Chris_Black (
17:04:25 Quit Rob2222 (
17:04:25 Quit dmb (
17:04:25 Quit perrikwp (
17:04:25 Quit J-23 (
17:04:26GodEaterDONO: that's the genius of apple's marketing I guess
17:04:39GodEaterDONO: on the outside they all look similar, on the inside, they're really not.
17:05:04Tornelinuxstb: i'm just wondering how come it knows it's an ipod to detect it, but not that it's an unsupported model
17:05:06NJoinRob2222 [0] (
17:05:06NJoinperrikwp [0] (i=4aa794a0@gateway/web/ajax/
17:05:06NJoinJ-23 [0] (
17:05:13DONOaha yeah , i understand that part ;) like their genius who designed iTunes
17:05:22NJoindmb [0] (n=dmb@unaffiliated/dmb)
17:05:25NJoinChris_Black [0] (n=Sedgewic@
17:05:27Tornedoes ipodpatcher still detect unsupported models, to be able to dump them etc?
17:05:32GodEaterDONO: you think he's a genius? I think I'd like to toilet paper his house....
17:05:41 Quit Chris_Black (Client Quit)
17:05:44DONOGodEater - was being sarcastic ;P haha
17:05:55GodEaterDONO: thank heavens :)
17:06:42DONOme and another Acer techie have a pact to roll him up in carpet and throw him into a pit of scarabs.
17:06:58GodEaterbit extreme
17:07:02n1slinuxstb: i listed the usb id for the 4g nano on that wiki page with usb ids so it should be easy to make a patch if someone can find it :)
17:07:03GodEaterbut I understand the sentiment
17:07:04*LambdaCalculus37 thinks we're veering off-topic here
17:07:08DONOnot really in my opinion.
17:07:29 Quit daurn (Read error: 110 (Connection timed out))
17:07:32GodEaterI thought apple bashing was definitely on-topic for #rockbox :)
17:07:32DONOin any case , is there any sort of , well putting it simple , hacks for the 4th NANO. ? heh
17:07:38GodEaterDONO: nope
17:07:44saratogawhy do we check for specific USB IDs as opposed to just checking for any unknown Apple id?
17:08:04saratogagetting an apple usbid thats unknown should trigger an unsupported device message IMO
17:08:35 Join Sedgewick [0] (n=Sedgewic@
17:08:40n1ssaratoga: very sensible suggestion
17:09:01linuxstbTorne: I don't think ipodpatcher supports the new ipods in any way - rbutil detects them via usb IDs.
17:09:03Torneexcept that won't work so well on a mac
17:09:07Torne"apple mouse"
17:09:09Torne"apple keyboard"
17:09:11DONOLook what ive done. lol
17:09:23TorneDONO: hey it's good to know these things
17:09:36Torneit's theoretically possible for rbutil to have told you that your ipod is not supported, and it didn't
17:09:42Torneso, maybe someone should fix that
17:09:51Tornethen the next guy won't have to come ask us
17:10:01GodEaterMr Someone already has a large todo list
17:10:12TorneThere is that
17:10:12DONOheh i see , saratoga's idea seems ideal
17:10:24n1sall the different ipod PIDs seem to start with 12 so maybe other apple usb stuff use other starting digits of the pid?
17:10:29Tornei dunno. you would at least want to make sure it was mass storage.
17:10:46GodEaterI agree
17:10:55GodEatergoing with all Apple Id's isn't quite the right idea
17:11:04n1sor we could rely on the users knowing what ipod they have ;)
17:11:07Tornejust doing the list on the wiki would be a good start
17:11:12GodEatern1s: epic fail
17:11:23n1sGodEater: evidently so :D
17:11:45n1sif anybody patches and builds rbutil i can test with my 4g nano
17:13:11saratogadoes rbutil fail to detect a supported player if an unsupported one is also attached?
17:13:24n1sno idea
17:13:31GodEaterdon't think anyone's tested that
17:13:59saratogathe logic looks as if it might
17:15:24pixelma2hmm... pf seems to be crashing now on hwcodec - and it also won't be able to start playback the way it is now
17:16:17kugelpixelma2: when does it crash?
17:17:52pixelma2a very short time after entering. It shows the first "screen" with the covers and then freezes. I'm not able to move left or right or go to the menu. If I press "off" now, Rockbox crashes completely and reboots
17:18:38kugelI pretty much only touched code which is used after entering the track list
17:18:40*n1s is very confused, the usbids in DeviceDetection for the ipods do not match those in rbutli.ini very well, the one the wiki states is for the classic rbutil uses for video, same with first/second gen nano...
17:18:55 Join SirFunk_ [0] (
17:19:18pixelma2kugel: I would have to try when it broke. But the playback thing is a fact
17:19:31 Join gregzx [0] (
17:20:08 Join funman [0] (n=fun@rockbox/developer/funman)
17:20:11 Join pixelma [50] (n=pixelma@rockbox/staff/pixelma)
17:20:11 Quit amiconn (Nick collision from services.)
17:20:13 Join amiconn_ [50] (n=jens@rockbox/developer/amiconn)
17:20:32pixelma2because the plugin grabs the audio buffer
17:20:33 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
17:21:12*n1s obviously needs more coffee, ignore
17:21:19*pixelma2 refers to amiconn for the details
17:21:35kugelI see that it uses plugin_get_audio_buffer for "low mem"
17:21:47 Quit pixelma2 (" .")
17:22:21 Join PaulJam [0] (
17:22:24kugelThat means playback has to be stopped before, but I thought you could start playback after again, but I'm not into the details of the get_buffer stuff either
17:22:26Tornen1s: hm, that's unfortunate.. presumably the ones in rbutil are right :)
17:22:36Tornen1s: since it seems to work for everyone..
17:22:49kugelthat doesn't seem like a HWCODEC restriction though
17:22:49n1sTorne: no, i was wrong, misread the ini file
17:22:53Torneoh, right.
17:23:14n1ssome devices in that has a "usbid" and some has "usbincompat"
17:23:18n1swhich confused me
17:23:38n1sfor example the ipod nano has no usbid but has usbincompat with the 2g nano...
17:24:48 Join jgarvey [0] (
17:24:56n1swhich means i have no idea how to add more incompatible models, or why a 2g nano is specifically ncompatible with a 1g nano etc...
17:25:00pixelmakugel: I believe there is something else but can't tell you the details
17:25:42 Join mirak [0] (
17:25:53Tornen1s: i would guess the usbincompat entries are on the closest model?
17:26:16 Quit PaulJam__ (Read error: 60 (Operation timed out))
17:26:45n1swell it seems weird to me, the 2g nano is incompatible with rockbox, it doesn't really matter that it is incompatible with the 1g nano
17:27:12Torneyah, but i mean the ini file is structured around models that *are* supported
17:27:15n1safaiu but maybe rbutil has some thing that relies on this so i'll not touch it
17:27:17Tornethere aren't seperate entries for things that don't work
17:27:30pixelmakugel: unfortunately testing on Ondio is always a bit tedious because unpacking a build takes ages :\ and if the plugin API changed I at least need the new firmware file and pictureflow.rock
17:27:40Torneso the entries are "this is the install config for ipod, except this usb id which is unsupported"
17:27:50n1sTorne: so it should get such a list imo
17:27:55Torneexcept it's complicated by there being lots of ipods
17:27:58Tornewell, perhaps
17:28:58funmanbertrik: I have seen Clipv2 1GB at 35€ here in Dijon
17:29:32n1swe could even create entries with names for them so rbutil could print "this is a fourth generation ipod nano which is not supported ny rockbox" instead of "unsupported ipod" or w.e.
17:29:56Tornewell, that sounds like a plan, but i have no idea how rbutil works
17:30:15*n1s neither
17:30:31n1sand the two guys working on it are not around
17:30:43Lloreankugel: I was wrong about Next/Prev working in-list on the S it seems. I could've sworn I'd used the dedicated buttons there, but it seems they don't work there (yet, they aren't mapped to anything and so should be mapped to it in the same way as the dedicated volume buttons are, and since it sounds like you're using the list context for the plugin anyway, I guess it shouldn't really affect you)
17:30:44saratogan1s: you want to add more usbids to rbutil?
17:30:56funmanIf only we had some way to contact them, like a developers mailing list ..
17:31:15Tornesaratoga: we're thinking it should at least include all the known ids for ipods, specifically so it can tell people that unsupported models aren't supported
17:31:29Tornemaybe saving some of the frequent questions :)
17:31:29n1ssaratoga: sounds liek the easiest way to tell users that their ipods are not supported reliably
17:31:49LloreanI thought RBUtil already tells you when you're using a newer-generation iPod
17:32:09linuxstbLlorean: It does for some, but apparently not all.
17:32:09n1sLlorean: only classic and 2g nano from what i can tell
17:32:56Tornen1s: you could try just duplicating the line and putting more ids in tehre
17:32:57kugelLlorean: yea, CONTEXT_TREE is chained in now. It's just a matter of implementing handlers for wps/stop/volume. Of course we have to take care whether pf doesn't already use that buttons
17:33:02 Quit MarcGuay (Read error: 110 (Connection timed out))
17:33:07Torneif you have one fo the ipods in question?
17:33:14Tornei don't, so not entirely trivial to test it
17:33:14 Quit SirFunk__ (Read error: 110 (Connection timed out))
17:33:17n1sTorne: yust found an example that should work, the sansas list multiple ids
17:33:24Lloreankugel: Since PF is basically just a visualized list, why would it need buttons that aren't normally needed for a list?
17:33:28Tornestuff the ones from the wiki in there, and try it ;)
17:33:35*n1s will do
17:33:53saratogan1s: can you actually put a message in somewhere?
17:34:13saratogayou don't need to duplicate the line it takes a list of IDs
17:34:19n1ssaratoga: not as it is now, you'll only get "unsupported ipod"
17:34:32n1ssaratoga: yes i know i saw how the sansas do it
17:34:35 Join shadoxx [0] (
17:34:52 Quit amiconn (Read error: 60 (Operation timed out))
17:35:06saratogashall i commit or you?
17:35:27 Quit pixelma (Nick collision from services.)
17:35:27 Join pixelma_ [50] (n=pixelma@rockbox/staff/pixelma)
17:35:28kugelLlorean: currently it has a dedicated quit button which may be be a button which is already used in lists (e.g. in the e200 case, power is quit but would stop in lists)
17:35:31n1si can do it, but i'll test with my 4g nano first
17:35:45 Nick pixelma_ is now known as pixelma (n=pixelma@rockbox/staff/pixelma)
17:35:54Lloreankugel: Does it need a dedicated quit button?
17:36:02LloreanIf it's ever going to be used in the database, it really shouldn't have one.
17:36:05saratogaif you are lazy just replace the ids for the nano with: usbincompat=0x05ac1260, 0x05ac1260, 0x05ac1263
17:36:06LloreanYou can't really "quit" the database.
17:36:10kugelalso, the list flips in PF, and left/right are used for scrolling in the album list
17:36:22 Join amiconn [0] (
17:36:25 Join pixelma2 [0] (n=Marianne@rockbox/staff/pixelma)
17:36:26saratogausbincompat=0x05ac1260, 0x05ac1262, 0x05ac1263
17:36:27kugelI don't think it does, you can quit with ACTION_STD_CANCEL also
17:36:35n1ssaratoga: i did that but with the correct one for the 3g ;)
17:36:42*n1s slow
17:37:03funmanbertrik: i can't get Clip LCD display something with FS #10048 patch :( , perhaps I should mention which revision I'm using when posting patches.
17:37:39saratogadetectUsb should probably also be fixed to check all available USB devices before returning an unsupported device error as well
17:38:03bertrikfunman, yeah, the screen went black here too just after the bootloader with that patch, no idea what causes it
17:38:07kugelI can 50fps in pf with FlynDice's latest patch
17:38:34bertrikI tried some delays in the lcd_write_xxx functions, but no luck (talking about a clip BTW)
17:38:49pixelma2kugel: I just noticed... it crashes when it wants to enter the tracklist. Thing is - it also hints on a keymap issue because now going "right" on the Ondio doesn't move right in th art screen but tries to enter the tracklist (and I didn't expect that as it once worked correctly)
17:39:08***Saving seen data "./dancer.seen"
17:39:30pixelma2so I can't test what it really does without the tracklist...
17:39:56bertrikregarding the recent fmradio problems on ams sansas that can't be solved by larger delays, I wonder if there's some kind of initialisation problem
17:41:12funmanbertrik: I tried that, also changing the DBOP frequency
17:41:14 Quit lyngaas (Remote closed the connection)
17:41:15bertrikI looked again in the OF and there something very weird there: there is an i2c delay function that is called with the amount delay in R0, but the delay parameter seems to be completely ignored by the function
17:41:25CIA-38New commit by nls (r21156): Add usb ids for third and fourth gen nanos to the list of unsupported targets
17:41:52kugelbut it stops loading covers quit soon (and I I regularly get theh F**K message which indicates retry in the sd transfers)
17:41:57funmanbertrik: perhaps the dynamic delay is replaced by a hardcoded value?
17:42:05kugelpixelma: what happens if you press up?
17:42:39funmankugel: the F**K message is only here to detect an infinite loop in resetting transfers (failed transfer often happen and this sounds normal to me)
17:43:19kugelwell, it stops loading covers at this point
17:43:26pixelma2kugel: the same. I'll bisect (as the keymap changes were a bit later)
17:43:29bertrikfunman, exactly
17:43:58kugeli.e. I can flip though the covers just fine, but they're all empty except the first 10 or so
17:44:51pixelma2kugel: what do you want to tell me with this? I'm stuck at cover #1...
17:45:15 Quit DONO ()
17:45:22funmankugel: sd transfers are not working (test_disk fails), so any plugin can function incorrectly without other reason than the code being corrupted.
17:45:28kugelnothing, just made sure they keymap conflict doesn't make it crash
17:45:37bertrikfunman, also the fmradio i2c delay is just a simple nop delay loop and doesn't use anything fancier that is independent of CPU speed (perhaps CPU speed in the OF is always unboosted with radio active)
17:45:48kugelfunman: "it sounds normal" sounded like everything is fine
17:46:11pixelma2kugel: maybe, but entering tracklist seems to do...
17:46:25 Quit advcomp2019 (Read error: 110 (Connection timed out))
17:46:34pixelma2which is the only thing I can do at the moment :\
17:47:11funmankugel: did you run test_disk to verify? I get sd transfers work correctly only when the data cache is disabled, and even then I have to use a DMA buffer aligned on cache line size (32 bytes)
17:47:29kugelpixelma: this should make it able to browse through covers again (I hope)
17:48:17 Join moos [0] (i=mustapha@rockbox/staff/moos)
17:48:23kugelfunman: No, I didn't.
17:48:36 Join PaulJam_ [0] (
17:49:35*n1s turned out to commit the rbutil thing without testing because i cant find any info on how it reads that ini file or where it stores any config info... :/
17:50:45funmanI'll look again at the USB code with the help of the AMD Geode CS5536 datasheet. With the hope to be able to communicate with our devices when the screen refuses working.
17:52:35pixelma2it also crashes with r21142 after a few seconds in the art view mode... :/
17:54:33kugelfunman: "creat() failed"
17:54:38 Quit HellDragon (Read error: 60 (Operation timed out))
17:57:42pixelma2seems I need to do it more carefully. With "catch mem access" I at least get an error address. Starting to test with r21108 now but have to do other (RL (TM)) too
17:58:07 Part pixelma2
18:02:18 Quit Sedgewick ("off")
18:05:26 Join JdGordon| [0] (i=ad748d7b@rockbox/developer/JdGordon)
18:05:54kugelhaha, test_disk apparently created several 2GB files, on my fuze which has only 2GB capacity :)
18:05:55bertrikhm, something looks fishy in i2c_start generic_i2c.c: SDA gets switched high then low without any explicit delay in between
18:07:02 Join bimbel [0] (n=Miranda@unaffiliated/bmbl)
18:09:50 Quit PaulJam (No route to host)
18:21:01 Quit bmbl (No route to host)
18:22:52 Nick bimbel is now known as bmbl (n=Miranda@unaffiliated/bmbl)
18:31:42 Quit VytenisS ("Quit")
18:33:34 Join wark [0] (
18:33:52 Join PaulJam [0] (
18:34:36 Join evans [0] (n=evans@
18:35:33 Quit JdGordon| (" ajax IRC Client")
18:41:21 Quit evans (Read error: 60 (Operation timed out))
18:41:37 Join evans [0] (n=evans@
18:42:00 Join kugel_ [0] (
18:42:06 Quit kugel (Nick collision from services.)
18:42:22 Nick kugel_ is now known as kugel (
18:43:07 Quit gregzx (Read error: 110 (Connection timed out))
18:44:07 Join toffe82 [0] (n=chatzill@
18:50:43 Quit PaulJam_ (Read error: 113 (No route to host))
18:51:10 Quit evans (Remote closed the connection)
18:54:09 Join VytenisS [0] (
18:56:20 Quit funman ("leaving")
18:56:41LambdaCalculus37linuxstb: Thanks for the earlier svn advice; I managed to get the three files I couldn't svn add before to now add themselves correctly to the trunk.
18:57:05LambdaCalculus37I now need to get the last couple of kinks out and I should be able to commit the new target.
19:04:51 Join killan [0] (
19:05:43 Quit killan_ (Read error: 104 (Connection reset by peer))
19:06:32LambdaCalculus37 <−− Now I know I have quite a bit of work to do. :)
19:09:03 Join advcomp2019 [0] (n=advcomp2@unaffiliated/advcomp2019)
19:10:39 Quit VytenisS ("Quit")
19:11:14linuxstbLambdaCalculus37: Looks like you have a port to write... ;)
19:12:14LambdaCalculus37linuxstb: So definitely not in a commitable state?
19:13:50LambdaCalculus37Or can I at least commit what I have now so that others can also help do some work on it?
19:15:25linuxstbI'm not the person to ask about committing not-working ports...
19:19:34bertrikjust my opinion: I think for ports that currently don't work yet it's no problem to commit new stuff even if it doesn't fully work yet (as long as it's an improvement over the current situation)
19:20:13bertrikand it shouldn't interfere with currently working ports of course
19:21:02LambdaCalculus37bertrik: Just to make sure none of my changes are going to bork anything, I'll try to build for other targets as well to make sure everything's still working.
19:21:22CIA-38New commit by kugel (r21157): Samsung YH*: Enable RTC on all, enable lcd contrast on 925.
19:21:57LambdaCalculus37bertrik: Besides the fact that the HDD6330 still needs more work in order to be ready to go.
19:22:20LambdaCalculus37There's no proper touchpad driver for it, and while audio playback works, it's still rather spotty.
19:23:53 Join kugel_ [0] (
19:24:37 Quit kugel (Nick collision from services.)
19:24:41 Nick kugel_ is now known as kugel (
19:25:03 Join JdGordon| [0] (i=836b0070@rockbox/developer/JdGordon)
19:26:00 Join merbanan [0] (
19:33:22CIA-38New commit by kugel (r21158): Move the ONDIO_PAD to the others which have the same keymap in the generic directions PLA context.
19:36:50 Join HellDragon [0] (
19:38:12 Join {phoenix} [0] (
19:39:12***Saving seen data "./dancer.seen"
19:39:18 Quit JdGordon| (" ajax IRC Client")
19:39:34 Join JdGordon| [0] (i=836b0055@gateway/web/ajax/
19:40:38 Join kachna [0] (
19:43:21 Join onlysoaa [0] (i=42248748@gateway/web/ajax/
19:44:18onlysoaaHello everyone. I just tried built tcctool in Ubuntu 9.04 to test out the D2 build on my P2, but when I run it, it throws an error:
19:44:24onlysoaa[ERR] usb_set_configuration failed (-1)
19:44:37gevaertsyou need to run it as root, e.g. with sudo
19:44:46onlysoaaAhh, thanks.
19:45:06onlysoaa*facepalm* how'd I not think of that?
19:45:59onlysoaaAh hah!
19:46:01onlysoaaIt loaded.
19:46:53onlysoaaPretty much the same as the last report, as I haven't made any changes to Rockbox yet. It's the same blue flickery screen as previously reported, but at least it runs.
19:47:42onlysoaaI have a feeling the D2 and the P2 use different power management chips; am I right?
19:48:08 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
19:49:28LambdaCalculus37onlysoaa: There should be some information about the P2 on the wiki.
19:49:37LambdaCalculus37That's your best bet to find out what the innards are.
19:49:46 Join PaulJam_ [0] (
19:50:19onlysoaaYeah, the wiki page doesn't have info on power management.
19:50:34 Join SeaWeed [0] (
19:52:04LambdaCalculus37onlysoaa: That's strange. Whoever wrote the wiki page must've forgotten to mention it, or probably didn't know.
19:52:10LambdaCalculus37So are you willing to find out? ;)
19:52:36onlysoaaI'd like to figure out the LCD first. :P
19:53:24kugelonlysoaa: there's some hardcoded numbers in the drivers related to the resolution
19:53:43 Quit PaulJam (Read error: 60 (Operation timed out))
19:53:48 Join PaulJam__ [0] (
19:53:52onlysoaakugel: I'm looking into that right now. The file in question is lcd-cowond2.c, right?
19:54:02kugeltry searching for the D2's resolution and it's hex-number variants and replace it with the P2 ones
19:54:50kugelalso search for the numbers-1 (i.e. 319 and 239 if the resolution is 320x240), those are likely resolution related too
19:55:11onlysoaaAhh, I see. Any suggestions for a good way to search across files?
19:55:30onlysoaaAlright. p:
19:55:39 Quit Llorean (Read error: 54 (Connection reset by peer))
19:55:47kugelbut lcd-cowond2.c should be the right file
19:56:07 Join Llorean [0] (
19:56:11kugeldon't forget to change config-cowond2.h accordingly too
19:56:21bertrikany e200v2 owners online that have problems with the radio?
19:58:24FlynDiceonly with the mmu enabled.... will that help?
19:59:02bertrikFlynDice, can you test my fmradio_i2c_fix patch?
19:59:18bertrikit's in FS #10267
19:59:58FlynDicesure, but my radio works with current svn, I'll try it with the mmu build though
20:00:16FlynDiceI'll be a few getting set up
20:01:08 Quit mt (Read error: 110 (Connection timed out))
20:01:24 Quit SirFunk_ (Read error: 60 (Operation timed out))
20:02:04bertrikFlynDice, there's no hurry to do it right now, but if/when you do it please let me know the result
20:02:15FlynDiceyou bet
20:02:37 Join PaulJam [0] (
20:05:23 Join SirFunk_ [0] (
20:09:42 Join mt [0] (n=MTee@rockbox/developer/mt)
20:10:38 Quit PaulJam_ (Read error: 113 (No route to host))
20:16:33 Quit PaulJam__ (No route to host)
20:24:24JdGordon|can you not do next-dir on the ipods?
20:24:43JdGordon|menu+next and select+next do anything
20:25:00JdGordon|or play+next
20:25:54LloreanDid you try short + long next?
20:26:29JdGordon|but thanks, that works
20:26:34LloreanYeah, I'm not fond of it
20:26:38LloreanIt trips me up when I try to seek sometimes
20:26:53JdGordon|its nice that it can be done one handed... but... yuk
20:27:14*JdGordon| tries to remember to checkout the ipod keymap and add nextdir tonight if possible
20:27:19 Quit Bagder ("*plopp*")
20:29:06*JdGordon| also thinks about adding an option to disable AA...
20:29:37LloreanAn option to disable AA?
20:29:59LloreanIsn't that just "use a WPS without AA"?
20:30:00Mikachujust don't put it in the wps?
20:30:41JdGordon|I'm using cabbie because im too lazy and dont care enough to find a different one
20:31:02JdGordon|and I dont think AA looks very good on the mini
20:32:28 Nick fxb__ is now known as fxb (
20:34:54 Join Bagder [241] (n=daniel@rockbox/developer/bagder)
20:35:15 Join robin0800 [0] (
20:36:27 Join Thundercloud [0] (
20:38:15 Join stoffel [0] (
20:40:03 Quit flydutch ("/* empty */")
20:41:28FlynDicebertrik: FS #10267 works fine for me patched to svn, but my radio was working in svn anyway. When I added in FS #10048 it does not work. But interestingly enough it improves playback with the mmu enabled. I'm playing mp3 at an average of 43 MHz and only boosting 7% of the time. Perhaps there is more gold in the mine? ;)
20:45:58FlynDiceI should say 192_32_32_dcache.patch from FS #10048 to be more precise
20:46:30 Join bagawk [0] (
20:49:58 Quit Galois (Remote closed the connection)
20:50:22 Quit FlynDice (Remote closed the connection)
20:50:23 Join Galois [0] (
20:56:43 Join gregzx [0] (
20:57:35 Quit bagawk_ (Read error: 110 (Connection timed out))
20:57:55 Join FlynDice [0] (
21:00:51 Join PaulJam_ [0] (
21:00:52 Quit daurn| (Read error: 110 (Connection timed out))
21:01:48 Join daurn| [0] (
21:07:57 Join wincent [0] (
21:09:46 Quit Horscht ("Verlassend")
21:10:54kugelFlynDice: but it still crashes early?
21:12:14FlynDiceYes it is does not fix it, but I was able to enter a proper boost/unboost mode simply by resuming play, I have not had that work before
21:12:22bertrikFlynDice, weird, the fmradio i2c stuff should have absolutely no effect on mp3 playback
21:12:53FlynDiceI don't claim to understand, I just observe at this point...
21:13:38bertrikthe weird thing is that the current delay in the fmradio i2c layer should already be long enough, so I can't explain either that it increasing the delay seems to make things work for some people
21:14:07 Join pyro_maniac [0] (
21:14:11 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
21:15:09FlynDiceYou see, I don't have the handicap of actually understanding everything completely. I just try lots of stuff and see what works ;)
21:15:28bertrikit is very much appreciated
21:15:28saratogadoes disabling boosting make any difference?
21:17:55bertrikFlynDice, the F*CK messages I got with the 192_32_32_dcache patch scared me a little ...
21:18:37kugelsaratoga: interesting question
21:18:54kugelThat reminds me that F/X don't have unboosting as well
21:19:04kugeland they have a quite similar processor
21:20:12kugelthough, they implemted a cpu_set_frequency() which does the same as our (switch to fastbus)
21:20:20 Quit PaulJam (No route to host)
21:20:33kugelapparently, at some point, unboosting was disabled, for some reason
21:20:54kugelmaybe kkurbjun or jhMikeS knows more
21:21:03saratogait didn't seem to make a difference
21:21:20saratogai think the chip has some sort of internal clock control that was as good as boosting, or something like that
21:21:34bertrikhm, the radio still works even when boosted with the 192_32_32_dcache patch on my clip
21:21:34kugelI'd think our chip has this as well
21:21:47saratogai don't think so
21:21:55saratogaits not part of the arm stuff
21:22:04kugelwouldn't that be a cpu feature?
21:22:15saratogaprobably built into the SOC
21:23:04saratogaand the F never had the most amazing battery life either
21:23:22toffe82I think there was no gain for batteriy life
21:23:50 Join funman [0] (n=fun@rockbox/developer/funman)
21:23:51kugelsaratoga: states pretty decent runtimes
21:24:01funmanr12228 "disable CPU scaling again because it actually made the power consumption wor
21:24:14gevaertssaratoga: 20+ hours isn't bad for a HD player
21:24:19bertrikwow, worse?
21:24:48kugel15h for with standard battery seems good for a hdd target with big screen
21:24:59 Quit robin0800 (Remote closed the connection)
21:25:16kugelI think we don't really have messured the impact of boosting on the AMSes
21:25:21saratogaits more like 15 hours with stock battery
21:25:27 Join robin0800 [0] (
21:25:37martian67actually im pretty sure that rockbox is beating apple standard firmware for me
21:25:47martian67or at least its very close
21:26:04gevaertsBoth of mine do 20+ hours. I got them on ebay, so I don't know if they have non-stock batteries
21:26:11saratogasignficantly worse then PP Ipods, and PP is crap
21:26:30martian67saratoga, i dont find that :)
21:26:52martian67if you mean that battery time is significantly worse
21:27:01martian67that was the case at one time :/
21:27:27saratogaregarding AMS, maybe the best way to debug the MMU would be to disable radio, boosting, etc and focus on just getting it working at one clock
21:27:32saratogathen enable more features once we have that
21:27:40kugelthat seems like a good idea
21:27:43saratogai feel like development is chasing 10 different bugs at once and getting no where
21:28:01saratogait seems very likely we have additional driver bugs which are just now being exposed
21:28:28saratogabut i have done no work so what do i know
21:30:15kugelsaratoga: I see only patched benches (with 24MHz etc) from Buschel. I don't think the ipod is much better
21:30:20funmanMMU itself works very fine.
21:30:34funmanThe problems come with caches, especially the data cache
21:30:50*LambdaCalculus37 is glad to hear FM radio working on the GoGear HDD6330
21:31:04bertrikLambdaCalculus37, is that a first?
21:31:16funmanWe don't use the MMU in SVN because then rockbox would be much slower, because of the disabled caches
21:32:59LambdaCalculus37bertrik: No, it just confirms that the hardware is the same as in the HDD1630.
21:34:27bertrikfunman, and the dcache problem manifests itself mainly when using dma?
21:35:14saratogakugel: buschel's test should be very close to stock SVN, 24MHz clock probably adds an hour or so
21:36:06kugelfunman: the dcache isn't enabled, so only the icache would make a difference
21:36:06funmanbertrik: yes, we should check what happens when using simple transfers from the SD fifo.
21:36:17kugelwhy's the icache not functioning with mmu?
21:36:45funmanthe icache functions "better" with cached memory.
21:37:00funmansince it's enabled i suppose it works fine
21:37:13kugelI mean, if we enable at least the mmu in svn, in theory, we shouldn't be slower?
21:37:15bertrikfunman, yes I would like to see how it behaves with just one DMA active (only PCM)
21:37:35funmankugel: perhaps virtual address translation takes more cycles to complete
21:37:40kugelI once enabled only the MMU and I had the same poor speed as when i disabled the icache
21:39:17***Saving seen data "./dancer.seen"
21:43:33 Quit Thundercloud (Read error: 113 (No route to host))
21:45:23CIA-38New commit by bertrik (r21159): Add missing delay in i2c_start of the generic_i2c driver, used by ams targets.
21:47:22 Quit JdGordon| (" ajax IRC Client")
21:47:48 Join JdGordon| [0] (i=836b0070@rockbox/developer/JdGordon)
21:50:25funmanbuild rockbox, zip, unzip, boot : works. modify test_disk, build, zip, unzip, boot : white screen.
21:50:30 Join mcuelenaere [0] (
21:51:19n1si think you miss a "unmount/eject" step in there... might cause weirdness from time to time ;)
21:51:42funmannope, safely unmounted
21:53:17n1sso random weirdness or a rockbox bug that only shows itself sometimes maybe?
21:53:26 Join robin0800_ [0] (
21:54:01funmann1s: a random bug that shows on AMS - with one of the patches attached to FS #10048 -
21:54:33funmanalright it booted, now to verify what test_disk has to say
21:54:34 Join Mathiasdm [0] (
21:55:33 Part wark ("Leaving")
21:59:12 Quit LambdaCalculus37 (" ajax IRC Client")
22:00:54kugelfunman: you have really no RTC in rockbox on the samsung?
22:00:55funmanit says something in the lines of "damn, you handle pretty well your data cache : i couldn't find any error transferring 300MB while the data cache is on"
22:01:07funmankugel: yeah really :) you don't trust me?
22:01:52kugelIt works just fine on my yh925, and it also doesn't show in the OF
22:02:13funmanperhaps some additional magic is required, like for making sound work
22:02:34kugelperhaps even the same magic? :)
22:02:43funmanyes perhaps
22:03:08funmani started disassembling but it's very different from the sansa ams
22:03:50funmanfor example the PP registers can often be loaded in a register with 1 mov, unlike on ams where the register offset is loaded from memory, and then you can find cross references to it
22:06:07funmanbertrik: the patch I attached to FS #10048 has to be edited
22:06:57funmanI'll try the correct version on the Clip. It is very stable on the Fuze except some white screens when booting
22:07:13funmantest_disk passes, and I'm listening to a Jimi Hendrix album right now.
22:07:20 Join barrywardell [0] (
22:09:02bertrikfunman, I'm interested in trying it
22:09:45funmanbertrik: uncomment declaration of uncached_buffer and memcpy() calls using it, and comment the other memcpy() calls. (all of this in ata_xx.c)
22:10:17funmandisable or enable data cache in system-as3525.c (I have it enabled here)
22:12:11 Join Xakh [0] (
22:12:26funmanI still have the "white" (=black because there is no backlight on the Clip OLED display) effect.
22:12:48XakhSo... The Sansa Clip, does it work with Rockbox?
22:12:52funmanI also modified the CCU_SCON register to be 0x1 (dma has the highest priority)
22:13:03funmanXakh: can you see it on the list of supported targets ?
22:13:04 Join froggyman [0] (
22:13:18XakhIunno where that is....
22:13:21 Quit robin0800 (Read error: 110 (Connection timed out))
22:13:47 Quit PaulJam_ (Read error: 60 (Operation timed out))
22:13:50Xakhwell, I didn't see it, but you guys were talking about it.
22:14:15funmanwe work on it, but it doesn't work yet
22:15:03bertrikand we don't know how long it will take
22:15:48 Join robin0800__ [0] (
22:20:36 Quit HellDragon (Read error: 60 (Operation timed out))
22:21:30 Quit domonoky (Read error: 54 (Connection reset by peer))
22:22:08bertrikfunman, I still get a blank screen on my clip
22:22:23funmanbertrik: same here
22:23:17 Join HellDragon [0] (
22:23:55 Quit n1s ("Lämnar")
22:25:50onlysoaaOh, so it turns out the P2's LCD is quite different from the D2's. It just happens that the D2's code made the P2 display something.
22:26:05 Join PaulJam [0] (
22:27:22bertrikfunman, I think I know what's wrong :P
22:27:48*funman holds his breath
22:28:05*Mikachu prepares the oxygen tank
22:28:40bertrikI added a map_section(0, 0, 4096, CACHE_NONE); to set a default non-cached linear mapping for all memory before mapping all of the other specific ranges
22:29:55bertrikI mean that helped to get display working again
22:30:12funmanhum where did that call go ? Oo
22:30:21 Join PaulJam_ [0] (
22:30:55funmani'm pretty sure it was there in the first patches ! thanks for noticing
22:31:15martian67rockbox summons demons! froggyman told me so
22:31:42onlysoaaOkay, shoot. The P2's LCD data sheet should have all the info necessary to write a driver, but I don't know how.
22:31:45 Quit robin0800_ (Read error: 110 (Connection timed out))
22:31:47Xakhso when you talk about the D2, do you mean Samsung's D2?
22:32:08onlysoaaXakh, Samsung P2, and Cowon D2.
22:32:27XakhHow well does the Cowon D2 work?
22:32:29 Join Thundercloud [0] (
22:32:36funmanbertrik: test_disk running on the Clip, can you test playback ? ;)
22:32:41onlysoaaI have a P2, and it has hardware similar to the D2, so I'm using the D2 port as a base. Not sure how to proceed though.
22:32:46 Join lyngaas [0] (
22:33:32bertriktest_disk only reads and should be relatively safe, right?
22:34:21funmantest_disk writes 300MB of data and then reads it back. It should be as safe as current SVN, except that it may break the filesystem
22:34:50Unhelpfulpixelma: that's very odd, as the album list config on pretty much everything but m3 maps BUTTON_RIGHT,BUTTON_NONE to PF_NEXT, BUTTON_RIGHT|BUTTON_REPEAT,BUTTON_NONE to PF_NEXT_REPEAT, and makes sure that BUTTON_RIGHT|BUTTON_REL and BUTTON_RIGHT|BUTTON_REPEAT with precondition BUTTON_NONE both map to ACTION_NONE.
22:35:21Unhelpfuldoes the same thing happen on ondio sim? might at least be able to trace how it matches the wrong mapping, then.
22:36:04funmanbertrik: the call was there in my 12th may patch, and absent in my 27th may patch :/
22:38:42funmanClip: test passed.
22:38:49bertrikBTW, do you know a way to disable database refresh on the clip?
22:38:58funmanno :/
22:39:55bertrikhm, "incompatible model" :/
22:40:27 Quit robin0800__ (Read error: 60 (Operation timed out))
22:40:28funmanbertrik: make clean? (allocation of room for the TTB changes the memory map)
22:40:42bertrikyes, doing that now
22:41:31Unhelpfulhrm, recorder sim is utterly broken, also.
22:46:04funmanHum I see the Clip crashing (twice when resuming playback from the main menu)
22:46:13 Quit tvelocity (Read error: 101 (Network is unreachable))
22:46:28 Join tvelocity [0] (
22:48:08bertrikfunman, I'm not familiar with test_disk but it's doing something now (it's going to take a while I think)
22:48:29 Quit PaulJam (Read error: 113 (No route to host))
22:48:42 Quit stoffel ("leaving")
22:48:49bertrikI think it's writing slightly faster than 1 MB/s
22:49:27funmanbertrik: it's computing a sequence of numbers and writing it to disk (mixing aligned/unaligned accesses and transfer sizes), and then will read it back and verify it.
22:50:28 Part Xakh
22:50:51 Quit merbanan (Read error: 110 (Connection timed out))
22:51:19 Quit pixelma (Nick collision from services.)
22:51:20 Join pixelma_ [50] (n=pixelma@rockbox/staff/pixelma)
22:51:38 Nick pixelma_ is now known as pixelma (n=pixelma@rockbox/staff/pixelma)
22:51:38 Quit amiconn (Read error: 60 (Operation timed out))
22:51:53bertrikfunman, so far so good, this is taking a while ..
22:51:58 Join amiconn [50] (n=jens@rockbox/developer/amiconn)
22:53:25funman~5 minutes
22:53:27 Join tessarakt [0] (
22:56:43Unhelpfuli think i see the trouble :/
22:57:04 Nick fxb is now known as fxb__ (
22:58:43pixelmaUnhelpful: haven't investigated further as I didn't have much time and concentrated on the crash. Revision 20800 worked and 21000 didn't (I know not much yet) so my gut feeling it that it has something to do with jpg and/or restructuring as overlay plugin
22:59:23Unhelpfulpixelma: CONTEXT_PLUGIN conflicts with ALLOW_SOFTLOCK
23:00:37CIA-38New commit by mcuelenaere (r21160): Onda VX747: try at implementing FM tuner support
23:00:42Unhelpfulso that under certain circumstances, core can up masking out CONTEXT_PLUGIN
23:01:37Unhelpfuli've fixed it locally, but i should bump min plugin version for this, so i need to figure out how to sort some things. :)
23:02:01bertrikfunman, not sure what I'm testing right now ... I took the patch, added the default linear uncached MMU mapping, edited ata_sd_as3525 to use the uncached buffer
23:02:07bertrikplayback hung after about 10 seconds
23:02:29Unhelpfulshould action_get_touchscreen_press go with touchscreen_set_mode or with the other action/button stuff?
23:02:33bertrik(and did the icache enable thing in system-as3525.c)
23:02:53 Join daurn [0] (n=daurnima@unaffiliated/daurnimator)
23:03:02funmanbertrik: well you are testing rockbox with both caches enabled
23:04:16bertrikthe HW info debug menu now says: MMU:op, Icache: op, Dcache:nop
23:07:07 Quit shadoxx (Read error: 113 (No route to host))
23:08:51CIA-38New commit by unhelpful (r21161): Fix PF keymap issue caused by conflict between new CONTEXT_PLUGIN and ALLOW_SOFTLOCK, by changing value of ALLOW_SOFTLOCK. Bump min plugin version as ...
23:09:29Unhelpfulpixelma: that should sort the keymap issue, at least it did on recorder sim for me. it might even happen to clear the crash, though i can't be certain of that.
23:09:56evilnickWhat's the current plan for (Rockbox) USB in bootloaders on PP targets for 3.3, are they to be included?
23:10:22pixelmaUnhelpful: I guess not, the crash happened with r21000 too which was from before the keymap changes
23:10:27funmanbertrik: just comment the code which disables data cache, or get my patch from FS #10048
23:11:01 Quit {phoenix} (Read error: 104 (Connection reset by peer))
23:12:30Unhelpfulpixelma: that's... quite old. does backing out the change in that revision fix it, or do you not know yet if the problem goes back even farther?
23:14:42 Join matsl [0] (
23:15:48Bagdermcuelenaere: Rockbox Blame Game!
23:15:58mcuelenaereI'm aware :)
23:18:41bertrikfunman, working nicely so far :)
23:19:00funmanbertrik: yup, i'm listening full albums with that patch :)
23:19:49 Quit daurn| (Read error: 110 (Connection timed out))
23:21:18bertrikwait, test_fps now claims 1591 fps for 1/1 greyscale ...
23:21:31CIA-38New commit by mcuelenaere (r21162): Fix reds & yellow
23:21:46 Join keverw [0] (
23:23:36 Join BHSPitLappy [0] (n=BHSPitLa@unaffiliated/bhspitmonkey)
23:25:08Unhelpfulmcuelenaere: apparently we have another report for "crash on PF track list enter". i've still no idea how buflib_buffer_out could do that.
23:25:22mcuelenaereFS# ?
23:25:29mcuelenaereand which target?
23:25:29funmanbertrik: how full is your Clip's battery? it would be interesting to run a battery bench with this patch applied
23:26:15Unhelpfulmcuelenaere: from pixelma, and on ondio. it apparently goes back to at least r21000 for her.
23:26:29mcuelenaereI'll look at it again on my target now
23:27:04*mcuelenaere has plenty of Rockbox bugs on his Onda :)
23:27:25funmanbertrik: apparently my Clip crashes everytime I use database or "resume playback"
23:28:05CIA-38New commit by amiconn (r21163): Speed up mono bitmap drawing on horizontally packed greyscale targets (greyscale iPods). Average speedup is 80% on PP5002, and 55% on PP502x. * ...
23:32:07mcuelenaereUnhelpful: yep, I can reconfirm it's l.796 (buflib_buffer_out); I'll look into it
23:32:36mcuelenaereUnhelpful: does that function gets called elsewhere?
23:32:42Unhelpfulmcuelenaere: is it possible to trace more closely inside buflib?
23:32:50mcuelenaerethat's what I'm trying to do:)
23:33:32Unhelpfuli believe there are two calls to in, both in create_track_index - one to initially force all free space out of the buflib buffer, the second to do so again after freeing a slide, if the track index runs out of space.
23:33:58 Part keverw
23:34:53 Join fdinel [0] (
23:35:53 Quit pyro_maniac ("Leaving.")
23:36:04mcuelenaereUnhelpful: buflib_buffer_shift(ctx, avail); seems to cause the hang (l.206, buflib.c)
23:37:13 Join webguest [0] (n=ad585e5d@gateway/web/cgi-irc/
23:37:16*mcuelenaere suspects the memmove call
23:37:58Bagderhm, we're now down to 20 build servers
23:37:58 Quit webguest (Client Quit)
23:38:16*mcuelenaere is right
23:38:30mcuelenaereUnhelpful: l.177 at buflib.c
23:38:49mcuelenaereI think memcpy == memmove on MIPS targets
23:38:51bertrikfunman, I'll try some things tomorrow
23:39:21***Saving seen data "./dancer.seen"
23:39:23Unhelpfulhrm. is it being called in some invalid fashion, though? zero or negative length, maybe? the buffer shoudln't really *be* empty...
23:39:26funmanBagder: do you know the rockbox git server is down ?
23:39:41Bagdernow I do!
23:39:52MikachuYou don't have permission to access / on this server.
23:40:24mcuelenaereUnhelpful: I'll try with a panicf
23:40:32mcuelenaereah no, plugins don't have access to that
23:40:56Bagderfunman: down as in how down?[0:]: errno=Connection refused
23:42:53Bagderls -l /usr/bin/git-daemon
23:42:53Bagderls: cannot access /usr/bin/git-daemon: No such file or directory
23:43:04Bagdercan that have anything to do with it? B)
23:43:26funmangit now (since several months already) doesn't use git-XX commands, but "git XX" only
23:43:29 Join robin0800 [0] (
23:43:50funmanthat must be why people using svn where seeing a post commit hook failed: "git-push failed"
23:43:55bertrikfunman, stopping playback and restarting it with Resume Playback works here, I'll try it across a reboot
23:44:14bertrikworks across a reboot too
23:44:14Mikachugit-daemon should still be there, the dashless thing only applies to interface commands
23:44:15Bagderfunman: yes, as it's done in the post commit hook for svn
23:44:32BagderMikachu: I assume the debian package is modified to not provide it anymore or something
23:44:33 Quit jfc (Read error: 54 (Connection reset by peer))
23:44:42Bagderor possibly zagor just made a mistake or something
23:44:47Mikachuthat said, i could be wrong
23:44:50 Join jfc [0] (
23:44:58BagderI'll change it to use the dash-less one
23:45:35funmanhum git-daemon is not part of my git installation, so I'm not sure if it uses the same naming convention
23:46:06Bagderit does
23:46:31Bagderat least "git daemon" is there and works
23:46:35mcuelenaereUnhelpful: rb->memmove(0x80FA37A0, 0x80F0C3B0, 395340); (RAM starts at 0x80004000 and is 16MB long)
23:46:37funmangit server is up, now I guess it will be synced at the next commit
23:46:52BagderI'll force a sync, hang on still doesn't work, if it should
23:49:03Bagderhehe, git-push was used as well
23:49:45Mikachuyou can add `git −−exec-path` to $path and dashed commands should work too
23:49:59funmanBagder: thanks!
23:53:21 Quit jgarvey ("Leaving")
23:55:21 Quit Mathiasdm ("Yuuw!")
23:57:47mcuelenaerehmm sorry, memmove != memcpy on MIPS (memcpy is assembler optimised while memmove is the C version)
23:58:18 Join safetydan [0] (n=deverton@rockbox/developer/safetydan)
23:58:54mcuelenaeresafetydan: hi, what do you think of having bitlib in Lua?

Previous day | Next day