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 2009-09-06

00:06:04 Join iCub [0] (
00:10:38 Quit mcuelenaere (Read error: 110 (Connection timed out))
00:15:36 Join mcuelenaere [0] (
00:20:01***Saving seen data "./dancer.seen"
00:22:26 Quit Hillshum (Read error: 110 (Connection timed out))
00:25:11 Quit DarkDefender ("Leaving")
00:31:41 Part domonoky1
00:44:06 Quit bah_ ()
00:44:23 Join Sajber^ [0] (
00:45:28 Quit Sajber^ (Read error: 104 (Connection reset by peer))
00:45:30CtcpPing from gevaerts!n=fg@rockbox/developer/gevaerts
00:46:54 Join Sajber^ [0] (
00:53:56 Join Hillshum [0] (
01:00:15 Quit ender` (" They could not know that this was because Greebo, as a matter of feline pride, would attempt to fight or rape absolutely an")
01:00:19 Quit Hillshum (Read error: 60 (Operation timed out))
01:00:26 Join kugel [0] (n=kugel@rockbox/developer/kugel)
01:01:13*kugel wonders when JdGordon is going to fix album art in viewports
01:02:03kugelit seems to be related to the %lh tag
01:03:19kugelunhelpfullaa is acting weird too
01:13:56 Join webguest99 [0] (n=4ac5dd3c@gateway/web/cgi-irc/
01:14:10kugelmcuelenaere: ping
01:14:16mcuelenaerekugel: pong
01:14:34kugelI have a patch to change touchscreen behavior in quickscreen a bit
01:14:57mcuelenaereyou need testers?
01:15:53kugelI tested in the sim, I just wanted to hear a touchscreen guy's opinion (testing can't be bad anyway)
01:16:18kugelthis is apparently just for the absolute mode, where I have no idea how you enter it anyway (on the onda) :p
01:16:57 Quit webguest99 (Client Quit)
01:17:13mcuelenaereit's a setting, but if you're using the sim you can always switch by pressing F4
01:17:18kugelso, the change is to add handling for the top item. and only accept the "button presses" if they hit the text, nor the area around
01:17:25 Join Eosphere46 [0] (n=4ac5dd3c@gateway/web/cgi-irc/
01:17:39kugelit=quickscreen in that sentence
01:20:21*mcuelenaere compiles
01:21:45kugelI don't think it changes anything if the crippled grid mode is used
01:23:59mcuelenaerehmm what's the difference between the top and bottom entry in the quick menu?
01:24:06mcuelenaerethe patch works great other then that
01:24:35kugelthe top item can be set to anything
01:24:45kugelit just mirrors the bottom one by default
01:26:03mcuelenaereit seems to always mirror the bottom one here
01:26:12mcuelenaereor does it depend on the right and/or left ones too?
01:28:00 Join Eosphere46_ [0] (i=IceChat7@
01:28:22 Quit Eosphere46 ("CGI:IRC (EOF)")
01:29:19kugelmcuelenaere: no
01:29:43kugelcan't you go to the context menu of a setting, and chose "set as top quickscreen item"?
01:30:33mcuelenaereahh yes, so it seems :)
01:30:39*mcuelenaere needs to keep up with new Rockbox features :)
01:30:46 Quit Eosphere46_ (Client Quit)
01:31:27*kugel thinks the check in wps.c at line 608+ is wrong
01:31:49kugelmcuelenaere: so you don't object if I commit?
01:32:02mcuelenaerekugel: nope, go for it
01:32:04 Join Eosphere46 [0] (i=IceChat7@
01:32:25CIA-43New commit by kugel (r22637): Improve and simplify touchscreen handling of the quickscreen in absolut mode - and implement handling for the top item.
01:34:47kugelmcuelenaere: do you have a wps using touchregions?
01:35:02mcuelenaerekugel: yes, cabbie
01:35:12mcuelenaere(and Pen&Paper, on the theme site)
01:36:30*kugel can't imagine that check is correct
01:38:20mcuelenaerekugel: hmm yes, that doesn't seem right
01:38:36mcuelenaerer->wvp->vp.{x,y} shouldn't be added there
01:38:47 Quit Eosphere46 (Read error: 104 (Connection reset by peer))
01:39:47 Join Eosphere46 [0] (i=IceChat7@
01:42:03 Join froggyman [0] (
01:42:04 Quit DerPapst ("Leaving.")
01:42:41saratogado the buttons on that theme work in the WPS?
01:42:49kugelmcuelenaere: exactly
01:43:07 Quit wincent_balin (Read error: 110 (Connection timed out))
01:43:24mcuelenaeresaratoga: cabbie? yes
01:46:12kugelmcuelenaere: maybe you can check if this patch breaks anything?
01:46:17kugelseems to work in the sim
01:46:43kugel(taking the chance to generalize that xy_within_viewport function :) )
01:47:16mcuelenaereif it works in the sim, it should work on target (won't test now, going to bed)
01:49:12 Quit Eosphere46 (Read error: 104 (Connection reset by peer))
01:50:03 Join Eosphere46 [0] (i=Eosphere@
01:50:58 Quit mcuelenaere ()
01:52:41 Join PaulJam [0] (
01:56:07 Join alexbobp [0] (
01:57:18CIA-43New commit by kugel (r22638): Factor out/generalise the function that checks whether a point is within a viewport to, and use it in the wps' touchregion code. This corrects the ...
02:00:17 Quit Eosphere46 (Read error: 104 (Connection reset by peer))
02:01:01 Join Eosphere46 [0] (i=Eosphere@
02:03:35 Nick fxb__ is now known as fxb (
02:03:54 Quit bertrik (Read error: 113 (No route to host))
02:06:19 Quit Eosphere46 (Read error: 104 (Connection reset by peer))
02:12:47 Quit froggyman (Read error: 104 (Connection reset by peer))
02:15:54 Join AndyIL [0] (n=pasha_in@
02:17:00 Quit Kinetix ()
02:17:05 Join Eosphere46 [0] (i=Eosphere@
02:17:59 Join froggyman [0] (
02:20:06***Saving seen data "./dancer.seen"
02:22:34 Quit Eosphere46 ("If you think nobody cares, try missing a few payments")
02:23:28 Join venn177 [0] (
02:23:58 Part venn177
02:28:18 Quit PaulJam (".")
02:29:21 Quit AndyI (Read error: 110 (Connection timed out))
02:30:59 Quit jordan` (Read error: 110 (Connection timed out))
02:37:06 Quit Tech-Guy ("( :: NoNameScript 4.02 :: )")
02:37:21 Join TechGuy [0] (
02:45:23 Nick fxb is now known as fxb__ (
02:45:48 Quit kugel (Read error: 110 (Connection timed out))
02:59:42 Quit dfkt ("-= SysReset 2.53=- Ph'nglui mglw'nafh Cthulhu R'lyeh wgah'nagl fhtagn.")
03:07:20 Join kugel [0] (n=kugel@rockbox/developer/kugel)
03:07:37 Quit kugel (Remote closed the connection)
03:07:58 Quit DrMoos ("Rockbox rules the DAP world")
03:09:07 Join Hillshum [0] (
03:16:28TheSevenok, I have a trivial read-only FTL implementation for the nano 2G, not, FYI
03:16:40TheSevenit's not thoroughly tested yet, but I think it works
03:24:46 Quit froggyman ("ChatZilla 0.9.85 [Firefox 3.5.2/20090729225027]")
03:33:51 Join froggyman [0] (
03:39:36*TheSeven can't spell his words today...
03:44:05 Quit froggyman ("ChatZilla 0.9.85 [Firefox 3.5.2/20090729225027]")
03:48:07 Join DrMoos [0] (i=mostafa@
03:50:33 Quit Thundercloud (Remote closed the connection)
03:50:47 Join than1 [0] (n=thanaton@
03:52:18 Quit Hillshum (Read error: 110 (Connection timed out))
03:58:09 Join grdxyxy [0] (n=chen@
04:03:15 Join toffe82_ [0] (
04:06:12 Quit TheSeven (Nick collision from services.)
04:06:27 Join The_Seven [0] (
04:06:32 Nick The_Seven is now known as TheSeven (
04:08:38 Join dys` [0] (
04:10:36 Quit toffe82 (Read error: 145 (Connection timed out))
04:20:08***Saving seen data "./dancer.seen"
04:21:16 Quit dys (No route to host)
04:28:10 Join antil33t [0] (
04:33:45 Join scorche [50] (n=scorche@rockbox/administrator/scorche)
04:42:02 Part toffe82_
04:46:32 Join decayedcell [0] (n=decayed_@
04:47:23grdxyxywhere to get the source code of lrcplay?
04:48:07 Quit Rondom (Nick collision from services.)
04:48:18 Join Rondom [0] (
04:51:06 Quit The3_14ed|r ()
05:05:28 Join Hillshum [0] (
05:09:06 Quit Llorean (Read error: 104 (Connection reset by peer))
05:09:27 Join Llorean [0] (
05:15:49 Join EvilJeff [0] (n=18b18de5@gateway/web/cgi-irc/
05:21:46saratogagrdxyxy: I think its on the patch tracker
05:22:54 Quit bluebrother (Nick collision from services.)
05:22:58 Join bluebroth3r [0] (n=dom@rockbox/developer/bluebrother)
05:25:01EvilJeffHi. Is it possible to have the "Database" menu reset to the mainmenu each time you enter it? When I re-enter it after doing something else, I get "lost in the jungle". Thanks.
05:26:49grdxyxysaratoga: thanks
05:36:42EvilJeffLet me try to explain better: Rockbox 3.3, iPod 5.5G 30GB, all possible features installed. I've got a few issues, trying to introduce this to my wife and she hands it back to me when something counterintuitive happens, such as the "database" menu - I have the "Files" menu configured so that it lists the entire folder structure. That way you neer get lost while navigating. The issue I'm having is that the Datab
05:40:20 Quit grdxyxy ("Leaving.")
05:44:19EvilJeffACTION Rockbox 3.3, iPod 5.5G 30GB, all possible features installed. I've got a few issues, trying to introduce this to my wife and she hands it back to me when something counterintuitive happens, such as the "database" menu - I have the "Files" menu configured so that it lists the entire folder structure. That way you neer get lost while navigating. The issue I'm having is that the Database menu does not indi
05:54:46saratogai think IRC posts are limited to a couple hundred characters
05:56:34*EvilJeff -Simplified- Can "Database" be configured to forget where you were last - start over each time from the root menu?
06:03:34JdGordonnot without doing your own build
06:08:29EvilJeffI'm assuming the same would be true if I wanted to see the entire tree structure when navigating, like when navigating files?
06:10:11JdGordonyou can of course hold down left in both and you will get to the start
06:11:58EvilJeffOK, thanks. I've got one other issue - "Playlist" doesnt add anything when "create playlist"; or only one song (Ive got 540). Cant use Playlist at all. Any ideas?
06:16:13 Quit fdinel ("Miranda IM! Smaller, Faster, Easier.")
06:19:21 Quit GeekShadow (Read error: 54 (Connection reset by peer))
06:20:10***Saving seen data "./dancer.seen"
06:31:09 Join chandoo [0] (
06:54:04 Quit Hillshum ("Ex-Chat")
06:55:22 Quit EvilJeff ("CGI:IRC 0.5.9 (2006/06/06)")
06:57:21soapyou invoking it on a file or a folder?
07:02:28 Quit TopyMobile (Read error: 113 (No route to host))
07:07:44 Join vitja [0] (n=vitja@
07:14:37 Part decayedcell
07:15:42 Quit panni__ (Read error: 54 (Connection reset by peer))
07:24:37 Join stoffel [0] (
07:27:26 Join viikatemies [0] (n=4772fe4b@gateway/web/cgi-irc/
07:28:28 Quit viikatemies (Client Quit)
07:34:19 Join viikatemies [0] (
07:37:02 Join Horschti [0] (n=Horscht2@xbmc/user/horscht)
07:37:19viikatemiesMight anyone know why an ipod 5g running 3.3 "locks up" when disconnecting from an ipod car radio adapter (it's a usa spec)?
07:46:37 Part viikatemies
07:55:43 Quit Horscht (Read error: 110 (Connection timed out))
08:11:59 Quit chandoo ("Leaving")
08:20:13***Saving seen data "./dancer.seen"
08:31:32 Join robin0800 [0] (
08:31:37 Join Rob2222 [0] (
08:33:28 Join at0m [0] (
08:43:29 Quit than1 ("Ex-Chat")
08:50:04 Quit Rob2223 (Read error: 110 (Connection timed out))
09:08:48 Quit antil33t ()
09:22:19 Join petur [0] (n=peter@rockbox/developer/petur)
09:25:42 Join TopyMobile [0] (
09:32:50 Join decayedcell [0] (
09:36:39 Join bertrik [0] (
09:44:18 Join GeekShadow [0] (n=Antoine@reactos/tester/GeekShadow)
09:51:19 Quit bertrik (Remote closed the connection)
09:54:36 Join bertrik [0] (
09:55:13 Join merbanan [0] (
10:01:54 Quit TopyMobile (Read error: 60 (Operation timed out))
10:20:17***Saving seen data "./dancer.seen"
10:21:07 Join Boopop [0] (
10:21:12BoopopHi all
10:21:16Boopopswitched on my iPod today with rockbox installed
10:21:25BoopopI get "Please use iTunes to restore"
10:21:31Boopopany help?
10:22:34bertrikBacklight brightness doesn't work on the c200v2 I have here
10:22:40advcomp2019Boopop, you came in at the wrong time because most people are sleeping or working
10:23:03Boopopgreat ='/
10:23:25bertrikI thought backlight was fairly simple, just being a setting in the DCDC converter of the ascodec, but apparently this one is different
10:23:57 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
10:26:23 Quit Boopop ("ChatZilla 0.9.85 [Firefox 3.5.2/20090729225027]")
10:33:03 Join TopyMobile [0] (
10:34:37 Join flydutch [0] (
10:44:22 Join TheSphinX^ [0] (
10:55:08GodEaterworking on a Sunday? Who are these fools ?
10:55:54 Join DarkDefender [0] (
10:56:20 Join z35 [0] (
11:00:55 Quit bluebroth3r (Read error: 113 (No route to host))
11:01:13 Join fyrestorm [0] (
11:08:46 Join pamaury [0] (n=pamaury@
11:12:36 Join antil33t [0] (n=Mudkips@
11:19:08 Quit fyre^OS (Read error: 110 (Connection timed out))
11:29:06 Quit pamaury ("exit(*(int *)0 / 0);")
11:33:28 Join mt [0] (n=MTee@rockbox/developer/mt)
11:33:51 Join pamaury [0] (n=pamaury@
11:35:44 Quit amiconn (Nick collision from services.)
11:35:44 Join amiconn_ [0] (i=quassel@rockbox/developer/amiconn)
11:35:49 Quit pixelma (Read error: 113 (No route to host))
11:35:50 Join pixelma_ [0] (i=quassel@rockbox/staff/pixelma)
11:36:09 Nick amiconn_ is now known as amiconn (i=quassel@rockbox/developer/amiconn)
11:36:11 Nick pixelma_ is now known as pixelma (i=quassel@rockbox/staff/pixelma)
11:51:00 Join bluebrother [0] (
11:52:13sinthetekmy microsd card works again :D
11:52:50sinthetekit looks like it was a rockbox problem since re-installing was what fixed it
11:53:37sinthetek(not to suggest i had no role in causing the problem)
12:03:28 Join Utchybann_ [0] (
12:03:28 Quit Utchybann (Read error: 104 (Connection reset by peer))
12:04:49 Quit pamaury (Read error: 113 (No route to host))
12:09:54 Join DerPapst [0] (n=DerPapst@
12:14:22 Join PaulJam [0] (
12:20:20***Saving seen data "./dancer.seen"
12:24:39 Join pamaury [0] (n=pamaury@
12:27:18 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
12:32:55 Quit TopyMobile (Remote closed the connection)
12:32:59 Join SIGSEGV [0] (n=user@
13:01:37 Quit mt (Read error: 104 (Connection reset by peer))
13:05:01 Join mt [0] (n=MTee@rockbox/developer/mt)
13:10:51 Quit FlynDice (Remote closed the connection)
13:13:07 Quit DerPapst (Read error: 145 (Connection timed out))
13:18:39 Join DerPapst [0] (
13:20:52 Join skyhunter [0] (
13:23:11skyhunterI tried to install RockBox on my Cowon D2 16 Gig but now I can't even start my D2.
13:23:14 Nick fxb__ is now known as fxb (
13:25:12GodEaterskyhunter: and you're aware it's a development build right?
13:25:40skyhunterYes but how can I reset that? I want my D2 to start again o.o
13:28:06LloreanIt depends on what you've done.
13:28:10GodEatermy guess would be to start it in USB mode, and then upload an original firmware to it
13:28:18GodEaterbut I've no real idea - I don't own a D2
13:28:25LloreanHaving installed Rockbox on a D2 myself, I can say that if you do it the "right" way, it should both work, and be dual-bootable.
13:28:45skyhunterI tried that tih Tcctool.
13:28:50 Quit stoffel (Read error: 113 (No route to host))
13:28:57LloreanDid you follow instructions from our site, or from some forum somewhere?
13:29:53skyhunterBut the problem is when I upload the files with Tcctool the screen turns on and it looks like it worked. But when I try to start it without usb mode it dont work.
13:30:20LloreanIt wasn't a yes/no question.
13:30:59skyhunter? You asked if I followed Instructions
13:31:30GodEateroh dear
13:31:39GodEaterLlorean: good luck :)
13:32:00Lloreanskyhunter: No, I asked where the instructions came from, but I can see how it was unclear.
13:32:31 Join stoffel [0] (
13:34:58GodEaterso not our instructions then
13:35:13Lloreanskyhunter: If you didn't follow our instructions, you really should seek help from the person who gave you the instructions.
13:36:35 Quit skyhunter (Remote closed the connection)
13:36:55 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
13:37:12GodEaterwell we fobbed him off nicely
13:37:33GodEaterhow many irc regulars do we have that actually own and are familiar with the D2 apart from you then Llorean ?
13:37:54LloreanOh, I'm not at all familiar with it. Nor am I a regular these days. :)
13:38:19LloreanI just know that, as of, oh, about a year and a few months ago (just before last DevCon) Rockbox was perfectly installable and dual-bootable by our wiki instructions.
13:38:46LloreanI think gevaerts may have one, and other than that, just the guys working on it (who aren't really on IRC)?
13:38:54GodEateronce in a blue moon
13:38:58*gevaerts may indeed have one
13:39:28GodEaterfor this reason I dislike the idea of thee D2 becoming one of these "unstable-but-supported" targets
13:39:33LloreanI imagine the touchscreen adds a few interesting support issues with regard to the WPS and a few other screens.
13:39:40GodEaterthere's virtually no-one around who know about them
13:39:52LloreanSince anywhere absolute controls are used, you can't just say "tap the cancel region" or whatever.
13:40:24gevaertsit's easy. "Hold power for a while" solves all problems
13:40:35LloreanI'm not fond of it being "supported" when it requires an SD card for regular use.
13:41:35LloreanBut that's just a personal quirk - I don't like the idea of the Rockbox install (or at least, is written data like configs, nvram, etc) being on something intended to be removed and/or swapped
13:41:42GodEatergevaerts: that instruction doesn't appear to be in the wiki page =(
13:41:44LloreanSame complaint I may have had about other projects.
13:41:49domonokymaking the D2 a "unstable" target might get us more people who know it :-)
13:42:19LloreanMaybe we shouldn't use the term "supported" then
13:42:40LloreanSomething more... like "available" or "provided"
13:43:03domonokywording is to be discussed. i would just use "stable" and "unstable" and remove the "supported" keyword :-)
13:43:03LloreanBuilds we compile regularly, since we don't promise actual "support" for any of them.
13:43:06GodEaterif Rob hung out here more I wouldn't mind so much
13:43:19*gevaerts prefers "supported", because the term "supported" will get used anyway. Using it officially will reduce confusion
13:44:07domonokyso "unstable" and "supported" and there can be "supported" ones without release ?
13:44:12Lloreangevaerts: Well, maybe just tinker with the wording so it's clear that Rockbox (the software) supports the hardwares, rather than "we support those builds"
13:44:12 Join Thundercloud [0] (
13:44:26domonokyor move the ones without releases to the "unstable" category ?
13:44:39GodEaterI thought the idea of this change was that we *would* provide support for them
13:44:47gevaertsLlorean: from what I understand, we do support them
13:45:03GodEatergevaerts: that was my understanding too
13:45:08gevaertswith less guarantees than for the stable builds of course
13:45:23GodEaterit just seems a little daft though, since so few of us actually *can* support them
13:45:40*domonoky think the important thing of this new category, is not that we support them, but that we no longer hide them.
13:45:48gevaertsthat's really not very different from e.g. mr100 or iaudio m3
13:46:18GodEaterwe hide them ?
13:46:34LloreanGodEater, gevaerts: Well, my thinking is more "we don't promise support for anything, really"
13:46:47LloreanThere's no guaranteed support staff, no experts on hand, so much as "whoever happens to be around might know"
13:47:02domonokyyes, they are not on the front page. we are essentially saying rockbox doesnt run on AMS (or other ) targets.
13:47:07GodEaterthe chances of that "might" seem to be diminishing
13:47:15LloreanSo "support" as in "fix bugs", sure, "support" as in "have support staff", not at all really
13:47:20gevaertsLlorean: the big difference between "supported" and "not supported" I think is whether we get to tell people to go away
13:47:53Lloreangevaerts: Yeah. I was just wondering if it were possible to make sure what we mean by it is what the public will expect from it.
13:48:26*domonoky thinks "support staff" is a bonus. we can not garantie that we always have people who know a specific target. even for releases targets.
13:49:15Lloreandomonoky: That was kinda my point.
13:49:27 Quit robin0800 (Remote closed the connection)
13:49:35gevaertsexactly, and that means that while it is indeed an important point, it's not at all related to whether or not we can call these new targets supported
13:49:52 Join robin0800 [0] (
13:50:22Lloreangevaerts: I was just wondering if there were better phrasing for it, is all.
13:50:23domonokywe should just choose new terms for our categorys, that should cleanup the confusion.
13:51:13gevaertsLlorean: I think that whichever phrasing you come up with, people will still call it "supported"
13:51:54gevaertsso I think the best thing to do is just calling it "supported", and making sure we clarify what we mean by that on the website
13:51:55domonokygevaerts: but wont people call both "stable" and "unstable" supported ?
13:52:04gevaertsdomonoky: yes
13:52:15*gevaerts was unclear
13:52:39Lloreangevaerts: Yes, but even as simple as writing "Rockbox supports these hardwares:" vs "We support Rockbox on these hardwares:" suggests too different meanings. I just think it needs kept in mind as anything is written, is all.
13:52:47gevaerts"Stable" and "unstable" are categories within "supported" in my view
13:53:17gevaertsLlorean: very true
13:53:26domonokyso "supported" means: its on the frontpage. "stable"/"unstable" means the current state of the code, then we have the ones which have a release, and the ones where we have support staff for :-)
13:53:53LloreanI don't think we should ever imply or promise support staff. There's no guarantee at any point anyway.
13:54:14gevaertsdomonoky: I would hope that "stable" means "will be in the next release" :)
13:54:45domonokygevaerts: then we have to move some current targets back to "unstable"
13:56:14domonokyfor example iaudio m3
13:56:14 Quit FOAD (Read error: 104 (Connection reset by peer))
13:56:19 Join FOAD [0] (
13:56:34Llorean"released" is a factor of documentation and installation, rather than stability, right?
13:56:46gevaertsyes and no. We currently don't have the stable/unstable division, so all we basically do is reexamining all supported targets and put them in the proper category. No real "moving back" involved
13:56:53 Quit stoffel (Read error: 113 (No route to host))
13:57:01LloreanOr, rather, in addition to stability.
13:57:57domonokyLlorean: yes.. thats why we have "stable" targets without release.
13:58:36domonokybut of course we could just include that requirement into the "stable" category
13:59:27LloreanWe could just have a separate way of denoting installation status.
13:59:48LloreanSo we have a table of Stable and Unstable. Ones with a * after them require manual installation, the others are installable with rbutil
14:00:09LloreanThat way you can include in rbutil unstable ones and not confuse things about releases at all.
14:00:22LloreanYou don't even need to throw the "release" term into the mix - it's just there for version numbering.
14:00:48domonokysounds good. how about documentation ?
14:01:54LloreanFor the future, for a version numbered release you need a usable manual. Beyond that, it doesn't factor into the front page display?
14:02:38LloreanI think manuals are very important. I also don't think we really need a complex way of telling people they're available - we should be funneling people through the manual as their first step anyway.
14:02:55domonokyso manual would be only relevant for making releases, and not for the target -status ?
14:03:14LloreanIn my opinion, yes. The status should be about Rockbox itself.
14:03:43LloreanTo get displayed on the front page you need to be "installable by an ordinary user" (no special hardware, or particularly difficult tricks) and "usable day to day"
14:03:56domonokyanother idea would be to make 3 categorys: "unstable","stable" concerning only code quality, rbutil/manual can be missing (displayed with *) and "gold" where we have everything.
14:04:03LloreanTo be "stable" you need to more or less be ready for release in terms of Rockbox functionality.
14:05:36 Quit robin0800 (Remote closed the connection)
14:05:55domonokyLlorean: i would think for "unstable" it would be enogh if "installable by user", "no hardware critical issues" (corruptions etc) and "plays music".
14:06:53LloreanThat's more or less what I meant by "usable day-to-day". It needs to play music (all available formats) without crashing (playlisting is included in this), and not risk damaging the player.
14:07:27LloreanBasically, it needs to succeed at the core "purpose" of Rockbox. Recording, FM Radio, plugins and viewers can all come later.
14:08:21domonokyi would think for "unstable" it would be even ok, to crash once a day and only provide some music formats..
14:08:28 Quit fyrestorm (Read error: 104 (Connection reset by peer))
14:09:28pcc1pamaury: just to let you know that I had a chance to work on MTP and my work is available in this git repository: git://
14:10:18pamaurypcc1: I currently working on MTP, I'm nearly having MTP works read-only on my computer, perhaps in five minutes it's done
14:11:13pcc1pamaury: oh. would have been good if you had told me
14:11:38pamauryI just worked a lot this night :)
14:12:07pcc1(I am probably also guilty of not telling you)
14:12:08pamauryI don't see the mtp code in the repo link you have me (with the web interface)
14:12:31pcc1you have to check the "mtp" tag
14:12:48pamauryAh yes
14:13:10pamauryWhat did you changed ?
14:13:50pamauryAh you added dircache. I fear we will have incompatible modification :)
14:13:59pcc1it's all there in the log
14:15:49pamaurypcc1: you changed exclusive storage thing but you shouldn't I think. It's related to a usb xfer size problem
14:16:06gevaertspamaury: in what way?
14:16:16pamaurypcc1: cpu boost
14:16:31pamaurygevaerts: cpu boost
14:16:53pcc1if I didn't turn that off then dircache would be disabled during the usb session
14:16:56gevaertsthat means you need to separate boost from exclusive storage, not request exclusive storage access when you don't need it
14:17:47pamauryYes for now it's a hack. I'll change it
14:17:53gevaertsyou also can't actually access the filesystem (or if you can, you shouldn't) if you ask for exclusive storage
14:19:37 Join stoffel [0] (
14:20:06pamaurypcc1: In the code I see a dircache_is_valid_ptr call. Does this function exists ?
14:20:24***Saving seen data "./dancer.seen"
14:20:27pcc1I added tht function to dircache
14:20:46pcc1it checks whether the pointer is in range and whether its pointers are also within range
14:21:02pcc1(of the dircache reserved area)
14:21:40pamauryAh that's really useful. I wanted to have such a function
14:23:51pamauryArg, my code still don't work. I have a problem with libmtp: it does really recognize the device as MTP but only as PTP. I changed some thing but it's not sufficient. I will commit my code anyway so you can see what I've done
14:29:37 Quit FOAD ("I'll be back")
14:29:40 Quit AlexP (Read error: 104 (Connection reset by peer))
14:29:52 Join AlexP [0] (n=alex@rockbox/staff/AlexP)
14:30:02 Join FOAD [0] (
14:32:41 Quit domonoky (Read error: 104 (Connection reset by peer))
14:35:29pamauryI've commited my changes plus some that you corrected
14:36:37 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
14:36:41pamaurypcc1: I've lookked at your implementation of dircache_get_root_ptr and I'm sure it's necessary to have it. A call to dircache_get_entry_ptr("/") is sufficient no ?
14:39:27pcc1pamaury: at the time after reading the code I wasn't sure if that would work. but if it worked for you then I guess it's fine
14:40:39pamauryThe thing is that it returns the first entry in the root directory anyway. It's not really well defined if I could say so
14:41:47 Join ender` [0] (
14:42:28 Part at0m ("more has been done with less")
14:45:43pamauryThe thing is that it returns the first entry in the root directory anyway. It's not really well defined if I could say so
14:45:46pcc1oh, you never made the out -> in change? how on earth did it (almost) work for you without that?
14:46:11pamauryI belive I did it five minutes ago. I'm also wondering how it worked :)
14:46:36 Quit FOAD ("I'll be back")
14:46:56pamauryI made two commites: one for my changes and one for your changes.
14:47:00 Join FOAD [0] (
14:47:40 Join crculver [0] (
14:48:18pcc1yeah. what should have happened is we should have merged our changes together
14:48:26pcc1oh well, never mind
14:48:32crculverWhen, while listening to a track, I navigate through the database and select "Insert next..." on an item, Rockbox accesses the disk. Yet my database is loaded to RAM and>>
14:48:48crculverthe playing song is nowhere near finished, so no immediate need to preload. Why then is there disk access?
14:49:03pamaurypcc1: that the problem of having two repos. And the thing is that I only an internet access at my school for now which is a pain
14:49:50*pamaury has just broken its sansa's fat filesystem
14:51:26pcc1pamaury: nah, that's only the problem of not undrstanding how DVCSs work
14:52:03pamaurypcc1: yes also
14:57:04*pamaury hates fsck which doesn't work on ubuntu
15:02:00pamaurypcc1: I have a working version of MTP (read-only)
15:03:10pamauryI will commit it but I don't understand why at some point I broke my filesystem. Perhaps there is a bug somewhere, perhaps not
15:04:39 Join robin0800 [0] (
15:06:09pamaurypcc1: could you explain me some thing about DVCSs, I agree I don't really undertsand everything but then I could learn
15:07:45 Quit stoffel (Read error: 113 (No route to host))
15:13:06 Part decayedcell
15:15:41 Quit pamaury ("exit(*(int *)0 / 0);")
15:16:44 Join decayedcell [0] (
15:17:03 Part decayedcell
15:21:42 Join pamaury [0] (n=pamaury@
15:23:28 Quit antil33t (Read error: 104 (Connection reset by peer))
15:23:42 Join antil33t [0] (n=Mudkips@
15:24:02 Join panni_ [0] (
15:25:53 Join teru [0] (
15:34:04 Quit merbanan (Read error: 110 (Connection timed out))
15:35:23domonokyACTION needs some ideas on how to best present the detected device(s) to the users in rbutil. I have the following info: 0..n devices. Each device has 1..n potential targets (if rbutil isnt sure) and 0..n mountpoints (again rbutil can not find everything, and might be unsure), and a status like "incompatible device" or "device in wrong mode". Additionaly we need to give the user a way to select a device (if there is more then 1) and a way
15:38:24 Quit ender` (" I find television very educating. Every time somebody turns on the set, I go into the other room and read a book. -- Grouc")
15:39:33domonokycurrent, very ugly UI: with many found devices:
15:46:51 Join fdinel [0] (
15:46:58bluebrotherdomonoky: have you written up a concept on how to perform detection?
15:47:23bluebrotherIMO the UI is the easiest part. More problematic is to figure how to actually do the detection best ...
15:48:09domonokybluebrother: i have a current attempt (work inprogresss) in the tracker, but i dont have a writeup for it at moment.
15:48:30domonokyi think the UI is something we need, needless of how we approach the detection.
15:48:40bluebrotherwell, I was throwing ideas around for a concept. Haven't gotten very far yet though.
15:49:03bluebrothersure, but the UI is easy compared to the rest ...
15:49:42domonokycurrent attempt goes like this: use usb-ids for a list of connected devices. map the targets to it -> list of device with potentially more the 1 one target per device.
15:50:57domonokythe scan potential mountpoints and update the list with this "in a clever way", and in the end, if there is still work todo. use traget specific detection methods and again merge the info cleverly with the target list.
15:52:00domonokyfinally there might be needed some cleanup for dublicate targets, and you have the list of connected devices, each with 1-n potential targets, and 0-n mountpoints and a status.
15:53:34domonokytwo things are common for all approaches to autodetection: a nice gui and a way to test it. (the tracker entry has a way to test autodetection on all targets, without ipods and sansas (need a way to fake them so ipod/sansapatcher accepts them).
15:56:02 Join ender` [0] (i=krneki@
15:57:26*bluebrother hands domonoky a p :)
15:58:18*domonoky takes the p and all other chars he might have missed :-)
15:59:35 Join Thundercloud_ [0] (
16:00:16 Quit Thundercloud (Read error: 101 (Network is unreachable))
16:00:55 Quit pamaury ("exit(*(int *)0 / 0);")
16:11:43 Join pamaury [0] (n=pamaury@
16:14:29CIA-43New commit by teru (r22639): brickmania, blackjack, jewels, bubbles: Remove save file only if player resumed the game loaded from the file.
16:17:01bertrikdomonoky, can you try a radio patch on your e200v2 this afternoon?
16:17:36 Quit Thundercloud_ (Remote closed the connection)
16:17:51 Join Thundercloud_ [0] (
16:19:20 Join chandoo [0] (
16:19:59*domonoky detects that the current incompatible device detection in rbutil is broken. the name entry in rbutil.ini should have "" if it contains spaces..
16:20:28***Saving seen data "./dancer.seen"
16:23:23CIA-43New commit by Domonoky (r22640): rbutil: correct several name entrys in the rbutil.ini file.
16:23:44robin0800domonoky: how do manufactures check they have the correct one when thy update the firmware
16:25:24domonokyrobin0800: i dont know, but they have it easier as they only need to detect their own, and some even dont do this, they just provide a upgrade image which you have to place on their player.
16:25:59domonokyif we had a crossplatform way to map usb-ids to moutpoints, this would be sooo much easier... :-)
16:26:50robin0800domonoky: the sansa updater at least knows what player you have
16:29:11domonokyfor pp-sansas and ipods its also easy, we have the patchers which can detect the target/mountpoint exactly (altough IMHO only if there is only one ipod/sansa connected)
16:29:32bertrikdomonoky, can you test this patch on your e200v2 and see if the radio still works?
16:30:17domonokyother targets are more difficult to detect, because we dont have a way to map usb-ids to mountpoints, and because there are dublicate usb-ids.
16:30:36bertrikthis fixes radio on the c200v2 I borrowed and still works on my clip (which has a slightly newer version of the tuner), I wonder if it still works on the e200v2
16:31:03robin0800domonoky: I would make only one connected a requirement
16:31:28bertrikI'll test it on my samsung yp-s3 too (which has a si4703, even newer, similar to the one in the gigabeat s)
16:34:29domonokyrobin0800: that doesnt really help, as there can be left-overs on possible mountpoints. currently rbutil just takes the first thing it finds, which can lead to wrongly detected mountpoints. and there are targets which rbutil can not distinguish, so rbutil should show a selection. so better try to show all that is connected.
16:39:11 Quit iCub (Read error: 60 (Operation timed out))
16:39:47bertrikyay, fixed si4700 driver works on my samsung yp-s3 too
16:40:46 Join Polo [0] (n=polobric@
16:45:30*domonoky can confirm that radio also works with this patch on e200v2
16:47:54 Join merbanan [0] (
16:50:07 Quit robin0800 (Remote closed the connection)
16:51:07 Join Lear [0] (i=chatzill@rockbox/developer/lear)
16:52:44CIA-43New commit by bertrik (r22641): Sansa c200v2: fix si4700 tuner initialisation to make it work for the c200v2
16:52:55 Join stoffel [0] (
16:53:01 Quit DrMoos ("Rockbox rules the DAP world")
16:53:49 Quit chandoo ("Leaving")
17:00:14 Quit Thundercloud_ (Remote closed the connection)
17:04:35 Join chandoo [0] (
17:05:01 Quit mt (Read error: 113 (No route to host))
17:05:22 Quit chandoo (Read error: 104 (Connection reset by peer))
17:05:58 Join chandoo [0] (
17:08:52 Quit Polo (Read error: 110 (Connection timed out))
17:09:04 Join polobricolo_ [0] (n=polobric@
17:09:09 Join robin0800 [0] (
17:10:24 Join Thundercloud [0] (
17:11:14 Quit AlexP (Remote closed the connection)
17:11:19robin0800domonoky: you probably know Linux has this program now just run sudo blkid
17:15:48 Quit pamaury (Remote closed the connection)
17:17:38 Quit maraz (Remote closed the connection)
17:19:15 Join maraz [0] (
17:20:30domonokyrobin0800: i just tried, but blkid doesnt list my e200v2 only my hd. and also remember that we need methods working win/lin/mac :-)
17:21:04 Quit teru ("Quit")
17:24:30robin0800domonoky: I thought it showed all mount points?
17:25:17domonokyrobin0800: for me it doesnt. but i tried it in my VM, so that might confuse it.
17:27:26robin0800domonoky: I know it doesn't see usb modems etc.
17:28:14domonokyand even if it did work, we would need the same on win and mac ..
17:29:38robin0800domonoky: can't you use the same methods and build it in?
17:29:59 Quit maraz (Remote closed the connection)
17:30:31 Quit polobricolo_ ("Good bye")
17:30:41domonokyrobin0800: no, because blkid probably uses methods specific for linux. usb is quite different on windows.
17:31:32 Join maraz [0] (n=maraz@
17:32:53 Quit GeekShadow (Read error: 113 (No route to host))
17:33:39 Join DrMoos [0] (i=mostafa@
17:35:49 Quit DrMoos (Client Quit)
17:43:54 Join barrywardell [0] (n=barrywar@rockbox/developer/barrywardell)
17:48:49 Quit gevaerts (Nick collision from services.)
17:48:57 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
17:59:50 Join AlexP [0] (n=alex@rockbox/staff/AlexP)
18:02:40 Quit robin0800 (Remote closed the connection)
18:03:09 Join robin0800 [0] (
18:03:52 Nick fxb is now known as fxb__ (
18:08:26 Join froggyman [0] (
18:17:30 Join Thundercloud_ [0] (
18:19:22 Quit Thundercloud (Read error: 101 (Network is unreachable))
18:20:29***Saving seen data "./dancer.seen"
18:24:29TheSevencan somebody tell me where the code to reboot an ipod to disk mode is? i bet i've seen that somewhere in the rockbox source a while ago...
18:25:50 Join jordan` [0] (i=gromit@
18:27:41 Join shotofadds [0] (n=rob@rockbox/developer/shotofadds)
18:32:53merbananTheSeven: IIRC it used to reboot to that mode before usb support and battery recharging was there
18:33:23TheSevenany idea where in the source tree this could be?
18:36:23merbanansorry no idea
18:38:11shotofaddsLlorean, GodEater: I may have misinterpreted what "unstable" entails - I wouldn't have expected them to be "supported" in any meaningful sense, other than acknowledging they exist and providing an easy(ish) install method.
18:38:43shotofaddsI think it's important to allow discussion of functional-but-not-supported targets in the forums, at least
18:39:29*bertrik wonders about the exact difference between the RADIO_FREQUENCY and RADIO_SCAN_FREQUENCY setting
18:41:21gevaertsshotofadds: in my view "supported" means "you don't get told to go away if you have a question about this". I think that includes both irc and the forums
18:42:25 Quit robin0800 (Remote closed the connection)
18:42:36shotofaddspersonally I don't see why we should tell anyone to go away. "I don't know" is a perfectly good answer in my book
18:44:43domonokyTheSeven: take a look a usb-fw-pp502.c
18:45:09 Join MG_Man [0] (
18:45:10 Join pamaury [0] (
18:45:13domonoky~line 48 is the reboot to discmode for this target.
18:45:14 Join robin0800 [0] (
18:45:24 Quit pamaury (Client Quit)
18:48:21JdGordonI'm going to do a slightly annoying commit in a few minutes... apparently at least the e200 doesnt compile the virtual LED for the sim because the #define is in the wrong place.... so to find out which other targets need fixing I'm going to cause red on them....
18:49:25domonokyJdGordon: boo ! the buildserver is not your testsystem :-)
18:49:38 Quit Dhraakellian (Read error: 104 (Connection reset by peer))
18:49:40JdGordonyou got a system fast enough to run every build quickly?
18:49:43*JdGordon sure doesnt
18:49:49n1sJdGordon: good, many targets have way too much stuff excluded for the sims
18:49:59gevaertsshotofadds: it depends I think. What if you do know but you also know that rockbox doesn't actually work properly yet on this target, the user gives a distinct impression of not going to be a developer anytime soon, and helping would take lots of time?
18:50:01domonokyno, its probably ok, if you revert it directly after it.
18:50:09JdGordonyeah, thats obvisouly the plan :)
18:50:15JdGordonI thought I should warn everyone its coming
18:50:19gevaertsJdGordon: how many builds, and how much tome do you have?
18:50:43JdGordonevery target... all day really, but preferably while I get breakfast happening :p
18:50:54shotofaddsgevaerts: true, but it doesn't take long to say "it doesn't work yet, sorry" :p
18:50:54*JdGordon has no qualms about using the build system for this
18:51:42*n1s agrees it's a fine use for the system
18:51:46*domonoky wants the build-table to stay nice and green, so i dont really like this...
18:52:19domonokyperhaps we should setup a seperate build-master for such things in the future ?
18:52:30*gevaerts can do all sim builds in ten minutes or so. The regular builds would take about 15 minutes
18:52:58*n1s would probably need several hours to build all sims
18:53:18n1s...slow netbook
18:53:31*JdGordon does it in a way that is actually a legit commit
18:53:53gevaertsI don't have a problem using the build system for this, but I don't really like these things in the svn history
18:54:11CIA-43New commit by jdgordon (r22642): Make the e200 compile the vled for the sim... other targets are probably doing the same thing so find out which.. commit will be immediatly followed ...
18:54:49CIA-43New commit by jdgordon (r22643): fix red :)
18:54:57 Join liar [0] (
18:55:22n1sJdGordon: why not just have config.h do #ifndef CONFIG_LED\n #define CONFIG_LED LED_VIRTUAL ?
18:55:23JdGordonit will actually be pretty funny if the e200 is the only one with that issue... but I doubt it :)
18:55:59JdGordonn1s: yes, thats probably a good way to do it
18:56:52 Quit Thundercloud_ (Remote closed the connection)
18:57:43 Join FOAD_ [0] (
18:58:28JdGordonthats quite a few broken targets :/
18:59:46domonoky1280 points for JdGordon !
19:01:08 Quit antil33t (Read error: 104 (Connection reset by peer))
19:01:25 Join antil33t [0] (n=Mudkips@
19:01:49*JdGordon decides to do n1s's suggestion
19:01:55JdGordonconsidering how many are broken
19:02:19 Quit FOAD (Read error: 110 (Connection timed out))
19:02:19 Nick FOAD_ is now known as FOAD (
19:03:06shotofaddsI'd like to get the D2 added to the theme site to encourage touchscreen WPS development (as per mrobe500 / ondas), but there's an issue re: touch and non-touch enabled themes as it can also use iPod Video 320x240 themes.
19:03:17shotofaddsAny ideas how to make sure touch themes don't end up in the iPod list?
19:03:41n1sJdGordon: that should probably be accompanied by a check to see that targets with real led doesn't have this mistake and gets the virtual led in the sim
19:03:41shotofadds..and how to label non-touch vs. touch themes on the D2 list
19:03:51 Join pamaury [0] (n=pamaury@
19:05:54JdGordonI dont know how targets with a real LED handle the sim
19:08:06 Quit robin0800 (Remote closed the connection)
19:08:11JdGordonapparently not at all
19:08:36domonokyJdGordon: real leds are not shown in the sim.
19:08:49JdGordonsomething to maybe add later then
19:11:54 Join notlistening [0] (
19:12:16CIA-43New commit by jdgordon (r22644): move the CONFIG_LED #define out into config.h as all but 3 targets actually have a real one (and apparently lots had it #defined in a !SIMULATOR ...
19:12:41notlisteningdomonoky, hi still plugging away just doing the comprehensive feedback model for the client and server
19:12:44JdGordonunfortunately that didnt fix the problem I was investigating :(
19:13:17notlisteningdomonoky, and another bit of good new is I have the vista TTS voice working under wine ;)
19:13:25domonokynotlistening: when you have a server ready, which i can try with rbutil. just tell me :-)
19:13:52notlisteningdomonoky, ok :)
19:14:21notlisteningdomonoky, just letting you know i am still here and working on it
19:14:41domonokythats nice :-)
19:19:41 Quit notlistening ("Leaving")
19:20:51JdGordonis anyone else seeing 2 copies of the AA being displayed?
19:24:12JdGordonUnhelpful: where does the AA get drawn by the code?
19:24:15*JdGordon cant find it
19:25:11JdGordonah, in aa.c
19:25:34*JdGordon punches kugel in the nose!
19:27:10JdGordonhmm... maybe I'll taker that back
19:27:51JdGordonapparently (I tihnk) the AA is drawn on the conditional %?C<> tag as well as the usual %C, both of which draw in the current viewport which is not a very good way to do it...
19:29:59 Join than0 [0] (
19:35:42 Join mitk [0] (
19:46:46 Quit pamaury ("exit(*(int *)0 / 0);")
19:47:12 Join Eosphere46 [0] (i=Eosphere@
19:50:03 Quit chandoo (Read error: 104 (Connection reset by peer))
19:50:55 Join chandoo [0] (
19:53:17 Quit chandoo (Read error: 54 (Connection reset by peer))
19:57:22 Join Thundercloud [0] (
20:10:44 Quit PaulJam (".")
20:13:46 Quit Eosphere46 (Read error: 54 (Connection reset by peer))
20:15:33 Join Eosphere46 [0] (i=Eosphere@
20:20:32***Saving seen data "./dancer.seen"
20:28:35 Quit Eosphere46 ("Do fish get thirsty?")
20:32:19 Quit flydutch ("/* empty */")
20:38:16 Join p3tur [50] (n=petur@rockbox/developer/petur)
20:38:44 Quit vitja ("Ex-Chat")
20:41:08 Join fyrestorm [0] (
20:44:23 Join Xerion [0] (
20:45:35 Join Grahack [0] (
20:47:07 Quit amiconn (Nick collision from services.)
20:47:09 Join amiconn_ [0] (i=quassel@rockbox/developer/amiconn)
20:47:21 Join pixelma_ [0] (i=quassel@rockbox/staff/pixelma)
20:47:21 Quit pixelma (Nick collision from services.)
20:47:29 Nick amiconn_ is now known as amiconn (i=quassel@rockbox/developer/amiconn)
20:47:41 Nick pixelma_ is now known as pixelma (i=quassel@rockbox/staff/pixelma)
21:07:27 Join Calaco [0] (n=464621f0@gateway/web/cgi-irc/
21:10:08Calacocan rockbox be configures to exclude directorys from the database system
21:10:31gevaertsyes. Look for database.ignore in the manual
21:14:27Calacok ty
21:26:39Calacocan any1 give me an example apon what kind of entry to put in database.ignore
21:27:32JdGordondont you just copy that file to any folder you dont want in the db?
21:28:35LearYep, the content of the file doesn't matter.
21:28:49Calacokk, works now thx =)
21:32:27 Quit Lear ("ChatZilla 0.9.85 [Firefox 3.5.2/20090729225027]")
21:35:01 Quit Calaco ("CGI:IRC (EOF)")
21:46:56GrahackHi, I noticed a typo in "recieves" in the rockbox-cvs section
21:47:43 Quit Sajber^ ("Leaving.")
21:48:40 Join AsaelReiter [0] (n=d59730fa@gateway/web/cgi-irc/
21:53:56 Quit TheSphinX^ (Read error: 104 (Connection reset by peer))
21:54:24 Join robin0800 [0] (
21:57:01AsaelReiterIn brickmania, shouldn't the new-balls speed been changed like r22582?
21:59:29bertrikAsaelReiter, I'm not really familiar with brickmania, but it does make sense indeed
22:04:34bertrikkkurbjun, did you notice FS #10585? there is a problem with the speed (sometimes being 0 apparently) of various objects in brickmania
22:12:47TheSevensaratoga: we have circumvented the ipod copyright problem now
22:13:33TheSevenwe just move the OFW, in its original encrypted state, to the filesystem, and replace the image with our own bootloader
22:14:16 Join webguest777 [0] (n=47ba0bf9@gateway/web/cgi-irc/
22:15:07 Quit froggyman (Nick collision from services.)
22:15:14saratogaTheSeven: can this be done from the notes exploit?
22:15:53 Nick webguest777 is now known as froggyman (n=47ba0bf9@gateway/web/cgi-irc/
22:16:38 Quit ehntoo (Read error: 110 (Connection timed out))
22:17:20TheSevensaratoga: we don't need an exploit any more
22:17:32TheSevenjust one time to bootstrap the rest
22:17:49TheSeventhen we can re-distribute an encrypted image which can be installed directly
22:19:16saratogaso you would load the exploit once, then copy over a new disk partition from image?
22:19:29saratogai'm trying to understand what the exact install procedure will be
22:20:01TheSevenyou only need to use the exploit (on any ipod) to encrypt the bootloader once. the encrypted bootloader can then be redistributed
22:20:05 Quit AsaelReiter ("CGI:IRC (Ping timeout)")
22:20:24 Join AsaelReiter [0] (n=d59730fa@gateway/web/cgi-irc/
22:20:34***Saving seen data "./dancer.seen"
22:20:36rasherAnd is that bootloader installed?
22:20:41rasherAnd how is that bootloader installed?
22:20:55TheSeventhe install procedure will just be reading out the encrypted OFW, storing it to a file, and replacing it by the encrypted bootloader
22:21:09TheSeventhat can all be done through access to the mass storage device
22:21:10rasherAh, so a disk operation
22:21:27rasherThat may be the simplest install procedure since the archos devices
22:21:29saratogaso the encrypted firmware is exposed over MSC?
22:21:57saratogarasher: gigabeat F is the same way, just overwrite its firmware image on the disk
22:22:16TheSeveneven on the newest ones - even though they tried to hide it, but you just have to use some custom scsi commands in order to access it
22:22:47saratogaTheSeven: are you interested in rockbox development or just reverse engineering?
22:23:18TheSevenand, of course, the bricking risk is near zero, as we're only rewriting the NAND firmware, and disk mode is stored in NOR. if the bootloader fails somehow, it will just reboot to disk mode to let the user fix it, without the hassle of pressing key combos
22:23:42 Quit stoffel (Remote closed the connection)
22:24:19TheSevensaratoga: well, not really development on the application side, but getting this target running - maybe... i have exams coming, though, so i won't have that much time during the next 6 weeks
22:25:29saratogawas going to ask if you're working on a real bootloader now
22:25:52TheSevenyou mean a rockbox bootloader?
22:26:13TheSeventhe question is whether we even need one, or whether we can fire up the core directly
22:26:41TheSevenif it's just that we need to copy that thing into memory and jump to it, then it's already done
22:27:03rasherTheSeven: We want a bootloader that loads Rockbox from disk, for easy updates
22:27:14rasherFrom the file system, I should say
22:27:18saratogawell a bootloader would be more or less generic much like how we share some with ipodlinux and sansalinux
22:27:34markunTheSeven: yes, that's all you need. Read the file .rockbox/rockbox.ipod and execute it
22:27:36saratogabut yeah, something that lets you update your code without having to reencrypt the firmware image
22:28:18TheSevenmarkun: if that can be changed to /iLoader/rockbox.bin then the bootloader is done, else, well, i would need to change the path in the source code :-P
22:28:49saratogafor now it doesn't matter, we can change it when it gets commited to our svn
22:29:32markunTheSeven: I'm really impressed by what you guys did. In fact I didn't think this port would ever happen :)
22:29:42saratogaalthough putting a bootloader and install directions on our wiki might be a nice way to interest other people in your work
22:29:54TheSevenwell, i was pretty much impressed myself that doing a bootloader was so easy
22:30:06markunbertrik: I'm back in Enschede this week and could mail you your meizu
22:30:21TheSevenafter i got a readonly FTL running yesterday (well, 5:30 am today, in fact), it was just putting pieces together and doing a fancy boot menu
22:31:12markunTheSeven: how did you figure out the FTL? Just by looking at the disassembly?
22:31:21TheSevendisassembly + openiboot source
22:31:49markunbertrik: do the meizu's have the same FTL as the ipods?
22:32:00TheSevenmarkun: looks like they don't
22:32:08TheSevenyou'll be able to reuse the lowlevel NAND driver, though
22:32:31markunstrange that these whimory versions are so different
22:32:32bertrikmarkun, yes that would be nice
22:32:41markunor is it just small changes?
22:32:58TheSevenwe couldn't find anything they have in common besides that name, in fact
22:33:18markuncrazy man
22:33:19bertrikI see the iphoneboot code expects some strings that I cannot find in the samsung/meizu NAND contents
22:33:42markunTheSeven: did you put your code online somewhere?
22:33:54bertrikI could zip up some NAND contents and put them online
22:36:17TheSevenmarkun: PMed you a link
22:36:37 Quit merbanan (Read error: 110 (Connection timed out))
22:38:05saratogaTheSeven: how complicated does the FTL look? do you think it will be feasible to reverse engineer write support ?
22:38:35TheSevensaratoga: in the long term, maybe... but this is a hell of a lot of code
22:39:03saratogado you think its worth looking into implementing our own FTL in the meantime?
22:39:03pixelmaJdGordon: from your description about the album art "problem" I would think that my WPS should be broken too but it's not (at least hasn't been so far/not that I noticed)
22:39:08TheSevenmy read driver is ~500 lines, write support will probably cross the 10.000 boundary
22:39:55TheSevensaratoga: using our own FTL means breaking compatibility with the NOR-based bootloader, so our code won't be booted any more, at least if we don't fake some whimory structures
22:40:15CtcpPing from gevaerts!n=fg@rockbox/developer/gevaerts
22:40:29saratogaand we have no way to disable the built in USB mode which would be quite dangerous
22:41:02 Join eveenstra [0] (
22:41:10 Part eveenstra
22:42:08saratogaTheSeven: eventually when you get a chance could you link a copy of your encrypted firmware image and instructions how to install it somewhere (or even to our wiki)?
22:44:21TheSevenbasic instructions: use some tool to extract the encrypted OFW, dd an image to the ipod device, create the data partition, copy some files (including the extracted firmware) to it
22:44:41 Join domonoky1 [0] (
22:45:29TheSevenyou could also do it without repartitioning, but this way you can reclaim ~70MB of space from the apple firmware
22:45:48TheSeven(and no, the 0 in there is not a typo!)
22:45:55rasherRockbox tries to maintain dualboot it possible
22:46:27 Nick dys` is now known as dys (
22:46:32TheSeventhis is dualboot-capable, only disadvantage: you lose the data partition contents if you repartition it the rude way
22:47:27TheSevenin fact I shrink the firmware partition to 126KB and then put a 6MB image into the filesystem, instead of having a 70MB garbage partition
22:47:34TheSevenyou'll lose Nikepod support, though :-P
22:48:00TheSeven(which could be worked around, too)
22:48:29GrahackI filed a bug report ( but just found a fix. Should I open a patch entry in FS with a link to the related bug report or just comment the first report attaching the patch ?
22:48:40rasherGrahack: the latter
22:48:53Grahackrasher: thanks
22:49:29 Quit domonoky1 (Client Quit)
22:50:14 Join domonoky1 [0] (
22:50:32 Join funman [0] (n=fun@rockbox/developer/funman)
22:50:56pixelmaJdGordon: at least it doesn't seem broken with r22608...
22:51:05 Quit Thundercloud (Read error: 104 (Connection reset by peer))
22:52:04Grahackargh, shame on me: I just did the contrary rasher told me to do. Now I think I should leave things like that and go to bed... Sorry.
22:52:28rasherGrahack: no big issue, both ways are fine, really
22:52:30 Join mt [0] (n=MTee@rockbox/developer/mt)
22:52:34TheSevenbtw, who else, besides linuxstb, is interested in pushing this forward? who has a nano2g?
22:54:01 Join Thundercloud [0] (
22:55:50 Quit Grahack ("Leaving.")
22:56:04 Quit AsaelReiter ("CGI:IRC (Ping timeout)")
22:56:06saratogai think bertrik and linuxstb were the main two people but i'm not really sure
22:56:30TheSevenwhere is linuxstb, btw? i haven't seen him saying something in a long time
22:56:40saratogai think hes on vacation
22:57:13bertriksaratoga, I don't have any ipod, I looked at meizu and samsung players with (almost) the same SoC
22:57:32saratogai know, but you're interested on the FTL anyway right?
22:57:32TheSevenbtw, cramming the bootloader into SRAM was a little tricky. the FTL needs ~45KB of heap and ~2KB of stack, the FAT32 driver needs ~33KB of stack, plus 10KB worth of code, so a total of 90KB
22:58:12JdGordonpixelma: it might be broken but your tag placements might hide it
22:58:26TheSeventhat thing has 176KB of SRAM, but the OFW seems to reuse some structures set up by the NOR bootloader, and is really picky concerning overwriting them
22:59:17 Quit mitk ("Leaving")
23:00:22pixelmaJdGordon: well it follows your example in the forums (with stuff in between)
23:00:53JdGordonok, oh well... ill try fixing it properly today anyway
23:02:07 Quit domonoky (Read error: 113 (No route to host))
23:06:49 Quit bluebrother ("leaving")
23:09:32 Quit p3tur ("Zzzzz")
23:09:45 Quit than0 ("Ex-Chat")
23:10:42 Quit petur ("Zzzzzz too")
23:14:49 Quit bmbl ("Bye!")
23:19:00funmanJdGordon: would you try the e200v2 recovery method on your Clip+?
23:19:12JdGordonthats the plan :)
23:19:22JdGordonwheres the instructions?
23:23:57 Nick fxb__ is now known as fxb (
23:25:17 Join DrMoos [0] (
23:25:52*JdGordon is having a bit of trouble getting the pcb off the front plastic
23:26:12JdGordonbut so far i havnt found any solder joints that might be the ones needed
23:26:49funmani can trigger this mode on the clip by shorting 2 specific pins of the flash, to make the chip think the flash isn't working
23:26:58bertrikI wonder what the 2x3 copper colored stripes are in the bottom right of the front of the PCB
23:26:59funmanbut then you wouldn't be able to access the flash where the firmware is supposedly stored
23:27:16bertrikmaybe for a JTAG header
23:27:57 Join iCub [0] (
23:29:48*JdGordon has it completly free now
23:30:25JdGordoni also see two unused solder points
23:30:35JdGordonwhere is the abi pics?
23:31:45JdGordonpic 14... just at the bottom left corner of the white label box...
23:32:29JdGordonthey seem to be about the right place
23:33:12TheSeventhis looks more like solder pads for an additional chip
23:33:38TheSevenanother set of such strips (with a chip) is on the bottom of the picture
23:34:30bertrikstill looks like a place to solder a header to me, not for a chip
23:35:15bertrikthe pads look way too big (long) for a chip to be placed there
23:35:18JdGordonfunman: what do you reckon? join those pins and connect to usb?
23:35:47funmanyes (making sure the clip+ is off before pluging it)
23:36:10JdGordondo they need to stay joined while in usb?
23:36:21JdGordoncan i do this with a screwdriver and only 2 hands?
23:36:38bertrikif you have soldering skills and equipment, I recommend to solder a small value resistor instead of a hard short
23:36:43funmanyes we can!
23:37:26funmanJdGordon: i would start plugging usb on the computer so only a little press with one hand is needed to make the connection, and the other hand to connect the 2 pads with a screwdriver
23:37:46 Quit HBK (Read error: 60 (Operation timed out))
23:37:52funmanbertrik: a resistor could prevent unwanted damage?
23:38:39JdGordondo they need to stay connected the whole time?
23:38:51funmanclip+ has its own forum section on sansa website
23:39:19funmanJdGordon: only until you can see a device connected on the computer ("while :;do sudo dmesg -c;sleep 1;done" on linux for example)
23:39:26 Join safetydan [0] (n=deverton@rockbox/developer/safetydan)
23:41:57CIA-43New commit by mcuelenaere (r22645): Lua plugin: fix bad behaviour for, 'w') ...
23:42:09JdGordonsoldering time...
23:42:28 Join HBK [0] (
23:44:05bertrikJdGordon, if it turns out that those pins were not meant to be connected, about 100 ohm (for example) would only cause a current of about 40 mA, i.e. a bit high but probably not going to burn anything out
23:44:23bertrikonly do this if you have soldering skills of course
23:45:00JdGordonwe'll find out :)
23:45:20bertrikif they _were_ meant to be connected, then about 100 ohm (same example) would probably still be low enough to activate the bridge
23:45:46JdGordonit didnt seem to do aything bad when they were connected while it was on...
23:46:01funman < a bug in clip+ which I hope would get fixed in a near firmware update
23:47:38bertrikfunman, :)
23:49:50 Quit bertrik ("De groeten")
23:49:55JdGordonwheres my multimeter gone?
23:50:53rasherfunman: I guess what we need is JdGordon to go bughunting to find as many embarassing flaws as he can
23:51:05rasherMaybe if he had experience in software QA
23:51:12*DerPapst pings JdGordon's multimeter and waits for an answer...
23:51:18funmanexploitable vulnerabilities are a plus
23:52:09 Join GreatBeaver [0] (
23:52:20GreatBeaverwhat do you guys think of archos 7? amazon is selling for half off now
23:53:24 Quit funman ("free(random());")
23:53:59krazykitGreatBeaver, they don't run rockbox, for one
23:54:05GreatBeaveroh no
23:54:13GreatBeaveri will never buy archos
23:54:27AlexPThen why bring it up?
23:55:11JdGordonnothing in dmesg with the 2 pins joined...
23:55:26GreatBeaverAlexP: until now i didnt know rockbox didnt run on archos
23:55:40AlexPIt does on some archos
23:55:42scorche|shit run on some archoses
23:55:58AlexPAnd has a cunning list for you to check
23:56:31GreatBeaverdoes the archos 7 have digital output?
23:56:37AlexPoff topic

Previous day | Next day