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 2009-05-17

00:01:29skull0Good night,
00:01:59skull0excuse me someone could lend me a hand about hardware info on the sony nwd-b105?
00:02:38 Quit Seedy ("cu, Andre")
00:03:27kkurbjunskull0: did you check the wiki? Usually if someone is actively working on a device, or considering it there's a wiki page with at least the hw information. I have not heard of anyone workong on that particular device offhand
00:03:55skull0the wiki has no info
00:03:57kkurbjunthe new ports forum sometimes has information that's not on the wiki too
00:04:04skull0only have the nw-1000 and like
00:04:59skull0ok! Ill check the forum
00:05:13skull0 thanks kkurbjun
00:05:38kkurbjunno problem
00:05:58skull0see ya!
00:06:01 Quit skull0 ("Leaving")
00:07:15 Quit robin0800_ (Remote closed the connection)
00:09:33 Join Seed [0] (
00:13:21dfkti'm having issues with the WPS default status bar, used in a conditional
00:14:02dfktin condition a it still flashes on and off, over the full-screen album art that should be there
00:14:03 Join Ubuntuxer [0] (
00:14:45 Quit {phoenix} (Remote closed the connection)
00:15:02 Join DrMoos [0] (
00:15:05 Quit moos (Read error: 104 (Connection reset by peer))
00:15:17 Join bertrik [0] (
00:15:29 Quit LambdaCalculus37 ("Fwump")
00:16:36 Quit kugel (Read error: 110 (Connection timed out))
00:18:26 Quit Ubuntuxer (Client Quit)
00:22:04 Quit renke ("leaving")
00:22:19Horschtdfkt, i would think that you can not use the stautsbar in a conditional...
00:23:07dfktdamn, that sets me back in my plans
00:23:31Horschtbut I am not realy sure. It doesn't specificaly say so in the wiki
00:24:03dfktit works "somewhat", but it flickers when AA is displayed, switching between status bar on and off
00:24:15dfktin no-AA mode it works perfectly fine, displaying the bar
00:24:36Horscht"The conditionals nest, so the text in the if and else part can contain all % commands, including conditionals."
00:24:45Horschtactualy, it says right there that it should work
00:25:24dfkti don't think i made a mistake:
00:26:23HorschtI am not a wps guy, realy.
00:26:29Horschtit's all cryptic to me
00:26:58dfktok, but thanks for verifying that it *should* work :)
00:27:05Horschtit's not very nice too look at a .wps file and trying to find out how it actualy looks.
00:27:15Horschtwell, the wiki says it does.
00:27:42dfktit's a x5/h10 wps, i can zip it up if it's any help
00:28:25Horschtno, I don't have a x5/h10 nor do I understand wps syntax
00:28:50dfktok, thanks though
00:29:01Horschtyou'll either have to stick around till someone with the knowledge comes arounf
00:29:05Horschtor ask in the forums
00:29:08dfktwill do
00:29:28 Quit tomers ("ChatZilla 0.9.84 [Firefox 3.0.10/2009042523]")
00:31:01Horschthm... try this, dfkt:
00:31:21 Quit PsiloX (Remote closed the connection)
00:32:00Horschti.e. do not specify a width and heighth
00:32:04dfktnope, still flickers
00:33:59 Join kushalone [0] (n=kushal@
00:34:44dfkthowever, using the same line for the non-aa conditional makes it flicker too... so you might be on to something
00:40:24dfktseems the status bar flickers as soon as i define a width/height other than 0
00:41:38Horschtwhich is weird as you are defining a viewport that is 0x0 in size...
00:41:58Horschtunless 0x0 means "as big as it needs to be"
00:42:22dfktindeed.. for example when i define the non-aa status bar as 160x1, only the first row of pixels flickers in non-aa
00:43:10dfkt160x2 = 2 lines flicker, etc
00:43:54Horschtstupid idea: what happens with this conditional:
00:44:48Horschtand commenting out the two statusbar viewports you defined earlier
00:45:27Horschtno wait... I don't think that'll work... You could try, but I think it wont
00:45:33dfktdoesn#t work, reverts to the default, i.e. error
00:45:45 Quit ZincAlloy ("CGI:IRC")
00:46:27 Quit Thundercloud (Remote closed the connection)
00:46:32Horscht"...they must be on their own line."
00:46:36Horschtforgot about that
00:46:42 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
00:46:48dfktthe %wd/we tags?
00:48:35Horschtmaking this line a lie:
00:48:36Horscht"The conditionals nest, so the text in the if and else part can contain all % commands, including conditionals."
00:49:47dfktseems the very specific hard coded status bar is a bit different than all other items that can be placed in a wps...
00:50:33Horschtso you probably can't use the statusbar in a conditional in any way, not even via viewports... you'd probably have to define your own custom statusbar (i.e. a Viewport containing all the wanted information in another way)
00:50:37dfktvery time consuming workaround would be to mimic the status bar by coding all items on it separately, with their graphics and all
00:50:44dfktjinx :)
00:51:37Horschtyeah, that's what I meant
00:52:01dfktthanks Horscht, i will look into it, the graphics aren't that hard to recreate - probably easier than to get them out of the source
00:52:12 Part mcuelenaere ("Leaving.")
00:54:17Horschtbut remember: I do not actualy have any knowledge about wps, I am just speculating :D
00:54:46dfktok :) soo.. if anyone knows what's going on, my question still stands :D
01:04:59CIA-38New commit by mcuelenaere (r20973): Try at implementing interrupt endpoints in the Ingenic Jz4740 USB driver
01:17:44***Saving seen data "./dancer.seen"
01:23:49 Quit rwcr (Read error: 113 (No route to host))
01:24:14 Quit kushalone ("This computer has gone to sleep. Macs are like babies. They need a lot of sleep. :)")
01:26:38 Join kugel [0] (n=kugel@rockbox/developer/kugel)
01:26:43 Quit kugel (Read error: 104 (Connection reset by peer))
01:26:52 Join kugel [0] (n=kugel@rockbox/developer/kugel)
01:29:31 Quit tessarakt (Read error: 110 (Connection timed out))
01:29:33 Quit flydutch ("/* empty */")
01:30:34 Join tessarakt [0] (
01:40:29 Quit Llorean (Read error: 104 (Connection reset by peer))
01:43:11 Join Horschti [0] (n=Horscht@xbmc/user/horscht)
01:43:37 Join Thundercloud [0] (
01:44:08 Quit ender` (" On the other hand, you have different fingers.")
01:51:08 Quit merbanan (Read error: 110 (Connection timed out))
01:55:32 Quit efyx_ (Remote closed the connection)
01:57:49 Join Horscht86 [0] (
01:58:38 Quit Horscht (Nick collision from services.)
01:58:45 Join barrywardell [0] (n=barrywar@
01:58:50 Nick Horscht86 is now known as Horscht (
02:06:17 Join Chex_ [0] (
02:07:58 Quit bertrik (Read error: 113 (No route to host))
02:09:01 Quit xnyhps ("Zzzzz...")
02:10:16 Quit amiconn (Nick collision from services.)
02:10:18 Join amiconn_ [50] (n=jens@rockbox/developer/amiconn)
02:10:38 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
02:11:29 Quit pixelma (Nick collision from services.)
02:11:32 Join pixelma_ [50] (n=pixelma@rockbox/staff/pixelma)
02:11:51 Nick pixelma_ is now known as pixelma (n=pixelma@rockbox/staff/pixelma)
02:12:08 Quit Horscht ("Verlassend")
02:16:39 Quit fdinel (Read error: 110 (Connection timed out))
02:16:44 Quit Horschti (Read error: 110 (Connection timed out))
02:17:13 Quit Chex (Read error: 110 (Connection timed out))
02:36:26 Join andrewcodemonkey [0] (
02:37:16andrewcodemonkeyis there any way to leave out certain parts when compiling?
02:37:32Unhelpfulwhat parts, exactly?
02:37:56andrewcodemonkeywell, I am only wanting the rockbox binary, so i don't need /lang or /bitmaps
02:38:03andrewcodemonkeyi'm not entirely sure
02:38:40 Join krazykit [0] (
02:38:41andrewcodemonkeybasically i'm impatient and every time i make a small change, rebuilding takes near ten minutes really do want /lang and /bitmaps :P
02:39:18andrewcodemonkeyhaha, what i mean is i dont want to *re* build them
02:39:31Unhelpfulit takes 10min even if you're not building from scratch? it should only be rebuilding changed files, or things that depend on them.
02:40:04Unhelpfulmake will already not redo things that don't have a changed dependency. if you want to improve things, you might try using ccache.
02:40:13andrewcodemonkeywell, i'm not making in the same directory, i am keeping different builds separate
02:44:12andrewcodemonkeyok let me put it a different way. is there any way to compile only the rockbox binary?
02:46:38Unhelpful*without* building the files in /bitmaps that are used to build the binary?
02:47:16andrewcodemonkeyok. point taken.
02:47:42andrewcodemonkeythanks =)
02:48:02Unhelpfulfirst of all, make -j might help a bit if you aren't using -j now
02:49:03 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
02:49:23andrewcodemonkeywell, i'm running in a VM on a single core laptop circa 2002 so it probably won't make much different
02:49:25Unhelpfulnot kidding about ccache, it will recover a lot of the time you're losing if you use a new dir each time you build
02:49:46andrewcodemonkeyah, ok, i have no knowledge of ccache. i'll have a look. thanks.
02:49:55 Quit DrMoos ("Rockbox rules the DAP world")
02:50:50 Join itcheg [0] (i=62db4767@gateway/web/ajax/
02:51:36Unhelpfulmake -j is also not useless with single cores, it can help make sure that whichever resource is the bottleneck *actually* gets saturated, instead of idling between building individual files.
02:51:52Unhelpfulif you're very short on memory, make -j2 or -j3 might be better than a bare -j
02:52:53andrewcodemonkeyok, thanks. i'll give it a try while ccache is installing
02:53:27Unhelpfulccache will probably be the larger gain, since it will save you rebuilding some of those files.
02:54:20andrewcodemonkeysure, that makes complete sense. re-building the same code twice is exactly what i'm trying to achieve, obviously, so a cache system is perfect
02:55:38Unhelpfulif you have several more ancient laptops you could try distcc ;)
02:56:50andrewcodemonkeyhaha even if i did, somehow i would guess the time taken to switch them all on and get connected would defeat any gain
02:58:09 Quit tvelocity (Read error: 110 (Connection timed out))
02:58:31Unhelpfulyes, you generally use distcc as a daemon on a system that is left running. but you *might* try to use it, if you have such a system, and build some cross-compilers for it.
02:58:35 Join tvelocity [0] (
03:02:11andrewcodemonkeyah, ok. sadly this is the only system i have.
03:02:33andrewcodemonkeyto be honest i really ought to save up some cash and buy a new computer
03:03:10andrewcodemonkeythen i could leave this one on permanently as a server
03:07:46andrewcodemonkeyright, ccache is installed. should i symlink gcc to it, or arm-elf-gcc, or both?
03:08:12CtcpIgnored 2 channel CTCP requests in 8 seconds at the last flood
03:08:12*Unhelpful is assuming that 3-4 more add or shift ops is worth getting rid of a 32x32->64 multiply in the scaler
03:09:49Unhelpfulbasically, if i take the scaler back to "dividing" the values into a fixed-point representation between the horizontal and vertical parts of the scale, i think i can the the multiply that's used to do that conversion for free.
03:10:15 Quit tessarakt ("Client exiting")
03:10:53andrewcodemonkeyi think there is possibly some grammar missing ^^ as i'm not sure what you are getting at
03:11:54scorchehe isnt talking to you
03:12:22andrewcodemonkeyi guessed that, obviously, i was just pointing out that there may be a mistake
03:13:12 Join Strife89 [0] (n=michael@
03:13:14 Nick andrewcodemonkey is now known as andrew288 (
03:17:48***Saving seen data "./dancer.seen"
03:26:09 Quit Strife89 ("Reboot.")
03:27:12 Quit miepchen^schlaf (Read error: 113 (No route to host))
03:36:29 Join Strife89 [0] (n=michael@
03:36:34 Quit Thundercloud (Remote closed the connection)
03:38:58 Quit barrywardell ()
03:46:35 Quit dfkt ("-= SysReset 2.53=- Ph'nglui mglw'nafh Cthulhu R'lyeh wgah'nagl fhtagn.")
03:56:14 Quit perrikwp (" ajax IRC Client")
03:57:09 Join miepchen^schla [0] (
03:57:15 Quit Strife89 ("Switching clients.")
03:58:17 Quit tvelocity (Read error: 110 (Connection timed out))
03:58:43 Join tvelocity [0] (
04:03:56 Join wpyh [0] (n=william@
04:05:05andrew288... I just discovered svn revert has a recursive function.
04:05:28andrew288which pretty much makes my above nonsense invalid.
04:11:30 Join toffe82_ [0] (
04:14:17 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
04:17:35 Quit Riku (Read error: 54 (Connection reset by peer))
04:28:12 Quit toffe82 (Read error: 110 (Connection timed out))
04:29:38 Quit itcheg (" ajax IRC Client")
04:35:48 Quit andrew288 ("ChatZilla 0.9.84 [Firefox 3.0.8/2009032600]")
04:36:20 Join killan_ [0] (
04:38:12 Quit killan (Read error: 104 (Connection reset by peer))
04:40:09 Join Llorean [0] (
04:43:26 Quit wpyh (SendQ exceeded)
04:47:07 Quit antil33t (Read error: 110 (Connection timed out))
04:48:02 Join itcheg [0] (i=62db4767@gateway/web/ajax/
04:50:12 Join cmwslw [0] (
04:51:00 Quit miepchen^schla (Read error: 101 (Network is unreachable))
05:07:10 Join andrew288 [0] (
05:07:48andrew288um, any idea why plugin chessclock.c would break my build?
05:08:43 Part cmwslw ("Ex-Chat")
05:08:49andrew288sorry, thats an ambiguous question. more specifically, i just got this error from make: "chessclock.c:568: error: duplicate case value"
05:09:19andrew288i've checked the line and i cant see how that has anything to do with what i've been changing (key bindings for the cowon d2)
05:09:55andrew288that file hasnt been changed in svn for five months either.
05:12:27andrew288is anybody awake?
05:12:42andrew288or is my IRC etiquette lacking?
05:14:16Lloreanandrew288: If someone has an answer, they'll answer
05:14:33LloreanBut there's 120 people in the room, any number of which could be AFK or asleep (considering the current time in Europe, especially)
05:14:52LloreanAnd it wouldn't make much sense for everyone to chime in with "I don't know, but I did see your message"
05:15:55andrew288ah, ok. thanks. i guess its just sometimes a little offputting when there is no response at all, tbh i would prefer if at least one person said that
05:16:54 Quit Horscht (Read error: 110 (Connection timed out))
05:17:37 Quit andrew288 ("ChatZilla 0.9.84 [Firefox 3.0.8/2009032600]")
05:17:49***Saving seen data "./dancer.seen"
05:18:37 Quit Seed ("cu, Andre")
05:19:32MarcGuay__andrew288: it probably means that the same button is defined twice in that plugin after your changes.
05:20:40 Quit kugel (Read error: 110 (Connection timed out))
05:43:24 Quit itcheg (" ajax IRC Client")
05:47:15 Join wooj [0] (n=wooj@unaffiliated/wooj)
05:47:25 Part wooj ("Leaving")
06:27:21 Join animalstyle [0] (
06:29:13 Part animalstyle
06:44:16 Nick JdGordon|afk is now known as JdGordon (n=jonno@rockbox/developer/JdGordon)
06:48:48 Join jck [0] (n=3b5d74e9@gateway/web/cgi-irc/
06:49:12jckcan i install rockbox on a transcend tsonic 820
06:49:32scorcheis that device listed on the front page?
06:50:34scorchethen it looks like you have an answer
06:52:44 Part wincent ("Kopete 0.12.7 :")
06:53:39 Quit CaptainKwel (Remote closed the connection)
07:09:07 Quit KBH (Read error: 110 (Connection timed out))
07:11:10 Quit jck ("CGI:IRC")
07:17:52***Saving seen data "./dancer.seen"
07:19:46 Quit tvelocity (Read error: 110 (Connection timed out))
07:20:14 Join tvelocity [0] (
07:29:11 Quit perrikwp (" ajax IRC Client")
07:36:25 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
07:49:02 Join kushalone [0] (n=kushal@
08:25:22CIA-38New commit by jdgordon (r20974): get repeat/hold actions working on the touchscreen WPS. prepending an action with & means that this will only get triggered if you hold down the area. ...
08:26:03JdGordontouchscreen wps is just about finished... the last thing to figure out is a slider mechanism for volume and seeking...
08:36:00 Join stoffel [0] (
08:47:20 Join Rob2223 [0] (
09:05:11 Quit Rob2222 (Read error: 110 (Connection timed out))
09:12:51 Join saratoga_lab [0] (n=9803c264@gateway/web/cgi-irc/
09:13:07 Join barrywardell [0] (n=barrywar@
09:17:56***Saving seen data "./dancer.seen"
09:22:43 Quit barrywardell ()
09:35:10 Quit tvelocity (Read error: 110 (Connection timed out))
09:35:46 Join tvelocity [0] (
09:54:08 Quit saratoga_lab ("CGI:IRC (EOF)")
10:07:01 Join ender` [0] (
10:12:14 Quit tvelocity (Read error: 110 (Connection timed out))
10:12:45 Join tvelocity [0] (
10:14:48 Join matsl [0] (
10:18:36 Join stripwax [0] (
10:23:48 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
10:24:53 Join n1s [0] (n=n1s@rockbox/developer/n1s)
10:34:01 Join bertrik [0] (
10:47:16 Quit tvelocity (Read error: 110 (Connection timed out))
10:47:45 Join tvelocity [0] (
11:00:56 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
11:05:16 Join Ubuntuxer [0] (
11:07:24 Join merbanan [0] (
11:12:59 Join antil33t [0] (n=Mudkips@
11:17:57***Saving seen data "./dancer.seen"
11:26:25UbuntuxerHi, Does anybody know a smart ability to avoid unused parameter warnings in a function, if the parameter is e.g. just used only on color displays.
11:27:00n1syou can just void cast them
11:27:11n1slike (void)param;
11:38:45 Join BXCracer [0] (
11:41:54Ubuntuxerthank you very much
11:54:26 Quit perrikwp (" ajax IRC Client")
11:55:08 Quit stoffel ("leaving")
11:57:24 Quit tvelocity (Read error: 110 (Connection timed out))
11:57:46 Join tvelocity [0] (
11:59:07 Join mt [0] (n=MTee@
12:00:10 Join miepchen^schla [0] (
12:04:29 Quit stripwax (Read error: 54 (Connection reset by peer))
12:07:39 Quit Ubuntuxer ("Leaving.")
12:07:39 Join stripwax [0] (
12:24:29 Join Jaykay [0] (
12:25:43JaykayAlexP: i found the "problem" with the patch - i used some tabs
12:26:00AlexPah, naughty naught
12:26:00Jaykaybtw searching for tabs in the rockbox trunk givex 18k results
12:26:27AlexPdoesn't matter as far as this is concerned
12:26:51AlexPYou definately shouldn't be introducing new ones
12:28:18AlexPAnyway, I'm still not convinced about the point of moving this all around - it doesn't add anything IMO and it just obscures who wrote it in the first place
12:28:22bertrikJaykay, IIRC last time I looked, especially the plugins and codecs contained tabs, only a few in firmware
12:29:53 Join petur [0] (
12:30:11Jaykaybertrik: searching in apps/codecs gives 14k results :)
12:35:34AlexPJaykay: Anyway, I think these are all things that the next time someone is editing that section can be changed. If I apply thid patch now, it'll show as me having written everything which won't be very helpful if people are trying to track down why something was written as it was
12:38:42JaykayAlexP: didn't you edit this section a few days ago?
12:38:54AlexPyes, little bits
12:39:01AlexPbut not big swathes
12:39:15Jaykayanyway - close it if you think it's helpful - imo the code really looks much better
12:39:32Jaykaydon't think....
12:39:33AlexPI don't think it really matters
12:47:05Jaykayhow about a big commit of nonsense and whitespace-changes - 18k tabs, //-comments, >80 character lines and so on
12:47:23Jaykayit was just an idea ;)
12:48:47n1sJaykay: the codecs contain a lot of tabs because we do generally not change the style of imported code
12:49:14n1sbecause that would make merging in new upstream versions a lot more work
12:51:49 Join moos [0] (i=mustapha@rockbox/staff/moos)
13:01:47 Join Makuseru [0] (
13:02:18 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
13:03:32 Join Lear [0] (i=chatzill@rockbox/developer/lear)
13:06:56 Join flydutch [0] (
13:17:58***Saving seen data "./dancer.seen"
13:22:48 Join Thundercloud [0] (
13:24:26 Quit Makuseru (Read error: 104 (Connection reset by peer))
13:26:13 Quit markun (Remote closed the connection)
13:35:13 Quit Thundercloud (Remote closed the connection)
13:52:45mcuelenaereis anyone here able to tell me how I can verify whether usb hid on Rockbox works (on Linux)?
13:53:32gevaertsyou need one more patch to actually send HID events
13:54:59gevaertsAfter that, run xev to see input events
13:56:10CIA-38New commit by bertrik (r20975): Clean up some unused #includes
13:57:47mcuelenaeregevaerts: I checked out tomers git repository so I didn't think I needed a patch
13:57:52mcuelenaerebut I'll try xev
13:58:03gevaertsah ok. In that case you should have everything
13:59:49mcuelenaerehmm do I have to enter some kind of special menu to trigger usb hid events to be sent to the host?
14:00:48gevaertsif you're running an MSC+HID connection, you should be able to press keys on the USB screen
14:01:59mcuelenaereand if a HID-only connection?
14:02:16*mcuelenaere suspects something's wrong with his USB driver + interrupt endpoints
14:02:20gevaertsthen the debug menu
14:03:44mcuelenaerehmm still nothing
14:04:22*mcuelenaere enables logf in the USB HID driver
14:05:43mcuelenaerehmm I don't see any HID: Sending 0x%x messages, so I think there's something wrong with intercepting the buttons
14:07:19 Quit tvelocity (Read error: 110 (Connection timed out))
14:07:52 Join tvelocity [0] (
14:09:25gevaertsmcuelenaere: is HID enabled properly? There's a debug menu item for that as well
14:10:00 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
14:10:10mcuelenaeregevaerts: it is, but I think the problem is me + git
14:10:43mcuelenaeresomehow the apps/ code doesn't contain tomers changes
14:12:22gevaertsmcuelenaere: do you have the hid branch or the master branch?
14:12:39mcuelenaereI only see the master branch
14:14:26gevaertshow did you clone?
14:14:43mcuelenaere : git clone git://
14:15:07gevaertsok. Does "git branch" list just master?
14:16:22mcuelenaereI tried git branch hid, but that just created a hid branch
14:17:46gevaertsI'm working a bit differently. I use a "normal" rockbox git checkout, with tomers' branches added into it
14:18:25 Join Thundercloud [0] (
14:18:53mcuelenaereI seem to have a ~/usb_hid/Rockbox-USB-stack-enhancement/.git/refs/remotes/origin/hid file.. (which contains the latest commit 430150e99f20d33c045fadfbcccb5cc78233d54e)
14:19:55gevaertsAfter that, "git checkout tomers_usb_hid" or "git checkout master" to switch
14:20:07*gevaerts thanks Unhelpful for his advice on this
14:22:55gevaertsmcuelenaere: I'm also having issues with HID-only
14:22:59 Quit krazykit ("Connection reset by beer")
14:23:44mcuelenaeregevaerts: mine were probably due to nothing HID-related being in apps/
14:24:17 Join stoffel [0] (
14:29:23 Quit stoffel ("leaving")
14:31:12*mcuelenaere has a lot of new USB HID-related items in his debug menu, but still none of them are working
14:32:02gevaertsmcuelenaere: what does lsusb -v -d <your-VID>:<your-PID> say?
14:32:26mcuelenaeregevaerts: if I do a sudo cat /dev/input/by-id/usb-Rockbox.org_Rockbox_media_player-event-mouse , should that display the HID messages?
14:32:44gevaertsyou get that? I don't...
14:34:34 Join GeneralTru [0] (
14:34:49GeneralTruEy people
14:35:09bertrikHI GeneralTru
14:35:43gevaertsmcuelenaere: how did you configure things? No USB_STORAGE and USB_HID?
14:35:48GeneralTruI got a question i heard that when you downgrade your sansa fuze firmware from v2 too v1 you can use rockbox i did so but it does not work did that person lie?
14:36:32mcuelenaeregevaerts: yes
14:36:46mcuelenaere(and USB_SERIAL, but I don't think that matters?)
14:36:52AlexPGeneralTru: Rockbox isn't supported on any Fuse
14:37:07gevaertshm, maybe it does...
14:37:14GeneralTruive read that that is cause it has v2 firmware but i thought that if you downgraded to v1 it might work mmm
14:37:23AlexPGeberAnd generally you cannot do that as the hardware is different
14:37:36AlexP-Geber @ GeneralTru
14:38:02GeneralTruwell i could instal rockbot but just not the bootloader ;p
14:38:34GeneralTruWell i suppose i have to be patience then
14:38:57AlexPInstalling Rockbox in that sense is just copying some files to the disk
14:39:15AlexPi.e. the files will be there but do nothing
14:39:21GeneralTruis there no way to copy the bootloader files too the disk?
14:39:35 Quit mt (Read error: 110 (Connection timed out))
14:41:55AlexPIt won't do anything - Rockbox need to be written for specific hardware
14:42:21GeneralTruSo i better give up right;p
14:42:25AlexPIt is like saying if I install the firmware from my microwave onto my toaster will the toaster become a microwave
14:42:39GeneralTrulol true
14:43:12GeneralTruOke thanks then i will stop doing this then
14:43:59GeneralTrui also heard someone had a alpha version or something for the rockbox sansa fuze version is that anywhere available as it was a random guy at some strange forum
14:48:48gevaertsmcuelenaere: do you have CHARGING_ONLY?
14:50:20mcuelenaeregevaerts: yes
14:53:21gevaertsGeneralTru: stop it
14:56:25 Quit GeneralTru ()
15:00:51tmztfuze is in svn though, it just has to be built and installed manually
15:01:31tmztnot supported means some specific in rockbox, there's no manual and the builds aren't listed on the build page as well as some other things
15:01:32tmztand he left
15:02:03n1sit also means that it might work poorly or not at all
15:02:14n1sor steal your lunch
15:02:19mcuelenaeregevaerts: sounds like a bug in USB driver management?
15:03:20gevaertseither that or me doing stupid things. I'm still not sure
15:03:22tmztwhat is usb hid used for currently?
15:03:33gevaertstmzt: testing :)
15:03:53tmztwheel or whatever work as buttons?
15:04:31 Join robin0800 [0] (
15:05:30 Quit tvelocity (Read error: 110 (Connection timed out))
15:05:38 Join tessarakt [0] (
15:05:42gevaertsyes. Also a better-working "dummy" driver
15:05:45mcuelenaeretmzt: Rockbox's coolness factor ;)
15:05:52 Join tvelocity [0] (
15:06:28 Quit robin0800 (Client Quit)
15:06:46 Join stoffel [0] (
15:07:27 Join robin0800 [0] (
15:08:08tmztwould it be possible/pratical to make fm radio work through usb (as sound card)?
15:08:53gevaertsyes, if someone does the work
15:09:23Unhelpfultmzt: it would be possible to do output over USB, either talking to a USB host and appearing to capture audio, or (only for those devices with host support) by talk to a USB audio device. but nobody has done it yet.
15:10:39 Quit robin0800 (Client Quit)
15:11:18tmztit requires iso support though, right?
15:11:27gevaertsit does, yes
15:12:39 Nick fxb__ is now known as fxb (
15:13:40bertriktmzt, some targets now play radio by simply enabling an analog input channel of the codec, it needs to be digital to be sent over usb
15:16:16n1sbertrik: most targets can record radio though so they should be able to handle the A/D conversion fine
15:17:04 Join robin0800 [0] (
15:17:06tmztthere are no wifi targets in progress?
15:18:03***Saving seen data "./dancer.seen"
15:18:13 Quit robin0800 (Remote closed the connection)
15:18:17tmztIm only asking because one way to test an IP stack is usb cdc-ethernet driver
15:19:34Unhelpful"This page served by RockWEB" :D
15:19:39gevaertsyou don't need wifi for that
15:19:57tmztyes, but it wouldn't be a practical result for the effort without it
15:20:07tmztto add an IP stack to rockbox I mean
15:20:51gevaertsWhat's the relation between a device-side cdc-ethernet implementation and wifi anyway?
15:21:33tmztwhat I just said. why have an ip stack if you don't have a target that needs it to implement an integral hardware feature
15:21:40tmztiPod Touch, Zune, etc.
15:22:05gevaertsdo they use device-side USB for wifi?
15:23:16mcuelenaerehmm usb_hid_send_consumer_usage() seems to be called but usb_hid_try_send_drv() not..
15:24:40Unhelpfultmzt: you assume that a USB stack, and cdc-ethernet, would not be useful. you could run them both in a plugin, if they're not deemed useful in core, and serve files while playback continues.
15:25:47 Join arohtar [0] (
15:26:18 Quit faemir (Read error: 104 (Connection reset by peer))
15:26:42tmztit's more of a question of the effort required. I understand why rockbox is not linux (kernel) based, but I also know that a lot of effort has gone into existing stacks and have no idea how to integrate one into a small os like rockbox
15:27:16tmztI mean I don't know what the effort required would be
15:28:03gevaertsit should be doable, with stacks like lwip
15:28:10tmztI also think they would be useful, for a podcast client, sending files to rockbox without using storage, etc.
15:28:13 Join moredhel [0] (n=faemir@
15:29:02mcuelenaerehmm somehow USB HID thinks it's not active while it is
15:29:24 Join robin0800 [0] (
15:30:16gevaertsmcuelenaere: check usb.c, and make sure usb_core_enable_driver() always enables HID. The debug entry basically doesn't work if usb.c handles enabling/disabling
15:31:31mcuelenaeregevaerts: what I said is related to the USB HID driver itself (the active variable), not usb.C
15:31:51 Quit arohtar (Read error: 54 (Connection reset by peer))
15:32:06 Join faemir [0] (
15:33:01 Join efyx_ [0] (
15:33:06gevaertsmcuelenaere: ignore me. You get a proper config descriptor, so the enabling part is indeed fine
15:33:49tmztgevaerts: I can't see where to enable these USB_ defines
15:34:34mcuelenaeretmzt: usb_core.h
15:35:24tmztdoes serial enable a console of some type?
15:35:29mcuelenaeregevaerts: if I add logf("USBHID: %d %d", active, length); on line 355 of usb_hid.c it always says active is 0, even though it should've only set it to true and never false..
15:35:41mcuelenaeretmzt: it forwards logf() calls
15:40:08 Join mt [0] (n=MTee@
15:45:16CIA-38New commit by gevaerts (r20976): Add HID related bits to the debug menu (work by Tomer Shalev)
15:46:26mcuelenaeregevaerts: shouldn't those HID commands be in a separate menu
15:46:53gevaertsmcuelenaere: they're not going to stay forever
15:47:15gevaertsbut as long as the apps/screens.c bit isn't committed, this is the only way to actually test anything
15:50:52 Quit Lear ("ChatZilla 0.9.84 [Firefox 3.5b5pre/20090515045515]")
15:51:17 Quit gevaerts (Nick collision from services.)
15:51:26 Join gevaerts [0] (
15:55:32gevaertsmcuelenaere: do the various init functions get called correctly?
15:55:57 Quit moredhel (Read error: 110 (Connection timed out))
15:59:57mcuelenaeregevaerts: yes they do, that's why it looks so weird to me
16:00:15mcuelenaereI even commented the singe active=false; statement out and it still has value 0
16:00:36mcuelenaereI commented the memset() calls out, but that gave me exceptions
16:09:35 Quit TheDJACR ("ZNC -")
16:09:37 Join itcheg [0] (i=62db4767@gateway/web/ajax/
16:12:00 Join Horschti [0] (n=Horscht@xbmc/user/horscht)
16:13:18 Quit tvelocity (Read error: 110 (Connection timed out))
16:13:53 Join tvelocity [0] (
16:14:37gevaertsmcuelenaere: maybe remove all active checking to see if that changes anything
16:14:43 Join TheDJACR [0] (n=TDJACR@Wikipedia/Thedjatclubrock)
16:15:02 Join __lifeless [0] (n=lifeless@
16:15:29mcuelenaereyes that fixes it, but it doesn't make interrupt endpoints work apparantly
16:15:37mcuelenaere(and it isn't a real 'fix')
16:15:54gevaertsindeed not, but it may help find more hints
16:16:35mcuelenaerebtw there's only one check, and that's this one in usb_hid_queue()
16:19:07 Quit Horscht (Read error: 60 (Operation timed out))
16:20:18 Join Lss [0] (
16:24:28 Quit petur ("reboot")
16:26:26 Quit stoffel ("leaving")
16:27:30 Quit _lifeless (Read error: 110 (Connection timed out))
16:29:53 Join petur [0] (
16:33:52 Quit petur (Remote closed the connection)
16:34:28 Join petur [0] (
16:41:44 Join CaptainKwel [0] (
16:42:23 Quit Thundercloud (Remote closed the connection)
16:50:23CIA-38New commit by bertrik (r20977): Fix duplicate #includes
16:51:51mcuelenaerehmm usbmon seems to indicate interrupt endpoints work
16:52:01bertrikI was just trying to do a test-build of the archos recorder v2 and I get a "region PLUGIN_RAM is full" during link of text_editor
16:52:56 Quit itcheg (" ajax IRC Client")
16:54:57bertrikhm, now I don't anymore, never mind
17:03:48 Quit timc (Read error: 60 (Operation timed out))
17:07:09 Join fdinel [0] (
17:10:25mcuelenaeregevaerts: what FIFO size does the interrupt endpoint is supposed to have? (64bytes?)
17:15:24dfkti think i found a bug with WPS tags -
17:15:45gevaertsmcuelenaere: it apparently sets a maxpacketsize of 8 bytes
17:15:55Lssquestion. will the 240gig ipod harddisk be compatible with rockbox?
17:17:13 Join Thundercloud [0] (
17:18:07***Saving seen data "./dancer.seen"
17:19:09mcuelenaeregevaerts: hmm how is the usb driver supposed to know this?
17:20:35gevaertshm, actually usb-drv-arc.c doesn't handle this properly either. It tells the hardware to use 512
17:21:09 Join timc [0] (n=aoeu@
17:22:27mcuelenaereaccording to the datasheet: The value written to bits 10:0 (multiplied by m in the case of high-bandwidth Isochronous transfers) must match the value given in the wMaxPacketSize field of the Standard Endpoint Descriptor for the associated endpoint (see USB Specification Revision 2.0, Chapter 9). A mismatch could cause unexpected results.
17:24:01gevaertsI guess we have to extend usb_drv_request_endpoint() to pass the required endpoint size. That means of course that the class drivers need to know the connection speed...
17:24:38mcuelenaereisn't that just a usb_drv_port_speed() call?
17:25:03 Join z35 [0] (
17:25:45gevaertsit should be just that, yes.
17:26:50mcuelenaerehmm it seems to work even though I set the endpoint size to 64
17:28:33gevaertsit probably works fine until for whatever reason the driver tries to send more
17:30:05 Join juane414 [0] (n=chatzill@
17:30:29juane414does anyone know if it is possible to have one viewport turn off when another viewport turns on?
17:31:37 Join wpyh [0] (n=william@
17:34:21MarcGuay__juane414: Have whatever turns the viewport on turn the other one off?
17:34:50 Join froggyman [0] (n=47ba0b80@gateway/web/cgi-irc/
17:35:12juane414well I'm trying to get one viewport to be visible during the first 50% of song progress and the other visible during the last 50% of the song... any ideas?
17:35:40 Quit kushalone ("Leaving. I cannot promise to be back but most likely will.")
17:35:44AlexPjust use the percentage conditional (i forget what it is as %?xx<viewporta|viewportb>
17:35:46gevaertsthat doesn't sound too hard
17:36:05juane414oh duh
17:36:12juane414that makes perfect sense thanks Alex
17:36:35juane414might have my wps available for testing later tonight :)
17:37:57juane414AlexP: thanks man appreciate it
17:38:04AlexPno problem
17:39:12juane414AlexP: one more thing, will it matter if my viewport Vda and Vdb are defined before or after that line?
17:39:41AlexPjuane414: The wiki page says "The conditional viewport display tag must come before the preload tag. "
17:39:55wpyhFS #9456 updated. Please check and see if the new patch is acceptable. I've tested on iPod Video 30GB without any side effect.
17:40:04juane414AlexP: figure that would be the case
17:40:25AlexPjuane414: If it doesn't wotk you can always just move it :)
17:41:08CIA-38New commit by mcuelenaere (r20978): Get interrupt endpoints fully working on Ingenic Jz4740 targets
17:43:11 Quit nibbler__ (Read error: 113 (No route to host))
17:44:56 Join Seed [0] (
17:59:47mcuelenaeredoes a codec handle .wav files in Rockbox or do these files get treated differently?
18:00:42 Join kugel [0] (i=kugel@rockbox/developer/kugel)
18:02:26 Join bighornram [0] (n=bighornr@
18:03:01juane414AlexP: I tried doing the conditional viewport with %px and I got an error saying "enable VP:0" know what that means?
18:03:01bighornramCan I install rockbox on a sandisk view?
18:03:40froggymanbighornram: the front page is up to date
18:04:47bighornramUnfortunately the identification info for this sansa is as poor as it's OS!! All I find for model is a sansa view. no model number etc.
18:05:02 Quit miepchen^schla ()
18:06:04juane414anyone know what "enable VP:0" means?
18:06:23 Part wpyh ("Kopete 0.12.7 :")
18:08:36CIA-38New commit by mcuelenaere (r20979): Don't define LOGF_ENABLE..
18:10:13 Join nibbler__ [0] (
18:10:43bighornramFroggyman: thanks!
18:11:30froggymanbighornram: next time try searching the wiki, or google
18:13:07bighornramfroggyman: have been for hours
18:13:23bighornrammaybe I need more whiskey
18:13:54 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
18:22:08 Quit kugel (Nick collision from services.)
18:22:11 Join kugel_ [0] (
18:22:19 Nick kugel_ is now known as kugel (
18:28:33 Quit bighornram ("Leaving")
18:32:12froggymanare there plans on finishing all of the features on the calculator?
18:36:34 Join markun [50] (n=markun@rockbox/developer/markun)
18:41:42 Join Rondom [0] (
18:42:40 Quit nibbler__ (Read error: 113 (No route to host))
18:45:44 Join saratogahome [0] (n=41becb3b@gateway/web/cgi-irc/
18:45:54 Quit faemir ("Leaving")
18:48:01 Quit Jaykay (Read error: 110 (Connection timed out))
18:48:13n1sfroggyman noone has touched it for a long time so i wouldn't guess so
18:48:33 Join renke [0] (
18:49:43 Join nibbler__ [0] (
18:51:06 Quit Thundercloud (Remote closed the connection)
18:52:54 Quit Galois ("Leaving")
18:53:44 Quit n1s (Read error: 60 (Operation timed out))
18:57:46 Join krazykit [0] (
19:05:48 Quit krazykit ("Connection reset by beer")
19:06:10 Join krazykit [0] (
19:06:24 Quit robin0800 (Read error: 110 (Connection timed out))
19:08:02 Join KBH [0] (
19:15:10AlexPjuane414: Do you want to pastebin your code?
19:15:32juane414AlexP: actually I got it figured out
19:15:56 Join Galois [0] (
19:16:03AlexPcool, just wrong order I suspect
19:16:50juane414AlexP: I'm trying to get a dial to turn one click everytime the song runs another 7%... but I'm having trouble trying to figure out the needle for 50%...
19:17:27juane414AlexP: I have a viewport for the images with the needle on the left of the dial and another viewport for the right side... can't figure out what to do for the middle %50 spot
19:18:09***Saving seen data "./dancer.seen"
19:18:18AlexPWhy do you need to split them?
19:19:05AlexPWhy not just have the %?px conditional display a different image for the needle for each step?
19:19:22AlexPThe dial would be part of the background and always there
19:19:51juane414that's exactly what I have
19:20:20juane414i split it up into 2 viewports to save image buffer
19:20:31AlexPI don't think that makes any difference
19:20:41AlexPThe image buffer is just the size it is
19:20:49AlexPAll used images have to be preloaded anyway
19:21:16gevaertsAlexP: you can still save buffer this way
19:21:24AlexPcan you?
19:21:30AlexPgevaerts: How does that work?
19:21:34juane414well if the needle is only in the top right 90% of the circle why waste the other 270%?
19:21:47juane414that's a lot of wasted pixels right?
19:22:00AlexPYou don't waste the rest if the image for the needle is only the size it needs to be
19:22:19gevaertsAlexP: that only works with nice square needles
19:22:34AlexPI'm not getting how viewports helps here
19:22:58 Join mehrab [0] (n=quassel@
19:23:00AlexPYou have to display the same needle image no matter what viewport it is in
19:23:28juane414the needle changes angle though, like on a clock, so the .bmp is different each time
19:23:34gevaertsAlexP: gevaerts/"> has a nice example of how splitting can help
19:24:15 Join Jaykay [0] (
19:24:17gevaertsand yes, I ran into buffer limits while working on that :)
19:24:26AlexPjuane414: yes, I know - I still don't get how it helps which viewport you put that bmp in
19:24:38AlexPBut give me a moment to look at gevaerts' example
19:24:58 Quit mehrab (Remote closed the connection)
19:32:10juane414isn't there a limit to how many pixels can be loaded into memory?
19:32:23gevaertsthere is
19:33:53juane414so it would make sense to have twelve 60x60 images split between two viewports rather than twelve 120x60 images with one viewport right?
19:34:51gevaertsOf course depending on the exact shape of what you're showing you could have some of them even smaller
19:34:54juane414the alternative would be to crop the needle rather than having a mapped .bmp list, but then i would have to screw with all of those coordinates...
19:35:13gevaertsbut that can get tricky
19:35:16juane414but it would save more pixels
19:36:07juane414well, i would say time consuming... not really all that tricky
19:36:13AlexPgevaerts: I see what you have done there - I think the problem is that I was picturing juane414's needle differently
19:36:34AlexPbut yes, I see what you mean now
19:36:43gevaertsAlexP: his is probably different, but I think it's the same sort of problem
19:36:58juane414my .wps is basically going to end up looking like a car dashboard
19:37:02gevaertsI also split to be able to have more than 52 effective values
19:37:35AlexPjuane414: In this case for your 50% one, you could just have one additional needle image for the inbetween case
19:38:02juane414I've considered that but which viewport would it go in?
19:38:34juane414it will throw off my balance so that one viewport with get 6 images and another 7
19:38:50juane414so the percentages will be thrown off and the needle movements won't be consistent
19:38:54AlexPjuane414: You could have another small one
19:39:03AlexPIf the viewports are conditional anyway
19:39:08juane414I didn't think about that
19:39:15 Join faemir [0] (
19:39:19juane414I could just add a 3rd viewport for that one image
19:41:31AlexPjuane414: What player is this for incidently?
19:41:40juane414ipod video
19:41:58AlexPdamn, I won't be trying it then :)
19:42:15juane414what do you have?
19:43:15juane414AlexP: you'd be more than welcome to try and port it for your player
19:43:20AlexPa few, but not that (Gigabeat S120, gigabeat F60, iriver h140, sanse e260, sansa c240, Meizu M3 and ATMT MP170)
19:43:34AlexPjuane414: I'll wait and see it first :)
19:44:48juane414AlexP: well this needle thing is going to take me ages :)
19:45:30 Quit moos (Read error: 104 (Connection reset by peer))
19:45:32juane414AlexP: If I add a third conditional viewport I would then have them split into thirds... so I would have to divide my images into three mapped lists... correct?
19:46:00 Quit Jaykay (Read error: 104 (Connection reset by peer))
19:46:19gevaertsif I understand what you're doing correctly, three could be a pain (unless of course you can then ignore the middle bit...
19:46:25 Join Jaykay [0] (
19:47:08 Nick Horschti is now known as Horscht (n=Horscht@xbmc/user/horscht)
19:47:37AlexPjuane414: The way I imagine it (but it is hard to know without seeing the images) is that you have two that cover each half, and a third that covers the middle slice (and therefore overlaps a bit of each the other. As long as they are conditional so that the overlapping ones aren't displayed at the same time, it might work
19:48:12 Join alexbobp_ [0] (
19:48:58 Quit tvelocity (Read error: 110 (Connection timed out))
19:49:23 Join tvelocity [0] (
19:49:30juane414AlexP: well I might revert back to that method, but for now I'm going to try cropping each needle image individually... I'll have to abandon the nice mapped lists and mess with lots of coordinates but I think it will work better
19:49:41AlexPgood luck :)
19:53:08 Join Thundercloud [0] (
19:53:22juane414AlexP: I think I'll need it
19:53:52 Quit kugel (Nick collision from services.)
19:53:55 Join kugel_ [0] (
19:54:03 Nick kugel_ is now known as kugel (
19:58:58 Quit alexbobp (Read error: 110 (Connection timed out))
20:00:28 Quit bmbl ("Woah!")
20:11:55 Quit tvelocity (Read error: 110 (Connection timed out))
20:12:06 Join itcheg [0] (i=62db4767@gateway/web/ajax/
20:12:24 Join tvelocity [0] (
20:13:27 Quit saratogahome ("CGI:IRC (Ping timeout)")
20:16:01 Join n1s [0] (n=n1s@rockbox/developer/n1s)
20:18:02 Join tomers [0] (
20:23:09mtIs there a way to change a pointer's value inside a function without the function having to take a pointer-to-pointer parameter ?
20:31:31juane414AlexP: got it to work... pain in the arse, gotta go to bed so I'll have to work on it more tomorrow
20:32:06kugelmt: what's bad with a pointer-to-pointer parameter?
20:32:07 Quit Seed ("cu, Andre")
20:32:28 Quit juane414 ("ChatZilla 0.9.84 [Firefox 3.0.10/2009042316]")
20:33:38mtkugel : nothing, it just looks a bit ugly to me. :)
20:35:27 Quit bertrik (Remote closed the connection)
20:42:51 Join salty-horse [0] (
20:43:24salty-horsehey. I managed to crash pictureflow (and freeze the machine with "data error"). how can I debug it? via the simulator?
20:47:42 Quit efyx_ (No route to host)
20:50:40 Join R[a]ndom [0] (
20:53:25 Join Strife89 [0] (n=michael@
20:53:34 Join efyx_ [0] (
20:54:22Strife89For some reason the Dircache setting won't stay on, even after I reset my settings. Sign of filesystem corruption?
20:58:54 Nick J-23 is now known as rielev (
20:58:57 Nick rielev is now known as J-23 (
20:59:04soapUnhelpful, regarding that album art "glitch" I mentioned the other day. Despite best efforts to reproduce I have been so far unable. (Thanks to Last.FM) I am not playing the exact same playlist in its entirety in a last-ditch effort to reproduce.
20:59:12 Quit krazykit (Remote closed the connection)
21:00:47 Nick J-23 is now known as DeSnajpa_VIII (
21:00:53 Nick DeSnajpa_VIII is now known as J-23 (
21:01:56 Join krazykit [0] (
21:05:42 Nick J-23 is now known as DeSnajpa_XI (
21:15:01 Quit Strife89 ("Off for now.")
21:15:57gevaertstomers: I think we need to add maxpacketsize to usb_drb_request_endpoint(), so the driver can set up the hardware correctly. That means that the class drivers need to get the connection speed of course, but that shouldn't be a problem
21:16:05gevaertsAny thoughts?
21:17:00 Quit mcuelenaere ("Leaving.")
21:17:25 Nick DeSnajpa_XI is now known as J-23 (
21:18:10***Saving seen data "./dancer.seen"
21:18:26 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
21:19:36 Quit Thundercloud (Remote closed the connection)
21:24:18 Nick J-23 is now known as kaskader (
21:24:21 Nick kaskader is now known as J-23 (
21:25:07 Join robin0800 [0] (
21:25:28 Nick J-23 is now known as daj_mnie (
21:25:31 Nick daj_mnie is now known as J-23 (
21:27:40 Join barrywardell [0] (n=barrywar@
21:31:58 Join paulk_ [0] (
21:34:07 Quit itcheg (" ajax IRC Client")
21:34:49 Quit paulk_ (Client Quit)
21:34:56 Join itcheg [0] (i=62db4767@gateway/web/ajax/
21:39:08 Quit itcheg (Client Quit)
21:40:55 Join itcheg [0] (i=62db4767@gateway/web/ajax/
21:41:36kugelsaratoga: Why's rbutil required now?
21:42:38 Join bluebrother [0] (
21:42:57saratogakugel: i thought we agreed it was required for an AMS launch?
21:43:28kugelDidn't we talk about it a few days ago? Llorean said it was required but no-one else seemed to agree with him
21:44:01kugelAnd I don't think a "AMS launch" should be different from any other launch
21:45:04Lloreankugel: I didn't say it was required
21:45:08LloreanI said I think it SHOULD be required.
21:46:12saratogakugel: i distinctly remember you saying you didn't want to release without rbutil
21:46:33kugelI never said that
21:46:48kugelthat's not my opinion at all
21:47:11 Quit itcheg (" ajax IRC Client")
21:47:15bluebrotherwhat's the state of mkamsboot anyway?
21:48:11kugeldomonoky has a patch which makes it possible to compile it as a lib, that's all AFAIR
21:48:28kugelmkamsboot is a sane tool nevertheless
21:49:00 Join itcheg [0] (i=62db4767@gateway/web/ajax/
21:49:14domonokybluebrother, kugel: rbutil can install ams bootloader with this patch, but it needs more cleanup (especially the mkamstool part)
21:49:22bluebrotherplease define "sane". Is it user-proof similar to ipodpatcher?
21:49:56kugelI'd say so
21:50:28bluebrotherand why don't we use it for end-user installation already? I.e. provide binaries for all major OS?
21:50:46kugelit checks the OF (per md5 sum and header), checks the bootloader (so that the model fits) and checks if the inserted bootloader won't exceed our max size
21:50:56domonokybecause the ams targets are still not ready ?
21:51:07domonokyits not only rbutil support missing..
21:51:10Lloreanmkamsboot isn't as user friendly as ipodpatcher
21:51:16LloreanIt still requires several parameters, doesn't it?
21:51:46bluebrotherwell, if an ipodpatcher-like bootloader installation is available and fool proof enough we can already go and create binaries from that and distribute them.
21:51:56*domonoky thinks mkamsboot is as userfriendly as needed. more comfortable installation will be done in rbutil :-)
21:52:00bluebrotherthat way we might get more testers for the AMS ports.
21:52:03*kugel agrees
21:52:23kugelwith domonoky
21:52:26*bluebrother thinks a command line tool similar to ipodpatcher would be good. Not all users can / want to install using a GUI tool
21:52:26 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
21:52:31 Quit barrywardell ()
21:52:35 Join paulk_ [0] (
21:52:38 Join Thundercloud [0] (
21:52:41Lloreanbluebrother: It's more like the old pre-ipodpatcher install process.
21:52:50saratogakugel: you're right you said you wanted rbutil for 3.3 support, not for a release
21:52:54bluebrotherLlorean: the ipod_fw one? Urgh.
21:53:08domonokybluebrother: sure, mkamsboot will stay for non-gui people, as we do with the other targets.
21:53:16Lloreanbluebrother: It's not *quite* as bad, but you still need to have several files, invoke the program pointing at them, etc.
21:53:34bluebrotherthat's nothing I'd consider good enough for end users using a cli tool
21:53:43kugelsaratoga: yes, if anything, then I said that rbutil support for a release (as in go with 3.x Rockbox releases), but not for putting it to "supported"
21:53:49 Join Xerion [0] (
21:54:18saratogai dont' think we should provide binaries for AMS until at least funman tells us if that file system corruption problem he thought he found is fixed or not
21:54:23domonokyLlorean: it mkamsboot can not behave the same as ipodpatcher. we patch a OF and not the disc.
21:54:28domonoky- it
21:54:53bluebrotherdomonoky: well, you could make it look for an OF file. The filename is know, so it can look for it in the same folder.
21:54:54Lloreandomonoky: This is true, but we can still reduce the number of parameters to the point where we can tell users "run mkamsboot in the same folder as (OF filename)"
21:55:03LloreanSo that they could even put the two in the same folder and double click it
21:55:04*domonoky is with saragota, we first need to fix the serious problems.
21:55:11kugelif we distinguish between supported and released, then I don't care about rbutil for a release. But I think we should not require rbutil for initially supporting a target (as in proving links, binaries, installation howto's etc)
21:55:31Lloreankugel: Right now our policy is that all supported targets get releases.
21:55:35 Join {phoenix} [0] (
21:55:35LloreanSo we don't distinguish between them.
21:55:52bluebrothera release doesn't depend on support in rbutil.
21:56:03kugelOk, then I don't want rbutil to be a requirement
21:56:09Lloreanbluebrother: I feel it should going forward, for new targets (unless support is impossible)
21:56:10bluebrotherit's something we should try to accomplish, though.
21:56:35 Join krazykit` [0] (
21:56:37Lloreanbluebrother: If support is possible, there should be a strong incentive for it. I don't want future ports to be split into "user friendly" and "not user friendly" subcategories.
21:56:40domonokyyes, but for supported, we need to fix the bad bugs, and release the bootloader. rbutil and better installation and release can come later
21:56:46 Quit mt (Read error: 110 (Connection timed out))
21:57:05Lloreandomonoky: Right now "release" is the same as "supported", though. We release supported targets every three months. That's the problem.
21:57:20kugelwho defines that rbutil support is possible? The rbutil devs that don't own the targets in question, or the target devs that have no idea of rbutil and its code at all?
21:57:27 Quit paulk_ ("DSOrganize IRC")
21:57:29domonokyLlorean: i can remember times, where we had supported but not released targets.. why not do this again ?
21:57:32saratogasome way to make it easier for other people to help with rbutil would be nice, maybe a wiki page explaining whats needed from a program like mkamsboot to be easily integreated into rbutil
21:57:38bluebrotherLlorean: sure. This doesn't mean that it will be done in time with a release ;-)
21:57:48Lloreandomonoky: We could do this again, yes. But we chose not to with the beginning of the 3.0 releases
21:57:48domonokykugel: its always possible :-)
21:58:06 Quit krazykit` (Client Quit)
21:58:09Lloreandomonoky: It may be impossible in a practical sense due to, for example, lack of libraries on one OS
21:58:12bluebrotherkugel: why should support be impossible?
21:58:25 Join krazykit` [0] (
21:58:37domonokyand AMS support in rbutil is nearly ready. so its no question for this targets :-)
21:58:39bluebrotherLlorean: what are you thinking about? beastpatcher?
21:58:44kugelWhy's e200r still unsupported then?
21:58:52Lloreanbluebrother: And similar.
21:58:59saratogai think its supported we just didn't do a release for it
21:59:11kugelapparently nobody can do the e200r support right know. And this will be the case for future targets too
21:59:26Lloreankugel: Possibly because certain people who worked on it stopped working as soon as it became a supported target, since rbutil wasn't a barrier for "supported" status?
21:59:28bluebrotherand which targets if I may ask?
22:00:14domonokykugel: thats not correct. you have to change a e200r to a e200 player, thats the only thing we havent automated. (not only because its hard, but because noone is interested)
22:00:35saratogai thought the problem with rbutil support for the e200r was that you had to hack up the windows drivers for it to work, hard to imagine we could automate that
22:00:58bluebrotherthe other problem with e200r is the driver issue on windows. It's still doable, but it would require a _lot_ of work
22:01:11domonokysaratoga: thats not really a problem, you can backup and install drivers without problems.. remeber what setups can do :-)
22:01:20bluebrotheryou need to install libusb on windows plus the inf for the filter driver.
22:01:45bluebrotherwhich means windows installer stuff. At least I have no idea about such things (apart from knowing that it is possible)
22:01:48domonokyyes, its hard. but not impossible
22:02:16bluebrotheranother problem is that the e200r isn't very common, and to develop an installer one kinda needs access to the player
22:02:17domonokybut the real problem is, that noone in
22:02:27domonokyis interested in it
22:03:16 Quit itcheg (" ajax IRC Client")
22:04:12saratogaok so we agree we need a final bootloader, MMU, and no disk issues for AMS support
22:04:33*kugel can wholeheartedly agree with that
22:04:36saratogawe could do the bootloader now, but it might change for either the MMU or disk issue
22:04:47domonokyjup, everything else can come afterwards.
22:05:01kugelthere's little point in pushing the bootloader if the other requirements aren't met
22:05:21saratogayeah its so well tested at this point it'll just be a formality
22:06:23bluebrotherthere's one point in a release: if you direct interested users to using exactly that bootloader you definitely know it isn't broken by a later change.
22:06:56 Join bertrik [0] (
22:07:18*kugel would still like to compile the bootloader into mkamsboot, and possibly patch the OF in-place
22:08:35domonokykugel: embedding the bootloader into mkamsboot is possible (like it is done with ipod/sansapatcher) but needs to be optional (rbutil always downloads the bootloader)
22:08:58 Quit krazykit (Read error: 110 (Connection timed out))
22:09:00kugeldomonoky: sure, optional.
22:11:29 Join itcheg [0] (i=62db4767@gateway/web/ajax/
22:14:58 Join Ubuntuxer [0] (
22:17:18 Quit itcheg (" ajax IRC Client")
22:17:54 Join miepchen^schla [0] (
22:18:00 Join itcheg [0] (i=62db4767@gateway/web/ajax/
22:18:12 Quit miepchen^schla (Remote closed the connection)
22:25:21 Quit itcheg (" ajax IRC Client")
22:26:32 Quit martian67 (Connection timed out)
22:28:54 Join itcheg [0] (i=62db4767@gateway/web/ajax/
22:33:01 Quit froggyman ("CGI:IRC (EOF)")
22:35:23 Quit itcheg (" ajax IRC Client")
22:36:17 Nick alexbobp_ is now known as alexbobp (
22:39:21 Quit J-23 ("ZNC -")
22:39:42 Join J-23 [0] (
22:40:51 Quit BXCracer (Remote closed the connection)
22:41:51 Join miepchen^schla [0] (
22:49:36pixelmakugel: I have doubts re. the restart buttons on Ondio and Player: on Ondio, I *think* you will rather quit instead of restart half of the times now if you don't check for pre conditions - and I *believe* that on the Player there are some button combs which are not possible, electrically (I'm not sure and wouldn't know which)
22:50:43pixelmathe latter is not simulated but you could check the first in a sim
22:50:49 Quit bluebrother (Read error: 113 (No route to host))
22:51:06 Quit tomers ("ChatZilla 0.9.84 [Firefox 3.0.10/2009042523]")
22:54:36 Quit tvelocity (Read error: 60 (Operation timed out))
22:54:56 Join tvelocity [0] (
22:56:26 Quit {phoenix} (Remote closed the connection)
23:02:13 Quit J-23 ("ZNC -")
23:02:22 Join J-23_ [0] (
23:02:37kugelpixelma: I checked in a sim, and found it fine
23:02:51kugelquitting instead of restarting isn't really an issue imo
23:03:05kugelyou can reenter and try again :)
23:03:24pixelmahow fun...
23:03:27kugelon the ondio I found it particularly ok
23:03:36kugelpixelma: well, better than no restart at all?
23:04:11pixelmaI don't have anything against restart but you could code it so that it works reliably
23:04:43kugelwhy should you be quitting instead of restarting?
23:05:16kugelfeel free to improve..
23:06:05kugelI found the combos I used quite good in the sims
23:06:08pixelmaquit is mapped to the simple Off button, restart is now mapped to Off+Menu
23:07:54 Join Chronon [0] (
23:08:33pixelmait's not quite the same as in the sim, because the placing of the buttons is a bit different. The players' drawings give a bit of an overview.
23:08:43pixelmain the manual
23:09:00pixelmaor even the sim background
23:09:24 Join froggyman [0] (n=47ba0b80@gateway/web/cgi-irc/
23:09:31 Quit Jaykay ("ChatZilla 0.9.84 [Firefox 3.0.10/2009042316]")
23:09:56kugelI don't understand
23:10:20kugelif you think there can be better restart buttons/combos then I'm not saying anything against.
23:11:57pixelmajust a general statement that te impression of a button combo's usability can differ between sim and target because of the physical placement of buttons
23:12:31kugelthe off button is in the top, right?
23:12:41kugel(for ondio), i.e. not on the front
23:12:58kugelthat's how I saw it in the sim
23:15:56pixelmabut the off button in the sim is not on the top side of your keyboard, I guess ;)
23:17:43kugelyes, but I imagined that the combo is still doable
23:18:13***Saving seen data "./dancer.seen"
23:18:49 Quit merbanan (Read error: 110 (Connection timed out))
23:20:41pixelmadoable but maybe not as easy as in the sim (you probably have to do it two-handedly). It's actually not my main point but I don't want to discuss further without doing some "homework" on it first
23:25:45 Join itcheg [0] (i=62db4767@gateway/web/ajax/
23:31:34 Quit itcheg (" ajax IRC Client")
23:33:00 Quit petur ("Zzzzz")
23:33:32CIA-38New commit by mcuelenaere (r20980): Make it possible to exit from the WPS on the Onda VX747
23:35:16 Join itcheg [0] (i=62db4767@gateway/web/ajax/
23:35:44 Quit robin0800 (Read error: 60 (Operation timed out))
23:36:56 Join robin0800 [0] (
23:38:33 Quit killan_ ("( :: NoNameScript 4.22 :: )")
23:38:58 Quit n1s (Remote closed the connection)
23:40:37 Quit itcheg (" ajax IRC Client")
23:41:52 Join itcheg [0] (i=62db4767@gateway/web/ajax/
23:45:21 Quit Rondom (Remote closed the connection)
23:46:39mcuelenaeretomers (logs): any idea why I need to comment lines 356-357 in usb_hid.c out to get USB HID to work? (Rockbox SVN)
23:51:39 Quit itcheg (" ajax IRC Client")
23:53:23 Quit renke ("leaving")
23:54:46 Join itcheg [0] (i=62db4767@gateway/web/ajax/
23:56:03 Quit matsl ("Leaving")
23:57:09 Quit tvelocity (Read error: 110 (Connection timed out))

Previous day | Next day