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 2013-01-28

00:00:04pamauryI've played with it, really you can't go anywhere with it
00:00:11saratogaah ok
00:00:52saratogaso realistically, we'd probably want to implement mouse based gestures in the sim, and then once that works, hook it into the touchpad driver
00:00:56pamauryit's highly unreliable, you don't know what is going on
00:00:57AlexPRaptors: Everything is either done from datasheets (but we don't have those for most things), or by reverse engineering
00:01:56pamaurysaratoga: yeah, that's why I want to make touchpad something really handled by rockbox, and also make the gesture handling device independent
00:02:16pamauryas much as possible at least
00:02:28saratogaGSOC project
00:03:30pamauryappart from that the touchpad is pretty good: you get 60 reports per second, that's more than enough to get all the information you need
00:04:06RaptorsSo wait, the touchpad is actually really good but Sandisk botched their firmware?
00:04:18RaptorsThat's pretty sad for a big company
00:04:18saratogadoes USB HID work on the Fuze+?
00:04:24AlexPRaptors: Also common
00:04:33saratogasandisk always screws up their firmware
00:04:47pamauryRaptors: it's good in the sense that it gives you a lot of information, if they had spent some time on it, they surely could have done much better
00:05:05pamaurysaratoga: yes, I wanted to implement touchpad support over HID also
00:05:19saratogathat would certainly make improving it easier
00:05:37 Quit prof_wolfff (Ping timeout: 260 seconds)
00:06:03pamaurysure, I begun some work but I'm not very familiar with USB HID and touchpads in particular
00:06:32saratogaand maybe not necessary since debugging with a mouse would be possible too
00:06:42pamauryif someones feels like working on this I could try to finish this HID thing to make it easier
00:07:10pamaurybut a mouse is not that bad ^^
00:19:20 Quit ender` (Quit: I went to a general store. They wouldn't let me buy anything specifically.)
00:38:19 Join Keripo [0] (
00:47:04 Join sakax [0] (
00:58:51 Quit Guest75883 (Read error: Connection reset by peer)
01:02:47 Quit Strife89 (Ping timeout: 260 seconds)
01:21:46 Quit lebellium (Quit: ChatZilla 0.9.89 [Firefox 19.0/20130123083802])
01:23:51 Quit fs-bluebot (Read error: Operation timed out)
01:26:04 Join fs-bluebot [0] (
01:30:56 Quit pamaury (Ping timeout: 252 seconds)
01:35:40[Saint]Can some PHP whiz with the correct access amend to include a statement that makes it clear that if a theme is re-submitted with the original details that it will replace the original?
01:36:34[Saint]Theme name - Real name - email address
01:37:09[Saint]If those values are the same as the original upload, if will replace the original (apparently this isn't terribly widely known)
01:38:09[Saint]Oh, should also match too, iirc.
01:38:40funmansaratoga: pamaury: i had closed that bug because of confusing title iirc
01:38:54[Saint]closed as "fixed"
01:40:06funmanmaybe one aspect of it was fixed
01:41:05funmane.g. when g#235 was superseded
01:41:08fs-bluebotGerrit review #235 at : FS #12639: remove skinparser_set_buffer by Rafaël Carré (changes/35/235/1)
01:41:39[Saint]I don't blame you, it is a nightmare of a bug. Both yourself and wodz have spent some time looking into this mess and turned up other bugs...but, apparently, not *the* bug.
01:41:51[Saint]It reappeared for another user today.
01:46:38 Quit sakax (Ping timeout: 245 seconds)
01:48:36JdGordonsomeone should just rewrite the damn thing!
01:50:11*[Saint] designates JdGordon as Chief Damn Thing Rewriter
01:50:23[Saint]....nowwwwwwwwwww, GO!
01:53:07funmanor we need heap corruption detection
01:53:26funmandid someone test these themes in the simulator?
01:54:04[Saint]I recall testing lebelliums's theme in the sim without issue.
01:54:22[Saint]One cannot tell if USB gets hosed in the sim, though.
01:56:38funmanrunning in valgrind
01:56:44funmanthat's how i made the first fix
01:57:27funmanyou need to use sdl-threads though iirc
01:57:42***Saving seen data "./dancer.seen"
01:59:40 Join sakax [0] (
02:40:26 Join krabador [0] (
02:42:47funmanJdGordon: FS #12639 still shows the exact same backtrace
02:42:48fs-bluebot Certain themes cause the WPS to not load properly and USB to not work (bugs, unconfirmed)
02:59:17JdGordonfunman: can you try with g#382 please?
02:59:18fs-bluebotGerrit review #382 at : Use the skin engine to draw the usb screen via the SBS by Jonathan Gordon (changes/82/382/2)
03:07:41 Quit sakax (Remote host closed the connection)
03:32:11*[Saint] doesn't see how that would fix the problem
03:32:21[Saint]or, is it not supposed to?
03:32:41[Saint]JdGordon: ^
03:33:04JdGordonit may...]
03:33:23[Saint]Only one aspect thereof.
03:33:47[Saint]There's still the "theme fails to apply correctly until reboot" part of it.
03:34:14JdGordondoes his theme use fonts?
03:35:54 Quit krabador (Remote host closed the connection)
03:36:15[Saint]It uses userfont from the looks.
03:36:37[Saint]Yeah, it uses userfont.
03:57:46***Saving seen data "./dancer.seen"
04:19:04 Quit saratoga (Quit: Page closed)
04:21:04 Join amiconn_ [0] (amiconn@rockbox/developer/amiconn)
04:21:05 Quit pixelma (Disconnected by services)
04:21:05 Quit amiconn (Disconnected by services)
04:21:05 Join pixelma_ [0] (pixelma@rockbox/staff/pixelma)
04:21:07 Nick amiconn_ is now known as amiconn (amiconn@rockbox/developer/amiconn)
04:21:07 Nick pixelma_ is now known as pixelma (pixelma@rockbox/staff/pixelma)
04:35:18 Join TheSphinX^ [0] (
04:38:37 Quit TheSphinX_ (Ping timeout: 252 seconds)
04:54:27 Quit thegeek_ (Read error: Connection reset by peer)
04:54:39 Join thegeek [0] (
05:13:51 Quit TheSeven (Disconnected by services)
05:13:59 Join [7] [0] (~quassel@rockbox/developer/TheSeven)
05:15:42 Quit Keripo (Read error: Connection reset by peer)
05:32:24 Join TheSphinX_ [0] (
05:34:50 Quit TheSphinX^ (Ping timeout: 255 seconds)
05:43:56 Join TheSphinX^ [0] (
05:46:44 Quit TheSphinX_ (Ping timeout: 240 seconds)
05:57:47***Saving seen data "./dancer.seen"
06:25:27 Join foolsh [0] (
06:31:16 Part foolsh ("Leaving")
06:32:22 Join foolsh [0] (
06:37:31 Join TheSphinX_ [0] (
06:40:51 Quit TheSphinX^ (Ping timeout: 245 seconds)
06:46:29 Join TheSphinX^ [0] (~briehl@
06:46:47 Quit TheSphinX_ (Read error: Operation timed out)
07:01:42JdGordon[Saint]: whats the RAM delta with the extra eq bands?
07:09:03 Join TheSphinX_ [0] (
07:11:00 Join TheSphin- [0] (
07:12:37 Quit TheSphinX^ (Ping timeout: 245 seconds)
07:13:40 Quit TheSphinX_ (Ping timeout: 248 seconds)
07:13:46 Quit Tuco1 (Read error: Connection reset by peer)
07:14:30 Join TheSphinX^ [0] (
07:17:36 Quit TheSphin- (Ping timeout: 264 seconds)
07:21:52 Quit Rower (Quit: Hmmm...)
07:32:11 Join |akaWolf| [0] (~akaWolf@
07:35:12 Join stoffel [0] (
07:35:19[Saint]JdGordon: NFI, how would one compare the two?
07:36:46JdGordondo a build, save the boot file, then do utils/analysis/
07:37:23JdGordonif it adds alot of bloat then the extra bands should be compile time confiurable
07:39:40[Saint]does just take the boot file as its argument...or?
07:41:43JdGordontwo boot files
07:41:50JdGordonor maybe the elfs, i cant remmber :p
07:48:34[Saint]if my local checkout is behind master, how do I diff against the current revision?
07:49:07[Saint]git diff HEAD^ > diff.diff?
07:49:30[Saint], that wouldn't be it.
07:52:00 Quit TheSphinX^ (Read error: Operation timed out)
07:52:30 Quit stoffel (Remote host closed the connection)
07:54:08kugel[Saint]: rockbox-info.txt shows both binsize and ram usage
07:54:18kugelno need for bloat-o-meter
07:57:51***Saving seen data "./dancer.seen"
08:00:02 Join TheSphinX^ [0] (
08:00:34[Saint]I wanted 'git diff <current_branch> origin/master
08:01:34[Saint]Another git-esque/gerrity kinda question:
08:02:36 Join melmothX [0] (~melmoth@unaffiliated/melmothx)
08:03:02[Saint]I haven't done so, but, if in theory...if I wiped out my local repo/branch/whatever, or nuked the last change somehow or accidentally committed so that 'git -amend' no longer works for adding to a gerrit task, is there a way I can pick that task back up again?
08:04:08[Saint]If I added the change id manually to the commit description, would that "Just Work" assuming the author details were the same?
08:04:36kugelyes. I believe the author information need not be the same
08:04:54[Saint]Ah, so I can even pick up others abandoned tasks that way?
08:05:02kugeliiuc the change-id line is the only significant one for gerrit
08:05:20[Saint]Good to know.
08:05:21kugeli don't know, worth a try in the sandbox
08:05:42[Saint]Is that still up? Hum. I didn't know.
08:06:26kugelI guess it'll be open forever since it doesnt harm and there will always be gerrit newcomers
08:07:23*[Saint] thinks it is a shame that so many contributers have still not adopted the git/gerrit workflow
08:07:56[Saint](and that a few are adament that they will not do so)
08:09:02 Join TheSphin- [0] (
08:10:38 Quit TheSphinX^ (Ping timeout: 252 seconds)
08:10:59 Quit SuperBrainAK (Quit: pbly going to sleep /_\)
08:14:02 Join TheSphinX^ [0] (
08:16:48 Quit TheSphin- (Ping timeout: 248 seconds)
08:18:14 Quit DonAman (Ping timeout: 240 seconds)
08:19:46[Saint]for android, do I need to advertise 'android15' or 'android16' in -archlist?
08:19:50[Saint]wrt: build client
08:22:26[Saint];a=blob_plain;f=buildserver/;hb=HEAD still lists 'android15', but;a=commit;h=d798c72 updated to API16, so I'm unsure.
08:24:20 Join DonAman [0] (~DocHollid@unaffiliated/phifedoc)
08:26:33 Join TheSphinX_ [0] (
08:27:00 Quit TheSphinX^ (Read error: Operation timed out)
08:31:06 Join ender` [0] (
08:31:51 Quit foolsh (Quit: Leaving)
08:34:35JdGordon[Saint]: Once those magic numbers are gone i'm happy to push the patch
08:35:03JdGordonI just tihnk its a bad idea to use 10 arbitrarily without making it easier for anyone to change the number
08:35:26JdGordoni dont tink its against the realm of possibility that some HA forum guy would want a 150 band eq :)
08:35:32 Quit ender` (Ping timeout: 252 seconds)
08:37:13 Join ender` [0] (
08:37:35JdGordonespecially if the ram hit for the extra 5 is problematic on the lowmem targets
08:37:54[Saint]Feel free to take it over.
08:39:33[Saint]I don't really see how it makes it any easier, though.
08:39:41*JdGordon would also love it if the menu entries wernt done like that, but it isnt really possible any other way
08:39:57[Saint]They're still going to have to go through and set the frequency ranges iiuc.
08:40:13JdGordonisnt that all just setting?
08:40:28[Saint]You still need to set the defaults.
08:41:40 Quit ender` (Ping timeout: 245 seconds)
08:42:06kugelthe number of bands could be a runtime setting
08:43:33JdGordon[Saint]: have you got other patches in?
08:43:58[Saint]Yes, but I could just as easily not.
08:44:18JdGordoni mean on the gerrit task
08:44:30JdGordonthe bloat output looks like there are ata changes also
08:45:10JdGordonah, not rebased on master
08:45:16[Saint]errr...I don't believe I do.
08:45:55[Saint]gerrit task says no.
08:46:22JdGordon+1800 bytes, not huuge, and could be shrunk if we broke the settings i tinhk
08:46:49[Saint]nope, no additional changes.
08:46:55[Saint]you made me check, bastard :)
08:47:04[Saint]I knew I pushed that from a clean branch.
08:47:25JdGordonyou didnt rebase before pushing
08:47:31[Saint]I arsed it up several times, but the alst patch set should be kosher
08:47:46JdGordonfunction old new delta
08:47:47JdGordonstatic.__compound_literal 4864 5284 +420
08:47:47JdGordonsettings 6048 6468 +420
08:47:47DBUGEnqueued KICK JdGordon
08:47:47JdGordoneq_data 292 576 +284
08:47:47[Saint]and no, I didn't.
08:48:00JdGordonthose are the only sizeable increases
08:48:05JdGordoni.e mostly strings
08:48:06kugel[Saint]: good question, forgot about the build farm
08:48:08 Join Zagor [0] (
08:48:08 Quit Zagor (Changing host)
08:48:08 Join Zagor [242] (~bjst@rockbox/developer/Zagor)
08:48:26JdGordon[Saint]: there is no problem rebasing before pushing to gerrit... i think you're even supposed to
08:48:29[Saint]kugel: not a problem, I was just curious. If I helped catch something, I'm glad I did.
08:48:44 Join ben007 [0] (
08:49:19[Saint]JdGordon: I still had a lot of uncommited stuff in the tree at the time, I only just cleaned up house now.
08:49:35[Saint]I knew it would still merge, so I wasn;t too bothered.
08:51:21[Saint]The original patch from ~09 still merged fairly cleanly, with only a minimal offset. Apparently that area rarely changes.
08:51:27JdGordoncould we change the eq band config value format to "eq band: <number>, <cutoff>, <q>, <gain>"?
08:52:25[Saint]I'm sure that could be done. But it isn't as verbose.
08:52:50JdGordondoes it need to be?
08:53:28[Saint]I guess not, I suppose people don't usually edit them bu hand...
08:54:20JdGordonI'll have a little play, it will also make adding/removing bands easier
08:54:25kugelit would be nice to not break peoples eq setups
08:54:38JdGordonso provide a script
08:54:45JdGordonits not like its hard to translate them
08:55:18kugelhow hard would it be to be able to read the old style?
08:55:54 Quit ben007 (Ping timeout: 260 seconds)
08:56:14JdGordonwell, we could acutally support both easily, and deprecate the old style, but that probably means the change is pointless
08:56:18JdGordonit would make the code nicer
08:56:56JdGordonwe can add it so the old and new are readable but always write in the new style
08:57:05 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
08:59:34 Join ender` [0] (
09:00:20JdGordonwhich part was the "yea" too?
09:00:50kugelthe sentence just above
09:02:53JdGordondo we have sscanf or atoi?
09:03:24kugelboth, also sscanf is currently not linked into the core (no user) but that can be easily changed
09:03:24 Join einhirn [0] (
09:04:22 Join ben007 [0] (
09:06:56 Join petur [0] (~petur@rockbox/developer/petur)
09:12:00JdGordonbah, screw it...
09:24:58 Join ruskie [0] (ruskie@sourcemage/mage/ruskie)
09:27:12 Quit bertrik (Ping timeout: 244 seconds)
09:38:10 Join kevku [0] (~kevku@2a01:d0:ffff:34a::8:3)
09:44:32ZagorI got a report from someone who got a page in swedish when clicking on our paypal donation link. Does it happen for anyone else? I get an english page when I try.
09:57:55***Saving seen data "./dancer.seen"
10:09:40 Join Wardo [0] (
10:24:06 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
10:25:47 Join mortalis [0] (
10:26:29mortalisZagor: swedish for me
10:29:35pamauryZagor: english here (and I'm in portugal right now if that matters)
10:34:33 Join wo [0] (
10:35:04woswedish here (Poland)
10:35:09 Quit pamaury (Ping timeout: 256 seconds)
10:39:56kugelswedish here
10:41:00Zagorweird. I wonder what the difference is.
10:45:04 Join LinusN [0] (
10:49:52Zagorkugel: please reload the page and try again now
10:51:05wostill swedish
10:52:15wothe link is
10:53:00Zagorwo: no, it's redirecting you there.
10:53:14woI know
10:53:24 Join lebellium [0] (
10:57:50mortalisZagor: <input type="hidden" name="lc" value="US">
10:58:24Zagormortalis: yeah, that's what I just added to try and force it. doesn't work.
10:59:41mortalisZagor: i don't see this code in front page
11:00:20Zagormortalis: then you are looking at an old version
11:01:03 Join redhot [0] (~kvirc@
11:01:06mortalisi disabled cache and updated the page
11:01:10redhotHi folks!
11:01:13mortalisstill old version
11:01:30redhotHave wiki registration issue been solved?
11:04:37 Quit wo (Quit: CGI:IRC (EOF))
11:08:47 Join wodz [0] (
11:12:21Zagorredhot: added your email now. try resetting your password.
11:13:56redhotZagor: souns good, thanks :) So I'll be able to upload by battery_bench comparison FLAC vs MP3
11:15:02 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
11:15:59redhotCool 8)
11:15:59kugelZagor: still sweidish
11:16:50Zagorthe paypal "developer" site is a sick joke
11:18:16mortalisZagor: those line works for me, I added it in chrome's "firebug"
11:19:45Zagormortalis: then you have an aggressive proxy or something. the line *is* on the front page. # curl -s | grep 'name="lc"'
11:20:37mortalisZagor: found it, it's in wrong form
11:21:43mortalismay be not
11:21:57mortalisi'm looking to the button in bottom
11:22:28mortalisbutton on left side works
11:22:53kugelyea, I was trying the bottom one as well
11:23:07wodzheh me too
11:23:56Zagorok, try that now
11:24:25wodznow better
11:25:08Zagorgreat. thanks.
11:25:37Zagorwhat about now? (bottom button)
11:25:47kugelenglish now. it also looks vastly different
11:31:02mortalisbottom button redirects me to japanese paypal (, but it's in english.
11:31:10mortalisleft to us version
11:31:19 Join lorenzo92 [0] (
11:32:14pamaurysame here, it's in english but I get the /jp/ page ^^ I used the left button
11:33:19JdGordonI got the us page
11:33:43mortalisbtw, when in forums left button give me swedish page
11:33:51wodzso maybe this some form of random seed
11:33:56JdGordonthe forums one gives me se also
11:34:24JdGordonwiki gives jp/
11:34:33JdGordontime to fix some links :)
11:35:03JdGordonthemes is jp also
11:35:07JdGordonscorche|sh: fix it!
11:35:30kugelmain page (both): us; wiki: jp (in english); forum: se
11:35:49JdGordon[Saint]: do you tinhk anyone cares if the eq band settings (q,cutoff,etc) cant be on the quickscreen?
11:36:04 Quit mt (Ping timeout: 252 seconds)
11:36:17 Join mt [0] (~quassel@rockbox/developer/mt)
11:36:55lorenzo92kugel: can you give me some hints to implement touchscreen? I'm a little lost with the various things to implement...
11:37:26pamaurylorenzo92: you need to enable #define have_button_data in config/target.h
11:37:36lorenzo92yes, done
11:37:51pamaurythen your button_read_device (don't remember the exact name) has to fill its data pointer with the touch position
11:37:59kugeldid you look at the android code? it's pretty simple
11:38:07pamauryand return buttons | BUTTON_TOUCH (or whatever)
11:38:24lorenzo92ok, yes I've looked at it, I just don't understand how a "press" is detected hum
11:38:28kugelthe java layer just tells the C code the position (x, y) and state (up, down) of the last press
11:38:29JdGordonpamaury: did you get anywhere with the touchpad thing?
11:38:32pamaurylook at the zenxfi2 code too
11:38:55kugelthe C code calls some generic touchscreen functions to translate to rockbox functions
11:38:56pamauryJdGordon: no, didn't have time, got distracted by some many things
11:39:18lorenzo92ok, in fact I'm using this right now key |= touchscreen_to_pixels(someData.xLow, someData.yLow, data);
11:39:55pamauryJdGordon: I begun some work but did not finish
11:40:02lorenzo92and something is moving on the screen, indeed, but I need then to get up/down state by myself I guess, or?
11:40:16kugellorenzo92: android handles the state detection. if the R1 doesn't have this feature you should use some (small timeout) after which unpressing is assumed
11:40:40lorenzo92kugel: okay I need to figure out if the driver supplies this data, thanks ;)
11:41:26pamaurylorenzo92: it's a i2c touchscreen controller right ? it probably gives you all the information you need and more
11:42:00lorenzo92pamaury: yes indeed it is, it give the Z axis as well as width and gesture, the worst thing is that I cannot find a datasheet :(
11:42:01JdGordonwhy do we have 3 different ways of fiddling with the eq settings in the ui?
11:42:25pamaurylorenzo92: there is a linux driver right ?
11:42:37lorenzo92yes yes
11:42:44lorenzo92I'm studying it indeed
11:43:00redhotZagor: Thanks! It has sent me password, Reset has worked
11:43:03kugelyou should be able to receive events through /dev/input/X, no?
11:43:09Zagorredhot: excellent
11:43:21lorenzo92kugel: nope, as usually samsung did not use standard things
11:43:32redhotZagor: yup, I'm very glad! Thanks a lot!
11:43:35lorenzo92I need to read the device and grab its content using a struct
11:43:42lorenzo92kugel: I'm already done with this
11:43:57redhotWho now can allow me to upload FM Presets and battery_bench results? :)
11:44:10kugelI'm not sure there is a standard for touchscreens on (embedded) linux
11:45:10redhotZagor: If you recently registered here, have you asked to be added to the WikiUsersGroup yet? I kindly ask you to add me to this group :)
11:45:11lorenzo92kugel: wait! I'll seek for a generic linux driver (who knows ^^) I need just to have some hints without the need of reversing code :D
11:46:01lorenzo92kugel: yes there is (in the net I mean)
11:47:18Zagorredhot: added!
11:47:32lorenzo92hahah pretty fun, Samsung did the same code twice (fortunately a different programmer ^^)
11:47:52pamaurywhat ?
11:49:34lorenzo92pamaury: if you are asking me, simply that samsung wrote twice (and different) code for the same touchscreen controller
11:49:53pamaurythat's good for you, two references :)
11:50:15kugelwhere is the other one?
11:50:56redhotZagor: Excellent! :)
11:50:57lorenzo92kugel: in the R1 open source package, under devdrv folder
11:51:07kugelI think writing things twice or even more is not that uncommon in linux business when trying to mainline stuff, due to the rigid review process
11:51:30lorenzo92hey but looking at "inputInfo" (btw I have this information in my struct) I guess from here I can grab touch state
11:52:09lorenzo92btw would be cool to add multitouch to rockbox :D
11:52:30wodzfeel free to code
11:52:39kugelnot sure, we can't even do single touch properly :p
11:53:03lorenzo92there is always space to improve ;)
11:54:05redhotGuys, where is the correct place to put comparison for battery running time for FLAC vs MP3
11:54:49Zagorredhot: I'd say
11:55:12Zagoror, hmm, that's more about cpu performance
11:55:45redhotHm.. Probably about device running time too
11:56:01redhotIt was suprisingly to me find about ~10% difference
11:56:31Zagorredhot: which device?
11:56:42redhotSanDisk Sansa Clip+
11:56:47redhotRockBox 3.12
11:57:15redhot"SanDisk Sansa Clip+ 4GB Red "to be exact
11:57:56***Saving seen data "./dancer.seen"
11:59:04Zagorthe numbers on that page suggest flac is far more efficient than mp3
11:59:12kugel ?
11:59:35Zagorkugel: that looks like a better place. thanks.
11:59:47wodzcpu wise yes but data stream is bigger so storage access occures more often
11:59:57redhotFLAC is better, yes
11:59:59lorenzo92kugel: hum still not very clear, looking at android code I see that you send touch events only when the touch is still "down". that sounds logic, but then how telling rockbox that user has indeed done a press?
12:00:05redhotwodz: I have benchmarked them ;)
12:00:14lorenzo92kugel: or is it automatically guessed?
12:00:29kugellorenzo92: android also sends event for unpressing
12:00:59lorenzo92kugel: okay but i cannot see this hum, I mean, it simply stops giving rockbox touch events, right?
12:01:50kugellorenzo92: Java_org_rockbox_RockboxFramebuffer_touchHandler() has a down parameter
12:02:16kugelit's false (== up) when the finger leaves the screen
12:02:27lorenzo92okay that's clear
12:02:37kugelI don't know how android collects that information, though
12:02:48kugelit comes from the framework
12:03:12lorenzo92yes that's not the problem, my problems is to understand how touchscreen is managed by rb
12:03:15lorenzo92but okay
12:03:46lorenzo92I see now that you always call touchscreen_to_pixels
12:03:47kugelthe state variabile is set
12:03:51lorenzo92even if the button is UP
12:04:06lorenzo92but only send button action when the button is down
12:05:19kugelsee button_read_device(). "touch" is only or'd into the return value if the last state is STATE_DOWN
12:05:45kugelif it's STATE_UP then it's not, and button_read_device() will return BUTTON_NONE (assuming no other button is pressed)
12:06:14kugelby returning BUTTON_NONE you tell the core that the user stopped pressing the screen
12:06:42lorenzo92kugel: ok
12:07:51lorenzo92kugel: nice! something's movin ;)
12:09:46 Join dfkt [0] (dfkt@unaffiliated/dfkt)
12:10:16lorenzo92lower right corner means "back" to previous menu, right?if so it's really working (badly atm)
12:17:23lorenzo92kugel: ahh pff now I understand why, X and Y are swapped because for now I'm still sticky to portrait but I will rotate the screen
12:19:33lorenzo92nice, now's better ;)
12:24:44copperZagor: there's a compromise: lossyFLAC
12:24:56coppercodec is FLAC and they're half the size
12:25:43soapor less if you're willing to endure audible noise.
12:26:22 Join pandrew [0] (~andrew@
13:03:02coppersoap: no point there in using a quality setting that's not transparent, you might as well use a regular lossy codec
13:03:42 Join stryaponoff_ [0] (
13:05:29copperZagor: I got 32 hours with my iPod Classic and LossyFLAC
13:05:37copper(of playback time)
13:06:09 Quit petur (Quit: *plop*)
13:06:37copperCorrect me if I'm wrong, but I don't think the Classic is very sensitive to what codec is used, as far as decoding efficiency goes.
13:07:03copperer, where you guys even talking about the Classic?
13:07:49coppereh, no
13:11:50copperBut just as the Classic is sensitive to HDD use, conversely, I doubt the Clip+ is very sensitive to flash memory use.
13:18:02redhotcopper: it should be benchmarked - that will be an answer
13:18:18redhotbenchamrk same files on internal memory then on Flash
13:18:21copperyeah, I benched my Classic but no my Clip+
13:18:28copperwait, what?
13:18:39copper"on internal memory then on Flash"?
13:18:50copperthe Clip's internal memory IS flash
13:19:02redhotI've meant microSD
13:19:06redhotyou know it well
13:19:22gevaertsredhot: that's *much* less relevant than flash vs disk
13:22:35copperI wonder if there's a significant difference in power usage between fast and slow-ass microsdhc cards though?
13:22:55copperthat's stupid
13:23:38redhotgevaerts: absolutely, but copper mentioned Clip+ here. I thought it was concerning different type of storages but I see he mentioned file types
13:25:10redhotSo sorry for my note, just ignore it
13:25:26redhotBTW, is it possible to connect external HDD to Clip+?
13:25:40redhotHust for fun
13:28:05coppernot that I know of
13:28:46copperit does work with 64 GB microsdxc cards formatted as FAT32, though, from what I've read
13:28:59copperI'm tempted to buy one.
13:29:27copperMy 32 GB microsdhc card is almost full, and it's really slow
13:30:18redhotbe careful with buying these cards on ebay. last my item received was fake :(
13:30:33redhotthere's a lot of fake microSD's there.
13:31:09copperI'd buy one off Amazon
13:31:34 Quit lorenzo92 (Quit: ChatZilla 0.9.89 [Firefox 18.0.1/20130117041235])
13:31:44redhotyou're lucky, probably living in U.S.A, right?
13:31:51redhotSo no delivery problems
13:32:35CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
13:32:35*gevaerts thinks that that's a strange assumption
13:32:44copperFrance, so yeah, no delivery problems
13:33:28redhotYep! Anyway you have better delivery option then me, here, in Ukraine :D
13:33:46 Join TheSphinX^ [0] (
13:37:01copperdamn internet, it's so easy to buy stuff on impulse!
13:37:26 Quit TheSphinX_ (Ping timeout: 276 seconds)
13:38:24redhotOh, yes, an then so fun to argue with seller trying to resolv the conflict :D
13:38:50copperI actually got fully refunded for my defective Clip+, without having to argue
13:39:13copperamazon warranty
13:40:26copperI've always been very satisfied with Amazon's customer service. They've shipped my item a second time after the first one failed to reach my doors, on several occasions.
13:40:38copperThey'd rather take the hit than lose a customer, apparently.
13:40:53wodzmove to -community please
13:44:51redhotsorry :(
13:50:33redhotbtw, anyine can tell me how to calibrate battery?
13:50:36redhoton Clip+
13:54:32 Join Slowbreh [0] (
13:54:44 Quit Slowbreh (Client Quit)
13:56:46[Saint]redhot: unless you added an aftermarket battery with a higher/lower capacity, no.
13:57:28redhot[Saint]: ok, thanks
13:57:58***Saving seen data "./dancer.seen"
13:58:15[Saint][23:35:49] <JdGordon> [Saint]: do you tinhk anyone cares if the eq band settings (q,cutoff,etc) cant be on the quickscreen? <−− if the reason is essentially arbitrary to them, then, yes.
13:58:40[Saint]Users won't care if the config and internal code is cleaner if they are denied a feature they're using.
14:17:16 Join mt_ [0] (~quassel@
14:18:47 Join jhMikeS [0] (~jethead71@
14:18:47 Quit jhMikeS (Changing host)
14:18:47 Join jhMikeS [0] (~jethead71@rockbox/developer/jhMikeS)
14:19:31 Quit mt (Ping timeout: 255 seconds)
14:42:20 Join nateloaf [0] (
14:42:47 Join amayer_ [0] (
14:47:51 Quit mortalis (Quit: Leaving)
14:48:09 Quit stryaponoff_ (Quit: Ex-Chat)
14:49:08 Quit [Saint] (Quit: Quit)
14:59:56 Quit mt_ (Read error: Connection reset by peer)
15:00:35 Join mt [0] (~quassel@rockbox/developer/mt)
15:10:38 Join [Saint] [0] (~saint@rockbox/user/saint)
15:18:41 Join Rower [0] (
15:39:47funmanJdGordon: no complaint from valgrind with g#382 indeed
15:39:49fs-bluebotGerrit review #382 at : Use the skin engine to draw the usb screen via the SBS by Jonathan Gordon (changes/82/382/2)
15:41:50[Saint]It disturbs me greatly that that fixes the issue.
15:42:06[Saint]Pity it won't get committed as-is.
15:42:32[Saint]well...not really. I'll rephrase that.
15:43:35[Saint]"Pity it uses the sbs instead of its own custom extension like other screens do, which is a source of contention blocking it"
15:45:25funmani didn't say that fixes the issue
15:45:42[Saint]Ah. :)
15:45:46funmanjust that after one plug/unplug cycle in valgrind there's no read from uninitialized data
15:45:51funmanmight be 100% unrelated
15:46:01funmansince it appears only on unplug anyway
15:46:33funmanworth putting that bug out of our way, still. at least we would be certain it is not related
15:48:47 Join stryaponoff [0] (
15:57:36 Nick stryaponoff is now known as stryaponov (
15:58:01***Saving seen data "./dancer.seen"
15:58:32 Part stryaponov ("I need to go. Have a nice day!")
15:59:17 Join stryaponoff [0] (
16:07:56funmanwhat would be cool is if we could make a full RAM dump after a data abort
16:08:10 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
16:19:56wodzfunman: thats easy :-) Its only a matter of fixing postmortem stub on n2g
16:20:08funmanah the bug happens also on n2g?
16:20:36wodzdon't know I am speaking about dumping mem on data abort
16:24:17[Saint]would N2G need USB for that to be relevant? :)
16:25:19funmanwe could work out something
16:25:42funmanhum btw since this happens on fuze+ and amsv2, it has nothing to do with usb driver, right?
16:26:03[Saint]One assumes so.
16:36:15 Quit kevku (Ping timeout: 245 seconds)
16:42:17pamaurydoes it happen on plug or unplug ?
16:55:14pamauryOn the fuze+ (usb-drv-arc) On unplug I could image one way to corrupt some buffers if the usb controller is not stopped in time (perhaps there is a race condition here) and some remaining dma transfers corrupt some memory. Appart from this I see no way to corrupt some memory, the driver seems pretty robust
16:56:19 Quit Zagor (Quit: Clint excited)
17:04:13 Quit lebellium (Read error: Connection reset by peer)
17:04:46 Join lebellium [0] (
17:23:01redhotok, guys, thanks!
17:25:49 Quit redhot (Quit: KVIrc 4.1.3 Equilibrium
17:29:09 Quit [Saint] (Remote host closed the connection)
17:31:25 Join [Saint] [0] (~saint@rockbox/user/saint)
17:38:33funmanpamaury: the valgrind trace i pasted on the bug happens on unplug
17:40:39 Join n1s [0] (
17:40:39 Quit n1s (Changing host)
17:40:39 Join n1s [0] (~n1s@rockbox/developer/n1s)
17:41:19[Saint]Who wants the joyous opportunity to tell me what I'm doing wrong?
17:42:18copperIf you're eating cocaine, you're doing it wrong.
17:43:41[Saint]relevant section of ~/.bashrc (slightly convoluted, but afaik fine):
17:43:41[Saint]proof exporting the env vars "works":
17:45:01pamauryfunman: can you point me to the bug report ?
17:45:17copper[Saint]: export PATH=$PATH:$ANDROID_NDK_PATH:$ANDROID_SDK_PATH
17:45:40copperif those indeed contain the binary that you want
17:46:09lebelliumpamaury: I'm done with gathering service manuals.
17:46:13[Saint]copper: shit, I missed that paste...yes, they're in my $PATH
17:47:19[Saint]saint@saint-desktop:~/Development/Rockbox/development/rockbox/build$ echo $PATH
17:47:32[Saint]that != what I wanted to do...
17:48:41 Join pretty_function [0] (~sigBART@
17:49:52pamaurylebellium: ok, thanks for letting me know
17:50:22copper[Saint]: does this return an included path? $ which arm-linux-androideabi-cpp
17:50:37[Saint]copper: yeppers
17:50:40copperand also, maybe "echo $PATH" in ./configure just to see if it matches what you expect
17:50:42*[Saint] is stumped
17:51:11copperand add "which arm-linux-androideabi-cpp" in ./configure too while you're at it
17:51:41*[Saint] is considering picking up the binaries and dumping them in ~/bin
17:51:56[Saint]very ugly kludge, but it seems to "just work".
17:52:54[Saint]Oh, hum...configure works, compilation fails.
17:54:14 Join kevku [0] (~kevku@2001:470:27:773:0:feed:c0f:fee)
17:54:48 Join Strife89 [0] (
17:55:24copper[Saint]: maybe a prayer to Saint Stallman would help
17:55:44copperI'm agnostic, so I'm with the "you never know" camp.
17:56:20pamauryfunman: so if you got the bug in the simulator, this seems to indicate that it is not related to the usb driver
17:56:42funmanpamaury: i didn't got *the* bug
17:56:50[Saint]*Saint* Stallman?!?
17:56:54funmanpamaury: fs#12639
17:56:56fs-bluebot Certain themes cause the WPS to not load properly and USB to not work (bugs, unconfirmed)
17:56:56[Saint], that's a new apporach :)
17:57:08copperHe knows not sin.
17:57:27pamauryfunman: yeah but you got something suspicious however
17:57:49funmantrue, no idea if it's related yet
17:58:05***Saving seen data "./dancer.seen"
17:58:10pamaurythere might be two bugs of course ^^
17:58:41funmanor more ;)
17:58:58pamaurymy ultimate project would be to be able to emulate the code on the pc (in qemu) while doing all the actual hardware access on the real hardware. This would be damn slow but very useful
18:00:39funmanyeah skyeye or qemu like would be useful
18:01:59pamauryI abandonned the idea of emulating the entire imx233 in qemu or skyeye, too much work and useless, we need the actual hardware. I recall the coreboot guys did something similar on some motherboards
18:02:16 Quit Wardo (Ping timeout: 255 seconds)
18:03:41pamauryOn the imx233 I think such a scheme would be possible by a running a small stub on the device and doing transfers over usb (if usb is not needed) or over debug uart (but then I only have one device where I know where debug uart pins are and the data bandwidth is really limited)
18:12:59 Quit ben007 (Ping timeout: 260 seconds)
18:15:13 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
18:15:38wodzisnt it easier to write gdb stub then?
18:17:45pamaurywodz: that's another possibility, I never wrote any code anyway ^^
18:17:53pamaurythat might be simpler yeah
18:18:55wodztruly complicated part would be usb
18:24:18pamaurywhy ?
18:25:35wodzok, for me :P
18:26:34wodzanyway this could not debug usb problems in our code. We will need different debug channel
18:30:41 Quit wodz (Quit: Leaving)
18:31:06 Quit pamaury (Quit: this->disconnect())
18:31:52 Quit stryaponoff (Quit: I need to go. Have a nice day!)
18:33:50 Quit pandrew (Ping timeout: 244 seconds)
18:36:10 Quit [Saint] (Read error: Connection reset by peer)
18:38:18 Join [Saint] [0] (~saint@rockbox/user/saint)
18:40:20lebelliumSI4709 is RDS-compatible right?
18:41:26 Join y4n [0] (~y4n@unaffiliated/y4ndexx)
18:59:05lebelliumHum why is YP-R0 in both unstable and unusable ports?
19:00:05 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
19:01:04funmanif we take 1 for unusable, 2 for unstable and 3 for stable; that makes it stable. thus i think it's an evil plan of pamaury
19:01:39funmanor maybe it made it to unstable but website was not updated ?
19:02:25lebelliumit has been unstable for 1 year, I don't see why it would be unusable now while nothing has become worse on it
19:02:35funmanbtw, I can take saratoga's nano2g at fosdem if one of you want it
19:02:45funmani also have another one
19:04:54*pamaury didn't make any evil plan for the r0
19:08:21 Quit melmothX (Quit: #)
19:14:17 Quit pamaury (Remote host closed the connection)
19:19:03 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
19:28:44[Saint]Is anyone sucessfully compiling RaaA on Ubuntu 64 bit?
19:28:55[Saint]RaaAoA, rather
19:34:11 Part LinusN
19:36:24[Saint]this makes no sense!
19:36:50[Saint]configure is bitching about something missing that clearly isn't...
19:37:02[Saint]Any suggestions?
19:37:24*[Saint] also likes:
19:37:36[Saint]./../tools/configure: 1: ./../tools/configure: arm-linux-androideabi-ld: not found
19:37:59[Saint]Using arm-linux-androideabi-ld
19:38:17[Saint]So, you didn't find it...but, you're using it? Good to know.
19:43:09n1s[Saint]: can you run arm-linux-androideabi-cpp, like arm-linux-androideabi-cpp -v ?
19:44:12[Saint]Ohhhh...what the smurf?
19:44:23 Join zamboni [0] (~bottledwa@unaffiliated/zamboni)
19:44:26[Saint]saint@saint-desktop:~/Development/Rockbox/development/rockbox/build$ arm-linux-androideabi-cpp -v
19:44:26[Saint]bash: /home/saint/android-ndk-r8d/toolchains/arm-linux-androideabi-4.4.3/prebuilt/linux-x86/bin/arm-linux-androideabi-cpp: No such file or directory
19:45:02[Saint]but, its there, dammit! :)
19:49:52 Part amayer_
19:56:39 Quit Strife89 (Quit: Heading out.)
19:57:01 Quit pretty_function (Ping timeout: 255 seconds)
19:58:00 Quit bebna (Quit: Leaving.)
19:58:09***Saving seen data "./dancer.seen"
19:58:13 Join pretty_function [0] (~sigBART@
20:08:18 Join [Saint_] [0] (~quassel@
20:08:26 Quit pretty_function (Remote host closed the connection)
20:08:43 Nick [Saint_] is now known as Guest88639 (~quassel@
20:15:24 Quit alexbobp (Remote host closed the connection)
20:16:16 Join alexbobp [0] (
20:23:38pamaurybertrik: ping
20:24:30 Join amayer_ [0] (
20:28:15bertrikhi pamaury
20:30:05pamauryhi, could you do something for me ? I would like upload a built bootloader for the zenxfi2 and zenxfi3 but the internet connection is way too slow here, could you build and upload them ?
20:30:19pamauryby bootloader Imean the complete sb file
20:30:21amayer_is tagcache just for the database?
20:30:32pamauryamayer_: yes
20:31:30amayer_so why is it using 7%(System->Debug Menu->View OS Stack) if i dont use the database as all?
20:32:03n1samayer_: the thread still exists i guess
20:32:19[Saint]amayer_: the thread still exists if you use it or not.
20:32:24[Saint]ohhh...too late.
20:32:30bertrikpamaury: ok, I'll have a go at that
20:32:38[Saint]and, disturbingly close to n1s
20:33:14pamaurybertrik: for the zenxfi3, you know how to do it I guess, other follow the instruction from the wiki ^^
20:34:19[Saint]amayer_: it's the same as "why does "scroll" use N% if nothing is currently scrolling?"
20:34:27amayer_[Saint]: n1s: can we make threads on demand? or do they have to exists all the time?
20:34:29amayer_what im asking is this something that can be improved(thread is dead/not initalized until needed)?
20:34:31[Saint]and various other examples.
20:34:54pamaurythey are ondemand
20:35:39pamauryin theory we could create the thread only if needed
20:35:55n1samayer_: the stack for each thread is statically allocated so not creating the thread won't save any memory and just perhaps an unnoticable amount of cpu cycles in the kernel
20:35:56amayer_pamaury: wouldnt that help with battery life?
20:36:01pamaurybut it won't change anything: the stack is preallocated, same thing for the thread structure
20:36:10pamauryno, the thread is never run
20:36:37bertrikpamaury: I don't have to do any special, do I? just ../tools/configure for Boot && make, right?
20:36:55pamaurythen mkimxboot
20:37:11pamauryI would like to build the complete files
20:38:42pamauryamayer_: afair, the tagcache blocks on queue which contains tagcache events and systems events. If you don't use tagcache, there are no former events, and system events only include thing like usb plug and unplug. So I expect it to be blocked 99.9% of the time
20:39:32amayer_[Saint]:when i go under Debug->database info it says Initalized: yes. but i dont have any databases on my hard drive. it this info wrong or does it pull from somewhere else?
20:39:48n1samayer_: looks like the tagcache stack is 20kB, which is less than 0.1% of the total ram on a 32 MB target
20:40:21amayer_n1s: i think the iclassic is 64mb(but i get your point)
20:42:05pamauryamayer_: if you enabled database, I *think* it will scan the disk, even if you don't use it
20:43:26pamaurybertrik: ^ that's why I don't have the bandwidth, the sb file includes the OF which 40MB for the zenxfi2 and 20MB for the zenxfi3 iirc
20:43:27bertrikpamaury: sorry, but I don't really know how to use imxboot, or what extra files I need
20:43:59bertrikI need the original .sb files then?
20:44:08amayer_pamaury: under settings->database everything is set to off/no
20:44:25[Saint]If the db claims it is initialized, there definitely should be database (.tcd) files present on the device.
20:44:29pamauryit's all explained, so yeah you need the OF. If don't want to spend time that's no problem, I'll upload then at my work tomorrow
20:45:05pamauryamayer_: I don't know then, perhaps this is always set to true when database is off, only the code can tell
20:46:01amayer_[Saint]: there are no .tcd files on my disk(i just checked to make sure)
20:46:13[Saint]It seems its a bug.
20:46:23[Saint]I can reproduce this behaviour.
20:46:48 Quit |akaWolf| (Ping timeout: 246 seconds)
20:46:58[Saint]Initialized: Yes - after wiping out the db and rebooting
20:47:18amayer_[Saint]: while we are talking about the debug menu. do you have your iclassic around?
20:47:20amayer_what does yours say under partion data?
20:47:29[Saint]I don't, no.
20:47:37[Saint]Ms [Saint]left it at work.
20:48:15[Saint]amayer_: where are/were you going with that?
20:48:36amayer_mine just has all 0(zero) for everything
20:49:03amayer_P:0 S:0
20:49:05amayer_T:0 0MB
20:49:07amayer_P:1 S:0
20:49:09amayer_T:0 0MB
20:49:22amayer_P:3 S:0
20:49:24amayer_T:0 0MB
20:49:26amayer_P:3 S:0
20:49:28amayer_T:0 0MB
20:50:03[Saint]p0 should be non-zero...but, I'm not too surprised.
20:50:09[Saint]It is still marked unusable.
20:50:51[Saint]Other iPods would display p0 as the OF firmware partition, but on the Classic, that doesn't exist so p0 would be storage.
20:51:23amayer_well what im getting at is they all say 0Mb. is that correct?
20:51:35pamaurybertrik: warning, if you are following the zenxfi2 instruction, I just fixed an error in the commands: the mkimxboot needs "-v zenxfi2-nand" and not "-v zenxfi2-sd"
20:51:47[Saint]No. But it isn't anything to worry about. It says keep out for a reason ;)
20:52:37pamauryThe database is a complex piece of software, and its debug menu has been known to be ...confusing
20:52:44bertrikpamaury: looking for the .exe right now
20:52:55amayer_[Saint]: ok. thats just one place i havent poked around in so i figured i would peek in.
20:53:18amayer_just trying to get to know as much of the project as i can
20:53:34pamaurycool ! For the zenxfi2, mkimxboot is happy with the exe file directly, for the zenxfi3 you'll need to cabextract, but it is all written on the wiki, that should be a good test for the wiki instruction :D
20:53:39[Saint]You'll find various wrong information scattered around the debug menu.
20:54:09[Saint]it isn;t really entirely useful, and should probably be a compile-time switch.
20:56:29*pamaury finds it very useful ;)
20:56:56*[Saint] forgot to add "for end users"
20:57:29[Saint]s/for end users/for end users - after a port is stable/
20:59:37amayer_[Saint]: i roll my own(with some test code)
21:00:00[Saint]and, have you added anything to the debug screen? :)
21:00:45[Saint]My point with the Classic is, you shouldn;t expect any of it to work. There's no officially supported way to install, it technically doesn't exist.
21:01:09[Saint]We could (rightly) deny support to all Classic users.
21:01:20[Saint]We don't. But we could.
21:03:26amayer_i get what your saying. and im not complaining about the debug menu. i just wanted to know if i didnt understand it(probably dont) or if it isnt behaving how it is intended to
21:03:51[Saint]It isn't behaving as intended, but, this isn;t entirely surprising.
21:04:16[Saint]For young ports, developers tend to get the main features working and then clean up the rest afterwards.
21:04:54pamaurythe debug menu has always been confusing for the database, like the 110% for progress and so on
21:05:11[Saint]we're onto the partition menu now ;)
21:05:17[Saint]keep up man!
21:05:36pamaurysorry ;) the partition menu is also a nice unreadable one
21:05:45[Saint]It is, yes.
21:06:25n1smost of it is just quickly hacked up for whatever was needed and sometimes improved a bit, like the buffering thread screen
21:07:45bertrikpamaury: is what I get, no success
21:08:20bertrikcabextract -d temp ZEN_X-Fi2_PCFW_L22_1_23_01.exe claims "no valid cabinets found"
21:08:43pamauryfor the xfi2 you don't need to cabextract
21:08:47bertrikand file says ZEN_X-Fi2_PCFW_L22_1_23_01.exe: PE32 executable (GUI) Intel 80386, for MS Windows
21:08:49pamaurybecause there is no cab in it ^^
21:09:16pamauryhowever I'm surprised by the mkximboot message
21:09:38pamaurymake sure your are at git HEAD and recompile dualboot and mkimxboot
21:10:56pamauryoh wait, you have the non-european installer, I didn't write the offsets in mkimxboot !!
21:10:59 Join sakax [0] (
21:11:31bertrikI got a .zip from some chinese website, couldn't actually understand anything but the .zip filename
21:12:09pamauryyou have two options: either get the european version or find the sb files in the installer, there are in plain binary form
21:12:15pamaurythere should be three of them
21:12:35pamauryI'm really asking you too much :(
21:13:00pamaurybluebrother: what ?
21:13:04 Join lebellium_ [0] (
21:13:06bluebrotherQNetworkDiskCache seems to be fragile :o
21:13:34bluebrothernew HttpGet implementation works. Trying with Qt5. Back to Qt4. Crashes on startup. Remove cache folder, works again.
21:13:58*bluebrother did expect something more from such a class :/
21:14:30 Quit lebellium (Ping timeout: 244 seconds)
21:14:43 Nick lebellium_ is now known as lebellium (
21:15:52bluebrotherI was hoping to get things stable and Qt5 compatible ...
21:16:02bluebrotherI guess I have to do something about this caching thing then :(
21:18:07bluebrotherpamaury: btw, any news on the libmspack issue? Should this go into mkimxboot or rather Rockbox Utility?
21:18:58pamauryI was wondering, perhaps it's better in Rockbox Utility: the installer class would unpack it and give the file to mkimxboot
21:19:07pamauryI'll try to make that work tonight
21:19:10bluebrotheryeah, would do
21:19:24bluebrotherok. Otherwise I can try to give it a look the next couple of days
21:19:50pamaurylet's see, if I fail tonight, you'll do the job !
21:20:07pamauryI know nearly nothing about rbutil but I'll learn
21:20:47bluebrotheron a first thought it might be a good idea to have that in BootloaderInstallBase::setOfFile()
21:21:05bluebrotherit already handles zip files, so handling cab doesn't sound like a bad idea :)
21:21:30bluebrother(and put as filename in rbutil.ini so it tries to find that file)
21:22:25pamauryhow does that work ? how does RbUtil knows when to unpack files ?
21:22:34 Quit lebellium (Read error: Connection reset by peer)
21:22:40pamauryfor zip files that's the extension perhaps but here it's an .exe file
21:23:03 Join lebellium [0] (
21:23:14bluebrotherno, for zip file we simply try to open the file as zip and check the result :)
21:23:36pamauryah ok, do you have a reference bootloader installer which does that so I can look and copy ?
21:24:05bluebrotherthis function prepares the bootloader
21:24:38bluebrothersince we detect zip files in this function pretty much every bootloader that needs an OF file will do
21:25:12 Join SuperBrainAK [0] (
21:25:37pamauryOk, i'll have a look and cry for help
21:26:00bluebrotherin rbutil.ini bootloaderfile names the filename. setOfFile() checks if the file provided by the user is a zip file and includes that file. If it does it extracts the named file and passes that along
21:26:56bluebrotherso you should be able to simply extend the if( with an else if(some_check_for_cab()) { ... }
21:30:43 Quit y4n (Quit: only amiga makes it possible)
21:34:55bertrikpamaury: no luck yet, downloaded another file now from
21:35:38pamaurythen, could you try to find the correct offsets into the file ? Otherwise just drop it
21:35:56funman[Saint]: you probably didn't install ia32-liba
21:36:17[Saint]which is?
21:36:23funman[Saint]: ia32-libs* (gcc is a 32bit binary, and "file not found" means "32 bits libc not found"
21:36:47funmantry ldd `which arm-...-ld`
21:37:35[Saint]I wiped the setup out to start clean in case I messed up.
21:37:47[Saint]I shall install said package and try again.
21:38:29[Saint]holy CRAP! That is a LOT of dependencies/additional packages.
21:39:00[Saint]277 new packages
21:39:18funmanmaybe you can install a few les
21:39:28funmanin any case the list of needed packages is in ndk doc
21:40:09funmanhmm or 'was'
21:41:11funman -> see 'Ubuntu'
21:41:43funmanthey moved it far away from the download page :/
21:42:06pamaurybertrik: ok, by chance I should have the same file on my hard-drive, I'll find the offsets and commit then
21:42:37[Saint]yep, I'll be ignoring /that/ one.
21:44:12[Saint]funman: thank you so much, I looked for something similar, but I did not expect to find it located in "setting up an IDE". The fact that an IDE is required, or they should think so, is hilarious.
21:44:31[Saint]This quite literally drove me nuts for about 2 hours.
21:46:30funmani don't know if openjdk works
21:46:31[Saint] is /quite/ out of date.
21:46:40[Saint]funman: It does, yes.
21:56:33 Join ben007 [0] (
21:57:18 Join olspookishmagus [0] (~pookie@
21:58:08olspookishmagushello... the 80GB HDD limitation of a 5th generation iPod (iPod Video) is a hardware or a software limitation?
21:58:12***Saving seen data "./dancer.seen"
21:58:29olspookishmagusor let me put this better, wouck rockbox support a HDD larger than 80GB for that device?
21:58:43[Saint]neither, it just happens to be the size of the HDD.
21:58:53[Saint]Aha, right, yes...yes it would.
21:59:21olspookishmagusfor example:
21:59:40olspookishmagusok cool, the it's Rockbox with an 240GB HDD
22:00:01[Saint]you're looking for a MK2431GAH
22:00:28olspookishmagusone more... does Rockbox support losslell audio formats?
22:00:41[Saint]lossless? Yes.
22:00:49olspookishmagusyeah, as in: .flac/.ape?
22:00:55[Saint]Flac, ALAC, et al
22:01:22AlexPwavpack, wav, ape, flac, alac
22:01:25olspookishmagusI can't recall with I removed rockbox sometime after putting it in my iPod some time ago
22:01:30AlexPprobably others I can't remember
22:01:43olspookishmagusyou got me covered
22:02:00[Saint]a MK2431GAH should be pretty easy to find.
22:02:15[Saint]It is a 240GB drive in the same form factor.
22:02:56[Saint]there 'ya go :)
22:03:29[Saint]holy hell that's expensive, though...
22:03:29pamaurybertrik: I'm committed the offsets, can you retry on that file please ?
22:03:46olspookishmagus[Saint]: true, check that eBay link I gave earlier ;D
22:03:57[Saint]Keep looking, you'll almost certainly find one cheaper.
22:04:22olspookishmagusalready did, now I'll have to prioritize this month's expenses
22:05:09olspookishmagusand what's the suggestion for managing the files? gtkpod?
22:05:28[Saint]drag and drop.
22:06:14[Saint]Unless you have other devices that rely on the iPod Database, in which case, yes...gtkpod.
22:06:39olspookishmagussounds like, rsync might do the job as well
22:06:55[Saint]Rockbox doesn't need any fancy syncing, drag & drop is fine for it.
22:07:18[Saint]It can, however, play audio that has already been synced to the iPod via iTunes et al.
22:07:42[Saint]But only via the Database, not via the File Browser
22:08:02[Saint]If you drag & drop, both the database and the file browser will "just work".
22:08:05 Quit XavierGr (Ping timeout: 252 seconds)
22:08:14olspookishmagusas I recall, if you do something *NOT* within iTunes you render it's db useless and then it asks for... re-format
22:09:42[Saint]the iPod OF will let you treat the device as removable storage.
22:09:52[Saint] just won't be able to play any media you put there that way.
22:10:34olspookishmaguswould you please define: 'OF'?
22:10:43[Saint]Original Firmware
22:11:18[Saint]ie. Apple's firmware.
22:11:24olspookishmagusrockbox on the other hand, will let you play anything
22:11:31olspookishmagusgot that too
22:12:00olspookishmagusif you put a .flac along with it's .cue would you be able to select individual tracks?
22:12:16olspookishmagusok now [Saint] one final question
22:12:22olspookishmagushow can I buy you a beer? ^^
22:12:23[Saint]we have .cue support (disabled by default)
22:13:49[Saint]that monster, is the project's paypay donation link.
22:13:56 Join psilonux [0] (~psilonux@2001:470:1f15:10f6:d95e:dad8:d40f:ed92)
22:16:32[Saint](ironically often referred to as The Beer Fund)
22:17:13 Join TheSphinX_ [0] (
22:17:39 Quit Guest88639 (Remote host closed the connection)
22:17:41olspookishmagus[Saint]: don't go posting your SESSION IDs
22:18:43[Saint] auto-shorten-long-links widget didn't work there. :-S
22:19:16[Saint]There are two donate links on the main page (both go to the same project) at
22:19:23olspookishmagusOK, got it
22:19:42olspookishmagusI'd hit the right one
22:20:19bertrikpamaury: ok, it runs now
22:20:21olspookishmagusfine, when I'm bying that HDD I will donate you some
22:20:28pamaurybertrik: cool :)
22:20:38bertrikpamaury: now I upload the, right?
22:20:39 Quit TheSphinX^ (Ping timeout: 252 seconds)
22:20:41[Saint]olspookishmagus: don't feel obligated to, but, thanks.
22:20:52[Saint]It will surely be appreciated.
22:20:54pamaurybertrik: just wait a minute, I'll check on mine, we never know ^^
22:24:08pamaurybertrik: you gave the "-v zenxfi2-nand" option to mkimxboot right ?
22:24:15pamauryok it works on mine, go ahead :)
22:25:13olspookishmagusis there a SMART utility for rockbox?
22:25:29pamauryit might be that some guy with a sb-based zenxfi2 use it and it doesn't work, but I don't know if such devices exist
22:25:43 Join pandrew [0] (~andrew@
22:25:52dfktolspookishmagus, SMART doesn't work over USB
22:26:32bertrikpamaury: the md5sum on mine is fc00211a4854b6258d7002aa237008db
22:26:58olspookishmagusdfkt: the HDD is connected via a USB controller?
22:27:13AlexPolspookishmagus: There isn't a SMART plugin, no
22:27:16pamaurybertrik: we can't compare the sums, the encryption key is randomly generated!
22:27:23dfktolspookishmagus, sorry, had a brainfart... internally it might work, but there's no plugin
22:27:27bertrikoh ok
22:27:33olspookishmagusdfkt: a brainfart?
22:27:38AlexPThere has been some talk by some poeple at some point, but nobody has done it as of yet
22:27:46*olspookishmagus opens the dfkt's brain-windows
22:27:59dfktolspookishmagus, i was thinking about reading out SMART data over USB on the computer... :)
22:28:12olspookishmagusdfkt, AlexP: thanks both of you
22:28:21AlexPolspookishmagus: g#385
22:28:23fs-bluebotGerrit review #385 at : iPod Classic/6G: reads HDD S.M.A.R.T. data by Cástor Muñoz (changes/85/385/1)
22:28:40AlexPNothing in the repos as of yet though
22:37:01pamaurybluebrother: did you investiaget which files in libmspack were really needed or not ?
22:37:35pamauryshould I include the entire library or just the necessary part ?
22:39:47 Quit psilonux (Quit: Leaving)
22:43:03 Join lebellium_ [0] (
22:43:19 Join prof_wolfff [0] (
22:44:42 Quit lebellium (Ping timeout: 240 seconds)
22:44:43 Nick lebellium_ is now known as lebellium (
22:48:13pamaurybluebrother: do I need to abstract the CAB with a ZipUtil like class ?
22:54:34 Quit ben007 (Ping timeout: 256 seconds)
22:59:17pamaurybluebrother: what happen in the (unlikely) case of, say, an installer which contains an CAB but in which the actual firmware is not in the CAB ?!
22:59:40pamaurysetOfFile will see it's a cab, extract it and fail because the file is not in it
23:00:08 Quit sakax (Remote host closed the connection)
23:00:51 Part amayer_
23:07:33 Join ben007 [0] (
23:19:26olspookishmagusis there an iPod back case/cover that features the rockbox logo? ^^
23:19:26 Quit ender` (Quit: Never put off til tomorow what you can skip altogether.)
23:21:29 Quit n1s (Quit: Ex-Chat)
23:36:53olspookishmagusdoes anyone recognise this skin/theme?
23:37:48dfktsince it's in the manual, i guess it is plain cabbie v2?
23:42:43olspookishmaguslooks like it, only that the font's thinner
23:44:26 Quit ben007 (Ping timeout: 256 seconds)
23:44:28dfktyou can always set the font independently of the theme :)
23:45:18olspookishmaguscan you also hot-swap themes?
23:45:40olspookishmagusand are there any UTF enabled fonts?
23:46:15gevaertsAll fonts do unicode
23:46:16dfktsure, no need to reboot... you can also mix&match different parts of different themes
23:46:57dfktall included ones, or the ones from the fontpack too?
23:48:46gevaertsThey don't all have an extremely wide coverage, but they all do unicode :)
23:48:54dfktaah :)
23:49:17*dfkt only has one song that uses runes in the title
23:53:02 Quit nateloaf (Quit: Leaving.)
23:55:09 Quit dfkt (Quit: -= SysReset 2.55=- Sic gorgiamus allos subjectatos nunc.)
23:57:37 Join ben007 [0] (
23:58:16***Saving seen data "./dancer.seen"

Previous day | Next day