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

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

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

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

#rockbox log for 2010-03-03

00:00:58JdGordon_gevaerts: when you built the sim for your phone, did you just use regular sdl?
00:01:07JdGordon_(regular being the one cross compiled for it)
00:01:18 Quit dfkt (Ping timeout: 245 seconds)
00:03:21JdGordon_I'm trying to find out what needs to be done to get it to use sdl's fbdev driver
00:03:47JdGordon_that should just be a matter of recompliling sdl with that driver right?
00:03:57gevaertsI think so, yes
00:05:02JdGordon_I'm not having much luck with X on my mini2440 so I figured I'd just skip it :p
00:05:33gevaertsYou know there is a native port, right? ;)
00:05:52JdGordon_native misses out on networking and usb
00:06:21JdGordon_cant put much music on a 4gb sd card :p
00:06:28gevaertsget busy then!
00:10:15 Quit ecio (Ping timeout: 265 seconds)
00:14:20 Quit Lss (Read error: Connection reset by peer)
00:14:57 Quit jgarvey (Quit: Leaving)
00:16:21 Quit bmbl (Quit: Bye!)
00:22:25 Join stripwax_ [0] (
00:23:59 Quit stripwax (Ping timeout: 260 seconds)
00:26:51 Quit ender` (Quit: But there, everything has its drawbacks, as the man said when his mother-in-law died, and they came down upon him for the funeral expenses.)
00:29:52 Join planetbeing_ [0] (
00:33:04 Quit planetbeing__ (Ping timeout: 245 seconds)
00:33:10 Join anewuser [0] (anewuser@unaffiliated/anewuser)
00:33:57 Part froggyman
00:41:01 Join Tomis [0] (~Tomis@
00:43:52 Quit Tomis (Client Quit)
00:55:26***Saving seen data "./dancer.seen"
00:57:10*JdGordon_ looks at FS #11058 and could have sworn the displayer was brain dead enough to not care if viewports are conditional
00:57:15JdGordon_thats very wierd behaviour
00:57:31 Quit Schmogel (Quit: Miranda IM! Smaller, Faster, Easier.
01:00:36 Join Tomis [0] (~Tomis@
01:09:12 Quit leavittx (Read error: Operation timed out)
01:14:34Stephen__gevaerts, I've fixed knightboxd3.
01:20:14 Quit dfkt_ (Quit: -= SysReset 2.53=- Ph'nglui mglw'nafh Cthulhu R'lyeh wgah'nagl fhtagn.)
01:28:38 Part toffe82
01:29:53 Join planetbeing__ [0] (
01:31:21 Quit JdGordon_ (Quit: Miranda IM! Smaller, Faster, Easier.
01:33:07 Quit planetbeing_ (Ping timeout: 240 seconds)
01:34:31 Join ecio [0] (~ecio@
01:35:04 Quit Stephen__ (Quit: Leaving)
01:36:47 Nick fxb__ is now known as fxb (
01:37:37 Join JdGordon1 [0] (~jonno@
01:37:59 Join karashata [0] (
01:45:07JdGordon1is checkwps built without debug symbols?
01:46:23 Join Rob2223 [0] (
01:48:35 Join CaptainKewl [0] (~jason@
01:50:22 Quit Rob2222 (Ping timeout: 276 seconds)
01:50:48JdGordon1gevaerts: FYI... that phi-infoscreen which segfaulted is because the viewport acually uses font 2! (typo)
01:51:00JdGordon1so im going to make this more forgiving and undo the revert
01:53:09CIA-5New commit by 03jdgordon (r24994): undo that last undo, make the font loader more forgiving to user errors
01:54:43CIA-5New commit by 03jdgordon (r24995): remove a nonsense outdate warning. there is no token limit anymore
02:05:47 Quit JdGordon1 (Ping timeout: 264 seconds)
02:13:36 Quit ecio (Quit: Colloquy for iPhone -
02:18:50 Quit komputes (Quit: I haven't slept for ten days, because that would be too long.)
02:21:27 Quit stripwax_ (Quit:
02:26:40 Quit m3dlg (Ping timeout: 268 seconds)
02:27:57 Nick fxb is now known as fxb__ (
02:30:55 Nick S_a_i_n_t_ is now known as S_a_i_n_t (S_a_i_n_t@
02:30:59TheSevennow undo the undo of the undo!
02:31:02 Join planetbeing_ [0] (
02:31:54S_a_i_n_tJdGordon: have you noticed that the Title-Bar lags slightly when using %Lt & %Li?
02:32:05S_a_i_n_t...I worked it out by the way \0/
02:32:26JdGordonI dont tihnk t matters if both are being used
02:32:29JdGordondoes it?
02:33:07S_a_i_n_tI'm not sure, oh...actually, from what I've doesn't seem to do it without %Li
02:33:18S_a_i_n_tjust the title seems fine (without a delay)
02:33:23 Quit planetbeing__ (Ping timeout: 264 seconds)
02:33:52S_a_i_n_tI *thought* centering the title would look coo, ...but it looks crap IMO
02:35:00JdGordonwhy does it look crap?
02:35:26JdGordonI need to get around to doing a sbs for my mini2g.. I want to make the title bold and centered, and move the bar to the right
02:35:42S_a_i_n_toh, and you probably figured it out but I noticed this morning that you missed a % in the string for %Li you tested last night so maybe that was why the order was goofed. I can't tell as all my icons are blue dots (all the same)
02:35:57JdGordonyeah, saw that
02:36:05JdGordonand missed the p option
02:36:54S_a_i_n_twhy does it look crap: I guess it would look better if the menu was a little more centered, but then there is too much overhang off hte screen with longer menu items
02:37:09S_a_i_n_t*I only have 176px to play with there*
02:38:28CIA-5New commit by 03jdgordon (r24996): some small debug info fixes, add %Li and %lt and make uppercase subimage labels display correctly
02:38:58S_a_i_n_theh, yes...I noticed that also.
02:39:58S_a_i_n_tIf/When it gets done...I'm guessing %Li will be replaced with "display appropriate icon"?
02:40:12JdGordonit doesnt have to be
02:40:47S_a_i_n_tthe string is HUGE, oh...and (in the SIM) at least...I can't get it to stop complaining about the font identifier being used by the WPS already.
02:40:58S_a_i_n_tI have to make the .sbs use font 5 to work
02:41:14S_a_i_n_tas I have 2, 3, and 4 in my WPS
02:41:29JdGordonyou didnt update then?
02:41:56S_a_i_n_t...this is based on last nights testing...and that was current SVN at the time
02:42:12JdGordonthat should be fixed now
02:42:30S_a_i_n_twell, when we were working on it last was doing it then
02:42:49 Join Tomis2 [0] (~Tomis@
02:43:04S_a_i_n_tit (the SIM) was having a cry about "loaded font in use" or whatever the output is.
02:43:13JdGordonlies! lies and slander!
02:43:19S_a_i_n_thehe :D
02:44:10S_a_i_n_tIt still seemd to work (if the font ID was in use already) but if I closed/reopened the sim for instance I had to reload the .sbs each time
02:44:28S_a_i_n_tbut changing the font ID to one that wasn't being called elsewhere fixed that.
02:45:26 Quit Tomis (Ping timeout: 276 seconds)
02:45:26 Nick Tomis2 is now known as Tomis (~Tomis@
02:47:21JdGordonwhat warning was safe to ignore
02:47:32JdGordonunless you actually did use the same id twice in the one skin
02:48:21S_a_i_n_tnope...the WPS uses 2, 3 and 4...and (to get it to work without having to reload it on each boot) the .sbs had to use 5
02:48:37S_a_i_n_tif I just used 2, it worked after loading the .sbs
02:48:43S_a_i_n_tbut a reboot dropped it
02:49:57JdGordonthats odd
02:50:04JdGordonsvn up and tell me if its fixed?
02:50:23S_a_i_n_twell, that's what I was didn't seem like "intended behaviour"
02:50:39S_a_i_n_tI have to run for an hour or so, but I will when I get back.
02:50:42 Quit TheSphinX^_ (Quit: XChat)
02:50:52 Quit karashata (Quit: The fluffy dragon has left completely!)
02:55:30***Saving seen data "./dancer.seen"
03:10:28CIA-5New commit by 03jdgordon (r24997): Fix FS #11058 - conditionl viewports cause sublines to go a bit crazy
03:12:16JdGordonS_a_i_n_t: double check that one please... using your test wps it looks fixed
03:20:44CIA-5New commit by 03jdgordon (r24998): fix FS #10926 - backlight turning off stops track info being updated
03:30:04 Join planetbeing__ [0] (
03:33:20 Quit planetbeing_ (Ping timeout: 240 seconds)
03:36:17CIA-5New commit by 03theseven (r24999): Fix numerous non-critical iPod Nano 2G FTL bugs
03:36:53CIA-5New commit by 03theseven (r25000): PANIC!!!
03:37:08scorcheTheSeven: cheater
03:37:53TheSevenscorche: why?
03:41:57 Join iq [0] (~iq@unaffiliated/iq)
03:43:00S_a_i_n_tyou stole 25000 from JdGordon hehe...he won't be impressed :P
03:53:44 Quit Adnyxo (Ping timeout: 264 seconds)
03:59:01S_a_i_n_tJdGordon: I will (attempt to at least) verify that r24997 fixes FS #11058, and that the font identfiers can be shared again building a 25000 SIM now, just got to run out for a half hour or so however.
04:36:07 Join Barahir [0] (
04:39:20 Quit Barahir_ (Ping timeout: 245 seconds)
04:40:02 Quit dys (Ping timeout: 276 seconds)
04:40:25 Join dys [0] (
04:44:02 Quit anewuser (Quit: Likely voting 114 entries =o)
04:44:24 Join ecio [0] (
04:45:55 Join anewuser [0] (anewuser@unaffiliated/anewuser)
04:53:25 Join Aslai [0] (
04:55:06 Quit Aslai (Client Quit)
04:55:34***Saving seen data "./dancer.seen"
05:00:52 Join FlynDice_ [0] (
05:01:28 Quit FlynDice (Ping timeout: 246 seconds)
05:04:20TheSevenanyone around who is annoyed by those FTL crashes?
05:05:17 Quit krabador (Quit: Sto andando via)
05:07:42 Quit antil33t ()
05:11:27 Quit MagusG (Quit: ( :: NoNameScript 4.22 :: ))
05:15:17 Join FlynDice [0] (
05:15:49 Quit FlynDice_ (Ping timeout: 246 seconds)
05:16:38S_a_i_n_tTheSeven: I think everyone is annoyed by them :P But do you mean is anyone still getting the frequently?
05:17:36TheSevenwell, until today one would get them after every unclean shutdown
05:17:42TheSeveni have a solution for that now :-)
05:20:24TheSevenhah, i just witnessed *diskmode* causing an ftl crash!
05:20:46TheSevenjust because i ejected the device *immediately* after I finished copying things
05:20:58TheSevenseems like it didn't have time to clean up
05:22:42 Join MagusG [0] (
05:30:16S_a_i_n_tTheSeven: I've had that problem before now...copy files, "safely (pffft! yeah right) remove hardware", remove usb...crash.
05:30:34S_a_i_n_t*if done too quickly
05:36:26 Quit togetic (Ping timeout: 256 seconds)
05:39:37 Join tmzt [0] (
05:41:49*S_a_i_n_t is curious as to what TheSeven's solution for this is...
05:42:07 Quit Horscht (Quit: Verlassend)
05:43:05S_a_i_n_tand ps: can you get crypt_firmware to work on your 2g? (I can't) it causes *PANIC* Stkov main and my main stack size is already two times the norm...
05:45:54S_a_i_n_tI suspected it may have been my (...Ahem) "adaptations" to the Nano2g source files...but it doesn't work on a clean svn either ;[
05:46:20S_a_i_n_tAnd I'd REALLY like to update my bootloader and be able to use OSOS install.
05:50:47 Join togetic [0] (~togetic@unaffiliated/ibuffy)
06:20:26 Quit anewuser (Quit: Likely voting 114 entries =o)
06:36:25 Join shiretoko [0] (
06:37:34 Join aguy [0] (
06:38:41aguymy wife just got a Lisong mp3 player off ebay, has anyone had any luck putting rockbox on one of those?
06:39:17tmztthere's a project for those type of players I think is it
06:39:42aguychecking into it, thanks
06:41:11aguyawesome. thanks
06:41:52tmztright sorry
06:42:07tmztLlorean: was there ever and faq or topic mention of that?
06:43:04CIA-5New commit by 03jdgordon (r25001): make %Fl (font load) more consistant with bmp loading and require the whole filename, i.e require the .fnt.
06:43:29Lloreantmzt: Not that I know of.
06:55:32 Quit nimak (Ping timeout: 260 seconds)
06:55:35***Saving seen data "./dancer.seen"
06:56:23 Quit aguy (Quit: Java user signed off)
06:56:53 Join nima [0] (
06:57:17CIA-5New commit by 03jdgordon (r25002): remove some gremlins... viewport_set_*() sets the font to the UI font for that screen which is bad for the parser.
07:22:17 Join fejfighter [0] (
07:23:42 Quit fejfighter (Remote host closed the connection)
07:26:14pixelmaJdGordon: trying to understand your statement - what part of a font has to fit into which buffer (and do you mean everything but the user font)?
07:26:43 Join einhirn [0] (
07:27:13JdGordonpixelma: you talking about the comment in #rockbox-community?
07:28:12JdGordonI was complaining because currently the skin buffer for the mini2g is ~32K, a skin font is 10K which means that with and cabbiev2.wps there isnt enough room for the theme
07:28:24JdGordonI modified them to use a second font for some info
07:29:34pixelmawhat is skin font then?
07:29:56JdGordona font used/loaded by a skin/sbs/wps/etc
07:32:48JdGordonnow I have a nice modified default theme and cant upload it :p
07:33:31pixelmadoes this mean it's some fixed reserved space for glyphs to be used from different fonts - glyphs are swapped in and out of this as needed? I'm trying to understand the mechanism
07:33:32 Nick shiretoko is now known as zz_shiretoko (
07:34:14 Join fejfighter [0] (
07:34:45JdGordoneach font that a skin wants to load gets its own 10K buffer from the skin buffer
07:42:09 Quit CaptainKewl (Remote host closed the connection)
07:53:46 Join stoffel [0] (
08:10:10 Quit jd (Ping timeout: 245 seconds)
08:15:59 Join pondlife [0] (~Steve@rockbox/developer/pondlife)
08:22:57 Join ender` [0] (
08:23:27 Join B4gder [0] (~daniel@rockbox/developer/bagder)
08:26:52 Join leavittx [0] (
08:27:55 Quit stoffel (Remote host closed the connection)
08:32:08*S_a_i_n_t guesses that tis is the time to concentrate on the resizable skin buffer then....
08:35:02 Quit leavittx (Read error: Connection reset by peer)
08:36:30S_a_i_n_tJdGordon: Going by 25001's comment, I'll need to load an additional font as %Fl|'blah'||'blahfont.fnt'| now?
08:36:43S_a_i_n_tunless I want my theme to break that is...
08:37:07 Quit planetbeing__ (Quit: planetbeing__)
08:37:15S_a_i_n_twithout the additional | however...
08:42:04 Join evilnick_ [0] (
08:42:44 Quit evilnick (Ping timeout: 265 seconds)
08:42:52 Quit feisar_ (Ping timeout: 260 seconds)
08:44:34 Join feisar [0] (
08:45:02 Nick feisar is now known as Guest41701 (
08:46:21amiconnAfter changing the filter calls in libdemac, I also get " Error: bad immediate value for offset" on ARMv6
08:47:11amiconnSeems I need to go back from nice .rept n / ... / .endr to the preprocessor macro mess
08:48:42 Join flydutch [0] (
08:48:57amiconnThis may help gevaerts to get the sim running on the n900 though
08:49:04 Join evilnick [0] (
08:51:35 Quit Guest41701 (Ping timeout: 276 seconds)
08:51:46 Quit evilnick_ (Ping timeout: 246 seconds)
08:54:31JdGordona quick not so nice warning.... if you fill up the skin buffer so there isnt enough room for the wps, and the hardcoded wps also, then Bad Things Happen
08:55:35JdGordonS_a_i_n_t: wtf? no, the only difference is .fnt is required at the end of the line, no other change
08:55:36***Saving seen data "./dancer.seen"
09:00:29 Join jd [0] (
09:00:29 Quit jd (Changing host)
09:00:29 Join jd [0] (~jd@Wikipedia/HellDragon)
09:02:00 Join petur [0] (~petur@rockbox/developer/petur)
09:04:33 Join feisar_ [0] (
09:16:41B4gderon March 25 we celebrate the 8th anniversary of Rockbox code as the first actual commits were made Mon Mar 25 15:21:30 2002 CET
09:17:50B4gderwhich makes it roughly 8.5 commits per day on average since then
09:19:21JdGordonhow many do we need to do before then to get that to 10?
09:20:10B4gder~4200 if you make them before March 25 ;-)
09:20:40JdGordoneasy :)
09:20:51B4gder8 years being 2922 days
09:21:20JdGordondoes that account for leap years? thats the type of thinking that causes big name electronics to fail!
09:21:36B4gderthat does include two leap days, yes ;-)
09:22:36S_a_i_n_tI guess my last question was a really backward way of saying "does a 'load font' fine now have to include the .fnt extension", to which the answer is yes. Thanks :D
09:24:20 Join DerPapst [0] (
09:30:33CIA-5New commit by 03jdgordon (r25003): FS #11072 - display the filename of the skin begin parsed in the debug output
09:49:34 Join pamaury [0] (
09:51:02 Quit fejfighter (Remote host closed the connection)
09:56:55 Join bmbl [0] (~Miranda@unaffiliated/bmbl)
09:57:35 Quit flydutch (Quit: /* empty */)
09:59:20 Join flydutch [0] (
10:07:17 Quit knine (Ping timeout: 256 seconds)
10:15:21 Quit bmbl (Ping timeout: 268 seconds)
10:16:19 Quit pamaury (Remote host closed the connection)
10:18:58 Join pamaury [0] (
10:20:51 Join efyx_ [0] (
10:32:12ZagorB4gder: did you include the cvs commits in your calculation?
10:32:39B4gderI counted from the start of the svn info, so it should include everything converted from cvs
10:33:03Zagoroh right, the conversion included the individual commits
10:33:16pixelmarevision numbers didn't start again with the change from cvs to svn, or am I remembering wronng?
10:33:23pixelmawrong too...
10:33:27 Quit pamaury (Ping timeout: 276 seconds)
10:34:38B4gderwell, we had no revision number in the cvs days
10:34:55*pixelma now remembers a whole lot of commits during the move to svn
10:35:00B4gderit was all date stamps then
10:35:04linuxstbBut CVS commit #1 is (more or less) SVN revision 1?
10:35:18B4gderand commit #4 is the first actual code
10:35:35 Join kaniini [0] (
10:36:16 Join pamaury [0] (
10:51:42 Quit Tomis (Quit: Tomis)
10:55:38***Saving seen data "./dancer.seen"
11:29:04 Quit DerPapst (Quit: Leaving.)
11:29:19stuckeyCould anyone reccommend me a music player? I'm looking at the SanDisk Sansa Fuze 4GB, but I don't really have anything to go on model wise, or version wise.
11:34:37 Join dfkt [0] (dfkt@unaffiliated/dfkt)
11:47:09 Join |DaMaGeD| [0] (~dam@
11:48:10 Quit |DaMaGeD| (Client Quit)
12:03:33 Quit efyx_ (Remote host closed the connection)
12:08:00 Quit scorche (Ping timeout: 265 seconds)
12:10:54 Join DerPapst [0] (
12:14:47B4gderstuckey: see the BuyersGuide wiki page on
12:20:03S_a_i_n_tffs! 2nd "please restore using iTunes" crash in 7 days....*sigh*
12:24:02S_a_i_n_tI didn't see it myself, (the GF had it at the time) but apparently she was playing jewel and it froze on a black screen for like 30 seconds, then rebooted and "please resotre using itunes" 'd.
12:31:20gibbon_well nevermind. i was getting a "bad checksum" on clip+ with the r25003, but then i noticed that i was using clipv2 rockbox... just to explain the weird "hm".
12:43:38linuxstbgibbon_: Yes, perhaps our bootloaders should give a more useful message when that happens...
12:46:13 Quit kaniini (Read error: Connection reset by peer)
12:46:20 Join kaniini [0] (
12:46:55gibbon_linuxstb: actually they do - at least if I had read more carefully, i would have noticed the "model: cliv2" (or similar) directly above the error message ;)
12:48:04gibbon_to me that seems useful enough at least for the case of wrong model. especially on the clip models there is not much space for verbosity
12:49:22gibbon_at least, if the possibility of harming the hardware at this point is small enough
12:49:38gibbon_but amazing progress on the samsa devices.
12:51:05stuckeyB4gder: Awesome. Is it up-to-date?
12:51:14linuxstbgibbon_: "Wrong model" is a shorter error message than "bad checksum" (just!)
12:51:55gibbon_linuxstb: since i don't know if this error occurs only on model mismatch, i didn't suggest that
12:52:02gibbon_but you are obviously right :)
12:52:31stuckeyThe Sansa Fuse isn't listed...
12:53:41gibbon_listed where?
12:53:46stuckeybuyer's guide wiki
12:54:09*stuckey is looking to buy a player
12:54:26 Join piotrekm [0] (~piotrek@unaffiliated/piotrekm)
12:54:52gibbon_Fuze is according to the wiki not a usable target yet
12:55:12gibbon_well ... much green though...
12:55:26Tornefuzev1 is stable/supported
12:55:34gibbon_if you are buying a target for daily use though i would suggest, sticking with release supported devices
12:55:38stuckeyI'm willing to deal with some bugs, if that means I can get a better peice of hardware that *will* be supported soon enough.
12:55:41***Saving seen data "./dancer.seen"
12:55:45gibbon_Torne: sorry for that
12:55:46Tornegibbon_: fuzev1 is stable and supported, *if* you can find a v1.
12:55:53Tornefuze v2 does not have a port at all yet
12:56:01stuckeyHow manys vs are there?
12:56:03Tornethough it's likely that much of the work on the clipv2/clip+ applies to it
12:56:06Tornestuckey: just two
12:56:15Tornestuckey: afaik you can't tell them apart without turning them on though
12:56:21Torneno way to tell from looking at the player/box
12:56:50gibbon_Torne: someone with write permission should note that on then, perhaps
12:57:06stuckeyis the fuze a good player?
12:57:08TorneIt is?
12:57:15Tornegibbon_: what doesn't it say that you expect it to?
12:57:37Tornestuckey: dunno, don't have one
12:57:49gibbon_the page itself is fine, if viewed independently
12:58:24Tornebuyers guide hasn't been updated for a year, so it is missing osme of the more recent targets
12:58:32gibbon_but some "unstable/unusable" pages link to that, so a column in the table would be nice perhaps
12:58:44Tornegibbon_: Ports have their own wiki page
12:58:51gibbon_"stable support" or the like
12:59:02TorneOh, i see wha tyou mean
12:59:05Tornethat might be good, yes
12:59:19Torneit says at the top of the table, though
12:59:23gibbon_i once had write access... i don't know if i have still
12:59:23stuckeyIs there a player that everyone agrees is either very good or the best?
12:59:33 Join karashata [0] (
12:59:45Tornegibbon_: we don't take it away or delete accounts so unless you've lost your password you do
12:59:58Tornestuckey: no, that's absurd :)
13:00:03gibbon_Torne: it has been before the TWiki switch iirc
13:00:05Tornepeople have different requirements
13:00:08Tornegibbon_: oh, dunno then
13:00:24Tornestuckey: depends if you want recording or not, depends how much storage you need, etc
13:00:40stuckeyTorne: I don't need either recording nor much storage.
13:00:50stuckeyLong battery would be nice... build quality is important to me
13:00:55Tornehow much is not much?
13:01:03stuckey2GB - 8GB should be sufficient
13:01:10Torneright, so you want a flash based player
13:01:51stuckeyAny chance there exists one that isn't made in china?
13:01:59Torneno idea
13:02:08Tornei would guess the vast majority of all mp3 players are
13:02:12Torneregardless of rockbox support :)
13:02:15Tornewhy do you care?
13:03:15Torneanyway, if all you want is that then you can basically pick any of the stable ports which is flash-based. whichever you can get most easily and like the look of
13:03:44stuckeyTorne: Just something I personally always try to do; buy stuff that isn't made in China.
13:03:53Tornefor stable ports battery life is mostly dependent on the battery capacity, as we're reasonably optimal for most of them
13:04:01Torneand battery capacity is mostly dependent on the physical size of the player :)
13:04:06Torneso, it's a pretty straight tradeoff
13:04:21gibbon_Torne: are you familiar with the Sansa AMS progress in concern of the clip+? i was wondering what the yellow Yes/Yes in the battery collumn stands for
13:04:24Tornestuckey: well, almost all consumer electronics are. so, you are likely to be out of luck
13:04:41stuckeyTorne: I got a camera that's made in Japan :)
13:05:07stuckeyMost all are, but if you look, you can find stuff that isn't.
13:05:11Tornestuckey: the final assembly perhaps but I would be *stunned* if major internal components were not made in china.
13:05:22Torneincluding entire subassemblies.
13:05:25stuckeyDamn chinese make everything
13:05:28Tornecapitalism works that way
13:05:43Tornegibbon_: it means that we think battery charging and monitoring works, mostly, probbaly. :)
13:06:26gibbon_then a good thing to do while i am here at work would be to test that and let this column have a reason to get green :)
13:06:59gibbon_would be nice to know, if battery control in these players works over 1wire
13:08:05stuckeyTorne: which player do you use?
13:08:37Tornei use ipod video 80gb
13:08:44Tornebecause i have a lot of music and thus need a resonably large hard disk based player
13:09:07Tornegibbon_: No, battery control in most of these players is done by the CPU itself
13:09:22Tornegibbon_: or by extremely dumb chips controlled via a couple of GPIO lines, or at most i2c
13:09:29Tornei don't think we have any targets with 1wire.
13:10:03TorneNone of them have smart batteries, they are all dumb cells attached to dumb charging chips, or even just to actual logic (and hten the cpu has to manage charging itself)
13:10:29 Join anewuser [0] (anewuser@unaffiliated/anewuser)
13:10:33gibbon_reduced size has a tradeoff
13:12:34 Join Adnyxo [0] (
13:18:13stuckeyWould it even be possible to load rockbox on a v2, or would the install process fail?
13:18:48 Quit anewuser (Quit: Likely voting 114 entries =o)
13:19:36gibbon_stuckey: they use an entirely different SoC, so it would fail
13:19:57gibbon_a different port is needed (bootloader, hardware control)
13:21:18stuckeyOkay, I'm looking at one for sale, and the guy says rockbox is loaded on it already.
13:27:38 Nick fxb__ is now known as fxb (
13:30:09Tornethen it's a v1 :)
13:32:07stuckeyIt's one year old
13:37:11 Join anewuser [0] (anewuser@unaffiliated/anewuser)
13:42:07 Join moos [0] (moos@rockbox/staff/moos)
13:44:42partwhat's the difference between sansa clip and clip+?
13:47:00Tornethe clip+ has a microSD slot, and its internal hardware is somewhat different
13:47:07Tornethe clip+ is similar to the clipv2 but not identical.
13:47:51 Join AsusFreak [0] (
13:48:33partoh, they seem to support microsdhc now
13:49:15partclip+ seems to support ogg vorbis and flac too
13:51:10 Join efyx_ [0] (
13:51:13gibbon_why would it not? or do you mean in OF?
13:51:26AsusFreakHi all, I just read a mail concerning my theme AF_Black. It is deleted, because there should be a new version. I uploaded Version V4.0 on Monday and Version V5.0 on Tuesday. Now both are gone. Please recover my last Version V5.0 or give me a sign to upload again. Thank you.
13:51:50part8GB clip+ is cheaper than an 8GB clip at a local store, I'm tempted to buy one
13:52:02partgibbon_: OF
13:52:14gibbon_part: ah.. then its special :)
13:52:33gibbon_part: i have bought a clip+, too... but not to use OF
13:53:27AsusFreakOh - in addition: two of my other themes disappeared as well. AF_Sand and AF_Wood. Therefor I didn't get an email.
13:54:44partgibbon_: oh, I'm not tempted by the OF
13:55:53*S_a_i_n_t thought he was going insane...but now wonders why 12-Nimbus and 14-Nimbus are both 9px high...
13:56:20S_a_i_n_tjust differen tspacing.
13:57:11gibbon_part: me neither
13:58:25 Join piotrm [0] (
13:58:25 Quit piotrm (Changing host)
13:58:25 Join piotrm [0] (~piotrek@unaffiliated/piotrekm)
13:58:57partI'm surprised that they've bothered with sdhc and vorbis+flac
13:59:08S_a_i_n_tI always though that the 8, 9, 10, 11 etc refered to 'X pixels high', various screendumps later...and I realise this is not the case at all...
13:59:49gevaertsAsusFreak: I can't see any traces, so I suspect that they were deleted instead of hidden... Was this about the multifont changes? If so, please just reupload, and I'll make sure to tell JdGordon to do it properly next time :)
14:00:35 Quit piotrekm (Ping timeout: 252 seconds)
14:00:35 Nick piotrm is now known as piotrekm (~piotrek@unaffiliated/piotrekm)
14:00:44S_a_i_n_twell, I assume it's the hight of the font including blank space top & bottom...but it does seem kinda odd that nimbus is the same hieght in 14 and 12 flavours.
14:00:57gevaertsAsusFreak: also, this is about the D2, right?
14:00:57TorneS_a_i_n_t: isn't it the point size it was converted from?
14:01:02S_a_i_n_tto me at least.
14:01:08gevaerts(or other 320x240)
14:01:14TorneS_a_i_n_t: which has nothing to do with pixel dimensions at all
14:01:32TorneI assumed 12-Nimbus was the Nimbus font, at 12pt, converted to bitmap
14:01:49S_a_i_n_tTorne: I'm realising this now...
14:01:59Tornei could be wrong though
14:02:24S_a_i_n_tbut (in my head at least) it made sense to me that a 12pt font would be smaller than a 14pt font
14:02:54S_a_i_n_tmy theory works for Helvetica...but not Nimbus
14:03:45AsusFreakgevaerts: yes it's for the D2
14:04:42AsusFreakIn all cases multifont was used. Is it a problem with the CheckWPS as a few days ago?
14:05:48AsusFreakAnd yes it's for D2 and other 320x240 like Ipod Video.
14:06:06 Join knine [0] (
14:06:21S_a_i_n_tis there a "make fontpack" command for a build environment?
14:06:33AsusFreakSo I have to reupload them again?
14:07:09gevaertsAsusFreak: I think the syntax was slightly changed, and you now have to add the .fnt extension to the filename as well
14:07:20B4gderS_a_i_n_t: 'make fontzip'
14:07:48S_a_i_n_tthanks, does it have to be in a particular dir? or can it just be in the build directory?
14:08:00gevaertsAsusFreak: so yes, please reupload
14:08:15B4gderS_a_i_n_t: its to be run in the build dir
14:08:51AsusFreakAdd .fnt extension to the filename - ups, that's pretty new to me, sorry. So I'll have too study the FAQs first.
14:09:24gevaertsAsusFreak: in the %Fl tag I mean
14:09:54S_a_i_n_tgevaerts: and the wps
14:10:08S_a_i_n_tmy sim just complained about it not having .fnt
14:10:28 Quit kaniini (Ping timeout: 248 seconds)
14:13:54gevaertsS_a_i_n_t: where do you think the %Fl tag is?
14:14:39S_a_i_n_tyeah, sorry...I was just looking for what made me think you were exclusively talking about the .sbs
14:14:41AsusFreakgevaerts: Is this documented elsewhere? I only found a mail of JdGordon in the forum.
14:14:54S_a_i_n_tnot surprisingly, I imagined it.
14:14:55gevaertsAsusFreak: the wiki page has been updated
14:16:19AsusFreakgevaerts: And now I see it, too. Thank you. I will change my code and reup them again.
14:20:15TheSevenS_a_i_n_t: i've never really used crypt_firmware, i usually use libipodcrypto to do this
14:20:40TheSevenbut with the new bootloader, there's no need for an osos install anyways
14:21:14S_a_i_n_twhen was the *new* bootloader updated?
14:21:37TheSevenS_a_i_n_t: it's the experimental thing we discussed via pm some days ago
14:21:37*S_a_i_n_t wonders if he already has it.
14:21:50S_a_i_n_tAhhh right.
14:22:21TheSeveni updated it yesterday to include a recovery tool that will pick up an FTL in a bad "unclean shutdown" state and repair it, which should work at least in most cases
14:23:24S_a_i_n_t...still the riculously complicated( yet necessary to avoid bricking) install process?
14:23:32pixelmaS_a_i_n_t: didn't we talk about the font names earlier? The number means the complete line height (so that you know how many lines fit on your screen or which font could be used instead without problems. How the basic character height changes is dependent on font - and remember that there is no antialiasing so it's not easy to just scale a font to one ore two pixels bigger and keep the same (or at least similar enough) look
14:25:02pixelmaand it was pointed out to me that 12- and 14-Nimbus are not completely the same although base character height is the same
14:25:30S_a_i_n_tpixelma: wasn't me...but thanks. I worked out it was the total line height. I was jsut going insane thinking something broke in my code because of the very small visaul difference between Nimbus 12 and 14
14:26:32S_a_i_n_tI was used to the height difference with the Helvetica fonts (for a multifont visual effect), and I thought I'd try Nimbus for a change, but didn't think it was actually working.
14:31:10TheSevenS_a_i_n_t: yes, it's still complicated, because a) i haven't written a proper installation program yet, b) because it wasn't tested on much hardware, so i don't trust it yet.
14:31:19 Join Schmogel [0] (
14:31:35TheSevenif we are sure that it works on all devices, we can just flash it blindly without testing it first, which can be fully automated
14:32:02TheSevenbut I would like to see it work on more than 5 ipods before doing that
14:40:49 Quit piotrekm (Quit: piotrekm)
14:46:17*S_a_i_n_t suspects there is no CygWin package for libipodcrypto
14:46:34TheSevenit's all python ;-)
14:47:39S_a_i_n_tSetting up/installing pyton to achieve this doesn't seem too painful
14:48:02S_a_i_n_twell, less painful than trying to figure out why crypt_firmware is failing
14:48:12TheSevenyou will have to do that anyways, sooner or later
14:48:38TheSevenyou'll need python 2.6, pyusb and the ibugger driver
14:48:49S_a_i_n_tI know..I had a pretty big hardware fail a few days ago...and I've been lazy
14:49:05S_a_i_n_tit used to have it set up
14:51:01 Quit ecio (Ping timeout: 265 seconds)
14:55:43***Saving seen data "./dancer.seen"
15:00:14 Quit AsusFreak (Quit: CGI:IRC (EOF))
15:05:34 Nick fxb is now known as fxb__ (
15:08:05TheSevenis there an on-target implementation of uclpack?
15:08:47LloreanIsn't it used in the compressed Archos in-flash bootloaders at least?
15:09:00LloreanI guess that would be unpacking.
15:09:39TheSeveni mean packing
15:10:20LloreanYeah I realized that. Don't know, sorry.
15:10:43TheSevenhow does ucl perform (in terms of compression ratio and unpacking speed) compared to other commonly-used compression methods?
15:12:01LloreanI don't know much about it at all, but I got the impression it was very fast to decompress and could have a very small decompressor, which was the major deciding factor.
15:12:31 Join jgarvey [0] (
15:12:46TheSevendecompressor size wouldn't hurt as long as it is compensated by better compression ;-)
15:13:09 Join kugel [0] (~kugel@rockbox/developer/kugel)
15:18:07 Join perfectdrug [0] (
15:19:29 Join wodz [0] (
15:19:34ZagorTheSeven: it's about on par with gzip
15:20:40wodzwhat are the criteria a port has to fullfil to be commited to svn?
15:21:29Zagorwodz: very little, actually. early ports are work-in-progress and described as "unusable"
15:21:32LloreanGenerally they need to be added to the build system properly, and have some code of their own (probably outside of simply a simulator)
15:21:44 Join kugel_ [0] (
15:21:54 Quit kugel (Disconnected by services)
15:21:58 Nick kugel_ is now known as kugel (
15:22:04 Quit kugel (Changing host)
15:22:04 Join kugel [0] (~kugel@rockbox/developer/kugel)
15:22:59 Quit karashata (Quit: The fluffy dragon has left completely!)
15:23:16 Join evilnick_B [0] (~0c140464@rockbox/staff/evilnick)
15:23:24wodzwould it be sufficient?
15:23:39Zagorwodz: absolutely
15:24:00wodzwhat is the procedure?
15:24:19Zagorcreate a patch and submit to the tracker
15:24:45wodzagainst svn head?
15:24:47Zagorhave you no way of loading the code onto the device?
15:25:02ZagorI mean other than bdm?
15:25:08wodzI have
15:25:11wodzno problem
15:25:44Zagorgreah. please extend that wiki page with installation instructions so other people can try the code.
15:28:11linuxstbTheSeven: There's no on-target ucl compression. Just decompression. You can easily test it - just compile "uclpack" in tool/ucl/
15:28:25linuxstbtools/ucl/ even...
15:28:59 Join mischasworld [0] (
15:29:41TheSevenwhere was the decompression stub hidden?
15:30:02 Join ecio [0] (
15:30:05wodzThanks. I will clean up my build tree, update sources and prepare patch. This will probably take me a week or so.
15:30:29linuxstbTheSeven: In rbutil/mkamsboot/
15:30:47TheSevenwodz: wow, great documentation work! i wish i had such detailed docs for the ipods...
15:31:59 Quit einhirn (*.net *.split)
15:31:59 Quit nima (*.net *.split)
15:31:59 Quit FlynDice (*.net *.split)
15:32:11 Quit wodz (Quit: CGI:IRC)
15:33:15linuxstbTheSeven: BTW, when you type "make" in tools/ucl/ the uclpack binary is created up one level in tools/
15:35:42 Join krabador [0] (
15:36:07 Quit ecio (Ping timeout: 240 seconds)
15:36:34 Join einhirn [0] (
15:36:34 Join nima [0] (
15:36:34 Join FlynDice [0] (
15:49:44 Join ecio [0] (
15:50:01TheSevenwow, uclpack even compiles for windows with a small makefile fix
15:51:39 Quit ecio (Client Quit)
15:51:55TheSevenare there any restrictions on which compression modes i may use for the stub to work?
15:55:54 Join ecio [0] (
15:56:02linuxstbTheSeven: I was wondering about that.... I think you just use the defaults.
15:58:16linuxstbAh no, I think it's −−nrv2e -10
15:59:13linuxstbmkamsboot.c contains its own compress function using libucl.a - it doesn't call uclpack
15:59:21 Quit efyx_ (Remote host closed the connection)
15:59:56TheSevenit doesn't like -10
16:00:23linuxstbAh sorry, it's −−10
16:01:21 Join karashata [0] (
16:06:29 Join Casainho [0] (
16:07:00 Part Casainho
16:07:04 Join Casainho [0] (
16:12:26 Join piotrekm [0] (
16:12:27 Quit piotrekm (Changing host)
16:12:27 Join piotrekm [0] (~piotrek@unaffiliated/piotrekm)
16:22:14 Join robin0800 [0] (
16:36:43 Join toffe82 [0] (~chatzilla@
16:37:13 Join captanikewlllll [0] (~2669ecc2@gateway/web/freenode/x-snheouokxnbinxbt)
16:44:10 Join Hillshum [0] (
16:46:47 Quit ender` (Read error: Operation timed out)
16:49:56 Join efyx_ [0] (
16:55:47***Saving seen data "./dancer.seen"
17:00:15 Quit flydutch (Quit: /* empty */)
17:03:54 Join kramer3d [0] (~kramer@unaffiliated/kramer3d)
17:04:28 Quit B4gder (Quit: It is time to say moo)
17:06:46 Join liar [0] (
17:09:56 Quit anewuser (Quit: Likely voting 114 entries =o)
17:13:38 Quit kramer3d (Quit: Leaving)
17:17:03 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
17:18:35 Quit krabador (Read error: Connection reset by peer)
17:22:33 Quit Schmogel (Read error: Connection reset by peer)
17:26:01 Join Schmogel [0] (
17:30:36 Quit DerPapst (Quit: Leaving.)
17:33:31 Quit pamaury (Quit: Quitte)
17:48:37 Part pondlife
17:49:33 Quit rvvs89 (Quit: Reconnecting)
17:49:39 Join rvvs89 [0] (
17:49:39 Quit rvvs89 (Changing host)
17:49:39 Join rvvs89 [0] (robotnik@pdpc/supporter/base/rvvs89)
17:51:04 Join trio [0] (
17:52:51triohi i was wondering if its possible to use rockbox with a trio v418 and trio v218 both are made by machspeed
17:53:35 Join bertrik [0] (
17:55:18triois it possible to use rockbox with a trio v418 and trio v218 both are made by machspeed because i badly need a ui replacement
17:55:23GodEatertrio - Rockbox works specifically ONLY on the players we mention on the front page at
17:55:32GodEaterso the answer is no
17:56:17triook ty is there any other ui replacements i could lookinto
17:56:36Tornerockbox is not a UI replacement, it's an operating system
17:57:35triook is there a different os i should know about so i can look into it
17:58:28Tornes1mp3 have an alternative firmware for certain very low spec chinese DAPs
17:58:38Torneother than that, not really
17:59:32triook ty is there anyway to develop one for my mp3 player
18:01:17Torneassuming it is powerful enough to run rockbox
18:01:34Torneotherwise probably not; s1mp3's work only works on things with S1 processors
18:01:45 Quit petur (Quit: work->home)
18:01:56triook ty
18:02:22Tornebe prepared for it to require a fairly high level of programming ability and hundreds of hours of work :)
18:03:01 Join stoffel [0] (
18:03:03triook ty
18:03:45 Join komputes [0] (~komputes@ubuntu/member/komputes)
18:05:38 Join ender` [0] (
18:07:23 Quit trio (Quit: CGI:IRC (EOF))
18:09:06stuckeyI can't register on the forums... keeps saying the letter I give are wrong.
18:09:39Tornewell, are they?
18:09:40 Nick YPSY is now known as Ypsy (
18:10:03Tornelooks like the captchas have gone unreadable again
18:10:14 Join kramer3d [0] (~kramer@unaffiliated/kramer3d)
18:10:16Torneok only some o fhtem
18:10:24stuckeyyes, it always displays s :)
18:10:25Tornestuckey: hit request another image until it's clearer :)
18:10:33Torneshould be 5 letters
18:10:41Tornesometimes it only draws the first letter
18:11:56 Join Farthen [0] (
18:18:25 Quit captanikewlllll (Quit: Page closed)
18:19:39 Join domonoky [0] (~Domonoky@rockbox/developer/domonoky)
18:31:00 Quit amiconn (Disconnected by services)
18:31:02 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
18:31:04 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
18:31:04 Quit pixelma (Disconnected by services)
18:31:07 Nick Ypsy is now known as YPSY (
18:31:21 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
18:31:26 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
18:39:37stuckeyTorne: Thanks! I was able to register. I made my first post in general.
18:43:46 Quit mischasworld (Remote host closed the connection)
18:48:49 Join funman [0] (~fun@rockbox/developer/funman)
18:48:58 Join DerPapst [0] (
18:50:49kugelfunman: what's the differences between as3525 and as3525v2 system init?
18:51:23kugelI tried writing an lcd driver for the fuzev2 without success. the OF enables some peripherals before calling lcd_init_device, would I need to do that too?
18:51:44funmankugel: in rockbox?
18:53:09funmanI think CCU_SRC bits are different (at least there are more bits)
18:53:38 Quit DerPapst (Client Quit)
18:53:44funmanCGU_PLLA is completely different (although we have a known setting), CGU_PROC is different, and CGU_PERI might be different
18:53:54funman(in the divider bits, the peripheral bits look the same)
18:54:14*funman draws bertrik's attention to the Sansa AMS forum thread
18:55:51***Saving seen data "./dancer.seen"
18:59:19kugelJdGordon: what target did you notice the slower theme loading on?
18:59:49kugeltheme loading has always been incredibly slow on my e200 when the music is playing, while it's blazing fast on my fuze
19:01:12 Quit kugel (Disconnected by services)
19:01:17 Join kugel [0] (
19:01:27 Quit kugel (Changing host)
19:01:27 Join kugel [0] (~kugel@rockbox/developer/kugel)
19:02:05funmankugel: did you identify which peripherals are enabled?
19:02:25kugeldbop and gpio I think
19:02:48funmanyou'd only need to enable dbop, gpio is already enabled by mkamsboot
19:03:09kugelalright, I'll try
19:03:36funmanwhat's in fuzev2/lcd-fuzev2.c:as3525_dbop_init() looks ok (CGU_DBOP = (1<<3) | div)
19:04:09funmanbertrik: are you still against adding i2c code in mkamsboot ?
19:04:16kugelfunman: the files in svn are plain copies from fuzev1
19:04:36bertrikfunman, yes, I think it adds a lot of not-so-simple-code
19:04:54bertrikbut if we absolutely need it, I guess we'll have no other choice
19:04:59funmanbertrik: ranmachan's diff looked not very long at first sight
19:05:08funman(I didn't read it carefully yet)
19:05:26bertriklucky we don't use i2c interrupts
19:06:00bertrikI thought there were two buttons that needed "assistance" from some other GPIO before being able to be read, have we fully explored those?
19:06:17funmanon c200v2 ?
19:07:29funmangibbon_: battery charging / monitoring seems to work on Clip+ but I didn't test extensively, so I marked them in yellow. Probably making a complete charge/discharge cycle would be enough to mark it as green
19:08:02bertrikfunman, actually I forgot which ams sansa target it was ...
19:08:12funmangibbon_: as3525/powermgmt-target.h has a FIXME for Clipv2/Clip+ settings
19:10:16funmanbertrik: I think we should use DBOP on c200v2/fuze/e200v2
19:10:55bertrikwhat target do we need the i2c code for then?
19:11:06funmanc200v2 and clip+ at least
19:11:29funmanbut it could replace the USB_PIN check for all targets
19:11:50gibbon_funman: is there a guideline for documentation of that cycle? the bench tool perhaps?
19:12:19gibbon_or would it suffice to just look what happens when discharging and charging again in rockbox? that would be an easy thing to do
19:12:49bertrikfunman, ok, the i2c code doesn't look that complex
19:14:04bertrikgibbon_, you should see the battery voltage slowly drop over time and have a reasonable value, between 3.4 and 4.25V
19:14:35gibbon_ok... i'll do that and come back with the results
19:15:09funmangibbon_: for finding the full voltage, you could fully charge from the OF, boot rockbox and note the voltage from the battery debug menu
19:15:14bertrikduring charging, you can check if it's still a reasonable value, and it should slowly increase up to 4.2 (no higher!) then stay at 4.2V for some time and finally a slight drop after some time (as the internal charger is switched off)
19:16:09funmanclipv2/clip+ use the battery levels from clipv1 (sansa-clip/powermgmt-clip.c)
19:16:55bertrikfunman, just to be clear, I'm no longer opposing the i2c code in dualboot
19:17:22kugelyea it doesn't look as scary as I feared
19:17:50funmanbertrik: thanks
19:18:42bertrikthat clipv1 battery bench could probably use an update, I think I made that one very early in the port when we were probably not running as efficiently as we are now
19:19:27 Quit efyx_ (Remote host closed the connection)
19:20:24 Join Horscht [0] (~Horscht2@xbmc/user/horscht)
19:20:38 Join JdGordon_ [0] (~Miranda@rockbox/developer/JdGordon)
19:21:08funmani will test the dualboot diff tomorrow hopefully
19:21:43 Quit funman (Quit: free(random());)
19:22:45 Join bluebrother [0] (
19:22:45 Quit bluebrother (Changing host)
19:22:45 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
19:23:54JdGordon_kugel: mini2g with music off (I tihnk) and CF disk
19:24:13JdGordon_boot time is still blazingly fast though so not too worried
19:26:37 Quit bluebroth3r (Ping timeout: 276 seconds)
19:31:01 Join planetbeing_ [0] (
19:37:27gibbon_the battery voltage at least changes over time until now
19:38:53bertrikit's usual to dip a little when something draws more power (e.g. display back light) and recover after that
19:40:35gibbon_would it make sense to adjust the cpu boost to speed up discharge?
19:41:12gibbon_does that even change the actual clock? i always wondered about that
19:41:45bertrikyes, it boosts the CPU clock, other clocks should remain unaffected
19:42:05bertriknot sure if that is already implemented on the clip+/clipv2 though
19:42:43gibbon_at least the clock value increases
19:43:12gibbon_its worth a try i guess
19:43:21bertrikgibbon_, but you are at least seeing reasonable battery values, right? It might take ten minutes or more to see some kind of downwards trend in battery voltage
19:44:04gibbon_bertrik: yes, i do
19:44:52 Quit Schmogel (Ping timeout: 265 seconds)
19:45:01gibbon_in the battery info screen the scale spans around 3.834v and 3.854v and i see a exponentioal (or similar) curve with some deviations
19:45:27gibbon_looks like the beginning of a discharge curve to me
19:46:46gibbon_the current value decreases
19:47:00JdGordon_where do I discuss a possible change to the cabbie theme?
19:47:07JdGordon_here or the forum thread?
19:47:16bertrikThere is no disk write support on the clip+/clipv2 right?
19:47:20gibbon_when i looked earlier, an attached charger was detected
19:47:28gibbon_bertrik: looks like that
19:47:43gibbon_bertrik: then the battery bench plugin would have been the choice, right?
19:48:46bertrikgibbon_, exactly. I think we can basically declare discharge monitoring as done now, and do a battery bench later
19:49:12LloreanJdGordon_: Probably start a new forum thread at this point
19:49:14gibbon_since i am sitting here anyway with nothing to do, i can also keep an eye on it
19:49:31LloreanSince if it's a cross-target change you'd probably need to recruit people, and the old thread is kinda long, cluttered, and irrelevant now.
19:49:34 Join scorche [0] (~scorche@rockbox/administrator/scorche)
19:50:16 Quit Xerion (Read error: Connection reset by peer)
19:50:24bertrikgibbon_, you could also check the charge behaviour, you may have to enable it first in the config file first
19:50:32 Join Schmogel [0] (
19:50:37gevaertsSpeaking of threads (or no), what do we do with 3.5.1?
19:50:51Lloreangevaerts: Is it ready?
19:50:56JdGordon_wtf is 3.5.1?
19:50:59gibbon_bertrik: pre- or post compile?
19:51:10LloreanJdGordon_: It fixes a significant buffering underrun issue
19:51:23gevaertsLlorean: kugel wanted one more backport I think. The skip-to-end-of-track crash
19:51:36JdGordon_oh ok
19:51:48Lloreangevaerts: I think we should backport as little as possible. Is that an easy to trigger/common crash?
19:51:48bertrikgibbon_, oh sorry, it looks like it's already enabled in SVN. Just remember to hold the center button when plugging in the charger
19:51:57 Join Strife89 [0] (~michael@
19:52:04gibbon_bertrik: i will
19:52:06gibbon_thanks so far
19:52:16bertrikthank you for testing :)
19:52:17gevaertsLlorean: FS #11004 IIRC
19:53:22gibbon_bertrik: i cant code that well... i am happy with any little piece i can do :)
19:53:28Lloreangevaerts: Does that even affect any of the release targets?
19:53:42gevaertsLlorean: well, fuze was declared stable
19:55:24LloreanIt sounds like a rather uncommon case. I'm really not sure how I feel about it either way - it seems to me we shouldn't make a habit of backporting fixes and only do it in the most critical cases. The buffering watermark one happens even without any user interaction other than starting playback, right?
19:55:40gevaertsit does, yes
19:55:57gevaertsI don't understand playback well enough to evaluate the risks
19:56:27LloreanCan the crash be worked around by skipping to the end of the song first, then skipping backward? It could just be added to known issues if that works.
19:56:50*gevaerts waits for kugel to comment
19:57:00pixelmabluebrother: while talk clip (and voice generation) doesn't hang on an Intel Mac running 10.4.11, it generates just a noisy burp. And my newly generated deutsch.voice doesn't want to work at all
19:57:17pixelmaVoiceOver sounds fine
19:58:20Lloreangevaerts: Another option is to just push out 3.5.1 and discuss whether that one merits another backport later, rather than waiting on one person?
19:59:11gevaertsWe could, but on the other hand he's online. so he likely will comment soon
20:00:33pixelmabluebrother: (with the dmg you gave for download)
20:01:49 Quit knine (Quit: E.T. go home.)
20:02:47 Join Xerion [0] (
20:05:57gevaertsdomonoky, bluebrother: can rbutil handle different stable revisions for different targets?
20:06:57LloreanDidn't we have a different one for Archoses than everything else a while back?
20:07:22Hillshum3.0 and 3.0.1. Yup
20:07:50gevaertsI'd have scored 10/10 too if I only had easy ones!
20:09:17gibbon_clip+ charger detection works, voltage then is measured as ~4.6V
20:09:58gibbon_battery level is shown as 100% directly after attaching the usb cable (for charging)
20:10:50gibbon_that does not seem right
20:10:53kugelLlorean: that is the crash
20:11:08kugeli.e. if you skip to the end, then back it crashes
20:11:23kugeland it has been reported to happen on other targets too.
20:11:24Lloreankugel: The description says if you seek to near the end, then skip back, it crashes
20:11:45LloreanI asked about skipping. That is, actually ending the song.
20:12:10*JdGordon_ has no problem with crash fixes getting backported
20:12:18kugelwell I meant skipping the near the end (with the skip to outro feature)
20:12:32kugelI think this one should be backported
20:12:44LloreanYou still haven't answered my question, though.
20:13:29JdGordon_kugel: I just tried again and with music off theme load is maybe 1s faster, but even then its only maybe 2s so not really a big deal
20:13:30kugelfull track skips don't crash
20:13:36JdGordon_not sure why I thought it was so bad last night
20:19:47gibbon_bertrik: i will try charging in OF
20:21:15bertrikgibbon_, it probably wasn't actually 4.6V, but rather some kind of an incorrect reading. We've had a similar issue on the e200v1 before, where we can get a normal value from another ADC channel
20:22:26 Join kramer3d_ [0] (~kramer@unaffiliated/kramer3d)
20:22:28gibbon_bertrik: i suspected that 4.6-4.7V where 5V USB supply voltage minus voltage drop because of charge current
20:22:34gibbon_but i am not sure about that
20:22:41 Quit kramer3d (Ping timeout: 256 seconds)
20:22:42bertrikyes, that sounds sensible
20:22:43bluebrothergevaerts: yes.
20:22:55gevaertsgood :)
20:23:05gevaertsI mean, HWCODEC isn't broken
20:23:13JdGordon_shame :p
20:23:21gibbon_i have NO idea about the internal circuits in the device :P
20:23:22JdGordon_i mean...
20:23:36bluebrotherpixelma: ok :/ Does the configuration dialog show the available voices on the system correctly at least?
20:23:36gevaertsJdGordon_: we all know what you mean! :)
20:25:11 Join petur [0] (
20:25:12 Quit petur (Changing host)
20:25:12 Join petur [0] (~peter@rockbox/developer/petur)
20:26:54gibbon_bertrik: the voltage rose to about 4.78v...
20:27:40linuxstbTheSeven: What are you using ucl for?
20:28:02TheSevenlinuxstb: compressing the nano2g nor contents
20:28:13linuxstbTheSeven: I assume you've realised that the ARM assembler version of the decoder is thumb code?
20:28:38linuxstbHow well does it compress?
20:28:59TheSevenslightly better than 1:2 on code, about 1:3 on graphics
20:29:53TheSevenLZMA is another 20% better on both, but it's a lot more complex
20:29:59 Join phanboy4 [0] (
20:31:38 Join planetbeing__ [0] (
20:34:31 Quit planetbeing_ (Ping timeout: 260 seconds)
20:35:25 Join efyx_ [0] (
20:39:21 Quit stoffel (Remote host closed the connection)
20:43:13pixelmabluebrother: yes, the configuration dialogue looked correctly on this Mac
20:43:29JdGordon_domonoky: request for the theme site, can you make it email (if the themer wants it) when it fails checkwps if it used to pass?
20:44:41domonokyuh, should be possible, but i am not motivated to do it at moment :-)
20:44:56*JdGordon_ beats domonoky with a motivation stick
20:46:47 Join Barahir_ [0] (
20:48:30JdGordon_kugel: got any ideas how we can have the value of one token used by another?
20:48:38JdGordon_or anyone else
20:48:53bluebrotherpixelma: have you tried the "Test TTS" button in the configuration? Did it produce static as well?
20:49:07 Quit Barahir (Ping timeout: 240 seconds)
20:49:46 Quit kramer3d_ (Ping timeout: 252 seconds)
20:51:26 Join wodz [0] (
20:51:57kugelJdGordon: some find_token() function maybe?
20:52:16wodzis there somewhere rockbox coding rules or something?
20:53:02*bluebrother wonders if the problem is in the aiff to wav conversion or somewhere else
20:53:04 Quit Barahir_ (Ping timeout: 276 seconds)
20:53:18wodzI mean name conventions, source identation, header format etc.
20:53:53bluebrotherwodz: read docs/CONTRIBUTING
20:53:57 Join FlynDice_ [0] (
20:55:52***Saving seen data "./dancer.seen"
20:56:55wodzWhat about header in source files? this big ascii-art ROCKBOX logo and friends
20:57:47JdGordon_that logo and the (c) notice should be there on new files
20:57:50pixelmabluebrother: yes, I tested it and think it worked correctly but the longer I think about it I get a bit unsure if it was just the VoiceOver (which was enabled still) announcing something of RbUtil. There was no static noise at least, so it worked correctly or it was silent
20:57:51JdGordon_not imported files
20:57:56 Quit FlynDice (Ping timeout: 264 seconds)
20:58:38*pixelma wonders if the VoiceOver running at the same time (as talk clip generation) has a bad effect
20:58:44wodzthats clear, but is there some standard way to produce that header in new files?
20:59:03JdGordon_ctrl-c ctrl-v
21:01:30wodzvi do not have ctrl-c :-) and what about this fancy line (for example) * $Id: i2c-coldfire.c 17847 2008-06-28 18:10:04Z bagder $
21:01:55kugelI think you must reduce that to "$id:$"
21:02:04kugelthe text between comes from subversion
21:02:05JdGordon_thats a special line which svn will fiddle with if you set the svn properties corectly
21:03:47pixelmabluebrother: I could imagine aiff played as wav (or vice versa) sounding this way. The "clips" seem to consist of "words" or what is left over
21:04:24pixelmaI can recognise pauses between "words"
21:04:35JdGordon_is [ and ] special chars in the skin language?
21:04:42JdGordon_i.e do they need % before them?
21:04:57pixelmaJdGordon_: not that I know of
21:05:03wodzNext question: what about derivate work - I mean I took some existing file from other port and tweaked it to work with the new port. Should I preserve original author in header?
21:06:21JdGordon_damn... I was thinking something likr %xda[%mp] to make the xda tag use the value from mp
21:06:45*gevaerts would like a final conclusion about 3.5.1
21:07:25pixelmaJdGordon_: you mean for this auto-expand?
21:07:37JdGordon_more than just autoexpand
21:07:44JdGordon_make it part of the actual language
21:07:51JdGordon_I'm not sure how useful it will be though
21:10:18pixelmamy first reaction was - if at all, I'd prefer a new %x* tag for it. Then it is more obvious that it's not the ususal "display bitmap" and you have to be careful.
21:10:23bluebrotherwodz: you can leave the $Id:$ line as it is. It's contents are replaced by svn if the svn:keywords property is set correctly
21:11:17pixelmaJdGordon_: what will happen in your example if the number of conditionals in the %mp tag changes?
21:11:56wodzbluebrother: and what about author
21:12:18bluebrotherpixelma: I could create you a binary that does leave the intermediate files around (which is aiff and wav). Could you test such a version? I'm suspecting the aiff -> wav conversion to misbehave
21:12:44bluebrotherwodz: that's also an svn keyword, it works the same
21:13:11pixelmabluebrother: sure, maybe tomorrow
21:13:22 Join Lear [0] (chatzilla@rockbox/developer/lear)
21:13:29bluebrothera keyword is always $Keyword: expansion$, and the expansion is replaced by svn. You need to write the Keyword correctly, though, and have it in the svn:keywords property
21:13:34gevaertsbluebrother: not that one I think
21:13:44wodzbluebrother: I was asking if I should preserve original author in header or I should credit him/her
21:13:54bluebrothersvn knows about a keyword Authors
21:14:14gevaertsyes, but I think wodz means the Copyright (c) 20... bit
21:14:29bluebrotherwodz: ah, that author :) If you change an existing file just leave it as is.
21:14:57pixelmaJdGordon_: the number of parts of the bitmap strip is still set through the %xl tag then?
21:15:30JdGordon_pixelma: yes, but I was more going for a generic way to do it, I dont tihnk image display is the only tag which would benefit from this
21:15:42JdGordon_although I cant really think of any others at the moment :/
21:16:17pixelmaanswer before question ;)
21:16:32JdGordon_%xda[%mp] is backwards to the rest of the system though
21:16:42wodzOnce again. I am preparing new port. Of course I reuse as much code as I can so there will be new files but based on some existing ones (tweaked to work with new target). My question is how to treat Author field in header in this case
21:16:56JdGordon_maybe and image autoexpand tag is just the easier way to go
21:17:10pixelmamaking [] special would break some of my WPSs :\
21:18:12*JdGordon_ can live with that though :D
21:20:14pixelmawhy not %xe|%mp| or so?
21:20:15 Join p3tur [0] (~petur@rockbox/developer/petur)
21:21:21pixelmaerr... %xeA|%mp| but you should get the idea
21:22:07JdGordon_because that ciould get confusing
21:22:35JdGordon_%?mp<%xeA> is most liekly the way to go
21:23:30JdGordon_and allowing for %?pv<%xdaa|%xeA|something else>
21:24:18JdGordon_%xeA5 ? (expand A 5 times)?
21:25:03JdGordon_%xeA|4|5 will just be very consufing inside another contiional
21:25:12kugelI'd think it would always expand to the number of strips in the bitmap?
21:25:40JdGordon_to keep it usable, I think it would have to
21:26:06pixelmacan't imagine at the moment what expand x-times would be used for and how that would work
21:26:19*kugel neither
21:28:24*gevaerts wishes he understood buffering.c
21:28:39 Join Kitr88 [0] (
21:28:49JdGordon_gevaerts: rewrite it so yoiu do!
21:29:29gevaertsJdGordon_: the primary goal right now would be to know if kugel's fix for FS #11004 is really safe. Rewriting won't help with that :)
21:30:15kugelwell, it's in for some time and no problems have been reported
21:30:24gevaertsI tend to think that it is, but (a) I don't understand the code, and (b) it's been in svn for a bit over ten days now. That's not enough to declare it "stable because nobody saw a bug"
21:30:48JdGordon_when is it long enough?
21:30:50kugelalso, matsch said I should commit it, and he's a buffering expert (more than anyone of us anyway)
21:31:14gevaertskugel: the "other" issue (the dropout-on-rebuffer thing) was broken in november and only reported after 3.5
21:31:39JdGordon_its only if you enable that wierd skip setting right?
21:31:42 Join planetbeing_ [0] (
21:31:43 Quit Kitar|st (Ping timeout: 258 seconds)
21:31:49kugeldidn't Horscht report it pre3.5?
21:31:57gevaertshm, yes
21:32:05Mode"#rockbox +o Bagder" by ChanServ (ChanServ@services.)
21:32:10Mode"#rockbox -b *!*benlar@193.13.246.*" by Bagder (~daniel@rockbox/developer/bagder)
21:32:31gevaertsJdGordon_: if I understand things correctly, that setting makes reproducing the issue a lot easier. I don't think it actually causes it
21:32:42Mode"#rockbox -o Bagder" by ChanServ (ChanServ@services.)
21:32:50kugelJdGordon_: I haven't been able to reproduce it in another way, but it was a general problem that handles are rewritten during a rebuffer event
21:33:15 Quit Kitr88 (Ping timeout: 258 seconds)
21:33:18gevaertskugel: will you prepare 3.5.2 if it breaks things? ;)
21:33:20 Join Kitar|st [0] (
21:33:56*gevaerts isn't entirely convinced either way, but he tends to think that it should go in
21:35:02 Quit planetbeing__ (Ping timeout: 245 seconds)
21:37:25kugelI'm still wondering why that fix for stuck playback when playing a deleted song caused a glitch
21:40:31gevaertskugel: let's include it. Will you merge 24755 to the 3.5 branch, or shall I?
21:40:57kugeldo it please, I don't have the checkout anymore
21:41:19*gevaerts admits that he still has all the 3.x checkouts
21:42:30 Part domonoky
21:43:00gevaertskugel: it doesn't merge cleanly :(
21:43:15kugelreally? it ought to
21:43:45kugelshall I have a look?
21:44:33 Quit wodz (Quit: CGI:IRC (EOF))
21:45:32gevaertsplease do
21:46:04 Join ecio1 [0] (~ecio@
21:47:13gevaertsIf I also merge r24440 and r24618, svn doesn't complain
21:47:38 Join froggyman [0] (
21:48:05gevaertsI'd really prefer a minimal backport though
21:49:08*gevaerts does it by hand
21:49:27 Quit ecio (Ping timeout: 246 seconds)
21:51:37kugelgevaerts: not sure why it conflicted...
21:52:04CIA-5New commit by 03amiconn (r25004): Get rid of .rept in inline asm() blocks where possible. Using .rept causes gcc to wrongly estimate the size of the asm(), leading to (potential) ...
21:52:09gevaertskugel: r24440 is Torne's alignment patch, and r24618 is your inline ringbuffer functions patch
21:53:25kugelah yea, the inline one is needed I think
21:55:07amiconngevaerts ^^
21:56:00 Quit Strife89 (Quit: Clocking out.)
21:56:44kugelgevaerts: I have backported 24618 and 24755, the diff looks alright
21:57:49gevaertskugel: I have a manual merge of only 24755, but maybe it's safer to go for both. Can you commit?
21:58:48kugelr24618 didn't contain a functional change so it would work either way
21:59:51gevaertsI know. I'd have liked to avoid merging too much, but I think I prefer this to more manual merge work
22:00:05kugelI also needed manual merge work
22:00:58gevaertsyes, but on the other hand they're your patches :)
22:01:08 Quit Adnyxo (Ping timeout: 276 seconds)
22:01:58kugelalright, I'll make a quick test and then commit
22:02:47bluebrotherpixelma: debug binary here:
22:03:52bluebrotherif you press the "Test TTS" button it will keep two intermediate files: the aiff and the wav of the spoken text. Close the configuration dialog after executing the "Test TTS" (give it a couple of seconds to acutally complete the talking), then check the system trace
22:04:28bluebrotherit will contains two lines starting with "WARNING:" somewhere near the end with the filename of the temporary files.
22:05:02bluebrotheryou can try playing the aiff, it should be the text "Rockbox Utility voice test". My suspicion is that the wav one will be broken.
22:05:32bluebrotherif that's the case it would be great if you could send me those two files
22:06:08gibbon_charging completed.
22:07:59gibbon_bertrik: after charging: 4.155v
22:08:53pixelmabluebrother: the system trace?
22:09:35gibbon_i don't know where to go with this information
22:09:45bluebrotherpixelma: no, the wav and aiff file. The system trace shouldn't contain anything that's important for that issue, at least nothing I can think of right now.
22:10:20bluebrotherbut if the wav file is broken and the aiff correct I need to check my aiff -> wav converter. Maybe something is different with the aiff files produced on your machine.
22:11:47pixelmaok, my question was a bit too short, but I was wondering where to find the system trace - do you mean the console on a Mac or something in RbUtil or something else?
22:12:35 Join sirdaniel [0] (
22:13:21bluebrotherpixelma: ah :) In the menu go to Help / Troubleshooting / System Trace.
22:13:32 Join Adnyxo [0] (
22:13:44pixelmathe RbUtil menu?
22:13:52 Quit Lear (Quit: ChatZilla 0.9.86 [Firefox 3.6/20100115144158])
22:14:34bluebrotherthe binary is a debug binary so it should also show that if you run it from a terminal. But that's a bit less convenient ;-)
22:20:15CIA-5New commit by 03amiconn (r25005): libdemac: Separate filtering calls per channel in preparation for the dual-core split on PP. This also means less inlining, and hence speeds up ...
22:22:46 Quit sirdaniel (Quit: ChatZilla 0.9.86 [Firefox 3.6/20100115144158])
22:23:01kugelphew, I almost thought the patch has a regression
22:23:12kugelbut it's a bug in trunk
22:23:19pixelmabluebrother: as I said, I'll try to test tomorrow if I get a chance
22:24:56 Quit Hillshum (Ping timeout: 264 seconds)
22:26:09bluebrotherpixelma: great, thanks for the testing.
22:27:32 Quit phanboy4 (Ping timeout: 245 seconds)
22:33:27kugelgevaerts: ok, I'm ready
22:34:26 Join phanboy4 [0] (
22:40:19CIA-5New commit by 03kugel (r25006): Backport r24618 and r24755 for 3.5.1. ...
22:43:42 Quit planetbeing_ (Quit: planetbeing_)
22:44:44 Quit iq (Quit: leaving)
22:48:04gevaertsbluebrother: does the font pack need to match the release version?
22:48:47JdGordon_fonts arnt target or version dependant
22:48:56 Quit GodEater (Ping timeout: 269 seconds)
22:49:26amiconnThe font format *did* change occasionally in the past though
22:49:47gevaertsI know, but I don't know what rbutil expects
22:50:10 Quit phanboy4 (Read error: Connection reset by peer)
22:50:11 Join GodEater [0] (
22:50:11 Quit GodEater (Changing host)
22:50:11 Join GodEater [0] (~bibble@rockbox/staff/GodEater)
22:50:18bluebrothergevaerts: I don't think so. IIRC rbutil doesn't even handle a "release" font pack
22:50:53 Join phanboy4 [0] (
22:53:01 Quit guymann (Quit: brb)
22:53:39bluebrothersomething that really should get changed. As well as handling checkwps results for theme installation
22:55:55***Saving seen data "./dancer.seen"
22:59:23bertrikgibbon_, looks good
23:00:53*kugel just got a prefetch abort at 0xCOEDBABE
23:01:21 Quit GodEater (Ping timeout: 256 seconds)
23:02:53 Join GodEater [0] (
23:02:53 Quit GodEater (Changing host)
23:02:53 Join GodEater [0] (~bibble@rockbox/staff/GodEater)
23:03:45 Join guymann [0] (
23:06:14gevaertsBagder, Zagor: could you move to the download server? I don't have release notes or an announce ready yet though
23:06:26 Join Tomis [0] (~Tomis@
23:06:45gevaertsIf anyone feels motivated right now (I'm too tired), this fixes FS #11004 and FS #10919
23:07:08gevaertsonly SWCODEC is included
23:08:43*Bagder gets them packages
23:08:54gibbon_bertrik: discharge looks good, too. even though i am irritated by a linear voltage drop on li-ion
23:10:06gibbon_bertrik: 0.1v pro Stunde
23:10:16bertrikoh no, the gradual voltage drop vs. discharge is really the beauty of li-ion/li-poly, it makes it easier to estimate the charge left
23:10:50 Join TMM [0] (~hp@pdpc/supporter/professional/TMM)
23:11:04gibbon_oh, welll ... see how much i know about battery chemistry
23:11:46gibbon_i always thought, li-ion was different
23:12:55gibbon_but you are right about beauty ... its a beautiful linear drop
23:13:08gevaertsBagder: will you also update build-info? I'll send a mail as soon as you tell me it's ready
23:15:26kugeldamn! I'm experiencing FS #11041 in 3.5.1
23:16:10CIA-5New commit by 03peter (r25007): Histogram display on recording screen. Based on the work of Jvo Studer in FS #5021 but reduced and reworked since the recording screen code changed ...
23:16:13gevaertsok, cancel 3.5.1 then for now until it's more clear?
23:16:18kugelI think sop
23:16:42gevaertsBagder: can you hold off updating build-info?
23:17:06BagderI don't even know where that is ;-)
23:17:11gevaertsApart from that, this is useful information :)
23:17:25Bagderah there it is
23:17:40gevaertsAnyway, we can't release with known regressions
23:17:51gevaertsnot for a two-bugfix update anyway
23:19:27kugelI've experienced that once in a current build, but that's long (well, relatively) ago, and the guy that opened FS #11041 doesn't experience it anymore as well
23:19:45gevaertsso something changed between 24755 and now
23:19:53amiconnSeems like PP5020 behaves as predicted wrt IRAM
23:20:10Bagderalso, it'll be better with a full release, ie all release-targets - as we don't really have a good way to do partial releases with our scripts
23:20:11JdGordon_petur: I bet it feels good to finally get that in? (you bastard :D )
23:20:29amiconnMoving small arrays to dram makes things faster (cache is faster than iram), while moving large arrays to dram makes things slower
23:20:40p3turJdGordon_: yeah, took too much time
23:20:45gevaertsBagder: ok. I'll do them all next time
23:20:59JdGordon_p3tur: it uses the peak values right? would it be possible to use it on the wps?
23:21:18p3turbut now we can start splitting the drawing code of the recording screen into parts
23:21:36p3turJdGordon_: I think so
23:23:46gevaertskugel: The earliest release builds I can do are tomorrow evening, so you do have some time :). Worst case we can always revert that patch I guess
23:25:16Bagderand the build-info is generated using the file as input
23:25:33Bagderso it should be sufficient to update that
23:25:53Bagderwww/ that is
23:26:06CIA-5New commit by 03bluebrother (r25008): Code police. ...
23:27:30JdGordon_I can build release builds if needed
23:27:56 Quit kugel (Read error: Operation timed out)
23:28:04 Quit liar (Ping timeout: 245 seconds)
23:32:42gevaertsJdGordon_: there's *slightly* more than building required. First r25006 would have to be reverted :)
23:33:15amiconnp3tur: Why not on H100?
23:33:19gevaertsI'm actually in favour of just reverting that. Whatever fix kugel comes up with, no matter how obviously correct, won't be tested enough I think
23:33:34 Join kugel [0] (~kugel@rockbox/developer/kugel)
23:34:18kugelwell, it looks like it should be reverted
23:34:37kugelthe only changes in that area are yellow fixes in DEBUGF statements, and the playback.c change that's already backported
23:34:41kugeland r24790 but I can't believe that changed/fixed anything
23:35:47gevaertsa bit risky anyway
23:35:57gevaertsso yes, let's revert
23:36:08p3turamiconn: true...
23:36:25p3turbut also other recording targets can enable it
23:36:41p3turI just only did it now for the targets I could test on
23:36:48amiconnProbably not if the screen is too small
23:37:02p3turindeed not
23:37:57*amiconn wonders whether this would work on hwcodec as well
23:38:03amiconnNot that I'd need it...
23:38:05 Quit AlexP (Remote host closed the connection)
23:39:12 Join AlexP [0] (~ap@rockbox/staff/AlexP)
23:39:13 Quit ecio1 (Read error: Connection reset by peer)
23:39:57CIA-5New commit by 03kugel (r25009): Revert previous commit. It appears to introduce a regression (FS #11041) which doesn't happen in trunk (anymore).
23:40:15 Quit bertrik (Quit: De groeten)
23:40:40gevaertsJdGordon_: if you want to build, go ahead!
23:40:51JdGordon_you can :p
23:40:54JdGordon_im at work
23:41:06kugelsvn diff -r PREV was too greedy
23:41:43 Quit ender` (Quit: Don't worry about the world coming to an end today. It's already tomorrow in Australia.)
23:42:04gevaertsJdGordon_: I didn't mean you have to do it right away :) It's just that I don't want to wait for a build anymore tonight
23:42:58 Quit phanboy4 (Read error: Connection reset by peer)
23:42:59kugelit seems to have reverted r24978 as well
23:46:47 Join Barahir [0] (
23:47:09CIA-5New commit by 03kugel (r25010): Backport r24968 again. svn diff -r PREV was too greedy in the previous commit. ...
23:50:30 Nick FlynDice_ is now known as FlynDice (
23:51:07kugelmay I commit the init code patch?
23:51:08 Join planetbeing_ [0] (~planetbei@
23:51:13 Quit linuxstb (Ping timeout: 256 seconds)
23:51:19JdGordon_only if you commit the skin resizing one also!
23:53:41 Join DerPapst [0] (
23:53:52 Quit evilnick_B (Quit: Page closed)

Previous day | Next day