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

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

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

#rockbox log for 2009-03-10

00:00:09bluebrotherpixelma: simply "Danke"?
00:00:19 Quit ibseco (Read error: 113 (No route to host))
00:00:45 Quit arohtar (Client Quit)
00:00:58 Join faemir [0] (
00:01:15pixelmahmm, another possibility
00:03:44 Quit FlynDice (Remote closed the connection)
00:04:15Bagderrasher: mispel setup now on your build host?
00:04:20Bagdermipsel even
00:04:25rasherOh right, yes
00:04:46Bagder5 hosts so far
00:07:12 Quit stripwax ("")
00:09:11MTeelinuxstb : ping
00:15:34ConicQuick question, what's the best device that Rockbox works with that doesn't have a touchwheel, only buttons?
00:15:57n1sConic: see the BuyersGuide page in the wiki
00:16:05ConicAh, thanks.
00:16:25*Bagder does the drumroll to ease the tension waiting for this build round to finish
00:18:20*bluebrother wonders why a new #ifdef in settings.h is required if the build system is building just fine
00:18:44 Join CaptainKewl [0] (
00:18:55ConicHah, it doesn't say anything about scroll wheels.
00:19:07ConicI guess I'll do my own research then.
00:19:19JdGordon|anything except the ipods or sansa e200's
00:19:39JdGordon|whats best isnt really useful because not all of them will still be easily avilable
00:20:05pixelmathe h10 has a touch strip and the M:Robe100 a kind of touch pad
00:20:25ConicThe iRiver h300 looks good
00:20:28*n1s wonders why the manual entry for "Interpret numbers when sorting" does not mention that it changes the way files with spaces in their names are sorted, it's not exactly clear from the name...
00:20:50*Conic enters 11 in configure
00:20:56bluebrothern1s: because some people just discovered that? ;-)
00:21:34ConicI'm going to try to port a simulator to the iPhone.
00:22:00JdGordon|then you proabbly really want the onda or mr500 which expect a touchscreen
00:22:35ConicOh, cool
00:22:39ConicThat makes it even easier
00:22:51ConicI didn't know there was a touchscreen player that worked with Rockbox.
00:22:52JdGordon|except dunno how well it is expected to work at the moment
00:23:10JdGordon|both are not supported fully yet
00:23:17ConicI've been dying to use Rockbox again after my Nano died a while ago. :)
00:23:27n1sbluebrother: surely it should have been documented if it was an intended feature all along?
00:23:33 Quit JdGordon| (" ajax IRC Client")
00:23:41 Quit amiconn (Nick collision from services.)
00:23:42 Join amiconn_ [50] (n=jens@rockbox/developer/amiconn)
00:23:43oboBagder: mips was already on mine (along with the other 3)
00:23:50 Join pixelma_ [50] (n=pixelma@rockbox/staff/pixelma)
00:23:50 Quit pixelma (Nick collision from services.)
00:24:00 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
00:24:03 Nick pixelma_ is now known as pixelma (n=pixelma@rockbox/staff/pixelma)
00:24:28n1sBagder: should i add the mipsel toolchain on my host too?
00:24:34bluebrothern1s: well ...
00:24:41Bagdern1s: please do!
00:24:58 Quit kushalone ("Leaving. I cannot promise to be back but most likely will.")
00:26:45n1sbluebrother: it just seems like a surprise to me that this setting changes more than the number sorting
00:27:06rasherBagder: What are we waiting for?
00:27:14Bagdern1s: agreed, but it turns out it is mostly a documentation error
00:27:25Bagderrasher: something is fishy
00:27:33bluebrothern1s: I didn't expect it to do more than that either. But it seems some(tm) people have different expectations of that and don't consider others having expectations differing from theirs
00:28:03rasherbluebrother: The name was poorly chosen in hindsight.
00:28:14advcomp2019Conic, if you want a cheap player, i will say the sansas might be good if you can get a pp sansas unless you want to wait for ams sansas
00:28:40ConicI'm not buying it.
00:28:54ConicI'm porting a simulator to my iPod touch
00:28:57advcomp2019o ok
00:29:05pixelmabuild hanging?
00:29:07ConicOr trying. I may not get very far. :P
00:30:02 Quit AndyI (Read error: 104 (Connection reset by peer))
00:31:00pixelmawhat's the touch's screen resolution - 320x240?
00:31:15 Join AndyI [0] (i=AndyI@
00:31:21ConicNot sure, I think it has a 4xx somewhere in it
00:31:48ConicAh, 480x320
00:31:59*Bagder was too slow
00:32:14ConicSo, anyone have an Onda?
00:32:21ConicWait, never mind
00:33:25ConicHuh, do these guys even have a websire?
00:33:45 Quit bluebrother ("leaving")
00:33:51Bagderin chinese, sure! ;-)
00:33:51mcuelenaereConic: I do
00:33:55kugelhm, what happened to the build table? :S
00:34:27ConicWhat Onda has the closest screen size to 480x320?
00:34:44mcuelenaereI think almost all of them
00:34:47mcuelenaerethey have 400x320
00:34:53ConicOh, that's good
00:35:02mcuelenaereI'm not sure what the VX767 has
00:35:16ConicI'm building the 474+
00:35:24mcuelenaerebut the VX747, VX747+, VX777 (and probably VX757 and older VX9xxx) have 400x320
00:35:36ConicGah, build errors
00:35:40mcuelenaereConic: the simulator?
00:35:52mcuelenaereI just committed it, so you'll need to do 'svn up'
00:36:21mcuelenaereoh wait, it probably won't work for the vx747+
00:36:26ConicI'm trying to get the iPhone toolchain set up in the same Cygwin install
00:36:34ConicIt downgraded Subversion.
00:36:45mcuelenaeretry the vx747 (the vx747+ doesn't have simulator setup yet)
00:36:49mcuelenaeres/setup/set up/
00:38:11 Quit MTee (Read error: 110 (Connection timed out))
00:39:36 Join gewuerzwiesel [0] (n=gewuerzw@unaffiliated/gewuerzwiesel)
00:40:25rasherBagder: oh dear
00:40:36mcuelenaereVX747+ simulator is now in SVN too
00:40:38***Saving seen data "./dancer.seen"
00:43:16 Join rocko [0] (
00:44:13Bagderhm, my script is missing now...
00:45:44 Quit Conic ()
00:45:59Bagderand so are other ones
00:46:14gewuerzwieselhi, where is it recommended to store the music-directories on the sd card? directely to the root dir, or in a dir called "MUSIC" or so?
00:46:22n1sBagder: ok, worked as well as ever and the path is added to so everything should be fine
00:46:36rasherBagder: Mine's still here..
00:46:42n1sgewuerzwiesel: it doesn't really matter
00:46:49gewuerzwieseln1s: ok, thx
00:47:15gewuerzwieselI had the problem, that after updating the database there were every track about 10 times in the list..
00:47:29Bagdern1s: great
00:47:34mcuelenaeregewuerzwiesel: preferably not in .rockbox/, but otherwise it doesn't care where you store your music
00:48:12gewuerzwieselmcuelenaere: ok, so..that "music" ... structure is the orig.firmware's layout?
00:48:29mcuelenaeregewuerzwiesel: probably yes, what device is this?
00:48:46gewuerzwieseloh sorry.. :) it's an sandisk sansa e250
00:49:38mcuelenaereI'm not familiar with the Sansa's, but if the folders keep re-appearing when you delete them, then probably the original firmware creates them
00:49:52amiconnBagder: Wth, is gone??
00:49:59gewuerzwieselyes, I think so
00:50:13pixelmagewuerzwiesel: yes, and the Sansa's original firmware also sets this directory to hidden (and so it won't be visible in Rockbox with default settings)
00:50:18Bagderamiconn: something with the mipsel stuff clearly
00:50:33amiconnWell, I have no copy to replace it...
00:50:45Bagder is a start
00:52:29*rasher suspects line 17
00:53:15 Join cool_walking_ [0] (i=cb3b81c3@gateway/web/ajax/
00:53:24BagderI did a mistake in the config file so I added a non-existing target
00:53:26amiconnBagder: Your path is wrong (apart from that I have a different base path
00:53:43Bagderoh right
00:53:50amiconnYour mipsel path has an extraneous 'bin' component
00:53:52gewuerzwieselok, so I freshly installed 3.1 the database is building
00:54:17 Join plutonian [0] (
00:54:42plutonianout of curiosty, are bitwise operations on the ipod's architecture faster than addition operations?
00:54:47gewuerzwieselok, that looks better now :) every track is there just one time :) fine
00:55:15mcuelenaereplutonian: AFAIK those are faster on all architectures
00:55:53amiconnadd/sub and bit manipulation are the same speed on all target architectures (single cycle when reg-to-reg)
00:55:58n1sadditions are typically cheap though
00:56:24*mcuelenaere was wrong then
00:56:51*Bagder sighs and restarts the script
00:56:57 Part toffe82
00:57:01amiconnActually I can know for sh, coldfire and arm. Not sure abozt mips
00:57:06plutonianSo basically the idea that bitwise operations are fast has no relevance to any modern computers
00:57:16amiconnBagder: How to fix line 17?
00:57:26 Quit ender` (" Trying to establish voice contact ... please yell into keyboard.")
00:57:31Bagderline 17 is fine as long as I do the right thing in my end
00:57:35 Join planetbeing [0] (
00:58:04*amiconn makes that file mode 0555
00:58:51mcuelenaereplutonian: bitwise shifting should be faster than multiplying (but you can only do this for powers of 2)
00:58:58n1splutonian: that depends on what you do and what you compare to
00:59:38n1sand the cpus in our daps are not very related to modern computers
00:59:54 Part gewuerzwiesel
00:59:57plutonianbut they're still computers
01:00:00plutonianand modern
01:01:32n1sok, so what is it that tou are asking?
01:02:26 Quit tvelocity (Remote closed the connection)
01:03:08 Join fdinel [0] (
01:03:12plutonianI read that addition and bitwise operations are now equal in speed
01:03:26plutonianI wondered if this was true for daps as well
01:03:31plutonianapparently, it is
01:04:18n1syes, that is a property of many cpus this does not mean that all cpus behave the same in different areas
01:06:49amiconnIt's not only a feature of modern CPUs. Even the Z80 needs the same cycle count for bit manipulation and add/sub
01:10:48 Quit n1s ("Lämnar")
01:16:28Bagder97 builds in 601 seconds
01:21:02 Quit mcuelenaere ("Gnight")
01:25:59 Join BHSPitMonkey [0] (n=stephen@unaffiliated/bhspitmonkey)
01:31:47 Join saratogahome [0] (n=41becb3b@gateway/web/cgi-irc/
01:31:50Bagderhm, just making the source archive takes 214 seconds on the build master
01:31:56saratogahomeBagder: I'm building the mips compiler now
01:32:15saratogahomealso, i installed the coldfire compiler a while ago, i'm not sure if i need to tell you that or not
01:32:35Bagderyou did tell me
01:32:47saratogahomeprobably not
01:33:32Bagderwell I have your server marked for m68k builds anyway ;-)
01:33:55*kugel wonders if a build server that isn't alway available would be useful
01:34:55Bagderkugel: sure, at least if it is running often enough to contribute
01:35:40kugelI'm not too sure about that ;)
01:35:51 Quit plutonian ("leaving")
01:38:22saratogahomeBagder: ok done compiling
01:39:20Bagderok, your server is now marked for mipsel builds too
01:39:38saratogahomea lot of the mips warnings seem trivial to fix
01:39:43saratogahomeanyone object to me doing it?
01:40:20*Bagder certainly doesn't
01:40:44kugelthe other problem is my changing ip
01:41:10saratogahometheres dns services you can use for free
01:41:28kugelyea, I use that already
01:41:33rasherBagder: isn't "often enough to contribute" just codeword for "more than 0"?
01:41:49*kugel remembers rasher having slight problems with ddns
01:42:00rasherkugel: Not at all
01:42:16rasherHalf the build servers are on some ddns service
01:43:06Bagderrasher: well, unless you just boot the machine to "nick" a build and then you shut it down again or similar
01:43:59rasherThat's could happen in a number of ways anyway. But yeah, if your usual uptime is less than 30 minutes it's probably not a good idea
01:45:41 Join funman [0] (n=fun@rockbox/developer/funman)
01:45:52Bagdercrap, my mess
01:46:05Bagder(the red)
01:46:10funmangevaerts:;rev2=65 < I looked at the linux source code and it doesn't seem to match the USB registers of AS3525 SoC ?
01:47:23 Quit funman (Client Quit)
01:51:16Bagderamiconn: your new countdown.js is now in use
01:51:40saratogahomeactually i wonder if we can just disable the warning for bitwise operations and parenthesis
01:59:15 Join yhuang [0] (n=yhuang@unaffiliated/yhuang)
02:01:55 Quit planetbeing ()
02:02:10 Quit nuonguy ("This computer has gone to sleep")
02:03:57 Quit gregzx ("ChatZilla 0.9.84 [Firefox 3.0.7/2009021910]")
02:04:33saratogahomeanyone object to -Wno-parentheses to bring MIPS inline with the other targets?
02:08:15 Quit kugel (Nick collision from services.)
02:08:20 Join kugel [0] (n=kugel@rockbox/developer/kugel)
02:08:38 Quit kugel (Nick collision from services.)
02:08:42 Join kugel [0] (n=kugel@rockbox/developer/kugel)
02:08:49 Quit kugel (Remote closed the connection)
02:09:07 Join kugel [0] (n=kugel@rockbox/developer/kugel)
02:23:27 Join jfc^3 [0] (
02:24:46 Join fyrestorm_ [0] (
02:24:49 Join Lss__ [0] (
02:24:49 Quit jfc (Read error: 60 (Operation timed out))
02:29:13 Join tmzt_ [0] (
02:29:35 Join blithe_ [0] (
02:30:21 Quit tmzt (Nick collision from services.)
02:30:36 Quit fyrestorm (Read error: 60 (Operation timed out))
02:30:41 Nick tmzt_ is now known as tmzt (
02:30:42 Nick fyrestorm_ is now known as fyrestorm (
02:32:42 Quit blithe (Read error: 104 (Connection reset by peer))
02:34:11 Join akur [0] (
02:34:35 Part akur
02:35:53 Join kkurbjun [0] (
02:36:16 Quit Riku (Read error: 104 (Connection reset by peer))
02:40:42***Saving seen data "./dancer.seen"
02:47:48kugelsaratogahome: hehe, adding the paranthesis was cleaner :p
02:48:03kugelI think our other compilers do warn on those
02:48:13 Quit efyx (Remote closed the connection)
02:48:43saratogahomekugel: the other compilers don't warn on those
02:49:04kugelare you sure? I got those warnings just today. amiconn recently too
02:49:15saratogahomea lot of them were in the codecs
02:50:37kugelsaratogahome: I don't see any -Wno-parentheses for the other compiler
02:50:50 Quit fyrestorm ("ChatZilla 0.9.84 [Firefox 3.0.7/2009021910]")
02:51:18saratogahomekugel: its not apart of -Wall for older gcc versions
02:51:40kugelI see
02:51:50kugelbut I definitely had those warnings
02:52:02kugelmaybe 4.1.x is more sensitive about them
02:52:10saratogahomefor targets or sims?
02:52:44kugelgood question. I had it in a sim today. But IIRC amiconn had them in a normal build (they appeared on the build server)
02:53:44saratogahomehmm build system is backed up
02:53:58saratogahomei'm not sure then
02:54:04saratogahomeIMO disabling them makes the most sense
02:54:17saratogahomei'd rather not dig through codecs just to satisfy silly warnings
03:01:01 Join nuonguy [0] (
03:05:52 Join timc [0] (n=aoeu@
03:05:52jhulstBagder: I saw your mail about the build server and just finished setting mine up at
03:06:28 Join LynuxDevil [0] (n=travis@unaffiliated/ferret0567/x-681468)
03:06:49LynuxDevilwhat is the deal with the iPod Nano 4G signed firmware?
03:06:59saratogahomeit doesn't run rockbox
03:07:10saratogahomeis anyone around with a MIPS target?
03:07:21LynuxDevilI know that
03:07:45LynuxDevilwhat encryption does it use? nobody knows?
03:08:32LynuxDevilor is that confidential?
03:08:36 Quit Alpha-Bait ()
03:09:01saratogahomeI don't think anyone knows for sure
03:09:04LynuxDevilok then, please tell me this, if you can
03:09:36LynuxDevilcan I create a new partition on the iPod and use it to put a LiveUSB of Linux onto it?
03:10:01LloreanLynuxDevil: That really has nothing to do with Rockbox.
03:10:04LynuxDevilalso, does this iPod support GPT (GUID) partition tables?
03:10:15LynuxDevilLlorean, well, I don't think anybody else would know
03:10:33saratogahomeand we don't either
03:10:34LloreanLynuxDevil: That doesn't make this the right place to ask. Especially since we clearly state where to take off-topic questions in our channel topic.
03:10:35LynuxDevilthis channel might know the answers
03:11:25 Quit MethoS- (Remote closed the connection)
03:11:42saratogahomethink anyone will get mad at me for blind adding MIPS asm opt to libmad?
03:12:04krazykit`probably not unless it cocks something up :)
03:12:06LloreanWe don't have any MIPS supported targets yet, so if you break something, it's not so bad.
03:12:31saratogahomewiki says sound doesn't work, so theres probably no harm in it
03:18:32 Quit midijunkie (Read error: 104 (Connection reset by peer))
03:29:53 Quit Aurix_Lexico (Remote closed the connection)
03:29:55saratogahomeBagder: [for the logs] watching the build process it seems to use relatively little CPU time on my quad processor system
03:30:15saratogahomeperhaps an easy way to speed up the build system would be to hand out pairs of build jobs and run them in parallel
03:41:36 Quit LynuxDevil (Read error: 104 (Connection reset by peer))
03:45:00 Quit rocko ("Leaving")
03:46:05kugelsaratogahome: I don't think that makes it much faster, clients can build with -j already
03:46:26saratogahomekugel: yeah but a lot of the time is still spent waiting on disk, network and svn
03:47:05kugelare you running with -j?
03:48:25kugelwhich client are you?
03:49:29 Quit nuonguy ("Leaving")
03:49:34kugelhm, rank 2, not bad :p
03:50:16kugelgiving paralell builds would also allow doing the next build, while the old .zip is still uploading?
03:51:09saratogahomei don't see why it couldn't just create two build dirs, and run them fully in parallel
03:59:18kugelsaratogahome: funman is receving a (your?) fuze?
04:02:45 Join blkhawk- [0] (
04:07:09saratogahomekugel: yeah i sent it a week ago
04:12:52 Quit lark-__ (Read error: 145 (Connection timed out))
04:18:52 Quit blkhawk (Read error: 113 (No route to host))
04:19:42 Nick blkhawk- is now known as blkhawk (
04:20:44kugelsaratogahome: somehow, your commits freed 30k ram usage on some devices
04:21:23kugelor was that the DM320 commit?
04:21:30saratogahomekugel: the dm320
04:21:42saratogahomethe build system was backed up so our commits got merged into one common build
04:21:45kugeloh, I didn't know the creatives have that too
04:22:56 Join lark-__ [0] (
04:23:32kugeltheir ram usage is still insane, compared to the other targets
04:25:20 Join AndyIL [0] (i=AndyI@
04:27:52 Quit kugel ("ChatZilla 0.9.84-rdmsoft [XULRunner 1.9/2009020407]")
04:30:21 Quit fdinel ("Miranda IM! Smaller, Faster, Easier.")
04:31:24 Quit saratogahome ("CGI:IRC (EOF)")
04:36:40 Quit AndyI (Read error: 110 (Connection timed out))
04:40:44***Saving seen data "./dancer.seen"
04:41:03 Quit BHSPitMonkey (Read error: 113 (No route to host))
04:46:54 Quit miepchen^schlaf (Read error: 101 (Network is unreachable))
04:49:25 Join Llorean1 [0] (
04:50:24 Quit Llorean (Nick collision from services.)
04:50:26 Nick Llorean1 is now known as Llorean (
04:54:05 Join Martyn_ [0] (n=martinb@
05:00:16 Join blkhawk- [0] (
05:06:47 Quit lark-__ (Read error: 145 (Connection timed out))
05:09:59 Join pondlife1 [0] (
05:10:42 Quit blkhawk (Read error: 104 (Connection reset by peer))
05:11:14 Nick blkhawk- is now known as blkhawk (
05:14:18 Join gartral [0] (
05:14:26 Part gartral
05:23:57 Join avis [0] (n=ident@pdpc/supporter/student/avis)
05:30:17 Join Darksair [0] (n=user@
05:32:26 Join AndyI [0] (i=AndyI@
05:32:27 Quit AndyIL (Read error: 104 (Connection reset by peer))
05:32:30 Quit Horscht ("Verlassend")
05:34:59 Join SoapGuam [0] (n=7937fcb9@gateway/web/cgi-irc/
05:35:26 Quit dmb (Remote closed the connection)
05:37:22 Join rocko [0] (
05:39:40 Quit SoapGuam (Client Quit)
05:44:16 Quit perrikwp (" ajax IRC Client")
05:44:20 Join perrikwp [0] (i=18ac0c41@gateway/web/ajax/
06:04:55 Nick Lss__ is now known as Lss (
06:07:49 Join BHSPitMonkey [0] (n=stephen@unaffiliated/bhspitmonkey)
06:10:51 Join LennyK [0] (n=cf97f730@gateway/web/cgi-irc/
06:24:50 Quit rocko ("Leaving")
06:25:03 Join rocko [0] (
06:28:59__BradCsaratoga, watching the build process, it appears that building the dependencies is what takes the most wall clock time, and that appears to be quite a bottleneck. Perhaps there is a way to streamline that somehow to speed up the builds
06:32:34 Quit scorche (Nick collision from services.)
06:33:21 Join scorche [50] (n=scorche@rockbox/administrator/scorche)
06:35:14 Quit CaptainKewl (Read error: 110 (Connection timed out))
06:36:33 Join midgey [0] (n=tjross@
06:40:48***Saving seen data "./dancer.seen"
06:51:58 Quit Martyn_ ()
06:57:18 Join homielowe [0] (
06:59:35 Quit BHSPitMonkey (Remote closed the connection)
07:09:21 Quit LennyK ("CGI:IRC (Ping timeout)")
07:12:27 Quit rocko ("Leaving")
07:21:17 Join jordoex_ [0] (
07:41:01 Quit Lss ()
07:42:11 Quit jordoex (Read error: 110 (Connection timed out))
07:47:41 Join nuonguy [0] (
08:04:21 Quit crwl (Read error: 110 (Connection timed out))
08:06:59 Quit cool_walking_ (" ajax IRC Client")
08:09:53 Join avis- [0] (
08:10:01 Quit avis (Read error: 145 (Connection timed out))
08:10:19lostlogicI think that building with #undef HAVE_TAGCACHE is broken again :(
08:10:34lostlogicsomething about the new LANG_SORT_INTERPRET_* stuff
08:11:55JdGordonsif not use the db :p
08:12:03JdGordonanyone know if the archos rec charges over usb?
08:12:21JdGordonas if...
08:12:22scorcheJdGordon: it doesnt
08:13:12scorchelostlogic: i actually might have some time this weeke to have a look at that =P
08:13:25lostlogichaha, I'm trying to figure it out ATM
08:13:35lostlogicI do still know how to code, you know :-P
08:13:44*scorche wants aware
08:13:48scorches/wants wasnt
08:15:57 Join lark-__ [0] (
08:17:11lostlogic'course ATM, I don't see what they would have to do with each other at all, so maybe I'm wrong... but that's the only thing "special" about my config
08:19:03lostlogic/home/lostlogic/rockbox/apps/settings_list.c:796: error: 'LANG_SORT_INTERPRET_NUMBERS' undeclared here (not in a function)
08:22:35 Join crwl [0] (
08:27:00 Quit perrikwp (" ajax IRC Client")
08:27:22 Join perrikwp [0] (i=18ac0c41@gateway/web/ajax/
08:27:46lostlogicugh, I think it may just have been some kind of weird timestamping issue that didn't regen lang.h
08:29:33 Join ender` [0] (
08:31:01 Quit avis- (Read error: 110 (Connection timed out))
08:37:02 Join einhirn [0] (
08:40:52***Saving seen data "./dancer.seen"
08:41:41 Join tvelocity [0] (
08:43:07 Nick JdGordon is now known as JdGordon|zzz (
08:49:06 Join Zagor [242] (n=bjorn@rockbox/developer/Zagor)
08:49:29 Join Rob2223 [0] (
08:53:19 Join Bagderr [241] (n=daniel@rockbox/developer/bagder)
08:53:52 Quit jordoex_ (Remote closed the connection)
08:58:50 Join draft [0] (n=draft@
08:58:58drafthow do i reboot iPod Mini?
09:01:05scorche|shas in a hard or soft reboot? a note, this sort of thing is covered in the manual..
09:03:15 Join FlynDice [0] (
09:06:58Zagorhmm? cat: /tmp/rbdev-build/summary-*: No such file or directory
09:07:02 Quit FlynDice ("ChatZilla 0.9.84 [Firefox 3.0.7/2009021910]")
09:07:27 Quit Rob2222 (Read error: 110 (Connection timed out))
09:07:45ZagorBagder: lillebror and lillebror2 are now mipsel ready
09:08:47draftscorche|sh: i mean normal iPod firmware
09:08:54 Quit kharo ("Leaving.")
09:09:26 Join MTee [0] (n=MTee@
09:10:34drafthow do i reboot original iPod Mini with original firmware?
09:13:21BagderrZagor: nice
09:13:32The-Compilerdraft: hold Menu and Select for some secs
09:13:33BagderrZagor: I noticed that summary thing too, we should fix...
09:14:21Bagderrnow 12 build servers are mipsel-ready
09:14:22draftThe-Compiler: it doesn't seem to reboot
09:14:43The-Compileris hold on?
09:15:40draftit booted
09:15:44draftthanks mates
09:15:55Bagderrbuild server coverage: sh 26, arm 28, m68k 28, mipsel 12, sdl 23
09:16:05 Quit midgey ()
09:16:33Bagderrthere are 28 ones in my list
09:17:00The-Compilerhuh why are there Bagderr and Bagder?
09:17:06 Nick Bagderr is now known as B4gder (n=daniel@rockbox/developer/bagder)
09:17:19B4gderbecause I don't know who I am!
09:18:25 Join midijunkie [0] (
09:19:11pondlife1What no B4gderr ?
09:19:14 Join stripwax [0] (
09:19:41B4gderthat's the human who runs me, but he's hardly every arond. I'm a grown-up bot now
09:19:53The-Compileras long it's not 84Gd3Rr1337 :D
09:20:58 Quit midijunkie (Client Quit)
09:21:03 Quit tessarakt ("Client exiting")
09:21:57*pondlife1 checks that FS #10000 passed without the End Of The Universe occurring.
09:22:10 Join midijunkie [0] (
09:22:20pondlife1Nope, still here
09:22:27 Join n1s [0] (n=n1s@rockbox/developer/n1s)
09:25:18 Join Thundercloud [0] (
09:25:30 Quit stripwax ("")
09:26:00 Join stripwax [0] (
09:26:36 Quit stripwax (Client Quit)
09:26:56scorcheI have added a few things here...any opinions?
09:27:50 Quit midijunkie (Read error: 104 (Connection reset by peer))
09:28:03 Join stripwax [0] (
09:32:26 Join GodEater [0] (n=ge@rockbox/staff/GodEater)
09:32:28 Quit draft ("Lost terminal")
09:34:06 Join Lss [0] (
09:34:44MTeelinuxstb : ping
09:35:34 Quit GodEater (Client Quit)
09:35:49 Join nibbler_ [0] (
09:36:36linuxstbMTee: Hi.
09:38:41MTeelinuxstb : Hello ! Yesterday I was able to write the audio data to an output file, the sample was mp3 in rm container, but it sounded crappy when I played it, could this be due to the encryption you told me about ?
09:40:43linuxstbI'm not sure - maybe that "encryption" was just with Cook streams. If there was the xor'ing going on, I would expect the audio to be unrecognisable.
09:41:25amiconnZagor: I observed a few quirks in the build system: (1) Why is rbspeexenc relinked everytime when building voices? (2) Why is rbspeexenc checked at all when building hwcodec voices?
09:42:16Zagoras I recall the voice building is still an "old style" build, i.e. without full dependencies
09:42:47amiconnrbspeexenc is in tools/
09:43:03MTeeweird things happen during playback, besides the audio being unrecognizable (there is sound, but gibberish), the duration of playback keeps changing, and it stops after about 8 seconds
09:44:35amiconnI'm not referring to the fact that 'make voice' always starts make in rbspeexenc, but the latter relinks every time, even if rbspeexenc is up to date.
09:45:27Zagoryes, it does that if the target does not have full dependencies
09:45:28 Quit Darksair ("People who are zhuangbility want to show their niubility but only reflect their shability.")
09:49:18linuxstbMTee: What are you using to play the mp3 file? "madplay" (with the -v option) can display information about the errors.
09:50:23MTeeok will try that
09:50:41linuxstbMTee: Or (armed with a copy of the description of the mpeg audio frame header), look at it in a hex editor and try and see if anything is wrong (i.e. packets truncated, or extra data inserted between packets)
09:51:58 Quit stripwax (Read error: 110 (Connection timed out))
09:52:37MTeelinuxstb : if madplay fails to report errors, I'll try the other option.
09:52:48linuxstbMTee: What test file are you using? It it one from the mplayer collection of audio samples?
09:53:04linuxstbI'm sure madplay will report errors, but I don't know how helpful those errors will be...
09:54:05MTeelinuxstb : yes, the one from mplayer collection, it's "I want candy"
09:56:08linuxstbYes, I'm just looking at that now ( ) - the readme talks about "mp3 data in adu format". I've no idea what that means...
09:57:01 Quit Thundercloud (Remote closed the connection)
09:57:36linuxstbApparently it's rfc3119 -
09:59:26linuxstbSo it seems it's not just an mp3 stream in a realmedia container - it reorganises the actual mpeg audio data...
10:01:43 Quit kachna (Read error: 113 (No route to host))
10:02:12MTeebut if a player is able to play the rm one it should be able to play the output audio data no matter what, correct ?
10:03:11linuxstbNot necessarily. It may only support it when it's inside the rm container.
10:06:00 Join avis- [0] (
10:06:53linuxstbMaybe it's worth trying a different format to mp3 - e.g. AAC:
10:07:20 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
10:07:52MTeelinuxstb : here's the output of madplay
10:08:16MTeeok will try that
10:10:13 Quit lark-__ (Read error: 145 (Connection timed out))
10:10:29linuxstbI've just tried "mplayer -dumpaudio iwantcandy.mp3.rm" and then "madplay -v stream.dump", and get identical output to your pastebin. So it seems your demuxing is correct.
10:11:10*MTee is happy
10:11:16linuxstbI tried the same with one of the aac files (using mplayer to play stream.dump), and it won't play it... So it seems there is always something extra going on...
10:12:24linuxstbSo I'm not sure what I would do next. Maybe start actually extracting one of the Real codecs from ffmpeg, and extending your demuxer to actually decode the audio, and write a WAV file.
10:13:46MTeeI thought I should do that. I believe cook is the one I should try ?
10:14:39linuxstbI think that's the most useful, yes. Although I think it's still floating-point only, so there will be work needed to convert it to fixed-point before it will work in Rockbox. But the first step will just be a simple standalone C program that converts Cook to WAV.
10:16:43linuxstbMTee: One thing you can do is to compare the output of your demuxer with "mplayer -dumpstream", to see if it is in fact identical.
10:20:35MTeeOk. one final question : what are the chances of continuing this as a SoC project ?
10:21:22 Quit homielowe ()
10:22:45linuxstbI've no idea - Rockbox hasn't even been accepted for SoC 2009 yet...
10:22:48n1si think extended codec support is always a good thing so i would give it a fairly high chance
10:23:12gevaertsWe haven't even *applied* yet (although we plan to)
10:23:29n1s(provided that we are accepted as a mentoring org, etc.. what they said ^ ;))
10:23:29linuxstbBut it will depend on the quality of the other students, and what the other projects are.
10:24:11linuxstbBut the fact that you've been around before the start of SoC and have been making progress on the project is a good thing.
10:25:18linuxstbBut the nice thing about realaudio is that it supports many codecs - so there will be lots of potential work, even if you manage to get Cook working before the summer.
10:25:41linuxstb(but I doubt you will get a fully optimised fixed-point Cook decoder working before the start of SoC)
10:27:08linuxstbMTee: BTW, do you own any Rockbox devices yet?
10:27:29MTeeyeah, me too. I'm supposed to study A LOT for school :)
10:27:57MTeeoh yes, I have owned it for about 2 years
10:28:06linuxstbGetting to the stage of a standalone C decoder using floating-point would be a nice basis for a port to Rockbox though.
10:28:41linuxstbi.e. it proves you're capable of doing something, and makes the actual project much easier.
10:29:22*linuxstb is reminded that we need to think of qualification tasks for this year's SoC, assuming we want them...
10:29:47MTeeok, I'll try working on it as much as I can
10:30:00*scorche has started up #rockbox-gsoc again...those who can, try and join?
10:30:22n1slinuxstb: what did they do last year? write and compile a simple plugin or something?
10:30:25 Join Bagderr [0] (n=daniel@gateway/web/cgi-irc/
10:30:49linuxstbn1s: I can't remember. Yes, I think one task was a .xpm viewer or something like that.
10:30:58linuxstb(i.e. a very simple bitmap format)
10:31:09 Quit saratoga (Read error: 104 (Connection reset by peer))
10:31:33 Quit LinusN (Read error: 60 (Operation timed out))
10:31:47 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
10:32:01 Quit B4gder (Nick collision from services.)
10:32:05 Nick Bagderr is now known as B4gder (n=daniel@gateway/web/cgi-irc/
10:33:12 Quit nuonguy ("This computer has gone to sleep")
10:37:33MTeeAbout the Arabic translation , it would be really hard to find translations for some of the technical terms (e.g : PCM wave). Also strings which are composed of numbers only should not be counted as missing strings in the translation.
10:38:43*linuxstb refers the above question to rasher...
10:39:40n1sMTee: if technical terms are not commonly translated, it is probably best to leave them as english
10:40:54***Saving seen data "./dancer.seen"
10:41:25B4gdergevaerts: did you spot funman's comment from tonight on the usb?
10:41:31B4gderon as3525
10:42:03linuxstbHmm, bad news then?
10:45:10gevaertsB4gder: I saw it. I'd like to see what differences he found before drawing conclusions though
10:45:21B4gderyeah me too
11:04:02rasherMTee: It's up to the person doing the translation really. Some translations leave these as-is, other (at least Danish) try to use native terms, even when they aren't very quite as commonly used.
11:06:12rasherI don't think it's something there needs to be a common Rockbox policy on.
11:08:22MTeerasher : ok. what about numerical strings being counted as missing in the translation ? (e.g : 19200 in Arabic would still be 19200)
11:10:12rasherMTee: Just "translate" them by not doing anything
11:10:37rasherMTee: The tool doesn't do anything to notice that they're numbers
11:13:14pixelmaisn't there a difference (different characters) in east arabian and west arabian digits?
11:16:55MTeepixelma : unless you mean Iran by east arabian, then no there's no difference.
11:17:17MTeeIranian digits looks a lot like arabic
11:17:28pixelmano, I remember different digits in Egypt
11:18:02MTeeI'm Egyptian :)
11:18:51MTeeanyway I have to go now
11:19:02*pixelma guesses a misunderstanding
11:19:30pixelmabut it's not really important, in case it's needed another lang can easily be added
11:20:08B4gderI hear tuxpaint has 80 languages, we need to step up a notch!
11:25:13 Join PaulJam [0] (
11:29:25B4gder15 build servers now grok mips
11:30:17B4gderwith only 2 mips builds I think we're pretty well covered already ;-)
11:45:04 Quit avis- (Remote closed the connection)
12:00:51 Quit PaulJam (".")
12:02:58MTeepixelma : There's a very old debate about the origin of arabic digits, some people say that latin numbers are the original arabic digits,others say it's hindi numbers that are so. You could actually find formal arabic documents with latin numbers and others with hindi. but hindi is the wide-spread one that most of the people believe it's actually arabic. That's what I think would be close to...
12:03:00MTee...what you're saying
12:08:28 Join webguest30 [0] (n=4fb51393@gateway/web/cgi-irc/
12:08:37 Quit webguest30 (Client Quit)
12:11:02MTeerasher : the first two columns is what I'm talking about, both are used, but the middle one (Egypt/Mideast) is the one used the most. the right-most column (Urdu/Persian) is not used in arabic-speaking countries
12:14:58MTeefrom wikipedia : "..Eastern Arabic numerals (used in Iran, Afghanistan, Pakistan and India) ". Arabic countries are 22 which do not include those 4, the article is a bit inaccurate.
12:18:31rasherI'm not really fussed, use whichever numerals seems most natural to you, and we'll see if anyone complains and take it from there
12:27:58pixelmaMTee: I meant the difference between the first two (as a European the first is what I think of as arabic - and how I read your remark about the bitrate number). I though in some arabic countries it's also common to use those, not the middle one as it seems to be typical for Egypt. But as I said it's not that important
12:29:23MTeeI see
12:31:31MTeein recording settings, what is AGC short for ?
12:32:40 Join avis [0] (n=ident@pdpc/supporter/student/avis)
12:34:48pixelmaautomatic gain control
12:40:58***Saving seen data "./dancer.seen"
12:45:12 Join kachna [0] (
12:47:37MTee<All tracks> when translated would mean either <All files> or <All songs> which is better ?
12:48:45linuxstbI would say "files"
12:51:39pixelmaI would expect "songs" as <All tracks> is currently used in the database browser and the database currently only indexes music files. Hmm, thinking about audio books though...
12:52:05linuxstbPlus classical music
12:52:50linuxstbOr even any kind of instrumental music.
12:54:03MTee"files" it is.
12:56:17 Join robin0800 [0] (
13:00:46 Quit Rob2223 ()
13:01:36 Quit faemir (Nick collision from services.)
13:01:59 Join arohtar [0] (
13:02:23 Join __lifeless [0] (n=lifeless@
13:02:25 Quit _lifeless (Remote closed the connection)
13:03:57 Join kushalone [0] (n=kushal@
13:04:33 Quit Lss (Read error: 131 (Connection reset by peer))
13:04:38 Join Riku [0] (
13:12:34 Join Rob2222 [0] (
13:21:36 Join cuser [0] (n=8d48731a@gateway/web/cgi-irc/
13:21:36 Quit robin0800 (Read error: 104 (Connection reset by peer))
13:23:03 Join robin0800 [0] (
13:23:20cuserhi all
13:23:38cuseris this player compatible with rockbox?
13:24:04linuxstbOnly the pages listed at the top of are compatible
13:24:11linuxstbI meant players listed
13:24:43gevaertsThere's work going on for the clip, but it's not done yet, and it's not ready enough to be generally usable
13:25:29cuserhmm ok
13:25:36cuserthx for the answer
13:25:49pixelmaisn't it even a "v2" Clip? I'm a bit lost in Sandisk's versions but it looks a bit unusual
13:25:50cuserhow long do you thing it takes you to get it running on the sansa?
13:26:03B4gderpixelma: there is, yes
13:27:51cuserlinuxstb: you are right with the list, but i didn't know if (Sansa c200, e200 and e200R series) is probably a sansa clip... that's why i asked
13:27:53pixelmaI mean the one in his link
13:28:47cuserpixelma: so you are developing the rockbox for the "v1" of the sansa clip, right?
13:30:42pixelmasome people are (I'm not) and I don't know how big the difference between the v1 and v2 Clips is, so whether the v2s of it would be a completely new port or not
13:31:31pixelmaabout the how long: noone can tell
13:32:14robin0800any one having problems with rockbox usb on linux?
13:34:23cuserpixelma: thank you for your help. bye!
13:39:57 Join CaptainKewl [0] (
13:41:11 Quit cuser ("CGI:IRC")
13:42:38gevaertsrobin0800: not me anyway
13:42:44linuxstbrobin0800: Someone reported Rockbox USB problems (with an e200) on all operating systems - transfers basically doesn't work for him...
13:43:30linuxstbgevaerts: Did you see that Martyn said that he got the problems on Linux as well - so it doesn't work on all three OSes, which he was running on 3 different computers. He also said he tried two different cables...
13:43:59 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
13:49:37 Quit perrikwp (Remote closed the connection)
13:49:37 Quit evilnick (Remote closed the connection)
13:49:56 Join LambdaCalculus37 [0] (n=rmenes@rockbox/staff/LambdaCalculus37)
13:50:51robin0800linuxstb: can't get jaunty alpha 5 to see c240 but quite happy with the OF usb wonder what the difference is
13:51:09linuxstbAh, I think someone else mentioned the alpha Ubuntu was unhappy.
13:51:25linuxstbOr maybe you...
13:51:57 Join perrikwp [0] (i=18ac0c41@gateway/web/ajax/
13:52:23robin0800linuxstb: the funny thing is it works in kde on alpha5 but not gnome
13:52:52*linuxstb checks the logs and sees it was robin0800 who mentioned this before... ;)
13:54:01robin0800linuxstb: would like to file a bug but not certain it is a rockbox error
13:55:40gevaertsrobin0800: if it works in kde but not gnome, I'm pretty sure it's not a rockbox bug...
13:56:26linuxstbrobin0800: So does the disk device (i.e. /dev/sdX) always appear, but only KDE will automount it?
13:57:35linuxstbrobin0800: Also, have you used your c240 in MTP mode in Gnome in the past?
13:57:42 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
14:01:01robin0800linuxstb: no only msc
14:02:07 Quit CaptainKewl (Read error: 54 (Connection reset by peer))
14:02:39 Join tyfoo [0] (
14:02:41robin0800linuxstb: dosn't get as far as dev, ill pastbin syslog
14:04:28linuxstbThat doesn't make sense to me - I wouldn't think the GUI would affect the presence (or not) of kernel device nodes...
14:06:33 Quit kushalone ("Leaving. I cannot promise to be back but most likely will.")
14:08:39LambdaCalculus37I need a little info on the GoGear port; tools/configure lists it as HDD1630/HDD1830, but does it also double for the HDD6330?
14:09:15B4gderthere was activity on the 6330 with lowlight but I didn't catch the end outcome
14:09:46B4gderthere was some minor diff only
14:10:43robin0800linuxstb: paste bin as promised
14:11:11LambdaCalculus37B4gder: I guess we'll have to pester lowlight later. :)
14:11:24B4gderrobin0800: normally you'd get more info later on if you dmesg again. I take it nothing further was logged?
14:12:01robin0800B4gder: correct it just stops there
14:12:15LambdaCalculus37I was thinking that there would be runtime detection of LCD and touchpad between the 16x0/63x0, since they're mostly the same internally save for LCD and touchpad.
14:12:22B4gderLambdaCalculus37: or grep the irc logs and see what the discussion went like last
14:12:46LambdaCalculus37B4gder: I'll check out the logs.
14:13:21 Join CaptainKewl [0] (
14:13:56robin0800linuxstb: I was thinking perhaps theres a gnome mount
14:14:48 Join kps [0] (n=40eb617d@gateway/web/cgi-irc/
14:15:47B4gder"<low_light> toffe82: good. There's a great deal of similarity between the hdd1630 & hdd6330. The lcd controller is different. I have a simple driver if you want to test it."
14:16:09LambdaCalculus37B4gder: Thanks!
14:16:12B4gder(Feb 05 20:23:01)
14:18:11linuxstbrobin0800: Maybe, I've no idea how involved Gnome/KDE get with kernel things nowadays...
14:21:48 Join midijunkie [0] (
14:22:20robin0800linuxstb: if it works the next line is....... usb-storage: device scan complete
14:23:33B4gderon my e200 that takes ~5 seconds
14:24:05linuxstbrobin0800: This bug report could be relevant -
14:24:11linuxstbThere's a suggested solution at the very end.
14:24:39 Quit CaptainKewl (Remote closed the connection)
14:28:29pixelmainteresting bug report
14:30:31LambdaCalculus37pixelma: Indeed.
14:30:37*LambdaCalculus37 goes to find his c250
14:31:13LambdaCalculus37pixelma: Can you reproduce that behavior from that forum post?
14:31:55 Join pyro_maniac [0] (
14:34:23 Quit kps ("CGI:IRC (EOF)")
14:34:44robin0800B4gder: tried bug solution don't have the file mentioned so created it it makes no difference
14:35:02*LambdaCalculus37 updates his c250's build
14:35:44 Quit Nico_P (Remote closed the connection)
14:36:21 Quit crwl (Read error: 110 (Connection timed out))
14:38:56pixelmaLambdaCalculus37: yes, reproducible
14:39:20pixelmaI mean that the radio keeps playing
14:39:22 Quit LambdaCalculus37 (Read error: 104 (Connection reset by peer))
14:39:39 Part B4gder
14:40:31 Join LambdaCalculus37 [0] (n=rmenes@rockbox/staff/LambdaCalculus37)
14:40:59***Saving seen data "./dancer.seen"
14:41:02pixelmaand when plugging USB from the recording screen the statusbar stays the recording screen one
14:41:14LambdaCalculus37pixelma: Sorry, had a kernel panic.
14:41:19*LambdaCalculus37 goes to see what pixelma said
14:43:23pixelmabut it doesn't seem to stay in recording mode (nothing to hear when monitoring and when I unplug USB again, I am put into the menu - only then the statusbar updates fully)
14:43:59amiconnStatusbar updates are still severely bugged
14:44:57pixelmatrue, saw a nice effect when entering the menu in the "star" plugin yesterday
14:45:02LambdaCalculus37Yes. I notice that when switching from the WPS to the menu, the very top portion of the WPS screen is still drawn on top of the statusbar.
14:46:45 Join Big_D [0] (n=d4f0b4d2@gateway/web/cgi-irc/
14:47:00pixelmaReally? I know this from before the recent statusbar changes (entering the file browser from the WPS), now the statusbar seems to be redrawn before the rest of the screen
14:47:02robin0800linuxstb: after about 50 connects and disconects usb is reconised I think its a timing problem
14:48:26amiconnStatusbar isn't redrawn for quite a while when entering the USB screen (most onticeable on non-RTC targets
14:48:37 Quit nibbler_ ("Ex-Chat")
14:49:00LambdaCalculus37amiconn: Happens on RTC targets, too.
14:49:12*LambdaCalculus37 just noticed it right now on his e280
14:51:00 Join FlynDice [0] (
14:51:30LambdaCalculus37Also on my c250.
14:52:28 Join kugel [0] (n=kugel@rockbox/developer/kugel)
14:53:47kugelBagder: I think you can add
15:05:42*robin0800 should every thing be stoped on usb conection ie play fm video etc?
15:06:06LambdaCalculus37robin0800: I think that's the desired behavior.
15:07:41robin0800LambdaCalculus37: yes unless new bootloader which connects from off
15:08:06LambdaCalculus37robin0800: Ahh, thanks for reminding me!
15:08:15*LambdaCalculus37 has to test the new bootloaders
15:09:03linuxstbI thought FM radio continued during USB? (at least I seem to remember my h140 doing that, I could be wrong...)
15:10:37robin0800linuxstb: not sure what usb stops I asume with new bootloader ethery nthing is off?
15:11:02robin0800evert thing
15:11:48robin0800every if i could only spell
15:12:10 Quit parafin (Read error: 104 (Connection reset by peer))
15:12:17*LambdaCalculus37 turns of the Dyslexia Ray so robin0800 can type correctly :)
15:17:08 Join parafin [0] (
15:19:06pixelmahmm, my Ondio keeps playing radio too when connecting USB. I usually stop everything, so never noticed.
15:19:58Big_DDoes everyone use linux with RockBox or are there some windows users? IF so, do you manage your music with the usual file management, or is there a piece of software you use?
15:20:28LambdaCalculus37Big_D: Everyone here uses whatever they feel comfortable with.
15:20:41LambdaCalculus37There's no right or wrong way of managing your music.
15:20:57pixelmalinuxstb: although I would have thought that with software USB things could be different...
15:21:12 Quit kachna (Connection timed out)
15:22:49Big_DOK, thanks. I think I will start by giving Amarok a try on a linux box.
15:22:55robin0800Big_D: sync toy in windows
15:23:23pixelmahaha, the radio stops playing after disconnecting my Ondio from USB and I'm back in the main menu
15:23:41Big_DAh, yes. Sync Toy, I'd forgotten about that one
15:23:58Big_DThanks Robin0800
15:24:38pixelmasame on the c200
15:25:02LambdaCalculus37pixelma: Same on my e200.
15:25:16pixelmaat least it's consitent ;)
15:26:42LambdaCalculus37What about other targets with FM radio? Do the exhibit the same behavior?
15:26:54*LambdaCalculus37 carnt spel
15:27:03pixelmanor read ;)
15:28:41pixelmathe the Ondio is the only other one I can test though
15:29:24pyro_maniaccan somebody help me with logf()? i tried to get FS #9253 to work but i have no device where rockbox runs on and i have no more idea to debug it.
15:30:41 Join midgey [0] (n=tjross@
15:34:15pyro_maniacthe file was correctly generated but i got no access to it
15:36:26kugelpyro_maniac: you have no device where rockbox runs on?
15:37:21pyro_maniacno, i am currently on porting the Samsung YH 920
15:37:31pyro_maniaci have no other device yet
15:38:31 Join kachna [0] (
15:39:01 Join nibbler_ [0] (
15:39:55 Join toffe82 [0] (n=chatzill@
15:40:26kugelpyro_maniac: how are you doing logf then?
15:42:53pyro_maniaci modified a h10 5g build which is partly running on my device. and an that i applied the patch:
15:49:54 Join jgarvey [0] (
15:51:25kugelpyro_maniac: really? haha
15:51:47kugelwhat means partly?
15:52:02pyro_maniacthe usb stack is functional
15:52:16 Quit robin0800 (Read error: 110 (Connection timed out))
15:52:23kugelbut no display or sound or buttons?
15:52:57pyro_maniacwith some help og gevaerts i got a serial conntection via usb
15:53:35pyro_maniacbut on the last tries my system freezed so i found that patch and tried it on my code
15:53:54 Join robin0800 [0] (
15:54:45*robin0800 could this patch FS #8663 - Data corruption on usb write on sansa (SD driver bugs) be closed now?
15:56:13 Join jaykay [0] (
15:57:12jaykaypixelma: am i allowed to discuss with you about "dank" and "dank an"?
15:58:15kugellinuxstb: what happened to FS #9948?
15:58:45lostlogicis it "normal" to still be getting DT_TEXTREL related warnings on sim builds? They go away if I stick -fPIC on my simgcc options...
15:59:04pixelmajaykay: depends ;)
15:59:04 Quit Big_D ("CGI:IRC (EOF)")
16:00:09jaykaypixelma: i thought credits is now more or less "eingedeutscht"... if it MUST be really german i would prefer mitwirkende
16:04:35 Quit nibbler_ (Read error: 110 (Connection timed out))
16:05:42pixelmahmm, well. First: I don't think everything really must be German but try to avoid Denglisch if I can (and it doesn't sound too unusal or too artificial). Then, "Mitwirkende" could work there too, maybe get some other opinions
16:06:15linuxstbkugel: It needs some work to make similar changes to other lang files (so translators don't get unnecessary conflicts). I haven't had chance to do that yet...
16:07:35kugelpixelma: "Mitwirkende" is way better. I also like "Ehrungungen" (which the translators on the tracker propose) better
16:10:17pixelma"Mitwirkende" yes, but "Ehrungen"? And could people search in the tracker before posting new patches? ;)
16:10:54 Quit freqmod_gq ("No Ping reply in 30 seconds.")
16:10:55lostlogicBagder: I can't seem to replicate the problems that y'all were having with my build server a while back −− got mipsel compiler setup if you want to put in the list (also updated and cleaned up some messes in rbclient's home dir)
16:10:57jaykaypixelma: it seems you chose an old patch, the translation for the speakers is not in you commit
16:10:58 Join freqmod_gq [0] (
16:11:18lostlogicI think we have a build dependency breakage −− logo isn't built early enough under heavy parallel build on my macbook for ipod video simulator
16:11:52pixelmajaykay: yeah, saw that too now
16:12:26jaykaypixelma: and LANG_CREDITS is there two times
16:13:02jaykayand theres a blank line.... and a ===== and >>>>>
16:13:10jaykayshould i make a patch? ;)
16:15:30pixelmano, the last things you said are not in SVN (and I am especially sure about the >> thing as I searched in the text file before commit)
16:15:59*jaykay does svn revert -R
16:16:38jaykaysorry, you are right
16:19:53 Part LinusN
16:22:58 Quit yhuang ("Leaving")
16:23:08 Quit robin0800 (Read error: 104 (Connection reset by peer))
16:29:52pixelmahmm, rasher's script catches a weirdness in a english.lang voice string's features as it is different in the german lang file (where it's not that weird). The phrase is deprecated though...
16:30:49pixelmagenlang doesn't care
16:31:04 Quit kachna (Remote closed the connection)
16:31:38rasherpixelma: manual inspection is always better :)
16:32:21amiconnZagor: Is there a reason why 'make reconf' rebuids dependencies? It's annoying, especially in a build dir not configured for ordinary builds (e.g. voice)
16:33:09pixelmarasher: I just ask myself what to do with it - I lean towards just keep on ignoring
16:33:11amiconnI have it disabled locally, is there anything that would make this undesirable in svn?
16:33:15Zagorthe reason is that a new configure may cause new dependencies (for example if you change from target to sim)
16:34:02amiconn'reconf' cannot change target or build type though
16:34:45Zagoryou're right. looks like a bug then.
16:35:20amiconnI added 'reconf' to the list of goals in line 32 of root.make
16:35:26amiconnI think this is correct?
16:37:07pixelmaok, I try my luck once more and ask about opinions of German speakers about "Mitwirkende", "Ehrungen" maybe still "Dank" as translation of "Credits"
16:39:40 Nick JdGordon|zzz is now known as JdGordon (
16:40:56 Join robin0800 [0] (
16:41:03***Saving seen data "./dancer.seen"
16:44:55 Part pondlife1
16:49:43 Quit Martyn ("Ex-Chat")
17:01:10 Quit Zagor ("Don't panic")
17:01:27 Quit _jhMikeS_ (Nick collision from services.)
17:01:31 Join jhMikeS [50] (n=jethead7@rockbox/developer/jhMikeS)
17:05:41ScothHey folks. Got rockbox running nicely on a 5G 60g iPod Video, but I find if I try to hook it up to USB after running a game/app/plugin, I get the "USB Device Not Recognized" error. Works fine after a reboot of the iPod until I load another app. Found a bug on the wiki about it not being recognized at all, but it looked like it was fixed
17:06:34linuxstbThat's consistent for every plugin you've tried? Have you tried many?
17:07:04ScothEvery one I've tried so far, from Doom to Pacman to lamp
17:07:26ScothThe iPod itself seems to go into disk mode (I get the picture of the USB plug) and it charges, but Winders doesn't recognize it
17:07:43pyro_maniackugel: what would be the best way for my code? should i post my patch under FS #9253 on the tracker or should i ask later again?
17:08:33 Quit perrikwp (" ajax IRC Client")
17:08:55gevaertsScoth: which revision are you using?
17:09:30ScothLessee... r20205-090304
17:09:52ScothDownloaded it a week ago or so, I think
17:12:30gevaertsScoth: can you try with the current version and if it's still not fixed file a bug report? I'll test later today to see if I can reproduce this
17:13:43ScothSure. Might be later before I can get it done, but I'll do it
17:14:28gevaertsThe USB code hasn'e changed in any possibly related way since that revision, but plugin code might...
17:15:41jaykayenglish.lang says "The target names used for target-specific strings are set in the ARCHOS variable in the root Makefile"...
17:15:47jaykayis this outdated?
17:18:29jaykayand in the header of the lang files are some things like "if you find errors, contact me" or last update-date or emails from some translators....
17:18:34jaykaycan they be removed?
17:20:21rasherjaykay: No, ARCHOS is still used. And why do you want to remove those?
17:21:27rasherjaykay: Alright, I take one of those back. ARCHOS should be replaced by MODELNAME
17:21:34jaykaybut it says below "The target names used are picked from the configure script and are set in the MODELNAME variable in the root makefile."
17:22:52jaykayand removing of those "additional information" is just for.... consistency?
17:23:20rasherConsistency with what? Where else would you put that information?
17:23:47rasherThe header is there for a reason.
17:23:58 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
17:24:25jaykaynowhere, i would delete them. consistency with the other lang files.
17:24:34jaykaybut ok, ill leave them there....
17:24:50rasherThank you.
17:27:50 Join saratoga [0] (n=41becb3b@gateway/web/cgi-irc/
17:29:43saratogagevaerts: I had three consective hard freezes in rockbox USB mode while copyign files on my Vista box
17:30:22LambdaCalculus37saratoga: I had a kernel panic earlier when I was copying something to my c250 in Rockbox USB mode.
17:30:26saratogathe transfer would fail, the the devices would show up as being empty in Windows, and if I unplugged they'd unmount in Windows but rockbox wouldn be completely deadlocked
17:30:26 Join _Auron_ [0] (
17:30:49saratogaI have no issues in XP or LInux though, and the OF works fine on all
17:30:54LambdaCalculus37Same here.
17:31:04LambdaCalculus37saratoga: Which device did that to you?
17:32:01saratogae200v1 8GB
17:32:11saratogai was copying to a 2GB SD card if that matters
17:32:34 Quit midijunkie (Read error: 101 (Network is unreachable))
17:32:38 Join jaykay_ [0] (
17:32:40saratogait would typically freeze before it had copied more then 10-15MB
17:33:19LambdaCalculus37I think it does, because that's how I had the kernel panic occur; I was copying about 30MB worth of data to the card. It finished, but my Mac paniced after I ejected the drive.
17:35:13*gevaerts doesn't have all those issues :(
17:35:43saratogais there some way to get debug info out of Windows?
17:35:47LambdaCalculus37gevaerts: That's because you wrote it, so it knows who its master is. :)
17:35:56rashergevaerts: Maybe you should start using Vista like a real man
17:38:23gevaertssaratoga: if you have enough backups, could you try reformatting the e200 and seeing if that changes anything?
17:38:31*gevaerts switches to voodoo debugging
17:42:38LambdaCalculus37Copying files to the internal memory is working just fine on my e280.
17:43:21Scothgevaerts: Hmm, I downloaded the latest rockbox off the site and seem to have ended up with a lower version that before O_o
17:43:52ScothWhen I plug it in, it drops into the Apple "Do not disconnect" screen
17:43:54 Join yhuang [0] (n=yhuang@unaffiliated/yhuang)
17:44:50rasherAm I the only one who wants to gouge my eyes out with a spork due to all the bold text on the WhyRockbox page?
17:45:08gevaertsScoth: did you install the latest release, or the current build?
17:45:10 Join kushalone [0] (n=kushal@
17:45:41saratogagevaerts: format the sd card or the whole memory?
17:46:00saratogai could do either, but does the main memory matter if i was writing to sd
17:46:05ScothAh, here we go. 'tis what I get for doing installs in different places without paying attention
17:46:28ScothLet's try this again...
17:46:29 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
17:46:47gevaertssaratoga: I'd try both, with tests in between (i.e. first sd, and if it still happens both)
17:49:31 Quit jaykay (Read error: 110 (Connection timed out))
17:50:46Scothgevaerts: Yep, still does it. Registering for the bug reporting db now :)
17:52:30 Nick jaykay_ is now known as jaykay (
17:52:44saratogaScoth: what version do you have installed
17:54:11ScothJust now d/led and installed it
17:56:33saratogagevaerts: i had it fail again, this time without crashing, coudl this be a bus reset issue?
17:58:44gevaertssaratoga: well, yes, but bus resets are just a symptom, not a cause...
18:00:18 Quit timc (Read error: 131 (Connection reset by peer))
18:00:30 Join timc [0] (n=aoeu@
18:01:20ScothHuh. It seems to be working now that I've undocked my computer and plugged straight into the computer. I wonder if it's an issue with some USB hubs
18:05:57 Join nuonguy [0] (
18:10:16 Part pyro_maniac ("Leaving.")
18:14:41 Quit LambdaCalculus37 ("back in a bit")
18:17:47 Join sharperguy [0] (
18:18:09sharperguyWas there ever a fix for the DC offset problem in the e200's?
18:23:41 Join midijunkie [0] (
18:24:48 Join faemir [0] (
18:24:49 Quit arohtar (Read error: 104 (Connection reset by peer))
18:25:13robin0800LambdaCalculus37 same here except on alpha5 and gnome but after 50 or so connections and disconnections it sometimes works
18:25:23robin0800LambdaCalculus37 same here except on alpha5 and gnome but after 50 or so connections and disconnections it sometimes works
18:25:43 Join planetbeing [0] (
18:25:49 Join LambdaCalculus37 [0] (n=rmenes@rockbox/staff/LambdaCalculus37)
18:27:41LambdaCalculus37Damn... I copied nearly 7GB worth of data to the internal memory of my e280, and it had caused my Mac to start going ballistic.
18:28:53*linuxstb starts to wonder if usb in 3.2 is sensible...
18:29:06 Join planetbeing_ [0] (
18:36:22 Join itcheg [0] (i=41d59de2@gateway/web/ajax/
18:36:33 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
18:37:12jaykaypixelma: if you still care... is for the speaker<−−>credits change....
18:37:20FlynDiceDoes anyone know the status of the mmu on the as3525 in the ams sansas? My understanding is that it's not currently working. I see the icache and dcache enabled for the bootloader but not the main build.
18:37:25jaykaythen you could also add the newline at the end :)
18:38:41pixelmajaykay: already did that myself and went with "Lautsprecher einschalten" from the other patch
18:39:33 Join MethoS- [0] (
18:39:36pixelmaand the second hunk is only there in patches made with rasher's site (and really useless) ;)
18:40:49saratogaFlynDice: kugel was pretty sure we don't have it setup correctly
18:41:05***Saving seen data "./dancer.seen"
18:41:10rashermcuelenaere: I've created a new theme site that's more or less at the same stage as the one in SVN, but (in my opinion of course) easier to make sense of.
18:41:31mcuelenaererasher: yes, I've seen it. nice job!
18:41:36 Quit FlynDice (Remote closed the connection)
18:41:41rashermcuelenaere: So you think it's worth replacing the current code?
18:42:06mcuelenaeredefinitely. it looks much better as source code (haven't tried it though)
18:42:14rasherDidn't want to do it without some second opinion because it might just be me..
18:42:31mcuelenaeredoes it have all the existing features?
18:43:12rasherI'm not entirely sure - haven't actually tried out the current one much
18:43:33mcuelenaeredo you have a demo site set up?
18:43:35 Quit planetbeing (Read error: 110 (Connection timed out))
18:43:53rasherHang on, I'll boot it
18:44:07 Join FlynDice [0] (
18:44:31 Join miepchen^schlaf [0] (
18:44:43 Join bertrik [0] (
18:45:13 Join crwl [0] (
18:46:12bertriksharperguy, what DC offset problem?
18:46:56rasherThe templates are rather rough around the edges, but I was hoping someone else could fix that
18:47:49mcuelenaerehmm there isn't any example target data?
18:48:05rasherI had some..
18:48:35rasherlooks like I deleted that
18:49:24FlynDicesaratoga: Sorry, hit the wrong button after updating.... Is kugel the one to talk to then?
18:50:20linuxstbrasher: How are you storing the data for the themes site? i.e. text files, MySQL, ... ?
18:50:32 Quit kushalone ("Leaving. I cannot promise to be back but most likely will.")
18:50:35mcuelenaerelinuxstb: SQLite AFAICS
18:50:58linuxstbAh, that doesn't surprise me with rasher ;)
18:50:59rashersqlite, but should be relatively easy to change if we want a "real" db
18:51:41linuxstbWell, scorche's server has mysql running (it's the forums server), so that's available. sqlite should be fine though.
18:51:51linuxstbThat's assuming you're not wanting to host it yourself?
18:52:14rasherI think the plan is to use scorche's server
18:52:30mcuelenaereWasn't there going to be a link between the forums and the theme site? (for developers login)
18:53:03saratogaFlynDice: that or search the logs if hes not around
18:53:07linuxstbI recall some discussion about sharing logins with something, although I can't remember the outcome...
18:53:17saratogai remember him talking to one of the gigabeat people [they use the same ARM MMU] about it a while ago
18:53:45 Quit LambdaCalculus37 ("Fwump")
18:54:30mcuelenaereFlynDice: I think he said in order to enable the I- & D-cache, the MMU needs to be enabled; and he didn't know how to do any of them
18:54:38FlynDicesaratoga: yes I was looking at that in the code wondering why we couldn't use that.
18:55:04ScothI don't suppose anyone in here is familiar with the ASAP port for playing .sap files?
18:55:10FlynDicemcuelenaere: It says you can use the icache without the mmu but not the dcache
18:55:26rashermcuelenaere: there's a theme uploaded now..
18:55:33rashertest/test for the admin area
18:55:42*rasher braces for vandalism
18:55:49mcuelenaereFlynDice: I'm not sure what he specifically said, I just remember that what I said :)
18:55:56mcuelenaererasher: ok
18:56:04*gevaerts already spots vandalism! Someone removed his targets
18:56:35mcuelenaeredid someone already change the admin password? :)
18:57:10rasherThe theme upload is fairly comprehensive. It checks the zip structure, runs checkwps with current and release version and errors if none of them work
18:57:25rasherThen it requires the user to confirm the upload by clicking a link in an email
18:57:30rashermcuelenaere: That shouldn't be possible
18:57:32linuxstbrasher: Is your "Themes" menu on the left just temporary, or is that how you see it working?
18:57:43mcuelenaererasher: then test/test doesn't work?
18:57:52mcuelenaereor I'm doing something wrong..
18:58:01rashermcuelenaere: no, it doesn't work..
18:58:13rasherNow it should
18:58:21rasherForgot to re-add when I re-created the db.
18:58:40gevaertsrasher: is shortname meant to be same as modelname in configure?
18:58:43rasherlinuxstb: That was just to have *something* there. I don't really have any opinion how it should work
18:59:00rashergevaerts: Yes - it's used when running checkwps
18:59:27mcuelenaereok it seems similar, only the admin panel seems to be missing some things (not sure whether it's important)
18:59:32saratogaFlynDice: the gigabeat code probably needs minor changes due to IRAM
18:59:49linuxstbrasher: What's the git command I need to download your themes source code?
19:00:46rasherlinuxstb: git clone git://
19:00:49rasherI *think*
19:01:12rashermcuelenaere: It currently doesn't mail people when hiding their themes - I think the current one does that? What else is missing
19:01:29mcuelenaererasher: no, current doesn't do any mailing at all AFAIK
19:01:43rasherAh, not sure where I picked that up
19:01:46mcuelenaereit's missing the 'edit themes' feature in the admin
19:02:04mcuelenaere+ it doesn't show the zip contents when approving a theme (minor)
19:02:05rasherAh, to modify theme details?
19:02:15FlynDicesaratoga: what's the best way to search the logs, grep the daylies or are there monthlies somewhere?
19:02:19mcuelenaerefor both to-be-added themes and already-existant themes
19:02:35mcuelenaere(you could do the first with the latter though)
19:02:51rasherAll seems like can-do features
19:03:12mcuelenaererasher:"> test:test
19:03:31rashermcuelenaere: Can't connect
19:03:37mcuelenaereyeah, me neither
19:03:39*mcuelenaere fixes
19:05:25 Join DerPapst [0] (
19:05:31DerPapsto/ :)
19:05:33gevaertsrasher: is the text field in the admin pages meant for admin comments?
19:05:40rashergevaerts: Yep
19:05:43saratogaFlynDice: I juse use the site search
19:05:55 Quit robin0800 (Remote closed the connection)
19:06:01rasherPerhaps there should be an irc-specific google-box
19:06:42mcuelenaererasher:"> test:test
19:06:48DerPapstrasher: while you're at it the target information should be editable as well. eg. the ipod video target has a 13bit display by accident :P
19:07:29rasherI was hoping admins wouldn't be quite that spastic and having it editable by editing the db manually would be enough, but I guess not
19:08:09gevaertsrasher: that only works with proper admins, not with test/test ones :P
19:09:02DerPapstwell, depends, if all admins have direct access to the database. but i guess not and someday someone will make a mistake.
19:09:02rasherShould be easy to fix of course
19:09:36*rasher scribbles on the TODO
19:10:39*DerPapst is doing similar suff for weeks now and no end in sight
19:10:57kugelFlynDice: the dcache is not enabled
19:11:11kugelicache is. but dcache can't be active without mmu
19:11:33kugelI tried setting up the mmu, but it's not trivial due to several factors (like our RAM mapping)
19:11:39 Quit midgey ()
19:12:20kugelin theory, the code could be more or less copy&pasted from F/X (nearly same CPU), but those do not use IRAM at all + they map the RAM differently
19:12:35FlynDicekugel: That's what it looked like to me. Is there a reason we can't use the gigabeat stuff in mmu-arm.c?
19:12:46*rasher pushes a few changes to make the site actually work
19:12:48kugelFlynDice: I'm sure we can
19:14:15 Join dabujo [0] (
19:14:20FlynDicekugel: Do we actually get the icache enabled through the bootloader right now, an I understanding that correctly?
19:14:21kugeljust calling the stuff doesn't help though (I tried, my fuze didn't boot). See F/X's crt0.S what that does. As it stands, it seems to be a problem that we have IRAM mapped to 0x0 and DRAM to 0x30000000. I'm not sure how to change that though without mmu
19:14:35kugelyes, and it's still active in the main build
19:15:45DerPapstrasher: and you should close that pre tag after the todo list :D
19:16:45FlynDicekugel: I tried that too... Does all this happen in system-as3525.c and crt0.s?
19:17:28rasherDerPapst: I already said the templates were rough
19:17:35rasherBut yes, obvious typo
19:18:26DerPapstwow.. i didn't know smarty has this debug console thingy :D
19:18:39bertrikkugel, do you have a link to the as3525 MMU documentation?
19:18:49bertrikor document name
19:19:10saratogai think its just the generic arm922T docs from
19:19:23kugelstandard arm922t datasheet
19:19:39rasherDerPapst: It's occasionally useful. I discovered it rather by accident
19:19:54bertrikkugel, oh I though you already read about it
19:20:52kugelbertrik: I did
19:21:08kugelthat's why I know that the dcache can't be active without mmu
19:22:11FlynDicekugel: how do we know that the icache is active?
19:22:22rasherIf anyone wants push access on that git repo just poke me. Or just commit the thing to SVN if enough agree that it's an improvement over the current code
19:22:26saratogai'll add that to the wiki
19:23:16kugel"You must only enable the DCache when the MMU is enabled."
19:23:17DerPapstrasher: pretty nifty. i'm reading about it right now. until now all i did was print_r debugging ;)
19:23:43 Join Thundercloud [0] (
19:25:15kugelFlynDice: I commented the activation out, rockbox was horribly slow
19:26:27 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
19:31:07saratogagevaerts: formatting did not help the USB issue on my sansa
19:33:30gevaertssaratoga: could you enable logf for usb_storage.c and see what it says when failing?
19:36:57 Quit mcuelenaere ()
19:38:08saratogagevaerts: is that as easy as configuring a log build, or do I need to edit code?
19:39:31 Quit midijunkie ("?(???~•~)?")
19:39:38rashersaratoga: I think you need to #include <logf.h> in usb_storage.c then enable logf in configure
19:42:47 Join MethoS-- [0] (
19:43:25 Join Wraith| [0] (
19:45:00 Quit dabujo ("( :: NoNameScript 4.2 :: )")
19:47:09 Join bluebrother [0] (n=dom@rockbox/developer/bluebrother)
19:54:01 Quit MethoS- (Read error: 110 (Connection timed out))
19:56:22 Quit amiconn (Nick collision from services.)
19:56:23 Join amiconn_ [50] (n=jens@rockbox/developer/amiconn)
19:56:31 Join pixelma_ [50] (n=pixelma@rockbox/staff/pixelma)
19:56:31 Quit pixelma (Nick collision from services.)
19:56:41 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
19:56:42gevaertssaratoga: uncomment the #include <logf.h> in usb_storage.c, and do a logf build
19:56:44 Nick pixelma_ is now known as pixelma (n=pixelma@rockbox/staff/pixelma)
19:59:23saratogagevaerts: done. now what? just transfer files until it happens?
19:59:43gevaertssaratoga: yes, and then do a logf dump from the debug menu
20:02:42saratogagevaerts: i've done that ow, but it doesn't look like it caputured anything interesting
20:02:55gevaertscan you pastebin it?
20:03:59saratogaexactly 1000 entires, is there a limit on how much it can record?
20:04:37gevaertsyes, 1000 entries :) Anyway it's a ring buffer, so the latest should be there
20:04:56 Quit MTee (Read error: 110 (Connection timed out))
20:05:55saratogagevaerts: that particular time i just copied files from the sansa until it disconnected in order to avoid locking up rockbox and losing the log
20:06:30gevaertscan you uncomment the logf on line 991?
20:06:49gevaertshm, maybe also the one on line 1102
20:07:19 Join petur [50] (n=petur@rockbox/developer/petur)
20:08:57 Quit MethoS-- (Read error: 113 (No route to host))
20:10:05 Join Reptile211 [0] (
20:11:10 Quit yhuang ("Leaving")
20:13:54saratogafor what its worth, the device probably remains in USB mode for 20 seconds between when it stops the transfer and when I get it unplugged, could the entries be overwritten by then?
20:14:14 Quit DerPapst (Read error: 110 (Connection timed out))
20:16:01gevaertssaratoga: no. This is the last one...
20:16:15gevaertsThere should be a CSW, but it's not sent apparently
20:16:59saratogahave other people tested vista?
20:17:29gevaertssaratoga: maybe just uncomment all commented logfs
20:17:38 Quit planetbeing_ ()
20:18:08gevaertsfrom what I can see now, this doesn't seem vista related
20:19:37 Quit kugel (Read error: 110 (Connection timed out))
20:20:07amiconnEh, what? "error: invalid suffix "pbbb5" on integer constant". In a file I didn't even touch!
20:21:34 Join Conic [0] (
20:22:06 Join kugel [0] (n=kugel@rockbox/developer/kugel)
20:25:00FlynDicekugel:I've been reading over the logs from mid Jan when you were discussing mmu issues with amiconn & jhMikeS. Did you ever try leaving the cache alone in the bootloader and then initing the mmu in the main program?
20:25:32 Quit kugel (Remote closed the connection)
20:25:50 Join ibseco [0] (
20:25:56saratogathis time explorer became unresponsive so i'm going to try again and see if i can duplicate the exact sort of failure from the last logf
20:26:47lostlogicit's too bad we don't have enough build servers to build each target twice and ignore it if the error/warning count doesn't match (or request a tie breaker build)
20:27:50saratogagevaerts: heres a repeat with all logfs enabled, and the same sort of failure i had previously:
20:29:45gevaertsok, so it's a transfer that never completes...
20:29:58Lloreanlostlogic: But we do have enough to build it a second time elsewhere only when there are errors or warnings.
20:30:09ConicOh, wow, you guys finally corrected the little Archos thing with the simulator.
20:30:25ConicNot that it makes any difference, but it's a nice little change.
20:31:06 Quit Tristan ("changing servers")
20:31:21gevaertssaratoga: can you uncomment the #define LOGF_ENABLE in firmware/target/arm/usb-drv-arc.c as well?
20:31:32amiconnLlorean: Hmm, that would mean if there are any legitimate warnings/errors, all builds could be handed out twice in the worst case
20:31:43saratogagevaerts: sure, do i also need to enable some logs?
20:31:48saratogauncomment some logfs
20:31:52 Join Tristan [0] (
20:32:21amiconnUnfortunately the build system can't be made smart enough to distinguish build server problems from legitimate errors
20:32:25gevaertssaratoga: might as well. Just add them all in
20:32:56lostlogicyeah, that's the challenge with having different people on different OSs with different compilers all operating build servers
20:33:24amiconnIt might be possible to detect the (somewhat frequent) "no space left on device" problem, but even that may be tricky given that error messages are localized on linux
20:33:46amiconnPerhaps we should make (or the server side) set LANG=C before building?
20:35:27 Quit Tristan (Client Quit)
20:35:53Lloreanamiconn: Maybe some comparison of *where* the errors occur. Like if 2+ servers get errors in the same line, that one's ignored for the purposes of repeating a build?
20:36:19ConicWhy doesn't the simulator build a .rockbox directory?
20:36:25ConicDo I need to do an extra step?
20:36:42saratogamake install
20:38:09ConicWait, are any apps even ported to the Onda?
20:38:37BigBambiyou mean plugins?
20:39:03ConicBecause this simulator doesn't have any, when I click Games I get "No Files"
20:39:28ConicI'm building a 747, if that matters.
20:40:38saratogagevaerts: [kind where explorer hangs]
20:41:09***Saving seen data "./dancer.seen"
20:41:37saratogahere is the kind where it simply disconnects mid transfer:
20:41:39 Quit killan_ ("( :: NoNameScript 4.22 :: )")
20:41:43LloreanConic: You can check the source to see if it's supposed to build plugins. Generally though, it's hit and miss as to what's done with incomplete ports.
20:44:05 Join Tristan [0] (
20:46:01gevaertssaratoga: are there changes in your tree other than these logf things?
20:46:08 Join planetbeing [0] (
20:46:12*gevaerts doesn't like this problem
20:46:44saratogagevaerts: no, and I get it with official build anyway
20:46:50linuxstbConic: As you discovered, plugins are currently disabled for the 747. So it needs someone to fix them - which normally just means assigning a keymap.
20:47:01 Join akur [0] (
20:47:04 Part akur
20:47:21pixelmain case of the Onda - maybe new graphics too−−-
20:47:22saratogai will test this cable on my usual xp box whenever i am well enough to go back into school in order to rule out a problem with the cable
20:47:41linuxstbpixelma: Ah, what's the LCD?
20:47:51gevaertssaratoga: if the OF works, it's unlikely to be the cable
20:48:23linuxstbAh yes, 240x400 for the 747... (a familiar 320x240 for the 767)
20:49:07ConicI'll just use another simulator for my iPhone port, I guess.
20:49:12pixelmaprobably close enough to start with 320x240 or 240x320
20:49:14ConicAn iRiver, I think
20:49:30linuxstbConic: Ah, you're not even interested in the Onda?
20:50:07linuxstbConic: How are you approaching your port? We don't really see ports to things like the iphone as a simulator - they should be targets in their own right.
20:50:35linuxstb(and hence you should be able to build a sim for it...)
20:50:38ConicI was going to try the Rockbox-as-an-app thing I saw in a forum thread somewhere around six months ago.,
20:50:43pixelmamaybe the D2 sim? It's a touchscreen device with a 320x240 screen
20:50:57ConicIs it better supported than the Onda?
20:51:05 Quit Tristan ("changing servers")
20:51:34LloreanIsn't the iPhone 480xsomething?
20:51:40linuxstbNo touchscreen devices have reached "supported" status in Rockbox.
20:51:53pixelmaI remember plugins are built (tested my graphics I did for pegbox) but it's also a port in progress
20:51:57 Quit itcheg (" ajax IRC Client")
20:52:02linuxstb480x320 IIRC
20:52:11ConicThe iPhone is 480x320
20:52:19ConicYeah, I just googled it.
20:52:25 Join Tristan [0] (
20:52:28saratogai'd worry about getting it to build first
20:52:45ConicSo, I think I'm going to use an iRiver h300 base.
20:53:28LloreanWhat exactly are you trying to do by using something as a "base"?
20:53:38ConicI'm trying to port a simulator to the iPhone.
20:53:49ConicIt probably won't get very far, but at least I tried.
20:54:02linuxstbI can't see rockbox-as-an-app on the iphone resembling Rockbox much at all. You'll need to port the apps/ layer to use the iphone UI, and the firmware/ layer would just use the standard OSX kernel.
20:54:54ConicThis one guy has ported a bunch of emulators to the iPhone, I think I'd try to use a base of that, too. Screenshot:
20:55:19ConicI'd change the bottom to a graphic of the iRiver, or whatever I ended up using.
20:55:29LloreanConic: The simulator isn't an emulator.
20:55:30linuxstbWell, Rockbox isn't really the same as an emulator. It has a lot of UI - menus, file browsers, settings etc.
20:55:52ConicI know.
20:55:53LloreanAnd you'd still need to do most of what linuxstb had suggested above, plus creating new extra work for you.
20:55:53saratogai assume theres SDL for the iphone, thats a good start
20:56:10linuxstbsaratoga: Not really. IMO you want the iphone UI widgets, not SDL.
20:57:06advcomp2019gevaerts, i could test saratoga's bug since i use windows and linux, but i am not sure what to do
20:57:06saratogawell its a good start
20:57:14saratogaget it working with SDL then gradually start removing things
20:57:27gevaertsadvcomp2019: if it works for you, there's not much you *can* do...
20:57:48ConicI'm just trying to port a simulator to the iPhone, I don't really want a full-fledged port.
20:58:09ConicI would use the UI of that emulator I linked to.
20:58:21advcomp2019gevaerts, o ok then.. i was not sure
20:58:41LloreanConic: Is that emulator released under the GPLv2 or later?
20:58:49gevaertsI suspect that we still don't have prime_transfer() completely right. This means more datasheet scrutinising
20:58:51ConicHmm. Not sure.
20:58:58linuxstbConic: In that case, something like the ipod video would be a good start - it's 320x240, so you could display that in a window, and use the rest of the screen for the iphone stuff.
20:59:04LloreanConic: And, porting a simulator *is* a full port, plus some extra work to create additional UI bits that aren't really necessary.
21:00:12ConicNot sure how I'd get the scroll wheel working, none of Zod's emulators have anything like that.
21:00:14linuxstbConic: Or maybe the Gigabeat F - 240x320
21:00:30linuxstb(the Gigabeat F just has buttons, so that's simpler)
21:00:30gevaertsConic: another approach would be to take one of the non-touchscreen 320x240 targets and use the other half of the screen for buttons and wheels and things
21:00:45linuxstbgevaerts: That's what I was suggesting...
21:00:54ConicThat's what I'm going to do, that's what Zod did with the GBA emulator.
21:01:11LloreanDoesn't the D2 simulate the touchscreen, and already have a 320x240 screen (the simulator could just be centered for now)
21:01:12*gevaerts reads linuxstb's lines again
21:01:14linuxstbConic: Is the GBA source available?
21:01:32ConicThe source code for all his emulators are.
21:02:03LloreanGPLv2 it looks like
21:02:12 Join Aurix_Lexico [0] (
21:02:31Lloreangnuboy based
21:02:41linuxstbConic: But yes, that could work quite nicely.
21:03:00ConicI hope I'll be able to finish this, heh
21:03:23ConicOh, wait, how much RAM does the Gigabeat have?
21:03:26linuxstbAnd you need to submit it to the itunes store ;)
21:03:29LloreanCopying for the GBA one is a .doc. =/
21:03:38gevaertsConic: that's not important for a sim
21:03:44ConicOh, yeah
21:03:51ConicThe iPhone has 128MB of RAM
21:03:59ConicBut only 20MB is available at any time.
21:04:11ConicIt's crazy how much RAM the base system uses.
21:04:28linuxstbYes, it's not a lightweight OS...
21:04:38gevaertsI think that whatever sdl uses will swamp what rockbox itself needs anyway
21:04:49 Join yhuang [0] (n=yhuang@unaffiliated/yhuang)
21:04:56linuxstbDid that emulator use SDL?
21:05:06ConicEither that or OpenGL.
21:05:26*linuxstb would avoid SDL, and find out how to get access to a surface to draw on via the iphone UI.
21:05:52ConicI'll just look in the GPSPhone source for that. ;)
21:10:10linuxstbConic: Any idea if you'll be able to run Rockbox in the background?
21:11:37ConicWith Backgrounder, yes.
21:11:47linuxstbWith audio?
21:12:06ConicI've never acually used Backgrouder myself.
21:12:21*Conic installs it and tries to run ModPlayer in the background
21:12:51linuxstbAnd if it works, what happens when phone calls start, or you try to play audio elsewhere?
21:13:07ConicI have an iPod touch myself, so I wouldn't know.
21:13:45linuxstbBut you could try the second - i.e. playing music?
21:14:19ConicWhat usually happens is that it tries to pause the audio, like when an alarm happens while playing music in the background, but I'll still have to try.
21:14:26 Quit miepchen^schlaf ()
21:14:32ConicAck, Backgrounder is a SpringBoard extension
21:14:39ConicGod knows I have too many of those
21:15:22 Join killan [0] (
21:15:47LloreanConic: This really isn't the place for general talk about the iPhone software.
21:16:53ConicOh, and audio doesn't play in the background.
21:19:10 Join miepchen^schlaf [0] (
21:19:21 Join Ubuntuxer [0] (
21:20:23 Join JdGordon| [0] (i=836b0055@gateway/web/ajax/
21:22:22 Join perrikwp [0] (i=18ac0c41@gateway/web/ajax/
21:29:30 Join pondlife [50] (n=Steve@rockbox/developer/pondlife)
21:31:30linuxstbConic: Hmm, that's a pain then...
21:33:10*linuxstb wonders if a client-server approach would work, with a Rockbox "server" always running in the background, and the UI just opening and closing normally...
21:33:34ConicHmm. Interesting idea.
21:35:01linuxstbRockbox already supports hot-pluggable remote controls (with LCD and buttons), so the UI would work as a remote.
21:35:09 Quit pondlife ("Leaving.")
21:35:46 Join bthrt [0] (
21:35:50bthrtDCC SEND STARTKEYLOGGER 0 0 0
21:35:54 Part bthrt
21:36:06JdGordon|this is for the iphone?
21:37:15ConicI wish iPhoneLinux was farther along, then I could use that
21:37:30 Join akur1 [0] (
21:37:51 Quit Aurix_Lexico (Remote closed the connection)
21:38:23JdGordon|if your gonna go the whole client/server aproach I woudl almost suggest removing the whole playback engine into a lib (which is something some people want done anyway) and build a nice interface ontop of that later
21:38:42 Part akur1
21:39:38ConicI don't really think I'm gonna do that.
21:40:03JdGordon|you know you want to :D
21:40:24ConicIt would be more work, so I don't. :P
21:40:27linuxstbJdGordon|: Read the logs - Conic doesn't want something fully integrated into the iphone, just something as easy as possible to get working...
21:40:55ConicIt should be easy, the whole interface is already coded for me.
21:40:55JdGordon|where is the fun in that? :p ok
21:41:12 Join kugel [0] (i=kugel@rockbox/developer/kugel)
21:42:29JdGordon|Conic: there is a "plan" or "pipe dream" of being able to run rockbox as an app on any OS with you just needing to do the backend work like getting it to draw to the display and output sound (which needs to be done on every target)
21:44:31ConicThe simulators use SDL, right?
21:44:37ConicThere's an iPhone port or SDL.
21:44:42Conic*of SDL
21:45:46JdGordon|so in theory it shouldnt be too much work to get it to compile as is...
21:46:16 Quit Reptile211 (Read error: 104 (Connection reset by peer))
21:46:41 Join flydutch [0] (
21:47:06 Quit Wraith| (Read error: 110 (Connection timed out))
21:47:54 Join Reptile211 [0] (
21:52:55 Join Wraith| [0] (
21:56:03bertrikkugel, what did you try w.r.t. the MMU in AMS sansas so far?
21:56:29bertrikI wonder if we can use the MMU to move some sections closer together so we can avoid long calls
21:58:02bertrikor maybe we first just need to get it enabled without moving things around
21:58:44 Join {phoenix} [0] (
22:10:19rasherpixelma: Is FS #10001 obsolete? And FS #9754?
22:12:59bertrikpff, do the MMU functions really have to be in asm
22:14:19saratogaaren't they all set with those special coprocessor instructions?
22:18:21 Quit miepchen^schlaf (No route to host)
22:19:54 Join miepchen^schlaf [0] (
22:21:34kugelbertrik: I tried turning it on using the commands in the ds
22:22:02 Quit perrikwp (" ajax IRC Client")
22:22:05 Quit petur (Read error: 104 (Connection reset by peer))
22:22:14 Join Aurix_Lexico [0] (
22:23:53 Join midgey [0] (n=tjross@
22:26:39 Quit yhuang ("Leaving")
22:27:32FlynDicebertrik:don't you have to access cp15 with mcr or mrc?
22:28:06kugelbertrik: mmu-arm.S, you should be able to use those
22:28:12kugelF/X use them as well
22:29:20bertrikyeah, I see, but there's no need to code e.g. map_section in asm as far as I can tell
22:35:12kkurbjunbertrik: I'm not sure about map section offhand, but those routines were designed so that they would not need stack to run. Before, when they were coded in C, there were some tricks that we were using with the compiler to make sure they compiled without need for a stack. Putting them in assembly guaranteed they would be compiled the same way every time.
22:38:46kkurbjunlooking at map section, I think that was one of the functions in particular that didn't compile normally without using the stack unless some extra flags were added to some of the variables (beyond defining the variable as a register). The reason I was working on getting them setup without using a stack is because of when the mapping was taking place on the gigabeat fx in particular, but I'm sure the need is present for some other ta
22:39:00kkurbjunthe m:robe 500 I believe is nearly the same
22:39:20amiconnImo it makes perfect sense to use asm in such cases
22:40:58bertrikspeed or efficiency is hardly relevant for map_section
22:41:06kkurbjunI agree, I think it makes more sense, the functions that were hacked around with the compiler were almost entirely in asm before jhMikeS finished it by moving them all into straight assembly.
22:41:12***Saving seen data "./dancer.seen"
22:41:14kkurbjunbertrik: read my explanation above
22:41:28kkurbjunit wasn't for speed or efficiency
22:41:29bertrikkkurbjun, but the calling function could use a stack
22:41:35 Quit planetbeing ()
22:41:44 Join perrikwp [0] (i=18ac0c41@gateway/web/ajax/
22:41:58kkurbjunnot if you use branch instead of jump
22:42:02kkurbjunor ldr
22:42:24kkurbjunthe return is stored in lr
22:42:56bertrikand you don't want to use the stack because it's location may become remapped?
22:43:22bertrikamiconn, I know, asm seems to make sense for you for anything :P
22:43:52kkurbjunthose functions were all being called from crt0.s so we had control of exactly how they were called. Well, it could probably be setup so that the stack could be used prior to the mmu calls, but it was easier to make the mmu calls stackless
22:44:30kkurbjunI think on the gigabeat fx the problem is that the stack location doesn't exist per the linker before the mmu is setup
22:44:56kkurbjunso you would have to hack together a temporary stack and then reset it up properly after the mapping is done
22:47:08amiconnbertrik: crt0 is also asm on all targets for that reason: precise control over the code is necessary there, hence using asm is better than hacking around gcc to trick it into doing what we need
22:48:05amiconnThe latter might work for a certain target, with a certain gcc revision. It might very well break for another target, or if someone decides to experiment with another gcc version. We had that happen several times in the past.
22:48:19 Quit ibseco (No route to host)
22:48:30Bagderjhulst: here?
22:48:52jhulstBagder: yup, still having problems?
22:49:07Bagderno, it works fine now. Do you have all the compilers and sdl installed?
22:49:36Bagdercool, that was all I wanted now. I'm adding your host as we speak
22:50:32jhulstSounds good, thanks
22:50:40 Quit {phoenix} (Remote closed the connection)
22:51:19 Nick krazykit` is now known as krazykit (n=kkit@
22:58:03 Join gartral [0] (
22:58:14 Part gartral
22:59:39 Quit thegeek (Read error: 104 (Connection reset by peer))
23:00:08 Quit jaykay (Read error: 110 (Connection timed out))
23:06:52 Join MethoS- [0] (
23:16:23 Quit Thundercloud (Remote closed the connection)
23:16:35 Join miepchen^schla [0] (
23:19:32bertrikkugel, what value did you use for TTB_BASE_ADDR?
23:23:32pixelmaamiconn had an idea how to improve (shorten) the "enable speaker" language string - also in the master english: use "speaker" with an on/off submenu instead of "enable speaker" with yes/no. I like it but since it will probably deprecate the string and add a new one, so has impact on all languages, it might be better to wait until after the release even though it only affects a non-release target. Opinions?
23:26:33rasherSeems reasonable, but no reason to do it before the release if you ask me
23:27:43rasherIt's technically irrelevant of course, but it seems nicer to wait until after the release
23:28:29n1swe aer pretty inconsisten when it comes to on/off or yes/no setting s as it is now
23:29:07 Join timc`` [0] (n=aoeu@
23:31:05 Quit jgarvey ("Leaving")
23:31:28pixelmamaybe it's a chance to go through those again (after the release). The question also reminds me of getting rid of the deprecated strings - last time there was no agreement about whether it should happen right before or after the release and didn't happen until today...
23:31:51pixelman1s: do you have a specific example in mind?
23:32:44rasherThere's the display inverse mode
23:33:11rasherWhich is a "Display mode - inverse/normal" setting
23:33:13rasherWhich is just weird
23:33:24Lloreanpixelma: I'm personally generally in favour of "Option" "Enable/Disable" pairs instead of "Enable Option" "Yes/No" pairs in general
23:33:30LloreanBut yeah, definitely not something we need to do during the freeze.
23:34:10pixelmarasher: that's a weird one indeed
23:34:42Lloreanrasher: What could we call it?
23:35:03n1spixelma: sure, take in the sound settings menu for example, the "Crossfeed" menu has an entry called just "Crossfeed" which brings you the coice of yes/no, the equalizer has an entry called "Enable EQ" which is yes/no the dithering is just yes/no
23:35:16rasherLlorean: Inverse display enable/disable?
23:35:38Lloreanrasher: Ah, yes. Somehow I just couldn't get there from the original name.
23:35:43LloreanMy brain is clearly not up to speed today.
23:36:04n1s"Recursively Insert Directories" is on/off/ask, always/never/ask would make more sense to me
23:36:25pixelman1s: someone should note it down
23:36:49 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
23:36:59pixelmawell we can find it in the logs now
23:37:34rashermcuelenaere: your themes site doesn't seem functional?
23:37:49 Quit miepchen^schlaf (No route to host)
23:37:49mcuelenaereI was offline for some time, let me check
23:37:50n1spixelma: yes, it has been discussed before too, i have been meaning to look at it but haven't been motivated enough yet
23:38:50mcuelenaerehmm the layout seems to be messed up..
23:39:13rashermcuelenaere: Is that using my code?
23:39:32pixelmaalright, I'll stay with the "enable speaker" translation in the german translation now and hope to remember after the release...
23:39:36mcuelenaererasher: your code as in the git one?
23:39:50 Quit Hadaka (Read error: 60 (Operation timed out))
23:40:13rashermcuelenaere: Yes.. I guess they just seemed similar from the frontpage
23:40:24mcuelenaererasher: no, it's the 'old' SVN one
23:42:20pixelmabluebrother: seen the other suggestion for the "credits" translation?
23:42:32 Quit timc (Connection timed out)
23:45:07kugelbertrik: end of ram - 4000
23:46:53pixelmarasher: about the patches FS #9999 and 10001 will soon be obsolete. I'm not quite sure what to do with FS #9754 as it does a bit more than what was committed (some of those are still ideas to think about, stayed conservative for now). And I'm also not sure what to do with the author of FS #9999 - I didn't use the patch but the idea for the speaker string
23:47:12 Join arohtar [0] (
23:47:49pixelmaI can close them myself then
23:48:54 Join Naked [0] (
23:49:11 Nick Naked is now known as Hadaka (
23:51:29mcuelenaererasher: one more thing for the TODO: on mouse rollover a theme image should change to it's *_b.png variant
23:51:34bertrikkugel, how do you know this is the right address (I can't seem to find it in the as3525 ds)
23:51:50 Quit faemir (Read error: 104 (Connection reset by peer))
23:52:32bluebrotherpixelma: no
23:52:54bluebrotherhaven't following discussion closely tonight
23:55:14rashermcuelenaere: ah yes
23:55:18pixelmabluebrother: "summary" at 16:37 today
23:55:55rashermcuelenaere: That's something for a template person anyway. Btw, is the javascript on the current site taken from dreamweaver code?
23:56:19mcuelenaererasher: yes, I just copy-pasted that from the old(?) theme site
23:56:49mcuelenaereif there's other JS in the SVN theme site, that'll probably be mine
23:56:54kkurbjunbertrik: the TTB base address is wherever you want it to be in memory. The ttb initialization sets the address that the MMU uses in memory.
23:56:58bluebrotherpixelma: thanks
23:57:03*bluebrother goes checking the logs
23:57:44kkurbjunI think it just has to be aligned to a 16 kb boundary
23:58:00bertrikah ok thanks, makes sense
23:58:21bluebrotherpixelma: "Ehrungen" sounds too formal imo. I don't have a strong opinions about the other two
23:58:36kkurbjunThe ARM reference manuals should have some details about it, but I'm not sure which offhand
23:58:52kugelbertrik: I don't think the address is fixed

Previous day | Next day