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 2009-10-17

00:01:02 Join KBH [0] (
00:01:14 Join spacecadet [0] (
00:02:49spacecadetIs there an official "relase" zip of ipodpatcher source, instead of needing svn?
00:03:52TheSevenwhat's the matter with svn?
00:04:08TheSevenhuh? it doesn't allow HTTP?
00:04:32spacecadetnormal http gets through, svn gets blocked
00:04:46linuxstbspacecadet: I'll make one for you...
00:04:49spacecadetor it doesn't work, anyways
00:05:16TorneTheSeven: our svn repo isn't http...
00:05:40Torne(svn:// is svnserve protocol)
00:05:58 Quit HBK (Read error: 110 (Connection timed out))
00:06:19TheSevenah? someone's still using svnserve?
00:06:47 Join mc2739 [0] (n=mc2739@rockbox/developer/mc2739)
00:08:23linuxstbspacecadet:"> - compile with "make RELEASE=1"
00:09:01 Quit KBH (Read error: 60 (Operation timed out))
00:09:20*TheSeven is getting annoyed by that freeze
00:09:42linuxstbspacecadet: Earlier versions of ipodpatcher had BSD binaries, but that person has disappeared and I don't now if it still compiles cleanly. Please let me know if you have any problems...
00:10:35spacecadetI'll let you know if it compiles. Do you want a binary if it does?
00:11:34spacecadetok, I'll go be aggravated by gcc now
00:12:15 Part spacecadet ("Leaving")
00:12:26 Join GeekShadow [0] (n=Antoine@reactos/tester/GeekShadow)
00:14:11 Quit jgarvey ("Leaving")
00:14:29CIA-85New commit by rob (r23217): Fix logic error in D2+ touchscreen driver.
00:15:00rhcp1253srry had to eat dinner 4gb
00:18:01*TheSeven will fall back to tweeter-based debugging now :-/
00:20:00 Quit rhcp1253 ("CGI:IRC (EOF)")
00:20:29 Join rhcp1253 [0] (
00:21:30 Quit robin0800 (Remote closed the connection)
00:22:17rhcp1253k i just had this problem and i remember it from before, my ipod video 5g keeps randomly restarting and repetedly. I've fixed it for now but what can i do if it keep happening
00:22:42 Join Hillshum [0] (
00:23:07 Quit Zagor ("Clint excited")
00:24:37 Join WilliamC [0] (
00:26:56TheSevenok, now I know at leas that it's not locking up inside the USB handler
00:27:03TheSeventhe tweeter is a really nice means of output :-)
00:27:34WilliamCI have replaygain applied to all of my music but some of them sound noticeably louder than others.
00:28:01 Join robin0800 [0] (
00:28:21linuxstbTheSeven: You couldn't use the UART to debug?
00:28:34TheSeveni don't have a cable, and it will be too slow
00:29:11TheSeveni'm having serious trouble with delays caused even by a i2c write to the PMU to flash the backlight in that int handler
00:29:25TheSeventhe tweeter has the advantage of just being a single reg write
00:29:42linuxstbWilliamC: Do those tracks have anything in common?
00:30:02 Join fdinel [0] (
00:30:32linuxstbWilliamC: Are you _sure_ ? You're not giving us much to go on...
00:30:49WilliamCThey're all MP3 files
00:30:56WilliamCThe majority of my files are OGG though
00:31:03TheSevencalling wakeup_wait with ints disabled will re-enable them?
00:31:27linuxstbWilliamC: That's more helpeful ;) So the MP3 files are sounding louder than the vorbis files?
00:31:49 Quit daggett ("Ex-Chat")
00:32:02linuxstbHow did you add replaygain to the MP3 files?
00:32:15WilliamCFoobar 2000
00:32:25WilliamCper track replay gain
00:32:54linuxstbDo you know if Foobar added the replaygain info in APE tags?
00:32:58rhcp1253just a sugestio from a noob try editing the codecs
00:33:13WilliamCNo, it's not.
00:33:43TheSevenbah. shooting at a moving target with only a single bit of feedback is really hard.
00:34:02TheSeventhe place it's locking up is moving :-/
00:34:22linuxstbWilliamC: When one of those tracks is playing in Rockbox, press and hold "select" to bring up the WPS context menu, then look in "Show track info" to see if the replaygain values are shown.
00:34:46 Join spacecadet [0] (
00:35:19gevaertsTheSeven: chase it until it leaves the code
00:35:39TheSeveni fear it won't do that
00:36:50spacecadetlinuxstb, gcc didn't like the makefile syntax, it wanted "."'s infront of the preprocessor statements, and it thought some of the parenthesis were mismatched. I guess I'll have to work on it.
00:37:18gevaertsspacecadet: have you tried gmake instead of make?
00:37:22linuxstbspacecadet: Are you sure you have GNU make? Try "gmake" instead of make.
00:37:34spacecadetthat'll be next :)
00:37:57spacecadetI do have gmake
00:38:25spacecadetget back to you later
00:38:35 Part spacecadet ("Leaving")
00:39:25WilliamCIt's not showing
00:39:42linuxstbWilliamC: Did you also check a vorbis file?
00:39:49WilliamCThat showed
00:40:19linuxstbOK, then that's the problem...
00:40:40WilliamCI wonder if it ever had replaygain
00:41:29linuxstbRockbox only supports id3/id3v2 tags on mp3 files, so if they were added as APE tags (which is quite common), Rockbox won't see them.
00:41:34WilliamCIt never had replaygain
00:41:48WilliamCI have APE disable in Foobar 2000
00:42:25 Quit roolku ()
00:43:23WilliamCThere, fixed
00:44:46 Join BBBradley [0] (
00:45:06 Quit LambdaCalculus37 ("Fwump")
00:47:12WilliamCHow come 2 drives show up when I plug in my MP3 player to my computer?
00:47:12WilliamCIs that something to do with the bootloader?
00:47:26 Quit domonoky (Read error: 54 (Connection reset by peer))
00:47:34TheSevenwhich kind of player?
00:47:43***Saving seen data "./dancer.seen"
00:47:58linuxstbDoes it have an SD card slot?
00:48:46WilliamCSansa e200
00:48:58linuxstbThat's why there are two drives.
00:49:16WilliamCI never even really thought about the micro SD slot
00:49:34kugelhaving a 2nd albumart slows down track skipping quite a bit on the fuze
00:50:26 Quit Thundercloud (Remote closed the connection)
00:50:27kugelin fact, having 1 albumart is already slower vs no aa. buffering albumart seems to take 1-2 seconds on the fuze
00:50:42gevaertswhat's the original size?
00:52:23kugeljpg, so halving it it almost free
00:53:21gevaertsthat's not unreasonably large...
00:53:49gevaertswell, people aren't forced to use AA status bars
00:54:06 Join Strife89 [0] (
00:54:23kugelthe issue is also that the fuze has so little memory that track skipping very often means buffering
00:54:41 Join Thundercloud [0] (
00:54:59 Nick liar2 is now known as liar (n=liar@
00:55:12 Quit bertrik ("Leaving")
00:57:12WilliamCKugel, that sounds very pathetic on whoever makes the Fuze.
00:58:03*gevaerts gives WilliamC a DAX
00:58:30*linuxstb gevaerts a DAX - you can't get rid of it that easily...
00:58:30WilliamCWhat's a DAX?
00:58:31TheSevengevaerts: it's leaving the interrupt handler and then never calling any usb_drv function again, disabling interrupts, and locking up somewhere :-/
00:58:41 Join stripwax [0] (
00:59:01 Quit stripwax (Client Quit)
00:59:37 Quit BBBradley ("CGI:IRC (EOF)")
01:00:43gevaertsWilliamC: Logik DAX
01:01:10 Quit ender` (" "Why are we hiding from the police dad?" "They use vi, my son. We use EMACS".")
01:01:22WilliamCI never heard, or saw them
01:01:41gevaertsthey have 1MB of RAM
01:01:53TheSevengevaerts: and sometimes (~20% of the tries) it passes to a stage where windows detects the vid/pid, the ipod tries to transmit or receive some data, occasionally there's an int, but it doesn't work out
01:01:58 Join HBK [0] (
01:01:59gevaertsseriously though, what's pathetic about not using more hardware than you need?
01:02:52WilliamCgevaerts, I bet they could make an MP3 player that is pure ram, though you'd be screwed if the battery died.
01:03:08 Quit gitster ("ERC Version 5.3 (IRC client for Emacs)")
01:03:37 Join spacecadet [0] (
01:04:06spacecadetlinuxstb, gmake worked but the code still needed some changes
01:04:20 Quit robin0800 (Remote closed the connection)
01:05:03kugelgevaerts: hrm, using a 300x300 AA with a wps that wants 75x75 isn't any faster, even though scaling is supposedly almost free
01:05:38linuxstbspacecadet: What was the problem?
01:05:43gevaertskugel: maybe try bmps of exactly the right size to see what the baseline is?
01:06:17spacecadetI had to replace refs to stdint.h, and then I've just been left with two sets of errors
01:06:23gevaertsTheSeven: I'd say there's something wrong with the hardware initialisation, but I really have no idea
01:06:46linuxstbspacecadet: What did you replace stdint.h with? inttypes.h?
01:06:50*kugel wonders if it can be sped up with some threading intelligence
01:07:14kugellike, yielding less often during scaling so that finishes faster. audio doesn't play before it finished anyway
01:07:37spacecadetipodpatcher.c: In function `GetSecurityBlockKey':
01:07:37gevaertskugel: that's a special case. Normally audio *does* play while scaling
01:07:37spacecadetipodpatcher.c:2036: syntax error before `int'
01:07:37spacecadetipodpatcher.c:2043: `r2' undeclared (first use in this function)
01:07:37DBUGEnqueued KICK spacecadet
01:07:37spacecadetipodpatcher.c:2043: (Each undeclared identifier is reported only once
01:07:37spacecadetipodpatcher.c:2043: for each function it appears in.)
01:07:38***Alert Mode level 1
01:07:38spacecadetipodpatcher.c:2044: `r12' undeclared (first use in this function)
01:07:40spacecadetipodpatcher.c:2045: `r_tmp' undeclared (first use in this function)
01:07:42***Alert Mode level 2
01:07:42spacecadetipodpatcher.c:2046: `r14' undeclared (first use in this function)
01:07:44***Alert Mode level 3
01:07:44spacecadetipodpatcher.c:2049: `r1' undeclared (first use in this function)
01:07:50gevaertsspacecadet: please use a pastebin for that
01:08:08kugelgevaerts: that may be because the pcmbuf holds enough
01:08:30kugelor doing the scaling in its own thread, allocating the max buffer the AA is going to need, so that audio buffering can start before scaling is finished
01:09:20gevaertsthat could work, but I'm not convinced that it will be worth the complexity
01:09:48linuxstbspacecadet: Did you find a fix for that?
01:10:03kugelmaybe Unhelpful has some ideas
01:10:31spacecadetI'm still going over it. I'm tired, for now, and there's sg-u on in a while :)
01:10:59gevaertslinuxstb, spacecadet: it looks like the declarations could just be moved to the start of the block
01:11:24gevaertsthis is just a compiler that's more than ten years out of date :)
01:11:35spacecadetthere was one other error, but since I can't find the pastebin, I won't copy it.
01:11:52spacecadetno the error is only 5 years out of date :)
01:12:04spacecadetno the compiler is only 5 years out of date :)
01:12:14gevaertsten. This is C99 code :)
01:14:02CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
01:14:02*TheSeven is annoyed by this weird bug and too tired to continue working on it, so he'll go to bed now
01:14:32*linuxstb wonders if he should tell TheSeven how long it took to find all the PP USB bugs...
01:14:55spacecadetI'll figure it out and come back another day
01:15:01gevaertslinuxstb: you mean we have them all?
01:15:03TheSevenlinuxstb: but i have a working driver in ibugger! i just can't find the difference...
01:15:41 Part spacecadet ("Leaving")
01:15:42TheSevenand i guess the pp one isn't documented at all
01:16:27gevaertsit's documented in a freescale datasheet, but most problems are with magic ram and cache setup things
01:17:06TheSeveni think i've caught all of these on the s5l8700
01:17:08 Join JdGordon [0] (i=836b0068@gateway/web/freenode/x-folazdutxhuqlyxb)
01:17:20TheSevenstill need to work out a solution for the nand driver alignment issues though
01:17:28TheSevendon't want to copy every buffer around all the time
01:17:30*gevaerts admires TheSeven's boundless optimism
01:17:45***Alert Mode OFF
01:18:31TheSevenwell, we have a detailed datasheet of that processor, and after i rewrote some things, the cache stuff is definitely working now, and we don't have any other memory management stuff to deal with
01:19:49TheSevenif you align everything properly on 16-byte boundaries and do a flush_dcache before every dma transaction and invalidate_dcache afterwards, nothing can really go wrong
01:20:05TheSevenand there are only 3 places where dma is used at all
01:20:20TheSeventhe NAND/ECC, the PCM driver, and of course USB
01:20:41TheSeven(and in that cryptfirmware plugin and the bootloader the AES/HASH stuff)
01:22:27*TheSeven is gone now
01:22:27 Quit rhcp1253 ("CGI:IRC (EOF)")
01:22:34gevaertsI wouldn't be so sure :)
01:23:06gevaertsAre you sure you won't run into timing issues if you do heavy things like cache flushes while DMA is in progress?
01:30:08 Join advcomp2019_ [0] (n=advcomp2@unaffiliated/advcomp2019)
01:32:17 Quit Hillshum (Read error: 110 (Connection timed out))
01:36:41 Quit fyrestorm (Read error: 54 (Connection reset by peer))
01:40:55 Join gitster [0] (
01:46:18 Quit TheSeven (Read error: 110 (Connection timed out))
01:46:54 Quit advcomp2019 (Read error: 110 (Connection timed out))
01:57:20 Quit bertrik_ (Read error: 113 (No route to host))
02:09:12Unhelpfulkugel: i said scaling was almost free when?
02:10:10 Join webguest71 [0] (n=60f125fc@
02:10:14kugelUnhelpful: scaling jpg /2, /4 and /8, wasn't it?
02:11:32webguest71Hey guys, I'm on a Sansa Fuze v1 and it's been working fine until I put in a 4gb SDHC uSD card. Now it's saying:
02:11:42webguest71"Loading Firmware"
02:11:55webguest71"Length: FFFFFFF8"
02:11:59Unhelpfulkugel: oh... yes, i hadn't compared the dimensions. 300x300->75x75 will be substantially faster than loading at native res. :)
02:12:08webguest71Read failed (chksum)
02:12:15 Join peter__b [0] (n=peter_b@
02:12:24 Join BHSPitMonkey [0] (n=stephen@unaffiliated/bhspitmonkey)
02:12:31Unhelpful*only* for jpeg, i should point out.
02:12:32webguest71I've formatted it, ran chkdsk, reinstalled the orig. bootloader
02:12:44kugelUnhelpful: yea, but it wasn't really faster
02:12:47webguest71which promptly hung after "Refreshing your Database"
02:12:55webguest71deleted the mtable.sys
02:13:02webguest71Reinstalled the rockbox bootloader
02:13:06webguest71Any other ideas?
02:14:08kugelUnhelpful: with my custom sb patch, track skipping on the fuze needs ~0.2s without AA, ~1.5s with 1 AA, and ~3s with 2 AA
02:14:52kugelthe 1.5s (with cabbiy) was the same for the 263x230 and 300x300 albumart (scaled to 75x75)
02:14:56Unhelpfulkugel: i'm assuming this is when the next track is not on buffer?
02:15:26kugelyes, that's almost always true on the fuze (only 4MB audio buffer here)
02:16:05 Join AndyI [0] (n=pasha_in@
02:16:29Unhelpful263x230 would load with 1:2 reduction, then the scaler would do the rest.
02:19:33 Quit peter-b (Read error: 145 (Connection timed out))
02:22:35 Join evilnick_ [0] (
02:25:58webguest71Any ideas guys?
02:29:16 Nick webguest71 is now known as Alex_Buie (n=60f125fc@
02:29:19 Quit AndyIL (Read error: 110 (Connection timed out))
02:34:57 Quit Alex_Buie ("CGI:IRC")
02:41:03 Quit evilnick (Read error: 110 (Connection timed out))
02:43:19 Quit n1s ("Lämnar")
02:44:55kugelUnhelpful: strange, now it's instant
02:45:39 Quit JdGordon ("Page closed")
02:46:57 Quit matsl (Read error: 110 (Connection timed out))
02:47:47***Saving seen data "./dancer.seen"
02:48:11Unhelpfulhm? rather
02:48:21kugeltrack skipping
02:48:26kugelafter a reboot
02:49:24*kugel needs to get arond unifying button handling for radio and wps
02:51:15 Quit StealthyXIIGer (Read error: 54 (Connection reset by peer))
02:51:20 Join StealthyXIIGer [0] (
02:59:14 Quit kugel (Remote closed the connection)
03:00:29 Quit amiconn (Nick collision from services.)
03:00:32 Join amiconn_ [0] (i=quassel@rockbox/developer/amiconn)
03:00:36 Quit pixelma (Nick collision from services.)
03:00:37 Join pixelma_ [0] (i=quassel@rockbox/staff/pixelma)
03:00:54 Nick amiconn_ is now known as amiconn (i=quassel@rockbox/developer/amiconn)
03:00:59 Nick pixelma_ is now known as pixelma (i=quassel@rockbox/staff/pixelma)
03:02:29 Join KBH [0] (
03:05:45 Quit BHSPitMonkey (Read error: 113 (No route to host))
03:06:53 Quit Thundercloud (Remote closed the connection)
03:13:04 Join Hillshum [0] (
03:16:50 Quit GeekShadow (Read error: 104 (Connection reset by peer))
03:19:18 Quit HBK (Read error: 110 (Connection timed out))
03:24:09 Quit batman52 (Read error: 104 (Connection reset by peer))
03:35:53 Quit fdinel ("Miranda IM! Smaller, Faster, Easier.")
03:40:19 Quit Hillshum (Read error: 60 (Operation timed out))
03:41:01 Quit efyx_ (Remote closed the connection)
03:45:23 Join kkurbjun [0] (n=kkurbjun@rockbox/developer/kkurbjun)
03:47:40 Join barrywardell_ [0] (
03:50:23 Quit n17ikh ()
04:03:14 Join BHSPitLappy [0] (n=BHSPitLa@unaffiliated/bhspitmonkey)
04:04:02 Join joojii [0] (n=joojii@
04:04:25 Quit barrywardell (Read error: 110 (Connection timed out))
04:04:25 Nick barrywardell_ is now known as barrywardell (n=barrywar@rockbox/developer/barrywardell)
04:18:09 Quit joojii (Read error: 145 (Connection timed out))
04:41:39 Join Strife1989 [0] (
04:43:12 Quit Rondom (Nick collision from services.)
04:43:23 Join Rondom [0] (
04:47:51***Saving seen data "./dancer.seen"
04:48:39 Quit evilnick_ (Read error: 110 (Connection timed out))
04:58:13 Quit Strife89 (Read error: 110 (Connection timed out))
05:02:41 Join HBK- [0] (
05:12:24 Join Tristan [0] (i=tristan@
05:17:26 Quit Strife1989 ("Bed.")
05:20:26 Quit KBH (Read error: 110 (Connection timed out))
06:34:32 Join n17ikh [0] (
06:47:55***Saving seen data "./dancer.seen"
06:56:24 Join phanboy4 [0] (
07:21:19 Quit StealthyXIIGer (Read error: 110 (Connection timed out))
07:23:15 Join lifeless__ [0] (n=lifeless@
07:24:13CIA-85New commit by midgey34 (r23218): Add support to genlang to sort strings based on their user. This is useful for translatable plugins. Currently, the .lng that is generated is ...
07:29:32 Join |DaMaGeD| [0] (n=dam@
07:39:56CIA-85New commit by tomers (r23219): USB HID: Use different context for each mode; Use constant for each action; ...
07:40:07 Quit lifeless_ (Read error: 113 (No route to host))
07:41:20 Join KBH [0] (
07:42:20 Join Horschti [0] (n=Horscht2@xbmc/user/horscht)
07:45:26 Quit HBK- (Read error: 60 (Operation timed out))
07:49:07 Quit Horscht (Read error: 60 (Operation timed out))
07:49:26 Quit |DaMaGeD| ("used jmIrc")
07:49:41CIA-85New commit by tomers (r23220): Fix red: Type in constant
07:50:33 Join |DaMaGeD| [0] (n=dam@
07:50:47 Quit |DaMaGeD| (Client Quit)
08:07:51CIA-85New commit by midgey34 (r23221): Change lang_init() to take a source, destination, and count. Needed for translatable plugins. Minor stylist changes as well.
08:08:04 Part WilliamC ("Leaving")
08:08:11 Join midgey [0] (n=tjross@rockbox/developer/midgey)
08:18:46 Quit Zarggg ()
08:24:10 Join midgey_ [0] (
08:26:30 Join Rob2223 [0] (
08:30:03 Quit midgey (Read error: 110 (Connection timed out))
08:30:03 Nick midgey_ is now known as midgey (n=tjross@rockbox/developer/midgey)
08:30:55 Quit sbhsu (Read error: 148 (No route to host))
08:31:57 Join sbhsu [0] (n=a6530466@
08:32:13 Join esperegu [0] (n=quassel@
08:33:22midgeyerr, i'm going to assume that red for the iaudio x5 is not my fault...
08:44:21 Quit Rob2222 (Read error: 110 (Connection timed out))
08:48:00***Saving seen data "./dancer.seen"
08:56:50 Join Zarggg [0] (
09:05:12 Quit panni_ ("( :: NoNameScript 3.81 :: )")
09:06:25 Join |DaMaGeD| [0] (n=|DaMaGeD@
09:15:54 Join bertrik [0] (
09:16:02 Join ShapeShifter499 [0] (
09:20:51 Join kyle6513 [0] (n=kyle@
09:24:42 Quit phanboy4 (Read error: 113 (No route to host))
09:34:51 Join stoffel [0] (
09:36:48 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
09:43:56 Quit ShapeShifter499 (Remote closed the connection)
09:46:41 Join ShapeShifter499 [0] (n=chatzill@
09:47:23 Quit Rob2223 (Read error: 104 (Connection reset by peer))
09:48:03 Join Rob2222 [0] (
09:48:27|DaMaGeD|give somebody plz an example of menu made through menu_api)
09:54:14 Join kyle6513_ [0] (n=kyle@
09:58:37 Nick Horschti is now known as Horscht (n=Horscht2@xbmc/user/horscht)
10:01:32 Quit BHSPitLappy ("Ex-Chat")
10:09:02|DaMaGeD|maah,this menu won't compile in any way
10:09:22 Quit midgey ()
10:10:59amiconngevaerts: I just tried a current swusb build (i.e. with hid) on my Amiga (Poseidon USB stack). It's working fine :)
10:12:21 Nick |DaMaGeD| is now known as |DaMaGeD|[noob] (n=|DaMaGeD@
10:12:23 Join polobricolo_ [0] (
10:12:46 Quit kyle6513 (Read error: 110 (Connection timed out))
10:13:04 Quit polobricolo (Read error: 110 (Connection timed out))
10:13:16 Nick polobricolo_ is now known as polobricolo (
10:13:38amiconnHehe, I can even scroll using the ipod wheel in AWeb (one of the few Amiga apps supporting a scroll wheel)
10:15:07 Part polobricolo
10:16:34 Quit kadoban (Remote closed the connection)
10:16:44 Quit gevaerts (Nick collision from services.)
10:16:56 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
10:18:48pixelmatomers: I don't understand your {} changes in the manual - and that you use % to end the line, I guess it's related... % is a comment mark and everything behind it will be ignored even a linefeed - that's why I think you added the {} there because somewhere in the output was a space missing?
10:20:25 Quit ShapeShifter499 ("ChatZilla 0.9.85 [Firefox 3.0.14/2009090216]")
10:21:02pixelmausually you don't need % at the end, only when breaking a long opt list ( \opt{a,b,c,d} what's happening in the curly braces) or when you get additional whitespace. In usual texts you don't need it because TeX works like e.g. html there
10:22:20 Join ender` [0] (
10:24:56pixelmatomers: so to me it looks like the proper solution would be to get rid of the % but I'm not sure if this is really what you were trying to solve
10:33:02pixelmaI'd also prefer if the manual stuff was an extra commit
10:38:41 Join Casainho [0] (
10:39:16 Join n1s [0] (n=n1s@rockbox/developer/n1s)
10:41:49CasainhoBob_C: I am starting working now on the JTAG cable. Did you found any other way to load Rockbox bootloader using the factory bootloader on the board?
10:42:11CasainhoBob_C: and did you build already the expansion board for buttons?
10:42:57 Join Dgby714 [0] (
10:47:50bluebrothera comment followed by an empty statement in the following line? Urgh.
10:48:01***Saving seen data "./dancer.seen"
10:48:54bluebrotherpixelma: those changes are really broken :(
10:50:14 Nick |DaMaGeD|[noob] is now known as |DaMaGeD|[test] (n=|DaMaGeD@
10:55:48tomerspixelma: (sorry for the late response) the { } is to force a space. I guess there is some better directive, something like \space.
10:56:20pixelmajust don't end the line above with a comment...
10:56:25 Join matsl [0] (
10:57:45bluebrothertomers: forcing spaces is really bad. We had issues with those in the past and I'd really like to not introduce that again ...
10:58:14 Join flydutch [0] (
10:58:24bluebrotheryou'll easily get duplicated spaces this way, especially opt is known to insert additional spaces at times
10:58:30tomerspixelma: Removing the % (comment indeed) at the end of line causes it to regard line break as some kind of space in the output. Since I use multiple \opt{..}, each in its own line, it causes more and more spaces to appear, depending on the amount of lines. Having the comment directive at the end of line prevent this from happening.
10:58:47bluebrotherwith duplicated spaces = multiple spaces in the output, which looks really bad
10:59:05tomersbluebrother: Totaly agree, but I couldn't find a better solution (although I was sure there was such0
10:59:39tomerspixelma: I'll check that
10:59:47Bob_Ccasainho: I downloaded some code into RAM and ran it, but that's all so far
11:00:19Bob_Ccasainho: I have built a keypad, yes
11:01:40bluebrothertomers: I can't follow this argument. For example when looking at line 208 you introduced that. It did work before, did it?
11:02:02pixelmatomers: just leave out the last right above that then
11:02:25bluebrotherbesides, looking at line 200 makes it really clear: the space is added because you add a space after the first \opt, and the second would add another. You need to have no trailing spaces and comments for all but the last opt
11:04:20tomersIt's all trial & error. I sometimes got the words adjacent to each other...
11:04:23 Join bertrik_ [0] (
11:04:30bluebrotherbtw, I'd really suggest discussing such stuff first. We already had space issues, and there is at least some knowledge on how to avoid it ...
11:04:47pixelmatomers: I like that you take care of the documentation at all though :)
11:05:00*bluebrother agrees with pixelma
11:05:48tomersbluebrother: Thanks :-) It just that there was so much code changes going on with that last commit of mine, that I wanted to put it all on SVN first
11:06:33pixelmaand I also experienced unexpected behaviour of TeX sometimes, so yes even if you already did some changes it stays trial and error
11:07:13tomersWill we ever go to XMLDOC?
11:07:28bluebrotherunfortunately broken solutions are the ones that tend to be around longer than the correct ones. One reason I don't like such things.
11:07:37bluebrotherwhat's the gain of that?
11:08:03 Quit bertrik_ (Remote closed the connection)
11:08:44bluebrotherand what exactly is xmldoc? I can't find any reasonable result with that search string. You're referring to docbook?
11:08:47tomersbluebrother: I don't actually know, besides the fact that I know people who switched to it in their businesses, and they did some research and got to the conclusion it worth (don't remember what exactly are the gains, though)
11:09:04tomersbluebrother: True ;-) docbook
11:09:49bluebrotherwell, from my point of view the actual problem is opt, not LaTeX by itself. I've used LaTeX quite a bit and never had big issues. Only opt is giving me headaches every now and then.
11:10:21bluebrothermy goal is to replace opt with a proper c preprocessor. Unfortunately the gcc preprocessor leaves empty lines around, which are a problem in LaTeX.
11:10:24Tornewell LaTeX works fine, until it doesn't :)
11:11:03bluebrotherand gpp (the one I've been trying to integrate a while back) does the job rather good −− until it comes to variadic macros which it doesn't understand at all :(
11:11:22CasainhoBob_C: "openocd.exe -f interface/olimex-jtag-tiny2.cfg -f board/mini2440.cfg" −−> I can't find the "limex-jtag-tiny2.cfg" on OpenOCD sources!
11:11:43CasainhoBob_C: did you work that file for yourself? can you send it to me?
11:12:04bluebrotheras building the manual needs a compiler since the introduction of features.txt there is no point in not using the preprocessor. I also consider it an advantage to be able writing #ifdef ;-)
11:12:25tomersbluebrother, pixelma: I got a nice patch which proves my point of not to remove the % at the end of line. Should I commit so you can experience it, and we gather around a solution, or should I upload it to or something like that?
11:12:41CasainhoBob_C: I mean, that file is not inside of "tools/OpenOCD" folder...
11:12:59bluebrothertomers: committing something broken is broken :) Use a pastebin.
11:13:18Bob_Ccasainho: Try interface/olimex-jtag-tiny.cfg ?
11:14:11amiconnbluebrother: Do you know where those empty lines are coming from?
11:14:24amiconnAre they coming from #ifdef, #endif etc themselves?
11:14:51tomers$ export DAP=ipodvideo ; rm -rf build_manual_${DAP}/ && mkdir build_manual_${DAP} && cd build_manual_${DAP} && printf "${DAP}\nm\n" | ../tools/configure && make manual ; cd ..
11:14:59bluebrotheramiconn: from #ifdef and the like. It's basically blocks the preprocessor removes getreplaced by empty lines.
11:15:06tomers$ acroread build_manual_${DAP}/*.pdf &
11:15:34amiconnbluebrother: Hmm, so everything between #ifdef and #endif as well?
11:15:44CasainhoBob_C: looks like you just uploaded to lyre hg the file for "triton" and not for theOlimex JATG:
11:16:07amiconnI.e. if I do #if 0 / for / bar / baz / #endif I will end up with 5 empty lines?
11:16:16pixelmatomers: what if you keep all % (and then right after the closing curly braces) except the one above e.g. the "and" in the first part?
11:16:20bluebrothertomers: simply reading that diff shows that this won't work. You need to terminate \opt lines with a comment mark (without trailing space) *except* the last opt line, which needs a space in front of the %
11:16:29CasainhoBob_C: I already build the cable and I am testing it now, I really need that file for Olimex JTAG
11:16:34bluebrotheramiconn: yes, as far as I've checked.
11:16:37 Nick |DaMaGeD|[test] is now known as |DaMaGeD|[hard_w (n=|DaMaGeD@
11:16:53Bob_Ccasainho: interface/olimex-jtag-tiny.cfg is the standard file that comes with OpenOCD
11:16:59tomersOh, that's explain this behaviour :-)
11:17:19pixelmabluebrother: the fun thing is, that it will also work without the % *sometimes*
11:17:28CasainhoBob_C: ok, then I will try to use that comming on OpenOCD
11:18:04*bluebrother points to his statement made at 15 minutes ago (11:02 local time)
11:18:05pixelmabluebrother: and where is the difference between not having a % at all in that last line or " %"?
11:18:42bluebrotherpixelma: true, omitting the % in the last line without having a space should have the same effect.
11:19:11CIA-85New commit by bluebrother (r23222): When cancelling the bootloader installation delete the bootloader install object.
11:20:57CIA-85New commit by tomers (r23223): Remove last comment directive at the end of line of the last \opt
11:25:14 Join polobric1lo [0] (
11:28:16 Quit polobric1lo (Client Quit)
11:29:43linuxstbbluebrother: Do you understand how the website works? It's using template files using C-style #defines and #if. And from what I can tell, there are no blank lines. See "Makefile" in the www module.
11:29:58|DaMaGeD|[hard_wis there someone with Cowon D2?
11:30:22 Nick |DaMaGeD|[hard_w is now known as |DaMaGeD|[code] (n=|DaMaGeD@
11:30:53tmzt|DaMaGeD|[code]: status changes seem to be against the guidelines in topic
11:31:08|DaMaGeD|[code]tmzt: sorry
11:31:31 Nick |DaMaGeD|[code] is now known as |DaMaGeD| (n=|DaMaGeD@
11:31:42bluebrotherlinuxstb: from a quick look it uses fcpp, not gcc. Plus, after running fcpp a sed call removes all lines that are completely empty. Nothing we can do for LaTeX sources :(
11:31:43 Join TheSeven [0] (n=theseven@rockbox/developer/TheSeven)
11:33:11 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
11:33:19 Join polobric1lo [0] (n=polobric@
11:34:03 Quit linuxstb (Nick collision from services.)
11:34:03 Quit polobric1lo (Client Quit)
11:34:08 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
11:34:35 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
11:35:07 Join ShapeShifter499 [0] (
11:35:08tomers|DaMaGeD|[hard_w: Just bought one yesterday. No Rockbox on it yet. Why?
11:35:18 Quit linuxstb (Client Quit)
11:35:31 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
11:35:37 Join polobric1lo [0] (
11:38:40|DaMaGeD|tomers: mean,it's not installing?
11:41:58 Nick polobric1lo is now known as polobricolo (
11:42:33 Join polobricolo_ [0] (
11:44:12tomers|DaMaGeD|: No, Just had no time to do it... :-) BTW how can I help with that port? What areas are still missing?
11:46:31|DaMaGeD|tomers: there's a feature in OF that uses hardware plus\minus\menu buttons as next\prevtrack\play while keys are on hold
11:47:56 Quit polobricolo ("leaving")
11:48:19|DaMaGeD|tomers: i'm doing the same thing,but for rockbox.there was a patch,that did it (FS #9875),but i'm doing a menu for enable\disable this
11:48:26 Join polobricolo [0] (
11:48:49tomersnice. But doesn't it voids the purpose of hold?
11:49:23|DaMaGeD|tomers: yes,but it very useful while on road.
11:50:34tomersSo that's the point of the new configuration setting... Sounds nice, and that it could be useful for the other DAPs as well
11:51:24|DaMaGeD|tomers: I want to add a menu configuring this for those as you:)
11:51:35 Quit |DaMaGeD| ("Back in reality.")
11:51:36TheSevenat least some other DAPs have hold switches that disable the button hardware directly
11:53:04tomersTheSeven: So we could use a new HAVE_SW_HOLD directive, or something similar
11:53:45*TheSeven wonders if HAVE_SW_HOLD or HAVE_HW_HOLD is the way to go
11:54:15pixelmathere are targets with pure software hold without hold switches...
11:54:28tomersTheSeven: Doesn't matter... HAVE_HW_HOLD sounds much better
11:54:36pixelmaand there is a define
11:55:33pixelmafor targets with hold switch
11:55:49*tomers got to go
11:56:51 Join |DaMaGeD| [0] (n=dam@
11:58:16|DaMaGeD|So,just built it,now will test.
11:58:55 Part polobricolo_
11:59:42 Quit bertrik ("Leaving")
12:00:24|DaMaGeD|How do you think,where to place this menu?
12:01:30 Join pamaury [0] (
12:02:43 Join robin0800 [0] (
12:04:57 Quit |DaMaGeD| ("used jmIrc")
12:06:25Casainhoanyone can say t ome waht is the e-mail of jdgordon?
12:08:46Casainhook, I got it from mailing list
12:26:34 Quit robin0800 (Remote closed the connection)
12:29:26 Quit polobricolo ("leaving")
12:30:29 Join robin0800 [0] (
12:30:59 Join Grahack [0] (
12:48:03***Saving seen data "./dancer.seen"
12:57:30 Join DerPapst [0] (
13:02:23 Nick kyle6513_ is now known as kyle6513 (n=kyle@
13:03:10 Join Thundercloud [0] (
13:07:38 Quit n1s (Read error: 60 (Operation timed out))
13:08:01 Join n1s [0] (n=n1s@rockbox/developer/n1s)
13:10:26tomers|DaMaGeD|: How is it going? Need some help?
13:18:04 Join |DaMaGeD| [0] (n=|DaMaGeD@
13:19:20 Join petur [0] (n=peter@rockbox/developer/petur)
13:28:05Casainhohere is the image showing Rockbox firmware running on a new target:
13:28:36 Quit bmbl (Read error: 145 (Connection timed out))
13:29:42|DaMaGeD|Casainho: hmm.very good
13:30:23 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
13:31:04Casainhowe need to work now the audio drivers to have audio. Later I would like to have the touchscreen working, to use it for navigate.
13:32:13|DaMaGeD|Casainho: how can i get mini2440 too?
13:36:01Casainhoyou can buy on Ebay, it costs just $85!
13:36:18Casainhoincluding the LCD and touchscreen!
13:36:27kyle6513so its a development board for ARM9?
13:36:29Dgby714ok thats kool
13:36:33Casainhoare are toons of sellers for that board!
13:37:13Casainhoyes, a dev. board but good enough to make a audio player for home/car −− we just need to provide power and put inside a box
13:37:31kyle6513what are the dimensions?
13:37:57Dgby714hmm that would be kool, put it in the dashboard
13:38:00Casainhokyle6513: see the main page: −− all datasheets and schematics are available!
13:38:27kyle6513ooo thanks casainho
13:38:30Dgby714you can also get it from
13:38:36Casainhoa few other Rockbox developers already bought that board, recently
13:38:53kyle6513:o camera interface
13:39:00|DaMaGeD|Casainho: thanks,will buy it)
13:39:08Dgby714Mini2440 with VGA Board O.o
13:39:18kyle6513yeah haha
13:39:21Casainho|DaMaGeD|: we hope to soon have Rockbox running on full!!
13:39:57 Quit yosafbridge ("Coyote finally caught me")
13:39:59Casainhodon't forget to visit our project page, we will post blog messages with Rockbox port advances:
13:40:06 Join yosafbridge [0] (
13:40:14kyle6513it could be made alot more compact i believe
13:43:25Bob_CI heard some disturbing news that there may be "counterfeit" Mini2440s around
13:43:36 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
13:44:35 Join MethoS- [0] (n=clemens@
13:44:42CasainhoBob_C: what you mean? not working boards?
13:46:15Bob_CI think they work, but maybe quality is not the best. According to Hiteg, who claim to make the "real" Mini2440
13:46:31kyle6513maybe less space ect
13:46:37 Quit linuxstb (Read error: 145 (Connection timed out))
13:47:33CasainhoI can think in problems of electronics circuits, maybe the high speed signals to SDRAM..?? who knows? :-)
13:47:35Bob_CIt looks like one Chinese company designed the original boards, and then the design has been ripped by other manufacturers. But I don't read Chinese, so it's hard to tell
13:48:08kyle6513it could be alot similar to the rip off ipods running apple's firmware, just without the firmware :D
13:48:40Dgby714hmm they have a 7" screen
13:48:43Bob_CThe only thing I can tell on my board is that the SDRAM is not Hynix but some noname
13:48:56kyle65137" screen is pretty good :D
13:49:01kyle6513imagine watching mewvies
13:49:17 Quit flydutch ("/* empty */")
13:49:18kyle6513has usb input
13:49:19Dgby7143.5 or 7 but the 7" one is $176
13:49:25kyle6513either way
13:49:30kyle65133.5 is awesome
13:49:41kyle6513compared to some media players that currently run rockbox
13:49:46kyle6513*looks at apple*
13:49:51kyle6513wait no
13:49:59kyle6513media players that play videos
13:50:01kyle6513thats better
13:50:28kyle6513hey it supports android
13:50:44Bob_CThese things are basically netbooks
13:50:53kyle6513yeah just realised it haha
13:51:00kyle6513pretty slick
13:51:22*Dgby714 wants one now...
13:51:28domonokyCasainho: so you sd-card works without problems ?
13:51:31*kyle6513 wonders if rockbox could have a camera plugin coded
13:51:47*kyle6513 wonders if the wlan could be implemented too
13:52:11*Dgby714 wonders if the usb could be used for wireless?
13:52:26kyle6513they have a little thing there just click accessories
13:52:48Bob_CThere is a USB wifi adaptor
13:52:49domonokykyle6513, Dgby714 everything is possible, it needs just somebody good enough to code it :-)
13:53:06kyle6513good point :)
13:53:19kyle6513haha i think this is getting a bit offtopic
13:53:26kyle6513is there a rockbox-community sort of thing?
13:53:33Dgby714i think there is
13:53:56*pixelma points at the topic...
13:54:09*kyle6513 facepalms self
13:54:49 Join DarkDefender [0] (
13:56:10CIA-85New commit by tomers (r23224): Manual: Add mrobe100's remote control support to the manual ...
13:57:11 Join bimbel [0] (n=Miranda@unaffiliated/bmbl)
13:57:12*tomers Getting "Build expected to complete in 20m 32s, at 14:01:05" but build ends in much shorter time
13:58:39*|DaMaGeD| looks at camera add-on,wondering
13:58:58kyle6513opens alot of possibilities dont it?
13:59:36|DaMaGeD|Mini2440 (on characteristics) is much better than Cowon Q5,but Q5 worth more then $500
13:59:47|DaMaGeD|kyle6513: yeeeah:)
14:00:46kyle6513|DaMaGeD|, look at the wlan module
14:01:05 Quit pamaury (Read error: 113 (No route to host))
14:01:49 Quit bmbl (Read error: 145 (Connection timed out))
14:01:59|DaMaGeD|kyle6513: that's paradise,isn't it?:D
14:01:59 Quit antil33t (Read error: 104 (Connection reset by peer))
14:02:08 Join antil33t [0] (n=Mudkips@
14:02:32Bob_CHow is USB OTG support in Rockbox?
14:02:55kyle6513|DaMaGeD|, imagine it, you can create a completely opensource iphone esque beast for the cost of a very cheap phone
14:03:11kyle6513(well an ipod touch, iphone is a tad out of the scope)
14:03:17pixelmatomers: what's your system clock (and what time zone). The estimated time in that line would be alright for me here.
14:03:35tomersI am UTC+2
14:04:10tomersMy time *now* is 14:04
14:04:29|DaMaGeD|kyle6513: heh,there's skype)
14:04:36kyle6513ah :D
14:04:46tomersThe IRC server time diffs in 16 minutes! That could explain it, doesn't it?
14:04:49pixelmaBob_C: not existent currently
14:04:53*|DaMaGeD| sings:i've got..two tickets to pa-aradise...
14:05:10kyle6513|DaMaGeD|: since we have those 3g internet dongles, think of the possibilities :D
14:05:11|DaMaGeD|sorry for offtop.but skype is good idea
14:05:13*tomers Are the IRC and build server are the same machine? Are their time is correct?
14:05:38*tomers Typo: Is their time correct?
14:05:43 Quit TheSeven (Read error: 110 (Connection timed out))
14:06:46|DaMaGeD|kyle6513: someday we will conquer the world with mini's:)
14:07:07*kyle6513 locks his shotgun and prepares for them to overthrow us
14:07:23pixelmatomers: IRC server doesn't have to do with that, it's an entirely differen system - and for me the time stamps in my client are correct. I'd rather guess for a problem on your machine then but don't know
14:08:34domonokykyle6513, |DaMaGeD|: please put offtopic talk to #rockbox-community
14:08:50|DaMaGeD|domonoky: sorry
14:09:06tomerspixelma: Found the error. I use VM machine, and its time is 16 minutes earlier than real time.
14:10:05pixelmatomers: also, while you are around and added some remote mappings. Could you take a look at my last suggestion on LatexGuidelinesTalk (so you are aware and maybe follow one style if you add things)? Hope we can agree on something...
14:10:23n1sbluebrother: i'd really love to be able to use a real preprocessor in the manual, \opt is just too limited... couldn't we mark intentional blank lines in some way and use sed to remove any blank lines before blanking those marked lines?
14:10:25pixelmabluebrother, others interested in the manual too?
14:10:32 Quit Thundercloud (Remote closed the connection)
14:11:26 Join ShapeShifter499_ [0] (
14:11:37n1sbluebrother: also i wonder about the tag table at the end of the database section in the browsing and playing chapter, what's it for? seems very out of place to me and is not referenced in the text and lacks figure description
14:11:48 Quit ShapeShifter499_ (Client Quit)
14:12:31 Join ShapeShifter499_ [0] (n=chatzill@
14:13:50 Quit bimbel (Read error: 145 (Connection timed out))
14:14:34pixelman1s: I think those are the tags (and virtual tags) you can use in tagnavi syntax. But it doesn't look very helpful without explanation indeed
14:22:07Casainhodomonoky: I don't know about SD Card. Ask to Bob_C, he wrote the code.
14:22:59Casainhodomonoky, Bob_C but I must say that firmware is very slow to load, when compared to the other dev. board I were working (with SPI BUS).
14:23:30Casainhodomonoky: so, did you found a way to flash Rockbox booltoader using factory bootloader?
14:23:47domonokyCasainho: no i justed h-JTAG
14:23:49Bob_CThe loading is slow because of the debug output in SD driver, but due to a bug it doesn't work with the debug off
14:23:57 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
14:24:12domonokyits probably slow because of uart, and big delays.
14:25:28n1spixelma: i think it should go into an appendix similar to the one about custom wps'es then
14:25:29 Quit ShapeShifter499 (Connection timed out)
14:30:51 Quit Grahack ("Leaving.")
14:30:51 Join teru [0] (
14:32:24CasainhoBob_C: ah, ok. We must work it then.
14:32:46Casainhodomonoky: so, you have firmware running already?
14:33:18domonokyCasainho: yes, i have it running, but it has problems finding a partition on my sd-card.
14:33:40Casainhobut are you using the Bob code, right?
14:33:51 Join phanboy4 [0] (
14:35:03domonokyCasainho: yes.
14:35:24Casainhodomonoky: is your card HDC?
14:35:45Casainhodomonoky: HC? −− because looks like this SoC do not support HC...
14:35:49domonokyCasainho: no, its a old 512MB sd card.
14:35:57 Quit lifeless__ (Read error: 104 (Connection reset by peer))
14:36:18Bob_CI think SDHC work , mine does
14:36:36Bob_CBut it may not support the high speed SDHC modes
14:36:37*domonoky thinks SDHC support should not depend on the SoC
14:37:32 Quit |DaMaGeD| ("Back in reality.")
14:37:38Casainhodomonoky: so, I hope soon see a patch on Rockbox SVN, for this port ;-)
14:38:07Bob_CIt may the opposite problem, the SD driver selects a speed too fast for the card
14:39:28Bob_CThe debug output may be useful here...
14:40:44Casainhodomonoky: are you thinking in making JTAG debug? how you usually do debug on Rockbox code?
14:41:34domonokyCasainho: for no i will just use Bob_Cs helpful uart_printf() :-)
14:42:24Casainhodomonoky: ok, and are you putting the firmware file to SD Card everytime you build a new code=
14:42:26domonokyBob_C: from the debug output, it looks like it can sucessfully init the card, but all transfered data is 0 ? (i am still debugging it, so i might be not correct)
14:42:55domonokyCasainho: SD doesnt work for me at moment, so i just flash a new bootloader with jtag to try code.
14:43:44Casainhodomonoky: ok. The question is that I think will be much more productive to use JTAG to load code (even firmware) do SDRAM and run/debug it
14:43:52Casainhodomonoky: what do you think?
14:44:46Bob_CI think storage_init first reads the sector 0 to get the MBR. If this fails or gets bad data, it says "No partition"
14:48:07***Saving seen data "./dancer.seen"
14:48:31domonokyBob_C: thanks i will try to debug this :-)
14:49:20domonokyCasainho: sure, being able to use a normal debugger will be helpfull, but the question is, is it worth the effort ?
14:50:09Bob_CThe SD driver may be missing code to do bank select, I couldn't find much info about that
14:50:42Casainhodomonoky: that's why I would like to know your opinion, since you have experience with developing Rockbox code. But, as for running the code directly and quick, would it be important?
14:50:59 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
14:51:20domonokyCasainho: thats just more comfortable, but i can deal with flashing/making a sd card for testing :-)
14:55:14Casainhodomonoky: ok. I will take some time to try having JTAG debugging using Eclipse
14:57:05Bob_CI'd like to get to grips with Eclipse too, if only cos I can put it on my CV, lol
15:01:00CIA-85New commit by nls (r23225): Use a wrapper function for the very common snprintf(); lcd_puts() sequence to save some binsize
15:02:26Casainhodomonoky: did you tough in put touchscreen working? instead of using the small buttons under the LCD? −− I would prefer, to have the LCD fully cover the board, or I will need to make a board with buttons but that's not very easy since I don't have a connector with that pitch needed.
15:03:31 Quit kyle6513 ("Leaving")
15:03:32domonokyCasainho: rockbox has touchscreen support. but i dont know if you can really control everything woithout buttons.
15:03:50gevaertsyou can. Some targets don't have any buttons
15:04:45Bob_Ctouchscreen should be easy I think, just need to read some ADCs
15:05:44Casainhook. Many work to do ;-)
15:06:00CasainhoI will take a rest now. bye bye
15:06:07 Quit Casainho ("ChatZilla 0.9.85 [Firefox 3.5.3/20091007095328]")
15:09:20tomersIs there is something to do with the build server's script to be tolerant with time incorrectness on the client?
15:09:21*tomers Going now...
15:12:57domonokyhm, strange. it correctly finds the bootsector signature, but all partition-info is 0. Does this mean my sd-card is still in superfloppy mode ? (my tries to find a hexeditor to verify the cards bootsector)
15:15:05 Join funman [0] (n=fun@rockbox/developer/funman)
15:15:53 Quit ShapeShifter499_ ("ChatZilla 0.9.85 [Firefox 3.0.14/2009090216]")
15:23:20Bob_CI dont what a superfloppy is...?
15:23:56gevaertsBob_C: superfloppy mode means a filesystem without a partition table
15:25:47 Join kugel [0] (n=kugel@rockbox/developer/kugel)
15:25:52Bob_COk, makes sense. Sounds like a formatting issue then
15:26:03kugeln1s: impressive delta :>
15:26:09bluebrothern1s: you mean the table with database tags? If seems somewhat out of place, indeed. Might be better placed somewhere in the appendix in some "configuring database views" section or similar.
15:26:16kugelbut why didn't you put it into the lcd driver?
15:26:20bluebrotherno idea why it was put there, but it wasn't me :)
15:26:27kugelso firmware/ code could use it also
15:27:05amiconnn1s: Nice delta, but yellow and red!
15:27:18n1samiconn: yes i know, working on it
15:27:31amiconnAlso, did you check whether some of the other lcd_puts* variants would also profit from this?
15:28:03kugelamiconn: so this was the overkill you were talking about? :p
15:28:22amiconnI didn't expect this combo to be so common
15:28:55amiconnn1s: I wonder why the Player sees zero delta from this. Doesn't it use this combo at all?
15:28:55 Quit linuxstb (Connection timed out)
15:28:55n1samiconn: no i didn't check that thoroughly but after some quick checking i don't think so and also that debug_menu was the only place where this was done a lot
15:29:26n1sbluebrother: ok
15:31:31 Quit krazykit (Remote closed the connection)
15:31:42 Join krazykit [0] (
15:31:50CIA-85New commit by nls (r23226): fix yellow by 'un-staticing' the printing wrapper function and typo causing red
15:32:48 Quit Dgby714 (Read error: 60 (Operation timed out))
15:34:40 Quit Topy44 (Read error: 60 (Operation timed out))
15:34:56n1samiconn: the player code in the debug menu usually only prints 2 lines so for example a big switch deciding what to snprintf is done and then a single lcd_puts, so no real gain from modifying that but i did change a few places
15:36:04 Quit funman ("free(random());")
15:36:58 Join |DaMaGeD| [0] (n=|DaMaGeD@
15:39:36*domonoky was completely wrong with his sd-problems. it correctly finds the partition, but fat_mount fails.
15:40:43CIA-85New commit by nls (r23227): Const correctness for output_dyn_value and unify some identical constants
15:42:28|DaMaGeD|is there some sort of batteries for mini2440?
15:45:19domonoky|DaMaGeD|: no it just wants 5V, (dont know how much A). But you could ofcourse build some sortof batterypack.
15:46:33 Quit phanboy4 ("Leaving")
15:47:34 Join AlexP_ [0] (n=alex@rockbox/staff/AlexP)
15:47:55|DaMaGeD|domonoky: thanks,will think on this...
15:49:06Bob_CIt needs at least 600mA with LCD+backlight, which is a lot of juice for a battery.
15:50:02*domonoky found his error, sd-card wasnt fat32 ! (debuging fat.c gave me "This is not FAT32, go away" ) :-)
15:50:27|DaMaGeD|Bob_C: don't know what chars has Cowon D2's battery?
15:50:41|DaMaGeD|sorry,bu i'm here through GPRS:(
15:51:53 Quit DerPapst ("Leaving.")
15:52:21domonokyBob_C: 600ma is pretty much, is this with everything enabled, or are we able to spare something by powering down not needed units ?
15:53:09 Quit daurn (Remote closed the connection)
15:53:12|DaMaGeD|domonoky: will we implement usb-host?it might be very power-hungry.
15:53:37gevaerts|DaMaGeD|: "we" will not implement anything
15:53:50|DaMaGeD|mean in Rockbox
15:53:52Bob_CThose figures I believe came from running Linux, usb-host would add on top
15:54:02domonoky|DaMaGeD|: usb-host is far away in the future, as rockbox currently doesnt has a usb-host stack.
15:54:16Bob_CThe wall adaptor is rated 2A, if that's any guide
15:54:16 Join daurnimator [0] (i=daurnima@freenode/staff/daurnimator)
15:54:50domonokywe should do some measurements in the future :-) First get it running and into rockbox svn :-)
15:55:42Bob_CA power measurement is on my list of things to do
15:58:01n1smaybe that thing will be a nice in-car system or something
15:58:17kugeln1s: did you see my question?
15:58:33 Join BigBambi [0] (n=alex@rockbox/staff/AlexP)
15:58:56n1skugel: sorry no, gtg for a while now, will look later
15:59:15 Join panni_ [0] (
15:59:30domonokyn1s: yes, the mini2440 is good for car and in-home use.
16:00:26Bob_CMy plan is to use for a car system
16:00:37 Quit AlexP (Read error: 110 (Connection timed out))
16:02:07 Join efyx_ [0] (
16:02:25domonokyis there someone with a gigabeat f here ? to get the mini2440 code into svn, we need to test the changes to the code common to mini2440 and gigabeatf.
16:02:29kugelI was wondering if it was possible to replace to touchscreen in the future
16:02:44kugel4-wire resistive isn't exactly high quality and durable
16:03:26 Nick BigBambi is now known as AlexP (n=alex@rockbox/staff/AlexP)
16:04:09domonokykugel: possible sure, but from my first tests with Qtopia, the touch seems to work pretty good.
16:09:04 Quit AlexP_ (Read error: 110 (Connection timed out))
16:10:04domonokygevaerts, AlexP: could you do a test of the modified drivers with your gigabeatfs ? ( i can build you a zip, if you want)
16:10:27AlexPWhich driver?
16:10:36AlexPah, I see
16:10:48*gevaerts goes to get his f60
16:10:50AlexPdomonoky: A zip would be handy
16:11:09domonokythe patch in the tracker isnt up-to-date. i will build a fresh zip
16:11:11 Join pamaury [0] (
16:13:04 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
16:13:35 Join DG-2 [0] (
16:13:37 Nick DG-2 is now known as Dgby714 (
16:15:33gevaertsdomonoky: anything in particular to watch?
16:16:23domonokynothing particular, as far as i know, the changes are to the lcd driver and system/kernel. so if it works and display is fine, its ok :-)
16:16:29gevaertsdomonoky: it misses the top few rows on the screen
16:17:32AlexPdomonoky: I'm going to have to leave it to gevaerts - I can't find my F :/
16:17:42gevaertsaudio works
16:19:09domonokyoki, so i gonna review the changes to the lcd-driver.
16:19:14 Quit Dgby714 (Nick collision from services.)
16:19:15 Join DG-2 [0] (
16:19:17 Nick DG-2 is now known as Dgby714 (
16:20:34*gevaerts wants kugel to commit the custom status bar
16:21:04AlexPI'm looking forward to that
16:21:50gevaertsI was running a build with it, and domonoky's test build doesn't have it :(
16:24:13 Join Grahack [0] (
16:31:06n1skugel: i wasn't sure if it would be used in other places and should remain local to the debug menu but maybe i should just create a lcd_putsf() like i initially planned to, do you know of any places where such a function would be useful?
16:32:18kugeldebug-<target>.c for ams, ipodnano2g (as _DEBUG_PRINTF _macro_) and gigabeats
16:32:44 Join Topy44 [0] (
16:33:46n1shmm, right, and the debug menu is in apps so that would count as a bad hack with the current location
16:34:01kugelgevaerts: what? *You*?
16:34:36kugelmy current plan is monday
16:35:09 Join uflops [0] (n=yogurt@
16:38:20 Join DG-2 [0] (
16:38:20 Quit Dgby714 (Nick collision from services.)
16:38:29 Nick DG-2 is now known as Dgby714 (
16:42:13kugelAlexP: How would you call an option for track skipping that causes to skip ~5s before the end of a song
16:42:52AlexPSo once it gets to 5s before the end, it skips to the next track?
16:43:22kugelno, if you press next, it jumps to 2:55 if the track was 3:00 minutes
16:43:31 Quit matsl (Remote closed the connection)
16:43:44AlexPWhy would you want this?
16:44:04AlexPLead out previous track on skip?
16:44:34AlexPThat's a difficult one to have a short name for - you pretty much have to describe it
16:45:15AlexPYou don't need previous there actually
16:45:28AlexPPlay lead out on skip?
16:45:36kugelwell, I have some gapless albums where the track transition is worth listening to
16:46:04AlexPWell, it is up to you - seems a pretty marginal option to me :)
16:46:08kugeland since we already have various track skip options, I figured "Why not?"
16:47:19kugelmaybe some outro?
16:47:37kugelas opposed to intro (not sure if outro is an actual word)
16:47:38AlexPPlay xxx on skip
16:48:03AlexPWhere xxx describes the last few seconds - lead out, outro, ...
16:48:09***Saving seen data "./dancer.seen"
16:48:46AlexPOutro isn't really a word, but is sometimes used for this
16:48:50AlexPas is lead out
16:49:08AlexPI don't like either lead out or outro, but I'm not sure if there is a better option
16:49:08 Quit Dgby714 (Read error: 60 (Operation timed out))
16:49:12*kugel has never heard lead out for this before
16:49:23AlexPas opposed to lead in
16:50:18kugelbut intro is more commonly used than lead in, isn't it?
16:50:33AlexPbut neither are very nice - intro works as it is short for introduction, and lead in you are leading into the song. You can't lead out as the verb lead doesn't work there, and outro is just a made up word
16:50:55AlexPBut as I say, I can't think of a better option
16:51:00AlexPMaybe others can :)
16:51:42 Join BBBradley [0] (
16:51:55 Quit mikroflops (Read error: 110 (Connection timed out))
16:51:55kugelI fear a "color of the bikeshed" discussion if I ask more people ;)
16:52:06AlexPkugel: That'd be colour :)
16:52:23AlexPDo we use outro anywhere else?
16:52:38AlexPWhat does crossfade use?
16:53:02kugelfade out
16:54:56AlexPkugel: Probably easiest just to go for outro
16:55:16domonokygevaerts: can you try again with please ?
16:55:26kugelwell, I'm fine with describing it proper in the manual, I just want a short descriptive name for the option in rockbox
16:55:34 Quit BBBradley (Client Quit)
16:55:54kugelAlexP: alright, thanks a lot
16:55:58AlexPkugel: I believe "coda" is the technical term in (classical) music
16:56:31AlexPkugel: Yeah, so maybe "Play outro on skip"
16:58:38gevaertsdomonoky: 404
16:58:48kugellet's see if I still have the code for that
16:59:55kugelI do, cool
16:59:56domonokygevaerts: up, try again no please
17:00:18domonokys/up/ups/ s/no/now/
17:01:18 Quit teru ("Quit")
17:01:19gevaertsdomonoky: the lcd problem seems to be gone
17:03:23domonoky Bob_C: there are a few mistakes in your lcd-target.h for gigabeat f: the values for LCD_UPPER_MARGIN and LCD_LOWER_MARGIN are swapped, and LCD_LEFT_MARGIN is 9 not 5 (as in the comment).
17:03:50 Quit |DaMaGeD| ("Back in reality.")
17:04:13domonokyBob_C: can you fix that in hg, and provide a fresh, uptodate patch to the tracker ? then i can review and commit it easier :-)
17:06:47Bob_Cdomonoky: thanks for the feedback, I will do that
17:07:47Bob_Cdomonoky: about the button names, I didn't see any other target using "BUTTON_ONE" etc.. but I can make that change as well
17:08:31 Join enthdegree [0] (
17:08:44enthdegreerockblox1d seems to be a rather stale game.
17:09:43domonokyBob_C: i am fine with how the buttons are in hg. and we can still change it if we want, when we have it in svn.
17:13:21Bob_Cok, I will fix up the LCD issues then
17:14:46 Join BBBradley [0] (n=40832529@
17:18:41 Quit BBBradley (Client Quit)
17:20:25CIA-85New commit by tomers (r23228): USB HID: Add mouse mode to h10 ...
17:21:36 Join midgey [0] (
17:26:14enthdegreeWill the Anti-Aliased fonts patch work on any device?
17:26:38kugelAlexP: "Skip to Outro" works
17:26:48kugelenthdegree: on color screens, yes
17:27:50enthdegreeOk, great. :D Also, bubbles is unplayable on the c200 (there's already a bug for this)
17:28:18 Join DG-2 [0] (
17:28:27 Nick DG-2 is now known as Dgby714 (
17:28:59 Quit tomers ("ChatZilla 0.9.85 [Firefox 3.0.14/2009090216]")
17:31:44pixelmalooks like it uses some wrong screen defines
17:33:26domonokyBob_C: another change: move debug-target.h into gigabeatf and mini2440 folder (or use a ifdef), then you can define your own DEBUG_CANCEL button, and dont need to change how you exit the debug screen.
17:37:03Bob_Cthat's well spotted, I missed that
17:39:50 Quit JackWinter (Remote closed the connection)
17:41:20 Join markatto_ [0] (n=markatto@
17:42:10 Join JackWinter [0] (
17:42:18 Join funman [0] (n=fun@rockbox/developer/funman)
17:44:09 Join |DaMaGeD| [0] (n=dam@
17:44:48pixelmathe "Next" line itself and the next bubble and the start of the current bubble are shifted down a bit
17:45:19 Join FOAD_ [0] (
17:45:20pixelmawhich is outside the screen for the latter two things
17:47:07 Quit markatto (Read error: 60 (Operation timed out))
17:48:31 Join Farthen [0] (
17:50:04 Quit bubsy (Read error: 60 (Operation timed out))
17:51:52 Quit |DaMaGeD| ("used jmIrc")
17:53:54 Join TheSeven [0] (n=theseven@rockbox/developer/TheSeven)
17:56:49kugelAlexP: "The \setting{Skip to Outro} option skips nearly to the end of a song, so that the
17:56:49kugel last few seconds are played."
17:58:00kugelor rather "The \setting{Skip to Outro} option changes the behavior so that buttons skip to the near end of a song, so that the last few seconds are played." ?
17:58:20AlexPkugel: I'd have "... skips to just before the end of the song, so that the last few seconds are played before the bext track"
17:58:56n1syou have a song/track inconsistency there
17:59:05AlexPsong/track/whatever :)
18:00:57kugelthat works
18:00:59AlexPIt was the kugel: I like "The \setting{Skip to Outro} option changes the behavior so that the buttons skip to just before the end of the song, so that the last few seconds are played before the next song"
18:01:10AlexPer, ignore "It was the" :)
18:01:25AlexPkugel: Also, it is "behaviour"
18:01:43kugelso, song not track?
18:02:09kugelthe existing text uses track, since it was (originally) made for audio books
18:02:13AlexPkugel: I don't care, but I guess it should be consistent
18:02:17kugelI rather keep track here
18:02:17 Quit FOAD (Read error: 110 (Connection timed out))
18:02:18 Nick FOAD_ is now known as FOAD (
18:02:22AlexPkugel: Sure :)
18:04:00enthdegreeWhere does "settings_dumper" dump my settings?
18:04:15AlexProot I'd guess, or /.rockbox
18:04:25kugelI wonder about the point of that plugin
18:04:45kugelwhat's the difference to manage settings->save config?
18:05:10AlexPone only does those that aren't default, the other does all IIRC
18:05:25CIA-85New commit by kugel (r23229): Add a option to skip length which lets you skip 5s before the end of a song, which can be useful on gapless albums.
18:05:28AlexPI guess so you can see what can be changed (for themes etc)
18:05:29CIA-85New commit by kugel (r23230): Update manual for the new "Skip to Outro" feature.
18:06:11*AlexP kicks kugel
18:06:15AlexPkugel: "behaviour"
18:06:25kugelI changed that, didn't I?
18:06:37kugelwhat the heck
18:06:54kugelI copy&pasted your correction, it's somehow gone
18:07:02AlexPhehe :)
18:07:15AlexPYou can increase your commit count by one :)
18:07:36kugelmaybe I didn't hit ctrl+s
18:07:57AlexPThere are some other misspellings of behaviour in there
18:08:03AlexPThe line just before for one
18:08:30kugelwhat would be correct?
18:10:02 Quit FOAD ("I'll be back")
18:10:13kugelI mean the other line
18:10:29AlexPkugel: I meant that behaviour was misspelt in the other line
18:10:31kugelI see what you mean
18:10:37kugelI got it wrong
18:10:42AlexPAnd once more in that file too
18:10:52AlexP(Unrelated to your commit, I just spotted them)
18:10:57AlexPI can fix if you like
18:11:08kugelnah, I'll do it
18:11:13AlexPOK, cool
18:11:26kugelAlexP: unless you have some more, then I'll let you
18:11:43AlexPGo ahead
18:11:52AlexPI'll check other files later, but I have to go very soon
18:12:09AlexPI spot three that need changing in playback_options.tex
18:12:36kugelAlexP: I just wanted to be consitent with the other typos, you know :P
18:12:46AlexPhehe, consistency is key :)
18:13:01n1shmm, doing a lcd_putsf in the lcd drivers presents a dilemma since one of the benefits of debug_printf is that it assumes the x arg to always be 0, doing this for lcd_putsf is ugly
18:13:54CIA-85New commit by kugel (r23231): Correct a few misspellings of behaviour.
18:14:19n1sso i can make it clean and replace debug_printf with lcd_putsf and waste binsize or have both functions which is ugly but wastes less binsize...
18:14:28kugeln1s: you can still have the wrapper in debug_menu
18:15:04kugelusing a macro or so
18:15:16 Join bubsy [0] (n=bubsy@
18:15:45n1skugel: sure but a macro will waste binsize just as calling lcd_putsf directly
18:15:49 Quit funman ("free(random());")
18:16:27kugelhow will that waste binsize? the additional parameter?
18:16:47n1skugel: yes
18:16:52 Join FOAD [0] (
18:17:18kugelaren't you exaggerating a bit?
18:17:35n1sno :)
18:17:39 Quit barrywardell ()
18:18:08kugelI think you are
18:18:27n1syeah, maybe, this would have been so easy if i could just pass on the variadic args easily
18:18:51kugelyou can with macros at least
18:19:44n1syes, but doing this with a macro will waste way more binsize than that extra param
18:20:07 Join funman [0] (n=fun@rockbox/developer/funman)
18:20:22*n1s goes for clean
18:22:25enthdegreeHow well does this antialias patch work? Anyone using it?
18:23:11 Join fdinel [0] (
18:23:20 Join toffe82 [0] (
18:23:47 Quit Grahack ("Leaving.")
18:24:36pixelmaon the c200 display?
18:25:17enthdegreeOn anything.
18:35:34 Quit bubsy ("Party time, EXCELLENT! *wring wring wring!!* Party on, Wayne! Party on, Garth!")
18:35:45 Join bubsy [0] (n=bubsy@
18:37:04enthdegree patch < ../anti-alias_fonts.patch asks me what file I want to patch.
18:39:05 Join moufsnafa [0] (i=18796437@gateway/web/freenode/x-eyrdccvqhbzcnwvr)
18:39:55enthdegreepatch -p1 < ../anti-alias_fonts.patch returns that patching of one file failed. What am I doing wrong?
18:39:56linuxstbenthdegree: You need a "-p" parameter - normally "patch -p0 < ...." works, if not try -p1
18:40:00n1senthdegree: you need to specify the correct strip level, such as -p0 for example depending on the patch
18:40:36linuxstbenthdegree: It's also likely that the patch is out of sync with current SVN - so you'll either need to fix the patch, or use it with an older version of Rockbox.
18:41:33 Join DerPapst [0] (
18:42:12 Quit funman ("leaving")
18:43:50moufsnafaAre there any plans for musepack sv8 support?
18:45:11 Join saratoga_ [0] (i=463f90ed@gateway/web/freenode/x-bilijkdelsuqoxqc)
18:45:41enthdegreeThis patch appears to be about a month old.
18:46:41saratoga_i don't think anyone is working on sv8 but it would be nice to have
18:46:51 Join BHSPitLappy [0] (n=BHSPitLa@unaffiliated/bhspitmonkey)
18:48:11***Saving seen data "./dancer.seen"
18:48:16moufsnafaI would work on it myself, but I have to learn C first.
18:54:19 Join BdN3504 [0] (
18:56:46 Part moufsnafa
18:57:14 Quit enthdegree ("Leaving")
18:59:41BdN3504i have an ipod 5.5g and think it has a hard disk failure. i want to replace it, but i can't find proper instructions on its disassembly. i searched the main page and the forums but i can't find any info
18:59:44 Quit kugel (Read error: 104 (Connection reset by peer))
18:59:56BdN3504could you point me to a guide or something?
19:02:28 Join HBK- [0] (
19:02:38 Quit bmbl ("Bye!")
19:03:25 Quit saratoga_ ("Page closed")
19:05:36 Join Grahack [0] (
19:09:19 Quit robin0800 (Remote closed the connection)
19:12:32 Quit BdN3504 ("CGI:IRC (EOF)")
19:18:55*TheSeven wonders why nobody seems to know about ifixit
19:19:36 Quit KBH (Read error: 110 (Connection timed out))
19:35:35 Join tomers [0] (n=chatzill@
19:36:38 Quit esperegu (Read error: 104 (Connection reset by peer))
19:36:51 Join esperegu [0] (n=quassel@
19:40:02 Join funman [0] (n=fun@rockbox/developer/funman)
19:47:56CIA-85New commit by tomers (r23232): Manual: Fix button tables according to guidelines ...
19:49:22 Quit bubsy ("Party time, EXCELLENT! *wring wring wring!!* Party on, Wayne! Party on, Garth!")
19:49:24 Join kyle6513 [0] (n=kyle@
19:50:02 Quit kyle6513 (Client Quit)
19:50:38 Quit toffe82 (Remote closed the connection)
19:56:31 Join toffe82 [0] (
20:01:38 Quit MethoS- (Remote closed the connection)
20:02:50CIA-85New commit by nls (r23233): Revise r23225 a bit, removing the debug_printf function and implementing more generic lcd_(remote)_putsf function(s) instead and use those in more ...
20:11:29gevaertsn1s: interesting deltas :)
20:12:26CIA-85New commit by nls (r23234): fix red
20:13:13n1sgevaerts: they actually make sense :)
20:13:47funmanif you continue to lower binsize, you'll render the Clip usable!
20:14:24n1sbasically size increased a little for every target due to the change in debug_menu but decreased for some targets which have target specific code that now use this putsf function
20:14:35n1sfunman: isn't the clip usable?
20:14:58funmanno, the very small audiobuffer highlights some bugs in buffering code (see FS #10605)
20:15:13funmanplayback crashes very fast in SVN on Clip/m200v4/c200v2
20:16:26n1saha, don't think shrinking the binary by a few kb's will help much though :)
20:18:41funmanI believe the problems happen when a buffered handle (like a codec) is bigger than half of audiobuffer, so a few kb's could make the difference
20:18:56n1sfunman: btw, maybe it's a good idea to use the -Os switch for gcc for those targets with very little ram
20:19:25n1sif we aren't already?
20:19:26funmanhmm can't remember if I tried to use it
20:21:03funmanconfigure does s/-O/-Os/ for m68k and sh
20:21:56n1syeah, we use Os for those
20:22:45saratogafunman: i don't think its just codec files though, since it'll still crash with codec buffering disabled
20:22:46n1s...for the core and most plugins, codecs and some plugins set their own switches
20:22:54saratogabut you probably knew that already
20:23:30 Join mc2739_ [0] (n=mc2739@rockbox/developer/mc2739)
20:23:42funmansaratoga: if we have spotted the buggy thing in the fs# , it would crash with any too big handle (a sound file which has enough data buffered already for example)
20:23:43 Quit mc2739 (Nick collision from services.)
20:23:45 Nick mc2739_ is now known as mc2739 (n=mc2739@rockbox/developer/mc2739)
20:24:23saratogaah misread what you had said
20:24:35saratogayeah i'm hoping thats the only serious bug too :)
20:24:43 Quit Grahack ("Leaving.")
20:24:57 Quit n17ikh ()
20:25:06funmanlet's hope dionoea will figure a good algorithm for this
20:25:21 Join p3tur [50] (n=petur@rockbox/developer/petur)
20:25:23saratogaits werid theres no standard function for doing this on ring buffers
20:25:27saratogaseems like a common problem
20:27:43funmanwe could look at existing ring buffers (only thing that comes to my mind is malloc() )
20:37:20pixelmadoes bubbles look correctly on an Ipod Video currently?
20:38:52 Quit stoffel (Remote closed the connection)
20:41:55 Quit saratoga ("Page closed")
20:48:08pixelmasim it is then :\
20:48:12***Saving seen data "./dancer.seen"
20:52:04Tornepixelma: define correctly, and i'll have a look?
20:52:52pixelmaif there is something that looks off in the y-axis, maybe the next bubble box or even offscreen
20:53:05 Quit tomers ("ChatZilla 0.9.85 [Firefox 3.0.14/2009090216]")
20:53:17pixelmahmm... shuffle the words around a bit ;)
20:54:10TorneYeah, the next bubble box
20:54:18Tornethe text "next" and the bubble is too low
20:54:21Tornethe box itself looks rightr
20:54:45Tornethe rest of the screen looks ok
20:55:05Tornethe top row of bubbles is right up against hte top of the screen but i dunno if that's riht. they aren't offscreen, there' s jut no top border
20:56:09pixelmaI guess the start position of the current bubble is too low too but if it is still on the screen, it won't lokk wrong, only if you compare too an older build or the screenshot in the manual
20:56:36 Join phanboy4 [0] (
20:56:43pixelma I know what's wrong then, now to find a "nice" way to fix it... and it's broken since 21664
20:57:04pixelmarevision that is ;)
20:57:34*pixelma wonders about typos today :/
20:58:09 Quit esperegu (Read error: 104 (Connection reset by peer))
20:58:25 Join esperegu [0] (n=quassel@
21:03:54 Join esperegu_ [0] (n=quassel@
21:12:15 Join dmb [0] (n=Dmb@unaffiliated/dmb)
21:16:54 Quit dmb (Read error: 104 (Connection reset by peer))
21:18:44 Quit esperegu (Read error: 113 (No route to host))
21:19:29 Quit phanboy4 (Read error: 104 (Connection reset by peer))
21:22:45 Join phanboy4 [0] (
21:27:14 Join bubsy [0] (n=bubsy@
21:30:50 Quit funman (Read error: 104 (Connection reset by peer))
21:41:40 Join Thundercloud [0] (
21:49:31pixelmakkurbjun: you around?
21:58:37 Nick fxb__ is now known as fxb (n=felixbru@
22:06:38 Join ShapeShifter499 [0] (
22:15:57 Quit phanboy4 (Read error: 104 (Connection reset by peer))
22:16:34 Join casainho [0] (
22:16:44 Quit HBK- ()
22:18:34casainhodomonoky: did you saw the picture I took of debugging using Eclipse?
22:19:02 Quit p3tur (Remote closed the connection)
22:19:20domonokycasainho: nice :-)
22:19:55casainhodomonoky: and you got arround the problems on SD Card drivers?
22:20:15domonokycasainho: yes, my card wasnt fat32 :-)
22:20:23casainhodomonoky: eheh :-)
22:21:12casainhodomonoky: so, for when the first code on Rockbox source? will you do it?
22:21:36casainhoI am curious if JdGordon made some advances...
22:22:19domonokycasainho: yes, i want to get it into rockbox svn fast..
22:22:49casainhodomonoky: nice :-)
22:23:13casainhodomonoky: what is the next task? work the audio drivers?
22:24:31domonokybesides getting it into svn, yes audio and touch, sd-driver cleanup.. there is much todo :-)
22:24:56 Join HBK [0] (
22:25:47domonokyfor getting it into svn, i need a new, uptodate patch, which the fixes i told Bob_C :-)
22:26:09casainhodomonoky: why can't you make that changes by yourslef?
22:27:01pixelmapetur: with your recording screen commit - are you aware you removed some convenience for at least the c200? It's also not what I expected from the commit after following the forum thread with the H100 (not a 100% sure if you posted the patch you committed then somewhen shortly before the actual commit)
22:27:20 Join Xerion [0] (
22:27:45domonokycasainho: lazyness ? :-)
22:28:36pixelmapetur: just saying because you were so "angry" about the same things by MarcGuay. I'm not sure yet what to think about the "consequences" for the c200 as it also gives a chance for some other improvements
22:28:44peturpixelma: the change has nothing to do with the patch I put on FS earlier, I rejected that myself and found a better solution. But not ok for c200?
22:29:02petursince the complaints were from ipod, I only checked with those :/
22:29:26casainhodomonoky: ok. There is also DMA, I think there none code done yet for DMA. Did you find any code drivers from FriendlyARM that can help you?
22:31:22domonokycasainho: the code from the 2440test application is pretty nice to get a grip of the target, but till now i am still exploring and lerning about this target..
22:31:23peturpixelma: looking at - what's the problem
22:32:10petura guy at work has a c200, I guess I could ask his player to try things if needed
22:32:10pixelmapetur: see... well the c200 has a Rec button and it's logical to use that to start recording but - the Rec button is so weird at the side of the c200 that it was also convenient to start a recording with the Select button which you now can't do anymore. Thing is that it might be possible to put the recording screen context menu on long select then (like other context menus) which I stepped back doing because it would have been easy to accidentally
22:32:10pixelma start the recording if you want to access the menu if it's on the same button
22:32:44pixelmaso I'm a bit torn what to do with that now
22:33:15casainhodomonoky: ok, sent you an e-mail with the list of files that Samsung provides but not public. You can try to register and get them if you are interested.
22:33:17peturwhat you're saying is that things were better before MarcGuay made his first change to it?
22:33:30peturie rec and select start recording
22:34:22pixelmaI don't know about both but I got used to the Select button
22:34:37petursorry bout that
22:35:00 Quit casainho ("ChatZilla 0.9.85 [Firefox 3.5.3/20090824085743]")
22:35:08pixelmaprobably my problem then. And as I said... my point was that it's not easy to think about everything
22:35:34peturindeed... keymaps are tricky
22:36:29peturgtg - my daughter needs me
22:36:33pixelmaby the way - your point about the ifdefs (in the forum thread). There must be some already because the OndioFM does not have the recording split action due to the lack of buttons
22:36:34 Quit petur ("Zzzzz")
22:48:16***Saving seen data "./dancer.seen"
22:53:48 Quit bluebrother (Nick collision from services.)
22:53:51 Join bluebroth3r [0] (n=dom@rockbox/developer/bluebrother)
23:02:26 Quit togetic ("WeeChat 0.3.0")
23:05:23 Join robin0800 [0] (
23:06:05 Join mrtok1 [0] (
23:06:22kkurbjunpixelma: what's up?
23:07:10mrtok1does datasheets of the pp5022c processor exist in th epublic?
23:08:08mrtok1i try to find out why running my code on the cop is producing artefacts...
23:08:21mrtok1ipod nano 1g
23:11:59mrtok1okay next time ;-)
23:12:11 Quit mrtok1 ()
23:14:11pixelmakkurbjun: your logic for getting the row height in bubbles (rev 21664) only works for the medium sizes - it's off for small display (you already kept some exceptions for the really small displays= and it's off for the big displays. Can you think of something else (other than more exceptions or maybe change the bitmap sizes)?
23:14:13 Join moos [0] (n=599ef66f@
23:14:49moosHello here
23:15:22 Quit ShapeShifter499 (Remote closed the connection)
23:15:39pixelmakkurbjun: at the moment the game is unplayable on the c200 (next bubble and the bubble to shoot) is completely offscreen and on the m3 those two are half offscreen
23:15:45moosI have an svn question please: How do I change the username using vmware here and logged as root?
23:16:51pixelmakkurbjun: I'm trying to think of something too but haven't come up with something yet that solves all three "categories"
23:17:44 Join bertrik [0] (n=bertrik@
23:18:20kkurbjun21664- that seems to be by mcnuelenaere? I'll have to take a look at the logic and see what values it is calculating. I am not sure offhand
23:19:02 Join ShapeShifter499 [0] (
23:19:06pixelmasorry, r21945
23:22:46moosggr it don't let me set a password for my username now
23:25:04CIA-85New commit by moos (r23235):
23:25:54pixelmano commit message?
23:26:19moosOh no, it didn't pick up the commit message :(
23:26:28moospixelma: not intentional
23:26:45 Join Jaykay [0] (
23:26:45moos"Updated the french translation" that was
23:27:30*moos isn't using his pc
23:28:18Jaykaymoos: and you overlooked one translation - it still says 6 to 1 :)
23:28:43pixelmakkurbjun: I analysed a little, for small displays row height = emblem height (up to the M3), then medium displays and even big ones except the M:Robe500 it's emblem height +2, the current logic breaks with gigabeat/video and up
23:29:17moosJaykay: oh my bad :(
23:29:52moosfixing thanks
23:31:44 Quit killan ("( :: NoNameScript 4.22 :: )")
23:32:40 Join togetic [0] (n=togetic@unaffiliated/ibuffy)
23:33:13kkurbjunhmm, it looks appropriate on the 500, H300, and gigabeat at least for the game board - it might be that different logic is needed for positioning the next shot and shooting bubble
23:33:20CIA-85New commit by moos (r23236): Updated french translation part II, sorry for the previous version where the commit message was skipped.
23:33:48kkurbjunI guess as the bubbles move down thoughthe inaccuracy might start to show
23:34:13 Quit domonoky (Read error: 104 (Connection reset by peer))
23:34:56 Quit ShapeShifter499 ("ChatZilla 0.9.85 [Firefox 3.0.14/2009090216]")
23:35:12pixelmakkurbjun: it didn't look correctly for me in a M:Robe500 sim - look at the placement of the "Next" and the actual next bubble compared to the box in the background picture
23:35:54pixelmaH300 is correct with the current logic - down to the Mini's display (138x110)
23:35:56kkurbjunyeah, I agree, that is definately wrong, but I think row height is also used to position the bubbles int eh game board
23:36:16 Quit moos ("Goodbye, see you soon")
23:36:34kkurbjunfor the game board the bubbles look right - I need to look into where ROW_HEIGHT is used more and how it is used
23:36:48TheSevengevaerts: around?
23:37:04TheSevenI caught that stupid bug that was fooling me for days! guess what it was...
23:37:21*bertrik guesses missing ;
23:37:27pixelmayes probably - and for Video and Gigabeat it's one off compared to before (it was 18 and is now 19 they way it is calculated)
23:37:44TheSevennot sure why this caused lockups, but at least I found the root cause of the whole trouble
23:38:12*bertrik guesses again: not clearing the USB interrupt correctly
23:38:19pixelmakkurbjun: the problem really is the c200 and M3 at the moment
23:38:22TheSevenbertrik: wrong
23:38:49*TheSeven guesses control ACK packets being STALLED, but isn't sure of it either, as he doesn't have a bus tracer
23:38:52bertrikbuffer overflow? stack too small?
23:39:39 Join killan [0] (n=nnscript@
23:39:44TheSevenno, it was things being done in the wrong order in the USB core, but this only shows up if certain timing constraints are met
23:40:34TheSevengevaerts: is the USB core expecting the driver to re-setup the control out pipe for the next SETUP transaction automatically? I can't see any api being called to do that.
23:40:57bertrikI thought only endpoints could be stalled, not ACK packets, and only from the device side
23:41:23TheSeventhis in combination with a receive request getting sent too late caused the whole mess
23:42:50TheSevenbertrik: host=>device: SETUP GET DESCRIPTOR, <pipe getting set up for the next SETUP packet>, device=>host: Descriptor data, host=>device: ACK (0-byte packet), device=>host: STALL (expecting a setup packet), usb core => usb driver: receive 0 byte packet
23:42:57TheSevensomething happening the wrong way round, huh?
23:43:14TheSeventhe way it's supposed to be is:
23:43:48TheSevenbertrik: host=>device: SETUP GET DESCRIPTOR, <pipe getting set up for the next SETUP packet>, usb core => usb driver: receive 0 byte packet, device=>host: Descriptor data, host=>device: ACK (0-byte packet), device=>host: ACK, <pipe getting set up for the next SETUP packet>
23:43:53pixelmakkurbjun: I doesn't look completely right to me in a D2 sim (320x240 as the Video) because there is a small gap between the bubbles
23:44:48 Join stoffel [0] (
23:44:53TheSeventhe above was happening if things took too long for the ACK to slip in before the recv() call was being done, which was happening because of debugging stuff
23:45:03TheSeven(enabling logf killed it)
23:45:36TheSevengevaerts: what's the proper way to solve this?
23:45:52 Quit bertrik (Read error: 131 (Connection reset by peer))
23:45:57 Join bertrik [0] (
23:46:11 Join phanboy4 [0] (
23:51:30tmztTheSeven: somebody is working on usb networking?
23:51:41TheSevenno idea
23:51:51tmztwhat is recv() that's not a socket call?
23:52:09TheSevenshorthand for usb_drv_recv()
23:54:16 Quit stoffel (Remote closed the connection)
23:58:03 Join Strife89 [0] (

Previous day | Next day