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 2008-06-12

00:00:32gibbon_i changed that to something more usable...
00:00:55gibbon_it builds without warnings and i'm going to test it...
00:01:08gibbon_thanks so far
00:01:31gibbon_i'm going to bed now... sleep well (in your appropriate timezone :))
00:06:39 Join sarixe [0] (
00:08:23 Quit davina (Remote closed the connection)
00:09:38 Join ender1 [0] (i=krneki@
00:09:50 Quit Schmogel ("Miranda IM! Smaller, Faster, Easier.")
00:14:38 Quit nplus (Remote closed the connection)
00:24:41amiconnlinuxstb: around?
00:26:10 Quit ender` (Read error: 110 (Connection timed out))
00:31:56 Quit gregzx ("ChatZilla [Firefox]")
00:41:06 Quit KevinClift ("CGI:IRC (EOF)")
00:45:16 Quit faemir ("Leaving")
00:50:20 Quit desowin ("KVIrc 4.0.0 Insomnia")
00:50:24 Quit bertrik ("bye")
00:51:19 Join Shaid [0] (
00:52:09 Quit jhMikeS (Read error: 110 (Connection timed out))
00:52:28linuxstbamiconn: Yes
00:53:04 Join Llorea1 [0] (
00:53:13 Join einhirn_ [0] (
00:53:27 Join jhMikeS [50] (n=jethead7@rockbox/developer/jhMikeS)
00:53:28 Quit Llorean (Read error: 113 (No route to host))
00:54:43 Quit jgarvey ("Leaving")
00:58:47*linuxstb leaves again
00:58:58amiconnlinuxstb: Did you see this:
00:59:20 Quit ender1 (" If I know of any heroes in the land, I will not under any circumstance kill their mentors, teachers, and/or best friends. -")
00:59:24amiconnI wonder whether this could be caused by some init code that's present in ipl but missing in rockbox
00:59:35amiconn(for type 0 lcd)
01:00:47 Quit troffo ("Leaving.")
01:00:54amiconnIt almost looks like swapped bytes, and that init code might change a controller mode to the opposite byte order
01:01:07amiconnUnfortunately we still don't know what tpye 0 actually is
01:09:05amiconnThis is the code I am referring to: (in fb.c)
01:09:32***Saving seen data "./dancer.seen"
01:23:27 Join Dhraakellian [0] (
01:24:53 Join krazykit` [0] (
01:33:04 Quit csc` (Read error: 104 (Connection reset by peer))
01:36:02 Part toffe82
01:38:24 Quit advcomp2019 (Read error: 104 (Connection reset by peer))
01:38:29 Quit krazykit (Read error: 110 (Connection timed out))
01:38:38 Join advcomp2019 [0] (n=advcomp2@unaffiliated/advcomp2019)
01:44:14 Join homielowe_ [0] (n=eric_j_l@
01:44:14 Quit homielowe (Read error: 104 (Connection reset by peer))
01:46:06 Join fdinel [0] (
01:46:47fdinelhey Bagder, are you online? :)
01:46:53 Join miepchen^schlaf [0] (
01:48:50 Quit advcomp2019 (Read error: 104 (Connection reset by peer))
01:48:52 Join advcomp2019_ [0] (n=advcomp2@unaffiliated/advcomp2019)
01:50:34 Nick advcomp2019_ is now known as advcomp2019 (n=advcomp2@unaffiliated/advcomp2019)
01:50:54 Quit knittl ("-- visit")
01:53:54pixelmafdinel: he's almost always online but probably not around now (almost 2am here in central europe)
01:54:22 Nick Llorea1 is now known as Llorean (
01:54:25 Quit audijack (Read error: 110 (Connection timed out))
01:54:52fdinelpixelma: ok thanks :) I'll send him a PM on the forums
01:59:45linuxstbamiconn: Yes, that seems to be the issue. The cvs log message that added those lines to IPL said "fix for ipod photo 1.1 color problem". Feel free to commit it - it's too late for me now, and I probably won't have chance tomorrow.
02:02:50 Quit dabujo (Read error: 104 (Connection reset by peer))
02:05:28 Quit vort3x (Remote closed the connection)
02:05:33 Quit efyx (Remote closed the connection)
02:07:02 Join d1sturb [0] (
02:07:46d1sturbAre you able to uninstall rockbox from 1st gen iPod Nano?
02:08:18LloreanAny device you can install it on, you can uninstall it from.
02:08:27LloreanThe instructions for both are in the manual
02:08:58d1sturbThanks. I just found this program now while googling a way to watch movies on it! =D
02:14:23 Quit PaulJam_ (".")
02:21:31 Nick krazykit` is now known as krazykit (
02:22:02 Join Zom_ [0] (
02:22:39 Quit Zom (Read error: 104 (Connection reset by peer))
02:22:43 Quit EspeonEefi ("さよなら")
02:24:51 Join EspeonEefi [0] (
02:25:40 Join mindheavy [0] (
02:27:42mindheavyi thought i might be able to hook a palm pilot to a usb hard drive with music and use the palm to browse music, and play thru my home stereo, from what i can tell, you cannot connect an external drive to a palm, would anyone have any idea's of something small i could use to browse and play music from this usb drive?
02:28:23d1sturbiPod nano 1st gen getting an error: Can't load rockbox.ipod: bad checksum
02:29:21 Join jac0b [0] (
02:30:08jac0bhey JdGordon
02:30:24Lloreanmindheavy: This has nothing to do with Rockbox, though...
02:30:28Lloreand1sturb: Try reinstalling
02:30:55JdGordonjac0b: about to head of... ill reply in a few hours
02:30:55 Quit sarixe ("(EE) Failed to load "quit" module")
02:31:04jac0bis the margins_condvp_merged.3.diff the conditional vp and lcd margins patches combined?
02:31:19mindheavyif possibly there was something i could use with rockbox, it could have something to do with it
02:31:43LloreanRockbox doesn't have USB Host.
02:31:58JdGordonjac0b: yes
02:32:07 Join cool_walking_ [0] (
02:32:11jac0bJdGordon: thanks
02:32:17 Quit jac0b (Client Quit)
02:35:42 Nick fxb is now known as fxb__ (
02:37:45d1sturbI reintsalled and it's still a no-go. Any suggestions?
02:39:38pixelmadid you also download the build again?
02:39:38 Join toffe82 [0] (
02:43:04 Join midgey [0] (
02:43:08d1sturbSo there's no other way I can get it to work?
02:47:27pixelmahmm... not many ideas as this is not a common error. Maybe you could try yesterday's "daily build" for once. And just making sure: are we talking about a Nano 1st generation?
02:50:10 Quit Nico_P (Remote closed the connection)
02:50:32 Join goffa [0] (n=goffa@
02:52:23 Quit goffa_ (Read error: 110 (Connection timed out))
03:00:12d1sturbIs it possible because my iPod is full that it isn't working/
03:05:05 Quit herrwaldo (Remote closed the connection)
03:05:59pixelmayou could try. Make sure that the build unzips completely.
03:06:30pixelmahope you get it working, I need to leave now
03:06:38 Part pixelma
03:06:41 Quit ali_as ()
03:06:47 Quit mindheavy ("ChatZilla [Firefox 3.0/2008060309]")
03:06:51d1sturbAny other staff here?
03:07:24 Quit thegeek (Read error: 104 (Connection reset by peer))
03:09:34***Saving seen data "./dancer.seen"
03:11:43 Join thegeek [0] (
03:19:37 Join VoiceCX [0] (n=VoiceCX@
03:24:37 Join Rob2222 [0] (
03:30:19d1sturbAnyone know of any other way I can watch videos on a iPod 1st Gen Nano? This program isn't working properly.
03:32:48 Quit VoiceCX ()
03:33:52 Join homielowe [0] (n=eric_j_l@
03:33:53 Quit homielowe_ (Read error: 104 (Connection reset by peer))
03:34:35cool_walking_Which program? What's going wrong?
03:34:36 Quit homielowe (Read error: 104 (Connection reset by peer))
03:34:37 Join homielowe_ [0] (
03:34:42 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
03:36:34d1sturbI did everything to install rockbox on my 1st gen iPod Nano, however, when I do a hard restart after installing everything, I'm getting a "Bad Checksum" error. PLEASE HELP!
03:36:53 Join BHSPitLappy [0] (n=BHSPitLa@unaffiliated/bhspitmonkey)
03:36:58 Quit einhirn_ (Read error: 104 (Connection reset by peer))
03:38:21cool_walking_is that the exact error?
03:39:02 Quit Thundercloud (Remote closed the connection)
03:40:09d1sturb...checksum : 2c77DFB / ...loading rockbox.ipod / sum :2C77DEC / Error! / Can;t load rockbox.ipod / Bad checksum
03:40:38cool_walking_rockbox.ipod is corrupt. Try downloading/extracting again.
03:41:15d1sturbI have...3 times =[
03:41:29saratogawell that rules out the download
03:41:32d1sturbI redownloaded it 3 times and reinstalled it
03:41:51d1sturbI also tried yesterday's build...same error
03:41:51saratogathat still leaves your ipod and whatever you extracted with as suspect
03:42:03d1sturbWhat do you suggest?
03:42:12cool_walking_Try a different upzip program
03:42:20d1sturbI'm using winrar
03:42:58 Quit Rob2223 (Read error: 113 (No route to host))
03:43:04d1sturbI'll try winzip
03:43:08cool_walking_I use winrar too, but better try another one just in case. Are you sure you're downloading the correct build from the download page?
03:44:03cool_walking_oh not, wait, that wouldn't matter... bad checksum
03:44:07d1sturbUnder "Downloads", Current Build
03:44:53d1sturbfine, I"m trying winzip
03:45:18cool_walking_Is your iPod's disk full?
03:45:56cool_walking_What Operating System are you using?
03:46:08cool_walking_oh.. winrar.. dogh
03:46:43d1sturbJust about
03:46:46d1sturbWin XP
03:47:53cool_walking_Close iTunes and anything else using your iPod, open a command prompt, and type "chkdsk i: /r /x", where "i:" is your iPod.
03:48:45cool_walking_Wait, is "just about" in answer to whether the disk is full?
03:49:09d1sturbyup, my ipod is just about full
03:49:13cool_walking_Try clearing some space and extracting again.
03:49:21d1sturbAnd I tried winzip didnt work.
03:49:23d1sturbk, one sec
03:51:17d1sturb81.7MB Free, good?
03:52:02cool_walking_Just want to make sure your zip program isn't being stupid and writing half of rockbox.ipod to the disk, then finding out there's no space, and leaving a half-file there.
03:53:14d1sturbno, same error
03:53:22d1sturbIs it the iPod?
03:53:26 Quit homielowe_ (Read error: 104 (Connection reset by peer))
03:53:27 Join JdGordon|uni [0] (i=768ae23d@gateway/web/ajax/
03:53:28cool_walking_Okay try the "chkdsk" thing I wrote above
03:53:32cool_walking_I don't think it's the iPod.
03:54:35 Join homielowe [0] (n=eric_j_l@
03:55:02d1sturbrun and then just type that thing above?
03:55:41cool_walking_no, you want to see what happens with the program. Run "cmd" to open a command prompt, then type the chkdsk command
03:56:34cool_walking_If you just type the "chkdsk .." directly into "Run", it'll close itself when it's finished and you won't be able to see what happened.
03:58:22d1sturb"The type of file system is RAW; CHKDSK is not available for RAW drives."
03:58:52cool_walking_That's not good, are you sure you have the right drive letter?
03:59:33d1sturbI did run/cmd/ and then typed chkdsk i: /r /x exactly
04:00:02cool_walking_the "i:" i wrote was just a placeholder. replace it with your iPod's drive letter.
04:00:07Shaidbut is I: the drive letter your iPod is using?
04:00:10d1sturbIt is I
04:00:36d1sturbSorry, I'm a n00b at this stuff =[
04:00:49cool_walking_Yet you can still browse the disk in explorer? strange.
04:00:58wpyhI found something interesting while browsing today:
04:01:10wpyhThere's an explanation of the ipod dock there
04:01:24wpyh(maybe everyone here knows it already, but who knows)
04:01:59wpyhand some more here, with the pin numbers reversed:
04:02:23wpyhit also appears that ipl knows about this (can't verify since their website is down)
04:02:47d1sturbcool walking, what about ipodlinux?
04:03:03cool_walking_I don't know anything about iPod Linux, I've never used it.
04:03:11d1sturbah, so nvm.
04:03:19scorchean ipod can take 22VDC? must be joking
04:03:25 Quit jhMikeS (Nick collision from services.)
04:03:31 Join jhMikeS [50] (n=jethead7@rockbox/developer/jhMikeS)
04:03:31cool_walking_What do you mean "what about" it? Do you have it installed as well?
04:03:34Lloreanscorche: probably missing a decimal.
04:03:36d1sturbI just found this program today while googling for a way to put videos on a nano
04:03:42Lloreanscorche: 2.2 is pretty common, no?
04:03:58scorcheLlorean: well, USB is 5
04:03:59cool_walking_d1sturb: Unless anyone else has any suggestions, I'd say backup everything from your iPod, and do a "restore" in iTunes, then do the "chkdsk", then install rockbox.
04:04:26Lloreanscorche: Oh, yeah, looking at that thread, probably not a typo
04:04:29wpyheven more here:
04:04:33scorcheLlorean: indeed
04:04:47d1sturbWas gonna ask ye. How can I backup the music on my Ipof
04:04:50wpyhso, maybe we can detect the presence of an accessory
04:04:57scorcheLlorean: were you thinking 3.3?
04:05:01scorche3.3 is the common one
04:05:13Llorean3.3 is common, but there's one below it. 2.1 I think.
04:05:29cool_walking_d1sturb: All the music is in a hidden folder called "iPod_Control"
04:05:40Lloreanwpyh: We need to know what's going on internally.
04:05:47LloreanPinouts aren't particularly helpfu
04:05:55cool_walking_d1sturb: do you know if your music has DRM?
04:06:05d1sturbOk, found it
04:06:07scorchewell, either way, i would love to see an ipod take 22VDC =P
04:06:14d1sturbNO, my music isn't protected
04:06:19cool_walking_d1sturb: It's just I don't use iTunes so I'm not sure if it'll let you import it back in if you just copy it off.
04:06:21BHSPitLappythose pins are pretty tiny
04:06:26BHSPitLappythat'd be a good chunk of current
04:06:27cool_walking_d1sturb: Okay, no issues then.
04:06:30wpyhLlorean: ok...
04:07:07d1sturbAlright, I'm trying what you said above
04:07:17scorcheit could be a firewire thing, i suppose
04:07:22d1sturbBy the way, thanks for helping me :)
04:07:49cool_walking_no problem
04:08:01 Quit maddler (Remote closed the connection)
04:08:29 Join maddler [0] (
04:08:53DhraakellianI'm pondering how to get some sort of moodbar support in Rockbox
04:09:34cool_walking_I had to google it too :)
04:09:48 Quit DerPapst (Read error: 113 (No route to host))
04:09:58scorcheDhraakellian: so something to read and display .mood files?
04:10:18Dhraakelliangiven that I don't care too much about album art, I think that I could probably hack up a .wps to use a long, thin bitmap
04:10:37scorchewell, it would likely be a plugin if anything
04:10:48Dhraakellianof course, this means that I have to figure out a way of converting existing .moods to bitmaps
04:13:42 Quit maddler (Remote closed the connection)
04:13:55 Join maddler [0] (
04:14:16 Quit conley (Read error: 104 (Connection reset by peer))
04:15:05 Join conley [0] (
04:15:22 Quit conley (Read error: 104 (Connection reset by peer))
04:15:27 Join conley [0] (
04:15:43Dhraakellianhmm... and probably using the "Frozen in the Arctic" or "Angry as Hell" colors just to be sure there's no magenta
04:17:41d1sturbdamn it, didnt work
04:17:45d1sturbWhatever, I give up
04:17:57cool_walking_did chkdsk still say it was raw?
04:18:53cool_walking_Did it find any errors?
04:19:09cool_walking_Are you still getting the same "bad checksum" error?
04:19:14d1sturbBut now I disconnected my iPod, and it says it's in disk mode
04:19:49d1sturbwait, nvm above message
04:19:53d1sturbits still getting the error
04:22:10 Quit maddler (Remote closed the connection)
04:23:41saratogaanyone tried out Buschel's GUI boost patch lately?
04:24:11cool_walking_d1sturb: I've got no idea.. try an older build.
04:24:57d1sturbI have =[
04:25:14saratogai doubt it has anything to do with builds
04:25:14d1sturbBy the way, are you one of the people that created the program?
04:25:25saratogawe would have noticed if builds stopped working at some point
04:25:34saratogaits obviously some problem with either his ipod or his computer
04:25:53d1sturbSo I don't have to report this as a bug...speaking to you is like I already did
04:25:55 Join Lynx [0] (
04:26:04d1sturbI can try it on a different computer
04:26:08d1sturbbe back in a minute
04:26:20cool_walking_d1sturb: nope. I'm not.
04:27:34 Join maddler [0] (
04:27:55 Join miepchen^schlaf_ [0] (
04:30:00 Join m0f0x [0] (
04:30:49 Join FOAD_ [0] (
04:32:24 Quit FOAD (Read error: 104 (Connection reset by peer))
04:32:24 Nick FOAD_ is now known as FOAD (
04:34:46 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
04:35:23 Quit nedd1 ("Leaving.")
04:36:00d1sturbIt worked on the other computer! =]
04:36:20d1sturbTHANK YOU!
04:37:25 Quit maddler (Remote closed the connection)
04:37:41 Join maddler [0] (
04:42:43 Quit Lynx_ (Read error: 110 (Connection timed out))
04:42:43 Nick Lynx is now known as Lynx_ (
04:43:52 Quit maddler (Remote closed the connection)
04:44:16 Join maddler [0] (
04:44:20 Quit JdGordon|uni (" ajax IRC Client")
04:46:16saratogafinally got the SDL sim playing audio
04:46:18wpyhd1sturb: do other devices behave the same way on your computer?
04:46:37wpyhsaratoga: I always get some kind of codec error
04:46:54saratogayou're running my patch?
04:48:22wpyhwhich patch?
04:48:47saratogathe one that creates adds an SDL target to rockbox
04:49:10d1sturbUsually not
04:49:10wpyhuh... tracker number?
04:50:49wpyhd1sturb: you may want to check for data corruption on other devices, like usb sticks
04:51:21wpyhI'd say that in a different way: you may want to check for data corruption on other devices, like such as.......
04:51:40 Join JdGordon|uni [0] (i=768ae21e@gateway/web/ajax/
04:52:02wpyhsaratoga: I didn't find anything resembling audio codec searching "sdl"
04:52:44saratogasdl is an api on unix, it doesn't have anything to do with codecs
04:52:54saratogawell api on various operating systems, including unix
04:56:23wpyhI'm running the sim right now, and when I try to load a song it will say "Codec failure"
04:56:47wpyhand creates a _temp_codec0.dll (I'm running linux btw, so I don't know why it would be dll)
04:58:23 Join DaCapn [0] (
05:00:01Lloreanwpyh: Did you compile the sim yourself?
05:00:55wpyhyes, I compiled it myself
05:01:12wpyhnot the latest svn though, maybe a week back
05:01:23LloreanSo you just did "make" then "make install" then ran it, and it didn't work? Nothing else in between?
05:01:28wpyhI thought it was an acceptable problem :p
05:01:45wpyhyes, I configure'd it, make
05:01:46wpyhoh wait
05:01:58wpyhmaybe I unzipped the into the archos subdir
05:02:07wpyhbut wouldn't that fail to run at all?
05:02:14LloreanNo, that'd just mess everything up.
05:02:36LloreanSince you'd be running a windows binary with player codecs. And it'd cause exactly the errors you mentioned.
05:02:36wpyhum... I can run the ui just fine
05:03:18LloreanThe UI is entirely in the executable file.
05:03:41saratogayou can run the UI without even an archos folder
05:04:22wpyhbut without the archos folder it would just say "incomplete rockbox installation"
05:04:44saratogayes I know
05:04:56Lloreanwpyh: His point was, the binary executable does not require the archos folder just to start.
05:05:07LloreanIn the archos folder should be codecs and plugins *compiled for windows*
05:05:09wpyhoh, ok
05:05:18LloreanYou get them there by way of "make install"
05:05:36LloreanYou do not get them there by extracting a build compiled for the player, as it's designed for an entirely different processor than your computer.
05:05:48wpyhright, I just realized that :)
05:06:26wpyhit plays audio fine now
05:06:38wpyhalthough with some clipping...
05:07:01 Quit linuxstb (Read error: 104 (Connection reset by peer))
05:07:06 Join nuonguy [0] (
05:07:09wpyhmaybe it's just my volume settings though
05:07:35saratogaLlorean: you had a windows ce phone right?
05:07:38 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
05:09:06Lloreansaratoga: Yes.
05:09:23saratogahave you looked at running any third party software on it?
05:09:33 Quit EspeonEefi ("さよなら")
05:09:34LloreanI run a bunch of third party apps.
05:09:35***Saving seen data "./dancer.seen"
05:09:36saratogai'm thinking about wince as a target, but i know nothing about it or it's phones
05:10:14LloreanHeck, there's a version of PuTTY for it. A little buggy compared to the desktop version, but usable.
05:11:24JdGordon|uniputty on it takes ages to connect... but still better to kick my mac than going up to my room where i can ssh in to it from my desktop :p
05:12:06JdGordon|unisaratoga: after my exam tomorow im going to have another stab at getting sdl to cross compile for it.... i dont tihnk it will be that hard anymore seen as it can be cross compiled for regular windows
05:12:10LloreanJdGordon|uni: It doesn't take too long for me. It did at first, but decided to get better.
05:12:40saratogaJdGordon|uni: i'm building cegcc now
05:12:41JdGordon|uniand it wont accept my keys so i have to use a password to login which is annoying
05:12:45saratogaso i'll try and take a look at it
05:12:50JdGordon|unisaratoga: cool... good luck
05:12:59saratogai assume its not too difficult?
05:13:21 Quit fdinel ("Miranda IM! Smaller, Faster, Easier.")
05:13:22JdGordon|unioh, i found a site which has a ubuntu image with cegcc and sdl for ce supposedly working but havnt grabbed the image yet...
05:13:31JdGordon|uniits being?
05:13:46saratogabuilding cegcc
05:13:50saratogabut i guess you haven't tried then
05:14:06saratogait looks easy enough anyway
05:14:20JdGordon|unii tried it a while ago and it wouldnt compile
05:14:56saratogafailing this i will look into that image
05:16:07JdGordon|uniif the firefox bookmarks are accessable over ssh I can get the link from here...
05:16:19saratogaif not just gtalk it to me whenever
05:16:35saratogai'm not going to try and figure out how to run a virtual machine in ubuntu tonight anyway
05:20:02 Join goffa_ [0] (n=goffa@
05:21:50 Quit Horscht ("We don't make mistakes, we just have happy little accidents")
05:23:17saratogaubuntu's package manager has a pocketpc-binutils but no pocketpc-gcc
05:23:22saratogai wonder if its included
05:23:45JdGordon|unithat vm is on a slower net connection than me ! 4 hours to download 227mb
05:25:45 Quit d1sturb ("I-n-v-i-s-i-o-n 2.0 Build 3515")
05:29:16 Quit goffa (Read error: 110 (Connection timed out))
05:34:11Lloreansaratoga: Did you check for cegcc?
05:34:36*Llorean sometimes thinks Ubuntu's packages don't have the most consistent names.
05:37:49saratogaLlorean: yeah its not there
05:38:08saratogathe ubuntu docs say theres a pocketpc-gcc but i can't get at it for whatever reason
05:38:12JdGordon|unijust download the pocketpc-gcc deb for a debian system
05:38:24saratogafrom where?
05:38:36 Quit saratoga ("CGI:IRC")
05:38:45LloreanMaybe there's not a package for 8.04 yet or something. =/
05:39:03 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
05:39:40saratogai'm not sure whats actually in that pocketpc-gcc package
05:40:18JdGordon|unisaratoga: have you had more luck removing the uisimulator/ dependancy from a regular build?
05:40:25saratogabut this svn checkout is ridiculous, its been running for almost 45 minutes
05:40:50JdGordon|uniit gets to over 800mb
05:40:54saratogaJdGordon|uni: no thats up next
05:41:28saratogai haven't looked to carefully, but it hopefully won't be too bad
05:42:17JdGordon|unimy cegcc/ is 1.6GB
05:42:26JdGordon|uniso halve that because of svn overhead
05:42:35saratogai wouldn't mind that at all if it'd tar before sending or something
05:42:47saratogabut doing it one file at a time is ridiculous
05:43:04JdGordon|uniyou know where the svn patch tracker is :p
05:43:56 Join goffa [0] (n=goffa@
05:48:55saratogaugh its downloading parts of gfortran
05:50:27 Quit JdGordon|uni (" ajax IRC Client")
05:55:13 Quit goffa_ (Read error: 110 (Connection timed out))
05:55:36 Join JdGordon|uni [0] (i=768ae23e@gateway/web/ajax/
05:56:40 Quit saratoga ("CGI:IRC (Ping timeout)")
06:02:15 Join goffa_ [0] (n=goffa@
06:04:32 Join EspeonEefi [0] (i=espeonee@STRATTON-THREE-THIRTY-SIX.MIT.EDU)
06:09:45 Quit goffa_ (Remote closed the connection)
06:09:46 Quit goffa (Read error: 110 (Connection timed out))
06:14:16 Join homielowe_ [0] (n=eric_j_l@
06:14:16 Quit homielowe (Read error: 104 (Connection reset by peer))
06:15:23 Join K4rP4D [0] (n=KrPD@unaffiliated/krpd)
06:26:45 Quit pabs (Read error: 104 (Connection reset by peer))
06:28:05 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
06:30:51 Join pabs [0] (
06:32:31 Join goffa [0] (n=goffa@
06:42:09 Quit linuxstb (Read error: 110 (Connection timed out))
06:46:11 Join goffa_ [0] (n=goffa@
06:56:01 Join midgey_ [0] (
06:56:01 Quit midgey (Read error: 104 (Connection reset by peer))
06:57:10 Join midgey [0] (
06:57:10 Quit midgey_ (Read error: 104 (Connection reset by peer))
07:01:22 Quit goffa (Read error: 110 (Connection timed out))
07:08:35 Quit JdGordon|uni (" ajax IRC Client")
07:09:37***Saving seen data "./dancer.seen"
07:13:34 Quit goffa_ (Read error: 110 (Connection timed out))
07:23:49 Join Bagderr [241] (n=daniel@rockbox/developer/bagder)
07:24:14 Nick Bagderr is now known as B4gder (n=daniel@rockbox/developer/bagder)
07:27:10 Join JdGordon|uni [0] (i=768adc91@gateway/web/ajax/
07:27:31 Quit JdGordon|uni (Client Quit)
07:30:52 Join goffa [0] (n=goffa@
07:33:27 Quit miepchen^schlaf_ ()
07:34:10 Quit advcomp2019 (Read error: 104 (Connection reset by peer))
07:34:23 Join advcomp2019 [0] (n=advcomp2@unaffiliated/advcomp2019)
07:39:57 Quit midgey ()
07:43:17 Quit m0f0x ()
07:45:35 Join linuxstb__ [0] (n=linuxstb@rockbox/developer/linuxstb)
07:51:24 Join midgey [0] (
07:59:39 Quit linuxstb_ (Read error: 110 (Connection timed out))
08:02:50 Join DerDome [0] (
08:10:44 Quit DerDome ("Leaving.")
08:10:54 Join rvvs89_ [0] (
08:19:37 Quit homielowe_ (Read error: 104 (Connection reset by peer))
08:20:09 Join homielowe [0] (n=eric_j_l@
08:20:58 Join homielowe_ [0] (
08:20:58 Quit homielowe (Read error: 104 (Connection reset by peer))
08:21:47 Join Buschel [0] (
08:22:32Buschelsaratoga: (i've just read the logs) any problems with the GUI boost patch? or does it work for you?
08:24:12*Buschel would like to submit FS #9076 the next days
08:24:27 Quit rvvs89 (Read error: 110 (Connection timed out))
08:25:12 Quit Buschel (Read error: 104 (Connection reset by peer))
08:25:53 Join Buschel [0] (
08:26:57Buschelbut I would like to have some opinions about removing the init of 2 pcf-registers for all ipod's (except video + nano)
08:28:10 Quit BigBambi (Read error: 113 (No route to host))
08:28:16Buschelimho the pcf's defaults should be fine for those players. furthermore i think that setting voltages, which we did not figure what they are used for, is no good idea
08:29:11Buschelwith FS #9076 the battery runtime of iPod nano was increased by ~45min in battery benchs of a user
08:31:27 Quit Buschel ()
08:34:27 Join GodEater [0] (i=c2cbc962@rockbox/staff/GodEater)
08:36:10wpyhBuschel: does the OF initialize or otherwise touch the PCF registers?
08:36:45 Join Rob2223 [0] (
08:37:42 Quit midgey ()
08:40:25 Join nplus [0] (
08:42:01 Part toffe82
08:45:51 Quit BHSPitLappy (Read error: 110 (Connection timed out))
08:47:22 Quit Reinhart (
08:47:22 Quit amiconn (
08:47:22 Quit jodedor (
08:47:22 Quit k4y (
08:47:22 Quit JdGordon (
08:47:22 Quit Bagder (
08:47:22 Quit Bjoern-Erik (
08:47:22 Quit blithe (
08:47:22 Quit courtc (
08:49:08NJoinReinhart [0] (
08:49:08NJoinamiconn [50] (n=jens@rockbox/developer/amiconn)
08:49:08NJoinjodedor [0] (
08:49:08NJoink4y [0] (
08:49:08 Join JdGordon [0] (i=jonno@rockbox/developer/JdGordon)
08:49:08 Join Bagder [241] (n=daniel@rockbox/developer/bagder)
08:49:08NJoinBjoern-Erik [0] (
08:49:08NJoincourtc [0] (n=court@unaffiliated/courtc)
08:49:08NJoinblithe [0] (
08:50:59 Join davina [0] (
08:54:18 Quit Rob2222 (Read error: 110 (Connection timed out))
08:56:33 Join ender` [0] (
09:03:50 Nick linuxstb__ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
09:06:28 Join PaulJam [0] (
09:09:38***Saving seen data "./dancer.seen"
09:23:22 Join Mathiasdm [0] (
09:26:50 Quit Rob2223 (Read error: 104 (Connection reset by peer))
09:27:12 Join Rob2222 [0] (
09:29:48 Quit Shaid (Read error: 110 (Connection timed out))
09:30:12 Join LinusN [0] (n=linus@gateway/web/cgi-irc/
09:34:09linuxstbcourtc: Are you around?
09:35:45linuxstbcourtc: (probably for the logs...) I was wondering if you recognised this issue with the iPod Photo LCD -
09:47:25 Quit midkay ("Leaving")
09:52:04 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
09:53:31 Join midkay [0] (n=midkay@rockbox/developer/midkay)
09:59:53 Join PaulJam_ [0] (
10:04:31 Quit linuxstb (Read error: 110 (Connection timed out))
10:04:46 Join pixelma [50] (i=pixelma@rockbox/staff/pixelma)
10:06:28 Quit PaulJam (Read error: 104 (Connection reset by peer))
10:17:22 Join faemir [0] (
10:27:24 Quit nuonguy ("This computer has gone to sleep")
10:31:15JdGordonwhats %t ?
10:32:31PaulJam_JdGordon: the time an alternating subline is shown.
10:32:32*JdGordon is confused by the 20gb h10 cabbie wps
10:35:16pixelmathat would be an X5 wps too (or just 160x128x16) ;)
10:35:47JdGordonthe next track lines are only shown if there is no AA? and its set as an alternating line but with nothing else?
10:35:57JdGordon%?C<%m|77||%ac>%t4Next Track:
10:36:15PaulJam_i don't think the %t4 makes sense there
10:38:12gibbon_ - thats what i commited to the flyspray patches page... its how far i got with my knowledge of rockbox
10:39:06JdGordonyou need to follow the coding guidlines...
10:39:18JdGordonimmeditaly obvious is the // style comments
10:39:39JdGordonand tab width
10:39:55pixelmaJdGordon: that looks like something's missing in the wps and actually that wps shouldn't load. The 160x128x2 shows what it should look like
10:39:57gibbon_will fixthis
10:42:37pixelmaJdGordon: and I agree that the %t4 doesn't make sense there
10:44:50pixelmaah no I was wrong... the colour variant is correct and makes the line shorter without duplicating.
10:46:25pixelmaneeded some time to understand that myself but the album art conditional just says "use left margin when aa is present (left align is default so doesn't have to be specified) - or use the whole line center aligned if not"
10:47:01pixelmastill the %t4 could be dropped IMO
10:48:00pixelmamaybe it's used to force a refresh of the line which seems to be necessary sometimes
10:56:04 Quit homielowe_ (Read error: 104 (Connection reset by peer))
10:57:46 Join homielowe [0] (n=eric_j_l@
10:57:53JdGordonpixelma: ah ok, thanks
11:01:53JdGordonpixelma: this good for the next track line on those displays? %s%?It<%?Ia<a - |>t|%Fn>
11:03:31PaulJam_JdGordon: could it be that your IDC client replaces "%I" with a spac3e?
11:04:34pixelmaor just drops everything that starts with %I
11:04:51JdGordonarg.. yeah looks like it tried italicizing the a
11:05:01pixelmahmm... no
11:05:02 Quit linuxstb_ (Read error: 110 (Connection timed out))
11:05:09JdGordon%s%?It<%?Ia<a - |>t|%Fn>
11:06:02JdGordon%s%?It<%?Ia<a - |>t|%Fn>
11:06:26PaulJam_i guess: %s%?It<%?Ia<%Ia - |>%It|%Fn>
11:06:26 Quit cool_walking_ (Remote closed the connection)
11:06:49JdGordoninstead of just the track name or filename
11:07:45pixelmaunderstood, guess that's more of a design decision though which I don't want to do because cabbiev2 is not my design...
11:08:32JdGordonok, ill put it in and see if anyone notices :p
11:09:13JdGordonbtw, shows why being able to use - in the viewport definitions is great
11:09:40pixelmathe line is more likely to scroll of course and some people don't like that. I like it displaying the filename instead of the title track if that's not present, I use it myself too :)
11:09:43***Saving seen data "./dancer.seen"
11:10:19JdGordonarg.. yeah, I should probably not fiddle with it... ok changing it back
11:11:06pixelmas/track/tag btw.
11:11:42PaulJam_JdGordon: what dimensions will the viewport have if you leave away the width and height?
11:11:53JdGordonthe rest of the screen
11:12:01pixelmait's a little cryptic...
11:12:26JdGordonno way... /me likey :p
11:13:24 Part LinusN
11:14:21JdGordonit also means that you dont need to change 2 numbers if you want to move the viewport
11:14:47PaulJam_JdGordon: i thought overlapping viewports that are shown at the same time are evil.
11:15:09pixelmathe - is hard to read and understand. I have no idea how the screen is used in that example
11:15:31JdGordonPaulJam_: not if nothing is displayed in the extra area
11:17:17pixelmaeven hard to understand in the colours part. Just an idea: what if one could specify "background" or "foreground" colour there?
11:19:41JdGordonyou really want to type the whole word each time?
11:19:57*JdGordon doesnt think - is really so hard to understand
11:20:08JdGordonof course, you have to remember the param order first though
11:21:12pixelmano I don't want to type the whole word all the time, only didn't come up with a shortened form right now.
11:22:10PaulJam_JdGordon: another thing, if i understood it correctly, then the viewports shown on condition a should cover the same space that the viewports for condition b cover. so shouldn't you make an extra viewport for the albumart that covers the remaining space of the non-AA viewport?
11:23:14JdGordonusually... yes.. but AA isnt going to disappear during a track, and its fully redrawn at atrack change so its safe
11:23:18pixelmaJdGordon: it is hard to understand especially if you read other people's wps (or try to understand your own wps after you've left it for a few months). I'll probably never use - for width and height
11:27:05JdGordonPaulJam_: ah, no.. the covers that space...
11:28:12PaulJam_i have the impression there is something missing in that sentence.
11:29:35JdGordon<percent>C got taken out
11:29:43*JdGordon curses konversation
11:30:11JdGordon%C test
11:30:13JdGordonyay :)
11:30:18JdGordonfound the bloody option
11:36:57 Join knittl [0] (n=knittl@
11:41:19pixelmaan hooray for those programs that are trying to be smarter than the user by default. kate, konversation ;)
11:52:27preglowwhy would it do that
11:55:25JdGordoncoz its bloody stupid
11:56:05JdGordonits for the script kiddies who somehow unbelivably managed to install linux and miss mirc
11:57:48 Quit goffa (Read error: 110 (Connection timed out))
11:59:56 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
12:01:17 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
12:02:17 Join dabujo [0] (
12:11:17 Join csc` [0] (n=csc@archlinux/user/csc)
12:16:44 Join ali_as [0] (
12:29:50 Join DerDome [0] (
12:29:51 Join goffa [0] (n=goffa@
12:32:21 Join MethoS- [0] (
12:38:36 Join MethoS_ [0] (
12:41:23 Join ikari [0] (n=5a850baf@gateway/web/cgi-irc/
12:42:35ikarihi. is it possible to switch from ipodlinux-bootloader2 to rockbox bootloader?
12:43:05JdGordonjust install our bootloader with either rbutil or ipodpatcher
12:43:23ikarii got ipodpatcher from ipodlinux
12:43:29ikariand i used it
12:43:57ikarido i need a different version of it?
12:44:20JdGordongrab the bootloader for your ipod from and use it with ippodpatcher
12:44:36*JdGordon doesnt remember which switch to use though
12:44:39ikarithank you very much :)
12:45:45*PaulJam_ reccommends the manual for installation instructions.
12:46:10 Nick PaulJam_ is now known as PaulJam (
12:46:21 Quit DerDome ("Leaving.")
12:48:22 Join Shaid [0] (
12:49:09 Join DerDome [0] (
12:54:43 Quit ikari ("CGI:IRC (EOF)")
12:56:50 Quit MethoS- (Read error: 113 (No route to host))
13:01:34GodEaterour copy of ipodpatcher has all our bootloaders built into it, you don't need to download the bootloaders seperately
13:02:04 Quit Shaid ("*toink*")
13:04:20 Quit DerDome ("Leaving.")
13:04:59 Join mf0102 [0] (n=michi@
13:07:23 Join DerDome [0] (
13:09:44***Saving seen data "./dancer.seen"
13:12:09 Nick Seedy is now known as Seed (
13:14:03 Quit linuxstb (Read error: 110 (Connection timed out))
13:16:55 Quit Seed ("cu, Andre")
13:17:34 Join Seed [0] (
13:29:21 Quit DerDome ("Leaving.")
13:31:20 Join homielowe_ [0] (n=eric_j_l@
13:31:20 Quit homielowe (Read error: 104 (Connection reset by peer))
13:31:39 Quit homielowe_ (Client Quit)
13:32:10 Join DerDome [0] (
13:37:00 Join desowin [0] (n=desowin@atheme/member/desowin)
13:45:57 Join Thundercloud [0] (
13:46:34dionoeagibbon_: you bipped me about audio in the sim ... but I've never worked on that so I don't know how I could have been of any help :)
14:03:46 Quit Thundercloud (Remote closed the connection)
14:06:02 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
14:11:38 Quit DerDome ("Leaving.")
14:15:32 Join _kch_ [0] (
14:15:36 Join DerDome [0] (
14:16:55gibbon_dionoea: you were anyway, there were some things about the callback you told me iirc
14:18:28 Quit [CBR]Unspoken|w ("!")
14:22:18 Part _kch_ ("Abandonando")
14:24:07dionoeagibbon_: hum ... ok . If you way so :) (I doubt it though)
14:26:14 Quit tvelocity (Read error: 110 (Connection timed out))
14:27:20gibbon_something else then?
14:27:26gibbon_nevermind, then
14:30:36pixelmawhy are the attachments on the wike WebHome reappearing once more?
14:35:44 Quit EspeonEefi ("さよなら")
14:35:53B4gderit's a mystery
14:37:17 Join Zom [0] (
14:39:42 Join Thundercloud [0] (
14:40:20 Quit goffa (Remote closed the connection)
14:41:29gibbon_dionoea: you pointed out DEBUGF ... so i may have confused you with someone else some hours later ;)
14:43:10 Nick daurnimator is now known as zz_daurnimator (n=daurn@unaffiliated/daurnimator)
14:43:13 Nick zz_daurnimator is now known as daurnimator (n=daurn@unaffiliated/daurnimator)
14:45:25 Quit nplus (Read error: 110 (Connection timed out))
14:46:16 Part B4gder
14:46:59 Quit Zom_ (Read error: 110 (Connection timed out))
14:49:04 Quit Zom (Read error: 104 (Connection reset by peer))
14:49:07 Join Zom_ [0] (
14:50:20 Join [CBR]Unspoken|w [0] (n=cbr@
14:51:36 Nick daurnimator is now known as zz_daurnimator (n=daurn@unaffiliated/daurnimator)
14:51:38 Nick zz_daurnimator is now known as daurnimator (n=daurn@unaffiliated/daurnimator)
14:57:38 Quit DerDome ("Leaving.")
15:01:50 Join nplus [0] (
15:02:06 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
15:09:36gibbon_i just (think i) corrected the code formating according to
15:09:43 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
15:09:45***Saving seen data "./dancer.seen"
15:11:39gibbon_is there anything else i should look for?
15:14:20 Join LambdaCalculus37 [0] (i=44a0430f@gateway/web/ajax/
15:15:16 Join PaulJam_ [0] (
15:17:28 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
15:26:51LambdaCalculus37gibbon_: Want to pastebin some of your code for review?
15:29:36gibbon_LambdaCalculus37: i wanted to upload it to the flyspray id, then i am finished with some minor tweaks in a minute
15:31:36LambdaCalculus37gibbon_: Okay, no problem.
15:32:12 Join DerPapst [0] (
15:33:06 Quit PaulJam (Read error: 110 (Connection timed out))
15:33:49 Join Siku [0] (
15:34:29gibbon_LambdaCalculus37: attached to first comment
15:34:53 Join MethoS- [0] (n=clemens@
15:34:57gibbon_when splitting codelines i was unsure if there was some policy, too... i just did it in a way that made sense to me
15:35:36LambdaCalculus37gibbon_: Wow! Nice idea for a plugin! :)
15:35:52gibbon_it even already works on my sansa
15:36:06gibbon_working on it for two days, having only minor experience with c...
15:36:10LambdaCalculus37I'll be glad to test it for you, but I forgot my laptop today so I can't compile a build right now.
15:36:12gibbon_i was surprised...
15:37:30gibbon_the features are still... well ... ehm ... limited
15:38:52gibbon_i want to implement "morsing from file" and some mode where you get a char and must decide which one of two (that are displayed) it is...
15:39:12 Nick rvvs89_ is now known as rvvs89 (
15:39:26LambdaCalculus37gibbon_: Heh... would be useful for HAM radio. :)
15:39:40gibbon_training at least... thats what i wrote it for
15:40:03gibbon_and i have to implement some tuning of the frequency..
15:40:10gibbon_currently thats not possible
15:40:21*amiconn wonders why that plugin reinvents several wheels, so to say
15:40:51gibbon_amiconn: point at them and they are gone
15:40:56amiconnlen_zeroterm() -> rb->strlen()
15:41:14amiconnrcw_clean_buffer() -> rb->memset()
15:41:56gibbon_will fix that, too
15:43:50amiconnThis plugin should also be possible on hwcodec
15:43:59amiconnThat's a separate step, of course
15:44:35gibbon_amiconn: yes, i read metronome.c ... it uses some kind of static mp3 snippet from ram, right?
15:46:35*JdGordon stumbles and trips over everyone...
15:46:44*JdGordon punches gevaerts for pushing me
15:47:03*gevaerts falls over
15:47:09wpyhwell, ok so we should continue here
15:47:13gibbon_amiconn: the strlen returns size_t ... how far is it from an integer?
15:47:59wpyhNico_P: the database can run in the background, just like battery_ bench
15:48:21Nico_PI don't see the point of doing that
15:48:33gevaertsHow does the database get its memory ? Does it use bufalloc() ?
15:48:48dionoeagibbon_: size_t is unsigned integer
15:48:52wpyhthat would be a good idea, using bufalloc()
15:48:52dionoeassize_t is signed
15:49:16gibbon_dionoea: thanks
15:49:30wpyhNico_P: I think it's not the objective, it's just the way to make the db a plugin
15:49:40Nico_Pgevaerts: no, I think it uses buffer_alloc, which is not the same
15:49:56Nico_Pbufalloc is not meant for that sort of thing, it's higher level
15:50:40Nico_Pthe buffering API (not bufalloc, but its cousins) is used to buffer audio files during playback
15:50:42*wpyh thinks all those buf.*alloc things will go away if we have malloc, but that's out of the question
15:50:50JdGordonNico_P: by the way... how safe is bufalloc? does it keep moving around the buffer so as long as you keep getting the address before accessing it your safe "for ever"?
15:51:22*LambdaCalculus37 whips out a cricket bat with wpyh's name on it
15:51:31wpyhif we can make the db a plugin then run it in the background, it won't create JdGordon's issue #1 : "1) cant use another plugin while in the db"
15:51:35 Quit MethoS_ (Read error: 113 (No route to host))
15:51:47*wpyh wonders what he did wrong...
15:51:54Nico_PJdGordon: using bufgetdata to get the pointer protects you from the moving, yeah. the data won't disappear until you call bufclose on it
15:51:57JdGordononly 1 plugin buffer... so you can only run 1 plugin at a time
15:52:09LambdaCalculus37malloc bad... bufalloc good...
15:52:12gevaertsCorrect me if I'm wrong, but could (assuming that playback is stopped or paused) buffer_alloc() force a flush and re-buffer, thereby avoiding having to reboot ?
15:52:16JdGordonNico_P: so in effect.. you have snuck malloc in?
15:52:36wpyhJdGordon: um... maybe we can make multiple plugin buffers then?
15:52:39Nico_PJdGordon: sort of :) it's even used in picture flow
15:52:49JdGordongevaerts: yes, and its been agreed that thats not a big deal in certain situations... its just not done
15:52:57wpyhLambdaCalculus37: yeah, so malloc is out of the question...
15:53:06JdGordonwpyh: then you lose the benefit of moving it out of the core
15:53:31LambdaCalculus37wpyh: Remember, GodEater has torches and pitchforks. :)
15:53:34Nico_PJdGordon: it's basically malloc with the audio buffer as a pool
15:53:51*GodEater denies owning torches or pitchforks
15:53:53gevaertsJdGordon: ok, so I assume that _if_ we want to get rid of the reboot, that would be the way to go
15:54:04wpyhwell, the main benefit of moving it out of the core I think is that it can reload itself without having to reboot
15:54:07JdGordondatabse and dircache
15:54:27amiconngibbon_: Yes. In case of the morse beeps, it could use a single mp3 frame and repeat it as needed
15:54:54JdGordongevaerts: except.. if you enable it, then disable, you cant get that ram back.... reenabling would get wasteful very quickly
15:55:31JdGordonIIRC there was some talk on adding a mechnism to make all buffer_alloc() users dump and reload their data so this wouldnt be a problem
15:55:40gevaertsJdGordon: you could disable while keeping track of the memory allocation, so you need to reboot if you want to disable _and_ reclaim the memory
15:55:44amiconnAdding the mas pcm codec to the core (another open task) will open another possibility: just using pcm like on swcodec. This would exclude the Player though.
15:56:09 Join MethoS_ [0] (
15:56:31JdGordongevaerts: yes, but you may need more ram when you renable it
15:56:47JdGordonbassically.. its not worth the effort
15:57:22*wpyh wonders (quietly) if softmmu is feasible
15:57:24gevaertsWhat does ram use depend on, and how does updating the database work around that ?
15:57:56JdGordonfor ramcache? (db in ram?)
15:58:18gevaertsFor everything that uses buffer_alloc
15:59:38JdGordonfirst example from find is filetypes.c uses it as a string storage instead of statically allocating it which would be either wasteful or possibly not enough
15:59:46amiconnwpyh: Certainly not. The main problem with malloc() is that it needs free ram (the malloc buffer) to allocate from
16:00:04amiconnIn rockbox, we want to use as much ram as possible for the main buffer
16:00:10JdGordonthe playlist and filetree buffers are both buffer_Alloced()
16:01:50wpyhyeah... that's one thing
16:02:01*wpyh goes to the bathroom to think over this problem
16:02:03*gevaerts will remember the buf_alloc() thing for if/when he does usb audio
16:02:33 Quit Horscht ("")
16:03:14 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
16:03:39JdGordongevaerts: depending how much you need... the plugin buffer would be better to steal
16:04:05gibbon_amiconn: did anything else come to mind that should be fixed about that code?
16:04:42gevaertsJdGordon: that's the other option, but I'd like to leave the thing as functional as possible
16:04:46 Quit Horscht (Client Quit)
16:06:50Nico_Pgevaerts: buf_alloc() is a mix between bufalloc() and buffer_alloc() :)
16:07:05JdGordonusing buf_alloc isnt a great idea... you cant free it.... what might be better is allowing the end of the audio buffer to be brought forward (Would need a rebuffer but meh), only allowing 1 thing to move the end would make it relativl safe and recoverable
16:07:14JdGordonand the buffer wouldnt move like buf_Alloc() would
16:08:53gevaertsThat would work
16:09:46gevaertsIt could also be useful for MTP. You could probably do the writes pretty directly from a small static buffer, but performance really increases if the buffer grows
16:10:25JdGordonMTP would allow you to keep listening while its connected yeah?
16:10:46gevaertsIt should. It works at file (and database, if we want to) level
16:11:51*gevaerts should really read a bit more about MTP, so he can get rid of the 'should's and 'may's
16:13:10JdGordonnot to be rude... but... MSC mode? still no clue how to fix the problems?
16:13:57 Quit MethoS- (Read error: 110 (Connection timed out))
16:14:26gevaertsNo, unfortunately
16:18:50 Join saratoga [0] (n=9803c50e@gateway/web/cgi-irc/
16:18:50Nico_Pgevaerts: have you been able to try the validation you mentioned ?
16:19:12gevaertsThe ramdisk ? Yes, no change (but great speed)
16:19:37Nico_Poh so it's not at ATA level?
16:20:23Nico_Pbut actually I was thinking of something I recall you mentionong... a kind of validation test against the spec, that you could get done through someone you know
16:22:26gevaertsThe electrical tests ? I can't figure out how to put the controller in test mode
16:23:44Nico_Pthat's probably what I recalled
16:23:52 Join Schmogel [0] (
16:29:11saratogai'm trying to build cegcc, but i get a syntax error in the build script
16:29:21saratogaubuntu should be able to run a .sh shell script right?
16:29:40Nico_Pyes, it should
16:29:46LambdaCalculus37Of course. :)
16:29:46JdGordonbut that doesnt get it much further
16:30:01*JdGordon rereads
16:30:05JdGordonrun it with sh
16:30:14JdGordonnot in bas
16:30:36Nico_PI think the default is dash
16:32:59saratogadoesn't seem to matter, i still get the same syntax error
16:34:32 Nick JdGordon is now known as JdGordon|zzz (i=jonno@rockbox/developer/JdGordon)
16:36:16wpyhDerPapst: did you have a problem with FAT32 formatting using mkfs.vfat?
16:36:45DerPapstbut only where the sectorsize is larger than 512bytes
16:36:52DerPapsti.e. 5.5G iPods
16:37:09wpyhI'm going to try it now
16:37:16wpyhwhat problems did you have?
16:37:26wpyhI'm using a cf card on my 5.5g with mkfs.vfat without problems
16:37:56DerPapstrockbox refused to read anything from the hdd, iPL worked fine and the OF refused to write files
16:38:23wpyhin my case,rockbox loads fine
16:38:28wpyhdidn't try ipl
16:38:38wpyhwhat do you mean by "the OF refused to write files"?
16:39:49DerPapstit was possible to read files but several functions didn't work "normally". iirc updating OTG-Playlists, Playcount, etc
16:40:23wpyhwas linux able to mount the device normally?
16:41:02DerPapstyup. read/write worked fine
16:41:32wpyhok... weird
16:41:39wpyhdid you use the "-S 2048" switch?
16:42:55wpyhwell.. ok I'll try it with my 80GB drive
16:43:06wpyhI suspect it's not the 2048 sector problem, but the 1024 sector problem
16:44:00DerPapstheh.. my fat partition still thinks it's 1GB larger than it really is ;)
16:45:33wpyhhow come?
16:45:41wpyhisn't it already 70GB+?
16:47:51wpyhhow do you know it thinks it's 1gb larger ?
16:48:15saratogacould someone help me out with using alien to install a package in ubuntu?
16:49:17*pixelma suggests asking that in the other channel
16:49:25DerPapstwpyh: i creaded a 1GB parititon at the end of the hdd for iPL and didn't reformat the fat32 partiiton. so the mbr disagreas with the fat superblock ;)
16:49:26LambdaCalculus37saratoga: Sure, hop in to #rockbox-community
16:50:00wpyhis that the source of your troubles then?
16:50:14saratogacan the web client do that?
16:50:28 Join bughunter2 [0] (n=Jelle@
16:50:30LambdaCalculus37saratoga: Yeah, I used to do it all the time.
16:50:48*GodEater still thinks regular webclient users would be better off with mibbit
16:51:46saratogaactually i think this package isn't going to work regardless due to 64 bit issues
16:52:22LambdaCalculus37saratoga: If you hop into the community channel, I can help you out.
16:52:45*LambdaCalculus37 is a sysadmin, after all.
16:53:29 Quit linuxstb (Read error: 110 (Connection timed out))
16:54:51 Join domonoky1 [0] (
16:55:37wpyhDerPapst: let's see...
16:55:51wpyhI've reformatted my fat32 partition using mkfs.vfat
16:58:40 Join petur [50] (n=petur@rockbox/developer/petur)
16:58:53wpyhDerPapst: I get even weirder results with mformat..
16:59:21DerPapstyou mean mtools?
16:59:50wpyhso, how did you format it? Windows?
17:00:15DerPapsti didn't that's my fat32 partition thinks it's 1gb larger ;)
17:01:54wpyhbut you said you were having problems with mkfs.vfat?
17:01:57wpyhhow did you solve it?
17:02:02wpyhI don't understand...
17:02:24DerPapsti restored it parititoned it and "forgot" to reformat the fat32 partition
17:03:09wpyhcan you please post the output of "file -Ls /dev/sdb2" (or whatever is that partition)?
17:04:03DerPapstno, i can't because i have no linux box here right now :P
17:04:17wpyhplease do it when you have time
17:04:30wpyhmine is here:
17:04:41wpyhcan you spot the differences?
17:04:52wpyhI mean between the original partition, mkfs.vfat and mformat?
17:05:02wpyhI think mformat gets the most fatal error
17:05:18wpyhbut strangely I can mount it on Linux
17:06:31DerPapsti guess mine matches your first.
17:06:58wpyhwell, the first one is the original ipod partition...
17:07:12wpyhnot mkfs.vfat or mformat...
17:07:27DerPapsti only touched the MBR and not the second partition. so this is what itunes creates.
17:08:00wpyhyes, it must be what itunes did
17:09:14 Quit domonoky (Read error: 104 (Connection reset by peer))
17:09:38wpyhso, I have OF and some music on my 5.5g, formatted with mkfs.vfat
17:09:46wpyhwhat should I do to reproduce your problem?
17:09:51***Saving seen data "./dancer.seen"
17:11:33DerPapstwell several features didn'T work and itunes complained too iirc. try OTG-Playlists, playcounts, alarm, hibernation... everything that writes to the disk
17:12:08wpyhok, the playlists work fine
17:12:13wpyhthe otg playlist also does work fine
17:12:24wpyhhibernation... how do I do that (I'm new to ipods)
17:12:52DerPapstwell you wait 18 hours without touching the iPod :P
17:13:28DerPapsthave you connected the iPod to itunes after creating an OTG playlists? so it converts it into a real one?
17:14:04wpyhno, what I did was save the otg playlist as a normal one, so it creates a "New Playlist 1"
17:14:22wpyhoh, _that_ hibernation... maybe I don't have that patience :p
17:14:32wpyhI don't have itunes btw
17:14:42DerPapstit's ok. ;)
17:15:17 Quit MethoS_ ("Konversation terminated!")
17:15:35wpyhso... judging from the created playlists, does it mean the OF can write to the disk fine?
17:16:04 Quit PaulJam_ (".")
17:16:16DerPapstseems so ;)
17:16:20 Quit DaCapn ()
17:16:41wpyhuh... then mkfs.vfat has no problems
17:16:47wpyhshall I update the wiki? ;)
17:17:03 Join nuonguy [0] (
17:17:55DerPapstyup.. if it works, why not :) maybe keep the old info though.
17:18:07wpyhyes, I'll keep the old info
17:18:09wpyhand uh
17:18:22wpyhhow about merging IpodManualRestore and ConversionToFat32?
17:20:04DerPapstwell, i'm not a big contributor to the rockbox wiki so i don't know. but i guess they're pretty similar....
17:20:16wpyhOK then
17:20:23wpyhmaybe we can ask the devs
17:21:30gevaertsI would think twice before merging them. One is for Mac users who are used to itunes, the other is for people who don't want to see itunes up close
17:22:44wpyhgevaerts: the "Manual Restore" part of ConversionToFAT32 is similar to IpodManualRestore
17:23:03wpyhso similar, that in fact I think they are some kind of knowledge duplication
17:23:07gevaertsThat's true.
17:23:26wpyhyeah, so let's merge it, no?
17:23:42scorchethey could be linked, i suppose, but i would rather keep them is much easier to point someone at a page of instructions to do rather than say "do this, but not this, etc"
17:24:02scorchefor support matters
17:24:04gevaertsMaybe restrict the ConversionToFAT32 to the 'use windows/itunes' bit, and refer to IpodManualRestore for the manual conversion ?
17:24:33wpyhgevaerts: good idea
17:24:55wpyhbut as I see it now, ConversionToFAT32 looks like it's written for OS X users
17:24:58wpyhlike you said :)
17:25:38gevaertsThen (if I'm not mistaken) the only difference is the need to restore the apple firmware, so that's only one conditional (and if someone does that when it's not needed, it's not that bad)
17:25:45wpyhBut I can easily imagine what I would write there
17:26:27wpyhthe steps would be 1. unmount the ipod, 2. write a new partition table, 3. write a new firmware, 4. format the new partition
17:26:44wpyhthen in each step we can branch for a) OSX, b) linux
17:27:05gevaertsAnd 3 is optional
17:27:27wpyhWe can keep the pages separate
17:27:50 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
17:27:50wpyhbut direct users who want the manual way (in IpodConversionToFAT32) to IpodManualRestore
17:27:53wpyhhow about that?
17:28:14 Nick Zom_ is now known as Zom (
17:28:46gevaertsThat's what I would do, yes
17:29:07wpyhok, I'll try to implement it
17:31:04*linuxstb thinks the page for Mac users needs to be left as a page for Mac users...
17:31:23LambdaCalculus37linuxstb: But should we label it as such?
17:31:50linuxstbLambdaCalculus37: Sure.
17:32:19linuxstbwpyh: Also, what are you planning to change about mkfs.vfat? Has it been shown to work with an 80GB disk?
17:32:32wpyhlinuxstb: yes, please see the logs
17:32:37wpyhit works fine with the OF
17:32:49LambdaCalculus37linuxstb: Which page is it?
17:33:11*wpyh thinks it's this one:
17:33:13linuxstbLambdaCalculus37: I'm talking about IpodConversionToFAT32
17:34:52*LambdaCalculus37 thinks that it should be "iPod Conversion to FAT32 under MacOS X"
17:36:03wpyhwell, what about us linux users?
17:36:22LambdaCalculus37wpyh: The commands are slightly different.
17:36:54wpyhLambdaCalculus37: yes, that's why I propose adding a branch at each step: a) for linux, do xxx , b) for osx, do yyy
17:37:12linuxstbwpyh: Why did you change "require linux" to "these instructions have been tested on linux" on the IpodManualRestore page? Where else will they work?
17:38:10wpyhlinuxstb: wouldn't it work on freebsd and such as (quoting miss teen south carolina :p)
17:38:40wpyhthe device names would be a little different, but you get the ida
17:38:49linuxstb"get the idea" isn't the same...
17:39:00wpyhyeah, I probably should make it clearer
17:39:35wpyhany *bsd users around here?
17:40:28LambdaCalculus37wpyh: And Mac OS X counts. :)
17:40:44wpyhyeah, that too
17:41:05wpyhI don't know much about OSX though, so I can't say much :p
17:41:31wpyhso what I'm proposing now is
17:42:09wpyhthe IpodManualRestore will contain instructions for Linux and OSX (taken from IpodConversionToFAT32), to manually restore / convert it into a winpod
17:42:32linuxstbwpyh: Converting to a winpod isn't the same as manually restoring. For conversion, you don't need to copy the firmware.
17:42:56wpyhthe IpodConversionToFAT32 contains the "easy way" , and the section "Manual Conversion" will direct users to the page IpodManualRestore
17:43:05wpyhlinuxstb: yes, and that part will be marked as optional
17:43:20wpyhand I'll explain that that step is not needed for manually restoring the ipod
17:43:30wpyhI mean for converting the ipod
17:43:34linuxstbwpyh: I just think the instructions are confusing enough for users, and the fewer options and choices users need to make, the better.
17:43:41linuxstbWhich is why there is a page specifically for Mac users.
17:44:03 Quit Seed (Remote closed the connection)
17:44:14wpyhlinuxstb: you may be right there
17:44:26wpyhbut what I see is just duplication of contents
17:44:33linuxstbe.g. you're already adding many options (dd or cat) to the IpodManualRestore page, which adds nothing...
17:44:45 Join Seed [0] (
17:45:01wpyhthat dd/cat part can go :p
17:45:12wpyhI personally like the "cat" but I realize it can't be done under sudo
17:45:24wpyhI hope we never have to write instructions for sudo :p
17:47:27gevaertswpyh: I fully agree about the duplication. I didn't see the mformat on IpodManualRestore, so I only fixed it on IpodConversionToFAT32
17:48:06wpyhit's on this line: For 5.5G iPods (with 2048-byte sectors): mformat -S 2048 -M 2048 -F DRIVE_LETTER:
17:48:44wpyhand by the way my mkfs.vfat formatted partition works with OF and RB
17:48:48gevaertsAh yes. That one doesn't specify the sectors per track, so it isn't confused
17:48:48wpyhdunno about ipl
17:49:15wpyhactually, I was a bit confused with mformat
17:49:25wpyhwhich is why I investigated mkfs.vfat, and found no problems with it
17:49:39wpyhand that is why I queried DerPapst about the problem
17:50:27linuxstbBut that still leaves us with a problem for Mac users - mkfs.vfat isn't there....
17:51:01wpyhlinuxstb: we can let osx users use mformat and linux users use mkfs.vfat
17:51:08wpyhmkfs.vfat is WAY simpler to use
17:51:23linuxstbwpyh: Sure. Another reason to keep the instructions separate though...
17:51:52gevaertsOr we can wait until the ipodpatcher author writes 2048 byte sector support ;)
17:52:54wpyhgevaerts: I thought ipodpatcher has got nothing to do with formatting the partition?
17:53:28wpyhlinuxstb: we can split the instructions early: A) for osx users, do 1. 2. 3. 4., B) for linux users, do 1. 2. 3. 4.
17:53:32DerPapstipodpatcher can format partitions
17:53:50wpyhthis will duplicate some information, but it will be minimal
17:53:56wpyhDerPapst: I didn't know that..
17:54:11*wpyh just noticed the "-f −−format" option
17:54:16DerPapstbut only 512bytes/sector
17:54:22wpyhwell, maybe we can take matters into our own hands :D
17:54:42wpyhand modify ipodpatcher behind the author's back :D
17:54:47wpyh(as long as we don't break it)
17:54:50*linuxstb turns away
17:55:09wpyhlinuxstb: huh?
17:55:22gevaertswpyh: guess who the author is
17:55:33wpyhDave Chapman
17:55:35wpyhwho's that?
17:55:41wpyhlinuxstb? :O
17:55:41LambdaCalculus37Yep. :)
17:55:44DerPapstnow guess who linuxstb is
17:56:19*wpyh wonders how to transform LINUXSTB into DCHAPMAN
17:56:21*pixelma points wpyh to the IrcNicks page ;)
17:56:32wpyhpixelms: let's see..
17:56:40linuxstbOr just query my nick in your IRC client...
17:57:08wpyhoh, it says "Real name: Dave Chapman"
17:57:08pixelmaor a /whois
17:57:25wpyhok, thanks :P
17:58:19linuxstbwpyh: But I would be very happy if somebody added 2048-byte formatting support for ipodpatcher....
17:58:29wpyhso... linuxstb aka Dave Chapman, is there a problem with 2048-byte support in ipodpatcher?
17:58:29linuxstbAnd then finished the "−−convert" option...
17:58:56 Join SUSaiyan` [0] (
17:58:58*wpyh goes to add himself to the IrcNicks page
17:59:03linuxstbwpyh: The formatting function only handles 512-byte sectors, so will refuse to format anything else.
17:59:14wpyhdoes "−−convert" mean convert to a winpod?
17:59:15linuxstb(the code was taken from fat32format.exe)
17:59:29wpyhlinuxstb: maybe we can copy code from mkfs.vfat
17:59:35wpyhI'll put another TODO item :p
17:59:39linuxstbwpyh: Yes. But that's not been tested very well, so I'm not 100% sure it works. Which is why it's not publicised.
18:00:45wpyhmaybe if we advertise it as "can convert any ipod into a winpod" then we'd get more testing?
18:02:09*linuxstb needs to go home
18:02:33wpyhis there any consensus on the wiki pages, btw?
18:08:29 Join Buschel [0] (
18:08:38 Quit SUSaiyan (Read error: 110 (Connection timed out))
18:13:31 Quit krazykit (Read error: 104 (Connection reset by peer))
18:16:43 Join krazykit [0] (
18:19:15 Join MethoS- [0] (
18:20:22 Quit linuxstb (Read error: 113 (No route to host))
18:20:43Buschelfar less than half present −− Germany is playing right now (UEFA)
18:21:11wpyhah, no wonder
18:23:17 Quit Nico_P (Remote closed the connection)
18:23:28 Join miepchen^schlaf [0] (
18:25:33 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
18:26:36wpyhwell, I guess I'll wait for a consensus / decision from the devs then
18:32:07 Join PRINCESS_FLUFF [0] (
18:33:11PRINCESS_FLUFFHi, I program a bit in Java, and I keep thinking I'd like to be able to code for stuff like rockbox or zsnes (a super nintendo emulator), and maybe other projects, or even my own projects... What stuff do I need to study to get into this? Just C programming? Or Assembler and some other stuff too?
18:33:45LambdaCalculus37For Rockbox, most everything is in C.
18:34:01saratogaBuschel: I don't get any real improvement using SSO in MAD
18:34:15PRINCESS_FLUFFWhen you want to support a new player, do you need to build in some interface code in Assembler or something like that?
18:34:33saratogai looked it up in the ARM7 manual, and the SMULAL is only 1 clock slower then MLA anyway
18:34:36LambdaCalculus37No, assembler is used for platform-specific performance bits.
18:34:36PRINCESS_FLUFFAnd is it just "the c programming language" or some libraries as well?
18:34:51PRINCESS_FLUFFI thought assembler was what everything used
18:34:55saratogaso the whole idea of the SSO is probably pointless on ARM7
18:35:05PRINCESS_FLUFFie. comptatible with everything
18:35:42saratogaassembler is the least compatable choice
18:35:55saratogait only runs on the CPU platform its written in
18:36:02LambdaCalculus37PRINCESS_FLUFF: Think of assembler as programming in the CPU's DNA.
18:36:20saratogaor better yet see wikipedia and lookup what it is
18:36:27PRINCESS_FLUFFWhen they talk about "chipsets", that's the set of assembly thingies that work with that motherboard or cpu right?
18:36:46LambdaCalculus37Chipsets are the chips used to design a computer.
18:36:58saratogaits usually just the chip connected most closely to the CPU, whatever that may be
18:38:40PRINCESS_FLUFFok thank you
18:38:58saratogaBuschel: unfortunately, that leaves me with no idea how to make MAD any faster
18:39:38saratogashort of coming up with some kind of algorithmic improvement to the decoder
18:39:46 Quit amiconn (Nick collision from services.)
18:39:53 Join amiconn [50] (n=jens@rockbox/developer/amiconn)
18:39:59LambdaCalculus37PRINCESS_FLUFF: If you really want to learn C, I suggest picking up the book "The C Programming Language" by Brian Kerrighan and Dennis Ritchie.
18:40:13LambdaCalculus37Also, download the Rockbox source, and look through it to see how C is structured.
18:42:54 Quit mf0102 (Remote closed the connection)
18:43:46 Join Enginerd3 [0] (n=407c962a@gateway/web/cgi-irc/
18:43:50 Quit pixelma (Read error: 110 (Connection timed out))
18:44:30 Part Seed
18:45:05 Join bertrik [0] (
18:45:53PRINCESS_FLUFFLambdaCalculus37: thank you, I asked someone in #c a while ago and ordered that same book since
18:46:21LambdaCalculus37PRINCESS_FLUFF: There's also plenty of free information online.
18:46:23 Join Seed [0] (
18:46:37LambdaCalculus37Trust me, once you get the hang of C, you'll find it's a rather easy language to work with.
18:47:13LambdaCalculus37But like I said, get the Rockbox source; it's the best information you can get.
18:47:24PRINCESS_FLUFFI already know java and did a bit of C++ a few years ago... is it accurate to say that C is at most as complicated as C++ minus some features?
18:47:24 Join PaulJam [0] (
18:48:33LambdaCalculus37C is less complicated, but enough about the programming languages, since it's off-topic.
18:48:37 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
18:50:04 Join SUSaiyan [0] (
18:50:16 Join MethoS_ [0] (
18:55:42 Part flynux_
18:55:58 Join herrwaldo [0] (
18:57:21 Quit Seed (Remote closed the connection)
18:59:19gibbon_for now, FS #9088 contains every suggestion i got so far...
18:59:58gibbon_i tested it to build on ipod mini and sansa c200 for now... if anyone interested could also test, that would be great.
19:00:40 Quit SUSaiyan` (Read error: 110 (Connection timed out))
19:02:10 Quit Siku ()
19:05:48 Quit MethoS- (Read error: 110 (Connection timed out))
19:06:26 Join PaulJam_ [0] (
19:06:33dionoeaOo they have unicode chars in morse?
19:06:53 Join BigBambi [0] (n=Alex@rockbox/staff/BigBambi)
19:09:52***Saving seen data "./dancer.seen"
19:11:04 Join ompaul [0] (n=ompaul@gnewsense/friend/ompaul)
19:12:32dionoeagibbon_: why don't you use 'a' in your code instead of 0x61? (same thing for all the chars)
19:13:04dionoeain your case conversion and validation functions
19:13:35gibbon_as a reminder that i deal with them as numbers
19:13:48 Join PaulJam__ [0] (
19:13:53gibbon_" 'a' - 0x20 " looks kind of odd to me
19:14:43 Quit PaulJam (Nick collision from services.)
19:14:49 Nick PaulJam__ is now known as PaulJam (
19:15:24dionoea- ('a' - 'A')
19:16:04*dionoea needs to try this plugin out ... so he can send an SOS next time he's near a radio
19:16:37wpyhrockbox has just kinda locked up
19:16:44wpyhit doesn't respond to button press
19:16:58wpyhdoesn't do anything when the cable is plugged in
19:16:59LambdaCalculus37MENU+SELECT until it reboots.
19:17:00wpyhyes, ipod 5.5g
19:17:06wpyhyep, that's what I did
19:17:12wpyhproblem is, I can't reproduce the situation
19:17:15wpyhany ideas?
19:17:22LambdaCalculus37What were you doing?
19:17:46wpyh(something completely unrelated) playing around with the skin
19:17:55wpyhI mean with the rubber skin
19:18:03wpyhso it should have nothing to do with the player
19:18:23wpyhby "playing around" I mean pulling the ipod out of the rubber skin and putting it in again (repeatedly)
19:18:59wpyhmaybe I pressed on something when doing that, but I can't reproduce it... :(
19:19:44wpyhhas anyone seen this error before?
19:20:01wpyhI'm thinking this is a software issue, because when I plugged in the usb cable it didn't do anything
19:20:15 Join tvelocity [0] (
19:21:38 Part PRINCESS_FLUFF
19:26:24gibbon_dionoea: would you please do that, yes? and don't mind the hundreds of people willing to help you at your backdoor :P
19:27:02 Join lolmaus [0] (i=lolmaus@
19:28:21 Join Seed [0] (
19:28:44gibbon_dionoea: that note is there, because SOS signals should by common sense never be automatically emitted... iirc its even in the radio regulations
19:30:56gibbon_its the same for all distress signals...
19:32:11 Quit PaulJam_ (Read error: 110 (Connection timed out))
19:32:18gibbon_bye... see you later
19:33:38 Quit MethoS_ (Read error: 104 (Connection reset by peer))
19:33:42 Join MethoS_ [0] (
19:44:31 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
19:46:12 Quit Seed (Remote closed the connection)
19:57:01*Buschel should better have done business planning at work rather than watching the German soccer team :(
19:57:19 Join OlivierBorowski [0] (
19:58:56Buschelsaratoga: you've got results which are not too promising on libmad
19:59:35 Quit MethoS_ (Remote closed the connection)
20:00:40Buschelwith mpc i get a large improvement on the dct stuff when using 32bit instead of 64bit −− but this has not been asm'ed yet
20:01:24saratogaBuschel: which mul instructions are you talking about?
20:01:56Buschelsaratoga: mpc's calc_new_V()
20:02:29saratogabut you're changing from 32x32=64 to 32x32=32 ?
20:02:37Buschelsaratoga: yes
20:02:54saratogathats only a single clock cycle savings though per op
20:03:05Buschelsaratoga: but maybe the gcc compiler produces better code with it?
20:03:44Buschelsaratoga: i'll compare the sso-like synthesis with the 64bit one in mpc −− both of which are asm'ed already.
20:03:58Buschelsaratoga: then we can compare the mpc synth with libmad synth
20:04:09saratogathat'd be nice
20:04:22saratogai'd love to know why the synth in mpc is so amazingly fast compared to mp3
20:04:44saratogawe need like 15-20MHz for synth alone in mp3
20:04:56 Quit nuonguy ("This computer has gone to sleep")
20:05:07Buschelsaratoga: you've measured 48% of 37MHz for the libmad synth, right?
20:05:23saratogathats more then quick and dirty all c IMDCT used in WMA needs
20:06:06saratogathough i get that normal is 39MHz
20:06:56Buschelsaratoga: that's about 18 MHz for synth −− mpc's synth needs about 11-14MHz (needs remeasurment though)
20:07:37gevaertsdionoea: I assume that in your FS #8663 comment 'with and without the patch' means the original udelay() removing patch ?
20:07:44saratogathe two codecs use the same synth process right (ignoring that MP3 have them out of order an all that)?
20:08:10Buschelsaratoga: correct. mpc is based upon MPEG 1 Layer2
20:08:22Buschel(filterbank wise)
20:08:53saratogai can't imagine just reordering the coefficients or whatever nonsense mp3 does really makes a 50+% difference
20:09:03saratogaso your decoder must be doing something smarter then mad
20:09:10Buschelsaratoga: how many percent did you measure for dct32?
20:10:26saratogaoff hand i can't remember if I included that in the synth figure though
20:10:42saratogaprobably not
20:10:43Buschelsaratoga: that's 6MHz (64bit). mpc needs 5.5MHz (32bit) or 7-8MHz (32bit)
20:11:34Buschelsaratoga: that's why i am interested in having an asm'ed dct32 for mpc −− should be nearly as fast as the 64bit-mpc solution
20:12:01saratogado you really benefit from faster mpc?
20:12:19saratogai thought it was fast enough that the CPU core is idle while you decode
20:12:28Buschelsaratoga: if you compare mpc's synth to libmad you'll see a lot of differences. e.g. the mpc synth is a tiny small simple loop.
20:12:59Buschelsaratoga: i am running my CPU @24MHz, I can lower it further and have some more runtime :)
20:13:25preglowsaratoga: did you get arm profiling working well?
20:13:50Buschelsaratoga: mpc uses totally different arrangement of arrays/samples for the synthesis
20:14:15saratogapreglow: no i never got it to work correctly, so i did run time tests with different functions commented out and then computed the %
20:14:35 Quit tvelocity ("Αποχώρησε")
20:14:45preglowif lostlogic ever surfaces again, we should ask him if he knows what's up with it
20:14:52saratogaBuschel: is it more efficient? as in fewer multiplies per sample?
20:15:22saratogapreglow: i got it to spit out profiles, but the addresses dont make sense in them so i haven't been able to use them
20:19:43 Quit OlivierBorowski (Read error: 110 (Connection timed out))
20:20:17 Join Seed [0] (
20:21:28Buschelsaratoga: it might be that mpc-implementation uses less ops. there is no different handling for eve or odd subband indices. also i can remember that the standard filterbank uses memmove for each processed subband sample, mpc only memmove's once per frame.
20:23:03Buschelsaratoga: changing the libmad filterbank is a lot of re-engineering and re-organizing internal buffers
20:23:17 Quit miepchen^schlaf ()
20:25:35 Join tvelocity [0] (
20:26:08saratogawould anyone be upset if I did a libmad commit that fixed it's tabs?
20:26:51amiconnIirc the consensus on imported code was to keep its formatting as-is, in order to ease merging of improvements
20:27:26saratogamad hasn't been updated in 4 years and the code is an absolute mess of spaces and tabs
20:29:03Buschelamiconn: as you seem to be available :o) -> FS #9076: do you mind to change the pcf init as done in this patch?
20:29:27*amiconn is a bit busy atm, will have a look later
20:29:43Buschelamiconn: thanks, come back with any comments on it
20:30:51saratogaBuschel: i hadn't noticed that you'd kept working on your boosting patch until yesterday
20:31:07saratogaare you aiming to get it commited soon?
20:31:30saratogai imagine it would be quite useful if done properly, particularly for flac users
20:32:22Buschelsaratoga: in fact i use changed clock since January. but i don't think it will be committed soon.
20:33:45Buschelsaratoga: the gui boost itself which is need to avoid sluggishness does boost on wheel scrolling, which may effect the clocking of plugins afaik
20:34:36Buschelsaratoga: btw, you can put the clocking to an extreme when using 15MHz normal clock (FLAC doesn't even boost then, except when buffering)
20:34:45 Join miepchen^schlaf [0] (
20:36:38saratogaBuschel: so you'd need a way to only enable this when theres no plugins running?
20:36:40preglowsaratoga: well, if it does get updated, it'll be hell to merge
20:36:48preglowso lets just keep whitespace as it is in imported code
20:37:17preglowsaratoga: the god awful mix of tabs and spaces work out if you set your tabstop to eight spaces
20:37:28preglowit's still a sin, but at least it's readable
20:38:18 Quit saratoga ("CGI:IRC (EOF)")
20:38:23 Join saratoga2 [0] (n=9803c50e@gateway/web/cgi-irc/
20:38:41Buschelsaratoga: i think we need a way to enable the boost for gui actions in general −− the current patch only boost for scrollwheels. the patch is more like a proof-of-concept for gui boosting. my 5.5G is more responsive than svn, even though it's clocked @24MHz normal clock :)
20:39:28saratoga2i haven't looked, but since the backlight has a mechanism to do that, i can't imagine it'd be too hard
20:39:59saratoga2i mean it manages to turn on and off consistently enough, so a similar mechanism for boosting shouldn't be too hard
20:50:54*preglow summons shotofadds
20:51:54Buschelsaratoga2: i need to correct myself -> the patch already is "generic" in terms of buttons −− so, it's not for scrollwheel only. but there is some special handling needed for scrollwheel targets
20:52:19Buschelsaratoga2: it boosts on each button activity and unboosts after timeout
20:52:34Buschelsaratoga2: simple but effective :o)
20:54:50Buschelsaratoga2: another idea comes up my mind regarding the smull-stuff in libmad -> doesn't smull/mlal need less time when the operand is small (like most significant bytes are = 0)?
20:55:45saratoga2Buschel: in theory yes, but in practice I haven't been able to get that to work on PP
20:56:16saratoga2i tried zeroing the lower 2 bytes on all my WMA window tables and there was no change at all in run time
20:56:27saratoga2i suspect that PP may have somehow broken that
20:56:39saratoga2though I'm not certain
20:56:47Buschelhmm, bad. otherwise we could just do some simple rescaling
20:57:20saratoga2though i didn't try scaling so that the top two bytes where 0
20:58:14Buschele.g. this could be done for the D-coefficients. of course proper rescaling on the results must be done (similar to sso)
20:58:52Buschelthe D-coefficients can be scaled by >>12 without loss
20:59:26preglowthis early termination of the multiplier only works for the right operand
20:59:34preglowthe arm docs specify it
20:59:52Buschellet me check
21:00:00saratoga2i tried both operands to be sure
21:00:13preglowsaratoga2: did you get any solid timings out of the mul instructions?
21:00:22saratoga2the spec says "bits [31:24]"
21:00:31saratoga2does that refer to the bottom or top byte?
21:00:31preglowand 23:16, 15:8
21:01:02 Join einhirn_ [0] (
21:01:07preglowone cycle is shaved off if the top byte is zero or all ones, another if the next byte is zeros/ones
21:01:12preglowthe lowest byte doesn't matter
21:01:17saratoga2preglow: the datasheet says 5 clocks for mul, but I've never tried to measure it
21:01:42saratoga2so small numbers are faster, but big numbers with lots of trailing zeros aren't?
21:01:43preglowsaratoga2: well, if pp somehow don't use arms regular implementation for the mul, it would be interesting indeed to find out what the timings are
21:01:54preglowsaratoga2: indeed
21:02:00saratoga2ok then i did it wrong
21:02:11Buschelpreglow: signed numbers are slow then in general, right?
21:02:11preglowsaratoga2: smaller numbers are better, both positive or negative
21:02:19saratoga2i just AND'ed the lookup table with 0xFFFF0000
21:02:27saratoga2no they're the same
21:02:28preglowsaratoga2: no, since it also does early termination if a byte is all ones
21:02:35preglowBuschel: that last was to you :)
21:02:45preglowsaratoga2: that won't matter at all, yes
21:02:46saratoga2not for signed
21:02:48Buschelpreglow: yep, got it
21:02:52saratoga2all 0 or all 1
21:03:14saratoga2thats really annoying that they do it that way, since it messes up fixed point operations
21:04:21Buschellibmad's synth uses the D-coefs as 2nd (=right) operand
21:04:32Buschelso downscaling should have some effect then
21:04:44saratoga2then why didn't I see an improvment from SSO?
21:05:16 Join PaulJam_ [0] (
21:05:16Buschelsaratoga2: good question
21:05:25saratoga2that should have saved 1 cycle for the mul, another for write back of the 64 bit value, and possibly a third for any shifts that were saved
21:05:38saratoga2instead i got maybe a 1-2% speedup of synth_full
21:05:54preglowBuschel: downscaling will have some effect, i also think the regular mul and mla instructions have this early termination
21:06:18*preglow likes the core in his d2, which has mul and smull which are both faster and fully deterministic
21:06:59saratoga2I should put a block in libmad that does 1 million muls and see how long it takes with various output argument widths
21:09:55***Saving seen data "./dancer.seen"
21:10:03Buschelhmm, mpc uses D-coefs in the left operand. maybe some more speed possible then? ;-)
21:10:59saratoga2let me know if you get any speed up
21:11:09preglowsaratoga2: would be very interesting
21:12:05Buschelgotta switch to dev-pc now
21:13:05 Join DerDome [0] (
21:14:27 Quit Seed (Remote closed the connection)
21:15:25 Join Seed [0] (
21:15:53 Quit jhMikeS (Read error: 104 (Connection reset by peer))
21:16:08 Quit desowin ("KVIrc 4.0.0 Insomnia")
21:19:52 Quit PaulJam (Read error: 110 (Connection timed out))
21:21:15preglowi like how the string "A:\" appears in the d2 firmware
21:21:54preglowone gold nugget among many
21:21:56LambdaCalculus37preglow: Heh... :)
21:22:04LambdaCalculus37Got any floppy disks?
21:25:24 Quit nplus (Read error: 104 (Connection reset by peer))
21:25:45preglowperhaps they hadn't yet decided between sd cards and floppy disks early in the design phase
21:26:14preglowi swear, 100kb could be shaved off this firmware if they removed all their bad assembly
21:26:27preglowif (r0 == 0) { r0 = 0; return; }
21:26:46 Join nplus [0] (
21:27:13linuxstbI didn't know they were using gcc...
21:27:15 Join desowin [0] (n=desowin@atheme/member/desowin)
21:28:00preglowfor some reason, there are also cascades of "bx lr" lines
21:28:05preglowperhaps they use them as padding
21:28:11preglowbut it doesn't look like it
21:28:17preglowunless they like padding to arbitrary boundaries
21:28:31linuxstbI think I've noticed those in other tcc firmwares
21:28:53LambdaCalculus37What? That they leave unessential code all over the place?
21:28:57 Join neddy1 [0] (n=john@nat/sun/x-29c82c3e45484236)
21:29:32 Join webguest67 [0] (n=5ab841d0@gateway/web/cgi-irc/
21:29:43 Quit neddy1 (Client Quit)
21:30:28 Quit webguest67 (Client Quit)
21:30:37 Join neddy1 [0] (n=john@nat/sun/x-3d05a4dd638f4b35)
21:31:35preglowlots of mentions of aac+ in here too
21:31:38preglowi didn't think this thing did aac
21:32:43linuxstbpreglow: I wonder if DMB uses it...
21:33:34 Part kennyj
21:34:10 Join pixelma [0] (i=pixelma@
21:34:46linuxstbYes, it does... AAC+, "the BSAC variant of AAC" and MP2
21:35:50preglowtoo bad they don't support the file version
21:35:56preglownot that i care...
21:36:56linuxstbShould be relatively easy for them to support DAB+ though.
21:41:29preglowif the reed solomon decoder can handle the dab+ encoding, then yes
21:41:38 Join Lear [0] (i=chatzill@rockbox/developer/lear)
21:41:42preglowsweet god, decoding 50++ "bx lr" now...
21:46:18amiconnlinuxstb: Is there some ViewVC for brosing the ipl cvs?
21:47:10 Join Buschel_ [0] (
21:47:35linuxstbamiconn: There seems to be something here -
21:47:47linuxstbLooks like they've moved to SVN at sourceforge...
21:49:53 Quit TMM (Remote closed the connection)
21:50:50amiconnHmm, this lcd_cmd_data() block in fb.c is the only change in ipl revision 650 ...
21:51:21preglowseems this thing supports fat12
21:57:26amiconnlinuxstb: Could you take a closer look at lcd-color_nano.c lines 95..103?
21:57:55amiconnIiuc gpio_a01 doesn't matter at all...
21:58:36 Quit csc` (Remote closed the connection)
21:58:40linuxstbamiconn: One moment... I was just reading the IPL IRC logs from the date of that commit -
21:58:45amiconnNot that this simplification would help solving the ipod photo bug.... but perhaps this check is still a bit wrong
21:58:57linuxstbSeems the problem was caused with the upgrade from v1.0 to v1.1 of the Photo OF
21:59:53 Quit LambdaCalculus37 (" ajax IRC Client")
22:00:09 Join csc` [0] (n=csc@archlinux/user/csc)
22:01:06 Quit Buschel_ ()
22:06:06amiconnlinuxstb: Do you have an idea why the fix doesn't work in rockbox?
22:06:26 Join powr-toc [0] (
22:06:44powr-tocHey, is anyone else having trouble downloading builds from
22:07:01linuxstbamiconn: No, I'm just looking at things now. And I think I agree - gpio_a01 is irrelvant - it's just "lcd_type = gpio_a04;"
22:08:10amiconnThat reminds me that I still need to do some research regarding the PP colour lcd bridge
22:08:29amiconnIt might pay off to split the color and nano lcd driver at some point, btw
22:09:00amiconnThe nano has some potential for optimising yuv blitting that cannot be applied to the color (because of the 2 lcd types)
22:09:40Buschelsaratoga2/preglow: i am not sure about my results on prescaling the "right operand" of mul/smull
22:10:33Buschelsaratoga2/preglow: >>8 seems to speed up by 1 cycle, but a lot of the operands are =0 in this case (measured with filterbank)
22:10:38 Quit powr-toc (Remote closed the connection)
22:11:11Buschelsaratoga2/preglow: but results on >>16 with smull/mlal are really strange. it's slower...
22:11:18Buschel(by 1 cycle)
22:11:19 Quit Enginerd3 ("CGI:IRC (EOF)")
22:12:28Buschelsaratoga2: maybe you'll get more precise results from your "1 million mul" test
22:12:45*amiconn bookmarked the ipl ViewVC now
22:12:52 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
22:19:28 Join mirak [0] (
22:25:13 Quit linuxstb (Read error: 113 (No route to host))
22:25:37 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
22:27:59 Join TMM [0] (
22:30:51 Quit nplus (Remote closed the connection)
22:31:50 Quit Lear ("ChatZilla [Firefox 3.0/2008052906]")
22:36:55preglowBuschel: weird
22:37:13preglowBuschel: have you looked at the actual compiled code to make sure gcc doesn't mess anything up?
22:37:52 Join jswetz [0] (
22:39:51 Join crope [0] (
22:41:25 Quit csc` (Remote closed the connection)
22:41:30 Join webguest56 [0] (n=4515baea@gateway/web/cgi-irc/
22:41:50Buschelpreglow: i am not sure what this came from, only did several code changes, rebuilds and test runs
22:42:49 Quit DerPapst (Read error: 110 (Connection timed out))
22:43:20 Join mraudio [0] (n=4515baea@gateway/web/cgi-irc/
22:46:56 Quit webguest56 ("CGI:IRC (Ping timeout)")
22:47:01Buschelbye, see you tomorrow
22:47:03 Quit Buschel ()
22:48:47 Quit mraudio ("CGI:IRC (Ping timeout)")
22:48:52 Join bluebrother [0] (n=dom@rockbox/staff/bluebrother)
22:49:19 Quit mirak ("Ex-Chat")
22:49:32 Quit miepchen^schlaf ()
22:51:27 Join miepchen^schlaf [0] (
22:52:45preglowwhy would the firmware contain http response strings...
22:55:11bluebrotherdomonoky1: around?
22:56:10bluebrotherI noticed sectorbuf and verbose being set as global variables in ipodpatcher.c and getting reused by sansapatcher.
22:56:36bluebrotherIs there a reason this is done that way? Makes a somewhat inconsistent impression to me
22:57:38bluebrotherI'm wondering if it's better to move the variable to the bootloader base class or to split it into two variables −− sansapatcher and ipodpatcher having its own
22:57:55bluebrotherthat would also make it easier to lib-ify the patchers if we want to do that later on
22:58:35 Quit PaulJam_ (".")
22:58:48domonoky1but i think you have to change the patchers to do that, i didnt want to modify much.... but feel free to change..
22:59:03domonoky1haveing them seperatly would probably be better..
23:01:00bluebrotherwell, I already made some small changed to the patchers, so I think I'd do that too
23:01:27 Join csc` [0] (n=csc@archlinux/user/csc)
23:05:41 Quit linuxstb (Read error: 104 (Connection reset by peer))
23:05:49 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
23:06:10 Quit neddy1 ("Leaving.")
23:06:52 Quit davina ("GNU/Linux the free alternative to Windows")
23:07:56 Quit domonoky1 (Read error: 104 (Connection reset by peer))
23:09:59***Saving seen data "./dancer.seen"
23:10:57 Quit lolmaus ()
23:18:13 Quit bluebrother ("Verlassend")
23:20:17 Quit Mathiasdm ("Yuuw!")
23:28:34 Quit jswetz (Remote closed the connection)
23:31:47 Quit bughunter2 ("bye")
23:32:45 Join SliMM [0] (n=SliMM@
23:33:19SliMMis the mediawiki plugin in the standard build?
23:33:45LloreanIt's a patch in the tracker.
23:34:16SliMMbut not in the standard build, right?
23:34:47 Join neddy1 [0] (n=john@nat/sun/x-3b951c0ecea18480)
23:34:53LloreanIf it were in the standard build, a patch to add the feature would no longer be necessary.
23:36:10 Quit desowin ("KVIrc 4.0.0 Insomnia")
23:38:15SliMMcan I use background images with viewports, or make a viewport's background transparent?
23:40:12LloreanBackground images still work how they always did (one fullscreen background image)
23:42:49 Quit XavierGr ()
23:43:12LloreanWhen there's a background image, colored viewports don't show their color. They're transparent automatically
23:43:18LloreanAt least, if i understand correctly
23:43:27 Quit petur ("*plop*")
23:49:05 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
23:53:27 Join powr-toc [0] (
23:54:03powr-tocI've jus updated rockbox on my ipod video 30G, and I keep getting codec error when I try and play mp3s... any ideas whats up?
23:54:43scorche|shpowr-toc: you likely didnt install correctly...did you extract the entire zip to the device?
23:55:04powr-tocI think so, but I'll try again
23:56:32 Join d1sturb [0] (
23:56:59powr-tocscorche|sh: cheers... I obviosly missed something :-)
23:57:09d1sturbThanks, I got it to work. Just wondering, is there any app you guys no that lets you play videos on 2nd gen nano?
23:57:48scorche|shd1sturb: rockbox doesnt work on the 2nd gen nano
23:58:52d1sturbI know, but there are no other apps that do?

Previous day | Next day