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 2010-09-06

00:11:00 Quit Judas_PhD (Quit: This is a quitting message)
00:18:31 Quit Jerom (Read error: Connection reset by peer)
00:22:41 Join Evilnick_ [0] (~Evilnick@cpe-24-193-43-185.nyc.res.rr.com)
00:24:31 Quit Evilnick (Ping timeout: 240 seconds)
00:25:54 Quit ender` (Quit: The sum of intelligence on the planet is a constant; the population is growing.)
00:35:04 Quit pamaury (Remote host closed the connection)
00:35:25 Quit Dreamxtreme (Quit: Don't follow me)
00:42:43 Quit _s1gma (Quit: irl ragequit)
00:44:58 Join krabador [0] (~krabador@host18-176-dynamic.33-79-r.retail.telecomitalia.it)
00:46:02 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
00:47:16 Quit Evilnick_ (Ping timeout: 245 seconds)
00:48:06 Join advcomp2019_ [0] (~advcomp20@unaffiliated/advcomp2019)
00:48:21 Quit advcomp2019 (Read error: Connection reset by peer)
00:48:33 Quit Stummi (Read error: Operation timed out)
00:48:35 Join shai_ [0] (~Shai@l192-117-110-233.cable.actcom.net.il)
00:48:35 Join lestatar__ [0] (~chatzilla@cpe-72-229-41-214.nyc.res.rr.com)
00:48:46 Join Stummi [0] (stummi@doppeldenk.org)
00:52:10 Quit lestatar (Ping timeout: 252 seconds)
00:52:10 Quit shai (Ping timeout: 252 seconds)
00:52:20 Nick lestatar__ is now known as lestatar (~chatzilla@cpe-72-229-41-214.nyc.res.rr.com)
00:54:30 Join JdGordon1 [0] (~jonno@111.220.211.253)
00:54:40 Join Evilnick [0] (~Evilnick@cpe-24-193-43-185.nyc.res.rr.com)
00:56:03 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
00:57:06 Join Dreamxtreme [0] (~Dreamxtre@92.30.24.203)
00:57:25 Quit bluebroth3r (Read error: Operation timed out)
00:57:32***Saving seen data "./dancer.seen"
00:58:08 Quit JdGordon (Ping timeout: 276 seconds)
01:00
01:04:12 Quit bmbl (Quit: Bye!)
01:10:29 Join Judas_PhD [0] (~kevin@misterfluffy.dsl.xmission.com)
01:20:19 Join Evilnick_ [0] (~Evilnick@cpe-24-193-43-185.nyc.res.rr.com)
01:22:58 Quit Evilnick (Ping timeout: 252 seconds)
01:26:21eclecticSystem: Host sting Kernel 2.6.31-14-generic i686 (32 bit) Distro Linux Mint 8 Helena - Xfce Community Edition
01:26:24eclecticSystem: Host sting Kernel 2.6.31-14-generic i686 (32 bit) Distro Linux Mint 8 Helena - Xfce Community Edition
01:26:48[Saint]eclectic: Yo! ...errr, don't, please
01:27:06eclecticsorry! didn't mean to!
01:30:15 Part eclectic ("Leaving")
01:30:48 Join eclectic [0] (~bill@pool-72-81-182-171.bltmmd.east.verizon.net)
01:33:44eclecticI've been running Rockbox since July on my v2 Sansa Fuze, and have become addicted to Doom. At some point Doom has started crashing whenever I try to enter "map 2"
01:34:11eclecticI have the error message Rockbox gives handy.
01:34:30[Saint]RAM, IIRC
01:34:51[Saint]not a lot that can be done about that, except find a wad with smaller maps
01:35:12[Saint]the V2 fuze has considerable less ram than the V1
01:35:35eclecticbut the next level was working fine for weeks?
01:36:09[Saint]Hmmm...different allocations of RAM elsewhere? pass.
01:36:52[Saint]I'm happy for someone to chime in and say "no, its this...", but I believe the issue is available RAm
01:37:00[Saint]*RAM
01:38:53eclecticI installed new versions of Rockbox, plus fonts and themes (that I don't really use). I wonder if all that gets loaded into ram, if so I could remove stuff.
01:39:40[Saint]as I understand it, it doesn't...but, again, happy to be corrected.
01:41:50[Saint]If DOOM is a critical function for you, roll back to a last known good version and then make a not of the behaviour on the forums, and/or the tracker.
01:41:59[Saint]*note
01:42:48[Saint]Doom is a bit of a burden for a lot of devs though, it sees very little development.
01:42:54[Saint]More of a novelty.
01:43:27eclecticI was wondering about that: my original Rockbox version was dated 7-10-10, but I have not figured out how to go back to that.
01:43:43mulenmarWait, do the themes/fonts get unloaded when Doom is loaded? (I've never used it)
01:44:03mulenmarIf not, and you're using a different theme/font, maybe it uses more memory than the old one?
01:44:03[Saint]mulenmar: I highly doubt that.
01:44:18eclecticI'm an odd case in that I've never gotten around to a game like Doom 'till Rockbox had it.
01:44:26[Saint]s/that/that they get unloaded/
01:45:14eclecticI tried fonts and themes but I went back to cabbie for daily use.
01:47:05eclecticI tried to in stall Doom on the computer and it somehow destroyed X window system.
01:49:47 Quit [Saint] (Ping timeout: 255 seconds)
01:50:29eclecticI've spent a good deal of time looking, but have not found instructions for reloading my original Rockbox version from July 10th
01:50:56 Join [Saint] [0] (S_a_i_n_t@203.184.2.148)
01:51:53mulenmareclectic: If you don't have the zipfile you downloaded still, you'll have to build it from SVN.
01:52:04eclecticthere is a file on my Fuze called rockbox.sansa dated 7-10-10; I wonder if that is the backup file from that version.
01:53:20eclecticoh, yeah I'm sure I have the orginal zip file. Is it necessary to uninstall Rockbox first, or just install over the current version?
01:53:36[Saint]just extract overtop
01:54:08eclecticcool. I'll try putting everything back to how it first was. And let you all know.
01:54:28eclecticThanks to you all for all your work, and your suggestions!!
01:55:10eclecticI LOVE Rockbox on my Fuze! Waited two years till I heard it was ready.
01:55:51 Quit Evilnick_ (Ping timeout: 240 seconds)
01:55:58 Join Evilnick [0] (~Evilnick@cpe-24-193-43-185.nyc.res.rr.com)
01:57:31 Join JdGordon| [0] (~jonno@vl10.gw.ok-labs.com)
01:57:31 Quit JdGordon| (Changing host)
01:57:31 Join JdGordon| [0] (~jonno@rockbox/developer/JdGordon)
02:00
02:03:13 Nick Guest13872 is now known as froggyman (~seth@pool-72-69-207-158.chi01.dsl-w.verizon.net)
02:03:21 Quit froggyman (Changing host)
02:03:21 Join froggyman [0] (~seth@unaffiliated/froggyman)
02:08:05 Join steve|m1 [0] (~steve@p4FD46842.dip.t-dialin.net)
02:08:05 Quit steve|m (Disconnected by services)
02:08:08 Nick steve|m1 is now known as steve|m (~steve@p4FD46842.dip.t-dialin.net)
02:08:40 Join alexbobP [0] (~alex@ppp-70-253-65-239.dsl.austtx.swbell.net)
02:09:59alexbobPI just got a sansa fuze, and it seems to have the latest firmware (2.3.33A)
02:10:11alexbobPis there still any way to load the rockbox bootloader?
02:10:18alexbobPor will I have to wait for another sansa firmware to come out?
02:10:44[Saint]http://download.rockbox.org/daily/manual/rockbox-sansafuzev2/rockbox-build.html
02:10:55[Saint]*should cover installation: ^
02:16:18alexbobP[Saint]: ooh, thanks
02:16:23alexbobPI had tried the auto installer and it didn't work
02:16:32alexbobPbut now I have rockbox running :D
02:16:34alexbobPawesome!
02:16:47JdGordon|IIRC there is an issue trying to install onto the latest OF version, you might have to do it manually
02:29:27 Join McDougal [0] (~McDougal@pool-173-48-186-93.bstnma.fios.verizon.net)
02:29:38McDougalhello again
02:29:56McDougali have another problem
02:30:31McDougalmy problem is with album art
02:30:55[Saint]McDougal: Well, since we're no psychic ;)
02:31:05[Saint]*not
02:31:19McDougali play a song or album, the album art shows on the now playing screen, but whenever i go to anywhere else (menu, etc.) and then go back to the now playing screen, the album art disappears
02:31:33McDougalthis happens with every theme except cabbiev2
02:31:47[Saint]McDougal: likely the theme doesn't display the AA in its own viewport.
02:31:51[Saint]which theme is this?
02:32:06[Saint]theme(s)
02:32:07McDougalthe theme I use most often is RED, and amarok
02:32:09*JdGordon| really doesnt like that answer
02:32:25JdGordon|there is absolutly no reason for it to work on first go but not after coming back from menus
02:32:44McDougalokay
02:32:56[Saint]JdGordon1: No, I agree...but, it happens.
02:33:15McDougalso is there a fix for a fuze v2 using the current build?
02:33:18JdGordon|is there a bug for this?
02:33:23[Saint]I've fixed the same thing many a time by "viewportifying" the AA/wps elements
02:33:40McDougalhow do you do that
02:33:59*JdGordon| should probably look into this one
02:34:02[Saint]JdGordon|: Not that I know of, I thought that the new parser basically meant "if it isn't in ts own viewport...expect amusing hillarity"
02:34:14JdGordon|no :)
02:34:15 Join CaptainKwel [0] (~jason@207-38-215-126.c3-0.nyr-ubr1.nyr.ny.cable.rcn.com)
02:35:12[Saint]wll, it seems to ;)
02:35:12[Saint]*well.
02:35:12*JdGordon| is really a joker IRL
02:35:12McDougalso how did you guys solve this problem
02:35:12[Saint]seems to stack up with cabbiev2 working, also.
02:35:12DBUGEnqueued KICK [Saint]
02:35:12[Saint]McDougal: Updating the theme to draw theme elements correctly.
02:35:27McDougalokay, how do i do that (im new to this stuff)
02:36:35[Saint]probably best to attempt to contact the theme author, inthat case.
02:36:35[Saint]*author(s)
02:36:35McDougalokay....
02:36:35[Saint]It is their responsibility, really...but you can still fix it yourself.
02:36:35CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
02:36:35*[Saint] digs up the wiki links
02:36:55McDougalwell if you tell me how i can fix it, i would be glad to do it myself
02:37:20JdGordon|put %V(0,0,-,-,1) above the line with %Cd and it should maybe work
02:37:31McDougalokay, i'll try that
02:37:57[Saint]the AA needs to be drawn in its own viewport, and not in the "default (fullscreen) viewport" if other viewports are used.
02:38:32[Saint]JdGordon|'s solution may work, but, could cause interesting issues with overlapping viewports and updating
02:38:41McDougalwait a sec i dont get it, what do i use to put the V%(0, 0, -, -, 1) above the line with %Cd?
02:38:55[Saint]any text editor
02:39:06McDougalokay, so what file do i open with notepad?
02:39:26[Saint]<themename>.wps
02:39:35McDougalokay, thanks
02:44:36krabador'night people, i've no hope to find an old release of the sansa bootloader?
02:44:41McDougali will do this later and report back when i can with the results
02:44:43[Saint]krabador: Not without compiling your own...no.
02:45:31krabador[Saint], i installed subversion, and rapidsvn, on ubuntu linux
02:46:20krabadori would download only bootloader sources, but in rapidsvn, i need user and password to scrool the svn
02:47:04krabadorit's a little bit annoying to download all rockbox sources, only for the bootloader
02:47:16[Saint]it will *really* improve your situation learning to compile your own builds.
02:47:52[Saint]krabador: I've heard you ask for older builds several times also, though...so, it would come in handy for you evidently.
02:48:20krabador[Saint], i only need an older version of sansa bootloader
02:48:44krabadori've some problem with the latest
02:49:01[Saint]wait..the actual Sansa bootloader, or, the RB sansa bootloader for your port?
02:49:07[Saint](big difference)
02:49:48krabador[Saint], the bootloader i must insert in the OF, for the dual boot
02:50:16[Saint]Right, you'll need to download the RB source.'
02:50:38[Saint]cherrypicking the source to get all the relevant code will be a royal pain in the ass.
02:50:48[Saint]best to just grab it all.
02:50:50krabadorthis http://download.rockbox.org/bootloader/sandisk-sansa/fuze/bootloader-fuze.sansa
02:50:55krabadorbut not the latest
02:51:32[Saint]Yeah, no real choice other than building it yourself.
02:51:32krabadoroh, my god....
02:51:56krabadoryou rockbox developers are a little bit bondage...
02:52:14[Saint]a: I'm not a developer...and b: what!?!
02:52:44[Saint]I'm simply telling you what you'll have to do, if that makes me "bondage", then so be it.
02:53:12krabadorwhat's the problem to take previous versions in dowload folder, it's stuff of 90k :((((ù
02:53:53[Saint]why, when its literally childs play to set up a dev environment and built it yourself?
02:54:00[Saint]why lose the bandwidth?
02:54:28mulenmarUnfortunately krabador, it's the easiest way to get it and build it. I wish I had an archived version of the Subversion repo, but I don't.
02:54:59krabadori'll cut my veins....
02:55:00krabador:)
02:55:14mulenmarThat won't help. >_>
02:55:16[Saint]Your loss.
02:55:41mulenmarIf you want to donate blood, the American Red Cross is always having blood drives. ;~)
02:55:53JdGordon|PATETS!
02:56:29mulenmarJdGordon|: a: Offtopic b: Misspelled. c: That level of funny is patented and d: What?
02:56:51[Saint]krabador: You have unix, you have SVN, all you need to do is download the source...once, ever. and you can build all the bootloaders and builds your heart desires.
02:56:57[Saint]why is that so hard, really?
02:57:27mulenmarThis brings up an idea I had, but dismissed because of the possible complexity: build scripts that allow only building PARTS of Rockbox (to not build some of the games, codecs, et cetra).
02:57:34***Saving seen data "./dancer.seen"
02:57:35krabador(i haven't enough space to download all RB surces.....) (don't shoot me...)
02:58:23[Saint]if you don't have the space to DL ~200MB of sources, your system has bigger problems to worry about.
02:58:49JdGordon|mulenmar: like make menuconfig in the linux kernel?
02:59:34[Saint]mulenmar: I think it is generally excepted that if you can build RB, you're not far away from figuring that out yourself.
02:59:48mulenmarJdGordon|: Yes −− like I said, I dismissed it thinking it would be too complicated.
02:59:51[Saint]but, it's not like the scripts would take up any space...so, plausible.
03:00
03:00:06[Saint]bah, not complicated at all.
03:00:13[Saint]just effort.
03:00:18JdGordon|lots of effort
03:00:20mulenmarkrabador: After running make clean in the build directory, the rockbox tree was 377MB.
03:00:38krabadorgreat....
03:00:39JdGordon|git it about half that iirc
03:00:43mulenmarkrabador: Depending on if you have the RAM, you could do the work in a ramdisk, maybe?
03:01:03krabadoryes, not bad
03:01:08mulenmarkrabador: Or you buy/borrow a USB flash drive.
03:01:18*[Saint] simply can't imagine a system with less than ~500mb free space total.
03:01:21mulenmarThen it's just the download.
03:01:35krabadormulenmar, i'm formatting one
03:01:37[Saint]which is a LOT less traffic than ~377MB
03:01:45[Saint](the DL)
03:01:48*mulenmar conjures up his first computer, an IBM AT with a 40MB hard drive, and gives it to [Saint]
03:02:06[Saint]mulenmar: A system capable of RB development
03:02:20alexbobP[Saint]: <3
03:02:27*alexbobP has a rockbox mp3 player now
03:02:47alexbobPyesterday I broke my mp3 player, it was a gigaware...
03:02:56alexbobPone of those weird brands hackers don't even know about
03:02:59[Saint]\o/ ...don't know what I did, but, happy to take the credit.
03:03:02*mulenmar conjures up his fourth computer, which had a Cyrix M-II (?) 120MHz cpu and a 1.2GB hard drive, and gives it to [Saint]
03:03:14alexbobPso I bougt a sansa fuze, since I had had one in the past but lost it before rockbox actually came out for it XD
03:03:17alexbobPnow I am a happy camper
03:03:26alexbobP[Saint]: ou linked me to the instructions :P
03:03:32alexbobP[Saint]: all I had done before that was try the RockboxUtility
03:03:39alexbobPthe manual install worked great
03:03:48[Saint]mulenmar: with a stripped *nix system, and the sources...that's still have more than ~500MB free :P
03:03:48mulenmarAnyway, good luck krabador. The wiki is very helpful on the whole checkout/build toolchain/configure/build process.
03:04:18[Saint]Dl source, run one script...wait...build.
03:04:19mulenmar[Saint] I know. It had Windows 98 and ~300MB free before I stripped it down. #98lite_rocks
03:05:53 Quit krabador (Ping timeout: 240 seconds)
03:13:49 Join krabador [0] (~krabador@host18-176-dynamic.33-79-r.retail.telecomitalia.it)
03:18:47krabadorok, i downloaded all the stuff in svn
03:21:13McDougalokay, i havent done it yet, but i can confirm that it is a theme problem. im too tired to look through all that code and do it tonight (I know, I know, so LAZY :D), but I will do it when I can. In the meantime, I found myself a new theme Darkness which works perfectly
03:22:37 Join bunnyboi [0] (~androgyne@cpe-72-224-17-58.nycap.res.rr.com)
03:22:56krabadorso, what part of sources i must look, to find bootloader sources?
03:25:28krabadori must compile sansa_as3525.c ?
03:26:25mulenmarHEY WAIT A SECOND −− krabador, did you look at http://www.rockbox.org/dl.cgi?bin=sansaclip by any chance?
03:26:44 Quit McDougal (Quit: Leaving)
03:27:24krabadormulenmar, no...
03:27:44mulenmarI don't know if the bootloader stuff is included in those archived builds (that I didn't know existed), since I'm a Nano2G user, but maybe that and the manual will help?
03:29:11krabadorin rockbox svn folder, i've a /bootloader/ folder, full of sources
03:31:57 Quit [Saint] (Ping timeout: 272 seconds)
03:32:04 Join [Saint] [0] (S_a_i_n_t@203.184.0.146)
03:33:25krabador[Saint], i got all, now, can you help me to understand how can i obtain a new/old bootloader-fuze.sansa?
03:34:12 Quit dfkt (Quit: -= SysReset 2.53=- Ph'nglui mglw'nafh Cthulhu R'lyeh wgah'nagl fhtagn.)
03:35:43eclecticwell, Doom still crashes instead of loading "map 2" with my original Rockbox firmware from July.
03:37:41JdGordon|so... anyone remember that feautre request which was bassically "jump to the next letter in the filebrowser"?
03:38:52JdGordon|well, it turns out to be trivially simple and I'm tihnking about making it an option to replace the current paged scrolling behaviour
03:39:26eclecticand, I remembered another thing that changed. I got a new Fuze since I started with Rockbox. Could Sansa have reduced the RAM on a new Fuze from one bought last year??
03:39:47eclecticboth v2.
03:41:10[Saint]eclectic: Nope.
03:41:24[Saint]I mean they could have, but, didn't...afaik
03:42:23eclecticI didn't think so. But, after many searches, I cannot find any clue online about actual Fuze CPU speed or RAM size.
03:42:47[Saint]v2 == 32 RAM iirc, v2 == 4
03:42:53[Saint]I *think*
03:43:05[Saint]so...big difference.
03:43:14[Saint]Guess they realised they didn't need it.
03:43:51[Saint]V2 is either 4 or 8 (can't remember, sorry...) but it is definitely substantially less than the V1
03:44:01eclecticso, Buffer in Rockbox Info app means RAM, then?
03:44:21[Saint]what are you looking at?
03:44:54[Saint]If you're looking at what I think, that's available skin buffer
03:44:57eclecticrockbox menu: system>rockbox info
03:44:59[Saint]used/available
03:45:17[Saint]skin buffer...I believe, don't think that lists RAM
03:45:50[Saint]krabador: You were handed http://www.rockbox.org/wiki/Main/LinuxSimpleGuideToCompiling earlier
03:45:54[Saint]that is all you need.
03:46:11eclecticit says Buffer: 5.03 MB, and skin RAM usage: 221/396 KB
03:46:27[Saint]run configure, select (B) for bootloader, run make...done
03:46:40krabador[Saint], yes, but how obtain perfectly bootloader-fuze.sansa
03:47:04krabadorlike the one in download.rockbox.org
03:47:10[Saint]run configure, select (B) for bootloader, run make...done
03:47:31krabadorok
03:47:34mulenmar*facepalms* I should've remembered that.
03:47:50[Saint]eclectic: That is audiobuffer, I _think_
03:48:17[Saint]I'm not sure you can view all available RAM anywhere...but, could be wrong.
03:48:18eclecticok, thanks
03:50:24eclecticso, why I ran Freedoom, map 2 fine in July and then it always crashes, remains a strange mystery. :(
03:50:28krabador[Saint], excuse me a thing, i downloaded the last revision , the 28001, if i compile something inside it , i'll compile the bootloader as the newest
03:50:40krabadornot?
03:54:48[Saint]yes.
03:56:06[Saint]"svn up -r XXXXX" is the command you want, if I remember.
03:56:22[Saint]where XXXXX is the build revision you want.
03:56:39[Saint]I use a script for this, perhaps an SVN expert will be able to confirm this.
04:00
04:00:08krabador"The bootloader only changes rarely" tells the fuze manual, i need only one before the 26 june
04:00:17krabadoror many times before
04:00:45krabador*july
04:01:29[Saint]so...use the commit logs/svn/whatever to find the revision you want...then use "svn up -r XXXXX" where XXXXX == revision number
04:01:37 Join funman [0] (~fun@rockbox/developer/funman)
04:01:56[Saint]to checkout that revision, and, build it.
04:02:32funmankrabador: if you want to use OF USB just press the left key to boot OF, no need to bother with building/upgrading/whatever
04:03:14[Saint]funman: he wants an older RB bootloader
04:03:31funmanwhy, i wonder
04:03:35JdGordon|Buffer on the info screen is the audio buffer, it is also the plugins maximum buffer space
04:03:39[Saint]"unstable USB"
04:03:54[Saint]funman: ^
04:04:12funmandon't let him shoot himself in the foot, no need to mess with the bootloaders to use OF usb
04:04:44[Saint]I'm letting him do what he wants to do...he's convinced the bootloader is the problem.
04:04:46[Saint]so.....
04:05:34[Saint]apparently USB went to hell with a bootloader upgrade, and, since I can't confirm or deny this...
04:06:20funmanit's not linked, the bootloader 'bootloads' and doesn't affect rockbox
04:06:59krabadorfunman, [Saint], no, i discovered that the problem isn't the bootloader, but if i plug , when the fuze is off, on a pc with ubuntu or other linux distro, rockbox runs and is unstable
04:07:11funmankrabador: press the left key then
04:07:20 Quit bunnyboi (Read error: Connection reset by peer)
04:08:13[Saint]ah, that update would have been nice.
04:08:36[Saint]but, I guess you want the behaviour of the older bootloader, as in, boot OF on USB plug?
04:09:06funmanthis has nothing to do with the bootloader, you need mkamsboot 1.3 to boot OF on USB plug
04:09:26 Join webguest412 [0] (~cf66404a@giant.haxx.se)
04:10:53webguest412is there an irc channel for the general user population?
04:11:04webguest412rockbox user population that is
04:11:15krabadorfunman, oh my god...
04:11:33krabadorfunman, this is what i wanted to know...
04:11:33[Saint]webguest412: as in?
04:11:49[Saint]webguest412: to talk about what, I mean?
04:12:07funmankrabador: next time read the manual: http://download.rockbox.org/daily/manual/rockbox-sansafuze/rockbox-buildch3.html#x5-280003.1.3
04:12:16webguest412well I just installed it and I'm going through the manual and there is some stuff I'm not getting
04:12:30[Saint]webguest412: this is the place, then.
04:12:56krabadorfunman, yes , i know how run the OF, but i don't wanted run at plug
04:13:55krabadorin the normal use of the reader, for now, with linux systems, would be annoying.
04:14:07webguest412Saint: OK. btw, I'm the same guy who had been pestering you for a few nights now cause I had trouble installing on my ipod 5.5g 80gb
04:14:48[Saint]webguest412: Don't remember, sorry...what is bothering you?
04:15:15 Quit amiconn (Disconnected by services)
04:15:17 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
04:15:32 Quit pixelma (Disconnected by services)
04:15:34 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
04:15:36 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
04:15:37 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
04:16:40krabadorbtw, it will be fixed soon , the usb plug on linux systems?
04:16:56funmanit works fine for me so no
04:17:03webguest412section 3.3 of the manual says "The player can be charged over USB without connecting to your computer by holding Menu while plugging in." But that is the same thing section 3.1.3 as a way of getting the ipod to boot back into the native os. That is in fact what happens. So how do I plug in for a charge without connecting (mounting I'm assuming this means)
04:18:17[Saint]webguest412: not quite.
04:18:43[Saint]when the player is booted into RB, hold menu, and plug USB and it will only charge, not mount.
04:18:55[Saint]if the player if OFF holding menu to boot will boot the OF
04:19:12 Quit TheSeven (Ping timeout: 252 seconds)
04:19:42krabadorfunman, when i plug, on linux , at rockbox on, usb mode are unstable, if i plug at rockbox off, are unstable too, and ubuntu 10.04, 9.10, try to mount it at loop
04:19:55[Saint]*if the player IS off, rather
04:20:05[Saint]webguest412: ^
04:20:18krabadorfunman, i can post some .log for it?
04:20:24[Saint]so, two different results, same key, different circumstances
04:20:25funmankrabador: first start by telling us what's on the screen
04:20:36webguest412saint: so then booting to RB, then holdidng menu whilst plugging the usb cable into the bottom, I could then turn it off and it would be charging unmounted?
04:20:59[Saint]no, you cannot turn it off when charging
04:21:09webguest412gotcha
04:21:20[Saint]however, RB uses sane charging for iPod now...so, it will *actually* charge
04:21:25 Join bunnyboi [0] (~androgyne@cpe-72-224-17-58.nycap.res.rr.com)
04:21:37krabadorfunman, i've the screen of usbmode, later 2-3 second it flip with rockbox menu
04:21:51[Saint]3.6 had a risk of running the battery flat, even while charging for HDD iPods
04:23:03 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
04:23:09[Saint]webguest412: Give it a go while you're here (booting the OF, and charging without mounting) while you're still here.
04:23:13krabadorfunman, then the system try to mount it
04:23:23krabadorbut without success
04:23:34funmanwhat is written on the fuze screen
04:24:17krabadorfunman, i've usb mode screen, flipping with rockbox menu
04:24:46webguest412cool, so I can play doom and charge at the same time!
04:24:51funmanok, can you paste the output of dmesg on www.pastie.org ?
04:24:53krabadorfunman, if i've multimedia mode
04:25:20krabadoriit works some seconds, later stops
04:25:24[Saint]webguest412: yep.
04:25:33krabadorthen flip the screen at rockbox menuù
04:26:00krabadorfunman, i haven't this in win7 home premium on a laptop
04:26:14funmanwin7 mounts ok ?
04:26:22krabadorfunman, yes
04:26:33krabadorall right, i can use all key modes
04:26:45krabadorand mounts ok
04:27:17funmanok, then paste the dmesg output of linux box (on www.pastie.org)
04:33:07[Saint]has anyone every brought up "why on earth is there three places you can log in on the RB main page?"?
04:33:19krabadorfunman, it's very long
04:33:52krabadorfunman, do you want something properly?
04:33:53JdGordon|[Saint]: and 2 on the forums
04:34:09[Saint]JdGordon|: Any idea why?
04:34:18[Saint]it seems excessive, and confusing.
04:34:29JdGordon|because people are lazy
04:34:39[Saint]what happened with new.rockbox.org for that matter I wonder...?
04:34:50[Saint]that looks *awesome*.
04:34:51funmankrabador: the last lines, please put them on pastie.org , and then paste the link here
04:35:30JdGordon|didnt it get in trouble with css?
04:35:42JdGordon|as in it became a fucking nightmare or something?
04:35:49[Saint]JdGordon|: pass.
04:36:30[Saint]I think the argument about the .svg pictures became too much, and, it fell in the "abandoned" hole
04:36:33[Saint]that was my impression, anyway.
04:37:00krabadorfunman, http://pastie.org/1140722
04:37:31 Quit Judas_PhD (Quit: This is a quitting message)
04:37:31funmankrabador: no sign of an USB connection
04:37:55[Saint]mulenmar: do you use pictureflow?
04:38:08[Saint]or, could you for the purposes of testing something?
04:38:53krabadorfunman, http://pastie.org/1140723
04:39:19funmanstill no sign of an USB connection
04:40:24[Saint]mulenmar: If you could test http://www.rockbox.org/tracker/task/11602?project=1&show_task=&order=dateopened&sort=desc with the patched build, and against vanilla SVN...I'd appreciate it.
04:40:40[Saint](with pictureflow, while playing audio)
04:43:24krabadorfun but now i'm using it...
04:44:20 Quit krabador (Quit: Sto andando via)
04:45:03 Join krabador [0] (~krabador@host18-176-dynamic.33-79-r.retail.telecomitalia.it)
04:45:43 Quit guymann (Quit: brb)
04:45:59 Join LambdaCalculus37 [0] (~rmenes@rockbox/staff/LambdaCalculus37)
04:46:58 Join Barahir_ [0] (~jonathan@frnk-590fc68f.pool.mediaWays.net)
04:48:20 Quit krabador (Client Quit)
04:49:45[Saint]wow, lol...I just discovered someone else did backlight brightness in FS #11546 the same way as I did in FS #11578 so either I did it correctly, or there is only one way of doing it and it's a kludge, or we're both sucky coders linked inexplicably in some fashion.
04:50:49 Quit Barahir (Ping timeout: 276 seconds)
04:54:01funmanclipv1 with pclk/unboosted fclk @ 31MHz gives 1 more hour runtime
04:55:41 Quit MethoS- (Read error: Connection reset by peer)
04:57:38***Saving seen data "./dancer.seen"
04:57:53mulenmar[Saint] I'm running batterybench right now as part of testing how much of an effect the lcd sleep patch has on power usage, but I'm putting it on the to-do list.
04:58:36mulenmar[Saint]: I'll do it right after this first drain finishes, though.
05:00
05:09:40[Saint]liar: I have noticed that your backlight patch has a really bad effect on the bootloader splash screen
05:11:00[Saint]there is a veery visible *blip* when switching from the bootloader to the main menu, where it used to display my custom bootsplash, and now flashes a white, garbled screen for a *very* brief, yet visible period.
05:21:10[Saint]funman: pictureflow is still *very* bugged on Nano2g
05:27:06[Saint]funmanL loading pictureflow from the .wps with hotkey (btw, should that even work if "pictureflow integration" == off?) 4 out of five times it decides to rebuild the albumart cache, and 9 out of ten times it cannot find any files to load.
05:27:39funmani don't know pictureflow
05:28:06[Saint]after it loads, trying to exit the plugin causes "undefined instruction", or, failing that stops playback and playback cannot be reatarted.
05:28:31funmanwhere's the undefined instruction in the .map ?
05:28:47[Saint]it is still "playing", but, cannot be stopped/reatarted, the progressbar doesn't move at all (nor elapsed time) and there is no audio.
05:29:00[Saint]funman: it isn't in the map, at all.
05:29:17[Saint]I asked you about it/mentioned it to you as you were the last to touch pictureflow
05:30:57[Saint]Hmmm...interesting, after playback "stops" the player needs to be reset, as there is no button response even though the play/pause icon changes....it requires a hard reset to exit it.
05:31:09funmando you have a sansa AMS to reproduce the bug?
05:31:12 Quit LambdaCalculus37 (Quit: Fwump)
05:31:13[Saint]And the playlist control file gets corrupted.
05:31:25[Saint]funman: sorry, no Sansa here.
05:32:21funmanbtw i didn't touch pictureflow, only the code where pictureflow returned
05:32:26[Saint]does this help? I get a "divide by zero" error if I manage to open a album from pictureflow.
05:32:32[Saint]Aha.
05:32:51funmanif you can find where in the .map / .c the divide by zero is, sure
05:32:55[Saint]I'll check the address the divide by zero gives in the map now
05:34:13 Quit ps-auxw (Ping timeout: 264 seconds)
05:34:51[Saint]Hmmm...what should I be looking for? the address the divide by zero gives is *completely* different to anything in the .map file
05:35:26[Saint]all I see in the .map is "0xXXXXXXXXXXXX" but I have 09F83C78"
05:35:50JdGordon|too many X's
05:35:54JdGordon|or too few
05:36:04[Saint]JdGordon1: yeah...you get the idea though..
05:36:18[Saint]the address the error gives is completely different.
05:36:30[Saint]no "0x<blah>" at all.
05:36:47[Saint]I can't find anything like it in the map...not even close.
05:42:00JdGordon|well that probably means its done a jump to nonesense and thinks it is at code
05:42:30mulenmarMeh, forget the battery benchmark, I messed it up. I'm going to test the vanilla Pictureflow and patched Pictureflow.
05:42:33 Quit Horscht (Quit: Verlassend)
05:44:11mulenmarI don't need to revert anything first, right? I don't have any other patches against pictureflow.c
05:45:14 Join ps-auxw [0] (~arneb@p4FF7F384.dip.t-dialin.net)
05:45:18[Saint]correct
05:49:47[Saint]mulenmar: I'm also interested to see if you can reproduce "skipping tracks *really* fast causes 'undefined instruction' error"
05:50:50mulenmar[Saint] Let me get the album art converted to a size that actually works, first. 8~/
05:51:05JdGordon|aa gets resized on load
05:51:09[Saint]mulenmar: that shouldn't necessary...at all.
05:51:50[Saint]it's pointless having it larger than the vertical height of the screen...but, it *could* be 1000x1000
05:52:31mulenmarhttp://www.rockbox.org/wiki/AlbumArt seems to indicate, at least in the bash script, that the thumbnails should be 130x130
05:52:47[Saint]well, if it does...its BS
05:53:01[Saint]AA can be any arbitrary size
05:53:23mulenmarThe album art (in the Diva Mirage albums I'm using for testing) are 251x251
05:53:27JdGordon|that was obviously written when it was needed :)
05:53:30[Saint]both pictureflow and thw .wps/.sbs resize it as needed.
05:53:33mulenmarWait, never mind, they're the WRONG FORMAT
05:53:38mulenmarJPEG instead of BMP
05:53:43mulenmar*facepalm*
05:53:43JdGordon|jpg is fine
05:53:47[Saint]that isn't wrong, either
05:53:48JdGordon|folder.jpg
05:54:14[Saint]so...shit knows what you're doing wrong.
05:54:26[Saint]jpg is supported for AA since ages ago.
05:54:44JdGordon|PF needs the database initialised
05:54:55[Saint]it dies indeed.
05:54:58[Saint]*does
05:54:59mulenmarWell then the wiki page needs a MASSIVE overhaul.
05:55:09[Saint]mulenmar: it's a wiki ;)
05:55:15mulenmarAnd I did initialize the database and reboot.
05:55:35mulenmarI know I've had cover art working before, it's just not a feature I use. (Too much space)
05:55:37[Saint]you shouldn't need to reboot, to commit the DB either
05:55:50 Join guymann [0] (~charlie@69.0.11.104)
05:55:54[Saint]btw...AA takes *very* little space.
05:55:59mulenmarWell the Database entry didn
05:56:20mulenmar't work −− it said to reboot
05:56:25[Saint]well, it takes very little space if you have it in a sane format/size
05:56:49[Saint]mulenmar: depends on your settings...it should init/commit in one step if it can.
05:56:55[Saint]I forget what governs that.
05:57:38[Saint]mulenmar: for my Nanos, I have my AA at 100px, and, it takes up very little space.
05:57:55[Saint]unless you have "per track" AA, which, is pointless.
05:58:10[Saint]you only need one "folder.jpg" in the album dir
05:59:09[Saint]mulenmar: you shouldn't technically need any AA present for this...anyway
05:59:18[Saint]it will simply show the default image.
05:59:28[Saint]It should still be reproducable with that.
06:00
06:01:59mulenmarWell, Pictureflow only displays the cover art for the one cover I converted to be "cover.bmp"
06:02:19mulenmarAnd that's after I forced it to reload its cache
06:02:33 Join user890104_ [0] (Venci@Venci-Notebook-LAN.ipv6.6bez10.info)
06:03:02 Quit Evilnick (Ping timeout: 245 seconds)
06:03:04mulenmar(Resize Covers -> Yes −− which is default it seems, then exit the plugin and then run it again.)
06:03:07funmanpamaury: http://pastie.org/1140803 <- i added a (triggered) assertion of linux
06:03:21[Saint]mulenmar: you're doing something wrong with your AA, then.
06:03:24 Quit user890104 (Ping timeout: 272 seconds)
06:03:31mulenmarAA = album art?
06:03:32[Saint]what, I have no idea...but, something.
06:03:40[Saint]yes. AA == album art.
06:03:47[Saint]does it show in the .wps?
06:04:33mulenmarYes, ONLY for the one I renamed and converted to bmp, though.
06:04:59mulenmarI'll test whether it's because of the conversion or just renaming −− give me a minute
06:05:01[Saint]Hmmmmm....I cannot use text_editor either, perhaps I have disk corruption, or this build is *really* fucked.
06:05:04JdGordon|are your jpgs umm.... thingamy
06:05:12JdGordon|not interlaced...
06:05:14JdGordon|umm...
06:05:52*[Saint] waits patiently to find what "thingamy" is/
06:05:53JdGordon|there is a JPG feature which isnt supported, which I cant remember right now
06:05:55JdGordon|Unhelpful: ?
06:06:38mulenmarAHHA!
06:06:55[Saint]mulenmar: what?
06:06:55mulenmarIt's not the JPEG vs bitmap thing, you were right.
06:07:02mulenmarIt's the NAME of the FILE.
06:07:12 Quit user890104_ (Ping timeout: 272 seconds)
06:07:19[Saint]indeed it is...well, depending on what the name *is*
06:07:33[Saint]what *is* the naming convention?
06:08:02JdGordon|folder.jpg
06:08:09mulenmarIn the Diva Mirage album "My Crazy Universe" I downloaded from Jamendo, the name is "[cover] Diva_Mirage - My Crazy Universe.jpg"
06:08:26mulenmarToo long/can't handle brackets?
06:08:29[Saint]I know that, well, that's only one of many...I meant "what is mulenmar using" ;)
06:08:53mulenmarThat's what it is downloaded direct from Jamendo.
06:08:53[Saint]mulenmar: completely wrong naming convention.
06:09:04[Saint]see the manual, under "advanced topics" IIRC
06:09:08JdGordon|mulenmar: there are half dozen specific formats which it searches for, none are like that.
06:09:20JdGordon|it is always best to just name them folder.jpg in the current folder
06:09:33[Saint]mulenmar: Jamendo hane nothing to do with RB, though.
06:09:49[Saint]assuming the same naming convention will work...is, odd.
06:10:14[Saint]<exact_file_name>.jpg works
06:10:24mulenmarMaybe it's the brackets?
06:10:31mulenmar[ ]
06:10:39mulenmarThey aren't escaped, after all.
06:10:52mulenmarAlso, album art is section B.20
06:10:55[Saint]No, it's the fact it differs from the album title in the metadata...even by one char
06:11:15[Saint]it is *way* easier to name them all "folder.jpg"
06:11:30[Saint]do a search, then batch rename them...piss easy
06:12:49 Quit panni_ (Read error: Connection reset by peer)
06:13:48mulenmarOkay, the brackets aren't special characters that have to be escaped somehow, right?
06:14:20[Saint]no...but, if its not in the file or album name...it WONT work
06:14:34[Saint]look at the naming conventions in the manual.
06:14:37mulenmarI am.
06:15:18[Saint]seriously, "folder.jpg" will save you all kinds of hassle
06:15:33mulenmarWhere is the parsing code, I wouldn't mind taking the chance here to help AND grow my C skills. The existing definitions would be a good example for me.
06:15:34[Saint]it means even if you rename the album or file, your AA won't break
06:15:53[Saint]parsing code for.....?
06:16:29mulenmarEr, the code for parsing the names, looking for certain naming conventions.
06:16:35mulenmarThe code that looks for the album art.
06:17:01funmandid someone already measured the time needed by Clip+ OF to charge ?
06:17:16[Saint]that's one for Unhelpful I'd say, mulenmar
06:17:35[Saint]he's fairly aware of the AA code..its all vodoo to me.
06:25:19[Saint]so, mulenmar....issues with pictureflow?
06:25:43mulenmarOops, got distracted.
06:26:27JdGordon|mulenmar: apps/recorder/albumart.c somewhere
06:27:25[Saint]is apps/recorder/ some legacy from the AJBR?
06:28:03JdGordon|yes
06:28:04mulenmarI only have five albums on here at the moment, they all have cover art.
06:28:24[Saint]try exiting it
06:28:25JdGordon|[Saint]: ditto apps/charcell
06:28:39mulenmarNo difficulty scrolling through on vanilla Rockbox without playback. Exiting now.
06:28:57mulenmarNo problem exiting.
06:28:59[Saint]and, also...theissue is using hotkey to enter pictureflow
06:29:22[Saint]not goingstraight to the plugin.
06:32:15mulenmarSo how do I set a hotkey for it?
06:32:33mulenmarConfig file?
06:32:49[Saint]settings - general settings - hotkey
06:33:10[Saint]wps hotkey - pictureflow
06:37:07mulenmarBetween that and the manual I figured how the hotkey and use it. WPS is when the song is loaded (playing or paused), right? I don't have a problem hotkeying into Pictureflow or exiting.
06:37:20mulenmarThen again, I ONLY have five albums and their cover art on here.
06:37:45mulenmarLet me download a few more, rename the AA, and try again, and I'll get back to you [Saint]
06:38:58[Saint]volume [of albums] shouldn't be an issue...looks like this is mine alone to debug, yay :/
06:39:19mulenmarThis is without the patch, though.
06:41:24[Saint]ah...i'm only really interested in *with* the patch
06:42:00mulenmarHold on, that build MUST be finished by now.
06:43:03[Saint]i'm doig a full disk check/repair just in case...and I also revertedenogh patches so tht ccache couldn't reuse nything, both are taking ages
06:46:19mulenmarAlso, I just got repeat results: on my Nano2G, a reboot is definitely required before the database is commited.
06:46:29mulenmar(According to the software)
06:49:10[Saint]not in all cases...no.
06:50:20[Saint]I neverhave to reboot to finish theiit, i suspet I have a morefinly tuned coni
06:50:37[Saint]* the init
06:51:30mulenmar[Saint]: By "skipping tracks *really* fast, do you mean from in Pictureflow's file list, between files?
06:51:45[Saint]no
06:51:52[Saint]in the .wps
06:52:59mulenmarAh, different problem, of PEBKAC class. Because if I do that very quickly about eight to ten times in a row, it stops loading the song. I can quit the plugin just fine though.
06:53:05mulenmarAs for the .wps . . .
06:53:08*mulenmar checks
06:54:35mulenmarUh-huh −−- I guess my error wasn't in PictureFlow. But deal with that later. I've seen this before, I'll have to reboot to get song playing workign again.
06:54:37 Quit webguest412 (Quit: CGI:IRC (EOF))
06:55:13[Saint]you can reproduce it?
06:55:17 Quit fdinel (Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org)
06:56:12*JdGordon| wonders what happened with seani's new vkeyboard? :(
06:57:05mulenmar[Saint] Not reliably, no. Like I said, I've seen it before I and I have NO idea what triggers it, even from what I was doing.
06:57:17mulenmarI can do the same thing again and it doesn't cause a problem.
06:57:39***Saving seen data "./dancer.seen"
06:59:02mulenmarAlso, that skipping back and forth quickly in just the ordinary WPS −− no problem with the OGG Vorbis files I have on here.
06:59:43mulenmarEven in the album Pictureflow screwed up in and I then wasn't able to play anything until I rebooted.
06:59:47[Saint]I'm talking about *really* fast.
06:59:58[Saint]like...tapping "prev" as fast as you can.
07:00
07:00:02mulenmarSo was I. The first note usually didn't even get to play.
07:00:29[Saint]I can reproduce *PANIC* tagcache with that, every time.
07:00:31[Saint]weird.
07:00:47[Saint]I've narrowed down my pictureflow issues to one of two patches..though.
07:00:54mulenmarI'll try coping all the songs into one directory, hold on.
07:01:06[Saint]either my pictureflow patch, of liar's LCD sleep patch
07:02:47[Saint]mulenmar: do you notice that without the pictureflow patch, and during playback, that pictureflow is REALLY laggy changing between slides?
07:02:54[Saint]that is what my patch attempts to fix.
07:03:31mulenmarNo, I didn't, but again that's with only five albums.
07:04:08mulenmarHow many do you have, that you're having issues with?
07:04:36[Saint]It doesn't matter...two, three, or 50 run like a bag of ass.
07:04:43[Saint](without the patch)
07:04:50[Saint](and during playback)
07:05:24mulenmarI have the same, decent performance with our without the patch.
07:06:00mulenmar*or
07:06:25[Saint]very odd...it is a known bug that pictureflow is affected by the wps integration.
07:08:10mulenmarSo let me sure I'm doing it right.
07:08:37mulenmarI set the hotkey to Pictureflow, load a song, and while in the WPS I hit the hotkey for Pictureflow?
07:08:51mulenmarThen scroll between albums in Pictureflow?
07:09:05[Saint]yes.
07:09:39[Saint]you should see reduced performance as opposed to launching the plugin directly, without playback, from the plugins menu
07:09:57 Quit anewuser ()
07:10:26mulenmarWhile the song is playing? That makes sense to happen −− the CPU is decoding the music at the same time.
07:10:38mulenmarWhat format and bitrate are you using?
07:10:44[Saint]nah...it should handle it fine.
07:11:04[Saint]the CPU boosts as necessary, and, the Nano2G is certainly not short on CPU power
07:11:14[Saint]it is one of the more powerful RB targets
07:11:38[Saint]not only should it handle it fine, but, it used to...prior to .wps integration
07:11:57mulenmarWell OGG Vorbis is listed in the Wiki as having "very efficient code", so *maybe* it's making more of a difference, combined with the effects of Pictureflow, than seems logical.
07:13:18mulenmarJust tested the rapid back-forth in the WPS again, this time with 31 songs in one directory. Still no issue.
07:13:48[Saint]wow...I can reproduce that with all my targets...perhaps it has something to do with .mp3
07:13:59[Saint]I have 6 Nano2Gs, all do it.
07:16:45mulenmar[Saint] I detest MP3s, for quality/filesize issues as well as the last few patents that are still in effect (can't wait for 2015), but I'll download a few MP3 torrents from Jamendo and test with that for you.
07:17:31[Saint]mulenmar: have I showed you/have you seen the "iLike Nano" theme?
07:18:17mulenmarI've used it.
07:18:29[Saint]Ah, cool.
07:18:54[Saint]'tis mine.
07:19:06mulenmarI saw that. Nice work.
07:20:25[Saint]I need to backport some fixes into the version on the themesite, though...just syntax xleanup, and a few cleaner/new .bmps
07:20:45[Saint]*cleanup
07:22:46mulenmarUgh, there's the ONE thing that annoys me in both the OFW and Rockbox −− plugging into the computer stops playback. Technical limitation?
07:23:15JdGordon|nothing can ytouch the disk while usb is connected as mass storage
07:23:55mulenmarI thought the currently playing song was loaded into memory?
07:24:06mulenmarWait, nevermind, that doesn't make sense.
07:24:18CIA-81New commit by funman (r28002): manual: the AMSv1 don't run OF if powered by USB
07:24:43JdGordon|it is, but it is alot eaiser everywhere to make sure we kill playback than to stop writing/reading
07:26:01CIA-81r28002 build result: All green
07:36:32mulenmar[Saint]: The closest I can come to your problems is if I rebuild the Pictureflow cache while a song is playing, quit the plugin, and then reenter it with the hotkey.
07:37:36mulenmarThen it won't scroll unless I hit Menu first, and then it scrolls perfectly.
07:38:27mulenmarRephrase: if I tell Pictureflow to rebuild, exit, re-hotkey into Pictureflow, let it rebuild the cache, and then display it. THEN I can't scroll. But if I hit Menu, then it works.
07:39:17mulenmarI'm going to compare album scrolling in Picture flow with MP3 playing and OGG Vorbis playing now.
07:46:57*[Saint] waves to mulenmar from #rockbox too
07:52:17mulenmarWell, I'm not seeing a real difference in Pictureflow between with 192K VBR MP3 and Q7 OGG Vorbis playing in the background.
07:52:52[Saint]weird...things like this make me think there is some types of unknown hardware substitutions that Apple uses.
07:53:10mulenmarThe only thing is not scrolling between the albums, immediately after rebuilding the cache and quitting -> reentering the plugin.
07:53:10[Saint]but then, what are the odds of me getting 6 "different" players.
07:53:18[Saint]I can reproduce this on all my players
07:53:35mulenmar[Saint] If I remember right, TheSeven determined my Nano2G has a revision 7 board in it.
07:54:02[Saint]it ws liar, and it was a type 7 LCD
07:54:11[Saint]two LCD versions are known.
07:55:02mulenmarMy LCD as read by Debug -> View hardware: "type 1, (7) LDS176"
07:56:08mulenmarI don't know what the LCD differences are.
07:56:33[Saint]basically, nothing.
07:57:08[Saint]well, a few slight differences that don;t really matter, just need to be treated slightly differently.
07:58:22mulenmarWell, let me see if I can pull out what code we looked at in #freemyipod and found my Nano2G to be a "type 7 board"...
07:59:15mulenmarAh yes −− in NORFlash.bak, after "rVwH", there are some 0x00 and then a 0x07.
07:59:53mulenmarIs there a hardware difference between my version and yours that could be making me not have the problem?
08:00
08:01:21[Saint]I doubt it...I just need to track down someone else that *is* having the pictureflow issues...there's about 3 or 4 people on the tracker
08:03:59 Join solexx_ [0] (~jrschulz@e176113027.adsl.alicedsl.de)
08:07:25 Quit solexx (Ping timeout: 258 seconds)
08:12:40funmanhalving pclk/unboosted fclk on clipv1 has an impressive effect on codecs
08:12:54funmanAAC is almost twice slower
08:12:59funmanvorbis os more than twice slower
08:15:14funmanand no luck using DRAM@62MHz and PCLK@31MHz
08:15:58 Quit bunnyboi (Quit: I was raided by the FBI and all I got to keep was this lousy quit message!)
08:20:05 Quit CaptainKwel (Ping timeout: 276 seconds)
08:33:42[Saint]Hmmmm...is this a bug?
08:34:25[Saint]If I start playback with the database, then open pictureflow with hotkey, then exit pictureflow, then press select, it opens the pictureflow instad of the database
08:37:01[Saint]"Start playback with the database, then open pictureflow with hotkey in the .wps, then exit pictureflow back to the .wps, then press select (while still in the .wps), it opens the pictureflow instead of the database (it usually opens the database, if it launched playback, or the File Browsers, if playback was launched from there)
08:39:04 Join Zagor [0] (~bjst@rockbox/developer/Zagor)
08:39:25 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
08:45:27mulenmar[Saint] I (obviously) don't know if it's a bug, but the same thing happens here when I try it.
08:46:30JdGordon|that is probably a bug
08:46:41[Saint]I suspect it is a bug...IIUC select is supposed to return you to the list that playback was launched from.
08:46:51JdGordon|select should go back to where you started the music from, which isnt pf if bah....
08:47:13[Saint]and, in this case, it is *not* pictureflow that launches playback...but it takes over.
08:49:07JdGordon|any opinions on http://www.rockbox.org/tracker/task/11084 ? I'm probably going to be commiting it soonish... it is pretty much required for a useable touch sbs
08:49:57JdGordon|and I'll fix it so the hotkey button can control it also on non touch targets
08:53:20mulenmarWell [Saint], all I can offer is that fact that it happens here on my Nano2G too −− I'm heading to bed. I'll check Flyspray tomorrow and test whatever you come up with to fix it, if you do. 8~J
08:53:58JdGordon|my plan for touch sbs is having a title bar ar the top (centered) with the rb logo in one corner which always gets back to the menu, ui viewport taking up the rest of the screen except a row at the bottom which will be used for buttons.. if playback is running it will show track info, otherwise just the button bar
08:54:29JdGordon|the button bar will need a button which will change which set of buttons is displayed (playback control, browser control, etc) depending on the screen
08:55:22JdGordon|the lists annoyingly cant use the popup system because the list will get redrawn over it, so we have to use some screen space permenantly fo rit
08:55:25JdGordon|for it*
08:55:54JdGordon|^ portait layout.. landscape will put the buttons on the side somewhere
08:56:22*JdGordon| will need some help with icons :)
08:56:48[Saint]s/some help/someone to do them for him/ ;)
08:57:19JdGordon|you've seen how good i am in the art department
08:57:42***Saving seen data "./dancer.seen"
08:58:14[Saint]Hmmmm...I don't even know where to START looking to figure out why pictureflow is stealing the "return to <blah>" function from select if it is launched after playback commences.
08:58:46[Saint]can anyone try this on a non-ipod?
08:59:03JdGordon|it will be because the hotkey handler is a bit daft :)
08:59:13[Saint]or...any target, to confirm it exists throughout and not just some odd Nano2G bug
08:59:47[Saint]CsSavery kinda dissappeared after dropping that commit on us.
09:00
09:01:44 Join ender` [0] (krneki@foo.eternallybored.org)
09:02:00JdGordon|[Saint]: whats the story with your 24px iconset? did it go in?
09:02:18[Saint]Nope.
09:02:52[Saint]I guess I need to finish that with a viewers iconset...but, I didn't seee there was much interest in it going in from kugel
09:03:02[Saint]he seemed to dislike it rather strongly
09:03:21JdGordon|meh, sdl-app is my media player now so I'm interested in getting it up to scratch
09:03:44[Saint]I'll make a viewers iconset for it also, as soon as I can.
09:04:05[Saint]I'll probably do that tonight as it is relatively small.
09:04:14[Saint]*then* it can go in, sure.
09:04:51JdGordon|how is that smaller than the touch popup image? :'(
09:05:55[Saint]as it's just a matter of resizing/fixing some things...possibly creating a few icons from scratch, I'm not too familiar with the viewers icons.
09:06:06[Saint]I don't even use icons at all.
09:06:13JdGordon|wouldnt that make it more effort?
09:06:45[Saint]Nah...Tango! already exists, I'll need to get inventive and create something from scratch for the popup.
09:07:21[Saint]*anyway*, and idea where there's no keyclicks on Nano2G unless there's audio playback?
09:07:32[Saint]meaning to mention that for a while...bugs me.
09:07:52JdGordon|piezo or software clicks?
09:08:05[Saint]keyclick only works if the is current playback...it never used to be that way.
09:08:11[Saint]software.
09:08:49JdGordon|keyclick could have been broken... it should be the same as voice... i.e playback working or STOPPED... paused will give nothing
09:10:18funman[Saint]: if it didn't used to be that way you can bisect and find the commit which caused this
09:10:27 Join Gatz85 [0] (~add9ec48@giant.haxx.se)
09:10:39 Join lashmikhalich [0] (~lashmikha@83.69.28.123)
09:10:57lashmikhalichhi
09:11:05 Join LinusN [0] (~linus@rockbox/developer/LinusN)
09:11:08 Part lashmikhalich
09:11:10Gatz85quick question any one know what part of the code handles the Mouse HID when pluged in
09:11:34funmanGatz85: look in firmware/usbstack/
09:11:49JdGordon|depends what you are actually asking about...
09:11:58JdGordon|which button does what? how it works?
09:12:53Gatz85na more along the lines where it is triggernd and you can select the keyboard presation or mose
09:12:53 Quit [Saint] (Read error: Connection reset by peer)
09:13:10Gatz85umm the remote code if that help
09:14:12Gatz85becuase i havee found in the usbstack but the code that starts that process by selection
09:15:29Gatz85is what i'm shooting for
09:17:41JdGordon|what do you want to do?
09:18:42JdGordon|the code to handle the buttons is in apps/gui/usb_screen.c i tihnk or apps/misc.c?
09:18:54 Join Rob2223 [0] (~Miranda@p4FDCA9DB.dip.t-dialin.net)
09:22:07Gatz85I'm trying to port the ps3groove , and have a selection chocie for the jailbreak and when selected it.
09:22:21 Quit bzed (Ping timeout: 260 seconds)
09:22:34 Join efyx [0] (~efyx@lap34-1-82-225-185-146.fbx.proxad.net)
09:22:45 Quit Rob2222 (Ping timeout: 272 seconds)
09:23:01gevaertsYes, you're right. The user interface is the hardest part of that, so that's where you have to start work on it
09:24:09 Join bzed [0] (~bzed@devel.recluse.de)
09:24:12Gatz85getting it think its a hub might be a problem
09:25:27Gatz85but i'm doing research right now trying to decode what was done on the teensy board and try to aply that to the usbstack of rockbox
09:26:07JdGordon|even if you are successful, we very likely wont accept any patches which are purely for piracy
09:27:10Gatz85well first don't want pircy useing the orginal code that did not let you run the backup manger
09:27:12 Quit frankS2 (Quit: leaving)
09:27:50gevaertsOr more likely, we won't accept patches with a very questionable benefit for us if they make our code less clean (and I don't see how you can do the stuff involved *cleanly*)
09:28:03 Quit bertrik (Ping timeout: 245 seconds)
09:29:10Gatz85well didn't expect it to get patched in i'm really just trying to gut rockbox and make just a jig out of it
09:29:20gevaertsSure. You're welcome to do that
09:29:36funmanTheSeven: ping
09:29:39gevaertsI just want to make sure we all know where we stand, to avoid disappointment later :)
09:29:56Gatz85yeah i know
09:30:11funmanTheSeven: what's the "DCFG = 4; /* Address 0 */" in usb-s3c6400x.c ? address should start on bit 4
09:31:07funmanbit 2 is "Non Zero Length Status OUT Handshake" i think
09:32:49gevaertsGatz85: my guess is that you basically have two options: either build a hub class implementation, and extend usb_core.c to handle multiple device "contexts" (and possibly the actual controller driver as well, although I'm not sure if you need many changes there), *or* rip apart usb_core.c and just have it "expect request A, reply with a', then expect B, reply with b', ..."
09:33:13gevaertsChanging the device address as appropriate
09:35:34Gatz85first option would be need becuase want to be able to flash new firmware as need
09:36:39Gatz85thanks for the suggestions though they were helpful
09:37:07JdGordon|seen as this is obviously not rockbox related (and very questionable to begin with) please take the discussion elsewhere...
09:38:22funman[21430.190080] usb 1-3: can't set config #1, error -110
09:38:50Gatz85well i came here asking about the firmware and nothing else but ok thanks again
09:39:05funmangevaerts: would you know in which state the device is when Linux issue such an error ? (-110 = ETIMEDOUT)
09:39:34gevaertsfunman: it's in a "not responding as expected" state
09:40:07funmandoes it mean some packets went through? (i.e. get descriptor or set address)
09:40:24gevaertsNo
09:40:36gevaertsYou can get that after any request I think
09:41:04funmani'm trying to use the ipod nano2g driver on clip+
09:41:48 Join petur [0] (d408b802@rockbox/developer/petur)
09:42:45funmanBus 001 Device 049: ID 0781:74d1 SanDisk Corp. Sansa Clip+ (msc)
09:42:58 Join einhirn [0] (~Miranda@bsod.rz.tu-clausthal.de)
09:43:25gevaertsIf you get that, at least one packet got through :)
09:43:48funmanmore than one ;)
09:43:54funmanthat's with charging only btw
09:44:45gevaertsright, so enumeration only
09:58:51Gatz85are the config for each mp3 player defined in /exporrt/config/**
09:59:16JdGordon|yes
09:59:28Gatz85thanks
10:00
10:00:03 Quit sasquatch (Quit: WeeChat 0.3.2)
10:00:27 Join sasquatch [0] (~username@p4FF2CAB1.dip.t-dialin.net)
10:02:25funmangevaerts: only EP0 would be needed for charging-only enumeration, right?
10:02:40gevaertsOnly EP0 is needed for any enumeration
10:03:19gevaertsYou declare that the others exist during enumeration, but nothing is actually done with them
10:13:25 Part LinusN
10:16:26 Nick shai_ is now known as shai (~Shai@l192-117-110-233.cable.actcom.net.il)
10:29:43 Quit Gatz85 (Quit: CGI:IRC (EOF))
10:31:15 Quit JdGordon| (Quit: leaving)
10:31:41 Join brime [0] (~brime@207-118-99-101.stat.centurytel.net)
10:39:28funmani suppose pamaury will see better than me what fails with the usb analyzer
10:39:39 Join LinusN [0] (~linus@rockbox/developer/LinusN)
10:52:26 Quit xavieran_ (Remote host closed the connection)
10:57:43***Saving seen data "./dancer.seen"
11:00
11:00:21 Join xavieran [0] (~xavieran@ppp118-209-21-172.lns20.mel4.internode.on.net)
11:09:30 Join swilde [0] (~wilde@aktaia.intevation.org)
11:13:18funmangevaerts: any random guess why reading device descriptor & config descriptor would work but not setting it?
11:13:45gevaertssettin "it"?
11:14:17funmansetting the configuration*
11:14:36funmaniiuc the device sends all its possible configurations and the host selects one
11:14:37gevaertssettting the address works?
11:15:26funmanhow can i know? i expected that the address be set before reading the device descriptor
11:15:49gevaertsYes and no
11:17:00*gevaerts looks for things
11:17:36funmani suppose the clip+ is stuck in some endless loop since when charging only the screen is frozen
11:18:13gevaertsNormally, first the device descriptor is read, then the address is set, then the device descriptor is read again (possibly more of it), then the configuration descriptor is read
11:19:48gevaertsThen the configuration descriptor is read again (the first time only reads a bit, mainly to get the "length" field), then string descriptors are read, and finally the configuration is set (a single number actually)
11:20:03 Join Judas_PhD [0] (~kevin@misterfluffy.dsl.xmission.com)
11:20:57funmanhm the device descriptor can't be read in charging-only now
11:21:31 Part brime ("Leaving")
11:22:09funmanhttp://pastie.org/1141108
11:23:50gevaertsfunman: I assume this is consistently repeatable?
11:24:05 Join mischasworld [0] (~quassel@f051108081.adsl.alicedsl.de)
11:24:20funmanyes
11:24:47funmanas long as i don't mess with the register bits
11:25:28gevaertsWeird. I seems to fail at getting the device qualifier descriptor, but that really isn't anything special
11:25:50funman"can't set config #1" happens before i run lsusb
11:25:58funmanthe following messages, after i run lsusb
11:27:05gevaertsAs far as I know, lsusb gets the device and configuration descriptors itself, so that apparently still works
11:28:24 Join bmbl [0] (~Miranda@unaffiliated/bmbl)
11:28:48funmanon nano2g the rx/tx fifos are set to 512 bytes but in the linux (and pamaury's) code i don't see them being set.
11:28:49gevaertsfunman: does wireshark tell you something useful?
11:34:18funmanwireshark causes different errors?
11:34:47funmanGET DESCRIPTOR Response CONFIGURATION[Malformed Packet]
11:35:33funmanhm i've seen a SCSI device appear & disappear!
11:39:30funmanit tells me abotu GE DESCRIPTOR Response STRING[Malformed Packet]
11:40:44 Join Luca_S [0] (~5d3fc54b@giant.haxx.se)
11:40:58funmanafaics it's always the same (malformed) response
11:41:34gevaertsok, so the string descriptors might be wrong
11:41:56funmanhttp://pastie.org/1141130
11:42:27funmanthere's 3 successfull request/response before those 2 failing
11:43:03gevaertsThat's the serial number I guess
11:45:07gevaertsfunman: is there an appropriate set_serial_descriptor() in usb_core.c for your target?
11:45:26funmanyes
11:46:16funmanhmm maybe not
11:47:02funmanlooks like the serial number is twice smaller on as3543 than on as3514 so we might be reading undefined i2c registers
11:48:00gevaertsright, that might cause some weirdness
11:48:20gevaertsMake sure usb_string_iSerial.bLength remains a plausible value
11:48:53gevaertsOr just make it fall through to the default no serial number case for now
11:51:28funmanconfig is ok but lsusb still fails
11:51:52funmanthe failing response is always the same (the last one in my paste)
11:53:26funmancool, i saw SCSI requests again
11:53:44funmanit might be something with the caches (seeing the randomness of the failing point)
11:55:10 Join MethoS- [0] (~clemens@134.102.106.250)
12:00
12:00:23funman[29949.753320] scsi 101:0:0:0: Direct-Access Rockbox Internal Storage 0.00 PQ: 0 ANSI: 4
12:00:26funman:))
12:00:37gevaerts\☺/
12:00:42 Join user890104 [0] (Venci@Venci-Notebook-LAN.ipv6.6bez10.info)
12:00:48 Join wg412 [0] (~cf664046@giant.haxx.se)
12:01:21funmaniiuc there is a reset and after the reset things go bad again
12:04:10wg412soap: last night you made a couple of suggestions to help me conserve power on my newly minted "rockpod". I found the option to turn of accessory power, and a handful of options to modify the behaviour of the backlight but nothing allowing me to have the actual screen turn off when the backlight goes out. Can you help find this option?
12:04:50funmanhm there's always this weird packet
12:08:08 Quit krazykit (Quit: Connection reset by beer)
12:12:27 Join [Saint] [0] (S_a_i_n_t@203.184.3.248)
12:19:06Tornewg412: see the manual, specifically http://download.rockbox.org/daily/manual/rockbox-ipodvideo/rockbox-buildch8.html#x11-1430008.4
12:19:12 Join kugel [0] (~kugel@rockbox/developer/kugel)
12:19:26kugelfunman: success?
12:19:29funmanhttp://pastie.org/1141187 < my diff
12:19:37funmankugel: nope, but i think we're close
12:19:48Tornefolks in general: should we maybe amend the text in the manual for that setting? it says "but turning on the display takes noticably longer than just turning on the backlight" - that's not true on every device which can turn the lcd off
12:19:59funmanperhaps pamaury will have a clue what this weird packet is
12:20:00Tornethe ipodvideo's lcd turns back on instantly, or at least instantly enough for my reaction time
12:20:20funmanor perhaps we need to change a register between nano2g & amsv2
12:21:46pixelmawg412: which Ipod o you have exactly?
12:21:55pixelma*do you
12:23:15funmankugel: you want to try on fuzev2? i lended mine to a friend
12:23:35kugelsure
12:26:07funmanhmm it works on clipv2
12:27:24funmanalthough slowly but no bug yet
12:29:01funman*very* slowly even
12:29:52funmanthere might be something specific to the clip+
12:32:21kugelhttp://pastie.org/1141206
12:33:09kugeloh that's from the OF not rockbox
12:33:37kugelso, nothing seems to happen here (fuzev2 only charges)
12:35:40funmandid you enable USE_ROCKBOX_USB in config/sansafuzev2.h ?
12:36:18funmancopying from the device : 1Mbpm (per minute)
12:36:45JdGordon1kugel: did you see my email?
12:36:54kugelyes
12:37:42kugelfunman: no, I didn't
12:37:56JdGordon1do you know why dir_uncached.h is all in !APPLICATION?
12:40:44 Quit wg412 (Quit: CGI:IRC (Ping timeout))
12:41:47kugelfunman: better: http://pastie.org/1141216
12:42:50funmannice
12:43:03funmannot sure why the µSD isn't here though
12:43:50funmankugel: does it work fine? (i have a panic on usb unplug)
12:44:57JdGordon1sdl app sim spits out LOTS of warnings....
12:45:10JdGordon1and fails to compile :/
12:45:35kugelfunman: yes, I have that panic but it seems to have transferred a 20MB file just fine (and even quite quickly)
12:46:26funmanUSB_HAS_BULK was not defined but mass storage worked anyway :o
12:46:33 Join wg412 [0] (~cf664045@giant.haxx.se)
12:47:27kugel20MB in 3.1 seconds
12:48:11kugelis that realistic or does OS caching mess up the numbers?
12:48:43gevaertsYes (to both)
12:49:29funmanHID doesn't work out of the box though (like on nano2g, which is a good thing)
12:49:38gevaerts6MB/second seems realistic, and OS caching usually does mess up the numbers
12:50:48funmani have around 6mbps too
12:51:51kugel53MB in 9.5s, so it looks real :)
12:52:21kugelmd5sum matches
12:52:48kugelfunman: didn't you say it's slow for you?
12:52:58funmanyes but now it's fast, go explain why ..
12:57:15 Quit wg412 (Quit: CGI:IRC (EOF))
12:57:45***Saving seen data "./dancer.seen"
12:59:19JdGordon1where are the cabbie backdrop images stored? not the one for the wps
12:59:49JdGordon1ah, backdrops :/
12:59:50JdGordon1how
12:59:55JdGordon1how'd I miss that :p
13:00
13:02:14 Quit mischasworld (Ping timeout: 245 seconds)
13:04:28 Join mischasworld [0] (~quassel@f051108081.adsl.alicedsl.de)
13:04:53funmani give up on clip+, that can comes later
13:05:11TheSevenfunman: HID worked out of the box on nano2g, it's just that it causes random weirdnesses when hid and storage packets meet each other :)
13:05:24funmanTheSeven: ah that's similar to what i've seen
13:05:48funmanTheSeven: did you try to use *_dcache_range() functions instead of invalidating the whole cache in usb-s3c6400x.c ?
13:05:48UnhelpfulJdGordon1: progressive. :)
13:05:52TheSevenbut this really usually worked and just failed if you're sending hid events while reading files
13:06:05JdGordon1:) thanks
13:06:21TheSevenfunman: the nano2g doesn't have an MMU, and a rather trivial MPU
13:06:32funmanooh ok
13:06:43funmani thought it was an arm926ej-s but it's an arm940t, right?
13:06:47TheSevenyep
13:06:50TheSevenwe could use some UNCACHED_ADDR magic though
13:06:58*TheSeven doesn't remember why he didn't do that
13:07:03funmanfor ctrlreq
13:07:33funmani plan to work on this file soon (after some rest) to support as3525v2, so unless you beat me to it i can do it
13:07:38TheSevenyeah, probably because I don't have control over all the buffers passed to it
13:07:48JdGordon1kugel: does sdl-app *not* look in /usr/local/share/rockbox/wps for user wps/sbs files?
13:08:27TheSevenif we use uncached_addr, we would need to make sure that *every* access to that buffer (except dma at least) uses the uncached_addr
13:08:41TheSevenso this affects about all USB code
13:08:58funmanyeah it's only easy to do for ctrlreq, not for the other buffers
13:09:21kugelJdGordon1: it does but only if you have no wpses in $HOME/.config/rockbox.org
13:09:58JdGordon1I dont...
13:10:12JdGordon1why isnt my sbs loading? :(
13:20:30 Join wg412 [0] (~cf664048@giant.haxx.se)
13:22:27JdGordon1any guess for a font which will fit nicely inside the cabbie top bar thingy?
13:24:37wg412sorry, got disconnected a couple of times. my ipod is a 5.5gen 80gb. I already had a look at http://download.rockbox.org/daily/manual/rockbox-ipodvideo/rockbox-buildch8.html#x11-1430008.4 but I just don't see anything there to turn off the actual lcd when the backlight goies out
13:24:52 Quit funman (Quit: free(random());)
13:25:05wg412s/goies/goes/
13:26:53pixelmaI'd guess the "Sleep (After Backlight Off)
13:26:56pixelma"
13:27:43JdGordon1http://imagebin.org/112852 :/
13:28:00wg412pixelma: but isn't that for the device to go to sleep, ie shutdown x seconds after backlight off?
13:28:33pixelmathe description says "This setting controls how long rockbox will wait before turning off the display after the backlight is turned off. Turning off the display saves battery power but turning on the display takes noticeably longer than just turning on the backlight. "
13:28:33wg412guess I'll fart around with it, not like i'll break it i guess<
13:30:28CIA-81New commit by kugel (r28003): Check for presence of the sdk and ndk (the corresponding env variables to be exact) in configure.
13:31:28kugelJdGordon1: seems to be a mistake of mine
13:31:43wg412got it
13:31:45wg412thanks
13:32:13CIA-81r28003 build result: All green
13:32:15JdGordon1kugel: :) which? the compile break or the themes?
13:32:23kugelthe first
13:32:33pixelmawg412: the device will not shutdown after a backlight off - wouldn't make much sense to make that dependent on backlight off. There is the "Idle Shutdown" which will turn the device off after the set amount of time if it's doing nothing, not playing music, which is not in the "Display" submenu though - and there is the "Sleep Timer" where you can tell Rockbox to turn the device off after a set amount of time but that's not a persistent setting
13:32:41JdGordon1sweet
13:32:52kugelbrowsing wps/sbs seems to work fine for me
13:36:05CIA-81New commit by kugel (r28004): Fix sdl application build
13:37:55CIA-81r28004 build result: All green
13:39:27JdGordon1bieber: ping?
13:39:59JdGordon1kugel: if you can read the android keyboard I have a rather patch which you can make work.... I'll commit it soon enough for sdl
13:40:06JdGordon1jump to letters in the lists
13:51:31 Quit JdGordon1 (Ping timeout: 240 seconds)
13:51:52 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
13:55:04 Join webguest68 [0] (~4f15eb05@giant.haxx.se)
13:55:11webguest68hi ther
13:55:22webguest68there*
13:55:58webguest68Have I ti follow this http://www.rockbox.org/wiki/WpsTokenizer guide to build a patch for adding a new token?
13:56:32JdGordonno, that is very out of date
13:56:36JdGordonwhat do you want to add?
13:56:57webguest68the purpose is to have a token and a touch region for the onda's players, to enable the internal speaker
13:57:06 Quit user890104 ()
13:57:45webguest68currently available from the audio settings menu
13:58:13JdGordonyou need to look at parse_touchregion() in apps/gui/skin_engine.c to make the touchregion work
13:58:31JdGordonthen apps/gui/wps.c to make the button actually do something
13:59:37webguest68there's no skin_engine.c
13:59:58JdGordonapps/gui/skin_engine/skin_parser.c sorry
14:00
14:00:17kugelmaybe a generic setting toggle touchregion would be nice
14:08:19 Join dfkt [0] (dfkt@unaffiliated/dfkt)
14:13:31 Quit kugel (Ping timeout: 240 seconds)
14:15:09 Join krabador [0] (~krabador@host137-179-dynamic.3-87-r.retail.telecomitalia.it)
14:18:09 Join panni_ [0] (hannes@ip-178-203-81-220.unitymediagroup.de)
14:18:19 Join robin0800 [0] (~robin0800@cpc2-brig8-0-0-cust964.3-3.cable.virginmedia.com)
14:23:07webguest68so, in skin_parser.c under ifdev HAVE_ToUCHSCREEN ?
14:23:44 Join DerPapst [0] (~Alexander@91-64-225-76-dynip.superkabel.de)
14:25:16 Quit Luca_S (Quit: CGI:IRC)
14:26:00 Quit wg412 (Quit: CGI:IRC (Ping timeout))
14:27:24 Quit JdGordon (Ping timeout: 276 seconds)
14:27:29CIA-81New commit by jdgordon (r28005): The glyph number is required if it is given (i.e you cantt use -)
14:27:39 Quit mischasworld (Remote host closed the connection)
14:27:41CIA-81New commit by jdgordon (r28006): Add a work-in-progress 320x400 touch enabled sbs. ...
14:29:18CIA-81r28005 build result: All green
14:30:08CIA-81New commit by jdgordon (r28007): woops, wrong dimensions
14:30:18webguest68and in wps.c add #ifdev HAVE_SPEAKER?
14:30:38webguest68sorry, I can't understand very well how it's working....
14:30:58CIA-81New commit by jdgordon (r28008): 3rd time lucky
14:31:06CIA-81r28006 build result: All green
14:32:54CIA-81r28008 build result: All green
14:33:16 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
14:33:46JdGordonwebguest68: yeah, its a bit complicated. you need to add a new ACTION_ to action.h, then add that case in skintouch_to_wps() in wps.c
14:33:58JdGordonand make the touchregion use that ACTION_ value in skin_parser.h
14:34:01JdGordon.c
14:34:24webguest68ok...I will try to look into it
14:34:25JdGordonlook how the ACTION_HOTKEY is done... close to that
14:36:35webguest68ACTION_WPS_SPEAKER,
14:37:24webguest68do i put an ifdev HAVE_SPEAKER in action.h?
14:40:20 Nick fxb__ is now known as fxb (~felixbrun@h1252615.stratoserver.net)
14:42:55 Join mindfreakonline [0] (~53e630cf@giant.haxx.se)
14:43:30mindfreakonline:
14:43:36 Quit mindfreakonline (Client Quit)
14:44:24 Join mindfreakonline [0] (53e630cf@gateway/web/freenode/ip.83.230.48.207)
14:45:27JdGordonwebguest68: you dont need to, but it might be better for completeness
14:46:56webguest68and also I need to addd ACTION_STD_SPEAKER and ACTION_WPS_Speaker?
14:47:06mindfreakonlineHey Guys (: Is there somwhere out a guide how to code plugins? I'm currently looking into the rockbox SVN and trying to understand the code - but a guide or something similiar would be also nice.
14:47:37JdGordonwebguest68: ACTION_STD_SPEAKER is good for both
14:47:49JdGordonmindfreakonline: not really...
14:47:59 Join Kohlrabi_ [0] (~kohlrabi@ks305220.kimsufi.com)
14:49:34 Quit Kohlrabi (Read error: Connection reset by peer)
14:49:44 Nick Kohlrabi_ is now known as Kohlrabi (~kohlrabi@ks305220.kimsufi.com)
14:50:32mindfreakonlineWhat's about a description of the rb-object?
14:51:26JdGordonapps/plugins.h has API
14:51:34JdGordonis the API even
14:52:04mindfreakonlineThank you - haven't looked into it until now.
14:52:11mindfreakonline(Sorry for my bad english)
14:53:39webguest68), case ACTION_WPS_SPEAKER:...
14:53:55webguest68ouch...sorry..
14:54:57webguest68then under wps.c , skintouch_to_wps..
14:55:13 Join fyrestorm [0] (~nnscript@cpe-68-173-233-99.nyc.res.rr.com)
14:55:49 Quit poopie75656 (Quit: Lost terminal)
14:57:46***Saving seen data "./dancer.seen"
15:00
15:01:39 Join kugel [0] (~kugel@rockbox/developer/kugel)
15:02:01webguest68JdGordon: I'm having some troubles...
15:02:12JdGordonwhat have you got?
15:02:48kugelmaybe it's time to add a secondary framebuffer for touchscreens
15:03:07JdGordonwhy?
15:03:09webguest68I'm not understanding what I'm writing... and also don't know what to add in wps.c and where
15:03:10kugelit could be used for pop ups, yesno screen, splashes
15:03:17webguest68must say that I'm not a coder
15:03:36webguest68not a good one...
15:04:52JdGordonhang on, I tihnk tis will just be easier if I whip up a quick patch and you can test it
15:05:54JdGordonand by that I mean a touch region to generally change any setting... need to figure out how this would work...
15:06:21webguest68ok, great
15:06:38webguest68this will be great
15:07:03 Join hebz0rl [0] (~hebz0rl@dslb-088-065-214-240.pools.arcor-ip.net)
15:12:33 Join giovanni [0] (~giovanni@host5-235-dynamic.21-79-r.retail.telecomitalia.it)
15:13:10 Quit webguest68 (Quit: CGI:IRC (EOF))
15:14:21 Quit DerPapst (Ping timeout: 255 seconds)
15:23:48 Join CaptainKwel [0] (~jason@207-38-215-126.c3-0.nyr-ubr1.nyr.ny.cable.rcn.com)
15:29:59 Join fdinel [0] (~Miranda@modemcable235.127-131-66.mc.videotron.ca)
15:30:01fdinel0cc
15:30:53mindfreakonline1dd
15:31:13 Quit mindfreakonline (Quit: Page closed)
15:33:53CIA-81New commit by jdgordon (r28009): 2 new touch region options... "settings_inc" and "settings_dec" which will increase or decrease most of the available settings. To use it put the ...
15:33:56JdGordonthat was nice and easy :)
15:35:38CIA-81r28009 build result: 0 errors, 10 warnings (jdgordon committed)
15:36:25CIA-81New commit by jdgordon (r28010): fix warning
15:37:29 Join anewuser [0] (anewuser@unaffiliated/anewuser)
15:38:16CIA-81r28010 build result: All green
15:39:03giovanniJdGordon: thank you!
15:39:10giovanniI will test it right now
15:39:16JdGordon:)
15:39:28JdGordonhopefully that explanation is good enough?
15:40:30 Join DerPapst [0] (~Alexander@91-64-225-76-dynip.superkabel.de)
15:41:12giovanniyes, so for the speake?
15:41:17giovannispeaker?
15:42:06JdGordonwhat is the setting name?
15:42:45giovannidon't know...
15:43:20 Quit hebz0rl (Quit: Ex-Chat)
15:43:44JdGordonyou need to find out :)
15:43:51JdGordonchange it and open the config.cfg file
15:44:41giovanninow it is not in the audio settings, simulator
15:45:20JdGordon... apparently it is speaker
15:45:40JdGordonso add %T(x,y,w,h, setting_inc, speaker) and fix the first 4
15:45:57giovanniok
15:46:12 Join Jaykay [0] (~chatzilla@p5DC570B0.dip.t-dialin.net)
15:46:46 Quit DerPapst (Quit: Leaving.)
15:48:48*JdGordon is somewhat surprised to see the sbs doesnt update track positions at all... and noone has said this before?
15:49:43JdGordonbieber: is there a problem with how the editor loads libskin_parser.a? it doesnt seem to pick up maybe tag_table.c changes after remaking?
15:50:33giovanniand it will be great also to have something similar to the repeat mode, so i can display different status image for this
15:51:00JdGordonyou can
15:51:11JdGordon%?St(speaker)<on>
15:52:13giovanniok
15:55:23giovannigreat!
15:55:40giovanniIt works!
15:55:44giovannithank you :D
15:57:10JdGordonwell of course it works! :)
15:58:07giovanni:D
16:00
16:02:42giovanniwill it work also for the fms?
16:02:53JdGordonyes
16:02:56JdGordonand sbs
16:02:58JdGordonwell, it should
16:04:43giovannigreat
16:05:10giovannii will update my full touch theme
16:05:28 Quit krabador (Ping timeout: 258 seconds)
16:08:50 Join Beta2K [0] (~Beta2K@d24-36-97-38.home1.cgocable.net)
16:14:13CIA-81New commit by gevaerts (r28011): Slightly reorganise the FAT driver to ensure that no more than one sector buffer is ever allocated on the stack simultaneously. Fixes FS #11433
16:14:23 Quit fyrestorm (Quit: lamers envy me like they envy bill g -- main boot xp, just the way it should be!)
16:15:51CIA-81r28011 build result: All green
16:16:27gevaertsThat's a fun delta...
16:19:28giovanniis there a precompiled theme editor to download? I've got problem with my netbook
16:20:00giovanni/usr/bin/ld: cannot find -lz
16:20:12giovanniI'm on Ubuntu
16:20:53Torneyou are probably missing the -dev package for zlib
16:21:06Tornehttp://packages.ubuntu.com/search?keywords=zlib1g-dev
16:21:10Tornethat one :)
16:22:17giovannithank you
16:24:05giovanniyes, it fixes the error :)
16:24:57 Join h3po [0] (~h3po@p4FDE81A9.dip0.t-ipconnect.de)
16:28:39 Quit anewuser ()
16:30:08 Quit h3po (Quit: Leaving.)
16:30:08 Join G4Oblivion [0] (~186e2dc1@giant.haxx.se)
16:30:55G4OblivionFS #11468 can be closed. It was about the nand crash which was just fixed.
16:31:54 Quit G4Oblivion (Client Quit)
16:44:18 Join jjmartin [0] (~jjmartin@c-98-203-170-50.hsd1.wa.comcast.net)
16:46:53jjmartini would like a player that will do video in good quality but cost under $100
16:54:00 Join [sko] [0] (~sko]@p57A9A891.dip0.t-ipconnect.de)
16:54:37 Part LinusN
16:57:47***Saving seen data "./dancer.seen"
16:58:10 Part DarkMalloc
17:00
17:04:25kugelpowermgmt.c is a complete mess
17:04:38kugelit's scares me away from implementing auto-exit on android
17:05:22jjmartinpastebin
17:05:24jjmartinplz
17:06:14 Part Zagor
17:06:59 Join FlynDice [0] (~FlynDice@64.134.138.71)
17:15:14 Join domonoky [0] (~Domonoky@rockbox/developer/domonoky)
17:19:08 Quit JdGordon (Ping timeout: 240 seconds)
17:20:01 Quit robin0800 (Remote host closed the connection)
17:22:42 Join advcomp2019__ [0] (~advcomp20@unaffiliated/advcomp2019)
17:23:23 Join anewuser [0] (anewuser@unaffiliated/anewuser)
17:23:30 Quit jjmartin (Quit: Leaving)
17:24:04 Join n1s [0] (~n1s@nl118-174-115.student.uu.se)
17:24:06 Quit n1s (Changing host)
17:24:06 Join n1s [0] (~n1s@rockbox/developer/n1s)
17:24:20 Quit advcomp2019_ (Ping timeout: 265 seconds)
17:27:54 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
17:30:57 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org)
17:33:23CIA-81New commit by kugel (r28012): Default to 35px adobe helvetica for big displays (height > 480).
17:35:17CIA-81r28012 build result: All green
17:38:49 Quit [sko] (Quit: Leaving.)
17:39:51gevaertsHas anyone tried to make a screenshot recently on an mr500?
17:40:14kugelI even doubt that anyone has recently used an mr500 :)
17:41:14gevaertss/recently/(in 640x480 configuration)/
17:41:31gevaertsI'm pretty sure it would get a stack overflow
17:41:36kugelIIUC it doesn't even have a cabbie port
17:42:18kugelgevaerts: don't you have one?
17:42:39gevaertspamaury has mine now
17:59:48 Quit giovanni (Quit: Sto andando via)
18:00
18:02:31 Join komputes [0] (~komputes@ubuntu/member/komputes)
18:04:31 Join toffe82 [0] (~chatzilla@adsl-71-132-80-154.dsl.sntc01.pacbell.net)
18:05:06 Quit l403 (Read error: Connection reset by peer)
18:09:00 Quit panni_ (Read error: Connection reset by peer)
18:09:16 Quit petur (Quit: Page closed)
18:09:44 Quit Kohlrabi (Remote host closed the connection)
18:10:37 Join Kohlrabi [0] (~kohlrabi@ks305220.kimsufi.com)
18:11:19pamauryDo you need do check something on the mr500 ?
18:12:29gevaertspamaury: yes. I'd like to know if screendumps work
18:12:36gevaertsI suspect they'll crash
18:12:40gevaertshm, wait
18:12:59gevaertsThere's no decent hard reset on the mr500, so that's not a nice thing to ask
18:13:00pamauryI can't check know, I'm still unpacking but tonight or tomorrow I'll try
18:13:12pamauryno hard resert ?
18:13:22pamauryhow do you restart ?
18:13:23pamauryhow do you restart ?
18:13:33gevaertsWell, there is, but it uses remote buttons, and I never had the remote
18:13:44gevaertsOpen the thing and disconnect the battery :)
18:14:01pamauryhuh, is it difficult to open ?
18:14:21gevaertsNot very. Four screws IIRC
18:14:35n1sis anyone still using the old vmware image? if so which version is its native gcc
18:14:37n1s?
18:14:40pamauryok, I'll check if I have a compatible screw driver before
18:14:44pamaury:D
18:15:52*pixelma points towards kkurbjun to get his attention ;)
18:28:16n1sif the vmware image can't build sims and is shipping with at least one out of date target toolchain, which just happens to be the one used for the majority of targets. Does it do more good than harm?
18:28:23 Quit Kohlrabi (Read error: Connection reset by peer)
18:29:19Tornesomeone could just make a new one or update it. ;)
18:29:49n1sbtw, as soon as i get a monitor for my desktop and some free time i'll try to get coldfire over to a newer gcc too so it would be even more out of date.
18:29:54n1sTorne: sure
18:30:54n1sactually jHmikeS (capitalization?) made one and posted in the forums, but i dunno the state of it (can it build manuals?)
18:35:25 Quit kugel (Remote host closed the connection)
18:36:36 Quit swilde (Remote host closed the connection)
18:38:03 Nick fxb is now known as fxb__ (~felixbrun@h1252615.stratoserver.net)
18:38:56bieberJdGordan: If you catch this later, I've never had an issue with it, but I remember you having the same problem a while ago. Perhaps try cleaning and rebuilding everything?
18:40:21 Quit lestatar (Max SendQ exceeded)
18:42:55 Join lestatar [0] (~chatzilla@cpe-72-229-41-214.nyc.res.rr.com)
18:50:51 Join oolongtea [0] (~1810a2d0@giant.haxx.se)
18:51:05 Join funman [0] (~fun@rockbox/developer/funman)
18:57:49***Saving seen data "./dancer.seen"
18:58:47 Join _s1gma [0] (~d.d.derp@77.107.164.131)
19:00
19:02:41funmanpamaury: did you see that we can mount clipv2/fuzev2 with rockbox usb?
19:06:17 Join ken147 [0] (~4846e546@giant.haxx.se)
19:06:54ken147hi everyone
19:07:05 Quit ken147 (Client Quit)
19:09:07 Quit advcomp2019__ (Quit: IRC is just multiplayer notepad)
19:09:37 Join advcomp2019 [0] (~advcomp20@unaffiliated/advcomp2019)
19:10:28 Quit oolongtea (Quit: CGI:IRC (EOF))
19:10:42 Join Jerom [0] (~heidi@95.171.131.249)
19:15:17 Join Kohlrabi [0] (~kohlrabi@ks305220.kimsufi.com)
19:16:47pamauryfunman: with my code ?
19:17:05funmanno with nano2g code
19:17:49 Quit antil33t (Ping timeout: 255 seconds)
19:17:57pamauryand lcip+
19:17:58pamaury?
19:18:01pamaury*clip
19:18:07 Quit [Saint] (Ping timeout: 272 seconds)
19:18:32funmanit somehow works but it looks like some responses on IN endpoint are replaced by a malformed packet (always the same)
19:19:46pamauryand what my code procudes on clipv2 and fuzev2 ? I can't test, I'm stuck with my clip+
19:20:44funmanafaics it was the same on clipv2
19:21:11 Join [Saint] [0] (S_a_i_n_t@203.184.3.248)
19:21:55 Join antil33t [0] (~Mudkips@124-197-51-80.callplus.net.nz)
19:21:57 Quit Judas_PhD (Quit: This is a quitting message)
19:22:13 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
19:22:25 Quit pamaury (Read error: Operation timed out)
19:23:25 Join pamaury [0] (~quassel@dhcp-129-136.residence.ens-lyon.fr)
19:23:25 Quit pamaury (Changing host)
19:23:25 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
19:24:10 Join [sko] [0] (~sko]@p57A9A891.dip0.t-ipconnect.de)
19:26:24pamaurythe clip+ must have something special but I can't explain the fundamental difference between nano2g code and mine that changes everything :) I should have more time soon, I'll try to solve this definitely
19:27:38funmanwell enumeration works so it must be something small
19:28:57pamaurymy next idea to try is a circular buffer for setup packet because all the problem is there, the setup packet is somehow corrupted or not written, or ...
19:30:37funmansetup works with nano2g code
19:30:52funmansometimes i see the /dev/sd* being created
19:31:03pamaurywith nano2g code ?
19:31:20funmanyes, read this morning logs
19:31:29pamaury"sometimes" ?
19:31:54amiconnn1s: *iirc* the vmware image is debian lenny x86, but I lost overview over the various images long ago...
19:31:59funmani see malformed responses in wireshark, but the malformed responses happen at various random times
19:32:11amiconnThere were several, and once made, nobody seems to ever update them
19:32:28funmani'm modifying the usb-s3c6400x.c so it works for both ams and nano2g (my current code is just a hack)
19:33:34 Join einhirn [0] (~Miranda@bsod.rz.tu-clausthal.de)
19:35:17pamauryok, I'll try with the modified code, or your patch if you have it, see if I can explain this weird packet. But I really would like to understand why my code doesn't work :) I don't have time now. If you have a patch (even a hacky one), poke me, so I check with the analyzer*
19:35:18 Quit einhirn (Client Quit)
19:36:21funmani pasted it around 12:00
19:38:27pamauryok
19:39:29funmanhm i think i understood something about the right -target.h file not being included
19:39:47funmanif the .c file is in arm/ gcc will pick up arm/x-target.h and not arm/target/x-target.h
19:43:54funmanTheSeven: what does DCFG=4; mean in usb-s3c6400x.c ?
19:44:03amiconnfunman: That shouldn't happen
19:44:56amiconn#include order is firmware/target/arch/manuf/device/ -> firmware/target/arch/manuf/ -> firmware/target/arch/
19:45:47funmanTheSeven: also how about renaming INT_USB_FUNC() to INT_USB() ?
19:46:01TheSevenFUNC in contrast to PHY
19:46:10 Join Luca_S [0] (~57064cf0@giant.haxx.se)
19:46:11TheSevenit should maybe be OTG instead
19:46:16 Join Horscht [0] (~Horscht@xbmc/user/horscht)
19:46:25TheSevenand regarding the 4 value... no idea
19:46:42TheSevenapple used that, and i have the impression that my docs aren't accurate on that one
19:48:47TheSeveni only care about the address bits of that reg :)
19:48:54funmancan you try setting it to 0 ?
19:52:19TheSevenseems to work fine at a first glance
19:54:23 Join krazykit [0] (~kkit@adsl-99-88-108-147.dsl.chcgil.sbcglobal.net)
19:55:36funmanah cool no more panics on unplug
19:56:44funmanalso, does the dummy address for empty packets need to be in the memory map?
19:57:13TheSeveni don't quite know why it is there in the first place
19:57:43TheSevenat apparently helped to prevent the usb core from spuriously overwriting the vector table
19:58:07TheSevenbut in theory it shouldn't be neccessary as it should never write anythin to that address in that case
19:58:48funmanok
19:58:57TheSevenIIRC it was liar who introduced that, and while it seemed to solve a problem on his iPod, it was working perfectly fine on mine even without that patch
19:59:15funmanand last thing: how are D?EPCTLx documented for you? what's bit 11 ?
19:59:50 Quit krazykit (Ping timeout: 240 seconds)
19:59:52funmanmax packet size is 11 bits here but your mask is 10 bits
20:00
20:00:18funmani mean what's bit 10
20:00:25 Quit liar (Read error: Connection reset by peer)
20:00:31 Join liar [0] (~liar@clnet-p09-185.ikbnet.co.at)
20:00:53 Join panni_ [0] (hannes@ip-178-203-81-220.unitymediagroup.de)
20:01:57TheSevenyes, the datasheet says bits 10:0
20:02:13funmanok so i'll change the mask
20:02:16 Quit ender` (Read error: Connection reset by peer)
20:02:35TheSeven9:0 will work because the maximum packet size that's ever possible is 1023
20:02:47TheSeven(for isochronous transfers)
20:03:40TheSevenmust have slipped in because of me thinking of usb packet sizes as 10 bit values
20:06:46funmanoh if i include with <> it works as expected
20:07:39 Join ender` [0] (krneki@foo.eternallybored.org)
20:09:09 Quit Kohlrabi (Remote host closed the connection)
20:09:52 Join Kohlrabi [0] (~kohlrabi@ks305220.kimsufi.com)
20:13:40funmanTheSeven: is there a macro to use uncached addresses on s5l8700 ?
20:13:53TheSevenUNCACHED_ADDR(x)
20:14:03funmanwhere is it defined?
20:14:19TheSevenhmm, i'd guess system-s5l8700.h or something
20:14:37TheSevenhowever, this will only work properly if every single access to that buffer goes through uncached_addr
20:14:51funmansuer
20:14:53funmansure*
20:15:36TheSevenbtw it is *not* needed (but won't hurt) to be applied to the address passed to the OTG
20:16:26funman/media/bordel/rockbox/firmware/target/arm/usb-s3c6400x.c:53: warning: implicit declaration of function ‘UNCACHED_ADDR’
20:16:51 Quit FlynDice (Remote host closed the connection)
20:17:26TheSevendoes that file include system.h?
20:17:55funmanyes
20:18:59funmani don't see UNCACHED_ADDR in s5l8700/sytem-target.h
20:19:33TheSevenhmm, ask grep...
20:19:40TheSeveni don't have a working copy around right now
20:19:42funmanit's definitely not there
20:19:52funmani'll use the same formula than pcm
20:19:57TheSeveni just know that it is *somewhere* :)
20:20:00funman+ addr = (void*)((uintptr_t)addr | 0x40000000);
20:20:03funmanit's not!
20:20:22TheSevenuncached_addr should be defined to be that formula
20:20:38TheSevenand IIRC it was used in the pcm driver one day
20:20:42TheSevendid someone eliminate it?
20:21:18funmanyou removed one occurence in r23308
20:22:14 Join bunnyboi [0] (~androgyne@cpe-72-224-17-58.nycap.res.rr.com)
20:22:45funmanok i have a patch
20:25:31 Join Lear [0] (chatzilla@rockbox/developer/lear)
20:25:40funmanFS #11607 - Enable USB on AMSv2 using Nano2g driver
20:26:21CIA-81New commit by funman (r28013): Fix set_serial_descriptor() for targets with as3543
20:26:25bunnyboioh that is awesome, i'll try it out asap
20:28:05CIA-81r28013 build result: All green
20:29:23 Quit Luca_S (Quit: CGI:IRC (EOF))
20:29:49funmanpamaury: 0000 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ........ ........
20:29:52funman0010 00 02 00 00 00 00 00 00 ........
20:30:01funmanthis is the malformed response which seems to happen at various times on clip+
20:30:40funmanwhen reading device descriptor, or reading config descriptor, or setting configuration, or when the host is sending SCSI commands ...
20:30:48TheSevenfunman: what's the status code of that one?
20:30:57funman'status code'?
20:31:31TheSevendid the transfer succeed from the chip's point of view, regardless of the data?
20:31:37funmanno idea
20:31:47funmani'm reading the packet in wireshark
20:32:00TheSevenyeah, that's showing a return code somewhere
20:32:04funmani suppose that yes since i dont' see duplicate packets
20:32:06TheSeventhe place where you see -ETIMEOUT sometimes
20:32:32funmanhttp://pastie.org/1141992 < sample log
20:33:59funmanURB status : -EOVERFLOW
20:34:10TheSevenaha
20:34:21TheSeventhe data shown might be just garbage in that case
20:34:29funmanit's nearly all zero
20:34:41TheSevenprobably just some uninitialized (host) memory
20:34:53TheSevennow what did an overflow error mean in usb terms?
20:35:04TheSevenpacket longer than requested?
20:35:17funmanhm another response says -ENOENT
20:36:04*funman spots fdinel in the logs
20:37:50funmani see this buffer in "__compound_literal.*" in rockbox.elf but i don't know where it is in the source
20:37:59funmanapps/settings_list.o
20:45:00funmanok for making clipv1 stable?
20:48:06CIA-81New commit by funman (r28014): Clipv1: stable status
20:49:37CIA-81r28014 build result: All green
20:50:04CIA-81New commit by funman (r28015): front page: move clipv1 to stable
20:50:38funmanif we get this USB patch committed & tested we could make the clipv2/fuzev2 stable (and clip+ if we can find the problem)
20:50:43 Join n00b81 [0] (~taylor@c-24-91-82-205.hsd1.ma.comcast.net)
20:50:58funmanonly bugs left is wakeup alarm sometimes buggy, and bad sound quality on clipv2
20:51:10 Join Freddy-PSP [0] (~alfredo@87.13.200.218)
20:52:28 Part n00b81 ("Cyaz")
20:52:47*TheSeven wonders when (or rather if at all) the nano2g will reach stable
20:53:18*funman injects 50cc of motivation to fix the last bugs in TheSeven's veins
20:53:23 Nick Freddy-PSP is now known as Freddy-PSPAWAY (~alfredo@87.13.200.218)
20:53:31TheSeventhe bugs are rather easy
20:53:34 Nick Freddy-PSPAWAY is now known as Freddy-PSP (~alfredo@87.13.200.218)
20:53:39TheSeventhe bigger problem is the bootloader dilemma
20:53:49TheSeveneasy installation/uninstallation vs. speed and stability
20:53:59 Nick Freddy-PSP is now known as CatchMyBalls (~alfredo@87.13.200.218)
20:54:01 Join Delbury9 [0] (185b52cd@gateway/web/freenode/ip.24.91.82.205)
20:54:01funmancan't embios have both?
20:54:08funman(embios-based tool)
20:54:10 Nick CatchMyBalls is now known as Freddy-PSP (~alfredo@87.13.200.218)
20:54:35 Quit Freddy-PSP (Quit: ab5000 pinged me)
20:54:38TheSevenas soon as we write to the nor flash, we will lose the "use itunes to restore" feature
20:55:00gevaertsTheSeven: is an iloader-based system very different from the coldfire irivers we support?
20:55:24funmani'd say we don't care about itunes if you document uninstallation
20:55:33TheSevenit's probably even safer, as a failure during the flashing process can't brick it
20:55:49funmanyou can write a CLI tool and integrate it in rbutil
20:56:03TheSevenwe can't do it directly from a PC
20:56:19 Join clone4crw [0] (~calvin@96-42-90-88.dhcp.roch.mn.charter.com)
20:56:34TheSevenwe need to install an installer to the firmware partition, which gets booted by the apple bootloader, and then reflashes the NOR
20:56:51gevaertsTo be honest, "harder to uninstall than other ipods" shouldn't mean anything if it's still better than other stable supported targets
20:57:18TheSevenhm, there are several possible ways to go:
20:57:51***Saving seen data "./dancer.seen"
20:58:04TheSeven- read the firmware partition in the nor-based bootloader and check for a flash update, apply it if we find one
20:58:17TheSeven(that would enable itunes to kick out our bootloader indirectly at least)
20:58:30clone4crwgevaerts: Hey I saw you uploaded an update to the resistor calculator. I actually am about to upload a new version myself, and will look at yours in a bit. I am pretty much done with the resistance to color codes part.
20:58:58Delbury9clone4crw: nice
20:59:07TheSeven- have it remove itself when it detects a problem
20:59:41TheSevenactually I'd like to install our bootloader in a way that's fully transparent to itunes, but kicked out by a restore
20:59:57TheSevenwhich means we need to:
21:00
21:00:05TheSeven- rename osos to osbk in ipodpatcher
21:00:12TheSeven- install the installer as osos
21:00:21TheSeven- have the installer reflash the nor and rename osbk to osos again
21:00:33funmanpamaury: 20:25 < funman> FS #11607 - Enable USB on AMSv2 using Nano2g driver
21:00:44TheSeven- have the bootloader check for a flash update and flash it if it found one
21:00:58gevaertsclone4crw: I changed a lot actually. I mainly meant it as a sort of example of how I would do it though, don't feel forced to keep any of those changes
21:01:34TheSeveni'd like to have a 2-stage bootloader for safety reasons
21:01:53TheSeventhe first stage being more or less a DFU stub, the second stage caring about booting and uninstallation
21:02:22TheSeventhe first stage would basically be embios loader
21:02:35TheSeventhe second stage would be a modified rockbox bootloader
21:02:41TheSevenwe would need to add the following features:
21:02:56TheSeven- scan the firmware partition for flash updates and flash them
21:03:06TheSeven- restore the FTL if it's unclean
21:03:29TheSevenand we would need to write an installer for it, which could possibly be based on the one i'm currently using for iloader
21:03:30 Join help-a-foo [0] (452a6708@gateway/web/freenode/ip.69.42.103.8)
21:03:34help-a-foohi
21:03:47help-a-fooquick compativbility question
21:03:59help-a-foodoes rockbox run on the Archos Studio 2
21:04:07help-a-fooor just the Archos Studio
21:04:11TheSevenis it listed on the front page?
21:04:18help-a-foowell, studio is listed
21:04:44help-a-fooand im not sure if the "studio 2" i am seeing is the same product or what
21:04:55bunnyboifunman, I just tried the patch on my Fuze V2. The internal and external disks show in My Computer, but they are empty. Doesn't show as raw, doesn't show as fat either. Just empty.
21:05:24help-a-fooim actually in Tiger Direct here in chicago on one of their kiosks, using webchat.freenode.net
21:05:39help-a-fooi notice they have used sansa fuze, and new archos studio - but the box has a 2 on the archos
21:05:42help-a-fooused = refurb
21:06:50help-a-fooits pretty cheap, $49 for the archos studio (2?) and $39 for the refurb sansa fuze
21:07:28 Quit Delbury9 ()
21:07:34help-a-fooi know for sure the fuze will work, but id rather buy something new if possible
21:07:49pixelmathe supported Archoses are quite old lines of models, I doubt you can get any of these in shops
21:08:41help-a-foopixelma: ya, but it does list studio on the wiki, plus this tiger direct has a bunch of older stuff
21:09:00help-a-fooreally its too bad one of those $20 MP3 playher manufacturers doesnt just use rockbox
21:09:02 Join Ramsey[LC]_ [0] (~RamseyLC]@adsl-70-234-236-78.dsl.rcsntx.sbcglobal.net)
21:09:12help-a-fooim surprised china hasnt just started putting rockbox on cheap players
21:09:22help-a-fooitd be better than the usual poorly dfesigned interface
21:09:34help-a-fooon those chinavasion.com players
21:09:58help-a-foooh well, this kiosk wont let me browse the rockbox.org in a separate window, so off i go! thanks for the help pixelma
21:10:09 Quit help-a-foo (Quit: Page closed)
21:12:20 Join fyrestorm [0] (~nnscript@cpe-68-173-233-99.nyc.res.rr.com)
21:15:03alexbobPhow do I get my rockbox sansafuze to play videos?
21:15:19alexbobPwhen I click a video in the file browser it just plays it's audio
21:15:31TheSevenwhat kind of video?
21:15:43alexbobPhmm... I'll check what codecs
21:16:02TheSevensee the wiki for instructions on how to encode videos for rockbox
21:16:04 Join kugel [0] (~kugel@rockbox/developer/kugel)
21:16:07alexbobPa lot of files don't even show up in the interface
21:16:22alexbobPTheSeven: oh... I thought rockbox would come with more codec support... do I really have to reencode all my crap?
21:16:35bunnyboiI have a bit of a broken bash script somehwere that'll do it automagically
21:16:39gevaertsalexbobP: no, only the good stuff
21:16:45TheSevenmore support for audio codecs, yes. not for video codecs.
21:16:45LloreanalexbobP: Rockbox is an *audio* player.
21:16:59TheSevenIIRC only a single video codec is supported through the mpegplayer plugin
21:17:12LloreanTheSeven: Well, MPEG-1 and MPEG-2 isn't it?
21:17:23alexbobPavc1 video and mp4a audio
21:17:30LloreanNot supported.
21:17:36alexbobPLlorean: aww
21:17:42alexbobPLlorean: is ther any way to install more video codecs on it?
21:17:49LloreanalexbobP: Yeah, write them.
21:18:00funmanbunnyboi: i didn't try on fuzev2, but kugel said it worked fine
21:18:03alexbobPI don't know how to do that XD
21:18:22LloreanalexbobP: The main problem is that the people interested in expanded video support don't know how to do it, so it never happens.
21:18:24funmanbunnyboi: i assume it's on windows?
21:18:25TheSevenand you probably won't manage to decode h.264 in realtime without hardware accelleration, which you would probably need to reverse engineer
21:18:38alexbobPsrsly, I'd happily port codecs to rockbox if it was within my sphere of competence, but it's probably not
21:18:41bunnyboifunman, yes i'm on windows right now. i can try it on linux though
21:18:41alexbobPLlorean: heh
21:18:44LloreanTheSeven: That depends on the player.
21:19:40alexbobPbunnyboi: can I please have your script?
21:20:02bunnyboialexbobP, you just put the files you want in the same directory and run the script: http://pastebin.com/fmJH74hq
21:20:06kugelamiconn: your #include order was correct, but "." is still the highest prority so the #include order doesn't always work from within the target tree
21:20:34bunnyboiyou also need mencoder
21:20:43alexbobPLlorean: funny thing is, I got rockbox specifically because I thought it would enable more video codecs than the fuze already played... didn't know that wasn't a design goal
21:20:58alexbobPLlorean: not that I'm not happy I rockboxed. all these awesome games and themes :O
21:21:06Tornenothing in the design prevents it playing video :)
21:21:08funmankugel: #include <> works fine
21:21:11alexbobPbunnyboi: thanks!
21:21:19Torneit's just noy something anyone has implemented
21:21:28Torneother than the mpegplayer plugin
21:21:54 Join hairyballs [0] (hairyballs@gateway/shell/bshellz.net/x-iiqyoktypuapllno)
21:22:07 Part hairyballs
21:22:08LloreanalexbobP: Well, Rockbox began on monochrome players. The original design focus was on expanded audio functionality. With a wider range of multimedia-capable players supported, expanded video functionality would probably be welcome, but not the reason most active developers became involved, so it's something likely to require new blood
21:22:16 Join n00b81 [0] (~taylor@c-24-91-82-205.hsd1.ma.comcast.net)
21:22:29 Join hairyballs [0] (hairyballs@gateway/shell/bshellz.net/x-iiqyoktypuapllno)
21:23:04CIA-81New commit by funman (r28016): debug-as3525.c : fix include, remove comment
21:23:19 Quit n00b81 (Quit: Cyaz)
21:23:50bunnyboifunman, It works fine on Debian. Everything shows.
21:24:23 Join cwilly [0] (cwilly@gateway/shell/bshellz.net/x-ppjwcvutwnktaejk)
21:24:29funmanbunnyboi: i'll try the clipv2 on windows vista later
21:24:42bunnyboiI'm running XP by the way, if that helps.
21:24:53CIA-81r28016 build result: All green
21:27:31 Join user890104 [0] (Venci@Venci-Notebook-LAN.ipv6.6bez10.info)
21:29:08 Nick hairyballs is now known as somethingbetter (hairyballs@gateway/shell/bshellz.net/x-iiqyoktypuapllno)
21:29:21amiconnfunman: If the bad sound quality on clipv2 has the same reason as on clip+, there is a patch that helps a lot
21:29:35*amiconn uses that for quite some time now on his clip+
21:30:58amiconnalexbobP: Regarding video codecs: Unlike with audio codecs, supporting more than one video codec doesn't make sense except on *very* powerful devices which can do realtime scaling
21:31:15cwillysomethingbetter: how's the 3g nano port?
21:31:20cwillyany progress?
21:31:49somethingbetteruhm...yes
21:32:00somethingbetternow i can see pr0nz 10 times faster
21:32:23amiconnFor all others, you need to prescale anyway so reencoding is no extra effort
21:32:43funmanbunnyboi: perhaps the logical volume manager (right click my computer -> manage iirc) can give some info?
21:32:55cwillysomethingbetter: good good
21:33:28TheSevengevaerts: to get back to the bootloader problem, what do you think is the best way?
21:34:11somethingbettercwilly: that's a good progress isn't it?
21:34:28cwillyYeah but you should really think about getting a kernel exploit.
21:34:46somethingbetteri haz one
21:34:47bunnyboifunman, Doesn't show up in Disk Management
21:34:58somethingbetteri found it in a pr0n image
21:35:25somethingbetterthe dick of the guy in the pic is too long,it causes a stack overflow
21:35:40somethingbetteri'm that guy btw
21:35:42gevaertssomethingbetter, cwilly: this *isn't* a random chat channel
21:35:54somethingbetterorly?
21:36:06somethingbetterare you serious?
21:36:08Mode"#rockbox +o Torne" by ChanServ (ChanServ@services.)
21:36:10Torneyes
21:36:22somethingbetteromai
21:36:26somethingbetteru gotta be kidding
21:36:42Mode"#rockbox +b *!*airyballs@gateway/shell/bshellz.net/x-iiqyoktypuapllno" by Torne (torne@rockbox/developer/Torne)
21:36:42Kick(#rockbox somethingbetter :somethingbetter) by Torne!torne@rockbox/developer/Torne
21:36:51Mode"#rockbox +b *!*cwilly@gateway/shell/bshellz.net/x-ppjwcvutwnktaejk" by Torne (torne@rockbox/developer/Torne)
21:36:51Kick(#rockbox cwilly :cwilly) by Torne!torne@rockbox/developer/Torne
21:37:28gevaertsTheSeven: actually, if you have a DFU stub that's reasonably easy to get to, I'm not sure if the second stage needs to be that fancy
21:38:21Mode"#rockbox +b *!*@gateway/shell/bshellz.net/*" by Torne (torne@rockbox/developer/Torne)
21:38:21WarnBanlist contains 18 bans!
21:38:27TheSeventhe problem with the dfu stub is that the average windows user will have to install special drivers to access it
21:38:46Mode"#rockbox -b *!*airyballs@gateway/shell/bshellz.net/x-iiqyoktypuapllno" by Torne (torne@rockbox/developer/Torne)
21:39:00funman[Saint]: can you try FS #11607 on your nano2g and see if USB still works fine?
21:39:04Mode"#rockbox -b *!*airyballs@gateway/shell/bshellz.net/x-iiqyoktypuapllno" by Torne (torne@rockbox/developer/Torne)
21:39:05gevaertsok, I think I'm beginning to see the full picture
21:39:17Torneoops
21:39:23gevaertsThis "bootloader" can load the Apple in-ROM disk mode?
21:39:34TheSevennot the dfu stub
21:39:38TheSeventhe second stage can do it
21:39:51gevaertsyes, quotes because I actually mean the whole lot :)
21:40:03Mode"#rockbox -b *!*cwilly@gateway/shell/bshellz.net/x-ppjwcvutwnktaejk" by Torne (torne@rockbox/developer/Torne)
21:40:08TheSevenspeaking of iloader, it can do it
21:40:16TheSeventhe rockbox bootloader would work similarly
21:40:39gevaertsok, so the more or less standard recovery would be to write a new image (or recognise a new itunes image) to disk mode, and if needed reflash from there
21:40:51gevaertsWill the DFU stub always remain in place?
21:40:55kugelif that driver is going to be shared we should maybe add a usb_otg subdir in drivers (like with audio and rtc), it's not the first usb driver to be shared IIRC
21:41:11bertrikamiconn, are you talking about the patch that changes some things in the codec of the clip+ (bias and "hi-q mode" IIRC)? I thought dfkt tested that and didn't find a major difference
21:41:25Mode"#rockbox +o scorche" by ChanServ (ChanServ@services.)
21:41:26amiconnbertrik: Yes I do mean that. For me it makes a *huge* difference
21:42:13TheSevengevaerts: the dfu stub will remain in place until it is uninstalled
21:42:14amiconnI'm not sure which of the two changes helps more, as I didn't try them separately
21:42:25TheSeveni see no need to ever update that one
21:42:39amiconnIt may depend on the phones. If it does, it's probably bias
21:42:56Mode"#rockbox -o Torne" by Torne (torne@rockbox/developer/Torne)
21:43:03gevaertsTheSeven: ok. I think you have everything covered then
21:43:22 Join giovanni [0] (~giovanni@host5-235-dynamic.21-79-r.retail.telecomitalia.it)
21:43:34TheSevenanother question would be if we want to add a way of booting the apple bootloader from ours
21:43:44TheSeventhere would be enough free space in the nor flash to do that in theory
21:43:48kugelbertrik: you're a bit into powermgmt aren't you? feel like working on the android port? :)
21:43:53gevaertsIs it useful?
21:44:14TheSevenprobably not
21:44:30funmankugel: the linux source code makes me think that it's used in some PPC SoC
21:44:34kugelI want the sleep & poweroff timers from powermgmt.c, but not the battery/voltage/estimated runtime things
21:44:45Mode"#rockbox -b *!*@*.se.biz.rr.com" by scorche (~scorche@rockbox/administrator/scorche)
21:44:50TheSevenif we're sure that end users won't ever get their hands on non-working bootloaders, that self-uninstalling way is probably safe
21:45:16Mode"#rockbox -b *!*@*.dialup.umc.net.ua" by scorche (~scorche@rockbox/administrator/scorche)
21:45:21gevaertsSafe enough, if the DFU stub survives that
21:45:21bertrikkugel, I did some power stuff but I'm not really enjoying it per se
21:45:25funmanamiconn: i tried pretty much every as3543 settings i could find on my clipv2
21:45:35gevaertsAnd even if not, as safe as the coldfire irivers
21:45:50funmani plan to go back at it again after USB works
21:45:54TheSeventhe dfu stuff can't survive that as it's overwritten by the apple bootloader
21:46:08gevaertsright
21:46:31gevaertsSo it's a matter of being careful then
21:46:36TheSevenhowever, in case garbage is flashed or the flasing process is interrupted, bootrom DFU will kick in
21:46:56TheSeventhe only real way to brick it is to flash correctly signed but non-working code
21:47:01gevaertsAh yes, only properly flashed broken code is actually dangerous :)
21:47:13TheSeventhis might actually happen while uninstalling, as the apple bootloader will be signed on-device
21:47:29kugelbertrik: it would be nice if rockbox could exit itself after some time (i.e. like it powers down on daps). while the app is quite undemanding it still adds a non-zero load on the system if doing nothing
21:47:44kugelbut powermgmt.c is a bloody mess
21:48:05TheSevenwe might want to sigcheck the AUPD to prevent a bad AUPD file from bricking it
21:48:17TheSevenbut if the uninstallation code is broken, that won't help
21:48:36funmankugel: where is the non-zero load, something is busy looping?
21:48:55kugelmainly the system tick
21:49:22Mode"#rockbox +q *!*@gateway/shell/bshellz.net/*" by scorche (~scorche@rockbox/administrator/scorche)
21:49:22funmanmake HZ variable! :)
21:49:30Mode"#rockbox -b *!*@gateway/shell/bshellz.net/*" by scorche (~scorche@rockbox/administrator/scorche)
21:50:03[Saint]funman: I can test that on my Nano2Gs, sure..what's it do?
21:50:20*[Saint] is lazy, just woke up, and there are a lot of logs...
21:50:33funman[Saint]: just moves things around so the driver can be used on AMSv2
21:50:46[Saint]Ohhhh, right. Ok, sure.
21:51:31gevaertsTheSeven: I think you're fairly safe once you reach that stage. More than safe enough, anyway
21:51:45[Saint]I'm also going to try and see if I can enable USB HID again on the Nano2G (afterward) and see if it has magicallly fixed itself.
21:51:52kugelfunman: wait_for_interrupt() works but it still interrupts every 10ms
21:52:29gevaerts[Saint]: you could of course test if new svn now works properly as is :)
21:52:49TheSeventhat patch probably won't have an influence on hid
21:52:55funmancould you disable the tick while rockbox is idle?
21:53:31funmanand be notified when rockbox is used again to enable it
21:54:32kugelI had a quick shot at not waking up and it didn't work well. disabling the tick well break every TIME_AFTER()/_BEFORE() so I don't like that idea really
21:54:54kugelanyway, quitting would give the least added load (exactly zero) so I would prefer that one
21:55:33kugelthe app starts quick enough to not be annoying
21:56:03[Saint]gevaerts: ...huh?
21:56:18[Saint]Did I miss something re: HID?
21:56:19gevaerts[Saint]: I committed my fat patch
21:56:22gevaertsNo, not HID
21:56:33 Quit Lear (Quit: ChatZilla 0.9.86 [Firefox 4.0b6pre/20100902090004])
21:56:46gevaertsThe nand stack
21:56:52[Saint]gevaerts: Oh, right..hmm.
21:57:19 Join heorshe [0] (hairyballs@gateway/shell/bshellz.net/x-rezcymwfuxcpuduu)
21:57:34[Saint]so I can drop my ridiculous (~3X greater than default( NAND stack?
21:57:42[Saint]+)
21:57:51 Part heorshe
21:57:54TheSevenhopefully yes
21:58:25 Join heorshe [0] (hairyballs@gateway/shell/bshellz.net/x-lfrlroroltdhyacj)
21:58:47 Join j0ek [0] (cwilly@gateway/shell/bshellz.net/x-ppjwcvutwnktaejk)
21:59:00gevaertsSomeone still needs to fix the screendump code though. It allocates a full line on the stack, which on mr500 means about 2K
21:59:03*[Saint] also wonders what TheSeven has on his mind for Nano2G botloader
21:59:39[Saint]gevaerts: So, should screendump work or not, on my 2G?
21:59:45TheSevenit should
21:59:54[Saint](I know it was previously well borked)
22:00
22:00:02TheSevenand regarding the bootloader, I'm wondering what to do about the rockbox bootloader
22:00:18pixelmaoh, good. With r28011 Rockbox consistently freezes when trying to load my theme on the M5 :\
22:00:21[Saint]to et itunes not know we've fiddled with it?
22:00:29[Saint]*let
22:00:36TheSevenshould we use the NOR-based one as the only supported way of installation, or keep the nand one and add a (recommended) option to flash the NOR one from within rockbox?
22:00:46gevaertsI think it should. The USB stack is 3KB, it *probably* needs a sector in fat.c, but that leavs 1KB for one line, which isn't *that* big on a nano
22:00:54TheSeventhe latter would allow users to have a quick glance if they like rockbox without flashing their nor
22:01:01TheSevenbut it may complicate things a bit
22:01:20TheSevenand on the other hand, having a nand-based bootloader won't have any advantage if everything works like i'm planning
22:01:29[Saint]my problem with the NOR install is it removes the "use itunes to restore" which is the safety net for all iPods at present.
22:01:41[Saint]But, if that can't be avoided...so be it.
22:02:00TheSeveni could make the bootloader detect a restore and uninstall itself in that situation
22:02:02kugelI think it would be nice if the itunes restore uninstallation/safety net can be kept
22:02:39[Saint]kugel: As do I, but, iLoader based booting it SOOOOOOOOOOOOOOOOO much cleaner
22:02:47[Saint]si, its 50/50
22:03:18[Saint]but, I'd really like to restore with iTunes, well, not me personally, but to be able to redirect people to do so safely.
22:04:44gevaerts[Saint]: "use itunes to restore" is nice, but if it comes at the cost of having to restore once a week, well...
22:05:10[Saint]gevaerts: well...that's a bit overstated
22:05:17[Saint]things are not as bad as they seem...
22:05:28[Saint]the Nano2G is REALLY stable...bootloader wise
22:05:33gevaertsOf course it is! Nobody has ever won an argument by being subtle@!
22:05:42[Saint]Mine has not had the "white screen of death" for...months
22:05:50[Saint]4, 5 months, easily.
22:05:58[Saint]and, I give them a LOT of abuse
22:05:58kugelall of them?
22:06:07[Saint]kugel: All of them.
22:06:08gevaertsThat's because you can't actually write any file because it overflows the stack first
22:06:47kugel[Saint]: btw, I'm not uninterested in your icons, but I don't like the introduced inconsistency
22:07:02kugelas I said I'm not opposed to fixing the svn iconset
22:08:12 Join Jerom1 [0] (~heidi@95.171.131.249)
22:08:12 Quit Jerom (Read error: Connection reset by peer)
22:08:45 Quit eclectic (Ping timeout: 240 seconds)
22:09:07[Saint]gevaerts: or mine...whatever works: NAND stack/fat fix should be plenty to make the Nano2G a WHOLE lot nicer for "average joe" to use.
22:09:24[Saint]despite what poeple might think...I only use iLoader on one Nano2G
22:10:31 Quit fdinel (Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org)
22:10:37[Saint]the RB bootloader I actually find faster even, unless iLoader is on fastboot theme.
22:10:57[Saint]I have NO idea why TheSeven's Nano takes so damn long to boot..no idea at all.
22:11:09TheSevendoesn't yours show an apple logo at all?
22:11:19[Saint]yes, it does.
22:11:23gevaerts[Saint]: having a default setup that *can* wipe your filesystem (even if it doesn't often happen) really isn't a nice option if there is an alternative
22:11:32[Saint]it still gets to the RB menu in ~2 seconds though.
22:11:36[Saint]all of them do.
22:11:45TheSevenmine shows an apple logo for about 5 seconds before doing anything else
22:11:59[Saint]gevaerts: I'm not convinced it still can, though.
22:12:01TheSevenand the iloader menu is shown in <1 second on mine
22:12:06[Saint]noone has reported this in AGES
22:12:24TheSeven[Saint]: try copying files to it via usb while holding the reset combo
22:12:33[Saint]If anything...it should happen to me...I give mine a LOT of torture
22:12:37TheSevencount how many tries you need to wsod it
22:12:38gevaerts[Saint]: I'm sure TheSeven will yell at me if I'm wrong, but I thought the root cause is well understood and still there
22:12:47[Saint]hasn't happened to me in...shit, 5 months, easy
22:13:09TheSevenhm, depends on what you define to be the root cause
22:13:27TheSevenwe know what's triggering it, but not why the apple bootloader fails to recover from that situation
22:13:30 Join saratoga [0] (9803c22e@gateway/web/freenode/ip.152.3.194.46)
22:13:50gevaertsTheSeven: right. Yes, finding *that* would be an even better solution of course
22:13:57saratogaamiconn: FWIW people tried the high quality patch with a few different headphones (high, low impedance) and no one found any measurable difference in RMAA
22:13:58TheSevenit also doesn't always fail, just sometimes, when there was a lot of writes before crashing
22:14:10[Saint]all I'm saying...is that if it is still there, it was exaggerated by something that seems to NOT still be there...anymore.
22:14:21saratogamy assumption was that it didn't actually work in whatever version of the chip sandisk used
22:14:32[Saint]I realy should see it, I write large ammounts to USB on them all daily...err, almost daily.
22:14:42 Quit Ramsey[LC]_ (Remote host closed the connection)
22:14:47[Saint]and change the RB build sometimes 3 ~ 4 times daily
22:14:50TheSevendoes it crash after that?
22:14:53[Saint]no WSODs for me.
22:15:07TheSevenit can only wsod if you hard reset it after writing
22:15:13TheSevenand before rolo'ing
22:15:15clone4crwgaverts: I just uploaded my new resistor calculator, and am looking at your version. I like what you did with the structures.
22:15:15[Saint]yeah...I know.
22:15:35clone4crwgavaerts: ^^
22:15:37TheSevenis that a situation you encounter often?
22:15:40[Saint]TheSeven: If rolo is the acid test...then, it is fine.
22:15:43TheSevenclone4crw: still wrong :P
22:16:03TheSevenno, rolo will make sure everything is clean before trying to boot the new build
22:16:11clone4crwgevaerts: Sorry. ^^
22:16:17[Saint]TheSeven: I'd use rolo on at least 2 nanos ...twice daily I'd say.
22:16:24[Saint]at least once on one.
22:16:33[Saint]and, no WSODs
22:16:36TheSevenwhat i'm saying is that rolo'ing is *preventing* it
22:16:38Mode"#rockbox -q *!*@gateway/shell/bshellz.net/*" by scorche (~scorche@rockbox/administrator/scorche)
22:16:40 Join Dreamxtreme_ [0] (Dreamxtrem@92.30.177.143)
22:16:58[Saint]the cause is still there..most definitely, but, whatever was pissing it off isn't..apparently.
22:17:31TheSeven[Saint]: how often do you hard-reset your ipod in a week?
22:17:50TheSevenafter writing something, and not rolo'ing
22:17:53[Saint]with the pictureflow troubles recently......
22:18:09[Saint]hmm, about 10 times daily for the lot of them for the past few days.
22:18:18[Saint]got *that* sorted in pictureflow though.
22:18:32TheSevenhave you written anything to the flash before triggering the bug?
22:18:39gevaertsclone4crw: judging from a very quick look, you mainly added the resistance_to_color bit, right?
22:18:52[Saint]"triggering the bug"?
22:19:00TheSeventhe pictureflow crash
22:19:10[Saint]oh, sure.
22:19:19[Saint]it manages to write the pictureflow cache
22:19:34TheSevenhm
22:19:37[Saint]then, crash...hard reset, no WSOD
22:19:38 Quit Dreamxtreme (Ping timeout: 255 seconds)
22:20:06TheSevenlet me know if you ever happen to see one again
22:20:12clone4crwgevaerts: yes, mostly that
22:20:14[Saint]actually thinking about it...remember how many people used to run in here, and the forums, complainng of it?
22:20:19gevaertsMaybe the Apple FTL recovery code is self-learning?
22:20:21[Saint]that's stoped, completely.
22:20:36gevaertsclone4crw: ok, in that case my changes shouldn't get in the way too much :)
22:20:47clone4crwgevaerts: Now I'm going to combine both of ours
22:21:07[Saint]TheSeven: I don;t sweep them under the rug ;)
22:21:22TheSeven[Saint]: we have not changed anything in the ftl since those times where they complained
22:21:23[Saint]I want them fixed too :P but, I honestly haven't seen one in...ages.
22:21:26gevaertsclone4crw: good luck! I actually have to go now, I guess I'll see the results later on
22:21:28[Saint]no mattter what I do.
22:21:29TheSevenmost of them are probably just using iloader
22:21:48[Saint]not the ones I know...but, ok.
22:21:55j0eklater gevaerts.
22:22:07[Saint]all (except one) of the 2G users I know use the RB loader
22:22:09 Join eclectic [0] (~bill@pool-72-81-186-207.bltmmd.east.verizon.net)
22:22:11TheSevenwell, if you have any explanation why they might have been gone...
22:22:29[Saint]I think there was something aggrivatring it we never saw
22:22:33[Saint]and, missed
22:22:46[Saint]and, has since departed us.
22:23:09*[Saint] shrugs...
22:23:22[Saint]that's my guess anyway...but, at least iLoader came from it.
22:23:56 Quit [sko] (Quit: Leaving.)
22:23:57amiconnsaratoga: Imo that's rather strange. I didn't measure, but for me the patch makes the difference between almost unusable and acceptable quality
22:24:31[Saint]re: <TheSeven> [Saint]: we have not changed anything in the ftl since those times where they complained
22:24:44[Saint]haven't yo umade multiple changes to the FTL since?
22:25:01bertrikamiconn, I'm curious about which part of the patch makes the most difference for you
22:25:13TheSeveni don't think any of those might have affected it
22:25:21TheSevenmost of it was internal code cleanup
22:25:25[Saint]well, you never know...
22:25:45[Saint]I often am surprised bt what affects what.
22:26:45 Part heorshe
22:26:58[Saint]TheSeven: maybe you could add a poll on the forums?
22:27:25[Saint]do you use A: iLoader B: RB loader...have yo had any :A WSOD B: unclean shutdown..etc
22:27:46 Part j0ek
22:29:38 Quit bunnyboi (Read error: Connection reset by peer)
22:31:29 Quit clone4crw (Quit: leaving)
22:34:57 Quit mc2739 (Ping timeout: 258 seconds)
22:35:07saratogaamiconn: maybe its only needed on some device revisions?
22:35:10 Join bunnyboi [0] (~androgyne@cpe-72-224-19-157.nycap.res.rr.com)
22:36:56 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
22:37:42 Quit [Saint] (Quit: Even if you're lying, please tell me everythings going to be fine.)
22:40:16 Quit Jaykay (Quit: ChatZilla 0.9.86 [Firefox 3.6.8/20100722155716])
22:43:45 Quit funman (Quit: free(random());)
22:44:43bertriksaratoga, there are indeed subtle changes between some clip+ models, e.g. with respect to the sd card controller
22:45:15alexbobPdoes rockbox have usb host support? could I plug in e.g. a flash drive with a modified cable and expect it to work?
22:45:26alexbobPI have a sansa fuze v2
22:46:49bunnyboialexbobP, no host support, not even the gigabeat S has host support in rockbox
22:46:50bertrikit depends on the specific target, but as far as I know, no current rockbox target acts as a USB host
22:48:02bunnyboialso, i seem to hit a snag. slack only sees the internal storage on my fuze v2
22:48:53 Quit Dreamxtreme_ (Ping timeout: 255 seconds)
22:52:59alexbobPaww!
22:53:06n1samiconn: i was asking because one person in the forum has trouble building the sim, first because of seemingly slightly different system headers (files using the SEEK_CURR etc symbols needed to include stdio.h) but then got an unresolved symbol from __builtin_clzl and that was (according to gcc manuals) included in 3.4
22:53:27n1slenny shouldn't have gcc < 3.4 i think
22:57:55n1sanyway, untill we have a verified working virtual machine or a decent guide for creating your own we should probably slap a big warning on that page telling people to download that old vmware image
22:57:56***Saving seen data "./dancer.seen"
22:59:09n1sactually, since installing the necessary tools and running rockboxdev.sh is about 2 commands maybe it doesn't make sense for us to create our own vm images anymore?
22:59:31bunnyboiwhat's needed for a proper virtual machine? i got a vm of Debian that i've been using. though that is just for the arm targets
22:59:34n1syou need to co svn too so 3 commands
23:00
23:00:21n1sbunnyboi: for there to be any point to it it would need all the target toolchains, the libs to build sism and whatever is needed to build the manual IMO
23:00:59n1ss/sism/sims
23:01:02bunnyboiok, i can just look at the manual in a bit, and just use the VM i have
23:02:07n1sbut is it worth the hassle of maintaining our own vm image rather than telling people to just get a random linux vm and use the usual linux way to do stuff?
23:11:05 Quit n1s (Quit: Lämnar)
23:15:07 Quit bunnyboi (Quit: Ex-Chat)
23:30:29 Quit BlakeJohnson86 (Quit: Leaving.)
23:31:23 Join BlakeJohnson86 [0] (~bjohnson@c-24-118-162-123.hsd1.mn.comcast.net)
23:34:58 Quit pamaury (Remote host closed the connection)
23:35:36 Quit kugel (Ping timeout: 245 seconds)
23:38:04 Quit BlakeJohnson86 (Remote host closed the connection)
23:38:23 Join BlakeJohnson86 [0] (~bjohnson@c-24-118-162-123.hsd1.mn.comcast.net)
23:38:44 Quit giovanni (Quit: Sto andando via)
23:39:31 Join [Saint] [0] (S_a_i_n_t@203.184.2.40)
23:41:23 Quit domonoky (Read error: Connection reset by peer)
23:46:28 Quit bmbl (Quit: Bye!)

Previous day | Next day