--- Log opened Sat Sep 04 01:18:02 2010
01:18 -!- fmibot [~fmibot@static.225.178.40.188.clients.your-server.de] has joined #freemyipod
02:21 -!- TheSeven [~TheSeven@rockbox/developer/TheSeven] has quit [Ping timeout: 255 seconds]
02:25 -!- TheSeven [~TheSeven@rockbox/developer/TheSeven] has joined #freemyipod
05:00 -!- clustur [~logger@c-76-127-58-39.hsd1.ga.comcast.net] has quit [Remote host closed the connection]
05:00 -!- clustur [~logger@c-76-127-58-39.hsd1.ga.comcast.net] has joined #freemyipod
05:04 -!- mulenmar [~mulenmar@74-132-43-158.dhcp.insightbb.com] has joined #freemyipod
05:07 < mulenmar> Hello. I'm not entirely sure this shouldn't be asked in #freemyipod-support or not, but I have a question about emBIOS.
05:08 < mulenmar> I noticed it replaces whatever displays the Apple logo during boot, and remains even if I delete the data partion /dev/sdX2
05:08 < mulenmar> (which I know shouldn't be done, I'm fiddling/breaking/fixing things)
05:09 < mulenmar> So, does the emBIOS get installed to the firmware partition, or what?
05:10 < mulenmar> I'm just a little confused -- not on how to DO it (install it), but on how it is DONE.
05:11 < mulenmar> I'm updating my blog's directions on how this new procedure works, and I don't want to explain things incorrectly. ^_^
05:13 < mulenmar> Well, I guess it CAN'T be the Apple firmware partition, since iLoader can completely remove that and everything still works. #facepalm
05:14 < mulenmar> So, I guess the most defined questions I have are:
05:14 < mulenmar> 1) Is emBIOS basically a kernel that completely replaces part of the iPod boot sequence?
05:15 < mulenmar> 2) Where in vacumn diagrams is it stored!?
05:15 < mulenmar> (and yes, I'm browsing the SVN -- I just don't really know what I'm looking for)
06:43 < mulenmar> Also, I seem to have found a bug.
06:45 < mulenmar> With my Nano 2G, after following the process of Restoring with iTunes -> extracting .rockbox to the Nano 2G -> running ipodpatcher, and then instructing iLoader to repartition to recover the Apple firmware partition, I cannot load Rockbox.
06:46 < mulenmar> I press the button for Rockbox to load, and it freezes at "Booting . . ." and displaying the Rockbox logo.
06:47 < mulenmar> This is with the r27995 revision of Rockbox. I'll try the process one more time, but without the repartitioning, to see if this is where the problem is.
07:02 < mulenmar> Nope, same problem without repartitioning.
07:03 < mulenmar> In both cases, however, the Apple firmware loads just fine.
07:58 < mulenmar> As a measure of last resort, I followed the Rockbox installation instructions to the letter -- which involved only running ipodpatcher, with NO options or files, and telling it to install the Rockbox bootloader.
08:00 < mulenmar> This hasn't worked for me since . . . r21xxx or so. But it works now. So, I guess it's definitely an iLoader bug. But I can't track it down any more specifically without step-by-step instructions on what to do.
08:01 < mulenmar> And since things were reversed -- the Rockbox bootloader somehow resulting in White Screens of Death and the old, pre-emBIOS iLoader being needed to make things work -- for a very long time, the bug's obviously somewhere in the new code.
08:03 < mulenmar> My C-fu is pathetic, my knowlege of assembly limited to the barest basics of x86 assembly, and my skills with ARM programming none-existent. So I doubt I can help with the actual coding.
08:32 -!- benedikt93 [~benedikt9@unaffiliated/benedikt93] has joined #freemyipod
08:33 < benedikt93> mulenmar, about 1) it's AFAIK loaded at the point when normally the Apple firmware would be loaded
08:33 < benedikt93> and AFAIK it resides in the nandflash, but I can't tell you where
08:36 < benedikt93> people who really now are most likely to be around in the evening (UTC+2h)
08:42 < liar> i think its not in the nand(if it still exists after removing all partitions like you said it cant be there).. its more likely in the NOR, where iLoader was installed, but i dont know much about embios
08:48 < mulenmar> Ah. Makes sense, from what I've managed to understand from reading through the more technical parts of the wiki.
08:50 < benedikt93> NOR is a flash where some init code (and some other stuff) resides that is executed before the apple firmware is loaded. It''s a seperate chip inside the iPod you can't directly access over USB
08:52 < mulenmar> So, it's like a hardware version of an initrd or initramfs in a GNU/Linux system, right?
08:55 < mulenmar> It contains the minimum code needed to continue booting the system?
08:56 < Farthen> mulenmar it resides in the NOR flash, the boot flash of the nano 2g
08:57 < mulenmar> Yeah, that answers my question. I'll still be in here if someone needs me to follow steps to debug the issue I found.
08:57 < Farthen> it has some code in it that initializes the hardware and stuff like that
08:57 < Farthen> and yes, if you reformat the ipod it can not load rockbox... or did i misunderstand your question?
08:59 < mulenmar> The issue? No, I followed the installations directs on the iLoader website, including installing Rockbox, and iLoader stuck at "Booting" and displaying the Rockbox logo instead of actually loading Rockbox.
08:59 * TheSeven tries to catch up
09:00 < TheSeven> 1) Is emBIOS basically a kernel that completely replaces part of the iPod boot sequence? << yes
09:01 < TheSeven> 2) Where in vacumn diagrams is it stored!? << in the 1MB-sized boot flash of the iPod (NOR flash)
09:02 < mulenmar> Thanks TheSeven, benedikt93 liar and Farthen answered my questions. ^_^ Just have the new-emBIOS-based iLoader not loading Rockbox r27995 problem to sort out.
09:03 < TheSeven> hmm
09:03 < TheSeven> someone else also reported something similar, but as it works just fine on mine, I have no idea
09:04 < TheSeven> the other guy also had a lot of trouble with the backlight behaving somewhat randomly in iloader, and it sometimes failed to access the flash
09:04 < TheSeven> even diskmode sometimes couldn't mount the flash, so something is being really weird
09:05 < TheSeven> as most of iloader was rewritten, and basically everything that accesses hardware is done by emBIOS now, we can't really bisect it
09:05 < mulenmar> Wait -- "disk mode" that is what comes up with the Select button, or the Disk Mode that comes up when you reset with the Apple firmware and hold Select+Play?
09:06 < mulenmar> Select button under emBIOS iLoader, that is.
09:06 < TheSeven> one of my guesses were that it has something to do with the I2C clock, which we increased a bit, but rockbox has been using that clock (~450kHz) all the time, and apparently it worked until now. also, reducing it seemed to actually worsen things
09:06 < TheSeven> mulenmar: both
09:06 < TheSeven> it's the same thing
09:07 < mulenmar> Huh -- one is actually labeled "Disk Mode" when you get there, the other isn't, but both are registered as "Recovery Mode" by the hated iTunes. That's what confused me. Hold on, I'll look it up again . . .
09:08 < TheSeven> I have no idea why the "Disk Mode" text is missing somethimes
09:08 < TheSeven> sometimes*
09:08 < TheSeven> seems to depend on the way it is booted
09:08 < TheSeven> when launched from iloader, it behaves as it it was triggered by a white screen of death
09:10 < mulenmar> Well, if it helps Rockbox's Debug Menu -> View HW info gives some specifics on what my Nano 2G has . . .
09:11 < mulenmar> LCD: type 1 (7) LDS176
09:11 < TheSeven> the same as mine
09:11 < mulenmar> NAND: banks 2 bank 0 id 2585D3AD bank 1 id: 2585D3AD
09:12 < benedikt93> TheSeven, btw, is there anything new about Nano 3G? I myself didn't do anything the last weeks as school started again
09:13 < TheSeven> serpilliere is writing some scripts to dissect the efi hell
09:13 < benedikt93> mmh
09:15 < mulenmar> Oh lovely, the ipodlinux site is down. >_>
09:15 < TheSeven> hmm, it has been down for some weeks
09:16 < TheSeven> http://ipl.derpapst.eu/ mirrors it :)
09:18 < mulenmar> I pretty much lost faith in that project after how they completely ignored the Nano 2G's Notes app breakthrough anyway. I just wanted at the wiki, so thanks! ^_^
09:29 < TheSeven> mulenmar: can you have a look at your norflash.bak file?
09:29 < TheSeven> search for "rVwH"
09:29 < mulenmar> One moment
09:30 < TheSeven> that will be followed by some 0x00 and 0x01 bytes, and eventually a 0x07 or 0x09
09:31 < mulenmar> Actually, it'll be several moments -- I took a break after finding the regular Rockbox bootloader actually works without causing WSoDs now. Give me about five minutes.
09:32 < TheSeven> the wsods are still around
09:32 < TheSeven> they only come up after unclean shutdowns though
09:42 < mulenmar> Okay, I just finished restoring, extracting over Rockbox, and then install iLoader according to the directions on the website.
09:42 < mulenmar> I've tried this with both repartitioning and not, so I'm not going to do that this time.
09:44 < mulenmar> Alright TheSeven, you wanted NORFlash.bak in the iLoader directory, right?
09:45 < benedikt93> he did
09:45 < TheSeven> yep
09:46 < mulenmar> Okay, I'm looking at it in GHex right now.
09:50 < mulenmar> Starting from and including rVwH, the number equivalents (you can tell I'm not a programmer in anything but the most basic of shell scripts 8~P) are as follows
09:51 < mulenmar> 72 56 77 48 00 00 00 00 07 00 10 00 00 00 00 00 00 00 00 00
09:53 < mulenmar> 23 64 6F 4D 4D 41 34 37 37 00 00 00 00 00 00 00 00 00 00 00
09:53 < mulenmar> Far enough? 8~\
10:01 < benedikt93> I think even the first line would have been enough, the second is some Mod# (Model number?)
10:02 < mulenmar> Well maybe that'll prove useful in figuring out why some Nano 2G's have the issue and others don't. <shrug/>
10:02 < TheSeven> ok, so that is a Rev. 7 board
10:03 < TheSeven> mine is a rev. 9
10:03 < mulenmar> And you said your doesn't have the problem, TheSeven?
10:03 < TheSeven> yep
10:03 < mulenmar> I guess we can just blame it on Apple then?  X~D
10:04 < mulenmar> Nah, too easy.
10:32 < TheSeven> mulenmar: but you say that apart from rockbox not booting everything else seems to work?
10:33 < mulenmar> Yes -- iLoader installs just fine, it can load the Apple firmware (even with the repartitioning option), and the VERY HELPFUL poweroff function works just fine.
10:33 < mulenmar> It just sticks at "Booting" and displaying the Rockbox logo.
10:33 < TheSeven> hm, no backlight trouble either?
10:34 < mulenmar> Nope. 8~)
10:34 < TheSeven> can you try downloading http://files.freemyipod.org/rockbox.ucl to /iLoader on the iPod and trying to boot rockbox again?
10:35 < mulenmar> I can.
10:36 < mulenmar> I did.
10:36 < mulenmar> I conquered. Works perfectly!
10:36 < mulenmar> Thanks!
10:36 < TheSeven> aha.
10:36 < TheSeven> so this is not a hardware related problem at all
10:37 < mulenmar> I was starting to get a sneaking suspicion that it wasn't finding something, but I still thought it was a hardware-access thing.
10:37 < mulenmar> Looks like it's just a case of missing files.
10:38 < mulenmar> Of course, there's still the issue of occasional random "*PANIC* Stkov nand" with the current Rockbox build, but that's not an emBIOS/iLoader issue -- that's just a Rockbox bug.
10:39 < mulenmar> I'll report that later -- I need food and coffee. ^_^
10:39 * TheSeven finally has a suspicion what could be the cause of the stkov
10:39 < TheSeven> not really the root cause or why it isn't happening for me, but what's happening and how to prevent it at least
10:41 < TheSeven> oh, and I just caught the "stuck at booting rockbox" bug :)
10:41 < mulenmar> Well last version of Rockbox I used before I updated was r24509, so the Stkov nand issue was introduced between then and now.
10:42 < mulenmar> (happens regardless of bootloader)
10:44 < TheSeven> yep
10:44 < TheSeven> the problem is coming up from the fat32 filesystem internals
10:44 < TheSeven> it might be rather trivial to fix :)
10:46 < mulenmar> Yeah, that reminds me of something.
10:46 < mulenmar> Adding other filesystems is on the NoDo list, and for good reason. But do the iPods' flash chips have any sort of wear-leveling?
10:47 < Farthen> the wear-leveling does not happen on a filesystem level
10:47 < mulenmar> I was wondering about that when I saw the Linux support for filesystems designed for solid state drives and flash devices, is all.
10:48 < mulenmar> I overoptimize my kernels sometimes, I'm afraid.
10:48 < Farthen> the flash translation layer (FTL) cares about wear-leveling and this stuff
10:48 < mulenmar> Oh, that's what I was wondering.
10:51 < mulenmar> Thanks for putting up with the newb. ^_^
10:54 * TheSeven is recompiling
10:54 < TheSeven> looks like iloader 0.2.1 will be released today
10:55 < TheSeven> uh oh
10:55 * TheSeven spots a makefile bug
10:55 < mulenmar> 0.2.1.1?
10:56 < TheSeven> no, that doesn't need a release
10:56 * mulenmar was joking
10:56 < TheSeven> sometimes it doesn't seem to remake all deps for some reason
10:56 < TheSeven> the installer makefile is just crazy
10:58 < TheSeven> mulenmar: can you download http://files.freemyipod.org/installer-nano2g.bin to /iLoader/installer.bin and http://dl.the-seven.tk/ipod/iloader/update.cfg to /iLoader/update.cfg, reboot, and choose "update iLoader" from the advanced menu?
10:58 < mulenmar> I'm wgetting right to it. ^_^
11:00 -!- clustur [~logger@c-76-127-58-39.hsd1.ga.comcast.net] has quit [Remote host closed the connection]
11:00 -!- clustur [~logger@c-76-127-58-39.hsd1.ga.comcast.net] has joined #freemyipod
11:02 < mulenmar> Okay, did it, backlight dimmed but nothing else seems to be happening.
11:03 < mulenmar> Doesn't look like it's doing anything at all. Doesn't respond to scrolling. Should I just let it sit for a few minutes?
11:10 < mulenmar> Still stuck. I'm going to try resetting it, redownloading the files, and trying again in case I made a mistake.
11:12 < TheSeven> hmm
11:13 < mulenmar> Ah, worked that time.
11:14 < TheSeven> remove the rockbox.ucl file again and try if it works without it
11:14 < mulenmar> No, I removed the files, redownloaded them, and copied the new versions over. Worked this time.
11:15 < mulenmar> Although after it finished flashing the screen faded to nothing but white instead of the usual shutting-off-and-needing-reset-to-turn-back-on deal . . . peculiar.
11:16 < TheSeven> hmm, that seems to happen sometimes
11:16 < TheSeven> like one in a thousand reboots
11:17 < mulenmar> I'm so tempted to make a "Shiny Zubat" joke here . . . 
11:17 < TheSeven> but it works without the ucl file now?
11:17 < mulenmar> Oh, I thought you meant "remove the file, maybe that's keeping the update from happening".
11:19 < mulenmar> Removed the rockbox.ucl file, re-did the update -- white screen again. Reset like always . . .
11:20 < mulenmar> "emBIOS v0.1.0 r187
11:20 < mulenmar> "*PANIC*
11:20 < mulenmar> "No known Rockbox checksum format for this target"
11:20 < TheSeven> wtf
11:20 < mulenmar> Um . . . I'm putting the rockbox.ucl file back.
11:20 * TheSeven must have mistyped something
11:21 < TheSeven> ah, missing break in a switch statement -.-
11:21 < mulenmar> If it works fine with the rockbox.ucl file, why not just put that in as the default for now? Rockbox is really the only firmware that works on the Nano 2G anyway.
11:22 < TheSeven> I don't want to provide a new ucl for every new rockbox revision :)
11:22 < mulenmar> Oh. <facepalm/>
11:28 < TheSeven> can you re-download installer.bin
11:29 < benedikt93> btw, at http://logs.freemyipod.org/%23freemyipod/ the september logs are still not available, though at clustur they are
11:30 < TheSeven> mulenmar: should be fixed in the newest installer
11:30 < TheSeven> (the checksum format panic)
11:31 < TheSeven> no idea what causes the ipod to enter DFU mode after flashing though
11:31 < mulenmar> I've read conflicting things on the DFU mode -- some say soldering required, others only damaged firmware . . .
11:31 < mulenmar> Anyway, trying it . . . . NOW
11:32 < mulenmar> Hey, no lockup . . .
11:32 < mulenmar> And . . . BINGO, we have Rockbox.
11:32 < TheSeven> soldering might be required to enter dfu mode if a bad image was flashed
11:32 < TheSeven> however, you problem is the other way round, it enters DFU mode with no apparent reason
11:33 < TheSeven> there must be some glitch while trying to read out the boot flash immediately after writing to it
11:33 < mulenmar> Oh, so that's what the screen-black/screen-white thing was?
11:33 < mulenmar> It didn't do it this time. 8~/
11:33 < TheSeven> yes, that "fading away to white" is usually a sign of DFU
11:34 < mulenmar> Huh. Usually it just goes completely black, and I have to reset it anyway.
11:34 < TheSeven> well, if the backlight is off, it might seem to be black :)
11:34 < mulenmar> No, I can tell with  a bright light.
11:34 < mulenmar> Not EASILY, mind you, but there's nothing on the screen.
11:34 * TheSeven hasn't ever observed lockups with a black screen
11:35 < mulenmar> It looked exactly like there was no power at all to the LCD or backlight. But now that it first switched to going white, and then not doing it at all, I can't really troubleshoot it.
11:36 < fmibot> New commit by 3theseven (r188): Fix a couple of annoying iLoader bugs: Make it recognize the rockbox checksum correctly, and stop the theme from locking up if booting rockbox fails.
11:36 < fmibot> r188 build result: All green!
11:37 < fmibot> New commit by 3theseven (r189): Release iLoader v0.2.1
11:37 < fmibot> r189 build result: All green!
11:44 -!- benedikt93 [~benedikt9@unaffiliated/benedikt93] has quit [Quit: Bye ;)]
11:45 < TheSeven> ok, this fix has been officially released :)
11:46 < mulenmar> I updated my blog post to credit your fixing the issue, TheSeven. ^_^
11:46 < TheSeven> which blog btw?
11:46 < mulenmar> mulenmar.wordpress.com
11:46 < mulenmar> This'll make it so iLoader will recognize all versions of Rockbox from here on, right?
11:48 < TheSeven> yes
11:49 < TheSeven> btw, restoring using itunes before installing iloader is not needed at all
11:49 < mulenmar> I figured as much, but I like to make sure everyone starts from the same place. Less frustration that way.
11:50 < mulenmar> It's really just for people already using Rockbox or an older version of iLoader -- those just discovering it can skip that step.
11:51 < TheSeven> if the RB bootloader was installed, there's no need to do it either. ipodpatcher will care about that case.
11:51 < TheSeven> if the old iloader was installed before, it must be uninstalled manually, which involves restoring using itunes
11:53 < mulenmar> So -- restoring is only actually NECESSARY if an older version of iLoader was used, but NOT if the Rockbox loader was used?
11:53 < TheSeven> yes
11:53 < mulenmar> Got it -- editing for clarity
11:54 < TheSeven> and the uninstallation instructions for the old iloader mention it themselves
11:55 < mulenmar> I try to condense things to a few easily understood steps, while giving links to the more technical info -- my blog is mainly to jog my own memory, while designed so that others can benefit.
11:57 < mulenmar> "Older version of iLoader" -- does that mean anything pre-emBIOS, or any version whatsoever?
11:58 < TheSeven> anything pre-embios
11:58 < mulenmar> Got it.
11:58 < TheSeven> but you can't update iloader through ipodpatcher
11:59 < mulenmar> Only through the "Update iLoader" option in the advanced menu, right?
11:59 < TheSeven> yes
11:59 < mulenmar> That's more advanced stuff, that isn't even in the wiki yet.
12:00 < TheSeven> hm?
12:00 < TheSeven> http://the-seven.tk/ipod/iloader/updating.php
12:02 < mulenmar> I just followed your directions on that -- I didn't see anything in the freemyipod Wiki. And I completely missed that on your site.
12:02 < mulenmar> <facepalm/>
13:03 -!- funman [~fun@rockbox/developer/funman] has joined #freemyipod
14:06 -!- funman [~fun@rockbox/developer/funman] has quit [Quit: free(random());]
15:13 < TheSeven> regarding the backlight: embios seems to be writing to only 3 of 4 registers related to the backlight
15:13 < TheSeven> so apple probably screws up the fourth
15:13 < TheSeven> we should just set that to a sane value on boot
15:14 < mulenmar> Oh boy, two version updates in a matter of hours? 8~P
15:14 < mulenmar> I'll be here to test it. ^_^
15:15 < TheSeven> this one will affect even more components
15:15 < mulenmar> Oh boy.
15:15 < TheSeven> embios, embios loader and the installer this time
15:16 < TheSeven> however, i ignored it some hours ago because it doesn't affect iloader directly
15:17 < mulenmar> It's connected somehow -- after the FTL cleanup . .. whatever it is after an unclean shutdown, merely TOUCHING the trackwheel sends it reeling into OFW.
15:18 < mulenmar> Other than these issues, I don't see any other problems with the emBIOS and iLoader yet. *knocks on wood*
15:18 < TheSeven> let's go after the backlight one first
15:19 < mulenmar> Alright. Do all the iPods use the same number of registers? It might be easiest to set them all to a sane default in one go.
15:19 < mulenmar> Say, about half? Easy to read, but not killing the battery?
15:19 < mulenmar> Just to be safe.
15:20 < mulenmar> I don't know how much work that would be though -- still haven't sorted through actual code yet, and would have difficulty knowing what I was looking at if I did.
15:23 < fmibot> New commit by 3theseven (r190): Initialize the backlight properly (on nano2g)
15:24 < fmibot> r190 build result: All green!
15:24 < TheSeven> in case you're curious: http://websvn.freemyipod.org/revision.php?repname=freemyipod&path=%2Fembios%2F&isdir=1&
15:25 < fmibot> New commit by 3theseven (r191): Urgh, copy&paste bug
15:25 < fmibot> r191 build result: All green!
15:28 < fmibot> New commit by 3theseven (r192): Also add backlight_init to backlight.h
15:29 < fmibot> r192 build result: All green!
15:36 < mulenmar> So, stupid question: what's the process for me to try this new emBIOS build? Is iLoader built against it somehow, or what?
15:36 < mulenmar> Or do I use ipodpatcher -a to send the emBIOS bin over?
15:36 < TheSeven> this will require flashing a new version of embios
15:37 < TheSeven> if you want to do it yourself, you'll need to compile an installer
15:37 < TheSeven> so i'll probably make one and upload it :)
15:37 < TheSeven> this will be a regular iloader update like every other one
15:38 < mulenmar> I've got a toolchain compiled for ARM, with the Rockbox script. I'm just not entirely comfortable with how to use it yet.
15:39 < TheSeven> check out the freemyipod svn
15:39 < TheSeven> cd to apps/installer-nano2g and run:
15:40 < TheSeven> make flashfiles && make
15:40 < TheSeven> your ipod needs to be connected, powered on and running embios during that process
15:40 < TheSeven> and you need to have python and pyusb (debian package python-usb) installed
15:40 < TheSeven> and parts of it will probably need to run as root
15:40 < TheSeven> (everything accessing the ipod)
15:41 * TheSeven has not yet bothered to figure out which device node permissions need to be changed
15:41 < TheSeven> probably usbfd
15:41 < TheSeven> usbfs*
15:42 < mulenmar> Well, for me the iPod is /dev/sdb*
15:42 < mulenmar> The storage partition being /dev/sdb2
15:43 < TheSeven> it won't be using the mass storage driver to talk to it :)
15:46 < TheSeven> http://www.netfrag.org/cgi-bin/dwww/usr/share/doc/libusb-0.1-4/README.Debian
15:47 < mulenmar> Yeah yeah, let me figure out how to get make using the ARM toolchain I built earlier. X~P
15:47 < TheSeven> so you apparently need permissions to access /dev/bus/usb
15:47 < TheSeven> CROSS=arm-elf-eabi- make flashfiles etc. :)
15:50 < mulenmar> That's "arm-elf-eabi-gcc", right? Sorry, I've been awake about 21 hours.
15:50 < mulenmar> Oh, guess not.
15:55 < mulenmar> Got a compile error, I'm going to try putting it up on pastebin.
15:56 < mulenmar> Okay, it's http://pastebin.com/FLMPH5VS
15:56 < mulenmar> It don't know if URLs get removed here.
15:57 < mulenmar> Important part looks like "ImportError: No module named core"
16:04 < TheSeven> hmm
16:04 < mulenmar> Probably something on my end. I installed python-usb first.
16:05 < TheSeven> apparently you're using libusb-0.1
16:05 < mulenmar> Ubuntu 10.04.
16:05 < mulenmar> (I prefer Arch, but I was working on another experiment, so . . .)
16:05 < mulenmar> What version of libusb do I need?
16:06 < TheSeven> http://sourceforge.net/projects/libusb/files/libusb-1.0/libusb-1.0.8/libusb-1.0.8.tar.bz2/download
16:06 < mulenmar> Wait a second, I don't have libusb-1.0-0-dev installed. >_>
16:07 < mulenmar> It has to be 1.0.8? Alright . . .
16:07 < TheSeven> any 1.0.x version will probably work on linux
16:07 < TheSeven> on windows, one needs something very recent
16:08 < mulenmar> Well even with the -dev package installed, it doesn't work. Might as well try the 1.0.8 version.
16:09 * TheSeven is mixing up things
16:09 < TheSeven> actually pyusb seems to be the culprit
16:10 < TheSeven> http://download.berlios.de/pyusb/pyusb-0.4.2.tar.gz
16:12 < mulenmar> Oh joy -- in reading the README, it shows me I nead python-dev -- which I don't have. Maybe that's the issue, we'll see.
16:12 < mulenmar> This is why I don't like Debian-likes' separation of program and -dev
16:16 < mulenmar> Followed the directions, built and install pyusb-0.4.2 successfully, even completely removed the source tree and re-checked out. Still same error.
16:18 < mulenmar> Built and installed libusb-1.0.8. Still no go.
16:18 < mulenmar> *sigh*
16:19 < mulenmar> I guess you'll have to compile something together.
16:30 < TheSeven> http://files.freemyipod.org/installer-nano2g.bin as usual
16:32 -!- benedikt93 [~benedikt9@unaffiliated/benedikt93] has joined #freemyipod
16:33 < fmibot> New commit by 3theseven (r193): Restore attributes of the flashfiles folder
16:33 < fmibot> r193 build result: All green!
16:43 < mulenmar> I don't know if I've completely forgotten the way to cleanly shut down the default firmware (NOT just got to sleep) or what, but after a hard-reset the backlight is still off.
16:44 < mulenmar> That not-really-turning-off thing is the first thing that sent me looking for replacement software for it, actually.
17:00 -!- clustur [~logger@c-76-127-58-39.hsd1.ga.comcast.net] has quit [Remote host closed the connection]
17:00 -!- clustur [~logger@c-76-127-58-39.hsd1.ga.comcast.net] has joined #freemyipod
17:37 < mulenmar> Summary of that last statement of mine: the fix didn't work, the backlight is still off after hard-reset.
17:38 * TheSeven doesn't get why
17:38 < TheSeven> this didn't happen with the old iloader, did it?
17:38 < mulenmar> Nope.
17:44 -!- funman [~fun@rockbox/developer/funman] has joined #freemyipod
17:47 < fmibot> New commit by 3theseven (r194): Also fix the backlight init in the loader.
17:47 < fmibot> r194 build result: All green!
17:53 < fmibot> New commit by 3theseven (r195): Fixup constant in the installer boot stub
17:53 < fmibot> r195 build result: All green!
18:47 < benedikt93> TheSeven, funman , do you think it would be useful to find out how exactly the code from nor is decrypted by bootrom (Nano 3G)?
18:48 < TheSeven> IIRC someone has already analyzed that
18:48 < TheSeven> might have been planetbeing...
18:49 < TheSeven> and it isn't really that interesting, because we can decrypt it, and won't need to encrypt it (thanks to pwnage 2.0)
18:53 < benedikt93> whom is planetbeing?
18:54 < funman> planetbeing is iphone hacker
18:54 < funman> TheSeven: what's pwnage 2.0 and what does it do?
18:55 < funman> http://blog.iphone-dev.org/post/42858313/thanks-for-waiting ?
18:56 < funman> benedikt93: http://linuxoniphone.blogspot.com/
18:56 < funman> ah no it's not his blog
18:56 < funman> well in fact yes but not only him ^^
18:58 < funman> i was thinking encrypting is mandatory to install on nor
19:00 < benedikt93> I think TheSeven means finding out how to encrypt frimware files and than place one with emBIOS in it on the firmware partition to be run by norflash
19:00 < benedikt93> *then
19:02 < benedikt93> funman, do you then have any idea what to do on Nano 3G, or just wait for serpilliere to finish the efi parsing scripts?
19:03 < funman> no idea, there's a lot of things i ignore on nano3g
19:03 < funman> i vaguely know that the NOR is using an EFI format
19:04 < benedikt93> what are you doing then?
19:04 < funman> nothing :)
19:04 < funman> i had a quick look at the first part of NOR which is in cleartext but didn't see anything that looked related to LCD
19:05 < benedikt93> it's just some strings, each 20 bytes long
19:05 < funman> which strings?
19:06 < benedikt93> until 0xb7 in nor, there are firmware version, hw version, and some more
19:06 < benedikt93> or, wher did you mean in nor?
19:06 < benedikt93> *where
19:07 < funman> ah
19:07 < funman> i have the part which starts with the exception vectors
19:07 < benedikt93> where is that (address)?
19:09 < funman> 0x8800 it seems
19:10 < benedikt93> that's the code section loaded by bootrom
19:10 < benedikt93> I thought that was encrypted
19:13 < funman> it is but i decrypted it with serpi's code
19:14 < funman> IIUC the other part of the NOR, even when decrypted, is still obfuscated. I didn't try to decrypt it
19:15 < benedikt93> could you send me serpillieres decrypting code then, mine did for some reason fail?
19:18 < funman> let me try dcc
19:21 < benedikt93> didn't work, pchat failing on my side
19:21 < funman> i'm behind NAT
19:22 < funman> copy your nor dump to dump_spi0.bin
19:23 < funman> arm-elf-gcc -c core.S ; arm-elf-ld -T ls.x core.o -o core.elf; arm-elf-objcopy -O binary core.elf core.bin; ./tool.py
19:32 -!- user890104 [Venci@Venci-Notebook-LAN.ipv6.6bez10.info] has joined #freemyipod
19:39 -!- ScorchedIcee [ScorchedIc@CPE001d7e6ba8cb-CM000a739a8c13.cpe.net.cable.rogers.com] has joined #freemyipod
19:40 < ScorchedIcee> 160gb need testers?  I am an programming & Analysis student, good at coding...
19:41 < benedikt93> ScorchedIcee, actually there is nothing to test on the classics
19:42 < benedikt93> nobody 'til now investigated in these
19:42 < ScorchedIcee> :( welll..... I want to :(
19:43 < benedikt93> though work on nano 3g is atm slowly starting and the nano 3g is very similiar to the classics which might lead to some reusable code
19:52 < ScorchedIcee> what are the main hurdles to get over?
19:53 < benedikt93> reverse engineer the bootloader(s)
19:53 < ScorchedIcee> is there any encryption on these that need to be removed?
19:58 < funman> benedikt93: you have experience with the nano2g ?
19:59 < benedikt93> no, I don't
20:00 < benedikt93> ScorchedIcee, yep, and it' needs to be figured out how to access hardware
20:02 < funman> can't find 0x38100000 in s5l8700x datasheet or http://code.google.com/p/iphone-elite/wiki/MemoryMap
20:08 < funman> nothing in s3c6400x either :/
20:18 < funman> i see something which goes to 0x30800000
20:21 < funman> at 0x2200E662 : str r3, [r0], r3 is 0x61000 << 15
20:22 < funman> at 0x22001048A : branch to [0x30800000, #0x10]
20:22 < funman> 0x2201048A*
20:22 < funman> in supervisor mode
20:24 < TheSeven> 381 might be the AES unit IIRC
20:25 < TheSeven> and 308 points into nowhere
20:26 < funman> TheSeven: AES is not not 38C00000 ?
20:26 < TheSeven> AFAIK the only used memory ranges on those are 00000000 to 01ffffff (mappable), 08000000 to 09ffffff (sdram), 20000000 to 2000c7ff (bootrom), 22000000 to 22020000 (sram), 38000000 to 3bffffff (ahb) and 3c000000 to 3fffffff (apb)
20:27 < TheSeven> you're right, 38c is aes
20:27 < TheSeven> what was 381... might have also been usb
20:28 < TheSeven> hm no, that's 384 and 3c4
20:28 < TheSeven> i bet i've seen 381 somewhere before
20:32 < TheSeven> hm, can't find it anywhere
21:57 -!- benedikt93 [~benedikt9@unaffiliated/benedikt93] has quit [Quit: Bye ;)]
22:00 -!- funman [~fun@rockbox/developer/funman] has quit [Quit: free(random());]
22:51 -!- ScorchedIcee [ScorchedIc@CPE001d7e6ba8cb-CM000a739a8c13.cpe.net.cable.rogers.com] has quit []
23:00 -!- clustur [~logger@c-76-127-58-39.hsd1.ga.comcast.net] has quit [Remote host closed the connection]
23:00 -!- clustur [~logger@c-76-127-58-39.hsd1.ga.comcast.net] has joined #freemyipod