**** BEGIN LOGGING AT Sun May 02 02:59:57 2021 May 02 13:27:55 hey there! i've just got my first n900, reflashed with latest fremantle and installed cssu-thumb, but having issues trying to get u-boot installed to play with leste May 02 13:28:28 it's complaining about the kernel-modules package, and if I try to install that it threatens to basically obliterate all the packages on my system, which sounds like a very bad idea; is there any way around this? May 02 13:30:04 if it's fresh device, you can always reflash, no? May 02 13:30:25 but most likely you are doing something wrong May 02 13:31:07 yeah, could reflash again; was a bit of a pain dealing with VMs and such though May 02 13:31:26 I'm assuming there's some issue with the fact I'm trying to install u-boot after upgrading to cssu May 02 13:35:25 reflash time it is then :p May 02 13:35:27 thanks May 02 13:36:04 What is the actual message you get? May 02 13:36:36 It's somewhat ok for u-boot to threaten to overwrite your existing kernel May 02 13:37:18 Because that's actually what it does ... sits in kernel area and loads a real kernel that you've defined. May 02 13:37:50 I'd say install u-boot, and only reflash if there are issues May 02 13:38:04 yeah; here it's saying there's an unmet dependency, kernel-modules, and refuses to continue (one second, I'll grab a pastebin link) May 02 13:38:59 sources.list and apt output when trying to install u-boot flasher -- https://paste.debian.net/1196021/ May 02 13:39:12 Because you are on thumb, after installing u-boot and u-boot-tools, ensure you set up the thumb kernel to boot by default. See the u-boot thread on talk.maemo.org for details. May 02 13:39:40 Make sure you do this BEFORE you power off or reboot the device May 02 13:40:52 thanks; main issue is not really sure how to even get u-boot installed at all May 02 13:42:29 since it seems I can't install u-boot-flasher without dealing with this kernel-modules issue, as installing kernel-modules as it requests would hose my system: https://paste.debian.net/1196022/ May 02 13:42:32 Install the kernel and modules it wants :-) May 02 13:42:52 Those should be thumb enabled, iirc May 02 13:43:42 something seems wrong though since I don't think installing the kernel-modules package should usually threaten to remove pretty much every package on the system, including apt May 02 13:43:43 In any case, since you'll be setting up a default boot menu entry (as I mentioned), then you can still point that entry to the original kernel May 02 13:44:06 maemo is just like that May 02 13:44:33 there is single metapackage per release May 02 13:44:59 is one of the packages that it depends on gets uninstalled, it takes the rest with it May 02 13:45:22 that's why you should be cautious when hacking around with apt/dpkg\ May 02 13:45:58 yeah, reminded me of the ubuntu-desktop metapackages on ubuntu which had similar traits May 02 13:46:07 Show the output which wants to remove everything May 02 13:46:14 here: https://paste.debian.net/1196022/ May 02 13:46:28 Also, for Maemo Leste, you may want to join #maemo-leste May 02 13:47:01 Okay that looks bad :-p May 02 13:47:07 yeah hahaha May 02 13:47:10 I forgot how to handle that one safely May 02 13:47:40 There was a way though ... but I haven't played with installing u-boot in a while now May 02 13:48:03 hmm, lemme check what's installed on my phone May 02 13:51:55 If all fails, reflash to cssu (not thumb), install u-boot, then only afterwards go thumb May 02 14:02:10 thejsa: http://pastebin.notk.org/pastebin.php?show=mc2cc6b May 02 14:02:41 that's cssu-stable (iirc) with a few additions though, not thumb May 02 14:03:04 Hello! Has somebody N8x0/770 device in hands? I would like to see first 2kB of kernel MTD partition... May 02 14:03:52 it can be read (dumped to file header.bin) by following command on N8x0/770 device: nanddump --omitoob -s 0 -l 2048 -f header.bin /dev/mtd2 May 02 14:08:39 sicelo: thanks! I think I'll go for a reflash as you suggested, probably easier and quicker than trying to untangle all this May 02 14:09:44 bencoh: thanks for grabbing that, reflashing w/ stable now May 02 14:12:32 * enyc meows May 02 14:13:04 thejsa: hrrm not sru ewhat i did but i can say u-boot-flasher went ok May 02 14:13:15 thejsa: i would say I got kernel power for u-boot installed first May 02 14:13:18 inrluhing modules May 02 14:13:24 which presumuboly yo uhave to have for thumb to behave May 02 14:13:46 installed kernel power for u-boot a lil while ago but didn't seem to make much difference May 02 14:13:57 well, to the u-boot situation anyway May 02 14:18:53 hmm, KotCzarny used to build oscp for n8x0 as well, maybe he still has one around May 02 15:20:43 unfortunatelly my n8x0 is at remote location May 02 15:20:49 so i cant get to it May 02 15:20:56 but maybe i have mtd dump somewhere May 02 15:22:45 yup, got it, let me dd the block May 02 15:24:54 https://mrrau.eu:20281/n800/n800-p2-2k.bin May 02 15:24:56 here you gu May 02 15:24:59 *go May 02 15:25:15 but i dont remember what i used to dump it May 02 15:26:13 it might have been done with dd and /dev/mtdblock May 02 16:22:09 KotCzarny: thanks! May 02 16:22:34 can you "dd" also whole partition? May 02 16:27:20 just to check if your kernel image in that partition has size 1530880 bytes May 02 16:27:44 it is 1.45MB so seems plausible (below 2MB limit) May 02 16:44:49 now I pushed lot of changes to https://github.com/pali/0xFFFF flasher and fiasco image unpacker/generator May 02 16:45:15 feel free to test it! now it supports unpacking also /home partition from vanilla/mmc fiasco image May 02 16:45:45 (yes, /home partition is included in vanilla/mmc image as tarball, not as ext3) May 02 23:42:03 pali: May 02 23:42:11 https://mrrau.eu:20281/n800/n800-p2.bin May 02 23:43:41 if you need anything else shout, i'm heading to bed now though May 02 23:46:02 kernel in that dump ends at offset 0x001763c8 May 02 23:46:18 aligned offset is 0x001763d0 May 02 23:46:32 which is 1530880 (in decimal) May 02 23:46:54 keep in mind those are dd dumps of mtdblock May 02 23:47:08 dont know if it matters May 02 23:47:10 so perfecly matches what I predicted May 02 23:47:16 that's good then May 02 23:47:31 thanks for confirmation) May 02 23:47:45 np, and thanks for still hacking at those old babies May 02 23:48:40 I have already pushed commit into 0xFFFF https://github.com/pali/0xFFFF/commit/a5279bffaf4ac2a97c5a443978a2abb4189eb765 which parses alse these dumps **** ENDING LOGGING AT Mon May 03 02:59:56 2021