**** BEGIN LOGGING AT Sun Jul 03 23:59:56 2005 **** BEGIN LOGGING AT Mon Jul 04 00:13:30 2005 Jul 04 04:24:25 silly question perhaps.. the kernel in /initrd/boot (after turnup), can one replace it and the slug will boot with new kernel? Jul 04 04:24:47 or.. at least try to :) Jul 04 04:25:07 no, but you can delete it. And it will still boot. Jul 04 04:25:24 ok Jul 04 04:25:38 that's what I thought perhaps was the case Jul 04 04:25:57 The kernel isn't in jffs2 Jul 04 04:26:10 right, I suspected so much Jul 04 04:26:11 If you run APEX, you can have the image in jffs2 Jul 04 04:26:32 never heard of apex :) Jul 04 04:26:46 alternative bootstrapper? Jul 04 04:27:06 alternativ bootloader, yes. Instead of RedBoot. Jul 04 04:27:12 ok Jul 04 04:28:25 :) Jul 04 04:29:02 when I do make it says "openslug-image-1.0", is that correct? Should it not be 2.0? Jul 04 04:29:23 that's correct, I think Jul 04 04:29:25 hm.. 2.0 is sti9ll in beta, perhaps that's why Jul 04 04:29:31 ok Jul 04 04:30:01 * NAiL goes away to try to upgrade system Jul 04 04:30:01 I had it compiling over night, but it stopped at libpam Jul 04 04:30:03 bbl Jul 04 04:30:59 re [g2] :) Jul 04 04:31:06 <[g2]> kolla_, hey Jul 04 04:31:25 <[g2]> I saw solar showed up later but I was gone by then Jul 04 04:33:33 he's sort of a mythical person to me :) Jul 04 04:34:19 but I'm building openslug myself now Jul 04 04:35:07 I assume openembedded/packages/linux/openslug-kernel-2.6.11.2/defconfig is the kernel config I can change to add my own stuff? Jul 04 04:35:15 (need ipv6) Jul 04 04:36:18 <[g2]> yup Jul 04 04:36:34 <[g2]> all built fine on the amd64 ? Jul 04 04:36:57 only got a P4 here Jul 04 04:37:32 <[g2]> ok that'll work fine :) Jul 04 04:37:51 but it doesnt build "just fine" though, it exits on libpam :) Jul 04 04:37:54 <[g2]> sorry I was talking with someone else late last night that was building with amd64 Jul 04 04:38:18 yes, NAiL perhaps.. his nick was NAiL-AMD64 a few minutes ago :) Jul 04 04:38:45 <[g2]> what's the error on libpam ? Jul 04 04:39:03 /usr/bin/openjade:0:1:25:W: cannot generate system identifier for document type "LINUXDOC" Jul 04 04:39:43 and then some pages with various errors from openjade Jul 04 04:40:05 <[g2]> I dunno what that is :( Jul 04 04:40:26 <[g2]> i'd guess LINUXDOC relates though Jul 04 04:40:31 mm Jul 04 04:40:32 it sounds like a SGML Error. Jul 04 04:40:43 <[g2]> moring sunshine Jul 04 04:40:51 <[g2]> or morning Jul 04 04:40:56 <[g2]> :) Jul 04 04:42:10 you may want to verify that docbook (or whatever SGML Frontend youre using) actually works. Jul 04 04:42:32 [g2] Actually, its goodnight for me. Jul 04 04:43:03 do one still use bitkeeper to update the source?... I'm full of stupid questions.. Jul 04 04:43:09 <[g2]> Well sweet dreams Jul 04 04:43:13 dyoung-zzzz: that's true Jul 04 04:43:19 <[g2]> kolla_, we've switched to monotone Jul 04 04:43:37 ah, right.. I did know that, heh Jul 04 04:46:04 export ac_cv_prog_HASDOCBOOK=no may help. Jul 04 05:04:20 kolla_: http://www.nslu2-linux.org/Makefile Jul 04 05:56:27 Haven't built on AMD64 yet. I've barely installed wintendo on it so far :P Jul 04 05:57:22 <[g2]> NAiL, I've built on amd64 Jul 04 05:57:31 <[g2]> all builds but I haven't run it Jul 04 05:57:37 ah Jul 04 05:58:03 I'll have to spend some time getting this box up before I say anything about amd64 builds Jul 04 06:23:47 hm, function do_rootfs failed.. wonder what's wrong now Jul 04 06:24:02 all I did was changing defconfig Jul 04 06:27:24 but now it worked, hm Jul 04 15:42:16 rwhitby-asleep: The memcpy is of the number of bytes in the header, not number-16 Jul 04 15:42:24 That applies both to the kernel and the ramdisk. Jul 04 15:43:30 I'll update the wiki. Jul 04 15:53:09 ok, so slugimage was right, but the wiki and kern_header were 16 bytes too many? Jul 04 16:09:21 hm, so the slug only runs at 133MHz.. I thought it was 266MHz already Jul 04 16:59:04 kolla_, http://www.nslu2-linux.org/wiki/HowTo/OverClockTheSlug Jul 04 16:59:17 more precisely, its how to de-underclock it. Jul 04 17:05:16 rwhitby: I found the wiki was a little unclear, in fact the evidence quoted there suggested the data length was L, not L-16 - because there was a non-zero byte in the kernel block after L-16. Jul 04 17:05:18 right, that's what I read :) Jul 04 17:37:08 jbowler-away: dunno about what is right. all I know is that slugimage works :-) Jul 04 17:46:53 heh **** ENDING LOGGING AT Mon Jul 04 23:59:56 2005