**** BEGIN LOGGING AT Tue Dec 12 02:59:57 2006
Dec 12 08:42:23 <likewise> morning
Dec 12 08:55:03 <Caelian|work> rwhitby, NAiL ... ping
Dec 12 08:55:56 <rwhitby> Caelian|work: gu
Dec 12 08:56:44 <Caelian|work> rwhitby: moin ... for about a week or two now .. i haven't really had much time to look at it yet) i am unable to build debianslug any longer ...
Dec 12 08:56:59 <Caelian|work> building reiserfsprogs bails out during do_compile stage
Dec 12 08:57:28 <Caelian|work> if ccache arm-linux-gcc -march=armv5te -mtune=xscale -mthumb-interwork -mno-thumb -DHAVE_CONFIG_H -I. -I/home/pascal/slug/debianslug/tmp/work/reiserfsprogs-3.6.19-r0/reiserfsprogs
Dec 12 08:57:28 <Caelian|work>         then mv -f ".deps/do_balan.Tpo" ".deps/do_balan.Po"; else rm -f ".deps/do_balan.Tpo"; exit 1; fi
Dec 12 08:57:28 <Caelian|work> In file included from ../include/reiserfs_lib.h:14,
Dec 12 08:57:28 <Caelian|work>                  from includes.h:12,
Dec 12 08:57:29 <Caelian|work>                  from do_balan.c:20:
Dec 12 08:57:31 <Caelian|work> ../include/reiserfs_fs.h:41:27: error: asm/unaligned.h: No such file or directory
Dec 12 08:58:08 <Caelian|work> i assume it has something to do with the semi-recent change of kernel-version ... though i am not exactly sure how
Dec 12 09:02:01 <rwhitby> Caelian|work: have you reported the failure on #oe ?
Dec 12 09:02:14 <rwhitby> (as it's a standard OE package, which we don't modify at all)
Dec 12 09:02:29 <Caelian|work> not yet ... i wasn't sure if it was an OE or a SlugOS issue
Dec 12 09:03:05 <Caelian|work> like i said before .. i haven't had much time to really look at it yet
Dec 12 09:03:35 <Caelian|work> i'll drop by in #OE and see what i can find out there
Dec 12 09:04:05 <rwhitby> it could be a 2.6.17 to 2.6.19 thing.  dunno.
Dec 12 09:08:15 <Caelian|work> that's my idea as well ... it used to build on the 2.6.18 patched to a 2.6.19 kernel
Dec 12 09:08:21 <Caelian|work> but i think ever since 2.6.19 proper got imported ... it stopped building
Dec 12 09:10:12 <Caelian|work> (i ran a make clobber at the time though ... so i'm not sure it wasn't broken before that ... since it's been a while since i clobbered before that)
Dec 12 09:11:47 <rwhitby> check if that include file got changed in 2.6.19 or something
Dec 12 09:12:08 <rwhitby> if we can't solve it, we'll drop reiserfs
Dec 12 09:12:49 <likewise> See here: http://www.mail-archive.com/blfs-dev@linuxfromscratch.org/msg05931.html
Dec 12 09:13:05 <Caelian|work> /home/pascal/slug/debianslug/tmp/staging/arm-linux/kernel/include/asm/unaligned.h   does seem to exist
Dec 12 09:13:20 <likewise> Fix in second email in that thread
Dec 12 09:13:53 <Caelian|work> likewise: aah nice catch
Dec 12 09:14:18 <likewise> Caelian|work: I'll check the build without and with patch (custom ixp4xx build, openslug distro)
Dec 12 09:15:02 <Caelian|work> likewise: thanks ... debianslug is essentially the exact same thing .. just the opposite endianness
Dec 12 09:15:18 <Caelian|work> so if it works on openslug it should work on debianslug as well
Dec 12 09:17:34 <rwhitby> can one of you check in that patch if it works?
Dec 12 09:18:00 <Caelian|work> i don't have commit
Dec 12 09:18:11 <Caelian|work> so that would have to be likewise in that case :)
Dec 12 09:23:07 <likewise> Caelian|work: Is your development host kernel Linux, if so, what release?
Dec 12 09:23:46 <Caelian|work> Linux faramir.design-center.offis.uni-oldenburg.de 2.6.17-10-generic #2 SMP Fri Oct 13 18:45:35 UTC 2006 i686 GNU/Linux
Dec 12 09:24:07 <Caelian|work> i wish i could build under FreeBSD instead ... but that's an impossibility with OE
Dec 12 09:25:10 <likewise> the build works without a fix here, so reiserfsprogs must pickup the include from elsewhere... hmm I would like to known what happens exactly, although the patch looks ok, as it makes reiserfsprogs independent of the unaligned.h header
Dec 12 09:26:21 <Caelian|work> makes sense ... though it picking up the include elsewhere sounds like something that should not happen either
Dec 12 09:27:47 <likewise> ah, the linux header package, what release is in your /tmp/work?
Dec 12 09:28:03 <Caelian|work> now that i think about it though ... i would not be surprised i don't have any of the kernel-headers / kernel-sources packages installed on my build host ....
Dec 12 09:28:06 <Caelian|work> let me look
Dec 12 09:28:16 <Caelian|work> host /tmp/work ... or debianslug/temp/work ?
Dec 12 09:29:31 <Caelian|work> none
Dec 12 09:29:45 <Caelian|work> at least ... as far as i can tell
Dec 12 09:30:14 <Caelian|work> debianslug/tmp/work/ixp4xx-kernel-2.6.19-r1.605
Dec 12 09:30:48 <Caelian|work> is that what you meant ?
Dec 12 09:32:16 <likewise> linux-libc-headers-2.6.x rather.
Dec 12 09:32:27 <Caelian|work> linux-libc-headers-2.6.18-r1/
Dec 12 09:32:41 <likewise> Caelian|work: it seems that reiserfsprogs .bb is missing a DEPENDS on the kernel headers.
Dec 12 09:33:14 <likewise> Caelian|work: so besides the unaligned.h kernel header, we first need to make sure if it needs other kernel headers.
Dec 12 09:33:57 <Caelian|work> what do you want me to test ?
Dec 12 09:36:22 <likewise> nothing, it's just that this package needs more care than just this patch. I'll open a bug at #oe and try to solve it today.
Dec 12 09:36:35 <Caelian|work> allright :)
Dec 12 09:41:25 <likewise> follow the bug here: http://bugs.openembedded.org/show_bug.cgi?id=1675
Dec 12 09:42:58 <Caelian|work> will do :)
Dec 12 09:56:45 * rwhitby now has Debian etch natively running on two nas100d's :-)
Dec 12 09:57:00 <rwhitby> Now to get all the patches upstream, and get the installer to work.
Dec 12 11:27:38 <likewise> Caelian|work: it *seems* that the patch alone was enough.
Dec 12 12:03:57 <CIA-8> 03oleo * r4667 10optware/trunk/make/nano.mk: nano: disable wide char and ncursesw depends - on some platforms ncursesw were linked and not ncurses
Dec 12 12:06:38 <oleo> This egalisation may not be the best way, but it is a quick fix. Loosing wcha seems not to be so important .
Dec 13 01:48:41 <dumfrac> ~seen rwhitby
Dec 13 01:49:00 <rwhitby> dumfrac: gu
Dec 13 01:49:06 <dumfrac> hi
Dec 13 01:49:34 <dumfrac> sorry about that bug with the ethenet adpaters in Debian
Dec 13 01:49:42 <jbot> rwhitby is currently on #oe (7h 2m 2s) #openmoko (7h 2m 2s) #nslu2-linux (7h 2m 2s) #openjtag (7h 2m 2s) #openslug (7h 2m 2s), last said: 'Now to get all the patches upstream, and get the installer to work.'.
Dec 13 01:49:54 <rwhitby> dumfrac: why are you sorry?
Dec 13 01:50:08 <dumfrac> unfortunately, we need to get the installer working again on the slug to be able to debug the ethernet problem :-)
Dec 13 01:50:31 <dumfrac> I guess it is an expression, an introduction into the topic :-)
Dec 13 01:51:14 <rwhitby> dumfrac: I've been trying to build the installer for days :-)
Dec 13 01:51:33 <dumfrac> rwhitby: have you had any success ?
Dec 13 01:52:11 <dumfrac> rwhitby: the problem we are having is that apt keeps on segfaulting during the installation - this only happens on the slug, not other arm machines
Dec 13 01:52:48 <dumfrac> rwhitby: and it appears to be kernel specific, so we are now going through the kernel revisions to see what change to the kernel broke apt
Dec 13 01:53:24 <rwhitby> dumfrac: I have debian unstable running on the nas100d now from internal IDE disk, and am now building the installer there as root.
Dec 13 01:53:43 <dumfrac> rwhitby: for instance, 2.6.18-5 works (when last I tested it) but 2.6.18-6 doesn't work :-( I am busy testing 2.6.18 subversion revision 7815
Dec 13 01:54:02 <dumfrac> rwhitby: you are building the installer natively ?
Dec 13 01:54:09 <rwhitby> I am now, yes.
Dec 13 01:54:23 <dumfrac> how is it going ?
Dec 13 01:54:28 <rwhitby> (just bringing up the second nas100d now to act as the test machine)
Dec 13 01:54:52 <rwhitby> I was able to build the netboot once - now working on building the packages. haven't made any changes or tested anything yet.
Dec 13 01:55:21 <rwhitby> (getting all the dependencies installed for all the packages under d-i takes a while on a slug-like device)
Dec 13 01:55:22 <dumfrac> is the plan to get the nas100d support into etch ?
Dec 13 01:55:36 <rwhitby> no - post etch
Dec 13 01:55:56 <rwhitby> I think there's only two or three people in the world who care about it, since the nas100d is no EOL by Iomega.
Dec 13 01:56:09 <dumfrac> ah
Dec 13 01:56:11 <rwhitby> (but I have two of them)
Dec 13 01:56:53 <dumfrac> does Iomega have a replacement to which this work would apply ?
Dec 13 01:57:47 <rwhitby> I think the storcenter, but I think it's freescale pcc based.  Not sure if there is an ixp version.
Dec 13 01:58:04 <rwhitby> this work applies also to the dlink dsm-g600 rev A, but it's EOL too :-)
Dec 13 02:55:01 <[g2]> did you guys see the Gigabit D-link box ?
Dec 13 02:55:30 <[g2]> I think it was d-link
Dec 13 02:56:39 <[g2]> d-link DNS 323
Dec 13 02:58:16 <[g2]> runs linux source and build tools are out
**** ENDING LOGGING AT Wed Dec 13 02:59:57 2006