**** BEGIN LOGGING AT Mon Apr 15 02:59:57 2013
Apr 15 08:38:55 <ant_work> Jay7: I was surprised to see this cmdline passed by kexecboot
Apr 15 08:38:57 <ant_work> Kernel command line: root=/dev/sda1 rootfstype=ext2 rootwait mtdparts=sharpsl-ng
Apr 15 08:39:42 <ant_work> then I realized minicom messed up the line.... the g is the last letter (debug)
Apr 15 08:39:44 <ant_work> :)
Apr 15 08:40:37 <ant_work> I even googled for sharpsl-ng...
Apr 15 08:42:07 <bluelightning> ant_work: merged your patch, thanks
Apr 15 08:42:20 <bluelightning> morning btw
Apr 15 08:45:58 <ant_work> hi
Apr 15 08:47:01 <ant_work> strange failures on spitz reported by Jama...
Apr 15 08:47:27 <ant_work> ...but I could boot it yesterday     http://paste.debian.net/249559/
Apr 15 08:51:36 <ant_work> doh... By default, minicom will not wrap longs lines of text, which causes them to run off the edge of your terminal screen. To avoid this, use the '-w' command line option
Apr 15 08:52:08 <bluelightning> minicom is evil... I much prefer picocom
Apr 15 08:52:23 <ant_work> good to know
Apr 15 08:53:38 <ant_work> I just got rid of gtkterm and went with the first one in memory
Apr 15 08:55:49 <ant_work> about Zaurus, well now any Zaurus is booting 3.8 from MMC/SD
Apr 15 08:55:52 <ant_work> :/
Apr 15 08:56:00 <ant_work> pxa models:  pxa2xx-mci.0 supply vmmc not found, using dummy regulator
Apr 15 08:56:18 <ant_work> (it's me who introduced the dummy regulator...)
Apr 15 08:56:34 <ant_work> collie: need to import locomo_spi patch
Apr 15 08:57:10 <ant_work> better ask dromede :)
Apr 15 09:23:18 * Jay7 is using 'cu' instead of mini/picocom :)
Apr 15 09:23:24 <ant_work> bluelightning: wrt postinst, anything special building distroless defaults
Apr 15 09:23:31 <ant_work> Jay7: hi there
Apr 15 09:23:49 <Jay7> I leave my work today :)
Apr 15 09:24:03 <ant_work> fo a better one ofc...
Apr 15 09:24:08 <ant_work> ?
Apr 15 09:24:14 <Jay7> no :) just to be free :)
Apr 15 09:24:24 <Jay7> have some plans anyway :)
Apr 15 09:24:34 <ant_work> I see, good luck
Apr 15 09:25:28 <bluelightning> ant_work: there shouldn't be... are you seeing a problem then?
Apr 15 09:25:30 <Jay7> so I'll complete proposed tasks for kexecboot during May I think
Apr 15 09:25:36 <bluelightning> hi Jay7
Apr 15 09:25:41 <Jay7> hi bluelightning
Apr 15 09:35:58 <ant_work> bluelightning: no, I didn't see evident issues
Apr 15 09:36:14 <ant_work> but it was short testing, indeed
Apr 15 09:36:26 <ant_work> fwiw there is one issue, though
Apr 15 09:36:38 <bluelightning> ant_work: what's that?
Apr 15 09:36:46 <ant_work> if you change the kernel provider from linux-yocto to linux or viceversa
Apr 15 09:37:16 <ant_work> there are oddities with the modules. JaMa seems knowing
Apr 15 09:38:12 <ant_work> JaMa	ant_home: there are some wrong dependencies between modules
Apr 15 09:38:19 <ant_work> JaMa	ant_home: they depend only on kernel-3.8.4-yocto-standard, not on required modules like they did with linux.inc based kernels (and maybe other l-y versions too)
Apr 15 09:40:12 <bluelightning> er that would be a bug
Apr 15 09:41:06 <ant_work> I did that while testing kernel (linu.inc, no-yocto is much faster to build)
Apr 15 09:41:24 <ant_work> but I beautifully ignored all warnings
Apr 15 09:42:19 <ant_work> but the image is created incorrectly..wrong modules iirc
Apr 15 09:42:50 <ant_work> sorry, I was just testing kernels...can't be more precise
Apr 15 09:43:23 <ant_work> no issues building from scratch with either linux-yocto or linux
Apr 15 09:46:32 <ant_work> bluelightning: here it is
Apr 15 09:46:35 <ant_work> 16:02.15	JaMa	interesting, meta-handheld has: conf/machine/include/zaurus.inc:MACHINE_ESSENTIAL_EXTRA_RDEPENDS_append = " kernel-base kernel-image"
Apr 15 09:46:35 <ant_work> 16:02.28	*** join/#oe andyross (~andy@c-67-171-188-207.hsd1.or.comcast.net)
Apr 15 09:46:35 <ant_work> 16:02.38	JaMa	but packagegroup.*boot is not rebuild when new kernel version is built
Apr 15 09:46:35 <ant_work> 16:03.06	JaMa	causing do_rootfs to fail
Apr 15 09:46:35 <ant_work> 16:03.07	JaMa	|  * satisfy_dependencies_for: Cannot satisfy the following dependencies for packagegroup-core-boot:
Apr 15 09:46:37 <ant_work> 16:03.10	JaMa	|  *    kernel-3.4.0 *  kernel-image-3.4.0 *
Apr 15 09:48:07 <bluelightning> ant_work: ah :/ this is a generic problem we have with allarch unfortunately
Apr 15 09:48:21 <bluelightning> RP and I have been discussing it this morning actually
Apr 15 09:48:35 <bluelightning> still not sure what the proper solution is :/
Apr 15 09:48:37 <ant_work> ok, is on some to-do. flushed here ;)
Apr 15 09:48:54 <bluelightning> unfortunately the package manager can't be relied upon to support virtual dependencies
Apr 15 09:49:54 <ant_work> but I suppose an upgrade from l-y_3.2 to l-y_3.8 should run flawlessy
Apr 15 09:50:41 <bluelightning> it ought to yes
Apr 15 10:02:41 <ant_work> Jay7: JaMa asked
Apr 15 10:02:43 <ant_work> > Could you prepare some easy way to enable debug in all 3 components
Apr 15 10:02:43 <ant_work> > (linux-kexecboot, kexecboot itself, linux)?
Apr 15 10:02:50 <ant_work> I see little point in that
Apr 15 10:03:08 <ant_work> I would answer, forget linux-kexecboot, it's working
Apr 15 10:03:34 <ant_work> just add 'debug' or 'ignore_loglevel' in boot.cfg of testing kernel
Apr 15 10:27:59 <Jay7> ant_work: I don't know how to set this at once
Apr 15 10:28:15 <Jay7> some global knob?
Apr 15 10:28:29 <ant_work> CMDLINE_DEBUG is a weak assignment
Apr 15 10:28:42 <ant_work> http://cgit.openembedded.org/meta-openembedded/tree/meta-initramfs/recipes-bsp/kexecboot/kexecboot-cfg_0.1.bb
Apr 15 10:28:58 <ant_work> I always flash image then edit boot.cfg...
Apr 15 10:33:48 <ant_work> bluelightning: oh, I forgot to send oe-core patch for importing cramfs
Apr 15 10:34:41 <ant_work> I has a small chat with RP and it sorted out cramfs is the only helper missing in oe-core for the creation of images listed in IMAGE_TYPES
Apr 15 10:34:49 <ant_work> not ethe recipe in meta-oe is broken
Apr 15 10:36:06 <ant_work> http://cgit.openembedded.org/meta-openembedded/tree/meta-oe/recipes-support/cramfs/cramfs_1.1.bb
Apr 15 10:36:15 <ant_work>  -SRC_URI = "http://sourceforge.net/projects/cramfs/files/cramfs/1.1/${PN}-${PV}.tar.gz"
Apr 15 10:36:26 <ant_work>  +SRC_URI = "http://sourceforge.net/projects/cramfs/files/cramfs/1.1/cramfs-${PV}.tar.gz"
Apr 15 10:36:41 <ant_work> oherwise it tries to fetch cramfs-native
Apr 15 10:37:11 <ant_work> bluelightning: this is just because collie can boot core-image-minimal.
Apr 15 10:37:21 <ant_work> I need to mak one RO
Apr 15 10:37:40 <ant_work> brb
Apr 15 10:39:49 <bluelightning> ant_work: should use BPN
Apr 15 10:40:23 <bluelightning> although I wonder if there is any point, "cramfs" would be fine
Apr 15 15:06:34 <ant_work> bluelightning: wrt release, do you think it's too late now to merge cramfs in oe-core?
Apr 15 15:13:14 <bluelightning> ant_work: probably yes
Apr 15 15:16:58 <ant_work> np
Apr 15 15:17:27 <ant_work> btw, which image should I build to get a RO small image?
Apr 15 15:19:45 <ant_work> is it just some setting in local.conf?
Apr 15 15:22:19 <bluelightning> ant_work: read-only-rootfs should be in IMAGE_FEATURES
Apr 15 15:22:33 <bluelightning> core-image-minimal works
Apr 15 15:23:16 <ant_work> ok, thx. core-image-base is just a bit too large for collie...just a few kb
Apr 15 15:28:13 <bluelightning> yeah core-image-base pulls in everything based upon MACHINE_FEATURES and DISTRO_FEATURES
Apr 15 15:28:26 <bluelightning> and the defaults for those variables enable a fair amount of stuff
**** ENDING LOGGING AT Tue Apr 16 02:59:58 2013