**** BEGIN LOGGING AT Sat Jul 05 02:59:59 2014
**** BEGIN LOGGING AT Sat Jul 05 04:42:45 2014
Jul 05 10:25:48 <ant__> Jay7, one recent commit to kernel.bbclass has changed the symlink
Jul 05 10:26:03 <ant__> I don't know who's right now...
Jul 05 10:27:23 <ant__> root@mizar:/var/tmp# ls -al boot-bad/
Jul 05 10:27:23 <ant__> total 2076
Jul 05 10:27:23 <ant__> drwxr-xr-x 2 root root    4096 lug  5 01:38 .
Jul 05 10:27:23 <ant__> drwxrwxrwt 4 root root    4096 lug  5 12:26 ..
Jul 05 10:27:23 <ant__> -rw-r--r-- 1 root root     831 lug  5 01:24 boot.cfg
Jul 05 10:27:24 <ant__> -rw-r--r-- 1 root root    1322 lug  5 01:24 icon.xpm
Jul 05 10:27:26 <ant__> lrwxrwxrwx 1 root root      34 lug  5 12:26 zImage -> /boot/zImage-3.14.5-yocto-standard
Jul 05 10:27:28 <ant__> -rw-r--r-- 1 root root 2106832 lug  5 01:20 zImage-3.14.5-yocto-standard
Jul 05 10:27:30 <ant__> root@mizar:/var/tmp# ls -al boot-good/
Jul 05 10:27:32 <ant__> total 2076
Jul 05 10:27:34 <ant__> drwxr-xr-x 2 root root    4096 lug  5 12:20 .
Jul 05 10:27:36 <ant__> drwxrwxrwt 4 root root    4096 lug  5 12:26 ..
Jul 05 10:27:38 <ant__> -rw-r--r-- 1 root root     831 lug  5 01:24 boot.cfg
Jul 05 10:27:40 <ant__> -rw-r--r-- 1 root root    1322 lug  5 01:24 icon.xpm
Jul 05 10:27:42 <ant__> lrwxrwxrwx 1 root root      28 lug  5 12:26 zImage -> zImage-3.14.5-yocto-standard
Jul 05 10:27:44 <ant__> -rw-r--r-- 1 root root 2106848 lug  5 12:16 zImage-3.14.5-yocto-standard
Jul 05 10:27:48 <ant__> root@mizar:/var/tmp#
Jul 05 10:39:44 <ant__> maybe one extra iterative symlink to boot...
Jul 05 10:40:01 <ant__> was it for lilo ? it was eons ago..
Jul 05 12:32:02 <ant__> ok, I'll add it to the cpio
Jul 05 19:11:56 <Jay7> hm
Jul 05 19:11:59 <Jay7> evening
Jul 05 19:12:30 <Jay7> i.e. now symlink have absolute path inside
Jul 05 19:13:07 <Jay7> which is inaccessible when partition is mounted to /mnt
Jul 05 19:14:25 <ant__> hi
Jul 05 19:14:33 <ant__> I've sent a mail to oe-core
Jul 05 19:14:43 <ant__> we could also mount in /boot
Jul 05 19:15:38 <ant__> I think that *any* boot-script would fail
Jul 05 19:15:54 <ant__> not only kexecboot
Jul 05 19:16:14 <ant__> never tried dracut but we have it
Jul 05 19:17:02 <Jay7> yeah, it may introduce some unexpected side effects like our case
Jul 05 19:17:22 <ant__> do you remember long ago the symlink in /boot -> . for lilo?
Jul 05 19:17:30 <ant__> maybe we need smthg like
Jul 05 19:17:31 <ant__> pkg_postinst_${PN} () {
Jul 05 19:17:32 <ant__>     ln -sf ${bindir}/kexecboot $D/init
Jul 05 19:17:32 <ant__>     ln -sf /mnt $D/mnt/boot
Jul 05 19:17:32 <ant__> }
Jul 05 19:19:01 <ant__> but the /mnt in the initramfs is 'overwritten' once we mount any partition over it
Jul 05 19:19:26 <ant__> so *if* we have to do, we should add it to the rootfs I think
Jul 05 19:25:03 <ant__> http://sources.debian.net/src/debianutils/4.4/installkernel
Jul 05 19:26:29 <ant__> that's 'make install'
Jul 05 19:26:48 <Jay7> I've used FreeBSD that times )
Jul 05 19:27:04 <ant__> fair enough
Jul 05 19:27:11 <Jay7> when I've switched to Linux there was grub everywhere )
Jul 05 19:27:47 <ant__> it was just
Jul 05 19:27:50 <ant__> cd /boot
Jul 05 19:28:05 <Jay7> but I understand that point
**** ENDING LOGGING AT Sun Jul 06 02:59:59 2014