[meta-ti] AM335x EVM and core-image-sato
Denys Dmytriyenko
denys at ti.com
Fri Oct 18 19:39:54 PDT 2013
On Fri, Oct 18, 2013 at 06:09:23PM -0700, Rudolf Streif wrote:
> Well, I rebuilt with a newer version of meta-ti
> (master, b3df09902da0d2469a02e9975d404e26c4c5ec0e). Now the /dev/fb0 device
> is in the root file system. However:
>
> - Kernel image is now zImage but u-boot environment still looks for
> uImage; work around that and then
u-boot has changed the preferred default from uImage to zImage somewhere
between 2013.10-rc4 and the final 2013.10 release. The kernel now builds
zImage by default and there's corresponding u-boot update:
http://git.yoctoproject.org/cgit/cgit.cgi/meta-ti/commit/?id=3f3bad888972776b4740f7210e709b74362631f5
Looks like you hit it at the wrong time... Sorry, that sometimes happens
when there are multiple components being actively developed and the API
between them changes...
> - kernel panic - not syncing: Attempting to kill init! exitcode=0x0b.
Have you tried the latest update to 3.12-rc5? That combination seems to be
building and booting fine, according to our automated reports:
http://article.gmane.org/gmane.linux.embedded.yocto.meta-arago/1245
> One step forward, two steps back. Pulled again, and this time build fails
> because branch ti-linux-312.y is not available.
The branch was and is called ti-linux-3.12.y and it hasn't changed since the
beginning of 3.12 integration and is still there:
http://git.ti.com/gitweb/?p=ti-linux-kernel/ti-linux-kernel.git;a=shortlog;h=refs/heads/ti-linux-3.12.y
> What branch of meta-ti and what commit is working?
Both master and dylan are currently in sync and always are, when it comes to
3.12-rc5 kernel and 2013.10 u-boot. Other recipes can differ between master
and dylan, but the integration trees are the same. The above link to automated
build and test report used dylan branch of meta-ti and the latest commit ID of
f61008a20340373247ebfd8f381e377b21f11afe. If you want to use meta-ti master,
then the corresponding commit ID is c11504664990f007c3903d349e7cc135a904f8e2.
--
Denys
More information about the meta-ti
mailing list