I have successfully built various angstrom kernel and root filesystem images using bitbake, but I still feel like I’m not fully understanding things. I’ve been reading the various documents and wikis, but so far I haven’t found the answer to the following questions. If anyone could provide an answer or a lead to finding the answer, I’d appreciate it.
-
As far as I can tell, after I do a bitbake, like “bitbake console-image”, the results are left in directories under the “tmp” subdir, and they appear to be intermixed with lots of intermediate build files. Is there a way to extract just the final product into a clean tree, or am I just not looking in the right place?
-
If I am building a different kernel (like MontaVista), using ‘make’ instead of ‘bitbake’, I can configure the kernel using “make gconfig” to edit the .config file. Is there a similar mechanism using bitbake, or do I need to edit the config files (which I think are stored as ‘defconfig’) by hand?
-
Is there an automatic way to extract the kernel source from bitbake into a tree that is buildable via the standard kernel build tools using “make”, or is that not something that is designed to be possible?
I apologize if any of this stuff is obvious and I’m just missing it, but after days of reading, I haven’t found the answers.
Thanks again for any help,
Brett