Re: another questions about early initrd issues

["Vegard Nossum" <vegard.nossum@xxxxxxxxx>]

Hi,

On Dec 4, 2007 2:59 PM, Robert P. J. Day <rpjday@xxxxxxxxxxxxxx> wrote:
> all right, let me make sure i understand what's happening here.  if
> you look at the tail end of init/initramfs.c, you can see in the
> routine populate_rootfs() where that "checking if image is
> initramfs..." message is being printed.
>
> but *above* that, there has *already* been a call to
> unpack_to_rootfs() to unpack the in-kernel cpio initramfs (the very
> first thing in that routine, in fact).
>
> so the question remains -- what was *that* unpacked into?  are you
> saying it was an initial /dev/ram?  if so, when was that created
> during the boot process?  is this question making any sense?

I hope I am understanding the question correctly :) If you look at the
function unpack_to_rootfs(), it makes repeated calls to
write_buffer(). This is the "destination" of the unpack.
write_buffer() is a state machine that calls different functions
depend on the current state. These functions call vfs functions like
sys_open() with O_CREAT (in do_name()) and sys_write() with the
unpacked data (in do_copy()). I do not know the state of the vfs at
this moment, however.

> rday

Vegard

--
To unsubscribe from this list: send an email with
"unsubscribe kernelnewbies" to ecartis@xxxxxxxxxxxx
Please read the FAQ at http://kernelnewbies.org/FAQ