On 5/16/06, Randy McMurchy <[EMAIL PROTECTED]> wrote:

But couldn't you just run it make sure it worked? (as if a bunch of
mknod commands aren't going to work?) :-)

Would you really know it worked, though?  Presumably, if udev was
working, all the necessary device nodes already exist.  Unless you
started from an empty /dev, how would you know that the script did
what it was supposed to do?  I suppose you could run the script on a
different directory and check what devices it created.  DIY does this
in the chroot:

http://www.diy-linux.org/x86-reference-build/chroot.html#c-MAKEDEV

I'm just trying to do what I can so that we don't have instructions
to reboot the machine somewhere in our book. Linux/Unix machines
simply shouldn't have to be rebooted. All I'm asking is that we
explore every possible avenue before putting the Windows fix-all
in our book.

This is a very different situation.  Windows asks you to reboot after
any silly change to the registry.  This change affects how your
hardware interfaces to the kernel.  I'd want to know if my system
could bootstrap correctly since that is precisely the time when a
static device creation script is needed.

If you never plan to reboot your system, then why do you need to test
device node creation?  They're already created unless your system is
broken.

--
Dan
--
http://linuxfromscratch.org/mailman/listinfo/blfs-dev
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page

Reply via email to