Bug#491376: recipe for BIOS-based boot on gpt

[Robert Millan <rmh@aybabtu.com>]

On Fri, Jul 25, 2008 at 05:03:01PM +0200, Robert Millan wrote:
> On Sun, Jul 20, 2008 at 09:08:56AM +0200, Robert Millan wrote:
> > 
> > Ok, so here are two patches.  One of them using recipes-x86, and the other
> > simply modifiing recipes.
> > 
> > The first one expects recipes-x86 has been created previously via "svn copy".
> 
> So is either of these okay?

Hi folks,

I don't want to be annoying, but I really think this should be sorted out in
time for lenny.  Not doing so will mean that boot will be less robust; for
example if one does:

  mkdir /tmp/boot
  mv /boot/* /tmp/boot/
  cp -a /tmp/boot/* /boot/

on a system where core.img was embedded, nothing will happen.  OTOH if you do
the same on a system where core.img was referenced via blocklists, boot will
start crashing as soon as the (now free) blocks that held the old core.img
are overwritten.

-- 
Robert Millan

<GPLv2> I know my rights; I want my phone call!
<DRM> What good is a phone call… if you are unable to speak?
(as seen on /.)