Bug#239111: Grub is shockingly bad code

[Robert Millan]

On Mon, Jan 12, 2009 at 07:32:49PM +0100, Robert Millan wrote:
> 
> I (and upstream in general) believe that the only right way to rely on a
> hardcoded list of blocks that live inside a filesystem is _not to_.

Grmf.  I was making wrong assumptions.  This is not about block lists
(I still think block lists suck, but let's be fair...):

#ifdef GRUB_UTIL
          /* In the grub shell, access the Stage 2 via the OS filesystem
             service, if possible.  */
          if (stage2_os_file)
            {
[...]
              if (fwrite (stage2_buffer, 1, SECTOR_SIZE, fp)
                  != SECTOR_SIZE)
[...]
            }
          else
#endif /* GRUB_UTIL */
            {
              if (! devwrite (saved_sector - part_start, 1, stage2_buffer))
                goto fail;
            }

So we freeze the filesystem and afterwards try to write to it.  Not a good
idea...

#239111 initial report claims GRUB hangs before we added xfs_freeze.diff, but
according to what Rob says, not freezing makes it work for him.

Rob, can you test removing the patch completely?

-- 
Robert Millan

  The DRM opt-in fallacy: "Your data belongs to us. We will decide when (and
  how) you may access your data; but nobody's threatening your freedom: we
  still allow you to remove your data and not access it at all."