Hi,

> > Judging from a short glimpse into sd.c and wd.c it looks like write-enabling
> > the label or not doesn't make any difference at all, if you access the disk
> > through the raw partition.  The only thing it does is prevent write access
> > to the label sector through any other partition than the raw one.
> 
> This needs to be fixed, then ... but should'nt this be handled by a common
> (possibly MI) function, rather than in each disk driver ?

Hmm, at the very minimum the check for the raw partition should probably
be placed in the bounds_check_with_label function and ripped out of the
various disk drivers.

BTW, the disk drivers seem to support the illusion that they can have
different partitioning schemes from the standard for a port (note the
#define	WDPART in the wd driver for example).  However, redefining those
away from the current simple reference to DISKPART will not work, as e.g.
the bounds_check_with_label that is called from this driver will select
the partition with the generic scheme.  Either the WDPART etc. macros
should be removed and DISKPART used instead, or at least those macros
should be accompanied with an appropriate comment.

Ciao,
Wolfgang
-- 
ws@TooLs.DE     (Wolfgang Solfrank, TooLs GmbH) 	+49-228-985800