Subject: Re: Bus-Master DMA error: missing interrupt
To: Christoph Badura <email@example.com>
From: Manuel Bouyer <firstname.lastname@example.org>
Date: 01/04/1999 15:10:53
On Dec 31, Christoph Badura wrote
> If you take the time to read the sources, you'll find that there doesn't
> appear to be an "official semantic".
> Anyway, SCSI and ATAPI disks now exactly how many blocks are addressable
> and can be queried for that information. There is no reason to allow a
> process to do IO past that limit. Especially if it is know that this
> leads to lossage.
This is true for recent IDE disks too.
I still maintain this check should not be in the disk drivers, but at a higther
level (possibly with a hook in each driver to tell the higther level how big
the disk is)
Manuel Bouyer, LIP6, Universite Paris VI. Manuel.Bouyer@lip6.fr