On Tue, Jun 04, 2002 at 01:35:53PM -0700, Bill Studenmund wrote:
> On 4 Jun 2002 eeh@netbsd.org wrote:
> 
> > | As I understand what Chuck did, all i/os have to be in terms of underlying
> > | disk sectors. So you won't get two writes to sub-sector parts because you
> > | can't write less than a sector.
> >
> > Actually, I thought it was the other way around.  All IOPs are done in terms
> > of DEV_BSIZE.  If they were all done in terms of the device's block size we
> > wouldn't be having any of this conversation and 2KB devices would just work.
> 
> As I understand it, all IOPs are expressed in units of DEV_BSIZE, but they
> have to map to an integer number of real sectors.
> 
> So for instance on a 1k device when DEV_BSIZE is 512, all i/o has to be in
> an even number of DEV_BSIZE sectors, and start on even DEV_BSIZE sector
> boundries.


yup, that's it exactly.

-Chuck