On Tue, 25 Jun 2002, Trevin Beattie wrote:

> >DEV_BSIZE from sectors doesn't mean that "db" needs to mean one or the
> >other.
>
> Then what, pray tell, would it mean?  It can't be both at the same time!

I meant to say that we got to choose which one it was, and our choice was
not forced one way or the other.  My apologies for my imperfect use of my
native tongue....

> required those units.  But the lfs code seemed to be just the opposite;
> using fsbtodb where DEV_BSIZE was required and something else when sectors

In sys/ufs/lfs I count 18 times where fsbtodb() is used to convert from
"fsb" to DEV_BSIZE (as a setup for bread(), VOP_BWRITE, or VOP_STRATEGY),
once where I'm not sure why we bother to convert (in lfs_fits), and *zero*
times to convert to sectors.  Likewise with dbtofsb()  (though there add
decoding bi_daddr elements from BLOCK_INFO objects, and results from
ufs_bmaparray()).

(Now, note that I am assuming that newfs_lfs should be changed to base
this constant on DEV_BSIZE, not on d_secsize as it does currently.)

> I'll get to in RSN. :)

Cool, looking forward to it :^)
						Konrad Schroder
						perseant@hhhh.org