On Sun, Feb 02, 2003 at 01:04:57PM +1030, Greg 'groggy' Lehey wrote:
> On Saturday,  1 February 2003 at 23:22:13 +0100, Manuel Bouyer wrote:
> > On Sat, Feb 01, 2003 at 07:54:27PM +0100, Pavel Cahyna wrote:
> >>>
> >>> Right, this is a deficiency in how we handle layered disk I/O.  Right
> >>> now we are limited to MAXPHYS (64k) for each "disk".  If this "disk"
> >>
> >> Why this limit? Is this a heritage from PDP-11 with its (i think) 64kB address
> >> space?
> >
> > Some popular disk standart (IDE comes to mind, there may be others) don't
> > support transfers > 64k.
> 
> No, that's not correct:
> 
>   #  dd if=/dev/ad0s1c of=/dev/null bs=128k &
>   # iostat ad0 1
>         tty             ad0             cpu
>    tin tout  KB/t tps  MB/s  us ni sy in id
>    164  158  0.00   0  0.00   9  0  4  3 85
>      1  340 128.00 190 23.71   3  0  8 12 78
>      0   44 128.00 188 23.52   1  0  7  5 88
>      0  155 128.00 187 23.37   2  0  5  5 88
>      0  155 128.00 191 23.91   2  0  6  5 88
>      0  154 128.00 191 23.89   6  0  6  5 82
>      0  155 128.00 191 23.89   6  0  6  3 84

Sorry, I miss remembered. The limit for IDE is 128k, not 64 (and much highter
with the extended read/write commands).
There are devices which don't support more than 64k though, but I don't
remember which one(s).

-- 
Manuel Bouyer <bouyer@antioche.eu.org>
     NetBSD: 24 ans d'experience feront toujours la difference
--