On Mon, Jul 24, 2000 at 12:54:53AM +0200, Pavel Machek wrote:
> > > > Fortunately all UDMA transfers are protected by safe enough CRC to see
> > > > if a transfer fails. This way you can know your cable won't do UDMA/66
> > > > and switch to something slower. (UDMA/44 seems to work on most 40-wire
> > > > cables just fine). If that doesn't work, go even lower. Something like
> > > > modem autobauding ...
> > >
> > > If it'll work like modem autobauding here (connect with speed limit=14400
> > > and get 1.5Kbytes/second, connect with 33600 and spend 95% of time for
> > > retrains and thus transfer with 0.1Kbytes/second speed in the end) then
> > > better not...
> > >
> > > P.S. Sorry, could not resist...
> >
> > :)))))))) It'd be like this if you let UDMA/66 enabled on a 40-wire
> > cable. What I meant is that after a *single* CRC error the speed would be
> > lowered to the next lower step (UDMA 100, 66, 44, 33, 16) and if even
> > UDMA/16 doesn't work, then fall back to PIO.
>
> That seems wrong. If you have server capable of UDMA/66 and you run it
> for year or so, you are likely to find it in PIO mode. Not good.
No. You're not supposed to get CRC errors with UDMA. Not a single one.
If you do, it'd mean you'd get data corruption without the CRCs.
MWDMA/16 runs at the same speed as UDMA/16 and doesn't have CRC
protection. Many people are using MWDMA/16. If your expectations (there
will be a bit error now and then) were true, many people would see
filesystem corruption.
-- Vojtech Pavlik SuSE Labs- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Mon Jul 31 2000 - 21:00:16 EST