cvs commit: src/sys/dev/ata ata-all.c ata-all.h ata-chipset.cata-dma.c ata-pci.c ata-pci.h

[Wilko Bulte]

On Tue, Apr 08, 2003 at 07:15:04PM +0200, Dag-Erling Smorgrav wrote:
> Nate Lawson <nate at root.org> writes:
> > Tagged queuing is orthogonal to caching.  TQ allows multiple transactions
> > to be passed to the drive, allowing it to order and aggregate the requests
> > as it thinks best.  TQ can be done with no cache because all the drive is
> > tracking is handles to transactions (typically 16 bit ids).  Think of it
> > as more like a pipeline than a cache.
> 
> It can't be done without cache, because you can't queue and reorder
> transactions without some sort of cache.

But you could do read buffering (to avoid the C word) only, so 
no write cache which has the volatility issues which have been discussed
over and over again in combination with ATA drives. I still don't get
it why ATA drives cannot be built to the same spec as SCSI drives
as far as write caching etc are concerned 

Probably cost ;) ATA is for el-cheapo stuff. Serial ATA might see new
uses, say in arrays.

W/

-- 
|   / o / /_  _   		wilko at FreeBSD.org
|/|/ / / /(  (_)  Bulte