Re: Switchover to CAM ATA?

On Apr 27, 2010, at 7:34 AM, Dag-Erling Smørgrav wrote:
> Maxim Sobolev <sobomax_at_FreeBSD.org> writes:
>> Richard Tector <richardtector_at_thekeelecentre.com> writes:
>>> Could I also add that the removal of ataraid would affect those
>>> users who dual-boot with Windows and rely on the psuedo-raid
>>> provided by most Intel chipsets to be able to share the same pair of
>>> disks.
>> Well, this won't be a problem if we have GEOM classes that can
>> understand metadata created by the ATA RAID BIOS(es).
> 
> Most pseudo-raid kit has nifty features like checksum offloading,
> composite writes etc. which can improve performance considerably.  You
> can't access those from GEOM.

If my "most" you mean "a small subset of vendors and products", then yes, you're correct.  For the vast remaining majority, all you get is a special BIOS that can do striping and mirroring at boot.  Sometimes that special BIOS requires a special hook in the controller chip, which is why you have ICHxxR vs ICHxx chips, for example.  All the 'R' means is that the silicon supports an external ROM.

Scott