>  * 
>  * In general, using queue tagging can only help performance (unless
>  * you're doing something dumb like trying to use too many tags; the
>  * number of tags supported is a drive-dependent value).  If you consider
>  * the worst case of a drive's handling of tagged commands, it will
>  * process them in the order received.  That's equivalent to sending them
>  * down one at a time.  (OK, actually, if you send them down one at a
>  * time, the extra delay in sending them to the drive may get you some
>  * disksort() wins, but it's not clear that it would really help.)  In
>  * the best case, the drive can optimize the handling of the requests so
>  * that they're handled in the 'best' order.
>  * 
> 
> So, in other words, disksort() is now obsolete and should be replaced.

That's not what i said...


> Are there any drives newer than, say, two years old which don't handle
> tagging?

I dunno if there are drives that "don't handle queueing," but there
certainly are drives which support a very small number of tags.  (In
the limit, that would be '1', but i dunno whether tagged queueing is
optional or 'mandatory,' etc. in SCSI 2.)

> Should not the SCSI driver attempt to determine whether or not the drive
> handles tagging, and disable tagging for that drive if tagging fails?

The big issue is "how many tags can be used," and as far as i know,
there's no way to determine that in a standard way. (!!  I might be
wrong; it might be in a standard location in the mode pages, or
something, but i'm just not familiar enough with the SCSI spec to
know.)


chris