Re: LSF Papers online?

[Jeff Garzik <jeff@garzik.org>]

Bartlomiej Zolnierkiewicz wrote:
> On Tuesday 14 April 2009 12:14:00 Jeff Garzik wrote:


> The SATA features that needed SCSI infrastructure came 2 years later.
> 
>> Moving libata out of SCSI is now a long term, far off goal.  A goal that 
>> implies many intermediate steps, cleanups to block, libata, IDE, SCSI 
>> and other block drivers.
> 
> "far off"?
> 
> The fact that it is much harder to do nowadays than in 2004-2005 (without
> ATAPI support, PATA support and heavy dependence on SCSI infrastructure)
> is only _your_ fault.

Of course it is.  Use of SCSI driver infrastructure was a sound 
technical decision, I'll happily defend.  Key reasons SCSI core was used:

* ATA-SCSI convergence was clear when libata began.  Time has proven 
this true:

ATAPI was always SCSI-like.  SAS is plug-compatible with SATA [for some 
SAS plugs], and SAS transmits SATA frames from SAS expanders and SATA 
port multipliers.  T10 and T13 standards committees actively 
collaborate.  SCSI even has a specification, SAT, that describes how to 
best co-mingle ATA with SCSI.

* SCSI driver infrastructure was the only one advanced enough to support 
controller hotplug, device hotplug, and all sorts of queueing contortions.

* SCSI was the only infrastructure that _guaranteed_ it would work with 
existing installers and distros.  For users, there is a clear level of 
difference in support between /dev/hdXX, /dev/sdXX, and every other 
block device in the kernel.

SCSI had a higher Just Works(tm) value at the time.

	Jeff