Linux kernel 2.6.19 ATA drive support boobytrap

Lennart Sorensen lsorense-1wCw9BSqJbv44Nm34jS7GywD8/FfD2ys at public.gmane.org
Mon Feb 19 16:57:08 UTC 2007


On Mon, Feb 19, 2007 at 12:08:48AM -0500, Evan Leibovitch wrote:
> Isn't there something wrong with this picture?
> 
> Serial ATA is the new(er) technology -- "parallel ATA" is just a new
> name for what we've always called IDE, whose support under Linux I
> always know to be mature and not at all experimental.

The IDE subsystem is mature (although has lots of bad error handling
issues and uses it's own device node naming system).  The libata PATA
support is new (mainly written by Alan Cox), and use the same libata
interface as current SATA drivers (some early SATA drivers were based on
the ide subsystem design).  So yes libata PATA support is new and
experimental, libata SATA is production code, and ide subsystem is
mature, but eventually headed for deprecation.

Advantages for the libata interface is a more consistent support for
ATAPI and other things, and the fact all disks will be /dev/sd* devices
no matter what controller they connect to.  You also won't have an
ide-scsi issue with the bugs it contained, as everyting will simple be
scsi style interface.  This is of course how BSD has done it for years.

--
Len Sorensen
--
The Toronto Linux Users Group.      Meetings: http://gtalug.org/
TLUG requests: Linux topics, No HTML, wrap text below 80 columns
How to UNSUBSCRIBE: http://gtalug.org/wiki/Mailing_lists





More information about the Legacy mailing list