Re: [PATCH scsi-misc-2.6 04/08] scsi: remove meaningless volatilequalifiers from structure definitions

[Jeff Garzik]

James Bottomley wrote:
> On Wed, 2005-03-23 at 11:14 +0900, Tejun Heo wrote:
>
> > 	scsi_device->device_busy, Scsi_Host->host_busy and
> > 	->host_failed have volatile qualifiers, but the qualifiers
> > 	don't serve any purpose.  Kill them.  While at it, protect
> > 	->host_failed update in scsi_error for consistency and clarity.
>
>
> Well ... the data here is volatile so what you're advocating is a move
> away from a volatile variable model to a protected variable one ... did
> you audit all users of both of these to make sure we have protection on
> all of them?  It looks like the sata strategy handlers would still rely
> on the volatile data.

volatile is almost always (a) buggy, or (b) hiding bugs.  At the very 
least, barriers are usually needed.

Almost every case really wants to be inside a spinlock, or atomic_t, or 
similarly protected.

Specifically for SATA, I am making the presumption that SCSI is smart 
enough not to mess with host_failed until my error handler completes.

	Jeff


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/