James Bottomley wrote:
SCSI needs this for our scheme to avoid executing generic device release calls from interrupt context (SCSI patch using this to follow).
Shouldn't we rather fix the SCSI low level drivers or SCSI transport layers to trigger device releases only from process context? (Instead of SCSI core transparently falling back to a workqueue, that is.)
I know only one of these drivers, hence I don't know which are actually affected and how much work that would be.
-- Stefan Richter -=====-=-==- --=- -=--- http://arcgraph.de/sr/ - : send the line "unsubscribe linux-scsi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
