Subject: Re: change bus_space_barrier() semantic
To: None <tech-kern@netbsd.org>
From: Michael van Elst <mlelstv@serpens.de>
List: tech-kern
Date: 01/10/2006 20:43:00
ignatios@cs.uni-bonn.de (Ignatios Souvatzis) writes:

>> What about doing the opposite? Making all accesses synchronized
>> by default and allowing the driver an exception to postpone
>> the synchronization until after the last access.

>How would it do this? Define bus_space_write_N_foo_async() ?

Yes, modulo the name. Would save us from changing the drivers
that assume implicit barriers.

-- 
-- 
                                Michael van Elst
Internet: mlelstv@serpens.de
                                "A potential Snark may lurk in every tree."