From mboxrd@z Thu Jan 1 00:00:00 1970 From: Nick Piggin Date: Wed, 29 Mar 2006 02:35:40 +0000 Subject: Re: Fix unlock_buffer() to work the same way as bit_unlock() Message-Id: <4429F27C.6020404@yahoo.com.au> List-Id: References: <200603281853.k2SIrGg28290@unix-os.sc.intel.com> <4429ADBC.50507@free.fr> <4429CFCA.7010201@yahoo.com.au> In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Christoph Lameter Cc: Zoltan Menyhart , "Chen, Kenneth W" , akpm@osdl.org, linux-kernel@vger.kernel.org, linux-ia64@vger.kernel.org Christoph Lameter wrote: >On Wed, 29 Mar 2006, Nick Piggin wrote: > > >>However, I think it might be reaonsable to use bit lock operations for >>in places like page lock and buffer lock (ie. with acquire and relese >>semantics). It improves ia64 without harming other architectures, and >>also makes the code more expressive. >> > >How would be express the acquire and release semantics? > Hmm, not sure. Maybe a few new bitops with _lock / _unlock postfixes? For page lock and buffer lock we'd just need test_and_set_bit_lock, clear_bit_unlock, smp_mb__after_clear_bit_unlock. I don't know, _for_lock might be a better name. But it's getting long. -- Send instant messages to your online friends http://au.messenger.yahoo.com