From mboxrd@z Thu Jan  1 00:00:00 1970
From: paulmck@linux.vnet.ibm.com (Paul E. McKenney)
Date: Wed, 21 Sep 2011 16:42:42 -0700
Subject: [ltt-dev] [URCU PATCH] cmm: provide lightweight rmb/wmb on PPC
In-Reply-To: <4E78C1C5.2080001@redhat.com>
References: <1316502769-576-1-git-send-email-pbonzini@redhat.com>
	<20110920163125.GA31802@Krystal> <4E78C1C5.2080001@redhat.com>
Message-ID: <20110921234242.GG2394@linux.vnet.ibm.com>

On Tue, Sep 20, 2011 at 06:39:33PM +0200, Paolo Bonzini wrote:
> On 09/20/2011 06:31 PM, Mathieu Desnoyers wrote:
> >I don't think lwsync orders non-cacheable memory operations. Therefore,
> >is it the right choice for cmm_rmb/cmm_wmb ?
> 
> I think you're right. "eieio;lwsync" is good for rmb, lwsync is good
> for wmb/smp_rmb/smp_wmb.

But eieio;lwsync won't order non-cacheable memory operations against
cacheable memory operations.  This means that your MMIO accesses can
slip out of your lock-based critical section, which is a very bad
thing indeed.

For general-purpose use, I strongly recommend full "sync" for mb(),
rmb(), and wmb().  For memory-only use, the developer should not
be using these, but smp_mb(), smp_rmb(), and smp_wmb() instead.

							Thanx, Paul