From mboxrd@z Thu Jan  1 00:00:00 1970
From: cl@linux.com (Christoph Lameter)
Date: Fri, 10 Dec 2010 15:45:15 -0600 (CST)
Subject: [BUG] 2.6.37-rc3 massive interactivity regression on ARM
In-Reply-To: <1292016165.2746.14.camel@edumazet-laptop>
References: <20101208142814.GE9777@n2100.arm.linux.org.uk>
	<1291851079-27061-1-git-send-email-venki@google.com>
	<1291899120.29292.7.camel@twins>
	<AANLkTikd3X4pLs5De-LO6=XD+JqUz3-Ku7YM8+0-oK8t@mail.gmail.com>
	<1291917330.6803.7.camel@twins>
	<AANLkTin+eO_jPceSOHL4e7vqd+tf0ZCMHJ-USoz8aZL1@mail.gmail.com>
	<1291920939.6803.38.camel@twins>
	<AANLkTim+v21RpWobdw6exEbazDJwm4JjemPVB2+3PuRJ@mail.gmail.com>
	<1291936593.13513.3.camel@laptop>
	<AANLkTi=_qf8NAgj2HHMP0qG-ibnH-726NzoXbxd-bS1N@mail.gmail.com>
	<1291975704.6803.59.camel@twins> <1291987065.6803.151.camel@twins>
	<1291987635.6803.161.camel@twins> <1291988866.6803.171.camel@twins>
	<1292001500.3580.268.camel@edumazet-laptop>
	<1292003346.13513.30.camel@laptop>
	<1292004859.3580.387.camel@edumazet-laptop>
	<alpine.DEB.2.00.1012101236300.13986@router.home>
	<1292006788.13513.43.camel@laptop>
	<alpine.DEB.2.00.1012101344570.13986@router.home>
	<1292011644.13513.61.camel@laptop>
	<alpine.DEB.2.00.1012101410571.13986@router.home>
	<1292013590.2746.2.camel@edumazet-laptop>
	<1292014165.2746.9.camel@edumazet-laptop>
	<alpine.DEB.2.00.1012101505250.13986@router.home>
	<1292016165.2746.14.camel@edumazet-laptop>
Message-ID: <alpine.DEB.2.00.1012101533400.13986@router.home>
To: linux-arm-kernel@lists.infradead.org
List-Id: linux-arm-kernel.lists.infradead.org

On Fri, 10 Dec 2010, Eric Dumazet wrote:

> > There is certainly a major difference in that execution of a stream of
> > instructions on the same cpu is guaranteed to have a coherent view of
> > the data. That is not affected by interrupts etc.
> >
>
> We dont care of interrupts. We care of doing a transaction over a
> complex set of data, that cannot be done using an atomic op (or we need
> a spinlock/mutex/rwlock), and should not because of performance.

So what? The cpu gives you incoherent view of data somewhere when only
processing data from a single cpu?

If you have remote data accesses (loop summing the data?) and you have to
be concerned about data coherence then you CANNOT use this_cpu_ops since
they are not guaranteed to be coherent to other processors.