Message-ID: <553A6A94.1010508@redhat.com>
Date: Fri, 24 Apr 2015 12:08:52 -0400
From: Rik van Riel <riel@redhat.com>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.4.0
MIME-Version: 1.0
To: Christoph Lameter <cl@linux.com>,
        "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
CC: Benjamin Herrenschmidt <benh@kernel.crashing.org>,
        Jerome Glisse <j.glisse@gmail.com>, linux-kernel@vger.kernel.org,
        linux-mm@kvack.org, jglisse@redhat.com, mgorman@suse.de,
        aarcange@redhat.com, airlied@redhat.com,
        aneesh.kumar@linux.vnet.ibm.com,
        Cameron Buschardt <cabuschardt@nvidia.com>,
        Mark Hairgrove <mhairgrove@nvidia.com>,
        Geoffrey Gerfin <ggerfin@nvidia.com>,
        John McKenna <jmckenna@nvidia.com>, akpm@linux-foundation.org
Subject: Re: Interacting with coherent memory on external devices
References: <20150421214445.GA29093@linux.vnet.ibm.com> <alpine.DEB.2.11.1504211839120.6294@gentwo.org> <1429663372.27410.75.camel@kernel.crashing.org> <20150422005757.GP5561@linux.vnet.ibm.com> <1429664686.27410.84.camel@kernel.crashing.org> <alpine.DEB.2.11.1504221020160.24979@gentwo.org> <20150422163135.GA4062@gmail.com> <alpine.DEB.2.11.1504221206080.25607@gentwo.org> <1429756456.4915.22.camel@kernel.crashing.org> <alpine.DEB.2.11.1504230925250.32297@gentwo.org> <20150423185240.GO5561@linux.vnet.ibm.com> <alpine.DEB.2.11.1504240929340.7582@gentwo.org>
In-Reply-To: <alpine.DEB.2.11.1504240929340.7582@gentwo.org>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1174
Lines: 28

On 04/24/2015 10:30 AM, Christoph Lameter wrote:
> On Thu, 23 Apr 2015, Paul E. McKenney wrote:
> 
>> If by "entire industry" you mean everyone who might want to use hardware
>> acceleration, for example, including mechanical computer-aided design,
>> I am skeptical.
> 
> The industry designs GPUs with super fast special ram and accellerators
> with special ram designed to do fast searches and you think you can demand page
> that stuff in from the main processor?

DRAM access latencies are a few hundred CPU cycles, but somehow
CPUs can still do computations at a fast speed, and we do not
require gigabytes of L2-cache-speed memory in the system.

It turns out the vast majority of programs have working sets,
and data access patterns where prefetching works satisfactorily.

With GPU calculations done transparently by libraries, and
largely hidden from programs, why would this be any different?

-- 
All rights reversed
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/