From mboxrd@z Thu Jan 1 00:00:00 1970
Received: from eggs.gnu.org ([2001:4830:134:3::10]:56525)
by lists.gnu.org with esmtp (Exim 4.71)
(envelope-from
) id 1YxrOX-00025N-TB
for qemu-devel@nongnu.org; Thu, 28 May 2015 02:34:22 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
(envelope-from ) id 1YxrOT-0000Be-MI
for qemu-devel@nongnu.org; Thu, 28 May 2015 02:34:21 -0400
Received: from mailout2.w1.samsung.com ([210.118.77.12]:49815)
by eggs.gnu.org with esmtp (Exim 4.71)
(envelope-from ) id 1YxrOT-0000BM-GD
for qemu-devel@nongnu.org; Thu, 28 May 2015 02:34:17 -0400
Received: from eucpsbgm2.samsung.com (unknown [203.254.199.245])
by mailout2.w1.samsung.com
(Oracle Communications Messaging Server 7.0.5.31.0 64bit (built May 5
2014)) with ESMTP id <0NP100EZCRL2ED10@mailout2.w1.samsung.com> for
qemu-devel@nongnu.org; Thu, 28 May 2015 07:34:14 +0100 (BST)
From: Pavel Fedin
References: <1430921082-16779-1-git-send-email-shlomopongratz@gmail.com>
<1430921082-16779-2-git-send-email-shlomopongratz@gmail.com>
<20150527181202.2ef26ca6@nial.brq.redhat.com>
<021a01d098a7$7fd6ac40$7f8404c0$@samsung.com>
<556666F5.1020701@linaro.org>
In-reply-to: <556666F5.1020701@linaro.org>
Date: Thu, 28 May 2015 09:34:13 +0300
Message-id: <007a01d09910$507ed440$f17c7cc0$@samsung.com>
MIME-version: 1.0
Content-type: text/plain; charset=us-ascii
Content-transfer-encoding: 7bit
Content-language: ru
Subject: Re: [Qemu-devel] [PATCH RFC V2 1/4] Use Aff1 with mpidr
List-Id:
List-Unsubscribe: ,
List-Archive:
List-Post:
List-Help:
List-Subscribe: ,
To: 'Shannon Zhao' , 'Igor Mammedov' , shlomopongratz@gmail.com
Cc: peter.maydell@linaro.org, qemu-devel@nongnu.org, 'Shlomo Pongratz'
Hi!
> IIUC, your implementation also uses 8 cpus per cluster which is not
> consistent with CPU spec.
Yes, i have inherited the value from Shlomo's implementation. To tell the truth i don't
have time to read through all the docs. And i don't know in which PDF i should look for
this info. But, my implementation is very easy to tune. Just change the definition.
Actually it could be even made into static variable so that machines could set it before
instantiating CPUs.
And my real HW has 16 CPUs per cluster, but i thought it's implementation-specific. Isn't
it ?
> I guess we should give people time to do the respin or other things.
I am giving it. A week has passed. Actually one of goals of my project here at Samsung is
to upstream our changes, that's why i worry about it.
Kind regards,
Pavel Fedin
Expert Engineer
Samsung Electronics Research center Russia