Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 06 Sep 2010 19:33:21 +0300
From:      Andriy Gapon <avg@freebsd.org>
To:        Jeremy Chadwick <freebsd@jdc.parodius.com>
Cc:        freebsd-current@freebsd.org, freebsd-stable@freebsd.org
Subject:   Re: patch for topology detection of Intel CPUs
Message-ID:  <4C8517D1.3050603@freebsd.org>
In-Reply-To: <20100906162230.GA1444@icarus.home.lan>
References:  <AANLkTinYUz0V%2B2nSWBMYLf2fL8HnUQ-fvXT0q-5WY4bb@mail.gmail.com> <4C782D3B.6020407@icyb.net.ua> <201008271743.29393.jkim@FreeBSD.org> <4C7835E6.6070309@icyb.net.ua> <4C7A2782.5040009@freebsd.org> <4C84DBE6.5010809@freebsd.org> <20100906122325.GA78727@icarus.home.lan> <4C84E4E1.8010709@freebsd.org> <20100906131203.GA80251@icarus.home.lan> <4C84EC62.6020205@freebsd.org> <20100906162230.GA1444@icarus.home.lan>

next in thread | previous in thread | raw e-mail | index | archive | help
on 06/09/2010 19:22 Jeremy Chadwick said the following:
> On Mon, Sep 06, 2010 at 04:28:02PM +0300, Andriy Gapon wrote:
>> on 06/09/2010 16:12 Jeremy Chadwick said the following:
>>> Great, thanks!  I'll be testing this out on two separate systems, both
>>> RELENG_8:
>>>
>>> - Supermicro X7SBA     + Intel C2D E8400 (stepping 10)
>>> - Supermicro X7SBL-LN2 + Intel C2D E6600 (stepping 6)
>>>
>>> I'll make sure to provide what the topology looks like before and after.
>>> Is CPU-relevant dmesg output sufficient?
>>
>> If you mean something like the below, then yes.  Thanks!
>> [...]
> 
> All done.  Good news (I think): there's no difference in the CPU-related
> topology on either system with your patch, aside from kernel build date.
> The topologies are still detected correctly.  In case you want them:
> 

Thanks a lot for the test!

[test results snipped]

> All other systems I have are C2D and C2Q-based, but I can't easily test
> on those given their production roles.  If there's a particular Intel
> processor family/model you're interested in, let me know and I can dig
> around to see if I have access to one.

No particular models in mind.
If you have systems with more complex topologies, like multiple physical packages
or HTT enabled, I will be interested in seeing test results for those.
Thanks again.
-- 
Andriy Gapon



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4C8517D1.3050603>