Re: Kernel v4.7-rc5 - performance degradation upto 40% after disabling and re-enabling a core

From	Peter Zijlstra <peterz@infradead.org>
Newsgroups	linux.kernel
Subject	Re: Kernel v4.7-rc5 - performance degradation upto 40% after disabling and re-enabling a core
Date	2016-06-29 12:00 +0200
Message-ID	<rPkrM-2AI-33@gated-at.bofh.it> (permalink)
References	<rPasp-4UA-3@gated-at.bofh.it> <rPjm1-1VD-3@gated-at.bofh.it> <rPkrM-2AI-35@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw

On Wed, Jun 29, 2016 at 11:47:56AM +0200, Jirka Hladky wrote:
> Hi Peter,
> 
> I think Cluster on Die technology was introduced in Haswell generation. The
> server I'm using is equipped with 4x Intel E5-4610 v2 (Ivy Bridge). I have
> double checked the BIOS and there is no cluster on die setting.

Oh right, that's E5v3..

> The authors of the paper have reported the issue on AMD Bulldozer CPU which
> also does not have COD technology.

The Opteron 6272, which they use, is an Interlagos, that has something
similar in that each package contains two nodes.

And their patch touches exactly that part of the x86 topo setup, the
match_die() && !same_node() condition, IOW same package, different node.

That's not a path an Intel chip would trigger without COD support.

Back to linux.kernel | Previous | Next — Previous in thread | Find similar | Unroll thread

Thread

Kernel v4.7-rc5 - performance degradation upto 40% after disabling  and re-enabling a core Jirka Hladky <jhladky@redhat.com> - 2016-06-29 01:20 +0200
  Re: Kernel v4.7-rc5 - performance degradation upto 40% after  disabling and re-enabling a core Peter Zijlstra <peterz@infradead.org> - 2016-06-29 10:50 +0200
    Re: Kernel v4.7-rc5 - performance degradation upto 40% after  disabling and re-enabling a core Peter Zijlstra <peterz@infradead.org> - 2016-06-29 12:00 +0200

csiph-web