xen-devel.lists.xenproject.org archive mirror
 help / color / mirror / Atom feed
From: He Chen <he.chen@linux.intel.com>
To: Andrew Cooper <andrew.cooper3@citrix.com>
Cc: wei.liu2@citrix.com, ian.campbell@citrix.com,
	stefano.stabellini@eu.citrix.com, ian.jackson@eu.citrix.com,
	jbeulich@suse.com, xen-devel@lists.xenproject.org, keir@xen.org
Subject: Re: [PATCH 0/5] Intel Code/Data Prioritization(CDP) feature enabling
Date: Sun, 6 Sep 2015 14:49:32 +0800	[thread overview]
Message-ID: <20150906064932.GA29010@HE> (raw)
In-Reply-To: <55E6E6C1.5050906@citrix.com>

On Wed, Sep 02, 2015 at 01:08:33PM +0100, Andrew Cooper wrote:
> On 02/09/15 09:27, He Chen wrote:
> > Hi all,
> >
> > Code/Data Prioritization(CDP) is offered in Intel Broadwell and later server
> > platforms, which is an extension of CAT. CDP enables isolation and separate
> > prioritization of code and data fetches to the L3 cache in a software
> > configurable manner, which can enable workload prioritization and tuning of
> > cache capacity to the characteristics of the workload. CDP extends Cache
> > Allocation Technology (CAT) by providing separate code and data capacity bit
> > masks(CBM) per Class of Service (COS). CDP is used on VM basis in the Xen
> > implementation.
> >
> > More information about CDP, please refer to Intel SDM, Volumn 3, section 17.16
> > http://www.intel.com/content/dam/www/public/us/en/documents/manuals/64-ia-32-architectures-software-developer-manual-325462.pdf
> >
> > This patch series enables CDP feature in Xen based on CAT code, including
> > extending CBM operation functions and introducing new commands to enable/disable
> > CDP dynamically. For all the changes, please see in each patch.
> >
> > This patchset has been tested on Intel Broadwell server platform.
> >
> > To make this patchset better, any comment or suggestion is welcomed, I would
> > really appreciate it.
> 
> I have taken a look at patches 1-3.  For the most part, it looks good.
> 
> The main point I have is on patch 2, as to whether it is sensible to
> permit enabling/disabling cdp at runtime.  I suggest that it is not
> sensible, and should be a command line parameter instead.
> 
> If this is agreed as ok going forwards, patches 3 through 5 should
> become rather more simple.
> 
> ~Andrew

Thanks for your patient review and valuable suggestions.

About permitting enabling/disabling CDP at runtime, I agree with you to
use command line parameter instead, it really makes code simple and
reliable.

For caution's sake, hardware support confingure CDP dynamically at any
point during normal system operation according to Intel SDM (see section
17.16.2), and that is why I wrote patch 2.

Anyway, since there is few cases to change CDP at runtime, I think it is
better to make this a boot-time parameter. I would resend v2 patch soon
and thanks again~

      parent reply	other threads:[~2015-09-06  6:49 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-09-02  8:27 [PATCH 0/5] Intel Code/Data Prioritization(CDP) feature enabling He Chen
2015-09-02  8:27 ` [PATCH 1/5] x86: detect Intel CDP feature He Chen
2015-09-02 11:12   ` Andrew Cooper
2015-09-02  8:27 ` [PATCH 2/5] x86: Support enable/disable CDP dynamically and get CDP status He Chen
2015-09-02 11:39   ` Andrew Cooper
2015-09-02 14:07     ` Jan Beulich
2015-09-02  8:28 ` [PATCH 3/5] x86: add domctl cmd to set/get CDP code/data CBM He Chen
2015-09-02 11:59   ` Andrew Cooper
2015-09-06  7:15     ` He Chen
2015-09-06 16:29       ` Andrew Cooper
2015-09-02  8:28 ` [PATCH 4/5] tools: add tools support for Intel CDP He Chen
2015-09-02 13:32   ` Wei Liu
2015-09-02  8:28 ` [PATCH 5/5] docs: add document to introduce CDP command He Chen
2015-09-02 13:32   ` Wei Liu
2015-09-02 12:08 ` [PATCH 0/5] Intel Code/Data Prioritization(CDP) feature enabling Andrew Cooper
2015-09-06  1:18   ` Chao Peng
2015-09-06  6:49   ` He Chen [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20150906064932.GA29010@HE \
    --to=he.chen@linux.intel.com \
    --cc=andrew.cooper3@citrix.com \
    --cc=ian.campbell@citrix.com \
    --cc=ian.jackson@eu.citrix.com \
    --cc=jbeulich@suse.com \
    --cc=keir@xen.org \
    --cc=stefano.stabellini@eu.citrix.com \
    --cc=wei.liu2@citrix.com \
    --cc=xen-devel@lists.xenproject.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).