From mboxrd@z Thu Jan 1 00:00:00 1970 From: Dario Faggioli Subject: Re: [PATCH v5 01/24] docs: create L2 Cache Allocation Technology (CAT) feature document Date: Wed, 18 Jan 2017 10:11:15 +0100 Message-ID: <1484730675.7492.66.camel@citrix.com> References: <1484704967-5609-1-git-send-email-yi.y.sun@linux.intel.com> <1484704967-5609-2-git-send-email-yi.y.sun@linux.intel.com> Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="===============5665156825098617726==" Return-path: Received: from mail6.bemta5.messagelabs.com ([195.245.231.135]) by lists.xenproject.org with esmtp (Exim 4.84_2) (envelope-from ) id 1cTmH9-0003ge-1R for xen-devel@lists.xenproject.org; Wed, 18 Jan 2017 09:11:27 +0000 In-Reply-To: <1484704967-5609-2-git-send-email-yi.y.sun@linux.intel.com> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Errors-To: xen-devel-bounces@lists.xen.org Sender: "Xen-devel" To: Yi Sun , xen-devel@lists.xenproject.org Cc: wei.liu2@citrix.com, he.chen@linux.intel.com, andrew.cooper3@citrix.com, konrad.wilk@oracle.com, ian.jackson@eu.citrix.com, mengxu@cis.upenn.edu, jbeulich@suse.com, chao.p.peng@linux.intel.com List-Id: xen-devel@lists.xenproject.org --===============5665156825098617726== Content-Type: multipart/signed; micalg=pgp-sha256; protocol="application/pgp-signature"; boundary="=-eTll5stZxBijpj+bjkYI" --=-eTll5stZxBijpj+bjkYI Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable On Wed, 2017-01-18 at 10:02 +0800, Yi Sun wrote: > This patch creates L2 CAT feature document in doc/features/. > It describes details of L2 CAT. >=20 > Signed-off-by: Yi Sun > --- > Hey, it is very very useful to put _RIGHT_HERE_ a summary of what changed, within, this patch, wrt the previous version. That helps reviewers quite a bit, especially in making sure that you hav considered and addressed comments made during the previous iterations. > =C2=A0docs/features/intel_psr_l2_cat.pandoc | 347 > ++++++++++++++++++++++++++++++++++ > =C2=A01 file changed, 347 insertions(+) > =C2=A0create mode 100644 docs/features/intel_psr_l2_cat.pandoc > --- /dev/null > +++ b/docs/features/intel_psr_l2_cat.pandoc > @@ -0,0 +1,347 @@ > +% Intel L2 Cache Allocation Technology (L2 CAT) Feature > +% Revision 1.0 > + > +\clearpage > + > +# Basics > + > +---------------- ------------------------------------------------- > --- > +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0Status: **Tech Pre= view** > + > +Architecture(s): Intel x86 > + > +=C2=A0=C2=A0=C2=A0Component(s): Hypervisor, toolstack > + > +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0Hardware: Atom codename Goldmo= nt and beyond CPUs > +---------------- ------------------------------------------------- > --- > + > +# Overview > + > +L2 CAT allows an OS or Hypervisor/VMM to control allocation of a > +CPU's shared L2 cache based on application priority or Class of > Service > +(COS). Each CLOS is configured using capacity bitmasks (CBM) which > +represent cache capacity and indicate the degree of overlap and > +isolation between classes. Once L2 CAT is configured, the processor > +allows access to portions of L2 cache according to the established > +class of service. > + Well, considering that even here in the overview, acronyms are used, that are then defined in the section below, it would look better to me to move the Terminology section up (and let it be the first one). > +## Terminology > + > +* CAT=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0Cache Allocat= ion Technology > +* CBM=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0Capacity BitM= asks > +* CDP=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0Code and Data= Prioritization > +* COS/CLOS=C2=A0=C2=A0=C2=A0=C2=A0Class of Service > +* MSRs=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0Machine Specific R= egisters > +* PSR=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0Intel Platfor= m Shared Resource > +* VMM=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0Virtual Machi= ne Monitor > So, the term 'hypervisor' is (must be!) quite a well known one, for anyone reading docs in this directory. And, in the docs we already have around in here, it is far more common, AFAICT, to call an hypervisor 'hypervisor', rather than 'Virtual Machine Monitor'. Therefore, I don't think there is the need to define the term 'VMM' in this section, and, at the same time, I'd replace occurrences of 'Hypervisor/VMM' to just 'hypervisor' (the only one of which appear to be the one above, in the Overview). > +# User details > + > +* Feature Enabling: > + > +=C2=A0=C2=A0Add "psr=3Dcat" to boot line parameter to enable all support= ed level > CAT > +=C2=A0=C2=A0features. > + > +* xl interfaces: > + > +=C2=A0=C2=A01. `psr-cat-show [OPTIONS] domain-id`: > + > +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0Show domain L2 or L3 CAT CBM. > + > +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0New option `-l` is added. > +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0`-l2`: Show cbm for L2 cache. > +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0`-l3`: Show cbm for L3 cache. > + > +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0If neither `-l2` nor `-l3` is given, show = both of them. If any > one > +=C2=A0=C2=A0=C2=A0=C2=A0=C2=A0is not supported, will print error info. > + I actually think the best behavior would be: =C2=A0- if -lX is specified, and LX is not supported =3D=3D> print error; =C2=A0- if no -l is specified =3D=3D> print info about the supported levels= . (See comment on patch 21.) > +## The relationship between L2 CAT and L3 CAT/CDP > + > +L2 CAT is independent of L3 CAT/CDP, which means L2 CAT would be > enabled > +while L3 CAT/CDP is disabled, or L2 CAT and L3 CAT/CDP are all > enabled. > + I find 'would be enabled' and 'are all enabled' a bit confusing. Maybe: "which means L2 cat can be enabled while L3 CAT/CDP is disabled, or L2 CAT and L3 CAT/CDP can be all enabled" ? Regards, Dario --=20 <> (Raistlin Majere) ----------------------------------------------------------------- Dario Faggioli, Ph.D, http://about.me/dario.faggioli Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK) --=-eTll5stZxBijpj+bjkYI Content-Type: application/pgp-signature; name="signature.asc" Content-Description: This is a digitally signed message part Content-Transfer-Encoding: 7bit -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQIcBAABCAAGBQJYfzEzAAoJEBZCeImluHPucqYQANP8W/wXVkx6FhdFtht1r3Fj CzDfFpkTpXpbO9u9ElHVclg28AtncFtdLhBiXRJx1ZYpjFt3+VscPq2oH0/fI+FU q4lGel43+pqrSo+v7kxrbUgVjPAgnnT++dxLIe766uJutBliYqgdmhbue3CUW/3V PnEnq97O6HMgAxUQmjFKc1801zaDnAMz+4S4pbYgml4pryKQBbrSBn2nwVsW78Rm 4pwrtqIpGvJMiX/EHd5xU14Em9lRmEl95fCLTVAciEI8Zv0Q+qCvwf3AMUVtD65x PEDl8PTRTMUMwcqNeg6T4FD4TLs0u8Hyg3xtHOW7rGtcyMYCHh2xNLzUv340CI1K Nfrjg6isClIctPyAnppLkUoQY+/dEmL50w5/8aC7/j0jMLjX/eSkJ7ViJBkH9JJM g4l8mReBgTAlCg/Jjc6X9PhKtz8kvrg3xqjsGFyxUpHalFHpLYL0apJWXojmk8to f1nlFcVZz1xtLP0J2MDMg0LHfLZdykzLG2AZsQnR4yF0FJr1kwLJHiZBMYjNPOuh hi/7ILzcH1q2ttAszHP7zWgnWfLAXxCDx85ySwd5Qi5lP3jGdKwAVBr3kKPbtB2Z Th+IgFHmNbBJy8Rh2g9P9zarQ7AIRkR+oL/LVXNOPQ7Hb+oKAxQKxhFCVqB1wIaY O/alrUgS3XIxO/pg0+vt =SmUG -----END PGP SIGNATURE----- --=-eTll5stZxBijpj+bjkYI-- --===============5665156825098617726== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: base64 Content-Disposition: inline X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX18KWGVuLWRldmVs IG1haWxpbmcgbGlzdApYZW4tZGV2ZWxAbGlzdHMueGVuLm9yZwpodHRwczovL2xpc3RzLnhlbi5v cmcveGVuLWRldmVsCg== --===============5665156825098617726==--