Date: Sat, 9 Jun 2012 12:07:40 +0300 From: Konstantin Belousov <kostikbel@gmail.com> To: Mikolaj Golub <trociny@freebsd.org> Cc: Alan Cox <alc@freebsd.org>, Wojciech Puchar <wojtek@wojtek.tensor.gdynia.pl>, Robert Watson <rwatson@freebsd.org>, Ivan Voras <ivoras@freebsd.org>, freebsd-hackers@freebsd.org Subject: Re: SuperPages utilization survey Message-ID: <20120609090740.GL85127@deviant.kiev.zoral.com.ua> In-Reply-To: <86obosluvk.fsf@kopusha.home.net> References: <jqac8f$6mn$1@dough.gmane.org> <alpine.BSF.2.00.1206011429160.20357@wojtek.tensor.gdynia.pl> <CAF-QHFXsi3SaF-rmFK-qM89LfnStKwRWN1-HRM4yg4Bt_SFAOg@mail.gmail.com> <86sje5kkl6.fsf@kopusha.home.net> <20120609083822.GJ85127@deviant.kiev.zoral.com.ua> <86obosluvk.fsf@kopusha.home.net>
next in thread | previous in thread | raw e-mail | index | archive | help
--vkzoT2Bc9VsM/9ev Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Sat, Jun 09, 2012 at 12:03:43PM +0300, Mikolaj Golub wrote: >=20 > On Sat, 9 Jun 2012 11:38:22 +0300 Konstantin Belousov wrote: >=20 > KB> On Sat, Jun 09, 2012 at 10:31:17AM +0300, Mikolaj Golub wrote: > >>=20 > >> On Fri, 1 Jun 2012 14:54:48 +0200 Ivan Voras wrote: > >>=20 > >> IV> On 1 June 2012 14:35, Wojciech Puchar <wojtek@wojtek.tensor.gdyn= ia.pl> wrote: > >> >>> http://people.freebsd.org/~ivoras/stuff/spsurvey.py > >>=20 > >> ... > >>=20 > >> IV> If anyone posts more data, I'll analyse it. I'm more worried abo= ut the > >> IV> granularity of procstat, where it marks the entire region if a s= ingle > >> IV> superpage exists in it - it means any such analysis is only > >> IV> approximate. > >>=20 > >> Here is a patch (for kernel and procstat) that allows to see amount o= f pages > >> mapped to superpages. > >>=20 > >> http://people.freebsd.org/~trociny/procstat-superpages.cnt.1.patch > >>=20 > >> Not sure it is useful enough to be committed. >=20 > KB> Superpage aggregates mappings for several normal-sized pages. > KB> As a consequence, when you iterate over small pages in > KB> sysctl_kern_proc_vmmap(), you account each superpage as many time as > KB> much constituent small pages it contains. >=20 > This is exactly what my intention was to count: how much memory are handl= ed by > superpages (using normal-sized page as a measurement unit), not amount of > superpages. And I think this is what Ivan wanted to know. Do you think it= is > better to return number of superpages? >=20 Well, if I see a report informing me that some 2M region contains 512 super pages, how should I interpret it ? For me, it is only one superpage (mappin= g) that can be created in one 2M region. --vkzoT2Bc9VsM/9ev Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.12 (FreeBSD) iEYEARECAAYFAk/TElwACgkQC3+MBN1Mb4jaTgCdH1mYx2sqVHLz4D8yon9TSpFX rGoAoNDIlr8KQGNDXLL/2nLZROLqgNM8 =LXsu -----END PGP SIGNATURE----- --vkzoT2Bc9VsM/9ev--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20120609090740.GL85127>