Date: Mon, 17 Apr 2006 12:22:17 -0400 From: Kris Kennaway <kris@obsecurity.org> To: Surer Dink <surerlistmail@gmail.com> Cc: smp@freebsd.org, current@freebsd.org, Kris Kennaway <kris@obsecurity.org> Subject: Re: Anomalous performance increase from mutex profiling Message-ID: <20060417162216.GA90886@xor.obsecurity.org> In-Reply-To: <b00a10c30604170054r57d13768u4aeb79e91c436d51@mail.gmail.com> References: <b00a10c30604170054r57d13768u4aeb79e91c436d51@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
--BOKacYhQ+x31HxR3 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Mon, Apr 17, 2006 at 03:54:07AM -0400, Surer Dink wrote: > On Mon, 17 Apr 2006, Kris Kennaway wrote: >=20 > > On Mon, Apr 17, 2006 at 01:14:40AM -0400, Kris Kennaway wrote: > > > >> * Our best guess is that mutex profiling is doing something that > >> reduces contention on this very heavily contended mutex (unp), but I'd > >> like to know what is happening precisely so I can maybe make use of > >> it. > >> > >> Can anyone think of what may be happening that I've missed? > > > > I think it is just doing effectively the same thing as my exponential > > spin backoff patch, namely introducing delays with the effect of > > reducing common memory accesses. When I turn the maximum spin backoff > > limit *way* up (from 1600 to 51200) I get performance that slightly > > exceeds what I see from mutex profiling alone (adding mutex profiling > > again on top of this gives a small further increase, but only a few % > > and so probably achievable by further increasing the backoff limit). > > > > A limit of 51200 is not an appropriate default since it penalizes the > > common case of light to moderate contention. The point is that here > > basically all 12 CPUs are spinning on a single lock > > (kern/uipc_usrreq.c:308), so it's massively over-contended and all we > > can do is mitigate the effects of this. > > > > On this system, the maximum supersmack performance (3700 queries/sec) > > comes when there are only 6 clients, so (as jasone eloquently put it) > > with 10 clients the difference between 2300 queries/sec (with absurdly > > high backoff limits or mutex profiling) and 1450/sec (with reasonable > > backoff limits) is the difference between "slow" and "ass slow". >=20 > Please excuse if this is a stupid question - but might using MCS or > QOLB locks in this situation be useful? What are they? Kris --BOKacYhQ+x31HxR3 Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.3 (FreeBSD) iD8DBQFEQ8C4Wry0BWjoQKURAvt5AKCMRE+4/1wrRTGDt0LTHcHXKmtldQCg3XLw ZiMsQve4FuiR6QoL+N9ClhI= =WCKf -----END PGP SIGNATURE----- --BOKacYhQ+x31HxR3--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20060417162216.GA90886>