Date: Sat, 22 Dec 2007 07:37:32 +0200 From: Kostik Belousov <kostikbel@gmail.com> To: Alfred Perlstein <alfred@freebsd.org> Cc: freebsd-net@freebsd.org, freebsd-stable@freebsd.org Subject: Re: Packet loss every 30.999 seconds Message-ID: <20071222053732.GR57756@deviant.kiev.zoral.com.ua> In-Reply-To: <20071222002432.GK16982@elvis.mu.org> References: <D50B5BA8-5A80-4370-8F20-6B3A531C2E9B@eng.oar.net> <20071217102433.GQ25053@tnn.dglawrence.com> <CD187AD1-8712-418F-9F49-FA3407BA1AC7@eng.oar.net> <20071220011626.U928@besplex.bde.org> <814DB7A9-E64F-4BCA-A502-AB5A6E0297D3@eng.oar.net> <20071219171331.GH25053@tnn.dglawrence.com> <20071221200810.GY16982@elvis.mu.org> <20071221234347.GS25053@tnn.dglawrence.com> <20071222002432.GK16982@elvis.mu.org>
next in thread | previous in thread | raw e-mail | index | archive | help
--soWJpSPh+l8Y6Fy7 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Fri, Dec 21, 2007 at 04:24:32PM -0800, Alfred Perlstein wrote: > * David G Lawrence <dg@dglawrence.com> [071221 15:42] wrote: > > > > Unfortunately, the version of the patch that I sent out isn't go= ing to > > > > help your problem. It needs to yield at the top of the loop, but vp= isn't > > > > necessarily valid after the wakeup from the msleep. That's a proble= m that > > > > I'm having trouble figuring out a solution to - the solutions that = come > > > > to mind will all significantly increase the overhead of the loop. > > >=20 > > > I apologize for not reading the code as I am swamped, but a technique > > > that Matt Dillon used for bufs might work here. > > >=20 > > > Can you use a placeholder vnode as a place to restart the scan? > > > you might have to mark it special so that other threads/things > > > (getnewvnode()?) don't molest it, but it can provide for a convenient > > > restart point. > >=20 > > That was one of the solutions that I considered and rejected since it > > would significantly increase the overhead of the loop. > > The solution provided by Kostik Belousov that uses uio_yield looks l= ike > > a find solution. I intend to try it out on some servers RSN. >=20 > Out of curiosity's sake, why would it make the loop slower? one > would only add the placeholder when yielding, not for every iteration. Marker is already reinserted into the list on every iteration. --soWJpSPh+l8Y6Fy7 Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (FreeBSD) iD8DBQFHbKKcC3+MBN1Mb4gRAoT1AKCBmHGiMjE36UzMRadhMvV4puYuSwCfQby3 vGqKmgClkrEd3/x3ytBSUao= =HhMC -----END PGP SIGNATURE----- --soWJpSPh+l8Y6Fy7--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20071222053732.GR57756>