Date: Thu, 7 Jun 2007 13:42:55 -0500 From: Astrodog <astrodog@gmail.com> To: "Chuck Swiger" <cswiger@mac.com> Cc: freebsd-current@freebsd.org, Abdullah Ibn Hamad Al-Marri <almarrie@gmail.com> Subject: Re: Why not remove polling(4) from 7.0? Message-ID: <2fd864e0706071142u127b275ahda831db2751a810f@mail.gmail.com> In-Reply-To: <DDC35B19-1873-4BC4-8817-C08C9A8232AA@mac.com> References: <499c70c0706070210v39f7016hbd80e9780902e992@mail.gmail.com> <20070607093027.GA4784@heff.fud.org.nz> <499c70c0706070236x28d781e6yb8ba4c8ccd251372@mail.gmail.com> <29EFA5CA-6232-45AA-A10D-0A45BB3E2100@mac.com> <499c70c0706071118l6410dc5dqce32b86fd919de3@mail.gmail.com> <DDC35B19-1873-4BC4-8817-C08C9A8232AA@mac.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On 6/7/07, Chuck Swiger <cswiger@mac.com> wrote: > > On Jun 7, 2007, at 11:18 AM, Abdullah Ibn Hamad Al-Marri wrote: > [ ... ] > > So it's recommended to use polling if I run pf and deals with DDoS? > > polling tends to handle a DDoS attack well compared with the default > interrupt-driven approach. Whether you are using pf or some other > firewall doesn't make any particular difference. > > -- > -Chuck > As an aside, in the rare cases where the extra system load is worth it, polling combined with a high value for HZ can also be a good way to insure consistent low latency, even under heavy load.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?2fd864e0706071142u127b275ahda831db2751a810f>