Date: Tue, 27 Dec 2011 08:27:48 -0800 From: perryh@pluto.rain.com To: egrosbein@rdtc.ru Cc: iwc2010005@gmail.com, freebsd-net@freebsd.org, jhell@DataIX.net Subject: Re: Can we do perform a C style file Read/Write from within a ARP module Message-ID: <4ef9f204.c2ZP2f73gFwo7lUU%perryh@pluto.rain.com> In-Reply-To: <4EF975FC.5080604@rdtc.ru> References: <CA%2BDF=vy3s1mEGaCoQdZWaws%2BE0=D7JvNO6BRbFeJv6VDxS=JXA@mail.gmail.com> <20111226143421.GB17435@DataIX.net> <4ef9c41f.rC8YkQQx738CJ3EN%perryh@pluto.rain.com> <4EF975FC.5080604@rdtc.ru>
next in thread | previous in thread | raw e-mail | index | archive | help
Eugene Grosbein <egrosbein@rdtc.ru> wrote: > 27.12.2011 20:11, perryh@pluto.rain.com ?????: > > Jason Hellenthal <jhell@DataIX.net> wrote: > >> > >> See siftr(4). This module writes to a file. > > > > Is siftr(4) new since 8.1? > > HISTORY > SIFTR first appeared in FreeBSD 7.4 and FreeBSD 8.2. which explains why there's no manpage for it in 8.1 :) It turns out that siftr(4) does not directly manipulate its logfiles, instead using alq(9) which has been around since 5.0. To (partly) answer the original question, it looks as if alq(9) can handle opening/creating and writing to the file -- and siftr(4) will serve as an example of using alq(9) -- but on a brief perusal it does not look as if alq(9) includes a mechanism to read back the data. One kernel operation that does directly process data from files -- as opposed to passing the data to a userland process for processing -- is execve(2), which must examine the file header to identify the type of executable being invoked.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4ef9f204.c2ZP2f73gFwo7lUU%perryh>