Date: Fri, 07 Apr 2006 21:17:41 -0700 From: Frank Mayhar <frank@exit.com> To: Sam Leffler <sam@errno.com> Cc: arch@freebsd.org Subject: Re: ifnet cloning changes Message-ID: <1144469861.21943.13.camel@realtime.exit.com> In-Reply-To: <44370CD6.9090006@errno.com> References: <44370CD6.9090006@errno.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, 2006-04-07 at 18:07 -0700, Sam Leffler wrote: > modifies the SIOCIFCREATE api in the kernel so that an opaque parameter > block can be passed down to ifnet cloning routines (actually the user > address is passed down and the caller must do the copyin since it alone > knows the size of the parameter block). Isn't that a recipe for potential breakage? Why not make the api receive a vector, i.e. buffer/length? The "caller" will know the actual layout of the parameter block but other kernel routines will at least be able to deal with it, since they'll know the size. As an aside, I just recently got this reference regarding "self-describing" data: http://www.multicians.org/thvv/marking.html Interesting. -- Frank Mayhar frank@exit.com http://www.exit.com/ Exit Consulting http://www.gpsclock.com/ http://www.exit.com/blog/frank/
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?1144469861.21943.13.camel>