Date: Tue, 1 Apr 1997 12:46:06 -0700 (MST) From: Terry Lambert <terry@lambert.org> To: nate@mt.sri.com (Nate Williams) Cc: proff@suburbia.net, terry@lambert.org, hackers@freebsd.org Subject: Re: Internal clock Message-ID: <199704011946.MAA11874@phaeton.artisoft.com> In-Reply-To: <199704011920.MAA04841@rocky.mt.sri.com> from "Nate Williams" at Apr 1, 97 12:20:39 pm
next in thread | previous in thread | raw e-mail | index | archive | help
> > The sad reality is if these things are not incorporated in -current > > then they fall by as the original authors move onto other projects. > > Then adding the code into the tree is a 'bad thing', since it becomes > unsupported. If no-one is willing to incorporate/support the code, then > it shouldn't be incorporated. Better that it be lost forever? Code shouldn't need a hell of a lot of maintenance, if the interfaces for plugging the code in are fairly static and well enough designed that they can remain that way. Seems to me that the issue is the fluidity of the kernel interfeaces, not the module code, that is at fault. Define a spanning set, and access it via macros, and the underlying implementation can change as much as you want without damaging the utlity of the unmaintained code. Network interfaces are a good example of one place where this should be happening. The RT scheduling question that started this particular thread is another. Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199704011946.MAA11874>