Date: Tue, 2 Jun 1998 20:36:22 +0000 (GMT) From: Terry Lambert <tlambert@primenet.com> To: dyson@FreeBSD.ORG Cc: mike@dingo.cdrom.com, mike@smith.net.au, hackers@FreeBSD.ORG Subject: Re: kernfs/procfs questions... Message-ID: <199806022036.NAA06082@usr06.primenet.com> In-Reply-To: <199806021650.LAA02867@dyson.iquest.net> from "John S. Dyson" at Jun 2, 98 11:50:32 am
next in thread | previous in thread | raw e-mail | index | archive | help
> > Sure; but can't these sort of improvements be made to the methods for > > manipulating procfs nodes? What other drawbacks are there to the FS > > model? > > It is bogus that writing to a file "controls" something inside > the system, that is kind-of what SNMP is for. Now, if someone > wants a kernfs that is compatible with our sysctl, they should > be able to use the sysctl info to build the kernfs. I think the sysctl infrastructure implementation lacks sufficient dynamism. There is no concept of a subschema entry (for example) which would allow for runtime extension rather than linktime agregation (which is a much simpler problem). The non-dynamic portions of the framework, inherited from the 4.4 code base, are also quite problematic. Not that I advocate a kernfs... but a very dynamic procfs could "know" from the accessor what ABI it was being invoked from, if it had the necessary parametric hints (which it currently lacks). Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199806022036.NAA06082>