Date: Thu, 23 Sep 1999 11:40:53 -0400 (EDT) From: Zhihui Zhang <zzhang@cs.binghamton.edu> To: freebsd-hackers@freebsd.org Subject: NFS and RPC Message-ID: <Pine.GSO.3.96.990923101148.17856B-100000@sol.cs.binghamton.edu>
next in thread | raw e-mail | index | archive | help
NFS is said to be built on RPC. However, NFS daemons seldom uses RPC library calls before they enter into the kernel forever (The nfsd daemon only calls pmap_set(). The nfsiod daemon invokes no RPC call at all! The mountd daemon uses a lot of RPC calls, but it does not enter the kernel forever and it is not used during normal reads/writes). Once inside the kernel, the NFS daemons can not use RPC library any more, they must create/interprete RPC format messages themselves. My guess this is for performance reason and because there is no kernel-to-kernel RPC. Since the kernel part of NFS code does not use RPC library routines, why FreeBSD still conforms to the RPC format for NFS requests/replies? Is this for compatibily with other NFS servers/clients that are implemented entirely as user-level code and with RPC library routines? One more question is about how to assembly a RPC request from several mbufs? I notice that there is a check for 0x80000000 in the routine nfsrv_getstream() for the last fragment. But I assume that all mbufs linked together via their m_next field should consist of one RPC request. I also assume that the protocol layer has ripped off any TCP/IP headers from the mbufs before it pass the mbufs up to the socket layer. Any help is appreciated. -------------------------------------------------- Zhihui Zhang. Please visit http://www.freebsd.org -------------------------------------------------- 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?Pine.GSO.3.96.990923101148.17856B-100000>