Path: utzoo!attcan!uunet!lll-winken!lll-tis!helios.ee.lbl.gov!pasteur!ucbvax!NORUNIT.BITNET!HANCHE From: HANCHE@NORUNIT.BITNET Newsgroups: comp.protocols.tcp-ip.ibmpc Subject: (none) Message-ID: <8807221503.aa12894@Louie.UDEL.EDU> Date: 23 Jul 88 09:14:11 GMT Sender: daemon@ucbvax.BERKELEY.EDU Organization: The Internet Lines: 46 X-Unparsable-Date: Fri, 22 Jul 88 19:16:26 ECT Date: 22 July 1988, 19:14:45 ECT From: Harald Hanche-Olsen +47-7-593515 HANCHE at NORUNIT To: PCIP@UDEL at EDU Subject: NCSA and PC-NFS drivers for NI5*10 Background: We have a number of PCs on an ethernet today. The ethernet cards are a mixture of NI5010 and NI5210 cards. I have tried NCSA telnet and am impressed. We are also considering the purchase of a Sun 4 and want to use it for a file server for the PCs. (Not for that purpose alone of course - this is to be a first step out of the PC world into the Unix world - but we do need a file server for all those PCs!) Needless to say, this seems to require PC-NFS. A good point seems to be that PC-NFS and NCSA telnet can coexist on the same card at least as long as not both of them try to acces the net at the same time. However, there is a problem: NCSA telnet supports the NI5210 but not the NI5010. PC-NFS supports the NI5010 but not the NI5210. Upon my inquiry, Tim Krauskopf (timk@ncsa.uiuc.edu) tells me that I don't know of any planned support for the NI5010 in NCSA Telnet. Source code for pc-ip exists, so someone could adapt that driver fairly easily (as these things go). Has anyone out there done this? Or planned to? I looked at both drivers. The NCSA Telnet driver for NI5210 is all assembly and uses no interrupts. The PC-IP driver for NI5010 is mostly C and does use interrupts. I don't know if this would make it too slow and/or incompatible with NCSA telnet. Any helpful suggestions are welcome. On the other hand, our Sun sales rep tells me that although PC-NFS does not support the NI5210, I am more than welcome to write my own driver. Has anyone tried that? Do such drivers exist, either commercially or in the public domain? Finally - any other suggestions as to how I might solve our problems are most welcome. Please reply directly to me (hanche@norunit.bitnet). I will summarize to the list if any interesting/useful information comes out of it. - Harald Hanche-Olsen Division of Mathematical Sciences The Norwegian Institute of Technology