Date: Wed, 29 Jun 94 04:30:08 PDT From: Advanced Amateur Radio Networking Group Errors-To: TCP-Group-Errors@UCSD.Edu Reply-To: TCP-Group@UCSD.Edu Precedence: Bulk Subject: TCP-Group Digest V94 #133 To: tcp-group-digest TCP-Group Digest Wed, 29 Jun 94 Volume 94 : Issue 133 Today's Topics: NOS and the PC (3 msgs) Router Project Router Project (TNC from Hell) Send Replies or notes for publication to: . Subscription requests to . Problems you can't solve otherwise to brian@ucsd.edu. Archives of past issues of the TCP-Group Digest are available (by FTP only) from UCSD.Edu in directory "mailarchives". We trust that readers are intelligent enough to realize that all text herein consists of personal comments and does not represent the official policies or positions of any party. Your mileage may vary. So there. ---------------------------------------------------------------------- Date: Tue, 28 Jun 94 09:58:15 EDT From: crompton@nadc.nadc.navy.mil (D. Crompton) Subject: NOS and the PC To: ssampson@sabea-oc.af.mil, tcp-group@UCSD.EDU DMA vs. Serial I/O?? - Do you mean DMA vs. Polled I/O?? That makes more sense. In any event it is not clear which is better on the latest fast PC's. Most Ethernet cards DO NOT use DMA and the IOMEGA drives run FASTER in polled I/O mode than DMA. Of course this is very dependent on hardware and software design. Doug ------------------------------ Date: Tue, 28 Jun 1994 13:01:41 -0500 (CDT) From: ssampson@sabea-oc.af.mil (Steve Sampson) Subject: NOS and the PC To: crompton@NADC.NAVY.MIL (D. Crompton) > > Most Ethernet cards DO NOT use DMA . . . That's true. The NE2000 compatible cards just use serial I/O, but do bulk transfers if using the Crynwr drivers. The latest cards (I purchased two Accton cards during their $70 deal) do have a DMA mode (or shared memory as they call it). Which brings up a couple of points. The first is that Linux probably does NE2000 mode, and that some standards must exist on something called an IP-TNC. That is, that the IRQ and Addresses of I/O devices should be standardized so only one kernel is needed. -- Steve ------------------------------ Date: Tue, 28 Jun 94 16:58:07 EDT From: crompton@nadc.nadc.navy.mil (D. Crompton) Subject: NOS and the PC To: crompton@nadc.nadc.navy.mil, ssampson@sabea-oc.af.mil Steve, You keep using the term "serial I/O" - actually it is parallel I/O 8 or 16 bits from the ethernet card to the processor buss. The term Polled I/O vs DMA can be applied to this connected Parallel I/O port. In other words how is the data retrieved - direct to memory or handled by the processor in a polled fashion. Not being an expert I can only relate what I have heard and that is that DMA, although it sounds great does not always give the results you would expect. Many of the cards perform better, on fast processors, in polled mode. Doug ------------------------------ Date: Tue, 28 Jun 1994 07:20:31 -0400 From: "Brandon S. Allbery" Subject: Router Project To: tcp-group@ucsd.edu In your message of Tue, 28 Jun 1994 02:59:02 EDT, you write: +--------------- | >Ron, I don't understand this. If you're running at 38400, then the BEST you | >could do is 3,840 bytes/second. | | I always thought the thruput in NOS was measured in bytes/sec too. | I just did an ftp from Linux to the JNOS/Linux and this is what I got: +------------->8 The speed isn't used on ptys --- why should it? The kernel would have to introduce artificial delays into what is essentially a memory-to-memory copy. Ptys go at full bore; you're essentially measuring Linux's context switch and kernel-to-user-space copy time. ++Brandon -- Brandon S. Allbery kf8nh@kf8nh.ampr.org bsa@kf8nh.wariat.org Friends don't let friends load Windows NT. Linux iBCS2 emulation ------------------------------ Date: Tue, 28 Jun 1994 16:34:56 -0400 (EDT) From: DJ Gregor Subject: Router Project (TNC from Hell) To: agodwin@acorn.co.uk (Adrian Godwin) Adrian Godwin said: > > > From: ssampson@sabea-oc.af.mil (Steve Sampson) > > > > Develop code for an MS-DOS computer using Borland C++ which > > has the following components: > > But encapsulate the OS dependencies with a view to removing the OS > altogether : on a system that has no applications, what does DOS > do to justify it's memory usage after it's loaded NOS ? > Freedom from DOS would give freedom to use the hardware as a real > embedded system - cross-compiled, protected mode. I have been thinking of something like this. If everyone could get together and write a good amateur radio BBS program for Linux -- **NOT** a JNOS port or something like that, one that operates like a UNIX shell -- we could then strip down the Linux kernel, and integrate everything. Make it easy to configure and use--for the *AVERAGE* BBS operator, like the ones that run MSYS now. If we do that, we would have a NICE and POWERFULL BBS that would run on a 386 with 4 megabytes of memory. Most of the MSYS operators that I have seen have that kind of system already. This already exists, kind of. It is the N0ARY BBS. It was written by Bob Arasmith, and is meant to run on a Sun workstation. There are a few people trying to port it over to Linux right now, and I am one of them. Anyone else want to help??? ----- DJ Gregor, N8QLB dgregor@bronze.coil.com "...oh, you use DOS, sorry to hear that..." ------------------------------ End of TCP-Group Digest V94 #133 ******************************