Return to BSD News archive
Path: sserve!newshost.anu.edu.au!munnari.oz.au!news.Hawaii.Edu!ames!olivea!gossip.pyramid.com!decwrl!csus.edu!netcom.com!duncan From: duncan@netcom.com (Donald K. Duncan) Newsgroups: comp.os.386bsd.bugs Subject: Re: Software Solution to "com: silo overflow" Summary: kernel won't boot - size > 640KB Message-ID: <duncanC4B7DF.zI@netcom.com> Date: 22 Mar 93 21:22:27 GMT References: <GENE.93Mar15225407@stark.stark.uucp> Organization: Netcom - Online Communication Services (408 241-9760 guest) Lines: 32 In article <GENE.93Mar15225407@stark.stark.uucp>, gene@stark.uucp (Gene Stark) writes: > FINALLY -- A SOFTWARE SOLUTION TO THE SILO OVERFLOW PROBLEM > > OK, folks, here is a set of software patches that almost completely > licks the "com: silo overflow" problem without having to install any > 16550's. I have tested it on my system (486DX/33) with a 9600 baud > terminal and a 9600 baud modem running UUCP and it reduces the number > of silo overflows almost to zero. I don't have SLIP, so I haven't > tested it with that, but maybe someone will let me know whether it > works. > > ... > > Good luck! > > Gene Stark > stark@cs.sunysb.edu > > March 15, 1993 > I patched my kernel sources (pl.0.2.1) with your patches and compiled without difficulty. The kernel however is around 710KB total. The Bss segment nearly doubled in size! Is this correct? The buffers you added shouldn't be anywhere near this size. I get errors trying to boot the kernel as there is a requirement that it fit in the lower 640K (braindead!). Do you know of a way around this problem? -- duncan@zycad.com (Donald Duncan)