Return to BSD News archive
Received: by minnie.vk1xwt.ampr.org with NNTP id AA825 ; Mon, 08 Feb 93 08:01:53 EST Path: sserve!manuel.anu.edu.au!munnari.oz.au!sgiblab!swrinde!gatech!udel!sbcs.sunysb.edu!stark.UUCP!gene From: gene@stark.uucp (Gene Stark) Newsgroups: comp.unix.bsd Subject: Re: [386bsd] Console Devices?? Date: 5 Feb 93 07:51:13 Organization: Gene Stark's home system Lines: 46 Message-ID: <GENE.93Feb5075114@stark.stark.uucp> References: <1993Feb4.123606@axion.bt.co.uk> NNTP-Posting-Host: stark.uucp In-reply-to: lessen@axion.bt.co.uk's message of 4 Feb 93 12:36:06 GMT Lee Essen writes: >I've recently managed to install 386bsd 0.1 on my noname 386DX/40. After >a few initial teething problems it all seems to be ok. > >(Just a side note: the amount of teething problems experienced with 386bsd > were far less that with SCO, support is better too!) > >During my kernel builds (sometimes) the compilation of a module can take >*ages* (sometimes is just stops - no disk access), if I stop the make and start again the modules compilation can be really quick. At the moment I >only have 4MB, and am using the default 5MB swap - (i'de like to enlarge it but only have one drive and no backup medium and don't like the thought of starting again!). Is this small memory/swap the cause of this! Yes, small memory is the problem. You also don't say whether you are running the kernel with patchkit-0.2 installed. You should be. I compiled a kernel yesterday on a 386 machine with 4MB and default swap and it took a long time but completed without difficulty. >Although I've got a vt100 terminal attached I thought that virtual consoles >would be nice. I installed the 'syscon' package (on ref.tfs.com, in >sos-syscons - or something like that!) > >I found a few problems with this, has anyone else experienced them, and >Is there a better/improved virtual console driver around? > >1. The kernel build failed - can't find 'get_char' when linking. This > routine is declared static in the syscons.c file (not static in > pccons.c) Allowing the function to be exported works but should > I have to change the source? Yes, this is a problem with syscons-0.1. The author is aware of it and says it is fixed in the 0.2 version. >I'm about to start trying to add virtual console support to the standard >console driver unless something is available! (emphasis on the word >'trying' in the last bit!) This sounds like a waste of time, since there are now several virtual console drivers out there. Besides syscons, there is "codrv" by Holger Veit and another driver (I forgot what it's called) by Hellmuth Michaelis. I am using syscons-0.1. If you don't like something about it, rather than rewriting from scratch a more effective use of your time might be to find out what is wrong, fix it, post the changes and/or send them to the author. - Gene Stark -- stark@cs.sunysb.edu