Date: Thu, 16 Mar 95 14:03:56 MST From: terry@cs.weber.edu (Terry Lambert) To: bde@zeta.org.au (Bruce Evans) Cc: bde@zeta.org.au, aw2t+@andrew.cmu.edu, freebsd-hackers@freefall.cdrom.com Subject: Re: installing on a thinkpad 750 Message-ID: <9503162103.AA24601@cs.weber.edu> In-Reply-To: <199503161735.DAA32571@godzilla.zeta.org.au> from "Bruce Evans" at Mar 17, 95 03:35:54 am
next in thread | previous in thread | raw e-mail | index | archive | help
> >> The problem may be that the APM BIOS uses memory that it has reserved > >> below 640K. FreeBSD doesn't honor BIOS reserved memory. > > >This theory doesn't make sense. The problem with it is that the BIOS > >is not accessed after BSD boots, and BSD doesn't load into the BIOS > >reserved area while the boot blocks are running before that. > > The APM BIOS is accessed whenever BSD is idle. How? Who makes the BIOS call, and what is "idle"? I actually don't think the BIOS is used when running in protected mode. In the ones I've disassembled, the APM stuff is basically accessed through the DOS-Not-Busy interrupt being hooked to count, and then others are hooked to cause the count to be reset. When the count rolls to a particular value, the APM stuff is invoked, assuming the keyboard code has not been entered. That is, the hook is quite similar to how you implement TSR's, but is intrinsic to the BIOS code for the hooked operations. The major kink is that the state save is very processer specific, since a lot of Intel and IBM chips, for instance, used the undocumented ICEBP which is available on all Intel processers to do the state save. AMD and Cyrix have their own (documented-but-different) non-standard instructions to do the same job. I think that the APM code is no more likely to run "when it needs to" once BSD is up than DOS TSR's are likely to run. 8-). > >Anyway, back to the keyboard. The problem is that by default the > >ThinkPad uses PS/2 scan code mode. > > Doesn't it accept the console driver's setting of the old scan code mode? Not according to Joerg's message. Really, we need a definitive post from someone who has one working; I thought that's what the quoted article did, although it was hardly something you could do anything about without another machine to build a kernel for you... a poor soloution, but the only one I had to offer. 8-(. Terry Lambert terry@cs.weber.edu --- Any opinions in this posting are my own and not those of my present or previous employers.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?9503162103.AA24601>