From owner-freebsd-mobile Wed Oct 29 22:00:32 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id WAA16964 for mobile-outgoing; Wed, 29 Oct 1997 22:00:32 -0800 (PST) (envelope-from owner-freebsd-mobile) Received: from twinlark.arctic.org (twinlark.arctic.org [204.62.130.91]) by hub.freebsd.org (8.8.7/8.8.7) with SMTP id WAA16955 for ; Wed, 29 Oct 1997 22:00:30 -0800 (PST) (envelope-from dgaudet-list-freebsd-mobile@arctic.org) Received: (qmail 31625 invoked by uid 500); 30 Oct 1997 06:05:11 -0000 Date: Wed, 29 Oct 1997 22:05:11 -0800 (PST) From: Dean Gaudet To: John Polstra cc: Nate Williams , mobile@FreeBSD.ORG Subject: Re: Suspend/resume in -current: still no joy In-Reply-To: <199710300148.RAA03943@austin.polstra.com> Message-ID: Organization: Transmeta Corp. MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-mobile@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk I haven't been following this thread closely ... but lemme mention an oddness about the thinkpad 560 apm bios that I discovered while getting linux to work cleanly on it. For some reason the bios will return suspend events "forever" if you continue to query it after having received one suspend event. It seems to want the apm client to give it one of the cookies that says "yeah yeah I heard ya". Unless you've changed how the freebsd apm stuff works you shouldn't be running into this problem -- I actually looked at it to figure out why suspend/resume worked fine under freebsd but not under linux. And freebsd was fine because it only dealt with one apm event at a time; whereas linux builds a queue of outstanding apm events. The other thing is that on AC power it never reports user suspend events if there's a powered PCMCIA card. I haven't tried system requested suspending ... I'll do it when I get home. And also if your tp560 bios is older than, um, april 97 or so then try upgrading. There were some apm glitches in earlier versions that messed up some folks trying my linux solution. Dunno if that helps any. IBM will give out their SMAPI docs which includes all the info needed to write the "PS2.EXE" utility... so someone could write a freebsd version, which would me no booting into win95 :) I've got a copy of the docs, but never got around to doing it :( Dean On Wed, 29 Oct 1997, John Polstra wrote: > Oooh, I've got a big clue! But first ... > > > > What I'm getting at is this: The manual for the ThinkPad 560 says > > > that an incoming call on the modem can wake up the machine when > > > it is in the suspended state. I'm having problems only with the > > > modem card, not with the ethernet card. > > > > On most machines, this can be disabled in the BIOS. (I think you > > can get to it in Win95). Can you try disabling it? > > OK, I tried. And I want you to be fully aware of the extreme personal > cost at which I did that. Because every time I have to interact with > Windows 95, I damn near have a stroke. This exercise probably took 2 > years off my life span. > > Anyway, in Windows 95 I found a couple of places where you could > enable/disable waking up the machine on an incoming call. Both were > already disabled. > > Back in Unix again (whew!), I made it fail again. I did that by > "apmconf -e" followed by "zzz". It resumed all by itself right away, > though it didn't seem to hang this time. Here are the messages it > produced: > > sio2: unload,gone > Return IRQ=10 > Slot 0, unfielded interrupt (0) > Card disabled, slot 0 > resumed from suspended mode (slept 00:00:01) > Card inserted, slot 0 > Entire system suspend failure: errcode = 96 > > Now here's the clue. I can make it fail like that reliably when the > AC adapter is plugged in. But if I unhook the AC so that it's using > battery power, suspend/resume works perfectly every time! Plug in the > AC, and it fails again. Unplug it, and it works again. > > What do you think about that? :-) > > John > -- > John Polstra jdp@polstra.com > John D. Polstra & Co., Inc. Seattle, Washington USA > "Self-knowledge is always bad news." -- John Barth >