Anyways, here goes via sunsite.auc.dk...
-- Mike. ------- Start of forwarded message ------- To: "Adam D. Bradley" <artdodge@cs.bu.edu> Cc: linux-kernel@vger.rutgers.edu Subject: Re: Are people still having fun with psaux and 2.1.12/3x ? References: <Pine.LNX.4.02.9812051700170.32099-100000@dawntreader.bu.edu> From: Mike McEwan <mike@lotusland.demon.co.uk> Date: 07 Dec 1998 12:36:21 +0000 Message-ID: <m3yaok9liy.fsf@lotusland.demon.co.uk> Mime-Version: 1.0"Adam D. Bradley" <artdodge@cs.bu.edu> writes:
> I'm having similar problems, except the mouse going ballistic usually > coincides with heavy IDE disk use. This is an SMP system (dual PII), > mouse is a logitech 3-button, server is Accelerated-X. > > I've been meaning to try backing out the "grand ps/2 unification" patch to > see if this fixes it, but am coming up on the last week of the semester > and will be a little short on time until this coming Friday (when I plan > to be watching the new Star Trek movie). If you get a chance, just back > out those changes and see if that fixes it...
Well, I'm off work today with a cold (tsh, tsh), and decided to give this a go. It wasn't that straight forward to reverse the patches as a few sources are affected, some of them hit by several different things (e.g. ksyms.c). A structure referenced in psaux.c would appear to have changed - I've changed the affected line of code to mirror almost identical code in qpmouse.c (by the same author). Also it would appear there has been a second patch to pc_keyb.c since the 'unifications' that took place *officially* in patch-2.1.126.
So far things are looking good with this in place, kernel 2.1.131. I'll have to stress things a little longer of-course. Let me know if you're interested in a patch, although I'm not sure how much good it will do other than proving our recent problems with PS/2 mice do indeed stem from the code merge of the PS/2 mouse driver with the keyboard code :-(. Some comments to be seen in psaux.c show that Linus was very keen to see this code merge take place, so any 'fixing' of the problem will probably have to be in a forward direction.
-- Mike. ------- End of forwarded message -------- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/