From owner-freebsd-hackers Sat Aug 25 11:19:26 2001 Delivered-To: freebsd-hackers@freebsd.org Received: from earth.backplane.com (earth-nat-cw.backplane.com [208.161.114.67]) by hub.freebsd.org (Postfix) with ESMTP id 90F1337B401 for ; Sat, 25 Aug 2001 11:19:23 -0700 (PDT) (envelope-from dillon@earth.backplane.com) Received: (from dillon@localhost) by earth.backplane.com (8.11.6/8.11.2) id f7PIIeO10488; Sat, 25 Aug 2001 11:18:40 -0700 (PDT) (envelope-from dillon) Date: Sat, 25 Aug 2001 11:18:40 -0700 (PDT) From: Matt Dillon Message-Id: <200108251818.f7PIIeO10488@earth.backplane.com> To: Mike Silbersack Cc: Julian Elischer , Subject: Re: Possible race in i386/i386/pmap.c:pmap_copy() References: Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG : :On Sat, 25 Aug 2001, Julian Elischer wrote: : :> AH yes, it's a race for KSe, but we are 1:1 still so it's not a problem (yet :-) :> ( at least, not the one that's hitteng me at the moment) : :Well, don't get frustrated, look on the bright side of things: Even if :you don't have KSE running in the next few weeks, you will have fixed :*many* bugs in the kernel. :) : :Actually, wouldn't this bug also affect linuxthreads operation? I'm not :sure if anyone actually uses that, but if they do and were experiencing :problems, this could explain it. : :Mike "Silby" Silbersack Yes, it could effect linuxthreads. I think it will be fairly easy to fix - I just have to add a check after it gets past the blocking condition to see if it needs to reload the alternative page table. - I found the KSE bug that was stumping us last night. Julian will be a happy camper when he wakes up! I found a bunch of minor issues as well, things like buildworld erroring out in libkvm due to the structural changes, but I expect he will be able to commit the KSE patchset very soon now that we've found the bug that was making such a huge mess of the machine. It turns out that some assembly in swtch.s got munged and pmap->pm_active was not getting set, having the side effect of causing most of the TLB invalidation routines to believe that they didn't have to do anything. You can imagine the result! -Matt To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message