From owner-freebsd-bugs Thu Jul 20 20:19:33 1995 Return-Path: bugs-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.11/8.6.6) id UAA04871 for bugs-outgoing; Thu, 20 Jul 1995 20:19:33 -0700 Received: from Root.COM (implode.Root.COM [198.145.90.1]) by freefall.cdrom.com (8.6.11/8.6.6) with ESMTP id UAA04863 for ; Thu, 20 Jul 1995 20:19:31 -0700 Received: from corbin.Root.COM (corbin [198.145.90.18]) by Root.COM (8.6.11/8.6.5) with ESMTP id UAA00562; Thu, 20 Jul 1995 20:19:06 -0700 Received: from localhost (localhost [127.0.0.1]) by corbin.Root.COM (8.6.11/8.6.5) with SMTP id UAA07026; Thu, 20 Jul 1995 20:20:06 -0700 Message-Id: <199507210320.UAA07026@corbin.Root.COM> To: Matt Dillon cc: bugs@freebsd.org Subject: Re: verification of ffs_vget() bug In-reply-to: Your message of "Thu, 20 Jul 95 17:29:43 PDT." <199507210029.RAA21522@shell1.best.com> From: David Greenman Reply-To: davidg@Root.COM Date: Thu, 20 Jul 1995 20:20:05 -0700 Sender: bugs-owner@freebsd.org Precedence: bulk > Here are some real numbers... the collision occurs much more often > then I would have thought in a heavily loaded system! I'll say! John and I went through the possible failure scenarios and we think that severe problems resulting in a system crash would be rare on most systems (which is obviously proven true), but that cache coherency problems and other strange behavior might be likely. > (my temporary ffs_vget() code is included below as well) Actually, you're fix looks pretty good. After talking with Kirk awhile about your fix, other possible fixes, etc, I think we've concluded that doing the global lock is as good a fix as any for the problem, and appears to be the lowest overhead solution. Thanks for the fix... -DG