From owner-freebsd-hackers Tue Mar 7 12:21:00 1995 Return-Path: hackers-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id MAA24438 for hackers-outgoing; Tue, 7 Mar 1995 12:21:00 -0800 Received: from Root.COM (implode.Root.COM [198.145.90.1]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id MAA24428 for ; Tue, 7 Mar 1995 12:20:56 -0800 Received: from corbin.Root.COM (corbin.Root.COM [198.145.90.18]) by Root.COM (8.6.8/8.6.5) with ESMTP id MAA00643; Tue, 7 Mar 1995 12:20:30 -0800 Received: from localhost (localhost [127.0.0.1]) by corbin.Root.COM (8.6.9/8.6.5) with SMTP id MAA01516; Tue, 7 Mar 1995 12:20:30 -0800 Message-Id: <199503072020.MAA01516@corbin.Root.COM> X-Authentication-Warning: corbin.Root.COM: Host localhost didn't use HELO protocol To: "Paul F. Werkowski" cc: hackers@FreeBSD.org Subject: Re: SYSV{SHM|MSG} kills network?? In-reply-to: Your message of "Tue, 07 Mar 95 14:41:13 EST." <199503071941.OAA01546@snoopy.mv.com> From: David Greenman Reply-To: davidg@Root.COM Date: Tue, 07 Mar 1995 12:20:24 -0800 Sender: hackers-owner@FreeBSD.org Precedence: bulk >I was running a stock 2.0 kernel connected to ethernet by DE435 PCI >NIC. Kernel was also built with SYSVSHM. This system got along just >fine on our small LAN. A few weeks ago I added options SYSVMSG and >SYSVSEM and rebuilt the kernel. The new system then had network >problems which caused repeated NFS hangs, etc. Ping from the FreeBSD >box to any other system resulted in OK performance for about 45 >seconds followed by 45 seconds of lost packets, followed by 45 seconds >of ok response, followed by 45 seconds of lost packets, etc.. From >TCPDUMP it appeared that nothing was received for about 50% of the >time. Rebuilding the kernel without SYSVMSG and SYSVSEM returned the >network to OK performance. > >So, is this fixed in -current or can anyone confirm this behavior? Yes, it's fixed in -current. It was caused by a bad assumption about allocated memory being physically contiguous at startup. -DG