From owner-freebsd-questions Fri Oct 5 10:14:23 2001 Delivered-To: freebsd-questions@freebsd.org Received: from navgwout.symantec.com (navgwout.symantec.com [198.6.49.12]) by hub.freebsd.org (Postfix) with ESMTP id D9C4037B403 for ; Fri, 5 Oct 2001 10:14:20 -0700 (PDT) Received: from navgwout.symantec.com (navgwout [198.6.49.12]) by navgwout.symantec.com (8.9.3+Sun/8.9.3) with SMTP id KAA07833 for ; Fri, 5 Oct 2001 10:14:15 -0700 (PDT) Received: from mailer.symantec.com ([198.6.49.176]) by navgwout.symantec.com (NAVGW 2.5.1.13) with SMTP id M2001100510141400474 for ; Fri, 05 Oct 2001 10:14:14 -0700 Received: from uscu-smtp02.symantec.com (uscu-smtp02.symantec.com [155.64.74.114]) by mailer.symantec.com (8.9.3+Sun/8.9.3) with ESMTP id KAA21021 for ; Fri, 5 Oct 2001 10:14:14 -0700 (PDT) Subject: Program suddenly becomes stuck in getblk and biowr? To: freebsd-questions@FreeBSD.ORG X-Mailer: Lotus Notes Release 5.0.7 March 21, 2001 Message-ID: From: "Jay Rossiter" Date: Fri, 5 Oct 2001 10:11:45 -0700 X-MIMETrack: Serialize by Router on USCU-SMTP02/SYMSMTP(Release 5.0.6a |January 17, 2001) at 10/05/2001 10:07:16 AM MIME-Version: 1.0 Content-type: text/plain; charset=us-ascii Sender: owner-freebsd-questions@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG A program that I work on suddenly started having problems when I moved to 4.4-RELEASE the other day. A program that used to take an hour or two to do its work is now taking six to eight hours to do the same amount of work. Running top, the first thing I noticed was that the app appears as if it has suddenly become completely IO bound where this was never the case before. (It is very I/O intensive, but not to the degree that it's performing now). Previously this program would knock off 90%+ of the CPU for itself and do its work in a RUN state. Now it's bouncing between GETBLK and BIOWR and taking 5% of the CPU with an occasional hiccup that takes it to 15% for a couple seconds. Since this was a drastic change, I'm doubting it's hardware. (Just for reference, it's a P4 1.4GHz with ATA100 drives) Was there some drastic change to I/O code between 4.3 and 4.4? To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message