From owner-freebsd-questions@FreeBSD.ORG Fri Aug 29 22:21:43 2003 Return-Path: Delivered-To: freebsd-questions@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 1344D16A4BF for ; Fri, 29 Aug 2003 22:21:43 -0700 (PDT) Received: from smtp-2.paradise.net.nz (smtp-2a.paradise.net.nz [202.0.32.195]) by mx1.FreeBSD.org (Postfix) with ESMTP id E6E8F43FE1 for ; Fri, 29 Aug 2003 22:21:41 -0700 (PDT) (envelope-from markir@paradise.net.nz) Received: from paradise.net.nz (202-0-44-68.adsl.paradise.net.nz [202.0.44.68]) by smtp-2.paradise.net.nz (Postfix) with ESMTP id E01A69EA19 for ; Sat, 30 Aug 2003 17:21:39 +1200 (NZST) Message-ID: <3F50335D.1030807@paradise.net.nz> Date: Sat, 30 Aug 2003 17:17:17 +1200 From: Mark Kirkwood User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.4) Gecko/20030630 X-Accept-Language: en-us, en MIME-Version: 1.0 To: questions@freebsd.org Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Subject: IO To IDE Blocking In 5.1 X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 30 Aug 2003 05:21:43 -0000 I am running Freebsd 5.1 on a Dell 410 with 2xPII and 2xIDE drives (each on its own channel). Drive #1 as 1 filesystem mounted on /, drive #2 has 1 filesystem, mounted on /data1 (both have softupdates enabled) i) start a process doing some io on drive #1 ii) start iostat 5 iostat blocks until the process is completed. If I start iostat *first*, then it does not block, and I can see the io activity generated by the process when it starts. If I try this test on drive #2, then neither case blocks. I am running the SMP kernel, but can reproduce this using GENERIC. Any ideas ? P.s : This came to my attention whilst experimenting with the smp kernel, I was able to get nice 2 processer scaling with 2 running processes *except* if they both needed to do IO (even if they were doing it on different drives). Note : Have previously run 5.1 on a HP VE6, and dont recall encountering this issue. regards Mark