From owner-freebsd-hackers Fri Oct 30 02:07:52 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id CAA13276 for freebsd-hackers-outgoing; Fri, 30 Oct 1998 02:07:52 -0800 (PST) (envelope-from owner-freebsd-hackers@FreeBSD.ORG) Received: from alcanet.com.au (border.alcanet.com.au [203.62.196.10]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id CAA13263 for ; Fri, 30 Oct 1998 02:07:48 -0800 (PST) (envelope-from peter.jeremy@auss2.alcatel.com.au) Received: by border.alcanet.com.au id <40347>; Fri, 30 Oct 1998 21:07:12 +1100 Date: Fri, 30 Oct 1998 21:07:36 +1100 From: Peter Jeremy Subject: Re: sync writes with softupdates enabled To: mike@smith.net.au Cc: hackers@FreeBSD.ORG Message-Id: <98Oct30.210712est.40347@border.alcanet.com.au> Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG I wrote: > All I can think of is that the MMAP'd history file is causing the > kernel to unnecessarily perform sync writes to try and keep the > MMAP and VFS images in sync. Mike Smith wrote: >That's almost certainly correct. Try running without mmap. It used to >be faster that way; maybe it is again? I tried rebuilding it without DBZ-MMAP and it no longer does any sync writes to /usr. I haven't tried following up Luoqi's suggestion as to the possible cause, but this behaviour would seem undesirable - it would seem to nullify many of the advantages of using mmap(2) [I accept that there's a fair amount of black magic involved in making it work at all without a unified buffer cache]. Peter To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message