From mboxrd@z Thu Jan 1 00:00:00 1970 From: Florian Weimer Subject: Re: [rfc] fsync_range? Date: Thu, 22 Jan 2009 22:18:22 +0100 Message-ID: <871vuv3uqp.fsf@mid.deneb.enyo.de> References: <20090120164726.GA24891@wotan.suse.de> <20090120183120.GD27464@shareable.org> <20090121012900.GD24891@wotan.suse.de> <20090121031500.GA2354@shareable.org> <20090121041604.GI24891@wotan.suse.de> <20090121045921.GA3944@shareable.org> <20090121062306.GK24891@wotan.suse.de> <20090121121308.GA31253@mit.edu> <20090121123711.GA10637@shareable.org> <20090121141207.GD31253@mit.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Jamie Lokier , Nick Piggin , linux-fsdevel@vger.kernel.org To: Theodore Tso Return-path: Received: from mail.enyo.de ([212.9.189.167]:39245 "EHLO mail.enyo.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751974AbZAVVze (ORCPT ); Thu, 22 Jan 2009 16:55:34 -0500 In-Reply-To: <20090121141207.GD31253@mit.edu> (Theodore Tso's message of "Wed, 21 Jan 2009 09:12:07 -0500") Sender: linux-fsdevel-owner@vger.kernel.org List-ID: * Theodore Tso: > Actually, I take that back; Oracle (and most other enterprise > databases; the world is not just Oracle --- there's also DB2, for > example) generally uses Direct I/O, so I wonder if they are using > sync_file_range() at all. Recent PostgreSQL might use it because it has got a single-threaded background writer which benefits from non-blocking fsync(). I'll have to check to be sure, though.