From owner-freebsd-fs@FreeBSD.ORG Thu Dec 20 23:51:42 2007 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 4C33916A417; Thu, 20 Dec 2007 23:51:42 +0000 (UTC) (envelope-from peter.schuller@infidyne.com) Received: from smtp.infidyne.com (ds9.infidyne.com [88.80.6.206]) by mx1.freebsd.org (Postfix) with ESMTP id 0AD5713C45B; Thu, 20 Dec 2007 23:51:41 +0000 (UTC) (envelope-from peter.schuller@infidyne.com) Received: from c-8216e555.03-51-73746f3.cust.bredbandsbolaget.se (c-8216e555.03-51-73746f3.cust.bredbandsbolaget.se [85.229.22.130]) by smtp.infidyne.com (Postfix) with ESMTP id 3D20677887; Fri, 21 Dec 2007 00:51:40 +0100 (CET) From: Peter Schuller To: Julian Elischer Date: Fri, 21 Dec 2007 00:51:33 +0100 User-Agent: KMail/1.9.7 References: <200712210036.49040.peter.schuller@infidyne.com> <476AFDBC.9040301@elischer.org> In-Reply-To: <476AFDBC.9040301@elischer.org> MIME-Version: 1.0 Content-Type: multipart/signed; boundary="nextPart1325381.rZX7Dr3quE"; protocol="application/pgp-signature"; micalg=pgp-sha1 Content-Transfer-Encoding: 7bit Message-Id: <200712210051.41936.peter.schuller@infidyne.com> Cc: freebsd-fs@freebsd.org, ticso@cicely.de, Ivan Voras Subject: Re: readv: parallel or sequential? X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 20 Dec 2007 23:51:42 -0000 --nextPart1325381.rZX7Dr3quE Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Content-Disposition: inline > multiple reads and writes to the same file *From different file > descriptors* (same process or not) might proceed in "parallel" but readv > and writev will be implemented serially to the filesystem. now IF THE > FILESYSTEM IS NOT DOING SYNCHRONOUS DISK ACCESSES the reads and writes > might proceed in parallel or be grouped, clustered or otherwise rearanged. If the original claims were just as applied to readv/writev, I don't really= =20 have a problem since there is no guarantee that it will be performed in=20 parallel (though even then, assuming non-synchronous writes, it would=20 certainly be nice if it did pass it along to the device layer in a fashion= =20 that allowed concurrency). But when I/O is clearly being done with the intent of concurrency (by using= =20 AIO or by performing the I/O in different threads/processes), it is more=20 important. =2D-=20 / Peter Schuller PGP userID: 0xE9758B7D or 'Peter Schuller ' Key retrieval: Send an E-Mail to getpgpkey@scode.org E-Mail: peter.schuller@infidyne.com Web: http://www.scode.org --nextPart1325381.rZX7Dr3quE Content-Type: application/pgp-signature; name=signature.asc Content-Description: This is a digitally signed message part. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.4 (FreeBSD) iD8DBQBHawANDNor2+l1i30RApKtAJ42aWOvXKnYbC73AoXaZvza/I1MVwCdFei1 0FAfRm9x1swgNtx3YDTVmww= =NQ+4 -----END PGP SIGNATURE----- --nextPart1325381.rZX7Dr3quE--