Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751595AbbEZXIa (ORCPT ); Tue, 26 May 2015 19:08:30 -0400 Received: from cantor2.suse.de ([195.135.220.15]:54289 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751868AbbEZXI2 (ORCPT ); Tue, 26 May 2015 19:08:28 -0400 Date: Wed, 27 May 2015 09:06:40 +1000 From: NeilBrown To: Alasdair G Kergon Cc: Ming Lin , Mike Snitzer , lkml , Christoph Hellwig , Kent Overstreet , Jens Axboe , Dongsu Park , Christoph Hellwig , Al Viro , Ming Lei , dm-devel@redhat.com, Lars Ellenberg , drbd-user@lists.linbit.com, Jiri Kosina , Geoff Levand , Jim Paris , Joshua Morris , Philip Kelleher , Minchan Kim , Nitin Gupta , Oleg Drokin , Andreas Dilger , Dave Chinner Subject: Re: [PATCH v4 01/11] block: make generic_make_request handle arbitrarily sized bios Message-ID: <20150527090640.44c346e2@notabene.brown> In-Reply-To: <20150526153414.GA31182@agk-dp.fab.redhat.com> References: <1432318723-18829-1-git-send-email-mlin@kernel.org> <1432318723-18829-2-git-send-email-mlin@kernel.org> <20150526143626.GA4315@redhat.com> <20150526153414.GA31182@agk-dp.fab.redhat.com> X-Mailer: Claws Mail 3.10.1-162-g4d0ed6 (GTK+ 2.24.25; x86_64-suse-linux-gnu) MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; boundary="Sig_/cp2xcSEZDnlt4B7CsO6fyKB"; protocol="application/pgp-signature" Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2609 Lines: 69 --Sig_/cp2xcSEZDnlt4B7CsO6fyKB Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: quoted-printable On Tue, 26 May 2015 16:34:14 +0100 Alasdair G Kergon wrote: > On Tue, May 26, 2015 at 08:02:08AM -0700, Ming Lin wrote: > > Now bio_add_page() can build lager bios. > > And blk_queue_split() can split the bios in ->make_request() if needed. >=20 > But why not try to make the bio the right size in the first place so you > don't have to incur the performance impact of splitting? Because we don't know what the "right" size is. And the "right" size can change when array reconfiguration happens. Splitting has to happen somewhere, if only in bio_addpage where it decides = to create a new bio rather than add another page to the current one. So moving the split to a different level of the stack shouldn't necessarily change the performance profile. Obviously testing is important to confirm that. NeilBrown >=20 > What performance testing have you yet done to demonstrate the *actual* im= pact > of this patchset in situations where merge_bvec_fn is currently a net ben= efit? >=20 > Alasdair >=20 > -- > To unsubscribe from this list: send the line "unsubscribe linux-kernel" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html > Please read the FAQ at http://www.tux.org/lkml/ --Sig_/cp2xcSEZDnlt4B7CsO6fyKB Content-Type: application/pgp-signature Content-Description: OpenPGP digital signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQIVAwUBVWT8gTnsnt1WYoG5AQLPchAAhrwL4XhtBInjN+nQIz3d88KQWPGzTZRx 5MEeBMrfDOHC4ZksJgdwcMb7x5SGz5Q86JOKuNfDP/f1njGWnhlgcuTcdhnHtMik eJ0CtZ3X0asVh/1j+kdDYKD8Bz4oi9TYLIe5bA6mNAFhWtU5f4OhV3ahHNvwfery ai99HYZORwf7/3LlwSvi8juWUK06iFqdowDkfnxmUnoHpXd8Enk0B7dcTHi9HBu6 z/xA3VSCcclEU8f7eSRn5/G+hV9626MPFknXhvAFfGSxDyLwSyWLdeVbbaqnbSDH q3cZqtZk4whb5EobZVptZofTOna5q7/5u+FRPVESVeTSeov/c5eAvEndUaLxMNLQ MsnOAk/KlVlSdU3CHPm0Rej7+5CpXaiLDia0zUo/vgOSAS7JzqHuYnSVvsH5yNQ2 Vkv7VZRr2WinZD5Z+l9SwE416A2POtvgTXrljIGcjon1ZUZQrUTk/zIUtVOIKflC kmFOw+F1ApY3uio81hl3SZ4ABUqhlrRzRzJ/FP2hpYfnXV+QFo+lVXC0fODzyLS3 KS3ULL056oT1dMKf139INFGbgTLZXXYPzo4FQHr7d+rMlgB0p0f0zqh+kmEGdDhU KBFt7pHcjKJWMV8kvlplLWfz9K/e3QtO5uScY3xJQTCJkqySxw+NRCQ4wGIdZOyw DGnUcTRiXUo= =YtUu -----END PGP SIGNATURE----- --Sig_/cp2xcSEZDnlt4B7CsO6fyKB-- -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/