From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757466Ab3CSBSi (ORCPT ); Mon, 18 Mar 2013 21:18:38 -0400 Received: from shadbolt.e.decadent.org.uk ([88.96.1.126]:40952 "EHLO shadbolt.e.decadent.org.uk" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754222Ab3CSBSh (ORCPT ); Mon, 18 Mar 2013 21:18:37 -0400 Message-ID: <1363655903.3937.350.camel@deadeye.wl.decadent.org.uk> Subject: Re: [ 04/21] target/pscsi: Fix page increment From: Ben Hutchings To: Asias He Cc: "Nicholas A. Bellinger" , linux-kernel@vger.kernel.org, stable@vger.kernel.org, Greg Kroah-Hartman , target-devel Date: Tue, 19 Mar 2013 01:18:23 +0000 In-Reply-To: <20130319005659.GB15879@hj.localdomain> References: <20130312223245.109098379@linuxfoundation.org> <20130312223245.664885170@linuxfoundation.org> <1363399822.3937.161.camel@deadeye.wl.decadent.org.uk> <20130318053519.GE6232@hj.localdomain> <1363640434.3156.36.camel@haakon2.linux-iscsi.org> <1363649447.3937.348.camel@deadeye.wl.decadent.org.uk> <20130319005659.GB15879@hj.localdomain> Content-Type: multipart/signed; micalg="pgp-sha512"; protocol="application/pgp-signature"; boundary="=-Q0o54vILwLAc2gWg+fCO" X-Mailer: Evolution 3.4.4-2 Mime-Version: 1.0 X-SA-Exim-Connect-IP: 2001:470:1f08:1539:d98f:da4e:f620:7bea X-SA-Exim-Mail-From: ben@decadent.org.uk X-SA-Exim-Scanned: No (on shadbolt.decadent.org.uk); SAEximRunCond expanded to false Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org --=-Q0o54vILwLAc2gWg+fCO Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable On Tue, 2013-03-19 at 08:56 +0800, Asias He wrote: > On Mon, Mar 18, 2013 at 11:30:47PM +0000, Ben Hutchings wrote: > > On Mon, 2013-03-18 at 14:00 -0700, Nicholas A. Bellinger wrote: > > > On Mon, 2013-03-18 at 13:35 +0800, Asias He wrote: > > > > On Sat, Mar 16, 2013 at 02:10:22AM +0000, Ben Hutchings wrote: > > > > > On Tue, 2013-03-12 at 15:44 -0700, Greg Kroah-Hartman wrote: > > > > > > 3.0-stable review patch. If anyone has any objections, please = let me know. > > > > > >=20 > > > > > > ------------------ > > > > > >=20 > > > > > > From: Asias He > > > > > >=20 > > > > > > commit 472b72f2db7831d7dbe22ffdff4adee3bd49b05d upstream. > > > > > >=20 > > > > > > The page++ is wrong. It makes bio_add_pc_page() pointing to a w= rong page > > > > > > address if the 'while (len > 0 && data_len > 0) { ... }' loop i= s > > > > > > executed more than one once. > > > > > >=20 > > > > > > Signed-off-by: Asias He > > > > > > Signed-off-by: Nicholas Bellinger > > > > > > Signed-off-by: Greg Kroah-Hartman > > > > > >=20 > > > > > > --- > > > > > > drivers/target/target_core_pscsi.c | 1 - > > > > > > 1 file changed, 1 deletion(-) > > > > > >=20 > > > > > > --- a/drivers/target/target_core_pscsi.c > > > > > > +++ b/drivers/target/target_core_pscsi.c > > > > > > @@ -1210,7 +1210,6 @@ static int __pscsi_map_task_SG( > > > > > > bio =3D NULL; > > > > > > } > > > > > > =20 > > > > > > - page++; > > > > > > len -=3D bytes; > > > > > > data_len -=3D bytes; > > > > > > off =3D 0; > > > > >=20 > > > > > So in case a fragment crosses a page boundary, we wrap around to = the > > > > > beginning of the same page? That doesn't look right. > > > >=20 > > > > If the fragment crosses a page boundary, what is the correct page > > > > for it? > > > >=20 > > > > Nicholas, can we assume sg->length + sg->offset should be less than= PAGE_SIZE here? > > > >=20 > > >=20 > > > sg->length + sg->offset can be less than or equal to PAGE_SIZE here. > > > > > > For everything other than tcm_loop + tcm_vhost using externally > > > allocated SGLs, we can expect fragments to never cross the page > > > boundary. > > >=20 > > > For tcm_loop + tcm_vhost, there are a few special cases with control = CDB > > > paylaods (usually going through scsi-generic) where we can have a non > > > zero sg->offset, but at least in the cases I've seen this is still no= t > > > using SGL elements that exceed PAGE_SIZE. > > > > > > So, I think this logic is OK for SGLs that cross page boundries, give= n > > > that it's done outside of the inner loop where *page is set during ea= ch > > > for_each_sg(). > >=20 > > The page is set using sg_page() in the outer loop and was then > > incremented in the inner loop. > >=20 > > for_each_sg(sgl, sg, sgl_nents, i) { > > page =3D sg_page(sg); > > off =3D sg->offset; > > len =3D sg->length; > >=20 > > while (len > 0 && data_len > 0) { > > bytes =3D min_t(unsigned int, len, PAGE_SIZE - off); > > bytes =3D min(bytes, data_len); > > ... > > /* page++; */ > > len -=3D bytes; > > data_len -=3D bytes; > > off =3D 0; > > } > > } > >=20 > > The inner loop is apparently meant to iterate over pages of a segment, > > but is now just wrapping around a single page. >=20 > Yes, but we never loop more than once in the inner loop. If you're sure of that, then: > So, how about > 1) fail on sg->offset + sg->length > PAGE_SIZE (we can not find a > proper page address in this case) > 2) remove the inner while loop, run the code was in the loop only once. this is fine, but then I don't understand why you sent a no-op change to stable in the first place. Ben. --=20 Ben Hutchings When you say `I wrote a program that crashed Windows', people just stare ..= . and say `Hey, I got those with the system, *for free*'. - Linus Torvalds --=-Q0o54vILwLAc2gWg+fCO Content-Type: application/pgp-signature; name="signature.asc" Content-Description: This is a digitally signed message part -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.12 (GNU/Linux) iQIVAwUAUUe83+e/yOyVhhEJAQoMlhAAzXSh0LvVc3FwmgPk8KCSIq5Pv3J7gHC8 lGK/BK1oRBxNjkuLrc5sS7FCPeFibZK/cjR/geg34tZ8FPPCGqEle7X+Nf2c6RF3 4QZfFveMfcmxlhsG0Gp3iB2Q+DKvV1nD6PXxD6BEN4ue7Uta8qqiDK3PbGxMaiaA GHqgxtpzDjNS+0ZHHtpGR1eWywuQQ0qnQzjM/AzX1O31cKP3L4damZQOtXwcgHDn Gugwm3BAvJ+R8P4h/usMiyYLcnkDG1HndQzeeJEPc0bRUXQN5fDnsJOOncB4MyqF lw6/4RE4fxBXK15cTIElafUOorBy8IQUVbkw1YkVi3DcI5mGiGrfXcmXnAIpurAH xhvUvkdCcaka/F8IK9T1VT6wkx5P+bKO/6tJtlJN1WK3OKgnoxkAzaijX4Bhvctg JqhpMRMM09bpKtEgYPpuES+lScDlUNsiex1oOcZHnE6jzRxTsMVcXJEH6LVKFjpV Qb9guYjW1h5te1BQhvr4F7BAeE/gjzHhJJnfXVx7Z78vxELxFsUs8vjlvGYoBtCc MBOcnaMDRtRita1BySBcyFvoTYJdrAhjMujbTdbgdeLs0FVrtlR4XNu0xDv7aFTU PwS3fOE2gjeA8bJ266faKOuq18Ln91f1mapJ1pxSQksKTUFoBgSm640deU+LI+mv D//doulkL+A= =WOnf -----END PGP SIGNATURE----- --=-Q0o54vILwLAc2gWg+fCO--