From mboxrd@z Thu Jan 1 00:00:00 1970 From: Neil Horman Subject: Re: [Bug #15196] kmem_cache_create: duplicate cache ccid2_h Date: Mon, 1 Feb 2010 06:55:18 -0500 Message-ID: <20100201115518.GA25094@hmsreliant.think-freely.org> References: <7b6bb4a51001311930i7d72f60dj3b15ca3e1a715607@mail.gmail.com> <20100131.232050.266652459.davem@davemloft.net> Mime-Version: 1.0 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Content-Disposition: inline In-Reply-To: <20100131.232050.266652459.davem@davemloft.net> Sender: linux-kernel-owner@vger.kernel.org List-ID: Content-Type: text/plain; charset="iso-8859-1" To: David Miller Cc: xtfeng@gmail.com, rjw@sisk.pl, htd@fancy-poultry.org, linux-kernel@vger.kernel.org, kernel-testers@vger.kernel.org On Sun, Jan 31, 2010 at 11:20:50PM -0800, David Miller wrote: > From: Xiaotian Feng > Date: Mon, 1 Feb 2010 11:30:02 +0800 >=20 > > On Mon, Feb 1, 2010 at 8:22 AM, Rafael J. Wysocki wro= te: > >> This message has been generated automatically as a part of a repor= t > >> of recent regressions. > >> > >> The following bug entry is on the current list of known regression= s > >> from 2.6.32. =A0Please verify if it still should be listed and let= me know > >> (either way). > >> > >> > >> Bug-Entry =A0 =A0 =A0 : http://bugzilla.kernel.org/show_bug.cgi?id= =3D15196 > >> Subject =A0 =A0 =A0 =A0 : kmem_cache_create: duplicate cache ccid2= _h > >> Submitter =A0 =A0 =A0 : Heinz Diehl > >> Date =A0 =A0 =A0 =A0 =A0 =A0: 2010-01-30 18:33 (2 days old) > >> References =A0 =A0 =A0: http://marc.info/?l=3Dlinux-kernel&m=3D126= 487640324942&w=3D4 > >=20 > > Cced Neil, > >=20 > > I think this one is introduced by commit > > de4ef86cfce60d2250111f34f8a084e769f23b16, > > passing char *slab_name_fmt as function parameter, but vsnprintf is > > using sizeof(slab_name_fmt), > > which is 8 (or 4 in 32bit kernel) instead of 32 as old version. > >=20 > > Does following patch resolve this bug, Heinz? >=20 > There seems to be even more to this than that. Neils > patch seems to need completely reverting. >=20 > See the patch set posted by Gerrit Renker: >=20 > http://marc.info/?l=3Dlinux-netdev&m=3D126500585823775&w=3D2 > http://marc.info/?l=3Dlinux-netdev&m=3D126500591923880&w=3D2 >=20 Dave, some of this doesn't make the least bit of sense to me. I get th= e sizeof error, thats clear (and I apologize, I should have seen that), but Gerr= its revert of the dccp_probe changes is non-sensical. I'm not sure I even = follow the comments: >Previously (during about 4 years of this module's history) there had n= ever >been a problem with the 'silent dependency' that the commit tried to f= ix: >this dependency is deliberate and required, since dccp_probe performs = probing >of dccp connections and hence needs to know about dccp internals. He claims this dependency is deliberate and requires, to which I agree,= but he would seem to fix that by making the dccp_probe module error out in the= event that dccp wasn't loaded. Why bother with that? Neil