From mboxrd@z Thu Jan 1 00:00:00 1970 From: Paul Moore Subject: Re: [RFC PATCH 0/5] Overhaul the audit filename handling Date: Mon, 12 Jan 2015 16:03:18 -0500 Message-ID: <2809190.iPDDC9gGz8@sifl> References: <20150108164514.11316.85331.stgit@localhost> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7Bit Cc: linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org To: linux-audit@redhat.com, viro@zeniv.linux.org.uk Return-path: In-Reply-To: <20150108164514.11316.85331.stgit@localhost> Sender: linux-kernel-owner@vger.kernel.org List-Id: linux-fsdevel.vger.kernel.org On Thursday, January 08, 2015 11:50:23 AM Paul Moore wrote: > There have been some patches added to v3.19-rcX to fix various > problems in the way audit handles filenames but they have been hacks > on top of hacks, not really something we want long term. This > patchset reworks the way audit handles filenames, removing a lot of > nasty hacks added recently, and fixing a few bugs that still remain. > > Most significant to folks outside of audit, patch 5/5 does away with > the nasty getname()/putname() kludge in favor of a less ugly reference > count approach. > > Comments and feedback are welcome. If nothing ugly pops up on review > I'll see if Linus is interested in taking this for the next v3.19-rcX > release, otherwise I'll toss it into linux-next for v3.20. > > -Paul > > --- > > Paul Moore (5): > fs: rework getname_kernel to handle up to PATH_MAX sized filenames > fs: create proper filename objects using getname_kernel() > audit: enable filename recording via getname_kernel() > audit: fix filename matching in __audit_inode() and > __audit_inode_child() > audit: replace getname()/putname() hacks with reference counters > > > fs/exec.c | 11 +++ > fs/namei.c | 98 ++++++++++++++++++----------- > fs/open.c | 11 +++ > include/linux/audit.h | 3 - > include/linux/fs.h | 9 +-- > kernel/audit.h | 17 +---- > kernel/auditsc.c | 167 ++++++++------------------------------------- > 7 files changed, 115 insertions(+), 201 deletions(-) Al/fs dev: any chance you guys can take a quick look at the fs relevant portions of this patchset and give it a ACK/NACK? I'll carry the patchset in the audit tree, but I'd like to see a thumbs up from someone in fs land before I merge these patches. Thanks, -Paul -- paul moore security @ redhat