From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Tobin C. Harding" Subject: Re: [RFC 0/3] kallsyms: don't leak address when printing symbol Date: Tue, 28 Nov 2017 12:50:41 +1100 Message-ID: <20171128015041.GT17858@eros> References: <1511821819-5496-1-git-send-email-me@tobin.cc> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: kernel-hardening@lists.openwall.com, LKML , Network Development , Steven Rostedt , Tycho Andersen To: Kees Cook Return-path: Content-Disposition: inline In-Reply-To: Sender: linux-kernel-owner@vger.kernel.org List-Id: netdev.vger.kernel.org On Mon, Nov 27, 2017 at 04:52:21PM -0800, Kees Cook wrote: > On Mon, Nov 27, 2017 at 2:30 PM, Tobin C. Harding wrote: > > This is an RFC for two reasons. > > > > 1) I don't know who this patch set may break? > > 2) Patch set includes a function that is not called. Function is there > > to facilitate fixing breakages. > > > > _If_ no one gets broken then we can remove the unused function. > > > > Thanks for looking at this. > > > > Currently if a pointer is printed using %p[ssB] and the symbol is not > > found (kallsyms_lookup() fails) then we print the actual address. This > > potentially leaks kernel addresses. We could instead print something > > _safe_. If kallsyms is made to return an error then callers of > > sprint_symbol() can decide what to do. > > > > In the case of vsprintf we can print '' (patch 2). > > > > In the case of trace we want the address so we can check the return code > > and print the address if no symbol is found (patch 3). > > > > Design for this set loosely suggested by Steve Rostedt (so as not to > > break ftrace). > > If tracing is the only place this is needed, this series seems reasonable to me. Noob question: how do we _know_ this. In other words how do we know no userland tools rely on the current behaviour? No stress to answer Kees, this is a pretty general kernel dev question. thanks, Tobin.