linux-kernel - Re: [PATCH v3 39/51] x86/dumpstack: convert show_trace_log

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20160815164421.tfi7anwzaqrwtsvz@treble>
Date:	Mon, 15 Aug 2016 11:44:21 -0500
From:	Josh Poimboeuf <jpoimboe@...hat.com>
To:	Andy Lutomirski <luto@...capital.net>
Cc:	Thomas Gleixner <tglx@...utronix.de>,
	Ingo Molnar <mingo@...nel.org>,
	"H . Peter Anvin" <hpa@...or.com>, X86 ML <x86@...nel.org>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	Linus Torvalds <torvalds@...ux-foundation.org>,
	Steven Rostedt <rostedt@...dmis.org>,
	Brian Gerst <brgerst@...il.com>,
	Kees Cook <keescook@...omium.org>,
	Peter Zijlstra <peterz@...radead.org>,
	Frederic Weisbecker <fweisbec@...il.com>,
	Byungchul Park <byungchul.park@....com>,
	Nilay Vaish <nilayvaish@...il.com>
Subject: Re: [PATCH v3 39/51] x86/dumpstack: convert show_trace_log_lvl() to
 use the new unwinder

On Sun, Aug 14, 2016 at 01:13:54AM -0700, Andy Lutomirski wrote:
> On Fri, Aug 12, 2016 at 7:28 AM, Josh Poimboeuf <jpoimboe@...hat.com> wrote:
> > Convert show_trace_log_lvl() to use the new unwinder.  dump_trace() has
> > been deprecated.
> 
> >
> > Another change here is that callers of show_trace_log_lvl() don't need
> > to provide the 'bp' argument.  The unwinder already finds the relevant
> > frame pointer by unwinding until it reaches the first frame after the
> > provided stack pointer.
> 
> I still think that the best long-term solution is to change the sp and
> bp arguments to an optional state argument and to add a helper to
> capture the current state for future unwinding, but this is okay too.
> (If nothing else, this may improve DWARF's ability to recover function
> arguments and such that are available when the trace is requested but
> that are gone by the time the unwinder runs.  But mainly because it
> seems simpler and more direct to me and therefore seems like it will
> be less likely to get confused and skip too many frames.)
> 
> But I'm okay with this for now.

Long term, maybe something like an unwind_capture_state() helper might
be best (though I'm not yet convinced either way).  However it wouldn't
quite work with today's code because show_stack(), which is implemented
for all architectures, takes 'sp' as an argument.

-- 
Josh