linux-kernel - Re: [RFC/PATCH] ftrace: add set_graph

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: <20131011085809.GD4154@osiris>
Date:	Fri, 11 Oct 2013 10:58:09 +0200
From:	Heiko Carstens <heiko.carstens@...ibm.com>
To:	Namhyung Kim <namhyung@...nel.org>
Cc:	Steven Rostedt <rostedt@...dmis.org>,
	Frederic Weisbecker <fweisbec@...il.com>,
	Ingo Molnar <mingo@...nel.org>,
	LKML <linux-kernel@...r.kernel.org>,
	Namhyung Kim <namhyung.kim@....com>,
	Martin Schwidefsky <schwidefsky@...ibm.com>
Subject: Re: [RFC/PATCH] ftrace: add set_graph_notrace filter

On Fri, Oct 11, 2013 at 05:34:19PM +0900, Namhyung Kim wrote:
> Hi Heiko,
> 
> On Fri, 11 Oct 2013 09:21:56 +0200, Heiko Carstens wrote:
> > On Fri, Oct 11, 2013 at 12:17:17AM -0400, Steven Rostedt wrote:
> >> Isn't there a way we could pass the state? Hmm, I think we could use
> >> depth to do that. As depth is a pointer to trace.depth and not used
> >> before then. We could make it negative and then check that.
> >> 
> >> /me looks at other archs.
> >> 
> >> Darn it, s390 calls ftrace_push_return_trace() before
> >> ftrace_graph_entry(). They got fancy, as they must have noticed that
> >> trace.depth is set by ftrace_push_return_trace() and they just figured
> >> to let the ftrace_push_return_trace() do the work.
> >> 
> >> Hmm, we could add a config to do the check on one side or the other
> >> depending on how the arch handles it.
> >> 
> >> It needs to be well commented though.
> >
> > I don't remember I why decided to do that more than four years ago, however
> > on the other hand I really don't think we need a config for this.
> > So I just committed the patch below... it compiles and doesn't crash
> > immediately, so must be good.
> > It will be merged during the next merge window.
> 
> Nice!
> 
> >
> > From 8c539d10400ee2efe000d324497a0661a2143dbf Mon Sep 17 00:00:00 2001
> > From: Heiko Carstens <heiko.carstens@...ibm.com>
> > Date: Fri, 11 Oct 2013 08:55:57 +0200
> > Subject: [PATCH] s390/ftrace: prepare_ftrace_return() function call order
> >
> > Steven Rostedt noted that s390 is the only architecture which calls
> > ftrace_push_return_trace() before ftrace_graph_entry() and therefore has
> > the small advantage that trace.depth gets initialized automatically.
> 
> Well, this is not true.  AFAICS arm, x86 and powerpc have same order, and
> other archs have different (inverted) order.  And it'd really great if
> all archs agreed on a single ordering.

Ooops. Anyway.. I just realized that we also have an out-of-tree patch
that relies on the now new order on s390. So that patch got a bit smaller
as an added bonus ;)

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/