[Top] [All Lists]

Re: [PATCH 3/5] v2 seccomp_filters: Enable ftrace-based system call filt

To: Peter Zijlstra <>
Subject: Re: [PATCH 3/5] v2 seccomp_filters: Enable ftrace-based system call filtering
From: Ingo Molnar <>
Date: Sun, 29 May 2011 22:17:13 +0200
Cc: Thomas Gleixner <>, Will Drewry <>, Steven Rostedt <>, Frederic Weisbecker <>, James Morris <>,, Eric Paris <>,,, "Serge E. Hallyn" <>, Ingo Molnar <>, Andrew Morton <>, Tejun Heo <>, Michal Marek <>, Oleg Nesterov <>, Jiri Slaby <>, David Howells <>, Russell King <>, Michal Simek <>, Ralf Baechle <>, Benjamin Herrenschmidt <>, Paul Mackerras <>, Martin Schwidefsky <>, Heiko Carstens <>,, Paul Mundt <>, "David S. Miller" <>, "H. Peter Anvin" <>,, linux-arm-kernel <>,,,,,, Linus Torvalds <>
In-reply-to: <1306345402.21578.100.camel@twins>
Original-recipient: rfc822;
References: <> <> <> <> <> <1306254027.18455.47.camel@twins> <> <alpine.LFD.2.02.1105242239230.3078@ionos> <> <1306345402.21578.100.camel@twins>
User-agent: Mutt/1.5.20 (2009-08-17)
* Peter Zijlstra <> wrote:

> But face it, you can argue until you're blue in the face,

That is not a technical argument though - and i considered and 
answered every valid technical argument made by you and Thomas.
You were either not able to or not willing to counter them.

> [...] but both tglx and I will NAK any and all patches that extend 
> perf/ftrace beyond the passive observing role.

The thing is, perf is *already* well beyond the 'passive observer' 
role: we already generate lots of 'action' in response to events. We 
generate notification signals, we write events - all of which can 
(and does) modify program behavior.

So what's your point? There's no "passive observer" role really - 
it's apparently just that you dislike this use of instrumentation 
while you approve of other uses.



<Prev in Thread] Current Thread [Next in Thread>