linux-kernel - Re: [PATCH 1/2] perf: Add persistent event facilities

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: <20120518125505.GK20215@aftab.osrc.amd.com>
Date:	Fri, 18 May 2012 14:55:05 +0200
From:	Borislav Petkov <bp@...64.org>
To:	Ingo Molnar <mingo@...nel.org>
Cc:	Peter Zijlstra <peterz@...radead.org>,
	Frederic Weisbecker <fweisbec@...il.com>,
	Ingo Molnar <mingo@...e.hu>,
	Steven Rostedt <rostedt@...dmis.org>,
	LKML <linux-kernel@...r.kernel.org>,
	Borislav Petkov <borislav.petkov@....com>
Subject: Re: [PATCH 1/2] perf: Add persistent event facilities

On Fri, May 18, 2012 at 01:59:57PM +0200, Ingo Molnar wrote:
> 
> * Peter Zijlstra <peterz@...radead.org> wrote:
> 
> > On Fri, 2012-05-18 at 13:03 +0200, Borislav Petkov wrote:
> > > >  - you allow an arbitrary place in debugfs; this might make finding
> > > > them 'interesting'. Should we put them all in the same place?
> > > 
> > > My take on this is that we want to be able to make the same events we
> > > have now, persistent. Basically not trace for the duration of a child
> > > process but in a process-agnostic way, system-wide.
> > 
> > This would argue against per-task persistent events.. 
> 
> Yeah, 'persistency' is something that's per CPU at minimum.

Yeah, forget the per-task thing - that's me not understanding the code
fully. They need to be per CPU and "taskless."

> > Note the plural there, it might be very desirable to allow 
> > multiple events into a single persistent buffer.
> 
> very, very, very much so. One (per CPU) buffer, with many events 
> multiplexed into it.

Which begs the other question: mce could use buffers which are RO - at
least, I don't see a usecase where we want to delete entries from them -
but other persistent events users might want to delete those events to
free up the buffers.

It is an interesting question how we handle that RO/RW thing? Maybe per
event and do something of a kernel-side filtering which Peter didn't
like in Robert's IBS stuff.

Hmmm...

-- 
Regards/Gruss,
Boris.

Advanced Micro Devices GmbH
Einsteinring 24, 85609 Dornach
GM: Alberto Bozzo
Reg: Dornach, Landkreis Muenchen
HRB Nr. 43632 WEEE Registernr: 129 19551
--
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/