linux-kernel - Re: bisected: 'perf top' causing soft lockups under Xen

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 for Android: free password hash cracker in your pocket

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

Date:	Fri, 10 Feb 2012 18:24:29 +0100
From:	Peter Zijlstra <a.p.zijlstra@...llo.nl>
To:	Konrad Rzeszutek Wilk <konrad.wilk@...cle.com>
Cc:	Steven Noonan <steven@...inklabs.net>,
	linux-kernel@...r.kernel.org, Paul Mackerras <paulus@...ba.org>,
	Ingo Molnar <mingo@...e.hu>,
	Arnaldo Carvalho de Melo <acme@...stprotocols.net>,
	Jeremy Fitzhardinge <jeremy@...p.org>
Subject: Re: bisected: 'perf top' causing soft lockups under Xen

On Fri, 2012-02-10 at 11:12 -0500, Konrad Rzeszutek Wilk wrote:
> Hm, during bootup what does perf say about the CPU availability? Is
> it that it can only do perf via NMI's? 

We used to pass down if the event came from NMI context or not in order
to avoid doing certain things that were not NMI-safe, like wakeups.

That patch changed the code so that it always assumed the worst case
(NMI) and thus avoids doing such things.

All non-NMI-safe operations are pushed out to IRQ context through
kernel/irq_work.c. If there's no particular platform support for that,
it falls back to doing that work from the timer tick.

Now that all _should_ just work but clearly there's a problem there.

I haven't tried making sense of the traces yet..

--
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/