Re: [RFC 1/3] Unified NMI delayed call mechanism

From: huang ying <huang.ying.caritas@gmail.com>
To: Ingo Molnar <mingo@elte.hu>
Cc: Andi Kleen <andi@firstfloor.org>,
	Peter Zijlstra <peterz@infradead.org>,
	Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>,
	Huang Ying <ying.huang@intel.com>,
	"Fr??d??ric Weisbecker" <fweisbec@gmail.com>,
	Don Zickus <dzickus@redhat.com>, "H.PeterA" <hpa@zytor.com>,
	linux-kernel@vger.kernel.org
Subject: Re: [RFC 1/3] Unified NMI delayed call mechanism
Date: Sat, 19 Jun 2010 22:07:42 +0800	[thread overview]
Message-ID: <AANLkTilyV7XkZCsD5uNGhdTAGUUEvR5xIuvbwQGiMYvP@mail.gmail.com> (raw)
In-Reply-To: <20100619105352.GA10186@elte.hu>

On Sat, Jun 19, 2010 at 6:53 PM, Ingo Molnar <mingo@elte.hu> wrote:
> The proper, generic approach would be to enable softirq notifications (on x86)
> from NMI contexts as well (it's actually possible without overhead),

Yes. I will do that. And I think self interrupt can be used as the
short-cut for soft_irq if available. The next soft_irq may be too late
if there is too few interrupts.

> and to
> extend user return notifiers with the logical next step: nmi return notifiers.
> If presented in such a form then those could use softirqs for atomic callbacks
> and per cpu kthreads for sleepable callbacks, etc.

NMI return notifiers fired in soft_irq?

Best Regards,
Huang Ying