Re: [PATCH v4 2/2] ksm: provide support to use deferrable timers for scanner thread

From: Peter Zijlstra <peterz@infradead.org>
To: Hugh Dickins <hughd@google.com>
Cc: Chintan Pandya <cpandya@codeaurora.org>,
	akpm@linux-foundation.org, linux-mm@kvack.org,
	linux-arm-msm@vger.kernel.org, linux-kernel@vger.kernel.org,
	Thomas Gleixner <tglx@linutronix.de>,
	John Stultz <john.stultz@linaro.org>,
	Ingo Molnar <mingo@redhat.com>,
	Frederic Weisbecker <fweisbec@gmail.com>,
	Paul McKenney <paulmck@linux.vnet.ibm.com>
Subject: Re: [PATCH v4 2/2] ksm: provide support to use deferrable timers for scanner thread
Date: Wed, 10 Sep 2014 10:10:00 +0200	[thread overview]
Message-ID: <20140910081000.GN6758@twins.programming.kicks-ass.net> (raw)
In-Reply-To: <alpine.LSU.2.11.1409091225310.8432@eggly.anvils>

[-- Attachment #1: Type: text/plain, Size: 1345 bytes --]

On Tue, Sep 09, 2014 at 01:14:50PM -0700, Hugh Dickins wrote:
> On Mon, 8 Sep 2014, Peter Zijlstra wrote:
> > >  		switch_mm(oldmm, mm, next);
> > > +		wake_ksm = ksm_switch(mm);
> > 
> > Is this the right mm?
> 
> It's next->mm, that's the one I intended (though the patch might
> be equally workable using prev->mm instead: given free rein, I'd
> have opted for hooking into both prev and next, but free rein is
> definitely not what should be granted around here!).
> 
> > We've just switched the stack,
> 
> I thought that came in switch_to() a few lines further down,
> but don't think it matters for this.

Ah, yes. Got my task and mm separation messed up.

> > so we're looing at next->mm when we switched away from current.
> > That might not exist anymore.
> 
> I fail to see how that can be.  Looking at the x86 switch_mm(),
> I can see it referencing (unsurprisingly!) both old and new mms
> at this point, and no reference to an mm is dropped before the
> ksm_switch().  oldmm (there called mm) is mmdropped later in
> finish_task_switch().

Well, see the above confusion about switch_mm vs switch_to :-/

So if this were switch_to(), we'd see next->mm as before the last
context switch. And since that switch fully happened, it would also
already have done the finish_task_switch() -> mmdrop().

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]