Date: Thu, 27 Oct 2016 10:46:38 +0200
From: Pavel Machek <pavel@ucw.cz>
To: Peter Zijlstra <peterz@infradead.org>
Cc: acme@redhat.com, kernel list <linux-kernel@vger.kernel.org>,
        mingo@redhat.com, alexander.shishkin@linux.intel.com
Subject: Re: Getting interrupt every million cache misses
Message-ID: <20161027084637.GA19469@amd>
References: <20161026204748.GA11177@amd>
 <20161027082801.GE3568@worktop.programming.kicks-ass.net>
MIME-Version: 1.0
Content-Type: multipart/signed; micalg=pgp-sha1;
        protocol="application/pgp-signature"; boundary="17pEHd4RhPHOinZp"
Content-Disposition: inline
In-Reply-To: <20161027082801.GE3568@worktop.programming.kicks-ass.net>
User-Agent: Mutt/1.5.23 (2014-03-12)
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1874
Lines: 62


--17pEHd4RhPHOinZp
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

Hi!

> > I'd like to get an interrupt every million cache misses... to do a
> > printk() or something like that. As far as I can tell, modern hardware
> > should allow me to do that. AFAICT performance events subsystem can do
> > something like that, but I can't figure out where the code is / what I
> > should call.
> >=20
> > Can someone help?
>=20
> Can you go back one step and explain why you would want this? What use
> is a printk() on every 1e6-th cache miss.

First, thanks for quick reply.

And actually, printk() is not needed, udelay(50msec) is. Reason is,
that DRAM becomes unreliable if about milion cache misses happen in
under 64msec -- so I'd like to slow the system down in such cases to
prevent bug from biting me.

(Details are here
https://googleprojectzero.blogspot.cz/2015/03/exploiting-dram-rowhammer-bug=
-to-gain.html
). Bug is exploitable to get local root; it is also exploitable to
gain local code execution from javascript... so it is rather severe.

> That is, why doesn't:
>=20
>  $ perf record -e cache-misses -c 1000000 -a -- sleep 5
>=20
> suffice?

Thanks for the pointer... I'd really like to do this from kernel, so
that I can "almost synchronously" stop the execution when excessive
cache isses happen.

Best regards,
									Pavel
--=20
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blo=
g.html

--17pEHd4RhPHOinZp
Content-Type: application/pgp-signature; name="signature.asc"
Content-Description: Digital signature

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1

iEYEARECAAYFAlgRvuwACgkQMOfwapXb+vJHnwCbB3wD0dGv+II9dHO5Yz3bzFqI
W7sAoLJMrZiptolUp7tjv+FRBHLB3mYp
=Q1VF
-----END PGP SIGNATURE-----

--17pEHd4RhPHOinZp--