From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:34209) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1aYaCB-00036D-HF for qemu-devel@nongnu.org; Wed, 24 Feb 2016 09:13:40 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1aYaCA-0004Pb-LP for qemu-devel@nongnu.org; Wed, 24 Feb 2016 09:13:39 -0500 Received: from mx1.redhat.com ([209.132.183.28]:48665) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1aYaCA-0004PS-Gs for qemu-devel@nongnu.org; Wed, 24 Feb 2016 09:13:38 -0500 Date: Wed, 24 Feb 2016 14:13:36 +0000 From: Stefan Hajnoczi Message-ID: <20160224141336.GD23533@stefanha-x1.localdomain> References: <1455744555-22101-1-git-send-email-hollis_blanchard@mentor.com> <1455744555-22101-2-git-send-email-hollis_blanchard@mentor.com> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="C1iGAkRnbeBonpVg" Content-Disposition: inline In-Reply-To: <1455744555-22101-2-git-send-email-hollis_blanchard@mentor.com> Subject: Re: [Qemu-devel] [PATCH 2/2] trace: separate MMIO tracepoints from TB-access tracepoints List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Hollis Blanchard Cc: pbonzini@redhat.com, qemu-devel@nongnu.org --C1iGAkRnbeBonpVg Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Wed, Feb 17, 2016 at 01:29:15PM -0800, Hollis Blanchard wrote: > Memory accesses to code which has previously been translated into a TB sh= ow up > in the MMIO path, so that they may invalidate the TB. It's extremely conf= using > to mix those in with device MMIOs, so split them into their own tracepoin= t. >=20 > Signed-off-by: Hollis Blanchard >=20 > --- > It took many hours to figure out why some RAM accesses were coming throug= h the > MMIO path instead of being handled inline in the TBs. >=20 > On IRC, Paolo expressed some concern about performance, but ultimately ag= reed > that adding one conditional to an already heavy codepath wouldn't have mu= ch > impact. > --- > memory.c | 25 +++++++++++++++++++++++++ > trace-events | 2 ++ > 2 files changed, 27 insertions(+) Reviewed-by: Stefan Hajnoczi --C1iGAkRnbeBonpVg Content-Type: application/pgp-signature; name="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQEcBAEBAgAGBQJWzbqQAAoJEJykq7OBq3PIxyIH/Ri72jq2mxBs9js52axKlHXh VCeb36xoJsnGpTQoT/bfBsIBDbKMQQ6dQ5bR/P56urr1gCE70Rulo85kcEDvfR1N KRDRZtTmNYLD3mwC6qzXW0Zy/Lw8cu3YUVrPXqJVZ/CFORADXIfzoqjc5tK9KmmX eN9jU5R01sFBTlTHoF2pfz1ny9SU+igksvVwtm3Jpd28+eUeWCZ36xRgo1Q/WdAe mZ4UhzwCLhIjDPsVKuk1Gg+hZr/uzRYQdD2hj3Hbqskf2b8rJFzaaAAQnkh/aEC3 VFasJdbJ1LsAx0r/EXBvKyf6OZKBJVFvDVLkwpQqO1afQkr/95TDZ0BVLTZIqck= =pAWU -----END PGP SIGNATURE----- --C1iGAkRnbeBonpVg--