From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753315AbcDZVKp (ORCPT ); Tue, 26 Apr 2016 17:10:45 -0400 Received: from sauhun.de ([89.238.76.85]:52364 "EHLO pokefinder.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752469AbcDZVKn (ORCPT ); Tue, 26 Apr 2016 17:10:43 -0400 Date: Tue, 26 Apr 2016 23:10:38 +0200 From: Wolfram Sang To: Jan Glauber Cc: linux-kernel@vger.kernel.org, linux-i2c@vger.kernel.org, David Daney , Peter Swain Subject: Re: [PATCH v7 08/15] i2c: octeon: Improve performance if interrupt is early Message-ID: <20160426211038.GC4587@katana> References: <1461594824-7215-1-git-send-email-jglauber@cavium.com> <1461594824-7215-9-git-send-email-jglauber@cavium.com> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="oJ71EGRlYNjSvfq7" Content-Disposition: inline In-Reply-To: <1461594824-7215-9-git-send-email-jglauber@cavium.com> User-Agent: Mutt/1.5.24 (2015-08-30) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org --oJ71EGRlYNjSvfq7 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Mon, Apr 25, 2016 at 04:33:37PM +0200, Jan Glauber wrote: > From: Peter Swain >=20 > There is a race between the TWSI interrupt and the condition > that is required before proceeding: >=20 > Low-level: interrupt flag bit must be set > High-level controller: valid bit must be clear >=20 > If the interrupt comes too early and the condition is not met > the wait will time out, and the transfer is aborted leading > to very poor performance. >=20 > To avoid this race retry for the condition ~80 =C2=B5s later. > The retry is avoided on the very first invocation of > wait_event_timeout() (which tests the condition before entering > the wait and is therefore always wrong in this case). >=20 > EEPROM reads on 100kHz i2c now measure ~5.2kB/s, about 1/2 what's > achievable, and much better than the worst-case 100 bytes/sec before. >=20 > While at it remove the debug print from the low-level wait function. >=20 > Signed-off-by: Peter Swain > Signed-off-by: Jan Glauber Applied to for-next, thanks! Assigned 'true' instead of '1' to bool... --oJ71EGRlYNjSvfq7 Content-Type: application/pgp-signature; name="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQIcBAEBAgAGBQJXH9lOAAoJEBQN5MwUoCm2ttAP/Ry083VaTvtlryRiGUkAtyu3 QDVDrrPsHVGf00zuVndaSFGjSnBra222GIin7UmWHiGV0BYvbds4F3dLZgge5s2K XPlLaL1b2g6zv+rhR7/cmVSp7qvZERYNqco0nTtsJ0rYvPrOaBzw3odDZVU5d/Wv ibHsruAg9l9X+8u1Sl2Dvx8qtQeEWG/TgdGk9Xzuv/lGkVc+Gykgqu0Gw1dXYn/9 1b0QICEoC/1KWzy1r8gNi2tJ67N12nZQoRrTot7DexV8Wz3PFfF+WRziJ0BasQ4y RIUaE5jk7lcEA3tTuvbMHcARYFKbgxrCGCTGkVGNri1+gXki4duMv2x0/Egw1p/P qA4SFobdTH1T5MBW4Or9uTuUMWY6deNt4tgidP6IgwFXj1xS6wcSq2ChpyIruREe lGLOdrgObD8HZKIKRPfQR58rap6Z9GKgs1MAJ4bmb6w9j73qyIj0N8gzcJN2gpt0 4RQLAy+Jae+Pkqd+rayg4m4BDg3+Hx3vM3JRjg58ftsOHyLLTYJj5PmhD3ew5xVZ P0TkxnPkDP1vYHRJunVJrnw4ObsuFYTjQEGReqpH3sZXY1JkT3HShRGtyuXJB5y4 8jpvdTdRNhfI0NDhqvS8Dux0B85GL5TpLPUmJ4B/h/dPoE+IiuqE9VoI0q6mVoHO prqAAtcQtYctPiLppZ1m =qFYf -----END PGP SIGNATURE----- --oJ71EGRlYNjSvfq7--