From mboxrd@z Thu Jan 1 00:00:00 1970 From: Tomi Valkeinen Subject: Re: v3.4-rc4 DSS PM problem (Was: Re: Problems with 3.4-rc5) Date: Thu, 24 May 2012 15:35:02 +0300 Message-ID: <1337862902.13723.4.camel@deskari> References: <1336033721.14378.2.camel@deskari> <1336050442.14378.10.camel@deskari> <1336139415.2552.4.camel@deskari> <1336140072.2552.6.camel@deskari> <1336143281.2552.21.camel@deskari> <1336143500.2552.23.camel@deskari> <1336483598.5761.45.camel@deskari> <1336982138.2532.32.camel@lappyti> <1337159300.7692.22.camel@deskari> <4FB37CE5.1040707@ti.com> Mime-Version: 1.0 Content-Type: multipart/signed; micalg="pgp-sha1"; protocol="application/pgp-signature"; boundary="=-nvYyGW0lJu9boJoIwucg" Return-path: Received: from na3sys009aog116.obsmtp.com ([74.125.149.240]:38229 "EHLO na3sys009aog116.obsmtp.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751876Ab2EXMfJ (ORCPT ); Thu, 24 May 2012 08:35:09 -0400 Received: by lban1 with SMTP id n1so7115199lba.24 for ; Thu, 24 May 2012 05:35:06 -0700 (PDT) In-Reply-To: <4FB37CE5.1040707@ti.com> Sender: linux-omap-owner@vger.kernel.org List-Id: linux-omap@vger.kernel.org To: "Cousson, Benoit" Cc: Paul Walmsley , Joe Woodward , khilman@ti.com, Archit Taneja , linux-omap@vger.kernel.org --=-nvYyGW0lJu9boJoIwucg Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable On Wed, 2012-05-16 at 12:09 +0200, Cousson, Benoit wrote: > Hi Tomi, >=20 > On 5/16/2012 11:08 AM, Tomi Valkeinen wrote: > > Disabling DPLL3 autoidle fixes the problem. Disabling DPLL4 autoidle > > doesn't affect the problem. >=20 > The issues your are facing seems to be the well known DSS low power=20 > refresh mode we've been trying to use since OMAP2 :-). Hmm, so are you saying no one has managed to get fifomerge and autoidle working reliably? If so, no point for me to even try it =3D). If so, I wonder which is better to have: fifomerge or autoidle...=20 > > I also suspect that this could be just a plain DSS bug. The default FIF= O > > low/high thresholds are 960/1023 bytes (i.e. DSS starts refilling the > > FIFO when there are 960 or less bytes in the fifo, and stops at 1023. > > The fifo is 1024 bytes). The values are calculated with fifo_size - > > burst_size and fifo_size - 1. > > > > We are now using FIFO merge features, which combines multiple fifos int= o > > one when possible, making the fifo size 1024*3 =3D 3072. Using the same > > low threshold and increasing the high threshold to 960/3071 works fine. > > Changing the high threshold to 3008 causes underflows. Increasing the > > low threshold to ~1600 makes DSS work again. >=20 > That's weird, in theory what should matter is only the diff between the= =20 > high and low. Well the low value should be as high as possible as well= =20 > to support the wakeup latency. Yep. That makes me think there's some kind of problem with DSS accessing the memory with particular fifo thresholds. Tomi --=-nvYyGW0lJu9boJoIwucg Content-Type: application/pgp-signature; name="signature.asc" Content-Description: This is a digitally signed message part Content-Transfer-Encoding: 7bit -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.11 (GNU/Linux) iQIcBAABAgAGBQJPvir2AAoJEPo9qoy8lh71WzIP/0aEa48COv1rt+oYnixodh+T LvFosvs4QF2v/Eb4gpx+OgfCBT/9rt6iTifeDxjhMyKK4G4Sj8Yn/H4mCVx/C9wC Z8E2QKpJQWkCar2bL/jEvNrmU9b2OIPZPwyR9Qpu582WF7RasWfGZzqsApz7A66o uQNYSNEm6S+hO0rgHUhgaOQU4ZIkzPPOqaw1XchBC/uyIq3lmnWMCzCU6fYwYhof WK0Qna1lX10E4utcMTwRpjfpCjuHTVFTTvPlfv1f3siUnSEuxhSb1tn1d4ma4eud TDMI3iLmOZVYkyaR9wFqhThRJG4MpiXHEH3p7cL9gCktYPcntuPrWR66iFsudMud zuGbTlDCbsrwIK0oI9yAblb6541+UpjsPY/7MPndBxiBz7+t6eVfwB2V381PlMD7 s8M/GKU/t29eGrOBx9RtCy8MqXHhLkOd00LGccyeNJSXZ2qn7Gwk5x4zEOZfPZ/a v+eMgJ4kBlcDvP9NOaYYslVLBsTR2P30mYsTiC7L4g5U9GI0f57OpfWzJC7sv3RS C68XmkVI5min5cDiqsOgu0IWH/aDXYU8Dvi0WJuYCXv5Bo1zY9f+wHCttpO7+Z5y HjU8jgBfLv9TxcTN2lkRLVBdEcR0q8dqPBARIyMMNlaeMtWNa98TZUA8VRlYvZWS tDEqRhIwqW1AdTnSN5Eh =g3cW -----END PGP SIGNATURE----- --=-nvYyGW0lJu9boJoIwucg--