From mboxrd@z Thu Jan 1 00:00:00 1970 From: Russell King - ARM Linux Subject: Re: [PATCHv2 2/2] pwm: imx: support polarity inversion Date: Thu, 23 Jan 2014 12:33:41 +0000 Message-ID: <20140123123341.GW15937@n2100.arm.linux.org.uk> References: <1389859585-14006-1-git-send-email-LW@KARO-electronics.de> <1389859585-14006-3-git-send-email-LW@KARO-electronics.de> <20140116160356.GY16215@pengutronix.de> <20140123083714.3c6e86ae@ipc1.ka-ro> <20140123115203.GV15937@n2100.arm.linux.org.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Content-Disposition: inline In-Reply-To: <20140123115203.GV15937@n2100.arm.linux.org.uk> Sender: linux-doc-owner@vger.kernel.org To: Lothar =?iso-8859-1?Q?Wa=DFmann?= Cc: Mark Rutland , linux-pwm@vger.kernel.org, Sascha Hauer , Pawel Moll , Ian Campbell , Sascha Hauer , linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org, devicetree@vger.kernel.org, Rob Herring , Thierry Reding , Rob Landley , Kumar Gala , Shawn Guo , linux-arm-kernel@lists.infradead.org List-Id: devicetree@vger.kernel.org On Thu, Jan 23, 2014 at 11:52:03AM +0000, Russell King - ARM Linux wrot= e: > On Thu, Jan 23, 2014 at 08:37:14AM +0100, Lothar Wa=DFmann wrote: > > This wouldn't buy much without a material change to of_pwm_get(). > > The function of_parse_phandle_with_args() called by of_pwm_get() > > requires the number of args in the pwms property be greater or equa= l to > > the #pwm-cells property in the pwm node. Thus, the interesting case= of > > having #pwm-cells =3D <3> without changing the existing users is > > prohibited by of_parse_phandle_with_args(). >=20 > I really don't think that's a problem we need to be concerned with at > the moment. What we need is for the kernel to be able to parse files > with #pwm-cells =3D <2> with the pwms property containing two argumen= ts, > and when they're updated to #pwm-cells =3D <3> with the pwms property > containing three arguments. >=20 > Yes, that means all the board dt files need to be updated at the same > time to include the additional argument, but I don't see that as a bi= g > problem. >=20 > What we do need to do is to adjust the PWM parsing code such that it'= s > possible to use either specification without causing any side effects= =2E >=20 > I would test this, but as u-boot is rather fscked at the moment and t= he > networking has broken on my cubox-i as a result... and it seems that = the > u-boot developers have pissed off cubox-i u-boot hackers soo much tha= t > they've dropped u-boot in favour of barebox... Oh, and another reason... the u-boot video settings are totally and utt= erly buggered to the point that it doesn't produce correct timings, and it s= eems that u-boot people have zero interest in fixing that, so u-boot mainlin= e is basically refusing to fix this - another reason to stay away from it. (1024x768 @ 60Hz produces 70Hz refresh on iMX6Q here - I've seen it pro= duce 51Hz on iMX6S, both of which are far enough out that lots of display de= vices will not accept it as a valid signal.) --=20 =46TTC broadband for 0.8mile line: 5.8Mbps down 500kbps up. Estimation in database were 13.1 to 19Mbit for a good line, about 7.5+ for a bad. Estimate before purchase was "up to 13.2Mbit".