From mboxrd@z Thu Jan 1 00:00:00 1970 From: tomasz.figa@gmail.com (Tomasz Figa) Date: Wed, 21 Aug 2013 00:32:06 +0200 Subject: [PATCH 01/16] pwm: samsung: Update DT bindings documentation to cover clocks In-Reply-To: <5213D2D6.50603@wwwdotorg.org> References: <1377019903-14614-1-git-send-email-t.figa@samsung.com> <1377019903-14614-2-git-send-email-t.figa@samsung.com> <5213D2D6.50603@wwwdotorg.org> Message-ID: <11380970.mYsvoRAzs5@flatron> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On Tuesday 20 of August 2013 14:34:30 Stephen Warren wrote: > On 08/20/2013 11:31 AM, Tomasz Figa wrote: > > PWM driver consumes at least one and up to three clocks, which need to > > be specified in device tree when used. This patch updates bindings > > documentation to add information about clocks. > > > > diff --git a/Documentation/devicetree/bindings/pwm/pwm-samsung.txt > > b/Documentation/devicetree/bindings/pwm/pwm-samsung.txt > > > > +- clock-names: should contain all following required clock names: > > + - "timers" - PWM base clock used to generate PWM signals, > > + and any subset of following optional clock names: > > + - "pwm-tclk0" - first external PWM clock source, > > + - "pwm-tclk1" - second external PWM clock source. > > + Note that not all IP variants allow using all external clock > > sources. + Refer to SoC documentation to learn which clock source > > configurations + are available. > > It might be nice to explicitly enumerate which variants (or rather, I > suppose which exact compatible values) support which optional clocks. > However, I suppose it's fine to just say "go read the HW manual" > instead. I don't have enough information about some of the variants, so I decided to put it this way, instead of making an incomplete list or speculating about things. > So, this patch, > Acked-by: Stephen Warren Thanks. > (although this patch isn't backwards-compatible since before now, DT > nodes didn't need to provide any entries in clocks) The thing is that the clocks were supposed to be there from the beginning, but in a hurry I forgot to specify them in original bindings documentation and things were running over compatibility clkdev aliases anyway, so nobody (including me) noticed. Best regards, Tomasz