From mboxrd@z Thu Jan 1 00:00:00 1970 From: Mark Rutland Subject: Re: [PATCH 4/7] spi: pl022: attempt to get sspclk by name Date: Wed, 12 Feb 2014 11:47:40 +0000 Message-ID: <20140212114740.GE21992@e106331-lin.cambridge.arm.com> References: <1392118632-11312-1-git-send-email-mark.rutland@arm.com> <201402111508.06267.arnd@arndb.de> <20140212103329.GC21992@e106331-lin.cambridge.arm.com> <201402121221.51233.arnd@arndb.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <201402121221.51233.arnd-r2nGTMty4D4@public.gmane.org> Sender: devicetree-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Arnd Bergmann Cc: Mark Brown , "devicetree-u79uwXL29TY76Z2rM5mHXA@public.gmane.org" , "linux-arm-kernel-IAPFreCvJWM7uuMidbF8XUB+6BGkLq7r@public.gmane.org" , "robh+dt-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org" , Pawel Moll , Linus Walleij List-Id: devicetree@vger.kernel.org On Wed, Feb 12, 2014 at 11:21:50AM +0000, Arnd Bergmann wrote: > On Wednesday 12 February 2014, Mark Rutland wrote: > > To me it feels odd to require the last clock in the list (apb_pclk) to > > be named, and the rest to be in a particular order. For the dt case it > > seems saner to add new clocks with names as it allows arbitrary subsets > > of clocks to be wired up and described (though obviously in this case a > > missing sspclk would be problematic). > > Yes, good point about the missing clocks, and I also agree mixing ordered > and named clocks in one device is rather odd and can lead to trouble. > > I guess there isn't really a good way out here, and I certainly won't > ask for it to be done one way or the other if someone else has a > good argument which way it should be implemented. Having thought about it, all dts that for the ssp_pclk must have some name for the sspclk (though the specific name is arbitrary). I could get the driver to try each of those before falling back to the index (perhaps with a helper that takes a list of known aliases), so all existing dts (that we are aware of) would work with the clock requested by name. I assume that for the non-dt case it's possible to name clock inputs to a device without the clock being associated with the name globally? If so we could get rid of the index usage entirely in this case. Does that sound workable? Thanks, Mark. -- To unsubscribe from this list: send the line "unsubscribe devicetree" in the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org More majordomo info at http://vger.kernel.org/majordomo-info.html From mboxrd@z Thu Jan 1 00:00:00 1970 From: mark.rutland@arm.com (Mark Rutland) Date: Wed, 12 Feb 2014 11:47:40 +0000 Subject: [PATCH 4/7] spi: pl022: attempt to get sspclk by name In-Reply-To: <201402121221.51233.arnd@arndb.de> References: <1392118632-11312-1-git-send-email-mark.rutland@arm.com> <201402111508.06267.arnd@arndb.de> <20140212103329.GC21992@e106331-lin.cambridge.arm.com> <201402121221.51233.arnd@arndb.de> Message-ID: <20140212114740.GE21992@e106331-lin.cambridge.arm.com> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On Wed, Feb 12, 2014 at 11:21:50AM +0000, Arnd Bergmann wrote: > On Wednesday 12 February 2014, Mark Rutland wrote: > > To me it feels odd to require the last clock in the list (apb_pclk) to > > be named, and the rest to be in a particular order. For the dt case it > > seems saner to add new clocks with names as it allows arbitrary subsets > > of clocks to be wired up and described (though obviously in this case a > > missing sspclk would be problematic). > > Yes, good point about the missing clocks, and I also agree mixing ordered > and named clocks in one device is rather odd and can lead to trouble. > > I guess there isn't really a good way out here, and I certainly won't > ask for it to be done one way or the other if someone else has a > good argument which way it should be implemented. Having thought about it, all dts that for the ssp_pclk must have some name for the sspclk (though the specific name is arbitrary). I could get the driver to try each of those before falling back to the index (perhaps with a helper that takes a list of known aliases), so all existing dts (that we are aware of) would work with the clock requested by name. I assume that for the non-dt case it's possible to name clock inputs to a device without the clock being associated with the name globally? If so we could get rid of the index usage entirely in this case. Does that sound workable? Thanks, Mark.