From mboxrd@z Thu Jan 1 00:00:00 1970 From: arnd@arndb.de (Arnd Bergmann) Date: Mon, 01 Sep 2014 14:11:12 +0200 Subject: [PATCH 6/6 v4] ARM: realview: basic device tree implementation In-Reply-To: <20140901120313.GI30401@n2100.arm.linux.org.uk> References: <1406294628-16079-1-git-send-email-linus.walleij@linaro.org> <9626823.572BIYcQL9@wuerfel> <20140901120313.GI30401@n2100.arm.linux.org.uk> Message-ID: <5409848.tU8vLO6RTd@wuerfel> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On Monday 01 September 2014 13:03:13 Russell King - ARM Linux wrote: > On Fri, Jul 25, 2014 at 05:58:24PM +0200, Arnd Bergmann wrote: > > On Friday 25 July 2014 16:50:10 Russell King - ARM Linux wrote: > > > > > > I toyed with the idea of adding the standard cache size specifications > > > to the L2C code, it sounds like there's a reason to do that now. > > > > > > Let's not mess around with adding stuff like the above to work around > > > the L2 code not parsing the appropriate specs from DT, let's do the > > > right thing and add bindings to allow DT to provide the cache geometry. > > > > Yes, please! > > > > That would actually reduce the realview to just the dma_zone_size > > definition, and I think someone also posted a patch to get that from > > DT recently. Having realview booting with an empty board file would > > be ideal. > > I did have a patch in my L2C cleanup series to provide these properties, > but it got discarded - even though we have existing platforms which > provide these values. > > The reason for that was because requiring new properties causes a DT > breakage - previous versions of the DT suddenly stop working because > the values necessary to setup the L2C have been removed from the > kernel, and the older DT files do not contain the neccessary values > either. But in case of realview or any other platform that does not have mainline DT support, nothing would break. There is no need to remove the settings from the kernel, I would just prefer not to add any other ones. Arnd