From mboxrd@z Thu Jan 1 00:00:00 1970 From: hdoyu@nvidia.com (Hiroshi Doyu) Date: Mon, 1 Jul 2013 10:18:32 +0200 Subject: [PATCH 03/23] ARM: dt: tegra30: iommu: Add "nvidia,memory-clients" In-Reply-To: <1372244646.2986.17.camel@jlo-ubuntu-64.nvidia.com> References: <20130626101028.GD27083@manwe><20130626.133314.1520719774895092173.hdoyu@nvidia.com><1372244646.2986.17.camel@jlo-ubuntu-64.nvidia.com> Message-ID: <20130701.111832.2076981043593380845.hdoyu@nvidia.com> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org Joseph Lo wrote @ Wed, 26 Jun 2013 13:04:06 +0200: > On Wed, 2013-06-26 at 18:33 +0800, Hiroshi Doyu wrote: > > Thierry Reding wrote @ Wed, 26 Jun 2013 12:10:30 +0200: > > > > > * PGP Signed by an unknown key > > > > > > On Wed, Jun 26, 2013 at 12:28:06PM +0300, Hiroshi Doyu wrote: > > > > Add "nvidia,memory-clients" to identify which swgroup ID a device > > > > belongs to. > > > > > > Why not call the property "nvidia,swgid" instead? That seems a lot more > > > intuitive. > > > > Tegra powerdomain also needs this info. Joseph? > > Ah, yes. We need some kind of this information to know the relation in > the power domain when we wants to power down it. But it didn't upstream > yet and still in very early stage of the design. I believe you can > ignore that and do the good design here for your usage. We can re-use > the binding if it point to the same HW behind that. Ok, at least, I'll move this "tegra-swgid.h" to under "include/dt-bindings/memory/". This look to suite better than "include/dt-bindings/iommu/" since this isn't specific to iommu but others needs from memory client POV.