From: Grant Likely Subject: Re: [PATCH 1/2 v2] of: move phandle/ihandle into types.h and export to userspace Date: Fri, 8 Oct 2010 12:27:45 -0600 Message-ID: <20101008182745.GI3863@angua.secretlab.ca> References: <20101008171210.GE3863@angua.secretlab.ca> <20101008.101759.193728130.davem@davemloft.net> <20101008173650.GH3863@angua.secretlab.ca> <20101008.104557.59696706.davem@davemloft.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <20101008.104557.59696706.davem@davemloft.net> Sender: sparclinux-owner@vger.kernel.org To: David Miller Cc: dilinger@queued.net, sparclinux@vger.kernel.org, linux-kernel@vger.kernel.org, sam@ravnborg.org, devicetree-discuss@lists.ozlabs.org List-Id: devicetree@vger.kernel.org On Fri, Oct 08, 2010 at 10:45:57AM -0700, David Miller wrote: > From: Grant Likely > Date: Fri, 8 Oct 2010 11:36:50 -0600 > > > Weird. Yeah, no other platforms expect to get a phandle type > > definition from the kernel headers. The only thing driving this > > conversation is from arch/sparc/include/asm/Kbuild: > > > > header-y += openprom.h > > It was probably at one point for the sake of asm/openpromio.h but > that header has no dependencies on openprom.h > > The only hit I can find in google code search, for non-kernel code, is > the SILO bootloader. > > But that tree includes it's own copy of include/asm/openprom.h so > the actual kernel copy isn't even used. > > I'd say we can stop exporting that header and also therefore not > worry about making phandle_t visible to userspace. Yay! That simplifies everything, and I believe it also means that the phandle/ihandle definitions can remain where they currently are in linux/of.h Andres, can you post an updated series that includes removing openprom.h from the header export list? Thanks, g.