From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from [209.85.132.245] (helo=an-out-0708.google.com) by linuxtogo.org with esmtp (Exim 4.63) (envelope-from ) id 1HRa7u-0007s5-HO for openembedded-devel@lists.openembedded.org; Wed, 14 Mar 2007 21:31:15 +0100 Received: by an-out-0708.google.com with SMTP id c5so307346anc for ; Wed, 14 Mar 2007 13:31:13 -0700 (PDT) Received: by 10.100.133.9 with SMTP id g9mr2481800and.1173904273279; Wed, 14 Mar 2007 13:31:13 -0700 (PDT) Received: from cube ( [82.193.98.7]) by mx.google.com with ESMTP id r1sm12436685nzd.2007.03.14.13.31.11; Wed, 14 Mar 2007 13:31:12 -0700 (PDT) Date: Wed, 14 Mar 2007 22:31:13 +0200 From: Paul Sokolovsky X-Priority: 3 (Normal) Message-ID: <1259297402.20070314223113@gmail.com> To: openembedded-devel@lists.openembedded.org In-Reply-To: <1173869049.5834.16.camel@localhost.localdomain> References: <1173869049.5834.16.camel@localhost.localdomain> MIME-Version: 1.0 Subject: Re: RFC: Kernel image packaging X-BeenThere: openembedded-devel@lists.openembedded.org X-Mailman-Version: 2.1.9 Precedence: list Reply-To: openembedded-devel@lists.openembedded.org List-Id: Using the OpenEmbedded metadata to build Distributions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 14 Mar 2007 20:31:23 -0000 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hello Richard, Wednesday, March 14, 2007, 12:44:08 PM, you wrote: > Hi, > The kernel packaging has given various people concern for a while. The > problem is some devices don't want a kernel binary installed into the > root filesystem. We can do this at the moment but the kernel doesn't end > up packaged and we would like those packages to be available even if > they're not installed by default on a given device. Thanks for going for this issue! [] > For a device that doesn't want the kernel in the rootfs, it would simply > set: > RDEPENDS_kernel-base = "" > in the appropriate kernel file. Perhaps the definition below should be > a ?= and the above could then be set in the machine.conf file. > Does anyone see a nicer way to handle this or have an alternative? I guess this is just the correct solution which handles this setting where it should be, and doesn't compromise packaging. > If nobody objects, this will probably get committed at the weekend. Hope to see it soon! > Cheers, > Richard [] -- Best regards, Paul mailto:pmiscml@gmail.com