From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:40164) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Y1Axd-0000qe-1I for qemu-devel@nongnu.org; Wed, 17 Dec 2014 04:32:08 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1Y1AxV-0001Yr-Hm for qemu-devel@nongnu.org; Wed, 17 Dec 2014 04:32:00 -0500 Received: from cantor2.suse.de ([195.135.220.15]:37548 helo=mx2.suse.de) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1Y1AxV-0001Yi-Ah for qemu-devel@nongnu.org; Wed, 17 Dec 2014 04:31:53 -0500 Message-ID: <54914D87.6080601@suse.de> Date: Wed, 17 Dec 2014 10:31:51 +0100 From: Alexander Graf MIME-Version: 1.0 References: <1418399932-7658-1-git-send-email-lersek@redhat.com> <1418399932-7658-2-git-send-email-lersek@redhat.com> <20141216134858.GD3283@hawk.usersys.redhat.com> <5490815E.8@redhat.com> <54908CB6.8030501@redhat.com> <549090D8.5010006@redhat.com> <549098D2.5020001@redhat.com> <54910F67.2000804@redhat.com> <54914B85.7070605@redhat.com> In-Reply-To: <54914B85.7070605@redhat.com> Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [PATCH v4 1/8] fw_cfg: max access size and region size are the same for MMIO data reg List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Paolo Bonzini , Laszlo Ersek , Andrew Jones Cc: peter.maydell@linaro.org, qemu-devel@nongnu.org On 17.12.14 10:23, Paolo Bonzini wrote: > > > On 17/12/2014 06:06, Laszlo Ersek wrote: >>>> That said, the standalone selector is used by BE platforms only, so we >>>> know that the standalone selector is always DEVICE_BIG_ENDIAN. >> This series exposes the standalone selector (as MMIO) to ARM guests as >> well; and in "Documentation/devicetree/bindings/arm/fw-cfg.txt" for the >> kernel I'm saying that the selector is little endian. >> >> Therefore I think that the standalone selector is not only (going to be) >> used by BE platforms (or I don't understand your above statement >> correctly). > > It's not going to be used only by BE platforms. But so far it is used > by BE platforms only. If you want to keep everything consistent, you > can make it (and the wide datum) BE; and swizzle data in the firmware. > Otherwise, NATIVE_ENDIAN is fine. For the sake of keeping myself sane, I would really prefer not to have NATIVE_ENDIAN. Once you start thinking about little endian guests on PPC sharing code with little endian guests on x86 your head will start to explode otherwise. If code becomes simpler by making the mmio accessors BIG_ENDIAN, that's fine with me as well. Alex